Archaeal DNA uracil repair via direct strand incision: A minimal system reconstituted from purified components

DNA Repair 9 (2010) 438–447

Contents lists available at ScienceDirect

DNA Repair journal homepage: www.elsevier.com/locate/dnarepair

Archaeal DNA uracil repair via direct strand incision: A minimal system reconstituted from purified components Lars Schomacher a,1 , K. Anke Schürer a , Elena Ciirdaeva a , Paul McDermott b,2 , James P.J. Chong b , Wilfried Kramer a , Hans-Joachim Fritz a,∗ a

Abteilung Molekulare Genetik und Präparative Molekularbiologe, Institut für Mikrobiologie und Genetik, Georg-August-Universität Göttingen, Grisebachstr. 8, 37077 Göttingen, Germany Department of Biology (Area 5), P.O. Box 373, University of York, YO10 5YW, York, UK

b

a r t i c l e

i n f o

Article history: Received 11 November 2009 Received in revised form 21 December 2009 Accepted 5 January 2010 Available online 2 February 2010 Keywords: Hydrolytic deamination DNA uracil repair Strand incision repair DNA uridine endonuclease Thermophilic archaea Methanothermobacter thermautotrophicus H

a b s t r a c t Hydrolytic deamination of DNA cytosine residues results in U/G mispairs, pre-mutagenic lesions threatening long-term genetic stability. Hence, DNA uracil repair is ubiquitous throughout all extant life forms and base excision repair, triggered by a uracil DNA glycosylase (UDG), is the mechanistic paradigm adopted, as it seems, by all bacteria and eukaryotes and a large fraction of archaea. However, members of the UDG superfamily of enzymes are absent from the extremely thermophilic archaeon Methanothermobacter thermautotrophicus H. This organism, as a hitherto unique case, initiates repair by direct strand incision next to the DNA-U residue, a reaction catalyzed by the DNA uridine endonuclease Mth212, an ExoIII homologue. To elucidate the detailed mechanism, in particular to identify the molecular partners contributing to this repair process, we reconstituted DNA uracil repair in vitro from only four purified enzymes of M. thermautotrophicus H. After incision at the 5 -side of a 2 -d-uridine residue by Mth212 DNA polymerase B (mthPolB) is able to take over the 3 -OH terminus and carry out repair synthesis generating a 5 -flap structure that is resolved by mthFEN, a 5 -flap endonuclease. Finally, DNA ligase seals the resulting nick. This defines mechanism and minimal enzymatic requirements of DNA-U repair in this organism. © 2010 Elsevier B.V. All rights reserved.

1. Introduction The long-term in vivo conservation of DNA structure – and hence of genetic information – requires expenditure of metabolic energy in both proof-reading and DNA repair. The mispaired U/G opposition is one of the most common pre-mutagenic DNA lesions [1]; it originates from spontaneous hydrolytic deamination of DNA cytosine residues and, left unrepaired, results in C/G to T/A transition mutations in 50% of the progeny. Due to fundamental laws of chemical reaction kinetics, this genetic threat is greatly exacerbated in thermophilic and hyperthermophilic organisms.

Abbreviations: BER, base excision repair; DNA-U repair, DNA uracil repair; (DNA-)U endonuclease, DNA uridine endonuclease; PolB, DNA Polymerase B; PCNA, Proliferating Cell Nuclear Antigen; FEN, 5 -flap endonuclease; PPi , inorganic pyrophosphate. ∗ Corresponding author. Tel.: +49 551 39 3801; fax: +49 551 39 3805. E-mail address: [email protected] (H.-J. Fritz). 1 Present address: Division of Molecular Embryology, German Cancer Research Center, Im Neuenheimer Feld 280, 69120 Heidelberg, Germany. 2 Present address: Department of Biological Sciences, University of Essex, Wivenhoe Park, Colchester, CO4 3SQ, UK. 1568-7864/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.dnarep.2010.01.004

DNA uracil repair is thought to be a mandatory element of the macromolecular metabolism of all present-day organisms. In bacteria and eukarya, U/G mispairs are invariably channeled into the base excision repair (BER) pathway [2] by hydrolytic removal of the uracil residue, catalyzed by a member of the UDG superfamily of DNA uracil glycosylases. Frequently, several UDG enzymes of one or several of the UDG families 1–5 [3–5] are present in a given organism. In archaea, the situation concerning DNA uracil repair is in stark contrast to the uniform picture in bacteria and eukarya. In particular, the complete absence from a number of sequenced archaeal genomes of genes coding for members of any known UDG family, e.g. Methanothermobacter thermautotrophicus H [6] and Methanopyrus kandleri AV19 [7], used to be a puzzle in view of the typically high growth temperatures of these organisms. Recently, we described a novel pathway of DNA uracil repair operating in the thermophilic archaeon M. thermautotrophicus H (optimal growth temperature: 65–70 ◦ C). Repair studies in vitro employing whole-cell extracts unambiguously demonstrated that in this organism DNA uracil repair is initiated by direct strand incision next to the mispaired uracil residue and that this reaction is catalyzed by the enzyme Mth212, a member of the ExoIII family of AP endonucleases [8]. Within this family, Mth212 is uniquely equipped with the additional function of a DNA uridine endonu-

L. Schomacher et al. / DNA Repair 9 (2010) 438–447

439

Table 1 DNA polymerase classes and homologous proteins in M. thermautotrophicus H. Class

DNA polymerases in other organisms

Homologues in M. th.

Polymerizing function in

Additional activities

A B C D X Y

Pol I (E. coli), ␥, ␪ (H. sapiens) Pol II (E. coli), ␣, ␦, ␧, ␨ (H. sapiens) Pol III (E. coli) Pol II (Pyrococcus furiosus) ␤, ␭, ␮, TdT (H. sapiens) Pol IV, Pol V (E. coli) ␩, ␫, ␬, Rev1 (H. sapiens)

None present mthPolB (Mth208 + Mth1208) None present mthPol II (Mth1405 + Mth1536) mthPolX (Mth550) None present

Repair/replication Replication/repair Replication Replication Repair Translesion synthesis

3 → 5 exonuclease; 5 nuclease 3 → 5 exonuclease 3 → 5 exonuclease 3 → 5 exonuclease 5 phosphatase Unknown

Classification, functions and list of additional activities beyond DNA 5 → 3 polymerase activity were adapted from the literature [25–27]. Whenever possible for each class of DNA polymerases enzymes from E. coli and Homo sapiens are listed. Note: In contrast to Burgers et al. [25] human polymerase ␴ was not included since the yeast orthologue of the enzyme was found to be a poly(A) RNA polymerase instead of a DNA polymerase [28]. Class D DNA polymerases have been identified only in archaea [29,30], class X DNA polymerases are unknown in bacteria, M. th.: M. thermautotrophicus H. TdT: terminal deoxynucleotidyl transferase.

clease [9]. Furthermore, we showed that in M. thermautotrophicus H general repair of DNA uracil residues completely depends on this newly discovered pathway and that Mth212 is its sole initiator [8]. As to the nature of other contributing biochemical components, a number of plausible suggestions could be made, but no proof was furnished. In particular, the complexity of the set of enzymes defining the core of the entire pathway remained an open question. To date, no methods of genetic experimentation exist for M. thermautotrophicus H and the questions posed above must therefore be approached biochemically. Here we show that a minimal set of four enzymes (Mth212, mthPolB, mthFEN and mthDNA ligase), purified to apparent homogeneity, is sufficient to bring about complete repair of DNA uracil residues in vitro. 2. Materials and methods 2.1. Strains Escherichia coli DH5␣ (Invitrogen, Carlsbad, CA), E. coli BL21 UX [9] and E. coli BL21 UXX [8]. M. thermautotrophicus H (DSM 1053) was grown as described earlier [10]. Cells were harvested in an early stationary phase. 2.2. Plasmids pET B 001/mth212 [9]. pET-28a/mth1633 [8]. pET-28a was from Novagen (San Diego, CA). pGexRB was a kind gift from M. Konrad (Max Planck Institute for Biophysical Chemistry, Göttingen), it is based on pGex-2T (GE Healthcare, Uppsala, Sweden) but differs in the multiple cloning site. pCR-Blunt II-TOPO vector for cloning of blunt-ended PCR-products was purchased from Invitrogen (Carlsbad, CA) as part of the ‘Zero Blunt® TOPO’ kit. 2.3. Antibodies Custom rabbit antisera to heterologously produced and highly purified mthPolB [11] and Mth212 were as described previously [8]. Anti-rabbit IgG conjugated with Horseradish peroxidase was purchased from Sigma–Aldrich (Munich, Germany). 2.4. Enzymes and chemicals Restriction endonucleases, T4 DNA ligase, Klenow fragment exo− and Pfu DNA Polymerase were from Fermentas (St. Leon-Rot, Germany), chemicals were from either Roth (Karlsruhe, Germany) or Merck (Darmstadt, Germany). 2.5. Substrates All oligonucleotides were purchased from PURIMEX (Grebenstein, Germany) in preparative polyacrylamide gel electrophoresis purified grade. F: fluorescein moiety, U: 2 -d-uridine residue, P: phosphate residue.

HINDIII U T1 (30mer) 5 F-CCTGCCGAGTGCACCTGCGAAGUTTGATGT 3 HINDIII C T1 (30mer) 5 F-CCTGCCGAGTGCACCTGCGAAGCTTGATGT 3 HINDIII T T1 (30mer) 5 F-CCTGCCGAGTGCACCTGCGAAGTTTGATGT 3 HINDIII T2 (30mer) 5 P-ACATGCAGGGTCGCACGCTGTTACTGATAA 3 HINDIII LIG F (20mer) 5 F-CCTGCCGAGTGCACCTGCGA3 HINDIII LIG P (40mer) 5 P–AGCTTGATGTACATGCAGGGTCGCACGCTGTTCATGATAA 3 HINDIII 1 nt-Gap (39mer) 5 P-GCTTGATGTACATGCAGGGTCGCACGCTGTTCATGATAA 3 HINDIII 55-G (55mer) 5 CAGCGTGCGACCCTGCATGTACATCAAGCTTCGCAGGTGCACTCGGCAGGTCTAG 3 HINDIII 23-flap (23mer) 5 GGCTAGCCTCCGCTGCTGAGCTC 3 HINDIII 55-G-flap (55mer) 5 CAGCGTGCGACCCTGCATGTACATCAAGAGCTCAGCAGCGGAGGCTAGCCTCTAG 3 HINDIII M 212 (22mer) 5 F-CCTGCCGAGTGCACCTGCGAAG 3 HINDIII C 50 F (50mer) 5 F-CCTGCCGAGTGCACCTGCGAAGCTTGATGTACATGCAGGGTCGCACGCTG 3 Preparation of fluorescein-labeled 60mer oligonucleotides via ligation of HINDIII T2 with HINDIII U T1, HINDIII C T1 and HINDIII T T1, respectively, and assembly of substrates for enzymatic assays were essentially as described earlier [8]. Similarly, the 1 nt gap repair substrate was prepared from 50 pmol HINDIII LIG F, 150 pmol HINDIII 1 nt-Gap and 75 pmol of HINDIII 55 G in 100 ␮l SSC (150 mM NaCl, 15 mM trisodiumcitrate) and diluted to a total volume of 250 ␮l with H2 O. 2.6. Preparation of cell extract, immunodepletion of mthPolB and Western blot Preparation of whole-cell extract from 4 g of wet cell mass of M. thermautotrophicus H and immunodepletion of mthPolB from 1 mg of cell extract using anti-mthPolB total IgG fraction instead of anti-Mth212 total IgG fraction was essentially as described previously [8] except that the extract was subjected to four consecutive rounds of immunodepletion for DNA-U repair assay. For Western blot analysis whole-cell extracts subjected to only three consecutive rounds of immunodepletion were employed. To this end 15 ␮l whole-cell extract (150 ␮g) and equal amounts of the depleted extracts were loaded onto a 12.5% SDS-PAGE and semi-dry blotted onto a nitrocellulose membrane. For development the membrane was incubated over night with either rabbit anti-mthPolB total IgG fraction (1:1000) or anti-Mth212 total IgG fraction as control (1:1000), followed by incubation with anti-rabbit IgG Horseradish peroxidase conjugated (1:5000) as secondary antibody. Signals were detected with enhanced chemiluminescence reaction according to manual instructions (Pierce ECL Western Blotting Substrate, Thermo Scientific, Rockford, IL) and the membrane exposed to

440

L. Schomacher et al. / DNA Repair 9 (2010) 438–447

X-ray film. As marker Biorad Prestained Precision Plus Dual Color was used. 2.7. Isolation of genes by PCR and insertion into expression vector DNA from M. thermautotrophicus H was prepared by phenol/chloroform extraction and ethanol precipitation from 100 mg

wet cell mass that was resuspended in 2 ml SCE buffer (1 M sorbitol, 100 mM trisodiumcitrate, 60 mM EDTA) and sonicated on ice for 3 min with a Branson Sonifier 250 (microtip, output level 5, duty cycle 50%). For standard PCR amplification of open reading frames (orfs) from genomic DNA the following primer pairs were used (each first primer flanks 5 -region, second 3 -region of the respective genes):

Fig. 1. DNA uracil repair in cell extracts from M. thermautotrophicus H immunochemically depleted for mthPolB. (A) Schematic representation of substrate used for DNA uracil repair assay. 3 -overhangs are of 5 nt (left site) and 10 nt length (right site). F: fluorescein moiety, nt: nucleotides. Arrows hint at sites of cleavage by HindIII in case of X = C or U endonuclease in case of X = U (for rational of repair assay see Fig. 5). (B) Untreated (‘−’) or mthPolB depleted (‘+’) whole-cell extracts from M. thermautotrophicus H were analyzed for the presence of mthPolB and Mth212 (control) by Western blot as described in Section 2. Molecular masses (×10–3 ) of marker proteins are indicated on the left. Calculated relative molecular masses for the two subunits of mthPolB [11] are 67,950 and 25,500, respectively, and for Mth212 30,350. (C) Gel electrophoretic analysis of products after incubation of repair substrate with 50 ␮g M. thermautotrophicus H cell extract in presence of 20 ␮M dNTPs and 1 mM ATP in 50 ␮l assay buffer. HindIII: no incubation (‘−’) or incubation (‘+’) of reaction products with HindIII. Lengths of 5 -fluorescein-labeled oligonucleotides are indicated above the respective peaks in lanes labeled ‘marker’, running time between the markers is shown below the chromatograms. mthPolB immunodepletion: cell extracts that were immunochemically depleted of mthPolB (‘+’) or untreated (‘−’) similarly to those shown in Fig. 1B. mthPolB re-addition: cell extract after addition of 0.25 pmol of purified mthPolB. The arrows point at the HindIII cleavage product which correlates with the amount of repaired U/G mismatch in each reaction. For reaction details refer to Section 2.

L. Schomacher et al. / DNA Repair 9 (2010) 438–447

441

Table 2 Candidate repair proteins: overview of constructs, relative molecular masses, expression strains and purification procedure. orf enzyme

Plasmid construct purification tag

Mr

Expression strain

Purification

mth212 U endonuclease

pET B 001/mth212 C-terminal His6

31,380

BL21 UX

IMAC, Heparin, MonoQ

mth1208 + mth208 mthPolB

pET-28a/polB N-terminal His6 at Mth1208

70,190 25,530

BL21 UXX

IMAC, MonoQ, gel filtration

mth1312 mthPCNA

pET-28a/mth1312 N-terminal His6

30,120

BL21 UXX

IMAC, MonoQ, gel filtration

mth263 mthPPi ase

pET-28a/mth263 N-terminal His6

25,900

BL21 UXX

IMAC, MonoQ gel filtration

mth1633 mthFEN

pET-28a/mth1633 N-terminal His6

39,340

BL21 UXX

IMAC, MonoQ

mth1580 mthDNA ligase

pGexRB/mth1580 N-terminal GST

63,380

BL21 UX

GAC, MonoQ, gel filtration

Orf: open reading frame as annotated [6]. Mr : calculated monomeric relative molecular mass for each heterologously produced polypeptide including purification tag. His6 : hexa-histidine tag. GST: glutathione-S-transferase. IMAC: immobilized metal ion affinity chromatography. Heparin: Heparin affinity chromatography. MonoQ: anion exchange chromatography. Gel filtration: gel filtration chromatography. GAC: glutathione affinity chromatography. For details of production and purification refer to Section 2.

Orf mth208 (polB2): 5 CCATGGCCCAGTTGTCTAAGGTGGA 3 and 5 GGATCCTTAAAAGAACGCGTCGAGGC 3 (NcoI and BamHI sites are underlined) Orf mth1208 (polB1): 5 CATATGGAAGATTACAGAATGGTCCTCCTC 3 and 5 GGATCCCTATCCCCTGTAGGTTGCATAGAA 3 (NdeI and BamHI sites are underlined) Orf mth1312 (pcna): 5 CATATGTTCAAGGCAGAATTGAATGACC 3 and 5 GGATCCTTATTCCTCTGCCTCTATTCTT 3 (NdeI and BamHI sites are underlined) Orf mth263 (pyrophosphatase): 5 CATATGAATCTGTGGAAGGATATTG 3 and 5 GGATCCTTACTCCATATATTTTTTCC 3 (NdeI and BamHI sites are underlined) Orf mth1580 (DNA ligase): 5 GATTACTTAGCATATGAAGGAATTGCTC 3 and 5 TATTTAAATAAGATCTCTAACCTGGTTG 3 (NdeI and BglII sites are underlined) Blunt-ended PCR products were gel-purified and cloned into pCR-Blunt II-TOPO vector as described by the supplier’s manual. Genes were isolated from pCR-BluntII by cleavage with the restriction enzymes indicated. Genes mth208 (polB2), mth1208 (polB1), mth1312 (pcna) and mth263 (pyrophosphatase) were inserted into pET-28a via the corresponding restriction sites. Gene mth1580 (DNA ligase) was inserted into NdeI- and BamHI-treated pGexRB. For construction of a plasmid encoding both subunits of mthPolB as described earlier [11] frame mth1208 (polB1) together with the T7 promoter and the ribosomal binding site was isolated from pET-28a/mth1208 by BglII- and BamHI-treatment and subsequently inserted into BamHI-treated pET-28a/mth208. The resulting construct pET-28a/polB encodes for a N-terminally His-tagged PolB1 and a none tagged PolB2 subunit. Correct sequences of the cloned genes were verified by sequence analyses.

chromatography (MonoQ column 5/50 GL, GE Healthcare, Uppsala, Sweden) using 20 mM Tris/HCl pH 7.8 as running buffer with a gradient from 0 to 1 M NaCl. PolB, mthPCNA, mthPPi -ase and mthFEN were further purified by anion exchange chromatography as described above. To isolate correctly assembled proteins an additional gel filtration chromatography was carried out with mthPolB (heterodimer [11]), mthPCNA (homotrimer [12]) and mthPPi -ase (presumed as homotetramer [13]) on a HiLoad 16/60 Superdex 200 prep grade gel filtration column (GE Healthcare, Uppsala, Sweden) with 50 mM Tris/HCl pH 7.5, 100 mM KCl as running buffer at a flow rate of 1 ml/min. DNA ligase (Mth1580) was overexpressed in BL21 UX. Cleared cell lysate (25 ml) containing Glutathione-S-Transferase (GST) fused mthDNA ligase was applied to 5 ml of PBS (10 mM Na2 HPO4 , 1.8 mM KH2 PO4 pH 7.3, 140 mM NaCl, 2,7 mM KCl) equilibrated Glutathione Sepharose 4 Fast Flow (GE Healthcare, Uppsala, Sweden) and incubated under gentle shaking for 60 min at room temperature. Subsequently the mixture was passed through a 10 ml plastic syringe fitted with a plastic frit. The matrix was washed three times with 25 ml PBS. Bound proteins

2.8. Overexpression of cloned genes and purification of repair proteins Overexpression of Mth212 (U endonuclease, containing Cterminal His-tag) in BL21 UX and mthPolB (Mth1208 + Mth208), mthPCNA (Mth1312), mthPPi -ase (Mth263) and mthFEN (Mth1633) (all four containing N-terminal His-tags) in BL21 UXX, respectively, preparation of cleared cell lysate and immobilized metal ion affinity chromatography (IMAC) were essentially as described [8,9]. Additional purification steps for the proteins were as follows: Mth212 (U endonuclease) was further purified by Heparin affinity chromatography [9] and by anion exchange

Fig. 2. Purified candidate repair proteins. Final purification state of proteins heterologously produced in Ung-deficient E. coli strains is shown by a SDS-polyacrylamide gel stained with Coomassie brilliant blue R-250. M: Marker proteins (Fermentas) with corresponding molecular masses (×10−3 ). The calculated relative molecular mass (Mr ) of each protein is indicated below their names, for production and purification details refer to Section 2. Note that mthPolB consists of two different subunits.

442

L. Schomacher et al. / DNA Repair 9 (2010) 438–447

were eluted three times by 10 min incubation with 2.5 ml 50 mM Tris/HCl pH 8.0, 10 mM reduced glutathione. DNA ligase (Mth1580) was further purified by MonoQ and gel filtration chromatography as described above. Protein concentrations were determined by Bradford assays [14] using bovine serum albumin (BSA) as the standard.

2.9. Enzyme assays In general 0.25 pmol of oligonucleotide substrates were preincubated at 65 ◦ C for 10 min in a total volume of 45 ␮l 44 mM Tris/HCl pH 7.8, 5.5 mM MgCl2 , 11% PEG 6000, 110 mM NH4 Cl (assay buffer) with or without 1 nmol dNTPs (each), 100 nmol ATP or 5 nmol

Fig. 3. Individual enzymatic properties of candidate proteins for DNA uracil repair. (A, C, E, G, and I) Schematic representation of substrates and expected reaction products of individual assays (F: fluorescein moiety, nt: nucleotides). (B, D, F, H, and J) Gel electrophoretic analysis of reaction products. Sizes of marker oligonucleotides are indicated above each panel, running time between the markers is shown below each panel. All assays were carried out with 0.25 pmol of the respective substrate as explained in detail in Section 2. (A and B) DNA uridine endonuclease assay with three substrates differing in the nature of residue ‘X’ (U, C or T) and bearing 3 -overhang structures of 5 nt (left site) and 10 nt (right site). The assay was performed with 0.5 pmol of purified Mth212. (C and D) DNA polymerase assay with 20 ␮M dNTPs (each) in assay buffer, 0.125 pmol purified mthPolB and 0.625 pmol mthPCNA when indicated. 3 -overhang as in (A). (E and F) Pyrophosphatase assay. The rational of the assay is explained in the main text. Briefly, pyrophosphorolysis of a primer (same as in (C)) by Klenow fragment lacking 3 → 5 exonuclease activity is prevented by preincubation of primer/pyrophosphate mixture with an inorganic pyrophosphatase (illustrated schematically by a grey cross and grey letters in the right part of the scheme). The assay was done with 0.5 u Klenow fragment, exo− (Fermentas), 20 ␮M dNTPs (each), 100 ␮M Na2 P2 O7 and 20 pmol of mthPPi -ase as indicated in assay buffer. (G and H) 5 -flap endonuclease assay with 2.5 pmol of mthFEN as described [8]. (I and J) DNA ligase assay. P: 5 -phosphate residue, 3 -overhangs as in (A). The assay was performed with 0.25 pmol of mthDNA ligase in presence of 2 mM ATP in assay buffer. For detailed reaction conditions see Section 2.

L. Schomacher et al. / DNA Repair 9 (2010) 438–447

443

Fig. 3. (Continued ).

Na2 P2 O7 as indicated in the figures. 5 ␮l of cell extract (50 ␮g), the respective purified protein or mixtures of purified proteins (the exact amounts of proteins used are indicated in the figure legends) were added and the reaction incubated for 30 min at 65 ◦ C (final volume: 50 ␮l). The reaction was stopped by extraction with 1 vol. of TE-saturated phenol followed by a phenol/chloroform (1:1, v:v) extraction. DNA was precipitated with ethanol and dissolved in 20 ␮l H2 O. For detection of single activities, 10 ␮l loading dye (95% formamide, 20 mM EDTA pH 8.0, 50 mg/ml dextran Blue) were added to the mixture and the reaction products were subjected to gel electrophoretic analysis using an ALF DNA sequencer (GE Healthcare, Uppsala, Sweden) and the software ‘Fragment Manager’ from Pharmacia essentially as previously described [8]. To detect complete repair reactions, 10 ␮l of the reaction products dissolved in water (see above) were incubated with 10 u HindIII in a total volume of 15 ␮l reaction buffer (10 mM Tris/HCl pH 8.5, 10 mM MgCl2 , 100 mM KCl, 0.1 mg/ml BSA) for 60 min at 37 ◦ C. The reaction was stopped with 7.5 ␮l of loading dye and the products were analyzed by gel electrophoresis. 3. Results 3.1. Choice of DNA polymerase DNA repair initiated by strand incision necessarily requires the action of a DNA polymerase. Table 1 gives an overview of the different classes of DNA polymerases identified to date and their respective homologues in M. thermautotrophicus H. Obviously, mthPolB and mthPolX are both candidates for being involved in DNA uracil repair—with the two options not necessarily being mutually exclusive.

In our hands, mthPolX proved insoluble when produced by heterologous gene expression in E. coli. Furthermore, repair assays in the presence of the class-selective DNA polymerase inhibitor aphidicolin yielded inconclusive results. The relevance of mthPolB for DNA-U repair was therefore established by in vitro assays using whole-cell extract [8] combined with immunodepletion (see Fig. 1). DNA-U repair acting on the substrate shown in Fig. 1, panel (A), leads to either a 50mer (in case of run-off DNA synthesis) or a 60mer (in case of ligation “on the fly”)—also compare Fig. 5. Both products are subject to shortening by exonuclease present in the extract as observed previously [8]. The 60mer repair product can be distinguished from unreacted starting material by virtue of a HindIII restriction site introduced in the course of repair. Clearly, repair does occur with the complete extract (panel (C), right column, lane 5, compare with corresponding lane in left column and with control experiments: C/G and T/G oppositions, respectively, shown in lanes 6 and 7). Repair is abolished after four rounds of immunodepletion with polyclonal antibody against mthPolB (panel (C), right column, lane 8) and is restored after re-addition of mthPolB to the depleted extract (lane 11). The latter result proves that immunodepletion did not compromise the repair competence of the extract in an unspecific manner. Clearly, mthPolB is sufficient to sustain DNA-U repair. While an additional contribution of mthPolX is not strictly ruled out, this would either have to be quite small or be concealed in our experiment by co-depletion together with mthPolB. The latter possibility, however, seems unlikely for the following reasons: (i) The two enzymes share no significant sequence similarity, (ii) highly purified mthPolB, produced by heterologous gene expression in E. coli was used for raising the antibody and (iii) no material in the relative

444

L. Schomacher et al. / DNA Repair 9 (2010) 438–447

molecular mass region of mthPolX (Mr : 61,310) lights up in Western blot analysis using the same antibody as in immunodepletion (Fig. 1, panel (B)). 3.2. Isolation and characterization of assay components The completely sequenced genome of M. thermautotrophicus H [6] was used as the source of genes; the set of candidate enzymes chosen for the reconstitution of DNA-U repair in vitro is compiled in Table 2. PCNA was shown to enhance long patch base excision repair in combination with a B-type polymerase in vitro [15] (human Pol␦, see also Table 1). Clamp loader was not included because of the linear topology of the substrates used. Inorganic pyrophosphatase (mthPPi -ase) was selected as part of the repair machinery to account for the strong 3 → 5 exonuclease activity of Mth212 [9] which, in principle, can compete against repair synthesis and, hence, lead to a high accumulation of PPi resulting in synthesis inhibition during repair. To exclude possible contamination with host uracil-DNA glycosylase Ung, all proteins were produced in ung E. coli strains and purified to apparent homogeneity as described in Section 2. Their final purification status is illustrated in Fig. 2. One minor unaccountable band each in the preparations of mthPCNA and mthDNA ligase were neglected. Before combining the six proteins to a purified system of DNAU repair, their activities were individually assessed as described below and documented in Fig. 3.

of the mutant. If inorganic pyrophosphate is added to the mixture instead of dNTPs (lane 5), the primer is degraded to shorter oligonucleotides. This reaction can be abolished by incubating the pyrophosphate-containing assay mixture with mthPPi -ase prior to addition of the Pol I derivative (lane 6). Finally, synthesis of full-length product is restored if dNTPs are present together with inorganic pyrophosphate in the mixture before sequential incubation with pyrophosphatase and DNA polymerase. 3.2.4. 5 -Flap endonuclease (mthFEN) The assay of 5 -flap endonuclease (Orf mth1633 [17]) is illustrated in Fig. 3, panels (G) and (H); for a detailed discussion of substrate design refer to our previous publication [8]. The assay attests to the competence of the enzyme in removing the protruding 5 -flap of the substrate in an endonucleolytic fashion. 3.2.5. DNA ligase (mthDNA ligase) DNA ligase (Orf mth1580 [18]) smoothly joins two oligonucleotides, properly phosphorylated and aligned directly adjacent to one another by hybridization to a splint DNA strand (Fig. 3, panels (I) and (J)).

3.2.1. DNA-U endonuclease Mth212 Mth212 was purified as described earlier [9]. Fig. 3 (panels (A) and (B)) documents the final product of the reaction catalyzed specifically on the U/G substrate: incision at U producing a 22mer, followed by exonucleolytic trimming of the intermediate until the product is too short to hybridize to the opposite strand (Mth212 is a homologue of ExoIII). For a more detailed dissection of the multi-step reaction refer to our earlier paper [9]. 3.2.2. DNA polymerase B (mthPolB) and mthPCNA PolB (Orfs mth1208 coding for subunit PolB1 and mth208 coding for subunit PolB2) was produced using an artificial operon construct as described by Kelman et al. [11]. PolB and mthPCNA (Orf mth1312 [12]) were tested in the same combined assay (Fig. 3, panels (C) and (D)). PolB alone (panel (D), second lane from bottom) elongates most of the labeled primer to the end of the template strand. A second group of peaks corresponds to primer molecules that were shortened to lengths too small to hybridize to the template (compare to Section 3.2.1); this reaction is most easily explained by the known 3 → 5 -exonuclease activity of PolB [11]. With addition of mthPCNA (bottom lane) a qualitatively similar picture emerges, but with a pronounced shift of product distribution in favour of fulllength product. The presence of mthPCNA seems to favour primer elongation over 3 → 5 exonucleolytic recession. 3.2.3. Inorganic pyrophosphatase (mthPPi -ase) Inorganic pyrophosphatase (Orf mth263 [6]) eluted from the final gel filtration column as a mixture of dimer and trimer (against expectation, since an inorganic pyrophosphatase from M. thermautotrophicus H had previously been described as a homotetramer [13]). Experimental data are accessible online under Supplementary Fig. 1. The assay (Fig. 3, panels (E) and (F)) rests on pyrophosphorolysis of a labeled primer, catalyzed by an E. coli Pol I derivative that is deficient in both its 5 → 3 and 3 → 5 exonucleolytic activities [16]. Synthesizing activity of the DNA polymerase is demonstrated by extension of the primer to full-length product (panel (F), lane 4). With no dNTPs added, the primer is stable (lane 3) – a consequence of the 3 → 5 exonuclease deficiency

Fig. 4. Repair of a gapped DNA duplex. (A) Schematic representation of the single nucleotide gapped DNA duplex used for incision independent repair assays. P: 5 -phosphate residue, F: fluorescein moiety, nt: nucleotides, 3 -overhangs as in Fig. 1. (B) Gel electrophoretic analysis of reaction products after incubation of gapped DNA duplex in assay buffer containing dNTPs (20 ␮M each) and ATP (2 mM) with 0.25 pmol mthPolB, 2.5 pmol mthFEN, 1.25 pmol mthDNA ligase and 1.25 pmol mthPCNA as indicated. The arrows point at reaction products that represent the ligated repair fraction. For reaction details refer to Section 2.

L. Schomacher et al. / DNA Repair 9 (2010) 438–447

3.3. Repair starting from a pre-formed strand interruption In a first round of experiments, the concerted action of downstream repair functions was tested separately from strand incision. The substrate used (Fig. 4, panel (A)) is identical to the ligation substrate (Fig. 3, panel (I)), with the exception of the right upper oligonucleotide being shortened at its 5 -terminus by one residue. The one nucleotide gap in the corresponding ternary duplex makes sure that DNA ligation can only occur after some preceding DNA synthesis. PolB alone (panel (B), lane 5) leads to run-off product (50mer and 49mer), together with a series of short oligonucleotides, not unlike the situation seen with the mthPolB assay in which binary duplex was used as a substrate (Fig. 3, panels (C) and (D)). This confirms that mthPolB is not only capable of gap-filling reaction, but is also able to carry DNA synthesis into double stranded regions. Again, the simultaneous presence of mthPCNA reduces the fraction of short oligonucleotides (lane 6). If both mthDNA ligase and 5 -flap endonuclease (mthFEN) are present together with mthPolB (lane 9), a considerable portion of labeled substrate is converted to material in the full-length region (mostly 58mer). These molecular species must result from ligation “on the fly”. The shortening of the full-length product by a few residues is not surprising in view of the known 3 → 5 exonuclease activity of mthPolB even against single strands [11]—also compare Fig. 6, lane 8. The addition of mthPCNA (lane 10) makes no significant difference. This is somewhat surprising since mthPCNA did show a

445

significant effect in the PolB assay (Fig. 3, panels (C) and (D)) and one might have expected mthPCNA to increase tract lengths of DNA synthesis and hence to favour the run-off fraction. If mthFEN is omitted from the reaction mixture, the yield of fulllength products decreases to only trace amounts (compare lanes 7 and 9 or, respectively, lanes 8 and 10). This implies that the extension of mthPolB synthesis tracts into double stranded areas (compare above) is accompanied by strand displacement—not surprisingly in view of the known lack of 5 → 3 exonuclease activity of mthPolB [11]. The small amount of 58mer in lanes 7 and 8 is most simply explained as resulting from ligation immediately after elongation of the primer by just one nucleotide (i.e. before entering into strand displacement reaction). In summary, this series of experiments defines mthPolB, mthFEN and mthDNA ligase as a set of functions both necessary and sufficient to promote all salient steps of DNA repair starting from a pre-formed strand interruption. In the following experiments of complete DNA-U repair (starting from intact DNA strands and including DNA-U endonuclease Mth212) mthPolB, mthFEN and mthDNA ligase were therefore always employed as a fixed set. 3.4. Reconstitution of DNA-U repair The logic of the DNA-U repair assay is outlined in Fig. 5; data are compiled in Fig. 6. Lanes 5 to 7 in both left and right half document the expected behavior of Mth212 and HindIII on substrate and control duplexes. PolB alone produces a series of shortened

Fig. 5. Presumed DNA-U repair pathway and rational of the assay. The assay is essentially as described in our preceding publication [8]. Briefly, repair of a U/G mismatch (a) by the indicated core enzymes to a C/G base pair (e and f) is monitored by HindIII susceptibility of the reaction products (g and h). Complete repair including ligation leads to a 60mer product (e) which is distinguishable from run-off synthesis (50mer reaction product (f)). DNA synthesis is indicated by grey lines and italics.

446

L. Schomacher et al. / DNA Repair 9 (2010) 438–447

Fig. 6. Reconstitution of DNA uracil repair with purified components. Analysis of reaction products by gel electrophoresis after incubation of repair substrate with the denoted proteins in assay buffer supplemented with 20 ␮M dNTPs (each) and 2 mM ATP. HindIII: no incubation (‘−’) or incubation (‘+’) of reaction products with HindIII. Other labels as in Fig. 1. The assay was performed with 0.125 pmol Mth212 (U endonuclease), 0.25 pmol mthPolB, 2.5 pmol mthFEN, 1.25 pmol mthDNA ligase, 1.25 pmol mthPCNA and 2.5 pmol mthPPi -ase as indicated. Note: The amount of repaired U/G mismatch is represented by the HindIII cleavage product (20mer) in the right half of the figure. Detailed reaction conditions are depicted in Section 2.

substrate DNAs (Fig. 6, left half, lane 8, also compare preceding section). Under assay conditions, however, this degradation does not extend appreciably into the double stranded part of the substrate duplex. This finding is significant since it rules out the possibility that repair may be feigned by mthPolB occasionally backing up beyond the uridine residue before engaging in DNA synthesis. Lanes 9 of Fig. 6 illustrates repair brought about by the minimal set of enzymes. The group of peaks in the 50mer region (left half, lane 9) corresponds to run-off product. Upon HindIII-treatment, these peaks – as expected – disappear completely in favour of a newly formed 20mer (right half, lane 9). In contrast, a larger portion of full-length product is HindIII resistent; this fraction represents substrate that never entered the first reaction of the pathway. Tuning the assay for optimal U endonuclease activity is a delicate task because, with oligonucleotide substrates, the 3 → 5 -exonuclease activity easily becomes overwhelming (compare Fig. 3, panels (A) and (B)). Competence of the three downstream-enzymes mthPolB, mthFEN and mthDNA ligase in concert to bring about ligation “on the fly” is documented in Fig. 4, lane 9. Final qualitative proof of full repair, including ligation, came from first isolating from the gel the full-length product of the repair reactions and subsequently treating it with HindIII (Supplementary Fig. 2). In comparison to repair in whole-cell extract [8], the purified system illustrated here shifts the ratio of full-length to run-off product in favour of the latter. This could be taken as pointing to a possible involvement of a DNA polymerase other than mthPolB in bringing about repair in

the crude extract (and hence in vivo). While this possibility cannot rigorously be ruled out, we find it unlikely in view of the results of the immunodepletion experiment illustrated in Fig. 1, where an ␣-mthPolB antibody completely abolished repair which was subsequently fully recovered upon re-addition of purified mthPolB (also compare Section 3.1). Hence, we favour an unknown auxiliary factor present in the crude extract which might reduce processivity of mthPolB or otherwise facilitate ordered hand-over of the growing primer end from polymerase to ligase. Possible packaging of DNA in archaeal histones or other DNA-binding proteins [19] is one of several alternative/additional differences between whole-cell extract and purified system. For the time being, this question is left open. The addition of mthPCNA to the assay mixture has no conspicuous consequences (lanes 10 on both left and right half, compare to lanes 9) as was already observed in the preceding experiment of repair starting from a pre-formed strand interruption (compare to Section 3.3 and Fig. 4). Finally, a possible influence of mthPPi -ase on yield and product distribution was investigated—without and with the additional presence of mthPCNA (left and right half, lanes 11 and 12). Again, mthPCNA makes no difference but pyrophosphatase increases repair yield—selectively, as it seems, via run-off product. This observation cannot be explained in a straightforward fashion as a consequence of the chemical action of the enzyme alone—especially since dNTPs are present in large excess over substrate. As a set, the four M. thermautotrophicus H enzymes Mth212, mthPolB, mthFEN and mthDNA ligase are both necessary and sufficient to bring about complete repair of pre-mutagenic U/G lesions.

L. Schomacher et al. / DNA Repair 9 (2010) 438–447

4. Discussion Until recently, the exchange of a uracil residue in a premutagenic U/G mispair in DNA by cytosine was thought to be accomplished by base excision repair exclusively throughout all domains of life. In contrast to this paradigm, we demonstrated previously an alternative repair entry mechanism that omits uracil excision but rather initiates repair by direct strand incision thus simplifying DNA uracil repair [8]. The reconstitution of the entire repair pathway from purified components presented in this study corroborates this finding and identifies the core molecular components involved downstream of strand incision for repair completion. Repair synthesis is achieved by mthPolB, a member of class B DNA polymerases which are known to be involved in long patch base excision repair [15,20], nucleotide excision repair [21] and mismatch repair [22] in eukaryotic organisms. Also, the requirement of a flap endonuclease to trim the 5 -flap intermediate generated by repair synthesis is in accordance with the eukaryotic repair system [15]. Hence, our findings add yet another element to the growing list of similarities between macromolecular DNA metabolism in the archaeal and eukaryotic domains of life [23]. Additionally, we observed a stimulating influence of inorganic pyrophosphatase on the amount of repair products. Since pyrophosphorolysis is unlikely to play a significant role in our assay, the effect may be explained by a structural or regulatory role of this protein on the core repair enzymes. A similar role of inorganic pyophosphatase was already proposed in connection with a eukaryotic nucleosome remodeling complex [24]. Further investigation is needed to clarify this interesting feature in more detail. In contrast to eukaryotic long patch base excision repair [15,20] we could not observe a significant influence of mthPCNA on repair yield and product distribution. This absence of a measurable effect may, however, be simply due to the somewhat arbitrarily chosen reaction conditions or to the oligonucleotides employed in the assay being too short. Taken together the results of the present work and our previous study [8] establish a second (non-BER) paradigm of DNA uracil repair in which glycosylytic uracil excision is bypassed by direct strand incision next to the uracil residue. Interestingly, this hitherto unique function has been implemented in the major APendonuclease of the cell (Mth212, an ExoIII homologue), i.e. the general strand incision enzyme in BER. Future work has to clarify, if this repair mechanism represents just one evolutionary maverick or whether it reflects in a more general way special requirements of genetic adaptation posed by conditions such as life at elevated temperatures. Conflict of interest statement The authors declare that there are no conflicts of interest. Acknowledgement We thank Chritiane Preiß for expert technical assistance. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.dnarep.2010.01.004. References [1] T. Lindahl, Instability and decay of the primary structure of DNA, Nature 362 (1993) 709–715. [2] E. Seeberg, L. Eide, M. Bjoras, The base excision repair pathway, Trends Biochem. Sci. 20 (1995) 391–397.

447

[3] L.H. Pearl, Structure and function in the uracil-DNA glycosylase superfamily, Mutat. Res. 460 (2000) 165–181. [4] V. Starkuviene, H.J. Fritz, A novel type of uracil-DNA glycosylase mediating repair of hydrolytic DNA damage in the extremely thermophilic eubacterium Thermus thermophilus, Nucleic Acids Res. 30 (2002) 2097–2102. [5] A.A. Sartori, S. Fitz-Gibbon, H. Yang, J.H. Miller, J. Jiricny, A novel uracil-DNA glycosylase with broad substrate specificity and an unusual active site, EMBO J. 21 (2002) 3182–3191. [6] D.R. Smith, L.A. Doucette-Stamm, C. Deloughery, H. Lee, J. Dubois, T. Aldredge, R. Bashirzadeh, D. Blakely, R. Cook, K. Gilbert, et al., Complete genome sequence of Methanobacterium thermoautotrophicum deltaH: functional analysis and comparative genomics, J. Bacteriol. 179 (1997) 7135–7155. [7] A.I. Slesarev, K.V. Mezhevaya, K.S. Makarova, N.N. Polushin, O.V. Shcherbinina, V.V. Shakhova, G.I. Belova, L. Aravind, D.A. Natale, I.B. Rogozin, et al., The complete genome of hyperthermophile Methanopyrus kandleri AV19 and monophyly of archaeal methanogens, Proc. Natl. Acad. Sci. U.S.A. 99 (2002) 4644–4649. [8] L. Schomacher, J.P. Chong, P. McDermott, W. Kramer, H.J. Fritz, DNA uracil repair initiated by the archaeal ExoIII homologue Mth212 via direct strand incision, Nucleic Acids Res. 37 (2009) 2283–2293. [9] J. Georg, L. Schomacher, J.P. Chong, A.I. Majernik, M. Raabe, H. Urlaub, S. Muller, E. Ciirdaeva, W. Kramer, H.J. Fritz, The Methanothermobacter thermautotrophicus ExoIII homologue Mth212 is a DNA uridine endonuclease, Nucleic Acids Res. 34 (2006) 5325–5336. [10] C.P. Guy, A.I. Majernik, J.P. Chong, E.L. Bolt, A novel nuclease-ATPase (Nar71) from archaea is part of a proposed thermophilic DNA repair system, Nucleic Acids Res. 32 (2004) 6176–6186. [11] Z. Kelman, S. Pietrokovski, J. Hurwitz, Isolation and characterization of a split B-type DNA polymerase from the archaeon Methanobacterium thermoautotrophicum deltaH, J. Biol. Chem. 274 (1999) 28751–28761. [12] Z. Kelman, J. Hurwitz, A unique organization of the protein subunits of the DNA polymerase clamp loader in the archaeon Methanobacterium thermoautotrophicum deltaH, J. Biol. Chem. 275 (2000) 7327–7336. [13] G.J. van Alebeek, J.T. Keltjens, C. van der Drift, Purification and characterization of inorganic pyrophosphatase from Methanobacterium thermoautotrophicum (strain delta H), Biochim. Biophys. Acta 1206 (1994) 231–239. [14] M.M. Bradford, A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem. 72 (1976) 248–254. [15] A. Klungland, T. Lindahl, Second pathway for completion of human DNA base excision-repair: reconstitution with purified proteins and requirement for DNase IV (FEN1), EMBO J. 16 (1997) 3341–3348. [16] V. Derbyshire, P.S. Freemont, M.R. Sanderson, L. Beese, J.M. Friedman, C.M. Joyce, T.A. Steitz, Genetic and crystallographic studies of the 3 ,5 exonucleolytic site of DNA polymerase I, Science 240 (1988) 199–201. [17] M.W. Kaiser, N. Lyamicheva, W. Ma, C. Miller, B. Neri, L. Fors, V.I. Lyamichev, A comparison of eubacterial and archaeal structure-specific 5 -exonucleases, J. Biol. Chem. 274 (1999) 21387–21394. [18] V. Sriskanda, Z. Kelman, J. Hurwitz, S. Shuman, Characterization of an ATP-dependent DNA ligase from the thermophilic archaeon Methanobacterium thermoautotrophicum, Nucleic Acids Res. 28 (2000) 2221– 2228. [19] K. Sandman, J.N. Reeve, Archaeal chromatin proteins: different structures but common function? Curr. Opin. Microbiol. 8 (2005) 656–661. [20] M. Stucki, B. Pascucci, E. Parlanti, P. Fortini, S.H. Wilson, U. Hübscher, E. Dogliotti, Mammalian base excision repair by DNA polymerases delta and epsilon, Oncogene 17 (1998) 835–843. [21] A. Aboussekhra, M. Biggerstaff, M.K. Shivji, J.A. Vilpo, V. Moncollin, V.N. Podust, M. Protic, U. Hübscher, J.M. Egly, R.D. Wood, Mammalian DNA nucleotide excision repair reconstituted with purified protein components, Cell 80 (1995) 859–868. [22] M.J. Longley, A.J. Pierce, P. Modrich, DNA polymerase delta is required for human mismatch repair in vitro, J. Biol. Chem. 272 (1997) 10917– 10921. [23] A.I. Majernik, E.R. Jenkinson, J.P. Chong, DNA replication in thermophiles, Biochem. Soc. Trans. 32 (2004) 236–239. [24] D.A. Gdula, R. Sandaltzopoulos, T. Tsukiyama, V. Ossipow, C. Wu, Inorganic pyrophosphatase is a component of the Drosophila nucleosome remodeling factor complex, Genes Dev. 12 (1998) 3206–3216. [25] P.M. Burgers, E.V. Koonin, E. Bruford, L. Blanco, K.C. Burtis, M.F. Christman, W.C. Copeland, E.C. Friedberg, F. Hanaoka, D.C. Hinkle, et al., Eukaryotic DNA polymerases: proposal for a revised nomenclature, J. Biol. Chem. 276 (2001) 43487–43490. [26] J. Filee, P. Forterre, T. Sen-Lin, J. Laurent, Evolution of DNA polymerase families: evidences for multiple gene exchange between cellular and viral proteins, J. Mol. Evol. 54 (2002) 763–773. [27] I.K. Cann, Y. Ishino, Archaeal DNA replication: identifying the pieces to solve a puzzle, Genetics 152 (1999) 1249–1267. [28] L. Haracska, R.E. Johnson, L. Prakash, S. Prakash, Trf4 and Trf5 proteins of Saccharomyces cerevisiae exhibit poly(A) RNA polymerase activity but no DNA polymerase activity, Mol. Cell Biol. 25 (2005) 10183–10189. [29] T. Uemori, Y. Sato, I. Kato, H. Doi, Y. Ishino, A novel DNA polymerase in the hyperthermophilic archaeon, Pyrococcus furiosus: gene cloning, expression, and characterization, Genes Cells 2 (1997) 499–512. [30] Y. Ishino, K. Komori, I.K. Cann, Y. Koga, A novel DNA polymerase family found in Archaea, J. Bacteriol. 180 (1998) 2232–2236.