Docetaxel pharmacogenetics and its influence on the pharmacokinetics pharmacodynamics of docetaxel in asian nasopharyngeal cancer patients

The following genes involved in the biochemical pathways of docetaxel were studied: regulatory nuclear receptors PXR, CAR, RXRα, HNF4α, drug metabolism enzymes CYP3A4, CYP3A5 and drug tr

Trang 1

DOCETAXEL PHARMACOGENETICS AND ITS INFLUENCE ON THE PHARMACOKINETICS / PHARMACODYNAMICS OF DOCETAXEL IN ASIAN NASOPHARYNGEAL CANCER PATIENTS

Chew Sin Chi

B Health Sciences (Hons), University of Adelaide

A THESIS SUBMITTED FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY DEPARTMENT OF PHARMACOLOGY

NATIONAL UNIVERSITY OF SINGAPORE

2013

Trang 2

i

DECLARATION

I hereby declare that this thesis is my original work and it has been written by

me in its entirety I have duly acknowledged all the sources of information

which have been used in the thesis

This thesis has also not been submitted for any degree in any university

previously

- Chew Sin Chi (21 November 2013)

Trang 3

ii

ACKNOWLEDGEMENTS

First and foremost, I would like to thank my supervisor, Professor Balram Chowbay for his support, guidance and invaluable insights throughout my postgraduate studies Four-year stint under his guidance has helped me to mature as a student and a researcher I would also like to express my gratitude

to Professor Edmund Lee for his constructive and insightful comments during project discussions I was very fortunate to be acquainted with another PhD candidate, Joanne Lim Siok Liu, who has shared weal and woe with me throughout the training process Completing this work would have been more difficult were it not for the support and friendship that she offered I am also grateful to the colleagues in Clinical Pharmacology Laboratory, National Cancer Centre Singapore who have provided technical support in my work I

am thankful to Dr Tan Eng Huat and Dr Darren Lim Wan Teck from the Department of Medical Oncology for providing the clinical samples for this study, as well as the patients who were involved in this study For the financial support, I would like to thank National Medical Research Council Singapore for study grants (NMRCB1011 and NMCCG10122) and National University

of Singapore for the research scholarship Last but not least, I would like to express my deepest thanks to my family and people who are close to me for their unwavering support and encouragement Their presence has provided me

the strength to persevere through the difficult times

Trang 4

iii

TABLE OF CONTENTS

1.2.1.1.Pharmacogenetics of cytochrome (CYP) P450 enzymes

32 1.2.1.1.1.Cytochrome P450, family 3,

subfamily A, polypeptide 4

(CYP3A4)

32

1.2.1.1.2.Cytochrome P450, family 3, subfamily A, polypeptide 5

(CYP3A5)

35

1.2.2.1.Pharmacogenetics of phase II drug metabolising enzymes

39 1.2.2.1.1.Uridine diphosphate

1.3.1.2.Role of hepatic organic anion transporter polypeptide member 1B1 (OATP1B1) and 1B3 (OATP1B3) in drug disposition

44

1.3.1.3.Pharmacogenetics of influx

transporters

45

Trang 5

drug disposition and chemoresistance

52 1.3.2.2.1.ATP-binding cassette,

subfamily B, member 1 (ABCB1)

subfamily C, member 2 (ABCC2)

subfamily G, member 2 (ABCG2)

54 1.3.2.3.Pharmacogenetics of drug efflux

transporters

subfamily B, member 1 (ABCB1)

subfamily C, member 2 (ABCC2)

subfamily G, member 2 (ABCG2)

1.5.2.Nasopharyngeal cancer chemotherapy and

regimens

79 1.5.3.Taxane-based chemotherapy in metastatic

Trang 6

2.4.3.Amplification of target genes by polymerase chain reactions (PCR) and primer design

2.4.7.Pharmacogenetics of regulatory nuclear receptors

2.4.9.1.Solute carrier organic anion transporter, family member 1B3

(SLCO1B3)

119

2.4.10.1.ATP Binding Cassette, subfamily

B, member 1 (ABCB1)

121 2.4.10.2.ATP Binding Cassette, subfamily

C, member 2 (ABCC2)

122 2.4.10.3.ATP Binding Cassette, subfamily

Trang 7

vi

2.5.1.1.1.Instrumentation and chromatographic conditions

123 2.5.1.1.2.Standard stock solutions,

calibration and quality control samples

124

2.5.1.1.3.Sample preparation and analysis

124 2.5.1.1.4.Estimation of

129

3.1.Patient demographic and clinical characteristics (N = 50) 131

pharmacodynamic-pharmacogenetic correlations

151

3.4.2.1.CAR pharmacogenetics in healthy

derivation

169 3.4.2.2.Pharmacokinetic /

176

Trang 8

derivation

with docetaxel pharmacokinetics

derivation

with docetaxel pharmacokinetics

Trang 9

(ABCB1, ABCC2 and ABCG2)

328

3.7.1.1.ABCB1, ABCC2 and ABCG2

pharmacogenetics in healthy and patient populations

328

3.7.1.2.Pharmacokinetic / pharmacodynamic-pharmacogenetic correlations

333

3.8.2.Demographic and baseline clinical characteristics of pharmacokinetic outliers

347

3.8.3.Pharmacodynamics of pharmacokinetic outliers

348 3.8.4.Investigation of the genetic basis of

Trang 10

ix

SUMMARY

This is a study aimed at elucidating the genetic basis of inherited differences in docetaxel pharmacokinetics and pharmacodynamic in local Chinese nasopharyngeal cancer patients (N = 50) The following genes involved in the biochemical pathways of docetaxel were studied: regulatory nuclear receptors

(PXR, CAR, RXRα, HNF4α), drug metabolism enzymes (CYP3A4, CYP3A5) and drug transporters (ABCB1, ABCC2, ABCG2, SLCO1B3) The docetaxel

pharmacokinetic profile of the patients was characterised by high degree of interindividual variability, with approximately 4- to 6-fold variations observed

in Cmax, AUC0-∞ and CL Grade 3/4 haematological and non-haematological toxicity events across all cycles occurred in 30% of the patients (N = 15) An exposure-toxicity relationship was observed, in which patients with grade 3/4 toxicities exhibited a trend towards higher Cmax, AUC0-∞ and lower CL in the first cycle compared to those who did not have toxicities Wide interpatient difference in percentage decrease of nadir haemoglobin, absolute neutrophil count and platelet from baseline in cycle 1 was also observed

Individual SNP association tests revealed 5 SNPs each in RXRα

[IVS5+288C>T (rs7861779), IVS7+70A>G (rs1536475), IVS8+106A>G

(rs35443779), *846G>A (rs4240711) and *+4458G>A (rs3132291)] and

HNF4α [-728A>C (rs1800963), IVS6+141A>G (rs6103731), IVS7-88T>C

(rs2273618), IVS9+354G>T (rs3818247) and IVS9–145T>C (rs3746574)] to

be significantly correlated with altered docetaxel pharmacokinetics

Subsequent haplotype association analysis has identified the RXRα LD block 2

Trang 11

x

AG haplotype [*+4458G>A (rs3132291) and *+4988A>G (rs4842198)] to be significantly associated with altered pharmacokinetics whereby patients carrying two copies of AG haplotypes had approximately 20% decreased Cmax , AUC0-∞ and increased CL compared to those who carried 0 or 1 copy of the haplotype

With regard to the transporters, we have identified the significant relationship

between SLCO1B3 haplotypes containing four polymorphisms [IVS4+76G>A

pharmacokinetics ABCG2 421C>A (rs2231142; Gly141Lys) was correlated to

decrease in Cmax of docetaxel Notably, this effect was relatively weak

In this study, we also evaluated the effects of the investigated SNPs on the percentage decrease of nadir neutrophil, haemoglobin and platelet from

baseline in cycle 1 The polymorphisms in ABCB1 [2677G>T/A (rs2032582; Ala893Ser/Thr) and 3435C>T (rs1045642; Ile1145Ile)] as well as PXR

(rs6413517) and IVS8+106A>G (rs35443779)], HNF4α [-1623A>G

(rs321272), IVS7-88T>C (rs2273618), IVS9–145T>C (rs3746574) and

*1132C>T (rs6130615)] and SLCO1B3 [-1951T>C (rs1356149 ), -1390A>G

Trang 12

xi

(rs2138334), IVS7-76T>C (rs3829311), *347_*348insA (rs3834935) and

*+1502C>T (rs12581998)] were found to be linked to these parameters Most

of the relationships were however independent of pharmacokinetic changes, possibly to be due to the altered local efflux of the docetaxel in blood cells in the presence of variants

Our data showed that several SNPs and haplotypes in the genes along the biochemical pathway of docetaxel might contribute to the interindividual variability in the pharmacokinetics and pharmacodynamics of docetaxel Apart from pharmacogenetics, environmental factors and clinical characteristics are among the plausible factors that can affect the pharmacokinetic and pharmacodynamic phenotypes Therefore, our results have to be validated in a larger cohort of patients with different backgrounds Genome-wide association studies will provide further insight on the mechanistic basis of the variability

in pharmacokinetics and pharmacodynamics of this agent

Trang 13

EH, Lim WT and Chowbay B Cancer Chemother Pharmacol 67,

1471-1478 (2011)

2. Influence of SLCO1B3 haplotype-tag SNPs on docetaxel disposition in

Chinese nasopharyngeal cancer patients Chew, SC, Sandanaraj E, Singh O,

Chen X, Tan EH, Lim, WT, Lee, EJD and Chowbay B Br J Clin

Pharmacol 73, 606-618 (2012)

3. Genetic variations of NR1I3 and NR2B1 in Asian populations Chew SC,

Lim, J, Singh O, Wong M, Lee, EJD and Chowbay B Drug Metab

Pharmacokinet DOI: 10.2133/dmpk.DMPK-12-SC-060 Epub ahead

of print (2012)

4. Genetic variations of NR2A1 in Asian populations: Implications in

pharmacogenetics studies Chew SC, Lim, J, Lee, EJD and Chowbay B

Drug Metab Pharmacokinet DOI: 10.2133/dmpk.DMPK-12-SH-114 Epub ahead of print (2012)

5. Pharmacogenetic effects of regulatory nuclear receptors (PXR, CAR, RXRα and HNF4α) on docetaxel disposition in Chinese nasopharyngeal cancer

patients Eur J Clin Pharmacol (In press 2013)

Trang 14

xiii

Abstracts

Overseas

1 American Society of Clinical Oncology (ASCO), USA 2010 The effects

of SLCO1B3, ABCG2, MDR1, and CYP3A5 single nucleotide polymorphisms on the pharmacokinetics and pharmacodynamics of docetaxel in nasopharyngeal carcinoma patients Chew SC, Li HH, Singh

O, Lim WT, Tan EH, Lee EJD and Chowbay B

2 European Society of Medical Oncology (ESMO), Italy 2010 The

influence of SLCO1B3 polymorphisms on docetaxel disposition in Asian

nasopharyngeal carcinoma patients Chew SC, Singh O, Ramasamy RD, Lim WT, Tan EH, Lee EJD and Chowbay B

3 American Society of Clinical Oncology (ASCO), USA 2012 Docetaxel

pharmacogenetics: The influence of RXRα and HNF4α genetic variations

on docetaxel disposition in Asian nasopharyngeal carcinoma patients Chew SC, Singh O, Chen X, Lim WT, Tan EH, Lee EJD and Chowbay B

Local

1 Singhealth Duke-NUS Scientific Congress, Singapore 2010 Impact of

SLCO1B3 ht-SNPS on docetaxel pharmacokinetics in Chinese nasopharyngeal carcinoma patients Chew SC, Singh O, Chen X, Ramasamy RD, Lim WT, Tan EH, Lee EJD and Chowbay B

2 Annual Scientific Meeting (ASH), SGH, Singapore 2011 Population

linkage modelling of SLCO1B3 haplotype-tag SNPs (htSNPs) on docetaxel

disposition in Chinese nasopharyngeal cancer patients Chew SC, Sandanaraj E, Singh O, Chen X, Mahajan A, Soh XJ, Ho PY, Osman I and Chowbay B

Trang 15

xiv

LIST OF TABLES

Table 1.1 Substrates and functions of human CYP gene

families

31

expression location and function

50 Table 2.1 List of reagents and chemicals used in the

experiments

95

Table 2.6 Components and their respective volumes in a Sanger

sequencing reaction

106

Table 2.8 Primer details and PCR conditions used for sequence

analysis of PXR gene polymorphisms

108 Table 2.9 Primer details and PCR conditions used for sequence

analysis of CAR gene polymorphisms

analysis of RXRα gene polymorphisms

112 Table 2.11 PCR components and their respectively volumes for

RXRα fragment 3

114

Table 2.13 PCR components and their respectively volumes for

RXRα fragment 4

114

Table 2.15 Sanger cycling sequencing conditions for RXRα

fragment 3 and 4

analysis of HNF4α gene polymorphisms

analysis of CYP3A4 gene polymorphisms

analysis of CYP3A5 gene polymorphisms

analysis of SLCO1B3 gene polymorphisms

analysis of ABCB1 gene polymorphisms

analysis of ABCC2 gene polymorphism

analysis of ABCG2 gene polymorphism

122

Trang 16

xv

Table 3.1 Demographic and clinical characteristics of Chinese

nasopharyngeal cancer patients (N = 50)

132 Table 3.2 Summary of docetaxel pharmacokinetics in Chinese

134 Table 3.3 Summary of percentage decrease of nadir blood

count from baseline in cycle 1 in Chinese

141

Table 3.4 Summary of the incidence of grade 3 & 4 docetaxel

toxicity in Asian nasopharyngeal cancer patients (N =

polymorphisms in local Chinese nasopharyngeal

cancer patients (N = 50)

152

docetaxel in Chinese nasopharyngeal cancer patients

(N = 50)

153

Table 3.8 Genotype and allelic frequencies of CAR

polymorphisms in healthy local populations

(Chinese, Malays, Indians; N = 56 each) and Chinese

163

Table 3.9 In-silico functional predictions of CAR

polymorphisms

171

(N = 50)

177

189

polymorphisms

203

(N = 50)

211

Table 3.14 RXRα LD block haplotypes and diplotypes in

Chinese nasopharyngeal cancer patients (N = 50)

222

Table 3.15 Influence of RXRα haplotypes on docetaxel

pharmacokinetics

225

Table 3.16 Genotype and allelic frequencies of HNF4α

232

Trang 17

xvi

polymorphisms

244

(N = 50)

254

Table 3.19 HNF4α LD block haplotypes and diplotypes in

Chinese nasopharyngeal cancer patients (N = 50)

268

Table 3.20 Influence of HNF4α haplotypes on docetaxel

pharmacokinetics

270

Table 3.21 Genotype and allelic frequencies of CYP3A4 and

CYP3A5 polymorphisms in healthy local

populations* (Chinese, Malays, Indians; N = 56

each) and Chinese nasopharyngeal cancer patients (N

= 50)

280

(N = 50)

281

Table 3.23 Genotype and allelic frequencies of SLCO1B3

284

Table 3.24 In-silico functional predictions of SLCO1B3

polymorphisms

304

Table 3.25 SLCO1B3 polymorphisms and their relationships

with pharmacokinetics and pharmacodynamics of

Table 3.27 Genotype and allelic frequencies of ABCB1, ABCC2

and ABCG2 polymorphisms in healthy local

populations* (Chinese, Malays, Indians; N = 56

each) and Chinese nasopharyngeal cancer patients (N

= 50)

330

Table 3.28 ABCB1 LD block haplotypes and diplotypes in

healthy local populations* and Chinese NPC patients

(N = 50)

331

Table 3.29 ABCB1, ABCC2 and ABCG2 polymorphisms and

their relationships with pharmacokinetics and

nasopharyngeal cancer patients (N = 50)

335

Table 3.30 ABCB1 diplotypes and the relationships with

pharmacokinetics and pharmacodynamics of

docetaxel in Chinese NPC patients (N=50)

339

Trang 18

xvii

Table 3.31 The comparison of pharmacokinetic parameters

between pharmacokinetic outliers (N = 5) and the

remaining patients (N = 50)

346

Table 3.33 Comparison of demographics and baseline

biochemistry test values between pharmacokinetic

outliers (N = 5) and the rest of the patients (N = 50)

348

Table 3.34 Comparison of percentage decrease in nadir blood

counts in first cycle between pharmacokinetic

outliers (N = 5) and the rest of the patients (N = 50)

349

Table 3.35 Comparison of genotype proportion between

pharmacokinetic outliers (N = 5) and the rest of the

patients (N = 50)

350

Trang 19

xviii

LIST OF FIGURES

Figure 1.1 Mechanisms of transcriptional regulation of drug

metabolism enzymes and efflux transporters by regulatory nuclear receptors

6

Figure 2.1 The pharmacogenetic analysis of regulatory nuclear

receptors in healthy subjects

98 Figure 2.2 The pharmacogenetic analysis of drug metabolism

enzymes and transporters in healthy subjects

99

and internal standard (I.S.) paclitaxel

125 Figure 3.1 Median plasma concentration versus time profile in

Chinese nasopharyngeal cancer patients (N = 50)

134

Figure 3.2 Scatter plot of docetaxel AUC0-∞ versus percentage

decrease in absolute neutrophil count (ANC) in the same course

142

Figure 3.3 The relationship between docetaxel (a) Cmax, (b)

AUC0-∞ and (c) CL and overall toxicity outcomes (Whiskers: 5-95 percentile)

145

Figure 3.4 Graphical representations of identified polymorphisms

in the PXR gene

149 Figure 3.5 Graphical representations of identified polymorphisms

in the CAR gene

162 Figure 3.6 (a) Linkage disequilibrium plot of CAR

polymorphisms in healthy Chinese

173

(b) Linkage disequilibrium plot of CAR

polymorphisms in healthy Malays

174

(c) Linkage disequilibrium plot of CAR

polymorphisms in healthy Indians

175 Figure 3.7 Graphical representations of identified polymorphisms

in the RXRα gene

188 Figure 3.8 (a) Linkage disequilibrium plot of RXRα

206

(b) Linkage disequilibrium plot of RXRα

207

(c) Linkage disequilibrium plot of RXRα

208 Figure 3.9 Linkage disequilibrium blocks of investigated RXRα

polymorphisms in Chinese nasopharyngeal cancer patients

221

docetaxel (a) Cmax , (b) AUC0-∞ and (c) CL

226

Trang 20

xix

in the HNF4α gene

231 Figure 3.12 (a) Linkage disequilibrium plot of HNF4α

248

(b) Linkage disequilibrium plot of HNF4α

249

(c) Linkage disequilibrium plot of HNF4α

250 Figure 3.13 Linkage disequilibrium blocks of investigated HNF4α

polymorphisms in Chinese nasopharyngeal cancer patients

267

in the SLCO1B3 gene

283 Figure 3.15 (a) Linkage disequilibrium plot of SLCO1B3

299

(b) Linkage disequilibrium plot of SLCO1B3

300

(c) Linkage disequilibrium plot of SLCO1B3

301 Figure 3.16 SLCO1B3 haplotypes in healthy (a) Chinese, (b)

Malay and (c) Indian populations

302 Figure 3.17 SLCO1B3 htSNPs in the healthy Asian populations (a)

Chinese, (b) Malays and (c) Indians

303

docetaxel clearance

321

(GAG*347wt) in comparison to the wild-type haplotype on docetaxel (a) AUC0-∞ and (b) CL

323

Figure 3.20 Effects of ABCG2 421C>A on docetaxel C

Figure 3.21 Docetaxel plasma concentration versus time profile of

pharmacokinetic outliers (N = 5) and the remaining patients (N = 50)

346

Figure 3.22 Genotypic distribution of HNF4α -728A>C

(rs1800963) in 50 subjects and 5 outliers

352 Figure 3.23 Overall findings on the genetic factors that may

partially account for the wide interpatient variability in docetaxel pharmacokinetics and pharmacodynamics

357

Trang 21

xx

LIST OF ABBREVIATIONS

ABBREVIATIONS

AUC0-∞

Area under the plasma concentration-time curve from time zero

to infinity

CCRP Cytoplasmic constitutive androstane receptor retention protein

Trang 22

xxi

ABBREVIATIONS

Trang 23

xxii

ABBREVIATIONS

SMRT Silencing mediator of retinoic acid and thyroid hormone receptor

Trang 24

1

Chemotherapeutic agents are characterised by narrow therapeutic indices and wide interindividual variability in drug response and toxicity While some individuals have the desired therapeutic effects, others can obtain suboptimal

or no response Moreover, the extent of adverse effects can greatly differ from mild to life-threatening among individuals too In general, the therapeutic and toxic effects of a drug are governed by the drug exposure at the site of action Due to feasibility issue, plasma exposure level is commonly measured as a surrogate for exposure at the target site Plasma exposure level can vary significantly between individuals, which can be attributed to a wide range of factors such as age, gender, comorbidity, liver and renal dysfunction, compliance, drug-drug interaction, environment and lifestyle Notably, genetic variations have been identified as important factors which account for 20 to

95 percent of the variability in drug response and toxicity (Kalow et al., 1998) There are two types of genetic variations, namely somatic and germline genetic variations Germline variations are present in the germ cells and passed on to the offspring, while somatic variations are acquired later in life and can be localized in any cells except germ cells Somatic variations are common in tumour cells and results in a cancer cell genome that differs from the host cell genome (Marsh et al., 2005)

Pharmacogenetics is the study of how genetic variations in the candidate genes

polymorphism refers to the genetic alteration that occurs with a minor allelic frequency of more than 1% in a population, otherwise it is known as mutation

Trang 25

2

These genetic variations can present in the form of a single base change [also known as single nucleotide polymorphism (SNP)], insertions, deletions, tandem repeats, and even whole gene deletions and copy number variants SNP is the most common type of variation; with the occurrence rate of one in every 1000-2000 base pairs (Sachidanandam et al., 2001) The initial screening of the human genome discovered 1.4 million SNP, in which more than 4% of them were located in the coding region of genes (Sachidanandam

et al., 2001) SNPs are known to be in linkage disequilibrium (LD), which refers to the non-random association of alleles at two or more loci at the same chromosome This phenomenon is common across the genome and results in the formation of haplotype blocks that consist of the particular combination of alleles (Gabriel, 2002)

The tendency of a genetic variant in causing functional effects depends on its type and location Firstly, variants in the promoter region can affect the transcriptional activity of the gene, resulting in altered level of gene expression Secondly, genetic alterations in the coding region of a gene can affect protein function Non-synonymous changes refer to alterations that cause either amino acid substitutions (missense), early stop codons (nonsense),

or disruption of reading frames (frame-shift) As proteins are encoded from exonic sequences, non-synonymous variants located in these regions are likely

to result in altered protein structures and functions On the other hand, genetic variants that do not result in a change in the encoded amino acids are termed synonymous variants Although these variants are regarded as silent, their importance has been increasingly recognised due to their functions in

Trang 26

Pharmacokinetics refers to the study of the time course of absorption, distribution, metabolism and excretion of a drug molecule and its metabolites Pharmacodynamics, on the other hand, is a study on the biochemical and physiological effects of a drug molecule and its metabolites The perturbations

in these parameters are the consequences of polygenic traits Alterations in the expression or function of drug metabolism enzymes, drug transporters or drug targets due to the presence of cis-acting genetic polymorphisms have been known to correlate with interindividual and interethnic differences in drug disposition, treatment efficacy and toxicity (Zhou et al., 2008) The transcriptional activation of several key drug metabolism enzymes and drug transporters are in turn governed by regulatory nuclear receptors, which act as master xenosensors (Xu et al., 2005) By virtue of the central role of these master regulators in drug disposition, polymorphisms in the genes encoding these receptors can result in differential expression and function of the

pharmacokinetics and pharmacodynamics

Trang 27

4

1.1 Regulatory nuclear receptors

Nuclear receptors belong to a superfamily comprised of evolutionary related ligand-dependent transcription factors These nuclear receptors respond to exogenous or endogenous ligands and mediate the transcriptional activation of downstream target genes involved in important physiological functions Over the past decade, about 40 receptors have been identified and characterised, which shed light into the many aspects of physiology (Aranda and Pascual, 2001) Some of the receptors do not have known putative ligands, and were termed ‘orphan nuclear receptors’ (Aranda and Pascual, 2001) These receptors remain in this category despite the subsequent characterisation of specific ligands

Typically, nuclear receptors have three main functional domains: an amino terminal domain that contains activation function 1 (AF-1), a highly conserved DNA binding domain (DBD), and a carboxyl terminal ligand binding domain (LBD) that contains dimerisation motif and ligand-dependent transcriptional AF-2 (Edwards, 2000) In the absence of ligand, some nuclear receptors such

as thyroid hormone receptor, retinoic acid receptor, vitamin D3-receptor and peroxisome proliferator-activated receptor (PPAR) constitutively bind DNA and silence the target genes through interaction with co-repressors such as nuclear receptor co-repressor (N-CoR) and silencing mediator of retinoic acid and thyroid hormone receptor (SMRT) that recruit histone deacetylases (HDAC) containing complexes (Chen and Evans, 1995; Horlein et al., 1995; Moehren et al., 2004) Unlike the classical mechanism mentioned above, some unliganded receptors are located in the cytoplasm in association with

Trang 28

5

chaperones such as cytoplasmic constitutive androstane receptor retention protein (CCRP) and heat-shock proteins (HSP) (Pratt and Toft, 1997) Upon ligand binding, the complex undergoes conformational changes The receptor then disassociates from the chaperones and translocates into the nucleus The receptor forms homodimer, or heterodimer, followed by binding to the response element in the promoter region of target gene This triggers a cascade

of molecular activities that initiates chromatin remodelling, such as the recruitment of co-activators with histone acetyltransferases (HAT) activity HATs transfer acetyl groups to the lysine residues of the histones and neutralize the positive charge on it This modulates the relaxation of the chromatin structure, leading to the transcriptional activation of the target gene

Among the wide array of nuclear receptors, pregnane X receptor (PXR), constitutive androstane receptor (CAR), hepatocyte nuclear receptor (HNF4α) and retinoid X receptor alpha (RXRα) are being increasingly recognised for their roles in regulating the transcription of drug metabolism enzymes and drug transporters These master xenosensors detect the presence of drug compounds and initiate the transcription of downstream genes involved in drug disposition In addition, these receptors also cross-talk among themselves Figure 1.1 depicts the mechanism of transcriptional activation by these regulatory nuclear receptors

Trang 29

Figure 1.1 Mechanisms of transcriptional regulation of

enzymes and efflux transporters by regulatory nuclear receptors

1.1.1 Pregnane X receptor (PXR)

Pregnane X receptor (PXR; NR1I

receptor belonging to the NR1I family Several biological functions have been

ascribed for this receptor such as

(Nakamura et al., 2007; Zhai et al., 2007

xenobiotics disposition The human

q13.3 spanning about 40 kbp It encodes for

coding exons and one non

(PXR1-3) are present in the hepatic tissues

Stephen et al., 2004; Tompkins et al., 2008

translates into 434 amino acids Another hepatic mRNA isoform,

distinct 5' region from PXR

isoform was shown to account for 2

possessed similar ability to

(Gardner-Stephen et al., 2004

Mechanisms of transcriptional regulation of drug metabolism

transporters by regulatory nuclear receptors

Pregnane X receptor (PXR)

Pregnane X receptor (PXR; NR1I2) is categorised as an orphan nuclear

receptor belonging to the NR1I family Several biological functions have been

ascribed for this receptor such as tissue homeostasis and energy metabolism

Zhai et al., 2007), in addition to its well-known role in

xenobiotics disposition The human PXR gene is located at chromosome 3q12

q13.3 spanning about 40 kbp It encodes for PXR mRNA consisting of eight

coding exons and one non-coding exon Three mRNA transcript variants

) are present in the hepatic tissues (Hustert et al., 2001; Gardner

Tompkins et al., 2008) The major mRNA, PXR translates into 434 amino acids Another hepatic mRNA isoform, PXR-2, has a

PXR-1, and codes for 473 amino acids This alternative

isoform was shown to account for 2-5% of the total PXR expression, and

possessed similar ability to PXR-1 in inducing the expression of target genes

Stephen et al., 2004; Tompkins et al., 2008) PXR-3 contains an in

mRNA consisting of eight

transcript variants

Gardner-PXR-1

, has a , and codes for 473 amino acids This alternative

5% of the total PXR expression, and

in inducing the expression of target genes

contains an

Trang 30

in-7

frame deletion of 37 amino acids in the ligand-binding domain The lack of ligand binding ability might be responsible for the failure of this transcript in transactivating the target genes (Gardner-Stephen et al., 2004)

PXR is ubiquitously expressed in the xenobiotics detoxification organs, such

as liver, intestine and kidney While the DNA binding domain is highly conserved at 96% similarity, the ligand binding of the receptor is highly

divergent among species The amino acid identity between human PXR and mouse Pxr is only about 80% (Zhang et al., 2004) The cross-species variation

of amino acid identity in the ligand binding domain accounts for the different ligand recognition and activation profile of PXR in both human and mouse (Bertilsson et al., 1998; Xie et al., 2000) PXR binds to a wide range of structurally unrelated exogenous and endogenous substrates Examples of prescription drugs include dexamethasone, paclitaxel, phenobarbital and rifampicin PXR also responds to herbal derivatives such as hyperforin, as well

as pesticides and environmental contaminants Endogenous ligands of PXR include corticosterone, 17a-hydroxypregnenolone, progesterone, estrogen and bile acids (Kliewer et al., 1998; Moore et al., 2000)

PXR functions as a xenosensor which detects the presence of foreign substances and activates the expression of a multitude of enzymes and transporters to abate toxic insults In the absence of exogenous inducers, PXR also mediates the basal transcription of downstream genes (Hustert et al.,

2001) PXR-null mice experienced no CYP3A induction by prototypic mouse CYP3A inducers, although the genetic deficiency of PXR did not alter the

basal CYP3A expression, suggesting the existence of endogenous factors that

Trang 31

8

maintain the basal expression level of important housekeeping genes (Xie et al., 2000) Upon ligand activation, PXR dissociates from the complex consisting of CCRP and heat-shock chaperone HSP90 (Squires et al., 2004) This is followed by nuclear translocation, retinoid X receptor alpha (RXRα) heterodimerisation and recruitment of co-activator molecules of the p160 family, including steroid receptor coactivator-1 (SRC-1) and glucocorticoid receptor-interacting protein-1 (GRIP1) (Synold et al., 2001; Johnson et al., 2006) Synold et al (2001) demonstrated that paclitaxel activated PXR and induced CYP2C8, CYP3A4 and ABCB1 expressions, leading to auto-induction of its own metabolism and efflux (Synold et al., 2001) CYP3A4 is a major enzyme involved in the metabolism of a broad range of clinical drugs Thus, the administration of PXR ligands can lead to accelerated drug clearance and drug-drug interaction This is of clinical importance in view of the promiscuous nature of PXR in binding ligands, and its extensive regulatory effects on majority of the disposition genes

PXR modulates the expressions of a wide range of disposition genes To

characterise the PXR network of gene targets, several in-vitro studies as well

as microarray gene expression profiling analyses have identified Phase I enzymes (CYP2B6, 2C8/9, 3A4/5/7), Phase II enzymes (UGT1A1/3/4/6) and Phase III transporters (ABCB1, ABCC2, ABCC3, ABCG2 and OATP2) as PXR gene targets (Geick et al., 2001; Gerbal-Chaloin et al., 2001; Staudinger

et al., 2001; Synold et al., 2001; Kast et al., 2002; Rosenfeld et al., 2003; Sugatani et al., 2005; Jigorel et al., 2006) The upstream PXR hormone response elements have been identified in the promoter region of these genes

Trang 32

9

For example, the PXR-RXRα heterodimer binds to CYP3A4 or CYP3A5

promoter region at the proximal DR3 [direct repeats of two consensus AG (G/T) TCA separated by three bases] or ER6 [two inverted repeats separated

by 6 nucleotides], depending on the species (Bertilsson et al., 1998) Goodwin

et al (1996) later revealed distal DR3 and ER6 sites that are present at -7k

base pairs upstream in the promoter region of CYP3A4, and concluded that

both the proximal and distal sites are required for transcriptional induction (Goodwin et al., 1999) A response element has also been found in the

upstream enhancer region of ABCB1, where the heterodimer binds to two

consensus AG(G/T)TCA motifs separated by 4 nucleotides (DR4) (Geick et al., 2001)

In light of the central role of PXR in drug disposition and its broad ligand / substrate specificity, functional genetic polymorphisms in the gene encoding PXR may have pronounced impact on drug pharmacokinetics Therefore, the

exploration of the pharmacogenetics influences of PXR may provide insight on

the interindividual variability in the disposition of PXR ligands substrates

1.1.1.1. Pharmacogenetics of PXR

Functional variants in the gene encoding PXR are likely to affect the transactivation potential of the downstream genes, leading to variable basal and/or induced expression levels of gene products Several naturally occurring

exonic variants [V140M (PXR*10), D163G (PXR*11), and A370T)] have been shown to affect basal and/or induced CYP3A transactivation in the in-

vitro setting (Hustert et al., 2001) These missense variants were located

Trang 33

10

within or at the boundary of the putative ligand binding domain of PXR, and

were specific to either Caucasians or Africans with allelic frequencies below

0.02 (Hustert et al., 2001) Another exonic polymorphism PXR P27S (PXR*2)

localized to DNA binding domain was unique to the African American population, with reported allelic frequencies ranging from 0.15 to 0.20 (Hustert et al., 2001; Zhang et al., 2001) Despite its prevalence, this variant did not exhibit significant effects on hepatic CYP3A4 expression compared to wild-type (Hustert et al., 2001; Zhang et al., 2001) Zhang et al (2001) identified a rare Caucasian-specific non-synonymous coding variant, R122Q

(PXR*4), which led to weakened DNA binding affinity and ligand activation

(Zhang et al., 2001) Screening in 205 Japanese individuals identified R98S

(PXR*5) to be present at the allele frequency of 0.0024 Investigation using

electrophoretic mobility shift assay revealed that this variant lacked DNA binding and transactivation properties (Koyano et al., 2004) Healthy screening in the local Asian populations did not detect any of the aforementioned variants, indicating the high degree of conservation in the exon region (Sandanaraj et al., 2008) Due to the rarity of the functional exons

in PXR, the studies hitherto were restricted to in-vitro settings but not in-vivo

phenotype evaluations

A number of studies have also demonstrated the potential pharmacogenetics implications of variants located in non-coding regions For example, the variants -1570C>T (rs3814055) and -298G>A (rs2276706) were correlated with higher erythromycin breath test values after rifampicin treatment, an indicator of higher hepatic CYP3A4 activity As these two SNPs were found

Trang 34

11

to be in high LD, the causative effect was likely to be due to -1570C>T (rs3814055) which was predicted to lie in the nuclear factor-ĸB binding

element in the promoter region of PXR (Zhang et al., 2001) These two SNPs

were found to be common in the Caucasian and Indian populations with frequencies of around 0.30, and lower in Chinese and Malays with frequencies ranging from 0.17 to 0.21 (Zhang et al., 2001; Lamba et al., 2008; Sandanaraj

et al., 2008) Another intronic SNP, -6994T>C (rs2472677) was present in the

putative HNF3β site and associated with lower PXR mRNA expression In concordance to the in-vitro observation, patients harbouring two copies of the

wild-type allele (TT) had higher atazanavir clearance (CL/F) This SNP was present at the frequency of 0.38 in Caucasians (Lamba et al., 2008; Schipani et al., 2010)

In summary, distinct interethnic variabilities in the allelic frequencies and the

in-vitro and/or in-vivo effects of PXR SNPs have been demonstrated Larger

studies in different ethnic groups of patients receiving putative PXR drug substrates are required to validate these results in order to prove their clinical

utility

Trang 35

12

1.1.2. Constitutive androstane receptor (CAR)

Constitutive androstane receptor (CAR; NR1I3) is another orphan nuclear receptor in the same family as PXR It modulates PXR cross-talk and regulates genes involved in drug disposition In addition, it also maintains other housekeeping functions such as bile acids and endocrine homeostasis (Swales

and Negishi, 2004) CAR is located in chromosome one (1q23.3) and spans

across a genomic distance of 8.5 kbp It contains one non-coding exon and

eight coding exons In line with its role in drug disposition, CAR mRNA has

been found to be highly expressed in the liver and kidney (Nishimura et al., 2004) However, a wide range of transcript sizes was identified, indicating the presence of alternatively spliced transcripts (Baes et al., 1994) Approximately

26 splice variants of CAR have been identified so far with tissue-specific

expression and altered function (Arnold et al., 2004; Lamba et al., 2004) Similar to other nuclear receptors, there is a carboxyl-terminal ligand binding domain and an amino-terminal DNA-binding domain CAR forms a heterodimer with RXRα (Suino et al., 2004) The ligand binding domain of CAR is 40% similar to PXR, attributing to the overlapping substrate specificity between both receptors (Willson and Kliewer, 2002) Nevertheless, the range of CAR ligands is not as broad as PXR, owing to the smaller ligand binding pocket in CAR (Suino et al., 2004) Large inter-species divergence in the ligand binding domain was observed, as evidenced by only 70% amino acid identity This contributes to a marked cross-species difference in xenobiotics response (Choi et al., 1997)

Trang 36

13

CAR is considered as a constitutive transcriptional activator due to the independent high basal activity observed in the cell-based reporter (Baes et al., 1994; Choi et al., 1997) Despite this, the identification of CAR endogenous ligands, such as androstane steroidal compounds and bilirubin, and exogenous ligands such as chlorpromazine, phenobarbital, phenytoin, planar hydrocarbon

6-(4-

chlorophenyl)imidazo[2,1-b][1,3]thiazole-5-carbaldehydeO-(3,4-dichlorobenzyl)oxime (CITCO), suggests that CAR can also be regulated by ligands (Sueyoshi et al., 1999; Willson and Kliewer, 2002) There is a high degree of similarity in the compounds that bind and activate PXR and CAR (Moore et al., 2000) For example, phenobarbital and TCPOBOP are activators

of both PXR and CAR (Moore et al., 2000)

Similar to PXR, CAR is sequestered in the cytoplasm with CCRP and HSP90

as a mean of controlling the otherwise constitutive CAR activity (Kobayashi et al., 2003; Swales and Negishi, 2004) The activation mechanism of CAR is novel and can be mediated through direct or indirect ligand binding While CITCO binds directly to human CAR, ligands like phenobarbital and bilirubin

do not bind to CAR directly Instead, they evoke a cascade of phosphorylation leading to CAR translocation without physical contact with CAR (Swales and Negishi, 2004) Studies later identified the role of cAMP-dependent protein kinase A (PKA) and protein phosphatase PP2A signalling pathways in CAR activation (Sidhu and Omiecinski, 1997; Honkakoski and Negishi, 1998) In the nucleus, CAR binds to xenobiotics response element as a heterodimer with RXRα, which leads to the transactivation of genes The role of CAR was first

Trang 37

14

discovered through its involvement in phenobarbital-induced CYP2B gene

(Sueyoshi et al., 1999; Sueyoshi and Negishi, 2001; Sugatani et al., 2005) Subsequent studies revealed that it also modulates various Phase I to III genes involved in drug disposition, such as CYP1A1/2 (Yoshinari et al., 2010), CYP2B6 (Sueyoshi et al., 1999), CYP3A4 (Goodwin et al., 1999), UGT1A1

(Sugatani et al., 2001) and MRP2 (Kast et al., 2002) In UGT1A and CYP2B

genes, CAR binds to the 51 base pairs phenobarbital–responsive enhancer module (PBREM) located at the distal element and consists of two direct repeat motifs (DR-4) The PBREM also responds to other PB-type CAR inducers such as planar hydrocarbon TCPOBOP and chlorpromazine (Honkakoski et al., 1998; Ganem et al., 1999) CAR also shares the common proximal and distal xenobiotics responsive modules with PXR, such as DR3 and ER6 in response to prototypical CYP3A inducers (Sueyoshi et al., 1999) Co-activators facilitating the transactivation activity of CAR include GRIP1, SRC-1 and PPAR-γ co-activator 1 (PGC-1) (Muangmoonchai et al., 2001; Min et al., 2002; Shiraki et al., 2003) The overlapping in ligand specificities and target genes between PXR and CAR suggests the cross-talk between these two receptors in the regulatory network of gene induction, highlighting the complexity underlying the body defence mechanism against xenobiotics insults

A striking interindividual difference of up to 240-fold in the expression level

of CAR mRNA was observed previously (Chang et al., 2003) In concordance,

this observation highly correlated with a 278-fold variability observed in the

mRNA level of CYP2B6, a CAR target gene (Chang et al., 2003) Additionally,

Trang 38

15

CAR induction by phenobarbital led to a markedly different gene expression pattern between Hispanics and Whites in a transcriptome study (Finkelstein et al., 2006) These findings suggest that the existence of differential CAR expression can contribute to altered target gene induction leading to interethnic or interindividual variability in drug disposition Therefore, knowledge of the pattern of genetic variability and ethnic diversity of the genes encoding CAR is necessary to discern the variations in drug disposition

1.1.2.1. Pharmacogenetics of CAR

A total of 26 single nucleotide polymorphisms (SNPs) in CAR were previously

reported from the screening of 74 Japanese diabetes patients (Ikeda et al., 2003) Among the four rare non-synonymous exonic variants [398T>G (Val133Gly), 737A>G (His246Arg), 923T>C (Leu308Pro) and 968A>G

(Asn323Ser)] that mapped to ligand-binding domain of CAR, 737A>G

(His246Arg) and 923T>C (Leu308Pro) resulted in significantly diminished

transactivation activities of CAR (Ikeda et al., 2005) The common exonic

polymorphism 540C>T (rs2307424; Pro180Pro)] located in the ligand binding domain was previously screened in a cohort of local Asian breast cancer patients (Hor et al., 2008; di Masi et al., 2009) The allelic frequencies of the polymorphism in the breast cancer patient population made up of Chinese, Malays and Indians were 0.52-0.64 While this SNP was shown to correlate with worse docetaxel-induced neutropenia in interaction with hepatocyte

nuclear factor 4-alpha (HNF4α) variant IVS1+308 G>A (rs2071197), the

effects of the SNPs located in other regions were not investigated as the study limited the SNP screening to exons and splice-site junctions (Hor et al., 2008)

Trang 39

16

Recently, the variant allele of 540C>T (rs2307424; Pro180Pro) was associated with lower efavirenz plasma concentrations (Swart et al., 2012) Caucasian HIV patients harbouring the homozygous wild-type genotype of this SNP had earlier discontinuation within 3 months of efavirenz therapy compared to others Antiviral efavirenz undergoes extensive phase I metabolism by CYP2B6, which is the main gene target of CAR The findings suggest the potential impact of the exonic SNP on efavirenz pharmacokinetics, probably mediated through the effect of CAR on its downstream target

As mentioned above, the CAR mRNA has several alternatively spliced

isoforms (Arnold et al., 2004; Lamba et al., 2004) Notably, the intronic regions of CAR are rather polymorphic, in comparison to the well-conserved exonic regions, possibly suggesting the susceptibility of these genes to differential splicing effects Although many intronic SNPs have been reported for CAR, not much is known about their biological significance (Ikeda et al., 2003) The variant IVS3-99C>T (rs2502815) was previously reported to be present at the allelic frequency of 0.40 in 548 healthy Japanese postmenopausal women In the same study, this SNP was statistically associated with bone mineral density; however the functional basis is not

known (Urano et al., 2009)

The collinearity between the mRNA levels of CAR and its target genes implicates that the variability in the expression level of the regulatory genes can possibly lead to the altered induction and activity of the target genes (Wortham et al., 2007) Nonetheless, the potential of the genetic variations in this regulatory gene in affecting target gene expression and drug disposition

Định dạng
Số trang	420
Dung lượng	13,68 MB