Study on the absorption characteristics of euscaphic acid and tiliroside in fruits of Rosa laxa Retz.

Materials and reagents used in the experiment

FRL was collected in August 2019 from Yamalik Mountain in Urumqi city (87°54′E, 43°79′N; altitude, 1,120.0 m) with a temperature of 27 °C. It was identified by Professor Xu Haiyan of Xinjiang Medical University as the dried fruit of Rosa laxa Retz., a plant from the Rosaceae family. The Caco-2 cell line was purchased from Wuhan Boster Biological Engineering Co., Ltd (item number: CX0064; Wuhan, China). The passage numbers of the cells used in the experiments were between 20 and 40 (product number: CX0064). Euscaphic acid (YRY097-210101, ≥98%) and gallic acid (YRM048, ≥98%) were purchased from Chengdu Yijie Rui Biotechnology Co., Ltd (Chengdu, China). Transwell^® 12-well transport chambers (5662), 12-well plates, 96-well plates, and culture flasks were all purchased from the Costar Corporation (Corning, NY, USA). Fetal bovine serum (FBS) (FND622) was purchased from EKOSAI Biotechnology Co., Ltd (Shanghai, China). Dulbecco’s Modified Eagle Medium (DMEM) culture medium (Batch No.: 7121572), non-essential amino acids (11140050), penicillin-streptomycin dual-antibiotic solution (10378016), trypsin (25200056), and D-Hank’s balanced salt solution (88284) were all purchased from the Gibco (Waltham, MA, USA). Fluorescein sodium (F8140), Verapamil (IV0040), Cyclosporin A (SC5120), ethylenebis (oxyethylenenitrilo) tetraacetic acid (EGTA) (E8050), and sterile dimethyl sulfoxide (DMSO) (D8371) were all purchased from Solarbio Technology Co., Ltd (Beijing, China). Methanol and acetonitrile (high-performance liquid chromatography (HPLC) grade) were purchased from the Thermo Fisher Scientific (Waltham, MA, USA).

Establishment of the determination method for euscaphic acid and tiliroside

Prepare euscaphic acid and tiliroside to make a stock solution of 100 μg/mL, stored for future use (in the subsequent addition of the drug to the cell culture solution, the amount of DMSO should not exceed 0.1% of the volume of the cell culture solution).

Establishment of a response surface model

Establishment of the cell model

The evaluation of the Caco-2 cell model primarily involves using fluorescein sodium as a marker. The cells are seeded in a 12-well cell plate (10⁵ cells/well) and observed and photographed every two days under a fluorescence inverted microscope. The transepithelial electrical resistance (TEER) is measured using a resistance meter (Millipore Millicell ERS-2; Millipore Sigma, Darmstadt, Germany). A TEER value of >400 Ω·cm² indicates that Caco-2 cells have formed a tight junction monolayer, suitable for transmembrane transport experiments (Srinivasan et al., 2015). Markers used for detection mainly include mannitol, inulin, and fluorescein sodium. When the apparent permeability coefficient (Papp) of the marker is <10⁻⁶ cm/s, it indicates that the integrity of the monolayer cells is good (Li et al., 2021).

MTT assay

The MTT colorimetric method is used to investigate cell toxicity. The detection principle is that succinate dehydrogenase in the mitochondria can reduce MTT to water-insoluble blue-violet formazan crystals, while dead cells lack this function. Caco-2 cells at a density of 1 × 10⁴ cells/mL in the logarithmic growth phase are inoculated into each well of a 96-well plate (200 µL per well) and cultured for 24 h. After removing the culture medium, a complete culture medium with different concentrations of FRLE is added to each well, with six replicates for each concentration. The control group is given a complete culture medium without the drug. After incubation for 24, 48, and 72 h, 20 µL of 5 mg/mL MTT solution is added to each well and cultured for another 4 h. The culture medium is then discarded, and 150 µL of DMSO is added to dissolve the blue-violet formazan crystals in each well. After thorough mixing, the absorbance is measured at 570 nm using a microplate reader, and the cell survival rate is calculated (Alizadeh-Navaei et al., 2016).

Absorption characteristics study

A precise amount of FRLE reserve solution is diluted to concentrations of 50, 100, and 150 μg/mL, representing the low, medium, and high concentrations of FRLE. Additionally, solutions are prepared containing 0.100 μmoL/mL Verapamil (Ver), 0.099 μmoL/mL Cyclosporine A (CsA), and 0.025 μmoL/mL EGTA at low, medium, and high concentrations for bidirectional transport.

Caco-2 cells are seeded at a density of 2.5 × 10⁵ cells/mL, with 0.5 mL inoculated into each well of a Transwell^® 12-well plate. After continuous culturing for 18 days, the cell model is established. Bidirectional transport experiments include AP→BL (B→A, absorption transport) and BL→AP (A→B, secretion transport).

Taking absorption transport as an example: 0.5 mL of drug-containing Hank’s Balanced Salt Solution (HBSS) solution is added to the apical (AP) side (supply side), while 1.5 mL of blank solution is added to the basolateral (BL) side (receiving side). At time points 15, 30, 60, 90, 120, 150, and 180 min, 200 µL is drawn from the BL side and simultaneously supplemented with 200 µL of HBSS buffer. The samples are then evaporated in a water bath, re-dissolved in methanol, and centrifuged at 12,000 rpm for 10 min. The supernatant is taken and assayed under the conditions described in “Establishment of the determination method for euscaphic acid and tiliroside”. The effects of different times, concentrations, inhibitors, chelating agents, and pH values on the absorption characteristics of euscaphic acid and tiliroside in FRLE are investigated. The secretion transport procedure is identical to the absorption transport (Artursson & Karlsson, 1991).

Calculation formulas and statistical analysis

The side with the added blank solution is considered the receiving side, and the amount of drug in the receiving side is regarded as its absorption amount (Qr). In the A→B transport test, the amount of drug absorbed by the B-side chamber at any time point is denoted as (QrBi). In the B→A transport test, the amount of drug absorbed by the A-side chamber at any time point is (QrAi). The specific calculations are provided in the following formulas:

A-B transport: Q_rBi = 0.2 × (C_r1 + C_r2 + Λ + C_r(i-1)) + 1.5 × C_ri

B-A transport: Q_rdi = 0.2 × (C_r1 + C_r2 + Λ + C_r(i-1)) + 0.5 × C_di

In the formula, 1.5 represents the volume of the test solution added to the B-side chamber (mL), 0.5 represents the volume of the test solution added to the A-side chamber (mL), and 0.2 is the volume of each sample taken (mL). Cri is the actual concentration in the receiving chamber at the i-th time point (μg/mL). The apparent permeability coefficient Papp (cm/s) is calculated using the following formula (Lu et al., 2020):

$$Papp={\displaystyle \frac{dQ}{dt}\times {\displaystyle \frac{1}{A\times C_0}}}$$dQ/dt represents the amount of drug transported per unit of time (μg/s); A is the surface area of the 12-well Transwell^® membrane (1.12 cm²), and C₀ is the initial concentration of the drug in the donor side (μg/cm) (Qu et al., 2021).

Exhaust ratio (R_B-A/A-B)) is calculated using the following formula: $R_{B-A/A-B}={\displaystyle \frac{Papp\left(B-A\right)}{Papp\left(A-B\right)}}$

SPSS 20.1 software (IBM, Armonk, NY, USA) was used for analysis of variance (ANOVA) on the data. Results are represented by $\overline{\mathrm{x}}\pm \mathrm{s}$. The t-test was employed for intergroup difference comparison. When P < 0.05, the results are considered statistically significant.

Methodology establishment for euscaphic acid, tiliroside, and total polyphenols

The standard curve for euscaphic acid is A = 30,214X + 19,246 (r = 0.9999; 0.1 $\sim$32 μg/mL; LOD = 0.03 μg/mL, LOQ = 0.10 μg/mL). The standard curve for tiliroside is A = 573833x − 3040.9 (r = 0.999; 0.005∼1 μg/mL; LOD = 0.0015 μg/mL, LOQ = 0.005 μg/mL). The standard curve for gallic acid is A = 108.03x + 0.1896 (r = 0.999; 10.30 ~ 92.70 μg/mL; LOD = 3.09 μg/mL, LOQ = 10.30 μg/mL) (Figs. 1, 2).

The precision, repeatability, and stability of euscaphic acid, tiliroside, and gallic acid were evaluated, with all RSD values being <2.00%, indicating good consistency. Recovery studies were conducted by spiking known amounts of each standard into the sample matrix. The average recovery rate for euscaphic acid was (99.75 ± 1.14)% with an RSD of 1.14%, for tiliroside it was (100.22 ± 2.78)% with an RSD of 2.78%, and for gallic acid it was (99.08 ± 0.56)% with an RSD of 0.57%. These results demonstrate the high accuracy of the analytical method, confirming both the efficiency of the extraction and the reliability of the analysis (Table 4).

Response surface experiment

The first step in the response surface experiment is a single-factor investigation. This mainly includes solid-liquid ratio, ethanol concentration, and extraction time. When the solid-liquid ratio increased from 1:15 (g/mL) to 1:30 (g/mL), the composite score significantly increased, rising from 75.53 to 92.38. Further increasing the solid-liquid ratio resulted in a significant decrease in the composite score. Using 60% ethanol as the extraction solvent yielded the highest composite score of 91.49. As the reflux extraction time extended, the composite score first increased and then decreased, with the highest score of 93.34 achieved at 150 min (Fig. 3A–3C).

Subsequently, the Box-Behnken design using Design Expert 11 software was used to design the experimental scale (Table 5) and calculate the composite scores. Based on the established model, 3D response surface plots and contour plots illustrating the effects of various influencing factors on the composite score were generated (Fig. 3D–3I). The figures show that the interaction of feed/liquid ratio and extraction time had a significant impact on extraction efficiency, followed by the interaction of ethanol concentration and extraction time. While drawing the contour plots, a binomial regression model was also obtained. The formula for the composite score is:

$$\begin{array}{rl}\mathrm{C}\mathrm{o}\mathrm{m}\mathrm{p}\mathrm{o}\mathrm{s}\mathrm{i}\mathrm{t}\mathrm{e}\mathrm{S}\mathrm{c}\mathrm{o}\mathrm{r}\mathrm{e}=& 85.45+0.0122\times \mathrm{A}+0.2461\times \text{B}-0.0002\times \text{C}\text{-}1.13\times \text{AB}\\ & +0.2982\times \text{AC}\text{-}1.29\times \text{BC}\text{-}1.24\times \text{A}2\text{-}4.89\times \text{B}2\text{-}4.24\times \text{C}2\end{array}$$

Based on the model, it can be deduced that A, B, C, A2, AB, AC, and BC are not significant; B2 and C2 are significant (Table 6). Based on the above results, the optimal extraction conditions were determined to maximize extraction yield, total polyphenol content, and Euscaphic acid content. The initial conditions were identified as a solid-liquid ratio of 1:35.72 g/mL, 67.14% ethanol, and extraction time of 135.34 min. For practical convenience, the parameters were adjusted to a solid-liquid ratio of 1:35 g/mL, 65% ethanol, and a reflux time of 135 min.

The optimized extraction process was verified by conducting three repetitions. The results showed that the average extraction yield was 51.00 ± 1.07%, the average total polyphenol content was 126.55 ± 2.61 mg/g, and the average euscaphic acid content was 2.90 ± 0.08 mg/g. The composite score, which serves as a comprehensive measure of extraction efficiency for all target components, was 82.38, with an average deviation of 1.01% from the predicted value of 80.217. This indicates that the extraction model meets the experimental criteria and is suitable for achieving the desired outcomes across multiple response variables.

Evaluation of the Caco-2 cell model and MTT assay

The cultured Caco-2 cells were observed and photographed under a fluorescent inverted microscope on days 5, 9, 13, and 18 (Fig. 4A). The results show that by the 18th day, the cells were closely aligned, indicating the successful establishment of the Caco-2 cell monolayer model. TEER value results demonstrate a continuous rise with increased culture time, stabilizing around the 20th day. With a TEER value >400 Ω/cm², it suggests that the electrical resistance of the Caco-2 cell monolayer model meets the experimental requirements (Fig. 4B). Concurrently, the Papp of fluorescein sodium was determined through four repeated measurements, showing a Papp of (4.06 ± 0.17) × 10⁻⁷ cm/s. The results of these three validations indicate that the Caco-2 cell model was successfully established. The cell survival rate was calculated as follows: (OD value of experimental wells—OD value of blank wells)/(OD value of control wells-OD value of blank wells) × 100%. The safe concentration range of FRLE is 0.5 ~ 100 μg/mL (Fig. 4C).

Absorption characteristics experiment of euscaphic acid

The Qr of euscaphic acid in FRLE is proportional to its concentration. There is a significant difference between the Qr from A to B (A→B) and the Qr from B to A (B→A). The Papp of euscaphic acid is greater than 10⁻⁶, indicating that euscaphic acid is well absorbed. The Papp value of transport from A→B is significantly different from the Papp value of transport from B→A. There is no significant difference in the absorption amount on the AP side compared to the BL side, indicating that the transport mechanism of euscaphic acid is passive (Table 7, Figs. 5A–5C). After the addition of the P-glycoprotein (P-gp) inhibitor, the Qr and Papp of euscaphic acid in FRLE significantly increased, with an R_B→A/A→B value around 1.20. This suggests that some components in FRLE combine with euscaphic acid to form substrates for P-gp, resulting in an increased Qr of euscaphic acid when the P-gp inhibitor is added. Moreover, the results also show that there isn’t a significant difference in bidirectional transport (Table 7, Figs. 5D–5F). After the addition of the EGTA chelating agent, the transport amount and Papp value of euscaphic acid in FRLE showed a significant increase, with an R_B→A/A→B value ranging between 0.8 and 1.4. This indicates that some component in FRLE interacts with euscaphic acid, serving as a substrate for the EGTA chelating agent, facilitating paracellular transport (Table 7, Figs. 5G–5I). In an acidic environment (pH 6.0), the Qr and Papp values of euscaphic acid in FRLE are the highest, with an R_B→A/A→B value around 1.0. The absorption is lowest in a neutral solution (pH 7.2), indicating that an acidic environment is favorable for the transport of euscaphic acid (Table 7, Figs. 5J–5O).

Absorption characteristics experiment of tiliroside

The Qr and Papp of tiliroside in FRLE are proportional to its concentration, and the Papp is greater than 10⁻⁶, indicating that tiliroside is well absorbed. The transport from A to B (A-B) is significantly lower than the Papp value of transport from B to A (B-A), suggesting that tiliroside’s transport mechanism is passive (Table 8, Figs. 6A–6C). Upon adding the P-gp inhibitor, the bidirectional transport Qr of tiliroside in FRLE significantly increased, with an R_B→A/A→B value ranging from 0.7 to 1.3. This indicates that tiliroside in FRLE is a substrate for P-gp, and there’s no significant difference in its bidirectional transport (Table 8, Figs. 6D–6F).

After the addition of the EGTA chelating agent, the Qr and Papp of tiliroside in FRLE significantly increased. The R_B→A/A→B value of tiliroside in FRLE ranged between 0.9 and 1.2, suggesting no noticeable difference in bidirectional transport (Table 8, Figs. 6G–6I). In an acidic environment (pH 6.0), the Qr and Papp of tiliroside in FRLE were the highest, while the absorption was lowest in a neutral solution (pH 7.2). The R_B→A/A→B value of tiliroside in FRLE was around 1.0, indicating that an acidic environment is favorable for the transport of tiliroside, with no noticeable directionality in its transport (Table 8, Figs. 6J–6O).