* Center for Experimental Therapeutics and Reperfusion Injury, Brigham and Women’s Hospital; and
Combined Program in Pediatric Gastroenterology and Nutrition, Children’s Hospital, and Harvard Medical School, Boston, Massachusetts
Correspondence: Sean P. Colgan, Ph.D., Brigham and Women’s Hospital, Center for Experimental Therapeutics and Reperfusion Injury, Thorn 704, 75 Francis Street, Boston, MA 02115. E-mail: colgan{at}zeus.bwh.harvard.edu
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: interleukin-8 • mucosal disorders • hypoxia
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A common feature of many disease processes is tissue hypoxia. Recent studies have shown that hypoxia induces a number of genes through diverse mechanisms and likely contributes significantly to inflammatory processes [5 ]. Tissue hypoxia may orchestrate leukocyte recruitment through induction of chemotactic cytokines (chemokines), polypeptides that function as leukocyte activators and chemoattractants in vitro and in vivo [6 ]. Chemokines can function as soluble mediators or can specifically bind to polysaccharide components of cell-surface and extracellular-matrix proteoglycans [7 ]. Proteoglycan binding sites on chemokines are generally distinct from the receptor binding site, utilize basic amino acid residues (lysine, arginine, or histidine) for binding, and proteoglycans can bind multiple chemokine molecules [8 , 9 ]. Some studies have suggested that chemokine activity is enhanced when bound to matrix components, particularly the proteoglycan heparan sulfate [7 , 10 ], possibly through oligomerization of the chemokine [8 ]. Proteoglycan structure can vary significantly, even between tissues, and it has been hypothesized that these structural differences may account for tissue localization of specific chemokines and, as a result, specific leukocyte populations [11 ].
The chemokine interleukin-8 (IL-8) has an established role in mucosal disease [12 ], and intestinal epithelia are primary sources of IL-8 [13 ]. Mechanistic in vitro studies suggest that epithelial-derived IL-8 serves as a recruitment signal for PMN. For example, apical colonization of epithelia with pathogenic bacteria induces the basolateral release of IL-8 and imprints the extracellular matrix with chemokine [14 ]. This insoluble, matrix-associated chemokine serves as a haptotactic recruitment signal for PMN to the basolateral epithelial surface [14 , 15 ]. Less is known, however, about the subsequent steps in PMN transmigration (i.e., after initial PMN-epithelial contact). PMN adhesive interactions to an as yet unidentified epithelial ligand are mediated primarily by CD11b/18, whereas transmigration through the epithelial paracellular space is predominantly CD47-dependent [16 ]. This epithelial paracellular pathway provides a unique environment for leukocyte movement. Unlike the thin, flat substrate of the vascular endothelium, transmigration through columnar epithelium mandates that PMN move greater than two cell lengths through the paracellular space. It is possible, therefore, that epithelia orchestrate PMN movement through the paracellular space by providing chemokine in a surface-expressed fashion. At present, this pathway has not been studied. We demonstrate here that epithelial cells pre-exposed to cellular hypoxia express functional IL-8 in a surface-expressed manner. Such induction of surface chemokine required the parallel activation of both IL-8 release and the surface up-regulation of perlecan. Thus, under defined conditions, the epithelial cell surface may provide a haptotactic substrate for PMN movement.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Epithelial cultures were exposed to hypoxia as previously described [15 ]. Briefly, growth media was replaced with fresh media equilibrated with hypoxic gas mixture and cells were placed in the hypoxic chamber (Coy Laboratory Products, Ann Arbor, MI). Oxygen concentrations were as indicated in torr (normoxia equal to pO2 of 150 torr) with the balance made up of nitrogen, carbon dioxide (constant pCO2 35 torr), and water vapor from the humidified chamber. These culture conditions are non-toxic to epithelia [15 ].
Quantitation of IL-8
Supernatants from normoxic and hypoxic epithelial monolayers
were quantitated for IL-8 secretion with the use of a capture sandwich
enzyme-linked immunosorbent assay (ELISA) and an IL-8 standard curve as
described before [15
]. Values are reported as IL-8 per
106 cells in a 1-mL volume of conditioned supernatant.
Detection of surface-associated IL-8 or surface-associated heparan sulfate was determined using intact monolayers of normoxic or hypoxic T84 cells or by flow cytometry of cells that were lifted from the permeable support. Briefly, confluent T84 cells were grown to electrical confluence on permeable supports, incubated in normoxia or hypoxia as indicated. Analysis of surface chemokine or proteoglycan on intact cell monolayers was performed by ELISA on insert-grown cells (0.33 cm2) as described previously [19 ] (values are reported as IL-8 per 106 cells relative to a standard curve). For assessment by flow cytometry, intact monolayers (5 cm2) were transferred to a solution of modified Hanks’ balanced salt solution (HBSS; without Ca2+ or Mg2+) containing 5 mM ethyleneglycol-bis(ß-aminoethyl ether)-N, N, N’, N’-tetraacetate (EGTA, Sigma Chemical) at 37°C. Monolayers were agitated for 2 h at 37°C and vigorously washed to remove cells, as described elsewhere [14 ]. Cells elicited in this fashion were cooled to 4°C, blocked with 100 µL media containing 10% bovine calf serum (BCS) for 1 h at 4°C, followed by addition of rabbit anti-human IL-8 polyclonal antibody (20 µg/mL in media, Endogen, Boston, MA) or control rabbit serum (Sigma, 1:500 dilution in media) for 2 h at 4°C. Cells were extensively washed with cold HBSS followed by addition of fluorescein-conjugated goat anti-rabbit secondary Ab (1:1000 in media, Cappel, West Chester, PA) and analyzed by flow cytometry. Control cells to determine background were incubated with only the secondary antibody. Cells were analyzed for surface expression of IL-8 or CD11b, as indicated, by flow cytometry as described before [20 ] with a FACScan flow cytometer (Becton Dickinson Immunocytometry Systems, Mountain View, CA) and CellQuestTM software.
Transcriptional analysis
The transcriptional profile of epithelial cells exposed to
ambient hypoxia was assessed in RNA derived from control or hypoxic
epithelia (T84 cells at 6 or 18 h hypoxia) using quantitative
genechip expression arrays (Affymetrix) [21
]. Reverse
transcriptase-polymerase chain reaction (RT-PCR) analysis of mRNA
levels was performed using DNAse-treated total RNA as previously
described [22
] using primers specific for IL-8 (5’-ATG
ACT TCC AAG CTG GCC GTG GCT-3’ and antisense primer 5’-TCT CAG CCC
TCT TCA AAA ACT TCT C-3’, 289-bp fragment), epican (5’-GCA GAA TGT
GGA CAT GAA GA-3’, and antisense primer 5’-ATG CTA AAA AAG ATT CGC
AAT G-3’, 150-bp fragment), perlecan (5’-CTG CCC CTC GTA GGC ACC-3’
and antisense primer 5’-CAC TCC CAT CCA ACC TGC-3’, 231-bp fragment),
syndecan-1 (5’-GAA CCA AAT CTG GAA GCC AA-3’ and antisense primer
5’-GAC ACA GGC ACG ACT GCT T-3’, 182-bp fragment), or control
ß-actin [22
] (5’-ATG ACT TCC AAG CTG GCC GTG GCT-3’
and antisense primer 5’-TCT CAG CCC TCT TCA AAA ACT TCT C-3’, 661-bp
fragment). Each primer set was amplified using 25 cycles of 94°C for
1 min, 60°C for 2 min, 72°C for 4 min, and a final extension of
72°C for 7 min. The PCR reactions were then visualized on a 1.5%
agarose gel containing 5 µg/mL of ethidium bromide.
Chemokine assessment in biotinylated plasma membrane fractions
Epithelial cells were grown to confluence on 45-cm2
permeable supports, and exposed to hypoxia or normoxia as indicated.
Plasma membranes were isolated using nitrogen cavitation (200 psi, 8
min, 4°C) as previously described [16
]. Plasma
membrane fractions were labeled with biotin (1 mM in HBSS) as
previously described [16
] and re-pelleted by
ultracentrifugation to remove excess biotin. Recombinant human IL-8
(100 ng/mL) was directly biotinylated [1 mM NHS-biotin (Pierce
Chemical, Rockford, IL) in HBSS] and excess biotin was removed by
multiple washes on a 5-kDa cut-off membrane filter (Amicon, Beverly,
MA). Fractions were pre-cleared with 50 µL pre-equilibrated protein-G
Sepharose (Pharmacia, Uppsala Sweden). Immunoprecipitation of IL-8 was
performed with rabbit polyclonal anti-IL-8 followed by addition of 50
µL pre-equilibrated protein-G Sepharose and overnight incubation.
Washed immunoprecipitates were boiled in non-reducing sample buffer
[2.5% sodium dodecyl sulfate (SDS), 0.38 M Tris pH 6.8, 20%
glycerol, and 0.1% bromophenol blue], resolved by non-reducing
SDS-polyacrylamide gel electrophoresis (PAGE; 18% polyacrylamide gel),
transferred to nitrocellulose, and blocked overnight in blocking
buffer. Biotinylated proteins were labeled with streptavidin-peroxidase
(Pierce Chemical) and visualized by enhanced chemiluminescence (ECL;
Amersham, Arlington Heights, IL).
Antisense oligonucleotide treatment of cells
Electrically confluent T84 cells grown as monolayers on
permeable supports were washed three times in warm T84 cell media. IL-8
phosphorothioate antisense oligonucleotide (40 base oligonucleotide)
derived from the 5’ untranslated sequence of exon 1
[23
], purchased from Oncogen Sciences (Cambridge, MA),
were diluted to 200 nM in serum-free media containing
N-[1-(2,3-dioleyloxy)propyl]-N, N,
N-trimethylammonium chloride (DOTMA, Lipofectin solution at
10 µg/mL, GIBCO) and applied to cells (150 µL) to the basolateral
surface only. Serum-containing media were used in the opposing well
(i.e., apical surface). Cells were incubated in normoxia or hypoxia, as
indicated, for 24 h followed by addition of 100 µL
serum-containing medium to the basolateral surface. Cells were allowed
to incubate for an additional 24 h, then used as described above
to assess IL-8 secretion or PMN transmigration. Controls consisted of
media only or media containing DOTMA without antisense
oligonucleotides.
Heparinase treatment of intact epithelia
Monolayers of the human intestinal epithelial cell line T84 were
grown as inverted monolayers on permeable supports. Confluent
monolayers were exposed to hypoxia (pO2 20 torr, 48 h), EGTA-elicited (as described above), and incubated in the presence
or absence of heparinase III (1 U/mL, Sigma) for 2 h at 37°C.
Monolayers were extensively washed at 4°C, and processed for flow
cytometry as described above.
Immunoprecipitation of biotinylated heparan sulfate
T84 cells were grown to confluence on 5-cm2
polycarbonate rings, exposed to experimental hypoxia conditions, as
indicated, washed with HBSS followed by labeling of extracellular cell
surface proteins with biotin (NHS-Biotin, 1 mM, Pierce), as described
previously [24
]. Unbound biotin was quenched with
NH4Cl (50 mM) in HBSS. Labeled T84 cells were lysed with
lysing buffer (150 mM NaCl, 25 mM Tris, 1 mM MgCl2, 1%
Triton X-100, 1% Nonidet P-40, 5 mM EDTA, 5 µg/mL chymostatin, 2
µg/mL aprotinin, and 1.25 mM PMSF, all from Sigma). Cell debris was
removed by centrifugation (10,000 g, 5 min). Lysates were
pre-cleared with 50 µL pre-equilibrated protein G-Sepharose
(Pharmacia) for 2 h. Immunoprecipitation of heparan sulfate was
performed by addition of anti-heparan sulfate monoclonal antibody
(clone A7L6, rat IgG2a from Upstate Biotechnology, Lake Placid, NY) for
2 h followed by 50 µL pre-equilibrated protein G-Sepharose
overnight on an end-over-end rotator. Washed immunoprecipitates were
boiled in non-reducing sample buffer (2.5% SDS, 0.38 M Tris, pH 6.8,
20% glycerol, and 0.1% bromophenol blue), separated by SDS-PAGE (10%
linear gel) under non-reducing conditions and transferred to
nitrocellulose through the use of standard protocols. Biotinylated
proteins were labeled with streptavidin-peroxidase and visualized by
enhanced chemiluminescence (ECL; Amersham). Resulting heparan sulfate
bands were quantified from scanned images using NIH Image software
(Bethesda, MD).
PMN transepithelial migration
PMN transmigration into and across confluent intestinal
epithelial monolayers was examined as previously described in detail
[25
, 26
]. Briefly, human PMN were isolated
from normal human volunteers by a gelatin sedimentation technique
[27
] and suspended in modified HBSS (without
Ca2+ and Mg2+, with 10 mM HEPES, pH 7.4, Sigma)
at a concentration of 5 x 107/mL. T84 monolayers,
plated in the inverted position to allow PMN interaction with the
physiologically relevant basolateral surface, were washed free of media
with HBSS (containing Ca2+ and Mg2+).
Transmigration assays were performed by the addition of PMN to the
upper chambers after chemoattractant (1 µM fMLP) was added to the
opposing (lower) chambers. At time 0, 1 x
106 PMN were added and transmigration was allowed to
proceed for 2 h at 37°C. In subsets of experiments, PMN were
pre-incubated with anti-IL-8 R mAb (anti-CXCR-1, clone 42705.111,
subclass IgG2a from R & D Sysytems, Minneapolis, MN) at indicated
concentrations for 30 min at 4°C before addition to epithelia. All
experiments were performed at 37°C. PMN migration into epithelial
monolayers was quantified by assaying for the PMN azurophilic granule
marker myeloperoxidase (MPO) as described previously
[26
]. After each transmigration assay,
nonadherent/loosely adherent PMN were extensively washed from the
surface of the monolayer and PMN cell equivalents, estimated from a
standard curve, were assessed as the number of PMN associated with the
monolayer. Previous studies have clearly demonstrated that the
morphological location of monolayer-associated PMN under these
conditions is within the epithelial paracellular space and not adherent
to the permeable support or the apical membrane surface
[25
, 28
].
PMN CD11b/18 expression
The influence of epithelia on PMN CD11b/18 was assessed by flow
cytometry during co-incubation of PMN with intestinal epithelial cells.
Briefly, 2.5 x 106 PMN were co-incubated with
1.25 x 107 T84 cells (pre-exposed to hypoxia or
hypoxia with added antisense oligonucleotides) for 20 min at 37°C.
Cells were immediately washed in HBSS, and incubated with anti-CD11b
monoclonal antibody (clone 44a, subclass IgGsa obtained from American
Type Tissue Collection) followed by FITC-conjugated secondary antibody.
CD11b/18 was analyzed on PMN from co-incubations by gating flow
cytometer on PMN.
Data presentation
ELISA data were compared by analysis of variance (ANOVA) or by
Student’s t test. Values are expressed as the mean and
SEM of n experiments. Flow cytometry data were
expressed as mean fluorescence intensity (MFI) and analyzed by the
Kolmogorov-Smirnov (K-S) two-sample test.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (44K): [in a new window] |
Figure 1. Hypoxia induces soluble and monolayer-associated IL-8, and IL-8 mRNA.
Monolayers of the human intestinal epithelial cell line T84 were grown
on polycarbonate permeable supports. Confluent monolayers were exposed
to hypoxia (pO2 20 torr, 0–48 h). (A) Soluble supernatants
were collected from the basolateral surface and assayed for IL-8 by
capture ELISA. (B) Washed monolayers were assessed for
monolayer-associated IL-8 by ELISA. Results represent the mean ±
SEM for 10–12 individual monolayers in each condition.
Values are reported as IL-8 per 106 cells relative to a
standard curve. (C) RT-PCR was used to determine IL-8 message levels in
response to hypoxia. After exposure to indicated periods of hypoxia,
total RNA was isolated, DNAse I treated, and amplified by RT-PCR using
IL-8 specific primers. As indicated, lanes represent (from left to
right); normoxia, 2-, 6-, 12-, and 24-h exposure to hypoxia,
respectively. In the bottom panel, corresponding ß-actin mRNA
expression are shown to demonstrate equivalent loading. Data shown are
representative of three experiments.
|
 View larger version (33K): [in a new window] |
Figure 2. Evaluation of epithelial membrane-associated IL-8. Monolayers of T84
cells were grown on polycarbonate permeable supports. Confluent
monolayers were exposed to hypoxia (pO2 20 torr, 0–48 h,
as indicated). (A) Surface IL-8 was assessed from EGTA-elicited
epithelial cells by flow cytometry. Also indicated is the secondary
antibody only control (2° only). (B) Plasma membrane preparations
were generated from epithelial cells pre-exposed to indicated hypoxia.
Membranes were biotinylated, re-pelleted, and IL-8 was
immunoprecipitated from solubilized fractions. Washed
immunoprecipitates were boiled in non-reducing sample buffer, resolved
by non-reducing SDS-PAGE (18% polyacrylamide gel), transferred to
nitrocellulose, and blocked overnight in blocking buffer. Biotinylated
proteins were probed with streptavidin-peroxidase and visualized by
enhanced chemiluminescence (ECL). Also indicated is a biotinylated IL-8
standard for reference. Results are representative of four experiments
(A) and three experiments (B).
|
Influence of IL-8 antisense oligonucleotides on cell-surface
chemokine
To better define the role of hypoxia in induction of IL-8
surface-bound chemokine, IL-8 mRNA was blocked in hypoxic cells with
antisense oligonucleotide probes. As we have demonstrated previously
[15
], the conditions for loading oligonucleotides do not
influence cell viability (judged morphologically and assessed by
transepithelial resistance, a sensitive measure of epithelial
viability, data not shown). As shown in Figure 3
, hypoxic epithelia incubated with IL-8 antisense oligonucleotides
for 48 h result in significantly decreased surface IL-8 (MFI of
620.5 vs. 12.8 for 48 h hypoxia and 48 h hypoxia pre-loaded
with IL-8 anti-sense oligonucleotides, respectively, P < 0.001). As a control, pre-loading epithelia with a scrambled
oligonucleotide did not influence surface IL-8 expression (data not
shown). No differences were observed between control normoxia cells and
48 h hypoxia pre-loaded with IL-8 anti-sense oligonucleotides (MFI
of 9.3 vs. 12.8, respectively, P = not significant).
These data indicate that IL-8 antisense oligonucleotides prevent
hypoxia-induced surface IL-8.
 View larger version (25K): [in a new window] |
Figure 3. Pre-loading epithelia with IL-8 antisense oligonucleotides (ASO) blocks
hypoxia-induced surface IL-8. Monolayers of T84 cells were grown on
polycarbonate permeable supports. Confluent monolayers were pre-loaded
with IL-8 ASO (200 nM) or mock treated with DOTMA followed by exposure
to hypoxia (pO2 20 torr, 48 h, as indicated). Surface
IL-8 was assessed from EGTA-elicited epithelial cells by flow
cytometry. Also indicated is the secondary antibody only control (2°
only) and the normoxic epithelial control (CTL). Results are
representative of three experiments.
|
 View larger version (28K): [in a new window] |
Figure 4. Heparinase treatment of epithelia decreases surface expression of IL-8
elicited by hypoxia. Monolayers of the human intestinal epithelial cell
line T84 were grown on polycarbonate permeable supports. Confluent
monolayers were exposed to hypoxia (pO2 20 torr, 48 h), EGTA-elicited (as described above), and incubated in the presence
or absence of heparinase III (1 U/mL) for 2 h at 37°C.
Monolayers were extensively washed at 4°C, and assessed for
surface-expressed IL-8 by flow cytometry. Also indicated is the
secondary antibody-only control (2° only) and the normoxic epithelial
control (exposed to similar conditions of heparinase, CTL). Results are
representative of three experiments.
|
50- and 90-kDa proteins were
present in control normoxic cells, but significantly induced by hypoxia
at 36–48 h (determined by densitometry, increase of 3.7 ± 0.9
and 4.2 ± 1.1 for the 50- and 90-kDa proteins, respectively,
P < 0.02 for both). The 68-kDa protein associated
with 24- and 36-h hypoxic cells was inconsistently observed.
 View larger version (56K): [in a new window] |
Figure 5. Hypoxia induces mRNA and surface expression of proteoglycans. T84
monolayers were cultured to confluency on 5-cm2 permeable
supports and exposed to indicated periods of hypoxia (pO2
20 torr). (A) Total RNA was isolated and examined for perlecan, epican,
and syndecan transcript by RT-PCR. ß-Actin transcript was used as an
internal control. (B) Cells were cooled to 4°C and labeled (apical
and basolateral) with biotin followed by immunoprecipitation of heparan
sulfate using clone A7L6 mAb. Immunoprecipitates were subjected to
SDS-PAGE on a 10% polyacrylamide gel followed by Western blotting,
incubation with streptavidin-peroxidase, and development by enhanced
chemiluminescence. Prominent 50- and 90-kDa bands are observed (open
arrows). The 68-kDa protein associated with 24 and 36 h hypoxic
cells was inconsistently observed. Results are representative of three
experiments.
|
 View larger version (20K): [in a new window] |
Figure 6. Hypoxia-induced surface expression of perlecan is diminished by the
cAMP agonist forskolin. Monolayers of the human intestinal epithelial
cell line T84 were grown on polycarbonate permeable supports. Confluent
monolayers were exposed to normoxia (pO2 150 torr, 48 h) or hypoxia (pO2 20 torr, 48 h) in the presence or
absence of indicated concentrations of the adenylate cyclase agonist
forskolin. Washed monolayers were assessed for monolayer-associated
perlecan by ELISA. Results represent the mean ± SEM
for 8–10 individual monolayers in each condition.
|
 View larger version (26K): [in a new window] |
Figure 7. Functional assessment of hypoxia-induced surface IL-8. (A) Monolayers
of the T84 cells were grown as inverted monolayers (i.e., basolateral
surface upward) on permeable supports. Confluent monolayers were
exposed to normoxia or hypoxia (pO2 20 torr, 48 h) and
reoxygenated with buffer containing PMN (1 x
106/monolayer) pre-incubated with indicated concentrations
of anti-IL-8 R mAb (CXCR-1). PMN were allowed to migrate for 2 h
at 37°C. Transmigration was quantitated by assaying for
monolayer-associated (the number of PMN that remain firmly associated
with washed monolayers) PMN myeloperoxidase. Results represent the
mean ± SEM for seven to nine individual monolayers in
each condition. (B) Monolayers of T84 cells were grown on polycarbonate
permeable supports. Confluent monolayers were pre-loaded with IL-8 ASO
(200 ng/mL) or mock treated with DOTMA followed by exposure to hypoxia
(pO2 20 torr, 48 h). Cells were EGTA-elicited and
co-incubated with PMN (5:1 epithelia-to-PMN ratio) for 15 min at
37°C. Washed co-incubations were assessed for PMN CD11b/18 (Mac-1) by
gating on PMN with flow cytometry. Also indicated is the secondary
antibody-only control (2° only) and baseline PMN Mac-1 expression
(basal). Results are representative of three experiments.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Leukocyte recruitment to tissue sites is highly ordered. After
extravasation from the vascular space, leukocytes migrating within
mucosal tissue respond to locally generated signals for successful
transit to the epithelium. These signals may exist in soluble forms
(e.g., fMLP) or may present themselves bound to matrix proteoglycans
(e.g., chemokines). Given its high negative charge, heparan sulfate
readily interacts with proteins and such interactions are increasingly
implicated in a variety of physiological processes [10
].
Previous studies, for instance, have indicated that chemokine
bioactivity in vitro may not be predictive in
vivo [11
], suggesting that additional molecules
contribute to chemokine function. A likely mechanism is the selective
association of chemokines with tissues specialized to recruit specific
leukocyte subpopulations. We have previously demonstrated that
epithelial hypoxia activates basolateral release of IL-8, and that IL-8
binds to epithelial extracellular matrix (i.e., is detected in
acellular monolayers) [15
]. The findings of chemokine
binding to proteoglycans are not unique. A number of studies have
directly demonstrated chemokine binding to glycosaminoglycan
subpopulations on numerous cell types [7
,
9
]. In some cases, proteoglycan-bound chemokine may show
enhanced bioactivity relative to the soluble forms of the molecule
[7
], and in one published finding, the chemokine
interferon-
-inducible protein-10 (IP-10) was demonstrated to use
proteoglycans as a signal-transducing receptor [40
].
Similarly, nuclear magnetic resonance analysis of IL-8-IL-8 receptor
complexes revealed that the heparin-binding residues of IL-8 are
exposed, thus allowing for the distinct possibility that surface
proteoglycans can present chemokine in a form that binds to the
receptor [41
]. Our studies indicate that epithelial
surface-bound IL-8 recruits PMN to the basolateral surface of intact
epithelia and that such surface-tethered chemokine provides a pathway
for induction of PMN CD11b/18 expression. We do not know, however,
whether chemokine binding to surface proteoglycan enhances chemokine
activity. Accurate estimates of chemokine concentrations expressed on
the cell surface are difficult to obtain with this model, limiting the
direct comparison of PMN CD11b/18 activation by soluble IL-8. Thus, at
the present time, we do not know whether IL-8 binding to epithelial
surface heparan sulfate enhances IL-8 bioactivity.
Previous studies indicate that hypoxia induces chemokine generation in
endothelia [42
] and in epithelia [15
]. At
present the mechanism(s) of such activation are not certain, but likely
involves activation mediated by the transcription factors cAMP response
element binding protein (CREB) [22
] and nuclear
factor-
B [22
, 42
]. Indeed, our analysis
of the cloned IL-8 gene [23
] revealed that it too bears
a CRE (TTTCGTCA) 150 base pairs upstream from the TATA signal in the
human IL-8 gene. Consistent with this observation, it was recently
revealed that cellular hypoxia induces activation of a number of genes
bearing CREB binding sites at or near the promoter [22
,
34
], including E-selectin [32
], tumor
necrosis factor 
[22
, 31
], COX-2
[43
], and MHC class II [22
,
31
]. A similar hypoxia-induced activation pattern was
observed with perlecan, which also bears a CRE (TGACGTGG) upstream of
the transcription start site [30
]. Moreover, perlecan
has been demonstrated to be cAMP responsive [37
]. Of
note, transcriptional analysis also revealed that epican is prominently
induced by hypoxia (Fig. 5)
, although mechanisms of such induction
remain unclear. Taken together, these studies indicate the likelihood
that hypoxia activates a cAMP-regulated IL-8 and perlecan expression in
epithelia.
A strong correlation between reperfusion injury and the acute inflammatory response exists [3 ]. The direct role of PMN has been implicated in tissue damage resulting from reperfusion injury, although mechanisms remain only partially understood [3 ]. In the intestine for instance, PMN accumulation at the level of the epithelium has been shown to play a central role in mucosal injury during intestinal reperfusion injury [44 45 46 ]. In disorders such as Crohn’s disease, significant evidence exists that ischemia may substantially contribute to tissue damage [47 48 49 ]. Thus, it seems likely that intestinal ischemia and PMN-epithelial interactions may be related events. Here, we demonstrate that hypoxia-induced surface IL-8 functionally activates PMN CD11b/18 expression. This observation has significance for a number of reasons. First, CD11b/18 has been extensively studied as an important adhesion molecule for PMN adhesion to, and transmigration across intestinal epithelia [50 ], and that hypoxia enhances PMN transmigration. Second, although we have previously demonstrated that PMN transmigration across epithelia pre-exposed to hypoxia is CD11b/18-dependent [15 ], the ligand for CD11b/18 on epithelia remains elusive. Third, the proteoglycan heparin (a glycosaminoglycan structurally related to heparan sulfate) is a demonstrated ligand for PMN CD11b/18 [51 ], thus it is tantalizing to speculate that heparan sulfate may also serve as a surface ligand for PMN CD11b/18. Using monoclonal antibodies directed against perlecan, we have not observed inhibition of PMN transmigration (data not shown). However, this approach is limited by the fact that these monoclonals may not bind to the perlecan epitope necessary to block function.
In conclusion, epithelial exposure to hypoxia activates the cAMP-regulated induction of both heparan sulfate and IL-8, resulting in the expression of a surface-bound form of chemokine that functionally activates PMN migration and CD11b/18. Thus, these studies contribute to the growing body of evidence that cAMP-responsive genes may play an important role in mediating crosstalk between hypoxia and inflammatory events.
Received January 12, 2000; revised March 31, 2000; accepted April 10, 2000.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
associates with heparan sulfates through the first beta-strand of the chemokine J. Biol. Chem. 274,23916-23925 J. Biol. Chem. 274,19447-19450 (TNF-) Gastroenterology 114,657-668[Medline] is an epithelial secretagogue J. Clin. Invest. 102,1161-1172[Medline]This article has been cited by other articles:
 |
|
 |
|
  W. Zheng, J. Kuhlicke, K. Jackel, H. K. Eltzschig, A. Singh, M. Sjoblom, B. Riederer, C. Weinhold, U. Seidler, S. P. Colgan, et al. Hypoxia inducible factor-1 (HIF-1)-mediated repression of cystic fibrosis transmembrane conductance regulator (CFTR) in the intestinal epithelium FASEB J, January 1, 2009; 23(1): 204 - 213. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Solic, J. Wilson, S. J. Wilson, and J. K. Shute Endothelial Activation and Increased Heparan Sulfate Expression in Cystic Fibrosis Am. J. Respir. Crit. Care Med., October 1, 2005; 172(7): 892 - 898. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. GOTTE Syndecans in inflammation FASEB J, April 1, 2003; 17(6): 575 - 591. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Narravula and S. P. Colgan Hypoxia-Inducible Factor 1-Mediated Inhibition of Peroxisome Proliferator-Activated Receptor {{alpha}} Expression During Hypoxia J. Immunol., June 15, 2001; 166(12): 7543 - 7548. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. T. Furuta, J. R. Turner, C. T. Taylor, R. M. Hershberg, K. Comerford, S. Narravula, D. K. Podolsky, and S. P. Colgan Hypoxia-inducible Factor 1-dependent Induction of Intestinal Trefoil Factor Protects Barrier Function during Hypoxia J. Exp. Med., April 30, 2001; 193(9): 1027 - 1034. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
