Single-Cell Phospho-Profiling in Pediatric B-Cell Precursor Acute Lymphoblastic Leukemia Reveals Signaling Differences in Cytogenetic and Prognostic Subgroups.

Abstract 2653

Poster Board II-629

Oncogenesis and tumor progression are supported by alterations in cellular signaling. We used phospho-specific antibodies in combination with surface staining in flow cytometry to analyze specific signaling profiles of leukemia cells at a single cell level. We anayzed 22 xenograft samples derived from NOD/SCID-mice transplanted with primary pediatric B- cell precursor acute lymphoblastic leukemia (BCP- ALL) cells. The cells were isolated from the spleens of leukemia bearing mice and stimulated ex vivo in vitro with different stimulants and cytokines. Activation of various phosphoepitopes was analyzed by flow cytometry and compared to the basal state of unstimulated samples. TEL/AML1 fusion and MLL-rearrangements are the most common cytogenetic aberrations in childhood BCP- ALL and are associated with a good or very poor prognosis, respectively. Although there were no differences detectable in basal phosphorylation between the different cytogenetic subgroups, TEL/AML1- positive samples (n= 5) displayed a significantly lower phosphorylation of extracellular regulated kinase (ERK1/2) after stimulation with PMA (Phorbol-12-myristat-13-acetate, activator of protein kinase C) or interleukin 7 (IL-7), while they showed a significantly higher activation of p38 after stimulation with PMA, compared to samples without translocation (n= 13). Additionally, the fusion gene negative samples showed a downregulation of STAT1-phosphorylation after stimulation with interleukin 10 (IL-10) whereas the TEL/AML1-positive samples showed no change. Interestingly, the MLL- positive samples (n= 3) also did not show a difference in STAT1-phosphorylation after IL-10, but showed significantly stronger STAT1 activation in response to interferon alpha (IFN-a) compared to samples without fusion genes. Moreover, the MLL- positive samples also displayed a weaker reaction in ERK-phosphorylation after IL-7 compared to the leukemia samples without cytogenetic aberrations. Differences in other prognostic subgroups analysed include a weaker phosphorylation of p38 and JNK after anisomycin in samples where the patient initially presented with hyperleucocytosis (> 100.000 WBC/μl) (n= 3), an indicator of poor prognosis. A decrease in STAT3- activation after IL-10 was observed in samples where the patients displayed bone marrow remission on day 15 of therapy (n= 8), compared to no change in the samples of patients with > 5% residual blasts (n= 8), indicative of therapy resistance, at this timepoint. Similar to the results for the cytogenetic subgroups, there were no differences detectable at basal phosphorylation levels between the prognostic subgroups. Taken together, these data show that basal phosphorylation states of specific signaling molecules do not discriminate between the different prognostic subgroups of BCP- ALL analyzed. Cytogenetic and other prognostic subgroups however display specific profiles of signaling networks after stimulation. This strategy will prove helpful to identify mechanisms by which different subgroups with distinct clinical outcomes interpret environmental signals and hereby define pathways important for continued survival, proliferation and resistance eventually leading to novel biomarkers and targeted therapies.