Comparative Microarray Analysis of Basal Gene Expression in Mouse Hepa-1c1c7 Wild-Type and Mutant Cell Lines
http://www.100md.com
《毒物学科学杂志》
Department of Biochemistry and Molecular Biology, National Food Safety and Toxicology Center, Center for Integrative Toxicology
Department of Pharmacology and Toxicology Michigan State University, East Lansing, Michigan, 48824
ABSTRACT
Hepa-1c1c7 wild-type and benzo[a]pyrene-resistant derived mutant cell lines have been used to elucidate pathways and mechanisms involving the aryl hydrocarbon receptor (AhR). However, there has been little focus on other biological processes which may differ between the isolated lines. In this study, mouse cDNA microarrays representing 4858 genes were used to examine differences in basal gene expression between mouse Hepa-1c1c7 wild-type and c1 (truncated Cyp1a1 protein), c4 (AhR nuclear translocator, ARNT, deficient), and c12 (low AhR levels) mutant cell lines. Surprisingly, c1 mutants exhibited the greatest number of gene expression changes compared to wild-type cells, followed by c4 and c12 lines, respectively. Differences in basal gene expression were consistent with cell line specific variations in morphology, mitochondrial activity, and proliferation rate. MTT and direct cell count assays indicate both c4 and c12 mutants exhibit increased proliferative activity when compared to wild-type cells, while the c1 mutants exhibited decreased activity. This study further characterizes Hepa-1c1c7 wild-type and mutant cells and identifies significant differences in biological processes that should be considered when conducting comparative mechanistic studies with these lines.
Key Words: benzo[a]pyrene; aryl hydrocarbon receptor; Hepa-1c1c7.
INTRODUCTION
The mouse Hepa-1c1c7 cell line was derived from a BW 7756 hepatoma which arose in a C57L mouse and propagated in C57L/J mice (Bernhard et al., 1973). Although parent Hepa-1 cells lack some liver-specific activities (Darlington et al., 1980), the isolated Hepa-1c1c7 line exhibits hepatic transferrin secretion (Papaconstantinou et al., 1978) and haptoglobin synthesis (Baumann and Jahreis, 1983). However, the vast majority of studies use Hepa-1 and Hepa-1c1c7 cells to examine their characteristic highly inducible aryl hydrocarbon hydroxylase activity (AHH) (Hankinson, 1979).
Hepa-1c1c7 mutants were selected and isolated based on their resistance to benzo[a]pyrene (B[a]P) toxicity and compromised AHH induction (Bartholomew et al., 1975; Hankinson, 1979; Thompson et al., 1974). Three isolated complementation groups (A, B, and C) were identified and used to investigate the mechanism of action of the aryl hydrocarbon receptor (AhR) (Van Gurp and Hankinson, 1984). C1 mutants (Group A) are characterized by a deficiency in Cyp1a1 activity as a result of a premature stop codon located after Asn 414, resulting in a loss of a C-terminal heme-binding site necessary for enzyme activity (Kimura et al., 1987). C12 mutants (Group B) express only 3% of the AHH activity of wild-type cells (Van Gurp and Hankinson, 1984) due to compromised AhR expression, possibly from a defective factor involved in chromatin remodeling (Hankinson et al., 1985; Zhang et al., 1996). Hepa-1c1c7 c4 (Group C) mutant cells are devoid of any AHH activity, due to a G326D point mutation in the ARNT (aryl hydrocarbon receptor nuclear translocator) coding region, which causes a marked reduction in DRE sequence binding by the AhR/ARNT complex (Numayama-Tsuruta et al., 1997).
Studies using Hepa-1c1c7 wild-type and mutant cells have significantly contributed to the elucidation of the mechanism of AhR-mediated gene expression (Boverhof et al., 2005; Hoffman et al., 1991; Probst et al., 1993; Reyes et al., 1992), including the identification of responsive detoxification genes (Liu et al., 1994). However, isogenicity of loci beyond Cyp1a1 is not expected, given the selection method and criteria used for mutant cell line isolation. Despite their use in numerous studies (Jeong and Lee, 1998; Merchant et al., 1992; Miranda et al., 2000; Pollenz, 1996; Seubert et al., 2002; Zacharewski et al., 1994), only one has examined differences in constitutive gene expression between Hepa-1c1c7 wild-type and mutant cells using differential display PCR in wild-type and the related but independently derived ARNT-deficient Hepa-1 cell line derivative, BPRc1 (Seidel and Denison, 1999). Our study uses cDNA microarrays to further examine differences in basal gene expression in mutant and wild-type cells with regards to variant morphology and growth rates. Differences in basal gene expression were also linked to cellular physiological observations to identify mechanistic networks that may be affected in the mutant hepatoma cell lines. These data indicate that differences between wild-type and mutant cells may not be attributed solely to differences in AhR/ARNT functionality and Cyp1a1 inducibility.
MATERIALS AND METHODS
Cell maintenance and sample collection.
Hepa-1c1c7 wild-type and mutant cells (c1, c4, and c12) (gifts from O. Hankinson, University of California, Los Angeles) were maintained in phenol-red-free DMEM/F12 media (Invitrogen, Carlsbad, CA) supplemented with 5% fetal bovine serum (FBS) (Serologicals Corporation, Norcross, GA), 50 μg/ml gentamycin, 2.5 μg/ml amphotericin B, 100 U/ml penicillin, and 100 μg/ml streptomycin (Invitrogen, Carlsbad, CA). 1 x 106 cells were seeded into T75 culture flasks (Sarstedt, Newton, NC) and harvested at 90% cell confluency by scraping in Trizol (Invitrogen, Carlsbad, CA). RNA was isolated as per manufacturer's instructions and stored in RNA storage solution (Ambion Inc., Austin, TX) at –80°C. RNA quality was examined on a denaturing 1% agarose gel as well as by A260/280 ratio.
Images were captured with Nikon Eclipse TE300 microscope and GelExpert software (San Carlos, CA). Measurements were conducted using ImageJ v132j (NIH, USA) and significance determined using SAS version 8.0. One-way ANOVA analysis followed by Dunnett's t-tests were performed on measurements where = 0.05.
Cell growth rate assessments.
For MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] assays, 10,000 cells were seeded in a 96-well tissue-culture plate (Corning, MI) in 5% FBS and antibiotic supplemented media. Thaizylol blue (Sigma, St. Louis, MO) was added to cells, incubated for 3 h at 37.5°C and 5% CO2, and absorbance at 595 nm was determined using Softmax software (Molecular Devices, Sunnyvale, CA) at 3, 6, 12, 24, 36, and 48 h on an Emax 96-well microplate reader (Molecular Devices, Sunnyvale, CA). Experiments were completed in quadruplicate. Relative viability was standardized to the blank wells containing media only.
For direct cell counts, 3 x 105 cells were seeded in T25 culture flasks (Corning, MI) in 5% FBS and antibiotic supplemented media. Cells were incubated at 37.5°C and 5% CO2 for 6, 12, 24, and 48 h. Cells were then trypsinized and subjected to manual counting, in duplicate, using a hemocytometer (Hausser Scientific Co., Horsham, PA). Experiments were completed in quadruplicate.
Significance was determined using SAS v.8.0. Two-way ANOVA analysis followed by Tukey's HSD post hoc test was performed on MTT assay data where = 0.05. For the direct cell counts, a repeated measures ANOVA was performed followed by a Tukey's HSD post hoc test where = 0.05. Comparisons were made between each mutant and wild-type cell line at each time point.
Microarray processing.
cDNA arrays containing 6376 features (representing 4858 unique genes based on mouse Unigene build 138), were spotted in duplicate, on epoxy-coated glass slides (SCHOTT Nexterion, Germany) using an Omnigrid arrayer (GeneMachines, San Carlos, CA) equipped with 16 (4 x 4) Chipmaker 2 pins (Telechem International Inc., Sunnyvale, CA) at the Genomics Technology Support Facility (http://genomics.msu.edu). Three biological replicates for each cell line were examined using a 6-slide independent reference design that incorporated dye-swap hybridizations (Fig. 1) for a total of 18 microarray hybridizations analyzed. Clones for the array were initially selected for the examination of testis-specific genes, and the library has since been expanded to include genes known to be responsive to estrogens, androgens, and dioxin. Selected clones were obtained from EPAMAC sequence verified mouse clones (Rockett and Dix, 1999), Research Genetics IMAGE clones, and the National Institute of Aging 15K mouse clone set.
Detailed protocols for processing of microarrays are available at http://dbzach.fst.msu.edu:8050/dbZachProtocolsInterfaceVersions/ProtocolsInterfaceFront. Briefly, 15 μg of total RNA were reverse transcribed with Superscript II (Invitrogen, Carlsbad, CA) to incorporate Cy3- or Cy5-conjugated dUTP (Amersham Biosciences, Piscataway, NJ). Labeled samples were mixed, purified on a Qiagen PCR purification column (QIAGEN, Valencia, CA), dried down using a vacuum centrifuge, and resuspended in 33 μl of hybridization buffer (56% formamide, 32% 20x SSPE, 8% 50x Denhardt's Solution, 4% 20% SDS, 20 μg poly(A), 20 μg mouse CoT-1 DNA, 10 μg yeast tRNA carrier) for overnight 42°C hybridization on arrays. Slides were washed sequentially in solutions containing SSC and decreasing concentrations of SDS and scanned using a 428 Affymetrix Scanner (Santa Clara, CA). Images were quantified using Analyzer DigitalGENOME (MolecularWare, Cambridge, MA) software. All data was stored in dbZach (http://dbzach.fst.msu.edu), a Minimum Information About Microarray Experiments (MIAME)-compliant relational database (Brazma et al., 2001) running under Windows 2003/Oracle 9i that currently supports microarray data storage, retrieval, and querying, as well as facilitates data analysis, sharing, and reporting (Burgoon et al., 2003; Mattes et al., 2004).
Quantitative RT-PCR.
A sample of total RNA isolated from each replicate was set aside for verification using SYBR Green-based quantitative real-time PCR (QRT-PCR). Briefly, 1 μg of RNA was primed by random hexamers (Roche, Switzerland) and reverse transcribed using Superscript II in a 20-μl reaction as described. From a 5-fold dilution of this cDNA solution, 3 μl was used in a 30-μl PCR reaction containing 1x SYBR Green PCR buffer, 3 mM MgCl2, 0.33 mM dNTPs, 0.5 IU AmpliTaq Gold (Applied Biosystems, Foster City, CA) and 0.15 mM forward and reverse primer. All primers were designed by submitting RefSeq sequences to Primer3 (http://www-genome.wi.mit.edu/cgi-bin/primer/primer3_www.cgi) to identify an approximately 125bp amplicon (Table 1). PCR amplification was conducted in 96-well MicroAmp Optical plates (Applied Biosystems) on an Applied Biosystems PRISM 7000 Sequence Detection System under the following conditions: 10 min denaturation and enzyme activation at 95°C, followed by 40 cycles of 95°C for 15 s and 60°C for 1 min. A 30-min dissociation protocol, after amplification, was conducted to assess primer specificity and product uniformity. Each plate contained duplicate standards of purified PCR product of known template concentration over the range of eight orders of magnitude to generate a log template concentration standard curve. No template controls (NTC) were included on each plate where samples with a Ct value within 2 SD of the mean Ct values for the NTCs were considered below the limits of detection. Plots were visualized and thresholds determined using ABI Prism 7000 SDS Software (Applied Biosystems, Foster City, CA). Results were normalized to beta-actin to control for differences in RNA loading, quality and cDNA synthesis. Statistical significance of expression differences between wild-type line and mutant lines were assessed using a standard one-way ANOVA followed by Tukey's HSD post hoc analysis using SAS version 8.0 (Cary, NC).
Microarray analysis and clustering.
A general linear mixed model was used to normalize microarray data (Boverhof et al., 2004) and generate t-value statistic prioritized lists for each mutant line. Lists were used to identify active features different from wild-type basal expression by filtering on a t-statistic cut-off (t > |2.77|). Hierarchical clustering of 723 genes was completed using the heatmap() function in R, version 1.9.0, and R, version 1.9.1, was used to compute the Pearson's correlation coefficient between DNA microarray data and QRT-PCR results.
RESULTS
Cellular Physiological Differences between Hepa-1c1c7 Wild-Type and Mutant Lines
Morphological comparisons were made between Hepa-1c1c7 wild type (WT) and mutant cell lines at comparable times and cell densities (Fig. 2). Wild-type and c1 mutant cells have a more compact cell body, whereas c4 and c12 mutant cell lines are narrower (p < 0.05) and thus appear more elongated (Table 2). The c4 and c12 mutant cells also possess significantly fewer number of processes (p < 0.05) extending from the main body; subsequently the wild-type and c1 cells appear more stellate in nature. In addition, c4 mutants exhibit longer process lengths (p < 0.05) compared to wild-type cells.
MTT colorimetric assays indicate significant (p < 0.05) differences in mitochondrial activity between wild-type and mutant cells at each time point (Fig. 3A). By 24 h, increased MTT activity was observed in c12 cell lines, while c1 cells exhibited decreased activity. At 36 h, c4 cells exhibit a statistically significant increase in mitochondrial activity. These trends were consistent throughout the time course.
Direct cell counts were conducted to assess proliferative rates (Fig. 3B). Although a significant decrease in proliferation rate was only observed in c1 cells at 48 h compared to wild type, generally, there was good trend agreement between MTT and direct cell count data. Statistical significance was not observed for c4 mutant cells; however, the trend observed indicates that a significant difference may be observed beyond 48 h. A more extensive time course was not conducted because the cells become confluent, thereby confounding proliferative kinetics.
Microarray Quality Assessment and Statistical Analysis
All arrays within this study were compared to a historical data set of established high-quality arrays. Background signal intensity, feature signal intensity, feature versus background signal intensity ratios, number of features with background intensities greater than the feature intensity for each array, and relationships between feature and background signal intensities were examined. Comparisons indicate that both background and signal intensities were higher than the historical data set; however, all arrays that were used for further analysis met the quality standard requirements (Burgoon et al., submitted).
Basal Gene Expression Differences between Hepa-1c1c7 and Wild-Type and Mutant Cell Lines
Statistical analysis using a model-based t-statistic test (t-value > |2.97|) identified a total of 946 active features (spots on an array) across c1, c4, and c12 mutant cell lines when compared to wild-type cells, 723 of which are unique and represent 499 unique LocusLink annotated genes. The purpose of the t-statistic is to rank and prioritize clones to be initially examined and a cut-off selected to generate a manageable clone list for additional functional investigation.
C1 mutants exhibited the largest number of differentially expressed basal level of genes (312) when compared to wild-type cells, followed by c4 (220) and c12 (134) cells, respectively (Fig. 4A). A total of 30 genes exhibited differential basal levels of expression in all mutant lines versus the wild type, while 63, 35, and 9 genes demonstrated differential basal levels in sets of two cell lines, c1 and c4 lines, c1 and c12 lines, and c4 and c12 lines, respectively, relative to wild-type (Fig. 4B).
Higher basal levels of gene expression represented approximately 60%, 70%, and 49% of the c1, c4, and c12 active gene lists, respectively (Fig. 4C). Gene expression differences ranged from the 3.3-fold increase in RIKEN cDNA 1700027N10 gene transcripts to 16.1-fold down regulation of albumin 1 (Alb1) in c1 cells relative to wild-type, 2.67-fold increase in B-cell translocation gene 1, anti-proliferative (Btg1) to 2.4-fold decrease in Alb1 in c4 cells relative to wild-type, and 2.3-fold increase in cytochrome P450, family 1, subfamily a, polypeptide 1 (Cyp1a1) to 4.7-fold decrease in lipocalin 2 (Lcn2) in c12 cells relative to wild-type (see Supplementary Data).
Agglomerative hierarchical clustering revealed two main clusters: (1) WT with c12 and (2) c1 with c4 (data not shown). This result correlates with the number of basal gene expression differences observed where the c12 line exhibited the least number of differences, while both c1 and c4 lines each exhibited approximately twice the number of differences of the c12 line (Fig. 4).
Functional roles for constitutively expressed genes identified through Gene Ontology annotations (www.geneontology.org) and complemented with literature reports suggest that Hepa-1c1c7 mutant cell lines likely carry multiple mutations affecting a number of divergent pathways. Differentially regulated genes were organized into biological processes including DNA replication, adhesion, and carbohydrate metabolism (data not shown). Genes involved in cytoskeletal organization, bioenergetics, and cell cycle were associated with the morphological, mitochondrial activity, and proliferative rate differences observed between the wild-type and mutant lines (Tables 3–5).
Fourteen genes, representing different pathways, were selected for verification by SYBR Green-based QRT-PCR. Analysis of cDNA microarray and QRTPCR data show good correlation between techniques in eight genes (p 0.5) (Fig. 5A), moderate correlation in three genes (0.5 > p > 0.1), and poor correlation in the remaining three genes (p > 0.1). Figures 5B and 5C illustrate one example each of the latter two categories. In general, the magnitudes of gene expression differences detected through the validation were demonstrated to be within 1-fold of the microarray data. However, larger fold-differences were elucidated using QRT-PCR, as microarray data has been reported to compress gene expression changes (Yuen et al., 2002).
DISCUSSION
Hepa-1c1c7 mutant lines were selected through a chemical resistance screen, and therefore it is not surprising that significant differences in basal gene expression affecting multiple pathways were detected beyond the characterized mutations in Cyp1a1, AhR, and ARNT (Hankinson et al., 1985; Hoffman et al., 1991; Legraverend et al., 1982; Miller et al., 1983; Reyes et al., 1992; Van Gurp and Hankinson, 1984). Surprisingly, c1 mutants, with their characterized truncated Cyp1a1 protein and minimal AHH activity, exhibited the largest number of gene expression differences relative to wild-type cells. Several studies have reported the identification of photooxidized tryptophan products in culture media are potent AhR ligands and readily metabolized by Cyp1a1 (Kocarek et al., 1993; Segner et al., 2000; Wei et al., 2000). The significant differences in gene expression observed in c1 cells may be due to presence of photooxidized tryptophan products that are not cleared, as a result of the expression of a nonfunctional Cyp1a1, and therefore are available to elicit AhR-mediated gene expression, which is intact in this mutant line. Comparisons of gene expression data collected on TCDD treated Hepa-1c1c7 wild-type cells (data not published) show 14% overlap of genes with the c1 active gene list. C4 and c12 mutants are not similarly affected, since their AhR signal transduction machinery is compromised by mutations affecting ARNT and AhR, respectively. Moreover, various additional endogenous metabolites may be similarly affected and influence basal gene expression of a variety of genes.
To further assess differences in constitutive gene expression between wild-type and mutant cells, functional annotation for differentially expressed genes was extracted from Gene Ontology, LocusLink, and the published literature. Gene expression differences were classified into mechanistic pathways and associated with cellular characteristics such as morphology and proliferation. Overall, the differences in constitutive gene expression were consistent with cellular characteristics that defined a mutant cell line.
Cellular Morphology
Differences in morphology observed between wild-type and mutant lines were reflected in differentially expressed genes involved in cytoskeletal organization. Although c4 and c12 mutants exhibited fewer changes in basal expression of cytoskeletal organization genes, they are morphologically distinct from the wild-type cells with a narrower morphology and longer processes. All differentially regulated cytoskeletal organization genes in c4 cells had higher basal expression compared to wild-type, including vimentin (Vim), which is involved in lymphocyte dendrite extension (Sumoza-Toledo and Santos-Argumedo, 2004). Intermediate filament and microtubule genes, represented by integrin beta 4 binding protein (Itgb4bp), microtubule-actin crosslinking factor 1 (Macf1) and alpha 4 tubulin (Tuba4), may also be involved, as orientation of these components facilitates process extension (Sumoza-Toledo and Santos-Argumedo, 2004) and suggests a coordinated effort of these genes to contribute to c4 elongation.
In c12 cells, increased basal expression of select genes may also lead to increased actin filament formation. Higher basal expression of phosphatidylinositol 4-kinase type 2 alpha (Pi4k2a) generates the secondary messenger phosphoinositol 4-phosphate, PI4P, which is essential for normal actin organization through Cdc42 GTPases (Wild et al., 2004) and is representative of phosphoinositide regulated pathways (reviewed in Lemmon, 2003). However, this activity may be counteracted by the lower basal expression of erythrocyte protein band 4.1-like 5 (Epb4.1l5). Epb4.1l5 belongs to a family of proteins which interact with the actin filament (Takeuchi et al., 1994) through a phosphoinositide binding regulatory domain for phosphoinositol 4,5-bisphosphate (PI4,5-P2), a product of PI4P phosphorylation (reviewed in Lemmon, 2003). Also increased basally in c12 cells is vinculin (Vcl), a focal adhesion protein which has been shown to be up-regulated upon actin disruption (Quinlan, 2004). Although there are fewer cytoskeletal gene expression changes in c4 and c12 cells, overall they are consistent with the apparent stretched morphology compared to the wild-type population.
Despite c1 cells exhibiting the most gene expression changes associated with cell morphometry, this was not reflected in a significant difference in its size, shape, or number of processes compared to wild type. These changes in basal expression may be offset by other genes resulting, in an overall absence of net change to cellular morphology. For example, actin is a key player in determining cellular morphology and undergoes constant polymerization and redistribution. Microarray analysis shows lower basal expression of annexins (Anxa3 and Anxa4) in c1 mutants compared to wild-type. Annexins are regulators of actin dynamics (Hayes et al., 2004), and their decreased basal expression indicates depolymerization of actin filaments. Dystroglycan 1 (Dag1), a scaffold for MEK/ERK signaling to maintain actin levels (Natalicchio et al., 2004; Spence et al., 2004) and Septin 2 (Sept2), a cytoskeletal component whose function is associated with actin-based structures during cytokinesis (Kinoshita et al., 1997), are also found to have lower basal expression in c1 cells. The activity of these genes may be compensated by the greater basal expression of annexin 7 (Anxa7) (Hayes et al., 2004) and cortactin (Cttn), which are involved in actin-mediated motility (Patel et al., 1998), and Rac GTPase-activating protein 1 (Racgap1), responsible for actin-based spreading (Wells et al., 2004), in an effort to maintain actin levels. As a result, the cumulative balance may not be shifted and, thus, may have little influence on overall morphology. This is consistent with the observation that the c1 cells are morphologically similar to wild-type cells. However, regulation of actin is also largely dependent on ATP hydrolysis and PIP2 signaling (reviewed in Gungabissoon and Bamburg, 2003) that cannot be detected by microarrays.
Mitochondrial Activity
The MTT assay evaluates mitochondrial activity through dehydrogenase conversion of soluble thaizylol blue to an insoluble dye for colorimetric analysis. Both c4 and c12 cells exhibited higher mitochondrial activity when compared to the wild type. Examination of the differentially regulated genes indicates that c4 cells exhibit the largest number of gene expression changes associated with mitochondrial activity. ATP synthase subunits (Atp5g1 and Atp5o) (Chen et al., 1995; Li et al., 2001), cytochrome c oxidase subunit (Cox6b) (Ohtsu et al., 2001), and NADH dehydrogenase Fe-S subunits (Ndufs2 and Ndufs4) (Loeffen et al., 1998; Papa et al., 2002) are all components of the mitochondrial electron transport chain and show higher basal expression in the c4 cells, consistent with increased basal mitochondrial activity. However, only Cox8a, a subunit of cytochrome c oxidase (Huttemann et al., 2003), exhibited a significant increase in basal expression that could be associated with the increased mitochondrial activity of the c12 cell line. This indicates the augmented activity may involve other genes not represented on the array or by other nontranscriptionally mediated factors, such as changes in enzymatic activity due to posttranslational signaling events. In addition, it could also reflect the incomplete status of the available gene annotation.
In the c1 mutants, Cox6b exhibited lower basal expression, in agreement with the lower MTT activity. However, the higher basal expression of NADH dehydrogenase subunits (Ndufa1 and Ndufv1) (Au et al., 1999; Schuelke et al., 2002) suggests higher mitochondrial activity should have been observed.
Expression differences in mitochondrial transcripts may address some of the differences in mitochondrial activity detected by the MTT assay, but the activity itself also correlated with the observed proliferative rates. Thus, changes in mitochondrial gene expression in mutant cells may be attributed to different energy needs related to their proliferative rate. Additional mechanisms may also be involved with the differences in basal mitochondrial activity, as transcript levels do not reflect changes in protein levels or enzymatic activity.
Proliferation
Despite exhibiting a slower proliferative rate, numerous proliferative genes exhibited higher basal expression in c1 cells, including insulin-like growth factor 1 (Igf1) and thymopoietin (Tmpo). Igf1 is a well-known mediator of cell proliferation through receptor signaling (reviewed in Furstenberger and Senn, 2002), while Tmpo may have a role in cell cycle control, as its transcripts are more abundant in cancerous cell lines compared to tissues in a slower proliferative state (Weber et al., 1999). Ornithine decarboxylase (Odc), involved in the synthesis of polyamines which support cell proliferation (reviewed in Luk and Casero, 1987), exhibited increased basal expression, while the lower transcript levels of glypican 1 (Gpc1), a membrane-anchored polyamine receptor, restrict polyamine uptake, consistent with the ligand-mediated feedback mechanism associated with this receptor (Belting et al., 2003). Collectively, this suggests that the c1 mutants would have a higher proliferative rate in comparison to the wild-type line. Furthermore, lower basal expression of cyclin-dependent kinase inhibitor 2d (Cdkn2i), an inhibitor of cell cycle kinases (Hirai et al., 1995), should also facilitate proliferation. However, the decrease in basal transcripts of syndecan 1 (Sdc1), a growth hormone receptor involved with adhesion, is consistent with the decrease proliferation observed with the c1 mutants (Maeda et al., 2004) in addition to its role in tumor invasion (Langford et al., 2005). The lower proliferative rate of c1 cells suggests the lower basal expression of Sdc1 may dominate, while changes in gene expression supportive of proliferation serve a compensatory role in an attempt to reestablish wild-type proliferation rates. These inconsistencies between changes in gene expression and functional assays illustrate the importance of phenotypic anchoring in the interpretation of gene expression data.
Numerous basal expression differences in c4 cells were found to be associated with proliferation. Although c4 cell proliferation was not found to be significantly different from WT, the data trend indicates that significance may be detected beyond 48 h. Studies beyond 48 h were not conducted, as cell cultures approached confluency which may confound the interpretation due to different proliferation kinetics. Higher transcript levels of cyclin D2 (Ccnd2) and cyclin-dependent kinase 4 (Cdk4) are required for the G1 S transition (Ando et al., 1993; Ando and Griffin, 1995), suggesting that the cells are positively regulated for growth. RanGTPase hydrolytic activity is essential for DNA synthesis, and its activity is mediated by RAN GTPase activating protein 1 (Rangap1), which is subsequently promoted through RAN-binding protein 1 (Ranbp1) interaction (Guarguaglini et al., 2000). Higher basal expression of Rangap1 and Ranbp1 may be factors regulating the elevated proliferation observed in the c4 mutants.
Other increases in basal transcript levels appear to limit proliferation, such as ornithine decarboxylase antizyme (Oaz1), which represses polyamine synthesis by promoting Odc degradation (Schipper et al., 2004), growth arrest specific 1 (Gas1), an inducible proliferation inhibitor (Mellstrom et al., 2002), and ubiquitin-conjugating enzyme E2C (Ube2c), which degrades cyclin B1 in plants (Criqui et al., 2002). Although these anti-proliferative signals are present, the overall pro-proliferative effects prevail.
General Conclusion
Our study demonstrates that numerous genes in the B[a]P-resistant mutant cell lines derived from the Hepa-1c1c7 line are differentially regulated at the basal level in comparison to the wild-type parent. We have observed changes in gene expression that may contribute to differences in morphology, mitochondrial activity, and cell proliferation rates, suggesting that B[a]P selection introduced mutations in addition to the well-characterized affects on Cyp1a1, AhR, and ARNT. However, the results are limited by the lack of an array with cDNA representative of all mouse genes, the lack of confirmation of increases or decreases in basal protein expression and/or enzymatic activity, and the realization that the available functional annotation is limited and likely biased. Nevertheless, these results indicate that conclusions regarding reported differences between wild-type and mutant cell lines may be partially attributed to factors other than mutations to Cyp1a1, AhR, and ARNT alone.
SUPPLEMENTARY DATA
Supplementary data on gene expression differences are available at Toxicological Sciences online as well as at http://dbzach.fst.msu.edu.
ACKNOWLEDGMENTS
Thanks to Matthew Ramer for assistance in cell maintenance and completing microarray experiments as well as to Dr. Jeremy Burt, Darrell Boverhof, and Edward Dere for aiding in the editing and proofing of the manuscript. Funding for this study was provided by NIEHS grant ES 011271 and ES 12245 to TRZ as well as fellowships from the MSU Biochemistry and Molecular Biology Department to CJF. TRZ is partially supported by the Michigan Agricultural Experiment Station.
Conflict of interest: none declared.
REFERENCES
Ando, K., Ajchenbaum-Cymbalista, F., and Griffin, J. D. (1993). Regulation of G1/S transition by cyclins D2 and D3 in hematopoietic cells. Proc. Natl. Acad. Sci. U.S.A. 90, 9571–9575.
Ando, K., and Griffin, J. D. (1995). Cdk4 integrates growth stimulatory and inhibitory signals during G1 phase of hematopoietic cells. Oncogene 10, 751–755.
Au, H. C., Seo, B. B., Matsuno-Yagi, A., Yagi, T., and Scheffler, I. E. (1999). The NDUFA1 gene product (MWFE protein) is essential for activity of complex I in mammalian mitochondria. Proc. Natl. Acad. Sci. U.S.A. 96, 4354–4359.
Bartholomew, J. C., Salmon, A. G., Gamper, H. B., and Calvin, M. (1975). Benzo(alpha)pyrene effects on mouse epithelial cells in culture. Cancer Res. 35, 851–856.
Baumann, H., and Jahreis, G. P. (1983). Regulation of mouse haptoglobin synthesis. J. Cell Biol. 97, 728–736.
Belting, M., Mani, K., Jonsson, M., Cheng, F., Sandgren, S., Jonsson, S., Ding, K., Delcros, J. G., and Fransson, L. A. (2003). Glypican-1 is a vehicle for polyamine uptake in mammalian cells. A pivotal role for nitrosothiol-derived nitric oxide. J. Biol. Chem. 278, 47181–47189.
Bernhard, H. P., Darlington, G. J., and Ruddle, F. H. (1973). Expression of liver phenotypes in cultured mouse hepatoma cells: Synthesis and secretion of serum albumin. Dev. Biol. 35, 83–96.
Boverhof, D. R., Burgoon, L. D., Tashiro, C., Chittim, B., Harkema, J. R., Jump, J. B., and Zacharewski, T. R. (2005). Temporal and dose-dependent hepatic gene expression patterns in mice provide new insights into TCDD-mediated hepatotoxicity. Toxicol Sci. 85, 1048–1063.
Boverhof, D. R., Fertuck, K. C., Burgoon, L. D., Eckel, J. E., Gennings, C., and Zacharewski, T. R. (2004). Temporal and dose-dependent hepatic gene expression changes in immature ovariectomized mice following exposure to ethynyl estradiol. Carcinogenesis. 25, 1277–1291.
Brazma, A., Hingamp, P., Quackenbush, J., Sherlock, G., Spellman, P., Stoeckert, C., Aach, J., Ansorge, W., Ball, C. A., Causton, H. C., et al. (2001). Minimum information about a microarray experiment (MIAME)-toward standards for microarray data. Nat. Genet. 29, 365–371.
Burgoon, L. D., Boutros, P. C., Dere, E., Pai, S. S., Sun, Y., Aiyar, R., Vakharia, J., and Zacharewski, T. R. (2003). dbZach: Relational toxicogenomics database system. In 42nd Annual Meeting of the Society of Toxicology. Society of Toxicology, Salt Lake City, UT.
Burgoon, L. D., Eckel-Passow, J. E., Gennings, C., Boverhof, D. R., Burt, J. W., and Zacharewski, T. R. (2005). Protocols for the Assurance of Microarray Data Quality and Process Control. (submitted).
Chen, H., Morris, M. A., Rossier, C., Blouin, J. L., and Antonarakis, S. E. (1995). Cloning of the cDNA for the human ATP synthase OSCP subunit (ATP5O) by exon trapping and mapping to chromosome 21q22.1-q22.2. Genomics 28, 470–476.
Criqui, M. C., de Almeida Engler, J., Camasses, A., Capron, A., Parmentier, Y., Inze, D., and Genschik, P. (2002). Molecular characterization of plant ubiquitin-conjugating enzymes belonging to the UbcP4/E2-C/UBCx/UbcH10 gene family. Plant Physiol. 130, 1230–1240.
Darlington, G. J., Bernhard, H. P., Miller, R. A., and Ruddle, F. H. (1980). Expression of liver phenotypes in cultured mouse hepatoma cells. J. Natl. Cancer Inst. 64, 809–819.
Furstenberger, G., and Senn, H. J. (2002). Insulin-like growth factors and cancer. Lancet Oncol. 3, 298–302.
Guarguaglini, G., Renzi, L., D'Ottavio, F., Di Fiore, B., Casenghi, M., Cundari, E., and Lavia, P. (2000). Regulated Ran-binding protein 1 activity is required for organization and function of the mitotic spindle in mammalian cells in vivo. Cell Growth Differ. 11, 455–465.
Gungabissoon, R. A., and Bamburg, J. R. (2003). Regulation of growth cone actin dynamics by ADF/cofilin. J. Histochem. Cytochem. 51, 411–420.
Hankinson, O. (1979). Single-step selection of clones of a mouse hepatoma line deficient in aryl hydrocarbon hydroxylase. Proc. Natl. Acad. Sci. U.S.A. 76, 373–376.
Hankinson, O., Andersen, R. D., Birren, B. W., Sander, F., Negishi, M., and Nebert, D. W. (1985). Mutations affecting the regulation of transcription of the cytochrome P1-450 gene in the mouse Hepa-1 cell line. J. Biol. Chem. 260, 1790–1795.
Hayes, M. J., Rescher, U., Gerke, V., and Moss, S. E. (2004). Annexin-actin interactions. Traffic 5, 571–576.
Hirai, H., Roussel, M. F., Kato, J. Y., Ashmun, R. A., and Sherr, C. J. (1995). Novel INK4 proteins, p19 and p18, are specific inhibitors of the cyclin D-dependent kinases CDK4 and CDK6. Mol. Cell. Biol. 15, 2672–2681.
Hoffman, E. C., Reyes, H., Chu, F. F., Sander, F., Conley, L. H., Brooks, B. A., and Hankinson, O. (1991). Cloning of a factor required for activity of the Ah (dioxin) receptor. Science 252, 954–958.
Huttemann, M., Schmidt, T. R., and Grossman, L. I. (2003). A third isoform of cytochrome c oxidase subunit VIII is present in mammals. Gene 312, 95–102.
Jeong, H. G., and Lee, S. S. (1998). Suppressive effects of estradiol on 2,3,7,8-tetrachlorodibenzo-p-dioxin-mediated transcriptional activation of murine Cyp1a-1 in mouse hepatoma Hepa 1c1c7 cells. Cancer Lett. 133, 177–184.
Kimura, S., Smith, H. H., Hankinson, O., and Nebert, D. W. (1987). Analysis of two benzo[a]pyrene-resistant mutants of the mouse hepatoma Hepa-1 P(1)450 gene via cDNA expression in yeast. EMBO J. 6, 1929–1933.
Kinoshita, M., Kumar, S., Mizoguchi, A., Ide, C., Kinoshita, A., Haraguchi, T., Hiraoka, Y., and Noda, M. (1997). Nedd5, a mammalian septin, is a novel cytoskeletal component interacting with actin-based structures. Genes Dev. 11, 1535–1547.
Kocarek, T. A., Schuetz, E. G., and Guzelian, P. S. (1993). Transient induction of cytochrome P450 1A1 mRNA by culture medium component in primary cultures of adult rat hepatocytes. In Vitro Cell Dev. Biol. 29A, 62–66.
Langford, J. K., Yang, Y., Kieber-Emmons, T., and Sanderson, R. D. (2005). Identification of an invasion regulatory domain within the core protein of syndecan-1. J. Biol. Chem. 280, 3467–3473.
Legraverend, C., Hannah, R. R., Eisen, H. J., Owens, I. S., Nebert, D. W., and Hankinson, O. (1982). Regulatory gene product of the Ah locus. Characterization of receptor mutants among mouse hepatoma clones. J. Biol. Chem. 257, 6402–6407.
Lemmon, M. A. (2003). Phosphoinositide recognition domains. Traffic 4, 201–213.
Li, H. S., Zhang, J. Y., Thompson, B. S., Deng, X. Y., Ford, M. E., Wood, P. G., Stolz, D. B., Eagon, P. K., and Whitcomb, D. C. (2001). Rat mitochondrial ATP synthase ATP5G3: Cloning and upregulation in pancreas after chronic ethanol feeding. Physiol. Genomics 6, 91–98.
Liu, R. M., Vasiliou, V., Zhu, H., Duh, J. L., Tabor, M. W., Puga, A., Nebert, D. W., Sainsbury, M., and Shertzer, H. G. (1994). Regulation of [Ah] gene battery enzymes and glutathione levels by 5,10-dihydroindeno[1,2-b]indole in mouse hepatoma cell lines. Carcinogenesis 15, 2347–2352.
Loeffen, J., van den Heuvel, L., Smeets, R., Triepels, R., Sengers, R., Trijbels, F., and Smeitink, J. (1998). cDNA sequence and chromosomal localization of the remaining three human nuclear encoded iron sulphur protein (IP) subunits of complex I: The human IP fraction is completed. Biochem. Biophys. Res. Commun. 247, 751–758.
Luk, G. D., and Casero, R. A., Jr. (1987). Polyamines in normal and cancer cells. Adv. Enzyme Regul. 26, 91–105.
Maeda, T., Alexander, C. M., and Friedl, A. (2004). Induction of syndecan-1 expression in stromal fibroblasts promotes proliferation of human breast cancer cells. Cancer Res. 64, 612–621.
Mattes, W. B., Pettit, S. D., Sansone, S. A., Bushel, P. R., and Waters, M. D. (2004). Database development in toxicogenomics: Issues and efforts. Environ. Health Perspect. 112, 495–505.
Mellstrom, B., Cena, V., Lamas, M., Perales, C., Gonzalez, C., and Naranjo, J. R. (2002). Gas1 is induced during and participates in excitotoxic neuronal death. Mol. Cell. Neurosci. 19, 417–429.
Merchant, M., Morrison, V., Santostefano, M., and Safe, S. (1992). Mechanism of action of aryl hydrocarbon receptor antagonists: Inhibition of 2,3,7,8-tetrachlorodibenzo-p-dioxin-induced CYP1A1 gene expression. Arch. Biochem. Biophys. 298, 389–394.
Miller, A. G., Israel, D., and Whitlock, J. P., Jr. (1983). Biochemical and genetic analysis of variant mouse hepatoma cells defective in the induction of benzo(a)pyrene-metabolizing enzyme activity. J. Biol. Chem. 258, 3523–3527.
Miranda, C. L., Aponso, G. L., Stevens, J. F., Deinzer, M. L., and Buhler, D. R. (2000). Prenylated chalcones and flavanones as inducers of quinone reductase in mouse Hepa 1c1c7 cells. Cancer Lett. 149, 21–29.
Natalicchio, A., Laviola, L., De Tullio, C., Renna, L. A., Montrone, C., Perrini, S., Valenti, G., Procino, G., Svelto, M., and Giorgino, F. (2004). Role of the P66SHC isoform in IGF-I receptor signal through MEK/ERK and regulation of actin cytoskeleton in rat myoblasts. J. Biol. Chem. 279, 43900–43909.
Numayama-Tsuruta, K., Kobayashi, A., Sogawa, K., and Fujii-Kuriyama, Y. (1997). A point mutation responsible for defective function of the aryl-hydrocarbon-receptor nuclear translocator in mutant Hepa-1c1c7 cells. Eur. J. Biochem. 246, 486–495.
Ohtsu, K., Nakazono, M., Tsutsumi, N., and Hirai, A. (2001). Characterization and expression of the genes for cytochrome c oxidase subunit VIb (COX6b) from rice and Arabidopsis thaliana. Gene 264, 233–239.
Papa, S., Sardanelli, A. M., Scacco, S., Petruzzella, V., Technikova-Dobrova, Z., Vergari, R., and Signorile, A. (2002). The NADH: Ubiquinone oxidoreductase (complex I) of the mammalian respiratory chain and the cAMP cascade. J. Bioenerg. Biomembr. 34, 1–10.
Papaconstantinou, J., Hill, R. E., Gibson, W. H., and Rao, E. Y. (1978). Synthesis and secretion of transferrin by cultured mouse hepatoma cells. Differentiation 10, 139–145.
Patel, A. S., Schechter, G. L., Wasilenko, W. J., and Somers, K. D. (1998). Overexpression of EMS1/cortactin in NIH3T3 fibroblasts causes increased cell motility and invasion in vitro. Oncogene 16, 3227–3232.
Pollenz, R. S. (1996). The aryl-hydrocarbon receptor, but not the aryl-hydrocarbon receptor nuclear translocator protein, is rapidly depleted in hepatic and nonhepatic culture cells exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin. Mol. Pharmacol. 49, 391–398.
Probst, M. R., Reisz-Porszasz, S., Agbunag, R. V., Ong, M. S., and Hankinson, O. (1993). Role of the aryl hydrocarbon receptor nuclear translocator protein in aryl hydrocarbon (dioxin) receptor action. Mol. Pharmacol. 44, 511–518.
Quinlan, M. P. (2004). Vinculin, VASP, and profilin are coordinately regulated during actin remodeling in epithelial cells, which requires de novo protein synthesis and protein kinase signal transduction pathways. J. Cell Physiol. 200, 277–290.
Reyes, H., Reisz-Porszasz, S., and Hankinson, O. (1992). Identification of the Ah receptor nuclear translocator protein (Arnt) as a component of the DNA binding form of the Ah receptor. Science 256, 1193–1195.
Rockett, J. C., and Dix, D. J. (1999). Application of DNA arrays to toxicology. Environ. Health Perspect. 107, 681–685.
Schipper, R. G., Cuijpers, V. M., De Groot, L. H., Thio, M., and Verhofstad, A. A. (2004). Intracellular localization of ornithine decarboxylase and its regulatory protein, antizyme-1. J. Histochem. Cytochem. 52, 1259–1266.
Schuelke, M., Detjen, A., van den Heuvel, L., Korenke, C., Janssen, A., Smits, A., Trijbels, F., and Smeitink, J. (2002). New nuclear encoded mitochondrial mutation illustrates pitfalls in prenatal diagnosis by biochemical methods. Clin. Chem. 48, 772–775.
Segner, H., Behrens, A., Joyce, E. M., Schirmer, K., and Bols, N. C. (2000). Transient induction of 7-ethoxyresorufin-O-deethylase (EROD) activity by medium change in the rainbow trout liver cell line, RTL-W1. Mar. Environ. Res. 50, 489–493.
Seidel, S. D., and Denison, M. S. (1999). Differential gene expression in wild-type and arnt-defective mouse hepatoma (Hepa1c1c7) cells. Toxicol. Sci. 52, 217–225.
Seubert, J. M., Darmon, A. J., El-Kadi, A. O., D'Souza, S. J., and Bend, J. R. (2002). Apoptosis in murine hepatoma hepa 1c1c7 wild-type, C12, and C4 cells mediated by bilirubin. Mol. Pharmacol. 62, 257–264.
Spence, H. J., Dhillon, A. S., James, M., and Winder, S. J. (2004). Dystroglycan, a scaffold for the ERK-MAP kinase cascade. EMBO Rep. 5, 484–489.
Sumoza-Toledo, A., and Santos-Argumedo, L. (2004). The spreading of B lymphocytes induced by CD44 cross-linking requires actin, tubulin, and vimentin rearrangements. J. Leukoc. Biol. 75, 233–239.
Takeuchi, K., Kawashima, A., Nagafuchi, A., and Tsukita, S. (1994). Structural diversity of band 4.1 superfamily members. J. Cell Sci. 107(7), 1921–1928.
Thompson, E. B., Benedict, W. F., Owens, I. S., and Nebert, D. W. (1974). Aryl hydrocarcon hydroxylase and tyrosine aminotransferase activities in somatic-cell hybrids derived from hepatoma tissue culture HTC (rat) cells and 3T3 (mouse) benzo(a)pyrene-resistant cells. Biochem. Biophys. Res. Commun. 56, 605–610.
Van Gurp, J. R., and Hankinson, O. (1984). Isolation and characterization of revertants from four different classes of aryl hydrocarbon hydroxylase-deficient hepa-1 mutants. Mol. Cell. Biol. 4, 1597–1604.
Weber, P. J., Eckhard, C. P., Gonser, S., Otto, H., Folkers, G., and Beck-Sickinger, A. G. (1999). On the role of thymopoietins in cell proliferation. Immunochemical evidence for new members of the human thymopoietin family. Biol. Chem. 380, 653–660.
Wei, Y. D., Bergander, L., Rannug, U., and Rannug, A. (2000). Regulation of CYP1A1 transcription via the metabolism of the tryptophan-derived 6-formylindolo[3,2-b]carbazole. Arch. Biochem. Biophys. 383, 99–107.
Wells, C. M., Walmsley, M., Ooi, S., Tybulewicz, V., and Ridley, A. J. (2004). Rac1-deficient macrophages exhibit defects in cell spreading and membrane ruffling but not migration. J. Cell. Sci. 117, 1259–1268.
Wild, A. C., Yu, J. W., Lemmon, M. A., and Blumer, K. J. (2004). The p21-activated protein kinase-related kinase Cla4 is a coincidence detector of signaling by Cdc42 and phosphatidylinositol 4-phosphate. J. Biol. Chem. 279, 17101–17110.
Yuen, T., Wurmbach, E., Pfeffer, R. L., Ebersole, B. J., and Sealfon, S. C. (2002). Accuracy and calibration of commercial oligonucleotide and custom cDNA microarrays. Nucleic Acids Res. 30, e48.
Zacharewski, T. R., Bondy, K. L., McDonell, P., and Wu, Z. F. (1994). Antiestrogenic effect of 2,3,7,8-tetrachlorodibenzo-p-dioxin on 17 beta-estradiol-induced pS2 expression. Cancer Res. 54, 2707–2713.
Zhang, J., Watson, A. J., Probst, M. R., Minehart, E., and Hankinson, O. (1996). Basis for the loss of aryl hydrocarbon receptor gene expression in clones of a mouse hepatoma cell line. Mol. Pharmacol. 50, 1454–1462.(C. J. Fong, L. D. Burgoon)
Department of Pharmacology and Toxicology Michigan State University, East Lansing, Michigan, 48824
ABSTRACT
Hepa-1c1c7 wild-type and benzo[a]pyrene-resistant derived mutant cell lines have been used to elucidate pathways and mechanisms involving the aryl hydrocarbon receptor (AhR). However, there has been little focus on other biological processes which may differ between the isolated lines. In this study, mouse cDNA microarrays representing 4858 genes were used to examine differences in basal gene expression between mouse Hepa-1c1c7 wild-type and c1 (truncated Cyp1a1 protein), c4 (AhR nuclear translocator, ARNT, deficient), and c12 (low AhR levels) mutant cell lines. Surprisingly, c1 mutants exhibited the greatest number of gene expression changes compared to wild-type cells, followed by c4 and c12 lines, respectively. Differences in basal gene expression were consistent with cell line specific variations in morphology, mitochondrial activity, and proliferation rate. MTT and direct cell count assays indicate both c4 and c12 mutants exhibit increased proliferative activity when compared to wild-type cells, while the c1 mutants exhibited decreased activity. This study further characterizes Hepa-1c1c7 wild-type and mutant cells and identifies significant differences in biological processes that should be considered when conducting comparative mechanistic studies with these lines.
Key Words: benzo[a]pyrene; aryl hydrocarbon receptor; Hepa-1c1c7.
INTRODUCTION
The mouse Hepa-1c1c7 cell line was derived from a BW 7756 hepatoma which arose in a C57L mouse and propagated in C57L/J mice (Bernhard et al., 1973). Although parent Hepa-1 cells lack some liver-specific activities (Darlington et al., 1980), the isolated Hepa-1c1c7 line exhibits hepatic transferrin secretion (Papaconstantinou et al., 1978) and haptoglobin synthesis (Baumann and Jahreis, 1983). However, the vast majority of studies use Hepa-1 and Hepa-1c1c7 cells to examine their characteristic highly inducible aryl hydrocarbon hydroxylase activity (AHH) (Hankinson, 1979).
Hepa-1c1c7 mutants were selected and isolated based on their resistance to benzo[a]pyrene (B[a]P) toxicity and compromised AHH induction (Bartholomew et al., 1975; Hankinson, 1979; Thompson et al., 1974). Three isolated complementation groups (A, B, and C) were identified and used to investigate the mechanism of action of the aryl hydrocarbon receptor (AhR) (Van Gurp and Hankinson, 1984). C1 mutants (Group A) are characterized by a deficiency in Cyp1a1 activity as a result of a premature stop codon located after Asn 414, resulting in a loss of a C-terminal heme-binding site necessary for enzyme activity (Kimura et al., 1987). C12 mutants (Group B) express only 3% of the AHH activity of wild-type cells (Van Gurp and Hankinson, 1984) due to compromised AhR expression, possibly from a defective factor involved in chromatin remodeling (Hankinson et al., 1985; Zhang et al., 1996). Hepa-1c1c7 c4 (Group C) mutant cells are devoid of any AHH activity, due to a G326D point mutation in the ARNT (aryl hydrocarbon receptor nuclear translocator) coding region, which causes a marked reduction in DRE sequence binding by the AhR/ARNT complex (Numayama-Tsuruta et al., 1997).
Studies using Hepa-1c1c7 wild-type and mutant cells have significantly contributed to the elucidation of the mechanism of AhR-mediated gene expression (Boverhof et al., 2005; Hoffman et al., 1991; Probst et al., 1993; Reyes et al., 1992), including the identification of responsive detoxification genes (Liu et al., 1994). However, isogenicity of loci beyond Cyp1a1 is not expected, given the selection method and criteria used for mutant cell line isolation. Despite their use in numerous studies (Jeong and Lee, 1998; Merchant et al., 1992; Miranda et al., 2000; Pollenz, 1996; Seubert et al., 2002; Zacharewski et al., 1994), only one has examined differences in constitutive gene expression between Hepa-1c1c7 wild-type and mutant cells using differential display PCR in wild-type and the related but independently derived ARNT-deficient Hepa-1 cell line derivative, BPRc1 (Seidel and Denison, 1999). Our study uses cDNA microarrays to further examine differences in basal gene expression in mutant and wild-type cells with regards to variant morphology and growth rates. Differences in basal gene expression were also linked to cellular physiological observations to identify mechanistic networks that may be affected in the mutant hepatoma cell lines. These data indicate that differences between wild-type and mutant cells may not be attributed solely to differences in AhR/ARNT functionality and Cyp1a1 inducibility.
MATERIALS AND METHODS
Cell maintenance and sample collection.
Hepa-1c1c7 wild-type and mutant cells (c1, c4, and c12) (gifts from O. Hankinson, University of California, Los Angeles) were maintained in phenol-red-free DMEM/F12 media (Invitrogen, Carlsbad, CA) supplemented with 5% fetal bovine serum (FBS) (Serologicals Corporation, Norcross, GA), 50 μg/ml gentamycin, 2.5 μg/ml amphotericin B, 100 U/ml penicillin, and 100 μg/ml streptomycin (Invitrogen, Carlsbad, CA). 1 x 106 cells were seeded into T75 culture flasks (Sarstedt, Newton, NC) and harvested at 90% cell confluency by scraping in Trizol (Invitrogen, Carlsbad, CA). RNA was isolated as per manufacturer's instructions and stored in RNA storage solution (Ambion Inc., Austin, TX) at –80°C. RNA quality was examined on a denaturing 1% agarose gel as well as by A260/280 ratio.
Images were captured with Nikon Eclipse TE300 microscope and GelExpert software (San Carlos, CA). Measurements were conducted using ImageJ v132j (NIH, USA) and significance determined using SAS version 8.0. One-way ANOVA analysis followed by Dunnett's t-tests were performed on measurements where = 0.05.
Cell growth rate assessments.
For MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] assays, 10,000 cells were seeded in a 96-well tissue-culture plate (Corning, MI) in 5% FBS and antibiotic supplemented media. Thaizylol blue (Sigma, St. Louis, MO) was added to cells, incubated for 3 h at 37.5°C and 5% CO2, and absorbance at 595 nm was determined using Softmax software (Molecular Devices, Sunnyvale, CA) at 3, 6, 12, 24, 36, and 48 h on an Emax 96-well microplate reader (Molecular Devices, Sunnyvale, CA). Experiments were completed in quadruplicate. Relative viability was standardized to the blank wells containing media only.
For direct cell counts, 3 x 105 cells were seeded in T25 culture flasks (Corning, MI) in 5% FBS and antibiotic supplemented media. Cells were incubated at 37.5°C and 5% CO2 for 6, 12, 24, and 48 h. Cells were then trypsinized and subjected to manual counting, in duplicate, using a hemocytometer (Hausser Scientific Co., Horsham, PA). Experiments were completed in quadruplicate.
Significance was determined using SAS v.8.0. Two-way ANOVA analysis followed by Tukey's HSD post hoc test was performed on MTT assay data where = 0.05. For the direct cell counts, a repeated measures ANOVA was performed followed by a Tukey's HSD post hoc test where = 0.05. Comparisons were made between each mutant and wild-type cell line at each time point.
Microarray processing.
cDNA arrays containing 6376 features (representing 4858 unique genes based on mouse Unigene build 138), were spotted in duplicate, on epoxy-coated glass slides (SCHOTT Nexterion, Germany) using an Omnigrid arrayer (GeneMachines, San Carlos, CA) equipped with 16 (4 x 4) Chipmaker 2 pins (Telechem International Inc., Sunnyvale, CA) at the Genomics Technology Support Facility (http://genomics.msu.edu). Three biological replicates for each cell line were examined using a 6-slide independent reference design that incorporated dye-swap hybridizations (Fig. 1) for a total of 18 microarray hybridizations analyzed. Clones for the array were initially selected for the examination of testis-specific genes, and the library has since been expanded to include genes known to be responsive to estrogens, androgens, and dioxin. Selected clones were obtained from EPAMAC sequence verified mouse clones (Rockett and Dix, 1999), Research Genetics IMAGE clones, and the National Institute of Aging 15K mouse clone set.
Detailed protocols for processing of microarrays are available at http://dbzach.fst.msu.edu:8050/dbZachProtocolsInterfaceVersions/ProtocolsInterfaceFront. Briefly, 15 μg of total RNA were reverse transcribed with Superscript II (Invitrogen, Carlsbad, CA) to incorporate Cy3- or Cy5-conjugated dUTP (Amersham Biosciences, Piscataway, NJ). Labeled samples were mixed, purified on a Qiagen PCR purification column (QIAGEN, Valencia, CA), dried down using a vacuum centrifuge, and resuspended in 33 μl of hybridization buffer (56% formamide, 32% 20x SSPE, 8% 50x Denhardt's Solution, 4% 20% SDS, 20 μg poly(A), 20 μg mouse CoT-1 DNA, 10 μg yeast tRNA carrier) for overnight 42°C hybridization on arrays. Slides were washed sequentially in solutions containing SSC and decreasing concentrations of SDS and scanned using a 428 Affymetrix Scanner (Santa Clara, CA). Images were quantified using Analyzer DigitalGENOME (MolecularWare, Cambridge, MA) software. All data was stored in dbZach (http://dbzach.fst.msu.edu), a Minimum Information About Microarray Experiments (MIAME)-compliant relational database (Brazma et al., 2001) running under Windows 2003/Oracle 9i that currently supports microarray data storage, retrieval, and querying, as well as facilitates data analysis, sharing, and reporting (Burgoon et al., 2003; Mattes et al., 2004).
Quantitative RT-PCR.
A sample of total RNA isolated from each replicate was set aside for verification using SYBR Green-based quantitative real-time PCR (QRT-PCR). Briefly, 1 μg of RNA was primed by random hexamers (Roche, Switzerland) and reverse transcribed using Superscript II in a 20-μl reaction as described. From a 5-fold dilution of this cDNA solution, 3 μl was used in a 30-μl PCR reaction containing 1x SYBR Green PCR buffer, 3 mM MgCl2, 0.33 mM dNTPs, 0.5 IU AmpliTaq Gold (Applied Biosystems, Foster City, CA) and 0.15 mM forward and reverse primer. All primers were designed by submitting RefSeq sequences to Primer3 (http://www-genome.wi.mit.edu/cgi-bin/primer/primer3_www.cgi) to identify an approximately 125bp amplicon (Table 1). PCR amplification was conducted in 96-well MicroAmp Optical plates (Applied Biosystems) on an Applied Biosystems PRISM 7000 Sequence Detection System under the following conditions: 10 min denaturation and enzyme activation at 95°C, followed by 40 cycles of 95°C for 15 s and 60°C for 1 min. A 30-min dissociation protocol, after amplification, was conducted to assess primer specificity and product uniformity. Each plate contained duplicate standards of purified PCR product of known template concentration over the range of eight orders of magnitude to generate a log template concentration standard curve. No template controls (NTC) were included on each plate where samples with a Ct value within 2 SD of the mean Ct values for the NTCs were considered below the limits of detection. Plots were visualized and thresholds determined using ABI Prism 7000 SDS Software (Applied Biosystems, Foster City, CA). Results were normalized to beta-actin to control for differences in RNA loading, quality and cDNA synthesis. Statistical significance of expression differences between wild-type line and mutant lines were assessed using a standard one-way ANOVA followed by Tukey's HSD post hoc analysis using SAS version 8.0 (Cary, NC).
Microarray analysis and clustering.
A general linear mixed model was used to normalize microarray data (Boverhof et al., 2004) and generate t-value statistic prioritized lists for each mutant line. Lists were used to identify active features different from wild-type basal expression by filtering on a t-statistic cut-off (t > |2.77|). Hierarchical clustering of 723 genes was completed using the heatmap() function in R, version 1.9.0, and R, version 1.9.1, was used to compute the Pearson's correlation coefficient between DNA microarray data and QRT-PCR results.
RESULTS
Cellular Physiological Differences between Hepa-1c1c7 Wild-Type and Mutant Lines
Morphological comparisons were made between Hepa-1c1c7 wild type (WT) and mutant cell lines at comparable times and cell densities (Fig. 2). Wild-type and c1 mutant cells have a more compact cell body, whereas c4 and c12 mutant cell lines are narrower (p < 0.05) and thus appear more elongated (Table 2). The c4 and c12 mutant cells also possess significantly fewer number of processes (p < 0.05) extending from the main body; subsequently the wild-type and c1 cells appear more stellate in nature. In addition, c4 mutants exhibit longer process lengths (p < 0.05) compared to wild-type cells.
MTT colorimetric assays indicate significant (p < 0.05) differences in mitochondrial activity between wild-type and mutant cells at each time point (Fig. 3A). By 24 h, increased MTT activity was observed in c12 cell lines, while c1 cells exhibited decreased activity. At 36 h, c4 cells exhibit a statistically significant increase in mitochondrial activity. These trends were consistent throughout the time course.
Direct cell counts were conducted to assess proliferative rates (Fig. 3B). Although a significant decrease in proliferation rate was only observed in c1 cells at 48 h compared to wild type, generally, there was good trend agreement between MTT and direct cell count data. Statistical significance was not observed for c4 mutant cells; however, the trend observed indicates that a significant difference may be observed beyond 48 h. A more extensive time course was not conducted because the cells become confluent, thereby confounding proliferative kinetics.
Microarray Quality Assessment and Statistical Analysis
All arrays within this study were compared to a historical data set of established high-quality arrays. Background signal intensity, feature signal intensity, feature versus background signal intensity ratios, number of features with background intensities greater than the feature intensity for each array, and relationships between feature and background signal intensities were examined. Comparisons indicate that both background and signal intensities were higher than the historical data set; however, all arrays that were used for further analysis met the quality standard requirements (Burgoon et al., submitted).
Basal Gene Expression Differences between Hepa-1c1c7 and Wild-Type and Mutant Cell Lines
Statistical analysis using a model-based t-statistic test (t-value > |2.97|) identified a total of 946 active features (spots on an array) across c1, c4, and c12 mutant cell lines when compared to wild-type cells, 723 of which are unique and represent 499 unique LocusLink annotated genes. The purpose of the t-statistic is to rank and prioritize clones to be initially examined and a cut-off selected to generate a manageable clone list for additional functional investigation.
C1 mutants exhibited the largest number of differentially expressed basal level of genes (312) when compared to wild-type cells, followed by c4 (220) and c12 (134) cells, respectively (Fig. 4A). A total of 30 genes exhibited differential basal levels of expression in all mutant lines versus the wild type, while 63, 35, and 9 genes demonstrated differential basal levels in sets of two cell lines, c1 and c4 lines, c1 and c12 lines, and c4 and c12 lines, respectively, relative to wild-type (Fig. 4B).
Higher basal levels of gene expression represented approximately 60%, 70%, and 49% of the c1, c4, and c12 active gene lists, respectively (Fig. 4C). Gene expression differences ranged from the 3.3-fold increase in RIKEN cDNA 1700027N10 gene transcripts to 16.1-fold down regulation of albumin 1 (Alb1) in c1 cells relative to wild-type, 2.67-fold increase in B-cell translocation gene 1, anti-proliferative (Btg1) to 2.4-fold decrease in Alb1 in c4 cells relative to wild-type, and 2.3-fold increase in cytochrome P450, family 1, subfamily a, polypeptide 1 (Cyp1a1) to 4.7-fold decrease in lipocalin 2 (Lcn2) in c12 cells relative to wild-type (see Supplementary Data).
Agglomerative hierarchical clustering revealed two main clusters: (1) WT with c12 and (2) c1 with c4 (data not shown). This result correlates with the number of basal gene expression differences observed where the c12 line exhibited the least number of differences, while both c1 and c4 lines each exhibited approximately twice the number of differences of the c12 line (Fig. 4).
Functional roles for constitutively expressed genes identified through Gene Ontology annotations (www.geneontology.org) and complemented with literature reports suggest that Hepa-1c1c7 mutant cell lines likely carry multiple mutations affecting a number of divergent pathways. Differentially regulated genes were organized into biological processes including DNA replication, adhesion, and carbohydrate metabolism (data not shown). Genes involved in cytoskeletal organization, bioenergetics, and cell cycle were associated with the morphological, mitochondrial activity, and proliferative rate differences observed between the wild-type and mutant lines (Tables 3–5).
Fourteen genes, representing different pathways, were selected for verification by SYBR Green-based QRT-PCR. Analysis of cDNA microarray and QRTPCR data show good correlation between techniques in eight genes (p 0.5) (Fig. 5A), moderate correlation in three genes (0.5 > p > 0.1), and poor correlation in the remaining three genes (p > 0.1). Figures 5B and 5C illustrate one example each of the latter two categories. In general, the magnitudes of gene expression differences detected through the validation were demonstrated to be within 1-fold of the microarray data. However, larger fold-differences were elucidated using QRT-PCR, as microarray data has been reported to compress gene expression changes (Yuen et al., 2002).
DISCUSSION
Hepa-1c1c7 mutant lines were selected through a chemical resistance screen, and therefore it is not surprising that significant differences in basal gene expression affecting multiple pathways were detected beyond the characterized mutations in Cyp1a1, AhR, and ARNT (Hankinson et al., 1985; Hoffman et al., 1991; Legraverend et al., 1982; Miller et al., 1983; Reyes et al., 1992; Van Gurp and Hankinson, 1984). Surprisingly, c1 mutants, with their characterized truncated Cyp1a1 protein and minimal AHH activity, exhibited the largest number of gene expression differences relative to wild-type cells. Several studies have reported the identification of photooxidized tryptophan products in culture media are potent AhR ligands and readily metabolized by Cyp1a1 (Kocarek et al., 1993; Segner et al., 2000; Wei et al., 2000). The significant differences in gene expression observed in c1 cells may be due to presence of photooxidized tryptophan products that are not cleared, as a result of the expression of a nonfunctional Cyp1a1, and therefore are available to elicit AhR-mediated gene expression, which is intact in this mutant line. Comparisons of gene expression data collected on TCDD treated Hepa-1c1c7 wild-type cells (data not published) show 14% overlap of genes with the c1 active gene list. C4 and c12 mutants are not similarly affected, since their AhR signal transduction machinery is compromised by mutations affecting ARNT and AhR, respectively. Moreover, various additional endogenous metabolites may be similarly affected and influence basal gene expression of a variety of genes.
To further assess differences in constitutive gene expression between wild-type and mutant cells, functional annotation for differentially expressed genes was extracted from Gene Ontology, LocusLink, and the published literature. Gene expression differences were classified into mechanistic pathways and associated with cellular characteristics such as morphology and proliferation. Overall, the differences in constitutive gene expression were consistent with cellular characteristics that defined a mutant cell line.
Cellular Morphology
Differences in morphology observed between wild-type and mutant lines were reflected in differentially expressed genes involved in cytoskeletal organization. Although c4 and c12 mutants exhibited fewer changes in basal expression of cytoskeletal organization genes, they are morphologically distinct from the wild-type cells with a narrower morphology and longer processes. All differentially regulated cytoskeletal organization genes in c4 cells had higher basal expression compared to wild-type, including vimentin (Vim), which is involved in lymphocyte dendrite extension (Sumoza-Toledo and Santos-Argumedo, 2004). Intermediate filament and microtubule genes, represented by integrin beta 4 binding protein (Itgb4bp), microtubule-actin crosslinking factor 1 (Macf1) and alpha 4 tubulin (Tuba4), may also be involved, as orientation of these components facilitates process extension (Sumoza-Toledo and Santos-Argumedo, 2004) and suggests a coordinated effort of these genes to contribute to c4 elongation.
In c12 cells, increased basal expression of select genes may also lead to increased actin filament formation. Higher basal expression of phosphatidylinositol 4-kinase type 2 alpha (Pi4k2a) generates the secondary messenger phosphoinositol 4-phosphate, PI4P, which is essential for normal actin organization through Cdc42 GTPases (Wild et al., 2004) and is representative of phosphoinositide regulated pathways (reviewed in Lemmon, 2003). However, this activity may be counteracted by the lower basal expression of erythrocyte protein band 4.1-like 5 (Epb4.1l5). Epb4.1l5 belongs to a family of proteins which interact with the actin filament (Takeuchi et al., 1994) through a phosphoinositide binding regulatory domain for phosphoinositol 4,5-bisphosphate (PI4,5-P2), a product of PI4P phosphorylation (reviewed in Lemmon, 2003). Also increased basally in c12 cells is vinculin (Vcl), a focal adhesion protein which has been shown to be up-regulated upon actin disruption (Quinlan, 2004). Although there are fewer cytoskeletal gene expression changes in c4 and c12 cells, overall they are consistent with the apparent stretched morphology compared to the wild-type population.
Despite c1 cells exhibiting the most gene expression changes associated with cell morphometry, this was not reflected in a significant difference in its size, shape, or number of processes compared to wild type. These changes in basal expression may be offset by other genes resulting, in an overall absence of net change to cellular morphology. For example, actin is a key player in determining cellular morphology and undergoes constant polymerization and redistribution. Microarray analysis shows lower basal expression of annexins (Anxa3 and Anxa4) in c1 mutants compared to wild-type. Annexins are regulators of actin dynamics (Hayes et al., 2004), and their decreased basal expression indicates depolymerization of actin filaments. Dystroglycan 1 (Dag1), a scaffold for MEK/ERK signaling to maintain actin levels (Natalicchio et al., 2004; Spence et al., 2004) and Septin 2 (Sept2), a cytoskeletal component whose function is associated with actin-based structures during cytokinesis (Kinoshita et al., 1997), are also found to have lower basal expression in c1 cells. The activity of these genes may be compensated by the greater basal expression of annexin 7 (Anxa7) (Hayes et al., 2004) and cortactin (Cttn), which are involved in actin-mediated motility (Patel et al., 1998), and Rac GTPase-activating protein 1 (Racgap1), responsible for actin-based spreading (Wells et al., 2004), in an effort to maintain actin levels. As a result, the cumulative balance may not be shifted and, thus, may have little influence on overall morphology. This is consistent with the observation that the c1 cells are morphologically similar to wild-type cells. However, regulation of actin is also largely dependent on ATP hydrolysis and PIP2 signaling (reviewed in Gungabissoon and Bamburg, 2003) that cannot be detected by microarrays.
Mitochondrial Activity
The MTT assay evaluates mitochondrial activity through dehydrogenase conversion of soluble thaizylol blue to an insoluble dye for colorimetric analysis. Both c4 and c12 cells exhibited higher mitochondrial activity when compared to the wild type. Examination of the differentially regulated genes indicates that c4 cells exhibit the largest number of gene expression changes associated with mitochondrial activity. ATP synthase subunits (Atp5g1 and Atp5o) (Chen et al., 1995; Li et al., 2001), cytochrome c oxidase subunit (Cox6b) (Ohtsu et al., 2001), and NADH dehydrogenase Fe-S subunits (Ndufs2 and Ndufs4) (Loeffen et al., 1998; Papa et al., 2002) are all components of the mitochondrial electron transport chain and show higher basal expression in the c4 cells, consistent with increased basal mitochondrial activity. However, only Cox8a, a subunit of cytochrome c oxidase (Huttemann et al., 2003), exhibited a significant increase in basal expression that could be associated with the increased mitochondrial activity of the c12 cell line. This indicates the augmented activity may involve other genes not represented on the array or by other nontranscriptionally mediated factors, such as changes in enzymatic activity due to posttranslational signaling events. In addition, it could also reflect the incomplete status of the available gene annotation.
In the c1 mutants, Cox6b exhibited lower basal expression, in agreement with the lower MTT activity. However, the higher basal expression of NADH dehydrogenase subunits (Ndufa1 and Ndufv1) (Au et al., 1999; Schuelke et al., 2002) suggests higher mitochondrial activity should have been observed.
Expression differences in mitochondrial transcripts may address some of the differences in mitochondrial activity detected by the MTT assay, but the activity itself also correlated with the observed proliferative rates. Thus, changes in mitochondrial gene expression in mutant cells may be attributed to different energy needs related to their proliferative rate. Additional mechanisms may also be involved with the differences in basal mitochondrial activity, as transcript levels do not reflect changes in protein levels or enzymatic activity.
Proliferation
Despite exhibiting a slower proliferative rate, numerous proliferative genes exhibited higher basal expression in c1 cells, including insulin-like growth factor 1 (Igf1) and thymopoietin (Tmpo). Igf1 is a well-known mediator of cell proliferation through receptor signaling (reviewed in Furstenberger and Senn, 2002), while Tmpo may have a role in cell cycle control, as its transcripts are more abundant in cancerous cell lines compared to tissues in a slower proliferative state (Weber et al., 1999). Ornithine decarboxylase (Odc), involved in the synthesis of polyamines which support cell proliferation (reviewed in Luk and Casero, 1987), exhibited increased basal expression, while the lower transcript levels of glypican 1 (Gpc1), a membrane-anchored polyamine receptor, restrict polyamine uptake, consistent with the ligand-mediated feedback mechanism associated with this receptor (Belting et al., 2003). Collectively, this suggests that the c1 mutants would have a higher proliferative rate in comparison to the wild-type line. Furthermore, lower basal expression of cyclin-dependent kinase inhibitor 2d (Cdkn2i), an inhibitor of cell cycle kinases (Hirai et al., 1995), should also facilitate proliferation. However, the decrease in basal transcripts of syndecan 1 (Sdc1), a growth hormone receptor involved with adhesion, is consistent with the decrease proliferation observed with the c1 mutants (Maeda et al., 2004) in addition to its role in tumor invasion (Langford et al., 2005). The lower proliferative rate of c1 cells suggests the lower basal expression of Sdc1 may dominate, while changes in gene expression supportive of proliferation serve a compensatory role in an attempt to reestablish wild-type proliferation rates. These inconsistencies between changes in gene expression and functional assays illustrate the importance of phenotypic anchoring in the interpretation of gene expression data.
Numerous basal expression differences in c4 cells were found to be associated with proliferation. Although c4 cell proliferation was not found to be significantly different from WT, the data trend indicates that significance may be detected beyond 48 h. Studies beyond 48 h were not conducted, as cell cultures approached confluency which may confound the interpretation due to different proliferation kinetics. Higher transcript levels of cyclin D2 (Ccnd2) and cyclin-dependent kinase 4 (Cdk4) are required for the G1 S transition (Ando et al., 1993; Ando and Griffin, 1995), suggesting that the cells are positively regulated for growth. RanGTPase hydrolytic activity is essential for DNA synthesis, and its activity is mediated by RAN GTPase activating protein 1 (Rangap1), which is subsequently promoted through RAN-binding protein 1 (Ranbp1) interaction (Guarguaglini et al., 2000). Higher basal expression of Rangap1 and Ranbp1 may be factors regulating the elevated proliferation observed in the c4 mutants.
Other increases in basal transcript levels appear to limit proliferation, such as ornithine decarboxylase antizyme (Oaz1), which represses polyamine synthesis by promoting Odc degradation (Schipper et al., 2004), growth arrest specific 1 (Gas1), an inducible proliferation inhibitor (Mellstrom et al., 2002), and ubiquitin-conjugating enzyme E2C (Ube2c), which degrades cyclin B1 in plants (Criqui et al., 2002). Although these anti-proliferative signals are present, the overall pro-proliferative effects prevail.
General Conclusion
Our study demonstrates that numerous genes in the B[a]P-resistant mutant cell lines derived from the Hepa-1c1c7 line are differentially regulated at the basal level in comparison to the wild-type parent. We have observed changes in gene expression that may contribute to differences in morphology, mitochondrial activity, and cell proliferation rates, suggesting that B[a]P selection introduced mutations in addition to the well-characterized affects on Cyp1a1, AhR, and ARNT. However, the results are limited by the lack of an array with cDNA representative of all mouse genes, the lack of confirmation of increases or decreases in basal protein expression and/or enzymatic activity, and the realization that the available functional annotation is limited and likely biased. Nevertheless, these results indicate that conclusions regarding reported differences between wild-type and mutant cell lines may be partially attributed to factors other than mutations to Cyp1a1, AhR, and ARNT alone.
SUPPLEMENTARY DATA
Supplementary data on gene expression differences are available at Toxicological Sciences online as well as at http://dbzach.fst.msu.edu.
ACKNOWLEDGMENTS
Thanks to Matthew Ramer for assistance in cell maintenance and completing microarray experiments as well as to Dr. Jeremy Burt, Darrell Boverhof, and Edward Dere for aiding in the editing and proofing of the manuscript. Funding for this study was provided by NIEHS grant ES 011271 and ES 12245 to TRZ as well as fellowships from the MSU Biochemistry and Molecular Biology Department to CJF. TRZ is partially supported by the Michigan Agricultural Experiment Station.
Conflict of interest: none declared.
REFERENCES
Ando, K., Ajchenbaum-Cymbalista, F., and Griffin, J. D. (1993). Regulation of G1/S transition by cyclins D2 and D3 in hematopoietic cells. Proc. Natl. Acad. Sci. U.S.A. 90, 9571–9575.
Ando, K., and Griffin, J. D. (1995). Cdk4 integrates growth stimulatory and inhibitory signals during G1 phase of hematopoietic cells. Oncogene 10, 751–755.
Au, H. C., Seo, B. B., Matsuno-Yagi, A., Yagi, T., and Scheffler, I. E. (1999). The NDUFA1 gene product (MWFE protein) is essential for activity of complex I in mammalian mitochondria. Proc. Natl. Acad. Sci. U.S.A. 96, 4354–4359.
Bartholomew, J. C., Salmon, A. G., Gamper, H. B., and Calvin, M. (1975). Benzo(alpha)pyrene effects on mouse epithelial cells in culture. Cancer Res. 35, 851–856.
Baumann, H., and Jahreis, G. P. (1983). Regulation of mouse haptoglobin synthesis. J. Cell Biol. 97, 728–736.
Belting, M., Mani, K., Jonsson, M., Cheng, F., Sandgren, S., Jonsson, S., Ding, K., Delcros, J. G., and Fransson, L. A. (2003). Glypican-1 is a vehicle for polyamine uptake in mammalian cells. A pivotal role for nitrosothiol-derived nitric oxide. J. Biol. Chem. 278, 47181–47189.
Bernhard, H. P., Darlington, G. J., and Ruddle, F. H. (1973). Expression of liver phenotypes in cultured mouse hepatoma cells: Synthesis and secretion of serum albumin. Dev. Biol. 35, 83–96.
Boverhof, D. R., Burgoon, L. D., Tashiro, C., Chittim, B., Harkema, J. R., Jump, J. B., and Zacharewski, T. R. (2005). Temporal and dose-dependent hepatic gene expression patterns in mice provide new insights into TCDD-mediated hepatotoxicity. Toxicol Sci. 85, 1048–1063.
Boverhof, D. R., Fertuck, K. C., Burgoon, L. D., Eckel, J. E., Gennings, C., and Zacharewski, T. R. (2004). Temporal and dose-dependent hepatic gene expression changes in immature ovariectomized mice following exposure to ethynyl estradiol. Carcinogenesis. 25, 1277–1291.
Brazma, A., Hingamp, P., Quackenbush, J., Sherlock, G., Spellman, P., Stoeckert, C., Aach, J., Ansorge, W., Ball, C. A., Causton, H. C., et al. (2001). Minimum information about a microarray experiment (MIAME)-toward standards for microarray data. Nat. Genet. 29, 365–371.
Burgoon, L. D., Boutros, P. C., Dere, E., Pai, S. S., Sun, Y., Aiyar, R., Vakharia, J., and Zacharewski, T. R. (2003). dbZach: Relational toxicogenomics database system. In 42nd Annual Meeting of the Society of Toxicology. Society of Toxicology, Salt Lake City, UT.
Burgoon, L. D., Eckel-Passow, J. E., Gennings, C., Boverhof, D. R., Burt, J. W., and Zacharewski, T. R. (2005). Protocols for the Assurance of Microarray Data Quality and Process Control. (submitted).
Chen, H., Morris, M. A., Rossier, C., Blouin, J. L., and Antonarakis, S. E. (1995). Cloning of the cDNA for the human ATP synthase OSCP subunit (ATP5O) by exon trapping and mapping to chromosome 21q22.1-q22.2. Genomics 28, 470–476.
Criqui, M. C., de Almeida Engler, J., Camasses, A., Capron, A., Parmentier, Y., Inze, D., and Genschik, P. (2002). Molecular characterization of plant ubiquitin-conjugating enzymes belonging to the UbcP4/E2-C/UBCx/UbcH10 gene family. Plant Physiol. 130, 1230–1240.
Darlington, G. J., Bernhard, H. P., Miller, R. A., and Ruddle, F. H. (1980). Expression of liver phenotypes in cultured mouse hepatoma cells. J. Natl. Cancer Inst. 64, 809–819.
Furstenberger, G., and Senn, H. J. (2002). Insulin-like growth factors and cancer. Lancet Oncol. 3, 298–302.
Guarguaglini, G., Renzi, L., D'Ottavio, F., Di Fiore, B., Casenghi, M., Cundari, E., and Lavia, P. (2000). Regulated Ran-binding protein 1 activity is required for organization and function of the mitotic spindle in mammalian cells in vivo. Cell Growth Differ. 11, 455–465.
Gungabissoon, R. A., and Bamburg, J. R. (2003). Regulation of growth cone actin dynamics by ADF/cofilin. J. Histochem. Cytochem. 51, 411–420.
Hankinson, O. (1979). Single-step selection of clones of a mouse hepatoma line deficient in aryl hydrocarbon hydroxylase. Proc. Natl. Acad. Sci. U.S.A. 76, 373–376.
Hankinson, O., Andersen, R. D., Birren, B. W., Sander, F., Negishi, M., and Nebert, D. W. (1985). Mutations affecting the regulation of transcription of the cytochrome P1-450 gene in the mouse Hepa-1 cell line. J. Biol. Chem. 260, 1790–1795.
Hayes, M. J., Rescher, U., Gerke, V., and Moss, S. E. (2004). Annexin-actin interactions. Traffic 5, 571–576.
Hirai, H., Roussel, M. F., Kato, J. Y., Ashmun, R. A., and Sherr, C. J. (1995). Novel INK4 proteins, p19 and p18, are specific inhibitors of the cyclin D-dependent kinases CDK4 and CDK6. Mol. Cell. Biol. 15, 2672–2681.
Hoffman, E. C., Reyes, H., Chu, F. F., Sander, F., Conley, L. H., Brooks, B. A., and Hankinson, O. (1991). Cloning of a factor required for activity of the Ah (dioxin) receptor. Science 252, 954–958.
Huttemann, M., Schmidt, T. R., and Grossman, L. I. (2003). A third isoform of cytochrome c oxidase subunit VIII is present in mammals. Gene 312, 95–102.
Jeong, H. G., and Lee, S. S. (1998). Suppressive effects of estradiol on 2,3,7,8-tetrachlorodibenzo-p-dioxin-mediated transcriptional activation of murine Cyp1a-1 in mouse hepatoma Hepa 1c1c7 cells. Cancer Lett. 133, 177–184.
Kimura, S., Smith, H. H., Hankinson, O., and Nebert, D. W. (1987). Analysis of two benzo[a]pyrene-resistant mutants of the mouse hepatoma Hepa-1 P(1)450 gene via cDNA expression in yeast. EMBO J. 6, 1929–1933.
Kinoshita, M., Kumar, S., Mizoguchi, A., Ide, C., Kinoshita, A., Haraguchi, T., Hiraoka, Y., and Noda, M. (1997). Nedd5, a mammalian septin, is a novel cytoskeletal component interacting with actin-based structures. Genes Dev. 11, 1535–1547.
Kocarek, T. A., Schuetz, E. G., and Guzelian, P. S. (1993). Transient induction of cytochrome P450 1A1 mRNA by culture medium component in primary cultures of adult rat hepatocytes. In Vitro Cell Dev. Biol. 29A, 62–66.
Langford, J. K., Yang, Y., Kieber-Emmons, T., and Sanderson, R. D. (2005). Identification of an invasion regulatory domain within the core protein of syndecan-1. J. Biol. Chem. 280, 3467–3473.
Legraverend, C., Hannah, R. R., Eisen, H. J., Owens, I. S., Nebert, D. W., and Hankinson, O. (1982). Regulatory gene product of the Ah locus. Characterization of receptor mutants among mouse hepatoma clones. J. Biol. Chem. 257, 6402–6407.
Lemmon, M. A. (2003). Phosphoinositide recognition domains. Traffic 4, 201–213.
Li, H. S., Zhang, J. Y., Thompson, B. S., Deng, X. Y., Ford, M. E., Wood, P. G., Stolz, D. B., Eagon, P. K., and Whitcomb, D. C. (2001). Rat mitochondrial ATP synthase ATP5G3: Cloning and upregulation in pancreas after chronic ethanol feeding. Physiol. Genomics 6, 91–98.
Liu, R. M., Vasiliou, V., Zhu, H., Duh, J. L., Tabor, M. W., Puga, A., Nebert, D. W., Sainsbury, M., and Shertzer, H. G. (1994). Regulation of [Ah] gene battery enzymes and glutathione levels by 5,10-dihydroindeno[1,2-b]indole in mouse hepatoma cell lines. Carcinogenesis 15, 2347–2352.
Loeffen, J., van den Heuvel, L., Smeets, R., Triepels, R., Sengers, R., Trijbels, F., and Smeitink, J. (1998). cDNA sequence and chromosomal localization of the remaining three human nuclear encoded iron sulphur protein (IP) subunits of complex I: The human IP fraction is completed. Biochem. Biophys. Res. Commun. 247, 751–758.
Luk, G. D., and Casero, R. A., Jr. (1987). Polyamines in normal and cancer cells. Adv. Enzyme Regul. 26, 91–105.
Maeda, T., Alexander, C. M., and Friedl, A. (2004). Induction of syndecan-1 expression in stromal fibroblasts promotes proliferation of human breast cancer cells. Cancer Res. 64, 612–621.
Mattes, W. B., Pettit, S. D., Sansone, S. A., Bushel, P. R., and Waters, M. D. (2004). Database development in toxicogenomics: Issues and efforts. Environ. Health Perspect. 112, 495–505.
Mellstrom, B., Cena, V., Lamas, M., Perales, C., Gonzalez, C., and Naranjo, J. R. (2002). Gas1 is induced during and participates in excitotoxic neuronal death. Mol. Cell. Neurosci. 19, 417–429.
Merchant, M., Morrison, V., Santostefano, M., and Safe, S. (1992). Mechanism of action of aryl hydrocarbon receptor antagonists: Inhibition of 2,3,7,8-tetrachlorodibenzo-p-dioxin-induced CYP1A1 gene expression. Arch. Biochem. Biophys. 298, 389–394.
Miller, A. G., Israel, D., and Whitlock, J. P., Jr. (1983). Biochemical and genetic analysis of variant mouse hepatoma cells defective in the induction of benzo(a)pyrene-metabolizing enzyme activity. J. Biol. Chem. 258, 3523–3527.
Miranda, C. L., Aponso, G. L., Stevens, J. F., Deinzer, M. L., and Buhler, D. R. (2000). Prenylated chalcones and flavanones as inducers of quinone reductase in mouse Hepa 1c1c7 cells. Cancer Lett. 149, 21–29.
Natalicchio, A., Laviola, L., De Tullio, C., Renna, L. A., Montrone, C., Perrini, S., Valenti, G., Procino, G., Svelto, M., and Giorgino, F. (2004). Role of the P66SHC isoform in IGF-I receptor signal through MEK/ERK and regulation of actin cytoskeleton in rat myoblasts. J. Biol. Chem. 279, 43900–43909.
Numayama-Tsuruta, K., Kobayashi, A., Sogawa, K., and Fujii-Kuriyama, Y. (1997). A point mutation responsible for defective function of the aryl-hydrocarbon-receptor nuclear translocator in mutant Hepa-1c1c7 cells. Eur. J. Biochem. 246, 486–495.
Ohtsu, K., Nakazono, M., Tsutsumi, N., and Hirai, A. (2001). Characterization and expression of the genes for cytochrome c oxidase subunit VIb (COX6b) from rice and Arabidopsis thaliana. Gene 264, 233–239.
Papa, S., Sardanelli, A. M., Scacco, S., Petruzzella, V., Technikova-Dobrova, Z., Vergari, R., and Signorile, A. (2002). The NADH: Ubiquinone oxidoreductase (complex I) of the mammalian respiratory chain and the cAMP cascade. J. Bioenerg. Biomembr. 34, 1–10.
Papaconstantinou, J., Hill, R. E., Gibson, W. H., and Rao, E. Y. (1978). Synthesis and secretion of transferrin by cultured mouse hepatoma cells. Differentiation 10, 139–145.
Patel, A. S., Schechter, G. L., Wasilenko, W. J., and Somers, K. D. (1998). Overexpression of EMS1/cortactin in NIH3T3 fibroblasts causes increased cell motility and invasion in vitro. Oncogene 16, 3227–3232.
Pollenz, R. S. (1996). The aryl-hydrocarbon receptor, but not the aryl-hydrocarbon receptor nuclear translocator protein, is rapidly depleted in hepatic and nonhepatic culture cells exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin. Mol. Pharmacol. 49, 391–398.
Probst, M. R., Reisz-Porszasz, S., Agbunag, R. V., Ong, M. S., and Hankinson, O. (1993). Role of the aryl hydrocarbon receptor nuclear translocator protein in aryl hydrocarbon (dioxin) receptor action. Mol. Pharmacol. 44, 511–518.
Quinlan, M. P. (2004). Vinculin, VASP, and profilin are coordinately regulated during actin remodeling in epithelial cells, which requires de novo protein synthesis and protein kinase signal transduction pathways. J. Cell Physiol. 200, 277–290.
Reyes, H., Reisz-Porszasz, S., and Hankinson, O. (1992). Identification of the Ah receptor nuclear translocator protein (Arnt) as a component of the DNA binding form of the Ah receptor. Science 256, 1193–1195.
Rockett, J. C., and Dix, D. J. (1999). Application of DNA arrays to toxicology. Environ. Health Perspect. 107, 681–685.
Schipper, R. G., Cuijpers, V. M., De Groot, L. H., Thio, M., and Verhofstad, A. A. (2004). Intracellular localization of ornithine decarboxylase and its regulatory protein, antizyme-1. J. Histochem. Cytochem. 52, 1259–1266.
Schuelke, M., Detjen, A., van den Heuvel, L., Korenke, C., Janssen, A., Smits, A., Trijbels, F., and Smeitink, J. (2002). New nuclear encoded mitochondrial mutation illustrates pitfalls in prenatal diagnosis by biochemical methods. Clin. Chem. 48, 772–775.
Segner, H., Behrens, A., Joyce, E. M., Schirmer, K., and Bols, N. C. (2000). Transient induction of 7-ethoxyresorufin-O-deethylase (EROD) activity by medium change in the rainbow trout liver cell line, RTL-W1. Mar. Environ. Res. 50, 489–493.
Seidel, S. D., and Denison, M. S. (1999). Differential gene expression in wild-type and arnt-defective mouse hepatoma (Hepa1c1c7) cells. Toxicol. Sci. 52, 217–225.
Seubert, J. M., Darmon, A. J., El-Kadi, A. O., D'Souza, S. J., and Bend, J. R. (2002). Apoptosis in murine hepatoma hepa 1c1c7 wild-type, C12, and C4 cells mediated by bilirubin. Mol. Pharmacol. 62, 257–264.
Spence, H. J., Dhillon, A. S., James, M., and Winder, S. J. (2004). Dystroglycan, a scaffold for the ERK-MAP kinase cascade. EMBO Rep. 5, 484–489.
Sumoza-Toledo, A., and Santos-Argumedo, L. (2004). The spreading of B lymphocytes induced by CD44 cross-linking requires actin, tubulin, and vimentin rearrangements. J. Leukoc. Biol. 75, 233–239.
Takeuchi, K., Kawashima, A., Nagafuchi, A., and Tsukita, S. (1994). Structural diversity of band 4.1 superfamily members. J. Cell Sci. 107(7), 1921–1928.
Thompson, E. B., Benedict, W. F., Owens, I. S., and Nebert, D. W. (1974). Aryl hydrocarcon hydroxylase and tyrosine aminotransferase activities in somatic-cell hybrids derived from hepatoma tissue culture HTC (rat) cells and 3T3 (mouse) benzo(a)pyrene-resistant cells. Biochem. Biophys. Res. Commun. 56, 605–610.
Van Gurp, J. R., and Hankinson, O. (1984). Isolation and characterization of revertants from four different classes of aryl hydrocarbon hydroxylase-deficient hepa-1 mutants. Mol. Cell. Biol. 4, 1597–1604.
Weber, P. J., Eckhard, C. P., Gonser, S., Otto, H., Folkers, G., and Beck-Sickinger, A. G. (1999). On the role of thymopoietins in cell proliferation. Immunochemical evidence for new members of the human thymopoietin family. Biol. Chem. 380, 653–660.
Wei, Y. D., Bergander, L., Rannug, U., and Rannug, A. (2000). Regulation of CYP1A1 transcription via the metabolism of the tryptophan-derived 6-formylindolo[3,2-b]carbazole. Arch. Biochem. Biophys. 383, 99–107.
Wells, C. M., Walmsley, M., Ooi, S., Tybulewicz, V., and Ridley, A. J. (2004). Rac1-deficient macrophages exhibit defects in cell spreading and membrane ruffling but not migration. J. Cell. Sci. 117, 1259–1268.
Wild, A. C., Yu, J. W., Lemmon, M. A., and Blumer, K. J. (2004). The p21-activated protein kinase-related kinase Cla4 is a coincidence detector of signaling by Cdc42 and phosphatidylinositol 4-phosphate. J. Biol. Chem. 279, 17101–17110.
Yuen, T., Wurmbach, E., Pfeffer, R. L., Ebersole, B. J., and Sealfon, S. C. (2002). Accuracy and calibration of commercial oligonucleotide and custom cDNA microarrays. Nucleic Acids Res. 30, e48.
Zacharewski, T. R., Bondy, K. L., McDonell, P., and Wu, Z. F. (1994). Antiestrogenic effect of 2,3,7,8-tetrachlorodibenzo-p-dioxin on 17 beta-estradiol-induced pS2 expression. Cancer Res. 54, 2707–2713.
Zhang, J., Watson, A. J., Probst, M. R., Minehart, E., and Hankinson, O. (1996). Basis for the loss of aryl hydrocarbon receptor gene expression in clones of a mouse hepatoma cell line. Mol. Pharmacol. 50, 1454–1462.(C. J. Fong, L. D. Burgoon)