(HepG2) cells - Springer Link

15 downloads 0 Views 152KB Size Report
Han-Saem Choi1, Youn-Jung Kim2, Mee Song1,. Mi-Kyung Song1 & Jae-Chun Ryu1. 1Cellular and Molecular ... 2Department of Applied Chemistry, Kyung Hee University, ..... Tai, H. L., Dehn, P. F. & Olson, J. R. Use of rat H4IIE and human ...
Available online at www.ehs.or.kr

Identification of Hepatotoxicity Related Genes Induced by Chlordane in Human Hepatocellular Carcinoma (HepG2) Cells Han-Saem Choi1, Youn-Jung Kim2, Mee Song1, Mi-Kyung Song1 & Jae-Chun Ryu1 1

Cellular and Molecular Toxicology Laboratory, Korea Institute of Science & Technology P.O. Box 131, Cheongryang, Seoul 130-650, Korea 2 Department of Applied Chemistry, Kyung Hee University, Yongin 449-701, Korea Correspondence and requests for materials should be addressed to J.-C. Ryu ([email protected]) Accepted 1 June 2010

Abstract Chlordane is a bioaccumulative, persistent, and toxic pollutant. Chlordane is one of the 12 priority of Persistent Organic Pollutants (POPs) intended for global action by the United Nations Environment Program (UNEP) Governing Council. POPs are organic compounds resistant to environmental degradation through chemical, biological, and photolytic processes. Chlordane is ubiquitous in air, water, soil, and biological matrices, as well as in major environmental compartments. Chlordane has effects on various organs such as thyroid, bone, skin, kidneys, and blood cells and especially, revealed strong toxicity to liver. In this study, we identified genes related to hepatotoxiciy induced by chlordane in human hepatocellular carcinoma (HepG2) cells using microarray and gene ontology (GO) analysis. Through microarray analysis, we identified 524 up- and 440 down-regulated genes changed by more than 2.0-fold and P-values 0.05 by chlordane. And after GO analysis, we determined several key pathways which known as related to hepatotoxicity such as metabolism of xenobiotics by cytochrome P450, cell cycle, and apoptosis. Thus, our present study suggests that genes expressed by chlordane may provide a clue for hepatotoxic mechanism of chlordane. Keywords: Persistent Organic Pollutants (POPs), Chlordane, Microarray, Gene ontology (GO)

The major research goals developing biomarkers are the development and validation of biomarkers that permit the prediction of the risk of disease. Toxicogenomic study has been widely used to characterize toxicological properties of disease. There are thousands of Persistent Organic Pollutants (POPs) chemicals, often coming from certain series or ‘families’ of chemicals (e.g there are theoretically 209 different polychlorinated biphenyls, differing from each other by level of chlorination and substitution position). POPs are persistent in the environment, having long half-lives in soils, sediments, air, or biota. Several POPs have been used as pesticides and some others are used in industrial processes and in the production of a range of goods such as solvents, polyvinyl chloride, and pharmaceuticals. Though there are a few natural sources of POPs, most POPs are created by humans in industrial processes, either intentionally or as byproducts1. Chlordane was introduced in the 1940s as the first chlorinated cyclodiene insecticide and used extensively for the control of numerous agricultural pests until banned in the United States, Canada, and Western Europe in the 1980s. Chlordane is a nongenotoxic murine hepatocarcinogen that is believed to be associated with increased risk of reproductive problems, immune dysfunction, and cancer. It is listed as “possibly carcinogenic to humans” (IARC group 2B)2. Chlordane has a 10-20 year half-life in soil around dwellings treated for termites3. Chlordane is highly toxic, stable, highly lipid soluble with a long biological half-life and shows a high degree of bioaccumulation and biomagnifications within food chains. Although the use of commercial chlordane has been banned since 1986 in japan, chlordane isomer and related compounds are still being detected in environmental sample around the world4. Chlordane acts as an endocrine disruption chemical in humans and wildlife. In animal studies, it was shown to exhibit an estrogen-like effect and to produce detrimental effects on reproductive systems5-10. Although many developing and developed countries have prohibited the use of this compound, it is consistently detected in ecosystems11-13. Therefore, impacts of chlordane on human and other wildlife have raised

Toxicogenomic Study of Chlordane

Cytotoxicity of Chlordane in HepG2 Cells To determine the optimal concentration, cytotoxicity of chlordane was assessed by MTT [3-(4,5-dimethylthizol-2-yl)-2,5-diphenyltetrazolium bromide] assay. The survival percentage relative to solvent control (DMSO) was determined as a percentage of optical density values measured after treatment. Based on results of cytotoxicity assay, nontoxic and 20% cell viability inhibitory concentrations (NT and IC20) of chlordane were calculated. Dose-dependent cell viability curve was obtained after 48 h exposure of chlordane in HepG2 cells as shown in Figure 1. The NT and IC20 values for chlordane were 120.0 μM and 158.2 μM, respectively. Gene Expression Analysis HepG2 cells were treated with 120.0 μM (NT) and 158.2 μM (IC20) chlordane for 48 h, and the total RNA was subjected to microarray analysis. Gene expression changes analyzed by comparing with treated group and control group using a statistical criteria of ›2.0-fold changes with P⁄0.05. In microarray analysis, NT group treated with 120.0 μM chlordane for 48 h expressed 1,714 genes. And 2,501 genes expressed in IC20 group treated with 158.2 μM chlordane for 48 h. Commonly expressed genes from the two groups

120 100

Cell viability (%)

global concerns. Chlordane is a widely distributed contaminant in environments such as aquatic systems, air, soil, sediments, seabird, fish, and breastmilk14-19. Although chlordane has been studied with respect to environmental distribution, concentration, and harmful effects on organisms, little is known regarding the effects of chlordane on genes function. Like these, many researchers are performing risk assessments and toxicological studies of chlordane with various animals and organs by means of physical and chemical measurements, but such physicochemical analysis may not be sufficient to provide detailed information on how chlordane affects the cells on a molecular level. Therefore, a toxicological study looking at the effects of chlordane on the molecular level is required. The aim of this study is the identification of potential gene-based markers on hepatotoxicity of chlordane. We examined global gene expression in a small number of well-matched exposed-control subject pairs. Genes with differential expression were then ranked and selected for further examination using several forms of statistical analysis. The identification of differentially expressed genes (DEGs) may assist in the identification of potential biomarker and may understand molecular toxicological mechanisms of chlordane in human hepatocytes.

169

80 60 40 20 0

Nontoxic dose: 120 μm IC20: 158.2 μM 0

100

200

300

Concentration (μM)

Figure 1. Cell viability measured by MTT assay. HepG2 cells were exposed to different concentrations of chlordane for 48 h. After exposure, cell viability for each treatment was determined based on spectrometry of formazan formation, and represented the viability percentage relative to control (DMSO) exposure.

Number of genes responsive to NT dose

390

Number of genes responsive to IC20 dose

1324

1177

Figure 2. Venn diagram of differentially expressed genes responsive to different concentrations of chlordane identified by oligonucleotide microarray analysis.

are 1,324 genes. Considering the genes which increases depending on the increase of the concentration of chlordane in these common genes, there are 524 up-regulated genes and 440 down-regulated genes (Figure 2). To confirm microarray data and mRNA expression for 16 genes (C9orf150, IL8, JAG1, DUSP1, HMGA1, LGALS1, LIF, ADH4, APOH, CPB2, CCNB2, CDC2, GSTA1, PCSK9, SLC13A3, and TNFRSF19), total RNA from chlordane-exposed HepG2 cells was analyzed using quantitative real-time RT-PCR with SYBR green fluorescent dye. Expression of these genes showed similar patterns in the quantitative real time RT-

170

Toxicol. Environ. Health. Sci. Vol. 2(3), 168-174, 2010

Log2 ratio (Quantitative real-time RT-PCR)

8

LIF C9orf150 JAG1 LGALS1

r2=0.958

6

DUSP1

IL8

4

HMGA1

2 0 SLC13A3 TNFRSF19 ADH4 CCNB2 APOH CDC2 CPB2

2 4 6

GSTA1

8 6

PCSK9

4

2 0 2 Log2 ratio (microarray)

4

6

Figure 3. Comparison of DNA microarray and quantitative real-time RT-PCR data in chlordane-induced expressed genes. The fold change in the mean mRNA ratio for chlordane were log2 transformed and plotted with microarray versus real-time RT-PCR (r2; correlation coefficient).

(A) Endocytosis Jak-STAT signaling.. Focal adhesion

skeleton, focal adhesion, JAK-STAT signalling pathway, and endocytosis are prominently annotated with up-regulated genes. Also cell cycle, metabolism of xenobiotics by cytochrome P450, oocyte meiosis, DNA replication, PPAR signalling pathway, fatty acid metabolism, and p53 signalling pathway are prominently annotated with down-regulated genes. We investigated an enrichment of GO annotations in the up-regulated and down-regulated genes. EASE analysis was performed on genes to find significantly represented biological process and to identify any biological terms that response to chlordane using program in http://david.abcc.ncifcrf.gov/. And then, the biological process terms were condensed to the most common parent term without going higher than the fourth GO level below biological process. The categories of the function in expressed genes are presented in Table 1. The biological process profile could be subdivided up- and down-regulated to top 10, cell cycle, transport, transcription, cell proliferation, proteolysis, DNA repair, fatty acid metabolic process, apoptosis, RNA splicing, cell differentiation, transport, signal transduction, transcription, apoptosis, cell proliferation, cell cycle, cell differentiation, cell adhesion, MAPK activity, and cell motility.

Regulation of actin.. MAPK signaling..

Discussion

Pathways in cancer 0

5

10

15

20

Number of up-regulated genes

(B) p53 signaling pathway Fatty acid metabolism PPAR signaling.. DNA replication Oocyte meiosis Metabolism of.. Cell cycle 0

5

10

15

20

Number of down-regulated genes

Figure 4. The pathway analysis of differentially expressed up-regulated (A) and down-regulated (B) genes using KEGG pathway database.

PCR as microarray data. Correlation coefficient (r2) between microarray and quantitative real-time RTPCR is 0.958 (Figure 3). Up- and down-regulated genes were classified according to KEGG pathway to analyse molecular mechanisms (Figure 4). In KEGG pathway analysis, cancer, MAPK signalling pathway, regulation of actin cyto-

In mammals, metabolism of chlordane is very slow and occurs primarily in the liver. Reductive dechlorination of chlordane catalyzed by enzymes located in the microsomal fraction of liver, lung, kidney, and intestine appears to be an important metabolic pathway13. As shown in Figure 3, the expressed genes in chlordane-treated HepG2 cells are related to pathways such as metabolism of xenobiotics by cytochrome P450, cell cycle, and apoptosis. Little data exists on pesticide impacts on the toxicological parameters selected in this study in humans, human tissues, or human cell lines. Exposure concentrations in this study were above values typically encountered in the real world, but necessitated by the model chosen, as HepG2 cells have lower levels of cytochrome P450 enzymes and hence lower sensitivities to toxicants than rat hepatocytes and cell lines or human hepatocytes20,21. Recent studies indicate that the development and progress of hepatocellular carcinoma (HCC) are caused by the accumulation of altered gene expression. A huge number of genes that might contribute to hepatocarcinogenesis have been obtained from studies using high throughput technology, such as microarray22-26. However, the key genes that are crucial for HCC and the molecular mechanisms for hepatocarcinogenesis

Toxicogenomic Study of Chlordane

171

Table 1. GO annotations for chlordane-induced genes. DOWN

UP Genes

GO-Biological Process (top 10 out of 221 total)

Genes

38 37 37 36 30 29 27 26 24 23

Cell cycle Cell division Mitosis Organelle fission DNA metabolic process Oxidation reduction Cellular response to stress Macromolecular complex subunit organization Regulation of apoptosis Regulation of programmed cell death

74 39 37 37 37 34 32 27 27 27

Genes

GO-Cellular Component (top 10 out of 56 total)

Genes

108 49 45 45 25 25 24 18 17 14

Intracellular non-membrane-bounded organelle Membrane-enclosed lumen Cytoskeleton Cytosol Mitochondrion Cytoskeletal part Chromosome Nucleoplasm Endoplasmic reticulum Extracellular region part

79 66 47 46 42 41 38 38 37 37

GO-Moleculer Function (top 10 out of 42 total)

Genes

GO-Moleculer Function (top 10 out of 76 total)

Genes

Protein dimerization activity Cytoskeletal protein binding Protein heterodimerization activity Protein homodimerization activity Transcription repressor activity GTPase activity Enzyme inhibitor activity Growth factor activity Protein complex binding Phosphoprotein phosphatase activity

35 30 19 17 14 12 12 11 11 10

GO-Biological Process (top 10 out of 248 total) Phosphate metabolic process Regulation of cell proliferation Regulation of apoptosis Homeostatic process Cell cycle Apoptosis Chemical homeostasis Cell proliferation Intracellular transport Response to wounding GO-Cellular Component (top 10 out of 39 total) Plasma membrane Cytosol Cell fraction Cytoskeleton Vesicle Cell projection Membrane-bounded vesicle Soluble fraction Internal side of plasma membrane Basolateral plasma membrane

remain to be elucidated. The subsequent results indicated that SOD1 and GSTA1 were down-regulated at both mRNA and protein levels in tree shrew and human HCC, and the down-regulations of SOD1 and GSTA1 in human HCC samples were closely correlated with the histopathological grading (P⁄0.05)27. These reports appears a consistency with our data that chlordane was down-regulated in the mRNA level of GSTA1 gene and genes related to xenobiotics by cytochrome P450. In conclusion, chlordane has been implicated in a broad range of adverse human environmental health effects, including impaired reproduction and neurological dysfunction, immunologic effects, hepatotoxic, skin effects, and cancer. As long as chlordane-contain-

Nucleotide binding ATP binding Cofactor binding Lipid binding Identical protein binding Coenzyme binding ATPase activity Carboxylic acid binding ATPase activity, coupled Monocarboxylic acid binding

80 61 24 24 24 16 14 12 12 8

ing products are being produced, and the convincing of substantive evidences for the actual and potential toxic impact to both human health and the environment, international cooperation in addressing the presence of chlordane in the global environment must be achieved. Our results showed that the gene expression patterns were associated with hepatotoxicity induced by chlordane. And classifying the gene alterations, analyzing the gene expression patterns, and understanding mechanism associated with chlordane-induced toxicity should allow earlier identification of environmentally relevant toxicological findings in compound screening and aid in the development of new chemical to reduce hepatotoxicity.

172

Toxicol. Environ. Health. Sci. Vol. 2(3), 168-174, 2010

Materials and Methods Chemicals and Reagents Chlordane, dimethyl sulfoxide (DMSO), and 3-(4,5dimethylthizol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) were purchased from Sigma (USA). Dulbecco’s Modified Eagle Medium (DMEM), Dulbecco’s Phosphate Buffered Saline (PBS), 0.5% trypsin-EDTA, and Fetal Bovine Serum (FBS) were the products of GIBCOTM (USA). Trizol reagent was produced by Invitrogen (USA) and RNeasy mini kit and RNasefree DNase set were purchased from Qiagen (USA). All other chemicals used were of analytical grade or the highest grade available. Cell Lines and Culture Human hepatocellular carcinoma cell line (HepG2) used throughout the study was purchased from Korean Cell Line Bank (Korea). HepG2 cells were grown in DMEM medium supplemented with 10% inactivated FBS, 0.044 M sodium bicarbonate, 10 mM sodium pyruvate and 1% penicillin at 37 C in 5% CO2 atmosphere. For cell growth, the medium was renewed every two or three days. HepG2 cells were approximately 80% confluence achieved by plating 6 106 cells/mL in 100 mm culture dish. Determination of Cell Viability To determine the cytotoxicity and effects on cell growth, MTT assay was performed28. HepG2 cells were seeded at a seeding density of 80 104 cells/mL on a well in 500 μL of media in 24-well plate. And cells were exposed to various concentrations of chlordane in culture medium at 37 C for 48 h exposure time. After exposure, the cells were incubated for 3 h with 4 mg/mL MTT in PBS. To quench the reaction, the medium was removed and DMSO was added and transferred to 96-well plate. The optimal density (O.D) of the purple formazan product was measured at a wavelength of 540 nm. Nontoxic (NT) dose selected the concentration that cell starts being died. The 20% inhibitory concentration (IC20) of cell proliferation was defined as the concentration that causes a 20% reduction in the cell viability versus the solvent treated control. The IC20 and NT values were directly determined from the linear dose-response curves. RNA Extraction Total RNA was extracted from the HepG2 cells after 120.0 and 158.2 μM chlordane exposure for 48 h, using the Trizol reagent and purified using RNeasy mini kit according to the manufacturer’s instructions. Genomic DNA was removed using RNase-free DNase set during RNA purification procedure. The amount of each total

RNA was quantified using NanoDrop ND 1000 spectrophotometer (NanoDrop Technologies Inc., USA). Only samples with an A260/A280 ratio between 1.9 and 2.2 were considered for suitable use and its quality was checked by ExperionTM (Bio-Rad, USA).

Oligonucleotide Microarray Hybridization Gene expression analysis was conducted on the RNA samples using 4 44 K whole human genome microarray (Agilent Technologies, USA). Triplicate analysis was simultaneously performed. Labeling and hybridization were performed by instruction of Platinum Biochip Reagent Kit (GenoCheck Co. Ltd, Korea). This was followed by the coupling of the Cy3 dye for the controls (DMSO) and Cy5 dye for the treated samples. Hybridization was performed in a hybridization oven at 62 C for 12 h. After washing (2 SSC/ 0.1% SDS for 2 min at 58 C, 1 SSC for 2 min at RT and 0.2 SSC for 3 min at RT), the slide was dried by centrifugation at 800 rpm for 3 min at RT. Hybridization images on the slides were scanned by GenePix 4000B (Axon Instruments, USA). Scanned images were analyzed with GenePix 4.1 software (Axon Instruments, USA) to obtain gene expression ratios. Data Analysis After analyzing of scanned images, spots that adjudged as substandard via the visual examination of each slide were flagged and excluded from further analysis. Spots that harboured dust artifacts or spatial defects were manually flagged and excluded. In an attempt to filter out the unreliable data, spots with signal-to-noise (signal-background-background SD) ratios below 10 were not included in the data. Data were normalized via global, lowless, print-tip, and scaled normalization methods. Obtained data were represented to volcano plot of genes that behaved similarly across the chlordane treatment using GeneSpring GX 7.3.1 software. We utilized an algorithm based on the Pearson’s correlation to separate genes exhibiting similar patterns29. Functional Grouping and Clustering Analysis In order to classify the selected genes into groups with a similar pattern of expression, each gene was assigned to an appropriate category according to its main cellular function. The necessary information to categorize each gene was obtained from several databases particularly the database located at http://david. abcc.ncifcrf.gov/home.jsp. Quantitative Real-time PCR Messenger RNA expression levels for the genes of interest were analyzed via quantitative real time reverse

Toxicogenomic Study of Chlordane

173

Table 2. Primer sequences. Gene

Accession No.

Primer sequence (5′

3′)

C9orf150

NM_203403

F R

TTACGTGGCAGCTACAACAGCCTA TGAGGGAGCTGTCACTGAATTGGT

IL8

NM_000584

F R

CCAGGAAGAAACCACCGGA GAAATCAGGAAGGCTGCCAAG

JAG1

NM_000214

F R

TGAGATCAATGGCTACCGGTGTGT TGCAAGGTCTCCCTGAAACTTCCT

DUSP1

NM_004417

F R

GGCCCCGAGAACAGACAAA GTGCCCACTTCCATGACCAT

HMGA1

NM_145904

F R

GTCCTCCTGCTTTGGCATC AGCAGGTGGAAGAGTGATGG

LGALS1

NM_002305

F R

TCGCCAGCAACCTGAATCTCAAAC TTGTTGCTGTCTTTGCCCAGGTTC

LIF

NM_002309

F R

TATCACCATCTGTGCCTTTGCTGC TCTGCCAGATTGTTCCTATGCCCA

ADH4

NM_000670

F R

TGAATTGACCAAGGGAGGTGTGGA TTGTACAGTCCAGGGCTGCTTTCA

APOH

NM_000042

F R

AATGCCCATTCCCATCAAGACCAG TTTCTTCCGGGCCATCCAGAGAAT

CPB2

NM_001872

F R

TCAGCATGCATTCATACTCCCAGC TAGCACGAACTGCTTCACTGGCTA

CCNB2

NM_004701

F R

AGACTCTGTACATGTGCGTTGGCA TTGGAAGCCAAGAGCAGAGCAGTA

CDC2

NM_001786

F R

CAGAGCTTTGGGCACTCCCAATAA ATGGGATGCTAGGCTTCCTGGTTT

GSTA1

NM_145740

F R

AGGAAGCCTCCCATGGATGAGAAA TTCTTGGCCTCCATGACTGCGTTA

PCSK9

NM_174936

F R

AGGAACTGAGCCAGAAACGCAGAT ATCAGGCCTGGGTGATAACGGAAA

SLC13A3

NM_001011554

F R

AGACAGCCCAGTGTCATCAACAGT TTGGAACATTGCAGCCCAGGAAAG

TNFRSF19

NM_148957

F R

TGATGAAAGTAGGCAGGGCTGTGT TCTCCAGCCAGTGTTTGTCCTTGA

transcription polymerase chain reaction (RT-PCR) using a Bio-Rad iCycler system (Bio-Rad, USA). Total RNA was reverse-transcribed into cDNA using an Omniscript RT kit (Qiagen). Primer specificity was tested by running a regular PCR for 40 cycles (95 C for 20 s and 60 C for 1 min). Real time RT-PCR was performed using a SYBR supermix kit (Bio-Rad). Samples were subjected to 45 cycles of 95 C for 20 s and 60 C for 1 min. PCR efficiency was determined by running serial dilutions of template cDNA and melting curve data were collected to assure PCR specificity. Each cDNA sample was analyzed in triplicate and the corresponding no-RT mRNA sample was included as a negative control. A glyceraldehyde-3-phosphate dehydrogenase (GAPDH) primer was included in every plate as an internal loading control. The mRNA

level of each sample for each gene was normalized against that of GAPDH mRNA. The relative mRNA level was determined as 2[(Ct/GAPDH Ct/gene of interest)]. All data were presented as the mean standard deviation (SD) of three separate experiments. The primers used for the quantitative real time RT-PCR are listed in Table 2.

Acknowledgements This subject is supported by the Korea Research Foundation grants from Korea Ministry of Environment as “The Eco-technopia 21 project”, Ryu, J.C. of the Republic of Korea.

174

Toxicol. Environ. Health. Sci. Vol. 2(3), 168-174, 2010

References 1. UNEP, www.chem.unep.ch/pops/ritter/en/ritteren.pdf (1995). 2. Malarkey, D. E., Devereux, T. R., Dinse, G. E., Mann, P. C. & Maronpot, R. R. Hepatocarcinogenicity of chlordane in B6C3F1 and B6D2F1male mice: evidence for regression in B6C3F1 mice and carcinogenesis independent of ras proto-oncogene activation. Carcinogenesis 16, 2617-2625 (1995). 3. Bennett, G. W. et al. Persistence and distribution of chlordane anddieldrin applied as termiticide. Bull. Environ. Contam. Toxicol. 11, 64-69 (1977). 4. Barber, L. B., Keefe, S. H., Antweiler, R. C., Taylor, H. E. & Wass, R. D. Accumulation of contaminants in fish from wastewater treatment wetlands. Environ. Sci. Technol. 15, 603-611 (2006). 5. Longnecker, M. P. & Rogan, W. J. Persistent organic pollutants in children. Pediatr. Res. 50, 322-323 (2001). 6. Abelsohn, A., Gibson, B. L., Sanborn, M. D. & Erica, W. Identifying and managing adverse environmental health effects: 5. Persistent organic pollutants. CMAJ 166, 1549-1552 (2002). 7. Mills, P. K. & Yang, R. Bioassay of chlordane for possible carcinogenicity. national cancer institute carcinogenesis technical report series. National Toxicology Program 8, 1-123 (1977). 8. Colt, J. S. et al. Breast cancer risk in Hispanic agricultural workers in California. Int. J. Occup. Med. Environ. Health 11, 123-131 (2005). 9. Colt, J. S. et al. Residential insecticide use and risk of non-Hodgkin’s lymphoma. Cancer Epidemiol. Biomarkers Prev. 15, 251-257 (2006). 10. Colborn, T., Vom-Saal, F. S. & Soto, A. M. Developmental effects of endocrine-disrupting chemicals in wildlife and humans. Environ. Health Perspect. 101, 378-384 (1993). 11. Taiwan.EPA. Standard methods to study and detect bioaccumulation in fish and shellfish (EPA-88-150203-01). Taipei (1999). 12. Dietz, R. et al. Comparison of contaminants from different trophic levels and ecosystems. Sci. Total Environ. 245, 221-231 (2000). 13. Bondy, G. et al. Toxicity of the chlordane metabolite oxychlordane in female rats: clinical and histopathological changes. Food Chem. Toxicol. 41, 291-301 (2003). 14. Monirith, I., Ueno, D. & Takahashi, S. Asia-Pacific mussel watch: monitoring contamination of persistent organochlorine compounds in coastal waters of Asian countries. Mar. Pollut. Bull. 6, 81-100 (2003). 15. Jaward, F. M. et al. Passive air sampling of polychlorinated biphenyls, organochlorine compounds, and polybrominated diphenyl ethers across Asia. Environ. Sci. Technol. 39, 8638-8645 (2005). 16. Kurt-Karakus, P. B., Bidleman, T. F. & Jones, K. C. Chiral organochlorine pesticide signatures in global

background soils. Environ. Sci. Technol. 39, 86718677 (2005). 17. Bakan, G. & Ariman, S. Persistent organochlorine residues in sediments along the coast of mid-Black Sea region of Turkey. Mar. Pollut. Bull. 8, 1031-1039 (2004). 18. Braune, B. M., Donaldson, G. M. & Hobson, K. A. Contaminant residues in seabird eggs from the Canadian Arctic. Part I. temporal trends 1975-1998. Environ. Pollut. 114, 39-54 (2001). 19. Sudaryanto, A., Kunisue, T., Tanabe, S., Niida, M. & Hashim, H. Persistent organochlorine compounds in human breast milk from mothers living in Penang and Kedah, Malaysia. Arch. Environ. Contam. Toxicol. 49, 429-437 (2005). 20. Schrenk, D., Stüven, T., Gohl, G., Viebahn, R. & Bock, K. W. Induction of CYP1A and glutathione Stransferase activities by 2,3,7,8-tetrachlorodibenzo-pdioxin in human hepatocyte cultures. Carcinogenesis 16, 943-946 (1995). 21. Tai, H. L., Dehn, P. F. & Olson, J. R. Use of rat H4IIE and human HepG2 cells as models to investigate the metabolism of 2,3,7,8-tetrachlorodibenzofuran (TCDF). Toxicologist 14, 271 (1994). 22. Smith, M. W. et al. Identification of novel tumor markers in hepatitis C virus-associated hepatocellular carcinoma. Cancer Res. 63, 859-864 (2003). 23. Iizuka, N. et al. Molecular signature in three types of hepatocellular carcinoma with different viral origin by oligonucleotide microarray. Int. J. Oncol. 24, 565574 (2004). 24. Mao, H. J., Li, H. N., Zhou, X. M., Zhao, J. L. & Wan, D. F. Monitoring microarray-based gene expression profile changes in hepatocellular carcinoma. World J. Gastroenterol. 11, 2811-2816 (2005). 25. Yang, L. Y., Wang, W., Peng, J. X., Yang, J. Q. & Huang, G. W. Differentially expressed genes between solitary large hepatocellular carcinoma and nodular hepatocellular carcinoma. World J. Gastroenterol. 10, 3569-3573 (2004). 26. Kurokawa, Y. et al. Molecular features of non-B, non-C hepatocellular carcinoma: a PCRarray gene expression profiling study. J. Hepatol. 39, 1004-1012 (2003). 27. Yuan, Li. et al. Candidate genes responsible for human hepatocellular carcinoma identified from differentially expressed genes in hepatocarcinogenesis of the tree shrew (Tupaia belangeri chinesis). Hepatol. Res. 38, 8595 (2008). 28. Mosmann, T. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. J. Immunol. Methods 65, 55-63 (1983). 29. Park, H. W. et al. Gene expression patterns of environmental chemicals in human cell lines using HazChem human array. BioChip J. 3, 65-70 (2009).