| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
B in Leukocyte Recruitment1
*
Center for Cancer Research and Department of Biology, Massachusetts Institute of Technology, Cambridge, MA 02139;
Physiology Program, Harvard School of Public Health, Boston, MA 02115;
Division of Immunology Research, Department of Pathology, Brigham and Women’s Hospital, and Division of Emergency Medicine, Children’s Hospital, Boston, MA 02115;
Department of Pathology, Tufts University School of Medicine and Veterinary Medicine, Boston MA 02111;
¶ Department of Biological Sciences, Columbia University, New York, NY 10027;
||
Biogen, Cambridge, MA 02142;
#
California Institute of Technology, Pasadena, CA 91125.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B binding sites are present in the promoter regions of many
acute phase and inflammatory response genes, suggesting that NF-B
plays an important role in the initiation of innate immune responses.
However, targeted mutations of the various NF-B family members have
yet to identify members responsible for this critical role.
RelA-deficient mice die on embryonic day 15 from TNF-
-induced liver
degeneration. To investigate the importance of RelA in innate immunity,
we genetically suppressed this embryonic lethality by breeding the RelA
deficiency onto a TNFR type 1 (TNFR1)-deficient background.
TNFR1/RelA-deficient mice were born healthy, but were susceptible to
bacterial infections and bacteremia and died within a few weeks after
birth. Hemopoiesis was intact in TNFR1/RelA-deficient newborns, but
neutrophil emigration to alveoli during LPS-induced pneumonia was
severely reduced relative to that in wild-type or TNFR1-deficient mice.
In contrast, radiation chimeras reconstituted with RelA or
TNFR1/RelA-deficient hemopoietic cells were healthy and demonstrated no
defect in neutrophil emigration during LPS-induced pneumonia. Analysis
of RNA harvested from the lungs of mice 4 h after LPS insufflation
revealed that the induction of several genes important for neutrophil
recruitment to the lung was significantly reduced in
TNFR1/RelA-deficient mice relative to that in wild-type or
TNFR1-deficient mice. These results suggest that TNFR1-independent
activation of RelA is essential in cells of nonhemopoietic origin
during the initiation of an innate immune
response. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B/Rel family of transcription factors is believed to be an
important regulator of innate immunity in species as diverse as insects
and mammals (reviewed in Refs. 1, 2, 3). The mammalian
transcription factors are homodimeric and heterodimeric complexes of
five family members, p50 (NF-B1), p52 (NF-B2), c-Rel, RelB, and
RelA (p65), that are held inactive in the cytoplasm by association with
IB inhibitory proteins. Activation of the cell with a wide variety
of different stimuli leads to degradation of IB and nuclear
translocation of NF-B, resulting in the transcription of multiple
target genes necessary for acute phase, inflammatory and immune
responses (reviewed in Refs. 4, 5, 6). Mice deficient for one
or more of these family members have been generated, revealing both
redundant and nonredundant biological roles for the different proteins
(7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17). These roles include promoting cell survival,
regulating hemopoiesis, and controlling innate and adaptive immune
responses (reviewed in Refs. 18 and 19). The role of RelA in immune protection remains unclear because targeted mutation of RelA results in lethality on embryonic day 15 (E15)5 from extensive liver degeneration (7). Studies of the RelA-deficient immune system using radiation chimeras generated with fetal liver-derived hemopoietic progenitors revealed that a RelA-deficient hemopoietic system develops normally, but that lymphocyte responses are impaired (9, 20). Nonetheless, these chimeras were relatively healthy, suggesting that RelA function is not essential in hemopoietic cells for innate immune protection. However, this does not obviate the potential importance of RelA in the response of nonhemopoietic tissue to infection.
Mouse embryonic fibroblast cells and 3T3 cells cultured from
RelA-deficient animals undergo apoptosis when treated with TNF-
(21), as do Jurkat cells (22), human and
mouse fibroblasts (22), and HT1080 cells (23)
that express dominant-negative mutants of IB. These results
revealed that TNF--induced activation of NF-B, and specifically
RelA, protects cells from the cytocidal effects of TNF- in vitro and
suggested that genetically abrogating the apoptotic TNF- signal
might suppress the extensive apoptosis in the RelA-deficient fetal
liver and rescue the RelA-deficient mice. However, it was unclear which
TNF receptor, TNFR1 (p55) or TNFR2 (p75), elicited the apoptotic
response that must be counteracted by NF-B activity for cells to
survive. Both can mediate activation of NF-B, and both have been
implicated in programmed cell death.
The extensive apoptosis and subsequent liver degeneration observed in
the RelA-deficient fetuses can be genetically suppressed by breeding
the RelA deficiency into a TNF--deficient background
(24), thereby illustrating that TNF- is cytotoxic to
cells in the absence of RelA in vivo as well as in vitro. Furthermore,
it has been shown that the absence of TNFR1 suppresses RelA-like fetal
liver apoptosis observed in mice deficient for IKK2 (25),
one of the signaling molecules implicated in NF-B activation.
Finally, it has recently been reported that the absence of TNFR1
genetically suppresses the phenotype of the RelA-deficient mouse,
indicating that TNFR1 is the mediator of this cytotoxic TNF- signal
(26), and that the resulting TNFR1/RelA-deficient mouse is
susceptible to endogenous hepatic infection. We report these results as
well, but also extend our observations to demonstrate that this
susceptibility to infection is not liver specific, but, rather, is a
more global phenomenon, revealing a critical role for the NF-B
family member RelA in the innate immune response.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
TNFR1/RelA-deficient mice were established from TNFR1-deficient and RelA heterozygous mice (both 129/Sv x C57BL/6J) and subsequently maintained by intercrosses (27, 28). Mice were genotyped by PCR amplification of tail DNA. DNA was prepared by digesting samples 6 h to overnight at 55°C in 400 ml 67 mM Tris (pH 8.8), 16.6 mM ammonium sulfate, 6.5 mM MgCl2, 0.5% Triton X-100, 1% 2-ME, and 0.2 mg/ml proteinase K, then heating at 95°C for 5 min. One microliter of the supernatant was used per 25 µl reaction. Amplification of the RelA locus was by PCR (1 min at 94°C and 2.5 min at 66°C for 30 cycles) using three primers, 5'-AAT CGG ATG TGA GAG GAC AGG-3', 5'-CCT ATA GAG GAG CAG CGC GGG-3', and 5'-AAA TGT GTC AGT TTC ATA GCC TGA AGA ACG-3', that recognized the intact RelA locus (oligonucleotides 1 and 2) and the targeted RelA-neo locus (oligonucleotides 2 and 3). Amplification of the TNFR1 locus was achieved by PCR (1 min at 94°C, 0.5 min at 63°C, and 1.5 min at 72°C for 30 cycles) using three primers, 5'-TGT GAA AAG GGC ACC TTT ACG GC-3', 5'-GGC TGC AGT CCA CGC ACT GG-3', and 5'-ATT CGC CAA TGA CAA GAC GCT GG-3', that recognized the intact TNFR1 locus (oligonucleotides 1 and 2) and the neo-targeted TNFR1 locus (oligonucleotides 2 and 3).
TUNEL assay
Wild-type, RelA+/-, and TNFR1-/-RelA+/- females were impregnated by matched males and sacrificed 15.5, 16.5, and 17.5 days postcoitum by cervical dislocation. Livers were harvested from fetuses and fixed in buffered formalin at 4°C for 2 h, embedded in paraffin, and sectioned (4–6 µm). Sections were deparaffinized and washed in water. They were then preincubated for 5 min in 10 mM Tris (pH 8.0)-20 mM EDTA, incubated for 10 min in 10 µg/ml proteinase K in 10 mM Tris (pH 8.0)-20 mM EDTA, and rinsed in water. Fragmented ends of DNA were labeled by incubating 1 h at 37°C in 1x TdT buffer with 15 U TdT (Life Technologies, Gaithersburg, MD) and 5 nM biotinylated dUTP (Roche, Indianapolis, IN). The reaction was stopped by washing twice in 2x SSC and rinsing in PBS. Sections were blocked for 1 h with 2% BSA in PBS, then rinsed again in PBS. Biotinylated dUTP was visualized using an alkaline phosphatase Vectastain ABC kit (Vector Laboratories, Burlingame, CA).
Tissue culture
Viability assays of fibroblastic and 3T3 cells were performed
essentially as previously described (21). Briefly, 200,000
cells of each genotype were plated per well of a six-well dish in DMEM
supplemented with 10% FBS. Twenty-four hours later the medium was
replaced with DMEM-0.1% FBS with or without 10 ng/ml mouse TNF-
(mTNF-; sp. act., 6.0 x 107 U/mg; Roche)
and incubated 8 or 24 h. The cells were trypsinized and assayed
for trypan blue exclusion. Percentage of viability was defined as the
number of cells remaining per well following serum starvation in the
presence of mTNF- divided by the number of cells remaining per well
following serum starvation in the absence of mTNF-. Two cell lines
were used per genotype, and each experiment was performed twice.
Histology
Moribund mice were sacrificed by CO2 inhalation, and tissues were fixed by immersion in Bouin’s solution (Sigma, St. Louis, MO) for 3 days. Tissues were embedded in paraffin, sectioned (4–6 µm thick), stained with H&E, and examined by light microscopy.
Flow cytometry
Spleen and thymus were harvested and prepared as single-cell suspensions by crushing between two slides and filtering through sterile mesh cell strainers (Applied Scientific, South San Francisco, CA). Bone marrow was harvested from the femur, passaged through a 26-gauge needle to make a single-cell suspension, and filtered through a cell strainer. Blood was collected from the severed necks of postnatal day (P) 3–5 pups and from the inferior vena cava in radiation chimeras. Bronchoalveolar lavage of radiation chimeras was performed as previously described (29). All samples were subjected to red cell lysis with ammonium chloride before staining.
For flow cytometric analysis, cells were first incubated with Fc-block
(anti-CD32/CD16, Fc
II/IIIR, 2.4G2; BD PharMingen, San Diego, CA)
for 5 min. They were then incubated with combinations of the following
primary and secondary Abs: anti-GR-1-biotin (Ly-6G; BD PharMingen),
anti-Mac-1-FITC (M1/70; BD PharMingen), anti-TER-119-biotin (BD
PharMingen), anti-Pan-NK-FITC (Dx5; BD PharMingen),
anti-B220-biotin (RA3-6B; BD PharMingen), anti-IgM-FITC
(R6-60.2; BD PharMingen), anti-CD4-PE (RM4-5; BD PharMingen),
anti-CD8-FITC (Ly-2; Caltag, South San Francisco, CA),
anti-F480-biotin (Caltag), anti-CD45.1-FITC (A20; BD
PharMingen), anti-CD45.2-FITC (104; BD PharMingen), streptavidin-PE
(BD PharMingen), and streptavidin CyChrome (BD PharMingen; used with
anti-F480-biotin only). Three mice of each genotype were analyzed;
shown is a representative plot of each genotype.
Reconstitution of bone marrow with fetal liver cells
For radiation chimera experiments, TNFR1+/-RelA+/- males and females were crossed to generate TNFR+/+RelA+/- and TNFR1-/-RelA+/- mice. Donor embryos were then generated by crossing TNFR+/+RelA+/- males to females and TNFR1-/-RelA+/- males to females. Fetal livers were harvested from day 14.5 embryos, and prepared and genotyped as previously described (12). Meanwhile, 6- to 8-wk-old C57BL/6 CD45.1+ hosts were delivered two doses of irradiation (800 and 400 rad, separated by 3 h) using a 137Cs source. Mice were anesthetized with avertin (2.5% solution of 2,2,2-tribromoethanol-tert amyl alcohol, 12 µl/g mouse) immediately after the second irradiation and transplanted with 1 x 106 liver cells from wild-type, TNFR1-deficient, RelA-deficient, or TNFR1/RelA-deficient fetuses in 200 µl medium by retro-orbital injection with a 26-gauge needle. The extent of reconstitution was analyzed by flow cytometry 4 wk after transplantation, and LPS-pneumonia experiments were performed 6 wk after transplantation.
Thioglycolate-induced peritonitis
Mice, aged 14–16 days, were anesthetized with an i.p. injection of avertin and retro-orbitally bled with Unopettes (BD Biosciences, Mountain View, CA). Blood cells were quantitated with a hemocytometer according to the manufacturer’s instructions to determine total circulating leukocyte counts. Immediately after being bled, mice received an i.p. injection of sterile 2.98% thioglycolate broth (50 µl/g mouse; Sigma) to induce peritonitis. Mice were sacrificed with CO2 6 or 48 h later, and the peritonea were washed with 3 ml ice-cold lavage solution (0.1% BSA-0.65 mM EDTA-heparin at 20 U/ml) delivered with a 26-gauge needle. After massaging for 30 s, 2 ml lavage solution was harvested from each peritoneum with an 18-gauge needle. Concentrations of leukocytes and cellular differentials were quantitated by hemocytometer counts and cytospin preparations, respectively.
LPS-induced pneumonia
Pneumonia was induced in mice, aged 3–5 days, by intranasal insufflation. Mice were anesthetized by inhalation of 2% halothane. Ten microliters of a solution of Escherichia coli LPS (2 mg/ml; Sigma L-2880) and colloidal carbon (5%) in PBS was placed into the nares and allowed to be inhaled. Six hours later, mice were sacrificed by an overdose of halothane. Lungs were removed, intratracheally instilled with 6% glutaraldehyde under 22 cm H2O pressure, tied off, and submerged in 6% glutaraldehyde overnight. Fixed tissue was embedded in paraffin, sectioned (5–7 µm), and stained with H&E.
Pneumonia was induced in radiation chimeras by intratracheal instillation as previously described (30). Mice were anesthetized by i.m. injection of ketamine hydrochloride (100 mg/kg) and acepromazine maleate (5 mg/kg). The tracheas were surgically exposed, and 50 µl of a solution of E. coli LPS (2 mg/ml) and colloidal carbon (5%) was instilled intratracheally. Six hours later lungs were removed and prepared for analysis as described above.
Neutrophil emigration was quantitated by morphometry of histological sections essentially as previously described (30). Briefly, a multipurpose test system containing 42 points was reflected onto the microscope field using a drawing tube. Fields of pneumonic peripheral lung were randomly selected for analysis. As LPS is only heterogeneously deposited into lungs following intranasal insufflation, sections that did not contain colloidal carbon-laden macrophages were not scored, and another region was randomly selected. The 42 points of the counting grid were classified as landing on 1) air space or tissue and 2) neutrophil or not a neutrophil. Ten fields (420 points) were assayed per lung. Statistical significance was determined by ANOVA.
Emigration was quantitated in at least six P3–5 mice per genotype, and in at least four radiation chimeras per donor genotype. P3–5 mice that were runted and lethargic were excluded from analysis, as were those with chronic organizing pneumonia, as these observations suggested preexisting illness. In addition, lungs of mice that had received a poor instillation of LPS, as defined by the macroscopic absence of colloidal carbon, were not analyzed.
RNA analysis
Mice, aged P3–5, were intranasally insufflated with E. coli LPS as described above. Four hours later pups were sacrificed by decapitation, and the lungs were excised and snap-frozen in liquid nitrogen. When all tissue samples had been dissected and frozen, samples were simultaneously thawed and homogenized with a Polytron homogenizer (Brinkmann, Westbury, NY) in 1 ml TRIzol reagent (Life Technologies). Total RNA was extracted and precipitated according to the manufacturer’s instructions.
For Northern blot analysis, 20 µg RNA from five animals of each genotype was electrophoresed separately on a 1.2% agarose gel containing 0.38 M formaldehyde, and transferred overnight by standard capillary action to a Hybond N+ membrane filter. ICAM-1 message was detected using a probe generated by RT-PCR that spanned from the third Ig repeat to the 3' untranslated region. Superscript II reverse transcriptase (Life Technologies) and an oligo(dT) primer were used for first-strand synthesis of 10 µg total RNA from an LPS-treated wild-type lung sample. ICAM-1 cDNA was amplified using 1/25th of this reaction mix, Pfu polymerase (Stratagene, La Jolla, CA), and oligonucleotide primers (5'-GCG GAT CCG ATC TTC CAG CTA CCA TCC CAA AG-3' and 5'-GCG AAT TCG TTC TGT GAC AGC CAG AGG AAG TG-3'). Probes for macrophage inflammatory protein 2 (MIP2) and KC were those used previously (31). The GAPDH transcript was detected with a probe to human cDNA (Clontech, Palo Alto, CA). All probes were randomly labeled to a sp. act. of at least 1 x 109 dpm/µg with a Prime-It RmT Random Primer Labeling Kit (Stratagene) and were cleaned over a Sephadex G-25 column (Roche). Prehybridization and hybridization were performed with Church buffer plus 10% dextran sulfate. Hybridization of all probes was conducted overnight at 65°C. After hybridization, blots of ICAM-1 and GAPDH were washed at 65°C with two 15-min low stringency washes in 2x SSC/1% SDS, followed by two 30-min high stringency washes in 0.1x SSC/1% SDS. Blots of MIP2 and KC received two low stringency washes and one high stringency wash. Blots were exposed to film as well as scanned on a phosphor imager for quantitation with ImageQuant software (BD Biosciences). Area was quantified by a line 11 pixels wide, after which background was subtracted. Three animals were quantitated per genotype.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RelA heterozygous mice (7) were bred with TNFR1 homozygous mutant mice (27), and the F1 progeny subsequently bred to generate TNFR1/RelA homozygous mutant mice, which were born in the expected Mendelian ratios. Gross dissection and serial tissue sectioning and staining with H&E revealed the absence of lymph nodes and Peyer’s patches and a disorganized splenic white pulp.6 No other histological or morphological abnormalities were observed.
The viability of the TNFR1/RelA-deficient mice demonstrated that TNFR1
mediates the signals that induce embryonic lethality in the
RelA-deficient mice. To determine whether and to what extent apoptosis
in the fetal liver was suppressed, fetal livers from day 15–17
wild-type, RelA-deficient, or TNFR1/RelA-deficient embryos were
harvested and analyzed by TUNEL and H&E staining. As previously
reported (7), significant apoptosis was readily observed
in the livers of most E15.5 RelA-deficient fetuses, and no viable E16
RelA fetuses could be recovered. In contrast, livers from wild-type and
TNFR1/RelA-deficient E15.5, E16, and E17 fetuses displayed
substantially less apoptosis (Fig. 1
A). This suggests that TNFR1
mediates cytotoxicity of TNF- in the RelA-deficient liver. To
determine whether this were true in other cell types, we treated
wild-type, RelA-deficient, TNFR1-deficient, and TNFR1/RelA-deficient
mouse embryonic fibroblast and 3T3 lines with mTNF-, which
stimulates both TNFR1 and TNFR2, and assayed for viability by trypan
blue exclusion. RelA-deficient cultures were significantly apoptotic
8 h following treatment with mTNF-, but TNFR1/RelA-deficient
mouse embryonic fibroblasts (data not shown) and 3T3s (Fig. 2) were as resistant to the cytotoxic
potential of mTNF- as wild-type or TNFR1-deficient cells.
|
|
Although TNFR1/RelA-deficient mice were born healthy, they began
to appear runted, lethargic, and unkempt within a few days after birth,
and 95% died by P20. Necropsy of moribund animals revealed the
development of organized pneumonia (Fig. 3A) as well as bacterial
aggregates associated with minimal inflammatory responses in the
pharynx, s.c. fat, marrow, and gut (Fig. 3, B and
C). Blood, lung, and liver tissue collected from five
moribund mice was cultured for microorganisms; all preparations yielded
Pasteurella pneumotropica.
|
D)
instead of the lung. These lesions exhibited a paucity of neutrophils
and an unusual accumulation of eosinophils. All five of the animals
assayed from this cleaner facility were bacteremic as well, although
blood microcultures yielded 
-hemolytic streptococcus and
Staphylococcus aureus instead of Pasteurella.
Thus, the premature mortality of the TNFR1/RelA-deficient mice may be
due to a susceptibility to infection by both common and uncommon mouse
pathogens. Hematopoiesis in TNFR1/RelA-deficient mice
The poor immune response to infection in TNFR1/RelA-deficient mice
could be due to the aberrant production of leukocyte populations, as
has been observed in PU.1-deficient (32) and
C/EBP-deficient mice (33). To determine whether
hemopoiesis was intact in the absence of TNFR1 and RelA, the cellular
compositions of primary and secondary hemopoietic tissues from P3
wild-type, TNFR1-deficient, and TNFR1/RelA-deficient mice were compared
by flow cytometry (Fig. 4). The data
indicate no substantial difference in the relative distribution of
leukocyte populations in thymus, spleen, and peripheral blood in
TNFR1/RelA-deficient mice. Total circulating leukocyte counts in
peripheral blood were also comparable among newborns of the three
genotypes (Table I), indicating no
deficiency in generating appropriate numbers of circulating cells.
However, leukocytosis was observed in the blood of TNFR1/RelA-deficient
mice by P14 (Table I). In contrast, radiation chimeras generated with
TNFR1/RelA-deficient hemopoietic progenitor cells did not have
increased circulating leukocyte counts relative to those generated with
wild-type or TNFR1-deficient progenitors (data not shown), suggesting
that the leukocytosis in the TNFR1/RelA-deficient mice was not due to a
cell intrinsic defect in hemopoietic cells.
|
|
An attenuated inflammatory response to the spontaneous bacterial
overgrowth observed in soft tissues could be due to a difficulty
advancing leukocytes to sites of infection. To determine whether
leukocyte recruitment was impaired in TNFR1/RelA-deficient mice, we
measured neutrophil emigration into the peritoneal cavity of P15 mice
following peritonitis induced by the sterile irritant thioglycolate.
The number of leukocytes recovered by lavage before thioglycolate
injection was elevated 2-fold in the TNFR1/RelA-deficient mice,
although this was predominantly due to an increased number of
eosinophils (data not shown). Substantial neutrophil accumulation in
the peritoneal cavity of wild-type mice could be observed 6 h
after thioglycolate injection; by 48 h these numbers had decreased
to some extent (Fig. 5). In contrast,
significantly fewer emigrated neutrophils were recovered from
TNFR1-deficient mice at 6 h, and numbers comparable to wild-type
numbers were recovered at 48 h (Fig. 5). These data demonstrated
that TNFR1 is an important mediator of thioglycolate-induced neutrophil
recruitment into the peritoneal cavity. TNFR1/RelA-deficient mice
displayed a similar defect in neutrophil emigration, suggesting that
RelA activity independent of TNFR1 signaling is not important for
neutrophil recruitment to the peritoneum.
|
To determine whether TNFR1-independent RelA activity was important
for leukocyte recruitment to the lungs, we analyzed neutrophil
emigration into pulmonary air spaces of P3–5 mice following delivery
of E. coli LPS by intranasal insufflation. Before treatment,
neutrophils were not observed in the uninflamed alveolar air spaces of
the P3–5 TNFR1/RelA-deficient mice or wild-type controls (data not
shown). Six hours after delivery, LPS-induced neutrophil emigration was
observed in wild-type and TNFR1-deficient mice and did not
significantly differ between these genotypes. In contrast, mice
deficient for both TNFR1 and RelA showed significantly less neutrophil
emigration compared with either wild-type or TNFR1-deficient mice (Fig. 6), suggesting that RelA activity
independent of TNFR1 signaling is essential for initiating maximal
neutrophil emigration in response to LPS in the pulmonary air
spaces.
|
To determine whether TNFR1/RelA-deficient neutrophils were capable
of being recruited into inflammatory foci, lethally irradiated C57BL/6
CD45.1+ hosts were engrafted with wild-type or
TNFR1-, RelA-, or TNFR1/RelA-deficient fetal liver hemopoietic
progenitor cells, and the reconstituted mice were assayed for their
ability to recruit neutrophils during LPS-induced pneumonia. Flow
cytometry confirmed that virtually all peripheral blood neutrophils
were donor derived (CD45.2+; Fig. 7A). Six hours after delivery
of LPS, neutrophil emigration was comparable in all radiation chimeras
(Fig. 7B), suggesting that the emigration defect in the
TNFR1/RelA-deficient mice is not a cell intrinsic defect of
RelA-deficient neutrophils. Interestingly, the vast majority of
alveolar macrophages, which are critical for inciting responses to
bacteria in the lungs (34, 35), were donor derived as well
(Fig. 7A). This suggests that RelA is not specifically
required in alveolar macrophages for the induction of neutrophil
emigration into the lungs in response to LPS.
|
ICAM-1, KC, and MIP2 often regulate neutrophil recruitment to the
lungs, particularly in response to LPS (36, 37, 38). To
determine whether the transcriptional induction of these and other
inflammatory genes by LPS was altered in the absence of RelA, RNA was
collected from the lungs of wild-type, TNFR1-deficient, and
TNFR1/RelA-deficient mice following insufflation of LPS and was
analyzed by Northern blot. Four hours after LPS insufflation, levels of
expression of ICAM-1, KC, and MIP2 were somewhat reduced in
TNFR1-deficient mice relative to those in wild-type controls and were
substantially reduced in TNFR1/RelA-deficient mice relative to those in
wild-type TNFR1-deficient mice (Fig. 8).
These data demonstrate that in lung stromal cells, representative genes
for chemokine and adhesion proteins show reduced expression in the
absence of RelA.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-mediated apoptosis.
Taken together, these results indicate that TNFR1 is the primary
receptor mediating TNF- cytotoxicity in RelA-deficient cells.
We also report here that the antibacterial host defenses of
TNFR1/RelA-deficient mice were severely compromised. Unlike mice with
targeted deficiency of the NF-B family members RelB (8, 16) or p50/RelB (17) or the IB family member
IB (39), which develop chronic inflammation in the
absence of bacterial infections, all moribund TNFR1/RelA-deficient mice
presented with numerous inflammatory lesions and septicemia. The
strains of infecting bacteria, the sites of infection, and the age of
mortality were constant among animals raised in the same facility, but
varied from facility to facility (data not shown). Therefore,
progression of the infection was most likely directed by the type of
bacteria to which the animal was first exposed, rather than by a
susceptibility to bacterial infection in a specific organ or tissue.
TNF-/RelA-deficient mice exhibit a similar immune phenotype,
although the onset of symptoms is delayed by several weeks
(24). Given the variation in phenotype observed in our
colony depending on the infecting pathogen, it is likely that the
differences in phenotypes of the TNF-/RelA-deficient and
TNFR1/RelA-deficient mice are due to differences in pathogen exposure
as well.
Although the development of secondary lymphoid organs was perturbed in TNFR1/RelA-deficient mice, such abnormalities should not compromise the innate immune response to the extent observed in these mice, suggesting that other aspects of host defense were also perturbed in the absence of TNFR1 and RelA. Large bacterial colonies devoid of leukocytes were present in various soft tissues despite normal or elevated circulating leukocyte counts, suggesting that leukocytes were unable to migrate to infected sites efficiently in the absence of TNFR1 and RelA. Therefore, we measured leukocyte recruitment in the TNFR1/RelA-deficient mice in two different tissues. Neutrophil emigration induced in the peritoneum by thioglycolate injection was significantly reduced by deficiency of TNFR1 alone, demonstrating that TNFR1 plays an important role in this inflammatory process. Thus, if RelA activity is important for neutrophil recruitment to the peritoneum, it is in response to TNFR1-mediated signaling. In contrast, inflammatory responses to LPS in the lungs were not reduced in the absence of TNFR1, allowing us to assess the importance of TNFR1-independent RelA activity in this inflammatory process. TNFR1/RelA-deficient newborns displayed a significant impairment in their ability to recruit neutrophils to the lung during a 6-h exposure to LPS relative to wild-type or TNFR1-deficient newborns, suggesting that early neutrophil emigration to the lung requires RelA activity that is independent of TNFR1 signaling.
To determine whether RelA-deficient neutrophils were capable of being recruited, neutrophil emigration was assayed in radiation chimeras. These experiments demonstrated that RelA is not intrinsically required in neutrophils or alveolar macrophages for neutrophil recruitment. Although a population of radiation-resistant host-derived T lymphocytes remained in the bronchoalveolar lavage fluid, which could have contributed to eliciting neutrophil emigration, these results more likely suggest that the production of recruitment mediators by parenchymal cells is sufficient for neutrophil recruitment and that the recruitment defect in TNFR1/RelA-deficient mice lies in this nonhemopoietic compartment. We attempted to confirm that the production of recruitment mediators by parenchymal cells was necessary by performing the reciprocal radiation chimera experiment, in which wild-type, TNFR1-deficient, and TNFR1/RelA-deficient mice were reconstituted with wild-type hemopoietic cells. However, although wild-type and TNFR1-deficient hosts were successfully reconstituted and appeared healthy 6 wk following transplant, transplanted TNFR1/RelA-deficient siblings did not survive the 2 wk necessary for reconstitution to be completed (E. Alcamo, unpublished observations), preventing further analysis of the importance of RelA in nonhemopoietic cells.
Previous studies have demonstrated the presence of TNFR1-independent pathways for recruitment in the lung (40, 41, 42, 43). This study suggests that RelA is an important mediator of these pathways. In support of this, the induction of ICAM-1, MIP2, and KC, proteins that are required for LPS-induced neutrophil emigration into the lung, was reduced to a greater degree in the absence of RelA and TNFR1 than in the absence of TNFR1 alone. Thus, in addition to the TNFR1-dependent mechanisms for mounting an innate immune responses that may or may not require RelA, RelA is essential for regulating TNFR1-independent mechanisms of innate immunity that are indispensable for protection against pathogens.
Although leukocyte recruitment to the lungs 6 h following LPS
insufflation is a RelA-dependent process, the spontaneous pneumonia
that developed in TNFR1/RelA-deficient mice clearly demonstrates the
presence of RelA-independent pathways mediating neutrophil emigration.
It is possible that although the initiation of pulmonary inflammatory
responses to LPS are RelA dependent, RelA-independent mechanisms for
recruiting neutrophils develop over time, just as neutrophil emigration
elicited by sterile irritants or bacteria in the peritoneal cavity
progresses from being dependent on E-selectin, P-selectin, ICAM-1, and
2 integrins at early time points to becoming independent
of these same molecules after 24 h (44, 45, 46).
Alternatively, it is possible that although LPS-induced inflammatory
responses are RelA dependent, inflammatory responses induced by other
organisms are not. For example, previous studies indicate that
neutrophil emigration in the lung is CD18 dependent in response to LPS
or E. coli, but CD18 independent in response to
Streptococcus pneumoniae or S. aureus (reviewed
in Ref. 47).
In conclusion, the work presented here demonstrates that TNFR1 is
responsible for mediating TNF- signals that in the absence of RelA
are cytotoxic to fetal hepatocytes and result in embryonic lethality.
Furthermore, the results demonstrate the importance of RelA in inducing
an innate immune response. The innate host defense system is
responsible for providing protection against the vast majority of
infectious micro-organisms by recognizing and removing pathogens within
hours of contact and activating the adaptive arm of the immune system
should infection persist (reviewed in Refs. 48, 49, 50). As
past studies have shown, deficiencies for hemopoietic cells of the
myeloid lineage or molecules that are involved in mounting an innate
immune response can be fatal (33, 51, 52, 53, 54). The results
presented here demonstrate that RelA plays a critical role in
activating the transcription of molecules that are important for
leukocyte emigration, an essential step in the development of an innate
immune response.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Current address: Department of Biological Sciences, Stanford University, Stanford, CA 94305.
3 Current address: Rainbow Babies and Children’s Hospital, Cleveland, OH 44106.
4 Address correspondence and reprint requests to Dr. David Baltimore, MC 204-31, California Institute of Technology, Pasadena, CA 91125. E-mail address: baltimo{at}its.caltech.edu
5 Abbreviations used in this paper: Ex, embryonic day x; TNFR1, TNF receptor 1; Px, postnatal day x; MIP2, macrophage inflammatory protein 2; mTNF-, mouse TNF-.
6 E. Alcamo, R. O. Hynes, and D. Baltimore. Submitted for publication.
Received for publication December 12, 2000. Accepted for publication May 17, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B and Rel proteins: evolutionarily conserved mediators of immune responses. Annu. Rev. Immunol. 16:225.[Medline]
B and Rel proteins in innate immunity. Adv. Immunol. 58:1.[Medline]
B puzzle?. Curr. Biol. 8:R19.[Medline]
B. Immunol. Today 19:80.[Medline]
B/IB family: intimate tales of association and dissociation. Genes Dev. 9:2723.B. Nature 376:167.[Medline]
B RelA-deficient lymphocytes: normal development of T cells and B cells, impaired production of IgA and IgG1 and reduced proliferative responses. J. Exp. Med. 185:953.B/p52 present with defects in humoral responses, germinal center reactions, and splenic microarchitecture. J. Exp. Med. 187:147.B-deficient fetal liver cells. Immunity 6:765.[Medline]
B1 and NF-B2. Nat. Med. 3:1285.[Medline]
B leads to multifocal defects in immune responses. Cell 80:321.[Medline]
B/Rel family. Cell 80:331.[Medline]
B complexes partially compensate for the absence of RelB: severely increased pathology in p50-/-relB-/- double-knockout mice. J. Exp. Med. 185:1359.B: a pivotal transcription factor in chronic inflammatory diseases. N. Engl. J. Med. 336:1066.B/Rel transcription factor: [Published erratum appears in 1998 J. Exp. Med. 187:661.]. J. Exp. Med. 187:143.B is redundant with p50 during B cell proliferative responses, and is required for germline CH transcription and class switching to IgG3. J. Immunol. 162:1941.B in preventing TNF--induced cell death. Science 274:782.B: association with NF-B and the PEST sequence in IB are not required. Mol. Cell. Biol. 16:6037.[Abstract]
B. Science 274:784.B kinase 2 gene. Science 284:321.B/Rel DNA-binding complexes by tumor necrosis factor. Oncogene 9:1487.[Medline]
B in regulation of multiple immune-response genes and in Fas-induced cell death. J. Exp. Med. 189:999.
-deficient mice. Proc. Natl. Acad. Sci. USA 94:13187.B1) but expressing p50. J. Exp. Med. 187:985.This article has been cited by other articles:
|
|
 |
|
  Y. Xu, F. Fang, D. K. St. Clair, P. Sompol, S. Josson, and W. H. St. Clair SN52, a novel nuclear factor-{kappa}B inhibitor, blocks nuclear import of RelB:p52 dimer and sensitizes prostate cancer cells to ionizing radiation Mol. Cancer Ther., August 1, 2008; 7(8): 2367 - 2376. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Gratz, M. Siller, B. Schaljo, Z. A. Pirzada, I. Gattermeier, I. Vojtek, C. J. Kirschning, H. Wagner, S. Akira, E. Charpentier, et al. Group A Streptococcus Activates Type I Interferon Production and MyD88-dependent Signaling without Involvement of TLR2, TLR4, and TLR9 J. Biol. Chem., July 18, 2008; 283(29): 19879 - 19887. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. V. Kravchenko, G. F. Kaufmann, J. C. Mathison, D. A. Scott, A. Z. Katz, D. C. Grauer, M. Lehmann, M. M. Meijler, K. D. Janda, and R. J. Ulevitch Modulation of Gene Expression via Disruption of NF-{kappa}B Signaling by a Bacterial Small Molecule Science, July 11, 2008; 321(5886): 259 - 263. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Dong, E. Jimi, H. Zhong, M. S. Hayden, and S. Ghosh Repression of gene expression by unphosphorylated NF-{kappa}B p65 through epigenetic mechanisms Genes & Dev., May 1, 2008; 22(9): 1159 - 1173. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. E. Simmonds and B. M. Foxwell Signalling, inflammation and arthritis: NF-{kappa}B and its relevance to arthritis and inflammation Rheumatology, May 1, 2008; 47(5): 584 - 590. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Vaira, T. Johnson, A. C. Hirbe, M. Alhawagri, I. Anwisye, B. Sammut, J. O'Neal, W. Zou, K. N. Weilbaecher, R. Faccio, et al. RelB is the NF-{kappa}B subunit downstream of NIK responsible for osteoclast differentiation PNAS, March 11, 2008; 105(10): 3897 - 3902. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. P. Mizgerd Acute Lower Respiratory Tract Infection N. Engl. J. Med., February 14, 2008; 358(7): 716 - 727. [Full Text] [PDF]
|
|
|
|
, Wild-type cultures; 
, TNFR1-deficient
cultures; 
, RelA-deficient cultures; and 
, TNFR1/RelA-deficient
cultures.
