Disintegration of nascent replication bubbles during thymine starvation triggers RecA- and RecBCD-dependent replication origin destruction

Defects in the central metabolism prevent thymineless death in Escherichia coli, while still allowing significant protein synthesis

Starvation of E. coli thyA auxotrophs for the required thymine or thymidine leads to cessation of DNA synthesis and, unexpectedly, to thymineless death (TLD). Previously, TLD-alleviating defects were identified by the candidate gene approach, for their contribution to replication initiation, fork repair, or SOS induction. However, no TLD-blocking mutations were ever found, suggesting a multifactorial nature of TLD. Since (until recently) no unbiased isolation of TLD suppressors was reported, we used enrichment after insertional mutagenesis to systematically isolate TLD suppressors. Our approach was validated by isolation of known TLD-alleviating mutants in recombinational repair. At the same time, and unexpectedly for the current TLD models, most of the isolated suppressors affected general metabolism, while the strongest suppressors impacted the central metabolism. Several temperature-sensitive (Ts)-mutants in important/essential functions, like nadA, ribB or coaA, almost completely suppressed TLD at 42°C. Since blocking protein synthesis completely by chloramphenicol prevents TLD, while reducing protein synthesis to 10% alleviates TLD only slightly, we measured the level of protein synthesis in these mutants at 42°C and found it to be 20-70% of the WT, not enough reduction to explain TLD prevention. We conclude that the isolated central metabolism mutants prevent TLD by affecting specific TLD-promoting functions.

Thymine-starvation-induced chromosomal fragmentation is not required for thymineless death in Escherichia coli

Thymine or thymidine starvation induces robust chromosomal fragmentation in E. coli thyA deoCABD mutants, and is proposed to be the cause of thymineless death (TLD). However, fragmentation kinetics challenges the idea that fragmentation causes TLD, by peaking before the onset of TLD and disappearing by the time TLD accelerates. Quantity and kinetics of fragmentation also stays unchanged in hyper-TLD-exhibiting recBCD mutant, making its faster and deeper TLD independent of fragmentation as well. Elimination of fragmentation without affecting cellular metabolism did not abolish TLD in the thyA mutant, but reduced early TLD in the thyA recBCD mutant, suggesting replication-dependent, but undetectable by pulsed field gel, double-strand breaks contributed to TLD. Chromosomal fragmentation, but not TLD, was eliminated in both the thyA and thyA recBCD mutants harboring deoCABD operon. Expression of a single gene, deoA, encoding thymidine phosphorylase, was sufficient to abolish fragmentation, suggesting thymidine-to-thymine interconversion during T-starvation being a key factor. Overall, this study reveals that chromosomal fragmentation, a direct consequence of T-starvation, is either dispensable or redundant for the overall TLD pathology, including hyper-TLD in the recBCD mutant. Replication forks, unlike chromosomal fragmentation, may provide minor contribution to TLD, but only in the repair-deficient thyA deoCABD recBCD mutant.

Oxidative damage blocks thymineless death and trimethoprim poisoning in Escherichia coli

Cells that cannot synthesize one of the DNA precursors, dTTP, due to thyA mutation or metabolic poisoning, undergo thymineless death (TLD), - a chromosome-based phenomenon of unclear mechanisms. In E. coli, thymineless death is caused either by denying thyA mutants thymidine supplementation or by treating wild type cells with trimethoprim. Two recent reports promised a potential breakthrough in TLD understanding, suggesting significant oxidative damage during thymine starvation. Oxidative damage in vivo comes from Fenton's reaction, when hydrogen peroxide meets ferrous iron to produce hydroxyl radical. Therefore, TLD could kill via irreparable double-strand breaks behind replication forks, when starvation-caused single-strand DNA gaps are attacked by hydroxyl radicals. We tested the proposed Fenton-TLD connection, in both thyA mutants denied thymidine, as well as in trimethoprim-treated WT cells, under three conditions: 1) intracellular iron chelation; 2) mutational inactivation of hydrogen peroxide (HP) scavenging; 3) acute treatment with sublethal HP concentrations. We found that TLD kinetics are affected by neither iron chelation, nor HP stabilization in cultures, indicating no induction of oxidative damage during thymine starvation. Moreover, acute exogenous HP treatments completely block TLD, apparently by blocking cell division - which may be a novel TLD prerequisite. Separately, the acute trimethoprim sensitivity of the rffC and recBCD mutants demonstrates how bactericidal power of this antibiotic could be amplified by inhibiting the corresponding enzymes. Importance Mysterious thymineless death strikes cells that are starved for thymine and therefore replicating their chromosomal DNA without dTTP. After 67 years of experiments testing various obvious and not so obvious explanations, thymineless death is still without a mechanism. Recently, oxidative damage via in vivo Fenton's reaction was proposed as a critical contributor to the irreparable chromosome damage during thymine starvation. We have tested this idea by either blocking in vivo Fenton's reaction (expecting no thymineless death) or by amplifying oxidative damage (expecting hyper thymineless death). Instead, we found that blocking Fenton's reaction has no influence on thymineless death, while amplifying oxidative damage prevents thymineless death altogether. Thus, oxidative damage does not contribute to thymineless death, while the latter remains enigmatic.

Electron Microscopy Reveals Unexpected Cytoplasm and Envelope Changes during Thymineless Death in Escherichia coli

Bacterial rod-shaped cells experiencing irreparable chromosome damage should filament without other morphological changes. Thymineless death (TLD) strikes thymidine auxotrophs denied external thymine/thymidine (T) supplementation. Such T-starved cells cannot produce the DNA precursor dTTP and therefore stop DNA replication. Stalled replication forks in T-starved cells were always assumed to experience mysterious chromosome lesions, but TLD was recently found to happen even without origin-dependent DNA replication, with the chromosome still remaining the main TLD target. T starvation also induces morphological changes, as if thymidine prevents cell envelope or cytoplasm problems that otherwise translate into chromosome damage. Here, we used transmission electron microscopy (TEM) to examine cytoplasm and envelope changes in T-starved Escherichia coli cells, using treatment with a DNA gyrase inhibitor as a control for "pure" chromosome death. Besides the expected cell filamentation in response to both treatments, we see the following morphological changes specific for T starvation and which might lead to chromosome damage: (i) significant cell widening, (ii) nucleoid diffusion, (iii) cell pole damage, and (iv) formation of numerous cytoplasmic bubbles. We conclude that T starvation does impact both the cytoplasm and the cell envelope in ways that could potentially affect the chromosome. IMPORTANCE Thymineless death is a dramatic and medically important phenomenon, the mechanisms of which remain a mystery. Unlike most other auxotrophs in the absence of the required supplement, thymidine-requiring E. coli mutants not only go static in the absence of thymidine, but rapidly die of chromosomal damage of unclear nature. Since this chromosomal damage is independent of replication, we examined fine morphological changes in cells undergoing thymineless death in order to identify what could potentially affect the chromosome. Here, we report several cytoplasm and cell envelope changes that develop in thymidine-starved cells but not in gyrase inhibitor-treated cells (negative control) that could be linked to subsequent irreparable chromosome damage. This is the first electron microscopy study of cells undergoing "genetic death" due to irreparable chromosome lesions.

Exopolysaccharide defects cause hyper-thymineless death in Escherichia coli via massive loss of chromosomal DNA and cell lysis

Thymineless death in Escherichia coli thyA mutants growing in the absence of thymidine (dT) is preceded by a substantial resistance phase, during which the culture titer remains static, as if the chromosome has to accumulate damage before ultimately failing. Significant chromosomal replication and fragmentation during the resistance phase could provide appropriate sources of this damage. Alternatively, the initial chromosomal replication in thymine (T)-starved cells could reflect a considerable endogenous dT source, making the resistance phase a delay of acute starvation, rather than an integral part of thymineless death. Here we identify such a low-molecular-weight (LMW)-dT source as mostly dTDP-glucose and its derivatives, used to synthesize enterobacterial common antigen (ECA). The thyA mutant, in which dTDP-glucose production is blocked by the rfbA rffH mutations, lacks a LMW-dT pool, the initial DNA synthesis during T-starvation and the resistance phase. Remarkably, the thyA mutant that makes dTDP-glucose and initiates ECA synthesis normally yet cannot complete it due to the rffC defect, maintains a regular LMW-dT pool, but cannot recover dTTP from it, and thus suffers T-hyperstarvation, dying precipitously, completely losing chromosomal DNA and eventually lysing, even without chromosomal replication. At the same time, its ECA⁺ thyA parent does not lyse during T-starvation, while both the dramatic killing and chromosomal DNA loss in the ECA-deficient thyA mutants precede cell lysis. We conclude that: 1) the significant pool of dTDP-hexoses delays acute T-starvation; 2) T-starvation destabilizes even nonreplicating chromosomes, while T-hyperstarvation destroys them; and 3) beyond the chromosome, T-hyperstarvation also destabilizes the cell envelope.

The Roles of Bacterial DNA Double-Strand Break Repair Proteins in Chromosomal DNA Replication

It is well established that DNA double-strand break (DSB) repair is required to underpin chromosomal DNA replication. Because DNA replication forks are prone to breakage, faithful DSB repair and correct replication fork restart are critically important. Cells, where the proteins required for DSB repair are absent or altered, display characteristic disturbances to genome replication. In this review, we analyze how bacterial DNA replication is perturbed in DSB repair mutant strains and explore the consequences of these perturbations for bacterial chromosome segregation and cell viability. Importantly, we look at how DNA replication and DSB repair processes are implicated in the striking recent observations of DNA amplification and DNA loss in the chromosome terminus of various mutant Escherichia coli strains. We also address the mutant conditions required for the remarkable ability to copy the entire E. coli genome, and to maintain cell viability, even in the absence of replication initiation from oriC, the unique origin of DNA replication in wild type cells. Furthermore, we discuss the models that have been proposed to explain these phenomena and assess how these models fit with the observed data, provide new insights and enhance our understanding of chromosomal replication and termination in bacteria.

Thymineless Death in Escherichia coli Is Unaffected by Chromosomal Replication Complexity

Thymineless death (TLD) is a rapid loss of viability of unclear mechanism in cultures of thyA mutants starved for thymine/thymidine (T starvation). It is accepted that T starvation repeatedly breaks replication forks, while recombinational repair restores them, but when the resulting futile breakage-repair cycle affects the small replication bubbles at oriC, the origin is degraded, killing the cell. Indeed, cells with increased chromosomal replication complexity (CRC), expressed as an elevated origin/terminus (ori/ter) ratio, die more extensively during TLD. Here we tested this logic by elevating the CRC in Escherichia coli thyA mutants before T starvation, anticipating exaggerated TLD. Unexpectedly, TLD remained unaffected by a CRC increase to either the natural limit (ori/ter ratio, ∼6) or the functional limit (ori/ter ratio, ∼16). Moreover, when we forced the CRC over the functional limit (ori/ter ratio, ∼30), TLD lessened. Thus, prior overinitiation does not sensitize cells to TLD. In contradiction with the published results, even blocking new replication initiations by the dnaA(Ts) defect at 42°C fails to prevent TLD. Using the thyA dnaA(Ts) mutant in a new T starvation protocol that excludes new initiations, we show that at 42°C, the same degree of TLD still occurs when chromosomes are demonstrably nonreplicating. Remarkably, 80% of the chromosomal DNA in these nonreplicating T-starved cells is still lost, by an unclear mechanism.IMPORTANCE Thymineless death kills cells of any type and is used in anticancer and antimicrobial treatments. We tested the idea that the more replication forks there are in the chromosome during growth, the more extensive the resulting thymineless death. We varied the number of replication forks in the Escherichia coli chromosome, as measured by the origin-to-terminus ratio, ranging it from the normal 2 to 60, and even completely eliminated replication forks in the nonreplicating chromosomes (ori/ter ratio = 1). Unexpectedly, we found that thymineless death is unaffected by the intensity of replication or by its complete absence; we also found that even nonreplicating chromosomes still disappear during thymine starvation. We conclude that thymineless death can kill E. coli independently of chromosomal replication.

Sources of thymidine and analogs fueling futile damage-repair cycles and ss-gap accumulation during thymine starvation in Escherichia coli

Thymine deprivation in thyA mutant E. coli causes thymineless death (TLD) and is the mode of action of popular antibacterial and anticancer drugs, yet the mechanisms of TLD are still unclear. TLD comprises three defined phases: resistance, rapid exponential death (RED) and survival, with the nature of the resistance phase and of the transition to the RED phase holding key to TLD pathology. We propose that a limited source of endogenous thymine maintains replication forks through the resistance phase. When this source ends, forks undergo futile break-repair cycle during the RED phase, eventually rendering the chromosome non-functional. Two obvious sources of the endogenous thymine are degradation of broken chromosomal DNA and recruitment of thymine from stable RNA. However, mutants that cannot degrade broken chromosomal DNA or lack ribo-thymine, instead of shortening the resistance phase, deepen the RED phase, meaning that only a small fraction of T-starved cells tap into these sources. Interestingly, the substantial chromosomal DNA accumulation during the resistance phase is negated during the RED phase, suggesting futile cycle of incorporation and excision of wrong nucleotides. We tested incorporation of dU or rU, finding some evidence for both, but DNA-dU incorporation accelerates TLD only when intracellular [dUTP] is increased by the dut mutation. In the dut ung mutant, with increased DNA-dU incorporation and no DNA-dU excision, replication is in fact rescued even without dT, but TLD still occurs, suggesting different mechanisms. Finally, we found that continuous DNA synthesis during thymine starvation makes chromosomal DNA increasingly single-stranded, and even the dut ung defect does not completely block this ss-gap accumulation. We propose that instability of single-strand gaps underlies the pathology of thymine starvation.

Near-continuously synthesized leading strands in Escherichia coli are broken by ribonucleotide excision

In vitro, purified replisomes drive model replication forks to synthesize continuous leading strands, even without ligase, supporting the semidiscontinuous model of DNA replication. However, nascent replication intermediates isolated from ligase-deficient Escherichia coli comprise only short (on average 1.2-kb) Okazaki fragments. It was long suspected that cells replicate their chromosomal DNA by the semidiscontinuous mode observed in vitro but that, in vivo, the nascent leading strand was artifactually fragmented postsynthesis by excision repair. Here, using high-resolution separation of pulse-labeled replication intermediates coupled with strand-specific hybridization, we show that excision-proficient E. coli generates leading-strand intermediates >10-fold longer than lagging-strand Okazaki fragments. Inactivation of DNA-repair activities, including ribonucleotide excision, further increased nascent leading-strand size to ∼80 kb, while lagging-strand Okazaki fragments remained unaffected. We conclude that in vivo, repriming occurs ∼70× less frequently on the leading versus lagging strands, and that DNA replication in E. coli is effectively semidiscontinuous.

Genomic Analysis of DNA Double-Strand Break Repair in Escherichia coli

Counting DNA whole genome sequencing reads is providing new insight into DNA double-strand break repair (DSBR) in the model organism Escherichia coli. We describe the application of RecA chromatin immunoprecipitation coupled to genomic DNA sequencing (RecA-ChIP-seq) and marker frequency analysis (MFA) to analyze the genomic consequences of DSBR. We provide detailed procedures for the preparation of DNA and the analysis of data. We compare different ways of visualizing ChIP data and show that alternative protocols for the preparation of DNA for MFA differentially affect the recovery of branched DNA molecules containing Holliday junctions.

Incomplete base excision repair contributes to cell death from antibiotics and other stresses

Numerous lethal stresses in bacteria including antibiotics, thymineless death, and MalE-LacZ expression trigger an increase in the production of reactive oxygen species. This results in the oxidation of the nucleotide pool by radicals produced by Fenton chemistry. Following the incorporation of these oxidized nucleotides into the genome, the cell's unsuccessful attempt to repair these lesions through base excision repair (BER) contributes causally to the lethality of these stresses. We review the evidence for this phenomenon of incomplete BER-mediated cell death and discuss how better understanding this pathway could contribute to the development of new antibiotics.

Replication-transcription conflicts trigger extensive DNA degradation in Escherichia coli cells lacking RecBCD

Bacterial chromosome duplication is initiated at a single origin (oriC). Two forks are assembled and proceed in opposite directions with high speed and processivity until they fuse and terminate in a specialised area opposite to oriC. Proceeding forks are often blocked by tightly-bound protein-DNA complexes, topological strain or various DNA lesions. In Escherichia coli the RecBCD protein complex is a key player in the processing of double-stranded DNA (dsDNA) ends. It has important roles in the repair of dsDNA breaks and the restart of forks stalled at sites of replication-transcription conflicts. In addition, ΔrecB cells show substantial amounts of DNA degradation in the termination area. In this study we show that head-on encounters of replication and transcription at a highly-transcribed rrn operon expose fork structures to degradation by nucleases such as SbcCD. SbcCD is also mostly responsible for the degradation in the termination area of ΔrecB cells. However, additional processes exacerbate degradation specifically in this location. Replication profiles from ΔrecB cells in which the chromosome is linearized at two different locations highlight that the location of replication termination can have some impact on the degradation observed. Our data improve our understanding of the role of RecBCD at sites of replication-transcription conflicts as well as the final stages of chromosome duplication. However, they also highlight that current models are insufficient and cannot explain all the molecular details in cells lacking RecBCD.

Broken replication forks trigger heritable DNA breaks in the terminus of a circular chromosome

It was recently reported that the recBC mutants of Escherichia coli, deficient for DNA double-strand break (DSB) repair, have a decreased copy number of their terminus region. We previously showed that this deficit resulted from DNA loss after post-replicative breakage of one of the two sister-chromosome termini at cell division. A viable cell and a dead cell devoid of terminus region were thus produced and, intriguingly, the reaction was transmitted to the following generations. Using genome marker frequency profiling and observation by microscopy of specific DNA loci within the terminus, we reveal here the origin of this phenomenon. We observed that terminus DNA loss was reduced in a recA mutant by the double-strand DNA degradation activity of RecBCD. The terminus-less cell produced at the first cell division was less prone to divide than the one produced at the next generation. DNA loss was not heritable if the chromosome was linearized in the terminus and occurred at chromosome termini that were unable to segregate after replication. We propose that in a recB mutant replication fork breakage results in the persistence of a linear DNA tail attached to a circular chromosome. Segregation of the linear and circular parts of this "σ-replicating chromosome" causes terminus DNA breakage during cell division. One daughter cell inherits a truncated linear chromosome and is not viable. The other inherits a circular chromosome attached to a linear tail ending in the chromosome terminus. Replication extends this tail, while degradation of its extremity results in terminus DNA loss. Repeated generation and segregation of new σ-replicating chromosomes explains the heritability of post-replicative breakage. Our results allow us to determine that in E. coli at each generation, 18% of cells are subject to replication fork breakage at dispersed, potentially random, chromosomal locations.

Degradation of RNA during lysis of Escherichia coli cells in agarose plugs breaks the chromosome

The nucleoid of Escherichia coli comprises DNA, nucleoid associated proteins (NAPs) and RNA, whose role is unclear. We found that lysing bacterial cells embedded in agarose plugs in the presence of RNases caused massive fragmentation of the chromosomal DNA. This RNase-induced chromosomal fragmentation (RiCF) was completely dependent on the presence of RNase around lysing cells, while the maximal chromosomal breakage required fast cell lysis. Cell lysis in plugs without RNAse made the chromosomal DNA resistant to subsequent RNAse treatment. RiCF was not influenced by changes in the DNA supercoiling, but was influenced by growth temperature or age of the culture. RiCF was partially dependent on H-NS, histone-like nucleoid structuring- and global transcription regulator protein. The hupAB deletion of heat-unstable nucleoid protein (HU) caused increase in spontaneous fragmentation that was further increased when combined with deletions in two non-coding RNAs, nc1 and nc5. RiCF was completely dependent upon endonuclease I, a periplasmic deoxyribonuclease that is normally found inhibited by cellular RNA. Unlike RiCF, the spontaneous fragmentation in hupAB nc1 nc5 quadruple mutant was resistant to deletion of endonuclease I. RiCF-like phenomenon was observed without addition of RNase to agarose plugs if EDTA was significantly reduced during cell lysis. Addition of RNase under this condition was synergistic, breaking chromosomes into pieces too small to be retained by the pulsed field gels. RNase-independent fragmentation was qualitatively and quantitatively comparable to RiCF and was partially mediated by endonuclease I.

Replication fork progression is paused in two large chromosomal zones flanking the DNA replication origin in Escherichia coli

Although the speed of nascent DNA synthesis at individual replication forks is relatively uniform in bacterial cells, the dynamics of replication fork progression on the chromosome are hampered by a variety of natural impediments. Genome replication dynamics can be directly measured from an exponentially growing cell population by sequencing newly synthesized DNA strands that were specifically pulse-labeled with the thymidine analogue 5-bromo-2'-deoxyuridine (BrdU). However, a short pulse labeling with BrdU is impracticable for bacteria because of poor incorporation of BrdU into the cells, and thus, the genomewide dynamics of bacterial DNA replication remain undetermined. Using a new thymidine-requiring Escherichia coli strain, eCOMB, and high-throughput sequencing, we succeeded in determining the genomewide replication profile in bacterial cells. We also found that fork progression is paused in two ~200-kb chromosomal zones that flank the replication origin in the growing cells. This origin-proximal obstruction to fork progression was overcome by an increased thymidine concentration in the culture medium and enhanced by inhibition of transcription. These indicate that DNA replication near the origin is sensitive to the impediments to fork progression, namely a scarcity of the DNA precursor deoxythymidine triphosphate and probable conflicts between replication and transcription machineries.

Transcriptome Analysis of Escherichia coli during dGTP Starvation

Our laboratory recently discovered that Escherichia coli cells starved for the DNA precursor dGTP are killed efficiently (dGTP starvation) in a manner similar to that described for thymineless death (TLD). Conditions for specific dGTP starvation can be achieved by depriving an E. coli optA1 gpt strain of the purine nucleotide precursor hypoxanthine (Hx). To gain insight into the mechanisms underlying dGTP starvation, we conducted genome-wide gene expression analyses of actively growing optA1 gpt cells subjected to hypoxanthine deprivation for increasing periods. The data show that upon Hx withdrawal, the optA1 gpt strain displays a diminished ability to derepress the de novo purine biosynthesis genes, likely due to internal guanine accumulation. The impairment in fully inducing the purR regulon may be a contributing factor to the lethality of dGTP starvation. At later time points, and coinciding with cell lethality, strong induction of the SOS response was observed, supporting the concept of replication stress as a final cause of death. No evidence was observed in the starved cells for the participation of other stress responses, including the rpoS-mediated global stress response, reinforcing the lack of feedback of replication stress to the global metabolism of the cell. The genome-wide expression data also provide direct evidence for increased genome complexity during dGTP starvation, as a markedly increased gradient was observed for expression of genes located near the replication origin relative to those located toward the replication terminus.

IMPORTANCE

Control of the supply of the building blocks (deoxynucleoside triphosphates [dNTPs]) for DNA replication is important for ensuring genome integrity and cell viability. When cells are starved specifically for one of the four dNTPs, dGTP, the process of DNA replication is disturbed in a manner that can lead to eventual death. In the present study, we investigated the transcriptional changes in the bacterium E. coli during dGTP starvation. The results show increasing DNA replication stress with an increased time of starvation, as evidenced by induction of the bacterial SOS system, as well as a notable lack of induction of other stress responses that could have saved the cells from cell death by slowing down cell growth.

Static and Dynamic Factors Limit Chromosomal Replication Complexity in Escherichia coli, Avoiding Dangers of Runaway Overreplication

We define chromosomal replication complexity (CRC) as the ratio of the copy number of the most replicated regions to that of unreplicated regions on the same chromosome. Although a typical CRC of eukaryotic or bacterial chromosomes is 2, rapidly growing Escherichia coli cells induce an extra round of replication in their chromosomes (CRC = 4). There are also E. coli mutants with stable CRC∼6. We have investigated the limits and consequences of elevated CRC in E. coli and found three limits: the "natural" CRC limit of ∼8 (cells divide more slowly); the "functional" CRC limit of ∼22 (cells divide extremely slowly); and the "tolerance" CRC limit of ∼64 (cells stop dividing). While the natural limit is likely maintained by the eclipse system spacing replication initiations, the functional limit might reflect the capacity of the chromosome segregation system, rather than dedicated mechanisms, and the tolerance limit may result from titration of limiting replication factors. Whereas recombinational repair is beneficial for cells at the natural and functional CRC limits, we show that it becomes detrimental at the tolerance CRC limit, suggesting recombinational misrepair during the runaway overreplication and giving a rationale for avoidance of the latter.

Fork restart protein, PriA, binds around oriC after depletion of nucleotide precursors: Replication fork arrest near the replication origin

Arrest of replication fork progression is one of the most common causes for increasing the genomic instability. In bacteria, PriA, a conserved DEXH-type helicase, plays a major role in recognition of the stalled forks and restart of DNA replication. We took advantage of PriA's ability to specifically recognize stalled replication forks to determine the genomic loci where replication forks are prone to stall on the Escherichia coli genome. We found that PriA binds around oriC upon thymine starvation which reduces the nucleotide supply and causes replication fork stalling. PriA binding quickly disappeared upon readdition of thymine. Furthermore, BrdU was incorporated at around oriC upon release from thymine starvation. Our results indicate that reduced supply of DNA replication precursors causes replication fork stalling preferentially in the 600 kb segment centered at oriC. This suggests that replication of the vicinity of oriC requires higher level of nucleotide precursors. The results also point to a possibility of slow fork movement and/or the presence of multiple fork arrest signals within this segment. Indeed, we have identified rather strong fork stall/pausing signals symmetrically located at ∼50 kb away from oriC. We speculate that replication pausing and fork-slow-down shortly after initiation may represent a novel checkpoint that ensures the presence of sufficient nucleotide supply prior to commitment to duplication of the entire genome.

A DinB Ortholog Enables Mycobacterial Growth under dTTP-Limiting Conditions Induced by the Expression of a Mycobacteriophage-Derived Ribonucleotide Reductase Gene

Mycobacterium species such as M. smegmatis and M. tuberculosis encode at least two translesion synthesis (TLS) polymerases, DinB1 and DinB2, respectively. Although predicted to be linked to DNA repair, their role in vivo remains enigmatic. M. smegmatis mc(2)155, a strain commonly used to investigate mycobacterial genetics, has two copies of dinB2, the gene that codes for DinB2, by virtue of a 56-kb chromosomal duplication. Expression of a mycobacteriophage D29 gene (gene 50) encoding a class II ribonucleotide reductase in M. smegmatis ΔDRKIN, a strain derived from mc(2)155 in which one copy of the duplication is lost, resulted in DNA replication defects and growth inhibition. The inhibitory effect could be linked to the deficiency of dTTP that resulted under these circumstances. The selective inhibition observed in the ΔDRKIN strain was found to be due solely to a reduced dosage of dinB2 in this strain. Mycobacterium bovis, which is closely related to M. tuberculosis, the tuberculosis pathogen, was found to be highly susceptible to gene 50 overexpression. Incidentally, these slow-growing pathogens harbor one copy of dinB2. The results indicate that the induction of a dTTP-limiting state can lead to growth inhibition in mycobacteria, with the effect being maximum in cells deficient in DinB2.

IMPORTANCE

Mycobacterium species, such as M. tuberculosis, the tuberculosis pathogen, are known to encode several Y family DNA polymerases, one of which is DinB2, an ortholog of the DNA repair-related protein DinP of Escherichia coli. Although this protein has been biochemically characterized previously and found to be capable of translesion synthesis in vitro, its in vivo function remains unknown. Using a novel method to induce dTTP deficiency in mycobacteria, we demonstrate that DinB2 can aid mycobacterial survival under such conditions. Apart from unraveling a specific role for the mycobacterial Y family DNA polymerase DinB2 for the first time, this study also paves the way for the development of drugs that can kill mycobacteria by inducing a dTTP-deficient state.

Thymineless Death Lives On: New Insights into a Classic Phenomenon

The primary mechanisms by which bacteria lose viability when deprived of thymine have been elusive for over half a century. Early research focused on stalled replication forks and the deleterious effects of uracil incorporation into DNA from thymidine-deficient nucleotide pools. The initiation of the replication cycle and origin-proximal DNA degradation during thymine starvation have now been quantified via whole-genome microarrays and other approaches. These advances have fostered innovative models and informative experiments in bacteria since this topic was last reviewed. Given that thymineless death is similar in mammalian cells and that certain antibacterial and chemotherapeutic drugs elicit thymine deficiency, a mechanistic understanding of this phenomenon might have valuable biomedical applications.

A balanced perspective on unbalanced growth and thymineless death

The early history of the esoteric phenomenon of thymineless death (TLD) is recounted, from the pioneering discovery by Seymour Cohen and Hazel Barner, through my graduate studies at Yale and postdoctoral research in Copenhagen. My principal contribution was the discovery that restricted synthesis of protein and RNA permits cultures of Escherichia coli to complete their DNA replication cycles without initiating new ones, and that cells held in this physiological state are immune to the lethality of thymine deprivation; unbalanced growth is not the fundamental cause of TLD. The successful synchronization of the DNA replication cycle contributed to formulation of the replicon concept. Studies at Stanford revealed a specific requirement for transcription and led to the discovery of a TLD-resistant mutant in a new gene, termed recQ, with important homologs in humans and most other organisms. The lessons learned from research on TLD underscore the value of basic research in bacterial systems that can have profound implications for human health.

Thymineless death, at the origin

Thymineless death (TLD) in bacteria has been a focus of research for decades. Nevertheless, the advances in the last 5 years, with Escherichia coli as the model organism, have been outstanding. Independent research groups have presented compelling results that establish that the initiation of chromosome replication under thymine starvation is a key element in the scenario of TLD. Here we review the experimental results linking the initiation of replication to the lethality under thymine starvation and the proposed mechanisms by which TLD occurs. The concept of this relationship was ‘in the air,’ but approaches were not sufficiently developed to demonstrate the crucial role of DNA initiation in TLD. Genome-wide marker frequency analysis and Two Dimensional agarose gel electrophoresis have been critical methods employed to reveal that initiation events and the degradation of the oriC region occur during thymine starvation. The relationships between these events and TLD have established them to be the main underlying causes of the lethality under thymine starvation. Furthermore, we summarize additional important findings from the study of different mutant strains, which support the idea that the initiation of chromosomal replication and TLD are connected.

Perturbed states of the bacterial chromosome: a thymineless death case study

Spatial patterns of transcriptional activity in the living genome of Escherichia coli represent one of the more peculiar aspects of the E. coli chromosome biology. Spatial transcriptional correlations can be observed throughout the chromosome, and their formation depends on the state of replication in the cell. The condition of thymine starvation leading to thymineless death (TLD) is at the "cross-roads" of replication and transcription. According to a current view, e.g., (Cagliero et al., 2014), one of the cellular objectives is to segregate the processes of transcription and replication in time and space. An ultimate segregation would take place when one process is inhibited and another is not, as it happens during thymine starvation, which results in numerous molecular and physiological abnormalities associated with TLD. One of such abnormalities is the loss of spatial correlations in the vicinity of the origin of replication. We review the transcriptional consequences of replication inhibition by thymine starvation in a context of the state of DNA template in the starved cells and opine about a possible significance of normal physiological coupling between the processes of replication and transcription.

End of the beginning: elongation and termination features of alternative modes of chromosomal replication initiation in bacteria

In bacterial cells, bidirectional replication of the circular chromosome is initiated from a single origin (oriC) and terminates in an antipodal terminus region such that movement of the pair of replication forks is largely codirectional with transcription. The terminus region is flanked by discrete Ter sequences that act as polar, or direction-dependent, arrest sites for fork progression. Alternative oriC-independent modes of replication initiation are possible, one of which is constitutive stable DNA replication (cSDR) from transcription-associated RNA–DNA hybrids or R-loops. Here, I discuss the distinctive attributes of fork progression and termination associated with different modes of bacterial replication initiation. Two hypothetical models are proposed: that head-on collisions between pairs of replication forks, which are a feature of replication termination in all kingdoms of life, provoke bilateral fork reversal reactions; and that cSDR is characterized by existence of distinct subpopulations in bacterial cultures and a widespread distribution of origins in the genome, each with a small firing potential. Since R-loops are known to exist in eukaryotic cells and to inflict genome damage in G1 phase, it is possible that cSDR-like events promote aberrant replication initiation even in eukaryotes.

Replication fork inhibition in seqA mutants of Escherichia coli triggers replication fork breakage

SeqA protein negatively regulates replication initiation in Escherichia coli and is also proposed to organize maturation and segregation of the newly replicated DNA. The seqA mutants suffer from chromosomal fragmentation; since this fragmentation is attributed to defective segregation or nucleoid compaction, two-ended breaks are expected. Instead, we show that, in SeqA's absence, chromosomes mostly suffer one-ended DNA breaks, indicating disintegration of replication forks. We further show that replication forks are unexpectedly slow in seqA mutants. Quantitative kinetics of origin and terminus replication from aligned chromosomes not only confirm origin overinitiation in seqA mutants, but also reveal terminus under-replication, indicating inhibition of replication forks. Pre-/post-labelling studies of the chromosomal fragmentation in seqA mutants suggest events involving single forks, rather than pairs of forks from consecutive rounds rear-ending into each other. We suggest that, in the absence of SeqA, the sister-chromatid cohesion 'safety spacer' is destabilized and completely disappears if the replication fork is inhibited, leading to the segregation fork running into the inhibited replication fork and snapping the latter at single-stranded DNA regions.

dGTP starvation in Escherichia coli provides new insights into the thymineless-death phenomenon

Starvation of cells for the DNA building block dTTP is strikingly lethal (thymineless death, TLD), and this effect is observed in all organisms. The phenomenon, discovered some 60 years ago, is widely used to kill cells in anticancer therapies, but many questions regarding the precise underlying mechanisms have remained. Here, we show for the first time that starvation for the DNA precursor dGTP can kill E. coli cells in a manner sharing many features with TLD. dGTP starvation is accomplished by combining up-regulation of a cellular dGTPase with a deficiency of the guanine salvage enzyme guanine-(hypoxanthine)-phosphoribosyltransferase. These cells, when grown in medium without an exogenous purine source like hypoxanthine or adenine, display a specific collapse of the dGTP pool, slow-down of chromosomal replication, the generation of multi-branched nucleoids, induction of the SOS system, and cell death. We conclude that starvation for a single DNA building block is sufficient to bring about cell death.

Rifampicin suppresses thymineless death by blocking the transcription-dependent step of chromosome initiation

Thymineless death (TLD), a phenomenon in which thymine auxotrophy becomes lethal when cells are starved of thymine, can be prevented by the presence of rifampicin, an RNA polymerase inhibitor. Several lines of evidence link TLD to chromosome initiation events. This suggests that rifampicin-mediated TLD suppression could be due to the inhibition of RNA synthesis required for DNA chromosomal initiation at oriC, although other mechanisms cannot be discarded. In this work, we show that the addition of different rifampicin concentrations to thymine-starved cells modulates TLD and chromosomal initiation capacity (ChIC). Time-lapse experiments find increasing levels of ChIC during thymine starvation correlated with the accumulation of simple-Y, double-Y and bubble arc replication intermediates at the oriC region as visualized by two-dimensional DNA agarose gel electrophoresis. None of these structures were observed following rifampicin addition or under genetic-physiological conditions that suppress TLD, indicating that abortive chromosome replication initiations under thymine starvation are crucial for this lethality. Significantly, the introduction of mioC and gid mutations which alter transcription levels around oriC, reduces ChIC and alleviates TLD. These results show that the impairment of transcription-dependent initiation caused by rifampicin addition, is responsible for TLD suppression. Our findings here may provide new avenues for the development of improved antibacterial treatments and chemotherapies based on thymine starvation-induced cell death.

Uracil DNA glycosylase (UNG) loss enhances DNA double strand break formation in human cancer cells exposed to pemetrexed

Misincorporation of genomic uracil and formation of DNA double strand breaks (DSBs) are known consequences of exposure to TS inhibitors such as pemetrexed. Uracil DNA glycosylase (UNG) catalyzes the excision of uracil from DNA and initiates DNA base excision repair (BER). To better define the relationship between UNG activity and pemetrexed anticancer activity, we have investigated DNA damage, DSB formation, DSB repair capacity, and replication fork stability in UNG^+/+ and UNG^−/− cells. We report that despite identical growth rates and DSB repair capacities, UNG^−/− cells accumulated significantly greater uracil and DSBs compared with UNG^+/+ cells when exposed to pemetrexed. ChIP-seq analysis of γ-H2AX enrichment confirmed fewer DSBs in UNG^+/+ cells. Furthermore, DSBs in UNG^+/+ and UNG^−/− cells occur at distinct genomic loci, supporting differential mechanisms of DSB formation in UNG-competent and UNG-deficient cells. UNG^−/− cells also showed increased evidence of replication fork instability (PCNA dispersal) when exposed to pemetrexed. Thymidine co-treatment rescues S-phase arrest in both UNG^+/+ and UNG^−/− cells treated with IC₅₀-level pemetrexed. However, following pemetrexed exposure, UNG^−/− but not UNG^+/+ cells are refractory to thymidine rescue, suggesting that deficient uracil excision rather than dTTP depletion is the barrier to cell cycle progression in UNG^−/− cells. Based on these findings we propose that pemetrexed-induced uracil misincorporation is genotoxic, contributing to replication fork instability, DSB formation and ultimately cell death.

The phtC-phtD locus equips Legionella pneumophila for thymidine salvage and replication in macrophages

The phagosomal transporter (Pht) family of the major facilitator superfamily (MFS) is encoded by phylogenetically related intracellular gammaproteobacteria, including the opportunistic pathogen Legionella pneumophila. The location of the pht genes between the putative thymidine kinase (tdk) and phosphopentomutase (deoB) genes suggested that the phtC and phtD loci contribute to thymidine salvage in L. pneumophila. Indeed, a phtC(+) allele in trans restored pyrimidine uptake to an Escherichia coli mutant that lacked all known nucleoside transporters, whereas a phtD(+) allele did not. The results of phenotypic analyses of L. pneumophila strains lacking phtC or phtD strongly indicate that L. pneumophila requires PhtC and PhtD function under conditions where sustained dTMP synthesis is compromised. First, in broth cultures that mimicked thymidine limitation or starvation, L. pneumophila exhibited a marked requirement for PhtC function. Conversely, mutation of phtD conferred a survival advantage. Second, in medium that lacked thymidine, multicopy phtC(+) or phtD(+) alleles enhanced the survival of L. pneumophila thymidylate synthase (thyA)-deficient strains, which cannot synthesize dTMP endogenously. Third, under conditions in which transport of the pyrimidine nucleoside analog 5-fluorodeoxyuridine (FUdR) would inhibit growth, PhtC and PhtD conferred a growth advantage to L. pneumophila thyA(+) strains. Finally, when cultured in macrophages, L. pneumophila required the phtC-phtD locus to replicate. Accordingly, we propose that PhtC and PhtD contribute to protect L. pneumophila from dTMP starvation during its intracellular life cycle.

Replication of the Escherichia coli chromosome in RNase HI-deficient cells: multiple initiation regions and fork dynamics

DNA replication in Escherichia coli is normally initiated at a single origin, oriC, dependent on initiation protein DnaA. However, replication can be initiated elsewhere on the chromosome at multiple ectopic oriK sites. Genetic evidence indicates that initiation from oriK depends on RNA-DNA hybrids (R-loops), which are normally removed by enzymes such as RNase HI to prevent oriK from misfiring during normal growth. Initiation from oriK sites occurs in RNase HI-deficient mutants, and possibly in wild-type cells under certain unusual conditions. Despite previous work, the locations of oriK and their impact on genome stability remain unclear. We combined 2D gel electrophoresis and whole genome approaches to map genome-wide oriK locations. The DNA copy number profiles of various RNase HI-deficient strains contained multiple peaks, often in consistent locations, identifying candidate oriK sites. Removal of RNase HI protein also leads to global alterations of replication fork migration patterns, often opposite to normal replication directions, and presumably eukaryote-like replication fork merging. Our results have implications for genome stability, offering a new understanding of how RNase HI deficiency results in R-loop-mediated transcription-replication conflict, as well as inappropriate replication stalling or blockage at Ter sites outside of the terminus trap region and at ribosomal operons.

Thymineless death is inhibited by CsrA in Escherichia coli lacking the SOS response

Thymineless death (TLD) is the rapid loss of colony-forming ability in bacterial, yeast and human cells starved for thymine, and is the mechanism of action of common chemotherapeutic drugs. In Escherichia coli, significant loss of viability during TLD requires the SOS replication-stress/DNA-damage response, specifically its role in inducing the inhibitor of cell division, SulA. An independent RecQ- and RecJ-dependent TLD pathway accounts for a similarly large additional component of TLD, and a third SOS- and RecQ/J-independent TLD pathway has also been observed. Although two groups have implicated the SOS-response in TLD, an SOS-deficient mutant strain from an earlier study was found to be sensitive to thymine deprivation. We performed whole-genome resequencing on that SOS-deficient strain and find that, compared with the SOS-proficient control strain, it contains five mutations in addition to the SOS-blocking lexA(Ind(-)) mutation. One of the additional mutations, csrA, confers TLD sensitivity specifically in SOS-defective strains. We find that CsrA, a carbon storage regulator, reduces TLD in SOS- or SulA-defective cells, and that the increased TLD that occurs in csrA(-) SOS-defective cells is dependent on RecQ. We consider a hypothesis in which the modulation of nucleotide pools by CsrA might inhibit TLD specifically in SOS-deficient (SulA-deficient) cells.

Low-molecular-weight DNA replication intermediates in Escherichia coli: mechanism of formation and strand specificity

Chromosomal DNA replication intermediates, revealed in ligase-deficient conditions in vivo, are of low molecular weight (LMW) independently of the organism, suggesting discontinuous replication of both the leading and the lagging DNA strands. Yet, in vitro experiments with purified enzymes replicating sigma-structured substrates show continuous synthesis of the leading DNA strand in complete absence of ligase, supporting the textbook model of semi-discontinuous DNA replication. The discrepancy between the in vivo and in vitro results is rationalized by proposing that various excision repair events nick continuously synthesized leading strands after synthesis, producing the observed LMW intermediates. Here, we show that, in an Escherichia coli ligase-deficient strain with all known excision repair pathways inactivated, new DNA is still synthesized discontinuously. Furthermore, hybridization to strand-specific targets demonstrates that the LMW replication intermediates come from both the lagging and the leading strands. These results support the model of discontinuous leading strand synthesis in E. coli.

Metabolic and functional genomic studies identify deoxythymidylate kinase as a target in LKB1-mutant lung cancer

The LKB1/STK11 tumor suppressor encodes a serine/threonine kinase, which coordinates cell growth, polarity, motility, and metabolism. In non-small cell lung carcinoma, LKB1 is somatically inactivated in 25% to 30% of cases, often concurrently with activating KRAS mutations. Here, we used an integrative approach to define novel therapeutic targets in KRAS-driven LKB1-mutant lung cancers. High-throughput RNA interference screens in lung cancer cell lines from genetically engineered mouse models driven by activated KRAS with or without coincident Lkb1 deletion led to the identification of Dtymk, encoding deoxythymidylate kinase (DTYMK), which catalyzes dTTP biosynthesis, as synthetically lethal with Lkb1 deficiency in mouse and human lung cancer lines. Global metabolite profiling showed that Lkb1-null cells had a striking decrease in multiple nucleotide metabolites as compared with the Lkb1-wild-type cells. Thus, LKB1-mutant lung cancers have deficits in nucleotide metabolism that confer hypersensitivity to DTYMK inhibition, suggesting that DTYMK is a potential therapeutic target in this aggressive subset of tumors.