Host Susceptibility to Brucella abortus Infection Is More Pronounced in IFN-γ knockout than IL-12/β2-Microglobulin Double-Deficient Mice

Brucella abortus is a facultative intracellular bacterial pathogen that causes abortion in domestic animals and undulant fever in humans. IFN-γ, IL-12, and CD8+ T lymphocytes are important components of host immune responses against B. abortus. Herein, IFN-γ and IL-12/β2-microglobulin (β2-m) knockout mice were used to determine whether CD8+ T cells and IL-12-dependent IFN-γ deficiency would be more critical to control B. abortus infection compared to the lack of endogenous IFN-γ. At 1 week after infection, IFN-γ KO and IL-12/β2-m KO mice showed increased numbers of bacterial load in spleens; however, at 3 weeks postinfection (p.i.), only IFN-γ KO succumbed to Brucella. All IFN-γ KO had died at 16 days p.i. whereas death within the IL-12/β2-m KO group was delayed and occurred at 32 days until 47 days postinfection. Susceptibility of IL-12/β2-m KO animals to Brucella was associated to undetectable levels of IFN-γ in mouse splenocytes and inability of these cells to lyse Brucella-infected macrophages. However, the lack of endogenous IFN-γ was found to be more important to control brucellosis than CD8+ T cells and IL-12-dependent IFN-γ deficiencies.


Introduction
Brucella is a Gram-negative bacterium which is pathogenic to humans and animals [1]. The establishment on infection depends of entrance of this bacterium through the nasal, oral, and/or conjunctival mucosa. After entering into the host cells, Brucella has the ability to infect and multiply in phagocytic and nonphagocytic cells [2,3]. However, macrophages are considered the main cells of Brucella residence in the host [4]. The immune response against Brucella infection involves many molecules and cells to trigger a Th1 immune response and activation of CD8+ T cells [5][6][7].
IFN-γ is a critical cytokine for host control of Brucella infection [8][9][10]. The importance of IFN-γ to control Brucella was first shown in vivo with monoclonal antibodies that depleted or neutralized IFN-γ in mice [10][11][12]. Subsequently, a more dramatic role was shown by using IFN-γ KO mice when both BALB/c and C57BL/6 mice died after infection with B. abortus strain S2308 [8]. CD4+ T cells are the major producers of IFN-γ in brucellosis, although other subsets such as CD8+ T cells also contribute [7,13]. A number of studies have demonstrated a role for either CD4+ or CD8+ T cells in the control of brucellosis [7,14]. In adoptive transfer studies, CD8+ and CD4+ T cells have been shown to be equally protective for resistance to infection with virulent B. abortus [15]. Using β2-microglobulin (β2-m) gene KO mice, our group has demonstrated that CD8+ T cells have an additional role of lysing infected macrophages and thus either killing intracellular Brucella or exposing them to IFNγ-activated macrophages [7].
IL-12 is a proinflammatory cytokine that has a profound effect on the induction of IFN-γ-producing type 1 pattern of immune response during Brucella infection [9,16]. Since CD8+ T cells and IL-12 are important immunological components during brucellosis, we decided to investigate the course of Brucella infection in IL-12/β2-microglobulin double KO mice compared to IFN-γ KO animals. This study was designed to determine whether CD8+ T cells and IL-12-dependent IFN-γ deficiencies would be more critical to control B. abortus infection compared to the lack of endogenous IFN-γ. Our results revealed that IFN-γ and IL-12/β2-m KO mice died from Brucella infection. However, all IFN-γ KO were dead at day 16 postinfection (p.i.) whereas death within the IL-12/β2-m KO group was delayed and occurred at day 32 until day 47. These results suggest that lack of endogenous IFN-γ is more important than CD8+ T cells and IL-12-dependent IFN-γ deficiencies to control murine brucellosis.

2.2.
Bacteria. Brucella abortus S2308 strain was obtained from our laboratory collection [17]. The strain S2308 was grown in Brucella Broth liquid medium (BB) (DIFCO) at 37 • C under constant agitation (200 rpm). After three days of growth, the bacterial culture was centrifuged and the pellet was resuspended in saline (NaCl 0.8% wt/vol), divided in aliquots, and frozen in 20% glycerol (vol/vol). Aliquots of these cultures were serially diluted and plated on BB medium containing 1.5% bacteriological agar (wt/vol). After incubation for 72 hours at 37 • C, bacterial numbers were determined by counting colony forming units (CFU).

Infection and Brucella Counting in Spleens.
Five mice of each strain (IL-12/β2-m −/− , IFN-γ −/− , or C57BL/6) were infected intraperitoneally with 1 × 10 6 CFU of B. abortus virulent strain S2308. These mice were sacrificed at 1-and 3weeks after infection. The spleen harvested from each animal was macerated in 10 mL of saline (NaCl 0.8%, wt/vol), and it was used for counting of CFU and splenocyte culture. For CFU determination, spleen cells were serially diluted and were plated in duplicate on BB agar. After 3 days of incubation at 37 • C in air with 5% CO 2 , the number of colony forming units (CFU) was determined. Results were expressed as the mean log CFU of each group. The experiment was repeated three times.

Cytotoxic Assay.
To determine the cytolytic activity of splenocytes from Brucella-infected mice, we used the CytoTox 96 Nonradioactive Cytotoxicity Assay (Promega, Madison, USA) that is based on the colorimetric detection of the released levels of the LDH enzyme. Macrophages differentiated (5 × 10 5 cells/well) from IL-12/β2-m −/− , IFNγ −/− , and C57BL/6 mice were infected with B. abortus (MOI 100 : 1). After 24 hours of infection extracellular bacteria was removed. Macrophages infected were used as target cells for cytotoxic assay. Splenocytes (1 × 10 6 cells/well) obtained from IL-12/β2-m −/− , IFN-γ −/− and C57BL/6 mice at one week p.i. were used as effector cells and were cocultured with macrophages in 24 well plates in DMEM medium. Effector cells were added to target cells in duplicate at 2 : 1 ratio. Culture was maintained at 37 • C in 5% CO 2 for 24 hours, and 50 μL of supernatants were harvested and placed in 96-well flat-bottom plate. Controls for spontaneous LDH release from effector and target cells, as well as target maximum release, were also added in the experiment. The cell supernatants were assayed for lactate dehydrogenase (LDH) activity following the manufacturer's protocol. The percentage of specific lysis was calculated according to the following formula: [(Experimental−Effector Spontaneous−Target Spontaneous)/(Target Maximum−Target Spontaneous)] × 100%.

Statistical Analysis.
The results of this study were analyzed using the Student's t-test, using GraphPad Prism 4 (GraphPad Software, Inc). The level of significance in the analysis was P < 0.05.

IFN-γ −/− and IL-12/β2-m −/− Mice Showed Reduced Specific Type 1 Immune Response to B. abortus.
Protective immunity against infection by B. abortus is directly related to the induction of a type 1 pattern of immune response. IL-12 and IFN-γ are key cytokines involved in this type of immunity [20]. Thus, we evaluated the production of IFN-γ, TNFα, and NO in spleen cells from IFN-γ −/− and IL-12/β2m −/− mice. As expected, no detectable IFN-γ production was observed in IFN-γ −/− and also in IL-12/β2-m −/− mice when compared to wild-type animals at one week after infection (Figure 3(b)). Furthermore, a dramatic reduction on TNF-α and NO production was observed in IFN-γ −/− and IL-12/β2m −/− mice when compared to wild-type animals (Figures 3(a) and 3(c)). Additionally, the levels of NO produced by IFN-γ −/− cells were reduced when compared to IL-12/β2m −/− mice. These results demonstrate that type 1 cytokine profile is compromised in IFN-γ −/− and IL-12/β2-m −/− mice during Brucella infection.

IFN-γ −/− and IL-12/β2-m −/− Mice Produce Normal Levels of TNF-α Levels in Macrophages. The recognition of
Brucella by innate immune cells, such as macrophages and dendritic cells, results in activation and the concomitant production of proinflammatory cytokines [21]. In this study, we evaluated the proinflammatory cytokine production by macrophages from bone-marrow cells of IFN-γ −/− and IL-12/β2-m −/− mice when stimulated with live B. abortus or E. coli LPS. As shown in Figure 4, no IL-12 was detected in IL-12/β2-m −/− mice as expected but normal levels of this cytokine were measured in IFN-γ −/− cells. Regarding TNFα, no statistically significant difference in production of this proinflammatory mediator was detected in knockout mice compared to C57BL/6.  Figure 4: IL-12 and TNF-α production induced by B. abortus in IFN-γ KO or IL-12/β2-m KO macrophages. Bone marrow from C57BL/6, IFN-γ −/− , and IL-12/β2-m −/− mouse cells were differentiated in macrophages and stimulated with B. abortus S2308 (MOI 100 : 1) or E. coli LPS (1 μg/mL). Supernatants were harvested for measuring IL-12 (a) and TNF-α (b) after 24 hrs by ELISA. Significant difference in relation to nonstimulated cells is denoted by the symbol # and an in relation to C57BL/6 mice is denoted by an asterisk (P < 0.05). Results are means ± standard deviations of experiments performed with three animals. Data shown are representative of two different experiments.

Discussion
Typical host immunity to Brucella is based on a Th1dependent immune response. Previously immunity to intra- cellular bacteria was considered to be exclusively dependent on CD4+ T cells [22]. However, later studies have emphasized the role of CD8+ T cells in protection against Brucella infection. [7,8]. The purpose of this study was to compare the susceptibility of IFN-γ KO versus IL-12/β2m KO animals, defining the importance of these immune components on host resistance to B. abortus infection.
Previous studies have demonstrated that IFN-γ was indeed crucial for the control of Brucella infection [8,10]. Additionally, our group and others have established that CD8+ T cells are critical components of host resistance to Brucella [7,15]. Herein, we determined that IFN-γ KO mice had increased numbers of Brucella CFU compared to IL-12/β2-m KO at one week postinfection. Furthermore, all IFN-γ KO died of infection at 16 days p.i. whereas death within the IL-12/β2-m KO group was delayed and occurred at 32 days until 47 days postinfection. In contrast, at 50 days p.i., 100% of C57BL/6 mice were still alive. Ko et al. [23] have previously demonstrated that IL-12 KO mice infected with B. abortus did not control infection and maintained high plateau of bacteria; however, the animals did not die at fours week postinfection. In contrast, in our study, IL-12/β2m mice succumbed to infection as a result of combined IL-12 and β2-m deficiencies. Taken together, these results suggest that IFN-γ and IL-12/β2-m are important components to host control of Brucella infection. However, the lack of endogenous IFN-γ is more important than CD8+ T cells and IL-12-dependent IFN-γ deficiencies to induce immunity to brucellosis.
In order to determine which factors could be involved with enhanced susceptibility to Brucella infection in IFNγ KO and IL-12/β2-m KO mice, we determined the concentration of IFN-γ, TNF-α, and NO in spleen cells of these animals. Splenocytes from both KO mice stimulated with live Brucella produced undetectable levels of IFN-γ and reduced amounts of TNF-α and NO. In the case of NO, this reduction was prominent in IFN-γ KO. Recently, Norman et al. have identified IFN-γ-based mechanisms that regulate NO production [24]. Furthermore, Yagi et al. [25] have demonstrated that deletion of Gata 3 allowed the appearance of IFN-γ-producing cells in the absence of IL-12. Thus, the Runx3-mediated pathway, actively suppressed by GATA3, induces IFN-γ production in a STAT4-and T-betindependent manner. Another study using Listeria monocytogenes at low dose revealed that splenocytes of IL-12 KO mice produced only 10% of the amount of IFN-γ detected in wild-type mice in response to antigen [26]. They suggested that NK cells or other cells have the potential to produce residual but substantial amounts of IFN-γ independent of IL-12. Since Listeria-infected mice showed enhanced IL-18 expression, this cytokine may stimulate NK cells for IFN-γ production in the absence of IL-12. Additionally, Freudenberg et al. [27] demonstrated the existence of an IL-12-independent pathway of IFN-γ induction by Gramnegative bacteria in mice in which IFN-β and IL-18 act synergistically. Brucella has induced the production of IL-18 and IFN-β in mice [28,29]. Therefore, this pathway could be used to produce IFN-γ during Brucella infection in absence of IL-12. Even though IFN-γ can be produced independently of IL-12, we did not detect this cytokine in IL-12/β2-m KO spleen cells activated with live Brucella at one week postinfection. It is possible that IFN-γ increases in IL-12/β2-m KO after one week postinfection. Another possibility is that IFN-γ production by NK and other cells that are present in low numbers in spleens is underestimated when we analyzed whole splenocytes. Further, it is possible that other cell types present in other organs than spleen are responsible for residual IFN-γ production in IL-12/β2-m KO.
Macrophages are key elements in innate immune responses and recognition of Brucella components [30]. Herein, we investigated the involvement of IFN-γ and IL-12/β2m in Brucella-induced IL-12 and TNF-α production by macrophages. As expected, macrophages from IL-12/β2-m KO mice showed no production of IL-12 when they were stimulated with live Brucella compared to normal synthesis of this cytokine by IFN-γ KO and wild-type cells. As for TNF-α production, no statistically significant difference was observed between KO mouse macrophages compared to C57BL/6. Since macrophages are considered the main cells of Brucella residence in the host, we infected these cells and tested them as targets for primed splenocytes from KO and wild-type mice in a cytotoxic assay. Pathogenesis induced by Brucella is the product of a complex series of interactions between the bacteria and different components of the immune system. One interaction of interest is between CD8+ CTL and Brucella-infected macrophages. In this study, specific lysis of infected macrophages was detected in wildtype and IFN-γ KO but not in IL-12/β2-m KO mice.
IL-12/β2-m KO mice fail to assemble and express MHC class I molecules on the cell surface, and, therefore, these animals are devoid of functional CD8+ αβ T cells. Thus, the lack of functional CD8+ T cells might be the reason why we did not detect macrophage lysis by IL-12/β2-m KO splenocytes. Recently, Durward et al. [31] have identified two CD8+ T cell epitopes in B. melitensis that induced IFN-γ production and specific killing in vivo. Their work reinforced the important aspect of inducing Brucella-specific CD8+ T cells to achieve an efficient host response to this pathogen.
Collectively, we have demonstrated that IFN-γ and IL-12/β2-m are important components of host immune response to control Brucella infection. However, lack of endogenous IFN-γ is more crucial to immunity against this pathogen than lack of functional CD8+ T cells and IL-12.