Evidence for the potentiation of epidermal growth factor receptor tyrosine kinase activity by association with the detergent-insoluble cellular cytoskeleton: analysis of intact and carboxy-terminally truncated receptors

Abstract

The ligand-stimulated tyrosine kinase activity of the normal human epidermal growth factor (EGF) receptor and a truncated EGF receptor lacking 164 carboxy-terminal (C-terminal) amino acids was examined in intact cells and after Triton X-100 extraction into Triton-soluble and -insoluble (cytoskeletal) preparations. Detergent extraction of the intact and truncated receptors appeared complete using 0.3% Triton as demonstrated by anti-EGF receptor immunoblots, tyrosine kinase assays, and marker enzyme (alkaline phosphatase) solubilization. Higher Triton concentrations yielded no additional EGF receptor extraction and began to inhibit EGF-stimulated kinase activity toward angiotensin II (AII). Furthermore, the tyrosine kinase activity of the truncated EGF receptor exhibited increased sensitivity to Triton extraction, suggesting a lower affinity or a more labile association of this receptor with the cytoskeleton. However, both EGF receptor forms had altered catalytic activity when associated with the cytoskeletal fraction, as evidenced by the increased phosphorylation of the exogenous substrates: AII, src-peptide, and [Val5]AII. Kinetic analyses of both receptor types revealed that the cytoskeletal fractions obtained using 0.3% Triton contain EGF receptor activity that exhibits a Michaelis-Menten constant (Km) for AII that is 2- to 3-fold more favorable than that calculated for the soluble receptor forms. EGF treatment of intact cells containing either the intact or truncated receptor revealed similar phosphorylated proteins in the soluble fraction of both cell types, although there was evidence for the enhanced phosphorylation of certain proteins (e.g. 115 and 50 kilodalton proteins) in cells containing the truncated receptor. There was also a greater number of tyrosine-phosphorylated proteins in the Triton-insoluble fraction of cells containing the truncated receptor, suggesting an altered specificity of this receptor toward selected cytoskeletal proteins. This work indicates that EGF receptor-cytoskeletal interaction may be an important consideration in the control of receptor-kinase activity and has examined the detergent sensitivity of this association. These studies also suggest that the C-terminal domain of the EGF receptor may affect cytoskeletal interaction in addition to influencing the receptor's catalytic capacity.