Repair of DNA Double-strand Breaks and Homologous Recombination

Repair of DNA double-strand breaks and homologous recombination 16.1 an overview of DNA double-strand break repair and homologous recombination Double-strand breaks are repaired by two different mechanisms Double-strand breaks in DNA, unlike other types of DNA damage, must be repaired in the absence of an undamaged template strand that could otherwise have been used to direct the repair process In the first of these, non-homologous end joining (NHEJ), the broken ends, often damaged by limited degradation, are directly rejoined The second mechanism of repair, homology-directed repair involves (as its name implies) sequence-specific pairing between the damaged DNA and intact homologous sequences in another chromosome, which serve as the template for repair Although NHEJ has the advantage that it does not require a template and can therefore occur during G0, G1, and early S phases of the cell cycle, when no sister chromed is available, it has the disadvantage that it can be error-prone and cause mutations Nucleotides at the breakage site are often lost through degradation by cellular nucleases before repair is initiated, and further loss frequently happens because of trimming that occurs during the repair process itself In homology-directed repair, sequence-specific pairing between DNA strands from a broken DNA duplex and an intact homologous DNA duplex allows the use of the intact duplex as a template for repair of the broken DNA In eukaryotes, the majority of homology-directed repair occurs during late S phase or G2 phase of the cell cycle only, when closely apposed sister chromatids are available to serve as a template Homology-directed repair generally results in repair without mutation or loss of sequences because it is most often involves copying intact information from an undamaged identical sister chromatid More rarely, a homologous chromosome may provide the template, instead of a sister chromed, but this process may not restore the original sequence to the damaged chromosome, because the sequences of the two homologous chromosomes may not be identical Activation of the KU/DNA-PKcs pathway leads to NHEJ, while activation of the MRN/ATM pathway often leads to homology-directed repair The double-strand break repair machineries also mediate homologous recombination and underlie mechanisms to generate diversity in the immune system The homology-directed repair machinery has been adapted to mediate the DNA strand pairing interactions between homologous DNA molecules that underlie homologous recombination The conserved molecular components and mechanisms of the homology-directed repair and homologous recombination pathways highlight the close relationship between these reactions In eukaryotes, homologous recombination is intimately involved in the production of gametes during meiosis and also occurs occasionally during mitosis Homologous recombination also occurs in bacteria, often during conjugation, when DNA is actively transferred from one cell to another and can be integrated into the genome of the acceptor cell via homologous recombination Incorporation of these exogenous DNAs into the genome by homologous recombination is relatively common and is a significant source of the horizontal gene transfer that accounts for much bacterial diversification in the course of evolution Homologous recombination can, however, occasionally occur between two stretches of homologous DNA located anywhere in the genome, potentially leading to deletions or translocations, both of which can have grave consequences for the cell In addition to rejoining the ends of DNA exposed by DNA damage, the NHEJ pathway also provides the means for the DNA joining steps we see during the programmed DNA rearrangement of specialized cells of the immune system 16.2 Double-strand break repair by NHEJ The NHEJ pathway is often mutagenic The exact nature of the repair reaction is highly influenced by the fact that different types of double-strand breaks result in different end structures In some cases, the DNA ends have simple overhangs that can be simply rejoined by ligation, with little or no sequence loss. In other cases, particularly when radiation-mediated breaks occur, the ends are rejoined only after considerable processing of the ends at the break In one pathway, the ends at the break undergo only limited processing with perhaps the loss of a few base pairs, and two blond ends are ligated together. in the other pathway, resection reveals short complementary regions that occur by chance anywhere in the tails NHEJ can thus be mutagenic and may be deleterious if, for example, it occurs in a region of functional DNA sequence such that important coding information is lost The NHEJ pathway depends on conserved end-binding and ligating proteins The NHEJ process in bacteria and eukaryotes requires a protein, KU, a dimer of KU70 and KU80 subunits, which binds to broken DNA ends and recruits the kinases, nucleases, polymerases, phosphatases, and ligases that mediate the joining reaction First, the KU complexes slide onto each end of the DNA double-strand break and act as a scaffold for the recruitment of DNA repair enzymes, in particular A protein kinase subunit, DNA-PKcs, to form DNA-PK Second, DNA-PKcs is activated by binding to KU, can autophosphorylate, and also phosphorylates and controls other important repair proteins Third, other DNA end-processing enzymes, including DNA polymerases and/or exonucleases and DNA kinases and/or phosphatases, further process the ends before ligation 16.3 Homology-directed repair of double-strand breaks Accurate repair of double-strand DNA breaks involves pairing of homologous duplexes The reactions that underlie homology-directed double-strand break repair occur within four main phases: Generation of single-stranded DNA at the site of a double-strand break Pairing by one of the single-stranded ends with the intact duplex by invasion of the duplex to form heteroduplex DNA Repair of the damaged duplex by DNA synthesis using strands from the undamaged duplex as a template Separation of the two duplexes The formation of the heteroduplex is known as synapsis, and the steps that precede or follow it are referred to as presynaptic and post-synaptic Homology-directed repair requires long single-stranded DNA regions The first step in homology-directed repair is the generation of singlestranded 3' ends at the double-strand break by specialized helicases and nucleases by degradation of the 5' ends of the broken molecule Homology-directed repair requires long regions of pomology at the break for base-pairing to occur. Consequently, the single-stranded tails at the break are usually at least 50 nucleotides long These long tails are needed to ensure accurate pairing with the template DNA in another chromosome, and thus to ensure error-free repair Formation of a d-loop containing heteroduplex DNA provides a template for DNA repair synthesis In the second step in homology-directed repair, strand invasion, the single-stranded DNA from one end of the damaged duplex locates and pairs with its complementary strand in the related unbroken duplex to form a DNA heteroduplex -a region in which a single strand from one DNA molecule is base-paired with a single strand from a different DNA molecule Homology-directed repair can proceed via synthesis-dependent strand annealing Once specific heteroduplex pairing has occurred between the invading strand and the undamaged duplex to form the dloop, homology-directed repair can proceed in any several ways Repair by this pathway [synthesis-dependent strand annealing] begins with DNA synthesis that is initiated in the dloop, using the 3'-OH end of the invading strand as a primer and the undamaged strand as a template When it is long enough (namely, past the site of the double-strand break), the newly synthesized strand pairs with the complementary strand of the damaged duplex on the other side of the break Notably, the DNA that now bridges the double-strand gap is newly synthesized DNA, generated from the undamaged template duplex; there is no permanent formation of heteroduplex DNA between the damaged duplex and the template duplex As a result, in homology-dependent repair by SDSA, no DNA strand from the undamaged duplex is joined to, or incorporated into the repaired duplex, nor i's there any change to either duplex outside of the repaired strands A notable feature of SDSA is that the repaired duplex acquires a copy of the sequence of the template duplex. If the template duplex is the sister chromatid, there is no change in the genetic information in the repaired duplex 16.4 generation of single-stranded DNA by helicases and nucleases The RecBCD complex in bacteria generates the single-stranded DNA needed for strand invasion This step [presynaptic step] depends both on DNA helicases, which separates the two strands of DNA by unwinding, and on nucleases that degrade one of the DNA strands at the broken DNA, namely the strands with the 5' ends at the break In E. coli, there are two distinct presynaptic pathways by which 3'-ended single strands are generated: the RecBCD and the recF pathways, named after their constituent proteins This complex [RecBCD pathway], which has combined helicase and nuclease activities, attaches at a double-strand break and translates along the DNA, its helicase activity separating the two strands and its nuclease activity cleaving the unwound strands into small fragments When the RecBCD complex arrives at a chi sequence, the nuclease activity is reduced, and the enzyme now preferentially degrades the 5’-end strand, leaving a duplex DNA with a single-stranded 3’ tan containing the chi sequence RecBCD also actively facilitates the loading of the strand exchange recombinase RecA protein onto the singlestranded 3’-end DNA generated at a chi site RecA then promotes pairing between the single-stranded DNA and a homologous DNA duplex that will provide the template for repair, preferentially recognizing the single-stranded 3'-DNA end that participates in strand invasion to form a d-loop In addition to RecBCD, the other presynaptic system in E. coli 1s known as the RecF pathway, which includes a helicase called RecQ and a nuclease called RecJ that, together, generate single-stranded DNA This pathway [RecF pathway] appears to be used mainly in the repair of replication forks that have stalled as a result of single-strand lesions in the DNA The MRN complex and nucleases in eukaryotes generate the single-stranded DNA needed for strand invasion In eukaryotes, the generation of the 3' single-stranded DNA tails that are the substrates for strand invasion is carried out by several proteins whose activity is highly regulated by inputs from several other cellular processes, including the cell cycle Activation of 5' end resection to generate the long 3' ends, which are the substrate for strand pairing, is the key step that commits a double strand break to homology-directed repair or homologous recombination, rather than NHEJ Notably, the first step of resection requires a phosphorylated activator protein, Sae2 in yeast or its vertebrate homolog CtIP, which activates the nuclease activity of MRN that arrives early at the damage site Because the cell-cycle regulatory CDK1 kinase that phosphorylates Sae2/CtIP is only active in S and G2 phases of the cell cycle, homology-dependent repair thus occurs only in these phases when a sister chromatid is available as the template for repair In the first step of repair, the phosphorylated Sae2/CtIP-MRN nuclease introduces nicks near the end (within about 270 bp) of the 5' strand At the double-strand break site 16.5 The mechanism of DNA strand pairing and exchange DNA strand pairing is promoted by specialized recombinases Synapsis involves sequence-specific recognition events between single-stranded DNA and DNA duplexes, and the exchange of the single strand with one of the strands of the duplex to form heteroduplex DNA Bacterial cells can live without RecA and yeast cells can live without Rad51, but they grow poorly and are highly sensitive to DNA-damaging agents. Mammals without Rad51 die in utero These growth defects likely reflect the fact that homology-dependent repair reactions are required to reassemble broken replication forks resulting from DNA damage The key step in the strand-exchange process is the identification of homology between an invading single strand and a DNA duplex. Strand exchange begins with the loading of the RecA monomers onto single-stranded DNA to form the presynaptic filament This process requires ATP, which binds at the interface between the monomers of RecA and drives a conformational change in the RecA molecules that is essential for DNA binding and assembling RecA filament DNA binding, in turn, promotes the binding of further RecA monomers; therefore, the assembly of the presynaptic filament occurs through cooperative binding between RecA, ATP, and DNA Binding of the RecA nucleoprotein filament to the DNA in the invaded intact duplex distorts the duplex DNA molecule so that stretches of extended and untwisted DNA alternate with three-nucleotide segments that assume the conformation of normal b-DNA Untwisting of the extended duplex DNA is thought to enable the interrogation of the duplex for DNA sequence homology by the invading single strand, while the three-nucleotide segments of b-DNA promote correct base-pairing between the invading and the recipient strands, since incorrect pairing would disrupt the stacking interactions of the DNA Thus, binding is strongly disfavored by mismatched bases, which is thought to account for the requirement for very high sequence similarity between the interacting DNA molecules, and hence the accuracy of repair Once homologous pairing is established between the single strand of the nucleofilament and the complementary strand of the duplex DNA, strand exchange follows and heteroduplex DNA is formed Accessory proteins assist the strand-exchange recombinases Although RecA and Rad51/RAD51 recombinases can execute homologous pairing and strand exchange in the absence of other proteins in vitro, many other proteins are involved in recombination in vivo When single-stranded DNA persists, it also recruits accessory proteins that bind to the junction of single-stranded and double-stranded DNA. These proteins then catalyze the loading of the recombinase and the displacement of the single-strand binding protein Moreover, BRCA1 is a ubiquitin ligase and is also thought to be important in the regulated recruitment of repair proteins to sites of DNA damage. This recruitment is driven by the addition of ubiquitin tags to molecules at the damage site that are recognized by components of repair pathways BRCA1 also influences repair by directly interacting with CtIP, the enzyme involved in resection of 5' ends to make their cognate 3' single-ended tails, while BRCA2 interacts directly with RAD51 16.6 Gene conversion through homology-directed repair Homology-directed repair at a double-strand break can change gene sequence by replacing one version of a gene with another For breaks that occur in G0, G1, and early S phases of the cell cycle, when sister chromatids are not available, a homologous chromosome may serve as the template for repair Sister chromatids are exact replicas of each other, and if one serves as a template for repair of the other, the damaged chromatid will, in effect, be restored to its pre-damage sequence This process is called gene conversion - a sequence on one homologous chromosome is converted into the sequence of the other, which itself remains unchanged Note also that this exchange of sequence is specifically targeted to a particular region that is at the site of the double-strand break that initiates homology-directed repair However, note that the relationship and identity of the DNA segments that flank the double-strand break do not change - the A and C versions of these genes are still on one parental duplex, and the a and c (lowercase) versions on the other In addition to its role in repairing damaged chromosomes, this doublestrand break-directed repair mechanism has been exploited in some organisms to carry out specific changes in gene structure and expression and by researchers to carry out directed genome editing in a wide variety of organisms Yeast mating-type switching is the best understood example of directed gene conversion In some strains of S. cerevisiae, the haploid cells can switch mating type by a process involving directed gene conversion by homology-directed repair The transfer of DNA sequence information from the silent sites to the MAT expression site occurs by directed gene conversion, which 1s mediated by the homology-directed double-strand break repair pathway The HML-alpha and HMRA silent loci serve alternately as donors, so that the expression of A or alpha cell-type information, and thus the mating type of the cell, alternates in a cell lineage Targeted genome editing is achieved by researcher-directed DNA double-strand breaks Genome editing has recently been revolutionized by the development of systems that use a synthetic RNA to program a double-strand nuclease to act at particular positions in the genome through the selectivity of base-pairing, which targets DNA repair These advances in genome editing exploit adaptive immune systems found widely among bacteria and archaea that can store DNA sequence information from invading foreign DNA plasmids or phages in CRISPR (clustered regularly interspaced short palindromic repeats) sequences using associated Cas gene products Transcription of these stored sequences then generates an RNA copy that programs Cas nucleases via base-pairing interactions to cleave and destroy complementary DNA or RNA from these plasmids or pages Once cleavage of the target DNA by a double-strand break occurs, the break may be repaired by end joining by NHEJ, which is often mutagenic, or by homology-directed repair that uses a co-provided DNA as a template for repair 16.7 homologous recombination Large segments of DNA can be exchanged between two DNA molecules Although homology-directed repair of a double-strand break can result in local gene conversion, it does not change the relative positions of the DNA segments flanking the initiating double-strand break This process is called homologous recombination and can result in the reciprocal exchange of large segments of DNA between homologous diva duplexes Homologous recombination in meiosis plays a key role in the chromosome segregation process that culminates in the production of haploid gametes; it is important both for recognition between homologous chromosomes and for the exchange of genetic information between them that results in new combinations of genetic information that underlie genera variation in humans and most other eukaryotes Homologous recombination can occur between any DNA molecules with extensive regions of identical sequence (as with two sister chromatids) or very similar sequence (as with two homologs); no specialized recombination sequences are required Such exchanges generate diversity by reshuffling segments of the genome to make new combinations of existing characteristics, rather than through the introduction or new random mutations or by the translocation of discrete DNA segments Homologous recombination is initiated by a DNA double-strand break In meiotic recombination, this break is made by the programmed action of specialized nucleases. The break then undergoes reactions closely related to those underlying homology-directed repair, beginning with the formation of a d-loop between one of the broken duplexes and a homologous duplex The broken DNA ends are degraded to generate single-stranded 3’ tails (step 1) The resulting nucleoprotein filament then invades the intact duplex (step 2) The next step is DNA synthesis, with the invading 3' end serving as a primer and its complementary strand in the participating homologous duplex as a template (step 3) Subsequent events depart from those of SDSA, however; instead of unwinding from the template strand, the newly synthesized strand remains paired to the template in the homologous duplex, and the d-loop is extended by continued DNA synthesis This DNA synthesis displaces more of the non-template strand in the recipient duplex, and the displaced strand pairs with the complementary strand from the broken duplex (step 4) In this way, the remaining broken strand on the other side of the break also has an intact template for repair and synthesis proceeds from the 3' OH end of the damaged duplex across the break, using the displaced strand from the other duplex as a template The capture of DNA on the other side of the break by the displaced strand that results from d-loop extension is key to homologous recombination; without capture, the broken chromosome will be permanently fragmented These junctions [Holliday junctions] flank the position of d-loop formation by one end of the broken DNA and the position of capture of the other broken end by the displaced strand of the d-loop The formation and resolution of Holliday junctions is the basis for homologous recombination The central act in homologous recombination that changes the relationship of genetic information on one parental duplex to genetic information on the other parental duplex is covalent linkage followed by separation of DNA strands from one parental duplex to DNA strands of the other parental duplex The key intermediate in homologous recombination that defects this exchange of information and DNA strands between duplexes is the Holliday junction A Holliday junction is formed by the reciprocal exchange via d-loop formation of DNA strands between two duplexes to form regions of heteroduplex DNA Resolution [name of a step] can occur by cleavage of the DNA strands at the Holliday junction at the point where the molecules are joined, and is promoted by structure-specific nucleases called resolvases Cleavage of Holliday junctions can result in permanent exchange of DNA between two duplexes There are two ways in which Holliday junctions can be resolved by cleavage; only one of these results in the permanent exchange of flanking DNA - that is, homologous recombination - between the participating DNA molecules Holliday junctions adopt a planar, unfolded structure when they interact with recombination proteins. The Holliday junction has two axes of symmetry, one depicted here as horizontal and the other as vertical If cleavage occurs in the horizontal direction, the duplexes are separated without exchange of the DNA outside of the Holliday junction. These products are called non-recombinant In contrast, if resolution occurs in the vertical direction, the DNA outside of the Holliday junction is exchanged or crossed over between the two parental duplexes Such recombinant are also said to have undergone a crossover - that is genetic information has been exchanged between the two parental duplexes and the product duplexes are called crossover duplexes or crossovers Branch migration can change the position of a Holliday junction When two identical duplexes are joined in a Holliday junction (for example, where sister chromatids become joined in the course of homologous recombination), the junction can move and change position along the DNAs without any net change in base pairing. Such movement is known as branch migration It is important to note that branch migration can either increase or decrease the region of heteroduplex DNA between the linked duplexes Branch migration can also occur between two homologous, but non-identical, DNA molecules, but large regions of non-homology will inhibit migration Thus, the process of branch migration can, by changing the positions of the Holliday junction (that is, how far it migrated from the original point of exchange), determine the point at which the junction is resolved and the point at which DNA strand exchange between the participating chromosomes can occur The nature of resolution of the double Holliday junction intermediate determines whether exchange of the DNA that flanks the junctions occurs During homologous recombination in meiosis, and sometimes during mitosis, however, two Holliday junctions are often formed and resolved In each case, the same pair of strands is cut at the left junction, but different pairs are cut at the right junction, The consequence of the different cleavages at the right-hand junction is a change in the relationship of the DNA segments outside of the junctions When the junctions are resolved symmetrically - that is, both resolved on the same axis, the product duplexes are non-recombinant - that is, non-crossovers When the junctions are resolved symmetrically - that is, when both junctions are resolved on the same axis, either both horizontally or both vertically, the product duplexes are non-recombinant Thus, recombination, a change in the spatial relationship of the DNA segments that flank a double-strand break, results from Holliday junction formation and resolution By contrast, homology-directed repair by SDSA does not involve covalent linkage between strands from the parental duplexes to form a Holliday junction so that no exchange (crossover) of flanking DNAs is possible Specific enzymes drive branch migration RuvA is a tetramer that binds at the Holliday junction, keeping it in the unfolded and open conformation. RuvA also recruits RuvB to the junction RuvB is a helicase that uses the energy of ATP hydrolysis to move the Holliday junction along the DNA, breaking and re-forming base in one direction, and increasing or decreasing the region of heteroduplex in its wake, depending on the direction in which the junction is moved Branch migration, followed by topoisomerase action, can dissolve double Holliday junctions without the exchange of flanking DNA In another processing pathway, double Holliday junctions can be “dissolved” by multi-protein complexes containing a helicase, a topoisomerase, and modulatory proteins to also yield two separate duplexes These proteins promote the branch migration of the Holliday junctions toward each other to form a hemidesmosomecantenane linking the two duplexes, which is the. “dissolved” by topoisomerase-mediated breakage and joining Notably, in this pathway, there is no exchange of flanking DNA segments - that is, no crossing over - between the two parental duplexes because no DNA strands remain covalently linked in the product duplexes Several types of proteins mediate Holliday junction separation Studies in a wide variety of organisms have revealed extensive structural diversity in resolvases, which can cleave the branched DNA in Holliday junctions and thereby separate duplexes The “dissolvasome” complexes that lead to dissolution of double Holliday junctions to separate duplexes that do not have crossovers - that is, those that are in the parental non-recombinant configuration - include a helicase, a type 1a topoisomerase, and Rm Proteins that associate with both the helicase and the topoisomerase Mismatch repair of the heteroduplex regions in Holliday junctions can result in changes in DNA sequence The formation of a Holliday junction involves the pairing of strands from different DNA helices to form heteroduplex DNA, and branch migration of the Holliday junction influences the extent and location of this heteroduplex DNA If heteroduplex DNA forms between strands from duplexes that are slightly different in sequence (as may occur between homologs, for example), the heteroduplex DNA will contain mismatches, which can be acted upon by the mismatch repair machinery, thus changing or “correcting” the sequence of DNA In the case of mismatch correction of recombination intermediates, however, correction is not biased toward one strand or the other, and either DNA strand can serve as the template for repair Overall, we see that formation of a Holliday junction and branch migration can lead to significant changes in the sequence of a DNA molecule - the formation of heteroduplex DNA may lead to sequence changes in the parental duplexes by gene conversion through mismatch correction and can also lead to the exchange of the DNA Segments that flank the junction, depending on how the Holliday junction is resolved 16.8 repair of damaged replication forks by homology-directed repair DNA damage can collapse replication forks Base lessons, abasic sites or cross-links, or defects in the replication complex itself can stall or block progression of the replication fork, again leading to DSB Collisions of the replisome with transcription complexes can also lead to fork collapse. Another source of Replisome instability is the presence of particular sequences that are difficult to replicate, such as fragile sites Homology-directed repair can restore a collapsed replication fork In such instances [Replication fork collapse], interactions between the undamaged and damaged arms of the fork mediated by homology-directed repair can restore the fork to allow resumption of DNA synthesis Indeed, this repair of replication forks is likely the primary role of homology-directed repair and systems for homologous recombination evolved from this primarily repair process Following degradation to expose 3’ single-stranded DNA, the broken arm of a replication fork interacts with the undamaged arm through the invasion of a single DNA strand from the damaged arm into the undamaged arm to form a d-loop Thus, single DNA strands from each parental duplex are reciprocally exchanged between the parental duplexes. A replisome can then be assembled, with the resumption of DNA synthesis on the lagging strand The exchanged strands that connect the duplexes, which are Holliday junctions, can then be cleared by A resolvase that recognizes such reciprocally exchanged strands to separate the arms of the restored fork The actual resumption of replication requires reassembly of the replisome. Also, replication restart will need to occur wherever for collapse has occurred - that is, without a specific DNA sequence that binds a specific initiator protein Bacteria have solved this problem by using a specialized protein called PriA, which can recognize a d-loop and promote loading of the replicative helicase and other replication components onto the d-loop, and thus promote initiation from a DNA intermediate assembled by homology-dependent interactions Replication fork reversal can repair a blocked replication fork However, other repair pathways related to the replication fork restart, including one related to the homologydirected repair pathway described earlier, can also facilitate the resumption of DNA synthesis The restoration of a damaged replication fork containing unrepaired lessons, such as an abasic site or a pyrimidine dimer, can be achieved by a pathway that involves the pairing of the two newly synthesized strands of DNA with each other, instead of with the template strands This strand re-pairing provides another opportunity for lesion repair by systems such as base and nucleotide excision repair, which use the complementary DNA strand as a template Once repair has occurred, the replication fork may be restarted in several ways (i) first, The regressed fork may be cleaved to yield a collapsed fork, which may be repaired by homologydirected repair (ii) second, the newly synthesized annealed strands may be degraded by a nuclease (iii) third, the regression of the fork may be reversed 16.9 chromosome rearrangements during aberrant recombination and repair Misdirected recombination and repair reactions can be damaging When they are misdirected, however, they can cause genetic changes that can be lethal to the cell or can considerably increase genome instability, and thus the risk of cancer In homologous recombination, misfiring interactions can cause deletions and duplications of DNA on the participating chromosomes. NHEJ reactions can inappropriately join broken ends from unrelated chromosomes, also with damaging consequences Recombination between repeated sequences can result in chromosome rearrangements Homologous recombination between non-allelic repeated sequences can lead to deletions, duplications, or translocations, depending on the relative position of the repeats Recombination may be intrachromosomal, occurring between misaligned repeats on the same chromosome, or interchromosomal, occurring between repeats on sister chromatids, homologous chromosomes, or non-homologous chromosomes Recombination between misaligned repeats on sister chromatids or on homologous chromosomes, often called unequal crossing over, can also occur, resulting in deletions and duplication on the different partners Homologous recombination and homology-directed repair in somatic cells can lead to loss of heterozygosity Diploid organisms are heterozygous at many of their genes - that is, there are likely to be different alleles of any given gene on any two homologous chromosomes However, when one of the abeles is defective, the heterozygous cell depends on the normal allele to grow and behave normally. Therefore, loss of the functional copy of the gene - that is, loss of heterozygosity - can have grave consequences After chromosomes have replicated, recombination may occur between two chromatids from different homologs, resulting in the reciprocal exchange of information between the two chromatids so that the defective gene is now present in one chromatid of each homolog on the same template strand After the chromatids separate and segregate at mitosis, one daughter cell could receive two copies of the normal gene, while the other cell receives two copies of the defective gene Another source of loss of heterozygosity is homology-directed repair that initiates within the chromosome and erroneously extends to the end of a chromosome, in a process called break-induced replication An entire member of a chromosome par can be lost as a result of a malfunction of the cellular machinery that distributes copies of chromosomes evenly to the two daughter cells at mitosis