Elsevier

DNA Repair

Volume 2, Issue 11, 21 November 2003, Pages 1199-1210
DNA Repair

Strand-specific processing of 8-oxoguanine by the human mismatch repair pathway: inefficient removal of 8-oxoguanine paired with adenine or cytosine

https://doi.org/10.1016/S1568-7864(03)00140-XGet rights and content

Abstract

Genomic DNA and its precursors are susceptible to oxidation during aerobic cellular metabolism, and at least five distinct repair activities target a single common lesion, 7,8-dihydro-8-oxoguanine (8-oxoG). The human mismatch repair (MMR) pathway, which has been implicated in an apoptotic response to covalent DNA damage, is likely to encounter 8-oxoG in both the parental and daughter strand during replication. Here, we show that lesions containing 8-oxoG paired with adenine or cytosine, which are most likely to arise during replication, are not efficiently processed by the mismatch repair system. Lesions containing 8-oxoG paired with thymine or guanine, which are unlikely to arise, are excised in an MSH2/MSH6-dependent manner as effectively as the corresponding mismatches when placed in a context that reflects the daughter strand during replication. Using a newly developed assay based on methylation sensitivity, we characterized strand-excision events opposite 8-oxoG situated to reflect placement in the parental strand. Lesions that efficiently trigger strand excision and resynthesis (8-oxoG paired with thymine or guanine) result in adenine or cytosine insertion opposite 8-oxoG. These latter pairings are poor substrates for further action by mismatch repair, but precursors for alternative pathways with non-mutagenic outcomes. We suggest that the lesions most likely to be encountered by the human mismatch repair pathway during replication, 8-oxoG·A or 8-oxoG·C, are likely to escape processing in either strand by this system. Taken together, these data suggest that the human mismatch repair pathway is not a major contributor to removal of misincorporated 8-oxoG, nor is it likely to trigger repeated attempts at lesion processing.

Introduction

The mismatch repair (MMR) pathway plays a role in mutation avoidance and genomic stability that is widely conserved [1], [2]. One of its major contributions is recognizing DNA mispairs during replication, which results in an excision and resynthesis event targeted to the nascent DNA strand. In addition to recognizing base–base or insertion/deletion mismatches, the prokaryotic MutS homodimers and eukaryotic MutS heterodimers are competent to bind specific types of damaged DNA. For example, human MSH2/MSH6 (MutSα) binds specific methylated or platinated DNA lesions [3] as well as the 7,8-dihydro-8-oxoguanine (8-oxoG) lesion within specific base contexts [4]. The former examples are of particular interest because they can arise during anticancer therapies and trigger a MMR-dependent apoptotic response, presumably resulting from recognition of DNA lesions in the parental strand during replication [2]. A similar outcome has been reported for oxidative DNA damage resulting from exposure to low-level radiation [5] or peroxide [6]. More recently, increased numbers of persistent 8-oxoG lesions in DNA were identified when the MSH2 gene was deleted in both mouse embryonic fibroblasts and in tumor cell lines [7]. In each of the latter studies involving oxidative damage to DNA, an important role for the human MMR pathway in responding to oxidized DNA bases, primarily 8-oxoG, has been inferred but not demonstrated.

In aerobic organisms, both the genomic DNA and the deoxynucleotide pools that support replication are susceptible to oxidative damage such as guanine oxidation [8], [9]. A hierarchy of protective repair mechanisms in human cells either excludes or results in the removal of 8-oxoG from DNA. Many of the strategies for mutation suppression are mechanistically analogous to the better-studied prokaryotic pathways [10]. For example, MutT from Escherichia coli and its human homolog hMTH1 [11] are thought to scavenge deoxy-8-oxoG triphosphate from the deoxynucleotide triphosphate pools by hydrolysis to the monophosphate, although this conclusion has been challenged [12]. When 8-oxoG triphosphate is instead misincorporated by a replicative polymerase, the first line of defense is likely an editing exonuclease activity [13]. Without proofreading, adenine is likely to be the most frequent templating base [14], [15]; repair of such lesions is discussed below. An 8-oxoG·C mispair can also result from oxidative damage to DNA, and 8-oxoG therefore represents a templating base when it persists to be replicated. Whether the 8-oxoG·C pairing arises from oxidative damage or by replicative bypass using a specialized polymerase such as pol eta [16], it is perhaps the least likely to be mutagenic, given that the hOGG1 glycosylase [17] targets this substrate and excises the lesion.

The 8-oxoG·A pairing presents a more difficult problem because it can arise in either strand during replication, and non-mutagenic processing depends on successfully distinguishing the origin of the lesion. Incorporation of dAMP opposite 8-oxoG in the template strand requires replacement of the adenine with cytosine to avoid a mutation, and the human homolog of the bacterial MutY protein, hMYH [18], is capable of initiating this repair event. Conversely, avoiding mutations when deoxy-8-oxoG monophosphate is incorporated opposite adenine requires removal of the deoxy-8-oxoG while preserving the templating adenine. Colussi et al. have suggested a protective role for the MMR pathway during replication, based in part on the observation that transfection of cell lines with human MTH1 suppresses the number of persistent 8-oxoG lesions [7]. In addition, an activity characterized as hOGG2 has been proposed to target the 8-oxoG for excision in this context [19]. Precisely how processing of 8-oxoG·A mispairs is coordinated during replication remains to be resolved.

This work describes our characterization of how and when the human MMR pathway contributes to processing mispairs containing 8-oxoG. By introducing site-specific nicks in either DNA strand 40 nucleotides away from such mispairs, mismatch correction was targeted to either the lesion-containing or its complementary strand. Misincorporation of deoxy-8-oxoG triphosphate into the nascent strand during replication was modeled by placing the lesion and nick in the same strand, which allows lesion removal by the MMR pathway. Conversely, when the nick and lesion reside in complementary strands, the MMR pathway is targeted to the strand opposite 8-oxoG. These substrates were used to model strand-excision and resynthesis events that are triggered when 8-oxoG lesions originate in genomic DNA and persist to be replicated. Based on our data, we propose a model that predicts how the MMR pathway complements alternative repair pathways without triggering an apoptotic event when the 8-oxoG lesion is encountered.

Section snippets

Cell culture and extract preparation

HeLa and LoVo cells were supplied by the National Cell Culture Center (Minneapolis MN). Cultures of MT1 and TK6 were grown as described [20]. Nuclear extracts were prepared using published protocols [21] and the protein concentration determined using the Bradford assay (Bio-Rad) using BSA as a standard. For each extract, the efficiency of repair using a substrate containing a G·T mismatch was titered as a function of protein concentration to ensure that the extract was not limiting with respect

Results

We first determined whether MutSα (MSH2/MSH6), the heterodimer that recognizes base–base mismatches and initiates repair [22], was competent to bind mispairs containing 8-oxoG. We used a mobility shift assay [23] to obtain a relative affinity of MutSα (MSH2/MSH6) for 8-oxoG paired with each possible base partner (Fig. 1) within oligo duplexes. We found that MutSα recognized 8-oxoG paired with either guanine or thymine with an affinity similar to the corresponding G·G or G·T mispairs. By

Discussion

Based on our results, we propose a model that describes the contribution of the MMR pathway to the global response to 8-oxoG in human cells (Fig. 6). When 8-oxoG evades glycolytic repair and persists to be replicated, the bases most likely to be incorporated are adenine [33] or cytosine [16], [33], depending upon the specific polymerase that effects the bypass. Both the 8-oxoG·A and 8-oxoG·C pairs represent poor substrates for the MMR pathway, and thus are likely to escape recognition and

Acknowledgements

The authors wish to thank Pat Foster for critical comments on the manuscript and the National Cell Culture Center (Minneapolis, MN) for providing the HeLa and LoVo cells used in this work. We are grateful to the NIH Genetics, Cellular and Molecular Sciences Training Grant GM 07757-21 for supporting EDL and the NIH for supporting this research.

References (41)

  • D.C. Thomas et al.

    Heteroduplex repair in extracts of human HeLa cells

    J. Biol. Chem.

    (1991)
  • J. Genschel et al.

    Human exonuclease I is required for 5′ and 3′ mismatch repair

    J. Biol. Chem.

    (2002)
  • Y. Gu et al.

    Human MutY homolog, a DNA glycosylase involved in base excision repair, physically and functionally interacts with mismatch repair proteins human MutS homolog 2/human MutS homolog 6

    J. Biol. Chem.

    (2002)
  • W.H. Fang et al.

    Human strand-specific mismatch repair occurs by a bidirectional mechanism similar to that of the bacterial reaction

    J. Biol. Chem.

    (1993)
  • G.L. Dianov et al.

    Base excision repair in nuclear and mitochondrial DNA

    Prog. Nucleic Acid Res. Mol. Biol.

    (2001)
  • M.J. Longley et al.

    DNA polymerase delta is required for human mismatch repair in vitro

    J. Biol. Chem.

    (1997)
  • T.T. Ni et al.

    MSH2 and MSH6 are required for removal of adenine misincorporated opposite 8-oxo-guanine in S. cerevisiae

    Mol. Cell

    (1999)
  • A.B. Buermeyer et al.

    Mammalian DNA mismatch repair

    Annu. Rev. Genet.

    (1999)
  • D.R. Duckett et al.

    Human MutSalpha recognizes damaged DNA base pairs containing O6-methylguanine, O4-methylthymine, or the cisplatin-d(GpG) adduct

    Proc. Natl. Acad. Sci. U.S.A.

    (1996)
  • T.L. DeWeese et al.

    Mouse embryonic stem cells carrying one or two defective Msh2 alleles respond abnormally to oxidative stress inflicted by low-level radiation

    Proc. Natl. Acad. Sci. U.S.A.

    (1998)
  • Cited by (31)

    • The impact of oxidative DNA damage and stress on telomere homeostasis

      2019, Mechanisms of Ageing and Development
      Citation Excerpt :

      This effect was reduced when MTH1 was overexpressed, indicating MMR processes 8-oxoG misincorporated during replication or repair events. Interestingly, the MMR complex MutSα recognizes 8-oxoG opposite A or C very poorly, but has strong affinity for 8-oxoG opposite T or G (Larson et al., 2003; Mazurek et al., 2002). While these mispairings are unlikely, 8-oxoG opposite C is the preferred substrate of OGG1, and A opposite 8-oxoG is the preferred substrate of MUTYH.

    • Pathways controlling dNTP pools to maintain genome stability

      2016, DNA Repair
      Citation Excerpt :

      In budding yeast, MMR appears to function as a MUTYH equivalent, recognising dA:8-oxodG mis-pairs following error-prone replication of genomic 8-oxodG and subsequently removing the non-damaged dA [124]. In contrast, human MutSα (MSH2:MSH6 heterodimer) inefficiently recognises 8-oxodG:dA/dC mis-pairs [125] and in agreement, these were poorly corrected by MMR in human cell extracts [123,126]. This apparent discrepancy between in vitro and in vivo studies could be potentially explained by the sequence context of the mis-incorporated 8-oxodGTP, as it was shown that MutSα could efficiently recognise 8-oxodG within slipped/mis-paired repeat sequences resembling frameshift intermediates [127].

    • Influence of oxidized purine processing on strand directionality of mismatch repair

      2015, Journal of Biological Chemistry
      Citation Excerpt :

      That both fragments a and b were labeled with similar intensity indicates that the repair patch spanned at least the distance of 361 nucleotides between the nick and the mispair. Only background amounts of [32P]dAMP were detected in fragments a and b of the control homoduplex, GO/A, or GO/C substrates (lanes 1, 3, and 4), which confirmed that the GO/C and GO/A pairs failed to activate nick-dependent long patch excision and resynthesis characteristic of MMR (38, 39) and exemplified in lane 2. Having shown that MMR was active in the extracts, but that it did not address the GO/A and GO/C substrates, we wanted to see whether they were processed by BER.

    • Structural, molecular and cellular functions of MSH2 and MSH6 during DNA mismatch repair, damage signaling and other noncanonical activities

      2013, Mutation Research - Fundamental and Molecular Mechanisms of Mutagenesis
      Citation Excerpt :

      Indeed, it has now been well documented that therapeutic levels of monofunctional alkylating agents do not affect the cell cycle until the second replication phase [102,137]. MutSα can also bind damaged bases that arise from normal cellular metabolism, such as 8-oxoGuanine [138–141]. Processing of 8-oxoG:G and 8-oxoG:T mispairs by the MMR pathway is similar to the high repair efficiency of G:G and G:T mispairs. [22,141].

    View all citing articles on Scopus
    1

    Present address: Department of Immunology HSB, University of Washington, Seattle, WA, USA.

    View full text