Download Regulated MIP-3/CCL20 production by human intestinal epithelium

Am J Physiol Gastrointest Liver Physiol
280: G710–G719, 2001.
Regulated MIP-3␣/CCL20 production by human intestinal
epithelium: mechanism for modulating mucosal immunity
ARASH IZADPANAH, MICHAEL B. DWINELL, LARS ECKMANN,
NISSI M. VARKI, AND MARTIN F. KAGNOFF
Laboratory of Mucosal Immunology, Department of Medicine, University of California,
San Diego, La Jolla, California 92093-0623
Received 13 July 2000; accepted in final form 19 October 2000
of the human intestine forms
a physical barrier that separates the internal milieu of
the host from luminal contents. During the course of
intestinal inflammation or microbial infection, intestinal epithelial cells can, in addition to their normal
absorptive and secretory functions, develop additional
characteristics usually attributed to classic inflamma-
tory cell types. Thus, after stimulation with proinflammatory mediators such as tumor necrosis factor
(TNF)-␣ or interleukin (IL)-1␣ or in response to infection with enteric pathogens, intestinal epithelial cells
upregulate a program of genes whose products can
signal the onset of an acute mucosal inflammatory
response characterized by an influx of neutrophils and
monocytes (9, 10, 21, 23, 29, 32, 37, 42). Many of the
genes, including several of the chemokine genes that
are activated in intestinal epithelial cells in response to
agonist stimulation or bacterial infection, are target
genes of the transcription factor nuclear factor (NF)-␬B
(13, 19, 20, 32). In this regard, NF-␬B can be viewed as
a central regulator of the intestinal epithelial cell response to a set of signals, activated by proinflammatory
stimuli and bacterial infection, that are thought to be
important for the initiation of mucosal innate immune
responses and acute mucosal inflammatory responses
(13). Whereas intestinal epithelial cells do not produce
the cytokines interferon (IFN)-␥, IL-2, IL-4, and IL-5,
which are essential components of host adaptive immune responses (9, 21), recent studies have shown that
they do produce three IFN-inducible T cell chemoattractants that may play a role in physiological inflammation characteristic of the normal intestinal mucosa
and in the chemoattraction of T helper (Th)1-type
CD4⫹ T cells within the intestinal mucosa (8).
Chemokines are low-molecular-weight chemotactic
cytokines that have a diverse set of activities after
binding and signaling through their cognate receptors
on target cells (2, 30). Chemokines play a key role in
the directional trafficking of leukocytes and dendritic
cells (DCs) and may play a further role in angiogenesis,
hematopoiesis, organogenesis, and viral pathogenesis
(2, 25, 30). Chemokines signal target cells through G
protein-coupled seven-transmembrane-spanning receptors that, in many cases, are promiscuous in that a
single chemokine receptor frequently binds several different chemokines. In addition, several known chemokines can bind to more than one chemokine receptor (2,
26, 30). The chemokine superfamily can be divided into
four groups based on the number and spacing of the
Address for reprint requests and other correspondence: M. F.
Kagnoff, Laboratory of Mucosal Immunology, Dept. of Medicine,
Univ. of California, San Diego, 9500 Gilman Dr., La Jolla, CA
92093-0623.
The costs of publication of this article were defrayed in part by the
payment of page charges. The article must therefore be hereby
marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734
solely to indicate this fact.
chemokines; dendritic cells; infectious immunity; inflammation; T lymphocytes
THE EPITHELIAL CELL LINING
G710
0193-1857/01 $5.00 Copyright © 2001 the American Physiological Society
http://www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.3 on June 17, 2017
Izadpanah, Arash, Michael B. Dwinell, Lars Eckmann, Nissi M. Varki, and Martin F. Kagnoff. Regulated
MIP-3␣/CCL20 production by human intestinal epithelium:
mechanism for modulating mucosal immunity. Am J Physiol
Gastrointest Liver Physiol 280: G710–G719, 2001.—Human
intestinal epithelial cells secrete an array of chemokines
known to signal the trafficking of neutrophils and monocytes
important in innate mucosal immunity. We hypothesized
that intestinal epithelium may also have the capacity to play
a role in signaling host adaptive immunity. The CC chemokine macrophage inflammatory protein (MIP)-3␣/CCL20 is
chemotactic for immature dendritic cells and CD45RO⫹ T
cells that are important components of the host adaptive
immune system. In these studies, we demonstrate the widespread production and regulated expression of MIP-3␣ by
human intestinal epithelium. Several intestinal epithelial
cell lines were shown to constitutively express MIP-3␣
mRNA. Moreover, MIP-3␣ mRNA expression and protein
production were upregulated by stimulation of intestinal
epithelial cells with the proinflammatory cytokines tumor
necrosis factor-␣ or interleukin-1␣ or in response to infection
with the enteric bacterial pathogens Salmonella or enteroinvasive Escherichia coli. In addition, MIP-3␣ was shown to
function as a nuclear factor-␬B target gene. In vitro findings
were paralleled in vivo by increased expression of MIP-3␣ in
the epithelium of cytokine-stimulated or bacteria-infected
human intestinal xenografts and in the epithelium of inflamed human colon. Mucosal T cells, other mucosal mononuclear cells, and intestinal epithelial cells expressed CCR6,
the cognate receptor for MIP-3␣. The constitutive and regulated expression of MIP-3␣ by human intestinal epithelium is
consistent with a role for epithelial cell-produced MIP-3␣ in
modulating mucosal adaptive immune responses.
INTESTINAL EPITHELIAL CELLS PRODUCE MIP-3␣
MATERIALS AND METHODS
Reagents. Recombinant human (rh)IL-1␣, IFN-␥, MIP-3␣,
and TNF-␣ were from PeproTech (Rocky Hill, NJ). Biotinconjugated affinity-purified goat anti-human MIP-3␣, murine monoclonal antibody (MAb) to human CCR6 (clone
53103.111), and murine MAb to human MIP-3␣ (clone
67310.111) were from R&D Systems (Minneapolis, MN).
Mouse IgG1 and IgG2b as well as MG-132 were from Sigma
Chemical (St. Louis, MO). Rabbit anti-human CD3 was from
Dako (Carpenteria, CA). Alexa 488-conjugated goat antirabbit IgG was from Molecular Probes (Eugene, OR), and
Cy3-conjugated goat anti-mouse IgG was from Jackson ImmunoResearch Laboratories (West Grove, PA).
Cell culture and stimulation protocols. The human colon
adenocarcinoma cell lines HT-29, HCA-7, LS174T, HCT-8,
and I-407 were grown in DMEM supplemented with 10%
heat-inactivated FCS and 2 mM L-glutamine as described
previously (7), and the Caco-2 cell line was grown in DMEM
supplemented with 15% heat-inactivated FCS. T84 human
colon carcinoma cells were grown in 50% DMEM-50% Ham’s
F12 medium supplemented with 5% newborn calf serum and
2 mM L-glutamine (21). HT-29, LS174T, HCT-8, I-407, and
Caco-2 were from the American Type Culture Collection,
HCA-7 colony 29 was a gift from S. C. Kirkland (Royal
Postgraduate Medical School, London, UK), and T84 was
initially obtained from K. Dharmsathaphorn (UCSD). Cells
were cultured at 37°C under 5% CO2-95% air.
For agonist stimulation, confluent monolayers of HT-29,
Caco-2, or T84 cells grown in six-well plates (Costar, Cambridge, MA) or Caco-2 cells grown in Transwell cultures
(24-mm diameter, 0.4-␮m pore size; Costar) were stimulated
with TNF-␣ (20 ng/ml), IL-1␣ (20 ng/ml), or IFN-␥ (40 ng/ml).
For bacterial infections, confluent HT-29 or T84 monolayers
in six-well plates were incubated with Salmonella dublin or
enteroinvasive Escherichia coli O29:NM at a multiplicity of
infection (MOI) of 100 and 500 for 1 h as described previously
(10, 11), after which the medium was removed and cells were
washed and incubated with fresh gentamicin (50 ␮g/ml)containing medium to kill remaining extracellular bacteria.
Human fetal intestinal xenografts. Human fetal intestine
(obtained from Advanced Biosciences Resources, Alameda,
CA), gestational age 12–18 wk, was transplanted subcutaneously onto the backs of C57BL/6 severe combined immunodeficiency (SCID) mice as described previously (11, 17, 23).
Human fetal intestinal xenografts were allowed to develop
for 10 wk after implantation, at which time the epithelium,
which is strictly of human origin, is fully differentiated (35).
Littermate SCID mice were injected subcutaneously with
⬃106 HT-29 cells, and tumors were allowed to develop for 5
wk. Mice carrying mature xenografts or HT-29 tumors were
injected intraperitoneally with 1 ␮g of human IL-1␣ in 200 ␮l
of PBS or with 200 ␮l of PBS alone. Intestinal xenografts and
HT-29 tumors were removed 5 h later, and adjacent segments of intestine and HT-29 tumors were frozen in liquid
nitrogen for RNA isolation or were embedded in optimum
cutting temperature compound (TissueTek, Torrance, CA)
and frozen in isopentane-dry ice or fixed in 10% neutral
buffered formalin for immunohistochemical analysis. In additional experiments, intestinal xenografts were infected
with ⬃5 ⫻ 107 of an attenuated aroA aroC S. typhi, in
DMEM-F12 medium at a 100-␮l volume, injected intraluminally by subcutaneous injection (17, 27). Those xenografts
were removed 6 h after infection, after which mucosal scrapings were prepared and immediately frozen in liquid nitrogen. These studies were approved by the University of California, San Diego Human and Animal Subjects Committees.
Adenovirus constructs and adenovirus infection. Recombinant adenovirus 5 (Ad5) containing an I␬B␣-AA superrepressor (Ad5I␬B-A32/36) or the E. coli ␤-galactosidase gene
(Ad5LacZ) was constructed as described previously (13, 20).
Ad5I␬B-A32/36 expresses a hemagglutinin (HA) epitopetagged mutant form of I␬B␣ in which serine residues 32 and
36 are replaced by alanine residues. The mutant I␬B␣ cannot
be phosphorylated at positions 32 and 36 and acts as a
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.3 on June 17, 2017
amino-terminal cysteines. In CC chemokines, two aminoterminal cysteines are adjacent, whereas in CXC chemokines, the two amino-terminal cysteines are separated by an intervening amino acid (2, 30).
Macrophage inflammatory protein (MIP)-3␣/CCL20
(31), also known as liver and activation-regulated chemokine (LARC) (15) or Exodus (16), is a member of the
CC chemokine subfamily initially noted to be expressed in human liver, lung, appendix, and tonsillar
crypts (5, 6, 15, 16, 38). MIP-3␣ is selectively chemotactic for CD34⫹ bone marrow cell-derived immature
DCs and CD45RO⫹ memory T cells that express the
cognate receptor CCR6 (1, 14, 24, 28). MIP-3␣ produced at sites of inflammation may chemoattract
CCR6-expressing immature DCs to the subepithelial
region of mucosal surfaces (5, 18, 34). As immature
DCs capture antigen at mucosal surfaces, they undergo
a functional and phenotypic change that includes a
decrease in CCR6 expression and a concomitant increase in expression of CCR7, the receptor for secondary lymphoid tissue chemokine (SLC) and MIP-3␤.
This change enables DCs to traffic out of the tissues,
where they encounter antigen and migrate to the blood
and secondary lymphoid organs (5, 34, 40). Almost all
of the memory T cells in the circulation that express
the ␣4␤7 integrin characteristic of mucosal homing
lymphocytes in humans (3) also coexpress CCR6 (24),
suggesting that MIP-3␣ may also be important for
chemoattracting mucosal T cells.
The orthologue of human MIP-3␣ in mice, termed
mLARC, was recently cloned and shown to have a
selective distribution in follicle-associated epithelium
overlying murine Peyer’s patches and other mucosal
lymphoid follicles (38). Unlike human MIP-3␣, mLARC
is not expressed in liver or lung (38), and mLARC
differs functionally from human MIP-3␣ in that
mLARC, but not human MIP-3␣, is chemotactic for
CCR6-expressing B cells (24). mRNA for MIP-3␣ has
been demonstrated in epithelial cells in human appendix and in inflamed epithelial crypts of tonsils (5, 38),
but little is known regarding the overall distribution
and regulation of MIP-3␣ production in the normal or
inflamed human intestinal tract. We hypothesized that
human intestinal epithelial cells might have the capacity to link mucosal innate and acquired immunity
through the regulated production of MIP-3␣, a chemokine capable of signaling immature DCs and CD45RO⫹
T cells. The studies herein describe the constitutive
and regulated expression and production of MIP-3␣
mRNA and protein by human intestinal epithelial
cells, using in vitro and in vivo models, and the presence of cells in the intestinal mucosa that express
CCR6, the cognate receptor for MIP-3␣.
G711
G712
INTESTINAL EPITHELIAL CELLS PRODUCE MIP-3␣
tory bowel disease were embedded in optimum cutting temperature compound and snap frozen in isopentane-dry ice as
described previously (7, 27). Serial cryostat sections (5 ␮m)
were prepared, fixed in acetone for 10 min, and blocked in
PBS-1% BSA for 1 h at room temperature. For MIP-3␣ and
CCR6 immunostaining, sections were incubated overnight at
4°C with 5 ␮g/ml mouse MAb to human MIP-3␣ or control
mouse IgG1 or 5 ␮g/ml mouse MAb to human CCR6 or control
mouse IgG2b, and sections were subsequently stained with
Cy3-conjugated goat anti-mouse IgG antibody. The isotypematched control antibodies were used at the same concentrations as the MIP-3␣ and CCR6 antibodies, respectively. For
CD3 immunostaining, sections were incubated as above with
rabbit anti-human CD3 or normal rabbit serum and subsequently stained with Alexa 488-conjugated goat anti-rabbit
IgG.
RESULTS
MIP-3␣ expression by epithelium of normal and inflamed adult human colon. MIP-3␣ is a chemokine
known to chemoattract populations of immature DCs
and CD45RO⫹ T cells. To test whether epithelial cells
are the major source of MIP-3␣ production in human
colon, sections of normal and inflamed human colon
were immunostained for MIP-3␣. As shown in Fig. 1B,
MIP-3␣ is minimally expressed by normal colon epithelium. In contrast, epithelial MIP-3␣ immunostaining
was markedly increased in sections of inflamed human
colon and the epithelium was the major site of MIP-3␣
production (Fig. 1A). No immunostaining was seen in
sections of inflamed human colon incubated with an
isotype-matched control antibody.
Constitutive and regulated MIP-3␣ mRNA expression in human intestinal epithelial cell lines. To better
characterize the expression of MIP-3␣ seen in vivo, we
used several human colon epithelial cell lines. As
shown in Fig. 2A, several human colon epithelial cell
lines, namely, HT-29, Caco-2, LS174T, and, to a lesser
extent, I-407, constitutively expressed MIP-3␣ mRNA.
In contrast, none of the cell lines expressed mRNA for
MIP-3␤, a chemoattractant for CCR7-expressing mature DCs. As shown in Fig. 2B, MIP-3␣ mRNA levels in
HT-29 and Caco-2 cells were upregulated after stimulation with the proinflammatory mediators TNF-␣ or
IL-1␣, which are cytokines produced by mononuclear
cells in the intestinal mucosa during the course of
mucosal inflammation. MIP-3␣ mRNA levels were not
upregulated by stimulation of the human colon epithelial cell lines with IFN-␥, granulocyte-macrophage colony-stimulating factor (GM-CSF), or IL-4 (data not
shown). None of the cytokines tested (i.e., TNF-␣, IL-1␣,
IFN-␥, GM-CSF, or IL-4) upregulated expression of the
mature DC chemoattractant MIP-3␤ (data not shown).
Regulated production of MIP-3␣ by human intestinal
epithelial cell lines. We next assessed whether the
constitutive and regulated expression of MIP-3␣
mRNA by the epithelial cell lines was paralleled by
protein production, using an ELISA to measure secreted levels of MIP-3␣. As shown in Fig. 3, unstimulated HT-29, Caco-2, and T84 cell lines secreted 2.1,
18.6, and 1.4 ng/ml MIP-3␣, respectively, as determined after 18-h incubation. Stimulation of those cells
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.3 on June 17, 2017
superrepressor of NF-␬B activation (13, 20). The HA epitope
tag enables identification of the exogenous superrepressor
with anti-hemagglutinin antibodies. Viral titers were determined by plaque assay. Recombinant virus was stored in PBS
containing 10% (vol/vol) glycerol at ⫺80°C.
HT-29 cells grown to confluence in six-well tissue culture
plates were infected with Ad5I␬B-A32/36 or Ad5LacZ in
serum-free medium at a MOI of 100 for 16 h as described
previously (13, 27). At this MOI, Ad5I␬B-A32/36 or Ad5LacZ
infected ⬎80% of HT-29 cells and infected cells expressed
I␬B␣-A32/36 and ␤-galactosidase, respectively, at high levels
as assessed by staining for ␤-galactosidase and immunostaining for HA-tagged I␬B␣-A32/36 (data not shown). After
infection, adenovirus was removed by washing, fresh medium containing serum was added, and cells were incubated
for an additional 12 h before bacterial infection or IL-1␣ or
TNF-␣ stimulation. The I␬B␣-AA superrepressor inhibited
NF-␬B activation in HT-29 cells, as assessed by electrophoretic mobility shift assay, and inhibited cytokine-induced
and bacteria-induced upregulation of IL-8 and intercellular
adhesion molecule-1 expression in HT-29 cells but did not
alter ␤-actin mRNA levels in the same cells (data not shown;
see Ref. 13).
RNA isolation. Total cellular RNA was extracted using an
acid guanidinium-phenol-chloroform method (TRIzol Reagent, Gibco Life Technologies, Grand Island, NY) and
treated with RNase-free DNase (Stratagene, La Jolla, CA) (7,
21). For RT-PCR, 1 ␮g of total cellular RNA was reverse
transcribed and cDNA was amplified as described previously
(21). The primers for human MIP-3␣ (sense 5⬘-ACC ATG
TGC TGT ACC AAG AGT TTG-3⬘ and antisense 5⬘-CTA AAC
CCT CCA TGA TGT GCA AGT GA-3⬘) and MIP-3␤ (sense
5⬘-CAG CCT GCT GGT TCT CTG GAC TTC-3⬘ and antisense
5⬘-GCC CCT CAG TGT GGT GAA CAC TAC-3⬘) were designed from available sequences from GenBank (NM004591
and AJ223410) and yielded PCR products of 390 bp and 276
bp, respectively. The amplification profile was 30 cycles of
45-s denaturation at 95°C and 2.5-min annealing and extension at 60°C for cell lines and tissues. Negative control
reactions had no added RNA in the RT and subsequent PCR
amplification. Peripheral blood mononuclear cells stimulated
overnight with 10 ␮g/ml phytohemagglutinin-P and 2 ␮g/ml
lipopolysaccharide (E. coli O111:B4) were used as a source of
cDNA for positive controls. The MIP-3␣ primers did not
amplify sequences in murine or human genomic DNA or
murine cDNA. After amplification, aliquots of the PCR reaction were size separated on a 1.0% agarose gel containing
ethidium bromide and photographed.
ELISA. Polystyrene 96-well plates (Immulon-4, Dynex
Technologies, Chantilly, VA) were coated with murine MAb
to human MIP-3␣ diluted in carbonate buffer as the capture
antibody. Affinity-purified biotinylated goat anti-human
MIP-3␣ diluted in PBS, 1.0% BSA, and 0.1% Tween-20 was
used as the detection antibody. The second step reagent was
horseradish peroxidase (HRP)-conjugated streptavidin.
Bound HRP was visualized with 3,3⬘,5,5⬘-tetramethylbenzidine dihydrochloride and H2O2 diluted in sodium acetate
buffer (pH 6.0), the color reaction was stopped by addition of
1.2 M H2SO4, and absorbance was measured at 450 nm.
MIP-3␣ concentration was calculated from a standard curve
using rhMIP-3␣. The ELISA was sensitive to 50 pg/ml.
Immunohistochemistry. Segments of human colon from the
histologically normal-appearing resection margin of a surgical specimen from an individual undergoing partial colectomy for diverticular disease, mucosal biopsies from four
healthy individuals obtained during screening colonoscopy,
and mucosal biopsies from three individuals with inflamma-
INTESTINAL EPITHELIAL CELLS PRODUCE MIP-3␣
G713
Fig. 2. Constitutive and regulated expression of MIP-3␣ mRNA by
human intestinal epithelial cell lines. A: total RNA was isolated from
the indicated cell lines, and MIP-3␣ and MIP-3␤ mRNA were amplified by RT-PCR using 30 amplification cycles. Peripheral blood mononuclear cells (PBMC) were stimulated with lipopolysaccharide and
phytohemagglutinin. Negative controls had no added RNA. B: total
RNA isolated from control HT-29 and Caco-2 cells or HT-29 and
Caco-2 cells stimulated for 3–12 h with tumor necrosis factor
(TNF)-␣ and interleukin (IL)-1␣ was amplified as in A.
with IL-1␣ or TNF-␣ markedly increased MIP-3␣ secretion within 3 h, with MIP-3␣ levels ranging as high
as 200–500 ng/ml at 18 h after stimulation. Consistent
with the lack of effect on mRNA expression, stimulation of the same cell lines with IFN-␥, GM-CSF, or IL-4
did not increase MIP-3␣ secretion (data not shown).
Consistent with the data for T84, the cell lines HCA-7
and HCT-8, which did not constitutively express
MIP-3␣ mRNA, could also be induced to upregulate the
production of MIP-3␣ protein in response to IL-1␣ or
TNF-␣ stimulation (data not shown).
The human intestinal epithelium is polarized into
phenotypically and functionally distinct apical and basolateral domains. For MIP-3␣ produced by epithelial
cells to imprint a chemotactic gradient and act as a
chemoattractant for mucosal T cells or DCs, it predictably would be secreted by epithelial cells in the physiologically relevant basolateral direction rather than
apically into the intestinal lumen. To determine
whether MIP-3␣ was vectorially secreted by intestinal
epithelial cells, we used Caco-2 cells grown on microporous filter supports as a polarized monolayer in
Transwell cultures. Polarized Caco-2 cells were stimulated with IL-1␣ added to the basal chamber, as intestinal epithelium in vivo would normally be exposed to
IL-1 produced in the intestinal mucosa at the basolateral membrane. These experiments revealed that polarized intestinal epithelial cells mainly secreted
MIP-3␣ into the basal rather than the apical Transwell
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.3 on June 17, 2017
Fig. 1. Epithelial macrophage inflammatory
protein (MIP)-3␣ immunostaining of normal
and inflamed human colon. Sections of inflamed (A) and normal (B) human colon were
immunostained for MIP-3␣. Adjacent sections
from inflamed (C) and normal (D) human colon were stained with hematoxylin and eosin.
As shown, MIP-3␣ immunostaining was heterogeneous and was most marked in the inflamed colon (A) and restricted to epithelial
cells. No immunostaining was seen in sections stained with an isotype-matched mouse
IgG1 antibody at the same concentration used
as a control (data not shown). The nonspecific
background staining was seen in sections incubated with either specific or nonspecific antibodies. Original magnification ⫻200.
G714
INTESTINAL EPITHELIAL CELLS PRODUCE MIP-3␣
Fig. 3. Secretion of MIP-3␣ by HT-29, Caco-2, and T84 cells stimulated with TNF-␣ or IL-1␣. Confluent HT-29 (A), Caco-2 (B), or T84
(C) cells were stimulated with TNF-␣ (E) or IL-1␣ () for 3–18 h,
whereas control cultures (F) remained unstimulated. MIP-3␣ in
culture supernatants was assayed by ELISA. Results are means ⫾
SE of 3–5 repeated experiments.
Table 1. MIP-3␣ secretion in HT-29 and T84 cells infected with Salmonella dublin
or Escherichia coli O29:NM
MIP-3␣ Secretion, ng/ml
⫹ S. dublin
No bacteria
⫹ E. coli O29:NM
Cell Line
6h
12 h
6h
12 h
6h
12 h
HT-29
T84
0.9 ⫾ 0.2
0.3 ⫾ 0.07
2.8 ⫾ 0.03
0.6 ⫾ 0.1
42.3 ⫾ 4.2
36.1 ⫾ 2.0
65.2 ⫾ 5.8
41.6 ⫾ 2.9
16.9 ⫾ 1.9
ND
59.6 ⫾ 7.8
ND
Values are means ⫾ SE of 3–5 repeated experiments. MIP, macrophage inflammatory protein; ND, not done.
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.3 on June 17, 2017
chamber. Thus MIP-3␣ secretion in the IL-1␣-stimulated Caco-2 cultures was 173.0 ⫾ 12.2 ng in the basal
chamber compared with 25.1 ⫾ 2.9 ng in the apical
chamber. This contrasts with 6.9 ⫾ 0.1 ng in the basal
chamber and 0.5 ⫾ 0.2 ng in the apical chamber of
unstimulated cultures. Values are means ⫾ SD from
four or five replicate wells in a single representative
experiment. Similar results were obtained in two repeated experiments.
We next sought to determine whether infection of
intestinal epithelial cells with enteric bacterial pathogens that are known to upregulate the expression of
epithelial cell proinflammatory genes whose products
chemoattract neutrophils and monocytes increases epithelial cell MIP-3␣ secretion. For these studies, HT-29
and T84 cells were infected with S. dublin or enteroinvasive E. coli O29:NM. As shown in Table 1, MIP-3␣
secretion increased as much as 20- to 120-fold in bacteria-infected cultures.
I␬B␣ superrepressor blocks inducible MIP-3␣ production in HT-29 cells. Many of the genes upregulated
in response to bacterial infection of intestinal epithelial
cells, or after stimulation of those cells with TNF-␣ or
IL-1␣, are NF-␬B target genes (13, 20), although the
role, if any, of NF-␬B in the transcriptional regulation
of the CC chemokine MIP-3␣ has not previously been
reported. To determine whether MIP-3␣ in intestinal
epithelial cells functions as an NF-␬B target gene, we
first assessed whether blocking NF-␬B activation with
a proteasome inhibitor, MG-132, decreased MIP-3␣
secretion in response to TNF-␣ or IL-1␣ stimulation.
Treatment of HT-29 cells with 50 ␮M of MG-132 before
stimulation with TNF-␣ decreased MIP-3␣ secretion by
90% compared with cells not pretreated with MG-132
(data not shown). Because pharmacological agents are
not always completely specific, we also used an additional approach to block NF-␬B activation. In this
approach, cells were infected with a recombinant adenovirus expressing a mutant I␬B␣ protein that has
serine-to-alanine substitutions at positions 32 and 36
(Ad5I␬B-A32/36). Ad5I␬B-A32/36 acts as a superrepressor of NF-␬B activation by preventing signal-induced I␬B␣ phosphorylation (13, 20). After infection
with this recombinant adenovirus, HT-29 cells were
stimulated with TNF-␣ or IL-1␣ or infected with S.
dublin or enteroinvasive E. coli. As shown in Table 2,
MIP-3␣ secretion was markedly inhibited in agoniststimulated or bacteria-infected HT-29 cells infected
with recombinant adenovirus expressing the super-
INTESTINAL EPITHELIAL CELLS PRODUCE MIP-3␣
Table 2. MIP-3␣ is a NF-␬B target gene
MIP-3␣ Secretion, ng/ml
Stimulus
⫹Ad5I␬B-A32/36
superrepressor
⫹Ad5 LacZ control
virus
None
IL-1␣
TNF-␣
S. dublin
E. coli O29:NM
0.8
3.4
5.2
7.3
5.4
2.4
23.7
40.7
57.5
30.9
G715
with human IL-1␣. As shown in Fig. 4, MIP-3␣ mRNA
expression increased in the xenografts of IL-1␣-injected mice, although constitutive background levels of
MIP-3␣ varied from xenograft to xenograft. MIP-3␣
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.3 on June 17, 2017
Values are means of replicate wells from a representative experiment. Similar results were obtained in 2 repeated experiments.
HT-29 cells were infected at a multiplicity of infection of 100 with a
recombinant adenovirus (Ad5I␬B-A32/36) expressing a mutant I␬B␣
protein that acts as a superrepressor of nuclear factor-␬B activation
or the same recombinant adenovirus expressing the LacZ gene
(Ad5LacZ). After infection, cells were stimulated for 6 h with tumor
necrosis factor (TNF)-␣ or interleukin (IL)-1␣ or infected with enteroinvasive bacteria as described in MATERIALS AND METHODS.
repressor but not in cells infected with the same recombinant adenovirus expressing the LacZ gene
(Ad5LacZ).
MIP-3␣ expression is upregulated in human intestinal xenografts in response to IL-1␣ stimulation or Salmonella infection. To determine whether regulated
MIP-3␣ expression by the human intestinal epithelial
cell lines was paralleled in vivo, expression of MIP-3␣
was assessed in human intestinal xenografts. As
shown in Fig. 4, we first showed that HT-29 cells,
which respond to cytokine stimulation in vitro, also do
so in vivo when implanted subcutaneously in SCID
mice. We next used human intestinal xenografts implanted subcutaneously in SCID mice and cytokine
treatment to assess regulated MIP-3␣ production by
normal human intestinal epithelium. For these studies, we first assessed MIP-3␣ mRNA expression in
xenografts after intraperitoneal injection of SCID mice
Fig. 4. Expression of MIP-3␣ mRNA in human intestinal xenografts.
Total RNA isolated from intestinal xenografts or HT-29 tumors
implanted subcutaneously in severe combined immunodeficiency
(SCID) mice was amplified by RT-PCR using 30 amplification cycles.
Data in A, B, and C are from three different stimulated or infected
xenografts. Although constitutive expression of MIP-3␣ by the xenografts varied from experiment to experiment, MIP-3␣ expression was
consistently upregulated in the xenografts in response to cytokine
stimulation or bacterial infection.
Fig. 5. Epithelial cell expression of MIP-3␣ in response to IL-1␣ stimulation of human intestinal xenografts. Sections of human intestinal
xenografts from SCID mice injected with human IL-1␣ (A and B) were
immunostained for MIP-3␣ (A) or stained with Hoechst dye 33258 to
reveal nuclear morphology (B). Note that MIP-3␣ immunostaining is
restricted to the intestinal epithelium. C: xenograft section of the same
donor fetal intestine implanted in parallel in a SCID mouse that was
not injected with IL-1␣ before xenograft harvesting. Adjacent sections
stained with isotype-matched control mouse IgG1 revealed no immunostaining (data not shown). Original magnification ⫻400.
G716
INTESTINAL EPITHELIAL CELLS PRODUCE MIP-3␣
mRNA expression was also increased in xenografts
infected with an aroA aroC mutant of S. typhi. We note
that IL-1␣ is not species specific and may have stimulated additional murine mediators that potentially can
act on human cells in the xenografts. Nonetheless, the
approach used herein allows the assessment of early
changes in epithelial cell chemokine gene expression
and regulation in response to a limited set of stimuli in
human intestine in vivo, which is not possible when
studying naturally infected human subjects or individuals with ongoing acute or chronic intestinal mucosal
inflammation.
To show that increased chemokine mRNA expression was accompanied by epithelial protein production,
adjacent segments of the intestinal xenografts were
immunostained for MIP-3␣. As shown in Fig. 5, there
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.3 on June 17, 2017
Fig. 6. CCR6 and CD3 expression in
normal human colon. Adjacent sections
of normal adult human colon were immunostained for CCR6 (A), for CD3 (C),
or with an isotype control antibody for
CCR6 (D). An additional section of normal human colon, not associated with a
lymphoid follicle, was also immunostained for CCR6 (B). As shown in A,
numerous CCR6-staining cells are
present in the T cell-rich marginal
zone, with fewer in the B cell-containing germinal center (GC), of a lymphoid
follicle and in the adjacent lamina propria. The colonic epithelium also
stained positively for CCR6 as shown
in the top and left of A and in B. As
shown in C, CD3 immunostaining is
seen in mononuclear cells in the lamina
propria and T cell-abundant marginal
zone of the lymphoid follicle and only
scattered CD3-positive cells are
present in the GC. Colocalization studies showed that some CCR6-positive
cells in the lamina propria were CD3
positive, whereas others were CD3 negative. The intestinal lumen is oriented
to the top in A, C, and D and on the
right in B. Original magnification
⫻200.
was little MIP-3␣ immunostaining in unstimulated
xenografts but a marked increase in epithelial cell
MIP-3␣ immunostaining that was localized in the intestinal epithelium of the human xenografts from
IL-1␣-injected mice. The relatively low level of epithelial MIP-3␣ immunostaining in unstimulated xenografts
is consistent with the minimal constitutive expression
seen in normal human colon (compare with Fig. 1B).
CCR6, the cognate receptor for MIP-3␣, is expressed
by mononuclear cells and epithelial cells in human
colon mucosa. The normal human intestinal mucosa
contains an abundant population of T cells and DCs.
Because CCR6 is the only currently known cellular
receptor for MIP-3␣, we sought to determine whether
cells in human colon mucosa express CCR6. As shown
in Fig. 6A, cells within the lamina propria and a lym-
INTESTINAL EPITHELIAL CELLS PRODUCE MIP-3␣
DISCUSSION
We report here that the CC chemokine MIP-3␣ is
widely expressed and regulated in human intestinal
epithelium. Moreover, epithelial expression of MIP-3␣
is upregulated by stimulation with the proinflammatory cytokines IL-1␣ and TNF-␣, cytokines that can be
produced by mononuclear cells in the intestinal mucosa during acute inflammation, or in response to bacterial infection. Intestinal epithelial cells activated by
proinflammatory stimuli, infected with enteric pathogens, or stimulated with bacterial products have been
shown to produce signals for the chemoattraction of
neutrophils and monocytes, cells important in host
innate mucosal immunity (22, 29, 32, 37, 39, 42). Because MIP-3␣ is known to chemoattract CCR6-expressing memory T cells and immature DCs (6, 24), our data
extend that current view and suggest that intestinal
epithelial cells also have the capacity to play a role in
orchestrating antigen-specific acquired mucosal immune responses. The pattern of expression of MIP-3␣
throughout the epithelium of human intestine and its
expression in other sites such as human lung and liver
contrasts with its more localized expression in follicleassociated epithelium in mice and suggests that the
functional activity of this chemokine goes beyond a
proposed role in the genesis and function of mucosal
lymphoid follicles (5, 6, 15, 16, 18, 38). In addition, the
upregulated expression of human MIP-3␣ in response
to TNF-␣ or IL-1 stimulation of human intestinal epithelial cells contrasts with the lack of response of
MIP-3␣ production in J774 mouse monocytes to those
stimuli (38), indicating that there may be differences in
the signal transduction pathways important for activating MIP-3␣ in different cell types and/or species.
Our finding of upregulated epithelial cell MIP-3␣ in
cytokine-stimulated or bacteria-infected human intestinal xenografts, and in inflamed human colon, indicates that human intestinal epithelial cells regulate
the expression of a chemokine whose major known
function is to chemoattract cells important for antigen
presentation and the development of the host adaptive
immune response. Thus, in response to inflammatory
stimuli, intestinal epithelial cells may develop the capacity to chemoattract immature DCs and memory T
cells in close proximity to the single layer of intestinal
epithelium that separates the intestinal lumen from
the host’s internal milieu. After capturing antigen,
immature DCs at sites of inflammation undergo a
phenotypic change and migrate to regional lymph
nodes. As part of this process, they downregulate surface expression of several chemokine receptors (i.e.,
CCR1, CCR5, and CCR6), concurrently lose their responsiveness to the chemokine ligands for those receptors (i.e., MIP-1␣, MIP-1␤, RANTES, and MIP-3␣) (33,
34, 36), upregulate the expression of CXCR4, CCR4,
and CCR7, and become responsive to chemokines expressed in blood vessels and secondary lymphoid organs [e.g., SLC, MIP-3␤, stromal cell-derived factor
(SDF-1), macrophage-derived chemokine, thymus- and
activation-regulated chemokine] (5, 33, 34, 36). Consistent with this phenotypic and functional change in
DCs, intestinal epithelial cells do not express the mature DC chemoattractant MIP-3␤. In addition to highlevel MIP-3␣ production after stimulation, it is possible
that constitutive low levels of epithelial expression of
MIP-3␣ may serve to maintain immature DCs and
memory T cells in close proximity to the epithelial
surface, the first site of host contact with antigen in the
intestinal lumen.
MIP-3␣ is shown to function as a NF-␬B target gene
in human intestinal epithelial cells, and its expression
is upregulated by TNF-␣ or IL-1␣ stimulation or infection with enteric bacteria. The same is true of IL-8 and
growth-related protein-␣ (GRO-␣) that, like MIP-3␣,
are upregulated in the intestinal epithelium by IL-1 or
TNF-␣ stimulation or infection with enteric bacterial
pathogens (13, 19, 22, 42). However, MIP-3␣ mainly
signals immature DC and CD45RO⫹ T cells important
in host adaptive immune responses, whereas IL-8 and
GRO-␣ signal neutrophils that can play a role in mucosal innate immune defense. We recently described (8)
the expression of three additional T cell chemoattractants expressed by human intestinal epithelium, the
CXC chemokines IP-10, Mig, and I-TAC, which, in
contrast to MIP-3␣, are known to signal activated/
memory CD4⫹ CD45RO⫹ T cells that express the receptor CXCR3. Unlike MIP-3␣, those chemokines are
preferentially upregulated by the Th1 cytokine IFN-␥
and are only minimally, if at all, regulated by TNF-␣,
IL-1, or bacterial infection in the absence of IFN-␥.
Together, these data suggest that the intestinal epithelium can play a role in regulating both innate and
acquired mucosal immune responses and demonstrate
differential regulation of epithelial cell-produced chemokines that act on different target populations of T
cells.
The human intestinal epithelial cell lines tested in
our studies secreted up to 200–500 ng/ml of MIP-3␣ in
response to IL-1␣ stimulation. It is not currently possible to determine the effective concentrations of
MIP-3␣ in the microenvironment of the intestinal mu-
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.3 on June 17, 2017
phoid follicle, most markedly in the marginal zone
rather than the germinal center, expressed CCR6. In
addition, human colon epithelial cells expressed CCR6
(Fig. 6, A and B), consistent with our prior report (7) of
the constitutive expression of CCR6 mRNA by several
human intestinal epithelial cell lines. Whereas many
CCR6-positive mononuclear cells also stained for the T
cell marker CD3, a number of CCR6-expressing cells in
the lamina propria, some of which were in close proximity to the epithelium, did not stain for CD3 (compare
Fig. 6C with Fig. 6A). This finding is consistent with
the reported expression of CCR6 by both T cells and
immature DCs (5, 24). In contrast to MIP-3␣ mRNA
expression and protein production, CCR6 mRNA expression was not significantly increased in intestinal
epithelial cell lines infected with Salmonella, as we
reported previously (12), and epithelial CCR6 immunostaining was not increased in sections of inflamed
human colon (data not shown).
G717
G718
INTESTINAL EPITHELIAL CELLS PRODUCE MIP-3␣
We thank J. Leopard and D. McCole for helpful assistance and R.
Lara for final preparation of the manuscript.
This work was supported by National Institute of Diabetes and
Digestive and Kidney Diseases (NIDDK) Grant DK-35108. A. Izadpanah is a Howard Hughes Medical Institute Medical Student Research Training Fellow; M. B. Dwinell was supported by a Research
Career Development Award from the Crohn’s and Colitis Foundation
of America (CCFA) and NIDDK Grant K01-DK-02808; and L. Eckmann was supported, in part, by a research grant from the CCFA.
REFERENCES
1. Baba M, Imai T, Nishimura M, Kakizaki M, Takagi S,
Hieshima K, Nomiyama H, and Yoshie O. Identification of
CCR6, the specific receptor for a novel lymphocyte-directed CC
chemokine LARC. J Biol Chem 272: 14893–14898, 1997.
2. Baggiolini M, Dewald B, and Moser B. Human chemokines:
an update. Annu Rev Immunol 15: 675–705, 1997.
3. Butcher EC, Williams M, Youngman K, Rott L, and Briskin
M. Lymphocyte trafficking and regional immunity. Adv Immunol 72: 209–253, 1999.
4. Colotta F, Dower SK, Sims JE, and Mantovani A. The type
II “decoy” receptor: a novel regulatory pathway for interleukin 1.
Immunol Today 15: 562–566, 1994.
5. Dieu MC, Vanbervliet B, Vicari A, Bridon JM, Oldham E,
Ait-Yahia S, Briere F, Zlotnik A, Lebecque S, and Caux C.
Selective recruitment of immature and mature dendritic cells by
distinct chemokines expressed in different anatomic sites. J Exp
Med 188: 373–386, 1998.
6. Dieu-Nosjean MC, Vicari A, Lebecque S, and Caux C. Regulation of dendritic cell trafficking: a process that involves the
participation of selective chemokines. J Leukoc Biol 66: 252–262,
1999.
7. Dwinell MB, Eckmann L, Leopard JD, Varki NM, and
Kagnoff MF. Chemokine receptor expression by human intestinal epithelial cells. Gastroenterology 117: 359–367, 1999.
8. Dwinell MB, Lügering N, Eckmann L, and Kagnoff MF.
Regulated production of interferon-inducible T cell chemoattractants by human intestinal epithelial cells. Gastroenterology 120:
49–59, 2001.
9. Eckmann L, Jung HC, Schurer-Maly C, Panja A, Morzycka-Wroblewska E, and Kagnoff MF. Differential cytokine
expression by human intestinal epithelial cell lines: regulated
expression of interleukin 8. Gastroenterology 105: 1689–1697,
1993.
10. Eckmann L, Kagnoff MF, and Fierer J. Epithelial cells
secrete the chemokine interleukin-8 in response to bacterial
entry. Infect Immun 61: 4569–4574, 1993.
11. Eckmann L, Stenson WF, Savidge TC, Lowe DC, Barrett
KE, Fierer J, Smith JR, and Kagnoff MF. Role of intestinal
epithelial cells in the host secretory response to infection by
invasive bacteria. Bacterial entry induces epithelial prostaglandin H synthase-2 expression and prostaglandin E2 and F2␣
production. J Clin Invest 100: 296–309, 1997.
12. Eckmann L, Smith JR, Housley MP, Dwinell MB, and
Kagnoff MF. Analysis by high-density cDNA arrays of altered
gene expression in human intestinal epithelial cells in response
to infection with the invasive enteric bacteria Salmonella. J Biol
Chem 275: 14084–14094, 2000.
13. Elewaut D, DiDonato JA, Kim JM, Truong F, Eckmann L,
and Kagnoff MF. NF-␬B is a central regulator of the intestinal
epithelial cell innate immune response induced by infection with
enteroinvasive bacteria. J Immunol 163: 1457–1466, 1999.
14. Greaves DR, Wang W, Dairaghi DJ, Dieu MC, Saint-Vis B,
Franz-Bacon K, Rossi D, Caux C, McClanahan T, Gordon
S, Zlotnik A, and Schall TJ. CCR6, a CC chemokine receptor
that interacts with macrophage inflammatory protein 3␣ and is
highly expressed in human dendritic cells. J Exp Med 186:
837–844, 1997.
15. Hieshima K, Imai T, Opdenakker G, Van DJ, Kusuda J, Tei
H, Sakaki Y, Takatsuki K, Miura R, Yoshie O, and
Nomiyama H. Molecular cloning of a novel human CC chemokine liver and activation-regulated chemokine (LARC) expressed
in liver. Chemotactic activity for lymphocytes and gene localization on chromosome 2. J Biol Chem 272: 5846–5853, 1997.
16. Hromas R, Gray PW, Chantry D, Godiska R, Krathwohl M,
Fife K, Bell GI, Takeda J, Aronica S, Gordon M, Cooper S,
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.3 on June 17, 2017
cosa because this depends on several factors, including
the biological half-life of MIP-3␣ in the mucosa, the
extent to which it binds proteoglycans and is available
for binding to CCR6 on target cells, and its effective
diffusion distance from the epithelium. Nevertheless,
the quantity of MIP-3␣ secreted by human intestinal
epithelial cell lines is within the range shown in vitro
to be chemotactic for lymphocytes. Thus 1 ␮g/ml recombinant MIP-3␣ was maximally chemotactic for freshly
isolated human peripheral blood lymphocytes, and significant chemotaxis was observed at concentrations of
100 ng/ml (1).
The present studies report on the regulated expression of MIP-3␣ by human intestinal epithelium. It will
be important for future studies to assess the functional
activity of this chemokine on target cells in the complex
microenvironment of the human intestinal mucosa.
Relevant to MIP-3␣ function in the intestinal tract, a
recent study in mice indicated that a specific subset of
mucosal DC within the subepithelial region of Peyer’s
patches express CCR6 and migrate in response to
MIP-3␣ (18). However, we also note that CCR6, the
cognate receptor for MIP-3␣, was abundantly expressed by epithelial cells in normal human colon. This
is consistent with our prior report (7) that several
human intestinal epithelial cell lines express CCR6
mRNA as well as functional receptors for several other
CC and CXC chemokines, most notably CXCR4, which
binds the CXC chemokine SDF-1, and CCR5, which
binds the CC chemokines MIP-1␣, MIP-1␤, and
RANTES. Like CXCR4 and CCR5, CCR6 on intestinal
epithelial cells acts as a functional signaling receptor
after interaction with its ligand MIP-3␣ (unpublished
data). Thus, in addition to paracrine effects on DCs and
T cells, MIP-3␣, like several other mediators produced
by intestinal epithelial cells, may mediate autocrine
effects on the intestinal epithelium. In contrast to
MIP-3␣, CCR6 was not upregulated by intestinal epithelial cells in inflamed colon.
Finally, we note that CCR6 has been reported to act
as a functional receptor for a second epithelial cellproduced ligand, the antimicrobial peptide human ␤
defensin-2 (hBD-2) (41). In addition, we have reported
(27) the upregulated expression of hBD-2 by human
intestinal epithelial cells in vitro and in vivo in response to proinflammatory cytokine stimulation or
bacterial infection. However, we found (present study
and Ref. 27) that MIP-3␣ secretion by agonist-stimulated or bacteria-infected human intestinal epithelial
cell lines was ⬃100-fold greater than that of hBD-2 in
response to identical proinflammatory stimuli. Coupled with a lower affinity of hBD-2 binding to CCR6
compared with MIP-3␣ (41), hBD-2 produced by intestinal epithelium may not compete effectively with
MIP-3␣ for binding to CCR6 on immature DC or mucosal T cells or to CCR6 expressed by intestinal epithelium.
INTESTINAL EPITHELIAL CELLS PRODUCE MIP-3␣
17.
18.
19.
20.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
gene expression in intestinal epithelial cells and inflammatory
bowel disease mucosal. Gastroenterology 108: 40–50, 1995.
Rollins BJ. Chemokines. Blood 90: 909–928, 1997.
Rossi DL, Vicari AP, Franz-Bacon K, McClanahan TK, and
Zlotnik A. Identification through bioinformatics of two new
macrophage proinflammatory human chemokines: MIP-3␣ and
MIP-3␤. J Immunol 158: 1033–1036, 1997.
Savkovic DS, Koutsouris A, and Hecht G. Activation of
NF-␬B in intestinal epithelial cells by enteropathogenic Escherichia coli. Am J Physiol Cell Physiol 273: C1160–C1170, 1997.
Sallusto F and Lanzavecchia A. Mobilizing dendritic cells for
tolerance, priming, and chronic inflammation. J Exp Med 189:
611–614, 1999.
Sallusto F, Schaerli P, Loetscher P, Schaniel C, Lenig D,
Mackay CR, Qin S, and Lanzavecchia A. Rapid and coordinated switch in chemokine receptor expression during dendritic
cell maturation. Eur J Immunol 28: 2760–2769, 1998.
Savidge TC, Morey AL, Ferguson DJ, Fleming KA, Shmakov AN, and Phillips AD. Human intestinal development in a
severe-combined immunodeficient xenograft model. Differentiation 58: 361–371, 1995.
Sozzani S, Allavena P, D’Amico G, Luini W, Bianchi G,
Kataura M, Imai T, Yoshie O, Bonecchi R, and Mantovani
A. Differential regulation of chemokine receptors during dendritic cell maturation: a model for their trafficking properties.
J Immunol 161: 1083–1086, 1998.
Steiner TS, Nataro JP, Poteet-Smith CE, Smith JA, and
Guerrant RL. Enteroaggregative Escherichia coli expresses a
novel flagellin that causes IL-8 release from intestinal epithelial
cells. J Clin Invest 105: 1769–1777, 2000.
Tanaka Y, Imai T, Baba M, Ishikawa I, Uehira M,
Nomiyama H, and Yoshie O. Selective expression of liver and
activation-regulated chemokine (LARC) in intestinal epithelium
in mice and humans. Eur J Immunol 29: 633–642, 1999.
Thorpe CM, Hurley BP, Lincicome LL, Jacewicz MS,
Keusch GT, and Acheson DW. Shiga toxins stimulate secretion of interleukin-8 from intestinal epithelial cells. Infect Immun 67: 5985–5993, 1999.
Willimann K, Legler DF, Loetscher M, Roos RS, Delgado
MB, Clark-Lewis I, Baggiolini M, and Moser B. The chemokine SLC is expressed in T cell areas of lymph nodes and mucosal
lymphoid tissues and attracts activated T cells via CCR7. Eur
J Immunol 28: 2025–2034, 1998.
Yang D, Chertov O, Bykovskaia SN, Chen Q, Buffo MJ,
Shogan J, Anderson M, Schroder JM, Wang JM, Howard
OM, and Oppenheim JJ. ␤-defensins: linking innate and adaptive immunity through dendritic and T cell CCR6. Science 286:
525–528, 1999.
Yang SK, Eckmann L, Panja A, and Kagnoff MF. Differential and regulated expression of C-X-C, C-C, and C-chemokines
by human colon epithelial cells. Gastroenterology 113: 1214–
1223, 1997.
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.3 on June 17, 2017
21.
Broxmeyer HE, and Klemsz MJ. Cloning and characterization of exodus, a novel ␤-chemokine. Blood 89: 3315–3322, 1997.
Huang GT, Eckmann L, Savidge TC, and Kagnoff MF.
Infection of human intestinal epithelial cells with invasive bacteria upregulates apical intercellular adhesion molecule-1
(ICAM-1) expression and neutrophil adhesion. J Clin Invest 98:
572–583, 1996.
Iwasaki A and Kelsall BL. Localization of distinct Peyer’s
patch dendritic cell subsets and their recruitment by chemokines
macrophage inflammatory protein (MIP)-3␣, MIP-3␤, and secondary lymphoid organ chemokine. J Exp Med 191: 1381–1393,
2000.
Jobin C, Holt L, Bradham CA, Streetz K, Brenner DA, and
Sartor RB. TNF receptor-associated factor-2 is involved in both
IL-1␤ and TNF-␣ signaling cascades leading to NF-␬B activation
and IL-8 expression in human intestinal epithelial cells. J Immunol 162: 4447–4454, 1999.
Jobin C, Panja A, Hellerbrand C, Iimuro Y, DiDonato J,
Brenner DA, and Sartor RB. Inhibition of proinflammatory
molecule production by adenovirus-mediated expression of a
nuclear factor ␬B super-repressor in human intestinal epithelial
cells. J Immunol 160: 410–418, 1998.
Jung HC, Eckmann L, Yang SK, Panja A, Fierer J, Morzycka-Wroblewska E, and Kagnoff MF. A distinct array of
proinflammatory cytokines is expressed in human colon epithelial cells in response to bacterial invasion. J Clin Invest 95:
55–65, 1995.
Kagnoff MF and Eckmann L. Epithelial cells as sensors for
microbial infection. J Clin Invest 100: 6–10, 1997.
Laurent F, Eckmann L, Savidge TC, Morgan G, Theodos C,
Naciri M, and Kagnoff MF. Cryptosporidium parvum infection
of human intestinal epithelial cells induces the polarized secretion of C-X-C chemokines. Infect Immun 65: 5067–5073, 1997.
Liao F, Rabin RL, Smith CS, Sharma G, Nutman TB, and
Farber JM. CC-chemokine receptor 6 is expressed on diverse
memory subsets of T cells and determines responsiveness to
macrophage inflammatory protein 3␣. J Immunol 162: 186–194,
1999.
Locati M and Murphy PM. Chemokines and chemokine receptors: biology and clinical relevance in inflammation and AIDS.
Annu Rev Med 50: 425–440, 1999.
Mantovani A. The chemokine system: redundancy for robust
outputs. Immunol Today 20: 254–257, 1999.
O’Neil DA, Porter EM, Elewaut D, Anderson GM, Eckmann L, Ganz T, and Kagnoff MF. Expression and regulation
of the human ␤-defensins hBD-1 and hBD-2 in intestinal epithelium. J Immunol 163: 6718–6724, 1999.
Power CA, Church DJ, Meyer A, Alouani S, Proudfoot AE,
Clark-Lewis I, Sozzani S, Mantovani A, and Wells TN.
Cloning and characterization of a specific receptor for the novel
CC chemokine MIP-3␣ from lung dendritic cells. J Exp Med 186:
825–835, 1997.
Reinecker HC, Loh EY, Ringler DJ, Mehta A, Rombeau JL,
and MacDermott RP. Monocyte-chemoattractant protein 1
G719