JOURNAL OF BACTERIOLOGY, Oct. 2011, p. 5683–5691 0021-9193/11/$12.00 doi:10.1128/JB.00428-11 Copyright © 2011, American Society for Microbiology. All Rights Reserved.
Vol. 193, No. 20
Brucella melitensis Cyclic di-GMP Phosphodiesterase BpdA Controls Expression of Flagellar Genes䌤† Erik Petersen,1 Pallab Chaudhuri,2 Chris Gourley,1 Jerome Harms,1 and Gary Splitter1* Department of Pathobiological Sciences, University of Wisconsin—Madison, 1656 Linden Dr., Madison, Wisconsin 53706,1 and Genetic Engineering of Bacteria Lab, Division of Bacteriology, Indian Veterinary Research Institute, Izatnagar, India2 Received 29 March 2011/Accepted 7 August 2011
Brucella melitensis encounters a variety of conditions and stimuli during its life cycle—including environmental growth, intracellular infection, and extracellular dissemination—which necessitates flexibility of bacterial signaling to promote virulence. Cyclic-di-GMP is a bacterial secondary signaling molecule that plays an important role in adaptation to changing environments and altering virulence in a number of bacteria. To investigate the role of cyclic-di-GMP in B. melitensis, all 11 predicted cyclic-di-GMP-metabolizing proteins were separately deleted and the effect on virulence was determined. Three of these cyclic-di-GMP-metabolizing proteins were found to alter virulence. Deletion of the bpdA and bpdB genes resulted in attenuation of virulence of the bacterium, while deletion of the cgsB gene produced a hypervirulent strain. In a Vibrio reporter system to monitor apparent alteration in levels of cyclic-di-GMP, both BpdA and BpdB displayed a phenotype consistent with cyclic-di-GMP-specific phosphodiesterases, while CgsB displayed a cyclic-di-GMP synthase phenotype. Further analysis found that deletion of bpdA resulted in a dramatic decrease in flagellar promoter activities, and a flagellar mutant showed similar phenotypes to the bpdA and bpdB mutant strains in mouse models of infection. These data indicate a potential role for regulation of flagella in Brucella melitensis via cyclic-di-GMP.
Recently, work in several systems has identified the bacterial secondary messenger cyclic-di-GMP (c-di-GMP) as a powerful mediator used by bacteria to respond to varied environments (75). c-di-GMP is generated by c-di-GMP synthases (diguanylate cyclases) containing GGDEF domains and degraded by c-di-GMP-specific phosphodiesterases containing EAL or HDGYP domains (60, 62, 63). The number of c-di-GMP-metabolizing proteins in bacterial species varies widely, from a single protein in some species up to nearly 100 in others (59). This complexity in signaling is also demonstrated in the variety of phenotypes that are reportedly regulated by c-di-GMP, including motility, biofilm formation, cell division, virulence, and intracellular survival (reviewed in references 58 and 67). To date, the role of c-di-GMP in B. melitensis has not been examined. The presence of 11 c-di-GMP-metabolizing proteins encoded by B. melitensis suggests a system of previously unknown regulation in Brucella. Further analysis of the identified proteins responsible for c-di-GMP synthesis and degradation found that deletion of two genes, BMEI1453 (Brucella phosphodiesterase A [bpdA]) and BMEI1448 (Brucella phosphodiesterase B [bpdB]), resulted in the attenuation of virulence during mouse infection, while the deletion of a third gene, BMEI1520 (c-di-GMP synthase of Brucella [cgsB]), resulted in increased virulence. Expression of the BpdA and BpdB proteins in a Vibrio reporter system for apparent alterations of c-di-GMP levels resulted in phenotypes typical of c-di-GMPspecific phosphodiesterases. CgsB expression in the same Vibrio system resulted in a phenotype consistent with that of a c-di-GMP synthase. Furthermore, deletion of the bpdA gene resulted in a dramatic downregulation of flagellar promoter activities and a deletion mutant with mutation of several fla-
Brucella spp. are Gram-negative, facultative intracellular pathogens and the causative agents responsible for the zoonotic disease brucellosis. Brucella melitensis is endemic in many areas of the world, causing sterility, abortions, and chronic disease in infected animals. Human infection with B. melitensis progresses from an initial period of undulating fever to a prolonged period of chronic infection potentially resulting in endocarditis, osteomyelitis, and meningitis (45). Brucella spp. are routinely passed from animals to humans through the consumption of contaminated milk or the inhalation of aerosolized bacteria generated from aborted placenta or during butchering practices (64). While the economic, environmental, and health costs of Brucella spp. are high, more remains to be known about Brucella virulence factors and their regulation. Brucella spp. must encounter several different environments during the course of an infection. The bacteria first adjust during transmission from the reservoir (either an infected animal or an environmental contamination) to a new host. Later, during infection, Brucella spp. must adapt to intracellular life in a wide variety of cellular types and tissues, including both professional and nonprofessional phagocytes and organs ranging from spleen and heart to placenta (6, 24, 26, 48, 56, 57). To survive this wide range of hostile encounters, Brucella spp. have evolved to swiftly adapt in response to a rapidly changing environment. * Corresponding author. Mailing address: Department of Pathobiological Sciences, University of Wisconsin—Madison, 203 AHABS, 1656 Linden Dr., Madison, WI 53706. Phone: (608) 262-1837. Fax: (608) 262-7420. E-mail:
[email protected]. † Supplemental material for this article may be found at http://jb .asm.org/. 䌤 Published ahead of print on 19 August 2011. 5683
5684
PETERSEN ET AL.
gellar genes also resulted in attenuation of virulence in our mouse models of infection. MATERIALS AND METHODS Bacterial strains, plasmids, and growth conditions. The strains and plasmids used in this study are listed in Table S1 in the supplemental material. Escherichia coli strain DH5␣ was used for propagation of cloning plasmids. ccdB Survival E. coli (Invitrogen) was used to propagate plasmids with the Gateway recombination region or ccdB gene. E. coli was grown in Luria-Bertani broth at 37°C. Brucella melitensis strain 16M (ATCC 23456) was used as the wild-type strain. All B. melitensis strains were maintained on brucella agar and grown in brucella broth (Becton Dickinson) at 37°C. Vibrio parahaemolyticus strains were grown at 30°C in heart infusion medium (Becton Dickinson) supplemented with 15g/liter NaCl. Kanamycin (50 g/ml), zeocin (50 g/ml for E. coli, 250 g/ml for Brucella), spectinomycin (100 g/ml), ampicillin (100 g/ml), chloramphenicol (20 g/ml), and gentamicin (25 g/ml) were added to the medium as needed. Identification of B. melitensis c-di-GMP-metabolizing domains. Protein-protein BLAST analysis of the B. melitensis 16M genome was conducted to identify c-di-GMP-metabolizing domains (3). The c-di-GMP synthase GGDEF domain from Caulobacter crescentus protein PleD and the c-di-GMP-specific phosphodiesterase EAL and HD-GYP domains from Pseudomonas aeruginosa protein PA2567 and Xanthomonas campestris protein RpfG were used to identify the respective domains in B. melitensis proteins (22, 32, 60). Further analysis of identified B. melitensis proteins was conducted using the SMART domain tool (39). Three genes were found to show a phenotype in this study, and to ease further discussion, these three genes were renamed BMEI1520 (cyclic-di-GMP synthase of Brucella [cgsB]), BMEI1453 Brucella phosphodiesterase A [bpdA]), and BMEI1448 Brucella phosphodiesterase B [bpdB]). Generation of deletion mutants of B. melitensis c-di-GMP-metabolizing proteins. Deletion mutants of each of the 11 B. melitensis genes found to encode a c-di-GMP-metabolizing domain were generated by double homologous recombination, as described previously (50). The oligonucleotides used to generate the deletion vectors are listed in Table S2 in the supplemental material. In short, the gene of interest and approximately 1 kb of DNA flanking either end of the gene were amplified by PCR. This PCR product was cloned into the pZErO-1 vector using KpnI and XhoI restriction sites found in the vector and oligonucleotides containing these restriction sites and sequence complementary to the targeted gene. An inverse PCR (IPCR) of the resulting vector was used to remove ⬎95% of the gene of interest, leaving only the pair of 1 kb of flanking DNAs. A kanamycin resistance cassette was cloned into the place of the gene of interest using PstI sites found flanking the kanamycin resistance cassette in pUC4K (Amersham) and in the IPCR oligonucleotides. The resulting deletion vector was electroporated into B. melitensis strain 16M. Kanamycin-resistant clones were selected for the loss of the pZErO-1 vector backbone by testing for zeocin sensitivity, indicating that replacement of the targeted gene with the kanamycin cassette had occurred. The resulting clones were tested by PCR for confirmation of deletion of the gene of interest. For in vivo imaging experiments, bpdA, bpdB, cgsB, and flgBCG fliE gene mutants were made in the GR023 background. GR023 is a strain of B. melitensis that carries the transposon-based Photorhabdus luminescens luxCDABE operon providing luminescence to the bacterium while preserving virulence (51). Mutant strains were generated as described above, using either a zeocin or ampicillin marker for deletion as GR023 is kanamycin resistant. All mutant strains were found to produce luminescence at the same levels as the wild-type GR023 strain (data not shown). Mouse infection with B. melitensis c-di-GMP-metabolizing mutants. The virulence of the B. melitensis c-di-GMP-metabolizing mutants was evaluated using an IRF-1⫺/⫺ murine model of infection as described previously (52). One million CFU of overnight brucella broth-grown culture in 0.2 ml of phosphate-buffered saline (PBS) of each of the 11 B. melitensis c-di-GMP-metabolizing mutants was injected intraperitoneally (i.p.) into groups of 6- to 8-week-old IRF-1⫺/⫺ mice (n ⫽ 7 or 8). Mice were monitored for mortality up to 60 days. Mutants with mutation of three genes (BMEI1448 [bpdB], BMEI1453 [bpdA], and BMEI1520 [cgsB]) were found to have altered virulence in the IRF-1⫺/⫺ murine model. These three mutants and wild-type strain 16M were analyzed for splenic colonization of IRF-1⫺/⫺ mice. Mice were infected as described above. At specified time points, mice (n ⫽ 3) were euthanized and their spleens were removed. The spleens were ground in 3 ml of PBS, and dilutions were plated to determine the amount of CFU in each spleen. Survival of IRF-1⫺/⫺ mice was analyzed using the Mantel-Cox test, and splenic CFU values were analyzed by Student’s t test. BALB/c mice (n ⫽ 4) were infected with luminescent bpdA, bpdB, and cgsB mutants and compared to wild-type B. melitensis GR023 infections. Mice were
J. BACTERIOL. infected as described above. Imaging of the mice was done using the Caliper Life Science in vivo imaging system. Quantification of the level of luminescence of each mouse was conducted at each imaging point using Living Image software (Caliper Life Science). Analysis of the data was conducted using analysis of variance (ANOVA) with Tukey’s comparison and compared to the wild-type level at each time point. Generation of expression plasmids for a Vibrio parahaemolyticus c-di-GMP reporter system. To determine the activity of the B. melitensis c-di-GMP-metabolizing proteins identified to play a role in virulence, each protein was expressed in a Vibrio parahaemolyticus reporter system. V. parahaemolyticus strain LM5984 contains a genomic transcriptional lacZ fusion to the capsular polysaccharide synthase (cps) promoter (8, 17, 20, 27). The V. parahaemolyticus cps promoter in this cps::lacZ strain has been shown to alter levels of LacZ in response to changing c-di-GMP concentrations induced by expression of c-di-GMP-metabolizing proteins (17, 27). Plasmid pLM1877 has been used in V. parahaemolyticus to express proteins from the IPTG(isopropyl--D-thiogalactopyranoside)-inducible Ptac promoter (7). To ease cloning of genes into pLM1877, the Gateway recombination region from pRH003 was cloned into pLM1877 downstream of the Ptac promoter, giving plasmid pEP95 (21). Proper orientation was confirmed by restriction enzyme digestion. Full-length B. melitensis proteins were transferred by Gateway reaction (Invitrogen) into pEP95 from pDONR201-based plasmids generated as part of the B. melitensis ORFeome (14). As positive controls for the V. parahaemolyticus cps::lacZ system, the c-di-GMP synthase PleD* and the c-di-GMP-specific phosphodiesterase PA2567 were also expressed. PleD* is a constitutive variant of the c-di-GMP synthase from Caulobacter crescentus and has been shown in several in vitro experiments to synthesize c-di-GMP (44, 46, 65). PA2567 is a c-di-GMPspecific phosphodiesterase from Pseudomonas aeruginosa that has been shown experimentally to degrade c-di-GMP (54, 60). Mutation of bpdA, bpdB, and cgsB to generate inactive (null) proteins for analysis in the Vibrio reporter system. The genes encoding the BpdA, BpdB, and CgsB proteins were mutated in an effort to determine whether inactivation of the protein would result in loss of a phenotype in the Vibrio reporter system. As the BpdB and CgsB proteins only carry transmembrane domains and the c-di-GMPmetabolizing domain, conserved signature motifs in each domain were mutated by alanine substitution of the active site residues. Site-directed mutagenesis (SDM) oligonucleotides were generated that spanned the active site residues, and nucleotides were replaced in the oligonucleotide that would result in a conversion of the active site residues to alanine residues. The EAL residues of BpdB were mutated to AAA, while the RMGGEEF residues of CgsB were mutated to AMGAAEF, both of which have been shown to abrogate activity in other domains (9, 28, 55, 68). pDONR-based plasmids containing either the bpdB or cgsB gene were extracted from a dam/dcm methylation-competent background to ensure parent DNA was methylated. PCR of the entire plasmid using the SDM oligonucleotides followed by methylation-specific restriction digest of the parent plasmid resulted in bpdB and cgsB active site mutations that were confirmed by DNA sequencing. These plasmids were then used in Gateway-based recombination for future use in the Vibrio reporter system. As BpdA contains several regulatory domains in addition to the EAL phosphodiesterase domain, the EAL domain was simply removed from the protein, leaving the transmembrane domain cluster, PAS fold, and inactive GGDEF domain. Oligonucleotides corresponding to the start codon and to the codons immediately upstream of the EAL domain were used to clone the C-terminal truncated bpdA gene into the plasmid pCR8 (Invitrogen). DNA sequencing confirmed deletion of the EAL domain. This plasmid was also used in Gatewaybased recombination reactions for use in the Vibrio reporter system. Analysis of B. melitensis c-di-GMP-metabolizing protein activity using the Vibrio reporter system. The expression plasmids generated above were transformed into the V. parahaemolyticus cps::lacZ strain (LM5984). Each strain was spread on a heart infusion agar plate, harvested after 18 h of growth at 30°C, and assayed for LacZ expression as described by Miller (43). Briefly, the A600 of the harvested bacteria was determined, and a 1:10 dilution was lysed using Koch’s lysis solution (49). The resulting preparation was incubated with ortho-nitrophenyl--galactoside (ONPG) at a final concentration of 0.67 mg/ml. After the reaction, the breakdown of ONPG was determined by measuring the A420. Miller units were quantified as (1,000 ⫻ A420)/(A600 ⫻ 0.1 ⫻ t), where t is the time of reaction, and relative Miller units were calculated by determining the fold difference from the vector-only control. As positive controls for the Vibrio assay, the genes encoding the constitutively active Caulobacter crescentus protein PleD* and the Pseudomonas aeruginosa PA01 protein PA2567 were cloned into the Gateway-ready donor vector pCR8 according to the manufacturer’s instructions (Invitrogen). The PleD* and
VOL. 193, 2011
CYCLIC DI-GMP IN BRUCELLA MELITENSIS
PA2567 proteins, respectively, generate and degrade c-di-GMP in both in vitro and in vivo assays (2, 44, 46, 54, 60). All samples were measured in triplicate, and the results of three replicate experiments are graphed. Data were analyzed using Student’s t test. Generation of luminescent reporter plasmids to monitor activity of three flagellar promoters within Brucella. Expression from three flagellar promoters in the bpdA, bpdB, and cgsB mutants and wild-type strain 16M were measured using the luminescent reporter vector pEP3 (53). Plasmid pEP3 contains a promoterless luxCDABE operon and a multiple-cloning site (MCS) upstream of the operon for insertion of promoters of interest. Three separate flagellar promoters (fliF, flgE, and flgB) were selected for analysis, and the infA (initiation factor A) promoter was used as a negative control (53). Approximately 500 bp upstream of the respective start codon was cloned into the MCS of pEP3 upstream of the luxCDABE operon in order to encompass a majority of the potential transcription factor binding sites (11). The four reporter plasmids were electroporated into the bpdA, bpdB, and cgsB mutants and the 16M wild-type strain for analysis. Flagellar promoter analysis in the bpdA, bpdB, and cgsB mutants using a luminescent reporter system. Strains were grown to stationary phase and then diluted to an optical density (OD) of 0.1 in brucella broth. Samples were plated in triplicate into a 96-well plate, and then the plate was transferred to a DTX 880 plate reader (Beckman Coulter). Samples were grown to an OD595 of 0.2, at which point the luminescence and OD595 values were recorded. The experiment was repeated in triplicate from three separate cultures grown from a streaked plate. To quantify luminescence of the promoter strains, relative light units (RLU) were calculated by determining the luminescence/OD595. The ratio of promoter activity in each mutant to wild-type activity was then calculated to give the fold difference in RLU for each promoter. Data were analyzed by Student’s t test.
5685
FIG. 1. BLAST analysis of the B. melitensis 16M genome identified 11 genes that encode proteins predicted to regulate levels of c-diGMP. Six genes encode a c-di-GMP synthase (GGDEF) domain, two genes encode a c-di-GMP phosphodiesterase (EAL) domain, and three genes encode both domains. Several other regulatory domains are also present, including REC domains (response regulator phosphorylation), PAS/PAC domains (cofactor regulatory domain, i.e., heme, flavin, etc.), and GAF domains (cyclic nucleotide or small molecule binding). Horizontal bars indicate predicted transmembrane helices.
RESULTS Brucella melitensis encodes 11 c-di-GMP-metabolizing proteins. c-di-GMP has been identified within several bacterial species as a secondary signaling molecule that regulates a shift between various lifestyles. To determine what role c-di-GMP plays in B. melitensis, BLAST analysis was conducted on the published 16M genome to identify c-di-GMP-metabolizing proteins (3). c-di-GMP is generated by c-di-GMP synthases (diguanylate cyclases) containing a GGDEF domain, while degradation of c-di-GMP is governed by c-di-GMP-specific phosphodiesterases containing either an EAL domain or an HD-GYP domain. Analysis of the B. melitensis 16M genome identified 11 genes that encode a GGDEF and/or EAL domain and none with an HD-GYP domain (Fig. 1). Six proteins contain a GGDEF domain, two proteins contain an EAL domain, and three proteins contain both a GGDEF domain and an EAL domain. Analysis of the latter group identified one gene, bpdA (BMEI1453), in which the usually conserved residues in the c-di-GMP synthase active site (GG[D/E]EF) instead encode the amino acids SSDQF. Mutation of GGDEF domain active site residues abrogates c-di-GMP synthase activity in other proteins and was therefore predicted to be inactive in BpdA (9, 28, 40). Two c-di-GMP-metabolizing B. melitensis mutants are attenuated and one mutant is hypervirulent in a mouse model of Brucella infection. Mutants with deletion of each of the 11 c-di-GMP-metabolizing proteins were generated in the B. melitensis 16M parent strain as described in Materials and Methods. Growth of each of the mutants in brucella broth and a minimal medium was identical to that of the parent strain. To gauge any attenuation in the mutant strains to infect and replicate within phagocytic cells, each was tested for intracellular levels after infection of RAW 264.7 macrophages. Each strain exhibited similar levels to the parent strain during macrophage infection, indicating that intracellular infection and growth
were not dramatically affected (data not shown). To further gauge the level of virulence of each of the mutants, an IRF1⫺/⫺ mouse model assay was conducted. IRF-1⫺/⫺ mice have been used previously to successfully determine whether Brucella mutants are still virulent, as these immunocompromised mice succumb to wild-type B. melitensis infection but survive infection with a variety of attenuated mutants (29, 30). For this assay, IRF-1⫺/⫺ mice (n ⫽ 8) were infected with each of the 11 mutants or the parent strain, and the time point to death after infection was determined. Three c-di-GMP-metabolizing mutants possessed a significant change in virulence (Fig. 2A). Mice infected with the wild-type strain 16M have a mean time to death of 10.7 ⫾ 0.5 days. The bpdA mutant was avirulent, with all mice living until the end of the experiment at 60 days. Mice infected with the bpdB mutant possessed intermediate virulence, with mice dying on average a week later than the parent strain-infected mice, resulting in a mean time to death of 18.0 ⫾ 1.3 days. cgsB mutant-infected mice displayed a hypervirulent phenotype, dying several days earlier than the parent strain-infected mice, with a mean time to death of 8.3 ⫾ 0.5 days. Complementation with a plasmid-based copy of the deleted gene fully restored the phenotype of the bpdA and cgsB mutants to wild-type levels and restored the phenotype of the bpdB mutant to a statistically nonsignificant level (see Fig. S1 in the supplemental material). Further characterization of the phenotype of the bpdA, bpdB, and cgsB mutants was accomplished by determining levels of bacteria in infected spleens at time points during IRF1⫺/⫺ mouse infection (Fig. 2B). Consistent with the shorter time to death and increase in virulence, mice infected with the cgsB hypervirulent mutant contained higher initial levels of splenic bacteria than the parent strain. Mice infected with the bpdB mutant had very low initial levels of splenic bacteria, but the splenic bacterial levels eventually increased and killed the
5686
PETERSEN ET AL.
FIG. 2. bpdA and bpdB mutants are attenuated in IRF-1⫺/⫺ mice, while the cgsB mutant is hypervirulent. (A) IRF-1⫺/⫺ mice (n ⫽ 7 or 8) were infected with mutants containing mutations of each of the 11 identified c-di-GMP-metabolizing genes. Three mutants possessed a significant change in the amount of time that it took for the mice to succumb to infection. (B) Spleens from IRF-1⫺/⫺ mice (n ⫽ 3 or 4/time point) infected with each of the three mutants were removed at the specified time points and plated for CFU determinations. The ⌬cgsB strain had a higher initial level of CFU, the ⌬bpdB strain had lower initial level until eventually reaching the wild-type level, and the ⌬bpdA strain had lower but persistent level throughout the course of infection. Survival data were analyzed using the Mantel-Cox test, and CFU data were analyzed using Student’s t test.
mice. bpdA mutant-infected mice also had lower levels of splenic bacteria, but those levels stayed fairly consistent for 2 weeks before declining over the course of the 60-day infection. However, these remaining levels of splenic bacteria are far greater than those exhibited by strains exhibiting rough lipopolysaccharide mutations or mutations in the type IV secretion system gene virB, two commonly studied avirulent Brucella mutants (data not shown). bpdA and bpdB mutants exhibit decreased dissemination within immunocompetent mice. bpdA, bpdB, and cgsB deletion mutants were generated in the GR023 luminescent background (51). These mutants were used to infect BALB/c mice to examine the virulence and dissemination of the mutants in an immunocompetent mouse (Fig. 3). Unlike the immunocompromised IRF-1⫺/⫺ mice, BALB/c mice typically exhibit a peak level of bacteria between 5 and 7 days postinfection, followed by the subsequent clearance below detectable levels. The wildtype GR023 strain possessed classical dissemination and virulence levels, initially localizing near the site of injection (lower
J. BACTERIOL.
left abdomen), spreading to the rest of the mouse, peaking at day 5, and declining by the second week. The hypervirulent cgsB mutant infected at a higher level than the wild type, disseminating and peaking before the wild type at day 3. The bpdB mutant spread throughout the mouse similarly to the wild type, but at a lower level of infection, similar to observations in the IRF-1⫺/⫺ mice. The highly attenuated bpdA mutant never disseminated far from the site of injection before rapidly decreasing. CgsB exhibits the c-di-GMP synthase phenotype, while BpdA and BpdB exhibit the c-di-GMP phosphodiesterase phenotype in a Vibrio reporter system. After determining that mutants with deletion of the three B. melitensis c-di-GMPmetabolizing proteins BpdA, BpdB, and CgsB possessed a significant phenotype in mouse models of Brucella infection, the functional activity of the proteins toward c-di-GMP was determined. Vibrio spp. have been well studied with regard to c-di-GMP due to the very recognizable phenotypes displayed under high-c-di-GMP (biofilm formation) and low-c-di-GMP (motility) conditions. A V. parahaemolyticus strain with a lacZ transcriptional fusion to the capsular polysaccharide synthase promoter (cps) was used for analysis of cps promoter activity (20). Several previously published reports have indicated that the level of LacZ production from the cpsA3::lacZ-Genr fusion is directly tied to the level of cyclic-di-GMP that has been altered by the expression of c-di-GMP-metabolizing proteins (17, 27). Two control proteins (the constitutive c-di-GMP synthase PleD* and the phosphodiesterase PA2567) and the three B. melitensis proteins (BpdA, BpdB, and CgsB) were expressed within the V. parahaemolyticus cpsA3::lacZ-Genr strain (Fig. 4). PleD* is a constitutively active c-di-GMP synthase from Caulobacter crescentus that has been shown in several assays to generate c-di-GMP (44, 46, 65). The increased level of LacZ activity seen with expression of PleD* is consistent with the predicted result of expression of a c-di-GMP synthase which generates c-di-GMP, increasing activity from the cps promoter and generating a higher level of LacZ activity. Similarly, expression of CgsB supported the fact that it was a functional c-di-GMP synthase as it exhibited higher levels of LacZ activity, similar to that of the control PleD*. Expression of the protein PA2567 from Pseudomonas aeruginosa resulted in a decrease in LacZ activity, consistent with the expected results of expression of a c-di-GMP-phosphodiesterase in which a lowering of c-di-GMP levels by PA2567 reduces the activity of the cps promoter (54, 60). Similarly, expression of both BpdA and BpdB decreased levels of LacZ activity similar to that of PA2567, supporting the case that both of these proteins are c-di-GMP phosphodiesterases. To further characterize the BpdA, BpdB, and CgsB proteins, null mutants of each were generated and expressed in the Vibrio reporter system. The BpdB and CgsB null mutants each contained alanine substitutions in the active site residues of their respective c-di-GMP-metabolizing domains. The BpdA null mutant had the entire EAL domain truncated from the C-terminal portion of the protein, leaving the transmembrane domains, PAS fold, and the inactive GGDEF domain. All three null mutants had their activity reduced to the level of the vector-only control, further supporting our characterization of
VOL. 193, 2011
CYCLIC DI-GMP IN BRUCELLA MELITENSIS
5687
FIG. 3. BALB/c mice (n ⫽ 4) were infected intraperitoneally with bpdA, bpdB, and cgsB mutants generated in the luminescent GR023 background. (A) In vivo imaging of infected mice showed that the cgsB mutant rapidly reached a high level of bacteria, while the bpdA and bpdB mutants both had decreased levels across the entire time of infection. A scale shows the results in photons per second per cm2 per steradian. (B) Quantification of luminescence levels from infected BALB/c mice indicates that the cgsB mutant peaked 2 days before the wild type, while the bpdA and bpdB mutants both peaked at much lower levels. Luminescence of the entire mouse was analyzed by ANOVA (Tukey’s) and compared to that of the wild type at each time point.
the CgsB protein as a c-di-GMP synthase and the BpdA and BpdB proteins as c-di-GMP-specific phosphodiesterases. Flagellar transcription is strongly downregulated in the bpdA mutant. Work in several different bacterial species has identified that c-di-GMP is a common regulator of flagellar production (2, 5, 31, 47). In these cases, low levels of c-di-GMP
FIG. 4. B. melitensis proteins BpdA, BpdB, and CgsB were overexpressed in a Vibrio parahaemolyticus model system to determine their ability to regulate levels of c-di-GMP via a LacZ reporter under the control of the c-di-GMP-regulated capsular polysaccharide gene (cps). A known c-di-GMP synthase (PleD*) and phosphodiesterase (PA2567) were expressed as controls. BpdA and BpdB both decreased cps promoter activity similar to the c-di-GMP phosphodiesterase PA2567, while CgsB increased cps promoter activity similar to the c-di-GMP synthase PleD*. Null mutants of all three B. melitensis proteins returned promoter levels to that of the vector-only control. Miller unit measurements are presented relative to the vector-only control. The data presented are the results from three experiments measured in triplicate and were analyzed using Student’s t test.
promoted motility and increased flagellar transcription, while high levels of c-di-GMP promoted sessility. Although B. melitensis is nonmotile, it has been shown to produce a flagellum, and flagellar mutants are attenuated in mouse and goat models of infection, suggesting a role for flagella in B. melitensis virulence (18, 38, 76). To determine if the identified proteins BpdA, BpdB, and CgsB regulate flagellar transcription in B. melitensis, the promoter activities of three separate flagellar operons were assayed in each mutant. Promoter regions upstream of the fliF, flgE, and flgB genes were tested for expression in wild-type strain 16M and each mutant. These promoters were selected as (i) they have been shown to be differentially regulated in other B. melitensis mutants, (ii) mutagenesis of fliF has led to reduced virulence, and (iii) they are predicted to occur at the furthest endpoint of flagellar regulation before the final flagellin regulation that has been shown to be atypical in B. melitensis (15, 16, 38, 53, 73). In all three mutant strains, regulation of the negative control promoter infA was unchanged. Furthermore, none of the flagellar promoters showed a significant change (greater than 2-fold change from wild type) in either the ⌬bpdB or ⌬cgsB strains. However, all three flagellar promoters were downregulated approximately 3- to 7-fold in the ⌬bpdA strain compared to wild-type strain 16M (Fig. 5). These data together with the predicted function of the BpdA protein suggest that B. melitensis under certain elevated c-di-GMP concentrations represses flagellar production. Flagellar mutant exhibits reduced virulence in mouse models of infection. The promoter activity of flagellar genes was strongly downregulated in the bpdA mutant, suggesting that levels of flagella were also downregulated. To examine the relationship between the BpdA protein and flagella in B. melitensis, a flagellar mutant was generated in the luminescent
5688
PETERSEN ET AL.
J. BACTERIOL.
data are in contrast to ⌬bpdA mutant-infected mice that never succumbed to infection, possibly indicating that additional factors are regulated by the BpdA protein in addition to flagella that result in additional attenuation of virulence. DISCUSSION
FIG. 5. Loss of BpdA results in a 3- to 7-fold decrease in activity of three flagellar promoters. Three separate flagellar operons were tested for promoter activity in the wild-type strain 16M and mutant strains using a luminescent reporter assay. All three operons were highly downregulated in the ⌬bpdA strain, indicating that the BpdA protein regulates flagellar gene expression in B. melitensis. Promoter levels were graphed as fold difference relative to wild-type levels of promoter activity. The data presented are the results from three experiments, each measured in triplicate and analyzed using Student’s t test.
GR023 background. A four-gene operon encoding flagellar proteins that constitute the transmembrane flagellar rod (flgBCGfliE, corresponding to BMEI1086 to -9, respectively) was deleted, and the mutant was examined alongside the bpdA mutant in both the BALB/c and IRF-1⫺/⫺ mouse models of Brucella infection (Fig. 6). In vivo imaging of BALB/c mice (n ⫽ 4) with the luminescent mutant shows that initial infection with the flagellar mutant is also restricted to the site of injection (lower left, i.p.). While the flagellar mutant eventually lightly disseminated outside the site of injection, it never reached the level of wild-type infection (Fig. 6A). Infection of immunocompromised IRF1⫺/⫺ mice with the flagellar mutant resulted in attenuated virulence, with the mice eventually succumbing to infection, with a mean time to death of 19.0 ⫾ 4.8 days (Fig. 6B). These
The identification of c-di-GMP-metabolizing proteins that play a role in virulence of B. melitensis suggests a previously unidentified mechanism(s) of regulation of virulence. Signature-tagged mutagenesis of B. melitensis during acute infection first identified the “EAL-domain hypothetical protein” BpdB (BMEI1448) as a potential regulator of virulence (38). This work has identified a second c-di-GMP-specific phosphodiesterase, BpdA (BMEI1453), that shows an even further level of attenuation of virulence, and a c-di-GMP synthase, CgsB (BMEI1520), that shows increased virulence in mouse models of infection. Together with the phenotypes seen during expression of the B. melitensis proteins in the Vibrio reporter system for c-diGMP regulation, we can begin to hypothesize what role c-diGMP may play in Brucella virulence. Deletion of the cgsB c-di-GMP synthase would be expected to lower potential levels of c-di-GMP, leading to an increase in virulence. On the other hand, deletion of the bpdA or bpdB c-di-GMP phosphodiesterases would potentially increase the level of c-di-GMP and result in a corresponding decrease in virulence. This hypothesis would be consistent with that proposed for other models in which production of virulence factors (i.e., cholera toxin and flagella) requires a reduction in c-di-GMP and that loss of these virulence factors is accompanied by a decrease in virulence (5, 10, 35, 71). Further analysis of the bpdA mutant indicated that activation levels of several flagellar promoters were strongly downregulated compared to wild-type levels, suggesting that flagellar production is decreased in the mutant. Previously published results indicate that flagellar mutants of B. melitensis do not
FIG. 6. Deletion of B. melitensis flagellar genes results in attenuation in virulence in mouse models of infection. A four-gene operon encoding flagellar components of the central rod (flgB, flgC, flgG, and fliE) was deleted in the GR023 luminescent, virulent B. melitensis strain. (A) Infection of BALB/c mice (n ⫽ 4) with the flagellar mutant strain results in an attenuated phenotype in which the bacteria fail to fully disseminate and instead localize at the site of infection. A scale shows the results in photons per second per cm2 per steradian. (B) Infection of IRF-1⫺/⫺ mice with the flagellar mutant results in prolonged survival but eventual death of the mice.
VOL. 193, 2011
show a phenotype different from the wild type in cellular models of infection, similar to the bpdA mutant, but do show a phenotype in mice (18, 38). To establish a correlation between flagellar production and virulence, a flagellar deletion mutant was generated and compared to the bpdA mutants. While the flagellar mutant disseminates similar to the bpdA mutant, IRF1⫺/⫺ mice infected with the flagellar mutant still succumb to infection while ⌬bpdA mutant-infected mice do not, potentially indicating a role of the BpdA protein in virulence in addition to flagellar regulation (18, 38). These data suggest a possible role for c-di-GMP in the regulation of virulence through flagellar expression in B. melitensis. The role of flagella in B. melitensis is still an evolving field. Lack of a chemotactic system, truncation of several flagellar genes, and historical classification as a nonmotile bacterium meant that identification of flagellar genes within the genome of B. melitensis was initially hypothesized to either be the result of cryptic remnants or as part of a type III secretory system (1, 13). More recent work found that not only does B. melitensis produce a sheathed flagellum under the proper conditions, but deletion of flagellar genes results in attenuation of virulence in both mouse and goat models of infection (18, 38, 76). Rather than motility, it is possible that B. melitensis uses flagella for attachment, secretion, or as a source of immune system activation through either Toll-like receptor 5 (TLR5) or IpaF (42). The exact role of c-di-GMP in the regulation of flagella in B. melitensis is still unknown. Flagella are often regulated in tiers, with the proteins of the hook-basal body occupying tier II, downstream of regulatory factors in tier I but upstream of the chemotaxis and filament proteins of tier III (12). The master regulator FtcR was found to regulate expression of several flagellar genes, possibly under the control of the quorum-sensing regulator VjbR (37). A second quorum-sensing regulator, BlxR, also regulates expression levels of flagellar genes (53). Recent work has furthered our knowledge of the regulation of flagella in B. melitensis, identifying both a sigma factor (RpoE1) that represses expression of flagellar genes and a flagellin activator responsible for regulation of the fliC gene (FlbT) (15, 16). As the flagellin filament protein FliC was shown to be produced even in B. melitensis mutants lacking assembly of the flagellar hook, promoter activity of third tier proteins was not examined in this study (16). By selecting structural proteins downstream of several regulatory checkpoints for our assay, the cumulative effect of c-di-GMP on flagellar expression was determined. However, detailed work at each of the checkpoints in flagellar synthesis will be required to pinpoint the location(s) of c-di-GMP involvement. The mechanism of signaling achieved by c-di-GMP in B. melitensis is also unknown. Three separate mechanisms have been shown in other bacterial species, although it is likely that more remain. The first is at the protein level, where c-di-GMP binds directly to a protein and activates an adjacent enzymatic or regulatory domain. These proteins include PilZ domainencoding proteins in components of cellulose synthase and type IV pilin apparatus and the PelD polysaccharide regulator (4, 36, 61, 74). Unfortunately, none of these c-di-GMP-binding domains show homology to any B. melitensis protein. The second method of regulation is the recently discovered c-di-GMP riboswitch. This upstream mRNA element regulates terminator formation in the mRNA based upon binding to c-di-GMP
CYCLIC DI-GMP IN BRUCELLA MELITENSIS
5689
(66). Again though, previous sequence analysis of the B. melitensis genome did not identify any c-di-GMP-responsive riboswitches (66). The third mechanism of c-di-GMP regulation is through the binding of c-di-GMP to transcription factors like the FleQ flagellar synthesis regulator and cyclic AMP receptor protein (CRP)-like proteins (Clp) (23). While B. melitensis doesn’t encode a protein with any similarity to the FleQ protein, it encodes several proteins with homology to the Clp family of proteins. A subset of these B. melitensis Clp proteins conserve a glutamate residue shown to be required for c-di-GMP binding, but further analysis of these proteins would be required to suggest that they possess the ability to bind c-di-GMP (69). In addition to regulation of flagella, it is likely that other cellular processes are controlled by c-di-GMP in B. melitensis. CgsB and BpdB do not appear to affect flagellar production in our assay under the conditions tested, but both contribute to the virulence of the bacterium. BMEII0660 encodes a putative c-di-GMP synthase with homology to the family of PleD response regulators whose activity has been shown to play a role in motility and replication in Caulobacter crescentus and host cell infection and survival in Anaplasma phagocytophilum and Ehrlichia chaffeensis (22, 33, 34). Additional c-di-GMP-metabolizing proteins encoded by B. melitensis also contain domains like PAS and GAF domains that have been found in other instances to respond to oxygen levels, redox, and cyclic nucleotides (41, 70). Recent reports have also identified the ability of B. melitensis to form biofilms after certain genetic modifications, another network commonly regulated by c-di-GMP (19, 25, 72). While there have been no identified instances in which B. melitensis is naturally found in biofilms, it is possible that B. melitensis uses c-di-GMP to regulate production of a biofilm in specific, yet unidentified, environments. In conclusion, we have identified three B. melitensis proteins that actively regulate levels of the secondary messenger c-diGMP, opening up a wealth of new possibilities for B. melitensis regulation. A c-di-GMP-specific phosphodiesterase, BpdA, was found to regulate the expression of several flagellar components, suggesting a role for c-di-GMP in the regulation of the B. melitensis flagellum. Further experimentation will seek to identify the mechanisms and stimuli of this regulation. The presence of an additional 8 predicted c-di-GMP-metabolizing proteins indicates that there may be several additional pathways regulated by c-di-GMP that are yet to be uncovered. ACKNOWLEDGMENTS This work was supported by NIH 1R01AI073558, NIH R21 AI088038, and BARD grant US-4378-11. Erik Petersen was supported by a Molecular Biosciences Training grant (NIH T32 GM07215-34). We also thank Urs Jenal at the University of Basel, Switzerland, for the gift of the PleD* gene, Linda McCarter at the University of Iowa for the Vibrio parahaemolyticus strains, and Susan West at the University of Wisconsin—Madison for the Pseudomonas aeruginosa PA01 strain. Plasmid construction was aided by Diogo Magnani. Vibrio parahaemolyticus Miller assays were accomplished with help from Michael Cho. REFERENCES 1. Abdallah, A. I., et al. 2003. Type III secretion homologs are present in Brucella melitensis, B. ovis, and B. suis biovars 1, 2, and 3. Curr. Microbiol. 46:241–245. 2. Aldridge, P., R. Paul, P. Goymer, P. Rainey, and U. Jenal. 2003. Role of the
5690
3. 4. 5.
6.
7.
8.
9.
10.
11.
12.
13. 14. 15.
16. 17.
18.
19.
20.
21.
22.
23.
24.
25. 26. 27.
28.
29.
30.
31.
PETERSEN ET AL.
GGDEF regulator PleD in polar development of Caulobacter crescentus. Mol. Microbiol. 47:1695–1708. Altschul, S. F., W. Gish, W. Miller, E. W. Myers, and D. J. Lipman. 1990. Basic local alignment search tool. J. Mol. Biol. 215:403–410. Amikam, D., and M. Y. Galperin. 2006. PilZ domain is part of the bacterial c-di-GMP binding protein. Bioinformatics 22:3–6. Beyhan, S., A. D. Tischler, A. Camilli, and F. H. Yildiz. 2006. Transcriptome and phenotypic responses of Vibrio cholerae to increased cyclic di-GMP level. J. Bacteriol. 188:3600–3613. Billard, E., C. Cazevieille, J. Dornand, and A. Gross. 2005. High susceptibility of human dendritic cells to invasion by the intracellular pathogens Brucella suis, B. abortus, and B. melitensis. Infect. Immun. 73:8418–8424. Boles, B. R., and L. L. McCarter. 2000. Insertional inactivation of genes encoding components of the sodium-type flagellar motor and switch of Vibrio parahaemolyticus. J. Bacteriol. 182:1035–1045. Boles, B. R., and L. L. McCarter. 2002. Vibrio parahaemolyticus scrABC, a novel operon affecting swarming and capsular polysaccharide regulation. J. Bacteriol. 184:5946–5954. Christen, M., B. Christen, M. Folcher, A. Schauerte, and U. Jenal. 2005. Identification and characterization of a cyclic di-GMP-specific phosphodiesterase and its allosteric control by GTP. J. Biol. Chem. 280:30829–30837. Claret, L., et al. 2007. The flagellar sigma factor Flia regulates adhesion and invasion of Crohn’s disease-associated Escherichia coli via a c-di-GMP-dependent pathway. J. Biol. Chem. 282:33275–33283. Collado-Vides, J., B. Magasanik, and J. D. Gralla. 1991. Control site location and transcriptional regulation in Escherichia coli. Microbiol. Rev. 55: 371–394. Dasgupta, N., et al. 2003. A four-tiered transcriptional regulatory circuit controls flagellar biogenesis in Pseudomonas aeruginosa. Mol. Microbiol. 50:809–824. DelVecchio, V. G., et al. 2002. Brucella proteomes—a review. Vet. Microbiol. 90:593–603. Dricot, A., et al. 2004. Generation of the Brucella melitensis ORFeome version 1.1. Genome Res. 14:2201–2206. Ferooz, J., J. Lemaire, M. Delory, X. De Bolle, and J. J. Letesson. 2011. RpoE1, an extracytoplasmic sigma factor, is a repressor of the flagellar system in Brucella melitensis. Microbiology 157:1263–1268. Ferooz, J., J. Lemaire, and J. J. Letesson. 2011. Role of FlbT in flagellin production in Brucella melitensis. Microbiology 157:1253–1262. Ferreira, R. B., L. C. Antunes, E. P. Greenberg, and L. L. McCarter. 2008. Vibrio parahaemolyticus ScrC modulates cyclic dimeric GMP regulation of gene expression relevant to growth on surfaces. J. Bacteriol. 190:851–860. Fretin, D., et al. 2005. The sheathed flagellum of Brucella melitensis is involved in persistence in a murine model of infection. Cell. Microbiol. 7:687–698. Godefroid, M., et al. 2010. Brucella melitensis 16M produces a mannan and other extracellular matrix components typical of a biofilm. FEMS Immunol. Med. Microbiol. 59:364–377. Guvener, Z. T., and L. L. McCarter. 2003. Multiple regulators control capsular polysaccharide production in Vibrio parahaemolyticus. J. Bacteriol. 185: 5431–5441. Hallez, R., J. J. Letesson, J. Vandenhaute, and X. De Bolle. 2007. Gatewaybased destination vectors for functional analyses of bacterial ORFeomes: application to the Min system in Brucella abortus. Appl. Environ. Microbiol. 73:1375–1379. Hecht, G. B., and A. Newton. 1995. Identification of a novel response regulator required for the swarmer-to-stalked-cell transition in Caulobacter crescentus. J. Bacteriol. 177:6223–6229. Hickman, J. W., and C. S. Harwood. 2008. Identification of FleQ from Pseudomonas aeruginosa as a c-di-GMP-responsive transcription factor. Mol. Microbiol. 69:376–389. Hong, P. C., R. M. Tsolis, and T. A. Ficht. 2000. Identification of genes required for chronic persistence of Brucella abortus in mice. Infect. Immun. 68:4102–4107. Jonas, K., O. Melefors, and U. Romling. 2009. Regulation of c-di-GMP metabolism in biofilms. Future Microbiol. 4:341–358. Kim, S., et al. 2005. Interferon-gamma promotes abortion due to Brucella infection in pregnant mice. BMC Microbiol. 5:22. Kim, Y. K., and L. L. McCarter. 2007. ScrG, a GGDEF-EAL protein, participates in regulating swarming and sticking in Vibrio parahaemolyticus. J. Bacteriol. 189:4094–4107. Kirillina, O., J. D. Fetherston, A. G. Bobrov, J. Abney, and R. D. Perry. 2004. HmsP, a putative phosphodiesterase, and HmsT, a putative diguanylate cyclase, control Hms-dependent biofilm formation in Yersinia pestis. Mol. Microbiol. 54:75–88. Ko, J., A. Gendron-Fitzpatrick, T. A. Ficht, and G. A. Splitter. 2002. Virulence criteria for Brucella abortus strains as determined by interferon regulatory factor 1-deficient mice. Infect. Immun. 70:7004–7012. Ko, J., A. Gendron-Fitzpatrick, and G. A. Splitter. 2002. Susceptibility of IFN regulatory factor-1 and IFN consensus sequence binding protein-deficient mice to brucellosis. J. Immunol. 168:2433–2440. Kuchma, S. L., et al. 2007. BifA, a c-di-GMP phosphodiesterase, inversely
J. BACTERIOL.
32.
33.
34.
35.
36. 37.
38.
39. 40.
41.
42.
43. 44.
45. 46.
47. 48.
49.
50.
51.
52.
53.
54.
55.
56. 57.
58.
regulates biofilm formation and swarming motility by Pseudomonas aeruginosa PA14. J. Bacteriol. 189:8165–8178. Kulesekara, H., et al. 2006. Analysis of Pseudomonas aeruginosa diguanylate cyclases and phosphodiesterases reveals a role for bis-(3⬘-5⬘)-cyclic-GMP in virulence. Proc. Natl. Acad. Sci. U. S. A. 103:2839–2844. Kumagai, Y., J. Matsuo, Y. Hayakawa, and Y. Rikihisa. 2010. Cyclic di-GMP signaling regulates invasion by Ehrlichia chaffeensis of human monocytes. J. Bacteriol. 192:4122–4133. Lai, T. H., Y. Kumagai, M. Hyodo, Y. Hayakawa, and Y. Rikihisa. 2008. The Anaplasma phagocytophilum PleC histidine kinase and PleD diguanylate cyclase two-component system and role of cyclic di-GMP in host-cell infection. J. Bacteriol. 191:693–700. Lee, H. S., F. Gu, S. M. Ching, Y. Lam, and K. L. Chua. 2010. CdpA is a Burkholderia pseudomallei cyclic di-GMP phosphodiesterase involved in autoaggregation, flagellum synthesis, motility, biofilm formation, cell invasion, and cytotoxicity. Infect. Immun. 78:1832–1840. Lee, V. T., et al. 2007. A cyclic-di-GMP receptor required for bacterial exopolysaccharide production. Mol. Microbiol. 65:1474–1484. Leonard, S., et al. 2007. FtcR is a new master regulator of the flagellar system of Brucella melitensis 16M with homologs in Rhizobiaceae. J. Bacteriol. 189:131–141. Lestrate, P., et al. 2003. Attenuated signature-tagged mutagenesis mutants of Brucella melitensis identified during the acute phase of infection in mice. Infect. Immun. 71:7053–7060. Letunic, I., T. Doerks, and P. Bork. 2009. SMART 6: recent updates and new developments. Nucleic Acids Res. 37:D229–D232. Liu, N., T. Pak, and E. M. Boon. 2010. Characterization of a diguanylate cyclase from Shewanella woodyi with cyclase and phosphodiesterase activities. Mol. Biosyst. 6:1561–1564. Martinez, S. E., J. A. Beavo, and W. G. Hol. 2002. GAF domains: two-billionyear-old molecular switches that bind cyclic nucleotides. Mol. Interv. 2:317– 323. Miao, E. A., E. Andersen-Nissen, S. E. Warren, and A. Aderem. 2007. TLR5 and Ipaf: dual sensors of bacterial flagellin in the innate immune system. Semin. Immunopathol. 29:275–288. Miller, J. F. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Neunuebel, M. R., and J. W. Golden. 2008. The Anabaena sp. strain PCC 7120 gene all2874 encodes a diguanylate cyclase and is required for normal heterocyst development under high-light growth conditions. J. Bacteriol. 190:6829–6836. Pappas, G., N. Akritidis, M. Bosilkovski, and E. Tsianos. 2005. Brucellosis. N. Engl. J. Med. 352:2325–2336. Paul, R., et al. 2004. Cell cycle-dependent dynamic localization of a bacterial response regulator with a novel di-guanylate cyclase output domain. Genes Dev. 18:715–727. Pesavento, C., et al. 2008. Inverse regulatory coordination of motility and curli-mediated adhesion in Escherichia coli. Genes Dev. 22:2434–2446. Pizarro-Cerda, J., et al. 1998. Brucella abortus transits through the autophagic pathway and replicates in the endoplasmic reticulum of nonprofessional phagocytes. Infect. Immun. 66:5711–5724. Putnam, S. L., and A. L. Koch. 1975. Complications in the simplest cellular enzyme assay: lysis of Escherichia coli for the assay of beta-galactosidase. Anal. Biochem. 63:350–360. Rajashekara, G., D. A. Glover, M. Banai, D. O’Callaghan, and G. A. Splitter. 2006. Attenuated bioluminescent Brucella melitensis mutants GR019 (virB4), GR024 (galE), and GR026 (BMEI1090-BMEI1091) confer protection in mice. Infect. Immun. 74:2925–2936. Rajashekara, G., D. A. Glover, M. Krepps, and G. A. Splitter. 2005. Temporal analysis of pathogenic events in virulent and avirulent Brucella melitensis infections. Cell. Microbiol. 7:1459–1473. Rajashekara, G., et al. 2005. Unraveling Brucella genomics and pathogenesis in immunocompromised IRF-1⫺/⫺ mice. Am. J. Reprod. Immunol. 54:358– 368. Rambow-Larsen, A. A., G. Rajashekara, E. Petersen, and G. Splitter. 2008. Putative quorum-sensing regulator BlxR of Brucella melitensis regulates virulence factors including the type IV secretion system and flagella. J. Bacteriol. 190:3274–3282. Rao, F., et al. 2009. The functional role of a conserved loop in EAL domainbased cyclic di-GMP-specific phosphodiesterase. J. Bacteriol. 191:4722– 4731. Rao, F., Y. Yang, Y. Qi, and Z. X. Liang. 2008. Catalytic mechanism of cyclic di-GMP-specific phosphodiesterase: a study of the EAL domain-containing RocR from Pseudomonas aeruginosa. J. Bacteriol. 190:3622–3631. Reguera, J. M., et al. 2003. Brucella endocarditis: clinical, diagnostic, and therapeutic approach. Eur. J. Clin. Microbiol. Infect. Dis. 22:647–650. Rittig, M. G., M. T. Alvarez-Martinez, F. Porte, J. P. Liautard, and B. Rouot. 2001. Intracellular survival of Brucella spp. in human monocytes involves conventional uptake but special phagosomes. Infect. Immun. 69:3995–4006. Romling, U. 2009. Cyclic di-GMP (c-di-GMP) goes into host cells: c-di-GMP signaling in the obligate intracellular pathogen Anaplasma phagocytophilum. J. Bacteriol. 191:683–686.
VOL. 193, 2011 59. Romling, U., M. Gomelsky, and M. Y. Galperin. 2005. C-di-GMP: the dawning of a novel bacterial signalling system. Mol. Microbiol. 57:629–639. 60. Ryan, R. P., et al. 2006. Cell-cell signaling in Xanthomonas campestris involves an HD-GYP domain protein that functions in cyclic di-GMP turnover. Proc. Natl. Acad. Sci. U. S. A. 103:6712–6717. 61. Ryjenkov, D. A., R. Simm, U. Romling, and M. Gomelsky. 2006. The PilZ domain is a receptor for the second messenger c-di-GMP: the PilZ domain protein YcgR controls motility in enterobacteria. J. Biol. Chem. 281:30310– 30314. 62. Ryjenkov, D. A., M. Tarutina, O. V. Moskvin, and M. Gomelsky. 2005. Cyclic diguanylate is a ubiquitous signaling molecule in bacteria: insights into biochemistry of the GGDEF protein domain. J. Bacteriol. 187:1792–1798. 63. Schmidt, A. J., D. A. Ryjenkov, and M. Gomelsky. 2005. The ubiquitous protein domain EAL is a cyclic diguanylate-specific phosphodiesterase: enzymatically active and inactive EAL domains. J. Bacteriol. 187:4774–4781. 64. Seleem, M. N., S. M. Boyle, and N. Sriranganathan. 2009. Brucellosis: a re-emerging zoonosis. Vet. Microbiol. 140:392–398. 65. Spangler, C., A. Bohm, U. Jenal, R. Seifert, and V. Kaever. 2010. A liquid chromatography-coupled tandem mass spectrometry method for quantitation of cyclic di-guanosine monophosphate. J. Microbiol. Methods 81:226– 231. 66. Sudarsan, N., et al. 2008. Riboswitches in eubacteria sense the second messenger cyclic di-GMP. Science 321:411–413. 67. Tamayo, R., J. T. Pratt, and A. Camilli. 2007. Roles of cyclic diguanylate in the regulation of bacterial pathogenesis. Annu. Rev. Microbiol. 61:131–148.
CYCLIC DI-GMP IN BRUCELLA MELITENSIS
5691
68. Tamayo, R., A. D. Tischler, and A. Camilli. 2005. The EAL domain protein VieA is a cyclic diguanylate phosphodiesterase. J. Biol. Chem. 280:33324– 33330. 69. Tao, F., Y. W. He, D. H. Wu, S. Swarup, and L. H. Zhang. 2010. The cyclic nucleotide monophosphate domain of Xanthomonas campestris global regulator Clp defines a new class of cyclic di-GMP effectors. J. Bacteriol. 192:1020–1029. 70. Taylor, B. L., and I. B. Zhulin. 1999. PAS domains: internal sensors of oxygen, redox potential, and light. Microbiol. Mol. Biol. Rev. 63:479–506. 71. Tischler, A. D., and A. Camilli. 2005. Cyclic diguanylate regulates Vibrio cholerae virulence gene expression. Infect. Immun. 73:5873–5882. 72. Uzureau, S., et al. 2007. Mutations of the quorum sensing-dependent regulator VjbR lead to drastic surface modifications in Brucella melitensis. J. Bacteriol. 189:6035–6047. 73. Weeks, J. N., et al. 2010. Brucella melitensis VjbR and C12-HSL regulons: contributions of the N-dodecanoyl homoserine lactone signaling molecule and LuxR homologue VjbR to gene expression. BMC Microbiol. 10:167. 74. Weinhouse, H., et al. 1997. c-di-GMP-binding protein, a new factor regulating cellulose synthesis in Acetobacter xylinum. FEBS Lett. 416:207–211. 75. Yildiz, F. H. 2008. Cyclic dimeric GMP signaling and regulation of surfaceassociated developmental programs. J. Bacteriol. 190:781–783. 76. Zygmunt, M. S., S. D. Hagius, J. V. Walker, and P. H. Elzer. 2006. Identification of Brucella melitensis 16M genes required for bacterial survival in the caprine host. Microbes Infect. 8:2849–2854.
