8fsi

Publication Abstract from PubMed

Radical S-adenosyl-l-methionine (SAM) enzymes are ubiquitous in nature and carry out a broad variety of difficult chemical transformations initiated by hydrogen atom abstraction. Although numerous radical SAM (RS) enzymes have been structurally characterized, many prove recalcitrant to crystallization needed for atomic-level structure determination using X-ray crystallography, and even those that have been crystallized for an initial study can be difficult to recrystallize for further structural work. We present here a method for computationally engineering previously observed crystallographic contacts and employ it to obtain more reproducible crystallization of the RS enzyme pyruvate formate-lyase activating enzyme (PFL-AE). We show that the computationally engineered variant binds a typical RS [4Fe-4S](2+/+) cluster that binds SAM, with electron paramagnetic resonance properties indistinguishable from the native PFL-AE. The variant also retains the typical PFL-AE catalytic activity, as evidenced by the characteristic glycyl radical electron paramagnetic resonance signal observed upon incubation of the PFL-AE variant with reducing agent, SAM, and PFL. The PFL-AE variant was also crystallized in the [4Fe-4S](2+) state with SAM bound, providing a new high-resolution structure of the SAM complex in the absence of substrate. Finally, by incubating such a crystal in a solution of sodium dithionite, the reductive cleavage of SAM is triggered, providing us with a structure in which the SAM cleavage products 5'-deoxyadenosine and methionine are bound in the active site. We propose that the methods described herein may be useful in the structural characterization of other difficult-to-resolve proteins.

Computational engineering of previously crystallized pyruvate formate-lyase activating enzyme reveals insights into SAM binding and reductive cleavage.,Moody JD, Hill S, Lundahl MN, Saxton AJ, Galambas A, Broderick WE, Lawrence CM, Broderick JB J Biol Chem. 2023 Jun;299(6):104791. doi: 10.1016/j.jbc.2023.104791. Epub 2023 , May 6. PMID:37156396^[1]

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine.