Atomic resolution structures of S1 nuclease complexes reveal details of RNA interaction with enzyme in spite of unusual crystal defect

K. Adámková^1,2, T. Kovaľ¹, L. H. Østergaard³, J. Dohnálek¹

¹Institute of Biotechnology of the Czech Academy of Sciences, v.v.i., Biocev, Průmyslová 595, 252 50 Vestec, Czech Republic

²University of Chemistry and Technology Prague, Department of Biochemistry and Microbiology, Technická 5, 166 28 Prague 6, Czech Republic

³Novozymes A/S, Biologiens Vej 2, 2800 Kgs. Lyngby, Denmark

adamkovak@ibt.cas.cz

S1 nuclease from Aspergillus oryzae is a zinc-dependent phosphoesterase from the S1-P1 family of nucleases cleaving nucleic acids to nucleotides. Enzymes of the family can be found in many species having various biological functions with potential utilization in biotechnology and biochemistry. These enzymes have the phospholipase C/P1 nuclease fold and almost the same active site features, including the trinuclear zinc cluster, yet their substrate specificity differs from the strictly single-strand-specific to the non-specific nucleases able to cleave RNA, ssDNA, as well as dsDNA. Even though, the cleavage mechanism of the S1-P1 family enzymes was already described and there are several structures of complexes with products/ligands bound in the active site, there are still questions related to their substrate specificity and preferences [1, 2].

To further increase our understanding of the mechanisms behind the substrate specificity, we performed co-crystallization of S1 nuclease with various products of DNA and RNA cleavage. We obtained two novel structures of S1 nuclease in complex with RNA cleavage products in the active site. A complex with cytidine‑5′-monophosphate at 1.05 Å resolution and a complex with uridine at 1.10 Å resolution. The atomic resolution of both structures enables observation of fine details of ligand binding in the active site of the S1 nuclease and also a detailed comparison with already known complexes.

Both structures feature several unexpected strong peaks at 10σ level of difference electron density (mF_O‑DF_C), incompatible with standard molecular geometry constraints and not interpretable within the coordinates of the enzyme and its ligands. A closer analysis suggests that they correspond to zinc ions of translated protein molecules along the c axis of the unit cell. The translated molecules are present at low partial occupancy in two different positions and can be most likely interpreted as an intriguing case of order-disorder in protein crystal.

1. Kovaľ T, Dohnálek J, Biotechnology Advances, 2018, 36(3): 603-612

2. Kovaľ T. et al., PLoS ONE, 2016, 11(12): e0168832

This work was supported by the Academy of Sciences of the Czech Republic (RVO: 86652036) and European Regional Development Fund (Project CIISB4HEALTH, No. CZ.02.1.01/0.0/0.0/16_013/0001776 and ELIBIO, No. CZ.02.1.01/0.0/0.0/15_003/0000447).