Differential binding and internalization of insulin-like growth factor (IGF)-I in cultured human trophoblast and JEG-3 cells: possible modulatory effect of IGF binding proteins (BP)

Abstract

The present study was undertaken to characterize and identify the insulin-like growth factor binding proteins (IGF BPs) secreted by placental cells and their possible modulatory effect on IGF-I binding to cell surface receptors. The experimental approach taken was comparative characterization of binding and internalization of IGF-I and its analog, [Gln³, Ala⁴, Tyr¹⁵, Leu¹⁶ (QAYL)]IGF-I, with reduced affinity for IGF BP, in two different placental cell culture models. One was human placental trophoblast in primary culture and the other, JEG-3 cells, a human choriocarcinoma cell line, representing placental trophoblasts. Binding of [¹²⁵I]IGF-I in both trophoblast and JEG-3 cells was time and temperature dependent. At 37°C, the plateau of [¹²⁵I]IGF-I binding to both the cells (1–2% specific binding per 10⁵ cells) was reached by 40–60 min. At 4°C, the time required to reach the plateau in both cells was increased to ∼4h. The maximum binding of [¹²⁵I]IGF-I to trophoblasts, however, was ∼2 times higher than at 37°C, whereas in JEG-3 cells binding remained the same. Internalization of [¹²⁵I]IGF-I in trophoblast cells was low and temperature independent. At both 37 and 4°C, ≤30% of the total cell-associated [¹²⁵I]IGF-I was internalized. In contrast, internalization of [¹²⁵I]IGF-I in JEG-3 cells was rapid and temperature dependent. At 37°C, ≥60% of the total cell-associated [¹²⁵]IGF-I was internalized by 40–60min. At 4°C, internalization was slow and did not exceed 10% of the total cell-associated radioactivity. Binding of [¹²⁵I-QAYL]IGF-I to trophoblasts, in comparison to [¹²⁵I]IGF-I, was significantly different. The binding was undetectable at 37°C and it was low at 4°C. In JEG-3 cells, however, the binding and internalization of [¹²⁵I]-QAYL]IGF-I at both the temperatures were comparable to that of [¹²⁵I]IGF-I. Further characterization of the two [¹²⁵I]IGF-I bindings to the different placental cells was achieved by binding competition studies using unlabelled IGF-I, [QAYL]IGF-I and [Leu]IGF-I, another analog of IGF-I, [Leu²⁴, 1–62]IGF-I with reduced affinity for the IGF-I receptor, and α-IR3, a monoclonal antibody to the IGF-I receptor. The different potencies of IGF-I and its analogs, and α-IR3 in competing binding of two [¹²⁵I]IGF-Is in the different cells suggested that binding of IGF-I to JEG-3 cells was predominantly to IGF-I receptor, whereas to trophoblast cells it was to IGF BP. This was confirmed by affinity cross-linking studies. The major affinity cross-linked [¹²⁵I]IGF-I complex in trophoblast cells was shown to be a protein of Mr. ∼43 kDa, corresponding to IGF BP-3. In JEG-3 cells, the major cross-linked [¹²⁵I]IGF-I and-[QAYL]IGF-I complexes were proteins of Mr ∼130 kDa and >260 kDa, corresponding to the monomeric and multimeric forms of IGF-I receptor. The ∼43 kDa complex in trophoblast was confirmed to be IGF BP-3 by identification of the characteristics of the IGF BP secreted by trophoblast by Western ligand and immunoblots of the conditioned media. JEG-3 cells did not secrete IGF BP. In conclusion, the membrane associated IGF BP-3 in trophoblast cells, shown here, imply anin vivo modulatory effect of membrane bound IGF BP-3 on IGF-I action in placenta. JEG-3 cells, not secreting IGF-BP, offer an attractive model to study the interactive mechanism of IGF-I and IGF BP-3 actions on the placenta.