InterPro: IPR008255 Pyridine nucleotide-disulphide oxidoreductase, class-II, active site

The pyridine nucleotide-disulphide reductases (PNDR) use the isoalloxazine ring of FAD to shuttle reducing equivalents from NAD(P)H to a Cys residue that is usually a part of a redox-active disulphide bridge. In a second step, the reduced disulphide reduces the substrate. On the basis of sequence and structural similarities [1], PNDR can be categorised into 2 groups.

Class II includes: prokaryotic and eukaryotic thioredoxin reductases [2, 3]; bacterial alkyl hydroperoxide reductases [4]; bacterial NADH:dehydrogenases [5]; a probable oxidoreductase encoded in the Clostridium pasteurianum rubredoxin operon [6]; and yeast hypothetical protein YHR106w.

The 3D structure of Escherichia coli thioredoxin reductase (TR) has been solved [1, 7]. The protein exists as a homodimer, with 3 domains per monomer, which correspond to the FAD-binding, NAD(P)H-binding and central domains of glutathione reductase (GR) (cf. signature PNDRDTASEI). However, TR lacks the domain that provides the dimer interface in GR, and forms a completely different dimeric structure. The relative orientation of these domains is very different in the 2 enzymes: when the FAD-binding domains of TR and GR are superimposed, the NADPH-binding domain of one is rotated by 66 degrees with respect to the other. The FAD- and NAD(P)H-binding domains have a similar doubly-wound alpha/beta fold, suggesting they evolved by gene duplication [8]. While in GR the redox active disulphide is located in the FAD-binding domain, in TR it lies in the NADPH-binding domain. This suggests that the enzymes diverged from an ancestral nucleotide-binding protein and acquired their disulphide reductase activities independently [1].

The sequence around the two cysteines involved in the redox-active disulphide bond is conserved, and is covered by this pattern.