lable at ScienceDirect

Microbial Pathogenesis 49 (2010) 323e329

Contents lists avai

Microbial Pathogenesis

journal homepage: www.elsevier .com/locate/micpath

Screening of nitrosative stress resistance genes in Coxiella burnetii: Involvementof nucleotide excision repair

Sung-Hyun Park, Hyun-Wook Lee, Weiguo Cao*

Department of Genetics and Biochemistry, South Carolina Experiment Station, Clemson University, Room 219 Biosystems Research Complex, 51 New Cherry Street, Clemson,SC 29634, USA

a r t i c l e i n f o

Article history:Received 19 May 2010Received in revised form28 July 2010Accepted 2 August 2010Available online 10 August 2010

Keywords:Nitric oxideReactive nitrogen intermediatesNitrosative stressuvrBNucleotide excision repairDNA repair

* Corresponding author. Tel.: þ1 864 656 4176; faxE-mail address: wgc@clemson.edu (W. Cao).

0882-4010/$ e see front matter � 2010 Elsevier Ltd.doi:10.1016/j.micpath.2010.08.001

a b s t r a c t

Coxiella burnetii, an obligate intracellular Gram-negative bacterium, is the etiological agent of Q fever.This work takes advantage of a hypersensitive Escherichia coli genetic system to identify genes involvedin resistance to nitrosative stress imposed by reactive nitrogen intermediates. Among the ten candidategenes identified, the transposase, UvrB and DNA topoisomerase IV are involved in DNA transaction; thesigma-32 factor and the putative DNA-binding protein may be involved in transcriptional regulation; IF-2is involved in protein translation; malate dehydrogenase and carbamoyl-phosphate synthase aremetabolic enzymes; and the ABC transporter is a membrane-bound protein. In addition, a hypotheticalprotein was identified. The role of the DNA repair gene uvrB in resistance to RNI was further confirmed byinvestigating the sensitivity of uvrB deletion mutant and complementation by C. burnetii uvrB. Deletion oftwo other components of the UvrABC nuclease, uvrA and uvrC also renders the cell sensitive to RNI. Therelationship between UvrABC and nitrosative stress is discussed.

� 2010 Elsevier Ltd. All rights reserved.

1. Introduction

Coxiella burnetii, an obligate intracellular Gram-negative bacte-rium, is the etiological agent of Q fever [1e5]. The typical means oftransmission is through inhalation of contaminated aerosols,enabling its spread to large population. Once inhaled, C. burnetiitargets alveolar macrophages, is internalized by phagocytosis andresides in acidic phagolysosomes [6,7]. Upon infection, macro-phages arewell known to trigger “respiratory burst” to release largeamount of reactive radicals known as reactive oxygen intermediate(ROI) which imposes oxidative stress and reactive nitrogen inter-mediate (RNI) which imposes nitrosative stress to kill or containinvading pathogens [8,9]. Despite of such a powerful host defensesystem, some macrophage-residing pathogens such as the facul-tative intracellular Mycobacterium tuberculosis and the obligateintracellular C. burnetii remain viable after the assault by thesereactive radicals that inflict a variety of damage to essentialmacromolecules such as DNA and proteins. Evidently, theseorganisms have developed an unusual ability to resist the killingeffects caused by ROI and RNI. It is of great interest to understandthe underlying resistance mechanisms that confer pathogen’s

: þ1 864 656 0393.

All rights reserved.

ability to survive in the host, for study of pathogen biology,prevention, and therapeutic interaction.

How bacteria cope with oxidative stress imposed by ROI hasbeen a subject of intense investigation for many years [10,11].Resistance of nitrosative stress imposed by RNI was not recognizeduntil the role of nitric oxide synthase (NOS) in host defenseemerged (for review see Ref. [12]). Since then, studies in Myco-bacterium, Salmonella and other bacteria started to unravel bacte-rial defense mechanisms against RNI [13,14]. Yet, compared withROI, our understanding of resistance to RNI appears still in a veryearly stage. Use of M. tuberculosis as a model system to studybiochemical basis of resistance to RNI is facilitated by the ability tocultivate it outside the host cells and availability of tools for geneticmanipulation. Given its unusual ability to survive in host macro-phage cells, C. burnetii serves as a unique model system to under-stand its resistance mechanism to RNI. However, the obligate hostrequirement for growth and lack of a manipulatable genetic systempresent a challenge [15]. Because many genetic screening andselectionmethodologies are not applicable due to these limitations,alternative strategies have to be explored.

To investigate the resistance mechanisms to RNI adopted inC. burnetii, we developed an Escherichia coli-based genetic systemto allow for screening of genes involved in resistance to RNI. Thissystem takes advantage of the hypersensitivity of an E. coli triplemutant strain to nitrosative stress imposed by RNI. In this study, we

S.-H. Park et al. / Microbial Pathogenesis 49 (2010) 323e329324

screened a C. burnetii genomic library for candidate genes involvedin resistance to RNI. Among the genes positively identified, uvrBencodes a helicase as part of the UvrABC nuclease involved innucleotide excision repair. Based on the finding of uvrB gene, weexamined the sensitivity of uvrA and uvrC to RNI-mediated stress.Both uvrA and uvrC deletion mutants were sensitive to nitrosativestress. To our knowledge, this is the first report examining therelationship between all three components of UvrABC nuclease(uvrA, uvrB and uvrC) and nitrosative stress.

2. Results

2.1. E. coli-based genetic system

To investigate the resistance mechanisms to RNI adopted bybacteria pathogens, we sought to develop a genetic system to allowfor screening of genes involved in resistance to RNI. Sinceendonuclease V (nfi), alkyladenine DNA glycosylase (alkA) andendonuclease VIII (nei) are involved in repair of DNA damagecaused by RNI, the lack of these repair genes may render the cellssensitive to nitrosative stress. To test the sensitivity of this triplemutant strain to nitrosative stress, we treated it with differentconcentrations of acidified sodium nitrite. The triple mutant strainBW1739 that lacks nfi alkA and nei was significantly more sensitivethan the corresponding wild type (wt) strain BW1466 (Fig. 1). Aftertreating with more than 20 mM of acidified nitrite, BW1739completely lost viability, indicating that the triple mutant strainBW1739 becomes hypersensitive under nitrosative stress (Fig. 1).These results were consistent with previous observations ofreduced survival of Pseudomonas aeruginosa, and Mycobacteriumsmegmatis, and Salmonella typhimurium in glycosylase-deficientmutants under nitrosative stress conditions [16,17].

2.2. Screening of candidate genes in C. burnetii

We reasoned that genes involved in resistance to RNI maycomplement the hypersensitive phenotype demonstrated in thetriple mutant strain. Since C. burnetii is known for its ability tosurvive under nitrosative stress condition in the host, we thoughtthis genetic systemmay allow us to identify RNI resistance genes inC. burnetii. We thus constructed a genomic expression library using

Fig. 1. Sensitivity of E. coli strains BW1466 and BW1739 to acidified sodium nitritetreatment.Colonies from each strain was grown to saturation in 3.5 ml of LB medium.Cell density was approximated using a spectrophotometer and divided into six aliquotsof 0.5 ml (containing approximately 1 � 108 cell numbers). Each aliquot was treatedwith 0, 10, 20, 30, 40, and 50 mM acidified nitrite (pH 4.6) and then incubated at 37 �Cfor 10 min. Dilution and plating were performed as described in Materials andmethods. (C) wt E. coli strain (BW1466); (B) DNA repair deficient E. coli strain(BW1739).The data are the means � SD of at least three independent experiments.

a Lambda ZAP Express Vector (Stratagene). The quality of thelibrary was assessed by the lacZ-based blue/white color selectionassay on X-gal plates. Approximately 98% of clones containedinserts, indicating a high ratio between recombinant clones andself-ligated plasmid.

The genomic expression library was electroporated into thehypersensitive BW1739 cells. The screening was carried out bytreating the electroporated BW1739 cells with ascending concen-trations of acidified sodium nitrite at pH 4.6. As shown in Fig. 2, thecells with the empty plasmid were still hypersensitive to RNI asthey could not survive the treatment above 20 mM level. However,a large number of colonies were observed in the C. burnetii libraryat concentrations above 20 mM. Some cells were viable at 40 mMand even 50 mM concentration (Fig. 2B).

To verify these results, each individual surviving colonies from40 to 50mM concentrations were selected, grown, and treatedwith0e50mM acidified sodium nitrite at pH 4.6 again. Colonies that stillmaintained viability at 40 or 50 mM treatment were kept forfurther confirmation while those did not were discarded. Toconfirm that the survival is caused by the candidate genes insteadof a change in the host, plasmids were isolated from the clones andretransformed back to the hypersensitive BW1739 cells. The cellsthat again showed complementation to the triple mutant cells wereconsidered confirmed candidate genes. After these screening andconfirmation steps, ten genes from C. burnetii were identified(Table 1). Among the list, the transposase, UvrB and DNA top-oisomerase IV are involved in DNA transaction; the sigma-32 factorand the putative DNA-binding protein may involved in transcrip-tional regulation; IF-2 is involved in protein translation; malatedehydrogenase and carbamoyl-phosphate synthase are metabolicenzymes; and the ABC transporter is a membrane-bound protein.In addition, a hypothetical protein was identified.

2.3. UvrB and RNI resistance

The DNA repair gene uvrB gene caught our attention because ofits role in nucleotide excision repair. To assess the relationshipbetween uvrB gene and the RNI resistance, we measured thesensitivity of an E. coli uvrB deletion strain. While the isogenic wildtype BW25113 strain showed resistance to RNI, the survival ofΔuvrB strain was subdued at high acidified nitrite concentrations(Fig. 3). The cell counts were substantially reduced by 30 mMtreatment and reduced at zero by 40 and 50mM treatments. To testwhether C. burnetii uvrB can complement the phenotype, wecloned the gene into the pBK-CMV plasmid and transformed it intothe ΔuvrB strain. The cell counts of the survival colonies weresubstantially increased in all treatments (Figs. 3 and 4). A largenumber of uvrB-containing cells survived 30 mM treatment. Evenat 40 or 50 mM concentration, some of the uvrB-containing cellsstill survived (Fig. 3). These results suggest that the DNA repair geneuvrB is involved in resistance to RNI.

2.4. UvrA, UvrC and RNI resistance

Since UvrA and UvrC, along with UvrB, consist of the UvrABCnuclease, we were curious whether uvrA and uvrC are also involvedin resistance to RNI.We, therefore, tested the sensitivity of uvrA anduvrC deletion strains to acidified nitrite treatment. The uvrA dele-tion strain became very sensitive to RNI, as indicated by completeloss of survival at 40 or 50 mM and reduction of colony count toseveral orders of magnitude at 30 mM concentration (Fig. 5).Similarly, the survival of the uvrC deletion strain was also reduced(Figs. 5 and 6), as indicated by substantial decrease of colony countsfrom 20 to 50 mM. However, uvrA appeared far more sensitive than

Fig. 2. Screening of Coxiella burnetii genomic library treated with acidified sodium nitrite. See Materials and methods for details. A. E. coli strain (BW1739) containing empty pBK-CMV vector. B. E. coli strain (BW1739) containing C. burnetii genomic DNA library.

S.-H. Park et al. / Microbial Pathogenesis 49 (2010) 323e329 325

uvrC to the treatment. The possible reason accounting for thedifference is discussed later.

3. Discussion

3.1. Candidate genes and nitrosative stress

C. burnetii, as an obligate intracellular pathogen, possesses anunparalleled ability to replicate in phagolysosomes and to beresistant to various stresses. However, understanding of theunderlying molecular mechanisms of resistance is hindered by thelack of a manipulatable genetic system. This work takes advantageof a hypersensitive E. coli genetic system to identify genes involvedin resistance to nitrosative stress imposed by RNI. Among the tengenes identified, several of them are known for their involvementin stress resistance, while the mechanisms of other genes, such asputative DNA-binding protein and hypothetical protein, areunknown. Previously, transposases are found up-regulated inresponse to heat, oxidative stress although the exact mechanism isnot known [18,19]. Sigma-32 factor (rpoH) is a regulator of

Table 1Genes identified by screening C. burnetii genomic library.

C. burnetii # GenBank # Protein Name Gene Name

CBU_1640 AAO91136 Transposase (IS1111A)CBU_1770 AAO91264 ABC transporter, ATP-binding proteinCBU_0518 AAO90064 UvrB protein uvrBCBU_1241 AAO90750 Malate dehydrogenase mdhCBU_1281 AAO90788 Carbamoyl-phosphate synthase carBCBU_1432 AAO90929 Translation initiation factor IF-2 infBCBU_1686 AAO91182 Hypothetical protein, CBU1686CBU_1866 AAO91357 DNA topoisomerase IV, A subunit parCCBU_1249 AAO90758 DNA-binding protein, putativeCBU_1909 AAO91400 RNA polymerase sigma-32 factor rpoH

expression of heat shock proteins [20]. The deletion of the rpoHgene sensitize E. coli cells to oxidative stress. IF2 is up-regulated inresponse to heat in Porphyromonas gingivalis, an anaerobic oralbacterial pathogen [18]. Interestingly, it is proposed that IF2 acts asa chaperone to renature proteins under stress conditions [21].Increased expression of several ABC transporter genes is observedin Desulfovibrio vulgaris in response to nitrate stress [22]. Deletionof the malate dehydrogenase in S. typhimurium causes significantattenuation in BALB/c mice [23]. These analyses indicate thatC. burnetii shares some similar mechanisms with other microor-ganisms, but it also employs novel mechanisms to confer resistanceto RNI. A caveat of this study is that the screening was performed ina heterologous system, which may differ in the physiology of thenative C. burnetii system.

3.2. Role of UvrB, UvrA and UvrC

This work, for the first time, links nucleotide excision repair(NER) and RNI resistance in C. burnetii. UvrB, along with UvrA andUvrC, consists of a three-component protein complex calledUvrABC nuclease to initiate NER [24,25]. The UvrA dimer, incomplex with UvrB, is responsible for recognition of damaged DNA.UvrB is an ATP-dependent helicase that opens a short stretch of thelesion to facilitate recognition and incision by UvrC. UvrC containsthree domains: a central UVR domain to interact with UvrB and twoendonuclease domains located at the N-terminus and C-terminusto make 30 and 50 incisions, respectively [25]. Evidently, UvrB isessential for the function of UvrABC nuclease. Indeed, deletion ofuvrB gene alone renders E. coli hypersensitive to nitrosative stress(Figs. 3 and 4). Addition of an ectopic copy of C. burnetii uvrB partlyrescues the phenotype (Fig. 4). These results indicate that C. burnetiiuvrB is functional with E. coli UvrA and UvrC. It is interesting that

Fig. 3. Role of uvrB in resistance to nitrosative stress. See Materials and methods for details. A. Wild type E. coli strain (BW25113). Cells from 0 mM to 20 mM reaction were diluted5 � 106 times, from 30 mM reaction were diluted 5 � 105 times, from 40 mM reaction were diluted 5 � 104 times and from 50 mM reaction were diluted 5 � 103 times prior toplating. B. E. coli uvrB deletion strain (JW0762-2). C. E. coli uvrB deletion strain (JW0762-2) containing pBK-CMV-Cbu-uvrB vector.

Fig. 4. Plot of sensitivity of E. coli uvrB strains under nitrosative stress. (B) Wild type E.coli strain (BW25113); (6) E. coli uvrB deletion strain (JW0762-2); (7) E. coli uvrBdeletion strain (JW0762-2) containing pBK-CMV-Cbu-uvrB vector. The data are themeans � SD of at least three independent experiments.

S.-H. Park et al. / Microbial Pathogenesis 49 (2010) 323e329326

uvrB also confers sensitivity to RNI in M. tuberculosis. In a trans-poson insertion screening, uvrB is identified as one of the genes thatrenders M. tuberculosis sensitive to RNI [26]. A subsequent studyshows that a uvrB mutant is attenuated in wt mice but remainsvirulent in mice lacking iNOS [27]. These results suggest a commonDNA repair mechanism in conferring resistance to RNI in intracel-lular pathogens. Given the need for a new generation of antibiotics,this study, combined with previous studies on M. tuberculosis,indicates that NER may be a candidate to exploit non-traditionalpoints of vulnerability [28].

The finding on uvrB prompted us to investigatewhether uvrA anduvrC would exhibit a similar effect. As expected, uvrA deletion ishypersensitive to nitrosative stress (Figs. 5 and 6), indicating that likeuvrB, uvrA also confers resistance to RNI [29]. Deletion of uvrC alsorenders the strain more sensitive to nitrosative stress but to a lesserdegree than the uvrA strain (Fig. 6). The discovery of a second func-tional homolog of uvrC may provide an answer to the differentialeffect. In addition to uvrC, E. coli genome contains an uvrC homolo-gous gene Cho (UvrC homologue) [30]. Cho interacts with UvrB andcleaves at the 30 side of a damaged site [30]. Thus, in the absence ofUvrC, Cho may partially substitute for the function of uvrC in NER.

Fig. 5. Role of uvrA and uvrC in resistance to nitrosative stress. See Materials and methods for details. A. Wild type E. coli strain (BW25113). Cells from 0 mM to 20 mM reactionwerediluted 5 � 106 times, from 30 mM reaction were diluted 5 � 105 times, from 40 mM reaction were diluted 5 � 104 times and from 50 mM reaction were diluted 5 � 103 times priorto plating. B. E. coli uvrA deletion strain (JW4019-2). C. E. coli uvrC deletion strain (JW1898-1).

Fig. 6. Plot of sensitivity of E. coli uvrB strains under nitrosative stress. (B) Wild typeE. coli strain (BW25113); (6) E. coli uvrA deletion strain (JW4019-2); (7) E. coli uvrCdeletion strain (JW1898-1). The data are the means � SD of at least three independentexperiments.

S.-H. Park et al. / Microbial Pathogenesis 49 (2010) 323e329 327

DNA damage caused by RNI includes deamination of nucleotidebases and formation of cross-linked products [31e37]. While DNAbase deamination is usually repaired by DNA glycosylases, bulkycross-linked damage is removed by NER. Given that interstrandcross-links can block a variety of DNA transactions includingreplication, repair, recombination and transcription, their presencein the genomemay have a profound effect on the cell’s survivability.The sensitivity of uvrA, uvrB, and uvrC mutants to nitrosative stressindicates the importance of NER to maintain its genetic integrityand cellular survival. Consistent with this notion, uvr genes ofM. tuberculosis are up-regulated in response to phagocytosis inhuman macrophages [38]. This study, thus, underscores thepotential role of UvrABC nuclease for maintaining survivability ofbacterial pathogens under the attack of host defense.

4. Materials and methods

4.1. Reagents, media and strains

All routine chemical reagents were purchased from SigmaChemicals (St. Louis, MO), Fisher Scientific (Suwanee, GA), or VWR

S.-H. Park et al. / Microbial Pathogenesis 49 (2010) 323e329328

(Suwanee, GA). Restriction enzymes, Taq DNA polymerase and T4DNA ligase were purchased from New England Biolabs (Beverly,MA). BSA and dNTPs were purchased from Promega (Madison, WI).LB and SOC media were prepared according to standard recipes[39]. Sodium nitrite and sodium acetate trihydrate were purchasedfrom Fisher Scientific (Suwanee, GA). E. coli strains BW1466 (CC106)and BW1739 (BW1466 but Dnfi::(FRT-Spc-FRT) nei-1::cat alkA1) arekind gifts from Dr. Bernard Weiss at Emory University. The E. colistrains ΔuvrA (JW4019-2), E. coli ΔuvrB (JW0762-2), E. coli ΔuvrC(JW1898-1) and their isogenic wild type strain (BW25113) wereobtained from E. coli Genetic Stock Center at Yale University. C.burnetii prototype strain RSA 493 (phase I) genomic DNA (bothchromosomal DNA and plasmid) was a kind gift from Dr. JamesSamuel at Texas A&M University System Health Science Center.

4.2. Construction of C. burnetii genomic expression library

The C. burnetii RSA 493 genomic DNA (500 ng) was partiallydigested with 1 unit of Sau3AI at 37 �C for 20 min to generate DNAfragments predominantly ranging from 0.5 kb to 2 kb. Afterquenching the reaction with 2 ml of 0.5 M EDTA, the partiallydigested DNA was purified by phenol-chloroform-isoamylalcohol(25:24:1) extraction and ethanol precipitation. The DNA was driedand then dissolved in 1 mM TriseHCl (pH 8.0). The partiallydigested genomic DNA (150 ng) was inserted into a BamHI-CIAPpredigested Lambda ZAP Express vector (Stratagene) using 400units of T4 DNA ligase. Phage packaging and in vivo excision werecarried out according to the manufacturer’s instructional manual.Excised pBK-CMV plasmid was purified using Plasmid Mega Kit(Qiagen). BW1739 cells were made electroporation competent bywashing with sterile water once and with 10% glycerol three times[39]. The plasmid library was electroporated into BW1739 cellsusing MicroPulser (Bio-Rad). Briefly, aliquot of cells (1.6 � 109) wasmixed with 5 mg of the plasmid Library and incubated on ice for 3min. The mixture was then placed in a sterile 100 ml (1 mm gap)electroporation cuvette (USA Scientific) and electroporated withthree 8-ms pulses at 500 V. The cells were then immediately addedto 1 ml of sterile prewarmed SOC medium and incubated at 37 �Cfor 1 h with shaking at 250 rpm.

4.3. Screening of C. burnetii genomic expression library

The electroporated BW1739 cells, divided into approximately1�108/tube, were treated with 0, 20, 30, 40 and 50 mM of acidifiedsodium nitrate (pH 4.6) and then incubated at 37 �C for 10 min.Cells from 0 mM reaction were diluted 5 � 106 times prior toplating. Cells from 20 mM to 50 mM reactions were plated withoutdilution. Cells were plated onto LB plates with Kanamycin (50 mg/ml) and IPTG (0.5 mM). Plates were incubated for 48 h at 37 �C. BW1739 cells containing empty vctor pBK-CMV (without insert) wereused as a control. The sequences of the candidate genes wereidentified by dideoxy sequencing and BLAST search of GeneBankdatabases.

4.4. Cloning of Cbu uvrB

The C. burnetii uvrB genewas amplified by PCR using the forwardprimer Cbu.uvrB F (50-AAC TGC AGATGT CAA AAG CGT TTA AAT TAACA- 30; the PstI site is underlined) and the reverse primer Cbu.uvrB R(50-CGGAAT TCT CAT TCT TTAACT TCCAAAAGCCC-30; the EcoRI siteis underlined). The PCR reaction mixture (50 ml) consisted of 8 ng ofC. burnetii genomic DNA, 200 nM forward primer Cbu.uvrB F andreverse primer Cbu.uvrB R, 1 � Taq PCR buffer (New England Biol-abs), 200 mM each dNTP, and 5 units of Taq DNA polymerase (NewEngland Biolabs). The PCR procedure included a predenaturation

step at 94 �C for 5 min; 35 cycles of three-step amplification witheach cycle consisting of denaturation at 94 �C for 1min, annealing at45 �C for 1min andextension at 72 �C for 2min; andafinal extensionstep at 72 �C for 10 min. The PCR product was purified by phenolextraction and ethanol precipitation. The purified PCR product andplasmid pBK-CMV were digested with PstI and EcoRI, purified withphenol/chloroform extraction and ligated using T4 DNA ligase. Theligation mixture was transformed into E. coli strain JM109 compe-tent cells by electroporation. The sequence of the C. burnetii uvrBgene in the resulting plasmid (pBK-CMV-Cbu-uvrB) was confirmedby DNA sequencing. The plasmid pBK-CMV-Cbu-uvrB was thentransformed into the E. coli ΔuvrB strain (JW0762-2).

Acknowledgements

This project was supported in part by CSREES/USDA (SC-1700274,technical contribution No. 5797), and DOD-Army Research Office(W911NF-05-1-0335 and W911NF-07-1-0141). We thank Dr. JamesSamuel for providing C. burnetii genomic DNA and discussions. Wealso thank members of Cao laboratory for stimulating discussions.

References

[1] Baca OG, Paretsky D. Q fever and Coxiella burnetii: a model for host-parasiteinteractions. Microbiol Rev 1983;47:127e49.

[2] Cutler SJ, Bouzid M, Cutler RR. Q fever. J Infect 2007;54:313e8.[3] Waag DM. Coxiella burnetii: host and bacterial responses to infection. Vaccine

2007;25:7288e95.[4] Mege JL, Maurin M, Capo C, Raoult D. Coxiella burnetii: the ‘query’ fever

bacterium. A model of immune subversion by a strictly intracellular micro-organism. FEMS Microbiol Rev 1997;19:209e17.

[5] Woldehiwet ZQ. fever (coxiellosis): epidemiology and pathogenesis. Res VetSci 2004;77:93e100.

[6] Baca OG, Li YP, Kumar H. Survival of the Q fever agent Coxiella burnetii in thephagolysosome. Trends Microbiol 1994;2:476e80.

[7] Russell-Lodrigue KE, Zhang GQ, McMurray DN, Samuel JE. Clinical and path-ologic changes in a guinea pig aerosol challenge model of acute Q fever. InfectImmun 2006;74:6085e91.

[8] Babior BM. Phagocytes and oxidative stress. Am J Med 2000;109:33e44.[9] Chakravortty D, Hensel M. Inducible nitric oxide synthase and control of

intracellular bacterial pathogens. Microbes Infect 2003;5:621e7.[10] Hassett DJ, Cohen MS. Bacterial adaptation to oxidative stress: implications for

pathogenesis and interaction with phagocytic cells. FASEB J 1989;3:2574e82.[11] Zahrt TC, Deretic V. Reactive nitrogen and oxygen intermediates and bacterial

defenses: unusual adaptations in Mycobacterium tuberculosis. AntioxidRedox Signal 2002;4:141e59.

[12] Nathan C, Shiloh MU. Reactive oxygen and nitrogen intermediates in therelationship between mammalian hosts and microbial pathogens. Proc NatlAcad Sci U S A 2000;97:8841e8.

[13] Kaufmann SH, Cole ST, Mizrahi V, Rubin E, Nathan C. Mycobacterium tuber-culosis and the host response. J Exp Med 2005;201:1693e7.

[14] Fang FC. Perspectives series: host/pathogen interactions. Mechanisms of nitricoxide-related antimicrobial activity. J Clin Invest 1997;99:2818e25.

[15] Thompson HA, Suhan ML. Genetics of Coxiella burnetii. FEMS Microbiol Lett1996;145:139e46.

[16] Suvarnapunya AE, Lagasse HA, Stein MA. The role of DNA base excision repairin the pathogenesis of Salmonella enterica serovar Typhimurium. MolMicrobiol 2003;48:549e59.

[17] Venkatesh J, Kumar P, Krishna PS, Manjunath R, Varshney U. Importance ofuracil DNA glycosylase in Pseudomonas aeruginosa and Mycobacteriumsmegmatis, G þ C-rich bacteria, in mutation prevention, tolerance to acidifiednitrite, and endurance in mouse macrophages. J Biol Chem 2003;278:24350e8.

[18] Shelburne CE, Gleason RM, Coulter WA, Lantz MS, Lopatin DE. Differentialdisplay analysis of Porphyromonas gingivalis gene activation response to heatand oxidative stress. Oral Microbiol Immunol 2005;20:233e8.

[19] Zhang W, Culley DE, Nie L, Brockman FJ. DNA microarray analysis of anaerobicMethanosarcina barkeri reveals responses to heat shock and air exposure.J Ind Microbiol Biotechnol 2006;33:784e90.

[20] Kogoma T, Yura T. Sensitization of Escherichia coli cells to oxidative stress bydeletion of the rpoH gene, which encodes the heat shock sigma factor.J Bacteriol 1992;174:630e2.

[21] Caldas T, Laalami S, Richarme G. Chaperone properties of bacterial elongationfactor EF-G and initiation factor IF2. J Biol Chem 2000;275:855e60.

[22] Redding AM, Mukhopadhyay A, Joyner DC, Hazen TC, Keasling JD. Study ofnitrate stress in Desulfovibrio vulgaris Hildenborough using iTRAQ proteo-mics. Brief Funct Genomic Proteomic 2006;5:133e43.

S.-H. Park et al. / Microbial Pathogenesis 49 (2010) 323e329 329

[23] Tchawa Yimga M, Leatham MP, Allen JH, Laux DC, Conway T, Cohen PS. Role ofgluconeogenesis and the tricarboxylic acid cycle in the virulence of Salmo-nella enterica serovar Typhimurium in BALB/c mice. Infect Immun2006;74:1130e40.

[24] Sancar ADNA. excision repair. Annu Rev Biochem 1996;65:43e81.[25] Van Houten B, Croteau DL, DellaVecchia MJ, Wang H, Kisker C. ‘Close-fitting

sleeves’: DNA damage recognition by the UvrABC nuclease system. Mutat Res2005;577:92e117.

[26] Darwin KH, Ehrt S, Gutierrez-Ramos JC, Weich N, Nathan CF. The proteasomeof Mycobacterium tuberculosis is required for resistance to nitric oxide.Science 2003;302:1963e6.

[27] Darwin KH, Nathan CF. Role for nucleotide excision repair in virulence ofMycobacterium tuberculosis. Infect Immun 2005;73:4581e7.

[28] Nathan C. Antibiotics at the crossroads. Nature 2004;431:899e902.[29] Hartman Z, Henrikson EN, Hartman PE, Cebula TA. Molecular models that may

account for nitrous acid mutagenesis in organisms containing double-stranded DNA. Environ Mol Mutagen 1994;24:168e75.

[30] Moolenaar GF, van Rossum-Fikkert S, van Kesteren M, Goosen N. Cho,a second endonuclease involved in Escherichia coli nucleotide excision repair.Proc Natl Acad Sci U S A 2002;99:1467e72.

[31] Burney S, Caulfield JL, Niles JC, Wishnok JS, Tannenbaum SR. The chemistry ofDNA damage from nitric oxide and peroxynitrite. Mutat Res 1999;424:37e49.

[32] Dedon PC, Tannenbaum SR. Reactive nitrogen species in the chemical biologyof inflammation. Arch Biochem Biophys 2004;423:12e22.

[33] Spencer JP, Whiteman M, Jenner A, Halliwell B. Nitrite-induced deaminationand hypochlorite-induced oxidation of DNA in intact human respiratory tractepithelial cells. Free Radic Biol Med 2000;28:1039e50.

[34] Wink DA, Kasprzak KS, Maragos CM, Elespuru RK, Misra M, Dunams TM, et al.DNA deaminating ability and genotoxicity of nitric oxide and its progenitors.Science 1991;254:1001e3.

[35] Dong M, Wang C, Deen WM, Dedon PC. Absence of 2’-deoxyoxanosine andpresence of abasic sites in DNA exposed to nitric oxide at controlled physi-ological concentrations. Chem Res Toxicol 2003;16:1044e55.

[36] Caulfield JL, Wishnok JS, Tannenbaum SR. Nitric oxide-induced interstrandcross-links in DNA. Chem Res Toxicol 2003;16:571e4.

[37] Kirchner JJ, Sigurdsson ST, Hopkins PB. Interstrand cross-linking of duplexDNA by nitrous-Acid e covalent structure of the Dg-to-Dg cross-link at thesequence 5’-Cg. J Am Chem Soc 1992;114:4021e7.

[38] Graham JE, Clark-Curtiss JE. Identification of Mycobacterium tuberculosisRNAs synthesized in response to phagocytosis by human macrophages byselective capture of transcribed sequences (SCOTS). Proc Natl Acad Sci U S A1999;96:11554e9.

[39] Sambrook J, Russsell DW. Molecular cloning e a laboratory manual. ColdSpring Harbor, New York: Cold Spring Harbor Laboratory Press; 2001.