DAPK1 showed an inverse correlation between promoter methylation and gene expression consistent with epigenetic gene regulation ( Figure 2a, Table 1). The predictive value for true DAPK1 methylation was 1.0, sensitivity 0.86 and accuracy of the classifier 0.95. T-ALL cell lines on an average showed more than 100-fold higher DAPK1 expression levels than did NK, NK/T-and T-cell lymphoma cell lines (Figure 2b). Western blot analysis confirmed cell type-specific DAPK1 expression also on the protein level ( Figure 2c, Table 1).To verify the epigenetic regulation of DAPK1, we applied the DNA demethylating agent 5-Aza-2 0 deoxycytidine (Aza). Aza treatment for 4 days at 5 mM induced a DAPK1 mRNA expression in the DAPK1-methylated cell lines (KARPAS-299, L-82, YT) more than 200-fold. It has recently been described that the transcriptional repressor DAXX suppresses DAPK1 by the induction of DNA hypermethylation. 17 We performed quantitative PCR to find out whether the DAPK1 hypermethylated cell lines (NK, NK/T and T-ALCL) had substantially higher expression levels of DAXX than the unmethylated T-ALL cell lines. This is not the case; DAXX is constitutively, and to a similar extent, expressed in all cell lines tested (Table 1). These results suggest that the differences in DAPK1 methylation between T-ALL and NK cell lines are not the result of a differential DAXX expression.In summary, the hypermethylation of DAPK1 differentiates NK cell lines from T-ALL derived cell lines. The different DAPK1 methylation patterns in the lymphomatous entities cannot simply be explained by different expression levels of DAXX, a transcriptional repressor that recruits DNA methyltransferases to target promoters.