Am J Physiol Gastrointest Liver Physiol
280: G1217–G1226, 2001.
Production of MDC/CCL22 by human
intestinal epithelial cells
M. CECILIA BERIN, MICHAEL B. DWINELL, LARS ECKMANN, AND MARTIN F. KAGNOFF
Laboratory of Mucosal Immunology, Department of Medicine, University of California
at San Diego, La Jolla, California 92093
Received 13 September 2000; accepted in final form 26 January 2001
mucosa; cytokines; Th1/Th2; chemotaxis
forms a barrier that restricts
entry of intestinal luminal contents and microbes into
the host. Nonetheless, the lamina propria underlying
the surface epithelium is “physiologically inflamed” in
that it contains abundant numbers of B and T cells,
monocytes, and dendritic cells, as well as variable
numbers of eosinophils and mast cells. T cells that
encounter luminally derived antigens within the organized lymphoid tissue of the Peyer’s patches can exit
from those structures, recirculate, and eventually
home to the intestinal lamina propria. T cells in the
lamina propria are mostly CD45RO⫹, consistent with a
memory phenotype, and also express the mucosal homing ␣4␤7-integrin as well as human leukocyte antigen-DR and CD25, reflecting an activated state compared with T cells in other peripheral sites (27).
THE INTESTINAL EPITHELIUM
Address for reprint requests and other correspondence: M. F.
Kagnoff, Univ. of California, San Diego, Dept. of Medicine (0623D),
9500 Gilman Drive, La Jolla, California 92093-0623.
http://www.ajpgi.org
Lamina propria T cells play an important role in host
defense but can also contribute to the pathophysiology
of intestinal inflammation. The cytokine profile of lamina propria lymphocytes is predominantly Th1 in nature [i.e., interferon (IFN)-␥ producing] under physiological conditions (21) as well as in Crohn’s disease (18)
and most murine models of intestinal inflammation (5).
Nonetheless, interleukin (IL)-4-producing Th2 cells are
also present in the normal human intestinal mucosa
and are increased in certain murine models of colitis,
such as the TCR ␣/␣ mutant mouse (25, 35) and the
oxazolone-induced colitis model (6). Despite the association of those models of colitis with elevated Th2
cytokines, most evidence supports the hypothesis that
Th2 cytokines, primarily IL-4 and IL-10, are antiinflammatory in the intestinal mucosa through their
actions on macrophages and Th1 cells (3, 23).
Little is known about the signals responsible for the
recruitment of T cells within the intestinal mucosa.
Although an initial interaction between ␣4␤7-integrin
on T cells and mucosal addressin cell adhesion molecule on venules is required for homing of T cells to the
gut, additional activation signals provided by chemokines are also thought to be necessary (8, 31). T cells
express a number of cell surface chemokine receptors,
and different subsets of T cells express different chemokine receptors, allowing for differential recruitment
of those subsets. For example, Th2-type cells (IL-4,
IL-5, IL-10, and IL-13 producing) express the receptor
CCR4 on their surface (11, 37), whereas Th1-type cells
(IFN-␥ and IL-2 producing) express the receptor
CXCR3.
Intestinal epithelial cells are an important source of
chemokines in the intestinal mucosa. Epithelial cells
are the first line of defense against luminal pathogens
and, in response to bacterial invasion, can secrete an
array of neutrophil and macrophage chemoattractants
(e.g., IL-8/CXCL8, ENA-78/CXCL5, GRO␣/CXCL1,
MCP-1/CCL2, and MIP-1␣/CCL3) (14, 42). Human intestinal epithelium can also produce IP-10/CXCL10,
Mig/CXCL9, and I-TAC/CXCL11 (13, 40), chemokines
that are known to chemoattract CXCR3-expressing T
cells that have a Th1 phenotype (i.e., IFN-␥-producing
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.
0193-1857/01 $5.00 Copyright © 2001 the American Physiological Society
G1217
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 16, 2017
Berin, M. Cecilia, Michael B. Dwinell, Lars Eckmann, and Martin F. Kagnoff. Production of MDC/CCL22
by human intestinal epithelial cells. Am J Physiol Gastrointest Liver Physiol 280: G1217–G1226, 2001.—The intestinal
mucosa contains a subset of lymphocytes that produce Th2
cytokines, yet the signals responsible for the recruitment of
these cells are poorly understood. Macrophage-derived chemokine (MDC/CCL22) is a recently described CC chemokine
known to chemoattract the Th2 cytokine producing cells that
express the receptor CCR4. The studies herein demonstrate
the constitutive production of MDC/CCL22 in vivo by human
colon epithelium and by epithelium of human intestinal
xenografts. MDC/CCL22 mRNA expression and protein secretion was upregulated in colon epithelial cell lines in response to proinflammatory cytokines or infection with enteroinvasive bacteria. Inhibition of nuclear factor (NF)-␬B
activation abolished MDC/CCL22 expression in response to
proinflammatory stimuli, demonstrating that MDC/CCL22 is
a NF-␬B target gene. In addition, tumor necrosis factor-␣induced MDC/CCL22 secretion was differentially modulated
by Th1 and Th2 cytokines. Supernatants from the basal, but
not apical, side of polarized epithelial cells induced a MDC/
CCL22-dependent chemotaxis of CCR4-positive T cells.
These studies demonstrate the constitutive and regulated
production by intestinal epithelial cells of a chemokine
known to function in the trafficking of T cells that produce
anti-inflammatory cytokines.
G1218
REGULATED MDC/CCL22 PRODUCTION BY HUMAN INTESTINAL EPITHELIUM
MATERIALS AND METHODS
Reagents. Recombinant human (rh) tumor necrosis factor
(TNF)-␣, IL-1␣, MDC/CCL22, and transforming growth factor-␤1 were from R&D Systems (Minneapolis, MN). rhIFN-␥
was from Biosource International (Camarillo, CA), and
rhIL-4 and IL-13 were from PeproTech (Rocky Hill, NJ).
Mouse anti-human MDC/CCL22 monoclonal antibody (MAb)
(IgG2b) and affinity-purified goat anti-human TNF-␣ were
from R&D Systems, and rabbit anti-human MDC/CCL22 IgG
was from PeproTech. MG-132 was from Sigma Chemical (St.
Louis, MO).
Cell culture. The human colon adenocarcinoma cell lines
HT-29 (ATCC HTB 38, Rockville, MD) and HCA-7 colony 29
(a gift from S. C. Kirkland) were grown in DMEM supplemented with 10% heat-inactivated FCS and 2 mM L-glutamine. Epithelial cells were grown to confluence in six-well
plates before stimulation with cytokines or bacterial infection. To obtain polarized monolayers, HCA-7 cells were
seeded onto tissue culture-treated transwell filters (0.4 ␮m
pore size, 1.2 cm2 surface area; Costar, Cambridge, MA) and
allowed to grow for ⬃7 days, at which time a mean resistance
of ⬃500 ⍀ 䡠 cm2 was established. Peripheral blood monocytes
were prepared by separation of whole blood on a Ficoll gradient, followed by adherence of the harvested buffy coat to
tissue culture plastic wells overnight. Monocyte-derived macrophages were prepared as previously described (32).
Bacterial infection. Enteroinvasive Escherichia coli strain
O29:NM was obtained from the ATCC (43892), and Salmo-
nella dublin was provided by Dr. J. Fierer (University of
California, San Diego). For infection, HT-29 cells were grown
to confluence in six-well plates and incubated with 108 S.
dublin or 5 ⫻ 108 enteroinvasive E. coli for 1 h to allow
invasion to occur, after which the extracellular bacteria were
removed by washing and cells were incubated for an additional 24 h in the presence of 50 ␮g/ml of the non-membranepermeant antibiotic gentamicin to kill remaining extracellular bacteria (15). Culture supernatants were removed 24 h
after infection and stored at ⫺20°C before measurement of
MDC/CCL22 by ELISA.
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 before (17). Ad5I␬BA32/36 expresses a hemagglutinin (HA) epitope-tagged 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 superrepressor of nuclear factor (NF)-␬B activation (17). The HA
epitope tag enables identification of the exogenous superrepressor with anti-HA 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 multiplicity of infection (MOI) of 100
for 16 h as described before (17). 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
EMSA, and inhibited cytokine-induced and bacteria-induced
upregulation of IL-8 and intracellular adhesion molecule-1
expression in HT-29 cells but did not alter ␤-actin mRNA
levels in the same cells (17).
RT-PCR. Total cellular RNA was extracted with TRIzol
reagent (GIBCO BRL, Gaithersburg, MD), treated with
RNase-free DNase (Stratagene, La Jolla, CA) to remove contaminating genomic DNA, and reverse transcribed using 1
␮g of total RNA and 2.5 units Superscript II RT (GIBCO
BRL). Sequences were amplified from cDNA with primers
specific for human MDC/CCL22 and ␤-actin. Primer sequences for MDC/CCL22 were 5⬘-GAGACATACAGGACAGAGCATGGC T-3⬘ (sense) and 5⬘-ATGGAGATCAGGGAATGCAGAGAG-3⬘ (antisense), and those for ␤-actin were 5⬘TGACGGGGTCACCCACACTGTGCCCATCTA-3⬘ (sense)
and 5⬘-CTAGAAGCATTGCGGTGGACGATGGAGGG-3⬘ (antisense). Five microliters of RT reaction were used for PCR
amplification in a volume of 50 ␮l of PCR buffer (GIBCO
BRL), including 25 pM of each primer, 1.5 mM MgCl2, and
200 ␮M of each dNTP. After an initial hot start at 95°C for 5
min, 4.0 units Taq polymerase (GIBCO BRL) was added. The
amplification profile consisted of 45 s denaturation at 95°C,
2.5 min annealing, and extension at 64°C (MDC/CCL22) or
72°C (␤-actin) for 38 (MDC/CCL22) or 28 (␤-actin) cycles.
Each experiment included negative controls in which RNA
was omitted from the RT mixture and cDNA was omitted
from the PCR reactions. RNA from human monocyte-derived
macrophages was used as a positive control for MDC/CCL22.
Real-time PCR. Real-time PCR was performed using an
ABI Prism 7700 sequence detection system (PE Applied Bio-
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 16, 2017
CD4⫹ T cells) and MIP-3␣/CCL20, a chemoattractant
for CD45RO⫹ T cells and immature dendritic cells (26).
These findings suggest that epithelial cells may have a
role not only in the acute host response to infection but
also in T cell trafficking in the intestine.
The CC chemokine receptor CCR4 is preferentially
expressed by T lymphocytes that produce Th2 cytokines (11, 37). The production of a chemokine specific
for CCR4 by the intestinal epithelium could provide a
signal for the specific recruitment of Th2 cells. Macrophage-derived chemokine (MDC/CCL22) is a recently
identified chemokine of the CC chemokine family that
uses CCR4 as its receptor and is known to selectively
chemoattract Th2 cytokine-producing cells (2, 20).
MDC/CCL22 is constitutively expressed by macrophages, mature dendritic cells, and B cells, and upregulated expression of MDC/CCL22 has been noted in
T lymphocytes that concomitantly produce the Th2
cytokines IL-4, IL-5, and IL-6 and in monocytes stimulated with the Th2 cytokines IL-4 and IL-13 (2, 19).
We hypothesized that human intestinal epithelial
cells may have the capacity to signal Th2 cells through
the production of the CCR4 ligand MDC/CCL22. In the
studies herein, we report on the constitutive expression of MDC/CCL22 by epithelium in normal human
colon and in human intestinal xenografts in vivo and
the regulated expression of MDC/CCL22 by cultured
human intestinal epithelial cells. These data suggest
the notion that intestinal epithelial cells have the capacity to regulate mucosal T cell trafficking through
the release of T cell chemoattractants and the potential
to modulate the local mucosal cytokine milieu through
the recruitment of specific cytokine-producing T cell
subsets.
REGULATED MDC/CCL22 PRODUCTION BY HUMAN INTESTINAL EPITHELIUM
Chemotaxis assay. To determine whether epithelial cellderived MDC/CCL22 chemoattracts T lymphocytes, chemotaxis assays were performed using the CEM T cell line, which
expresses CCR4 as we verified by RT-PCR using previously
described primers (12). Supernatants were obtained from the
apical or basolateral chambers of polarized HCA-7 cells cultured in Transwells in RPMI-0.1% BSA 24 h after stimulation with TNF-␣ or IL-1␣ plus IFN-␥. Recombinant human
MDC/CCL22 or HCA-7 supernatants were added to the bottom well of ChemoTx chemotaxis microplates (Neuroprobe,
Gaithersburg, MD), and a 5-␮m-pore filter plate was placed
over the bottom wells. To assess chemotaxis, 1.25 ⫻ 105 CEM
cells labeled with the fluorescent dye calcein-AM (Molecular
Probes, Eugene, OR) in a total volume of 25 ␮l were added to
the top of the filter. After 2 h incubation, fluorescence in the
bottom wells was measured with a fluorescence microplate
reader and number of migrated cells were calculated from a
standard curve of labeled cells. To determine the role of
MDC/CCL22 in supernatant-induced chemotaxis, 5 ␮g/ml of
a neutralizing chicken IgY anti-MDC/CCL22 antibody (R&D
Systems) or an equivalent concentration of control IgY (Jackson ImmunoChemical Laboratories, West Grove, PA) were
added to the supernatants before addition to the chemotaxis
plate.
All studies involving human and animal protocols were
approved by the UCSD Human Subjects Committee and the
Animal Subjects Program.
RESULTS
Constitutive MDC/CCL22 expression by epithelial
cells in human colon and human intestinal xenografts.
To determine if human intestinal epithelium in vivo
expresses MDC/CCL22, we first immunostained sections of normal human colon for that chemokine. As
shown in Fig. 1A, MDC/CCL22 was constitutively expressed by epithelial cells in normal human colon.
Sections stained with an isotype-matched control antibody at the same concentration (data not shown) or
anti-MDC/CCL22 antibody preabsorbed with rhMDC/
CCL22 (Fig. 1B) did not demonstrate any specific immunoreactivity. Surface epithelium as well as crypt
epithelium stained with the anti-MDC/CCL22 antibody.
To determine if MDC/CCL22 production in vivo depends on the exposure of the intestinal epithelium to a
conventional luminal bacterial flora, we assessed
MDC/CCL22 expression by the epithelium of human
intestinal xenografts grown subcutaneously in SCID
mice, since the xenograft lumen is sterile. As shown in
Fig. 2, like normal human colon epithelium in vivo,
epithelium in the xenografts produced MDC/CCL22,
indicating that the normal colon flora was not required
for constitutive epithelial cell MDC/CCL22 expression.
Constitutive and regulated MDC/CCL22 mRNA expression in human intestinal epithelial cells. To determine if the expression of MDC/CCL22 by intestinal
epithelial cells is regulated, we took advantage of cultured human colon epithelial cell lines. As shown in
Fig. 3A, MDC/CCL22 mRNA expression was upregulated in the human colon epithelial cell line (HT-29) by
3 h and continued to increase up to 12 h after stimulation with TNF-␣ or IL-1␣. MDC/CCL22 mRNA expression also increased after IFN-␥ stimulation, but
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 16, 2017
systems, Foster City, CA). Each reaction contained 25 ␮l of
2⫻ SYBR Green Master Mix (containing 200 nM dATP,
dGTP, and dCTP, 400 nM dUTP, 2 mM MgCl2, 0. 25 units
uracil N-glycosylase, and 0. 625 units Amplitaq Gold DNA
polymerase), 25 pmol of the sense and antisense primers, and
2 ␮l of cDNA in a final volume of 50 ␮l. MDC/CCL22 primers
used were those described above, and ␤-actin primers were as
follows: sense, 5⬘-CAAAGACCTGTACGCCAACAC-3⬘; antisense, 5⬘-CATACTCCTGCTTGCTGATCC-3⬘. Reactions were
incubated at 50°C for 2 min followed by 95°C for 10 min. The
amplification profile was 95°C for 15 s and 62°C for 1 min for
40 cycles. Amplification of the expected single PCR product
was confirmed by electrophoresis of the product on a 1%
agarose gel that was stained with ethidium bromide. Data
analysis was performed using PE Biosystems 7700 sequence
detection system software. Real-time PCR data were plotted
as the ⌬Rn fluorescence signal vs. the cycle number. ⌬Rn was
calculated using the equation ⌬Rn ⫽ Rn⫹ ⫺ Rn⫺, where Rn⫹ is
the fluorescence signal of the product and Rn⫺ is the fluorescence signal of the baseline emission. Ct is the PCR cycle at
which an increase in reporter fluorescence above the baseline
signal can first be detected, as determined by the sequence
detection system software. The fold increase in MDC/CCL22
mRNA concentration was calculated as 2 to the power of the
difference between Ct values for two samples.
MDC/CCL22 ELISA. Polystyrene 96-well plates (Immulon-4; Dynex Technologies, Chantilly, VA) were coated with
mouse anti-human MDC/CCL22 MAb (R&D Systems), diluted in carbonate buffer, as the capture antibody. Affinitypurified rabbit polyclonal anti-human MDC/CCL22 (PeproTech) was diluted in PBS, 1.0% BSA, and 0.1% Tween 20 and
used as the detection antibody. The second step reagent was
horseradish peroxidase-conjugated donkey anti-rabbit IgG
(Amersham Life Science, Arlington Heights, IL). Bound
horseradish peroxidase was visualized with tetramethylbenzidine 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. MDC/CCL22 concentrations were calculated from standard curves using rhMDC/
CCL22 (R&D Systems). The MDC/CCL22 ELISA was sensitive to 50 pg/ml.
Human intestinal xenografts. Human fetal intestine, gestational age 14–20 wk (Advanced Biosciences Resources,
Alameda, CA), was transplanted subcutaneously onto the
backs of C57BL/6 SCID mice as described before (16, 24).
Human fetal intestinal xenografts were allowed to develop
for at least 10 wk following transplantation, at which time
the epithelium, which is strictly of human origin, is fully
differentiated (38). Intestinal xenografts were removed, and
segments were embedded in optimum cutting temperature
(OCT) compound and frozen in isopentane/dry ice for immunohistochemical analysis.
Immunohistochemical detection of MDC/CCL22 in human
colon and human intestinal xenografts. Sections of histologically normal human colon from surgical specimens from
individuals undergoing partial colectomy were imbedded in
OCT and snap frozen in isopentane/dry ice. Cryostat sections
(5 ␮m) of human colon or intestinal xenografts were fixed in
acetone for 10 min and blocked in PBS-1% BSA with 10%
goat serum for 1 h at room temperature. Sections were
incubated overnight at 4°C with 2.5 ␮g/ml mouse anti-human
MDC/CCL22 MAb, an isotype-matched control MAb (mouse
IgG2b, 2.5 ␮g/ml, Sigma), or anti-MDC/CCL22 antibody that
was preabsorbed with rhMDC/CCL22 (2.5 ␮g/ml), and sections were subsequently stained with Cy3-labeled goat antimouse IgG. Sections were counterstained with the nuclear
dye Hoechst 33258 (Calbiochem, San Diego, CA).
G1219
G1220
REGULATED MDC/CCL22 PRODUCTION BY HUMAN INTESTINAL EPITHELIUM
increased expression was delayed for up to 12 h after
IFN-␥ stimulation. Furthermore, costimulation of
HT-29 cells with a combination of IFN-␥ plus TNF-␣ or
IFN-␥ plus IL-1␣ resulted in a synergistic increase
Fig. 2. Immunohistochemical detection of MDC/CCL22 in human
intestinal xenografts. Sections of human intestinal xenografts were
incubated with MDC/CCL22 MAb (A) or an isotype-matched control
MAb at the same concentration (B). Original magnification, ⫻400.
MDC/CCL22 mRNA expression (Fig. 3). Similar data
were obtained using HCA-7 cells (data not shown).
Real-time PCR was performed to quantify the increase
in MDC/CCL22 mRNA expression after cytokine stimulation. These data confirmed that costimulation of
HT-29 cells with TNF-␣ or IL-1␣ plus IFN-␥ induced a
synergistic increase in MDC/CCL22 mRNA expression
(Fig. 3, B and C). Thus IFN-␥ and TNF-␣ costimulation
induced a 123-fold increase in MDC/CCL22 expression
over unstimulated controls, compared with a 35-fold
increase after TNF-␣ stimulation and an 8-fold increase after IFN-␥ stimulation.
Cytokine-stimulated MDC/CCL22 secretion. To determine if agonist-stimulated MDC/CCL22 mRNA expression was accompanied by increased MDC/CCL22
protein secretion, HT-29 cells were left unstimulated
or were stimulated for up to 24 h with TNF-␣, IL-1␣, or
IFN-␥, after which MDC/CCL22 levels in culture supernatants were assayed by ELISA. As shown in Fig.
4A, TNF-␣ stimulation markedly increased MDC/
CCL22 secretion (130-fold increase over unstimulated
cells at 24 h), and this was also the case, although to a
lesser extent, for IL-1␣ stimulation (36-fold increase).
In contrast, MDC/CCL22 secretion in response to
IFN-␥ stimulation was significantly less than that induced by TNF-␣ or IL-1␣ (10-fold increase over control
at 24 h).
As shown in Fig. 4B, IFN-␥ synergistically increased
TNF-␣-stimulated MDC/CCL22 secretion. IFN-␥ induced a 10- to 15-fold increase in TNF-␣-induced MDC/
CCL22 secretion at TNF-␣ concentrations ranging
from 0. 1 to 10 ng/ml and a lesser 3-fold increase at 100
ng/ml, the highest concentration of TNF-␣ tested. Synergy between IFN-␥ and TNF-␣ or IL-1␣ was not ob-
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 16, 2017
Fig. 1. Immunohistochemical detection of macrophage-derived chemokine
MDC/CCL22 in normal human colon.
A: normal human colon immunostained with a MDC/CCL22 monoclonal antibody (MAb). B: adjacent section of normal human colon immunostained with MDC/CCL22 MAb that
was preabsorbed with recombinant human MDC/CCL22. Preabsorption of
anti-MDC/CCL22 with recombinant
human MDC/CCL22 removed all specific epithelial cell staining. C and D:
nuclear staining of sections shown in A
and B, respectively. Original magnification, ⫻400.
REGULATED MDC/CCL22 PRODUCTION BY HUMAN INTESTINAL EPITHELIUM
G1221
served 3 h after stimulation but was present at each
time point examined thereafter (8, 12, 16, and 24 h)
(data not shown).
MDC/CCL22 secretion by peripheral blood monocytes was determined to compare levels of MDC/CCL22
production between epithelial cells and monocytes,
which had previously been described to produce MDC/
CCL22 (2). Monocytes cultured at the same cell density
as intestinal epithelial cells constitutively produced
slightly higher levels of MDC/CCL22 than intestinal
epithelial cells (1.5 ng/ml for monocytes compared with
0.5 ng/ml for intestinal epithelial cells). However, maximal MDC/CCL22 production by monocytes, which was
observed after TNF-␣ plus IL-4 stimulation, increased
by only fourfold over constitutive levels (up to 5.6
ng/ml), whereas HT-29 cells stimulated with TNF-␣
plus IFN-␥ produced ⬎300 ng/ml MDC/CCL22.
In contrast to MDC/CCL22, the CCR4 ligand, TARC/
CCL17, which has been shown by us (4) and others (39)
to be expressed by airway epithelial cells, was not
detected in supernatants from intestinal epithelial
cells after stimulation with proinflammatory cytokines
(TNF-␣, IL-1␣) with or without costimulation with
IFN-␥, IL-4, or IL-13 (data not shown).
Polarized secretion of MDC/CCL22. Intestinal epithelial cells in vivo are polarized with functionally
distinct apical and basolateral domains. For epithelial
MDC/CCL22 to act as a physiological chemoattractant
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 16, 2017
Fig. 3. Regulation of MDC/CCL22 expression in human
intestinal epithelial cell lines. A: HT-29 cells were stimulated with cytokines for 3 or 12 h before isolation of
total RNA and analysis of mRNA expression for MDC/
CCL22 and ␤-actin by RT-PCR using 38 and 28 cycles of
amplification, respectively. Cells were incubated with
media alone (None), or the indicated cytokines at a
concentration of 20 ng/ml. B and C: MDC/CCL22
mRNA expression was assessed by real-time PCR after
stimulation with the indicated cytokines for 3 h (left) or
12 h (right). Results shown are mean values of triplicates. TNF, tumor necrosis factor; IL, interleukin; IFN,
interferon.
G1222
REGULATED MDC/CCL22 PRODUCTION BY HUMAN INTESTINAL EPITHELIUM
Table 1. Polarized MDC/CCL22 secretion
by HCA-7 cells
MDC/CCL22 Secretion, ng/well
Stimulus
Apical
Basolateral
Total
None
TNF-␣
IL-1␣
IFN-␥
TNF-␣ ⫹ IFN-␥
IL-1␣ ⫹ IFN-␥
⬍0.05
0.06 ⫾ 0.01 (10)
0.08 ⫾ 0.02 (11)
0.20 ⫾ 0.02 (9)
0.58 ⫾ 0.03 (6)
0.49 ⫾ 0.02 (7)
0.26 ⫾ 0.04
0.55 ⫾ 0.01 (90)
0.68 ⫾ 0.04 (89)
2.03 ⫾ 0.17 (91)
8.64 ⫾ 0.64 (94)
6.30 ⫾ 0.51 (93)
⬍0.31
0.61
0.76
2.23
9.22
6.79
basal chamber of Transwell-grown HCA-7 cells stimulated with IL-1␣ and IFN-␥ induced an increase in T
cell chemotaxis compared with media from unstimulated monolayers (18.6 vs. 8.3% of input cells migrated). This was comparable to the level of chemotaxis
induced by similar concentrations of rhMDC/CCL22.
In contrast, supernatants from the apical chamber of
filter-grown, stimulated HCA-7 cells did not induce any
increase in chemotaxis (7.7% of input cells migrated).
IL-1 and IFN-␥ at a concentration of 20 ng/ml in RPMI
did not affect chemotaxis of CEM cells (7.3% of input
cells migrated). Incubation of the basal chamber super-
Fig. 4. MDC/CCL22 Secretion by HT-29 Cells. A: time course of
MDC/CCL22 secretion. HT-29 cells were grown to confluence and left
unstimulated (control) or were stimulated with TNF-␣, IL-1␣, or
IFN-␥, each at 20 ng/ml. MDC/CCL22 in culture supernatants was
determined by ELISA. B: effect of IFN-␥ on TNF-␣-stimulated MDC/
CCL22 secretion. HT-29 cells were left unstimulated or were stimulated for 24 h with varying concentrations of TNF-␣ in the presence
or absence of 20 ng/ml IFN-␥. Values are means ⫾ SE of 3 or more
independent experiments.
for target cells in the lamina propria, it predictably
would be secreted in a polarized manner from the
basolateral epithelial cell membrane rather than apically into the intestinal lumen. To determine if this
was the case, HCA-7 cells were grown as polarized
monolayers and stimulated with TNF-␣ or IL-1␣ either
alone or in combination with IFN-␥. As shown in Table
1, for each stimulus, ⬃90% of MDC/CCL22 was secreted into the basal compartment.
Epithelial cell-derived MDC/CCL22 induces T cell
chemotaxis. To determine if MDC/CCL22 secreted by
agonist-stimulated epithelial cells could contribute to
the recruitment of T cells, chemotaxis assays were
performed using supernatants obtained from polarized
HCA-7 cells (Fig. 5). rhMDC/CCL22 induced a dosedependent increase in chemotaxis of the CCR4-expressing T cell line CEM. Supernatants taken from the
Fig. 5. Chemotaxis of CEM T cells in response to MDC/CCL22
secreted by intestinal epithelium. The bottom wells of a chemotaxis
chamber were loaded with recombinant human MDC (rhMDC) or
apical or basal supernatants (SN) from polarized HCA-7 cells stimulated for 24 h with IL-1␣ and IFN-␥ (20 ng/ml each). MDC/CCL22
concentrations in the supernatants as measured by ELISA were 4.1
ng/ml in the basal supernatant and 0.9 ng/ml in the apical supernatant. CEM T cells (1.25 ⫻ 105 cells) were added to the top well of
chemotaxis chambers in a final volume of 25 ␮l, and cells were
allowed to migrate for 120 min. To determine the contribution of
MDC/CCL22 to supernatant-induced chemotaxis of CEM cells, supernatants were preincubated with anti-MDC/CCL22 antibody (anti-MDC) or an isotype control antibody (No Ab). Values are means ⫾
SE of quadruplicates of a representative experiment.
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 16, 2017
Values are means ⫾ SE of 3 independent experiments. Values in
parentheses are % of total. Supernatants were collected 24 h after
stimulation with 20 ng/ml of each of the indicated cytokines, and
MDC/CCL22 concentrations were determined by ELISA. MDC, macrophage-derived chemokine; TNF, tumor necrosis factor; IL, interleukin; IFN, interferon.
REGULATED MDC/CCL22 PRODUCTION BY HUMAN INTESTINAL EPITHELIUM
natants from TNF-␣- and IFN-␥-stimulated HCA-7
cells with a neutralizing anti-MDC/CCL22 antibody,
but not an isotype-matched control antibody, completely abrogated chemotaxis induced by the supernatants (8.8% of input cells migrated in the presence of
anti-MDC compared with 15.1% of input cells in the
presence of an isotype-matched control antibody).
Epithelial MDC/CCL22 secretion in response to infection with enteroinvasive bacteria. Epithelial cells
can function as sensors for microbial infection and, in
response to bacterial infection, secrete chemokines
that can chemoattract neutrophils and monocyte/macrophages within the intestinal mucosa (22, 29, 34). To
determine if epithelial cell infection with enteroinvasive bacteria upregulates MDC/CCL22 secretion,
HT-29 cells were infected with S. dublin or enteroinvasive E. coli O29:NM. As shown in Fig. 6, infection
with those bacteria increased MDC/CCL22 secretion.
To determine if IFN-␥ synergized with bacterial infection to increase MDC/CCL22 production, HT-29 cells
were stimulated with IFN-␥ immediately after Salmonella infection. As shown in Fig. 6, MDC/CCL22 secretion synergistically increased when bacteria-infected
cultures were stimulated with IFN-␥.
Since infection of HT-29 cells with S. dublin or enteroinvasive E. coli can induce the release of TNF-␣
(i.e., 35–100 pg/ml) (28), which might increase epithelial cell MDC/CCL22 secretion by an autocrine or paracrine effect, anti-TNF-␣ (1 ␮g/ml) was added to bacteria-infected cultures to block the possible effects of
TNF-␣ released by infected cells. Anti-TNF-␣ did not
inhibit MDC/CCL22 secretion in response to S. dublin
or E. coli O29:NM infection (e.g., S. dublin-infected, 4.8
ng/ml; S. dublin plus anti-TNF-␣, 4.4 ng/ml) but completely blocked MDC/CCL22 secretion in response to
added TNF-␣ (20 ng/ml) (17.1 ng/ml vs. 0.3 ng/ml in
TNF-␣-stimulated cultures in the absence or presence
of anti-TNF-␣, respectively).
MDC/CCL22 expression is regulated by NF-␬B. The
transcription factor NF-␬B plays a central role in the
activation of several genes that are upregulated in
intestinal epithelial cells in response to stimulation
with proinflammatory cytokines or bacterial infection
(17). However, little is known about the transcriptional
regulation of MDC/CCL22 in different cell types. To
determine if MDC/CCL22 functions as an NF-␬B target gene in intestinal epithelial cells, HT-29 cells were
treated for 1 h with the proteasome inhibitor MG-132
(10 ␮M). MG-132 completely abrogated TNF-␣-stimulated MDC/CCL22 mRNA expression (Fig. 7A) and
protein secretion (data not shown). Since pharmacological agents are not always completely specific, we used
an additional approach in which HT-29 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) and
acts as a superrepressor of NF-␬B activation (17). As
shown in Fig. 7B, MDC/CCL22 secretion was markedly
inhibited in Ad5I␬B-A32/36-infected cells, but not in
cells infected with control adenovirus, in response to
TNF-␣ or IL-1␣ stimulation or S. dublin or E. coli
O29:NM infection. Real-time RT-PCR was performed
on mRNA isolated 12 h after cytokine stimulation to
examine the effect of inhibiting NF-␬B activation at a
time when maximal changes were observed in MDC/
Fig. 7. Effect of NF-␬B inhibition on MDC/CCL22 expression. A:
HT-29 cells were pretreated with 10 ␮M of the proteasome inhibitor
MG-132 for 60 min before replacement of the media and stimulation
with TNF-␣ for 8h. MDC/CCL22 and ␤-actin mRNA expression were
assessed by RT-PCR as described in MATERIALS AND METHODS. B:
HT-29 cells were left uninfected (open bars) or infected with an
adenovirus encoding an I␬B␣ superrepressor (Ad5I␬B-A32/36;
hatched bars) or ␤-galactosidase (Ad5LacZ, black bars) as a virus
control. Cells were then stimulated with cytokines (20 ng/ml) or
infected with S. dublin or E. coli O29:NM. MDC/CCL22 in supernatants was determined by ELISA 6 h after cytokine stimulation or
bacterial infection. Values are means of 2 independent experiments.
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 16, 2017
Fig. 6. MDC/CCL22 secretion in response to infection with enteroinvasive bacteria. HT-29 cells were grown to confluence and were
infected with E. coli O29:NM or S. dublin at a multiplicity of
infection of 500 and 100, respectively. After infection, cells were
incubated for a further 24 h with media alone or IFN-␥ (20 ng/ml).
Values are means ⫹ SE of 3 or more independent experiments.
G1223
G1224
REGULATED MDC/CCL22 PRODUCTION BY HUMAN INTESTINAL EPITHELIUM
DISCUSSION
The normal adult human intestine is considered to
be “physiologically inflamed.” Thus the normal intestinal mucosa contains B and T lymphocytes, mononuclear phagocytes, and dendritic cells, as well as smaller
numbers of eosinophils and mast cells that, if present
in other tissues, would be regarded as an abnormal
chronic inflammatory cell infiltrate. A requirement for
the presence of cells producing downregulatory immune mediators in the intestinal mucosa is evident in
the IL-10 knockout mouse, which develops spontaneous inflammation of the colonic mucosa (30). We hypothesized that the intestinal epithelium may play a
role in regulating physiological mucosal inflammation
by producing signals with the capacity to chemoattract
T cells that can, in turn, produce anti-inflammatory
cytokines (e.g., IL-4, IL-10) capable of downregulating
mucosal inflammation (3, 23). As we demonstrate
herein, normal human intestinal epithelial cells produce MDC/CCL22, a CC chemokine that could fulfill
that role.
Table 2. Th2 cytokines inhibit TNF␣-stimulated
MDC/CCL22 secretion by HT-29 cells
MDC/CCL/22 Secretion, ng/ml
Cytokines Added
None
IL-4
IL-13
No TNF-␣
⫹TNF-␣
0.30 ⫾ 0.12
0.48 ⫾ 0.25
0.37 ⫾ 0.11
57.5 ⫾ 9.5
20.6 ⫾ 3.8
20.5 ⫾ 3.4
Values are means ⫾ SE of 3 independent experiments. IL-4, IL-13,
and TNF-␣ were each used at a concentration of 20 ng/ml. Supernatants were collected after 24 h of culture. MDC/CCL22 concentrations were assayed by ELISA.
Using cultured human intestinal epithelial cells, we
demonstrated that the regulation of MDC/CCL22 production differs between intestinal epithelial cells and
macrophages and monocytes. Whereas TNF-␣ and IL-1
were potent agonists for MDC/CCL22 protein production by cultured human intestinal epithelial cells, this
was not the case for other MDC/CCL22-expressing
cells (including monocytes, macrophages, and B cells
from human peripheral blood) in vitro (2), although one
report described small increases of MDC/CCL22
mRNA expression in TNF-␣- or IL-1-stimulated macrophages (36). In contrast, IL-4 and IL-13 increased
MDC/CCL22 production by macrophages (2, 7) but had
no effect by themselves on MDC/CCL22 secretion by
cultured intestinal epithelial cells and modestly inhibited TNF-␣-stimulated MDC/CCL22 secretion.
Whereas IFN-␥ alone stimulated low levels of MDC/
CCL22 production in human intestinal epithelial cells
and synergistically increased TNF-␣- and IL-1␣stimulated MDC/CCL22 production in those cells,
IFN-␥ inhibited MDC/CCL22 secretion by macrophages and monocytes (7). Together, these findings
suggest that MDC/CCL22 production by different cell
types is dependent on the cytokine milieu in their
microenvironment, with inflammatory stimuli appearing to favor epithelial but not macrophage-derived
MDC/CCL22 production. The predicted outcome of this
epithelial cell response would be an influx of T cells
known to produce cytokines that can potentially downregulate mucosal inflammation. The observation of
constitutive MDC/CCL22 protein expression in the human intestinal xenograft tissue demonstrates that epithelial-derived MDC/CCL22 is not completely dependent on the presence of an inflammatory state and may
be stimulated by low basal levels of cytokines released
by resident immune cells or by other factors that remain to be identified.
Expression of the chemoattractants MDC/CCL22
and TARC/CCL17, which are both ligands for CCR4,
appears to differ between different mucosal sites. We
show herein that MDC/CCL22, but not TARC/CCL17,
is produced by intestinal epithelial cells in response to
proinflammatory stimuli. In addition, chemotactic activity for the CCR4-expressing T cell line CEM in
supernatants of intestinal epithelial cells was completely abolished by neutralizing anti-MDC/CCL22 antibodies. This finding indicates that MDC/CCL22 may
be the sole intestinal epithelial-derived factor responsible for recruiting CCR4-positive T cells. In contrast to
intestinal epithelial cells, bronchial epithelial cells produce TARC/CCL17, but not MDC/CCL22, after stimulation with the same proinflammatory cytokines (4,
39). The significance of this regional compartmentalization of chemokine expression is unclear since both
chemokines are known to signal through CCR4 as their
receptor and both can act as Th2-type T cell chemoattractants. Nonetheless, these findings suggest that
there may be divergent functions between MDC/
CCL22 and TARC/CCL17.
The regulated production of MDC/CCL22 by intestinal epithelial cells shares similarities with several
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 16, 2017
CCL22 mRNA expression. TNF-␣ induced a 15-fold
increase in MDC/CCL22 mRNA expression in HT-29
cells infected with control virus compared with a 4-fold
increase in cells infected with virus containing the
Ad5I␬B-A32/36 superrepressor. At 12 h after stimulation, MDC/CCL22 protein secretion induced by TNF-␣
was reduced by 91% in Ad5I␬B-A32/36-infected HT-29
cells compared with cells infected with the control
virus. Similar results were obtained in HT-29 cells
stimulated with TNF-␣ plus IFN-␥ at 12 h. Inhibition
of NF-␬B had no effect on the induction of MDC/CCL22
expression in response to stimulation of cells with
IFN-␥ alone (data not shown), which is consistent with
the lack of NF-␬B activation in these cells after IFN-␥
stimulation.
Modulation of MDC/CCL22 secretion by Th2 cytokines. IL-4 and IL-13 increase MDC/CCL22 expression
by monocytes and macrophages (2, 7). To determine if
this is the case also for MDC/CCL22 production by
intestinal epithelial cells, HT-29 cells were stimulated
with IL-4 or IL-13, alone or in the presence of TNF-␣.
IL-4 or IL-13 alone had no significant effect on constitutive MDC/CCL22 secretion. However, IL-4 and IL-13
inhibited TNF-␣-stimulated MDC/CCL22 secretion by
⬃60% (Table 2).
REGULATED MDC/CCL22 PRODUCTION BY HUMAN INTESTINAL EPITHELIUM
We thank John Leopard for expert technical assistance with
immunohistochemistry and thank Dr. D. Guiney for providing monocyte-derived macrophages.
This work was supported by National Institutes of Health Grants
DK-35108, DK-58960, and DK-02808, a Research Grant from the
Crohn’s and Colitis Foundation of America, and a Canadian Institute
of Health Research Fellowship (M. C. Berin).
REFERENCES
1. Agace WW, Roberts AI, Wu L, Greineder C, Ebert EC, and
Parker CM. Human intestinal lamina propria and intraepithelial lymphocytes express receptors specific for chemokines induced by inflammation. Eur J Immunol 30: 819–826, 2000.
2. Andrew DP, Chang MS, McNinch J, Wathen ST, Rihanek
M, Tseng J, Spellberg JP, and Elias CG 3rd. STCP-1 (MDC)
CC chemokine acts specifically on chronically activated Th2
lymphocytes and is produced by monocytes on stimulation with
Th2 cytokines IL-4 and IL-13. J Immunol 161: 5027–5038, 1998.
3. Barbara G, Xing Z, Hogaboam CM, Gauldie J, and Collins
SM. Interleukin 10 gene transfer prevents experimental colitis
in rats. Gut 46: 344–349, 2000.
4. Berin MC, Eckmann L, Broide DH, and Kagnoff MF. Regulated production of the T helper 2-type T cell chemoattractant
TARC by human bronchial epithelial cells in vitro and in human
lung xenografts. Am J Resp Cell Mol Biol 24: 382–389, 2001.
5. Bhan AK, Mizoguchi E, Smith RN, and Mizoguchi A. Colitis
in transgenic and knockout animals as models of human inflammatory bowel disease. Immunol Rev 169: 195–207, 1999.
6. Boirivant M, Fuss IJ, Chu A, and Strober W. Oxazolone
colitis: a murine model of T helper cell type 2 colitis treatable
with antibodies to interleukin 4. J Exp Med 188: 1929–1939,
1998.
7. Bonecchi R, Sozzani S, Stine JT, Luini W, D’Amico G,
Allavena P, Chantry D, and Mantovani A. Divergent effects
of interleukin-4 and interferon-␥ on macrophage-derived chemokine production: an amplification circuit of polarized T helper 2
responses. Blood 92: 2668–2671, 1998.
8. Butcher EC, Williams M, Youngman K, Rott L, and Briskin
M. Lymphocyte trafficking and regional immunity. Adv Immunol 72: 209–253, 1999.
9. Campbell JJ, Haraldsen G, Pan J, Rottman J, Qin S,
Ponath P, Andrew DP, Warnke R, Ruffing N, Kassam N,
Wu L, and Butcher EC. The chemokine receptor CCR4 in
vascular recognition by cutaneous but not intestinal memory T
cells. Nature 400: 776–780, 1999.
10. Chvatchko Y, Hoogewerf AJ, Meyer A, Alouani S, Juillard
P, Buser R, Conquet F, Proudfoot AE, Wells TN, and
Power CA. A key role for CC chemokine receptor 4 in lipopolysaccharide-induced endotoxic shock. J Exp Med 191: 1755–
1764, 2000.
11. D’Ambrosio D, Iellem A, Bonecchi R, Mazzeo D, Sozzani S,
Mantovani A, and Sinigaglia F. Selective up-regulation of
chemokine receptors CCR4 and CCR8 upon activation of polarized human type 2 Th cells. J Immunol 161: 5111–5115, 1998.
12. Dwinell MB, Eckmann L, Leopard JD, Varki NM, and
Kagnoff MF. Chemokine receptor expression by human intestinal epithelial cells. Gastroenterology 117: 359–367, 1999.
13. Dwinell MB, Luegering N, Eckmann L, and Kagnoff MF.
Regulated production of interferon-inducible T-cell chemoattractants by human intestinal epithelial cells. Gastroenterology 120:
49–59, 2001.
14. 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.
15. 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.
16. 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.
17. 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.
18. Fuss IJ, Neurath M, Boirivant M, Klein JS, de la Motte C,
Strong SA, Fiocchi C, and Strober W. Disparate CD4⫹ lam-
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 16, 2017
other chemokines that are expressed by the intestinal
epithelium. Like MDC/CCL22, the CXC chemokines
IL-8/CXCL8 and GRO␣/CXCL1 and the CC chemokines MCP-1/CCL2 and MIP-3␣/CCL20 are upregulated in intestinal epithelial cells in response to TNF-␣
and IL-1␣ stimulation or bacterial infection (26, 42)
and, like the genes that encode those other chemokines
(17), we show that MDC/CCL22 is a NF-␬B target
gene. Furthermore, synergy between IFN-␥ and TNF-␣
or IL-1␣ was observed for MDC/CCL22 regulation,
similar to that noted for IL-8 (14), RANTES, MCP-1
(41), IP-10, and I-TAC (13). Therefore, when the cytokine milieu includes the proinflammatory stimuli
TNF-␣ or IL-1 together with IFN-␥, chemoattractants
for both CCR4-expressing (i.e., MDC/CCL22) and
CXCR3-expressing (i.e., IP-10, Mig, and I-TAC) cells
could be secreted by the intestinal epithelium.
The constitutive production of MDC/CCL22 by intestinal epithelium in vivo suggests that this chemokine
may contribute to the trafficking of CCR4-expressing
CD4⫹ T lymphocytes within the microenvironment of
the normal intestinal mucosa. Nonetheless, the cytokine milieu in the normal human intestinal mucosa is
thought to be predominantly Th1 in nature, with 10fold more IFN-␥-producing than IL-4-producing T cells
(21). This may reflect the low expression of CCR4,
relative to CXCR3, on ␣4␤7-integrin mucosal homing T
cells (1, 9). A recent study demonstrated that ⬃7–10%
of ␣4␤7-integrin-positive peripheral blood T cells express CCR4 (9), which approximately corresponds with
the fraction of Th2 cells found within the intestinal
mucosa (21). It is clear that Th2 cytokine-producing
cells are present in the intestinal mucosa, under both
basal conditions (21) and in sufficient numbers to induce pathology, as seen in the TCR ␣/␣ knockout
mouse (25) and in the oxazolone model of murine colitis
(6). However, MDC/CCL22 may have broader functions
beyond the chemoattraction of Th2 cells. Recently, a
CCR4 knockout mouse was described that exhibited
impaired immune responses and increased mortality
after lipopolysaccharide challenge, associated with a
decrease in macrophage recruitment (10). Furthermore, in vivo administration of MDC/CCL22 has been
shown to increase bacterial clearance in a murine
model of sepsis through the increased recruitment of
macrophages (33). These findings suggest that MDC/
CCL22 has a role in macrophage trafficking and that
epithelial cell-derived MDC/CCL22 may also function
to chemoattract mononuclear cells other than CCR4expressing T cells to the subepithelial regions in the
intestinal mucosa. Alternatively, the relative paucity
of CCR4-expressing T cells in the intestinal mucosa
could suggest that epithelial cell-derived MDC/CCL22
can also act on a chemokine receptor other than CCR4.
G1225
G1226
19.
20.
21.
23.
24.
25.
26.
27.
28.
29.
30.
ina propria (LP) lymphokine secretion profiles in inflammatory
bowel disease. Crohn’s disease LP cells manifest increased secretion of IFN-␥, whereas ulcerative colitis LP cells manifest
increased secretion of IL-5. J Immunol 157: 1261–1270, 1996.
Galli G, Chantry D, Annunziato F, Romagnani P, Cosmi L,
Lazzeri E, Manetti R, Maggi E, Gray PW, and Romagnani
S. Macrophage-derived chemokine production by activated human T cells in vitro and in vivo: preferential association with the
production of type 2 cytokines. Eur J Immunol 30: 204–210,
2000.
Godiska R, Chantry D, Raport CJ, Sozzani S, Allavena P,
Leviten D, Mantovani A, and Gray PW. Human macrophagederived chemokine (MDC), a novel chemoattractant for monocytes, monocyte-derived dendritic cells, and natural killer cells.
J Exp Med 185: 1595–1604, 1997.
Hauer AC, Breese EJ, Walker-Smith JA, and MacDonald
TT. The frequency of cells secreting interferon-␥ and interleukin-4, -5, and -10 in the blood and duodenal mucosa of children
with cow’s milk hypersensitivity. Pediatr Res 42: 629–638, 1997.
Hecht G and Savkovic SD. Review article: effector role of
epithelia in inflammation—interaction with bacteria. Aliment
Pharmacol Ther 11, Suppl 3: 64–68, 1997.
Hogaboam CM, Vallance BA, Kumar A, Addison CL, Graham FL, Gauldie J, and Collins SM. Therapeutic effects of
interleukin-4 gene transfer in experimental inflammatory bowel
disease. J Clin Invest 100: 2766–2776, 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.
Iijima H, Takahashi I, Kishi D, Kim JK, Kawano S, Hori M,
and Kiyono H. Alteration of interleukin 4 production results in
the inhibition of T helper type 2 cell-dominated inflammatory
bowel disease in T cell receptor ␣ chain-deficient mice. J Exp
Med 190: 607–615, 1999.
Izadpanah A, Dwinell MB, and Kagnoff MF. Regulated MIP3␣/CCL20 production by human intestinal epithelium: a mechanism for modulating mucosal immunity. Am J Physiol Gastrointest Liver Physiol 280: G710–G719, 2001.
James SP and Kiyono H. Gastrointestinal lamina propria T
cells. In: Mucosal Immunology (2nd ed.), edited by Ogra PL,
Mestecky J, Lamm ME, Strober W, Bienenstock J, and McGhee
JR. San Diego: Academic, 1999, p. 381–396.
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.
Kuhn R, Lohler J, Rennick D, Rajewsky K, and Muller W.
Interleukin-10-deficient mice develop chronic enterocolitis. Cell
75: 263–74, 1993.
31. Kunkel EJ, Campbell JJ, Haraldsen G, Pan J, Boisvert J,
Roberts AI, Ebert EC, Vierra MA, Goodman SB, Genovese
MC, Wardlaw AJ, Greenberg HB, Parker CM, Butcher EC,
Andrew DP, and Agace WW. Lymphocyte CC chemokine receptor 9 and epithelial thymus-expressed chemokine (TECK)
expression distinguish the small intestinal immune compartment: Epithelial expression of tissue-specific chemokines as an
organizing principle in regional immunity. J Exp Med 192:
761–768, 2000.
32. Libby SJ, Lesnick M, Hasegawa P, Weidenhammer E, and
Guiney DG. The Salmonella virulence plasmid spv genes are
required for cytopathology in human monocyte-derived macrophages. Cellular Microbiology 2: 49–58, 2000.
33. Matsukawa A, Hogaboam CM, Lukacs NW, Lincoln PM,
Evanoff HL, and Kunkel SL. Pivotal role of the CC chemokine,
macrophage-derived chemokine, in the innate immune response.
J Immunol 164: 5362–5368, 2000.
34. McCormick BA, Parkos CA, Colgan SP, Carnes DK, and
Madara JL. Apical secretion of a pathogen-elicited epithelial
chemoattractant activity in response to surface colonization of
intestinal epithelia by Salmonella typhimurium. J Immunol
160: 455–466, 1998.
35. Mizoguchi A, Mizoguchi E, and Bhan AK. The critical role of
interleukin 4 but not interferon ␥ in the pathogenesis of colitis in
T-cell receptor ␣ mutant mice. Gastroenterology 116: 320–326,
1999.
36. Rodenburg RJ, Brinkhuis RF, Peek R, Westphal JR, Van
Den Hoogen FH, van Venrooij WJ, and van de Putte LB.
Expression of macrophage-derived chemokine (MDC) mRNA in
macrophages is enhanced by interleukin-1␤, tumor necrosis factor ␣, and lipopolysaccharide. J Leukoc Biol 63: 606–611, 1998.
37. Sallusto F, Lenig D, Mackay CR, and Lanzavecchia A.
Flexible programs of chemokine receptor expression on human
polarized T helper 1 and 2 lymphocytes. J Exp Med 187: 875–
883, 1998.
38. 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.
39. Sekiya T, Miyamasu M, Imanishi M, Yamada H, Nakajima
T, Yamaguchi M, Fujisawa T, Pawankar R, Sano Y, Ohta
K, Ishii A, Morita Y, Yamamoto K, Matsushima K, Yoshie
O, and Hirai K. Inducible expression of a Th2-type CC chemokine thymus- and activation-regulated chemokine by human
bronchial epithelial cells. J Immunol 165: 2205–2213, 2000.
40. Shibahara T, Wilcox JN, Couse E, and Madara JL. Characterization of epithelial chemoattractants for human intestinal
intraepithelial lymphocytes. Gastroenterology 120: 60–70, 2001.
41. Warhurst AC, Hopkins SJ, and Warhurst G. Interferon ␥
induces differential upregulation of ␣ and ␤ chemokine secretion
in colonic epithelial cell lines. Gut 42: 208–213, 1998.
42. 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 10.220.32.246 on June 16, 2017
22.
REGULATED MDC/CCL22 PRODUCTION BY HUMAN INTESTINAL EPITHELIUM