Search

Brucella spp. infection is frequently acquired through contaminated aerosols. The role of interleukin-1 beta (IL-1β) in the early pulmonary response to respiratory Brucella infection is unknown. As shown here, IL-1β levels in lung homogenates and bronchoalveolar lavage fluid (BALF) of mice intratracheally inoculated with B. abortus were increased at 3 and 7 days p.i. At 7 days p.i., pulmonary CFU numbers were higher in IL-1 receptor (IL-1R) knockout (KO) mice than in wild type (WT) mice. At different times p.i. CFU in lungs and BALF were higher in mice lacking some inflammasome components (caspase-1, AIM2, NLRP3) than in WT mice. At 2 days p.i. pulmonary levels of IL-1β and CXCL1 (neutrophils chemoattractant) were lower in caspase-1/11 KO mice. At day 3 p.i., neutrophils counts in BALF were lower in caspase-1/11 KO mice than in WT mice. During in vitro infections, IL-1β secretion was lower in alveolar macrophages from caspase-1/11, NLRP3 or AIM2 KO mice than in WT controls. Similarly, IL-1β production by B. abortus-infected alveolar epithelial cells was reduced by pretreatment with a specific caspase-1 inhibitor. This study shows that IL-1R, probably through IL-1β action, and the NLRP3 and AIM2 inflammasomes are involved in pulmonary innate immune protective mechanisms against respiratory B. abortus infection.

Although inhalation of infected aerosols is a frequent route for Brucella infection in humans, it rarely causes pulmonary clinical manifestations, suggesting a mild or nearly absent local inflammatory response. The goal of this study was to characterize the early innate immune response to intratracheal infection with Brucella abortus in mice and to evaluate whether it is modulated by this pathogen. After infection with 106 CFU of B. abortus, the pulmonary bacterial burden at 7 days post-infection (p.i.) was comparable to the initial inoculum, despite an initial transient decline. Brucella was detected in spleen and liver as early as 1 day p.i. IL-1β and MCP-1 increased at 3 days p.i., whereas IL-12, KC, TNF-α, and IFN-γ only increased at 7 days p.i. Histological examination did not reveal peribronchial or perivascular infiltrates in infected mice. Experiments were conducted to evaluate if the limited inflammatory lung response to B. abortusis caused by a bacterial mechanism of TLR signaling inhibition. Whereas inoculation of E. coli LPS to control mice [phosphate-buffered saline (PBS)/LPS] caused lung inflammation, almost no histological changes were observed in mice preinfected intratracheally with B. abortus (WT/LPS). We speculated that the Brucella TIR-containing proteins (Btps) A and B, which impair TLR signaling in vitro, may be involved in this modulation. After LPS challenge, mice preinfected with the B. abortus btpAbtpB double mutant exhibited a stronger pulmonary polymorphonuclear infiltrate than WT/LPS mice, although milder than that of the PBS/LPS group. In addition, lungs from B. abortus btpAbtpB-infected mice presented a stronger inflammatory infiltrate than those infected with the WT strain, and at day 7 p.i., the pulmonary levels of KC, MCP-1, and IL-12 were higher in mice infected with the mutant. This study shows that B. abortus infection produces a mild proinflammatory response in murine lungs, partially due to immune modulation by its Btp proteins. This may facilitate its survival and dissemination to peripheral organs.

Brucella canis infection is an important cause of late-term abortion in pregnant bitches. The pathophysiological mechanisms leading to B. canis-induced abortion are unknown, but heavily infected trophoblasts are consistently observed. As trophoblasts responses to other pathogens contribute to placental inflammation leading to abortion, the aim of the present study was to characterize the cytokine response of canine trophoblasts to B. canis infection. To achieve this, trophoblasts isolated from term placenta of healthy female dogs were infected with B. canis, culture supernatants were harvested for cytokine determinations, and the load of intracellular viable B. canis was determined at different times post-infection. Additionally, cytokine responses were assessed in non-infected trophoblasts stimulated with conditioned media (CM) from B. canis-infected canine monocytes and neutrophils. Finally, cytokine response and bacteria replication were assessed in canine placental explants infected ex vivo. B. canis successfully infected and replicated in primary canine trophoblasts, eliciting an increase in IL-8 and RANTES (CCL5) secretion. Moreover, the stimulation of trophoblasts with CM from B. canis-infected monocytes and neutrophils induced a significant increase in IL-8, IL-6 and RANTES secretion. B. canis replication was confirmed in infected placental explants and the infection elicited an increased secretion of TNF-α, IL-8, IL-6 and RANTES. This study shows that canine trophoblasts produce proinflammatory cytokines in response to B. canis infection and/or to stimulation with factors produced by infected monocytes and neutrophils. These cytokines may contribute to placental inflammation leading to abortion in B. canis-infected pregnant bitches.

Both CCL20 and human β-defensin 2 (hBD2) interact with the same membrane receptor and display chemotactic and antimicrobial activities. They are produced by airway epithelia in response to infectious agents and proinflammatory cytokines. Whereas Brucella spp. can infect humans through inhalation, their ability to induce CCL20 and hBD2 in lung cells is unknown. Here we show that B. abortus induces CCL20 expression in human alveolar (A549) or bronchial (Calu-6) epithelial cell lines, primary alveolar epithelial cells, primary human monocytes, monocyte-derived macrophages and the monocytic cell line THP-1. CCL20 expression was mainly mediated by JNK1/2 and NF-kB in both Calu-6 and THP-1 cells. CCL20 secretion was markedly induced in A549, Calu-6 and THP-1 cells by heat-killed B. abortus or a model Brucella lipoprotein (L-Omp19) but not by the B. abortus lipopolysaccharide (LPS). Accordingly, CCL20 production by B. abortus-infected cells was strongly TLR2-dependent. Whereas hBD2 expression was not induced by B. abortus infection, it was significantly induced in A549 cells by conditioned media from B. abortus-infected THP-1 monocytes (CMB). A similar inducing effect was observed on CCL20 secretion. Experiments using blocking agents revealed that IL-1β, but not TNF-α, was involved in the induction of hBD2 and CCL20 secretion by CMB. In the in vitro antimicrobial assay, the lethal dose (LD) 50 of CCL20 for B. abortus (>50 μg/ml) was markedly higher than that against E. coli (1.5 μg/ml) or a B. abortus mutant lacking the O polysaccharide in its LPS (8.7 ug/ml). hBD2 did not kill any of the B. abortus strains at the tested concentrations. These results show that human lung epithelial cells secrete CCL20 and hBD2 in response to B. abortus and/or to cytokines produced by infected monocytes. Whereas these molecules do not seem to exert antimicrobial activity against this pathogen, they could recruit immune cells to the infection site.