| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Laboratory of Cellular and Molecular Immunology, Department of Animal Health and Biomedical Sciences, and
Enzyme Institute, Research Animal Resource Center, University of Wisconsin, Madison, WI 53706
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
is a key cytokine controlling Brucella
infection, and the diverse functions of this cytokine are mediated by
IFN regulatory factors (IRFs) such as IRF-1, IRF-2, and IFN consensus
sequence binding protein (ICSBP). However, the roles of these three
IRFs in Brucella infection have not been investigated.
The infection of each IRF-deficient mouse strain provides an
opportunity to determine not only the significance of each IRF molecule
but also the crucial immune components necessary for host defense
during in vivo infection, because respective IRF-deficient mouse
strains contain unique immunodeficient phenotypes.
Brucella abortus S2308-infected
IRF-1-/- mice were dead within 2 wk postinfection, while
IRF-2-/- mice contained less splenic
Brucella CFU than wild-type mice at the early stage of
infection. Infected ICSBP-/- mice maintained a plateau of
splenic Brucella CFU throughout the infection.
Additional infection of IL-12p40-, NO synthase 2-, and
gp91phox-deficient mice indicates that these
immune components are crucial for Brucella immunity and
may contribute to the susceptibility of IRF-1-/- and
ICSBP-/- mice. Immunologic and histopathological analyses
of infected IRF-1-/- mice indicate that the absence of
IL-12p40 induction and serious hepatic damage are involved in the death
of IRF-1-/- mice. These results indicate that 1) IRF-1
and ICSBP are essential transcriptional factors for IFN--mediated
protection against Brucella; 2) IL-12, reactive nitrogen
intermediates, and reactive oxygen intermediates are crucial immune
components against Brucella, and their absence may
contribute to the susceptibility of IRF-1-/- and
ICSBP-/- mice; and 3) hepatic damage caused by
Brucella virulence contributes to the death of
IRF-1-/- mice. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
One of the crucial cytokines involved in Brucella immunity
is IFN-, a Th-1 cytokine produced by activated
CD4+ and CD8+ T cells
(5, 6). The major function of IFN- in
Brucella immunity is the stimulation of bactericidal
activity in macrophages, host cells of Brucella spp.
However, the function of IFN- is more diverse than the induction of
bactericidal function and includes the stimulation of Ag presentation
through class I and class II MHC molecules, the orchestration of
leukocyte-endothelium interactions, the effects on cell proliferation
and apoptosis, as well as stimulation and repression of a variety of
genes whose functional significance remains obscure (7).
The implementation of pleiotropic effects by a single cytokine is
accomplished by complex patterns of cell-specific gene regulation, and
additionally several IFN- regulatory genes are themselves components
of transcription factors that create a complex picture for IFN-
functions.
IFN regulatory factor
(IRF)3 is a family of
secondary transcriptional factors containing the unique tryptophan
cluster DNA-binding region located at the N terminus and includes
IRF-1, IRF-2, IRF-3, IRF-4, IRF-5, IRF-6, IRF-7, IFN consensus sequence
binding protein (ICSBP; IRF-8), and p48 (IRF-9) (8, 9).
IRFs can be induced by IFN-
and/or IFN-, and among the nine
IRFs, the expressions of IRF-1, IRF-2, ICSBP, and p48 are induced by
IFN-. Thus, to understand the roles of IRFs induced by IFN- in
brucellosis in vivo, currently available IRF-1, IRF-2, and ICSBP gene
knockout (KO) mice were introduced. Intriguingly, respective KO mice
contain unique multiple immunodeficient phenotypes (Table I
). For example,
IRF-1-/- mice are deficient in the induction of
inducible NO synthase (iNOS), IL-12 p40, CD8+ T
cells, and NK cells as well as IFN-, while
IRF-2-/- mice possess a deficiency in NK cells
and the dysregulation of IL-12p40 induction (10, 11, 12, 13).
ICSBP-/- mice are deficient in the induction of
IL-12p40 and IRF-2 as well as the production of reactive oxygen
intermediates (ROI) (14, 15, 16). Therefore, the infection of
these three IRF-deficient mouse strains provides a unique opportunity
to determine not only the significance of each IRF molecule induced by
IFN- but also the crucial immune components during the infection.
For instance, IRF-1-/- mice were susceptible to
Mycobacterium bovis (13), Leishmania
major (17), and Toxoplasma gondii
(18), due to the absence of iNOS induction and/or IL-12p40
induction, but were resistant to Listeria monocytogenes
infection (14). IRF-2-/- mice were
susceptible to L. monocytogenes (14) and
L. major (10) infection.
ICSBP-/- mice were susceptible to T.
gondii (15), L. major (16),
and L. monocytogenes (14) infection due to the
absence of IL-12p40 induction or ROI production. Thus, to additionally
elucidate the immune components that may contribute to the
susceptibility of respective IRF-deficient mice,
IL-12p40-/-, NOS2-/-,
and gp91phox-/- mice were also
infected with Brucella.
|
during Brucella infection in vivo. Here,
we report 1) the susceptibility of Brucella infection in
IRF-1-/- and ICSBP-/-
mice, 2) the involvement of IL-12, reactive nitrogen intermediates
(RNI), and ROI, which may contribute to the susceptibility of
IRF-1-/- and ICSBP-/-
mice, in Brucella immunity, and 3) the correlation between
the level of hepatic damage and acute death of
IRF-1-/- mice induced by virulent B.
abortus strains. These results suggest the significant roles of
IRF-1 and ICSBP during Brucella infection and the ascription
of hepatic damages to the death of IRF-1-/-
mice infected with virulent B. abortus strains. |
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Brucella abortus S2308, the virulent wild-type strain, was provided by Dr. B. Martin (Veterinary Services, National Animal Disease Center, Ames, IA). RB51, the current live attenuated U.S. vaccine strain derived from S2308 deficient in O-chain LPS, was a gift from Dr. G. G. Schurig (Virginia-Maryland Regional College of Veterinary Medicine, Virginia Polytechnic Institute and State University, Blacksburg, VA) (19).
Mice
Breeding pairs of IRF-1-/- and IRF-2+/- mice on the C57BL/6 background were provided by Dr. T. W. Mak (Amgen Institute, Ontario Cancer Institute, University of Toronto, Toronto, Canada) (20). Breeding pairs of ICSBP+/- mice on the DBA/2J background were donated by Dr. H. C. Morse III (National Institute of Allergy and Infectious Diseases, Bethesda, MD) (16). These three strains were heterozygously bred, and genotypes were determined by PCR (20, 21). Whenever possible, IRF-1+/+, IRF-2+/+, and ICSBP+/+ mice derived from heterozygous littermates were used as controls; otherwise, C57BL/6 (H-2b) or DBA/2J (H-2d) mice were used (The Jackson Laboratory, Bar Harbor, ME). Breeding pairs of IL-12p40-/- (22), NOS2-/- (23), and gp91phox-/- (24) on the C57BL/6 background (The Jackson Laboratory) were homozygously bred, and C57BL/6 mice were used as controls. The founders of double KO NOS2-/-/ICSBP-/-, NOS2-/-, ICSBP-/-, and wild-type mice on the C57BL/6 x DBA/2J background were generated by cross-breeding F1 (NOS2+/-ICSBP+/-) mice and were used for breeding. Mice were genotyped by PCR if necessary, and 6- to 9-wk-old mice were used for experimental infection. All the mice were bred at the Department of Animal Health and Biomedical Sciences animal care facilities, and mice infected with B. abortus were housed in the Biosafety Level 3 facilities in the School of Veterinary Medicine. Infected IRF-1-/- mice were monitored daily for survival.
Infection and enumeration of Brucella from spleens and livers
Mice were injected i.p. with B. abortus S2308 or RB51 at 5 x 105 CFU in 200 µl PBS. To count residual Brucella CFUs in the spleens or livers of mice, five mice from each group were examined at each sampling period. Spleens or livers were homogenized in plastic bags with 10 ml sterile PBS using a Stomacher Lab Blender (Tekmar, Cincinnati, OH). To enumerate viable Brucella, homogenized spleen or liver samples were serially diluted 10-fold with PBS and plated on Brucella agar (Difco, Detroit, MI). Brucella colonies were counted after a 3-day incubation at 37°C with 5% CO2. CFU numbers <100/organ could not be detected using this method.
Bioactive rIL-12p70 injection
IRF-1-/- mice were i.p. injected with mouse rIL-12p70 (Sigma-Aldrich, St. Louis, MO) for the initial 8 days of infection at a concentration of 0.33 mg/mouse in 200 µl PBS daily beginning at the time of challenge with B. abortus S2308. This concentration of IL-12 has been shown previously to be effective in significantly reducing the mortality of IRF-1-/- or wild-type mice due to acute parasite infection (18, 25).
Medication of iNOS inhibitor
C57BL/6 and IL-12p40-/- mice were medicated with an iNOS inhibitor, #67, and L-arginine analogs (Dr. P. T. Manning, Searle, St. Louis, MO) in the drinking water (1 mg/ml) from 3 days before infection throughout the completion of the experiment (26). The solution was completely replaced three times per week.
GOT, GPT, and albumin assays
Mice were exsanguinated from the orbital sinus, and sera were collected using Microtainer (BD Biosciences, Rutherford, NJ). Glutamic oxalacetic transaminase (GOT) and glutamic pyruvic transaminase (GPT) activities as well as albumin concentration were quantitatively measured by colorimetric determination with diagnostic transaminase reagents and albumin reagent (M. bovis), respectively, following the manufacturer’s directions (Sigma-Aldrich). The activities of GOT and GPT were measured with 1/5 diluted sera with PBS.
Histology
Mice infected with 5 x 105 CFU B. abortus S2308 or RB51 and were killed 12 days postinfection (p.i.). Organs were immersed in 10% formalin, embedded, microsectioned, mounted on microscopic slides, and stained with H&E. The H&E-stained slides were microscopically observed.
Statistical analysis
Data from infected mice were analyzed independently and compared using ANOVA and the generalized Wilcoxon rank sum test (LIFETEST) in the SAS program (SAS Institute, Cary, NC). Infection of IRF-1-/- mice was performed twice independently, and then the results were pooled to perform the data analysis. Values of p < 0.05 were considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
All IRF-1-/- mice infected with B.
abortus wild-type strain S2308 at 5 x
105 CFU were dead within 2 wk p.i., and 50% of
IRF-1-/- mice were dead at 9 days p.i. (Fig. 1A). However, no death
occurred in S2308 (see Fig. 3)- or RB51 (data not shown)-infected
IRF-1+/+ mice or in RB51-infected
IRF-1-/- mice throughout 4 wk p.i., suggesting
that the death of IRF-1-/- mice is caused by
the reciprocal relation between the immunocompromised phenotypes of
IRF-1-/- mice and the level of virulence in
Brucella strains (Fig. 1A). The death of
IRF-1-/- mice caused by B. abortus
wild-type strain infection is notable, because even immunocompromised
SCID (27) or nude (28) mice control
Brucella infection. Dead IRF-1-/-
mice previously infected with the virulent B. abortus strain
S2308 contained ascitic fluids, suggesting that hepatic damage might
have occurred. IRF-2-/- mice contained fewer
splenic S2308 CFUs than IRF-2+/+ mice during the
initial 2 wk p.i., and ICSBP-/- mice maintained
a plateau of splenic S2308 CFU throughout the experiment similar to
IL-12p40-/- mice (Fig. 1B). IL-12
was not detected by ELISA in IRF-1-/- and
ICSBP-/- mice during the initial 5 days p.i.
(data not shown). IRF-2-/- mice had very low
amounts of IL-12 on day 3 p.i. RNI and ROI were measured in
peritoneal macrophages obtained from uninfected KO mice treated with
LPS and/or IFN-. ROI was deficient in
ICSBP-/- mice, and RNI was defective in
IRF-1-/- mice. However,
IRF-2-/- mice were normal in RNI and ROI
activities (data not tabulated). These results suggest that the roles
of IRF-1 and ICSBP in Brucella immunity are critical, and
the death of IRF-1-/- mice correlates with the
level of virulence in B. abortus strains.
|
|
IRF-1-, IRF-2-, and ICSBP-deficient mouse strains possess unique
immunodeficient phenotypes (Table I
). However, the roles of these
defective immune components in Brucella immunity in vivo
have not been fully elucidated. The importance of IL-12 and
CD8+ T cells in Brucella immunity in
vivo has been tested using anti-IL-12 Ab-injected mice
(29) and MHC class I-deficient
2-microglobulin-/-
mice (30), respectively; however, IL-12-deficient mice
have not been tested in brucellosis. Additionally, although the
significance of ROI has been delineated in vitro (31), the
roles of iNOS and ROI have not been investigated during
Brucella infection in vivo. The NK cell-depleted mice could
control Brucella infection as wild-type mice do, suggesting
that the role of NK cells is minor (32). Thus, to
demonstrate a definitive role of each immune component during the
Brucella infection in vivo,
IL-12p40-/-, NOS2-/-,
and gp91phox-/- mouse strains were
infected.
IL-12 is a 70-kDa heterodimeric cytokine formed by two covalently
linked glycosylated chains of p35 and p40 and produced mostly by
phagocytic cells in response to intracellular bacteria
(33). In vivo IL-12 induces IFN- production from NK and
T cells during early infection and favors Th-1 cell differentiation by
promoting CD4+ T cells for high IFN-
production. The crucial role of IL-12 in other intracellular pathogen
infections was demonstrated using IL-12p40-/-
mice (34, 35). IL-12p40-/- mice
infected with B. abortus S2308 could not control infection
and maintained a high plateau of bacteria for the duration of the
experiment, suggesting that the Th-1 immune response is required to
control Brucella infection (Fig. 2).
|
). At 2 wk p.i. splenic CFU reaches a peak,
and a 100-fold CFU difference was observed between C57BL/6 mice and
NOS2-/- or gp91phox-/-
mice. However, after 2 wk p.i. a dramatic reduction of CFU was observed
for the last 2 wk, indicating that other compensatory bactericidal
mechanisms are probably activated to control the infection in
NOS2-/- and
gp91phox-/- mice. In fact,
NOS2/gp91phox double-KO mice showed delayed
control of splenic Brucella similar to NOS2 or
gp91phox single-KO mice, suggesting that other
sources of ROI and RNI besides NADPH oxidase and NOS2, respectively,
are involved in a Brucella killing mechanism (data not
shown). However, death did not occur in any of these immunodeficient
mouse strains, confirming that death in
IRF-1-/- mice is unusual. Taken together, these
results suggest that IL-12, RNI, and ROI are crucial innate immune
components involved in Brucella immunity in the early stage
of infection, and the absence of these immune components may contribute
to the susceptibility of IRF-1-/- and
ICSBP-/- mice. Bioactive rIL-12p70-supplemented IRF-1-/- mice infected with B. abortus S2308
The death of IRF-1-/- mice from
Brucella virulence is an unusual feature in brucellosis.
Thus, we hypothesized that the death of
IRF-1-/- mice results from the multiple
immunodeficiencies required for immunity to intracellular pathogens.
The persistent Brucella infection occurring in
IL-12p40-/- mice indicated the significance of
IL-12 in B. abortus infection in vivo. Therefore, we tested
whether the absence of IL-12p40 induction contributes to the death of
IRF-1-/- mice by administering IL-12. Bioactive
rIL-12p70 (0.33 mg/mouse in 200 µl PBS) was injected i.p. to
IRF-1-/- mice for the initial 8 days of
infection. rIL-12 administered to IRF-1-/- mice
delayed death, indicating the absence of IL-12p40 induction is involved
in the death of IRF-1-/- mice as well as
confirming the significance of IL-12 in Brucella immunity
(Fig. 3).
Phenotypic NOS2/IL-12p40 double-immunodeficient mice infected with B. abortus S2308
The results obtained from NOS2-/- and
IL-12p40-/- mice indicate the importance of
iNOS and IL-12 in B. abortus immunity; however, this single
immunodeficient mice with high splenic Brucella CFUs failed
to die (Fig. 2). In addition, the delayed death of
IRF-1-/- mice supplemented with rIL-12
indicated that the absence of IL-12 contributed to the death of
IRF-1-/- mice (Fig. 3). Thus, we hypothesized
that multiple immunodeficiencies including IL-12p40 induction
contribute to the death of IRF-1-/- mice.
Therefore, mice deficient in both iNOS and IL-12p40 were tested for
their ability to control Brucella infection (Fig. 4). Because the genes encoding NOS2 and
IL-12p40 are located on the same chromosome, NOS2/IL-12p40 double-KO
mice are difficult to generate at the gene level. Therefore,
double-immunodeficient phenotype mice, iNOS and IL-12p40 deficiency,
were generated by medicating IL-12p40-/- mice
with an iNOS inhibitor, #67, followed by Brucella infection
(Fig. 4A). The iNOS inhibitor-treated C57BL/6 mice were
delayed in controlling Brucella infection similar to
NOS2-/- mice, with nearly a 1-log CFU increase
at 2 wk p.i. Also, treated IL-12p40-/- mice had
more splenic Brucella CFU than nontreated
IL-12p40-/- mice. These results suggest both NO
and IL-12 play important roles, but this double immunodeficiency was
not sufficient to explain the death of IRF-1-/-
mice. Also, the reduction of splenic CFU in treated C57BL/6 mice, but
not in treated IL-12p40-/- mice, suggests that
the control mechanism after 2 wk p.i. in NOS2-deficient mice is IL-12
dependent (Fig. 2).
|
B). Each single-immunodeficient mouse strain on the
C57BL/6 x DBA/2J background demonstrated an infection similar to
those in NOS2-/- mice on the C57BL/6 background
and ICSBP-/- mice on the DBA/2J background.
NOS2/ICSBP double-immunodeficient mice infected with B.
abortus did not die, and the growth of Brucella CFU in
spleens was similar to that in IL-12p40-/- mice
treated with the iNOS inhibitor. CFUs from double-immunodeficient mice
were 1.5 log higher than those in ICSBP-/- mice
and 2.5 log higher than those in wild-type and
NOS2-/- mice. Taken together, the cause of
death in IRF-1-/- mice may result from a
multiple immunodeficiencies, and NOS2/IL-12p40 double immunodeficiency
was not sufficient to induce the death of infected mice. Increased B. abortus CFUs in the livers of IRF-1-/- mice
The level of systemic infection in murine brucellosis is
detectable by enumerating the number of residual Brucella
CFU in liver and spleen (6). Thus,
IRF-1+/+ and IRF-1-/-
mice were infected with B. abortus virulent S2308 and
avirulent RB51 strains, and hepatic and splenic CFUs were counted every
third day for investigating the cause of death in
IRF-1-/- mice (Fig. 5).
|
A). From day 6 p.i.
IRF-1+/+ mice began to control S2308 CFU in the
liver. However, IRF-1-/- mice failed to control
hepatic B. abortus, and at 12 days p.i.
IRF-1-/- mice had 2.5-log greater hepatic CFUs
than IRF-1+/+ mice (p <
0.005). Unlike S2308-infected mice, IRF-1+/+ and
IRF-1-/- mice infected with 5 x
105 CFU of RB51 had a parallel pattern of hepatic
CFU clearance. Although a delay of bacterial clearance was detected
from IRF-1-/- mice compared with
IRF-1+/+ mice, ultimately RB51 was cleared from
the livers of both mouse strains within 12 days p.i. These results
indicate that a virulence-dependent bacterial hepatic burden in
IRF-1-/- mice that induces hepatic damage
contributed to death in this mouse strain.
The spleens from IRF-1+/+ and
IRF-1-/- infected mice had a similar plateau of
S2308 CFUs for the 12 days p.i., except for 3 and 6 days p.i. where 0.5
log higher CFU was detected from IRF-1-/- mice
(Fig. 5B). The splenic RB51 CFUs from
IRF-1+/+ and IRF-1-/-
mice demonstrated a delay of clearance in
IRF-1-/- mice. At 9 days p.i., >100-fold CFU
difference was detected in IRF-1-/- mice
compared with IRF-1+/+ mice, but splenic RB51
CFUs in IRF-1-/- mice declined by day
12.
Taken together, the hepatic Brucella CFU difference between IRF-1+/+ and IRF-1-/- mice was dramatically increased in S2308-infected mice, but not in RB51-infected mice. Thus, a certain level of bacterial virulence is required to induce such a hepatic bacterial burden, or the basal immunity in IRF-1-/- mice can control the infection caused by RB51 virulence. This continual hepatic bacterial burden in IRF-1-/- mice may lead to fatal hepatic damage.
Histological analysis of IRF-1-/- mice infected with B. abortus
Brucellosis is a chronic infection inducing granulomatous
formation in the livers and spleens of humans and mice (1, 28, 40). To determine whether differences in the levels of hepatic
Brucella CFUs detected in IRF-1+/+ and
IRF-1-/- mice affect the degree of hepatic
damage, livers from infected IRF-1+/+ and
IRF-1-/- mice were examined histologically.
Compared with RB51 infection, S2308 induced more granulomas in the
livers of IRF-1+/+ and
IRF-1-/- mice at 12 days p.i. Also, more
extensive hepatic granulomas were detected in
IRF-1-/- mice compared with
IRF-1+/+ mice. Thus,
IRF-1-/- mice infected with S2308 showed the
greatest hepatic granuloma formation (Fig. 6A), while RB51-infected
IRF-1-/- mice had fewer and less extensive
microfocal granulomas (Fig. 6B). These results suggest that
the virulence of B. abortus correlates qualitatively and
quantitatively with granuloma formation in
IRF-1-/- mice and can be interpreted as the
level of bacterial hepatic burden correlates with the level of liver
damage in IRF-1-/- mice based on the previous
results (Fig. 5A). Additional analyses with brain, lung,
heart, kidney, spleen, as well as genital and gastrointestinal organs
did not show significant histopathological manifestations (data not
shown).
|
Histological analyses identified different levels of hepatic
granuloma formation in the IRF-1+/+ and
IRF-1-/- mice infected with Brucella
strains containing different degrees of virulence. Thus, to quantify
the level of liver damage, sera GOT and GPT activities were measured
from the infected mice. Both uninfected IRF-1+/+
and IRF-1-/- mice maintained normal GOT and GPT
activities in serum (37 and 19 IU/l, respectively; Fig. 7, A and B)
(41). Also, RB51-infected IRF-1+/+
and IRF-1-/- mice did not possess dramatically
increased GOT and GPT activity in sera. In contrast, S2308-infected
IRF-1+/+ and IRF-1-/-
mice possessed increased GOT and GPT activities, indicating that acute
hepatitis occurred in both mouse strains. However, the extent of liver
damage between the two mouse strains was discernable. Serum GOT
activity in S2308-infected IRF-1+/+ and
IRF-1-/- mice was 81 and 166 IU/l, and GPT
activity was 39 and 81 IU/l, respectively, indicating that 2- and
4-fold increased liver damage had occurred in each mouse strain. These
results indicate that there is virulence-dependent hepatic damage in
both mouse strains, and the level of hepatic damage in
IRF-1-/- mice was more severe than that in
IRF-1+/+ mice.
|
C) (41). S2308-infected
IRF-1+/+ and IRF-1-/-
mice had decreased albumin, indicating that dysosmoregulation occurred
in both mouse strains. Serum albumin in S2308-infected
IRF-1+/+ mice and
IRF-1-/- mice was 2.5 and 1.8 g/dl,
respectively, indicating that 23 and 47% albumin reduction had
occurred in each mouse strain. The albumin concentration in ascitic
fluids in IRF-1-/- mice was 1.5 g/dl,
suggesting that a 56% reduction occurred compared with albumin in
serum (data not shown). Additional analyses of the ascitic fluid
obtained from S2308-infected IRF-1-/- mice at
12 days p.i. determined that the ascitic pellet was 90–95%
macrophages with the remaining cells consisting of small lymphocytes,
neutrophils, as well as a rare plasma cell, and 1.6 x
102–5.9 x 105 CFU/ml
S2308 bacteria were present in the ascitic fluid, providing evidence of
peritonitis (data not shown). As a result, GOT and GPT as well as albumin assays identified serious liver damage and dysosmoregulation in IRF-1-/- mice infected with the B. abortus virulent S2308 strain, suggesting that the cause of death in IRF-1-/- mice involved liver damage, and that a certain level of bacterial virulence is required to induce liver damage.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, and IFN- activates
phagocytic functions such as ROI and RNI in macrophages to kill
intracellular bacteria. However, the actual functions induced by
IFN- to control the intracellular infection are diverse, and these
various implementations are facilitated by secondary transcriptional
factors such as IRFs (7). Thus, to investigate the roles
of IFN- during Brucella infection, the roles of IRFs
induced by IFN- such as IRF-1, IRF-2, ICSBP, and p48 require
elucidation. Currently, IRF-1, IRF-2, and ICSBP gene-disrupted mice are
available, and these individual gene KO mouse strains are useful tools
to determine the function of respective factors during the infection in
vivo. Unlike other gene KO mice that are deficient in cytokines or
innate immune components such as IL-12p40, iNOS, or
gp91phox, these three transcriptional
factor-deficient mice possess multiple immunodeficient phenotypes
(Table I). Thus, to understand the immune responses in IRF-deficient
mice during Brucella infection, the roles of individual
immune components deficient in respective mouse strains also need to be
elucidated.
IRF-1-/- mice infected with B.
abortus virulent strains died within 12 days p.i.
IRF-2-/- mice contained less splenic
Brucella CFU than IRF-2+/+ mice during
the initial 2 wk p.i. ICSBP-/- mice maintained
a plateau of splenic Brucella CFU throughout the experiment.
These results clearly indicate that the roles of IRF-1 and ICSBP are
essential for Brucella immunity. However, the defective
immune components in IRF-1-/- and
ICSBP-/- mice that contribute to
Brucella susceptibility are still unknown. The lower splenic
CFU detected in IRF-2-/- compared with
IRF-2+/+ mice for the first 2 wk p.i. was
unexpected, because the dysregulation of IL-12p40 induction in
IRF-2-/- mice was anticipated to prolong
bacterial clearance (Fig. 1B) (10), and very
low concentrations of IL-12 were detected by ELISA from the sera of
B. abortus S2308-infected IRF-2-/-
mice during the initial 5 days p.i. (data not shown). However,
IRF-2-/- mice are phenotypically different from
IL-12p40-/- mice. For example, IRF-2 functions
to down-regulate the genes activated by IRF-1. Thus,
IRF-2-/- mice may have increased activation of
the genes induced by IRF-1 that play an important role in controlling
Brucella infection. In contrast to our result,
IRF-2-/- mice were highly susceptible to
L. monocytogenes (14) and L. major
(10) infection, suggesting that the immune response in
respective IRF-deficient mice may be pathogen specific. However,
IRF-2-/- (9) and
ICSBP-/- (21) mice demonstrate an
autoimmunity and a chronic myelogenous leukemia-like syndrome later in
life, respectively, suggesting that the progression of these
manifestations in even young mice may influence the in vivo results of
Brucella infection obtained from these mouse strains.
IL-12p40-, NOS2-, or gp91phox-deficient
mice were also susceptible to Brucella infection, suggesting
the significance of IL-12, RNI, and ROI in Brucella immunity
that may contribute to the susceptibility of
IRF-1-/- and ICSBP-/-
mice. We hypothesize that the combination of these immune components
produced the susceptibility of IRF-1-/- and
ICSBP-/- mice to Brucella infection.
In fact, rIL-12-supplemented IRF-1-/- mice
infected with B. abortus survived much longer than
PBS-injected IRF-1-/- mice, indicating that the
absence of IL-12 contributes to the death of
IRF-1-/- mice. This result is coincident to the
result obtained after T. gondii infection that suggests the
induction of an IFN--independent mechanism of protection against
this opportunistic pathogen by IL-12 (18).
In contrast to the rapid death of Brucella-infected
IRF-1-/- mice (Fig. 1A), others have
recently reported that IFN--/- mice have
increased Brucella replication compared with wild-type mice,
but survive for 6–10 wk (43). This difference in
IRF-1-/- mouse susceptibility to
Brucella suggests that multiple immune components contribute
to IRF-1-/- mouse susceptibility. In addition,
the fact that immunocompromised SCID (27) and nude
(28) mice control Brucella infection
significantly better than IRF-1-/- mice
indicates the possibility of an IFN-- and T cell-independent
mechanism(s) for controlling brucellosis. TNF-, produced by
macrophages and NK cells in SCID, nude, or IFN- KO mice, may induce
bactericidal mechanisms in macrophages that are T cell independent. The
significant role of TNF- in murine brucellosis has been delineated
(44). Therefore, IRF-1-/- mice
defective in IFN- and TNF- production (45) would
lack IFN--dependent as well as independent mechanisms to activate
the phagocytic function of macrophages to control Brucella
infection. Thus, more severe symptoms would be anticipated in
Brucella-infected IRF-1-/- mice than
in SCID, nude, and IFN- KO mice.
The critical question regarding hepatic damage is why B.
abortus virulent strains were not cleared and possessed a
hepatotropic phenotype in IRF-1-/- mice, while
the attenuated strains were controlled in the liver (Fig. 5A). Future experiments are required to address the
hepatotropic growth of virulent B. abortus strains in
IRF-1-/- mice. Kupffer cells may be unable to
control virulent infection in IRF-1-/- mice
because the levels of IFN- and TNF- in
IRF-1-/- mice are much lower than those in
wild-type mice (45). Thus, dysregulation of these
cytokines as well as insufficient IFN- signaling in
IRF-1-/- mice may not activate the phagocytic
activity of Kupffer cells. Thus, Kupffer cells in
IRF-1-/- mice would be a niche for B.
abortus proliferation in livers.
To understand the pathogenic mechanisms in
IRF-1-/- mice, immunological and
histopathological analyses were performed with mice infected with
diverse B. abortus strains containing different levels of
virulence. The contribution of defective IL-12p40 induction to the
death of IRF-1-/- mice was demonstrated by
rIL-12p70 administration (Fig. 3). However, phenotypic NOS2/IL-12p40
double-immunodeficient mice did not die of Brucella
infection, suggesting that other immunodeficient combinations may
contribute to death (Fig. 4). Histopathological analysis demonstrated
that the level of hepatic damage correlated with the degree of
virulence of the B. abortus strain (Fig. 6). One
manifestation of brucellosis is acute hepatitis (46). Most
animals infected with Brucella overcome this acute hepatitis
and maintain a chronic infection, and this transition is dependent on
the degree of virulence in Brucella strains
(1). This shared symptom of hepatitis between natural
hosts and IRF-1-/- mice supports the possible
role of IRF-1-/- mice as a relevant animal
model for detecting Brucella virulence. In addition, the
level of hepatic bacterial burden (Fig. 5A) correlated with
the level of granuloma formation (Fig. 6) as well as increasing
activities of GOT and GPT and decreasing albumin concentration in sera
(Fig. 7), indicating that hepatic damage contributes to the death of
IRF-1-/- mice. IRF-1-/-
mice were susceptible to M. bovis (13),
L. major (17), and T. gondii
(18), but resistant to L. monocytogenes
(14) infection, suggesting that pathogen-specific host
immune mechanisms are required for protection. The death of
IRF-1-/- mice occurred with T.
gondii, but liver damage was not described (18).
Thus, the death of IRF-1-/- mice with liver
damage and ascitic fluids may be a unique feature of brucellosis. Also,
the acute death of IRF-1-/- mice that manifest
hepatic damage is dependent on the virulence of Brucella
(Fig. 1A). These results suggest that the rapidity of death
in IRF-1-/- mice may be an alternative
yardstick to determine the level of Brucella virulence. In
fact, IRF-1-/- mice were infected with 5
x 105 CFU of diverse B. abortus
strains containing different degrees of virulence, and the correlation
between the rapidity of death in IRF-1-/- mice
and the level of virulence was
confirmed.4
In summary, using IRF-1-, IRF-2-, and ICSBP-deficient mice, we demonstrated that 1) the roles of IRF-1 and ICSBP are essential for Brucella immunity; 2) IL-12, RNI, and ROI, whose absence may contribute to the susceptibility of IRF-1-/- and ICSBP-/- mice, are important innate immune components in brucellosis; 3) the absence of IL-12p40 and hepatic damage contribute to the death of IRF-1-/- mice; and 4) the level of hepatic damage in IRF-1-/- mice is dependent on the bacterial burden in livers and B. abortus virulence.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Gary A. Splitter, Laboratory of Cellular and Molecular Immunology, Department of Animal Health and Biomedical Sciences, University of Wisconsin, 1656 Linden Drive, Madison, WI 53706. E-mail address: splitter{at}ahabs.wisc.edu
3 Abbreviations used in this paper: IRF, IFN regulatory factor; GOT, glutamic oxalacetic transaminase; GPT, glutamic pyruvic transaminase; ICSBP, IFN consensus sequence binding protein; iNOS, inducible NO synthase; KO, knockout; p.i., postinfection; RNI, reactive nitrogen intermediate; ROI, reactive oxygen intermediate.
4 J. Ko, T. A. Ficht, and G. A. Splitter. Brucella abortus dose and virulence dependent responses in interferon regulatory factor-1 deficient mice. Submitted for publication.
Received for publication July 27, 2001. Accepted for publication January 9, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. Annu. Rev. Immunol. 15:749.[Medline]
production in IFN regulatory factor-1 knockout mice during endotoxemia is secondary to a loss of both IL-12 and IL-12 receptor expression. J. Immunol. 165:3970. production and type 1 cytokine responses. Immunity 4:471.[Medline]
/) and type 2 nitric oxide synthase regulate the innate immune response to a protozoan parasite. Immunity 8:77.[Medline]
production in T cells mediating chronic resistance to the intracellular pathogen. Toxoplasma gondii. J. Immunol. 165:628. is crucial for surviving a Brucella abortus infection in both resistant C57BL/6 and susceptible BALB/c mice. Immunology 103:511.[Medline]
and interleukin-12 contribute to resistance to the intracellular bacterium Brucella abortus by different mechanisms. Infect. Immun. 64:2782.[Abstract]
in mice lacking IFN regulatory factor-1. J. Immunol. 163:6820.This article has been cited by other articles:
|
|
 |
|
  A. A. Rambow-Larsen, G. Rajashekara, E. Petersen, and G. Splitter Putative Quorum-Sensing Regulator BlxR of Brucella melitensis Regulates Virulence Factors Including the Type IV Secretion System and Flagella J. Bacteriol., May 1, 2008; 190(9): 3274 - 3282. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. C. Macedo, D. M. Magnani, N. B. Carvalho, O. Bruna-Romero, R. T. Gazzinelli, and S. C. Oliveira Central Role of MyD88-Dependent Dendritic Cell Maturation and Proinflammatory Cytokine Production to Control Brucella abortus Infection J. Immunol., January 15, 2008; 180(2): 1080 - 1087. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Turcotte, S. Gauthier, D. Malo, M. Tam, M. M. Stevenson, and P. Gros Icsbp1/IRF-8 Is Required for Innate and Adaptive Immune Responses against Intracellular Pathogens J. Immunol., August 15, 2007; 179(4): 2467 - 2476. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Copin, P. De Baetselier, Y. Carlier, J.-J. Letesson, and E. Muraille MyD88-Dependent Activation of B220-CD11b+LY-6C+ Dendritic Cells during Brucella melitensis Infection J. Immunol., April 15, 2007; 178(8): 5182 - 5191. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. P. Jimenez de Bagues, S. Loisel-Meyer, J.-P. Liautard, and V. Jubier-Maurin Different Roles of the Two High-Oxygen-Affinity Terminal Oxidases of Brucella suis: Cytochrome c Oxidase, but Not Ubiquinol Oxidase, Is Required for Persistence in Mice Infect. Immun., January 1, 2007; 75(1): 531 - 535. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Loisel-Meyer, M. P. Jimenez de Bagues, E. Basseres, J. Dornand, S. Kohler, J.-P. Liautard, and V. Jubier-Maurin Requirement of norD for Brucella suis Virulence in a Murine Model of In Vitro and In Vivo Infection Infect. Immun., March 1, 2006; 74(3): 1973 - 1976. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Pappas, N. Akritidis, M. Bosilkovski, and E. Tsianos Brucellosis N. Engl. J. Med., June 2, 2005; 352(22): 2325 - 2336. [Full Text] [PDF]
|
|
|
|
|
|
S.-H. Baek, G. Rajashekara, G. A. Splitter, and J. P. Shapleigh Denitrification Genes Regulate Brucella Virulence in Mice J. Bacteriol., September 15, 2004; 186(18): 6025 - 6031. [Abstract] [Full Text] [PDF]
|
|
2.5-log difference at 12 days p.i. (p
< 0.005), indicating the serious bacterial burden in
IRF-1-/- mice. However, attenuated RB51 was controlled by
IRF-1-/- mice.
