MSH2 is essential for the preservation of genome integrity and prevent
http://www.100md.com
《核酸研究医学期刊》
Station de Génétique et d'Amélioration des Plantes, INRA Route de St Cyr, 78026 Versailles, France 1Département de biologie moléculaire végétale, Université de Lausanne CH-1015 Lausanne, Switzerland
*To whom correspondence should be addressed. Tel: +33 1 30833009; Fax: +33 1 30833319; Email: nogue@versailles.inra.fr
ABSTRACT
MSH2 is a central component of the mismatch repair pathway that targets mismatches arising during DNA replication, homologous recombination (HR) and in response to genotoxic stresses. Here, we describe the function of MSH2 in the moss Physcomitrella patens, as deciphered by the analysis of loss of function mutants. Ppmsh2 mutants display pleiotropic growth and developmental defects, which reflect genomic instability. Based on loss of function of the APT gene, we estimated this mutator phenotype to be at least 130 times higher in the mutants than in wild type. We also found that MSH2 is involved in some but not all the moss responses to genotoxic stresses we tested. Indeed, the Ppmsh2 mutants were more tolerant to cisplatin and show higher sensitivity to UV-B radiations. PpMSH2 gene involvement in HR was studied by assessing gene targeting (GT) efficiency with homologous and homeologous sequences. GT efficiency with homologous sequences was slightly decreased in the Ppmsh2 mutant compared with wild type. Strikingly GT efficiency with homeologous sequences decreased proportionally to sequence divergence in the wild type whereas it remained unaffected in the mutants. Those results demonstrate the role of PpMSH2 in the maintenance of genome integrity and in homologous and homeologous recombination.
INTRODUCTION
The mismatch repair (MMR) system is evolutionarily highly conserved and plays an essential role in maintaining genome stability (1). In all organisms, MMR is best known for its role in the post-replication repair of DNA polymerization errors. It keeps the rate of mutations due to nucleotide misincorporation and polymerase slippage at an acceptable low level (2). MMR proteins also recognize mismatches in heteroduplex recombination intermediates. In somatic cells, they display an antirecombination activity that inhibits recombination between homeologous sequences present in a single genome (such as allelic genes in diploid cells and members of multigene families) and thus preserves genome integrity (2). This activity also prevents genetic recombination between different but related species and thus contributes to the definition of species. Additionally, MMR plays a role in some types of nucleotide excision repair, which are responsible for repair of physical and chemical damage to DNA. It also participates in a cell-cycle checkpoint control system by recognizing certain types of DNA damage and promoting cell-cycle arrest or triggering apoptosis pathways (3,4).
The methyl-directed MMR system of Escherichia coli is composed of MutS and MutL proteins (1,3). The MutS proteins are conserved from bacteria to mammals. In eukaryotes MutS homologs (MSH) are encoded by small multigene families, whose members can associate into heterodimers that have discrete roles in MMR-related processes (2). For example, MSH2/MSH6 heterodimers recognize and stimulate the repair of single base-pair mismatches while MSH2/MSH3 heterodimers recognize small insertion/deletion loops (2). The MSH2 protein is a central component of the eukaryotic MMR system and is present in all heterodimers. It is crucial for the repair of all mismatched lesions, whereas other MSH proteins modulate the function of MSH2 depending on the different lesion types or developmental stages.
The msh2 mutants in yeast and mammals show (i) microsatellite instability, (ii) a mutator phenotype characterized by a high spontaneous mutation rate, (iii) an increased recombination frequency between diverged DNA sequences (homeologous) and (iv) tolerance to certain types of chemical and physical treatments that damage DNA (5–8). In Caenorhabditis elegans, loss of MSH2 is associated with microsatellite instability, a mutator phenotype and reduced fertility (9).
However, in plants, the effect of MSH2 mutations is still poorly understood. In Arabidopsis thaliana, Hoffman et al. (10) have shown rapid accumulation of mutations and microsatellite instability during seed-to-seed propagation of Atmsh2 defective lines. However, the molecular tools are not available to evaluate the role of MSH2 in the mechanism of homologous recombination (HR) in A.thaliana and other higher plants. This type of study in plants is currently only possible in the moss Physcomitrella patens.
The moss P.patens is unique in the plant kingdom in that it allows high efficiency of gene targeting (GT) via HR (11). Thus, it is possible to use gene knock-out and allele replacement methods in this organism. This allows to examine for the first time the importance of MMR in a multicellular organism in which targeted integration by HR is a major transformation pathway . Recently, we have isolated the PpMSH2 gene (12). We now have generated Ppmsh2 disruptants to examine the role of MSH2 in the DNA metabolism of P.patens. Here, we show that PpMSH2 is involved in the control of genome integrity and that it is a barrier to recombination between divergent DNA sequences. Furthermore, in the absence of PpMSH2, moss development is dramatically affected most likely as a consequence of a strong mutator phenotype that is directly revealed by the dominance of the haplophase during its development.
MATERIALS AND METHODS
Plant material and culture conditions
The Gransden wild-type strain of P.patens (13) was used in this study. Protonemal tissue was propagated on PpNO3 medium (14), supplemented with 2.7 mM NH4-tartrate (standard medium). Cultures were grown in 9 cm Petri dishes on medium solidified with 0.7% Agar (Biomar) and overlaid with a cellophane disk (Cannings, Bristol). Cultures were illuminated with a light regime of 16 h light/8 h darkness and a quantum irradiance of 80 μE m–2 s–1 (standard conditions). Sporogenesis was performed in Magenta box in which the tested strains were grown aside the self sterile but cross fertile strain nicB5ylo6 (13) on minimal PpNO3 medium. Crosses were grown in standard conditions for 6–8 weeks (i.e. until about 50 gametophores reached full development), then irrigated with sterile water and transferred for 2 weeks in growth chambers set at 15°C with 10 h of light per day and a quantum irradiance of 15 μE m–2 s–1. The development of archegonia, antheridia and spore capsules was followed visually during the next month. Antheridia and archegonia were manually dissected for further microscopic observation.
Molecular cloning
We used standard methods for all molecular cloning (15). Using genomic P.patens DNA as starting template, we amplified a 2340 bp PCR fragment covering an internal fragment of the MSH2 genomic region (Gene Bank: DQ117988) and inserted it into the TA-cloning vector pCR?II (Invitrogen, Groningen, The Netherlands) to produce PpMSH2 (oligonucleotide sequences can be provided on request). To obtain PpMSH2-KO, a 465 bp 5' targeting fragment (coordinate 2557–3022 in DQ117988 ) and a 598 bp 3' targeting fragment (coordinate 4614–5212 in DQ117988 ) were recovered by restriction from PpMSH2 and inserted upstream (5') and downstream (3') of the loxP sites flanking the resistance cassette of pBSK35S-NPTII-lox (supplied by Dr Laloue, INRA Versailles). Vector pBSK35S-NPTII-lox carries a neomycin phosphotransferase resistance gene driven by the 35S CaMV promoter and flanked by two LoxP sites in direct orientation. The antibiotic resistance cassette is in the same orientation with regard to the transcription of PpMSH2.
To obtain PpAPT-KO, a 3995 bp Xba1 and EcoRI fragment covering the APT genomic sequence (Gene Bank: DQ117987), was inserted in pBlueScript KS (Stratagene), forming PpAPT1. The ATG codon is 1217 bp 3' of the XbaI site and the stop codon is 881 bp 5' of the EcoRI site in this PpAPT fragment. An internal 1102 bp BamHI fragment was deleted by digestion of the PpAPT1 vector, to produce PpAPT. We generated homeologous variants of this targeting cassette by mutating the PpAPT sequence by mutagenic PCR using the Cadwell and Joyce (16) protocol. Successive APT variants were generated until we reached 3% divergence within the PpAPT original fragment. A BamHI fragment containing the hygromycin phosphotransferase gene (HptII) driven by the 35S CaMV promoter was recovered from plasmid pGL2 (17) and ligated into the BamHI site of PpAPT and of the variants to produce the homologous and homeologous PpAPT-KO vectors (Figure 4).
Figure 4 Gene targeting PpAPT fragments used for homologous and homeologous recombination studies. The hygromycin gene resistance cassette is represented between both PpAPT fragments by a horizontal interrupted line. (A) PpAPT fragments in the PpAPT-KO cassette in which the sequences are exactly identical to the locus PpAPT used as reporter gene. (B) Alignments of PpAPT fragments in homeologous recombination cassettes; each vertical line indicates mismatch between mutated PpAPT fragments and the original locus.
Protoplasts isolation and regeneration
Protoplasts were isolated from 6-day-old protonema by incubation for 40 min in 1% Driselase (Fluka 44585) dissolved in 0.47 M mannitol. The suspension was filtered successively through 80 and 40 μm stainless steel sieves. Protoplasts were sedimented by low-speed centrifugation (600 g for 5 min at 20°C) and washed twice with 0.47 M mannitol. Protoplasts were then resuspended at 1.2 x 106 protoplasts/ml in MMM solution (0.47 M mannitol, 15 mM MgCl2 and 0.1% MES, pH 5.6) for transformation. For growth assay and phenotypic analyses, protoplasts were resuspended at a density of ca. 1–3 x 104 protoplasts/ml in mannitol 0.47 M containing 0.5% agar. Two ml of suspension were spread on cellophane disks covering standard medium supplemented with 0.5% glucose and 0.36 M mannitol (PPM). Regeneration was conducted in standard growth conditions. After 1 week of regeneration, protoplasts were transferred to standard medium without mannitol.
Transformation of protoplasts
Moss protoplasts (1.2 x 106) were transformed as described previously (18) with PpMSH2-KO digested with AgeI and AatII, which generate DSBs within the genomic MSH2 sequence. Primary transformants were selected in the presence of 50 mg/l paramomycin (Duchefa P0141, Haarlem, The Netherlands). Small pieces of the protonema tissue (2 mm in diameter) of primary P.patens transformants were then transferred onto standard medium without selection and cultured for 15 days. A second round of selection allowed the isolation of stable transformants, by transferring small pieces of protonema tissue from the previous colonies onto standard medium containing 50 mg/l G418 (Duchefa G0175, Haarlem, The Netherlands) (paramomycin and G418 belong to the kanamycin family of antibiotics).
Molecular analysis of Msh2 knockouts
Moss DNA was extracted from 200 mg of fresh protonema tissue ground in 220 μl of extraction buffer (220 mM Tris–HCl, pH 7.5, 250 mM NaCl, 25 mM EDTA and 0.5% SDS) with an electric pestle in the presence of sand. After centrifugation, DNA was precipitated with isopropanol at room temperature. The pellet was dried and resuspended with 100 μl sterile water. Stable disruptants of the PpMSH2 gene were screened by PCR for the incorporation of the resistance cassette into the MSH2 locus. We used primers specific for the resistance cassette (Uc: TATTTTTGGAGTAGACAAGCGTGTCGT and Lc: CCGCTTCCTCGTGCTTTACGGTAT) and primers specific for the PpMSH2 sequence outside the transformation construct (PpMSH2#4: CAATGTCACTGAGAGCAAAACG and PpMSH2#5: CACACTGATCTCTGCTCGGTCG).
Moss RNA was isolated using the RNAeasy kit (Qiagen) starting with 100 mg of fresh protonema. We obtained doubled-stranded cDNAs corresponding to mRNAs expressed in 6-day-old protonema cultures using the SMART-PCR cDNA Synthesis Kit (Clontech). The primers PpMSH2#4 and PpMSH2#2 (CACCAACACGCGCAAAGATG) were used for PCR on these cDNAs.
Generation of clean deletion by transient Cre recombinase expression
Direct repeats of LoxP recombination sites from the Cre/lox system (19) were located on either side of the 35S-nptII resistance cassette in PpMSH2-KO. Transient expression of a constitutive Cre expression cassette (20) in protoplasts allowed excision of the resistance gene from the PpMSH2 recombinant locus (D.G. Schaefer, A. Finka and J.P. Zr?d, manuscript in preparation). Protoplasts were regenerated at a low density on non-selective medium for 2 weeks and fragments of single protoplast-derived colonies were then replicated on selective (i.e. G418 50 mg/l) and non-selective medium. Loss of antibiotic resistance was determined after an additional week of growth. Replicates growing on non-selective medium were used for subsequent amplification and molecular analyses. Clone Ppmsh2 was selected this way and used for further analyses. We analyzed by PCR the recombinant locus with the primers PpMSH2#4 and PpMSH2#2.
Cytology
Binocular observations were made with a Nikon SMZ1000, and microscopic observations with a Leitz (type 090-122.012) inverted microscope.
Evaluation of spontaneous mutation frequency
Mutations in the APT gene confer resistance to 2-FA (2-Fluoroadenine, from Dr Laloue, INRA Versailles, France), a toxic compound for cells. The number of 2-FA resistant colonies reflects the frequency of spontaneous mutations. Protoplasts of wild-type and Ppmsh2 mutants were regenerated for 6 days regeneration on PPM (105/Petri dish) and then transferred on standard solid medium supplemented with 10 μM 2-FA. After 2 weeks, the number of resistant colonies was counted. We repeated this experiment three times. The APT genomic sequence from 2-FA resistant clones was subsequently PCR amplified and sequenced to identify the mutation responsible for the resistance (oligonucleotide sequences can be provided on request).
Sensitivity to cisplatin/MNU and /UV-B radiations
Cisplatin (cis-diamminedichloroplatinum(II)) (Calbiochem, Darnstadt, Germany), MNU (N-methyl-N-nitrosourea) and O6-bzGua (O6-benzylguanine) (Sigma-Aldrich, Steinheim, Germany) were dissolved in dimethyl sulfoxide at 100 mM each for storage at –20°C and at 25 mM for O6-bzGua immediately before use.
Wild-type and Ppmsh2 mutant protoplasts in liquid PPM were treated for 20 h with 0–5 mM cisplatin or MNU (2 x 105 protoplasts at each concentration). From 1 h before exposure to MNU and throughout the whole procedure, 25 μM O6-bzGua was present in the medium to inhibit the removal of methyl groups from the O6 position of guanine by endogenous methyl-transferase activity. After 20 h in the dark, protoplasts were washed and spread on solid medium at low concentrations.
Protoplasts in liquid mannitol medium were exposed to ionizing radiation (1 Gy/s) using a 137Cs irradiator (IBL-637 CIS-Bio-International, Institut Curie, Orsay). After 20 h in the dark, the protoplasts were spread on solid medium at low concentrations.
Protoplasts spread on solid medium (5 x 104 protoplasts at each concentration in four Petri dishes) were exposed to UV-B light (60 J/m2/s) from a 312 nm TFX lamp. We calculated the flux with a UV-Elektronik GmbH dosimeter. The protoplasts were then left in the dark for 20 h.
After 6 days regeneration, protoplasts were transferred onto standard solid medium. After 1 week, the number of survivors was counted. We repeated these experiments two to three times. Statistical analyses have been made with the 2 test.
APT targeting frequency with homologous and homeologous DNA
Targeting APT fragments were produced by digestion of PpAPT-KO vector with NotI and EcoRI enzymes. We selected stable transformants after transformation of wild-type and Ppmsh2 mutant protoplasts with homologous and homeologous APT disruption fragments. Small pieces of protonema tissue (2 mm in diameter) from stable transformants were then transferred onto standard medium containing 10 μM of 2-Fluoroadenine to detect APT gene targeting events. Experiments were repeated three times.
RESULTS
Targeted disruption of the MSH2 gene in P.patens
Moss protoplasts were transformed with vector PpMSH2-KO and four independent clones were selected for further investigation (Ppmsh2#1 to Ppmsh2#4). In each clone, disruption of the chromosomal PpMSH2 locus was controlled by PCR and absence of the PpMSH2 transcript was established by RT–PCR (see Ppmsh2#1 for example Figure 1B–D). In order to obtain a simple deletion mutant, the resistance cassette inserted at the PpMSH2 locus of the Ppmsh2#1 clone was eliminated by the CRE recombinase to produce clone Ppmsh2 (see Materials and Methods and Figure 1A). In this clone, we confirmed deletion of exon 7 of the PpMSH2 gene by PCR and checked modification of the transcript by RT–PCR (Figure 1C and D). The four independent PpMSH2 knockouts (Ppmsh2#1 to Ppmsh2#4) and the deletion mutant (Ppmsh2) behaved identically under all circumstances tested except in aged cultures. Therefore, Ppmsh2 mutant was used for further characterization except for late developmental phenotypes (see below). To avoid the genetic drift resulting to the expected mutator phenotype associated with msh2 mutations, all the experiments were performed using the primary stock of protonema suspension from Ppmsh2 stored at 4°C as starting material.
Figure 1 Molecular analysis of Ppmsh2 mutants. (A) Schematic representation of PpMSH2 genomic locus, schematic representation of PpMSH2 disrupted locus with a double crossover insertion and schematic representation of PpMSH2 disrupted locus without resistance gene. Black open boxes represent exons, black lines represent introns, gray open boxes are homology regions used to disrupt the gene, black boxes are Lox sites for Cre recombination and arrows show the primers used for PCR. Ppmsh2? is mutant without the resistance gene, resulting from Ppmsh2#1 after Cre recombinase expression. (B) PCR of the recombinant borders with a primer external to the targeting sequences and one on the resistance gene . (C) PCR with primers flanking recombination sites (primers: PpMSH2#4/2), (D) RT–PCR for PpMSH2 transcripts (primers: PpMSH2#4/2), PpAPT was used as control.
Ppmsh2 mutants display pleiotropic developmental defects
The haploid gametophytic development of P.patens from regenerating protoplasts starts with a filamentous branched structure called the protonema that grows by apical cell divisions and lateral division of subapical cells. There are two cell types in protonema filaments: chloronema and caulonema. Regenerating protoplasts first develop into photosynthetically active chloronemal filaments, formed by cells filled with chloroplasts and separated by walls perpendicular to the axis of the filaments. After 5 days, chloronemal apical cells occasionally differentiate into caulonemal cells characterized by oblique cell walls and smaller and fewer chloroplasts. After 8–10 days, primitive meristems called buds differentiate from caulonemal side branch initials and further differentiate into leafy shoots, also known as gametophores. Each leafy shoot resembles a small plant with an unbranched stem, filamentous basal and aerial rhizoids and leaves displaying clear phyllotaxis along the stem axis (21). We found that the msh2 mutants had several pronounced phenotypes affecting the development of both protonema and leafy shoots.
Gametophytic developmental failures
Preliminary observations of msh2 mutant colonies regenerated from protoplasts indicated that these colonies were larger than those obtained from wild-type protoplasts (data not shown). To further quantify this observation, we compared the size of protonema filaments by measuring the length of the three apical cells from 14-day-old regenerated colonies. We found a significant difference in the mean length between the wild-type and the Ppmsh2 mutant (4.2 ± 1 versus 7.2 ± 2, arbitrary units, t-test P < 0.0001, n = 40) (Figure 2A). These data suggest that the Ppmsh2 mutant cells elongate 1.5 times faster than the wild-type cells. We next evaluated the cell division rate in the two backgrounds by counting the number of cell divisions after 8 days from regenerated protoplasts. We found a significant difference for mean cell division numbers between the wild-type and for the Ppmsh2 mutant (5 ± 0.5 versus 6 ± 0.8, t-test P < 0.0001) (Figure 2B), showing that the number of cell divisions was slightly increased in the mutant.
Figure 2 Comparison of wild type and Ppmsh2 mutants apparent phenotypes: (A) Comparison of the length of the last three cells of protonema filaments; arrows show the cell walls. (B) Number of cell divisions after 8 days protoplast regeneration; arrows show the cell walls. (C) Wild-type and Ppmsh2 mutant colonies after 3 weeks development. (D) Male antheridia of wild-type and of Ppmsh2 mutant, (E) female archegonia of wild-type and of Ppmsh2 mutant. (F) Wild-type, Ppmsh2#1, Ppmsh2#3 and Ppmsh2#2 mutant colonies after 12 weeks development, a representative leafy shoot from those colonies and enlargement of the top of the same gametophore.
Differentiation of chloronema cells into caulonema cells looks normal in the mutant compared with wild type. After 10 days of growth, differentiation of buds and further development of leafy shoots could clearly be observed in WT colonies whereas, at the same time-point, the Ppmsh2 mutant colonies had neither developed buds nor leafy shoots and the protonema was more widespread (data not shown). After 3 weeks, wild-type colonies had numerous well-developed leafy shoots, whereas Ppmsh2 filaments had just developed a few leafy shoots that were thinner, shorter and occasionally distorted (Figure 2C). We therefore concluded that loss of function of MSH2 seriously impairs the normal development of P.patens at different levels: protonema cell growth is accelerated, protonema cell cycle may be shorter, differentiation of buds is delayed and further development of leafy shoots is architecturally affected.
Reproductive developmental failures
The late developmental stages of P.patens life cycle are the differentiation of reproductive organs, fertilization to produce the diploid sporophyte and meiosis of spore mother cells to produce haploid spores. Therefore, we examined whether the developmental abnormalities seen at the protonema and leafy shoot stages could also be observed in antheridia (male) and archegonia (female), the reproductive organs of P.patens that differentiate at the apex of leafy shoots. After 10 weeks, the reproductive organs of wild type were fully developed on leafy shoot (Figure 2D and E) and the sporophyte further developed normally to produce viable spores (data not shown). In all five Ppmsh2 mutants, abnormal reproductive organs were formed on the few leafy shoots (Figure 2C), with aberrant number and shape of cells (Figure 2D and E). Consequently we never obtained differentiated capsules from the mutants and concluded that loss of function of MSH2 led to complete sterility in P.patens. To determine whether sterility was due to a problem in male and/or female gametes, reciprocal out-crosses were performed between Ppmsh2 mutants and strain NicB5ylo6. We did not obtain any capsules from these crosses and thus concluded that Ppmsh2 mutants were male and female sterile as a consequence of abortive development of reproductive organs.
Late developmental defects
In order to check the phenotype of late growing colonies, we inoculated similar size filaments (5–10 cells) from 3-week-old single protoplasts derived colonies on standard medium. After 6 weeks, the mutant colonies started to become chlorotic and a significant fraction of the cells were dead (data not shown). After 12 weeks, wild-type colonies were healthy (Figure 2F, WT) whereas those from independent Ppmsh2 knockouts presented various developmental aberrations from leaf shape abnormalities (Figure 2F, Ppmsh2#1) or supernumerary leaves and ‘callus’-like structure (Figure 2F, Ppmsh2#3) to completely disorganized necrotic shoots with numerous aerial rhizoids (Figure 2F, Ppmsh2#2). Additionally, the frequency of dead cells in these colonies was directly proportional to the intensity of the phenotype. In the most dramatic case (Ppmsh2#2) the leafy shoots were completely dead since isolated gametophores plated on fresh standard medium were not able to regenerate viable protonema, which was feasible with WT leafy shoots isolated from the same culture (data not shown). We also observed that the intensity of the phenotypes increased proportionally to the number of subcultures the original clone has undergone (data not shown), suggesting that random mutations accumulated within Ppmsh2 mutant clones in successive subcultures. Those late phenotypes could clearly be associated with major developmental processes, such as the establishment of shoot architecture, early senescence and eventually active cell death. The diversity of the phenotypes observed in Ppmsh2 mutants supports a strong mutator phenotype that is immediately detected in the moss haploid gametophyte. For this reason all the following experiments have been carried out using 3–4 individual replicates of the primary isolated knockout clones (stored at 4°C) as starting material.
Ppmsh2 mutants have a mutator phenotype
The strong developmental defects observed in Ppmsh2 mutants could be related to accumulation of spontaneous mutations. This phenotype has already been observed in msh2 mutants of other organisms (2). To further quantify this mutator phenotype, we investigated the effect of the loss of the PpMSH2 gene on mutation frequency of the adenine phosphoribosyl transferase (PpAPT) reporter gene. Spontaneous mutations leading to loss of APT activity confers resistance to adenine analogues, such as 2-fluoroadenine (2-FA) and can be selected on standard medium supplemented with 10 μM 2-FA. We tested a total of 107 Ppmsh2 protoplast-derived colonies and 3 x 106 wild-type colonies for resistance to 2-FA.
We observed no resistant colonies in the wild type and a total of 398 2-FA resistant colonies in the Ppmsh2 mutant. APT mutation frequencies were thus lower than 3 x 10–7 in wild type and 4 x 10–5 ± 1.2 x 10–5 in Ppmsh2. To confirm that 2-FA was effectively resulting from mutation in the APT gene, we partially sequenced the APT gene of randomly chosen 2-FA resistant clones. Point mutations in the APT sequence were effectively detected in more than 75% of the clones tested (Table 1). Mutations resulting from transitions and transversions were recovered, but we did not identify mutations resulting from insertion or deletion. These modifications were matching with highly conserved domain or amino acid of the APT protein and some of them were identified several times in independent 2-FA resistant clones. These results demonstrate that loss of MSH2 gene leads to the accumulation of point mutations in the moss genome and that the spontaneous mutation rate is increased at least by two orders of magnitude (130 times) in the Ppmsh2 mutant compared with the wild type. This last value is probably underestimated since the screen used here has selected only mutations that confer 2-FA resistance. Additionally, the type of mutations identified suggests that they result from defaults of post DNA replication proofreading processes.
Table 1 Point mutations identified in the APT genomic sequence in Ppmsh2 2-FA resistant clones
Msh2 mutants display altered responses to genotoxic stresses
MMR is known to recognize different types of DNA damages depending on the lesion and on the organism. MMR is responsible for the hypersensitivity to agents, such as radiation or some genotoxic chemicals that create DNA lesions (22). Loss of MMR may prevent cell death normally resulting from exposure to such genotoxic agents, thus conferring tolerance to these stresses. Therefore, we treated Ppmsh2 mutant protoplasts with UV-B and -radiation and chemicals such as cisplatin and MNU to determine whether P.patens cells with defective MMR would be affected or not. Survival was calculated as the ratio of protoplasts surviving after 15 days regeneration following treatment to the number of protoplasts undergoing normal regeneration without treatment.
Ppmsh2 mutants are equally sensitive to -radiations than wild type
Ionizing radiation can cause many different lesions in DNA, including strand breaks and base or sugar damage (23,24). Oxidized base damages induced by ionizing radiation may therefore be substrate for MMR (22). We compared the effect of -radiation on the survival of Ppmsh2 mutant and wild-type protoplasts. We found that both wild-type cells and Ppmsh2-deficient cells were similarly affected by -radiation (Figure 3A). For both genotypes, the LD50 (Lethal Dose 50%) was reached at 600 Gray. All -radiation dose comparisons were not significantly different.
Figure 3 Survival fraction after different genotoxic stresses. (A) -radiation, (B) UV-B, (C) Cisplatin, (D) MNU. Wild-type survival is represented with triangles and dashed line, and Ppmsh2 mutant survival is represented with squares and solid line. Error bars indicate SD based on at least two independent experiments in all cases.
Ppmsh2 mutants are more sensitive to UV-B radiations than wild type
A high UV-B flux can introduce a number of different lesions into the genome, mostly cyclobutane pyrimidine dimers (CPDs) and pyrimidine-6-4-pyrimidinone adducts (6-4PPs) (25). The elimination of these DNA photoproducts is essential for cell survival because both RNA- and DNA-polymerase are unable to read through both of these classes of dimers. Therefore, we tested the effect of increasing UV-B doses on the survival of regenerated wild-type and Ppmsh2 mutant protoplasts. We found that wild-type protoplasts were affected by high doses of UV-B, with an LD50 of 36 KJ/m2. In the Ppmsh2 background, the lethal dose was only 18 KJ/m2 (Figure 3B). All UV-B dose comparisons were significantly different. These results show that the Ppmsh2 mutant is more sensitive to UV-B than the wild type.
Ppmsh2 mutants are more tolerant to cisplatin than wild type
Cisplatin is a DNA-damaging drug that forms bifunctional covalent adducts with DNA and is a well-recognized mutagen that affects the survival of mammalian cells (26). We compared survival of wild-type and Ppmsh2 mutant regenerated protoplasts after exposure to increasing concentrations of cisplatin. The LD50 for the wild type was 250 μM cisplatin (Figure 3C), whereas the Ppmsh2 cells were more resistant than wild-type cells, with an LD50 of about 500 μM. All concentration comparisons were significantly different. These experiments show that the Ppmsh2 moss cells have a cisplatin tolerant phenotype.
Ppmsh2 mutants are more tolerant to MNU than wild type
The cytotoxicity of methylating agents such as MNU (N-methyl-N-nitrosourea) is primarily caused by their ability to introduce O6-methylguanine (O6-meGua) into the DNA (27). Persistent O6-meGua in the DNA is normally lethal. Therefore, we determined survival of wild-type and Ppmsh2mutant cells after exposure to increasing concentration of MNU. Wild-type survival was slightly affected by 1 mM MNU and dramatically affected by 5 mM MNU treatment. These values are comparable with those reported for MNU-induced DNA damage in Barley (28). The survival of Ppmsh2 mutant cells was slightly better than wild type at 1 mM MNU and significantly higher at 5 mM (Figure 3D). The LD50 was reached at 500 μM for both genotypes. These results show that Ppmsh2 moss cells have a weak resistance to MNU at high doses.
Gene targeting is decreased and homeologous recombination is not repressed in Ppmsh2 mutants
In E.coli (29), Saccharomyces cerevisiae (5), Trypanosoma brucei (30) and mouse embryogenic stem cells (7), depletion of MutS or MSH2 leads to an increased frequency of recombination between homeologous DNA sequences. We examined the involvement of PpMSH2 in HR by measuring GT efficiency with homologous or homeologous DNA sequences at the APT locus. Wild-type and Ppmsh2 cells were transformed with APT constructs presenting 0, 1, 2 and 3% divergence to the PpAPT genomic sequence, respectively (Figure 4). Hygromycin resistant transgenic clones were initially selected and tested for 2-FA resistance. GT frequency was calculated as the ratio of 2-FA resistant clones to the total number of Hygromycin resistant transformants.
We observed a significant decrease (1.8-fold) in GT frequency for mutant cells compared to wild type (P = 0.001) when the recombination substrate was fully homologous to its genomic target (Table 2). In wild type, GT frequency decreased as the percentage of divergence increased, whereas GT frequency remained stable for the Ppmsh2 mutant whatever the divergence level (Table 2). In wild type GT frequency significantly decreased 22-fold with 3% mismatches compared to 0% mismatches (2 test, P < 0.00001), whereas in Ppmsh2 mutant, the GT was as efficient with homeologous fragments as with homologous fragment (P = 0.40). Thus, GT efficiency with a homeologous fragments containing 3% divergence was increased 15-fold in Ppmsh2 mutant compared to the wild type. These data demonstrate that loss of function of MSH2 allows GT in P.patens with up to 3% divergent homeologous sequences, hence that the moss MSH2 protein has an antirecombination activity that prevents recombination between slightly divergent sequences.
Table 2 Gene targeting frequency in wild type and Ppmsh2 mutant
DISCUSSION
Ppmsh2 mutants show strong developmental defects, are sterile and have a mutator phenotype
Ppmsh2 mutant colonies develop strong developmental defects, including sterility, after a few weeks. This sterility may be due to the msh2 associated mutator phenotype increasing the rate of mutations that cause developmental problems in complex structures such as leafy shoots, antheridia and archegonia. Similar phenotypes have been observed in Arabidopsis thaliana msh2 mutant lines (10). After five generations, plants showed abnormalities in morphology and development, fertility, germination efficiency, seed/silique development and seed set. In human and mice, mutations in MSH2 result in a greatly increased likelihood of developing certain types of tumors (31). As in other eukaryotes, during growth and development, MSH2 in moss seems to be essential for the maintenance of genome integrity. Accordingly, we scored a 130-fold increase in the spontaneous mutation rate of the APT gene in Ppmsh2 mutants compared with wild type. This increased mutation rate is comparable with those observed in E.coli mutS mutant (20-fold increase) (32), or S.pombe (15-fold increase) (33) and S.cerevisiae (200-fold increase) (8) and other higher eukaryotes msh2 mutants (from 30- to several hundred-fold increase (7,10,34,35). This mutation rate can explain the phenotypes observed in the Ppmsh2 mutants. Assuming that the amount of mutations increases with the number of rounds of DNA replication, abnormalities should increase with the number of generations of vegetative propagation and this is what we observed after multiple successive subcultures of the colonies. Nevertheless, the phenotypes we observed in the mutant background are particularly strong, and two explanations can be proposed to account for this observation. The first one is linked to the fact that the major part of the moss life cycle is haploid. Thus, the deleterious effects of mutations in development are greater in moss than for diploid mice or A.thaliana, where the allelic locus would mask the phenotype resulting from randomly occurring mutations, at least in the first generation. Moreover, we cannot discard the hypothesis that, as in budding yeast, efficient GT in P.patens reflects the predominance of the HR pathway in the repair of DNA DSB. In this case, alteration of the MMR system, which participates in this DNA repair pathway, would lead to more drastic effects than in other higher eucaryotes, where HR is not the major pathway for DNA DSB repair. This is reminiscent of the strong reduction in spore viability observed in the msh2 yeast mutant (36) an organism where HR is also predominant.
Is PpMSH2 protein involved in cell-cycle checkpoints and apoptosis-induced by genotoxic stresses?
The sensitivity of cells to agents that create lesions on DNA has been attributed to attempts of MMR to correct the resulting damage (22). Depending on the type of DNA damage, loss of MMR may result in increased mutagenesis, loss of cell-cycle control and resistance to apoptosis (37). The resistance of msh2-deficient cells to DNA damage is not so clear because it depends on the organism and on the specificity of the lesions. In C.elegans, MSH2 confers sensitivity to ionizing radiation (9). In mouse cells, conflicting results have been reported, although the effects are generally minor. Msh2-nullizygous cells were found to exhibit slight higher levels of survival after exposure to ionizing radiation (38,39). However, it has recently been shown that both mouse embryogenic fibroblasts (MEFs) and mouse colorectal carcinoma cells defective in MSH2 were slightly more sensitive (1.5-fold) to X-ray irradiation than their MSH2-non-defective counterparts (40). In contrast, in moss, as in drosophila and human cells, the loss of MSH2 does not affect the sensitivity to -radiation, suggesting that MSH2 is not involved in the detection of lesions due to -radiation (41,42).
In E.coli, T.brucei, mouse and human cells, MSH2 confers sensitivity to MNU (27,38,43,44). In contrast, in the yeast S.cerevisiae, the MMR system did not affect sensitivity to methylating agents (45). In our experiments, Ppmsh2 mutants showed only a weak resistance to MNU at high concentration. This suggests that only a fraction of the MNU cytotoxicity in moss cells may be dependent on MSH2 function. Other mechanisms must be involved to repair DNA lesion in conjunction or not with MMR. HR can repair DNA lesions, like it has been shown in yeast for O6-meGua damages (45). Absence of MMR in S.cerevisiae did not affect sensitivity to methylating agents, while HR inactivation sensitized cells to this agent. In HR-deficient cells, defects in the MSH2 gene rescue the original sensitivity (45). These authors conclude that methylation damage processing and repair by HR masks the sensitivity of MMR-proficient S.cerevisiae cells toward methylating agents (45). In our experiments, Ppmsh2 mutants showed a weak resistance to MNU only at high concentration, suggesting that only a fraction of the MNU cytotoxicity in moss cells may be dependent on MSH2 function. Like in yeast, moss MMR could be overwhelmed by other mechanisms such as HR, thus preventing the cytotoxicity due to methylating agents to be detectable in wild type.
Peters et al. (37) have shown that in MEF (mouse embryogenic fibroblasts), MSH2 may sense damage due to photoproducts or may contribute to the processing of UV-B-induced DNA damage, thereby initiating apoptosis and influencing the activation of p53 (a tumor suppressor which functions as a transcription factor involved in cell-cycle checkpoints and apoptosis). They showed that Msh2-null MEFs were partially resistant to the cytotoxic effects of UV-B, and that this resistance led to a reduction in apoptosis. In contrast, we have shown that the Ppmsh2 mutant is more sensitive to UV-B radiation than the wild type. Therefore, MSH2 in moss may play a more important role in the recognition of UV-B lesions than that reported in MEF.
Experiments in E.coli have shown that loss of MMR activity, due to mutations in mutS, was correlated with cisplatin resistance (46,47). There are conflicting results in the response of cisplatin treatment in higher eukaryotes. It was thought that mouse MSH2 conferred sensitivity to cisplatin (48). However, Claij and te Riele (49) have recently shown that msh2-deficient mouse cells do not respond specifically to cisplatin lesions compared with wild type (49). In contrast, data from Trypanosoma cruzei have demonstrated a role of MSH2 in sensitivity to cisplatin (50), like in prokaryotes and in yeast (46,51). Our results in moss show that Ppmsh2 mutants are tolerant to cisplatin.
The fact that Ppmsh2 mutants have opposite responses to lesions due to UV-B (sensitivity) and cisplatin (resistance) suggests that moss may have different pathways for managing lesions involving PpMSH2 proteins. PpMSH2 may signal DNA lesions and induce cell-cycle arrest followed by either DNA repair or apoptosis to prevent accumulation of a large number of mutations. In the absence of PpMSH2, we suggest that, depending on the type of DNA lesions, different responses may be engaged. When the lesion is too severe (exposure to UV-B), the cells are unable to repair them and die. For other lesions, they can be bypassed by a translesion synthesis polymerase (52,53) or be repaired by other mechanisms such as base excision repair, nucleotide excision repair or by HR (45,54,55). In these cases, the cells can survive almost normally, as observed with moss msh2 mutant cells exposed to cisplatin. We suggest that in wild-type cells, the DNA lesions caused by cisplatin trigger cell death and/or prolonged cell-cycle arrest through PpMSH2. We have evidence that the cell cycle is perturbed in the absence of PpMSH2: during the regeneration of protoplasts, Ppmsh2 mutants showed a slightly higher rate of cell division compared with the wild type. This suggests that Ppmsh2 mutants may have an accelerated cell cycle, which may be explained if there was shorter arrest at cell-cycle checkpoints. Indeed, in mice it has been shown that after DNA damage, the absence of MSH2 results in premature release from the early G2/M arrest and an inability to activate properly the checkpoint kinase proteins (40). It was suggested that an active MSH2 is required for a correct response to DNA damage in the G2 phase of the cell cycle, possibly connecting DSB repair to checkpoint signaling. Moreover, in human, MMR is required for S-phase and G2-M checkpoint regulation (56,57). Consistent with these findings, we suggest that PpMSH2 is required for efficient cell-cycle arrest induced by DNA damage and may be a key component acting on DNA damage-activated checkpoints.
PpMSH2 is involved in HR and is essential to homeologous recombination
We examined the influence of MMR on HR in P.patens by assaying the frequency of recombination between a foreign DNA fragment released in vivo and a genomic target sequence. We found that the frequency of GT via HR between identical DNA was decreased 1.8-fold in the Ppmsh2-defective cells compared with the wild type. In mice, msh2 mutants show a slight decrease of GT efficiency with homologous fragment compared with wild-type mice, and this decrease is comparable with our results (1.5-fold) (7). In yeast, a recent study has shown that the frequency of GT was reduced 3- to 4-fold in the msh2 strain (58). Together with our results, these suggest that MSH2 in P.patens is probably involved in mechanisms that regulate GT of perfectly matching sequences in the genome. Further analyses are needed to determine the mechanisms involved.
We tested the influence of MMR on homeologous recombination in P.patens by assaying the frequency of recombination between a mutated DNA fragment and a genomic target sequence. Our results show that DNA sequence identity is important to determine the efficiency of gene replacement in the wild type. We have also shown that lack of PpMSH2 allows homeologous sequences with up to 3% divergence to be as efficiently targeted as perfectly identical sequences to a genomic locus. This shows that PpMSH2 plays a major role to prevent recombination between homeologous sequences. This is consistent with the similar role described for MMR in E.coli (29), in Trypanosoma (30), in yeast (5) and in mouse cells (7,59,60), suggesting that this role of MSH2 is strongly preserved during evolution. However, this is the first time that MSH2 involvement in HR and homeologous recombination is shown in plants.
In conclusion, we have found that in moss, MSH2 is involved in mismatch recognition and repair that most likely results from DNA replication errors, and in the response to certain genotoxic agents. Moreover, MSH2 plays a role in HR and in the recognition of mismatches that occur during heteroduplex formation between two not perfectly complementary DNA strands. These mechanisms are essential for maintaining genome integrity in all organisms and to establish interspecific barrier. This is particularly true for plants as they are unable to move and must cope with exposure to environmental mutagens. Thus, an efficient mechanism for genomic stability is needed in plants, and even more in mosses, where the major developmental stage of the life cycle is the haploid gametophyte. This is emphasized by the dramatic deleterious effect on development observed in an msh2 P.patens mutant compared with its A.thaliana counterpart. Our results show for the first time, the crucial role of MMR in a pluricellular organism where HR seems to be one of the major pathways to repair DNA lesions.
ACKNOWLEDGEMENTS
The authors thank Mr Favodon, (Institut Curie, Orsay, France), for help with the 137Cs irradiator for gamma radiations assays. The authors thank Dr Michel Laloue (INRA Versailles, France) for providing the 2-Fluoroadenine. The authors thank Marie-Pascale Doutriaux, Mathilde Grelon, Christine Mézard and Rapha?l Mercier for suggestions and discussions on the manuscript. This work was supported by grant from the Institut National de la Recherche Agronomique and Cellectis SA (Romainville, France). Funding to pay the Open Access publication charges for this article was provided by Institut National de la Recherche Agronomique.
REFERENCES
Modrich, P. and Lahue, R. (1996) Mismatch repair in replication fidelity, genetic recombination, and cancer biology Annu. Rev. Biochem, . 65, 101–133 .
Harfe, B.D. and Jinks-Robertson, S. (2000) DNA mismatch repair and genetic instability Annu. Rev. Genet, . 34, 359–399 .
Kolodner, R. (1996) Biochemistry and genetics of eukaryotic mismatch repair Genes Dev, . 10, 1433–1442 .
Schofield, M.J. and Hsieh, P. (2003) DNA mismatch repair: molecular mechanisms and biological function Annu. Rev. Microbiol, . 57, 579–608 .
Negritto, M.T., Wu, X., Kuo, T., Chu, S., Bailis, A.M. (1997) Influence of DNA sequence identity on efficiency of targeted gene replacement Mol. Cell. Biol, . 17, 278–286 .
Sia, E.A., Kokoska, R.J., Dominska, M., Greenwell, P., Petes, T.D. (1997) Microsatellite instability in yeast: dependence on repeat unit size and DNA mismatch repair genes Mol. Cell. Biol, . 17, 2851–2858 .
de Wind, N., Dekker, M., Berns, A., Radman, M., te Riele, H. (1995) Inactivation of the mouse Msh2 gene results in mismatch repair deficiency, methylation tolerance, hyperrecombination, and predisposition to cancer Cell, 82, 321–330 .
Greene, C.N. and Jinks-Robertson, S. (2001) Spontaneous frameshift mutations in Saccharomyces cerevisiae: accumulation during DNA replication and removal by proofreading and mismatch repair activities Genetics, 159, 65–75 .
Degtyareva, N.P., Greenwell, P., Hofmann, E.R., Hengartner, M.O., Zhang, L., Culotti, J.G., Petes, T.D. (2002) Caenorhabditis elegans DNA mismatch repair gene msh-2 is required for microsatellite stability and maintenance of genome integrity Proc. Natl Acad. Sci. USA, 99, 2158–2163 .
Hoffman, P.D., Leonard, J.M., Lindberg, G.E., Bollmann, S.R., Hays, J.B. (2004) Rapid accumulation of mutations during seed-to-seed propagation of mismatch-repair-defective Arabidopsis Genes Dev, . 18, 2676–2685 .
Schaefer, D. (2002) A new moss genetics: targeted mutagenesis in Physcomitrella patens Annu. Rev. Plant Biol, . 53, 477–501 .
Brun, F., Gonneau, M., Doutriaux, M.P., Laloue, M., Nogué, F. (2001) Cloning of the PpMSH-2 cDNA of Physcomitrella patens, a moss in which gene targeting by homologous recombination occurs at high frequency Biochimie, 83, 1003–1008 .
Ashton, N.W. and Cove, D.J. (1977) The isolation and preliminary characterisation of auxotrophic and analogue resistant mutants of the moss, Physcomitrella patens Mol. Gen. Genet, . 87–95 .
Asthon, N.W., Grimsley, N., Cove, D.J. (1979) Analysis of gametopytic development in the moss, Physcomitrella patens, using auxin and cytokinn resistants mutants Planta, 427–435 .
Sambrook, J., Frisch, E.F., Maniatis, T. Molecular Cloning: A Laboratory manual, (1989) NY Cold Spring Harbor Laboratory Press .
Cadwell, R.C. and Joyce, G.F. (1994) Mutagenic PCR PCR Methods Appl, . 3, S136–S140 .
Bilang, R., Iida, S., Peterhans, A., Potrykus, I., Paszkowski, J. (1991) The 3' terminal region of the hygromycin-B-resistance gene is important for its activity in Escherichia coli and Nicotiana tabacum Gene, 100, 247–250 .
Schaefer, D.G. and Zr?d, J.P. (1997) Efficient gene targeting in the moss Physcomitrella patens Plant J, . 11, 1195–1206 .
Sauer, B. (1993) Manipulation of transgenes by site-specific recombination: use of Cre recombinase Methods Enzymol, . 225, 890–900 .
Albert, H., Dale, E.C., Lee, E., Ow, D.W. (1995) Site-specific integration of DNA into wild-type and mutant lox sites placed in the plant genome Plant J, . 7, 649–659 .
Schaefer, D.G. and Zryd, J.P. (2001) The moss Physcomitrella patens, now and then Plant Physiol, . 127, 1430–1438 .
Fedier, A. and Fink, D. (2004) Mutations in DNA mismatch repair genes: implications for DNA damage signaling and drug sensitivity (review) Int. J. Oncol, . 24, 1039–1047 .
Hutchinson, F. (1985) Chemical changes induced in DNA by ionizing radiation Prog. Nucleic Acid Res. Mol. Biol, . 32, 115–154 .
Ward, J.F. (1988) DNA damage produced by ionizing radiation in mammalian cells: identities, mechanisms of formation, and reparability Prog. Nucleic Acid Res. Mol. Biol, . 35, 95–125 .
Jasen, M.A.K., Gaba, V., Greengerg, B.M. (1998) Higher plants and UV-B radiation: balancing damage, repair and acclimation Trends Plant Sci, . 3, 131–135 .
Aebi, S., Kurdi-Haidar, B., Gordon, R., Cenni, B., Zheng, H., Fink, D., Christen, R.D., Boland, C.R., Koi, M., Fishel, R., et al. (1996) Loss of DNA mismatch repair in acquired resistance to cisplatin Cancer Res, . 56, 3087–3090 .
Humbert, O., Fiumicino, S., Aquilina, G., Branch, P., Oda, S., Zijno, A., Karran, P., Bignami, M. (1999) Mismatch repair and differential sensitivity of mouse and human cells to methylating agents Carcinogenesis, 20, 205–214 .
Jovtchev, G., Menke, M., Schubert, I. (2001) The comet assay detects adaptation to MNU-induced DNA damage in barley Mutation Res, . 493, 95–100 .
Rayssiguier, C., Thaler, D.S., Radman, M. (1989) The barrier to recombination between Escherichia coli and Salmonella typhimurium is disrupted in mismatch-repair mutants Nature, 342, 396–401 .
Bell, J.S. and McCulloch, R. (2003) Mismatch repair regulates homologous recombination, but has little influence on antigenic variation, in Trypanosoma brucei J. Biol. Chem, . 278, 45182–45188 .
Buermeyer, A.B., Deschenes, S.M., Baker, S.M., Liskay, R.M. (1999) Mammalian DNA mismatch repair Annu. Rev. Genet, . 33, 533–564 .
Schaaper, R.M. and Dunn, R.L. (1987) Spectra of spontaneous mutations in Escherichia coli strains defective in mismatch correction: the nature of in vivo DNA replication errors Proc. Natl Acad. Sci. USA, 84, 6220–6224 .
Rudolph, C., Kunz, C., Parisi, S., Lehmann, E., Hartsuiker, E., Fartmann, B., Kramer, W., Kohli, J., Fleck, O. (1999) The msh2 gene of Schizosaccharomyces pombe is involved in mismatch repair, mating-type switching, and meiotic chromosome organization Mol. Cell. Biol, . 19, 241–250 .
Denver, D.R., Feinberg, S., Estes, S., Thomas, W.K., Lynch, M. (2005) Mutation rates, spectra, and hotspots in mismatch repair-deficient Caenorhabditis elegans Genetics, .
Malkhosyan, S., McCarty, A., Sawai, H., Perucho, M. (1996) Differences in the spectrum of spontaneous mutations in the hprt gene between tumor cells of the microsatellite mutator phenotype Mutat Res, . 316, 249–259 .
Reenan, R.A.G. and Kolodner, R.D. (1992) Characterization of insertion mutations in the Saccharomyces cerevisiae Msh1 and Msh2 genes—evidence for separate mitochondrial and nuclear functions Genetics, 132, 975–985 .
Peters, A.C., Young, L.C., Maeda, T., Tron, V.A., Andrew, S.E. (2003) Mammalian DNA mismatch repair protects cells from UVB-induced DNA damage by facilitating apoptosis and p53 activation DNA Repair (Amst.), 2, 427–435 .
Fritzell, J.A., Narayanan, L., Baker, S.M., Bronner, C.E., Andrew, S.E., Prolla, T.A., Bradley, A., Jirik, F.R., Liskay, R.M., Glazer, P.M. (1997) Role of DNA mismatch repair in the cytotoxicity of ionizing radiation Cancer Res, . 57, 5143–5147 .
DeWeese, T.L., Shipman, J.M., Larrier, N.A., Buckley, N.M., Kidd, L.R., Groopman, J.D., Cutler, R.G., te Riele, H., Nelson, W.G. (1998) 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. USA, 95, 11915–11920 .
Franchitto, A., Pichierri, P., Piergentili, R., Crescenzi, M., Bignami, M., Palitti, F. (2003) The mammalian mismatch repair protein MSH2 is required for correct MRE11 and RAD51 relocalization and for efficient cell cycle arrest induced by ionizing radiation in G2 phase Oncogene, 22, 2110–2120 .
Aquilina, G., Crescenzi, M., Bignami, M. (1999) Mismatch repair, G(2)/M cell cycle arrest and lethality after DNA damage Carcinogenesis, 20, 2317–2326 .
Flores, C. and Engels, W. (1999) Microsatellite instability in Drosophila spellchecker1 (MutS homolog) mutants Proc. Natl Acad. Sci. USA, 96, 2964–2969 .
Karran, P. and Marinus, M.G. (1982) Mismatch correction at O6-methylguanine residues in E.coli DNA Nature, 296, 868–869 .
Bell, J.S., Harvey, T.I., Sims, A.M., McCulloch, R. (2004) Characterization of components of the mismatch repair machinery in Trypanosoma brucei Mol. Microbiol, . 51, 159–173 .
Cejka, P., Mojas, N., Gillet, L., Schar, P., Jiricny, J. (2005) Homologous recombination rescues mismatch-repair-dependent cytotoxicity of S(N)1-type methylating agents in S.cerevisiae Curr. Biol, . 15, 1395–1400 .
Calmann, M.A., Nowosielska, A., Marinus, M.G. (2005) Separation of mutation avoidance and antirecombination functions in an Escherichia coli mutS mutant Nucleic Acids Res, . 33, 1193–1200 .
Fram, R.J., Cusick, P.S., Wilson, J.M., Marinus, M.G. (1985) Mismatch repair of cis-diamminedichloroplatinum(II)-induced DNA damage Mol. Pharmacol, . 28, 51–55 .
Fink, D., Zheng, H., Nebel, S., Norris, P.S., Aebi, S., Lin, T.P., Nehme, A., Christen, R.D., Haas, M., MacLeod, C.L., et al. (1997) In vitro and in vivo resistance to cisplatin in cells that have lost DNA mismatch repair Cancer Res, . 57, 1841–1845 .
Claij, N. and te Riele, H. (2004) Msh2 deficiency does not contribute to cisplatin resistance in mouse embryonic stem cells Oncogene, 23, 260–266 .
Augusto-Pinto, L., Teixeira, S.M., Pena, S.D., Machado, C.R. (2003) Single-nucleotide polymorphisms of the Trypanosoma cruzi MSH2 gene support the existence of three phylogenetic lineages presenting differences in mismatch-repair efficiency Genetics, 164, 117–126 .
Drotschmann, K., Topping, R.P., Clodfelter, J.E., Salsbury, F.R. (2004) Mutations in the nucleotide-binding domain of MutS homologs uncouple cell death from cell survival DNA Repair (Amst.), 3, 729–742 .
Lehmann, A.R. (2005) Replication of damaged DNA by translesion synthesis in human cells FEBS Lett, . 579, 873–876 .
Hoffmann, J.S., Locker, D., Villani, G., Leng, M. (1997) HMG1 protein inhibits the translesion synthesis of the major DNA cisplatin adduct by cell extracts J. Mol. Biol, . 270, 539–543 .
Lindahl, T. and Wood, R.D. (1999) Quality control by DNA repair Science, 286, 1897–1905 .
Liu, Z., Hossain, G.S., Islas-Osuna, M.A., Mitchell, D.L., Mount, D.W. (2000) Repair of UV damage in plants by nucleotide excision repair: Arabidopsis UVH1 DNA repair gene is a homolog of Saccharomyces cerevisiae Rad1 Plant J, . 21, 519–528 .
Brown, K.D., Rathi, A., Kamath, R., Beardsley, D.I., Zhan, Q., Mannino, J.L., Baskaran, R. (2003) The mismatch repair system is required for S-phase checkpoint activation Nature Genet, . 33, 80–84 .
Yan, T., Schupp, J.E., Hwang, H.S., Wagner, M.W., Berry, S.E., Strickfaden, S., Veigl, M.L., Sedwick, W.D., Boothman, D.A., Kinsella, T.J. (2001) Loss of DNA mismatch repair imparts defective cdc2 signaling and G(2) arrest responses without altering survival after ionizing radiation Cancer Res, . 61, 8290–8297 .
Langston, L.D. and Symington, L.S. (2005) Opposing roles for DNA structure-specific proteins Rad1, Msh2, Msh3, and Sgs1 in yeast gene targeting EMBO J, . 24, 2214–2223 .
Elliott, B. and Jasin, M. (2001) Repair of double-strand breaks by homologous recombination in mismatch repair-defective mammalian cells Mol. Cell. Biol, . 21, 2671–2682 .
Abuin, A., Zhang, H., Bradley, A. (2000) Genetic analysis of mouse embryonic stem cells bearing Msh3 and Msh2 single and compound mutations Mol. Cell. Biol, . 20, 149–157 .(Bénédicte Trouiller, Didier G. Schaefer1)
*To whom correspondence should be addressed. Tel: +33 1 30833009; Fax: +33 1 30833319; Email: nogue@versailles.inra.fr
ABSTRACT
MSH2 is a central component of the mismatch repair pathway that targets mismatches arising during DNA replication, homologous recombination (HR) and in response to genotoxic stresses. Here, we describe the function of MSH2 in the moss Physcomitrella patens, as deciphered by the analysis of loss of function mutants. Ppmsh2 mutants display pleiotropic growth and developmental defects, which reflect genomic instability. Based on loss of function of the APT gene, we estimated this mutator phenotype to be at least 130 times higher in the mutants than in wild type. We also found that MSH2 is involved in some but not all the moss responses to genotoxic stresses we tested. Indeed, the Ppmsh2 mutants were more tolerant to cisplatin and show higher sensitivity to UV-B radiations. PpMSH2 gene involvement in HR was studied by assessing gene targeting (GT) efficiency with homologous and homeologous sequences. GT efficiency with homologous sequences was slightly decreased in the Ppmsh2 mutant compared with wild type. Strikingly GT efficiency with homeologous sequences decreased proportionally to sequence divergence in the wild type whereas it remained unaffected in the mutants. Those results demonstrate the role of PpMSH2 in the maintenance of genome integrity and in homologous and homeologous recombination.
INTRODUCTION
The mismatch repair (MMR) system is evolutionarily highly conserved and plays an essential role in maintaining genome stability (1). In all organisms, MMR is best known for its role in the post-replication repair of DNA polymerization errors. It keeps the rate of mutations due to nucleotide misincorporation and polymerase slippage at an acceptable low level (2). MMR proteins also recognize mismatches in heteroduplex recombination intermediates. In somatic cells, they display an antirecombination activity that inhibits recombination between homeologous sequences present in a single genome (such as allelic genes in diploid cells and members of multigene families) and thus preserves genome integrity (2). This activity also prevents genetic recombination between different but related species and thus contributes to the definition of species. Additionally, MMR plays a role in some types of nucleotide excision repair, which are responsible for repair of physical and chemical damage to DNA. It also participates in a cell-cycle checkpoint control system by recognizing certain types of DNA damage and promoting cell-cycle arrest or triggering apoptosis pathways (3,4).
The methyl-directed MMR system of Escherichia coli is composed of MutS and MutL proteins (1,3). The MutS proteins are conserved from bacteria to mammals. In eukaryotes MutS homologs (MSH) are encoded by small multigene families, whose members can associate into heterodimers that have discrete roles in MMR-related processes (2). For example, MSH2/MSH6 heterodimers recognize and stimulate the repair of single base-pair mismatches while MSH2/MSH3 heterodimers recognize small insertion/deletion loops (2). The MSH2 protein is a central component of the eukaryotic MMR system and is present in all heterodimers. It is crucial for the repair of all mismatched lesions, whereas other MSH proteins modulate the function of MSH2 depending on the different lesion types or developmental stages.
The msh2 mutants in yeast and mammals show (i) microsatellite instability, (ii) a mutator phenotype characterized by a high spontaneous mutation rate, (iii) an increased recombination frequency between diverged DNA sequences (homeologous) and (iv) tolerance to certain types of chemical and physical treatments that damage DNA (5–8). In Caenorhabditis elegans, loss of MSH2 is associated with microsatellite instability, a mutator phenotype and reduced fertility (9).
However, in plants, the effect of MSH2 mutations is still poorly understood. In Arabidopsis thaliana, Hoffman et al. (10) have shown rapid accumulation of mutations and microsatellite instability during seed-to-seed propagation of Atmsh2 defective lines. However, the molecular tools are not available to evaluate the role of MSH2 in the mechanism of homologous recombination (HR) in A.thaliana and other higher plants. This type of study in plants is currently only possible in the moss Physcomitrella patens.
The moss P.patens is unique in the plant kingdom in that it allows high efficiency of gene targeting (GT) via HR (11). Thus, it is possible to use gene knock-out and allele replacement methods in this organism. This allows to examine for the first time the importance of MMR in a multicellular organism in which targeted integration by HR is a major transformation pathway . Recently, we have isolated the PpMSH2 gene (12). We now have generated Ppmsh2 disruptants to examine the role of MSH2 in the DNA metabolism of P.patens. Here, we show that PpMSH2 is involved in the control of genome integrity and that it is a barrier to recombination between divergent DNA sequences. Furthermore, in the absence of PpMSH2, moss development is dramatically affected most likely as a consequence of a strong mutator phenotype that is directly revealed by the dominance of the haplophase during its development.
MATERIALS AND METHODS
Plant material and culture conditions
The Gransden wild-type strain of P.patens (13) was used in this study. Protonemal tissue was propagated on PpNO3 medium (14), supplemented with 2.7 mM NH4-tartrate (standard medium). Cultures were grown in 9 cm Petri dishes on medium solidified with 0.7% Agar (Biomar) and overlaid with a cellophane disk (Cannings, Bristol). Cultures were illuminated with a light regime of 16 h light/8 h darkness and a quantum irradiance of 80 μE m–2 s–1 (standard conditions). Sporogenesis was performed in Magenta box in which the tested strains were grown aside the self sterile but cross fertile strain nicB5ylo6 (13) on minimal PpNO3 medium. Crosses were grown in standard conditions for 6–8 weeks (i.e. until about 50 gametophores reached full development), then irrigated with sterile water and transferred for 2 weeks in growth chambers set at 15°C with 10 h of light per day and a quantum irradiance of 15 μE m–2 s–1. The development of archegonia, antheridia and spore capsules was followed visually during the next month. Antheridia and archegonia were manually dissected for further microscopic observation.
Molecular cloning
We used standard methods for all molecular cloning (15). Using genomic P.patens DNA as starting template, we amplified a 2340 bp PCR fragment covering an internal fragment of the MSH2 genomic region (Gene Bank: DQ117988) and inserted it into the TA-cloning vector pCR?II (Invitrogen, Groningen, The Netherlands) to produce PpMSH2 (oligonucleotide sequences can be provided on request). To obtain PpMSH2-KO, a 465 bp 5' targeting fragment (coordinate 2557–3022 in DQ117988 ) and a 598 bp 3' targeting fragment (coordinate 4614–5212 in DQ117988 ) were recovered by restriction from PpMSH2 and inserted upstream (5') and downstream (3') of the loxP sites flanking the resistance cassette of pBSK35S-NPTII-lox (supplied by Dr Laloue, INRA Versailles). Vector pBSK35S-NPTII-lox carries a neomycin phosphotransferase resistance gene driven by the 35S CaMV promoter and flanked by two LoxP sites in direct orientation. The antibiotic resistance cassette is in the same orientation with regard to the transcription of PpMSH2.
To obtain PpAPT-KO, a 3995 bp Xba1 and EcoRI fragment covering the APT genomic sequence (Gene Bank: DQ117987), was inserted in pBlueScript KS (Stratagene), forming PpAPT1. The ATG codon is 1217 bp 3' of the XbaI site and the stop codon is 881 bp 5' of the EcoRI site in this PpAPT fragment. An internal 1102 bp BamHI fragment was deleted by digestion of the PpAPT1 vector, to produce PpAPT. We generated homeologous variants of this targeting cassette by mutating the PpAPT sequence by mutagenic PCR using the Cadwell and Joyce (16) protocol. Successive APT variants were generated until we reached 3% divergence within the PpAPT original fragment. A BamHI fragment containing the hygromycin phosphotransferase gene (HptII) driven by the 35S CaMV promoter was recovered from plasmid pGL2 (17) and ligated into the BamHI site of PpAPT and of the variants to produce the homologous and homeologous PpAPT-KO vectors (Figure 4).
Figure 4 Gene targeting PpAPT fragments used for homologous and homeologous recombination studies. The hygromycin gene resistance cassette is represented between both PpAPT fragments by a horizontal interrupted line. (A) PpAPT fragments in the PpAPT-KO cassette in which the sequences are exactly identical to the locus PpAPT used as reporter gene. (B) Alignments of PpAPT fragments in homeologous recombination cassettes; each vertical line indicates mismatch between mutated PpAPT fragments and the original locus.
Protoplasts isolation and regeneration
Protoplasts were isolated from 6-day-old protonema by incubation for 40 min in 1% Driselase (Fluka 44585) dissolved in 0.47 M mannitol. The suspension was filtered successively through 80 and 40 μm stainless steel sieves. Protoplasts were sedimented by low-speed centrifugation (600 g for 5 min at 20°C) and washed twice with 0.47 M mannitol. Protoplasts were then resuspended at 1.2 x 106 protoplasts/ml in MMM solution (0.47 M mannitol, 15 mM MgCl2 and 0.1% MES, pH 5.6) for transformation. For growth assay and phenotypic analyses, protoplasts were resuspended at a density of ca. 1–3 x 104 protoplasts/ml in mannitol 0.47 M containing 0.5% agar. Two ml of suspension were spread on cellophane disks covering standard medium supplemented with 0.5% glucose and 0.36 M mannitol (PPM). Regeneration was conducted in standard growth conditions. After 1 week of regeneration, protoplasts were transferred to standard medium without mannitol.
Transformation of protoplasts
Moss protoplasts (1.2 x 106) were transformed as described previously (18) with PpMSH2-KO digested with AgeI and AatII, which generate DSBs within the genomic MSH2 sequence. Primary transformants were selected in the presence of 50 mg/l paramomycin (Duchefa P0141, Haarlem, The Netherlands). Small pieces of the protonema tissue (2 mm in diameter) of primary P.patens transformants were then transferred onto standard medium without selection and cultured for 15 days. A second round of selection allowed the isolation of stable transformants, by transferring small pieces of protonema tissue from the previous colonies onto standard medium containing 50 mg/l G418 (Duchefa G0175, Haarlem, The Netherlands) (paramomycin and G418 belong to the kanamycin family of antibiotics).
Molecular analysis of Msh2 knockouts
Moss DNA was extracted from 200 mg of fresh protonema tissue ground in 220 μl of extraction buffer (220 mM Tris–HCl, pH 7.5, 250 mM NaCl, 25 mM EDTA and 0.5% SDS) with an electric pestle in the presence of sand. After centrifugation, DNA was precipitated with isopropanol at room temperature. The pellet was dried and resuspended with 100 μl sterile water. Stable disruptants of the PpMSH2 gene were screened by PCR for the incorporation of the resistance cassette into the MSH2 locus. We used primers specific for the resistance cassette (Uc: TATTTTTGGAGTAGACAAGCGTGTCGT and Lc: CCGCTTCCTCGTGCTTTACGGTAT) and primers specific for the PpMSH2 sequence outside the transformation construct (PpMSH2#4: CAATGTCACTGAGAGCAAAACG and PpMSH2#5: CACACTGATCTCTGCTCGGTCG).
Moss RNA was isolated using the RNAeasy kit (Qiagen) starting with 100 mg of fresh protonema. We obtained doubled-stranded cDNAs corresponding to mRNAs expressed in 6-day-old protonema cultures using the SMART-PCR cDNA Synthesis Kit (Clontech). The primers PpMSH2#4 and PpMSH2#2 (CACCAACACGCGCAAAGATG) were used for PCR on these cDNAs.
Generation of clean deletion by transient Cre recombinase expression
Direct repeats of LoxP recombination sites from the Cre/lox system (19) were located on either side of the 35S-nptII resistance cassette in PpMSH2-KO. Transient expression of a constitutive Cre expression cassette (20) in protoplasts allowed excision of the resistance gene from the PpMSH2 recombinant locus (D.G. Schaefer, A. Finka and J.P. Zr?d, manuscript in preparation). Protoplasts were regenerated at a low density on non-selective medium for 2 weeks and fragments of single protoplast-derived colonies were then replicated on selective (i.e. G418 50 mg/l) and non-selective medium. Loss of antibiotic resistance was determined after an additional week of growth. Replicates growing on non-selective medium were used for subsequent amplification and molecular analyses. Clone Ppmsh2 was selected this way and used for further analyses. We analyzed by PCR the recombinant locus with the primers PpMSH2#4 and PpMSH2#2.
Cytology
Binocular observations were made with a Nikon SMZ1000, and microscopic observations with a Leitz (type 090-122.012) inverted microscope.
Evaluation of spontaneous mutation frequency
Mutations in the APT gene confer resistance to 2-FA (2-Fluoroadenine, from Dr Laloue, INRA Versailles, France), a toxic compound for cells. The number of 2-FA resistant colonies reflects the frequency of spontaneous mutations. Protoplasts of wild-type and Ppmsh2 mutants were regenerated for 6 days regeneration on PPM (105/Petri dish) and then transferred on standard solid medium supplemented with 10 μM 2-FA. After 2 weeks, the number of resistant colonies was counted. We repeated this experiment three times. The APT genomic sequence from 2-FA resistant clones was subsequently PCR amplified and sequenced to identify the mutation responsible for the resistance (oligonucleotide sequences can be provided on request).
Sensitivity to cisplatin/MNU and /UV-B radiations
Cisplatin (cis-diamminedichloroplatinum(II)) (Calbiochem, Darnstadt, Germany), MNU (N-methyl-N-nitrosourea) and O6-bzGua (O6-benzylguanine) (Sigma-Aldrich, Steinheim, Germany) were dissolved in dimethyl sulfoxide at 100 mM each for storage at –20°C and at 25 mM for O6-bzGua immediately before use.
Wild-type and Ppmsh2 mutant protoplasts in liquid PPM were treated for 20 h with 0–5 mM cisplatin or MNU (2 x 105 protoplasts at each concentration). From 1 h before exposure to MNU and throughout the whole procedure, 25 μM O6-bzGua was present in the medium to inhibit the removal of methyl groups from the O6 position of guanine by endogenous methyl-transferase activity. After 20 h in the dark, protoplasts were washed and spread on solid medium at low concentrations.
Protoplasts in liquid mannitol medium were exposed to ionizing radiation (1 Gy/s) using a 137Cs irradiator (IBL-637 CIS-Bio-International, Institut Curie, Orsay). After 20 h in the dark, the protoplasts were spread on solid medium at low concentrations.
Protoplasts spread on solid medium (5 x 104 protoplasts at each concentration in four Petri dishes) were exposed to UV-B light (60 J/m2/s) from a 312 nm TFX lamp. We calculated the flux with a UV-Elektronik GmbH dosimeter. The protoplasts were then left in the dark for 20 h.
After 6 days regeneration, protoplasts were transferred onto standard solid medium. After 1 week, the number of survivors was counted. We repeated these experiments two to three times. Statistical analyses have been made with the 2 test.
APT targeting frequency with homologous and homeologous DNA
Targeting APT fragments were produced by digestion of PpAPT-KO vector with NotI and EcoRI enzymes. We selected stable transformants after transformation of wild-type and Ppmsh2 mutant protoplasts with homologous and homeologous APT disruption fragments. Small pieces of protonema tissue (2 mm in diameter) from stable transformants were then transferred onto standard medium containing 10 μM of 2-Fluoroadenine to detect APT gene targeting events. Experiments were repeated three times.
RESULTS
Targeted disruption of the MSH2 gene in P.patens
Moss protoplasts were transformed with vector PpMSH2-KO and four independent clones were selected for further investigation (Ppmsh2#1 to Ppmsh2#4). In each clone, disruption of the chromosomal PpMSH2 locus was controlled by PCR and absence of the PpMSH2 transcript was established by RT–PCR (see Ppmsh2#1 for example Figure 1B–D). In order to obtain a simple deletion mutant, the resistance cassette inserted at the PpMSH2 locus of the Ppmsh2#1 clone was eliminated by the CRE recombinase to produce clone Ppmsh2 (see Materials and Methods and Figure 1A). In this clone, we confirmed deletion of exon 7 of the PpMSH2 gene by PCR and checked modification of the transcript by RT–PCR (Figure 1C and D). The four independent PpMSH2 knockouts (Ppmsh2#1 to Ppmsh2#4) and the deletion mutant (Ppmsh2) behaved identically under all circumstances tested except in aged cultures. Therefore, Ppmsh2 mutant was used for further characterization except for late developmental phenotypes (see below). To avoid the genetic drift resulting to the expected mutator phenotype associated with msh2 mutations, all the experiments were performed using the primary stock of protonema suspension from Ppmsh2 stored at 4°C as starting material.
Figure 1 Molecular analysis of Ppmsh2 mutants. (A) Schematic representation of PpMSH2 genomic locus, schematic representation of PpMSH2 disrupted locus with a double crossover insertion and schematic representation of PpMSH2 disrupted locus without resistance gene. Black open boxes represent exons, black lines represent introns, gray open boxes are homology regions used to disrupt the gene, black boxes are Lox sites for Cre recombination and arrows show the primers used for PCR. Ppmsh2? is mutant without the resistance gene, resulting from Ppmsh2#1 after Cre recombinase expression. (B) PCR of the recombinant borders with a primer external to the targeting sequences and one on the resistance gene . (C) PCR with primers flanking recombination sites (primers: PpMSH2#4/2), (D) RT–PCR for PpMSH2 transcripts (primers: PpMSH2#4/2), PpAPT was used as control.
Ppmsh2 mutants display pleiotropic developmental defects
The haploid gametophytic development of P.patens from regenerating protoplasts starts with a filamentous branched structure called the protonema that grows by apical cell divisions and lateral division of subapical cells. There are two cell types in protonema filaments: chloronema and caulonema. Regenerating protoplasts first develop into photosynthetically active chloronemal filaments, formed by cells filled with chloroplasts and separated by walls perpendicular to the axis of the filaments. After 5 days, chloronemal apical cells occasionally differentiate into caulonemal cells characterized by oblique cell walls and smaller and fewer chloroplasts. After 8–10 days, primitive meristems called buds differentiate from caulonemal side branch initials and further differentiate into leafy shoots, also known as gametophores. Each leafy shoot resembles a small plant with an unbranched stem, filamentous basal and aerial rhizoids and leaves displaying clear phyllotaxis along the stem axis (21). We found that the msh2 mutants had several pronounced phenotypes affecting the development of both protonema and leafy shoots.
Gametophytic developmental failures
Preliminary observations of msh2 mutant colonies regenerated from protoplasts indicated that these colonies were larger than those obtained from wild-type protoplasts (data not shown). To further quantify this observation, we compared the size of protonema filaments by measuring the length of the three apical cells from 14-day-old regenerated colonies. We found a significant difference in the mean length between the wild-type and the Ppmsh2 mutant (4.2 ± 1 versus 7.2 ± 2, arbitrary units, t-test P < 0.0001, n = 40) (Figure 2A). These data suggest that the Ppmsh2 mutant cells elongate 1.5 times faster than the wild-type cells. We next evaluated the cell division rate in the two backgrounds by counting the number of cell divisions after 8 days from regenerated protoplasts. We found a significant difference for mean cell division numbers between the wild-type and for the Ppmsh2 mutant (5 ± 0.5 versus 6 ± 0.8, t-test P < 0.0001) (Figure 2B), showing that the number of cell divisions was slightly increased in the mutant.
Figure 2 Comparison of wild type and Ppmsh2 mutants apparent phenotypes: (A) Comparison of the length of the last three cells of protonema filaments; arrows show the cell walls. (B) Number of cell divisions after 8 days protoplast regeneration; arrows show the cell walls. (C) Wild-type and Ppmsh2 mutant colonies after 3 weeks development. (D) Male antheridia of wild-type and of Ppmsh2 mutant, (E) female archegonia of wild-type and of Ppmsh2 mutant. (F) Wild-type, Ppmsh2#1, Ppmsh2#3 and Ppmsh2#2 mutant colonies after 12 weeks development, a representative leafy shoot from those colonies and enlargement of the top of the same gametophore.
Differentiation of chloronema cells into caulonema cells looks normal in the mutant compared with wild type. After 10 days of growth, differentiation of buds and further development of leafy shoots could clearly be observed in WT colonies whereas, at the same time-point, the Ppmsh2 mutant colonies had neither developed buds nor leafy shoots and the protonema was more widespread (data not shown). After 3 weeks, wild-type colonies had numerous well-developed leafy shoots, whereas Ppmsh2 filaments had just developed a few leafy shoots that were thinner, shorter and occasionally distorted (Figure 2C). We therefore concluded that loss of function of MSH2 seriously impairs the normal development of P.patens at different levels: protonema cell growth is accelerated, protonema cell cycle may be shorter, differentiation of buds is delayed and further development of leafy shoots is architecturally affected.
Reproductive developmental failures
The late developmental stages of P.patens life cycle are the differentiation of reproductive organs, fertilization to produce the diploid sporophyte and meiosis of spore mother cells to produce haploid spores. Therefore, we examined whether the developmental abnormalities seen at the protonema and leafy shoot stages could also be observed in antheridia (male) and archegonia (female), the reproductive organs of P.patens that differentiate at the apex of leafy shoots. After 10 weeks, the reproductive organs of wild type were fully developed on leafy shoot (Figure 2D and E) and the sporophyte further developed normally to produce viable spores (data not shown). In all five Ppmsh2 mutants, abnormal reproductive organs were formed on the few leafy shoots (Figure 2C), with aberrant number and shape of cells (Figure 2D and E). Consequently we never obtained differentiated capsules from the mutants and concluded that loss of function of MSH2 led to complete sterility in P.patens. To determine whether sterility was due to a problem in male and/or female gametes, reciprocal out-crosses were performed between Ppmsh2 mutants and strain NicB5ylo6. We did not obtain any capsules from these crosses and thus concluded that Ppmsh2 mutants were male and female sterile as a consequence of abortive development of reproductive organs.
Late developmental defects
In order to check the phenotype of late growing colonies, we inoculated similar size filaments (5–10 cells) from 3-week-old single protoplasts derived colonies on standard medium. After 6 weeks, the mutant colonies started to become chlorotic and a significant fraction of the cells were dead (data not shown). After 12 weeks, wild-type colonies were healthy (Figure 2F, WT) whereas those from independent Ppmsh2 knockouts presented various developmental aberrations from leaf shape abnormalities (Figure 2F, Ppmsh2#1) or supernumerary leaves and ‘callus’-like structure (Figure 2F, Ppmsh2#3) to completely disorganized necrotic shoots with numerous aerial rhizoids (Figure 2F, Ppmsh2#2). Additionally, the frequency of dead cells in these colonies was directly proportional to the intensity of the phenotype. In the most dramatic case (Ppmsh2#2) the leafy shoots were completely dead since isolated gametophores plated on fresh standard medium were not able to regenerate viable protonema, which was feasible with WT leafy shoots isolated from the same culture (data not shown). We also observed that the intensity of the phenotypes increased proportionally to the number of subcultures the original clone has undergone (data not shown), suggesting that random mutations accumulated within Ppmsh2 mutant clones in successive subcultures. Those late phenotypes could clearly be associated with major developmental processes, such as the establishment of shoot architecture, early senescence and eventually active cell death. The diversity of the phenotypes observed in Ppmsh2 mutants supports a strong mutator phenotype that is immediately detected in the moss haploid gametophyte. For this reason all the following experiments have been carried out using 3–4 individual replicates of the primary isolated knockout clones (stored at 4°C) as starting material.
Ppmsh2 mutants have a mutator phenotype
The strong developmental defects observed in Ppmsh2 mutants could be related to accumulation of spontaneous mutations. This phenotype has already been observed in msh2 mutants of other organisms (2). To further quantify this mutator phenotype, we investigated the effect of the loss of the PpMSH2 gene on mutation frequency of the adenine phosphoribosyl transferase (PpAPT) reporter gene. Spontaneous mutations leading to loss of APT activity confers resistance to adenine analogues, such as 2-fluoroadenine (2-FA) and can be selected on standard medium supplemented with 10 μM 2-FA. We tested a total of 107 Ppmsh2 protoplast-derived colonies and 3 x 106 wild-type colonies for resistance to 2-FA.
We observed no resistant colonies in the wild type and a total of 398 2-FA resistant colonies in the Ppmsh2 mutant. APT mutation frequencies were thus lower than 3 x 10–7 in wild type and 4 x 10–5 ± 1.2 x 10–5 in Ppmsh2. To confirm that 2-FA was effectively resulting from mutation in the APT gene, we partially sequenced the APT gene of randomly chosen 2-FA resistant clones. Point mutations in the APT sequence were effectively detected in more than 75% of the clones tested (Table 1). Mutations resulting from transitions and transversions were recovered, but we did not identify mutations resulting from insertion or deletion. These modifications were matching with highly conserved domain or amino acid of the APT protein and some of them were identified several times in independent 2-FA resistant clones. These results demonstrate that loss of MSH2 gene leads to the accumulation of point mutations in the moss genome and that the spontaneous mutation rate is increased at least by two orders of magnitude (130 times) in the Ppmsh2 mutant compared with the wild type. This last value is probably underestimated since the screen used here has selected only mutations that confer 2-FA resistance. Additionally, the type of mutations identified suggests that they result from defaults of post DNA replication proofreading processes.
Table 1 Point mutations identified in the APT genomic sequence in Ppmsh2 2-FA resistant clones
Msh2 mutants display altered responses to genotoxic stresses
MMR is known to recognize different types of DNA damages depending on the lesion and on the organism. MMR is responsible for the hypersensitivity to agents, such as radiation or some genotoxic chemicals that create DNA lesions (22). Loss of MMR may prevent cell death normally resulting from exposure to such genotoxic agents, thus conferring tolerance to these stresses. Therefore, we treated Ppmsh2 mutant protoplasts with UV-B and -radiation and chemicals such as cisplatin and MNU to determine whether P.patens cells with defective MMR would be affected or not. Survival was calculated as the ratio of protoplasts surviving after 15 days regeneration following treatment to the number of protoplasts undergoing normal regeneration without treatment.
Ppmsh2 mutants are equally sensitive to -radiations than wild type
Ionizing radiation can cause many different lesions in DNA, including strand breaks and base or sugar damage (23,24). Oxidized base damages induced by ionizing radiation may therefore be substrate for MMR (22). We compared the effect of -radiation on the survival of Ppmsh2 mutant and wild-type protoplasts. We found that both wild-type cells and Ppmsh2-deficient cells were similarly affected by -radiation (Figure 3A). For both genotypes, the LD50 (Lethal Dose 50%) was reached at 600 Gray. All -radiation dose comparisons were not significantly different.
Figure 3 Survival fraction after different genotoxic stresses. (A) -radiation, (B) UV-B, (C) Cisplatin, (D) MNU. Wild-type survival is represented with triangles and dashed line, and Ppmsh2 mutant survival is represented with squares and solid line. Error bars indicate SD based on at least two independent experiments in all cases.
Ppmsh2 mutants are more sensitive to UV-B radiations than wild type
A high UV-B flux can introduce a number of different lesions into the genome, mostly cyclobutane pyrimidine dimers (CPDs) and pyrimidine-6-4-pyrimidinone adducts (6-4PPs) (25). The elimination of these DNA photoproducts is essential for cell survival because both RNA- and DNA-polymerase are unable to read through both of these classes of dimers. Therefore, we tested the effect of increasing UV-B doses on the survival of regenerated wild-type and Ppmsh2 mutant protoplasts. We found that wild-type protoplasts were affected by high doses of UV-B, with an LD50 of 36 KJ/m2. In the Ppmsh2 background, the lethal dose was only 18 KJ/m2 (Figure 3B). All UV-B dose comparisons were significantly different. These results show that the Ppmsh2 mutant is more sensitive to UV-B than the wild type.
Ppmsh2 mutants are more tolerant to cisplatin than wild type
Cisplatin is a DNA-damaging drug that forms bifunctional covalent adducts with DNA and is a well-recognized mutagen that affects the survival of mammalian cells (26). We compared survival of wild-type and Ppmsh2 mutant regenerated protoplasts after exposure to increasing concentrations of cisplatin. The LD50 for the wild type was 250 μM cisplatin (Figure 3C), whereas the Ppmsh2 cells were more resistant than wild-type cells, with an LD50 of about 500 μM. All concentration comparisons were significantly different. These experiments show that the Ppmsh2 moss cells have a cisplatin tolerant phenotype.
Ppmsh2 mutants are more tolerant to MNU than wild type
The cytotoxicity of methylating agents such as MNU (N-methyl-N-nitrosourea) is primarily caused by their ability to introduce O6-methylguanine (O6-meGua) into the DNA (27). Persistent O6-meGua in the DNA is normally lethal. Therefore, we determined survival of wild-type and Ppmsh2mutant cells after exposure to increasing concentration of MNU. Wild-type survival was slightly affected by 1 mM MNU and dramatically affected by 5 mM MNU treatment. These values are comparable with those reported for MNU-induced DNA damage in Barley (28). The survival of Ppmsh2 mutant cells was slightly better than wild type at 1 mM MNU and significantly higher at 5 mM (Figure 3D). The LD50 was reached at 500 μM for both genotypes. These results show that Ppmsh2 moss cells have a weak resistance to MNU at high doses.
Gene targeting is decreased and homeologous recombination is not repressed in Ppmsh2 mutants
In E.coli (29), Saccharomyces cerevisiae (5), Trypanosoma brucei (30) and mouse embryogenic stem cells (7), depletion of MutS or MSH2 leads to an increased frequency of recombination between homeologous DNA sequences. We examined the involvement of PpMSH2 in HR by measuring GT efficiency with homologous or homeologous DNA sequences at the APT locus. Wild-type and Ppmsh2 cells were transformed with APT constructs presenting 0, 1, 2 and 3% divergence to the PpAPT genomic sequence, respectively (Figure 4). Hygromycin resistant transgenic clones were initially selected and tested for 2-FA resistance. GT frequency was calculated as the ratio of 2-FA resistant clones to the total number of Hygromycin resistant transformants.
We observed a significant decrease (1.8-fold) in GT frequency for mutant cells compared to wild type (P = 0.001) when the recombination substrate was fully homologous to its genomic target (Table 2). In wild type, GT frequency decreased as the percentage of divergence increased, whereas GT frequency remained stable for the Ppmsh2 mutant whatever the divergence level (Table 2). In wild type GT frequency significantly decreased 22-fold with 3% mismatches compared to 0% mismatches (2 test, P < 0.00001), whereas in Ppmsh2 mutant, the GT was as efficient with homeologous fragments as with homologous fragment (P = 0.40). Thus, GT efficiency with a homeologous fragments containing 3% divergence was increased 15-fold in Ppmsh2 mutant compared to the wild type. These data demonstrate that loss of function of MSH2 allows GT in P.patens with up to 3% divergent homeologous sequences, hence that the moss MSH2 protein has an antirecombination activity that prevents recombination between slightly divergent sequences.
Table 2 Gene targeting frequency in wild type and Ppmsh2 mutant
DISCUSSION
Ppmsh2 mutants show strong developmental defects, are sterile and have a mutator phenotype
Ppmsh2 mutant colonies develop strong developmental defects, including sterility, after a few weeks. This sterility may be due to the msh2 associated mutator phenotype increasing the rate of mutations that cause developmental problems in complex structures such as leafy shoots, antheridia and archegonia. Similar phenotypes have been observed in Arabidopsis thaliana msh2 mutant lines (10). After five generations, plants showed abnormalities in morphology and development, fertility, germination efficiency, seed/silique development and seed set. In human and mice, mutations in MSH2 result in a greatly increased likelihood of developing certain types of tumors (31). As in other eukaryotes, during growth and development, MSH2 in moss seems to be essential for the maintenance of genome integrity. Accordingly, we scored a 130-fold increase in the spontaneous mutation rate of the APT gene in Ppmsh2 mutants compared with wild type. This increased mutation rate is comparable with those observed in E.coli mutS mutant (20-fold increase) (32), or S.pombe (15-fold increase) (33) and S.cerevisiae (200-fold increase) (8) and other higher eukaryotes msh2 mutants (from 30- to several hundred-fold increase (7,10,34,35). This mutation rate can explain the phenotypes observed in the Ppmsh2 mutants. Assuming that the amount of mutations increases with the number of rounds of DNA replication, abnormalities should increase with the number of generations of vegetative propagation and this is what we observed after multiple successive subcultures of the colonies. Nevertheless, the phenotypes we observed in the mutant background are particularly strong, and two explanations can be proposed to account for this observation. The first one is linked to the fact that the major part of the moss life cycle is haploid. Thus, the deleterious effects of mutations in development are greater in moss than for diploid mice or A.thaliana, where the allelic locus would mask the phenotype resulting from randomly occurring mutations, at least in the first generation. Moreover, we cannot discard the hypothesis that, as in budding yeast, efficient GT in P.patens reflects the predominance of the HR pathway in the repair of DNA DSB. In this case, alteration of the MMR system, which participates in this DNA repair pathway, would lead to more drastic effects than in other higher eucaryotes, where HR is not the major pathway for DNA DSB repair. This is reminiscent of the strong reduction in spore viability observed in the msh2 yeast mutant (36) an organism where HR is also predominant.
Is PpMSH2 protein involved in cell-cycle checkpoints and apoptosis-induced by genotoxic stresses?
The sensitivity of cells to agents that create lesions on DNA has been attributed to attempts of MMR to correct the resulting damage (22). Depending on the type of DNA damage, loss of MMR may result in increased mutagenesis, loss of cell-cycle control and resistance to apoptosis (37). The resistance of msh2-deficient cells to DNA damage is not so clear because it depends on the organism and on the specificity of the lesions. In C.elegans, MSH2 confers sensitivity to ionizing radiation (9). In mouse cells, conflicting results have been reported, although the effects are generally minor. Msh2-nullizygous cells were found to exhibit slight higher levels of survival after exposure to ionizing radiation (38,39). However, it has recently been shown that both mouse embryogenic fibroblasts (MEFs) and mouse colorectal carcinoma cells defective in MSH2 were slightly more sensitive (1.5-fold) to X-ray irradiation than their MSH2-non-defective counterparts (40). In contrast, in moss, as in drosophila and human cells, the loss of MSH2 does not affect the sensitivity to -radiation, suggesting that MSH2 is not involved in the detection of lesions due to -radiation (41,42).
In E.coli, T.brucei, mouse and human cells, MSH2 confers sensitivity to MNU (27,38,43,44). In contrast, in the yeast S.cerevisiae, the MMR system did not affect sensitivity to methylating agents (45). In our experiments, Ppmsh2 mutants showed only a weak resistance to MNU at high concentration. This suggests that only a fraction of the MNU cytotoxicity in moss cells may be dependent on MSH2 function. Other mechanisms must be involved to repair DNA lesion in conjunction or not with MMR. HR can repair DNA lesions, like it has been shown in yeast for O6-meGua damages (45). Absence of MMR in S.cerevisiae did not affect sensitivity to methylating agents, while HR inactivation sensitized cells to this agent. In HR-deficient cells, defects in the MSH2 gene rescue the original sensitivity (45). These authors conclude that methylation damage processing and repair by HR masks the sensitivity of MMR-proficient S.cerevisiae cells toward methylating agents (45). In our experiments, Ppmsh2 mutants showed a weak resistance to MNU only at high concentration, suggesting that only a fraction of the MNU cytotoxicity in moss cells may be dependent on MSH2 function. Like in yeast, moss MMR could be overwhelmed by other mechanisms such as HR, thus preventing the cytotoxicity due to methylating agents to be detectable in wild type.
Peters et al. (37) have shown that in MEF (mouse embryogenic fibroblasts), MSH2 may sense damage due to photoproducts or may contribute to the processing of UV-B-induced DNA damage, thereby initiating apoptosis and influencing the activation of p53 (a tumor suppressor which functions as a transcription factor involved in cell-cycle checkpoints and apoptosis). They showed that Msh2-null MEFs were partially resistant to the cytotoxic effects of UV-B, and that this resistance led to a reduction in apoptosis. In contrast, we have shown that the Ppmsh2 mutant is more sensitive to UV-B radiation than the wild type. Therefore, MSH2 in moss may play a more important role in the recognition of UV-B lesions than that reported in MEF.
Experiments in E.coli have shown that loss of MMR activity, due to mutations in mutS, was correlated with cisplatin resistance (46,47). There are conflicting results in the response of cisplatin treatment in higher eukaryotes. It was thought that mouse MSH2 conferred sensitivity to cisplatin (48). However, Claij and te Riele (49) have recently shown that msh2-deficient mouse cells do not respond specifically to cisplatin lesions compared with wild type (49). In contrast, data from Trypanosoma cruzei have demonstrated a role of MSH2 in sensitivity to cisplatin (50), like in prokaryotes and in yeast (46,51). Our results in moss show that Ppmsh2 mutants are tolerant to cisplatin.
The fact that Ppmsh2 mutants have opposite responses to lesions due to UV-B (sensitivity) and cisplatin (resistance) suggests that moss may have different pathways for managing lesions involving PpMSH2 proteins. PpMSH2 may signal DNA lesions and induce cell-cycle arrest followed by either DNA repair or apoptosis to prevent accumulation of a large number of mutations. In the absence of PpMSH2, we suggest that, depending on the type of DNA lesions, different responses may be engaged. When the lesion is too severe (exposure to UV-B), the cells are unable to repair them and die. For other lesions, they can be bypassed by a translesion synthesis polymerase (52,53) or be repaired by other mechanisms such as base excision repair, nucleotide excision repair or by HR (45,54,55). In these cases, the cells can survive almost normally, as observed with moss msh2 mutant cells exposed to cisplatin. We suggest that in wild-type cells, the DNA lesions caused by cisplatin trigger cell death and/or prolonged cell-cycle arrest through PpMSH2. We have evidence that the cell cycle is perturbed in the absence of PpMSH2: during the regeneration of protoplasts, Ppmsh2 mutants showed a slightly higher rate of cell division compared with the wild type. This suggests that Ppmsh2 mutants may have an accelerated cell cycle, which may be explained if there was shorter arrest at cell-cycle checkpoints. Indeed, in mice it has been shown that after DNA damage, the absence of MSH2 results in premature release from the early G2/M arrest and an inability to activate properly the checkpoint kinase proteins (40). It was suggested that an active MSH2 is required for a correct response to DNA damage in the G2 phase of the cell cycle, possibly connecting DSB repair to checkpoint signaling. Moreover, in human, MMR is required for S-phase and G2-M checkpoint regulation (56,57). Consistent with these findings, we suggest that PpMSH2 is required for efficient cell-cycle arrest induced by DNA damage and may be a key component acting on DNA damage-activated checkpoints.
PpMSH2 is involved in HR and is essential to homeologous recombination
We examined the influence of MMR on HR in P.patens by assaying the frequency of recombination between a foreign DNA fragment released in vivo and a genomic target sequence. We found that the frequency of GT via HR between identical DNA was decreased 1.8-fold in the Ppmsh2-defective cells compared with the wild type. In mice, msh2 mutants show a slight decrease of GT efficiency with homologous fragment compared with wild-type mice, and this decrease is comparable with our results (1.5-fold) (7). In yeast, a recent study has shown that the frequency of GT was reduced 3- to 4-fold in the msh2 strain (58). Together with our results, these suggest that MSH2 in P.patens is probably involved in mechanisms that regulate GT of perfectly matching sequences in the genome. Further analyses are needed to determine the mechanisms involved.
We tested the influence of MMR on homeologous recombination in P.patens by assaying the frequency of recombination between a mutated DNA fragment and a genomic target sequence. Our results show that DNA sequence identity is important to determine the efficiency of gene replacement in the wild type. We have also shown that lack of PpMSH2 allows homeologous sequences with up to 3% divergence to be as efficiently targeted as perfectly identical sequences to a genomic locus. This shows that PpMSH2 plays a major role to prevent recombination between homeologous sequences. This is consistent with the similar role described for MMR in E.coli (29), in Trypanosoma (30), in yeast (5) and in mouse cells (7,59,60), suggesting that this role of MSH2 is strongly preserved during evolution. However, this is the first time that MSH2 involvement in HR and homeologous recombination is shown in plants.
In conclusion, we have found that in moss, MSH2 is involved in mismatch recognition and repair that most likely results from DNA replication errors, and in the response to certain genotoxic agents. Moreover, MSH2 plays a role in HR and in the recognition of mismatches that occur during heteroduplex formation between two not perfectly complementary DNA strands. These mechanisms are essential for maintaining genome integrity in all organisms and to establish interspecific barrier. This is particularly true for plants as they are unable to move and must cope with exposure to environmental mutagens. Thus, an efficient mechanism for genomic stability is needed in plants, and even more in mosses, where the major developmental stage of the life cycle is the haploid gametophyte. This is emphasized by the dramatic deleterious effect on development observed in an msh2 P.patens mutant compared with its A.thaliana counterpart. Our results show for the first time, the crucial role of MMR in a pluricellular organism where HR seems to be one of the major pathways to repair DNA lesions.
ACKNOWLEDGEMENTS
The authors thank Mr Favodon, (Institut Curie, Orsay, France), for help with the 137Cs irradiator for gamma radiations assays. The authors thank Dr Michel Laloue (INRA Versailles, France) for providing the 2-Fluoroadenine. The authors thank Marie-Pascale Doutriaux, Mathilde Grelon, Christine Mézard and Rapha?l Mercier for suggestions and discussions on the manuscript. This work was supported by grant from the Institut National de la Recherche Agronomique and Cellectis SA (Romainville, France). Funding to pay the Open Access publication charges for this article was provided by Institut National de la Recherche Agronomique.
REFERENCES
Modrich, P. and Lahue, R. (1996) Mismatch repair in replication fidelity, genetic recombination, and cancer biology Annu. Rev. Biochem, . 65, 101–133 .
Harfe, B.D. and Jinks-Robertson, S. (2000) DNA mismatch repair and genetic instability Annu. Rev. Genet, . 34, 359–399 .
Kolodner, R. (1996) Biochemistry and genetics of eukaryotic mismatch repair Genes Dev, . 10, 1433–1442 .
Schofield, M.J. and Hsieh, P. (2003) DNA mismatch repair: molecular mechanisms and biological function Annu. Rev. Microbiol, . 57, 579–608 .
Negritto, M.T., Wu, X., Kuo, T., Chu, S., Bailis, A.M. (1997) Influence of DNA sequence identity on efficiency of targeted gene replacement Mol. Cell. Biol, . 17, 278–286 .
Sia, E.A., Kokoska, R.J., Dominska, M., Greenwell, P., Petes, T.D. (1997) Microsatellite instability in yeast: dependence on repeat unit size and DNA mismatch repair genes Mol. Cell. Biol, . 17, 2851–2858 .
de Wind, N., Dekker, M., Berns, A., Radman, M., te Riele, H. (1995) Inactivation of the mouse Msh2 gene results in mismatch repair deficiency, methylation tolerance, hyperrecombination, and predisposition to cancer Cell, 82, 321–330 .
Greene, C.N. and Jinks-Robertson, S. (2001) Spontaneous frameshift mutations in Saccharomyces cerevisiae: accumulation during DNA replication and removal by proofreading and mismatch repair activities Genetics, 159, 65–75 .
Degtyareva, N.P., Greenwell, P., Hofmann, E.R., Hengartner, M.O., Zhang, L., Culotti, J.G., Petes, T.D. (2002) Caenorhabditis elegans DNA mismatch repair gene msh-2 is required for microsatellite stability and maintenance of genome integrity Proc. Natl Acad. Sci. USA, 99, 2158–2163 .
Hoffman, P.D., Leonard, J.M., Lindberg, G.E., Bollmann, S.R., Hays, J.B. (2004) Rapid accumulation of mutations during seed-to-seed propagation of mismatch-repair-defective Arabidopsis Genes Dev, . 18, 2676–2685 .
Schaefer, D. (2002) A new moss genetics: targeted mutagenesis in Physcomitrella patens Annu. Rev. Plant Biol, . 53, 477–501 .
Brun, F., Gonneau, M., Doutriaux, M.P., Laloue, M., Nogué, F. (2001) Cloning of the PpMSH-2 cDNA of Physcomitrella patens, a moss in which gene targeting by homologous recombination occurs at high frequency Biochimie, 83, 1003–1008 .
Ashton, N.W. and Cove, D.J. (1977) The isolation and preliminary characterisation of auxotrophic and analogue resistant mutants of the moss, Physcomitrella patens Mol. Gen. Genet, . 87–95 .
Asthon, N.W., Grimsley, N., Cove, D.J. (1979) Analysis of gametopytic development in the moss, Physcomitrella patens, using auxin and cytokinn resistants mutants Planta, 427–435 .
Sambrook, J., Frisch, E.F., Maniatis, T. Molecular Cloning: A Laboratory manual, (1989) NY Cold Spring Harbor Laboratory Press .
Cadwell, R.C. and Joyce, G.F. (1994) Mutagenic PCR PCR Methods Appl, . 3, S136–S140 .
Bilang, R., Iida, S., Peterhans, A., Potrykus, I., Paszkowski, J. (1991) The 3' terminal region of the hygromycin-B-resistance gene is important for its activity in Escherichia coli and Nicotiana tabacum Gene, 100, 247–250 .
Schaefer, D.G. and Zr?d, J.P. (1997) Efficient gene targeting in the moss Physcomitrella patens Plant J, . 11, 1195–1206 .
Sauer, B. (1993) Manipulation of transgenes by site-specific recombination: use of Cre recombinase Methods Enzymol, . 225, 890–900 .
Albert, H., Dale, E.C., Lee, E., Ow, D.W. (1995) Site-specific integration of DNA into wild-type and mutant lox sites placed in the plant genome Plant J, . 7, 649–659 .
Schaefer, D.G. and Zryd, J.P. (2001) The moss Physcomitrella patens, now and then Plant Physiol, . 127, 1430–1438 .
Fedier, A. and Fink, D. (2004) Mutations in DNA mismatch repair genes: implications for DNA damage signaling and drug sensitivity (review) Int. J. Oncol, . 24, 1039–1047 .
Hutchinson, F. (1985) Chemical changes induced in DNA by ionizing radiation Prog. Nucleic Acid Res. Mol. Biol, . 32, 115–154 .
Ward, J.F. (1988) DNA damage produced by ionizing radiation in mammalian cells: identities, mechanisms of formation, and reparability Prog. Nucleic Acid Res. Mol. Biol, . 35, 95–125 .
Jasen, M.A.K., Gaba, V., Greengerg, B.M. (1998) Higher plants and UV-B radiation: balancing damage, repair and acclimation Trends Plant Sci, . 3, 131–135 .
Aebi, S., Kurdi-Haidar, B., Gordon, R., Cenni, B., Zheng, H., Fink, D., Christen, R.D., Boland, C.R., Koi, M., Fishel, R., et al. (1996) Loss of DNA mismatch repair in acquired resistance to cisplatin Cancer Res, . 56, 3087–3090 .
Humbert, O., Fiumicino, S., Aquilina, G., Branch, P., Oda, S., Zijno, A., Karran, P., Bignami, M. (1999) Mismatch repair and differential sensitivity of mouse and human cells to methylating agents Carcinogenesis, 20, 205–214 .
Jovtchev, G., Menke, M., Schubert, I. (2001) The comet assay detects adaptation to MNU-induced DNA damage in barley Mutation Res, . 493, 95–100 .
Rayssiguier, C., Thaler, D.S., Radman, M. (1989) The barrier to recombination between Escherichia coli and Salmonella typhimurium is disrupted in mismatch-repair mutants Nature, 342, 396–401 .
Bell, J.S. and McCulloch, R. (2003) Mismatch repair regulates homologous recombination, but has little influence on antigenic variation, in Trypanosoma brucei J. Biol. Chem, . 278, 45182–45188 .
Buermeyer, A.B., Deschenes, S.M., Baker, S.M., Liskay, R.M. (1999) Mammalian DNA mismatch repair Annu. Rev. Genet, . 33, 533–564 .
Schaaper, R.M. and Dunn, R.L. (1987) Spectra of spontaneous mutations in Escherichia coli strains defective in mismatch correction: the nature of in vivo DNA replication errors Proc. Natl Acad. Sci. USA, 84, 6220–6224 .
Rudolph, C., Kunz, C., Parisi, S., Lehmann, E., Hartsuiker, E., Fartmann, B., Kramer, W., Kohli, J., Fleck, O. (1999) The msh2 gene of Schizosaccharomyces pombe is involved in mismatch repair, mating-type switching, and meiotic chromosome organization Mol. Cell. Biol, . 19, 241–250 .
Denver, D.R., Feinberg, S., Estes, S., Thomas, W.K., Lynch, M. (2005) Mutation rates, spectra, and hotspots in mismatch repair-deficient Caenorhabditis elegans Genetics, .
Malkhosyan, S., McCarty, A., Sawai, H., Perucho, M. (1996) Differences in the spectrum of spontaneous mutations in the hprt gene between tumor cells of the microsatellite mutator phenotype Mutat Res, . 316, 249–259 .
Reenan, R.A.G. and Kolodner, R.D. (1992) Characterization of insertion mutations in the Saccharomyces cerevisiae Msh1 and Msh2 genes—evidence for separate mitochondrial and nuclear functions Genetics, 132, 975–985 .
Peters, A.C., Young, L.C., Maeda, T., Tron, V.A., Andrew, S.E. (2003) Mammalian DNA mismatch repair protects cells from UVB-induced DNA damage by facilitating apoptosis and p53 activation DNA Repair (Amst.), 2, 427–435 .
Fritzell, J.A., Narayanan, L., Baker, S.M., Bronner, C.E., Andrew, S.E., Prolla, T.A., Bradley, A., Jirik, F.R., Liskay, R.M., Glazer, P.M. (1997) Role of DNA mismatch repair in the cytotoxicity of ionizing radiation Cancer Res, . 57, 5143–5147 .
DeWeese, T.L., Shipman, J.M., Larrier, N.A., Buckley, N.M., Kidd, L.R., Groopman, J.D., Cutler, R.G., te Riele, H., Nelson, W.G. (1998) 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. USA, 95, 11915–11920 .
Franchitto, A., Pichierri, P., Piergentili, R., Crescenzi, M., Bignami, M., Palitti, F. (2003) The mammalian mismatch repair protein MSH2 is required for correct MRE11 and RAD51 relocalization and for efficient cell cycle arrest induced by ionizing radiation in G2 phase Oncogene, 22, 2110–2120 .
Aquilina, G., Crescenzi, M., Bignami, M. (1999) Mismatch repair, G(2)/M cell cycle arrest and lethality after DNA damage Carcinogenesis, 20, 2317–2326 .
Flores, C. and Engels, W. (1999) Microsatellite instability in Drosophila spellchecker1 (MutS homolog) mutants Proc. Natl Acad. Sci. USA, 96, 2964–2969 .
Karran, P. and Marinus, M.G. (1982) Mismatch correction at O6-methylguanine residues in E.coli DNA Nature, 296, 868–869 .
Bell, J.S., Harvey, T.I., Sims, A.M., McCulloch, R. (2004) Characterization of components of the mismatch repair machinery in Trypanosoma brucei Mol. Microbiol, . 51, 159–173 .
Cejka, P., Mojas, N., Gillet, L., Schar, P., Jiricny, J. (2005) Homologous recombination rescues mismatch-repair-dependent cytotoxicity of S(N)1-type methylating agents in S.cerevisiae Curr. Biol, . 15, 1395–1400 .
Calmann, M.A., Nowosielska, A., Marinus, M.G. (2005) Separation of mutation avoidance and antirecombination functions in an Escherichia coli mutS mutant Nucleic Acids Res, . 33, 1193–1200 .
Fram, R.J., Cusick, P.S., Wilson, J.M., Marinus, M.G. (1985) Mismatch repair of cis-diamminedichloroplatinum(II)-induced DNA damage Mol. Pharmacol, . 28, 51–55 .
Fink, D., Zheng, H., Nebel, S., Norris, P.S., Aebi, S., Lin, T.P., Nehme, A., Christen, R.D., Haas, M., MacLeod, C.L., et al. (1997) In vitro and in vivo resistance to cisplatin in cells that have lost DNA mismatch repair Cancer Res, . 57, 1841–1845 .
Claij, N. and te Riele, H. (2004) Msh2 deficiency does not contribute to cisplatin resistance in mouse embryonic stem cells Oncogene, 23, 260–266 .
Augusto-Pinto, L., Teixeira, S.M., Pena, S.D., Machado, C.R. (2003) Single-nucleotide polymorphisms of the Trypanosoma cruzi MSH2 gene support the existence of three phylogenetic lineages presenting differences in mismatch-repair efficiency Genetics, 164, 117–126 .
Drotschmann, K., Topping, R.P., Clodfelter, J.E., Salsbury, F.R. (2004) Mutations in the nucleotide-binding domain of MutS homologs uncouple cell death from cell survival DNA Repair (Amst.), 3, 729–742 .
Lehmann, A.R. (2005) Replication of damaged DNA by translesion synthesis in human cells FEBS Lett, . 579, 873–876 .
Hoffmann, J.S., Locker, D., Villani, G., Leng, M. (1997) HMG1 protein inhibits the translesion synthesis of the major DNA cisplatin adduct by cell extracts J. Mol. Biol, . 270, 539–543 .
Lindahl, T. and Wood, R.D. (1999) Quality control by DNA repair Science, 286, 1897–1905 .
Liu, Z., Hossain, G.S., Islas-Osuna, M.A., Mitchell, D.L., Mount, D.W. (2000) Repair of UV damage in plants by nucleotide excision repair: Arabidopsis UVH1 DNA repair gene is a homolog of Saccharomyces cerevisiae Rad1 Plant J, . 21, 519–528 .
Brown, K.D., Rathi, A., Kamath, R., Beardsley, D.I., Zhan, Q., Mannino, J.L., Baskaran, R. (2003) The mismatch repair system is required for S-phase checkpoint activation Nature Genet, . 33, 80–84 .
Yan, T., Schupp, J.E., Hwang, H.S., Wagner, M.W., Berry, S.E., Strickfaden, S., Veigl, M.L., Sedwick, W.D., Boothman, D.A., Kinsella, T.J. (2001) Loss of DNA mismatch repair imparts defective cdc2 signaling and G(2) arrest responses without altering survival after ionizing radiation Cancer Res, . 61, 8290–8297 .
Langston, L.D. and Symington, L.S. (2005) Opposing roles for DNA structure-specific proteins Rad1, Msh2, Msh3, and Sgs1 in yeast gene targeting EMBO J, . 24, 2214–2223 .
Elliott, B. and Jasin, M. (2001) Repair of double-strand breaks by homologous recombination in mismatch repair-defective mammalian cells Mol. Cell. Biol, . 21, 2671–2682 .
Abuin, A., Zhang, H., Bradley, A. (2000) Genetic analysis of mouse embryonic stem cells bearing Msh3 and Msh2 single and compound mutations Mol. Cell. Biol, . 20, 149–157 .(Bénédicte Trouiller, Didier G. Schaefer1)