Abstract

Colony Stimulating Factor-1 (CSF-1) acts through the CSF-1R to regulate proliferation, survival and differentiation of mononuclear phagocytes (MNPs). GAB2, a scaffolding protein, modulates signals from numerous receptors, through recruitment of phosphatidylinositol 3-kinase (PI3K) and SHP2 phosphatase. Previously we reported at this meeting (

Lee et al.,
Blood
110
:
2197
,
2007
) that under steady conditions, GAB2−/− mice showed a significant reduction in the number of bone marrow (BM) tissue macrophages (Mφs) and dramatic decreases in the number of CSF-1 dependent colony forming units (CFU-Cs) from total BM and from BM cells that have been induced to commit to the MNP lineage. We also showed by retroviral transduction that GAB2 relies on the recruitment of PI3K and an unidentified pathway to promote CFU-C formation. These findings raised additional questions:

  1. What is the identity of the BM progenitor that GAB2 regulates?

  2. What is the identity of the second pathway?

  3. Does GAB2 have a role in MNP recruitment under in vivo situations of increased MNP demands, e.g. in acute inflammation?

To address these questions, we developed methods to isolate BM progenitors committed to the MNP lineage and flow cytometric assays to monitor intracellular phosphorylation in complex BM populations. Based on our CFU-C results, we predicted that GAB2 would have a major role in the expansion of the CD31highLy6C subset, which should contain CSF-1 responsive early progenitors. We enriched for these cells from ex vivo expanded day 2 BM cultures by magnetic bead depletion of lymphocytes, granulocytes and erythroid cells. The LIN*CD31highLy6C subset showed a ≈ 30-fold enrichment of CFU-Cs compared to total BM. Those from Gab2−/− mice gave rise to ≈ 4-fold fewer CFU-Cs and exhibited increased apoptosis and diminished cell cycle progression. To verify that our findings also applied to MNP progenitors isolated directly from the BM, we purified early myeloid progenitors that can proliferate and differentiate in response to CSF- 1. These are lineage depleted (LIN) cKit+CD31highLy6C (LK31C) cells and they showed a 51-fold enrichment of CFU-Cs compared to total BM. LK31C cells from GAB2−/− mice also showed a 9-fold reduction in CFU-Cs, relative to GAB2+/+ cells. By CFSE tracking, LK31C cells from GAB2−/− mice demonstrated diminished proliferation. Thus, GAB2 is required for CSF-1 dependent expansion of CD31highLy6C BM progenitors. To biochemically identify GAB2-regulated signaling pathways that promote expansion of CSF-1 responsive progenitors, we utilized flow cytometry to assay intracellular phosphorylation in day 2 LIN* cells. We found that the percentage of responding cells as well as the level of phosphorylation of Akt, Erk and S6 were significantly diminished in the LIN*CD31highLy6C subset from GAB2−/− cells. We also found that optimal S6 phosphorylation was dependent on both the PI3K and Erk pathway, indicating that Gab2 could play an important role in protein translation in CSF-1 responsive progenitors. Since GAB2’s recruitment of SHP2 may be important for optimal Erk activation, we investigated the role of GAB2-SHP2 in CFU-C formation. When a GAB2 mutant that cannot recruit SHP2 was overexpressed in GAB2−/− day 2 BM cells, the number of CFU-Cs was only 40% of that obtained with wildtype GAB2. Together with our previous findings, they indicate that both the PI3K and SHP2-Erk pathways recruited by GAB2 were required for optimal CFU-C formation. Since Gab2−/− mice have a significant defect in CSF-1 dependent MNP development, we reasoned that this deficit could be exacerbated in inflammatory conditions with a prominent MNP infiltration. To test this hypothesis, we used a well-established sterile peritonitis model. GAB2−/− mice showed a 75% (p=0.003) and 67% (p=0.005) reduction in recruited Mφs at 24h and 72h respectively after thioglycollate (TG) challenge. Moreover, there was a 68% increase in CFU-Cs from GAB2+/+ BM cells at 24h (p=0.0001) post TG challenge, whereas GAB2−/− BM cells showed no change. These findings indicate that GAB2−/− mice are much less able to elicit a MNP response after an inflammatory challenge. Altogether, our results reveal that GAB2 utilizes both PI3K-Akt and SHP2-Erk to maximally promote CSF-1 mediated expansion along the MNP lineage by targeting the lineage depleted CD31highLy6C subset. In the absence of GAB2, the BM is unable to respond maximally to inflammatory challenges.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author