INFECTION AND IMMUNITY, Jan. 2007, p. 471–480
Copyright © 2007, American Society for Microbiology. All Rights Reserved.
Vol. 75, No. 1
Trefoil Family Factor 2 Is Expressed in Murine Gastric and Immune
Cells and Controls both Gastrointestinal Inflammation and
Systemic Immune Responses?
Evelyn A. Kurt-Jones,1* LuCheng Cao,1Frantisek Sandor,1,2Arlin B. Rogers,3Mark T. Whary,3
Prashant R. Nambiar,3Anna Cerny,1Glennice Bowen,1Jing Yan,1Shigeo Takaishi,4
Alfred L. Chi,4George Reed,1JeanMarie Houghton,1,4
James G. Fox,3and Timothy C. Wang1,4*
Department of Medicine, University of Massachusetts Medical Center, 364 Plantation Street, Lazare Research Building, Worcester,
Massachusetts 016051; Department of Immunology, Comenius University School of Medicine, University Hospital Bratislava,
Department of Oncology, Clinic of Pneumology, Bratislava, Slovak Republic2; Division of Comparative Medicine,
Massachusetts Institute of Technology, Cambridge, Massachusetts 021393; and Division of Gastroenterology,
Department of Medicine, University of Massachusetts Medical Center, 364 Plantation Street,
Lazare Research Building, Worcester, Massachusetts 016054
Received 19 December 2005/Returned for modification 8 February 2006/Accepted 26 October 2006
Trefoil family factor 2 (TFF2), also known as spasmolytic peptide, is a low-molecular-weight protein that is
upregulated in gastric tissues infected with Helicobacter or having other inflammatory conditions, but a precise
function is yet to be elucidated. The role of TFF2 in the development of gastritis, colitis, and inflammatory cytokine
responses was examined both in vivo and in vitro using wild-type and TFF2 knockout mice. TFF2 knockout and
and wild-type mice by administering dextran sodium sulfate (DSS) in drinking water. Histopathology, clinical
disease (colitis), and antibody levels (H. felis) were examined. TFF2 expression in tissues was determined by reverse
transcriptase PCR, and the inflammatory and proliferative responses of TFF2-expressing macrophages and spleen
cells were examined by cytokine enzyme-linked immunosorbent assay, thymidine incorporation, and gene array
studies. TFF2 knockout mice have increased susceptibility to H. felis-induced gastritis, with enhanced gastric
inflammation. They were also more susceptible to DSS-induced colitis, with prolonged colonic hemorrhage and
persistent weight loss. Remarkably, TFF2 expression was not limited to the gastrointestinal tract, as suggested in
previous studies, but was also present in macrophages and lymphocytes. The inflammatory and proliferative
to interleukin 1 beta stimulation but showed normal responses to lipopolysaccharide, suggesting a specific role for
TFF2 in interleukin 1 receptor but not Toll-like receptor 4 signaling via their Toll-interleukin 1 resistance domains.
TFF2?/?lymphocytes also produced higher levels of interleukin 2 than wild-type cells. Thus, TFF2 was expressed
in the gastrointestinal cells and in immune cells and was a negative regulator of gastrointestinal inflammation and
immune cell cytokine responses. Our studies suggest that TFF2 not only controls gastrointestinal repair but also
regulates mononuclear cell inflammatory responses.
Trefoil factor family 2 (TFF2), also known as spasmolytic
polypeptide, is a small peptide upregulated in association with
gastric Helicobacter infection and other diverse inflammatory
conditions of the gastrointestinal tract. There are three known
members of the TFF family (TFF1, -2, and -3), which all
contain one or more three-looped structural motifs resembling
a three-leaf clover or trefoil (36). Under normal conditions,
TFF2 is expressed and secreted by gastric mucosal neck cells
and antral mucous cells, and it appears to play a role in gastric
cytoprotection, at least in part through promotion of epithelial
cell migration. More recently, studies with TFF2-deficient mice
showed evidence of decreased gastric cell proliferation and
increased susceptibility to gastric mucosal injury (10).
While a number of potential biological roles for TFF2 have
been reported, a precise understanding of the function of
TFF2 has remained elusive. In addition, the cellular mecha-
nisms by which TFF2 exerts its effects remain uncertain, since
cell surface receptors for the TFFs have yet to be identified.
Recently, the possibility has been raised that TFF2 may be
involved in some aspect of immune function. Cook et al. re-
ported that TFF2 (and TFF3) could be detected at low levels
in rat spleen, lymph nodes, and bone marrow (2). To investi-
gate the role of TFF2 in immunity, we undertook studies of
immune cells isolated from wild-type and TFF2-deficient mice.
These studies reveal a specific role for TFF2 in the negative
regulation of interleukin 1 (IL-1) receptor-mediated signaling.
MATERIALS AND METHODS
Animals: TFF2?/?mice (Farrell). Specific-pathogen-free TFF2-deficient
(SPKO) and wild-type littermate controls on a mixed C57BL/6 ? 129Sv back-
* Corresponding author. Mailing address for E.K.-J.: University of
Massachusetts Medical School, Department of Medicine, 364 Planta-
tion Street, Lazare Research Building Rm. 226, Worcester, MA 01605.
Phone: (508) 856-3531. Fax: (508) 856-6176. E-mail: evelyn.kurt-jones
@umassmed.edu. Present address for T.C.W.: Division of Digestive
and Liver Diseases, Department of Medicine, College of Physicians
and Surgeons, Columbia University, 630 West 168th Street, Box 83,
New York, NY 10032. Phone: (212) 342-3412. Fax: (212) 305-6443.
?Published ahead of print on 13 November 2006.
ground were generated as previously reported (10). The mice were maintained in
an Association for Assessment and Accreditation of Laboratory and Animal
Care (AAALAC)-approved facility under barrier conditions. Animal protocols
were approved by the Institutional Animal Use and Care Committee of the
University of Massachusetts Medical School.
DSS colitis. Mice received 4% dextran sodium sulfate (DSS) in their drinking
water for 4 days, followed by 18 days of plain water. A second cycle of DSS
treatment, for 3 days, was followed by an additional 25 days of plain water. Mice
were weighed, and blood in the stool was monitored using Hemoccult strips (n ?
11 mice per group for wild-type and knockout mice). For statistical analysis, the
P values were obtained by testing a group ? time interaction in a linear random
effects model (where the animal is a random effect) to test whether the time
trends were significantly different between the groups. A likelihood ratio test was
used to compare nested models (with and without the interation term) after day
23 (11). The entire length of the colon was examined histologically and scored by
a gastrointestinal (GI) pathologist blinded to the treatment group. Statistical
analysis of histologic scores were performed using the Mann-Whitney test. Scor-
ing criteria are given in Table 1.
Helicobacter felis infection studies. An H. felis (ATCC 49179) strain was used
for oral inoculation as previously described (41). The organism was grown for
48 h at 37°C under microaerobic conditions on 5% lysed horse blood agar,
harvested, and resuspended in phosphate-buffered saline. Bacteria were assessed
by Gram stain and phase microscopy for purity, morphology, and motility. Mice
were orally infected with 108CFU of H. felis in 0.3 ml phosphate-buffered saline
given three times every other day. At 3 and 6 months postchallenge, mice were
sacrificed, and gastric tissues were collected from the corpus and antrum and
used for histopathologic evaluation (n ? 4 for the wild type; n ? 8 for TFF2
knockout mice). Murine tissues were harvested, processed, and blindly scored
by a comparative pathologist, using an established scoring rubric which we
have previously described in detail (31). The mice were 70% B6 and 30%
SV/129, but the controls were genetically matched. The inflammation in the
wild-type (littermate) controls is slightly less in mice on a completely SV129
Evaluation of serum antibody responses to H. felis. Serum was collected at
intervals up to 12 months postinfection and evaluated by enzyme-linked
immunosorbent assay (ELISA) for serum immunoglobulin G2c (IgG2c)
(IgG2a-equivalent Th1-promoted isotype in C57BL/6 mice) (25) and IgG1
(Th2-promoted isotype in all strains of mice) using an outer membrane
antigen preparation of H. felis as previously described (13, 30). Antigen (10
?g/ml) was coated on Immulon II plates (Thermo Labsystems, Franklin,
MA), and sera were diluted 1:100. Biotinylated secondary antibodies were
monoclonal antimouse antibodies produced by clones A85-1 and 5.7 (Phar-
mingen, San Diego, CA) for detecting IgG1 and IgG2c, respectively (25).
Assays were developed with extravidin peroxidase (Sigma), followed by the
2,2?-azinobis(3-ethylbenzthiazolinesulfonic acid substrate (Kirkegaard &
Perry Laboratories, Gaithersburg, MD). Data were analyzed using the two-
sample Student t test, assuming equal variances.
Peritoneal macrophages. Peritoneal exudate cells (PECs) were harvested by peri-
toneal lavage of mice. In some experiments, 1 ml of 4% thioglycolate was adminis-
tered intraperitoneally, and cells were harvested 4 days later (elicited macrophages).
PECs were cultured at 106per well in 24-well plates in RPMI 1640 plus 10%
heat-inactivated fetal calf serum and stimulated with lipopolysaccharide (LPS) or
human interleukin 1 beta (IL-1?). LPS from Escherichia coli serotype O111:B4 was
purchased from Sigma (St. Louis, Missouri) and phenol reextracted to remove
contaminating lipopeptides (21). IL-1? was purchased from R&D Systems (Minne-
apolis, MN). Culture supernatants were harvested 18 h after stimulation, and cyto-
kine levels were determined by ELISA (OptEIA; BD-Pharmingen).
For analysis of cytokine gene transcription, resident (nonelicited) peritoneal
macrophages were incubated in polypropylene tubes with or without IL-1? (100
ng/ml) for 2 h. The cells were harvested and lysed, and RNA was prepared using
a commercial RNA extraction kit (RNeasy; QIAGEN, Valencia, CA). Total
mRNA (5 ?g per blot) was reverse transcribed, labeled with biotin, and hybrid-
ized to a Mouse NF-?B Signaling Pathway SuperArray according to the manu-
facturer’s instructions (SuperArray, Inc., Frederick, MD). (The experiment was
repeated on three separate occasions using cells pooled from two to three mice.)
Chemiluminescent images on film were transilluminated, scanned, and analyzed
using GEArray software.
Spleen cell assays. Spleen cells were isolated using lymphocyte separation
medium M and cultured in 96-well plates at 2 ? 105per well with concanavalin
A (ConA) (2.5 ?g/ml) and IL-1? (1 to 100 ng/ml). [3H]thymidine (1 ?Ci per well)
was added for the last 16 h of a 96-h incubation. (The analysis was repeated on
at least three separate occasions using spleen cells from pools of two to three
mice.) For anti-CD3 stimulation, HighProtein binding plates (Corning) were
coated with anti-CD3 antibody (Ab) at 400 ng/ml together with anti-CD28 at 100
ng/ml (BD-Pharmingen). Spleen cells were plated at 1 ? 106per well and
incubated for 18 h. Cytokine levels were measured in culture supernatants by
ELISA. Statistical analysis was performed using Student’s t test (JMP software).
For fluorescence-activated cell sorting analysis, spleen cells were harvested, and
red blood cells were lysed with Tris-ammonium chloride. Cells were incubated
with allophycocyanin- or phycoerythrin-labeled antibody specific for CD3, CD4,
CD8, CD11b, or CD19 (BD-Pharmingen) and analyzed using a BD FACScan
RT-PCR detection of TFF2 and TFF1 mRNA. RNA was extracted using
RNeasy (QIAGEN) reagents and treated with RNase-free DNase (Ambion).
Reverse transcriptase PCR (RT-PCR) analysis was performed on tissues from
individual mice and on tissues from pools of two to three animals on at least
three separate occasions with identical results. Murine embryonic fibroblasts
were prepared as previously described (22). RNA was DNase treated, reverse
transcribed, and PCR amplified using the QIAGEN One-Step/RT-PCR kit. PCR
conditions for all three genes were as follows: 94°C for 2 min, followed by 30
cycles at 94°C for 30 s plus 55°C for 30 s plus 72°C for 30 s and one cycle at 72°C
for 10 min. Primers were TFF2 (5?-GCAGTGCTTTGATCTTGGATGC-3? and
5?-TCAGGTTGGAAAAGCAGCAGTT-3?), TFF1 (5?-TGTGTCCTCGCTGT
GGTCCTCA-3? and 5?-CGATGGCCATGGGGTGGAAGC-3?), and glyceral-
dehyde-3-phosphate dehydrogenase (5?-GACATCAAGAAGGTGGTGAAG
C-3? and 5?-GTCCACCACCCTGTTGCTGTAG-3?).
TFF2?/?mice show increased inflammatory responses to H.
felis. TFF2 is thought to play a role in protection of the
gastrointestinal tract from injury. Exogenous administration
of TFF2 protects mice and rats from GI injury in several
models of gastrointestinal inflammatory disease. We exam-
ined the effect of TFF2 on gastrointestinal pathology using
knockout mice and studied two models of gastrointestinal
inflammation, Helicobacter felis-induced gastritis and DSS-
Wild-type and TFF2?/?mice were inoculated with H. felis
and assessed at 3 and 6 months postinfection. TFF2?/?mice
showed more severe inflammation in response to H. felis than
TABLE 1. Scoring criteria for statistical analysis of histology
Mild; focal or multifocal
Moderate; focal or multifocal
Marked; hemorrhagic and diffuse
Mild, focal erosion
Multifocal erosion or focal ulcer ? submucosal
edema (?2 foci)
Diffuse erosion or multifocal/segmental
ulcer with submucosal edema
Diffuse ulcer and marked submucosal edema
Mild; focal or multifocal
4Severe; hemorrhagic and diffuseSevere; diffuse
472KURT-JONES ET AL.INFECT. IMMUN.
wild-type mice (Fig. 1a and b). Histological scoring revealed
significant differences between wild-type and TFF2?/?mice in
the degree of inflammation, atrophy, mucus metaplasia, and
dysplasia at both the 3- and 6-month time points (Table 2). In
addition, there was a significantly greater degree of hyperplasia
and intestinal metaplasia at 6 months in the TFF2?/?mice
(P ? 0.01) (Table 2).
TFF2?/?mice (like wild-type mice) showed a strong Th1-
polarized T-cell response to H. felis as revealed by IgG subclass
analysis. In fact, the levels of IgG2c were significantly higher in
TFF2?/?mice than in wild-type mice (P ? 0.029) (Fig. 1b).
IgG1 levels were not significantly different (P ? 0.12) between
TFF2?/?and wild-type mice.
TFF2?/?mice are susceptible to DSS colitis: delayed recov-
ery compared to wild-type mice. Intermittent DSS treatment is
thought to model recurrent inflammatory bowel disease by
FIG. 1. TFF2?/?mice show more-severe inflammatory responses to Helicobacter felis. (a) Histopathology from wild-type and TFF2?/?mice
infected for 6 months with H. felis, showing marked inflammation in TFF2?/?cells compared to wild-type cells. (b) H. felis-specific IgG1 and IgG2c
levels in serum (12 months postinfection) determined by ELISA. (IgG2c is the IgG2a-equivalent allele that is uniquely expressed in C57BL/6 mice
.) IgG2c levels were significantly higher in TFF2?/?mice than in wild-type mice (P ? 0.029).
TABLE 2. Histological scores of gastric corpus mucosa in TFF2?/?
mice and wild-type mice with Helicobacter felis
infection for 3 or 6 months
H. felis infection
3 mo 6 mo
0.9 ? 0.5
0.1 ? 0.1
0.3 ? 0.3
0.0 ? 0.0
0.0 ? 0.0
0.0 ? 0.0
2.3 ? 0.6b
1.4 ? 1.0a
1.6 ? 1.3a
0.4 ? 0.4
0.5 ? 0.5
0.5 ? 0.4a
1.1 ? 0.8
1.0 ? 0.6
0.3 ? 0.3
0.0 ? 0.0
0.0 ? 0.0
0.0 ? 0.0
2.4 ? 0.9a
2.3 ? 1.0a
1.6 ? 1.1b
0.6 ? 0.4b
0.7 ? 0.5b
0.6 ? 0.4b
aP ? 0.05; results for TFF2?/?mice compared with those for wild-type mice.
bP ? 0.01; results for TFF2?/?mice compared with those for wild-type mice.
cn ? 8 for TFF2?/?mice with H. felis infection for 6 months. n ? 4 for other
VOL. 75, 2007IL-1 RESPONSE AND INFLAMMATION IN TFF2-DEFICIENT MICE 473
subjecting the colonic epithelium to acute injury followed by a
period of epithelial healing and regeneration (3, 4, 9, 20, 35).
Mice were subjected to two cycles of DSS treatment and were
weighed daily and scored for colonic hemorrhage. Weight loss
was minimal (?10%) during the first cycle of DSS, and no
significant differences were noted between wild-type and
TFF2?/?mice. However, following the second cycle of DSS,
TFF2?/?mice showed greater colonic inflammation and injury
than wild-type mice, as measured clinically by sustained weight
loss (Fig. 2a) (P ? 0.0001) and by increases in the frequency
and persistence of colonic hemorrhage (Fig. 2b) (P ? 0.004).
Pathological examination of colons from mice recovering from
DSS colitis revealed evidence of persistent inflammation, crypt
atrophy, erosion, and ulceration at 3 weeks after the second
cycle of DSS (Fig. 3a to d). Histological scores showed a
tendency for greater inflammatory changes and ulceration in
the colons of TFF2?/?mice than in those of wild-type mice
(Fig. 3e), but the differences did not achieve statistical signif-
icance (P ? 0.05).
Expression of TFF2 in murine tissues. We next examined
TFF2 expression in gastric and nongastric cells. First, we con-
firmed that TFF2 was expressed in the stomach. TFF2 mRNA
could be detected by RT-PCR in stomach tissue from wild-type
and heterozygous (TFF2?/?) mice but not in that from ho-
FIG. 2. TFF2?/?mice are susceptible to DSS colitis. TFF2?/?mice showed delayed recovery from DSS colitis (slower weight gain) and
prolonged inflammation (occult blood) compared to results for wild-type mice. Data shown are means for 11 animals per group starting at day 22,
1 day prior to the start of the second cycle of DSS. (a) Weights relative to starting weight. (b) Blood in stool was monitored both visually and by
Hemoccult strip. Scoring for blood in stool was as follows: 0, none; 1, trace using Hemoccult strips; 2, strong positive using Hemoccult strips; 3,
474 KURT-JONES ET AL.INFECT. IMMUN.
mozygous TFF2?/?knockout animals (Fig. 4a). TFF2 was also
expressed in spleen cells (consistent with studies of Cook et al.
), suggesting that TFF2 expression was not limited to the GI
tract and could be present in cells of the immune system. In
addition to spleen tissue, TFF2 was also expressed in PECs,
primarily macrophages, and in murine embryonic fibroblast
cell lines (Fig. 4a). Interestingly, TFF2 was not well expressed
in the thymus relative to expression in the spleen of wild-type
FIG. 3. Inflammatory responses in colons of TFF2?/?and wild-type mice treated with DSS. Histopathology for TFF2?/?(a, c, d) or wild-type
(b) mice treated with two cycles of DSS followed by 25 days of water. Colons show inflammatory cell infiltration (asterisk), separation and atrophy
of crypts (open arrow), and focal ulceration (solid arrows). (e) Histological scoring of colons from individual wild-type and TFF2?/?mice (P ?
VOL. 75, 2007 IL-1 RESPONSE AND INFLAMMATION IN TFF2-DEFICIENT MICE475
FIG. 4. TFF2 expression in gastric and immune cells. (a) PCR-amplified RNA from wild-type, TFF2?/?, and TFF2?/?mice. TFF2 is expressed
in stomach, splenocytes, PEC, and murine embryonic fibroblasts (MEFs) from wild-type mice but not in thymic cells nor in corresponding cells
from TFF2-deficient mice. Dose-dependent proliferation of spleen cells (b) and thymocytes (c) cultured with ConA plus various doses of IL-1?
(1 to 100 ng/ml). Proliferation of lymphocytes was measured by ?3H]thymidine incorporation in cultures incubated for 96 h with ConA plus IL-1?.
(d) IL-2 and (e) IL-4 cytokine secretion into culture supernatants 24 h after spleen cell challenge with ConA and IL-1?. (f) Gamma interferon
secretion from spleen cells cultured on anti-CD3-coated plates plus anti-CD28.
476 KURT-JONES ET AL.INFECT. IMMUN.
animals, suggesting lesser expression in immature T cells. The
gene encoding TFF1, which is closely related to that encoding
TFF2, lies within the same region of the chromosome as the
gene for TFF2 and is an excellent internal control for RNA
quality and for PCR and gene targeting specificity, i.e., the
gene for TFF2 was specifically targeted in the knockout; thus,
TFF2?/?mice should be wild type for TFF1 expression. As
predicted, TFF1 was expressed in the thymus as well as in the
spleen and stomach of both wild-type and TFF2?/?mice. Fur-
ther, thymic TFF1 levels were higher in TFF2?/?mice than in
Splenic T cells but not thymocytes from TFF2?/?mice are
hyperresponsive to ConA-plus-IL-1?-induced proliferative
signals. The expression of TFF2 in leukocytes in the spleen
and peritoneal cavity raised the possibility that TFF2 could
regulate inflammation by directly affecting mononuclear cells
FIG. 5. Macrophages from TFF2?/?mice show increased responses to IL-1? but not TLR4. (a) Elicited macrophages (PEC) from wild-type
and TFF2?/?mice were challenged with IL-1? 10 to 100 ng/ml or with LPS, a TLR4 ligand, at 10 to 100 ng/ml. Secretion of IL-6 into culture
supernatants was measured by ELISA. (b) Resting macrophages from aged mice with constitutive activation of cytokine genes. Resting (resident)
peritoneal macrophages were cultured in medium alone or challenged with IL-1? for 2 h. Inflammatory cytokine mRNA levels were determined
using a SuperArray NF-?B Signaling Pathway Array membrane. Membranes probed with wild type or TFF2?/?mRNA with or without IL-1?
challenge. Bar graph: Densitometry measurements of selected inflammatory cytokine genes.
VOL. 75, 2007IL-1 RESPONSE AND INFLAMMATION IN TFF2-DEFICIENT MICE 477
(lymphocytes and macrophages) in inflammatory infiltrates.
TFF2?/?splenic T cells exhibited an enhanced proliferative
response compared to that of wild-type T cells when stimulated
with ConA alone or ConA plus IL-1? (Fig. 4b). The hyperre-
sponse of spleen cells was dose dependent on the IL-1? con-
centration, suggesting that TFF2 may have a direct effect on
IL-1R signaling in mature T cells. Interestingly, thymocytes
(immature T cells) from TFF2?/?and wild-type mice had very
similar proliferative responses to ConA plus IL-1? (Fig. 4c),
consistent with the lack of TFF2 expression in these cells (Fig.
4a). Consistent with the enhanced proliferative response of
mature T cells, TFF2?/?splenic T cells secreted higher levels
of IL-2 and IL-4 (Fig. 4d and e) than wild-type T cells when
cultured with ConA or ConA plus IL-1?. TFF2?/?cells also
exhibited enhanced proliferation and cytokine secretion when
activated via their T-cell receptors using anti-CD3 plus CD28
as stimulants (Fig. 4f).
Lymphocyte differentiation in TFF2?/?mice is normal. Our
studies suggested that TFF2 was expressed in the spleen but
not in the thymus and that splenic responses mediated via
IL-1R were dysregulated in TFF2?/?mice. TFF2 deficiency
did not affect the differentiation of lymphocytes in the
TFF2?/?mice. Fluorescence-activated cell sorting analysis of
spleen cells stained for CD3?cells (T and NK cells), CD3?
CD4?cells (T helper cells), CD3?CD8?cells (cytolytic T
cells), CD19?cells (B cells), and CD11b?cells (dendritic cells
and splenic macrophages) indicated that the percentage of
each of these subpopulations was indistinguishable when wild-
type and TFF2?/?spleen cells were analyzed by flow cytometry
(data not shown). Similarly, the percentage of CD4?CD8?
double- and single-positive thymic lymphocytes was unchanged
for TFF2?/?mice compared to that for wild type mice (data
not shown). Thus, TFF2 deficiency did not affect the differen-
tiation of lymphocytes in the TFF2?/?mice, although the
proliferative response of these cells to IL-1? challenge (i.e.,
IL-1R signaling) was enhanced.
Wild-type and TFF2?/?macrophages responded similarly
to TLR4 LPS but were hyperresponsive to IL-1? stimulation
in vitro. Peritoneal macrophages, like spleen cells, expressed
TFF2 mRNA. We therefore wanted to determine if the hyper-
responsiveness to IL-1? we found in TFF2?/?spleen cells was
limited to lymphocytes or occurred in other IL-1-responsive
cell types. Peritoneal macrophages isolated from wild-type and
TFF2?/?mice were challenged with IL-1? to engage IL-1R,
and the secretion of IL-6 into culture supernatants was mea-
sured. TFF2?/?macrophages secreted much higher levels of
IL-6 than wild-type macrophages in response to IL-1R stimu-
lation (Fig. 5a, left panel). In contrast, wild-type macrophages
secreted low levels of IL-6 even at the highest IL-1? concen-
tration tested (100 ng/ml), suggesting that IL-1R activation was
limited in wild-type mice. Interestingly, macrophages from
TFF2?/?heterozygous mice secreted IL-6 at levels intermedi-
ate between those of homozygous knockout cells and wild-type
cells, suggesting a gene dosage effect of TFF2 on IL-1R sig-
naling (Fig. 5a).
Unlike the case with the IL-1? response, wild-type and
TFF2?/?mice had very similar responses to LPS (Fig. 5a, right
panel), suggesting that the hyperresponsiveness of TFF2?/?
mice was specific for IL-1?/IL-1R signaling pathways despite
the fact that the IL-1R pathway shares many common adapters
with the LPS/TLR4 pathway (1).
Resident (resting) macrophages from TFF2?/?mice are
highly activated compared to wild-type macrophages. Our in
vivo and in vitro studies suggested that TFF2 deficiency pro-
duced a dysregulated, proinflammatory response. We exam-
ined the inflammatory gene expression profile of resident peri-
toneal macrophages from older adult mice (?8 weeks of age).
RNA was prepared from freshly isolated, resident macro-
phages, labeled with biotin, and hybridized to a commercial
NF-?B signaling pathway gene array (SuperArray). Wild-type
resident macrophages expressed few inflammatory cytokine
genes, but the expression of these genes could be induced by a
2-h incubation with IL-1? (Fig. 5b). In contrast, TFF2?/?res-
ident macrophages expressed a large number of inflammatory
genes even in the absence of exogenous IL-1? (Fig. 5b). In fact,
the pattern of gene expression in unstimulated TFF2?/?mac-
rophages is remarkably similar to the gene expression pattern
in wild-type macrophages incubated with IL-1?. This included
changes in genes such as those encoding IL-1?, IL-1?, mono-
cyte chemoattractant protein 1, etc. (Fig. 5B). Thus, TFF2?/?
macrophages appeared to have been activated via their IL-1R
in vivo, consistent with the hyperinflammatory phenotype of
Since 1989, it has been well documented that TFFs are
expressed along the length of the normal gastrointestinal tract,
with TFF2 expression being primarily localized to the stomach
(27, 28). In previously published experiments using either rats
or mice, our group has shown that TFF2 is constitutively ex-
pressed in the GI mucosa within specialized epithelial cells and
is markedly upregulated in response to inflammation, injury,
and repair, as demonstrated using immunohistochemisty of
tissue sections (18, 34). Here we demonstrate that immune
cells are an important additional source of TFF2, a novel and
potentially important finding, and that immune cell TFF2 is an
important regulator of the innate immune response in these
In the gut, TFFs are typically expressed in mucus-secreting
cells, and consequently, most of the focus has been on their
role in cytoprotection and reparation of the mucosal lining of
the gut (29, 37, 40). However, given that administration of
exogenous TFFs reduces inflammation in the gastrointestinal
tract (40), Cook et al. (2) first explored the possibility of ex-
pression of TFFs in nongastric cells. They reported the expres-
sion of TFF2 in rat spleen and lymphoid tissues (2). Thus, the
current study extends these findings but also provides the first
demonstration that TFF2 functions as an anti-inflammatory
peptide with specificity for IL-1? signaling and immune cell
In this study, we demonstrate a novel role for TFF2 in the
regulation of IL-1 signaling. We confirmed that TFF2 is ex-
pressed not only in gastric epithelial cells (18, 34) but also in
spleen cells and that TFF2 deficiency is associated with in-
creased cytokine secretion by macrophages and T-cell prolif-
eration in response to IL-1?. These in vitro responses corre-
lated with increased inflammatory responses to LPS injection,
H. felis infection, and DSS models of inflammatory bowel dis-
478KURT-JONES ET AL.INFECT. IMMUN.
ease. The effect of TFF2 deficiency on macrophages was highly
specific for the IL-1 pathway, since no alteration was seen in
the responses of PECs to the TLR4 ligand LPS, which shares
many/most of the same signaling pathways.
Several previous observations suggested that TFF2 might
play a role in controlling or dampening inflammation. Several
groups have reported that application of recombinant trefoils
can reduce inflammatory indices in animal models of colitis
(16, 33, 40). For example, luminal application of recombinant
TFF2/spasmocytic peptide in a rat model of colitis revealed
markedly accelerated colonic mucosal rebuilding and reduced
inflammatory indices (40) through mechanisms that could pos-
sibly include inhibition of inducible nitric oxide synthase and
NO in monocytes (17). In addition, TFF2 is clearly upregulated
in epithelial cells in chronic inflammatory conditions of the
gastrointestinal tract, such as peptic ulcer disease and Crohn’s
disease, and in other organ systems, such as the lung; in fact,
TFF2 may be regulated directly by a variety of cytokines (26).
TFF2 may also function in promoting restitution of the epi-
thelium, and it has always been assumed that such increased
epithelial expression functioned mainly in the context of epi-
thelial regeneration and repair. Recently, TFF2 has been
shown to affect the development of asthma in mice (26). How-
ever, given our results, it would seem reasonable to suggest
that TFF2 may also represent a signal by the epithelium to the
immune system to down-regulate or otherwise dampen an
acute inflammatory response. Thus, TFF2 would serve as a key
player in the epithelial-immune signaling system.
IgG subclass responses have been used as estimates of Th
cell type, with IgG1 being produced in Th2 processes and
IgG2c (the C57BL/6 allelic equivalent of IgG2a) in Th1 re-
sponses of C57BL/6 mice (25, 32, 38). Thus, the IgG1/IgG2c
ratio represents a quantifiable measure of balance between
mucosal Th2/Th1 activity, and in previous studies by our group,
a low ratio (indicating a predominant Th1 response) has cor-
related well with the tendency toward neoplastic progression
(12). TFF2?/?mice, like wild type mice, had a Th1-dominated
response to H. felis and in fact had significantly higher levels of
IgG2c than wild-type mice following H. felis infection. Thus,
the Ab response to infection was consistent with an enhanced
proinflammatory response in the TFF2?/?mice.
The expression of TFF2 in spleen cells raised the possibility
that TFF2 might affect lymphocyte responses. We examined
the effect of TFF2 deficiency on spleen cell responses to B- and
T-cell mitogens. Splenic B cells proliferate in response to
Pokeweed mitogen, while splenic T cells proliferate and se-
crete cytokines in response to ConA. IL-1? (an inflammatory
cytokine which has been linked to human gastrointestinal dis-
ease and gastric cancer progression) synergizes with ConA to
stimulate T-cell proliferation. T cells are also activated by
cross-linking T-cell receptors with plate-bound anti-CD3 Ab.
T cells from TFF2?/?mice exhibited an enhanced prolifer-
ative response to ConA and ConA plus IL-1 and to anti-CD3-
coated plates. Thymocytes, on the other hand, were unaffected
by TFF2 deficiency. Thus, the hyperproliferative response of
the TFF2?/?T cells was limited to mature cells and/or periph-
eral lymphoid tissues, i.e., mature T lymphocytes in the spleen
but not immature T cells in the thymus. The sensitivity to TFF2
follows the pattern of TFF2 gene expression in these lymphoid
organs. In addition to excess proliferation, TFF2?/?lympho-
cytes secreted increased levels of IL-2 and IL-4, suggesting a
generalized enhancement of T-lymphocyte activation. Al-
though TFF2 deficiency markedly increased T-cell prolifera-
tive responses in the spleen, TFF2 deficiency did not affect
B-cell proliferation induced by Pokeweed mitogen (data not
shown). Thus, mature T cells may be the major target of TFF2
regulation in the spleen.
Our current study suggests that in addition to excess cyto-
kine production and lymphocyte proliferation, a major pheno-
type associated with a deficiency of TFF2 is unrestrained
IL-1R signaling. In fact, the TFF2?/?mouse resembled in
some ways a mouse with constitutive IL-1R activation. IL-1?
signals via the type I IL-1R. The cytoplasmic domain of IL-1RI
contains a Toll-interleukin 1 resistance (TIR) domain which
mediates receptor signaling via MyD88-dependent pathways
leading to cytokine secretion. Like IL-1R, TLR4 utilizes TIR
domain-mediated and MyD88-dependent signaling cascades.
Thus, LPS-induced signaling via TLR4 engages many of the
same downstream adapters that are triggered by IL-1?/IL-1R
engagement. To determine if TFF2 deficiency globally en-
hanced TIR signaling, we examined the response of wild-type
and TFF2 knockout macrophages to LPS. Interestingly,
TFF2?/?mice were hyperresponsive to IL-1 receptor signaling
but not to LPS signaling, suggesting a selective enhancement of
Interleukin 1 receptor signaling has been shown to be par-
ticularly important in chronic inflammatory diseases of the
gastrointestinal tract, including both inflammatory bowel dis-
ease and Helicobacter pylori gastritis. IL-1R signaling induces
the expression and secretion of multiple inflammatory cyto-
kines and chemokines, including IL-6, monocyte chemoattrac-
tant protein 1, and tumor necrosis factor alpha, as well as
autocrine secretion of IL-1?. These inflammatory cytokines
were up-regulated in resident peritoneal macrophages from
older TFF2 knockout mice, suggesting an ongoing inflamma-
tory response in vivo. Mononuclear cells isolated from the
intestinal mucosa of patients with Crohn’s disease or ulcerative
colitis show enhanced production of IL-1? (24), and patients
with ulcerative colitis show an increased carrier rate for the
intron 2 IL-1 receptor antagonist polymorphism (39). Infection
with H. pylori also results in a strong inflammatory response led
by the early release of IL-1? along with numerous other cyto-
kines and chemokines (5). Further, we have noted that H. felis
infection in C57BL/6 mice induces a significant IL-1? response
(12). However, the progression of gastric inflammation to gas-
tric atrophy and cancer is closely related to the overall balance
between pro- and anti-inflammatory cytokines and specifically
to the expression of IL-1?. Thus, proinflammatory polymor-
phisms in the IL-1B (511 T/T) and IL-1RN (*2) genes are
associated with an increased risk for gastric carcinoma in H.
pylori-infected individuals (6–8, 14, 23). H. pylori-infected car-
riers of these polymorphisms have been shown to exhibit
higher mucosal levels of IL-1? and more severe inflammation
than H. pylori-infected noncarriers (19). Recently, mice lacking
SIGIRR/Tir8, an important regulator of IL-1 signaling, were
shown to exhibit increased susceptibility to inflammatory bowel
disease (15). Given the central role of IL-1 in the inflammatory
process and the potential harm associated with excessive IL-
1-dependent signaling, it would be reasonable to speculate that
vertebrates have evolved diverse mechanisms, including TFF2,
VOL. 75, 2007IL-1 RESPONSE AND INFLAMMATION IN TFF2-DEFICIENT MICE 479
to counterregulate the response to IL-1?. The recognition of Download full-text
trefoil peptides as a novel class of cytokines may provide new
insight into the modulation of the immune system.
This work was supported by grants from the National Institutes of
Health (RO1 AI51415 [to E.K.-J.], RO1 AI37750 [to J.G.F.], and RO1
CA93405 [to T.C.W.]) and a grant from the Slovak Ministry of Edu-
cation, 1/3437/06 (to F.S.).
1. Adachi, O., T. Kawai, K. Takeda, M. Matsumoto, H. Tsutsui, M. Sakagami,
K. Nakanishi, and S. Akira. 1998. Targeted disruption of the MyD88 gene
results in loss of IL-1- and IL-18-mediated function. Immunity 9:143–150.
2. Cook, G. A., M. Familari, L. Thim, and A. S. Giraud. 1999. The trefoil
peptides TFF2 and TFF3 are expressed in rat lymphoid tissues and partici-
pate in the immune response. FEBS Lett. 456:155–159.
3. Cooper, H. S., S. Murthy, K. Kido, H. Yoshitake, and A. Flanigan. 2000.
Dysplasia and cancer in the dextran sulfate sodium mouse colitis model.
Relevance to colitis-associated neoplasia in the human: a study of histopa-
thology, B-catenin and p53 expression and the role of inflammation. Carci-
4. Cooper, H. S., S. N. Murthy, R. S. Shah, and D. J. Sedergran. 1993. Clini-
copathologic study of dextran sulfate sodium experimental murine colitis.
Lab. Investig. 69:238–249.
5. El-Omar, E. M. 2001. The importance of interleukin 1beta in Helicobacter
pylori associated disease. Gut 48:743–747.
6. El-Omar, E. M., M. Carrington, W. H. Chow, K. E. McColl, J. H. Bream,
H. A. Young, J. Herrera, J. Lissowska, C. C. Yuan, N. Rothman, G. Lanyon,
M. Martin, J. F. Fraumeni, Jr., and C. S. Rabkin. 2000. Interleukin-1
polymorphisms associated with increased risk of gastric cancer. Nature 404:
7. El-Omar, E. M., M. Carrington, W. H. Chow, K. E. McColl, J. H. Bream,
H. A. Young, J. Herrera, J. Lissowska, C. C. Yuan, N. Rothman, G. Lanyon,
M. Martin, J. F. Fraumeni, Jr., and C. S. Rabkin. 2001. The role of inter-
leukin-1 polymorphisms in the pathogenesis of gastric cancer. Nature 412:99.
8. El-Omar, E. M., C. S. Rabkin, M. D. Gammon, T. L. Vaughan, H. A. Risch,
J. B. Schoenberg, J. L. Stanford, S. T. Mayne, J. Goedert, W. J. Blot, J. F.
Fraumeni, Jr., and W. H. Chow. 2003. Increased risk of noncardia gastric
cancer associated with proinflammatory cytokine gene polymorphisms. Gas-
9. Elson, C. O., R. B. Sartor, G. S. Tennyson, and R. H. Riddell. 1995. Exper-
imental models of inflammatory bowel disease. Gastroenterology 109:1344–
10. Farrell, J. J., D. Taupin, T. J. Koh, D. Chen, C. M. Zhao, D. K. Podolsky, and
T. C. Wang. 2002. TFF2/SP-deficient mice show decreased gastric prolifer-
ation, increased acid secretion, and increased susceptibility to NSAID injury.
J. Clin. Investig. 109:193–204.
11. Fitzmaurice, G. M., N. M. Laird, and J. H. Ware. 2004. Applied longitudinal
analysis. John Wiley & Sons, Hoboken, N.J.
12. Fox, J. G., P. Beck, C. A. Dangler, M. T. Whary, T. C. Wang, H. N. Shi, and
C. Nagler-Anderson. 2000. Concurrent enteric helminth infection modulates
inflammation and gastric immune responses and reduces helicobacter-in-
duced gastric atrophy. Nat. Med. 6:536–542.
13. Fox, J. G., A. B. Rogers, M. Ihrig, N. S. Taylor, M. T. Whary, G. Dockray, A.
Varro, and T. C. Wang. 2003. Helicobacter pylori-associated gastric cancer in
INS-GAS mice is gender specific. Cancer Res. 63:942–950.
14. Furuta, T., E. M. El-Omar, F. Xiao, N. Shirai, M. Takashima, H. Sugimura,
and H. Sugimurra. 2002. Interleukin 1beta polymorphisms increase risk of
hypochlorhydria and atrophic gastritis and reduce risk of duodenal ulcer
recurrence in Japan. Gastroenterology 123:92–105.
15. Garlanda, C., F. Riva, N. Polentarutti, C. Buracchi, M. Sironi, M. De
Bortoli, M. Muzio, R. Bergottini, E. Scanziani, A. Vecchi, E. Hirsch, and A.
Mantovani. 2004. Intestinal inflammation in mice deficient in Tir8, an in-
hibitory member of the IL-1 receptor family. Proc. Natl. Acad. Sci. USA
16. Giraud, A. S. 2000. X. Trefoil peptide and EGF receptor/ligand transgenic
mice. Am. J. Physiol. Gastrointest. Liver Physiol. 278:G501–G506.
17. Giraud, A. S., P. M. Pereira, L. Thim, L. M. Parker, and L. M. Judd. 2004.
TFF-2 inhibits iNOS/NO in monocytes, and nitrated protein in healing colon
after colitis. Peptides 25:803–809.
18. Houghton, J., C. Stoicov, S. Nomura, A. B. Rogers, J. Carlson, H. Li, X. Cai,
J. G. Fox, J. R. Goldenring, and T. C. Wang. 2004. Gastric cancer originating
from bone marrow-derived cells. Science 306:1568–1571.
19. Hwang, I. R., T. Kodama, S. Kikuchi, K. Sakai, L. E. Peterson, D. Y.
Graham, and Y. Yamaoka. 2002. Effect of interleukin 1 polymorphisms on
gastric mucosal interleukin 1beta production in Helicobacter pylori infec-
tion. Gastroenterology 123:1793–1803.
20. Kitajima, S., S. Takuma, and M. Morimoto. 2000. Histological analysis of
murine colitis induced by dextran sulfate sodium of different molecular
weights. Exp. Anim. 49:9–15.
21. Kurt-Jones, E. A., L. Mandell, C. Whitney, A. Padgett, K. Gosselin, P. E.
Newburger, and R. W. Finberg. 2002. Role of toll-like receptor 2 (TLR2) in
neutrophil activation: GM-CSF enhances TLR2 expression and TLR2-me-
diated interleukin 8 responses in neutrophils. Blood 100:1860–1868.
22. Kurt-Jones, E. A., F. Sandor, Y. Ortiz, G. N. Bowen, S. L. Counter, T. C.
Wang, and R. W. Finberg. 2004. Use of murine embryonic fibroblasts to
define Toll-like receptor activation and specificity. J. Endotoxin Res. 10:419–
23. Machado, J. C., C. Figueiredo, P. Canedo, P. Pharoah, R. Carvalho, S.
Nabais, C. Castro Alves, M. L. Campos, L. J. Van Doorn, C. Caldas, R.
Seruca, F. Carneiro, and M. Sobrinho-Simoes. 2003. A proinflammatory
genetic profile increases the risk for chronic atrophic gastritis and gastric
carcinoma. Gastroenterology 125:364–371.
24. Mahida, Y. R., K. Wu, and D. P. Jewell. 1989. Enhanced production of
interleukin 1-beta by mononuclear cells isolated from mucosa with active
ulcerative colitis of Crohn’s disease. Gut 30:835–838.
25. Martin, R. M., J. L. Brady, and A. M. Lew. 1998. The need for IgG2c specific
antiserum when isotyping antibodies from C57BL/6 and NOD mice. J. Im-
munol. Methods 212:187–192.
26. Nikolaidis, N. M., N. Zimmermann, N. E. King, A. Mishra, S. M. Pope, F. D.
Finkelman, and M. E. Rothenberg. 2003. Trefoil factor-2 is an allergen-
induced gene regulated by Th2 cytokines and STAT6 in the lung. Am. J.
Respir. Cell Mol. Biol. 29:458–464.
27. Nomura, S., T. Baxter, H. Yamaguchi, C. Leys, A. B. Vartapetian, J. G. Fox,
J. R. Lee, T. C. Wang, and J. R. Goldenring. 2004. Spasmolytic polypeptide
expressing metaplasia to preneoplasia in H. felis-infected mice. Gastroen-
28. Otto, W. R., and K. Patel. 1999. Trefoil factor family (TFF)-domain peptides
in the mouse: embryonic gastrointestinal expression and wounding response.
Anat. Embryol. (Berlin) 199:499–508.
29. Poulsen, S. S., J. Thulesen, L. Christensen, E. Nexo, and L. Thim. 1999.
Metabolism of oral trefoil factor 2 (TFF2) and the effect of oral and paren-
teral TFF2 on gastric and duodenal ulcer healing in the rat. Gut 45:516–522.
30. Pronovost, A. D., S. L. Rose, J. W. Pawlak, H. Robin, and R. Schneider. 1994.
Evaluation of a new immunodiagnostic assay for Helicobacter pylori antibody
detection: correlation with histopathological and microbiological results.
J. Clin. Microbiol. 32:46–50.
31. Rogers, A. B., N. S. Taylor, M. T. Whary, E. D. Stefanich, T. C. Wang, and
J. G. Fox. 2005. Helicobacter pylori but not high salt induces gastric intra-
epithelial neoplasia in B6129 mice. Cancer Res. 65:10709–10715.
32. Romagnani, S. 1997. The Th1/Th2 paradigm. Immunol. Today 18:263–266.
33. Soriano-Izquierdo, A., M. Gironella, A. Massaguer, F. E. May, A. Salas, M.
Sans, R. Poulsom, L. Thim, J. M. Pique, and J. Panes. 2004. Trefoil peptide
TFF2 treatment reduces VCAM-1 expression and leukocyte recruitment in
experimental intestinal inflammation. J. Leukoc. Biol. 75:214–223.
34. Stoicov, C., X. Cai, H. Li, K. Klucevsek, J. Carlson, R. Saffari, and J.
Houghton. 2005. Major histocompatibility complex class II inhibits Fas an-
tigen-mediated gastric mucosal cell apoptosis through actin-dependent inhi-
bition of receptor aggregation. Infect. Immun. 73:6311–6321.
35. Strober, W., I. J. Fuss, and R. S. Blumberg. 2002. The immunology of
mucosal models of inflammation. Annu. Rev. Immunol. 20:495–549.
36. Taupin, D., and D. K. Podolsky. 2003. Trefoil factors: initiators of mucosal
healing. Nat. Rev. Mol. Cell. Biol. 4:721–732.
37. Thim, L. 1997. Trefoil peptides: from structure to function. Cell Mol. Life
38. Toellner, K. M., S. A. Luther, D. M. Sze, R. K. Choy, D. R. Taylor, I. C.
MacLennan, and H. Acha-Orbea. 1998. T helper 1 (Th1) and Th2 charac-
teristics start to develop during T cell priming and are associated with an
immediate ability to induce immunoglobulin class switching. J. Exp. Med.
39. Tountas, N. A., V. Casini-Raggi, H. Yang, F. S. Di Giovine, M. Vecchi, L.
Kam, L. Melani, T. T. Pizarro, J. I. Rotter, and F. Cominelli. 1999. Func-
tional and ethnic association of allele 2 of the interleukin-1 receptor antag-
onist gene in ulcerative colitis. Gastroenterology 117:806–813.
40. Tran, C. P., G. A. Cook, N. D. Yeomans, L. Thim, and A. S. Giraud. 1999.
Trefoil peptide TFF2 (spasmolytic polypeptide) potently accelerates healing
and reduces inflammation in a rat model of colitis. Gut 44:636–642.
41. Wang, T. C., C. A. Dangler, D. Chen, J. R. Goldenring, T. Koh, R. Ray-
chowdhury, R. J. Coffey, S. Ito, A. Varro, G. J. Dockray, and J. G. Fox. 2000.
Synergistic interaction between hypergastrinemia and Helicobacter infection
in a mouse model of gastric cancer. Gastroenterology 118:36–47.
Editor: J. L. Flynn
480KURT-JONES ET AL.INFECT. IMMUN.