Brucella species are Gram-negative, facultative intracellular bacteria that cause zoonotic brucellosis. Survival and replication inside macrophages is critical for establishment of chronic Brucella infection. Virulent smooth B. abortus strain 2308 inhibits programmed macrophage cell death and replicates inside macrophages. Cattle B. abortus vaccine strain RB51 is an attenuated rough, lipopolysaccharide O antigen-deficient mutant derived from smooth strain 2308. B. abortus rough mutant RA1 contains a single wboA gene mutation in strain 2308. Our studies demonstrated that live RB51 and RA1, but not strain 2308 or heat-killed Brucella, induced both apoptotic and necrotic cell death in murine RAW264.7 macrophages and bone marrow derived macrophages. The same phenomenon was also observed in primary mouse peritoneal macrophages from mice immunized intraperitoneally with vaccine strain RB51 using the same dose as regularly performed in protection studies. Programmed macrophage cell death induced by RB51 and RA1 was inhibited by a caspase-2 inhibitor (Z-VDVAD-FMK). Caspase-2 enzyme activation and cleavage were observed at the early infection stage in macrophages infected with RB51 and RA1 but not strain 2308. The inhibition of macrophage cell death promoted the survival of rough Brucella cells inside macrophages. The critical role of caspase-2 in mediating rough B. abortus induced macrophage cell death was confirmed using caspase-2 specific shRNA. The mitochondrial apoptosis pathway was activated in macrophages infected with rough B. abortus as demonstrated by increase in mitochondrial membrane permeability and the release of cytochrome c to cytoplasm in macrophages infected with rough Brucella. These results demonstrate that rough B. abortus strains RB51 and RA1 induce apoptotic and necrotic murine macrophage cell death that is mediated by caspase-2. The biological relevance of Brucella O antigen and caspase-2-mediated macrophage cell death in Brucella pathogenesis and protective Brucella immunity is discussed.
Citation: Chen F, He Y (2009) Caspase-2 Mediated Apoptotic and Necrotic Murine Macrophage Cell Death Induced by Rough Brucella abortus. PLoS ONE 4(8): e6830. doi:10.1371/journal.pone.0006830
Editor: Robin Charles May, University of Birmingham, United Kingdom
Received: April 30, 2009; Accepted: July 29, 2009; Published: August 28, 2009
Copyright: © 2009 Chen, He. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: These studies and F.C. were in part supported by a NIAID (http://www3.niaid.nih.gov/) R-21 grant (1R21AI057875-01) and a BMRC grant from the University of Michigan (http://www.umich.edu/) to Y.H., and in part by the NIH/NIAID Regional Center of Excellence for Bio-defense and Emerging Infectious Diseases Research (RCE, http://www.glrce.org/) Program. The authors wish to acknowledge membership within and support from the Region V ‘Great Lakes’ RCE (NIH award 1-U54-AI-057153). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Brucella species are Gram-negative, facultative intracellular bacteria that cause brucellosis in humans and animals . Human brucellosis remains the most common zoonotic disease worldwide with more than 500,000 new cases annually . B. abortus infects primarily cattle and is one of the common Brucella species that afflict humans . The brucellae disseminate or spread via the blood and lymphatics where they multiply inside phagocytes and cause bacteremia. Unlike many pathogenic bacteria, Brucella lacks most of the classical virulence factors such as invasive proteases, exotoxins, capsules, fimbriae, virulence plasmids, and lysogenic phages . Instead, Brucella virulence relies on its ability to survive and replicate in the vacuolar phagocytic compartments of macrophages. Brucella lipopolysaccharide (LPS) is a Brucella virulence factor . The LPS has three domains: lipid A, the core oligosaccharide, and the O antigen (O-polysaccharide, or O-side chain). The brucellae exhibit two phenotypes, i.e., smooth and rough. Smooth Brucella strains include a complete LPS, while rough Brucella strains do not contain or produce very low level O-antigen. Smooth virulent Brucella strains, e.g., B. abortus strain 2308 (S2308), inhibit programmed cell death of infected human and mouse macrophages , , . For example, smooth B. suis infection inhibits spontaneously occurring apoptosis in human macrophages , , . The inhibition of cell death facilitates the survival and replication of smooth Brucella strains inside macrophages. In contrast, many rough derivatives of B. abortus, B. melitensis, and B. suis, deficient in the LPS O antigen (O-side chain), can not survive inside macrophages and are therefore attenuated , . A number of B. abortus rough strains are cytotoxic to macrophages and induce macrophage cell death , , , , . The underlying details of the molecular mechanism of macrophage death induced by rough Brucella strains remain obscure.
Several types of programmed cell death have been defined. These include apoptosis, autophagy, and pyroptosis , , , . These categories are based on criteria such as morphological alterations, initiation of a death signal, and the involvement of caspases. The most thoroughly studied aspect of programmed cell death is apoptosis. Apoptosis is mediated by a family of evolutionarily conserved proteases known as caspases . Activation of caspases leads to cell shrinkage, nuclear condensation, and oligonucleosomal DNA fragmentation. There are two types of apoptotic caspases: initiator (apical) caspases and effector (executioner) caspases. Initiator caspases (e.g., caspase-2, 8, 9, 10) activate effector caspases (e.g., caspase-3, 6, 7) that digest specific proteins and/or activate other specific caspases (e.g., caspase-1, 4, 5, 11, 12, 13, 14). Two well-studied pathways of apoptosis are the mitochondrial (intrinsic) cell death pathway and the cell surface death receptor (extrinsic) pathway . The mitochondrial cell death pathway is characterized by increased mitochondrial membrane permeability and the release of cytochrome c . Necrosis refers to the sum of changes that have occurred secondary to cell death (e.g., apoptosis) after dead cells have reached equilibrium with their surroundings. According to recent nomenclature recommendations, necrosis is the end product of a cell death process instead of a form of cell death in itself , , .
Although macrophages are highly adept at destroying bacteria, modulation of macrophage cell death by some species of bacteria is emerging as an important pathogenesis mechanism. One such mechanism, pyroptosis, is a caspase-1-dependent pro-inflammatory cell death . Caspase-1 is required for induction of macrophage cell death infected by many microbial species including Salmonella typhimurium , , , Shigella flexneri , Legionella pneumophila , and Mycobacterium tuberculosis . The participating of caspase-2 has also been reported in Salmonella-infected macrophages . Our previous microarray study indicates smooth virulent B. melitensis strain 16 M down-regulates caspase-2 transcriptional levels and inhibits transcription of genes involving mitochondria activities at an early stage of infection . No caspases were up-regulated in strain 16M-infected macrophages. These results suggest that smooth virulent Brucella modulate macrophage cell death by inhibiting mitochondrial cell death pathway and caspase activation. The caspase gene expression profiles in macrophages infected with rough Brucella strains have not been studied.
We hypothesized that rough B. abortus strains would induce macrophage cell death through a caspase-dependent pathway. To test this hypothesis, we used rough B. abortus strains RB51  and RA1  as compared with their parent smooth virulent strain 2308 (S2308). The live attenuated rough strain RB51 serves as the official Brucella cattle vaccine and has been used in the US since 1996 . The use of RB51 as a vaccine has the advantage that it does not stimulate Brucella O antigen specific antibodies and thus does not interfere with the diagnosis of wild type Brucella infection in the field. The rough phenotype of RB51 is due to the mutation of wboA that encodes a glycosyltransferase . Compared to its parent strain S2308, RB51 has at least one additional mutation of rpoB . B. abortus RA1 is a single wboA Tn5 rough mutant of S2308 . Just as to RB51, RA1 lacks the O side chain and is attenuated. Elucidation of the mechanisms underlying of Brucella rough strains RB51 and RA1 induced macrophage response will improve our understanding of Brucella vaccine immunity and foster our goal of developing an effective vaccine against human brucellosis. The results presented here demonstrate that RB51 and RA1 induce a novel apoptotic and necrotic macrophage cell death pathway mediated by caspase-2.
RB51 and RA1 induce apoptotic and necrotic macrophage cell death
The programmed cell death of Brucella-infected macrophages was first analyzed by Annexin V (green) and propidium iodide (PI, red) staining. Fluorescein-conjugated Annexin V detects translocation of phosphatidylserine from the inner cell membrane to the outer cell membrane of cells at the early stage of apoptosis. PI stains the DNA of necrotic cells and/or cells at late stage of apoptosis . The results of fluorescence staining of infected macrophages indicated that rough B. abortus strains RB51 and RA1 induce both apoptotic (stained green) and necrotic (stained green and red) macrophage cell death. At 24 h post infection with RB51 or RA1, 61.5%±8.7% or 74.7%±5.6% of the infected macrophages, respectively, underwent either apoptotic or necrotic cell death (Figure 1A, Table 1). The cytopathic effect of rough Brucella on macrophages was further confirmed by the lactate dehydrogenase (LDH) release assay, a nonradioactive cytotoxicity assay used to monitor the LDH release from dying membrane-damaged macrophages (Figure 1B).
(A) RB51 and RA1 induced 61.5%±8.7% and 74.7%±5.6% macrophage cells death, respectively, at a MOI of 200 at 24 h post infection. B. abortus strain 2308 induced limited macrophage cell death even at the MOI of 2,000. (B) LDH release from RB51, RA1 and S2308 infected macrophages. The data represent the means±standard deviations for three independent experiments. The asterisk (*) represents the significant differences (P<0.05) of the LDH release level from macrophages infected by smooth S2308 compared to that from macrophages infected with rough strain RB51 or RA1. (C) Growth kinetics of RB51, RA1 and S2308 in macrophages. The number of internalized cells of S2308 at MOI 2,000 found was similar to the number of cells of internalized RB51 and RA1 at a MOI 200 at 1 h post infection. The asterisk (*) represents the significant differences (P<0.05) in survival level of smooth S2308 compared to that of rough strain RB51 or RA1 inside infected macrophages.
The kinetic profiles of intracellular survival of different Brucella strains were analyzed in parallel with the cell death of infected macrophages (Figure 1C). Macrophage cells were infected with RB51 or RA1 at a multiplicity of infection (MOI) of 200 that has been commonly used for Brucella infection studies , . Neither RB51 nor RA1 survived inside macrophages as demonstrated by continuous decline of survived bacterial numbers inside macrophages. In contrast to these rough Brucella strains, a less amount of smooth strain 2308 (S2308) cells was taken up by macrophages at a MOI of 200 (data not shown). This observation is consistent with other reports , . At 1 h post infection, the number of internalized S2308 cells found inside macrophages at the MOI of 2,000 was similar to the number of internalized RB51 or RA1 cells at a MOI 200 (Figure 1C). At 24 h post infection, the number of living S2308 cells was significantly higher than that of S2308 at 6 h post infection and that of RB51 or RA1 at 24 h post infection. These results indicated that S2308 survived and replicates inside macrophages. However, S2308 failed to induce any macrophage cell death, even with a MOI of 2,000, at 24 h post infection (Figure 1A and 1B).
Programmed cell death of Brucella-infected macrophages was confirmed and further analyzed by Hoechst 33342 and Annexin V straining (Figure 2). Apoptotic cells were detectable as early as 2 h post infection with RB51 (Figure 2). The level of cell death increased to about 16.3%±2.1 at 4 h post infection. Approximately, 19.5%±2.9% macrophage cells exhibited both Annexin-V and PI positive (necrotic cell) among the dead cells population at 4 h post infection. At 6 h after infection, 26.6%±3.5% of cells was membrane-damaged macrophages as established by a lactate dehydrogenase (LDH) releasing test (Figure 1B). Of the dying cells, 70.3%±5.9% were necrotic or late-apoptotic cells. Others are early apoptotic cells (Figure 2A). RB51-induced apoptotic macrophage cell death was confirmed by staining with Hoechst 33342. The RB51- and RA1-infected macrophages exhibited typical signs of apoptosis, i.e., condensed nuclei and shrunken cells (Figure 2B).
(A) Annexin V/PI staining shows both apoptosis (solid arrow) and necrosis (hollow arrow) of macrophages at as early as 2 h post infection. (B) Staining with Hoechst 33342 shows cell shrinkage and nuclear condensation (diamond head arrow) of infected macrophages at 6 h post infection at a MOI of 200. The results are a representative of three independent experiments.
To specifically study Brucella internalization by individual macrophages, a recombinant strain RB51GFP that expresses the green fluorescent protein (GFP)  was used to infect macrophages. Over 95% of macrophages were infected with RB51GFP at any tested MOI (20, 200, and 2000) at 1 h post infection (Figure 3A). The bacteria infection rate for individual macrophages was uneven in any tested MOI. Some macrophages were infected with more bacteria than others. With a low MOI of 20, the majority of RB51GFP-infected macrophages contained 2–12 green fluorescence dots per macrophage (Figure 3A). The higher MOIs usually resulted in increased numbers of infected bacteria per macrophages. The PI staining of RB51GFP-infected macrophages allowed us to simultaneously monitor macrophage cell death and Brucella numbers inside macrophages (Figure 3B). It was found that most cell death occurred in Brucella-infected macrophages. For example, in macrophages infected with RB51GFP for 6 h (MOI: 200), most dying macrophages contained intracellular bacteria shown by green fluorescence (Figure 3B). A small number of PI-positive dying cells were not surrounded by green fluorescence probably due to the disruption of cell membrane leading to the release of bacteria. This possibility was observed by comparing the images of light microscopy and fluorescence microscopy.
(A) Uneven macrophage internalization of RB51GFP with different infection MOIs observed at 1 h post infection. (B) Most cell death occurred in macrophages infected with RB51GFP (MOI: 200) at 6 h post infection. The results are a representative of three independent experiments.
To further determine the relationship between the infection dose of Brucella and macrophage cell death, a kinetic experiment was conducted (Table 1). In this experiment, macrophages were infected with different MOIs and cell death evaluated at different time points (4, 8, 24, and 48 h) following infection. It was found that the level and speed of macrophage cell death were dependent on the infection doses of rough Brucella. The macrophage cell death induced by RB51 with the MOI of 2,000 developed much faster than that with a MOI of 200 or 20 (Table 1). Heat killed RB51 (HK-RB51) only caused limited cell death (∼10%) at a MOI of 2000 at 24 and 48 h post infection. S2308 did not induce obvious cell death at any tested MOI.
To test if the same cell death also occurs in bone marrow derived macrophages (BMDM) infected with rough Brucella strains, BMDM were prepared using BALB/c mice and infected with different Brucella strains. Live RB51 or RA1 was also capable of inducing apoptotic and necrotic cell death in BMDM. Heat-killed RB51 or RA1 did not cause obvious cell death with a MOI of 200. It suggests that viable rough strains are required to induce macrophage death. Live smooth strain S2308 did not induce cell death of infected BMDM (Figure 4).
RB51 induces apoptosis and necrosis in primary peritoneal macrophages isolated from RB51-vaccinated mice
Vaccine strain RB51 has been widely used for studying protective mechanisms against brucellosis using mice , . In order to determine if RB51 or S2308 induces similar cell death in primary macrophages in RB51-infected mice, BALB/c mice were injected intraperitoneally (i.p.) with RB51, HK-RB51 or S2308 at a dose of 5×108 CFU. This is the same dosage and injection route used for other typical RB51 mouse studies , . Peritoneal macrophages were collected and monitored for induction of apoptosis and necrosis. In total, 38.4%±4.0% of the peritoneal macrophages from RB51-vaccinated mice showed apoptotic or necrotic staining at 24 h after infection (Figure 5). Macrophages infected with Brucella S2308 appeared to be activated as demonstrated by the larger size and irregular shape with long pseudopodia, as compared with noninfected macrophages that showed smaller size and oval shape with less short cytoplasmic processes. Similar observation was found in macrophages treated with HK-RB51. In addition, no obvious cell death was observed in macrophages from control mice injected with saline, HK-RB51 or S2308. These results indicate that RB51 induces apoptosis and necrosis in primary macrophages from RB51-vaccinated mice.
Annexin V/PI staining shows both apoptosis and necrosis of peritoneal macrophages at 24 h after infection with RB51. No obvious cell death was observed in peritoneal macrophages from mice injected (i.p.) with S2308 or heat-killed RB51 (HK-RB51). The results are a representative of three independent experiments.
Inhibition of caspase-2 inhibits RB51 and RA1-induced macrophage cell death
The role of caspases in rough B. abortus-induced macrophage cell death was examined. Peptide-based substrates and inhibitors have been used extensively to monitor the association of caspases with cell death and to identify the caspases involved , . Pretreatment of macrophages with the pan-caspase inhibitor Z-VAD-FMK prior to infection with RB51 or RA1 resulted in an approximate 69.4%±4.9% or 63.2%±4.8%, respectively, decrease in macrophage cell death at 24 h post infection compared to untreated infected macrophages (Figure 6). These results suggest that RB51 and RA1-induced macrophage cell death is mediated by a caspase(s). To determine the potential role of individual caspases in the macrophage death, a series of nine caspase inhibitors were tested for inhibition of cell death. The caspase-2 specific inhibitor (Z-VDVAD-FMK) blocked macrophage cell death caused by RB51 or RA1 98.2 %±0.5% or 80.3%±5.2%, respectively, compared to untreated controls at 24 h post infection (Figure 6). The caspase-2 inhibitor appears to exert a greater influence in preventing cell death in RB51-infected macrophages than in RA1-infected macrophages; it might be attributed to the differences in the gene mutations of these two strains. The difference in inhibition of macrophage cell death provided by the caspase-2 inhibitor over that of the pan-caspase inhibitor may be attributed to the difference in the specificities of these inhibitors. The inhibitors of caspase -1, 3, 4, 6, 8, 9, 10, and 13 failed to significantly inhibit RB51 and RA1-induced macrophage cell death in our studies. These results support our hypothesis that the rough Brucella abortus strains induce macrophage cell death is mainly mediated by caspase-2.
Macrophages were pretreated with individual caspase inhibitor and then infected with RB51 or RA1 at a MOI of 200. No inhibition was observed with any of the other caspase inhibitor. The results are a representative of three independent experiments.
Caspase-2 is activated at an early stage of infection
Caspase-2 activation was analyzed by caspase colorimetric assay and procaspase cleavage assay. The caspase colorimetric assay showed that caspase-2 enzyme activity of RB51 and RA1 infected macrophages increased approximately 59.9%±8.3% (p<0.05) and 55.7%±1.9% (p<0.05) respectively at 1 h post infection. This increase was transient. It was followed by a rapid return to the base line at 2 h post infection (Figure 7A). Capsase-2 activation was analyzed further by detection of the cleavage of procaspase-2 protein . As ascertained by western blot analysis, transient decrease in the level of procaspase-2 (48 kDa) in RB51 or RA1-infected macrophages was observed at 1 h, 2 h and 6 h post infection (Figure 7B). This confirmed that the activation of caspase-2 occurs at the early stage of infection. In contrast, S2308 did not induce significant changes in the caspase-2 enzyme activity or the level of procaspase-2 (Figure 7A and B). Caspase-1 activity was also examined in RB51 or RA1-infected macrophages. No significant changes were observed (data not shown).
(A) Analysis of caspase-2 activities by a colorimetric assay. The asterisk (*) represents significant differences (P<0.05) compared to the medium control group. (B) Western blot analysis of precaspase-2 in infected macrophages. The numbers shown in the figure represent the amounts of procaspase-2 quantified by densitometry and normalized to the β-actin content. The data represent the means±standard deviations from three independent experiments.
The above results indicate that rough strains RB51 and RA1 induce similar apoptotic and necrotic cell death in macrophages and that the cell death is mediated by caspase-2. We later focused on comparing RB51 and its parent smooth strain 2308 to further study the apoptotic and necrotic cell death mechanism induced by rough B. abortus.
Inhibition of caspase-2 activity increases RB51 survival in macrophages
To confirm the inhibition of programmed macrophage cell death after treatment with the caspase-2 inhibitor, the LDH release assay was used to monitor cytopathic cells (Figure 8A). The LDH released from RB51-infected macrophages reached a value of 54.6%±6.5% of the total LDH available in both living and dying cells at 8 h post infection, and 80.5%±6.4% at 24 h post infection. In contrast, after pretreatment with the caspase-2 inhibitor, the yield of LDH released was 9.3%±6.0% and 24.7%±13.3% at 8 h and 24 h post infection, respectively.
Macrophages pretreated with Z-VDVAD-FMK were infected with RB51 for 24 h at a MOI of 200. (A) Macrophages pretreated with the caspase-2 inhibitor and infected with RB51 released less LDH than infected cells lacking inhibitor. (B) Growth kinetics of RB51 in macrophages. The number of RB51 cells surviving inside macrophages pretreated with the caspase-2 inhibitor was approximately 120 times greater than the number of cells found in untreated macrophages (P<0.05). The asterisk (*) represents significant difference (P<0.05) in the LDH release or bacterial CFUs from RB51-infected macrophages with caspase-2 inhibitor treatment compared to that without caspase-2 inhibitor. Data represent the means±standard deviations from three independent experiments.
Inhibition of caspase-2 activity also increased the number of RB51 survived inside macrophages (Figure 8B). The inhibitor did not change the phagocytosis of RB51 by macrophage cells as demonstrated by similar number of living brucellae with or without the caspase-2 inhibition at 1 h post infection. At 8 h post infection, approximately 5% of RB51 survived inside infected macrophage cells. However, after the pretreatment of the macrophages with the caspase-2 inhibitor prior to RB51 infection, approximately 36% of brucellae survived in the macrophages. At 24 h post infection, the number of RB51 surviving inside the macrophages pretreated with the caspase-2 inhibitor was approximately 120 times greater than that inside untreated macrophages (Figure 8B). This phenomenon might be related to more cells survived due to the inhibition of macrophage cell death.
Knockdown of caspase-2 expression decreases RB51-induced macrophage cell death
Using different specific caspase inhibitors, the present studies demonstrate that caspase-2 plays an important role in rough B. abortus induced macrophage cell death. To address potential concerns as to the specificity of the caspase inhibitors , the RNA interference (RNAi) technique  was used to further identify the role of capase-2. RNAi is a natural, evolutionarily conserved regulatory mechanism that is mediated by the introduction of dsRNA into the cytoplasm of a host cell. RNAi has provided a unique tool for sequence-specific silencing to develop practical strategies for studying gene function, biological processes and pathway analysis . Short hairpin RNA (shRNA) is a sequence of RNA that makes a tight hairpin turn that silences gene expression via RNAi . Macrophages with specific gene knockdown were generated by transfection of macrophages with lentivirus vectors expressing gene-specific shRNA . Four caspase-2 knockdown clones (RAW264.7-Casp2KD) and two control clones (RAW264.7-GipZ) were selected based on EGFP fluorescence detection. The protein expression of caspase-2 was further monitored by Western blot analysis (Figure 8A). RAW264.7-GipZ cells did not exhibit any difference in procaspase-2 expression compared with the original RAW264.7 cells. The expression levels of procaspase-2 protein in four RAW264.7-Casp2KD clones decreased by 34–64%. When infected with RB51 at a MOI of 200 for 6 h, the control RAW264.7-GipZ clones demonstrated a similar rate of cell death compared to the original RAW264.7 (68–74% vs. 71%). In contrast, the RAW264.7-Casp2KD clones showed a significantly lower rate of cell death rate (5–44%) (P<0.05) (Figure 9B). Of the four RAW264.7-Casp2KD clones, clone two had the lowest cell death rate (5%±0.6%) and is correlated with the lowest caspase-2 protein expression (Figure 9). These results confirm that RB51-induced apoptosis is mediated specifically by caspase-2.
Macrophages with or without caspase-2 shRNA were infected for 6 h with RB51 at a MOI of 200. (A) Caspase-2 expression in different clones was detected by Western blot analysis. The numbers shown in the figure represent the amounts of procaspase-2 that were quantified by densitometry and normalized to the β-actin data. (B) The LDH assay was used to detect differential cell death in different macrophage clones. The asterisk (*) represents significant differences (P<0.05) of LDH releases from different clones compared to that from the parent RAW264.7 cells. The data represent the means±standard deviations from three independent experiments.
Mitochondrial membrane potential (MMP) decreases following cytochrome c release
Mitochondria are viewed as the major players in both necrotic and apoptotic cell death , . A key feature of apoptotic and necrotic cascades is the initiation of the mitochondrial permeability transition (MPT) followed by the release of cytochrome c from mitochondria. We hypothesized that RB51 induced apoptotic macrophage cell death by activating a mitochondria-dependent apoptotic pathway. Increased MPT results from a decrease of MMP. Decreased MMP promotes a decrease in the retention of the lipophilic cationic dye 3,3′-dihexyloxycarbocyanine iodide (DiOC6). This phenomenon is frequently used to quantify MMP through flow cytometry  and was applied in this study. A MMP decrease of 23.5%±5.8% (P<0.05) was detected in RB51-infected macrophages at 1 h post infection. At 6 h post infection, the level of MMP decreased to 28.5%±0.7% (P<0.05) of the value of uninfected macrophages. The MMP of RB51-infected macrophages remained at a low level for 24 h post infection (Figure 10A). The MMP of S2308-infected macrophages increased 33.0%±10.5% (P<0.05) at 1 h post infection but after 2 h post infection returned to the level found with the uninfected macrophages (Figure 10A).
Infection with S2308 at a MOI of 2,000 did not cause decrease of MMP or cytochrome c release. (A) The results of flow cytometry results were summarized in graphic form, and represent the means±standard deviations of three independent experiments. The asterisk (*) represents significant differences (P<0.05) compared to the medium control group. (B) Cytochrome c release was analyzed by Western blot analysis. β-actin was used as a reference. Only cytoplasmic portion of cell lysates was assayed. The results are a representative of three independent experiments.
A direct effect of the increased MPT and decreased MMP is the release of cytochrome c from the mitochondria to the cytoplasm. To test this effect, the cytochrome c levels from the cytosol of infected and uninfected macrophages from the same experiment were analyzed (Figure 10B). Release of cytochrome c into the cytoplasm of RB51-infected macrophage was detectable as early as 2 h post infection (Figure 10B). No cytochrome c was detected in the cytoplasm of S2308-infected macrophage before 24 h post infection.
The results reported here indicated that attenuated rough B. abortus strains RB51 or RA1, but not their parent smooth virulent strain 2308, induced both apoptosis and necrosis in RAW264.7 macrophage cells, bone marrow-derived macrophages, and peritoneal macrophages. The programmed cell death induced by rough B. abortus is mediated primarily by caspase-2 and through a mitochondria-dependent cell death pathway. Since rough strains RB51 and RA1 have deficient LPS O antigen found in their parent strain S2308, the rough Brucella phenotype with deficient O antigen plays a critical role in caspase-2-mediated macrophage cell death induced by rough Brucella.
Cell death of macrophages infected with rough Brucella strains has been discovered by several groups , , , , . The discovery of cell death induced by rough Brucella was first reported by Freeman et al in 1961 . Fernandez-Prada et al later found that a rough B. melitensis mutant WRR51 lacking a wboA gene carried by smooth virulent B. melitensis strain 16 M resulted in apoptotic cell death in human monocytes from healthy volunteers . Our results are consistent with this finding. Pei et al also showed that several rough B. abortus strains obtained from a transposon mutation induced necrotic and oncotic macrophage cell death . No apoptotic macrophage cell death was observed in this report. One possible reason for the difference in the observed cell death patterns might be that different rough mutants were used in each study , . The rough phenotype can result from mutations affecting O antigen precursor synthesis, its polymerization and transport, or from defects in the inner core oligosaccharide . The structure of rough lipopolysaccharide present in each mutant may vary. Different levels of necrotic/oncotic macrophage cell death may also be induced by rough B. abortus mutants with distinct gene mutations. For example, at the MOI of 200, macrophage cell death was induced by rough strain CA180 (mutation of manA encoding phosphomannomutase) but not by rough strain CA533 (mutation of a rfbD-like gene encoding a putative GDP-D-mannose dehydratase) at 8 h post infection . The observation that RB51 and RA1 induce similar patterns of macrophage cell death suggests that the wboA gene mutation in RB51 is a primary player in the apoptotic and necrotic cell death.
The absence of Brucella O antigen may not directly contribute to the induction of macrophage cell death by rough Brucella. While RB51 does not induce detectable antibodies in animals against Brucella O antigen, RB51 has been found to produce a low but detectable level of O antigen . Gonzalez et al.  generated rough Brucella mutants by transposon mutagenesis of more than 10 genes involved in O antigen (or O chain) or the core oligosaccharide biosynthetic pathways. These rough mutants are classified into three types based on the decrease in LPS molecular weight. Therefore, rough Brucella strains are defined by deficient O antigen instead of completely deleted O antigen. The reason of macrophage cell death induced by rough Brucella does not rely on the complete deletion of O antigen. Instead, the cell death is most likely due to the pleiotropic effect of removing the majority of the O antigen that results in the change of the Brucella cell envelope structure. The LPS defects alter key topological, physiochemical and biological surface properties of rough Brucella . The outer membrane proteins and lipids in rough Brucella cells are readily exposed to the host environment, making them much easier to interact with the host cells. It has also been found that the surface of rough Brucella out membrane is highly hydrophobic and negatively charged . Such a surface may allow multiple nonspecific interactions with eukaryotic membranes. Therefore, Brucella O antigen probably acts as a negative modulator of non specific adherence which allows receptors to function . It has been observed that rough mutants penetrate more actively and display higher adherence to macrophages than smooth strains , , . Rough and smooth Brucella strains enter macrophages through different mechanisms. Virulent smooth Brucella cells use lipid rafts to enter murine macrophages while rough Brucella cells do not . We found similar numbers of internalized brucellae inside macrophages when rough strain MOI was 200 and smooth strain MOI was 2000 (Figure 1). The MOI of 200 of rough strain induced significant cell death, while smooth strain 2308 did not induce obvious cell death even with an MOI of 2000. The results also suggest that different outcomes of macrophage cell death is primarily due to the difference of rough vs. smooth phenotypes instead of the different numbers of phagocytosed Brucella cells.
Pei et al , , , ,  found that viable intracellular rough Brucella were required for inducing cytopathic effect on infected macrophages. Although heat-killed rough brucellae were internalized as efficiently as live bacteria, they did not induce macrophage cell death , , , , . Our study is consistent with the conclusion although a very high MOI of 2,000 induced limited but detectable macrophage cell death at late infection stage (Table 1). Pei et al.  further found that protein synthesis from live rough Brucella cells is required for induction of macrophage cell death. It was shown that inhibition of bacterial synthesis and replication by chloramphenicol treat prevented cell death of infected macrophage . It is interesting that viable but not killed smooth Brucella inhibit spontaneously occurring apoptotis in human macrophages , , . How live rough Brucella promotes apoptotis and live smooth Brucella inhibits apoptosis is an important question for better understanding Brucella pathogenesis and host responses against Brucella infections.
The relationship between rough Brucella-induced apoptotic and necrotic macrophage cell death may be complex. Necrosis refers to the sum of changes secondary to cell death by another mechanism such as apoptosis , , . Our study showed that macrophage cell death induced by rough Brucella was dose-dependent since more cell death was observed with higher MOIs (Table 1, Figure 3). Meanwhile, the numbers of internalized bacteria in infected macrophages might not be even. Some macrophages phagocytosed more rough bacteria than others with the same MOI (Figure 3). Macrophages infected with more rough Brucella cells tended to induce more and quicker macrophage cell death. Necrosis occurred more than apoptotis in the late stage of Brucella-induced macrophage cell death. Therefore, it is likely that apoptotic and necrotic cell death of Brucella-infected macrophages represents different stages of the same cell death pathway, from early and late apoptosis to eventual necrosis. Alternatively, a non-apoptotic pathway(s) of cell death might be induced in some infected macrophages, which results in necrosis. Further investigation is needed to test this possibility.
Caspase-2 is activated in response to embryonic development  and many biological stresses, including DNA damage, aging, UV irradiation, tropic factor withdrawal, endoplasmic reticulum stress, death receptor activation and heat shock , , . Deficient caspase-2 activation has been observed in gastric cancer, and acute myeloblastic leukemias , . Caspase-2 activation has also been observed in host cells infected with Salmonella  and Epstein-Barr virus . Salmonella typhimurium induced primarily caspase-1-mediated pro-inflammatory pyroptosis , , . The action of caspase-2 was also needed for Salmonella-induced apoptosis as shown by the results of experiments using caspase-2 knockout macrophages and chemical caspase-2 inhibitors . Unlike with S. typhimurium, rough (but not smooth) B. abortus-induced macrophage cell death exhibits caspase-2 activation and is noticeable hampered by a caspase-2 inhibitor and caspase-2-specific shRNA (Figure 6 and 9). On the other hand, the macrophage cell death induced by rough B. abortus strains RB51 and RA1 were not obviously inhibited by a caspase-1 inhibitor (Figure 6). These findings suggest that caspase-1 plays minimal role in the Brucella-induced cell death. Loss of mitochondrial integrity and release of cytochrome c do not occur during pyroptosis  but was observed in macrophages infected with RB51 and RA1. This difference indicates that pyroptosis does not play a major role in rough Brucella-induced cell death. Unlike virulent Salmonella and Shigella strains which induce apoptosis in infected macrophages, smooth virulent Brucella strains inhibit programmed macrophage cell death and replicate in macrophages , , , , . The current study further indicates that B. abortus smooth strain S2308 does not induce caspase-2 activation or macrophage death. Therefore, the caspase-2-mediated macrophage apoptosis induced by rough attenuated strains but inhibited by the parent wild type appears to be a novel findings not as yet found in other bacterial pathogen.
How caspase-2 mediates programmed cell death of macrophages infected by rough Brucella is unclear and deserves further investigations. Across species caspase-2 is one of the most conserved caspases. Its function in cell death signaling is still an enigma . Caspase-2 is unique among the caspases in that it has features of both upstream and downstream caspases. Caspase-2 contains a long CARD prodomain that can be used to interact with adaptor proteins. This is typical of initiator caspases. On the other hand, the predicted cleavage specificity of caspase-2 appears to place it to be an effector caspase similar to caspases-3 and -7 . The present studies reveals that caspase-3 or caspase-8 inhibitor slightly yet not significantly inhibits RB51- and RA1-induced macrophage cell death. The caspase-3 inhibitor (Z-DEVD-FMK) was reported to also inhibit caspase-7 activity . It is possible that caspase-2 functions as an initiator caspase and regulates apoptotic activities of other effector caspases (e.g., caspase-3, -7 and -8). It is also possible that caspase-2 regulates Brucella-induced cell death directly by acting as an effector caspase. This hypothesis is supported by the observation that caspase-3 and caspase-8 have the potential to regulate downstream caspase-2 activity , , . An alternative possibility is that caspase-2 may function as an initiator caspase and effector caspase at the same time.
The study provides new evidence to support the hypothesis that smooth virulent Brucella has the ability to control or influence mitochondrial functions, especially those related to programmed cell death in macrophage cells, via a mechanism that is a crucial survival mechanism in this pathogen. Virulent Brucella strain S2308 appears to inhibit macrophage cell death by inhibiting mitochondrial membrane transition and cytochrome c release (Figure 10). The inhibition of macrophage cell death allows Brucella survive and replicate inside macrophages. In contrast, rough Brucella strains induce macrophage cell death through the release of cytochrome c from the mitochondria. The dying macrophages release Brucella to a more hostile environment outside macrophages. It is interesting to note that the Brucella species are classified as α-Proteobacteria, the modern-day relatives of an ancient endosymbiont that gave rise to mitochondria , . Our previous study indicates virulent B. melitensis strain 16 M inhibited mitochondria activities at an early stage of infection . The interactions between mitochondrial permeability transition and caspase-2 activation in the apoptotic cell death are still unclear. It has been repeatedly proposed that caspase-2 can act upstream of mitochondrial events as an initiator that triggers cytochrome c release and subsequent apoptosome formation, either directly or through generation of tBid , , . Other studies place caspase-2 neither upstream nor downstream but suggest it acts in concert with the mitochondrial as well as the death receptor pathways of apoptosis . Our results suggest that programmed cell death in macrophages infected with rough B. abortus is at least partially due to the mitochondrial permeability transition and cytochrome c release. The relationship between mitochondrial permeability transition and caspase-2 activation in the apoptotic cell death of Brucella-infected macrophages requires further investigation.
The caspase-2-mediated cell death of macrophages infected with vaccine strain RB51 and vaccine candidate RA1 may help us to better understand the process of protective Brucella immunity. Programmed cell death of macrophages infected with many pathogens (e.g., M. tuberculosis, Salmonella, Influenza virus) plays an important role in antigen presentation to T cells through a process called cross-priming . It is likely that rough Brucella-induced macrophage cell death may play an important role in protective immunity against brucellosis through cross-priming or other mechanisms. Further understanding of Brucella-induced programmed cell death will undoubtedly help decipher Brucella pathogenesis and promote rationale Brucella vaccine development.
Materials and Methods
Bacterial strains and macrophage cell culture
B. abortus vaccine strain RB51, strain RA1 and the parent wild type strain S2308 were originally obtained from Dr. Gerhardt G. Schurig's laboratory at Virginia Tech. B. abortus strain RB51GFP that contains a GFP plasmid (pNSGroE/GFP) was a gift from Dr. Nammalwar Sriranganathan's laboratory at Virginia Tech . These bacterial strains were grown either in tryptic soy broth or on tryptic soy agar plates. Heat-killed Brucella was prepared by boiling bacteria for 20 min. The characteristics of these strains were confirmed with AMOS (abortus-melitensis-ovis-suis) PCR reactions , crystal violet staining, antibiotic resistance, urease test, and thionin and fuchsin susceptibility . To minimize naturally mutated rough phenotypes in smooth strain S2308, one single colony of smooth phenotype S2308 was selected based on crystal violet staining and used for bacterial amplification. All studies utilizing Brucella strains were performed in a Biosafety level 3 (BSL-3) laboratory. Murine macrophage cell line RAW264.7 (ATCC #TIB-71) was cultured at 37°C with 5% CO2 in complete tissue culture medium (c-DMEM) consisting of Dulbecco's modified Eagle's medium (DMEM; ATCC) supplemented with 10% heat-inactivated fetal bovine serum (HyClone, Logan, Utah). Macrophages between passages 4 through 15 were used in these experiments. Bone marrow-derived macrophages (BMDM) were prepared from BALB/C mice and were cultured at 37°C with 5% CO2 in RPMI-1640 supplemented with 10% heat-inactivated fetal serum and 30% L-Cell Medium as previously described . Bacterial and macrophage cells were maintained under conditions that sustained exponential growth.
Macrophage infection assays
Macrophage cells were cultured in 96-well, 24-well, or six-well plates. In the macrophage infection experiments, 5×104 cells per well were seeded in 96-well plate and 2.5×105 cells per well were seeded in 24-well plate. For flow cytometry assays, 1×106 cells were seeded per well in six-well plate. The macrophages were cultured overnight prior to infection with RB51, RA1 or S2308 cells. Macrophages were infected with Brucella at a multiplicity of infection (MOI) of 20, 200 or 2,000. The plates were centrifuged at 300 x g for 5 min. All incubations with Brucella were conducted at 37°C in an atmosphere containing 5% (vol/vol) CO2. After 30 min the cells were washed three times with Dubecco's phosphate buffered saline (DPBS), and incubated in fresh DMEM supplemented with 50 µg/ml of gentamicin to kill extracellular bacteria. A specific caspase inhibitor might be included in the cell culturing. For caspase inhibition studies, a caspase inhibitor sample pack (R&D systems Inc, Minneapolis, US) was used which contains a pancaspase inhibitor (Z-VAD-FMK) and nine specific caspase inhibitors including those for inhibition of caspase-1 (Z-WEHD-FMK), caspase-2 (Z-VDVAD-FMK), caspase-3 (Z-DEVD-FMK), caspase-4 (Z-YVAD-FMK), caspase-6 (Z-VEID-FMK), caspase-8 (Z-IETD-FMK), caspase-9 (Z-LEHD-FMK), caspase-10 (Z-AEVD-FMK), and caspase-13 (Z-LEED-FMK). The macrophages were pretreated with the individual caspase inhibitors at a final concentration of 20 µM for 1 h prior to infection.
Determination of programmed macrophage cell death
Macrophages were infected in the presence, or absence, of a caspase inhibitor as described above. Apoptotic or necrotic macrophages were detected using two approaches. In the first approach, cells were stained with Annexin V (green dye) and propidium iodide (PI, red dye) using an Annexin V-FLUOS staining kit (Roche Diagnostics Corporation, Indianapolis, Ind.). RB51, RA1 or S2308-infected macrophages were incubated with Annexin V and PI at room temperature for 20 min and observed by fluorescence microscopy (Nikon TE2000-S microscope). Images were photographed with an RT Slide Spot digital camera. Apoptotic and necrotic cell numbers were counted in representative fields containing at least 200 cells. In the second approach, the nuclei of RB51 infected cells and untreated control cells were stained with the blue-fluorescent Hoechest 33342 (Invitrogen, OR)at 1 µg/ml in DPBS at room temperature for 15 min and observed by fluorescence microscopy. Gliotoxin-treated cells were employed as a positive control for apoptotic cell death, and t-butyl-hydroperoxide (TBH)-treated cells served as a positive control for necrotic cell death , .
Quantitation of cell viability
Cells were cultured in triplicate in 96-well plates infected with RB51, RA1 or S2308 as described above. The culture supernatants were collected at various time points, and the lactate dehydrogenase (LDH) released determined by a CytoTox 96 nonradioactive cytotoxicity assay (Promega, Madison, Wis.) according to the manufacturer's instructions. To reduce the LDH background from fetal bovine serum, the supernatants were diluted 1∶1 with phosphate buffered saline (PBS) before the assay. The percent of membrane-damaged dying cells is expressed as a percentage of maximum LDH release, i.e., 100× (optical density at 490 nm [OD490] of infected cells - OD490 of uninfected cells)/(OD490 of lysed uninfected cells - OD490 of uninfected cells). The data presented represent the average±standard deviation of at least three separate experiments.
Assay of Brucella survival inside macrophages
Macrophages were seeded in a 24-well plate with 2.5×105 cells per well and incubated overnight as described above. RB51, RA1 and 2308 were used to infect macrophages with varying MOIs (see above). To assess intracellular survival of Brucella inside macrophages, the cells were lysed with 1 ml 0.1% (vol/vol) Triton X-100 in sterile water at selected time points (1, 6 and 24 h). The colony forming units (CFUs) were obtained by plating a series of dilutions on TSA plates . All experiments were conducted in triplicate.
Western blot analysis
Cells cultured in six-well plates were infected with RB51, RA1 or S2308 as described above. Cell lysis of the cytosolic fraction was prepared with a Mitochondria/Cytosol Fractionation Kit (Biovision) according to the manufacturer's instructions. Lysates (50 µg per sample) were analyzed by Laemmli SDS-PAGE and transferred onto Immobilon P membranes (Millipore) at 150 V for 1 h in the presence of transfer buffer (25 mM Tris, 192 mM glycine, 15% methanol, pH 8.3). After transfer, the membranes were blocked for 1 h a blocking solution (5% low fat dried milk dissolved in TBS-T (25 mM Tris-HCl, 150 mM NaCl, 0.5% Tween 20, pH 7.4) and were probed overnight at 4°C with the mouse anti-caspase-2 (Cell signaling, CA), Rabbit anti-β-actin (Biovision, CA) or mouse anti-cytochrome c (NeoMarkers, CA) antibodies. Membranes were washed with PBS-T and incubated with HRP conjugated to either goat anti-rabbit IgG or goat anti-mouse IgG (Biovision, CA) at room temperature for 1 h. After extensive washing with TBS-T, the antigen was visualized using an ECL Western Blotting Substrate (Pierce). The results were quantified by densitometry and normalized to the β-actin data. This experiment was conducted in triplicate.
Caspase-1 and -2 activity assays
Macrophages were cultured in T-25 flasks infected with RB51 or S2308. The caspase-1 and -2 activities were determined by monitoring proteolysis of the appropriate colorimetric substrates using the Caspase Colorimetric Assays (R&D Systems, CA). WEHD and VDVAD served as substrates employed to detect caspase-1 and -2 enzyme activities, respectively. All experiments were conducted in triplicate. The whole-cell lysate was added to a buffer containing 200 μM of the appropriate caspase-specific substrate conjugated to p-nitroanilide. After an incubation of 37°C for 3 h, caspase activity was quantified spectrophotometrically at a wavelength of 405 nm using a VERSAmax microplate reader (Molecular Devices).
Preparation of stable caspase-2 knockdown macrophage cell lines
Lentivirus vectors Lenti-Casp2miRNA-VSVG (Open Biosystems OligoID V2LMM_16681) used for producing anti-caspase-2 shRNA and empty lentivirus vector Lenti-GipZ-VSVG were generated at the University of Michigan shRNA Core Facility. For transduction, RAW264.7 cells were planted into six-well tissue culture plates at 2.5×105 per well. On the following day, the cell medium was aspirated off and replaced with 1.25 ml viral supernatant (MOI of 10) containing 8 µg/ml polybrene into each well. The plates were gently rocked and centrifuged at 2,500 rpm for 90 min at 30°C using an Eppendorf 5810R centrifuge. The medium was replaced after 24 h, and the cells were monitored for enhanced green fluorescent protein (EGFP) expression. To select the stably transfected cells, at 48 h after transfection, cells were scraped down and planted to 10 cm cell culture dish, selected with c-DMEM containing 10 µg/ml puromycin. Stably transfected cells were selected with the presence of puromycin for over 4 weeks. Puromycin-resistant clones were picked and expanded. The caspase-2 expression of each clone was examined by Western blot analysis.
Analysis of mitochondrial membrane potential (MMP)
The loss of mitochondrial membrane potential (ΔΨm) results in a decrease in the retention of 3,3′-dihexyloxycarbocyanine iodide (DiOC6). The ΔΨm was evaluated by measuring the retention of the lipophilic cationic dye DiOC6 in mitochondria . Briefly, cells cultured in six-well plates were infected with RB51 or S2308 in triplicate as described above. At various time points post infection, the cells were preloaded with 1.5 nM DiOC6 in c-DMEM incubated at 37°C for 20 min, washed and fixed with 1% paraformaldehyde at 25°C for 20 min. The cells were then dislodged with a rubber policeman, pelleted at 500 x g, washed, and resuspended in PBS containing 1% paraformaldehyde. Flow cytometry to detect cells with diminished fluorescence was performed under FL-1 logarithmic amplification using a FACSCanto Flow Cytometer (BD Bioscience). Macrophages stained with an irrelevant isotype antibody source served as controls.
Analysis of cell death of peritoneal primary macrophage vaccinated with RB51
Female BALB/c mice (three per treatment group) were injected intraperitoneally with 0.2 ml of a 0.15 M NaCl saline solution (controls) or 0.2 ml of saline containing approximately 5×108 CFUs of strain RB51, heat-killed RB51 (HK-RB51), or S2308. This dose was routinely used for RB51 vaccine efficacy studies , . At 2 h post infection, peritoneal macrophages were isolated from both RB51-immunized and saline-injected mice by washing the cavity with cold PBS. Cell suspensions were washed twice with cold complete tissue culture medium (RPMI 1640 with 10% fetal calf serum and 50 µg/ml gentamicin). Isolated peritoneal cells were counted with 0.4% trypan blue and plated in complete tissue culture medium at a concentration of 1×106 cells per well in a six-well plate. After incubation for 60 min at 37°C in a 5% CO2 atmosphere, the nonadherent cells were removed by washing them five-time in 2 ml PBS hold at 37°C. The peritoneal macrophages were incubated in fresh RPMI 1640 supplemented with 50 µg/ml of gentamicin. Annexin V-FLUOS staining kit (Roche Diagnostics Corporation, Indianapolis, Ind.) was used to detect the dead peritoneal macrophage cells 24 h post infection.
Suggestions and critical review of this manuscript by Drs. George W. Jourdian, Michele Swanson, and Gabriel Nunez at the University of Michigan are gratefully acknowledged.
Conceived and designed the experiments: FC YH. Performed the experiments: FC. Analyzed the data: FC YH. Contributed reagents/materials/analysis tools: YH. Wrote the paper: FC YH.
- 1. Schurig GG, Sriranganathan N, Corbel MJ (2002) Brucellosis vaccines: past, present and future. Vet Microbiol 90: 479–496.
- 2. Pappas G, Papadimitriou P, Akritidis N, Christou L, Tsianos EV (2006) The new global map of human brucellosis. Lancet Infect Dis 6: 91–99.
- 3. Roop RM , Gee JM, Robertson GT, Richardson JM, Ng WL, et al. (2003) Brucella stationary-phase gene expression and virulence. Annu Rev Microbiol 57: 57–76.
- 4. Lapaque N, Moriyon I, Moreno E, Gorvel JP (2005) Brucella lipopolysaccharide acts as a virulence factor. Curr Opin Microbiol 8: 60–66.
- 5. Gross A, Terraza A, Ouahrani-Bettache S, Liautard JP, Dornand J (2000) In vitro Brucella suis infection prevents the programmed cell death of human monocytic cells. Infection and immunity 68: 342–351.
- 6. He Y, Reichow S, Ramamoorthy S, Ding X, Lathigra R, et al. (2006) Brucella melitensis triggers time-dependent modulation of apoptosis and down-regulation of mitochondrion-associated gene expression in mouse macrophages. Infection and immunity 74: 5035–5046.
- 7. Tolomeo M, Di Carlo P, Abbadessa V, Titone L, Miceli S, et al. (2003) Monocyte and lymphocyte apoptosis resistance in acute and chronic brucellosis and its possible implications in clinical management. Clin Infect Dis 36: 1533–1538.
- 8. Rittig MG, Kaufmann A, Robins A, Shaw B, Sprenger H, et al. (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.
- 9. Fernandez-Prada CM, Zelazowska EB, Nikolich M, Hadfield TL, Roop RM , et al. (2003) Interactions between Brucella melitensis and human phagocytes: bacterial surface O-Polysaccharide inhibits phagocytosis, bacterial killing, and subsequent host cell apoptosis. Infection and immunity 71: 2110–2119.
- 10. Freeman BA, Pearson GR, Hines WD (1964) Host-parasite relationships in brucellosis. 3. Behavior of avirulent Brucella in tissue culture monocytes. J Infect Dis 114: 441–449.
- 11. Freeman BA, Kross DJ, Circo R (1961) Host-parasite relationships in brucellosis. II. Destruction of macrophage cultures by Brucella of different virulence. J Infect Dis 108: 333–338.
- 12. Pei J, Ficht TA (2004) Brucella abortus rough mutants are cytopathic for macrophages in culture. Infection and immunity 72: 440–450.
- 13. Pei J, Turse JE, Wu Q, Ficht TA (2006) Brucella abortus rough mutants induce macrophage oncosis that requires bacterial protein synthesis and direct interaction with the macrophage. Infection and immunity 74: 2667–2675.
- 14. Bras M, Queenan B, Susin SA (2005) Programmed cell death via mitochondria: different modes of dying. Biochemistry (Mosc) 70: 231–239.
- 15. Hetz CA, Torres V, Quest AF (2005) Beyond apoptosis: nonapoptotic cell death in physiology and disease. Biochem Cell Biol 83: 579–588.
- 16. Edinger AL, Thompson CB (2004) Death by design: apoptosis, necrosis and autophagy. Curr Opin Cell Biol 16: 663–669.
- 17. Bergsbaken T, Fink SL, Cookson BT (2009) Pyroptosis: host cell death and inflammation. Nat Rev Microbiol 7: 99–109.
- 18. Reed JC (2004) Apoptosis mechanisms: implications for cancer drug discovery. Oncology (Williston Park) 18: 11–20.
- 19. Colussi PA, Kumar S (1999) Targeted disruption of caspase genes in mice: what they tell us about the functions of individual caspases in apoptosis. Immunol Cell Biol 77: 58–63.
- 20. Mignotte B, Vayssiere JL (1998) Mitochondria and apoptosis. Eur J Biochem 252: 1–15.
- 21. Fink SL, Cookson BT (2005) Apoptosis, pyroptosis, and necrosis: mechanistic description of dead and dying eukaryotic cells. Infect Immun 73: 1907–1916.
- 22. Majno G, Joris I (1995) Apoptosis, oncosis, and necrosis. An overview of cell death. Am J Pathol 146: 3–15.
- 23. Lara-Tejero M, Sutterwala FS, Ogura Y, Grant EP, Bertin J, et al. (2006) Role of the caspase-1 inflammasome in Salmonella typhimurium pathogenesis. The Journal of experimental medicine 203: 1407–1412.
- 24. Raupach B, Peuschel SK, Monack DM, Zychlinsky A (2006) Caspase-1-mediated activation of interleukin-1beta (IL-1beta) and IL-18 contributes to innate immune defenses against Salmonella enterica serovar Typhimurium infection. Infection and immunity 74: 4922–4926.
- 25. Santos RL, Tsolis RM, Baumler AJ, Smith R 3rd, Adams LG (2001) Salmonella enterica serovar typhimurium induces cell death in bovine monocyte-derived macrophages by early sipB-dependent and delayed sipB-independent mechanisms. Infect Immun 69: 2293–2301.
- 26. Sansonetti PJ, Phalipon A, Arondel J, Thirumalai K, Banerjee S, et al. (2000) Caspase-1 activation of IL-1beta and IL-18 are essential for Shigella flexneri-induced inflammation. Immunity 12: 581–590.
- 27. Zamboni DS, Kobayashi KS, Kohlsdorf T, Ogura Y, Long EM, et al. (2006) The Birc1e cytosolic pattern-recognition receptor contributes to the detection and control of Legionella pneumophila infection. Nature immunology 7: 318–325.
- 28. Ciaramella A, Cavone A, Santucci MB, Amicosante M, Martino A, et al. (2002) Proinflammatory cytokines in the course of Mycobacterium tuberculosis-induced apoptosis in monocytes/macrophages. J Infect Dis 186: 1277–1282.
- 29. Jesenberger V, Procyk KJ, Yuan J, Reipert S, Baccarini M (2000) Salmonella-induced caspase-2 activation in macrophages: a novel mechanism in pathogen-mediated apoptosis. J Exp Med 192: 1035–1046.
- 30. Schurig GG, Roop RM , Bagchi T, Boyle S, Buhrman D, et al. (1991) Biological properties of RB51; a stable rough strain of Brucella abortus. Veterinary microbiology 28: 171–188.
- 31. McQuiston JR, Vemulapalli R, Inzana TJ, Schurig GG, Sriranganathan N, et al. (1999) Genetic characterization of a Tn5-disrupted glycosyltransferase gene homolog in Brucella abortus and its effect on lipopolysaccharide composition and virulence. Infect Immun 67: 3830–3835.
- 32. Schurig GG, Sriranganathan N, Corbel MJ (2002) Brucellosis vaccines: past, present and future. Vet Microbiol 90: 479–496.
- 33. Vemulapalli R, McQuiston JR, Schurig GG, Sriranganathan N, Halling SM, et al. (1999) Identification of an IS711 element interrupting the wboA gene of Brucella abortus vaccine strain RB51 and a PCR assay to distinguish strain RB51 from other Brucella species and strains. Clin Diagn Lab Immunol 6: 760–764.
- 34. Marianelli C, Ciuchini F, Tarantino M, Pasquali P, Adone R (2004) Genetic bases of the rifampin resistance phenotype in Brucella spp. J Clin Microbiol 42: 5439–5443.
- 35. van Engeland M, Ramaekers FC, Schutte B, Reutelingsperger CP (1996) A novel assay to measure loss of plasma membrane asymmetry during apoptosis of adherent cells in culture. Cytometry 24: 131–139.
- 36. Martin-Martin AI, Caro-Hernandez P, Orduna A, Vizcaino N, Fernandez-Lago L (2008) Importance of the Omp25/Omp31 family in the internalization and intracellular replication of virulent B. ovis in murine macrophages and HeLa cells. Microbes Infect 10: 706–710.
- 37. Seleem MN, Vemulapalli R, Boyle SM, Schurig GG, Sriranganathan N (2004) Improved expression vector for Brucella species. Biotechniques 37: 740, 742, 744.
- 38. He Y, Vemulapalli R, Zeytun A, Schurig GG (2001) Induction of specific cytotoxic lymphocytes in mice vaccinated with Brucella abortus RB51. Infect Immun 69: 5502–5508.
- 39. Stevens MG, Olsen SC, Pugh GW (1995) Comparison of spleen cell proliferation in response to Brucella abortus 2308 lipopolysaccharide or proteins in mice vaccinated with strain 19 or RB51. Infect Immun 63: 3199–3205.
- 40. Lee IK, Olsen SC, Kehrli M, Bolin CA (1999) The adjuvant effect of a single dose of interleukin-12 on murine immune responses to live or killed Brucella abortus strain RB51. Can J Vet Res 63: 284–287.
- 41. Talanian RV, Quinlan C, Trautz S, Hackett MC, Mankovich JA, et al. (1997) Substrate specificities of caspase family proteases. J Biol Chem 272: 9677–9682.
- 42. Thornberry NA, Rano TA, Peterson EP, Rasper DM, Timkey T, et al. (1997) A combinatorial approach defines specificities of members of the caspase family and granzyme B. Functional relationships established for key mediators of apoptosis. J Biol Chem 272: 17907–17911.
- 43. Troy CM, Shelanski ML (2003) Caspase-2 redux. Cell Death Differ 10: 101–107.
- 44. Chang K, Elledge SJ, Hannon GJ (2006) Lessons from Nature: microRNA-based shRNA libraries. Nat Methods 3: 707–714.
- 45. Chen Y, Cheng G, Mahato RI (2008) RNAi for treating hepatitis B viral infection. Pharm Res 25: 72–86.
- 46. Fewell GD, Schmitt K (2006) Vector-based RNAi approaches for stable, inducible and genome-wide screens. Drug Discov Today 11: 975–982.
- 47. Kroemer G (2003) Mitochondrial control of apoptosis: an introduction. Biochem Biophys Res Commun 304: 433–435.
- 48. Rodriguez-Enriquez S, He L, Lemasters JJ (2004) Role of mitochondrial permeability transition pores in mitochondrial autophagy. Int J Biochem Cell Biol 36: 2463–2472.
- 49. Chen M, Gan H, Remold HG (2006) A mechanism of virulence: virulent Mycobacterium tuberculosis strain H37Rv, but not attenuated H37Ra, causes significant mitochondrial inner membrane disruption in macrophages leading to necrosis. J Immunol 176: 3707–3716.
- 50. Gonzalez D, Grillo MJ, De Miguel MJ, Ali T, Arce-Gorvel V, et al. (2008) Brucellosis vaccines: assessment of Brucella melitensis lipopolysaccharide rough mutants defective in core and O-polysaccharide synthesis and export. PLoS One 3: e2760.
- 51. Cloeckaert A, Zygmunt MS, Guilloteau LA (2002) Brucella abortus vaccine strain RB51 produces low levels of M-like O-antigen. Vaccine 20: 1820–1822.
- 52. Jimenez de Bagues MP, Terraza A, Gross A, Dornand J (2004) Different responses of macrophages to smooth and rough Brucella spp.: relationship to virulence. Infect Immun 72: 2429–2433.
- 53. Fernandez-Prada CM, Zelazowska EB, Nikolich M, Hadfield TL, Roop RM , et al. (2003) Interactions between Brucella melitensis and human phagocytes: bacterial surface O-Polysaccharide inhibits phagocytosis, bacterial killing, and subsequent host cell apoptosis. Infect Immun 71: 2110–2119.
- 54. Detilleux PG, Deyoe BL, Cheville NF (1990) Entry and intracellular localization of Brucella spp. in Vero cells: fluorescence and electron microscopy. Vet Pathol 27: 317–328.
- 55. Porte F, Naroeni A, Ouahrani-Bettache S, Liautard JP (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.
- 56. Bergeron L, Perez GI, Macdonald G, Shi L, Sun Y, et al. (1998) Defects in regulation of apoptosis in caspase-2-deficient mice. Genes Dev 12: 1304–1314.
- 57. Lassus P, Opitz-Araya X, Lazebnik Y (2002) Requirement for caspase-2 in stress-induced apoptosis before mitochondrial permeabilization. Science 297: 1352–1354.
- 58. Zhivotovsky B, Orrenius S (2005) Caspase-2 function in response to DNA damage. Biochemical and biophysical research communications 331: 859–867.
- 59. Zhang Y, Padalecki SS, Chaudhuri AR, De Waal E, Goins BA, et al. (2007) Caspase-2 deficiency enhances aging-related traits in mice. Mech Ageing Dev 128: 213–221.
- 60. Faderl S, Thall PF, Kantarjian HM, Talpaz M, Harris D, et al. (1999) Caspase 2 and caspase 3 as predictors of complete remission and survival in adults with acute lymphoblastic leukemia. Clin Cancer Res 5: 4041–4047.
- 61. Yoo NJ, Lee JW, Kim YJ, Soung YH, Kim SY, et al. (2004) Loss of caspase-2, -6 and -7 expression in gastric cancers. APMIS 112: 330–335.
- 62. Zhang X, Uthaisang W, Hu L, Ernberg IT, Fadeel B (2005) Epstein-Barr virus-encoded latent membrane protein 1 promotes stress-induced apoptosis upstream of caspase-2-dependent mitochondrial perturbation. Int J Cancer 113: 397–405.
- 63. Galdiero E, Romano Carratelli C, Vitiello M, Nuzzo I, Del Vecchio E, et al. (2000) HSP and apoptosis in leukocytes from infected or vaccinated animals by Brucella abortus. The new microbiologica: official journal of the Italian Society for Medical, Odontoiatric, and Clinical Microbiology (SIMMOC) 23: 271.
- 64. Krumschnabel G, Sohm B, Bock F, Manzl C, Villunger A (2009) The enigma of caspase-2: the laymen's view. Cell Death Differ 16: 195–207.
- 65. Dahmer MK (2005) Caspases-2, -3, and -7 are involved in thapsigargin-induced apoptosis of SH-SY5Y neuroblastoma cells. J Neurosci Res 80: 576–583.
- 66. Li H, Bergeron L, Cryns V, Pasternack MS, Zhu H, et al. (1997) Activation of caspase-2 in apoptosis. J Biol Chem 272: 21010–21017.
- 67. Gray MW, Burger G, Lang BF (1999) Mitochondrial evolution. Science 283: 1476–1481.
- 68. Dyall SD, Brown MT, Johnson PJ (2004) Ancient invasions: from endosymbionts to organelles. Science 304: 253–257.
- 69. Enoksson M, Robertson JD, Gogvadze V, Bu P, Kropotov A, et al. (2004) Caspase-2 permeabilizes the outer mitochondrial membrane and disrupts the binding of cytochrome c to anionic phospholipids. J Biol Chem 279: 49575–49578.
- 70. Sidi S, Sanda T, Kennedy RD, Hagen AT, Jette CA, et al. (2008) Chk1 suppresses a caspase-2 apoptotic response to DNA damage that bypasses p53, Bcl-2, and caspase-3. Cell 133: 864–877.
- 71. Baumann S, Nasser Eddine A, Kaufmann SH (2006) Progress in tuberculosis vaccine development. Curr Opin Immunol 18: 438–448.
- 72. Bricker BJ (2002) PCR as a diagnostic tool for brucellosis. Vet Microbiol 90: 435–446.
- 73. Timoney JF, Gillespie JH, Scott FW, Barlough JE (1988) The Genus Brucella. Hagan and Bruner's microbiology and infectious diseases of domestic animals. Eighth edition. ed: Cornell University Press.135–152.
- 74. Celada A, Gray PW, Rinderknecht E, Schreiber RD (1984) Evidence for a gamma-interferon receptor that regulates macrophage tumoricidal activity. J Exp Med 160: 55–74.
- 75. Barquero-Calvo E, Chaves-Olarte E, Weiss DS, Guzman-Verri C, Chacon-Diaz C, et al. (2007) Brucella abortus uses a stealthy strategy to avoid activation of the innate immune system during the onset of infection. PLoS ONE 2: e631.