Brucella: Molecular and Cellular Biology Chapter Abstracts
Chapter 1
DNA Polymorphism and Taxonomy of Brucella Species
Axel Cloeckaert and Nieves Vizcaino
Abstract
Six species are currently recognized within the genus Brucella: B. melitensis, B. abortus, B. suis, B. neotomae, B. ovis, and B. canis. This classification is mainly based on differences in pathogenicity and in host preferences (sheep and goats, cattle, swine, desert rat, rams, and dogs respectively). Although the six species can be differentiated by conventional phenotypic tests, these species display a high degree of DNA homology in DNA-DNA hybridization assays (> 90% identity). Therefore it has been proposed that the Brucella genus should comprise only one species i.e. B. melitensis and that the other species should be considered as biovars. However, several molecular genotyping methods has been used to show that Brucella species display significant DNA polymorphism allowing the species to be differentiated thereby justifying the current species classification. This is also true for the recent marine mammal Brucella isolates for which two new species names have been proposed, i.e. B. pinnipediae and B. cetaceae, according to the classical criteria of host preferentialism (pinnipeds and cetaceans respectively) and specific molecular markers.
Chapter 2
Molecular Diagnostics of Animal Brucellosis: A Review of PCR-Based Assays and Approaches
Betsy J. Bricker
Abstract
Most of the countries that are faced with the economic losses and public health issues caused by animal brucellosis have governmental programs for the eradication or control of the disease. Accurate diagnostic procedures are critical for the success of these programs. This review examines PCR-based diagnostics for animal brucellosis in three contexts: 1) for screening large populations of livestock to identify infected herds, 2) for confirming the presence of infection in herds and identifying the species involved so that appropriate regulatory actions are taken, and 3) for identifying epizootic strains to help epidemiologists trace-back infections to their sources. Emphasis is placed on the diversity of methods that have been developed thus far, how the tests are used, and which tests are currently being tested and used in the field.
Chapter 3
Polymerase Chain Reaction: A Powerful New Approach for the Diagnosis of Human Brucellosis
Juan D. Colmenero, María I. Queipo-Ortuño and Pilar Morata
Abstract
Brucellosis is a zoonosis transmittable to humans that shows a high degree of morbidity. More than 500,000 new cases of human brucellosis are reported each year, and according to the World Health Organisation, this figure underestimates the magnitude of the problem. Owing to the heterogeneous and poorly specific clinical symptomatology of the disease, its diagnosis always requires laboratory confirmation. The conventional microbiological methods used for the diagnosis of human brucellosis have important limitations. Although blood culture provides the best results in microbiological diagnosis, its sensitivity is considerably reduced in patients with long-term clinical courses or with focal complications. Furthermore, blood cultures are time-consuming and handling of the microorganisms represents a high risk for laboratory personnel, since Brucella spp. are Class III pathogens. Serological diagnosis lacks specificity in areas where the disease is endemic and in those persons exposed professionally to Brucella. Moreover, cross-reactions with other bacteria can also occur. In order to overcome some of the limitations of these conventional techniques, assays based on the polymerase chain reaction (PCR) have been proposed as a very useful tool for the diagnosis of human of brucellosis. Our group has recently reported that PCR methods applied to blood samples provide better results than conventional culture techniques for the diagnosis of both primary infection and relapses, as well as for focal complications of the disease. In the present Chapter, we review the current status and future contribution of PCR techniques for the diagnosis and follow-up of human brucellosis.
Chapter 4
A Brucella melitensis Genomic Database
Xavier De Bolle, Christophe Lambert, Eric Depiereux and Jean-Jacques Letesson
Abstract
The availability of genomic sequence information for the B. melitensis 16M and B. suis 1330 strains opened the way to a variety of predictions, ranging from the coding sequences delimitation to predictions of function and comparative genomics. In this chapter we briefly summarise the functional predictions made from the putative coding sequences of B. melitensis as obtained using homology search algorithms. Comparisons were made to information stored in a database, which also contains predictions of subcellular localisation and tridimensional structure. The conservation of each putative coding sequence in other a Proteobacteria genomes is also reported, allowing the first steps into comparative genomics within this diversified family. The database described here also allows simple genome-wide investigations, and some examples of such analyses are described. The database is user-friendly and would benefit from a close interaction with experimental work.
Chapter 5
Comparative Genomics of Brucella melitensis, B. suis, and B. abortus
Shirley M. Halling, Cynthia Gibas and Stephen Boyle
Abstract
The genomes of the classical Brucella species and their biovars have two chromosomes with the exception of B. suis biovar 3 strain 686 which has a single chromosome. The larger chromosome has approximately 2.1 Mbp and has a bacterial origin of replication. The smaller chromosome has approximately 1.2 Mbp and has plasmid replication functions. There is a large inversion within the small chromosome of B. abortus biovars 1, 2, and 4. There is a single large unique genetic island among the genomic sequences of B. melitensis, B. suis, and B. abortus. This island resides in the small genome of B. suis and encodes homologous of transfer functions and phage related genes. Given the high similarity among the genomic sequences of Brucellae, differences among them with regards to host preference, virulence and infectious cycle could be due to subtle variations in the conserved DNA and differential expression of conserved genes, rather than due to unique genomic DNA fragments. Detailed comparative sequence analysis identified common and unique regions and diverged regions within conserved genes, and suggests sequence targets to be used in a comparative approach to functional genomics experiments.
Chapter 6
Comparative Proteomics of Brucellae Species
Vito G. DelVecchio and Cesar V. Mujer
Abstract
The field of proteomics is the most prominent post-genomic discipline. Although this field is relatively young compared to other established sciences, proteomics has provided vast amounts of information crucial to a comprehensive understanding of system biology in various organisms. Proteomics provide the necessary tools to elucidate the interplay of metabolism, function and phenotypic expression. Ultimately proteomics will aid in the understanding of evolutionary relatedness and interrelationships among various species. Comparative proteomics has been used for identifying, characterising and comparing the proteomes of several nomen species of Brucellae. A global overview of B. melitensis proteomes has been accomplished in a relatively short period of time primarily because a completely sequenced and annotated genome of this organism is now available. Consequently, proteins that are differentially expressed between the virulent 16M and vaccine Rev 1 strains of B. melitensis have been identified and have led to a better understanding of key metabolic pathways that may be crucial in attenuating virulence and the production of vaccine strains. Soluble and membrane-bound proteomes of all the recognised nomen species (B. melitensis, B. abortus, B. suis, B. ovis, B. canis and B. neotomae) are currently being examined to identify proteins involved in host preference and virulence. Additionally, the secretomes of B. abortus and attenuated virB mutants have been investigated. Protein identification in these mutants will allow identification of protein candidates for use in vaccine development and as potential drug targets.
Chapter 7
Brucella Virulence: A Matter of Control
Jean-Jaques Letesson and Xavier De Bolle
Abstract
The publication of whole genomic data from several Brucella species allowed us to make a survey of the genomic information available on the regulation of gene expression and hence on the possible control of virulence. We identified 146 transcriptional regulators in the genome of B. melitensis. More than fifty percent of them are localised on the small chromosome that, however, represents only one third of the total genome. Concerning the two-component signal transduction systems, B. melitensis contains nineteen predicted histidine kinases and twenty one predicted response regulators that are able to form between ten and twelve pairs. The unmatched ones either belong to a phosphorelay or are orphans. A phosphoenolpyruvate-dependent phosphotransferase transduction pathway seems also present and predicted to function in a way resembling the phosphotransferase systems of Gram-positive bacteria. Concerning the specialised global regulators, B. melitensis has five sigma factors in addition to the housekeeping sigma 70 factor, two quorum-sensing linked regulators and one regulator involved in the stringent response. Where possible, we compared the organisation of the mutated locus in the aProteobacteria and tried to predict putative functions by using as the starting point the previously reported attenuation of B. melitensis mutated in regulatory factors.
Chapter 8
Brucella Lipopolysaccharide: Structure, Biosynthesis and Genetics
Maite Iriarte, David González, Rose M. Delrue, Daniel Monreal, Raquel Conde, Ignacio López-Goñi, Jean-Jacques Letesson, and Ignacio Moriyón
Abstract
Brucella lipopolysaccharide is one of the key molecules involved in Brucella virulence. Despite this, its structure has not being fully elucidated. This structure was reviewed on the light of structural and genomic comparisons, with emphasis on the core and lipid A sections. Brucella has homologues of the ORFs necessary to synthesise lipid IVA and to acylate it with long-chain fatty acids. Despite the overall similarity with R. leguminosarum, the only homologue to those that modify the disaccharide backbone of the latter is a putative phosphatase which may act at position 1 to generate monophosphoryl lipid A. The comparisons support the absence of heptose, galacturonic acid and phosphate in the core, or of arabinosamine in lipid A. An ORF putatively involved in phosphoethanolamine transfer was found in B. melitensis and B. suis but carried a missense mutation in B. abortus, suggesting species differences. The evidence supports the hypothesis that Brucella is unique among aProteobacteria in that the lipopolysaccharide core lacks all negatively charged groups but 3-deoxy-D-manno-2-octulosonic acid while keeping amino-compounds. It is postulated that this reduced negative charge plus the presence of long chain fatty acids are the basis for the altered pathogen-associated molecular pattern of Brucella lipopolysaccharide.
Chapter 9
Brucella Cyclic b-1,2-Glucans: Structure, Biosynthesis, Biological Activities and Role in Virulence
Nora Iñón de Iannino and Rodolfo A. Ugalde
Abstract
Cyclic b-1,2-glucans are unique periplasmatic polysaccharides produced by Brucella, Rhizobium, and Agrobacterium, bacteria that belong to the Proteobacteria group and display intimate interactions with plant or animal eukaryotic cells. Cyclic b1,2-glucans are homopolymers with a degree of polymerisation ranging from 17 to 24 glucose residues substituted with a variety of non-glycosidic residues. Under low osmolarity conditions cyclic glucans are highly accumulated in the periplasmatic space reaching values ranging from 5 to 20 % of the total cellular dry weight, thus suggesting that they may play a role in adaptation and/or protection of the bacterial cells under certain environmental conditions. Moreover, Rhizobium and Agrobacterium cyclic b-1,2-glucans are required for effective plant interaction (symbiosis and virulence, respectively). In addition, Brucella cyclic b-1,2-glucan mutants displayed reduced intracellular multiplication and virulence, thus suggesting that cyclic b-1,2-glucan may be a common feature required for effective interaction of the bacteria with animal and plant cells. Cyclic b-1,2- glucans are synthesised by a glucosyl transferase, cyclic glucan synthase (Cgs), that uses UDP-glucose as sugar donor. The synthesis proceeds through a novel mechanism in which Cgs, an integral membrane protein of 300 kDa, acts itself as protein intermediate and contains all the enzymatic activities required for the synthesis: i.e. initiation, elongation and cyclization. Cyclic glucans are synthesised on the cytoplasmatic side of the inner membrane and secreted into the periplasmic space by inner membrane-bound ABC-transporters (S. meliloti ndvA, A. tumefaciens chvA or B. abortus cgt). A. tumefaciens, S. meliloti and B. abortus cyclic glucan ABC-transporters are highly similar and functionally interchangeable. Moreover ndvA, chvA and cgt null mutants, unable to transport cyclic glucan into the periplasm, display defective host interaction, thus suggesting that secretion into the periplasmic space is required to exert its action. Little is known on the mechanism of action of cyclic glucan. It has been proposed that the glucan itself may be required for effective cell attachment. Other authors have proposed that the observed phenotype might be associated to the protection of the cells under stressing conditions, as for example low osmolarity. However, several evidences suggest that the mechanism may be more complicated than that originally thought. For example several pleiotropic effects on the structure and function of cellular envelope were observed in cgs mutants: i) defective assembly of the flagella in motile species like Rhizobium and Agrobacterium, ii) increased sensitivity to antibiotics and detergents, and iii) reduced stability of several components of the type IV secretion system. These results suggest that cyclic b-1,2-glucan may be involved in the correct assembly of membrane macromolecular structures required for effective host interaction.
Chapter 10
The Brucella bvrS/bvrR and Related Two-Component Regulatory Systems of the a-2 Proteobacteria: Common Regulatory Strategies of Animal and Plant Pathogens and Endosymbionts.
Ignacio López-Goñi, Lorea Manterola and Shen Q. Pan
Abstract
The a-2 Proteobacteria include pathogenic intracellular and extracellular bacteria of animals and plants as well as plant endosymbionts sharing conserved two-component regulatory systems. Brucella bvrS/bvrR is similar to the chvG(ExoS)/ChvI of Agrobacterium and Sinorhizobium, and to the BatS/BatR putative system of Bartonella. Moreover, the structure around the genes encoding these systems is essentially the same. Dysfunction of these systems alters the cell envelopes: bvrS/bvrR mutants display increased surface hydrophobicity, sensitivity to bactericidal peptides and detergents, and altered lipid A acylation; chvG mutants are sensitive to detergents and acidic pH. Moreover, bvrS/bvrR regulates transcription of at least Omp25 and Omp22, chvG/ChvI regulates the acidic pH-inducible outer membrane protein Aop (and also genes katA, virB and virE), and S. meliloti ExoS regulates the production of succinoglycan. These systems are critical in the interaction with eukaryotic cells: bvrS and bvrR mutants are avirulent in mice, sensitive to serum, and hampered in cell penetration and intracellular trafficking; chvG and chvI mutants lack tumour inducing ability; and S. meliloti succinoglycan-deficient mutants cannot establish symbiosis. These similarities suggest that these systems have a common ancestor that, while evolving to sense pericellular or intracellular signals, has kept the control of cell surface characteristics. Moreover, they suggest a co-evolution of the systems and the cognate surface molecules.
Chapter 11
Erythritol Metabolism and Virulence in Brucella
Juan M. García-Lobo and Félix Sangari García
Abstract
Erythritol metabolism by bacteria in the genus Brucella has been recognised from long time ago as a peculiar trait related to the capability of the members in this genus to produce abortions in ruminants. This association has been explained by the high concentration of erythritol in foetal tissues linked to the capability of Brucella to use erythritol preferentially over other nutrients. The pathway for erythritol degradation in Brucella is well known and more recently the genes encoding an operon for erythritol catabolism have been identified and characterised. This finding, along with the availability of genomic data and genetic tools, allows a closer analysis of the mechanisms underlying the role of erythritol metabolism in the virulence of Brucella.
Chapter 12
Iron Metabolism in Brucella
R. Martin Roop II, Bryan H. Bellaire, Eric Anderson and James T. Paulley
Abstract
Like most pathogenic bacteria, the Brucella spp. encounter extreme iron deprivation in their mammalian hosts. Two siderophores have been described for these bacteria, the simple catechol 2,3-dihydroxybenzoic acid (2,3-DHBA) and the more complex 2,3-DHBA-based siderophore brucebactin. Experimental evidence has clearly shown that 2,3-DHBA production is required for the wild type virulence of Brucella abortus 2308 in the reproductive tract of pregnant ruminants. In contrast, lack of catechol siderophore production does not compromise the capacity of this bacterium to establish and maintain chronic spleen infection in the mouse model. Surveys of the genome sequences of Brucella melitensis 16M and Brucella suis 1330 and experimental evidence obtained in the laboratory suggest that transport systems allowing the Brucellae to utilise heme and ferric dicitrate as iron sources may be important for sustaining the intracellular lifestyle of the Brucellae in host macrophages. The accumulation of excess intracellular iron can enhance oxidative damage to bacterial cells as a consequence of Fenton chemistry. Iron toxicity could be particularly problematic for the Brucellae, as oxidative killing appears to be one of the primary mechanisms employed by host phagocytes to control the intracellular replication of these bacteria. Accordingly, experimental evidence and analysis of genome sequence data indicate that the Brucella spp. employ multiple transcriptional regulators including a ferric uptake regulator (Fur) to strictly control iron uptake in response to intracellular iron levels.
Chapter 13
Brucella Interaction with Membrane Lipids of the Host Cell
Masahisa Watarai
Abstract
Brucella species are facultative intracellular bacteria capable of surviving inside macrophages and are thought to actively modify their phagosomes to avoid lysosomal fusion for intracellular survival. Both entry and intracellular growth of Brucella are dependent on interaction with microdomains of the cellular membranes. These microdomains, commonly referred to as lipid rafts, are enriched in glycosylphosphatidylinositol (GPI)-anchored proteins, glycosphingolipids and cholesterol. Lipid raft-associated molecules are selectively incorporated into phagosomes containing Brucella and treatment of raft-disrupting agents inhibits bacterial internalisation and intracellular replication. As lipid rafts participate in the signalling pathway in immune cells, entry processes associated with lipid rafts may lead Brucella into compartments that avoid fusion with the lysosomal network in the early stage of infection. Plasma membrane cholesterol of macrophages is also required for bacterial proliferation in mice. Thus, lipid raft microdomains not only influence bacterial internalisation and intracellular replication, but also contribute to the establishment of Brucella infection.
Chapter 14
Invasion, Intracellular Trafficking and Replication of Brucella Organisms in Professional and Non-Professional Phagocytes
Edgardo Moreno and Jean-Pierre Gorvel
Abstract
Brucella organisms are intracellular parasites of mammals, including humans. Initially bacteria seem to bind lipid rafts and to different membrane receptors of macrophages. In these cells, the cyclic-AMP/protein kinase-A pathway is activated followed by phosphorylation of transcription factors. In epithelial cells, the bacterium activates small GTPases of the Rho subfamily and attains a modest recruitment of actin cytoskeletal structures. In macrophages, most of the ingested Brucella are routed to phagolysosomes and only a few bacteria arrive at endoplasmic reticulum, compartment that constitutes the Brucella replicating niche. In epithelial cells, on the contrary, most of the bacteria are directed to the endoplasmic reticulum and not to lysosomes. For internalisation, Brucella requires the competence of the bvrS/bvrR regulatory system, while the virB type IV secretion apparatus is needed for intracellular trafficking. The expression of stationary phase genes seems to be required during the replicating stage. During this period Brucella are able to prevent apoptosis. Replicating bacteria release large quantities of lipopolysaccharide (LPS) within the host cells. The LPS recycles to the cell membrane forming stable complexes with MHC-II proteins in lipid mega-rafts. These surface lipid-LPS-protein macrodomains hamper the presentation of peptides to T-cells. The Brucella LPS also triggers regulatory T cells recognising MHC-II-LPS macrodomains on the surface of antigen presenting cells.
Chapter 15
The Intramacrophagic Environment of Brucella spp. and Their Replicative Niche
Véronique Jubier-Maurin, Séverine Loisel, Jean-Pierre Liautard, and Stephan Köhler
Abstract
The intracellular bacteria Brucella spp. replicate within their host cell, the macrophage. This relationship is slowly getting elucidated, especially the characterization of the compartment containing the pathogen, and the adaptive response of the latter to this environment. Entry of Brucellae via LPS O chain-lipid raft interaction is critical and leads to inhibition of early phagosome-lysosome fusion. Rough strains, devoid of O antigen, do not use lipid rafts for entry and cannot avoid fusion. Major stress proteins allow the pathogen to resist at this stage of infection. In a second step, activation of specific genes such as virB in an acidic compartment deprived of amino acids results in the establishment of the replicative niche termed "brucellosome". Escape into this final compartment allows the onset of multiplication and necessitates expression of genes we defined as the intramacrophagic "virulome". Its analysis allows an indirect characterization of the brucellosome, resulting in the conclusion that the environment encountered by Brucella inside the macrophage is deprived in amino acids, poor in nutrients, and characterised by low oxygen tension. Brucellae live in a specific, isolated niche, and infection does not seem to affect the main functions of the cell. These observations contributed to their nickname, "stealthy bacteria".
Chapter 16
Host Cellular Immune Responses Against Brucella spp. Evaluated Using the Mouse Model
Cynthia L. Baldwin and Radhika Goenka
Abstract
Brucella spp. are able to survive phagocytosis by several adaptations to intracellular life including the ability to prevent phagolysosomal fusion in professional phagocytes such as macrophages and dendritic cells and by activation of a set of genes in response to the acidic environment. In this way they are functionally related to other intracellular prokaryotes such as the acid-fast bacteria Mycobacterium tuberculosis and intracellular eukaryotes such as the protozoa Leishmania major and Toxoplasma gondii that also live within membrane-bound vacuoles in phagocytic cells albeit using different mechanisms to survive in such an environment. Because these organisms are replicating inside of host cells it is logical that an effective immune response must include a cell-mediated immune response. Immunity is based most importantly on production of interferon-g (IFN-g) since in its absence mice die of brucellosis. Production of IFN-g is controlled by IL-12 in vivo and its effective functioning for activation of macrophages depends upon TNF-g. It is likely that IFN-g is made by both CD4 and CD8 T cells in response to infections with the attenuated B. abortus strain 19 but only by CD4 T cells in response to the virulent strain 2308. Both reactive oxygen intermediates and nitric oxide contribute to control within macrophages and IFN-g serves to increase anti-Brucella activities. The involvement of CD8 T cells in control of infections with attenuated but not virulent strains of B. abortus and the implications with regard to cross-presentation of antigens is discussed.
Chapter 17
Development of New Brucella Vaccines by Molecular Methods
David L. Hoover, Mikeljon P. Nikolich, Mina J. Izadjoo, Richard H. Borschel, and Apurba K. Bhattacharjee
Abstract
Brucellosis causes substantial morbidity in humans and exacts a considerable economic toll on both the health care and livestock industries. Recently, the long-recognised potential use of Brucella as a bioweapon coupled with growing concern over bioterrorism has led to its classification by the United States Centers for Disease Control as a Category B biological threat agent. Elimination of brucellosis in food animals is the preferred method to prevent naturally acquired disease in humans, but this approach would not protect against illicit use of the organism as a bioweapon. Development of a human vaccine would be valuable both as a biodefense strategy and as an interim solution for prevention of naturally acquired human disease in situations where economic or sociological factors prevent application of an effective animal disease control program. Indeed, the need to protect occupationally high-risk workers and other susceptible populations has led to a number of efforts at human vaccine development. These efforts have met with variable success, but none has resulted in a well-accepted product. The recent explosion in genetic information on Brucella as a result of complete sequencing of the B. suis, B. melitensis and B. abortus genomes may provide an opportunity to make safe and effective vaccines for humans and improve on those available for animals. Novel molecular based vaccine strategies, which may include DNA vaccination, new adjuvants and selection of immunogens based on genomic and proteomic screening methods may lead to development of subunit vaccines, with potential safety advantages over living organisms. In the near future, however, live attenuated vaccines will be the gold standard for efficacy. In this chapter, we will focus on live, attenuated human vaccines. We will review previous human studies and examine aspects of interactions between Brucella and its host that are relevant for development of appropriate animal models for testing and for vaccine design. We will then discuss work in nonhuman primates and summarise results from our own program, which has established mucosal challenge models of infection to assess vaccine efficacy and uses molecular methods to develop live, attenuated vaccines. Finally, we will survey recent work that identifies other interesting targets for vaccine development.
Chapter 18
Alternative Ovine Brucellosis Vaccine: Experiences with Drug Delivery Systems
Carlos Gamazo, Maite Estevan, Maribel Murillo, M. Jesús Grilló, Clara M. Marín, Montserrat Barberán, M. Mar Goñi, José M. Blasco, and Juan M. Irache
Abstract
It is generally assumed that live vaccines are better than inactivated or acellular vaccines, being less expensive and inducing a more persistent immunity. However, the microencapsulation of antigens in biodegradable polymers may turn the balance in favour of the last ones. In this chapter, we describe the properties of drug-vaccine delivery systems based on the use of such polymers, and its application in the control of experimental brucellosis. We have demonstrated that micelles of major membrane antigens from Brucella ovis (HS extract) can be microencapsulated in poli-(e-caprolactone), and that this preparation (HS-PEC) induce a protective effect against an experimental challenge against experimental brucellosis in mice and rams. The resulting microparticles displayed sub-3 mm sizes, containing unaltered significant amounts of the antigenic complex, being released from the microparticles in a pulsated way. These microparticles were injected orally or subcutaneously in BALB/c mice in order to observe the protection conferred against experimental infection with the virulent strains B. abortus 2308 or B. ovis PA. The results showed that subcutaneous administration of HS-PEC microparticles eliciting high amounts of IFN-g and IL-2 but low quantities of IL-4, and protected mice against any of the challenge strains used. Such protection was similar to that provided by the reference living attenuated B. melitensis Rev 1 vaccine. Similarly, oral immunization was also able to protect mice challenged with B. ovis. Additional research was performed in rams, where HS-PEC was innocuous and, in contrast to Rev 1 immunized animals, did not induce antibodies against smooth lipopolysaccharide. To establish the protective value of this rough subcellular vaccine, rams were challenge with B. ovis, rendering a similar protection than that conferred by Rev 1. In conclusion, protection against experimental infection in mice and rams after one single shoot, and its potential for mucosal vaccination, suggest that HS-PEC microparticles may represent a serious alternative to the conventional attenuated vaccines against brucellosis.
Current Books: