Determination of the three-dimensional solution structure of the C-terminal domain of cellobiohydrolase I from Trichoderma reesei. A study using nuclear magnetic resonance and hybrid distance geometry-dynamical simulated annealing

Abstract

The solution structure of a synthetic 36-residue polypeptide comprising the C-terminal cellulose binding domain of cellobiohydrolase I (CT-CBH I) from Trichoderma reesei was investigated by nuclear magnetic resonance (NMR) spectroscopy. The 1H NMR spectrum was completely assigned in a sequential manner by two-dimensional NMR techniques. A large number of stereospecific assignments for .beta.-methylene protons, as well as ranges for the .vphi., .psi. and .chi.1 torsion angles, were obtained on the basis of sequential and intraresidue nuclear Overhauser enhancement (NOE) and coupling constant data in combination with a conformational data base search. The structure calculations were carried out in an iterative manner by using the hybrid distance geometry-dynamical simulated annealing method. This involved computing a series of initial structures from a subset of the experimental data in order to resolve ambiguities in the assignments of some NOE cross-peaks arising from chemical shift degeneracy. Additionally, this permitted us to extend the stereospecific assignments to the .alpha.-methylene protons of glycine using information on .vphi. torsion angles derived from the initial structure calculations. The final experimental data set consisted of 554 interproton distance restraints, 24 restraints for 12 hydrogen bonds, and 33 .vphi., 24 .psi., and 25 .chi.1 torsion angle restraints. CT-CBH I has two disulfide bridges whose pairing was previously unknown. Analysis of structures calculated with all three possible combinations of disulfide bonds, as well as without disulfide bonds, indicated that the correct disulfide bridge pairing was 8-25 and 19-35. Forty-one structures were computed with the 8-25 and 19-25 disulfide bridges, and the average atomic rms difference between the individual structures and the mean structure obtained by averaging their coordinates was 0.33 .+-. 0.04 .ANG. for the backbone atoms and 0.52 .+-. 0.06 .ANG. for all atoms. The protein has a wedgelike shape with an amphiphilic character, one face being predominantly hydrophilic and the other mainly hydrophobic. The principal elements of secondary structure is made up of an irregular triple-stranded antiparallel .beta.-sheet composed of residues 5-9 (.beta.1), 24-28 (.beta.2), and 33-36 (.beta.3) in which strand .beta.3 is hydrogen bonded to the other two strands. Strand .beta.1 is preceded by and overlaps with a type II turn (residues 3-6), and strands .beta.2 and .beta.3 are connected by a type I turn (residues 29-32). Strands .beta.1 and .beta.2 are connected by a loop (residues 14-19) via type II turns (residues 10-13 and 20-23). A number of side-chain interactions are discussed in the light of structure.