Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2015 Apr 22;5:9281.
doi: 10.1038/srep09281.

The Hypomethylating Agent Decitabine Causes a Paradoxical Increase in 5-hydroxymethylcytosine in Human Leukemia Cells

Free PMC article

The Hypomethylating Agent Decitabine Causes a Paradoxical Increase in 5-hydroxymethylcytosine in Human Leukemia Cells

Basudev Chowdhury et al. Sci Rep. .
Free PMC article


The USFDA approved "epigenetic drug", Decitabine, exerts its effect by hypomethylating DNA, demonstrating the pivotal role aberrant genome-wide DNA methylation patterns play in cancer ontology. Using sensitive technologies in a cellular model of Acute Myeloid Leukemia, we demonstrate that while Decitabine reduces the global levels of 5-methylcytosine (5mC), it results in paradoxical increase of 5-hydroxymethylcytosine (5hmC), 5-formylcytosine (5fC) and 5-carboxylcytosine (5caC) levels. Hitherto, the only biological mechanism known to generate 5hmC, 5fC and 5caC, involving oxidation of 5mC by members of Ten-Eleven-Translocation (TET) dioxygenase family, was not observed to undergo any alteration during DAC treatment. Using a multi-compartmental model of DNA methylation, we show that partial selectivity of TET enzymes for hemi-methylated CpG dinucleotides could lead to such alterations in 5hmC content. Furthermore, we investigated the binding of TET1-catalytic domain (CD)-GFP to DNA by Fluorescent Correlation Spectroscopy in live cells and detected the gradual increase of the DNA bound fraction of TET1-CD-GFP after treatment with Decitabine. Our study provides novel insights on the therapeutic activity of DAC in the backdrop of the newly discovered derivatives of 5mC and suggests that 5hmC has the potential to serve as a biomarker for monitoring the clinical success of patients receiving DAC.


Figure 1
Figure 1. The effect of DAC on 5mC and 5hmC in HL-60 cells.
(a) The scheme of mammalian ‘active demethylation’ pathway. (b) Immunocytochemistry for 5mC (green channel) and 5hmC (red channel) performed on untreated and 3 μM DAC treated HL-60 cells. The scale bar denotes 5 μm. (c & d) Global levels of 5mC and 5hmC by EIA respectively in untreated, 0.5 μM and 3 μM DAC treated HL-60 cells. The limit of detection of 5mC was 5 pg/100 ng of added DNA, while that for 5hmC was 2 pg/200 ng of added DNA. (e & f) LC-MS/MS quantitation of levels of 5mC and 5hmC in terms of ratios of 5-methyl-2′-deoxycytidine (5mdC) or 5-hydroxymethyl-2′-deoxycytidine (5hmdC) to those of deoxycytidine (dC) respectively in untreated, 0.5 μM and 3 μM DAC treated HL-60 cells. Limits of Detection (LOD) of 5mdC and 5hmdC were 0.09 and 0.11 fmol respectively.
Figure 2
Figure 2. Molecular profiling of changes occurring during DAC treatment.
(a & b) The transcriptional changes, measured by qPCR, of in DNMT1, DNMT3A, DNMT3B, TET1, TET2 and TET3 in 0.5 μM and 3 μM DAC treated HL-60 cells, respectively, compared to their levels in untreated HL-60 cells. All values have been normalized with GAPDH/β-actin (c) Western Blot analysis to understand the precise effect on DNMTs. β-actin is shown as a protein loading control.
Figure 3
Figure 3. Mathematical simulation using a multi-compartmental model.
(a) A schematic representation of the original six compartmental model components (black) and the extended model components (red). Solid lines indicate epigenetic modification by enzymes. Broken lines indicate the effects of cell division. The effects of DAC treatment (at time = 0) in machina on the relative abundance of unmethylated (red), methylated (blue), and hydroxymethylated (green) CpGs (b) in cells with non-selective TET; (c) in cells with TET proteins which are intrinsically fully selective for hemi-methylated CpG dinucleotides; (d) in cells with TET partially selective (6-fold) for hemi-methylated CpG dinucleotides; (e) and in cells with TET partially selective for hemi-methylated CpG dinucleotides demonstrating the effect on abundance of 5fC (purple) and 5caC (brown).
Figure 4
Figure 4. FCS measurement of TET-CD dynamics in MCF7 cells upon DAC treatment.
(a) Representative autocorrelation function of TET-CD-GFP protein dynamics in early stage of DAC treatment. (b) Representative autocorrelation function of TET-CD-GFP protein dynamics in late stage of DAC treatment. (c) The percentage of bound TET-CD-GFP was calculated at different time points (n > 40 measurements for each point). (d) The corresponding changes of 5mC and 5hmC were determined after 40 hours of DAC treatment.
Figure 5
Figure 5. Our proposed model to explain the increase of 5hmC, 5fC and 5caC upon DAC treatment.
The TET proteins appear to be partially selective for hemi-methylated CpGs, and in absence of DNMT1 (due to covalent binding with DAC), can convert the hemi-methylated CpGs into hemi-hydroxymethylated, hemi-formylated or hemi-carboxylated CpGs. Semi-conservative DNA replication in the presence of DAC may give rise to a condition where some methylated-CpGs incorporate DAC in place of Cytidine in the newly synthesized strand while the parent strand maintains the original 5mC mark (i.e. resulting in 5mC-G/G-DAC dinucleotide). In absence of DNMTs (DAC induced trapping and degradation) or knowledge of non-TET mediated active demethylation pathway in mammalian cells, it may be likely that the TETs can act on the 5mC of the parent strand, converting it to the other derivatives in the active demethylation pathway (5mC-G/G-DAC → 5hmC-G/G-DAC → 5fC-G/G-DAC → 5caC-G/G-DAC). Similarly, 5hmC or 5fC of the hydroxymethylated-CpGs and formylated-CpGs respectively incorporating DAC in the daughter strand could get converted to downstream derivatives of the pathway.

Similar articles

See all similar articles

Cited by 10 articles

See all "Cited by" articles


    1. Robertson K. D. DNA methylation and human disease. Nature Reviews Genetics 6, 597–610, 10.1038/nrg1655 (2005). - DOI - PubMed
    1. Baylin S. B. & Jones P. A. A decade of exploring the cancer epigenome - biological and translational implications. Nat. Rev. Cancer 11, 726–734, 10.1038/nrc3130 (2011). - DOI - PMC - PubMed
    1. Suzuki M. M. & Bird A. DNA methylation landscapes: provocative insights from epigenomics. Nature reviews. Genetics 9, 465–476, 10.1038/nrg2341 (2008). - DOI - PubMed
    1. Moran-Crusio K. et al. Tet2 loss leads to increased hematopoietic stem cell self-renewal and myeloid transformation. Cancer Cell 20, 11–24 (2011). - PMC - PubMed
    1. Pronier E. et al. Inhibition of TET2 Conversion of 5-Methylcytosine to 5-Hydroxymethylcytosine Disturbs Myelopoiesis and Granulo-Monocytic Differentiation. Blood 116, 669–669 (2010). - PMC - PubMed

Publication types

MeSH terms