The Role of Keratin 15 in Small Intestinal Homeostasis

Aus der Klinik für Innere Medizin A (Direktor Univ. Prof. Dr. Markus M. Lerch) der Universitätsmedizin der Ernst-Moritz-Arndt-Universität Greifswald

The role of keratin 15 in small intestinal homeostasis

Inaugural - Dissertation zur Erlangung des akademischen Grades eines Doktors der Medizin der Universitätsmedizin der Ernst-Moritz-Arndt-Universität Greifswald 2018

Vorgelegt von: Julien Orlando Stephan Geboren am 01.06.1991 in Bethesda (USA)

Dekan: Prof. Dr. med. Karlhans Endlich

1. Gutachter: Prof. Dr. Barthlen 2. Gutachter: Prof. Dr. Lerch 3: Gutachter: Prof. Dr. Singer

Ort, Raum: Greifswald, Seminarraum Innere Medizin A Tag der Disputation: 12.12.2019

Diese Arbeit widme ich meinen Eltern, für deren unermüdliche Unterstützung und Liebe.

ZUSAMMENFASSUNG

Aus der Klinik und Poliklinik für Innere Medizin A - Direktor: Univ.Prof. Dr. M. M. Lerch - der Medizinischen Fakultät der Universität Greifswald

ZUSAMMENFASSUNG The role of keratin 15 in small intestinal homeostasis

- Julien Orlando Stephan-

Die vorliegende Arbeit untersucht die Relevanz des Intermediärfilamentes Zytokeratin 15 (K15) als potentiellen epithelialen Stammzellmarker im Darm und den Einfluss K15 positiver Zellreihen auf die Krypta Homöostase. Zwei Hauptstammzellpools regulieren den schnellen Zellumsatz im Darmepithel. Dies sind einerseits schnell proliferierende Lgr5 positive Stammzellen, welche zwischen den Paneth-Zellen an der Krypta-Basis vorgefunden werden. Anderseits gibt es vermutlich langsamer wachsende Bmi1 positive Zellen, welche sich an der +4-Position oberhalb der Krypta-Basis befinden. Im Haarfollikel und im Ösophagusepithel stellt das Intermediärfilament K15 einen Marker für Stammzellen dar, die zur Gewebereparatur beitragen. In dieser Arbeit haben wir gezeigt, dass K15 im Darm langlebige Kryptazellen mit Multipotenz- und Selbsterneuerungspotenzial markiert. K15 positive Krypta-Zellen sind resistent gegen hochdosierte ionisierende Strahlung und tragen zur Kryptaexpansion bei. Die hier vorgestellten Ergebnisse zeigen nun erstmals im Darm eine langlebige, multipotente K15 exprimierende Kryptazellpopulation, die eine Selbsterneuerungskapazität besitzt. Insbesondere führt der Verlust des Tumorsuppressor Gens Apc in K15 positiven Zellen zur Adenombildung, die potentiell zum Adenokarzinomen fortschreiten können. Wir erörtern die Hypothese, dass K15 eine Gruppe langlebiger, strahlenresistenter Stammzellen markiert, die die Homöostase der Krypta und die Regenerationsfähigkeit maßgeblich beeinflussen.

TABLE OF CONTENTS

ZUSAMMENFASSUNG ...... 4

TABLE OF CONTENTS ...... 5

LIST OF FIGURES ...... 8

LIST OF ABBREVIATIONS ...... 10

1. ABSTRACT ...... 12

2. REVIEW OF THE LITERATURE ...... 13 2.1. OVERVIEW OF INTERMEDIATE FILAMENTS ...... 13 2.1.1. Main intermediate filament functions ...... 13 2.2. KERATINS ...... 14 2.2.1. The role of SEKs in the small intestine ...... 14 2.3. OVERVIEW OF THE SMALL INTESTINAL EPITHELIUM ...... 15 2.3.1. The stem cell niche and their components ...... 16 2.3.2. The role of stem cells in cancer initiation ...... 18

3. AIMS OF THE STUDY ...... 20

4. MATERIAL AND METHODS ...... 21 4.1. CHEMICALS AND REAGENTS ...... 21 4.1.1. Immunohistochemistry and Immunofluorescence ...... 21 4.1.2. Primary Antibodies ...... 22 4.1.3. Secondary Antibodies ...... 22 4.1.4. PCR primers ...... 22 4.1.5. Materials used in cell culture ...... 23 4.1.6. Instruments and software ...... 23 4.1.7. Consumable supplies ...... 25 4.1.8. Animal care ...... 25 4.2. ANIMAL EXPERIMENTAL METHODOLOGY ...... 26 4.2.1. Origin of transgenic mice used in the experiments ...... 26 4.2.2. Animal care ...... 26 4.2.3. Genotyping procedure ...... 27 4.2.4. Euthanasia and organ harvest ...... 27 4.2.5. Cre / LoxP recombination ...... 27

TABLE OF CONTENTS

4.2.6. Experimental strategy for the Krt15-crePR1; R26mT/mG mouse...... 27 4.2.8. Experimental strategy for Krt15-crePR1; R26Tom mice...... 28 4.2.9. Experimental strategy for irradiation of Krt15-crePR1; R26mT/mG and Krt15-/- mice...... 29 4.2.10. Experimental strategy for Krt15-CrePR1;Apcfl/fl;R26mT/mG mice ...... 29 4.3. TISSUE FIXATION ...... 30 4.4. IMMUNOHISTOCHEMISTRY (IHC) ...... 30 4.4.1. IHC in combination with the M.O.M.-kit ...... 31 4.5. IMMUNOFLUORESCENCE (IF) ...... 31 4.5.1. Immunofluorescence using the TSA system ...... 31 4.6. ALCIAN BLUE STAINING (AB) ...... 32 4.7. SINGLE CELL ISOLATION ...... 32 4.8. CRYPT ISOLATION AND 3D ORGANOID CULTURE ...... 32 4.9. IRRADIATION ...... 33 4.10. HISTOMETRIC ANALYSIS ...... 33 4.10.1. Crypt length analysis ...... 33 4.10.2 Crypt microcolony assay ...... 34 4.10.3. Software and statistical analysis ...... 34

5. RESULTS ...... 35 5.1. ENDOGENOUS EXPRESSION OF KRT15 IN THE SMALL INTESTINE...... 35 5.1.1. Endogenous expression of Krt15 in WT mice ...... 35 5.1.2. Localization of Krt15+ cells in Krt15-crePR1; R26mT/mG mice...... 37 5.2. KRT15 EXPRESSION IN VARIOUS EPITHELIAL CELL TYPES PRESENT IN THE INTESTINE...... 40 5.2.1. Krt15 expression in enteroendocrine cells...... 41 5.2.2. Krt15 expression in the goblet cell population...... 42 5.2.3. Krt15 expression in the Paneth cell population...... 42 5.2.4. Cell-cycling activity among Krt15+ cells...... 44 5.3. EVALUATION OF PROLIFERATIVE POTENTIAL WITHIN THE KRT15+ CELL POPULATION...... 45 5.3.1. Lineage tracing experiment on Krt15-crePR1; R26mT/mG mice...... 45 5.3.2. Colocalization of Krt15+ cells with proliferating cells in a lineage tracing setting...... 49 5.3.3. Colocalization of Krt15+ cells with enteroendocrine cells in a lineage tracing setting...... 50 5.3.4. Colocalization of Krt15+ cells with goblet cells in a lineage tracing setting...... 51 5.3.5. Colocalization of Krt15+ cells with Paneth cells in a lineage tracing setting...... 52 5.3.6. Krt15+ cells depict multipotency in 3D organoids ...... 52

TABLE OF CONTENTS

5.4. THE ROLE OF KRT15+ CELLS IN RESPONSE TO RADIATION INJURY AND THEIR REGENERATION CAPABILITIES...... 56 5.4.1. The role of Krt15+ cells in tissue recovery...... 56 5.4.2. The importance of Krt15 expressing cells in irradiation models compared to Krt15-/- ...... 60 5.4.3. Microcolony assay on Krt15 expressing cells...... 64 5.5. ROLE OF KRT15 IN CANCER INITIATION...... 66

6. DISCUSSION ...... 69

7. CONCLUSION ...... 75

8. ACKNOWLEDGEMENT ...... 76 8.1. AUTHOR CONTRIBUTION ...... 76

9. BIBLIOGRAPHY ...... 77

10. ORIGINAL PAPER ...... 85

EIDESSTATTLICHE ERKLÄRUNG ...... 98

CURRICULUM VITAE ...... 99

DANKSAGUNG ...... 101

LIST OF FIGURES

Figure 1: The regulatory niche of intestinal stem cells ...... 17 Figure 2: Breeding strategy for the Krt15-crePR1; R26mGFP mouse...... 28 Figure 3: Experimental strategy for Krt15-crePR1; R26mT/mG mouse irradiation...... 29 Figure 4:Experimental strategy for 3D organoid structures ...... 33 Figure 5: Control of Krt15 knockout in Krt15-/- mice versus WT mice...... 35 Figure 6: Krt15 is expressed in intestinal epithelial cells...... 36 Figure 7: Krt15+ cells can consistently be identified in the intestinal epithelium...... 37 Figure 8: Krt15-crePR1; R26mT/mG mice reliably label Krt15+ cell in the small intestine. ... 38 Figure 9: Krt15+ cells are predominantly located in the stem cell compartment...... 39 Figure 10: Krt15+ cells are predominantly located in the stem cell compartment...... 40 Figure 11: Colocalization of Krt15+ cells with enteroendocrine cells...... 41 Figure 12: Krt15+ cells colocalize with goblet cells in isolated events...... 42 Figure 13: Krt15+ cells are located above crypt base, showing no colocalization with secretory paneth cells...... 43 Figure 14: Krt15+ cells are proliferating cells located at a +4 position...... 44 Figure 15: Timetable depicting experimental setup of the lineage tracing experiment over 150 days...... 45 Figure 16: Evolution of Krt15+ progeny cells within the crypt...... 46 Figure 17: Lineage tracing over 150 days depicts Krt15+ as proliferative, long-lived epithelial cells...... 47 Figure 18: Krt15+ progeny cells colonize the entire crypt as well as adjacent crypt and villi after 150 days...... 48 Figure 19: Krt15 marks a long-lived consistently expanding crypt population...... 49 Figure 20: Krt15+ progeny cells maintain the ability to differentiate in to enteroendocrine cells...... 50 Figure 21: Krt15+ progeny cells reliably differentiate into goblet cells over time...... 51 Figure 22: Krt15+ cells can differentiate into paneth cells over time...... 52 Figure 23: live 3D organoids formed by Tomato+(Krt15-derived) ...... 53 Figure 24: Fixated 3D organoids formed by Tomato+(Krt15-derived) sf ...... 53

LIST OF FIGURES

Figure 25: Co-staining of Tomato with Ki-67, Mucin 2, Chromogranin A and Lysozyme in fixated Organoids ...... 54 Figure 26: live 3D organoids formed by isolated single Tomato+(Krt15-derived) cell...... 55 Figure 27: GFP marks radioresistant Krt15 positive cells capable of repopulating intestinal crypts after radiation injury ...... 58 Figure 28: Krt15+ cells vitally expand post radiation injury, labeling the majority of the crypt population...... 59 Figure 29: Proliferative capacity of Krt15+ progeny cells within labeled crypts, post radiation injury...... 59 Figure 30: Depressed ileal crypt rehabilitation in Krt15-/- mice, post irradiation injury. .... 61 Figure 31: Discrepancies in crypt length in the ileum of WT mice and Krt15-/- mice 5 days post radiation injury...... 61 Figure 32: Jejunal crypts unchallenged post irradiation in Krt15-/- mice...... 62 Figure 33: Discrepancies in crypt length between WT mice and Krt15-/- mice 5 days post radiation injury...... 63 Figure 34: Krt15-/- mice, portrayed a significant reduction of proliferating ileal crypt cells...... 64 Figure 35: Significant reduction in absolute number of live cells per crypt in wildtype and Krt15-/- mice...... 65 Figure 36: Number of microcolonies per 50 crypts, in wildtype vs. Krt15-/- mice post irradiation...... 65 Figure 37: Timetable depicting experimental setup for the cancer initiation model ...... 66 Figure 38: Representative image of intestinal tumors in Krt15-CrePR1;Apcfl/fl;R26mT/mG mice ...... 67 Figure 39: Representative histology of flat adenoma and invasive adeno carcinoma in Krt15- CrePR1;Apcfl/fl;R26mT/mG mice...... 67

LIST OF ABBREVIATIONS

-/- Knock-out +/- Heterozygote +/+ Wild type AAALCA Association for Assessment and Accreditation of Laboratory Animal Care AB Alcian Blue ABSL Animal Biosafety Level BMP bone morphogenetic protein bp Base pair BrdU Bromodeoxyuridine BSA Bovine serum albumin C Colon CBC Crypt based columnar cDNA Complementary deoxyribonucleic acid Chr A Chromogranin A CO2 Carbon dioxide CTL Control CY Cyanine dye DAB Diaminobenzene DAPI Diamidino-2-Phenylindole DMEM Dulbecco's Modification of Eagle's Medium DNA Deoxyribonucleic acid DPBS Dulbecco's Phoasphate buffered Saline DT Diphtheria toxin DTR Diphtheria toxin receptor EGF epidermal growth factor EtOH Ethanol GFP Green fluorescent protein H2O Water HRP Horseradish peroxidase iDTR Inducible diphtheria toxin receptor

LIST OF ABBREVIATIONS

IF Immunofluorescence IFL Intermediate filament IgG Immunoglobulin G IHC Immunohistochemistry IP Intra peritoneal ISC Intestinal stem cell JIR Jackson Immune Research K15 Cytokeratin / Keratin 15 (protein) K19 Cytokeratin / Keratin 19 KO Knock out mice Krt 15 Keratin 15 promoter (DNA) Lgr5 Leucine-rich repeat-containing G-protein coupled receptor 5 LYZ Lysozyme milliQ H2O Purified water Muc 2 Mucin 2 PBS Phosphate buffered saline PBT Phosphate buffered saline with 1% Tween 20 PCR Polymerase chain reaction PFA Paraformaldehyde qPCR Quantitative polymerase chain reaction R26 ROSA 26 locus Rb Rabbit RT Room temperature SEK Simple epithelial Keratins SI Small intestine SOX9 Sex-determining region y-box 9 TSA Tyramide signal amplification ULAR University Laboratory Animal Resource WT Wild type mice

1. ABSTRACT

Two principal stem cell pools orchestrate the fast cell turnover in the intestinal epithelium. Rapidly cycling Lgr5+ stem cells are intercalated between the Paneth cells at the crypt base (CBCs) and a putative slower cycling Bmi1+ cells are located at the +4 position above the crypt base. In the hair follicle and the esophageal epithelium, the intermediate filament Keratin 15 (Krt15) marks stem cells contributing to tissue repair. Herein, we demonstrated that Krt15 labels long-lived crypt cells harboring multipotency and self-renewing potential. Krt15+ crypt cells are resistant to high-dose radiation and contribute to crypt expansion following injury. These results suggest that Krt15 annotate a long-lived, multipotent crypt cell population harboring self-renewal capacity. Notably, Apc loss in Krt15+ cells lead to adenoma formation that can occasionally progress to adenocarcinoma.

2. REVIEW OF THE LITERATURE

2.1. Overview of intermediate filaments The eukaryotic cytoskeleton is composed of three major protein families, i.e. microfilaments, microtubules and intermediate filaments (IFL) which can be differentiated from one another by size and function. IFL have an average size of 10 nm and are thereby ranged in size between microfilaments (7nm) and microtubule (25nm). In contrast to microfilaments and microtubules, IFL have a widespread expression profile formed by the transcription of 70 functional genes with alternative splicing and post-transcriptional modifications (Szeverenyi et al., 2008). These modifications bear extremely versatile protein structure with individualized properties. Whereas microfilaments and microtubule are polymers of a single type of protein, IFL are heteropolymers composed of a variety of proteins. More than 50 different proteins have been identified and classified into 6 types of IFL, based on similarities in their primary protein structure (Cooper, 2000). Type I and II are keratins, expressed in epithelial cells and divided into hard epithelial keratins and soft epithelial keratins (SEK). Hard keratins are present in structures such as hairs, nails and horns, whereas SEK are contained in the cytoplasm of epithelial cells. Type III IFLs include vimentin and desmin and are especially expressed in fibroblast und muscle tissue. Type IV IFLs consist mainly of neurofilaments expressed in neural cells. Type V IFLs are nuclear laminins, forming an orthogonal meshwork that is crucial for nuclear stability. Type VI IFLs mainly consist of nestine proteins expressed in neural stem cells.

2.1.1. Main intermediate filament functions During the 1990s, the main role of IFL was believed to be maintaining the structural and mechanical integrity of cells. Only recent studies unraveled IFL as a highly dynamic structure involved in organizing various cellular processes (Eriksson et al., 2009). Diseases associated with isolated IFL mutations are linked to a broad range of symptoms, unraveling the vast impact of these underestimated proteins. IFLs can act as functional determinants signaling pathways and posttranslational protein targeting. The cytoprotective effects of IFLs are related to their mechanical properties facilitating cells to cope with mechanical and non-mechanical stresses.

2. REVIEW OF THE LITERATURE

However, cytoprotection may also arise from the capacity of IFs to interact with signaling pathways involved in determining cell survival and cell fate (Sahlgren et al., 2003).

2.2. Keratins Keratins are type I and II IFLs expressed in epithelial cells of all vertebrates. Like all types of IFL, they play a crucial role in cellular integrity, cell signaling, protein targeting, apoptosis and protection against stresses. Keratins form obligate heteropolymers (one type I and one type II) and share a common structure that consists of a central coiled-coil α-helical rod domain that is flanked by non-α-helical head and tail domains. Different types of keratins can be distinguished, regarding their biochemical constitution or according to keratin-producing cells and tissues. Epithelial cells of in simple epithelia regularly synthesize K5 or K14, whereas K8 and K18 predominate in stratified epithelia. These epithelial cells can also produce other keratins in addition to or instead of the primary keratins and these keratins are referred to as secondary keratins, such as K7/K19 in simple epithelia or K15 and K6/K16 in stratified epithelia (Bragulla and Homberger, 2009). For example, Krt15 is expressed in the basal layers of the stratified epithelium as well as in keratinoblasts of the hair follicle. The distinction between keratin types can help distinguish between various levels of differentiation of epithelial cells. Recent findings link Krt15 expression patterns to undifferentiated stem cells in the bulge of the hair follicle as well as in epidermal cell layers (Bose et al., 2013; Liu et al., 2003). Krt15+ cells in the bulge of the hair follicle are long- lived and possess a greater colony-forming ability, establishing them as putative follicular stem cells (Inoue et al., 2009). Krt15+ cells have also been shown to contribute in epidermal renewing and repair after wounding (Ito et al., 2005; Li et al., 2013). Furthermore, our team identified Krt15 expressing cells to mark a long-lived subpopulation of basal cells in the mouse esophagus capable of generating all states of squamous lineages (Giroux et al., 2017).

2.2.1. The role of SEKs in the small intestine One of the challenges in modern biomedicine is to explore the functional significance of keratins in the intestine as various keratin types appear limited to the type of tissue and cell differentiation. Specific keratins seem to predominate in undifferentiated epithelial cells, presumably providing a selection advantage. A collaborating group investigated the role of Krt19 as potential stem cell marker in the colon epithelium. Their findings indicated that Krt19 marks a line of radiation-

2. REVIEW OF THE LITERATURE resistant stem cell located in the intestinal crypt (Asfaha et al., 2015a). Similar to K15, K19 was reported as biochemical marker for cutaneous stem cells and is believed to be distinctly expressed in undifferentiated progenitor cells (Driskell et al., 2015; Kloepper et al., 2008; Michel et al., 1996). Furthermore, stem cells located in the bulge of the hair follicle were shown to express Krt19 especially in the wounded epidermis. Based on these observations, both K15 and K19 could be used as markers of stem cells to segregate them from the differentiated cell types. Nonetheless, the role of keratins in the intestinal homeostasis remains mainly unchartered.

2.3. Overview of the Small intestinal epithelium The gastrointestinal tract is the largest organ in the human organism and covers more than 32 square meters of epithelial surface (Helander and Fändriks, 2014). Specialized cells form a highly complex and adaptable barrier between the exterior world and the mammalian host. The small intestinal epithelium is formed by a single layer of columnar epithelial cells organized in villi and crypts. These folded structures as well as the microvilli present at the surface of the enterocytes generate a large contact surface enabling the organism to absorb a maximum amount of nutrients and water. The gastrointestinal tract is colonized by commensal bacteria that aid in digestion and influence the development and function of the mucosal immune system. However, microbial colonization carries the risk of infection and inflammation as well as any potential damage by toxins. Epithelial homeostasis is maintained by an interaction of the mucosal immune system and the intestinal epithelial cells. Epithelial cells can be categorized into absorptive (enterocytes) and secretory lineages (goblet cells, enteroendocrine cells and Paneth cells). Most epithelial cells consist of absorptive enterocytes essential for nutrient and water uptake. Mucin-producing goblet cells are mostly located in the villi and are responsible for covering the epithelium with a protective layer of mucus. Enteroendocrine cells compose only a small proportion of the differentiated cells, but play a crucial role by producing active hormones and cytokines acting locally and systemically. At the bottom of each small intestinal crypt can be found Paneth cells. These cells are extraordinarily versatile exocrine cells responsible for producing antimicrobial factors, such as alpha-defensine and locally active mediators. These cells are believed to be essential to uphold the microenvironment necessary for the crypt-based stem cell population. Residing at the base of the crypt are the intestinal stem cells (ISCs) responsible for continuous turnover of epithelial cells every 4 to 7 days. ISCs give rise to transient amplifying cells which will then proliferate and

2. REVIEW OF THE LITERATURE terminally differentiate while undergoing an upward migration. Differentiated cells migrate along the crypt/villi axis to the top of the villi where they undergo anoikis. This rapidly cycling system makes the small intestine the ideal location to study stem cell activity.

2.3.1. The stem cell niche and their components The term stem cell niche exists since 1978 (Schofield, 1978) and describes the unique microenvironment crucial for stem cell survival. This stem cell niche is a complex multicellular construct essential for stem cell maintenance, differentiation and proliferation. The ISC niche consists of different cellular components, namely endothelial cells, Paneth cells, myofibroblasts, neural cells, lymphocytes and adjacent smooth muscle tissue. Each component contributes by producing unique cell-cell ligands, local growth factors or cytokines (Barker, 2014; Sailaja et al., 2016). Wnt-signaling, bone morphogenetic protein (BMP), Notch and epidermal growth factor (EGF) signaling pathways have been reported to regulate stem cell activity and are gradually expressed along the length of the villi and crypts (Carulli et al., 2015; Kuhnert et al., 2004; Pellegrinet et al., 2011; Vries et al., 2010). At this point, two main stem cell states have been categorized. The rapidly proliferating population of crypt base columnar (CBC) cells are intercalated between the Paneth cells. CBC cells express Lgr5, are highly proliferative pluripotent radio-sensitive stem cells (Barker et al., 2007). CBC give rise to transient amplifying cells, which terminally differentiate into secretory and absorptive lineages. Alongside Lgr5 a variety of different CBC cell markers were found consisting of: Ascl2, Olfm4, Sox9-EGFP(lo), Rnf43, Znrf3, Smoc2, Troy, Prom1, and Msi1 (Fafilek et al., 2013; Hao et al., 2012; Muñoz et al., 2012; Roche et al., 2015; Schuijers et al., 2014, 2015; Snippert et al., 2009). Yet Lgr5 has been predominantly studied, due to the robust expression in CBC cells and the availability of reliable tracing systems. The second stem cell state, the reserve stem cells are located at the +4 position, meaning 4 cell positions above the crypt base. A variety of different stem cell markers for the +4 position were found, comprising: p-PTEN, Bmi1, mTert, Hopx and Lrig1 (He et al., 2007; Montgomery et al., 2011; Powell et al., 2012; Sangiorgi and Capecchi, 2008a; Wong et al., 2012). The first functional validation of +4 stem cells was achieved via lineage tracing, using an in vivo mouse model. Contrary to Lgr5+ CBC cells, Bmi1 labeled cells are slow cycling and radioresistant cells. Bmi1+ stem cells seem to be mainly insensitive to Wnt-signaling variations proliferating only weekly under homeostatic conditions (Yan et al., 2012). However, ablation of CBC following radiation

2. REVIEW OF THE LITERATURE activates quiescent +4 stem cells causing them to rapidly proliferate and repopulate the entire crypt population. Experimental studies of Lgr5+ cells ablation were designed using Lgr5-DTR mice, thus specifically deleting Lgr5 expressing cells (Yan et al., 2012). Lgr5-DTR mice express the diphtheria toxin receptor under the control of the Lgr5 promoter, allowing targeted deletion of Lgr5+ cells. After ablation of CBC cells, +4 stem cells rapidly proliferated, replacing vacant crypt spots with Bmi1+ progeny cells. Remarkably, Bmi1+ progeny cells replaced ablated CBC cells and began to express Lgr5, confirming the potential of Bmi1+ stem cells to differentiate into all intestinal cell lineages.

Figure 1: The regulatory niche of intestinal stem cells

Picture derived from an illustration taken out of The Journal of Physiology, Volume 594, Issue 17, pages 4827-4836, 28 JUL 2016 DOI: 10.1113/JP271931. The stem cells, consisting of CBC and +4 stem cells, reside at the bottom of the crypt. Rapidly dividing transit-amplifying cells arise from these stem cells and differentiate into absorptive lineages (enterocytes) or secretory lineages (enteroendocrine cells, goblet cells and Paneth cells). The niche consists of multiple components and cell types, including extracellular matrix, fibroblasts,

2. REVIEW OF THE LITERATURE myofibroblasts, smooth muscle cells, neural cells, endothelial cells, lymphocytes and macrophages along with various secreted factors. Wnt, BMP, Notch, Hedgehog, and EGF signaling pathways are regulating clonogenic activity (Sailaja et al., 2016).

A collaborating group identified Keratin 19 as a marker of a new type of stem cell located from the +4 position and reaching up to the crypt isthmus. This novel cell type is Lgr5-, radio resistant and gives rise to all epithelial cell lineages, including Lgr5+ CBC cells (Asfaha et al., 2015a). It could also be associated with cancer initiation of radioresistant neoplastic tumor cells. These findings were particularly interesting for our purpose, as K19 marks the first intestinal stem cell marker labeled by intermediate filaments. K19 and K15 are putative biomarkers in the epidermal tissue of the hair follicle, however their involvement in the intestinal epithelium has yet to be sufficiently identified (Inoue et al., 2009; Ma et al., 2004).

2.3.2. The role of stem cells in cancer initiation Cancer of the gastrointestinal tract, such as the colorectal cancer, is one of the most common cancer worldwide and carries the second highest mortality rate in developed countries (Jemal et al., 2011). The traditional view of cancer initiation argues that all mutated cells have the equal potential to proliferate and drive tumor growth, based on the principal of stochastic probability. However, the traditional cancer model has recently been challenged by the cancer stem cell model. The cancer stem cell model postulates that tumors are organized hierarchically, with cancer stem cells as a proliferative driving force (Dick, 2009). Adult stem cells possess essential characteristics such as self-renewal capacity and long-term replication ability, bearing similitudes to cancer cells. During normal homeostasis, these capacities are tightly regulated by diverse growth factors and signaling pathways such as Wnt signaling, mutated stem cells are far more likely to transform to cancer stem cells, than any other cell (Espersen et al., 2015). These theories fueled recent studies to identify putative intestinal stem cell markers and their potential role in cancer initiation. Some intestinal stem cell markers are now believed to be linked with colorectal cancer initiation including SOX9, Bmi1 and Lgr5 (Espersen et al., 2015; Seshagiri et al., 2012; Takahashi et al., 2011). In addition, the first proven stem cell marker Lgr5 has been extensively studied and increased expression in adenomas as well as invasive colorectal cancer was confirmed in several studies (Fan et al., 2010, 2010; Takeda et al., 2011). High expression of Lgr5 is usually linked with Wnt signaling. R- spondin, a Wnt-signaling agonist was linked to increased Lgr5 expression in colorectal cancer via positive feedback loops (de Lau et al., 2011; Seshagiri et al., 2012). Next to Lgr5+ CBC cells,

2. REVIEW OF THE LITERATURE

Bmi1+ stem cells are located at the +4 position and are functionally distinct by their radioresistant and quiescent role in the intestinal crypt. The exact implication of Bmi1 in colon cancer initiation remains unclear, yet studies report some overexpression of Bmi1 in human colon cancer (Li et al., 2010; Reinisch et al., 2006; Tateishi et al., 2006). Induction of b-catenin in Bmi1+ ISC has shown to generate rapidly growing adenomas (Sangiorgi and Capecchi, 2008a). Nonetheless, the exact implication of stem cells on homeostasis and potential impact on cancer initiation has yet to be clarified.

3. AIMS OF THE STUDY

The main objective of this study is to unravel the roles of keratin 15 in intestinal homeostasis. Specific aims are as follow:

A. Determine the endogenous expression of K15 in the mouse small intestinal epithelium

B. Identify which small intestinal cell types express K15

C. Evaluate the proliferative potential of the K15 positive cell population

D. Investigate the role of K15 expressing cells in an injury model and the regeneration

capability of these cells

E. Determine whether K15 cells are crucial for small intestinal homeostasis

F. Determine the implication of K15 in cancer initiation

4. MATERIAL AND METHODS

4. MATERIAL AND METHODS 4.1. Chemicals and Reagents

4.1.1. Immunohistochemistry and Immunofluorescence Zinc Formalin Fixative Polyscience Inc. 21516-3.75

Paraffin: TissuePrep embedding compound Fisher Scientific T565 Ethanol 200 Proof 100% Decon Laboratories #2701

Ethanol 150 Proof 95% Decon Laboratories #2801

Xylene VWR 89370-088

10X PBS Growcells MRGF-6235

Hydrogen peroxide solution 30% in H2O Sigma-Aldrich 216763 Citric Acid Monohydrate Fisher Scientific A104-500

Starting Block T20 Fisher Scientific PI-37539

Diamidino-2-Phenylindole (DAPI) Fisher Scientific D1306 Grill III Hematoxylin Leica Surgipath 3801540

M.O.M. Kit Vector BMK-2202

Triton X-100 Fisher Scientific BP151-500

Tween 20 - was part of the washing buffer Amresco M147-1L Mounting media for IF KPL 71-00-16

Cytoseal 60 Richard-Allan Corp 8310-4

3 % Acetic Acid Self-made -

1% Alcian Blue in 3% Acetic acid R&D Corp S111A-8OZ

Nuclear fast RED Kernechtrot 0.1% R&D Corp S248-8OZ Normal Donkey Serum JIR 017-000-121

BSA - Bovine Serum Albumin Fract V Fisher Biotech CAS 9048-46-8 Biotin Sigma-Aldrich B4501

Avidin from egg Sigma-Aldrich A9275

Reagent A (Avidin DH) Vectastain PK-6100

Reagent B (Biotinylated HRP) Vectastain PK-6100

TSA Perkin Elmer NEL700A001K

Biotinyl tyramide reagent Perkin Elmer FP1019

4. MATERIAL AND METHODS

Streptavidin-HRP Perkin Elmer FP1047

4.1.2. Primary Antibodies Lysozyme Rabbit Diagnostic Biosystems RP-028 Mucin 2 Glycoprotein Mouse Vector Labs VP-M656 Mucin 2 Glycoprotein Rabbit Abcam ab133555 SP-1 Chromogranin A Rabbit ImmunoStar 20085 Ki67 Rabbit Abcam ab16667 BrdU (Bu 20a) mouse IgG1 cell signaling 5292S GFP Goat pAb Abcam ab6673 K15 Mouse Vector Labs VP-C411 K15 Rabbit Abcam ab52816

4.1.3. Secondary Antibodies Biotinylated Anti-RB IgG Goat Vector BA-1000 Biotinylated Anti-Mouse IgG Horse Vector BA-2001 Biotinylated Anti-Goat IgG Rabbit Vector BA-5000 CY 3-conjugated Anti-Rabbit Donkey JIR 711-165-152 CY 3-conjugated Anti-mouse Donkey JIR 715-165-150 CY 2-conjugated Anti-Rabbit Donkey JIR 711-225-152 CY 2- conjugated Streptavidin Donkey JIR 016-220-084 CY 3- conjugated Streptavidin Donkey JIR 016-160-084 CY 3-conjugated Anti-Chicken Donkey JIR 703-165-155

4.1.4. PCR primers Cre-neu-LP (F): 5’-CAGGGTGTTATAAGCAATCCC-3’ Cre-neu-UP (R): 5’-CCTGGAAAATGCTTCTGTCCG-3’ R26RFP wt (F): 5’-AAGGGAGCTGCAGTGGAGTA-3’ R26RFP wt (R): 5’-CCGAAAATCTGTGGGAAGTC-3’ GFP Common: 5’-AAAGTCGCTCTGAGTTGTTAT-3’

4. MATERIAL AND METHODS

GFP Wildtype: 5-GGAGCGGGAGAAATGGATATG-3’ Krt15 wt (F): 5’-GCTGGTATTGGTGTCAGAGAAG-3’ Krt15 wt (R): 5’-CCTGCACCAGACACTTAGATTT-3’

4.1.5. Materials used in cell culture Biolite 6 Well Thermo Scientific 130184 Biolite 12 Well Thermo Scientific 130185 Biolite 24 Well Thermo Scientific 930186 Sterile filter - Puradisc 25 mm Whatman 6780-2504 Sterile filter - Minisart 0.2 µm sartorius stedim 16532 DPBS Life technologies 14190-136 DMEM Corning cellgro 10-013-CV 0.05% Trypsin-EDTA Life technologies 25300-054 Penicillin Streptomycin Life technologies 15140-122 CMF-HBSS - Hank's Balanced Salt Solution Life technologies 88284 GlutamaxTM Thermo Scientific g1870127 Corning® Matrigel® matrix Corning® 356234 N-2 Supplement Thermo Scientific 17502048 B-27 Supplement Thermo Scientific 17504044 recombinant mEGF R&D Systems 2028-EG CHIR99021 Cayman Chemical 13122

4.1.6. Instruments and software Microscope - Eclipse E600 Nikon E600

Mecury Lamp power supply Nikon C-SHG1

MAC5000 power base - manual shutter LUDL electronics 73005001 Camera 12bit Q IMAGING 01-QIC-M12C

NIS - Imaging Software Nikon iVision - Imaging Software Bio Vision Technologies

Tissue-Tek VIP 5 Sakura

Thermo Shandon Histocentre 2 Sakura RM2255

4. MATERIAL AND METHODS

Automated Microtome Leica 22-047-239

TBS Tissue Floating Bath Fisher Scientific 6846

Isotemp Incubator Fisher Scientific 11-690-637D pH Meter - accumet basic Fisher Scientific AB15 pH Buffer 7.00 Thermo Scientific Orion 910107 pH Buffer 4.01 Thermo Scientific Orion 910104

Microwave oven General Electric Co. JES1460DN1

Pressure cooker-Antigenretriever Aptum Biologics #R2100US

Isotemp Incubator Fisher Scientific 11-690-650D

Millipore UV Synergy SYNSV0000

Nano Drop 2000 Thermo Scientific E112352 Universal Hood ii (visualizing gels) BIO RAD 721BR08505 Real-Time PCR System life technologies 272002766 2720 Thermal Cycler life technologies 4359659 Centrifuge Kendro Laboratory 75004377 Hoshizaki ice maker Hoshizaki F344-HAF UV-VIS spectrophotometer Beckmann Coulter DU 640 Freezer -120 Thermo Electron Corp. 7404 Electrophoresis System Fisher Scientific FB 105 Millipore UV Synergy SYNSV0000 Balance Mettler Toledo PM4000 Small incubator for single cell suspension Fisher Scientific Isotemp 202 PIPETTMANN P2 Gilson F144801 PIPETTMANN P10 Gilson F144802 PIPETTMANN P20 Gilson F123600 PIPETTMANN P100 Gilson F123615 PIPETTMANN P1000 Gilson F123602

4. MATERIAL AND METHODS

4.1.7. Consumable supplies 25 mL Tube Thermo Scientific 339652 50 mL Tube Thermo Scientific 339650 GeneJET Plasmid Miniprep Kit Thermo Scientific K0503 GeneJET Gel Extraction Kit Thermo Scientific K0692 GeneJET Genomic DNA Purification Kit Thermo Scientific K0721 GeneJET PCR Purification Kit Thermo Scientific K0702 GeneJET RNA Purification Kit Thermo Scientific K0732 Microcentrifuge Tube DOT scientific inc. RN1700-OMT 1-10 µl Pipet Tips DOT scientific inc. ESN-P10XLGS 1-200 µl Pipet Tips DOT scientific inc. ESN-0200NBG Clear Adhesive Film Micro Amp 43011 High sided Low profile PCR Plate Thermo Scientific AB-1900 Fast optical 96-well reaction plate Life technologies 4346906 TaqMan Reverse Transcriptase Fisher Scientific N8080234 Power SYBR Green Master Mix Life technologies 4367659

4.1.8. Animal care Plastic mouse cage Ancare AN75 Mouse cage lid Ancare N10 Stainless steel card holder Ancare HH35 Micro filter top Ancare R20MBT Rodent diet Pico Labs #5053 Irradiator RS2000 RAD source

4. MATERIAL AND METHODS

4.2. Animal experimental methodology The Institutional Animal Care and Use Committee of the University of Pennsylvania approved all animal studies. All experiments were designed and executed in respect to mouse protocol #805215.

4.2.1. Origin of transgenic mice used in the experiments • Krt15-crePR1 mice were provided by Dr. George Cotsarelis, a collaborating partner at the University of Pennsylvania and recently published in (Morris et al., 2004) • Krt15-/- mice were provided by Dr. George • C57BL/J6 wild-type mice (WT) were purchased from Jackson Laboratories, Lot Nr. 000664 and recently published in (Mouse Genome Sequencing Consortium et al., 2002) • Rosa26mTomato/mGFP (R26mT/mG) mice were purchased from Jackson Laboratories, Lot Nr. 007576 and recently published in (Lozano et al., 2012) • Apcfl/fl mice were obtained from the NCI Mouse Repository (Kuraguchi et al., 2006) • Rosa26LSL-td Tomato(R26Tom) were kindly provided by Dr. Christopher Lenger (University of Pennsylvania)

4.2.2. Animal care Krt15-crePR1 and Krt15-/- mice were generated by a collaborating team at the Transgenic and Chimeric Mouse Facility of the University of Pennsylvania School of Medicine and described previously by a collaborating team (Morris et al., 2004). C57BL/J6 wild-type mice (WT), Rosa26mTomato/mGFP (R26mT/mG) reporter mice and inducible DTR Rosa26iDTR (R26iDTR) mice were obtained from Jackson Laboratories. All animals were held in the animal care facility of the biomedical department of the University of Pennsylvania. The facility fits ABSL2 standard and was managed by the AAALAC approved University Laboratory Animal Resource (ULAR) division. The animals were subject of 12-hour night and day cycle and had free access to water and food. The mice were housed in AN75 Mouse Cages offering 75 square inches of floor space on dust free softwood kindling. The room temperature was of 22 ± 2 °C und the relative air humidity set to 55%.

4. MATERIAL AND METHODS

4.2.3. Genotyping procedure Tissue samples were collected from the tail of the mice and placed on ice. We extracted DNA from the tissue using the GeneJET Genomic DNA Purification Kit. Individual PCR was performed to genotype each allele. Each sample DNA was tested via PCR against a positive and negative control.

4.2.4. Euthanasia and organ harvest Mice were sacrificed at different time points using a CO2 inhalation chamber. Following CO2 inhalation, we ensured the successful sacrifice via cervical dislocation. We harvested tongue, esophagus, stomach, small intestine and colon via sagittal incision from the abdomen to the thorax. Harvested tissue was immediately prepped for tissue fixation.

4.2.5. Cre / LoxP recombination The Cre / LoxP recombination is a common method in molecular biology used for gene insertion, deletion or translocation at a specific LoxP site in genome. The Cre recombinase is an enzyme derived from P1 Bacteriophages (Abremski and Hoess, 1984) catalyzing DNA recombination between two short DNA recognition sequences. Those sequences are called loxP (locus of {X} over P1) and consist of 34 base pairs (bp) specific for each loxP site. Each LoxP site consists of two 13-bp palindromic sequences that are aligned on both sides of an asymmetric 8-bp spacer region (Gierut et al., 2014). The basic strategy for Cre/LoxP-directed gene knockout experiments is to flank, or ‘‘flox’’ the target gene with two loxP sites; the orientation of the two loxP sites determines whether the DNA sequence is being excised or inverted by the Cre. By using ligand depended Cre such as CreERT2 or CrePR1 in vivo, inducibility can be achieved. Inducible Cres are commonly linked to an estrogen or progesterone receptor and inactive until induction. Synthetic receptor ligands such as RU486 for the CrePR1 can penetrate the cell nucleus and activating the inducible Cre.

4.2.6. Experimental strategy for the Krt15-crePR1; R26mT/mG mouse. Krt15-crePR1 mice were bred with R26mT/mG and 6-week old mice were used for lineage tracing experiments. Cre recombination was induced in Krt15-crePR1; R26mT/mG mice by intraperitoneal (IP) injection of 0.5 mg of RU486 (Sigma-Aldrich). RU486 was dissolved in ethanol and then

4. MATERIAL AND METHODS diluted in peanut oil. RU486 was administered daily for five continuous days and mice were then sacrificed at different time points. Tongue, esophagus, forestomach and intestinal tissues were removed and transferred into zinc formalin fixative. Appropriate control was achieved with untreated Krt15-crePR1; R26mT/mG mice harvested on the same time points as the experimental mice.

Krt15 Promoter Cre PR1 R26 mTomato R26 mGFP LoxP LoxP

Krt15 Promoter Cre PR1 R26 mTomato R26 mGFP

Inject with RU486

mGFP Krt15 Promoter Cre PR1 R26

Figure 2: Breeding strategy for the Krt15-crePR1; R26mGFP mouse.

Krt15-crePR1 mice were bred with the Rosa2 mTomato/mGFP reporter mice. At 6 weeks of age the mice were injected with RU486, a synthetic progesterone receptor agonist also known as Mifepristone. RU486 induces the ligand dependent CrePR1 which would then excise the R26 mTomato Sequence at the loxP site. After deletion of the mTomato cassette only recombined Cells with active Krt15 promoter will express mGFP, thus enabling lineage tracing.

4.2.8. Experimental strategy for Krt15-crePR1; R26Tom mice. Krt15-CrePR1 mice were bred with R26Tom mice and the progeny mice were used for 3D organoid culture. Cre recombination was induced in six- to eight-week old Krt15-CrePR1;R26Tom mice by

4. MATERIAL AND METHODS a single I.P. injection of 0.5mg RU486. Mice were sacrificed 24 hr. later and the small intestines were harvested. Tissues were processed as described below for crypt culture and/or single cell sorting.

4.2.9. Experimental strategy for irradiation of Krt15-crePR1; R26mT/mG and Krt15-/- mice. Krt15-crePR1; R26mT/mG were injected with 0.5 mg of RU486 to induce recombination. C57BL/J6 wild-type mice and Krt15-/- knockout mice did not receive Ru486 treatment. Furthermore, all mice were administered one injection of BrdU (Sigma-Aldrich) 1,5 h prior to sacrifice (Kalabis et al., 2008). After 5 consecutive days of RU486 treatment we irradiated the mice with a 12 Gy whole body dose of radiation. Mice were sacrificed 2 and 5 days after irradiation treatment. The gut was removed and transferred to zinc formalin fixative. After following fixation protocol, the tissues were embedded in paraffin and cut using the microtome in sections 5 μm of debts for IF and IHC staining.

Inject with RU486 Irradiation for 5 days with 12 Grey

Krt15 Promoter Cre PR1 R26 mGFP Sacrifce and analysis 2 & 5 days following irradiation

Figure 3: Experimental strategy for Krt15-crePR1; R26mT/mG mouse irradiation.

The Krt15-crePR1; R26mT/mG mice as well as the WT mice were treated with RU486 for 5 days to ensure high recombination rate. After 5 days the mice were irradiated at 12 Grey ablating most radiosensitive intestinal stem cells. Following irradiation, the mice were sacrificed at 2 and 5 days post treatment.

4.2.10. Experimental strategy for Krt15-CrePR1;Apcfl/fl;R26mT/mG mice Krt15-crePR1; R26mT/mG were bred with Apcfl/fl mice to obtain Krt15-CrePR1;Apcfl/fl;R26mT/mG mice. Cre recombination was induced using daily injections of 0.5mg of RU486 for five consecutive days. Experimental mice were sacrificed 150 days post recombination or at prior time point in concert with severe weight loss.

4. MATERIAL AND METHODS

4.3. Tissue fixation Extracted tissues were splayed open and sandwiched between pieces of filter paper to fix flat. The tissues, wedged in between filter paper, were then placed in small tissue fixation cassettes and emerged in a 4% paraformaldehyde (PFA) solution at 4 °C overnight. The following day, we washed the tissue samples with 1X PBS for 60 minutes and then transferred the samples to a 70% Ethanol solution at 4°C. The cassettes could then be processed with Tissue-TEK VIP for paraffin embedding. The paraffin embedded tissue could then be cut in sections 5 μm of debts, using the microtome. Each section was placed on a glass slide and left to dry at 60°C over night.

4.4. Immunohistochemistry (IHC) To dewax the slides, we preheated them using a 60°C incubator for 15 minutes, after which the slides were placed in a Xylene bath. The slides were kept in Xylene for 3 minutes and then moved to a second Xylene bath for the same time-period. Subsequently, the slides were then moved into 100% Ethanol twice for two minutes per interval. Following this step, the slides were then moved from 95% Ethanol to 80% and then to 70% every 1 minute. Finally, the slides were rinsed in milliQ water for 1 minute before antigen retrieval. Antigen retrieval was performed by pressure cooking, immersing the slides in citric acid (pH 6.0) buffer. We incubated the slides for 20 minutes within the pressure cooker and then left them to cool for 40 minutes. Once the slides cooled to room temperature, we gently rinsed them with milliQ water for 5 minutes. After that, the slides were quenched in 30% hydrogen peroxide for 15 minutes, subsequently eliminating the endogenous peroxidase reaction. Before blocking, slides were rinsed off with milliQ water and 1 X PBS for 5 minutes each. We used a hydrophobic pen to restrict the volume of the reagents applied to each section. For IHC blocking we used a Biotin/Avidin system. This system is based on the fact that the rather large Glycoprotein Avidin depicts a strong affinity to the Biotin molecule, thus forming Avidin/Biotin structures on non-specific antibody loci (Hsu et al., 1981). Avidin D blocking reagent (Vecta) was applied first for a time-period of 15 minutes and then rinsed off with 1 X PBS for 5 minutes. Following Avidin D we applied Biotin blocking reagent (Vecta) for another 15 minutes and rinsed with 1 X PBS. As final step before antibody treatment we used unspecific Protein Blocking Agent for 10 min. Primary antibodies were diluted in PBT and incubated overnight at 4°C. Following 3 washing steps with 1 X PBS of 5 minutes each slide was then

4. MATERIAL AND METHODS incubated with biotinylated secondary antibodies diluted in PBT for 30 min at 37°C. Next we used the ABC Reagent (Vector Laboratories) for an additional 30 min at 37°C. Finally, slides were treated with DAB substrate, making sure that consistent exposure time within each slide was set and counterstained with hematoxylin for 5 seconds.

4.4.1. IHC in combination with the M.O.M.-kit The M.O.M.™ immunodetection Kit was specifically designed by Vector®, to reduce endogenous mouse IgG staining when using mouse primary antibodies on mouse tissue (He et al., 2013). Specific mouse horse radish peroxidase (HRP) polymers significantly reduce background from mouse antigens and thereby allow us to use mouse K15 antibodies on our WT and K15-/- mice.

4.5. Immunofluorescence (IF) Sections were rehydrated in the same fashion as described in 2.4 and underwent the same antigen retrieval steps. Contrary to the IHC slides, IF sections were blocked using an adapted blocking buffer with 1% BSA, 0,3% Triton X and 5% Donkey serum in 1X PBS for 1h at room temperature. Following blocking steps, the slides were rinsed off with milliQ Water and incubated with primary antibodies diluted in 1% BSA overnight at 4°C. Sections were then incubated with secondary antibodies for 1h at room temperature and counterstained with DAPI for 1 minute.

4.5.1. Immunofluorescence using the TSA system The Tyramide Signal Amplification (TSA™) is a HRP-mediated method that was developed in the late 1990s to detect proteins and polynucleotides in situ (van Gijlswijk et al., 1996). Sensitivity levels using TSA systems are greatly higher than the common HRP based methods (Liu et al., 2006). TSA Biotin Kits use HRP to catalyze covalent deposition of biotin labels directly adjacent to the immobilized enzyme. The labeling reaction is quick (less than 10 minutes) and deposited labels can be detected with streptavidin conjugates for imaging in brightfield or fluorescence microscopy. After overnight incubation with primary antibodies and consecutive washing steps with PBS, we added biotinylated secondary antibodies, incubating them for 30 minutes at 37°C. Following washing, we added the streptavidin-HRP (Perkin Elmer) for 30 minutes at RT. Then, we washed the slides with PBS and added the biotinylated Tyramide Reagent (Perkin Elmer) for 10 minutes at RT. We then added the CY2 conjugated Streptavidin binding the Tyramide.

4. MATERIAL AND METHODS

Subsequently, this enabled us to stain another marker using a tertiary antibody from another species and CY3 conjugated for antibodies.

4.6. Alcian Blue staining (AB) Sections were rehydrated in the same fashion as described in 4.4. and underwent the same antigen retrieval step. The slides were then quenched with hydrogen peroxide and blocked with Avidin, Biotin and BSA as described in 4.4. Primary antibodies were diluted in PBT and applied overnight at 4°C. Following the washing steps with 1 X PBS, the slides were then incubated with biotinylated secondary antibodies diluted in PBT for 30 min at 37°C. Next we used the ABC Reagent (Vector Laboratories) for an additional 30 min at 37°C. The slides were developed with DAB substrate for 10 seconds and immediately washed in milliQ H2O. The slides were then placed in 3% acetic acid for 3 minutes. The samples were then treated with 1% Alcian Blue in 3% acetic acid (pH2.5) for 30 minutes. After a 10-minute wash in milliQ water we counterstained the samples with Nuclear fast RED 0.1% for 40 seconds. After counterstaining, we dehydrated the slides and placed them in Xylene for mounting.

4.7. Single cell isolation Crypt single cells were isolated as described previously by our collaborators (Hamilton et al., 2015). Briefly, the small intestine was opened longitudinally and washed with cold PBS. Tissue was incubated 20 min on ice in PBS-EDTA-DTT. Tissues were then incubated in PBS-EDTA at 37°C. Crypts were isolated from the epithelial fraction using a 70μm filter. Crypts were then dissociated to a single cell suspension using PBS/dispase at 37°C.

4.8. Crypt isolation and 3D organoid culture Crypts were isolated from the mouse small intestine. Tissues were chilled in Ca2+-Mg2+-free HBSS (CMF-HBSS) with 1mM N-acetyl-cysteine (NAC). Tissues were then incubated in CMF- HBSS/1mM NAC/10mM EDTA for 45min at 4°C. Epithelial cells were then dissociated through vortexing/resting cycles. Crypts were separated from epithelial dissociation with 70μm filter. Crypts were pelleted and resuspended in Basal Media (Advanced DMEM/F12, 2mM GlutamaxTM, 10mM HEPES, 1X Penicillin/Streptomycin, 5μM CHIR99021 (Cayman Chemical), 1mM NAC,

4. MATERIAL AND METHODS

1x N2 Supplement (Gibco) and 1X B27 Supplement (Gibco)). Approximately 500 crypts were then embedded in 80% Matrigel/20% ENR (Basal Media, 50ng/ml recombinant mEGF(R&D Systems) and 1% Noggin/R-Spondin conditioned media). Enteroids were cultured in ENR.

Tissue extraction & Transfer isolated crypts into crypt isolation 3D organotypic culture

Figure 4:Experimental strategy for 3D organoid structures

Krt15-CrePR1;R26mTom/mGFP mice were injected with 1 dose of RU486 sacrificed 24h post injection. Intestinal tissue was then extracted and crypts were isolated following the Hamilton protocol. Isolated crypts were then transferred into matrigel matrix for 3D organoid culturing.

4.9. Irradiation Six to eight-week old mice were subjected to 12Gy whole-body gamma-irradiation (Gammacell 40 Cesium 137 Irradiation Unit). Radiation was used as the cytotoxic model, as dose delivery throughout the epithelial tissue is known to induce stem cell renewal (Otsuka et al., 2013). Mice were sacrificed 5 days later and the small intestines were harvested and fixed for histology. Surviving crypts, crypt length and villi height were measured on H&E stained ileum sections. Twenty-five high magnification fields were analyzed for each mouse and a minimum of 100 crypts or villi was measured for each mouse.

4.10. Histometric analysis

4.10.1. Crypt length analysis Measurements were performed on small intestinal crypts and villi using an Eclipse E600 microscope with NIS imaging software. For each individual mouse, crypt depth was measured in 10–15 intestinal circumferences. An intestinal circumference is a convenient section of intestinal epithelium representing a relative standard length and width of the individual mouse (Booth et al., 2004). Out of those 10–15 circumferences we measured 120 crypts per animal. Where possible, the crypts and villi measured were continuously indicating a true longitudinal section. Crypt and

4. MATERIAL AND METHODS villus height were measured via ImageJ and the data presented as a group which means a change in size from WT control.

4.10.2 Crypt microcolony assay The term microcolony was first used in 1970 (Withers and Elkind, 1970) to describe regenerative crypts post injury which were believed to repopulate the intestinal epithelium. The number of surviving and regenerating crypts per intestinal circumference was scored and the average per mouse and per group determined. A surviving crypt was defined as one that had 10 or more BrdU+ cells. Only regions that were orientated correctly and did not contain Peyers patches were scored (Booth et al., 2015; Jones et al., 2015; Potten et al., 1994). Cytotoxic damage caused by ionizing radiation accelerated stem cell renewal, thus accumulating BrdU + proliferating cells (Roe et al., 1996).

4.10.3. Software and statistical analysis All tables presented in this dissertation were analyzed for statistical relevance, using the unpaired student T-Test. Calculations were done via GraphPad Prism® 6.0, applying a confidence interval of 99%.

5. RESULTS

5. RESULTS 5.1. Endogenous expression of Krt15 in the small intestine. Our first aim is directed at proving Krt15 expression in the intestinal epithelium of the small intestine. Furthermore, we were interested in the exact location of Krt15 expressing cells within the structure of the intestinal crypt and villi. To minimize the risk of false positive data, we planned two separate experiments targeted at the same inquiry. On one hand, we looked at the endogenous expression of Krtr15 in WT mice via IHC and on the other hand, we inspected the expression of GFP via IF in Krt15-crePR1; R26mT/mG mice.

5.1.1. Endogenous expression of Krt15 in WT mice We sacrificed the 4 mice at the age of 6 months, harvested the small intestine and conducted an IHC staining as described previously. We stained directly for K15 using a rabbit antibody created by Abcam (ab52816). As a negative control, we used a Krt15 knockout mouse provided by Dr. George Cotsarelis. To ensure validity of our antibodies, we used sections of mouse esophagus as positive controls.

WT 10 X K15 KO 10 X

Figure 5: Control of Krt15 knockout in Krt15-/- mice versus WT mice.

IHC pictures of mouse esophagi, taken with a 10X magnification lens. Sections were stained for K15 comparing WT mice (left) to Krt15 knockout mice (right). Krt15 knockout in Krt15-/- mice is complete showing no signs of Krt15 expression.

5. RESULTS

Krt15-/- mice demonstrated a complete knockout of Krt15 contrary to the consistent expression of Krt15 in basal compartment of the esophageal epithelium. We concluded that the antibody used specifically recognize K15. Small intestinal staining demonstrated similar specificity apart from background staining in the stroma of the villi. These results demonstrate the existence of Krt15 expressing cells in the intestinal epithelium. Krt15+ cells were found at two different locations within the intestinal epithelium: One group of Krt15 expressing cells was found in the villi, the second group of Krt15 expressing cells in the crypt.

WT 20 X K15 KO 20 X

Figure 6: Krt15 is expressed in intestinal epithelial cells.

IHC pictures of mouse intestinal epithelia, taken with a 20X magnification lens. We compared WT mice (left) to Krt15-/- control mice (right) and stained the tissue with rabbit K15 antibodies. WT mice expressed Krt15 at various locations in the villi as well as in the crypt. We suspected some degree of false positive staining at crypt base, due to antibody interaction with the Paneth cell population.

To verify our findings, we performed the same experiment using mouse antibody instead of rabbit antibody to stain for K15. To minimize endogenous background staining we used the M.O.M.® Immunodetection Kit specifically designed for using mouse primary antibodies on mouse tissue.

5. RESULTS

We could demonstrate a similar localization pattern of Krt15+ epithelial cells. Both experimental approaches show a distinct localization of Krt15 expressing cells at the bottom of the crypt. However, previous approaches did show interaction between the K15 antibody and the Paneth cell population, resulting in partially unspecific crypt staining. To avoid this problematic, we established a new mouse model (Krt15-crePR1; R26mT/mG) directly labeling the Krt15+ cells with GFP.

WT 20 X K15 KO 20 X

Figure 7: Krt15+ cells can consistently be identified in the intestinal epithelium.

Pictures of mouse intestinal epithelia, taken with a 20X magnification lens. We compared WT mice (left) to Krt15-/- control mice (right) using a mouse antibody combined with the M.O.M. kit. Location of Krt15+ cells with K15 mouse antibodies is consistent to that identified with rabbit antibodies. We suspect some degree of false positive staining at crypt base due to antibody interaction with the Paneth cell population.

5.1.2. Localization of Krt15+ cells in Krt15-crePR1; R26mT/mG mice. To further characterize the localization of Krt15+ epithelial cells, we examined GFP labeling on Krt15-crePR1; R26mT/mG mice. We analyzed sections of the small intestine via IF using GFP Antibodies from Abcam (ab6673) and E-Cadherin from Cell signaling (3195).

5. RESULTS

Krt15-crePR1; R26mT/mG 20 X Krt15-crePR1; R26mT/mG 20 X WT 20 X

Krt15 GFP Krt15 GFP Krt15 GFP E-Cadherin E-Cadherin E-Cadherin

Figure 8: Krt15-crePR1; R26mT/mG mice reliably label Krt15+ cell in the small intestine.

IF pictures of mouse intestinal epithelia, taken with a 20X magnification lens. Krt15+ cells are successfully labeled with GFP in Krt15-crePR1; R26mT/mG mice, E-Cadherin is labeled in red and nuclei in blue. Krt15+ cells are epithelial cells, showing same localization patterns as previously seen with direct K15 antibody targeting.

We examined four different Krt15-crePR1; R26mT/mG mice and two control mice. All mice were sacrificed at same time point and organs were harvested and fixated as described previously. To insure adequate Cre-recombination, the mice were treated with one single dose of RU486 and sacrificed 24 hours post injection. Only cells with an active Krt15 promoter will express GFP, thus enabling us visualize Krt15+ cells. We used combined staining with GFP and E-Cadherin, a transmembrane protein present in epithelial cells allowing us to distinguish between epithelial und stromal tissue. As shown in Figure 7 we recognized a similar pattern of localization as previously seen in IHC picture from WT mice. GFP labeled cells are present in the epithelium of the villi and in the crypt. Considering our interest in the role of Krt15 as a potential stem cell marker, we decided to focus our research on the Krt15 expression in the crypt. We analyzed 100 different intestinal crypts to determine the localization of crypt bound Krt15+ cells. High magnification pictures of intestinal crypt show GFP+ cells localized throughout the crypt, but predominantly in the stem cell compartment i.e. below the +4 position. This position is known to host slow cycling stem cells, often labeled with Bmi1, mTert or Hopx, an aspect we are planning to investigate promptly.

5. RESULTS

Percentage of GFP cells 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Crypt position

Figure 9: Krt15+ cells are predominantly located in the stem cell compartment.

The graph illustrates the localization of GFP+cells throughout the crypt. Krt15+ cells are located at various position within the crypt. However, they are predominantly located in-between second and sixth position. This compartment is known to host multipotent stem cells.

5. RESULTS

Krt15-crePR1; R26mT/mG 40 X Krt15-crePR1; R26mT/mG 40 X

5. 4. 3. 6. 2. 5. 1. 4. 3. Krt15 GFP Krt15 GFP 2. 1. E-Cadherin E-Cadherin

Figure 10: Krt15+ cells are predominantly located in the stem cell compartment.

IF pictures of mouse intestinal epithelia, taken with a 40X magnification lens. GFP is labeled in green, E-Cadherin in red and nuclei in blue. Krt15+ cells are predominantly located in-between the second and fifth position above crypt base. This compartment is known to host radioresistant quiescent stem cells such as Bmi1, Lrig1 or K19.

One key aspect of multipotent stem cells is the ability to give rise to all types of intestinal cell lineages. Our next aim focuses on this principal, to further investigate the role of Krt15.

5.2. Krt15 expression in various epithelial cell types present in the intestine. We formulated aim B to identify which cell types express Krt15 in the intestinal epithelium. We co-stained intestinal sections of Krt15-crePR1; R26mT/mG mice with common markers for intestinal cell lineages and GFP. Due to the lack of specific antibodies for ISCs, we can not perform co- staining for ISCs in the Krt15-crePR1; R26mT/mG mice. However, all secretory lineages have well established immunolabeling markers. Subsequently, we analyzed intestinal sections from four Krt15-crePR1; R26mT/mG mice.

5. RESULTS

5.2.1. Krt15 expression in enteroendocrine cells. Enteroendocrine cells produce a variation of cytokines impacting local and systemic metabolism. A known marker protein for enteroendocrine cells is Chromogranin A (Sei et al., 2011) which is a precursor protein to several functional peptides. Colocalization of both markers would indicate active Krt15 promoter in hormone producing enteroendocrine cells. We stained the sections with SP-1 Chromogranin A antibody from ImmunoStar (20085) and anti-GFP. We demonstrated occasional colocalization of K15 and Chromogranin A, suggesting a prolonged expression of Krt15 even in terminally differentiated neuroendocrine cells.

Chromogranin A 20 X Chromogranin A 40 X

Krt15 GFP Krt15 GFP Chromogranin A Chromogranin A

Figure 11: Colocalization of Krt15+ cells with enteroendocrine cells.

IF pictures were taken with 20X and 40X magnification lens. Enteroendocrine cells are targeted by chromogranin A and labeled in red, Krt15+ cells appear in green. Colocalization of K15 and Chromogranin A can be shown in isolated events, suggesting a potential differentiation of Krt15+ cells into enteroendocrine cells.

5. RESULTS

5.2.2. Krt15 expression in the goblet cell population. The intestinal epithelium is covered by a protective mucus gel consisting of mucin glycoproteins, produced and secreted by Goblet cells. The mucus layer is believed to be a dynamic defense barrier against gut bacteria and other pathogens (Gaskins et al., 2008). Goblet cells are the most common secretory cells within the intestinal epithelium. In this experimental step, we used Alcian Blue staining as a well-established method to visualize goblet cells in combination with immunolabeled GFP. We could show that a fraction of the goblet cells actively expressed Krt15.

GFP / AB 20 X GFP / AB 40 X

Krt15 GFP Krt15 GFP Mucin Mucin

Figure 12: Krt15+ cells colocalize with goblet cells in isolated events.

IHC pictures taken with 20X and 40X magnification. Alcian blue labels goblet cells in blue, Krt15+ cells appear brown. Colocalization of GFP and mucins can be shown in isolated events. This colocalization indicates that Krt15+ cells may also differentiate into goblet cells.

5.2.3. Krt15 expression in the Paneth cell population. Paneth cells are epithelial cells located at the bottom of the crypt, secreting antimicrobial proteins such as alpha-defensine and lysozyme. Paneth cells also produce local acting growth hormone and

5. RESULTS different modulating proteins essential for the stem cell niche. The niche constitutes a unique microenvironment which supports and maintains the self-renewal capacity of intestinal stem cells. Intestinal stem cells reside at the base of the crypt, generating progeny cells that differentiate into various cell lineages. Paneth cells play a predominant role in maintaining the delicate microclimate of the stem cell niche (Sailaja et al., 2016; Tan and Barker, 2014). We co-stained for Paneth cells by using lysozyme antibody from Diagnostic Biosystems (RP-028) and GFP antibody. Results show no colocalization between the Paneth cell population and the Krt15+ cells. These findings are particularly relevant considering the difficulties of K15 antibody interaction with Paneth cells in the direct approach taken in aim A. The pictures prove that Krt15 marks a distinct cell population located in the crypt that is not the Paneth cell.

Lysozyme 20 X Lysozyme 40 X

Figure 13: Krt15+ cells are located above crypt base, showing no colocalization with secretory paneth cells.

IF pictures taken with 20X and 40X magnification lens. Lysozyme staining marks paneth cells in red, Krt15+ positive cells are labeled in green. Krt15+ cells form a distinct population seemingly located above paneth cells. There is no colocalization between Krt15 expressing cells and the paneth cell population.

5. RESULTS

5.2.4. Cell-cycling activity among Krt15+ cells. Being unable to stain directly for known stem markers such as Lgr5, Bmi1 or K19, we envisaged to visualize actively proliferating cells. The expression of the Ki67 protein is strictly associated with cell proliferation. Ki67 is present in all active phases of the cell cycle (G1, S, G2 and mitosis) but absent from resting cells (Scholzen and Gerdes, 2000). We stained the sections using Ki67 antibodies from Abcam (ab16667) and anti GFP. Results demonstrated numerous crypt bound Krt15+ cells colocalized with Ki67, thus proving active that Krt15+ cells can occasionally overlap with transit-amplifying cells.

Ki 67 20 X Ki 67 40 X

Krt15 GFP Krt15 GFP Ki67 Ki67

Figure 14: Krt15+ cells are proliferating cells located at a +4 position.

IF pictures taken with 20X and 40X magnification lens. Ki67 labels proliferating cells in red, GFP expressing Krt15+ cells are labeled in green. Crypt bound Krt15+ cells colocalized with Ki67, proving them to be proliferative cells. As shown previously their positions within the crypt is located above crypt ground.

5. RESULTS

5.3. Evaluation of proliferative potential within the Krt15+ cell population.

To determine the growth of Krt15+ cells, it is necessary to examine Krt15+ progeny cells. A common characteristic of adult stem cells it that they are multipotent, i.e. able to give rise to all differentiated lineages throughout a tissue. The setting of a lineage tracing experiment would enable us to investigate the Krt15+ population and all progeny cells over a long period of time.

5.3.1. Lineage tracing experiment on Krt15-crePR1; R26mT/mG mice. We induced LoxP recombination via Ru486 injections in 6-week-old mice and sacrificed the mice at different time points. Experimental mice were injected with Ru486 for 5 consecutive days to insure maximum recombination. We sacrificed a total number of 49 mice over a period of 150 days at different time points (Days: 1,2,3,5,7,10,14,21,28,56,150). This setup enables us to determine the fate of the Krt15+ population at day one post injections up to their progeny cells at the latest time point, 150 days post induction. Considering that the small intestinal epithelium is renewed every 3 days in the mouse except for the Paneth cells, all GFP expressing cells past this day constitute Krt15+ progeny cells. Krt15+ progeny cells are cells arising from the Krt15+ origin cells. Any labeled cells beyond a week shows that Krt15+ have self- renewing capacity and are long-lived cells.

Figure 15: Timetable depicting experimental setup of the lineage tracing experiment over 150 days.

Experimental mice were treated during 5 consecutive days with RU486 to maximize induction rate. After successful induction, mice were sacrificed at specific time points for IF contemplation. The timeframe is set to determine the vivacity and endurance of the Krt15+ population and evaluate the fate of any daughter cells.

At early time points (day: 1,2,3,5), we observed a small population of Krt15+ cells located within the crypt. Following time points show a graduate expansion of GFP expressing Krt15+ progeny cells. Day 14 marks the first date on which crypts with Krt15+ cells consist of more than 50 percent

5. RESULTS of Krt15+ progeny cells. To determine the growth capacity of Krt15+ cells, we defined any crypt containing a minimum of 1 GFP expressing Krt15+ cells as “labeled crypt”. We evaluated 3 mice per time point (after 1,10 and 56 days) analyzing 150 crypts from intestinal cross sections per specimen. Only 1 in 4 crypts harbored Krt15+ cells, thus resulting in ~ 20 percent of labeled crypts per specimen. Yet Krt15+ progeny cells, within their labeled crypt proliferated to the point of representing the majority cell type after 56 days. Krt15+ cells demonstrated a remarkable perseverance as well as a long-lived cell pool capable of continued cell renewal.

Figure 16: Evolution of Krt15+ progeny cells within the crypt.

The graph depicts the evolution of crypt bound GFP labeled Krt15+ cells over a period of 56 days. The absolute amount of GFP labeled crypts represents around 20% of all crypts. However, the Krt15+ population within labeled crypts increases significantly in the time frame of 56 days.

Moreover, Krt15+ progeny cells showed signs of upward migration populating large portion of adjacent villi. Daughter cell capable of vertical differentiation into terminally differentiated epithelial cells.

5. RESULTS

CTL K15 GFP Day 1 Day 2 Day 3 E-Cadherin DAPI Figure for dissertation

Day 5 Day 7 Day 10 Day 14

Day 21 Day 28 Day 56 Day 150

Figure 17: Lineage tracing over 150 days depicts Krt15+ as proliferative, long-lived epithelial cells.

IF pictures taken with a 20X magnification lens over a time-period of 150 days. E-Cadherin labels epithelial cells in red, GFP expressing Krt15+ progeny cells appear green. As previously established Krt15+ cells are located at a +4 position or above, capable of cell cycling. Krt15+ progeny cells show signs of upward migration populating the entire crypt as well as adjacent villi.

5. RESULTS

Day 150 10 X

Krt15 GFP DAPI

Figure 18: Krt15+ progeny cells colonize the entire crypt as well as adjacent crypt and villi after 150 days.

IF pictures taken with a 10X magnification lens 150 days’ post RU486 recombination. Krt15+are labeled with GFP in green. Krt15+ progeny cells show signs of vertical and horizontal migration, colonizing adjacent crypts. For the first time Krt15 expressing cells can be seen at crypt base, suggesting that Krt15+ cells may give rise to CBC cells. Furthermore, the expansion to adjacent crypts underlines our theory of Krt15+ cells differentiating into all cell types.

5. RESULTS

5.3.2. Colocalization of Krt15+ cells with proliferating cells in a lineage tracing setting. We could demonstrate the high proliferative capacity of crypt bound Krt15+ in previous experiments. Yet, we intended to prove that Krt15 markes long-lived, robustly proliferative cell types. This could be achieved by demonstrating a persistent expression of cell cycling markers. We chose to analyze the same time-points, depicting a consistent expansion of the Krt15+ population. Krt15+ cells reliably colocalized with proliferation marker and maintained co- expression over time. The late time points still showed highly proliferative Krt15+ progeny cells, thus depicting a brand of long lived proliferative cells.

Ki67 CTL Ki67 Day 1 Ki67 Day 10 Ki67 Day 56

Figure 19: Krt15 marks a long-lived consistently expanding crypt population.

IF photos were taken with 20X magnification lens labeling Krt15+ progeny cells in green and Ki67 expressing, proliferative cells in red. Crypt bound Krt15+ progeny cells were consistently proliferating and demonstrated a vertical and horizontal expansion.

5. RESULTS

5.3.3. Colocalization of Krt15+ cells with enteroendocrine cells in a lineage tracing setting. As previously shown, Krt15+ cells can differentiate into all known classes of epithelial cells. The question remains whether K15+ progeny cells maintain the ability of vertical differentiation over time. Persistent co-staining shows that Krt15+cells can give rise to enteroendocrine cells. We could determine a persistent colocalization of GFP and Chromogranin A over a period of 56 days. This underlines our theory of multipotent Krt15+ cells, capable to differentiate into highly specified enteroendocrine cells

Chromogranin A 40 X Chromogranin A 40 X Chromogranin A 40 X Chromogranin A 40 X

Figure 20: Krt15+ progeny cells maintain the ability to differentiate in to enteroendocrine cells.

IF pictures taken with a 40X magnification lens at various time-points post RU486 recombination. Enteroendocrine cells are labeled in red, Krt15+ progeny cells are labeled in green. Differentiated enteroendocrine cells actively express Krt15 and demonstrate persistent expression pattern over time.

5. RESULTS

5.3.4. Colocalization of Krt15+ cells with goblet cells in a lineage tracing setting. As demonstrated in the previous experiment, intestinal goblet cells express Krt15, however the fate of this subset of Krt15+ progeny cells remains unclear. The expansion of the crypt bound Krt15+ progeny cells indicate an upwards migration and maturation. The increased colocalization between K15 and Mucins proves a persistent ability of Krt15+ progeny cells to differentiate into goblet cells.

GFP / AB CTL GFP / AB Day 1 GFP / AB Day 10 GFP / AB Day 56

Figure 21: Krt15+ progeny cells reliably differentiate into goblet cells over time.

IHC pictures taken with a 40X magnification lens at various time-points post RU486 recombination. Alcian blue staining labels Goblet cells in blue, Krt15+ cells turn up brown. Crypt bound Krt15+ cells proliferate and demonstrate a vertical expansion, differentiating into goblet cells, while maintaining Krt15 expression.

5. RESULTS

5.3.5. Colocalization of Krt15+ cells with Paneth cells in a lineage tracing setting. Previous experiments established no interaction between the Krt15+ population and the Paneth cell population. Nonetheless, the expanding crypt bound Krt15+ progeny cells seem to populate the totality of the crypt after 56 days including Paneth cells. This evidence consolidates our hypothesis that Krt15 expressing cells can differentiate into all cell types present in the intestinal epithelium.

Lysozyme CTL Lysozyme Day 1 Lysozyme Day 10 Lysozyme Day 56

Figure 22: Krt15+ cells can differentiate into paneth cells over time.

IF pictures taken with a 40X magnification lens at various time-points post RU486 recombination. Paneth cells are labeled in red, GFP expressing Krt15+ cells are labeled in green. Crypt bound Krt15+ cells proliferate and demonstrate a vertical and horizontal expansion, seemingly co-localizing with paneth cells.

To sum up, the lineage tracing experiments indicate that Krt15 marks a type of long lived epithelial cell located in the stem cell compartment of the crypt. Furthermore, we could show a colocalization with transient amplifying cells and depict the ability to differentiate into all epithelial lineages.

5.3.6. Krt15+ cells depict multipotency in 3D organoids To further investigate the self-renewing capacity of Krt15+ crypt cells, we used Krt15- CrePR1;Rosa26LSL-tdTomato (Krt15-CrePR1;R26Tom) mice. A single injection of RU486 was used to induce recombination and mice were sacrificed 24 hrs. later. Crypts were isolated and grown as 3D organoids. Krt15+ (Tomato+) cells were detected in 3D organoid crypts and these cells can expand and form ribbon as observed in vivo.

5. RESULTS Brightfield Krt15-crePR1; R26 Tomato Krt15-crePR1; R26 Tomato omato td T

Tomato Tomato Krt15-crePR1; R26 Krt15-crePR1; R26

Figure 23: live 3D organoids formed by Tomato+(Krt15-derived)

Live microscopy of 3D organoids taken with a 40X magnification lens in brightfield and CY 3. Krt15-derived Tomato+ are labeled in red. Krt15+ can form functioning organoid structures.

omato td T Krt15-crePR1; R26 Tomato Krt15-crePR1; R26 Tomato

Figure 24: Fixated 3D organoids formed by Tomato+(Krt15-derived) sf

IF pictures taken with a 20X magnification lens depicting various organoids. Krt15+ cells are labeled in red. Krt15+ cells can form functioning organoid structures.

Interestingly, following 3D organoid passaging, we observed crypts that were entirely labeled with Tomato+ cells (Figures 20 and 21). Multipotency of Krt15+ cells was confirmed in these 3D organoids with Tomato+ proliferative cells, goblet cells, enteroendocrine cells and Paneth cells.

5. RESULTS

Tom Tom KI67 MUC2

Krt15-crePR1; R26 Tomato Krt15-crePR1; R26 Tomato Tom Tom CHGA LYS

Krt15-crePR1; R26 Tomato Krt15-crePR1; R26 Tomato

Figure 25: Co-staining of Tomato with Ki-67, Mucin 2, Chromogranin A and Lysozyme in fixated Organoids

IF pictures taken with a 40X magnification lens depicting various organoids. Tomato+ Krt15 derived cells are labeled in red whereas colocalized markers appear in green. Tomato+ show signs of multipotency, giving rise to all epithelial lineages.

Furthermore, single crypt cell suspension was prepared from Krt15-CrePR1;R26Tom mice 24 hrs. following Cre recombination. Suspended cells were then transferred into a matrigel matrix to determine their proliferative capabilities in isolation. Krt15+ cells were able to grow as single cell culture and multicellular organoid after 10 days. These results suggest that Krt15+ crypt cells display “stemness” features.

5. RESULTS

tdTomato Brightfield D1 D2 D4 D6 D10

Figure 26: live 3D organoids formed by isolated single Tomato+(Krt15-derived) cell.

Live microscopy of organoid taken with a 40X magnification lens in brightfield and CY 3 at various time point. Krt15-derived Tomato+ cells are labeled in red. Isolated Krt15 derived cells form functioning organoid structures from a single cell.

All together, these results demonstrate that Krt15+ crypt cells are long-lived and multipotent cells with self-renewing capacity, characteristics consistent with stem cells.

5. RESULTS

5.4. The role of Krt15+ cells in response to radiation injury and their regeneration capabilities. Our earlier findings demonstrated the proliferative capacities of crypt bound Krt15+ cells, yet it remains unclear to what extend this influences epithelial homeostasis. Radiation injury is known to initiate stem cell proliferation as epithelial repair becomes necessary (May et al., 2008; Yan et al., 2012). The remarkably quick turnover of the intestinal epithelium is driven by multipotent stem cells, crucial for upholding epithelial homeostasis. Durability and resistance to extracorporeal noxae is a key feature for stem cells and is predominantly expressed in quiescent +4 positioned stem cells. Lgr5+ CBC stem cells seem to be radiosensitive to the point of stem cell sterilization (van Es et al., 2012; Yan et al., 2012), whereas K19+ and Bmi1+ stem cells are radioresistant (Asfaha et al., 2015a; Yan et al., 2012). Past experiments have shown dormant +4 positioned stem cells capable of replacing ablated CBC cells after irradiation. We hypothesized K15 would label a type of radioresistant cell type.

5.4.1. The role of Krt15+ cells in tissue recovery. We sought to investigate the self-renewal properties of Krt15+ regarding their sensitivity to radiation. The key aspect was to investigate the intestinal epithelium in the process of convalescence, following radiation injury. Thus, we treated a total of 12 adult Krt15-crePR1; R26mT/mG mice with RU486 for 5 consecutive days. The mice were then divided into two groups, one receiving a 12 Gy whole-body irradiation, the other being the control group. This dose is sufficient to cause severe crypt destruction and stem cell ablation in the majority of the crypts (May et al., 2008). Both experimental groups consisted of 6 mice each, which we then sacrificed after 2 and 5 days post irradiation. As depicted in figure 19, we compared irradiated and non- irradiated Krt15-crePR1; R26mT/mG mice 2 and 5 days post treatment. It appears that two days post irradiation, the effect of epithelial cell damage still predominates and only little cell proliferation can be seen. Nonetheless, self-renewal activity peaks 5 days post irradiation treatment and crypts fill with rapidly proliferating cells. Injured epithelial cells are rapidly replaced by cells emerging from crypt stem cell populations. Crypts in the process of recovery appear enlarged and deepened containing a predominant number of proliferating cells. Experiments suggest that the vivacity of the epithelial tissue after injury correlates with the crypt structure (Al Alam et al., 2015; Kantara

5. RESULTS et al., 2015). Krt15+ cells prove to be resilient to radiation injury and persist in their crypt location. The expeditious expansion of Krt15 progeny cells after radiation injury demonstrates the implication of Krt15 expressing stem cells in the process of recovery. To quantify our findings, we conducted a count of crypt cells, establishing a ratio of GFP+ cells to all crypt cells. We counted 50 of intact crypts per specimen, focusing on GFP labeled crypts consequently concentrating on crypts with Krt15+ cells, which constituted only 10% of all crypts. Table 2 depicts the significant increase of the Krt15+ population 5 days post irradiation underlining their resilience to radiation damage and their capability to repopulate the crypt. We compared the percentage of Krt15+ crypt cells 2 and 5 days post irradiation and standardized it to non-irradiated experimental mice. Two days post treatment the percentage of GFP+ cells in both irradiated and non-irradiated mice remained around 18%. Five days post treatment, the number of GFP+ cells drastically increased in the irradiated mice to around 90 % of the crypt population. Moreover, we investigated the proliferation rate via Ki67 depicted in table 3. We determined the amount of KI67+ cells within the Krt15+ cell population, by establishing a ratio of GFP+ and Ki67+ co-stained crypt cells versus all Ki67+ crypt cells. Five days post irradiation an overwhelming majority of all dividing cells co- stained with GFP, thus supporting our thesis of Krt15+ cells being involved in tissue regeneration. Furthermore, the comparison of irradiated intestinal tissue to non-irradiated tissue depicted a disparity of proliferation patterns of Krt15+ cells. In healthy tissue Krt15+ cells appeared to divide at relatively moderate pace. However, irradiation seemingly eliminating most rivalry stem cells causes a major influx in Krt15+ propagation.

5. RESULTS

2 Days after treatment 5 Days after treatment

Ki 67 20 X Ki 67 40 X Ki 67 20 X Ki 67 40 X Non irradiated

Ki 67 20 X Ki 67 40 X Ki 67 20 X Ki 67 40 X Irradiated

DAPI GFP Ki 67

Figure 27: GFP marks radioresistant Krt15 positive cells capable of repopulating intestinal crypts after radiation injury

IF pictures were taken with a 20X and 40X magnification lens and stained with GFP in green and Ki67 in red. Crypt bound Krt15+ cells are radioresistant and show increased proliferation post irradiation. 5 days post irradiation Krt15+ progeny cells label the entire crypt. Whereas Krt15+ cells in non-irradiated mice remain at a +4 and above position, Krt15+ progeny cells in irradiated mice seem capable of repopulating the entire crypt including ablated CBC cells.

5. RESULTS

Figure 28: Krt15+ cells vitally expand post radiation injury, labeling the majority of the crypt population.

We counted crypt cells and established a ratio of GFP+ cells to all crypt cells, two and five days post irradiation. Two days post treatment the percentage of GFP+ cells in both irradiated and non-irradiated mice remained around 18%. Five days post treatment the number of GFP+ cells drastically increased in the irradiated mice to around 90 % of the crypt population.

Figure 29: Proliferative capacity of Krt15+ progeny cells within labeled crypts, post radiation injury.

We compared GFP+ and Ki67+ co-stained crypt cells to all Ki67+ cells to reiterate the portion of Krt15 expressing cells in the proliferating cell population. Two days post irradiation, 50% of all proliferating cells co-stained with GFP, yet five days post irradiation that ratio increases to almost 95%.

5. RESULTS

5.4.2. The importance of Krt15 expressing cells in irradiation models compared to Krt15-/- In our previous experiments, we established K15 as a marker for radiation resistant stem-cells, capable of repopulating intestinal crypt after injury. Nonetheless, the question remains whether Krt15+ stem-cells are required for crypt homeostasis and injury management. To this purpose, we compared 4 adult Krt15-/- mice to 3 adult Wildtype mice. Prenatal deletion of Krt15 in mice did not disrupt intestinal epithelial homeostasis and all specimens were phenotypically unobtrusive. Manifestly, the organisms were capable of replacing Krt15+ expressing cells, thus upholding intestinal homeostasis. To examine the injury management of Krt15-/- mice, we designed a similar experiment as described earlier, irradiating the mice and screening for intestinal tissue damage. Mice were treated with 12 Gy or ionizing radiation at the age of 8 weeks and dissected 5 days post treatment. Small intestinal crypts were then examined, analyzing 100 intact crypts per specimen. We analyzed crypt-length of the ileum and jejunum, as a parameter of tissue rehabilitation in Krt15- /- mice and Wildtype. Histometric analysis of the crypt length showed significant differences in crypt length of treated Krt15-/- and Wildtype mice, as demonstrated in figure 20. We compared the average crypt length of all measurements and ran a T-test, detecting a highly significant length difference. The considerably shortened crypt and dysmorphic villi of the Krt15-/- mice indicated an inferior regenerative capability. The loss of radiation resistant stem-cells caused delayed injury recovery.

5. RESULTS

WT Ileum 20 X K15 - KO Ileum 20 X K15 - KO Ileum 20 X

97 µ m

m 58 µ

151 µ

129 µ m 92 µ

Figure 30: Depressed ileal crypt rehabilitation in Krt15-/- mice, post irradiation injury.

H&E pictures of the ileum were taken with a 20X magnification lens, depicting a significant difference in crypt-length between Krt15-/- mice and Wildtype mice. Crypt length was used as a marker for crypt homeostasis and cell renewal post injury. Inferior cell renewal properties presented in the Krt15-/- mice would indicate a deficit of radiation resistant stem cells.

Average crypt length in Ileum

250 **

200

150

100

Ileum crypt length in µm 0 Wildtype K15-KO

Figure 31: Discrepancies in crypt length in the ileum of WT mice and Krt15-/- mice 5 days post radiation injury.

We measured the average crypt length in the ileum post irradiation, to compare crypt vivacity of Krt15 knockout mice and wildtype. Crypt length of Krt15-/- mice was significantly shorter than that of wildtype mice. Crypt length 5 days post irradiation, was set as a criterion for crypt regeneration. Significant discrepancies between mice models would indicate a reduced activity of stem cells

5. RESULTS necessary to repopulate the crypt after injury. The loss of radiation resistant Krt15+ cells impacts the organism’s ability to cope with radiation injury.

We divided intestinal sections into duodenum, ileum and jejunum to determine the impact of Krt15 loss within the entire small intestine. The fact that Krt15 loss was impacting the ileum, to a greater extent, than the jejunum or duodenum could be explained by a more predominant role of Krt15+ stem-cells in the distal small intestine. Variation in stem cell expression along the small intestine has not yet been sufficiently described and should represent a promising field of research for upcoming experiments.

WT Jejunum 20 X K15-KO Jejunum 20 X K15-KO Jejunum 20 X

m 110 µ

95 µ

134 µ

144 µ m m

Figure 32: Jejunal crypts unchallenged post irradiation in Krt15-/- mice.

Pictures were taken with a 20X magnification lens, depicting various jejunal crypts with no significant difference in crypt length, between both specimens. The impact Krt15 loss seems to be more vigorous in the ileum compared to jejunum. These finding would indicate an uncontinued expression of Krt15 along the length of the small intestine.

5. RESULTS

Average crypt length in the Jejenum

250

200

150

100

50 Crypt length in µm in length Crypt

0 WT K15-KO

Figure 33: Discrepancies in crypt length between WT mice and Krt15-/- mice 5 days post radiation injury.

We measured the average crypt length in the jejunum post irradiation, to compare crypt vivacity of Krt15 knockout mice and wildtype. There was no significant difference in crypt length between wildtype mice and Krt15-/- mice. These findings would indicate a subordinated role of Krt15 in the jejunum compared to the ileum.

5. RESULTS

5.4.3. Microcolony assay on Krt15 expressing cells. A microcolony consist of a crypt with surviving clonogenic cells post irradiation injury. For experimental purposes, we defined a microcolony as a crypt with more than 10 proliferating non- Paneth cells. Hence using the amount of microcolonies as a predictive factor for the survival of clonogenic cells. We analyzed sections of wildtype and Krt15-/- mice, specifically studying the Ileum. Comparable to previous sets of experiments, we expected a more proliferative situation post injury in the wildtype specimen. Remarkably, the number of microcolonies in the Krt15-/- mice decreased as well as the overall amount of live cell within each crypt. Further analysis of the produced data indicates a highly significant deduction of clonogenic crypt cells and thereby the overall crypt population. Krt15 expressing stem cells seem to be relevant to crypt homeostasis in situations of systemic injuries. Only a small portion of all stem cells are radiation resistant, thus limiting the number of potential stem cells capable of tissue regeneration (Asfaha et al., 2015a; Roche et al., 2015). The data suggests that Krt15 is required for epithelial regeneration after radiation injury.

Irradiated WT Irradiated K15-KO

WT 20 X K15-KO 20 X K15-KO 20 X

Figure 34: Krt15-/- mice, portrayed a significant reduction of proliferating ileal crypt cells.

Picture of Ki67 stains taken with a 20X magnification lens, comparing wildtype mice to Krt15-/- mice, 5 days following irradiation. Wildtype mice have considerably more and larger microcolonies than Krt15-/- mice. The loss of radioresistant Krt15+ cells influences the epithelial repair greatly.

5. RESULTS

Figure 35: Significant reduction in absolute number of live cells per crypt in wildtype and Krt15-/- mice.

We counted surviving crypt epithelial cells as a parameter for crypt restoration. Krt15-/- mice have a significant reduction of the absolute cell population after radiation injury, indicating a deficit in active stem cells. The loss of radioresistant Krt15+ cell in the knockout mice is portrayed in a delayed reproduction of intestinal stem cells.

Number of microcolonies per 50 crypts

50 WT

s K15-KO

t 40 p y r c

30 o f r

e 20 b m u

N 10

0 WT K15-KO

Figure 36: Number of microcolonies per 50 crypts, in wildtype vs. Krt15-/- mice post irradiation.

We defined microcolonies as surviving crypt with more than 10 proliferating cells. The presence of microcolonies can be used as an indicator for tissue survival and vivacity post radiation injury. Krt15-/- mice have significantly fewer microcolonies indicating a decrease in tissue restoration post radiation injury.

5. RESULTS

5.5. Role of Krt15 in cancer initiation. Tumor-initiating capacity is an important feature in intestinal stem cells. Lgr5+ cells, Lrig1+ cells and Krt19+ cells have been described as the cell-of-origin for intestinal tumorigenesis following Apc loss (Asfaha et al., 2015a; Barker et al., 2009; Powell et al., 2012). Also, Bmi1+ cells can form adenomas following induction of a stable form of b-catenin in these cells (Sangiorgi and Capecchi, 2008b). In order to determine if Krt15+ cells can initiate tumor formation in the intestine, we bred Apcfl/fl mice with our Krt15-CrePR1;R26mT/mG mice. Cre recombination was induced using daily injections of 0.5mg RU486 for five consecutive days. The control mice consisted of two WT mice and one Krt15-CrePR1;R26mT/mG who did not receive any RU486 injections. All mice (n=11 mice) were sacrificed 6 months after Cre recombination or at prior time points in concert with severe weight loss.

Krt15-CrePR1;Apcfl/fl;R26mT/mG

RU486

-4 -3 -2 -1 0 150 Time (d)

Figure 37: Timetable depicting experimental setup for the cancer initiation model

Krt15-CrePR1;Apcfl/fl;R26mT/mG mice were injected daily with 0.5mg RU486 for five consecutive days to induce Cre recombination and mice were sacrificed when sick or 150 days following Cre induction.

Gross lesions were observed in all Krt15-CrePR1;Apcfl/fl;R26mT/mG mice but not in control mice. All mice developed lesions in the small intestine, while colon lesions were observed in 37 % of the mice. GFP labeling confirmed that the observed lesions were derived from Krt15+ cells (Figure 28).

5. RESULTS

Krt15-CrePR1;Apcfl/fl;R26mT/mG 5 X Krt15-CrePR1;Apcfl/fl;R26mT/mG 10 X

Figure 38: Representative image of intestinal tumors in Krt15-CrePR1;Apcfl/fl;R26mT/mG mice

The left image depicts a representative intestinal tumor observed in Krt15-CrePR1;Apcfl/fl;R26mT/mG mice with 5x magnification lens. On the right-hand side, we have image, taken with a 10X magnification lens of a GFP+ intestinal tumor observed in Krt15- CrePR1;Apcfl/fl;R26mT/mG mice.

Tumor lesions were assessed by the Molecular pathology and Imaging Core of university of Pennsylvania. Mice developed between 4 to 33 tumors measuring on average 3.17 mm. Most observed lesions were adenomas but invasive adenocarcinomas were occasionally observed (Figure 29). Flat Adenoma Adenocarcinoma

Figure 39: Representative histology of flat adenoma and invasive adeno carcinoma in Krt15-CrePR1;Apcfl/fl;R26mT/mG mice.

The left image depicts a representative flat adenoma observed in Krt15-CrePR1;Apcfl/fl;R26mT/mG mice with 5x magnification lens. On the right-hand side, we have image of an adenocarcinoma, occasionally overserved in Krt15-CrePR1;Apcfl/fl;R26mT/mG mice.

5. RESULTS

These results suggest that Krt15+ crypt cells are tumor-initiating and lesions derived from Krt15+ cells can even progress to invasive adenocarcinoma, which would appear to distinguish such cells from other ISCs.

6. DISCUSSION

Krt15 has been described as a putative epidermal stem cell marker and is thought to play a pivotal role in cell homeostasis and carcinogenesis of the cutaneous epithelium. Krt15+ hair bulge cells are slow cycling long-lasting stem cells, preferentially proliferating in combined expression with Krt19. Recent publication of Krt19 expression in the intestinal epithelium, lead to the theory that both keratins could be used as stem cell markers in the intestine. Our first approach was to prove Krt15 expression in the intestine by directly targeting Krt15+ cells. We examined the small intestine of WT-mice with K15 antibodies using Krt15-/- as a negative control. Although our control mice depicted an absolute Krt15 knockout in the esophagus, we encountered occasional false positive staining at crypt ground. These false positives could be explained by an interaction of K15 antibodies with unknown elements of the Paneth cell population. To confirm these observations, we designed a transgenic mouse model Krt15-crePR1; R26mT/mG, enabling us to label Krt15 expressing cells directly. By repeated Cre induction via RU486, we established a high recombination rate, allowing us to trace the Krt15+ population. This unique mouse model permitted us to determine the fate of any recombined Krt15+ cell and all daughter cells. In a series of experiments on Krt15-crePR1; R26mT/mG mice, we established a robust expression of Krt15 in the small intestine at various locations. The intestinal epithelium can be divided functionally into two compartments: the crypt, containing long-lived stem cells and mainly undifferentiated cells and the villi consisting of highly differentiated short-lived cells. We identified Krt15+ cells in the crypt mainly between the second and fifth position, which is consistent with the location of most ISCs. Nevertheless, a series of differentiated cells expressed Krt15 in the villi which was unexpected since Krt15 expression in the epidermal epithelium is mostly restricted to undifferentiated basal epithelial cells. Vertical differentiation involves the process of progressive terminal differentiation of basal cells in the epidermal epithelium as well as transient amplifying cell in the intestine. Intestinal daughter cells undergo a vertical migration and simultaneous differentiation. Collaborating groups proposed Krt15 regulation by two main putative mechanisms in keratinocytes: one that drives its expression in the basal layer, mediated primarily by FOXM1, and another that induces its expression in the suprabasal layers, involving PKC/AP-1 signaling (Bose et al., 2013). Diverse Krt15 regulation would explain the presence in terminally differentiated epithelial cells. Another possible explanation could be the age at which we sacrificed the mice (6- week-old). Suprabasal Krt15 expression has been reported in squamous epithelium of the

6. DISCUSSION esophagus of neonatal mice and sheep, adult specimens however, presented Krt15 expression confined to the basal layer (Porter et al., 2000; Waseem et al., 1999). Consistent with that theory, intestinal probes taken from adult mice in the lineage tracing experiment did not show any sign of isolated Krt15+ cells in the villi. Therefore, suggesting an alternate role of Krt15 in the adult epithelia compatible with selective expression pattern. As yet, the exact regulation of Krt15 in adult epithelia is not understood. Although 6-week-old mice seemingly expressed Krt15 both the villi and the crypt, proliferative Krt15+ cells were restricted to the crypt. Krt15+ cells could reliably be detected between second fifth crypt position, known to form the stem cell compartment. Consequently, we decided to trace our Krt15+ population to determine their fate and differentiation capability. The mouse model with the inducible Cre-LoxP allowed clear recognition of Krt15+ progeny cells. Robust Cre-recombination was achieved by consecutive Ru486 administration for 5 days on 49 mice, sacrificing them at various time points up to 150 days. After recombination, only Krt15+ progeny cells would express GFP, thus labeling all daughter cells. We quantified Krt15+ crypt analysis, selecting 4 mice per time point (1, 10 and 56 days) and counting crypt cells of 150 intact crypts per specimen. The number of 150 analyzed crypts per time point mitigated the risk of statistical variation. Less than 20% of all crypts contain Krt15+ cells after 56 days, yet they constitute the majority of cells in those crypts. ISC markers are known to label only a small portion of all intestinal crypts, suggesting distinct gene variation of ISCs with otherwise similar cell functions. A stem cell marker reliably labeling all CBC or +4 positioned stem cells has yet to be found. Continued progression of the Krt15+ population marks a long-lasting type of stem cell. Although Krt15+ cells appear to be slow cycling, they proliferate to the extent of labeling entire crypts and villi. We could also establish that Krt15+ progeny cells differentiate into secretory and absorptive cell lines. To prove multipotency in Krt15+ cells we designed an experimental approach transferring in vivo gained crypt into in vitro 3D organoid. We extracted intestinal crypts from Krt15-CrePR1;Rosa26LSL-tdTomato mice and transferred set crypts into a matrigel matrix. Krt15+ crypt bound cells proved capable of creating organoids as well as giving rise to all cell lineages in an in vitro setting. Moreover, the single cell suspension demonstrates that isolated Krt15+ cells can form new organoids, without the support of a stem cell niche. Regrettably, we could not elaborate any interaction between Krt15+ cells and other known stem cell, since there exist no adequate antibodies reliably targeting Lgr5+ CBC stem cells (Yamazaki

6. DISCUSSION et al., 2015). Nonetheless, it would be most relevant to cross our Krt15-crePR1; R26mT/mG mouse with an Lgr5 reporter mice, to demonstrate the implication of Krt15 in CBC cell homeostasis. Krt15+ cells were shown to colocalize with Lgr5+ cells found in the mouse epidermal bulge of hair follicle which acted as stem cells regenerating new hair follicles and maintaining all cell lineages of the follicle (Jaks et al., 2008). Yet another approach could be to cell sort Krt15+ and run a qPCR to establish mRNA expression level of known stem cell markers. These experimental approaches remain however future projects. Similar markers such as Krt19 label a radioresistant slow cycling stem cell, capable of crypt regeneration after irradiation. The position of Krt15+ cells and their slow cycling quiescent nature puts forth the theory that they could have similar properties. As experimental approach for this aim, we chose an injury model irradiating lineage tracing mice at 12 Gy. 12 Gy irradiation was previously established as sufficiently strong, for complete ablation of radiosensitive Lgr5+ CBC stem cells (Asfaha et al., 2015a; van Es et al., 2012). Irradiation damage and the loss of highly proliferate Lgr5+ stem cells severely impact epithelial homeostasis initiating intestinal stem cell division during epithelial repair (May et al., 2008; Yan et al., 2012). Hence, it appears to be the ideal setting to examine the influence of Krt15+ on epithelial homeostasis and injury response. We addressed the question at hand with a cohort of 12 Krt15-crePR1; R26mT/mG mice, irradiating half and using the other half as non-irradiated control group. 3 specimens out of each cohort were then sacrificed after 2 and 5 days, ensuring an adequate number of treated animals for each time point. Irradiation effect was then quantified by crypt analysis via cell counting and crypt length analysis. Results of crypt analysis were highly significant and depicted a radioresistant Krt15+ cell population with highly increased cell division. The first time-point (2 days post irradiation) illustrates a severally damaged epithelial structure, with limited proliferation of Krt15+ cells. After 5 days however, cell division of Krt15+ cells is increased to the point of labeling entire crypts and adjacent villi. Such rapid cell proliferation admirably exhibits the important role of Krt15+ cells on epithelial homeostasis after irradiation injury. Comparable epidermal injury models show an initial upregulation of Krt15 in the basal layer expanding up to suprabasal layers of the follicular epidermis, followed by Krt15 down-regulation and re-establishing basal specific expression (Troy et al., 2011). Presumably, Krt15+ cells are capable of restoring crypt-integrity by replacing ablated radiosensitive stem cells. Krt15+ cells were shown to colocalize with Lgr5+ cells found in the mouse epidermal bulge of hair follicle, which acted as stem cells regenerating new hair follicles

6. DISCUSSION and maintaining all cell lineages of the follicle (Jaks et al., 2008). To prove this assumption, we need to confirm interaction between Krt15+ cells and CBC stem cells, before and after irradiation. Thus, demonstrating the ability of Krt15+ cells to give rise to new CBC stem cells after crypt sterilization via irradiation. Since Lgr5+ CBC cells cannot be visualized by normal immunohistochemistry due to a lack of adequate antibodies, this could be realized by insito Hybridization targeting Lgr5 expression or other known CBC stem cell markers. Yet, this remains an experimental approach for future projects. To verify the impact of Krt15 on epithelial homeostasis, we designed an irradiation assay on Krt15- /- mice comparing them to WT mice. Early systemic loss of Krt15 can commonly be compensated by other keratins and result in significant disruption of epidermal epithelial function. The impact on the intestinal epithelium however, has not yet been identified. Epithelial conditions were quantified 5 days post irradiation, comparing Krt15-/- mice to WT mice. Normal uninjured intestinal epithelium was not disrupted by loss of Krt15, neither was the crypt nor villi structure altered nor morphologically aberrant. However, epithelial repair after irradiation injury appeared segmentally impaired by Krt15 loss. Similarly, epithelial repair in the ileum was significantly impaired by Krt15 deletion, as measured by microcolony assay and crypt-length analysis. Microcolony assay is a commonly used method to quantify epithelial repair after irradiation. It was first established in the 1970s and has been used as a widely accepted experimental tool ever since. To minimize detection-bias, we analyzed 150 intact ileal crypts per specimen repeating the same procedure for jejunal crypts. Furthermore, we evaluated crypt length, as a parameter of successful epithelial repair, using the same n as in the previous approach. Both approaches produced similar significant results suggesting predominant expression of Krt15 in the ileum compared to the jejunum. Presumably the impact of Krt15 on epithelial homeostasis could variate along the length of the small intestine. Graduate expression of Krt15 along the length of the small intestine has not yet been reported, however considering the stretch of the intestinal epithelium, localized expression of any marker could be considered as biological diversity. qPCR analysis of epithelial probes, taken at various sections of the intestine, could directly target this question and display differentiated expression patterns. In any case, the reduced epithelial repair due to Krt15 knockout supports our theory of Krt15 playing a crucial role for epithelial homeostasis. Although substantial research has been done to identify intestinal stem cells, the reliability of stem cell markers in clinical application remains debatable. A possible clinical application in the field

6. DISCUSSION of stem cell research focuses on defining cancer initiating stem cells. Tumor growth is mainly independent from external growth factors, either producing growth factors autonomously or by inadequate overexpression of hormone receptors. Intestinal as well as epidermal stem cells are predominantly regulated by Wnt signaling. The presence of Wnt signaling is one of the driving forces of CBC cell proliferation (Haegebarth and Clevers, 2009). Loss of the tumor suppressor gene APC results in adenomatous transformation of the epithelium, driven by Wnt-signaling and is considered to be the principal cause of colon cancer in humans (Korinek et al., 1997). The role of Wnt-signaling further underlines differences between +4 and CBC stem cells. The Bmi1+ stem cell population was found inherently insensitive to global gain- and loss-of-function modulation compared to Lgr5+ CBC stem cells (Yan et al., 2012). The cancer stem cell theory assigns cancer specific attributes, such as inadequate growth and pluripotency to mutated stem cell forming the origin of cancer genesis. Many studies were conducted to identify the cell-of-origin of intestinal cancers. Apc loss in Lgr5+ cells (Asfaha et al., 2015a; Barker et al., 2009), Lrig1+ cells (Powell et al., 2012, 2014) and Krt19+ cells(Asfaha et al., 2015b) lead to adenoma formation. Adenomas are also observed following induction of stable β-catenin in Bmi1+ cells (Sangiorgi and Capecchi, 2008a). Yet none of these lesions have been reported to progress to adenocarcinoma. Herein, we observed that mice harboring Apc loss in Krt15+ cells develop adenomas as well as occasionally invasive adenocarcinomas. These results suggest that Krt15+ cells are tumor-initiating cells that also have the capacities to progress to more invasive phenotypes. Krt15-CrePR1;Apcfl/fl mice might also represent a new mouse model for intestinal cancer. It would be interesting to analyze cell growth in the context of hormone signaling such as Wnt signaling. One of the most elegant ways to determine Wnt signaling sensibility is by creating 3D entero-organoids. Our next experimental step would consist of sorting Krt15+ cells from Krt15- crePR1; R26mT/mG mice via FACS and transferring them to a matrigel in vitro structure, thus establishing an elaborate entero-genesis assay. The in vitro assay would enable us to determine the potency of isolated Krt15+ cells while depicting relevant implications of known growth signaling pathways such as Wnt-signaling and Notch-signaling. By variating external growth factors or niche signaling supplemented by Paneth cells, we could determine dependency of Krt15+ cells on external parameters. We assume that Krt15+ cells compete with different stem cells and may play a subordinate role in normal crypt homeostasis. The role of Krt15+ cells could rather be explained by that of quiescent resilient stem cells vital for restoring homeostasis after injury. It is certainly

6. DISCUSSION possible that Krt15 may only mark a subset of quiescent stem cells and our results do not exclude overlapping expression with populations identified by other intestinal stem cell markers.

7. CONCLUSION

Undifferentiated epidermal cells selectively express Krt15 as dominating intermediate filaments subsequently evoking a link between the level of cell differentiation and cytoskeletal components (Bose et al., 2013). As K15 is one of the basal-specific cytoskeletal proteins, its presence in intestinal crypts may indicate the necessity of structural integrity for developing cells. In contrast to the squamous epithelium, the involvement of cytoskeletal proteins in proliferation and homeostasis in the intestine has scarcely been identified. We proved the presence of K15 in small intestine and demonstrated a robust expression of Krt15 in intestinal crypts and villi. Furthermore, we showed that Krt15, a known stem cell and differentiation marker of the epidermal epithelium, is expressed in long-lived, radioresistant, proliferating epithelial cells in the small intestine. Krt15 expressing+ cells survive well beyond 150 days and show lineage tracing capacity even after irradiation. We could also prove multipotency in 3D organotypic organoids. In which Krt15+ cells could recreate crypt like formation in vitro even from a single isolated Krt15+ cell. Due to the location of Krt15+ cells at the bottom of crypt we expect a certain overlap between the Lgr5+ and Krt15+ population, however both cell types are functionally distinct by their relative radioresistance. The direct interaction of Krt15+ cells with the Lgr5+ CBC cell population is yet to be analyzed. The lineage tracing assays indicated that Krt15+ progeny cells can differentiate into all known cell types with continuously active Krt15 promoter. Moreover, loss of Apc the Krt15+ cells seems to trigger cell degeneration and cancer initiation. Krt15-CrePR1;Apcfl/fl mice all developed adenomas as well as occasional progress to adenocarcinoma. Since it is believed that most intestinal carcinomas origin from mutated stem cell, a more in depth knowledge of specific markers becomes crucial. We propose that Krt15 marks a type of long-lived radioresistant stem cells, inherently influencing crypt homeostasis and regenerative capacity

8. ACKNOWLEDGEMENT

I am grateful for the continued support and scientific training by Prof. Dr. Markus M. Lerch as well as all members of Lerch lab (University of Greifswald). Special thanks goes to Dr. Anil K. Rustgi and PhD Veronique Giroux for the substantial support. Furthermore, I am grateful to Dr. Christopher Lengner and Dr. Ning Li as well as members of the Rustgi lab for discussions and comments in the manuscript. We thank the Molecular Pathology and Imaging Core, Human Microbial Analytic Depository Core, Cell Culture and iPS Core, Genetic and Modified Mouse Core and FACS/sorting Core facilities (University of Pennsylvania). We thank Dr. George Cotsarelis (University of Pennsylvania) for Krt15-CrePR1 and Krt15-/- mice and Dr. Christopher Lengner (University of Pennsylvania) for RosaLSL-tdTomato mice. This work was supported by NCI P01-CA098101 (VG, AJKS, AKR), NIH/NIDDK P30-DK050306 Center of Molecular Studies in Digestive and Liver Diseases, American Cancer Society (AKR), Fonds de recherche en santé du Québec P-Giroux-27692 and P-Giroux-31601 (VG), NIH NIDDK K01-DK100485 (KEH) and Crohn’s and Colitis Foundation Career Development Award (KEH).

8.1. Author contribution Veronique Giroux (VG), Katheryn E. Hamilton (KEH), Anil K. Rustgi (AKR) and Julien Stephan (JS) designed the study. Veronique Giroux (VG) performed the 3D organoid experiments as well as the cancer initiation model, while all other experiments were performed by Julien Stephan (JS). Ben Rhoades (BR) maintained the mouse colony.

9. BIBLIOGRAPHY

9. BIBLIOGRAPHY Abremski, K., and Hoess, R. (1984). Bacteriophage P1 site-specific recombination. Purification and properties of the Cre recombinase protein. J. Biol. Chem. 259, 1509–1514.

Al Alam, D., Danopoulos, S., Schall, K., Sala, F.G., Almohazey, D., Fernandez, G.E., Georgia, S., Frey, M.R., Ford, H.R., Grikscheit, T., et al. (2015). Fibroblast growth factor 10 alters the balance between goblet and Paneth cells in the adult mouse small intestine. Am. J. Physiol. - Gastrointest. Liver Physiol. 308, G678–G690.

Asfaha, S., Hayakawa, Y., Muley, A., Stokes, S., Graham, T.A., Ericksen, R.E., Westphalen, C.B., von Burstin, J., Mastracci, T.L., Worthley, D.L., et al. (2015a). Krt19+/Lgr5− Cells Are Radioresistant Cancer-Initiating Stem Cells in the Colon and Intestine. Cell Stem Cell 16, 627– 638.

Asfaha, S., Hayakawa, Y., Muley, A., Stokes, S., Graham, T.A., Ericksen, R.E., Westphalen, C.B., von Burstin, J., Mastracci, T.L., Worthley, D.L., et al. (2015b). Krt19(+)/Lgr5(-) Cells Are Radioresistant Cancer-Initiating Stem Cells in the Colon and Intestine. Cell Stem Cell 16, 627– 638.

Barker, N. (2014). Adult intestinal stem cells: critical drivers of epithelial homeostasis and regeneration. Nat. Rev. Mol. Cell Biol. 15, 19–33.

Barker, N., van Es, J.H., Kuipers, J., Kujala, P., van den Born, M., Cozijnsen, M., Haegebarth, A., Korving, J., Begthel, H., Peters, P.J., et al. (2007). Identification of stem cells in small intestine and colon by marker gene Lgr5. Nature 449, 1003–1007.

Barker, N., Ridgway, R.A., van Es, J.H., van de Wetering, M., Begthel, H., van den Born, M., Danenberg, E., Clarke, A.R., Sansom, O.J., and Clevers, H. (2009). Crypt stem cells as the cells- of-origin of intestinal cancer. Nature 457, 608–611.

Booth, C., Booth, D., Williamson, S., Demchyshyn, L.L., and Potten, C.S. (2004). Teduglutide ([Gly2]GLP-2) protects small intestinal stem cells from radiation damage. Cell Prolif. 37, 385– 400.

Booth, C., Tudor, G.L., Katz, B.P., and MacVittie, T.J. (2015). The Delayed Effects of Acute Radiation Syndrome: Evidence of long-term functional changes in the clonogenic cells of the small intestine. Health Phys. 109, 399–413.

Bose, A., Teh, M.-T., Mackenzie, I.C., and Waseem, A. (2013). Keratin K15 as a Biomarker of Epidermal Stem Cells. Int. J. Mol. Sci. 14, 19385–19398.

Bragulla, H.H., and Homberger, D.G. (2009). Structure and functions of keratin proteins in simple, stratified, keratinized and cornified epithelia. J. Anat. 214, 516–559.

Carulli, A.J., Keeley, T.M., Demitrack, E.S., Chung, J., Maillard, I., and Samuelson, L.C. (2015). Notch receptor regulation of intestinal stem cell homeostasis and crypt regeneration. Dev. Biol. 402, 98–108.

9. BIBLIOGRAPHY

Cooper, G.M. (2000). The Cell (Sinauer Associates).

Dick, J.E. (2009). Looking ahead in cancer stem cell research. Nat. Biotechnol. 27, 44–46.

Driskell, I., Oeztuerk-Winder, F., Humphreys, P., and Frye, M. (2015). Genetically Induced Cell Death in Bulge Stem Cells Reveals Their Redundancy for Hair and Epidermal Regeneration. Stem Cells Dayt. Ohio 33, 988–998.

Eriksson, J.E., Dechat, T., Grin, B., Helfand, B., Mendez, M., Pallari, H.-M., and Goldman, R.D. (2009). Introducing intermediate filaments: from discovery to disease. J. Clin. Invest. 119, 1763– 1771. van Es, J.H., Sato, T., van de Wetering, M., Lyubimova, A., Yee Nee, A.N., Gregorieff, A., Sasaki, N., Zeinstra, L., van den Born, M., Korving, J., et al. (2012). Dll1+ secretory progenitor cells revert to stem cells upon crypt damage. Nat. Cell Biol. 14, 1099–1104.

Espersen, M.L.M., Olsen, J., Linnemann, D., Høgdall, E., and Troelsen, J.T. (2015). Clinical Implications of Intestinal Stem Cell Markers in Colorectal Cancer. Clin. Colorectal Cancer 14, 63– 71.

Fafilek, B., Krausova, M., Vojtechova, M., Pospichalova, V., Tumova, L., Sloncova, E., Huranova, M., Stancikova, J., Hlavata, A., Svec, J., et al. (2013). Troy, a tumor necrosis factor receptor family member, interacts with lgr5 to inhibit wnt signaling in intestinal stem cells. Gastroenterology 144, 381–391.

Fan, X.-S., Wu, H.-Y., Yu, H.-P., Zhou, Q., Zhang, Y.-F., and Huang, Q. (2010). Expression of Lgr5 in human colorectal carcinogenesis and its potential correlation with beta-catenin. Int. J. Colorectal Dis. 25, 583–590.

Gaskins, H.R., Croix, J.A., Nakamura, N., and Nava, G.M. (2008). Impact of the Intestinal Microbiota on the Development of Mucosal Defense. Clin. Infect. Dis. 46, S80–S86.

Gierut, J.J., Jacks, T.E., and Haigis, K.M. (2014). Strategies to Achieve Conditional Gene Mutation in Mice. Cold Spring Harb. Protoc. 2014, 339–349. van Gijlswijk, R.P., Wiegant, J., Vervenne, R., Lasan, R., Tanke, H.J., and Raap, A.K. (1996). Horseradish peroxidase-labeled oligonucleotides and fluorescent tyramides for rapid detection of chromosome-specific repeat sequences. Cytogenet. Cell Genet. 75, 258–262.

Giroux, V., Lento, A.A., Islam, M., Pitarresi, J.R., Kharbanda, A., Hamilton, K.E., Whelan, K.A., Long, A., Rhoades, B., Tang, Q., et al. (2017). Long-lived keratin 15+ esophageal progenitor cells contribute to homeostasis and regeneration. J. Clin. Invest. 127.

Giroux, V., Stephan, J., Chatterji, P., Rhoades, B., Wileyto, E.P., Klein-Szanto, A.J., Lengner, C.J., Hamilton, K.E., and Rustgi, A.K. (2018). Mouse Intestinal Krt15+ Crypt Cells Are Radio- Resistant and Tumor Initiating. Stem Cell Rep.

9. BIBLIOGRAPHY

Haegebarth, A., and Clevers, H. (2009). Wnt Signaling, Lgr5, and Stem Cells in the Intestine and Skin. Am. J. Pathol. 174, 715–721.

Hamilton, K.E., Crissey, M.A.S., Lynch, J.P., and Rustgi, A.K. (2015). Culturing Adult Stem Cells from Mouse Small Intestinal Crypts. Cold Spring Harb. Protoc. 2015, pdb.prot078303.

Hao, H.-X., Xie, Y., Zhang, Y., Charlat, O., Oster, E., Avello, M., Lei, H., Mickanin, C., Liu, D., Ruffner, H., et al. (2012). ZNRF3 promotes Wnt receptor turnover in an R-spondin-sensitive manner. Nature 485, 195–200.

He, J., Stewart, K., Kinnell, H.L., Anderson, R.A., and Childs, A.J. (2013). A developmental stage- specific switch from DAZL to BOLL occurs during fetal oogenesis in humans, but not mice. PloS One 8, e73996.

He, X.C., Yin, T., Grindley, J.C., Tian, Q., Sato, T., Tao, W.A., Dirisina, R., Porter-Westpfahl, K.S., Hembree, M., Johnson, T., et al. (2007). PTEN-deficient intestinal stem cells initiate intestinal polyposis. Nat. Genet. 39, 189–198.

Helander, H.F., and Fändriks, L. (2014). Surface area of the digestive tract - revisited. Scand. J. Gastroenterol. 49, 681–689.

Hsu, S.M., Raine, L., and Fanger, H. (1981). Use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase techniques: a comparison between ABC and unlabeled antibody (PAP) procedures. J. Histochem. Cytochem. Off. J. Histochem. Soc. 29, 577–580.

Inoue, K., Aoi, N., Sato, T., Yamauchi, Y., Suga, H., Eto, H., Kato, H., Araki, J., and Yoshimura, K. (2009). Differential expression of stem-cell-associated markers in human hair follicle epithelial cells. Lab. Investig. J. Tech. Methods Pathol. 89, 844–856.

Ito, M., Liu, Y., Yang, Z., Nguyen, J., Liang, F., Morris, R.J., and Cotsarelis, G. (2005). Stem cells in the hair follicle bulge contribute to wound repair but not to homeostasis of the epidermis. Nat. Med. 11, 1351–1354.

Jaks, V., Barker, N., Kasper, M., van Es, J.H., Snippert, H.J., Clevers, H., and Toftgård, R. (2008). Lgr5 marks cycling, yet long-lived, hair follicle stem cells. Nat. Genet. 40, 1291–1299.

Jemal, A., Bray, F., Center, M.M., Ferlay, J., Ward, E., and Forman, D. (2011). Global cancer statistics. CA. Cancer J. Clin. 61, 69–90.

Jones, J.W., Tudor, G., Li, F., Tong, Y., Katz, B., Farese, A.M., MacVittie, T.J., Booth, C., and Kane, M.A. (2015). Citrulline as a biomarker in the murine total-body irradiation model: correlation of circulating and tissue citrulline to small intestine epithelial histopathology. Health Phys. 109, 452–465.

Kalabis, J., Oyama, K., Okawa, T., Nakagawa, H., Michaylira, C.Z., Stairs, D.B., Figueiredo, J.- L., Mahmood, U., Diehl, J.A., Herlyn, M., et al. (2008). A subpopulation of mouse esophageal basal cells has properties of stem cells with the capacity for self-renewal and lineage specification.

9. BIBLIOGRAPHY

Kantara, C., Moya, S.M., Houchen, C.W., Umar, S., Ullrich, R.L., Singh, P., and Carney, D.H. (2015). Novel Regenerative Peptide TP508 Mitigates Radiation-Induced Gastrointestinal Damage By Activating Stem Cells and Preserving Crypt Integrity. Lab. Investig. J. Tech. Methods Pathol. 95, 1222–1233.

Kloepper, J.E., Tiede, S., Brinckmann, J., Reinhardt, D.P., Meyer, W., Faessler, R., and Paus, R. (2008). Immunophenotyping of the human bulge region: the quest to define useful in situ markers for human epithelial hair follicle stem cells and their niche. Exp. Dermatol. 17, 592–609.

Korinek, V., Barker, N., Morin, P.J., van Wichen, D., de Weger, R., Kinzler, K.W., Vogelstein, B., and Clevers, H. (1997). Constitutive transcriptional activation by a beta-catenin-Tcf complex in APC-/- colon carcinoma. Science 275, 1784–1787.

Kuhnert, F., Davis, C.R., Wang, H.-T., Chu, P., Lee, M., Yuan, J., Nusse, R., and Kuo, C.J. (2004). Essential requirement for Wnt signaling in proliferation of adult small intestine and colon revealed by adenoviral expression of Dickkopf-1. Proc. Natl. Acad. Sci. U. S. A. 101, 266–271.

Kuraguchi, M., Wang, X.-P., Bronson, R.T., Rothenberg, R., Ohene-Baah, N.Y., Lund, J.J., Kucherlapati, M., Maas, R.L., and Kucherlapati, R. (2006). Adenomatous Polyposis Coli (APC) Is Required for Normal Development of Skin and Thymus. PLOS Genet. 2, e146. de Lau, W., Barker, N., Low, T.Y., Koo, B.-K., Li, V.S.W., Teunissen, H., Kujala, P., Haegebarth, A., Peters, P.J., van de Wetering, M., et al. (2011). Lgr5 homologues associate with Wnt receptors and mediate R-spondin signalling. Nature 476, 293–297.

Li, D., Tang, H., Fan, J., Yan, D., Zhou, C., Li, S., Wang, X., and Peng, Z. (2010). Expression level of Bmi-1 oncoprotein is associated with progression and prognosis in colon cancer. J. Cancer Res. Clin. Oncol. 136, 997–1006.

Li, S., Park, H., Trempus, C.S., Gordon, D., Liu, Y., Cotsarelis, G., and Morris, R.J. (2013). A keratin 15 containing stem cell population from the hair follicle contributes to squamous papilloma development in the mouse. Mol. Carcinog. 52, 751–759.

Liu, G., Amin, S., Okuhama, N.N., Liao, G., and Mingle, L.A. (2006). A quantitative evaluation of peroxidase inhibitors for tyramide signal amplification mediated cytochemistry and histochemistry. Histochem. Cell Biol. 126, 283–291.

Liu, Y., Lyle, S., Yang, Z., and Cotsarelis, G. (2003). Keratin 15 Promoter Targets Putative Epithelial Stem Cells in the Hair Follicle Bulge. J. Invest. Dermatol. 121, 963–968.

Lozano, R., Naghavi, M., Foreman, K., Lim, S., Shibuya, K., Aboyans, V., Abraham, J., Adair, T., Aggarwal, R., Ahn, S.Y., et al. (2012). Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 380, 2095–2128.

Ma, D.R., Yang, E.N., and Lee, S.T. (2004). A review: the location, molecular characterisation and multipotency of hair follicle epidermal stem cells. Ann. Acad. Med. Singapore 33, 784–788.

9. BIBLIOGRAPHY

May, R., Riehl, T.E., Hunt, C., Sureban, S.M., Anant, S., and Houchen, C.W. (2008). Identification of a Novel Putative Gastrointestinal Stem Cell and Adenoma Stem Cell Marker, Doublecortin and CaM Kinase-Like-1, Following Radiation Injury and in Adenomatous Polyposis Coli/Multiple Intestinal Neoplasia Mice. STEM CELLS 26, 630–637.

Michel, M., Török, N., Godbout, M.J., Lussier, M., Gaudreau, P., Royal, A., and Germain, L. (1996). Keratin 19 as a biochemical marker of skin stem cells in vivo and in vitro: keratin 19 expressing cells are differentially localized in function of anatomic sites, and their number varies with donor age and culture stage. J. Cell Sci. 109 ( Pt 5), 1017–1028.

Montgomery, R.K., Carlone, D.L., Richmond, C.A., Farilla, L., Kranendonk, M.E.G., Henderson, D.E., Baffour-Awuah, N.Y., Ambruzs, D.M., Fogli, L.K., Algra, S., et al. (2011). Mouse telomerase reverse transcriptase (mTert) expression marks slowly cycling intestinal stem cells. Proc. Natl. Acad. Sci. 108, 179–184.

Morris, R.J., Liu, Y., Marles, L., Yang, Z., Trempus, C., Li, S., Lin, J.S., Sawicki, J.A., and Cotsarelis, G. (2004). Capturing and profiling adult hair follicle stem cells. Nat. Biotechnol. 22, 411–417.

Mouse Genome Sequencing Consortium, Waterston, R.H., Lindblad-Toh, K., Birney, E., Rogers, J., Abril, J.F., Agarwal, P., Agarwala, R., Ainscough, R., Alexandersson, M., et al. (2002). Initial sequencing and comparative analysis of the mouse genome. Nature 420, 520–562.

Muñoz, J., Stange, D.E., Schepers, A.G., van de Wetering, M., Koo, B.-K., Itzkovitz, S., Volckmann, R., Kung, K.S., Koster, J., Radulescu, S., et al. (2012). The Lgr5 intestinal stem cell signature: robust expression of proposed quiescent “+4” cell markers. EMBO J. 31, 3079–3091.

Otsuka, K., Hamada, N., Magae, J., Matsumoto, H., Hoshi, Y., and Iwasaki, T. (2013). Ionizing radiation leads to the replacement and de novo production of colonic Lgr5(+) stem cells. Radiat. Res. 179, 637–646.

Pellegrinet, L., Rodilla, V., Liu, Z., Chen, S., Koch, U., Espinosa, L., Kaestner, K.H., Kopan, R., Lewis, J., and Radtke, F. (2011). Dll1- and dll4-mediated notch signaling are required for homeostasis of intestinal stem cells. Gastroenterology 140, 1230-1240.e1-7.

Porter, R.M., Lunny, D.P., Ogden, P.H., Morley, S.M., McLean, W.H., Evans, A., Harrison, D.L., Rugg, E.L., and Lane, E.B. (2000). K15 expression implies lateral differentiation within stratified epithelial basal cells. Lab. Investig. J. Tech. Methods Pathol. 80, 1701–1710.

Potten, C.S., Merritt, A., Hickman, J., Hall, P., and Faranda, A. (1994). Characterization of radiation-induced apoptosis in the small intestine and its biological implications. Int. J. Radiat. Biol. 65, 71–78.

Powell, A.E., Wang, Y., Li, Y., Poulin, E.J., Means, A.L., Washington, M.K., Higginbotham, J.N., Juchheim, A., Prasad, N., Levy, S.E., et al. (2012). The Pan-ErbB Negative Regulator Lrig1 Is an Intestinal Stem Cell Marker that Functions as a Tumor Suppressor. Cell 149, 146–158.

9. BIBLIOGRAPHY

Powell, A.E., Vlacich, G., Zhao, Z.-Y., McKinley, E.T., Washington, M.K., Manning, H.C., and Coffey, R.J. (2014). Inducible loss of one Apc allele in Lrig1-expressing progenitor cells results in multiple distal colonic tumors with features of familial adenomatous polyposis. Am. J. Physiol. Gastrointest. Liver Physiol. 307, G16-23.

Reinisch, C., Kandutsch, S., Uthman, A., and Pammer, J. (2006). BMI-1: a protein expressed in stem cells, specialized cells and tumors of the gastrointestinal tract. Histol. Histopathol. 21, 1143– 1149.

Roche, K.C., Gracz, A.D., Liu, X.F., Newton, V., Akiyama, H., and Magness, S.T. (2015). SOX9 Maintains Reserve Stem Cells and Preserves Radioresistance in Mouse Small Intestine. Gastroenterology 149, 1553-1563.e10.

Roe, D.J., Alberts, D.S., Wargovich, M.J., Bostick, R.M., Garewal, H.S., Einspahr, J., Fosdick, L., Ramsey, L., Woods, C., and McGee, D.L. (1996). Reproducibility of the measurement of colonic proliferation using bromodeoxyuridine across laboratories. Cancer Epidemiol. Biomark. Prev. Publ. Am. Assoc. Cancer Res. Cosponsored Am. Soc. Prev. Oncol. 5, 349–353.

Sahlgren, C.M., Mikhailov, A., Vaittinen, S., Pallari, H.-M., Kalimo, H., Pant, H.C., and Eriksson, J.E. (2003). Cdk5 regulates the organization of Nestin and its association with p35. Mol. Cell. Biol. 23, 5090–5106.

Sailaja, B.S., He, X.C., and Li, L. (2016). The regulatory niche of intestinal stem cells. J. Physiol. 594, 4827–4836.

Sangiorgi, E., and Capecchi, M.R. (2008a). Bmi1 is expressed in vivo in intestinal stem cells. Nat. Genet. 40, 915–920.

Sangiorgi, E., and Capecchi, M.R. (2008b). Bmi1 is expressed in vivo in intestinal stem cells. Nat. Genet. 40, 915–920.

Schofield, R. (1978). The relationship between the spleen colony-forming cell and the haemopoietic stem cell. Blood Cells 4, 7–25.

Scholzen, T., and Gerdes, J. (2000). The Ki-67 protein: from the known and the unknown. J. Cell. Physiol. 182, 311–322.

Schuijers, J., van der Flier, L.G., van Es, J., and Clevers, H. (2014). Robust Cre-Mediated Recombination in Small Intestinal Stem Cells Utilizing the Olfm4 Locus. Stem Cell Rep. 3, 234– 241.

Schuijers, J., Junker, J.P., Mokry, M., Hatzis, P., Koo, B.-K., Sasselli, V., van der Flier, L.G., Cuppen, E., van Oudenaarden, A., and Clevers, H. (2015). Ascl2 acts as an R-spondin/Wnt- responsive switch to control stemness in intestinal crypts. Cell Stem Cell 16, 158–170.

Sei, Y., Lu, X., Liou, A., Zhao, X., and Wank, S.A. (2011). A stem cell marker-expressing subset of enteroendocrine cells resides at the crypt base in the small intestine. Am. J. Physiol. Gastrointest. Liver Physiol. 300, G345-356.

9. BIBLIOGRAPHY

Seshagiri, S., Stawiski, E.W., Durinck, S., Modrusan, Z., Storm, E.E., Conboy, C.B., Chaudhuri, S., Guan, Y., Janakiraman, V., Jaiswal, B.S., et al. (2012). Recurrent R-spondin fusions in colon cancer. Nature 488, 660–664.

Snippert, H.J., van Es, J.H., van den Born, M., Begthel, H., Stange, D.E., Barker, N., and Clevers, H. (2009). Prominin-1/CD133 marks stem cells and early progenitors in mouse small intestine. Gastroenterology 136, 2187-2194.e1.

Szeverenyi, I., Cassidy, A.J., Chung, C.W., Lee, B.T.K., Common, J.E.A., Ogg, S.C., Chen, H., Sim, S.Y., Goh, W.L.P., Ng, K.W., et al. (2008). The Human Intermediate Filament Database: comprehensive information on a gene family involved in many human diseases. Hum. Mutat. 29, 351–360.

Takahashi, H., Ishii, H., Nishida, N., Takemasa, I., Mizushima, T., Ikeda, M., Yokobori, T., Mimori, K., Yamamoto, H., Sekimoto, M., et al. (2011). Significance of Lgr5+ve Cancer Stem Cells in the Colon and Rectum. Ann. Surg. Oncol. 18, 1166–1174.

Takeda, K., Kinoshita, I., Shimizu, Y., Matsuno, Y., Shichinohe, T., and Dosaka-Akita, H. (2011). Expression of LGR5, an intestinal stem cell marker, during each stage of colorectal tumorigenesis. Anticancer Res. 31, 263–270.

Tan, D.W.-M., and Barker, N. (2014). Chapter Three - Intestinal Stem Cells and Their Defining Niche. In Current Topics in Developmental Biology, M. Rendl, ed. (Academic Press), pp. 77–107.

Tateishi, K., Ohta, M., Kanai, F., Guleng, B., Tanaka, Y., Asaoka, Y., Tada, M., Seto, M., Jazag, A., Lianjie, L., et al. (2006). Dysregulated expression of stem cell factor Bmi1 in precancerous lesions of the gastrointestinal tract. Clin. Cancer Res. Off. J. Am. Assoc. Cancer Res. 12, 6960– 6966.

Troy, T.-C., Arabzadeh, A., and Turksen, K. (2011). Re-Assessing K15 as an Epidermal Stem Cell Marker. Stem Cell Rev. Rep. 7, 927–934.

Vries, R.G.J., Huch, M., and Clevers, H. (2010). Stem cells and cancer of the stomach and intestine. Mol. Oncol. 4, 373–384.

Waseem, A., Dogan, B., Tidman, N., Alam, Y., Purkis, P., Jackson, S., Lalli, A., Machesney, M., and Leigh, I.M. (1999). Keratin 15 expression in stratified epithelia: downregulation in activated keratinocytes. J. Invest. Dermatol. 112, 362–369.

Withers, H.R., and Elkind, M.M. (1970). Microcolony survival assay for cells of mouse intestinal mucosa exposed to radiation. Int. J. Radiat. Biol. Relat. Stud. Phys. Chem. Med. 17, 261–267.

Wong, V.W.Y., Stange, D.E., Page, M.E., Buczacki, S., Wabik, A., Itami, S., van de Wetering, M., Poulsom, R., Wright, N.A., Trotter, M.W.B., et al. (2012). Lrig1 controls intestinal stem-cell homeostasis by negative regulation of ErbB signalling. Nat. Cell Biol. 14, 401–408.

Yamazaki, M., Kato, A., Zaitsu, Y., Watanabe, T., Iimori, M., Funahashi, S., Kitao, H., Saeki, H., Oki, E., and Suzuki, M. (2015). Intensive Immunofluorescence Staining Methods for Low

9. BIBLIOGRAPHY

Expression Protein: Detection of Intestinal Stem Cell Marker LGR5. Acta Histochem. Cytochem. 48, 159–164.

Yan, K.S., Chia, L.A., Li, X., Ootani, A., Su, J., Lee, J.Y., Su, N., Luo, Y., Heilshorn, S.C., Amieva, M.R., 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.

10. ORIGINAL PAPER

Mouse Intestinal Krt15+ Crypt Cells Are Radio-Resistant and Tumor Initiating

Giroux, V., Stephan, J., Chatterji, P., Rhoades, B., Wileyto, E.P., Klein-Szanto, A.J., Lengner, C.J., Hamilton, K.E., and Rustgi, A.K. (2018).

Stem Cell Reports: 2018 Jun 5; 10(6): 1947–1958.

10. ORIGINAL PAPER

Stem Cell Reports Article

Mouse Intestinal Krt15+ Crypt Cells Are Radio-Resistant and Tumor Initiating

Ve´ronique Giroux,1,2 Julien Stephan,1,2 Priya Chatterji,1,2 Ben Rhoades,1,2 E. Paul Wileyto,3 Andres J. Klein-Szanto,4 Christopher J. Lengner,5 Kathryn E. Hamilton,1,2,6 and Anil K. Rustgi1,2,7,* 1Division of Gastroenterology, Department of Medicine, University of Pennsylvania, Perelman School of Medicine, 951 BRBII/III, 421 Curie Boulevard, Philadelphia, PA 19104, USA 2Abramson Cancer Center, University of Pennsylvania, Philadelphia, PA 19104, USA 3Department of Biostatistics and Epidemiology, University of Pennsylvania, Philadelphia, PA 19104, USA 4Department of Pathology and Cancer Biology Program, Fox Chase Cancer Center, Philadelphia, PA 19111, USA 5Department of Biomedical Sciences, School of Veterinary Medicine, Institute for Regenerative Medicine, University of Pennsylvania, Philadelphia, PA 19104, USA 6Department of Pediatrics, Division of Gastroenterology, Hepatology and Nutrition, Children’s Hospital of Philadelphia, Philadelphia, PA 19104, USA 7Department of Genetics, University of Pennsylvania, Philadelphia, PA 19104, USA *Correspondence: [email protected] https://doi.org/10.1016/j.stemcr.2018.04.022

SUMMARY

Two principal stem cell pools orchestrate the rapid cell turnover in the intestinal epithelium. Rapidly cycling Lgr5+ stem cells are intercalated between the Paneth cells at the crypt base (CBCs) and injury-resistant reserve stem cells reside above the crypt base. The intermediate ﬁlament Keratin 15 (Krt15) marks either stem cells or long-lived progenitor cells that contribute to tissue repair in the hair follicle or the esophageal epithelium. Herein, we demonstrate that Krt15 labels long-lived and multipotent cells in the small intestinal crypt by lineage tracing. Krt15+ crypt cells display self-renewal potential in vivo and in 3D organoid cultures. Krt15+ crypt cells are resistant to high- dose radiation and contribute to epithelial regeneration following injury. Notably, loss of the tumor suppressor Apc in Krt15+ cells leads to adenoma and adenocarcinoma formation. These results indicate that Krt15 marks long-lived, multipotent, and injury-resistant crypt cells that may function as a cell of origin in intestinal cancer.

INTRODUCTION eth and enteroendocrine cells, located in the crypt and express Lgr5 (Buczacki et al., 2013). Subsequent work Epithelial stem cells are multipotent cells with self- showed the label-retaining secretory precursor cells to be renewal capacity that ensure normal epithelial renewal a distinct population from the reserve ISCs labeled by and tissue regeneration in response to injury (e.g., radia- CreER knockin reporters (Li et al., 2016). tion). The intestinal epithelium is highly proliferative, be- While a body of work has illuminated the distinct ing renewed every few days in the mouse (Bjerknes and nature of these two populations, certain controversies Cheng, 1999; Wong et al., 1999; Barker et al., 2012). persist. For example, in contrast to Bmi1-CreER+ cells, Two main intestinal stem cell (ISC) pools orchestrate Bmi1-GFP+ cells may represent an enteroendocrine pro- maintenance of the intestinal homeostatic epithelium genitor cell population (Jadhav et al., 2017). Furthermore, (Beumer and Clevers, 2016). First, an actively proliferative the heterogeneity of these populations makes interpreta- Lgr5+ crypt base columnar cell population (CBC) is driven tion of genetic labeling challenging at times. For example, by the activity of the canonical Wnt pathway (Barker the RNA binding protein Mex3a marks a subpopulation et al., 2007; Kretzschmar and Clevers, 2017). CBCs are of Lgr5+ cells displaying characteristics consistent with also characterized by the expression of Ascl2, Olfm4 (van reserve-like stem cells (Barriga et al., 2017). Other alleles der Flier et al., 2009), and Smoc2 (Munoz et al., 2012). Sec- can broadly mark several cell types; for example, Lrig1 ond, a slower cycling reserve crypt stem cell population is marks Lgr5+ cells (Wong et al., 2012) and reserve ISCs (Po- located around the +4 position above the crypt base and well et al., 2012). However, the populations marked by lacks regulation by the canonical WNT signaling pathway Lrig1 can vary greatly depending on whether the readout (Sangiorgi and Capecchi, 2008). Speciﬁcally, reserve ISCs is endogenous mRNA, protein (which may be antibody are marked by CreER insertions into the Bmi1 (Sangiorgi dependent), or reporter alleles (Poulin et al., 2014; Powell and Capecchi, 2008) or Hopx loci (Takeda et al., 2011), et al., 2012; Wong et al., 2012). The Sox9-CreER allele as well as by a Tert-CreER transgene mouse (Montgomery also marks reserve ISCs and CBCs (Roche et al., 2015). et al., 2011). Reserve ISCs were originally associated The transcripts of certain reserve stem cell markers are with label-retention capacities (Potten et al., 1978). The expressed in other crypt cells, notably CBCs, thereby identity and function of intestinal label-retaining cells complicating analysis (Li et al., 2014; Munoz et al., (LRCs) remain to be fully understood, but recent work 2012; Grun et al., 2015). Nevertheless, single-cell proﬁling shows that intestinal LRCs are secretory precursors of Pan- has revealed that Bmi1-CreER+ cells and Hopx-CreER+ cells

Stem Cell Reports Vol. 10 1947–1958 June 5, 2018 ª 2018 The Authors. 1947 This is an open access article under the CC BY-NC-NDj licensej (http://creativecommons.org/licenses/by-nc-nd/4.0/j j ).

10. ORIGINAL PAPER

are transcriptionally distinct from Lgr5-CreER + cells and RESULTS generate CBCs under homeostatic conditions ( Li et al., 2014 ). Also, it has been demonstrated that Bmi1-CreER- Krt15 Marks Proliferating Cells in the Small Intestinal labeled cells could replenish Lgr5+ stem cell population af- Crypt ter diphtheria toxin (DT)-mediated ablation ( Tian et al., Krt15 + stem/progenitor cells were described originally in 2011 ). Hopx-CreER -labeled cells could also give rise to the bulge of the hair follicle ( Morris et al., 2004 ) and, CBCs ( Takeda et al., 2011 ). more recently, within the basal compartment of the esoph- Interestingly, Lgr5+ cells are sensitive to DNA damage ageal squamous epithelium ( Giroux et al., 2017 ). Taken and largely ablated with high-dose irradiation ( Yan et al., together, this suggests an important role for Krt15+ cells 2012; Hua et al., 2012; Metcalfe et al., 2014; Tao et al., in the maintenance of squamous epithelia and append- 2015 ), whereas Bmi1-CreER cells ( Yan et al., 2012 ), Hopx- ages. In contrast to the multi-layered squamous epithelium CreER cells ( Yousefi et al., 2016 ), and Lrig1-CreER cells ( Po- in the skin and esophagus, a single layer of the columnar well et al., 2012 ) are resistant to high-dose radiation injury. epithelium is present in the mammary gland, pancreas, Following radiation, reserve ISCs can give rise to CBCs and intestine. We hypothesized that Krt15+ cells in the (Montgomery et al., 2011; Yan et al., 2012; Yousefi et al., small intestine might annotate a unique subpopulation 2016 ). Although Lgr5+ cells are sensitive to injury, ablation of cells with properties that contribute to tissue mainte- of Lgr5+ cells concomitant with or following radiation nance and regeneration. results in failed regeneration, suggesting that de novo First, we noted that endogenous K15 protein is detected generation of new Lgr5+ cells is required for efficient tissue in some crypt cells and occasionally in cells within the villi repair ( Metcalfe et al., 2014 ). Interestingly, despite the exis- (Figure 1 A) of the small intestine. K15 protein can be de- tence of Wnt-negative, injury-resistant reserve ISCs that tected in all small intestinal segments with slightly higher contribute to intestinal epithelial regeneration, evidence expression in the ileum. We confirmed these results using a exists for plasticity in more diferentiated intestinal cells. mouse expressing a progesterone receptor (PR)-fused Cre For example, Dll1+ secretory progenitor cells can revert to recombinase downstream of Krt15 promoter ( Krt15-CrePR1 a stem cell state and give rise to Lgr5+ cells ( van Es et al., mouse) ( Morris et al., 2004 ) crossed withareporter mouse 2012 ). More recently, Asfaha et al. ( 2015 ) identified expressing the mTomato-LSL-mGFP construct from the Krt19+/Lgr5- radio-resistant and cancer-initiating cells in Rosa26 locus ( R26 mT/mG ) (Muzumdar et al., 2007 ). A single the small intestine located above the crypt base. Similarly, injection of the PR agonist, RU486, was used for Cre induc- alkaline-phosphatase-positive transit-amplifying cells can tion and mice were sacrificed 24 hr later to visualize Krt15+ regenerate CBCs after their genetic ablation with Lgr5- (GFP+) cells. Krt15+ cells were observed in some crypts and DTR ; however, it remains unknown whether this mecha- occasionally in the villi of the small intestine ( Figure 1 B). In nism is employed in DNA damaging injury under physio- the villi, 10% of the GFP+ cells also stained positive for logical conditions, which might be expected to also ablate Alcian blue, a goblet cell marker, and 3 % also expressed the very rapidly cycling transit-amplifying cells ( Tetteh chromograninA(CHGA), an enteroendocrine cell marker et al., 2016 ). (Figures S1 A and S1B), suggesting that the majority of In the present study, we describe the role of Keratin 15 Krt15+ cells in the villi are enterocytes. Since the Krt15+ (Krt15 )-labeled cells in the mouse small intestinal epithe- villi cells could result from migration of a recombination lium. Krt15 -labeled cells were characterized initially as event in the top of the crypt, we injected Krt15- stem cells in the hair follicle bulge contributing to wound CrePR1;R26 mT/mG mice with a single dose of RU486 and healing and the development of squamous papilloma mice were sacrificed 8 hr later to minimize the potential (Morris et al., 2004; Ito et al., 2005 ; Li et al., 2013 ). We for migration ofabeled cells. We observed that most described recentlyalong-lived Krt15+ progenitor cell pop- labeled cells were located in the lower third of the villi, ulation in the mouse esophageal epithelium ( Giroux et al., thereby suggesting that recombination happened in the 2017 ). Herein, we identify and describe a long-lived Krt15+ crypt or at the crypt/villi junction ( Figure S1 C). cell population in the small intestinal crypt using genetic Interestingly, approximately 40% of Krt15 -labeled cells lineage tracing in mice. Krt15+ crypt cells give rise to all in the crypt are localized in the stem cell compartment the intestinal lineages and have self-renewal capacity. Ra- (i.e., +4 position and below) ( Figures 1 C and S1D). We dio-resistant Krt15+ cells contribute to tissue regeneration confirmed that the GFP+ cells located between positions +1 after radiation-mediated injury. Interestingly, Apc loss in and +4 from the crypt base are not Paneth cells by lysozyme Krt15+ cells leads to adenoma and adenocarcinoma forma- staining ( Figures 1 D and 1E). GFP can also be detected in tion in the small intestine, as well as occasional adenoma some Ki-67 + transit-amplifying cells as a result of possible formation in the colon, demonstrating the tumor-initi- stem cell division and/or residual Krt15 promoter activity ating potential of these cells.

1948 Stem Cell Reportsj Vol. 10 j 1947–1958 j June 5, 2018

10. ORIGINAL PAPER

Figure 1. Krt15 Is Expressed in a Subpop- ulation of Crypt and Villi Cells (A) Representative immunohistochemistry (IHC) for K15 in the small intestine (ileum) of adult mice. Small intestinal tissues from Krt15 / mice were used as negative control. Insets for higher magnification of (A0) crypt and (A00) villi staining. Arrows mark cells expressing K15 protein (n = 4 mice). (B–E) Krt15-CrePR1;R26mT/mG mice were administered a single injection of 0.5 mg RU48 6 to induce Cre recombination and were sacrificed 2 4 hr later (n = 4 mice). (B) Representative visualization ofKrt15+ cells by GFP-targeted IF. Arrow marks a GFP + crypt cell. (C) Percentage of GFP+ cells at each crypt position. Graph represents mean± SEM (n =3mice, 50 crypts/mouse were counted). (D and E) Representative co-localization of Krt15+(GFP+) crypt cells with lysozyme (LYS, Paneth cell marker) and Ki-67 (proliferative cell marker). (E) Graph represents percentage of GFP+ crypt cells that are positive for LYS or Ki-67 , mean ± SEM (n =4mice, 50 crypts/ mouse were counted). (F) Krt15 mRNA expression inLgr5+ sorted cells versus unsorted cells. Graph represents mean± SEM (n = 3–5 samples). *Represents statistical significance p<0.05 using the rank-sum (Wilcoxon) test. (G) Co-localization of Krt15+ (GFP+) cells with OLFM4 (CBC marker). Arrow indicates a co-localization event. Scale bars, 50mm except for inserts (10mm). See alsoFigure S1.

in these cells ( Figures 1 D and 1E). Interestingly, the +5 results suggest that Krt15- labeled cells in the crypt overlap to +10 region may contain progenitor/stem cells as illus- with CBCs, as well as transit-amplifying cells. trated by Krt19 genetic labeling of multipotent and self-renewing cells ( Asfaha et al., 2015 ). In order to evaluate the Krt15 Marks Long-Lived Multipotent Crypt Cells with crypt cell types that express Krt15 , we used Lgr5-EGFP Self-Renewal Capacity mice. Krt15 expression is enriched in Lgr5+ cells compared We next performed lineage-tracing experiments using with unsorted crypt cell samples ( Figure 1 F). Finally, small Krt15-CrePR1;R26 mT/mG mice to investigate the self-renew- intestines were harvested from Krt15-CrePR1;R26 mT/mG ing capacity of Krt15+ crypt cells. Cre recombination was mice sacriﬁced 24 hr after Cre recombination. OLFM4, induced by daily injection of RU48 6 for 5 days ( Figure 2 A). another marker of Lgr5+ stem cells, expression was evalu- Krt15 -derived cells completely labeled the small intestinal ated. OLFM4 co-expressing GFP+ cells were observed in crypt-villus axis by 14 days. In the early time points, we the bottom of the crypt ( Figure 1 G). In addition, we inves- rarely observed clusters of cells at the crypt base, but found tigated the potential overlap between Krt15+ cells and +4 them mostly in the transit-amplifying zone, which sug- reserve ISCs. Bmi1 and Hopx expression is signiﬁcantly gests that the Krt15+ cells present in that region might lower in Krt15+ cells versus Krt15 cells ( Figure S1 E). These be at the origin of the long-term tracing events as was

Stem Cell Reportsj Vol. 10 j 1947–1958 j June 5, 2018 1949

10. ORIGINAL PAPER

Figure 2. Krt15 Marks Long-Lived Cells in the Small Intestinal Crypt as Well as Some Cells in the Villi (A and B) Six- to 8-week-old Krt15-CrePR1;R26mT/mG mice were injected daily with 0.5 mg RU486 (PR agonist) to induce Cre recombination for 5 consecutive days and sacrificed at diferent time points (days post- recombination = D). (B)Krt15+cells were visualized by GFP immunofluorescence (n = 4–7 mice/time point). (C and D)Krt15-CrePR1;R26Confmice were injected twice a day with 1 mg RU486 for 1 0 consecutive days to induce recombination. Mice were sacrificed 2 months following recombination (n = 4 mice). (D) Represen- tative images of YFP+ and RFP+Krt15-derived clones visualized by immunofluorescence. Scale bars, 50mm. See alsoFigures S2 and S3 .

1950 Stem Cell Reportsj Vol. 10 j 1947–1958 j June 5, 2018

10. ORIGINAL PAPER

observed with Krt19+/Lgr5 cells ( Asfaha et al.,2015 ). La- gated the contribution of Krt15+ cells to tissue regeneration beling persists beyond 6 months,thereby suggesting that following high-dose g-irradiation. First, we used Krt15- some Krt15+ crypt cells are long-lived stem cells ( Figures CrePR1;R26 mT/mG mice to determine if Krt15+ cells are resis- 2B and S2A). Notably,7% of the crypts were labeled 1 day tant to high-dose radiation. Following Cre recombination, after the last RU186 injection and 3% after 2 months ( Fig- mice were subjected to whole-body g-irradiation (12 Gy) ure S2 B). Co-localization of GFP with Ki-67,Alcian blue, (Figure 4 A). After irradiation, Krt15+ cells were present in chromogranin A,and lysozyme demonstrate that Krt15+ the highly proliferative crypts that underwent regenera- cells can give rise to proliferative,transit-amplifying cells, tion and designated as microcolonies ( Figure 4 B). Five goblet cells,enteroendocrine cells,and Paneth cells,respec- days following irradiation,the percentage of GFP+ cells tively ( Figures S3 A–S3D). Krt15-CrePR1 mice were also per crypt was increased in the irradiated mice when crossed with RosaConfetti mice. Every Krt15 -labeled ribbon compared with non-irradiated mice ( Figure 4 C). Further- observed was monochromatic,suggesting that the progeny more, Krt15+ crypt cells were more proliferative in irradi- of Krt15+ cells may arise from monoclonal units ( Figures 2 C ated animals versus non-irradiated mice ( Figures 4 D and and 2D). Interestingly, Krt15 marked clones in the colon 4E). These results are consistent with an expansion epithelium as well ( Figure S3 E). Taken together,these re- of Krt15+ cells in response to irradiation. Finally, we sults indicate that Krt15 can mark long-lived and multipo- induced Cre recombination in Krt15-CrePR1;R26 mT/mG tent crypt stem cells. mice following whole-body irradiation and noticed that To investigate the self-renewing capacity of Krt15+ crypt Krt15+ cells expanded in response to injury in these condi- cells, we used Krt15-CrePR1;Rosa26 LSL-tdTomato (Krt15- tions as well ( Figures S4 B and S4C). Thus,we demonstrate CrePR1;R26 Tom) mice. A single injection of RU486 was that Krt15 -labeled cells are radio-resistant and expand in used to induce recombination and mice were sacrificed response to high-dose irradiation. 24 hr later. Crypts were isolated and grown as 3D organo- Due to the multiple roles of keratins in cell migration and ids. Krt15+ (Tomato+) cells were detected in 3D organoid proliferation,we sought to characterize the potential role crypts and these cells expanded to form lineage-traced of K15 protein itselfn tissue regeneration following injury. ribbons as observed in vivo (Figure 3 A). Interestingly, We subjected Krt15 +/+ and Krt15 / mice to whole-body following 3D organoid passaging,we observed organoids g-irradiation (12 Gy) and sacrificed them 5 days later ( Fig- consisting entirely of Tomato+ cells ( Figures 3 A and 3B). ure 5 A). Regeneration of the ileum from Krt15 / mice The multipotency of Krt15+ crypt cells was confirmed in was impaired following irradiation ( Figure 5 B). First,fewer these 3D organoids with Tomato+ proliferative cells,goblet residual crypts were observed in the ileum from Krt15 / cells,enteroendocrine cells,and Paneth cells ( Figure 3 C). compared with Krt15 +/+ irradiated mice ( Figure 5 C). Sec- Furthermore,single crypt cell suspensions were prepared ond,crypt length and villus height were both significantly from Krt15-CrePR1;R26 Tom mice 24 hr following Cre recom- reduced in Krt15 -deficient mice when compared with bination. Krt15+ cells were able to grow as single cell cul- Krt15 +/+ mice ( Figures 5 D and 5E). Third,crypt regenerative tures,suggesting that Krt15+ crypt cells may display ‘‘stem- capacity was determined using bromodeoxyuridine (BrdU) ness’’features. The organoid formation efficiency of Krt15+ incorporation. Crypts containing >10 BrdU+ cells were crypt cells is 0.01%,which is less than the 6% efficiency of considered as microcolonies. Krt15 deficiency significantly Lgr5+ cells originally reported ( Sato et al.,2011 ),but similar reduced the percentage of microcolonies present in the to the 0.01%–1.2% efficiency reported by other groups ileum 5 days following irradiation ( Figures 5 F and 5G). (Xian et al.,2017;Qi et al.,2017 ). Of note,these diferences Finally,the percentage of BrdU+ cells per crypt was lower in organoid formation capacity of Krt15+ and Lgr5+ cells in Krt15 / versus Krt15 +/+ ileum. These results suggest a may not correlate with distinct properties in vivo. Taken possible role for K15 protein in intestinal tissue regenera- together, these results demonstrate that Krt15 marks tion. We appreciate this is likely diferent from the proper- some long-lived and multipotent crypt cells with self-re- ties of the Krt15+ cells. newing capacity,characteristics consistent with stem cells. Apc Loss in Krt15+ Cells Leads to Adenoma and Krt15+ Cells Are Radio-Resistant and Expand in Adenocarcinoma Formation in the Small Intestine Response to High-Dose Radiation and Colon Reserve stem cells are believed to be radio-resistant and Tumor-initiating capacity is an important feature of ISCs. contribute to tissue regeneration following high-dose radi- Lgr5+, Lrig1+, and Krt19+/Lgr5 populations have all ation ( Sangiorgi and Capecchi, 2008; Yan et al., 2012; been described to contain a cell of origin for intestinal Takeda et al., 2011; Yousefi et al., 2016 ). Interestingly, tumorigenesis following Apc loss ( Barker et al.,2009;Powell Krt15 mRNA expression is increased in small intestinal cells et al.,2012;Asfaha et al.,2015 ). Also, Bmi1-CreER+ cells can following irradiation ( Figure S4 A). Therefore,we investi- form adenomas following induction of a stable form of

Stem Cell Reportsj Vol. 10 j 1947–1958 j June 5, 2018 1951

10. ORIGINAL PAPER

Figure 3. Krt15 Lineage-Labeled Cells Have Clonogenic Potential (A–D) Krt15-CrePR1;R26Tommice were injected with 0.5 mg RU48 6 for Cre induction and sacriﬁced 2 4 hr later. 3 D organoids were grown from isolated small intestinal (ileum) crypts (P0) and passaged once (P1 ) (n = 4 mice). (A and B) Tomato+ (Krt15-derived) cells (A) were visualized on live 3 D organoids as well as (B) by IF on sections of ﬁxed organoids. (C) Representative co-staining of Tomato with Ki-6 7 (proliferative cell marker), Mucin2 (MUC2 , goblet cell marker), Chromogranin A (CHGA, enteroendocrine cell marker), and lysozyme (LYS, Paneth cell marker). Arrows mark co-localization. (D) Single suspensions of Tomato+ cells were sorted and seeded in Matrigel and grown for 1 0 days. Representative pictures of 3 D organoids growing fromKrt15+cells at diferent time points following seeding (D = days) (n = 4 mice, 4 experimental replicates/mouse). Arrows mark spheres that are forming. Scale bars, 50mm.

b-catenin in these cells ( Sangiorgi and Capecchi, 2008 ). In ing daily injection of 0.5 mg RU48 6 for 5 consecutive order to determine if Krt15+ cells can initiate tumor forma- days. Mice were sacrificed 6 months after Cre recombination in the intestine, we bred Apcfl/fl mice with the Krt15- tion or at prior time points if severe weight loss necessitated CrePR1;R26 mT/mG mice. Cre recombination was induced us- sacrifice ( Figure 6 A). Gross lesions were observed in the

1952 Stem Cell Reportsj Vol. 10 j 1947–1958 j June 5, 2018

10. ORIGINAL PAPER

Figure 4. Krt15+ Cells Are Radio-Resis- tant and Contribute to Tissue Regener- ation (A–E) Krt15-CrePR1;R26mT/mG mice were treated with 0.5 mg RU48 6 daily for 5 consecutive days to induce Cre recombination. Mice were subjected or not to 1 2 Gy whole-bodyg-irradiation 1 day after Cre induction and sacrificed 2 (D2 ) or 5 days (D5) after irradiation (NI = Non-irradiated) (n = 3–4 mice/group). (B) GFP labeling ofKrt15- derived cells during regeneration following irradiation. (C) Percentage of GFP+ cells in labeled crypts following irradiation in comparison to normal conditions. Graph represents mean ± SEM (n = 3–4 mice/group). *Represents statistical significance with p < 0.05 versus D2 and#p<0.05 versus non- irradiated animals, using Student’s t test). (D and E) Ki-6 7 staining of GFP+ regenerative crypts. (E) Percentage of Ki-6 7 positivity in crypt GFP+ cells. Graph represents mean± SEM (n = 3–4 mice/group, 50 crypts/mouse were analyzed). *Represents statistical significance with p < 0.05 versus D2 and#p < 0.05 versus non-irradiated animals, using Student’s t test. Scale bars, 50mm. See alsoFigure S4.

intestinal tract of Krt15-CrePR1;Apc fl/fl ;R26 mT/mG mice but DISCUSSION not in control mice (n = 8 mice) ( Figure 6 B). All mice developed lesions in the small intestine while colon lesions We report that some Krt15+ lineage-labeled small intestinal were observed in 37 % of the mice ( Figure 6 C). Mice devel- crypt cells have self-renewal capacity, give rise to all difer- oped between 4 and 33 tumors measuring on average entiated cell lineages, and participate in tissue regeneration 3.17 mm. Most observed lesions were adenomas, but inva- following high-dose radiation injury. In combination with sive adenocarcinomas were occasionally observed. Interest- Apc loss, Krt15+ cells can give rise to adenomas and even ad- ingly, half of the Krt15-CrePR1;Apc fl/fl ;R26 mT/mG mice devel- enocarcinomas. While the spatial localization of Krt15+ oped at least one adenocarcinoma ( Figures 6 D and 6E). The cells in intestinal crypts is not limited to CBCs or +4 ISCs, lesions were highly proliferative, as demonstrated by a our results suggest that Krt15 could annotate a subpopula- large number of Ki-67 + cells ( Figure 6 F) and had also evi- tion of Lgr5+ cells. dence of nuclear b-catenin staining ( Figure 6 G). TP53 Krt15+ cells are predominantly detected at the crypt base mutations are observed at later stages of colon cancer and where Lgr5+ stem cells reside ( Barker et al., 2007 ), while the typically lead to nuclear accumulation of p5 3 ( Fearon and Bmi1+ or Hopx+ reserve stem cells are localized at the +4 po- Vogelstein, 1990 ). We observed nuclear p5 3 staining in sition above the crypt base ( Sangiorgi and Capecchi, 2008; some tumor regions ( Figure 6 H). These results demonstrate Takeda et al., 2011 ). Sorted Lgr5+ cells also harbor higher that tumor-initiating cells exist within the Krt15+ popula- Krt15 mRNA expression and some Krt15- labeled cells tion, and lesions derived from Krt15+ cells can even prog- also express OLFM4 . These results suggest at least partial ress to invasive adenocarcinoma. overlap between Krt15+ cells and Lgr5+ cells. Analysis of

Stem Cell Reportsj Vol. 10 j 1947–1958 j June 5, 2018 1953

10. ORIGINAL PAPER

Figure 5. Krt15 Loss Impairs Intestinal Tissue Regeneration Following Gamma- Irradiation (A–H) Six- to 8-week-old Krt15+/+ and Krt15 / mice were subjected to 12 Gy whole-body g-irradiation and sacrificed 5 days later. Tissue injury and regeneration were compared in ileal sections. (B) Representative H&E staining from ileal sections. (C) Residual crypts by high-power field (HPF) images were quantified, graph represents mean± SEM (n = 3–4 mice/group, 2 5 HPFs/ mouse were analyzed. *Represents statistical significance with p<0.05 using Student’s t test). (D and E) Crypt length (D) and villi height (E) were measured in ileal sections, graph represents mean± SEM (n = 3–4 mice/group, 50 crypts and villus/mouse were measured. *Represents statistical significance with p < 0.05 using Student’s t test). (F–H) Mice were injected with BrdU 1.5 hr prior to sacrifice and BrdU+ cells were detected by IHC. (G) The percentage of microcolonies (crypt with >10 BrdU + cells/total number of crypts by HPF) was quantified, graph represents mea±n SEM (n = 3–4 mice/group, 2 5 HPFs/mouse were analyzed. *Represents statistical significance with p < 0.05 using Student’sttest). (H) The percentage of BrdU+ cells/crypt was quantified, graph represents mean± SEM (n = 3–4 mice/group, 50 crypts/mouse were analyzed. *Represents statistical significance with p<0.05 using Student’s t test). Scale bars, 50mm.

published datasets also suggests upregulation of Krt15 tran- et al., 2012; Metcalfe et al., 2014 ) and these might be script in Lgr5+ cells versus Lgr5 cells (1.0084 2 versus marked by Krt15 promoter activity. Recently, a subpopula- 0.0258, p = 0.052) ( Yan et al., 2012 ). Some Krt15+ cells tion of Lgr5+ cells was described as expressing a high level are located above the +4 position, in the transit-amplifying of Mex3a , suggesting heterogeneity in the Lgr5+ cell popu- zone similar to Krt19+/Lgr5 stem-like cells ( Asfaha et al., lation ( Barriga et al., 2017 ). However, Mex3a is not diferen- 2015 ). This suggests heterogeneity throughout the Krt15+ tially expressed between Krt15+ and Krt15 cells (data not crypt population. Nevertheless, we observed multipotency shown). Furthermore, analysis of published datasets re- and self-renewing capacity in Krt15+ cells, indicating that vealed no enrichment of Krt15 in previously described at least some Krt15+ crypt cells have properties consistent Lgr5+ subpopulations of cells ( Barriga et al., 2017 ; Yan with stem cells or long-lived progenitor cells. et al., 2012; Munoz et al., 2012 ). Krt15+ cells located above Lgr5+ CBCs are thought to be highly sensitive to high- the +4 position might also be radio-resistant, as has been dose radiation, while the reserve stem cells are resistant to described for the similarly located Krt19+/Lgr5 cells( As- such an insult ( Hua et al., 2012; Metcalfe et al., 2014 ; Yan faha et al., 2015 ). et al., 2012; Tao et al., 2015 ). Herein, we demonstrate that Inadequate regeneration following radiation was intestinal Krt15+ crypt cells are resistant to high-dose radi- observed in the Krt15- deﬁcient mice, suggesting a possible ation. Due to the possible overlap between Krt15+ cells and role for the K15 protein. Keratins confer structural support Lgr5+ cells, Krt15 could mark a subpopulation of Lgr5+ cells in epithelial cells and are also involved in sensing cues from that could be resistant to radiation. Although radiation ab- the microenvironment. Keratins formastrong complex lates the vast majority of Lgr5+ cells, a small fraction of with integrins and other extracellular-matrix-associated Lgr5+ cells persists ( Yan et al., 2012; Tao et al., 2015; Hua proteins. It is believed that the stem cell niche often

1954 Stem Cell Reportsj Vol. 10 j 1947–1958 j June 5, 2018

10. ORIGINAL PAPER

Figure 6. Krt15+ Cells Are Tumor-Initi- ating Cells in the Intestine (A) Krt15-CrePR1;Apcfl/fl ;R26mT/mGmice were injected daily with 0.5 mg RU48 6 for 5 consecutive days to induce Cre recombination and mice were sacrificed when sick or 150 days following Cre induction (n = 8 mice). (B) Representative image of small intestinal tumors observed in Krt15-CrePR1;Apcfl/fl ; R26mT/mGmice. Scale bar, 1 mm. (C) Incidence ofesions in small intestine (SI) and colon of Krt15-CrePR1;Apcfl/fl ; R26mT/mGmice (n = 8 mice). (D) Incidence of adenomas and adenocarcinomas inKrt15-CrePR1;Apcfl/fl ;R26mT/mGmice (n = 8 mice). (E–H) Representative histology of flat adenoma and invasive adenocarcinoma observed in Krt15-CrePR1;Apcfl/fl ;R26mT/mG mice. (F) Ki-67 , (G) b-catenin, and (H) p53 staining of small intestinal tumors observed inKrt15- CrePR1;Apcfl/fl ;R26mT/mG mice. Scale bar, 50 mm.

involves extrinsic cells or extracellular proteins or mole- also observed following induction of stable b-catenin in cules that will modify the immediate environment of Bmi1-CreER+ cells ( Sangiorgi and Capecchi, 2008 ). None stem cells. Interestingly, integrin signaling has been shown of these studies have reported progression to adenocarci- to be essential for ISC maintenance and proliferation in noma. Herein, we observed that mice harboring Apc loss Drosophila (Lin et al., 2013; You et al., 2014 ). It has been in Krt15-CrePR1+ cells develop adenomas and invasive ad- shown that keratins can interact with several kinases and enocarcinomas. These results suggest that Krt15+ cells are other cytoplasmic proteins to regulate cell proliferation tumor-initiating cells, spanning benign to malignant and cell metabolism ( Bragulla and Homberger, 2009 ). For lesions. Loss of Apc in Krt15+ cells leads to more severe le- example, K19 interacts with b-catenin/RAC1 complex in sions that what was reported for Lgr5+, Lrig1+, or Krt19+ breast cancer cells inducing nuclear translocation of b-cat- cells. Mice presenting Apc loss in Krt15+ cells survive enin( Saha et al., 2017 ). K8 and K18 interact with TNFR2 longer in contrast to other reported mice, which could afecting nuclear factor kB signaling ( Caulin et al., 2000 ) mean that the latter mice do not survive long enough and K17 regulates the AKT/mTOR pathway by interacting to display adenocarcinomas. Alternatively, this could with 14-3-3 s , resulting in protein synthesis and cell growth suggest diferent cells of origin for adenomas and adeno- upregulation ( Kim et al., 2006 ). Herein, we demonstrate carcinomas. Interestingly, in inherited colon cancer syn- that mice lacking Krt15 recover more slowly in response dromes, one may see variable presentation of adenoma- to high-dose radiation, thereby suggesting that K15 protein tous polyps or hamartomatous polyps with or without could also play a role in tissue regeneration. progression to colon cancer ( Rustgi, 2007 ) and this may Several studies have been conducted to identify the cell underlie the potential heterogeneity of cells of origin of origin ofntestinal cancers. Apc loss in Lgr5-CreER+ cells that may be implied in the various mouse models. Finally, (Barker et al., 2009 ; Asfaha et al., 2015 ), Lrig1-CreER+ cells extensive adenoma formation like observed in mice car- (Powell et al., 2012, 2014 ), and Krt19-CreER+ cells ( Asfaha rying loss of Apc in Lgr5+ cells could require culling before et al., 2015 ) lead to adenoma formation. Adenomas are cancer progression.

Stem Cell Reportsj Vol. 10 j 1947–1958 j June 5, 2018 1955

10. ORIGINAL PAPER

The lesions formed by Krt15+ cells most likely originate Single-Cell Isolation from the cells located at the crypt base and/or those located Crypt single cells were isolated as described previously ( Hamilton in the transit-amplifying zone. We also cannot exclude et al.,2015 ). Briefly,the small intestine was opened longitudinally the possibility that Krt15+ villi cells might have given rise and washed with cold PBS. Tissue was incubated 20 min on ice in to the lesions since it has been demonstrated that cells in PBS-EDTA-DTT. Tissues were then incubated in PBS-EDTA at 37 C. Crypts were isolated from the epithelial fraction using a 70- mm fil- the villi can initiate tumorigenesis ( Davis et al.,2015 ). In ter. Crypts were then dissociated to a single-cell suspension using that model,aberrant expression of the BMP antagonist, PBS/dispase at 37 C. GREM1,led to formation ofntestinal tumors histologically similar to polyps observed in hereditary mixed polyposis Crypt Isolation and 3D Organoid Culture syndrome. Nevertheless, Krt15-CrePR1;Apc fl/fl mice might Crypts were isolated from the mouse small intestinal ileum. Tissues represent a new mouse model for intestinal cancer as a foun- were chilled in Ca 2+-Mg 2+-free Hank’s balanced salt solution (CMF- dation for understanding the interrelationship between HBSS) with 1 mM N-acetyl-cysteine (NAC). Tissues were then incu- normal intestinal cells and malignant transformation. bated in CMF-HBSS/1 mM NAC/10 mM EDTA for 45 min at 4 C. Overall,we demonstrate that Krt15+ crypt cells are self- Epithelial cells were then dissociated through vortexing/resting renewing,multipotent,radio-resistant,and tumor initi- cycles. Crypts were separated from epithelial dissociation with a ating in concert with Apc loss. Krt15+ cells could represent 70- mm filter. Crypts were pelleted and resuspended in Basal Media a radio-resistant subpopulation of Lgr5+ cells. RNA (Advanced DMEM/F12,2 mM Glutamax,10 mM HEPES,1 3 peni- profiling of Krt15+ esophageal long-lived progenitor cells cillin/streptomycin,5 mM CHIR99021 [Cayman Chemical],1 mM revealed enrichment for gene sets associated with Wnt/ NAC, 1 3 N2 Supplement [Gibco], and 1 3 B27 Supplement b-catenin pathway and DNA repair ( Giroux et al.,2017 ). [Gibco]). Approximately 500 crypts were then embedded in 80% It is possible,if not likely,that the Wnt/ b-catenin pathway Matrigel/20% ENR (Basal Media;50 ng/mL recombinant mouse epidermal growth factor [R&D Systems] and 1% Noggin/R-Spon- is involved in the proliferative capacity of Krt15+ intestinal din conditioned media). Enteroids were cultured in ENR. crypt cells and their tumor-initiating capacities. Further- more,Krt15+ cells could also display higher DNA repair ca- Fluorescence-Activated Cell Sorting pacity to ensure genomic stability in response to radiation Crypt single-cell suspension was prepared from Krt15-CrePR1; by example. R26Tom small intestines as described above. The fluorescence-activated cell sorting sorter Jazz (BD Biosciences) was used to sort EXPERIMENTAL PROCEDURES Tomato+ and Tomato-crypt cells. Sorting was conducted at the University of Pennsylvania Flow Cytometry and Cell Sorting Facil- ity. Cells were sorted in Advanced DMEM/F12 media supple- Mouse Experimental Design mented with 1 3 penicillin-streptomycin, 1 3 GlutaMAX, 1 3 Krt15-CrePR1 (Morris et al.,2004 ) and Krt15 / mice were pro- HEPES,and 10 mM Rock inhibitor Y27632. Cells were then used vided by Dr. George Cotsarelis (University of Pennsylvania). for 3D organoid formation or RNA extraction. Rosa26mTomato/mGFP (R26mT/mG ) mice ( Muzumdar et al.,2007 ) were Details on mouse experimental design,immunohistochemistry purchased from Jackson Laboratories. Rosa26Confetti (R26Conf) mice and immunofluorescence,RNA extraction,qPCR,and statistical were provided by Dr. Ben Stanger (University of Pennsylvania). analyses are available in Supplemental Information . Rosa26LSL-tdTomato (R26Tom) mice were provided by Dr. Christopher Lengner (University of Pennsylvania). Apcfl/fl mice were obtained from the NCI Mouse Repository ( Kuraguchi et al.,2006 ). All exper- SUPPLEMENTAL INFORMATION iments were performed with 6- to 10-week-old mice. Intraperito- Supplemental Information includes Supplemental Experimental neal injection of PR agonist,RU486,was performed to induce Cre Procedures,four figures,and two tables and can be found with recombination. The Institutional Animal Care and Use Committee this article online at https://doi.org/10.1016/j.stemcr.2018.04.022 . of the University of Pennsylvania approved all animal studies. Detailed information regarding mouse experimental design and AUTHOR CONTRIBUTIONS treatment are in the Supplemental Information . V.G.,K.E.H.,C.J.L.,and A.K.R. designed the study. V.G.,J.S.,and P.C. performed the experiments. B.R. maintained the mouse col- Irradiation ony. E.P.W. assisted with statistical analysis. A.J.K.-S. assisted with Six- to eight-week-old mice were subjected to 12 Gy whole-body histology analysis. V.G. and A.K.R. wrote the manuscript. J.S. is gamma-irradiation (Gammacell 40 Cesium 137 Irradiation Unit). currently affiliated with University Medicine Greifswald (Depart- Mice were sacrificed 5 days later,and the small intestines were har- ment of Medicine). vested and fixed for histology. Surviving crypts,crypt length,and villi height were measured on H&E-stained ileum sections. ACKNOWLEDGMENTS Twenty-five high-magnification fields were analyzed for each mouse and a minimum of 100 crypts or villi were measured for We are grateful to members of the Rustgi lab for discussions and each mouse. comments on the manuscript. We thank the Molecular Pathology

1956 Stem Cell Reportsj Vol. 10 j 1947–1958 j June 5, 2018

10. ORIGINAL PAPER

and Imaging Core, Human Microbial and Analytic Depository ing cells are secretory precursors expressing Lgr5. Nature 495 , Core, Cell Culture and iPS Core, Genetic and Modified Mouse 65–69 . Core, and FACS/sorting Core facilities (University of Pennsylva- Caulin, C., Ware, C.F., Magin, T.M., and Oshima, R.G. (2000). Ker- nia). We thank Dr. George Cotsarelis (University of Pennsylvania) atin-dependent, epithelial resistance to tumor necrosis factor- / for Krt15-CrePR1 and Krt15 mice, Dr. Ben Stanger (University of induced apoptosis. J. Cell Biol. 149 , 17–22 . Pennsylvania) for RosaConfetti mice, and Dr. Christopher Lengner Davis, H., Irshad, S., Bansal, M., Raferty, H., Boitsova, T., Bardella, (University of Pennsylvania) for RosaLSL-tdTomato mice. We thank C., Jaeger, E., Lewis, A., Freeman-Mills, L., Giner, F.C., et al. (2015). MaryAnn Crissey and Dr. John Lynch for providing Lgr5+ sorted Aberrant epithelial GREM1 expression initiates colonic tumorigen- cells. This work was supported by NCI P01-CA098101 (V.G., esis from cells outside the stem cell niche. Nat. Med. 21, 62–70 . A.J.K.-S., and A.K.R.), NIH/National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) P30-DK050306 Center of Fearon, E.R., and Vogelstein, B. (1990). A genetic model for colo- Molecular Studies in Digestive and Liver Diseases, NIH R01- rectal tumorigenesis. Cell 61, 759–767 . DK056645 (A.K.R.), American Cancer Society (A.K.R.), Fonds de re- Giroux, V., Lento, A.A., Islam, M., Pitarresi, J.R., Kharbanda, A., cherche en sante´ du Que´bec P-Giroux-27692 and P-Giroux-31601 Hamilton, K.E., Whelan, K.A., Long, A., Rhoades, B., Tang, Q., (V.G.), NIH NIDDK K01-DK100485 (K.E.H.), and Crohn’s and Co- et al. (2017). Long-lived keratin 15+ esophageal progenitor cells litis Foundation Career Development Award (K.E.H.). contribute to homeostasis and regeneration. J. Clin. Invest. 127 , 2378–2391 . Received: July 7, 2017 Grun, D., Lyubimova, A., Kester, L., Wiebrands, K., Basak, O., Sa- Revised: April 23, 2018 saki, N., Clevers, H., and van Oudenaarden, A. (2015). Single-cell Accepted: April 24, 2018 messenger RNA sequencing reveals rare intestinal cell types. Na- Published: May 24, 2018 ture 525 , 251–255 . Hamilton, K.E., Crissey, M.A., Lynch, J.P., and Rustgi, A.K. (2015). REFERENCES Culturing adult stem cells from mouse small intestinal crypts. Cold Spring Harb. Protoc. 2015 , 354–358 . Asfaha, S., Hayakawa, Y., Muley, A., Stokes, S., Graham, T.A., Erick- Hua, G., Thin, T.H., Feldman, R., Haimovitz-Friedman, A., Clevers, sen, R.E., Westphalen, C.B., von Burstin, J., Mastracci, T.L., Worth- H., Fuks, Z., and Kolesnick, R. (2012). Crypt base columnar stem ley, D.L.,, et al. (2015). Krt19(+)/Lgr5(-) cells are radioresistant can- cells in small intestines of mice are radioresistant. Gastroenter- cer-initiating stem cells in the colon and intestine. Cell Stem Cell ology 143 , 1266–1276 . 16, 627–638 . Ito, M., Liu, Y., Yang, Z., Nguyen, J., Liang, F., Morris, R.J., and Cot- Barker, N., Ridgway, R.A., van Es, J.H., van de Wetering, M., sarelis, G. (2005). Stem cells in the hair follicle bulge contribute to Begthel, H., van den Born, M., Danenberg, E., Clarke, A.R., San- wound repair but not to homeostasis of the epidermis. Nat. Med. som, O.J., and Clevers, H. (2009). Crypt stem cells as the cells-of- 11, 1351–1354 . origin ofntestinal cancer. Nature 457 , 608–611 . Jadhav, U., Saxena, M., O’Neill, N.K., Saadatpour, A., Yuan, G.C., Barker, N., van Es, J.H., Kuipers, J., Kujala, P., van den Born, M., Co- Herbert, Z., Murata, K., and Shivdasani, R.A. (2017). Dynamic reor- zijnsen, M., Haegebarth, A., Korving, J., Begthel, H., Peters, P.J., ganization of chromatin accessibility signatures during dediferen- et al. (2007). Identification of stem cells in small intestine and co- tiation of secretory precursors into Lgr5+ intestinal stem cells. Cell lon by marker gene Lgr5. Nature 449 , 1003–1007 . Stem Cell 21, 65–77.e5 . Barker, N., van Oudenaarden, A., and Clevers, H. (2012). Identi- Kim, S., Wong, P., and Coulombe, P.A. (2006). A keratin cytoskel- fying the stem cell of the intestinal crypt: strategies and pitfalls. etal protein regulates protein synthesis and epithelial cell growth. Cell Stem Cell 11, 452–460 . Nature 441 , 362–365 . Barriga, F.M., Montagni, E., Mana, M., Mendez-Lago, M., Her- Kretzschmar, K., and Clevers, H. (2017). Wnt/beta-catenin nando-Momblona, X., Sevillano, M., Guillaumet-Adkins, A., Ro- signaling in adult mammalian epithelial stem cells. Dev. Biol. driguez-Esteban, G., Buczacki, S.J.A., Gut, M., et al. (2017). 428 , 273–282 . Mex3a marks a slowly dividing subpopulation of Lgr5+ intestinal stem cells. Cell Stem Cell 20, 801–816.e7 . Kuraguchi, M., Wang, X.P., Bronson, R.T., Rothenberg, R., Ohene- Baah, N.Y., Lund, J.J., Kucherlapati, M., Maas, R.L., and Kucherla- Beumer, J., and Clevers, H. (2016). Regulation and plasticity ofn- pati, R. (2006). Adenomatous polyposis coli (APC) is required for testinal stem cells during homeostasis and regeneration. Develop- normal development of skin and thymus. PLoS Genet. 2, e146 . ment 143 , 3639–3649 . Li, N., Nakauka-Ddamba, A., Tobias, J., Jensen, S.T., and Lengner, Bjerknes, M., and Cheng, H. (1999). Clonal analysis of mouse in- C.J. (2016). Mouse label-retaining cells are molecularly and func- testinal epithelial progenitors. Gastroenterology 116 , 7–14 . tionally distinct from reserve intestinal stem cells. Gastroenter- Bragulla, H.H., and Homberger, D.G. (2009). Structure and func- ology 151 , 298–310.e7 . tions of keratin proteins in simple, stratified, keratinized and corni- Li, N., Yousefi, M., Nakauka-Ddamba, A., Jain, R., Tobias, J., Ep- fied epithelia. J. Anat. 214 , 516–559 . stein, J.A., Jensen, S.T., and Lengner, C.J. (2014). Single-cell anal- Buczacki, S.J., Zecchini, H.I., Nicholson, A.M., Russell, R., Vermeu- ysis of proxy reporter allele-marked epithelial cells establishes len, L., Kemp, R., and Winton, D.J. (2013). Intestinal label-retain- intestinal stem cell hierarchy. Stem Cell Reports 3 , 876–891 .

Stem Cell Reportsj Vol. 10 j 1947–1958 j June 5, 2018 1957

10. ORIGINAL PAPER

Li, S., Park, H., Trempus, C.S., Gordon, D., Liu, Y., Cotsarelis, G., with beta-catenin/RAC1 complex to regulate NUMB-dependent and Morris, R.J. (2013). A keratin 15 containing stem cell popula- NOTCH signaling pathway and breast cancer properties. Onco- tion from the hair follicle contributes to squamous papilloma gene 36, 332–349 . development in the mouse. Mol. Carcinog. 52, 751–759 . Sangiorgi, E., and Capecchi, M.R. (2008). Bmi1 is expressed in vivo Lin, G., Zhang, X., Ren, J., Pang, Z., Wang, C., Xu, N., and Xi, R. in intestinal stem cells. Nat. Genet. 40, 915–920 . (2013). Integrin signaling is required for maintenance and prolifer- Sato, T., van Es, J.H., Snippert, H.J., Stange, D.E., Vries, R.G., van ation ofntestinal stem cells in Drosophila. Dev. Biol. 377 , 177–187 . den Born, M., Barker, N., Shroyer, N.F., van de Wetering, M., and Metcalfe, C., Kljavin, N.M., Ybarra, R., and de Sauvage, F.J. (2014). Clevers, H. (2011). Paneth cells constitute the niche for Lgr5 Lgr5+ stem cells are indispensable for radiation-induced intestinal stem cells in intestinal crypts. Nature 469 , 415–418 . regeneration. Cell Stem Cell 14, 149–159 . Takeda, N., Jain, R., LeBoeuf, M.R., Wang, Q., Lu, M.M., and Ep- Montgomery, R.K., Carlone, D.L., Richmond, C.A., Farilla, L., Kra- stein, J.A. (2011). Interconversion between intestinal stem cell nendonk, M.E., Henderson, D.E., Bafour-Awuah, N.Y., Ambruzs, populations in distinct niches. Science 334 , 1420–1424 . D.M., Fogli, L.K., Algra, S., et al. (2011). Mouse telomerase reverse Tao, S., Tang, D., Morita, Y., Sperka, T., Omrani, O., Lechel, A., Sakk, transcriptase (mTert) expression marks slowly cycling intestinal V., Kraus, J., Kestler, H.A., Kuhl, M., et al. (2015). Wnt activity and stem cells. Proc. Natl. Acad. Sci. USA 108 , 179–184 . basal niche position sensitize intestinal stem and progenitor cells Morris, R.J., Liu, Y., Marles, L., Yang, Z., Trempus, C., Li, S., Lin, J.S., to DNA damage. EMBO J. 34, 624–640 . Sawicki, J.A., and Cotsarelis, G. (2004). Capturing and profiling Tetteh, P.W., Basak, O., Farin, H.F., Wiebrands, K., Kretzschmar, K., adult hair follicle stem cells. Nat. Biotechnol. 22, 411–417 . Begthel, H., van den Born, M., Korving, J., de Sauvage, F., van Es, Munoz, J., Stange, D.E., Schepers, A.G., van de Wetering, M., Koo, J.H., et al. (2016). Replacement ofost Lgr5-positive stem cells B.K., Itzkovitz, S., Volckmann, R., Kung, K.S., Koster, J., Radulescu, through plasticity of their enterocyte-lineage daughters. Cell S., et al. (2012). The Lgr5 intestinal stem cell signature: robust Stem Cell 18, 203–213 . expression of proposed quiescent ’+4’ cell markers. EMBO J. 31, Tian, H., Biehs, B., Warming, S., Leong, K.G., Rangell, L., Klein, 3079–3091 . O.D., and de Sauvage, F.J. (2011). A reserve stem cell population Muzumdar, M.D., Tasic, B., Miyamichi, K., Li, L., and Luo, L. in small intestine renders Lgr5-positive cells dispensable. Nature (2007). A global double-fluorescent cre reporter mouse. Genesis 478 , 255–259 . 45, 593–605 . van der Flier, L.G., van Gijn, M.E., Hatzis, P., Kujala, P., Haegebarth, Potten, C.S., Hume, W.J., Reid, P., and Cairns, J. (1978). The segre- A., Stange, D.E., Begthel, H., van den Born, M., Guryev, V., Oving, gation of DNA in epithelial stem cells. Cell 15, 899–906 . I., et al. (2009). Transcription factor achaete scute-like 2 controls Poulin, E.J., Powell, A.E., Wang, Y., Li, Y., Franklin, J.L., and Cofey, intestinal stem cell fate. Cell 136 , 903–912 . R.J. (2014). Using a new Lrig1 reporter mouse to assess diferences van Es, J.H., Sato, T., van de Wetering, M., Lyubimova, A., Nee, between two Lrig1 antibodies in the intestine. Stem Cell Res. 13, A.N., Gregorief, A., Sasaki, N., Zeinstra, L., van den Born, M., Korv- 422–430 . ing, J., et al. (2012). Dll1+ secretory progenitor cells revert to stem Powell, A.E., Vlacich, G., Zhao, Z.Y., McKinley, E.T., Washington, cells upon crypt damage. Nat. Cell Biol. 14, 1099–1104 . M.K., Manning, H.C., and Cofey, R.J. (2014). Inducible loss of one Apc allele in Lrig1-expressing progenitor cells results in Wong, M.H., Stappenbeck, T.S., and Gordon, J.I. (1999). Living and multiple distal colonic tumors with features ofamilial adenoma- commuting in intestinal crypts. Gastroenterology 116 , 208–210 . tous polyposis. Am. J. Physiol. Gastrointest. Liver Physiol. 307 , Wong, V.W., Stange, D.E., Page, M.E., Buczacki, S., Wabik, A., Itami, G16–G23 . S., van de Wetering, M., Poulsom, R., Wright, N.A., Trotter, M.W., Powell, A.E., Wang, Y., Li, Y., Poulin, E.J., Means, A.L., Washington, et al. (2012). Lrig1 controls intestinal stem-cell homeostasis by M.K., Higginbotham, J.N., Juchheim, A., Prasad, N., Levy, S.E., et al. negative regulation of ErbB signalling. Nat. Cell Biol. 14, 401–408 . (2012). The pan-ErbB negative regulator Lrig1 is an intestinal Xian, L., Georgess, D., Huso, T., Cope, L., Belton, A., Chang, Y.T., stem cell marker that functions as a tumor suppressor. Cell 149 , Kuang, W., Gu, Q., Zhang, X., Senger, S., et al. (2017). HMGA1 am- 146–158 . plifies Wnt signalling and expands the intestinal stem cell Qi, Z., Li, Y., Zhao, B., Xu, C., Liu, Y., Li, H., Zhang, B., Wang, X., compartment and paneth cell niche. Nat. Commun. 8, 15008 . Yang, X., Xie, W., et al. (2017). BMP restricts stemness ofntestinal Yan, K.S., Chia, L.A., Li, X., Ootani, A., Su, J., Lee, J.Y., Su, N., Luo, Lgr5(+) stem cells by directly suppressing their signature genes. Y., Heilshorn, S.C., Amieva, M.R., et al. (2012). The intestinal stem Nat. Commun. 8, 13824 . cell markers Bmi1 and Lgr5 identify two functionally distinct pop- Roche, K.C., Gracz, A.D., Liu, X.F., Newton, V., Akiyama, H., and ulations. Proc. Natl. Acad. Sci. USA 109 , 466–471 . Magness, S.T. (2015). SOX9 maintains reserve stem cells and pre- You, J., Zhang, Y., Li, Z., Lou, Z., Jin, L., and Lin, X. (2014). serves radioresistance in mouse small intestine. Gastroenterology Drosophila perlecan regulates intestinal stem cell activity via cell- 149 , 1553–1563.e10 . matrix attachment. Stem Cell Reports 2, 761–769 . Rustgi, A.K. (2007). The genetics of hereditary colon cancer. Genes Yousefi, M., Li, N., Nakauka-Ddamba, A., Wang, S., Davidow, K., Dev. 21, 2525–2538 . Schoenberger, J., Yu, Z., Jensen, S.T., Kharas, M.G., and Lengner, Saha, S.K., Choi, H.Y., Kim, B.W., Dayem, A.A., Yang, G.M., Kim, C.J. (2016). Msi RNA-binding proteins control reserve intestinal K.S., Yin, Y.F., and Cho, S.G. (2017). KRT19 directly interacts stem cell quiescence. J. Cell Biol. 215 , 401–413 .

1958 Stem Cell Reportsj Vol. 10 j 1947–1958 j June 5, 2018

Eidesstattliche Erklärung

Hiermit erkläre ich, dass ich die vorliegende Dissertation selbständig verfasst und keine anderen als die angegebenen Hilfsmittel benutzt habe.

Die Dissertation ist bisher keiner anderen Fakultät, keiner anderen wissenschaftlichen Einrichtung vorgelegt worden.

Ich erkläre, dass ich bisher kein Promotionsverfahren erfolglos beendet habe und dass eine Aberkennung eines bereits erworbenen Doktorgrades nicht vorliegt.

Datum: Unterschrift:

CURRICULUM VITAE

Personal data: Name Julien Orlando STEPHAN Primary Address Baustraße 12, 17489 Greifswald (Germany) Telephone +49 17661314872 Email [email protected] Birthday and -place 1st June 1991 in Bethesda, Maryland (USA) Nationality USA, Germany, France

School career: 08/1997 – 07/1998 Primary School Bogenstraße, Solingen (Germany) 08/1998 – 07/2001 Primary School Gräfin-Imma, Bochum (Germany) 08/2001 – 07/2007 Gymnasium Schiller-Schule, Bochum (Germany) 07/2007 – 07/2008 Collegio Loretto, La Paz (Bolivia) 08/2008 – 06/2010 Gymnasium John-Lennon, Berlin (Germany) Abitur 2010 Graduation score: 1.9 focuses on Math and Biology

(US equivalent: high school + one year)

University career: 10/2011 – 11/2018 School of Medicine, Universität Greifswald (Germany) 10/2014 – 10/2015 School of Medicine, University of Pennsylvania (USA)

Received scholarship: 08/2007 – 07/2008 Rotary Scholarship, La Paz (Bolivia)

International experiences: 08/2007 – 07/2008 Rotary Scholarship, La Paz (Bolivia) 08/2011 – 11/2011 School of Medicine, Université de Nantes (France) 10/2014 – 11/2015 School of Medicine, University of Pennsylvania (USA)

CURRICULUM VITAE

Laboratory experience: 02/2014 – 10/2014 Department of Gastroenterology, University of Greifswald (Germany) 10/2014 – 11/2015 Department of Gastroenterology, University of Pennsylvania (USA)

Publication: Giroux, V., Stephan, J., Chatterji, P., Rhoades, B., Wileyto, E.P., Klein-Szanto, A.J., Lengner, C.J., Hamilton, K.E., and Rustgi, A.K. (2018). Mouse Intestinal Krt15+ Crypt Cells Are Radio- Resistant and Tumor Initiating. Stem Cell Rep.

Volunteering: 08/2007 – 07/2008 Social worker in humanitarian affairs, La Paz (Bolivia) 10/2010 – 10/2011 Vivantes Hospital, Berlin (Germany)

Work experiences: 02/2012 – 06/2014 Schlothauer & Wauer (Germany & Austria) - Student assistant in civil engineering 02/2014 – 10/2014 Department of gastroenterology, University of Greifswald - Laboratory technician 05/2016 – 03/2017 Department of pharmacology, University of Greifswald - Medical assistant in clinical trials since 01/2019 Department of gynaecology, University of Greifswald - Resident physician

Language skills: German fluent, mother tongue French fluent, mother tongue English fluent Spanish very good working knowledge

Datum: Unterschrift:

100

DANKSAGUNG

Dank gebührt Prof. Dr. Markus. M. Lerch und den Mitgliedern der wissenschaftlichen Arbeitsgruppe der Inneren Medizin A für meine Einarbeitung in die erforderlichen Labormethoden, meiner kontinuierlichen Unterstützung, wie auch meiner Entsendung in die Arbeitsgruppe von Dr. Anil K. Rustgi. Desweiteren danke ich Dr. Anil K. Rustgi und PhD Veronique Giroux, sowie allen Mitgliedern des Rustgi Labors für die wissenschaftliche Anleitung und stetige Begleitung dieser Experimente. Darüber hinaus danke ich den Mitarbeiter dem Molecular Pathology and Imaging Core, Human Microbial Analytic Depository Core, Cell Culture and iPS Core, Genetic, sowie dem Modified Mouse Core der University of Pennsylvania. Besonderer Dank gilt Dr. George Cotsarelis (University of Pennsylvania) für die Bereitstellung von Krt15-CrePR1 und Krt15-/- Mäusen, wie auch Dr. Christopher Lengner (University of

Pennsylvania) für RosaLSL-tdTomato Mäuse.

101