Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Cell Cycle. Author manuscript; available in PMC 2007 December 6.
Published in final edited form as:
Published online 2007 April 19.
PMCID: PMC2117899

Identification of Primary Transcriptional Regulation of Cell Cycle-Regulated Genes upon DNA Damage


The changes in global gene expression in response to DNA damage may derive from either direct induction or repression by transcriptional regulation or indirectly by synchronization of cells to specific cell cycle phases, such as G1 or G2. We developed a model that successfully estimated the expression levels of >400 cell cycle-regulated genes in normal human fibroblasts based on the proportions of cells in each phase of the cell cycle. By isolating effects on the gene expression associated with the cell cycle phase redistribution after genotoxin treatment, the direct transcriptional target genes were distinguished from genes for which expression changed secondary to cell synchronization. Application of this model to ionizing radiation (IR)-treated normal human fibroblasts identified 150 of 406 cycle-regulated genes as putative direct transcriptional targets of IR-induced DNA damage. Changes in expression of these genes after IR treatment derived from both direct transcriptional regulation and cell cycle synchronization.

Keywords: microarray, cell cycle, ionizing radiation, human fibroblasts, EPIG


Cell proliferation is a fundamental biological activity that is regulated through many gene products, including cyclin-dependent kinases (CDKs), enzymes for DNA replication and repair, enzymes for chromatin condensation and segregation, proto-oncogenes, and tumor suppressor genes.1,2 Many cell cycle regulators undergo dramatic changes in levels of expression and activity to propel or regulate progression through the cell division cycle.3 Identification of cell cycle-regulated genes will facilitate understanding of not only normal biological processes during cell division but also mechanisms of pathologic response to environmental toxicants.

Microarray technology enables researchers to determine changes in expression of thousands of genes in cells and tissues in response to treatment with exogenous chemicals or drugs.4,5 A commonly addressed question in microarray experiments is which genes are direct transcriptional targets of carcinogen-induced DNA damage, and which are altered secondarily to inhibition of cell cycle progression following DNA damage? It is not uncommon for hundreds or even thousands of genes to be changed significantly after a specific drug or toxicant treatment. However, toxic agents that cause DNA damage, such as IR, UV and chemotherapeutic drugs for cancer, usually inhibit cell proliferation by activating cell cycle checkpoint functions and causing growth arrest at specific cell cycle phases.2,6,7 The synchronizing effect of the cell cycle checkpoint response to DNA damage can affect the expression of cell cycle- or growth-regulated genes at later times8,9 masking direct transcriptional targets.

To quantify the effects on gene expression resulting from changes in the proportions of cells in G1, S, G2 and M phases of the cell division cycle, a baseline level for all transcripts in each of these phases must first be established, and then gene expression in a population of cells can be estimated when the proportionate distribution of cells in the various cycle compartments is known. Although global transcriptional regulation during the cell cycle has been widely investigated in many organisms using microarray technology,1,3,1015 none of the current methods used to synchronize cells results in 100% of the cells in a specific cell cycle phase. Furthermore, the cells become asynchronous with time after release from the synchronizing block. Therefore, the measured transcript levels do not truly represent a pure cell population.

In the present study, a model was developed to determine “true” expression levels of genes in each pure cell cycle phase. Cell-cycle-specific gene expression patterns and cell cycle-regulated genes were further identified by EPIG method.16 Once the levels of cell cycle-regulated genes in pure G1, S, G2 or M phase cells were obtained, we were able to isolate effects of DNA damage-induced redistribution of cell cycle phases on expression of cell cycle-regulated genes, and then to identify direct transcriptional targets of a genotoxin.


Cell lines and culture

Normal human fibroblasts, NHF1, NHF3 and NHF10, were derived from neonatal foreskins and established in culture according to established methods.17 Immortalized cell lines were obtained by ectopic expression of human telomerase (hTERT), as previously described.18,19 Fibroblasts were cultured in DMEM high-glucose medium (Invitrogen, Carlsbad, CA USA) supplemented with 2 mM L-glutamine (Invitrogen, Carlsbad, CA, USA) and 10% fetal bovine serum (FBS, Sigma Chemical Co, St. Louis, MO, USA). All cell lines were maintained at 37°C in a humidified atmosphere of 5% CO2 and were routinely tested and shown to be free of mycoplasma contamination using a commercial kit (Gen-Probe, San Diego, CA, USA).

Cell irradiation

Cells were exposed to ionizing radiation in their culture medium, using a 137Cs source (Gammacell of Canada) at a dose rate of 0.84 Gy/min. Sham-treated controls were subjected to the same movements in and out of incubators as irradiated cells.

Cell cycle compartment determination

For quantification of G1, S and G2 phase cells, 10 μM 5′-bromo-2′-deoxyuridine (BrdU, Sigma Chemical Co.) was added to culture medium at various times after irradiation to label S phase cell DNA and cultures were incubated for another 2 hours. Cells were harvested, washed with phosphate-buffered saline (PBS), and fixed with 67% ethanol in PBS. Cells were stained with fluorescein isothiocyanate (FITC)-conjugated anti-BrdU antibody (BD Biosciences, San Jose, CA, USA) and propidium iodide (PI, Sigma Chemical Co., St. Louis, MO, USA), then G1, S and G2 phase cells were enumerated by flow cytometry as previously described.2022 For analysis of M phase cells, cells were harvested at various times after irradiation, washed with PBS, fixed in 1% formaldehyde in PBS for 30 min, and then fixed in 67% ethanol in PBS. Cells were incubated in 0.5 μg/100μl anti-phospho-histone H3 antibody (Upstate Biotechnology Inc., Lake Placid, NY, USA) for 2 h, and then stained with FITC-conjugated anti-rabbit antibody (Santa Cruz) and propidium iodide.2325 Flow cytometric analyses to enumerate mitotic cells were done using a FACScan flow cytometer and Summit software (DakoCytomation, Fort Collins, CO, USA).

Cell synchronization

NHF1, NHF3 and NHF10 cells were synchronized as previously described.19,26,27 Briefly, cells were plated at a density of 1.3 × 104/cm2 and allowed to grow for 8 days to confluence-arrest in G0 phase. During this time cell growth medium was changed on days 3 and 5 post-seeding. Arrested cells were released from dishes with trypsin, reseeded at 1.3 × 104/cm2 and then incubated for 8 h to allow them to reach mid G1. Cells similarly released from G0 were incubated with the DNA polymerase inhibitor, aphidicolin (APC), at a concentration of 2 μg/ml for 24 h to collect cells at the beginning of S. APC was washed out and cells incubated for 3 hr to allow accumulation to mid S phase. Cells that were released from APC treatment also were incubated for 8 h to accumulate in G2. Colcemid at 100 ng/ml was added to such G2 cells and cells were incubated an additional 4 h to collect cells in mitosis (M). Ten million cells were harvested at each cell cycle phase for RNA isolation. Two million cells were sampled for flow cytometric analysis of DNA content by staining with propidium iodide. Modfit commercial software (Verity Software House Inc.,Topsham, ME, USA) was used to determine the proportions of cells in G0/G1, S and G2/M. The proportions of cells in M were determined by flow cytometric enumeration of phospho-histone H3-expressing cells with 4N DNA content.

Oligo DNA microarray

Total RNA was isolated from synchronized cell populations using a Qiagen RNeasy kit. The quality of all RNA samples was confirmed using an Agilent 2100 Bioanalyzer. Microarray analysis was then performed through a contract with Icoria Corporation as follows. Briefly, 1 μg of sample RNA and global reference RNA (Stratagene, La Jolla, CA, USA) were converted to cDNA with reverse transcriptase then amplified using T7 RNA polymerase while labeling with either Cy3-dUTP or Cy5-dUTP (Agilent’s Low RNA Input Linear Amplification Kit, Agilent Techologies, USA). The quality of each labeled cRNA was evaluated using an Agilent 2100 Bioanalyzer prior to hybridization. 750 ng of Cy3 and Cy5-labeled cRNA were used in the hybridization. The labeled cRNA from samples was hybridized with the labeled global reference cRNA on an Agilent 22k human 1A array (Agilent Techologies, USA) in a hybridization oven (Robbins Scientific Model 400, 1040-60-1AG, Sunnyvale, CA, USA) at 60°C for 17 h. Hybridization of sample RNA against reference RNA was done twice with a fluorophore reversal (dye swap). Following hybridization, the arrays were scanned using the Agilent DNA Microarray Scanner with SureScan® Technology and microarray images were analyzed using Agilent Feature Extraction Software (v7.1).

Computational analysis of microarray data

(1) Cell cycle-regulated gene extraction. As previously reported26 and shown in (Table 1), the synchronization methods yielded populations of cells that were highly enriched in selected cycle-phase compartments but the synchronization was imperfect. Cultures synchronized in G1 phase displayed 82–91% 2N DNA content with small proportions of cells in S, G2 and M. Cultures that were synchronized to S, G2 and M displayed 46–71%, 54–70% and 23–32% of cells in these phases, respectively. Global gene expression profiles were generated from the synchronized cell populations. To calculate theoretically “pure” G1-, S-, G2- or M-specific gene expression levels, four simultaneous equations were derived for each gene based on the microarray-determined expression levels and the proportions of cells in each cycle phase for each of the synchronized samples. The expression levels of a gene in G1, S, G2, or M, assuming that it has a linear relationship with the proportion of cells in each cycle compartment, may be expressed as:

Table 1
Synchronization of three normal human diploid fibroblast lines (%)

where genei,j is the measured expression level of gene i in cultures synchronized in phase j that is an average of the duplicate arrays (C3 or C5 labeling), j [set membership] {G1,G2,S,Mfj,G1, is the fraction of G1 cells in cultures synchronized in phase j ; conversely for fj,S, fj,G2 and fj,M · GENEi,j is the expression level of gene i if the gene was expressed in a perfectly synchronized phase j. The equations were simultaneously solved to calculate “pure” G1-, S-, G2- and M-specific expression levels. In this model, interactions between genes or cell cycle compartments were assumed to have no effect on the equations.

A targeted evaluation of 45 well-known cell cycle-regulated genes demonstrated that the model expanded the range of variation in expression levels across the cell cycle. The levels of transcripts specific to M cells were most impacted by the model (Table 2). This was to be expected as synchronization yielded an M population representing only 22–32% of the cells, while the model estimated transcript levels for a theoretically pure sample of mitotic cells.

Table 2
Comparison of expression levels of cell-cycle-regulated genes before and after phase correction in NHF1-hTERT fibroblasts*

GENEi,j, expression level of a gene with purity correction, was applied in EPIG analysis for extraction of patterns of cell cycle-regulated gene expression patterns and significant genes. In the EPIG method, the extracted log2 pixel intensity ratio values (sample vs. reference) were preprocessed, which included systematic variation normalization, dye-swap correction, and cell line alignment.28 Three parameters, the correlation coefficient within a specific pattern, the magnitude of change, and the signal-to-noise ratio (SNR), were employed for selection of significant genes.16

(2) Identification of primary transcriptional targets. Gene expression in sham- or IR-treated cell populations was estimated using GENEi,j weighted by the proportions of cells in G1, S, G2 and M. The estimated values were then compared with the observed values from microarrays to distinguish changes in gene expression caused by alterations in cell cycle distribution from changes caused directly by IR treatment. The comparisons were done by paired t-test to identify significant genes showing difference between the estimated values and the observed values.

(3) Gene ontology analysis of significant genes. Categories of genes that were over-represented in a selected gene list compared to what was represented in the microarray were analyzed using EASE ( Such over-represented categories represent biological “themes” of a given list.

(4) Identification of transcription factors (TFs) in cell cycle-regulated genes. Computational promoter analysis was done using the PRIMA software, described in detail by Elkon et al.29 Briefly, given a target set and a background set of promoters, PRIMA performs statistical tests, based on hypergeometric distribution, aimed at identifying TFs whose binding site signatures are significantly more prevalent in the target set than in the background set. In this study, each of the two sets of cell cycle-regulated genes (IR-target genes and non-IR-target genes) was considered a target set and the entire set of genes represented on the microarray served as the background set. PRIMA uses position weight matrices (PWMs) as models for regulatory sites that are bound by TFs. PWMs that represent human or mouse TF binding sites were obtained from the TRANSFAC database.30 Putative promoter sequences corresponding to all known human genes were extracted from the human genome, masking out repetitive elements (Ensembl, version 27).31 PRIMA promoters’ scan was done on both strands and was confined to the region from 600 bp upstream to 100 bp downstream to the putative genes’ transcription start site.


Extraction of cell cycle-regulated genes

From the purity-corrected microarray data, EPIG was able to identify nine cell cycle-specific expression patterns (Fig. 1) and 2410 genes or ESTs as cell-cycle-specific based on criteria of three parameters: correlation coefficient within a specific pattern (r ≥ 0.64, n = 24, p < 1×10−6), the magnitude of change (log2 ratio of sample vs. reference > 0.4), and the signal-to-noise ratio (SNR > 3, p <0.01) (Chou, Zhou, Bushel, Kaufmann and Paules, in preparation). The number of genes in each pattern ranged from 19 to 473 (Table 3). All expected patterns of gene expression across the cell cycle were detected. About 500 genes were expressed at highest levels in G1 and at lower levels in S, G2 and M (Patterns 1 and 2). Over 1100 genes were expressed at highest levels in S and at variably lower levels in G1, G2 and M (Patterns 3–7). No pattern displayed a peak of gene expression in G2 but over 700 genes in Patterns 8 and 9 displayed maximal expression in M. Among the genes that were expressed at highest levels in G1 were the well-known growth-enhancing transcription factor Myc, and GOS2, a gene known to be expressed highly in G1 cells and to regulate the G0 to G1 transition.32 Among the genes expressed at highest levels in S were many whose protein products are known to be required for DNA replication such as ORC3, ORC4, ORC6, CDC45L, CDC6, CDC7L, MCM2, MCM3, MCM6, MCM7, POLA2, PCNA, RFC2, RFC3, and RFC5. In patterns 8 and 9 with highest expression in M were BUB1B, CCNB1, CDC20, CDC25C and PLK that are known to regulate cell progression through G2 and M. Thus, many genes that displayed cell-cycle-specific gene expression encode proteins that are expected to function during the phases of maximal expression.

Figure 1
Patterns of cell cycle-specific gene expression that were extracted using the EPIG method in three immortalized normal human fibroblast cell lines, NHF1, NHF3 and NHF10. Gene expression level is presented as log2 ratio of sample RNA against global reference ...
Table 3
Gene number and selected genes in each expression cycle-phase-specific pattern

However, cycle-specific changes in gene expression in synchronized cell populations may not reflect the changes that occur in physiologically cycling cell populations. Some genes selected by above method, although showing pattern-specific expression, may not be truly cell-cycle-regulated genes. Changes in these genes might simply reflect the G0/G1 transition (highest expression in G1), a response to aphidicolin (highest expression in S), or a response to colcemid (highest expression in M). For those truly cell-cycle-regulated genes, changes in expression levels should reflect alterations in proportions in each phase of the cell cycle.

To further test the truly cell cycle-regulated genes, a second experiment was performed in which NHF1 cells were cultured and harvested at seven different times between days 2 to 3 after reseeding. These cultures contained different proportions of cells in each cycle-phase (Table 4) and gene expression levels were determined for each culture by microarray. The hypothesis for this experiment was that expression levels of cell cycle-regulated genes were linearly correlated with the proportions of cells in each cycle phase and variations in these proportions in normally cycling cell population would alter the levels of expression of cell cycle-regulated genes (given that expression levels of a cell cycle-regulated gene in each pure cell cycle phase are stable and known). Based on the measured proportions of cells in each cycle phase in each culture, the expression levels of the 2410 cell cycle-specific genes identified earlier were estimated to compare with the microarray-determined expression levels. A gene was identified as cell-cycle-regulated only when its expression level was well-estimated (r > 0.75, P < 0.05) by the proportions of cells in the various cycle phases. A total of 406 genes were confirmed as cell cycle-regulated by displaying a significant high correlation between the estimated and the microarray-determined expression levels. GO analysis of the 406 genes showed that all cell-cycle-related categories were significantly over-represented. The same analysis applied to the remainder of the 2410 genes yielded 12 categories related to intracellular processes, nucleotide and nucleic acid metabolism, and RNA binding, but no cell cycle-related categories.

Table 4
The proportions of cell cycle compartments in seven sham-treated NHF1 cells

Identification of transcriptional targets of ionizing radiation

NHF1, NFH3 and NHF10 human fibroblast lines in logarithmic growth were treated with 1.5 Gy IR, a dose that inhibited colony formation by 40–45%. Cells were harvested at 2, 6 and 24 h post-IR for microarray analysis of gene expression. Quantitative analysis of cell cycle compartments after IR treatment indicated that the proportions of G1, S, G2 and M cells changed markedly, especially at 24 h when the G1 population was substantially enriched and the S and M compartments were severely depleted (Table 5, Fig. 2). These changes in the distribution of cells across the cell cycle were primarily the consequence of the p53-dependent G1 checkpoint response.33 The cell-synchronizing effects of IR leading to redistribution of cell cycle compartments were expected to affect the expression profiles of all 406 cell cycle-regulated genes.

Figure 2
Dynamic changes in cell cycle compartments at 2, 6 and 24 hr post-IR. Status of DNA synthesis through the cell cycle was determined by BrdU incorporation. S represents S phase cell population, G1 represents G1 phase cell population and G2 represents G ...
Table 5
Changes of cell cycle compartments after IR irradiation

According to the determined proportions of cells in each of the cycle compartments in sham- and IR-treated cells (Table 5) and the expression levels of the 406 genes in each presumed pure cell cycle phase, expression levels of these cell cycle-regulated genes were estimated. Changes in expression levels (log2 ratios of sample vs reference) of the 406 genes at 2, 6 and 24 h after IR-irradiation relative to sham-treated controls were obtained by subtracting expression level of sham-treated samples from IR-treated samples and the estimated values and the measured values were compared (Table 9, Supplement).

If the values were equivalent, it would suggest that changes in the expression levels of cell- cycle-regulated genes were associated with the IR-induced cell synchronization. A total of 193 genes with changes in expression levels showing no significant difference between the estimated values and the measured values were identified as non-targets of IR.

Alternatively, if changes in the expression levels of cell-cycle-regulated genes were associated with IR directly affecting transcription in addition to cell synchronization, the measured expression levels would be impacted by both of these effects, and the measured and estimated values would be different. Accordingly, 150 genes with measured transcript levels different than estimated values were identified as IR-target genes based on a statistical significance criterion (paired t-test, P < 0.05) and fold change threshold (log2 ratios of sample vs reference >0.4). These 150 genes represent putative direct transcriptional targets of IR, of which 134 were repressed and 16 were induced. Most of the IR-target genes (131) were recognized at 24 hr post-treatment, with 44 targets recognized at 6 hr and 3 targets recognized at 2 hr.

Gene ontology analysis of the 150 target genes yielded over-represented categories related to cell cycle, cell proliferation, DNA metabolism and DNA repair. Some well-known genes were BUB1, CCNB1, CCNB2, CDC2, CDC20, E2F1, KNSL1, KNSL7, MCM2, MCM5, MCM7, PCNA, POLD1, RFC4, RFC5, TIMELESS, TOP2A, FANCA, RAD18, and RAD51 (Figs. 3a–c). Among the 193 non-target genes, CCNE1, CDC45L, CDC6, CDKN2D, CENPE, CHEK1, LIG1, MSH2, ORC6L, POLA2, and POLE were included (Fig. 3d) and over-represented categories from gene ontology also included biological process related to cell cycle, cell proliferation, DNA metabolism (Table 6). The 63 genes showing significant differences between observed and predicted but with fold changes less than 0.4 need be confirmed in further experiments.

Figure 3
Comparison of changes in expression level of selected target and non-target genes post IR-irradiation. Actual changes in cell cycle-regulated genes post IR-irradiation were measured by microarray (left), effects of cell synchronization on cell cycle-regulated ...
Table 6
Gene ontology analysis of biological processes in 150 potential target and 163 non-target genes

Transcription factors involved in target and non-target cell cycle-regulated genes

Analysis of transcription factor binding sites in the promoters of the target and non-target cell-cycle-regulated genes showed that two factors, E2F and NF-Y, were enriched in both groups while cell cycle-dependent element (CDE) was significantly enriched in the target group and upstream stimulating factor (USF) was significantly enriched in the non-target group (Table 7).

Table 7
Comparison of transcription factors (TFs) between target and non-target genes a


Cell synchronization helps identify phase-specific genes3,12,15 but does not provide quantitatively accurate expression levels of cell cycle-specific genes because of the lack of perfect synchrony. By applying simultaneous equations, the expression levels of genes in theoretically pure G1, S, G2 or M cell cycle phases were estimated. These gene expression levels were thereafter used to identify cell cycle-specific genes and to estimate the expression levels of cell cycle-regulated genes in an asynchronous cell population based on the proportions of G1, S, G2 and M cells. The results provide another image of the regulation of gene expression as human cells synchronously pass through the cell division cycle, and suggest that many cell cycle-regulated genes are actively repressed in response to IR-induced DNA damage.

It must be noted that artificially synchronized cell cycle phases do not biologically equal the phases existing in physiologically cycling cells. Although a total of 2410 genes or ESTs showing cell cycle-specific expression were identified based on nine extracted patterns by using EPIG method, they were not necessarily cycle-regulated genes. Many of these genes may have been selected because of serum stimulation after release from G0 into G1,12 or by a direct response to aphidicolin or colcemid treatment.27 Only genes whose expression levels could be correctly estimated in asynchronous cell populations based on the proportions of cells in the cycle compartments were accepted as cell cycle-regulated. A total of 406 among the 2410 genes were identified as cell cycle-regulated genes.

One of the purposes of the present study was to identify the transcriptional targets of a genotoxin among the cell cycle-regulated genes. It has been well-documented that IR induces DNA double-strand-breaks and activates checkpoints to arrest cells in G1 and G2.2,34 Hundreds of cell-cycle-regulated genes were observed to change in response to DNA damage.9,16,35,36 Genes that were not cell-cycle- regulated were easy to identify as transcriptional targets because expression of these genes was not affected by the DNA damage-induced synchronization effect. However, it is impossible in these studies to distinguish genes that were changed by primary transcriptional regulation from genes changed resulting from cell synchrony caused by the checkpoint response to DNA damage. Our model was developed to address this question. The methodology used here enabled estimation of the changes of gene expression resulting from the cell synchrony effect and identification of transcriptional targets of IR whose changes in gene expression were greater than that from cell synchrony alone.

Although the G2 checkpoint was clearly activated at 2 hr post-IR irradiation, the expression of genes that are important for the G2/M transition, such as CDC2 and CCNB1, did not show any difference between the measured and the estimated values, indicating that post-translational modifications account for the G2 checkpoint function at this time point and changes in expression of these genes were due to accumulation of the G2 population (Table 5, Fig. 3b). Starting at 6 hr post-IR, the time when G1 arrest was clearly observed, the expression of some S phase genes, like MCM2 and MCM3, showed greater repression than estimated, suggesting a direct transcriptional repression mechanism in G1 arrest. Further at 24 h post-IR, the expression of many important cell cycle-regulated genes, including CCNB1 and 2, CDC2, CDC20, CDC7L1, CDK2, MCM2, 3 and 7, RFC4, TIMELESS and TOP2α showed greater repression in the observed values also suggesting direct inhibition by IR in addition to the cell synchrony effects. These results suggest that expression of some cell-cycle-regulated genes is actively changed in response to DNA damage to contribute to cell cycle arrest. Many other cell-cycle-regulated genes were just down-regulated passively in response to DNA damage secondary to cell cycle arrest. The mechanisms of primary down-regulation of the target genes are still not clear. p53 has been reported to play an important role in trans-repression of cell-cycle-regulatory genes.35,37,38 Well-known p53-responsive genes include CDC2, CCNB1, and TOP2α.3941 Despite the transcription factors E2F and NF-Y being enriched in both groups, cell cycle-dependent element (CDE) was significantly enriched in the target group and upstream stimulating factor (USF) was enriched in the non-target group. Previous analyses of CDC2 and CCNB1 promoters suggested that p53 interacted with NF-Y to mediate trans-repression,42 although a subsequent study suggested that p53 trans-repressed through interaction with SP1.43 Conversely several groups have reported that genes containing CDE/CHR in their promoters, such as CCNB1, CCNB2, CDC2, TOP2A, RAD51, CENPA, CDC25C, can be directly trans-repressed by IR-induced DNA damage through a p53-dependent signaling pathway.4446 Two independent mechanisms, direct binding or via CDE/CHR element, are involved in p53-dependent transcriptional repression of some cell cycle-regulated genes.44 Six of the above-mentioned target genes appeared in our target gene list (Fig. 3a). The results presented here suggest that a large set of as many as 150 cell-cycle-regulated genes may be subject to p53-dependent trans-repression in response to DNA damage.

Analysis of gene ontology in the 150 potential target genes showed that although they distributed in various categories of biological process, cellular component and molecular functions, almost all of the over-represented categories were related to cell cycle regulation, DNA metabolism and cell proliferation (Table 7). Beside the post-translational modifications of checkpoint sensors, transducers, mediators and effectors, the repression of cell cycle-regulated genes may play an important role in cell cycle checkpoint function in response to IR-induced DNA damage.

In our previous publication, 1811 IR-responding genes or ESTs were identified using the EPIG method with the same criteria for significant gene extraction.16 Among the 1811 genes, 328 of the 406 cell cycle-regulated genes were identified, and all of the 150 IR-target genes were on the list. The other 1483 genes that were IR-responsive but not cell cycle-regulated included early DNA damage response genes, such as CDKN1A, BTG2, GADD45A, PLK2, PLK3 and PPMD1 that initiate or regulate cell cycle checkpoint functions. Such genes may work together with targeted repression in transcription of cell cycle-regulated genes to cause and maintain cell cycle arrest, and with genes passively responding to G0-like growth quiescence.16

Identification of transcriptional targets of the DNA damage response helps us understand the mechanisms of cell cycle arrest caused by direct regulation of cell cycle-regulated genes at the level of transcription, in addition to those checkpoint functions regulated by post-translational modification of protein. Both responses appear to play important roles in maintaining cell cycle arrest. By accounting for the changes in gene expression that are indirect manifestations of cell synchronization, this model may improve elucidation of the mechanisms of genotoxicity by environmental chemicals or therapeutic drugs that directly induce or repress transcription of target genes.


Supported by PHS grants ES11391, N01-ES-25497, and ES10126. We thank George Wu, Dong Xiang and Dr. Leping Li for help with microarray data analysis. We acknowledge PHS’s contribution of the Software EASE used in our data analysis.


Previously published online as a Cell Cycle E-publication:


1. van der Meijden CM, Lapointe DS, Luong MX, Peric-Hupkes D, Cho B, Stein JL, van Wijnen AJ, Stein GS. Gene profiling of cell cycle progression through S-phase reveals sequential expression of genes required for DNA replication and nucleosome assembly. Cancer Res. 2002;62:3233–43. [PubMed]
2. Abraham RT. Cell cycle checkpoint signaling through the ATM and ATR kinases. Genes Dev. 2001;15:2177–96. [PubMed]
3. Cho RJ, Huang M, Campbell MJ, Dong H, Steinmetz L, Sapinoso L, Hampton G, Elledge SJ, Davis RW, Lockhart DJ. Transcriptional regulation and function during the human cell cycle. Nat Genet. 2001;27:48–54. [PubMed]
4. Nuwaysir EF, Bittner M, Trent J, Barrett JC, Afshari CA. Microarrays and toxicology: The advent of toxicogenomics. Mol Carcinog. 1999;24:153–9. [PubMed]
5. Waters MD, Olden K, Tennant RW. Toxicogenomic approach for assessing toxicant-related disease. Mutat Res. 2003;544:415–24. [PubMed]
6. Melo J, Toczyski D. A unified view of the DNA-damage checkpoint. Curr Opin Cell Biol. 2002;14:237–45. [PubMed]
7. Kaufmann WK, Paules RS. DNA damage and cell cycle checkpoints. Faseb J. 1996;10:238–47. [PubMed]
8. Jen KY, Cheung VG. Transcriptional response of lymphoblastoid cells to ionizing radiation. Genome Res. 2003;13:2092–100. [PubMed]
9. Tusher VG, Tibshirani R, Chu G. Significance analysis of microarrays applied to the ionizing radiation response. Proc Natl Acad Sci USA. 2001;98:5116–21. [PubMed]
10. Cho RJ, Campbell MJ, Winzeler EA, Steinmetz L, Conway A, Wodicka L, Wolfsberg TG, Gabrielian AE, Landsman D, Lockhart DJ, Davis RW. A genome-wide transcriptional analysis of the mitotic cell cycle. Mol Cell. 1998;2:65–73. [PubMed]
11. Ishida S, Huang E, Zuzan H, Spang R, Leone G, West M, Nevins JR. Role for E2F in control of both DNA replication and mitotic functions as revealed from DNA microarray analysis. Mol Cell Biol. 2001;21:4684–99. [PMC free article] [PubMed]
12. Iyer VR, Eisen MB, Ross DT, Schuler G, Moore T, Lee JC, Trent JM, Staudt LM, Hudson J, Jr, Boguski MS, Lashkari D, Shalon D, Botstein D, Brown PO. The transcriptional program in the response of human fibroblasts to serum. Science. 1999;283:83–7. [PubMed]
13. Ren B, Cam H, Takahashi Y, Volkert T, Terragni J, Young RA, Dynlacht BD. E2F integrates cell cycle progression with DNA repair, replication, and G2/M checkpoints. Genes Dev. 2002;16:245–56. [PubMed]
14. Shedden K, Cooper S. Analysis of cell-cycle -specific gene expression in human cells as determined by microarrays and double-thymidine block synchronization. Proc Natl Acad Sci USA. 2002;99:4379–84. [PubMed]
15. Whitfield ML, Sherlock G, Saldanha AJ, Murray JI, Ball CA, Alexander KE, Matese JC, Perou CM, Hurt MM, Brown PO, Botstein D. Identification of genes periodically expressed in the human cell cycle and their expression in tumors. Mol Biol Cell. 2002;13:1977–2000. [PMC free article] [PubMed]
16. Zhou T, Chou JW, Simpson DA, Zhou Y, Mullen TE, Medeiros M, Bushel PR, Paules RS, Yang X, Hurban P, Lobenhofer EK, Kaufmann WK. Profiles of global gene expression in ionizing-radiation-damaged human diploid fibroblasts reveal synchronization behind the G1 checkpoint in a G0-like state of quiescence. Environ Health Perspect. 2006;114:553–9. [PMC free article] [PubMed]
17. Maher VM, Heflich RH, McCormick JJ. Repair of DNA damage induced in human fibro-blasts by N-substituted aryl compounds. Natl Cancer Inst Monogr. 1981:217–22. [PubMed]
18. Heffernan TP, Simpson DA, Frank AR, Heinloth AN, Paules RS, Cordeiro-Stone M, Kaufmann WK. An ATR- and Chk1-dependent S checkpoint inhibits replicon initiation following UVC-induced DNA damage. Mol Cell Biol. 2002;22:8552–61. [PMC free article] [PubMed]
19. Deming PB, Cistulli CA, Zhao H, Graves PR, Piwnica-Worms H, Paules RS, Downes CS, Kaufmann WK. The human decatenation checkpoint. Proc Natl Acad Sci USA. 2001;98:12044–9. [PubMed]
20. Kastan MB, Onyekwere O, Sidransky D, Vogelstein B, Craig RW. Participation of p53 protein in the cellular response to DNA damage. Cancer Res. 1991;51:6304–11. [PubMed]
21. Kaufmann WK, Schwartz JL, Hurt JC, Byrd LL, Galloway DA, Levedakou E, Paules RS. Inactivation of G2 checkpoint function and chromosomal destabilization are linked in human fibroblasts expressing human papillomavirus type 16 E6. Cell Growth Differ. 1997;8:1105–14. [PubMed]
22. Kaufmann WK, Heffernan TP, Beaulieu LM, Doherty S, Frank AR, Zhou Y, Bryant MF, Zhou T, Luche DD, Nikolaishvili-Feinberg N, Simpson DA, Cordeiro-Stone M. Caffeine and human DNA metabolism: The magic and the mystery. Mutat Res. 2003;532:85–102. [PubMed]
23. Juan G, Traganos F, James WM, Ray JM, Roberge M, Sauve DM, Anderson H, Darzynkiewicz Z. Histone H3 phosphorylation and expression of cyclins A and B1 measured in individual cells during their progression through G2 and mitosis. Cytometry. 1998;32:71–7. [PubMed]
24. Kaufmann WK, Levedakou EN, Grady HL, Paules RS, Stein GH. Attenuation of G2 checkpoint function precedes human cell immortalization. Cancer Res. 1995;55:7–11. [PubMed]
25. Xu B, Kim ST, Lim DS, Kastan MB. Two molecularly distinct G2/M checkpoints are induced by ionizing irradiation. Mol Cell Biol. 2002;22:1049–59. [PMC free article] [PubMed]
26. Unsal-Kacmaz K, Mullen TE, Kaufmann WK, Sancar A. Coupling of human circadian and cell cycles by the timeless protein. Mol Cell Biol. 2005;25:3109–16. [PMC free article] [PubMed]
27. Cordeiro-Stone M, Boyer JC, Smith BA, Kaufmann WK. Effect of benzo[a]pyrene-diol-epoxide-I on growth of nascent DNA in synchronized human fibroblasts. Carcinogenesis. 1986;7:1775–81. [PubMed]
28. Chou JW, Paules RS, Bushel PR. Systematic variation normalization in microarray data to get gene expression comparison unbiased. Journal of Bioinformatics and Computational Biology. 2005;3:225–41. [PubMed]
29. Elkon R, Linhart C, Sharan R, Shamir R, Shiloh Y. Genome-wide in silico identification of transcriptional regulators controlling the cell cycle in human cells. Genome Res. 2003;13:773–80. [PubMed]
30. Matys V, Fricke E, Geffers R, Gossling E, Haubrock M, Hehl R, Hornischer K, Karas D, Kel AE, Kel-Margoulis OV, Kloos DU, Land S, Lewicki-Potapov B, Michael H, Munch R, Reuter I, Rotert S, Saxel H, Scheer M, Thiele S, Wingender E. TRANSFAC: Transcriptional regulation, from patterns to profiles. Nucleic Acids Research. 2003;31:374–8. [PMC free article] [PubMed]
31. Birney E, Andrews TD, Bevan P, Caccamo M, Chen Y, Clarke L, Coates G, Cuff J, Curwen V, Cutts T, Down T, Eyras E, Fernandez-Suarez XM, Gane P, Gibbins B, Gilbert J, Hammond M, Hotz HR, Iyer V, Jekosch K, Kahari A, Kasprzyk A, Keefe D, Keenan S, Lehvaslaiho H, McVicker G, Melsopp C, Meidl P, Mongin E, Pettett R, Potter S, Proctor G, Rae M, Searle S, Slater G, Smedley D, Smith J, Spooner W, Stabenau A, Stalker J, Storey R, Ureta-Vidal A, Woodwark KC, Cameron G, Durbin R, Cox A, Hubbard T, Clamp M. An overview of Ensembl. Genome Res. 2004;14:925–8. [PubMed]
32. Russell L, Forsdyke DR. A human putative lymphocyte G0/G1 switch gene containing a CpG-rich island encodes a small basic protein with the potential to be phosphorylated. DNA Cell Biol. 1991;10:581–91. [PubMed]
33. Agarwal ML, Agarwal A, Taylor WR, Stark GR. p53 controls both the G2/M and the G1 cell cycle checkpoints and mediates reversible growth arrest in human fibroblasts. Proc Natl Acad Sci USA. 1995;92:8493–7. [PubMed]
34. Iliakis G, Wang Y, Guan J, Wang H. DNA damage checkpoint control in cells exposed to ionizing radiation. Oncogene. 2003;22:5834–47. [PubMed]
35. Burns TF, El-Deiry WS. Microarray analysis of p53 target gene expression patterns in the spleen and thymus in response to ionizing radiation. Cancer Biol Ther. 2003;2:431–43. [PubMed]
36. Amundson SA, Bittner M, Meltzer P, Trent J, Fornace AJ., Jr Induction of gene expression as a monitor of exposure to ionizing radiation. Radiat Res. 2001;156:657–61. [PubMed]
37. Sax JK, El-Deiry WS. p53 downstream targets and chemosensitivity. Cell Death Differ. 2003;10:413–7. [PubMed]
38. Mirza A, Wu Q, Wang L, McClanahan T, Bishop WR, Gheyas F, Ding W, Hutchins B, Hockenberry T, Kirschmeier P, Greene JR, Liu S. Global transcriptional program of p53 target genes during the process of apoptosis and cell cycle progression. Oncogene. 2003;22:3645–54. [PubMed]
39. Taylor WR, DePrimo SE, Agarwal A, Agarwal ML, Schonthal AH, Katula KS, Stark GR. Mechanisms of G2 arrest in response to overexpression of p53. Mol Biol Cell. 1999;10:3607–22. [PMC free article] [PubMed]
40. Taylor WR, Stark GR. Regulation of the G2/M transition by p53. Oncogene. 2001;20:1803–15. [PubMed]
41. de Toledo SM, Azzam EI, Keng P, Laffrenier S, Little JB. Regulation by ionizing radiation of CDC2, cyclin A, cyclin B, thymidine kinase, topoisomerase IIalpha, and RAD51 expression in normal human diploid fibroblasts is dependent on p53/p21Waf1. Cell Growth Differ. 1998;9:887–96. [PubMed]
42. Manni I, Mazzaro G, Gurtner A, Mantovani R, Haugwitz U, Krause K, Engeland K, Sacchi A, Soddu S, Piaggio G. NF-Y mediates the transcriptional inhibition of the cyclin B1, cyclin B2, and cdc25C promoters upon induced G2 arrest. J Biol Chem. 2001;276:5570–6. [PubMed]
43. Innocente SA, Lee JM. p53 is a NF-Y- and p21-independent, Sp1-dependent repressor of cyclin B1 transcription. FEBS Lett. 2005;579:1001–7. [PubMed]
44. St Clair S, Giono L, Varmeh-Ziaie S, Resnick-Silverman L, Liu WJ, Padi A, Dastidar J, DaCosta A, Mattia M, Manfredi JJ. DNA damage-induced downregulation of Cdc25C is mediated by p53 via two independent mechanisms: One involves direct binding to the cdc25C promoter. Mol Cell. 2004;16:725–36. [PubMed]
45. Badie C, Itzhaki JE, Sullivan MJ, Carpenter AJ, Porter AC. Repression of CDK1 and other genes with CDE and CHR promoter elements during DNA damage-induced G2/M arrest in human cells. Mol Cell Biol. 2000;20:2358–66. [PMC free article] [PubMed]
46. Lange-zu Dohna C, Brandeis M, Berr F, Mossner J, Engeland K. A CDE/CHR tandem element regulates cell cycle-dependent repression of cyclin B2 transcription. FEBS Lett. 2000;484:77–81. [PubMed]