Identification of a Putative Chaperone Involved in Stress Resistance and Virulence in Francisella tularensis

Abstract

Francisella tularensis is a highly infectious bacterium causing the zoonotic disease tularemia. This facultative intracellular bacterium replicates in vivo mainly inside macrophages and therefore has developed strategies to resist this stressful environment. Here, we identified a novel genetic locus that is important for stress resistance and intracellular survival of F. tularensis. In silico and transcriptional analyses suggest that this locus (genes FTL_0200 to FTL_0209 in the live vaccine strain [LVS]) constitutes an operon controlled by the alternative sigma factor σ³². The first gene, FTL_0200, encodes a putative AAA+ ATPase of the MoxR subfamily. Insertion mutagenesis into genes FTL_0200, FTL_0205, and FTL_0206 revealed a role for the locus in both intracellular multiplication and in vivo survival of F. tularensis. Deletion of gene FTL_0200 led to a mutant bacterium with increased vulnerability to various stress conditions, including oxidative and pH stresses. Proteomic analyses revealed a pleiotropic impact of the ΔFTL_0200 deletion, supporting a role as a chaperone for FTL_0200. This is the first report of a role for a MoxR family member in bacterial pathogenesis. This class of proteins is remarkably conserved among pathogenic species and may thus constitute a novel player in bacterial virulence.

Francisella tularensis is a Gram-negative bacterium causing the disease tularemia in a large number of animal species. This highly infectious bacterial pathogen can be transmitted to humans in numerous ways (4), including direct contact with sick animals, inhalation, ingestion of contaminated water or food, or bites from ticks, mosquitoes, or flies. Four different subspecies of F. tularensis that differ in virulence and geographic distribution exist, designated F. tularensis subsp. tularensis (type A), F. tularensis subsp. holarctica (type B), F. tularensis subsp. novicida, and F. tularensis subsp. mediasiatica. F. tularensis subsp. tularensis is the most virulent subspecies, causing severe disease in humans, whereas F. tularensis subsp. holarctica causes a disease that is similar but less severe (30). Because of its high infectivity and lethality, F. tularensis is considered a potential bioterrorism agent (23).

F. tularensis is a facultative intracellular pathogen that is believed to replicate in vivo mainly inside infected macrophages (23, 30). In order to survive in this stressful environment, the bacterium has developed stress resistance mechanisms, including the production of chaperones to assist proper protein folding. In this respect, we have shown previously (21) that the folding chaperone ClpB, a member of the AAA+ (ATPases Associated with diverse cellular Activities) family of ATPases, plays a critical role in the pathogenesis of F. tularensis. Expression of clpB, as well as that of other genes encoding predicted heat shock proteins, is under the control of sigma-32 (σ³² or RpoH), the unique alternate sigma factor of F. tularensis (13). σ³² is the primary regulator that controls the transcription of genes during heat shock and some other general stress conditions. Our recent transcriptional analysis of the F. tularensis heat stress response (13) showed that an important proportion of heat stress-induced genes are also important for intracellular survival and/or virulence (including clpB itself). These data support an indirect relationship between heat stress and virulence, as infections in warm-blooded hosts occur at higher temperatures than F. tularensis encounters in the environment.

We report here the functional characterization of a novel putative chaperone involved in F. tularensis virulence. Identification of the gene FTL_0200, encoding a predicted AAA+ ATPase of the MoxR family, occurred serendipitously from the earlier analysis of the F. tularensis clpB mutant (21). Indeed, one of the proteins whose expression and/or stability depended on ClpB is encoded by gene FTL_0207. Examination of the genetic region containing FTL_0207 revealed the presence of a putative 10-gene cluster (FTL_0200 to FTL_0209). Furthermore, the first gene of the locus, FTL_0200, encoded a protein that possessed several predicted protein motifs in common with ClpB, i.e., the AAA_3, AAA_5, AAA, Mg_chelatase, and Sigma54_activation PFAM motifs (PFAM, database of protein families and domains [11]).

We report here that gene FTL_0200, as well as two other genes of the same locus, participates in stress resistance and virulence of F. tularensis. Genes encoding MoxR-like protein are widespread among the bacterial genomes, occurring alone or within gene clusters. These proteins might constitute a novel class of bacterial virulence attributes.

MATERIALS AND METHODS

Bacterial strains and plasmids.

The Francisella tularensis live vaccine strain (LVS) was grown on premade chocolate agar (bioMérieux SA, Marcy l'Etoile, France) plates prepared from GC medium base, IsoVitaleX vitamins and hemoglobin (BD Biosciences, San Jose, CA), in Schaedler-vitamin K3 broth (Schaedler-K3; bioMérieux) at 37°C. All bacterial strains, plasmids, and primers used in this study are listed in Table 1.

TABLE 1.

Strains, plasmids, and primers

Strain, plasmid, or primer	Primer code	Description or sequence (5′ → 3′)	Reference or source
E. coli strains
DH5α		F− φ80lacZΔM15 recA1 endA1 hsdR17 supE44 thi-1 gyrA96 relA1 (lacZYA-argF)U169	Laboratory strain collection
DH5α λpir		F− φ80lacZΔM15 recA1 end A1 hsdR17 supE44 thi-1 gyrA96 relA1 (lacZYA-argF)U169 λpir lysogen	Laboratory strain collection
S17-1		F− RP4-2-Tc::Mu aph::Tn7	Laboratory strain collection
S17-1 λpir		F− RP4-2-Tc::Mu aph::Tn7 λpir lysogen	Laboratory strain collection
DH5α(pPV)		DH5α containing plasmid pPV	This study
S17-1(pPV-AB)		S17-1 containing plasmid pPV-AB	This study
DH5α(pFNLTP6 pgro-moxR)		DH5α containing plasmid pFNLTP6pgro-moxR	This study
DH5α λpir(pSW29T-200ins)		DH5α λpir containing plasmid pSW29T-200ins	This study
DH5α λpir(pSW29T-205ins)		DH5α λpir containing plasmid pSW29T-205ins	This study
DH5α λpir(pSW29T-206ins)		DH5α λpir containing plasmid pSW29T-206ins	This study
F. tularensis strains
LVS		Live vaccine strain	A. Sjöstedt strain collection
LVS ΔFTL_0200		LVS with gene FTL_0200 deleted	This study
LVS ΔFTL_0200(pFNLTP6 pgro-moxR)		LVSΔFTL_0200 containing plasmid pFNLTP6pgro-moxR	This study
LVS FTL_0200-ins		LVS with insertion of pSW29T-205ins in gene FTL_0200	This study
LVS FTL_0205-ins		LVS with insertion of pSW29T-205ins in gene FTL_0205	This study
LVS FTL_0206-ins		LVS with insertion of pSW29T-206ins in gene FTL_0206	This study
Plasmids
pPV		E. coli-F. tularensis shuttle vector; Cmr AmprsacB; oriT RP4; ori pUC19	12
pPV-AB		pPV containing the upstream (A) and the downstream (B) portions of gene FTL_0200, both linked	This study
pFNLTP6gro		E. coli-F. tularensis shuttle vector with groE operon promoter; Kmr Ampr; constructed from pFNLTP6gro-gfp by deleting gfp gene	19, 20
pFNLTP6pgro-moxR		pFNLTP6gro containing LVS gene FTL_0200 and its promoter (p)	This study
pSW29T		Suicide vector, Kmr, oriT RP4, oriV R6Kg	9
pSW29T-200ins		pSW29T containing the middle of gene FTL_0200	This study
pSW29T-205ins		pSW29T containing the middle of gene FTL_0205	This study
pSW29T-206ins		pSW29T containing the middle of gene FTL_0206	This study
Primers
200end-FW	1	CGCCATAGGATACTATTAACTT
201start-RV	2	CTCATCAAAATCCATACCACG
201end-FW	3	CTAGCAAACAGCAAAGTGGTA
202start-RV	4	ACCAAACCAACGATAATAGCC
202end-FW	5	AAGCTAATATGCTTGGCAATAG
203start-RV	6	CAATTTGAGATTCTAGCTCTTC
203end-FW	7	GAGCCAATAGAGTCGGATAAA
204start-RV	8	TCACAATACTTAGCCCAACCA
204end-FW	9	CAGCAAGTAGGCTCGACAAA
205start-RV	10	ATTCATGCCCTGTTGGTCTTT
205end-FW	11	GTTGATCAGTGTCAAAAAGAAC
206start-RV	12	CAAGATGAGTTCTATCGACACT
206end-FW	13	GCTGCAATATATGCTGTAAAAG
207start-RV	14	GGGTTTATCTTCTCACCACC
207end-FW	15	TTTATACATGTCCCGTATGCG
208start-RV	16	GATAGCTAGATGTAGATAATCC
208end-FW	17	AATCAAGTTGGCCAGCTCAAT
209start-RV	18	GACAACTTCAGTAATCAGCAC
MoxR-Ft-AF	19	ACCGCCTAGGGACAGGGTCACAGCAATATAAAC
MoxR-Ft −2	20	AGATTTACTCCAAAATCTTAAAGTACC
MoxR-Ft −3	21	GGTACTTTAAGATTTTGGAGTAAATCTTAGAGATTAGCCTAGATGAAAAGTTAT
MoxR-Ft-BR	22	ACCGGTCGACTAGTTTGCATAGCTAGCTATCCCA
ΔmoxR-FW	23	GACTAGCATTATTTGAGCAGC
ΔmoxR-RV	24	GCATTAGACGGATATCATCAC
5′up-moxR	25	TCTTCATCATGCTTGTCACCA
3′down-moxR	26	AAATATCCTTTAATTGTTCTAAAAG
TUL4-435	27	GCTGTATCATCATTTAATAAACTGCTG
TUL4-863	28	TTGGGAAGCTTGTATCATGGCACT
MoxR-Ft_x	29	GAAATAATGGCCCAGACTGAA
MoxR-Ft _y	30	TCTCATCAGCAAGTAATAA
MoxR-Ft _z	31	TTCAAAGCTTTGTTGCTGTGG
Compl-moxR-FW	32	GGAATTCCATATGTAACTAGGGTAAGTGTAGCTAA
Compl-moxR-RV	33	ATAAGAATGCGGCCGCCTTTTCATCTAGGCTAATCT CT
5′EcoRI-200ins	34	CCGGAATTCGAATTTCAGTCTGGGCCAT
3′XbaI-200ins	35	GCTCTAGAGCTCTAATGCTTGAGAAACTT
5′EcoRI-205ins	36	CCGGAATTCATGGGATGATTTATGGTTGA
3′XbaI-205ins	37	GCTCTAGAGTCTTTGTTGTTATTTTGTG
5′EcoRI-206ins	38	CCGGAATTCGCAACAATTCCAAACACTTT
3′XbaI-206ins	39	GCTCTAGAAACTGGCATATAACCAATTT
200ins-FW	40	TATATACCACAGCAACAAAGC
200ins-RV	41	TCTAGTTGCTAATATTAGCTCA
205ins-FW	42	CAGATTCCTCACCATTGCCAC
205ins-RV	43	GTGTCCTATCATCCTCTGG
206ins-FW	44	GGCGAGACATTTGAGCTAGT
206ins-RV	45	CAATGACTTCTTTAGCGAC
pSW29T_T7	46	GTAATACGACTCACTATAGGGC
pSW29T_T3	47	AATTAACCCTCACTAAAGGG
5′0227qRT	48	GAACTGGTAGAGCACATCCT
3′0227qRT	49	CAAGCACAGTAATATTAGCTG C
5′1433qRT	50	CATGTTGACAAAGAAGCATGC
3′1433qRT	51	GCAATTGCCAAAGCGTCTCC
5′1947qRT	52	GAAGGTATCCCATTCAAAGGT
3′1947qRT	53	GCTTTTTGATGAGCAGTTTCAC

Construction of LVS-200-ins, LVS-205-ins, and LVS-206-ins insertion mutants.

Inactivation of genes FTL_0200, FTL_0205, and FTL_0206 was performed by insertion of a recombinant suicide plasmid as described previously (21). Briefly, an internal fragment of each gene (396 bp, 357 bp, and 359 bp, for FTL_0200, FTL_0205, and FTL_0206, respectively) was amplified by PCR using primers 34 to 39 (Table 1). The fragments were cloned into the XbaI and EcoRI sites of the suicide vector pSW29T (9), and the resulting plasmid was mobilized into F. tularensis LVS by conjugation from Escherichia coli strain S17-1 λpir. After 16 h of incubation at room temperature, the bacterial material was scraped off the plate and incubated in Schaedler-K3 broth containing 100 μg ml⁻¹ polymyxin B. After 3 h of incubation at 37°C with agitation, bacteria were spread onto chocolate agar plates containing 100 μg ml⁻¹ polymyxin B and 5 μg ml⁻¹ kanamycin to select for F. tularensis transconjugants. Single colonies appeared after 3 to 6 days and were repurified. Plasmid insertion in either FTL_0205 or FTL_0206 was verified by PCR using a combination of a plasmid primer (pSW29T_T3 or pSW29T_T7) and a gene-specific primer that anneal outside the region cloned on the plasmid. The presence of a single plasmid insertion was further checked by Southern blotting.

Construction of a chromosomal LVSΔFTL_0200 deletion mutant.

The counterselectable plasmid pPV-ΔFTL_0200, used to generate a nonpolar deletion mutation of FTL_0200 in strain LVS, was constructed by overlap PCR. Primers 19 and 20 amplify the 936-bp region upstream of position +1 of moxR, and primers 21 and 22 amplify the 925-bp region immediately downstream of the moxR stop codon. Primers 20 and 21 have an overlapping sequence of 23 nucleotides, resulting in complete deletion of the FTL_0200 coding sequence after crossover PCR. PCR with primers 19/20 and 21/22 was performed with Taq polymerase Dynazyme (Finnzymes), and the products were purified using the QIAquick PCR purification kit (Qiagen, CA). Two microliters of each purified product was included in a PCR mix without primers and treated with 20 cycles of PCR (94°C for 40 s, 48°C for 40 s, and 72°C for 90 s) to anneal and extend the crossover product. Next, 2 μl of the extended crossover product was used as a template for PCR with primers 19 and 22. The gel-purified 1.8-kb fragment was digested with AvrII and SalI (New England BioLabs) and cloned into XbaI-SalI-digested pPV (12). The plasmid was introduced into F. tularensis LVS by conjugation from E. coli strain S17-1. S17-1/pPV-MoxR and LVS were grown to exponential phase, mixed (at a ratio of approximately 1:20) in 0.9% NaCl, and spotted onto Schaedler medium agar plates (Bio-Rad) supplemented with MEM vitamins (Sigma-Aldrich). Plates were incubated at 25°C for 16 h, and bacterial material was recovered in Schaedler-K3 medium (bioMérieux) supplemented with 100 μg ml⁻¹ polymyxin B (Sigma). After 3 h of incubation at 37°C (to eliminate the majority of E. coli), bacteria were spread on chocolate agar plates containing 100 μg ml⁻¹ polymyxin B and 1.75 μg ml⁻¹ chloramphenicol. Colonies appeared after 5 to 6 days of incubation at 37°C and were subsequently passed once on plates with selection, followed by a passage in liquid medium without selection (to allow recombination to occur). Next, bacteria were passed once on agar plates containing 5% sucrose. Isolated colonies were checked for loss of the wild-type FTL_0200 gene by size analysis of the fragment obtained after PCR using primers 25 and 26 (which anneal to the regions outside the fragment cloned in pPV-ΔFTL_0200). Furthermore, colonies were verified to be F. tularensis using F. tularensis primers 27 and 28. One colony harboring an FTL_0200 deletion, as determined by PCR analysis, was used for further studies. Genomic DNA was isolated and used as the template in a PCR with the 5′up-moxR and 3′down-moxR primer pair. The PCR product was directly sequenced using primers ΔmoxR-FW and ΔmoxR-RV to verify the complete deletion of the FTL_0200 gene.

Functional complementation.

The plasmid used for complementation of the ΔFTL_0200 mutant, pFNLTP6pgro-moxR, was constructed by amplifying a 1,176-bp fragment (corresponding to the sequence 200 bp upstream of the FTL_0200 start codon and to 20 bp downstream of the stop codon) using primers Compl-moxR-FW and Compl-moxR-RV (Table 1), followed by digestion with NdeI and NotI and cloning into plasmid pFNLTP6pgro (19). The plasmids pFNLTP6pgro (empty plasmid) and pFNLTP6pgro-moxR (complementing plasmid) were introduced into LVS and the ΔFTL_0200 mutant by electroporation as described previously (21).

Growth kinetics in broth and stress survival assays.

Stationary-phase bacterial cultures of wild-type LVS(pFNLTP6pgro) and the LVS ΔFTL_0200(pFNLTP6pgro) and LVS ΔFTL_0200(pFNLTP6pgro-moxR) mutant strains were diluted at a final optical density at 600 nm (OD₆₀₀) of 0.1 in Schaedler-K3 broth. Every hour or 9 h, the OD₆₀₀ of the culture was measured. Exponential-phase bacterial cultures were diluted to a final concentration of 10⁸ bacteria ml⁻¹ and subjected to the following stress conditions: 10% ethanol, 0.03% H₂O₂, pH 4.0, 50°C, or 0.05% SDS. The number of viable bacteria was determined by plating appropriate dilutions of bacterial cultures on chocolate Polyvitex plates at the start of the experiment and after the indicated durations. For experiments at 50°C, aliquots of 0.25 ml in a 1.5-ml tube were placed at 50°C statically, and at the indicated times the tubes were placed on ice and the contents were serially diluted and plated. For experiments with ethanol, H₂O₂, pH 4.0, 50°C, or 0.05% SDS, cultures (5 ml) were incubated at 37°C with rotation (100 rpm) and aliquots were removed at indicated times, serially diluted, and plated immediately. Bacteria were enumerated after 72 h of incubation at 37°C. Experiments were repeated independently at least twice, and data represent the averages from all experiments.

Multiplication in macrophages.

J774 (ATCC TIB67) and THP1 (ATCC TIB-202) cells were propagated in RPMI or Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal calf serum. Cells were seeded at a concentration of ∼2 × 10⁵ cells per well in 12-well cell tissue plates, and monolayers were used at 24 h after seeding. THP1 cells were differentiated by treatment with 200 ng ml⁻¹ phorbol myristate acetate (PMA).

J774 and THP1 macrophage monolayers were incubated for 90 min at 37°C with the bacterial suspensions (approximate multiplicities of infection, 100) to allow the bacteria to enter. Where indicated, 10 U ml⁻¹ gamma interferon (IFN-γ) was added to cells 18 h before bacteria (18). After washing (time zero of the kinetic analysis), the cells were incubated in fresh culture medium containing gentamicin (10 μg ml⁻¹) to kill extracellular bacteria. At several time points, cells were washed three times in RPMI or phosphate-buffered saline (PBS), macrophages were lysed by addition of, water and the titer of viable bacteria released from the cells was determined by spreading preparations on chocolate agar plates. For each strain and time in an experiment, the assay was performed in triplicate. Each experiment was independently repeated at least three times, and the data presented originate from one typical experiment.

Electron microscopy.

To perform electron microscopy of infected cells, we infected J774 murine macrophages with LVS and LVS ΔFTL_0200 bacteria. Samples for electron microscopy were prepared using the thin-sectioning procedure as described previously (21).

Isolation of total RNA and reverse transcription.

Bacteria were centrifuged for 2 min in a microcentrifuge at room temperature, and the pellet was quickly resuspended in Trizol solution (Invitrogen, Carlsbad, CA). Samples were either processed immediately or frozen and stored at −80°C. Samples were treated with chloroform, and the aqueous phase was used in the RNeasy Cleanup protocol (Qiagen, Valencia, CA) with an on-column DNase digestion of 30 min.

RT-PCR.

Reverse transcription-PCR (RT-PCR) experiments were carried out with 1 μg of RNA and 2.5 pmol of specific reverse primers for each amplification. After denaturation at 65°C for 10 min, 12 μl of a mixture containing 2 μl of deoxynucleoside triphosphate (dNTP) (25 mM), 4 μl of 4× buffer, 2 μl of dithiothreitol (DTT), 1 μl of RNase OUT (Invitrogen), and 1 μl of Superscript II (Invitrogen) was added. Samples were incubated for 60 min at 42°C, heated at 75°C for 15 min, and chilled on ice. Samples were diluted with 30 μl of H₂O and stored at −20°C. PCR conditions were identical for all reactions. The 50-μl reaction mixtures consisted of 2 μl of template, 10 pmol of each primer, 1 μl of dNTP (10 mM), 5 μl of 10× buffer, and 0.5 μl of Taq polymerase.

The following pair of primers was used to amplify the mRNA corresponding to the transcript of both the FTL_0200 and FTL_0201 genes: primer 1 (in the distal part of FTL_0200, coding strand) and primer 2 (in the proximal part of FTL_0201, complementary strand). Similarly, to amplify the mRNA corresponding to the transcript of two (or more) consecutive genes, one primer was chosen in the distal part of the first gene and the other in the proximal part of the last gene (Table 1).

Quantitative real-time RT-PCR.

F. tularensis LVS and the LVS ΔFTL_0200 mutant strain were grown at 37°C to an OD₆₀₀ of ∼0.1 in Schaedler-K3 broth. After 4 h of incubation, samples were harvested and RNA was isolated. One microgram of RNA was reverse transcribed using random hexamers and Superscript II reverse transcriptase (Invitrogen) according to the protocol provided by the manufacturer. Real-time RT-PCR was performed with gene-specific primers using an ABI PRISM 7700 and SYBR green PCR master mix (Applied Biosystems, Foster City, CA). To calculate the amount of gene-specific transcript, a standard curve was plotted for each primer set using a series of diluted genomic DNA from LVS. The amounts of FTL_0227, FTL_1433, and FTL_1947 transcripts were normalized to the helicase gene (FTL_1656) transcript, whose expression is only slightly changed during growth (6).

Determination of transcription start site.

The 5′ end of the FTL_0200 mRNA was determined by employing the method of rapid amplification of cDNA ends (5′-RACE) using the 5′/3′-RACE kit from Invitrogen. One microgram of RNA isolated from LVS grown in Schaedler-K3 medium in exponential phase was used as the template for cDNA synthesis with the gene-specific primer 29 (Table 1). The cDNA was purified using the QIAquick PCR purification kit (Qiagen, Valencia, CA), poly(C) tailed at the 3′ end, and amplified by PCR using the gene-specific primer 29 and the poly(A) tail-specific primer oligo(dG) anchor. An aliquot of the PCR product was used as the template in a second PCR with primer 30 and PCR anchor primers. An aliquot of the second PCR product was then used as the template in a third PCR with primer 31. The PCR product obtained was directly sequenced and gave position −29 as the 5′ end relative to the translational start.

Isobaric tag for relative and absolute quantitation (iTRAQ) proteomic analyses.

Four individual replicates of the LVS and LVS ΔFTL_0200 strains were cultivated in chemically defined medium and grown overnight under constant agitation at 37°C. The overnight cultures were diluted with fresh medium to an OD₆₀₀ of 0.15 and cultivated for 4 h to an OD₆₀₀ of approximately 0.75. Under these conditions, the growth curves of the two strains were essentially undistinguishable (data not shown).

Bacteria were collected by centrifugation, washed twice with ice-cold phosphate-buffered saline, resuspended in 8 M urea, and vortexed thoroughly for 1 min. Bacteria were disrupted in a French pressure cell (Thermo Scientific, Milford, MA) by two passages at 16,000 lb/in². The lysates were then incubated on ice for 30 min, unbroken cells were removed by centrifugation, and supernatants were filtered through a 0.22-μm syringe filter. Aliquots of the lysates were diluted 10 times with 500 mM triethylammonium bicarbonate buffer (pH 8.5) and concentrated on an Amicon ultracentrifugal filter device (regenerated cellulose, 3,000-molecular-weight cutoff; Millipore) to decrease the urea concentration. RapiGest (Sigma) was added to a final concentration of 0.15% (wt/vol). Aliquots corresponding to 80 μg of proteins were reduced, alkylated, digested, and subsequently labeled with iTRAQ Reagents-8plex (Applied Biosystems) according to the manufacturer's protocol. Labeling of reaction mixtures was quenched by adding an excess of water, and the samples were mixed. RapiGest was removed according to the manufacturer's protocol. Prior to separation, the sample was desalted using an Oasis HLB 1cc (30 mg) extraction cartridge (Waters) and vacuum dried.

Peptides were separated by two-dimensional liquid chromatography (LC). In the first dimension, separation in basic pH was performed and fractions were collected (LC system, Alliance 2695 [Waters]; column, C₁₈ Gemini [3 μm, 100 Å, 2 by 150 mm] [Phenomenex]; trap column, C₁₈ Gemini [2 by 4 mm]; flow rate, 0.16 ml min⁻¹; UV detection at 215 nm; mobile phases, water [A], acetonitrile [B], and 200 mM ammonium formate, pH 10 [C]; gradient, linear from 5 to 55% of B in 62 min with constant 10% of C; sample load, 152 μg of peptides; fraction collection, 2-min fractions manually collected between min 10 and 54). The fractions were acidified with trifluoroacetic acid and vacuum dried. The second dimension standard nano-LC separation in acidic pH was carried out on an Ultimate 3000 system (Dionex, Sunnyvale, CA) (column, Atlantis dC₁₈ [3 μm, 0.1 by 150 mm] [Waters]; trap column PepMap 100 [0.3 by 5 mm] [Dionex]). The column effluent was continuously mixed with matrix solution (3 mg ml⁻¹ of α-cyano-4-hydroxycinnamic acid in 70% acetonitrile and 0.1% trifluoroacetic acid), and fractions were online spotted on a matrix-assisted laser desorption ionization (MALDI) plate by Probot (Dionex).

The mass spectrometry (MS) analysis was performed on a 4800 MALDI-time-of-flight (TOF)/TOF instrument (Applied Biosystems). Spectra were acquired from m/z 1,000 to 4,000 in the reflector positive-ion mode. Fragmentation of automatically selected precursor ions was performed with argon as a collision gas. One MS/MS spectrum was accumulated from 3,000 laser shots. Data acquisition and processing were done using the 4000 Series Explorer software, v. 3.5 (Applied Biosystems).

Protein identification and quantification were conducted using the ProteinPilot v. 3.0 software (Applied Biosystems) equipped with a Paragon searching algorithm (2, 29) and a Proteomics System Performance Evaluation Pipeline script that permits false-positive discovery rate (FDR) estimation (33). The data were searched against the F. tularensis subsp. holarctica LVS database (NCBInr derived, version 20081022). For protein identification and quantification and FDR determination, data were searched against a concatenated database that contained the forward and reversed protein sequences. Relative changes in protein expression were obtained from comparison between the LVS and LVS ΔFTL_0200 strains. Proteins were accepted to be significantly differently expressed when the P value was ≤0.05 and the relative change was ≥1.1 for upregulated genes or ≤0.909 for downregulated genes. Only proteins that met both criteria in at least three out of four replicates were considered.

RESULTS

Genetic organization of the F. tularensis moxR locus.

The LVS chromosomal region from bp 200644 to 208964 (GenBank accession no. AM233362) comprises 10 consecutive genes in the same orientation (FTL_0200 to FTL_0209), either overlapping or separated by very short intergenic regions (Fig. 1 A). The entire locus (flanked by intergenic regions of 145 bp and 131 bp, respectively) is highly conserved in all the genomes of the different subspecies of F. tularensis sequence thus far (Table 2). Interestingly, the proximal portion of the locus (Fig. 1A, genes FTL_0200 to FTL_0206) is homologous to the batL locus of Bacteroides fragilis (34). The first gene, FTL_0200, encodes a predicted P-loop nucleoside triphosphatase (NTPase) of 318 amino acids belonging to the subclass of AAA+ ATPases designated MoxR (according to the InterPro integrated database of predictive protein signature available at http://www.ebi.ac.uk/interpro/IEntry?ac=IPR016366). The MoxR AAA+ family consists of a large group of ATPases that have been poorly studied so far. The name MoxR (mox, for methanol-oxidizing system) was initially attributed to a protein from Paracoccus denitrificans, a Gram-negative bacterium capable of growing on methanol as a carbon source (1, 24, 36). For simplification, the locus (FTL_0200 to FTL_0209) will be designated the moxR locus in the rest of this paper.

FIG. 1.

The moxR locus of F. tularensis. (A) Schematic organization. Upper line, the moxR locus of F. tularensis LVS (genes FTL_0200 to FTL_0209) comprises 10 consecutive genes in the same orientation (black arrows), either overlapping or separated by very short intergenic regions (+12 bp between FTL_0200 and FTL_0201, −8 bp between FTL_0201 and FTL_0202, −4 bp between FTL_0202 and FTL_0203, +1 bp between FTL_0203 and FTL_0204, −10 bp between FTL_0204 and FTL_0205, −11 bp between FTL_0205 and FTL_0206, +6 bp, between FTL_0206 and FTL_0207, +7 bp between FTL_0207 and FTL_0208, and 33 bp between FTL_0208 and FTL_0209). Lower line, the region of the batL locus of B. fragilis (strain YCH46) is shown for comparison (white arrows). Gene BF2422, encoding the MoxR ortholog, is preceded by a 243-bp intergenic region. The sizes of the intergenic regions between the following genes are as follows: +56 bp between BF2422 and BF2421, + 8 bp between BF2421 and BF2420, +40 bp between BF2420 and BF2419 (or batA), +26 bp between batA and batB, +2 bp between batB and batC, +14 bp between batC and batD, and +18 bp between batD and batE. Finally, gene batE is followed by a 210-bp intergenic region. The percentages between the F. tularensis and B. fragilis loci indicate the amino acid identity between the corresponding protein orthologs. (B) Conserved general gene structure of the MoxR proper subfamily. Panel B was derived from the model of Snider and Houry (31) with permission of the publisher. Upper part, major conserved gene organization present around the moxR gene (red arrows) in the Francisella LVS and Schu S4 genomes. Lower part, gene organization of the MoxR proper subfamily, as defined in reference 31. MRP-AAA+, MoxR proper (red arrows). VWA, protein containing von Willenbrand factor type a domain (pink arrows). DUF58, protein containing the conserved domain 58 of unknown function (purple arrows). TPR, protein containing tetratricopeptide repeat domains (yellow arrows).

TABLE 2.

iTRAQ proteomic analysis

Protein name or predicted function	FTL locusa	Test 1		Test 2		Test 3		Test 4		Identified in screensd
		ΔMut/WTb	Pc	ΔMut/WT	P	ΔMut/WT	P	ΔMut/WT	P
Ribosome recycling factor	FTL_0227	0.9648	0.5666	0.8927	0.0039	0.9060	0.0031	0.9039	0.0001	No
Arabinose phosphate isomerase	FTL_1433	0.8073	0.0335	0.8875	0.0292	0.9087	0.0085	0.9575	0.3676	No
Putative ABC transporter ATP binding protein, YjjK	FTL_1947	0.9314	0.1473	0.8779	0.0329	0.8686	0.0000	0.8318	0.0019	Yes
IglBe	FTL_0112, FTL_1158	1.2786	0.0000	1.2397	0.0168	1.2209	0.0007	1.4439	0.0000	Yes
IglCe	FTL_0113, FTL_1159	1.2983	0.0001	1.1490	0.0887	1.2882	0.0000	1.4547	0.0000	Yes
IglDe	FTL_0114, FTL_1160	1.2338	0.0009	1.3384	0.3260	1.3971	0.0437	1.3949	0.0000	Yes
PdpCe	FTL_0116, FTL_1162	1.1219	0.0327	1.1593	0.1691	1.3173	0.0001	1.2661	0.0000	Yes
Hypothetical proteine	FTL_0118, FTL_1164	1.1316	0.0026	1.1435	0.0940	1.2584	0.0001	1.3209	0.0000	Yes
PdpBe	FTL_0125, FTL_1171	1.1192	0.0076	1.2416	0.0643	1.3574	0.0015	1.2790	0.0000	Yes
PdpAe	FTL_0126, FTL_1172	1.2117	0.0075	1.2571	0.0027	1.3848	0.0000	1.3559	0.0019	Yes
Glutamate dehydrogenase	FTL_0269	1.7717	0.0000	1.3190	0.0024	1.3156	0.0000	0.9979	0.9640	No
Hypothetical protein	FTL_0569	1.1568	0.0034	1.0532	0.4131	1.1346	0.0003	1.2374	0.0000	No
Fusion protein, FadB/AcbP	FTL_0584	1.1225	0.0443	1.0961	0.4571	1.1672	0.0393	1.1054	0.0010	Yes
Acyl-CoA dehydrogenase, FadE	FTL_0585	1.1486	0.0044	1.1329	0.0717	1.1550	0.0001	1.1256	0.0062	No
Peptidase, M24 family protein	FTL_0877	1.2459	0.0149	1.3361	0.0140	1.0778	0.1556	1.2346	0.0104	No
Glutathione reductase	FTL_1248	1.4928	0.0010	1.4575	0.0067	1.2160	0.0040	1.3749	0.0036	No

^aName of the gene locus assigned from F. tularensis subsp. holarctica (accession no. NC_007880).

^bAverage ratio for the protein (ΔmoxR mutant/LVS), corrected for experimental bias.

^cDetermined by Student's t test. Values of less than 0.05 are considered significant.

^dMutants with mutations in the corresponding genes reported in previous in vivo or in vitro genome-scale screens.

^eGene comprised within the FPI.

The proximal part of this locus (Fig. 1B) encodes proteins containing either von Willenbrand type A (VWA) motifs (3) or tetratricopeptide repetition (TPR) motifs (15). The gene FTL_201 determines a protein of 303 amino acid residues that shares 30% identity with BF2421 of B. fragilis. The proteins FTL_0201, FTL_0203, and FTL_0204 contain VWA motifs (Fig. 1B). FTL_0204 also contains TPR motifs, like FTL_0205 (Fig. 1B). These structural motifs, which are associated with various biological processes, are generally found in proteins involved in the formation of protein-protein interactions. The proteins FTL_0203, FTL_0204, and FTL_0205 share 32.3%, 28.1%, and 28.1% amino acid identity with BatA (BF2419), BatB (BF2418), and BatC (BF2417) of B. fragilis, respectively. Notably, the two genes FTL_0204 and FTL_0205 are fused into a single gene in F. tularensis type A strain Schu S4 (FTT0294). Finally, FTL_0206 determines a protein of 541 amino acid residues that shares 22% identity with BatD of B. fragilis (BF2416).

The Francisella MoxR protein (FTL_0200) shares 52.8% identity with the protein BF2422 of B. fragilis strain YCH46, a putative protein of 331 amino acid residues annotated as a putative magnesium chelatase. FTL_0200 also shares 46.2% identity with the protein CHU_153 of Cytophaga hutchinsonii (330 amino acids), whose crystallographic three-dimensional structure has recently been determined at 2-Å resolution (available at the internet address http://www.pdb.org/pdb/explore/explore.do?structureId=2R44). Multiple-protein alignments of the LVS MoxR and other selected MoxR orthologs reveal the presence of conserved canonical Walker A and Walker B ATP binding sites as well as of a conserved sensor 1 domain characteristic of MoxR-type AAA+ ATPases (Fig. 2).

FIG. 2.

The MoxR-like protein of LVS. Multiple alignments were performed using the ClustalW 1.81 program and CL sequence viewer version 5.1.1. software. Upper line, LVS FTL_0200; middle line, B. fragilis MoxR; lower line, Cytophaga hutchinsonii MoxR. The consensus sequence is shown below the alignment. The Walker A and Walker B ATP binding sites are boxed and shaded in gray. The sensor I domain is boxed.

Transcriptional analysis.

We first monitored transcription of the moxR locus by RT-PCR in the LVS wild-type strain grown under laboratory conditions (Schaedler-K3 broth in exponential growth phase). The 10 genes of the locus were transcribed (not shown), suggesting that they encoded functional proteins. RT-PCR analyses further confirmed that all the genes were also cotranscribed 2 by 2 and 3 by 3 (Fig. 3 A). Together, these results strongly suggest that this locus constitutes an operon.

FIG. 3.

Transcriptional analysis of the moxR locus. (A) Cotranscription of genes in the locus FTL_0200 to FTL_0209. RT-PCR was performed with several pairs of primers amplifying two or three genes each. Each PCR product is indicated by a letter (A to N), with its extent shown by a dotted arrow. Negative controls (without reverse transcriptase enzyme) are indicated on the gel image with a − and RT-PCRs with a +. (B) Promoter sequence upstream of FTL_0200. The transcription start is shown by a vertical arrow (+1). The predicted σ³²-dependent −10 and −35 sequences are boxed. The predicted translation start codon of FTL_0200 is underlined, and the preceding putative Shine-Dalgarno sequence (SD) is in italics and underlined.

We then performed rapid amplification of cDNA ends (5′-RACE) to determine the 5′ end of the FTL_0200 mRNA. This revealed that transcription is initiated 29 nucleotides upstream of the translational start (GTG) (Fig. 3B). Inspection of the sequence immediately upstream of the transcriptional start identified putative −10 and −35 promoter elements that share striking homology to the consensus site recognized by the alternative sigma factor σ³² in F. tularensis (13). The same elements are found in the promoter regions of clpB as well as many heat shock genes in F. tularensis LVS (such as the groES, dnaK, and grpE genes). Of interest, the genes FTL_0201, FTL_0202, and FTL_0203 are listed among the genes whose expression was significantly upregulated after temperature upshift (13), further supporting the notion of a σ³²-dependent expression of the moxR locus.

Construction and characterization of insertion mutants.

We first tested whether genes of the moxR locus could be involved in pathogenesis by individually inactivating several genes of the locus. We chose FTL_0200, the first gene of the locus; FTL_0205, in the middle of the locus (encoding one of the two proteins with TPR motifs); and FTL_0206, the BatD gene ortholog. For this, we used the simple plasmid insertion mutagenesis described previously (21) (see Materials and Methods for details). For each construct, we performed RT-PCR on the gene lying immediately downstream. In all three cases, a transcript of the downstream gene was detected, suggesting a lack of major polar effect of the insertion (data not shown). In broth, the growth of the three insertion mutants was undistinguishable from that of wild-type LVS, indicating that in these strains, the transposon insertion had no deleterious impact on bacterial multiplication and viability (data not shown). In vitro infection studies of LVS and the LVS FTL_0200-ins, LVS FTL_0205-ins, and LVS FTL_0206-ins mutant strains were performed, using THP1 and J774 cells (Fig. 4). In THP1 cells, intracellular multiplication of the three insertion mutants was impaired. An average 10-fold reduction in the number of intracellular bacteria was observed after 24 h and 48 h, compared to LVS (Fig. 4A, left). In the cell line J774, a 10-fold reduction of the number of bacteria was recorded at 24 h for two of the three mutant strains, with a plateau reached at this time point. However, at 48 h, the number of LVS bacteria decreased and was similar to that of the mutants (Fig. 4A, right). These data confirmed the importance of the moxR locus in intracellular survival and/or multiplication of the bacterium.

FIG. 4.

LVS and insertion mutants. (A) Intracellular replication. Intracellular bacterial replication of LVS, LVS FTL_0200-ins, LVS FTL_0205-ins, and LVS FTL_0206-ins bacteria was monitored over a 48-h period in THP1 human macrophages (left) and J774 macrophage-like cells (right). Results are shown as the averages of log₁₀ (CFU well⁻¹) ± standard deviation. *, P < 0.05; **, P < 0.01 (as determined by Student's t test). (B) Virulence in mice. Groups of five BALB/c mice were infected with LVS, LVS ΔFTL_0200, LVS FTL_0205-ins, or LVS FTL_0206-ins bacteria at different doses by the i.p. route. The log₁₀ of the numbers of bacteria used for infection is indicated for each survival curve. The survival of mice (percent) was followed for 10 days after infection.

To determine if the moxR locus plays a role in the ability of F. tularensis to cause disease, we infected 6- to 8-week old BALB/c mice with the LVS and LVS insertion mutant strains. Groups of five mice were inoculated by the intraperitoneal (i.p.) route with different numbers of bacteria, and survival of the mice was followed for 12 days (Fig. 4B). The 50% lethal doses (LD₅₀) of LVS, LVS FTL_0200-ins, LVS FTL_0205-ins, and LVS FTL_0206-ins were calculated (26) to be 10^0.85, 10², 10³, and 10^5.2 bacteria, respectively. These results indicated that the three genes carried in the moxR locus were required for full virulence of F. tularensis LVS in the mouse model.

Construction and characterization of a ΔFTL_0200 deletion mutant.

As mentioned above, FTL_0200 is the only protein encoded by the proximal portion of the locus that shares sequence similarity with other proteins with experimentally determined functions. We therefore decided to focus on the role of gene FTL_0200 and constructed a chromosomal deletion mutant of LVS lacking the entire FTL_0200 gene. The mutant strain was designated LVS ΔFTL_0200. We also constructed a complemented strain by introducing, in LVS ΔFTL_0200, a wild-type copy of the FTL_0200 gene under the control of the promoter of the groES gene (designated pgro). We examined the ability of LVS(pFNLTP6pgro), LVS ΔFTL_0200(pFNLTP6pgro), and the complemented strain LVS ΔFTL_0200(pFNLTP6pgro-moxR) to multiply in THP1 and J774 macrophage cells lines over a 48-h period. The ΔFTL_0200 mutant confirmed the data obtained with insertion mutant, showing a growth defect in both cell types whether or not the empty plasmid pFNLTP6pgro was included. In the human macrophage cell line THP1, the ΔFTL_0200 mutant showed a 10-fold reduction of intracellular bacteria after 24 h and 34-fold less after 48 h (Fig. 5 A1). In murine macrophage-like J774 cells, the mutant showed an approximately 5-fold reduction of intracellular bacteria after 24 h and a 21-fold reduction after 48 h (Fig. 5A, top). Introduction of the complementing plasmid (pFNLTP6pgro-moxR) restored bacterial viability to the same level as in the wild-type parent in both cell types, demonstrating the specific involvement of gene FTL_0200 in intracellular survival.

FIG. 5.

Intracellular replication of LVS and ΔFTL_0200 strains. (A) Intracellular replication of LVS, LVS ΔFTL_0200, and the complemented strain was monitored over a 48-h period in THP1 human macrophages (top) and J774 murine macrophages (bottom). Results are shown as the averages of log₁₀ (CFU/well) ± standard deviations. (B) Intracellular replication of LVS and LVS ΔFTL_0200 was monitored over a 48-h period in J774 murine macrophages with or without addition of IFN-γ. *, P < 0.05; **, P < 0.01 (as determined by Student's t test).

We further examined if there was any difference in the abilities of the LVS and LVS ΔFTL_0200 strains to survive in activated macrophages (Fig. 5B) by treating J774 cells with 100 U ml⁻¹ gamma interferon (IFN-γ) before infection. As observed before (18), LVS was sensitive to activation of macrophages by IFN-γ. Indeed, a strong decrease of multiplication was observed after 24 h in the presence of IFN-γ with both LVS and LVS ΔFTL_0200 (4 and 3 log units, respectively). After 48 h, LVS ΔFTL_0200 was no longer detectable, while LVS bacteria could still be observed in an amount similar to that at 24 h.

Infection of J774 cells was further followed by thin-section electron microscopy, as described previously (21). As expected, bacterial replication was observed in the cytosol of infected cells at 24 h postinfection with wild-type LVS. Of 443 cells counted, 46 (10.4%) were infected by LVS. Most of the infected cells (93%) contained more than 5 bacteria (including 56% with 5 to 10 bacteria and 37% with more than 10 bacteria). With the LVS ΔFTL_0200 mutant strain, the percentage of infected cells was comparable to that with LVS (65/438 cells, 14.8%), suggesting a normal entry process. Bacterial multiplication had also occurred at 24 h in these cells. However, only 40% of the infected cells contained more than five bacteria. Hence, the fraction of cells infected with fewer than five bacteria was much higher with LVS ΔFTL_0200 than with LVS (60% versus 7%), suggesting impaired intracytosolic growth of LVS ΔFTL_0200.

Role of MoxR in stress resistance.

The growth kinetics of LVS and LVS ΔFTL_0200 demonstrated that deletion of FTL_0200 had no impact on bacterial growth in broth (data not shown). To determine if the F. tularensis MoxR protein is involved in stress tolerance, we monitored the survival of the LVS(pFNLTP6pgro), LVS ΔFTL_0200 (pFNLTP6pgro), and LVS ΔFTL_0200(pFNLTP6pgro-moxR) strains under various conditions.

Reactive oxygen species have been reported to play a role in control of F. tularensis infections (10). We therefore first examined the tolerance of the two strains to oxidative stress by determining the number of viable bacteria after exposure to 0.03% H₂O₂ (approximately 10 mM H₂O₂) (Fig. 6 A). The LVS ΔFTL_0200(pFNLTP6pgro) strain was much more sensitive to oxidative stress than the wild-type strain. It showed a 2-fold decrease in viable numbers already after 10 min, and a 1,330-fold decrease was observed after 1 h of exposure to 10 mM H₂O₂. The LVS ΔFTL_0200 mutant appears to be more sensitive to oxidative stress than the previously characterized clpB mutant of LVS (21). Indeed, no significant survival defect was detected in that mutant at up to 2 h after incubation with the same concentration of H₂O₂. To ensure that the observed survival phenotype resulted from inactivation of FTL_0200 and not from polar effects on downstream genes, we complemented the mutant strain with a plasmid-carried copy of FTL_0200. Introduction of the plasmid (pFNLTP6pgro-moxR) restored bacterial viability to the same level as in the wild-type parent (Fig. 6A), confirming the specific involvement of FTL_0200.

FIG. 6.

Stress resistance of the LVS and LVS ΔFTL_0200 strains. Exponential-phase bacteria were diluted to a final concentration of 10⁸ bacteria ml⁻¹ in fresh Schaedler-K3 broth and subjected to oxidative stress (10 mM H₂O₂) (A), acidic stress (pH 4) (B), heat stress (50°C) (C), alcohol stress (10% ethyl alcohol) (D), and 0.05% SDS (E). The bacteria were plated on chocolate agar plates at different times, and viable bacteria were monitored 3 days after. Data are the average CFU ml⁻¹ for two independent experiments for each condition.

F. tularensis may also encounter a low-pH environment when it transiently resides inside a phagosome after entry into macrophages, although the importance of this process in bacterial phagosomal escape is debated (7, 8, 27). We determined whether the LVS ΔFTL_0200 mutant was impaired in its ability to endure acid stress by incubating the LVS and LVS ΔFTL_0200 strains in normal growth medium adjusted to pH 4.0 (Fig. 6B). Both LVS(pFNLTP6pgro) and LVS ΔFTL_0200(pFNLTP6pgro) were sensitive to low pH, but the ΔFTL_0200 mutant strain showed a 33-fold increase of mortality after 4 h. The complemented strain exhibited viability at the same level as the wild-type parent. Sensitivity of the ΔFTL_0200 mutant was similar to that recorded previously for the LVS clpB mutant (21). We also evaluated the impact of the ΔFTL_0200 deletion on bacteria subjected to heat, ethanol, and SDS stresses. When subjected to high temperature (50°C), the number of viable bacteria of the LVSΔFTL_0200 mutant was 62-fold lower than that of the wild-type strain after 1 h (Fig. 6C). In medium containing a high concentration of ethanol (10%), the number of mutant bacteria was 40-fold lower than that of LVS bacteria after 2 h (Fig. 6D), and in medium containing 0.05% SDS (Fig. 6E), the mutant resisted 16-fold less than LVS after 4 h.

Together, these assays demonstrate that the MoxR-like protein of F. tularensis contributes to the adaptation to a variety of stressful conditions.

Proteomic analysis.

To evaluate whether the MoxR-like protein of LVS might be involved in protein folding and/or stability, we monitored the impact of moxR inactivation on the proteome of LVS. For this, we used the isobaric tag for relative and absolute quantitation (iTRAQ) technique. Four independent biological samples of LVS and ΔFTL_0200 mutant strains were prepared essentially as described previously (17) and compared (see Materials and Methods). In all, 994 proteins were identified and quantified using four iTRAQ data points and were submitted for exploratory statistical analysis (see Table S1 in the supplemental material). Only 16 proteins presented a ΔFTL_0200/LVS ratio different from 1, and this was statistically significant in at least three out of four comparative analyses (Table 2). As observed previously, differences in protein abundance were rather small (17), confirming that the values recorded by the iTRAQ technique can only be used for a relative quantitative evaluation of protein abundance.

As expected, the lowest values were recorded for FTL_0200 (mutant/wild-type ratios of 0.0994, 0.2311, 0.2610, and 0.2631, respectively [see Table S1 in the supplemental material]). These values were not null, probably due to background detection in the assay. Since these values were not statistically interpretable, they are not listed in Table 2. The amounts of three proteins (FTL_0227, FTL_1433, and FTL_1947) were significantly reduced in the ΔFTL_0200 mutant strain compared to LVS, suggesting that they required MoxR for proper expression and/or stability. We performed real time RT-PCR on these three genes in the LVS and LVS ΔFTL_0200 strains to evaluate a possible impact of MoxR at the transcriptional level. The data revealed that the expression of the three genes was almost identical in LVS and in the deletion strain (data not shown), further supporting a role of MoxR in protein stability.

FTL_0227 is a predicted ribosome-recycling protein. FTL_1433 is a predicted arabinose 5-phosphate isomerase, possibly involved in an early step of lipopolysaccharide (LPS) biogenesis. FTL_1947 is a putative ATP binding protein of an ABC transporter. Notably, two in vivo negative screens (32, 38) have previously identified the corresponding gene FTL_1947 (also designated yjjK), suggesting that this protein is involved in F. tularensis virulence. Unexpectedly, 13 proteins were also found at higher levels in the ΔFTL_0200 mutant, including 7 proteins encoded by the Francisella pathogenicity island (FPI). Among the other proteins, FTL_0584 (a predicted fusion product of 3-hydroxacyl coenzyme A [CoA] dehydrogenase and acyl-CoA binding protein, FadB/AcbP) has been identified in previous screens for attenuated mutants (32, 38).

DISCUSSION

We report here the identification of a novel virulence locus of F. tularensis that is likely controlled by the alternative sigma factor σ³². We show that FTL_0200, the first gene of this locus, encodes a MoxR-like AAA+ family member that contributes to stress resistance, intracellular multiplication, and virulence of F. tularensis. Inactivation of gene FTL_0200 has a pleiotropic impact on protein expression and/or stability. F. tularensis MoxR is a member of a highly conserved class of chaperones that may contribute to pathogenesis in other bacterial species.

MoxR, a putative chaperone involved in stress response.

We have shown recently that the chaperone ClpB was involved in resistance to multiple stresses as well as in virulence in F. tularensis (21). We found here that the F. tularensis genomes encode a MoxR ortholog. MoxR AAA+ ATPase proteins are predicted to participate in the proper maturation of proteins or protein complexes. AAA+-type ATPases possess conserved Walker A and Walker B motifs mediating ATP binding and hydrolysis (22). They generally function as oligomers, often forming hexameric rings (14, 16). However, their exact functions remain poorly understood. The stress sensitivity of the F. tularensis ΔFTL_0200 mutant prompted us to use a proteomic approach to evaluate a possible role for MoxR as a protein chaperone. Proteomic techniques are now currently used to gain information on the molecular bases of bacterial pathogenesis. Gel-based approaches have thus far been the most widely used approaches in quantitative proteomic studies. However, several other quantitative methods have recently been developed, including iTRAQ, a powerful non-gel-based technology that allows simultaneous quantification of up to four (or even eight) different samples in one analysis (25). Very recently, iTRAQ has been successfully applied to study the heat response of F. tularensis strains LVS and Schu S4 (17). iTRAQ analysis of the LVS ΔFTL_0200 mutant indicated that only three proteins were found at lower levels in the LVS ΔFTL_0200 mutant strain than in wild-type LVS, suggesting that they required MoxR for expression and/or stability. Real-time RT-PCR assays confirmed that MoxR had no significant impact on the expression of these genes at the transcriptional level (data not shown).

Interestingly, the gene encoding the third protein, FTL_1947 (also designated yjjK), has been identified in two different in vivo screens for virulence genes (32, 38). The precise role of FTL_1947, encoding a putative ATP binding protein of an ABC transporter, is currently unknown. Unexpectedly, the majority of proteins (13 proteins) were found at higher levels in the moxR mutant than in the wild-type strain, including 7 proteins encoded by the FPI. The impact on proteins of the FPI is presumably very indirect and might be a consequence of the complex regulation that governs its expression (20). Several hypotheses can account for the absence, in the proteomic analysis, of proteins that could be directly associated with the oxidative stress response (such as superoxide dismutases, catalases, and peroxidases): (i) FTL_0200 inactivation may alter enzymatic activity but not protein stability, (ii) FTL_0200 may play only an indirect role in resistance to H₂O₂, (iii) the growth conditions used may play a role, or (iv) a technical limitation (the low sensitivity of the iTRAQ methodology to detect minor changes in protein amount) may be involved.

A link between stress resistance and pathogenesis.

As mentioned earlier, the proximal part of the moxR operon shares the same gene organization as the batL operon (batA to batE) of B. fragilis (34), a Gram-negative obligatory anaerobic bacterium (5, 28) that can survive exposure to atmospheric oxygen for a prolonged period of time. The screening of a bank of B. fragilis transposon (Tn) insertion mutants for growth in an oxygenated tissue culture monolayer model (34) allowed the isolation of a Tn insertion mutant, which grew poorly in the tissue cultures and showed a markedly reduced level of aerotolerance. The Tn insertion occurred within batD, the ortholog of FTL_0206. Here, we found that inactivation of either FTL-0200, FTL_0205, or FTL_0206 led to increased susceptibility to oxidative stress as well as to all the other stresses that we tested, supporting a role of this region in stress resistance in general.

Remarkably, similar loci are found in a broad number of bacterial species, including pathogenic species (gene cluster analysis available at the KEGG database http://ssdb.genome.jp/ssdb-bin/ssdb_gclust?org_gene=ftl:FTL_0200). Regarding pathogenic bacteria, the entire moxR operon (FTL_0200 to FTL_0209) is conserved only in Francisella genomes. In several genomes (e.g., Pseudomonas, Vibrio, and Bordetella species), the proximal part of the operon, comprising FTL_0200 to FTL_0206, is conserved. In other genomes (e.g., Mycobacterium and Borrelia species) genes orthologous to only FTL_0200, FT_0201, and FTL_0203 are found. In pathogenic bacteria such as Salmonella enterica serovar Typhimurium (gene STL3879), Shigella dysenteriae (gene sdy_4002), and Yersinia pestis (gene ypo0005), the moxR gene is comprised in a two-gene cluster and the downstream gene encodes a protein with a VWA domain (not shown). Finally, in Listeria monocytogenes, the gene encoding a MoxR ortholog is not associated with any gene encoding a protein containing a TPR or VWA motif. Hence, moxR can be either located within a cluster of genes encoding one or several proteins bearing VWA (and, in some cases, TPR) motifs or not associated with any VWA-containing protein. MoxR-like proteins may be active either alone or in combination with one or more “helper” proteins to constitute active chaperone complexes.

In conclusion, the present study identifies a novel F. tularensis virulence locus and confirms the important link between stress response and virulence. Interestingly, several genes of the F. tularensis moxR locus have been shown to be upregulated upon infection of bone marrow murine macrophages (37) in the F. tularensis Schu S4 strain. Further work will be required to elucidate the contribution of this locus to the F. tularensis intracellular lifestyle.