ICSBP promoter methylation in myelodysplastic syndromes and acute myeloid leukaemia

Interferon consensus sequence-binding protein (ICSBP) was first described as a negative regulator of interferon-inducible genes, implying an important role in interferon (IFN)γ signalling and immune response.1, 2 Later, ICSBP-deficient mice were shown to develop a disease resembling human chronic myeloid leukaemia,3, 4 suggesting a tumour suppressor function of ICSBP. ICSBP functions as a molecular switch factor directing the differentiation of bi-potential myeloid precursors toward the monocytic lineage,5 possibly by forming heterodimeric complexes with PU.1.6 Moreover, ICSBP has pro-apoptotic properties,7 and modulates proliferation of leukaemic cells, requiring the chemokines CCL6 and CCL9.8, 9 Despite these important roles of ICSBP shown in the differentiation of haematopoietic cells, relatively less is known about its inactivation in human myeloid malignancies. Patients with chronic myeloid leukaemia and acute myeloid leukaemia (AML) show reduced ICSBP-mRNA expression.10 Furthermore, reduced ICSBP expression correlates with poor cytogenetic response to IFNα treatment.11 Gene expression profiling of CD34+ haematopoietic stem/progenitor cells revealed that therapy-related AML with loss of chromosome 5 or deletion of 5q (−5/del(5q)), in addition to upregulation of growth-promoting genes, had no expression of ICSBP.12 Using a novel approach to determine the methylation status of candidate tumour suppressor genes called methyl-binding (MB)-PCR, the ICSBP gene was identified as a target of hypermethylation in AML.13

To obtain a first estimation of whether and to what extent ICSBP transcription is downregulated and whether there is an association between downregulation of ICSBP and recurrent cytogenetic aberrations of myelodysplastic syndrome (MDS) and AML, we screened the PubMed GEO database (http://www.ncbi.nlm.gov/geo/accession number GSE425/8653, Supplementary Figure S1a and b).14, 15 By gene-expression data mining, a significant downregulation of ICSBP in AML with t(15;17) and with t(8;21), and a tendency of ICSBP downregulation in –7/del(7q) was found (Supplementary Figure S1c). Furthermore, an ICSBP downregulation was published for therapy-induced AML with –5/del(5q).12 To substantiate these findings, we investigated a cohort of 100 patients with myeloid malignancies that had been characterized cytogenetically (Supplementary Table S3). Informed consent was obtained according to the Declaration of Helsinki. The study was approved by the Institutional Ethics Committee of Hannover Medical School under No. 2899 (26.02.2002).

Mutations within the DNA-binding domain (DBD, exons 2 and 3) and the IRF-association domain (IAD, exons 7, 8 and 9) lead to a loss of wild-type function in in vitro and in vivo models.5, 6, 16 We analysed these exons by direct sequencing in 68 patients from whom material was available (Supplementary Materials and Methods). We found unaltered sequences in almost all analysed patients. Clearly, we cannot rule out that mutations in exons 4, 5 and 6 might also play a role. Two MDS patients with del(5q) harboured the silent single-nucleotide substitution c.864C>T; p.Phe288Phe in exon 7 (Supplementary Table S1). It seemed unlikely that this nucleotide substitution caused downregulation of ICSBP. Furthermore, we investigated the frequency of deletions of the ICSBP chromosomal locus 16q24.1. A dual-colour fluorescence in situ hybridisation probe to enable detection of micro-deletions was generated (Supplementary Materials and Methods and Supplementary Figure S2). In 26 patients from different cytogenetic subgroups, material was available, but no deletion or structural rearrangement was identified (Supplementary Table S1). To extend these analyses, we carried out an in silico ICSBP deletion screening in 69 MDS and 112 AML patients.17, 18, 19 Loss of one copy of the ICSBP locus in 16q24.1 was seen in 5% of the analysed AML patients (Supplementary Figure S3). All of them had a complex karyotype.

In Table 1 and Supplementary Table S2, the methylation status of the ICSBP promoter, as determined by methylation-specific PCR (MSP; Figure 1a and Supplementary Materials and Methods) is summarised. Overall, 12 of 59 (20%) patients with MDS, 3 of 12 (25%) patients with de novo AML, and 8 of 29 (28%) patients with secondary AML (Table 1) presented the methylated MSP product and were declared as methylated positive (Figure 1b). No methylation was seen in the eight healthy controls. Male patients had a significantly higher chance for ICSBP promoter methylation than female patients (OR=2.9 (0.34; 8.78); P=0.043). Both gender groups exhibit comparable age pattern. ICSBP methylation was found in all cytogenetic subgroups of MDS and AML except for cytogenetically normal (CN) MDS. Accordingly, the chance for ICSBP methylation was 3.57-fold higher in the patients with aberrant karyotype (24%) compared with CN patients (10%)(OR=3.57 (0.597; 68.658); P=0.24). This is true both for patients with MDS and with secondary AML (22 versus 0% for MDS; 30 versus 17% for secondary AML)(Supplementary Table S2). In more detail, patients carrying a single chromosomal aberration, for example, del(5q), –7/del(7q) or trisomy 8, presented a higher ICSBP methylation rate than CN patients or patients with complex karyotype. Considering all cytogenetic subgroups, the highest ICSBP promoter methylation was observed in secondary AML (Supplementary Table S2, Figure 1c), with a trend in increasing ICSBP methylation in the CN patient groups (0% MDS versus 17% secondary AML), and in the trisomy 8 patient groups (29% MDS versus 50% secondary AML) (Supplementary Table S2). In conclusion, a substantial proportion of patients with MDS and AML showed ICSBP promoter methylation, particularly those with secondary AML.

Table 1 Methylation of ICSBP promoter related to clinical, diagnostic and cytogenetic characteristics of 100 patients with MDS or AML
Figure 1

Methylation of the ICSBP promoter in patients with MDS or AML. (a) The CpG island of ICSBP: each vertical bar represents a single CpG site. Arrows indicate the start of the non-coding exon 1 and the start codon. Positions of amplicons for either methylated PCR product (m) or unmethylated PCR product (u) are indicated. Amplicon analysed by pyrosequencing is indicated as a grey bar. (b) PCR products are visualised by ethidium bromide staining after gel electrophoresis in 2% agarose. Representative results of MSP are shown for selected patients: the methylated (m) 180-bp product is present in +con (HL-60) and in patients P73 and P96; the unmethylated (u) 184-bp product is seen in –con (U937) and in patients P50, P51, P60, P61, P62, P63, P64, P97, P75, P76, P77 and P78. Number of patients correlates to the patient number given in Supplementary Table S3 (Excel). (c) Percentage of promoter methylation in cytogenetically defined subgroups of MDS and secondary AML, for example, cytogenetically normal, loss of 7 or 7q, trisomy 8, loss of 5q, complex karyotype and in all patients (total). Patients with secondary AML have an increased rate of ICSBP promoter methylation compared with patients with MDS. n.c.=no case. (d) Pyro-sequencing was carried out on selected cases with methylated ICSBP and on cases with unmethylated ICSBP according to MSP. Percentage of methylation is presented as scatterplots including mean and s.e.m. Statistics are calculated by natural logarithm and unpaired t-test, with Welch's correction ***P=0.0001. (e) Real-time RT-PCR was carried out to determine ICSBP expression in cases with methylated ICSBP promoter in comparison with relevant controls. White bars correlate to bone marrow (BM) control and grey bars correlate to peripheral blood (PB) control. Sample IDs are the same as in Supplementary Table S3 (Excel).

To verify the results of MSP, we pyrosequenced fragments of the ICSBP CpG island (Figure 1a and Supplementary Materials and Methods) in 15 patients presenting a methylated MSP product and eight patients with an unmethylated MSP product (Figure 1d and Supplementary Figure S4). Quantification of the CpG methylation status correlated strongly with the MSP results. Patients with methylated MSP products had significantly higher rates of CpG methylation than patients with unmethylated MSP products (P=0.00019). Interestingly, all patients with an unmethylated MSP product showed a very low percentage of methylated fragment 01 (5%; Figure 1d) in pyrosequencing. Calculating a cut-off line by the mean plus 2 s.d., that is, 6.44%, 13 out of 15 patients with methylated MSP product had higher methylation values of the 5′ region of ICSBP compared with the unmethylated patients. Furthermore, ICSBP expression of 14 patients presenting a methylated MSP product was analysed by real-time reverse-transcription PCR (Supplementary Materials and Methods). ICSBP expression levels were lower in patients with ICSBP promoter methylation than in the relevant controls (Figure 1d), reinforcing that CpG island methylation may lead to a downregulation of ICSBP in myeloid malignancies.

To determine the functional consequences of ICSBP promoter methylation in myeloid malignancies, we investigated ICSBP promoter methylation in the AML cell lines F-36P, HL-60, Kasumi-1, ME-1, MV4-11, NB-4 and U937 (Supplementary Figure S5). As Kasumi-1 presented only the methylated MSP product and lacked ICSBP protein, this cell line was chosen for further functional analyses. After treatment with the demethylating agent 5-aza-2′-deoxycytidine (DAC), ICSBP was demethylated as demonstrated by an MSP (Figure 2a) leading to ICSBP re-expression (Figures 2b and c). The ICSBP expression was further increased by addition of IFNγ (P=0.0301, Figure 2b), which is also needed for the ICSBP protein function.20 Although DAC treatment induced growth inhibition (Figure 2d) and apoptosis (P=0.0252, Figure 2e), the addition of IFNγ had no obvious effects on proliferation or apoptosis induction.

Figure 2

Re-expression of ICSBP by treatment of the Kasumi-1 cells with DAC and IFNγ. (a) MSP products in 2% agarose gel the methylated (m) and the unmethylated (u) products in Kasumi-1. (b) Increasing DAC concentrations and IFNγ-induced ICSBP re-expression as determined by real-time RT-PCR. *P=0.0301 (Students t-test). (c) Western blot analysis to show ICSBP re-expression on protein level. Representative blot of three experiments. (d) Proliferation curves obtained by WST-1 assay: Curves 1 and 3 relate to cells treated with IFNγ (arrow indicates time point of addition), curves 2 and 4 relate to cells without IFNγ treatment. n=3. (e) Rate of apoptosis as determined by the caspase-3 and -7 activity normalised to WST-1. Students t-test *P=0.0252. IFNγ treatment had no influence on the rate of apoptosis; n=3.

However, ICSBP is known to be involved in the regulation of distinct apoptotic pathways and sensitises cells for apoptosis induced by chemotherapeutic agents like etoposide,7 a topoisomerase-II inhibitor inducing DNA double-strand breaks. As described above, ICSBP re-expression alone, either by DAC treatment or by ectopic expression after ICSBP transfection had no dramatic effect on apoptosis in contrast to untreated cells (Figures 3a and b). To induce apoptosis, cells were treated with etoposide for 18 h. The rate of apoptosis increased 10.5-fold after addition of etoposide (P=0.0035, n=3). Cells expressing high levels of ICSBP, induced by DAC and IFNγ treatment, still presented a 2.7-fold higher rate of apoptosis after addition of etoposide compared with cells treated only with DAC and etoposide (P=0.0174, n=3, Figure 3a). Moreover, to rule out that global hypomethylation by DAC treatment and not the ICSBP re-expression, apoptosis was measured in transfected cells expressing ICSBP ectopically (Figure 3d). Again, the rate of etoposide-induced apoptosis was 5.9-fold higher in ICSBP-transfected and IFNγ-treated cells after addition of etoposide than in ICSBP-transfected cells without etoposide (P=0.0485; Figure 3b). As demonstrated by caspase-3 cleavage, caspase-3 is involved in ICSBP-induced apoptosis (Figures 3c and d). This is in agreement with previous reports on caspase-3 cleavage by etoposide treatment in U937 cells overexpressing ICSBP.7

Figure 3

Etoposide treatment (20 μg/ml) induces caspase-3-dependent apoptosis depending on ICSBP expression in Kasumi-1 cells. (a) Rate of apoptosis after ±etoposide treatment determined by the caspase-3 and -7 activity normalised to WST-1. ICSBP expression is induced by DAC and IFNγ treatment. n=3; *P=0.0174; **P=0.0035 (Student's t-test). (b) Rate of apoptosis after transfection of an ICSBP expression vector and empty vector as control and treatment with IFNγ and etoposide; *P<0.05. (c) RIPA lysates of cells treated with different combinations of DAC, IFNγ and etoposide (see a), and (d) RIPA lysates of cells transfected with an ICSBP expressing construct (see b): Caspase-3 cleavage was analysed by western blot using an anti-caspase-3 antibody. The expression of ICSBP was controlled by an anti-ICSBP antibody. Anti-GAPDH/β actin antibody was used as loading control. Representative blots of three experiments.

In conclusion, we showed that the promoter of ICSBP, an important regulator of differentiation and survival of haematopoietic cells,21 was methylated in a substantial proportion of patients with MDS and AML. ICSBP promoter methylation seems to be the main mechanism of ICSBP inactivation in myeloid malignancies and may be functionally important for accumulation of chromosome aberrations during leukemic progression. We also demonstrate that resistance to DNA damage-induced apoptosis mediated by ICSBP promoter hypermethylation can be overcome by treatment with DAC and may re-sensitize myeloid malignancies for chemotherapy with topoisomerase inhibitors.


  1. 1

    Driggers PH, Ennist DL, Gleason SL, Mak WH, Marks MS, Levi BZ et al. An interferon gamma-regulated protein that binds the interferon-inducible enhancer element of major histocompatibility complex class I genes. Proc Natl Acad Sci USA 1990; 87: 3743–3747.

  2. 2

    Weisz A, Marx P, Sharf R, Appella E, Driggers PH, Ozato K et al. Human interferon consensus sequence binding protein is a negative regulator of enhancer elements common to interferon-inducible genes. J Biol Chem 1992; 267: 25589–25596.

  3. 3

    Holtschke T, Lohler J, Kanno Y, Fehr T, Giese N, Rosenbauer F et al. Immunodeficiency and chronic myelogenous leukemia-like syndrome in mice with a targeted mutation of the ICSBP gene. Cell 1996; 87: 307–317.

  4. 4

    Turcotte K, Gauthier S, Tuite A, Mullick A, Malo D, Gros P . A mutation in the Icsbp1 gene causes susceptibility to infection and a chronic myeloid leukemia-like syndrome in BXH-2 mice. J Exp Med 2005; 201: 881–890.

  5. 5

    Hashmueli S, Gleit-Kielmanowicz M, Meraro D, Azriel A, Melamed D, Levi BZ . A truncated IFN-regulatory factor-8IFN consensus sequence-binding protein acts as dominant-negative, interferes with endogenous protein-protein interactions and leads to apoptosis of immune cells. Int Immunol 2003; 15: 807–815.

  6. 6

    Meraro D, Hashmueli S, Koren B, Azriel A, Oumard A, Kirchhoff S et al. Protein-protein and DNA-protein interactions affect the activity of lymphoid-specific IFN regulatory factors. J Immunol 1999; 163: 6468–6478.

  7. 7

    Gabriele L, Phung J, Fukumoto J, Segal D, Wang IM, Giannakakou P et al. Regulation of apoptosis in myeloid cells by interferon consensus sequence-binding protein. J Exp Med 1999; 190: 411–421.

  8. 8

    Burchert A, Cai D, Hofbauer LC, Samuelsson MK, Slater EP, Duyster J et al. Interferon consensus sequence binding protein (ICSBP; IRF-8) antagonizes BCR/ABL and down-regulates bcl-2. Blood 2004; 103: 3480–3489.

  9. 9

    Heuser M, Park G, Moon Y, Berg T, Xiang P, Kuchenbauer F et al. Extrinsic signals determine myeloid-erythroid lineage switch in MN1 leukemia. Exp Hematol 2010; 38: 174–179.

  10. 10

    Schmidt M, Nagel S, Proba J, Thiede C, Ritter M, Waring JF et al. Lack of interferon consensus sequence binding protein (ICSBP) transcripts in human myeloid leukemias. Blood 1998; 91: 22–29.

  11. 11

    Schmidt M, Hochhaus A, Nitsche A, Hehlmann R, Neubauer A . Expression of nuclear transcription factor interferon consensus sequence binding protein in chronic myeloid leukemia correlates with pretreatment risk features and cytogenetic response to interferon-alpha. Blood 2001; 97: 3648–3650.

  12. 12

    Qian Z, Fernald AA, Godley LA, Larson RA, Le Beau MM . Expression profiling of CD34+ hematopoietic stem/progenitor cells reveals distinct subtypes of therapy-related acute myeloid leukemia. Proc Natl Acad Sci USA 2002; 99: 14925–14930.

  13. 13

    Gebhard C, Schwarzfischer L, Pham TH, Andreesen R, Mackensen A, Rehli M . Rapid and sensitive detection of CpG-methylation using methyl-binding (MB)-PCR. Nucleic Acids Res 2006; 34: e82.

  14. 14

    Bullinger L, Rucker FG, Kurz S, Du J, Scholl C, Sander S et al. Gene-expression profiling identifies distinct subclasses of core binding factor acute myeloid leukemia. Blood 2007; 110: 1291–1300.

  15. 15

    Bullinger L, Ehrich M, Dohner K, Schlenk RF, Dohner H, Nelson MR et al. Quantitative DNA methylation predicts survival in adult acute myeloid leukemia. Blood 2010; 115: 636–642.

  16. 16

    Yang D, Thangaraju M, Browning DD, Dong Z, Korchin B, Lev DC et al. IFN regulatory factor 8 mediates apoptosis in nonhemopoietic tumor cells via regulation of Fas expression. J Immunol 2007; 179: 4775–4782.

  17. 17

    Parkin B, Erba H, Ouillette P, Roulston D, Purkayastha A, Karp J et al. Acquired genomic copy number aberrations and survival in adult acute myelogenous leukemia. Blood 2010; 116: 4958–4967.

  18. 18

    Scharpf RB, Ruczinski I, Carvalho B, Doan B, Chakravarti A, Irizarry RA . A multilevel model to address batch effects in copy number estimation using SNP arrays. Biostatistics 2011; 12: 33–50.

  19. 19

    Smith AE, Mohamedali AM, Kulasekararaj A, Lim Z, Gaken J, Lea NC et al. Next-generation sequencing of the TET2 gene in 355 MDS and CMML patients reveals low-abundance mutant clones with early origins, but indicates no definite prognostic value. Blood 2010; 116: 3923–3932.

  20. 20

    Huang W, Horvath E, Eklund EA . PU.1 interferon regulatory factor (IRF) 2, and the interferon consensus sequence-binding protein (ICSBP/IRF8) cooperate to activate NF1 transcription in differentiating myeloid cells. J Biol Chem 2007; 282: 6629–6643.

  21. 21

    Tamura T, Ozato K . ICSBP/IRF-8: its regulatory roles in the development of myeloid cells. J Interferon Cytokine Res 2002; 22: 145–152.

Download references


We thank Friederike Grundstedt and Sabine Schreek for their excellent technical assistance. Further, we thank Gillian Teicke for editing the manuscript. This study was supported by grants from the BMBF (BMFS-Netzwerk ‘Chromosomal instability as a basic genetic mechanism of tumorigenesis in congenital bone marrow failure syndromes’ 01 GM 0618). NO receives a fellowship from the Deutsche José Carreras Leukämie-Stiftung eV.

Author information

Correspondence to N Otto.

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

Otto, N., Manukjan, G., Göhring, G. et al. ICSBP promoter methylation in myelodysplastic syndromes and acute myeloid leukaemia. Leukemia 25, 1202–1207 (2011). https://doi.org/10.1038/leu.2011.61

Download citation

Further reading