A decline in plasma concentrations of both growth hormone and IGF-I occurs during aging of humans and rodents, and this is accompanied by involution of the thymus gland. Exogenous growth hormone induces the synthesis of IGF-I, which acts on bone marrow-derived hematopoietic progenitors of the myeloid and lymphoid lineages to promote their replication and survival. The increase in survival of these cells is caused by the ability of IGF-I to inhibit their apoptotic death. In contrast to the multipotential colony-stimulating-factor IL-3, inhibition of apoptosis by IGF-I requires the activation of the critical intracellular effector PI 3-kinase. These data establish that hematopoietic progenitors can use more than one intracellular signaling pathway in order to maintain their survival. The data also extend the original hypothesis that IGF-I shares with the colony-stimulating factors the properties of promoting DNA synthesis and inhibiting programmed cell death. Collectively, these data establish that hematopoietic progenitor cells are important targets for IGF-I, and this is likely to be important in understanding thymic aging.

Insulin receptor substrate-2 (IRS-2) is phosphorylated on tyrosine by a number of cytokine receptors and is implicated in the activation of phosphatidylinositol 3′-kinase (PI3-kinase). Here, we demonstrate that induction of granulocytic differentiation of human promyeloid HL-60 cells leads to an increase in the amount of IRS-2 that is phosphorylated in response to insulin-like growth factor (IGF)-I. Although PI3-kinase is often activated following interaction with IRS-1, we could not detect IRS-1 protein, IRS-1 mRNA, or IRS-1-precipitable PI3-kinase enzymatic activity. However, PI3-kinase activity that was coimmunoprecipitated with either anti-phosphotyrosine or anti-IRS-2 following IGF-I stimulation was increased 100-fold. Heightened tyrosine phosphorylation of IRS-2 during granulocytic differentiation was not caused by an increase in expression of the tyrosine kinase IGF-I receptor, as measured by the amount of both the α- and β-subunits. Instead, immunoblotting experiments with an Ab to IRS-2 revealed that induction of granulocytic differentiation caused a large increase in IRS-2, and this occurred in the absence of detectable IRS-1 protein. These IRS-2-positive cells could not differentiate into more mature myeloid cells in serum-free medium unless IGF-I was added. These data are consistent with a model of granulocytic differentiation that requires at least two signals, the first of which leads to an increase in the cytoplasmic pool of IRS-2 protein and a second molecule that acts to tyrosine phosphorylate IRS-2 and enhance granulocytic differentiation.

Flow cytometry was used to examine the expression of type I insulin-like growth factor receptors (IGF-IR) on three types of human hematopoietic cells that represent different stages of myeloid lineage development. Both HL-60 (promyeloid) and U-937 (monocytic) cells express abundant IGF-IR protein (> 79% cells positive for the IGF-IR), whereas KG-1 myeloblasts express negligible levels of IGF-IR (< 1% IGF-IR-positive cells). Exogenous IGF-I, IGF-II, and an IGF-I analog that binds poorly to IGF-binding protein-3 (des-IGF-I) increased DNA synthesis of HL-60 and U-937 cells in a dose-dependent (1-25 ng/ml) fashion by 2- to 4-fold in serum-free medium, whereas KG-1 cells did not respond to any of these growth factors. The IGF-induced increase in proliferation of HL-60 promyeloid cells was inhibited by soluble IGF-binding protein-3 (500 ng/ml) when these cells were stimulated with 10 ng/ml of either IGF-I (53 +/- 8%) or IGF-II (59 +/- 8%), but not with des-IGF-I (3 +/- 1%). In contrast, the anti-IGF-IR monoclonal antibody (mAb; alpha IR-3) inhibited the DNA synthesis caused by 10 ng/ml exogenous IGF-I (67 +/- 6%), IGF-II (72 +/- 8%), and des-IGF-1 (82 +/- 9%). Proliferation of KG-1 myeloblasts, however, was neither stimulated by the IGFs nor inhibited by the anti-IGF-IR mAb. In the absence of exogenous IGF-I, the mAb directed against the IGF-IR significantly suppressed basal DNA synthesis of HL-60 promyeloid (72 +/- 5%) and U-937 monocytic (39 +/- 7%) cells, but did not affect DNA synthesis of KG-1 myeloblasts (8 +/- 1%) compared to an isotype-matched control mAb. Similarly, the alpha IR-3 mAb abrogated vitamin D3-induced differentiation of the HL-60 cells into macrophages in serum-free medium, as assessed by expression of the leucam surface protein, CD11b. As the alpha IR-3 mAb inhibits DNA synthesis in the presence and absence of exogenous IGF-I on receptor-bearing cells, but not IGF-IR-negative cells, these data demonstrate that both endocrine and autocrine IGF-I are potent growth factors in human myeloid cells where expression of the surface receptor, rather than the ligand, is the critical control element. More importantly, these data support the hypothesis that autocrine IGF-I may play a significant role in the differentiation of promyeloid cells into macrophages.