INTRODUCTION. The importance of IL-7R signaling has been thoroughly studied in the context of both normal T-cell development and leukemogenesis. Upon IL-7 induced heterodimerization of the IL-7Rα and the γc chain, downstream signaling routes regulate survival and proliferation of differentiating T-cells. JAK-STAT, MAPK-ERK and PI3K-AKT pathways are aberrantly activated in T-cell acute lymphoblastic leukemia (T-ALL) patients by recurrent gain-of-function mutations of the IL-7R signaling pathway. IL-7R signaling mutations are a result of mutations in IL7Ra, JAK1/3, PTPN11, NF1, STAT5B, N/KRAS, PI3K or AKT that are found in nearly 35% of pediatric T-ALL patients and are associated with steroid resistance and inferior event-free survival. Here we focus on downstream AKT activation in relation to cellular transformation and steroid resistance. We previously demonstrated that AKTE17K transforms IL3-dependent BaF3 cells in contrast to wild type AKT, while wild type AKT drives steroid resistance in SUP-T1 T-ALL cells in contrast to mutant AKTE17K.

AIM. Our aim was to identify functional differences between AKT and AKTE17K molecules in respect to cellular transformation and steroid resistance. We assessed activation level, cellular localization and downstream signaling events that may explain these functional differences.

RESULTS. For this study, we used doxycycline-inducible SUP-T1 cell lines that overexpress wild type AKT or mutant AKTE17K molecules. Using immunohistochemistry, we revealed abundant cytoplasmic localization of wild type AKT but predominant plasma membrane localization of AKTE17K and extends previous findings by others. Furthermore, we found evidence for phosphorylation of wild type AKT on serine-473 (S473) and threonine-308 (T308) residues in both cytoplasmic and nuclear fractions, indicating that part of induced AKT becomes activated. In contrast, expression of AKTE17K was accompanied with abundant phosphorylation of T308 while having relatively low levels of S473 phosphorylation. The T308/S473 phosphorylation ratio was 5 times higher for AKTE17K compared to wild type AKT. These results are in line with current views on AKT activation that S473 phosphorylation facilitates the opening of the AKT pleckstrin homology domain and membrane docking, after which AKT is phosphorylated (activated) on the T308 residue by PDK1. The constitutive plasma membrane localization and open conformation of AKTE17K as a result of its E17K pleckstrin homology domain mutation leaves it independent of S473 phosphorylation and results in enhanced T308 phosphorylation by PDK1. Consequently, AKTE17K mutant cells may exhibit reduced cellular sensitivity to allosteric AKT inhibitors but not to type I ATP-competitive inhibitors. Indeed, SUP-T1 cells expressing wild type AKT or mutant AKTE17K are equally sensitivity to the ATP-competitive AZD5363 inhibitor, but AKTE17K cells display reduced sensitivity to the MK2206 allosteric inhibitor compared to AKT wild type cells.

In an attempt to explain enhanced steroid resistance by expression of wild type AKT but not mutant AKTE17K, we found no differences in AKT-mediated S134 phosphorylation of the NR3C1 steroid receptor or impaired nuclear import of NR3C1 following steroid treatment (Piovan et al., Cancer Cell 2013). This indicates that AKT induces steroid resistance by other mechanisms in our SUP-T1 cell system.

CONCLUSIONS. In contrast to wild type AKT overexpression, we demonstrated that AKTE17K has (i) transforming activity, (ii) does not yield cellular resistance to steroid treatment, (iii) is predominantly localized at the plasma membrane, (iv) has high T308 over S473 phosphorylation and (v) is less sensitive to the MK2206 allosteric AKT-inhibitor. These observed differences are likely caused by the constitutively open AKTE17K conformation. Since no differences in NR3C1 phosphorylation or nuclear shuttling were observed in our model, we hypothesize other mechanisms underlie the provoked steroid resistance by AKT wild type overexpression. To identify these novel mechanisms, we are currently performing an unbiased phospho-protein array to identify differentially phosphorylated AKT substrates.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.