Catalytic mechanism of Escherichia coli endonuclease VIII: roles of the intercalation loop and the zinc finger

Abstract

Endonuclease VIII (Nei) excises oxidatively damaged pyrimidines from DNA and shares structural and functional homology with formamidopyrimidine-DNA glycosylase. Although the structure of Escherichia coli Nei is solved [Zharkov et al. (2002) EMBO J. 21, 789-800], the functions of many of its amino acid residues involved in catalysis and substrate specificity are not known. We constructed a series of Nei mutants that interfere with eversion of the damaged base from the helix (QLY69-71AAA, DeltaQLY69-71) or perturb the conserved zinc finger (R171A, Q261A). Steady-state kinetics were measured with these mutant enzymes using substrates containing 5,6-dihydrouracil, two enantiomers of thymine glycol, 8-oxo-7,8-dihydroguanine, and an abasic site positioned opposite each of the four canonical DNA bases. To some extent, all Nei mutants were deficient in processing damaged DNA, with mutations in the zinc finger generally having a more profound effect. Wild-type Nei showed prominent opposite-base specificity (G > C approximately = T > A) when the lesion was 5,6-dihydrouracil or cis-(5S,6R)-thymine glycol but not for other lesions tested. Mutations in the Q69-Y71 loop eliminated this effect. Only wild-type Nei and Nei-Q261A mutants could be reductively cross-linked to damaged base-containing DNA. Experiments involving trapping with NaBH4 and the kinetics of DNA cleavage catalyzed by Nei-Q261A suggested that this mutant was deficient in regenerating free enzyme from the Nei-DNA covalent complex formed during the reaction. We conclude that the opposite-base specificity of Nei is primarily governed by residues in the Q69-Y71 loop and that both this loop and the zinc finger contribute significantly to the substrate specificity of Nei.

Fig. 1. Alignment of the sequences of Nei and Fpg proteins from different bacterial species

Protein sequences of Nei (top panel) and Fpg (bottom panel) were aligned using ClustalW (65). The light grey boxes indicate the intercalating residues inserted in the DNA helix; the dark grey box indicates the Arg residue conserved in Nei but not in Fpg (Arg-173 in E. coli Nei); the white box indicates the absolutely conserved Gln residue. Numeration of the residues in both Nei and Fpg is given according to the E. coli sequences.

Fig. 2. Cleavage of a DHU:T substrate by wild-type Nei

(A), A gel of a representative experiment is shown. S, substrate oligodeoxyribonucleotide; P, products (Nei releases products of both β- and δ-elimination, visible as the lower-mobility and the higher-mobility species, respectively, in the product doublet band). (B), The reaction velocity vs substrate concentration was plotted (3-4 experiments for each concentration, mean ± s.e.m. shown) and fit with a rectangular hyperbola to obtain the kinetic values listed in Table 2.

Fig. 3. Borohydride trapping of Nei and its mutants

Representative gels for wild-type Nei and Nei mutants are shown for DHU (left column), 8-oxoG (central column) and AP site (right column) opposite all bases. Arrows indicate the cross-link (Nei·DNA) and free oligonucleotides (DNA). The time of incubation was 5 min; the reactions were performed at 37°C as described in Materials and Methods.

Fig. 4. Dead-end complex formation in the Nei-Q261A mutant

(A), A representative time course of product accumulation during cleavage of DHU:G (50 nM) with wild-type Nei (open circles) and Nei-Q261A (filled circles). The concentration of the enzymes was 10 nM. Arrows indicate additions of incremental 10 nM Nei-Q261A. Solid lines show a fit of the experimental data to a model with an initial burst followed by an exponential increase to a maximum. (B), Time course of accumulation of cross-linked products from DHU:G by wild-type Nei and Nei-Q261A. (-), no enzyme added. Arrows correspond to the full-length cross-linking product (1), cleaved cross-linked product (2) and free oligonucleotides (3); incubation time is indicated on the figure.