SHORT COMMUNICATION

Herb-drug interaction between irinotecanand psoralidin-containing herbs

Xi-Shan Zhang • Zhi-Qiang Zhao • Zhen-Sheng Qin •

Kun Wu • Tian-Fang Xia • Li-Qun Pang

Received: 30 May 2014 / Accepted: 18 August 2014

� Springer International Publishing Switzerland 2014

Abstract Herb-drug interaction strongly limits the clini-

cal utilization of herbs and drugs. Irinotecan-induced

diarrhea is closely related with the UDP-glucuronosyl-

transferase 1A1-catalyzed glucuronidation of SN-38 which

has been widely regarded to be the toxic substance basis of

irinotecan. The present study aims to determine the influ-

ence of herbal component psoralidin toward the toxicity of

irinotecan. In vitro inhibition potential of psoralidin toward

the glucuronidation of SN-38 was firstly investigated using

human intestinal microsomes incubation system. Dose-

dependent inhibition of psoralidin toward SN-38 glucu-

ronidation was observed. Furthermore, Dixon plot showed

that the intersection point was located in the second

quadrant, indicating the competitive inhibition of psorali-

din toward the glucuronidation of SN-38. Through the data

fitting using competitive inhibition fitting equation, the

inhibition kinetic parameter (Ki) was calculated to be

5.8 lM. The translation of these in vitro data into the

in vivo situation showed that pre-treatment with psoralidin

significantly increased the toxicity of irinotecan, as indi-

cated by the increased body weight loss and more severe

colon histology damage. All these data indicated the herb-

drug interaction between irinotecan and psoralidin-con-

taining herbs.

Keywords Irinotecan � Psoralidin � Herb-drug

interaction � UDP-glucuronosyltransferase (UGT) 1A1

1 Introduction

Drug–drug interaction has become the most important

reason to limit the utilization of clinical drugs, and the

R&D of new chemical entities as drug candidates (Egan

et al. 2014). The herb-drug interaction due to the intro-

duction of herb medicine into the western medicinal system

further complicated the drug–drug interaction. Many fac-

tors contributed to the drug–drug interaction, and the

influence of activity of drug-metabolizing enzymes has

been widely accepted as the most important reason. The

components in herbs have been demonstrated to exhibit

inhibition toward multiple drug-metabolizing enzymes,

including phase I and phase II enzymes. For example, the

ginsenosides components have been reported to strongly

inhibit the activity of cytochrome P450 (CYP) and UDP-

glucuronosyltransferases (UGTs) (Liu et al. 2006; Fang

et al. 2013). Danshen components cryptotanshinone and

dihydrotanshinone I have been reported to exhibit strong

inhibition toward the glucuronidation of propofol (Cong

et al. 2013).

The utilization of anti-tumor drugs can always induce

various adverse effects. For example, the adverse effects of

first-line anti-tumor drug paxlitaxel contain nausea, vom-

iting, and loss of appetite. Irinotecan, also called as CPT-

11, is the first-line anti-colon cancer drug sold by Prizer

under the trade name Camptosar (Liang et al. 2014).

Diarrhea is the major adverse effect of irinotecan, and

X.-S. Zhang, Z.-Q. Zhao equally contributed to this work.

X.-S. Zhang

Department of General Surgery, Lian’shui County People’s

Hospital, Lian’shui, Jiangsu 223400, China

Z.-Q. Zhao

Department of Neurology, Huai’an Second People’s Hospital,

Huai’an, Jiangsu 223002, China

Z.-S. Qin � K. Wu � T.-F. Xia � L.-Q. Pang (&)

Department of General Surgery, Huai’an First People’s Hospital,

Nanjing Medical University, Huai’an, Jiangsu 223300, China

e-mail: hetangyuese132@163.com

Eur J Drug Metab Pharmacokinet

DOI 10.1007/s13318-014-0223-8

strongly limits the clinical utilization of irinotecan (Saliba

et al. 1998). Irinotecan’s diarrhea is closely related with the

intestinal UGT1A1 activity (Chen et al. 2013).

The present study aims to predict herb-drug interaction

between irinotecan and psoralidin using in vitro human

intestinal microsomal incubation system and in vivo mice

model.

2 Materials and methods

2.1 Description of reaction mixture

The used materials for the reaction are as follows: Tris–

HCl, MgCl2, alamethicin from Trichoderma viride,7-ethyl-

10-hydroxycamptothecin (SN-38) (purity C98 %), and

uridine-50-diphosphoglucuronic acid (UDPGA) (trisodium

salt). All these materials were purchased from Sigma-

Aldrich (St Louis, MO, USA). Pooled human intestinal

microsomes (HIMs) were purchased from BD Gentest

(Woburn, MA). Psoralidin was purchased from Weikeqi

Biotechnology Co. Ltd. (Sichuan, China). All other

reagents were of HPLC grade or of the highest grade

commercially available. The glucuronide of SN-38 was

generated using the following reaction system. In brief, the

mixture contained 5 mM MgCl2, 10 mM UDPGA, 4 lM

of SN-38, 50 lg/mL alamethicin, and 0.05 mg/mL of

HIMs in the Tris–HCl buffer (50 mM, pH = 7.4). After

60 min incubation, the reaction was terminated using

100 lM of 4-methylumbelliferone-b-D-glucuronide. Cen-

trifuged at 14,0009g for 30 min, 5 lL of supernatant was

used for LC/MS/MS analysis (Yu et al. 2014). A Shim-

pack XR-ODS column (100 mm, 2.0 mm, 2.2 lm, Shi-

madzu) was kept at 37 �C, and used for separation of

parent compound and metabolites. The mobile phase con-

tained acetonitrile (A) and H2O containing 0.2 % acetic

acid (B). The gradient conditions were as follows:

0–2 min, 95–83 % B; 2–7 min, 83–76 %; 7–9.5 min, 10 %

B; 9.5–12.5 min, 95 % B. The flow rate was 0.3 mL/min.

The MS conditions were as follows: voltage, 4 kV; inter-

face voltage, 40 V; nebulizing gas, 1.5 L/min; drying gas,

0.06 MPa. The select ion monitoring in the negative mode

was employed to determine SN-38 ([M-H]- = 391), SN-

38G ([M-H]- = 567), and internal standard ([M-

H]- = 351).

2.2 Inhibition kinetic analysis

The reaction velocity was determined in different con-

centrations of SN-38 and psoralidin, and calculated as the

formed glucuronide per reaction time per protein concen-

trations. The determination of inhibition type and

parameters was performed according to the previous liter-

atures (Xing and Che 2012; Yu et al. 2012).

2.3 Animal treatment

All animals care and experimental procedures complied

with the protocols approved by the Lian’shui County

People’s Hospital. The mice were maintained under a

standard 12 h light/12 h dark cycle with water and chow

provided ad libitum. A total of 15 male Sv/129 mice (6-

to 8-weeks age, 20–25 g body weight) were used in the

present study, including control group (n = 5), irinotecan

group (n = 5), and irinotecan ? psoralidin group

(n = 5). The diarrhea model was created through intra-

peritoneal (i.p.) injection of 50 mg/kg irinotecan for

8 days. In the irinotecan ? psoralidin group, 500 mg/kg

of psoralidin was given 7 days before the administration

of irinotecan, and continued to the end of the experi-

ment. The body weight was measured, and colon tissues

were taken for histology analysis after the sacrifice of

the mice.

2.4 Statistical method

All the data were given as mean plus SD, and the statistical

difference was compared using two-tailed student t test.

3 Results

The representative chromatography was given in Fig. 1a.

SN-38G, I.S., and SN-38 were eluted at 1.4, 1.8, and

2.5 min, respectively. The dose-dependent inhibition

behavior was observed for the inhibition of psoralidin

toward the glucuronidation of SN-38, with the residual

activity of 83.5, 77.0, 69.1, 60.3, 53.5, 43.3, 33.4, and

18.3 % at 1, 5, 10, 20, 40, 60, 80, and 100 lM of psoral-

idin, respectively (Fig. 1b). The inhibition kinetic type was

determined using Dixon plot which was drawn using

1/reaction velocity versus the concentrations of psoralidin.

As shown in Fig. 1c, the intersection point in Dixon plot

was located in the second quadrant, indicating the com-

petitive inhibition of psoralidin toward the glucuronidation

of SN-38. Through the data fitting using competitive

inhibition fitting equation, the inhibition kinetic parameter

(Ki) was calculated to be 5.8 lM. Treatment with irino-

tecan (50 mg/kg, i.p.) significantly decreased the body

weight (p \ 0.01), and co-administration with psoralidin

significantly increased the irinotecan-induced body weight

loss (p \ 0.05) (Fig. 2a). The pretreatment with psoralidin

can increase the tissue damage of colon shown in histology

results (Fig. 2b).

Eur J Drug Metab Pharmacokinet

Fig. 1 The in vitro inhibition

evaluation of psoralidin toward

the glucuronidation of SN-38.

a The representative UFLC-MS

chromatography for the

separation of SN-38G, internal

standard, and SN-38G.

b Concentration-dependent

inhibition of psoralidin toward

human intestinal microsomes

(HIMs)-catalyzed

glucuronidation of SN-38. Each

data point represents the mean

value of duplicate experiments.

c Determination of inhibition

kinetic type using Dixon plot.

Each data point represents the

mean value of duplicate

experiments. Dixon plot was

drawn using the 1/reaction

velocity versus the

concentrations of psoralidin.

The regression coefficients were

0.91, 0.95, 0.97, and 0.99 for the

fitting line of 2, 4, 6, 8 lM of

SN-38

Fig. 2 Pre-treatment with psoralidin increased the toxicity of irino-

tecan. a Pre-treatment with psoralidin increased the loss of body

weight induced by irinotecan treatment. * p \ 0.05; ** p \ 0.01;

*** p \ 0.001. b Histology analysis of colon damage in control,

irinotecan, and irinotecan ? psoralidin groups

Eur J Drug Metab Pharmacokinet

4 Discussion

The anti-tumor drugs can often co-administered with other

drugs, including other kinds of anti-tumor drug and antico-

agulant drugs. Therefore, the drug–drug interaction related

with anti-tumor drugs is frequent. Because cytochrome

P450s (CYPs) are the most important drug-metabolizing

enzymes (DMEs) involved in the metabolism of clinical

drugs (Lawrence et al. 2014), the inhibition of CYPs’ activity

is also used to explain the drug–drug interaction. For

example, the interaction between anti-tumor drug noscapine

and anticoagulant drug warfarin can be well explained by the

inhibition of noscapine toward the activity of CYP3A4 and

CYP2C9 which are the DMEs mainly involved in the

metabolism of warfarin (Fang et al. 2010).

More and more recent studies have reported the

importance of UDP-glucuronosyltransferases (UGTs) in

the metabolism of clinical drugs and endogenous sub-

stances, and the inhibition of UGTs provide a new insight

into the explanation of drug–drug interaction and patho-

genesis of diseases. For example, the experiment per-

formed by Mutlib et al. (2006) showed that UGT1A1, 1A6,

1A9, and 2B15 significantly contributed to the glucuroni-

dation of acetaminophen, and the inhibition of these

enzymes’ activity can increase the toxicity of APAP. The

elevation of unconjugated bile acids after the administra-

tion of HIV therapeutic drug indinavir has been well

explained by the inhibition of indinavir toward UGT1A1-

catalyzed bilirubin glucuronidation (Zucker et al. 2001).

Sorafenib-induced elevation of serum bilirubin can also be

explained through inhibition of UGT1A1 activity (Meza-

Junco et al. 2009).

The present study aims to investigate the influence of

psoralidin toward the toxicity of irinotecan. Firstly, the

in vitro HIM reaction system was employed to demonstrate

the inhibition of psoralidin toward the glucuronidation of

SN-38 which is the active metabolite of irinotecan. The

competitive inhibition of psoralidin toward SN-38 glucu-

ronidation was demonstrated, and the inhibition kinetic

parameter is relatively low. Furthermore, these in vitro

results can well translated into the in vivo results. Pre-

treatment with psoralidin can increase the toxicity of iri-

notecan. All these results were beneficial for the guidance

for the clinical co-utilization of irinotecan and psoralidin-

containing herbs.

Conflict of interest There is no conflict of interests.

References

Chen S, Yueh MF, Bigo C, Barbier O, Wang K, Karin M, Nguyen N,

Tukey RH (2013) Intestinal glucuronidation protects against

chemotherapy-induced toxicity by irinotecan (CPT-11). Proc

Natl Acad Sci 110(47):19143–19148

Cong M, Hu CM, Cao YF, Fang ZZ, Tang SH, Wang JR, Luo JS

(2013) Cryptotanshinone and dihydrotanshinone I exhibit strong

inhibition towards human liver microsome (HLM)-catalyzed

propofol glucuronidation. Fitoterapia 85:109–113

Egan G, Hughes CA, Ackman ML (2014) Drug interactions between

antiplatelet or novel oral anticoagulant medications and antiret-

roviral medications. Ann Pharmacother 48(6):734–740

Fang ZZ, Zhang YY, Ge GB, Huo H, Liang SC, Yang L (2010) Time-

dependent inhibition (TDI) of CYP3A4 and CYP2C9 by

noscapine potentially explains clinical noscapine-warfarin inter-

action. Br J Clin Pharmacol 69(2):193–199

Fang ZZ, Cao YF, Hu CM, Hong M, Sun XY, Ge GB, Liu Y, Zhang

YY, Yang L, Sun HZ (2013) Structure-inhibition relationship of

ginsenosides towards UDP-glucuronosyltransferases (UGTs).

Toxicol Appl Pharmacol 267(2):149–154

Lawrence SK, Nguyen D, Bowen C, Richards-Peterson LR, Skordos

KW (2014) The metabolic drug-drug interaction profile of

dabrafenib: in vitro investigations and quantitative extrapolation

of the P450-mediated DDI risk. Drug Metab Dispos. doi:10.

1124/dmd.114.057778

Liang HL, Hu AP, Li SL, Liu JY (2014) Combining bevacizumab and

panitumumab with irinotecan, 5-fluorouracil, and leucovorin

(FOLFIRI) as second-line treatment in patients with metastatic

colorectal cancer. Med Oncol 31(6):976

Liu Y, Zhang JW, Li W, Ma H, Sun J, Deng MC, Yang L (2006)

Ginsenoside metabolites, rather than naturally occurring ginse-

nosides, lead to inhibition of human cytochrome P450 enzymes.

Toxicol Sci 91(2):356–364

Meza-Junco J, Chu QS, Christensen O, Rajagopalan P, Das S,

Stefanyschyn R, Sawyer MB (2009) UGT1A1 polymorphism

and hyperbilirubinemia in a patient who received sorafenib.

Cancer Chemother Pharmacol 65(1):1–4

Mutlib AE, Goosen TC, Bauman JN, Williams JA, Kulkarni S,

Kostrubsky S (2006) Kinetics of acetaminophen glucuronidation

by UDP-glucuronosyltransferases 1A1, 1A6, 1A9, 2B15. Poten-

tial implications in acetaminophen-induced hepatotoxicity.

Chem Res Toxicol 19(5):701–709

Saliba F, Hagipantelli R, Misset JL, Bastian G, Vassal G, Bonnay M,

Herait P, Cote C, Mahjoubi M, Mignard D, Cvitkovic E (1998)

Pathophysiology and therapy of irinotecan-induced delayed-

onset diarrhea in patients with advanced colorectal cancer: a

prospective assessment. J Clin Oncol 16(8):2745–2751

Xing J, Che W (2012) 20(S)-protopanaxatriol (ppt) exhibits inhibition

towards UDP-glucuronosyltransferase (UGT)-catalyzed zidovu-

dine glucuronidation. Lat Am J Pharm 31(4):628–631

Yu J, Wang CQ, Fang LH, Li M, Wang XY, Dai LL, Gao YJ (2012)

Tacrolimus (Tacro) strongly inhibits intestinal UDP-glucurono-

syltransferase (UGT) 1A8. Lat Am J Pharm 31(4):625–627

Yu J, Han JC, Gao YJ (2014) Biotransformation of glucoaurantio-

obtusin towards aurantio-obtusin increases the toxicity of

irinotecan through increased inhibition towards SN-38 glucu-

ronidation. Phytother Res. doi:10.1002/ptr.5162

Zucker SD, Qin X, Rouster SD, Yu F, Green RM, Keshavan P,

Feinberg J, Sherman KE (2001) Mechanism of indinavir-induced

hyperbilirubinemia. Proc Natl Acad Sci 98(22):12671–12676

Eur J Drug Metab Pharmacokinet