Conditional knockout of RAD51-related genes in Leishmania major reveals a critical role for homologous recombination during genome replication
Authors:
Jeziel D. Damasceno aff001; João Reis-Cunha aff002; Kathryn Crouch aff001; Dario Beraldi aff001; Craig Lapsley aff001; Luiz R. O. Tosi aff003; Daniella Bartholomeu aff002; Richard McCulloch aff001
Authors place of work:
The Wellcome Centre for Integrative Parasitology, University of Glasgow, Institute of Infection, Immunity and Inflammation, Sir Graeme Davies Building, 120 University Place, Glasgow, United Kingdom
aff001; Laboratório de Imunologia e Genômica de Parasitos, Departamento de Parasitologia, Instituto de Ciências Biológicas, Universidade Federal de Minas Gerais, Belo Horizonte, Minas Gerais, Brasil
aff002; Department of Cell and Molecular Biology, Ribeirão Preto Medical School, University of São Paulo; Ribeirão Preto, SP, Brazil
aff003
Published in the journal:
Conditional knockout of RAD51-related genes in Leishmania major reveals a critical role for homologous recombination during genome replication. PLoS Genet 16(7): e32767. doi:10.1371/journal.pgen.1008828
Category:
Research Article
doi:
https://doi.org/10.1371/journal.pgen.1008828
Summary
Homologous recombination (HR) has an intimate relationship with genome replication, both during repair of DNA lesions that might prevent DNA synthesis and in tackling stalls to the replication fork. Recent studies led us to ask if HR might have a more central role in replicating the genome of Leishmania, a eukaryotic parasite. Conflicting evidence has emerged regarding whether or not HR genes are essential, and genome-wide mapping has provided evidence for an unorthodox organisation of DNA replication initiation sites, termed origins. To answer this question, we have employed a combined CRISPR/Cas9 and DiCre approach to rapidly generate and assess the effect of conditional ablation of RAD51 and three RAD51-related proteins in Leishmania major. Using this approach, we demonstrate that loss of any of these HR factors is not immediately lethal but in each case growth slows with time and leads to DNA damage and accumulation of cells with aberrant DNA content. Despite these similarities, we show that only loss of RAD51 or RAD51-3 impairs DNA synthesis and causes elevated levels of genome-wide mutation. Furthermore, we show that these two HR factors act in distinct ways, since ablation of RAD51, but not RAD51-3, has a profound effect on DNA replication, causing loss of initiation at the major origins and increased DNA synthesis at subtelomeres. Our work clarifies questions regarding the importance of HR to survival of Leishmania and reveals an unanticipated, central role for RAD51 in the programme of genome replication in a microbial eukaryote.
Keywords:
DNA – DNA replication – Cell cycle and cell division – Polymerase chain reaction – DNA extraction – DNA synthesis – Leishmania – Homologous recombination
Introduction
Homologous recombination (HR) has critical roles in the genome maintenance of all organisms, mainly through repair of double stranded DNA breaks [1]. HR is a multistep repair process initiated by resection of the ends of double-stranded DNA breaks to generate single stranded DNA overhangs. This processing provides access to a key player in HR: the Rad51 recombinase (RecA in bacteria, RadA in archaea)[2], which catalyses invasion of the single-stranded DNA into intact homologous duplex DNA, allowing template-directed repair of the broken DNA site. During evolution, duplications of the Rad51 gene have resulted in so-called Rad51 paralogues [3], a set of factors that are found in variable numbers in different organisms and whose spectrum of roles remain somewhat undefined, at least in part because they can belong to a number of protein complexes. Nonetheless, Rad51 paralogues have been implicated in directly modulating HR, acting on Rad51 HR intermediates [4–6], and in wider repair activities for cell cycle progression [7, 8]. HR reactions mediated by Rad51 [9–12] and modulated by the Rad51 paralogues [13] are also required for resolving DNA replication impediments, by promoting protection and restart of stalled replication forks during replication stress. An even more intimate association between HR and DNA replication has been described in bacteria and archaea, where RecA [14–17] and RadA [18] can mediate DNA replication when origins (the genome sites where DNA synthesis begins during replication) have been removed.
In addition to its roles in promoting genome stability, HR can drive to genome variation, which can cause diseases [19], as well as being a means for targeted sequence change during growth, such as during mating type switching in yeast [20]. Genome variation due to HR is found widely in trypanosomatid parasites, which are single-celled microbes that cause human and animal diseases worldwide. In Trypanosoma brucei, HR factors have been clearly implicated in the directed recombination of Variant Surface Glycoprotein (VSG) genes during host immune evasion by antigenic variation [21], as well as in maintenance of the massive subtelomeric VSG gene archive [22, 23]. In T. cruzi, HR has been suggested to be a driver of variability in multigene families [24, 25] and in cell hybridisation [26]. Finally, in Leishmania, HR related factors have been implicated in mediating the formation or maintenance of episomes, which appear to form stochastically, arise genome-wide and have been implicated in acquisition of drug resistance [27–32]. Whether the same roles for HR extend to widespread, stochastic formation of aneuploidy is unknown [33], but this other form of genome-wide variation has also been implicated in adaptation of the parasite, such as during life cycle transitions and in response to drug pressure [34–38].
Despite emerging evidence for HR roles in Leishmania genome change, is it possible that the reaction has wider and deeper functions in genome maintenance and transmission in the parasite? One reason for asking this question is recent observations suggesting that origin number and distribution in Leishmania is unusual, since one study detected only a single site of DNA replication initiation per chromosome [39], while a later study suggested >5000 sites [40]. These data indicate either a pronounced paucity or huge overabundance of origins relative to all other eukaryotes. Alternatively, since in neither study was DNA replication mapping correlated with binding of replication initiation factors, the disparity between the datasets may be due to a widespread, unconventional route for initiation of DNA synthesis acting alongside a small number of conventional origins, perhaps indicating novel strategies for DNA replication that may link with genome plasticity [41]. A second reason for asking about the importance of HR for genome transmission in Leishmania is other work that has led to uncertainty about the importance of HR factors for survival of the parasite. Leishmania encodes a highly conserved, canonical Rad51 recombinase [30, 42, 43], as well three Rad51 paralogues, referred to as RAD51-3, RAD51-4 and RAD5-6 [29], a slightly smaller repertoire of non-meiotic Rad51-related proteins than is found in T. brucei [44–47]. In L. donovani [48], unlike in L. infantum [30], it has proved impossible to make RAD51 null mutants. Furthermore, while null mutants of RAD51-4 are viable in L. infantum, RAD51-3 has been described as being essential, and RAD51-6 nulls were not recovered in the same experiments [29]. Mutation in Rad51 or its relatives has never been shown to be lethal in any single celled eukaryote, notably including T. brucei [49], or in prokaryotes, making these observations in Leishmania particularly striking.
In this work we sought to resolve the question marks over essentiality of HR factors in Leishmania and to test for roles in DNA replication by using conditional gene knockout (KO), comparing the short- and long-term effects of ablating RAD51 and each of its three RAD51 paralogues. Our data show that loss of each gene is, over time, increasingly detrimental to Leishmania fecundity, demonstrating that black and white definitions of essential or non-essential are too limiting for HR genes in the parasite. In addition, we show that the functions provided by the RAD51 paralogues are non-overlapping in Leishmania, and we reveal that RAD51 plays an unexpected, central role in genome replication, since in its absence the normal programme of DNA replication is profoundly altered.
Results
A combined CRISPR/Cas9 and DiCre approach for assessment of gene function in L. major
In order to compare the effects of ablating RAD51 and each of the three known L. major RAD51 paralogues, we adopted a rapid approach to generate cell lines for conditional induction of a gene KO. For this, we used a cell line constitutively expressing Cas9 and DiCre (Fig 1A). In this strategy, we first used CRISPR/Cas9-mediated HR to exchange the endogenous copy of the genes by a copy flanked by loxP sites. In addition, each construct translationally fused copies of the HA epitope to the C-terminus of the gene’s ORF. PCR showed this approach to be very efficient for RAD51 and the three RAD51 paralogues, since selection using only puromycin resulted in all wild type (WT) copies of each gene being replaced by floxed and tagged versions after a single transformation (S1 Fig). Because RAD51 and the RAD51 paralogue mutants may generate similar phenotypes [44, 46], since each contributes to HR [29], we used the same approach to modify the L. major gene encoding the orthologue of T. brucei PIF6. This factor has not been functionally examined in Leishmania, but in T. brucei PIF6 is the sole known nuclear Pif1 helicase homologue [50]. Since Pif1 helicases have been implicated in modulating DNA replication passage through barriers and during termination[51, 52], and thus operate in distinct aspects of nuclear genome maintenance compared with Rad51 paralogues, we considered this gene could provide a valuable control for the effects of conditional gene KO of the Rad51-related proteins. A number of attempts failed to replace all WT endogenous L. major PIF6 gene copies with HA-tagged versions, whereas each copy could be floxed with untagged gene variants (S1 Fig). Growth curves showed that addition of loxP sites or the HA tag did not lead to any significant growth impairment for any of the four RAD51 paralogues or PIF6 (S2 Fig). However, growth of cells expressing HA-tagged RAD51 was impaired, suggesting the epitope impedes activity of the recombinase (S3 Fig), and so we generated cells with floxed copies of untagged RAD51 (S1 Fig), which grew normally (S2 Fig) and were used in all subsequent experiments.
Next, KO induction of each gene was attempted by rapamycin-mediated DiCre activation in logarithmically growing cultures of each cell line (Fig 1A, S4 Fig). PCR analysis using DNA from cells after a number of induction rounds (‘passages’), where cells were grown from low to high density and repeated by dilution, showed that complete gene excision was achieved after passage 2 for all genes (S4 Fig). Controls without addition of rapamycin showed no gene excision, and unexcised gene copies were undetectable even after >15 passages in the presence of rapamycin (S4 Fig). The rapidity of DiCre mediated loss of the gene products was confirmed by western blotting (RAD51 paralogues) and RT-PCR (RAD51, PIF6): signal for all HA-tagged proteins was no longer detectable after 48 h of the second round of KO induction (Fig 1C, S3B Fig), and RAD51 or PIF6 cDNA could not be PCR-amplified (Fig 1B and 1D). KO induction of RAD51 and of each of the RAD51 paralogues, but not of PIF6, resulted in increased levels of γH2A [53] in western blotting analysis (Fig 1E), suggesting accumulation of nuclear DNA damage after loss of any L.major RAD51-like protein, but not after ablation of PIF6 (at least during unperturbed growth; see below). To attempt to answer the question of whether or not RAD51 and the RAD51 paralogues are essential in Leishmania [29, 48], we measured growth of the parasites for a prolonged period after DiCre-induced gene excision (Fig 1F). At least until passage 4, no growth defect was seen due to loss of RAD51, any of the RAD51 paralogues or PIF6 in HOMEM medium. However, when kept in culture for longer periods (Fig 1F), the RAD51 KO cells and each of the RAD51 paralogues KO cells, but not the PIF6 KO cells, showed marked growth defects, suggesting HR factors that contribute to the catalysis of homology-directed strand exchange might be critical for long-term Leishmania viability when cultured in HOMEM. In fact, when grown in M199 medium, growth defects were observed more rapidly after excision of RAD51 or a RAD51 paralogue, though again not after excision of PIF6 (Fig 1F). Accordingly, flow cytometry analysis showed that prolonged cultivation after KO induction of RAD51 and the RAD51 paralogues, but not PIF6, resulted in an increased proportion of cells with less than 1C DNA (S5 Fig), suggesting increased genomic instability, perhaps reflecting the increased levels of γH2A. Taken together, the phenotypes seen after induced loss of the five genes suggest some overlap in functions of RAD51 and its relatives, but a distinct role for PIF6. In addition, the PIF6 data demonstrate that prolonged exposure to rapamycin, or effects of DiCre and Cas9 expression, have a negligible effect on growth in these conditions.
Loss of RAD51 or RAD51-3, but not RAD51-4 or RAD51-6, impairs DNA synthesis in L. major
To ask if the impaired growth seen in four of the five induced KO cells is due to a common defect, we tested the extent of DNA synthesis after each gene deletion. To do this, rapamycin induced and uninduced cells were subjected to a short pulse of IdU labelling followed by immunostaining under denaturing conditions and flow cytometry detection, allowing us to track the level and pattern of DNA synthesis in each cell cycle stage (Fig 2A). Loss of RAD51 or RAD51-3, but not loss of RAD51-4, RAD51-6 or PIF6, resulted in a reduced percentage of IdU-positive cells in the population. Quantification of IdU signal in individual S-phase cells at 48 and 72 h of the second round of KO induction confirmed these effects (Fig 2B): a significant reduction in IdU fluorescence was found at both time points in the rapamycin-induced RAD51 and RAD51-3 KO cells compared with their cognate uninduced controls, whereas no such reduction was seen after KO of RAD51-4, RAD51-6 or PIF6. These data suggest that only loss of RAD51 or RAD51-3 affects DNA synthesis, meaning the growth impairment seen after loss of RAD51 and its relatives, though similar in extent, might not have a common basis, or loss of DNA synthesis is not the main reason for growth reduction after the induced KO of RAD51 or RAD51-3.
Next, we asked if the observed increases in DNA damage after induced KO of the RAD51-like genes all had the same basis by examining γH2A levels across the cell cycle. For this, rapamycin induced and uninduced cells were arrested in G1 using 5 mM hydroxyurea (HU) and then released from arrest by removing HU, sampling at the point of arrest and at various times after release for western blotting (Fig 2C). The patterns of γH2A accumulation revealed notable differences in the cell cycle functions of the five genes (Fig 2C). KO induction of RAD51 or RAD51-3 resulted in a pronounced increase of γH2A levels in cells navigating through S-phase up to G2/M, suggesting roles related to the resolution, before cell division, of genome injuries that arise during DNA replication. In contrast, loss of RAD51-4 or RAD51-6 did not show any clear evidence for increased γ H2A signal compared with uninduced controls after HU release, suggesting a more limited contribution to tackling replication-associated damage, which seems consistent with the absence of changes in IdU uptake after KO. PIF6 KO displayed a yet further difference, with increased levels of γH2A only ~6 h after HU release (Fig 2C), when much of the population had passed through S-phase (S6 Fig). These data indicate that loss of PIF6 does in fact result in nuclear damage, but this is more limited than is seen after loss of the RAD51-like genes, and suggests that if the putative helicase has a role in resolving replication problems, this is concentrated in the final steps of DNA replication or even in post-replication steps of the cell cycle.
DNA content in all the samples analysed for γH2A levels was next analysed by flow cytometry (S6 Fig). Intriguingly, no pronounced changes in cell cycle progression after release from HU arrest were observed upon KO induction of any of the HR genes or PIF6 (S6 Fig), perhaps suggesting cell cycle checkpoints are not enacted by the gene KOs, despite clear DNA damage accumulation after loss of RAD51 or its paralogues. Altogether, these data suggest that Leishmania RAD51 and RAD51-3, specifically amongst the genes examined, have roles in promoting effective DNA synthesis and their absence results in increased nuclear genome damage during S-phase.
Ablation of RAD51 or RAD51-3, but not RAD51-4, results in widespread mutagenesis
We next sought to determine if loss of the HR genes results in genome instability by using short-read Illumina sequencing of DNA from RAD51, RAD51-3 and RAD51-4 KO cells after growth for two and six passages in the presence of rapamycin, as well as in the same cells grown without rapamycin (Fig 3A). In each case, mapping of reads to the genome showed specific loss of sequence around the loxP-flanked gene, confirming KO induction (Fig 3B). To understand the effects of HR factor loss, we measured the number of single nucleotide polymorphisms (SNPs) in the induced and uninduced cells after passage two and six (P2 and P6) by comparing these genome sequences to the reference genome; in addition, to clearly determine the effect of DiCre excision, SNPs that were common to the two time points in the induced and uninduced cells were discarded (Fig 3C). Irrespective of whether or not gene loss was induced, SNPs accumulated during growth of L. major and the consequences of loss of the three HR factors was not equivalent (Fig 3D). Loss of RAD51 reduced the level of SNPs relative to uninduced cells at P2, while a significantly increased accumulation of SNPs was seen by P6. In contrast, loss of RAD51-3 caused a small but significant increase of SNPs at P2, which was no longer detectable at P6. Finally, there was no evidence that loss of RAD51-4 increased SNP accumulation relative to uninduced cells at either passage. To ask if this pattern of mutagenesis is only seen with SNPs, we also measured insertions and deletions (InDels) in the same cells (S7 Fig), with the same differential patterns seen in the three different mutants (S7C Fig).
Strikingly, when new SNP or InDel density was plotted individually for each of the 36 chromosomes, it became apparent that the increase in SNPs with or without gene KO was not random across the genome, as the smaller chromosomes tended to present a higher density of new SNPs and InDels than the larger chromosomes (Fig 3E, S7D Fig). To ask further if there is localised accumulation of mutations in the L. major genome, we examined SNP density proximal to the ‘strand switch regions’ (SSRs) within each chromosome where multigene transcription initiation and/or termination occurs (Fig 3F), and a subset of which are where MFA-seq has mapped DNA replication initiation (i.e. are predicted replication origins)[39]. This mapping revealed that SNP density peaks around the SSRs, suggesting these loci are hotspots for mutation. Furthermore, SSR proximity mapping of SNPs confirmed a difference between the three gene KOs. Following loss of RAD51 a decrease in SNP accumulation was seen relative to uninduced controls, and this was accounted for by reduced levels pf SNPs at origin-active SSRS, with little evidence of a change at origin-inactive SSRs. The opposite effect was observed after RAD51-3 KO, with increased SNP levels at origin active SSRs, and no change in SSR-proximal SNP density was found after RAD51-4 KO. Thus, the two HR factors whose loss was found to affect global DNA synthesis were also seen to affect SNP accumulation around SSRs, unlike RAD51-4 KO, which did not show an effect on DNA synthesis. A clear peak of new InDels in the uninduced cells was less apparent than was seen for new SNPs (compare S7E Fig with Fig 3F). However, InDels appeared to accumulate to a greater extent around origin-active than origin-inactive SSRs upon RAD51 KO (S7E Fig), while such effects were less clearly seen upon RAD51-3 KO, and no difference was seen in RAD51-4 KO cells compared with controls. In total, therefore, InDel accumulation proximal to SSRs upon HR KO was more modest than SNPs, but again any distinct accumulation around SSRs was most clearly detected after loss of RAD51 or RAD51-3, the two factors our data implicate in global DNA synthesis. Moreover, loss of RAD51 appears to have a distinct effect on SNP and InDel accumulation, which is primarily seen at predicted origin-active SSRs.
Given that excision of RAD51 and RAD51-3 altered the frequency of SNP accumulation, we next examined the types of mutation that resulted in the SNPs (S8 Fig). Though the data showed some bias towards specific forms of base substitution, it was not clear that induced loss of any of the HR genes altered this pattern relative to uninduced controls and nor was any difference seen when comparing loss of RAD51, RAD51-3 or RAD51-4.
To examine the effects of loss of either RAD51 or RAD51-3 further, we examined survival of the KO cells relative to uninduced controls in the presence of increasing concentrations of phleomycin and camptothecin, both of which cause DNA double-strand breaks, and HU, which inhibits ribonucleotide reductase and impairs DNA synthesis (S9 Fig). As expected for predicted DNA double-strand break repair factors, KO induction of either RAD51 or RAD51-3 lead to increased sensitivity to phleomycin and camptothecin. However, only RAD51-3 KO led to increased sensitivity to HU. Consistent with this growth difference, levels of SNP accumulation after exposure to and release from HU treatment (S10A and S10B Fig) also differed between the two KOs. Both genome-wide (S10C Fig) and in each chromosome (S10D Fig), fewer HU-induced new SNPs were detected in RAD51 KO cells exposed to HU compared with uninduced, whereas KO of RAD51-3 did not have such an effect after HU exposure. Despite this differential effect, HU treatment did not alter the SNP mutation profile, with or without gene excision (S11 Fig: compare to S8 Fig). Moreover, unlike for SNPs, loss of either HR gene caused a significant increase in InDel levels, with or without exposure to HU (S10C and S10D Fig), and HU-induced replication stress did not clearly change the SNP or InDel patterns around SSRs upon RAD51 or RAD51-3 KO (S10E Fig) relative to the changes seen in the absence of HU (Fig 3, S7 Fig). Altogether, these data reinforce the view that, despite loss of RAD51 or RAD51-3 both causing impaired DNA synthesis, the roles of the two factors in maintaining the L. major genome differ.
Generation of conditional double gene mutants by CRISPR/Cas9 and DiCre in Leishmania
To date, the analysis of single gene conditional KOs has implicated only RAD51 and RAD51-3 amongst the four HR factors in DNA synthesis. One explanation may be that RAD51-4 and RAD51-6 act redundantly in Leishmania. To test this, we used the combined CRISPR/Cas9, DiCre approach to attempt to make conditional double gene KOs of the Rad51-paralogues in all possible combinations (Fig 4A). First, CRISPR/Cas9 was used to generate floxed copies of either RAD51-3 or RAD51-4 using gene-specific puromycin-resistance constructs (described above). Next, the resulting cell lines were subjected to a second round of CRISPR/Cas9 engineering to delete the endogenous copies of RAD51-4 or RAD51-6 in the RAD51-3-HAflox cells, or RAD51-6 in the RAD51-4-HAflox cells. PCR on G418-resistant clones for each of the three cell lines (S12 Fig) showed that it was possible to delete all copies of either RAD51-4 or RAD51-6, retaining floxed copies of RAD51-3-HA or RAD51-4-HA. DiCre-mediated KO of the floxed gene was then induced by addition of rapamycin. PCR analysis (S13 Fig) showed complete excision of RAD51-3-HA or RAD51-4-HA after the second round of rapamycin-mediated DiCre induction in all three cell lines, thereby generating cells devoid of two L. major Rad51-paralogues genes simultaneously (RAD51-4 and RAD51-3; RAD51-6 and RAD51-3; or RAD51-6 and RAD51-4). Complete loss of the genes was maintained after cells were grown for more than 15 passages (S13 Fig).
Western blotting (Fig 4B) showed that HA-tagged RAD51-3 or RAD51-4 protein in the null mutants was undetectable after rapamycin induction of gene KO. The RAD51-4 and RAD51-6 null mutants, prior to induction of conditional excision of the second gene, showed increased levels of γH2A compared with the background cell line (Fig 4C). DiCre-mediated gene excision, leading to the three different double gene KOs, did not result in further increases in the levels of the phosphorylated histone. The lack of further impairment when two genes are lost relative to one was broadly consistent with growth curves (S14A Fig) and with flow cytometry analysis of DNA content (S14B Fig): conditional excision of the second gene did not lead to worsening of growth or to increased numbers of aberrant cells with less than 1C content compared with uninduced single gene null cells. Intriguingly, the RAD51-4 and RAD51-6 null mutants, even without induction of the second gene KO and having been selected as clones, appeared to recapitulate the long-term growth impairment seen upon prolonged cultivation after DiCre excision of a single HR gene (compare S14A Fig with Fig 1F).
IdU labelling followed by FACS analysis of each of the three cell lines (Fig 4D), before and after rapamycin induction, did not show a change in the proportion of cells that incorporated the nucleotide analogue. Quantification of IdU fluorescence levels of individual S-phase cells (Fig 4E) did not reveal a consistent pattern of fluorescence decrease upon simultaneous KO of two RAD51 paralogs. However, a signal reduction was detected 72 h after RAD51-3 KO induction in both the RAD51-4 and RAD51-6 null backgrounds, consistent with loss of only RAD51-3 amongst the paralogues impeding DNA synthesis (Fig 2A and 2B). In contrast, though a mild loss of fluorescence was seen at 48 h when KO of RAD51-4 was induced in the RAD51-6 null background, this was not seen at 72 h (Fig 4E). These data reinforce the suggestion that RAD51-3 alone among the three RAD51 paralogues plays a role in L. major DNA replication, and indicate that RAD51-4 and RAD51-6 do not obviously act redundantly in such a function.
To ask if the induction of a double RAD51 paralogue gene KO had similar effects on genome instability relative to what was seen after induction of single gene KOs (Fig 3), we again performed Illumina sequencing of DNA from the RAD51-4 null RAD51-3-HAflox cells after two and six passages of growth with and without rapamycin (Fig 5A and 5C). Read mapping to the genome showed the RAD51-4 gene was absent and that DiCre induction lead to removal of RAD51-3 (Fig 5B). Like what was seen in the RAD51-3 KO (Fig 3D), SNPs accumulated to a greater extent across the genome (Fig 5C) and in each chromosome (Fig 5D) in the RAD51-3 RAD51-4 double KO cells at P2, but no difference was seen at P6. Furthermore, increased SNP accumulation around SSRs upon induction of double a RAD51-4 RAD51-3 KO seemed comparable with the RAD51-3 KO, and distinct from RAD51-4 KO cells (Fig 5E). Finally, broadly comparable changes in levels and patterns of InDel accumulation were seen after induction of the double RAD51-4 RAD51-3 KO (S15A–S15C Fig) and RAD51-3 single KO (S7 Fig). Thus, these data, allied to previous DNA synthesis and cell cycle-dependent damage analysis, suggest a complex mixture of common and separate functions for these two RAD51 paralogues, probably related to epistatic interactions among them.
Loss of RAD51 leads to changes in the DNA replication landscape of Leishmania
Our analyses until this point indicated impairment of DNA synthesis after loss of either RAD51 or RAD51-3, including evidence for distinct effects of the gene KOs. However, if and how the two factors might contribute to the programme of DNA replication in L. major was not known. To address this question, we performed MFA-seq analysis, comparing read depth across the chromosomes in DNA extracted from replicating and non-replicating cells, thereby identifying sites of replication initiation and patterns of fork progression [39, 54, 55]. To do this, we used a modified version of MFA-seq compared to what we have described previously: rather than using FACS to isolate S- and G2-phase cells, we isolated DNA from L. major cells in logarithmic (log) or in stationary phase growth, sequenced each using Illumina technology, and mapped the ratio of reads in the two populations. Fig 6A shows these data as Z-scores across two chromosomes, where positive signal represents regions where the mean read depth in the log phase cells is greater than the mean of the stationary cells (S16 Fig shows further chromosomes). This modified MFA-seq approach will be described in detail elsewhere (BioRXiv 10.1101/799429), but two primary features of Leishmania replication are revealed relative to MFA-seq based on S/G2 phase read depth ratios using cell cycle-sorted cells [39]. First, we confirm the predominant use of single origins in each chromosome, most of which are centrally disposed in the molecules and each coinciding with SSRs. Second, we now detect weaker sites of replication initiation that were not seen previously and are proximal to one or both telomeres in the chromosomes (see -RAP data in Fig 6A, which are highly comparable with WT cells; BioRXiv 10.1101/799429).
To ask about the effect of RAD51 or RAD51-3 loss on DNA replication, we performed MFA-seq in cells induced for KO by growth in rapamycin (48 h, second round of induction), as well as in control cells without rapamycin grown for the same time. Comparing induced RAD51 KO cells with uninduced cells revealed a striking variation in MFA-seq pattern: loss of MFA signal was seen at the main origin that had previously been mapped within each chromosome [39], while an increase in MFA-seq signal was seen at the extremities of the chromosomes (Fig 6A, S16 Fig). In the induced RAD51-3 KO cells, the same differential effect was less obvious: though there was potentially some loss of MFA-seq signal at the main origin, increase in subtelomeric signal was not apparent. To examine this genome-wide, we generated metaplots of the MFA-seq signal in the different cells (Fig 6B and 6C). Profiling of MFA signal around the main origin (Fig 6B) revealed considerable consistency in amplitude and width of the peaks, both for the 36 origins within one cell and between the two uninduced cells, confirming there is little variation in timing or efficiency of DNA synthesis initiation at all main origins in the WT L. major population [39]. The same profiling confirmed that loss of MFA-seq signal around the main origins was a genome-wide effect upon RAD51 KO, with lowered amplitude and width, and was less pronounced upon KO of RAD51-3. These data indicate that loss of RAD51 leads to a more pronounced decrease in replication initiation activity at the main origins compared with RAD51-3 loss, which appears consistent with decreased SNP accumulation at SSRs after RAD51 KO but not after RAD51-3 KO (Fig 3F). Profiling of the MFA-seq mapping at the chromosome ends (Fig 6C) showed that loss of RAD51, but not RAD51-3, resulted in a gain of signal in the subtelomeres, which again was strikingly consistent in amplitude and width across all chromosomes. Altogether, these data indicate that loss of DNA replication initiation at the main, chromosome-central origins due to ablation of RAD51 is accompanied by increased DNA replication in the subtelomeres, a shift in the replication programme that is not clearly seen after loss of RAD51-3. In addition, these observations further confirm that RAD51 and RAD51-3 play distinct roles in Leishmania DNA replication.
It remains possibile that the above effects could be due to replication stress and DNA damage accumulation upon RAD51 KO (Fig 1E). Though this possibility appears to be argued against by the effects of RAD51-3 KO, which also resulted in increased levels of DNA damage but did not lead to an alteration in the MFA-seq profile, we attempted to test this idea further by searching for changes in MFA-seq profile upon treatment with 5 mM HU for 8 hours, which also causes replication stress associated with DNA damage levels (Fig 2C). HU treatment resulted in a narrower MFA-seq peak at the centre of the origin-active SSRs, surrounded by two shoulders (Fig 6B). This effect may be explained by impairment of bi-directional replication fork movement around the origins, since the cell cycle is mainly arrested at the G1/S transition by this treatment (S6 Fig). Only subtle changes in this HU-induced MFA-seq signal were observed after loss of RAD51 or RAD51-3, consistent with the FACS analysis showing no evidence for defective cell cycle arrest upon absence of the HR factors (S6 Fig). These data may indicate that the effect of RAD51 loss on DNA replication around origins in untreated cells is not due to roles at the early stages of S phase, but rather later in the cell cycle. MFA-seq signal near the chromosome ends upon RAD51 KO was very comparable between untreated and HU-treated cells, suggesting these replication profile changes are mainly concentrated at the early stages of S phase. Overall, HU treatment suggested the pronounced change in replication profile upon RAD51 loss cannot be simply be accounted for replication stress or DNA damage accumulation, suggesting RAD51 is more intimately involved in the L. major DNA replication programme.
Discussion
Homologous recombination is known to be important for episome formation in Leishmania [29–31], but the depth and breadth of how the process contributes to genome plasticity and transmission has not been fully explored. Here, we sought to answer two main questions. First, are RAD51 and RAD51 paralogues essential in Leishmania, given the conflicting data on the ability to generate and propagate null mutants in different species? Second, given the potentially novel distribution of mapped origins in the Leishmania genome, does HR play a central role in Leishmania genome duplication? Using a rapid, conditional knockout approach that combines CRISPR/Cas9 gene modification and DiCre gene excision, we show that loss of any RAD51-like gene is not immediately lethal to promastigote L. major but is increasingly detrimental over time, with phenotype penetrance dependent on cultivation conditions. Thus, we suggest that binary definitions of essential or non-essential for HR-related genes inadequately describe their contribution to parasite biology. In addition, our data reveal that loss of RAD51 and RAD51-3, uniquely among the four HR proteins examined, impairs DNA replication, though their roles are distinct, with RAD51 playing an unexpected and central role in DNA replication initiation.
Null mutants of RAD51 and RAD51-4 have been described in L. infantum [29, 30], whereas RAD51 mutants could not be recovered by CRISPR/Cas9 gene targeting in L. donovani [48]. In addition, RAD51-3 and RAD51-6 null mutants were not recovered by two-step homology-directed gene deletion in L. infantum [29]. The conditional KO approach used here aids understanding of this complex biology, since we show that excision of any of these HR factors has no immediate impact of parasite fitness, but rather causes a progressive slowing of growth, presumably due to accrual of problems. Nonetheless, it is clear that Leishmania null mutants of RAD51 (in L. infantum)[30], and the RAD51 paralogues RAD51-4 and RAD51-6 (this work, in L. major), can be generated. Similarly, null mutants of each RAD51-related gene have been described in T. brucei [44, 46]. It is possible such variation in importance reflects species-specific aspects of HR function. However, it is also conceivable that conventional, two step transformation approaches to generate null mutant clones allows for selection of cells possessing compensatory changes that lead to survival—an adaptation that may not always be recovered when using CRISPR-Cas9 to simultaneously ablate both alleles [48], or that emerged in the timeframe we have used during conditional gene excision (this work). What such adaptations might be is unclear, but the abundant mutations we describe after loss of HR genes may provide a genetic basis for their generation. In addition, RAD51-directed HR is not the sole route for repair of DSBs in any trypanosomatid [56–62], though whether such pathways can increase in activity in the absence of RAD51-directed HR has not been tested. What aspects of Leishmania genome function degrade with time after loss of RAD51 and its relatives remain to be fully characterised, though our data suggest these effects may differ for the different RAD51 paralogues, and for RAD51.
Conditional loss of RAD51 and RAD51-3 led to genome-wide increases in SNPs, as well as increased amounts of in InDels. Loss of RAD51-4 did not lead to such obvious mutation levels and the effect of RAD51-6 loss remains to be determined. Nonetheless, these data demonstrate a widespread role for HR related proteins in Leishmania genome stability. In addition, we did not attempt to test for larger genome changes, but translocations have been described in MRE11 [27], which can guide RAD51 function. Our mapping of SNPs also revealed two unusual features of the patterns of mutagenesis in L. major. First, there was a notable chromosome size-dependence on levels of SNP accumulation, with smaller chromosomes tending to have higher SNP densities when compared with the larger chromosomes. This was common to each conditional gene KO, indicating it is general feature of L. major chromosome biology. However, the basis of this effect is unclear. Might it relate to differences in gene expression or nucleosome occupancy? No data that we are aware of supports such a suggestion, and the commonality of multigene transcription in all chromosomes appears to argue against it [63]. If not gene expression, then the chromosome size-dependence of SNP density might reflect the limitations of predominant DNA replication initiation at a single origin [39]. The second feature was pronounced accumulation of SNPs around SSRs. Here, loss of the HR factors has somewhat different effects: for RAD51 KO, a decrease in SSR-proximal SNP accumulation was seen, whereas an increase was seen upon RAD51-3 KO and no change was found after KO of RAD51-4. These data suggest that the SNPs reflect the effects of differing roles of the proteins on damage repair. Given that both RAD51 and RAD51-3 loss affects DNA synthesis, while loss of RAD51-4 does not, an explanation could be that the SNPs are generated due to clashes between the transcription and replication machineries at SSRs, providing a source for such pronounced mutation rates at these sites. In T. brucei it is known that the Origin Recognition Complex (ORC) binds to potentially all SSRs and its loss affects levels of RNA at these loci [54]. In addition, RNA-DNA hybrids, which have been mapped to sites of replication-transcription clashes in other eukaryotes [64], form prominently at transcription start sites in T. brucei [65]. Thus, it seems conceivable that SSRs are also sites of such interaction in Leishmania, though no equivalent mapping of ORC or RNA-DNA hybrids has been reported.
Loss of IdU uptake after induced KO of RAD51 or RAD51-3 indicates a role for both HR factors in DNA synthesis and DNA replication, but several lines of evidence suggest these roles are not the same: distinct cell cycle timing of γH2A accumulation, distinct patterns of SNP accumulation around SSRs, and differing changes in DNA replication dynamics after their loss. Impaired nucleotide uptake after RAD51 loss appears to be explained by a shift in the programme of L. major DNA replication, with loss of efficient DNA replication initiation at the single primary origin in each chromosome and increased subtelomeric DNA replication initiation. The finding that RAD51 KO cells are impaired in growth and nucleotide uptake argues that increased replication from the subtelomeres is insufficient to compensate for loss of the primary initiation events. In the case of RAD51-3, the very modest loss of replication at the main origin, and no clear increase in subtelomeric replication, seems most readily explained by a widespread impairment in genome replication, perhaps because the RAD51 paralogue is needed to guide processes involved in promoting replication in the face of genome-wide impediments. In this regard, the replication phenotypes after loss of RAD51-3 may be comparable with effects seen after mutation of T. brucei MCM-BP [66], a poorly understood factor that modulates activity of the replicative MCM helicase [67, 68]. In addition, such a role for L. major RAD51-3 may be akin to roles for mammalian Rad51 paralogues on stalled DNA replication forks, with the differing effects of loss of L. major RAD51-4 or RAD51-6 suggesting similar functional compartmentalisation amongst the paralogues [13].
How might RAD51 provide a central role in Leishmania DNA replication? One possibility is that DNA replication initiation at the single SSR-localised origin in each chromosome is directly driven by RAD51, perhaps even by catalysing HR. Such an origin-specific function for RAD51 or RAD51-directed HR has no precedent and, indeed, would be distinct from origin activity in T. brucei, where these sites are conventionally defined by binding ORC [54, 69]. Nonetheless, ORC binding has not been mapped in any Leishmania genome and precedents for recombination-directed replication initiation exist in viruses [70–72], bacteria [17], polypoid archaea [18] and in Tetrahymena [73], albeit normally without a focus on defined genome sites. In addition, human Rad51 acts with MCM8-9, an alternate MCM helicase complex, to initiate DNA replication [74]; though this DNA synthesis pathway is origin-independent in humans, no work has tested MCM8 or MCM9 function in any trypanosomatid. Therefore, it remains to be tested if RAD51 associates with DNA replication initiation complexes and/or with the origins of replication.
Alternatively, and perhaps more likely, RAD51 may play a more indirect role in Leishmania genome replication. The earliest acting origin-active SSRs in T. brucei co-localize with centromeres [54, 75], and recent work has mapped one putative component of the centromere-binding kinetochore, KKT1, to each MFAseq-mapped origin SSR in L. major [76]. These loci might then be vulnerable to breakage, such as during chromosome segregation, and RAD51-directed HR may be required to repair any breaks, such as after mitosis in order to allow proper licensing and firing of origins at these loci. Such a scenario could be compatible with data from other eukaryotes of important roles for RAD51-directed HR is maintaining centromere function [77–79]. In T. brucei, the presence of further ORC-defined origins in each chromosome may compensate for loss of RAD51 causing impaired centromere-focused origin function, but such a mutation in Leishmania might be more detrimental, given the presence of only a single major centromere-focused origin in each chromosome [39].
A distinct suggestion for an indirect role of RAD51 is that the recombinase does not play an origin- or centromere-focused role, but instead is needed to support DNA replication genome-wide, given previous suggestions that replication from a single major origin in each Leishmania chromosome would be insufficient to replicate all chromosomes in S-phase [39, 41]. Such a function could be compatible with origin-independent roles of HR-directed replication (discussed above) and, in the absence of RAD51, complete chromosome replication is lost during S-phase, leading to impaired mitosis and reduced numbers of cells that license the main origins. Alternatively, loss of RAD51 may slow progress of replication forks emanating from the single central origin in each chromosome, preventing completion of chromosome duplication in S phase. Either scenario may explain the increased levels of MFA-seq reads in the subtelomeres. Increased subtelomeric MFA-seq signal in RAD51 KO cells may reflect greater numbers of cells stalling in late- or post-S phase. Alternatively, subtelomeric DNA synthesis may result from the activation of dormant origins in these regions, serving as back-up for the primary SSR-focused DNA replication initiation reaction; thus, when the primary reaction is compromised by loss of RAD51, subtelomere initiation assumes greater importance. As it stands, we do not know the nature of the subtelomeric DNA synthesis, but these data indicate it is distinct from DNA replication emanating from the main SSR-focused origins and is RAD51-independent. Whether or not this subtelomeric DNA replication relates to a recently proposed form of telomere maintenance [34] is worthy of further investigation.
Methods
Parasite culture
Cell lines were derived from Leishmania major strain LT252 (MHOM/IR/1983/IR). Promastigotes were cultured at 26°C in M199 or HOMEN medium supplemented with 10% heat-inactivated fetal bovine serum. Transfections were performed with exponentially growing cells with Amaxa Nucleofactor II, using the pre-set program X-001. Transfectants were selected by limiting dilution in 96-well plates in the presence of appropriate antibiotic. DiCre-expressing cells were selected with 10 μg.mL-1 blasticidin. DiCre/Cas9/T7-expressing cells were selected with 10 μg.mL-1 blasticidin and 20 μg.mL-1 hygromycin. Cells expressing the gene of interest (GOI) flanked by LoxP sites (GOIFlox cells) were selected with 10 μg.mL-1 blasticidin, 20 μg.mL-1 hygromycin and μg.mL-1 puromycin.
DNA constructs and cell line generation
A background cell line was established in which DiCre is expressed from the ribosomal locus, while both Cas9 and T7 polymerase are expressed from the tubulin array. For this, WT cells were transfected with plasmid pGL2339 [80], previously digested with PacI and PmeI, to generate DiCre-expressing cells. Correct integration into the ribosomal locus was confirmed by PCR. Then, the DiCre-expressing cells were transfected with plasmid pTB007 [81], previously digested with PacI, to generate the DiCre/Cas9/T7-expressing cell line. Correct integration of Cas9/T7-encoding cassette was confirmed by PCR. In this way, the Cas9/T7 system, as previously described [81], was used to flank all copies of a GOI with LoxP sites, in a single round of transfection to generate the GOIFlox cell lines used here. Deletion of the GOI was induced by rapamycin-mediated DiCre activation, as previously reported [82, 83].
Donor fragments for Cas9-mediated genome editing were generated by PCR (S1 Table). For this, the ORFs encoding each GOI were PCR-amplified using genomic DNA as template. PCR products of RAD51 (LmjF.28.0550), RAD51-4 (LmjF.11.0230), RAD51-6 (LmjF.29.0450) and PIF6 (LmjF.21.1190) were cloned between NdeI and SpeI restriction sites in the vector pGL2314 [83]. PCR product of RAD51-3 (LmjF.33.2490) was cloned into the SpeI restriction site of vector pGL2314. The resulting constructs contained the GOI flanked by LoxP sites (pGL2314GOIFlox) and were used as templates in PCR reactions to generate the donor fragments flanked by sequences homologues (30 nucleotides) to the targeting integration sites. PCR products were ethanol precipitated and transfected into the DiCre/Cas9/T7-expressing cell line. Correct integration into the expected locus was confirmed by PCR analysis.
The strategy used to generate sgRNAs was essentially as previously described[81]. Briefly, sgRNAs were generated in vivo upon transfection with appropriate DNA fragment generated by PCR. These fragments contained the sequence for the T7 polymerase promoter, followed by the 20 nucleotides of sgRNA targeting site and 60 nucleotides of sgRNA scaffold sequence. The Eukaryotic Pathogen CRISPR guide RNA/DNA Design Tool (http://grna.ctegd.uga.edu) was used to generate the 20 nucleotide sequences for sgRNA targeting sites. The default parameters and the highest scoring 20 nucleotide sgRNA sequences were chosen.
Western blotting
Whole cell extracts were prepared by collecting cells and boiling them in NuPAGE LDS Sample Buffer (ThermoFisher). Extracts were resolved on 4–12% gradient Bis-Tris Protein Gels (ThermoFisher) and then transferred to Polyvinylidene difluoride (PVDF) membranes (GE Life Sciences). Before probing for specific proteins, membranes were blocked with 10% (w/v) non-fat dry milk in phosphate-buffered saline supplemented with 0.05% Tween-20 (PBS-T). Primary antibody incubation was performed for 2 h at room temperature with PBS-T supplemented with 5% non-fat dry milk. Membranes were washed with PBS-T and then incubated with secondary antibodies in the same conditions as the primary antibodies. For HRP-conjugated secondary antibodies, ECL Prime Western Blotting Detection Reagent (GE Life Sciences) was used for band detection as visualized with Hyperfilm ECL (GE Life Sciences). For IRDye-conjugates secondary antibodies, Odyssey Imaging Systems (Li-COR Biosciences) was used for band detection and visualization.
Antibodies
Generation of affinity-purified antibodies against γH2A (1:1000) from rabbit serum was previously described [84]. Commercial primary antibodies used here were: mouse anti-HA (1: 5000, Sigma), anti-EF1α (1: 40 000, Merck Millipore) and anti-BrdU clone B44 (1: 500, BD Bioscience). Commercial secondary antibodies used here were: goat anti-Rabbit IgG HRP-conjugated (ThermoFisher), goat anti-Rabbit IgG HRP-conjugated (ThermoFisher), goat anti-Rabbit IgG Alexa Fluor 488-conjugated (ThermoFisher), goat anti-Rabbit IgG IRDye 800CW-conjugated (Li-COR Biosciences) and goat anti-Mouse IgG IRDye 680CW-conjugated (Li-COR Biosciences).
Genome sequencing and bioinformatics analysis
Total DNA extraction was performed with a DNeasy Blood & Tissue Kit (QIAGEN) following the manufacturer’s instructions. Genomic DNA libraries were prepared using a Nextera NGS Library Preparation Kit (QIAGEN). Libraries were sequenced at Glasgow Polyomics (www.polyomics.gla.ac.uk/index.html), using a NextSeq 500 Illumina platform, generating paired end reads of 75 nucleotides. Processing of sequencing data was performed at the Galaxy web platform (usegalaxy.org)[85]. FastQC (http://www.bioinformatics.babraham.ac.uk/projects/fastqc/) and trimomatic [86] were used for quality control and adapter sequence removal, respectively. BWA-mem[87] was used to map processed reads to the reference genome (Leishmania major Friedlin v39, available at Tritrypdb - http://tritrypdb.org/tritrypdb/). Reads with mapping quality score < 30 were discarded using SAMtools [88]. Single nucleotide polymorphisms (SNPs) and InDels were detected using GATK [89] and freebayes [90]. Only those SNPs and InDels with DP of at least 5 and map quality 30 were considered. VCFtools was used to calculate SNP and InDel density, with SNPdensity function[91]. Heatmaps, violin plots and metaplots were generated using Prism Graphpad. Underlying data for metaplots and coverage tracks were generated using deepTools[92]. Mutational SNP signature analysis was performed as previously described [93].
Marker Frequency Analysis (MFAseq)
After processing, reads were compared essentially using methods described previously [39], though with modifications. Briefly, the number of reads in 0.5 kb windows along chromosomes was determined. The number of reads in each bin was then used to calculate the ratio between exponentially growing and stationary phase cells, scaled for the total size of the read library. Ratio values were converted into Z scores values in a 5 kb sliding window (steps of 500bp), for each individual chromosome. MFAseq profiles for each chromosome were represented in a graphical form using Gviz[94].
Detection of cells in S phase
Exponentially growing cells were incubated for 30 min with 150 μM IdU. Cells were collected by centrifugation and washed with 1x PBS. Fixation was performed at -20 ºC with a mixture (7:3) of ethanol and 1x PBS for at least 16 h. Then, cells were collected by centrifugation and rinsed with washing buffer (1x PBS supplemented with 1% BSA). DNA denaturation was performed for 30 minutes with 2N HCL and reaction was neutralized with phosphate buffer (0.2 M Na2HPO4, 0.2 M KH2PO4, pH 7.4). Cells were collected by centrifugation, further incubated in phosphate buffer for 30 mins at room temperature and centrifuged again. To detect incorporated IdU, cells were incubated for 1h at room temperature with anti-BrdU antibody (diluted in washing buffer supplemented with 0.2% Tween-20), collected by centrifugation and washed with washing buffer. Cells were incubated for 1 h at room temperature with anti-mouse secondary antibody conjugated with Alexa Fluor 488 (diluted in washing buffer supplemented with 0.2% Tween-20), collected by centrifugation and washed with washing buffer. Finally, cells were re-suspended in 1xPBS supplemented with 10 μg.mL-1 Propidium Iodide and 10 μg.mL-1 RNAse A and passed through a 35 μm nylon mesh. Data was acquired with FACSCelesta (BD Biosciences) and further analysed with FlowJo software. Negative control cells, in which anti-BrdU antibody was omitted during IdU detection step, were included in each experiment. Negative control cells were used to draw gates to discriminate positive and negative events.
Supporting information
S1 Table [tiff]
Primers used in this study.
S1 Fig [a]
Combining CRISPR/Cas9 and DiCre to rapidly generate cell lines for inducible knockout.
S2 Fig [tiff]
Growth profile analysis.
S3 Fig [a]
Analysis of RAD51 C-terminally tagged with HA.
S4 Fig [a]
Dynamics of KO induction.
S5 Fig [g1]
Analysis of DNA content profile upon prolonged cultivation after KO induction of homologous recombination factors.
S6 Fig [g1]
Cell cycle progression analysis after replication stress upon KO induction of homologous recombination factors.
S7 Fig [a]
Whole genome analysis of InDel accumulation patterns upon KO induction of single or combined homologous recombination factors.
S8 Fig [tiff]
SNP mutation signature upon KO of RAD51 related genes.
S9 Fig [a]
Genotoxic stress resistance profiles upon KO induction of RAD51 and RAD51-3.
S10 Fig [a]
Whole genome analysis of SNPs and InDel accumulation patterns after replication stress upon KO induction of RAD51 and RAD51-3.
S11 Fig [tiff]
SNP mutation signature upon replication stress after KO induction of RAD51 and RAD513.
S12 Fig [a]
Combining CRISPR/Cas9 and DiCre to rapidly generate double KO cell lines.
S13 Fig [a]
Dynamics of KO induction.
S14 Fig [a]
Effects of double KO of RAD51 paralogues.
S15 Fig [a]
Whole genome analysis of InDels accumulation patterns upon double KO of RAD51-3 and RAD51-4.
S16 Fig [tiff]
Genome-wide mapping of replication initiation upon RAD51 and RAD51-3 KO and HU treatment.
Zdroje
1. Scully R, Panday A, Elango R, Willis NA. DNA double-strand break repair-pathway choice in somatic mammalian cells. Nature reviews Molecular cell biology. 2019. doi: 10.1038/s41580-019-0152-0 31263220.
2. Bell JC, Kowalczykowski SC. RecA: Regulation and Mechanism of a Molecular Search Engine. Trends in biochemical sciences. 2016;41(6):491–507. doi: 10.1016/j.tibs.2016.04.002 27156117; PubMed Central PMCID: PMC4892382.
3. Sullivan MR, Bernstein KA. RAD-ical New Insights into RAD51 Regulation. Genes (Basel). 2018;9(12). doi: 10.3390/genes9120629 30551670; PubMed Central PMCID: PMC6316741.
4. Zhang S, Wang L, Tao Y, Bai T, Lu R, Zhang T, et al. Structural basis for the functional role of the Shu complex in homologous recombination. Nucleic Acids Res. 2017;45(22):13068–79. doi: 10.1093/nar/gkx992 29069504; PubMed Central PMCID: PMC5727457.
5. Gaines WA, Godin SK, Kabbinavar FF, Rao T, VanDemark AP, Sung P, et al. Promotion of presynaptic filament assembly by the ensemble of S. cerevisiae Rad51 paralogues with Rad52. Nature communications. 2015;6:7834. doi: 10.1038/ncomms8834 26215801; PubMed Central PMCID: PMC4525180.
6. Taylor MRG, Spirek M, Chaurasiya KR, Ward JD, Carzaniga R, Yu X, et al. Rad51 Paralogs Remodel Pre-synaptic Rad51 Filaments to Stimulate Homologous Recombination. Cell. 2015;162(2):271–86. doi: 10.1016/j.cell.2015.06.015 26186187; PubMed Central PMCID: PMC4518479.
7. Rosenbaum JC, Bonilla B, Hengel SR, Mertz TM, Herken BW, Kazemier HG, et al. The Rad51 paralogs facilitate a novel DNA strand specific damage tolerance pathway. Nature communications. 2019;10(1):3515. doi: 10.1038/s41467-019-11374-8 31383866; PubMed Central PMCID: PMC6683157.
8. Saxena S, Somyajit K, Nagaraju G. XRCC2 Regulates Replication Fork Progression during dNTP Alterations. Cell reports. 2018;25(12):3273–82 e6. doi: 10.1016/j.celrep.2018.11.085 30566856.
9. Pond KW, de Renty C, Yagle MK, Ellis NA. Rescue of collapsed replication forks is dependent on NSMCE2 to prevent mitotic DNA damage. PLoS genetics. 2019;15(2):e1007942. doi: 10.1371/journal.pgen.1007942 30735491; PubMed Central PMCID: PMC6383951.
10. Malacaria E, Pugliese GM, Honda M, Marabitti V, Aiello FA, Spies M, et al. Rad52 prevents excessive replication fork reversal and protects from nascent strand degradation. Nature communications. 2019;10(1):1412. doi: 10.1038/s41467-019-09196-9 30926821; PubMed Central PMCID: PMC6441034.
11. Bhat KP, Krishnamoorthy A, Dungrawala H, Garcin EB, Modesti M, Cortez D. RADX Modulates RAD51 Activity to Control Replication Fork Protection. Cell reports. 2018;24(3):538–45. doi: 10.1016/j.celrep.2018.06.061 30021152; PubMed Central PMCID: PMC6086571.
12. Petermann E, Orta ML, Issaeva N, Schultz N, Helleday T. Hydroxyurea-Stalled Replication Forks Become Progressively Inactivated and Require Two Different RAD51-Mediated Pathways for Restart and Repair. MolCell. 2010;37(4):492–502.
13. Somyajit K, Saxena S, Babu S, Mishra A, Nagaraju G. Mammalian RAD51 paralogs protect nascent DNA at stalled forks and mediate replication restart. Nucleic Acids Res. 2015;43(20):9835–55. doi: 10.1093/nar/gkv880 26354865; PubMed Central PMCID: PMC4787763.
14. Kogoma T, von Meyenburg K. The origin of replication, oriC, and the dnaA protein are dispensable in stable DNA replication (sdrA) mutants of Escherichia coli K-12. The EMBO journal. 1983;2(3):463–8. 11894964; PubMed Central PMCID: PMC555155.
15. Asai T, Sommer S, Bailone A, Kogoma T. Homologous recombination-dependent initiation of DNA replication from DNA damage-inducible origins in Escherichia coli. The EMBO journal. 1993;12(8):3287–95. 8344265; PubMed Central PMCID: PMC413596.
16. Asai T, Kogoma T. D-loops and R-loops: alternative mechanisms for the initiation of chromosome replication in Escherichia coli. Journal of bacteriology. 1994;176(7):1807–12. doi: 10.1128/jb.176.7.1807-1812.1994 8144445; PubMed Central PMCID: PMC205281.
17. Kogoma T. Stable DNA replication: interplay between DNA replication, homologous recombination, and transcription. Microbiology and molecular biology reviews: MMBR. 1997;61(2):212–38. 9184011; PubMed Central PMCID: PMC232608.
18. Hawkins M, Malla S, Blythe MJ, Nieduszynski CA, Allers T. Accelerated growth in the absence of DNA replication origins. Nature. 2013;503(7477):544–7. doi: 10.1038/nature12650 24185008; PubMed Central PMCID: PMC3843117.
19. Piazza A, Heyer WD. Homologous Recombination and the Formation of Complex Genomic Rearrangements. Trends in cell biology. 2019;29(2):135–49. doi: 10.1016/j.tcb.2018.10.006 30497856; PubMed Central PMCID: PMC6402879.
20. Lee CS, Haber JE. Mating-type Gene Switching in Saccharomyces cerevisiae. Microbiology spectrum. 2015;3(2):MDNA3-0013-2014. doi: 10.1128/microbiolspec.MDNA3-0013-2014 26104712.
21. McCulloch R, Morrison LJ, Hall JP. DNA Recombination Strategies During Antigenic Variation in the African Trypanosome. Microbiology spectrum. 2015;3(2):MDNA3-0016–2014. doi: 10.1128/microbiolspec.MDNA3-0016-2014 26104717.
22. Trenaman A, Hartley C, Prorocic M, Passos-Silva DG, van den Hoek M, Nechyporuk-Zloy V, et al. Trypanosoma brucei BRCA2 acts in a life cycle-specific genome stability process and dictates BRC repeat number-dependent RAD51 subnuclear dynamics. Nucleic Acids Res. 2013;41(2):943–60. doi: 10.1093/nar/gks1192 23222131; PubMed Central PMCID: PMC3553974.
23. Hartley CL, McCulloch R. Trypanosoma brucei BRCA2 acts in antigenic variation and has undergone a recent expansion in BRC repeat number that is important during homologous recombination. MolMicrobiol. 2008;68(5):1237–51.
24. Moraes Barros RR, Marini MM, Antonio CR, Cortez DR, Miyake AM, Lima FM, et al. Anatomy and evolution of telomeric and subtelomeric regions in the human protozoan parasite Trypanosoma cruzi. BMC Genomics. 2012;13:229. doi: 10.1186/1471-2164-13-229 22681854; PubMed Central PMCID: PMC3418195.
25. Weatherly DB, Peng D, Tarleton RL. Recombination-driven generation of the largest pathogen repository of antigen variants in the protozoan Trypanosoma cruzi. BMC Genomics. 2016;17(1):729. doi: 10.1186/s12864-016-3037-z 27619017; PubMed Central PMCID: PMC5020489.
26. Alves CL, Repoles BM, da Silva MS, Mendes IC, Marin PA, Aguiar PHN, et al. The recombinase Rad51 plays a key role in events of genetic exchange in Trypanosoma cruzi. Scientific reports. 2018;8(1):13335. doi: 10.1038/s41598-018-31541-z 30190603; PubMed Central PMCID: PMC6127316.
27. Laffitte MC, Leprohon P, Hainse M, Legare D, Masson JY, Ouellette M. Chromosomal Translocations in the Parasite Leishmania by a MRE11/RAD50-Independent Microhomology-Mediated End Joining Mechanism. PLoS genetics. 2016;12(6):e1006117. doi: 10.1371/journal.pgen.1006117 27314941; PubMed Central PMCID: PMC4912120.
28. Laffitte MN, Leprohon P, Papadopoulou B, Ouellette M. Plasticity of the Leishmania genome leading to gene copy number variations and drug resistance. F1000Res. 2016;5:2350. doi: 10.12688/f1000research.9218.1 27703673; PubMed Central PMCID: PMC5031125.
29. Genois MM, Plourde M, Ethier C, Roy G, Poirier GG, Ouellette M, et al. Roles of Rad51 paralogs for promoting homologous recombination in Leishmania infantum. Nucleic Acids Res. 2015;43(5):2701–15. doi: 10.1093/nar/gkv118 25712090; PubMed Central PMCID: PMC4357719.
30. Ubeda JM, Raymond F, Mukherjee A, Plourde M, Gingras H, Roy G, et al. Genome-wide stochastic adaptive DNA amplification at direct and inverted DNA repeats in the parasite Leishmania. PLoS biology. 2014;12(5):e1001868. doi: 10.1371/journal.pbio.1001868 24844805; PubMed Central PMCID: PMC4028189.
31. Laffitte MC, Genois MM, Mukherjee A, Legare D, Masson JY, Ouellette M. Formation of linear amplicons with inverted duplications in Leishmania requires the MRE11 nuclease. PLoS genetics. 2014;10(12):e1004805. doi: 10.1371/journal.pgen.1004805 25474106; PubMed Central PMCID: PMC4256157.
32. Genois MM, Paquet ER, Laffitte MC, Maity R, Rodrigue A, Ouellette M, et al. DNA repair pathways in trypanosomatids: from DNA repair to drug resistance. Microbiology and molecular biology reviews: MMBR. 2014;78(1):40–73. doi: 10.1128/MMBR.00045-13 24600040; PubMed Central PMCID: PMC3957735.
33. Lachaud L, Bourgeois N, Kuk N, Morelle C, Crobu L, Merlin G, et al. Constitutive mosaic aneuploidy is a unique genetic feature widespread in the Leishmania genus. Microbes and infection / Institut Pasteur. 2014;16(1):61–6. doi: 10.1016/j.micinf.2013.09.005 24120456.
34. Bussotti G, Gouzelou E, Cortes Boite M, Kherachi I, Harrat Z, Eddaikra N, et al. Leishmania Genome Dynamics during Environmental Adaptation Reveal Strain-Specific Differences in Gene Copy Number Variation, Karyotype Instability, and Telomeric Amplification. mBio. 2018;9(6). doi: 10.1128/mBio.01399-18 30401775; PubMed Central PMCID: PMC6222132.
35. Dumetz F, Imamura H, Sanders M, Seblova V, Myskova J, Pescher P, et al. Modulation of Aneuploidy in Leishmania donovani during Adaptation to Different In Vitro and In Vivo Environments and Its Impact on Gene Expression. mBio. 2017;8(3). doi: 10.1128/mBio.00599-17 28536289; PubMed Central PMCID: PMC5442457.
36. Prieto Barja P, Pescher P, Bussotti G, Dumetz F, Imamura H, Kedra D, et al. Haplotype selection as an adaptive mechanism in the protozoan pathogen Leishmania donovani. Nat Ecol Evol. 2017;1(12):1961–9. doi: 10.1038/s41559-017-0361-x 29109466.
37. Zackay A, Cotton JA, Sanders M, Hailu A, Nasereddin A, Warburg A, et al. Genome wide comparison of Ethiopian Leishmania donovani strains reveals differences potentially related to parasite survival. PLoS genetics. 2018;14(1):e1007133. doi: 10.1371/journal.pgen.1007133 29315303; PubMed Central PMCID: PMC5777657.
38. Ubeda JM, Legare D, Raymond F, Ouameur AA, Boisvert S, Rigault P, et al. Modulation of gene expression in drug resistant Leishmania is associated with gene amplification, gene deletion and chromosome aneuploidy. Genome biology. 2008;9(7):R115. doi: 10.1186/gb-2008-9-7-r115 18638379; PubMed Central PMCID: PMC2530873.
39. Marques CA, Dickens NJ, Paape D, Campbell SJ, McCulloch R. Genome-wide mapping reveals single-origin chromosome replication in Leishmania, a eukaryotic microbe. Genome biology. 2015;16(1):230. doi: 10.1186/s13059-015-0788-9 26481451.
40. Lombrana R, Alvarez A, Fernandez-Justel JM, Almeida R, Poza-Carrion C, Gomes F, et al. Transcriptionally Driven DNA Replication Program of the Human Parasite Leishmania major. Cell reports. 2016;16(6):1774–86. doi: 10.1016/j.celrep.2016.07.007 27477279.
41. Marques CA, McCulloch R. Conservation and Variation in Strategies for DNA Replication of Kinetoplastid Nuclear Genomes. Current genomics. 2018;19(2):98–109. doi: 10.2174/1389202918666170815144627 29491738; PubMed Central PMCID: PMC5814967.
42. McKean PG, Keen JK, Smith DF, Benson FE. Identification and characterisation of a RAD51 gene from Leishmania major. MolBiochemParasitol. 2001;115(2):209–16.
43. Genois MM, Mukherjee A, Ubeda JM, Buisson R, Paquet E, Roy G, et al. Interactions between BRCA2 and RAD51 for promoting homologous recombination in Leishmania infantum. Nucleic Acids Res. 2012;40(14):6570–84. doi: 10.1093/nar/gks306 22505581; PubMed Central PMCID: PMC3413117.
44. Proudfoot C, McCulloch R. Distinct roles for two RAD51-related genes in Trypanosoma brucei antigenic variation. Nucleic Acids Res. 2005;33(21):6906–19. doi: 10.1093/nar/gki996 16326865
45. Proudfoot C, McCulloch R. Trypanosoma brucei DMC1 does not act in DNA recombination, repair or antigenic variation in bloodstream stage cells. MolBiochemParasitol. 2006;145(2):245–53.
46. Dobson R, Stockdale C, Lapsley C, Wilkes J, McCulloch R. Interactions among Trypanosoma brucei RAD51 paralogues in DNA repair and antigenic variation. MolMicrobiol. 2011;81(2):434–56.
47. Peacock L, Ferris V, Sharma R, Sunter J, Bailey M, Carrington M, et al. Identification of the meiotic life cycle stage of Trypanosoma brucei in the tsetse fly. Proceedings of the National Academy of Sciences of the United States of America. 2011;108(9):3671–6. doi: 10.1073/pnas.1019423108 21321215; PubMed Central PMCID: PMC3048101.
48. Zhang WW, Lypaczewski P, Matlashewski G. Optimized CRISPR-Cas9 Genome Editing for Leishmania and Its Use To Target a Multigene Family, Induce Chromosomal Translocation, and Study DNA Break Repair Mechanisms. mSphere. 2017;2(1). doi: 10.1128/mSphere.00340-16 28124028; PubMed Central PMCID: PMC5244264.
49. McCulloch R, Barry JD. A role for RAD51 and homologous recombination in Trypanosoma brucei antigenic variation. Genes & development. 1999;13(21):2875–88. doi: 10.1101/gad.13.21.2875 10557214; PubMed Central PMCID: PMC317127.
50. Liu B, Wang J, Yaffe N, Lindsay ME, Zhao Z, Zick A, et al. Trypanosomes have six mitochondrial DNA helicases with one controlling kinetoplast maxicircle replication. MolCell. 2009;35(4):490–501.
51. Byrd AK, Raney KD. Structure and function of Pif1 helicase. Biochemical Society transactions. 2017;45(5):1159–71. doi: 10.1042/BST20170096 28900015; PubMed Central PMCID: PMC5870758.
52. Deegan TD, Baxter J, Ortiz Bazan MA, Yeeles JTP, Labib KPM. Pif1-Family Helicases Support Fork Convergence during DNA Replication Termination in Eukaryotes. Molecular cell. 2019;74(2):231–44 e9. doi: 10.1016/j.molcel.2019.01.040 30850330; PubMed Central PMCID: PMC6477153.
53. Glover L, Horn D. Trypanosomal histone gammaH2A and the DNA damage response. MolBiochemParasitol. 2012;183(1):78–83.
54. Tiengwe C, Marcello L, Farr H, Dickens N, Kelly S, Swiderski M, et al. Genome-wide analysis reveals extensive functional interaction between DNA replication initiation and transcription in the genome of Trypanosoma brucei. Cell reports. 2012;2(1):185–97. doi: 10.1016/j.celrep.2012.06.007 22840408; PubMed Central PMCID: PMC3607257.
55. Muller CA, Nieduszynski CA. Conservation of replication timing reveals global and local regulation of replication origin activity. Genome research. 2012;22(10):1953–62. doi: 10.1101/gr.139477.112 22767388; PubMed Central PMCID: PMC3460190.
56. Conway C, Proudfoot C, Burton P, Barry JD, McCulloch R. Two pathways of homologous recombination in Trypanosoma brucei. Molecular microbiology. 2002;45(6):1687–700. doi: 10.1046/j.1365-2958.2002.03122.x 12354234.
57. Glover L, Jun J, Horn D. Microhomology-mediated deletion and gene conversion in African trypanosomes. Nucleic Acids Res. 2011;39(4):1372–80. doi: 10.1093/nar/gkq981 20965968
58. Glover L, McCulloch R, Horn D. Sequence homology and microhomology dominate chromosomal double-strand break repair in African trypanosomes. Nucleic Acids Res. 2008;36(8):2608–18. doi: 10.1093/nar/gkn104 18334531; PubMed Central PMCID: PMC2377438.
59. Zhang WW, Matlashewski G. CRISPR-Cas9-Mediated Genome Editing in Leishmania donovani. mBio. 2015;6(4):e00861. doi: 10.1128/mBio.00861-15 26199327; PubMed Central PMCID: PMC4513079.
60. Sollelis L, Ghorbal M, MacPherson CR, Martins RM, Kuk N, Crobu L, et al. First efficient CRISPR-Cas9-mediated genome editing in Leishmania parasites. Cellular microbiology. 2015;17(10):1405–12. doi: 10.1111/cmi.12456 25939677.
61. Lander N, Li ZH, Niyogi S, Docampo R. CRISPR/Cas9-Induced Disruption of Paraflagellar Rod Protein 1 and 2 Genes in Trypanosoma cruzi Reveals Their Role in Flagellar Attachment. mBio. 2015;6(4):e01012. doi: 10.1128/mBio.01012-15 26199333; PubMed Central PMCID: PMC4513075.
62. Peng D, Kurup SP, Yao PY, Minning TA, Tarleton RL. CRISPR-Cas9-mediated single-gene and gene family disruption in Trypanosoma cruzi. mBio. 2015;6(1):e02097–14. doi: 10.1128/mBio.02097-14 25550322; PubMed Central PMCID: PMC4281920.
63. Iantorno SA, Durrant C, Khan A, Sanders MJ, Beverley SM, Warren WC, et al. Gene Expression in Leishmania Is Regulated Predominantly by Gene Dosage. mBio. 2017;8(5). doi: 10.1128/mBio.01393-17 28900023; PubMed Central PMCID: PMC5596349.
64. Costantino L, Koshland D. Genome-wide Map of R-Loop-Induced Damage Reveals How a Subset of R-Loops Contributes to Genomic Instability. Molecular cell. 2018;71(4):487–97 e3. doi: 10.1016/j.molcel.2018.06.037 30078723; PubMed Central PMCID: PMC6264797.
65. Briggs E, Hamilton G, Crouch K, Lapsley C, McCulloch R. Genome-wide mapping reveals conserved and diverged R-loop activities in the unusual genetic landscape of the African trypanosome genome. Nucleic Acids Res. 2018;46(22):11789–805. doi: 10.1093/nar/gky928 30304482; PubMed Central PMCID: PMC6294496.
66. Kim HS. Genome-wide function of MCM-BP in Trypanosoma brucei DNA replication and transcription. Nucleic Acids Res. 2019;47(2):634–47. doi: 10.1093/nar/gky1088 30407533; PubMed Central PMCID: PMC6344857.
67. Santosa V, Martha S, Hirose N, Tanaka K. The fission yeast minichromosome maintenance (MCM)-binding protein (MCM-BP), Mcb1, regulates MCM function during prereplicative complex formation in DNA replication. The Journal of biological chemistry. 2013;288(10):6864–80. doi: 10.1074/jbc.M112.432393 23322785; PubMed Central PMCID: PMC3591596.
68. Nishiyama A, Frappier L, Mechali M. MCM-BP regulates unloading of the MCM2-7 helicase in late S phase. Genes & development. 2011;25(2):165–75. doi: 10.1101/gad.614411 21196493; PubMed Central PMCID: PMC3022262.
69. Marques CA, Tiengwe C, Lemgruber L, Damasceno JD, Scott A, Paape D, et al. Diverged composition and regulation of the Trypanosoma brucei origin recognition complex that mediates DNA replication initiation. Nucleic Acids Res. 2016;44(10):4763–84. doi: 10.1093/nar/gkw147 26951375; PubMed Central PMCID: PMC4889932.
70. Recombination Mosig G. and recombination-dependent DNA replication in bacteriophage T4. Annual review of genetics. 1998;32:379–413. doi: 10.1146/annurev.genet.32.1.379 9928485.
71. Mosig G, Gewin J, Luder A, Colowick N, Vo D. Two recombination-dependent DNA replication pathways of bacteriophage T4, and their roles in mutagenesis and horizontal gene transfer. Proceedings of the National Academy of Sciences of the United States of America. 2001;98(15):8306–11. doi: 10.1073/pnas.131007398 11459968; PubMed Central PMCID: PMC37436.
72. Wilkinson DE, Weller SK. The role of DNA recombination in herpes simplex virus DNA replication. IUBMB Life. 2003;55(8):451–8. doi: 10.1080/15216540310001612237 14609200.
73. Lee PH, Meng X, Kapler GM. Developmental Regulation of the Tetrahymena thermophila Origin Recognition Complex. PLoS genetics. 2015;11(1):e1004875. doi: 10.1371/journal.pgen.1004875 25569357; PubMed Central PMCID: PMC4287346.
74. Natsume T, Nishimura K, Minocherhomji S, Bhowmick R, Hickson ID, Kanemaki MT. Acute inactivation of the replicative helicase in human cells triggers MCM8-9-dependent DNA synthesis. Genes & development. 2017;31(8):816–29. doi: 10.1101/gad.297663.117 28487407; PubMed Central PMCID: PMC5435893.
75. Akiyoshi B, Gull K. Discovery of unconventional kinetochores in kinetoplastids. Cell. 2014;156(6):1247–58. doi: 10.1016/j.cell.2014.01.049 24582333; PubMed Central PMCID: PMC3978658.
76. Garcia-Silva MR, Sollelis L, MacPherson CR, Stanojcic S, Kuk N, Crobu L, et al. Identification of the centromeres of Leishmania major: revealing the hidden pieces. EMBO reports. 2017;18(11):1968–77. doi: 10.15252/embr.201744216 28935715; PubMed Central PMCID: PMC5666652.
77. Forsburg SL, Shen KF. Centromere Stability: The Replication Connection. Genes (Basel). 2017;8(1). doi: 10.3390/genes8010037 28106789; PubMed Central PMCID: PMC5295031.
78. Onaka AT, Toyofuku N, Inoue T, Okita AK, Sagawa M, Su J, et al. Rad51 and Rad54 promote noncrossover recombination between centromere repeats on the same chromatid to prevent isochromosome formation. Nucleic Acids Res. 2016;44(22):10744–57. doi: 10.1093/nar/gkw874 27697832; PubMed Central PMCID: PMC5159554.
79. McFarlane RJ, Humphrey TC. A role for recombination in centromere function. Trends in genetics: TIG. 2010;26(5):209–13. doi: 10.1016/j.tig.2010.02.005 20382440.
80. Santos R, Silva GLA, Santos EV, Duncan SM, Mottram JC, Damasceno JD, et al. A DiCre recombinase-based system for inducible expression in Leishmania major. Molecular and biochemical parasitology. 2017;216:45–8. doi: 10.1016/j.molbiopara.2017.06.006 28629935.
81. Beneke T, Madden R, Makin L, Valli J, Sunter J, Gluenz E. A CRISPR Cas9 high-throughput genome editing toolkit for kinetoplastids. R Soc Open Sci. 2017;4(5):170095. doi: 10.1098/rsos.170095 28573017; PubMed Central PMCID: PMC5451818.
82. Damasceno JD, Obonaga R, Silva GLA, Reis-Cunha JL, Duncan SM, Bartholomeu DC, et al. Conditional genome engineering reveals canonical and divergent roles for the Hus1 component of the 9-1-1 complex in the maintenance of the plastic genome of Leishmania. Nucleic Acids Res. 2018;46(22):11835–46. doi: 10.1093/nar/gky1017 30380080; PubMed Central PMCID: PMC6294564.
83. Duncan SM, Myburgh E, Philipon C, Brown E, Meissner M, Brewer J, et al. Conditional gene deletion with DiCre demonstrates an essential role for CRK3 in Leishmania mexicana cell cycle regulation. Molecular microbiology. 2016;100(6):931–44. doi: 10.1111/mmi.13375 26991545; PubMed Central PMCID: PMC4913733.
84. Stortz JA, Serafim TD, Alsford S, Wilkes J, Fernandez-Cortes F, Hamilton G, et al. Genome-wide and protein kinase-focused RNAi screens reveal conserved and novel damage response pathways in Trypanosoma brucei. PLoS pathogens. 2017;13(7):e1006477. doi: 10.1371/journal.ppat.1006477 28742144; PubMed Central PMCID: PMC5542689.
85. Afgan E, Baker D, Batut B, van den Beek M, Bouvier D, Cech M, et al. The Galaxy platform for accessible, reproducible and collaborative biomedical analyses: 2018 update. Nucleic Acids Res. 2018;46(W1):W537–W44. doi: 10.1093/nar/gky379 29790989; PubMed Central PMCID: PMC6030816.
86. Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics. 2014;30(15):2114–20. Epub 2014/04/04. doi: 10.1093/bioinformatics/btu170 24695404; PubMed Central PMCID: PMC4103590.
87. Li H, Durbin R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics. 2009;25(14):1754–60. Epub 2009/05/20. doi: 10.1093/bioinformatics/btp324 19451168; PubMed Central PMCID: PMC2705234.
88. Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, et al. The Sequence Alignment/Map format and SAMtools. Bioinformatics. 2009;25(16):2078–9. Epub 2009/06/10. doi: 10.1093/bioinformatics/btp352 19505943; PubMed Central PMCID: PMC2723002.
89. McKenna A, Hanna M, Banks E, Sivachenko A, Cibulskis K, Kernytsky A, et al. The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome Res. 2010;20(9):1297–303. Epub 2010/07/21. doi: 10.1101/gr.107524.110 20644199; PubMed Central PMCID: PMC2928508.
90. Garrison E, Marth G. Haplotype-based variant detection from short-read sequencing. arXiv e-prints [Internet]. 2012 July 01, 2012. Available from: https://ui.adsabs.harvard.edu/abs/2012arXiv1207.3907G.
91. Danecek P, Auton A, Abecasis G, Albers CA, Banks E, DePristo MA, et al. The variant call format and VCFtools. Bioinformatics. 2011;27(15):2156–8. Epub 2011/06/10. doi: 10.1093/bioinformatics/btr330 21653522; PubMed Central PMCID: PMC3137218.
92. Ramirez F, Ryan DP, Gruning B, Bhardwaj V, Kilpert F, Richter AS, et al. deepTools2: a next generation web server for deep-sequencing data analysis. Nucleic Acids Res. 2016;44(W1):W160–5. Epub 2016/04/16. doi: 10.1093/nar/gkw257 27079975; PubMed Central PMCID: PMC4987876.
93. Nik-Zainal S, Davies H, Staaf J, Ramakrishna M, Glodzik D, Zou X, et al. Landscape of somatic mutations in 560 breast cancer whole-genome sequences. Nature. 2016;534(7605):47–54. Epub 2016/05/03. doi: 10.1038/nature17676 27135926; PubMed Central PMCID: PMC4910866.
94. Hahne F, Ivanek R. Visualizing Genomic Data Using Gviz and Bioconductor. Methods Mol Biol. 2016;1418:335–51. Epub 2016/03/24. doi: 10.1007/978-1-4939-3578-9_16 27008022.
Článek vyšel v časopise
PLOS Genetics
2020 Číslo 7
- Prof. Libor Grubhoffer: Nová varianta mpox je příležitostí pokročit v poznání původce varioly
- Co obsahují nové pokyny WHO pro dozor nad antibiotickou rezistencí?
- AI místo lékaře? Výzkum zkoumal ochotu Čechů konzultovat své zdraví s umělou inteligencí
- Prof. Jan Škrha: Metformin je bezpečný, ale je třeba jej bezpečně užívat a léčbu kontrolovat
- Ibuprofen jako alternativa antibiotik při léčbě infekcí močových cest
Nejčtenější v tomto čísle
- Holocentric chromosomes
- Repression of tick microRNA-133 induces organic anion transporting polypeptide expression critical for Anaplasma phagocytophilum survival in the vector and transmission to the vertebrate host
- Brassinosteroids regulate root meristem development by mediating BIN2-UPB1 module in Arabidopsis
- Genetic and metabolomic architecture of variation in diet restriction-mediated lifespan extension in Drosophila