Cholic acid (CA), glycocholic acid (GCA), taurocholic acid (TCA), chenodeoxycholic acid (CDCA), glycochenodeoxycholic acid (GCDCA), taurochenodeoxycholic acid (TCDCA), ursodeoxycholic acid (UDCA), glycoursodeoxycholic acid (GUDCA), tauroursodeoxycholic acid (TUDCA), deoxycholic acid (DCA), glycodeoxycholic acid (GDCA), taurodeoxycholic acid (TDCA), taurolithocholic acid (TLCA), and cortisone acetate were purchased from Sigma-Aldrich (St. Louis, MO, USA). α-Muricholic acid (α-MCA) and tauromuricholic acid (TMCA) were purchased from Steraloids Inc (Newport, RI, USA). All other reagents and solvents were of high performance liquid chromatography (HPLC) grade. Deionized water was purified using a Milli-Q system (Millipore, Bedford, MA, USA). S. baicalensis Georgi (Hebei Province), G. uralensis Fisch (Inner Mongolia of China), P. lactiflora Pall (Anhui Province), and Z. jujuba Mill (Henan Province) were authenticated by Dr. Ehu Liu (State Key Laboratory of Natural Medicine, China Pharmaceutical University, China).

CPT-11 (Jaripharm Inc, Lianyungang, China) was obtained as irinotecan hydrochloride. It was prepared, according to manufacturer’s (Camptosar^TM, Pfizer Inc., USA) directions and the published papers, to duplicate the clinical formulation: 20 mg/mL CPT-11, 45 mg/mL sorbitol (Sigma, St Louis, MO, USA), and 0.9 mg/mL lactic acid (Sigma, St Louis, MO, USA) dissolved in deionized water and subsequently heated to 70–90°C until the solution turned to clear. The pH was adjusted to 3.5 using 1 M NaOH. The resulting solution was sterile-filtered and stored protected from light until the moment of administration (Hardman et al., 1999; Trifan et al., 2002; Yang et al., 2006).

HQD was prepared by decocting S. baicalensis Georgi, G. uralensis Fisch, P. lactiflora Pall, and Z. jujuba Mill at ratio of 3:2:2:2 by dry weight. The filtrate was collected then concentrated as a decoction of 1 g/mL crude drug, stored in the refrigerator before intragastric administration to rats.

Male Sprague-Dawley rats (Sino-British SIPPR/BK Lab Animal Ltd., Shanghai, China) initially weighing 200 ± 20 g were used in present study. Animals were housed in a routine laboratory conditions with a temperature-controlled (25°C) environment and a 12 h dark/light cycle throughout the acclimatization and experimental periods. The experimental protocols were in accordance to the Guide for the Care and Use of Laboratory Animals and approved by the Animal Ethics Committee of China Pharmaceutical University (License Number: SYXK 2012-0035). After 1 week adaptation, rats were randomly divided into three groups either received vehicle alone, CPT-11 alone, or CPT-11 and HQD. The toxicity model was constructed by treatment of animals with CPT-11 at 150 mg/kg by intravenous injection (i.v.) once a day for two consecutive days (Trifan et al., 2002). Animals in control group injected with the same dose of vehicle. HQD was given orally twice a day from 1 day before the start of CPT-11 injection (day 1) for six consecutive days at a dose of 10 g/kg. While animals in control and CPT-11 group received equivalent Milli-Q water. All the gavage were 30 min prior to the dosing of CPT-11 on days 1 and 2. The mortality, weight loss, and food intake of animals were monitored daily. Scoring of diarrhea was conducted twice daily (8 a.m. and 8 p.m.). The severity of delayed diarrhea was scored as described by Kurita et al. (2000).

Diarrhea observed immediately (occur within 2 h) after the dosing of CPT-11 and approximately 48 h after the final administration were considered to be acute diarrhea and late-onset diarrhea, respectively. Serum and tissues (liver, jejunum, ileum, cecum, colon, and rectum) were collected on day 5, the most severe day of late-onset diarrhea. After coagulation (1 h) and centrifugation (8,000 rpm for 10 min), serum samples were aliquoted and then stored at -80°C until needed for preparation and analysis. Some tissue samples were used for histological examination and others were used for quantitative analysis. The scheme of the whole experiment is shown in Figure 1.

Tissue sections (liver, jejunum, ileum, cecum, and colon) were dissected on the most severe day of late-onset diarrhea (day 5), fixed in 10% neutral formalin and stained with hematoxylin–eosin. The routine histopathology assessment was performed in a blinded manner.

Metabolomic profiling of different classes of BAs in biological samples was assayed by a LC-MS/MS-validated method. Serum samples were thawed at room temperature. A total of 200 μL methanol and 10 μL cortisone acetate [internal standard (IS), 100 μg/mL] were added to 50 μL serum and the mixture was vortexed for 5 min. After centrifuged twice at 16,000 rpm (4°C) for 10 min, the supernatant was used for LC-MS/MS analysis. Approximately 100–150 mg tissue samples, including liver and intestines, were homogenized in 10 volumes of methanol. After 10 min of centrifugation (16,000 rpm, 4°C), 50 μL supernatant was spiked with 10 μL IS and 150 μL methanol, vortexed, centrifuged, and then used for further analysis. The separation was finally carried out on a ZORBAX Eclipse XDB-C18 (2.1 × 150 mm, 3.5 μm) (Agilent Technologies). The mobile phase consisted of (A) acetonitrile and (B) 0.1% formic acid. The gradient elution started with 25% A for 20 min, linearly increased to 60% B in 65 min and brought back to 25% A in 5 min followed by 10 min of re-equilibration. The mass spectrometry (MS) analysis was performed in a triple quadruple TSQ Quantum mass spectrometer via electrospray ionization (ESI) interface (Thermo Fisher, Palo Alto, CA, USA). The parameters were as follows: ESI (-), spray voltage, 3.8 kV; capillary temperature, 380°C; scan width for multiple-reaction monitoring (MRM), 0.1 m/z. Nitrogen was used as sheath (40 arb) and auxiliary (25 arb) gas. Argon was used as the collision gas (1.0 mTorr). BAs with no reference (β-MCA and ω-MCA) were semi-quantified by using calibration curve of homologous standards (α-MCA). The main parameters for MS/MS detection of BAs as well as the internal standard cortisone acetate are summarized in Table 1.

The method used in this study was developed in terms of linearity, accuracy, precision, extraction recovery, matrix effects, and stability following the compliance criteria described by Food and Drug Administration (FDA) guidance for Bioanalytical Method Validation. Charcoal-treated serum and tissues homogenates were served as biological matrices for the preparation of standards and quality control (QC) samples. Calibration curves were constructed by plotting the peak–area ratio of each BA to IS versus the concentrations of the BA in each corresponding matrix. Intraday and interday accuracy and precision were determined using 3QC concentrations distributed throughout the calibration range for each analyte. Extraction recoveries only took into account the efficiency of the extraction procedure. Matrix effects were determined by adding the analytes to each blank matrix after extraction and compared with the same amount of analytes in pure solvents. The freeze–thaw, short-term, long-term, and post-preparative stability of the BAs were also investigated in current method.

All the results are present as mean ± standard deviation (SD). The multivariate data analysis, principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA), were conducted by SIMCA-P (version 13.0, Umetrics, Sweden). Statistical evaluations were performed using non-parametric Mann–Whitney U-test (SPSS Inc., Chicago, USA). Differences were considered the results with P-values <0.05.

A fully validated method is appropriate for all the determinations. The accuracy ranges from 86.85 to 111.81% in all tested samples. Precision, measured as coefficient of variation (CV), was below 14.37 and 13.30% for intraday and interday for all BAs, respectively. In terms of linearity, the regression coefficients for all the calibration curves of BAs were higher than 0.99. Extraction recoveries at all QC concentrations were >83.42% in serum, >71.79% in liver, and >78.46% in intestine. No significant matrix effects for BAs were observed. In addition, stability studies indicated that BAs were stable in serum and tissues for at least 8 h at room temperature, at least three freeze–thaw cycles, in the -80°C for 2 weeks, and also in the reconstitution solution at 16°C for 12 h.

As anticipated, marked intestinal pathological changes were observed in rats received CPT-11 alone, including epithelial degeneration, villus shortening, gland atrophy, inflammatory cellular infiltration, and submucosal edema. The histologic lesions were extremely severe in ileum, followed by cecum, and minimal in jejunum and colon. Conversely, liver histology appeared normal. These intestinal damages were significantly alleviated in rats co-treated with HQD (Figure 2). Histological sections verified that the CPT-11-induced toxicity model was successfully produced in present study and co-treatment with HQD effectively ameliorated the intestinal damages induced by CPT-11.

Rats repeated dosing CPT-11 (150 mg/kg/day i.v.) for two consecutive days appeared serious anorexia and rapid decrease in body weight, reached 85.57 ± 2.20% of initial body weight by day 5. However, rats co-administered with HQD significantly attenuated body weight loss (91.49 ± 2.54% by day 5) and had a higher food intake level as compared with rats only received CPT-11. In addition, severe acute (occurred within 2 h after CPT-11 injection on days 1–2) and late-onset (chronically present during days 4–5) diarrhea were observed in rats treated with CPT-11 alone. Rats co-administered with HQD significantly improved the late-onset diarrhea induced by CPT-11, but failed to prevent the acute diarrhea on days 1 and 2. As expected, rats received the vehicle alone did not show any significant signs of toxicity. None of the rats died during the experiment (Table 2).

To targeted assay of BAs profiles throughout the enterohepatic system, concentrations of 17 individual BAs, including seven free BAs, six taurine-conjugated BAs, and four glycine-conjugated BAs were simultaneously determined and quantified in biological samples using a sensitive high-throughput LC-MS/MS (Supplementary Figure 1). A previously uncharacterized compartment-specific dysregulation feature of BAs was identified in different groups of rats. CPT-11-treated animals showed very different profiles from controls, with a marked reduction of CA, TCA, α-MCA, TMCA, UDCA, and TUDCA in serum. Levels of TCA and TUDCA were also decreased in liver and certain intestinal segments. Additionally, β-MCA was increased in liver and colon, while its conjugate—TMCA, was decreased in ileum, colon, and rectum. Notably, the levels of hydrophobic BAs—CDCA, GCDCA, TCDCA, especially DCA, GDCA, TDCA, and TLCA were significantly increased in almost all of the tested tissues. Contrary to what we expected, the most obvious disturbance of BAs homeostasis was shown in serum rather than in target organs, such as liver and intestine. This may be because circulating BAs comprehensively reflected the hepatic input and intestinal reabsorption (Supplementary Table 1). Meanwhile, as shown in PCA and PLS-DA score plots, the CPT-11-treated group serum and tissues could be clearly distinguishable from the control group (Figure 3 and Supplementary Figure 2).

Compared with only CPT-11-treated animals, HQD co-treatment could partly restore the homeostasis of BAs. Although there were some overlaps among three groups in PCA score plots (Supplementary Figure 2), the control group, CPT-11-treated group, and HQD co-treated group were separately clustered and could be clearly distinguished in rat serum and tissues based on PLS-DA model (Figure 3). The quantitative results showed that HQD mainly regulated the hydrophobic BAs (Figure 4). Hydrophobicity is an important determinant of the biological activity exhibited by BAs (Powell et al., 2001). Although the levels of cytoprotective UDCA and its conjugates remained unchanged, that of cytotoxic BAs, including CDCA, GCDCA, TCDCA, DCA, GDCA, TDCA, and TLCA, were markedly decreased in multiple compartments of rats received HQD. In fact, both CPT-11-treated and HQD co-treated rats had a slightly altered BAs pool sizes in serum and tissues. However, there was no statistically significant difference in BAs pool size between CPT-11 group and HQD group (Supplementary Figure 3). These data may reflect the fact that HQD was more likely to shift the balance of BAs rather than influence the BAs pool size. Meanwhile, it also suggested that, in addition to total BAs concentration, the composition of BAs is more important regarding BAs-related functions in the body. The profound remodeling of the BAs composition may be one of the mechanisms of HQD to reduce the toxicity of CPT-11.