Topology of an amiloride-binding protein

Abstract

Amiloride and structurally related compounds inhibit many transport proteins, enzymes, and drug or hormone receptors, although the topology of amiloride binding sites on these proteins has not been defined. We have previously raised and characterized a monoclonal antiamiloride antibody (mAb BA7.1) which is similar to epithelial Na+ channels in its specificity of binding of amiloride and amiloride analogs, suggesting that their amiloride binding sites may be similar in topology, mAb BA7.1 was used as a model system to analyze the three-dimensional conformation of an amiloride binding site. The photoactive amiloride analog 2'-methoxy-5'-nitrobenzamil specifically labeled the heavy chain of mAb BA7.1, suggesting that the heavy chain participates in amiloride binding. The nucleotide sequences of the variable regions of the heavy and light chains of mAb BA7.1 were determined and amino acid sequences deduced to analyze the structure of the amiloride binding site. A comparative modeling approach was used to construct a model of the amiloride binding domain of mAb BA7.1, and a docking procedure was used to place amiloride within this domain. The model indicated that planar aromatic amino acid resides form a pocket into which amiloride, a planar molecule, inserts. Constraints on amiloride binding predicted by this model correlated with the measured specificity of binding of amiloride analogs with mAb BA7.1. These results provide a potential guide for the identification of motifs or amino acid contact residues present within other amiloride-sensitive proteins.