dentified Sp1, NF1 and E2F Tideglusib site response elements in the promoter of the DDB2 gene, and showed that mutations of these response elements reduced strongly the basal transcription of the DDB2 gene. In addition, it has been found that p53 and BRCA1 were able to activate the DDB2 gene. In the present study, we observed that the DDB2 gene was upregulated in ER-positive breast cancer cells, compared to the very low expression of DDB2 in the nontumorigenic epithelial mammary HMEC cell line. The mechanism by which DDB2 expression is dysregulated in ER-breast cancer cells is not known. Moreover, the molecular mechanism involved in the loss of DDB2 gene expression in ER-negative breast cancer cells will 11904527 need to be defined in the future. One hypothesis would suggest the involvement of BRCA1. The ER-positive breast cancer cells, such as MCF-7 and T47D cells, express BRCA1, whereas the ERnegative tumor cells, such as SKBR3 and MDA-MB231 cells are BRCA1 negative. Involvement of ER and other transcription factors or other molecular 19380825 mechanisms are not excluded and future investigations will need to elucidate the regulation of DDB2 expression during breast tumor progression. The surprising evidence that the high DDB2 content correlated with the high proliferation rate of MCF-7 and T47D cells compared to MDA-MB231 and SKBR3 cells, along with a number of studies reporting a role of DDB2 in the cell cycle regulation of normal cells, led us to investigate the role of this protein in tumor growth. The result of the inhibition of DDB2 the DDB2 deficient MCF-7 cells 3 h after the addition of serum. Similar to the finding by cell cycle analysis after PI staining, no 5 BrdU incorporation was quantified for both DDB2-deficient MCF-7 cell lines, whereas the LI for the Wt and siRNA control MCF-7 cells revealed important Sphase fractions in these lines. Then, % 5 BrdU-positive cells corresponding to the LI for DDB2 deficient cells was strongly increased and was similar to the control MCF-7 cells at 12 and 18 h after release from serum depletion. Compared to that of the control cells, this LI indicated an important pool of DDB2deficient cells which started to re-enter the cell cycle and which corresponded to an essentially G1/S subpopulation. The DDB2deficient MCF-7 cell clones 2 and 3 showed S-phase fractions respectively 6.8- and 4.2-fold less than that of the control MCF-7 cells, at 12 and 18 h after release from serum depletion. In addition, no G2 fraction was detected for both DDB2-deficient MCF-7 cell clones. These results demonstrate that DDB2 knockdown led to a delayed G1/S transition phase entry and a slowed MCF-7 cell progression through the S phase. These results were confirmed by an investigation of the PCNA protein level. Regardless of the time from release of serum depletion, the PCNA protein level was greatly reduced in both DDB2-deficient MCF-7 cell lines, compared to that of tubulin, used as a loading control. DDB2 and Breast Tumor Growth independent experiments were expressed as the % of colony formation = 6100%. Statistically significant differences from the parental cell value are indicated as P,0.05. doi:10.1371/journal.pone.0002002.g003 expression, through the strategy of small interfering RNA, gave a significant reduction of the growth rate and clonogenicity of 
the MCF-7 cells and an increased cell doubling time. Inversely, introduction of the DDB2 gene into MDA-MB231 cells increased their growth rate and clonogenicity and decreased their cell doubl