|
|
||||||||
1


*
Department of Biochemistry, University of Texas Health Center, Tyler, TX 75710; and
Department of Medicine, Saga Medical School, Saga, Japan
| Abstract |
|---|
|
|
|---|
mAb inhibited both the
cytotoxicity and IL-8 accumulation completely. When HMVEC-L were
incubated with supernatants from SEA-treated PBMCs, HMVEC-L
cytotoxicity was comparable with HMVEC-L incubated with SEA and PBMCs
at the same time. Although high concentrations of purified PMNs induced
HMVEC-L lysis in a dose-dependent manner, the effect of PMNs was not
changed in the presence of SEA. These findings suggest that TNF-
secreted by SEA-stimulated PBMCs plays a leading role in HMVEC-L
injury. | Introduction |
|---|
|
|
|---|
and IL-1ß in human monocytic cells
(12, 13, 14, 15). Proinflammatory cytokines such as TNF-
and IL-1ß can
stimulate the production of IL-8 (15), and there are reports that
describe TNF-
increasing the permeability of the endothelium
(16, 17, 18) and being associated with the acute lung injury (19). IL-8, a
potent PMN chemotactic and activating factor, has also been implicated
in the pathogenesis of ARDS (20). Furthermore, we have found that IL-8
concentrations are higher in the lungs of patients with ARDS associated
with sepsis than in nonseptic ARDS patients (21), and that i.v. SEA
increases the IL-8 concentration of plasma and epithelial lining fluid
in rabbits (22). In these studies, we examined the mechanism of lung
endothelial cell injury associated with SEA. | Materials and Methods |
|---|
|
|
|---|
All work involving human subjects was approved by the Institutional Human Subjects Committee at the University of Texas Health Center.
Cell culture
Human lung microvascular endothelial cells (HMVEC-L) (Clonetics, San Diego, CA) were maintained in EGM medium containing human epidermal growth factor (10 ng/ml), bovine brain extract (12 mg/ml), gentamicin sulfate (50 mg/ml), amphotericin-B (50 ng/ml) (Clonetics), and 10% FCS (Sigma, St. Louis, MO) at 37°C in a humidified atmosphere containing 5% CO2. HMVEC-L were grown as monolayers in tissue culture flasks. Cells were passaged when they reached 7080% confluence using trypsin (0.025%)/EDTA (0.01%) in HBSS (Clonetics), centrifuged at low speed (220 x g for 5 min), and resuspended in fresh medium. HMVEC-L were maintained for no longer than 3 wk.
Preparation of human PMNs and PBMCs
Human blood from healthy donors was anticoagulated with heparin
(Elkins-Sinn, Cherry Hill, NJ). PMNs were isolated by dextran
(Pharmacia, Piscataway, NJ) sedimentation and E lysis using the method
of Boyum (23) as modified in our earlier studies (24, 25) and were
further purified in gradients of Ficoll-Hypaque (density 1.114; ICN
Biomedicals, Costa Mesa, CA) (26) for cytotoxic assays. The isolated
PMNs were
99% pure and viable. PBMCs were also isolated in gradients
of Ficoll-Hypaque. The isolated PBMCs were
98% pure and 99% viable.
51Cr release cytotoxicity assay
The assay was performed as described previously (27). In brief,
HMVEC-L monolayers in 96-well plates were incubated with 2 Ci/well of
Na251CrO4 (DuPont-New England
Nuclear, Wilmington, DE) alone or with indicated concentrations of SEA
(Toxin Technology, Sarasota, FL) overnight at 37°C. Following the
incubation, the wells were washed three times and incubated in culture
medium for an additional 30 min at 37°C to allow spontaneous lysis of
marginally viable cells. After washing twice, a 100-µl aliquot of
SEA, purified mouse IgG1 anti-IL-8 mAb (R&D Systems, Minneapolis,
MN), anti-TNF-
mAb, anti-IL-1ß mAb (Biosource
International, Camarillo, CA), freshly isolated PBMCs, and/or PMNs were
added to each well. The cells were cultured for 21 h, and the
radioactivity in the supernatants was counted using a gamma radiation
spectrometer. Each well received culture medium alone or 2% SDS (EM
Industries, Cherry Hill, NJ) to determine spontaneous and maximum
release, respectively. Percent lysis was calculated using the following
formula: % Lysis = ([experimental cpm - spontaneous
cpm]/[maximum cpm - spontaneous cpm]) x 100.
Quantitation of IL-8
IL-8 accumulation in the supernatants was quantitated using an
ELISA as described previously (28, 29). The assay employed an
anti-IL-8 mAb (IgG1) (purified from ascites that had been developed
using HB9467 hybridoma cells (American Type Culture Collection,
Manassas, VA, with permission from Dr. E. J. Leonard, National
Cancer Institute, Frederick, MD)) (28) and rabbit polyclonal
anti-human IL-8 polyclonal antiserum (Upstate Biotechnology, Lake
Plasid, NY) followed by swine anti-rabbit Igs conjugated with
horseradish peroxidase (Dako, Carpinteria, CA). The immunoassay
was specific for IL-8 and did not cross-react with other members of the
-chemokine family (29).
Flow cytometric analysis
Flow cytometric analysis was performed as described previously with some modifications (30). Briefly, PMNs that had been freshly isolated using dextran and Ficoll-Hypaque were incubated with FITC-labeled anti-CD16 mAbs (Exalpha, Boston, MA) for 30 min at 4°C to identify the PMN population. Cells were then washed three times with cold PBS and incubated with phycoerythrin-labeled mAbs for Mac-1 (CD11b; Monosan, Uden, Netherlands) or L-selectin (CD62L; Exalpha) for an additional 30 min at 4°C. Cells were washed again three times and analyzed by FACScan (Becton Dickinson, Mountain View, CA). Control leukocytes were prepared by hypotonic lysis of freshly isolated blood and incubated with mAbs as described above.
Mean fluorescence intensity (MFI) was calculated, and the percent stimulation of expression was calculated using the following formula: % Stimulation = ([MFI of purified PMNs - MFI of control leukocytes]/MFI of control leukocytes) x 100.
Statistics
Data are expressed as mean values ± SD. Significant
differences between the means of two groups were assessed using the
Student t test. Data were considered statistically
significant if p values were
0.05. The experiments were
performed at least twice with at least four replicate cultures per
experiment.
| Results |
|---|
|
|
|---|
Endothelial cell injury was estimated as the release of chromium
from prelabeled HMVEC-L. When HMVEC-L were incubated with PBMCs and
SEA, HMVEC-L cytotoxicity increased significantly compared with the
incubation with PBMCs alone (p < 0.0001) (Fig. 1
). Alternatively, when HMVEC-L were
incubated with PMNs in the presence of SEA, there was no increase in
endothelial cell lysis as compared with the incubation with PMNs alone.
There was also no additional increase in cytotoxicity when PBMCs and
SEA were incubated with PMNs. However, when HMVEC-L and SEA were
incubated with PBMCs, the percentage of cell lysis was increased in a
dose-dependent manner. There were significant differences from control
cultures (incubated without PBMCs) at concentrations
7.5 x
104 cells/ml of PBMCs (p < 0.007)
(Fig. 2
A). When HMVEC-L and
PBMCs were incubated with SEA, the percentage of cell lysis also
increased in a dose-dependent manner. There were significant
differences from control cultures (incubated without SEA) at
concentrations
0.1 ng/ml of SEA (p < 0.01)
(Fig. 2
B).
|
|
SEA contributed to IL-8 production and secretion when it was
incubated with PBMCs (Fig. 3
). When
HMVEC-L and PBMCs were incubated with SEA, IL-8 accumulation was
increased significantly compared with incubation in the absence of SEA
(p = 0.0003). The addition of PMNs did not
contribute to the accumulation of IL-8 regardless of the presence of
SEA or PBMCs. Interestingly, HMVEC-L produced and secreted IL-8 in the
absence of any additional stimuli. When HMVEC-L and SEA were incubated
with PBMCs, IL-8 accumulation was increased and was related to the
number of PBMCs added; there were significant differences from control
cultures (incubated without PBMCs) at concentrations
1.5 x
104 cells/ml of PBMCs (p < 0.0001)
(Fig. 4
A). When HMVEC-L
and PBMCs were coincubated in the presence of SEA, IL-8 accumulation
also increased in a dose-dependent manner, and there were significant
differences from control cultures (incubated without SEA) at
concentrations
0.001 ng/ml of SEA (p <
0.0001) (Fig. 4
B).
|
|
To determine whether the cytotoxic factor and/or the stimulus for
IL-8 production was present in the medium, or whether cell contact was
required, PBMCs were incubated for 21 h in the presence or absence
of SEA. Next, supernatants were collected and incubated with HMVEC-L.
The cytotoxic effects induced by the coincubation of HMVEC-L, PBMCs,
and SEA were due to soluble factors released into the medium. When
HMVEC-L were incubated with SEA alone (no PBMCs), there was no
significant increase in cytotoxicity compared with the incubation with
medium alone (Fig. 5
A). When
HMVEC-L were incubated with supernatants that had been incubated with
PBMCs in the presence or absence of SEA, the percentage of cell lysis
was increased significantly compared with the incubation with medium
alone (p < 0.0001 and p =
0.0261, respectively). Furthermore, supernatants from SEA-stimulated
PBMCs were equally cytotoxic to HMVEC-L compared with when SEA, PBMCs
and HMVEC-L were present at the same time (p =
0.61). IL-8 accumulation in the supernatants before and after
incubation with HMVEC-L is shown in Fig. 5
B. Both PBMCs
and HMVEC-L produced and secreted IL-8, and SEA increased the
accumulation only when PBMCs had been incubated previously in the
medium.
|
51Cr-labeled HMVEC-L were incubated with SEA and PBMCs
in the presence of neutralizing Abs. When anti-IL-8 mAb was used,
the percentage of cell lysis did not change significantly (Fig. 6
A). However, when
anti-TNF-
mAb was incubated with HMVEC-L, SEA, and PBMCs, the
percentage of cell lysis was decreased in a dose-dependent manner; in
addition, there were significant differences from control cultures
(incubated without anti-TNF-
mAb) at concentrations
5 µg/ml
of anti-TNF-
mAb (p < 0.02) (Fig. 6
B). The SEA-induced cytotoxicity was completely inhibited
when anti-TNF-
mAb was coincubated at concentrations of
20
µg/ml (p = 0.267). Anti-IL-ß mAb, another
proinflammatory cytokine neutralizing Ab, was also tested (Fig. 6
C). In this case, there was no significant change in
cytotoxicity from control cultures grown in the absence of
anti-IL-1ß mAb.
|
mAb was
also quantitated (Fig. 7
mAb in a
dose-dependent manner and was completely inhibited at concentrations
1 µg/ml.
|
HMVEC-L were incubated with various concentrations of PMNs (Fig. 8
). In the absence of any added
stimulants, the percentage of cell lysis increased with the number of
PMNs added; there were significant differences from control cultures
(incubated without PMNs) at concentrations
1 x 106
cells/ml of PMNs (p < 0.002).
|
Flow cytometry was performed to determine whether PMNs were
activated by the purification procedure (Table I
). The expression of Mac-1 (CD11b) was
increased and the expression of L-selectin (CD62L) was decreased in
CD16+ cells.
|
| Discussion |
|---|
|
|
|---|
In this study, we describe a mechanism for SEA-induced endothelial
damage. TNF-
, which was produced and secreted by SEA-stimulated
PBMCs, is an essential component of endothelial injury. To our
knowledge, this is the first report that demonstrates a mechanism of
SEA-induced endothelial damage.
Many previous reports have described PMN-dependent endothelial damage and increase in endothelial permeability (9, 16, 32). However, ARDS has been reported in patients who are neutropenic (33, 34, 35, 36), suggesting that PMNs are not essential for its development. We found that the SEA-induced cytotoxic effect occurred when the toxin was incubated with PBMCs alone and was independent of the presence of PMNs. However, the cytotoxicity was dependent upon the concentrations of both SEA and PBMCs. Furthermore, supernatants from PBMCs incubated with SEA also induced the same level of cytotoxicity in HMVEC-L. These data support the hypothesis that the ability of SEA to cause endothelial cell lysis is not related to the presence of PMNs, suggesting a possible mechanism of endothelial injury in neutropenic patients.
IL-8 has been identified as a potent PMN chemotactic and activating factor (37, 38). Our previous study showed that the concentration of IL-8 in the airspaces is elevated in patients with ARDS. Additionally, the IL-8 in the lungs of patients with ARDS associated with sepsis reached greater concentrations than in nonseptic ARDS patients (20, 21). In this study, we found that IL-8 was produced and secreted by both PBMCs and HMVEC-L without any added stimulants. Although the accumulation of IL-8 was increased in parallel with the cytotoxicity when HMVEC-L were incubated with PBMCs and SEA, neutralizing Ab for IL-8 did not reduce the cytotoxicity. These data suggest that IL-8 is not directly responsible for HMVEC-L injury.
TNF-
and IL-1ß, which are known IL-8 inducers, are produced and
released by monocytic cells in response to SEA (12, 13, 14, 15). Also, both
TNF-
and IL-1ß have been reported to be associated with ARDS (19).
In particular, TNF-
increases pulmonary vascular permeability
independent of neutrophils (39). Although the addition of
anti-IL-1ß mAb did not affect HMVEC-L cytotoxicity, when
anti-TNF-
mAb was coincubated with HMVEC-L, SEA, and PBMCs, the
cytotoxicity decreased in a dose-dependent manner and was inhibited
completely by adding 20 µg/ml of the Ab. The accumulation of IL-8 was
also inhibited completely by adding 1 µg/ml of anti-TNF-
mAb.
These data suggest that TNF-
is essential for the HMVEC-L cytotoxic
reaction and the increase in the accumulation of IL-8. Since TNF-
also induces gene expression and the secretion of monocyte
chemoattractant protein-1 by human endothelial cells (40), it is also
possible that the migration of monocytes to a focus of inflammation
accelerates the production and secretion of TNF-
, which could worsen
the endothelial injury.
The interaction between toxic products from PMNs such as
myeloperoxidase and pulmonary endothelium is thought to increase
microvascular permeability to plasma proteins, and the resultant lung
edema is considered to be a major component of ARDS (41, 42).
Furthermore, it has been shown previously that i.v. administration of
IL-8 to rabbits induced changes in the lung histology that were
consistent with ARDS (43), and that IL-8 also plays a significant role
in PMNs adherence to and transmigration through vascular endothelium
(44). As shown in Fig. 8
, high concentrations of purified PMNs induced
HMVEC-L lysis in a dose-dependent manner. Our PMN-purification
protocol, which is a standard method for in vitro study, caused an
increase of Mac-1 as described previously (45), as well as a decrease
of L-selectin expression on the cell surface (Table I
). These changes
in adhesion molecules were also noted on IL-8-activated PMNs (46).
Therefore, it is possible that the activation of PMNs by the
purification procedure participates in the PMN-induced cytotoxic
effect. However, as shown in Fig. 1
, despite any purification-induced
activation, PMNs did not enhance the cytotoxic effect of PBMCs for
endothelial cells. These data indicate that PBMCs play an important
role in the pulmonary endothelial cytotoxicity induced by SEA.
In conclusion, we have demonstrated a mechanism of SEA-induced human
lung endothelium injury. TNF-
, which is secreted by SEA-induced
PBMCs, injures HMVEC-L and stimulates the production and
secretion of IL-8 from PBMCs and HMVEC-L. Because PMNs can be cytotoxic
to the endothelium, it is suggested that the accumulation of PMNs due
to IL-8 may accelerate the cytotoxicity of HMVEC-L. From the findings
we have reported here, it is expected that antagonists of TNF-
may
have an important role in the treatment of SEA-induced pulmonary
injury.
| Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Edmund J. Miller, Department of Biochemistry, University of Texas Health Center, Highway 271 at Highway 155, Tyler, TX 75710. ![]()
3 Abbreviations used in this paper: ARDS, acute respiratory distress syndrome; PMN, polymorphonuclear cell; SEA, staphylococcal enterotoxin A; HMVEC-L, human lung microvascular endothelial cell(s); MFI, mean fluorescence intensity. ![]()
Received for publication December 1, 1997. Accepted for publication July 7, 1998.
| References |
|---|
|
|
|---|
, LPS, and IL-1ß. Science 243:1467.
promoter by superantigen in human monocytic cells. J. Immunol. 155:902.[Abstract]
, lipopolysaccharide, and interleukin-1ß. Am. J. Respir. Cell Mol. Biol. 2:321.
induces endothelial cell F-actin depolymerization, new actin synthesis, and barrier dysfunction. Am. J. Physiol. 264:C894.
-chemokines inhibits the growth of melanoma cell lines. J. Clin. Invest. 99:2581.[Medline]
. Am. J. Pathol. 145:913.[Abstract]
This article has been cited by other articles:
![]() |
S. Ren, D. S. Terman, G. Bohach, A. Silvers, C. Hansen, H. Colt, and S. A. Sahn Intrapleural Staphylococcal Superantigen Induces Resolution of Malignant Pleural Effusions and a Survival Benefit in Non-Small Cell Lung Cancer Chest, November 1, 2004; 126(5): 1529 - 1539. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. L. Hale, S. B. Margolin, T. Krakauer, C. J. Roy, and B. G. Stiles Pirfenidone Blocks the In Vitro and In Vivo Effects of Staphylococcal Enterotoxin B Infect. Immun., June 1, 2002; 70(6): 2989 - 2994. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Ben-Ari, D. Gozal, R. J. Dorio, C. M. Bowman, A. Reiff, and S. M. Walker Superantigens and Cystic Fibrosis: Resistance of Presenting Cells to Dexamethasone Clin. Vaccine Immunol., July 1, 2000; 7(4): 553 - 556. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |