C-Terminal Flap Endonuclease (rad27) Mutations: Lethal Interactions With a DNA Ligase I Mutation (cdc9-p) and Suppression by Proliferating Cell Nuclear Antigen (POL30) in Saccharomyces cerevisiae

During lagging-strand DNA replication in eukaryotic cells primers are removed from Okazaki fragments by the flap endonuclease and DNA ligase I joins nascent fragments. Both enzymes are brought to the replication fork by the sliding clamp proliferating cell nuclear antigen (PCNA). To understand the relationship among these three components, we have carried out a synthetic lethal screen with cdc9-p, a DNA ligase mutation with two substitutions (F43A/F44A) in its PCNA interaction domain. We recovered the flap endonuclease mutation rad27-K325* with a stop codon at residue 325. We created two additional rad27 alleles, rad27-A358* with a stop codon at residue 358 and rad27-pX8 with substitutions of all eight residues of the PCNA interaction domain. rad27-pX8 is temperature lethal and rad27-A358* grows slowly in combination with cdc9-p. Tests of mutation avoidance, DNA repair, and compatibility with DNA repair mutations showed that rad27-K325* confers severe phenotypes similar to rad27Δ, rad27-A358* confers mild phenotypes, and rad27-pX8 confers phenotypes intermediate between the other two alleles. High-copy expression of POL30 (PCNA) suppresses the canavanine mutation rate of all the rad27 alleles, including rad27Δ. These studies show the importance of the C terminus of the flap endonuclease in DNA replication and repair and, by virtue of the initial screen, show that this portion of the enzyme helps coordinate the entry of DNA ligase during Okazaki fragment maturation.CELLULAR maintenance of genomic integrity is essential for the continued viability of all organisms. The fidelity of DNA replication has to be maintained and DNA insults have to be repaired to ensure that deleterious mutations are not passed on to progeny or cause cancerous growth. A number of cellular proteins have multiple roles in DNA replication, mutation avoidance, and repair. In Saccharomyces cerevisiae, the flap endonuclease, proliferating cell nuclear antigen (PCNA), and DNA ligase I encoded by RAD27, POL30, and CDC9, respectively, are all required for proper replication and also function to avoid mutation and to facilitate repair.The flap endonuclease, FEN-1 in humans, is a highly conserved structure-specific nuclease that has both endonuclease and 5′–3′ exonuclease activity. During lagging-strand replication these activities function to remove primers from Okazaki fragments, either by endonucleolytic cleavage of a flap made by strand displacement (Liu et al. 2004) or by sequential exonucleolytic removal of single nucleotides at the 5′ end of the primer (Murante et al. 1994).While deletion of RAD27 is not lethal to yeast cells, the rad27Δ mutant exhibits temperature-sensitive growth, is a mutator, and undergoes genomic instability (Johnson et al. 1995; Reagan et al. 1995; Tishkoff et al. 1997b; Chen and Kolodner 1999). In addition, its sensitivity to low doses of the methylating agent methylmethane sulfonate (MMS) implicates the participation of the enzyme in base excision repair (BER) (Reagan et al. 1995; Wu and Wang 1999). rad27Δ mutants have been reported to be either mildly sensitive to UV light or not sensitive to UV light (Reagan et al. 1995; Sommers et al. 1995). In the strain background that the mutant is mildly sensitive, its combination with rad2Δ yields a double mutant more sensitive than each single mutant, implying that the enzyme does not participate in RAD2-mediated nucleotide excision repair (NER) (Reagan et al. 1995). The flap endonuclease has also been implicated in double-strand break (DSB) repair by virtue of the incompatibility of rad27Δ with mutations of the DSB repair pathways (Tishkoff et al. 1997b; Symington 1998). In addition, either the yeast enzyme or its human ortholog has been shown to participate in reactions of homologous recombination, nonhomologous end joining, and telomere maintenance (Parenteau and Wellinger 1999, 2002; Wu et al. 1999; Wang et al. 2004; Kikuchi et al. 2005). Curiously, the rad27Δ mutant is not sensitive to gamma radiation but is sensitive to high doses of MMS that are thought to act as a radiomimetic agent (Reagan et al. 1995; Sommers et al. 1995).PCNA is the replicative clamp that acts as a scaffold to facilitate the loading of DNA replication and repair proteins, including DNA ligase I and the flap endonuclease to DNA (Warbrick 2000, 2006; Maga and Hubscher 2003). PCNA (POL30) is essential for cell viability, which is indicative of its central role in DNA metabolism. Biochemical characterization of its effect on the flap endonuclease shows that it stimulates its activity ∼50-fold, evidencing the productive nature of the interaction (Gomes and Burgers 2000; Tom et al. 2000; Frank et al. 2001; Stucki et al. 2001). The ability of DNA ligase to efficiently catalyze the formation of phosphodiester bonds in the DNA backbone may also be facilitated by its binding to PCNA. Tom et al. (2001) showed that, in vitro, PCNA enhances the ligation reaction 5-fold and that the stable association of DNA ligase with nicked duplex DNA requires PCNA.Both DNA ligase and the flap endonuclease bind to PCNA via their respective PCNA interactive peptide domains (PIP box). The PIP box is a conserved sequence motif of the amino acids QXXLXXFF. The PIP box fits into the interdomain connector loop (IDCL) of PCNA to provide a protein–protein interaction surface (Gomes and Burgers 2000; Chapados et al. 2004; Sakurai et al. 2005; Pascal et al. 2006). Mutations in the PIP box or the IDCL that impair the interaction of DNA ligase and the flap endonuclease to PCNA lead to genomic instability (Amin and Holm 1996; Eissenberg et al. 1997; Gary et al. 1999; Refsland and Livingston 2005; Subramanian et al. 2005). We have reported that the double mutants made by combinations of cdc9-p, rad27-p, and pol30-90—mutations with alterations of the PIP box or the IDCL in the respective proteins—have synergistic phenotypes with respect to MMS sensitivity and to trinucleotide repeat instability (Refsland and Livingston 2005). These results suggest that the two enzymes function in a concerted manner that is facilitated by PCNA.The precise nature of how PCNA coordinates the entry of the flap endonuclease and DNA ligase into the replication fork is not well understood. Biochemical and structural studies have begun to elucidate a possible ordering of these PCNA-mediated interactions. The possibility of such an ordering is underscored by the observation that DNA ligase adopts a toroidal conformation by completely encircling duplex DNA while interacting with PCNA (Pascal et al. 2004). Moreover, both PCNA and DNA ligase may be loaded onto the DNA in a mechanism utilizing the replication clamp loader replication factor C (RFC) (Levin et al. 2004; Vijayakumar et al. 2009), again suggesting a complete encirclement of the DNA by DNA ligase as well as by PCNA. PCNA and DNA ligase are similar in size and their interaction is likely to extend along the face of PCNA in a manner that would prevent other proteins such as the flap endonuclease from binding to the IDCL (Pascal et al. 2004, 2006). A biochemical study with purified yeast proteins showed that the two enzymes cannot bind simultaneously to PCNA (Subramanian et al. 2005). These studies suggest that a coordinated sequential interaction among PCNA, DNA ligase, and the flap endonuclease is important for replication and repair.Alternatively, both the flap endonuclease and DNA ligase may bind to the same molecule of PCNA. Since PCNA is a homotrimer, DNA ligase can potentially bind to one monomer while the flap endonuclease binds to another, using its extended C-terminal tail in a conformation allowing it to be tethered to PCNA concurrently with DNA ligase (Gomes and Burgers 2000; Sakurai et al. 2005). DNA ligase could also bind to PCNA in an extended conformation while the flap endonuclease cleaves the DNA. Sulfolobus solfataricus DNA ligase has been shown to have an open, extended conformation while binding to PCNA (Pascal et al. 2006). Presumably, once the flap endonuclease has removed the 5′ flap, DNA ligase acquires a closed, ring-shaped conformation to catalyze the joining of Okazaki fragments (Pascal et al. 2006).Exactly how the interaction of these enzymes with PCNA is coordinated in vivo, whether singly or concurrently, is not well understood. To further elucidate how the interaction of DNA ligase with PCNA is ordered, we performed a genetic screen to identify mutations that are synthetically lethal with cdc9-p (F44A/F35A), an allele of DNA ligase that has impaired binding to PCNA (Refsland and Livingston 2005; Subramanian et al. 2005). We postulated that genes recovered from this screen would function in DNA repair, replication, and recombination or would be involved in ordering the DNA ligase–PCNA interaction. From the screen we recovered a truncated allele of RAD27, rad27-K325*. This allele encodes a protein that lacks the PIP box and the entire C-terminal domain of the enzyme but retains the N terminus containing the nuclease activities. We have characterized this allele and compared it to two other rad27 alleles in which we have created different alterations of the C-terminal end of the flap endonuclease.     

Lambda Red Recombineering in Escherichia coli Occurs Through a Fully Single-Stranded Intermediate

The phage lambda-derived Red recombination system is a powerful tool for making targeted genetic changes in Escherichia coli, providing a simple and versatile method for generating insertion, deletion, and point mutations on chromosomal, plasmid, or BAC targets. However, despite the common use of this system, the detailed mechanism by which lambda Red mediates double-stranded DNA recombination remains uncertain. Current mechanisms posit a recombination intermediate in which both 5′ ends of double-stranded DNA are recessed by λ exonuclease, leaving behind 3′ overhangs. Here, we propose an alternative in which lambda exonuclease entirely degrades one strand, while leaving the other strand intact as single-stranded DNA. This single-stranded intermediate then recombines via beta recombinase-catalyzed annealing at the replication fork. We support this by showing that single-stranded gene insertion cassettes are recombinogenic and that these cassettes preferentially target the lagging strand during DNA replication. Furthermore, a double-stranded DNA cassette containing multiple internal mismatches shows strand-specific mutations cosegregating roughly 80% of the time. These observations are more consistent with our model than with previously proposed models. Finally, by using phosphorothioate linkages to protect the lagging-targeting strand of a double-stranded DNA cassette, we illustrate how our new mechanistic knowledge can be used to enhance lambda Red recombination frequency. The mechanistic insights revealed by this work may facilitate further improvements to the versatility of lambda Red recombination.OVER the past decade, lambda Red recombination (“recombineering”) has been used as a powerful technique for making precisely defined insertions, deletions, and point mutations in Escherichia coli, requiring as few as 35 bp of homology on each side of the desired alteration (Thomason et al. 2007a; Sharan et al. 2009). With this system, single-stranded DNA (ssDNA) oligonucleotides have been used to efficiently modify E. coli chromosomal targets (Ellis et al. 2001; Costantino and Court 2003), BACs (Swaminathan et al. 2001), and plasmids (Thomason et al. 2007b), as well as to rapidly optimize a metabolic pathway coding for the production of lycopene (Wang et al. 2009). Furthermore, linear double-stranded DNA (dsDNA) recombineering has been used to replace chromosomal genes (Murphy 1998; Murphy et al. 2000), to disrupt gene function (Datsenko and Wanner 2000), and to develop novel cloning methods (Lee et al. 2001; Li and Elledge 2005). Large-scale dsDNA recombineering projects include creating a library of single-gene knockout E. coli strains (Baba et al. 2006) and removing 15% of the genomic material from a single E. coli strain (Posfai et al. 2006). Linear dsDNA recombineering has also been used to insert heterologous genes and entire pathways into the E. coli chromosome (Zhang et al. 1998; Wang and Pfeifer 2008) and BACs (Lee et al. 2001; Warming et al. 2005), including those used for downstream applications in eukaryotes (Chaveroche et al. 2000; Bouvier and Cheng 2009). However, despite the broad use of this method, the mechanism of lambda Red recombination has not achieved scientific consensus, particularly in the case of dsDNA recombination. A clearer understanding of the mechanism underlying this process could suggest ways to improve the functionality, ease, and versatility of lambda Red recombination.Three phage-derived lambda Red proteins are necessary for carrying out dsDNA recombination: Gam, Exo, and Beta. Gam prevents the degradation of linear dsDNA by the E. coli RecBCD and SbcCD nucleases; lambda exonuclease (Exo) degrades dsDNA in a 5′ to 3′ manner, leaving single-stranded DNA in the recessed regions; and Beta binds to the single-stranded regions produced by Exo and facilitates recombination by promoting annealing to the homologous genomic target site (Sawitzke et al. 2007). Current mechanisms claim that Exo binds to both 5′ ends of the dsDNA and degrades in both directions simultaneously to produce a double-stranded region flanked on both sides by 3′ overhangs (Sharan et al. 2009; Szczepanska 2009). However, a comprehensive explanation of how this construct ultimately recombines with the chromosome has not yet been advanced.Initially, it was proposed that this recombination occurs via strand invasion (Thaler et al. 1987). However, it has more recently been shown that strand invasion is unlikely to be the dominant mechanism in the absence of long regions of homology, as recombination remains highly proficient in a recA^- background (Yu et al. 2000). Furthermore, a detailed analysis of lambda Red recombination products showed characteristics consistent with strand annealing rather than a strand invasion model (Stahl et al. 1997). Finally, lambda Red dsDNA recombination has been shown to preferentially target the lagging strand during DNA replication, which suggests strand annealing rather than strand invasion (Lim et al. 2008; Poteete 2008).To explain these results, Court et al. (2002) proposed a strand-annealing model for insertional dsDNA recombination (Figure 1A), in which one single-stranded 3′ end anneals to its homologous target at the replication fork. The replication fork then stalls, due to the presence of a large dsDNA nonhomology (i.e., the insertion cassette). The stalled replication fork is ultimately rescued by the other replication fork traveling in the opposite direction around the circular bacterial chromosome. The other 3′ end of the recombinogenic DNA anneals to the homology region exposed by the second replication fork, forming a crossover structure, which is then resolved by unspecified E. coli enzymes (Court et al. 2002).Open in a separate windowFigure 1.—Previously proposed lambda Red-mediated dsDNA recombination mechanisms. Heterologous dsDNA is shown in green; Exo is an orange oval, and Beta is a yellow oval. In both mechanisms the recombination intermediate is proposed to be a dsDNA core flanked on either side by 3′ ssDNA overhangs. (A) The Court mechanism posits that (1) Beta facilitates annealing of one 3′ overhang to the lagging strand of the replication fork. (2) This replication fork then stalls and backtracks so that the leading strand can template switch onto the synthetic dsDNA. The heterologous dsDNA blocks further replication from this fork. (3) Once the second replication fork reaches the stalled fork, the other 3′ end of the integration cassette is annealed to the lagging strand in the same manner as prior. Finally, the crossover junctions must be resolved by unspecified E. coli enzymes (Court et al. 2002). (B) The Poteete mechanism suggests that (1) Beta facilitates 3′ overhang annealing to the lagging strand of the replication fork and (2) positions the invading strand to serve as the new template for leading-strand synthesis. This structure is resolved by an unspecified host endonuclease (red triangle), and (3) the synthetic dsDNA becomes template for both lagging and leading-strand synthesis. A second template switch must then occur at the other end of the synthetic dsDNA (Poteete 2008). The figure was adapted from the references cited.The Court mechanism was challenged by Poteete (2008), who showed that the dsDNA recombination of a linear lambda phage chromosome occurs readily onto a unidirectionally replicating plasmid, which does not have the second replication fork required by the Court mechanism (Court et al. 2002). Thus, Poteete proposed an alternate mechanism (Poteete 2008), termed “replisome invasion” (Figure 1B), in which a 3′ overhang of the Exo-processed dsDNA first anneals to its complementary sequence on the lagging strand of the recombination target. Subsequently, this overhang displaces the leading strand, thereby serving as the new template for leading-strand synthesis. The resulting structure is resolved by an unspecified endonuclease, after which the recombinogenic DNA becomes the template for the synthesis of both new strands. In the context of recombineering using a linear dsDNA cassette, the author indicates that a second strand-switching event must occur at the other end of the incoming dsDNA.While Poteete''s mechanism addresses some of the weaknesses of the Court mechanism, it remains largely speculative. This mechanism does not identify the endonuclease responsible for resolving the structure after the first template switching event, nor does it explain how the recombinogenic DNA and replication machinery form a new replication fork. Additionally, this template-switching mechanism would have to operate two times in a well-controlled manner, which may not be consistent with the high-recombination frequencies often observed (Murphy et al. 2000) for lambda Red-mediated dsDNA insertion. Finally, little experimental evidence has been advanced to directly support this hypothesis.To address the deficiencies in these mechanisms, we propose that lambda Red dsDNA recombination proceeds via a ssDNA intermediate rather than a dsDNA core flanked by 3′ overhangs (Figure 2). In this mechanism, Exo binds to one of the two dsDNA strands and degrades that strand completely, leaving behind full-length ssDNA. This ssDNA then anneals to its homology target at the lagging strand of the replication fork and is incorporated as part of the newly synthesized strand as if it were an Okazaki fragment. This process is analogous to the accepted mechanism for the lambda Red-mediated recombination of ssDNA oligonucleotides (Court et al. 2002) and, therefore, unifies the mechanisms for ssDNA and dsDNA recombination. Notably, our mechanism uses one replication fork for the incorporation of a full-length heterologous cassette, thereby addressing Poteete''s criticism of the Court mechanism.Open in a separate windowFigure 2.—Lambda Red mediated dsDNA recombination proceeds via a ssDNA intermediate. Instead of a recombination intermediate involving dsDNA flanked by 3′-ssDNA overhangs, we propose that one strand of linear dsDNA is entirely degraded by Exo (orange oval). Beta (yellow oval) then facilitates annealing to the lagging strand of the replication fork in place of an Okazaki fragment. The heterologous region does not anneal to the genomic sequence. This mechanism could account for gene replacement (as shown) or for insertions in which no genomic DNA is removed.The degradation of an entire strand by lambda Exo is feasible, given the highly processive nature of the enzyme (Subramanian et al. 2003). Whereas previously proposed mechanisms assume that both dsDNA ends are degraded approximately simultaneously, our hypothesis implies that some dsDNA molecules will be entirely degraded to ssDNA before a second Exo can bind to the other end. In this article, we demonstrate that single-stranded DNA is a viable recombinogenic intermediate with lagging-strand bias. Furthermore, we show that genetic information from one strand of a recombinogenic dsDNA cassette cosegregates during lambda Red-mediated recombination. These results provide strong support of our proposed mechanism.     

The Dot1 Histone Methyltransferase and the Rad9 Checkpoint Adaptor Contribute to Cohesin-Dependent Double-Strand Break Repair by Sister Chromatid Recombination in Saccharomyces cerevisiae

Genomic integrity is threatened by multiple sources of DNA damage. DNA double-strand breaks (DSBs) are among the most dangerous types of DNA lesions and can be generated by endogenous or exogenous agents, but they can arise also during DNA replication. Sister chromatid recombination (SCR) is a key mechanism for the repair of DSBs generated during replication and it is fundamental for maintaining genomic stability. Proper repair relies on several factors, among which histone modifications play important roles in the response to DSBs. Here, we study the role of the histone H3K79 methyltransferase Dot1 in the repair by SCR of replication-dependent HO-induced DSBs, as a way to assess its function in homologous recombination. We show that Dot1, the Rad9 DNA damage checkpoint adaptor, and phosphorylation of histone H2A (γH2A) are required for efficient SCR. Moreover, we show that Dot1 and Rad9 promote DSB-induced loading of cohesin onto chromatin. We propose that recruitment of Rad9 to DSB sites mediated by γH2A and H3K79 methylation contributes to DSB repair via SCR by regulating cohesin binding to damage sites. Therefore, our results contribute to an understanding of how different chromatin modifications impinge on DNA repair mechanisms, which are fundamental for maintaining genomic stability.IN eukaryotic cells, genomic integrity is ensured by the action of the DNA damage checkpoint. This checkpoint coordinates the cellular response to DNA damage, triggering cell cycle arrest and activating DNA repair mechanisms, thus providing time for the cell to repair the damage before resuming cell cycle progression (Harrison and Haber 2006). DNA double-strand breaks (DSBs) are among the most dangerous genomic lesions and, if they are not properly repaired, they can lead to fatal consequences. DSBs can be repaired either by homologous recombination (HR) or by nonhomologous end joining (NHEJ), but only HR with the sister chromatid ensures that the fidelity of genetic information is mantained. Thus, sister chromatid recombination (SCR) is the preferred mechanism of DSB repair in mitotic cells (Kadyk and Hartwell 1992; Johnson and Jasin 2000; González-Barrera et al. 2003). Since SCR occurs between identical DNA molecules, its analysis by genetic or physical methods is difficult but, recently, a physical assay to monitor the repair by SCR of a single DSB generated during replication has been developed in budding yeast (González-Barrera et al. 2003; Cortes-Ledesma and Aguilera 2006). This SCR assay is based on a circular minichromosome harboring an internal mini-HO site, which is cleaved mainly in one strand producing ∼10% DSBs during replication, in contrast to the direct and efficient DSB induction at the full-length HO site. In this way, upon HO induction, the DSB occurs only in one chromatid and the other one remains intact and available for repair (see Figure 1A). Although this assay has been used mainly to monitor unequal SCR events, it has been demonstrated that it is an accurate indicator of the proficiency in total SCR (González-Barrera et al. 2003; Cortes-Ledesma and Aguilera 2006). Using this physical assay, it has been established that Rad52, Rad59, Rad51, and Rad54, but not Rdh54/Tid1, are involved in SCR (Cortes-Ledesma et al. 2007b). Also, SMC (structural maintenance of chromosomes) proteins including the cohesin complex and the Smc5/6 complex are required for efficient SCR (Cortes-Ledesma and Aguilera 2006; De Piccoli et al. 2006; Cortes-Ledesma et al. 2007a).Open in a separate windowFigure 1.—Dot1 is required for efficient SCR. (A) Schematic of the physical assay used to monitor repair by SCR of an HO-induced DSB in the centromeric plasmid pRS316-TINV. Fragments generated after XhoI–SpeI digestion, detected by the LEU2 probe (line with asterisks) are indicated with their corresponding sizes. Since other recombination events can also lead to the 2.9-kb fragment, only the 4.7-kb band is used to measure SCR. (B) Kinetics of HO-induced DSB formation and its repair in wild-type (MKOS-3C) and dot1 (YP764) cells incubated in galactose for the indicated times. A representative Southern blot is presented showing the different fragments detected. The 3.8-kb band corresponds to the intact plasmid and equal SCR events, the 1.4-kb and 2.4-kb fragments arise after HO cut, the 2.9-kb band results from unequal SCR and IC-BIR and the 4.7-kb band is specific for unequal SCR. (C) Quantification of DSBs (1.4-kb plus 2.4-kb bands) and SCR (4.7-kb band) relative to the total DNA. Averages and standard deviations are shown. In some cases, such as the dot1 SCR values, the error bars are hidden by the graph symbols.Detection, signaling and repair of genomic lesions occur in the context of chromatin; therefore, factors regulating chromatin structure, such as histone modifications and chromatin remodelers, play important roles in the DNA damage response (Peterson and Cote 2004; Lydall and Whitehall 2005; van Attikum and Gasser 2005; Downs et al. 2007). Mec1- and Tel1-dependent phosphorylation of histone H2A at serine 129 (hereafter referred to as γH2A) is required for DSB repair by NHEJ and likely HR (Downs et al. 2000) and also mediates recruitment of cohesin to DSB sites (Unal et al. 2004). Another chromatin modification involved in the DNA damage response is the methylation of lysine 79 of histone H3 (H3K79) mediated by Dot1 (van Leeuwen et al. 2002). During meiosis, Dot1 is required for the meiotic recombination checkpoint (San-Segundo and Roeder 2000) and, in mitotic cells, Dot1-dependent H3K79 methylation is involved in the Rad9-mediated activation of the Rad53 checkpoint kinase (Giannattasio et al. 2005; Wysocki et al. 2005). Moreover, genetic analyses of the response to different DNA damaging agents, such as ionizing radiation (IR), methyl methanesulfonate (MMS), and UV, have suggested that Dot1 modulates multiple DNA repair pathways (Game et al. 2006; Toh et al. 2006; Bostelman et al. 2007; Conde and San-Segundo 2008) and also controls DSB resection (Lazzaro et al. 2008). To gain further insight in the molecular mechanisms of DNA repair impacted by Dot1 function we have used a physical assay to monitor DSB repair by SCR as a manifestation of HR repair. We provide molecular and genetic evidence indicating that Dot1, together with γH2A, promotes SCR by Rad9-mediated recruitment of cohesin to DSB sites.     

TEN1 Is Essential for CDC13-Mediated Telomere Capping

Telomere binding proteins protect chromosome ends from degradation and mask chromosome termini from checkpoint surveillance. In Saccharomyces cerevisiae, Cdc13 binds single-stranded G-rich telomere repeats, maintaining telomere integrity and length. Two additional proteins, Ten1 and Stn1, interact with Cdc13 but their contributions to telomere integrity are not well defined. Ten1 is known to prevent accumulation of aberrant single-stranded telomere DNA; whether this results from defective end protection or defective telomere replication is unclear. Here we report our analysis of a new group of ten1 temperature-sensitive (ts) mutants. At permissive temperatures, ten1-ts strains display greatly elongated telomeres. After shift to nonpermissive conditions, however, ten1-ts mutants accumulate extensive telomeric single-stranded DNA. Cdk1 activity is required to generate these single-stranded regions, and deleting the EXO1 nuclease partially suppresses ten1-ts growth defects. This is similar to cdc13-1 mutants, suggesting ten1-ts strains are defective for end protection. Moreover, like Cdc13, our analysis reveals Ten1 promotes de novo telomere addition. Interestingly, in ten1-ts strains at high temperatures, telomeric single-stranded DNA and Rad52-YFP repair foci are strongly induced despite Cdc13 remaining associated with telomeres, revealing Cdc13 telomere binding is not sufficient for end protection. Finally, unlike cdc13-1 mutants, ten1-ts strains display strong synthetic interactions with mutations in the POLα complex. These results emphasize that Cdc13 relies on Ten1 to execute its essential function, but leave open the possibility that Ten1 has a Cdc13-independent role in DNA replication.GENOME stability is critically dependent upon functional telomeres. DNA ends that lack telomeres, or that have dysfunctional telomeres, are metabolized by DNA repair processes; without an appropriate repair template, such chromosome ends can be resected or joined inappropriately with other chromosome ends. Thus, genomic integrity can be significantly compromised by telomere dysfunction, particularly in proliferating cells where cycles of instability may ensue due to creation of dicentric chromosomes (Bailey and Murnane 2006). Protein complexes that bind to the duplex and single-stranded telomere repeats are key for stabilizing the chromosome ends (de Lange 2005). In proliferating cells, this job is complicated not only because the terminal chromatin must be opened during the process of chromosome replication, but also because additional processes that metabolize DNA ends are active. For example, while nonhomologous end joining processes are preferentially used in repair of DNA double-strand breaks in G1, homologous recombination is preferentially used for this repair in S and G2 (Ferreira and Cooper 2004; Zierhut and Diffley 2008). Given these complexities, it is not surprising that our molecular understanding of how telomere proteins protect chromosomes ends is incomplete.Budding yeast has been useful for dissecting how cells correctly metabolize their chromosome ends. In Saccharomyces cerevisiae, the terminal DNA comprises approximately 300 bp of TG_1-3/C_1-3A sequences, ending with a short single-stranded overhang of the G-rich repeats. This 3′ overhang is ∼12–14 nucleotides, although during the late S/G2 phase of the cell cycle, it becomes longer, >30 nucleotides in length (Wellinger et al. 1993b; Dionne and Wellinger 1996; Larrivee et al. 2004). Central among factors that prevent inappropriate telomere degradation in S. cerevisiae is Cdc13, a protein that binds to single-stranded telomere G-rich repeats (Garvik et al. 1995; Lin and Zakian 1996; Nugent et al. 1996). Reducing Cdc13 function through either the cdc13-1 temperature sensitive (ts) allele or the cdc13-td conditional null (degron) allele results in telomere C-strand loss, with degradation continuing into the subtelomeric chromosomal regions (Garvik et al. 1995; Vodenicharov and Wellinger 2006). Correspondingly, homologous recombination at chromosome termini increases in cdc13-1 strains (Carson and Hartwell 1985; Garvik et al. 1995). The loss of Cdc13 unmasks the telomeres, provoking activation of the DNA damage checkpoint (Weinert and Hartwell 1993; Garvik et al. 1995). This protective role of Cdc13 is most likely its essential function.A thorough, mechanistic understanding of how Cdc13 mediates chromosome end protection is hampered in part because the activities responsible for the loss of the telomere C strand are not fully known. At normal telomeres, the Mre11-Rad50-Xrs2 complex has a role regulating resection required for telomere addition, whereas the Exo1 nuclease, Rad9 and Rad24 checkpoint proteins each influence the resection process at uncapped telomeres (Lydall and Weinert 1995; Maringele and Lydall 2002; Larrivee et al. 2004; Zubko et al. 2004). The 5′-to-3′ resection of both normal and uncapped telomeres is regulated by the activity of Cdk1, the yeast cyclin-dependent kinase (Frank et al. 2006; Vodenicharov and Wellinger 2006). Similar to the activities that promote 5′-to-3′ degradation of DNA ends at double-strand breaks (Aylon et al. 2004; Ira et al. 2004), the activities that lead to telomere resection are active in late S and G2 cell cycle phases (Wellinger et al. 1993a, 1996; Marcand et al. 2000; Vodenicharov and Wellinger 2006). Interestingly, Cdc13 is required to prevent degradation at telomeres only in proliferating cells and not when cells are blocked in stationary phase (Vodenicharov and Wellinger 2006). Additional factors, such as the S. cerevisiae Rap1 protein, prevent chromosome fusions by nonhomologous recombination during the G1 phase of the cell cycle (Pardo and Marcand 2005; Marcand et al. 2008).At least two additional proteins, Stn1 and Ten1, aid the capping role of Cdc13. Like CDC13, both STN1 and TEN1 are essential, and loss of their function leads to excessive single-stranded telomeric DNA (Grandin et al. 1997, 2001; Petreaca et al. 2007). STN1 was originally identified as a high copy suppressor of cdc13-1 temperature sensitivity (Grandin et al. 1997), and TEN1 was similarly isolated as a dosage suppressor of stn1-13 (Grandin et al. 2001). Combining either the cdc13-1 allele with stn1 mutations or the ten1-31 allele with stn1-13 is lethal (Grandin et al. 2001; Petreaca et al. 2007). The essential nature of these genes makes it difficult to clearly differentiate whether these genes operate in the same, or in parallel pathways to protect telomeres. A compelling argument that Cdc13, Stn1, and Ten1 likely function in a common pathway is that, in addition to these genetic interactions, Stn1 and Ten1 proteins interact with one another both in vivo and in vitro (Grandin et al. 2001; Gao et al. 2007), and each associates with Cdc13 in the yeast two-hybrid assay (Grandin et al. 1997, 2001; Petreaca et al. 2007). From these data, Cdc13, Stn1, and Ten1 are suggested to function as a single complex that mediates chromosome end protection in S. cerevisiae. Such a complex would share some similarities with the single-stranded DNA binding complex RPA (Gao et al. 2007). Whether these proteins normally operate exclusively as a heterotrimeric complex is still not entirely clear. Stn1 and Ten1 can make contributions to capping that are independent of Cdc13, as shown in experiments where overproducing the Stn1 essential domain with Ten1 replaced the essential function of Cdc13 (Petreaca et al. 2006). In addition, while the Schizosaccharomyces pombe Stn1 and Ten1 homologs are critical for telomere protection, they do not interact with Pot1, the single-stranded telomere binding protein that is also critical for telomere capping (Martin et al. 2007).The role of Ten1 in maintaining both telomere integrity and length homeostasis is not understood. It has been assumed that Stn1 and Ten1 play the same role as Cdc13 in maintaining telomere integrity, namely, preventing inappropriate terminal resection. However, whether this is in fact the case is not entirely clear. For one, disrupting the DNA replication machinery can give rise to an excess of terminal single-stranded DNA, although in this case, the ssDNA accumulation is attributed to a failure to synthesize the lagging DNA strand rather than removing a block to telomere resection (Diede and Gottschling 1999; Adams Martin et al. 2000). Although both Cdc13 and Stn1 are thought to act as capping proteins, each can interact with Polα subunits (Qi et al. 2003; Grossi et al. 2004; Petreaca et al. 2006), making it important to evaluate Ten1 function more carefully. Our goal here was to compare how Cdc13 and Ten1 promote chromosome end protection, first by testing whether Ten1 acts to prevent telomere resection from activities comparable to those that degrade telomeres in cdc13-1, and second by determining the impact of ten1 dysfunction upon Cdc13. The cdc13-1 allele has been extremely useful in analyzing the CDC13 essential function; TEN1 analysis has been hindered by a lack of equivalent genetic reagents. Here we have created a collection of ten1-ts alleles useful for probing the essential role of TEN1. Analysis of these alleles, which show constitutive telomere elongation, reveals that Ten1 promotes telomere capping with a similar cell cycle dependency as Cdc13, protecting ends during the period in which mitotic forms of Cdk1 are active. Critically, by showing that single-stranded DNA is generated in ten1-ts strains under conditions where semi-conservative replication is complete, we conclude that Ten1 truly can function as a capping protein. Moreover, the ten1-ts strains fail to restrain degradation of chromosome ends and induce formation of Rad52 repair foci, despite the association of wild-type Cdc13 with telomeres, indicating not only that Cdc13 binds telomeres independent of Ten1 function, but also that Cdc13 telomere localization is not sufficient for end protection. Finally, although the ten1-ts capping-deficient phenotypes parallel cdc13-1, only the ten1-ts strains are highly sensitive to impaired POL1 function, leaving open the possibility that TEN1 function additionally impacts terminal replication.     

Lethal Mutagenesis in Viruses and Bacteria

In this work we study how mutations that change physical properties of cell proteins (stability) affect population survival and growth. We present a model in which the genotype is presented as a set folding free energies of cell proteins. Mutations occur upon replication, so stabilities of some proteins in daughter cells differ from those in the parent cell by amounts deduced from the distribution of mutational effects on protein stability. The genotype–phenotype relationship posits that the cell''s fitness (replication rate) is proportional to the concentration of its folded proteins and that unstable essential proteins result in lethality. Simulations reveal that lethal mutagenesis occurs at a mutation rate close to seven mutations in each replication of the genome for RNA viruses and at about half that rate for DNA-based organisms, in accord with earlier predictions from analytical theory and experimental results. This number appears somewhat dependent on the number of genes in the organisms and the organism''s natural death rate. Further, our model reproduces the distribution of stabilities of natural proteins, in excellent agreement with experiments. We find that species with high mutation rates tend to have less stable proteins compared to species with low mutation rates.MUTATION rates play an important role in the evolution and adaptation of bacteria and viruses. Considerable experimental evidence suggests that high mutation rates in RNA virus populations have powered their rapid evolution (Eggers and Tamm 1965; Domingo et al. 1978; de la Torre et al. 1990; Domingo 2000). However, artificially elevated mutation rates were shown to have deleterious effects on the fitness of RNA viruses and to eventually lead to extinction of the viral population beyond certain mutation rate thresholds (Loeb et al. 1999; Sierra et al. 2000; Pariente et al. 2001; Grande-Perez et al. 2002; Anderson et al. 2004; Freistadt et al. 2004; Bull et al. 2007; Graci et al. 2007, 2008; Zeldovich et al. 2007). This observation is called lethal mutagenesis for RNA viruses. Several authors proposed to use lethal mutagenesis to cure or control infection with RNA viruses, using certain mutagens (Anderson et al. 2004; Freistadt et al. 2004). The possibility of lethal mutagenesis in bacteria was also suggested and studied recently (Gessler 1995; Andre and Godelle 2006; Bull and Wilke 2008).Previously, many attempts have been made, using population genetics, to theoretically describe the effect of mutation rates on the survival of a population (Muller 1964; Haigh 1978; Gessler 1995). The effect has frequently been described within the paradigm of Muller''s ratchet (Muller 1964; Haigh 1978; Andersson and Hughes 1996), where the genome of an asexual organism accumulates stochastic deleterious mutations in an irreversible manner, leading to the systematic decrease in the fitness of the organism. The concept of Muller''s ratchet applies to finite, asexual populations. It states that if back mutations cannot occur, eventually any finite asexual population will accumulate deleterious mutations and the mutation-free wild-type genotype would be lost. While such models provided some useful insights into the phenomenon of lethal mutagenesis, they often assume a single fitness peak and absence of back or compensating mutations and depend heavily on arbitrary parameters, such as selection coefficients or deleterious mutation rates. Such analyses therefore lack a more fundamental connection between the physical properties of the proteins within the organism, the metabolic network of the organism, and the feedback relationship between the mutation rate and organismal fitness.In recent years, several theories of lethal mutagenesis have been proposed (Guo et al. 2004; Bull et al. 2007; Zeldovich et al. 2007; Bull and Wilke 2008). In a marked departure from earlier phenomenological approaches Zeldovich et al. (2007) suggested a model assuming that the loss of protein stability would lead to the loss of essential functions within the organism and therefore to a lethal phenotype. The evolution of a population in this model was mapped to a diffusion process in a multidimensional hypercube where each dimension represented stability of proteins encoded by an essential gene and adsorbing boundary conditions at ΔG=0 boundaries [where ΔG is the difference between free energy of the folded and unfolded proteins, which is the thermodynamic measure of protein stability (Zeldovich et al. 2007)] were imposed to account for the fact that loss of stability confers a lethal phenotype on an organism. This model differs from previous approaches in that, instead of depending on arbitrarily calibrated parameters such as selection coefficients or deleterious mutation rates, it is based only on the statistical distribution of proteins'' folding free energy change after point mutations, which was directly derived from in vitro experiments. Furthermore, it predicts a lethal mutagenesis threshold that is consistent with experimental findings and discovers a deep relation between fundamental biophysical properties of proteins, the mutation rate, and organismal fitness.However, despite these insights, the model proposed by Zeldovich et al. (2007) is based on a number of simplifying assumptions. First, it assumes a uniform mutation supply within the population, meaning that at any time, mutations could occur in any organism in the population with an equal and constant probability. However, in real biological systems mutations are coupled to replication. While the formalism developed by Zeldovich et al. (2007) allows one to consider the case when mutations are coupled to replication (see Methods in Zeldovich et al. 2007), this formalism gives numerically accurate predictions for the coupled mutation–replication only in the limit of low mutation rates. However, lethal mutagenesis occurs when the mutation rate is relatively high (approximately six mutations per genome per replication according to Crotty et al. 2001 and Zeldovich et al. 2007). Second, the model developed in Zeldovich et al. (2007) assumes a very simple “Θ-function-like” fitness landscape whereby fitness is the same for all protein stabilities as long as proteins are stable; i.e., it is flat for all ΔG<0. However, in reality as proteins become less stable they spend a greater fraction of time in the unfolded state, reducing therefore the effective concentration of functional proteins, which may affect fitness. Our study overcomes these limitations in a new computational model as outlined below.If the organism has a conservative replication mechanism, as is the case for RNA viruses, then mutations would occur, with certain probabilities, only in the descendant copy, while the parent copy would remain unchanged (that may also be the case in organisms with double-stranded genomes, where the methylation mechanism keeps a master copy of the genome preserved). If the organism has semiconservative replication, as in bacteria and DNA viruses and double-stranded RNA viruses, then mutations could happen with certain probabilities in both daughter organisms.Here we present a detailed study of the fitness effect of protein stability changing mutations, based on the coupled mutation–replication scenario and more explicit physical consideration of the effect of protein stability on fitness. Simulating evolution and population growth in this model, we observe lethal mutagenesis in conservative and semiconservative replicating populations. Further, we show how stationary distribution of protein stabilities (folding free energies) emerges and discuss the physical and evolutionary reasons for the observed moderate stability of proteins.     

High Sensitivity to Auxin is a Common Feature of Hairy Root

The responses to auxin of Lycopersicon esculentum roots transformed by (T_l+T_r)-DNA of the Ri plasmid of agropine-type Agrobacterium rhizogenes strain 15834 and Catharanthus trichophyllus roots transformed by the (T_l+T_r)-DNA, and by T_l- or T_r- DNA alone of the same bacterial strain were compared to that of their normal counterparts. The transmembrane electrical potential difference of root protoplasts was measured as a function of the concentration of exogenous naphthalene acetic acid. The sensitivity to auxin expressed by this response was shown to be independent of the measurement conditions and of the basal polarization of isolated protoplasts. According to this electrical response, as well as to the modulation by auxin of proton excretion by root tips and root tip elongation, roots transformed by (T_l+T_r) DNA are 100 to 1000 times more sensitive to exogenous auxin than normal roots, as is the case with normal and transformed roots from Lotus corniculatus (WH Shen, A Petit, J Guern, J Tempé [1988] Proc Natl Acad Sci USA 85: 3417-3421). Further-more, transformed roots of C. trichophyllus are not modified in their sensitivity to fusicoccin, illustrating the specificity of the modification of the auxin sensitivity. Roots transformed by the T_r-DNA alone showed the same sensitivity to auxin as normal roots, whereas the roots transformed by the T_l-DNA alone exhibited an auxin sensitivity as high as the roots transformed by (T_l+T_r)-DNA. It was concluded that the high sensitivity to auxin is controlled by the T_l-DNA in agropine type Ri plasmids.     

Sexual Isolation in Acinetobacter baylyi Is Locus-Specific and Varies 10,000-Fold Over the Genome

Naturally transformable bacteria acquire chromosomal DNA from related species at lower frequencies than from cognate DNA sources. To determine how genome location affects heterogamic transformation in bacteria, we inserted an nptI marker into random chromosome locations in 19 different strains of the Acinetobacter genus (>24% divergent at the mutS/trpE loci). DNA from a total of 95 nptI-tagged isolates was used to transform the recipient Acinetobacter baylyi strain ADP1. A total of >1300 transformation assays revealed that at least one nptI-tagged isolate for each of the strains/species tested resulted in detectable integration of the nptI marker into the ADP1 genome. Transformation frequencies varied up to ∼10,000-fold among independent nptI insertions within a strain. The location and local sequence divergence of the nptI flanking regions were determined in the transformants. Heterogamic transformation depended on RecA and was hampered by DNA mismatch repair. Our studies suggest that single-locus-based studies, and inference of transfer frequencies from general estimates of genomic sequence divergence, is insufficient to predict the recombination potential of chromosomal DNA fragments between more divergent genomes. Interspecies differences in overall gene content, and conflicts in local gene organization and synteny are likely important determinants of the genomewide variation in recombination rates between bacterial species.HORIZONTAL gene transfer (HGT) contributes to bacterial evolution by providing access to DNA evolved and retained in separate species or strains (Cohan 1994a,b; Bergstrom et al. 2000; Ochman et al. 2000; Feil et al. 2001; Koonin 2003; Lawrence and Hendrickson 2003; Fraser et al. 2007). Multilocus sequence typing (MLST) has provided strong evidence for frequent transfer and recombination of chromosomal DNA between related bacterial strains within the same species (Maiden et al. 1998; Enright et al. 2002). HGT occurring by natural transformation allows bacteria to exploit the presence of nucleic acids in their environment for the purposes of nutrition, DNA repair, reacquisition of lost genes, and/or acquisition of novel genetic diversity (Redfield 1993; Mehr and Seifert 1998; Dubnau 1999; Claverys et al. 2000; Szöllösi et al. 2006; Johnsen et al. 2009). It can be inferred from observations of the presence of extracellular DNA in most environments that bacteria are constantly exposed to DNA from a variety of sources, without such exposure necessarily producing observable changes in the genetic compositions of bacterial populations over evolutionary time (Thomas and Nielsen 2005; Nielsen et al. 2007a,b).The absence of sequence similarity between the donor DNA and the DNA of the recipient bacterium is the strongest barrier to the horizontal acquisition of chromosomal genes in bacteria (Matic et al. 1996; Vulic et al. 1997; Majewski 2001; Townsend et al. 2003) as illegitimate recombination occurs only at extremely low frequencies in bacteria (Hülter and Wackernagel 2008a). Single-locus transfer models have been extensively applied and have demonstrated a log-linear decrease in recombination frequencies with increasing sequence divergence for Bacillus subtilis (Roberts and Cohan, 1993; Zawadzki et al. 1995), Acinetobacter baylyi (Young and Ornston 2001), Escherichia coli (Shen and Huang 1986; Vulic et al. 1997), and Streptococcus pneumoniae (Majewski et al. 2000). For instance, heterogamic transformation between nonmutator isolates at the rpoB locus of B. mojavensis is undetectable at sequence divergences >16.7% (Zawadzki et al. 1995) and between S. pneumoniae isolates with sequence divergences >18% (Majewski et al. 2000). In A. baylyi, the nonmutator sequence divergence limit for detectable transformation at the pcaH locus of strain ADP1 was found to be 20% (Young and Ornston 2001), and up to 24% overall divergence yielded transformants at 16S rRNA loci in strain DSM587 (Strätz et al. 1996).Several recent studies also show that short stretches (<200 bp) of DNA sequence identity can facilitate additive or substitutive integration of longer stretches (>1000 bp) of heterologous DNA in bacteria (Prudhomme et al. 1991, 2002; de Vries and Wackernagel 2002; Hülter and Wackernagel 2008a). Thus, the uptake of DNA in bacteria can facilitate larger substitutions within gene sequences and the integration of additional DNA material on the basis of recombination initiated in flanking DNA stretches (either at one or both ends) with high sequence similarity (Nielsen et al. 2000). On the other hand, segments of heterologous DNA interrupting the synteny of homologous DNA have also been shown to be a barrier in intraspecies transformation in S. pneumoniae (Pasta and Sicard 1996, 1999).The various studies of the interspecies transfer potential of single genes demonstrate that the immediate local sequence divergence of the transferred locus is of high importance in determining recombination frequencies in hosts up to 20% divergent (at the housekeeping gene level). However, it can be hypothesized that the broader structural, organizational, and biochemical properties of the genome region surrounding a particular locus will determine its transfer potential to more divergent host species (Cohan 2001; Lawrence 2002). The interspecies transfer potential of various genome regions/loci between more diverged species (>20% at the housekeeping gene level) may therefore differ substantially from a log-linear model (determined experimentally for more closely related species) as local gene organization becomes less conserved with evolutionary time. The barriers to gene exchange between divergent bacterial species is likely a combination of inefficient recombination due to both mismatched base pairs (the main determinator in the log-linear model) and conflicting gene order and organization across the local recombining DNA regions. In addition, selective barriers due to negative effects on host fitness of the transferred DNA regions may become increasingly important for the removal of recombination events from the bacterial population. Recent bioinformatics-based genome analysis of E. coli and Salmonella genomes suggests various parts of the bacterial genome may have different suceptibilities to undergo evolutionarily successful recombination leading to temporal fragmentation of speciation (Lawrence 2002; Retchless and Lawrence 2007). Nevertheless, few studies have experimentally tested the effect of variable species and chromosome locations of genes on their transfer potential between bacteria (Ravin and Chen 1967; Ravin and Chakrabarti 1975; Siddiqui and Goldberg 1975; Cohan et al. 1991; Huang et al. 1991; Fall et al. 2007).Here, we determine to what extent genome location contributes to sexual isolation between the recipient A. baylyi strain ADP1 and 19 sequence divergent (24–27% divergent at the mutS/trpE loci) donor Acinetobacter strains and species (carrying a selectable nptI gene in a total of 95 random genome locations).     

Deficiency in l-Serine Deaminase Interferes with One-Carbon Metabolism and Cell Wall Synthesis in Escherichia coli K-12

Escherichia coli K-12 provided with glucose and a mixture of amino acids depletes l-serine more quickly than any other amino acid even in the presence of ammonium sulfate. A mutant without three 4Fe4S l-serine deaminases (SdaA, SdaB, and TdcG) of E. coli K-12 is unable to do this. The high level of l-serine that accumulates when such a mutant is exposed to amino acid mixtures starves the cells for C₁ units and interferes with cell wall synthesis. We suggest that at high concentrations, l-serine decreases synthesis of UDP-N-acetylmuramate-l-alanine by the murC-encoded ligase, weakening the cell wall and producing misshapen cells and lysis. The inhibition by high l-serine is overcome in several ways: by a large concentration of l-alanine, by overproducing MurC together with a low concentration of l-alanine, and by overproducing FtsW, thus promoting septal assembly and also by overexpression of the glycine cleavage operon. S-Adenosylmethionine reduces lysis and allows an extensive increase in biomass without improving cell division. This suggests that E. coli has a metabolic trigger for cell division. Without that reaction, if no other inhibition occurs, other metabolic functions can continue and cells can elongate and replicate their DNA, reaching at least 180 times their usual length, but cannot divide.The Escherichia coli genome contains three genes, sdaA, sdaB, and tdcG, specifying three very similar 4Fe4S l-serine deaminases. These enzymes are very specific for l-serine for which they have unusually high K_m values (3, 32). Expression of the three genes is regulated so that at least one of the gene products is synthesized under all common growth conditions (25). This suggests an important physiological role for the enzymes. However, why E. coli needs to deaminate l-serine has been a long-standing problem of E. coli physiology, the more so since it cannot use l-serine as the sole carbon source.We showed recently that an E. coli strain devoid of all three l-serine deaminases (l-SDs) loses control over its size, shape, and cell division when faced with complex amino acid mixtures containing l-serine (32). We attributed this to starvation for single-carbon (C₁) units and/or S-adenosylmethionine (SAM). C₁ units are usually made from serine via serine hydroxymethyl transferase (GlyA) or via glycine cleavage (GCV). The l-SD-deficient triple mutant strain is starved for C₁ in the presence of amino acids, because externally provided glycine inhibits GlyA and a very high internal l-serine concentration along with several other amino acids inhibits glycine cleavage. While the parent cell can defend itself by reducing the l-serine level by deamination, this crucial reaction is missing in the ΔsdaA ΔsdaB ΔtdcG triple mutant. We therefore consider these to be “defensive” serine deaminases.The fact that an inability to deaminate l-serine leads to a high concentration of l-serine and inhibition of GlyA is not surprising. However, it is not obvious why a high level of l-serine inhibits cell division and causes swelling, lysis, and filamentation. Serine toxicity due to inhibition of biosynthesis of isoleucine (11) and aromatic amino acids (21) has been reported but is not relevant here, since these amino acids are provided in Casamino Acids.We show here that at high internal concentrations, l-serine also causes problems with peptidoglycan synthesis, thus weakening the cell wall. Peptidoglycan is a polymer of long glycan chains made up of alternating N-acetylglucosamine and N-acetylmuramic acid residues, cross-linked by l-alanyl-γ-d-glutamyl-meso-diaminopimelyl-d-alanine tetrapeptides (1, 28). The glucosamine and muramate residues and the pentapeptide (from which the tetrapeptide is derived) are all synthesized in the cytoplasm and then are exported to be polymerized into extracellular peptidoglycan (2).In this paper, we show that lysis is caused by l-serine interfering with the first step of synthesis of the cross-linking peptide, the addition of l-alanine to uridine diphosphate-N-acetylmuramate. This interference is probably due to a competition between serine and l-alanine for the ligase, MurC, which adds the first l-alanine to UDP-N-acetylmuramate (7, 10, 15). As described here, the weakening of the cell wall by l-serine can be overcome by a variety of methods that reduce the endogenous l-serine pool or counteract the effects of high levels of l-serine.     

On the inference of complex phylogenetic networks by Markov Chain Monte-Carlo

For various species, high quality sequences and complete genomes are nowadays available for many individuals. This makes data analysis challenging, as methods need not only to be accurate, but also time efficient given the tremendous amount of data to process. In this article, we introduce an efficient method to infer the evolutionary history of individuals under the multispecies coalescent model in networks (MSNC). Phylogenetic networks are an extension of phylogenetic trees that can contain reticulate nodes, which allow to model complex biological events such as horizontal gene transfer, hybridization and introgression. We present a novel way to compute the likelihood of biallelic markers sampled along genomes whose evolution involved such events. This likelihood computation is at the heart of a Bayesian network inference method called SnappNet, as it extends the Snapp method inferring evolutionary trees under the multispecies coalescent model, to networks. SnappNet is available as a package of the well-known beast 2 software.Recently, the MCMC_BiMarkers method, implemented in PhyloNet, also extended Snapp to networks. Both methods take biallelic markers as input, rely on the same model of evolution and sample networks in a Bayesian framework, though using different methods for computing priors. However, SnappNet relies on algorithms that are exponentially more time-efficient on non-trivial networks. Using simulations, we compare performances of SnappNet and MCMC_BiMarkers. We show that both methods enjoy similar abilities to recover simple networks, but SnappNet is more accurate than MCMC_BiMarkers on more complex network scenarios. Also, on complex networks, SnappNet is found to be extremely faster than MCMC_BiMarkers in terms of time required for the likelihood computation. We finally illustrate SnappNet performances on a rice data set. SnappNet infers a scenario that is consistent with previous results and provides additional understanding of rice evolution.