Cdkn2a reactivation plays a critical role in UNC1999-mediated growth inhibition. (A) Change in p16Ink4a and p19Arf gene expression following a 3-day treatment with 3 μM UNC1999 among 6 independent murine leukemia lines, either freshly immortalized by MLL-AF9 or MLL-ENL (left) or derived from MLL-AF9–induced primary murine leukemia (right). Y-axis represents fold-change in gene expression after normalization to GAPDH and DMSO treatment, and error bars represent SD of triplicates. *P < .05; **P < .01; ***P < .001. (B) RT-qPCR shows time-dependent derepression of p16Ink4a and p19Arf by UNC1999 in a leukemia line derived from MLL-AF9–induced primary tumors. *P < .05; **P < .01; ***P < .001. (C) Summary of cell-cycle status of MLL-AF9–transformed murine leukemia progenitors following 2-day or 7-day treatment with DMSO, or 3 μM UNC2400 or UNC1999. (D) Representative histograms showing DNA contents measured by PI staining of MLL-AF9–transformed leukemia cells after treatment with 3 μM of compounds for 2 days. (E) Relative proliferation of MLL-AF9–transformed murine leukemia cells, either wild-type (red) or p16Ink4a^−/−/p19Arf^−/− (purple), after treatment with various concentrations of UNC1999 for 12 days. Y-axis, presented as the mean of triplicates ± SD, represents the relative percentage of cell numbers after normalization to DMSO treatment. (F) Summary of apoptotic induction in MLL-AF9–transformed murine leukemia progenitors, either wild-type (WT) or p16Ink4a/p19Arf-deficient, following a 6-day treatment with 3 μM of compounds as assayed by PI staining. **P < .01; ***P < .005. WT, wild type.