To the best of our knowledge, no data have been published so far about aberrant CpG island hypermethylation of SFRP-1, -2, -4, and -5 in the context of JAK2V617F in Ph-MPN.
In contrast to CML, a number of novel mutations have been described in BCR-ABL1-negative MPN. Among those are mutations involving JAK2, MPL, TET2, ASXL1, CBL, IDH, and IKZF1[17, 20]. Wnt signaling plays an important role in stem-cell self-renewal as well as in differentiation and proliferation of hematopoietic progenitor cells. Recently, it has been reported that Wnt signaling could be one of the mechanisms that shape a niche in the bone marrow supportive of hematopoiesis
. Since aberrant activation of the Wnt pathway has been demonstrated to contribute to leukemogenesis and until now, in contrast to solid tumors, no activating mutations in the genes of the Wnt pathway have been described in myeloid or lymphatic malignancies, we investigated epigenetic regulation of SFRP-1, -2, -4, and -5 as negative regulators of Wnt signaling. Our data indicate that CpG island hypermethylation of SFRP genes is a frequent event in Ph-MPN with aberrant methylation of at least one of the four genes being detected in 30% of the primary patient samples. Compared to the SFRP methylation patterns in other hematopoietic malignancies, aberrant promoter methylation in our study occurred less frequently than in AML
. We could show a preferential hypermethylation of SFRP-1 (4%) and of SFRP-2 (25%), in particular. This is in contrast to the methylation frequency of SFRP genes in acute lymphatic leukemia (ALL)
, where a more balanced methylation frequency of the four genes could be shown, but confirms findings in AML showing a preferential hypermethylation of SFRP-1 (29%) and SFRP-2 (19%).
We could show a significant association of SFRP-1 hypermethylation and diagnosis of ET. Regarding the limited number of patient samples methylated at the SFRP-1 promoter (2/57), those results require further confirmation. Furthermore, a study in 52 ET samples did not find any evidence of hypermethylation of soluble Wnt inhibitors
, whereas our analysis, also using MSP and the same primer sequences, revealed a rate of 40% of ET samples being methylated.
In contrast to SFRP-1, SFRP-2 hypermethylation in our study occurred in all Ph-MPN subtypes without any association with a specific entity. However, there was a significant correlation between aberrant SFRP-2 promoter methylation and the presence of JAK2V617F.
The autoactivating JAK2V617F mutation is responsible for constitutive signaling of the JAK2 protein with subsequent hypersensitivity to hematopoietic growth factors
. When comparing the incidence of the JAK2V617F mutation and the frequency among the MPN subtypes with published data, the relative distribution (80% PV, 50% ET, 52% MF) we obtained in our study is in accordance with previous reports
. Limited data are available about epigenetic disturbances in MPN. Increasing evidence suggests that loss-of-function mutations in genes involved in epigenetic regulation, TET2, ASXL1, and EZH2, as well as JAK2V617F-mediated phosphorylation of protein Arg N-methyltransferase (PRMT5) may play a role in disease pathogenesis, either as ‘pre-JAK2 events’ or occurring in late phase disease
Aberrant CpG island methylation has been reported for ABL, p14, p15, p16, RARß2, and SDF1 receptor CXCR4. SOCS proteins are the most thoroughly studied inhibitors of JAK-STAT pathways. Silencing of SOCS-1 and -3 in association with promoter hypermethylation has been reported by different groups
[12, 16, 27]. However, Capello et al. reported that SOCS-3 methylation was more frequent among JAK2V617F-negative patients. Nevertheless, the results suggest that epigenetic inactivation seems to play a role in Ph-MPN. Epigenetics might be a complementary or alternative mechanism to the JAK2V617F mutation in the pathogenesis of Ph-MPN, leading to dysregulation of JAK-STAT signal transduction and thus to growth factor hypersensitivity. Our findings are in accordance with this hypothesis showing a trend towards an association between JAK2V617F and SFRP-2 hypermethylation, thus, an association between genetic and epigenetic aberrations. Interesting in this context is the comparison of Ph-MPN with CML. The identified disease-causing mutation in CML which codes for a fusion-protein with tyrosine kinase activity provides a potential therapeutic target. The clinical success of the tyrosine kinase inhibitor imatinib suggests that the mutation might be one of the main events leading to the pathogenesis of CML. However, regarding the reported Wnt pathway activation in CML cancer stem cells
 and the reported correlation of SFRP-1 promoter methylation and resistance to imatinib
 the impact of hypermethylation-associated gene silencing in CML in contrast to Ph-MPN has to be further elucidated. According to our study, SFRP promoter hypermethylation in CML seems to be an uncommon event.