Top Banner
Involvement of multiple molecular events in pyrogallol-induced hepatotoxicity and silymarin-mediated protection: Evidence from gene expression profiles Ghanshyam Upadhyay a , Manindra Nath Tiwari a , Om Prakash b , Anurag Jyoti a , Rishi Shanker a , Mahendra Pratap Singh a, * a Indian Institute of Toxicology Research (Council of Scientific and Industrial Research), Lucknow 226 001, UP, India b Banaras Hindu University, Varanasi 221 005, India article info Article history: Received 15 February 2010 Accepted 26 March 2010 Keywords: Pyrogallol Silymarin Differential gene expression profiling Hepatotoxicity Hepatoprotection abstract In this study, the involvement of various molecular events in pyrogallol-mediated hepatotoxicity was deciphered by differential mRNA transcription profiles of control and pyrogallol treated mice liver. The modulatory effects of silymarin on pyrogallol-induced differentially expressed transcripts were also looked into. Swiss albino mice were treated with or without pyrogallol. In some set of experiments, mice were also treated with silymarin 2 h prior to pyrogallol. Total RNA was isolated from liver and polyaden- ylated RNA was reverse-transcribed into Cye 3 or Cye 5 labelled cDNA. Equal amounts of labelled cDNA from two different groups were mixed and hybridized with mouse 15 k array. The hybridized arrays were scanned, analyzed and the expression level of each transcript was calculated. The differential expression was validated by quantitative real time polymerase chain reaction. Comparative transcription pattern showed an alteration in the expression of 183 transcripts (150 up-regulated and 33 down-regulated) associated with oxidative stress, cell cycle, cytoskeletal network, cell–cell adhesion, extra-cellular matrix, inflammation, apoptosis, cell-signaling and intermediary metabolism in pyrogallol-exposed liver and silymarin pre-treatment modulated the expression of many of these transcripts. Results obtained thus suggest that pyrogallol induces multiple molecular events leading to hepatotoxicity and silymarin effec- tively counteracts pyrogallol-mediated alterations. Ó 2010 Elsevier Ltd. All rights reserved. 1. Introduction Liver plays a major role in homeostasis as well as in detoxifica- tion of drugs and xenobiotics (Sturgill and Lambert, 1997). Liver dysfunction leads to impairment in dynamic equilibrium of homeostasis resulting in excessive production of reactive oxygen species (ROS), thereby oxidative stress (Upadhyay et al., 2008). Environmental exposures to drugs, pharmaceuticals and toxic sub- stances cause liver damage, one of the major causes of mortality and morbidity. Drug-induced hepatotoxicity has been a major con- cern in modern biology and a number of drugs have been with- drawn from the market due to hepatotoxicity (Bakke et al., 1995). Drug-induced liver injury accounts for 10% cases of hepatitis in adults, 40% of cases in the individuals above the age of 50 years and 25% cases of fulminant liver failure (Lewis and Zimmerman, 1989). Pyrogallol is naturally present as decomposition product of hydrolysable tannins in many plants, such as oak and eucalyptus. Although pyrogallol possesses anti-fungal property and is com- monly used as anti-psoriatic drug, it exhibits hepatotoxicity due to its potential to generate ROS (Shukla et al., 1999; Budavari, 1996; Upadhyay et al., 2008; Cao et al., 1997; Hayakawa et al., 1997; Mochizuki et al., 2002; Akagawa et al., 2003; Gupta et al., 2004). Pyrogallol reduces ferric iron from ferritin and mobilizes the release of iron from ferritin core, which sensitises the cells to oxidative stress (Agrawal et al., 2001; Bauer and Bauer, 2002). Pyrogallol modulates the expression and catalytic activity of some xenobiotic metabolizing genes and antioxidant enzymes (Upadhyay et al., 2007, 2008). Silymarin offers protection against pyrogallol-induced changes in hepatic damage markers, oxidative stress and liver histology (Upadhyay et al., 2007). Silymarin has been selected over other hepatoprotective agents against pyrogallol-induced hepatotoxicity due to its plant origin, antioxidant property, easy availability and most importantly lack of toxicity even at moderate doses (Upadhyay et al., 2007). Although reverse transcriptase-polymerase chain reac- tion (RT-PCR), nuclease protection assay, enzyme-linked immuno- sorbent assay, in situ hybridization and immunohistochemistry 0278-6915/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.fct.2010.03.041 * Corresponding author. Address: Indian Institute of Toxicology Research, Mahatma Gandhi Marg, P.O. Box 80, Lucknow 226 001, UP, India. Tel.: +91 522 2620106/2614869x337; fax: +91 522 2628227. E-mail address: [email protected] (M.P. Singh). Food and Chemical Toxicology 48 (2010) 1660–1670 Contents lists available at ScienceDirect Food and Chemical Toxicology journal homepage: www.elsevier.com/locate/foodchemtox
11
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Food Chem toxicol

Food and Chemical Toxicology 48 (2010) 1660–1670

Contents lists available at ScienceDirect

Food and Chemical Toxicology

journal homepage: www.elsevier .com/locate/ foodchemtox

Involvement of multiple molecular events in pyrogallol-induced hepatotoxicityand silymarin-mediated protection: Evidence from gene expression profiles

Ghanshyam Upadhyay a, Manindra Nath Tiwari a, Om Prakash b, Anurag Jyoti a,Rishi Shanker a, Mahendra Pratap Singh a,*

a Indian Institute of Toxicology Research (Council of Scientific and Industrial Research), Lucknow 226 001, UP, Indiab Banaras Hindu University, Varanasi 221 005, India

a r t i c l e i n f o

Article history:Received 15 February 2010Accepted 26 March 2010

Keywords:PyrogallolSilymarinDifferential gene expression profilingHepatotoxicityHepatoprotection

0278-6915/$ - see front matter � 2010 Elsevier Ltd. Adoi:10.1016/j.fct.2010.03.041

* Corresponding author. Address: Indian InstituMahatma Gandhi Marg, P.O. Box 80, Lucknow 2262620106/2614869x337; fax: +91 522 2628227.

E-mail address: singhmahendrapratap@rediffmail.

a b s t r a c t

In this study, the involvement of various molecular events in pyrogallol-mediated hepatotoxicity wasdeciphered by differential mRNA transcription profiles of control and pyrogallol treated mice liver. Themodulatory effects of silymarin on pyrogallol-induced differentially expressed transcripts were alsolooked into. Swiss albino mice were treated with or without pyrogallol. In some set of experiments, micewere also treated with silymarin 2 h prior to pyrogallol. Total RNA was isolated from liver and polyaden-ylated RNA was reverse-transcribed into Cye 3 or Cye 5 labelled cDNA. Equal amounts of labelled cDNAfrom two different groups were mixed and hybridized with mouse 15 k array. The hybridized arrays werescanned, analyzed and the expression level of each transcript was calculated. The differential expressionwas validated by quantitative real time polymerase chain reaction. Comparative transcription patternshowed an alteration in the expression of 183 transcripts (150 up-regulated and 33 down-regulated)associated with oxidative stress, cell cycle, cytoskeletal network, cell–cell adhesion, extra-cellular matrix,inflammation, apoptosis, cell-signaling and intermediary metabolism in pyrogallol-exposed liver andsilymarin pre-treatment modulated the expression of many of these transcripts. Results obtained thussuggest that pyrogallol induces multiple molecular events leading to hepatotoxicity and silymarin effec-tively counteracts pyrogallol-mediated alterations.

� 2010 Elsevier Ltd. All rights reserved.

1. Introduction

Liver plays a major role in homeostasis as well as in detoxifica-tion of drugs and xenobiotics (Sturgill and Lambert, 1997). Liverdysfunction leads to impairment in dynamic equilibrium ofhomeostasis resulting in excessive production of reactive oxygenspecies (ROS), thereby oxidative stress (Upadhyay et al., 2008).Environmental exposures to drugs, pharmaceuticals and toxic sub-stances cause liver damage, one of the major causes of mortalityand morbidity. Drug-induced hepatotoxicity has been a major con-cern in modern biology and a number of drugs have been with-drawn from the market due to hepatotoxicity (Bakke et al.,1995). Drug-induced liver injury accounts for 10% cases of hepatitisin adults, 40% of cases in the individuals above the age of 50 yearsand 25% cases of fulminant liver failure (Lewis and Zimmerman,1989).

ll rights reserved.

te of Toxicology Research,001, UP, India. Tel.: +91 522

com (M.P. Singh).

Pyrogallol is naturally present as decomposition product ofhydrolysable tannins in many plants, such as oak and eucalyptus.Although pyrogallol possesses anti-fungal property and is com-monly used as anti-psoriatic drug, it exhibits hepatotoxicity dueto its potential to generate ROS (Shukla et al., 1999; Budavari,1996; Upadhyay et al., 2008; Cao et al., 1997; Hayakawa et al.,1997; Mochizuki et al., 2002; Akagawa et al., 2003; Gupta et al.,2004). Pyrogallol reduces ferric iron from ferritin and mobilizesthe release of iron from ferritin core, which sensitises the cells tooxidative stress (Agrawal et al., 2001; Bauer and Bauer, 2002).Pyrogallol modulates the expression and catalytic activity ofsome xenobiotic metabolizing genes and antioxidant enzymes(Upadhyay et al., 2007, 2008).

Silymarin offers protection against pyrogallol-induced changesin hepatic damage markers, oxidative stress and liver histology(Upadhyay et al., 2007). Silymarin has been selected over otherhepatoprotective agents against pyrogallol-induced hepatotoxicitydue to its plant origin, antioxidant property, easy availability andmost importantly lack of toxicity even at moderate doses (Upadhyayet al., 2007). Although reverse transcriptase-polymerase chain reac-tion (RT-PCR), nuclease protection assay, enzyme-linked immuno-sorbent assay, in situ hybridization and immunohistochemistry

Page 2: Food Chem toxicol

G. Upadhyay et al. / Food and Chemical Toxicology 48 (2010) 1660–1670 1661

are used to assay single gene to understand its role in a molecularnetwork, analysis of multiple genes and proteins have been possibleonly after the advent of high throughput genomic and proteomictechnologies (Chavan et al., 2006). Gene expression profiling isused to understand the molecular network, how multiple genes per-form a particular function and how environment influences theexpression patterns of these genes (Ramos and Olden, 2008). Thedifferential expressions of cytochrome P-450 1A2 (CYP1A2),CYP2E1, glutathione-S-transferase-ya (GST-ya) and GST-yc genes,after pyrogallol exposure, are reported by RT-PCR, biochemicalassays and western blot analyses (Upadhyay et al., 2007, 2008).Although biochemical and molecular evidences highlighted the roleof oxidative stress in the initiation and progression of pyrogallol-mediated hepatotoxicity, multiple gene expression profiles mayprovide a simultaneous and coordinated ways to elucidate thebetter picture of the molecular mechanism. Gene expression profil-ing has been consistently used to elucidate the mechanism of toxic-ity of drugs and environmental toxicants (Amin et al., 2002; Waringet al., 2002, 2001a,b; Burczynski et al., 2000). Hepatic cells dynami-cally and rapidly respond to pyrogallol and express complement ofgenes involved in toxicity progression (Upadhyay et al., 2007, 2008;Chavan et al., 2006).

Several studies have been performed to delineate the biochem-ical, cellular and molecular events leading to pyrogallol-inducedhepatic damage (Gupta et al., 2002, 2004; Upadhyay et al., 2007,2008), pyrogallol-induced alterations on hepatic gene expressionprofiles and its mechanism of hepatotoxicity have not yet beendemonstrated. The identification and functional characterizationof differentially expressed genes are expected to help in under-standing the toxic events leading to pyrogallol-induced hepaticdamage and silymarin-mediated protection.

2. Materials and methods

2.1. Chemicals

Agarose, bromophenol blue, tris-base, ethylene-diamine-tetra-acetic acid(EDTA), chloroform, dimethyl sulfoxide (DMSO), DL-dithiothreitol (DTT), ethidiumbromide, ethanol and TRI reagent were purchased from Sigma–Aldrich, St. Louis,USA. Pyrogallol and silymarin were procured from ICN biomedicals, USA. Mouse15 k pre-arrayed slides were obtained from University Health Network, Toronto,Canada. RT-PCR kits and SYBR green master mixture were purchased from MBI Fer-mentas, USA. The cDNA labelling kits were procured from GE Healthcare, Europe.PCR primers were procured from Invitrogen, USA. Other chemicals were of analyt-ical grade and procured locally.

2.2. Animal treatment

Swiss albino male mice (20–25 g) were obtained from the animal colony ofthe Indian Institute of Toxicology Research (IITR), Lucknow. The Institutionalethics committee for use of laboratory animals approved the study. The animalswere treated intraperitoneally with pyrogallol (40 mg/kg), daily for 4 weeks.

Table 1A table showing sequences of forward and reverse primers, primer bank IDs, mid point melt

Name of the gene Primer sequence

BCL2-associated athanogene 4 FP 50AGTGACGGCCCTTCTTRP 50CCGAGGGGTAGTAGC

Caspase 9 FP 50TCCTGGTACATCGAGARP 50AAGTCCCTTTCGCAGA

Complement component 1, s subcomponent FP 50GCTGCTCACGTTTTGGRP 50ACGTTTACTGTAGAGT

Proliferating cell nuclear antigen FP 50TTTGAGGCACGCCTGARP 50GGAGACGTGAGACGA

Tyrosine kinase, non-receptor, 2 FP 50AGGTGCAGCTACAACARP 50GCTCATCCATGACTTG

Some set of animals was pre-treated with silymarin daily (40 mg/kg, daily, i.p.)2 h prior to pyrogallol treatment for 4 weeks. Control animals were treatedwith an equal volume of vehicle (Upadhyay et al., 2007). As pyrogallol and silym-arin were dissolved in 0.1% DMSO and 99.9% normal saline, therefore, thevehicles were treated intraperitoneally, daily with 0.1% DMSO and 99.9% normalsaline.

2.3. Determination of alanine amino transaminase (ALT), aspartate amino transaminase(AST) and bilirubin contents

Blood was collected from the animals anaesthetized under ether. ALT, AST andbilirubin contents in blood serum were measured with commercial kits procuredfrom Merck, India to confirm pyrogallol-induced hepatic damage and silymarin-mediated protection (Upadhyay et al., 2007).

2.4. RNA isolation

Animals were sacrificed via cervical dislocation and liver was dissected out inliquid nitrogen. Total RNA was extracted from the liver of control and treatedanimals using standard procedure. In brief, liver was homogenized using TRI re-agent in ice-cold condition. Chloroform was added to the homogenate (0.2 ml perml of TRI reagent used), mixed, and kept at room temperature for 10 min. Thehomogenate was centrifuged at 12,000g for 15 min at 4 �C. The aqueous phasewas collected in a fresh tube and isopropanol (IPA) was added to it (0.5 mlper ml of TRI reagent used). The mixture was mixed gently and kept at roomtemperature for 10 min. The samples were centrifuged again at 12,000g for10 min at 4 �C. The RNA pellet was washed with 75% ethanol and centrifugedat 7500g for 10 min. The RNA pellet was dried under electric lamp and dissolvedin diethyl pyrocarbonate (DEPC)-treated water (RNase free water). Quantificationand integrity assessment of RNA were performed, as described previously (Patelet al., 2008).

2.5. Labelling of cDNA and microarray experimentation

Microarray experimentation and cDNA labelling were performed, as describedelsewhere (Patel et al., 2008). In brief, incorporating Cye 3/5-labeled dUTP in areaction mixture during reverse transcription made the labelled cDNA. The la-belled cDNA were purified using purification kit and concentrated by speed vacconcentrator. The concentrated cDNA was re-suspended in nuclease free waterand pooled (equal amounts of labelled cDNA from control and treated ones) in asingle tube. The combined labelled cDNA was heated for 5 min at 95 �C andhybridized with mouse 15 k arrays at 42 �C for 18 h. The hybridized slides werestringency washed at room temperature. At least three independent experimentswere performed for each group. Similarly, three dye swap technical replicateexperiments were also performed with aliquots of the same RNA samples to min-imize inconsistencies due to incorporation of dye and also to avoid other technicalproblems. Hybridized 15 k arrays were scanned using a laser scanner under green(550 nm for Cye 3) and red laser (675 nm for Cye 5). GeneTac Integrator 4.0(Genomic Solutions, USA) and Array Vision 8.0 (GE Healthcare, Europe) softwarewere used to analyze the scanned output files. Expression data were filtered usingLOWESS-normalization and background correction. Hierarchical clustering wasperformed through average linkage for identification of genes with similar expres-sion patterns using Cluster_vers_2.11 software and tree view was generated usingTreeView_vers_1.60 software.

DNA microarrays provide large quantities of data that are subject to potentialvariations owing to biological and technical variabilities but such variations weremaximally minimized, as described previously (Singh et al., 2010). In brief, equalnumber of normal and dye swap experiments were performed to minimize theinconsistencies and technical variabilities arising due to incorporation of dyes.

ing temperatures (Tm), size of PCR amplicons and name of genes selected for qRT-PCR.

Tm Size Primer bank ID

ACG30 61.4 196 17975504a1CATC30 60.6

CCTTG30 60.6 109 31560479a1AACAG30 60.5

AGAAA30 61.4 105 21450097a1CTCTGGG30 60.8

TCC30 62.3 135 7242171a1GTCCAT30 63.0

GTATTTC30 60.5 185 30851439a1CGTTTG30 62.3

Page 3: Food Chem toxicol

Table 2A table showing the summary of differentially expressed transcripts with their clone ID, possible association with biological pathways and relative fold changes in expressionlevel. The relative changes are expressed as control (C) versus (vs.) pyrogallol (P), P vs. P + silymarin (S) and C vs. P + S. Negative value indicates down-regulation and positive valueindicates up-regulation.

Clone ID Gene name Fold change

C vs. P P vs. P + S C vs. P + S

Signal transductionH3099G09 Calmodulin-like (Calml4) 4.02 ± 0.30 �1.34 ± 0.17 2.98 ± 0.41H3123H07 NADH dehydrogenase (ubiquinone) 1a subcomplex, 13 (Ndufb13) 3.92 ± 0.32 �1.61 ± 0.30 2.42 ± 0.52H3093E01 Reticulon 4 (Rtn4) 4.89 ± 0.48 �1.12 ± 0.21 4.32 ± 0.87H3121F07 Suppressor of cytokine signaling 6 (Socs6) 8.57 ± 1.32 �3.87 ± 0.91 2.21 ± 0.32H3026B02 Catenin (cadherin associated protein), a-like 1 (ACRP/Ctnnal1) �2.46 ± 0.21 1.92 ± 0.09 �1.27 ± 0.12H3153A07 Striatin, calmodulin binding protein 4 (Strn4) 2.48 ± 0.67 �1.56 ± 0.23 1.58 ± 0.21H3120E01 Cadherin 11 (Cdh11) 2.08 ± 0.18 �2.29 ± 0.32 �1.11 ± 0.07H3117E08 V-raf-leukemia viral oncogene 1 (Raf1) 3.26 ± 0.66 �1.4 ± 0.32 2.31 ± 0.49H3123D01 Synaptotagmin 4 (Syt4) 3.13 ± 0.41 �1.32 ± 0.08 2.36 ± 0.33H3099C02 Adaptor protein complex AP-1, c1 subunit (Ap1g1) 2.61 ± 0.54 �1.43 ± 0.32 1.81 ± 0.21H3041C08 Casein kinase 2, beta polypeptide (Csnk2b) 2.38 ± 0.33 �2.28 ± 0.21 1.04 ± 0.12H3026B07 ADP-ribosylation factor 1 (Arf1) 2.07 ± 0.20 �1.85 ± 0.37 1.11 ± 0.18H3126B12 Protein kinase inhibitor, gamma (Pkig) �2.27 ± 0.30 2.47 ± 0.62 1.09 ± 0.22H3010G07 Adenylate kinase 3 alpha-like 1 (Ak3l1) 2.05 ± 0.27 �2.74 ± 0.55 �1.33 ± 0.17H3033G01 Stathmin-like 3 (Stmn3) 4.43 ± 0.74 �1.17 ± 0.32 3.76 ± 0.52H3043A02 IQ motif containing GTPase activating protein 1 (Iqgap1) 4.26 ± 0.59 �1.44 ± 0.22 2.95 ± 0.77H3075F03 Complement component 1, s subcomponent (C1s) 3.11 ± 0.61 �2.03 ± 0.23 1.53 ± 0.12H3109H01 Leucine-rich and death domain containing (Lrdd) 3.16 ± 0.56 �1.04 ± 0.11 3.02 ± 0.40H3113E07 Related RAS viral (r-ras) oncogene homolog 2 (Rras2) 2.88 ± 0.44 �1.38 ± 0.18 2.07 ± 0.39H3124D04 Caspase 9 (Casp9) 2.94 ± 0.63 �2.22 ± 0.23 1.32 ± 0.10H3131D02 Tyrosine kinase, non-receptor, 2 (Tnk2) 2.69 ± 0.51 �2.01 ± 0.15 1.33 ± 0.23H3005E07 Peroxisomal biogenesis factor 14 (Pex14) �4.74 ± 1.32 2.16 ± 0.67 �2.19 ± 0.36H3007B07 Toll-interleukin 1 receptor (TIR) domain containing adaptor protein (Tirap) 3.73 ± 0.50 �1.27 ± 0.23 2.92 ± 0.71H3007D02 Guanine nucleotide binding protein (G protein), b1 (Gnb1) 4.67 ± 1.11 �2.01 ± 0.11 2.32 ± 0.24H3018F02 Kinesin-associated protein 3 (Kifap3) 2.03 ± 0.20 �1.03 ± 0.11 1.95 ± 0.29H3026A02 Sorting nexin 9 (Snx9) 2.24 ± 0.31 �1.70 ± 0.22 1.31 ± 0.15H3028E02 Syntrophin, acidic 1 (Snta1) 2.08 ± 0.23 �1.70 ± 0.14 1.21 ± 0.09H3092G01 Mitogen-activated protein kinase 13 (Mapk13) 5.82 ± 1.23 �1.53 ± 0.33 3.81 ± 0.67H3079D08 BCL2-associated athanogene 4 (Bag4) �6.19 ± 1.44 8.85 ± 1.91 1.42 ± 0.40

Cell cycleH3107F02 Cyclin-dependent kinase 4 (Crk3; p34/cdk4) �2.09 ± 0.21 4.71 ± 1.01 2.25 ± 0.43H3113H08 Cullin 3 (Cul3) 2.29 ± 0.37 �1.27 ± 0.16 1.80 ± 0.23H3157C01 Prefoldin 1(Pfdn1) 3.24 ± 0.77 �1.32 ± 0.22 2.43 ± 0.93H3097D03 Cyclin-dependent kinase inhibitor 1C (P57) (CDKI; p57Kip2) �7.29 ± 1.59 8.36 ± 1.82 1.14 ± 0.08H3116B09 Anaphase promoting complex subunit 1 (Anapc1/mcpr) 2.79 ± 0.44 �1.39 ± 0.13 2.01 ± 0.33H3150G07 Retinoblastoma binding protein 4 (Rbbp4) 5.83 ± 1.02 �1.31 ± 0.21 4.44 ± 0.83H3152D01 Cyclin D2 (Ccnd2) 2.13 ± 0.21 �1.86 ± 0.13 1.14 ± 0.10

Energy metabolismH3133B05 Cytochrome c oxidase, subunit Vib polypeptide 1 (Cox6b1) �2.34 ± 0.31 1.26 ± 0.23 �1.85 ± 0.25H3065E07 UDP-glucose dehydrogenase (Ugdh) 2.44 ± 0.55 �1.66 ± 0.10 1.46 ± 0.07H3027E07 Enolase 1, alpha non-neuron (Eno1) 2.46 ± 0.39 �2.11 ± 0.23 1.16 ± 0.05H3153F01 Succinate dehydrogenase complex, subunit A, flavoprotein (Fp) (Sdha) 6.71 ± 1.23 �3.34 ± 0.99 2.00 ± 0.32H3027E09 Enolase 1, alpha non-neuron (Eno1) 2.87 ± 0.29 �2.12 ± 0.29 1.35 ± 0.22H3097B08 Protein disulfide isomerase associated 3 (Pdia3) 2.36 ± 0.52 1.06 ± 0.12 2.50 ± 0.45H3124H01 NADH dehydrogenase (ubiquinone) 1b subcomplex, 10 (Ndufb10) 2.91 ± 0.49 1.27 ± 0.24 3.71 ± 0.46H3032C01 Cytochrome c, somatic (Cycs) 2.25 ± 0.30 �2.21 ± 0.18 1.01 ± 0.07H3096A03 Cytochrome c oxidase subunit IV isoform 1 (Cox4i1) 6.31 ± 1.12 �3.4 ± 0.16 1.85 ± 0.17H3013H02 Cytochrome c oxidase, subunit VIIc (Cox7c) 2.48 ± 0.27 �2.02 ± 0.65 1.22 ± 0.23

MetabolismH3109F04 Fumarylacetoacetate hydrolase (Fah) �2.61 ± 0.33 8.30 ± 1.39 3.18 ± 0.65H3114C07 Bmi1 polycomb ring finger oncogene (Bmi1) 2.38 ± 0.50 1.16 ± 0.08 2.76 ± 0.78H3124A07 Eukaryotic translation initiation factor 4E member 2 (Eif4E2) 2.19 ± 0.44 �2.29 ± 0.33 �1.04 ± 0.11H3124H02 Ring finger protein 11 (Rnf11) 3.72 ± 0.66 �6.21 ± 1.20 �1.67 ± 0.21H3016E02 Stem-loop binding protein (Slbp) 2.71 ± 0.32 �1.65 ± 0.11 1.63 ± 0.16H3019D09 Histidine triad nucleotide binding protein 1 (Hint1) 4.05 ± 0.67 �1.94 ± 0.23 2.07 ± 0.35H3043A10 Apolipoprotein A-I (Apoa1) 3.53 ± 0.98 �1.41 ± 0.24 2.49 ± 0.55H3128F07 Ring finger protein 4 (Rnf4) 3.97 ± 0.63 �2.02 ± 0.40 1.95 ± 0.30H3135D02 Lysosomal-associated membrane protein 2 (Lamp2) 3.32 ± 0.41 �2.08 ± 0.25 1.59 ± 0.13H3116D07 Membrane-bound transcription factor peptidase, site 1 (Mbtps1) 2.74 ± 0.33 �1.09 ± 0.09 2.51 ± 0.39H3117H07 Adenosine monophosphate deaminase 2 (isoform L) (Ampd2) 3.66 ± 0.47 �1.51 ± 0.18 2.41 ± 0.37H3118A07 Selenoprotein W, muscle 1 (Sepw1) 2.87 ± 0.44 �1.78 ± 0.24 1.61 ± 0.19H3118G07 Eukaryotic translation elongation factor 1b2 (Eef1b2) 4.26 ± 0.89 �1.04 ± 0.06 4.08 ± 0.78H3129A01 Aldehyde dehydrogenase family 1, subfamily A1 (Aldh1a1) 2.63 ± 0.23 �4.57 ± 0.68 �1.73 ± 0.21H3129C02 Eukaryotic translation elongation factor 2 (Eef2) �3.28 ± 0.55 3.64 ± 0.47 1.11 ± 0.11H3129G02 Serine palmitoyltransferase, long chain base subunit 1 (Sptlc1) 3.94 ± 0.88 �7.03 ± 1.13 �1.78 ± 0.10H3131H01 Rho GDP dissociation inhibitor (GDI)a (Arhgdia) 3.74 ± 0.64 �2.47 ± 0.33 1.51 ± 0.23H3133D01 Deoxyribonuclease II a (Dnase2a) 3.98 ± 0.70 �3.69 ± 0.60 1.07 ± 0.09H3145A08 Aldo–keto reductase family 1, member A4 (aldehyde reductase) (Akr1a4) 4.61 ± 0.94 �2.23 ± 0.41 2.06 ± 0.32

1662 G. Upadhyay et al. / Food and Chemical Toxicology 48 (2010) 1660–1670

Page 4: Food Chem toxicol

Table 2 (continued)

Clone ID Gene name Fold change

C vs. P P vs. P + S C vs. P + S

H3140B02 Ubiquitin specific peptidase 10 (Usp10) 3.21 ± 0.37 �2.93 ± 0.26 1.09 ± 0.15H3141A01 Phosphoribosylglycinamide formyltransferase (Gart) 3.07 ± 0.34 �1.49 ± 0.14 2.05 ± 0.27H3145G02 Prohibitin (Phb) 4.04 ± 0.48 �1.74 ± 0.13 2.31 ± 0.32H3154F02 Asparagine synthetase (Asns) �2.44 ± 0.35 2.71 ± 0.19 1.11 ± 0.08H3157B07 Adenosine deaminase, RNA-specific, B1 (Adarb1) �3.01 ± 0.22 1.51 ± 0.21 �1.98 ± 0.21H3097F02 ST8 alpha-N-acetyl-neuraminide alpha-2,8-sialyltransferase 2 (St8sia2) �3.07 ± 0.79 4.43 ± 0.76 1.44 ± 0.16H3115C01 Carbonic anhydrase 8 (Car8) 2.53 ± 0.25 1.13 ± 0.23 2.86 ± 0.53H3123D05 Liver glycogen phosphorylase (Pyg1) 3.16 ± 0.66 �1.52 ± 0.11 2.06 ± 0.20H3047A08 3-phosphoglycerate dehydrogenase (Phgdh) 7.94 ± 1.41 �10.28 ± 2.01 �1.29 ± 0.12H3049D01 SUMO/sentrin specific peptidase 6 (Senp6) 4.12 ± 1.01 �1.52 ± 0.25 2.71 ± 0.34H3134B06 Protease (prosome, macropain) 26S subunit, ATPase 1 (Psmc1) �2.25 ± 0.23 2.75 ± 0.38 1.22 ± 0.07H3146D01 Phosphorylase kinase alpha 2 (Phka2) �2.24 ± 0.19 1.80 ± 0.22 �1.23 ± 0.20H3002B07 Sepiapterin reductase (Spr) 3.66 ± 0.66 2.68 ± 0.32 9.81 ± 1.89H3002F01 Heterogeneous nuclear ribonucleoprotein D-like (Hnrpd1) 2.29 ± 0.17 �1.67 ± 0.16 1.36 ± 0.18H3007G08 Histidyl-tRNA synthetase (Hars) �3.46 ± 0.43 �3.01 ± 0.32 �10.4 ± 1.79H3024B03 Chaperonin subunit 3 (gamma) (Cct3) 2.15 ± 0.22 �1.67 ± 0.20 1.28 ± 0.12H3017F07 Hypoxanthine guanine phosphoribosyl transferase 1 (Hprt1) 2.33 ± 0.25 �1.61 ± 0.21 1.44 ± 0.15H3011D07 Seryl-aminoacyl-tRNA synthetase (Sars) 3.86 ± 0.53 1.03 ± 0.06 3.98 ± 0.66H3019H07 Alkaline phosphatase, liver/bone/kidney (Alp/Akp2) 2.56 ± 0.35 �2.13 ± 0.23 1.19 ± 0.15H3059F01 Ubiquitin carboxy-terminal hydrolase L1 (Uchl1) 5.78 ± 0.99 �1.75 ± 0.27 3.29 ± 0.29

TransportH3081F02 Eukaryotic translation initiation factor 4E nuclear import factor 1 (Eif4enif1) �2.61 ± 0.26 1.39 ± 0.20 �1.87 ± 0.18H3027D08 Nucleoporin like 1 (Nupl1) 2.37 ± 0.36 �1.61 ± 0.26 1.47 ± 0.17H3058D07 Matrix metallopeptidase 23 (Mmp23) 2.21 ± 0.17 1.58 ± 0.19 3.51 ± 0.46H3077B02 Solute carrier family 12, member 2 (Slc12a2) �2.15 ± 0.28 �1.07 ± 0.10 �2.31 ± 0.25H3122G01 Sterol carrier protein 2, liver (Scp2) 2.43 ± 0.36 �5.76 ± 1.03 �2.37 ± 0.34H3123A01 Haemoglobin alpha, adult chain 1 (Hba1) 2.58 ± 0.22 �1.41 ± 0.16 1.82 ± 0.20H3125H07 Haemoglobin alpha, adult chain 1 (Hba1) 2.68 ± 0.20 �1.17 ± 0.10 2.28 ± 0.30H3126E02 Voltage-dependent anion channel 1 (Vdac1) 3.58 ± 0.40 �1.37 ± 0.11 2.59 ± 0.21H3142G07 Secretory carrier membrane protein 2 (Scamp2) 7.02 ± 1.33 �1.93 ± 0.34 3.63 ± 0.51H3125F07 Translocase of inner mitochondrial membrane 17b (Timm17b) 2.72 ± 0.23 �1.88 ± 0.18 1.44 ± 0.14H3007H01 Down syndrome critical region gene 3 (Dscr3) �3.12 ± 0.71 3.40 ± 0.60 1.09 ± 0.09

Regulatory genesH3054F07 Hepatoma up-regulated protein (Hurp-pending) 2.25 ± 0.41 �1.51 ± 0.31 1.48 ± 0.12H3105G07 Bone morphogenetic protein receptor, type 1A (Bmpr1a) 5.87 ± 0.85 �2.07 ± 0.33 2.82 ± 0.54H3056A07 Nucleotide binding protein 1 (Nubp1) 2.34 ± 0.33 �1.31 ± 0.13 1.77 ± 0.13H3073F07 Splicing factor proline/glutamine rich (polypyrimidine tract binding protein associated) (Sfpq) 2.11 ± 0.31 �1.39 ± 0.08 1.51 ± 0.24H3143B08 Eukaryotic translation initiation factor 4, c1 (Eif4g1) 2.46 ± 0.42 �1.69 ± 0.21 1.45 ± 0.10H3139C01 Requiem (Req) �3.77 ± 0.64 1.72 ± 0.32 �2.27 ± 0.20H3108D07 Zinc finger protein 35 (Gfp35) 2.26 ± 0.26 2.58 ± 0.22 5.85 ± 0.99H3125C01 Hepatoma-derived growth factor, related protein 2 (Hdgfrp2) 7.56 ± 1.56 �2.15 ± 0.34 3.51 ± 1.05H3158F07 GATA binding protein 4 (Gata4) �2.92 ± 0.44 6.96 ± 1.15 2.38 ± 0.12H3128H12 Single-stranded DNA binding protein 3 (Ssbp3) 2.06 ± 0.19 �1.48 ± 0.17 1.38 ± 0.15H3021F12 Proliferating cell nuclear antigen (Pcna) 3.45 ± 0.58 1.47 ± 0.12 4.63 ± 0.23H3119F02 Transcription factor like 1/vacuolar protein sorting 72 (yeast) (Tcfl1/Vps72) �3.28 ± 0.45 3.72 ± 0.55 1.13 ± 0.16H3002E01 Selenoprotein X 1 (Sepx1) 2.35 ± 0.30 �1.48 ± 0.12 1.57 ± 0.14H3004F07 Protein tyrosine phosphatase 4a2 (Ptp4a2/prl-2) 2.26 ± 0.25 �1.75 ± 0.30 1.29 ± 0.20H3008B02 Nik related kinase (Nrk) �4.87 ± 0.63 3.30 ± 0.40 �1.47 ± 0.12H3015B01 Kruppel-like factor 4 (gut) (Klf4) 2.71 ± 0.23 �2.04 ± 0.21 1.32 ± 0.14H3019B01 ATPase, Na+/K+ transporting, b 1 polypeptide (Atp1b1) 4.12 ± 0.58 1.24 ± 0.27 5.13 ± 1.07H3021D01 Ciliary neurotrophic factor (Zfp91) 4.91 ± 0.79 �3.15 ± 0.56 1.55 ± 0.34H3033B01 Rogdi homolog (Drosophila) (Rogdi) 3.66 ± 0.47 �8.39 ± 1.22 �2.29 ± 0.18H3051F01 Ankyrin repeat domain 6 (Ankrd6) 3.14 ± 0.39 �1.28 ± 0.13 2.44 ± 0.33

HomeostasisH3120F01 Synaptic vesicle glycoprotein 2 a (Sv2a) 2.99 ± 0.29 �3.75 ± 0.48 �1.25 ± 0.19H3092H08 RalBP1 associated Eps domain containing protein (Reps1) 2.47 ± 0.34 �1.35 ± 0.17 1.82 ± 0.22

Structural genesH3012B02 Ribosomal protein L23 (Rpl23) 2.13 ± 0.22 �1.50 ± 0.13 1.41 ± 0.12H3139C02 Adducin 1 (alpha) (Add1) 4.29 ± 0.65 �3.12 ± 0.49 1.37 ± 0.15H3013F03 Ribosomal protein L23 (Rpl23) 3.50 ± 0.46 �1.28 ± 0.20 2.73 ± 0.23H3021B02 Keratin 18 (Krt18) 4.11 ± 0.68 �1.62 ± 0.14 2.53 ± 0.33H3036B09 Mitochondrial ribosomal protein S25 (Mrps25) 2.21 ± 0.27 1.63 ± 0.22 3.61 ± 0.78H3119G01 Vimentin (Vim) 4.81 ± 0.96 �11.99 ± 2.32 �2.49 ± 0.43H3075G08 Ribosomal protein SA (Rpsa) 2.16 ± 0.22 3.33 ± 0.32 7.19 ± 1.02H3079A02 Ribosomal protein L12 (Rpl12) 2.89 ± 0.37 �1.59 ± 0.21 1.81 ± 0.18H3126B01 Ribosomal protein L31 (Rpl31) 2.11 ± 0.23 �2.25 ± 0.26 �1.06 ± 0.09H3119H03 Procollagen, type I, alpha 1 (Coll a1) 2.07 ± 0.26 2.26 ± 0.22 4.68 ± 0.79H3124E08 Troponin T1, skeletal, slow (Tnnt1) �2.61 ± 0.33 3.50 ± 0.15 1.34 ± 0.17H3001E01 Kinesin family member 5B (Kif5b) 7.09 ± 1.38 �10.35 ± 1.23 �1.45 ± 0.19H3030E08 Ribosomal protein L26 (Rpl26) �6.33 ± 1.45 5.10 ± 1.01 �1.23 ± 0.23H3141A09 T-cell immunomodulatory protein (Cda08-pending) 2.87 ± 0.22 �1.89 ± 0.11 1.51 ± 0.24H3028D02 Ribosomal protein L5 (Rpl5) �2.94 ± 0.39 2.37 ± 0.33 �1.23 ± 0.21

(continued on next page)

G. Upadhyay et al. / Food and Chemical Toxicology 48 (2010) 1660–1670 1663

Page 5: Food Chem toxicol

Table 2 (continued)

Clone ID Gene name Fold change

C vs. P P vs. P + S C vs. P + S

H3116A03 Similar to ribosomal protein L10a (Rpl10a) 4.24 ± 0.65 �2.73 ± 0.25 1.55 ± 0.16H3028D03 Ribosomal protein L5 (Rpl5) 2.68 ± 0.22 1.58 ± 0.30 4.23 ± 0.67H3030G02 Discoidin, CUB and LCCL domain containing 2 (Dcbld2) �2.07 ± 0.16 2.25 ± 0.20 1.08 ± 0.08H3112B02 Ribosomal protein S12 (Rps12) �4.01 ± 0.73 2.45 ± 0.43 �1.63 ± 0.23H3015H04 Ribosomal protein L8 (Rpl8) 2.08 ± 0.20 1.91 ± 0.19 3.97 ± 0.66H3030E07 Ribosomal protein L26 (Rpl26) 7.31 ± 0.99 �7.96 ± 1.32 1.08 ± 0.11H3084H01 Ribosomal protein S20 (Rps20) 3.21 ± 0.89 �1.64 ± 0.21 1.94 ± 0.23H3058E07 Microtubule-associated protein 4 (Mtap4) 5.25 ± 1.01 �2.40 ± 0.25 2.18 ± 0.22H3014G03 Nucleosome assembly protein 1-like 1 (Nap1l1) 4.37 ± 0.69 �1.81 ± 0.20 2.40 ± 0.31

StressH3139E01 Heat shock protein 8 (Hspa8) 3.12 ± 0.44 �10.01 ± 1.59 �3.21 ± 0.32H3023C07 Peroxiredoxin 6 (Prdx6) 2.89 ± 0.27 �1.79 ± 0.22 1.61 ± 0.22H3007F12 Peroxiredoxin 1 (Prdx1) 3.81 ± 0.56 �2.27 ± 0.25 1.67 ± 0.23H3012A07 Heat shock protein 8 (Hspa8) 3.71 ± 0.32 �2.99 ± 0.49 1.23 ± 0.14H3023G01 Heat shock protein 90, alpha (cytosolic), class A member 1 (Hsp90 aa1) 4.01 ± 0.44 �2.09 ± 0.27 1.91 ± 0.31H3140D07 Peroxiredoxin 3 (Prdx3) 4.88 ± 0.88 �2.09 ± 0.24 2.33 ± 0.33

Immune responseH3019C07 Macrophage migration inhibitory factor (Mif/Glif) �3.21 ± 0.30 1.54 ± 0.21 �2.07 ± 0.27H3085C02 Traf and Tnf receptor associated protein (Ttrap) 4.86 ± 0.63 �1.39 ± 0.12 3.49 ± 0.45H3108F08 Programmed cell death 10 (Pdcd10) 2.56 ± 0.33 �1.34 ± 0.20 1.89 ± 0.17H3051H08 B-cell translocation gene 4 (Btg4) 4.87 ± 0.74 �1.03 ± 0.10 4.72 ± 0.69

Protein folding and modificationH3033G03 FK506 binding protein 4 (Fkbp4) 2.24 ± 0.25 �1.50 ± 0.16 1.48 ± 0.09H3070A07 Peptidylprolyl isomerase (cyclophilin)-like 4 (Ppil4) 3.30 ± 0.50 �1.83 ± 0.21 1.79 ± 0.23H3124H08 Ubiquitin C (Ubc) 2.38 ± 0.24 �1.44 ± 0.13 1.65 ± 0.17H3003D07 Ubiquitin B (Ubb) 2.49 ± 0.24 �1.55 ± 0.14 1.61 ± 0.26H3083B01 Ubiquitination factor E4B, UFD2 homolog (S. cerevisiae) (Ube4b) 3.55 ± 0.35 �2.57 ± 0.29 1.38 ± 0.18

OthersH3052F02 Dynein, axonemal, light chain 4 (Dnalc) 2.38 ± 0.40 �1.74 ± 0.30 1.36 ± 0.17H3065F07 Tumor endothelial marker 7 related precursor (Tem7r-pending) 2.09 ± 0.31 �1.06 ± 0.12 1.96 ± 0.32H3004H08 Alpha-2-macroglobulin (A2 m; A2mp) 2.62 ± 0.22 �7.22 ± 1.23 �2.75 ± 0.38H3016B08 Esterase D/formylglutathione hydrolase (Esd) 4.41 ± 0.87 �1.69 ± 0.17 2.60 ± 0.36H3018A07 Protein phosphatase 4, catalytic subunit (Ppc4c) �3.28 ± 0.29 3.35 ± 0.40 1.02 ± 0.06H3018G01 DEAD/H (Asp-Glu-Ala-Asp/His) box polypeptide 3, X-linked (ddx3x) 2.05 ± 0.16 �3.44 ± 0.56 �1.68 ± 0.22H3021D07 Prolactin family 7, subfamily d, member 1 (Prl7d1) 3.66 ± 0.44 �1.66 ± 0.24 2.19 ± 0.25H3058H01 20–50 oligoadenylate synthetase 1C (Oas1c) 4.77 ± 0.43 1.16 ± 0.11 5.54 ± 1.07H3108A01 Interferon-stimulated protein (Isg20) 2.03 ± 0.25 �1.91 ± 0.20 1.06 ± 0.10H3114A08 Eukaryotic translation initiation factor 1A (Eif1a) �3.43 ± 0.87 3.49 ± 0.42 1.01 ± 0.09

Unknown functionsH3005C07 Ubiquitin-associated protein 2 (Ubap2) 3.72 ± 0.65 �1.65 ± 0.22 2.25 ± 0.26H3120E02 Integral membrane protein 2B (Itm2b) 3.45 ± 0.39 �1.49 ± 0.22 2.31 ± 0.31H3017G01 TWIST neighbour (Twistnb) 4.57 ± 0.91 �1.76 ± 0.32 2.59 ± 0.37H3014B02 Coiled-coil-helix-coiled-coil-helix domain containing 2/ethanol induced 6 (Chchd2/Etohi6) �3.01 ± 0.81 1.93 ± 0.32 �1.55 ± 0.15H3117C08 Clathrin, heavy polypeptide (Hc) (Cltc) 2.99 ± 0.11 2.00 ± 0.20 5.98 ± 0.97H3123F07 Prostate tumor over expressed gene 1 (Ptov1) 9.42 ± 1.52 �3.30 ± 0.33 �3.32 ± 0.32H3102E01 Potassium channel modulatory factor 1 (Kcmf1) 5.21 ± 0.95 �1.50 ± 0.13 3.45 ± 0.43H3141G10 X-linked lymphocyte-regulated complex (Xlr) �2.38 ± 0.26 �1.92 ± 0.15 �4.56 ± 0.67H3001F07 Erythroblast membrane-associated protein (Ermap) 4.61 ± 0.46 1.81 ± 0.18 8.35 ± 1.83H3106E01 Similar to Zinc finger protein 422, related sequence 1 (Krox-25–2) 2.58 ± 0.25 �3.11 ± 0.36 �1.21 ± 0.12H3074A07 Preferentially expressed antigen in melanoma like 5 (Prame l5) 5.01 ± 0.51 �1.59 ± 0.15 3.14 ± 0.31H3085G07 Trans-golgi network protein (Tgoln1) 7.31 ± 0.98 �2.46 ± 0.24 2.96 ± 0.29H3023G02 Proteolipid protein 2 (Plp2) 2.11 ± 0.21 1.65 ± 0.17 3.48 ± 0.38H3079D07 RB1-inducible coiled-coil 1 (Rb1cc1) 2.68 ± 0.30 �1.60 ± 0.16 1.66 ± 0.16H3010H07 Pellino 1 (Peli1) �4.02 ± 0.42 6.37 ± 0.63 1.58 ± 0.15H3097G07 Retinoblastoma binding protein 9 (Rbbp9) 7.11 ± 1.71 �1.92 ± 0.20 3.68 ± 0.40

1664 G. Upadhyay et al. / Food and Chemical Toxicology 48 (2010) 1660–1670

Spots were individually quantified and local backgrounds were calculated fromthe corners between the spots. Signal intensity for each spot was determinedby subtracting the background from intensity. The fold changes for the differen-tially expressed genes were calculated from the ratios of intensities of the twotest groups. Results were further validated using GeneTac Integrator Version4.0 software. Genes showing consistency in the patterns of expression betweennormal and dye swap experiments were selected (Singh et al., 2010).

2.6. Functional annotation

Possible involvement of molecular pathways was assessed by datasets obtainedfor some of the differentially expressed genes (Dahlquist et al., 2002). Inference for

functional annotation of genes was obtained by considering the informationavailable at NIA (http://Igsun.grc.nia.nlm.gov), NCBI (http://ncbi.nlm.nih.gov/),SOURCE (http://smd.stanford.edu/cgi-bin/source/sourceBatchSearch), and Swiss-Prot (http://www.expasy.org/sprot/) databases.

2.7. Real time PCR

The expressions of some transcripts were validated using quantitative RT-PCR(qRT-PCR). Primers were synthesized and PCR amplifications of Bag4, Casp9, C1s,proliferating cell nuclear antigen (PCNA), Tnk2 and a housekeeping gene GAPDHwere performed. Primer sequences for above-mentioned genes were retrievedfrom the online PrimerBank database (http://pga.mgh.harvard.edu/primerbank/

Page 6: Food Chem toxicol

Fig. 1. Clustered display of differentially expressed transcripts of mouse liver after pyrogallol and pyrogallol + silymarin treatment for 4 weeks. (a) In the first cluster (fromleft), the differentially expressed transcripts after pyrogallol treatment as compared with control are displayed, in the intermediate cluster, the differentially expressedtranscripts after pyrogallol + silymarin treatment are arranged as compared with control and in the third cluster (third from left and first from right), the differentiallyexpressed transcripts after pyrogallol + silymarin treatment as compared with pyrogallol are arranged. Only those transcripts were selected for clustering, whose differentialexpression level was at 2.0-fold in pyrogallol treated group as compared with control. Red indicates up-regulated and green represents down-regulated transcripts. Thetranscript/clone/gene IDs are also mentioned in the figure. (b) In the first cluster (from left), the differentially expressed transcripts after pyrogallol treatment as comparedwith control are displayed, in the second cluster, the differentially expressed transcripts after pyrogallol + silymarin treatment are arranged as compared with pyrogallol.Only those transcripts were selected for clustering, whose differential expression level was at 2.0-fold in both comparison groups. Red indicates up-regulated and greenrepresents down-regulated transcripts. The transcript/clone/gene ID are also mentioned in the figure. (For interpretation of the references to colour in this figure legend, thereader is referred to the web version of this article.)

G. Upadhyay et al. / Food and Chemical Toxicology 48 (2010) 1660–1670 1665

Page 7: Food Chem toxicol

(a)

(b)

-10

-5

0

5

10

15

20

25

30

35

Up regulated Down regulated

No

of g

enes

Signal TransductionCell Cycle

Energy MetabolismMetabolismTransportRegulatory Genes

HomeostasisStress genesImmune ResponseProtein Folding and Modification

Structural genesGenes with unknown functions

16%

4%

5%

22%

6%11%1%

3%

2%

3%

13%

14%Signal TransductionCell CycleEnergy MetabolismMetabolismTransportRegulatory GenesHomeostasisStress genesImmune ResponseProtein Folding and ModificationStructural genesGenes with unknown functions

(c) 14%

3%

8%

24%

3%11%

1%

6%

1%

18%

11%

Signal TransductionCell CycleEnergy Metabolism

Metabolism

TransportRegulatory GenesHomeostasisStress genesProtein Folding and Modification

Structural genesGenes with unknown functions

Fig. 2. Pie diagram showing the % representation of various pathways among differentially expressed transcripts and number of up- and down-regulated genes of eachpathway. (a) and (b) represent percentages and number (both up- and down-regulated) respectively for differentially expressed genes after pyrogallol treatment in mouseliver as compared with control; similarly (c) and (d) represent pyrogallol + silymarin treatment as compared with pyrogallol and (e) and (f) represent pyrogallol + silymarintreatment as compared with control.

1666 G. Upadhyay et al. / Food and Chemical Toxicology 48 (2010) 1660–1670

index.html) (also shown in Table 1). RNA (2–4 lg) was reverse-transcribed usingoligo-dT primers. The qRT-PCR was carried out using cDNA (100 ng), primers(400 nM each) and SYBR green master mixture (25 ll total). Level of qRT-PCRproduct was measured using SYBR green fluorescence in an iCycler, BIO-RADDetection System (Bio-Rad, Hercules, CA, USA). The qRT-PCR results were ana-

lyzed by iCycle iQ real-time detection system software version 3.0A. A cyclethreshold (Ct) was assigned at the beginning of logarithmic phase of PCR ampli-fication and difference in Ct values of control and treated groups were used todetermine the relative expressions. Each reaction was subjected to melting pointanalysis to confirm single amplified products (Singh et al., 2010).

Page 8: Food Chem toxicol

(e)

(f)

-5

0

5

10

15

20

Up regulated Down regulated

No

of g

enes

17%

5%

4%

19%

7%11%

2%

4%

12%

19%

Signal TransductionCell Cycle

Energy MetabolismMetabolismTransportRegulatory Genes

HomeostasisStress genesImmune ResponseStructural genesGenes with unknown functions

Signal TransductionCell Cycle

Energy MetabolismMetabolismTransportRegulatory Genes

HomeostasisStress genesImmune ResponseStructural genesGenes with unknown functions

(d)

-15

-10

-5

0

5

10

Up regulated Down regulated

No

of g

enes

Signal TransductionCell Cycle

Energy MetabolismMetabolismTransportRegulatory Genes

HomeostasisStress genesProtein Folding and Modification

Structural genesGenes with unknown functions

Fig. 2 (continued)

G. Upadhyay et al. / Food and Chemical Toxicology 48 (2010) 1660–1670 1667

Page 9: Food Chem toxicol

1668 G. Upadhyay et al. / Food and Chemical Toxicology 48 (2010) 1660–1670

2.8. Statistical analysis

The data are expressed as means ± standard error of means (SEM). One-wayanalysis of variance (ANOVA) followed by Newman–Keuls post-test or student’st-test was used, wherever applicable for comparisons. The difference was consid-ered statistically significant when ‘P’ value was <0.05.

3. Results

3.1. ALT, AST and bilirubin content

The patterns of alterations in ALT, AST and bilirubin levels inmouse liver treated with or without pyrogallol in presence or ab-sence of silymarin showed similar trends as reported previously(Upadhyay et al., 2007).

3.2. Differential expression analysis

Gene/transcript expression patterns showed the differentialexpression of 183 transcripts (150 up-regulated, 33 down-regu-lated) over two folds in pyrogallol treated mouse liver as comparedwith control. The expression patterns of pyrogallol and silymarinpre-treated liver showed the differential expression of 79 genes/transcripts (27 up-regulated, 52 down-regulated) as comparedwith pyrogallol, however, 84 genes/transcripts (72 up-regulated,12 down-regulated) have been found to be differentially expressedin silymarin treated mouse as compared with control (Table 2).Furthermore, out of 150 up-regulated genes/transcripts in pyrogal-lol treated groups as compared with control, 51 were down-regu-lated in silymarin pre-treated animals. However, out of 33 genes/transcripts that were down-regulated in pyrogallol-treated ani-

0

0.5

1

1.5

2

2.5

3

3.5

4

Fold

Cha

nge

Bag4 Casp9 C1s

* *

0

0.5

1

1.5

2

2.5

Fold

Cha

nge

Bag4 Casp9 C1s

*

*

(b)

(a)

Fig. 3. qRT-PCR validation of some differentially expressed transcripts. (a) Bar diagrampyrogallol treated mouse liver as compared with controls. (b) Bar diagrams showpyrogallol + silymarin-treated mouse liver as compared with pyrogallol. Data were noand significant changes are expressed as [*(P < 0.05) and **(P < 0.01)].

mals as compared with control, 22 were up-regulated in silymarintreated group (Table 2). The expression of five genes/transcriptswas increased in pyrogallol + silymarin treated groups as com-pared with pyrogallol alone, however, pyrogallol alone also in-creased their level as compared with vehicle control (Table 2).

3.3. Gene clustering, pathway analysis and functional characterization

Only those transcripts were selected for clustering, when theirdifferential expression was 2.0 folds or higher in pyrogallol treatedgroup as compared with control (Fig. 1a and 1b). Comparative eval-uation of the expression patterns of control, pyrogallol and pyro-gallol + silymarin treated groups exhibited a significant alterationin the expression of transcripts associated with various pathwaysthat include, cell-signaling, oxidative stress, apoptosis, cell cycle,inflammatory response, ribosomal proteins, stress proteins, pro-teins involved in extra-cellular matrix remodelling, proteasomaldegradation, transport, immune response and energy metabolism(Fig. 2a–f). Silymarin pre-treatment modulated a number of af-fected genes/transcripts associated with energy metabolism, signaltransduction, cell cycle, apoptosis and oxidative stress (Fig. 1a andb).

3.4. Quantitative real time polymerase chain reaction (qRT-PCR)

The differential expressions of few transcripts whose expres-sions were altered by more than two folds in microarray experi-mentations were confirmed by qRT-PCR. Results obtained fromthe qRT-PCR exhibited similar expression patterns as obtainedwith microarray experimentation (Fig. 3a and b). Althoughmicroarray data underestimated the level of gene expression as

ControlPyrogallol

PCNA Tnk2

*

**

PyrogallolPyrogallol + silymarin

PCNA Tnk2

**

*

s show differential expression in Bag4, Casp9, C1s, PCNA and Tnk2 transcripts indifferential expression in Bag4, Casp9, C1s, PCNA and Tnk2 transcripts in

rmalized with reference to GAPDH. Values are expressed as means ± SEM (n = 4)

Page 10: Food Chem toxicol

G. Upadhyay et al. / Food and Chemical Toxicology 48 (2010) 1660–1670 1669

compared with qRT-PCR, this is not an unusual phenomenon, asqRT-PCR is a quantitative method to estimate gene expression levelwhile microarray is a semi-quantitative tool.

4. Discussion

The doses and times of exposure for these agents were selectedas reported previously (Upadhyay et al., 2007). The level of hepaticdamage markers, such as ALT, AST and bilirubin were reduced sig-nificantly after pyrogallol exposure, however, silymarin co-treat-ment restored ALT, AST and bilirubin levels almost up to thecontrol values. The experimental findings validated our previousobservations that pyrogallol induces hepatotoxicity and silymarinoffers hepatoprotection against pyrogallol-induced liver toxicity(Upadhyay et al., 2007, 2008).

Microarray based expression profiles of control and pyrogallol-treated animals suggest that pyrogallol exposure resulted in thealtered expression of 183 genes (150 up-regulated, 33 down-regulated), which are directly or indirectly related to liver injury.These transcripts are associated with several pathways that in-clude, oxidative stress, cell cycle, cytoskeletal network, cell-celladhesion, extra-cellular matrix, immunological, inflammation,apoptosis, cell-signaling, intermediary metabolism and homeosta-sis. Genes involved in cell-signaling, cell cycle, apoptosis andmetabolism have been known to maintain the dynamic homeo-static equilibrium and the differential expression of these genesshowed an altered homeostasis after pyrogallol treatment. Recep-tor and non-receptor-mediated regulation of signaling cascadecontribute in toxicity response (Frank et al., 2003). Several receptorand non-receptor regulators were differentially expressed in trea-ted conditions. A non-receptor tyrosine kinase-Tnk2 that bindsspecifically to a Rho GTPase and cell division cycle 42 (Cdc42) thatinvolves in the rearrangement of the actin cytoskeleton andremodelling of extra-cellular matrix (Manser et al., 1993; Ahmedet al., 2004), were altered after pyrogallol treatment. Over-expres-sion of activated CDC42 kinase (ACK42) (a Cdc42 binding domainof Tnk2) inhibits cell growth and movement (Ahmed et al.,2004). An up-regulation of Tnk2 in pyrogallol-exposed animalsindicated the loss of membrane integrity and damage (Manseret al., 1993; Ahmed et al., 2004). Bag4 (anti-apoptotic proteinand silencer of death domains), a 60 kDa cytosolic protein, bindswith death domains of tumor necrosis factor receptor 1 (TNFR1)and death receptor 3 (DR3) and prevents cell death signaling andnuclear factor-kappa B (NF-kB) induction by suppressing ligandindependent receptor oligomerization (Jiang et al., 1999). Anincreased expression of cytochrome c and caspase 9 in the pyrogal-lol-treated animals was observed as compared with vehicle controlshowing the involvement of apoptosis. This is in accordance withthe fact that caspase activation in response to mitochondrial cyto-chrome c release is a common phenomenon in the programmedcell death (Sasnauskiene et al., 2009). Although, apoptosis and pro-liferation are involved in the maintenance of homeostasis, theimpairment may further lead to liver damage (van der Horstet al., 2005). The reduced expression of Bag4 and the inducedexpression of pro-apoptotic protein (Casp9) indicated the involve-ment of apoptotic cell death in the pyrogallol-induced hepatotox-icity. An increased level of PCNA, an indicator of cell division andproliferation (Assy et al., 1998), was also observed in pyrogalloltreated mouse liver. It is not surprising, as liver possesses regener-ating capacity and it could be a compensatory response againstapoptotic cell death (Fan et al., 1998; Assy et al., 1998). It suggeststhat an increase in cell loss through apoptosis could be balanced byregeneration mediated through the signal transducer and activatorof transcription 3 (STAT3) and provides additional support for theregeneration in the damaged liver (Park et al., 2008). Free radical

generation is a critical event in pyrogallol-mediated toxicity lead-ing to the activation of complement system via a non-enzymaticmechanism (Upadhyay et al., 2007; Vogt et al., 1989). C1s, an initi-ator molecule of classical pathway of the complement system, wasup-regulated, showing increased complement activation in re-sponse to oxidative stress. An increased apoptosis may lead tocomplement activation and could be responsible for the clearanceof apoptotic bodies (Gullstrand et al., 2009).

Silymarin pre-treatment resulted in the differential expressionof 79 genes/transcripts (27 up-regulated and 52 down-regulated)over two folds as compared with pyrogallol treated group. Silyma-rin treatment caused attenuation of 51 genes/transcripts that wereaugmented in pyrogallol-treated animals; however, 22 genes/tran-scripts that were down-regulated in pyrogallol alone treatmentgroups as compared with control were augmented in silymarinpre-treated group. In particular, silymarin pre-treatment causeddown-regulation of Tnk2, caspase 9, C1s, cytochrome c and up-reg-ulation of Bag4 and PCNA in mouse liver as compared with pyro-gallol-treated animals. This suggests that silymarin offers liverprotection against pyrogallol-mediated toxicity by reducing apop-tosis, diminishing complement activation and regulating extra-cel-lular matrix remodelling in addition to its antioxidant capacity.Furthermore, augmentation of PCNA provides an additional sup-port that probably silymarin provides cell regeneration capacityin pyrogallol treated mouse liver, as PCNA is a marker of cell pro-liferation. The microarray data showed an active involvement ofvarious pathways including, oxidative stress, apoptosis, cell cycle,inflammatory response, ribosomal proteins, stress proteins, pro-teins involved in extra-cellular matrix remodelling, proteasomaldegradation, transport, immune response and energy metabolismin pyrogallol-induced hepatotoxicity and silymarin-mediatedhepatoprotection. Based on some of the modulated genes i.e.,Tnk2, Ttrap, Cycs, Mapk13, Bag4, caspase 9, Cdkn1c, PCNA andC1s, GenMAPP software generated the possibility of the involve-ment of Fas signaling, complement activation, TNF signaling andMAPK signaling in pyrogallol-induced hepatotoxicity and silyma-rin-mediated hepatoprotection. The results of the study high-lighted that multi-faceted effects of silymarin could be due to itsmultiple function in addition to its antioxidant property. Altera-tions in some of the transcripts whose roles are not well estab-lished were also identified; their functional characterization andsubsequent role in these processes needs further investigation.

Conflicts of Interest

The authors declare that there are no conflicts of interest.

Acknowledgements

Authors sincerely thank Council of Scientific and Industrial Re-search, New Delhi for providing research fellowship to GhanshyamUpadhyay, Manindra Nath Tiwari and Anurag Jyoti. The IITR Com-munication number of this article is 2758.

References

Agrawal, R., Sharma, P.K., Rao, G.S., 2001. Release of iron from ferritin bymetabolites of benzene and superoxide radical generating agents. Toxicology168, 223–230.

Ahmed, I., Calle, Y., Sayed, M.A., Kamal, J.M., Rengaswamy, P., Manser, E., Meiners, S.,Nur-E-Kamal, A., 2004. Cdc42-dependent nuclear translocation of non-receptortyrosine kinase. ACK Biochem. Biophys. Res. Commun. 314, 571–579.

Akagawa, M., Shigemitsu, T., Suyama, K., 2003. Production of hydrogen peroxide bypolyphenol and polyphenol-rich beverages under quasi-physiologicalconditions. Biosci. Biotechnol. Biochem. 67, 2632–2640.

Amin, R.P., Hamadeh, H.K., Bushel, P.R., Bennett, L., Afshari, C.A., Paules, R.S., 2002.Genomic interrogation of mechanism(s) underlying cellular responses totoxicants. Toxicology 181, 555–563.

Page 11: Food Chem toxicol

1670 G. Upadhyay et al. / Food and Chemical Toxicology 48 (2010) 1660–1670

Assy, N., Gong, Y., Zhang, M., Pettigrew, N.M., Pashniak, D., Minuk, G.Y., 1998. Use ofproliferating cell nuclear antigen as a marker of liver regeneration after partialhepatectomy in rats. J. Lab. Clin. Med. 131, 251–256.

Bakke, O.M., Manocchia, M., de Abajo, F., Kaitin, K.I., Lasagna, L., 1995. Drug safetydiscontinuations in the United Kingdom, the United States, and Spain from1974 through 1993: a regulatory perspective. Clin. Pharmacol. Ther. 58, 108–117.

Bauer, M., Bauer, I., 2002. Heme oxygenase-1: redox regulation and role in thehepatic response to oxidative stress. Antioxid. Redox Signal 4, 749–758.

Budavari, S., 1996. Pyrogallol. In: The Merck Index, 12th ed. Merck & Co., Inc.,Whitehall, NJ, pp. 1375–1376.

Burczynski, M.E., McMillian, M., Ciervo, J., Li, L., Parker, J.B., Dunn II, R.T., Hicken, S.,Farr, S., Johnson, M.D., 2000. Toxicogenomics-based discrimination of toxicmechanism in HepG2 human hepatoma cells. Toxicol. Sci. 58, 399–415.

Cao, G., Sofic, E., Prior, R.L., 1997. Antioxidant and prooxidant behavior offlavonoids: structure-activity relationships. Free Radic. Biol. Med. 22, 749–760.

Chavan, P., Joshi, K., Patwardhan, B., 2006. DNA microarrays in herbal drug research.Evid. Based Complement. Altern. Med. 3, 447–457.

Dahlquist, K.D., Salomonis, N., Vranizan, K., Lawlor, S.C., Conklin, B.R., 2002.GenMAPP, a new tool for viewing and analyzing microarray data on biologicalpathways. Nat. Genet. 31, 19–20.

Fan, G., Kren, B.T., Steer, C.J., 1998. Regulation of apoptosis-associated genes in theregenerating liver. Semin. Liver Dis. 18, 123–140.

Frank, G.D., Mifune, M., Inagami, T., Ohba, M., Sasaki, T., Higashiyama, S., Dempsey,P.J., Eguchi, S., 2003. Distinct mechanisms of receptor and nonreceptor tyrosinekinase activation by reactive oxygen species in vascular smooth muscle cells:role of metalloprotease and protein kinase C-delta. Mol. Cell. Biol. 23, 1581–1589.

Gullstrand, B., Martensson, U., Sturfelt, G., Bengtsson, A.A., Truedsson, L., 2009.Complement classical pathway components are all important in clearance ofapoptotic and secondary necrotic cells. Clin. Exp. Immunol. 156, 303–311.

Gupta, Y.K., Sharma, M., Chaudhary, G., Katiyar, C.K., 2004. Hepatoprotective effectof New Livfit, a polyherbal formulation, is mediated through its free radicalscavenging activity. Phytother. Res. 18, 362–364.

Gupta, Y.K., Sharma, M., Chaudhary, G., 2002. Pyrogallol-induced hepatotoxicity inrats: a model to evaluate antioxidant hepatoprotective agents. Methods Find.Exp. Clin. Pharmacol. 24, 497–500.

Hayakawa, F., Kimura, T., Maeda, T., Fujita, M., Sohmiya, H., Fujii, M., Ando, T., 1997.DNA cleavage reaction and linoleic acid peroxidation induced by tea catechinsin the presence of cupric ion. Biochim. Biophys. Acta 1336, 123–131.

Jiang, Y., Woronicz, J.D., Liu, W., Goeddel, D.V., 1999. Prevention of constitutive TNFreceptor 1 signaling by silencer of death domains. Science 283, 543–546.

Lewis, J.H., Zimmerman, H.J., 1989. Drug-induced liver disease. Med. Clin. North Am.73, 775–792.

Manser, E., Leung, T., Salihuddin, H., Tan, L., Lim, L., 1993. A non-receptor tyrosinekinase that inhibits the GTPase activity of p21cdc42. Nature 363, 364–367.

Mochizuki, M., Yamazaki, S., Kano, K., Ikeda, T., 2002. Kinetic analysis andmechanistic aspects of autooxidation of catechins. Biochim. Biophys. Acta1569, 35–44.

Park, J.Y., Cho, M.O., Leonard, S., Calder, B., Mian, I.S., Kim, W.H., Wijnhoven, S., vanSteeg, H., Mitchell, J., van der Horst, G.T., Hoeijmakers, J., Cohen, P., Vijg, J., Suh,Y., 2008. Homeostatic imbalance between apoptosis and cell renewal in theliver of premature aging Xpd mice. PLoS ONE 3, e2346.

Patel, S., Singh, K., Singh, S., Singh, M.P., 2008. Gene expression profiles of mousestriatum in control and maneb + paraquat-induced Parkinson’s diseasephenotype: validation of differentially expressed energy metabolizingtranscripts. Mol. Biotechnol. 40, 59–68.

Ramos, R.G., Olden, K., 2008. Gene-environment interactions in the development ofcomplex disease phenotypes. Int. J. Environ. Res. Public Health 5, 4–11.

Sasnauskiene, A., Kadziauskas, J., Vezelyte, N., Jonusiene, V., Kirveliene, V., 2009.Apoptosis, autophagy and cell cycle arrest following photodamage tomitochondrial interior. Apoptosis 14, 276–286.

Singh, K., Singh, S., Singhal, N.K., Sharma, A., Parmar, D., Singh, M.P., 2010. Nicotine-and caffeine-mediated changes in gene expression patterns of MPTP-lesionedmouse striatum: implications in neuroprotection mechanism. Chem. Biol.Interact. doi:10.1016/j.cbi.2010.03.015.

Shukla, Y.N., Srivastava, A., Kumar, S., Kumar, S., 1999. Phytotoxic and antimicrobialconstituents of Argyreia speciosa and Oenothera biennis. J. Ethnopharmacol. 67,241–245.

Sturgill, M.G., Lambert, G.H., 1997. Xenobiotic-induced hepatotoxicity: mechanismsof liver injury and methods of monitoring hepatic function. Clin. Chem. 43,1512–1526.

Upadhyay, G., Kumar, A., Singh, M.P., 2007. Effect of silymarin on pyrogallol- andrifampicin-induced hepatotoxicity in mouse. Eur. J. Pharmacol. 565, 190–201.

Upadhyay, G., Singh, A.K., Kumar, A., Prakash, O., Singh, M.P., 2008. Resveratrolmodulates pyrogallol-induced changes in hepatic toxicity markers, xenobioticmetabolizing enzymes and oxidative stress. Eur. J. Pharmacol. 596, 146–152.

van der Horst, E.H., Degenhardt, Y.Y., Strelow, A., Slavin, A., Chinn, L., Orf, J., Rong, M.,Li, S., See, L.H., Nguyen, K.Q., Hoey, T., Wesche, H., Powers, S., 2005. Metastaticproperties and genomic amplification of the tyrosine kinase gene ACK1. Proc.Natl. Acad. Sci. USA 102, 15901–15906.

Vogt, W., Damerau, B., von Zabern, I., Nolte, R., Brunahl, D., 1989. Non-enzymaticactivation of the fifth component of human complement, by oxygen radicals.Some properties of the activation product, C5b-like C5. Mol. Immunol. 26,1133–1142.

Waring, J.F., Ciurlionis, R., Jolly, R.A., Heindel, M., Ulrich, R.G., 2001a. Microarrayanalysis of hepatotoxins in vitro reveals a correlation between gene expressionprofiles and mechanisms of toxicity. Toxicol. Lett. 120, 359–368.

Waring, J.F., Gum, R., Morfitt, D., Jolly, R.A., Ciurlionis, R., Heindel, M., Gallenberg, L.,Buratto, B., Ulrich, R.G., 2002. Identifying toxic mechanisms using DNAmicroarrays: evidence that an experimental inhibitor of cell adhesionmolecule expression signals through the aryl hydrocarbon nuclear receptor.Toxicology 181, 537–550.

Waring, J.F., Jolly, R.A., Ciurlionis, R., Lum, P.Y., Praestgaard, J.T., Morfitt, D.C.,Buratto, B., Roberts, C., Schadt, E., Ulrich, R.G., 2001b. Clustering of hepatotoxinsbased on mechanism of toxicity using gene expression profiles. Toxicol. Appl.Pharmacol. 175, 28–42.