Analysis of Differentially Activated Signaling Pathways in Myeloproliferative Disease Using Kinomics Chip Profiling

In a multitude of cases, oncogenic mutations are gain of function mutations that confer a constitutively activated gene product. Currently, evidence from a large body of experimental studies suggests that oncogenic transformation induced by activating kinase mutations is not sufficiently explained by constitutive kinase activation alone but is a result of aberrantly activated signaling pathways in affected cells. The JAK2V617F-mutation is a highly prevalent molecular marker in Ph-negative myeloproliferative disease (MPD). In vitro, Ba/F3-cells expressing both erythropoietin receptor (EpoR) and the JAK2V617F-mutation show constitutive activation of the JAK-STAT pathway and cytokine independent growth. Multiple in-vitro and in-vivo studies demonstrate that the JAK2V617F-mutation mediates many of the phenotypic characteristics of the MPDs. Nevertheless, until now it is largely unclear, which signaling pathways in particular are involved in the process of malignant transformation in JAK2V617F-positive cells. Therefore, we applied a kinomics chip approach to screen for activated intracellular signaling pathways. We used a commercially available peptide chip containing 960 synthetic kinase substrate peptides spotted in triplicates and covering peptide substrates for approximately 50% of the human kinome. Peptides have been selected for their biologic relevance in physiologic processes such as stress response, growth and cell differentiation. With this approach, a broad spectrum of intracellular signaling pathways and kinases can be investigated simultaneously in a single experiment.

As a proof of principle, we performed kinomics chip analysis of Ba/F3-cells stably transfected with EpoR and either the JAK2V617F-mutant (Ba/F3-EpoR-VF) or wildtype JAK2 (Ba/F3-EpoR-WT). In brief, cells were seeded and treated with erythropoietin. One chip per cell lysate was incubated with an activation buffer containing the cell lysate and radioactively labelled ATP for a defined time period and washed several times afterwards. The chip was then analyzed by means of autoradiography using a phospho-storage-screen and a phospho-imager. Chip analysis was performed using standard microarray software and Microsoft Excel software. Chip experiments were performed simultaneously for Ba/F3-EpoR-VF and Ba/F3-EpoR-WT and in duplicate.

Analysis revealed differential activation of known pathways such as Ras/Raf/MEK/ERK, JAK/STAT, and PI3-Kinase with pronounced activation seen in Ba/F3-EpoR-VF cells as compared to Ba/F3-EpoR-WT cells. This was not a surprising result but strongly underlines the feasibility and validity of this approach and therefore served as an internal control. Differential regulation of a number of other signaling nodes that have not yet been described in the context of mutant JAK2 signaltransduction have been detected. To select for relevant hits among these potential targets, we first excluded all substrates from further analysis that are known to be involved in lymphocyte-specific pathways. For the remaining hits we performed a literature search to learn more about their known functions and their potential impact in JAK2V617F-positive MPD. Validation of selected signaling molecules by means of Western blotting analysis and functional investigations such as siRNA knock-down experiments are currently under way. In addition to the widely used lymphoid Ba/F3-model, we also established a novel cell culture model with simultaneous expression of EpoR and either mutant or wildtype JAK2 in a myeloid 32D-cell background. This model will be helpful to us to determine false-positive results due to cell-line specific changes.

We conclude, that kinomic profiling using the above mentioned chip-technology is a valid method to comprehensively investigate differential activation of signaling pathways in cell lysates. In our cell line model, we were able to detect activation of well known signaling pathways in JAK2V617F-positive cells. Furthermore, we were able to identify candidate proteins that appear to be specifically involved in JAK2V617F-signaling.