In addition, CD56 + IL-15 DCs fall within the flow cytometric scatter gate of DCs, but not of lymphocytes. The co-expression of the myeloid DC lineage markers 3,4,5-Trimethoxyphenylacetic acid BDCA-1 and CD11c along with the absence of CD7 expression, which allows their accurate discrimination from NK cells, lends further support to the notion that IL-15 DCs are unrelated to NK cells in spite of their partial positivity for CD56. To corroborate these phenotypic data and to confirm that IL-15 DCs also functionally qualify as DCs, we performed an allo-MLR as well as an antigen presentation assay. Both CD56 + and CD56�C IL-15 DCs are able to stimulate allogeneic T cell proliferation, thereby fulfilling one of the basic functional criteria for being qualified as DCs. Furthermore, in this study, we also show that IL-15 DCs, as would be expected from DCs, are capable of processing and presenting the WT1 tumor antigen to CD8 + T cells. Together with previous observations from our group and others, these data confirm that IL-15 DCs are “authentic” myeloid DCs not only from the phenotypic but also from the functional point of view. Strikingly, in the WT1 antigen presentation assay, CD56 + IL-15 DCs were found to have a superior antigen-presenting capacity over their CD56�C counterparts. Both fractions had comparable expression of the WT1 protein following electroporation, suggesting that CD56 + DCs have a higher intrinsic ability to process and present endogenously synthesized antigen to T cells. Although the precise functional role of CD56 on DCs remains to be elucidated, the above data suggest that the expression of CD56 on DCs is linked with superior immunostimulatory activity. This mirrors the situation in NK cells and CD56-expressing T cells where CD56 expression and antigen density correlate with activation status and enhanced immune function. Further support for this statement comes from the phenotypic data presented in Table 1, which show that CD56 + IL-15 DCs are in a more differentiated and activated modus as compared to their CD56�C counterparts. The observation that CD56 + IL-15 DCs, in 4-(Benzyloxy)phenol addition to being potent allostimulatory and antigen-presenting cells, are endowed with a cytotoxic capacity is a novel finding that adds to the growing body of evidence that DCs can adopt an “unconventional” cytotoxic effector function and act as killer cells. Among the stimuli capable of triggering such ��killer DC�� function are type I and II IFNs, TLR ligands and, as shown here and in another recent study, IL15. The fact that IL-15, a known growth factor for NK cells, was used in this study for DC differentiation as well as the fact that IL-15 DCs were found to be cytotoxic against the NK prototype target K562, prompted us to perform rigorous culture purity assessments in order to exclude the possibility that the observed cytotoxic effects were due to the presence of contaminating NK cells. Based on the model proposed by Stary et al., at least 10% of NK cell contaminants would have been needed to account for the cytotoxic activity reported in the present study. The trace contamination of IL-15 DC cultures by lymphocytes was thus far too low to account for the observed cytotoxic effects and was therefore considered negligible. This was also further supported by the finding that neither CD562 nor CD56 + IL-15 DC preparations had cytotoxic activity against the U937 cell line, another well-recognized NK-sensitive target. The lack of cytotoxicity against U937 identifies a second point of difference between IL-15 DCs and NK cells.