Download Dysregulation of intestinal crypt cell proliferation and villus cell

Am J Physiol Gastrointest Liver Physiol 292: G1757–G1769, 2007.
First published March 22, 2007; doi:10.1152/ajpgi.00013.2007.
Dysregulation of intestinal crypt cell proliferation and villus cell migration
in mice lacking Krüppel-like factor 9
Frank A. Simmen,1,2 Rijin Xiao,1,2 Michael C. Velarde,1,2 Rachel D. Nicholson,2
Margaret T. Bowman,2 Yoshiaki Fujii-Kuriyama,3 S. Paul Oh,4 and Rosalia C. M. Simmen1,2
Department of 1Physiology and Biophysics and 2Arkansas Children’s Nutrition Center, University of Arkansas for Medical
Sciences, Little Rock, Arkansas; 3TARA Center, University of Tsukuba, 1-1-1 Tennodai Tsukuba 305-8577, Japan;
and 4Department of Physiology and Functional Genomics, University of Florida, Gainesville, Florida
Submitted 6 January 2007; accepted in final form 14 March 2007
Krüppel-like; Igfbp4; Ptk6; stem cells; colon; smooth muscle
KRüPPEL-LIKE FACTOR 9
(Klf9), previously designated basic transcription element binding protein (Bteb) 1, is a transcriptional
regulator whose primary sequence is highly conserved among
vertebrate organisms (11, 44). Klf9 belongs to a family of
transcriptional mediators [specificity protein 1 (SP1)-like/
Krüppel-like factors, SP/KLF family], defined by the presence
of an 81-amino-acid DNA-binding region located in the carboxy terminus and comprised of three contiguous C2-H2 zinc
fingers (2, 12, 38). There are 25 SP/KLF genes/proteins known
at present. SP/KLF proteins bind to GC/GT boxes (consensus
binding site: 5⬘-NGGGNGNGG-3⬘) resident in promoter and
enhancer/silencer regions of multiple chromosomal genes (38).
Although each SP/KLF family member is homologous to all
others by virtue of the Krüppel-like DNA-binding domain,
each is unique with respect to the sequence amino terminal to
this domain. Some KLFs are widely expressed in multiple
tissues, whereas others are more restricted in their expression
(38). KLFs function as transcriptional activators and/or repressors depending on target gene, cis element(s), and tissue
context and by partnering with other nuclear proteins (12, 49).
Data, albeit limited, suggest nonoverlapping functional roles
for SP/KLF proteins in tissue growth, morphogenesis, and stem
cell biology (38).
Address for reprint requests and other correspondence: F. A. Simmen,
Arkansas Children’s Nutrition Center, 1120 Marshall St., Little Rock, AR
72202 (e-mail: [email protected]).
http://www.ajpgi.org
Klf9 was first isolated as a transcriptional inducer of the
hepatic CYP1A1 gene (11). Later work identified stimulatory
roles in vitro for Klf9 in neuronal cell differentiation (7) and in
endometrial cell proliferation (35, 48). In support of the latter,
subsequent studies described a role for Klf9, in concert with
progesterone receptor, in determining the program of endometrial secretory protein gene expression (34, 36, 42, 47). Studies
of the Klf9-null mutant (Klf9⫺/⫺) mouse have borne out findings from in vitro studies, namely that this transcription factor
has a functional role in the developing brain (mutant mice have
deficits in cerebellum function as gauged by certain behavioral
tests) (22) and in female reproductive function, the latter at the
level of the maternal uterus (37, 41). Female Klf9⫺/⫺ mice
exhibit reduced numbers of implantation sites, smaller uteri,
and developmental asynchrony of embryos and uterus and as a
consequence are subfertile compared with wild-type (WT)
dams (37, 41).
While elucidating the reproductive phenotype of Klf9 mutant
mice, we observed a significant degree of mortality of Klf9⫺/⫺
newborn mice as well as mild growth retardation of preweaning Klf9⫺/⫺ mice relative to WT counterparts (37). This,
coupled with the well-characterized roles for two other Klf
proteins (Klf4, Klf5) in intestinal crypt-villus growth and
differentiation (5, 30, 31), prompted us to characterize the
expression and in vivo function of Klf9 in mouse intestine
during postnatal ontogeny. In this report, we document the
expression of the Klf9 gene in small and large intestine smooth
muscle and describe a subtle, albeit significant, intestine mucosal phenotype in mice lacking this gene. In addition, we
identify potential downstream (direct and indirect) gene targets
and pathways of Klf9 by using microarrays. Our results suggest that Klf9 controls elaboration, from intestine smooth
muscle, of molecular mediator(s) of crypt cell proliferation,
villus cell migration, and Paneth and goblet cell differentiation.
MATERIALS AND METHODS
Animals. Animal use protocols were approved by the Institutional
Animal Care and Use Committee at the University of Arkansas for
Medical Sciences. Klf9 mutant mice (Klf9lacZ) in the C57BL/6J
background were previously generated by insertion of the bacterial
␤-galactosidase (LacZ) gene in frame within exon 1 of the mouse Klf9
gene (22). Mice were maintained on a 12-h light/12-h dark schedule
with ad libitum access to food and water. Animals were genotyped as
previously described (37).
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.
0193-1857/07 $8.00 Copyright © 2007 the American Physiological Society
G1757
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
Simmen FA, Xiao R, Velarde MC, Nicholson RD, Bowman
MT, Fujii-Kuriyama Y, Oh SP, Simmen RC. Dysregulation of
intestinal crypt cell proliferation and villus cell migration in mice
lacking Krüppel-like factor 9. Am J Physiol Gastrointest Liver
Physiol 293: G1757–G1769, 2007. First published March 22, 2007;
doi:10.1152/ajpgi.00013.2007.—Krüppel-like factor 9 (Klf9), a zincfinger transcription factor, is implicated in the control of cell proliferation, cell differentiation, and cell fate. Using Klf9-null mutant
mice, we have investigated the involvement of Klf9 in intestine
crypt-villus cell renewal and lineage determination. We report
the predominant expression of Klf9 gene in small and large intestine
smooth muscle (muscularis externa). Jejunums null for Klf9 have
shorter villi, reduced crypt stem/transit cell proliferation, and altered
lineage determination as indicated by decreased and increased numbers of goblet and Paneth cells, respectively. A stimulatory role for
Klf9 in villus cell migration was demonstrated by bromodeoxyuridine
labeling. Results suggest that Klf9 controls the elaboration, from
intestine smooth muscle, of molecular mediator(s) of crypt cell proliferation and lineage determination and of villus cell migration.
G1758
Klf 9 IN INTESTINAL GROWTH
a constant target intensity of 500, and the data from each GeneChip
were log2 transformed and normalized by using median values within
each treatment group (genotype) (13). Differentially expressed transcripts were identified by the t-test function in Spotfire DecisionSite
(Somerville, MA) using P ⬍ 0.05. Differentially expressed transcripts
were further filtered by using a fold-change cutoff of 1.3, a value
approaching the practical limit for detectable differences with
Affymetrix microarrays (28). An additional filter required that all
transcripts expressed at higher levels in WT jejunum had to be called
“present” on all five corresponding GeneChips, whereas transcripts
expressed at higher levels in Klf9⫺/⫺ jejunums had to be called
“present” on all five corresponding GeneChips. Unsupervised nearestneighbor hierarchical clustering was performed by using Spotfire
DecisionSite. Gene lists were annotated by using NETAFFX (http://
www.affymetrix.com/analysis/index.affx), Gene Ontology (http://
www.geneontology.org/), and NCBI (http://www.ncbi.nlm.nih.gov/).
The complete microarray dataset was deposited in Gene Expression
Omnibus (http://www.ncbi.nlm.nih.gov/geo/) under accession number
GSE6443.
Real time RT-PCR. Primer sequences designed by using Primer
Express (Applied Biosystems, Foster City, CA) were as follows: Klf9,
forward 5⬘-CGTTGCCCACTGTGTGAGAA-3⬘, reverse 5⬘-TTGATCATGCTGGGATGGAA-3⬘; Klf5, forward 5⬘-TCCGTCCTATGCCGCTACAA-3⬘, reverse 5⬘-CCAGATCCGGGTTACTCCTTCT-3⬘;
Klf13, forward 5⬘-ACACAGGTGAGAGGCCTTTCG-3⬘, reverse 5⬘AGCATGCCTGGGTGGAAG-3⬘; Klf16, forward 5⬘-CCTTACTCCCACTGGGTTAGGG-3⬘, reverse 5⬘-AGCACATGACGGCAGACCA-3⬘; Klf4, forward 5⬘-AGAGGAGCCCAAGCCAAAGA-3⬘,
reverse 5⬘-AGTTCGCAGGTGTGCCTTGA-3⬘; IGF binding protein
gene Igfbp4, forward 5⬘-CCAAACTGTGACCGCAACG-3⬘, reverse
5⬘-CCAAACCCCCAGGAAGCTT-3⬘; protein tyrosine kinase gene
Ptk6, forward 5⬘-TCCCAAGTGCTGGGATCAAA-3⬘, reverse 5⬘TCCACAAGGCCTGTTGCCTA-3⬘; pituitary homeobox 2 (Pitx2),
forward 5⬘-AACCTTACGGAAGCCCGAGTC-3⬘, reverse 5⬘CCCAAAGCCATTCTTGCACA-3⬘; cyclin D2 gene Ccnd2, forward
5⬘-CTTTGTGGTAGGACGGTGGGT-3⬘, reverse 5⬘-TGTGCAGTGCGTGAGCTCTG-3⬘; myotubularin 1 (Mtm1), forward 5⬘TGTCTCAAGATGGAGTCAGT-3⬘, reverse 5⬘-GACCATAGGAATTTTCTCCTC-3⬘; CEA-related cell adhesion molecule gene
Ceacam1, forward 5⬘-TTGTTGTCTTCAGCAACCTGG-3⬘, reverse
5⬘-AGGACTACTGCTCACAGCCTC-3⬘; Igfbp5, forward 5⬘GCTCGCCGTAGCTCTTTTC-3⬘, reverse 5⬘-GGTTCTTTCGTGCACTGTGA-3⬘; cyclophilin A gene Ppia, forward 5⬘-TGTGCCAGGGTGGTGACTTTA-3⬘, reverse 5⬘-AGATGCCAGGACCTGTATGCTT-3⬘. One microgram of RNA from each jejunum was
reverse transcribed by using random hexamers and MultiScribe Reverse Transcriptase (Applied Biosystems). Real-time PCR was performed with an ABI Prism 7000 Sequence Detector as described
previously (45).
Statistical analysis. Statistical analysis was performed by using
SigmaStat for Windows (SPSS, Chicago, IL). Values are presented as
means ⫾ SE or as box plots. Differences between treatment means
were considered significant at P ⬍ 0.05, whereas 0.05 ⬍ P ⬍ 0.1
indicated a tendency for a difference.
RESULTS
Effects of Klf9⫺/⫺ on gastrointestinal tissue weight at postweaning. We previously observed reduced body weights for
young Klf9⫺/⫺ mice (37). To elucidate the physiological basis
for this observation, the gastrointestinal tracts of Klf9⫺/⫺ and
heterozygous mutant Klf9⫹/⫺ male mice and their corresponding WT counterparts were evaluated at early postweaning. At
28 days of age, Klf9⫺/⫺ mice had reductions in body weight
compared with heterozygous mutant and WT counterparts
(Table 1), in agreement with our previous report (37). Gastro-
AJP-Gastrointest Liver Physiol • VOL
292 • JUNE 2007 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
Tissue collection. The small intestine was divided into three equal
segments that were operationally defined as duodenum, jejunum, and
ileum. Tissue samples from the midpoint of jejunum and ileum were
placed in formalin for later histochemistry or immunohistochemistry
analyses or were cryopreserved (see below) for 5-bromo-4-chloro-3indolyl-␤-D-galactopyranoside (X-gal) staining. The colon was divided into two halves, proximal and distal, and the midpoint of each
half-segment was fixed in 10% neutral-buffered formalin or was
cryopreserved. Fixed tissues were subsequently embedded in paraffin.
Histochemistry. Fixed tissues were embedded in paraffin. Sections
(5 ␮m) were deparaffinized in xylene, rehydrated through a series of
alcohols, stained with hematoxylin and eosin (H&E), and coverslipped. Differentiated cell types within intestinal crypts and villi were
identified by histochemistry. Grimelius stain (18) was used to identify
enteroendocrine cells; the Lendrum’s phloxine-tartrazine stain/procedure (1) was used to identify Paneth cells; and Alcian Blue (pH 2.5)
(3),mucicarmine(http://www.ihcworld.com/_protocols/special_stains/
southgate_mucicarmine_ellis.htm), or H&E stains were used for goblet cell identification. For purposes of X-gal staining, freshly isolated
gastrointestinal tissues were soaked in PBS containing 20% sucrose,
embedded in optimum cutting temperature medium (Fisher Scientific), and frozen in liquid nitrogen. Sections (7 ␮m) were cut by using
a cryostat, postfixed in PBS containing 2% paraformaldehyde and
0.2% glutaradehyde for 5 min, and then stained with X-gal (1 mg/ml)
at 37°C overnight, followed by counterstaining with neutral red (1%)
in sodium acetate (50 mM, pH 3.3). Lengths of villi, depths of crypts,
and muscularis thicknesses were measured by using MCID software
(Interfocus, Linton, UK).
Immunohistochemistry. Thirty-day-old male mice were injected
intraperitoneally with 150 ␮l of bromodeoxyuridine (BrdU) labeling
reagent (Zymed Laboratories, South San Francisco, CA) at 2 and 48 h
before tissue collection. Tissues were fixed in 10% neutral-buffered
formalin and were embedded in paraffin. Sections (4 ␮m) were cut,
applied to ProbeOn Plus slides, dewaxed in xylene, and rehydrated
through a series of graded alcohols. Antigen retrieval was done in 1⫻
Citra-plus (Biogenics, Napa, CA). After being cooled, slides were
washed in 1⫻ PBS and H2O, quenched in 3% H2O2, and blocked with
10% donkey serum-1⫻ PBS. Anti-BrdU mouse IgG1 monoclonal
antibody (Roche Diagnostics, Indianapolis, IN) was diluted in blocking buffer (1:500) and was incubated on slides overnight at 4°C. After
a wash with PBS, biotin-SP-conjugated affinity-purified F(ab⬘)2 fragment donkey anti-mouse antibody (Jackson ImmunoResearch Laboratories, West Grove, PA) was diluted (1:100) in blocking buffer and
was applied to the slides for 30 min at 37°C. Fluorescein (DTAF)conjugated streptavidin-RITC (Jackson ImmunoResearch) was diluted in blocking buffer (1:100) and was applied to slides for 30 min
at 37°C. After being washed, slides were counterstained with 0.01%
Evans blue and were viewed on an Axiovert 200M microscope.
Images were captured by AxioCam HRc (Zeiss Oberkochen) and
were processed by Axiovision software release 4.5 SPI (03-2006)
(Zeiss Oberkochen). BrdU-labeled cells in crypts were counted manually in a blinded fashion. Immunohistochemistry of PCNA followed standard methodologies described previously for other antigens (41, 45). Mouse monoclonal antibody to PCNA was from
Dako (Carpinteria, CA).
Microarrays. Total RNA was extracted in parallel from the jejunums of five WT and five Klf9⫺/⫺ male mice at postnatal day (PND)
30 by using Trizol reagent (Invitrogen, Carlsbad, CA). Conversion of
each RNA preparation to the corresponding fragmented cRNA probe
was as previously described (45). Fifteen micrograms of each cRNA
were hybridized for 16 h to an Affymetrix mouse 430A GeneChip.
Ten GeneChips (each corresponding to a single animal) were hybridized, washed, and scanned in parallel. Following the wash, signalamplification, and signal-detection steps, GeneChips were scanned
(Agilent GeneArray laser scanner) and the resultant images were
quantified by using Affymetrix MAS 5.0 software. The average of the
fluorescent intensities of all probe sets on a given array was scaled to
G1759
Klf 9 IN INTESTINAL GROWTH
Table 1. Body weights and gastrointestinal tissue weights of postweaning male mice
WT
Klf9⫹/⫺
Klf9⫺/⫺
Body wt (n)
Stomach wt (n)
Small Intestine wt (n)
Colon wt (n)
11.78⫾1.19 (16)
11.84⫾2.03 (26)
9.45⫾2.61a,b (35)
0.125⫾0.005 (5)
0.135⫾0.0027 (4)
0.097⫾0.0047a,b (8)
0.723⫾0.041 (5)
0.762⫾0.044 (4)
0.643⫾0.035 (8)
0.192⫾0.01 (5)
0.205⫾0.012 (4)
0.146⫾0.013c,d (8)
Values are means ⫾ SE on postnatal day (PND) 28 given in grams; n ⫽ number of mice examined. Effects of genotype on body weight were examined by
ANOVA; organ weight was normalized to body weight, and differences were analyzed by multiple-comparison Holm-Sidak test. aP ⬍ 0.01 vs. wild-type (WT);
b
P ⬍ 0.001 vs. Krüppel-like factor 9 (Klf9)⫹/⫺; cP ⬍ 0.05 vs. WT; dP ⬍ 0.01 vs. Klf9⫹/⫺.
Effects of Klf9 gene ablation on intestinal crypts and villi.
We measured small intestine villi and crypts as indices of
mucosal tissue growth. In jejunums of male mice at 4 and 6 wk
of age, absence of Klf9 resulted in significant reductions (by up
to 40%) in villus length, without any alteration in crypt depth
(Fig. 1, A–D). Interestingly, differences in villus length were
not apparent at 6 mo of age, because knockout jejunum villus
lengths had reached that of WT (Fig. 1C). At 6 mo of age,
Fig. 1. Shorter villi in Krüppel-like factor
(Klf) 9 knockout (Klf9⫺/⫺) mice at 4 wk and
6 wk but not 6 mo of age. Representative
sections of jejunum from wild-type (WT)
[A, postnatal day (PND) 29] and Klf9⫺/⫺
(B, PND 31) males were hematoxylin and
eosin-stained; me, muscularis externa; vi, villus. C and D: villus length and crypt depth
were measured for a minimum of 20 cryptvillus units per slide, with 2–3 slides analyzed per animal. Data were analyzed for
effects of genotype within each age group
(3– 8 animals per age/genotype) with the Student’s t-test. Crypt depths did not differ between genotypes (P ⬎ 0.1) at 4 and 6 wk of
age, but a reduction in length of Klf9⫺/⫺
crypts, which approached statistical significance (P ⫽ 0.06), was found for animals of 6
mo of age. E: 3 or 4 tissue sections from each
WT (n ⫽ 4) and Klf9⫺/⫺ (n ⫽ 3) male mouse
jejunum, respectively (ages ranged from
PND 25 to PND 33) underwent immunohistochemistry for PCNA. F: 3 or 4 sections
from each WT (n ⫽ 2) and Klf9⫺/⫺ (n ⫽ 4)
male mouse ileum, respectively (PND 25–
33) stained for PCNA. PCNA-positive cells
(exhibiting nuclear staining) in crypt-villus
epithelium were counted, and data (means ⫾
SE per crypt-villus unit) were analyzed by
Student’s t-test.
AJP-Gastrointest Liver Physiol • VOL
292 • JUNE 2007 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
intestinal tissue weights (normalized to individual body
weight) of Klf9⫺/⫺ mice were reduced compared with heterozygous mutant and WT counterparts (Table 1). For the
stomach and colon, numerical differences in wet weight were
statistically significant; by contrast, numerical reductions in
small intestine weight of Klf9⫺/⫺ mice were not. Klf9 protein
abundance was reduced and absent in heterozygous and homozygous null jejunums, respectively (data not shown).
G1760
Klf 9 IN INTESTINAL GROWTH
however, knockout jejunum crypt depth tended (P ⫽ 0.06) to
be reduced compared with WT (Fig. 1D). Quantification of
number of cells staining positive for PCNA (a marker primarily
of the crypt transit cell population) at postweaning revealed
decreases (⬃40% for jejunum, ⬃29% for ileum) with absence
of Klf9 (Fig. 1, E and F). The measured width (in crosssection) of the muscularis externa in jejunum was unaffected
by Klf9 gene deletion (data not shown). The data indicate
reductions in villus length that can be predicted to result in
lower capacity for nutrient absorption and hence lower body
weight.
Mitosis of stem cells deep within crypts of the small intestine yields transit cells that further divide (over several generations) to provide villus enterocyte, goblet, and endocrine cell
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
Fig. 2. Lineage determination is perturbed in jejunum villus epithelium of Klf9⫺/⫺ mice. A and B: representative sections of jejunum stained with Lendrum’s
reagent to reveal Paneth cells (arrows). Note the typical granular appearance of Paneth cell cytoplasm. C: quantification of Paneth cells in jejunum crypts. Paneth
cells were counted in a minimum of 20 well-oriented crypts per animal (n ⫽ 3 animals/genotype), and results were analyzed by Student’s t-test. D and E: goblet
cell numbers per crypt and villus, respectively, for WT (n ⫽ 3) and Klf9⫺/⫺ (n ⫽ 3) jejunums at PND 30. Goblet cell numbers within crypt, but not villus,
epithelium were reduced with Klf9 knockout.
AJP-Gastrointest Liver Physiol • VOL
292 • JUNE 2007 •
www.ajpgi.org
G1761
Klf 9 IN INTESTINAL GROWTH
Effects of Klf9 gene deletion on colon histomorphology.
Tissue sections from midpoints of proximal and distal colons
of ⬃ 4-wk-old mice were subjected to H&E staining, followed
by determination of crypt depth, goblet cell number per crypt,
and cross-sectional thickness of the muscularis externa
(Fig. 3, A–C). Results demonstrated a tendency (P ⫽ 0.054) for
reduced (by ⬃12%) colon crypt depths in knockout animals
with no interaction of genotype and colon location. Goblet cell
number per crypt was increased (by 2–3 cells/crypt, P ⬍ 0.05)
in proximal colons of Klf9 KO animals (Fig. 3D), with no such
differences observed for distal colons. There was no difference
in thickness of the muscularis with genotype (P ⫽ 0.928;
Fig. 3E).
Tissue-expression domains of Klf9. In view of the above
results, it was important to establish the cellular location of
Klf9 expression. To address this, we took advantage of the
targeted insertion of the Escherichia coli LacZ gene within
Klf9 exon 1, which allows for in situ staining of Klf9 gene
Fig. 3. A and B: representative hematoxylin and eosin-stained sections of proximal colon from WT (PND 27) and Klf9⫺/⫺ (PND 33) male mice. Crypt depths
and muscularis externa thicknesses were measured for proximal and distal colons (PND 30 ⫾ 3 days; minimum of 40 crypts and 5 muscle measurements per
slide, 1 slide per region/animal). C: data corresponding to proximal colon (n ⫽ 10 WT and n ⫽ 9 Klf 9⫺/⫺ animals) and distal colon (n ⫽ 6 WT and n ⫽ 6 Klf 9⫺/⫺
animals) were analyzed by using 2-way ANOVA. Crypt depths did not differ between colon regions (P ⫽ 0.139), but a tendency for a difference between
genotypes (P ⫽ 0.054, both regions combined) was found. D: in proximal colon only, Klf 9⫺/⫺ crypts had more goblet cells than did WT crypts (P ⫽ 0.015).
E: smooth muscle thickness did not differ by genotype; however, smooth muscle of the proximal colon was thicker than that of the distal colon (P ⫽ 0.018).
AJP-Gastrointest Liver Physiol • VOL
292 • JUNE 2007 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
lineages (cell differentiation commences with migration out of
the crypts) or Paneth cells (migrate to the crypt base). Because
Klf9 knockout resulted in a reduced number of PCNA-positive
cells in small intestine crypt epithelium (Fig. 1E), we examined
if cell lineage determination was consequently perturbed.
Quantification of enteroendocrine cells in tissue sections subjected to Grimelius argyrophil stain revealed no differences in
numbers of this cell type for WT vs. Klf9⫺/⫺ jejunum (data not
shown). Paneth cells were identified histochemically by using
Lendrum’s procedure. Interestingly, a small but significant
increase in number of Paneth cells was found for Klf9⫺/⫺
jejunum crypts compared with WT (Fig. 2, A–C). Klf9⫺/⫺
jejunum manifested fewer numbers of goblet cells in crypt but
not in villus epithelium (Fig. 2, D–E). Thus absence of Klf9 led
to reduced crypt transit cell proliferation (reduced number of
PCNA-positive cells) and subtle changes in goblet and Paneth
cell specification processes/pathways within the small intestine.
G1762
Klf 9 IN INTESTINAL GROWTH
expression with X-gal (22). WT tissues served as negative
controls because they lack the inserted LacZ gene. As expected, WT tissue sections had no X-gal-stained (blue) cells
(Fig. 4A). In Klf9⫹/⫺ and Klf9⫺/⫺ jejunums, X-gal staining was
observed throughout the muscularis externa, with rare stained
cells observed in the lamina propria (Fig. 4, B and C). In colon,
the muscularis externa was strongly stained, as were cells
comprising the surface epithelium (Fig. 4D). Sporadic X-galstained lamina propria cells were apparent also (Fig. 4D). The
identities of the X-gal stained lamina propria cells in small and
large intestines are unknown.
Gene transcripts that are differentially expressed between
WT and Klf9⫺/⫺ small intestines. We next performed gene
expression profiling to identify underlying mechanism(s) of
Klf9 action. Initial efforts used quantitative real-time RT-PCR
to examine relative abundance of key Klf9-related genes in
small and large intestine. We chose Klf5 (previously referred to
as intestinal Klf, Bteb2), Klf13(Bteb3, most similar in sequence
to Klf9), Klf16 (Bteb4), and Klf4 (gut Klf, Gklf) for analysis. In
WT mice, colon Klf9 and Klf4 transcript expression exceeded
that for jejunum (Fig. 5A and E). No effects of Klf9 gene
ablation (heterozygous or homozygous) on Klf gene expression
in jejunum were observed, thus eliminating the possibility of
compensatory expression of one or more of these Klf-related
genes in the tissue (Fig. 5, B–E). In colon, there was a tendency
(P ⫽ 0.073) for a small increase in Klf16 transcript abundance
with loss of Klf9 allele(s) (Fig. 5D). There was no detectable
expression of Klf14 (Bteb5) mRNA in tissue of either WT or
knockout animals (data not shown).
We also employed Affymetrix microarray technology to
monitor global gene expression in postweaning (PND 30) WT
and knockout mouse jejunums. Unsupervised hierarchical clustering of the resultant data demonstrated a major effect of Klf9
on jejunum gene-expression profile (data not shown), in agreement with the phenotypic changes observed above. Loss of
Klf9 affected expression of multiple genes, with inductions
(185 transcripts) and repressions (204 transcripts) in abundance
of mRNAs observed in knockout jejunums relative to WT
(Supplementary Tables S1 and S2; see the online version of
this article for supplemental data). Knockout of Klf9 led to
downregulation of genes encoding cell-signaling proteins, transcription factors, angiogenic/vasculogenic factors, defense/
pathogen-related molecules, cell adhesion and matrix proteins,
cell cycle and migration proteins, Wnt pathway members, and
transporters (partial list in Table 2). Transcripts upregulated
with Klf9 knockout encode signaling molecules and transcription factors, muscle and mitochondrial proteins, adhesion and
proliferation proteins, Wnt pathway members, and transporters
(partial list in Table 3). The increased expression of muscle and
mitochondrial transcripts in knockout jejunum is consistent
with less mucosal mass (i.e., shorter villi) and more smooth
muscle transcript representation in the tissue. The complete
gene lists with fold changes and functional annotation for
AJP-Gastrointest Liver Physiol • VOL
292 • JUNE 2007 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
Fig. 4. Klf9 gene promoter activity in jejunum and colon. A: WT jejunum, PND 22; B: Klf9 heterozygous (Klf9⫹/⫺) jejunum, PND 27; C: Klf9⫺/⫺ jejunum, PND
25. 5-Bromo-4-chloro-3-indolyl-␤-D-galactopyranoside (X-gal)-stained muscularis externa was apparent in mutant but not WT jejunum (B; large arrow, inner
circular layer; small arrows, outer longitudinal layer). Rare X-gal-stained cells were observed within the villus lamina propria (C, arrows) of mutant animals.
D: proximal colon of Klf9⫺/⫺ animals at PND 27, stained with X-gal. Stained cells were prevalent in the luminal surface epithelium (left, large arrow; image
corresponds to upper red-boxed area of middle) and muscularis externa (right, arrows; image corresponds to lower red-boxed area of middle) with lesser staining
observed in the upper crypts and some lamina propria cells. No X-gal-stained cells were observed in negative control (WT) colon (not shown).
G1763
Klf 9 IN INTESTINAL GROWTH
differentially expressed RNA transcripts (1.3-fold cutoff, P ⬍
0.05) are in Supplementary Tables S1 and S2.
Molecular basis for Klf9 knockout small intestine phenotype.
The PCNA immunohistochemistry data suggested that crypt
stem/transit cell proliferation is reduced with Klf9 knockout.
We confirmed this by analyzing BrdU-labeled jejunum of WT
and Klf9⫺/⫺ mice at 2 h after BrdU administration. There were
fewer BrdU-positive cells in knockout than in WT crypts at 2 h
after BrdU treatment (Fig. 6, A–C, G–I, and M), confirming
reduced crypt cell proliferation. Interestingly, however, knockout villi exhibited more BrdU-positive cells than WT villi at
48 h after BrdU (Fig. 6, D–F, J–L, and N). Closely migrating
cohorts of BrdU-labeled cells in villus epithelium were apparent near the termini of WT villi at 48 h (Fig. 6, D–F).
Corresponding cells in knockout villi were not as advanced in
relative position (Fig. 6, J–L and P), demonstrating a reduced
rate of villus epithelial cell migration toward the villus tips.
Staining of lamina propria cells was observed also; however,
this was also apparent for animals that did not receive BrdU
and hence was judged to be nonspecific (Fig. 6O). Terminal
deoxynucleotidyl transferase nick-end labeling staining revealed no differences in numbers of apoptotic cells in villus
epithelium of WT and knockout animals (data not shown).
Six transcripts identified as differentially expressed by microarray were examined by real-time RT-PCR of an expanded
number of PND 30 jejunums (Fig. 7A). Ptk6 and Ceacam1
were confirmed to be more highly expressed in WT than
knockout jejunum, and Ccnd2 (cyclin D2) was confirmed to be
more highly expressed in knockout than WT jejunum, whereas
Pitx2 and Mtm1 did not differ by genotype and are thus false
positives by microarray. Igfbp4 was elevated in knockout
jejunum, in agreement with the microarray results, although the
difference by RT-PCR was not statistically significant. Igfbp5
was not differentially expressed by microarray, and this was
AJP-Gastrointest Liver Physiol • VOL
292 • JUNE 2007 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
Fig. 5. Klf mRNA abundance in WT and Klf9
heterozygous (Klf9⫹/⫺) and Klf9⫺/⫺ mutant mouse
jejunums and proximal colons (PC). A: relative
abundance of Klf9 transcript in WT and Klf9⫹/⫺
jejunum and proximal colon. B–E: relative abundance of Klf5, Klf13, Klf16, and Klf4 mRNAs in
PND 30 male mouse jejunum (n ⫽ 3, 3, and 4 for
WT, heterozygous, and null-mutant animals, respectively) and PC (n ⫽ 3, 4, and 4 for WT,
heterozygous, and null-mutant animals, respectively) was evaluated by real-time quantitative RTPCR. All data were normalized to cyclophilin A
(Ppia) mRNA. There was no detectable amplification of Klf14 mRNA in any samples.
G1764
Klf 9 IN INTESTINAL GROWTH
Table 2. Partial list of transcripts identified by microarray analysis to be downregulated in the jejunum of Klf9⫺/⫺ mice
Protein
Full Name
CEA-related cell adhesion molecule 1
Disabled homolog 2 (Drosophila)
Chemokine (C-X-C motif) ligand 13
Nuclear receptor interacting protein 1
Fyn-related kinase
Diacylglycerol kinase, theta
CDC2-related kinase 7
Protein tyrosine kinase 6
Erbb2 interacting protein
Vascular endothelial zinc finger 1
Alpha thalassemia/mental retardation syndrome X-linked homolog
Trans-acting transcription factor 3
Zinc finger protein 306
X-box binding protein 1
Nuclear receptor subfamily 3, group C, member 1
Transducin-like enhancer of split 4, homolog of Drosophila E (spl)
Transcription factor 12
HECT, UBA, and WWE domain containing 1
Zinc finger protein 36, C3H type-like 1
Stomatin
Vascular endothelial zinc finger 1
CEA-related cell adhesion molecule 1
Phosphatidic acid phosphatase type 2B
Suppressor of cytokine signaling 3
CEA-related cell adhesion molecule 1
CD38 antigen
Natural killer cell group 7 sequence
Serum amyloid A 1
Pancreatitis-associated protein
Regenerating islet-derived 3 gamma
Chemokine (C-X-C motif) ligand 13
Traf2 binding protein
Phospholipase A2, group V
Mucin 13
CD2-associated protein
CEA-related cell adhesion molecule 1
Lin 7 homolog c (Caenorhabditis elegans)
Pancreatitis-associated protein
Cyclin-dependent kinase inhibitor 1B (P27)
Epidermal growth factor receptor pathway substrate 8
Cyclin G1
CEA-related cell adhesion molecule 1
CD2-associated protein
CAP, adenylate cyclase-associated protein 1 (yeast)
Disabled homolog 2 (Drosophila)
Myosin VI
Epidermal growth factor receptor pathway substrate 8
Pancreatitis-associated protein
Regenerating islet-derived 3 gamma
Phosphatidic acid phosphatase type 2B
Disabled homolog 2 (Drosophila)
Transducin-like enhancer of split 4, homolog of Drosophila E (spl)
Retinitis pigmentosa 2 homolog (human)
CAP, adenylate cyclase-associated protein 1 (yeast)
Epidermal growth factor receptor pathway substrate 8
Lin 7 homolog c (C. elegans)
Transferrin receptor
Cystic fibrosis transmembrane conductance regulator homolog
Solute carrier family 40 (iron-regulated transporter), member 1
Solute carrier organic anion transporter family, member 2a1
Solute carrier family 26, member 3
Fold cutoff of 1.3; P ⬍ 0.05. Note, some genes are listed in more than one category due to multiple functions.
AJP-Gastrointest Liver Physiol • VOL
292 • JUNE 2007 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
Signaling molecules
Ceacam1
Dab2
Cxcl13
Nrip1
Frk
Dgkq
Crk7
Ptk6
Erbb2ip
Transcription factors
Vezf1
Atrx
Sp3
Zfp306
Xbp1
Nr3c1
Tle4
Tcf12
Huwe1
Zfp3611
Vasculogenesis
Stom
Vezf1
Ceacam1
Ppap2b
Defense/pathogens/acute phase response/TNF-related/GALT
Socs3
Ceacam1
Cd38
Nkg7
Saa1
Pap
Reg3 g
Cxcl13
T2bp
Pla2 g5
Muc13
Cell adhesion/matrix proteins
Cd2ap
Ceacam1
Lin7c
Cell proliferation/cell cycle
Pap
Cdkn1b
Eps8
Ccng1
Cell migration
Ceacam1
Cd2ap
Cap1
Dab2
Myo6
Eps8
Wnt pathway-related
Pap
Reg3 g
Ppap2b
Dab2
Tle4
Polarized epithelial phenotype
Rp2 h
Cap1
Eps8
Lin7c
Transport
Tfrc
Cftr
Slc40a1
Slco2a1
Slc26a3
G1765
Klf 9 IN INTESTINAL GROWTH
Table 3. Partial list of transcripts identified by microarray analysis to be upregulated in jejunum of Klf9⫺/⫺ mice
Protein
Protein phosphatase 2A, regulatory subunit B (PR 53)
Nodal modulator 1
Delta/notch-like EGF-related receptor
Armadillo repeat containing, X-linked 2
Insulin-like growth factor binding protein 4
Inhibitor of growth family, member 1
Transforming growth factor-␤1-induced transcript 4
Serum response factor
High-mobility group AT-hook 1
Zinc finger, AN1-type domain 2A
Smoothelin
EGL nine homolog 3 (C. elegans)
Tropomyosin 2␤
Desmin
Lectin, galactose binding, soluble 1
Interferon-related developmental regulator 1
Serum response factor
ATPase family, AAA domain containing 3A
Acyl-CoA synthetase short-chain family member 1
Sirtuin 3 (silent mating-type information regulation 2, homolog) 3 (S. cerevisiae)
Mitochondrial ribosomal protein L38
Mitochondrial carrier homolog 2 (C. elegans)
Claudin 4
Adipocyte adhesion molecule
CDK2 (cyclin-dependent kinase 2)-associated protein 1
Cyclin-dependent kinase 4
Cyclin D2
Inhibitor of growth family, member 1
Cyclin D2
Potassium intermediate/small conductance calcium-activated channel, subfamily N, member 4
Fold cutoff of 1.3; P ⬍ 0.05. Note, some genes are listed in more than one category due to multiple functions.
borne out by RT-PCR. Ptk6 gene is expressed in transit cells as
they exit the crypt (40); therefore, reduction in this mRNA’s
abundance in knockouts may be reflective of fewer transit
cells per crypt or fewer numbers of cells exiting crypts.
Augmentation of cyclin D2 expression may be a compensatory response to reduced Wnt signaling in Klf9 knockout
crypts, as this gene product functions in Wnt/Disheveled
signaling pathways in other proliferative tissues (16). Expression of mushashi-1 mRNA (a presumptive marker of
intestinal stem cells; Ref. 26) in jejunum was unaffected by
altered genotype as analyzed by microarray and RT-PCR
(data not shown). The differential expression (knockout ⬎
WT) of Igfbp4 mRNA was confirmed in jejunum smooth
muscle tissue obtained by dissection (Fig. 7B).
DISCUSSION
Our results show that postnatal intestine development is
perturbed in the Klf9⫺/⫺ mouse. The original report of these
mice (22) described a neurobehavioral phenotype for the
knockout animals. Subsequent work demonstrated subfertility
for Klf9-deficient dams, characterized by perturbations in embryo-maternal signaling during peri-implantation (37, 41). The
present study further implicates Klf9 in control of cell growth,
differentiation, and migration and in long-range signaling
within the small intestine and colon. The emerging complexyet-subtle phenotype of the Klf9⫺/⫺ mouse is consistent with
the postnatal induction of expression of Klf9 gene in multiple
tissues (22).
Klf9 mRNA is expressed in epithelial and mesenchymal
layers of the embryonic mouse gut (20). During embryogenesis, Klf4 and Klf5 genes are coexpressed in this tissue (20, 24).
Our results for postweaning and adult mice did not indicate
significant epithelial expression of Klf9 gene in small intestine;
however, epithelial and smooth muscle expression of this gene
were retained in the postnatal and adult colon. The tissue
differences in postnatal Klf9 expression may be functionally
significant, although further characterization of the colon of the
Klf9 knockout is required to examine this possibility. An
important question raised by the current body of work is
whether loss of epithelial Klf9 expression in small intestine is
coincident with crypt-villus morphogenesis during preweaning. Our results, taken in combination with those for embryo
development (20), suggest that this is possible, and in one
scenario, crypt formation and villus growth during development are accompanied by restrictions in cellular expression
domains for Klf9 and other Klf genes. This process might serve
AJP-Gastrointest Liver Physiol • VOL
292 • JUNE 2007 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
Signaling molecules
Ppp2r4
Nomo1
Dner
Armcx2
Igfbp4
Transcription factors
Ing1
Tgfb1i4
Srf
Hmga1
Zfand2a
Muscle markers
Smtn
Egln3
Tpm2
Des
Lgals1
Ifrd1
Srf
Mitochondrial transcripts
Atad3a
Acss1
Sirt3
Mrpl38
Mtch2
Cell adhesion/matrix proteins
Cldn4
Acam
Cell proliferation/cell cycle
Cdk2ap1
Cdk4
Ccnd2
Ing1
Wnt pathway-related
Ccnd2
Transport
Kcnn4
Full Name
G1766
Klf 9 IN INTESTINAL GROWTH
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
Fig. 6. Crypt cell proliferation and villus
epithelial cell migration are inhibited in
Klf9⫺/⫺ mouse jejunum. Animals were injected with bromodeoxyuridine (BrdU) at 2
and 48 h before tissue collection. BrdUpositive cells were identified by immunohistochemistry (n ⫽ 4 and 3 of WT and nullmutant animals, respectively, with 2 sections
used per animal). A, B, C, G, H, and I
represent animals injected with BrdU 2 h
before tissue collection to identify proliferating cells in S-phase (C is a magnification of
an area of A). D, E, F, J, K, and L represent
animals injected with BrdU 48 h before tissue collection to monitor cell migration over
that period. Crypt cell proliferation was reduced in knockout jejunum (M). At 48 h,
there was a significant increase (P ⫽ 0.001)
in BrdU-labeled cells in villi of Klf9⫺/⫺ mice
(N). The crypt-villus unit was measured, and
the midpoint was established (arrows);
BrdU-labeled cells in the bottom half were
counted (P). A representative negative control (jejunum from a mouse that did not
receive BrdU; O) was subjected to BrdU
immunohistochemistry, and nonspecific
binding was observed in the lamina propria
(boxed areas). Nonspecific binding did not
interfere with quantitation of BrdU-labeled
cells in crypt and villus epithelium.
to establish nonoverlapping expression but functionally interactive actions of these transcription factors, consequently driving small intestine morphogenesis as well as cell renewal in the
maturing gut.
In mature mice, Klf4 mRNA abundance is maximal in the
colon, with the transcript localized to surface and middle/upper
crypt epithelium as well as scattered lamina propria cells (30,
31); this pattern of expression resembles that of Klf9 gene in
colon mucosa (present study). In vivo, Klf4 knockout is lethal
at PND 1 of mouse development because of loss of skin barrier
function (14, 29). Moreover, Klf4knockout mice had complete
loss of the goblet cell lineage in colon, without differences in
cell proliferation or apoptosis (14), whereas the Klf9 knockout
proximal colon had more goblet cells compared with WT (this
study). The phenotypic differences for Klf4 and Klf9 knockout
clearly indicate a lack of functional redundancy for these
transcription factors in the colon. Interestingly, however, Klf4
gene ablation resulted in lineage perturbations in the pit-gland
units of the gastric mucosa (15), and we observed reduced
stomach weights for Klf9⫺/⫺ mice. Studies to examine possible
functional interactions or overlaps of gastric Klf4 and Klf9 will
be interesting in this regard.
The Klf5 (Bteb2, Iklf) gene encodes another highly expressed Klf of the mouse gastrointestinal tract (5). This transcript is localized to the lower-third region of crypts in small
intestine and colon (5), thus spatially distinguishing it from
Klf9 and Klf4. Klf5⫺/⫺ embryos die in utero before embryonic
day 8.5 (32). Heterozygous Klf5⫹/⫺ animals survive until
adulthood but exhibit misshapen villi and a sparse submucosal
mesenchyme (32). Although extensive studies of Klf5 action in
AJP-Gastrointest Liver Physiol • VOL
292 • JUNE 2007 •
www.ajpgi.org
G1767
Klf 9 IN INTESTINAL GROWTH
small intestine crypt-villus epithelium have not been reported,
it seems reasonable to speculate that this Klf promotes crypt
stem and/or transit cell proliferation in concert with Klf9, the
latter acting at a distance.
Evidence for the above is provided by the Ptk6 and Ceacam1
genes. Ptk6 (Brk/Sik) gene encodes an intracellular src-related
tyrosine kinase highly expressed in the gastrointestinal tract
(19, 40). In the murine small intestine, this gene is expressed in
villus epithelial cells as they migrate out of crypts to begin
differentiation (40). Genetic disruption of Ptk6 resulted in
longer intestinal villi and an expanded crypt proliferative zone,
indicating an inhibitory role in cell-cycle arrest and enterocyte
differentiation (10). The Klf9 knockout manifested reduced
Ptk6 expression and shorter intestinal villi, results that appear
inconsistent with that for the Ptk6 knockout. However, Cdkn1b
(p27 or Kip1) mRNA also was repressed in Klf9-null jejunum.
This gene, like Ptk6, is expressed in the upper crypt region,
where it stimulates enterocyte differentiation, perhaps independent of its actions as a cell-cycle inhibitor (27). Because we did
not observe Klf9 promoter activity (measured by X-gal staining) in this cellular location, we infer that Ptk6 and Cdkn1b
genes are indirect targets of Klf9 action. Reductions in mRNA
abundance of both genes may reflect, in part, fewer numbers of
transit cells initiating (or in the process of) differentiation
within the upper crypt and lower villus regions, and that would
correlate with reduced numbers of BrdU-labeled cells at 2 h
after BrdU administration. Furthermore, Ptk6 promotes EGFinduced cell migration (4), and its downregulation may underlie reductions in villus cell motility observed with loss of Klf9.
Similar arguments can be made for the Ceacam1 transcript and
encoded protein. Ceacam1 is a multifunctional cell-adhesion
protein that is upregulated in the early postproliferative, nondifferentiated phase of Caco-2 cell differentiation in vitro
(http://www.ncbi.nlm.nih.gov/projects/geo/gds/gds_browse.
cgi?gds⫽709). Thus its reduced expression in the Klf9 KO
may again reflect a deficiency/block in onset of crypt cell
differentiation in transit cells concomitant with their reduced
migration. Crypt and villus cell migration is likely driven by
stem/transit cell division (21); hence, diminished proliferation
as reported here would contribute to slower migration rates. In
colon, Ptk6 mRNA is manifest in epithelial cells of the upper
crypt, the postmitotic cells undergoing terminal differentiation
AJP-Gastrointest Liver Physiol • VOL
292 • JUNE 2007 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
Fig. 7. Differentially expressed genes in WT and Klf9⫺/⫺ jejunums. A: intact jejunum was obtained from WT (n ⫽ 12) and Klf9⫺/⫺ (n ⫽ 11) male mice (PND
29 –31). RNA was isolated from each animal’s tissue and was subjected to real-time RT-PCR for various mRNAs (each transcript was normalized to Ppia
mRNA). Ptk, protein tyrosine kinase; Ceacam, CEA-related cell adhesion molecule; Igfbp, IGF binding protein; Pitx, pituitary homeobox; Ccnd2, cyclin D2 gene;
Mtm, myotubularin. B: Klf9 gene knockout leads to increased Igfbp4 mRNA abundance in intestinal smooth muscle. Jejunum smooth muscle was obtained by
dissection from WT (n ⫽ 7) and Klf9⫺/⫺ (n ⫽ 8) male mice (PND 30). RNA was isolated from each tissue and was subjected to real-time RT-PCR for Igfbp4
mRNA (normalized to Ppia mRNA). Normalized values for each mouse sample are shown along with corresponding box plot. Expression data for each gene
in A or B were compared for effects of genotype by Student’s t-test or Mann-Whitney rank sum test, with resultant P values indicated.
G1768
Klf 9 IN INTESTINAL GROWTH
proliferation and goblet cell differentiation in this tissue. Further study of intestinal Klf9 and its downstream targets may
reveal novel pathways that can facilitate pharmacological approaches for treatment of short-bowel syndrome, mucositis,
colitis, and other intestinal pathologies of compromised cryptvillus cell renewal.
ACKNOWLEDGMENTS
We thank Dr. Yan Geng, Amanda Linz, Renea Eason, Reneé Till, Julie
Frank, and Charles Skinner for technical assistance; Dr. Bhuvanesh Dave for
suggestions regarding BrdU histochemistry; Dr. Rick Helm for manuscript
critique; and Bryan Hewlett for suggestions regarding staining of Paneth cells.
GRANTS
This work was supported by grants from the National Institute of Child
Health and Human Development (2-RO1-HD-21961), Arkansas Children’s
Hospital Research Institute Dean’s Research Development Fund, and Arkansas
Biosciences Institute.
REFERENCES
1. Bancroft JD, Stevens A. Theory and practice of histological techniques
(2nd ed.). London: Churchill Livingston, 1982.
2. Black AR, Black JD, Azizkhan-Clifford J. Sp1and Krüppel-like factor
family of transcription factors in cell growth regulation and cancer. J Cell
Physiol 188: 143–160, 2001.
3. Carson SL, Pickett JP. Histochemistry. In: Laboratory Medicine, edited
by Race GS. Philadelphia: Harper and Row, 1983.
4. Chen HY, Shen CH, Tsai YT, Lin FC, Huang YP, Chen RH. Brk
activates rac1 and promotes cell migration and invasion by phosphorylating paxillin. Mol Cell Biol 24: 10558 –10572, 2004.
5. Conkright MD, Wani MA, Anderson KP, Lingrel JB. A gene encoding
an intestinal-enriched member of the Krüppel-like factor family expressed
in intestinal epithelial cells. Nucleic Acids Res 27: 1263–1270, 1999.
6. Cunha GR, Hayward SW, Dahiya R, Foster BA. Smooth muscleepithelial interactions in normal and neoplastic prostatic development.
Acta Anat (Basel) 155: 63–72, 1996.
7. Denver RJ, Ouellet L, Furling D, Kobayashi A, Fujii-Kuriyama Y,
Puymirat J. Basic transcription element-binding protein (BTEB) is a
thyroid hormone-regulated gene in the developing central nervous system.
Evidence for a role in neurite outgrowth. J Biol Chem 274: 23128 –23134,
1999.
8. Diehl D, Hoeflich A, Wolf E, Lahm H. Insulin-like growth factor
(IGF)-binding protein-4 inhibits colony formation of colorectal cancer
cells by IGF-independent mechanisms. Cancer Res 64: 1600 –1603, 2004.
9. Dvorak B, Stephana AL, Holubec H, Williams CS, Philipps AF,
Koldovskoy O. Insulin-like growth factor-I (IGF-I) mRNA in the small
intestine of suckling and adult rats. FEBS Lett 388: 155–160, 1996.
10. Haegebarth A, Bie W, Yang R, Crawford SE, Vasioukhin V, Fuchs E,
Tyner AL. Protein tyrosine kinase 6 negatively regulates growth and
promotes enterocyte differentiation in the small intestine. Mol Cell Biol
26: 4949 – 4957, 2006.
11. Imataka H, Sogawa K, Yasumoto K, Kikuchi Y, Sasano K, Kobayashi
A, Hayami M, Fujii-Kuriyama Y. Two regulatory proteins that bind to
the basic transcription element (BTE), a GC box sequence in the promoter
region of the rat P-4501A1 gene. EMBO J 11: 3663–3671, 1992.
12. Kaczynski J, Cook T, Urrutia R. Sp1- and Krüppel-like transcription
factors. Genome Biol 4: 206, 2003.
13. Kao LC, Tulac S, Lobo S, Imani B, Yang JP, Germeyer A, Osteen K,
Taylor RN, Lessey BA, Giudice LC. Global gene profiling in human
endometrium during the window of implantation. Endocrinology 143:
2119 –2138, 2002.
14. Katz JP, Perreault N, Goldstein BG, Lee CS, Labosky PA, Yang VW,
Kaestner KH. The zinc-finger transcription factor Klf4 is required for
terminal differentiation of goblet cells in the colon. Development 129:
2619 –2628, 2002.
15. Katz JP, Perreault N, Goldstein BG, Actman L, McNally SR, Silberg
DG, Furth EE, Kaestner KH. Loss of Klf4 in mice causes altered
proliferation and differentiation and precancerous changes in the adult
stomach. Gastroenterology 128: 935–945, 2005.
16. Kioussi C, Briata P, Baek SH, Rose DW, Hamblet NS, Herman T,
Ohgi KA, Lin C, Gleiberman A, Wang J, Brault V, Ruiz-Lozano P,
AJP-Gastrointest Liver Physiol • VOL
292 • JUNE 2007 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
(10, 19). In the present study, these same cells exhibited Klf9
promoter activity. Thus it will be interesting to examine the
relationships between Klf9 and Ptk6 in colon cell migration
and goblet cell specification.
The present study implicates Klf9 and one or more of its
target genes as long-range effector(s) of cell growth and
migration in crypt-villus units. Signaling of epithelium by
adjacent smooth muscle cells as well as the reciprocal interaction are not unprecedented (6). Wnts represent possible candidates for this signaling, since this signal-transduction pathway
is well implicated in the maintenance of stem/transit cell
proliferation as well as in differentiation of the intestinal cell
lineages (25, 39). Absence of Klf9 may lead to dampening of
Wnt signals and consequent reductions in crypt stem/progenitor cell proliferation and differentiation. Consistent with this,
microarray results identified changes in mRNA abundance for
several Wnt pathway-related proteins (Dab2, Ccnd2) that
might underlie the Klf9⫺/⫺ phenotype, although there were no
apparent changes in expression of Wnt ligands, members of the
Frizzled and LDL receptor-related proteins, or the Wnt antagonists. We speculate that upregulation of Ccnd2 transcripts in
knockout jejunum may be indicative of a compensatory response to attenuated Wnt signaling, because this protein can
function downstream of the Wnt-Disheveled-␤-catenin pathway (16, 46).
The observed increase in Igfbp4 mRNA abundance in the
jejunum smooth muscle of Klf9⫺/⫺ mice is of interest from the
standpoint of hypothesized mesenchymal-derived signaling
mediators. The Igfbp4 gene is expressed in the muscularis
externa and lamina propria of rodent and human intestines (17,
33, 43). In vitro, this IGF binding protein is antiproliferative
via sequestration of IGF and/or IGF-independent mechanism(s) and can inhibit IGF-dependent cell invasion (motility)
in vitro (8). IGF-I is synthesized by differentiating enterocytes
and goblet cells within intestinal crypts before their movement
onto villi (9). Thus it is tempting to speculate that the induction
of Igfbp4 in the muscularis externa with Klf9 knockout elicits
a partial block in IGF-dependent proliferation and/or motility
of cells within the confines of the crypts by sequestering IGF-I
ligand. Nevertheless, this requires experimental verification.
Transgenic overexpression of Igfbp4 in mouse intestinal
smooth muscle caused hypoplasia of this tissue compartment
(43). However, our morphological analysis of intestinal smooth
muscle did not indicate reductions in tissue cross-sectional
thickness with Klf9 knockout, perhaps reflecting functional
antagonism by another Igfbp (23). However, the latter point
remains speculative, as we have not documented mRNA expression for Igfbp family members other than Igfbp4 and
Igfbp5 in WT and knockout intestines.
In summary, results demonstrate a regulatory role for Klf9 in
crypt cell proliferation and villus cell migration in small
intestine, which may be mediated by long-range signaling from
the muscularis externa. Absence of Klf9 resulted in smaller
villi, likely a consequence of reductions in proliferation of
crypt cells and in villus cell migration. We speculate that
blocks in cell proliferation/motility in crypts elicit compensatory increases in Wnt signaling that, although not overriding
these blocks, result in augmented Paneth cell numbers and
diminished goblet cell numbers in crypts. Knockout of Klf9
resulted in subtle changes in colon mucosal phenotype, which
may indicate effects of this transcription factor on crypt cell
G1769
Klf 9 IN INTESTINAL GROWTH
17.
18.
19.
20.
21.
22.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
AJP-Gastrointest Liver Physiol • VOL
transcription element-binding protein, and progesterone receptor in endometrial epithelial cells. Endocrinology 140: 2517–2525, 1999.
Simmen RCM, Zhang XL, Michel FJ, Min SH, Zhao G, Simmen FA.
Molecular markers of endometrial epithelial cell mitogenesis mediated by
the Sp/Krüppel-like factor BTEB1. DNA Cell Biol 21: 115–128, 2002.
Simmen RCM, Simmen FA. Progesterone receptors and Sp/Krüppel-like
family members in the uterine endometrium. Front Biosci 7: d1556 –
d1565, 2002.
Simmen RCM, Eason RR, McQuown JR, Linz AL, Kang TJ, Chatman L Jr, Till SR, Fujii-Kuriyama Y, Simmen FA, Oh SP. Subfertility,
uterine hypoplasia, and partial progesterone resistance in mice lacking the
Krüppel-like factor 9/basic transcription element-binding protein-1
(Bteb1) gene. J Biol Chem 279: 29286 –29294, 2004.
Suske G, Bruford E, Philipsen S. Mammalian SP/KLF transcription
factors: bring in the family. Genomics 85: 551–556, 2005.
van Es JH, Jay P, Gregorieff A, van Gijn ME, Jonkheer S, Hatzis P,
Thiele A, van den Born M, Begthel H, Brabletz T, Taketo MM,
Clevers H. Wnt signalling induces maturation of Paneth cells in intestinal
crypts. Nat Cell Biol 7: 381–386, 2005.
Vasioukhin V, Serfas MS, Siyanova EY, Polonskaia M, Costigan VJ,
Liu B, Thomason A, Tyner AL. A novel intracellular epithelial cell
tyrosine kinase is expressed in the skin and gastrointestinal tract. Oncogene 10: 349 –357, 1995.
Velarde MC, Geng Y, Eason RR, Simmen FA, Simmen RCM. Null
mutation of Krüppel-like factor 9/basic transcription element binding
protein-1 alters peri-implantation uterine development in mice. Biol Reprod 73: 472– 481, 2005.
Velarde MC, Iruthayanathan M, Eason RR, Zhang D, Simmen FA,
Simmen RCM. Progesterone receptor transactivation of the secretory
leukocyte protease inhibitor gene in Ishikawa endometrial epithelial cells
involves recruitment of Krüppel-like factor 9/basic transcription element
binding protein-1. Endocrinology 147: 1969 –1978, 2006.
Wang J, Niu W, Witte DP, Chernausek SD, Nikiforov YE, Clemens
TL, Sharifi B, Strauch AR, Fagin JA. Overexpression of insulin-like
growth factor-binding protein-4 (IGFBP-4) in smooth muscle cells of
transgenic mice through a smooth muscle ␣-actin-IGFBP-4 fusion gene
induces smooth muscle hypoplasia. Endocrinology 139: 2605–2614, 1998.
Wang Y, Michel FJ, Wing A, Simmen FA, Simmen RCM. Cell-type
expression, immunolocalization, and deoxyribonucleic acid-binding activity of basic transcription element binding transcription factor, an Sprelated family member, in porcine endometrium of pregnancy. Biol Reprod 57: 707–714, 1997.
Xiao R, Badger TM, Simmen FA. Dietary exposure to soy or whey
proteins alters colonic global gene expression profiles during rat colon
tumorigenesis. Mol Cancer 4: 1, 2005.
Yang R, Bie W, Haegebarth A, Tyner AL. Differential regulation of
D-type cyclins in the mouse intestine. Cell Cycle 5: 180 –183, 2006.
Zhang D, Zhang XL, Michel FJ, Blum JL, Simmen FA, Simmen
RCM. Direct interaction of the Krüppel-like family (KLF) member,
BTEB1, and PR mediates progesterone-responsive gene expression in
endometrial epithelial cells. Endocrinology 143: 62–73, 2002.
Zhang XL, Simmen FA, Michel FJ, Simmen RCM. Increased expression of the Zn-finger transcription factor BTEB1 in human endometrial
cells is correlated with distinct cell phenotype, gene expression patterns,
and proliferative responsiveness to serum and TGF-␤1. Mol Cell Endocrinol 181: 81–96, 2001.
Zhang XL, Zhang D, Michel FJ, Blum JL, Simmen FA, Simmen
RCM. Selective interactions of Krüppel-like factor 9/basic transcription
element-binding protein with progesterone receptor isoforms A and B
determine transcriptional activity of progesterone-responsive genes in
endometrial epithelial cells. J Biol Chem 278: 21474 –21482, 2003.
292 • JUNE 2007 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
23.
Nguyen HD, Kemler R, Glass CK, Wynshaw-Boris A, Rosenfeld MG.
Identification of a Wnt/Dvl/beta-Catenin3 Pitx2 pathway mediating celltype-specific proliferation during development. Cell 111: 673– 685, 2002.
Kuemmerle JF, Teng B. Regulation of IGFBP-4 levels in human intestinal muscle by an IGF-I-activated, confluence-dependent protease. Am J
Physiol Gastrointest Liver Physiol 279: G975–G982, 2000.
Lack EE, Mercer L. A modified Grimelius argyrophil technique for
neurosecretory granules. Am J Surg Pathol 1: 275–277, 1977.
Llor X, Serfas MS, Bie W, Vasioukhin V, Polonskaia M, Derry J,
Abbott CM, Tyner AL. BRK/Sik expression in the gastrointestinal tract
and in colon tumors. Clin Cancer Res 5: 1767–1777, 1999.
Martin KM, Metcalfe JC, Kemp PR. Expression of Klf9 and Klf13 in
mouse development. Mech Dev 103: 149 –151, 2001.
Meineke FA, Potten CS, Loeffler M. Cell migration and organization in
the intestinal crypt using a lattice-free model. Cell Prolif 34: 253–266,
2001.
Morita M, Kobayashi A, Yamashita T, Shimanuki T, Nakajima O,
Takahashi S, Ikegami S, Inokuchi K, Yamashita K, Yamamoto M,
Fujii-Kuriyama Y. Functional analysis of basic transcription element
binding protein by gene targeting technology. Mol Cell Biol 23: 2489 –
2500, 2003.
Ning Y, Schuller AG, Bradshaw S, Rotwein P, Ludwig T, Frystyk J,
Pintar JE. Diminished growth and enhanced glucose metabolism in triple
knockout mice containing mutations of insulin-like growth factor binding
protein-3, -4, and -5. Mol Endocrinol 20: 2173–2186, 2006.
Ohnishi S, Laub F, Matsumoto N, Asaka M, Ramirez F, Yoshida T,
Terada M. Developmental expression of the mouse gene coding for the
Krüppel-like transcription factor KLF5. Dev Dyn 217: 421– 429, 2000.
Pinto D, Gregorieff A, Begthel H, Clevers H. Canonical Wnt signals are
essential for homeostasis of the intestinal epithelium. Genes Dev 17:
1709 –1713, 2003.
Potten CS, Booth C, Tudor GL, Booth D, Brady G, Hurley P, Ashton
G, Clarke R, Sakakibara S, Okano H. Identification of a putative
intestinal stem cell and early lineage marker; musashi-1. Differentiation
71: 28 – 41, 2003.
Quaroni A, Tian JQ, Seth P, Ap Rhys C. p27(Kip1) is an inducer of
intestinal epithelial cell differentiation. Am J Physiol Cell Physiol 279:
C1045–C1057, 2000.
Sasik R, Woelk CH, Corbeil J. Microarray truths and consequences. J
Mol Endocrinol 33: 1–9, 2004.
Segre JA, Bauer C, Fuchs E. Klf4 is a transcription factor required for
establishing the barrier function of the skin. Nat Genet 22: 356 –360, 1999.
Shie JL, Chen ZY, O’Brien MJ, Pestell RG, Lee ME, Tseng CC. Role
of gut-enriched Krüppel-like factor in colonic cell growth and differentiation. Am J Physiol Gastrointest Liver Physiol 279: G806 –G814, 2000.
Shields JM, Christy RJ, Yang VW. Identification and characterization of
a gene encoding a gut-enriched Krüppel-like factor expressed during
growth arrest. J Biol Chem 271: 20009 –20017, 1996.
Shindo T, Manabe I, Fukushima Y, Tobe K, Aizawa K, Miyamoto S,
Kawai-Kowase K, Moriyama N, Imai Y, Kawakami H, Nishimatsu H,
Ishikawa T, Suzuki T, Morita H, Maemura K, Sata M, Hirata Y,
Komukai M, Kagechika H, Kadowaki T, Kurabayashi M, Nagai R.
Krüppel-like zinc-finger transcription factor KLF5/BTEB2 is a target for
angiotensin II signaling and an essential regulator of cardiovascular
remodeling. Nat Med 8: 856 – 863, 2002.
Shoubridge CA, Steeb CB, Read LC. IGFBP mRNA expression in small
intestine of rat during postnatal development. Am J Physiol Gastrointest
Liver Physiol 281: G1378 –G1384, 2001.
Simmen RCM, Chung TE, Imataka H, Michel FJ, Badinga L, Simmen
FA. Trans-activation functions of the Sp-related nuclear factor, basic