Krüppel-Like Factor 5 Mediates Transmissible Murine Colonic Hyperplasia Caused by Citrobacter rodentium Infection
2008; Elsevier BV; Volume: 134; Issue: 4 Linguagem: Inglês
10.1053/j.gastro.2008.01.013
ISSN1528-0012
AutoresBeth B. McConnell, Jan‐Michael A. Klapproth, Maiko Sasaki, Mandayam O. Nandan, Vincent W. Yang,
Tópico(s)Cancer-related gene regulation
ResumoBackground & Aims: Krüppel-like factor 5 (KLF5) is a transcription factor that is highly expressed in proliferating crypt cells of the intestinal epithelium. KLF5 has a pro-proliferative effect in vitro and is induced by mitogenic and stress stimuli. To determine whether KLF5 is involved in mediating proliferative responses to intestinal stressors in vivo, we examined its function in a mouse model of transmissible murine colonic hyperplasia triggered by colonization of the mouse colon by the bacteria Citrobacter rodentium. Methods: Heterozygous Klf5 knockout (Klf5+/−) mice were generated from embryonic stem cells carrying an insertional disruption of the Klf5 gene. Klf5+/− mice or wild-type (WT) littermates were infected with C rodentium by oral gavage. At various time points postinfection, mice were killed and distal colons were harvested. Colonic crypt heights were determined morphometrically from sections stained with H&E. Frozen tissues were stained by immunofluorescence using antibodies against Klf5 and the proliferation marker, Ki67, to determine Klf5 expression and numbers of proliferating cells per crypt. Results: Infection of WT mice with C rodentium resulted in a 2-fold increase in colonic crypt heights at 14 days postinfection and was accompanied by a 1.7-fold increase in Klf5 expression. Infection of Klf5+/− mice showed an attenuated induction of Klf5 expression, and hyperproliferative responses to C rodentium were reduced in the Klf5+/− animals as compared with WT littermates. Conclusion: Our study shows that Klf5 is a key mediator of crypt cell proliferation in the colon in response to pathogenic bacterial infection. Background & Aims: Krüppel-like factor 5 (KLF5) is a transcription factor that is highly expressed in proliferating crypt cells of the intestinal epithelium. KLF5 has a pro-proliferative effect in vitro and is induced by mitogenic and stress stimuli. To determine whether KLF5 is involved in mediating proliferative responses to intestinal stressors in vivo, we examined its function in a mouse model of transmissible murine colonic hyperplasia triggered by colonization of the mouse colon by the bacteria Citrobacter rodentium. Methods: Heterozygous Klf5 knockout (Klf5+/−) mice were generated from embryonic stem cells carrying an insertional disruption of the Klf5 gene. Klf5+/− mice or wild-type (WT) littermates were infected with C rodentium by oral gavage. At various time points postinfection, mice were killed and distal colons were harvested. Colonic crypt heights were determined morphometrically from sections stained with H&E. Frozen tissues were stained by immunofluorescence using antibodies against Klf5 and the proliferation marker, Ki67, to determine Klf5 expression and numbers of proliferating cells per crypt. Results: Infection of WT mice with C rodentium resulted in a 2-fold increase in colonic crypt heights at 14 days postinfection and was accompanied by a 1.7-fold increase in Klf5 expression. Infection of Klf5+/− mice showed an attenuated induction of Klf5 expression, and hyperproliferative responses to C rodentium were reduced in the Klf5+/− animals as compared with WT littermates. Conclusion: Our study shows that Klf5 is a key mediator of crypt cell proliferation in the colon in response to pathogenic bacterial infection. The mammalian gut epithelium is a dynamic tissue that plays an active role in maintaining tissue homeostasis in the face of rapid cell turnover and constant changes in the bacterial milieu. Maintaining the status quo requires the stringent regulation of pathways involving proliferation, differentiation, apoptosis, and inflammation. This balance becomes even more critical in instances of inflammatory bowel disease (IBD), in which dysregulation of these pathways can result in excessive tissue injury, inadequate tissue regeneration, and increased risk of developing cancer. Although the primary pathway involved in the normal regeneration of the intestinal epithelium has been relatively well characterized, little is known about the molecular events that regulate proliferation during pathogenic bacterial infection or exposure to other stressors. Krüppel-like factor 5 (KLF5) is a member of a family of zinc finger-containing transcription factors that function in the regulation of diverse cellular processes, including development, proliferation, and differentiation.1Dang D.T. Pevsner J. Yang V.W. The biology of the mammalian Krüppel-like family of transcription factors.Int J Biochem Cell Biol. 2000; 32: 1103-1121Crossref PubMed Scopus (371) Google Scholar, 2Bieker J.J. Krüppel-like factors: three fingers in many pies.J Biol Chem. 2001; 276: 34355-34358Crossref PubMed Scopus (540) Google Scholar, 3Kaczynski J. Cook T. Urrutia R. Sp1- and Krüppel-like transcription factors.Genome Biol. 2003; 4: 206Crossref PubMed Scopus (748) Google Scholar, 4Black A.R. Black J.D. Azizkhan-Clifford J. Sp1 and Krüppel-like factor family of transcription factors in cell growth regulation and cancer.J Cell Physiol. 2001; 188: 143-160Crossref PubMed Scopus (902) Google Scholar, 5Ghaleb A.M. Nandan M.O. Chanchevalap S. et al.Krüppel-like factors 4 and 5: the yin and yang regulators of cellular proliferation.Cell Res. 2005; 15: 92-96Crossref PubMed Scopus (255) Google Scholar, 6McConnell B.B. Ghaleb A.M. Nandan M.O. et al.The diverse functions of Krüppel-like factors 4 and 5 in epithelial biology and pathobiology.Bioessays. 2007; 29: 549-557Crossref PubMed Scopus (212) Google Scholar KLF5 is highly expressed in the intestinal epithelium and is found predominantly in the proliferating cells of the crypt,6McConnell B.B. Ghaleb A.M. Nandan M.O. et al.The diverse functions of Krüppel-like factors 4 and 5 in epithelial biology and pathobiology.Bioessays. 2007; 29: 549-557Crossref PubMed Scopus (212) Google Scholar, 7Conkright M.D. Wani M.A. Anderson K.P. et al.A gene encoding an intestinal-enriched member of the Krüppel-like factor family expressed in intestinal epithelial cells.Nucleic Acids Res. 1999; 27: 1263-1270Crossref PubMed Scopus (143) Google Scholar, 8Ohnishi S. Laub F. Matsumoto N. et al.Developmental expression of the mouse gene coding for the Krüppel-like transcription factor KLF5.Dev Dyn. 2000; 217: 421-429Crossref PubMed Scopus (76) Google Scholar suggesting that it has a positive growth regulatory role in the intestinal tissue. Indeed, several in vitro studies support a pro-proliferative role for KLF5 in nontransformed cultured epithelial cells.9Bateman N.W. Tan D. Pestell R.G. et al.Intestinal tumor progression is associated with altered function of KLF5.J Biol Chem. 2004; 279: 12093-12101Crossref PubMed Scopus (110) Google Scholar, 10Sun R. Chen X. Yang V.W. Intestinal-enriched Krüppel-like factor (Krüppel-like factor 5) is a positive regulator of cellular proliferation.J Biol Chem. 2001; 276: 6897-6900Crossref PubMed Scopus (138) Google Scholar, 11Chanchevalap S. Nandan M.O. Merlin D. et al.All-trans retinoic acid inhibits proliferation of intestinal epithelial cells by inhibiting expression of the gene encoding Krüppel-like factor 5.FEBS Lett. 2004; 578: 99-105Abstract Full Text Full Text PDF PubMed Scopus (72) Google Scholar In addition, ectopic expression of KLF5 in NIH3T3 cells has been shown to promote proliferation.10Sun R. Chen X. Yang V.W. Intestinal-enriched Krüppel-like factor (Krüppel-like factor 5) is a positive regulator of cellular proliferation.J Biol Chem. 2001; 276: 6897-6900Crossref PubMed Scopus (138) Google Scholar Moreover, KLF5 has been shown to mediate the transforming effects of oncogenic H-RAS.12Nandan M.O. Yoon H.S. Zhao W. et al.Krüppel-like factor 5 mediates the transforming activity of oncogenic H-Ras.Oncogene. 2004; 23: 3404-3413Crossref PubMed Scopus (119) Google Scholar, 13Nandan M.O. Chanchevalap S. Dalton W.B. et al.Krüppel-like factor 5 promotes mitosis by activating the cyclin B1/Cdc2 complex during oncogenic Ras-mediated transformation.FEBS Lett. 2005; 579: 4757-4762Abstract Full Text Full Text PDF PubMed Scopus (71) Google Scholar Transcriptional targets of KLF5 include a number of genes that encode pro-proliferative components of the cell-cycle machinery, including cyclin D1, cyclin B1, and Cdc2.12Nandan M.O. Yoon H.S. Zhao W. et al.Krüppel-like factor 5 mediates the transforming activity of oncogenic H-Ras.Oncogene. 2004; 23: 3404-3413Crossref PubMed Scopus (119) Google Scholar, 13Nandan M.O. Chanchevalap S. Dalton W.B. et al.Krüppel-like factor 5 promotes mitosis by activating the cyclin B1/Cdc2 complex during oncogenic Ras-mediated transformation.FEBS Lett. 2005; 579: 4757-4762Abstract Full Text Full Text PDF PubMed Scopus (71) Google Scholar Various external stimuli have been reported to activate KLF5 expression, including addition of fetal bovine serum to serum-deprived cells10Sun R. Chen X. Yang V.W. Intestinal-enriched Krüppel-like factor (Krüppel-like factor 5) is a positive regulator of cellular proliferation.J Biol Chem. 2001; 276: 6897-6900Crossref PubMed Scopus (138) Google Scholar and treatment of cultured cells with basic fibroblast growth factor14Kawai-Kowase K. Kurabayashi M. Hoshino Y. et al.Transcriptional activation of the zinc finger transcription factor BTEB2 gene by Egr-1 through mitogen-activated protein kinase pathways in vascular smooth muscle cells.Circ Res. 1999; 85: 787-795Crossref PubMed Scopus (84) Google Scholar and phorbol 12-myristate 13-acetate.14Kawai-Kowase K. Kurabayashi M. Hoshino Y. et al.Transcriptional activation of the zinc finger transcription factor BTEB2 gene by Egr-1 through mitogen-activated protein kinase pathways in vascular smooth muscle cells.Circ Res. 1999; 85: 787-795Crossref PubMed Scopus (84) Google Scholar, 15Nagai R. Suzuki T. Aizawa K. et al.Phenotypic modulation of vascular smooth muscle cells: dissection of transcriptional regulatory mechanisms.Ann N Y Acad Sci. 2001; 947: 56-67Crossref PubMed Scopus (27) Google Scholar Furthermore, in vivo studies conducted in mouse vascular tissue have shown an increase in KLF5 expression in response to physical stress caused by injury.16Shindo T. Manabe I. Fukushima Y. et al.Krüppel-like zinc-finger transcription factor KLF5/BTEB2 is a target for angiotensin II signaling and an essential regulator of cardiovascular remodeling.Nat Med. 2002; 8: 856-863Crossref PubMed Scopus (329) Google Scholar Induction of KLF5 has been shown to be downstream of activation of the mitogen-activated protein kinase/extracellular signal-regulated kinase (ERK)1/2 pathway, with KLF5 expression being driven by the early response gene, early growth response factor 1.14Kawai-Kowase K. Kurabayashi M. Hoshino Y. et al.Transcriptional activation of the zinc finger transcription factor BTEB2 gene by Egr-1 through mitogen-activated protein kinase pathways in vascular smooth muscle cells.Circ Res. 1999; 85: 787-795Crossref PubMed Scopus (84) Google Scholar, 17Nagai R. Shindo T. Manabe I. et al.KLF5/BTEB2, a Kruppel-like zinc-finger type transcription factor, mediates both smooth muscle cell activation and cardiac hypertrophy.Adv Exp Med Biol. 2003; 538: 57-66Crossref PubMed Google Scholar Recently, our laboratory reported that KLF5 expression is induced in IEC-6 rat intestinal epithelial cells after exposure to the bacterial component lipopolysaccharide.18Chanchevalap S. Nandan M.O. McConnell B.B. et al.Krüppel-like factor 5 is an important mediator for lipopolysaccharide-induced proinflammatory response in intestinal epithelial cells.Nucleic Acids Res. 2006; 34: 1216-1223Crossref PubMed Scopus (81) Google Scholar Thus, given the pro-proliferative activity of KLF5 in vitro and its activation in response to various mitogenic and stress stimuli, we hypothesize that KLF5 may play a role in hyperproliferative responses to external stressors in intestinal tissues. To address this hypothesis, the current study uses a mouse model of hyperproliferation known as transmissible murine colonic hyperplasia to examine the involvement of KLF5 in proliferative changes induced by enteric bacterial pathogenic infection. Transmissible murine colonic hyperplasia is caused by infection with Citrobacter rodentium, a gram-negative, noninvasive bacterial pathogen that colonizes the distal colon of mice by forming attaching and effacing lesions.19Luperchio S.A. Newman J.V. Dangler C.A. et al.Citrobacter rodentium, the causative agent of transmissible murine colonic hyperplasia, exhibits clonality: synonymy of C. rodentium and mouse-pathogenic Escherichia coli.J Clin Microbiol. 2000; 38: 4343-4350PubMed Google Scholar Infection is characterized by dramatic elongation of colonic crypts, hyperproliferation of epithelial cells, goblet cell depletion, and mucosal thickening.20Luperchio S.A. Schauer D.B. Molecular pathogenesis of Citrobacter rodentium and transmissible murine colonic hyperplasia.Microbes Infect. 2001; 3: 333-340Crossref PubMed Scopus (220) Google Scholar Over a 2-week period, the crypts in the distal colon double in height, reaching a maximum between 14 and 21 days postinfection (pi). Inflammatory responses with C rodentium infection are minimal, making this model an excellent tool for examining hyperproliferative changes in the colon. To determine the role of KLF5 in proliferative responses of the mouse colon to C rodentium infection, we have generated mice with heterozygous knockout of the Klf5 gene (Klf5+/−). Homozygous deletions of Klf5 are embryonic lethal; however, heterozygous Klf5 mouse models have been used in other studies to show key roles for Klf5 in cardiovascular remodeling and adipocyte differentiation.16Shindo T. Manabe I. Fukushima Y. et al.Krüppel-like zinc-finger transcription factor KLF5/BTEB2 is a target for angiotensin II signaling and an essential regulator of cardiovascular remodeling.Nat Med. 2002; 8: 856-863Crossref PubMed Scopus (329) Google Scholar, 21Oishi Y. Manabe I. Tobe K. et al.Krüppel-like transcription factor KLF5 is a key regulator of adipocyte differentiation.Cell Metab. 2005; 1: 27-39Abstract Full Text Full Text PDF PubMed Scopus (357) Google Scholar In this study, we compared hyperproliferative responses to C rodentium infection in wild-type (WT) and Klf5+/− mice. Results show that Klf5 in the colonic crypts is induced in response to C rodentium infection and that hyperproliferative responses are suppressed in the colons of Klf5+/− animals. These findings suggest that induction of KLF5 is a key event that contributes to epithelial cell hyperproliferation after infection with C rodentium and shows that KLF5 is an important mediator for colonic hyperproliferation in response to in vivo stressors. C57BL/6 mice were purchased from The Jackson Laboratory (Bar Harbor, ME). Mice strains were bred and housed in the Whitehead Animal Research Facility at Emory University. Animal care and in vivo research complied with all relevant Emory University institutional policies and federal guidelines. Mouse embryonic stem cells containing a disrupted allele for Klf5 (Klf5+/−), clone XB751, were obtained from BayGenomics, Inc (San Francisco, CA). The disrupted allele was generated by insertion of a gene trap vector into the first intron of the Klf5 gene. The insertional mutation created a fusion transcript containing exon 1 of Klf5 joined to a β-galactosidase/neomycin cassette and followed by a translational stop codon. Chimeras for expression of the disrupted Klf5 allele were generated by the Emory Transgenic Facility by injecting blastocysts with Klf5+/− embryonic stem cells and implanting the embryos into pseudopregnant females. Chimeric progeny were bred to C57BL/6 animals to produce Klf5+/− founder mice. Male Klf5+/− mice were backcrossed with C57BL/6 females (WT) for 4 generations to produce Klf5+/− mice on a homogeneous C57BL/6 background. Klf5+/− mice used for the described experiments were F5 generation mice. A 0.5-cm section of tail was removed and used to prepare genomic DNA with the Extract-N-Amp Tissue polymerase chain reaction (PCR) kit (Sigma Aldrich, St Louis, MO) according to the manufacturer’s instructions. To identify Klf5+/− mice, PCR analysis was conducted to amplify a portion of the β-galactosidase gene that was part of the insertional mutation. Primers used were as follows: forward: 5′-TTATCGATGAGCGTGGTGGTTATGC-3′ and reverse: 5′-GCGCGTACATCGGGCAAATAATATC-3′. The C rodentium strain, C rodentium deposited under the name C freundii (Braak) Werkman and Gillen, was obtained from the American Type Culture Collection. Six-week-old C57BL/6 WT or Klf5+/− littermates were infected with 100 μL of phosphate-buffered saline (PBS) containing 5 × 108 colony-forming units of C rodentium by oral gavage. Uninfected controls were given PBS alone. Animals were provided unlimited access to food and water throughout the experiment. Mice were weighed before infection to determine baseline weights, and mice were weighed and observed daily. At various time points pi, mice were euthanized by carbon dioxide asphyxiation and the distal colon was removed. Transmissible murine colonic hyperplasia was apparent by the presence of a significantly thickened distal colon and loose stool. Sections of distal colon were isolated from the region 3 cm proximal to the anal verge, fixed in 10% formalin solution for 48 hours, and processed for embedding in paraffin. Transverse sections (5 μm) were stained with H&E for morphologic evaluation. Photomicrographs were taken using a Zeiss Axioskop2 plus microscope, and crypt height measurements were determined with AxioVision 4.5 software (Carl Zeiss, Inc, Maple Grove, MN). Crypt measurements were conducted in a blinded fashion, taking 3 separate measurements per tissue sample on well-oriented crypts. For Klf5 and Ki67 staining, frozen tissues from distal colon were sectioned, air dried for 1 hour, and fixed in 4% ultrapure formaldehyde (Polysciences, Inc, Warrington, PA) for 5 minutes. Slides were washed 3 times in PBS, permeabilized in 1% Igepal in PBS, and incubated in block solution (0.02% Triton X-100, 3% bovine serum albumin in PBS, or bovine serum albumin in PBS) for 1 hour. Slides then were incubated with primary antibodies against KLF5 (H-30, 1:500; Santa Cruz Biotechnology, Santa Cruz, CA) or Ki67 (NCL-Ki67p, 1:500; Vision BioSystems, Inc, Norwell, MA) overnight at 4°C in a humidified chamber. For immunofluorescence detection, slides were incubated with either Alexa fluor 488 or Alexa fluor 568 goat anti-rabbit secondary antibodies (Invitrogen, Carlsbad, CA), and the nuclei were counterstained with Hoechst 33258. Fluorescent staining was viewed with a Zeiss Axioskop2 plus microscope and images from the various samples were captured using identical exposure settings for quantitative purposes. Images were analyzed for intensities of nuclear KLF5 staining using Metamorph Imaging software (Molecular Devices, Sunnyvale, CA). For cyclin D1 staining, paraffin-embedded sections were deparaffinized and antigen retrieval was performed by incubating slides at 125°C for 10 minutes in citrate buffer, pH 6.0. Tissue sections were blocked using avidin/biotin (Vector Labs, Burlingame, CA) in 2% milk/PBS. Sections then were stained with cyclin D1 polyclonal antibody at a 1:100 dilution (Biocare Medical, Concord, CA) and developed using Vectastain Elite ABC kit (Vector Labs) with DAB (Biocare Medical). Sections then were counterstained using Gill’s Hematoxylin (Vector Labs). Whole-tissue lysates were prepared by homogenization of colon tissues in Laemmli buffer. Tissue extracts (50 μg protein/lane) were subjected to denaturing polyacrylamide gel electrophoresis and transferred to a polyvinylidene difluoride membrane. Membranes were blocked with 5% nonfat dried milk in tween/tris-buffered salt solution (20 mmol/L Tris-HCl, 3 mmol/L KCl, 137 mmol/L NaCl, 0.1% Tween 20, pH 7.5) and incubated with the appropriate primary antibodies: KLF5 rabbit polyclonal antibody developed by our laboratory, or β-actin (AC15; Sigma Aldrich) as a loading control. Total RNA was isolated at 14 days pi from mucosal tissue scraped from the distal colon of WT and Klf5+/− mice either uninfected or infected with C rodentium. Expression of cyclin D1 and cyclin B1 was examined by quantitative reverse-transcription (RT)-PCR using the SYBR Green ER Two-Step qRT-PCR Kit for iCycler (Invitrogen) per the manufacturer’s instructions. PCR amplification was conducted with an initial denaturation step at 95°C for 5 minutes followed by 45 cycles of denaturation at 95°C for 30 seconds, annealing at 60°C for 45 seconds and extension at 72°C for 1 minute. A final extension was performed at 72°C for 5 minutes. A melting curve analysis was performed after amplification was completed. Primers used for murine cyclin D1 were as follows: forward: 5′-CAGACGTTCAGAACCAGATTC-3′ and reverse: 5′-CCCTCCAATAGCAGCGAAAAC-3′. Primers for murine cyclin B1 were as follows: forward: 5′-CAGTTGTGTGCCCAAGAAGA-3′ and reverse: 5′-CTACGGAGGAAGTGCAGAGG-3′. Expression of cyclin D1 and cyclin B1 was normalized to β-actin expression, and relative cyclin D1 and cyclin B1 expression was reported in relation to the WT control. All data represent an n of 6 mice per condition, except where noted specifically. Data from the experiments are presented as the mean ± SEM for 6 mice from each experimental group. Data values for each mouse were determined from the average of multiple measurements as described in the legends to Figures 4B, C, , 5B, and 6B. P values comparing the data for mice in 2 groups were calculated using a 2-tailed t test. Differences were considered to be significant if the P value was less than .05.Figure 5Induction of Klf5 is reduced in Klf5+/− mice compared with WT mice after C rodentium infection. (A) Frozen colonic tissues from uninfected and C rodentium–infected mice were isolated at 14 days pi and stained by immunofluorescence for Klf5 expression. (B) Relative fluorescence intensities per cell were determined by quantitative analysis of images of Klf5 staining. Results represent the mean value of 6 mice, with the value of each mouse being determined by averaging intensities from at least 200 cells. Results are shown as Klf5 fluorescence intensity per cell relative to intensities in WT mice. ***P < .001; n = 6. See supplementary Figure 3C (see supplementary material online at www.gastrojournal.org) for a scatter plot of the individual data values. (B) ■, WT; □, Klf5+/−.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Figure 6Proliferation is reduced in Klf5+/− mice compared with WT mice after infection with C rodentium. (A) Frozen colonic tissues from uninfected and C rodentium–infected mice were isolated at 14 days pi and stained by immunofluorescence for Ki67 expression. (B) Numbers of Ki67-positive cells were determined on a per-crypt basis. Results represent the mean values of 6 mice, with the value of each mouse being calculated by averaging counts from at least 6 crypts per mouse. **P < .01; n = 6. See supplementary Figure 3D (see supplementary material online at www.gastrojournal.org) for a scatter plot of the individual data values for this figure. (B) ■, WT; □, Klf5+/−.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Because KLF5 has been shown to promote proliferation in vitro, we first examined whether there was a correlation between expression patterns of Klf5 and the proliferation marker, Ki67, in vivo. Immunofluorescence staining of colon tissues from WT adult mice revealed localization of Klf5 to nuclei of epithelial cells extending from the base of the crypt through approximately two thirds of the crypt height, where active proliferation occurs (Figure 1). Staining of sequential tissue sections with Ki67 showed a strikingly similar pattern to that of Klf5, suggesting that KLF5 participates in proliferative signaling pathways in the colon. To examine alterations in Klf5 expression during hyperproliferative responses in colonic tissues, C57BL/6 mice were infected with C rodentium to induce transmissible murine colonic hyperplasia. At days 8 and 14 pi, crypt heights were increased up to 2-fold in response to infection (Figure 2A). With induction of hyperproliferation, total numbers of epithelial cells per crypt increased by 1.8-fold at 14 days pi (Figure 2B), and numbers of cells expressing Klf5 visibly increased with crypt heights (Figure 2A). Furthermore, fluorescence staining of Klf5 showed higher intensities of Klf5 staining on a per-cell basis in the infected colon (Figure 2A and C). Over the time course of infection, Klf5 expression remained confined to the lower two thirds of the crypt in a pattern similar to that of Ki67 at each time point examined (Figure 2A). To study the direct involvement of Klf5 in hyperproliferative responses triggered by C rodentium infection, Klf5-heterozygous knockout (Klf5+/−) mice were generated. Mouse embryonic stem cells with a disruption in one allele of the Klf5 locus were used to produce Klf5+/− mice on a C57BL/6 background. The Klf5+/− embryonic stem cell line was acquired from BayGenomics, Inc., and the disrupted allele was generated by random insertion of a gene trap vector into the first intron of the Klf5 gene (Figure 3A). Klf5+/− mice were healthy, fertile, and appeared phenotypically normal. In concordance with previous studies using Klf5 knockout mice, no Klf5−/− mice were acquired, confirming embryonic lethality of the Klf5-homozygous knockout. Initial characterization of colon tissues from Klf5+/− mice revealed that Klf5 protein levels indeed were reduced in the heterozygous animals compared with WT animals as determined by Western blot analysis and immunofluorescence staining of frozen tissue sections (Figure 3B and C). Examination of H&E-stained colon tissues revealed no overt morphologic differences in the Klf5+/− mice compared with WT mice except for a modest reduction in crypt heights as described later. Expression of Klf5 in the Klf5+/− mice also was examined in other regions of the gastrointestinal tract, including the squamous epithelium of the esophagus, the secretory tissues of the stomach, and the mucosa of the small intestine (supplementary Figure 1; see supplementary material online at www.gastrojournal.org). Similar to the colonic tissues, Klf5 levels were reduced significantly in each of these tissues in the Klf5+/− mouse, as determined by immunofluorescence staining. To assess the involvement of Klf5 in hyperproliferative effects triggered by colonization of the colon with C rodentium, crypt heights were examined in both C57BL/6 WT and Klf5+/− littermates at 14 days pi. In infected WT mice, crypt heights were increased 2-fold compared with uninfected controls (100% ± 5% increase), although this effect was suppressed significantly in the infected Klf5+/− animals with an increase of only 77% ± 5% over control (P < .01, n = 6) (Figure 4A). A difference in crypt heights also was seen in the uninfected Klf5+/− mice compared with their WT counterparts (an 11% reduction in crypt heights of Klf5+/− mice compared with WT); however, the impact of reduced Klf5 expression on crypt heights was more pronounced in the infected mice, with crypt heights being 21% lower in the Klf5+/− infected mice compared with WT infected animals (Figure 4B). Similar responses were seen when comparing epithelial cell counts for the various conditions with the exception that the difference between uninfected WT and Klf5+/− animals was not statistically significant (Figure 4C). With infection of WT mice, numbers of epithelial cells per crypt increased by 85% ± 3%, whereas epithelial cells in Klf5+/− mice increased by only 63% ± 2% in response to infection, with these differences being statistically significant (P < .01, n = 6). To confirm that changes in Klf5 levels correlated with hyperproliferation, Klf5 protein expression was examined by immunofluorescence staining of colonic tissues from uninfected and infected C57BL/6 and Klf5+/− mice (Figure 5A). As was quantified in Figure 4B, total numbers of epithelial cells were increased in the crypts of infected WT mice, with epithelial crowding at the base of the crypts. On examining intensities of Klf5 staining in colonic epithelial cells of infected mice, both WT and Klf5+/− mice responded to infection by increasing Klf5 levels. In both genotypes, Klf5 levels were increased by 70% over levels in respective uninfected controls (Figure 5B), although this increase was to a lesser absolute level in the Klf5+/− mice owing to the absence of one allele. Thus, the single wild-type allele in the Klf5+/− mice responded in a manner comparable with that of the 2 wild-type alleles, indicating that there were no compensatory changes in Klf5 expression in the infected heterozygous animals due to Klf5 haploinsufficiency. Comparing Klf5 intensities in WT vs Klf5+/− mice after C rodentium infection, Klf5 staining on a per-cell basis was reduced by 45% in the heterozygous animals compared with WT mice. Thus, reduced levels of Klf5 correlated with reduced crypt heights and reduced expansion of epithelial cells in the infected mice. In contrast, the loss of one Klf5 allele in the uninfected animals had a minimal effect on crypt heights and numbers of epithelial cells, suggesting that Klf5 is more critical for responding to external stimuli than for maintaining homeostasis in the crypts. In examining epithelial cell proliferation in WT mice infected with C rodentium, immunofluorescence staining of colon tissues from infected mice showed an 81% ± 3% increase in numbers of Ki67-positive epithelial cells per crypt in infected C57BL/6 mice compared with uninfecte
Referência(s)