Vandetanib, luteolin and regorafenib (used as internal standard, IS) used in the experiment were purchased from Shanghai Perfemiker Chemical Technology Co., Ltd. (Shanghai, China). Nicotinamide adenine dinucleotide phosphate (NADPH) was purchased from Shanghai Aladdin Biochemical Technology Co., Ltd. (Shanghai, China). Methanol and acetonitrile (HPLC-grade) were purchased from Merck (Darmstadt, Germany). Formic acid was purchased from Anaqua Chemicals Supply (ACS, United States). Ultrapure water was from the Milli-Q Water purification system (Millipore, Bedford, United States). RLM was prepared according to the literature (Wang et al., 2015). HLM was purchased from iPhase Pharmaceutical Services Co., Ltd. (Jiangsu, China). All other chemicals and biological products were of analytical grade or above.

The UPLC-MS/MS system had a Waters Acquity UPLC I-Class system (Milford, MA, United States) and a Waters Xevo TQ-S triple quadrupole tandem mass spectrometer (Milford, MA, United States). The Waters Acquity UPLC I-Class was used to separate vandetanib, N-demethyl vandetanib, and IS with an Acquity BEH C18 column (2.1 mm × 50 mm, 1.7 μm) at 40°C. The mobile phase was consisted of 0.1% formic acid (solution A) and acetonitrile (solution B) with a gradient elution at a flow rate of 0.4 mL/min. The total run time was 2.0 min, and the injection volume was 0.5 μL. The temperature conditions were set as follows: the column temperature was 40°C and the autosampler temperature was 10°C.

A Waters Xevo TQ-S triple quadrupole mass spectrometer was equipped with electrospray ionization (ESI) in positive ion mode, and multiple reaction monitoring (MRM) mode was selected for quantification. The parent and product ions of vandetanib, N-demethyl vandetanib and IS were m/z 475.02→112.03, m/z 461.05→363.92 and m/z 483.00→269.97, respectively. The optimal MS parameters were defined as follows: the cone voltages of vandetanib, N-demethyl vandetanib and IS were set to be 10 V, 10 V and 20 V, respectively, and the collision energies were set to be 10 eV, 10 eV and 30 eV, respectively. Masslynx 4.1 software (Milford, MA, United States) was used for data acquisition, and the chromatograms of analytes were shown in Figure 1.

In order to ensure the accuracy and reliability of the analytical methods, we conducted method validation to correctly provide information on the validation parameters. According to the latest guidelines from the FDA, the validation parameters included selectivity, linearity, lower limit of quantification (LLOQ), precision and accuracy, matrix effects, recovery and stability.

Male SD rats weighing 180–220 g were provided by the Laboratory Animal Center of the First Affiliated Hospital of Wenzhou Medical University (Zhejiang, China). They were housed under standard conditions with a temperature of 20°C–26°C, a relative humidity of 55% ± 15%, and a 12-h dark/light cycle. The rats had free access to water and no other drugs were given during the feeding period, and were prohibited from eating for 12 h before the experiment. The experimental procedures and protocols were in accordance with animal ethics standards and approved by the Institutional Animal Care and Use Committee of the First Affiliated Hospital of Wenzhou Medical University (Zhejiang, China).

The rats were randomly assigned to two groups: luteolin group (n = 5) and control group (n = 5). Vandetanib and luteolin were both dissolved in 0.5% sodium carboxyl methyl cellulose (CMC-Na) solution. The luteolin group was given luteolin alone (30 mg/kg) by gavage, while the control group was given the same volume of 0.5% CMC-Na solution. 30 min later, each rat was administered of vandetanib (25 mg/kg) by gavage. Regarding the dosage selection for luteolin and vandetanib, we have chosen 30 mg/kg of luteolin and 25 mg/kg of vandetanib based on references from relevant literature (Lin et al., 2014; Yang et al., 2024). Therefore, we adopted the same dosages to ensure the comparability and validity of the experimental results. 0.3 mL tail vein blood was collected at 0.5, 1, 1.5, 2, 3, 4, 6, 8, 12, 24, 48 and 80 h post-vandetanib administration. The blood samples were centrifuged at 4°C at 13,000 rpm for 10 min to obtain plasma, which was aspirated and stored at −80°C until analysis.

Considering that RLM or HLM microsome may have different metabolic rates, we conducted experiments on enzyme and incubation time before the start of the experiment. Regarding the determination of enzyme concentration, we added vandetanib (50 μM), NADPH, potassium phosphate buffer and different concentrations of RLM/HLM in a 200 μL system, controlling the incubation time to 30 min to measure the production of N-demethyl vandetanib. In the incubation time experiment, we determined the concentration of RLM/HLM to be 0.3 mg/mL and compared the production of N-demethyl vandetanib at different incubation times. A rate-time graph was created to select the most suitable concentration and incubation time within the linear range of variation. Taking into account the experimental cost and production yield, we selected an enzyme concentration of 0.3 mg/mL and an incubation time of 30 min. Thus, the total volume of the incubation mixture was 200 μL and was prepared as follows: 0.3 mg/mL RLM (HLM), 100 mM PBS (pH = 7.4), 1 mM NADPH, vandetanib and luteolin. The drug was dissolved in DMSO. The content of organic solvents was less than 1% (Li et al., 2010).

First, in the RLM system, a series of vandetanib at concentrations of 0.1, 0.5, 1, 10, 20, 50, and 100 μM were added to the reaction buffer to determine the Michaelis-Menten constant (K_m) value. The concentrations of vandetanib in HLM were 1, 10, 50, 100, 300, 800 and 1,000 μM. The relationship between reaction rate and substrate concentration based on the Michaelis-Menten equation was calculated, and the K_m value through nonlinear fitting was determined. For half-maximum inhibitory concentration (IC₅₀) assay, the concentration gradients of luteolin used were 0, 0.01, 0.1, 1, 10, 25, 50 and 100 μM both in RLM and HLM, and the final concentration of vandetanib was the K_m value in the corresponding system. The relationship between inhibitor concentration and reaction rate was plotted after the reaction rate at different inhibitor concentrations was determined, and then the IC₅₀ value through nonlinear regression analysis was calculated. Finally, to determine the inhibitory mechanism of vandetanib on luteolin, a series of luteolin concentrations were produced according to the IC₅₀ levels, where 0, 2.14, 4.28, 8.56 μM were for RLM and 0, 7.92, 15.84, 31.68 μM were for HLM, respectively. And, the concentrations of vandetanib were set according to the K_m value, where the values were 2.35, 4.71, 9.41, 18.82 μM in RLM, and 12.47, 24.94, 49.87, 74.81 μM in HLM, respectively.

The above mixture was incubated at 37°C for 5 min, after which NADPH was added to start the reaction process for 30 min, and it was finally cooled to −80°C to stop the reaction. 400 μL of acetonitrile and 20 μL of IS working solution (200 ng/mL) were incorporated into the mixture. Following vortexed and centrifugated at 13,000 rpm for 10 min, the supernatant was obtained for UPLC-MS/MS analysis.

The molecular structures of vandetanib, luteolin and itraconazole were obtained from the Pubchem database. PyMOL for structural optimization and AutoDock software for molecular docking verification and binding energy calculation were conducted. Finally, the binding interactions between drugs using PyMOL software was evaluated.

K_m, IC₅₀, Lineweaver-Burk plots and mean plasma concentration-time curves were generated using GraphPad Prism 9.5 software. Drug and Statistics (DAS, Chinese Committee of Mathematical Pharmacology, Shanghai, China) was intended to calculate the pharmacokinetic parameters of vandetanib and its metabolite, including time to peak (T_max), maximum plasma concentration (C_max), elimination half-life (t_1/2), area under the drug-time curve (AUC), and clearance (CL_z/F). SPSS (version 26.0; SPSS Inc., Chicago, IL, United States) was used for data processing, and t-test was employed to determine whether there was a significant difference. P value <0.05 indicated a significant difference compared to the control group.

Under the chromatographic conditions listed above, the retention times of vandetanib, the metabolite N-demethyl vandetanib, and IS were 1.16 min, 1.16 min and 1.51 min, respectively. As shown in Figure 1, no interference was detected in rat plasma between vandetanib, N-demethyl vandetanib and IS. The calibration curve of vandetanib was linear over the concentration range of 2–1,000 ng/mL with a typical regression equation of Y = 0.0012972 × X + 0.00147,298 (r ² = 0.996). And, N-demethyl vandetanib was measured from 0.5 to 50 ng/mL, with a regression equation of Y = 0.00119,975 × X + 0.00104,396 (r ² = 0.991). The LLOQ of vandetanib and N-demethyl vandetanib were 2 ng/mL and 0.5 ng/mL, respectively, in this developed UPLC-MS/MS method. The results of accuracy, precision, recovery rate and matrix effect were presented in Supplementary Table S1. Also, the stability results were complied with FDA guidelines. These findings demonstrated that the established method was accurate and dependable.

The mean plasma concentration-time curves of vandetanib and its major metabolite after administration in rats were shown in Figure 2. The detailed pharmacokinetic parameters of vandetanib and N-demethyl vandetanib were shown in Table 1 and Table 2, respectively. In this study, it was found that compared with the control group, the AUC_(0-t) of vandetanib in luteolin group was increased by 18.1% and C_max was increased by 32.8%, while there were no statistically significant differences in other pharmacokinetic parameters. Moreover, there was no significant change in the metabolite. These data suggested that luteolin could significantly increase the plasma concentration and exposure of vandetanib by inhibiting the metabolism of vandetanib in rats.

In vitro study, the results showed that the K_m of vandetanib was 9.41 μM and the IC₅₀ value of luteolin was 8.56 μM in RLM (Figure 3). This suggested that luteolin had an inhibitory effect on vandetanib in vitro. The Lineweaver-Burk plot revealed that the inhibition mechanism was a mixture of non-competitive inhibition and competitive inhibition, with the values of K_i = 2.33 μM and α = 6.68 μM. As shown in Figure 4, in the HLM incubation system, the K_m and IC₅₀ values were 49.87 μM and 15.84 μM, respectively. The Lineweaver-Burk plot showed that the inhibition type of luteolin on vandetanib was a mixture of non-competitive and un-competitive, with K_i and α values of 16.52 μM and 0.28 μM, respectively.

To gain a deeper insight into the interaction mechanism between vandetanib and luteolin, molecular docking was conducted in this study. The results (Figure 5) showed that vandetanib (pink) formed three hydrogen bonds with residues PHE-304 (3.3 Å), PHE-213 (2.1 Å) and LEU-216 (2.5 Å), and the binding energy was −6.98 kcal/mol. Luteolin (cyan) formed three hydrogen bonds with PHE-304, PHE-213 and LEU-215, with sites 2.9 Å, 2.0 Å and 2.1 Å apart, respectively, and binding energy of −8.46 kcal/mol. We also observed that vandetanib and luteolin have common binding sites: PHE-304 and PHE-213, and that they can spontaneously bind to the active catalytic cavity of CYP3A4. At the same time, we observed a hydrogen bonding interaction between itraconazole (slate) and the active site residue ARG-372 (3.4 Å) that stabilized CYP3A4. The binding energy of itraconazole to CYP3A4 was −5.44 kcal/mol.