Genomic 5-hydroxymethylcytosine levels correlate with TET2 mutations and a distinct global gene expression pattern in secondary acute myeloid leukemia

Article metrics

The TET proteins are 2-oxoglutarate- and Fe(II)-dependent oxygenase catalyzing the conversion of 5-methylcytosine (5mC) to 5-hydroxymethylcytosine (5hmC).1 The TET1 (ten–eleven translocation 1) gene was originally identified as an MLL fusion partner in rare cases of acute myeloid leukemia (AML) with a t(10;11)(q22;q23).2, 3 The definite function of 5hmC still remains elusive, but hydroxylation of 5mC has been suggested to be involved in the process of DNA demethylation. This suggests a possible role of 5hmC in epigenetic gene regulation. Recently, hemizygous deletions and mutations of TET2 were found in a wide range of myeloid malignancies, including myelodysplastic syndrome (MDS), myeloproliferative disorders such as chronic myelomonocytic leukemia (CMML) and in secondary AML (sAML).4, 5, 6 Interestingly, very recently, myeloid neoplasias harboring heterozygous TET2 mutations were shown to have decreased levels of 5hmC.7

To explore the relationship among TET2 mutations, global gene expression profiles (GEPs) and 5hmC levels, we measured 5hmC levels in the genomic DNA in a series of 30 sAML patients using a novel assay method employing β-glucosyltransferase from bacteriophage T4.8 In addition to the TET2 mutational status, we screened for IDH1/2 mutations (see Supplementary Material).

All patients had developed AML after a preceding MDS, refractory anemia with excess blast or CMML phase. The average age at diagnosis was 70.8 years. Eight patients had a normal karyotype (nk), eight patients had a complex aberrant karyotype (ak) with more than three chromosomal aberrations and the remainder of the patients had one or two chromosomal aberrations, which are typical of MDS, that is, del(5)(q) (two patients), +8 (five patients), −7 or del(7)(q) (six patients; see Supplementary Table 1). We sequenced the complete coding region of TET2 in all 30 patients. In all, 7 of the 30 patients (23.3%) had TET2 mutations. One patient (no. 16) had single-nucleotide deletions in both alleles of TET2 at amino-acid positions 218 and 519, which caused truncation of the protein after 250 and 533 amino acids, respectively. Two patients (nos. 15 and 26) had nonsense mutations at positions 1216 and 1274, and four patients (nos. 7, 14, 20 and 30) had missense mutations (see Table 1). All the TET2 mutations (except for patient no. 16) were heterozygous. We did not detect any deletions in patients with TET2 mutations using a commercially available fluorescence in situ hybridization probe for the TET2 locus. There was no significant association between TET2 mutational status and any particular chromosomal abnormality. Although there was a trend toward a higher frequency of TET2 mutations in patients with a nk in comparison with patients with an ak (50% (4 out of 8 nk patients) versus 13.6% (3 out of 22 ak patients), χ2-test: P=0.17).

Table 1 Overview of the clinical diagnosis, TET2 and IHD2 mutational status, as well as 5hmC levels of the 30 sAML patients analyzed

The analysis of the 5hmC levels of the patients’ DNA using the β-glucosyltransferase assay revealed a 5hmC content of the DNA, ranging from 0.006 to 0.054%. This range of 5hmC levels, about 9- to 14-fold difference between the lowest and highest measurements, agrees well with the measurements reported by Ko et al.,7 although very different patient groups were assayed. In contrast to the results presented by Ko et al.,7 we did not observe a clear bimodal distribution of 5hmC values. This could be due to the smaller sample number in our series (30 versus 88) and to different patient characteristics in the two studies (mainly sAML in our study versus a broader range of myeloid malignancies in the study of Ko et al.7).

When we compared the presence of TET2 mutation with the 5hmC levels, we found a significant clustering of patients with TET2 mutations in the lower half of 5hmC levels (Figure 1 a). All but one patient with a TET2 mutation belonged to the group of the 15 patients with the lowest 5hmC levels (χ2-test: P=0.03). This corresponded to 5hmC levels of <0.02%. Only one patient with a TET2 mutation, no. 26) had 5hmC levels of >0.020%. These results agree well with the observation of Ko et al.7 Interestingly, there were several patients with very low 5hmC levels that did not have a TET2 mutation. As it was reported recently that IDH1/2 mutations can impair TET2 function, which might also correlate with low 5hmC levels,9, 10 we determined the mutational status of the IDH1 and IDH2 genes in our patients. Only three patients (nos. 7, 17 and 18) had mutations at amino acid R140 in IDH2 (Table 1, Figure 1 a). No mutations in IDH1 were detected. Interestingly, one patient (no. 7) had both mutations in IDH2 and in TET2. In a much larger series of patients reported by Figueroa et al.,10 no patient with both a TET2 and an IDH1/2 mutation was discovered. Patient no. 17 who had an IHD2 mutation had the lowest 5hmC levels in our series. Patient no. 18 had intermediate 5hmC levels. However, there are still eight patients in the lower half of the 5hmC level range who have neither a TET2 nor an IDH1/2 mutation. There was no correlation between TET2 expression levels and 5hmC levels in our patients (data not shown).

Figure 1

(a) Bar graph of 5hmC (%) levels of the 30 patients of this study. Patients with a TET2 mutation or an IDH2 mutation are indicated with a red or a blue rectangle, respectively. The yellow bars indicate the samples that were used for the differential gene expression analysis in the comparison of low versus high 5hmC levels. Note that the patients 9 and 22 did not have gene expression data of sufficient quality to be included in this analysis. (b) Heatmaps of the 50 most significantly differentially expressed probe sets comparing samples with TET2 mutations (n=7) versus TET2 wild type (N=21; top) and samples with low (n=7) and high (n=7) 5hmC levels (bottom).

To determine the impact of TET2 mutations and 5hmC levels on cellular function, we obtained GEPs from 28 patients (all except patient nos. 9 and 22) and performed two comparisons for differential gene expression: (1) patients with TET2 mutations (7 patients) versus patients without TET2 mutations (21 patients) and (2) the 7 patients with the lowest versus the 7 patients with the highest 5hmC levels. The top differentially expressed genes in the high versus low 5hmC level comparison had a lower P-value and had a higher degree of deregulation than the differentially expressed genes from the comparison TET2 mutated versus wild type (Figure 1 b; Supplementary Figure 1 and Supplementary Tables 2 and 3). These results indicate that 5hmC levels are most likely a more relevant measurement to define biologically distinct secondary leukemia subtypes than the TET2 (or IDH1/2) mutational status. The fact that in some patient samples with low 5hmC levels neither TET2 nor IDH1/2 mutations could be identified suggests that additional genes might be directly or indirectly involved in the regulation of 5hmC levels. To further elucidate the regulation of 5hmC levels and their role in leukemogenesis, larger groups of sAML as well as de novo AML patients need to be studied.


  1. 1

    Tahiliani M, Koh KP, Shen Y, Pastor WA, Bandukwala H, Brudno Y et al. Conversion of 5-methylcytosine to 5-hydroxymethylcytosine in mammalian DNA by MLL partner TET1. Science 2009; 324: 930–935.

  2. 2

    Ono R, Taki T, Taketani T, Taniwaki M, Kobayashi H, Hayashi Y . LCX, leukemia-associated protein with a CXXC domain, is fused to MLL in acute myeloid leukemia with trilineage dysplasia having t(10;11)(q22;q23). Cancer Res 2002; 62: 4075–4080.

  3. 3

    Lorsbach RB, Moore J, Mathew S, Raimondi SC, Mukatira ST, Downing JR . TET1, a member of a novel protein family, is fused to MLL in acute myeloid leukemia containing the t(10;11)(q22;q23). Leukemia 2003; 17: 637–641.

  4. 4

    Mohamedali AM, Smith AE, Gaken J, Lea NC, Mian SA, Westwood NB et al. Novel TET2 mutations associated with UPD4q24 in myelodysplastic syndrome. J Clin Oncol 2009; 27: 4002–4006.

  5. 5

    Langemeijer SM, Kuiper RP, Berends M, Knops R, Aslanyan MG, Massop M et al. Acquired mutations in TET2 are common in myelodysplastic syndromes. Nat Genet 2009; 41: 838–842.

  6. 6

    Delhommeau F, Dupont S, Della Valle V, James C, Trannoy S, Massé A et al. Mutation in TET2 in myeloid cancers. N Engl J Med 2009; 360: 2289–2301.

  7. 7

    Ko M, Huang Y, Jankowska AM, Pape UJ, Tahiliani M, Bandukwala HS et al. Impaired hydroxylation of 5-methylcytosine in myeloid cancers with mutant TET2. Nature 2010; 468: 839–843.

  8. 8

    Szwagierczak A, Bultmann S, Schmidt CS, Spada F, Leonhardt H . Sensitive enzymatic quantification of 5-hydroxymethylcytosine in genomic DNA. Nucleic Acids Res 2010; 38: e181.

  9. 9

    Xu W, Yang H, Liu Y, Yang Y, Wang P, Kim SH et al. Oncometabolite 2-hydroxyglutarate is a competitive inhibitor of α-ketoglutarate-dependent dioxygenases. Cancer Cell 2011; 19: 17–30.

  10. 10

    Figueroa ME, Abdel-Wahab O, Lu C, Ward PS, Patel J, Shih A et al. Leukemic IDH1 and IDH2 mutations result in a hypermethylation phenotype, disrupt TET2 function, and impair hematopoietic differentiation. Cancer Cell 2010; 18: 553–567.

Download references


SKB, KS and HL are supported by grants from the Deutsche Forschungsgemeinschaft (SFB 684 and SPP 1463). SKB is supported by a grant from the German Ministry of Education and Research (BMBF) National genome research network (NGFNplus; PKL-01GS0876-6) and by institutional funding from the Helmholtz Zentrum Munich, German Research Center for Environmental Health. We thank Natalia Huk for technical assistance and Alexander Kohlmann, Torsten Haferlach and Claudia Haferlach for primary gene expression data and cytogenetic data.

Author Contributions

NK and FS designed and performed the mutation screening with the help of AD and BK, and wrote the manuscript. SB, AS and HL designed and performed the 5hmC measurements and wrote the manuscript. HL supervised the project. PMK and SS performed cytogenetics and fluorescence in situ hybridization analysis. TH and MM analyzed the GEPs. KS designed experiments and wrote the manuscript. SKB designed experiments, analyzed the data, supervised the project and wrote the manuscript.

Author information

Correspondence to H Leonhardt or S K Bohlander.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Additional information

Supplementary Information accompanies the paper on the Leukemia website

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Konstandin, N., Bultmann, S., Szwagierczak, A. et al. Genomic 5-hydroxymethylcytosine levels correlate with TET2 mutations and a distinct global gene expression pattern in secondary acute myeloid leukemia. Leukemia 25, 1649–1652 (2011) doi:10.1038/leu.2011.134

Download citation

Further reading