The DNA-repair protein AlkB, EGL-9, and leprecan define new families of 2-oxoglutarate- and iron-dependent dioxygenases

Abstract

Background: Protein fold recognition using sequence profile searches frequently allows prediction of the structure and biochemical mechanisms of proteins with an important biological function but unknown biochemical activity. Here we describe such predictions resulting from an analysis of the 2-oxoglutarate (2OG) and Fe(II)-dependent oxygenases, a class of enzymes that are widespread in eukaryotes and bacteria and catalyze a variety of reactions typically involving the oxidation of an organic substrate using a dioxygen molecule.

Results: We employ sequence profile analysis to show that the DNA repair protein AlkB, the extracellular matrix protein leprecan, the disease-resistance-related protein EGL-9 and several uncharacterized proteins define novel families of enzymes of the 2OG-Fe(II) oxygenase superfamily. The identification of AlkB as a member of the 2OG-Fe(II) oxygenase superfamily suggests that this protein catalyzes oxidative detoxification of alkylated bases. More distant homologs of AlkB were detected in eukaryotes and in plant RNA viruses, leading to the hypothesis that these proteins might be involved in RNA demethylation. The EGL-9 protein from Caenorhabditis elegans is necessary for normal muscle function and its inactivation results in resistance against paralysis induced by the Pseudomonas aeruginosa toxin. EGL-9 and leprecan are predicted to be novel protein hydroxylases that might be involved in the generation of substrates for protein glycosylation.

Conclusions: Here, using sequence profile searches, we show that several previously undetected protein families contain 2OG-Fe(II) oxygenase fold. This allows us to predict the catalytic activity for a wide range of biologically important, but biochemically uncharacterized proteins from eukaryotes and bacteria.

Figure 1

Multiple sequence alignment of the 2OG-Fe(II) dioxygenase superfamily. Individual protein families are separated by blank lines and a brief description of each family is given to the right of the alignment. The numbers at the ends of the alignment indicate the position of the first and last of the aligned residues in the respective protein sequences. The consensus secondary structure is shown above the alignment in uppercase letters. It was derived by taking those elements that are shared by the predicted structures of individual families and the experimentally determined structures; H indicates α helix and E indicates extended conformation (β strand). The lowercase letters represent extensions of the secondary structure elements that are seen in some, but not all, members of the superfamily. The conserved amino-terminal extensions that are specific only to a given family are separated from the rest of the alignment by vertical lines. The coloring of the alignment columns is according to the 85% consensus that is shown underneath the alignment and includes the following categories of amino acid residues: h,hydrophobic; l, aliphatic; a, aromatic (Y, F, W, H, L, I, V, M, A, all shaded yellow); s, small (S, A, G, T, V, P, N, H, D, shaded blue); b, big (K, R, E, Q, W, F, Y, L, M, I, shaded gray); +, positively charged (K, R, H; colored magenta). The (predicted) catalytic residues are indicated by asterisks and with reverse red shading. The proteins are designated by the protein/gene name, the species abbreviation and the gene identification (GI) number. Protein abbreviations are: CAS, clavaminic acid synthase; DAOCS, deacetoxycephalosporin C synthetase; EFE, ethylene-forming enzyme; FLAS, flavonol synthase; Ga20Ox, giberellin 20-oxidase; IPNS, isopenicillin N synthase; LDOX, leucoanthocyanidin hydroxylase; Lep, leprecan; P4HA, prolyl-4-hydroxylase; PLO, lysyl hydroxylase; SanF and SanC, enzymes involved in nikkomycin biosynthesis. The remaining names are the standard names of the genes that encode the respective proteins. Species abbreviations: At, Arabidopsis thaliana; Bb, Borrelia burgdorferi; Cc, Caulobacter crescentus; Ce, Caenorhabditis elegans; Ci, Ciona intestinalis; Dm, Drosophila melanogaster; Ec, Escherichia coli; Em, Emericella nidulans; Hs, Homo sapiens; Lc, Lysobacter lactamgenus; Le, Lycopersicon esculentum; Mtu, Mycobacterium tuberculosis; Nc, Neurospora crassa; Pa, Pseudomonas aeruginosa; Pet, Petunia hybrida; Rr, Rattus rattus; Sc, Saccharomyces cerevisiae; Sp, Schizosaccharomyces pombe; Sot, Solanum tuberosum; Scoe, Streptomyces coelicolor; Scan, Streptomyces ansochromogenes; Scla, Streptomyces clavuligerus; Ssp, Synechocystis; Vc, Vibrio cholerae; ASPV, apple stem pitting virus; ACLSV, apple chlorotic leaf spot virus; BSV, blueberry scorch virus; GLV, garlic latent virus; GVA, grapevine virus A; PBCV, Parameciumbursaria chlorella virus; PMV, papaya mosaic virus; SHVX, shallot virus X.

Figure 2

A structural model of the DSBH core of the 2OG-Fe(II) dioxygenase superfamily. This is based on the Emericella nidulans isopenicillin N synthase structure (PDB:1ips). The side chains of the amino acid residues implicated in catalysis and in substrate binding are shown (see text) and the Fe(II) ion is indicated by a red circle.

Figure 3

Topological diagrams for three members of the 2OG-Fe(II) dioxygenase superfamily. The diagrams are based on the experimentally determined structures for E. nidulans isopenicillin N synthase (PDB: 1ips) and structural models of prolyl-4-hydroxylase and AlkB. The amino acid residues of the active site and the Fe(II) ion are shown as in Figure 2.