KMO
Kynurenine 3-Monooxygenase (4J2W) from Saccharomyces cerevisiae
Created by: Shanna Su
Kynurenine 3-Moonxygenase (KMO – 4J2W) is classified as a member of Class A flavoprotein hydroxylase based on its primary structure. This family of enzymes has a tightly bound
flavin adenine dinucleotide (FAD) cofactor. These enzymes hydroxylate activated aromatic substances, reduce molecular oxygen by using four electrons, and require a reduced pyridine nucleotide phosphate (1). KMO has an isoelectric point of 7.12 and a molecular weight of 188904.3 Da. These values were determined by ExPASY, a bioinformatics resource portal whose usage allows access to scientific information needed to compute a protein’s isoelectric point and molecular weight (2).
KMO is found in the outer mitochondrial membrane. In the body, KMO is highly expressed in the liver, kidney, endothelial cells, and monocytes. It is also found in microglia and macrophages, but is expressed at a relatively low level (3). As a part of the eukaryotic tryptophan catabolic pathway, the function of this enzyme is to convert L-kynurenine (L-Kyn) to 3-hydroxykynureninen. KMO requires one molecule of both NADPH and O
2 and produces nicotinamide adenine dinucleotide during the conversion. This enzyme may be
inhibited by UPF 648, an inhibitor that binds to the FAD cofactor. When KMO is inhibited, its active site enzyme is changed and L-Kyn can no longer bind (1,3).
This enzyme is of biological importance because inhibition of KMO activity serves as a good strategy for treatment of acute and chronic neurological diseases. Research has shown the inhibition of this enzyme leads to improvement in Huntington’s disease in yeast, fruit fly, and mouse models. In addition, inhibition of KMO has also improved symptoms of Alzheimer’s disease in mice. However, more research is needed because most KMO inhibitors cannot cross the blood-brain barrier. If an inhibitor that can cross this barrier is found, the cure against neurodegenerative diseases will be obtained (1,4,5).
KMO is a monomer of two identical
subunits that contains one ligand: FAD, a cofactor needed for the function of the protein. Its structure is a classic Rossmann fold with five β sheets and four α helices. These
secondary structures are responsible for interactions with the FAD cofactor (2). Overall, KMO is composed of 24% beta sheets and 29% helices. The β sheets are antiparallel and consist of 22 strands. In addition KMO also contains some random coils (6).
The
active site of KMO is adjacent to the FAD re-face, which is connected to solvent by a narrow-water filled cavity perpendicular to the active-site. The active site consists of protein residues: Met-230, Phe-246, Pro-321, Gln-325, Tyr-323, Ala-53, Leu-234, Phe-322, Ile-232, Leu-221, Arg-83, Ile-104, Glu-102, and Tyr-97. Residues
Lys-48 and Tyr-195 protect the dimethylbenzene of FAD from solvent.
Pro-321 to Gln-325 forms a
loop located above the re-side of FAD to allow for the binding of oxygen. The substrate for KMO is L-Kyn. However, it is impossible to crystallize a KMO complex with L-Kyn. KMO can be crystallized with the inhibitor UPF 648
bound to the active site (1).
Residues
Arg-83 and Tyr-97 bind the carboxylate of UPF 648 to KMO. The aromatic dichlorobenzene of the inhibitor is bound by hydrophobic residues: Leu-221, Met-230, Ile-232, Leu-234, Phe-246, Pro-321, and Phe-322. When UPF 648 is bound, it induces structural changes in KMO. For example the orientation of the Pro 321-Gln 325 loop is changed. This is because the inhibitor contains vicinal chlorides, which the natural substrate lacks. Consequently extra space is needed to accommodate the chlorides. To do so,
Phe-322 moves away from the active site and occupies the position that is usually taken by Tyr-323. This causes the Pro-321 to Gln-325 loop to move and the oxygen binding site is destroyed. Also, the position of the six-stranded anti-parallel β sheet is changed with respect to FAD (1,3).
UPF 648 and the natural substrate are structurally similar. Thus, the effect of binding L-Kyn can be modeled. Results showed that the Pro 321-Gln325 loop will not be moved and the oxygen binding site will function. A
hydrophobic pocket that binds to L-Kyn is formed by hydrophobic residues: Leu-221, Met-230, Ile-232, Leu-234, Phe-246, Pro-321, and Phe-322. Polar interactions can be found between
Gln-325 and the L-Kyn carbonyl group and between the aniline nitrogen atom of L-Kyn and the O4 atom of FAD. Furthermore, the carboxylate of L-Kyn is bound by residues: Arg-83 and Tyr-97. The amino group of L-Kyn has no interactions with the protein. L-Kyn often forms a salt bridge with the side chain of
Glu-102. However, this interaction is not critical for enzyme activity because the enzyme will still function if it is removed (1,3).
PSI-Blast is a program used to find proteins in the Protein Data Bank (PDB) with a similar primary structure to KMO. Total sequence homology and gaps between KMO and other proteins are compared. Gaps are amino acids that do not exist in KMO, but exist in the comparison protein. An E-value is assigned based on these two characteristics. An E value less than 0.05 is significant for proteins. Generally the more similar two proteins are, the lower the E value. Thus gaps and mutations will decrease sequence homology. This results in an increase in the E value.
During the search for this particular protein, all of results were KMO enzymes from different organisms. The E values were either 0 or very close to 0. This meant that the sequences of KMO enzymes in other organisms are virtually identical to the one being studied (4J2W). PDB IDs were unavailable for these enzymes, thus accession IDs were used. KMO from Penicillium digitatum (EKV07902.1) had an E value of 7e-177. E values determined from comparing KMO from Aedes aegypti (XP_001662859.1) and Xanthomonas albilineans (YP_003376241.1) were 3e-178 and 1e-178, respectively. Because the E values from these three organisms are essentially zero, this suggests that KMO is conserved across species (7).
The Dali server was used to find a protein in the PDB with a similar tertiary structure to KMO. Similarity is determined by comparing intramolecular distances between the two proteins and assigning a Z score. The Z score is significant if it is above two. This means the protein has similar folds.
One protein that was very similar to KMO is
2-methyl-3-hydroxypyridine-5-carboxylic acid oxygenase (3ALK) from Mesorhizobium loti. The Z score is 33.8. Other proteins with a similar tertiary structure are probable salicylate monooxygenase and p-hydroxybenzoate hydroxylase. When compared,
probable salicylate monooxygenase (4BK2) from Rhodococcus jostii has a Z score of 32.1.
P-Hydroxybenzoate Hydroxylase (1IUT) from Pseudomonas aeruginosa has a Z score of 31.4 (8). Of the three proteins being compared, 2-methyl-3-hydroxypyridine-5-carboxylic acid oxygenase (3ALK) is the most structurally similar to KMO. Like KMO, it is also monomer and contains a
FAD ligand. However it has a beta-mercaptoethanol which is not present in KMO. According to the Dali results of multiple structural alignment, 2-methyl-3-hydroxypyridine-5-carboxylic acid oxygenase has fewer number of loops (9).
Currently there are no drugs on market that target KMO. However research has suggested that KMO (4J2W, 4J33, 4J36, 4J31, and 4K34) could be developed as a target for drugs to treat various neurodegenerative diseases. Hypothetically, drugs would inhibit the enzyme in the same manner as UPF 648. More specifically these drugs would decrease levels of neurotoxic metabolites and increase levels of neuroprotective KYNA produced by the eukaryotic tryptophan catabolic pathway. This would limit neuronal damage caused by neurodegenerative diseases. Thus drugs treating neurodegenerative diseases should be inhibitors of KMO. Furthermore, the drug should be able to cross the blood-brain barrier (BBB). The biggest problem currently involved in drug development is finding an inhibitor that is able to cross the BBB. Inhibitors such as 3,4-dimethoxy-N-[4-(3-nitrophenyl)thiazol-2-yl]benzenesulfonamide (Ro-61-8048) have been studied, but none are able to cross the BBB. Consequently, the an inhibitor must be found to begin the development an efficacious drug (1,3).