Structural and functional studies of fungal cellulose-binding domains by NMR spectroscopy: Dissertation

Cellulose has a important environmental role in the preservation of the global carbon cycle and commercial significance as a raw material for industry. To understand the biodegradation of cellulose on the atomic level, it is important to be able to relate enzyme activities with the three-dimensional (3D) structures of cellulases and cellulose. The cellulolytic system of the fungus Trichoderma reesei is one of the best understood of all cellulolytic systems. Most cellulases from T. reesei have a cellulose-binding domain (CBD) specialised for binding to cellulose. As a means to understanding the interactions between cellulose and CBDs of cellobiohydrolase I (CBHI) and endoglucanase I (EGI) from T. reesei, 3D structures of these domains were studied by two-dimensional (2D) 1H NMR (nuclear magnetic resonance) techniques. Structural effects of mutations of conserved amino acids in CBDCBHI, Y5A, P16R, N29A, Y31A, Y32A and Q34A, were evaluated by comparing chemical shifts, coupling constants and NOEs of the backbone protons of the mutants and wild-type CBDCBHI. In general, the substitutions did not alter significantly the secondary structures of these engineered peptides. The adsorption experiments on cellulose showed that Y5A, Y31A and Y32A had lost nearly all their affinity to cellulose. For Y31A and Y32A the 3D structures revealed small local changes around the mutation on the flat face of CBD, which was expected to bind to cellulose. Therefore the structural roles of Tyr-31 and Tyr-32 must be minor, but their functional importance is clear since the mutants lacking these residues did not bind strongly to cellulose. In the case of Y5A the disruption of the structural framework at the N-terminus and the complete loss of affinity to cellulose implied that Tyr-5 has both structural and functional significance. The 3D structure of a synthetic CBDEGI was also determined by NMR spectroscopy. The structure was very similar to that of wild-type CBDCBHI. Within the precision of the structures, even the cellulose-binding face of CBDEGI was similar to that of CBDCBHI, apart from the place of attachment of the different side-chain. The determined NMR structure was also in agreement with an earlier modelled structure of CBDEGI. Finally, soluble cello-oligosaccharides were used as model compounds for cellobiose chains to investigate the interaction between CBD and cellulose by NMR spectroscopy. CBDs caused line broadening effects and decreasing T2 relaxation times for certain sugar resonances, whereas there were no effects in the presence of a mutant that bound weakly to cellulose. Experiments showed that the interactions between CBD and cellobiose units of sugars are specific, supporting the model presented for the CBD binding to crystalline cellulose. It remained uncertain, however, how well the cello-oligosaccharides mimicked the binding of CBD to cellulose.