Gender Differences in the Hepatotoxicity and Toxicokinetics of Emodin

Jul 16, 2018 - This study aims to (a) estimate gender differences of the hepatotoxicity and ... Several in vitro studies demonstrated that emodin coul...
0 downloads 0 Views 2MB Size
Subscriber access provided by UNIV OF NEW ENGLAND ARMIDALE

Article

Gender differences in the hepatotoxicity and toxicokinetics of emodin: The potential mechanisms mediated by UGT2B7 and MRP2 Lili Wu, Weichao Han, Yulian Chen, Tao Zhang, Junjin Liu, Shilong Zhong, Han Liu, Congcong Han, Zhongyi Zhang, Shu-Wen Liu, and Lan Tang Mol. Pharmaceutics, Just Accepted Manuscript • DOI: 10.1021/acs.molpharmaceut.8b00387 • Publication Date (Web): 16 Jul 2018 Downloaded from http://pubs.acs.org on July 16, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 53 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Pharmaceutical Sciences, Southern Medical University, Guangzhou 510515, China.

ACS Paragon Plus Environment

Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1

Gender differences in the hepatotoxicity and toxicokinetics of emodin: The

2

potential mechanisms mediated by UGT2B7 and MRP2

3 4

Lili Wu1,2#, Weichao Han2#, Yulian Chen2, Tao Zhang2, Junjin Liu2, Shilong Zhong3, Han Liu2,

5

Congcong Han2, Zhongyi Zhang2, Shuwen Liu1,2*, Lan Tang1,2*

6 7

#

Lili Wu and Weichao Han contribute equally to this paper

8 9

Authors and affiliations:

10

1

11

Kidney Disease, Division of Nephrology, Southern Medical University, Guangzhou, China.

12

2

13

of Pharmaceutical Sciences, Southern Medical University, Guangzhou 510515, China.

14

3

15

Sciences, China.

State Key Laboratory of Organ Failure Research, National Clinical Research Center of

Biopharmaceutics, Guangdong Provincial Key Laboratory of New Drug Screening, School

Medical Research Center of Guangdong General Hospital, Guangdong Academy of Medical

16 17 18 19 20

*Corresponding Author Address: Department of Pharmaceutics, School of Pharmaceutical

21

Sciences, Southern Medical University, Guangzhou 510515, China. Tel. /fax: +86

22

20-61648596. E-mail addresses: [email protected], [email protected]

1

ACS Paragon Plus Environment

Page 2 of 53

Page 3 of 53 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

23

Abstract

24

Emodin is a main anthraquinone compound which exists in Chinese traditional medicines

25

including Polygonum multiflorum and Rhubarb. It is documented to have obvious liver and

26

kidney toxicity. This study aims to (a) estimate gender differences of hepatotoxicity and

27

toxicokinetics in rats after oral administration of emodin (60 and 150 mg/kg/d) for

28

consecutive 28 days; (b) clarify relative mechanisms caused by glucuronidation and

29

disposition. Hepatotoxicity was significantly higher in female rats than that in male rats as

30

evidenced by histopathological and biochemical tests. Similarly, the toxicokinetics profiles

31

of emodin had time and gender differences, which could cause time and gender differences

32

in hepatotoxicity. Metabolic and transcriptomics data of 55 human liver and 36 human

33

kidney samples demonstrated that UDP-Glucuronosyltransferase 2B7 (UGT2B7) was the

34

predominant enzyme for emodin glucuronidation. A genome-wide association studies

35

(GWAS) identified that rs11726899 located within ~50 kb of the transcript of UGT2B could

36

significantly affect emodin metabolism. Knockdown of UGT2B7 in HepG2 cells

37

significantly decreased emodin glucuronidation and increased cytotoxicity of emodin. The

38

gene expression and protein levels of UGT2B7 were decreased, but those of

39

multidrug-resistant-protein 2 (MRP2) were increased in HepG2 cells after treated with 50

40

µM emodin for 48 h. Long-term use of emodin could decrease the intrinsic clearance (CLint,

41

decreased by 18.5%-35.4%) values of zidovidue (UGT2B7 substrate) glucuronide in both

42

male and female liver microsomes from rats administrated with emodin for 28 days, thus

43

causing the accumulation of emodin. However, higher self-induced MRP2 expression and

44

lower hepatotoxicity were observed in emodin-treated male rats compared to that in female 2

ACS Paragon Plus Environment

Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

45

rats. Therefore, gender differences in hepatotoxicity and toxicokinetics of emodin are

46

potentially mediated by the coupling of UGT2B7 and MRP2 in vivo.

47

Key words: Emodin; hepatotoxicity; toxicokinetics; GWAS; UGT2B7; MRP2

3

ACS Paragon Plus Environment

Page 4 of 53

Page 5 of 53 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

48

1. Introduction

49

Emodin (6-methyl-1, 3, 8-trihydroxyanthraquinone), an active and toxic anthraquinone

50

mainly from Rhubarb and Polygonum multiflorum 1, widely exists in more than 800 kinds of

51

Chinese medicine preparations in Chinese pharmacopoeia. It has a multitude of remarkable

52

pharmacological effects such as antibacterial, antidiabetic, antiviral, and antitumor

53

activities

54

taken for long times, and there are increasing reports on the hepatotoxicity, nephrotoxicity,

55

and reproductive toxicity of emodin. Several in vitro studies demonstrated that emodin could

56

cause cytotoxicity and embryonic toxicity

57

stabilizing topoisomerase II-DNA cleavage complexes and inhibiting ATP hydrolysis of

58

topoisomerase II 9. In addition, emodin also exhibits toxicity in vivo. The US National

59

Toxicology Program conducts toxicology and carcinogenesis studies on emodin using F344/N

60

rat and B6C3F1 mice, results of which show that emodin exposure could result in increased

61

incidences of nephropathy and liver lesion 10. It is also reported that Shou-Wu Pian and diet

62

pills containing emodin could cause acute hepatotoxicity in individuals 11, 12.

2-5

. However, most Chinese medicine preparations containing emodin need to be

6-8

, and trigger DNA double-strand breaks by

63

Emodin undergoes extensive glucuronidation in liver and intestine. Three glucuronide

64

metabolites of emodin have been identified and emodin-3-O-β-D-glucuronide (Emodin-3-G)

65

accounted for 99.5% and 98.4% of the total glucuronides in human intestine microsomes

66

(HIMs) and human liver microsomes (HLMs), respectively

67

studies have demonstrated that emodin is rapidly metabolized into emodin-3-G, the major

68

metabolite of emodin observed via UPLC-MS/MS in serum

69

rate of emodin-3-G is commonly used to measure the glucuronidation of emodin. It is 4

ACS Paragon Plus Environment

13

. Similarly, pharmacokinetic

14, 15

. Therefore, the formation

Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 53

70

reported that emodin mainly glucuronidated by recombinant UGTs 1A1, 1A3, 1A8, 1A9,

71

1A10, 2B7, and 2B15 in vitro 13. However, the predominant UGT isoform catalyzing emodin

72

has not been clarified in different human organs. Hence, an individual-based model to predict

73

the individual glucuronidation behavior of emodin in 55 human liver and 36 kidney samples

74

was utilized as published previously

75

applied differential expression analysis to detect biologically relevant genes, which have

76

become the golden standard for transcriptome analysis in recent years

77

effective method for identifying the genetic variations strongly affected pharmacokinetics of

78

drugs

79

pharmacokinetics of many drugs are identified 20-22. Accordingly, RNA-Seq study and GWAS

80

were used to further investigate the predominant UGT isoform responsible for the metabolism

81

of emodin and underlying genetic variants involved in emodin glucuronidation.

18, 19

16

. Many RNA sequencing (RNA-Seq) studies have

17

. GWAS is an

. In recent GWAS, different UGT isoform polymorphisms associated with the

82

UGT enzymes and efflux transporter are generally regarded as important defense

83

mechanisms which can protect biological systems from various insults. The efflux transporter

84

MRPs (also known as ATP-binding cassette sub-family C, ABCC) play an important role in

85

the elimination of various drugs in vivo and thus may affect the efficacy and toxicity of drugs

86

23

87

membrane into the bile 24. Some drugs could affect the expression levels of efflux transporters,

88

and thus alter the function of hepatic transporters

89

main active ingredients in herbal laxative preparations and diet pills

90

taken by female for losing weight and can cause various side effects. A research shows that

91

incidence, severity and progression of inflammation diseases are gender-biased 28. Moreover,

. As one of MRPs members, MRP2 transports emodin glucuronides across the canalicular

25

. Emodin has purgative effect and is the

5

ACS Paragon Plus Environment

26, 27

. It is frequently

Page 7 of 53 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

92

gender could influence the efficacy of therapeutics, and side effects of drug are more common

93

in females than in males

94

detoxification and gender differences in hepatotoxicity of emodin. Intensive research on these

95

differences is essential to the rational use of emodin in both genders considering its promising

96

medical value. Our previous study has demonstrated that emodin glucuronidation has gender

97

differences

98

toxicokinetics profiles, and glucuronidation of emodin. Therefore, the final purpose of our

99

study is to investigate detoxification way of emodin and mechanisms of gender differences in

29, 30

. However, there is little information available regarding

31

. In this study, we need to clarify the relationship of hepatotoxicity,

100

hepatotoxicity and toxicokinetics.

101

2. Materials and methods

102

2.1. Chemicals and reagents

103

Emodin, acetaminophen (APAP), genistein, zidovudine, testosterone, and urethane were

104

purchased from Aladdin Industrial Corporation (China). Uridine diphosphate glucuronic

105

acid (UDPGA), alamethicin, D-saccharic-1, 4-lactone monohydrate, tris base, and

106

magnesium chloride were purchased from Sigma-Aldrich (USA). All commercial reagents

107

were of analytical grade, and the purity of each compound was not less than 98%.

108

Recombinant human UGTs were obtained from BD Gentest Corporation (USA).

109

Acetonitrile, methanol, and ammonium acetate (≥ 99%, HPLC grade) were purchased from

110

Merck Company (Germany). The assay kits for the measurements of biochemical

111

parameters were bought from Nanjing Jiancheng Biological Engineering Institute (China).

112

HEK293, HK-2, HepG2, and L-02 cell lines were purchased from American Tissue Culture 6

ACS Paragon Plus Environment

Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

113

Collection (USA). Dulbecco's modified eagle medium (DMEM) and fetal bovine serum

114

(FBS) were purchased from GIBCO Company (USA). The siRNA was purchased from

115

Genechem Corporation (China). Lipofectamine 3000 was purchased from Thermo Fisher

116

Scientific (USA). RNAprep Pure Tissue Kit was purchased from TIANGEN BIOTECH CO.,

117

LTD (China). PrimeScript RT reagent Kit was purchased from Takara Company (Japan).

118

SYBR Green PCR Master Mix was purchased from Promega Company (Promega, USA).

119

Rabbit polyclonal antibody against UGT2B7 (ab126269), MRP2 (ab203397), GAPDH

120

(ab22555), and HRP-conjugated goat anti-rabbit IgG (ab6721) were purchased from Abcam

121

Company (USA). ECL western blotting substrate was purchased from Millipore Company

122

(USA).

123

2.2. Samples collection and ethics statement

124

Human liver tissues from 55 patients who had undergone surgery for hepatocellular

125

carcinoma resection were obtained from Sun Yat-Sen Memorial Hospital (Guangzhou,

126

Guangdong, China). Human kidney tissues from 36 patients who had undergone radical

127

nephrectomy were obtained from Nanfang Hospital of Southern Medical University

128

(Guangzhou, Guangdong, China). Healthy tissues surrounding the primary tumor were

129

isolated, and only histologically non-tumorous tissues were used. All tissue samples were

130

stored in liquid nitrogen for later experiments. The clinical characteristics including age,

131

gender, disease, and medication history of the patients were recorded.

132

Approvals for liver and kidney tissues collection and in vitro drug metabolism studies were

133

obtained from the ethics committee of Sun Yat-Sen Memorial Hospital (ECSYS NO. 7

ACS Paragon Plus Environment

Page 8 of 53

Page 9 of 53 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

134

CS07095) and Nanfang Hospital of Southern Medical University (ECNFH NO. CN14184),

135

respectively. All participants have provided written informed consent. Animals were approved

136

by animal ethics committee of Southern Medical University (Guangzhou, Guangdong, China).

137

All animals received humane care in compliance with the animal care guidelines of the

138

Southern Medical University.

139

2.3. Animals and experiment design

140

Sprague-Dawley (S.D.) rats (SPF, 4-6 week, 180-200 g) were obtained from laboratory

141

animal center of Southern Medical University. Animals were kept under conditioned

142

environment (22-25 oC, 60-70% relative humidity and 12 h light/dark cycle) with free access

143

to food and water for at least 7 days. 48 rats (half males and half females) were randomly

144

divided into control group (0.5% CMC-Na), model group (500 mg/kg APAP), low-dose

145

emodin group (60 mg/kg), and high-dose emodin group (150 mg/kg). There were 12 rats in

146

each group (6 males and 6 females). In this study, the dosages of emodin were 60 mg/kg and

147

150 mg/kg, equivalent to 7 and 17.5 times of the suggested dosage for human in the 2015

148

edition of Chinese pharmacopoeia (8.6 mg/kg, converted to rat dose based on body surface

149

area conversion), respectively. The administered volume was controlled below 2 mL per day

150

for each rat. The experimental groups received intragastric administration of drugs dissolved

151

in 0.5% CMC-Na once a day for 4 weeks.

152

On the 1st and 28th day, blood samples were withdrawn at designed time points via orbital

153

vein and stored at −80 oC for biochemical and toxicokinetics assay. On the 29th day, the rats

154

were sacrificed. The fresh liver and kidney tissues were immediately removed. A small 8

ACS Paragon Plus Environment

Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 53

155

portion of the tissues were prepared for histopathological observations, and the rest were used

156

to determine the gene and protein expression levels via quantitative real-time PCR (qRT-PCR)

157

and western blotting, respectively.

158

2.4. Biochemical analysis and histopathological studies

159

The blood samples were centrifuged at 4 oC, 5000 rpm for 8 min, and the supernatant of

160

each sample was transferred to clean tubes for detection. Biochemical parameters including

161

alanine aminotransferase (ALT), aspartate aminotransferase (AST), blood urea nitrogen

162

(BUN), and creatinine (CREA) were determined by automatic microplate reader (Bio-Rad,

163

USA).

164

Liver and kidney tissues for histopathologic examination were immediately fixed in 10%

165

neutral buffered formalin and embedded in paraffin. Sections (2-3 µm) were stained with

166

hematoxylin and eosin (H&E). The stained sections were observed under a light microscope.

167

2.5. Toxicokinetics studies

168

Toxicokinetics experiments were performed on the 1st and 28th experimental day. During

169

experiments, 13% circulating blood from each rat within 24 h was collected, which was below

170

the maximum blood sample volumes (15%) allowed for multiple sampling in rats

171

oral administration of emodin (60 mg/kg and 150 mg/kg), blood samples were collected in a

172

gentle manner via orbital venous of rats at nine time points (0, 0.25, 0.5, 1, 2, 4, 6, 10, and 24

173

h). All blood samples were immediately centrifuged at 4 °C, 5000 rpm for 8 min. Internal

174

standard (IS) solution (genistein) was added into plasma sample, and the mixtures were 9

ACS Paragon Plus Environment

32

. After

Page 11 of 53 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

175

extracted from plasma by liquid-liquid extraction using ethylene acetate. The mixtures were

176

centrifuged at 4 oC, 13000 rpm for 30 min and analyzed by UPLC/MS/MS as described in our

177

previous study 15. A non-compartmental model (statistical moments) was used to calculate the

178

toxicokinetics parameters with WinNonLin 3.3.

179

2.6. Glucuronidation of emodin in HKMs, HLMs, recombinant UGTs, and HepG2 cell lysates

180

HLMs, human kidney microsomes (HKMs), and rat liver microsomes (RLMs) preparation 16, 33, 34

181

and incubation were performed as described previously

182

incubation system contained potassium phosphate (50 mM, pH 7.4), MgCl2 (0.88 mM),

183

D-saccharic-1,4-lactone monohydrate (4.4 mM), alamethicin (50 mg/mL), UDPGA (3.5 mM,

184

added last), thirteen UGT isoforms (UGTs 1A1, 1A3, 1A4, 1A6, 1A7, 1A8, 1A9, 1A10, 2B4,

185

2B7, 2B10, 2B15, and 2B17)/HLMs/HKMs/HepG2 cell lysates with final concentration of

186

0.01 mg/mL and emodin (2.5-100 µM) in a total volume of 400 µL. Incubations were carried

187

out for 15 min in shaking water bath (80 rpm) at 37 oC. Ice-cold acetonitrile containing IS

188

(genistein) was added to terminate the enzyme activity. All experiments were performed in

189

triplicate. The samples were then analyzed by UPLC, method of which was the same as

190

previous study 31.

191

2.7. RNA preparation and sequencing

. Briefly, a typical phase II

192

RNA preparation was as described previously 16. Libraries were subsequently constructed

193

using 3 µL total RNA and NEBNext UltraTM RNA Library Prep Kit for Illumina (NEB,

194

USA), following the manufacturer's recommendations. Samples were sequenced on the 10

ACS Paragon Plus Environment

Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

195

Hiseq4000 platform, using 150 bp paired-end reads. Before alignment, clean data were

196

obtained by removing reads containing adapter, reads containing ploy-N and low quality

197

reads from raw data. The reference genome human release 81 was downloaded from

198

Ensemble (http://asia.ensembl.org). The Index of the reference genome was built using

199

Bowtie v2.2.3

200

TopHat v2.0.12

201

counts were estimated using Cufflinks v2.2.1

202

expression levels are less reliable 38, transcripts with expression levels equal to or fewer than

203

five counts were removed from the dataset. The expected number of fragments per kilobase of

204

transcripts sequence per millions base pairs sequenced (FPKM) of each gene was calculated

205

based on the length of the gene and read counts mapped to this gene.

206

2.8. SNP genotyping and quality control in the GWAS

35

and paired-end clean reads were aligned to the reference genome using

36

. On average, 34108 genes were detected among the mapped reads. Read 37

. Because transcripts with extremely low

207

A total of 55 liver samples collected between 2013 and 2015 were genotyped using the

208

Illumina SNP chip (HumanOmniZhongHua-8v1). First, the genotyping rate, heterozygous rate,

209

inbreeding coefficient, and principle component analysis (PCA) were performed to filter

210

unqualified samples. Then we applied the following thresholds for SNPs quality control (QC)

211

in data cleaning: minor allele frequency (MAF) ≥ 5% for all samples, SNP call rate ≥ 95% for

212

all samples, Hardy-Weinberg equilibrium (HWE) test p-value ≥ 0.0001. A total of 55 samples

213

with 685595 SNPs on autosomal chromosomes passed the QC filters and were used for the

214

GWAS.

215

2.9. Cell culture and cytotoxicity of emodin and emodin-3-G in four cells 11

ACS Paragon Plus Environment

Page 12 of 53

Page 13 of 53 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

216

Molecular Pharmaceutics

Four different cell lines including normal human embryonic kidney cell lines (HEK 293) 39, 40

41

217

normal human renal epithelial cell lines (HK-2)

218

and normal human liver cell lines (L-02)

219

this study, these four cell lines were used to investigate the cytotoxicity of emodin and

220

emodin-3-G. HEK293, HK-2, HepG2, and L-02 cells were cultured in the DMEM medium

221

supplemented with 10% FBS, penicillin (100 IU/mL), and streptomycin (100 µg/mL). Cells

222

were maintained at 37 oC in a humidified atmosphere containing 5% CO2. MTT was used to

223

measure viable cell counts. According to the previously described method 33, emodin (50 µM)

224

was glucuronidated by pooled HLMs/HKMs and incubation for 15, 30, 45, and 60 min,

225

respectively. The products of the glucuronidation reaction were divided into two parts, one of

226

which was used to analyze the concentration changes of emodin and emodin-3-G with the

227

different incubation time by UPLC, the other was analyzed the toxicity of glucuronidation

228

products containing different proportions of emodin and emodin-3-G in four cell lines by

229

MTT cell viability assy. Four cells were exposed to glucuronidation products for 48 h,

230

following addition of 10 µL MTT solution at final concentration of 0.5 mg/mL in PBS (pH

231

7.4) to each well, and then cells were incubated for an additional 4 h. After the medium was

232

removed, the dye crystal was dissolved in 150 µL dimethylsulfoxide (DMSO). The cell

233

viability was detected by automatic microplate reader at 570 nm.

234

2.10. Transfection of siRNA in HepG2 cells

6

, human hepatoma cell lines (HepG2)

,

are often used to study the toxicity of emodin. In

235

HepG2 cells were seeded in 6-well plate at density of 4×105/well and then incubated for 24

236

h till 80%-90% confluence. The target sequence of siRNA for UGT2B7 knockdown and 12

ACS Paragon Plus Environment

Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

237

control were (5′ to 3′) GGTTCCAGTACCACTCTTT and TTCTCCGAACGTGTCACGT,

238

respectively. HepG2 cells were transfected with UGT2B7 siRNA and Lipofectamine 3000

239

following manufacturer’s instruction. HepG2 cells were harvested 48 h later for fluorescence

240

microscopy, qRT-PCR, and western blotting assays.

241

2.11. Quantitative real-time PCR and Western blotting

242

Tissue total RNA samples were extracted from liver and kidney using RNAprep Pure

243

Tissue Kit according to the manufacturer's instruction. The concentration of RNA samples

244

was quantified by a measurement of OD260/280 using Nano Drop 2000 (Thermo, USA).

245

Total RNA samples were reverse-transcribed into cDNA using Takara PrimeScript RT reagent

246

Kit. In brief, 2 µL of the cDNA was used in 20 µL of reaction mixture containing 10 µL of

247

SYBR Green PCR Master Mix and 0.4 mM of a pair of primers for the detection of the

248

mRNA. Quantitative real-time PCR was conducted on an ABI Step-One Sequence Detection

249

System (Applied Biosystems 7500, USA). The results were analyzed using the ∆∆Ct method

250

for relative quantitation. GAPDH was used as housekeeping gene. All the primer sequences

251

for the genes are listed in the Table 1.

252

Protein extracts (30 µg) of HepG2 cells and liver tissues were loaded onto a 10% sodium

253

dodecyl sulfate-polyacrylamide gel electrophoresis and then transferred to polyvinylidene

254

fluoride membranes (Millipore, USA) for 2 h through wet transfer method. Subsequently, the

255

membranes were incubated with antibody (1:1000 for UGT2B7, 1:500 for MRP2, 1:3000 for

256

GAPDH) overnight after blocking with 5% nonfat milk for 2 h at room temperature. Then the

257

HRP-conjugated goat anti rabbit secondary antibody (1:3000) was used for detection by ECL 13

ACS Paragon Plus Environment

Page 14 of 53

Page 15 of 53 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

258

substrates. GAPDH was used as an internal control. Each integrated optical density value of

259

stripe was measured and calculated by Image J Software (USA). All experiments were

260

repeated three times.

261

2.12. Cytotoxicity of emodin in HepG2 cells transfected with UGT2B7 siRNA

262

HepG2 cells were transfected with UGT2B7 siRNA and control siRNA and then exposed

263

to 50 µΜ emodin for 48 h. Supernatants were obtained for ALT and AST detection by

264

automatic microplate reader at 510 nm. Cytotoxicity of HepG2 cells transfected with

265

UGT2B7 siRNA and control siRNA were determined by MTT assay.

266

2.13. The effects of emodin on the expression of UGT2B7 and MRP2 in HepG2 cells

267

HepG2 cells were seeded in 6-well plate at density of 4×105/well and then incubated for 24

268

h till 80%-90% confluence. Cells were exposed to 50 µM emodin for 48 h and then harvested

269

for qRT-PCR and western blotting assays.

270

2.14. Effects of emodin on UGT2B7 activity in the rats

271

Zidovudine (3'-azido-3'-deoxythymidine) is a selective substrate of UGT2B7, and 42

272

formation rate of zidovudine-5'-glucuronide represents the activity of UGT2B7

273

reaction mixture contained 0.25 mg/mL RLMs, 50 mM tris buffer (pH 7.6), MgCl2 (0.88 mM),

274

D-saccharic-1,4-lactone monohydrate (4.4 mM), alamethicin, UDPGA (3.5 mM), and

275

zidovudine (6.25-600 µM) in a total volume of 400 µL. Mixtures were incubated at 37 oC for

276

60 min. Ice-cold acetonitrile containing IS (testosterone) was added to terminate the enzyme

14

ACS Paragon Plus Environment

. The

Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

277

activity. All experiments were performed in triplicate. Samples were detected by AB Sciex

278

4000 QTRAP and shimadzu UPLC/MS/MS system. Detection was in multiple reaction

279

monitoring (MRM) mode with an ESI interface for positive ions ([M + H]+) by recording ion

280

currents for the following transitions: m/z 268→127 for zidovudine, m/z 444→127 for

281

zidovudine-5'-glucuronide, m/z 289→109 for testosterone. Parameters were set as following:

282

collision energy, 40 eV; Declustering potential, 100 eV; Curtain gas, 30 psi. Chromatographic

283

separations were achieved on an Agilent SB-C18 column (2.1 mm × 50 mm, 1.8 µm). The

284

mobile phase consisted of purified water (A) and acetonitrile (B) with a gradient elution of

285

5% B at 0-0.3 min, 5%-95% B at 0.3-1 min, 95% B at 1-2.4 min, 95%-5% B at 2.4-2.5 min,

286

5% A at 2.5-4 min with a flowing rate of 0.4 mL/min. Enzymes kinetic parameters were

287

analyzed by Graphpad Prism 5 (USA).

288

2.15. Data analysis

289

2.15.1. Covariates selection

290

Initially, in order to select the variables to be included as covariates in the analyses,

291

univariate relations of sex, age, weight and medication history were estimated. The variables

292

showing significant univariate association (p < 0.05) with the formation rate of emodin-3-G in

293

HLMs/HKMs were analyzed in a multiple model. Finally, the covariates having p < 0.05 in

294

multiple models were included in the regression models for emodin glucuronidation.

295

2.15.2. Association analysis between UGTs gene expression, emodin glucuronidation rate,

296

and SNPs 15

ACS Paragon Plus Environment

Page 16 of 53

Page 17 of 53 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

297

The gene expression of thirteen UGT isoforms was obtained from RNA-Seq data.

298

Associations between gene expression levels and emodin glucuronidation rates were tested

299

with a linear model using R 3.4.4 (https://www.r-project.org/). The use of proton pump

300

inhibitors (PPIs) was coded 0/1 and selected as covariates in the models involved in HLMs.

301

The resulting p values were obtained from the moderated t-statistic. The significant level for

302

correlated genes was set at 0.05. GWAS were performed to detect the additive genetic effect

303

of SNPs on emodin glucuronidation rate using PLINK v1.07 43. The analyses were performed

304

using a linear model adjusting for the use of PPIs. In order to decide the threshold of

305

significant association for the GWAS, Benjamini-Hochberg method

306

the raw p values, and a significant association was determined when threshold of false

307

discovery rate (FDR) < 0.05 (p < 3.0 × 10-7).

308

2.15.3. Analysis combining the activities of UGT isoforms and corresponding gene expression

309

We consider that the quantity of each UGT isoform and their related catalytic capability are

310

two major factors affecting the activity of UGT isoform in vivo. A feature evaluation index

311

(Coni) was applied to estimate the contribution of enzymes to drug metabolism. According to

312

the activities of thirteen UGTs for emodin-3-G formation in vitro and their gene expression in

313

kidney and liver tissues, the Coni of these thirteen UGTs for emodin glucuronidation were

314

estimated by equation 1:

 =

 ×  × 100% ∑ (  ×  )

44

was applied to correct

(1)

315

where Coni is the contribution of UGT isoform i for drug metablism, Ai is the activity of a

316

given recombinant human UGT isoform i, Ei is the gene expression of a given UGT isoform i. 16

ACS Paragon Plus Environment

Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

317

Page 18 of 53

2.15.4. Enzyme kinetics of emodin in HepG2 cell lysates and zidovudine in RLMs

318

Metabolic rates of emodin/zidovudine were expressed as amount of metabolites formed per

319

mg protein per min (nmol/min/mg). Due to the lack of emodin glucuronide standard substance,

320

an emodin standard curve was used for quantitation of emodin glucuronide by using a

321

conversion factor (K), as described previously 31. Kinetic parameters were obtained based on

322

the profiles of Eadie-Hofstee plots, and the fit to various kinetic equations was shown below.

323

The Michaelis-Menten model (equation 2) and the sigmoidal model (equation 3), which

324

incorporates the Hill coefficient (n), were evaluated for best-fit of the data 45.  ×   + 

(2)

 ×   ′ +  

(3)

= =

325

where Km is the is the substrate concentration at half-maximal velocity, K' is equal to Km when

326

n = 1, Vmax is the maximum rate of metabolism, C is the concentration of substrate, and n is an

327

exponent indicative of the degree of curve sigmoidicity (or Hill coefficient). CLint value was

328

calculated as the Vmax/Km value for the Michaelis-Menten kinetics.

329

2.15.5. Statistical Analysis

330

Data were expressed as the means ± SD. One-way ANOVA with Turkey-Kramer post hoc

331

tests and Student’s t test were used to evaluate differences between control and treatment.

332

Differences were considered significant when p values were less than 0.05.

333

3. Results

334

3.1. Common changes 17

ACS Paragon Plus Environment

Page 19 of 53 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

335

In the experiments, the skin color of emodin-treated rats changed from white to yellow,

336

and there also occurred dark yellow urine and obvious diarrhea symptom, which showed up

337

earlier in female rats (on the 8th day) than that in male rats (on the 12th day) after repeated

338

administration. The body weight gain of female rats was significantly suppressed after

339

administration of 150 mg/kg emodin compared with the control group. However, the body

340

weight gain of male rats was increased after administration of emodin compared with the

341

control group (Supplementary Figure. 1). At the end of the study, food intake was 18-20 g

342

per day for each rat, and the weight of emodin-treated male rats (240 ± 5 g) was larger than

343

that of emodin-treated female rats (215 ± 8 g).

344

3.2. Biochemical and histopathological analysis

345

The results of blood biochemical analysis are shown in Fig. 1A, in which ALT and AST are

346

generally regarded as sensitive biochemical markers for hepatic damage. On the 28th day,

347

serum levels of ALT and AST in APAP group (increased by 54%-312%) and emodin group

348

(increased by 37%-162%) were significantly higher than that in control group (p < 0.05). In

349

addition, the levels of ALT and AST in female rats were 1.46-fold to 1.75-fold higher than

350

those in male rats after administration of emodin for 28 day (p < 0.05). In the experiments, the

351

analyses for markers of renal injury (CREA and BUN) and renal histopathology demonstrated

352

that emodin-treated rats had no significant renal injury (Fig. 1 A and Supplementary Figure 2).

353

Renal lesions were only found in female APAP group on the 28th day (Fig. 1 A and

354

Supplementary Figure 2).

355

Typical histopathological section photos of liver are shown in Fig. 1B. No liver lesion was 18

ACS Paragon Plus Environment

Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

356

found in control groups while evident liver lesions were observed in APAP groups on the 28th

357

day. After administration of emodin for 28 days, all rats exhibited different degrees of

358

hepatotoxicity symptoms.

359

3.3. Toxicokinetics studies in rats

360

The toxicokinetics profiles of emodin and emodin-3-G on the 1st and 28th day are shown

361

in Figs. 1C-1D. The main toxicokinetics parameters were calculated by non-compartment

362

model (statistical moments), and the results are shown in Table 2 and Table 3. Toxicokinetics

363

parameters of emodin and emodin-3-G had time-dependence. In female rats, the area under

364

the curve (AUC) of emodin on the 28th day was 1.09-fold higher than that on the 1st day. It

365

showed that, except for low-dose female group, the AUC of emodin-3-G on the 28th day was

366

2.37-fold to 3.09-fold higher than that on the 1st day. In addition, gender differences in

367

toxicokinetics parameters of emodin and emodin-3-G were observed. In low-dose groups, the

368

AUC of emodin in male rats was higher (78%) than that in female rats on the 1st day (p