The aldo-keto reductases (AKRs): Overview

Abstract

The aldo-keto reductase (AKR) protein superfamily contains > 190 members that fall into 16 families and are found in all phyla. These enzymes reduce carbonyl substrates such as: sugar aldehydes; keto-steroids, keto-prostaglandins, retinals, quinones, and lipid peroxidation byproducts. Exceptions include the reduction of steroid double bonds catalyzed by AKR1D enzymes (5β-reductases); and the oxidation of proximate carcinogen trans-dihydrodiol polycyclic aromatic hydrocarbons; while the (β-subunits of potassium gated ion channels (AKR6 family) control Kv channel opening. AKRs are usually 37 kDa monomers, have an (α/β)8-barrel motif, display large loops at the back of the barrel which govern substrate specificity, and have a conserved cofactor binding domain. AKRs catalyze an ordered bi bi kinetic mechanism in which NAD(P)H cofactor binds first and leaves last. In enzymes that favor NADPH, the rate of release of NADP+ is governed by a slow isomerization step which places an upper limit on kcat. AKRs retain a conserved catalytic tetrad consisting of Tyr55, Asp50, Lys84, and His117 (AKR1C9 numbering). There is conservation of the catalytic mechanism with short-chain dehydrogenases/reductases (SDRs) even though they show different protein folds. There are 15 human AKRs of these AKR1B1, AKR1C1-1C3, AKR1D1, and AKR1B10 have been implicated in diabetic complications, steroid hormone dependent malignancies, bile acid deficiency and defects in retinoic acid signaling, respectively. Inhibitor programs exist worldwide to target each of these enzymes to treat the aforementioned disorders. Inherited mutations in AKR1C and AKR1D1 enzymes are implicated in defects in the development of male genitalia and bile acid deficiency, respectively, and occur in evolutionary conserved amino acids. The human AKRs have a large number of nsSNPs and splice variants, but in many instances functional genomics is lacking. AKRs and their variants are now poised to be interrogated using modern genomic and informatics approaches to determine their association with human health and disease.