| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,
Divisions of
*
Critical Care Medicine, and
Infectious Diseases, Children’s Hospital Research Foundation, Cincinnati, OH 45229; and
Department of Pediatrics and Sealy Center for Vaccine Development, University of Texas Medical Branch, Galveston, TX 77555.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B
loss poststimulation and NF-B activation in Caco-2BBe cells and was
as potent as native-purified SD flagellin.
IFN-
-primed DLD-1 cells stimulated with 1 µg/ml of 6HIS
flag induced high levels of NO (60 ± 0.95 µM) comparable to the
combination of IL-1
and IFN- (77 ± 1.2) or purified native
SD flag (66.3 ± 0.98). Selected rSD
flagellin proteins representing the D1, D2, or D3 domains alone or in
combination were tested for proinflammatory properties. Fusion proteins
representing the D3, amino, or carboxyl regions alone did not induce
proinflammatory mediators. The results with a recombinant protein
containing the amino D1 and D2 and carboxyl D1 and D2 separated by an
Escherichia coli hinge (ND1-2/ECH/CD2) indicated that D1
and D2 were bioactive when coupled to an ECH element to allow protein
folding. This chimera, but not the hinge alone, induced IB
degradation, NF-B activation, and NO and IL-8 production in two
intestinal epithelial cell lines. ND1–2/ECH/CD2–1 also induced high
levels of TNF- (900 pg/ml) in human monocytes comparable to native
SD flagellin (991.5 pg/ml) and 6HIS flag (987 pg/ml).
The potent proinflammatory activity of flagellin, therefore, resides in
the highly conserved N and C D1 and D2 regions. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(4) in various host cells. The production of NO, as well
as other proinflammatory mediators, contributes to and/or causes host
cell damage (5, 6).
LPS is not the only bacterial component capable of eliciting an
inflammatory response. We (7), and others
(8, 9, 10, 11), have reported that bacterial flagellin proteins
are capable of inducing various inflammatory responses in a variety of
host cells. Recently, we demonstrated that flagellin from
Salmonella dublin
(SD),3
Escherichia coli, and Pseudomonas aeruginosa were
responsible for the induction of IB degradation in cultured
intestinal epithelial cells (IEC; Ref. 7). Further, we
demonstrated that purified SD flagellin is significantly
more potent than LPS in inducing IB degradation, NF-B
activation, and a variety of proinflammatory mediators, including NO,
IL-8, and IL-6 in IEC (7). Other in vitro studies have
demonstrated that P. aeruginosa flagellin induces IL-8
secretion in airway epithelial cells (8) and
Salmonella typhimurium flagellin was shown to induce TNF-
in monocytes (10).
Collectively, the flagellin data indicate that the inflammation cascade
is initiated via the IB/NF-B pathway. The transcription of
many of these proinflammatory mediators, such as NO and IL-8, is
reliant upon the nuclear translocation and activation of NF-B
(12). IEC stimulated by bacteria or various cytokines
induce degradation of IB, the protein inhibitor of NF-B that
sequesters the transcription factor in the cytoplasm of the
unstimulated cells (12, 13). Upon IB degradation,
NF-B translocates to the nucleus and regulates the transcription of
a number of inflammatory genes (12).
To expand upon recent findings, we tested purified SD
flagellin in vivo. A vigorous innate immune response was detected
following a single injection of SD flagellin into C57BL/6
mice (7). This activity was not due to LPS contamination
of the purified flagellin because C3H/HeJ mice that are LPS-resistant,
when injected with purified SD flagellin, also showed
induction of IFN- and TNF- (7). Further, no
inflammatory responses could be measured in LPS-resistant mice injected
with medium obtained from nonflagellated SD mutants
(7), indicating that purified flagellin is a potent
inducer of proinflammatory mediators in vitro and in vivo.
Flagellin is the primary protein component of the highly complex flagellar structures that extend from the outer membrane of Gram-negative organisms. Flagella are the propellers that move bacteria through their aqueous environment and have been shown to aid the bacteria in attaching to host cells, assisting in bacterial invasion and (14, 15) thereby contributing to the virulence of pathogenic bacteria. The flagellin gene sequences from different Salmonella sp., as well as a variety of other Gram-negative bacterial flagellin genes, have been compared and were found to share highly conserved regions at the amino terminus and C terminus (16, 17). A central hypervariable region was present in each sequence that allows for antigenic variation and evasion of the host immune response (18, 19). The flagellin protein has been crystallized (16), allowing structural positioning of amino acid residues and identification of three domains (16, 17, 18, 19). Domains 1 (D1) and 2 (D2) are discontinuous and are formed when residues in the amino terminus and C terminus are juxtaposed by the formation of a hairpin structure. The middle hypervariable domain (D3), serves as the "blade" of the flagellar paddle (17, 18, 19). Therefore, the linear arrangement of the domains is amino domain (ND)1, ND2, D3, carboxyl domain (CD)2, and CD1.
Flagellin from different Salmonella species (7, 9, 10) and other Gram-negative bacteria (8) share
similar inflammatory properties suggesting that some degree of sequence
or structure conservation is responsible for inflammatory response
induction. To determine whether the conserved D1 and D2 regions or the
structure of the variable D3 region contained bioactivity, the SD
fliC flagellin gene was used to generate a series of
recombinant proteins that were tested in several bioassays. A
6-histidine (6HIS) tag was used to assist in purification. Each of the
purified recombinant fusion proteins was tested for its ability to
elicit IB degradation, NF-B activation, IL-8 secretion, and NO
production in cultured human IEC and for TNF- induction in human
monocytes. The results clearly indicated that the bioactivity of
SD flagellin was localized to a region in the stem generated
by the amino- and carboxyl-conserved sequences.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
DLD-1 and Caco-2BBe (American Type Culture Collection,
Manassas, VA) human adenocarcinoma cell lines, were used between
passages 5 and 15. These cell lines were maintained in DMEM
supplemented with 10% FBS, 110 mg/L sodium pyruvate, and antibiotics
(Life Technologies, Grand Island, NY). DLD-1 cells were grown to
confluence on 96-well plates for NO analysis and Caco-2BBe cells were
grown to confluency on 6-well plates or 10-cm plates for IB,
IL-8, and NF-B analysis. Before treatment, the growth medium was
removed and replaced with DMEM without FBS.
The human myelomonocytic cell line, THP-1 (American Type Culture Collection) was cultured in RPMI 1640 medium containing 2 mM of L-glutamine (Life Technologies) supplemented with 10 mM of HEPES, 1 mM of sodium pyruvate, 4.5 g/L glucose, 1.5 g/L sodium bicarbonate, 10% FBS, and 0.05 mM of 2-ME at 37°C in 5% CO2. For experimental purposes, cells were resuspended in supplemented RPMI 1640 without FBS then were seeded at 5 x 106 cells/well in 24-well tissue-culture plates.
Generation of SD flagellin protein and protein fragment constructs
Oligonucleotide primers were designed (Table I
) based upon the SD
fliC flagellin gene sequence (GenBank accession
number M84973) and used for PCR amplimer generation. The amplimers were
generated from SD genomic DNA templates under the following
conditions: 30 cycles of 94°C for 30 s, 55°C for 30 s,
and 72°C for 60 s. A final 72°C 7-min extension completed
amplification. PCR mixes included 15 mM of MgCl2,
100 mM of dNTPs, and 10 U of Taq polymerase (Life
Technologies). Using primer-introduced restriction enzyme sites, the
amplimers were cloned into the pPROEX vector (Life Technologies). The
cloning strategy allowed for unidirectional amplimer insertion
downstream of and in-frame with a 6HIS residue tag and an isopropyl
-D-thiogalactoside-inducible promoter present in pPROEX.
Resulting plasmids were transformed into competent E. coli
(Protein Express, Cincinnati, OH), selected, and then confirmed by
sequence analysis (Cleveland Genomics, Cleveland, OH).
|
.
|
E. coli, containing
selected flagellin constructs, were grown overnight at 37°C in
Luria-Bertani broth with 100 µg/ml ampicillin. Stationary phase
bacteria were transferred to fresh broth, grown to log phase, and then
supplemented with 1 µl/ml of 0.1 M of isopropyl -D-thiogalactoside
to induce fusion protein synthesis. After 4–6 h of induction, the
bacteria were lysed in a buffer containing 50 mM of
NaH2PO4, 300 mM of NaCl, 20
mM of imidazole (pH 7.5). The fusion proteins then were purified from
the lysate by affinity chromatography on a Ni-NTA column (Qiagen,
Valencia, CA). Bound 6HIS-tagged protein was eluted with a buffer
containing 50 mM of NaH2PO4
(pH 8.0), 300 mM of NaCl, and 250 mM of imidazole. Each preparation was
analyzed by SDS-PAGE. Quantification of the purified fusion proteins
was completed by fluorescence using a NanoOrange Protein Quantitation
kit (Molecular Probes, Eugene, OR).
Purification of flagellin directly from SD
Flagellin was purified from SD as previously
described (7, 21). Briefly, SD (American Type
Culture Collection) isolates were incubated in brain-heart infusion
broth at 35°C in an orbital shaker incubator at 80 rpm for 
16 h.
Bacteria were collected by centrifugation at 5,000 x g
for 30 min then mixed with saline solution to form a moderately thick
suspension. The suspension was adjusted to pH 2.0 with 1 M of HCl and
maintained at that pH under constant stirring at room temperature for
30 min. The bacteria, devoid of flagella, then were separated by
centrifugation at 5,000 x g for 30 min. The
supernatant that contained detached flagellin in monomeric form, was
further centrifuged at 100,000 x g for 1 h at
4°C. The pH of the supernatant was adjusted to 7.2 with 1 M of NaOH.
Ammonium sulfate was added slowly with vigorous stirring to achieve
two-thirds saturation (2.67 M) and then the mixture was centrifuged at
15,000 x g for 15 min at 4°C. The precipitate was
dissolved in 5 ml of distilled water and dialyzed for 18 h at
4°C against distilled water using 50,000 m.w. cutoff dialysis tubing.
The dialyzed flagellin then was lyophilized, resuspended in sterile
water, and stored at -70°C until used. This flagellin will be
referred to as purified native SD flagellin throughout
the text.
LPS detection
The chromogenic Limulus amebocyte test (BioWhittaker, Walkersville, MD) was used to determine the level of LPS present in the purified flagellin fusion proteins. Any LPS present in the fusion protein preparations was eliminated by passage over an endotoxin removing gel column (Pierce, Rockford, IL). To further control for potential LPS effects, some studies included IEC pretreatment with 10 µg/ml polymyxin B (PB), an LPS scavenger, before the addition of the flagellin fusion proteins.
NO2-/NO3- detection
DLD-1 cells were pretreated with IFN- (100 U/ml) 2 h
before the addition of selected concentrations of flagellin proteins.
Cell supernatants were collected 18 h after the addition of the
flagellin proteins. The combined concentration of nitrate and nitrite,
the degradation products of NO, was determined in the supernatants by
the Griess reaction following nitrate reduction as described previously
(22). In the text total nitrate/nitrite is referred to
as NO.
IB Western blot analysis
Caco-2BBe cells, grown in 6-well plates, were stimulated with
purified SD flagellin, the full-length 6HIS SD
flagellin (6HIS flag) fusion protein, or one of the other flagellin
region proteins for selected times. Caco-2BBe cells then were lysed in
4°C buffer containing 50 mM of Tris (pH 8.0), 110 mM of NaCl, 5 mM of
EDTA, 1% Triton X-100, and 0.1 mM of PMSF. The amount of protein in
each sample was determined by the Bradford assay (Bio-Rad, Hercules,
CA). Individual cell lysates were boiled in 10 µl of loading buffer
(4% SDS, 20% glycerol, 125 mM Tris-HCl (pH 6.8), and 10% 2-ME), and
50 µg of each protein sample was loaded per lane on an 8–16%
Tris-glycine gradient gel (NOVEX, San Diego, CA). Electrophoresed
proteins were transferred to a nitrocellulose membrane (NOVEX) with the
NOVEX X-cell MiniGel system. Membranes were blocked with 10% nonfat
dried milk resuspended in TBS for 30 min before incubation with rabbit
polyclonal anti-IB antiserum (Santa Cruz Biotechnology, Santa
Cruz, CA) at a dilution of 1/200 for 3 h. Blots were washed twice
for 7 min in TBS supplemented with 0.1% Tween 20 + TBS, followed by
the addition of peroxidase-conjugated anti-rabbit IgG
(Sigma-Aldrich, St. Louis, MO) at a dilution of 1/10,000 for 30 min.
Blots were washed three times for 5 min each with Tween 20 + TBS and
then incubated for 1 min in ECL reagents (ECL kit; Amersham, Little
Chalfont, Buckinghamshire, U.K.). Processed blots were placed on x-ray
film (Kodak, Rochester, NY) for empirically optimized
exposures.
Nuclear protein extraction
Nuclear protein extracts were prepared from Caco-2BBe cells incubated with purified SD flagellin, the 6HIS flag fusion protein, or one of the other flagellin region proteins for 1 h. Cells were washed twice with ice-cold PBS and harvested by scraping into 1 ml of ice-cold PBS. After pelleting at 6,000 rpm for 5 min, cells were washed twice with cold PBS, resuspended in 1 pellet vol of lysis buffer (1.5 mM of MgCl2, 0.2% v/v Nonidet P-40, 10 mM of HEPES (pH 7.9), 10 mM of KCl, 0.1 mM of EDTA, 1 mM of DTT, and 0.1 mM of PMSF), and incubated for 5 min on ice. Nuclei were collected from the cell lysates by centrifuging at 6,000 rpm for 5 min and then were resuspended in 1 pellet vol of extraction buffer containing 1.5 mM of MgCl2, 25% v/v glycerol, 20 mM of HEPES (pH 7.9), 420 mM of NaCl, 0.1 mM EDTA, 1 mM of DTT, and 0.5% PMSF. After incubation on ice for 15 min, nuclear proteins were collected by centrifuging at 14,000 rpm for 15 min. To ensure the elimination of nuclear debris, supernatants were also collected.
EMSA
An NF-B oligonucleotide probe (5'-AGT TGA GGG GAC TTT CCC
AGG-3'; Santa Cruz Biotechnology) was labeled with
[-32P]ATP by using T4 polynucleotide kinase
(Life Technologies) and then purified on a Bio-Spin chromatography
column (Bio-Rad). Nuclear protein extracts were preincubated with EMSA
buffer (1 mM of EDTA, 1 mM of DTT, 12 mM of HEPES (pH 7.9), 4 mM of
Tris-HCl (pH 7.9), 5 mM of MgCl2, 25 mM of KCl,
12% glycerol, 50 ng/ml poly(dI-dC), and 0.2 mM of PMSF) on ice for 10
min, followed by a 20 min incubation with the radiolabeled probe. The
specificity of the binding reaction was determined by incubating
duplicate nuclear protein samples with a 100-fold excess of unlabeled
probe. Samples were resolved on a nondenaturing 5% polyacrylamide gel
and run in 0.5 x Tris/boric acid/EDTA buffer (1 mM of EDTA, 45 mM of
boric acid, and 45 mM of Tris-HCl) at a constant current (30 mA) for
1 h. Gels were transferred to 3M paper (Whatman, Clifton,
NJ), dried under a vacuum for 1 h at 80°C, and exposed to
film overnight at -70°C with the addition of an intensifying
screen.
IL-8 and TNF- ELISA
Following the stimulation of Caco-2BBe or THP-1 cells with
individual SD flagellin fusion protein constructs at a
concentration of 1 µg/ml, cell supernatants were collected at 22
h poststimulation of Caco-2BBe cells for IL-8 detection and 4 h
poststimulation in the THP-1 cells for TNF- detection. The amount of
IL-8 and TNF- in the supernatants was quantified by an ELISA (Pierce
and Endogen, Woburn, MA).
Statistical Analyses
Where applicable, results are expressed as the mean ± SEM of three to four experiments. The Student’s t test was used to compare mean values. Differences were considered significant when p values were <0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To begin the molecular analysis of SD flagellin, 6HIS
flag was tested for biological activity. Confluent Caco-2BBe cells were
stimulated with 1 µg/ml of 6HIS flag; in previous studies
(7), this dose of native purified SD flagellin
consistently provided easily detectable activity and, therefore, was
used throughout these experiments with all recombinant flagellin fusion
proteins. Western blotting of stimulated lysates showed that Caco-2BBe
cells underwent the loss of IB at both 20 and 30 min
poststimulation with replenishment at 60 min comparable to the native
purified SD flagellin protein (Fig. 2A). To confirm nuclear
translocation of NF-B, an expected consequence of IB
degradation, EMSA analyses were completed and showed that 6HIS flag
induced NF-B activation in Caco-2BBe cells at 1 h
poststimulation (Fig. 2B). When IFN--primed DLD-1 cells
were incubated with 1 µg/ml of 6HIS flag for 18 h, similar
levels of NO production were observed when compared with the purified
native SD flagellin (1 µg/ml) or the combination of
IL-1 (0.5 ng/ml) and IFN- (100 U/ml; Fig. 2C). These
levels were significantly greater than untreated control DLD-1 cells
(p < 0.001; Fig. 2C). Further,
Caco-2BBe cells secreted elevated levels of IL-8 following stimulation
for 22 h with 1 µg/ml of 6HIS flag that was equal to purified
native SD flagellin at the same dose (Fig. 2D).
|
-primed DLD-1 cells that were pretreated with 10
µg/ml LPS scavenger, PB, followed by the addition of 1 µg/ml 6HIS
flag or Salmonella LPS, had indistinguishable NO levels when
compared with cells treated with 6HIS flag (Fig. 2
C).
However, PB pretreatment abolished the induction of NO (Fig. 2C) in DLD-1 cells stimulated with 1 µg/ml LPS. Together the data indicate that the 6HIS flag protein retained biological activity comparable to native flagellin establishing the basic construct for further dissection of the flagellin gene.
A flagellin with a D3-region deletion retained biological activity; however, truncated flagellin protein constructs did not induce proinflammatory responses
Using the active 6HIS flag construct as a starting point, an
internal deletion was made by eliminating 100 aa in the D3 region.
This particular deletion was chosen due to convenient restriction
enzyme sites in the fliC gene. The resulting protein was
designated N/C (Fig. 1). Complete IB degradation was observed in
Caco-2BBe cells stimulated with 1 µg/ml 6HIS flag at 20 and 30 min
with replenishment at 60 min as analyzed by immunoblotting (Fig. 3A). Treatment of cells with
the N/C flag also resulted in the loss of IB at 30 min (Fig. 3A), although not as completely as 6HIS flag, and similarly
treated IFN--primed DLD-1 cells showed NO production comparable to
the 6HIS flag and the combination of IL-1 and IFN- (Fig. 3B). In addition, substantial IL-8 secretion also was
observed in Caco-2BBe cells treated for 22 h with 1 µg/ml N/C
flag comparable to the full-length 6HIS flag (Fig. 3C).
These data show that a SD flagellin protein containing a
deletion of a portion of the D3 region retains bioactivity evidenced by
the loss of IB and the production of NO.
|
). None of these three
proteins induced IB degradation (data not shown), NO production
(detection of nitrite/nitrate expressed as micromoles; amino third,
10 ± 0.67 µM; middle third, 15 ± 0.54 µM; and carboxyl
third, 7 ± 0.56 µM vs 6HIS flag, 58 ± 0.85 µM), or IL-8
production (amino third, 10 ± 0.07 pg/ml; middle third, 10
± 0.04 pg/ml; and carboxyl third, 3 ± 0.03 pg/ml vs 6HIS flag,
994 ± 0.85 pg/ml) in either of the IEC lines tested. Two
possible conclusions for these results were considered. First, improper
folding of the truncated proteins may have prevented the linear
bioactive domains from forming. Alternatively, a nonlinear bioactive
domain was not formed because the two portions were not contained in
one of the three proteins. These possibilities were tested with a final
set of constructs. Constructs containing the D3, ND1 and CD1, or ND2 and CD2 regions did not induce proinflammatory responses
The fusion proteins containing the ND1 and CD1 or the ND2 and CD2
regions, without the D3 region, did not induce IB degradation or
NO induction (data not shown). This could have resulted from an
inability to fold correctly or the lack of region formation due to
missing aa sequences. The data suggest that flagellin bioactivity
resides in a discontinuous domain of the protein formed by
juxtaposition of the amino and carboxyl protein regions. To better
establish whether a discontinuous domain generated by protein folding
was responsible for bioactivity, fusion proteins were generated from
constructs containing the D3 region or an ECH (20) cloned
between the ND1 and CD1 (designated ND1/D3/CD1 or ND1/ECH/CD1; Fig. 1)
or ND2 and CD2 regions (designated ND2/D3/CD2 or ND2/ECH/CD2; Fig. 1).
The ECH element has been used successfully in other protein engineering
designs to bring the amino and carboxyl elements into a stem structure
(20). The D3 region and ECH alone did not induce IB
degradation (Fig. 4A), NO
production (Fig. 4B), or IL-8 secretion (Fig. 4C)
in either of the IEC lines stimulated with 1 µg/ml of the fusion
proteins. No IB degradation (data not shown), NO production
(Fig. 4B), or IL-8 secretion (Fig. 4C) was
induced in IEC by stimulation with ND1/ECH/CD1, ND2/ECH/CD2,
ND1/D3/CD1, or ND2/D3/CD2.
|
A construct containing the D1 and D2 regions with the ECH retained flagellin inflammatory activities
Based on the results obtained from N/C flag, ND1/ECH/CD1, and
ND2/ECH/CD2 proteins, it was hypothesized that the bioactive domain was
contained in a region generated by a combination of the amino and
carboxyl D1 and D2 regions. A chimeric SD flagellin fusion
protein was produced that contained a combination of the ND1 and ND2
and CD1 and CD2 regions separated by the ECH. This fusion protein lacks
the entire D3 region and was designated ND1–2/ECH/CD2–1 (Fig. 1).
Similar to Caco-2BBe cells stimulated with 1 µg/ml 6HIS flag,
ND1–2/ECH/CD2–1 stimulation induced IB degradation at 20 and 30
min (Fig. 5A) and NF-B
activation at 1 h poststimulation (Fig. 5B).
IFN--primed DLD-1 cells stimulated for 18 h with
ND1–2/ECH/CD2–1 also showed significantly higher levels of NO
production vs untreated control cells (p <
0.001; Fig. 5C). This trend continued when substantial
levels of IL-8 were detected 22 h poststimulation with 1 µg/ml
ND1–2/ECH/CD2–1 comparable to purified native SD
flagellin, equal to 6HIS flag, and significantly higher than
unstimulated cells (p < 0.001; Fig. 5D).
|
secretion in a
human monocyte cell line
The bioactivity of the purified native SD flagellin,
6HIS flag, ND1–2/ECH/CD2–1, and each of the other recombinant
flagellin proteins also was established by stimulation of 5 x
106 THP-1 monocytes. Following 4 h of
stimulation, significantly higher levels of TNF- were detected in
THP-1 cells stimulated with purified native SD flagellin,
6HIS flag, ND1–2/ECH/CD2–1, and N/C flag vs unstimulated control
cells (p < 0.001; Fig. 6). N/C flag, however, stimulated
significantly less TNF- production in cells compared with
SD flagellin, 6HIS flag, and ND1–2/ECH/CD2–1
(p < 0.005; Fig. 6). None of the other
constructs induced TNF- production in these cells (Fig. 6), a
finding consistent with the IEC results.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
production in cultured human monocytes.
These findings give insight into the pathogenesis of
Salmonella and other flagellated bacterial infections of the
gut suggesting that the D1 and D2 regions of flagellin contribute to
the pathology associated with Salmonella infection that
leads to host tissue damage. The D1 and D2 regions of several Salmonella and other Gram-negative bacterial flagellin genes have been sequenced, compared, and reported to be highly conserved (16). In our own comparisons, performed by ClustalW protein sequence alignment, of the flagellin regions of SD (GenBank accession number Z15067), P. aeruginosa (M57501), and L. monocytogenes (X65624), we found that the ND1 regions contained 55% identical or conserved residues, 44% in the ND2 regions, 10% in the CD2 regions, 60% in the CD1 regions, and only 5% in the D3 regions. Because flagellin from different species of Gram-negative bacteria (7, 8, 9, 10) stimulate various inflammatory responses in a variety of host cells, we hypothesized that this common inflammatory activity likely would reside either in the conserved amino acid sequences or in conserved structural motifs. Consistent with this hypothesis, the chimeric construct containing the conserved D1 and D2 regions, designated ND1–2/ECH/CD2–1, retained bioactivity while the nonconserved D3 region did not. The D3 domain is on the surface of the flagellar filament and contains the major antigenic epitopes (17, 19, 23) as well as elements involved in protein folding (19, 23, 24). Because the D3 region is an immunological determinant, sequence changes are key for the bacteria to escape the immunological response of the host (17, 25). Our data with D3 region constructs showed no induction of proinflammatory responses, and, further proteins lacking the entire D3 region retained potent inflammatory properties. Collectively our data showed that the proinflammatory activity of flagellin resides in the conserved "stem" of the protein where the monomer anchors to the filament.
Our findings are contrary to the findings of an earlier publication by
McDermott et al. (10) who reported that the D3 region of
S. typhimurium flagellin, but not the amino and carboxyl
regions, was responsible for induction of TNF- in human monocytes.
We tested each of our recombinant flagellin proteins in the same human
monocyte cell line used by McDermott et al. (10)and found
TNF- production in response to the amino and carboxyl D1 and D2
regions (ND1–2/ECH/CD2–1) of SD flagellin. We also
generated and tested a S. typhimurium D3 fusion protein
according to a ClustalW multiple sequence alignment (aa 185–421) with
a structural map of flagellin (16) and showed no induction
of NO production in IECs compared with the 6HIS-tagged full-length
S. typhimurium flagellin fusion protein (data not shown). It
is unclear why there are discrepancies between these studies regarding
the bioactivity of the D3 region of flagellin. The lack of bioactivity
detected by McDermott et al. (10) in the conservative
regions of flagellin may be due to the inability of the fusion proteins
to fold correctly and form the domain responsible for bioactivity. Our
results with flagellin fusion proteins containing the ND1 and CD1 or
the ND2 and CD2, without the D3 region, lacked bioactivity consistent
with the report of McDermott et al. (10). These results
open the possibility that misfolding or the lack of protein folding
prevented the formation of the bioactive domain. To address this
possibility, we tested chimeric proteins that included a hinge element
for the E. coli MukB gene (20) that
had been used successfully by others to generate hairpin folding
similar to flagellin. As noted in Results, the
ND1–2/ECH/CD2–1 fusion protein showed a vigorous induction of a
variety of innate immune responses in human IEC and monocytes while the
ND1/ECH/CD1, ND2/ECH/CD2, and the ECH alone lacked activity. Currently,
we are further defining the proinflammatory region of SD
flagellin by testing additional constructs representing the D1/D2
borders.
Recently, a report providing exciting insight into the mechanism of
action of the proinflammatory properties of flagellin identified
Toll-like receptor 5 (TLR5) as a mammalian host cell receptor for
flagellin (26). The TLR are a family of cell surface
receptors that are identified by the conserved cytoplasmic signaling
domain. The activation of TLR5 by microbial ligands triggers an
intracellular signaling pathway that leads to the activation of NF-B
(26). In vitro studies reported by Aderem and coworkers
(26) found that flagellin from Gram-positive and negative
bacteria, including Salmonella, binds to the TLR5 activating
NF-B and stimulating TNF- production. Studies in our laboratory
currently are underway to assess the binding of the flagellin fusion
protein, ND1–2/ECH/CD2–1, to TLR5. Understanding this interaction
should lead to the development of novel anti-inflammatory and
antishock treatment strategies for patients with bacterial infections.
This work also will provide important information about TLR and TLR
ligands.
Understanding the region of flagellin that is responsible for inflammatory induction is necessary for the future design of effective therapeutics and vaccines. Recognizing potency of flagellin and its likely liberation during infection of the host, flagellin may be an important stimulus of in vivo inflammation. We (7) as well as others (8, 9, 10, 11) have shown that flagellin, in and of itself, is a potent inducer of inflammation; however, it is also possible that flagellin in combination with other bacterial components, such as LPS, may work synergistically to induce inflammation in vivo. We currently are in the process of investigating the role that purified flagellin, as well as our proinflammatory flagellin fusion proteins, play in inflammation and shock in vivo. These studies, like those published in our initial report on the proinflammatory activity of flagellin (7), will allow us to begin to address the question of the role of flagellin in the pathogenesis of Salmonella infections as well as other flagellated Gram-negative bacterial infections. These findings, together with our present results, will lead to a clearer understanding of host/bacterial interactions and pathogenic bacterial infections of the gastrointestinal tract.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Tonyia D. Eaves-Pyles, University of Texas Medical Branch, Children’s Hospital 2.300, 301 University Boulevard, Galveston, TX 77555-0366. E-mail address: tdeavesp{at}utmb.edu
3 Abbreviations used in this paper: SD, Salmonella dublin; IEC, intestinal epithelial cell; D1, domain 1; D2, domain 2; D3, middle hypervariable domain; ND, amino domain; CD, carboxyl domain; 6HIS, 6-histidine; ECH, Escherichia coli hinge; PB, polymyxin B; 6HIS flag, full-length recombinant 6HIS SD flagellin; TLR, Toll-like receptor.
Received for publication July 5, 2001. Accepted for publication October 9, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
through a post-transcriptional mechanism. J. Immunol. 166:6855.B degradation, induction of nitric oxide synthase, induction of proinflammatory mediators, and cardiovascular dysfunction. J. Immunol. 166:1248.B and IB proteins: new discoveries and insights. Annu. Rev. Immunol. 14:649.[Medline]
B in enterocytes. Infect. Immun. 67:800.This article has been cited by other articles:
|
|
 |
|
  C. Nempont, D. Cayet, M. Rumbo, C. Bompard, V. Villeret, and J.-C. Sirard Deletion of Flagellin's Hypervariable Region Abrogates Antibody-Mediated Neutralization and Systemic Activation of TLR5-Dependent Immunity J. Immunol., August 1, 2008; 181(3): 2036 - 2043. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. S. Yoon and J. J. Mekalanos Decreased Potency of the Vibrio cholerae Sheathed Flagellum To Trigger Host Innate Immunity Infect. Immun., March 1, 2008; 76(3): 1282 - 1288. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R.-M. Salazar-Gonzalez, A. Srinivasan, A. Griffin, G. Muralimohan, J. M. Ertelt, R. Ravindran, A. T. Vella, and S. J. McSorley Salmonella Flagellin Induces Bystander Activation of Splenic Dendritic Cells and Hinders Bacterial Replication In Vivo J. Immunol., November 1, 2007; 179(9): 6169 - 6175. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. S. Steiner How Flagellin and Toll-Like Receptor 5 Contribute to Enteric Infection Infect. Immun., February 1, 2007; 75(2): 545 - 552. [Full Text] [PDF]
|
|
|
|
|
|
L. Sfondrini, A. Rossini, D. Besusso, A. Merlo, E. Tagliabue, S. Menard, and A. Balsari Antitumor Activity of the TLR-5 Ligand Flagellin in Mouse Models of Cancer. J. Immunol., June 1, 2006; 176(11): 6624 - 6630. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. G. Torres, Y. Li, C. B. Tutt, L. Xin, T. Eaves-Pyles, and L. Soong Outer Membrane Protein A of Escherichia coli O157:H7 Stimulates Dendritic Cell Activation. Infect. Immun., May 1, 2006; 74(5): 2676 - 2685. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Sharma, S. Tesfay, F. L. Tomson, R. P. Kanteti, V. K. Viswanathan, and G. Hecht Balance of bacterial pro- and anti-inflammatory mediators dictates net effect of enteropathogenic Escherichia coli on intestinal epithelial cells Am J Physiol Gastrointest Liver Physiol, April 1, 2006; 290(4): G685 - G694. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Sun, F. M. Dunning, C. Pfund, R. Weingarten, and A. F. Bent Within-Species Flagellin Polymorphism in Xanthomonas campestris pv campestris and Its Impact on Elicitation of Arabidopsis FLAGELLIN SENSING2-Dependent Defenses PLANT CELL, March 1, 2006; 18(3): 764 - 779. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Verma, S. K. Arora, S. K. Kuravi, and R. Ramphal Roles of Specific Amino Acids in the N Terminus of Pseudomonas aeruginosa Flagellin and of Flagellin Glycosylation in the Innate Immune Response Infect. Immun., December 1, 2005; 73(12): 8237 - 8246. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. S. Lopez-Boado, L. M. Cobb, and R. Deora Bordetella bronchiseptica Flagellin Is a Proinflammatory Determinant for Airway Epithelial Cells Infect. Immun., November 1, 2005; 73(11): 7525 - 7534. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. N. Honko and S. B. Mizel Mucosal Administration of Flagellin Induces Innate Immunity in the Mouse Lung Infect. Immun., November 1, 2004; 72(11): 6676 - 6679. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Dons, E. Eriksson, Y. Jin, M. E. Rottenberg, K. Kristensson, C. N. Larsen, J. Bresciani, and J. E. Olsen Role of Flagellin and the Two-Component CheA/CheY System of Listeria monocytogenes in Host Cell Invasion and Virulence Infect. Immun., June 1, 2004; 72(6): 3237 - 3244. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. E. Sbrogio-Almeida, T. Mosca, L. M. Massis, I. A. Abrahamsohn, and L. C. S. Ferreira Host and Bacterial Factors Affecting Induction of Immune Responses to Flagellin Expressed by Attenuated Salmonella Vaccine Strains Infect. Immun., May 1, 2004; 72(5): 2546 - 2555. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Cuadros, F. J. Lopez-Hernandez, A. L. Dominguez, M. McClelland, and J. Lustgarten Flagellin Fusion Proteins as Adjuvants or Vaccines Induce Specific Immune Responses Infect. Immun., May 1, 2004; 72(5): 2810 - 2816. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Adamo, S. Sokol, G. Soong, M. I. Gomez, and A. Prince Pseudomonas aeruginosa Flagella Activate Airway Epithelial Cells through asialoGM1 and Toll-Like Receptor 2 as well as Toll-Like Receptor 5 Am. J. Respir. Cell Mol. Biol., May 1, 2004; 30(5): 627 - 634. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. G. K. Murthy, A. Deb, S. Goonesekera, C. Szabo, and A. L. Salzman Identification of Conserved Domains in Salmonella muenchen Flagellin That Are Essential for Its Ability to Activate TLR5 and to Induce an Inflammatory Response in Vitro J. Biol. Chem., February 13, 2004; 279(7): 5667 - 5675. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. S. Lopez-Boado, M. Espinola, S. Bahr, and A. Belaaouaj Neutrophil Serine Proteinases Cleave Bacterial Flagellin, Abrogating Its Host Response-Inducing Activity J. Immunol., January 1, 2004; 172(1): 509 - 515. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Donnelly and T. S. Steiner Two Nonadjacent Regions in Enteroaggregative Escherichia coli Flagellin Are Required for Activation of Toll-like Receptor 5 J. Biol. Chem., October 18, 2002; 277(43): 40456 - 40461. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. J. McSorley, B. D. Ehst, Y. Yu, and A. T. Gewirtz Bacterial Flagellin Is an Effective Adjuvant for CD4+ T Cells In Vivo J. Immunol., October 1, 2002; 169(7): 3914 - 3919. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. B. Mizel and J. A. Snipes Gram-negative Flagellin-induced Self-tolerance Is Associated with a Block in Interleukin-1 Receptor-associated Kinase Release from Toll-like Receptor 5 J. Biol. Chem., June 14, 2002; 277(25): 22414 - 22420. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
