Occurrence and repair of alkylating stress in the intracellular pathogen Brucella abortus

Poncin, Katy; Roba, Agnès; Jimmidi, Ravikumar; Potemberg, Georges; Fioravanti, Antonella; Francis, Nayla; Willemart, Kévin; Zeippen, Nicolas; Machelart, Arnaud; Biondi, Emanuele G.; Muraille, Eric; Vincent, Stéphane P.; De Bolle, Xavier

doi:10.1038/s41467-019-12516-8

Download PDF

Article
Open access
Published: 24 October 2019

Occurrence and repair of alkylating stress in the intracellular pathogen Brucella abortus

Katy Poncin^1,2,
Agnès Roba¹,
Ravikumar Jimmidi³,
Georges Potemberg¹,
Antonella Fioravanti ORCID: orcid.org/0000-0002-4065-0839^4,5,
Nayla Francis¹,
Kévin Willemart¹,
Nicolas Zeippen¹,
Arnaud Machelart ORCID: orcid.org/0000-0002-4444-0011^1,6,
Emanuele G. Biondi⁴,
Eric Muraille^7,8,
Stéphane P. Vincent² &
…
Xavier De Bolle ORCID: orcid.org/0000-0002-3845-3043¹

Nature Communications volume 10, Article number: 4847 (2019) Cite this article

3069 Accesses
15 Citations
25 Altmetric
Metrics details

Subjects

Abstract

It is assumed that intracellular pathogenic bacteria have to cope with DNA alkylating stress within host cells. Here we use single-cell reporter systems to show that the pathogen Brucella abortus does encounter alkylating stress during the first hours of macrophage infection. Genes encoding direct repair and base-excision repair pathways are required by B. abortus to face this stress in vitro and in a mouse infection model. Among these genes, ogt is found to be under the control of the conserved cell-cycle transcription factor GcrA. Our results highlight that the control of DNA repair in B. abortus displays distinct features that are not present in model organisms such as Escherichia coli.

Bacterial DNA excision repair pathways

Article 24 February 2022

Antibiotic-induced DNA damage results in a controlled loss of pH homeostasis and genome instability

Article Open access 10 November 2020

The ALPK1/TIFA/NF-κB axis links a bacterial carcinogen to R-loop-induced replication stress

Article Open access 09 October 2020

Introduction

On DNA, alkylating stress typically results in aberrant methylation patterns. Many positions of DNA can be targeted and these modifications range from innocuous to mutagenic or cytotoxic^1,2. Alkylating agents are ubiquitous in the environment and also have endogenous and dietary sources^3,4.

Three main endogenous sources of alkylating agents are reported in living organisms: (1) the ubiquitous methyl-donor S-adenosylmethionine (SAM)⁵, (2) lipid peroxidation-derived alkylating agents⁶, and (3) N-nitroso compounds resulting from the nitrosation of metabolites and being either direct alkylating agents or requiring metabolic activation^7,8,9. SAM is often cited as the primary cause of endogenous alkylation, but its role has probably been overestimated in prokaryotic cells. Indeed, a 100-fold change of SAM levels was achieved experimentally without causing any change in the mutation rate of Escherichia coli¹⁰. Additionally, lipid peroxidation predominantly occurs on polyunsaturated fatty acids¹¹, which are rarely present in bacteria^12,13. Prokaryotic cells are thus considered to face marginal lipid-derived alkylating stress. In fact, in E. coli, the majority of the spontaneous mutations resulting from alkylating stress was found to be generated via the endogenous formation of N-nitroso compounds^7,14.

One question that remains unanswered is whether alkylating agents are produced by immune cells to fight intracellular pathogens. Of relevance, macrophages and neutrophils are known to produce N-nitroso compounds^15,16. However, the occurrence of alkylating stress on intracellular bacteria has not been detected as yet. It has been hypothesized that N-nitroso compounds could be differentially produced in subcellular compartments¹⁷, suggesting that vacuoles containing bacteria may also be prone to accumulating such compounds. Indeed, many intracellular bacteria first travel through an endosomal-derived vacuole, before reaching their replicative compartment¹⁸, as in Brucella abortus, the causative agent of brucellosis in animals and Malta fever in humans^19,20. This class III α-proteobacterium is known to enter host cells to form an endosomal Brucella-containing vacuole (eBCV), which becomes acidified to a pH of 4.0–4.5, suggesting that the eBCV undergoes normal endosomal maturation process, with the acquisition of the LampI marker^21,22. Later, in most cell types, the bacterium reaches the endoplasmic reticulum, its replicative niche (rBCV)^19,23,24.

Both prokaryotic cells and eukaryotic cells are competent for N-nitrosation reactions. In eukaryotic cells, the production of N-nitroso compounds is dependent on an acidic pH and the intracellular concentration of reactive nitrogen species (RNS)^15,25. However, in E. coli, a side reaction of nitrate reductases leads to the N-nitrosation of amines at a neutral pH⁷. Three classes of E. coli mutants have been shown to be deficient in nitrosation: (1) narG, encoding the catalytic subunit of nitrate reductase A; (2) fnr, encoding a pleotropic activator that influences the expression of the narGHIJ operon; and (3) moa, the mutant with the strongest nitrosation deficiency, probably because the enzyme is involved in the synthesis of a molybdopterin cofactor, required by all three E. coli nitrate reductases⁷. Importantly, the endogenous production of N-nitroso compounds is known to be more important in anaerobic and resting E. coli²⁶. In this respect, during the first hours of infection, while still inside the eBCV, B. abortus is known to be blocked in the G1 stage of its cell cycle, characterized by a non-growing and non-replicating state²⁷.

In the late 1970s, a specific response to alkylating stress was described in vitro for E. coli²⁸. This system was called “adaptive response” and is based on the detection of a methylphosphotriester (meP3ester) modification on DNA by the Ada²⁹. Most bacteria possess an Ada-based adaptive response, with some variation in the target genes and their genomic organization¹. In E. coli, ada is stochastically expressed to produce on average only one Ada protein per generation³⁰. The detected meP3ester group is captured on the cysteine 38 (C38) residue of Ada, which becomes active as a transcription factor, upregulating the expression of a series of genes coding for proteins dedicated to the repair of alkylated DNA. These proteins comprise Ada itself, which can also directly repair O⁶-methylguanine and O⁴-methylthymine via the capture of the methyl group by the C321 residue of Ada; the dioxygenase AlkB, involved in the direct repair of the mutagenic lesions N¹-methyladenine and N³-methylcytosine, as well as the DNA glycosylase AlkA, which removes the most cytotoxic lesion N³-methyladenine via the base excision repair (BER) pathway³¹. The aidB gene is also overexpressed by the adaptive response, but its function still remains elusive^32,33. There are also two proteins constitutively produced and independent of the adaptive system, namely the methyltransferase Ogt, which has a C139 residue with similar function than the C321 of Ada, and the glycosylase TagA, which is functionally similar to AlkA³¹. Importantly, other DNA repair pathways can be involved in repairing alkylated DNA. For instance, in E. coli, the SOS pathway—which is repressed by LexA under non-stress conditions—is activated early following alkylating stress, before the adaptive response takes over³⁴.

Here, we demonstrate that a weak alkylating stress occurs when B. abortus is inside its eBCV in a macrophage cell line. We also show that genes responsible for the response against alkylating stress are required by B. abortus to survive following mice intranasal infection. Our data indicate that B. abortus does not possess a functional Ada-based adaptive system, but instead relies on redundant repair pathways, partially dependent on the methylation-sensitive transcription factor GcrA and the SOS response, to cope with alkylating stress and subvert potentially mutagenic host environments.

Results

Conservation of alkylated DNA repair genes in bacteria

We rationalized that if most intracellular bacteria face alkylating stress, there would be a significant conservation of some alkylated DNA repair genes. We found that many intracellular bacteria, including obligate pathogens such as Chlamydia pneumoniae and Coxiella burnettii, have genes homologous to some of known DNA repair genes (Fig. 1). Note that Brucella species are predicted to be particularly well equipped against alkylating stress (Fig. 1 and Supplementary Fig. 1).

Alkylation stress is encountered by Brucella inside host cells

Until now, strategies to detect the presence of alkylating stress inside host cells have been based on the survival of alkylation-specific DNA repair enzymes. These approaches have been uninformative, probably because repair systems are redundant, or because the stress is too weak to be detected by CFU counting^35,36. Here, we took advantage of the ability of the auto-regulated Ada protein from E. coli to detect meP3ester groups on DNA³⁷ to create a transcription-based fluorescent reporter system. As the ada gene is in operon with alkB in E. coli³⁸, we replaced alkB with a superfolder gfp on a medium-copy plasmid (Fig. 2a). A mutated version of the reporter system was used as a negative control, in which a C38A mutation was introduced in Ada to prevent the protein from capturing meP3ester groups.

The reporter system was first tested with the alkylating agent methyl methane sulfonate (MMS) in E. coli and in Salmonella enterica biovar Typhimurium, which does not possess an Ada-based functional adaptive system³⁹. In both bacteria, the reporter system was activated only in the presence of MMS and only with the functional version of E. coli Ada (Supplementary Fig. 2). In B. abortus, the reporter was more active in exponential phase cultures compared to stationary phase cultures (Supplementary Fig. 3a). The emitted fluorescence was also dependent on the time of exposure and concentration of MMS (Supplementary Fig. 3b). To check that the reporter system was not affected by the endogenous Ada production in B. abortus, the mean fluorescence intensity of the system in a B. abortus ∆ada1 ∆ada2 background was compared with results in a wild-type (WT) background. No statistical difference could be observed between the two experiments, supporting the notion that endogenous Ada proteins do not affect the activation of the reporter system (Supplementary Fig. 3c).

The reporter system was then tested at the single-cell level during infection of RAW 264.7 macrophages. The first time point was 5 h PI, which corresponds to the end of the first phase of the infection, when B. abortus is not growing and blocked in a G1-like phase inside the eBCV²⁷. The second time point chosen was at 24 h PI, when the bacteria are actively growing inside the rBCV²⁷. The ratio between the mean fluorescence intensities of the functional reporter system and the non-functional version was calculated, and the level of fluorescence was significantly (p < 0.01) higher at 5 h PI than at 24 h PI (Fig. 2b, Supplementary Fig. 3d). This suggests that bacteria encounter alkylating stress inside host cells, but mainly during the initial phase of the infection.

To investigate the potential mutagenic properties of the intracellular eBCV environment compared to other conditions, we sequenced the genome of several individual clones of B. abortus before and after infection of RAW 264.7 macrophages at 6 and 48 h, and after early infection of mice (60 h), as well as after a similar number of generations in liquid culture (48 h). The genomes of five individual clones resulting from these different conditions were sequenced and subsequent analyses indicated that the number of mutations was not increased in the infection conditions compared to the culture (Supplementary Fig. 4).

N-nitrosation events occur inside the eBCV

One of the main sources of alkylating agents is the N-nitrosation of metabolites⁸. Since the content of the eBCV is unknown, we developed a new tool to investigate the presence of N-nitrosation in this compartment. Succinimidyl ester groups have been successfully employed to label the outer membrane of bacteria with fluorescent molecules^27,40. Besides, Miao et al.⁴¹ established a highly specific probe emitting fluorescence upon N-nitrosation. The two techniques were combined to create a N-nitrosation-sensitive probe that was covalently attached to the surface of B. abortus, allowing us to follow whether N-nitrosation occurs inside the eBCV, before the growth of bacteria²⁷ (Fig. 3a).

Labeled bacteria were first tested for their fluorescence in presence of KNO₂, which generates the NO donor N₂O₃ in aqueous solution (Supplementary Fig. 5a). Autofluorescence of non-labeled B. abortus was also compared to the fluorescence of labeled bacteria in the absence of KNO₂ (Supplementary Fig. 5a). Results demonstrate that the probe emits fluorescence when N-nitrosated, as expected. Next, RAW 264.7 macrophages were infected with labeled bacteria and mean fluorescence intensities were calculated at 5 h post infection at the single-cell level (Fig. 3b, Supplementary Fig. 5b, c). As a negative control, the same experiment was conducted in the presence of 163 µM of ascorbate in the cell culture medium, as this concentration of antioxidant is known to inhibit N-nitrosation reactions in RAW 264.7 macrophages⁴². We observed that about a quarter (23.3 %) of the bacterial population was subjected to N-nitrosation inside the eBCV. Importantly, the addition of ascorbate markedly decreased the proportion of positive labeled bacteria (Fig. 3b, Supplementary Fig. 5c), indicating that N-nitrosation reactions can effectively be prevented by antioxidants. LampI labeling of BCVs confirmed that most B. abortus were in the endosomal stage (eBCV) of the infection at that time point (Supplementary Fig. 5d).

We also investigated the presence of alkylating stress due to the endogenous production of N-nitroso compounds. To do so, the Ada_{E. coli}-based reporter system was tested in different genetic backgrounds (WT, ∆narG, and ∆moaA) at 5 h post infection. The deletion of narG alone was not sufficient to reduce alkylating stress, whereas it was the case with the ∆moaA strain (Fig. 3c). Notably, the addition of 163 µM of ascorbate to the cell culture medium did not significantly reduce the extent of alkylating stress, suggesting that external N-nitrosation is not responsible for alkylating stress in these conditions. Overall, this indicates that during RAW 264.7 infection, alkylating stress is mainly produced endogenously by B. abortus metabolism.

Key actors against alkylating stress in B. abortus

To evaluate which DNA repair genes are required by B. abortus to counteract alkylating stress, deletion strains were constructed and plated on rich medium supplemented with alkylating agents (Fig. 4). Mutants were constructed for genes predicted to encode proteins involved in direct repair, BER, HR, nucleotide excision repair (NER), and mismatch repair (MMR). Two strains were also included as negative controls: the triple mutant ∆mutM ∆mutY ∆mutT, required for DNA repair following oxidative stress, and the ∆virB strain, lacking the B. abortus type IV secretion system. These strains were tested for their survival against the S_N1 agent methylnitronitrosoguanidine (MNNG) that reacts with DNA in two main steps via a unimolecular nucleophilic substitution, and the S_N2 agent MMS, which reacts in one step with biomolecules^43,44 (Fig. 4).

Interestingly, some genes were required against MMS only, such as alkB and the BER genes. A B. abortus ∆xthA1 endonuclease mutant had been previously reported to be sensitive to MMS⁴⁵, although the function of XthA2 remains unclear. Here, we show that XthA1 is the major endonuclease, since its deletion was sufficient to confer sensitivity to MMS, whereas it was not the case for the deletion of xthA2. However, the double mutant was approximately a hundred-fold more sensitive than the single ∆xthA1 (Fig. 4), indicating that the genes have partially redundant functions. Similarly, the ∆alkA mutant was not affected by MMS, while the ∆tagA mutant displayed a 35-fold decrease in bacterial survival recovery. The ∆tagA ∆alkA mutant was more markedly sensitive, with a decrease in CFU of 5 orders of magnitude compared to the control condition (Fig. 4). This result indicates that AlkA and TagA share an overlapping function, which is crucial for survival in the presence of MMS in B. abortus, as it is the case in E. coli⁴⁶. The ∆recA mutant was also strongly affected by MMS (as described previously⁴⁷), but only slightly in the presence of MNNG (Fig. 4). The ∆ogt mutant is particularly noteworthy, as it appeared to be very sensitive to MMS exposure. The triple mutant ∆ada1 ∆ada2 ∆ogt was only marginally more attenuated than the single ∆ogt mutant against MMS, and slightly more against MNNG (Fig. 4). This indicates that the presence of Ogt is a key factor for B. abortus survival against alkylating agents in these conditions, unexpectedly more than the two Ada proteins compared to the E. coli model. At the protein level, Ogt_{B. abortus} is predicted to be 33% identical to Ogt_{E. coli} with the conservation of the methyl-acceptor C139 residue (Supplementary Fig. 6a). In B. abortus, the residue corresponding to Ogt_{E. coli} S134 is a proline (Supplementary Fig. 6a). Remarkably, in E. coli, the mutation of S134 into a proline confers broader substrate specificity to the protein by increasing the size of its active site⁴⁸, so this could explain why Ogt seems to be the major actor amongst the three methyltransferases of B. abortus.

The two genes predicted to code for Ada also possess conserved C38 and C321 residues (Supplementary Fig. 6b). Nevertheless, the deletion of the ada1 and ada2 genes did not change drastically the sensitivity of B. abortus to alkylating agents (Fig. 4), Questioning whether the Ada proteins are functioning as transcription factors. Quantitative reverse transcription polymerase chain reaction (RT-qPCR) experiments were performed on B. abortus in the presence or absence of MMS, and several DNA repair genes were compared for their mRNA levels in these conditions (Fig. 5). Interestingly, the mRNA levels of the two ada genes were not statistically increased after MMS exposure. In fact, the only overexpressed alkylation-specific genes were alkA and tagA, which are predicted to code for proteins of similar function. The absence of induction of the ada genes (Fig. 5) and their marginal role in coping with alkylating stress in vitro (Fig. 4) suggest that B. abortus does not rely on a classical Ada-dependent adaptive system to subvert alkylating stress. It has been proposed that the absence of an adaptive response in S. enterica serovar Typhimurium could be due to the lack of an acidic residue in position 106th³⁹. Similarly, in B. abortus, the corresponding position is occupied by either a N116 (Ada1) or a V105 (Ada2) (Supplementary Fig. 6b), so this could also explain why B. abortus does not possess a functional Ada-based adaptive system.

Another gene that was overexpressed upon MMS exposure was lexA (Fig. 5). Interestingly, the early accumulation of the SOS repressor LexA is a marker of the activation of the SOS response, because lexA itself is part of its early regulon to prevent an uncontrolled over-activation of the SOS response⁴⁹. To determine which genes are part of the SOS regulon in B. abortus under alkylating stress, the fold induction of several genes was compared after MMS exposure in the WT and lexA overexpression (pBBRMCS1-p_lac-lexA) strains. In the overexpression strain, lexA itself was not further induced by MMS exposure, as its level of transcription was probably already maximal (Supplementary Fig. 7). Importantly, neither tagA nor alkA were significantly differentialy induced by MMS in the two conditions (Supplementary Fig. 7), indicating that their induction is dependent of a yet unknown factor. Among the two error-prone DNA polymerases of B. abortus, the imuABC operon⁵⁰ had a drop of induction upon lexA overexpression, in contrast to dinB, which encodes DNA polymerase IV (Supplementary Fig. 7). Of note, in α-proteobacteria, a few genes are downregulated following the activation of the SOS response^51,52. In B. abortus, our results suggest that it could also be the case for the NER endonuclease uvrA, as it was further induced in the lexA-overexpressing strain (Supplementary Fig. 7). Importantly, RecA is suspected to constitutively trigger a basal SOS response in B. abortus, even under non-stress conditions⁵³, which could explain the relatively low induction of many potential target genes in our experiment.

Involvement of GcrA against alkylation stress in vitro

One striking characteristic of ogt is the presence, right after the start codon, of a GANTC motif. This sequence is known to be a site of epigenetic regulation in α-proteobacteria⁵⁴. Indeed, GANTC sites have been shown to be methylated by CcrM in a cell-cycle-dependent manner in the α-proteobacterium Caulobacter crescentus^55,56, and probably also in B. abortus^57,58. Interestingly, the gene coding for the alkylation-specific DNA repair AlkB protein is regulated throughout the cell cycle in C. crescentus⁵⁹ and in this bacterium, alkB mRNA levels drop by more than two-fold in a GcrA depleted strain⁶⁰. Since the cell cycle-dependent transcription factor GcrA is known to be modulated by methylated GANTC sites on C. crescentus DNA^60,61, we identified the ortholog of gcrA in B. abortus (BAB1_0329), a gene previously shown to be essential⁶². ChIP-seq analysis of GcrA was performed to identify its targets in B. abortus. As many as 232 hits were found for the first chromosome of B. abortus, and 110 for the second one (available at https://figshare.com/articles/Summary_of_whole_genome_sequencing_for_B_abortus/9747653). This high number of targets is consistent with GcrA being associated with the housekeeping sigma factor (σ⁷⁰), similarly to C. crescentus⁶⁰. Among GcrA targets, several genes are involved in DNA repair (ogt, lexA, uvrA, and mutL but also aidB) (Fig. 6a). Compared to the rest of the chromosomes, we found a significantly higher (4.4- and 4.7-fold in chromosomes I and II, respectively) frequency of GANTC sites in the peaks of this ChIP-seq (p < 0.001 according to a Poisson distribution), which is consistent with methylation-dependent binding of GcrA in B. abortus.

To test if genes involved in DNA repair are regulated by GcrA, we constructed a strain with IPTG-inducible GcrA factor (∆gcrA pBI-gcrA). In the absence of IPTG, the ∆gcrA pBI-gcrA strain mainly formed Y-shaped bacteria within the first 6 h of growth, indicating that their division was impaired. Texas Red succinimidyl ester (TRSE) labeling⁴⁰ also suggested that bacterial growth was also slower at later time points (Fig. 6b). After 3 h in the absence of IPTG, the bacteria were efficiently depleted from leftover GcrA, as demonstrated by western blot analysis (Supplementary Fig. 8a). RT-qPCR was performed on DNA repair genes after culturing B. abortus ∆gcrA pBI-gcrA in the presence or absence of IPTG. This confirmed that GcrA regulates the expression of ogt and mutL (Fig. 6c). Of note, the induction of lexA still occurred after MMS exposure in the absence of GcrA (Supplementary Fig. 8b), indicating that the activation of the SOS response under exogenous stress is regulated through a GcrA-independent mechanism.

The WT and GcrA-depleted strains were grown in liquid medium supplemented or not with IPTG and/or MMS to test whether the presence of GcrA is crucial for survival and growth in alkylating conditions. Aliquots were taken at different time points and plated on rich media supplemented with IPTG to assess bacterial survival. As seen with TRSE labeling (Fig. 6b), the ∆gcrA pBI-gcrA strain cultured without IPTG was almost not multiplying but did survive (at least up to 2 days) as bacteria were recovered following plating on media supplemented with IPTG (Fig. 6d). When bacteria were cultured in the presence of MMS, both strains first underwent a strong drop of CFU, independent of IPTG. Subsequently, both the WT and the ∆gcrA pBI-gcrA strains were able to overcome the stress and recovered with time if supplemented with IPTG (Fig. 6d). When B. abortus was depleted of GcrA, bacteria were unable to recover (Fig. 6d), indicating that the presence of GcrA is required for B. abortus to efficiently cope with high exogenous alkylating stress.

Individual repair pathways are required for long-term infections

Three strains (∆ada1 ∆ada2 ∆ogt, ∆alkA ∆tagA, and ∆alkB) which displayed increased sensitivity to alkylating stress in vitro (Fig. 4) were tested during the infection of RAW 264.7 macrophages but failed to show attenuation (Supplementary Fig. 9). Similarly, an aidB mutant had previously been shown to be unaffected in infection⁶³. Other DNA repair mutants were also tested but none of them was attenuated in this model of infection (Supplementary Fig. 9).

According to RT-qPCR data, GcrA is involved in the regulation of at least two DNA repair pathways: direct repair through ogt and MMR through mutL (Fig. 6c). In RAW 264.7 macrophages, it was observed that the GcrA depleted strain maintained a stable number of CFU until 24 h post infection, before dropping at 48 h post infection (Fig. 7). This indicates that GcrA is required for survival in this model of infection. However, this attenuation might not be only due to the disruption of DNA repair pathways, as GcrA also regulates many other functions, which will require further investigation.

The Ada-based reporter system allowed us to determine that the alkylating stress occurring on B. abortus inside RAW 264.7 macrophages is very low. In addition, our assays have all focuses on early infection times. One possibility is that alkylation stress could be more important at later time points inside host cells and/or in more physiological infection conditions. For example, a B. abortus ∆recA mutant was previously shown to be attenuated in a mouse infection model⁴⁷ even though we could not detect any attenuation in RAW 264.7 macrophages (Supplementary Fig. 9). Therefore, we tested several mutant strains in an intranasal mice infection model⁶⁴ at 12 and 28 days (Fig. 8). In the lungs, there was a striking attenuation at both time points for the alkylation-specific direct repair ∆ada1 ∆ada2 ∆ogt mutant and the glycosylase (∆alkA ∆tagA) deficient strain. The double endonuclease ∆xthA1 ∆xthA2 mutant and the MMR ∆mutL ∆mutS mutant were also strongly attenuated. In spleen, the defects were mild, with the BER mutants being the most affected (Fig. 8).

Discussion

Many environmental and pathogenic bacteria possess an adaptive system against alkylating stress¹, indicating that this stress is widespread in the environment. Nevertheless, before this study, it was not known whether intracellular bacteria also face alkylation during infection. Here, we show that the intracellular pathogen B. abortus faces alkylation stress during infection, with functional DNA repair pathways for alkylation damage required in a mice model of infection, and that the control of the genes involved in the repair of alkylated DNA is different from the one reported in E. coli.

To investigate the occurrence of alkylating stress at early time points of infection, we developed a fluorescence-based reporter system to follow the occurrence of the stress on bacterial DNA at the single-cell level. To understand the source of alkylating stress, a probe was designed to be covalently attached at the bacterial surface and report N-nitrosation events occurring inside the eBCV. The combination of those two approaches allowed us to determine that alkylating stress occurs inside the eBCV mainly via the bacterial metabolism. Indeed, the addition of ascorbate, which quenches N-nitrosation on the bacterial surface, was not able to decrease alkylation damage detected with the Ada-based reporter. Moreover, the deletion of moaA, which is involved in the biosynthesis pathway of the molybdenum cofactor, decreased the intensity of alkylating stress. B. abortus possesses a single nitrate reductase, so it is likely that the phenotype of the moaA mutant comes from the simultaneous deficiency of Nar and other enzymes dependent on the molybdenum cofactor⁷. It could be the case of MSF, a nitrate–nitrite antiporter, as well as FdnG, a formate dehydrogenase involved in nitrate respiratory chain⁶⁵. Importantly, the occurrence of external N-nitrosation events is relevant for alkylating stress only if metabolites are present in the environment. Indeed, alkylating stress is generated by those modified metabolites, and not by N-nitrosation per se⁶⁶. It has been proposed that the eBCV is deficient in metabolites^67,68, which could explain why host cells do not produce detectable exogenous alkylating agents. Another important point is that N-nitrosation reactions are dependent on RNS levels²⁵ and B. abortus is known to be very weakly immunogenic, as it prevents proinflammatory responses in macrophages and neutrophils⁶⁹. In addition, B. abortus possesses a gene coding for a nitric oxide reductase, involved in detoxification of NO and nitrate respiration, which is required and overexpressed by the bacterium early during infection^68,70,71. One interesting hypothesis is that by avoiding exogenous sources of alkylating stress, B. abortus generates weak endogenous stress. Indeed, as B. abortus activates its nitrate respiration/denitrification pathway inside host cells, it must probably also generate endogenous N-nitroso compounds and hence endogenous alkylating agents. Importantly, alkylating stress does not induce a linear dose response in organisms⁷², which could explain why DNA repair mutants were not attenuated during macrophage infection, but was attenuated in the more physiologicaly and relevant intranasal mouse infection model.

E. coli deals with alkylating stress through various and dynamically regulated DNA repair pathways. Indeed, in the model developed by Uphoff³⁴, E. coli mainly relies on damage tolerance (via the SOS response) and constitutive repair (via TagA and Ogt), before the adaptive response takes over after prolonged alkylating stress. Our data indicate that B. abortus has a different strategy to cope with this stress. First, we found that an Ada-based adaptive system is absent in B. abortus, similarly to some other α-proteobacteria^59,73. Nevertheless, B. abortus does possess the ability to induce the expression of both alkA and tagA upon alkylating agent exposure, via a yet unknown mechanism. Secondly, we showed that, through the SOS response, the genes coding for the error-prone DNA polymerase imuABC were also overexpressed in conditions of high alkylating stress. Finally, B. abortus was found to rely on the essential and well-conserved transcription factor GcrA to control the expression of a series of genes involved in DNA repair, including mutL and ogt. The GcrA depleted strain was impaired for division, growth, and virulence, suggesting that it plays a role in B. abortus cell cycle regulation. Of note, in B. abortus, the promoter of tagA is directly bound by CtrA, another conserved cell cycle regulator⁵⁷. In E. coli, ada is known to be overexpressed in the stationary phase, independent of the methylation of its C38 residue but through the activity of the alternative sigma factor RpoS⁷. Since there appears to be no rpoS homolog in B. abortus and other α-proteobacteria⁷⁴, it is tempting to speculate that these bacteria have selected systems in which cell cycle regulators control DNA repair. A previous report⁵³ also indicates that genes of the SOS regulon have a high basal expression in B. abortus, which could be an additional way to ensure sufficient protection against endogenously produced stress. Interestingly, the absence of genome replication inside the eBCV also constitutes a direct advantage against genotoxic stresses, as DNA adducts do not fix mutations so long as replication has not occurred. Another characteristic of GcrA in C. crescentus is its ability to sense CcrM-dependent methylation on DNA⁶¹. Knowing that this epigenetic mark is cell cycle regulated in C. crescentus⁷⁵ and probably also in B. abortus^57,58, there could be a functional link between both damage-induced and epigenetic methylation.

This study provides new insights into a stress that was, until now, only hypothetically associated with intracellular bacterial pathogens. The discovery of alkylating stress on intracellular B. abortus suggests that other bacteria transiting through similar compartments could also be exposed to such a stress. Future investigations along these lines could generate a better understanding of host–pathogen interactions at the molecular level.

Methods

Bacterial strains and media

E. coli strains DH10B (Thermo Fisher Scientific) and S17-1 (ref. ⁷⁶) were grown in Luria-Bertani (LB) medium at 37 °C. B. abortus 544 Nal^R strain (Obtained from J.-M. Verger and M. Grayon, Institut National de la Recherche Agronomique, Laboratoire de Pathologie Infectieuse et d’Immunologie, Nouzilly, France) and its derivatives were grown in either 2YT-rich medium (1% yeast extract, 1.6% peptone, 0.5% NaCl) or TSB-rich medium (3% Bacto tryptic soy broth) at 37 °C. Antibiotics were used at the following concentrations: ampicillin, 100 µg mL⁻¹; kanamycin, 10 µg mL⁻¹ with integrative plasmids or 50 µg mL⁻¹ with replicative plasmids; chloramphenicol, 20 µg mL⁻¹; nalidixic acid, 25 µg mL⁻¹; rifampicin, 20 µg mL⁻¹; gentamicin, 10 or 50 µg mL⁻¹ as indicated. When required, isopropyl β-d-1-thiogalactopyranoside (IPTG) was used at a concentration of 1 mM in bacterial culture and at 10 mM in the culture medium during cellular infections.

B. abortus deletion strains were constructed by allelic exchange, via pNPTS138 vectors (M. R. K. Alley, Imperial College of Science, London, UK) carrying a kanamycin resistance cassette and a sucrose sensitivity cassette²⁷. Briefly, we selected bacteria which had integrated the plasmid containing upstream and downstream sequences (about 750 bp each) of our target gene by plating them on kanamycin-containing agar plates, then performed a counterselection on 5% sucrose-containing agar plates without kanamycin to allow plasmid curing. To confirm gene deletion, PCR was performed on colonies that were kanamycin-negative and sucrose-resistant with primers targeting sequences upstream and downstream of the plasmid-containing sequences. Note that in the case of the ∆xthA2 strain, 204 nucleic acids were kept on each side of the gene, as there was a previous report that the full deletion of the gene was not feasible⁴⁵. Primers, plasmids and ORF of the studied genes are listed in Supplementary Data 1.

B. abortus GcrA-depleted strain was constructed similarly than the CtrA-depleted strain⁵⁷. Briefly, the p_lacI-lacI-p_lac sequence was amplified from the pSRK-Kan plasmid⁷⁷ using Phusion High-Fidelity DNA Polymerase (New England BioLabs). The PCR product was then cloned into a pBBRMCS1 plasmid using SacI and BamHI restriction enzymes. This modified pBBRMCS1 is referred to as pBI. The gcrA coding sequence was amplified form B. abortus 544 genome with Phusion High-Fidelity DNA Polymerase (New England BioLabs) and then cloned into pBI using BamHI and KpnI enzymes in order to orient the insert opposite to the p_lac promoter already present in pBBRMCS1. This final plasmid (pBI-gcrA) was transferred to B. abortus by mating, after inserting the deletion plasmid (pNPTS138-∆gcrA), then gcrA was removed from the chromosome of B. abortus as described above.

The lexA sequence was amplified from B. abortus 544 genome with primers listed in Supplementary Data 1 and the PCR product was ligated into a pBBRMCS1 plasmid after ApaI and BamHI restiction, generating the pBBRMCS1-p_lac-lexA final plasmid.

Cloning of the reporter system for alkylating stress

The p_ada-ada_{E. coli} sequence, including the start codon of alkB_{E. coli}, was amplified from E. coli DH10B with Phusion High-Fidelity DNA Polymerase (New England BioLabs) using primers listed in Supplementary Data 1. A superfolder gfp coding sequence (Supplementary Note 1), with a XhoI sequence after the start codon and a PstI sequence after the stop codon, was adapted to fit the codon usage of B. abortus 2308 (http://www.kazusa.or.jp/codon/) and ordered as gBlocks gene fragment (Integrated DNA Technologies). Both DNA products were cloned into a pBBRMCS1 plasmid using PstI, XhoI, and SpeI restriction enzymes in a triple ligation to orient the p_ada-ada_{E. coli}-gfp fusion opposite to the p_lac promoter of pBBRMCS1.

Synthesis of N-nitrosation-sensitive probe and binding conditions

The probe was designed based on Mio et al.⁴¹, with the addition of a succinimidyl ester group in order to allow the binding of the probe on amines at the bacterial surface. The characterization of the probe can be found in Supplementary Note 2.

One milliliter of bacteria (DO₆₀₀ 0.5) was centrifuged at 7000 r.p.m. for 2 min and washed twice in phosphate buffered saline (PBS). They were incubated for 1 h at 37 °C with the probe (10 µM) in 1 mL of PBS supplemented with 100 µL of NaHCO₃ 1 M (pH 8.4). Bacteria were then washed three times with PBS and used either for RAW 264.7 infection or for experiments in culture. For experiments in culture, labeled bacteria were left for 1 h on wheel in the dark with 1 M of KNO₂ (Thermo Fisher scientific) and 20 µL of HCl 3 M, before to be washed twice with PBS and fixed with paraformaldehyde (PFA) 2% for 20 min at 37 °C.

Texas Red succinimidyl ester labeling

One milliliter of bacteria (DO₆₀₀ 0.5) was centrifuged at 7000 r.p.m. for 2 min and washed twice in PBS. Bacteria were resuspended in 1 mL of PBS and incubated with Texas Red succinimidyl ester (TRSE) at a final concentration of 1 µg mL⁻¹ (Invitrogen) for 15 min at room temperature. Bacteria were then washed three times with PBS.

RAW 264.7 macrophage culture and infection

RAW 264.7 macrophages (ATCC) were cultured at 37 °C in the presence of 5% CO₂ in DMEM (Invitrogen) supplemented with 4.5 g L⁻¹ glucose, 1.5 g L⁻¹ NaHCO₃, 4 mM glutamine, and 10% fetal bovine serum (Gibco). RAW 264.7 macrophages were seeded in 24-well plates (with coverslips for immunolabeling) at a concentration of 10⁵ cells per mL and left in the incubator overnight. The next morning, late exponential phase cultures of Brucella (DO₆₀₀ 06–0.9) were washed twice in PBS in order to remove antibiotics and traces of growth medium, then they were prepared in DMEM at a multiplicity of infection of 50. During that step, IPTG or ascorbate was added to the culture medium if required. Bacteria and cells were centrifuged at 400 g for 10 min at 4 °C and then incubated for 1 h at 37 °C with 5% CO₂ atmosphere before to be washed twice with PBS and then incubated in medium supplemented with 50 µg mL⁻¹ of gentamicin to kill extracellular bacteria. One hour later, the medium was replaced by fresh medium supplemented with 10 µg mL⁻¹ of gentamicin.

Immunolabeling of infected RAW 264.7 macrophages

Cells were washed twice in PBS before to be fixed for 20 min in 2% PFA pH 7.4 at 37 °C. They were then left in PBS in the dark at 4 °C overnight before to be permeabilized in PBS with 0.1% Triton X-100 (Prolabo) for 10 min. Cells were incubated for 45 min with primary antibodies in PBS containing 0.1% Triton X-100 and 3% (w/v) bovine serum albumin (BSA, Sigma-Aldrich). Next, cells were washed three times in PBS before to be incubated with secondary antibodies in PBS containing 0.1% Triton X-100 and 3% BSA. For LampI labeling experiments, the primary antibodies consisted in homemade anti-Brucella rabbit polyclonal antibodies²⁷ and anti-LampI rat antibodies (1D4B; Developmental Studies Hybridoma Bank, University of Iowa), and the secondary antibodies consisted in goat anti-rabbit antibodies coupled to Pacific Blue (Invitrogen, cat. no. P10994) and goat anti-rat antibodies coupled to Alexa Fluor 647 (Invitrogen, cat. no. A21247). For all other experiments, we used anti-Brucella LPS primary antibodies (A76-12G12, undiluted hybridoma culture supernatant⁷⁸) and goat anti-mouse secondary antibodies coupled to Texas Red (1:500) (Invitrogen, cat. no. T862). Coverslips were washed three times in PBS, once in ddH₂O and then mounted on Mowiol (Sigma).

Microscopy and analyses of fluorescence

We used a Nikon Eclipse E1000 (objective ×100, plan Apo) microscope connected to a ORCA-ER camera (Hamamatsu). The Hg lamp was set with ND filter at 4. Bacteria in culture (2 μL) were observed with the phase contrast on PBS-agarose (1%) pads. Bacteria inside host cells were observed with the TxRed channel (or the CFP channel for LampI experiments) (100 ms). The FITC channel (1 s) was used to detect either the N-nitrosation-sensitive probe or the GFP signal of the reporter system. LampI proteins were detected with the APC channel (800 msec). Pictures were encoded with NIS-element software and analyzed with the plug-in MicrobeJ in ImageJ⁷⁹. For bacteria on pads, mean fluorescence intensities (MFI) were obtained as the “mean_c” values with MicrobeJ for individual bacteria. For intracellular bacteria, MFI were obtained by subtracting the background fluorescence (defined here as the average value of fluorescence given by the Pixel Inspection Tool on three points randomly chosen around a bacterium) to the “mean” value of fluorescence obtained for each bacterium with MicrobeJ (see Supplementary Note 3 for a detailed protocol). Note that MFI values were considered as positive for bacteria labeled with the N-nitrosation-sensitive probe when they reached an arbitrary threshold of MFI = 100, as this value is above 86% of the MFI values on labeled bacteria (in blue in Supplementary Fig. 5a), and below 96% of the MFI values for the positive control (labeled bacteria + KNO₂, in orange in Supplementary Fig. 5a).

Chromatin immunoprecipitation with anti-GcrA antibodies

Cultures of 80 mL of B. abortus (OD₆₀₀ 0.8) were harvested by centrifugation and proteins were cross-linked to DNA with 10 mM sodium phosphate buffer (pH 7.6) and 1% (v/v) formaldehyde for 10 min at RT and 30 min on ice. Bacteria were centrifuged and washed twice in cold PBS before to be resuspended in lysis buffer (10 mM Tris-HCl pH 7.5, 1 mM EDTA, 100 mM NaCl, 2.2 mg mL⁻¹ lysozyme, 20 mL protease inhibitor solution from Roche). Bacteria were lysed, after the addition of 0.1 and 0.5 mm diameter Zirconia/Silica beads (Biospec Products), in the cell Disruptor Genie from Scientific Industries at maximal amplitude (2800) for 25 min at 4 °C. Bacteria were then incubated for 10 min in the presence of ChIP buffer (1.1% Triton X-100, 1.2 mM EDTA, 16.7 mM Tris-HCl pH 8.0, 167 mM NaCl, protease inhibitors). DNA fragments of about 300 base pairs were obtained by sonicating the lysate on ice (Branson Sonifier Digital cell disruptor S-450D 400 W) by applying 15 bursts of 20 s (50% duty) at 30% amplitude. Debris were excluded in the pellet by centrifugation at 14,000 r.p.m. for 3 min. The supernatant was normalized by protein content by measuring the absorbance at 280 nm and 7.5 mg of protein was diluted in 1 mL of ChIP buffer supplemented with 0.01% SDS and pre-cleared in 80 mL of protein A-agarose beads (Roche) and 100 µg BSA. Homemade anti-rabbit polyclonal GcrA antibodies (290.S3) were added to the supernatant (1:1000) and incubated overnight at 4 °C. The mix was then incubated with 80 mL of protein A-agarose beads pre-saturated with BSA for 2 h at 4 °C. Beads were then washed in the following order: once with low salt buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl pH 8.1, 150 mM NaCl), once with high salt buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl pH 8.1, 500 mM NaCl), once with LiCl buffer (0.25 M LiCl, 1% NP-40, 1% sodium deoxycholate, 1 mM EDTA, 10 mM Tris-HCl pH 8.1), and twice with TE buffer (10 mM Tris-HCl pH 8.1 and 1 mM EDTA) before to be eluted with 500 µL of elution buffer (1% SDS and 0.1 M NaHCO₃). The reverse-crosslinking was performed with 500 mL of 300 mM of NaCl overnight at 65 °C. Samples were then treated with Proteinase K (in 40 mM EDTA and 40 mM Tris-HCl pH 6.5) for 2 h at 45 °C and DNA was finally extracted with QIAGEN MinElute kit to be resuspended in 30 µL of Elution buffer.

Illumina MiSeq was used to sequence immunoprecipitated DNA. Data consisted of a number of reads per nucleotide. A Z-score for each base pair (i.e. the number of standard deviations from the average) was calculated based on average and variance in a window of 1 million base pairs. A threshold of Z-score above 4 was set to consider genomic regions as bound by GcrA. These sequences were mapped to the genome of B. abortus 2308 (available at https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc = GSE136733). The GcrA-binding peaks (.txt files) can also be visualized on Artemis (freely available at http://www.sanger.ac.uk/science/tools/artemis) with the genomic sequences (.gb files) available at https://figshare.com/s/0e580305b65f67619d36. To calculate the number of GANTC sequences in ChIP-seq peaks, we extracted peak sequences online with Emboss-extractseq (http://emboss.bioinformatics.nl/cgi-bin/emboss/extractseq) and looked for the presence of GANTC sites with the “pattern matching, dna-pattern” tool on RSATools⁸⁰ (http://embnet.ccg.unam.mx/rsa-tools/) on both strands and with allowing overlapping matches. Results were normalized according to peak size to obtain the number of GANTC sites per kb (GANTC/kb). A similar analysis was performed for whole chromosomes with the “pattern matching, genome-scale dna-pattern” tool on RSATools. Ratios were calculated between data obtained (in GANTC/kb) for the peaks and for the whole chromosomes.

Mouse infection

Ethics statement: The Animal Welfare Committee of the Université de Namur (UNamur, Belgium) reviewed and approved the complete protocols for Brucella infections (Permit Numbers: 05-558 for intraperitonneal infections and 19-330 for intranasal infections).

Mice were acquired from Harlan (Bicester, UK) and bred in the animal facility of the Gosselies campus of the Université Libre de Bruxelles (ULB, Belgium). For intraperitoneal infections, mice (C57BL/6, 10–12 weeks old females) were injected with a dose of 10⁵ CFU per mL of B. abortus in 500 µL of PBS. Infectious doses were validated by plating serial dilutions of inoculums. At 60 h post inoculation, mice were euthanized by cervical dislocation. Immediately after being killed, spleens were recovered in PBS with 0.1% Triton X-100 (Sigma) and plated on 2YT agar medium.

For intranasal mouse infection, mice (C57BL/6, 10–12 weeks old, mix of males and females) were anesthetized with a mix of Xylasine (9 mg kg⁻¹) and Ketamine (36 mg kg⁻¹) in PBS before being inoculated by an intranasal injection of 2 × 10⁴ CFU mL⁻¹ of B. abortus. Confirmation of the infectious doses was done by plating serial dilutions of the inoculums. Mice were sacrificed by cervical dislocation at 12 or 28 days post infection. Immediately after sacrifice, spleen and lungs were crushed and resuspended in PBS 0.1% Triton X-100 (Sigma-Aldrich). The bacterial load was evaluated by serial dilutions in RPMI and plated on TSB agar medium. The CFU were counted after 5 days of incubation at 37 °C. All experiments were done in a Biosafety level 3 facility.

Whole-genome sequencing after infection and liquid cultures

Two liquid cultures of WT B. abortus were inoculated in 2YT-rich medium from the same plate. One of them was divided into five subcultures and diluted (1:10) in liquid cultures twice every 24 h, before to be plated. The remaining original liquid culture was used to infect five C57BL/6 mice and RAW 264.7 macrophages. Mice were injected intraperitoneally as described above and were euthanized at 60 h post inoculation to recover spleens. RAW 264.7 macrophages were infected as described above and bacteria were plated at 6 and 48 h post infection. Five streaks were made from five isolated colonies obtained after passage in either liquid cultures, mice, or RAW 264.7 macrophages from different wells. The five streaks served for inoculation of liquid cultures, from which genomic DNA was extracted (NucleoSpin Tissue extraction kit, Macherey-Nagel). Samples were sequenced with the Illumina sequencing technique using NextSeq500 run Mid PE150 after preparing a TruSeq DNA library (performed by Genomics Core Leuven, Belgium). Sequencing hits were mapped on the genome of B. abortus 544 and, for each colony, all the nucleotides that were different from the reference strain were compiled in an Excel table (performed by Genomics Core Leuven, Belgium; data available at https://figshare.com/articles/Summary_of_whole_genome_sequencing_for_B_abortus/9747653). Mutations were counted for each isolated colony by excluding regions corresponding to microsatellites and with less than 10 reads.

Reverse transcription followed by quantitative PCR

Bacterial cultures were grown in rich medium to exponential phase (OD₆₀₀ 0.3), washed twice in PBS, and allowed to grow in rich medium supplemented or not with IPTG and/or MMS 2.5 mM (Sigma) for 5 h. Bacteria were washed twice in PBS, then collected by centrifugation, and immediately frozen and stored at −80 °C until processing. RNA was extracted using TriPure isolation reagent (Roche) according to the manufacturer’s instructions. Samples were treated with DNase I (Fermentas), then RNA was reverse transcribed with specific primers (Supplementary Data 1), using the High capacity cDNA Reverse Transcription kit (Applied Biosystems). Specific cDNAs were amplified on a LightCycler 96 Instrument (Roche) using FastStart Universal SYBR Green Master (Roche). Results were normalized using 16S RNA as a reference with the E^−∆∆Ct calculation method⁸¹, where E corresponds to the efficiency of the primers (determined by serial dilution).

Colony-forming units counts

For CFU counts after infection, RAW 264.7 macrophages were washed twice in PBS, then lysed with 0.01% Triton X-100 in PBS for 10 min at 37 °C. Several dilutions were plated on TSB supplemented with IPTG when required. Plates were incubated for 3 days at 37 °C.

For CFU counts in culture, wild-type B. abortus and GcrA-depleted strain supplemented with IPTG were grown to the mid exponential phase (OD₆₀₀ 0.3–0.6) and normalized to OD₆₀₀ 0.15. Cultures were divided into different aliquots to be tested with or without MMS 5 mM and with or without IPTG. Samples were taken at different time points (3, 6, 20, 24, and 48 h) and plated with serial dilutions on 2YT plates, supplemented with 1 mM IPTG for the depleted strain.

For CFU counts on plates supplemented with MMS or MNNG, 100 µL of bacteria in the mid exponential phase (OD₆₀₀ 0.3–0.6) were plated after normalization to OD₆₀₀ 0.1 of all bacterial cultures. All plates were prepared fresh, 1–2 h before each experiment. MMS was added at a concentration of 2.5 mM in TSB-agar-rich medium during plate preparation. MNNG was prepared in an acetate buffer (pH 5), and then added at a final concentration of 35 µM in TSB-agar-rich medium during plate preparation.

Western blot

Cultures of B. abortus in the late exponential phase (OD₆₀₀ 0.7–1) were concentrated to an OD₆₀₀ of 10 in PBS, then inactivated for 1 h at 80 °C. Loading buffer (1:4 of final volume) was added before to heat the sample at 95 °C for 10 min. Ten microliters of sample were loaded on each well of a 12% acrylamide gels. After migration, proteins were transferred with the semi-dry method onto a nitrocellulose membrane (GE Healthcare) which was blocked in PBS supplemented with 0.05% Tween 20 (VWR) and 5% (w/v) milk (Nestlé, foam topping) overnight. The membrane was incubated for 1 h with polyclonal anti-GcrA (290.S3) or monoclonal anti-Omp10 (A68/4B10/F05) primary antibodies (1:1000), then with secondary antibodies coupled to HRP (Dako Denmark) (1:5000), both in PBS 0.05% Tween 1% milk. The membrane was washed three times in PBS before to be revealed with The Clarity Western ECL Substrate (Biorad) and Image Quant LAS 4000 (General Electric).

Statistical analysis

Statistical tests were performed in Excel or in GraphPad Prism version 8.0. All statistical tests were one-sided.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.

Data availability

ChIP-seq data are available at the GEO database under accession code GSM4056904. The list of mutations for whole genome sequencing experiments is available at figshare [https://figshare.com/articles/Summary_of_whole_genome_sequencing_for_B_abortus/9747653]. The source data underlying Figs. 2b, 3b-c, 4, 5, 6c-d, 7, and 8 and Supplementary Fig. 3b-d, 5a-b, d, 8a-b, 7, and 9 are provided as a Source Data file. All other relevant data supporting the key findings of this study are available within the article and its Supplementary Information file or from the corresponding author upon request.

References

Mielecki, D., Wrzesinski, M. & Grzesiuk, E. Inducible repair of alkylated DNA in microorganisms. Mutat. Res. Rev. Mutat. Res. 763, 294–305 (2015).
Article CAS PubMed Google Scholar
Fu, D., Calvo, J. A. & Samson, L. D. Balancing repair and tolerance of DNA damage caused by alkylating agents. Nat. Rev. Cancer 12, 104–120 (2012).
Article CAS PubMed PubMed Central Google Scholar
Catsburg, C. E. et al. Dietary sources of N-nitroso compounds and bladder cancer risk: findings from the Los Angeles bladder cancer study. Int. J. Cancer 134, 125–135 (2014).
Article PubMed CAS Google Scholar
Zhu, Y. et al. Dietary N-nitroso compounds and risk of colorectal cancer: a case-control study in Newfoundland and Labrador and Ontario, Canada. Br. J. Nutr. 111, 1109–1117 (2014).
Article CAS PubMed Google Scholar
Rydberg, B. & Lindahl, T. Nonenzymatic methylation of DNA by the intracellular methyl group donor S-adenosyl-L-methionine is a potentially mutagenic reaction. EMBO J. 1, 211–216 (1982).
Article CAS PubMed PubMed Central Google Scholar
Tudek, B. et al. Lipid peroxidation in face of DNA damage, DNA repair and other cellular processes. Free Radic. Biol. Med. 107, 77–89 (2017).
Article CAS PubMed Google Scholar
Taverna, P. & Sedgwick, B. Generation of an endogenous DNA-methylating agent by nitrosation in Escherichia coli. J. Bacteriol. 178, 5105–5111 (1996).
Article CAS PubMed PubMed Central Google Scholar
Sedgwick, B. Nitrosated peptides and polyamines as endogenous mutagens in O6-alkylguanine-DNA alkyltransferase deficient cells. Carcinogenesis 18, 1561–1567 (1997).
Article CAS PubMed Google Scholar
Guttenplan, J. B. N-nitrosamines: bacterial mutagenesis and in vitro metabolism. Mutat. Res. 186, 81–134 (1987).
Article CAS PubMed Google Scholar
Posnick, L. M. & Samson, L. D. Influence of S-adenosylmethionine pool size on spontaneous mutation, dam methylation, and cell growth of Escherichia coli. J. Bacteriol. 181, 6756–6762 (1999).
CAS PubMed PubMed Central Google Scholar
Bielski, B. H., Arudi, R. L. & Sutherland, M. W. A study of the reactivity of HO2/O2- with unsaturated fatty acids. J. Biol. Chem. 258, 4759–4761 (1983).
CAS PubMed Google Scholar
Nichols, D. S. & McMeekin, T. A. Biomarker techniques to screen for bacteria that produce polyunsaturated fatty acids. J. Microbiol. Methods 48, 161–170 (2002).
Article CAS PubMed Google Scholar
Winczura, A., Zdzalik, D. & Tudek, B. Damage of DNA and proteins by major lipid peroxidation products in genome stability. Free Radic. Res. 46, 442–459 (2012).
Article CAS PubMed Google Scholar
Mackay, W. J., Han, S. & Samson, L. D. DNA alkylation repair limits spontaneous base substitution mutations in Escherichia coli. J. Bacteriol. 176, 3224–3230 (1994).
Article CAS PubMed PubMed Central Google Scholar
Iyengar, R., Stuehr, D. J. & Marletta, M. A. Macrophage synthesis of nitrite, nitrate, and N-nitrosamines: precursors and role of the respiratory burst. Proc. Natl Acad. Sci. USA 84, 6369–6373 (1987).
Article ADS CAS PubMed PubMed Central Google Scholar
Vermeer, I. T. M. et al. Neutrophil-mediated formation of carcinogenic N-nitroso compounds in an in vitro model for intestinal inflammation. Toxicol. Lett. 154, 175–182 (2004).
Article CAS PubMed Google Scholar
Espey, M. G., Miranda, K. M., Thomas, D. D. & Wink, D. A. Distinction between nitrosating mechanisms within human cells and aqueous solution. J. Biol. Chem. 276, 30085–30091 (2001).
Article CAS PubMed Google Scholar
Kumar, Y. & Valdivia, R. H. Leading a sheltered life: intracellular pathogens and maintenance of vacuolar compartments. Cell Host Microbe 5, 593–601 (2009).
Article CAS PubMed PubMed Central Google Scholar
Pizarro-Cerda, J. et al. Brucella abortus transits through the autophagic pathway and replicates in the endoplasmic reticulum of nonprofessional phagocytes. Infect. Immun. 66, 5711–5724 (1998).
CAS PubMed PubMed Central Google Scholar
Moreno, E. & Moriyon, I. The genus Brucella. Prokaryotes 5, 315–456 (2006).
Article Google Scholar
Porte, F., Liautard, J. P. & Kohler, S. Early acidification of phagosomes containing Brucella suis is essential for intracellular survival in murine macrophages. Infect. Immun. 67, 4041–4047 (1999).
CAS PubMed PubMed Central Google Scholar
Starr, T., Ng, T. W., Wehrly, T. D., Knodler, L. A. & Celli, J. Brucella intracellular replication requires trafficking through the late endosomal/lysosomal compartment. Traffic 9, 678–694 (2008).
Article CAS PubMed Google Scholar
Celli, J. et al. Brucella evades macrophage killing via VirB-dependent sustained interactions with the endoplasmic reticulum. J. Exp. Med. 198, 545–556 (2003).
Article CAS PubMed PubMed Central Google Scholar
Sedzicki, J., et al. 3D correlative electron microscopy reveals continuity of Brucella-containing vacuoles with the endoplasmic reticulum. J. Cell Sci. 131, https://doi.org/10.1242/jcs.210799 (2018).
Article PubMed CAS Google Scholar
Ohshima, H. et al. L-arginine-dependent formation of N-nitrosamines by the cytosol of macrophages activated with lipopolysaccharide and interferon-gamma. Carcinogenesis 12, 1217–1220 (1991).
Article CAS PubMed Google Scholar
Kunisaki, N. & Hayashi, M. Formation of N-nitrosamines from seconday animes and nitrite by resting cells of Escherichia coli B. Appl. Environ. Microbiol. 37, 279–282 (1979).
CAS PubMed PubMed Central Google Scholar
Deghelt, M. et al. The newborn Brucella abortus blocked at the G1 stage of its cell cycle is the major infectious bacterial subpopulation. Nat. Commun. 5, 4366 (2014).
Article ADS CAS PubMed Google Scholar
Samson, L. & Cairns, J. A new pathway for DNA repair in Escherichia coli. Nature 267, 281–283 (1977).
Article ADS CAS PubMed Google Scholar
McCarthy, J. G., Edington, B. V. & Schendel, P. F. Inducible repair of phosphotriesters in Escherichia coli. Proc. Natl Acad. Sci. USA 80, 7380–7384 (1983).
Article ADS CAS PubMed PubMed Central Google Scholar
Uphoff, S. et al. Stochastic activation of a DNA damage response causes cell-to-cell mutation rate variation. Science 351, 1094–1097 (2016).
Article ADS CAS PubMed PubMed Central Google Scholar
Lindahl, T., Sedgwick, B., Sekiguchi, M. & Nakabeppu, Y. Regulation and expression of the adaptive response to alkylating agents. Annu. Rev. Biochem. 57, 133–157 (1988).
Article CAS PubMed Google Scholar
Bowles, T., Metz, A. H., O’Quin, J., Wawrzak, Z. & Eichman, B. F. Structure and DNA binding of alkylation response protein AidB. Proc. Natl Acad. Sci. USA 105, 15299–15304 (2008).
Article ADS CAS PubMed PubMed Central Google Scholar
Rippa, V. et al. Preferential DNA damage prevention by the E. coli AidB gene: a new mechanism for the protection of specific genes. DNA Repair (Amst.) 10, 934–941 (2011).
Article CAS Google Scholar
Uphoff, S. Real-time dynamics of mutagenesis reveal the chronology of DNA repair and damage tolerance responses in single cells. Proc. Natl Acad. Sci. USA 115, E6516–E6525 (2018).
Article CAS PubMed PubMed Central Google Scholar
Durbach, S. I. et al. DNA alkylation damage as a sensor of nitrosative stress in Mycobacterium tuberculosis. Infect. Immun. 71, 997–1000 (2003).
Article CAS PubMed PubMed Central Google Scholar
Alvarez, G. et al. Relevance of DNA alkylation damage repair systems in Salmonella enterica virulence. J. Bacteriol. 192, 2006–2008 (2010).
Article CAS PubMed PubMed Central Google Scholar
Sedgwick, B. Molecular signal for induction of the adaptive response to alkylation damage in Escherichia coli. J. Cell Sci. Suppl. 6, 215–223 (1987).
Article CAS PubMed Google Scholar
Kondo, H. et al. Structure and expression of the alkB gene of Escherichia coli related to the repair of alkylated DNA. J. Biol. Chem. 261, 15772–15777 (1986).
CAS PubMed Google Scholar
Hakura, A., Morimoto, K., Sofuni, T. & Nohmi, T. Cloning and characterization of the Salmonella typhimurium ada gene, which encodes O6-methylguanine-DNA methyltransferase. J. Bacteriol. 173, 3663–3672 (1991).
Article CAS PubMed PubMed Central Google Scholar
Brown, P. J. et al. Polar growth in the Alphaproteobacterial order Rhizobiales. Proc. Natl Acad. Sci. USA 109, 1697–1701 (2012).
Article ADS CAS PubMed PubMed Central Google Scholar
Miao, J. et al. Fast-response and highly selective fluorescent probes for biological signaling molecule NO based on N-nitrosation of electron-rich aromatic secondary amines. Biomaterials 78, 11–19 (2016).
Article CAS PubMed Google Scholar
Kosaka, H., Wishnok, J. S., Miwa, M., Leaf, C. D. & Tannenbaum, S. R. Nitrosation by stimulated macrophages. Inhibitors, enhancers and substrates. Carcinogenesis 10, 563–566 (1989).
Article CAS PubMed Google Scholar
Beranek, D. T. Distribution of methyl and ethyl adducts following alkylation with monofunctional alkylating agents. Mutat. Res. 231, 11–30 (1990).
Article CAS PubMed Google Scholar
Smith, M. B. & March, J. March’s advanced organic chemistry: reactions, mechanisms, and structure. 6th edition. J. Med. Chem. 50, 2279–2280 (2007).
Article CAS Google Scholar
Hornback, M. L. & Roop, R. M. II The Brucella abortus xthA-1 gene product participates in base excision repair and resistance to oxidative killing but is not required for wild-type virulence in the mouse model. J. Bacteriol. 188, 1295–1300 (2006).
Article CAS PubMed PubMed Central Google Scholar
Kaasen, I., Evensen, G. & Seeberg, E. Amplified expression of the tag+ and alkA+ genes in Escherichia coli: identification of gene products and effects on alkylation resistance. J. Bacteriol. 168, 642–647 (1986).
Article CAS PubMed PubMed Central Google Scholar
Tatum, F. M., Morfitt, D. C. & Halling, S. M. Construction of a Brucella abortus RecA mutant and its survival in mice. Microb. Pathog. 14, 177–185 (1993).
Article CAS PubMed Google Scholar
Schoonhoven, N. M., et al. Altering residue 134 confers an increased substrate range of alkylated nucleosides to the E. coli OGT protein. Molecules 22, https://doi.org/10.3390/molecules22111948 (2017).
Article PubMed Central CAS Google Scholar
Little, J. W. & Mount, D. W. The SOS regulatory system of Escherichia coli. Cell 29, 11–22 (1982).
Article CAS PubMed Google Scholar
Galhardo, R. S., Rocha, R. P., Marques, M. V. & Menck, C. F. An SOS-regulated operon involved in damage-inducible mutagenesis in Caulobacter crescentus. Nucleic Acids Res. 33, 2603–2614 (2005).
Article CAS PubMed PubMed Central Google Scholar
Tapias, A., Fernandez, S., Alonso, J. C. & Barbe, J. Rhodobacter sphaeroides LexA has dual activity: optimising and repressing recA gene transcription. Nucleic Acids Res. 30, 1539–1546 (2002).
Article CAS PubMed PubMed Central Google Scholar
da Rocha, R. P., Paquola, A. C., Marques Mdo, V., Menck, C. F. & Galhardo, R. S. Characterization of the SOS regulon of Caulobacter crescentus. J. Bacteriol. 190, 1209–1218 (2008).
Article PubMed CAS Google Scholar
Roux, C. M. et al. RecA and RadA proteins of Brucella abortus do not perform overlapping protective DNA repair functions following oxidative burst. J. Bacteriol. 188, 5187–5195 (2006).
Article CAS PubMed PubMed Central Google Scholar
Mohapatra, S. S., Fioravanti, A. & Biondi, E. G. DNA methylation in Caulobacter and other Alphaproteobacteria during cell cycle progression. Trends Microbiol. 22, 528–535 (2014).
Article CAS PubMed Google Scholar
Collier, J., McAdams, H. H. & Shapiro, L. A. DNA methylation ratchet governs progression through a bacterial cell cycle. Proc. Natl Acad. Sci. USA 104, 17111–17116 (2007).
Article ADS CAS PubMed PubMed Central Google Scholar
Kozdon, J. B. et al. Global methylation state at base-pair resolution of the Caulobacter genome throughout the cell cycle. Proc. Natl Acad. Sci. USA 110, E4658–E4667 (2013).
Article CAS PubMed PubMed Central Google Scholar
Francis, N. et al. CtrA controls cell division and outer membrane composition of the pathogen Brucella abortus. Mol. Microbiol. 103, 780–797 (2017).
Article CAS PubMed Google Scholar
Robertson, G. T. et al. The Brucella abortus CcrM DNA methyltransferase is essential for viability, and its overexpression attenuates intracellular replication in murine macrophages. J. Bacteriol. 182, 3482–3489 (2000).
Article CAS PubMed PubMed Central Google Scholar
Colombi, D. & Gomes, S. L. An alkB gene homolog is differentially transcribed during the Caulobacter crescentus cell cycle. J. Bacteriol. 179, 3139–3145 (1997).
Article CAS PubMed PubMed Central Google Scholar
Haakonsen, D. L., Yuan, A. H. & Laub, M. T. The bacterial cell cycle regulator GcrA is a sigma70 cofactor that drives gene expression from a subset of methylated promoters. Genes Dev. 29, 2272–2286 (2015).
Article CAS PubMed PubMed Central Google Scholar
Fioravanti, A. et al. DNA binding of the cell cycle transcriptional regulator GcrA depends on N6-adenosine methylation in Caulobacter crescentus and other Alphaproteobacteria. PLoS Genet. 9, e1003541 (2013).
Article CAS PubMed PubMed Central Google Scholar
Sternon J. F., et al. Transposon sequencing of Brucella abortus uncovers essential genes for growth in vitro and inside macrophages. Infect. Immun. 86, e00312 (2018).
Dotreppe, D., Mullier, C., Letesson, J. J. & De Bolle, X. The alkylation response protein AidB is localized at the new poles and constriction sites in Brucella abortus. BMC Microbiol. 11, 257 (2011).
Article CAS PubMed PubMed Central Google Scholar
Hanot Mambres, D. et al. Identification of immune effectors essential to the control of primary and secondary intranasal infection with Brucella melitensis in mice. J. Immunol. 196, 3780–3793 (2016).
Article CAS PubMed Google Scholar
Stewart, V. Nitrate respiration in relation to facultative metabolism in enterobacteria. Microbiol. Rev. 52, 190–232 (1988).
CAS PubMed PubMed Central Google Scholar
Archer, M. C. Mechanisms of action of N-nitroso compounds. Cancer Surv. 8, 241–250 (1989).
CAS PubMed Google Scholar
Lamontagne, J. et al. Intracellular adaptation of Brucella abortus. J. Proteome Res. 8, 1594–1609 (2009).
Article CAS PubMed PubMed Central Google Scholar
Lestrate, P. et al. Attenuated signature-tagged mutagenesis mutants of Brucella melitensis identified during the acute phase of infection in mice. Infect. Immun. 71, 7053–7060 (2003).
Article CAS PubMed PubMed Central Google Scholar
Barquero-Calvo, E. et al. Brucella abortus uses a stealthy strategy to avoid activation of the innate immune system during the onset of infection. PLoS ONE 2, e631 (2007).
Article ADS PubMed PubMed Central CAS Google Scholar
Haine, V., Dozot, M., Dornand, J., Letesson, J. J. & De Bolle, X. NnrA is required for full virulence and regulates several Brucella melitensis denitrification genes. J. Bacteriol. 188, 1615–1619 (2006).
Article CAS PubMed PubMed Central Google Scholar
Rossetti, C. A., Galindo, C. L., Garner, H. R. & Adams, L. G. Transcriptional profile of the intracellular pathogen Brucella melitensis following HeLa cells infection. Microb. Pathog. 51, 338–344 (2011).
Article CAS PubMed PubMed Central Google Scholar
Thomas, A. D. et al. Influence of DNA repair on nonlinear dose-responses for mutation. Toxicol. Sci. 132, 87–95 (2013).
Article CAS PubMed PubMed Central Google Scholar
Kaufman, A. & Walker, G. C. A constitutive O6-methylguanine-DNA methyltransferase of Rhizobium meliloti. Mutat. Res. 235, 165–169 (1990).
Article CAS PubMed Google Scholar
Staron, A. & Mascher, T. General stress response in alpha-proteobacteria: PhyR and beyond. Mol. Microbiol. 78, 271–277 (2010).
Article CAS PubMed Google Scholar
Stephens, C., Reisenauer, A., Wright, R. & Shapiro, L. A cell cycle-regulated bacterial DNA methyltransferase is essential for viability. Proc. Natl Acad. Sci. USA 93, 1210–1214 (1996).
Article ADS CAS PubMed PubMed Central Google Scholar
Simon, R., Priefer, U. & Pühler, A. A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in Gram negative bacteria. Bio/Technology 1, 784–791 (1983).
Article CAS Google Scholar
Khan, S. R., Gaines, J., Roop, R. M. II & Farrand, S. K. Broad-host-range expression vectors with tightly regulated promoters and their use to examine the influence of TraR and TraM expression on Ti plasmid quorum sensing. Appl. Environ. Microbiol. 74, 5053–5062 (2008).
Article CAS PubMed PubMed Central Google Scholar
Cloeckaert, A., Zygmunt, M. S., Dubray, G. & Limet, J. N. Characterization of O-polysaccharide specific monoclonal antibodies derived from mice infected with the rough Brucella melitensis strain B115. J. Gen. Microbiol. 139, 1551–1556 (1993).
Article CAS PubMed Google Scholar
Ducret, A., Quardokus, E. M. & Brun, Y. V. MicrobeJ, a tool for high throughput bacterial cell detection and quantitative analysis. Nat. Microbiol. 1, 16077 (2016).
Article CAS PubMed PubMed Central Google Scholar
van Helden, J. Regulatory Sequence Analysis Tools. Nucleic Acids Res. 31, 3593–3596 (2003).
Article PubMed PubMed Central Google Scholar
Livak, K. J. & Schmittgen, T. D. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25, 402–408 (2001).
Article CAS PubMed Google Scholar

Download references

Acknowledgements

We thank members of the URBM and I. Matic for stimulating discussions. We also thank Y. Ashhab for his help regarding ChIP-seq analysis. We thank H. Lavender and C. M. Tang for their careful reading of our manuscript. We thank UNamur (https://www.unamur.be/) for financial and logistic support. This work was funded by The Fédération Wallonie-Bruxelles (ARC 17/22-087) and by the FRS-FNRS “Brucell-cycle” (PDR T.0060.15). K.P., A.M., and G.P. were supported by a FRIA Ph.D. fellowship. A.R. and N.F. held an Aspirant fellowship from FRS-FNRS. This work was also supported by the French Agence Nationale de Recherche (ANR-JCJC-2011-Castacc) (http://www.agence-nationale-recherche.fr/) and the Region Pas-De-Calais (http://www.nordpasdecalais.fr) CPER to A.F. and E.G.B.

Author information

Authors and Affiliations

URBM, Narilis, University of Namur, Namur, Belgium
Katy Poncin, Agnès Roba, Georges Potemberg, Nayla Francis, Kévin Willemart, Nicolas Zeippen, Arnaud Machelart & Xavier De Bolle
Sir William Dunn School of Pathology, University of Oxford, South Parks Road, Oxford, OX1 3RE, UK
Katy Poncin & Stéphane P. Vincent
Unité de Chimie Organique, University of Namur, 61 rue de Bruxelles, 5000, Namur, Belgium
Ravikumar Jimmidi
Unité de Glycobiologie Structurale et Fonctionnelle, UMR 8576 CNRS, Université de Lille, 50 Avenue Halley, Villeneuve d’Ascq, France
Antonella Fioravanti & Emanuele G. Biondi
VIB,Vrije Universiteit Brussel, Pleinlaan 2, 1050, Brussels, Belgium
Antonella Fioravanti
Université de Lille, CNRS, INSERM, CHU Lille, Institut Pasteur de Lille, U1019, UMR 8204, Center for Infection and Immunity of Lille, Lille, France
Arnaud Machelart
IMM, 31 Chemin Joseph Aiguier, 13009 Marseille, Aix-Marseille Université, Marseille, France
Eric Muraille
Laboratoire de Parasitologie, Faculté de Médecine, Université Libre de Bruxelles, Brussels, Belgium
Eric Muraille

Authors

Katy Poncin
View author publications
You can also search for this author in PubMed Google Scholar
Agnès Roba
View author publications
You can also search for this author in PubMed Google Scholar
Ravikumar Jimmidi
View author publications
You can also search for this author in PubMed Google Scholar
Georges Potemberg
View author publications
You can also search for this author in PubMed Google Scholar
Antonella Fioravanti
View author publications
You can also search for this author in PubMed Google Scholar
Nayla Francis
View author publications
You can also search for this author in PubMed Google Scholar
Kévin Willemart
View author publications
You can also search for this author in PubMed Google Scholar
Nicolas Zeippen
View author publications
You can also search for this author in PubMed Google Scholar
Arnaud Machelart
View author publications
You can also search for this author in PubMed Google Scholar
Emanuele G. Biondi
View author publications
You can also search for this author in PubMed Google Scholar
Eric Muraille
View author publications
You can also search for this author in PubMed Google Scholar
Stéphane P. Vincent
View author publications
You can also search for this author in PubMed Google Scholar
Xavier De Bolle
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

K.P. and X.D.B. wrote the manuscript. R.J. and S.P.V. created and characterized the N-nitrosation sensitive probe. A.M., G.P. and E.M. designed and performed mice infection experiments. K.P. and X.D.B. designed all other experiments, and E.G.B. contributed to the design of the ChIP-seq experiment. K.P., A.R., G.P., A.F., N.F., K.W. and N.Z. performed experiments.

Corresponding author

Correspondence to Xavier De Bolle.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Peer review information Nature Communications thanks Elżbieta Grzesiuk, R. Martin Roop II, and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Peer Review File

Description of Additional Supplementary Files

Supplementary Data 1

Reporting Summary

Source Data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Poncin, K., Roba, A., Jimmidi, R. et al. Occurrence and repair of alkylating stress in the intracellular pathogen Brucella abortus. Nat Commun 10, 4847 (2019). https://doi.org/10.1038/s41467-019-12516-8

Download citation

Received: 22 October 2018
Accepted: 10 September 2019
Published: 24 October 2019
DOI: https://doi.org/10.1038/s41467-019-12516-8

This article is cited by

Distinctive roles of translesion polymerases DinB1 and DnaE2 in diversification of the mycobacterial genome through substitution and frameshift mutagenesis
- Pierre Dupuy
- Shreya Ghosh
- Michael S. Glickman
Nature Communications (2022)

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.