Ssion of the reporter gene (Figure 5B), and confirmed that Kaiso was negatively regulating the minimal cyclin D1 promoter. Importantly, since the cyclin D1 promoter-reporter plasmid was propagated in dam2/dcm2 bacteria, the CpG sites were unmethylated. Thus it appears that Kaiso-mediated transcriptional repression of the cyclinD1 promoter-reporter was occurring via the sequence-specific KBS sites and not the CpG sites.Kaiso Represses Transcription from the Minimal cyclin D1 Promoter in a Methyl-CpG-specific MannerWe next examined how a change in the methylation status of the promoter may affect Kaiso’s ability to regulate expression of the minimal cyclin D1 promoter-reporter. Thus, the 21748 CD1 promoter-reporter construct was in vitro methylated using Sss1 methyltransferase prior to transfection. CpG methylation of the plasmid was confirmed by restriction digest with the methylationresistant enzyme HpaII (Figure 6A). Transfection of the unmethylated 21748CD1 wild-type promoter-reporter construct resulted in more than 25-fold increase in luciferase 301353-96-8 site activity compared to the control pGluc-Basic vector, while its methylated counterpart only exhibited an ,3.5-fold increase (Figure 6B). This is consistent with the MedChemExpress Licochalcone-A notion that methylation of promoter regions is involved in gene silencing. However, when the methylated or unmethylated 21748 CD1 promoter-reporters were individually co-transfected with Kaiso, a similar two-fold decrease in luciferase activity was observed for both promoters tested (Figure 6B, compare 2.0 mg Kaiso for each reporter). This data suggests that Kaiso’s ability to repress the cyclin D1 promoter is via at least three distinct mechanisms: (1) via binding to the KBS, (2) via binding to methylated CpG sites, or (3) via combinatorial use of both KBS and CpG sites. To further delineate Kaiso’s mechanism(s) of transcriptional repression of the cyclin D1 promoter-reporter, we mutated the KBSs and assessed luciferase activity from the unmethylated and methylated mutant reporters. The methylated but KBS mutated promoter-reporter (Met+KBSmut) exhibited a dose-dependent decrease in luciferase activity upon ectopic Kaiso expression (Figure 6C) while its unmethylated and KBS mutated counterpart (Met-KBSmut) remained relatively unchanged (Figure 6D). Together our data suggests that Kaiso regulates cyclin D1 via its dual-specificity DNA-binding.Figure 5. Kaiso represses expression of a minimal cyclin D1 promoter-reporter. (A) Kaiso overexpression decreased luciferase expression from the minimal cyclin D1 promoter-reporter in a dosedependent manner in MCF7 cells. (B) Depletion of endogenous Kaiso caused de-repression of the minimal cyclin D1 promoter-reporter in MCF7 cells. doi:10.1371/journal.pone.0050398.gblot analysis of Kaiso-depleted HCT 116 colon carcinoma cell lysates. Similar to Jiang et al. (2012), we found that Kaiso depletion resulted in increased cyclin D1 protein levels (Figure 7A). Conversely, transient overexpression of Kaiso in MCF7 cells resulted in decreased cyclinD1 protein levels (Figure S3). More importantly, the Kaiso-depleted cells displayed an ,2-fold increase in cell proliferation compared to the parental and control vector HCT 116 cells in three independent trials (Figure 7B). The increased cell proliferation observed in the HCT 116 Kaisodepleted cells strengthens our hypothesis that cyclin D1 is a bona fide Kaiso target gene.Kaiso Depletion Increases HCT116 Cell Proliferation and cyclinD1 Protein LevelsAs a f.Ssion of the reporter gene (Figure 5B), and confirmed that Kaiso was negatively regulating the minimal cyclin D1 promoter. Importantly, since the cyclin D1 promoter-reporter plasmid was propagated in dam2/dcm2 bacteria, the CpG sites were unmethylated. Thus it appears that Kaiso-mediated transcriptional repression of the cyclinD1 promoter-reporter was occurring via the sequence-specific KBS sites and not the CpG sites.Kaiso Represses Transcription from the Minimal cyclin D1 Promoter in a Methyl-CpG-specific MannerWe next examined how a change in the methylation status of the promoter may affect Kaiso’s ability to regulate expression of the minimal cyclin D1 promoter-reporter. Thus, the 21748 CD1 promoter-reporter construct was in vitro methylated using Sss1 methyltransferase prior to transfection. CpG methylation of the plasmid was confirmed by restriction digest with the methylationresistant enzyme HpaII (Figure 6A). Transfection of the unmethylated 21748CD1 wild-type promoter-reporter construct resulted in more than 25-fold increase in luciferase activity compared to the control pGluc-Basic vector, while its methylated counterpart only exhibited an ,3.5-fold increase (Figure 6B). This is consistent with the notion that methylation of promoter regions is involved in gene silencing. However, when the methylated or unmethylated 21748 CD1 promoter-reporters were individually co-transfected with Kaiso, a similar two-fold decrease in luciferase activity was observed for both promoters tested (Figure 6B, compare 2.0 mg Kaiso for each reporter). This data suggests that Kaiso’s ability to repress the cyclin D1 promoter is via at least three distinct mechanisms: (1) via binding to the KBS, (2) via binding to methylated CpG sites, or (3) via combinatorial use of both KBS and CpG sites. To further delineate Kaiso’s mechanism(s) of transcriptional repression of the cyclin D1 promoter-reporter, we mutated the KBSs and assessed luciferase activity from the unmethylated and methylated mutant reporters. The methylated but KBS mutated promoter-reporter (Met+KBSmut) exhibited a dose-dependent decrease in luciferase activity upon ectopic Kaiso expression (Figure 6C) while its unmethylated and KBS mutated counterpart (Met-KBSmut) remained relatively unchanged (Figure 6D). Together our data suggests that Kaiso regulates cyclin D1 via its dual-specificity DNA-binding.Figure 5. Kaiso represses expression of a minimal cyclin D1 promoter-reporter. (A) Kaiso overexpression decreased luciferase expression from the minimal cyclin D1 promoter-reporter in a dosedependent manner in MCF7 cells. (B) Depletion of endogenous Kaiso caused de-repression of the minimal cyclin D1 promoter-reporter in MCF7 cells. doi:10.1371/journal.pone.0050398.gblot analysis of Kaiso-depleted HCT 116 colon carcinoma cell lysates. Similar to Jiang et al. (2012), we found that Kaiso depletion resulted in increased cyclin D1 protein levels (Figure 7A). Conversely, transient overexpression of Kaiso in MCF7 cells resulted in decreased cyclinD1 protein levels (Figure S3). More importantly, the Kaiso-depleted cells displayed an ,2-fold increase in cell proliferation compared to the parental and control vector HCT 116 cells in three independent trials (Figure 7B). The increased cell proliferation observed in the HCT 116 Kaisodepleted cells strengthens our hypothesis that cyclin D1 is a bona fide Kaiso target gene.Kaiso Depletion Increases HCT116 Cell Proliferation and cyclinD1 Protein LevelsAs a f.