Crystal structure of a fungal polyglycine hydrolase using a RoseTTAFold model

Presented During:

07/31/2022: 4:20 PM - 4:40 PM
Portland Marriott Downtown Waterfront  

Conference:

2022: 72nd ACA Annual Meeting

Session Type:

Oral Talk 

Presenting Author :

Nicole Fraser  
University of Waterloo

Additional Author(s):

Todd Naumann  
United States Department of Agriculture (USDA ARS)
Neil Price  
United States Department of Agriculture (USDA ARS)
David Rose  
Dept of Biology, Univ of Waterloo

Abstract Body:

Chitinase-modifying proteins (CMP) are a subclass of polyglycine hydrolases that are secreted by fungi in response to plant-fungal defense. These CMPs are responsible for inactivating plant chitinases through cleavage of their inter-domain linker region. Our work focused on the structural investigation of a chitinase-modifying protein originating from Fusarium vanettenii. We used a combinatorial approach of in silico and in vitro methods to determine the structure solution. Previous attempts at homology modelling failed to generate a reasonable model, suggesting unique structural features. Owing to recent developments in structure prediction, we were able to generate a full-length homology model using RoseTTAFold1 for our target protein. In combination with a native data set, we were able to solve the structure to 2.2 Angstroms. The structure solution revealed two distinct domains. The amino terminal domain revealed a putatively novel fold, thus explaining previous homology modelling difficulties, whereas the carboxyl terminal domain revealed a beta-lactamase-like region. Further work has focused on the crystallization of the catalytic mutant and co-crystallization of the substrate complex.

1Baek, M., DiMaio, F., Anishchenko, I., Dauparas, J., Ovchinnikov, S., Lee, G. R., Wang, J., Cong, Q., Kinch, L. N., Schaeffer, R. D., Millán, C., Park, H., Adams, C., Glassman, C. R., DeGiovanni, A., Pereira, J. H., Rodrigues, A. V., Dijk, A. A. van, Ebrecht, A. C., … Baker, D. (2021). Accurate prediction of protein structures and interactions using a three-track neural network. Science, 373(6557). https://doi.org/10.1126/science.abj8754

Additional Information - ORAL (2022)

Are you registered for the ACA meeting?

Yes

Which session are you submitting your abstract to?

2.2.1 Random Walks in ML: General applications of Machine Learning for Structural Science