Expansion of base excision repair compensates for a lack of DNA repair by oxidative dealkylation in budding yeast

Abstract

The Mag1 and Tpa1 proteins from budding yeast (Saccharomyces cerevisiae) have both been reported to repair alkylation damage in DNA. Mag1 initiates the base excision repair pathway by removing alkylated bases from DNA, and Tpa1 has been proposed to directly repair alkylated bases as does the prototypical oxidative dealkylase AlkB from Escherichia coli However, we found that in vivo repair of methyl methanesulfonate (MMS)-induced alkylation damage in DNA involves Mag1 but not Tpa1. We observed that yeast strains without tpa1 are no more sensitive to MMS than WT yeast, whereas mag1-deficient yeast are ∼500-fold more sensitive to MMS. We therefore investigated the substrate specificity of Mag1 and found that it excises alkylated bases that are known AlkB substrates. In contrast, purified recombinant Tpa1 did not repair these alkylated DNA substrates, but it did exhibit the prolyl hydroxylase activity that has also been ascribed to it. A comparison of several of the kinetic parameters of Mag1 and its E. coli homolog AlkA revealed that Mag1 catalyzes base excision from known AlkB substrates with greater efficiency than does AlkA, consistent with an expanded role of yeast Mag1 in repair of alkylation damage. Our results challenge the proposal that Tpa1 directly functions in DNA repair and suggest that Mag1-initiated base excision repair compensates for the absence of oxidative dealkylation of alkylated nucleobases in budding yeast. This expanded role of Mag1, as compared with alkylation repair glycosylases in other organisms, could explain the extreme sensitivity of Mag1-deficient S. cerevisiae toward alkylation damage.

Keywords: AlkA; AlkB; DNA damage; DNA repair; Mag1; Saccharomyces cerevisiae; Tpa1; base excision repair (BER); hydroxylase; oxidative dealkylation; translation regulation.

Figure 1.

Alkylation damage to DNA bases and pathways for repair. A, examples of alkylated nucleobases. B, ϵA may be repaired by either oxidative dealkylation (upper pathway; DRR) or glycosylase-initiated BER (lower pathway).

Figure 2.

Cell killing induced by exposure of yeast strains to MMS. Exponentially growing yeast cells were exposed to 0.3% MMS (v/v) for 1 h, then plated on YPD, and evaluated for survival relative to untreated cells. Error bars represent S.D.

Figure 3.

Mag1 excises alkylated nucleobases that are known substrates for DRR by AlkB. A, Mag1 excises ϵA, 1mA, 3mC, and ϵC from duplex DNA. After 24-h incubation of 25mer DNA duplexes (Table S1) containing the indicated central bp with Mag1, the abasic sites were cleaved with sodium hydroxide, and the samples were analyzed on a 17.5% denaturing polyacrylamide gel. Control reactions (lanes 1, 3, 5, and 7) show that the oligonucleotides remain intact in the absence of Mag1. The two product bands in lane 8 result from the milder hydrolysis conditions used to cleave abasic sites in reactions containing ϵC and correspond to species with 3′-phosphate or 3′-deoxyribose phosphate termini, which migrate slightly differently (see “Experimental procedures”). B, representative time courses for single-turnover excision of ϵC from 5 nm ϵC-25mer by varying concentrations of Mag1. The data were fit by a single exponential. The average of duplicate reactions is shown, and error bars represent S.D. C, dependence of the single-turnover rate constant for alkylated base excision from ϵA-25mer, 1mA-25mer, 3mC-25mer, and ϵC-25mer substrates on the Mag1 concentration. Hyperbolic dependence of the single-turnover rate constant on the Mag1 concentration was observed for ϵA-25mer and ϵC-25mer with k_max of 0.014 ± 0.001 min⁻¹ and a K_1/2 value of 120 ± 30 nm for ϵA-25mer and k_max of 0.010 ± 0.001 min⁻¹ and a K_1/2 value of 48 ± 18 nm for ϵC-25mer. K_1/2 values were too low to be measured for 1mA-25mer and 3mC-25mer, but k_max of 0.013 ± 0.001 min⁻¹ for 3mC-25mer and k_max of 0.00026 ± 0.00003 min⁻¹ for 1mA-25mer were determined by averaging all measured k_obs values for each substrate. The inset shows a rescaled plot of the 1mA-25mer data. The average of duplicate reactions is shown, and error bars represent S.D.

Figure 4.

Representative data for determination of Mag1 relative k_cat/K_m values. A, a typical denaturing polyacrylamide gel showing a time course for Mag1-catalyzed glycosylase activity toward a mixture of 1mA-25mer and the reference substrate, ϵA-19mer (Table S1). Cleavage of 1mA-25mer gives a labeled 12mer product, and cleavage of the reference substrate gives a labeled 9mer product. Mag1 was 50 nm, 1mA-25mer was 300 nm, and ϵA-19mer was 500 nm. B, quantitation of time points shown in A and identical reactions. The average of duplicate reactions is shown, and error bars represent S.D. The relative k_cat/K_m value of 0.70 ± 0.10 for 1mA-25mer with respect to ϵA-19mer was determined from the linear initial rates and the initial substrate concentrations (see “Experimental procedures”). C, at any given time point in B, the ratio of the two products is the same within error.

Figure 5.

Tpa1 and AlkB are not functionally homologous. A, AlkB repairs both ss and ds 25mers containing a central ϵA, 1mA, 3mC, or ϵC, but no DNA repair activity is observed for Tpa1 using the same substrates and reaction conditions. Reactions contained 5 μm AlkB or Tpa1, 100 nm ss or ds 25mer DNA, and appropriate Fe(II) cofactor and 2OG cosubstrate and were incubated at 37 °C for 1 h. Reactions were performed in triplicate. B, MALDI-TOF MS analyses of 25 μm RPS23_47–84 peptide (calculated mass, 4136 Da) with or without 2 μm Tpa1 and appropriate Fe(II) cofactor and 2OG cosubstrate at room temperature for 10 min (see “Experimental procedures” for details). The mass increase of 16 Da corresponds to hydroxylation of the peptide by Tpa1.