Bacterial DNA repair genes and their eukaryotic homologues:

Recombinational repair is a well conserved DNA repair mechanism present in all living organisms. Repair by homologous recombination is generally accurate as it uses undamaged homologous DNA molecule as a repair template. In Escherichia coli homologous recombination repairs both the double-strand breaks and single-strand gaps in DNA. DNA double-strand breaks (DSB) can be induced upon exposure to exogenous sources such as ionizing radiation or endogenous DNA-damaging agents including reactive oxygen species (ROS) as well as during natural biological processes like conjugation. However, the bulk of double strand breaks are formed during replication fork collapse encountering an unrepaired single strand gap in DNA. Under such circumstances DNA replication on the damaged template can be resumed only if supported by homologous recombination. This functional cooperation of homologous recombination with replication machinery enables successful completion of genome duplication and faithful transmission of genetic material to a daughter cell. In eukaryotes, homologous recombination is also involved in essential biological processes such as preservation of genome integrity, DNA damage checkpoint activation, DNA damage repair, DNA replication, mating type switching, transposition, immune system development and meiosis. When unregulated, recombination can lead to genome instability and carcinogenesis.


INTRODUcTION
Genomic DNA is constantly subjected to damages.Some repair systems simply reverse DNA modifications, for instance Ada methyltrasferase (reviewed in Nieminuszczy & Grzesiuk, this issue), some of them need to excise modified nucleotide from DNA in a process called base excision repair (reviewed in Krwawicz, this issue) or remove whole DNA fragment containing a lesion, which is characteristic for nucleotide excision repair (NER) and methylation-directed mismatch repair (MMR) (reviewed in Maddukuri et al., this issue, and Arczewska & Kusmierek, this issue, respectively).
While mentioned repair systems have evolved to mend specific DNA modification and incorrectly paired bases, recombination is specialized in recognition and repair of DNA breaks.

2007
A. Nowosielska Homologous recombination consists of three stages which are common for prokaryotes and eukaryotes: presynapsis, where DSB or gap is formed and the resulting DNA end is being prepared for recombination; synapsis, where physical connection between the recombinogenic substrate and an intact homologous duplex DNA template is generated leading to formation of heteroduplex (hybrid) molecules; and postsynapsis, where DNA synthesis from the invading 3' end takes place followed by the resolution of junction intermediates.
In Escherichia coli there are two major mechanisms of homologous recombination: the RecB pathway, which fixes double-strand breaks, and the RecF pathway, which repairs daughter strand gaps (Fig. 1a and b, respectively).Both of them require RecA recombinase for homology recognition and DNA strand exchange.

RecA pROTeIN
The RecA protein has a DNA-dependent AT-Pase activity, both double-strand and single-strand DNA binding activity, homologous DNA pairing activity, and strand exchange activity.
In addition to its recombinational function, RecA is also important in the induction of the SOS response -global DNA repair and DNA damage tolerance (Little, 1991).The co-protease activity of the RecA filament formed on the 3' end of single-stranded DNA stimulates autocleavage of the LexA repressor, which inhibits expression of a variety of genes products involved in faithful DNA repair systems, among them some components of homologous recombination (ruvA, ruvB, recN and recA) (Friedberg et al., 1995).The co-protease activity of the RecA-ssDNA filament facilitates the autocatalytic cleavage of UmuD, a component of DNA polymerase V (Tang et al., 1999;Pham et al., 2002), and stimulates DNA synthesis on a damaged template by PolV (Schlacher et al., 2006).
The structure of RecA protein was elucidated in 1992 (Story et al., 1992).RecA protein first binds to the single-stranded or gapped DNA substrate, producing a right-handed helical filament containing one RecA monomer for every three nucleotides or base pairs of DNA.Next, a homologous duplex DNA is aligned to produce a nascent hybrid DNA.The RecA filament extension proceeds in the 5' to 3' direction along the DNA (Shan & Cox, 1996).Extension is rapid and occurs via cooperative addition of RecA monomers to the 3'-proximal end of the filament.RecA filament disassembly requires ATP hydrolysis and also occurs in the 5' to 3' direction.

pReSyNApSIS IN RecB pATHWAy
The key player in homologous recombination in E. coli is RecBCD, a 330 kDa protein, which processes blunt DNA double strand ends (DSE) and loads RecA protein on the 3' tail of single stranded DNA.Thanks to its complex architecture the RecBCD enzyme possess a nuclease and bipolar helicase activities.One of the components of RecBCD complex, RecB, is a nuclease and a 3'-5' helicase.Another one, RecD, also has helicase activity however it unwinds dsDNA in the 5' to 3' direction.Both helicases, have opposite polarities and travel in the same direction on both strand of the DNA duplex.RecD is the fast helicase acting on the 5' DNA end, while RecB is the slow helicase acting on the 3' end of DNA.The different processivities of these helicases lead to the formation of a long 5' strand and a short 3' strand with an expanding single stranded loop which has been observed in electron micrographs (Taylor & Smith, 1980).The resolution of the RecBCD crystal structure obtained by two laboratories (Dillingham et al., 2003;Taylor & Smith, 2003) indicates that the DNA duplex fed to the RecBCD holoenzyme is split across the RecC subunit and each single DNA strand is directed towards different helicase subunits.The 5' tail is fed to the RecD helicase and then onto the nuclease domain of RecB for digestion.The 3' tail is fed along a channel within the protein complex that emerges at the nuclease active site.As this strand goes directly to the nuclease active site it is digested with a higher processivity in comparison to the 5' tail which is located in a less favorable position.The RecC subunit is responsible for the DNA scanning and recognition of the Chi sequence: 5'-GCTGGT-GG-3' (a 'crossover hotspot instigator').Upon encountering a Chi site the final cleavage is introduced within a few bases from it and then RecC binds tightly to 3' end thus preventing its further digestion.From that moment the 5' tail is able to access the nuclease site more freely and is cleaved more frequently.These structural data elegantly explain why, after encountering Chi, the RecBCD complex pauses, its nuclease activity decreases and its polarity switches, the events for the first time visualized by Kowalczykowski's laboratory (Handa et al., 2005).In the next step RecBCD loads the RecA protein on the 3' end to initiate homologous pairing and strand exchange (Anderson et al., 1997;Arnold et al., 2000;Churchill et al., 2000).

pReSyNApSIS IN RecF pATHWAy
The RecF pathway, under natural circumstances, is responsible for single strand gap repair and is composed of the following proteins: RecF, RecO, RecR, The role of recombination in DNA repair and genome stability RecJ, RecQ and RecN.These single strand gaps (also called daughter strand gaps) in DNA are often formed when the replication fork encounters a non-coding lesion in a template DNA and reinitiates downstream from it.However when the RecBCD pathway is inactivated by mutation in one of the genes encoding its components, the recombinational defect is suppressed by mutations in sbcA, sbcB, or sbcCD genes, which activate the RecF pathway (Bidnenko et al., 1999;Kow-alczykowski et al., 1994;Kuzminov, 1999).In the RecF pathway, the 3' ssDNA is prepared by RecQ helicase and RecJ 5' to 3' exonuclease, while RecA loading on the 3' tail is achieved by RecF, RecO and RecR (Lovett & Kolodner, 1989;Kowalczykowski, 2000).
The RecF, RecO, and RecR proteins are involved in establishment of a RecA filament on DNA and modulate both its assembly and disassembly (Shan et al., 1997).E. coli RecF protein (40.5 kDa) A. Double strand breaks repair: 1.Double strand end (DSE) is generated in DNA and RecBCD binds to it; 2. RecBCD unwinds a duplex and degrades it; 3. Till encounters Chi site, then RecBCD switches from its exonuclease V activity to recombinase activity and loads RecA on the 3' single strand to produce RecA filament; 4. The RecA filament invades homologous DNA strand and Holliday junction is formed to which RuvABC resolvase or RecG helicase binds; 5.The Holliday junction is resolved, and resulting D-loop is acted upon by PriA, to allow replisome assembly.B. Single strand gap repair: 1. Single strand gap is formed in DNA; 2. RecQ and RecJ start to unwind and degrade single-stranded DNA region; 3. The RecFOR proteins bind to it and load RecA; 4. The RecA filament promotes strand-exchange and the Holliday junction is resolved by RuvABC or RecG proteins; 5. Nicked DNA strand is repaired.RecBCD (light green indented oval); the RecA filament (light blue ovals); PriA (yellow circle); RecJ (purple indented circle); RecQ (light yellow indented circle); RecF (green circle); RecO (pink circle); RecR (yellow circle).

2007
A. Nowosielska binds both dsDNA and ssDNA in vitro, and has a weak dsDNA-dependent ATPase activity in vivo (Sandler, 1996;Webb, 1999;Rangarajan et al., 2002).The RecF protein physically interacts with RecR (22 kDa) and the latter one also interacts with RecO (26 kDa).The RecOR proteins stimulate displacement of SSB proteins from DNA, thus facilitating RecA nucleation.The RecF protein seems to interfere with this function (Hobbs et al., 2007).However, RecF may position RecOR on specific DNA sites to initiate presynaptic complex formation (Sandler & Clark, 1994).RecFR proteins can also prevent the RecA filament from expanding beyond the single strand gap.It has been shown that RecF protein crystallized from Deinococcus radiodurans shares structural similarity with the eukaryotic Rad50 protein (Koroleva et al., 2007).
The RecQ protein is an ATP-dependent DNA helicase which translocates unidirectionally 3' to 5' along one strand of the duplex.In humans there are five homologs of RecQ helicases: RECQL, BLM, WRN, RECQ4 and RECQ5.The defects in eukaryotic RecQ helicases lead to premature aging and cancer predisposition, whereas a null mutation in E. coli recQ gene results in a 30-fold increase in illegitimate recombination (Hanada, 1997).RecJ is a 60 kDa protein with a 5'-3' exonucleolytic activity (Lovett & Kolodner, 1989).It has also been shown that RecJ and RecQ proteins process replication forks, before resumption of replication, thus preventing stalled replication forks from unnecessary recombination event (Hanawalt & Courcelle, 2001).The enzymatic activities of RecQ and RecJ have a key role in enabling the RecF pathway to act on DSBs in the absence of an active RecB pathway (Amundsen, 2003).
Another E. coli protein belonging to the RecF pathway is encoded by the recN gene.The RecN protein is a member of the structural maintenance of chromosomes (SMC) family (Rostas et al., 1987).RecN is strongly induced during the SOS response and has been implicated in DNA double strand break repair (Meddows et al., 2005) Its concentration in a cell is tightly regulated, as the RecN protein has a short half-life and its degradation is dependent on the cytoplasmic protease ClpXP (Nagashima et al., 2006).

RegUlATION OF RecA FIlAmeNT FORmATION
In addition to the RecFOR and RecBCD complexes mentioned above, many other proteins control the formation of the RecA filament in E. coli.The major competitors of the RecA filament formation are single-stranded DNA binding (SSB) proteins, which as their name implies, bind to single-stranded DNA (ssDNA) to protect it from degradation.SSB proteins are essential to DNA metabolism in all organisms.In E. coli, the ssb gene is indispensable for cell viability (Meyer et al., 1990;Lohman et al., 1994;Curth et al., 1996).The eukaryotic counterpart of SSB is the heterotrimeric replication protein A (RPA) (Brill & Stillman, 1991;Ogawa et al., 1993).The SSB as well as RPA proteins inhibit RecA filament formation if they coat DNA before RecA binding (Lavery et al., 1990;Umezu et al., 1994;Shan et al., 1997;Shinohara & Ogawa, 1998;New et al., 1998;Bork et al., 2001).The inhibitory action of SSB proteins on RecA nucleation is overcome in the bacterial cell by the Re-cOR protein complex (Bork et al., 2001;Morimatsu et al., 2003;Hobbs et al., 2007).On the other hand, SSB protein binding to DNA eliminates secondary structures to which the RecA protein would not be able to bind efficiently.Therefore, SSB binding to DNA enables RecA to form a contiguous RecA filament on the DNA (Kowalczykowski et al., 1987).

pOSTSyNApTIc STAge OF RecB AND RecF pATHWAyS
The postsynaptic phase of homologous recombination requires housekeeping enzymes.In the last phase of the RecF pathway, gyrase and topoisomerase I are needed to relieve positive and negative The role of recombination in DNA repair and genome stability DNA supercoiling, generated during RecA filament interaction with the homologous double-stranded template (Cunningham et al., 1981;Casuto, 1984).Additionally helicase II (UvrD), DNA polymerase I (PolI), replicative DNA helicase (DnaB), a catalytic subunit of PolIII (PolC/DnaE) and ligase are needed to enable the filling and closing of single-strand gaps.Genetic studies showed that completion of double-strand break repair also requires DNA gyrase, DNA PolI and DNA ligase (reviewed in Kuzminov, 1999).

mIgRATION AND ReSOlUTION OF BRANcHeD DNA STRUcTUReS
The RuvAB protein complex is a molecular motor that can branch-migrate Holliday junctions, which result in extension of heteroduplex DNA between recombining DNA molecules (West, 2003).Similar reactions are catalyzed by RecG helicase (Lloyd & Sharples, 1993;McGlynn & Lloyd, 1999).The junction formed during strand invasion must be eventually resolved to restore linear duplexes.Depending on the configuration of Holliday junction breaks introduced by RuvC protein (a Holliday junction resolvase), the resulting DNA molecule may be identical with the parental one or changed if a crossover took place.
The RecA filament-promoted strand exchange generates a three strand junction such as a D-loop and the four strand junctions called Holliday junctions (Liu & West, 2004).RuvA (22 kDa) recognizes the Holliday junction structure and binds to it as a tetramer (Tsaneva et al., 1992).RuvA, together with the RuvB protein (37 kDa), promotes branch migration of Holliday junctions.The RuvB molecular motor is an intrinsic ATP-dependent DNA helicase with a hexameric ring structure (West, 1997).It has low intrinsic affinity to DNA, however, a direct interaction with RuvA targets RuvB to the junction.RuvA tetramers bind to the junction to open it into a square planar conformation while two RuvB rings bind to the opposite arms of the junction.RuvB rings pull duplex DNA through their holes, causing the junction to branch-migrate (Him & West, 1995) RuvC is a 19 kDa resolvase which binds to a Holliday junction as a dimer and introduces nicks on two DNA strands of the same polarity, at a degenerate sequence 5'-(A/T)TT ↓ (GC)-3' (Eggleston & West, 2000).The nicks introduced by RuvC are sealed by ligase.Another ATP-dependent DNA helicase which binds Holliday junctions and translocates them is RecG helicase.However, the RecG activity is much weaker in comparison to the RuvAB complex.While RuvAB tranlocates ssDNA in the 5' to 3' direction, RecG translocates it in the 3' to 5' direction.
The E. coli RuvC enzyme has high specificity for cleavage of Holliday junctions, but mutants lacking RuvC do not show a strong deficiency in conjugational recombination unless an additional mutation is present in recG (Lloyd, 1991;Benson & West, 1994).Another protein engaged in Holliday junction resolution in E. coli, RusA, is encoded by a defective prophage.It is a DNA structure-specific endonuclease which introduces symmetrically paired incisions 5' to CC (Sharples et al., 2002).
While in E. coli migration of the Holliday junction takes place with the help of the RuvABC and RecG proteins, in eukaryotes it is achieved by the Rad54 protein.Rad54 is a member of the Swi2/ Snf2 family of SF2 helicases (Pazin & Kadonaga, 1997), which translocates on dsDNA but it does not display a strand displacement activity typical for a helicase.Rad54 remodels DNA structure, chromatin structure and Rad51-dsDNA complexes (Heyer et al., 2006).
There have been intensive studies on identifying the eukaryotic counterparts of Holliday-junction endonucleases.Recently, it has been shown that Rad51 paralogs Rad51C and Xrcc3 participate in the Holliday junction resolution (Liu & West, 2004).Other studies in yeast and human cells have shown participation of a protein complex containing Mus81 with its partner MMS4 or Eme1 in resolving Holliday structure during meiosis (Boddy et al., 2001;Chen et al., 2001;Kaliraman et al., 2001;Constantinou et al., 2002;Ciccia et al., 2003;Gaillard et al., 2003).
Alternatively, Holliday junction in eukaryotes can be separated by the combined action of RecQlike helicases and a topoisomerase III (Heyer et al., 2003).

RepAIR OF STAlleD ReplIcATION FORKS
Replication blocks are quite frequent in the every living cell's life.Their causes range from the malfunction of the replicative machinery to the damage to DNA by the UV treatment.If replication forks encounter a lesion which prevents their progression, the cell employs restart systems in order to ensure replication completion.In E. coli stalled replication forks can be processed by DNA helicases, nucleases and recombinational proteins.Once rep-A.Nowosielska lication fork is reconstituted, the replisome can be reloaded, and the lesion, which imposed replication block, must be removed (Heller & Marians, 2006).In E. coli three proteins: PriA, Rep and UvrD are 3'-5' helicases involved in the restart of stalled replication forks (Heller & Marians, 2005;Flores et al., 2005).
The PriA protein is highly conserved in bacteria.Inactivation of this protein, which is important for replication restart, leads to reduced viability, slow growth, sensitivity to rich medium and induction of the SOS response.PriA contains a crucial 3'-termini binding pocket responsible for high affinity binding to D-loops and stalled fork structures that contain a nascent leading strand with the 3'-OH end near to the fork junction (Mizukoshi et al., 2003).PriA interaction with DNA induces binding of the PriB protein which stabilizes the PriA-DNA interaction and facilitates recruitment of DnaT (Liu & Marians, 1999).This multiprotein-DNA complex is responsible for recognizing the correct DNA structure and helps to remove recombination and other proteins such as RecA or SSB associated with the processed DNA structure, and unwinds duplex DNA to load the DnaB replicative helicase.Then, DnaB is loaded into the complex with a help of the DnaC protein, which does not retain in the complex.Next, DnaB interacts with PolIII holoenzyme and the DnaG primase to reconstitute replisome (Sandler, 2001).PriAdirected replication restart targets D loops, R loops and stalled replication forks with nascent leading strand.PriC-directed replisome loading, which depends on priC and rep genes products, is limited.It targets only a subset of stalled replication forks with a gap generated when the nascent leading strand encounters blocking lesion while the lagging strand continues to unwind (Heller & Marians, 2006).
Another DNA helicase, Rep, possessing a 3' to 5' translocation activity is required for the optimal progression of replication forks due to its ability to remove proteins in front of the replication forks.A lack of this protein also contributes to frequent replication fork stalling (Heller & Marians, 2005).Rep also takes part in the reconstitution of stalled replication forks by unwinding nascent lagging-strand DNA in a similar way to PriA.Rep is also proposed to act in the PriC pathway as its helicase activity is stimulated by PriC protein, therefore it is suggested that Rep is recruited to the stalled replication forks which are the substrate for the PriC helicase (Sandler, 2000;Heller & Marians, 2005).
The E. coli UvrD protein is a 3' to 5' helicase, which prefers to unwind DNA with a 3′ single-stranded overhang (Matson, 1986).UvrD is also able to unwind DNA from a nicked substrate and a blunt end (Runyon et al., 1990).The UvrD protein is a component of nucleotide excision repair (NER) and methyl-directed mismatch repair (Lahue et al., 1989;Dao & Modrich 1998).In addition, UvrD takes part in homologous recombination initiated by Rec-FOR in recBC sbcBC mutants (Mendonca et al., 1993).At blocked replication forks, UvrD dismantles the RecA filament, thus allowing replication fork reversal and proper replication restart (Flores et al., 2004).The Srs2 protein found in yeast is a homolog of E. coli Rep and UvrD helicases.Srs2 is an ATPdependent DNA helicase, that is required for DNA damage checkpoint responses and that modulates the efficiency of homologous recombination and (Van Komen, 2003).

INvOlvemeNT OF RecOmBINATION pROTeINS IN ReplIcATION FORK ReveRSAl (RFR)
Extensive studies in Benedicte Michel's laboratory showed that certain replication mutants: rep coding helicase, dnaE, encoding catalytic subunit of PolIII, dnaN, coding β clamp, and holD, which encodes ψ, a component of γ complex clamp loader, suffer frequent replication fork arrest (Grompone et al., 2002;Baharoglu et al., 2006).This group showed that these mutants were synthetically lethal with recBCD inactivation but not with recA deficiency, excluding dnaN mutant.Rep lethality was suppressed by the additional inactivation of ruvAB.These results contributed to the model of replication fork reversal in which nascent lagging and leading strand ends anneal creating double strand end adjacent to a Holliday structure.This structure is recognized by RuvAB proteins which bind to it.In RecBCD + cells RecBCD holoenzyme binds to DSE and degrades it up to the Holliday junction stabilized by RuvAB proteins.Next it displaces RuvAB complex leaving reconstituted RF, to which the replisome can be loaded by the PriA pathway.RecBCD enzyme can also encounter a Chi site during DSE degradation and initiate strand invasion and exchange, which also results in replication fork recovery.In the absence of RecBCD proteins a Holliday junction formed by a reversed fork is processed by RuvAB proteins which results in its breakage (Baharoglu et al., 2006).
This group also showed that the initiation of replication fork reveral can vary among studied replication mutants.
Studies in Michel's laboratory showed that, in dnaB mutants, defective in a replicative helicase, the initial stage of RFR requires RecA protein.They have proposed a model in which RecA alone is needed to direct RFR in this mutant (Seigneur et al., 2000).Further research revealed that in other temperature sensitive replication mutants, dnaE and holD, RFR was dependent on RuvAB, which probably by itself catalyzes the conversion of replication fork to Holliday junction.Only partial dependence on RuvAB was The role of recombination in DNA repair and genome stability observed for the rep mutant, which was defective in another DNA helicase.Studies on temperature sensitive mutants defective in catalytic subunit of PolIII (dnaE) and clamp (dnaN) showed a UvrD requirement for RFR in the presence of the RecF pathway.These results suggest a model in which RecFORJQ proteins promote RecA-binding at forks thus preventing RFR.It was concluded that fork processing by these proteins result in a lethal structure and that the deleterious action of RecQ, RecJ, RecF, RecO and RecR proteins in dnaE and dnaN mutants is counteracted by UvrD function (Flores et al., 2005).
Once the replication fork is restored it is targeted for replisome reassembly.
Depending on the way the replication fork is processed the replisome loading on reconstituted replication fork is coordinated by the PriA or PriCdependent pathways (Sandler et al., 2001).

Cisplatin
(cis-diamminedichlorideplatinum (II)) is a popular chemotherapeutic drug widely used in cancer treatment, particularly effective against testicular tumors (Einhorn, 2002).Cisplatin binds to the N7 atom of purine bases in DNA to form predominantly 1,2-d(GpG), 1,2-d(ApG) and 1,3-d(GpNpG) intrastrand cross-links, and a small percentage of interstrand crosslinks (Eastman, 1983;Fichtinger-Schepman, 1985), suggesting that cisplatin intrastrand crosslinks between adjacent purines are the biologically important adducts since they efficiently block progression of DNA polymerases in vitro and in vivo (Pinto & Lippard, 1985).Studies on the cytotoxic effect of this drug in E. coli showed that homologous recombination contributes to cisplatin resistance (Zdravesky et al., 2000;Nowosielska & Marinus, 2004).Subsequent studies on dam strains, deficient in Dam methyltransferase, revealed that the extensive number of DNA double strand breaks is generated in response to cisplatin.This observation led to conclusion, that the majority of cisplatin-triggered breaks are formed as a result of replication fork collapse on gaps introduced by mismatch repair processing platinated DNA (Nowosielska et al., 2005;Nowosielska & Marinus, 2007).
It has been shown recently that homologous recombination protects also from cytotoxicity induced by methylating agents (Nowosielska et al., 2006).These results suggested that single-strand gaps and DSBs are produced by the replication fork collapse at blocking lesions or at single-strand nicks produced by AP-endonucleases (Nowosielska et al., 2006).However, in contrast to cisplatin, only a small fraction of methylation-induced DSBs is generated during replication (Nowosiel-ska & Marinus, 2007) while the majority of them resulted from the interference of two repair systems processing methylated DNA (Nowosielska & Marinus, 2007).

HUmAN DISeASeS cAUSeD By mUTATIONS IN RecOmBINATIONAl RepAIR
The complexity and significance of homologous recombination in preservation of genome integrity can be better realized while studying human disorders caused by its malfunction.Increased or decreased frequencies of HR have been found in cancer cells and cancer-prone hereditary human disorders characterized by mutations in genes playing a role in HR, such as ATM, BRCA, BLM, and WRN genes.
Ataxia telangiectasia mutated (ATM) is a serine/threonine-specific protein kinase that is recruited and activated by DNA double-strand breaks.ATM kinase deficiency causes ataxia telangiectasia (AT), a syndrome characterized by increased sensitivity to ionizing radiation, cerebellar degeneration, oculocutaneous telangiectasia, immunodeficiency, aging and increased risk of cancers such as lymphoma and leukemia (Frappart & McKinnon, 2006).
Mutations in BRCA1 and BRCA2 were identified as they predisposing to breast cancer (Miki et al., 1994;Wooster et al., 1995).Brca2 controls DNA binding by Rad51.Mutations in BRCA2 cause Fanconi anemia (Mathew, 2006).This genetic disorder increases susceptibility to several types of leukemia, and to cancers affecting ovaries, prostate and pancreas.It has been shown recently that the Brca2 protein interacts with the Dss1 protein, a 70-amino-acid protein that has been associated with the developmental disorder split hand/split foot malformation (Yang et al., 2002).The Dss1 protein is also involved in recombinational repair, and mutation in the DSS1 gene has the same effect as defects in the BRCA2 homolog (Kojic et al., 2003).
NBS1 mutations cause Nijmegen breakage syndrome, characterized by microcephaly, immunodeficiency and high incidence of cancer.The NBS1 gene product associates in vivo with Mre11 and Rad50 proteins to form the Mre11-Rad50-Nbs1 complex which plays pivotal roles in eukaryotic DNA double strand break repair, meiotic recombination and telomere maintenance (Digweed & Sperling, 2004).
Mutations in three human homologs of RecQ helicase: BLM, WRN and RECQ4 contribute to genetic diseases.Defects in BLM lead to Bloom's syndrome (BS), in WRN to Werner syndrome (WS), and mutations in RECQ4 lead to Rothmund-Thompson syndrome (RTS), RAPALIDINO, and Baller-Gerold syndrome (BGS).BS, WRN and RTS syndromes cause chromosomal instability, a predisposition to cancer and in the case of RTS, premature aging.People with RTS displays growth deficiency, photosensitivity with poikilodermatous skin changes.RAPALIDINO syndrome is an autosomal recessive disorder characterized by radial hypoplasia/aplasia, patellae hypoplasia/aplasia, and cleft or highly arched palate, little size, limb malformation, diarrhoea and dislocated joints, nose slender and normal intelligence.BGS is another recessive autosomal condition characterized by radial aplasia/hypoplasia and craniosynostosis (Sharma et al., 2006;Hanada & Hickson, 2007).RecQ has been also reported in the maintenance of telomeres.Both the BLM and WRN proteins have been shown to interact with TRF2 telomere-binding protein (Opresco et al., 2002).The yeast homolog of RecQ, Sgs1, participates in a Rad52-dependent recombinational pathway of telomere maintenance in telomerasenegative mutants (Azam et al., 2006).The Rad51D protein, which is a RAD51 paralog, was shown to associate with telomeres and prevent their dysfunction (Tarsounas, 2004) and Rad54, which belong to the chromatin remodeling family, was found to act at telomeres.Its deficiency resulted in telomere shortening and telomere fusions (Jaco, 2003).To explain role of homologous recombination in telomere protection and elongation of telomeres two mechanisms involving inter-and intra-telomere homologous recombination were recently proposed (Tarsounas & West, 2007).

Figure 1 .
Figure 1.Homologous recombination. A. Double strand breaks repair: 1.Double strand end (DSE) is generated in DNA and RecBCD binds to it; 2. RecBCD unwinds a duplex and degrades it; 3. Till encounters Chi site, then RecBCD switches from its exonuclease V activity to recombinase activity and loads RecA on the 3' single strand to produce RecA filament; 4. The RecA filament invades homologous DNA strand and Holliday junction is formed to which RuvABC resolvase or RecG helicase binds; 5.The Holliday junction is resolved, and resulting D-loop is acted upon by PriA, to allow replisome assembly.B. Single strand gap repair: 1. Single strand gap is formed in DNA; 2. RecQ and RecJ start to unwind and degrade single-stranded DNA region; 3. The RecFOR proteins bind to it and load RecA; 4. The RecA filament promotes strand-exchange and the Holliday junction is resolved by RuvABC or RecG proteins; 5. Nicked DNA strand is repaired.RecBCD (light green indented oval); the RecA filament (light blue ovals); PriA (yellow circle); RecJ (purple indented circle); RecQ (light yellow indented circle); RecF (green circle); RecO (pink circle); RecR (yellow circle).