Poster Board III-904
Juvenile Myelomonocytic Leukemia (JMML) is a lethal myeloproliferative disorder (MPD) of children, characterized by hyperproliferation of myelomonocytic cells and hypersensitivity to Granulocyte-Monocyte Colony-Stimulating Factor (GM-CSF). Most patients show hyperactivation of Ras via one or more mutations, including in the PTPN11 gene, which encodes the protein tyrosine phosphatase, Shp2. It has been demonstrated that gain-of-function mutant Shp2 (Shp2 E76K and Shp2 D61Y) causes hyperactivation of the Mitogen-Activated Protein Kinase (MAPK) pathway. Additionally, we have previously shown that macrophage progenitor cells transduced with Shp2 D61Y or Shp2 E76K showed elevated levels of phospho-Akt, both at baseline and following 1 hour of GM-CSF stimulation, indicating a role for the Phospho-Inositol-3-Kinase (PI3K)/Akt pathway in the phenotype of elevated proliferation and survival in mutant Shp2-expressing cells (Yang, et al, 2008). However, it remains to be elucidated how PI3K contributes to the phenotype of increased proliferation and survival in cells bearing gain-of-function mutations in Shp2. Class IA PI3K is a lipid kinase heterodimer consisting of one of three catalytic subunits (p110α, p110β, or p110δ) and one of two regulatory subunits (p85α or p85β). It has been demonstrated that knocking out the main regulatory subunit, p85α, abrogated the hyperproliferative phenotype in mast cell progenitors bearing an oncogenic mutation in Kit in a model of another MPD, Systemic Mastocytosis (Munugalavadla, et al, 2007). In order to examine whether eliminating expression of p85α would cause a similar correction in cells expressing gain-of-function mutant Shp2, we performed timed matings of mice heterozygous for the knock-out of Pik3r1, which encodes the p85α subunit as well as its isoforms, p55α and p50α, since homozygous Pik3r1-/- is lethal in utero. WT and Pik3r1-/- fetal liver cells were isolated from 14.5 day embryos and transduced with either WT Shp2 or mutant Shp2 E76K. Transduced cells were subjected to serum deprivation followed by a 24-hour treatment with increasing doses of GM-CSF, and proliferation was then measured with H3-thymidine incorporation assays. We found that the absence of all the Pik3r1 isoforms resulted in a significant but incomplete correction of GM-CSF hypersensitivity in Shp2 E76K-expressing cells. To further investigate this observation, WT Pik3r1 and Pik3r1-/- macrophage progenitors, transduced with WT Shp2 or mutant Shp2 E76K as described above, were serum- and growth factor-deprived and then stimulated for 1 hour with GM-CSF. Western blot analysis showed that there was an expected increase in phospho-Akt in WT Pik3r1 cells following GM-CSF stimulation and that this increase was larger in Shp2 E76K-expressing cells than in WT Shp2-expressing cells, as previously observed. Upon genetic disruption of Pik3r1, Akt activation was reduced in both WT Shp2- and Shp2 E76K-expressing cells; however, the phospho-Akt in the Shp2 E76K-expressing cells was not reduced to WT levels. The phospho-Akt levels mirrored the proliferation pattern displayed by these cells in the H3-thymidine incorporation assays, where a modest reduction in proliferation in Pik3r1-/-, Shp2 E76K cells corresponded to the modest reduction in phospho-Akt levels in the same cells. Additionally, we found that Pik3r1-/-, Shp2 E76K cells also showed a blunted increase in phospho-Erk levels following GM-CSF stimulation compared to the WT Pik3r1, Shp2 E76K cells, indicating that knocking out Pik3r1 affects the MAPK pathway as well, which likely also contributes to the reduction in proliferation seen in Pik3r1-/-, Shp2 E76K cells following GM-CSF stimulation. Based on these data, we conclude that: (1) gain-of-function Shp2 activity results in dysregulated PI3K signaling, contributing to the leukemic phenotype of increased proliferation and survival; (2) PI3K signaling is reduced but not completely normalized in the absence of the main regulatory subunit, p85α and its isoforms, in gain-of-function mutant Shp2-expressing cells; and (3) there is cross-talk between the PI3K and MAPK pathways in the presence of Shp2 activating mutations, which is revealed by knocking out Pik3r1.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.