Abstract
Neuroblastoma is the most common extra-cranial solid tumor in children. Its broad spectrum of clinical outcomes reflects the underlying inherent cellular heterogeneity. As current treatments often do not lead to tumor eradication, there is a need to better define therapy-resistant neuroblastoma and to identify new modulatory molecules. To this end, we performed the first comprehensive flow cytometric characterization of surface molecule expression in neuroblastoma cell lines. Exploiting an established clustering algorithm (SPADE) for unbiased visualization of cellular subsets, we conducted a multiwell screen for small molecule modulators of neuroblastoma phenotype. In addition to SH-SY5Y cells, the SH-EP, BE(2)-M17 and Kelly lines were included in follow-up analysis as in vitro models of neuroblastoma. A combinatorial detection of glycoprotein epitopes (CD15, CD24, CD44, CD57, TrkA) and the chemokine receptor CXCR4 (CD184) enabled the quantitative identification of SPADE-defined clusters differentially responding to small molecules. Exposure to bone morphogenetic protein (BMP)-4 was found to enhance a TrkAhigh/CD15-/CD184- neuroblastoma cellular subset, accompanied by a reduction in doublecortin-positive neuroblasts and of NMYC protein expression in SH-SY5Y cells. Beyond yielding novel marker candidates for studying neuroblastoma pathology, our approach may provide tools for improved pharmacological screens towards developing novel avenues of neuroblastoma diagnosis and treatment.