Biologia 63/6: 941—946, 2008 Section Botany DOI: 10.2478/s11756-008-0146-4
Is it possible to improve homologous recombination in Chlamydomonas reinhardtii?* Miroslava Slaninová1, Dominika Hroššová1, Daniel Vlček1 & Wolfgang Mages2 1
Comenius University, Faculty of Natural Sciences, Department of Genetics, Mlynská dolina, SK-84215 Bratislava 4, Slovakia; e-mail:
[email protected] 2 Lehrstuhl f¨ ur Genetik, Universit¨ at Regensburg, Germany
Abstract: Targeted modification of the genome has long been an aim of many geneticists and biotechnologists. Gene targeting is a main molecular tool to examine biological effects of genes in a controlled environment. Effective gene targeting depends on the frequency of homologous recombination that is indispensable for the insertion of foreign DNA into a specific sequence of the genome. The main problem associated with the development of an optimal procedure for gene targeting in a particular organism is the variability of homologous recombination (HR) in different species. Chlamydomonas reinhardtii is an attractive model system for the study of many cellular processes and is also an interesting object for the biotechnology industry. In spite of many advantages of this model system, C. reinhardtii does not readily express heterologous genes and does not allow targeted integration of foreign DNA into its genome easily. This paper compares data obtained from several different experiments designed for improving gene targeting in different organisms and reviews the suitability of particular techniques in C. reinhardtii cells. Key words: homologous recombination; heterologous gene; homing endonuclease; ends-in; ends-out
Introduction During the last decades the molecular mechanisms and regulation of different recombination processes have increasingly become a focus of study. The application of techniques ranging from simple methods of genetic analysis to more and more sophisticated molecular techniques has resulted in a huge increase of knowledge that has contributed significantly to a better understanding of the essential roles of recombination in the maintenance of genomes. The basic substrate for the action of recombination mechanisms is a double strand break (DSB), which represents the most critical damage that can be inflicted to cells. A single unrepaired DSB is sufficient for apoptosis in mammalian cells (Rich et al. 2000). DSBs are generated as a consequence of a replication block, by environmental factors such as ionizing radiation, by cellular metabolic products, and as intermediates of recombination processes in the genome. For the repair of DSB, cells employ two different recombination mechanisms, namely homologous recombination (HR) and non-homologous end joining (NHEJ). HR is any process in which two identical or similar sequences interact and exchange genetic information. This process is usually very precise as is exemplified by the most common natural occurrence of HR in meiotic recombination and is also a very important DNA repair mecha-
nism. Many different models have been created with the ambition to explain the mechanism of HR. Most of the data have come from genetic analyses of lower eukaryotes, but molecular biology has not yet elucidated the details of HR (for a review see Smith 2001). NHEJ simply re-ligates the ends of a DSB resulting in a deletion of a sequence with a size between a few bp to several kb. It has been found that NHEJ plays an important role in V(D)J recombination by ligating of programmed DSBs (Taccioli et al. 1993). In bacteria and some species of yeast, HR has long been known as a major pathway for the repair of DSBs. More recently, HR has also been shown as a major DSBs repair pathway in vertebrate cells, in addition to non-homologous end joining (Johnson & Jasin 2001; Sonoda et al. 2006). In vertebrate cells, HR is operative predominantly in the late S and G2 phases when sister chromatids are available, resulting in gene conversion that most of the time is not associated with crossing-over. NHEJ can function in all phases of the cell cycle and requires a regulatory mechanism to choose between HR and NHEJ for recombination in particular processes (Sonoda et al. 2006). In spite of conservation of both pathways their relative contribution to DSB repair varies considerably among species. Targeted modification of the genome has been a long-standing goal of many geneticists and gene target-
* Presented at the International Symposium Biology and Taxonomy of Green Algae V, Smolenice, June 26–29, 2007, Slovakia.
c 2008 Institute of Botany, Slovak Academy of Sciences
942 ing is a powerful molecular tool to examine biological effects of genes. Successful gene targeting depends on the frequency of HR that is crucial for the insertion of foreign DNA into a particular sequence of the genome. Chlamydomonas reinhardtii is an attractive model system for studying many cellular processes and it is also an interesting object for the biotechnology industry. Despite many advantages of this model system, C. reinhardtii does not readily express heterologous genes and is indisposed to targeted integration of foreign DNA into its genome. Homologous recombination has been shown to occur in vegetative Chlamydomonas cells (Gumpel et al. 1994; Sodeinde & Kindle 1994; Zorin et al. 2005; Mages et al. 2007) but up to now targeted disruption of only a single endogenous gene (NIT8) has been reported (Nelson & Lefebvre 1995). HR occurs rarely in Chlamydomonas and, therefore, it is not yet available as a routine tool for a knockout or targeted replacementof Chlamydomonas genes (Mages et al. 2007). This paper compares the results of several different experiments aimed at developing methods for gene targeting in different species, with special reference to techniques specifically developed for Chlamydomonas reinhardtii cells. Expression of the genes responsible for HR One possible approach to increase the rate of HR is to introduce an active exogenous recombinase into a cell. The pairing of homologous molecules and strand exchange are key events in HR. This process is usually promoted by RecA-like proteins and their homologues which are widely distributed among prokaryotic and eukaryotic species (Lloyd & Sharp 1993; Brendel, et. al. 1997). Some laboratories have examined the ability of the E. coli RecA protein to contribute the repair of DNA damage in wild type or recombination-defective mutant strains with various results (Brozmanová et al. 1991; Slaninová et al. 1996; Morais et al. 1998; Dudáš et al. 2003). RecA protein that was successfully expressed in yeast cells was not able to complement a defect of the RAD51 gene whose product is the first known E. coli RecA homologue. On the other hand RecA protein was able to complement a defect of RAD52p which does not display homology with RecA, but which has strong binding activity to both ssDNA and dsDNA and which mediates Rad51 dependent strand exchange (Dudáš et al. 2003; Brozmanová et al. 2004). The effect of RecA protein over-expression on stimulation of gene targeting has been not examined in these studies, because HR is very efficient in yeast cells anyway. Another laboratory tested the potential stimulative effect of RecA protein over-expression on HR in mammalian cells and found that the frequency of gene targeting at the hprt locus increased tenfold in somatic cells expressing RecA protein (Shcherbakova et al. 2000). Mitomycin C is known to intercalate into DNA, leading to cross-links with complementary strands and blocking of DNA replication. Heterologous expression of the E. coli RecA protein increased the resistance of transgenic tobacco cells to
M. Slaninová et al. mitomycin C and moreover HR was stimulated at least tenfold (Reiss et al. 1996). Later the same group reported that RecA stimulates only sister chromatid exchange and the fidelity of DSBs repair but not gene targeting in plants transformed by Agrobacterium (Reiss et al. 2000). As predicted by the endosymbiont theory the chloroplast recombination system is related to the eubacterial recombination system. Therefore many homologues of the E. coli RecA protein were found in chloroplast genomes of numerous plant species. C. reinhardtii possesses a homologue of the E. coli RecA protein encoded in the nucleus and operating on chloroplast DNA. This protein is immunologically related to E. coli RecA and moreover over-expression of E. coli RecA protein from chloroplast DNA stimulates more than fifteen fold the frequency of plastid DNA recombination (Cerutti et al. 1995). Although the C. reinhardtii genome contains a homologue of the E. coli recA gene and thus C. reinhardtii should be able to perform HR, we were not able to express the cloned E. coli recA gene in C. reinhardtii from native C. reinhardtii promoter and 3’ regulatory signals free of doubt, nor were we able to verify its predicted effect on HR (our unpublished results). A major problem probably resided in the 20%-difference of the codon usage of the E. coli recA gene as compared to the codon usage that is generally observed in Chlamydomonas genes. Some new developments in gene targeting using heterologous genes were reported recently (Shaked et al. 2005; Tzfira & White 2005). These authors demonstrated a high frequency of gene targeting in Arabidopsis plants by expressing the yeast DNA repair protein Rad54p which is a member of he Swi/Snf2 family of ATP-dependent chromatinremodelling factors. In four independent experiments when targeting the cruciferin-encoding gene they observed 5–62-fold increases in HR compared to the wild type control (Shaked et al. 2005). In the light of what is presently known, the prospects for an improvement of HR in Chlamydomonas cells using heterologous gene expression are not very optimistic. The effects of heterologous enzymes expressed in different species are unclear and moreover stimulation of HR is quite variable. It is probably related to the different degrees of contribution and competitive roles of HR and NHEJ to DSB repair in different organisms. Participation of each mechanism is very often dependent on the production of free ends generated by DSBs and the latter is dependent on the availability of processing enzymes typical for a particular mechanism (Sonoda et al. 2006). Additionally, when we consider well-known difficulties in heterologous gene expression in Chlamydomonas cells (Cerutti et al. 1997), this approach for improving HR in Chlamydomonas is not very promising. Stimulation of HR using site-specific (endo)nucleases Homing endonucleases are a diverse collection of sitespecific endonucleases recognizing 14–40 bp sequences
Homologous recombination in Chlamydomonas and tolerating various degrees of degeneracy of target sequences (Belford & Roberts 1997). In vivo they play a pivotal role as target selector proteins in the homing of introns and inteins in the genome of bacteriophages, bacteria, archaea and unicellular eukaryotes. They are grouped into four families of which the LAGLIDADG family is the largest with at least 200 members (see at http://rebase.neb.com/cgi-bin/azlist/homing). Much interest has focused on the homing endonucleases due to their potential applicability as a tool in gene targeting. The pioneering experiments using highly specific DNA cleavage to investigate stimulation of HR by DSBs in mammalian cells have been made more than 13 years ago (Rouet et al. 1994). The authors used the I-SceI homing nuclease from yeast which has an 18-bp recognition site for the stimulation of HR. When a SceI recognition site was inserted into a target gene and endonuclease was expressed, HR and gene targeting were stimulated over 1000-fold (Rouet et al. 1994). I-CreI is a homing endonuclease whose gene was discovered in the chloroplast genome of C. reinhardtii (Rochaix & Malnoe 1978). I-CreI endonuclease is encoded by the chloroplast ribosomal group I intron and cleaves specifically intronless copies of the large ribosomal RNA (23S) gene. To examine DSB-induced recombination in the chloroplast, a target site for I-CreI endonuclease embedded in cDNA of the 23S gene was integrated at an ectopic location and the gene for I-CreI endonuclease was deleted in one strain. Genetic analysis in haploid progeny performed after crossing with a wild type strain demonstrated strong stimulation of HR mediated by DSBs. Gene conversion was observed when the 23S cDNA and the neighboring copy of the 23S gene were in opposite orientation, leading to mobilization of the intron to the 23S cDNA (D¨ urrenberger et al., 1996). I-ApeI is an intron-encoded endonuclease from the hyperthermophilic archeon Aeropyrum pernix and belongs to the LAGLIDADG family. It recognizes a 20bp sequence of which only 7 bp are essential while the other positions are variable (Nomura et al. 2005). In a number of recent studies, site-specific DNA DSBs have been used to induce efficient gene targeting. Novel engineered so-called meganucleases are produced for DSB induction in different organisms (Epinat et al. 2003). Endonuclease-induced gene targeting has, however, one major limitation which is that the target locus must contain an endonuclease cleavage site. Therefore the first step in such strategies is to introduce the nuclease recognition site into the target gene. Considering the low frequency of HR this is a big problem. Perspectives for using natural endonucleases as tools for gene targeting depend on the possibility of finding enzymes whose target sequences are already contained at the target loci. Since the natural repertoire of homing endonucleases is currently limited to about 300 proteins, most of them are still hypothetical or uncharacterized. It thus becomes necessary to resort to novel, artificial meganucleases, for which specificity has been tuned according to target choice (Epinat et al. 2003). Zinc-finger nucleases ZFN (originally termed chimeric
943 restriction enzymes) are also one of the newest tools for targeted mutagenesis (Porteus & Carroll et al. 2005; Tzfira & White 2005). This kind of enzymes was first developed following the hypothesis that novel sequence specificities could be created by fusing the nonsequencespecific cleavage domain of the FokI type II restriction endonuclease to a new DNA-binding domain (Kim et al. 1996; Chandrasegaran & Smith 1999; Kandavelou et al. 2004). Custom-made pairs of ZFN genes can be specifically designed to recognize a unique combination of 18 nucleotides and therefore ZNFs are able to recognize nearly every sequence of the genome and thus they represent promising tools for achieving gene targeting (Tzfira & White 2005). Different DNA topologies in gene targeting In many organisms, such as higher eukaryotes like mammals, targeted insertion is limited by a strong bias against HR. Since this is not the case in yeast, the history of targeted insertion mutagenesis consequently started with the first successful integrative transformation in Saccharomyces cerevisiae (Hinnen et al. 1978). Targeted insertion is induced by in vivo insertion of suitable vector DNA into a homologous locus of a chromosome after transformation of the cells. Three groups of basic topologies of targeting vectors are routinely used for gene targeting. First there are common circular vectors with a selection marker and with a region of homology to the target sequence. In the second group are so-called ends-in vectors, which are vectors that are linearized prior to transformation by restriction digestion and which carry deletions inside the homology region. Finally, there are so-called ends-out vectors in which the selection marker is inserted between two parts of the target sequence (Fig. 1). Usually, the selection marker does not share homology with the recipient DNA sequence. The main differences between two linear vectors are the configuration of their free ends and differences in the pairing with the target DNA. In Saccharomyces cerevisiae, transformation efficiency is usually much higher with ends-in vectors than with circular and ends-out vectors (Hastings et al., 1993); the last two have comparable transformation efficiencies. Very often ends-in vectors have been demonstrated to integrate in multiple tandem repeats not typical for circular and ends-out vectors (Klinner & Sch¨ afer 2004). Mitotic stability is highest for ends-out vectors, because their integration is not accompanied by the formation of duplicated homologous regions. Despite the fact that Saccharomyces cerevisiae is generally appreciated for its ability to carry out HR, some authors have demonstrated that the frequency of gene replacement by transformation of linear exogenous DNA is surprisingly low (Leung et al. 1997). These authors believe that ends-in and ends-out recombinations proceed by different mechanisms. Some of the pertinent data suggest that during ends-out recombination only one of two strands of DNA is assimilated creating heteroduplex DNA over the entire length. This is dif-
M. Slaninová et al.
944
Fig. 1. Structure and recombination of ends-in and ends-out vectors. Ends-in vectors are linearized and create a gap in the plasmid target region. Integration of these vectors occurs additively giving rise to two tandem copies of the target gene (one of the two copies still carrying a potential mutation represented by a vertical bar) separated by the vector plasmid. Ends-out vectors contain a central selection cassette integrated into a copy of the target gene which is, therefore, split into two parts. In this case, the targeted region is replaced by the incoming vector via homologous recombination between the two parts of the vector copy of the target gene and the homologous regions of the chromosomal copy of the target gene. Open circle, marker for orientation; vertical bar in target gene, site of mutation.
ferent from the more limited formation of heteroduplex DNA at the ends of DNA fragments undergoing endsin recombination. The most popular explanation of the mechanism of gene conversion that is supposed to take place in such events is synthesis dependent strand annealing (SDSA), in which the newly synthesized strand is displaced as a single strand from the migrating replication bubble and the complementary strand can be synthesized. Integration of all kinds of vectors is very often indicated by drawing crossing-over symbols, although a number of results support the assumption that crossing-over essentially does not occur during integration, particularly in the case of ends-out vectors (Klinner & Sch¨afer 2004). It is known that the majority of transgenic animals are mosaics suggesting that integration of targeting vectors occurs during DNA replication (Smith et al., 2001). Some data indicate that the free ends of the transgene construct initiate recombination by invasion of a replication bubble by one of the two DNA strands of the construct (Smith et al., 2001). A very important rate-limiting step is the subsequent mismatch correction by which heteroduplexes between the incoming and the resident DNA strands are repaired. It has been reported that 85% of the mismatches in heteroduplexes were repaired in favour of the resident DNA strand, even if the resident DNA strand was the mutant strand (Leung et al. 1997). In summary, data about the efficiency of gene targeting in different organisms by using ends-in or endsout vectors are discrepant. It is conceivable that dif-
ferent, and independent, HR pathways may be operational during a single targeting event. This is most likely the case with replacement ends-in or ends-out transgene vector constructs in which the homology regions are present at both ends of the transgene, separated by heterologous DNA. In such cases, an independent HR event can be expected at each end of the vector (Smith et al. 2001). Other studies scrutinized the potential of ssDNA to increase the rate of HR in gene targeting experiments (Rauth et al. 1986; Simon & Moore 1987; Baur et al. 1990; Bilang et al. 1992; Zorin et al. 2005) and since single-stranded DNA plays a central role in all proposed recombination models, the use of single-stranded targeting vectors is a promising approach for improving HR. Taken together, most of these experiments demonstrated an increased rate of HR. Only in the study of Zorin et al. (2005) the use of ssDNA did not lead to an obvious increase in the rate of HR but instead caused a strong reduction of nonhomologous DNA integration (which is the major route for integration of transforming DNA into the Chlamydomonas genome) and thus nevertheless allowed isolation of homologous recombinants. Concluding remarks In contrast to prokaryotes and some lower eukaryotes, many (especially higher) organisms prefer nonhomologous integration of foreign DNA into the genomes over homologous recombination. The number of homologous
Homologous recombination in Chlamydomonas recombinants that can be obtained in a particular experiment is dependent on several factors such as choice between various recombination systems in the cells of particular species, transformation methods, the topology and length of homology regions within targeting vectors, and the usage of single-stranded versus doublestranded targeting constructs. HR in Chlamydomonas depends heavily on a better understanding of the control and balancing between NHEJ and HR during different phases of the cell cycle. Progress can be expected from the completion of the Chlamydomonas genome sequence and its annotation which has already given insights into which repair and recombination systems are present in this organism. This information may be very useful in designing HR strategies that could make use of native Chlamydomonas recombination proteins thus avoiding problems with heterologous gene expression that have been encountered in previous experiments (our unpublished results). Considering the information that is presently available, we conclude that HR can be achieved in Chlamydomonas in principle, but that it is currently not yet possible to propose an optimal way for improving it to the point where Chlamydomonas genes for which no direct selection for a knockout or targeted replacement is available will become accessible for functional analysis. To achieve this goal a substantial amount of work remains to be done, but these efforts are fully justified in the light of the simplicity, versatility and beauty of this organism. Acknowledgements The work on this paper was supported by Slovak grants Vega 1/3243/06 and APVT 20-003-704 and a grant from the International Bureau of the BMBF (grant SVK 01/011)
References Baur M., Potrykus I. & Paszkowski J. 1990. Intermolecular homologous recombination in plants. Mol. Cell. Biol. 10: 492– 500. Belfort M. & Roberts R. J. 1997. Homing endonucleases: keeping the house in order. Nucleic Acids Res. 25: 3379–3388. Bilang R., Peterhans A., Bogucki A. & Paszkowski J. 1992. Single-stranded DNA as a recombination substrate in plants as assessed by stable and transient recombination assays. Mol. Cell. Biol. 12: 329–336. Brendel V., Brocchieri L., Sandler S.J., Clark A.J. & Karlin S. 1997. Evolutionary comparisons of RecA-like proteins across all major kingdoms of living organisms. J. Mol. Evol. 44: 528–541. Brozmanová J., Černáková L., Vlčková V., Duraj J. & Fridrichová I. 1991. The Escherichia coli recA gene increases resistance of the yeast Saccharomyes cerevisiae to ionizing and ultraviolet radiation. Mol. Gen. Genet. 227: 473–480. Brozmanová J., Vlčková V. & Chovanec M. 2004. How heterologously expressed Escherichia coli genes contribute to understanding DNA repair processes in Saccharomyes cerevisiae. Curr. Genet. 46: 317–330. Cerutti H., Johnson A.M., Boynton J.E. & Gillham N.W. 1995. Inhibition of chloroplast DNA recombination and repair by
945 dominant negative mutants of Escherichia coli RecA. Mol. Cell. Biol. 15: 3003–3011. Cerutti H., Johnson A.M., Gillham N.W. & Boynton J.E. 1997. Epigenetic silencing of a foreign gene in nuclear transformants of Chlamydomonas. Plant Cell 9: 925–45. Chandrasegaran S. & Smith J. 1999. Chimeric restriction enzymes: what is the next? Biol. Chem. 380: 841–848. Dudáš A., Marková E., Vlasáková D., Kolman A., Bartošová Z., Brozmanová J. & Chovanec M. 2003. The Escherichia coli RecA protein complements recombination defective phenotype of the Saccharomyes cerevisiae rad52 mutant cells. Yeast 20: 389–396. D¨ urrenberger F., Thompson A.J., Herrin D.L. & Rochaix J.D. 1996. Double strand break-induced recombination in Chlamydomonas reinhardtii chloroplasts. Nucleic Acids Research 24: 3323–3331. Epinat J.C., Arnould S., Chames P., Rochaix P., Desfontaines D., Puzin C., Patin A., Zanghellini A., Pˆ aques F. & Lacroix E. 2003. A novel engineered meganuclease induces homologous recombination in yeast and mammalian cells. Nucl. Acids. Res. 31: 2952–2962. Gumpel N.J., Rochaix J.D. & Purton S. 1994. Studies on homologous recombination in the green alga Chlamydomonas reinhardtii. Curr. Genet. 26: 438–442. Hastings P.J., McGill C., Shafer B. & Strathern J.N. 1993. Endsin vs. ends-out recombination in yeast. Genetics 135: 973– 980. Hinnen A., Hicks J.B. & Fink G.R. 1978. Transformation of yeast. Proc. Natl. Acad. Sci. USA 75: 1929–1933. Johnson R.D. & Jasin M. 2001. Double-strand-break-induced homologous recombination in mammalian cells. Biochem. Soc. Trans. 29: 196–201. Kandavelou K., Mani M., Durai S. & Chandrasegaran S. 2004. Nucleic Acids and Molecular biology (Springer-Verlag, Heidelberg, Germany) 14: 413–434. Kim Y.G., Cha J. & Chandrasegaran S. 1996. Hybrid restriction enzymes: zinc finger fusion to FokI cleavage domain. Proc. Natl. Acad. Sci. USA 93: 1156–1160. Klinner U. & Sch¨ afer B. 2004. Genetic aspects of targeted insertion mutagenesis in yeasts. FEMS Microbiol. Rev. 28: 201– 223. Leung W.Y., Malkova A. & Haber J.E. 1997. Gene targeting by linear duplex DNA frequently occurs by assimilation of a single strand that is subject to preferential mismatch correction. Proc. Natl. Acad. Sci. USA94: 6851–6856. Lloyd A.T. & Sharp P.M. 1993. Evolution of the recA gene and the molecular phylogeny of bacteria. J. Mol. Evol. 37: 399– 407. Mages W., Heinrich O., Treuner G., Vlcek D., Daubnerova I. & Slaninova M. 2007. Complementation of the Chlamydomonas reinhardtii arg2 (arg7-8) point mutation by recombination with a truncated nonfunctional ARG7 gene. Protist 158: 435–446. Morais M.A., Vlčková V., Fridrichová I., Slaninová M., Brozmanová J. & Henriques J.A.P. 1998. Effect of bacterial recA expression on DNA repair in the rad51 and rad52 mutants of Saccharomyes cerevisiae. Genet. Mol. Biol. 21: 3–9. Nelson J.A.E. & Lefebvre P.A. 1995. Targeted disruption of the NIT8 gene in Chlamydomonas reinhardtii. Mol. Cell Biol. 15: 5762–5769. Nomura N., Morinaga Y., Shirai N. & Sako Y. 2005. I-ApeI: a novel intron-encoded LAGLIDADG homing endonuclease from the archaeon, Aeropyrum pernix K1. Nucl. Acids. Res. 33: 1–8. Porteus M.H. & Carroll D. 2005. Gene targeting using zinc finger nucleases. Nature Biotechnology 23: 967–973. Rauth S., Song K.Y., Ayares D., Wallace L., Moore P.D. & Kucherlapati R. 1986. Transfection and homologous recombination involving single-stranded DNA substrates in mammalian cells and nuclear extracts. Proc. Natl. Acad. Sci. USA 83: 5587–91. Reiss B., Klemm M., Kosak H. & Schell J. 1996. RecA protein stimulates homologous recombination in plants. Proc. Natl. Acad. Sci. USA 93: 3094–3098.
946 Reiss B., Schubert I., Kopchen K., Wendeler E., Schell J. & Puchta H. 2000. RecA stimulates sister chromatid exchange and the fidelity of double-strand break repair, but not gene targeting, in plants transformed by Agrobacterium. Proc. Natl. Acad. Sci. USA 97: 3358–3363. Rich T., Allen R.L. & Wyllie A.H. 2000. Defying death after DNA damage. Nature 407: 777–783. Rochaix J.D. & Malnoe P. 1978. Anatomy of the chloroplast ribosomal DNA of Chlamydomonas reinhardtii. Cell 15: 661–670. Rouet F., Smih F. & Jasin M. 1994. Expression of a site-specific endonuclease stimulates homologous recombination in mammalian cells. Proc. Natl. Acad. Sci. USA 91: 6064–6068. Shaked H., Melamed-Bessudo C. & Levy A.A. 2005. High frequency gene targeting in Arabidopsis plants expressing the yeast RAD54 gene. Proc. Natl. Acad. Sci. USA 102: 12265– 12269. Shcherbakova O.G., Lanzov V.A., Ogawa H. & Filatov M.V. 2000. Overexpression of bacterial RecA protein stimulates homologous recombination in somatic mammalian cells. Mutat. Res. DNA Repair 459: 65–71. Simon J.R. & Moore P.D. 1987. Homologous recombination between single-stranded DNA and chromosomal genes in Saccharomyces cerevisiae. Mol. Cell. Biol. 7: 2329–2334. Slaninová M., Vlčková V., Brozmanová J., Morais M.A. & Henriques J.A.P. 1996. Biological consequences of E.coli RecA protein expression in the repair defective pso4-1 and rad51::URA3 mutants of S. cerevisiae after treatment with N-methyl-N’-nitro-N–nitrosoguanidine. Neoplasma 43: 315– 319.
M. Slaninová et al. Smith K. 2001. Theoretical mechanisms in targeted and random integration of transgene DNA. Reprod. Nutr. Dev. 41: 465– 485. Sodeinde O.A. & Kindle K.L. 1993. Homologous recombination in the nuclear genome of Chlamydomonas reinhardtii. Proc. Natl. Acad. Sci. USA 90: 9199–9203. Sonoda E., Hochegger H., Saberi A., Taniguchi Y. & Takeda S. 2006. Differential usage of non-homologous end-joining and homologous recombination in double strand break repair. DNA Repair 5: 1021–1029. Taccioli G.E., Rathbun G., Oltz E., Stamato T., Jeggo P.A. & Alt F.W. 1993. Impairment of V(D)J recombination in doublestrand break repair mutants. Science 260: 207–210. Tzfira T. & White C. 2005. Towards targeted mutagenesis and gene replacement in plants. Trends Biotechnol. 23: 567–569. Zorin B., Hegemann P. & Sizova I. 2005. Nuclear-gene targeting by using single-stranded DNA avoids illegitimate DNA integration in Chlamydomonas reinhardtii. Eukaryot. Cell 4: 1264–1272. Received September 1, 2007 Accepted March 17, 2008