Skip to main content
NCBI home page
Toxicology Research logoLink to Toxicology Research
. 2016 Feb 11;5(3):761–772. doi: 10.1039/c5tx00250h

eIF3 regulates migration, invasion and apoptosis in cadmium transformed 16HBE cells and is a novel biomarker of cadmium exposure in a rat model and in workers†

Zhiheng Zhou a,‡, Qian Lu b,‡, Qinhai Huang a, Chanjiao Zheng a, Baoxin Chen a, Yixiong Lei a,✉
PMCID: PMC6060694  PMID: 30090387

graphic file with name c5tx00250h-ga.jpgTranslation (eukaryotic) initiation factor 3 (eIF3 or TIF3) has been found to be a proto-oncogene in cadmium (Cd) response both in vitro and vivo, but whether eIF3 may serve as a biomarker of Cd exposure is still unclear.

Abstract

Translation (eukaryotic) initiation factor 3 (eIF3 or TIF3) has been found to be a proto-oncogene in cadmium (Cd) response both in vitro and vivo, but whether eIF3 may serve as a biomarker of Cd exposure is still unclear. This study aimed to investigate whether eIF3 could serve as a novel biomarker of Cd toxicity in cells, animals and workers, and regulate the apoptosis, migration and invasion in human bronchial epithelial cell (16HBE cells) transformation with cadmium chloride (CdCl2). In CdCl2 transformed 16HBE cells, eIF3 expression increased gradually, and sequencing did not identify mutation and methylation of eIF3. In 16HBE cells with eIF3 silencing by siRNA and CdCl2 treated 16HBE cells of the 15th and 35th generations, the apoptosis, migration and invasion were significantly inhibited, and the expressions of relevant genes were also altered (P < 0.05). In CdCl2 treated rats, eIF3 mRNA expression increased to different extents in the blood, liver, kidney, heart and lung, and this increase was dependent on the Cd concentration (P < 0.05). The eIF3 mRNA expression was related to the mRNA expressions of AKT, BAX, BCL-2, E-CADHERIN, CASPASE-3, EGFR, FOXC2, STAT3, TGF-β1 and VIMENTIN (P < 0.05). In 181 workers with Cd exposure, the eIF3 mRNA expression was positively related to the blood Cd, urine Cd and β2-microglobulin content (P < 0.05). This study showed that abnormally expressed eIF3 may regulate the apoptosis, migration and invasion of 16HBE cells with Cd toxicity. This suggests that eIF3 may become a novel and valuable biomarker of Cd toxicity and Cd-induced effects, and may regulate apoptosis, migration and invasion of 16HBE cells. Thus, the detection of eIF3 expression is important for the monitoring of Cd toxicity in humans.

Introduction

Cadmium (Cd) and its compounds are considered as harmful pollutants worldwide.1–3 Cd has a long biological half-time (19–30 years), and can accumulate and be present in multiple organs for a long time. Cd is toxic to organs and can lead to a number of diseases including liver and kidney injury, respiratory diseases, neurological disorders, skeletal system damage and reproductive system disorders.4–6 Cd is the 7th priority toxicant according to the Agency for Toxic Substances and Disease Registry (ATSDR) of the United States.7 Based on the epidemiological and laboratory findings, Cd has been reported to cause cancer in many organs including the kidney, liver, lung, prostate, pancreas, bladder and breast.8–12 In 1993, Cd and its compounds were named as Group 1 carcinogens by the International Agency for Research on Cancer (IARC).13,14 Therefore, a sensitive and specific biomarker for Cd exposure is beneficial. Although some of the molecules involved in Cd tolerance have been identified, the potential mechanisms involved are still largely unknown.

There is evidence showing that Cd is a potent inducer of gene mutation and may cause uncontrollable gene expression. Some studies on toxicological properties of Cd show that Cd may alter the expressions of key functional genes in target organs, which has been focused upon in cells and animals exposed to Cd.15 In several studies on the Cd induced carcinogenesis, the expressions of some translation-related genes are abnormal. Translation factors involved in the protein expression in eukaryotic cells include translation initiation factor, translation elongation factor and translation termination factor. These factors play important roles in the growth, proliferation and malignant transformation of normal cells.16,17 eIF3 is the largest translation initiation factor in eukaryotic cells, participates in several steps of translation initiation and plays a central role in translation initiation.18,19

Studies have shown that eIF3 may form stable complexes with a 40S ribosomal subunit, which may prevent against early binding to a 60S ribosomal subunit. eIF3 is indispensable for the stable binding of the eIF2-GTP-MettRNA triple complex to the 40S subunit. eIF3 binds to the eIF4G subunit of a cap-binding protein complex (eIF4E) when the 40S subunit is brought to mRNA.20,21 In addition, eIF3 may interact with eIF4B, eIF5 and eIF1, which suggests that eIF3 plays a central role in the translation initiation via interaction with different initiation factors.18,19 Studies also reveal that abnormal eIF3 high expression may cause the malignant transformation of cells, and eIF3 is highly expressed in transformed cells, breast cancer cells and prostate cancer cells.22,23 The stable expression of the modified eIF3e(p48) gene in NIH3T3 cells may cause malignant transformation.24 Joseph and Lei et al.25,26 investigated the gene expression profiles in CdCl2 transformed BALB/c-3T3 cells. They found that eIF3 (GenBank accession number AF271072) in rats was highly expressed in CdCl2 transformed cells and eIF3 was identified as a proto-oncogene of Cd response. However, in Cd toxicity, the regulatory capability of eIF3, the mechanism underlying the abnormal expression of eIF3 and whether eIF3 may serve as a biomarker for Cd exposure are still unclear.

We previously established a model of morphological cell transformation with cadmium chloride (CdCl2) in human bronchial epithelial cells (16HBE)27 and a Cd exposure model in rats.28 To establish the CdCl2 transformed model, 16HBE cells were malignantly transformed by continuous treatment by CdCl2. Tumorigenic potential of transformed cells was identified by assays for anchorage-independent growth in soft agar and for tumorigenicity in nude mice. Reproducibly, 16HBE cells treated with CdCl2 for 35 passages can form robust colonies in soft agar and initiate xenograft tumors in nude mice indicating the fully malignant transformation. To establish the Cd exposure model in rats, specific-pathogen-free (SPF) Sprague-Dawley (SD) rats were chronically exposed to Cd by intra-peritoneal injection of CdCl2. Cd treatment was performed five times weekly for 14 weeks. These models are helpful to examine the molecular events occurring during Cd toxicity and carcinogenesis. This study aimed to investigate the change in eIF3 expression and its potential mechanism and to explore the regulatory effects of eIF3 on the apoptosis, proliferation, migration and invasion of cells. Moreover, we validated eIF3 as a novel biomarker of Cd toxicity in cells, animals and workers with Cd exposure.

Materials and methods

Cell culture and treatments

16HBE cells were morphologically transformed using CdCl2, as previously described.27 Un-transformed 16HBE cells (control group); Cd-transformed cells of the 5th (5 μmol L–1 Cd for 2 weeks), 15th (5 μmol L–1 Cd for 6 weeks) and 35th (5 μmol L–1 Cd for 14 weeks) passages were maintained in RPMI-1640 containing l-glutamine, 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin (Life Technologies) at 37 °C under a 5% CO2 humidified atmosphere. The cells were passaged twice weekly and cells in the logarithmic growth phase (2–5 × 105 cells per mL) were harvested for the following experiments.

Animals and cadmium exposure

Specific-pathogen-free (SPF) Sprague-Dawley (SD) rats (90 ± 10 g) were purchased from Guangdong Medical Laboratory Animal Center (License no.: SCXK 2008-0002, Guangdong, China) and maintained under pathogen-free conditions in the Laboratory Animal Center of Guangzhou Army General Hospital [License no.: SYXK (Military) 2007-33, 2008C1230034834, Guangdong, China]. Ninety-six SD rats (half male and half female) were randomly divided into 4 groups. Rats were chronically exposed to Cd by intra-peritoneal injection of CdCl2 (Sigma, St Louis, MO, USA) in normal saline at different concentrations (high-dose: 1.225 mg kg–1; mid-dose: 0.612 mg kg–1 and low-dose: 0.306 mg kg–1). Rats in the control group were intra-peritoneally injected with 0.5 mL of normal saline. Cd treatment was performed five times weekly. After 14 weeks, 24 h urine samples were collected. On the second day, rats were anesthetized and blood was collected from the heart and stored at 4 °C. The liver, kidney, heart and lung were harvested and stored in liquid nitrogen. The animal handling and experimental procedures were approved by the Animal Experimental Ethics Committee of Guangzhou Army General Hospital (Guangzhou, China).28

Study population

A total of 181 workers were recruited from the Cd refinery factory with the assistance of Center for Disease Control and Prevention, Institute for Health Supervision in Shenzhen, P.R. China. The workers included production workers, machine maintenance workers, product development personnel, management personnel and other personnel engaged in cleaning, service, security, and so on. Detailed information including the age, marital status, smoking status, alcohol consumption, professional and medical history was collected from each subject and evaluated by well-trained interviewers. In addition, the workers were asked to receive a comprehensive physical examination. The physical examination included detection of blood pressure and pulse rate, examination of the throat and pharynx, detection of lung function, electrocardiogram, liver and kidney ultrasonography, cardiopulmonary X-ray, and detection of blood cells, serum alanine aminotransferase (ALT), urinary Cd and creatinine (Cr). In this study, subjects who could not provide reliable information on the smoking history, had a smoking history or had a history of kidney or liver diseases were excluded. Finally, 181 non-smoking subjects (109 males and 72 females) aged 23–50 years were included for analysis.

Quantitative real-time PCR

Total RNA was isolated using the TRIzol reagent. The RNA purity, integrity, and concentration were assessed by agarose gel electrophoresis and NanoDrop ND-1000 (Thermo Fisher Scientific Inc., USA). Reverse transcription was performed using a TITANIUM real-time PCR (RT-PCR) kit (Clontech, Mountain View, CA) according to the manufacturer's instructions. The gene expression was quantified using a fluorescence-based RT-PCR. Data were normalized to β-actin levels according to manufacturer's instructions (Bio-Rad Laboratories). The sequences of primers used for RT-PCR are shown in Table 1.

Table 1. Primers used for PCR of selected mRNA and siRNA.

mRNAs Primers
eIF3 F:5′-gCCCTCTCCCCCAACTATgA-3′
R:5′-gTggTTACATCCATggCTTCCT-3′
EGFR F:5′-GGAGGCAAAGTGCCTATCAA-3′
R:5′-AGGTCATCAACTCCCAAACG-3′
STAT3 F:5′-TGTGCGTATGGGAACACCTA-3′
R:5′-AGAAGGTCGTCTCCCCCTTA-3′
FOXC2 F:5′-CCTCCTGGTATCTCAACCACA-3′
R:5′-GAGGGTCGAGTTCTCAATCCC-3′
VIMENTIN F:5′-AGTCCACTGAGTACCGGAGAC-3′
R:5′-GGTTCCTTTAAGGGCATCCAC-3′
E-CADHERIN F:5′-AAAGGCCCATTTCCTAAAAACCT-3′
R:5′ TGCGTTCTCTATCCAGAGGCT-3′
AKT F:5′-AGCCCACCCTTCAAGCCCCA-3′
R:5′-CTGCGCTCGCTGTCCACACA-3′
BCL-2 F:5′-GATAACGGAGGCTGGGATGC-3′
R:5′-CAGGCATGTTGACTTCACTTGTG-3′
CASPASE-3 F:5′-AGAGGGGATCGTTGTAGAAGTC-3′
R:5′-ACAGTCCAGTTCTGTACCACG-3′
BAX F:5′-TTGCTTCAGGGGATGATTG-3′
R:5′-CAAAGTAGAAAAGGGCGACA-3′
β-Actin F:5′-ACAGAGCCTCGCCTTTGCCGAT-3′
R:5′-CTTGCACATGCCGGAGCCGTT-3′
eIF3 siRNA-1 5′-UACUUAAUCUGCGUAAUGGAC-3′
5′-CCAUUACGCAGAUUAAGUAUA-3′
eIF3 siRNA-2 5′-UCAAAGAAGCUCACAAAGCAC-3′
5′-GCUUUGUGAGCUUCUUUGACC-3′
eIF3 siRNA-3 5′-ACUCAAAUUCGAAGUACUGUG-3′
5′-CAGUACUUCGAAUUUGAGUUU-3′
eIF3 siRNA-nc 5′-UUCUCCGAACGUGUCACGUTT-3′
5′-ACGUGACACGUUCGGAGAATT-3′
Open in a new tab

RNA interference

To inhibit eIF3, 50 nM of siRNA (siRNA eIF3-1, siRNA eIF3-2, siRNA eIF3-3), Shanghai Genepharma, China) were transfected into untreated 16HBE cells, Cd-transformed cells of the 15th passage and Cd-transformed cells of the 35th passage using Lipofectamine 2000 reagent according to the manufacturer's instructions. Cells transfected with scramble-control siRNA (negative control) were used as controls. Cells were harvested 72 h after transfection. Compared with controls, three siRNA eIF3 successfully decreased the expression of eIF3. The sequences of eIF3 siRNA and scramble control siRNA are listed in Table 1.

Cell migration and invasion assay

Cell migration was assessed using Control Cell Culture Inserts in two 24-well plates of 8 μm (BD Biosciences) according to the manufacturer's instructions. Briefly, 200 μl of serum-free medium containing 2 × 105 cells from each subgroup were added to the upper chamber, and 0.6 ml of 20% FBS-containing medium was then added to the lower chamber as a chemoattractant. Cells were incubated for 16 h at 37 °C in 5% CO2.

Matrigel Invasion Chambers in two 24-well plates of 8 μm (BD Bioscience) were used for the invasion assay according to the manufacturer's instructions. Briefly, 200 μl of serum-free medium containing 1 × 105 cells from each subgroup were added to the upper chamber and 0.6 ml of 20% FBS-containing medium was then added to the lower chamber as a chemoattractant. Cells were incubated for 40 h at 37 °C in 5% CO2.

After the incubation, cells on the upper surface of the membrane were removed with cotton swabs. Cells migrating to the bottom of the membrane were fixed and stained with 0.1% Crystal Violet. Cells on the bottom of the membrane were counted at five different microscopic fields and the average was calculated. All the above experiments were replicated three times.

Cytometric analysis of apoptotic cells

To explore the effect of eIF3 on Cd-transformed cells, detection of apoptosis in 16HBE cells, and Cd-transformed cells of the 15th and 35th passages was carried out after being transfected with siRNA eIF3 only for 72 h. Apoptotic cells were analyzed using a flow cytometer (CYTOMICS FC 500, Beckman Coulter) after incubating with a reagent containing Annexin V-FITC and propidium iodide (BD Bioscience, San Jose, CA) for 15 min in the dark at room temperature. Each study was repeated four times.

Sequencing of eIF3

DNA extraction kit (QIAGEN) was used to extract total DNA from un-transformed 16HBE cells (control group) and Cd-transformed cells of the 5th passage, 15th passage and 35th passage. Seven paired primers were designed on the basis of sequences obtained from ; http://frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi and processed for PCR. The maximum and minimum lengths of 7 segments were 1266 bp and 350 bp, respectively, and these segments covered the exons and promoter with 1 kb of the eIF3 gene. PCR products were subjected to purification with Shrimp alkaline enzyme (SAP, Promega) and exonuclease I (EXO I, Epicentre) and then processed with BigDye3.1 kit (ABI). The sequencing reactants were purified with ethanol and the sequences of exons and promoter of the eIF3 gene were determined with the ABI3130XL sequencer. Results were analyzed and compared with Polyphred software.

Detection of eIF3 methylation

Total DNA was extracted from un-transformed 16HBE cells (control group); Cd-transformed cells of the 5th passage, 15th passage and 35th passage. Then, 1 μl of DNA was subjected to 1% agarose electrophoresis for the detection of concentration and quality. EZ DNA Methylation-Gold kit (Zymo Research) was used for methylation treatment, and the methylated CpG island, full exons and promoter were sequenced. A total of 13 segments (11 at exons [2002–2076, 2170–2262, 3666–3753, 4042–4107, 5803–5952, 6035–6162, 8131–8241, 8359–8448, 8767–8840, 10559–10651, 10747–>11235] and 2 at promoter) were sequenced.

Network diagram of genes regulated by eIF3

Data were collected from mRNA microarray assay, microRNA microarray assay and lncRNA microarray assay and screening was done according to the fold change. The P < 0.05 and a fold change of ≥2 were used as thresholds for screening of differentially expressed mRNA, microRNA and lncRNA. Cytoscope software was employed to delineate the network diagram of eIF3-mRNA co-expression.

Network diagram of eIF3 regulated by ceRNA

When the miRNA was negatively related to mRNA and miRNA–mRNA expression (Pearson correlation coefficient <–0.8), the predicted network between miRNA and eIF3 targets was established as follows: the target gene of the known miRNA was predicted from the intersection of Mireap, miRanda, TargetScan: miRanda (; http://www.microrna.org/) with default parameters, and the input was a mature miRNA sequence and 3′UTR of the gene; TargetScan (; http://www.targetscan.org/) with default parameters, the input is a 7mer seed of mature RNA and 3′UTR of the gene; Mireap (; http://sourceforge.net/projects/mireap) with default parameters, the input was a mature miRNA sequence and 3′UTR of the gene. Then, the regulatory relationship between miRNA and lncRNA was established. The regulatory networks of miRNA–mRNA and miRNA–lncRNA obtained on the basis of above procedures were employed to delineate the regulatory network of lncRNA–miRNA–mRNA by using cytoscope software.

Cadmium determination and functional and pathological examinations of organs in cadmium-exposed rats

The cadmium level was determined using the cadmium standard solution (BZ/WJ/GB101/2009-1, Guangdong Occupational Health Inspection Center, Guangdong, China) by atomic absorption spectrometry (ZEENIT700, Analytik Jena, Jena, German). The concentration of urine cadmium was normalized by urinary creatinine (Cr). Tissue samples were fixed in 10% formalin and the pathological features were examined following Hematoxylin and Eosin (HE) staining. Serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) were used as biochemical markers of liver function. Blood urea nitrogen (BUN), serum creatinine (sCr) and 24 h urine protein (24 h Pro) were used to evaluate the renal function. ALT, AST, BUN, sCr and 24 h Pro were measured using the corresponding kits according to the manufacturer's instructions with an automatic biochemistry analyzer (Hitachi 7600-020/7170A: Tokyo, Japan). All animal experiments were performed in accordance with the principles of the Declaration of Helsinki. All experimental protocols were approved by the Research Ethic Committee of Guangzhou Medical University.

Collection and treatment of biological samples from cadmium-exposed workers

Venous blood was collected after fasting for 10–12 h and transferred into an anticoagulant and metal-free tube for blood cadmium (BCd) detection, blood routine examination, blood biochemical examination (ALT, AST, Cr and BUN), and detection of blood eIF3 and its target genes. BCd concentrations were measured by atomic absorption spectrometry (ZEENIT700; Analytik Jena, Jena, Germany). Blood biochemistry was done with an automatic biochemical analyzer (HITACHI7600-020/7170A; Hitachi, Tokyo, Japan). The expression of eIF3 and its target genes was measured by quantitative real-time PCR.

Urine samples were collected from all participants and transferred into a metal-free polyethylene bottle as per the guidelines of clinical chemistry division of International Union of Pure and Applied Chemistry. These samples were diluted with an equal volume of 0.3 mol L–1 HNO3 and stored at 4 °C until further analysis. The urine Cd concentration was measured by atomic absorption spectrometry (ZEENIT700; Analytik Jena, Jena, Germany). The Cd standard curve was linear up to 25 μg L–1 and the detection limit was 0.33 μg L–1. The internal standard of Cd was added to urine and analyzed, and a recovery rate of 98.2% was found. Urinary beta2-microglobulin (Uβ2-MG) levels were measured by using radio-immunoassay (RIA) (Pharmacia β2-micro RIA, Pharmacia Diagnostics AB, Sweden).

All the human experiments were performed in accordance with the principles of the Declaration of Helsinki. All experimental protocols were approved by Research Ethic Committee of Guangzhou Medical University. The personal information of samples involved in the study was not opened.

Statistical analysis

All the data are represented as rate (%) or mean ± standard deviation (SD; x[combining macron] ± s) of three or more independent experiments. Comparisons were done using the chi square test for rate (%) from several independent experiments and Student's t-test or analysis of variance (ANOVA) followed by Dunnett's test for mean ± SD. The correlation of the two groups was tested by Pearson or Spearman's correlation analysis. Statistical analysis was performed with SPSS version 13.0 software. A value of P < 0.05 was considered statistically significant.

Results

Abnormally high eIF3 expression in CdCl2 transformed 16HBE cells

Real time PCR was performed to detect the eIF3 mRNA expression in CdCl2 transformed 16HBE cells at different stages. Results showed that the eIF3 mRNA expression increased over time in CdCl2 transformed 16HBE cells. The eIF3 mRNA expression in 16HBE cells of the 5th passage, 16HBE cells of the 15th passage, and 16HBE cells of the 35th passage was 3.6, 3.0 and 9.1 times that in the control group of the same passage numbers (P < 0.05). These suggest that there is an abnormally high eIF3 expression in CdCl2 transformed 16HBE cells.

Regulatory effects of eIF3 on the invasion and migration of CdCl2 transformed 16HBE cells

The invasion and migration of CdCl2 transformed 16HBE cells increased gradually, while the invasion and migration reduced markedly in 16HBE cells, 16HBE cells of the 15th passage, and 16HBE cells of the 35th passage after eIF3 silencing, accompanied by changes in the mRNA expressions of genes related to cell migration and invasion. Of these genes, EGFR, STAT3, FOX2 and VIMENTIN showed markedly reduced mRNA expression, but E-CADHERIN and AKT presented with significantly increased mRNA expression (P < 0.05). Results showed that the invasion and migration of CdCl2 transformed 16HBE cells were abnormal and eIF3 may exert regulatory effects on the migration and invasion of these cells (Fig. 1A–C).

Fig. 1. eIF3 affects cell migration and invasion in Cd-transformed cells based on transwell and wound-healing assay. Untreated control 16HBE cells, Cd-transformed 15th passage and Cd-transformed 35th passage cells were knock-down eIF3. (A) Representative photographs of migratory cells on the membrane (magnification, 100×). (B) Representative photographs of invasion cells on the membrane (magnification, 100×). The right panels of each row in A and B were the average cell number of triplicate. *P < 0.05 vs. control cells (t test). (C) The mRNA expression of cell migration and invasion related genes were detected using qPCR. * P < 0.05 vs. control cells (t test).

Fig. 1

Open in a new tab

Regulatory effects of eIF3 on the apoptosis of CdCl2 transformed 16HBE cells

Flow cytometry was performed to detect the apoptosis of 16HBE cells before and after eIF3 silencing. When compared with the control group, the number of apoptotic cells reduced markedly in 16HBE cells, 16HBE cells of the 15th passage and 16HBE cells of the 35th passage after eIF3 silencing, accompanied by changes in the mRNA expressions of apoptosis related genes. Of these genes, BAX and caspase-3 showed markedly reduced mRNA expression, and BCL-2 had significantly increased mRNA expression (P < 0.05). These results suggest that eIF3 is able to regulate the apoptosis of CdCl2 transformed 16HBE cells (Fig. 2A–C).

Fig. 2. eIF3 affects cell apoptosis in Cd-transformed cells. Untreated control 16HBE cells, Cd-transformed 15th passage and Cd-transformed 35th passage cells were knock-down eIF3. Cell apoptosis was assayed after 72 h treated with eIF3-siRNA by flow cytometry using Annexin-V staining (A), data are expressed as the cell apoptosis rate (B), and the mRNA expression of cell apoptosis related genes was detected using qPCR. *P < 0.05 vs. control cells (t-test).

Fig. 2

Open in a new tab

Sequencing of exons and promoter of eIF3 in CdCl2 transformed 16HBE cells

The whole exons and promoter were sequenced in the un-transformed 16HBE cells (control group) and the Cd-transformed cells of the 5th passage, 15th passage and 35th passage (11 segments in exons and 2 segments in promoter). One known SNP was identified, and point mutation was not found. These suggest that the abnormal eIF3 expression is not related to the eIF3 gene mutation in CdCl2 transformed 16HBE cells.

Sequencing for eIF3 methylation in CdCl2 transformed 16HBE cells

The predicted CpG island was subjected to expansion and sequencing after pre-treatment in the un-transformed 16HBE cells (control group), and the Cd-transformed cells of the 5th passage, 15th passage and 35th passage (2 pairs of primers used for expansion of 2 segments [250 bp and 202 bp]). Results showed there was no gene methylation in 4 types of cells. This suggests that abnormal eIF3 expression is not related to DNA methylation in CdCl2 transformed 16HBE cells.

Bioinformatics analysis of eIF3 in CdCl2 transformed 16HBE cells

The mRNA, microRNA and lncRNA differential expression profiling (Tables S1–S3†) had been done in CdCl2 transformed 16HBE cells in our previous studies. On the basis of these expression profiles, cytoscope software was employed to construct the network diagram of eIF3-mRNA co-expression (Table S4†). Results showed, in CdCl2 transformed 16HBE cells, eIF3 was related to the genes involved in the proliferation, differentiation, apoptosis, invasion and migration (Fig. 3A). According to the correlation of targets of miRNA–mRNA and miRNA–lncRNA, cytoscope software was used to delineate the ceRNA regulatory network of lncRNA–miRNA-eIF3 (Table S5†). Results showed that hsa-miR-3941 was able to regulate the eIF3 expression, and hsa-miR-3941 served as a response element or endogenous RNA (ceRNA). There were 14 lncRNAs in AL096700.2, RP11-213G2.4 and DHX9P1, which were involved in the regulation of eIF3 (Fig. 3B).

Fig. 3. Bioinformatics analysis of eIF3 in Cd-induced 35th cells as compared to untreated 16HBE cells. eIF3-mRNA network (A) and LncRNA–miRNA-eIF3 network (B) were constructed based on the correlation analysis between differentially expressed lncRNAs, microRNA mRNAs and in Cd-induced 35th 16HBE cells as compared to untreated 16HBE cells.

Fig. 3

Open in a new tab

Correlation between eIF3 mRNA expression and Cd exposure in rats with chronic Cd exposure

eIF3 mRNA expression increased significantly in rats chronically exposed to low-dose, mid-dose and high-dose Cd, and this increase was Cd-dose dependent (P < 0.05). The eIF3 mRNA expression in the kidney, heart and lung in the low-dose group was comparable to that in the control group (P > 0.05) (Fig. 4). The eIF3 mRNA expression in the blood was positively related to the blood Cd content (r = 0.383, p = 0.440), Cd content of the urine (r = 0.427, P = 0.037), liver (r = 0.343, P = 0.320), kidney (r = 0.515, P = 0.020), heart (r = 0.400, P = 0.023) and lung (r = 0.402, P = 0.280).

Fig. 4. eIF3 mRNA expression in the kidney, liver, lung and heart of cadmium-exposed rats using the real time qPCR assay. The gene expression was validated in the kidney, liver, lung and heart of cadmium-exposed rats of the control, low dose, mid-dose and high dose cadmium-exposed rats by qPCR and normalized to that of β-actin. Data are expressed as mean ± SE. *p < 0.05 compared to the corresponding control group.

Fig. 4

Open in a new tab

Correlation of eIF3 mRNA expression with liver and kidney injury in rats chronically exposed to Cd

To explore the correlation of eIF3 mRNA expression with liver and kidney injury in Cd toxicity, Pearson correlation analysis was used. Results showed that the eIF3 mRNA expression in the rat blood was positively associated with sCr (r = 0.419, P = 0.033), BUN (r = 0.473, P = 0.017), 24 h Pro (r = 0.422, P = 0.004) and 24 h UCr (r = 0.475, P = 0.016).

Health status of the workers exposed to Cd

The subjects (median age, 31 years) were directly or indirectly exposed to Cd for less than 2 years with no history of exposure to other toxins. Only non-smokers were included in the present study. Urine Cd concentration normalized to the urine creatinine (Cr) showed a normal distribution. The median, maximum and minimum urine Cd concentrations were 1.61, 113.86 and 0.31 μg g–1 Cr, respectively. The 25th percentile and 75th percentile of urine Cd concentrations were 0.69 and 9.54 μg g–1 Cr, respectively. According to the urine Cd concentration, subjects were divided into three groups: [circle containing 1] ≤2 μg g–1 Cr, [circle containing 2] 2–5 μg g–1 Cr, and [circle containing 3] >5 μg g–1 Cr. The age, gender and years of employment were comparable among the three groups, suggesting that our results were not confounded by these factors.

eIF3 Expression was correlated with Cd exposure in Cd-exposed workers

In order to evaluate whether eIF3 serves as a biomarker of Cd exposure, the expression of eIF3 in the blood of Cd-exposed workers was detected by quantitative real-time PCR. According to the urine Cd concentration and blood Cd concentration, these studies were divided into 3 groups. The blood eIF3 expression increased with the increase in urine Cd concentration and blood Cd concentration. The eIF3 expression in workers with urine Cd concentration at 2–5 μg g–1 Cr and >5 μg g–1 Cr was 1.310 and 5.581 times that in the control group (urine Cd concentration: ≤2 μg g–1 Cr) (P < 0.05). A similar finding was identified in blood eIF3 expression in workers with different blood Cd concentrations (1.659-fold and 6.119-fold) when compared with the control group (P < 0.05) (Table 2). There was a significant positive correlation of eIF3 expression with blood Cd concentration (r = 0.713, P < 0.0001), urine Cd concentration (r = 0.459, P = 0.003) and urine β2-MG concentration (r = 0.501, P = 0.001) (Fig. 5). These findings indicate that eIF3 expression is correlated with Cd exposure in Cd-exposed workers.

Table 2. Blood eIF3 expression at different Cd exposure levels in Cd-exposed workers.

Exposure to Cd at different levels N eIF3 F P value
UCd levels
[circle containing 1] ≤2 μg g–1 Cr 150 0.979 ± 1.007 16.847 Inline graphic < 0.0001
[circle containing 2] 2–5 μg g–1 Cr 17 1.472 ± 1.149 P [circle containing 1][circle containing 2] = 0.101
[circle containing 3] >5 μg g–1 Cr 14 3.304 ± 2.338 P [circle containing 1][circle containing 3] < 0.0001
P [circle containing 2][circle containing 3] < 0.0001
 
Urine β2-MG
≤500 μg g–1 Cr 147 0.921 ± 0.944 17.684 Inline graphic < 0.0001
500–1000 μg g–1 Cr 19 1.843 ± 1.636 P [circle containing 1][circle containing 2] = 0.001
>1000 μg g–1 Cr 14 3.351 ± 1.987 P [circle containing 1][circle containing 3] < 0.0001
P [circle containing 2][circle containing 3] < 0.0001
 
BCd level
≤2 μg l–1 131 0.771 ± 0.578 19.134 Inline graphic < 0.0001
2–5 μg l–1 35 1.421 ± 1.099 P [circle containing 1][circle containing 2] < 0.0001
>5 μg l–1 15 3.944 ± 2.156 P [circle containing 1][circle containing 3] < 0.0001
P [circle containing 2][circle containing 3] < 0.0001
Open in a new tab

Fig. 5. Correlation analysis between eIF3 expression and Cd concentration in Cd-exposed workers. Correlation analysis between eIF3 expression and blood Cd concentration (A), urine Cd concentration (B) and urine β2-MG concentration (C). Blood eIF3 expression was calculated by the ratio of its expression to that of β-actin. The urine cadmium concentration was normalized by urine creatinine (μg L–1 Cr) and urine β2-MG (μg g–1 Cr). The linear relationship was analyzed by Pearson correlation analysis.

Fig. 5

Open in a new tab

eIF3 Expression was correlated with the target genes in Cd-exposed workers

There was a significant positive correlation between eIF3 expression and the expression of EGFR, STAT3, FOXC2, VIMENTIN, CASPASE-3 and BAX, while a significant negative correlation was noted between eIF3 expression and expression of E-CADHERIN, AKT and BCL-2 in Cd-exposed workers. In addition, the associations between eIF3 expression and expression of target genes were further evaluated after adjusting the urine Cd concentration and blood Cd concentration (two factors affecting eIF3 expression and its target gene expression) in Cd-exposed workers. There was also a significant correlation between eIF3 expression and its target gene expression (Table 3). These findings indicate that eIF3 expression correlates very well with its target gene expression in Cd-exposed workers.

Table 3. Correlation analysis between eIF3 expression and target gene expression in Cd-exposed workers.

Target genes Correlation coefficient (r) P Target genes Correlation coefficient (r) P
EGFR 0.714 <0.0001 AKT –0.417 0.003
STAT3 0.671 0.002 BCL-2 –0.687 <0.0001
FOXC2 0.418 <0.0001 CASPASE-3 0.684 <0.0001
VIMENTIN 0.758 <0.0001 BAX 0.710 0.001
E-CADHERIN –0.513 0.012
Open in a new tab

Discussion

Many cellular and molecular events are involved in the toxic effects of chemical carcinogens in humans.29,30 It is well recognized that an abnormal expression of eukaryotic translation factors facilitates the malignant transformation of primary cells and promotes carcinogenesis.31,32 Accumulating evidence has demonstrated the important roles of over-expression of translation factors in many cancers including pancreatic cancer, colon cancer, breast cancer, lung cancer and prostate cancer.33–35 Enhanced expression of eIF3 has been found in a variety of transformed cells, tumor cells and cancer tissues. Our previous study has identified eIF3 as a novel mouse cadmium-responsive proto-oncogene.25,26 In addition, our previous results also revealed that the expression of human eIF3 increased in CdCl2 transformed 16HBE cells at different stages. This study investigated whether eIF3 modulated cell apoptosis, migration and invasion in Cd-induced toxicity, and examined the eIF3 expression in Cd exposed rats and workers exposed to Cd. The present study further demonstrated that eIF3 could be a potential biomarker regulating cell apoptosis, migration and invasion in Cd toxicity and Cd-induced carcinogenesis.

A protein family involved in the translation initiation of proteins, and their changes in the structure and function may increase the sensitivity of cells to transformation, resulting in cell transformation and acquisition of tumorigenic ability. Studies have confirmed that high eIF3 expression may cause malignant transformation of cells, and antisense eIF3 expression reverses the carcinogenesis of Cd-transformed cells.36,37 However, the consequence of eIF3 siRNA is still unclear in Cd toxicity. To verify the role of eIF3 in Cd toxicity, the expression of eIF3 in untreated 16HBE cells and Cd-induced 35th cells was knocked down with small interfering RNA. Results showed that eIF3 knockdown significantly inhibited the cell apoptosis, migration and invasion in Cd-transformed 16HBE cells. Moreover, eIF3 knockdown also altered the mRNA expression of genes related to cell apoptosis, migration and invasion (EGFR, STAT3, FOXC2, VIMENTIN, E-CADHERIN, AKT, BCL-2, CASPASE-3 and BAX) in Cd-transformed cells. Our results suggest that eIF3 silencing may inhibit the apoptosis and compromise the malignancy of Cd-transformed cells.

There were pieces of evidence showing that the abnormal gene expression in Cd toxicity was related to the alteration of genetics and epigenetics. O'Connor et al.38 found that eIF3 mutation altered the initiation codon and the recognition of initiation tRNAs, resulting in abnormal regulation and deficiency and a mutation of the eIF3 gene could cause malignant transformation of cells. Nina et al.39 found that the mutation of eIF3-P40 was related to breast cancer and prostate cancer. However, the abnormally expressed mechanisms of eIF3 are still unclear and have never been reported. In the present study, the exons and promoter of the eIF3 gene were sequenced in untreated cells and Cd-transformed cells, and 1 known SNP was identified in the eIF3 gene, but point mutation was not found. The sequencing and expansion of the predicted CpG of the eIF3 gene failed to show the DNA methylation. In our previous study, lncRNA, miRNA and mRNA expression profiling was performed and differentially expressed lncRNA, miRNA and mRNA were screened. In this study, bioinformatics analysis was employed to construct the network diagram of eIF3-mRNA co-expression and the ceRNA regulatory network of lncRNA–miRNA-eIF3. The results showed that hsa-miR-3941 had a regulatory effect on eIF3 in Cd-transformed cells, hsa-miR-3941 served as a response element or endogenous RNA (ceRNA), and 14 lncRNAs in AL096700.2, RP11-213G2.4 and DHX9P1 were related to the regulation of eIF3. The above findings suggested that the abnormal expression of eIF3 in Cd toxicity was related to lncRNA, which was warranted to be confirmed in future studies.

Many cellular and molecular events are involved in the toxic effects of chemical carcinogens,29,40 but few studies have been conducted to investigate eIF3 as a new biomarker of Cd exposure. The present study was undertaken to investigate the role of eIF3 in Cd toxicity in animals and Cd-exposed workers. The animal model of chronic Cd exposure used in this study was established by continuous intra-peritoneal injection of CdCl2 for 14 weeks. The Cd toxicity was evaluated by the weight coefficient, histo-pathological examination and liver and renal function (ALT, AST, sCr, BUN and 24 h Pro) detection. The blood Cd concentration reflects the recent exposure, and urine Cd concentration represents the whole body burden after a long-term exposure, while tissue Cd concentration reflects the Cd accumulation and organ damage.41–43 In the present study, the eIF3 expression in the kidney, liver, lung and heart of Cd-treated rats was positively correlated with the Cd exposure and the severity of organ damage, suggesting that eIF3 reflects the accumulation of Cd in the body and the severity of organ damage. eIF3 expression in the body is useful in predicting the Cd-induced toxicity.

In addition, the eIF3 expression was also detected in the blood and urine of workers chronically exposed to Cd. Results showed a strong positive correlation of blood eIF3 expression with the urine Cd and the expression of eIF3 target genes, suggesting that blood eIF3 expression is potentially a novel biomarker of Cd-exposure in humans. The workers with detectable urine Cd exhibit a significantly higher blood eIF3 expression than those with undetectable urine Cd, suggesting that, even at a lower urine Cd concentration, the blood eIF3 expression may reflect the alteration in Cd accumulation.

Conclusion

Our study indicates that eIF3 is able to regulate cell migration, invasion and apoptosis in the presence of Cd toxicity. In addition, eIF3 may be a novel and valuable biomarker of Cd toxicity, and may become a significant biomarker for field investigations and risk assessment in humans exposed to occupational and environmental Cd.

Conflict of interest

The authors have declared that no competing interests exist.

Supplementary Material

Click here for additional data file. (1.4MB, xls)
Click here for additional data file. (1.1MB, xls)
Click here for additional data file. (233KB, xls)
Click here for additional data file. (1.7KB, xls)
Click here for additional data file. (3.2KB, xls)

Acknowledgments

The authors thank the officers for their support and assistance in the coordination of this study, the organization of the field work and providing some background information, and thank all respondents for their cooperation. This work was supported by the National Natural Science Foundation of China (Lei YX: 81373038 and 81072322; Zhou ZH: 81202236 and 81473001); Science and Technology Planning Project of Guangdong Province, China (Lei YX: 2013B021800095 and Zhou ZH: 2013B021800093); The Training Project of Outstanding Young College Teachers in Guangdong Province, China (no. Yq2013138). Science Foundation of Guangzhou Bureau of Education (Zhou ZH: 1201410830).

Footnotes

†Electronic supplementary information (ESI) available. See DOI: 10.1039/c5tx00250h

References

  1. Diao W. P., Ni W. Z., Ma H. Y., Yang X. E. Bull. Environ. Contam. Toxicol. 2005;75:731–738. doi: 10.1007/s00128-005-0812-y. [DOI] [PubMed] [Google Scholar]
  2. Swaddiwudhipong W., Mahasakpan P., Funkhiew T., Limpatanachote P. J. Med. Assoc. Thailand. 2010;93:1217–1222. [PubMed] [Google Scholar]
  3. Thun M. J., Elinder C. G., Friberg L. Am. J. Ind. Med. 1991;20:629–642. doi: 10.1002/ajim.4700200506. [DOI] [PubMed] [Google Scholar]
  4. Koyu A., Gokcimen A., Ozguner F., Bayram D. S., Kocak A. Mol. Cell. Biochem. 2006;20:1–5. doi: 10.1007/s11010-005-9017-2. [DOI] [PubMed] [Google Scholar]
  5. Satarug S., Garrett S. H., Sens M. A., Sens D. A. Environ. Health Perspect. 2010;118:182–190. doi: 10.1289/ehp.0901234. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Schöpfer J., Drasch G., Schrauzer G. N. Biol. Trace Elem. Res. 2010;134:180–187. doi: 10.1007/s12011-010-8636-y. [DOI] [PubMed] [Google Scholar]
  7. Fay R. M., Mumtaz M. M. Food Chem. Toxicol. 1996;34:1163–1165. doi: 10.1016/s0278-6915(97)00090-2. [DOI] [PubMed] [Google Scholar]
  8. Benbrahim-Tallaa L., Tokar E. J., Diwan B. A., Dill A. L., Coppin J. F., Waalkes M. P. Environ. Health Perspect. 2009;117:147–152. doi: 10.1289/ehp.0900999. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Kriegel A. M., Soliman A. S., Zhang Q., El-Ghawalby N., Ezzat F., Soultan A., Abdel-Wahab M., Fathy O., Ebidi G., Bassiouni N. Environ. Health Perspect. 2006;114:113–119. doi: 10.1289/ehp.8035. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Eriksen K. T., Halkjær J., Meliker J. R., McElroy J. A., Sørensen M., Tjønneland A., Raaschou-Nielsen O. BMC Cancer. 2015;15:177. doi: 10.1186/s12885-015-1153-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Song J. K., Luo H., Yin X. H., Huang G. L., Luo S. Y., Lin D. R., Yuan D. B., Zhang W., Zhu J. G. Sci. Rep. 2015;5:17976. doi: 10.1038/srep17976. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Kim H. S., Kim Y. J., Seo Y. R. J. Cancer Prev. 2015;20:232–240. doi: 10.15430/JCP.2015.20.4.232. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Huff J., Lunn R. M., Waalkes M. P., Tomatis L., Infante P. F. Int. J. Occup. Environ. Health. 2007;13:202–212. doi: 10.1179/oeh.2007.13.2.202. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. The International Agency for Research on Cancer (IARC), Beryllium, Cadmium, Mercury and Exposures in the Glass Manufacturing Industry, World Health Organization International Agency for Research on Cancer, Lyon, France, 1993. [Google Scholar]
  15. Waisberg M., Joseph P., Hale B., Beyersmann D. Toxicology. 2003;192:95–117. doi: 10.1016/s0300-483x(03)00305-6. [DOI] [PubMed] [Google Scholar]
  16. Cherian M. G., Huang P. C., Klaassen C. D., Liu Y. P., Longfellow D. G., Waalkes M. P. Cancer Res. 1993;53(4):922–925. [PubMed] [Google Scholar]
  17. Zimmer S. G., DeBenedetti A., Graff J. R. Anticancer Res. 2000;20:1343–1352. [PubMed] [Google Scholar]
  18. Asano K., Clayton J., Shalev A., Hinnebusch A. G. Genes Dev. 2000;4:2534–2546. doi: 10.1101/gad.831800. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Phan L., Schoenfeld L. W., Valasek L. EMBO J. 2001;20:2954–2965. doi: 10.1093/emboj/20.11.2954. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Etchison D., Smith K. J. Biol. Chem. 1990;265:7492–7500. [PubMed] [Google Scholar]
  21. Lamphear B. J., Kirchweger R., Skern T., Rhoads R. E. J. Biol. Chem. 1995;270:21975–21983. doi: 10.1074/jbc.270.37.21975. [DOI] [PubMed] [Google Scholar]
  22. Nupponen N. N., Porkka K., Kakkola L. Am. J. Pathol. 1999;154(6):1777–1783. doi: 10.1016/S0002-9440(10)65433-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Rothe M., Ko Y., Albers P., Wernert N. Am. J. Pathol. 2000;157:1597–1604. doi: 10.1016/S0002-9440(10)64797-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Mayeur G. L., Hershey J. W. FEBS Lett. 2002;514:49–54. doi: 10.1016/s0014-5793(02)02307-4. [DOI] [PubMed] [Google Scholar]
  25. Joseph P., Lei Y. X., Whong W. Z., Ong T. M. Cancer Res. 2002;62:703–707. [PubMed] [Google Scholar]
  26. Lei Y. X., Joseph P., Ong T. M. Teratog., Carcinog., Mutagen. 2002;22:403–409. doi: 10.1002/tcm.10037. [DOI] [PubMed] [Google Scholar]
  27. Lei Y. X., Wei L., Wang M., Wu G. R., Li M. Biomed. Environ. Sci. 2008;21:332–338. doi: 10.1016/S0895-3988(08)60051-3. [DOI] [PubMed] [Google Scholar]
  28. Qian L., Lei Y. X., He C. C., Lei Z. N. Int. J. Mol. Sci. 2013;14:5182–5197. doi: 10.3390/ijms14035182. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Thomas R. S., Bao W., Chu T. M. M., Bessarabova T., Nikolskaya T., Nikolsky Y. Toxicol. Sci. 2009;112:311–321. doi: 10.1093/toxsci/kfp233. [DOI] [PubMed] [Google Scholar]
  30. Vogelstein B., Kinzler R. W. Trends Genet. 1993;9:138–141. doi: 10.1016/0168-9525(93)90209-z. [DOI] [PubMed] [Google Scholar]
  31. Lee T., Pelletier J. Future Med. Chem. 2012;4:19–31. doi: 10.4155/fmc.11.150. [DOI] [PubMed] [Google Scholar]
  32. Sonenberg N. Curr. Opin. Cell Biol. 1993;5:955–960. doi: 10.1016/0955-0674(93)90076-3. [DOI] [PubMed] [Google Scholar]
  33. Khoury T., Alrawi S., Ramnath N., Li Q., Grimm M., Black J. Clin. Lung Cancer. 2009;10:58–66. doi: 10.3816/CLC.2009.n.009. [DOI] [PubMed] [Google Scholar]
  34. Siebke C., James T. C., Cummins R., O'Grady T., Kay E., Bond U. Cell Stress Chaperones. 2012;17:145–156. doi: 10.1007/s12192-011-0295-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Yoshizawa A., Fukuoka J., Shimizu S., Shilo K., Franks T. J., Hewitt S. M. Clin. Cancer Res. 2010;6:240–248. doi: 10.1158/1078-0432.CCR-09-0986. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Marchetti A., Buttitta F., Pellegrini S., Bertacca G., Callahan R. Int. J. Oncol. 2001;18:175–179. doi: 10.3892/ijo.18.1.175. [DOI] [PubMed] [Google Scholar]
  37. Eberle J., Krasagakis K., Orfanos C. E. Int. J. Cancer. 1997;71:396–401. doi: 10.1002/(sici)1097-0215(19970502)71:3<396::aid-ijc16>3.0.co;2-e. [DOI] [PubMed] [Google Scholar]
  38. O'Connor M., Gregory S. T., Rajbhandary U. L., Dahlberg A. E. RNA. 2001;7:969–978. doi: 10.1017/s1355838201010184. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Nina N. N., Kati P., Laura K., Minna T., Karin P., Ake B., Jorma I., Tapio V. AJP. 1999;154:1777–1783. [Google Scholar]
  40. Weng S., Wang W., Li Y., Li H. Toxicol. Lett. 2014;225(3):367–377. doi: 10.1016/j.toxlet.2014.01.012. [DOI] [PubMed] [Google Scholar]
  41. Lei Y. X., Wang M., Wei L., Li M., Lin H. Z. Biomed. Environ. Sci. 2010;23:151–157. doi: 10.1016/s0895-3988(10)60045-1. [DOI] [PubMed] [Google Scholar]
  42. Ikeda M., Ohashi F., Fukui Y., Sakuragi S., Moriguchi J. Int. Arch. Occup. Environ. Health. 2011;84:121–129. doi: 10.1007/s00420-010-0527-1. [DOI] [PubMed] [Google Scholar]
  43. Järup L., Akesson A. Toxicol. Appl. Pharmacol. 2009;238:201–208. doi: 10.1016/j.taap.2009.04.020. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Click here for additional data file. (1.4MB, xls)
Click here for additional data file. (1.1MB, xls)
Click here for additional data file. (233KB, xls)
Click here for additional data file. (1.7KB, xls)
Click here for additional data file. (3.2KB, xls)

Articles from Toxicology Research are provided here courtesy of Oxford University Press

RESOURCES