Indo5 was synthesized as described previously (US Patent Number: 9642839B2). Its purity was determined using high-performance liquid chromatography (HPLC), and the compound was used as the reference standard because of its high purity (>99%). The molecular structure of Indo5 was characterized using Fourier-transform infrared (FT-IR) spectroscopy, ¹H and ¹³C nuclear magnetic resonance (¹H and ¹³C NMR), mass spectrometry (MS), and elemental analyses. These analyses were conducted at a laboratory specializing in pharmaceutical analysis.

The internal standard (IS), terfenadine, was purchased from Sigma-Aldrich (St. Louis, MO, United States). HPLC-grade water was obtained from Watson’s Food and Beverage (Guangzhou, China). HPLC-grade acetonitrile and HPLC-grade methanol were purchased from Thermo Fisher Scientific (Waltham, MA, United States). HPLC-grade formic acid and dimethyl sulfoxide (DMSO) were obtained from Sigma-Aldrich. The anesthetic phenobarbital for veterinary use was acquired from Wuhan Dongkang Source Technology (Wuhan, China). Heparin (1000 UI/ml) and physiologic saline (0.9%) were purchased from Sinopharm Chemical Reagents (Shanghai, China).

Indo5 (10 mg) was dissolved in 10 ml DMSO. The formulation was a clear-yellow solution, which was vortexed for 3 min and sterilized by passing through a 0.22-μm filter. A final concentration of 1 mg/ml was obtained. The solution was stored at 4°C and used to manufacture standards.

Male Wistar rats (300 ± 20 g) and mice (5–6 weeks old) were purchased from Beijing Vital River Laboratory Animal Technology (Beijing, China) and allowed to acclimate to their surroundings for 7 days. All of the animals were housed in individually ventilated cages in specific pathogen-free conditions at the animal facility of our institute. They were exposed to a 12-h light–dark cycle and allowed free access to food and water. The protocol for animal experiments was approved by the Animal Ethics Committee of the Academy of Military Medical Science (Beijing, China), and all of the experiments were conducted in accordance with the National Institutes of Health guidelines.

Calibration standards were prepared daily by diluting stock solutions. Briefly, appropriate volumes of stock solution were spiked into 100 μl of blank plasma and tissue homogenate. Then, the samples were vortexed for 3 min. Next, appropriate volumes of acetonitrile were added until each sample reached a final volume of 500 μl; these samples were centrifuged at 10,000 g using an SL eight Small Benchtop Centrifuge (Thermo Fisher Scientific) for 10 min at room temperature. Finally, the supernatant (40 μl) was injected into the HPLC/MS system. Standard solutions were prepared in triplicate for seven calibration points, and the final concentrations obtained were 1, 2, 5, 20, 50, 100, and 500 ng/ml.

Calibration curves were constructed in triplicate by plotting known concentrations of the standard versus the detector response area. In the same way, a separate set of stock solutions was prepared for QC samples to determine accuracy and precision.

The developed method was validated in accordance with the bioanalytical guidance from the FDA (FDA, Guidance for Industry Bioanalytical Method Validation, 2018), with determination of the following parameters: linearity; within-run and between-run accuracy and precision; recovery; LLOQ; limit of detection (LOD); selectivity; and stability.

For intravenous (IV) treatment, Indo5 was dissolved in DMSO (solution, 1 mg/ml) and injected into the lateral tail vein of rats (1 mg/kg, n = 3). For intragastric treatment, Indo5 was dissolved in 5% CMC-Na (suspension, 20 mg/ml). The rats were orally administered with Indo5 (100 mg/kg, n = 3) using a 20-G gavage needle. All of the animals were given a standard diet 4 h after the dosing. Blood samples (approximately 400 μl) were collected from each rat and placed into heparinized tubes before the dosing, and at 5, 15, 30, 60, 120, 240, and 480 min after the dosing. Plasma (100 μl) was harvested by centrifuging blood samples at 3,600 g for 10 min at room temperature followed by immediate processing for the analyses.

Twelve male Wistar rats were given a single dose of Indo5 (1 mg/kg, IV). At 2, 5, 30, 60, 120, 480, and 720 min after the dosing, a group of animals (n = 4 for each treatment time) was sacrificed. The liver, kidneys, heart, lungs, stomach, small intestine, small-intestine content, large intestine, large-intestine content, spleen, brain, and testes were dissected rapidly and harvested. All of the tissues were rinsed thoroughly in ice-cold physiologic saline to eliminate blood and other content. The tissues and contents were processed by homogenization with 0.9% saline in a 1:3 (w/v) ratio. The preparation process for analyses was the same as that described for plasma.

Freshly separated liver was rinsed in ice-cooled phosphate-buffered saline (PBS) and then placed on a 40-μm filter. We added an appropriate amount of RPMI-1640 medium containing 2% fetal bovine serum (FBS) and used the piston of a 5-ml syringe to squeeze the tissue blocks to the bottom of the screen until only fibrous tissue remained. The volume of cell suspension collected was 40 ml. We centrifuged the cell suspension at 50 g for 5 min at 4°C and collected the supernatant. Next, we centrifuged at 800 g for 10 min at 4°C and discarded the supernatant. The precipitate was resuspended with a solution containing 40% Percoll™ (2 ml of 90% Percoll with 2.5 ml RPMI-1640 medium). We carefully added the suspension, using a capillary pipette, to a solution containing 70% Percoll (4 ml of 90% Percoll with 1.12 ml RPMI-1640 medium). Next, we centrifuged at 800 g for 20 min at 4°C, removed the upper layer, took the middle layer, and added an appropriate amount of 2% FBS in RPMI-1640 medium to reach 15 ml. Then, we undertook centrifugation at 800 g for 10 min at 4°C, removed the supernatant, and transferred it to a fresh 1.5-ml Eppendorf™ tube. Next, it was centrifuged at 1800 g for 5 min at room temperature and dried with absorbent paper. We added 300 μl of red blood cell (RBC) lysate for resuspension, lysed the RBCs for 3 min at room temperature, added 1 ml of RPMI-1640 medium with 2% FBS to stop the lysis, and centrifuged at 1800 g for 5 min at room temperature. An appropriate amount of PBS containing 2% FBS was added to resuspend the cells.

The cell suspension was added to a 1.5-ml Eppendorf tube, incubated with antibody for 30 min at 4°C in the dark, washed twice with 2% PBS, resuspended with 400 μl of 2% PBS, and analyzed by flow cytometry. The antibodies used for labeling were as follows: cluster of differentiation (CD)3-PerCP-Cy5.5, NK1.1-FITC, B220-APC, CD11c-FITC, CD11b-APC, CD4-APC, CD8-PE, and CD69-PE/Cy7; they were obtained from eBioscience (San Diego, CA, United States) or BD Biosciences (Franklin Lakes, NJ, United States).

The tissues were fixed in 10% formalin and embedded in paraffin. Then, 4-μm-thick sections were stained with hematoxylin and eosin for histopathological and morphological analyses.

All of the data were expressed as mean ± standard deviation (SD) of at least three independent experiments. For comparisons between two groups, two-tailed Student’s t-tests were performed. p < 0.05 was considered statistically significant. The data were analyzed and graphed using GraphPad Prism 6.

Selectivity. Six blank samples of each matrix (plasma and tissues) were analyzed to evaluate the selectivity of our method. The samples were prepared in accordance with the method we developed. The blank samples were spiked with a standard solution of Indo5 and IS, and then analyzed; typical MRM chromatograms are shown in Figure 1. Our method demonstrated high selectivity given that there was no interference in the analyte peaks with those obtained using standards or the IS.

Linearity. Linearity was tested using calibration curves by spiking blank plasma and tissues at eight concentration levels. All of the calibration curves for Indo5 in plasma or tissue were linear over the concentration range 1–500 ng/ml with a correlation coefficient (r) > 0.99 (Table 1).

Accuracy and precision. Accuracy and precision were evaluated at LLOQ, as well as low, medium, and high levels, with five replicates, on three separate days. The accuracy of detecting Indo5 in plasma and tissues was 85.5–112.5%. Within-run and between-run precision was <15% (Table 2). These results all met the criteria set by the FDA.

Recovery and matrix-effect. Recovery was calculated by comparing the analytical results for the extracted samples with the corresponding extracts of blanks spiked with Indo5 after extraction. The recovery effect of Indo5 at three QC levels ranged from 87.3 to 95.7% in plasma, and from 85.6 to 105.2% in the tested tissue homogenates (Table 3). The extraction recovery of IS in plasma and tissues was between 95.9 and 103.4%, indicating that protein precipitation as the sample preparation resulted in high and reproducible extraction efficiencies. The matrix effect of Indo5 at three QC levels ranged from 91.2 to 98.4% in plasma, and from 90.7 to 106.1% in the tissue homogenates, and the matrix effect of the IS in plasma and the tissues was between 93.6 and 104.5%, which indicated that there was no significant ion suppression or enhancement in the LC-MS/MS method (data not shown).

After IV (1 mg/kg in DMSO) or oral (100 mg/kg in 0.5% CMC-Na) administration of Indo5 to rats in each treatment group (n = 3), the plasma concentrations of Indo5 were determined using LC-MS/MS. The data obtained from each group were averaged. Figures 2, 3 show the mean plasma concentration–time curves for Indo5 after IV and oral administration, respectively. The corresponding pharmacokinetic parameters, calculated using a non-compartmental model are summarized in Table 4.

After IV administration, the plasma concentrations of Indo5 were well above the LLOQ within 2 h but, at 4 h, Indo5 was not detected. For IV injection, T_max was 1 min, C_max was 1,565.3 ± 286.2 ng/ml, AUC_0–t was 158.6 ± 27.9 h⋅ng/ml, AUC_0–∞ was 163.4 ± 28.3 h⋅ng/ml, MRT_0–t was 0.18 ± 0.04 h, and MRT_0–∞ was 0.26 ± 0.06 h. T_1/2 was 0.63 ± 0.11 h, indicating that Indo5 had a short half-life in vivo. Moreover, Indo5 displayed a high systemic CL (6,121.6 ± 1,124.8 ml/h/kg) at the tested dose.

After oral administration, the plasma concentrations of Indo5 were above the LLOQ within 8 h. For oral administration, T_max was 2.0 ± 0.48 h, C_max was 54.7 ± 10.4 ng/ml, t_1/2 was 1.25 ± 0.24 h, AUC_0–t was 251.5 ± 43.5 h⋅ng/ml, AUC_0–∞ was 255.7 ± 44.9 h⋅ng/ml, MRT_0–t was 2.87 ± 0.61 h, MRT_0–∞ was 2.99 ± 0.63 h, and CL was 391056.4 ± 77,945.8 ml/h/kg. According to the value for AUC_0–∞ for oral administration and IV injection at 100 mg/kg, the bioavailability of Indo5 was 1.59%.

These results suggested that Indo5 was rapidly cleared from the plasma and had low bioavailability.

The concentrations of Indo5 after IV administration (1 mg/kg) in the liver, kidneys, lungs, heart, spleen, stomach, large intestine, small intestine, brain, and testes at the indicated times in rats are shown in Figure 4. Indo5 was distributed rapidly and widely in the tested tissues in the following order: liver > kidneys ≈ heart > lungs ≈ large intestine ≈ small intestine ≈ stomach > spleen > brain ≈ testes. Indo5 was eliminated quickly from the tested tissues, and at 8 h, it was not detectable anymore in the kidneys, heart, lungs, large intestine, small intestine, stomach, spleen, brain, or testes. Notably, the hepatic Indo5 concentration was higher than that in other tissues, and it was still detectable in the liver 8 h post administration.

To verify the liver enrichment of Indo5, we measured the Indo5 concentration in the tissues after oral administration. Consistent with the results following IV administration, the highest concentration of Indo5 was found in the liver, and it was detectable 8 h post administration (Figure 5).

Taken together, these results suggested that Indo5 displayed an accumulated distribution in the liver in vivo.

Given that Indo5 was most distributed in the liver and had significant antitumor activity in a xenograft model and hepatic orthotopic model in mice (Luo et al., 2019), we investigated if continuous administration of Indo5 leads to liver injury. Indo5 was orally administered (p.o.) once daily for 21 days, and we evaluated its toxicity. No obvious weight loss or other symptoms were observed with continuous injection of Indo5. The overall health of the animals was not adversely affected (data not shown). The levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) in serum did not change significantly after Indo5 administration (Figures 6A,B; p > 0.05). Histological analysis of liver tissue revealed no pathological changes in Indo5-treated mice compared with the solvent-control group, i.e., a normal lobular architecture with a central vein and radiating hepatic cords were observed (Figure 6C). Often, liver injury leads to infiltration or activation of immune cells in liver tissue. We thus investigated the number of liver mononuclear cells and frequencies of various immune cells, including T cells, natural killer cells, natural killer T cells, B cells, dendritic cells, and myeloid cells. The frequencies of T-cell subsets (CD4⁺, CD8⁺) and activated T cells (CD4⁺CD69⁺, CD8⁺CD69⁺) were also measured. Continuous administration of Indo5 for 21 days did not affect the number or frequencies of immune cells in the liver (Figure 7; p > 0.05).

These results suggested that Indo5 did not show obvious toxicity.