Photochemical localization of the semotiadil binding region within the cardiac Ca2+ channel alpha1 subunit. Comparison with the skeletal muscle counterpart

[Identification of the ligand binding sites and novel drug target molecules by photoaffinity labeling]

Photoaffinity labeling is a useful and reliable method for 1) the identification of the ligand-target receptor and 2) the structural investigation of its binding site. Using photoaffinity labeling techniques, the binding sites of four typical calcium antagonists, 1,4-dihydropyridines, benzothiazepines, phenylalkylamines, and benzothiazines, were successfully identified within the primary structure of the skeletal muscle calcium channels. The results confirm pharmacological observations of the four antagonists, which had been proposed to interact allosterically with each other. Secondarily we demonstrated that human glutathione S-transferase class pi (GST pi) is specifically photolabeled by the antidiabetic agent sulfonylurea glibenclamide (GB) and it also inhibits the enzyme activities of glutathione conjugation by GB in a competitive manner for glutathione. These results indicate that GST pi is another target molecule of sulfonylurea since a subunit of ATP-sensitive potassium channels is well known to be a sulfonylurea receptor. This review focuses on photoaffinity labeling techniques as a useful tool for drug discovery and development.

Efficient identification of photolabelled amino acid residues by combining immunoaffinity purification with MS: revealing the semotiadil-binding site and its relevance to binding sites for myristates in domain III of human serum albumin

To identify photoaffinity-labelled amino acid residue(s), we devised an effective method utilizing immunoaffinity purification of photolabelled fragments, followed by matrix-assisted laser-desorption ionization-time of flight (MALDI-TOF) MS and nanoelectrospray ionization tandem MS (nano-ESI-MS/MS) analysis. Human serum albumin (HSA) was photolabelled with an azidophenyl derivative of semotiadil, FNAK [(+)-(R)-3,4-dihydro-2-[5-methoxy-2-[3-[N-methyl-N-[2-(3-azidophenoxy)-ethyl]amino]propoxyl]phenyl]-4-methyl-2H-1,4-benzothiazin-3-(4H)-one], since HSA is a major binding protein for semotiadil in serum. After lysyl endopeptidase digestion, photolabelled HSA fragments were adsorbed selectively on to Sepharose beads on which an anti-semotiadil antibody was immobilized, and fractions were eluted quantitatively by 50% acetonitrile/10 mM HCl. MALDI-TOF MS analysis of the eluted fraction showed that it contained two photolabelled fragments of m/z 2557.54 (major) and 1322.44 (minor), corresponding to Lys-414-Lys-432 and Ala-539-Lys-545, respectively. Further nano-ESI-MS/MS analysis revealed that Lys-414 was the photolabelled amino acid residue in fragment 414-432 and Lys-541 was a likely candidate in fragment 539-545. Based on the photolabelling results, we constructed a three-dimensional model of the FNAK-HSA complex, revealing that FNAK resides in a pocket that overlaps considerably with myristate (Myr)-binding sites, Myr-3 and -4, by comparison with crystallographic data of HSA-Myr complexes described in Curry, Mandelkow, Brick and Franks (1998) Nat. Struct. Biol. 5, 827-835. Moreover, addition of Myr increased photo-incorporation into Lys-414, whereas incorporation into Lys-541 decreased under conditions of [Myr]/[HSA]<1. Further addition of Myr, however, uniformly decreased photo-incorporation into both Lys residues. These results indicate that FNAK labelling can also be used to monitor Myr binding in domain III. An interpretation for the concomitant local conformational change of HSA is provided.