Limits...
Identification of candidate methylation-responsive genes in ovarian cancer.

Menendez L, Walker D, Matyunina LV, Dickerson EB, Bowen NJ, Polavarapu N, Benigno BB, McDonald JF - Mol. Cancer (2007)

Bottom Line: An overlapping subset of these genes was found to display significant differences in gene expression between normal ovarian surface epithelial cells and malignant cells isolated from ovarian carcinomas.While 40% of all human genes are associated with CGIs, > 94% of the overlapping subset of genes is associated with CGIs.The predicted change in methylation status of genes randomly selected from the overlapping subset was experimentally verified.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Genetics, University of Georgia, Athens, GA 30605, USA. lmenen@uga.edu

ABSTRACT

Background: Aberrant methylation of gene promoter regions has been linked to changes in gene expression in cancer development and progression. Genes associated with CpG islands (CGIs) are especially prone to methylation, but not all CGI-associated genes display changes in methylation patterns in cancers.

Results: In order to identify genes subject to regulation by methylation, we conducted gene expression profile analyses of an ovarian cancer cell line (OVCAR-3) before and after treatment with the demethylating agent 5-aza-deoxycytidine (5-aza-dC). An overlapping subset of these genes was found to display significant differences in gene expression between normal ovarian surface epithelial cells and malignant cells isolated from ovarian carcinomas. While 40% of all human genes are associated with CGIs, > 94% of the overlapping subset of genes is associated with CGIs. The predicted change in methylation status of genes randomly selected from the overlapping subset was experimentally verified.

Conclusion: We conclude that correlating genes that are upregulated in response to 5-aza-dC treatment of cancer cell lines with genes that are down-regulated in cancer cells may be a useful method to identify genes experiencing epigenetic-mediated changes in expression over cancer development.

Show MeSH

Related in: MedlinePlus

Methylation analysis of the EMP3, CAV1 and CXCL6 genes in ovarian cancer (EOC) and control (NOSE) tissues. Summary of sodium bisulfite sequencing analysis of CGIs associated with the EMP3, CAV1 and CXCL6 genes in ovarian ovarian cancer and control (NOSE) tissues. For EMP3, 16 clones for 4 cancer samples and 23 clones for 5 control (NOSE) samples were sequenced. For CAV 1, 17 clones for 4 cancer samples and 18 clones for 4 control (NOSE) samples were sequenced. For CXCL6, 19 clones for 4 cancer samples and 24 clones for 5 normal (NOSE) samples were sequenced. Each circle displays the percent methylation of all clones for a single CpG dinucleotide (open circle, 100% unmethylated; filled circle, 100% methylated). Average percent methylation for cancer (EOS) and control (NOSE) samples is shown next to each row of circles. (Vertical bars, CpG dinucleotides; arrow, position of the transcription start site; horizontal black rectangle, region amplified and sequenced after sodium bisulfite).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC1803786&req=5

Figure 3: Methylation analysis of the EMP3, CAV1 and CXCL6 genes in ovarian cancer (EOC) and control (NOSE) tissues. Summary of sodium bisulfite sequencing analysis of CGIs associated with the EMP3, CAV1 and CXCL6 genes in ovarian ovarian cancer and control (NOSE) tissues. For EMP3, 16 clones for 4 cancer samples and 23 clones for 5 control (NOSE) samples were sequenced. For CAV 1, 17 clones for 4 cancer samples and 18 clones for 4 control (NOSE) samples were sequenced. For CXCL6, 19 clones for 4 cancer samples and 24 clones for 5 normal (NOSE) samples were sequenced. Each circle displays the percent methylation of all clones for a single CpG dinucleotide (open circle, 100% unmethylated; filled circle, 100% methylated). Average percent methylation for cancer (EOS) and control (NOSE) samples is shown next to each row of circles. (Vertical bars, CpG dinucleotides; arrow, position of the transcription start site; horizontal black rectangle, region amplified and sequenced after sodium bisulfite).

Mentions: As an initial experimental test of the model, we randomly selected two genes from the list of candidate genes (CAV 1 and CXCL6) and one CGI-associated gene not on the list of candidate genes (EMP3) for sodium bisulfite sequencing analysis. Consistent with the prediction of the model, CAV-1 and CXCL6 were found to be hypermethylated in ovarian cancer relative to control tissues while essentially no difference in EMP3 methylation pattern was detected. The observed fold changes in gene expression between control (NOSE) and ovarian cancer (EOC) samples were significant (p < 0.01) for CAV 1 (-30.88) and CXCL6 (-4.22) [see Additional File 1] but not significant (p < 0.06) for EMP 3 (+1.10). Consistent with these relative changes in levels of gene expression, the relative increase in levels of methylation was greater for Cav 1 than CXCL6 and minimal for EMP3 (Figure 3).


Identification of candidate methylation-responsive genes in ovarian cancer.

Menendez L, Walker D, Matyunina LV, Dickerson EB, Bowen NJ, Polavarapu N, Benigno BB, McDonald JF - Mol. Cancer (2007)

Methylation analysis of the EMP3, CAV1 and CXCL6 genes in ovarian cancer (EOC) and control (NOSE) tissues. Summary of sodium bisulfite sequencing analysis of CGIs associated with the EMP3, CAV1 and CXCL6 genes in ovarian ovarian cancer and control (NOSE) tissues. For EMP3, 16 clones for 4 cancer samples and 23 clones for 5 control (NOSE) samples were sequenced. For CAV 1, 17 clones for 4 cancer samples and 18 clones for 4 control (NOSE) samples were sequenced. For CXCL6, 19 clones for 4 cancer samples and 24 clones for 5 normal (NOSE) samples were sequenced. Each circle displays the percent methylation of all clones for a single CpG dinucleotide (open circle, 100% unmethylated; filled circle, 100% methylated). Average percent methylation for cancer (EOS) and control (NOSE) samples is shown next to each row of circles. (Vertical bars, CpG dinucleotides; arrow, position of the transcription start site; horizontal black rectangle, region amplified and sequenced after sodium bisulfite).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC1803786&req=5

Figure 3: Methylation analysis of the EMP3, CAV1 and CXCL6 genes in ovarian cancer (EOC) and control (NOSE) tissues. Summary of sodium bisulfite sequencing analysis of CGIs associated with the EMP3, CAV1 and CXCL6 genes in ovarian ovarian cancer and control (NOSE) tissues. For EMP3, 16 clones for 4 cancer samples and 23 clones for 5 control (NOSE) samples were sequenced. For CAV 1, 17 clones for 4 cancer samples and 18 clones for 4 control (NOSE) samples were sequenced. For CXCL6, 19 clones for 4 cancer samples and 24 clones for 5 normal (NOSE) samples were sequenced. Each circle displays the percent methylation of all clones for a single CpG dinucleotide (open circle, 100% unmethylated; filled circle, 100% methylated). Average percent methylation for cancer (EOS) and control (NOSE) samples is shown next to each row of circles. (Vertical bars, CpG dinucleotides; arrow, position of the transcription start site; horizontal black rectangle, region amplified and sequenced after sodium bisulfite).
Mentions: As an initial experimental test of the model, we randomly selected two genes from the list of candidate genes (CAV 1 and CXCL6) and one CGI-associated gene not on the list of candidate genes (EMP3) for sodium bisulfite sequencing analysis. Consistent with the prediction of the model, CAV-1 and CXCL6 were found to be hypermethylated in ovarian cancer relative to control tissues while essentially no difference in EMP3 methylation pattern was detected. The observed fold changes in gene expression between control (NOSE) and ovarian cancer (EOC) samples were significant (p < 0.01) for CAV 1 (-30.88) and CXCL6 (-4.22) [see Additional File 1] but not significant (p < 0.06) for EMP 3 (+1.10). Consistent with these relative changes in levels of gene expression, the relative increase in levels of methylation was greater for Cav 1 than CXCL6 and minimal for EMP3 (Figure 3).

Bottom Line: An overlapping subset of these genes was found to display significant differences in gene expression between normal ovarian surface epithelial cells and malignant cells isolated from ovarian carcinomas.While 40% of all human genes are associated with CGIs, > 94% of the overlapping subset of genes is associated with CGIs.The predicted change in methylation status of genes randomly selected from the overlapping subset was experimentally verified.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Genetics, University of Georgia, Athens, GA 30605, USA. lmenen@uga.edu

ABSTRACT

Background: Aberrant methylation of gene promoter regions has been linked to changes in gene expression in cancer development and progression. Genes associated with CpG islands (CGIs) are especially prone to methylation, but not all CGI-associated genes display changes in methylation patterns in cancers.

Results: In order to identify genes subject to regulation by methylation, we conducted gene expression profile analyses of an ovarian cancer cell line (OVCAR-3) before and after treatment with the demethylating agent 5-aza-deoxycytidine (5-aza-dC). An overlapping subset of these genes was found to display significant differences in gene expression between normal ovarian surface epithelial cells and malignant cells isolated from ovarian carcinomas. While 40% of all human genes are associated with CGIs, > 94% of the overlapping subset of genes is associated with CGIs. The predicted change in methylation status of genes randomly selected from the overlapping subset was experimentally verified.

Conclusion: We conclude that correlating genes that are upregulated in response to 5-aza-dC treatment of cancer cell lines with genes that are down-regulated in cancer cells may be a useful method to identify genes experiencing epigenetic-mediated changes in expression over cancer development.

Show MeSH
Related in: MedlinePlus