Intestinal Hedgehog signaling in tumors and inflammation
Büller, N.V.J.A.
Publication date
2015
Document Version
Final published version
Link to publication
Citation for published version (APA):
Büller, N. V. J. A. (2015). Intestinal Hedgehog signaling in tumors and inflammation.
General rights
It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s)
and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open
content license (like Creative Commons).
Disclaimer/Complaints regulations
If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please
let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material
inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter
to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You
will be contacted as soon as possible.
TINAL
HEDGEHOG
SIGNALING
IN
TUMOR
S
AND
INFL
AMMA
TION
NIKÈ
V
JA
BÜLLER
NIKÈ VJA BÜLLER
INTESTINAL
HEDGEHOG
SIGNALING
IN TUMORS AND INFLAMMATION
in tumors and inflammation
ACADEMISCH PROEFSCHRIFT
ter verkrijging van de graad van doctor
aan de Universiteit van Amsterdam
op gezag van de Rector Magnificus
prof. dr. D.C. van den Boom
ten overstaan van een door het college voor promoties ingestelde
commissie, in het openbaar te verdedigen in de Aula der Universiteit
op vrijdag 22 mei 2015, te 11.00 uur
door
Nike Vittoria Justitia Anna Büller
geboren te Amsterdam
Intestinal Hedgehog Signaling in tumors and inflammation Thesis, University of Amsterdam, The Netherlands N.V.J.A. Büller, 2015
Cover: Britte Hietkamp
Layout: Persoonlijkproefschrift.nl, Robbert de Vries Printed by: Ipskamp Drukkers BV
ISBN 978-94-6259-629-0
All rights reserved. No part of this thesis may be reproduced or transmitted in any form or by any means, without prior permission of the author.
The financial support for printing of this thesis by the following foundations and companies are gratefully acknowledged:
Universiteit van Amsterdam, Nederlandse Vereniging van Gastroenterologie, Chipsoft B.V., Abbvie B.V.
Copromotores: Dr. V. Muncan Dr. M.E. Wildenberg
Overige leden: Prof. dr. E. Dekker Prof. dr. J.C.H. Hardwick Prof. dr. J.P. Medema Prof. dr. C.J.A. Punt Prof. dr. E.H.H.M. Rings Prof. dr. R. Toftgård
Faculteit der Geneeskunde
Chapter 1 General Introduction
Chapter 2 Loss of Indian Hedgehog activates multiple aspects of
a wound healing response in the mouse intestine
Chapter 3 Stromal Indian Hedgehog Signaling is required for
intestinal adenoma formation in mice
Chapter 4 Intestinal Hedgehog Signaling suppresses the
interferon response in stromal cells
Chapter 5 Protective effect of stromal Hedgehog signaling
against colon cancer progression
Chapter 6 Rage signaling promotes intestinal tumorigenesis
Chapter 7 Summary and Perspective
Nederlandse samenvatting List of abbreviations Contributing authors Dankwoord 7 25 57 83 103 125 139 145 151 155 161
CHAPTER 1
Parts of this introduction have been published in Physiology 2012 Jun; 27(3): 148-55 and Stem Cells and Cancer Stem Cells Volume 2, Springer 2012
1
9 8
Intestinal Homeostasis
The intestinal epithelium consists of a single layer of highly specialized cells (Figure 1) that serves to digest our food and absorb water and nutrients. At the same time this thin layer is the major barrier that separates us from the large variety of microorganisms present in the lumen of the gut. The epithelium of the intestine is entirely renewed every 3-5 days from a pool of continuously replicating stem cells which reside in the bottom of small mucosal invaginations called crypts.1
Stem cells can either self renew or generate daughter cells that migrate upwards to become transit amplifying cells which can rapidly proliferate but have lost the capacity to self-renew. When they reach the top of the crypt, transit amplifying cells withdraw from the cell cycle and differentiate into one of the four cell types of the small intestine (enterocyte, goblet cell, enteroendocrine cell or Paneth cell). These cells migrate further to their positions on the villus or in the crypt until they are shed off or undergo apoptosis (Figure 1).2 The morphology of the colon is similar to that of the
small intestine, but it lacks villi and Paneth cells.
PHYSIOLOGY
Issue: June 2012 Stem cells Enterocytes
Paneth cells
Transit amplifying cells Neuroendocrine cells Goblet cells
crypt Figure 1 | Crypt-villus compartment
In the adult intestine stem cells reside in the bottom of the crypt and give rise to transit amplifying cells. These cells either move downwards to form Paneth cells (only in the small intestine), or upwards to form the other intestinal cell lineages. This figure was published in Physiology Jun 2012,27(3).
Under homeostatic conditions there is a remarkable stability in the turnover time of the differentiated epithelial cells. The balance between the two compartments of differentiated and proliferating cells is a dynamic equilibrium. In conditions of inflammation or epithelial damage
the compartment of proliferating cells can rapidly expand and the rate of epithelial proliferation is substantially increased. For a large part this proliferative response serves to quickly replace damaged epithelial cells. Dynamic equilibria between different compartments exist by virtue of the presence of negative feedback loops. Such signaling circuits must exist in rapidly regenerating homeostatic tissues such as the epithelia of the gastrointestinal tract and the skin. This would mean that differentiated cells secrete signals that negatively regulate the size of the precursor cell compartment, the rate of proliferation or both. If differentiated cells are lost by damage or infection, the signal is diminished and proliferation is enhanced to regenerate the epithelium. As the differentiated cells increase again in numbers so does the negative feedback signal which will restore proliferation in the precursor cell compartment to its normal rate. This line of reasoning predicts that wound healing responses are controlled by the size of the differentiated cell compartment and the negative feedback signals that are normally secreted from this compartment. In this chapter I will elaborate on the evidence that this negative feedback loop depends on Indian Hedgehog (Ihh) secreted by differentiated enterocytes which act to maintain the expression of bone morphogenetic proteins and activins in the mesenchyme which subsequently negatively regulate stem cell fate in the epithelial layer. Furthermore, I will discuss the research on Hedgehog signaling and its role in tumorigenesis.
Morphogens
Multicellular organisms consist of a large variety of specialized cells that form organs with complex architecture and tasks. In order to shape or maintain an organ, cells must be tightly instructed to determine their number, position and function. To maintain such complexity, cell fate needs to be regulated at the population level by extrinsic signals. How this cell fate is determined and by what signals, remains one of the greatest questions in biology. Research has shown that the fate of a cell in an organ is dependent on its position along different spatial axes both during development and in rapidly regenerating tissues in the adult, such as the hematopoietic system, skin and the intestinal epithelium.3
Morphogens are soluble proteins that form a concentration gradient through a tissue and are key regulators of such position dependent cell fate regulation. A morphogen receiving cell has one or more concentration thresholds that are coupled to the expression of a distinct set of target genes which determine different outcomes in cellular fate. Therefore, dependent on the position of a cell in such concentration gradients, it can acquire different characteristics in response to the same signaling molecule. Four families of morphogenetic pathways can roughly be distinguished, the Wnt, Hedgehog, Tgf-b families and a large group of receptor tyrosine kinases such as fibroblast growth factor, platelet-derived growth factor, epidermal growth factor, which share similar intracellular signaling pathways. Several of these morphogenetic pathways have now been identified as key regulators of cell fate in the intestinal epithelium. In the intestine, fate and proliferation of stem cells is determined by the Wnt signaling pathway and in this chapter I will show that Hedgehog acts as a negative regulator of the Wnt signaling pathway.
1
Wnt signaling and the intestinal stem
The Wnt pathway is one of the best studied signaling pathways due to its relevance in biology and carcinogenesis. The knowledge on Wnt signaling was boosted when it was discovered that the APC (adenomatous polyposis coli) gene, an important regulator of Wnt signaling, was mutated in patients with familial adenomatous polyposis (FAP).4 Although FAP is a rare disease, mutations in
APC have been found in the majority of sporadic colorectal cancers.5 The pathway turns on cell type
specific gene expression programs due to the stabilization and nuclear localization of β-catenin. Wnt ligand binds to its cell surface receptor of the Frizzled family and prevents degradation of β-catenin by a complex of proteins including Axin 1/2 and APC. β-catenin can translocate to the nucleus and interact with TCF/LEF family transcription factors, resulting in activation of the downstream targets of the Wnt pathway.
In recent years, it has become clear that Wnt signaling is important for the specification of stem cell fate and for stem cell self renewal. Several Wnt targets such as LGR5 and ASCL2 are specifically expressed in the intestinal epithelial stem cells and are considered as stem cell markers.
The Hedgehog pathway
The Hedgehog pathway was first discovered by Nüsslein-Volhard and Wieschaus 6 in a genetic
screen in Drosophila that identified regulators of body segmentation during embryogenesis. Three Hedgehog genes exist in mouse and man, Sonic Hedgehog (Shh), Indian Hedgehog (Ihh) and Desert Hedgehog (Dhh).7 Hedgehog proteins are produced as a 45 kD precursor protein which is cleaved
and lipid modified to generate an active 19 kD N-terminal fragment that is responsible for all signaling.8-10 Dispatched (Disp) is required forrelease of Hedgehog from the cell membrane.11-13
Once released, Hedgehogs bind to Patched (Ptch)14-16, a 12-transmembrane receptor which does
not convey the Hedgehog signal to the intracellular components of the pathway itself like a conventional receptor. Rather, binding of Hedgehog to Ptch alleviates the inhibitory effect of Ptch on another membrane receptor, the 7-transmembrane protein Smo (Figure 2).17-19 Cdo, Boc and
Gas1 are co-receptors that interact with Ptch, facilitate binding and positively regulate signaling.20-22
All aspects of Hedgehog signaling are mediated via the Glioblastoma (Gli) transcription factors Gli1, Gli2, and Gli3.23 Processing of Gli2 and Gli3 from repressor to activator is believed to occur
in the primary cilium although the mechanism has not been entirely resolved. Gli2 acts mainly as a transcriptional activator whereas Gli3 can also be processed into a transcriptional repressor.24, 25
In the context of gut development however both Gli2 and Gli3 seem to act mainly as an activator as Gli2 and Gli3 are the main mediators of the Hedgehog signal.26-28 Several transcriptional targets
of the Hedgehog pathway are more or less universal and independent of the tissue or target cell type. The expression of Ptch1, Ptch2, Gli1 and Hhip can be used as read out of pathway activity.
PHYSIOLOGY
Issue: June 2012
Y-00003-12/Dr. van den Brink Figure 02 ScEYEnce Studios—4/15/12 Smo Ptch Gli Gli targets OFF Smo Ptch Gli targets ON Hh Gli
Figure 2 | Hedgehog pathway
A model of interaction between receptors Smo and Ptch. (A) In the absence of Hedgehog Ptch functions to inhibit the signaling receptor Smo. (B) Binding of Hedgehog to Ptch alleviates the suppression of Smo and results in activation of the pathway. This figure was published in Physiology Jun 2012,27(3).
The important function of Hedgehog signaling in the development of the small intestine and colon has been shown in studies with Shhand Ihh deficient mice which both die before or shortly after birth.Shh-/- mice display major abnormalities in foregut development but also show multiple
anomalies in the intestine such as intestinal malrotation, duodenal stenosis, reduction of smooth muscle development and abnormal innervation.29 In Ihh-/- mice the intestinal epithelium fails to
organize in a monolayer and there is a lack of good crypt organization.30 In addition the Ihh mutant
mouse show lack of neurons and dilated colon, both features of Hirschsprung’s disease.31 These
results show that Hedgehog signals are essential for organogenesis of the gastrointestinal tract.
Hedgehog signaling is from epithelium to mesenchyme in the adult intestine
Ihh is the main Hedgehog expressed in the small intestine32, 33 and colon.30, 34 Low levels of Shh
may be expressed at the base of the small intestinal and colonic crypts.34-36 A recent paper
showed that mice with a deletion of Shh from the intestinal epithelium by using the Villin-Cre promoter, displayed a shortened ileum, decreased numbers of goblet and Paneth cells. However, the differences found between wild type and Shh mutant mice are small and only in the ileum. Therefore, it is questionable if Shh plays a major role in the homeostasis of the adult intestine.37
Careful analysis of Hedgehog targets Ptch1 and Gli1 by in situ hybridization32-34, 38 and using various
reporter mice38 has shown that Hedgehog signaling is exclusively paracrine in the adult intestine,
1
13 12
muscle precursor cells, differentiated smooth muscle cells, myofibroblast-like cells and pericytes. Functional evidence for exclusive paracrine signaling was provided in Villin-Cre;Smoflox/flox mice
in which the Hedgehog signaling receptor Smo was specifically deleted in intestinal epithelial cells. These mice exhibited a normal intestinal and colonic architecture, normal cell lineages and unaffected Wnt signaling.39 In addition, it was found that there are also LacZ positive Cd11b and
Cd11c expressing cells suggesting that intestinal macrophages and dendritic cells may directly respond to Hedgehog signaling.40, 41 Thus, mesenchymal cells such as smooth muscle cells and
myofibroblast-like cells are the main Hedgehog responsive cells but the innate immune system is a potential alternative target of the pathway.
Hedgehog signaling regulates homeostasis of intestinal mesenchymal cells
Hedgehog signaling regulates the growth and specification of the gut tube mesenchyme during development.42 Several experiments have shown that Hedgehog signaling is also a key factor in
the homeostasis of mesenchymal cells in the adult intestine. Accumulation of mesenchymal cells was observed when Hedgehog signaling was increased using two different strategies. Our group used Rosa26CreERT2-Ptch1fl/fl mice in which the Hedgehog pathway can be conditionally activated
in a body-wide fashion upon injection of tamoxifen.34 These mice displayed an accumulation of
a-Sma positive cells in the mesenchyme. For another approach Zacharias et al.used Villin-Ihh transgenic mice, in which Ihh was overexpressed under the control of a 12.4 kB intestine specific Villin promoter. These mice showed an accumulation of smooth muscle precursors (desmin+, a-Sma-), differentiated smooth muscle cells (desmin+, a-Sma+) and myofibroblast like cells (desmin-, a-Sma+).43
Experiments in which Hedgehog signaling was conditionally lost in the adult intestine have shown that Hedgehog signaling is not only sufficient to expand mesenchymal cells but also required to maintain the existing smooth muscle cells and myofibroblast-like cells. One model combined two different 12.4 kB Villin promoter transgenes, the Villin-Cre and a Villin-LacZfl/fl-HhipDTM.41, 43 In this
latter mouse model Cre-mediated removal of the LacZ cassette results in a frame shift leading to the translation of the soluble Hedgehog inhibitor Hhip. Although the Villin-Cre is active from E12.5 the combination of the transgenes reportedly leads to delayed induction of Hhip expression and mice are born without apparent histological abnormalities. In the double transgenic HhipDTM mouse a progressive loss of smooth muscle cells and an accumulation of smooth muscle precursor cells was found.43 When the same authors cultured E18.5 mesenchyme in the absence of Hedgehog (which
is derived from the epithelium and therefore absent from cultured mesenchyme) they observed an accumulation of smooth muscle precursor cells. These precursor cells differentiated into smooth muscle cells and myofibroblast-like cells after treatment with recombinant Hedgehog.43
A similar observation was made in Cyp1a1Cre-Ihhfl/fl mice that we generated as described in
chapter 2. In these mice Ihh was conditionally deleted from the intestinal epithelium in adult mice. In a time series after deletion we found that two weeks after loss of Ihh both desmin and a-Sma positive cells changed their elongated appearance to adopt a spherical shape. We subsequently observed a sequential loss of first a-Sma (at 2 weeks) and then desmin expression (between 2 and 4 months) from the villus core.
In conclusion, Hedgehog signaling is both sufficient for the expansion of intestinal mesenchymal cells and necessary to maintain their homeostasis. The analysis of Ihh mutant mice suggests that prolonged loss of Ihh signaling ultimately also results in loss of smooth muscle precursor cells and complete loss of the villus core support structure.
Hedgehog signaling regulates the size of the crypt compartment
Ihh is produced by differentiated cells in the adult colon30, 34 and small intestine.33 Several lines of
evidence indicate that Ihh acts as a negative feedback signal to the proliferating cells in the crypt (Figure 3). The first experiments with blocking Hedgehog signaling were performed using anti-Hedgehog monoclonal antibody 5E144 or by overexpression of Hhip under control of the 12.4 kB
Villin promoter in developing mice.45 In both experiments this resulted in a substantial expansion
of the proliferating cell compartment and aberrant localization of crypt-like foci of proliferating cells on the small intestinal villi at birth. This indicated that Hedgehog signaling might negatively regulate epithelial proliferation and crypt size in adult mice. The question remained however, which Hedgehog was being blocked, since both 5E1 and Hhip block all Hedgehogs. We found that in Ihh-/- mice which survived until E16.5 the epithelium of the colon failed to organize into crypts
with superficial non proliferating cells but remained a multilayered proliferating epithelium.30
An analysis of later time points in development could be performed in Villin-Cre-Ihhfl/fl mice in
which Ihh was specifically deleted from the intestinal epithelium during intestinal development.39
This resulted in the same phenotype as the 5E1 anti-Hedgehog antibody treated and Villin-Hhip mice, showing that Ihh is mainly responsible for the negative regulation of the size of the crypt compartment both before birth and during postnatal development.
Hh
Wnt
Bmp/Activin
Figure 3 | Link between Hh and Wnt signaling
Proposed model of negative feedback signaling in the intestinal epithelial crypt. Ihh secretedn by differentiated epithelial cells acts on mesenchymal cells, which secrete factors that subsequently negatively regulate stem cells at the base of the crypt, possibly through Bmp and Activin signaling. This figure was published in Physiology Jun 2012,27(3).
1
To examine the role of Hedgehog signaling in crypt size compartment regulation in adult animals adult rats were treated with the Smo antagonist cyclopamine. In these rats an expansion of BrdU positive proliferating cells was found and the rats showed disturbed enterocyte maturation in the colon, with loss of enterocyte differentiation markers such as brush-border staining for villin and carbonic anhydrase IV. Also, target genes of Wnt signaling were up regulated in the treated rats.30
A conditional genetic approach was used to confirm this negative regulation of the precursor cell compartment. We used both the Rosa26CreERT2-Ptch1fl/fl mice in which Hedgehog signaling can
be activated and Cyp1a1Cre-Ihhfl/fl mice in which Ihh signaling can be lost conditionally in adult
mice. Analysis of these mice confirmed the role of Ihh signaling in the negative regulation of the size of the crypt compartment in adult mice under homeostatic conditions. Interestingly, when studying different time points after recombination in Ihh mutant mice we observed progressive lengthening of crypts. On the other hand we found that crypt fissioning, a mechanism of crypt multiplication through budding and elongation, was only transiently induced at two weeks after recombination. The crypt fissioning stopped when the crypts had reached an increased density at one month after recombination.33 This indicates that in addition to restraining crypt proliferation
and length, Ihh controls a signal that maintains crypt spacing and prevents crypt fissioning under homeostatic conditions. A second, Ihh independent signal must be present to prevent further crypt fissioning when crypt spacing is progressively reduced.
One of the situations in which expansion of the intestinal precursor cell compartment is induced is in case of a massive small bowel resection. These resections induce both crypt lengthening and fissioning in the remainder of the intestine via an unknown mechanism. Interestingly, it was shown that massive small bowel resection results in a remarkable reduction in the expression of Ihh and almost complete loss of Hedgehog signaling as measured by the expression of Hedgehog targets in the remaining intestine.46 Given the fact that it has been demonstrated in Ihh mutant mice that
loss of Ihh induces epithelial changes very similar to those observed after massive small bowel resection, it may well be that loss of Ihh expression is responsible for the adaptive response to intestinal resection.
In conclusion, Ihh is a signal that limits the size of the precursor cell compartment in the crypt. Ihh also prevents crypt fissioning under homeostatic conditions but there are additional unidentified levels of regulation of the rate of crypt fissioning that stop fissioning when crypt spacing is reduced.
Hedgehog induced mesenchymal signals that control epithelial proliferation
Since Hedgehog signaling is exclusively from the epithelium to the mesenchyme, this implies that a mesenchymal derived factor controls epithelial proliferation and stem cells in response to Hedgehog signaling.
Bone Morphogenetic Proteins (Bmp) and activins are part of the Tgfb family. In the normal colon Bmp signaling acts mainly on the differentiated enterocytes in the upper half of the crypt.47
We found that activation of Hedgehog signaling in the Rosa26CreERT2-Ptch1fl/fl mouse resulted
in increased expression of Bmps in the mesenchyme and expanded the range of Bmp signaling from the top towards the base of the crypt.34 Since Bmps negatively regulate intestinal epithelial
proliferation47 and seem to restrict the number of stem cells in the crypt48, 49 this could be one
of the important Hedgehog regulated signals that controls the size and activity of the epithelial precursor cell compartment. However, as Bmp signaling occurs mainly in the differentiated cells under homeostatic conditions, it is difficult to understand how loss of Ihh could result in increased proliferation. We therefore examined the expression pattern of the phosphorylated form of Smads2 and 3 that act downstream of Tgfb/activin signaling. We found that pSmad1,5,8 and pSmad2,3 are expressed in remarkable non-overlapping patterns in the small intestine with pSmad2,3 being expressed specifically in the proliferating cells in the crypts.33 Both pSmad1,5,8
and pSmad2,3 expression was almost completely lost in Ihh mutant animals suggesting that they depend on Ihh regulated mesenchymal factors. We found that loss of pSmad2,3 did not correlate with loss of Tgfb expression but with loss of activin expression.33 These data suggest that Ihh may
control the expression of both Bmps and activins in the intestinal mesenchyme and that both contribute to the negative regulation of intestinal precursor cells. It should be noted however that the cell types expressing the different activin subunits have not been carefully mapped in the intestine and that the role of activin signaling in the intestinal epithelium has not been studied in vivo. Also it is clear that additional Ihh controlled mesenchymal factors are likely to exist that have not yet been identified.
In conclusion, pSmad2,3 and pSmad1,5,8 display non overlapping expression patterns in the intestinal epithelium and are both dependent on Ihh regulated mesenchymal factors. Bmps control the activity of Smad1,5,8 and activins are important candidate intermediates in the regulation of Smad2,3 phosphorylation.
Hedgehog signaling suppresses a lamina propria immune response
Polymorphisms in the GLI1 gene have been associated with the risk to develop inflammatory bowel disease (IBD).40 The rs2228226 C to G variant associated with the development of IBD is a missense
mutation, which encodes a change from glutamine to glutamic acid (Q1100E) in the C-terminus of GLI1 near the transactivation domain and is present in 30% of healthy controls. Odds ratios for the homozygous presence of the Q1100E variant were 1.56, 1.79 and 1.41 for all IBD, Crohn’s disease and ulcerative colitis respectively. It should be noted however that the association with this SNP has not subsequently been replicated in the various genome wide association screens (GWAS) nor has any other component of the Hedgehog pathway been associated with IBD in these screens. It was shown that Q1100E is a hypomorphic variant as it shows lower transcriptional activity than the more prevalent allele. In addition to studying the association of GLI1 with human IBD, the authors used Gli1 mutant mice in an experimental model of colitis.40 Gli1 heterozygous or
even homozygous mutant mice have no evident phenotype. The authors then exposed Gli1+/ lacZ heterozygous animals to dextrane sodium sulfate (DSS), a chemical that activates an innate
immune response by causing damage to the epithelial barrier. They found that DSS treated Gli1 heterozygotes show more severe colitis and an exaggerated cytokine response. Since the animals had a LacZ inserted into the Gli1 knockout allele the cells responding to hedgehog signaling in the uninflamed and DSS treated intestine could be identified by their LacZ positivity. The authors observed LacZ positivity in both Cd11b and Cd11c positive cells suggesting that macrophages and/ or dendritic cells may respond directly to Hedgehog signaling.
1
In a follow up study the Gumucio lab found further evidence for the important role of Ihh in suppressing an inflammatory response of the mesenchyme.41 When primary mesenchymal cells
were cultured for 48 hours in the absence of epithelium, a loss of Hedgehog signaling was observed. When these cultures were then retreated with Hedgehog the major class of downregulated genes were pro-inflammatory cytokines such as Il-1b, Il-6 and a variety of chemokines. These data clearly indicated that epithelium derived Ihh is required to continuously repress a pro-inflammatory response of the underlying mesenchyme.
In conclusion, signaling by epithelium derived Ihh is required to suppress an inflammatory response by the underlying mesenchyme. Reduced signaling by Gli1 may be associated with human IBD, exaggerates the immune response and aggravates disease in the DSS model of intestinal epithelial damage.
Loss of Ihh activates multiple aspects of a wound healing response
As discussed above Ihh is produced by the superficial epithelium, suppresses a lamina propria immune response and controls mesenchymal factors that negatively regulate epithelial proliferation and crypt fissioning. After prolonged disrupted Ihh signaling we and others observed in addition to the epithelial remodeling, loss of smooth muscle cells from the villus core with subsequent loss of villi and development of villus atrophy. Following this loss of villi we found that mice developed an enteritis characterized by a mixed inflammatory infiltrate and influx of fibroblasts.33 Together these data suggest that loss of Ihh is sufficient to activate major aspects
of a wound healing response such as activation of an immune response, epithelial remodeling and recruitment of fibroblasts and macrophages in the absence of any damage to the epithelial layer. Thus the loss of Ihh expression which results from damage or dysfunction of the superficial epithelial layer may be one of the major mechanisms that triggers a wound healing response in the underlying mesenchyme. This indicates that Ihh may act as a critical molecular indicator of epithelial integrity in the intestine. Loss of differentiated epithelial cells will result in a reduction of the concentration of Ihh which subsequently results in recruitment of mesenchymal cells critical to the wound healing response and loss of inhibition of proliferation of stem cells at the base of the crypt.
Hedgehog signaling and carcinogenesis
Colon cancer develops in a well recognized sequence of events termed the adenoma to carcinoma sequence. Adenomas are non malignant neoplastic lesions that carry a risk to progress to cancer. The best studied mutation that can lead to adenoma development in mice and men is the Wnt pathway activating mutations in the APC gene.5, 50 Mutations in APC have been found in the
majority of sporadic colorectal cancers.4
Since the Ihh-Bmp/activin signaling loop is a negative feedback loop that negatively regulates Wnt signaling, this loop may act as a tumor supressor mechanism. Indeed, both the BMP and Activin pathways are frequently inactivated in colorectal cancer 51, 52 and occurrs mainly at the transition
from late adenoma to early carcinoma.53 Thus it is clear that the cancer cells become unresponsive
to the BMPs and Activins that are induced in the mesenchyme by Hedgehog signaling.
Activating mutations in the Hedgehog pathway play an important role in the development of basal cell carcinomas, medulloblastomas and rhabdomyosarcomas.54-56 The role of Hedgehog signaling
in tumorigenesis in other organs is often more complex. In the case of the intestine, it seems that mutations in Hedgehog signaling by itself cannot initiate tumor formation since no gastrointestinal cancers have developed in mouse models with altered Hedgehog signaling. The only exception is the pancreas where Hedgehog overexpression results in the formation of preneoplastic lesions.57
There are several reports on the role of Hedgehog signaling in CRC but data are often conflicting. This is partly due to problems with antibody specificity of most comercially available antibodies for components of the Hedgehog pathway and the fact that many results depend on the use of the widely used Smo inhibitor cyclopamine which has important off target effects when used at higher doses that were used in most studies.58
It has been shown that, in contrast to normal epithelium, mRNA levels of Ptch1, Smo and Gli1 are detectable at least by PCR in colorectal cancer cell lines and xenografts derived from patients with colon carcinoma.59 However, in the case of the xenografts the expression of Hedgehog
pathway components as detected by RT-PCR may well be derived from small contaminations of mesenchymal cells. Although it was found that the Smo antagonist cyclopamine resulted in apoptosis of colon cancer cell lines59 this ocurred at doses that were later found to be 100 fold
higher than the dose required for maximal inhibition of Hedgehog signaling and may be the result of significant off-target effects of cyclopamine used at high doses.58 On the other hand, Varnat et
al. found evidence for expression of Hedgehog pathway components in colorectal cancers.60 Since
Hedgehog signaling normally occurs in the mesenchyme, it may be that Hedgehog responsiveness is related to features of epithelial-to-mesenchymal transition (EMT) in colon cancer. Varnat and colleagues went on to show that proliferation of a variety of colon cancer cell lines may depend on Hedgehog signaling as their growth was impeded by expression of shRNA for Gli transcription factors or Smo and overexpression of a Gli3 repressor form.60 Furthermore overexpression of Gli1
and an shRNA against the inhibitory Hedgehog receptor Ptch1 stimulated growth of colon cancer cell lines. It was not shown if these interventions altered the activity of the Hedgehog pathway in terms of the expression of established targets of the pathway or a Gli-luciferase reporter assay. Also in some of the experiments using CD133 as a single marker to identify colon cancer stem cells it is unsure that there is no contamination of CD133+ cells from the mesenchyme. In a subsequent study Varnat et al. showed that CD133+ fraction of colon cancers lose the expression of Wnt targets such as LGR5, AXIN2 and c-MYC at more advanced TNM stages.61 Instead the
CD133+ cells acquired expression of Hedgehog targets such as GLI1 and HHIP. Again the caveat of this experiment is that CD133 was used as a single marker of colon cancer stem cells which risks contamination of mesenchymal cells. In an intriguing follow-up experiment however the authors show that activation of Hedgehog signaling can overcome the dependence on Wnt signaling in LS147T colon cancer cells. Based on their data the authors hypothesize that there is a switch from Wnt dependence to Hedgehog dependence as colon cancer cells acquire metastatic potential. However, clinical trials for treatment of colorectal cancer with Smo inhibitor vismodegib have been unsuccessfull, which argues strongly against a role for Hedgehog signaling in colorectal carcinomas.62
1
On the other hand the role of Hedgehog signaling in adenomas could be very different from its effect in carcinomas, but this has not been studied extensively. Arimura et al. were first to study the role of Hedgehog in intestinal adenoma formation in vivo by crossing mice that were heterozygous for Hedgehog receptor Smo to Apc mutant mice.63 The authors found that heterozygosity for Smo
protected against adenoma formation suggesting that Hedgehog signaling may promote adenoma formation. However, heterzygosity for Smo did not affect the expression of Hedgehog targets Gli1 or Hhip in the adenomas and the authors detected expression of Smo predominantly in adenoma epithelium. Arimura et al. therefore suggested that the effects could depend on non-canonical Hedgehog signaling through Smo.
In conclusion, it is clear that Hedgehog mediated negative feedback signaling by BMPs and activins is often inactivated in tumors. Data from Varnat et al. suggest that colon cancer cells may aquire Hedgehog responsiveness at higher TNM stages and that growth and metastatic potential depends on Hedgehog signaling. However, clinical studies with Hedgehog antagonists argue against an important role for Hedgehog signaling in colorectal cancer. Regarding the role of Hedgehog signaling and adenoma formation, the only in vivo study showed that Hedgehog signaling may promote adenoma formation.
Concluding remarks and outline of the thesis
Research over the past years has clearly established that the role of Hedgehog signaling is not restricted to intestinal development but that it is also an important factor in the maintenance of homeostasis of the adult small intestine and colon. Ihh is produced by the superficial epithelium, signals to the underlying mesenchyme where it targets smooth muscle cells, myofibroblast-like cells and possibly also myeloid cells. Ihh seems to be a key signal emitted by the superficial epithelium to indicate its integrity. Loss of this signal is in itself sufficient to trigger not only a wound healing response with activation of an epithelial repair program and influx of fibroblast and macrophages, but also to activate a mesenchymal immune response. Furthermore, loss of Ihh leads to expansion of the size of the precursor cell compartment. This was accompanied by increased Wnt signaling as evidenced by nuclear accumulation of β-catenin and increased expression of Wnt targets. These findings suggest that Ihh is an anti-proliferative signal and might behave as an tumor suppressor. On the other hand, it was shown that Hedgehog signaling promotes adenoma formation and that Hedgehog is expressed in colon cancer cells.
Several key questions remain: What factors drive Ihh expression? What is the exact nature of the mesenchymal factors that activate the epithelial repair program once Ihh expression is lost? How does Hedgehog suppress an immune response? Does the Hedgehog pathway control the mesenchymal immune response directly in myeloid cells or is this mediated via Hedgehog responsive smooth muscle cells or myofibroblasts? Secondly, does Hedgehog play a role in adenoma formation? Is its role different in carcinomas? Is Hedgehog signaling paracrine as in the normal intestine or has this changed to autocrine signaling in tumors? To answer the first questions it is necessary to specifically isolate the different Hedgehog responsive target cells individually and generate specific knockout mice that target Smo in the different mesenchymal cell types, such as myeloid cells, smooth muscle cells and myofibroblasts. For the questions about Hedgehog and
tumorigenesis it is paramount to look carefully at Hedgehog expression in tumors using state of the art tools, such as reporter mice and in situ hybridrization. Next, crossing Ihh mutant mice to Apc mutant mice or using chemical carcinogens will elucidate the role of Hedgehog in adenomas. In chapter 2 the data on the role of Hedgehog signaling in intestinal homeostasis as described in this introduction will be shown. By using an inducible Ihh knockout mouse we can show that Ihh signals via the mesenchyme to the proliferating cells in the crypt to attenuate proliferation and negatively regulate Wnt signaling. In chapter 3 we describe that, despite its anti-proliferative role in intestinal homeostasis, Ihh is mandatory for Apc mutant adenoma formation in mice. It seems that tumor epithelial cells need to secrete Ihh in order to maintain an intestinal stromal phenotype that is required to sustain adenoma development. In chapter 4 we used mouse models where we can either activate or inactivate the Hedgehog pathway conditionally. Analysis of gene arrays of colons from these mice show that Hedgehog regulates the expression of the Interferon response target genes. This finding can explain why Ihh mutant mice are more prone to develop a more severe colitis. The next question is how can Hedgehog regulate the interferon response? This question touches upon an issue that has been in the field for years; what cell is the Hedgehog target cell in the intestine? We exclude that the Hedgehog target cell is a myeloid cell and show by characterization of known mesenchymal markers that the Hedgehog target cell is a stromal cell. Furthermore, we show that Hedgehog’s effect on the expression of the Interferon response mediated genes, including chemokines seems to act via the stromal cells. In chapter 5 we studied again the role of Hedgehog signaling in adenoma formation as in chapter 3, but now studied the effect of Hedgehog signaling on colitis-associated carcinogenesis. We demonstrate that moderate alterations of Hedgehog activity does not influence tumor incidence, but complete epithelial loss of Hedgehog leads to an increase in tumor number and size in the colon. This contradictory finding can be explained by the enhanced sensitivity for inflammation in Ihh mutant mice and the different genetic origin of the tumors in the mouse model used. In chapter 6 we studied the role of receptor for advanced glycation endproducts (Rage) which is part of the innate immune system, in tumorigenesis. This receptor binds ligands that are derived from damaged cells such as S100 and Hmgb1 which can initiate and perpetuate an immune response. We show that mice lacking this receptor develop fewer intestinal adenomas, suggesting that blocking this part of the innate immune system may inhibit tumor formation.
1
REFERENCES
1. Potten CS, Loeffler M. Stem cells: attributes, cycles, spirals, pitfalls and uncertainties. Lessons for and from the crypt. Development 1990;110:1001-1020.
2. Potten CS. Radiation, the ideal cytotoxic agent for studying the cell biology of tissues such as the small intestine. Radiat.Res. 2004;161:123-136. 3. van den Brink GR, Offerhaus GJ. The
morphogenetic code and colon cancer development. Cancer Cell 2007;11:109-117. 4. Nishisho I, Nakamura Y, Miyoshi Y, et al.
Mutations of chromosome 5q21 genes in FAP and colorectal cancer patients. Science 1991;253:665-669.
5. Bienz M, Clevers H. Linking colorectal cancer to Wnt signaling. Cell 2000;103:311-320. 6. Nüsslein-Volhard C, Wieschaus E. Mutations
affecting segment number and polarity in Drosophila. Nature 1980;287:795-801. 7. Echelard Y, Epstein DJ, St-Jacques B, et al. Sonic
hedgehog, a member of a family of putative signaling molecules, is implicated in the regulation of CNS polarity. Cell 1993;75:1417-1430. 8. Bumcrot DA, Takada R, McMahon AP.
Proteolytic processing yields two secreted forms of sonic hedgehog. Mol.Cell Biol. 1995;15:2294-2303.
9. Lee JJ, Ekker SC, von Kessler DP, et al. Autoproteolysis in hedgehog protein biogenesis. Science 1994;266:1528-1537. 10. Porter JA, von Kessler DP, Ekker SC, et al. The
product of hedgehog autoproteolytic cleavage active in local and long-range signalling. Nature 1995;374:363-366.
11. Caspary T, Garcia-Garcia MJ, Huangfu D, et al. Mouse Dispatched homolog1 is required for long-range, but not juxtacrine, Hh signaling. Curr.Biol. 2002;12:1628-1632.
12. Kawakami T, Kawcak T, Li YJ, et al. Mouse dispatched mutants fail to distribute hedgehog proteins and are defective in hedgehog signaling. Development 2002;129:5753-5765. 13. Ma Y, Erkner A, Gong R, et al.
Hedgehog-mediated patterning of the mammalian embryo requires transporter-like function of dispatched. Cell 2002;111:63-75.
14. Chen Y, Struhl G. Dual roles for patched in sequestering and transducing Hedgehog. Cell 1996;87:553-563.
15. Marigo V, Davey RA, Zuo Y, et al. Biochemical evidence that patched is the Hedgehog receptor. Nature 1996;384:176-179. 16. Stone DM, Hynes M, Armanini M, et al. The
tumour-suppressor gene patched encodes a candidate receptor for Sonic hedgehog. Nature 1996;384:129-134.
17. Alcedo J, Ayzenzon M, Von OT, et al. The Drosophila smoothened gene encodes a seven-pass membrane protein, a putative receptor for the hedgehog signal. Cell 1996;86:221-232. 18. Taipale J, Cooper MK, Maiti T, et al. Patched
acts catalytically to suppress the activity of Smoothened. Nature 2002;418:892-897. 19. van den Heuvel M, Ingham PW. smoothened
encodes a receptor-like serpentine protein required for hedgehog signalling. Nature 1996;382:547-551.
20. Tenzen T, Allen BL, Cole F, et al. The cell surface membrane proteins Cdo and Boc are components and targets of the Hedgehog signaling pathway and feedback network in mice. Dev.Cell 2006;10:647-656.
21. Yao S, Lum L, Beachy P. The ihog cell-surface proteins bind Hedgehog and mediate pathway activation. Cell 2006;125:343-357.
22. Zhang W, Kang JS, Cole F, et al. Cdo functions at multiple points in the Sonic Hedgehog pathway, and Cdo-deficient mice accurately model human holoprosencephaly. Dev.Cell 2006;10:657-665.
23. Altaba A. Gli proteins encode context-dependent positive and negative functions: implications for development and disease. Development 1999;126:3205-3216.
24. Pan Y, Bai CB, Joyner AL, et al. Sonic hedgehog signaling regulates Gli2 transcriptional activity by suppressing its processing and degradation. Mol.Cell Biol. 2006;26:3365-3377.
25. Wang B, Fallon JF, Beachy PA. Hedgehog-regulated processing of Gli3 produces an anterior/posterior repressor gradient in the developing vertebrate limb. Cell 2000;100:423-434.
26. Kim JH, Huang Z, Mo R. Gli3 null mice display glandular overgrowth of the developing stomach. Dev.Dyn. 2005;234:984-991. 27. Mo R, Kim JH, Zhang J, et al. Anorectal
malformations caused by defects in sonic hedgehog signaling. Am.J.Pathol. 2001;159:765-774. 28. Motoyama J, Liu J, Mo R, et al. Essential
function of Gli2 and Gli3 in the formation of lung, trachea and oesophagus. Nat.Genet. 1998;20:54-57.
29. Litingtung Y, Lei L, Westphal H, et al. Sonic hedgehog is essential to foregut development. Nat.Genet. 1998;20:58-61.
30. van den Brink GR, Bleuming SA, Hardwick JC, et al. Indian Hedgehog is an antagonist of Wnt signaling in colonic epithelial cell differentiation. Nat.Genet. 2004;36:277-282.
31. Ramalho-Santos M, Melton DA, McMahon AP. Hedgehog signals regulate multiple aspects of gastrointestinal development. Development 2000;127:2763-2772.
32. Batts LE, Polk DB, DuBois RN, et al. Bmp signaling is required for intestinal growth and morphogenesis. Dev.Dyn. 2006;235:1563-1570. 33. van Dop WA, Heijmans J, Büller NV, et al. Loss of
Indian Hedgehog activates multiple aspects of a wound healing response in the mouse intestine. Gastroenterology 2010;139:1665-76, 1676.
34. van Dop WA, Uhmann A, Wijgerde M, et al. Depletion of the colonic epithelial precursor cell compartment upon conditional activation of the hedgehog pathway. Gastroenterology 2009;136:2195-2203.
35. van den Brink GR, Hardwick JC, Nielsen C, et al. Sonic hedgehog expression correlates with fundic gland differentiation in the adult gastrointestinal tract. Gut 2002;51:628-633. 36. Varnat F, Zacchetti G, Altaba A. Hedgehog
pathway activity is required for the lethality and intestinal phenotypes of mice with hyperactive Wnt signaling. Mech.Dev. 2010;127:73-81. 37. Gagne-Sansfacon J, Allaire JM, Jones C, et al.
Loss of Sonic hedgehog leads to alterations in intestinal secretory cell maturation and autophagy. PLoS.One. 2014;9:e98751. 38. Kolterud A, Grosse AS, Zacharias WJ, et al.
Paracrine Hedgehog signaling in stomach and intestine: new roles for hedgehog in gastrointestinal patterning. Gastroenterology 2009;137:618-628.
39. Kosinski C, Stange DE, Xu C, et al. Indian hedgehog regulates intestinal stem cell fate through epithelial-mesenchymal interactions during development. Gastroenterology 2010;139:893-903.
40. Lees CW, Zacharias WJ, Tremelling M, et al. Analysis of germline GLI1 variation implicates hedgehog signalling in the regulation of intestinal inflammatory pathways. PLoS.Med. 2008;5:e239.
41. Zacharias WJ, Li X, Madison BB, et al. Hedgehog is an anti-inflammatory epithelial signal for the intestinal lamina propria. Gastroenterology 2010;138:2368-77, 2377.
42. van den Brink GR. Hedgehog signaling in development and homeostasis of the gastrointestinal tract. Physiol Rev. 2007;87:1343-1375.
1
43. Zacharias WJ, Madison BB, Kretovich KE, etal. Hedgehog signaling controls homeostasis of adult intestinal smooth muscle. Dev.Biol. 2011;355:152-162.
44. Wang LC, Nassir F, Liu ZY, et al. Disruption of hedgehog signaling reveals a novel role in intestinal morphogenesis and intestinal-specific lipid metabolism in mice. Gastroenterology 2002;122:469-482.
45. Madison BB, Braunstein K, Kuizon E, et al. Epithelial hedgehog signals pattern the intestinal crypt-villus axis. Development 2005;132:279-289.
46. Tang Y, Swietlicki EA, Jiang S, et al. Increased apoptosis and accelerated epithelial migration following inhibition of hedgehog signaling in adaptive small bowel postresection. Am.J.Physiol Gastrointest.Liver Physiol 2006;290:G1280-G1288.
47. Hardwick JC, van den Brink GR, Bleuming SA, et al. Bone morphogenetic protein 2 is expressed by, and acts upon, mature epithelial cells in the colon. Gastroenterology 2004;126:111-121. 48. Haramis AP, Begthel H, van den Born M, et al.
De novo crypt formation and juvenile polyposis on BMP inhibition in mouse intestine. Science 2004;303:1684-1686.
49. He XC, Zhang J, Tong WG, et al. BMP signaling inhibits intestinal stem cell self-renewal through suppression of Wnt-beta-catenin signaling. Nat. Genet. 2004;36:1117-1121.
50. Su LK, Kinzler KW, Vogelstein B, et al. Multiple intestinal neoplasia caused by a mutation in the murine homolog of the APC gene. Science 1992;256:668-670.
51. Jung B, Doctolero RT, Tajima A, et al. Loss of activin receptor type 2 protein expression in microsatellite unstable colon cancers. Gastroenterology 2004;126:654-659. 52. Markowitz S, Wang J, Myeroff L, et al.
Inactivation of the type II TGF-beta receptor in colon cancer cells with microsatellite instability. Science 1995;268:1336-1338.
53. Kodach LL, Bleuming SA, Musler AR, et al. The bone morphogenetic protein pathway is active in human colon adenomas and inactivated in colorectal cancer. Cancer 2008;112:300-306. 54. Hahn H, Wicking C, Zaphiropoulous PG, et al.
Mutations of the human homolog of Drosophila patched in the nevoid basal cell carcinoma syndrome. Cell 1996;85:841-851. 55. Mao J, Ligon KL, Rakhlin EY, et al. A novel
somatic mouse model to survey tumorigenic potential applied to the Hedgehog pathway. Cancer Res. 2006;66:10171-10178.
56. Pietsch T, Waha A, Koch A, et al. Medulloblastomas of the desmoplastic variant carry mutations of the human homologue of Drosophila patched. Cancer Res. 1997;57:2085-2088.
57. Thayer SP, di Magliano MP, Heiser PW, et al. Hedgehog is an early and late mediator of pancreatic cancer tumorigenesis. Nature 2003;425:851-856.
58. Yauch RL, Gould SE, Scales SJ, et al. A paracrine requirement for hedgehog signalling in cancer. Nature 2008;455:406-410.
59. Qualtrough D, Buda A, Gaffield W, et al. Hedgehog signalling in colorectal tumour cells: induction of apoptosis with cyclopamine treatment. Int.J.Cancer 2004;110:831-837. 60. Varnat F, Duquet A, Malerba M, et al. Human
colon cancer epithelial cells harbour active HEDGEHOG-GLI signalling that is essential for tumour growth, recurrence, metastasis and stem cell survival and expansion. EMBO Mol. Med. 2009;1:338-351.
61. Varnat F, Siegl-Cachedenier I, Malerba M, et al. Loss of WNT-TCF addiction and enhancement of HH-GLI1 signalling define the metastatic transition of human colon carcinomas. EMBO Mol.Med. 2010;2:440-457.
62. Berlin J, Bendell JC, Hart LL, et al. A randomized phase II trial of vismodegib versus placebo with FOLFOX or FOLFIRI and bevacizumab in patients with previously untreated metastatic colorectal cancer. Clin.Cancer Res. 2013;19:258-267.
63. Arimura S, Matsunaga A, Kitamura T, et al. Reduced level of smoothened suppresses intestinal tumorigenesis by down-regulation of Wnt signaling. Gastroenterology 2009;137:629-638.
CHAPTER 2
Willemijn A. van Dop, Jarom Heijmans, Nikè V.J.A. Büller,
Susanne A. Snoek, Sanne L. Rosekrans, Elisabeth A. Wassenberg,
Marius A. van den Bergh Weerman, Beate Lanske, Alan R. Clarke,
Douglas J Winton, Mark Wijgerde, G. Johan Offerhaus, Daan W. Hommes,
James C. Hardwick, Wouter J. de Jonge, Izak Biemond,
Gijs R. van den Brink
Gastroenterology 2010 Nov; 139(5): 1665-1667
Loss of Indian Hedgehog activates multiple aspects
of a wound healing response in the mouse intestine
2
ABSTRACT
Background and aims: Indian Hedgehog (Ihh) is expressed by the differentiated epithelial cells
of the small intestine and signals to the mesenchyme where it induces unidentified factors that negatively regulate intestinal epithelial precursor cell fate. Recently genetic variants in the Hedgehog pathway have been linked to the development of inflammatory bowel disease.
Methods: We deleted Ihh from the small intestinal epithelium in adult mice using Cyp1a1-CreIhhfl/fl
conditional Ihh mutant mice. Intestines were examined by immunohistochemistry, in situ hybridization and real-time polymerase chain reaction.
Results: Deletion of Ihh from the intestinal epithelium initially resulted in a proliferative response
of the intestinal epithelium with lengthening and fissioning of crypts and increased Wnt signaling. The epithelial proliferative response was associated with loss of Bmp and Activin signaling from the epithelium of the villus and crypts respectively. At the same stage we observed a substantial influx of fibroblasts and macrophages into the villus core with increased mesenchymal Tgf-b signaling and deposition of extracellular matrix proteins. Prolonged loss of Ihh resulted in progressive leukocyte infiltration of the crypt area, blunting and loss of villi and the development of intestinal fibrosis.
Conclusions: Loss of Ihh initiates several events that are characteristic of an intestinal wound
repair response. Prolonged loss resulted in progressive inflammation, mucosal damage and the development of intestinal fibrosis. Ihh is a signal derived from the superficial epithelial cells that may act as a critical indicator of epithelial integrity.
INTRODUCTION
Differentiated cells in rapidly renewing tissues such as epithelia of the skin and the gastrointestinal tract are in a dynamic equilibrium with precursor cells in order to balance the rate of proliferation with cell loss at the epithelial surface. The balance between input and output in homeostatic dynamic equilibria depends on the presence of negative feedback loops. The fate and proliferation of intestinal precursor cells is regulated by Wnt signaling. Ihh is the major Hedgehog expressed in the colon and it is secreted by the mature enterocytes at the top of the crypt. Treatment of rats with Hedgehog inhibitor cyclopamine resulted in increased Wnt signaling and precursor cell proliferation whereas enterocyte differentiation was impaired.1 In Ihh-/- mice we observed a
failure of proliferating cells to differentiate and impaired crypt formation but these mice are not viable and died well before birth.1 Inhibition of Hedgehog signaling in the developing intestine
by transgenic expression of Hedgehog antagonist Hedgehog interacting protein (Hhip) similarly resulted in accumulation of proliferating cells, some in ectopic foci on the small intestinal villi, and increased Wnt signaling.2 Conversely, conditional activation of Hedgehog signaling in the adult
intestine by inducible deletion of Hedgehog binding receptor Ptch1 (which acts by repressing the Hedgehog signaling receptor Smoothened) resulted in inhibition of Wnt signaling and depletion of precursor cells which underwent premature differentiation into the enterocyte lineage.3 Thus
Hedgehog signaling seems to act as a negative feedback signal that contributes to the dynamic equilibrium between epithelial precursor cells and enterocytes in the intestinal epithelium but several important questions remain. First, although a genetic loss of function experiment has been performed in the developing intestine1 a genetic loss of function experiment in the fully developed
adult intestine still needs to be performed. Second, although we were unable to detect Shh in the mouse intestine by in situ hybridization3 it has been shown using a ShhGfp reporter mouse that low
levels of Shh may be expressed by rare cells at the crypt base and thus the relative contribution of Shh and Ihh to Hedgehog signaling in the adult intestine still needs to be addressed. Third, Hedgehog signaling is exclusively from the epithelium to the mesenchyme3, 4 and the mesenchymal
factors that negatively regulate precursor cells in response to Hedgehog signaling still await identification. Hedgehog signaling regulates the expression of Bone Morphogenetic Proteins (Bmps) in the developing and adult intestine1, 3, 5 and increased Hedgehog signaling in the adult
extends the range of Bmp signaling through Smads1,5 and 8 from the top of the crypt towards the base of the crypt.3 Since the Bmp pathway is not normally active at the base of the crypt it is
unlikely that Bmps are the major negative regulators of Wnt signaling or precursor cell fate. Indeed, a transgenic mouse that overexpressed the Bmp antagonist noggin in the intestinal epithelium did not have a phenotype until three weeks after birth6, 7 and these mice develop hamartomas,
polyps that are characterized by abnormal growth of the mesenchyme rather than the epithelium. Possible Hedgehog dependent expression of Transforming Growth Factor-β (Tgfβs) or Activins which signal through Smad2 and 3 has not been examined.
The role of Hedgehog signaling in the intestine may extend beyond negative regulation of epithelial precursor cells as a hypomorphic mutant of downstream transcription factor GLI1 has now been linked to the development of inflammatory bowel disease.8 Here we examine the role of
2
Ihh signaling in the adult small intestine using mice in which Ihh can be conditionally deleted from the epithelium of the small intestine.
MATERIALS AND METHODS
(see supplementary methods for details)Mice
The generation of Cyp1a1-Cre and Ihhfl/fl mice 9, 10 has previously been described. At four
weeks of age, Cyp1a1Cre-Ihhfl/fl mice received intraperitoneal injections with either 80 mg/kg
b-naphthoflavone (Sigma) or vehicle (corn oil) for 5 days in a row. Mice were injected with 100 mg/kg BrdU (Sigma) intraperitoneally 1 hour before sacrifice and examined at 2 weeks, and 1, 2, 4 and 6 months after recombination. Vehicle injected Cyp1a1Cre-Ihhfl/fl and β-naphthoflavone
injected Cyp1a1Cre-Ihhwt/wt mice served as controls. For examination of recombination efficiency
by Cre upon injections with b-naphthoflavone Cyp1a1Cre mice were crossed with Rosa26Stopfl/ flLacZ mice11. The experiments were approved by the Institutional Animal Care and Use Committee
of the University of Leiden and Amsterdam.
Immunohistochemistry, LacZ staining and In Situ Hybridization
Immunohistochemistry, LacZ staining, generation of probes and in situ hybridization were performed using standard protocols. See supplementary data.
RNA isolation, complementary DNA synthesis and quantitative RT-PCR
For isolation of RNA from the duodenuma small piece of proximal tissue was collected. A detailed description of RNA isolation and complementary DNA synthesis can be found in the supplementary methods. Quantitative RT-PCR was performed using Sybr Green (LightCycler 480 SYBR Green I Master, Roche, #04707516001) and pre-optimized primers from Qiagen. Gapdh was used as household gene. Gapdh expression was equally distributed between the wild-types and the Ihh mutant mice.
Statistics
Statistical analysis was performed with Prism 5.0 (GraphPad Software). All values were represented as the mean ± standard error of the mean (SEM). Samples were analyzed using a student’s t-test. For multiple comparisons, a one-way ANOVA was utilized followed by a Tukey’s post-hoc test. Differences were considered statistically significant at P < 0.05.
RESULTS
Loss of Ihh signaling in adult β-naphthoflavone injected Cyp1a1Cre-Ihhfl/fl mice
We examined expression of Ihh mRNA (Figure 1A) and protein (Figure 1B) in the small intestine of the mouse. Both Ihh protein and mRNA were exclusively expressed by the differentiated epithelial cells on the villi. To examine the role of Ihh signaling in the adult small intestinal mucosa we injected adult Cyp1a1-Ihhfl/flmice and Ihhfl/flcontrol mice with β-naphthoflavone. This resulted
in substantially reduced expression of Ihh protein (Figure1B, right panel). Quantitative RT-PCR
showed sustained loss of Ihh expression at different time points after recombination (Figure 1C, P < 0.001 for all groups, n = 4/group). Loss of Ihh expression correlated with almost complete loss of Hedgehog targets Gli1 (86% reduced, P < 0.0001) and Hhip (94% reduced, P < 0.0001) but more modest reduction in Ptch1 (36% reduced, P = 0.03) and Ptch2 (65% reduced, P = 0.03) expression two weeks after recombination (n = 8 controls versus 6 Ihh mutants). Thus expression of Ptch1 and Ptch2 may only partially depend on Hedgehog signaling. X-gal staining of the small intestines of Cyp1A1Cre-Rosa26Stopfl/flLacZ mice showed efficient recombination along the proximo-distal axis
of the small intestine (Figure 1E) as previously described.9
Control B E Ihh mutant Control A C dd il mutant control dd il Ihh -1 0 *** Control 2 weeks 1 month 2 months 4 months re la tiv e ex pr es si on Gli1 Hhip Ptch1 Ptch2 -1 0 * *** *** * re la tiv e ex pr es si on Ihh Ihh D E Gli1 Hhip Ptch1 Ptch2
Figure 1 | Loss of Ihh signaling from the small intestine in β-naphthoflavone injected Cyp1a1-Cre-Ihhfl/fl adult
mice. (A) In situ hybridization showed that Ihh mRNA was exclusively expressed by the epithelial cells on the villi. Expression was highest at the crypt villus junction (arrows in A) and diminished towards the villus tip. (B) Immunohistochemistry for Ihh showed expression of Ihh protein by the enterocytes on the villi in mice injected with solvent whereas β-naphthoflavone injected mice have lost Ihh expression at 2 weeks after treatment (B, right panel). Quantitative RT-PCR for Ihh at different time points (C) and Hh signaling targets Gli1, Hhip,Ptch1 andPtch2 two weeks after recombination (D, black bars) on intestinal homogenates of β-naphthoflavone
injected Ihhfl/flcontrol and Cyp1a1-Cre-Ihhfl/fl mice confirmed loss of Ihh expression. Original magnifications:
100x. (E) X-gal staining of the duodenum (dd) and ileum (il) of Cyp1A1-stopfl/flLacZ mice injected with vehicle
2
crypt length contr ol 2 wee ks 1 mon th 2 mon ths 4 mon ths 0 25 50 75 100 125 150 175 *** le nt gh ti nmm crypts fissioning contr ol 2 wee ks 1 mon th 2 mon ths 4 mon ths 0 5 10 15 ** % Ihh mutant Ihh mutantControl Ihh mutant
Control A 2 weeks 1 month BrdU C E
BrdU labeling index
control 2 weeks 1 month 0 10 20 30 40 50 60 % BrdU
control 2 weeks 1 month 0 2 4 6 8 10 12 14 16 *** ** po si tiv e ce lls pe r cr yp t B D F crypt density contr ol 2 wee ks 1 mon th 2 mon ths 4 mon ths 0 1 2 3 *** cr yp ts pe r 10 0 mm
Control 2 weeks 1 month Control 2 weeks 1 month
Figure 2 | Loss of Ihh is sufficient to initiate a regenerative response. (A,B) A strong increase in the rate of
crypt fissioning was observed (arrows in A show fissioning crypts), which was maximal two weeks after recombination (B). (C) H&E staining of the duodenum demonstrated increased crypt density, deepening of crypts and lengthening of the villi in the Ihh mutant mice one month after injection with β-naphthoflavone. (D) Measurements of crypt density and crypt length confirmed an increase in the Ihh mutant mice one month and four months after recombination respectively. (E) Immunohistochemistry for BrdU at one month after recombination. (F) BrdU positive cells were counted and set out both as absolute number of positive cells per crypt and as labeling index, an indication for the percentage of cells per crypt that were positive for BrdU. Original magnifications: 100x (A left panel and C) and 200x (A right panel and E).
Loss of Ihh is sufficient to initiate an epithelial wound healing response
Intestinal epithelial wound healing is characterized by increased epithelial proliferation and lengthening of crypts which multiply by a process of budding and elongation termed crypt fissioning in order to replace lost crypts. We observed a strong increase in the rate of crypt fissioning in the Ihh mutant mice two weeks after recombination (Figure 2A,B, 3.3% in control mice versus 13.7% in mutant mice P<0.001, n=5 per group). Two weeks later, when the crypts reached substantially increased density (1.7 and 3.0 crypts per 100 mm in the controls (n=8) versus Ihh mutant mice (n=7), P<0.001, Figure 2C), the rate of crypt fissioning returned to normal (3.3%, n=8, Figure 2B). Crypt fissioning returned again at 2 and 4 months in the context of chronic inflammation (see below). Crypts became progressively longer (from 66 mm in the controls to 154 mm in the Ihh
mutant mice at four months (n > 4/time point) P<0.001 for control vs 4 months) and a transient modest increase in villus length was observed (Supplementary Figure 1). Changes were measured in the duodenum but were similar in the rest of the small intestine (data not shown). Ihh mutant mice showed an increase in BrdU positive cells (Figure 2E,F) with 6.9 BrdU positive cells per crypt in controls (n=5), 11.7 at two weeks (n=5, P<0.01) and 14.2 (n=7, P<0.001) at one month. Although the total number of BrdU positive cells per crypt increased, the relative number of positive cells per crypt cell (labeling index) remained stable. As we previously found an inhibitory role of Hedgehog signaling on Wnt signaling1, 3 we investigated the effect of loss of Ihh on Wnt signaling 1 months
after recombination.
EphB2
Lgr5 b-catenin A CControl Ihh mutant
Control Ihh mutant
B
b-catenin
control Ihh mutant 0 1 2 ** p o si ti ve n u cl ei p er cr yp t
Control Ihh mutant
Olfm4
D
Control Ihh mutant
E
Alk. Phosph.
F
Control 1 month 2 months 4 months
EM
