Attenuated Bioluminescent Brucella melitensis Mutants GR019 (virB4), GR024 (galE), and GR026 (BMEI1090-BMEI1091) Confer Protection in Mice
http://www.100md.com
感染与免疫杂志 2006年第5期
Department of Animal Health and Biomedical Sciences, University of Wisconsin—Madison, 1656 Linden Drive, Madison, Wisconsin 53706
Department of Bacteriology, Kimron Veterinary Institute, P.O. Box 12, Bet Dagan 50250, Israel
INSERM U431, UFR Medecine, CS83021, Avenue Kennedy, 30908 Nimes Cedex 02, France
ABSTRACT
In vivo bioluminescence imaging is a persuasive approach to investigate a number of issues in microbial pathogenesis. Previously, we have applied bioluminescence imaging to gain greater insight into Brucella melitensis pathogenesis. Endowing Brucella with bioluminescence allowed direct visualization of bacterial dissemination, pattern of tissue localization, and the contribution of Brucella genes to virulence. In this report, we describe the pathogenicity of three attenuated bioluminescent B. melitensis mutants, GR019 (virB4), GR024 (galE), and GR026 (BMEI1090-BMEI1091), and the dynamics of bioluminescent virulent bacterial infection following vaccination with these mutants. The virB4, galE, and BMEI1090-BMEI1091 mutants were attenuated in interferon regulatory factor 1-deficient (IRF-1–/–) mice; however, only the GR019 (virB4) mutant was attenuated in cultured macrophages. Therefore, in vivo imaging provides a comprehensive approach to identify virulence genes that are relevant to in vivo pathogenesis. Our results provide greater insights into the role of galE in virulence and also suggest that BMEI1090 and downstream genes constitute a novel set of genes involved in Brucella virulence. Survival of the vaccine strain in the host for a critical period is important for effective Brucella vaccines. The galE mutant induced no changes in liver and spleen but localized chronically in the tail and protected IRF-1–/– and wild-type mice from virulent challenge, implying that this mutant may serve as a potential vaccine candidate in future studies and that the direct visualization of Brucella may provide insight into selection of improved vaccine candidates.
INTRODUCTION
Brucella species are important zoonotic pathogens affecting a wide variety of mammals. In agriculturally important domestic animals, these bacteria cause abortion and infertility and are of serious economic concern worldwide (6). In humans, Brucella species constitute potential biowarfare agents, and the infection results in undulant fever, which if untreated, can manifest as orchitis, osteoarthritis, spondylitis, endocarditis, and neurological disorders (12, 46). Currently, no vaccine exists to protect against human brucellosis. Treatment of brucellosis requires a prolonged combination of antibiotic therapy and remains problematic because of potential relapse.
Identifying Brucella virulence factors has been of great interest in understanding Brucella pathogenesis and immune evasion. Smooth lipopolysaccharide (LPS) was the first identified virulence factor (25). Brucella LPS has minimal endotoxic effect, blocks complement activation, and protects against bactericidal cationic peptides (28). The O-chain of LPS is also important for the entry of Brucella suis into macrophages through lipid rafts, which permits the Brucella-containing vacuole (BCV) to avoid interaction with the classical endocytic pathway (32, 39). After entry into macrophages, the BCV acidifies and then transiently interacts with EEA- and LAMP1-positive vesicles. After an endosome-like stage, the BCV enters a sustained interaction with the endoplasmic reticulum, forming the replication niche (8). Maturation of the BCV into the replication niche is dependent upon the VirB type IV secretion system (T4SS) (8, 9), and therefore, the VirB system constitutes an important virulence factor for intracellular survival of Brucella spp. (10, 15, 33). Recently, cyclic -1,2-glucan has been shown as an important factor required for intracellular survival of Brucella (3). Though T4SS, cyclic -1,2-glucan, and LPS are clear virulence factors for Brucella, the attenuated mutants with these factors are either considered not safe or not sufficiently studied as possible vaccines for animals and humans. This has necessitated identification of additional vaccine targets.
Several genetic loci have been identified that are required for Brucella replication in macrophages cultured in vitro (15, 23). In vitro conditions may not adequately reflect in vivo infection, and therefore, findings may have little or no in vivo relevance (45). In vivo screening methods have been used to identify Brucella genes required for survival and persistence (18, 26); however, these studies rely on determining the numbers of tissue-specific CFU from multiple animals at different times, which is labor-intensive and requires large numbers of animals. Because infection is a dynamic process and varies within individual mice, monitoring disease progression temporally within the same mouse would provide a more comprehensive picture of pathogenic events. Further, such real-time analysis may reveal virulence determinants responsible for tissue-specific replication of bacteria that would not be revealed using conventional CFU enumeration.
Bioluminescence imaging of mice allows direct visualization of the infection process and is highly useful for bacterial pathogenesis studies (11), because the intensity of bioluminescence strongly correlates with the number of bacteria in the infected organs (17, 40). Bioluminescence imaging is useful in analyzing subacute and chronic infections that are often difficult to appreciate using conventional approaches because of uncertain bacterial locations (17, 40). Here we report the pathogenicity associated with the three attenuated bioluminescent Brucella melitensis mutants, GR019 (virB4), GR024 (galE), and GR026 (BMEI1090-BMEI1091). Attenuated bacteria could be visualized in the later stages of infection in tissues that are not conventionally evaluated, thus providing an unabridged approach to understand brucellosis affecting multiple tissues. In addition, we describe the dynamics of virulent bioluminescent B. melitensis GR023 infection following vaccination with these attenuated mutants.
MATERIALS AND METHODS
Bacterial strains, plasmids, and growth conditions. Bacterial strains and plasmids used in this study are listed in Table 1. Strains GR019 (virB4), GR024 (galE), and GR026 (BMEI1090-BMEI1091) are EZ::Tnlux transposon insertional mutants of B. melitensis 16M containing the promoterless lux operon. Rev-1 is an attenuated strain of virulent B. melitensis 6056 (2, 14) and is used as a vaccine for brucellosis in small ruminants (5). BM710 is a spontaneous stable rough mutant of Rev-1 isolated from vaccinated sheep and is phenotypically identical to Rev-1 except for the rough LPS (4). GR023 is a virulent bioluminescent B. melitensis 16M mutant used for challenge studies. GR023 was isolated by random EZ::Tnlux transposon mutagenesis and has promoterless luxCDABE genes inserted between a hypothetical protein (BMEI0100) and cysteine synthase A (BMEI0101) (40). All Brucella strains were grown in brucella broth (Difco). Ampicillin (100 μg/ml), chloramphenicol (20 μg/ml), kanamycin (50 μg/ml), and zeocin (50 μg/ml for Escherichia coli and 250 μg/ml for Brucella) were added to the medium as necessary. Brucella strains were grown at 37°C with shaking unless otherwise stated. E. coli strain DH5 (Invitrogen) and EC100Dpir+ (Epicenter, Madison, WI) were grown in LB broth (Difco).
Suicide vectors pGR026-90K and pGR026-91K for generating deletions in BMEI1090 and BMEI1091, respectively, were created using pZErO-1. To construct pGR026-90K, approximately 1-kb DNA sequences upstream and downstream of the deletion target were amplified by PCR (upstream, forward primer, 5'atcaacggtaccCGTTCAGCGCGTCGAGATCG, and reverse primer, 5'gctctaggatccGACTGATAATTATGCCGTGCG; downstream, forward primer, 5'acagtcggatccATAACCGAAGCCTATTCCTTC, and reverse primer, 5'ggtaacctgcagCGAACGTGCCCGCATCAT) and cloned into pZErO-1 to generate plasmid pGR026-90. Appropriate restriction sites were included in the PCR primers to facilitate the insertion of the kanamycin resistance gene aph(3')-la (Kanr) from pUC4K between the two fragments to generate pGR026-90K. Bases in lowercase were added to facilitate cloning, and underlined bases are restriction sites. To construct pGR026-91K, the desired deletion target was amplified with approximately 1-kb upstream and downstream sequences using specific primers (forward, 5'agatacggtaccTCTTCCATCGTTCCGGGCCT; reverse, 5'catgcatctagaGACGCCGTTGATGTTCCATGTA) and cloned into pZErO-1 to generate pGR026-91. Then, inverse PCR was performed on pGR026-91 using primers (5'tcttgagaattcCCCAATGCGACCGCTT and 5'gattcagaattcTTTGGCGATCCGCCTGGCA) designed to amplify all but the deletion target. The inverse PCR product was digested and ligated to the Kanr gene fragment to generate the final suicide vector pGR026-91K.
To construct plasmids pBBVirB4, pBBGalE, and pBBI1087-1090, DNA sequences encoding the respective open reading frames (ORFs) plus the ribosome binding site but lacking their promoter sequences were amplified using primers (for VirB4, forward primer, 5'agagagggtaccCATGTTCATATTGCCGCTGATCG, and reverse primer, 5'agagagggatccTGCTGGTTACA GTCAGGGCGAAT; for GalE, forward primer, 5'agagagggtaccAAAGCCCGGTAAAACGATTGATG, and reverse primer, 5' agagagggatccGTTCCGGCATTTTCTGGCAAA; for 1087-90, forward, 5'agagagactagtTGTGCCGTCGTTTCCACCTG, and reverse, 5'agagagctcgagAGGGACGGGGATCGGGTTAT). PCR products were digested and ligated with similarly digested pBBR-MCS4 to generate the complementation plasmids. The genes of interest were directionally cloned into pBBR1-MCS4 to ensure that these genes are being transcribed from the lac promoter present in the plasmid.
Mapping of the EZ::Tnlux transposon insertion site and Southern blot analysis. The site of transposon insertion in GR019 (virB4), GR024 (galE), and GR026 (BMEI1090-BMEI1091) was identified by rescue cloning as previously described (40). For Southern hybridization, 10 μg of genomic DNA was digested with ClaI and separated in a 0.7% agarose gel. The single-copy insertion of the transposon at the expected location was detected using the Kanr gene as a probe as previously described (40).
Inactivation of BMEI1090 and BMEI1091. To generate specific deletions, suicide vectors pGR026-90K and pGR026-91K were electroporated into B. melitensis 16M. Cells were plated on brucella agar containing kanamycin. To select for double recombinants, the Kanr colonies were checked for sensitivity to zeocin (Zeos). The resulting Kanr and Zeos clones were streak purified, and one such purified clone was used for further study.
Macrophage infection. The macrophage-like RAW 264.7 cells were cultured in RPMI supplemented with 10% heat-inactivated fetal calf serum. For macrophage growth assays, 24-well microtiter plates were seeded with 5 x 105 macrophages/well and infected with different B. melitensis strains at a multiplicity of infection of 1:50. Cells were incubated for 1 h at 37°C in 5% CO2, extracellular bacteria were removed with three washes of phosphate-buffered saline (PBS), followed by gentamicin treatment (25 μg/ml) for 30 min. Then the cells were maintained with medium containing 5 μg of gentamicin/ml. At specified times, cells were washed with PBS three times, lysed with 0.1% Triton X, and plated on brucella agar to determine intracellular bacterial counts. All experiments were performed in duplicate.
IRF-1–/– mouse virulence assay. Groups of 6- to 9-week-old interferon regulatory factor 1-deficient (IRF-1–/–) mice (n = 4) were infected intraperitoneally (i.p.) with 1 x 107 CFU of virB4, galE, BMEI1090-BMEI1091, Rev-1, and BM710 strains. Infected mice were housed in a biosafety level 3 facility and monitored for survival. For imaging, mice were anesthetized with isoflurane, and bioluminescence was recorded with a 10-min integration time using a charge-coupled device camera (Xenogen, Alameda, CA). The livers and spleens were collected aseptically from the surviving mice, homogenized in PBS, and plated on brucella agar. Plates were incubated at 37°C for 4 days, and CFU counts were determined. For histology, portions of livers and spleens were collected, fixed in 10% formalin, and 5-μm sections were stained with hematoxylin and eosin.
Vaccination and challenge studies. IRF-1–/– mice (6 to 9 weeks old) (nine mice/group) were vaccinated with 1 x 107 CFU i.p. with B. melitensis virB4, galE, BMEI1090-BMEI1091, or BM710 strain in 200 μl PBS. Strain Rev-1 was not included, because it is lethal for these mice. For a control, 10 mice were injected with 200 μl PBS. C57BL/6 mice (20 mice/group) were vaccinated i.p. with 5 x 107 CFU with each of the above strains and with Rev-1. Mice were imaged daily using a charge-coupled device camera until challenge. After 60 days, both IRF-1–/– and C57BL/6 mice were challenged with 1 x 106 CFU of virulent bioluminescent B. melitensis GR023 i.p. Following challenge, mice were imaged with 10 min of integration, and dissemination of bioluminescent GR023 was monitored.
For IRF-1–/– mice, the survival was recorded in different groups following virulent challenge. At 44 days postchallenge, livers and spleens from surviving mice were processed for CFU enumeration. For C57BL/6 mice, four mice from each group were killed at weekly intervals to determine CFU in livers and spleens. Portions of the livers and spleens were weighed, homogenates were serially diluted in PBS and plated on brucella agar with or without antibiotic, and colonies were counted after 72 h of incubation at 37°C. CFU were determined per gram of each tissue. To determine the histological changes at each time, portions of livers and spleens were collected and processed as described above.
Statistical analyses. All statistical analyses were performed using Minitab 13.31 statistical software (Minitab Inc., State College, PA). Association between each group at different times was evaluated using one-way analysis of variance. For all statistical analyses, P values of <0.05 were considered significant.
RESULTS
Pathogenicity of the three attenuated bioluminescent B. melitensis mutants. Previously we described identification and dissemination of three attenuated bioluminescent B. melitensis mutants in mice using an in vivo imaging system (40). In order to determine virulence and pathology associated with these strains, we tested these bioluminescent mutants in IRF-1–/– mice. In addition, we also tested two other B. melitensis mutants, BM710, a stable rough strain and Rev-1, a vaccine strain, so that the bioluminescent mutants could be evaluated for their ability to confer protection against virulent challenge. IRF-1–/– mice (n = 4) infected with these strains were monitored for bacterial dissemination, persistence, and pathology induced in liver and spleen. Mice infected with all three bioluminescent strains (GR019 [virB4], GR024 [galE], and GR026 [BMEI1090-BMEI1091]) and strain BM710 appeared healthy and survived longer than 24 days, suggesting attenuation. However, all Rev-1-infected mice died by 7 days postinfection (p.i.). Although Rev-1 strain is a commercial vaccine strain, it was fully virulent in these mice. To determine the pathology associated with different attenuated mutants, the livers and spleens were processed for CFU and histopathology. Livers and spleens from GR019 (virB4)- or BM710-infected mice had lower CFU than GR024 (galE)- or GR026 (BMEI1090-BMEI1091)-infected mice (Table 2). However, except for the GR026 (BMEI1090-BMEI1091)-infected group, livers and spleens from other groups had no observable histological changes (Table 2 and data not shown). The GR026 (BMEI1090-BMEI1091)-infected mice displayed very few multifocal granulomas in livers and minor changes in the white pulp of spleens.
virB4 but not the galE or BMEI1090-BMEI1091 mutant is attenuated in RAW 264.7 macrophages. To test whether the bioluminescent mutants were attenuated in macrophages, we examined their growth in RAW 264.7 macrophage-like cells. All three strains grew at rates similar to that of strain 16M, with a duplication time of 2 h in brucella broth suggesting no general growth defect (Fig. 1A). RAW 264.7 macrophages were infected with each strain at a multiplicity of infection of 1:50, and the intracellular growth was monitored for 72 h. Interestingly, only strain GR019 (virB4) was defective in replication with a significant decrease in intracellular bacteria by 24 h p.i. compared to 16M (Fig. 1B). On the other hand, both GR024 (galE) and GR026 (BMEI1090-BMEI1091) displayed growth curves distinct from those of GR019 (virB4) and 16M. Both strains were phagocytosed more with no apparent intramacrophage replication during 24 h p.i. as bacterial levels remained constant. However, by 48 h their growth was similar to 16M (Fig. 1). The growth curves of GR024 (galE) and GR026 (BMEI1090-BMEI1091) mutants appeared as intermediate between smooth (e.g., 16M) and rough (e.g., perA) strains of Brucella (41). Rough strains of Brucella are phagocytosed more efficiently than smooth strains and persist intracellularly at higher levels for more than 3 days even though they are defective for intramacrophage replication (20, 34). This led us to suspect that both GR024 (galE) and GR026 (BMEI1090-BMEI1091) may have an LPS defect; this suspicion was supported by the fact that these strains agglutinated in the presence of acriflavin (7). As suggested by the macrophage growth pattern, both strains resulted in fine amorphous agglutination particles that were less intense compared to rough strains of Brucella (i.e., RB51, BM710, and GI-2 deletion mutant; data not shown).
Molecular analysis of bioluminescent mutants. EZ::Tnlux insertion was localized to virB4 (BMEII0028) for strain GR019, galE homolog (BMEI0921) for GR024, and in the intergenic region of two divergent ORFs (BMEI1090 and BMEI1091) for GR026 (Fig. 2). The Brucella strains with mutations in the virB genes encoding T4SS are attenuated in macrophages as well as in mice (10, 15, 33). Similarly, in GR024, the insertion disrupted the galE homolog that has been previously shown to attenuate Brucella (35). Analysis of the genomic context of insertion in GR026 suggested that BMEI1090 is the first gene in a cluster of genes that are transcribed in reverse orientation, whereas BMEI1091 is an independent transcriptional unit (Fig. 2). Further, Southern blot analyses confirmed the sequencing results and also revealed the single-copy insertion of the transposon in these strains (data not shown).
To determine the gene(s) likely responsible for the observed phenotype of BMEI1090-BMEI1091 mutant, we generated BMEI1090 and BMEI1091 deletion mutants by allelic replacement. The respective ORFs were replaced with a Kanr marker by homologous recombination, and the resulting strains, GR-BMEI1090 and GR-BMEI1091, were tested for virulence in IRF-1–/– mice. IRF-1–/– mice infected with GR-BMEI1091 died within 10 days similar to virulent strain16M; however, only two mice infected with GR-BMEI1090 died, and the remaining mice survived for at least 21 days (Table 2). The livers and spleens from the surviving mice had average CFU of 6.65E + 04 and 1.14E + 06, respectively. Therefore, inactivation of BMEI1090 resulted in partial attenuation, suggesting that the phenotype associated with BMEI1090-BMEI1091 is likely due to altered expression of BMEI1090 and downstream genes.
To confirm that the attenuation of bioluminescent mutants is due to disruption of transposon insertion targets and not due to secondary mutations, we complemented virB4, galE, and BMEI1090-BMEI1091 mutations with corresponding ORFs. Since strain GR019 (virB4) has a growth defect in RAW 264.7 macrophages, GR019 (virB4) containing either pBBVirB4 or pBBVirB was tested for growth in these macrophages. Addition of pBBVirB4 in strain GR019 (virB4) resulted in partial restoration of growth, as suggested by an increase in intracellular bacteria at 24 h p.i. (Fig. 3A). However, addition of pBBVirB, containing the entire virB operon (33), into GR019 (virB4) resulted in complete restoration of growth (Fig. 3). In addition, virB4 complemented with pBBVirB, but not pBBVirB4, killed IRF-1–/– mice and restored complete virulence (Fig. 3B). Consistent with the in vitro results, mice infected with GR019 (virB4)/pBBVirB4 did not die and contained more bacteria in livers and spleens than mice infected with GR019 (virB4) (Fig. 3C).
Since both strains GR024 (galE) and GR026 (BMEI1090-BMEI1091) agglutinated in the presence of acriflavin, we tested the galE- and BMEI1090-BMEI1091-complemented strains for agglutination. GR024 (galE) complemented with pBBGalE resulted in no agglutination in the presence of acriflavin (data not shown). Since our earlier results suggested that the attenuation of GR026 (BMEI1090-BMEI1091) is likely due to the altered expression of BMEI1090 and downstream genes, we complemented GR026 with a plasmid containing four ORFs likely to form an operon (13). Surprisingly, addition of pBBI1087-1090 to BMEI1090-BMEI1091 resulted in much pronounced agglutination as seen with rough strains of Brucella (data not shown). Consistent with the acriflavin agglutination results, both GR024 (galE) and GR026 (BMEI1090-BMEI1091) were partially resistant to smooth-type-specific Tbilisi (Tb) phage, and the addition of pBBGalE restored the susceptibility of GR024 (galE) to Tb phage. However, GR026 (BMEI1090-BMEI1091) complemented with pBBI1087-1090 was completely resistant to Tb phage, suggesting a rough phenotype of the complemented strain (data not shown).
The galE and BMEI1090-BMEI1091 mutants protect IRF-1–/– mice from virulent challenge. Although IRF-1–/– mice are immunocompromised, they are protected against virulent Brucella following vaccination with attenuated strains (21). Therefore, we tested the attenuated bioluminescent mutants to determine if they protect IRF-1–/– mice from virulent challenge. IRF-1–/– mice vaccinated with attenuated bioluminescent mutants were challenged when no bioluminescent bacteria were detectable. To evaluate vaccine candidates, virulent strain GR023 was used to visualize dissemination and tissue localization of virulent Brucella by temporally imaging individual mice. All mice vaccinated with either strain GR024 (galE) or GR026 (BMEI1090-BMEI1091) survived at least 44 days, whereas only two mice vaccinated with GR019 (virB4) and three mice vaccinated with BM710 survived for 44 days following challenge (Fig. 4). Fifty percent of GR019 (virB4)-vaccinated mice died by day 12, whereas 50% of the BM710-vaccinated mice died by day 9 following challenge. As expected, all unvaccinated mice died within 2 weeks following challenge, with 50% mice dead after 7 days (Fig. 4).
The livers and spleens from surviving mice vaccinated with different strains had very similar CFU (CFU ranges, 2.2E + 02 to 1.2E + 03/gram of tissue in the liver and 1.5E + 04 to 3.4E + 04/gram of tissue in the spleen). Bacteria recovered from the livers and spleens of mice vaccinated with bioluminescent strains were confirmed as strain GR023 by verifying the EZ::Tnlux insertion site using PCR (data not shown). Bioluminescence imaging of vaccinated mice following i.p. challenge revealed strikingly different dynamics of persistence and spread of virulent bacteria. Unlike the unvaccinated mice, in all vaccinated groups, bacterial spread was less extensive, but correlated with the ability of the vaccine strain to protect from challenge (Fig. 5). In both BM710- and GR019 (virB4)-vaccinated groups, bioluminescence was pronounced with systemic spread by day 10; however, in both GR024 (galE)- and GR026 (BMEI1090-BMEI1091)-vaccinated groups, bioluminescence was observed at the injection site and in the tail (Fig. 5). By day 44, GR024 (galE)- and GR026 (BMEI1090-BMEI1091)-vaccinated mice had no detectable bioluminescent bacteria, while BM710- and GR019 (virB4)-vaccinated survivors still exhibited bioluminescence (Fig. 5B). Consistent with survival data, the GR024 (galE)- and GR026 (BMEI1090-BMEI1091)-vaccinated IRF-1–/– mice had the least histological changes in livers and spleens (Fig. 5). The GR024 (galE)- and GR026 (BMEI1090-BMEI1091)-vaccinated mice had few focal granulomas (less than three per field of view at a magnification of x4) in liver sections, while the spleens of GR024 (galE)-vaccinated mice appeared normal and with minimal disorganization of the white pulp in GR026 (BMEI1090-BMEI1091)-vaccinated mice. Both GR019 (virB4)- and BM710-vaccinated survivors had more histological changes in livers and spleens than GR024 (galE)- and GR026 (BMEI1090-BMEI1091)-vaccinated groups (Fig. 5B).
IRF-1–/– mice are defective in multiple aspects of the immune system (44). To better correlate the immune protection provided by the different attenuated strains, we tested these mutants in wild-type C57BL/6 mice, the parental strain of IRF-1–/– mice. C57BL/6 mice clear virulent Brucella infection naturally and serve as a relevant model to study Brucella pathogenesis and immune protection. To assess the protection by different attenuated strains, we monitored bacterial clearance and histological changes in livers and spleens. In addition, the dynamics of infection by attenuated bioluminescent strains and their effects on virulent challenge were monitored by imaging. Similar to IRF-1–/– mice, GR019 (virB4)-vaccinated C57BL/ 6 mice had bioluminescence in livers and spleens by day 1 p.i.; however, in GR024 (galE)- or GR026 (BMEI1090-BMEI1091)-vaccinated mice, bioluminescence was detected primarily at the injection site (Fig. 6A). Bioluminescence began to diminish by day 5 in all groups, and by 2 weeks p.i. minimal or no bioluminescence was observed (Fig. 6A). However, after challenge the dynamics of virulent Brucella dissemination was similar in all vaccinated groups being limited primarily to the injection site, though bioluminescence was stronger in GR019 (virB4)- and BM710-vaccinated groups (Fig. 6B). Consistent with imaging data, all vaccinated groups had significantly fewer CFU (P < 0.05) in livers and spleens 1 week postchallenge compared to the unvaccinated group. In addition, GR024 (galE)-, GR026 (BMEI1090-BMEI1091)-, and Rev-1-vaccinated groups contained even lower numbers of CFU (P < 0.05) than GR019 (virB4)- or BM710-vaccinated group did (Fig. 7). Though the spleens from GR024 (galE)- and Rev-1-vaccinated mice had lower CFU at all times than the spleens from other groups (Fig. 7), CFU counts from spleens at 3 and/or 4 weeks postchallenge in GR024 (galE)- and Rev-1-vaccinated groups were significantly lower (P < 0.05). To correlate bacterial clearance with tissue damage, histological changes were assessed in livers and spleens from immunized mice following challenge. Consistent with bacterial clearance, GR024 (galE)- and GR026 (BMEI1090-BMEI1091)-vaccinated mice exhibited fewer granulomas in livers; however, livers from GR019 (virB4) and BM710-vaccinated mice contained more granulomas (Table 3). Surprisingly, only the livers from Rev-1-vaccinated mice had large grossly visible focal calcified granulomas (Fig. 8). On the other hand, histological changes in spleens were similar in all vaccinated groups but contained fewer changes compared to unvaccinated mice (Table 3).
DISCUSSION
In this report, we describe infection dynamics of three attenuated bioluminescent mutants and the effect of vaccination on the dynamics of virulent bacterial infection in mice. Strains GR019 (virB4), GR024 (galE), GR026 (BMEI1090-BMEI1091), and BM710 were all attenuated in IRF-1–/– mice; however, Rev-1 remained virulent in these mice. As previously reported, all three attenuated bioluminescent strains revealed striking differences in bacterial dissemination and persistence. GR019 (virB4), unlike GR024 (galE) or GR026 (BMEI1090-BMEI1091), spread systemically, and bioluminescence was observed in liver, spleen, testes, submandibular region, and extremities early in infection, confirming previous reports that the VirB system is not required for establishing early infection (18). However, the VirB system is required for Brucella persistence, as C57BL/6 mice cleared GR019 (virB4) infection faster than virulent Brucella did (data not shown), which implies that the presence of Brucella T4SS may provide a selective advantage for the bacterium against host defense during later stages of infection. The GR024 (galE) and GR026 (BMEI1090-BMEI1091) strains, on the other hand, were defective in systemic spread, with bioluminescence localized primarily to the injection site (40). Both GR024 (galE) and GR026 (BMEI1090-BMEI1091) have altered surface structure, likely in the LPS component, and therefore are unable to replicate efficiently and spread. Brucella strains defective in LPS structure are more susceptible to innate bactericidal effects of phagocytes and are defective in intracellular survival (25). Interestingly, in both GR024 (galE)- and GR026 (BMEI1090-BMEI1091)-infected mice, bioluminescence reappeared 12 days p.i. and localized in the joint-rich tail region during the later stages of infection (40; data not shown). Spinal involvement in brucellosis is more destructive and occurs in 20 to 65% of all patients with musculoskeletal brucellosis (27). Our results parallel those of human brucellosis and may provide a model to understand chronic localization of Brucella. These bioluminescent joints contained up to 104 bacteria. Though these strains localized to tail joints, they induced no histological changes in these locations as well as in livers and spleens. However, whether the recovered bacteria have altered LPS and whether that LPS plays a role in chronic localization of Brucella needs to be determined. Importantly, bioluminescence imaging revealed localization of LPS-defective Brucella in tissues that have not been previously examined by others, suggesting a requirement for a comprehensive examination of infected mice to assess bacterial clearance. In addition, temporal analysis of infection revealed patterns of growth and clearance as well as the reemergence of bacteria that is extremely difficult to observe with conventional methods. Thus, our study clearly demonstrates that bioluminescence imaging in conjunction with CFU enumeration provides better assessment of in vivo clearance of Brucella affecting multiple tissues. Importantly, only GR019 (virB4) was attenuated in RAW 264.7 macrophages (Fig. 1B). Therefore, imaging may provide a comprehensive approach to identify Brucella genes that are relevant to in vivo pathogenesis.
Both galE and BMEI1090-BMEI1091 mutants exhibited growth patterns in macrophages intermediate to smooth and rough strains of Brucella (41), suggesting that they may have an altered surface structure. Both strains produced very fine agglutination in the presence of acriflavin and were partially resistant to smooth-type-specific Tb phage (30). In strain GR024 (galE), the transposon insertion is in ORF BMEI0921, a NAD-dependent epimerase/dehydratase family member that is closely related to enterobacterial galE. In many gram-negative bacteria, galactose is converted to UDP-galactose and the galE gene product, UDP-galactose 4-epimerase, catalyzes reversible conversion of UDP-galactose to UDP-glucose. UDP-galactose serves as donor for both LPS core and O-antigen polysaccharide biosynthesis. Therefore, galE mutants are defective in LPS, and thus, the galE gene is an important virulence factor. The acriflavin agglutination and phage susceptibility tests suggest a defect in the GR024 (galE) LPS; however, GR024 (galE) was not sensitive to growth in galactose-containing medium (1, 37; data not shown). The galE mutants fed excess galactose accumulate UDP-galactose, which is toxic to the cell (1). The galE mutants of other bacteria display contrasting responses to galactose, with some being sensitive while others were not sensitive to galactose (16, 19, 29, 36). Strikingly, the galE mutants of Brucella abortus and B. melitensis have different responses to growth in galactose (35, 43). The B. melitensis galE mutant (Bm92) is not sensitive to galactose; however, a plasmid that encodes the B. melitensis galE ORF complemented a galE mutation in Salmonella enterica serovar Typhimurium LB5010, as shown by the restoration of smooth LPS, sensitivity to phage P22 infection, and restoration of UDP galactose-4-epimerase activity, suggesting a role for GalE in LPS biogenesis (35). Interestingly, LPS of B. melitensis mutant Bm92 has been reported to have no major differences compared to LPS from strain 16M (35). This implies that minor differences that were not readily discernible by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and Western blot analysis may exist, since extraction of smooth LPS from Brucella is difficult due to its peculiar composition (25). Analysis of the galE homolog of Yersinia enterocolitica indicates that although it can complement an E. coli galE mutant, its primary function in Y. enterocolitica is not in the production of UDP-galactose but, instead, some other nucleotide sugar required for LPS biosynthesis (36). However, without fine structural analysis of LPS, it is not known whether Y. enterocolitica LPS has galactose or whether galE catalyzes the formation of a closely related sugar in the LPS structure. Similarly, in Brucella GalE may modify a closely related sugar, but a detailed chemical analysis of the LPS composition will be required to identify the role of GalE in LPS biogenesis. Consistent with the above findings, our data also indicate minor changes in the GR024 (galE) LPS, as it was partially resistant to Tb phage and produced very fine agglutination particles instead of large particulate agglutination seen with rough mutants of Brucella (data not shown).
The GR026 strain has an insertion in the intergenic region of BMEI1090 and BMEI1091. Further, selective allelic replacement of these ORFs suggested that BMEI1090 and downstream genes are responsible for the attenuation of GR026 (BMEI1090-BMEI1091) (Table 1). Interrogation of the B. melitensis genome suggested that BMEI1090 and downstream genes (BMEI1087 to BMEI1090) likely form an operon. BMEI1087 encodes -hexosaminidase A, while BMEI1088 encodes soluble lytic murein transglycosylase, and these are involved in amino sugar metabolism and N-glycan biosynthesis (http://www.genome.ad.jp/kegg). Therefore, this operon may contribute to cell membrane biogenesis. Consistent with this observation, the acriflavin agglutination and Tb phage susceptibility tests suggested that GR026 (BMEI1090-BMEI1091) has a surface structure defect. Complementation of BMEI1090-BMEI1091 with a plasmid containing BMEI1087-BMEI1090 ORFs resulted in more pronounced agglutination and complete resistance to Tb phage, which suggests that the expression of these genes is under strict regulation.
Vaccination with both GR024 (galE) and GR026 (BMEI1090-BMEI1091) strains protected IRF-1–/– mice from virulent B. melitensis challenge, whereas strains GR019 (virB4) and BM710 failed to protect these mice. In addition, GR024 (galE)- and GR026 (BMEI1090-BMEI1091)-vaccinated mice displayed minimal changes in livers and spleens, and no bioluminescence was observed at 44 days postchallenge. IRF-1–/– mice are defective in multiple immune components, with reduced numbers of CD8+ T cells, functionally impaired natural killer cells, and dysregulation of interleukin-12 p40 and inducible nitric oxide synthase (44). Though these mice are severely immunocompromised, they mount an adaptive immune response sufficient to protect against virulent challenge, and the protection is vaccine strain dependent. Unlike GR019 (virB4), both GR024 (galE) and GR026 (BMEI1090-BMEI1091) produced a localized but persistent infection in these mice (40; data not shown) and induced a protective immune response against virulent Brucella. This result is consistent with Plommet's observation that survival of the vaccine strain in the host for a critical period determines the efficacy of Brucella vaccines (38). Similar results have been observed with two field vaccine strains, S19 and RB51 in that S19 has been shown to persist longer and is more protective than RB51 in mice and other models (22, 42). However, S19 still possesses residual virulence in domestic animals and in IRF-1–/– mice (21, 31), whereas RB51 is highly attenuated (21). The galE and BMEI1090-BMEI1091 mutants are highly attenuated in IRF-1–/– mice, similar to RB51, but cause no or very minimal pathological changes in livers and spleen and are protective. Consistent with the IRF-1–/– mouse data, both GR024 (galE) and GR026 (BMEI1090-BMEI1091) provided greater protection to C57BL/6 mice than GR019 (virB4) or BM710 did, which suggests that IRF-1–/– mice may serve as an important model to rapidly assess vaccine efficacy of Brucella strains. Interestingly, Rev-1-vaccinated mice had fewer CFU in both livers and spleens than GR024 (galE)- and GR026 (BMEI1090-BMEI1091)-vaccinated mice; however, Rev-1-vaccinated mice displayed severe liver damage with grossly visible lesions (Fig. 8) that was not observed in other groups. These lesions are likely vaccine induced, as they were apparent even at 1 week postchallenge. Rev-1 vaccine is used in domestic animals with various degrees of success in areas where B. melitensis is endemic (5). Although Rev-1 protected wild-type mice, Rev-1 was highly virulent to IRF-1–/– mice (Table 2) and caused severe liver damage in wild-type mice. This is in line with the fact that Rev-1 strain can still cause clinical brucellosis (38). In summary, our study revealed the contribution of Brucella genes to in vivo pathogenesis and identified a new set of virulence genes (BMEI1090 and downstream genes). Further, the GalE-deficient GR024 strain has altered LPS, results in no detectable tissue damage, and protects against virulent B. melitensis challenge, making it an interesting vaccine candidate for brucellosis.
ACKNOWLEDGMENTS
We thank David Warshauer and Tim Monson at the Wisconsin State Laboratory of Hygiene for help with phage typing of Brucella strains. We also thank Tajie Harris and Michael Krepps for assistance with mapping bioluminescent mutants and macrophage infection and Fue Vang for statistical assistance.
This work was supported by the NIH grant R01AI048490, NIH/NIAID RCE for Biodefense and Emerging Infectious Diseases Research Program grant 1-U54-AI-057153, and USDA grant 35204-14856.
REFERENCES
1. Adhya, S. 1987. The galactose operon, p. 1503-1512. In F. C. Neidhardt, K. B. Low, B. Magasanik, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella typhimurium: cellular and molecular biology. American Society for Microbiology, Washington, D.C.
2. Alton, G. G., S. S. Elberg, and D. Crouch. 1967. Rev. 1 Brucella melitensis vaccine. The stability of the degree of attenuation. J. Comp. Pathol. 77:293-300.
3. Arellano-Reynoso, B., N. Lapaque, S. Salcedo, G. Briones, A. E. Ciocchini, R. Ugalde, E. Moreno, I. Moriyon, and J. P. Gorvel. 2005. Cyclic beta-1,2-glucan is a Brucella virulence factor required for intracellular survival. Nat. Immunol. 6:618-625.
4. Bardenstein, S., M. Mandelboim, T. A. Ficht, M. Baum, and M. Banai. 2002. Identification of the Brucella melitensis vaccine strain Rev.1 in animals and humans in Israel by PCR analysis of the PstI site polymorphism of its omp2 gene. J. Clin. Microbiol. 40:1475-1480.
5. Blasco, J. M. 1997. A review of the use of B. melitensis Rev 1 vaccine in adult sheep and goats. Prev. Vet. Med. 31:275-283.
6. Boschiroli, M. L., V. Foulongne, and D. O'Callaghan. 2001. Brucellosis: a worldwide zoonosis. Curr. Opin. Microbiol. 4:58-64.
7. Braun, W., and A. E. Bonestell. 1947. Independent variation of characteristics in Brucella abortus variants and their detection. Am. J. Vet. Res. 8:386-390.
8. Celli, J., C. de Chastellier, D. M. Franchini, J. Pizarro-Cerda, E. Moreno, and J. P. Gorvel. 2003. Brucella evades macrophage killing via VirB-dependent sustained interactions with the endoplasmic reticulum. J. Exp. Med. 198:545-556.
9. Celli, J., and J. P. Gorvel. 2004. Organelle robbery: Brucella interactions with the endoplasmic reticulum. Curr. Opin. Microbiol. 7:93-97.
10. Comerci, D. J., M. J. Martinez-Lorenzo, R. Sieira, J. P. Gorvel, and R. A. Ugalde. 2001. Essential role of the VirB machinery in the maturation of the Brucella abortus-containing vacuole. Cell. Microbiol. 3:159-168.
11. Contag, C. H., P. R. Contag, J. I. Mullins, S. D. Spilman, D. K. Stevenson, and D. A. Benaron. 1995. Photonic detection of bacterial pathogens in living hosts. Mol. Microbiol. 18:593-603.
12. Corbel, M. J. 1997. Brucellosis: an overview. Emerg. Infect. Dis. 3:213-221.
13. DelVecchio, V. G., V. Kapatral, R. J. Redkar, G. Patra, C. Mujer, T. Los, N. Ivanova, I. Anderson, A. Bhattacharyya, A. Lykidis, G. Reznik, L. Jablonski, N. Larsen, M. D'Souza, A. Bernal, M. Mazur, E. Goltsman, E. Selkov, P. H. Elzer, S. Hagius, D. O'Callaghan, J. J. Letesson, R. Haselkorn, N. Kyrpides, and R. Overbeek. 2002. The genome sequence of the facultative intracellular pathogen Brucella melitensis. Proc. Natl. Acad. Sci. USA 99:443-448.
14. Elberg, S. S., and K. Faunce, Jr. 1957. Immunization against Brucella infection. VI. Immunity conferred on goats by a nondependent mutant from a streptomycin-dependent mutant strain of Brucella melitensis. J. Bacteriol. 73:211-217.
15. Foulongne, V., G. Bourg, C. Cazevieille, S. Michaux-Charachon, and D. O'Callaghan. 2000. Identification of Brucella suis genes affecting intracellular survival in an in vitro human macrophage infection model by signature-tagged transposon mutagenesis. Infect. Immun. 68:1297-1303.
16. Fry, B. N., S. Feng, Y. Y. Chen, D. G. Newell, P. J. Coloe, and V. Korolik. 2000. The galE gene of Campylobacter jejuni is involved in lipopolysaccharide synthesis and virulence. Infect. Immun. 68:2594-2601.
17. Hardy, J., K. P. Francis, M. DeBoer, P. Chu, K. Gibbs, and C. H. Contag. 2004. Extracellular replication of Listeria monocytogenes in the murine gall bladder. Science 303:851-853.
18. 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.
19. Houng, H. S., D. J. Kopecko, and L. S. Baron. 1990. Molecular cloning and physical and functional characterization of the Salmonella typhimurium and Salmonella typhi galactose utilization operons. J. Bacteriol. 172:4392-4398.
20. Jimenez de Bagues, M. P., A. Gross, A. Terraza, and J. Dornand. 2005. Regulation of the mitogen-activated protein kinases by Brucella spp. expressing a smooth and rough phenotype: relationship to pathogen invasiveness. Infect. Immun. 73:3178-3183.
21. 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.
22. Ko, J., and G. A. Splitter. 2003. Molecular host-pathogen interaction in brucellosis: current understanding and future approaches to vaccine development for mice and humans. Clin. Microbiol. Rev. 16:65-78.
23. Kohler, S., V. Foulongne, S. Ouahrani-Bettache, G. Bourg, J. Teyssier, M. Ramuz, and J. P. Liautard. 2002. The analysis of the intramacrophagic virulome of Brucella suis deciphers the environment encountered by the pathogen inside the macrophage host cell. Proc. Natl. Acad. Sci. USA 99:15711-15716.
24. Kovach, M. E., P. H. Elzer, D. S. Hill, G. T. Robertson, M. A. Farris, R. M. Roop II, and K. M. Peterson. 1995. Four new derivatives of the broad-host-range cloning vector pBBR1MCS, carrying different antibiotic-resistance cassettes. Gene 166:175-176.
25. Lapaque, N., I. Moriyon, E. Moreno, and J. P. Gorvel. 2005. Brucella lipopolysaccharide acts as a virulence factor. Curr. Opin. Microbiol. 8:60-66.
26. Lestrate, P., R. M. Delrue, I. Danese, C. Didembourg, B. Taminiau, P. Mertens, X. De Bolle, A. Tibor, C. M. Tang, and J. J. Letesson. 2000. Identification and characterization of in vivo attenuated mutants of Brucella melitensis. Mol. Microbiol. 38:543-551.
27. Madkour, M. M. 2001. Osteoarticular brucellosis, p. 74-84. In M. M. Makour (ed.), Madkour's brucellosis. Springer, Heidelberg, Germany.
28. Martinez de Tejada, G., J. Pizarro-Cerda, E. Moreno, and I. Moriyon. 1995. The outer membranes of Brucella spp. are resistant to bactericidal cationic peptides. Infect. Immun. 63:3054-3061.
29. Metzger, M., P. Bellemann, P. Bugert, and K. Geider. 1994. Genetics of galactose metabolism of Erwinia amylovora and its influence on polysaccharide synthesis and virulence of the fire blight pathogen. J. Bacteriol. 176:450-459.
30. Monreal, D., M. J. Grillo, D. Gonzalez, C. M. Marin, M. J. De Miguel, I. Lopez-Goni, J. M. Blasco, A. Cloeckert, and I. Moriyon. 2003. Characterization of Brucella abortus O-polysaccharide and core lipopolysaccharide mutants and demonstration that a complete core is required for rough vaccines to be efficient against Brucella abortus and Brucella ovis in the mouse model. Infect. Immun. 71:3261-3271.
31. Moriyon, I., M. J. Grillo, D. Monreal, D. González, C. Marín, I. Lopez-Goi, R. C. Mainar-Jaime, E. Moreno, and J. M. Blasco. 2004. Rough vaccines in animal brucellosis: structural and genetic basis and present status. Vet. Res. 35:1-38.
32. Naroeni, A., and F. Porte. 2002. Role of cholesterol and the ganglioside GM1 in entry and short-term survival of Brucella suis in murine macrophages. Infect. Immun. 70:1640-1644.
33. O'Callaghan, D., C. Cazevieille, A. Allardet-Servent, M. L. Boschiroli, G. Bourg, V. Foulongne, P. Frutos, Y. Kulakov, and M. Ramuz. 1999. Homologue of the Agrobacterium tumefaciens VirB and Bordetella pertussis Ptl type IV secretion systems is essential for intracellular survival of Brucella suis. Mol. Microbiol. 33:1210-1220.
34. Pei, J., and T. A. Ficht. 2004. Brucella abortus rough mutants are cytopathic for macrophages in culture. Infect. Immun. 72:440-450.
35. Petrovska, L., R. G. Hewinson, G. Dougan, D. J. Maskell, and M. J. Woodward. 1999. Brucella melitensis 16M: characterisation of the galE gene and mouse immunisation studies with a galE deficient mutant. Vet. Microbiol. 65:21-36.
36. Pierson, D. E., and S. Carlson. 1996. Identification of the galE gene and a galE homolog and characterization of their roles in the biosynthesis of lipopolysaccharide in a serotype O:8. strain of Yersinia enterocolitica. J. Bacteriol. 178:5916-5924.
37. Plommet, M. 1991. Minimal requirements for growth of Brucella suis and other Brucella species. Zentbl. Bakteriol. 275:436-450.
38. Plommet, M. 1995. Live vaccines: virulence, immunogenicity/protection and safety. Historical, theoretical and practical considerations applied to the Brucella melitensis Rev. 1 vaccine, p. 7-18. In B. Garin-Bastugi and A. Benkirane (ed.), FAO/WHO/OIE Round Table on the Use of Rev. 1 Vaccine in Small Ruminants and Cattle, 21 to 22 September 1995. Centre National d'Etudes Veterinaires et Alimentaires, Maison-Alfort, France.
39. Porte, F., A. Naroeni, S. Ouahrani-Bettache, and J. P. Liautard. 2003. Role of the Brucella suis lipopolysaccharide O antigen in phagosomal genesis and in inhibition of phagosome-lysosome fusion in murine macrophages. Infect. Immun. 71:1481-1490.
40. 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. 10:1459-1473.
41. Rittig, M. G., A. Kaufmann, A. Robins, B. Shaw, H. Sprenger, D. Gemsa, V. Foulongne, B. Rouot, and J. Dornand. 2003. Smooth and rough lipopolysaccharide phenotypes of Brucella induce different intracellular trafficking and cytokine/chemokine release in human monocytes. J. Leukoc. Biol. 74:1045-1055.
42. Schurig, G. G., N. Sriranganathan, and M. J. Corbel. 2002. Brucellosis vaccines: past, present and future. Vet. Microbiol. 90:479-496.
43. Scupham, A. J., and E. W. Triplett. 1997. Isolation and characterization of the UDP-glucose 4'-epimerase-encoding gene, galE, from Brucella abortus 2308. Gene 202:53-59.
44. Taniguchi, T., K. Ogasawara, A. Takaoka, and N. Tanaka. 2001. IRF family of transcription factors as regulators of host defense. Annu. Rev. Immunol. 19:623-655.
45. Unsworth, K. E., and D. W. Holden. 2000. Identification and analysis of bacterial virulence genes in vivo. Philos. Trans. R. Soc. Lond. B 355:613-622.
46. Young, E. J. 1995. An overview of human brucellosis. Clin. Infect. Dis. 21:283-289.(Gireesh Rajashekara, Davi)
Department of Bacteriology, Kimron Veterinary Institute, P.O. Box 12, Bet Dagan 50250, Israel
INSERM U431, UFR Medecine, CS83021, Avenue Kennedy, 30908 Nimes Cedex 02, France
ABSTRACT
In vivo bioluminescence imaging is a persuasive approach to investigate a number of issues in microbial pathogenesis. Previously, we have applied bioluminescence imaging to gain greater insight into Brucella melitensis pathogenesis. Endowing Brucella with bioluminescence allowed direct visualization of bacterial dissemination, pattern of tissue localization, and the contribution of Brucella genes to virulence. In this report, we describe the pathogenicity of three attenuated bioluminescent B. melitensis mutants, GR019 (virB4), GR024 (galE), and GR026 (BMEI1090-BMEI1091), and the dynamics of bioluminescent virulent bacterial infection following vaccination with these mutants. The virB4, galE, and BMEI1090-BMEI1091 mutants were attenuated in interferon regulatory factor 1-deficient (IRF-1–/–) mice; however, only the GR019 (virB4) mutant was attenuated in cultured macrophages. Therefore, in vivo imaging provides a comprehensive approach to identify virulence genes that are relevant to in vivo pathogenesis. Our results provide greater insights into the role of galE in virulence and also suggest that BMEI1090 and downstream genes constitute a novel set of genes involved in Brucella virulence. Survival of the vaccine strain in the host for a critical period is important for effective Brucella vaccines. The galE mutant induced no changes in liver and spleen but localized chronically in the tail and protected IRF-1–/– and wild-type mice from virulent challenge, implying that this mutant may serve as a potential vaccine candidate in future studies and that the direct visualization of Brucella may provide insight into selection of improved vaccine candidates.
INTRODUCTION
Brucella species are important zoonotic pathogens affecting a wide variety of mammals. In agriculturally important domestic animals, these bacteria cause abortion and infertility and are of serious economic concern worldwide (6). In humans, Brucella species constitute potential biowarfare agents, and the infection results in undulant fever, which if untreated, can manifest as orchitis, osteoarthritis, spondylitis, endocarditis, and neurological disorders (12, 46). Currently, no vaccine exists to protect against human brucellosis. Treatment of brucellosis requires a prolonged combination of antibiotic therapy and remains problematic because of potential relapse.
Identifying Brucella virulence factors has been of great interest in understanding Brucella pathogenesis and immune evasion. Smooth lipopolysaccharide (LPS) was the first identified virulence factor (25). Brucella LPS has minimal endotoxic effect, blocks complement activation, and protects against bactericidal cationic peptides (28). The O-chain of LPS is also important for the entry of Brucella suis into macrophages through lipid rafts, which permits the Brucella-containing vacuole (BCV) to avoid interaction with the classical endocytic pathway (32, 39). After entry into macrophages, the BCV acidifies and then transiently interacts with EEA- and LAMP1-positive vesicles. After an endosome-like stage, the BCV enters a sustained interaction with the endoplasmic reticulum, forming the replication niche (8). Maturation of the BCV into the replication niche is dependent upon the VirB type IV secretion system (T4SS) (8, 9), and therefore, the VirB system constitutes an important virulence factor for intracellular survival of Brucella spp. (10, 15, 33). Recently, cyclic -1,2-glucan has been shown as an important factor required for intracellular survival of Brucella (3). Though T4SS, cyclic -1,2-glucan, and LPS are clear virulence factors for Brucella, the attenuated mutants with these factors are either considered not safe or not sufficiently studied as possible vaccines for animals and humans. This has necessitated identification of additional vaccine targets.
Several genetic loci have been identified that are required for Brucella replication in macrophages cultured in vitro (15, 23). In vitro conditions may not adequately reflect in vivo infection, and therefore, findings may have little or no in vivo relevance (45). In vivo screening methods have been used to identify Brucella genes required for survival and persistence (18, 26); however, these studies rely on determining the numbers of tissue-specific CFU from multiple animals at different times, which is labor-intensive and requires large numbers of animals. Because infection is a dynamic process and varies within individual mice, monitoring disease progression temporally within the same mouse would provide a more comprehensive picture of pathogenic events. Further, such real-time analysis may reveal virulence determinants responsible for tissue-specific replication of bacteria that would not be revealed using conventional CFU enumeration.
Bioluminescence imaging of mice allows direct visualization of the infection process and is highly useful for bacterial pathogenesis studies (11), because the intensity of bioluminescence strongly correlates with the number of bacteria in the infected organs (17, 40). Bioluminescence imaging is useful in analyzing subacute and chronic infections that are often difficult to appreciate using conventional approaches because of uncertain bacterial locations (17, 40). Here we report the pathogenicity associated with the three attenuated bioluminescent Brucella melitensis mutants, GR019 (virB4), GR024 (galE), and GR026 (BMEI1090-BMEI1091). Attenuated bacteria could be visualized in the later stages of infection in tissues that are not conventionally evaluated, thus providing an unabridged approach to understand brucellosis affecting multiple tissues. In addition, we describe the dynamics of virulent bioluminescent B. melitensis GR023 infection following vaccination with these attenuated mutants.
MATERIALS AND METHODS
Bacterial strains, plasmids, and growth conditions. Bacterial strains and plasmids used in this study are listed in Table 1. Strains GR019 (virB4), GR024 (galE), and GR026 (BMEI1090-BMEI1091) are EZ::Tnlux transposon insertional mutants of B. melitensis 16M containing the promoterless lux operon. Rev-1 is an attenuated strain of virulent B. melitensis 6056 (2, 14) and is used as a vaccine for brucellosis in small ruminants (5). BM710 is a spontaneous stable rough mutant of Rev-1 isolated from vaccinated sheep and is phenotypically identical to Rev-1 except for the rough LPS (4). GR023 is a virulent bioluminescent B. melitensis 16M mutant used for challenge studies. GR023 was isolated by random EZ::Tnlux transposon mutagenesis and has promoterless luxCDABE genes inserted between a hypothetical protein (BMEI0100) and cysteine synthase A (BMEI0101) (40). All Brucella strains were grown in brucella broth (Difco). Ampicillin (100 μg/ml), chloramphenicol (20 μg/ml), kanamycin (50 μg/ml), and zeocin (50 μg/ml for Escherichia coli and 250 μg/ml for Brucella) were added to the medium as necessary. Brucella strains were grown at 37°C with shaking unless otherwise stated. E. coli strain DH5 (Invitrogen) and EC100Dpir+ (Epicenter, Madison, WI) were grown in LB broth (Difco).
Suicide vectors pGR026-90K and pGR026-91K for generating deletions in BMEI1090 and BMEI1091, respectively, were created using pZErO-1. To construct pGR026-90K, approximately 1-kb DNA sequences upstream and downstream of the deletion target were amplified by PCR (upstream, forward primer, 5'atcaacggtaccCGTTCAGCGCGTCGAGATCG, and reverse primer, 5'gctctaggatccGACTGATAATTATGCCGTGCG; downstream, forward primer, 5'acagtcggatccATAACCGAAGCCTATTCCTTC, and reverse primer, 5'ggtaacctgcagCGAACGTGCCCGCATCAT) and cloned into pZErO-1 to generate plasmid pGR026-90. Appropriate restriction sites were included in the PCR primers to facilitate the insertion of the kanamycin resistance gene aph(3')-la (Kanr) from pUC4K between the two fragments to generate pGR026-90K. Bases in lowercase were added to facilitate cloning, and underlined bases are restriction sites. To construct pGR026-91K, the desired deletion target was amplified with approximately 1-kb upstream and downstream sequences using specific primers (forward, 5'agatacggtaccTCTTCCATCGTTCCGGGCCT; reverse, 5'catgcatctagaGACGCCGTTGATGTTCCATGTA) and cloned into pZErO-1 to generate pGR026-91. Then, inverse PCR was performed on pGR026-91 using primers (5'tcttgagaattcCCCAATGCGACCGCTT and 5'gattcagaattcTTTGGCGATCCGCCTGGCA) designed to amplify all but the deletion target. The inverse PCR product was digested and ligated to the Kanr gene fragment to generate the final suicide vector pGR026-91K.
To construct plasmids pBBVirB4, pBBGalE, and pBBI1087-1090, DNA sequences encoding the respective open reading frames (ORFs) plus the ribosome binding site but lacking their promoter sequences were amplified using primers (for VirB4, forward primer, 5'agagagggtaccCATGTTCATATTGCCGCTGATCG, and reverse primer, 5'agagagggatccTGCTGGTTACA GTCAGGGCGAAT; for GalE, forward primer, 5'agagagggtaccAAAGCCCGGTAAAACGATTGATG, and reverse primer, 5' agagagggatccGTTCCGGCATTTTCTGGCAAA; for 1087-90, forward, 5'agagagactagtTGTGCCGTCGTTTCCACCTG, and reverse, 5'agagagctcgagAGGGACGGGGATCGGGTTAT). PCR products were digested and ligated with similarly digested pBBR-MCS4 to generate the complementation plasmids. The genes of interest were directionally cloned into pBBR1-MCS4 to ensure that these genes are being transcribed from the lac promoter present in the plasmid.
Mapping of the EZ::Tnlux transposon insertion site and Southern blot analysis. The site of transposon insertion in GR019 (virB4), GR024 (galE), and GR026 (BMEI1090-BMEI1091) was identified by rescue cloning as previously described (40). For Southern hybridization, 10 μg of genomic DNA was digested with ClaI and separated in a 0.7% agarose gel. The single-copy insertion of the transposon at the expected location was detected using the Kanr gene as a probe as previously described (40).
Inactivation of BMEI1090 and BMEI1091. To generate specific deletions, suicide vectors pGR026-90K and pGR026-91K were electroporated into B. melitensis 16M. Cells were plated on brucella agar containing kanamycin. To select for double recombinants, the Kanr colonies were checked for sensitivity to zeocin (Zeos). The resulting Kanr and Zeos clones were streak purified, and one such purified clone was used for further study.
Macrophage infection. The macrophage-like RAW 264.7 cells were cultured in RPMI supplemented with 10% heat-inactivated fetal calf serum. For macrophage growth assays, 24-well microtiter plates were seeded with 5 x 105 macrophages/well and infected with different B. melitensis strains at a multiplicity of infection of 1:50. Cells were incubated for 1 h at 37°C in 5% CO2, extracellular bacteria were removed with three washes of phosphate-buffered saline (PBS), followed by gentamicin treatment (25 μg/ml) for 30 min. Then the cells were maintained with medium containing 5 μg of gentamicin/ml. At specified times, cells were washed with PBS three times, lysed with 0.1% Triton X, and plated on brucella agar to determine intracellular bacterial counts. All experiments were performed in duplicate.
IRF-1–/– mouse virulence assay. Groups of 6- to 9-week-old interferon regulatory factor 1-deficient (IRF-1–/–) mice (n = 4) were infected intraperitoneally (i.p.) with 1 x 107 CFU of virB4, galE, BMEI1090-BMEI1091, Rev-1, and BM710 strains. Infected mice were housed in a biosafety level 3 facility and monitored for survival. For imaging, mice were anesthetized with isoflurane, and bioluminescence was recorded with a 10-min integration time using a charge-coupled device camera (Xenogen, Alameda, CA). The livers and spleens were collected aseptically from the surviving mice, homogenized in PBS, and plated on brucella agar. Plates were incubated at 37°C for 4 days, and CFU counts were determined. For histology, portions of livers and spleens were collected, fixed in 10% formalin, and 5-μm sections were stained with hematoxylin and eosin.
Vaccination and challenge studies. IRF-1–/– mice (6 to 9 weeks old) (nine mice/group) were vaccinated with 1 x 107 CFU i.p. with B. melitensis virB4, galE, BMEI1090-BMEI1091, or BM710 strain in 200 μl PBS. Strain Rev-1 was not included, because it is lethal for these mice. For a control, 10 mice were injected with 200 μl PBS. C57BL/6 mice (20 mice/group) were vaccinated i.p. with 5 x 107 CFU with each of the above strains and with Rev-1. Mice were imaged daily using a charge-coupled device camera until challenge. After 60 days, both IRF-1–/– and C57BL/6 mice were challenged with 1 x 106 CFU of virulent bioluminescent B. melitensis GR023 i.p. Following challenge, mice were imaged with 10 min of integration, and dissemination of bioluminescent GR023 was monitored.
For IRF-1–/– mice, the survival was recorded in different groups following virulent challenge. At 44 days postchallenge, livers and spleens from surviving mice were processed for CFU enumeration. For C57BL/6 mice, four mice from each group were killed at weekly intervals to determine CFU in livers and spleens. Portions of the livers and spleens were weighed, homogenates were serially diluted in PBS and plated on brucella agar with or without antibiotic, and colonies were counted after 72 h of incubation at 37°C. CFU were determined per gram of each tissue. To determine the histological changes at each time, portions of livers and spleens were collected and processed as described above.
Statistical analyses. All statistical analyses were performed using Minitab 13.31 statistical software (Minitab Inc., State College, PA). Association between each group at different times was evaluated using one-way analysis of variance. For all statistical analyses, P values of <0.05 were considered significant.
RESULTS
Pathogenicity of the three attenuated bioluminescent B. melitensis mutants. Previously we described identification and dissemination of three attenuated bioluminescent B. melitensis mutants in mice using an in vivo imaging system (40). In order to determine virulence and pathology associated with these strains, we tested these bioluminescent mutants in IRF-1–/– mice. In addition, we also tested two other B. melitensis mutants, BM710, a stable rough strain and Rev-1, a vaccine strain, so that the bioluminescent mutants could be evaluated for their ability to confer protection against virulent challenge. IRF-1–/– mice (n = 4) infected with these strains were monitored for bacterial dissemination, persistence, and pathology induced in liver and spleen. Mice infected with all three bioluminescent strains (GR019 [virB4], GR024 [galE], and GR026 [BMEI1090-BMEI1091]) and strain BM710 appeared healthy and survived longer than 24 days, suggesting attenuation. However, all Rev-1-infected mice died by 7 days postinfection (p.i.). Although Rev-1 strain is a commercial vaccine strain, it was fully virulent in these mice. To determine the pathology associated with different attenuated mutants, the livers and spleens were processed for CFU and histopathology. Livers and spleens from GR019 (virB4)- or BM710-infected mice had lower CFU than GR024 (galE)- or GR026 (BMEI1090-BMEI1091)-infected mice (Table 2). However, except for the GR026 (BMEI1090-BMEI1091)-infected group, livers and spleens from other groups had no observable histological changes (Table 2 and data not shown). The GR026 (BMEI1090-BMEI1091)-infected mice displayed very few multifocal granulomas in livers and minor changes in the white pulp of spleens.
virB4 but not the galE or BMEI1090-BMEI1091 mutant is attenuated in RAW 264.7 macrophages. To test whether the bioluminescent mutants were attenuated in macrophages, we examined their growth in RAW 264.7 macrophage-like cells. All three strains grew at rates similar to that of strain 16M, with a duplication time of 2 h in brucella broth suggesting no general growth defect (Fig. 1A). RAW 264.7 macrophages were infected with each strain at a multiplicity of infection of 1:50, and the intracellular growth was monitored for 72 h. Interestingly, only strain GR019 (virB4) was defective in replication with a significant decrease in intracellular bacteria by 24 h p.i. compared to 16M (Fig. 1B). On the other hand, both GR024 (galE) and GR026 (BMEI1090-BMEI1091) displayed growth curves distinct from those of GR019 (virB4) and 16M. Both strains were phagocytosed more with no apparent intramacrophage replication during 24 h p.i. as bacterial levels remained constant. However, by 48 h their growth was similar to 16M (Fig. 1). The growth curves of GR024 (galE) and GR026 (BMEI1090-BMEI1091) mutants appeared as intermediate between smooth (e.g., 16M) and rough (e.g., perA) strains of Brucella (41). Rough strains of Brucella are phagocytosed more efficiently than smooth strains and persist intracellularly at higher levels for more than 3 days even though they are defective for intramacrophage replication (20, 34). This led us to suspect that both GR024 (galE) and GR026 (BMEI1090-BMEI1091) may have an LPS defect; this suspicion was supported by the fact that these strains agglutinated in the presence of acriflavin (7). As suggested by the macrophage growth pattern, both strains resulted in fine amorphous agglutination particles that were less intense compared to rough strains of Brucella (i.e., RB51, BM710, and GI-2 deletion mutant; data not shown).
Molecular analysis of bioluminescent mutants. EZ::Tnlux insertion was localized to virB4 (BMEII0028) for strain GR019, galE homolog (BMEI0921) for GR024, and in the intergenic region of two divergent ORFs (BMEI1090 and BMEI1091) for GR026 (Fig. 2). The Brucella strains with mutations in the virB genes encoding T4SS are attenuated in macrophages as well as in mice (10, 15, 33). Similarly, in GR024, the insertion disrupted the galE homolog that has been previously shown to attenuate Brucella (35). Analysis of the genomic context of insertion in GR026 suggested that BMEI1090 is the first gene in a cluster of genes that are transcribed in reverse orientation, whereas BMEI1091 is an independent transcriptional unit (Fig. 2). Further, Southern blot analyses confirmed the sequencing results and also revealed the single-copy insertion of the transposon in these strains (data not shown).
To determine the gene(s) likely responsible for the observed phenotype of BMEI1090-BMEI1091 mutant, we generated BMEI1090 and BMEI1091 deletion mutants by allelic replacement. The respective ORFs were replaced with a Kanr marker by homologous recombination, and the resulting strains, GR-BMEI1090 and GR-BMEI1091, were tested for virulence in IRF-1–/– mice. IRF-1–/– mice infected with GR-BMEI1091 died within 10 days similar to virulent strain16M; however, only two mice infected with GR-BMEI1090 died, and the remaining mice survived for at least 21 days (Table 2). The livers and spleens from the surviving mice had average CFU of 6.65E + 04 and 1.14E + 06, respectively. Therefore, inactivation of BMEI1090 resulted in partial attenuation, suggesting that the phenotype associated with BMEI1090-BMEI1091 is likely due to altered expression of BMEI1090 and downstream genes.
To confirm that the attenuation of bioluminescent mutants is due to disruption of transposon insertion targets and not due to secondary mutations, we complemented virB4, galE, and BMEI1090-BMEI1091 mutations with corresponding ORFs. Since strain GR019 (virB4) has a growth defect in RAW 264.7 macrophages, GR019 (virB4) containing either pBBVirB4 or pBBVirB was tested for growth in these macrophages. Addition of pBBVirB4 in strain GR019 (virB4) resulted in partial restoration of growth, as suggested by an increase in intracellular bacteria at 24 h p.i. (Fig. 3A). However, addition of pBBVirB, containing the entire virB operon (33), into GR019 (virB4) resulted in complete restoration of growth (Fig. 3). In addition, virB4 complemented with pBBVirB, but not pBBVirB4, killed IRF-1–/– mice and restored complete virulence (Fig. 3B). Consistent with the in vitro results, mice infected with GR019 (virB4)/pBBVirB4 did not die and contained more bacteria in livers and spleens than mice infected with GR019 (virB4) (Fig. 3C).
Since both strains GR024 (galE) and GR026 (BMEI1090-BMEI1091) agglutinated in the presence of acriflavin, we tested the galE- and BMEI1090-BMEI1091-complemented strains for agglutination. GR024 (galE) complemented with pBBGalE resulted in no agglutination in the presence of acriflavin (data not shown). Since our earlier results suggested that the attenuation of GR026 (BMEI1090-BMEI1091) is likely due to the altered expression of BMEI1090 and downstream genes, we complemented GR026 with a plasmid containing four ORFs likely to form an operon (13). Surprisingly, addition of pBBI1087-1090 to BMEI1090-BMEI1091 resulted in much pronounced agglutination as seen with rough strains of Brucella (data not shown). Consistent with the acriflavin agglutination results, both GR024 (galE) and GR026 (BMEI1090-BMEI1091) were partially resistant to smooth-type-specific Tbilisi (Tb) phage, and the addition of pBBGalE restored the susceptibility of GR024 (galE) to Tb phage. However, GR026 (BMEI1090-BMEI1091) complemented with pBBI1087-1090 was completely resistant to Tb phage, suggesting a rough phenotype of the complemented strain (data not shown).
The galE and BMEI1090-BMEI1091 mutants protect IRF-1–/– mice from virulent challenge. Although IRF-1–/– mice are immunocompromised, they are protected against virulent Brucella following vaccination with attenuated strains (21). Therefore, we tested the attenuated bioluminescent mutants to determine if they protect IRF-1–/– mice from virulent challenge. IRF-1–/– mice vaccinated with attenuated bioluminescent mutants were challenged when no bioluminescent bacteria were detectable. To evaluate vaccine candidates, virulent strain GR023 was used to visualize dissemination and tissue localization of virulent Brucella by temporally imaging individual mice. All mice vaccinated with either strain GR024 (galE) or GR026 (BMEI1090-BMEI1091) survived at least 44 days, whereas only two mice vaccinated with GR019 (virB4) and three mice vaccinated with BM710 survived for 44 days following challenge (Fig. 4). Fifty percent of GR019 (virB4)-vaccinated mice died by day 12, whereas 50% of the BM710-vaccinated mice died by day 9 following challenge. As expected, all unvaccinated mice died within 2 weeks following challenge, with 50% mice dead after 7 days (Fig. 4).
The livers and spleens from surviving mice vaccinated with different strains had very similar CFU (CFU ranges, 2.2E + 02 to 1.2E + 03/gram of tissue in the liver and 1.5E + 04 to 3.4E + 04/gram of tissue in the spleen). Bacteria recovered from the livers and spleens of mice vaccinated with bioluminescent strains were confirmed as strain GR023 by verifying the EZ::Tnlux insertion site using PCR (data not shown). Bioluminescence imaging of vaccinated mice following i.p. challenge revealed strikingly different dynamics of persistence and spread of virulent bacteria. Unlike the unvaccinated mice, in all vaccinated groups, bacterial spread was less extensive, but correlated with the ability of the vaccine strain to protect from challenge (Fig. 5). In both BM710- and GR019 (virB4)-vaccinated groups, bioluminescence was pronounced with systemic spread by day 10; however, in both GR024 (galE)- and GR026 (BMEI1090-BMEI1091)-vaccinated groups, bioluminescence was observed at the injection site and in the tail (Fig. 5). By day 44, GR024 (galE)- and GR026 (BMEI1090-BMEI1091)-vaccinated mice had no detectable bioluminescent bacteria, while BM710- and GR019 (virB4)-vaccinated survivors still exhibited bioluminescence (Fig. 5B). Consistent with survival data, the GR024 (galE)- and GR026 (BMEI1090-BMEI1091)-vaccinated IRF-1–/– mice had the least histological changes in livers and spleens (Fig. 5). The GR024 (galE)- and GR026 (BMEI1090-BMEI1091)-vaccinated mice had few focal granulomas (less than three per field of view at a magnification of x4) in liver sections, while the spleens of GR024 (galE)-vaccinated mice appeared normal and with minimal disorganization of the white pulp in GR026 (BMEI1090-BMEI1091)-vaccinated mice. Both GR019 (virB4)- and BM710-vaccinated survivors had more histological changes in livers and spleens than GR024 (galE)- and GR026 (BMEI1090-BMEI1091)-vaccinated groups (Fig. 5B).
IRF-1–/– mice are defective in multiple aspects of the immune system (44). To better correlate the immune protection provided by the different attenuated strains, we tested these mutants in wild-type C57BL/6 mice, the parental strain of IRF-1–/– mice. C57BL/6 mice clear virulent Brucella infection naturally and serve as a relevant model to study Brucella pathogenesis and immune protection. To assess the protection by different attenuated strains, we monitored bacterial clearance and histological changes in livers and spleens. In addition, the dynamics of infection by attenuated bioluminescent strains and their effects on virulent challenge were monitored by imaging. Similar to IRF-1–/– mice, GR019 (virB4)-vaccinated C57BL/ 6 mice had bioluminescence in livers and spleens by day 1 p.i.; however, in GR024 (galE)- or GR026 (BMEI1090-BMEI1091)-vaccinated mice, bioluminescence was detected primarily at the injection site (Fig. 6A). Bioluminescence began to diminish by day 5 in all groups, and by 2 weeks p.i. minimal or no bioluminescence was observed (Fig. 6A). However, after challenge the dynamics of virulent Brucella dissemination was similar in all vaccinated groups being limited primarily to the injection site, though bioluminescence was stronger in GR019 (virB4)- and BM710-vaccinated groups (Fig. 6B). Consistent with imaging data, all vaccinated groups had significantly fewer CFU (P < 0.05) in livers and spleens 1 week postchallenge compared to the unvaccinated group. In addition, GR024 (galE)-, GR026 (BMEI1090-BMEI1091)-, and Rev-1-vaccinated groups contained even lower numbers of CFU (P < 0.05) than GR019 (virB4)- or BM710-vaccinated group did (Fig. 7). Though the spleens from GR024 (galE)- and Rev-1-vaccinated mice had lower CFU at all times than the spleens from other groups (Fig. 7), CFU counts from spleens at 3 and/or 4 weeks postchallenge in GR024 (galE)- and Rev-1-vaccinated groups were significantly lower (P < 0.05). To correlate bacterial clearance with tissue damage, histological changes were assessed in livers and spleens from immunized mice following challenge. Consistent with bacterial clearance, GR024 (galE)- and GR026 (BMEI1090-BMEI1091)-vaccinated mice exhibited fewer granulomas in livers; however, livers from GR019 (virB4) and BM710-vaccinated mice contained more granulomas (Table 3). Surprisingly, only the livers from Rev-1-vaccinated mice had large grossly visible focal calcified granulomas (Fig. 8). On the other hand, histological changes in spleens were similar in all vaccinated groups but contained fewer changes compared to unvaccinated mice (Table 3).
DISCUSSION
In this report, we describe infection dynamics of three attenuated bioluminescent mutants and the effect of vaccination on the dynamics of virulent bacterial infection in mice. Strains GR019 (virB4), GR024 (galE), GR026 (BMEI1090-BMEI1091), and BM710 were all attenuated in IRF-1–/– mice; however, Rev-1 remained virulent in these mice. As previously reported, all three attenuated bioluminescent strains revealed striking differences in bacterial dissemination and persistence. GR019 (virB4), unlike GR024 (galE) or GR026 (BMEI1090-BMEI1091), spread systemically, and bioluminescence was observed in liver, spleen, testes, submandibular region, and extremities early in infection, confirming previous reports that the VirB system is not required for establishing early infection (18). However, the VirB system is required for Brucella persistence, as C57BL/6 mice cleared GR019 (virB4) infection faster than virulent Brucella did (data not shown), which implies that the presence of Brucella T4SS may provide a selective advantage for the bacterium against host defense during later stages of infection. The GR024 (galE) and GR026 (BMEI1090-BMEI1091) strains, on the other hand, were defective in systemic spread, with bioluminescence localized primarily to the injection site (40). Both GR024 (galE) and GR026 (BMEI1090-BMEI1091) have altered surface structure, likely in the LPS component, and therefore are unable to replicate efficiently and spread. Brucella strains defective in LPS structure are more susceptible to innate bactericidal effects of phagocytes and are defective in intracellular survival (25). Interestingly, in both GR024 (galE)- and GR026 (BMEI1090-BMEI1091)-infected mice, bioluminescence reappeared 12 days p.i. and localized in the joint-rich tail region during the later stages of infection (40; data not shown). Spinal involvement in brucellosis is more destructive and occurs in 20 to 65% of all patients with musculoskeletal brucellosis (27). Our results parallel those of human brucellosis and may provide a model to understand chronic localization of Brucella. These bioluminescent joints contained up to 104 bacteria. Though these strains localized to tail joints, they induced no histological changes in these locations as well as in livers and spleens. However, whether the recovered bacteria have altered LPS and whether that LPS plays a role in chronic localization of Brucella needs to be determined. Importantly, bioluminescence imaging revealed localization of LPS-defective Brucella in tissues that have not been previously examined by others, suggesting a requirement for a comprehensive examination of infected mice to assess bacterial clearance. In addition, temporal analysis of infection revealed patterns of growth and clearance as well as the reemergence of bacteria that is extremely difficult to observe with conventional methods. Thus, our study clearly demonstrates that bioluminescence imaging in conjunction with CFU enumeration provides better assessment of in vivo clearance of Brucella affecting multiple tissues. Importantly, only GR019 (virB4) was attenuated in RAW 264.7 macrophages (Fig. 1B). Therefore, imaging may provide a comprehensive approach to identify Brucella genes that are relevant to in vivo pathogenesis.
Both galE and BMEI1090-BMEI1091 mutants exhibited growth patterns in macrophages intermediate to smooth and rough strains of Brucella (41), suggesting that they may have an altered surface structure. Both strains produced very fine agglutination in the presence of acriflavin and were partially resistant to smooth-type-specific Tb phage (30). In strain GR024 (galE), the transposon insertion is in ORF BMEI0921, a NAD-dependent epimerase/dehydratase family member that is closely related to enterobacterial galE. In many gram-negative bacteria, galactose is converted to UDP-galactose and the galE gene product, UDP-galactose 4-epimerase, catalyzes reversible conversion of UDP-galactose to UDP-glucose. UDP-galactose serves as donor for both LPS core and O-antigen polysaccharide biosynthesis. Therefore, galE mutants are defective in LPS, and thus, the galE gene is an important virulence factor. The acriflavin agglutination and phage susceptibility tests suggest a defect in the GR024 (galE) LPS; however, GR024 (galE) was not sensitive to growth in galactose-containing medium (1, 37; data not shown). The galE mutants fed excess galactose accumulate UDP-galactose, which is toxic to the cell (1). The galE mutants of other bacteria display contrasting responses to galactose, with some being sensitive while others were not sensitive to galactose (16, 19, 29, 36). Strikingly, the galE mutants of Brucella abortus and B. melitensis have different responses to growth in galactose (35, 43). The B. melitensis galE mutant (Bm92) is not sensitive to galactose; however, a plasmid that encodes the B. melitensis galE ORF complemented a galE mutation in Salmonella enterica serovar Typhimurium LB5010, as shown by the restoration of smooth LPS, sensitivity to phage P22 infection, and restoration of UDP galactose-4-epimerase activity, suggesting a role for GalE in LPS biogenesis (35). Interestingly, LPS of B. melitensis mutant Bm92 has been reported to have no major differences compared to LPS from strain 16M (35). This implies that minor differences that were not readily discernible by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and Western blot analysis may exist, since extraction of smooth LPS from Brucella is difficult due to its peculiar composition (25). Analysis of the galE homolog of Yersinia enterocolitica indicates that although it can complement an E. coli galE mutant, its primary function in Y. enterocolitica is not in the production of UDP-galactose but, instead, some other nucleotide sugar required for LPS biosynthesis (36). However, without fine structural analysis of LPS, it is not known whether Y. enterocolitica LPS has galactose or whether galE catalyzes the formation of a closely related sugar in the LPS structure. Similarly, in Brucella GalE may modify a closely related sugar, but a detailed chemical analysis of the LPS composition will be required to identify the role of GalE in LPS biogenesis. Consistent with the above findings, our data also indicate minor changes in the GR024 (galE) LPS, as it was partially resistant to Tb phage and produced very fine agglutination particles instead of large particulate agglutination seen with rough mutants of Brucella (data not shown).
The GR026 strain has an insertion in the intergenic region of BMEI1090 and BMEI1091. Further, selective allelic replacement of these ORFs suggested that BMEI1090 and downstream genes are responsible for the attenuation of GR026 (BMEI1090-BMEI1091) (Table 1). Interrogation of the B. melitensis genome suggested that BMEI1090 and downstream genes (BMEI1087 to BMEI1090) likely form an operon. BMEI1087 encodes -hexosaminidase A, while BMEI1088 encodes soluble lytic murein transglycosylase, and these are involved in amino sugar metabolism and N-glycan biosynthesis (http://www.genome.ad.jp/kegg). Therefore, this operon may contribute to cell membrane biogenesis. Consistent with this observation, the acriflavin agglutination and Tb phage susceptibility tests suggested that GR026 (BMEI1090-BMEI1091) has a surface structure defect. Complementation of BMEI1090-BMEI1091 with a plasmid containing BMEI1087-BMEI1090 ORFs resulted in more pronounced agglutination and complete resistance to Tb phage, which suggests that the expression of these genes is under strict regulation.
Vaccination with both GR024 (galE) and GR026 (BMEI1090-BMEI1091) strains protected IRF-1–/– mice from virulent B. melitensis challenge, whereas strains GR019 (virB4) and BM710 failed to protect these mice. In addition, GR024 (galE)- and GR026 (BMEI1090-BMEI1091)-vaccinated mice displayed minimal changes in livers and spleens, and no bioluminescence was observed at 44 days postchallenge. IRF-1–/– mice are defective in multiple immune components, with reduced numbers of CD8+ T cells, functionally impaired natural killer cells, and dysregulation of interleukin-12 p40 and inducible nitric oxide synthase (44). Though these mice are severely immunocompromised, they mount an adaptive immune response sufficient to protect against virulent challenge, and the protection is vaccine strain dependent. Unlike GR019 (virB4), both GR024 (galE) and GR026 (BMEI1090-BMEI1091) produced a localized but persistent infection in these mice (40; data not shown) and induced a protective immune response against virulent Brucella. This result is consistent with Plommet's observation that survival of the vaccine strain in the host for a critical period determines the efficacy of Brucella vaccines (38). Similar results have been observed with two field vaccine strains, S19 and RB51 in that S19 has been shown to persist longer and is more protective than RB51 in mice and other models (22, 42). However, S19 still possesses residual virulence in domestic animals and in IRF-1–/– mice (21, 31), whereas RB51 is highly attenuated (21). The galE and BMEI1090-BMEI1091 mutants are highly attenuated in IRF-1–/– mice, similar to RB51, but cause no or very minimal pathological changes in livers and spleen and are protective. Consistent with the IRF-1–/– mouse data, both GR024 (galE) and GR026 (BMEI1090-BMEI1091) provided greater protection to C57BL/6 mice than GR019 (virB4) or BM710 did, which suggests that IRF-1–/– mice may serve as an important model to rapidly assess vaccine efficacy of Brucella strains. Interestingly, Rev-1-vaccinated mice had fewer CFU in both livers and spleens than GR024 (galE)- and GR026 (BMEI1090-BMEI1091)-vaccinated mice; however, Rev-1-vaccinated mice displayed severe liver damage with grossly visible lesions (Fig. 8) that was not observed in other groups. These lesions are likely vaccine induced, as they were apparent even at 1 week postchallenge. Rev-1 vaccine is used in domestic animals with various degrees of success in areas where B. melitensis is endemic (5). Although Rev-1 protected wild-type mice, Rev-1 was highly virulent to IRF-1–/– mice (Table 2) and caused severe liver damage in wild-type mice. This is in line with the fact that Rev-1 strain can still cause clinical brucellosis (38). In summary, our study revealed the contribution of Brucella genes to in vivo pathogenesis and identified a new set of virulence genes (BMEI1090 and downstream genes). Further, the GalE-deficient GR024 strain has altered LPS, results in no detectable tissue damage, and protects against virulent B. melitensis challenge, making it an interesting vaccine candidate for brucellosis.
ACKNOWLEDGMENTS
We thank David Warshauer and Tim Monson at the Wisconsin State Laboratory of Hygiene for help with phage typing of Brucella strains. We also thank Tajie Harris and Michael Krepps for assistance with mapping bioluminescent mutants and macrophage infection and Fue Vang for statistical assistance.
This work was supported by the NIH grant R01AI048490, NIH/NIAID RCE for Biodefense and Emerging Infectious Diseases Research Program grant 1-U54-AI-057153, and USDA grant 35204-14856.
REFERENCES
1. Adhya, S. 1987. The galactose operon, p. 1503-1512. In F. C. Neidhardt, K. B. Low, B. Magasanik, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella typhimurium: cellular and molecular biology. American Society for Microbiology, Washington, D.C.
2. Alton, G. G., S. S. Elberg, and D. Crouch. 1967. Rev. 1 Brucella melitensis vaccine. The stability of the degree of attenuation. J. Comp. Pathol. 77:293-300.
3. Arellano-Reynoso, B., N. Lapaque, S. Salcedo, G. Briones, A. E. Ciocchini, R. Ugalde, E. Moreno, I. Moriyon, and J. P. Gorvel. 2005. Cyclic beta-1,2-glucan is a Brucella virulence factor required for intracellular survival. Nat. Immunol. 6:618-625.
4. Bardenstein, S., M. Mandelboim, T. A. Ficht, M. Baum, and M. Banai. 2002. Identification of the Brucella melitensis vaccine strain Rev.1 in animals and humans in Israel by PCR analysis of the PstI site polymorphism of its omp2 gene. J. Clin. Microbiol. 40:1475-1480.
5. Blasco, J. M. 1997. A review of the use of B. melitensis Rev 1 vaccine in adult sheep and goats. Prev. Vet. Med. 31:275-283.
6. Boschiroli, M. L., V. Foulongne, and D. O'Callaghan. 2001. Brucellosis: a worldwide zoonosis. Curr. Opin. Microbiol. 4:58-64.
7. Braun, W., and A. E. Bonestell. 1947. Independent variation of characteristics in Brucella abortus variants and their detection. Am. J. Vet. Res. 8:386-390.
8. Celli, J., C. de Chastellier, D. M. Franchini, J. Pizarro-Cerda, E. Moreno, and J. P. Gorvel. 2003. Brucella evades macrophage killing via VirB-dependent sustained interactions with the endoplasmic reticulum. J. Exp. Med. 198:545-556.
9. Celli, J., and J. P. Gorvel. 2004. Organelle robbery: Brucella interactions with the endoplasmic reticulum. Curr. Opin. Microbiol. 7:93-97.
10. Comerci, D. J., M. J. Martinez-Lorenzo, R. Sieira, J. P. Gorvel, and R. A. Ugalde. 2001. Essential role of the VirB machinery in the maturation of the Brucella abortus-containing vacuole. Cell. Microbiol. 3:159-168.
11. Contag, C. H., P. R. Contag, J. I. Mullins, S. D. Spilman, D. K. Stevenson, and D. A. Benaron. 1995. Photonic detection of bacterial pathogens in living hosts. Mol. Microbiol. 18:593-603.
12. Corbel, M. J. 1997. Brucellosis: an overview. Emerg. Infect. Dis. 3:213-221.
13. DelVecchio, V. G., V. Kapatral, R. J. Redkar, G. Patra, C. Mujer, T. Los, N. Ivanova, I. Anderson, A. Bhattacharyya, A. Lykidis, G. Reznik, L. Jablonski, N. Larsen, M. D'Souza, A. Bernal, M. Mazur, E. Goltsman, E. Selkov, P. H. Elzer, S. Hagius, D. O'Callaghan, J. J. Letesson, R. Haselkorn, N. Kyrpides, and R. Overbeek. 2002. The genome sequence of the facultative intracellular pathogen Brucella melitensis. Proc. Natl. Acad. Sci. USA 99:443-448.
14. Elberg, S. S., and K. Faunce, Jr. 1957. Immunization against Brucella infection. VI. Immunity conferred on goats by a nondependent mutant from a streptomycin-dependent mutant strain of Brucella melitensis. J. Bacteriol. 73:211-217.
15. Foulongne, V., G. Bourg, C. Cazevieille, S. Michaux-Charachon, and D. O'Callaghan. 2000. Identification of Brucella suis genes affecting intracellular survival in an in vitro human macrophage infection model by signature-tagged transposon mutagenesis. Infect. Immun. 68:1297-1303.
16. Fry, B. N., S. Feng, Y. Y. Chen, D. G. Newell, P. J. Coloe, and V. Korolik. 2000. The galE gene of Campylobacter jejuni is involved in lipopolysaccharide synthesis and virulence. Infect. Immun. 68:2594-2601.
17. Hardy, J., K. P. Francis, M. DeBoer, P. Chu, K. Gibbs, and C. H. Contag. 2004. Extracellular replication of Listeria monocytogenes in the murine gall bladder. Science 303:851-853.
18. 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.
19. Houng, H. S., D. J. Kopecko, and L. S. Baron. 1990. Molecular cloning and physical and functional characterization of the Salmonella typhimurium and Salmonella typhi galactose utilization operons. J. Bacteriol. 172:4392-4398.
20. Jimenez de Bagues, M. P., A. Gross, A. Terraza, and J. Dornand. 2005. Regulation of the mitogen-activated protein kinases by Brucella spp. expressing a smooth and rough phenotype: relationship to pathogen invasiveness. Infect. Immun. 73:3178-3183.
21. 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.
22. Ko, J., and G. A. Splitter. 2003. Molecular host-pathogen interaction in brucellosis: current understanding and future approaches to vaccine development for mice and humans. Clin. Microbiol. Rev. 16:65-78.
23. Kohler, S., V. Foulongne, S. Ouahrani-Bettache, G. Bourg, J. Teyssier, M. Ramuz, and J. P. Liautard. 2002. The analysis of the intramacrophagic virulome of Brucella suis deciphers the environment encountered by the pathogen inside the macrophage host cell. Proc. Natl. Acad. Sci. USA 99:15711-15716.
24. Kovach, M. E., P. H. Elzer, D. S. Hill, G. T. Robertson, M. A. Farris, R. M. Roop II, and K. M. Peterson. 1995. Four new derivatives of the broad-host-range cloning vector pBBR1MCS, carrying different antibiotic-resistance cassettes. Gene 166:175-176.
25. Lapaque, N., I. Moriyon, E. Moreno, and J. P. Gorvel. 2005. Brucella lipopolysaccharide acts as a virulence factor. Curr. Opin. Microbiol. 8:60-66.
26. Lestrate, P., R. M. Delrue, I. Danese, C. Didembourg, B. Taminiau, P. Mertens, X. De Bolle, A. Tibor, C. M. Tang, and J. J. Letesson. 2000. Identification and characterization of in vivo attenuated mutants of Brucella melitensis. Mol. Microbiol. 38:543-551.
27. Madkour, M. M. 2001. Osteoarticular brucellosis, p. 74-84. In M. M. Makour (ed.), Madkour's brucellosis. Springer, Heidelberg, Germany.
28. Martinez de Tejada, G., J. Pizarro-Cerda, E. Moreno, and I. Moriyon. 1995. The outer membranes of Brucella spp. are resistant to bactericidal cationic peptides. Infect. Immun. 63:3054-3061.
29. Metzger, M., P. Bellemann, P. Bugert, and K. Geider. 1994. Genetics of galactose metabolism of Erwinia amylovora and its influence on polysaccharide synthesis and virulence of the fire blight pathogen. J. Bacteriol. 176:450-459.
30. Monreal, D., M. J. Grillo, D. Gonzalez, C. M. Marin, M. J. De Miguel, I. Lopez-Goni, J. M. Blasco, A. Cloeckert, and I. Moriyon. 2003. Characterization of Brucella abortus O-polysaccharide and core lipopolysaccharide mutants and demonstration that a complete core is required for rough vaccines to be efficient against Brucella abortus and Brucella ovis in the mouse model. Infect. Immun. 71:3261-3271.
31. Moriyon, I., M. J. Grillo, D. Monreal, D. González, C. Marín, I. Lopez-Goi, R. C. Mainar-Jaime, E. Moreno, and J. M. Blasco. 2004. Rough vaccines in animal brucellosis: structural and genetic basis and present status. Vet. Res. 35:1-38.
32. Naroeni, A., and F. Porte. 2002. Role of cholesterol and the ganglioside GM1 in entry and short-term survival of Brucella suis in murine macrophages. Infect. Immun. 70:1640-1644.
33. O'Callaghan, D., C. Cazevieille, A. Allardet-Servent, M. L. Boschiroli, G. Bourg, V. Foulongne, P. Frutos, Y. Kulakov, and M. Ramuz. 1999. Homologue of the Agrobacterium tumefaciens VirB and Bordetella pertussis Ptl type IV secretion systems is essential for intracellular survival of Brucella suis. Mol. Microbiol. 33:1210-1220.
34. Pei, J., and T. A. Ficht. 2004. Brucella abortus rough mutants are cytopathic for macrophages in culture. Infect. Immun. 72:440-450.
35. Petrovska, L., R. G. Hewinson, G. Dougan, D. J. Maskell, and M. J. Woodward. 1999. Brucella melitensis 16M: characterisation of the galE gene and mouse immunisation studies with a galE deficient mutant. Vet. Microbiol. 65:21-36.
36. Pierson, D. E., and S. Carlson. 1996. Identification of the galE gene and a galE homolog and characterization of their roles in the biosynthesis of lipopolysaccharide in a serotype O:8. strain of Yersinia enterocolitica. J. Bacteriol. 178:5916-5924.
37. Plommet, M. 1991. Minimal requirements for growth of Brucella suis and other Brucella species. Zentbl. Bakteriol. 275:436-450.
38. Plommet, M. 1995. Live vaccines: virulence, immunogenicity/protection and safety. Historical, theoretical and practical considerations applied to the Brucella melitensis Rev. 1 vaccine, p. 7-18. In B. Garin-Bastugi and A. Benkirane (ed.), FAO/WHO/OIE Round Table on the Use of Rev. 1 Vaccine in Small Ruminants and Cattle, 21 to 22 September 1995. Centre National d'Etudes Veterinaires et Alimentaires, Maison-Alfort, France.
39. Porte, F., A. Naroeni, S. Ouahrani-Bettache, and J. P. Liautard. 2003. Role of the Brucella suis lipopolysaccharide O antigen in phagosomal genesis and in inhibition of phagosome-lysosome fusion in murine macrophages. Infect. Immun. 71:1481-1490.
40. 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. 10:1459-1473.
41. Rittig, M. G., A. Kaufmann, A. Robins, B. Shaw, H. Sprenger, D. Gemsa, V. Foulongne, B. Rouot, and J. Dornand. 2003. Smooth and rough lipopolysaccharide phenotypes of Brucella induce different intracellular trafficking and cytokine/chemokine release in human monocytes. J. Leukoc. Biol. 74:1045-1055.
42. Schurig, G. G., N. Sriranganathan, and M. J. Corbel. 2002. Brucellosis vaccines: past, present and future. Vet. Microbiol. 90:479-496.
43. Scupham, A. J., and E. W. Triplett. 1997. Isolation and characterization of the UDP-glucose 4'-epimerase-encoding gene, galE, from Brucella abortus 2308. Gene 202:53-59.
44. Taniguchi, T., K. Ogasawara, A. Takaoka, and N. Tanaka. 2001. IRF family of transcription factors as regulators of host defense. Annu. Rev. Immunol. 19:623-655.
45. Unsworth, K. E., and D. W. Holden. 2000. Identification and analysis of bacterial virulence genes in vivo. Philos. Trans. R. Soc. Lond. B 355:613-622.
46. Young, E. J. 1995. An overview of human brucellosis. Clin. Infect. Dis. 21:283-289.(Gireesh Rajashekara, Davi)