“…So far, crystallization experiments with this monomeric multidomain enzyme have been unsuccessful, and consequently attempts were initiated to dissect the full-length enzyme into stable, functional parts while retaining its enzymatic activity. Sequence comparisons, using sequences of enzymes in the hydratase/isomerase (crotonase) superfamily 21 and of mammalian mitochondrial L-3-hydroxyacyl-CoAdehydrogenases (HAD), 22,23 have suggested the presence of five domains in MFE-1, referred to as A, B, C, D, and E. 24 At the N terminus of MFE-1, the sequence analysis suggested a domain of the hydratase/isomerase fold (domain A), followed by a linker region (domain B) leading to a domain (domain C) that was predicted to be the dinucleotidebinding domain of the dehydrogenase part. This Rossmann-fold domain is followed by domain D, which corresponds to the dimerization domain of the human monofunctional dehydrogenase.…”