CC BY-NC-ND
0Stem Cell Research (2014) 14, 10-19
Available online atwww.sciencedirect.com
ScienceDirect
www.elsevier.com/locate/scr
KrUppel-like factor 5 is essential for (J)^^
proliferation and survival of mouse intestinal epithelial stem cells
Mandayam O. Nandan, Amr M. Ghaleb, Agnieszka B. Bialkowska, Vincent W. Yang *
Department of Medicine, Stony Brook University School of Medicine, HSC-T16 Room 020, Stony Brook, NY 11794, United States Received 31 July 2014; received in revised form 29 September 2014; accepted 29 October 2014
Abstract Kruppel-like factor 5 (KLF5) is a pro-proliferative transcription factor that is expressed in dividing epithelial cells of the intestinal crypt. Leucine-rich repeat-containing G-protein coupled receptor 5 (Lgr5) has been identified as a stem cell marker in both small intestinal and colonic epithelial cells. To determine whether KLF5 regulates proliferation of intestinal stem cells, we investigated the effects of Klf5 deletion specifically from the intestinal stem cells in adult mice. Mice with inducible intestinal stem cell-specific deletion of Klf5 (Lgr5-Klf5fl/fl) were injected with tamoxifen for 5 consecutive days to induce Lgr5-driven Cre expression. Intestinal and colonic tissues were examined by immunohistochemistry at various time points up to 112 days following start of tamoxifen treatment. Klf5 is co-localized in the crypt-based columnar (CBC) cells that express Lgr5. By 11 days following the start of tamoxifen treatment, Lgr5-positive crypts from which Klf5 was deleted exhibited a loss of proliferation that was accompanied by an increase in apoptosis. Beginning at 14 days following the start of tamoxifen treatment, both Klf5 expression and proliferation were re-established in the transit-amplifying epithelial cells but not in the Lgr5-positive CBC cells. By 112 days post-treatment, up to 90% of the Lgr5-positive cells from which Klf5 was deleted were lost from the intestinal crypts. These results indicate a critical role for KLF5 in the survival and maintenance of intestinal stem cells.© 2014 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Introduction
The mammalian intestinal epithelium is comprised of a single layer of columnar cells that are divided into two distinct compartments, a proliferative zone and a differentiated zone. The differentiated epithelial cells populate the villus and surface epithelium of the small intestine and the colon, respectively. The proliferative epithelial cells are confined
t__o* Corresponding author. Fax: +1 631 444 3144. tE-mail address: Vincent.Yang@stonybrookmedicine.edu (V.W. Yang).
he crypts and are further divided into stem cells and transit-amplifying (TA) cells. TA cells are capable of rapidly dividing after which they differentiate and move out of the proliferating zone. TA cells are subsequently replenished from the pool of stem cells that lie immediately beneath them (van der Flier and Clevers, 2009). Constant cell division in the proliferative zone results in upward movement of epithelial cells along the crypt-villus axis until they undergo apoptosis and are sloughed off at the surface of the epithelium. The differentiated epithelial cells are divided into absorptive (enterocytes) and secretory (goblet and enteroendocrine cells) lineages in both the small and large bowel (van der Flier and Clevers, 2009).
http://dx.doi.Org/10.1016/j.scr.2014.10.008
1873-5061/© 2014 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.Org/licenses/by-nc-nd/4.0/).
Adult stem cells have been traditionally defined as undiffer-entiated cells that have the capacity to undergo self-renewal and proliferate while retaining the ability to differentiate into multiple cell types (multipotency) in various situations including regeneration after injury (Potten and Loeffler, 1990). In the intestinal epithelial tissue, stem cells have long been considered to reside at the bottom of the crypt (Potten et al., 1974; Cheng and Leblond, 1974), within the first 4-5 cells from the very bottom (Bjerknes and Cheng, 1981). Monoclonality of intestinal crypts from single progenitors has also been well established (Ponder et al., 1985). Several labs have laid claim to identifying the location of intestinal adult stem cells. A convincing claim was that the cell at position +4 from the bottom of the crypt is the "actual stem cell" due to its label retention capacity, slow division rate and ability to regenerate upon crypt damage (Potten et al., 1974, 1997; Marshman et al., 2002). However, the crypt-based columnar (CBC) cells, first described in 1974 (Cheng and Leblond, 1974), were also proposed as the putative small intestinal stem cells (Bjerknes and Cheng, 1981). There were two proposed models for the maintenance of the stem cell population; immortal or asymmetric model and the stochastic or symmetric model. The immortal model was based on the idea of asymmetric segregation of DNA strands in the dividing stem cells giving rise to two distinct daughter cells, one stem cell and the other a TA cell (Potten et al., 2002). This model has been challenged by the stochastic model in which the segregation of DNA in the stem cells happens randomly (Schepers et al., 2011). It is suggested that the crypt stem cells possess similar fates and the crypts achieve monoclonality over time through "neutral drift" suggesting a pool of equipotent stem cells (Snippert et al., 2010; Lopez-Garcia et al., 2010).
Intestinal stem cells have been identified by various markers, such as musashi-1 (Msi1), hairy enhancer of split-1 (Hes1), CD133, leucine-rich repeats and immunoglobulin-like domains protein 1 (Lrig1) (Kayahara et al., 2003; Potten et al., 2003; O'Brien et al., 2007; Wong et al., 2012). More recently the leucine-rich repeat-containing G-protein coupled receptor 5 (Lgr5) has also been established as the protein marking homeostatic intestinal epithelial stem cells (Barker et al., 2007). Lgr5 is expressed exclusively in the CBC cell compartment. Using mice expressing Cre recombinase from the Lgr5 promoter in lineage tracing experiments, it was shown that Lgr5-positive CBC cells gave rise to all the differentiated cell types in the intestinal epithelium (Barker et al., 2007). Furthermore, several stem cell genes, including Ascl2 and Olfm4, were also expressed in the Lgr5-positive stem cell compartment (Munoz et al., 2012). Finally, single Lgr5-positive stem cells are capable of generating crypt-villus structures in vitro (Sato et al., 2009).
Krüppel-like factors (KLFs) are zinc finger-containing transcription factors that are involved in numerous cellular processes including differentiation, proliferation, cell cycle and apoptosis (Ghaleb et al., 2005; McConnell et al., 2007; McConnell and Yang, 2010). In the intestinal epithelium, KLF4 and KLF5 are expressed in the differentiated and proliferating compartments, respectively (Ghaleb et al., 2005; McConnell et al., 2007; McConnell and Yang, 2010). KLF5 has been shown to regulate proliferation in the intestinal crypts both during homeostasis (McConnell et al., 2011; Nandan et al., 2014) and during tumorigenesis (Nandan etal., 2008, 2010; McConnell et al., 2009). Both whole-body homozygous deletion of Kf5
(Shindo et al., 2002) and villin-Cre mediated intestine-specific deletion in mice (McConnell et al., 2011) were not well tolerated. Inducible intestine-specific deletion of Klf5 in mice led to a phenotype that showed impaired proliferation followed by a regenerative response (Nandan et al., 2014).
In the current study, we focused on the effects of inducible Klf5 deletion from the Lgr5-expressing intestinal stem cells using the Lgr5-EGFP-IRES-CreERT2 mice (Barker et al., 2007). The impact of such deletion was examined both at short-term, or acute (up to 11 days following the start of tamoxifen treatment), and long-term or chronic (up to 112 days). Upon acute deletion of Klf5 from intestinal stem cells, we observed a loss of Klf5 expression starting initially from the bottom and gradually extending to the top of the crypt. We also noticed a decrease in proliferation and a corresponding increase in apoptosis within the crypt epithelium when Klf5 was deleted. The impact of chronic Klf5 deletion from the mouse intestinal stem cells was characterized by the recovery of both Klf5 expression and proliferation in TA cells but not in the Lgr5-positive CBC cells, the majority of which were lost by 112 days following tamoxifen treatment. These results indicate an important role for Klf5 in the maintenance and survival of the intestinal stem cells.
Methods Mice
All studies involving mice were approved by the Stony Brook University Institutional Animal Care and Use Committee (IACUC) and maintained on a 12:12 h light-dark cycle. C57BL/6 J mice carrying Klf5 alleles flanked by loxP sites (Klf5fl/fl) were previously described (Takeda et al., 2010) and graciously provided by Dr. Ryozo Nagai. C57BL/6 mice carrying the inducible Cre recombinase gene under regulation of the Lgr5 promoter (Lgr5-EGFP-IRES-CreERT2 mice; henceforth designated as Lgr5-Cre) were purchased from Jackson labs (ME, USA). To establish Lgr5-EGFP-IRES-CreERT2/Klf5fl/fl (designated as Lgr5-Klf5fl/fl) mice, Klf5fl/fl mice were initially crossed with Lgr5-Cre mice and its progeny then backcrossed to Klf5fl/fl mice. For all experiments, Lgr5-Cre mice were used as controls. Tamoxifen treatment of mice was per protocol previously described (el Marjou et al., 2004) (Suppl. Fig. 1). Eight week old Lgr5-Klf5fl/fl and Lgr5-Cre mice were injected intraperitoneally (I.P.) with 1 mg of tamoxifen (10 mg/ml, dissolved in sterile corn oil, Sigma, MO, USA) for five consecutive days. Animals were sacrificed on days 3, 5, 7 and 11 after the start of tamoxifen injections for short-term (acute) tamoxifen-treatment study (Suppl. Fig. 1A) and on days 14, 28, 56 and 112 for the long-term (chronic) tamoxifen-treatment study (Suppl. Fig. 1B).
Antibodies
Antibodies used in the experiments were previously described (Nandan et al., 2014). Klf5 rabbit polyclonal antibody (Santa Cruz Biotechnologies, CA, USA; Cat# sc-22797) was used at 1:150 dilution. Rabbit monoclonal anti-Ki67 antibody was purchased from Biocare Medical (CA, USA; Cat#CRM 325) and used at 1:500 dilutions. Chicken polyclonal anti-GFP antibody was purchased from Millipore (MA, USA; Cat# AB16901) was
used for immunofluorescence staining at 1:500 dilution. Rabbit polyclonal cleaved caspase 3 antibody was purchased from Cell Signaling (MA, USA; Cat# 9664) and used at 1:250 dilution.
Immunohistochemistry and immunofluorescence
Immunohistochemical analysis was performed as previously described protocol with modifications (McConnell et al., 2011; Nandan etal., 2010, 2014). Intestinal tissues dissected from mice were flushed with modified Bouine's fixative (50% ethanol + 5% acetic acid in water), then fixed overnight in 10% buffered formalin (Fisher Scientific, PA, USA). The tissues were then paraffin-embedded using an automated processor and sectioned at 5 |im. The sections were then dried onto charged slides and used for staining. The slides were baked in a 65 °C oven for 1 h and were subsequently deparaffinized in xylene. Sections were incubated in a 3% hydrogen peroxide in methanol bath to block endogenous tissue peroxidases and were then rehydrated by incubation in a decreasing alcohol bath series (100%, 95%, 70%) followed by antigen retrieval in citrate buffer solution (10 mM sodium citrate, 0.05% Tween-20, pH 6.0) at 125 °C for 10 min using a decloaking chamber (Biocare Medical, CA, USA). Tissue sections were first incubated with blocking buffer (2.5% BSA in TBS-Tween) for 30 min at 37 °C and then with primary antibody at 4 °C overnight in a humidified chamber while shaking gently. Sections were washed and incubated with secondary antibodies (HRP-conjugated or fluorescent-tagged) at the appropriate concentration for 30 min at 37 °C. Betazoid DAB (Biocare Medical, CA, USA) was used to reveal IHC staining in tissues. For fluorescent sections, slides were washed after secondary antibody treatment and then stained with DAPI (Fisher Scientific, PA, USA) and mounted with Prolong gold antifade (Life Technologies, CA, USA). For EGFP co-staining with Klf5, EGFP fluorescent secondary antibody detection was performed first then followed by DAB for Klf5.
For Klf5/GFP/Ki-67 co-staining, slides were processed and Klf5 antibody staining was performed overnight, as described above. Mach3 rabbit AP polymer detection was applied to slides as per protocol (Biocare Medical, CA, USA). Klf5 staining was revealed using Warp Red chromogen (Biocare Medical, CA, USA). Antibody elution was performed using previously described protocol (Pirici et al., 2009). Slides were incubated at 50 °C in Glycine-SDS (pH 2.0) solution with agitation for 1 h. Slides were then thoroughly washed with distilled water and incubated with blocking buffer (2.5% BSA in TBS-Tween) for 30 min at 37 °C. Then Ki-67 and GFP antibodies were added and incubated at 37 °C for 1 h. Fluorescent detection was performed with appropriate secondary antibodies before staining with DAPI (Fisher Scientific, PA, USA) and mounted with Prolong gold antifade (Life Technologies, CA, USA). Slides were observed under a Nikon Eclipse 90i microscope (Nikon, NY, USA) and representative pictures were taken. Morphology of sections was observed upon staining 5 |im sections with hematoxylin and eosin (H&E) (Vector Labs, CA, USA).
Statistical analysis
Statistical comparisons between Klf5 positive and Klf5/Ki-67 co-positive cells in EGFP+ and EGFP- crypts (50 randomly chosen
from jejunum, n = 3) were performed using the paired 2-tailed Student's t-test (n = 3). The number of EGFP positive crypts in the Lgr5-Klf5tl/fl mice upon chronic Klf5 deletion was compared to Lgr5-Cre controls using 2-tailed Student's t-test (n = 3). Error bars in all graphs represent the standard error of mean.
Results
Klf5 is expressed in the stem cells of the mouse small intestine and colon
Lgr5 has been recently identified as a marker for intestinal stem cells through studies in an EGFP-tagged Lgr5-Cre mouse model (Barker et al., 2007). Expression of Lgr5 is localized to the actively and steadily proliferating columnar cells in the crypt base (CBC cells) (Barker et al., 2007). Because of its pro-proliferative effect, we first determined whether Klf5 is also expressed in Lgr5-expressing CBC cells. As seen in Fig. 1, Klf5 is localized in the nuclei throughout the crypt epithelial zone of both small intestine and colon (Figs. 1A, C, D and F). Lgr5-Cre (as marked by EGFP) is localized to the CBC cells in both tissues (Figs. 1B, C, Eand F). Importantly, Lgr5-positive CBC cells in the small intestine and colon co-stained for Klf5 (Figs. 1C and F). It was noted that the Klf5 staining in the CBC cells was weaker than that in the transit-amplifying (TA) region located immediately above the stem cell zone.
Klf5 is progressively lost from the CBC cells throughout the entire intestinal crypt during short-term treatment of Lgr5-Klf5fl/fl mice with tamoxfen
We then examined the effect of short-term or acute deletion of Klf5 (up to 11 days) from Lgr5-positive CBC cells in Lgr5-Klf5fl/fl mice treated with tamoxifen for five consecutive days (Suppl. Fig. 1A). Tamoxifen-treated Lgr5-Cre mice served as controls. There were no significant weight differences between tamoxifen-treated Lgr5-Klf5fl/fl mice and control mice during the acute period after tamoxifen treatment (data not shown). Due to the relatively low penetrance of Lgr5-Cre expression, deletion of Klf5 was limited to 5-15% of the small intestinal and colonic crypts. At day 0, both EGFP-stained green crypts (blue box, representing Lgr5-Cre expressing crypts) and non-green crypts (white box, representing crypts without Lgr5-Cre expression) of Lgr5-Klf5fl/fl mice displayed Klf5 expression (Fig. 2C and Suppl. Fig. 2C). At day 3 after the start of tamoxifen treatment, there was a loss of Klf5 expression at the bottom of green crypts corresponding to Lgr5-Cre green cells (blue box) (Fig. 2F and Suppl. Fig. 2F) compared to adjoining non-green crypts (white box). By 5 and 7 days post-tamoxifen induction, loss of Klf5 expression was observed extending up into the TA cell region and a large number of the Lgr5-Cre positive green crypts have lost Klf5 expression (blue boxes), compared to the non-green crypts (white boxes) (Figs. 2I, L and Suppl. Figs. 2I, L). Complete or near complete loss of Klf5 in all of the green crypts (blue boxes) compared to normal Klf5 expression in the non-green crypts (white boxes) was observed 11 days post-tamoxifen induction (Fig. 2O and Suppl. Fig. 2O). No significant differences in Klf5 expression was observed in the control mice at each time point (data not shown). Together these results show an incremental loss of Klf5 starting from the
CBC cells at day 3 and progressing upwards to the entire crypts by day 11 post-tamoxifen treatment of Lgr5-Klf5fl/fl mice (Suppl. Fig. 3). A similar trend was noted in the colonic tissues of tamoxifen-treated Lgr5-Klf5fl/fl mice (Suppl. Fig. 2) and Lgr5-Cre mice (data not shown).
Loss of Klf5 from intestinal stem cells is accompanied by a decrease in proliferation and an increase in apoptosis of crypt epithelial cells
We previously established that Klf5, as a pro-proliferative transcription factor (Sun et al., 2001; Nandan et al., 2004), is often co-expressed with proliferation markers, such as Ki-67 (Nandan et al., 2008). Consequently, we detected consistent co-staining of Klf5 and Ki-67 in Lgr5-Klf5fl/fl mice at day 0 in the crypt epithelial cell nuclei in both green and non-green crypts (blue and white boxes) (Fig. 3D; Suppl. Fig. 4). In the short-term tamoxifen treatment study, we noted a loss of Ki-67 staining (Figs. 3G, K, O and S; Fig. 4; Suppl. Fig. 4) only in the green crypts compared to the adjacent non-green crypts at each time point post-tamoxifen induction of Lgr5-Klf5fl/fl mice. Simultaneously, there was a reduction in number of Klf5/Ki-67 co-stained nuclei from green crypts (blue boxes) compared to adjacent non-green crypts (white boxes) (Figs. 3H, L, P and T; Fig. 4; Suppl. Fig. 4). The control mice did not show any loss of Klf5 or Ki-67 co-staining with tamoxifen treatment (data not shown). These results indicate that the loss of proliferative cells in the intestinal crypts is a consequence of Klf5 deletion from
EGFP-positive stem cells (Fig. 4). Similar data were obtained from the mouse colonic crypts (data not shown).
We then determined whether Klf5 loss from the intestinal stem cells is accompanied by apoptosis by staining tissues for cleaved caspase 3. We noticed increasing levels of cleaved caspase 3 staining in intestinal crypts from Lgr5-Klf5fl/fl mice (Figs. 5H, L, P and T) from 3 to 11 days following start of tamoxifen treatment when compared to uninduced control (Fig. 5D). It is interesting to note that a majority of the Lgr5-positive green crypts, including green CBC cells, displayed clear evidence of apoptosis at 7 and 11 days post-tamoxifen treatment (Figs. 5P and T). There was no induction of cleaved caspase 3 noted in the control mice upon tamoxifen treatment (data not shown). Similar data was obtained from tamoxifen-treated Lgr5-Klf5fl/fl and Lgr5-Cre mouse colonic crypts (data not shown). These results suggest that acute loss of Klf5 in intestinal crypt stem cells leads to a loss of proliferation and apoptosis.
Deletion of Klf5 from EGFP-positive CBC cells results in chronic loss of proliferation in stem cells
We examined the effect of chronic Klf5 deletion from intestinal stem cells by extending the study period up to 112 days following tamoxifen treatment of Lgr5-Klf5fl/fl mice. In Lgr5-Cre control mice treated with tamoxifen, both EGFP-positive (blue arrows) and EGFP-negative crypts (white arrows) co-stained for Klf5 and Ki-67 from days 14 to 112 following induction (Figs. 6D & L; Suppl. Fig. 5). In
Klf5 EGFP Merge
Figure 1 Klf5 is expressed in the stem cell zone of the mouse gut. Lgr5-Cre mouse small intestinal and colonic sections were stained for both Klf5 and EGFP (representative of Lgr5 positive epithelial cells). Panels A-C represent mouse small intestinal crypts and panels D-F represent colonic tissues. Panels A and D show Klf5 staining using DAB chromogen which stains the nucleus red while panels B and E display EGFP stain using fluorescence appearing green. Panels C and F are merged images of respective Klf5 and EGFP panels. White arrows mark the cells that stain positive for both Klf5 and EGFP. There is consistent co-staining among Klf5 and EGFP proteins signifying the expression of Klf5 in the putative mouse gut epithelial cells.
• ••»J1' > ••. - • * . V
■ -J i 4>
Wgm W .
* * - v-
. s ii * Jg. ,
Figure 2 Tamoxifen induction results in Lgr5-driven Klf5 deletion in the small intestinal epithelium of Lgr5-Klf5fl/fl mice. Representative images from untreated Lgr5-Klf5fl/fl (Panels A-C) followed by day 3, 5, 7 and 11 tamoxifen-treated Lgr5-Klf5fl/fl mice jejunum (Panels D-O). Panels A, D, G, J and M show red DAB-stained nuclei representing Klf5 expression. Panels B, E, H, K and N show EGFP fluorescent images (green stain) and panels C, F, I, L and O show merged images of both Klf5 and EGFP stains. Crypts marked in the merged image by the blue boxes represent Lgr5-Cre positive green crypts, while the white boxes represent the non-green crypts. Klf5 expression is incrementally lost in the green crypts beginning at day 3 to day 11.
contrast, Klf5/Ki-67 co-staining was absent from the EGFP-positive CBC stem cells (blue arrows) but present in the CBC cells of the adjoining non-green crypts (white arrows) (Figs. 6H & P; Suppl. Fig. 5) in Lgr5-Klf5fl/fl small intestinal tissue. Of note is that there was a restoration of Klf5/Ki-67 co-positive staining in the transit-amplifying (TA) region of the EGFP-positive crypts from day 14 to day 112 (Figs. 6H & P; Suppl. Fig. 5). The results indicate that despite persistent loss of proliferation of Lgr5-positive CBC cells due to Klf5 deletion, there is recovery of both Klf5 and Ki-67 expression in the TA cell population immediately above the stem cell zone during this time period.
Chronic deletion of Klf5 from Lgr5-Klf5fl/fl mouse intestinal tissues results in depletion of EGFP-positive Lgr5-Cre crypts
At 112 days post-tamoxifen treatment, we noted that there was a significant reduction in the number of EGFP-positive crypts in the small intestine and colon of Lgr5-Klf5fl/fl mice when compared to day 14 of treatment, which in turn was reduced when compared to Lgr5-Cre control mice (Suppl. Fig. 6). This trend became apparent when we quantified and compared the number of EGFP-positive crypts in the small intestine and colon of mice from tamoxifen-treated Lgr5-Klf5fl/fl and Lgr5-Cre mice from day 14 onwards (Fig. 7). There was no difference in the number of green
crypts of Lgr5-Cre (control) tissues at any given time point from the start of tamoxifen-treatment (Fig. 7). By day 112 post-tamoxifen treatment, approximately 90% of the EGFP-positive crypts were depleted from the intestinal tissues of Lgr5-Klf5fl/fl mice (Fig. 7). Hence, crypts that sustained long-term loss of Klf5 resulted in the disappearance of Lgr5-positive stem cells.
Discussion
The mammalian intestinal epithelium is among the most rapidly proliferating tissue in the body and this fast turnover helps protect the epithelial cells from genotoxic insults (Podolsky, 1993; Booth and Potten, 2001). Our results show that Klf5 is also expressed in the stem cells of the intestinal epithelia. The intestinal crypt epithelial cells are innately responsive to the Wnt pathway signals that regulate proliferation, cell fate and differentiation. Deregulation of the Wnt signaling pathway, especially p-catenin and its modulator proteins, is important for the initiation of tumorigenesis from the normal epithelium (Korinek et al., 1997; Morin et al., 1997; Barker etal., 2008).
Lgr5 was identified from a panel of Wnt target genes and established as an intestinal stem cell marker (Barker et al., 2007). Klf5 was also identified as a Wnt-responsive gene (Ziemer et al., 2001) and facilitates p-catenin nuclear translocation during Wnt pathway activation in intestinal
Klf5 EGFP Ki-67 Merge
Figure 3 Loss of proliferation in crypts concomitant with Klf5 deletion in Lgr5-Klf5f'/f' mice. The panels are representative of immunofluorescent staining of jejunal tissue from different time points after induction of tamoxifen in Lgr5-Klf5f'/f' mice. Panels A-D represent control jejunum from uninduced Lgr5-Klf5f'/f' mice. Panels E-T represent jejunums collected from mice after 3, 5, 7 or 11 days following the start of tamoxifen treatment. Panels A, E, I, M and Q show immunofluorescent Klf5 staining in red, panels B, F, J, N and R show EGFP expression in green and panels C, G, K, O and S show Ki-67 staining representative of proliferation in yellow. Panels D, H, L, P and T are merged images of Klf5, EGFP and Ki-67 staining with Klf5/Ki-67 co-expression showing as orange stain. Blue boxes in the merged image represent green crypts, while the white boxes represent the non-green crypts. Loss of Ki-67 expression corresponds with loss of Klf5 staining in Lgr5-Klf5f'/f' mice after tamoxifen induction.
Figure 4 Klf5 and Ki-67 co-positive cell number is reduced in mouse small intestinal tissue. Klf5/Ki-67 co-expressing cells in the crypt are plotted on the Y-axis against days post-tamoxifen-treatment on the X-axis. EGFP-positive green crypts from mice are represented as gray bars while non-
green crypts are represented as black bars. Klf5/Ki-67 co-positive cells are reduced in L.grS-M/S^' green crypts after tamoxifen induction compared to non-green crypts. (**—p < 0.05).
epithelial cells (McConnell et al., 2011). KLF5 has been shown to be a transcription factor highly enriched in this rapidly proliferating epithelial cells and its loss has been associated with increased mortality and severe phenotypic defects (McConnell et al., 2011; Shindo et al., 2002). The current study shows that Klf5 is expressed in both the intestinal stem cell and TA compartments and that Klf5 expression in the Lgr5-expressing stem cells is lower than the adjoining TA epithelial cell compartment (Fig. 1). The Lgr5-positive mouse intestinal stem cells are long-living and cycle about every 22 h (Schepers et al., 2011). In contrast, the TA cells are short-living, rapidly proliferate with a turnover rate of 12 h and complete 4-5 rounds of cell division before they differentiate and exit the crypts (Schepers et al., 2011). We postulated that since Klf5 is intimately involved in proliferation, its high expression is necessary for sustenance and division of the short-living TA cell compartment. The stem cells are relatively slower cycling compared to the TA cell compartment and hence express lower levels of Klf5 (Fig. 1).
In this study, we were able to successfully delete Kf5 from the Lgr5-expressing intestinal stem cells in adult mice. The short-term effect of Ki/5 deletion in the intestinal and colonic stem cells was a gradual disappearance of Klf5 expression starting in the EGFP-positive stem cells and moving upwards, eventually resulting in total loss of Klf5 expression from the entire crypt at 11 days post-tamoxifen treatment (Fig. 2; Suppl. Fig. 2). This result supports the lineage tracing
DAPI EGFP CI. Caspase 3 Merge
Figure 5 Increase in apoptotic cells upon acute Klf5 deletion in Lgr5-Klf5f'/f' small intestinal crypt stem cells. Apoptosis in jejunal crypts is represented by positive cleaved caspase 3 staining. Staining of control jejunum from uninduced Lgr5-Klf5f'/f' mice is represented in panels A-D. Jejunums collected from mice after 3, 5, 7 and 11 days following the start of tamoxifen treatment are represented in panels E through T. Panels A, E, I, M and Q show nuclear DAPI staining in blue, panels B, F, J, N and R show EGFP expression in green and panels C, G, K, O and S show cleaved caspase 3 staining in red. Panels D, H, L, P and T are merged images of cleaved caspase 3, EGFP and DAPI stains where yellow staining represents EGFP stained stem cells that are positive for cleaved caspase 3. EGFP-positive crypts are marked with a yellow box while non-green crypts are marked with a white box, in the merged images. Klf5 deletion results in increased apoptosis in the stem cell zone of Lgr5-Klf5f'/f' mice after tamoxifen induction compared with uninduced controls.
studies with the Lgr5-Cre mice where daughter cells arising from the stem cells give rise to adjoining TA cells and migrate towards the villus (Barker et al., 2007). Tamoxifen-treated Lgr5-Ki/5f'/f' mice do not display the morbidity or mortality associated with complete embryonic deletion of Kf5 (McConnell et al., 2011; Shindo et al., 2002), possibly due to the low penetrance of Lgr5-Cre expression. These Lgr5-Ki/5f'/f' mice do show a loss of proliferation in the crypts corresponding with Klf5 loss (Figs. 3 and 4; Suppl. Fig. 4). We also noted that the EGFP-positive crypts display an increase in apoptotic cells that accompany Ki/5 loss (Fig. 5). The combinatorial loss of proliferation and increased apoptosis in crypts following the loss of Kf5 suggests that Klf5 is crucial for the maintenance of intestinal stem cells in vivo.
The effect of chronic Kf5 deletion in the intestinal stem cells was studied for 112 days after tamoxifen treatment. We observed prolonged loss of Klf5 and Ki-67 expression in the EGFP-positive crypt stem cells for the duration of the study (Fig. 6; Suppl. Fig. 5). However, both Klf5 and Ki-67 expression recovered in the TA cell compartment by 14 days post-tamoxifen treatment and continued till 112 days. We also noted a progressive decrease in the number of EGFP-positive crypts in the intestinal tissues of tamoxifen-treated Lgrô-Ki/S^ f' mice (Fig. 7; Suppl. Fig. 6). About 80-90% of small intestinal
and colonic EGFP-positive crypts were lost in the tamoxifen-treated Lgrô-Ki/.?7^ tissues when compared with the tamoxifen-treated Lgr5-Cre tissues (Fig. 7; Suppl. Fig. 6). This result suggests that the non-dividing stem cell population caused by Ki/5 deletion has a finite life-time, after which they become depleted. Apart from the gradual loss of EGFP-positive crypts from Lgr5-Ki/5f'/f' tissues, we also observed a decrease in number of EGFP-positive cells per EGFP-positive crypt (data not shown). These results, together with the re-occurrence of Klf5 and Ki67 positive cells in the TA cells, support a model where the prolonged loss of Klf5 expression from intestinal stem cells results in their gradual replacement by alternative sources that serve as a reserve. A mouse model with a more robust intestinal stem cell-specific Cre recombinase expression could potentially display catastrophic consequences upon Ki/5 loss.
Our understanding of intestinal stem cells and their modes of action has been considerably expanded. The +4 label retaining cell population that was once considered as the stem cells (Sangiorgi and Capecchi, 2008) has been rechristened as the quiescent stem cells expressing Bmi1, among other factors (Munoz et al., 2012; Li and Clevers, 2010). The Bmi1-labeled quiescent stem cells have gained a new identity as a reserve stem cell population that can get induced upon intestinal injury
Klf5 EGFP Ki-67 Merge
Figure 6 Klf5 deletion in Lgr5-Klf5f'/f' mice persists long-term in Lgr5-Cre positive green crypts. The top and bottom set of images are representative small intestinal tissue from Lgr5-Cre control and Lgr5-Klf5f'/f' mice after 14 days and 112 days of tamoxifen treatment, respectively. Panels A-D and I-L are representative of staining from tissue collected from Lgr5-Cre control mice while panels E-H and M-P show staining representative of Lgr5-Klf5f'/f' mice. Panels A, E, I and M display Klf5 immunofluorescent staining in red; panels B, F, J and N show EGFP staining in green; panels C, G, K and O show Ki-67 staining in yellow and panels D, H, L and P show merged images of Klf5, EGFP and Ki-67 stains. Blue arrows point to green crypts while white arrows point to non-green crypts in the merged images. Lgr5-Klf5f'/f' mice show long-term loss of Klf5 only in the EGFP labeled CBC cells.
and replace the underlying Lgr5 stem cell population (Yan et al., 2012). In the current model, whether cells expressing Bmi1 or other known quiescent stem cell marker may serve as the
source of crypt cell replacement in crypts that have lost Klf5 from the Lgr5-expressing stem cell population is an open question that begs further investigation.
0 14 28 56 112 0 14 28 56 112
Days post tamoxifen Days post tamoxifen
Figure 7 Loss of Lgr5-positive green crypts in Lgr5-K7/:5f'/f' mice upon chronic K//5 deletion. Quantitation of Lgr5-positive green crypts is graphically represented in panels A & B with number of EGFP positive crypts on the Y-axis and days post-tamoxifen treatment on the X-axis. Quantitation of jejunal tissues are represented in panel A and colonic tissues in panel B. Tissues from control Lgr5-Cre mice are represented as black bars while Lgr5-K//5f'/f' tissues are shown as gray bars in both panels. There is no change in the number of green crypts at day 0, while there is a steady decrease in EGFP-positive crypt count starting from day 14 till day 112 in both small intestinal and colonic tissues. Results are the average of four sections assayed from two independent experiments. ** p < 0.05.
A recent study (Nakaya et al., 2014), published during the preparation of this manuscript, has reported similar defects upon Lgr5-mediated loss of Klf5 in mice. They observed that Klf5 was obligatory for survival and transformation of intestinal epithelial cells. They also showed a reduction of proliferation and increase in apoptotic bodies upon Klf5 deletion from the intestinal stem cells that was dependent on nuclear localization of p-catenin (Nakaya et al., 2014). The authors further defined the importance of Klf5 expression in promoting Apc/ p-catenin mediated tumorigenesis. They postulated that Klf5 was a regulator of homeostatic and neoplastic proliferation in intestinal epithelial cells (Nakaya et al., 2014).
Our study reports that the loss of Klf5 from intestinal stem cells leads to reduction in proliferation eventually resulting in the loss of stem cells. We propose three possible outcomes of Klf5 deletion in stem cells that could elucidate our findings. Firstly, the replenishment of the TA cell population in the EGFP positive crypts could be due to drift from the adjoining non green crypt stem cells. Secondly, the pre-existing quiescent stem cell (Bmi1 +) population could be activated due to the lack of proliferation from the native stem cell population and function as the source of TA cell population. Thirdly, in conjunction with the emerging "neutral drift" model for competing stem cells (Snippert et al., 2010; Lopez-Garcia et al., 2010), equipotent competition of stem cells within the EGFP positive crypts could result in clonality arising from a neighboring unlabeled stem cell. The identity of the origin of these replacing crypt epithelial cells remains to be explored.
Conclusion
We have shown that intestinal stem cell proliferation is dependent on Klf5 expression. Stem cells that have lost Klf5 expression show an increase in apoptotic bodies. Disappearance of stem cells was observed as a chronic consequence of Klf5 deletion. Overall, our study shows that Klf5 is necessary and critical for the survival of intestinal crypt cells.
Supplementary data to this article can be found online at http://dx.doi.org/10.1016Zj.scr.2014.10.008.
Acknowledgments
We thank Dr. Ryozo Nagai for generously providing the floxed Klf5 mice. We also thank the National Institutes of Health for grant support to VWY (DK52230, DK64399, and CA84197).
References
Barker, N., van Es, J.H., Kuipers, J., et al., 2007. Identification of stem cells in small intestine and colon by marker gene Lgr5. Nature 449, 1003-1007.
Barker, N., van de Wetering, M., Clevers, H., 2008. The intestinal
stem cell. Genes Dev. 22, 1856-1864. Bjerknes, M., Cheng, H., 1981. The stem-cell zone of the small intestinal epithelium. V. Evidence for controls over orientation of boundaries between the stem-cell zone, proliferative zone, and the maturation zone. Am. J. Anat. 160, 105-112. Booth, D., Potten, C.S., 2001. Protection against mucosal injury by growth factors and cytokines. J. Natl. Cancer Inst. Monogr. 16-20.
Cheng, H., Leblond, C.P., 1974. Origin, differentiation and renewal of the four main epithelial cell types in the mouse small intestine. V. Unitarian Theory of the origin of the four epithelial cell types. Am. J. Anat. 141, 537-561.
elMarjou, F., Janssen, K.P., Chang, B.H., etal., 2004. Tissue-specific and inducible Cre-mediated recombination in the gut epithelium. Genesis 39, 186-193.
Ghaleb, A.M., Nandan, M.O., Chanchevalap, S., et al., 2005. Kruppel-like factors 4 and 5: the yin and yang regulators of cellular proliferation. Cell Res. 15, 92-96.
Kayahara, T., Sawada, M., Takaishi, S., et al., 2003. Candidate markers for stem and early progenitor cells, Musashi-1 and Hes1, are expressed in crypt base columnar cells of mouse small intestine. FEBSLett. 535, 131-135.
Korinek, V., Barker, N., Morin, P.J., et al., 1997. Constitutive transcriptional activation by a beta-catenin-Tcf complex in APC -/- colon carcinoma. Science 275, 1784-1787.
Li, L., Clevers, H., 2010. Coexistence of quiescent and active adult stem cells in mammals. Science 327, 542-545.
Lopez-Garcia, C., Klein, A.M., Simons, B.D., et al., 2010. Intestinal stem cell replacement follows a pattern of neutral drift. Science 330, 822-825.
Marshman, E., Booth, C., Potten, C.S., 2002. The intestinal epithelial stem cell. Bioessays 24, 91-98.
McConnell, B.B., Yang, V.W., 2010. Mammalian Kruppel-like factors in health and diseases. Physiol. Rev. 90, 1337-1381.
McConnell, B.B., Ghaleb, A.M., Nandan, M.O., et al., 2007. The diverse functions of Kruppel-like factors 4 and 5 in epithelial biology and pathobiology. Bioessays 29, 549-557.
McConnell, B.B., Bialkowska, A.B., Nandan, M.O., et al., 2009. Haploinsufficiency of Kruppel-like factor 5 rescues the tumor-initiating effect of the Apc(Min) mutation in the intestine. Cancer Res. 69,4125-4133.
McConnell, B.B., Kim, S.S., Yu, K., etal., 2011. Kruppel-like factor 5 is important for maintenance of crypt architecture and barrier function in mouse intestine. Gastroenterology 141, 1302-1313 (313 e1301-1306).
Morin, P.J., Sparks, A.B., Korinek, V., et al., 1997. Activation of beta-catenin-Tcf signaling in colon cancer by mutations in beta-catenin or APC. Science 275, 1787-1790.
Munoz, J., Stange, D.E., Schepers, A.G., et al., 2012. The Lgr5 intestinal stem cell signature: robust expression of proposed quiescent '+4' cell markers. EMBO J. 31, 3079-3091.
Nakaya, T., Ogawa, S., Manabe, I., et al., 2014. KLF5 regulates the integrity and oncogenicity of intestinal stem cells. Cancer Res. 74, 2882-2891.
Nandan, M.O., Yoon, H.S., Zhao, W., et al., 2004. Kruppel-like factor 5 mediates the transforming activity of oncogenic H-Ras. Oncogene 23, 3404-3413.
Nandan, M.O., McConnell, B.B., Ghaleb, A.M., et al., 2008. Kruppel-like factor 5 mediates cellular transformation during oncogenic KRAS-induced intestinal tumorigenesis. Gastroenterology 134, 120-130.
Nandan, M.O., Ghaleb, A.M., McConnell, B.B., et al., 2010. Kruppel-like factor 5 is a crucial mediator of intestinal tumorigenesis in mice harboring combined ApcMin and KRASV12 mutations. Mol. Cancer 9, 63.
Nandan, M.O., Ghaleb, A.M., Liu, Y., et al., 2014. Inducible intestine-specific deletion of Kruppel-like factor 5 is characterized by a regenerative response in adult mouse colon. Dev. Biol. 387, 191-202.
O'Brien, C.A., Pollett, A., Gallinger, S., et al., 2007. A human colon cancer cell capable of initiating tumour growth in immunodefi-cient mice. Nature 445, 106-110.
Pirici, D., Mogoanta, L., Kumar-Singh, S., et al., 2009. Antibody elution method for multiple immunohistochemistry on primary antibodies raised in the same species and of the same subtype. J. Histochem. Cytochem. 57, 567-575.
Podolsky, D.K., 1993. Regulation of intestinal epithelial proliferation: a few answers, many questions. Am. J. Physiol. 264, G179-G186.
Ponder, B.A., Schmidt, G.H., Wilkinson, M.M., et al., 1985. Derivation of mouse intestinal crypts from single progenitor cells. Nature 313, 689-691.
Potten, C.S., Loeffler, M., 1990. Stem cells: attributes, cycles, spirals, pitfalls and uncertainties. Lessons for and from the crypt. Development 110, 1001-1020.
Potten, C.S., Kovacs, L., Hamilton, E., 1974. Continuous labelling studies on mouse skin and intestine. Cell Tissue Kinet. 7, 271 -283.
Potten, C.S., Booth, C., Pritchard, D.M., 1997. The intestinal epithelial stem cell: the mucosal governor. Int. J. Exp. Pathol. 78, 219-243.
Potten, C.S., Owen, G., Booth, D., 2002. Intestinal stem cells protect their genome by selective segregation of template DNA strands. J. Cell Sci. 115, 2381-2388.
Potten, C.S., Booth, C., Tudor, G.L., et al., 2003. Identification of a putative intestinal stem cell and early lineage marker; musashi-1. Differentiation 71, 28-41.
Sangiorgi, E., Capecchi, M.R., 2008. Bmi1 is expressed in vivo in intestinal stem cells. Nat. Genet. 40, 915-920.
Sato, T., Vries, R.G., Snippert, H.J., et al., 2009. Single Lgr5 stem cells build crypt-villus structures in vitro without a mesenchymal niche. Nature 459, 262-265.
Schepers, A.G., Vries, R., van den Born, M., et al., 2011. Lgr5 intestinal stem cells have high telomerase activity and randomly segregate their chromosomes. EMBOJ. 30, 1104-1109.
Shindo, T., Manabe, I., Fukushima, Y., et al., 2002. Kruppel-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.
Snippert, H.J., van der Flier, L.G., Sato, T., et al., 2010. Intestinal crypt homeostasis results from neutral competition between symmetrically dividing Lgr5 stem cells. Cell 143, 134-144.
Sun, R., Chen, X., Yang, V.W., 2001. Intestinal-enriched Kruppel-like factor (Kruppel-like factor 5) is a positive regulator of cellular proliferation. J. Biol. Chem. 276, 6897-6900.
Takeda, N., Manabe, I., Uchino, Y., et al., 2010. Cardiac fibroblasts are essential for the adaptive response of the murine heart to pressure overload. J. Clin. Invest. 120, 254-265.
van der Flier, L.G., Clevers, H., 2009. Stem cells, self-renewal, and differentiation in the intestinal epithelium. Annu. Rev. Physiol. 71, 241-260.
Wong, V.W., Stange, D.E., Page, M.E., et al., 2012. Lrig1 controls intestinal stem-cell homeostasis by negative regulation of ErbB signalling. Nat. Cell Biol. 14, 401-408.
Yan, K.S., Chia, L.A., Li, X., et al., 2012. The intestinal stem cell markers Bmi1 and Lgr5 identify two functionally distinct populations. Proc. Natl. Acad. Sci. U. S. A. 109, 466-471.
Ziemer, L.T., Pennica, D., Levine, A.J., 2001. Identification of a mouse homolog of the human BTEB2 transcription factor as a beta-catenin-independent Wnt-1-responsive gene. Mol. Cell. Biol. 21, 562-574.
