| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
B Transcription Factors in Mononuclear Phagocytes1
*
Pulmonary and Critical Care Medicine Division and
Hematology-Oncology Division, Department of Internal Medicine, University of Michigan, Ann Arbor, MI 48109
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B transcription factors. U937 cells differentiated with PMA in
nonadherent culture were shown to express two fibrinogen-binding
integrins, predominately CD11b/CD18, and to a lesser extent,
CD11c/CD18. Cells stimulated with fibrinogen (10–100
µg/ml)/Mn2+ (50 µM) for 2 h were examined by
electrophoretic mobility shift assay. NF-B activation, minimal in
unstimulated cells, was substantially up-regulated by fibrinogen.
Fibrinogen also caused activation of AP-1, but not SP1 or cAMP response
element-binding protein (CREB) factors. Blocking mAbs against CD18 and
CD11b abrogated fibrinogen-induced NF-B activation. To determine the
effects on transcriptional regulation, U937 cells were transfected with
a plasmid containing the HIV-1 enhancer (bearing two NF-B sites)
coupled to a chloramphenicol acetyltransferase (CAT) reporter. Cells
were subsequently stimulated with 1) PMA for 24 h, inducing CAT
activity by 2.6-fold, 2) fibrinogen/Mn2+ for 2 h,
inducing CAT activity by 3.2-fold, or 3) costimulation with fibrinogen
and PMA, inducing 5.7-fold the CAT activity induced by PMA alone. We
conclude that contact with fibrinogen-derived proteins may contribute
to mononuclear phagocyte activation by signaling through CD11b/CD18,
resulting in selective activation of transcriptional regulatory
factors, including NF-B. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
in association
with activation of NF-B transcription regulatory factors (6).
Engagement of ß1 integrins can trigger activation
signaling (7), although the capacity for activation signaling directly
through ß2 integrins is less clearly defined (5). CR3
(Mac-1, Mß2, CD11b/CD18), a
ß2 integrin critically involved in adhesion, locomotion,
chemotaxis, and phagocytosis, serves as a receptor for a diverse group
of ligands, including iC3b, ICAM-1, bacterial LPS, ß glucan,
coagulation factor X, and fibrinogen (5, 8, 9). Fibrinogen is also a
ligand for a limited number of other adhesion proteins, including the
ß2 integrin CR4 (p150/95, CD11c/CD18),
vß3, and platelet GPIIb/IIIa (5, 10).
Interactions between leukocytes and fibrinogen/fibrin may have special
importance, since leukocytes commonly encounter provisional matrices
rich in fibrinogen and fibrin at sites of inflammation. There is
considerable evidence that the products of coagulation and fibrinolysis
profoundly affect many facets of inflammatory reactions, wound healing,
and fibrosis (11, 12, 13). Likewise, it is well established that contact
with fibrinogen or fibrin matrix can dramatically alter the expression
of proinflammatory factors such as TNF-, IL-1ß, reactive oxygen
intermediates, and possibly IL-8 in vitro (12, 14, 15, 16, 17). However, the
precise pathways by which fibrinogen and its derivatives modulate
leukocyte function are far less understood. One plausible mechanism by
which activation signaling through fibrinogen/fibrin could broadly
influence leukocyte function is by activation of pleiotropic
transcription regulatory proteins. NF-B is a particularly important
candidate, since its activation affects a broad array of
immediate-early gene products, including TNF-, ILs, chemokines, and
colony stimulating factors (reviewed in Refs. 18–21). NF-B
activation has been demonstrated by engagement of other adhesion
proteins such as ß1 integrins, but not by fibrinogen
binding or by Ab ligation of ß2 integrins (22). In this
study, we sought to determine whether fibrinogen could trigger
activation of NF-B and other transcription regulatory proteins in
mononuclear phagocytes. We demonstrate selective activation of NF-B
and AP-1 by fibrinogen binding in PMA-differentiated U937 cells.
Moreover, it is shown that fibrinogen-triggered NF-B activation is
sufficient to serve as a potent stimulus for NF-B-dependent gene
transcription. |
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Plasminogen-free fibrinogen (>95% purity) was obtained from
Calbiochem (La Jolla CA). LPS from Escherichia coli 0111:B4
was obtained from Sigma, St. Louis MO. Where appropriate, all reagents
used during cell stimulation were pretreated with End-X beads
(Associates of Cape Cod, Woods Hole, MA) to adsorb any contaminating
endotoxins. This pretreatment yielded reagents that contained less than
0.01 endotoxin U/ml, as determined by a chromogenic Limulus
amebocyte lysate assay (Associates of Cape Cod). The
anti-ß2 integrin mAbs included anti-CD18 mAb
(TS1/18; IgG1), anti-CD11b mAbs (clone 44 (IgG2a), and clone OKM1
(IgG2b), generously provided by Dr. Thomas Mercolino, Ortho Diagnostic
Systems, Raritan NJ), anti-CD11c mAbs (clone 3.9 (IgG1) Leinco
Technologies, Ballwin, MO, and clone 2E1 (IgG2b)). Isotype matched
control mAbs were obtained from Leinco, and ICN, Costa Mesa, CA. The
vß3 mAb (clone LM 609; IgG1) was obtained
from Chemicon International, Temecula, CA. Anti-p65, p50, and c-Rel Abs
were obtained from Santa Cruz Biotechnologies, Santa Cruz, CA. NF-B
oligonucleotides were prepared by the DNA Core Facility, University of
Michigan Medical Center.
Cell culture and stimulation
The U937 monocytic leukemia cell line was obtained from the American Type Culture Collection (ATCC, Manassas, VA). Cells were propagated in 75-cm2 polystyrene flasks in standard medium consisting of RPMI 1640 (Life Technologies, Grand Island, NY) with penicillin (100 U/ml), streptomycin (100 µg/ml), gentamicin (100 µg/ml), glutamine (2 mM), and 5% FBS (HyClone, Logan, UT). To induce morphologic and functional differentiation, the U937 cells were transferred to 1000-ml Teflon tissue culture bags (Baxter Healthcare, Deerfield, IL; 5 x 107 cells in 500 ml of medium) for stimulation with 0.5 nM PMA (Sigma) in nonadherent culture and incubated at 37°C, 5% CO2 for 48 h. The cells were then washed and resuspended in 20 mM HEPES/140 mM NaCl/2 mg/ml glucose/10 µg/ml polymyxin B, with additives as indicated, and returned to 15-ml Teflon tissue culture bags (American Fluoroseal, Columbia, MD) until the cells were harvested to prepare samples for EMSA3 or CAT assays.
Immunofluorescence flow cytometry
Cells were resuspended in staining buffer (PBS with 1% BSA, 0.1% sodium azide, pH 7.4) and incubated with the relevant primary mAb for 30 min, 4°C, followed by R-phycoerythrin-conjugated goat anti-mouse IgG (30 min, 4°C). Immunofluorescence flow cytometric analysis was performed with a Coulter electronics EPICSC flow cytometer with a logarithmic amplifier (Coulter, Miami, FL; Flow Cytometry Core Facility, University of Michigan Medical Center). The percent of positive cells was determined, using cells stained in parallel with an irrelevant isotype-matched primary Ab as background controls.
EMSA
Activation of NF-B transcription factors was measured by EMSA
(adapted from 23 . Nuclear proteins were extracted by the method
of Schreiber et al. (24) with minor modifications. After a preliminary
wash in Tris-buffered saline, the cells were pelleted and washed twice
in 10 mM HEPES, pH 7.9, with 10 mM KCl, 0.1 mM EDTA, 0.1 mM EGTA, 1 mM
DTT, and 0.5 mM PMSF, and incubated in a 700-µl volume at 0°C for
15 min. Five microliters of 1% Nonidet P-40 was then added, and the
nuclei were pelleted at 15,000 x g. The pellet was
resuspended in 75 µl of 20 mM HEPES, pH 7.9, with 0.4 M NaCl, 1 mM
EDTA, 1 mM EGTA, 1 mM DTT, and 1 mM PMSF, and incubated for 30 min at
4°C. The nuclei were pelleted once again at 15,000 x
g, 4°C, and the supernatant was stored at -70°C.
Nuclear extracts (5–10 µg protein in 10 µl) were incubated for 15
min, 25°C, with 10 µl reaction mixture (100 mM Tris Cl, pH 7.5, 1 M
NaCl, 50 mM DTT, 10 mM EDTA, 40% glycerol, 1 mg/ml BSA (nuclease
free), 50 ng/ml double-stranded poly(dI·dC)), and incubated
with 32P end-labeled NF-B consensus oligonucleotide
(5'-AGT TGA GGG GAC TTT CCC AGG-3') (6), or AP-1 consensus oligo
(5'-CGC TTG ATG AGT CAG CCG GAA-3'), SP1 consensus oligo, (5'-ATT CGA
TCG GGG CGG GGC GAG C-3'), or CREB consensus oligo (5'-AGA GAT TGC CTG
ACG TCA GAG AGC TAG-3') (Santa Cruz Biotechnologies). The reaction
mixture was run on a 4% PAGE gel in high ionic strength buffer (0.05 M
Tris, pH 8.5, 380 mM glycine, 2 mM EDTA), and developed with a
Molecular Dynamics PhosphorImager (Sunnyvale, CA). Binding of nuclear
protein(s) to the labeled oligonucleotide was demonstrated by
visualizing discrete bands of DNA-protein complexes with retarded
mobility in the PAGE gel. Binding of labeled oligonucleotides was shown
to be sequence-specific by demonstrating that binding was 1) blocked by
an excess of unlabeled oligonucleotide and 2) unaffected by an excess
of a control oligonucleotide of identical length and base composition,
but a randomized sequence. All the key findings shown were confirmed
with EMSA of the nuclear extracts prepared from at least two
independently run experiments.
Chloramphenicol acetyltransferase (CAT) assays
U937 cells were harvested during logarithmic growth and
transfected with a plasmid containing the HIV-1 enhancer (bearing two
NF-B sites) linked to a CAT reporter, which was generously provided
by Dr. Gary Nabel, Howard Hughes Medical Institute, University of
Michigan (25). Transfection was achieved by electroporation, using 15
µg of DNA for 107 cells at 300 V and 1000 µF. The cells
were returned to culture for 24 h, at which time they were
transferred to 15-ml Teflon culture bags and stimulated either with PMA
(0–5 nM) ± fibrinogen (50 µg/ml) for 24 h, or fibrinogen (50
µg/ml) with Mn2+ (50 µM) for 2 h, in serum-free
medium (Mac-SFM, Life Technologies). Control cells cultured in parallel
were mock transfected or transfected with the HIV-1-CAT construct and
cultured in medium supplemented with 5% FBS. As a further control,
cells were also transfected with a B mutant HIV-1-CAT plasmid and
stimulated with PMA ± fibrinogen as above. After culture for
24 h, cell lysates were adjusted for protein content and assayed
for CAT activity as described previously (26). Briefly, lysates were
incubated with 0.025 µCi of
D-threo-[dichloroacetyl-1-14C]
chloramphenicol (Amersham, Arlington Heights, IL) and 1.38 mM acetyl
coenzyme A at 37°C for 24 h. The chloramphenicol was separated
from its acetylated products by thin layer chromatography (TLC). The
TLC plates were then analyzed with a PhosphorImager, and the percentage
of acetylation was determined with Image Quant software (Molecular
Dynamics).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Preliminary experiments verified that there was weak NF-B
activation in undifferentiated U937 cells in response to LPS (not
shown). Therefore, U937 cells were induced to differentiate by
stimulation with a low concentration of PMA (0.5 nM) for 48 h.
Under these conditions, nuclear NF-B activity remained very low,
provided that the cells were maintained in nonadherent culture (Fig. 1
), whereas contact with any adhesive
surfaces such as polystyrene induced significant NF-B activation
(not shown). In addition, the PMA-differentiated cells became quite
responsive to LPS, demonstrating a strong NF-B response (Fig. 1).
|
B
PMA-differentiated U937 cells were stimulated with fibrinogen (50
µg/ml) in nonadherent culture for 2 h, at which time nuclear
proteins were extracted for EMSA assay. Where indicated, the medium was
supplemented with MnCl2 (50 µM) to ensure that
cation-dependent integrins were optimally activated (27). As shown in
Figure 1
A, fibrinogen induced substantial NF-B activation
in the presence of Mn2+, even in comparison with LPS
stimulation, which was included as a positive control. The effect of
fibrinogen was dose related, since NF-B activation was first evident
with as little as 10 µg/ml, with a substantial response occurring
with 25 µg/ml, a maximal response with 50 µg/ml, and a plateau
effect extending through 100 µg/ml (Fig. 1B). Time course
studies indicate that fibrinogen-induced NF-B activation was rapid
and transient, with a maximal response after 1 and 2 h of
stimulation, and a prompt return to baseline within 4 h. By
contrast, LPS-induced NF-B activation was somewhat slower to evolve,
with a maximal response developing within the 1- and 2-h time points.
Like the response to fibrinogen, NF-B activation then returned to
baseline within 4 h (not shown).
All these experiments were conducted in the presence of polymyxin B to
block the effects of endotoxin, even though the reagents or media
contained no more than 0.01 EU/ml. This corresponded to less than 1
ng/ml of the LPS used as a positive control, an amount that produces
minimal, if any, NF-B activation (not shown). It was also confirmed
that polymyxin B did not affect the NF-B activation produced by
fibrinogen but significantly reduced the magnitude of the response to
LPS (100 ng/ml) (Fig. 2). In addition,
LPS-induced NF-B activation was unaffected by the absence of
Mn2+ (not shown), while Mn2+ was necessary for
fibrinogen stimulation (Fig. 5; see below). Finally, fibrinogen
activated NF-B in undifferentiated U937 cells (Fig. 8; see below),
while, as noted above, LPS produced a weak NF-B response unless the
cells were pretreated with PMA. These distinctions confirm that LPS
contamination was not responsible for fibrinogen-induced NF-B
activation.
|
|
|
B proteins
Experiments were next performed to characterize the specific
NF-B proteins induced by fibrinogen. First, cold competition studies
were used to determine which protein-DNA complexes on EMSA are sequence
specific. For both LPS- and fibrinogen-stimulated cells, a single
complex was formed, with the labeled NF-B consensus oligo, that was
completely blocked by an excess of cold oligo but was unaffected by a
cold oligo with a randomized sequence (Fig. 3A).
|
B complex (Fig. 3B). The
anti-p50 Ab, which cross-reacts with p52, eliminated the lower
portion of the complex, and slightly attenuated the density of the
upper portion of the complex. The anti-p65 (Rel A) Ab, which
cross-reacts with c-Rel, eliminated the upper portion of this complex,
leaving only a minor band constituting the lower portion of the
complex. Finally, inclusion of both Abs virtually eliminated the entire
NF-B band. Thus, the demonstrated band of NF-B-specific binding
stimulated by fibrinogen is comprised of a mixture of NF-B forms.
The less prominent, lower portion of the band consists of dimers
containing p50 and/or p52, while the majority of the NF-B proteins,
representing the upper portion of the band, consists of dimers
containing Rel-A and/or c-Rel. Given that the anti-p50/p52 Ab
slightly reduced the density of the upper portion of the band, it is
likely that some heterodimers containing p50/p52 and Rel-A/c-Rel are
present here as well. In these respects, the NF-B binding proteins
extracted from fibrinogen-stimulated cells appeared identical in
composition to those extracted from LPS-treated cells. Recent studies
have suggested that a distinct class of NF-B sites, represented in
the IL-8 and tissue factor genes, do not bind p65/p50 heterodimers but
selectively bind c-Rel/p65 dimers (28), offering a potential mechanism
for selective gene regulation by NF-B. To determine whether
fibrinogen and LPS induced similar proportions of p65 vs c-Rel,
supershifting experiments were performed with a c-Rel-specific Ab that
does not cross-react with Rel-A (Santa Cruz No. sc-1827). The
anti-p50/p52 Ab was always included in these experiments so only
the p65/c-Rel reactive band would remain. The density of the p65/c-Rel
band was then examined with and without supershifting by the
anti-c-Rel-specific Ab. Because the c-Rel Ab only partially reduced
the density of the p65/c-Rel band, the results were analyzed by
densitometry, using ImageQuant software (Molecular Dynamics).
Supershifting with anti-c-Rel reduced the density of the p65/c-Rel
band by 28.7 ± 6.8% in extracts of fibrinogen-stimulated cells,
and by 27.5 ± 6.1% in extracts of LPS-stimulated cells,
indicating that fibrinogen and LPS induce activation of c-Rel- vs
Rel-A-containing dimers in similar proportions.
Selectivity of NF-B activation
To determine whether fibrinogen/Mn2+ broadly induces
activation of transcriptional regulatory factors, EMSA assays were also
performed using consensus oligonucleotides for AP-1, SP1, and CREB. As
shown in Figure 4,
fibrinogen/Mn2+ produced a substantial increase in AP-1
activation that was highly consistent but not as prominent as the
effect produced by LPS. However, fibrinogen/Mn2+ did not
affect binding of SP1 or CREB, even though activation of both factors
was evident in nuclear extracts of LPS-stimulated cells. These findings
indicate that signaling through fibrinogen/Mn2+ yields a
limited and selective array of activated transcriptional regulatory
factors. Nonetheless, the broad representation of NF-B and AP-1
regulatory elements in the enhancer regions of immediate-early gene
products and other proinflammatory mediators justifies the expectation
that fibrinogen-induced activation of both NF-B and AP-1 families
should have wide-ranging effects on expression of genes that are
tightly regulated during inflammation and wound healing.
|
B activation
One of the more notable properties of fibrinogen-mediated NF-B
activation was that the effect was seen only in
Mn2+-supplemented medium (Fig. 5). Fibrinogen (50 µg/ml) in
HEPES-buffered saline/glucose without Mn2+ had no
appreciable effect on NF-B activation. Likewise, the presence of
MnCl2 (50 µM) alone did not induce NF-B activation.
Preliminary experiments indicated that the 50 µM concentration of
MnCl2 was optimal for supporting NF-B activation (not
shown). This was quite similar to the Mn2+ concentration
that maximally activates LFA-1 in the same buffer system (29), in
keeping with the suggestion that the requirement for Mn2+
reflected the conditions that optimized ligation to fibrinogen-binding
integrins.
Immunofluorescence flow cytometry was performed to assess the
expression of integrins capable of serving as fibrinogen receptors,
namely CR3 (CD11b/CD18), CR4 (CD11c/CD18), and
vß3. GPIIb/IIIa (CD41b/CD61), which shares
epitope(s) and function with CR3, was not assessed, since it is
expressed by platelets and megakaryocytes and not by mononuclear
phagocytes (30). Using clone 44 anti-CD11b mAb, it was determined
that 62 ± 11% of PMA-stimulated cells expressed CR3 (mean
± SEM), while CR4 expression, although significant, was limited to
34 ± 7% of cells (using clone 3.9 anti-CD11c mAb). The
staining intensities for both CD11b and CD11c were broadly distributed,
so these percentages may be underestimates, excluding cells with
relatively low levels of integrin expression and including only those
with fluorescence intensities clearly exceeding control cells stained
with irrelevant primary Abs. There was negligible expression of
vß3 (1.5 ± 0.3% positive cells).
Finally, expression of the urokinase receptor (uPAR; CD87) was also
assessed, since uPAR associates with many integrins, including CR3 and
CR4, and this association strongly influences integrin-mediated
adhesion, including CR3-mediated adhesion of human monocytes to
fibrinogen (31, 32, 33). Detectable uPAR was found on 71 ± 10% of
cells, confirming their availability for interactions with integrins.
The presence of these two fibrinogen-binding ß2
integrins, coupled with a requirement for a divalent cation
(Mn2+), which is characteristic of ß2
integrins, indirectly implicated CR3 and/or CR4 as candidate receptors
responsible for fibrinogen-induced NF-B.
Confirmatory studies were next performed to determine which integrin(s)
might be responsible for fibrinogen-induced NF-B activation by
blocking the response selectively with a series of mAbs. In these
experiments, the cells were pretreated with mAb (100 µg/ml) for 30
min before adding fibrinogen/Mn2+. Monoclonal Abs were also
added with the fibrinogen/Mn2+ to maintain the
concentration at 100 µg/ml throughout the 2-h incubation before
nuclear protein extraction. As shown in Figure 6, the addition of either an
anti-CD18 mAb (clone TS1/18) or an anti-CD11b mAb (clone OKM1)
completely abrogated the NF-B response to
fibrinogen/Mn2+. Control mAbs had no effect on NF-B
activation. Likewise, the anti-CD11c mAb (clone 2E1) was
ineffective, suggesting that CR3, and not CR4, is the receptor
predominately responsible for fibrinogen-induced NF-B activation. It
is interesting to note that while OKM1 blocked NF-B activation
completely, clone 44, an anti-CD11b mAb that reacts with an epitope
located in the I domain, had no effect at all (not shown). This agrees
closely with prior studies indicating that OKM1 preferentially blocks
binding of soluble fibrinogen to CR3, while I-domain Abs such as clone
44 preferentially block cellular adhesion to immobilized
fibrinogen (8, 9, 34).
|
To determine whether stimulation with fibrinogen is sufficient to
modulate NF-B-driven gene transcription, U937 cells were transfected
with an HIV-1 enhancer-CAT construct, as described in Materials
and Methods. Prior studies have shown that transcription of this
construct is refractory to further modulation once the cells are
differentiated into a more mature phenotype (25). Accordingly, these
experiments were performed without pretreating the cells with PMA.
Preliminary experiments confirmed that control cells (transfected, but
otherwise unstimulated) expressed very low levels of CAT activity
minimally above the background activities of mock-transfected cells
(electroporated, no plasmid), indicating that transfection with the
NF-B/CAT plasmid alone resulted in little cellular activation (not
shown). Figure 7A shows that
stimulation with PMA alone for 24 h had only a limited effect on
CAT activity. Only the highest concentration of PMA, 5 nM, produced a
statistically significant increase in CAT activity (2.6-fold), relative
to unstimulated controls. By contrast, cells stimulated with fibrinogen
(50 µg/ml), in the presence of Mn2+ (50 µM) for only
2 h and then returned to serum-free medium for the remainder of a
24-h culture, yielded a statistically significant increase in CAT
expression, 3.2-fold relative to unstimulated controls (Fig. 7A). Experiments were also performed to determine whether
fibrinogen was an effective costimulus with PMA in inducing CAT
expression. In these experiments, cells were cultured with varying
concentrations of PMA ± fibrinogen for 24 h in serum-free
medium. Mn2+ was omitted from these experiments since
concurrent stimulation with PMA was deemed adequate to activate
fibrinogen-binding integrins (35). As shown in Figure 7B,
costimulation with fibrinogen significantly enhanced CAT activity for
all concentrations of PMA tested. The peak effect of fibrinogen was
seen with 0.5 nM PMA, producing 5.7-fold the CAT activity achieved by
stimulation with PMA alone. This protocol was repeated with cells
transfected with a HIV-1 enhancer-CAT construct with both NF-B sites
inactivated (Fig. 7B). Under these circumstances, fibrinogen
costimulation did not influence CAT expression at all, indicating that
the effect is clearly NF-B dependent. Finally, we wished to
determine whether other serum proteins could serve the same
costimulatory function as fibrinogen. Therefore, cells transfected with
the NF-B-intact HIV enhancer-CAT construct were stimulated with PMA
(0.5 nM) with and without 5% FBS. The addition of serum had no
costimulatory effect on CAT expression (Fig. 6B), indicating
that fibrinogen has a selective effect on NF-B-dependent gene
expression that is not shared broadly by other circulating proteins.
|
B proteins at 2 h parallel the
demonstrated effects on CAT expression at 24 h. Previously
untreated U937 cells were treated with 5.0 and 0.5 nM PMA, or with
fibrinogen (50 µg/ml)/Mn2+ (50 µM), for 2 h. PMA
alone produced only a modest degree of NF-B activation by EMSA
assay, while fibrinogen/Mn2+ produced a substantial level
of NF-B activation, even in undifferentiated U937 cells (Fig. 8). The combination of PMA (0.5 nM) and
fibrinogen/Mn2+ reproducibly yielded a level of NF-B
activation that was only minimally greater than
fibrinogen/Mn2+ alone. Therefore, PMA and fibrinogen do not
have additive effects on NF-B activation that can account for the
enhanced effect of fibrinogen on CAT expression. However, conditions
were chosen to maximize the short-term effects of PMA and
fibrinogen/Mn2+, so it remains possible that costimulation
with PMA and fibrinogen over 24 h affects NF-B activation in
ways that are not reflected in short term stimulation analyzed by EMSA.
It is also possible that the costimulatory effect of PMA modulates
other transcriptional regulatory factors or downstream steps in CAT
synthesis that compliment fibrinogen-induced NF-B activation. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B transcription regulatory factors in
undifferentiated and PMA-differentiated U937 mononuclear phagocytes.
The composition of NF-B-binding proteins closely resembled the
profile induced by activation with LPS, in that dimers containing p65
and c-REL were the dominant forms, with a minor contribution by dimers
composed of p50 and p52 (Fig. 3B). Under the conditions
employed, fibrinogen was able to produce this response only in the
presence of Mn2+. which suggested that an integrin adhesion
protein serves as the fibrinogen receptor. Many prior studies have
indicated that CR3 (Mac-1; CD11b/CD18), a ß2 integrin, is
the dominant fibrinogen-binding receptor on mononuclear phagocytes (8, 9, 36), while another ß2 integrin, CR4 (150/95;
CD11c/CD18), can also bind fibrinogen (10). Flow cytometry experiments
indicate that the PMA-differentiated U937 cells used in most
experiments in the present study expressed predominately CR3 (62% of
cells), although there was an appreciable representation of CR4 as well
(34% of cells). The use of blocking mAbs further verify that CR3 is
the dominant, if not sole, receptor responsible for fibrinogen-induced
NF-B activation. The NF-B response to fibrinogen/Mn2+
was completely negated by Abs to CD18 and CD11b. By contrast,
fibrinogen-induced NF-B activation was preserved in the presence of
anti-CD11c mAb. The absence of a blocking effect with a single Ab
is not itself convincing, but, together with the absence of any
residual NF-B activation in the presence of anti-CD11b mAb, it
is unlikely that CR4 contributed significantly to fibrinogen-induced
NF-B activation. Given the overlap in ligand specificities and the
structural similarities between CR3 and CR4, these functional
dissimilarities are striking and offer the opportunity to further
explore the distinct roles these integrins may play in
adhesion/activation coupling. Alternatively, it is possible that the
level of CR4 expression achieved under these experimental conditions
was simply insufficient to mediate NF-B activation.
The functional implications of this NF-B activation was demonstrated
by the markedly enhanced transcriptional activity of the HIV-1 LTR-CAT
construct in response to fibrinogen, both as a lone stimulus (3.2-fold
the activity of unstimulated controls; Fig. 7) and as a potent
costimulus with PMA, where the resulting CAT activities were enhanced
by fibrinogen at all PMA concentrations tested, optimally augmenting
CAT expression by 5.7-fold. None of these effects were observed when
the cells were transfected with a construct lacking functional NF-B
sites. While it is certainly possible that both PMA and fibrinogen may
have effects other than NF-B activation that enhance expression of
the HIV-1 LTR-CAT construct, it is clear that the effect is entirely
dependent on the functionality of NF-B.
There are a few features of the model used in the EMSA experiments
(PMA-pretreatment of U937 cells for 48 h) that merit specific
comment. Preliminary experiments demonstrated that untreated U937 cells
exhibited poor NF-B activation over 2 h in response an
archetypal agonist such as LPS. Prior work has shown that pretreating
U937 cells with PMA for 24 h induces sustained and constitutive
NF-B (25). The key methodologic difference in our study is that the
U937 cells were maintained in strictly nonadherent culture throughout
this incubation. Preliminary experiments showed that PMA-pretreated
U937 cells were exquisitely sensitive to even transient exposure to any
adhesive surface (including protein-coated or uncoated plastic),
resulting in near-maximal NF-B activation. This indicates that the
emergence of constitutive NF-B expression during PMA-induced
differentiation is itself critically dependent on cellular adhesion.
The very low baseline expression of nuclear NF-B proteins, combined
with a differentiated phenotype (with relatively high levels of
integrin expression) and the preserved responsiveness to subsequent
stimulation (fibrinogen/Mn2+ or LPS), make this model a
highly attractive one for studying the effects of integrins on gene
expression in mononuclear phagocytes. Since adhesion to
fibrinogen-coated flasks produced the same high levels of NF-B
activation as contact with uncoated or albumin-coated plastic (not
shown), the model is limited in that it is applicable only to studying
soluble counterligands and precluded us from examining the effects of
immobilized fibrinogen. Another particularly interesting feature of
this model is that fibrinogen induces NF-B activation even while the
cells are kept meticulously in nonadherent conditions, in striking
contrast to prior studies suggesting that integrin-mediated activation
signaling requires integrin cross-linking, cellular adhesion, and
cytoskeletal rearrangement (5). We cannot dismiss the possibility that
intercellular contact, which can be enhanced by fibrinogen, also
contributed to NF-B activation (36). However, the cells were
routinely examined after fibrinogen treatment, and the cells emerged
from these cultures in uniform suspensions. Homotypic aggregation was
not observed (not shown).
The concept that leukocyte integrins can serve as conduits for
activation signaling first emerged from observations that these
proteins can sample the extracellular environment for extracellular
matrix (ECM) proteins, adjacent cells, LPS, and microbial
pathogens, and, through transmembrane domains, can provide a physical
link to the cytoskeleton and thereby to the machinery of gene
expression (4, 5). Subsequent studies have reinforced this concept by
demonstrating that integrin ligation can engage various pathways of
intracellular signal transduction. Many integrins can activate the
Na+/H+ antiporter with a resultant increase in
intracellular pH (5). Most integrins tested, including CR3, can also
trigger an increase in the concentration of intracellular calcium (5).
Ab binding of ß1 integrins has been shown to trigger
immediate-early gene expression and NF-B activation (7, 22), but a
similar approach to signaling through ß2 integrins was
unsuccessful (7). Thus far, evidence for activation signaling through
ß2 integrins has been indirect or limited to changes in
early signaling mediators such as intracellular calcium. Contact with
fibrinogen-related proteins can induce or enhance expression of
immediate-early genes in mononuclear phagocytes, including TNF-,
IL-1ß, and tissue factor (12, 14, 15). All these proteins are subject
to regulation by NF-B, but NF-B activation was not demonstrated
directly in any of these studies. However, recent work has shown that
CR3-transfected fibroblasts are capable of NF-B activation in
response to LPS (37).
Local activation of coagulation and fibrinolytic pathways, virtually
ubiquitous at inflammatory foci, is deeply insinuated in the
pathogenesis of inflammatory tissue injury and repair. Fibrinogen and
its derivative products are involved in the adhesion, spreading, and
movement of many cell types involved in wound repair, including
fibroblasts, endothelial, cells, and epithelial cells (38, 39, 40). It is
also expected that leukocytes will encounter fibrinogen-related
proteins at virtually every phase of the inflammatory response.
Circulating monocytes would certainly be exposed to high concentrations
of fibrinogen in plasma. Monocytes do express some nuclear NF-B
constitutively (41), which is clearly not maximal, and monocytes retain
responsiveness to fibrinogen in vitro (12, 14, 15, 16). Consistent with the
results of the present study (Fig. 5), it is possible that monocyte CR3
is relatively incapable of mediating NF-B activation until activated
by exposure to chemoattractants, early adhesion, or other signals.
These and other factors that may modulate monocyte responsiveness to
plasma fibrinogen certainly merit future study. Fibrin and platelet
deposits accumulate on injured endothelium, particularly under shear
stress, providing an adhesive surface for activated leukocytes (42).
Fibrinogen can significantly enhance adhesive reactions between
monocytes and endothelial cells by acting as a bridge molecule between
the CR3 of a mononuclear phagocyte and endothelial ICAM-1 (36).
Extravascular fibrinogen is rapidly converted into fibrin deposits that
persist as prominent features of many acute and chronic inflammatory
reactions. Fibrinogen may gain access to the extravascular compartment
either by exudation of circulating fibrinogen or by frank hemorrhage,
which has been shown to induce NF-B activation in vivo (43).
Fibrinogen can also be synthesized locally by epithelial cells (44).
Leukocytes contacting fibrin-rich matrices may be directed to alter
production of proinflammatory cytokines (12, 15), and, in addition,
fibrin degradation products can amplify the inflammatory reaction by
serving as chemotaxins (45, 46, 47). In the present studies, all
experiments were performed with soluble fibrinogen, so it will be
necessary to determine whether its derivative forms expected at sites
of inflammation, including fibrin monomer, insoluble fibrin polymer,
and fibrin degradation products, are equally capable of activating
NF-B and AP-1. The fibrinous stroma also influences the architecture
of inflamed tissue by limiting hemorrhage, entrapping leukocytes, and
providing a provisional matrix for the ingrowth of fibroblasts, and
evidence is mounting that impaired local fibrinolysis favors the
development of excessive fibrosis rather than restoration of normal
tissue structure (13). The results of the present study add to the
potential mechanisms by which local fibrin deposition can alter the
course of inflammation and healing. The broad array of genes that are
regulated by NF-B factors include a host of cytokines that
participate in leukocyte recruitment and activation, including
ILs (IL-1ß, IL-6, IL-12), TNF-, chemokines (IL-8, growth-related
oncogene protein (Gro)-, -ß, and -
, macrophage inflammatory
protein (MIP)-1, monocyte chemotactic protein (MCP)-1), inducible
nitric oxide synthase (iNOS), adhesion proteins (ICAM-1, VCAM-1), and
viruses (HIV, cytomegalovirus (CMV), and adenovirus) (18, 21, 48).
Thus, one can easily hypothesize that the same fibrinogen-integrin
interaction that facilitates immobilization and recruitment of
leukocytes can, by triggering NF-B activation, augment the
production of cytokines, regulate expression of adhesion proteins, and,
possibly, contribute to the persistence and proliferation of viruses.
Indeed, prior studies have shown that HIV-infected leukocytes exhibit
enhanced expression of CD11b, increased adhesion, and enhanced
replication of HIV when cells adhere to specific components of the
extracellular matrix (49, 50, 51). It is well recognized that HIV
replication can be enhanced by the effects of opportunistic infections,
and local deposition of fibrinogen-related proteins at these sites of
infection provide yet another potentially important stimulus by which
this may occur. Lastly, NF-B factors have been implicated in both
the induction and prevention of apoptosis, suggesting that fibrinous
stroma may influence the longevity of leukocytes and parenchymal cells
in inflamed or healing tissues (52).
In summary, this study demonstrates that U937 mononuclear phagocytes
react to exposure to fibrinogen in vitro with prompt activation of
pleiotropic transcription regulatory factors, NF-B and AP-1. The
NF-B response is mediated by CR3 but does not require cellular
adhesion. Finally, fibrinogen, either alone or in concert with PMA, is
capable of supporting substantially enhanced transcriptional activity
of a NF-B-regulated gene. Fibrinogen-related proteins are already
believed to be important participants in leukocyte trafficking and
tissue remodeling and now can also be implicated in the mechanisms
whereby altered coagulation/fibrinolysis homeostasis in inflamed tissue
influences activation signaling of mononuclear phagocytes.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Robert G. Sitrin, 6301 MSRB III, Box 0642, 1150 West Medical Center Drive, University of Michigan Medical Center, Ann Arbor, MI 48109. E-mail address:
3 Abbreviations used in this paper: EMSA, electrophoretic mobility shift assay; CAT, chloramphenicol acetyltransferase; CREB, cAMP response element-binding protein; LTR, long terminal repeat; Fbg, fibrinogen; wt, wild-type.
Received for publication December 11, 1997. Accepted for publication March 25, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
secretion. J. Clin. Invest. 95:2297.
chain of fibrinogen. Proc. Natl. Acad. Sci. USA 88:1044. responses of monocytes. J. Immunol. 150:2972.[Abstract]
B in cytokine gene regulation. Am. J. Respir. Cell Mol. Biol. 17:3.B and IB proteins: new discoveries and insights. Annu. Rev. Immunol. 14:649.[Medline]
B and Rel: participants in a multiform transcriptional regulatory system. Int. Rev. Cytol. 143:1.[Medline]
B. Nature 339:70.[Medline]
B subunit which stimulates HIV transcription in synergy with p65. Nature 352:733.[Medline]
B-like site. Mol. Cell. Biol. 14:3772.IIbß3 (glycoprotein IIbIIIa; CD41b/CD61) and mß2 (Mac-1; CD11b/CD18) detected by a monoclonal antibody. J. Immunol. 156:284.[Abstract]
B in cells of the monocyte lineage. Biochem. J. 304:87.
B in murine intraparenchymal lung mononuclear cells. Am. J. Respir. Cell Mol. Biol. 16:145.[Abstract]
B subunits in endothelial and epithelial cells. Mol. Cell. Biol. 13:6283.B: ten years after. Cell 87:13.[Medline]
This article has been cited by other articles:
|
|
 |
|
  R. Ratra, A. Kar-Roy, and S. K. Lal ORF3 protein of hepatitis E virus interacts with the B{beta} chain of fibrinogen resulting in decreased fibrinogen secretion from HuH-7 cells J. Gen. Virol., June 1, 2009; 90(6): 1359 - 1370. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Alfano, S. A. Mariani, C. Elia, R. Pardi, F. Blasi, and G. Poli Ligand-engaged urokinase-type plasminogen activator receptor and activation of the CD11b/CD18 integrin inhibit late events of HIV expression in monocytic cells Blood, February 19, 2009; 113(8): 1699 - 1709. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Vidal, A. L. Serrano, M. Tjwa, M. Suelves, E. Ardite, R. De Mori, B. Baeza-Raja, M. Martinez de Lagran, P. Lafuste, V. Ruiz-Bonilla, et al. Fibrinogen drives dystrophic muscle fibrosis via a TGF{beta}/alternative macrophage activation pathway Genes & Dev., July 1, 2008; 22(13): 1747 - 1752. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. P. de Visser, F. J. Walther, E. H. Laghmani, S. van Wijngaarden, K. Nieuwland, and G. T. M. Wagenaar Phosphodiesterase-4 inhibition attenuates pulmonary inflammation in neonatal lung injury Eur. Respir. J., March 1, 2008; 31(3): 633 - 644. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Mu, D. Qu, A. Bartczak, M. J. Phillips, J. Manuel, W. He, C. Koscik, M. Mendicino, L. Zhang, D. A. Clark, et al. Fgl2 deficiency causes neonatal death and cardiac dysfunction during embryonic and postnatal development in mice Physiol Genomics, September 11, 2007; 31(1): 53 - 62. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. A. Lefevre, W. R. Hein, Z. Stamataki, L. S. Brackenbury, E. A. Supple, L. G. Hunt, P. Monaghan, G. Borhis, Y. Richard, and B. Charleston Fibrinogen is localized on dark zone follicular dendritic cells in vivo and enhances the proliferation and survival of a centroblastic cell line in vitro J. Leukoc. Biol., September 1, 2007; 82(3): 666 - 677. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. A. J. ter Horst, F. J. Walther, B. J. H. M. Poorthuis, P. S. Hiemstra, and G. T. M. Wagenaar Inhaled nitric oxide attenuates pulmonary inflammation and fibrin deposition and prolongs survival in neonatal hyperoxic lung injury Am J Physiol Lung Cell Mol Physiol, July 1, 2007; 293(1): L35 - L44. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. W.H. Woo, Y. L. Siow, and K. O Homocysteine Activates cAMP-response Element Binding Protein in HepG2 Through cAMP/PKA Signaling Pathway Arterioscler Thromb Vasc Biol, May 1, 2006; 26(5): 1043 - 1050. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Mendicino, M. Liu, A. Ghanekar, W. He, C. Koscik, I. Shalev, M. Javadi, J. Turnbull, W. Chen, L. Fung, et al. Targeted Deletion of Fgl-2/Fibroleukin in the Donor Modulates Immunologic Response and Acute Vascular Rejection in Cardiac Xenografts Circulation, July 12, 2005; 112(2): 248 - 256. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. W. H. Woo, Y. L. Siow, G. N. Pierce, P. C. Choy, G. Y. Minuk, D. Mymin, and K. O Hyperhomocysteinemia induces hepatic cholesterol biosynthesis and lipid accumulation via activation of transcription factors Am J Physiol Endocrinol Metab, May 1, 2005; 288(5): E1002 - E1010. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Flick, X. Du, and J. L. Degen Fibrin(ogen)-{alpha}M{beta}2 Interactions Regulate Leukocyte Function and Innate Immunity In Vivo Experimental Biology and Medicine, December 1, 2004; 229(11): 1105 - 1110. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. A. J. ter Horst, G. T. M. Wagenaar, E. de Boer, M. A. van Gastelen, J. C. M. Meijers, B. J. Biemond, B. J. H. M. Poorthuis, and F. J. Walther Pentoxifylline reduces fibrin deposition and prolongs survival in neonatal hyperoxic lung injury J Appl Physiol, November 1, 2004; 97(5): 2014 - 2019. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. A. Armstrong, J. A. Major, A. Chudyk, and T. A. Hamilton Neutrophil chemoattractant genes KC and MIP-2 are expressed in different cell populations at sites of surgical injury J. Leukoc. Biol., April 1, 2004; 75(4): 641 - 648. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Kettritz, M. Choi, S. Rolle, M. Wellner, and F. C. Luft Integrins and Cytokines Activate Nuclear Transcription Factor-{kappa}B in Human Neutrophils J. Biol. Chem., January 23, 2004; 279(4): 2657 - 2665. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Alfano, N. Sidenius, F. Blasi, and G. Poli The role of urokinase-type plasminogen activator (uPA)/uPA receptor in HIV-1 infection J. Leukoc. Biol., November 1, 2003; 74(5): 750 - 756. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. W. Y. Chan, L. S. Kay, R. G. Khadaroo, M. W. C. Chan, S. Lakatoo, K. J. Young, L. Zhang, R. M. Gorczynski, M. Cattral, O. Rotstein, et al. Soluble Fibrinogen-Like Protein 2/Fibroleukin Exhibits Immunosuppressive Properties: Suppressing T Cell Proliferation and Inhibiting Maturation of Bone Marrow-Derived Dendritic Cells J. Immunol., April 15, 2003; 170(8): 4036 - 4044. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Abraham, S. Shapiro, N. Lahat, and A. Miller The role of IL-18 and IL-12 in the modulation of matrix metalloproteinases and their tissue inhibitors in monocytic cells Int. Immunol., December 1, 2002; 14(12): 1449 - 1457. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. T. Smiley, J. A. King, and W. W. Hancock Fibrinogen Stimulates Macrophage Chemokine Secretion Through Toll-Like Receptor 4 J. Immunol., September 1, 2001; 167(5): 2887 - 2894. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. B. Forsyth, D. A. Solovjov, T. P. Ugarova, and E. F. Plow Integrin {alpha}M{beta}2-Mediated Cell Migration to Fibrinogen and Its Recognition Peptides J. Exp. Med., May 21, 2001; 193(10): 1123 - 1134. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Rezzonico, V. Imbert, R. Chicheportiche, and J.-M. Dayer Ligation of CD11b and CD11c {beta}2 integrins by antibodies or soluble CD23 induces macrophage inflammatory protein 1{alpha} (MIP-1{alpha}) and MIP-1{beta} production in primary human monocytes through a pathway dependent on nuclear factor-{kappa}B Blood, May 15, 2001; 97(10): 2932 - 2940. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Rubel, G. C. Fernandez, G. Dran, M. B. Bompadre, M. A. Isturiz, and M. S. Palermo Fibrinogen Promotes Neutrophil Activation and Delays Apoptosis J. Immunol., February 1, 2001; 166(3): 2002 - 2010. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Shi, X. Zhang, Z. Chen, M. K. Robinson, and D. I. Simon Leukocyte Integrin Mac-1 Recruits Toll/Interleukin-1 Receptor Superfamily Signaling Intermediates to Modulate NF-{kappa}B Activity Circ. Res., November 9, 2001; 89(10): 859 - 865. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
100% of
control. * p < 0.02, n = 4.
B, left, Effects of stimulating with
fibrinogen (50 µg/ml)/Mn2+ (50 µM) concurrently with
PMA (0.05–5.0 nM) for 24 h. Cells were transfected with the
wild-type (wt) HIV-1 LTR/CAT construct. CAT activities of cell lysates
are normalized to controls stimulated with PMA alone (without
fibrinogen/Mn2+). Costimulation with
fibrinogen/Mn2+ significantly augmented CAT activities at
all PMA concentrations. n = 7.
Middle, Costimulation with fibrinogen (50
µg/ml)/Mn2+ (50 µM) had no effect on PMA-driven CAT
activity when cells were transfected with a HIV-1 LTR/CAT construct
bearing inactivated NF-