Figure 1. An alignment of the conserved [4Fe–4S]^2+/+ cluster coordination site from a selection of AdoMet radical enzymes. The sequence alignment shows three strictly conserved cysteine residues that coordinate to the FeS cluster (green) and a semiconserved aromatic residue that stacks against the AdoMet adenine ring (blue). Enzymes are grouped based on their reported enzymatic function, but show little sequence conservation outside of this short motif.

Figure 2. The chemical reactions catalysed by AdoMet radical enzymes. (a) All of the AdoMet radical enzymes catalyse reduction of the AdoMet sulfonium with an electron derived from a [4Fe–4S]⁺ cluster and a hydrogen atom abstracted from a C–H bond of the substrate. The substrate is thus oxidized to initially generate a carbon radical. (b) The role of the carbon radical varies – representative examples include substrate isomerization (lysine-2,3-aminomutase and spore photoproduct lyase), glycyl radical formation (PFL activase), sulfur insertion (biotin and lipoate synthase) and substrate oxidation (coproporphyrinogen III oxidase). Hydrogen atoms that are abstracted by the AdoMet-derived 5¢-deoxyadenosyl radical, and the general area where chemistry occurs, are highlighted in red.

Figure 3. AdoMet radical enzymes share a common structural architecture. Shown are ribbon depictions of the structures of monomer subunits of E. coli biotin synthase (BioB), E. coli coproporphyrinogen III oxidase (HemN), C. subterminale Sb4 lysine-2,3-aminomutase (KamA) and S. aureus MoaA. At the upper region of each protein is the AdoMet-[4Fe–4S]²⁺ complex coordinate to three conserved cysteine residues within an extended loop. The core structure has an ()₆ topology generating a six-stranded parallel sheet that forms a 3/4-barrel around the active site. The central cavity or cleft contains the substrates: dethiobiotin and a [2Fe–2S]²⁺ cluster for BioB, an empty cleft that presumably binds coproporphyrinogen III in HemN, lysine bound to a PLP cofactor in KamA and a second FeS cluster that assists in binding 5¢-GTP in MoaA (not observed in this structure). The native oligomeric states of BioB and MoaA are dimers, and that of KamA is a tetramer. Structures are drawn with PyMol using PDB files 1R30, 1OLT, 2A5H and 1TV8.

Figure 4. The complex between AdoMet and the [4Fe–4S]^2+/+ cluster. (a) The interaction of AdoMet with the FeS cluster is taken from crystal structure of MoaA. The carboxylate and amine of the methionyl substituents are coordinated to a unique Fe from the cluster, and together with hydrogen bonds from protein residues to the ribose and adenine (not shown), this interaction forces close proximity of the positively charged sulfonium and the FeS cluster. The views on the left show the alignment of AdoMet with atoms in the FeS cluster. The space-filling model on the right shows that the AdoMet ribose and sulfonium are in van der Waals contact with both sulfide and iron from the FeS cluster. (b) AdoMet forms a stable complex with the [4Fe–4S]²⁺ cluster in the presence of substrate. An electron transferred from an external electron donor into the FeS cluster is passed into the AdoMet sulfonium, triggering C–S bond cleavage and formation of a 5¢-deoxyadenosyl radical. In most enzymes, AdoMet also forms a stable complex with the [4Fe–4S]⁺ cluster in the absence of substrate. In this case, binding of substrate likely triggers electron transfer into the AdoMet sulfonium, also leading to formation of a 5¢-deoxyadenosyl radical.