Influence of Shenxiong Glucose Injection on the Activities of Six CYP Isozymes and Metabolism of Warfarin in Rats Assessed Using Probe Cocktail and Pharmacokinetic Approaches

Shenxiong glucose injection (SGI), a traditional Chinese medicine (TCM) preparation, has been widely used for the treatment of various cardiovascular and cerebrovascular diseases for many years. We assessed the potential influences of SGI on the activities of six CYP enzymes (CYP1A2, CYP2C11, CYP2C19, CYP2D4, CYP2E1, and CYP3A2) and on the pharmacokinetics of warfarin in rats. We compared plasma pharmacokinetics of six probe drugs (caffeine/CYP1A2, tolbutamide/CYP2C11, omeprazole/CYP2C19, metoprolol/CYP2D4, chlorzoxazone/CYP2E1, and midazolam/CYP3A2) and of warfarin between control and SGI-pretreated groups, to estimate the effect on the relative activities of the six isozymes and warfarin metabolism. There were no significant differences in the pharmacokinetic parameters of caffeine, omeprazole, metoprolol, chlorzoxazone, and midazolam between the SGI-pretreated and control groups. However, many pharmacokinetic parameters of tolbutamide in SGI-pretreated rats were affected significantly (p < 0.05), and indicated tolbutamide metabolism in the former group was markedly slower. Moreover, SGI reduced the clearance of warfarin. These results suggested SGI showed no effects on the enzyme activities of rat CYP1A2, CYP2C19, CYP2D4, CYP2E1, and CYP3A2, but inhibited the enzyme activity of CYP2C11, and improved the blood concentration of warfarin. This suggests that the dose of warfarin may need be adjusted when co-administrated with SGI.


Introduction
Shenxiong glucose injection (SGI) is a preparation containing the water extracts of Salvia miltiorrhiza Bunge and ligustrazine hydrochloride. The injection is widely used in China for the treatment of cardiovascular and cerebrovascular diseases, such as angina pectoris [1,2], coronary heart disease [3],

Effect of SGI on CYP1A2 in Rats
The effects of SGI in the different treatment groups on pharmacokinetic parameters (AUC(0-t), AUC(0-∞), MRT(0-t), MRT(0-∞), t 1/2 , CL, and Vd) of caffeine in rats are presented in Table 1. Little changes in pharmacokinetic parameters of caffeine were observed in rats pretreated with either SGI or vehicle. This indicates that SGI had no influence on CYP1A2 activity in vivo.

Effect of SGI on CYP2C11 in Rats
CYP2C11 activity was evaluated by assessment of tolbutamide pharmacokinetic behavior ( Table 2). The mean plasma concentration-time curves of tolbutamide ( Figure 1) in the indicated study groups are presented. Pretreatment with SGI caused significant increases in AUC(0-t), AUC(0-∞), MRT(0-t), MRT(0-∞), and CL of tolbutamide, compared with those observed in the BCG. In addition, significantly increased t 1/2 was observed in 14D-G group pre-treated with SGI. Taken together, these results indicate that SGI may inhibit CYP2C11 activity in vivo. BCG, blank control group (0.9% sodium chloride solution for 10 days); 7D-G, short-period group (5 mL/kg of SGI concentrated solution, iv, once daily for consecutive 7 days); 14D-G, long-period group (5 mL/kg of SGI concentrated solution, iv, once daily for consecutive 14 days). Values are represented as mean ± SD (n = 6).  BCG, blank control group (0.9% sodium chloride solution for 10 days); 7D-G, short-period group (5 mL/kg of SGI concentrated solution, iv, once daily for consecutive 7 days); 14D-G, long-period group (5 mL/kg of SGI concentrated solution, iv, once daily for consecutive 14 days). Values are represented as mean ± SD (n = 6). BCG, blank control group (0.9% sodium chloride solution for 10 days); 7D-G, short-period group (5 mL/kg of SGI concentrated solution, iv, once daily for consecutive 7 days); 14D-G, long-period group (5 mL/kg of SGI concentrated solution, iv, once daily for consecutive 14 days). Values are represented as mean ± SD (n = 6). * p < 0.05 when compared with related parameters of model rats. ** p < 0.01 when compared with related parameters of model rats.  . BCG, blank control group (0.9% sodium chloride solution for 10 days); 7D-G, short-period group (5 mL/kg of SGI concentrated solution, i.v., once daily for consecutive 7 days); 14D-G, long-period group (5 mL/kg of SGI concentrated solution, i.v., once daily for consecutive 14 days). Error bars represent SD.

Effect of SGI on CYP2C19 in Rats
Pharmacokinetic parameters AUC(0-t), AUC(0-∞), MRT(0-t), MRT(0-∞), t1/2, CL, and Vd of omeprazole in rats are listed in Table 3. There was no significant difference in the pharmacokinetic parameters in the animals between the treatment with SGI and vehicle, indicating that SGI showed little impact on CYP2C19 activity in vivo.

Effect of SGI on CYP2D4 in Rats
Change in CYP2D4 activity induced by SGI was assessed by comparing the pharmacokinetics of metoprolol in rats pretreated with SGI or vehicle, as shown in Table 4. No significant difference in the pharmacokinetic parameters were observed in animals given either SGI or vehicle. These results imply that SGI revealed little influence on CYP2D4 activity in vivo. BCG, blank control group (0.9% sodium chloride solution for 10 days); 7D-G, short-period group (5 mL/kg of SGI concentrated solution, iv, once daily for consecutive 7 days); 14D-G, long-period group (5 mL/kg of SGI concentrated solution, iv, once daily for consecutive 14 days). Values are represented as mean ± SD (n = 6).

Effect of SGI on CYP2E1 in Rats
Pharmacokinetic profiles of chlorzoxazone after SGI treatment were used to describe the activity of CYP2E1. As shown in Table 5, pharmacokinetic parameters AUC(0-t), AUC(0-∞), MRT(0-t), MRT(0-∞), t1/2, CL, and Vd of chlorzoxazone in rats pretreated with SGI were found to be similar with those of control animals. This indicates that SGI showed no influence on CYP2E1 activity in vivo.

Effect of SGI on CYP3A2 in Rats
Alternation of CYP3A2 activity by SGI was evaluated by monitoring the pharmacokinetics of midazolam, as shown in Table 6. No significant change in pharmacokinetic behavior of midazolam was observed in rats after pretreatment with SGI. This suggests that SGI exhibited no effect on CYP3A2 activity in vivo.

Effect of SGI on the Pharmacokinetics of Warfarin
To determine the interaction of SGI with warfarin, time-course plasma warfarin was monitored in rats with and without pretreatment with SGI. The mean plasma concentration-time curves are presented in Figure 2, and the pharmacokinetic parameters are summarized in Table 7. Pretreatment with SGI for 7 days in rats increased the AUC(0-t), AUC(0-∞), and C max of warfarin by 86.0%, 87.3%, and 81.4% (all P < 0.01), along with decreased CL/F and V/F by 42.9% and 41.2% (both p < 0.01), respectively. As expected, more increased AUC(0-t), AUC(0-∞), MRT(0-∞), and C max (122.7%, 125.9%, 34.5%, and 83.2%) and more decreased CL/F and V/F (57.1% and 52.9%) were observed in rats pretreated with SGI for 14 days. BCG, blank control group (0.9% sodium chloride solution for 10 days); 7D-G, short-period group (5 mL/kg of SGI concentrated solution, iv, once daily for consecutive 7 days); 14D-G, long-period group (5 mL/kg of SGI concentrated solution, iv, once daily for consecutive 14 days). Values are represented as mean ± SD (n = 6). * p < 0.05 when compared with related parameters of model rats. ** p < 0.01 when compared with related parameters of model rats. 7D-G, short-period group (5 mL/kg of SGI concentrated solution, iv., once daily for consecutive 7 days); 14D-G, long-period group (5 mL/kg of SGI concentrated solution, iv, once daily for consecutive 14 days). Error bars represent SD.

Discussion
Cardiovascular and cerebrovascular diseases that increase the likelihood of thromboembolic events usually require a chronic use of warfarin. In China, warfarin is often used in combination with traditional Chinese medicine (TCM) preparations with function of activating circulation to remove blood stasis, antiplatelet drugs, and non-steroidal anti-inflammatory agents. Prescription analysis showed that the rate of concomitant use of warfarin and TCM, such as SGI, Danshen ligustrazine injection, and xueshuantong injection, was over 50% [33,34]. According scientific research studies, the prothrombin time was prolonged after combined with warfarin and Salvia miltiorrhiza Bunge [34]. Recently, there was a report about serious bleeding resulting from warfarin in a patient, which might be caused by co-administration of SGI [35]. This drew our attention to SGI-mediated pharmacokinetic behavior alteration of warfarin, especially the inhibitory effects of SGI on CYP enzymes.
In this study, we chose rats as the experimental animal. Although humans differ from rats with regard to isoform composition, expression, and catalytic activities of drug-metabolizing enzymes, rats are common animal models for metabolic behavior studies [36]. Human CYP2D6 is homolog to rat CYP2D4, human CYP3A4 is homolog to rat CYP3A2, human CYP2C9 is homolog to rat CYP2C11, human CYP1A2, CYP2C19, and CYP2E1 are homologs to rat CYP1A2, CYP2C19, and CYP2E1 respectively. We investigated the potential influence of SGI on the enzyme activities of CYPs1A2, 2C11, 2C19, 2D4, 2E1, and 3A2 in rats, by examining the pharmacokinetic behaviors of their probe substrates-caffeine, tolbutamide, omeprazole, metoprolol, chlorzoxazone, and midazolam, respectively.
Water-soluble extracts of Salvia miltiorrhiza were reported to scavenge peroxides and to inhibit the expression of adhesion molecules in vascular endothelium and leukocytes [37]. Recent studies demonstrated that salvianolic acid B and tanshinol, found in water Salvia miltiorrhiza extracts showed weak inhibitory effects on CYP enzymes [38][39][40]. However, tanshinones, lipophilic components of Salvia miltiorrhiza, revealed considerable inhibition towards CYP1A2, 2C9 and 3A4 [41,42]. SGI is composed of the water extracts of Salvia miltiorrhiza and tetramethylpyrazine hydrochloride. The water extracts of Salvia miltiorrhiza Figure 2. The mean plasma concentration-time curves of warfarin (2.0 mg/kg, i.g.) in the different groups (n = 6). BCG, blank control group (0.9% sodium chloride solution for 10 days); 7D-G, short-period group (5 mL/kg of SGI concentrated solution, iv., once daily for consecutive 7 days); 14D-G, long-period group (5 mL/kg of SGI concentrated solution, iv, once daily for consecutive 14 days). Error bars represent SD.

Discussion
Cardiovascular and cerebrovascular diseases that increase the likelihood of thromboembolic events usually require a chronic use of warfarin. In China, warfarin is often used in combination with traditional Chinese medicine (TCM) preparations with function of activating circulation to remove blood stasis, antiplatelet drugs, and non-steroidal anti-inflammatory agents. Prescription analysis showed that the rate of concomitant use of warfarin and TCM, such as SGI, Danshen ligustrazine injection, and xueshuantong injection, was over 50% [33,34]. According scientific research studies, the prothrombin time was prolonged after combined with warfarin and Salvia miltiorrhiza Bunge [34]. Recently, there was a report about serious bleeding resulting from warfarin in a patient, which might be caused by co-administration of SGI [35]. This drew our attention to SGI-mediated pharmacokinetic behavior alteration of warfarin, especially the inhibitory effects of SGI on CYP enzymes.
In this study, we chose rats as the experimental animal. Although humans differ from rats with regard to isoform composition, expression, and catalytic activities of drug-metabolizing enzymes, rats are common animal models for metabolic behavior studies [36]. Human CYP2D6 is homolog to rat CYP2D4, human CYP3A4 is homolog to rat CYP3A2, human CYP2C9 is homolog to rat CYP2C11, human CYP1A2, CYP2C19, and CYP2E1 are homologs to rat CYP1A2, CYP2C19, and CYP2E1 respectively. We investigated the potential influence of SGI on the enzyme activities of CYPs1A2, 2C11, 2C19, 2D4, 2E1, and 3A2 in rats, by examining the pharmacokinetic behaviors of their probe substrates-caffeine, tolbutamide, omeprazole, metoprolol, chlorzoxazone, and midazolam, respectively.
Water-soluble extracts of Salvia miltiorrhiza were reported to scavenge peroxides and to inhibit the expression of adhesion molecules in vascular endothelium and leukocytes [37]. Recent studies demonstrated that salvianolic acid B and tanshinol, found in water Salvia miltiorrhiza extracts showed weak inhibitory effects on CYP enzymes [38][39][40]. However, tanshinones, lipophilic components of Salvia miltiorrhiza, revealed considerable inhibition towards CYP1A2, 2C9 and 3A4 [41,42]. SGI is composed of the water extracts of Salvia miltiorrhiza and tetramethylpyrazine hydrochloride. The water extracts of Salvia miltiorrhiza contain tanshinol, protocatechuic aldehyde, rosmarinic acid, salvianolic acid B, and salvianolic acid A as major components, specifically, 100 mL of SGI contains 20 mg tanshinol and 100 mg tetramethylpyrazine hydrochloride [43]. No information is available for the effect of tetramethylpyrazine hydrochloride on CYP450 activity in vivo. Whether the observed decrease in CYP2C11 activity by SGI resulted from tetramethylpyrazine hydrochloride, which SGI contained, needs further investigation.
Warfarin (R-and S-) is metabolized to 4-hydroxy-and 10-hydroxy-warfarin via CYP3A4; to 6-hydroxy-and 8-hydroxy-warfarin by CYP1A2; to 6-hydroxy-and 7-hydroxywarfarin, through CYP2C9 (in humans) or CYP2C11 (in rats) [30]. The present study demonstrated that co-administration of SGI caused inhibition of CYP2C11, and not of the other CYPs evaluated. It is likely that the observed enhancement of blood concentration of warfarin in rats co-treated with SGI may arise from the inhibition of CYP2C11 activity. Therefore, co-administration of SGI and warfarin may result in herb-drug interactions, which may be an important factor which leads to serious adverse reactions. Further study is needed to evaluate the pharmacodynamic changes and toxicity of warfarin induced by co-administration of SGI.
Furthermore, there is an intrinsic factor to affect warfarin pharmacokinetics (PK) and pharmacodynamics (PD), which is mainly the genetic polymorphisms influencing the expression of CYP2C9 and vitamin K epoxide reductase complex subunit 1 (VKORC1), respectively [44]. Importantly, CYP2C9 and VKORC1 are significantly different among Asians, Europeans, and Africans [45,46], and the Chinese populations are more sensitive to warfarin than Western populations [46,47]. For the genetic polymorphisms, the warfarin label was updated in 2010, and gave a recommendation for initial dosing ranges for patients with different combinations of CYP2C9 and VKORC1 genotypes [48]. Regular monitoring of INR (International Normalized Radio) should be performed on all treated patients.

Animals and Treatment
Male Sprague Dawley rats (200 ± 20 g) were purchased from the Laboratory Animal Center of Guizhou Medical University. All animal maintenance and experimental studies were based on the guidelines of the National Institutes of Health for the Care and Use of Animals, and were approved by the Experiment Animal Center of Guizhou Medical University. All rats were maintained in a room under controlled temperature and humidity and fed standard laboratory food and water.
The experimental animals were divided into CYP probe group and warfarin group. The herb/drug administration was designed as follows: the CYP probe group comprised three subgroups (n = 6 per subgroup), viz., the blank control group (BCG), short-period group (7D-G), and long-period group (14D-G). BCG received 0.9% sodium chloride solution for 10 days. The other two groups were administered with concentrated SGI solution at a dose of 5.0 mL/kg via caudal vein daily for 7 or 14 consecutive days, respectively. Warfarin group also comprised three sub-groups: the blank control group (BCG), short-period group (7D-G), and long-period group (14D-G). BCG received 0.9% sodium chloride solution for 10 days. The other two groups were treated with concentrated SGI solution at a dose of 5.0 mL/kg via caudal vein daily for 7 or 14 days. On either the 8th or 15th day, the animals of the CYP probe group were treated with probe cocktail solution (5.0 mL/kg) containing caffeine (1.0 mg/kg), tolbutamide (1.0 mg/kg), omeprazole (2.0 mg/kg), metoprolol (10 mg/kg), chlorzoxazone (4.0 mg/kg), or midazolam (4.0 mg/kg), via caudal vein. Similarly, on either the 8th or 15th day, the rats of the warfarin groups were given warfarin (2.0 mg/kg, i.g.) suspended in CMC-Na solution (5.0 g/L).
The doses selected for the animal study were based on the clinical dose of SGI, which is 100-200 mL as a single daily dose. Thus, the doses used for rats needed to be 10.8-21.7 mL/kg body weight, as calculated for conversion between species. However, the required injection volume would then be in excess of the regular injection volume for rats (10 mL/kg body weight). Therefore, a concentrated form of SGI was used (equivalent to 80 mg per 100 mL of salvianic acid and 400 mg per 100 mL of ligustrazine hydrochloride); this concentration was four times higher than that of the original SGI solution (equivalent to 20 mg per 100 mL of salvianic acid and 100 mg per 100 mL of ligustrazine hydrochloride). To mimic the clinical use of SGI, the effects of SGI on rats CYP enzymes were evaluated after the rats were intravenously administered with SGI for 7 (short period) or 14 consecutive days (long period).
All samples (50 µL) spiked with internal standard puerarin (final concentration: 4.0 µg/mL) were mixed with 400 µL of methanol. The resulting mixture was vortexed for 5 min and centrifuged at 20,000 × g for 10 min. The supernatants were transferred to a fresh tube and concentrated under nitrogen at 45 • C. The residues were reconstituted with 400 µL of the initial mobile phase for UPLC, followed by vortexing for 1 min and centrifugation at 20,000 × g for 5 min.
To 50 µL of rat plasma samples, IS 2 naproxen (50 µL, 20.0 µg/mL) and dilute hydrochloric acid (50 µL, 1 M) were added, prior to mixing with 350 µL of methanol. The rest of sample preparation protocols were similar as described for that of CYP probe-drug groups.

CYP Probe Groups
UPLC-MS analyses were carried out on a system consisting of a Waters ACQUITYTM UPLC (Waters, Milford, MA, USA) coupled to a Waters ACQUITY triple quadrupole mass spectrometer equipped with a Z-spray ESI source, which was operated in either positive-ion or negative-ion mode detection. Nitrogen was used as a nebulizing gas with a source temperature of 120 • C. Desolvation gas (nitrogen) was heated to 350 • C, and delivered at a flow rate of 650 L/h. Selected or single-ion recording (SIR) was chosen for quantification of the probe substrates ( Table 8). The operation of the UPLC-MS and data analysis was achieved using a MassLynxTM V4.1 workstation (Micromass, Manchester, UK).
Liquid chromatography analyses were performed in a gradient elution mode on Waters BEH C18 column (2.1 mm × 50 mm, 1. rate of 0.35 mL/min was run at 10-65% A over 0-3.0 min, 65-90% A over 3.0-3.5 min, 90-10% A over 3.5-4.0 min, and 10% A over 4.0-4.5 min. The samples were maintained at 4 • C CV in the auto-sampler, and a volume of 1 µL was injected into the UPLC system.

Warfarin
The mass spectrometer was operated in positive ion mode. The cone voltage for warfarin and naproxen (IS 2) was set at 40 and 30 V, along with the collision energy at 25 and 15 V for warfarin and naproxen, respectively. Argon was used as collision gas at a flow rate of 0.16 mL/min. Multiple reaction monitoring (MRM) was employed to perform mass spectrometric quantification. The MRM analysis was conducted by monitoring the precursor ion to product ion transitions of m/z 307.0→164.0 for warfarin and m/z 230.9→185.0 for naproxen.
All other analysis conditions were the same as described for that of CYP probe-drug groups.

Statistical Analysis
Data were presented as mean ± SD. Pharmacokinetic parameter calculations were carried out using the DAS 2.0 pharmacokinetic program (Chinese Pharmacological Society, Beijing China), and generated using a non-compartmental model (statistical moment). Statistically significant differences in the pharmacokinetic parameters between the treatment groups and the blank control group were assessed using one-way analysis of variance followed by Dunnett's test, with the level of statistical significance set at 0.05.

Conclusions
SGI demonstrated inhibitory effect on CYP2C11 in rats. Co-administration of SGI induced elevation in AUC and C max of warfarin, possibly resulting from inhibition of CYP2C11 activity. This suggests that it is potentially risky when SGI is co-administered with warfarin. Dose adjustment of warfarin may be required when co-administration with SGI is necessary.

SGI
Shenxiong glucose injection TCM the traditional Chinese medicine AUC the area under the curves MRT mean retention time Cmax maximum concentrations t 1/2 elimination half-lives CL the clearance V apparent volume of distribution BCG blank control group (0.9% sodium chloride solution for 10 days) 7D-G short-period group (5 mL/kg of SGI concentrated solution, i.v., once daily for consecutive 7 days) 14D-G long-period group (5 mL/kg of SGI concentrated solution, i.v., once daily for consecutive 14 days)