Crystallographic and SAXS studies of S-adenosyl-l-homocysteine hydrolase from Bradyrhizobium elkanii

• Main Authors: Tomasz Manszewski, Kamil Szpotkowski, Mariusz Jaskolski
• Format: Article
• Language: English
• Published: International Union of Crystallography 2017-05-01
• Series: IUCrJ
• Subjects: SAH; SAM; SAHase adenosine; adenine; 2′-deoxyadenosine; 3′-deoxyadenosine; cordycepin; homocysteine; nicotinamide adenine dinucleotide; NAD; molecular gate
• Online Access: http://scripts.iucr.org/cgi-bin/paper?S2052252517002433

S-Adenosyl-l-homocysteine hydrolase (SAHase) from the symbiotic bacterium Bradyrhizobium elkanii (BeSAHase) was crystallized in four ligand complexes with (i) mixed adenosine (Ado) and cordycepin (Cord; 3′-deoxyadenosine), (ii) adenine (Ade), (iii) Ado and (iv) mixed 2′-deoxyadenosine (2′-dAdo) and Ade. The crystal structures were solved at resolutions of 1.84, 1.95, 1.95 and 1.54 Å, respectively. Only the Ade complex crystallized with a dimer in the asymmetric unit, while all of the other complexes formed a crystallographically independent tetrameric assembly. In the Ado/Cord complex, adenosine is found in three subunits while the fourth subunit has cordycepin bound in the active site. In the Ade and Ado complexes only these ligand molecules are present in the active sites. The 2′-dAdo/Ade complex has Ade bound in two subunits and 2′-dAdo bound in the other two subunits. The BeSAHase fold adopted a closed conformation in the complexes with Ado, Ade and 2′-dAdo, and a semi-open conformation when cordycepin occupied the active site. An SAHase-specific molecular gate, consisting of residues His342 and Phe343, behaves differently in the different complexes, but there is no simple correlation with the ligand type. Additional small-angle X-ray scattering (SAXS) experiments confirm the tetrameric state of the protein in solution. The main conclusions from this work are (i) that the SAHase subunit does not simply oscillate between two discrete conformational open/closed states in correlation with the absence/presence of a ligand in the active site, but can also assume an intermediate form for some ligands; (ii) that the shut/open state of the molecular gate in the access channel to the active site is not correlated in a simple way with the open/closed subunit conformation or empty/occupied status of the active site, but that a variety of states are possible even for the same ligand; (iii) that a cation (typically sodium) coordinated in an intersubunit loop rigidifies a molecular hinge and thus stabilizes the closed conformation; (iv) that BeSAHase in solution is a tetramer, consistent with the model derived from crystallography.