Nuclear NPM-ALK Protects Myc from Proteosomal Degradation and Contributes to High Protein Expression in Cancer Stem-like Cells in ALK-Positive Anaplastic Large Cell Lymphoma

While Myc is one of the oncoproteins most frequently deregulated in human cancers, its expression and significance in ALK-positive anaplastic large cell lymphoma (ALK+ALCL) has not been extensively studied. Previous studies from our group have shown that a high level of Myc proteins is a characteristic feature of a small subset of ALK+ALCL cells that carry cancer stem-like features (labeled RR cells), and shRNA knockdown or pharmacological inhibition of Myc in these cells significantly decreased their cancer stemness and phenotypically converted them into 'bulk' cancer cells (labeled RU cells). In this study, we hypothesized that NPM-ALK, the key oncogenic driver in ALK+ALCL, is implicated in the differential protein expression of Myc between RU and RR cells, even though cells from both cell subsets are known to express a similar level of NPM-ALK proteins. We first tested how NPM-ALK may impact the proteasomal degradation of Myc (or PD-Myc) in HEK293 cells. We found that transfection of NPM-ALK effectively diminished PD-Myc and dramatically increased the Myc protein level. Since NPM-ALK is known to heterodimerize with full-length NPM1, a protein that carries a Myc-binding domain and functions as a key component of the PD-Myc pathway in other cell types, we speculated that the suppression of PD-Myc by NPM-ALK is mediated via the heterodimers formed by NPM-ALK and full-length NPM1 (i.e. NPM-ALK:NPM1). In keeping with this hypothetical model, while shRNA knockdown of full-length NPM1 (by targeting its C-terminal) in HEK293 cells increased the Myc protein level, the same treatment of NPM-ALK-transfected HEK293 cells showed the opposite effects. To further support and validate this model, we extended our studies to purified RU and RR cells derived from SupM2, an ALK+ALCL cell line. In keeping with our model, the degradation half-life of Myc was found to be substantially longer in RR cells as compared to RU cells. Similar to NPM-ALK-transfected HEK293 cells, RR cells showed a dramatic decrease in the Myc protein level upon shRNA knockdown of the full-length NPM1, again suggesting the importance of the NPM-ALK:NPM1 heterodimers in this context. Similarly, siRNA knockdown of ALK in RR cells resulted in substantial decrease in Myc. Immunoprecipitation experiments showed that full-length NPM1 proteins in RU cells form protein complexes that are normal components of PD-Myc, whereas full-length NPM1 in RR cells pulled down a large amount of NPM-ALK, indicating the relative abundance of the NPM-ALK:NPM1 heterodimers in these cells. To validate this model in ALK+ALCL patient samples, we performed immunofluorescence triple staining for full-length NPM1, ALK and Myc, and images of 24 individual ALK+ALCL cells per sample were analyzed using confocal microscopy and the ImageJ software. We found that the pixel density of the fusion signals of NPM-ALK and full-length NPM1 significantly correlated with that of the Myc signals (Pearson correlation analysis, r=0.81; p<0.0001). In conclusion, our findings have provided evidence in support of a novel oncogenic function of NPM-ALK exerted in the nucleus. Specifically, in the small subset of cancer stem-like cells (i.e. RR cells), nuclear NPM-ALK proteins often heterodimerize with full-length NPM1, thereby protecting Myc away from the PD-Myc machinery and inhibiting its degradation. In contrast, in the 'bulk' RU cells, NPM-ALK:NPM1 heterodimers are less abundant, and Myc proteins are unprotected and subjected to degradation. Thus, the control of the NPM-ALK:NPM1 heterodimer formation in the nucleus regulates the Myc protein levels and the degree of cancer stemness in individual tumor cells.

Wang:Abbvie: Honoraria.