Kru ¨ppel-Like Factor 5 Protects against Murine Colitis and
Activates JAK-STAT Signaling In Vivo
Marie-Pier Tetreault, Rami Alrabaa, Megan McGeehan, Jonathan P. Katz*
Division of Gastroenterology, Department of Medicine, University of Pennsylvania Perelman School of Medicine, Philadelphia, Pennsylvania, United States of America
Inflammatory bowel disease (IBD), which is characterized by chronic or recurring inflammation of the gastrointestinal tract,
affects 1.4 million persons in the United States alone. KLF5, a Kru ¨ppel-like factor (KLF) family member, is expressed within
the epithelia of the gastrointestinal tract and has been implicated in rapid cell proliferation, migration, and remodeling in a
number of tissues. Given these functions, we hypothesized that constitutive Klf5 expression would protect against the
development of colitis in vivo. To examine the role of KLF5 in vivo, we used the Villin promoter to target Klf5 to the entire
horizontal axis of the small intestine and colon. Villin-Klf5 transgenic mice were born at normal Mendelian ratios and
appeared grossly normal to at least 1 year of age. Surprisingly, there were no significant changes in cell proliferation or in
the differentiation of any of the intestinal lineages within the duodenum, jejunum, ileum, and colon of Villin-Klf5 mice,
compared to littermate controls. However, when Villin-Klf5 mice were treated with dextran sodium sulfate (DSS) to induce
colitis, they developed less colonic injury and significantly reduced disease activity scores than littermate controls. The
mechanism for this decreased injury may come via JAK-STAT signaling, the activation of which was increased in colonic
mucosa of DSS treated Villin-Klf5 mice compared to controls. Thus, KLF5 and its downstream mediators may provide
therapeutic targets and disease markers for IBD or other diseases characterized by injury and disruption of intestinal
Citation: Tetreault M-P, Alrabaa R, McGeehan M, Katz JP (2012) Kru ¨ppel-Like Factor 5 Protects against Murine Colitis and Activates JAK-STAT Signaling In
Vivo. PLoS ONE 7(5): e38338. doi:10.1371/journal.pone.0038338
Editor: Markus M. Heimesaat, Charite ´, Campus Benjamin Franklin, Germany
Received November 9, 2011; Accepted May 7, 2012; Published May 31, 2012
Copyright: ? 2012 Tetreault et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by: the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) of the National Institutes of Health through
R01 DK080031 and DK080031-02S1 to JPK; the NIDDK through P30 DK050306, the University of Pennsylvania Center for Molecular Studies in Digestive and Liver
Diseases, which includes the Molecular Pathology and Imaging Core, the Cell Culture Core, the Transgenic and Chimeric Mouse Facility, and the Molecular
Biology/Gene Expression Core; and the National Cancer Institute (NCI) of the National Institutes of Health through P01 CA098101 ("Mechanisms of Esophageal
Carcinogenesis"). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: email@example.com
Intestinal epithelial cells typically form a selective permeability
barrier to separate luminal contents from the underlying tissues
. This barrier is critical to protect the host against the numerous
pathogens and other insults contained within the intestinal lumen.
Disruption of the permeability barrier is seen in certain intestinal
diseases such as inflammatory bowel disease (IBD), which affects
1.4 million persons in the United States , as well as celiac
disease, ischemia, and intestinal infections [3,4,5]. IBD, which has
two main forms, ulcerative colitis and Crohn’s disease, is a chronic
disease characterized by persistent or recurring inflammation and
immune response within the gastrointestinal mucosa . Thus,
most treatments for IBD target the immune response. Yet, the
ability of the intestine to repair itself following acute, chronic, or
relapsing injury is also likely to be key for disease activity and
Following mucosal injury, the initial response is of the intestine
is epithelial restitution, with rapid migration of cells from the
wound edge to restore surface epithelial continuity [5,6]. Epithelial
cells then undergo increased proliferation and cell differentiation
to fully reconstitute the intestinal epithelium. Thus factors that
upregulate cell proliferation and migration are likely to be critical
for intestinal repair following inflammation and injury. A number
of regulatory pathways have been shown to be important in
intestinal injury, including the JAK-STAT pathways [1,3,6]. In
mammals, the JAK/STAT pathway is a key signaling mechanism
for growth factors and cytokines , and activation of STAT3, in
particular, protects against the development of colitis and
The zinc-finger transcription factor Kru ¨ppel-like factor 5 (KLF5;
also known as BTEB2) has been implicated in tissue repair,
including in the intestine [14,15]. KLF5 has an epithelial-specific
expression pattern in the intestine and is localized primarily to the
proliferating cells within the small intestinal crypts and the lower
third of colonic crypts [16,17]. KLF5 promotes rapid cell
proliferation, migration, and remodeling [18,19,20,21,22], and
mice with hemizygous deletion of Klf5 have greater sensitivity to
dextran sodium sulfate (DSS) colitis than wild-type mice .
However, KLF5 has also been suggested to be an oncogene in the
intestine [18,24,25] and thus the potential relevance of KLF5
activation in intestinal injury and repair is not certain. Interest-
ingly, when Klf5 was targeted to esophageal epithelia in mice, we
found no changes in the gross morphology and histology of the
epithelium and no obvious increase in malignant potential, despite
a clear increase in basal cell proliferation . However, to date,
the effect of increased KLF5 in intestinal epithelia in vivo has not
PLoS ONE | www.plosone.org1 May 2012 | Volume 7 | Issue 5 | e38338
been examined, and a protective effect of constitutive intestinal
KLF5 expression on colitis has not yet been demonstrated.
Here, using the Villin promoter, we targeted Klf5 throughout the
entire intestinal epithelium, from duodenum to colon, along the
entire crypt-villus axis, and throughout the colonic crypts. We then
treated Villin-Klf5 transgenic mice with DSS to induce colitis and
examine the effect of KLF5 on intestinal injury and restitution in
Materials and Methods
Generation of Villin-Klf5 mice
All animal studies were approved by the Institutional Animal
Care and Use Committee (IACUC) at the University of
Pennsylvania. To express Klf5 in intestinal epithelia, we cloned
the complete coding sequence of murine Klf5  into the
p12.4KVill plasmid (gift of Dr. Deborah Gumucio, University of
Michigan) . This construct was sequenced, and the Villin-Klf5
fragment was excised for injection. Derivation of transgenic mice
was accomplished by the University of Pennsylvania Transgenic
and Chimeric Mouse Facility. We documented transgene integra-
tion in nine Villin-Klf5 founder lines by PCR for the Villin
promoter. Offspring were then screened for transgene expression
by RNAse protection assays, Western blot, and immunohisto-
chemistry, as described below, and a total of four lines
demonstrated transgene expression. Since initial phenotypic
analyses revealed no differences among the lines, further studies
were carried out on a single Villin-Klf5 transgenic line. Mice were
backcrossed to C57BL/6 (Charles River Laboratories, Wilming-
ton, MA) for at least 10 generations.
DSS-induced acute injury
Two month-old control and Villin-Klf5 mice were given 3.5%
DSS (MP Biomedicals, Aurora, OH) in their drinking water for
7 days, whereas untreated mice received water alone. Mice were
monitored daily for weight loss and visible signs of rectal bleeding.
Occult bleeding was evaluated (Hemoccult; Beckmann Coulter,
Fullerton, CA) at day 7, and disease index was calculated by
assessing weight loss, occult blood, and stool consistency, as
previously described . Mice were sacrificed 0, 3, and 7 days
after initiation of DSS treatment.
Villin-Klf5 mice and littermate controls were injected with 5-
bromo-2-deoxyuridine (BrdU) Labeling Reagent (Life Technolo-
gies, Grand Island, NY) 60 min prior to sacrifice. For untreated
mice, small and large intestines were removed at 1, 3, 6, and
12 months of age, examined grossly, and then processed for
histology as previously described . Briefly, the small and large
intestine were dissected out, longitudinally splayed, flushed with
ice-cold PBS, swiss-rolled, and fixed in 4% paraformaldehyde
overnight at 4uC. Tissues were then embedded in paraffin, and 5-
mm sections were applied to Probe-on Plus slides (Fisher Scientific,
Pittsburgh, PA). Sample slides were stained with hematoxylin and
eosin. For Alcian blue staining, 3% aqueous acetic acid was
applied to the slides followed by addition of 1% Alcian blue in 3%
acetic acid, pH 2.5, then counterstained with 0.1% nuclear fast
red. Images were captured on a Nikon Eclipse E600 microscope
(Nikon Instruments, Melville, NY) with a Photometrics CoolSNAP
charge-coupled device camera (Roper Scientific, Tucson, AZ).
Total RNA was extracted from mouse intestine using the
RNeasy Mini Kit (Qiagen, Valencia, CA) following manufactur-
er’s instructions. Ribonuclease (RNase) protection assays were
performed as described previously using 1 mg of total RNA per
sample . Probes were designed to span the 39 end of the Villin-
Klf5 transgene, from the Klf5 cDNA to just upstream of the
polyadenylation signal, to protect fragments of 345 bp for
endogenous Klf5 and 459 bp for the Villin-Klf5 transgene. A 103-
nt cyclophilin probe was employed as an internal standard. RNA
fragments obtained were separated on a Novex 6% TBE-urea
acrylamide gel (Life Technologies, Carlsbad, CA) and the
radioactive bands visualized on a Storm 840 phosphorimager
(GE Healthcare Bio-Sciences, Piscataway, New Jersey). Quantita-
tive real-time PCR analysis was performed on an ABI Step One
Plus sequence detection system (Life Technologies) using condi-
tions and primer concentrations suggested by the SyBr Green
PCR master mix (Life Technologies) protocol. Reverse transcrip-
tion was performed with MaximaH First Strand cDNA synthesis
kit for RT-qPCR (Thermo Fisher Scientific). The TATA box
binding protein (TBP) or Glyceraldehyde-3-phosphate dehydro-
genase (GAPDH) genes were used as the internal control. Primer
sequences are available upon request.
Immunohistochemistry and quantitation of proliferative
We performed microwave antigen retrieval and processed the
tissues as previously described , followed by an incubation with
one of the following primary antibodies: 1:5,000 anti-KLF5, which
we generated previously ; 1:15,000 rat anti-BrdU (Accurate
Chemicals, Westbury, NY); 1:1,500 Ki-67 (Vector Laboratories,
Burlingame, CA); 1:5,000 Lysozyme (Dako, Carpinteria, CA); or
1:50 rabbit anti-phospho-Stat3 (Y705) (Cell Signaling Technology
DJI C DJI C
Figure 1. Villin-Klf5 transgene was expressed in small and large
intestinal epithelia. (A) RNAse protection assay demonstrated strong
transgene expression in the duodenum (D), jejunum (J), ileum (I), and
colon (C) of Villin-Klf5 mice, compared to their littermate controls.
Cyclophilin was used as a loading control. (B) At 1 month of age,
compared to controls (Cn), Villin-Klf5 mice (Vl) had increased KLF5
expression in the small intestine and colon by Western blot. (C–D)
Immunofluorescence of 6 month-old control mice revealed nuclear
KLF5 staining restricted to small intestinal crypts (C), with weak nuclear
staining in colonic crypt cells (D). (E-D) Villin-Klf5 mice at 6 months of
age had stronger KLF5 staining, with nuclear staining seen along the
entire crypt-villus axis of the small intestine (E) and increased numbers
of KLF5-positive cells in colonic mucosa (F). In both control and Villin-
Klf5 mice, cytoplasmic KLF5 staining was seen in the surface epithelia.
Scale bars: 50 mm.
KLF5 Protects against Colitis In Vivo
PLoS ONE | www.plosone.org2May 2012 | Volume 7 | Issue 5 | e38338
Inc., Beverly, MA). Species-specific secondary antibodies were
added, and antibody binding was detected as previously described
. For fluorescent labeling, 1:600 Cy3 (Jackson ImmunoR-
esearch, West Grove, PA) antibody was used. Alkaline phospha-
tase activity was analyzed by incubation with 5-bromo-4-chloro-3-
indolyl-phosphatase-4-nitro blue tetrazolium chloride. The prolif-
erative index was determined by counting the numbers of BrdU-
or Ki-67-labeled cells per crypt in a least five distinct regions of
intestine from at least five Villin-Klf5 mice and five littermate
controls at each time point. Results were expressed as the average
number of labeled cells 6 SEM.
Protein aliquots (20 mg from mouse tissue or cell lysates were
separated by SDS-PAGE on 4–10% gels and electrotransfered
onto PVDF membranes (EMD Millipore, Billerica, MA). Mem-
branes were blocked in PBS containing 5% powdered milk and
0.05% Tween-20% for 1 hour at 25uC then incubated overnight
at 4uC with primary antibody in blocking solution followed by
horseradish peroxidase-conjugated secondary antibody (1:10,000)
for 1 hour. Blots were visualized using the Immobilon ECL system
(EMD Millipore). Protein concentrations were assayed using BCA
Protein Assay Reagent (Thermo Fisher Scientific) with BSA as
Cells were wounded using the scratch assay technique .
IEC-6 cells, transduced with pFB-neo or pFB-Klf5 retrovirus ,
were seeded in 6-well plates and allowed to reach confluence, after
which the functional epithelial monolayers were wounded linearly
several times. Cells were harvested and then lysed in Triton X-100
sample buffer for use in Western blotting.
To evaluate the consequences of KLF5 overexpression in
intestinal epithelial cells in vivo, we used the 12.4 Kb Villin
promoter  to target Klf5 along the entire horizontal and
longitudinal axes of the intestine in mice. Villin-Klf5 mice were
born at the appropriate Mendelian ratio and appeared grossly
normal up to at least 12 months of age. Transgene expression
throughout the small and large intestine of Villin-Klf5 mice was
confirmed by RNAse protection assay (Figure 1A). Increased
KLF5 protein levels were demonstrated by Western blot
(Figure 1B) of small intestine and colon from Villin-Klf5 mice
compared to littermate controls. In control mice, KLF5 was
restricted to the small intestinal crypts (Figure 1C) and the lower
third of colonic crypts (Figure 1D), regions of active cell
proliferation, as demonstrated by immunofluorescence. In con-
trast, KLF5 expression in Villin-Klf5 mice was seen in both the
small intestinal crypts and villi (Figure 1E) and throughout colonic
crypts (Figure 1F).
Compared to littermate controls (Figure 2A), Villin-Klf5 mice
had no morphological changes in the small intestine (Figure 2B) or
colon (not shown) at 12 months of age. Moreover, there was no
evidence of dysplasia, polyp formation, or cancer up to at least
12 months of age (not shown). Given our recent demonstration of
KLF5 as a positive regulator of proliferation in esophageal
epithelia of mice , we examined whether intestinal epithelial
proliferation was similarly altered in Villin-Klf5 mice. Surprisingly,
we found no significant difference in the number of proliferating
cells between control and Villin-Klf5 mice in any intestinal segment
(Figure 2C–D); proliferating cells were restricted to the small
intestinal crypts and the base of the colonic crypts in both control
and Villin-Klf5 mice (not shown). We next examined if intestinal
cell differentiation and lineage specification were altered by
transgenic expression of Klf5. Numbers of enteroendocrine cells,
as determined by staining for chromagranin A (Figure 2E), were
unchanged in small and large intestines of Villin-Klf5 mice,
compared to littermate controls. Likewise, goblet cells, visualized
by Alcian Blue staining, were seen at similar numbers in control
and Villin-Klf5 mice (Figure 2F). Paneth cell and enterocyte
differentiation was also unaffected by increased expression of Klf5
in Villin-Klf5 mice (not shown). Thus, at baseline, increased Klf5
expression in intestinal epithelia is not sufficient to perturb normal
epithelial homeostasis in vivo.
KLF5 is known to control cell migration, wound repair, and
inflammatory responses [20,23,31]. Therefore, we examined
whether increased Klf5 expression protects intestinal epithelia
from tissue damage. Villin-Klf5 and control mice were challenged
with 3.5% DSS for 7 days to induce experimental colitis. Disease
activity index was determined by scoring changes in weight,
hemoccult positivity, and stool consistency  and was signifi-
cantly lower in Villin-Klf5 mice than littermate controls (Figure 3A).
Compared to colonic tissues from control mice (Figure 3B), colonic
sections from Villin-Klf5 mice (Figure 3C) revealed less tissue
damage and improved regeneration of the intestinal mucosa
following DSS-induced injury. Notably, crypt architectural distor-
tion and focal thinning of colonic surface epithelia were observed
in control mice, while crypt architecture was better preserved in
the regenerating colonic mucosa of Villin-Klf5 mice (Figure 3C).
Higher numbers of surface epithelial erosions were also observed
in the colon of control mice treated with DSS, compared to Villin-
Klf5 mice (not shown). At day 0 of DSS treatment, compared to
controls (Figure 3D), nuclear KLF5 expression was increased in
colonic mucosa of Villin-Klf5 mice (Figure 3E). KLF5 was also
expressed in colonic epithelia of control mice following DSS
treatment (Figure 3F), but Villin-Klf5 mice (Figure 3G) had
substantially more KLF5 in the epithelial cells of the regenerating
intestinal mucosa adjacent to the regions of ulceration. Moreover,
expression was clearly nuclear in the lower half of the crypts of
Villin-Klf5 mice, consistent with the established role of KLF5 as a
transcriptional regulator. To characterize changes in proliferation
in Villin-Klf5 mice challenged with DSS, we pulse-labeled colonic
epithelial cells with BrdU. Compared to controls (Figure 3H),
Villin-Klf5 mice (Figure 3I) had more proliferating cells by BrdU
labeling in areas of intestinal epithelial regeneration. Thus, Klf5
overexpression, while not altering colonic homeostasis in the
unchallenged state, protects against colonic injury and promotes
epithelial restitution following challenge with DSS.
KLF5 can regulate cytokine production in response to injury or
other stimuli [32,33,34], and multiple cytokine-induced signaling
pathways converge on the key transcription factor STAT3 .
Moreover, activated STAT3 is critical both for protection against
colitis and for the restoration of intestinal integrity during colitis
[8,9,10,11,12]. We hypothesized that the protective effect of KLF5
following intestinal mucosal injury by DSS was mediated by
activation of the JAK/STAT pathway. To assess the role of
STAT3 activation, we initially performed immunohistochemistry
for phosphorylated STAT3 in control and Villin-Klf5 mice
following DSS treatment. In control mice treated with DSS,
phosphorylated STAT3 was present in intestinal epithelial cells
and in the stroma (Figure 4A), but levels of phosphorylated
STAT3 were much higher in regenerating epithelial cells of Villin-
Klf5 mice (Figure 4B). These findings were confirmed by Western
blotting of colonic epithelial scrapings from mice treated with
DSS, and this increase in STAT3 activity in Villin-Klf5 mice also
correlated with elevated phosphorylated JAK2 (Figure 4C). Of
KLF5 Protects against Colitis In Vivo
PLoS ONE | www.plosone.org3May 2012 | Volume 7 | Issue 5 | e38338
note, no significant changes in STAT3 phosphorylation levels
were observed between control and Villin-Klf5 mice at days 0 and
3 of DSS treatment. These data suggest that KLF5 may regulate
intestinal epithelial regeneration in experimental colitis through
the JAK2/STAT3 pathway.
We next sought to determine how KLF5 regulates STAT3
activation in the context of mucosal injury. IEC-6 intestinal
epithelial crypt cells were retrovirally transduced with pFB-neo or
pFB-Klf5, and confluent monolayers were injured and examined at
different time points following wounding. Western blot analyses
showed that phospho-STAT3 levels increased slightly following
wounding compared to unstimulated cells (Figure 4D), consistent
with previous findings that STAT3 plays a role in epithelial cell
migration in vitro . Surprisingly, however, Klf5 overexpression
did not increase STAT3 phosphorylation compared to controls.
Thus, the mechanism of STAT3 activation by KLF5 in the
context of colonic mucosal wound healing appears to be non cell-
STAT3 activation in murine colitis is dependent on IL-22,
which is secreted by cells of the innate immune system . To
determine whether IL-22 signaling was increased in Villin-Klf5
mice, we analyzed IL-22 and IL-6 mRNA expression in lysates of
colonic stroma following removal of the epithelium. Compared to
control mice, Villin-Klf5 mice had significantly increased IL-22
expression, whereas IL-6 expression was unchanged (Figure 5A).
Consistent with these findings, Villin-Klf5 mice demonstrated
increased expression of IL-22 receptor mRNA in colonic epithelia,
compared to control mice (Figure 5B).
Maintenance of intestinal epithelial homeostasis depends on a
careful balance between cell proliferation, differentiation, migra-
tion, and apoptosis [36,37]. Moreover, the presence of a functional
mucosal barrier is an essential for protection against numerous
intestinal diseases . Numerous studies have illustrated the
importance of the Kru ¨ppel-like factor family member KLF5 in the
regulation of epithelial homeostasis and diseases . Recently,
mice with hemizygous deletion of Klf5 were found to have greater
sensitivity to DSS colitis than wild-type mice . Yet the
consequences of increased KLF5 expression in the intestine have
not been examined in vivo.
A large body of evidence implicates KLF5 as a positive
regulator of proliferation, including in non-transformed intestinal
epithelial cells in vitro [18,39,40]. Moreover, transgenic expression
of Klf5 in murine esophageal epithelia in vivo results in increased
basal cell proliferation . Surprisingly, transgenic expression of
Klf5 did not significantly alter intestinal morphology, cell
proliferation, or cell lineage allocation, suggesting that endogenous
levels of KLF5 are sufficient to control these processes. Interest-
ingly, KLF5 has also been reported to have oncogenic properties
and to mediate transformation by K-Ras during intestinal
tumorigenesis . Yet, no dysplasia, polyps, or cancers developed
in Villin-Klf5 mice up to at least 12 months of age, suggesting that
KLF5 alone is not sufficient to induce cancer in the intestine.
Thus, context is important for KLF5 function, and combined with
our recent finding that KLF5 loss in the context of p53 ablation
drives invasive progression of squamous cell cancer , it would
be interesting to examine whether similar cooperativity exists in
Patients with inflammatory bowel disease (IBD) show evidence
of intestinal mucosal barrier disruption, prolonged inflammatory
response, and severe mucosal lesions . Therefore, a better
understanding of the pathways involved in the regeneration of the
intestinal mucosa following injury is crucial. Several lines of
evidence suggest that KLF5 plays a role in response to injuries.
KLF5 is an important regulator of cardiovascular remodeling ,
and keratinocyte migration , and mice with Klf5 haploinsuffi-
Average # of BrdU
positive cells per crypt
Average # of enteroendocrine
cells per crypt/villus unit
Average # of goblet cells
per crypt/villus unit
Average # of Ki-67
positive cells per crypt
Figure 2. Villin-Klf5 transgenic mice had normal-appearing intestinal mucosa. (A–B) Compared with 1 year-old littermate controls (A),
hematoxylin and eosin (H&E)-stained ileal epithelia, along with underlying lamina propria and muscular layers, appeared normal in 1 year-old Villin-
Klf5 mice (B). Findings were similar in duodeum, jejunum, and colon of Villin-Klf5 mice. Scale bars: 50 mm. (C–D) Quantitation of BrdU-labeled (C) or
Ki-67 positive (D) cells revealed no changes in proliferation in the small intestine or colon of Villin-Klf5 mice at 1 year of age, compared to littermate
controls. (E) The number of chromogranin-A positive enteroendocrine cells per crypt-villus unit was unchanged between control and Villin-Klf5 mice
at 1 year of age. (F) By Alcian blue staining, no difference was observed in the number of goblet cells within small intestine or colon of 1 year-old
control and Villin-Klf5 mice. Similar results were observed for proliferation and differentiation of intestinal cell types in 1 month-, 3 month-, and
6 month-old mice. D= Duodenum, J=Jejunum, I=Ileum, C=Colon.
KLF5 Protects against Colitis In Vivo
PLoS ONE | www.plosone.org4May 2012 | Volume 7 | Issue 5 | e38338
ciency exhibit greater sensitivity to DSS-induced colonic injury
 but are protected from renal injuries . Taken together,
these observations show the importance of studying the role of
KLF5 in different cell types, tissues, and context. The results
presented here demonstrate that epithelial-specific expression of
KLF5 protects against the development of colitis in vivo. The
mechanism for this resistance to colitis may be attributable to
increased IL-22/JAK2/STAT3 signaling, since recent studies
indicate that activation of STAT3 within the intestinal epithelium
plays an important role in mucosal tissue repair through secretion
of IL-22 from innate immune cells [8,9,10,11,12]. Moreover,
JAK/STAT signaling has been implicated in intestinal homeosta-
Disease activity index
Figure 3. Villin-Klf5 mice are less susceptible to injury following
DSS treatment. (A–G) Eight-week-old mice were treated with 3.5%
DSS for 7 days. (A) Compared to littermate controls (Con), Villin-Klf5
mice (Vil-Klf5) had decreased sensitivity to experimental colitis with
DSS, as assessed by disease activity index (*p=0.005; n=6 pairs). (B, C)
DSS-treated control mice (B) had reduced intestinal epithelial repair and
increased susceptibility to colitis, compared to Villin-Klf5 mice (C). Note
the preservation of crypt architecture in Villin-Klf5 mice. (D–G)
Immunofluorescence confirmed that, at day 0, compared to controls
(D), Villin-Klf5 mice (E) had increased KLF5 expression in colonic
epithelial cells. At day 7 of DSS treatment, compared to controls (F),
nuclear KLF5 expression in colonic epithelia of Villin-Klf5 mice was
markedly increased (G). (H, I) Compared to colonic mucosa from
control mice (H), colonic mucosa of Villin-Klf5 mice (I) demonstrated
increased numbers of BrdU-positive cells adjacent to the sites of
ulceration. Scale bars: 50 mm.
Cn Vl Cn Vl Cn Vl
Day 0 Day 3 Day 7
0022446 6 24 24
Figure 4. Klf5 overexpression in intestinal epithelia increases
STAT3 phosphorylation in vivo. (A, B) Immunohistochemistry of
colonic mucosa from control (A) and Villin-Klf5 (B) mice following DSS
treatment revealed more phosphorylated STAT3 in regenerating colonic
epithelial cells from Villin-Klf5 mice. (C) Western blot of colonic
epithelial scrapings from control (Cn) and Villin-Klf5 (Vl) mice at the
indicated time points following DSS treatment revealed increased
phospho-STAT3 and phospho-JAK2 in Villin-Klf5 mice at day 7, while
phospho-STAT3 was decreased in Villin-Klf5 mice at day 0 and
unchanged at day 3. (D) Western blots confirmed increased KLF5
expression in IEC-6 cells transduced with pFB-Klf5 compared to pFB-neo.
However, following linear wounding, no induction of phosphorylated
STAT3 was observed in pFB-Klf5 infected cells, compared to control pFB-
neo infected cells.
Relative IL-22R1α mRNA
Figure 5. Klf5 overexpression increases IL-22 signaling in the
colonic microenvironment following DSS treatment. (A) By
quantitative real-time PCR, Villin-Klf5 mice had increased IL-22 mRNA
expression in the colonic stromal compartment after 7 days of DSS
treatment, compared to controls, while IL-6 mRNA expression was
unchanged. (B) Quantitative real-time PCR of epithelial scrapings from
control and Villin-Klf5 mice treated for 7 days with DSS showed
increased IL-22R1a expression in the epithelia of Villin-Klf5 mice.
KLF5 Protects against Colitis In Vivo
PLoS ONE | www.plosone.org5 May 2012 | Volume 7 | Issue 5 | e38338
sis by regulating proliferation of intestinal stem cells , raising
the possibility that KLF5 induction could play a role in
upregulating intestinal stem cell proliferation following injury.
Following DSS treatment, we observed increased intestinal
epithelial cell proliferation and localized STAT3 activation; this
STAT3 activation was confined to crypts adjacent to the wounded
areas. These results correlated with an increase in IL-22 mRNA
levels in the stroma and increased IL-22R in the intestinal epithelia
of Villin-Klf5 mice.
STAT3 activation can be triggered by numerous cytokines ,
and the context of STAT3 activation for inflammation is
particularly important. For example, STAT3 activation in
intestinal epithelial cells protects against the development of colitis
and enhances intestinal restitution, findings consistent with our
observations [10,12,45]. STAT3 deletion in macrophages and
neutrophils produces chronic enterocolitis , while STAT3
expression in T cells is essential for the induction of colitis . In
contrast to our findings, STAT3 activation was reported to
promote colitis and prevent wound healing , but this study
does not discriminate in which cell-type STAT3 is hyperactivated.
In aggregate, these studies highlight a cell-type specific role of
STAT3 in the regulation of mucosal healing following colitis.
Interestingly, STAT3 activation in enterocytes may also increase
the development of colitis-associated cancer ; thus the long-
term effects of KLF5-mediated STAT3 induction in colitis merit
Our results also demonstrate that STAT3 activation by KLF5
following DSS treatment is non cell-autonomous. We propose a
model in which KLF5 protects following colonic injury through
secretion of IL-22 by infiltrating immune cells and subsequent
activation of STAT3 signaling (Figure 6). KLF5 can stimulate the
expression of several different cytokines and chemokines [33,34],
and we propose that increased KLF5 expression in intestinal
epithelial cells leads to the induction and elaboration of specific
chemokines by these cells. In turn, these chemokines recruit innate
immune cells secreting IL-22, and IL-22 binds to the IL-22
receptor to activate STAT3 within intestinal epithelial cells. The
ability to augment these intestinal repair mechanisms may lead to
improved treatments for diseases characterized by injuries to the
intestinal epithelium. However, additional studies will be needed
to identify the responsible chemokines and better define the link
between KLF5 and IL-22/JAK/STAT3 signaling.
In sum, we have used a novel transgenic mouse model to
demonstrate an essential role for KLF5 in intestinal epithelial
homeostasis and in mucosal healing following induction of
experimental colitis. Following DSS treatment, Klf5 overexpres-
sion is associated with activation of STAT3 signaling in vivo.
Importantly, there are no obvious consequences of Klf5 overex-
pression on intestinal homeostasis in untreated mice up to
12 months of age. Thus KLF5 and STAT3 signaling provide
important areas for further study in colitis and represent potential
diagnostic and therapeutic targets for IBD and other diseases
characterized by injury and disruption of intestinal epithelia.
Conceived and designed the experiments: MT JPK. Performed the
experiments: MT RA MM. Analyzed the data: MT RA MM JPK. Wrote
the paper: MT JPK. Obtained funding for the experiments: JPK.
1. Iizuka M, Konno S (2011) Wound healing of intestinal epithelial cells.
World J Gastroenterol 17: 2161–2171.
Loftus EV Jr. (2004) Clinical epidemiology of inflammatory bowel disease:
Incidence, prevalence, and environmental influences. Gastroenterology 126:
Dignass AU (2001) Mechanisms and modulation of intestinal epithelial repair.
Inflamm Bowel Dis 7: 68–77.
Dignass AU, Baumgart DC, Sturm A (2004) Review article: the aetiopatho-
genesis of inflammatory bowel disease – immunology and repair mechanisms.
Aliment Pharmacol Ther 20 Suppl 4: 9–17.
Sturm A, Dignass AU (2008) Epithelial restitution and wound healing in
inflammatory bowel disease. World J Gastroenterol 14: 348–353.
Podolsky DK (1997) Healing the epithelium: solving the problem from two sides.
J Gastroenterol 32: 122–126.
Rawlings JS, Rosler KM, Harrison DA (2004) The JAK/STAT signaling
pathway. J Cell Sci 117: 1281–1283.
Alonzi T, Newton IP, Bryce PJ, Di Carlo E, Lattanzio G, et al. (2004) Induced
somatic inactivation of STAT3 in mice triggers the development of a fulminant
form of enterocolitis. Cytokine 26: 45–56.
Gibson DL, Montero M, Ropeleski MJ, Bergstrom KS, Ma C, et al. (2010)
Interleukin-11 reduces TLR4-induced colitis in TLR2-deficient mice and
restores intestinal STAT3 signaling. Gastroenterology 139: 1277–1288.
10. Goldsmith JR, Uronis JM, Jobin C (2011) Mu opioid signaling protects against
acute murine intestinal injury in a manner involving Stat3 signaling. Am J Pathol
11. Neufert C, Pickert G, Zheng Y, Wittkopf N, Warntjen M, et al. (2010)
Activation of epithelial STAT3 regulates intestinal homeostasis. Cell Cycle 9:
12. Pickert G, Neufert C, Leppkes M, Zheng Y, Wittkopf N, et al. (2009) STAT3
links IL-22 signaling in intestinal epithelial cells to mucosal wound healing. J Exp
Med 206: 1465–1472.
13. Aaronson DS, Horvath CM (2002) A road map for those who don’t know JAK-
STAT. Science 296: 1653–1655.
14. Lin S, Wang D, Iyer S, Ghaleb AM, Shim H, et al. (2009) The absence of LPA2
attenuates tumor formation in an experimental model of colitis-associated
cancer. Gastroenterology 136: 1711–1720.
15. McConnell BB, Ghaleb AM, Nandan MO, Yang VW (2007) The diverse
functions of Kru ¨ppel-like factors 4 and 5 in epithelial biology and pathobiology.
Bioessays 29: 549–557.
16. Ohnishi S, Laub F, Matsumoto N, Asaka M, Ramirez F, et al. (2000)
Developmental expression of the mouse gene coding for the Kru ¨ppel-like
transcription factor KLF5. Dev Dyn 217: 421–429.
17. Conkright MD, Wani MA, Anderson KP, Lingrel JB (1999) A gene encoding an
intestinal-enriched member of the Kru ¨ppel-like factor family expressed in
intestinal epithelial cells. Nucleic Acids Res 27: 1263–1270.
18. Sun R, Chen X, Yang VW (2001) Intestinal-enriched Kru ¨ppel-like factor (Kru ¨ppel-
like factor 5) is a positive regulator of cellular proliferation. J Biol Chem 276:
19. Yang Y, Goldstein BG, Nakagawa H, Katz JP (2007) Kru ¨ppel-like factor 5
activates MEK/ERK signaling via EGFR in primary squamous epithelial cells.
Faseb J 21: 543–550.
20. Yang Y, Tetreault MP, Yermolina YA, Goldstein BG, Katz JP (2008) Kru ¨ppel-
like factor 5 controls keratinocyte migration via the integrin-linked kinase. J Biol
Chem 283: 18812–18820.
21. Shindo T, Manabe I, Fukushima Y, Tobe K, Aizawa K, et al. (2002) Kru ¨ppel-like
zinc-finger transcription factor KLF5/BTEB2 is a target for angiotensin II
signaling and an essential regulator of cardiovascular remodeling. Nat Med 8:
Figure 6. Model for the protective role of KLF5 in intestinal
wound healing during colitis. KLF5 induces the expression of
specific cytokines by epithelial cells, leading to the recruitment of
immune cells and elaboration of IL22, which then activates STAT3
signaling within the colonic epithelial cells.
KLF5 Protects against Colitis In Vivo
PLoS ONE | www.plosone.org6May 2012 | Volume 7 | Issue 5 | e38338
22. Suzuki T, Sawaki D, Aizawa K, Munemasa Y, Matsumura T, et al. (2009) Download full-text
Kru ¨ppel-like factor 5 shows proliferation-specific roles in vascular remodeling,
direct stimulation of cell growth, and inhibition of apoptosis. J Biol Chem 284:
23. McConnell BB, Kim SS, Bialkowska AB, Yu K, Sitaraman SV, et al. (2011)
Kru ¨ppel-like factor 5 protects against dextran sulfate sodium-induced colonic
injury in mice by promoting epithelial repair. Gastroenterology 140: 540–549
24. Nandan MO, McConnell BB, Ghaleb AM, Bialkowska AB, Sheng H, et al.
(2008) Kru ¨ppel-like factor 5 mediates cellular transformation during oncogenic
KRAS-induced intestinal tumorigenesis. Gastroenterology 134: 120–130.
25. Nandan MO, Yoon HS, Zhao W, Ouko LA, Chanchevalap S, et al. (2004)
Kru ¨ppel-like factor 5 mediates the transforming activity of oncogenic H-Ras.
Oncogene 23: 3404–3413.
26. Goldstein BG, Chao HH, Yang Y, Yermolina YA, Tobias JW, et al. (2007)
Overexpression of Kru ¨ppel-like factor 5 in esophageal epithelia in vivo leads to
increased proliferation in basal but not suprabasal cells. Am J Physiol
Gastrointest Liver Physiol 292: G1784–1792.
27. Madison BB, Dunbar L, Qiao XT, Braunstein K, Braunstein E, et al. (2002) Cis
elements of the villin gene control expression in restricted domains of the vertical
(crypt) and horizontal (duodenum, cecum) axes of the intestine. J Biol Chem 277:
28. Cooper HS, Murthy SN, Shah RS, Sedergran DJ (1993) Clinicopathologic study
of dextran sulfate sodium experimental murine colitis. Lab Invest 69: 238–249.
29. Yang Y, Goldstein BG, Chao HH, Katz JP (2005) KLF4 and KLF5 regulate
proliferation, apoptosis and invasion in esophageal cancer cells. Cancer Biol
Ther 4: 1216–1221.
30. Tetreault MP, Chailler P, Beaulieu JF, Rivard N, Menard D (2008) Epidermal
growth factor receptor-dependent PI3K-activation promotes restitution of
wounded human gastric epithelial monolayers. J Cell Physiol 214: 545–557.
31. Chanchevalap S, Nandan MO, McConnell BB, Charrier L, Merlin D, et al.
(2006) Kru ¨ppel-like factor 5 is an important mediator for lipopolysaccharide-
induced proinflammatory response in intestinal epithelial cells. Nucleic Acids
Res 34: 1216–1223.
32. Frigo DE, Sherk AB, Wittmann BM, Norris JD, Wang Q, et al. (2009) Induction
of Kru ¨ppel-like factor 5 expression by androgens results in increased CXCR4-
dependent migration of prostate cancer cells in vitro. Mol Endocrinol 23:
33. Kumekawa M, Fukuda G, Shimizu S, Konno K, Odawara M (2008) Inhibition
of monocyte chemoattractant protein-1 by Kru ¨ppel-like factor 5 small interfering
RNA in the tumor necrosis factor- alpha-activated human umbilical vein
endothelial cells. Biol Pharm Bull 31: 1609–1613.
34. Fujiu K, Manabe I, Nagai R (2011) Renal collecting duct epithelial cells regulate
inflammation in tubulointerstitial damage in mice. J Clin Invest 121: 3425–3441.
35. Bollrath J, Greten FR (2009) IKK/NF-kappaB and STAT3 pathways: central
signalling hubs in inflammation-mediated tumour promotion and metastasis.
EMBO Rep 10: 1314–1319.
36. Katz JP, Kaestner KH (2002) Cellular and molecular mechanisms of
carcinogenesis. Gastroenterol Clin North Am 31: 379–394.
37. Karam SM (1999) Lineage commitment and maturation of epithelial cells in the
gut. Front Biosci 4: D286–298.
38. Turner JR (2009) Intestinal mucosal barrier function in health and disease. Nat
Rev Immunol 9: 799–809.
39. Bateman NW, Tan D, Pestell RG, Black JD, Black AR (2004) Intestinal tumor
progression is associated with altered function of KLF5. J Biol Chem 279:
40. Chanchevalap S, Nandan MO, Merlin D, Yang VW (2004) All-trans retinoic
acid inhibits proliferation of intestinal epithelial cells by inhibiting expression of
the gene encoding Kru ¨ppel-like factor 5. FEBS Lett 578: 99–105.
41. Yang Y, Nakagawa H, Tetreault MP, Billig J, Victor N, et al. (2011) Loss of
Transcription Factor KLF5 in the Context of p53 Ablation Drives Invasive
Progression of Human Squamous Cell Cancer. Cancer Res 71: 6475–6484.
42. Podolsky DK (1991) Inflammatory bowel disease (1). N Engl J Med 325:
43. Jiang H, Patel PH, Kohlmaier A, Grenley MO, McEwen DG, et al. (2009)
Cytokine/Jak/Stat signaling mediates regeneration and homeostasis in the
Drosophila midgut. Cell 137: 1343–1355.
44. Murray PJ (2007) The JAK-STAT signaling pathway: input and output
integration. J Immunol 178: 2623–2629.
45. Bollrath J, Phesse TJ, von Burstin VA, Putoczki T, Bennecke M, et al. (2009)
gp130-mediated Stat3 activation in enterocytes regulates cell survival and cell-
cycle progression during colitis-associated tumorigenesis. Cancer Cell 15:
46. Takeda K, Clausen BE, Kaisho T, Tsujimura T, Terada N, et al. (1999)
Enhanced Th1 activity and development of chronic enterocolitis in mice devoid
of Stat3 in macrophages and neutrophils. Immunity 10: 39–49.
47. Durant L, Watford WT, Ramos HL, Laurence A, Vahedi G, et al. (2010)
Diverse targets of the transcription factor STAT3 contribute to T cell
pathogenicity and homeostasis. Immunity 32: 605–615.
48. Suzuki A, Hanada T, Mitsuyama K, Yoshida T, Kamizono S, et al. (2001)
CIS3/SOCS3/SSI3 plays a negative regulatory role in STAT3 activation and
intestinal inflammation. J Exp Med 193: 471–481.
KLF5 Protects against Colitis In Vivo
PLoS ONE | www.plosone.org7 May 2012 | Volume 7 | Issue 5 | e38338