HPLC-grade acetonitrile and HPLC-grade methanol used in the study were purchased from Merck (Darmstadt, Germany). HPLC-grade acetic acid was purchased from Aladdin (Shanghai, China) and analytical-grade ethanol was obtained from Sinopharm Chemical Reagents (Shanghai, China). Free-fatty acid HSA (lyophilized powder) was acquired from Sigma-Aldrich (St. Louis, MO, United States). Reference standards of amentoflavone (batch number: 160,529) and chrysin (150,204) were purchased from Winherb (Shanghai, China). Delicaflavone, robustaflavone, 2″,3″-dihydro-3′,3‴-biapigenin, and 3′,3‴-binaringenin (purity ≥98%) were isolated from S. doederleinii and their molecular structures were fully characterized by FT-IR, UV, MS, ¹H-NMR, and ¹³C-NMR (Chen et al., 2019). Ultrapure water was purified using a Milli-Q system (MA, United States).

Dried S. doederleinii Hieron was obtained from a local traditional Chinese medicine store in Fuzhou, and identified by Pro. Yao (Fujian Medical University, Fuzhou, China). A voucher specimen (batch number: 201,909) has been deposited at the Phytochemistry Laboratory, Fujian Medical University (Fuzhou, China).

The process of obtaining TBESD and characterization of its main components were performed as previously described (Li et al., 2014). The contents of amentoflavone, delicaflavone, robustaflavone, 2″,3″-dihydro-3′,3‴-biapigenin, and 3′,3‴-binaringenin in TBESD, quantified by our previous HPLC method (Supplementary Figure S1), was 103.82, 35.12, 37.52, 44.40, and 53.36 mg/g, respectively.

Thirty healthy Sprague–Dawley rats (weighing 250 ± 20 g) were obtained from the Laboratory Animal Center of Fujian Medical University (Fuzhou, China). All rats were maintained in an environmentally controlled breeding room (temperature: 25 ± 2°C and relative humidity: 55 ± 5%) at a 12-/12-h light/dark cycle and were observed for 1 week before the experiment. The rats were provided rat feed and water ad libitum; before drug administration, they were fasted for 12 h but were allowed free access to water. All experiments were conducted in accordance with the Guidelines for the Care and Use of Laboratory Animals approved by the Animal Ethics Committee of Fujian Medical University.

HPLC-ESI-MS/MS was conducted on a Shimadzu LC-20AD HPLC system with a Shimadzu LC/MS-8040 instrument (Shimadzu, Kyoto, Japan) using an Ultimate^® XB-C18 column (50 mm × 4.6 mm, 3.5 μm; Welch Materials, Inc., Ellicott, MD, United States). The sample injection volume was 5 μL. Gradient elution was performed at 30°C with 0.5% glacial acetic acid in water (solvent A) and acetonitrile (solvent B) at a flow rate of 0.2 ml/min. The gradient elution was set as follows, 0–1 min (40–44% B); 1–14 min (44–49.5% B); 14–15 min (49.5–95% B); 15–16 min (95% B); 16–18 min (40% B) for equilibration. The sample injection volume was 5 μL.

Triple-quadrupole MS/MS detection was carried out on a Shimadzu LC/MS-8040 system with electrospray ionization (ESI). Multiple reaction monitoring analysis in the negative ion mode was conducted based on ion transitions of amentoflavone, delicaflavone, robustaflavone, 2″,3″-dihydro-3′,3‴-biapigenin, 3′,3‴-binaringenin, and chrysin (internal standard, IS) at m/z 537.08→375.00 (CE = 35 V), m/z 537.08→255.00 (CE = 55 V), m/z 537.08→309.10 (CE = 40 V), m/z 541.11→237.09 (CE = 30 V), m/z 539.10→387.15 (CE = 45 V), and m/z 253.24 → 143.00 (CE = 28 V), respectively.¹¹ The optimized MS parameters were set as follows: ion spray voltage, 6.0 kV; heat block temperature, 400°C; DL temperature, 250°C; drying gas flow, 12 L/min; nebulizer gas flow, 3 L/min. Data acquisition and processing were performed using LabSolutions LCMS Ver. 5.5 software.

Standard stock solutions of amentoflavone, delicaflavone, robustaflavone, 2″,3″-dihydro-3′,3‴-biapigenin, 3′,3‴-binaringenin, and chrysin were prepared by dissolving them individually in methanol to a final concentration of 1 mg/ml; standard working and QC solutions were obtained by further dilution in methanol. For tissue distribution and excretion analyses, standard working solutions were prepared by serial dilution to obtain a linear concentration gradient. Calibration standards of delicaflavone, robustaflavone, 2″,3″-dihydro-3′,3‴-biapigenin, and 3′,3‴-binaringenin were prepared by spiking drug-free blank biological samples with the working solutions to obtain concentrations of 2, 5, 10, 25, 50, 125, 250, and 500 ng/ml and QC samples of 6, 80, and 400 ng/ml; for amentoflavone, standards of 4, 10, 20, 50, 100, 250, 500, and 1,000 ng/ml, and QC samples of 12, 160, and 800 ng/ml were prepared. The chrysin stock solution was diluted to a final concentration of 50 ng/ml. All stock solutions were stored at 4°C until use.

To obtain tissue samples for analysis, each organ was weighed, diced into small pieces, and homogenized in three volumes of ice-cold physiological saline. Thereafter, 100 μL of the tissue homogenate was spiked with 10 μL of IS working solution, mixed by vortexing with 300 μL of methanol for 2 min, centrifuged at 15,000 × g for 20 min at 4°C to precipitate proteins, and the resultant supernatant (5 μL) was used for HPLC-MS/MS.

The feces were dried at 50°C to a constant weight and homogenized in three volumes of physiological saline-methanol (1:1, v/v). Thereafter, 100 μL of fecal and urine samples were processed as described above for tissue samples. The HPLC-ESI-MS/MS method developed and validated previously was only used for rat plasma samples. In the present study, the method was further verified for specificity, linearity, accuracy, precision, recovery, and stability of tissue, urine, and fecal samples.

To assess the tissue distribution of orally administered TBESD, 24 Sprague–Dawley rats (12 females and 12 males, weight 200 ± 20 g) were randomly divided into four groups (3 females and 3 males per group) and intragastrically administered a single dose of TBESD (600 mg/kg) dissolved in a mixture of ethanol (2%) and propylene glycol (3.5%) in physiological saline (pH adjusted to 7.5 with 0.5 M NaOH). The rats were sacrificed by cervical dislocation at 5, 15, 40, and 90 min after TBESD administration; the time points for tissue collection were determined based on the concentration–time curves. The brain, heart, kidney, liver, lung, ovary, spleen, testes, and muscle tissues were immediately harvested, thoroughly rinsed in ice-cold physiological saline, dried with filter paper, and stored at −80°C until analysis.

To investigate urinary and fecal excretion, 3 female and 3 male rats housed in individual stainless-steel metabolic cages were orally administered a single dose of TBESD (600 mg/kg), and urine and fecal samples were collected after 0–3, 3–6, 6–9, 9–12, 12–24, 24–36, and 36–48 h. After drying the fecal samples to a constant weight at 50°C and measuring urine sample volumes, the specimens were stored at −80°C.

The interactions of TBESD with HSA were analyzed by fluorescence spectroscopy. HSA (10 μM in phosphate buffer, pH 7.4) was incubated in the presence and absence of various TBESD concentrations (2.5, 5, 7.5, 10, and 20 μg/ml), and HSA intrinsic fluorescence spectra were recorded at 37°C. Fluorescence measurements were performed using a Cary Eclipse spectrofluorometer (CA, United States) with a thermostat bath using a 1.0-cm quartz cuvette. The width of the excitation and emission slits was set to 5 nm. An excitation wavelength of 295 nm was used in the experiments to avoid interference from tryptophan and/or tyrosine residues (Strugała et al., 2018). To avoid the inner filter effect, the fluorescence intensities were corrected using the following formula: F c o r r =   F o b s × a n t i l o g [ ( A e m + A e x ) 2 ] (1) where, A_em and A_ex are the absorbance of the test sample at the excitation and emission wavelengths, respectively; F_obs and F_corr are the observed and corrected fluorescence intensities, respectively.

Fluorescence quenching experiments were performed at different concentrations of TBESD (2.5, 5, 7.5, 10, and 20 μg/ml) to HSA (10 μM) at different temperatures (298, 303, 308, and 313 K). The emission spectra were recorded in the range of 300–480 nm and then calculated using the Stern–Volmer equation: F 0 F = 1 + K S V [ Q ] = 1 +   k q τ 0 [ Q ] (2) where, F and F₀ are HSA fluorescence intensities in the presence and absence, respectively, of the quencher (TBESD); [Q] is the concentration of TBESD; K_SV is the dynamic quenching constant; K_q is the apparent bimolecular quenching rate constant; and τ₀ is the fluorophore lifetime without TBESD (10^–8 s).

The apparent binding constant (K_b) and number of binding sites (n) per macromolecule at the corresponding temperature were analyzed using the modified Stern–Volmer equation: log { F 0 − F F } = log ⁡ K b + n ⁡ log [ Q ]   (3)

The thermodynamic parameters of the TBESD-HSA binding were used to determine the binding mode. The thermal dependency of the binding constant was investigated at 298, 303, 308, and 313 K, and enthalpy and entropy changes (ΔH° and ΔS°, respectively) were analyzed based on the van’t Hoff plot to determine the binding forces between TBESD and HSA: ln ⁡ K b = − ( Δ H ° R T ) + ( Δ S ° R ) (4) where, R is the universal gas constant and T is the absolute temperature.

Assuming that ΔH° is nearly constant in the studied temperature range, the free energy change (ΔG°) can be calculated from the Gibbs equation: Δ G ° =   Δ H ° − Δ S ° (5)

Data acquisition and processing were conducted using LabSolutions LCMS Ver.5.5 software. The concentrations of bioflavonoids in tissues, urine, and feces were determined using the calibration curve of each analysis batch. The pharmacokinetic parameters of bioflavonoids were calculated using a non-compartmental approach with DAS 3.0 Version (Shanghai, China).

All data are expressed as mean ± standard deviation (SD). Independent t-tests were performed to assess the differences between the groups; p < 0.05 was considered to indicate statistical significance.

A sensitive and reliable HPLC-ESI-MS/MS method for the simultaneous quantification of five bioflavonoids (amentoflavone, delicaflavone, robustaflavone, 2″,3″-dihydro-3′,3‴-biapigenin, and 3′,3‴-binaringenin) in rat tissue, urine, and fecal samples was successfully developed and validated according to the Food and Drug Administration guidelines (Kamble et al., 2021; Shen et al., 2021). The representative chromatograms of blank rat lung homogenate, blank lung homogenate spiked with QC and IS samples, and lung homogenate samples collected 5 min after TBESD (600 mg/kg) administration are shown in Figure 1. Amentoflavone, robustaflavone, 2″,3″-dihydro-3′,3‴-biapigenin, 3′,3‴-binaringenin, IS, and delicaflavone were eluted at approximately 7.2, 8.4, 10.8, 11.2, 12.5, and 14.7 min, respectively. No detectable interfering peaks were observed in tissue, urine, or fecal samples at the retention time close to that of the five bioflavonoids and IS. The detailed method validation parameters are summarized in Tables 1–4 and Supplementary Tables S1–S12. The linearity of all calibration curves was characterized with regression coefficients (R ²) of 0.99 or higher. The intra- and inter-day precision (RSD, %) and accuracy (RE, %) of the method for each bioflavonoid were within the acceptable limit of 15%, and the matrix effect and extraction recovery of each bioflavonoid were consistent among the three different concentrations (n = 5 at high, medium, and low) and constituted >80% of the response. The five bioflavonoids were stable on the bench for up to 8 h, in the autosampler for up to 12 h, at −80°C for up to 2 months at three different concentrations, and after up to three cycles of freezing and thawing (Du et al., 2020).

The tissue distribution profiles of the five bioactive ingredients in the rats after the oral administration of TBESD (600 mg/kg) are depicted in Figure 2. The highest concentration of the five bioflavonoids was observed in the lungs at 5 min, followed by the kidneys, muscle, ovary, spleen, and testes, suggesting that amentoflavone, delicaflavone, robustaflavone, 2″,3″-dihydro-3′,3‴-biapigenin, and 3′,3‴-binaringenin distributed rapidly in various tissues. The peak levels in the lungs, kidneys, muscle, ovary, spleen, and testes were significantly higher than those in the plasma. However, the content of the five bioflavonoids in the lungs significantly decreased, whereas that in the kidneys, ovary, and spleen initially increased and then significantly decreased. Only a small amount of the five bioflavonoids was detected in the brain, indicating that amentoflavone, delicaflavone, robustaflavone, 2″,3″-dihydro-3′,3‴-biapigenin, and 3′,3‴-binaringenin could not easily cross the blood–brain barrier. The highest content of the five bioflavonoids in the lungs after oral TBESD administration suggests that TBESD can be used to treat lung-related diseases (Sui et al., 2017). Clearly, the pharmacological effects of TBESD are results from multi-ingredients and multi-targets synergistic integration, thus, the integrated pharmacokinetic studies have carried out to reveal the overall in vivo process of TBESD. The half maximal inhibitory concentration (IC₅₀) of an ingredient against human lung cancer cell line (take A549 cell line for example) curve was used to the integrated pharmacokinetic studies of TBESD (Shi et al., 2018). The IC₅₀ of amentoflavone, delicaflavone, robustaflavone, 2″,3″-dihydro-3′,3‴-biapigenin, and 3′,3‴-binaringenin were 36.3, 13.2, 50.0, 40.2, and 19.3 μg/ml, respectively (Li et al., 2014; Sui et al., 2016). The tissue distribution profiles of multiple ingredients of TBESD are shown in Supplementary Figure S2. The weighting coefficient and integrated concentrations are calculated by following equations: ω j = 1 / I C 50 j ∑ 1 n 1 / I C 50 n (6) ∑ 1 n 1 I C 50 n = 1 I C 501 + 1 I C 502 + 1 I C 503 + ⋯ + 1 I C 50 n (7) C T = ω 1 C 1 + ω 2 C 2 + ω 3 C 3 + ⋯ + ω n C n (8) Where j represents ingredient j, ω _j represents the weighting coefficient of the ingredient j, n represents the number of ingredients studied, C _n represents the concentration of each ingredients at time point T, and C _T represented the integrated concentration, respectively.

The urinary and fecal excretion profiles of the five bioflavonoids after a single oral administration of TBESD are shown in Figure 3. The cumulative concentration of amentoflavone, delicaflavone, robustaflavone, 2″,3″-dihydro-3′,3‴-biapigenin, and 3′,3‴-binaringenin excreted in feces and urine was 23.93 and 0.82%, 29.11 and 0.43%, 29.58 and 0.63%, 29.31 and 0.25%, and 20.51 and 0.32%, respectively, up to 48 h, indicating a considerably higher excretion rate in feces than in urine. These results suggest fecal excretion as the main excretion route for the five bioflavonoids of TBESD, and this is consistent with the findings of a previous study on flavonoids (Zheng et al., 2019).

To the best of our knowledge, this is the first study to report that TBESD is excreted mainly through feces. In addition, the cumulative excretion of the five bioflavonoids in feces after the oral administration of TBESD was considerable, which may have contributed to their low bioavailability and could be attributed to the abundant expression of organic anion transporters and multidrug resistance-related proteins in the membrane of intestinal epithelial cells (Barreca et al., 2017).

The tryptophan residues in HSA are highly sensitive to polarity changes, making fluorescence spectroscopy a valuable method for studying HSA conformational shifts after binding to different ligands (Balaei and Ghobadi, 2019). TBESD exhibited very low fluorescence under the present experimental conditions. Quenching of intrinsic HSA fluorescence was measured in the presence and absence of different concentrations of TBESD at 298 K (Figure 4A). The results indicated that the fluorescence intensity of HSA decreased with the increase in TBESD concentration, suggesting the presence of tryptophan residues at or near the HSA-TBESD binding site.

There are two types of fluorescence quenching mechanisms: dynamic and static. The dynamic quenching occurs in the presence of high quencher concentrations, and the effective number of collisions increases at higher temperatures; therefore, the increase in temperature should cause an increase in the quenching constant (K_SV). In contrast, in static quenching, the stability of the formed complex decreases with the increase in temperature, which results in the decrease in K_SV (Strugała et al., 2021). To elucidate the mechanism underlying TBESD quenching of HSA fluorescence, we analyzed HSA fluorescence at four different temperatures (Figure 4 and Table 5). The Stern–Volmer plots for the TBESD-HSA complexes were linear, showing a directly proportional relationship between the examined concentration range and the temperature. The number of binding sites (n) for TBESD in HSA at different temperatures is shown in Table 5. The results indicated that there was one binding site, that is, only one interaction point between the TBESD components and HSA. According to the obtained results, the most probable mechanism of HSA fluorescence quenching by TBESD should be dynamic quenching, in which higher temperatures increase the stability of the TBESD-HSA complex as indicated by higher K_SV and K_b values.

To characterize the energy changes during the interaction between TBESD and HSA, the thermodynamic binding parameters ΔH°, ΔS°, and ΔG° were calculated from the van’t Hoff plot and Gibbs equation (Table 6). The ΔH° and ΔS° values were 29.10 ± 2.86 kJ (g ml-1)⁻¹ and 0.27 ± 0.01 J (g mL-1 K)⁻¹, respectively. The high negative ΔG° value accompanied by positive ΔS° is indicative of a spontaneous process. The positive ΔS° value signifies hydrophobic interactions, whereas the variations in ΔH° indicate that the binding process is enthalpy-driven and probably depends on the formation of hydrogen bonds between the TBESD components and HSA. Thus, thermodynamic analysis of the binding between TBESD and HSA revealed the role of hydrophobic interactions and hydrogen bonds in complex formation.