USE OF FACTORIAL EXPERIMENTAL-DESIGN TO DELINEATE THE STRONG CALCIUM-DEPENDENT AND PH-DEPENDENT CHANGES IN BINDING OF HUMAN SURFACTANT PROTEIN-A TO NEUTRAL GLYCOSPHINGOLIPIDS - A MODEL FOR STUDIES OF PROTEIN CARBOHYDRATE INTERACTIONS

Human surfactant protein-A (SP-A) is a C-type lectin belonging to the collection supergroup of mammalian lectins. It is produced by alveolar type II cells and has been shown to bind to lactosylceramide (R. A. C hilds, J. R. Wright, G. F. Boss, C-T Yuen, A. M. Lawson, C. Chai, K. D rickamer, and T. Feizi (1992) J. Biol. Chem. 287, 9972-9979), galactos ylceramide, and gangliotriaosylceramide (Y. Kuroki, S. Gasa, Y. Ogasaw ara, A. Makita, and T. Akino (1992) Arch. Biochem. Biophys. 299, 261-2 67), To evaluate the pH- and calcium-dependent binding of SP-A to neut ral glycosphingolipids we employed a microtiter plate technique and pe rformed a series of full factorial design experiments using lactosylce ramide, galactosylceramide, and gangliotetraosylceramide and native no nderivatized SPA. The optimal binding conditions were drastically diff erent for these three receptors. At pH 7.4 and at 5 mM Ca concentratio n the binding affinity of SP-A followed the order galactosylceramide > lactosylceramide > gangliotetraosylceramide. However, this was close to optimal conditions for binding of SP-A to lactosylceramide and at m inimum for binding to gangliotetraosylceramide. Binding of SP-A to gal actosylceramide was relatively insensitive to pH but increased signifi cantly with increasing Ca concentrations. These experiments using fact orial experimental design emphasize the importance of critical interpr etation of all earlier studies on protein-carbohydrate interactions es pecially when transferring experimental data into complex biological s ystems. (C) 1995 Academic Press, Inc.