Limits...
Hypoxia-sensitive epigenetic regulation of an antisense-oriented lncRNA controls WT1 expression in myeloid leukemia cells.

McCarty G, Loeb DM - PLoS ONE (2015)

Bottom Line: WT1 is a transcription factor expressed in hematopoietic stem cells and in most cases of myeloid leukemia.WT1 expression correlates with hypomethylation of the CpG island in Intron 1, and pharmacologic demethylation of this CpG island induces WT1 mRNA expression.These results reveal a novel model of hypoxia-mediated epigenetic gene regulation.

View Article: PubMed Central - PubMed

Affiliation: Department of Oncology, Division of Pediatric Oncology, Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University, Baltimore, MD, United States of America.

ABSTRACT
WT1 is a transcription factor expressed in hematopoietic stem cells and in most cases of myeloid leukemia. We investigated the roles of hypoxia and epigenetics in the regulation of WT1 expression in myeloid leukemia cells. WT1 expression correlates with hypomethylation of the CpG island in Intron 1, and pharmacologic demethylation of this CpG island induces WT1 mRNA expression. Hypoxia causes decreases in DNMT expression and activity and increased expression and activity of TET2 and TET3, resulting in demethylation of this CpG island and expression of WT1 mRNA. Demethylation of the CpG island, either from pharmacologic treatment or induction of hypoxia, results in transcription of an antisense-oriented lncRNA, and inhibiting lncRNA expression with shRNA blocks WT1 mRNA expression. These results reveal a novel model of hypoxia-mediated epigenetic gene regulation. In addition, this is the first report that TET2 and TET3, increasingly recognized as important epigenetic regulators of gene expression in stem cells and in cancer cells, can be regulated by hypoxia, providing a solid mechanistic link between hypoxia and epigenetic regulation of gene expression with important implications for the role of hypoxia in stem cell function.

No MeSH data available.


Related in: MedlinePlus

Hypoxia diminishes DNMT expression and activity and increases expression and activity of TET2 and TET3.(A) K562 and U937 cells were treated with or without 100 μM CoCl2 for 48 hours. Expression of DNMT1, DNMT3A and DNMT3B mRNA was evaluated by quantitative RT-PCR. Data are fold-change of CoCl2-treated cells compared with control. Error bar shows standard error of the mean of triplicate experiments. All changes were statistically significant, with p < 0.05. (B) U937, HL60, and K562 cells were treated overnight with or without 100 μM CoCl2. Protein lysates were assayed for DNMT1 activity. Error bars show standard error of the mean of triplicate experiments. Student’s t test was performed to test for statistical significance of the indicated comparisons, and all p values are < 0.0006. (C) U937 and K562 cells were treated for 48 hours with 100 μM CoCl2. Expression of TET1, TET2, and TET3 mRNA was evaluated by quantitative RT-PCR. Data shown are fold-change of CoCl2-treated cells compared with control. Error bars show standard error of the mean of triplicate experiments. Changes in U937 cells were statistically significant, with p < 0.05, but changes in K562 were not. (D) U937 and K562 cells were treated for 48 hours with 100 μM CoCl2. The percentage of unmodified cytosine, 5mC, and 5hmC in the WT1 locus were quantified as described. Data shown are the means of triplicate samples, and this experiment was repeated 3 times with similar results. No statistical analysis of the data presented in Fig. 2D was performed because the number of transformations applied to the raw data in order to calculate the presented results (see Materials and Methods) makes all standard parametric tests of significance meaningless. Nevertheless, one can calculate that the likelihood of the observed pattern (comparing control and CoCl2-treated samples in each cell line) occurring by chance in a single experiment is 1/26, or 1.5%, and the likelihood of seeing this pattern in 3 independent experiments is 0.0153, or 3.4 x 10-6. Each experiment in this figure was repeated at least 3 times with similar results.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4368825&req=5

pone.0119837.g003: Hypoxia diminishes DNMT expression and activity and increases expression and activity of TET2 and TET3.(A) K562 and U937 cells were treated with or without 100 μM CoCl2 for 48 hours. Expression of DNMT1, DNMT3A and DNMT3B mRNA was evaluated by quantitative RT-PCR. Data are fold-change of CoCl2-treated cells compared with control. Error bar shows standard error of the mean of triplicate experiments. All changes were statistically significant, with p < 0.05. (B) U937, HL60, and K562 cells were treated overnight with or without 100 μM CoCl2. Protein lysates were assayed for DNMT1 activity. Error bars show standard error of the mean of triplicate experiments. Student’s t test was performed to test for statistical significance of the indicated comparisons, and all p values are < 0.0006. (C) U937 and K562 cells were treated for 48 hours with 100 μM CoCl2. Expression of TET1, TET2, and TET3 mRNA was evaluated by quantitative RT-PCR. Data shown are fold-change of CoCl2-treated cells compared with control. Error bars show standard error of the mean of triplicate experiments. Changes in U937 cells were statistically significant, with p < 0.05, but changes in K562 were not. (D) U937 and K562 cells were treated for 48 hours with 100 μM CoCl2. The percentage of unmodified cytosine, 5mC, and 5hmC in the WT1 locus were quantified as described. Data shown are the means of triplicate samples, and this experiment was repeated 3 times with similar results. No statistical analysis of the data presented in Fig. 2D was performed because the number of transformations applied to the raw data in order to calculate the presented results (see Materials and Methods) makes all standard parametric tests of significance meaningless. Nevertheless, one can calculate that the likelihood of the observed pattern (comparing control and CoCl2-treated samples in each cell line) occurring by chance in a single experiment is 1/26, or 1.5%, and the likelihood of seeing this pattern in 3 independent experiments is 0.0153, or 3.4 x 10-6. Each experiment in this figure was repeated at least 3 times with similar results.

Mentions: The decrease in methylation of the Intron 1 CpG island caused by hypoxia could result from either active DNA demethylation, a reduction in activity of the DNMTs that maintain the methylation status, or both. Several DNMTs have been implicated in establishing and maintaining DNA methylation patterns. We used quantitative RT-PCR to determine whether CoCl2 treatment affects the expression of any of these enzymes. K562 and U937 cells were treated for 48 hours with or without 100 μM CoCl2. We found that this treatment causes a significant decrease in DNMT3A expression in K562 cells, as well as modest decreases in expression of DNMT1 and DNMT3B (Fig. 3A). In U937 cells, CoCl2 causes substantial decreases in the expression of all of these enzymes (Fig. 3A). Because DNMT1 has been implicated in the maintenance of established CpG island methylation patterns, we investigated the effect of CoCl2 on DNMT1 activity in these cell lines. Using an in vitro assay, we found that CoCl2 induces modest, reproducible decreases in DNMT1 activity in K562 and HL60 cells (treated k562 cells have 70.1% of control activity, and treated HL60 cells have 62% of control activity; Fig. 3B), whereas treated U937 cells have only 43% of control activity, and this is the line in which CoCl2 induces hypomethylation of the Intron 1 CpG island.


Hypoxia-sensitive epigenetic regulation of an antisense-oriented lncRNA controls WT1 expression in myeloid leukemia cells.

McCarty G, Loeb DM - PLoS ONE (2015)

Hypoxia diminishes DNMT expression and activity and increases expression and activity of TET2 and TET3.(A) K562 and U937 cells were treated with or without 100 μM CoCl2 for 48 hours. Expression of DNMT1, DNMT3A and DNMT3B mRNA was evaluated by quantitative RT-PCR. Data are fold-change of CoCl2-treated cells compared with control. Error bar shows standard error of the mean of triplicate experiments. All changes were statistically significant, with p < 0.05. (B) U937, HL60, and K562 cells were treated overnight with or without 100 μM CoCl2. Protein lysates were assayed for DNMT1 activity. Error bars show standard error of the mean of triplicate experiments. Student’s t test was performed to test for statistical significance of the indicated comparisons, and all p values are < 0.0006. (C) U937 and K562 cells were treated for 48 hours with 100 μM CoCl2. Expression of TET1, TET2, and TET3 mRNA was evaluated by quantitative RT-PCR. Data shown are fold-change of CoCl2-treated cells compared with control. Error bars show standard error of the mean of triplicate experiments. Changes in U937 cells were statistically significant, with p < 0.05, but changes in K562 were not. (D) U937 and K562 cells were treated for 48 hours with 100 μM CoCl2. The percentage of unmodified cytosine, 5mC, and 5hmC in the WT1 locus were quantified as described. Data shown are the means of triplicate samples, and this experiment was repeated 3 times with similar results. No statistical analysis of the data presented in Fig. 2D was performed because the number of transformations applied to the raw data in order to calculate the presented results (see Materials and Methods) makes all standard parametric tests of significance meaningless. Nevertheless, one can calculate that the likelihood of the observed pattern (comparing control and CoCl2-treated samples in each cell line) occurring by chance in a single experiment is 1/26, or 1.5%, and the likelihood of seeing this pattern in 3 independent experiments is 0.0153, or 3.4 x 10-6. Each experiment in this figure was repeated at least 3 times with similar results.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4368825&req=5

pone.0119837.g003: Hypoxia diminishes DNMT expression and activity and increases expression and activity of TET2 and TET3.(A) K562 and U937 cells were treated with or without 100 μM CoCl2 for 48 hours. Expression of DNMT1, DNMT3A and DNMT3B mRNA was evaluated by quantitative RT-PCR. Data are fold-change of CoCl2-treated cells compared with control. Error bar shows standard error of the mean of triplicate experiments. All changes were statistically significant, with p < 0.05. (B) U937, HL60, and K562 cells were treated overnight with or without 100 μM CoCl2. Protein lysates were assayed for DNMT1 activity. Error bars show standard error of the mean of triplicate experiments. Student’s t test was performed to test for statistical significance of the indicated comparisons, and all p values are < 0.0006. (C) U937 and K562 cells were treated for 48 hours with 100 μM CoCl2. Expression of TET1, TET2, and TET3 mRNA was evaluated by quantitative RT-PCR. Data shown are fold-change of CoCl2-treated cells compared with control. Error bars show standard error of the mean of triplicate experiments. Changes in U937 cells were statistically significant, with p < 0.05, but changes in K562 were not. (D) U937 and K562 cells were treated for 48 hours with 100 μM CoCl2. The percentage of unmodified cytosine, 5mC, and 5hmC in the WT1 locus were quantified as described. Data shown are the means of triplicate samples, and this experiment was repeated 3 times with similar results. No statistical analysis of the data presented in Fig. 2D was performed because the number of transformations applied to the raw data in order to calculate the presented results (see Materials and Methods) makes all standard parametric tests of significance meaningless. Nevertheless, one can calculate that the likelihood of the observed pattern (comparing control and CoCl2-treated samples in each cell line) occurring by chance in a single experiment is 1/26, or 1.5%, and the likelihood of seeing this pattern in 3 independent experiments is 0.0153, or 3.4 x 10-6. Each experiment in this figure was repeated at least 3 times with similar results.
Mentions: The decrease in methylation of the Intron 1 CpG island caused by hypoxia could result from either active DNA demethylation, a reduction in activity of the DNMTs that maintain the methylation status, or both. Several DNMTs have been implicated in establishing and maintaining DNA methylation patterns. We used quantitative RT-PCR to determine whether CoCl2 treatment affects the expression of any of these enzymes. K562 and U937 cells were treated for 48 hours with or without 100 μM CoCl2. We found that this treatment causes a significant decrease in DNMT3A expression in K562 cells, as well as modest decreases in expression of DNMT1 and DNMT3B (Fig. 3A). In U937 cells, CoCl2 causes substantial decreases in the expression of all of these enzymes (Fig. 3A). Because DNMT1 has been implicated in the maintenance of established CpG island methylation patterns, we investigated the effect of CoCl2 on DNMT1 activity in these cell lines. Using an in vitro assay, we found that CoCl2 induces modest, reproducible decreases in DNMT1 activity in K562 and HL60 cells (treated k562 cells have 70.1% of control activity, and treated HL60 cells have 62% of control activity; Fig. 3B), whereas treated U937 cells have only 43% of control activity, and this is the line in which CoCl2 induces hypomethylation of the Intron 1 CpG island.

Bottom Line: WT1 is a transcription factor expressed in hematopoietic stem cells and in most cases of myeloid leukemia.WT1 expression correlates with hypomethylation of the CpG island in Intron 1, and pharmacologic demethylation of this CpG island induces WT1 mRNA expression.These results reveal a novel model of hypoxia-mediated epigenetic gene regulation.

View Article: PubMed Central - PubMed

Affiliation: Department of Oncology, Division of Pediatric Oncology, Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University, Baltimore, MD, United States of America.

ABSTRACT
WT1 is a transcription factor expressed in hematopoietic stem cells and in most cases of myeloid leukemia. We investigated the roles of hypoxia and epigenetics in the regulation of WT1 expression in myeloid leukemia cells. WT1 expression correlates with hypomethylation of the CpG island in Intron 1, and pharmacologic demethylation of this CpG island induces WT1 mRNA expression. Hypoxia causes decreases in DNMT expression and activity and increased expression and activity of TET2 and TET3, resulting in demethylation of this CpG island and expression of WT1 mRNA. Demethylation of the CpG island, either from pharmacologic treatment or induction of hypoxia, results in transcription of an antisense-oriented lncRNA, and inhibiting lncRNA expression with shRNA blocks WT1 mRNA expression. These results reveal a novel model of hypoxia-mediated epigenetic gene regulation. In addition, this is the first report that TET2 and TET3, increasingly recognized as important epigenetic regulators of gene expression in stem cells and in cancer cells, can be regulated by hypoxia, providing a solid mechanistic link between hypoxia and epigenetic regulation of gene expression with important implications for the role of hypoxia in stem cell function.

No MeSH data available.


Related in: MedlinePlus