Identification of Three Tyrosine Phosphorylation Sites in the p53 Binding Domain of Apoptosis-Stimulating p53 Protein 2 (ASPP2) That Are Differentially Phosphorylated in Response to Chemotherapy in Patient Derived AML Blasts

Abstract 1396

Acute myeloid leukemias (AML) are difficult to treat, and risk-stratification for successful chemotherapy remains a major challenge. Inactivation of the p53 tumor suppressor pathway is a frequent event in many cancers that promotes tumorigenesis and resistance to chemotherapy. However, p53 mutations are rare in AML, and thus the p53-pathway must be inactivated by other mechanisms. ASPP2 is a haploinsufficient tumor suppressor that belongs to a family of p53-binding proteins that enhance apoptosis in part by stimulation of p53-transactivation of selected pro-apoptotic target genes. High ASPP2 expression levels in the absence of p53 mutations thereby argue for proper apoptosis induction capacity and thereby for better response rates. Indeed, low ASPP2 expression levels are correlated with aggressive courses of different tumors. As we have previously shown by qPCR (Kampa-Schittenhelm et al., ASH 2010) and confirm now by intracellular immunostaining in a larger patient cohort, ASPP2 expression levels vary widely in acute leukemias. In vitro silencing of ASPP2 transcription leads to abrogation of induction of apoptosis after application of chemotherapy, arguing for inferior in vivo response rates to therapy of patients lacking ASPP2 expression. Of note, the highest expression levels we have seen was in a patient with good prognosis core binding factor leukemia lacking an autoactivating KIT mutation.

The p53 core domain must interact with the ASPP2 C-terminus to fully stimulate apoptotic function. To further investigate how regulation of the p53-ASPP2 interaction may play a role in apoptosis induction in AML, we identified several highly conserved and highly predicted tyrosine phosphorylation sites at the ASPP2 C-terminus. To study whether these sites modulate the p53-ASPP2 interaction and apoptotic function, we developed phospho-specific antibodies against the three highest-scoring phosphorylation sites and confirmed. tyrosine phosphorylation at Y1029, Y1046 and Y1114 in ex vivo blasts from AML patients. Intriguingly, based on the crystal structure of the p53-ASPP2 complex, phosphorylation of all three tyrosines is predicted to disrupt p53-ASPP2 binding. Tantalizingly, we found that these phosphorylation expression patterns changed after in vitro treatment of native blasts with chemotherapy: blasts treated with daunorubicin revealed an early change of tyrosine phosphorylation patterns. Using these new phospho-specific antibodies, we are continuing to analyze changes in phosphorylation patterns in primary AML blasts (with and without ex vivo chemotherapy) and are performing univariate and multivariate analysis to correlate with available clinical data. Preliminary data suggests that altered ASPP2 tyrosine phosphorylation in AML may play an important role in modulating response to chemotherapy-induced apoptosis in the absence of inactivating p53 mutations.

Ongoing work is prospectively analyzing pY-ASPP2 in patients with acute leukemia during induction chemotherapy. These results aim to evaluate ASPP2 expression as an early-on prediction marker of therapy response in acute leukemia. Further, we aim to provide new and clinically relevant insight into p53 pathway inactivation in acute leukemia – which suggests a novel potential target for therapy to increase the effectiveness of chemotherapy in these patients.