Abstract
T-ALL cells can become resistant to γ-secretase inhibitors through reversible epigenetic alterations.
Major finding: T-ALL cells can become resistant to γ-secretase inhibitors through reversible epigenetic alterations.
Concept: Persister cells have compact, repressive chromatin and are dependent on BRD4 for survival.
Impact: Combined use of epigenetic modulators may improve the efficacy of γ-secretase inhibitors in leukemia.
Activating NOTCH1 mutations are common in T-cell acute lymphoblastic leukemia (T-ALL), but blockade of NOTCH1 cleavage and activation with γ-secretase inhibitors (GSI) has only led to modest, transient responses in clinical trials. To gain insight into mechanisms of GSI resistance in T-ALL, Knoechel and colleagues isolated drug-tolerant NOTCH1-dependent T-ALL cells after long-term GSI treatment and evaluated NOTCH pathway activity. The active, cleaved form of NOTCH1 was undetectable and NOTCH1 target gene expression was downregulated in these “persister” cells, but upon GSI withdrawal NOTCH1 cleavage and target gene expression were rapidly restored, suggesting that GSI resistance is reversible and likely epigenetic in nature. Compared with treatment-naïve cells, persister cells had markedly smaller nuclei and higher global levels of repressive histone modifications and heterochromatin-associated proteins, indicative of compact chromatin. Hypothesizing that this altered chromatin state might confer sensitivity to inhibition of epigenetic regulators, the authors screened a library of short hairpin RNAs targeting chromatin modifier genes for those that specifically reduced the survival of persister cells when depleted and identified bromodomain-containing 4 (BRD4) as a top hit. Consistent with this finding, persister cells were significantly more sensitive to the bromodomain inhibitor JQ1 than were treatment-naïve cells. Although BRD4 binding patterns were relatively consistent in treatment-naïve and persister cells, the largest BRD4 binding sites were in persister cells at enhancers near genes with known roles in T-ALL pathogenesis, raising the possibility that the compact chromatin state in persister cells renders enhancer activity more dependent on BRD4. Combined GSI and JQ1 treatment was better at inhibiting growth and survival of primary T-ALL cells than inhibition of either NOTCH1 or BRD4 alone and significantly blocked leukemic cell growth and prolonged survival in patient-derived T-ALL xenograft models. These findings implicate epigenetic resistance mechanisms in refractory T-ALL and suggest that combination treatment with epigenetic modulators may improve responses to NOTCH-targeted therapies.