Limits...
Homeostatic maintenance of allele-specific p16 methylation in cancer cells accompanied by dynamic focal methylation and hydroxymethylation.

Qin S, Li Q, Zhou J, Liu ZJ, Su N, Wilson J, Lu ZM, Deng D - PLoS ONE (2014)

Bottom Line: P16 expression status was determined using immuno-staining and RT-PCR.Furthermore, subclones of the monoclones consistently maintained the same p16 methylation pattern.Interestingly, transcription was not observed in p16 alleles that were hydroxymethylated with an antisense-strand-specific pattern.

View Article: PubMed Central - PubMed

Affiliation: Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education), Division of Cancer Etiology, Peking University Cancer Hospital & Institute, Beijing, China.

ABSTRACT

Aim: p16 Methylation frequently occurs in carcinogenesis. While it has been hypothesized that the p16 methylation states are dynamically maintained in cancer cells, direct evidence supporting this hypothesis has not been available until now.

Methods: A fusion cell model was established which reprogrammed the native DNA methylation pattern of the cells. The methylation status of the p16 alleles was then repeatedly quantitatively analyzed in the fusion monoclonal, parental cancer cell lines (p16-completely methylated-AGS and unmethylated-MGC803), and HCT116 non-fusion cell using DHPLC and bisulfite sequencing. Histone methylation was analyzed using chromatin immuno-precipitation (ChIP)-PCR. P16 expression status was determined using immuno-staining and RT-PCR.

Results: The methylation status for the majority of the p16 alleles was stably maintained in the fusion monoclonal cells after up to 60 passages. Most importantly, focal de novo methylation, demethylation, and hydroxymethylation were consistently observed within about 27% of the p16 alleles in the fusion monoclones, but not the homozygously methylated or unmethylated parental cells. Furthermore, subclones of the monoclones consistently maintained the same p16 methylation pattern. A similar phenomenon was also observed using the p16 hemi-methylated HCT116 non-fusion cancer cell line. Interestingly, transcription was not observed in p16 alleles that were hydroxymethylated with an antisense-strand-specific pattern. Also, the levels of H3K9 and H3K4 trimethylation in the fusion cells were found to be slightly lower than the parental AGS and MGC803 cells, respectively.

Conclusion: The present study provides the first direct evidence confirming that the methylation states of p16 CpG islands is not only homeostatically maintained, but also accompanied by a dynamic process of transient focal methylation, demethylation, and hydroxymethylation in cancer cells.

Show MeSH

Related in: MedlinePlus

Characterization of histone modifications in the p16 chromatin by ChIP-PCR assays.(A) The active H3K4me3 level within the p16 CpG islands in the fusion subclones and MGC803 cells was significantly higher than AGS cells, especially in the exon-1 region (monoclones vs. AGS, P<0.01). (B) The repressive H3K9me3 level in the fusion subclones and AGS cells was significantly higher than MGC803 cells, especially in the promoter region (monoclones vs. MGC803, P<0.01).
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4020935&req=5

pone-0097785-g007: Characterization of histone modifications in the p16 chromatin by ChIP-PCR assays.(A) The active H3K4me3 level within the p16 CpG islands in the fusion subclones and MGC803 cells was significantly higher than AGS cells, especially in the exon-1 region (monoclones vs. AGS, P<0.01). (B) The repressive H3K9me3 level in the fusion subclones and AGS cells was significantly higher than MGC803 cells, especially in the promoter region (monoclones vs. MGC803, P<0.01).

Mentions: ChIP-PCR was used to quantitatively determine the levels of active H3K4me3 and repressive H3K9me3 in the fusion cells (Figure 7). The H3K4me3 level in the heterogeneously methylated p16 alleles of the fusion subclones was found to be significantly lower than was seen in the completely unmethylated alleles of the MGC803 cells. Furthermore, the level of H3K4me3 in exon-1 chromatin was determined to be higher than was found in promoter chromatin. While the H3K4me3 was not detected in the AGS cells, the H3K9me3 level was significantly higher in this cell line than other cell lines. H3K9me3 level showed an inverse correlation with the H3K4me3 level among these cell lines.


Homeostatic maintenance of allele-specific p16 methylation in cancer cells accompanied by dynamic focal methylation and hydroxymethylation.

Qin S, Li Q, Zhou J, Liu ZJ, Su N, Wilson J, Lu ZM, Deng D - PLoS ONE (2014)

Characterization of histone modifications in the p16 chromatin by ChIP-PCR assays.(A) The active H3K4me3 level within the p16 CpG islands in the fusion subclones and MGC803 cells was significantly higher than AGS cells, especially in the exon-1 region (monoclones vs. AGS, P<0.01). (B) The repressive H3K9me3 level in the fusion subclones and AGS cells was significantly higher than MGC803 cells, especially in the promoter region (monoclones vs. MGC803, P<0.01).
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4020935&req=5

pone-0097785-g007: Characterization of histone modifications in the p16 chromatin by ChIP-PCR assays.(A) The active H3K4me3 level within the p16 CpG islands in the fusion subclones and MGC803 cells was significantly higher than AGS cells, especially in the exon-1 region (monoclones vs. AGS, P<0.01). (B) The repressive H3K9me3 level in the fusion subclones and AGS cells was significantly higher than MGC803 cells, especially in the promoter region (monoclones vs. MGC803, P<0.01).
Mentions: ChIP-PCR was used to quantitatively determine the levels of active H3K4me3 and repressive H3K9me3 in the fusion cells (Figure 7). The H3K4me3 level in the heterogeneously methylated p16 alleles of the fusion subclones was found to be significantly lower than was seen in the completely unmethylated alleles of the MGC803 cells. Furthermore, the level of H3K4me3 in exon-1 chromatin was determined to be higher than was found in promoter chromatin. While the H3K4me3 was not detected in the AGS cells, the H3K9me3 level was significantly higher in this cell line than other cell lines. H3K9me3 level showed an inverse correlation with the H3K4me3 level among these cell lines.

Bottom Line: P16 expression status was determined using immuno-staining and RT-PCR.Furthermore, subclones of the monoclones consistently maintained the same p16 methylation pattern.Interestingly, transcription was not observed in p16 alleles that were hydroxymethylated with an antisense-strand-specific pattern.

View Article: PubMed Central - PubMed

Affiliation: Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education), Division of Cancer Etiology, Peking University Cancer Hospital & Institute, Beijing, China.

ABSTRACT

Aim: p16 Methylation frequently occurs in carcinogenesis. While it has been hypothesized that the p16 methylation states are dynamically maintained in cancer cells, direct evidence supporting this hypothesis has not been available until now.

Methods: A fusion cell model was established which reprogrammed the native DNA methylation pattern of the cells. The methylation status of the p16 alleles was then repeatedly quantitatively analyzed in the fusion monoclonal, parental cancer cell lines (p16-completely methylated-AGS and unmethylated-MGC803), and HCT116 non-fusion cell using DHPLC and bisulfite sequencing. Histone methylation was analyzed using chromatin immuno-precipitation (ChIP)-PCR. P16 expression status was determined using immuno-staining and RT-PCR.

Results: The methylation status for the majority of the p16 alleles was stably maintained in the fusion monoclonal cells after up to 60 passages. Most importantly, focal de novo methylation, demethylation, and hydroxymethylation were consistently observed within about 27% of the p16 alleles in the fusion monoclones, but not the homozygously methylated or unmethylated parental cells. Furthermore, subclones of the monoclones consistently maintained the same p16 methylation pattern. A similar phenomenon was also observed using the p16 hemi-methylated HCT116 non-fusion cancer cell line. Interestingly, transcription was not observed in p16 alleles that were hydroxymethylated with an antisense-strand-specific pattern. Also, the levels of H3K9 and H3K4 trimethylation in the fusion cells were found to be slightly lower than the parental AGS and MGC803 cells, respectively.

Conclusion: The present study provides the first direct evidence confirming that the methylation states of p16 CpG islands is not only homeostatically maintained, but also accompanied by a dynamic process of transient focal methylation, demethylation, and hydroxymethylation in cancer cells.

Show MeSH
Related in: MedlinePlus