Limits...
Discovery pipeline for epigenetically deregulated miRNAs in cancer: integration of primary miRNA transcription.

Hulf T, Sibbritt T, Wiklund ED, Bert S, Strbenac D, Statham AL, Robinson MD, Clark SJ - BMC Genomics (2011)

Bottom Line: Cancer is commonly associated with widespread disruption of DNA methylation, chromatin modification and miRNA expression.We find miR-205, miR-21, and miR-196b to be epigenetically repressed, and miR-615 epigenetically activated in prostate cancer cells.We show that detecting changes in primary miRNA transcription levels is a valuable method for detection of local epigenetic modifications that are associated with changes in mature miRNA expression.

View Article: PubMed Central - HTML - PubMed

Affiliation: Epigenetics Laboratory, Cancer Research Program, Garvan Institute of Medical Research, 384 Victoria Street, Darlinghurst, NSW 2010, Australia.

ABSTRACT

Background: Cancer is commonly associated with widespread disruption of DNA methylation, chromatin modification and miRNA expression. In this study, we established a robust discovery pipeline to identify epigenetically deregulated miRNAs in cancer.

Results: Using an integrative approach that combines primary transcription, genome-wide DNA methylation and H3K9Ac marks with microRNA (miRNA) expression, we identified miRNA genes that were epigenetically modified in cancer. We find miR-205, miR-21, and miR-196b to be epigenetically repressed, and miR-615 epigenetically activated in prostate cancer cells.

Conclusions: We show that detecting changes in primary miRNA transcription levels is a valuable method for detection of local epigenetic modifications that are associated with changes in mature miRNA expression.

Show MeSH

Related in: MedlinePlus

Progressive steps in the identification of epigenetically regulated miRNAs in prostate cells. Venn diagrams show the number of miRNA loci changed in LNCaP (L) with respect to PrEC (P) cells called significant by tiling arrays analysis using a t-statistic ≤-2 or ≥2 for regions +/-500 bp from the relevant miRNA. (A, B) Overlap of pri- and mature miRNA expression levels identified by tiling array and TLDA, respectively (TLDA miRNA expression data was normalized to RNU48, using a significance cutoff at p < 0.05 and ≥1.5 -fold change). Concomitant changes in active (H3K9Ac) or repressive (DNA methylation) epigenetic marks were then considered for miRNAs displaying significantly changed expression. (C, D) miRNA loci significantly changed in MeDIP-chip and H3K9Ac ChIP-chip tiling array analysis. (E, F) Four-way Venn diagrams showing the overlap of tiling array data with mature miRNA level upon 5-Aza-CdR treatment. MiR-615 lies at the four-way intersection indicating epigenetic up-regulation and miR-205, miR-196b and miR-21 are at the intersection of epigenetic down-regulation in LNCaP cells.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3037319&req=5

Figure 2: Progressive steps in the identification of epigenetically regulated miRNAs in prostate cells. Venn diagrams show the number of miRNA loci changed in LNCaP (L) with respect to PrEC (P) cells called significant by tiling arrays analysis using a t-statistic ≤-2 or ≥2 for regions +/-500 bp from the relevant miRNA. (A, B) Overlap of pri- and mature miRNA expression levels identified by tiling array and TLDA, respectively (TLDA miRNA expression data was normalized to RNU48, using a significance cutoff at p < 0.05 and ≥1.5 -fold change). Concomitant changes in active (H3K9Ac) or repressive (DNA methylation) epigenetic marks were then considered for miRNAs displaying significantly changed expression. (C, D) miRNA loci significantly changed in MeDIP-chip and H3K9Ac ChIP-chip tiling array analysis. (E, F) Four-way Venn diagrams showing the overlap of tiling array data with mature miRNA level upon 5-Aza-CdR treatment. MiR-615 lies at the four-way intersection indicating epigenetic up-regulation and miR-205, miR-196b and miR-21 are at the intersection of epigenetic down-regulation in LNCaP cells.

Mentions: Looking at those loci whose primary transcript repression or activation in LNCaP was associated with a significant change in the mature miRNA, we identified 29 upregulated and 33 downregulated miRNAs (Figure 2A,B). We then looked for concomitant epigenetic changes at these loci in the tiling array data. A high proportion of the upregulated (31%; 9/29) and downregulated loci (33%; 11/33) showed an overlap with at least one of these epigenetic marks (Figure 2C,D). The majority (18/20) of these miRNAs have previously been identified as deregulated in various cancers, and five with prior evidence for epigenetic regulation (Table 1). A concordant change in all assays; expression, DNA methylation, and H3K9Ac restricted the total number of candidate miRNAs to 2/29 up- and 4/33 downregulated (Figure 2C,D). Given the large number of possible epigenetic modifications (for a recent review see [6]), it is possible that many of the differentially expressed loci that show no change in DNA methylation or H3K9Ac could undergo changes in other epigenetic marks. Furthermore, miRNAs may be regulated by DNA methylation at CpG islands outside the 2000 bp regions interrogated in this analysis.


Discovery pipeline for epigenetically deregulated miRNAs in cancer: integration of primary miRNA transcription.

Hulf T, Sibbritt T, Wiklund ED, Bert S, Strbenac D, Statham AL, Robinson MD, Clark SJ - BMC Genomics (2011)

Progressive steps in the identification of epigenetically regulated miRNAs in prostate cells. Venn diagrams show the number of miRNA loci changed in LNCaP (L) with respect to PrEC (P) cells called significant by tiling arrays analysis using a t-statistic ≤-2 or ≥2 for regions +/-500 bp from the relevant miRNA. (A, B) Overlap of pri- and mature miRNA expression levels identified by tiling array and TLDA, respectively (TLDA miRNA expression data was normalized to RNU48, using a significance cutoff at p < 0.05 and ≥1.5 -fold change). Concomitant changes in active (H3K9Ac) or repressive (DNA methylation) epigenetic marks were then considered for miRNAs displaying significantly changed expression. (C, D) miRNA loci significantly changed in MeDIP-chip and H3K9Ac ChIP-chip tiling array analysis. (E, F) Four-way Venn diagrams showing the overlap of tiling array data with mature miRNA level upon 5-Aza-CdR treatment. MiR-615 lies at the four-way intersection indicating epigenetic up-regulation and miR-205, miR-196b and miR-21 are at the intersection of epigenetic down-regulation in LNCaP cells.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3037319&req=5

Figure 2: Progressive steps in the identification of epigenetically regulated miRNAs in prostate cells. Venn diagrams show the number of miRNA loci changed in LNCaP (L) with respect to PrEC (P) cells called significant by tiling arrays analysis using a t-statistic ≤-2 or ≥2 for regions +/-500 bp from the relevant miRNA. (A, B) Overlap of pri- and mature miRNA expression levels identified by tiling array and TLDA, respectively (TLDA miRNA expression data was normalized to RNU48, using a significance cutoff at p < 0.05 and ≥1.5 -fold change). Concomitant changes in active (H3K9Ac) or repressive (DNA methylation) epigenetic marks were then considered for miRNAs displaying significantly changed expression. (C, D) miRNA loci significantly changed in MeDIP-chip and H3K9Ac ChIP-chip tiling array analysis. (E, F) Four-way Venn diagrams showing the overlap of tiling array data with mature miRNA level upon 5-Aza-CdR treatment. MiR-615 lies at the four-way intersection indicating epigenetic up-regulation and miR-205, miR-196b and miR-21 are at the intersection of epigenetic down-regulation in LNCaP cells.
Mentions: Looking at those loci whose primary transcript repression or activation in LNCaP was associated with a significant change in the mature miRNA, we identified 29 upregulated and 33 downregulated miRNAs (Figure 2A,B). We then looked for concomitant epigenetic changes at these loci in the tiling array data. A high proportion of the upregulated (31%; 9/29) and downregulated loci (33%; 11/33) showed an overlap with at least one of these epigenetic marks (Figure 2C,D). The majority (18/20) of these miRNAs have previously been identified as deregulated in various cancers, and five with prior evidence for epigenetic regulation (Table 1). A concordant change in all assays; expression, DNA methylation, and H3K9Ac restricted the total number of candidate miRNAs to 2/29 up- and 4/33 downregulated (Figure 2C,D). Given the large number of possible epigenetic modifications (for a recent review see [6]), it is possible that many of the differentially expressed loci that show no change in DNA methylation or H3K9Ac could undergo changes in other epigenetic marks. Furthermore, miRNAs may be regulated by DNA methylation at CpG islands outside the 2000 bp regions interrogated in this analysis.

Bottom Line: Cancer is commonly associated with widespread disruption of DNA methylation, chromatin modification and miRNA expression.We find miR-205, miR-21, and miR-196b to be epigenetically repressed, and miR-615 epigenetically activated in prostate cancer cells.We show that detecting changes in primary miRNA transcription levels is a valuable method for detection of local epigenetic modifications that are associated with changes in mature miRNA expression.

View Article: PubMed Central - HTML - PubMed

Affiliation: Epigenetics Laboratory, Cancer Research Program, Garvan Institute of Medical Research, 384 Victoria Street, Darlinghurst, NSW 2010, Australia.

ABSTRACT

Background: Cancer is commonly associated with widespread disruption of DNA methylation, chromatin modification and miRNA expression. In this study, we established a robust discovery pipeline to identify epigenetically deregulated miRNAs in cancer.

Results: Using an integrative approach that combines primary transcription, genome-wide DNA methylation and H3K9Ac marks with microRNA (miRNA) expression, we identified miRNA genes that were epigenetically modified in cancer. We find miR-205, miR-21, and miR-196b to be epigenetically repressed, and miR-615 epigenetically activated in prostate cancer cells.

Conclusions: We show that detecting changes in primary miRNA transcription levels is a valuable method for detection of local epigenetic modifications that are associated with changes in mature miRNA expression.

Show MeSH
Related in: MedlinePlus