| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,
*
Division of Pulmonary and Critical Care Medicine, and
Division of Rheumatology, Department of Medicine, New York University Medical Center, New York, NY 10016; and
Department of Rheumatology, Hospital for Joint Diseases, New York, NY 10003
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, IL-1ß, and PMA induced the release of GM-CSF
in HBECs. The robust response to PMA was not detected in SV40
adenovirus-transformed normal human bronchial epithelial cells
(BEAS-2B). PMA and TNF- stimulation of GM-CSF required activation of
PKC (inhibition by staurosporine and bisindolylmaleimide I). GM-CSF
expression was up-regulated by a nonphorbol PKC activator, but not by
an inactive PMA analogue. PMA-induced GM-CSF production in HBECs did
not require a Ca2+ ionophore and was not inhibited by
cyclosporin A. Activation of MAPKerk1/2 via PKC was
associated with and was required for GM-CSF production induced by PMA
and TNF-. The data demonstrate regulation of GM-CSF in HBECs by PKC
pathways converging on the MAPKerk1/2 pathway and further
define cell-specific regulation critical for local airway
responses. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Elevated levels of GM-CSF have been well described in bronchoalveolar lavage, endobronchial biopsy, and sputum samples from asthmatics (1, 7, 8, 9, 10, 11, 12). Elevated levels of GM-CSF, largely derived from epithelial cells, have been demonstrated to increase eosinophil activation and survival in asthmatics (13, 14, 15). GM-CSF stimulates the recruitment and activation of eosinophils via ß integrin-mediated adhesion to epithelial and endothelial cells (16) and prolongs eosinophil survival via inhibition of apoptosis (5, 17, 18). Moreover, GM-CSF stimulates activation of the 5-lipoxygenase pathway (19). In addition, murine models of asthma and diesel-induced hyper-responsiveness have demonstrated an association between epithelial cell-derived GM-CSF and airway hyper-responsiveness (20, 21).
GM-CSF may also participate in immune mechanisms of asthma via an effect on dendritic cells (DC), which are abundantly distributed in the submucosa and intraepithelial regions of the airway (22, 23, 24). DC along with other local but less potent APC are required for the initiation and perpetuation of the T cell activation involved in asthma. GM-CSF is a critical factor for the maturation of DC and enhances the expression of accessory molecules such as ICAM-1, B7-1 (CD80), and B7-2 (CD86) (22). In the human lung, elevated epithelial levels of GM-CSF have been demonstrated in association with the accumulation of DC (25). Murine models of asthma with epithelial cell transgene expression of GM-CSF describe DC cell recruitment in the airway as well as that of inflammatory cells (20, 26). In murine models, DC have been demonstrated to be essential for presenting inhaled Ag to previously primed Th2 cells, and thus for chronic eosinophilic airway inflammation (20, 27).
Despite the potential critical importance of epithelial cell-derived
GM-CSF, little is known about the regulation of GM-CSF production by
these cells. In general, regulation of expression of GM-CSF is cell and
stimulus specific and can be controlled at multiple levels. Most
studies of GM-CSF regulation have been performed in lymphocytes or
transformed, transfected cells overexpressing constitutively active or
dominant negative signaling components, the regulation of which may
differ from that of primary cells (28, 29). AEC release
GM-CSF in response to physiologic stimuli relevant for asthma,
including dust mite proteolytic allergens, human rhinovirus-14,
respiratory syncytial virus, and histamine (30, 31, 32). AEC
have also been demonstrated to secrete GM-CSF in response to cytokines
such as TNF-, IL-1, IL-4, and IL-13 (2, 3, 4, 33). These
disparate stimuli act via a multitude of signaling pathways and
therefore fail to provide clear clues to deciphering pathways that may
lead to optimal production of GM-CSF.
In view of the easy access of AEC to relevant stimuli and the importance of locally released GM-CSF in airway immune and inflammatory homeostasis, it is critical to understand the regulation of GM-CSF in these cells. We investigated whether regulation of GM-CSF production in primary culture human bronchial epithelial cells (HBECs) involved activation of PKC and subsequent activation of the MAPKerk1/2 pathways.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Basal cell culture medium was obtained from Clonetics (San
Diego, CA) and for routine cell growth was supplemented with the
following compounds: human epidermal growth factor (0.5 ng/ml),
hydrocortisone (0.5 µg/ml), insulin (5 µg/ml), transferrin (10
µg/ml), epinephrine (0.5 µg/ml), tri-iodothyronine (6.5 ng/ml),
gentamicin 50 µg/ml), amphotericin B (50 ng/ml), bovine pituitary
extract (13 µg/ml), and retinoic acid (0.1 ng/ml) (34).
LHC-9 was obtained from Biofluids (Rockville, MD) and was similarly
supplemented. PMA and A23187 were obtained from Sigma (St. Louis, MO).
Human recombinant TNF- and IL-1 were obtained from R&D (Minneapolis,
MN). TRIzol reagent was obtained from Life Technologies (Gaithersburg,
MD). Quik Hyb hybridization solution and Random Primer Labeling Kit
Prime-It II were obtained from Stratagene (La Jolla, CA). The 751-bp
hGM-CSF cDNA probe was provided by Dr. Steven Clark (Cambridge, MA). A
cDNA probe to a fragment GAPDH was generated by RT-PCR. All Abs
directed against phosphorylated and nonphosphorylated forms of
MAPKerk, MAPKjnk, and
MAPKp38 were obtained from Santa Cruz
Biotechnology (Santa Cruz, CA). Cyclosporin A, mezerein, 4-phorbol
12,13-didecanoate, bisindolylmaleimide I, staurosporine, PD98059, and
SB203580 were purchased from Calbiochem (La Jolla, CA) and were
dissolved in DMSO (Me2SO). The final
concentration of Me2SO did not exceed 0.1%
(v/v). ECL Plus kits were obtained from Amersham Pharmacia (Aylesbury,
U.K.).
Cell culture
Culture of normal HBECs from bronchial brush biopsies was performed using a modification of the methods of Wu and Turner (35), Kelson et al. (36), and Nakamura et al. (33). Briefly, cells were obtained during bronchoscopy of normal human subjects (New York University review board-approved protocol). Bronchial brushing was performed with a disposable brush (model BC-15C, Olympus, New Hyde Park, NY), which was introduced via the bronchoscope into subsegmental bronchi and rubbed against the epithelial surface with 10 strokes. The cells obtained by brushing were collected into serum-free, hormonally supplemented medium (Clonetics) containing amphotericin and gentamicin. The cells were plated in uncoated T25 tissue culture flasks and incubated (37°C, 5% CO2) for 7–10 days, during which time the cells were fed every 2 days. When cells reached 70% confluence, they were passaged into appropriate tissue culture plates required for specific experiments. All experiments were performed at passage 3, as additional passaging led to increased constitutive release of GM-CSF. Hydrocortisone, retinoic acid, and epinephrine (known to suppress GM-CSF production) were removed from the medium 24 h before each experiment. Epithelial cell phenotype was confirmed by appropriate staining with anti-human cytokeratin Abs (positive staining for CK7 and CK18, absence of CK20). For some experiments, HBECs were purchased from Clonetics, cultured in the same manner, and used at passage 3.
SV40 adenovirus-transformed normal human bronchial epithelial cells (BEAS-2B) were obtained from American Type Culture Collection (Manassas, VA), grown in LHC-9, and used between passages 14–25.
ELISA
Cells were grown to near confluence at passage 3 and stimulated with the specified agents (18 h, 37°C). Supernatants were subsequently collected, centrifuged (1000 rpm, 10 min), and diluted appropriately, and the concentration of GM-CSF was determined by ELISA (Endogen, Cambridge, MA). Measurements were performed in duplicate and were quantified at 450 nm (microplate reader, Bio-Rad, Richmond, CA).
RNA isolation and Northern analysis
Cells were grown to near confluence at passage 3 and were
stimulated with the appropriate agent (4 h, 37°C). Total RNA was
extracted with 4 M guanidinium-HCl and isolated by CsCl centrifugation
(5.7 M CsCl and 0.01 M EDTA, pH 7.5) or alternatively by TRIzol
reagent. Total RNA (20 µg/lane) was electrophoresed overnight on a
2% agarose-formaldehyde gel (22 V/cm). RNA was transferred onto Nytran
filters (Schleicher & Schuell, Keene, NH) by capillary action, and
filters were cross-linked (Stratagene cross-linker). Filters were
prehybridized (Quik Hyb, 10 min, 65°C), followed by hybridization
with a [-32P]dCTP random primer-labeled cDNA
probe for GM-CSF or GAPDH. Filters were subjected to three stringent
washes, placed between intensifying screens, and exposed to x-ray film
(Fuji, Tokyo, Japan; -70°C). Densitometry was performed with an
Ultroscan XL densitometer (LKB, Bromma, Sweden).
Immunoblotting with phosphospecific Ab probes
Activated MAPK species were detected using phosphospecific Abs directed against the dually phosphorylated forms of the protein. Cells were incubated in basal medium (4 h) before stimulation with defined agents for the times indicated in the figure legends. Lysates were prepared by treating cells with lysis buffer (20 mM Tris-HCl (pH 7.4), 150 mM NaCl, 0.5% Triton X-100, 1% sodium deoxycholate, 0.5 M PMSF, 2 mM Na3VO4, 50 mM NaF, 1 mM EGTA, 50 µg/ml aprotinin, 50 µg/ml chymostatin, and 25 µg/ml pepstatin). Lysates were centrifuged (13,000 rpm, 30 min) to sediment the particulate material. The protein concentration of the supernatant was measured by the BCA protein assay method (Pierce, Rockville, IL). Equal amounts of protein (50 µg/lane) were electrophoresed in 10% SDS-Tris glycine gels, and resolved proteins were transferred to polyvinylidene difluoride (PVDF) membranes. The membranes were blocked (0.5% nonfat dried milk) and probed with anti-phospho-MAPKerk1/2 (1/200) or anti-MAPKerk2 (1/200) followed by incubation with the appropriate HRP-conjugated secondary Ab (1/1000). Bound Abs were visualized using the ECL Western blot detection system according to the manufacturer’s instructions. Equal loading of samples was checked by using parallel blots or stripped blots for immunodetection of MAPKerk2 with phosphorylation state-independent pan-Abs, and reprobing with anti-MAPKerk2.
In vitro kinase assay for MAPKerk1/2
HBECS were lysed (20 mM Tris-HCl (pH 7.4), 1% Triton X-100, 10
mM ß-glycerophosphate, 0.5 mM PMSF, 2 mM
Na3VO4 20 mM NaF, 1 mM
EGTA, 1 mM DTT, 50 µg/ml aprotinin, 50 µg/ml leupeptin, and 25
µg/ml pepstatin). Cell lysates were precleared with protein
A-Sepharose beads, and immunoprecipitations were performed with
anti-MAPKerk1 and
MAPKerk2 together in the presence of protein
A-Sepharose beads. Beads were incubated (30 min) with 20 µCi of
[
-32P]ATP and myelin basic protein (500
µM) as peptide substrate in buffer (20 mM HEPES (pH 7.6), 200 mM
MgCl2, 20 µM ATP, 2 nM DTT, 100 µM sodium
orthovanadate, and 25 µM ß-glycerophosphate). Reactions were
stopped by the addition of 15% formic acid. The beads were
centrifuged, and supernatants were spotted onto phosphocellulose
papers. Filters were washed, and [-32P]MBP
was quantified by scintillation counting. Duplicate assays in the
absence of MBP peptide were performed to determine non-MAPK background
kinase activities, which was subtracted for the MBP assays to determine
true MBP phosphorylation.
Data analysis
Student’s t tests with statistical significance at the 0.05 level or, when appropriate, ANOVA for multiple comparisons were used for data analysis. Results are expressed as the mean ± SEM.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To evaluate the regulation of GM-CSF production in AEC, we first
confirmed the response of primary culture HBECs to cytokines previously
reported to enhance GM-CSF production in AEC. HBECS were grown to near
confluence, and the release of GM-CSF in the supernatant was determined
by ELISA under resting or stimulated conditions. As shown in Fig. 1
a, stimulation of HBECs with
TNF- elicited a dose-dependent increase in GM-CSF production; these
levels continued to increase at doses as high as 0.3 µM. The increase
in GM-CSF production at this concentration was 
400% above that in
unstimulated controls (1.24 ± 0.20 ng/106
cells; n = 12; p < 0.004). Stimulation
of HBECs with IL-1 also induced a dose-dependent stimulation of GM-CSF
production, with maximal activity observed at 0.6–6 µM. The maximal
observed increase in GM-CSF production at 0.6 µM was 300% above
the unstimulated control value (1.11 ± 0.39
ng/106 cells; n = 5;
p < 0.5).
|
and IL-1 may be
dependent upon the activation of PKC, we also tested the effects of
PMA, an analogue of diacylglycerol and a direct activator of PKC. PMA
elicited a dose-dependent increase in GM-CSF production even greater
than that stimulated by TNF- and IL-1, with maximal activity at 5
ng/ml (8 nM; Fig. 1
b). At this concentration, PMA resulted
in a >14-fold increase in GM-CSF release (14.35 ± 5.0
ng/106 cells; n = 6;
p < 0.03). These observations established the capacity
TNF- and IL-1 to stimulate GM-CSF release in HBECs. Moreover, these
studies suggested the capacity of PKC to signal the production of
GM-CSF in these cells, raising the possibility that GM-CSF production
in response to IL-1 and TNF- may be PKC dependent.
To confirm that increases in GM-CSF protein release were associated
with similar increases in gene expression, we determined the effects of
PMA (8 nM) and TNF- (0.3 µM) on steady-state levels of GM-CSF mRNA
in HBECs. As shown in the representative Northern blot in Fig. 1c, steady-state GM-CSF mRNA was barely detectable in
unstimulated HBECs. Consistent with the ability of PMA to stimulate a
larger GM-CSF release than TNF-, the amount of message observed was
greater in PMA-treated cells than in TNF--treated cells.
Because of the difficulties involved in obtaining and culturing primary
bronchial epithelial cells, many studies of airway epithelium employ
BEAS-2B cells, an SV-40 adenovirus-transformed cell line originally
derived from normal human bronchial epithelial cells. To determine
whether BEAS-2B cells respond in a manner similar to that of HBECs, and
thus whether they can serve as an adequate model for epithelial cell
GM-CSF production, the release of GM-CSF in response to stimuli was
determined for BEAS-2B cells (passages 14–25). Exposure of BEAS-2B to
either TNF- (0.3 µM) or IL-1 (0.6 µM) resulted in a roughly
2-fold increase in GM-CSF above the unstimulated control value
(223 ± 11 and 148 ± 20% above unstimulated control,
respectively; n = 3; p < 0.05). In
contrast to the exuberant response to PMA demonstrated in HBECs, the
addition of PMA (8 nM) to BEAS-2B cells resulted in negligible GM-CSF
(0.32 ± 0.1 ng GM-CSF/106 cells above
background; n = 3). Thus, BEAS-2B cells failed to
respond in a manner similar to that of HBECs. In view of the difference
in response between HBECs and BEAS-2B cells, we focused all subsequent
studies on HBECs. Because the effect of TNF- was similar to that of
IL-1, all further cytokine studies were performed with TNF-.
Protein kinase C (PKC) activation and GM-CSF in HBECs
PMA has classically been considered a specific activator of PKC
and has been shown to activate both conventional and novel PKC
isoenzymes (37, 38). However, recent studies have
suggested that PMA may also activate additional targets including
n-chimaerin and Vav (39, 40). To confirm that GM-CSF
production in response to PMA was PKC dependent, we tested the ability
of PKC inhibitors to block PMA-stimulated GM-CSF production. In
addition, we tested the ability of mezerein, a daphnane analogue and
nonphorbol PKC agonist, to stimulate GM-CSF protein release. As
demonstrated in Fig. 2a, the
PKC inhibitors staurosporine (20 nM) and bisindolylmaleimide I (5 µM)
almost completely abrogated the response to PMA (9.0 ± 9 and
1.8 ± 0.4% PMA response, respectively; n = 3;
p < 0.01). Furthermore, as shown in Fig. 2b, mezerein (100 nM) elicited a robust increase in GM-CSF
protein by HBECs, a response similar to that induced by PMA. In
contrast, 4-phorbol 12,13-didecanoate (27 nM), an analogue of PMA
that does not stimulate PKC, had no effect on GM-CSF production.
|
a, GM-CSF release by TNF--stimulated HBECs
was also inhibited by staurosporine. These data support the importance
of PKC activation as a critical step in the upstream signaling pathway
leading to GM-CSF production by HBECs. Role of calcium in PKC-dependent GM-CSF production by HBECs
In T cells and endothelial cells stimulated with PMA, GM-CSF mRNA
production occurs only in the presence of a second stimulus, consisting
of a Ca2+ ionophore (28, 29, 41, 42). Moreover, cyclosporin A (CsA), which blocks the activity of
the calcium-binding protein calcineurin, inhibits increases in GM-CSF
mRNA in some cell types (43). We therefore tested to what
extent PMA-stimulated GM-CSF production in HBECs was dependent upon the
ionophore-induced increase in intracellular Ca2+
(Fig. 3a). As noted
previously, stimulation of HBECs with PMA in the absence of a
Ca2+ ionophore elicited a dramatic increase in
GM-CSF production. In contrast, stimulation of HBECs with the
Ca2+ ionophore A23187 (10 µM) in the absence of
PMA elicited a minimal and nonsignificant, increase in the release of
GM-CSF above background levels. The addition of A23187 to PMA resulted
in a small increase in GM-CSF production relative to PMA alone, an
effect that failed to achieve statistical significance.
|
b, the addition of A23187 to PMA resulted in a
small, but statistically insignificant, increase in GM-CSF message
(160% PMA response; n = 3; not significant).
The above studies were extended by testing the effects of CsA on GM-CSF
protein expression in HBECs. Cells were stimulated with PMA with or
without A23187 (18 h) in the presence or the absence of CsA, and GM-CSF
release was determined by ELISA (Fig. 3c). CsA (0.1–1 µM)
failed to inhibit GM-CSF production stimulated by PMA alone (98.3
± 1.2 and 99.2 ± 20.3% of control value; n =
3). GM-CSF release elicited by the combination of PMA and A23187 was
modestly inhibited by the higher dose (1.0 µM) of CsA (66.3 ±
4.4% stimulated release; n = 3; p <
0.01). We next tested the effect of CsA (0.1 µM) on steady-state mRNA
in cells stimulated with PMA and A23187. As demonstrated in Fig. 3, CsA
did not decrease steady-state mRNA for GM-CSF in HBECs when data were
normalized to GAPDH. Thus, in contrast to data reported for T cells and
endothelial cells, up-regulation of GM-CSF in response to PMA in HBECs
did not appear to require the addition of a calcium costimulus.
MAPKerk activation by PMA and TNF- in HBECs
The ability of PMA to cause the phosphorylation and activation of
the extracellular signal-regulated MAPK kinases
(MAPKerk1/2) is well established and has been
shown to depend upon PKC-dependent activation of upstream elements of
the MAPKerk pathway, including Ras and Raf-1
(44). In contrast, whereas several investigators have
recently reported the capacity of TNF- to stimulate
MAPKerk in a variety of cell types, the pathways
by which MAPKerk activation occurs in response to
TNF- are poorly understood (45). We therefore asked
whether PMA and TNF- retained the capacity to activate
MAPKerk in HBECs, and whether TNF- stimulation
of MAPKerk was PKC dependent. In our initial
experiments MAPKerk activation was determined by
immunoblot analysis of lysates of HBECs using an Ab that recognizes the
dually phosphorylated, active state of MAPKerk1/2
(p-MAPKerk1/2). As demonstrated in Fig. 4, p-MAPKerk1/2 was
slightly detected in resting HBECs. PMA induced a rapid (5- to 15-min
exposure) increase in the amount of p-MAPKerk1/2.
Phosphoactivation of MAPKerk by PMA was
persistent and continued to increase for at least 60 min. TNF-
induced minimal activation of MAPKerk after brief
periods of exposure (5–15 min); however, phosphoactivation of
MAPKerk was observed by 60 min. These effects
were not due to changes in total MAPKerk, as
duplicate blots or stripped blots subsequently exposed to an Ab that
does not distinguish between the active and the inactive state of
MAPKerk2 showed no change in total levels. In
contrast to MAPKerk, we did not observe PMA-induced
phosphoactivation of two other members of the MAPK family,
MAPKp38 and MAPKjnk, at
these time points (data not shown).
|
induced activation of MAPKerk1/2. HBECs were grown to near
confluence and stimulated with PMA (8 nM) or TNF- (0.3 µM), and
lysates were immunoprecipitated with rabbit polyclonal Ab directed
against MAPKerk1/2. Kinase activities in the
immunoprecipitates were then determined by their capacity to
phosphorylate MBP. As demonstrated in Fig. 4b, PMA induced
rapid activation of MAPKerk1/2 at 15 min,
consistent with the Western blot information. The level of PMA
stimulation at 30 min was similar to that observed at 15 min, although
the power of our analysis did not allow us to detect a significant
difference from baseline. TNF--induced activation of
MAPKerk1/2 was detected by 60 min.
The ability of both TNF- and PMA to stimulate
MAPKerk1/2 activity in HBECs together with our
observation that GM-CSF production stimulated by TNF- was PKC
dependent suggested that the capacity of TNF- to stimulate
MAPKerk1/2 activation might also be PKC
dependent. We therefore tested the effects of PKC inhibitors on
MAPKerk1/2 activation in response to TNF-. As
shown in Fig. 5, both staurosporine (20
nM) and bisindolylmaleimide (5 µM) inhibited TNF- stimulation of
MAPKerk1/2 phosphoactivation, as determined by
immunoblot. Taken together, these data indicate that TNF- retained
the capacity to activate MAPKerk1/2 in HBECS via
a PKC-dependent pathway.
|
We next asked whether the activation of MAPKerk1/2 was
necessary for the stimulation of GM-CSF production/release by HBECs.
The cell-permeant molecule PD98059 is a selective inhibitor of the MAPK
kinases (MKK) MEK1,2 (MKKmek1/2), proximal
activators of MAPKerk1/2 (46). As
shown in Fig. 5, preincubation of HBECs with PD98059 (40 µM) resulted
in marked inhibition of MAPKerk1/2 activity
induced by PMA or TNF-. The availability of a
MAPKmek/MAPKerk inhibitor
thus allowed us to test the effects of
MAPKmek/MAPKerk1/2
inhibition on GM-CSF production. HBECs were grown to near confluence
and stimulated with PMA or TNF- in the presence or the absence of
PD98059. As demonstrated in Fig. 6a, PD98059 elicited a
dose-dependent decrease in PMA-induced GM-CSF production with a 50%
inhibitory dose of 40 µM. PD98059 (40 µM) inhibited TNF-
stimulated GM-CSF release profoundly (12.7 ± 6.0% TNF-
response; n = 3; p < 0.003). A
selective inhibitor of the MAPKp38 pathway,
SB203580 (0.1–10 µM), failed to inhibit PMA-induced GM-CSF
production by HBECs, although SB203580 (0.1 µM) moderately inhibited
TNF--stimulated GM-CSF release (57.3 ± 11.2% TNF-
response; n = 3; p < 0.02; data not
shown).
|
b, steady-state
GM-CSF mRNA induced in response to PMA was decreased in the presence of
PD98059 (25.5 ± 4.4% PMA response; n = 3;
p < 0.001). Thus, PMA and TNF- stimulation of
GM-CSF message expression and protein production in HBECs correlated
with and appeared to depend upon the activation of
MAPKerk1/2 by these agents. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our experiments focused on primary culture epithelial cells. This model
was used because of the demonstration that transformed HBEC (BEAS-2B)
responded in a manner different from that of primary culture HBECs.
Both BEAS-2B cells and HBECs released a similar quantity of GM-CSF
protein and up-regulated GM-CSF mRNA in response to TNF- and
IL-1ß, a response in accordance with that reported by Nakamura et al.
(33). In contrast, the response to PMA differed markedly
between HBECs and BEAS-2B cells; HBECs, but not BEAS-2B cells, produced
abundant (nanogram) quantities of GM-CSF protein in response to PMA.
Differences in the response between primary culture and transformed
bronchial epithelial cells have also been demonstrated for additional
stimuli such as IL-4 (33). In addition, studies of
integrin expression demonstrate a disparity between the expression of
vß6,
5ß1, and
3ß1 in HBECS and
BEAS-2B cells (47). Thus, our data served again to
reinforce the need to evaluate findings in primary cell cultures as
well as in transformed cell lines.
Among the agents we tested, PMA was a potent stimulus for expression of GM-CSF mRNA and protein. PMA activates conventional and novel isozymes of PKC leading to activation of the Ras and Raf-1 pathways that result in up-regulation of MAPKerk1/2 (44). Since the intracellular pathways stimulated by phorbol esters have been well described, the use of this agent provided us with an instrument to begin to dissect signaling pathways in HBECs. In addition to multiple PKC isozymes, recent studies have demonstrated that PMA has high affinity for and can also activate additional signaling proteins, including the GTPase-activating proteins n-chimaerin and Vav (39, 40). However, PMA stimulation of GM-CSF in our studies was unlikely to be due to these alternative pathways, because the use of multiple PKC inhibitors blocked GM-CSF production, and the nonphorbol PKC agonist stimulated GM-CSF. Finally, the inactive PKC analogue had no effect. These data support the importance of PKC activation in GM-CSF release in HBECs; the role of specific isozymes awaits further analysis.
Although the ability of PMA to signal via PKC pathways has been
abundantly described, the pathways stimulated by cytokines are complex,
and the interaction with PKC isozymes is less clear. Our data suggest a
role for TNF--induced PKC activation in GM-CSF production in primary
culture cells. In particular, inhibition of PKC by staurosporine
inhibited TNF--induced GM-CSF production. TNF- triggers biologic
effects via engagement of two receptors: p55(TNFR-1) and p75 (TNFR-II),
both of which have been described in airway cells
(48, 49, 50, 51). Biologic responses induced by TNF are due to the
interaction of the cytoplasmic domains of multimerized receptors with
families of cytoplasmic proteins (49). The variation in
responses induced by ligation of TNF depends on cell type, state of
cell differentiation, and transformation. TNF- induced migration of
bovine epithelial cells and mucin secretion in guinea pig tracheal
epithelial cells have been demonstrated to be associated with PKC
activation (51, 52). Our studies thus suggest an
additional association of TNF and PKC in epithelial-type cells.
The requirement of PKC activation for GM-CSF induction in HBECs together with the ability of PKC to activate MAPK cascades suggested that one or more of these MAPK cascades might lie downstream of PKC in GM-CSF signaling. The common motif of MAPK pathways consists of three kinases that are sequentially activated (53). In the MAPKerk1/2 cascade, PKC activates Ras and c-Raf-1, leading to stimulation of MKKmek1/mek2 and subsequent activation of MAPKerk1/2 (44, 53, 54). Our studies extend the function of this pathway into primary culture cells and, using two methods for determination of MAPKerk1/2 activation, confirm that PKC activation stimulated the MAPKerk1/2 pathway. It is unlikely that either the MAPKjnk or the MAPKp38 pathway was similarly involved in PMA-stimulated GM-CSF release in HBECs, because 1) PMA failed to elicit phosphorylation of MAPKjnk and MAPKp38 at time points during which MAPKerk1/2 was observed; and 2) PMA-induced GM-CSF release was not inhibited by SB203580, an agent that inhibits MAPKp38 at low concentrations and has recently been reported to inhibit MAPKjnk at higher ones (>10 µM).
The ability of TNF receptor family members to activate the
MAPKerk1/2 pathway has been recently reported,
but is less well established (45). Our data demonstrated
that TNF- elicited up-regulation of MAPKerk1/2
in HBECs. The response was slightly delayed compared with that induced
by PMA, but was clearly present by 60 min. Moreover, the activation of
MAPKerk1/2 by TNF- was both associated with
and necessary for GM-CSF release. The upstream signals by which TNF
receptors activate MAPKerk1/2 have not been fully
described. Whereas the MAPKerk1/2 pathway is
associated with Ras/Raf-1 activation, the interaction of TNF signaling
with this pathway is not as well described and may be cell type
specific. The TNF receptor has been observed to interact with the
adaptor protein Grb2 and the exchange factor SOS in response to
TNF-, linking the TNF-R1 to c-Raf-1 kinase (55).
TNF- has also been suggested to activate the
MAPKerk1/2 pathway through sequential activation
of ceramide-activated protein kinase and Raf-1 (56, 57).
Recently, a novel serine-threonine kinase, RIP2 has been described to
interact with members of the TNF receptor family and interact
cooperatively with Ras, Raf-1, and MAPKerk1/2
(58). Our studies raise the tantalizing possibility that
the TNF--induced activation of MAPKerk1/2 in
HBECs is mediated by the cooperative activity of this pathway.
Engagement of the TNF- receptors has been well described to activate
MKKKmekk1,3 and MKKmek4,7
and subsequently isoforms of MAPKjnk or
MAPKp38 in a TRAF-dependent,
Ras/Raf-1-independent manner (53, 59, 60). In contrast to
our findings with PMA, our studies do not preclude the involvement of
MAPKjnk or MAPKp38 pathways
in TNF--induced GM-CSF production, because GM-CSF production was
moderately decreased in the presence of the
MAPKp38 inhibitor. However, the degree of the
effects observed suggest that the MAPKerk1/2
pathway is probably of primary importance.
In T cells and endothelial cells, activation of PKC via PMA is not sufficient for GM-CSF production (28, 29, 41, 42). Rather, the integrated stimulation of PMA and a Ca2+ influx results in activation of transcription factors of the NF-AT (NF of activated T cells) family in a CsA-sensitive manner (61, 62, 63). Our data suggest that in HBECs the addition of a Ca2+ ionophore was modulatory, but was not required for PMA-stimulated generation of GM-CSF mRNA or protein. Moreover, PMA-stimulated GM-CSF production was not inhibited by CsA. The data suggest that the mechanism for PMA-induced up-regulation of GM-CSF did not require activation of NF-AT as has been described for T cells. Concentrations of CsA higher than those usually used to inhibit NF-AT in T cells did produce modest inhibition of GM-CSF protein in cells stimulated with both PMA and a Ca2+ ionophore. These higher concentrations of CsA have been demonstrated to be required in cells in which NF-AT is minimally expressed (63). Thus, the small increase in GM-CSF induced by the Ca2+ ionophore and its inhibition by a high dose of CsA suggest the recruitment of additional pathways that may involve NF-AT.
In summary, our studies suggest that human primary culture bronchial epithelial cells are a potential source of abundant quantities of GM-CSF. Moreover, activation of PKC isozymes and the MAPKerk1/2 pathway is necessary for up-regulation of GM-CSF steady-state mRNA and protein release in HBECs. The regulation of GM-CSF mRNA and protein production in HBECs is different from that described in T cells and endothelial cells. Because local GM-CSF production may participate in the development of an airway immune and inflammatory response, understanding the cell-specific mechanisms that regulate its production in HBECs may allow for selective modulation of this cytokine.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Joan Reibman, Division of Pulmonary and Critical Care Medicine, New York University Medical Center, 550 First Avenue, New York, NY 10016.
3 Abbreviations used in this paper: AEC, airway epithelial cells; DC, dendritic cells; HBECs, primary culture human bronchial epithelial cells; PKC, protein kinase C; MAPK, mitogen-activated protein kinase; erk, extracellular signal-regulated kinases; jnk, c-Jun N-terminal kinase; MKK, MAPK kinase; PVDF, polyvinylidene difluoride; MBP, myelin basic protein; CsA, cyclosporin A.
Received for publication May 26, 1999. Accepted for publication May 19, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. Immunology 77:330.[Medline]
modulates the expression of its p60 receptor and several cytokines in rat tracheal epithelial cells. J. Immunol. 157:1089.
stimulates mucin secretion and cyclic GMP production by guinea pig tracheal epithelial cells in vitro. Am. J. Respir. Cell Mol. Biol. 20:413. in bovine bronchial epithelial cells. Am. J. Physiol. 273:L1007.
rapidly activates the mitogen-activated protein kinase (MAPK) cascade in a MAPK kinase kinase-dependent, c-Raf-1-independent fashion in mouse macrophages. Proc. Natl. Acad. Sci. USA 92:1614.. J. Biol. Chem. 271:13094.-induced activation of c-Jun N-terminal kinase is mediated by TRAF2. EMBO J. 16:1080.[Medline]
This article has been cited by other articles:
|
|
 |
|
  H. Slevogt, L. Maqami, K. Vardarowa, W. Beermann, A. C. Hocke, J. Eitel, B. Schmeck, A. Weimann, B. Opitz, S. Hippenstiel, et al. Differential regulation of Moraxella catarrhalis-induced interleukin-8 response by protein kinase C isoforms Eur. Respir. J., April 1, 2008; 31(4): 725 - 735. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Bleck, D. B. Tse, I. Jaspers, M. A. Curotto de Lafaille, and J. Reibman Diesel exhaust particle-exposed human bronchial epithelial cells induce dendritic cell maturation. J. Immunol., June 15, 2006; 176(12): 7431 - 7437. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Oshiro, H. Otani, Y. Yagi, S. Fukuhara, and C. Inagaki Protease-Activated Receptor-2-Mediated Inhibition for Ca2+ Response to Lipopolysaccharide in Guinea Pig Tracheal Epithelial Cells Am. J. Respir. Cell Mol. Biol., June 1, 2004; 30(6): 886 - 892. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. L. Miakotina and J. M. Snyder Signal transduction events involved in TPA downregulation of SP-A gene expression Am J Physiol Lung Cell Mol Physiol, June 1, 2004; 286(6): L1210 - L1219. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Kawaguchi, F. Kokubu, S. Matsukura, K. Ieki, M. Odaka, S. Watanabe, S. Suzuki, M. Adachi, and S.-K. Huang Induction of C-X-C Chemokines, Growth-Related Oncogene {alpha} Expression, and Epithelial Cell-Derived Neutrophil-Activating Protein-78 by ML-1 (Interleukin-17F) Involves Activation of Raf1-Mitogen-Activated Protein Kinase Kinase-Extracellular Signal-Regulated Kinase 1/2 Pathway J. Pharmacol. Exp. Ther., December 1, 2003; 307(3): 1213 - 1220. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Reibman, Y. Hsu, L. C. Chen, B. Bleck, and T. Gordon Airway Epithelial Cells Release MIP-3{alpha}/CCL20 in Response to Cytokines and Ambient Particulate Matter Am. J. Respir. Cell Mol. Biol., June 1, 2003; 28(6): 648 - 654. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Page, J. Li, L. Zhou, S. Iasvoyskaia, K. C. Corbit, J.-W. Soh, I. B. Weinstein, A. R. Brasier, A. Lin, and M. B. Hershenson Regulation of Airway Epithelial Cell NF-{kappa}B-Dependent Gene Expression by Protein Kinase C{delta} J. Immunol., June 1, 2003; 170(11): 5681 - 5689. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Saba, G. Soong, S. Greenberg, and A. Prince Bacterial Stimulation of Epithelial G-CSF and GM-CSF Expression Promotes PMN Survival in CF Airways Am. J. Respir. Cell Mol. Biol., November 1, 2002; 27(5): 561 - 567. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Reibman, Y. Hsu, L. C. Chen, A. Kumar, W. C. Su, W. Choy, A. Talbot, and T. Gordon Size Fractions of Ambient Particulate Matter Induce Granulocyte Macrophage Colony-Stimulating Factor in Human Bronchial Epithelial Cells by Mitogen-Activated Protein Kinase Pathways Am. J. Respir. Cell Mol. Biol., October 1, 2002; 27(4): 455 - 462. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Wolf, C. Schulz, G.A.J. Riegger, and M. Pfeifer Tumour necrosis factor-{alpha} induced CD70 and interleukin-7R mRNA expression in BEAS-2B cells Eur. Respir. J., August 1, 2002; 20(2): 369 - 375. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Graness, C. E. Chwieralski, D. Reinhold, L. Thim, and W. Hoffmann Protein Kinase C and ERK Activation Are Required for TFF- peptide-stimulated Bronchial Epithelial Cell Migration and Tumor Necrosis Factor-alpha -induced Interleukin-6 (IL-6) and IL-8 Secretion J. Biol. Chem., May 17, 2002; 277(21): 18440 - 18446. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Kawaguchi, L. F. Onuchic, and S.-K. Huang Activation of Extracellular Signal-regulated Kinase (ERK)1/2, but Not p38 and c-Jun N-terminal Kinase, Is Involved in Signaling of a Novel Cytokine, ML-1 J. Biol. Chem., May 3, 2002; 277(18): 15229 - 15232. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. P. HALLSWORTH, L. M. MOIR, D. LAI, and S. J. HIRST Inhibitors of Mitogen-activated Protein Kinases Differentially Regulate Eosinophil-activating Cytokine Release from Human Airway Smooth Muscle Am. J. Respir. Crit. Care Med., August 15, 2001; 164(4): 688 - 697. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
0.01).
