Biol. Chem., Vol. 386, pp. 971–980, October 2005 • Copyright  by Walter de Gruyter • Berlin • New York. DOI 10.1515/BC.2005.113 2005/203 Article in press - uncorrected proof Review Structural and functional comparison of HemN to other radical SAM enzymes Gunhild Layer 1,a , Eric Kervio 2 , Gaby Morlock 2 , Dirk W. Heinz 3 , Dieter Jahn 1, *, Janos Retey 2, * and Wolf-Dieter Schubert 3 1 Institute of Microbiology, Technical University of Braunschweig, Spielmannstrasse 7, D-38106 Braunschweig, Germany 2 Institute of Organic Chemistry, University of Karlsruhe, Richard-Willsta ¨ tter-Allee, D-76128 Karlsruhe, Germany 3 Department of Structural Biology, German Research Center for Biotechnology (GBF), Mascheroder Weg 1, D-38104 Braunschweig, Germany * Corresponding authors e-mail: d.jahn@tu-bs.de; janos.retey@ioc.uka.de Abstract Radical SAM enzymes have only recently been recog- nized as an ancient family sharing an unusual radical- based reaction mechanism. This late appreciation is due to the extreme oxygen sensitivity of most radical SAM enzymes, making their characterization particularly ardu- ous. Nevertheless, realization that the novel apposition of the established cofactors S-adenosylmethionine and w4Fe-4Sx cluster creates an explosive source of catalytic radicals, the appreciation of the sheer size of this previ- ously neglected family, and the rapid succession of three successfully solved crystal structures within a year have ensured that this family has belatedly been noted. In this review, we report the characterization of two enzymes: the established radical SAM enzyme, HemN or oxygen- independent coproporphyrinogen III oxidase from Esche- richia coli, and littorine mutase, a presumed radical SAM enzyme, responsible for the conversion of littorine to hyoscyamine in plants. The enzymes are compared to other radical SAM enzymes and in particular the three reported crystal structures from this family, HemN, biotin synthase and MoaA, are discussed. Keywords: BioB; w4Fe-4Sx cluster; HemN; littorine mutase; MoaA; radical SAM enzymes; S-adenosyl-L- methionine. Introduction Radical-based enzyme reactions provide unique solu- tions to chemical problems difficult to solve by more con- ventional routes. Enzyme reactions involving radical Present address: Laboratoire de Chimie et Biochimie des Cen- a tres Re ´ dox Biologiques, CEA-Grenoble, 17 rue des Martyrs, F- 38054 Grenoble, France. intermediates are consequently more widespread and involved in more diverse biological processes than pre- viously anticipated (van der Donk et al., 1998; Banerjee, 2003a). In some cases, the route of the organic radical is fairly simple, in that the catalytic radical is generated, transferred to a substrate and induces chemical re- arrangement of the substrate, followed by the transfer of a single electron to a final electron acceptor, preventing further unwanted reactions. Within biological systems, the organic radical is not confined to the substrate. Instead, it is frequently generated within one enzyme, transferred to the substrate or to a second enzyme, which in turn catalyzes the reaction, and transferred to a terminal acceptor or back to the enzyme and either stored for the next round of catalysis or recombined into the pre-radical state. Residues that can harbor the radi- cal state include tyrosine, tryptophan, cysteine and gly- cine. The advantage of radical-based catalysis is the high reaction selectivity achieved (Retey, 1990) at the cost that enzymes need to protect themselves to prevent undesi- red side reactions. Several distinct mechanisms to generate free radical states have been identified. One of these involves the generation of a 59-deoxyadenosyl radical, which initiates radical catalysis in two different classes of enzymes: ade- nosylcobalamin-dependent enzymes induce homolytic cleavage of the weak C-Co bond of the B 12 cofactor, forming cob(II)alamin and the proposed, highly reactive 59-deoxyadenosyl radical that abstracts hydrogen atoms from non-activated C-H bonds to generate the corre- sponding substrate radicals (Gruber and Kratky, 2002; Banerjee, 2003b; Toraya, 2003; Speranza et al., 2004). In recent years a new class of radical enzymes has emerged, members of which similarly produce a 59- deoxyadenosyl radical: the radical SAM family of pro- teins, with over 600 putative members (Sofia et al., 2001). The members of this new enzyme class use a combi- nation of a reduced iron-sulfur cluster and S-adenosyl-L- methionine (SAM) in place of adenosylcobalamin to generate the 59-deoxyadenosyl radical that subsequently initiates diverse radical reactions (Frey and Magnusson, 2003; Jarrett, 2003; Marsh et al., 2004). Although common features had been recognized for some radical SAM enzymes over the last decade (Frey and Reed, 2000; Cheek and Broderick, 2001), it was only in 2001 that a computational study proposed a large new enzyme class of radical enzymes: the radical SAM family (Sofia et al., 2001). Members are found in all three king- doms of life and catalyze highly diverse reactions as part of DNA precursor, vitamin, cofactor, antibiotic and her- bicide biosynthesis, and in biodegradation pathways. Well-established members of the radical SAM family are