There are several severe disorders that may affect the. Imaging and Advanced Technology

Size: px

Start display at page:

Download "There are several severe disorders that may affect the. Imaging and Advanced Technology"

Justin Reeves
5 years ago
Views:

1 Imaging and Advanced Technology Ralf Kiesslich and Pankaj Jay Pasricha, Section Editors Assessment of Tumor Development and Wound Healing Using Endoscopic Techniques in Mice MARKUS F. NEURATH,* NADINE WITTKOPF,* ALEXANDRA WLODARSKI, MAXIMILIAN WALDNER,* CLEMENS NEUFERT,* STEFAN WIRTZ,* CLAUDIA GÜNTHER,* and CHRISTOPH BECKER* *Medical Clinic 1, Friedrich-Alexander-University, Erlangen, Germany; and Institute of Immunology, Johannes Gutenberg University, Mainz, Germany Mouse models of intestinal inflammation and colon cancer are valuable tools to gain insights into the pathogenesis of the corresponding human diseases. Recently, in vivo mouse endoscopy has been developed, allowing not only the high-resolution monitoring and scoring of experimental disease development, but also enables the investigator to perform manipulations, including local injection of reagents or the taking of biopsies for molecular and histopathologic analyses. Chromoendoscopic staining with methylene blue enables visualization of the crypt structure and allows discrimination between inflammatory and neoplastic changes. The development of endoscopic techniques in live mice opened new options for the investigation of disease mechanisms in the gut and for the preclinical testing of potential therapeutic effects of drug candidates. Finally, mouse endoscopy can help to reduce animal numbers needed to gain significant experimental data. There are several severe disorders that may affect the intestinal tract in humans. Among these disorders are diseases such as colorectal cancer (CRC), ulcerative colitis, and Crohn s disease whose etiology is incompletely understood. However, these diseases generate economic pressure on the health care systems because of their chronicity, complications, and potentially lethal outcome. In fact, in the United States, CRC is the second most common cancer death of cancers affecting both men and women. 1 Therefore, research on these diseases is essential to provide new approaches for early diagnosis and improved therapy. Much of what we know on the pathogenesis of inflammation and cancer in the gut has come from studies in mouse models of diseases. A broad spectrum of such mouse models is now available to investigate the development of chronic inflammation and cancer in the gut. In addition to their significance in basic research, they are also valuable tools to test drugs candidates in preclinical studies. Experimental models for CRC include the well-established azoxymethane/dextran sodium sulfate (AOM/DSS) model, transgenic mouse models, and adoptive transfer models. 2,3 For the investigation of colitis, experimental models are using either chemical inducers (trinitro benzene sulfonic acid colitis, oxazolone colitis, DSS colitis), genetic mouse models which spontaneously develop intestinal inflammation (eg, interleukin [IL]-10 deficient mice, NF-kappaB essential modulator [NEMO]-deficient mice), or immunologic models based on an adoptive transfer of pathogenic CD4 T cells into RAG-deficient or SCID mice (adoptive transfer colitis). Although there is a large number of animal models mimicking human diseases, until recently, techniques to monitor disease development have been either indirect or where restricted to post mortem analysis. Accordingly, most noninvasive procedures for monitoring mice vitality like animal weight, food and water intake, the presence of blood in the feces, or analysis of blood and urine components are useful but of indirect and variable nature. Only recently direct, noninvasive imaging technologies have been developed for the use in mice, including positron emission tomography and magnetic resonance imaging. 4 They enable the scientist to have a live-look, for example, on tumors, at every required time point during the running experiment. However, these methods are very cost intensive, time consuming, and only give a descriptive and imprecise view on the gut. Recently, a high-resolution endoscopic system for the use in live mice was developed that allows having a direct view on pathologic tissue changes in the gut. 5 This system has been used to determine the presence of colonic inflammation in vivo and a quantitative scoring system for the assessment of colonic inflammation has been developed (the MEICS score). This endoscopy system may be used for quantification of macroscopic colonic inflammation and longitudinal assessment of colitis activity. In addition, this system may be used in murine models of colon cancer and wound healing. Herein, we discuss the possibilities and advantages of high-reso by the AGA Institute /$36.00 doi: /j.gastro GASTROENTEROLOGY 2010;139:

lution mouse endoscopy and related procedures in living mice in the longitudinal analyses of experimental colon cancer and wound healing in the gut.

2 lution mouse endoscopy and related procedures in living mice in the longitudinal analyses of experimental colon cancer and wound healing in the gut. The use of the presented technologies gives the possibility to follow the development of inflammation and tumorigenesis in the colon. In addition, these techniques permit the analysis of changes in the colon over time as well as the assessment of reactions to experimental treatments. 1838

3 Endoscopic Procedure The high-resolution mouse endoscopic system contains of a miniature endoscope with an outer diameter of 1.9 mm. Furthermore, the endoscopic setup is composed of a xenon light source, a digital camera device, and an air pump to provide a continuous air flow necessary for the inflation of the mouse colon (Figure 1A). The camera device can be connected to a personal computer to allow recording of digital videos. Instruments, including a biopsy forceps or an injection needle, can be introduced through the working channel of the endoscope sheath. For the endoscopic procedure, mice were anesthetized by isoflurane. While mice were anesthetized, the miniendoscope was carefully inserted into the mouse colon as far as possible under visual control. The endoscope can usually be introduced about 4 cm using a rigid scope. During the procedure, the colon was continuously insufflated by an air flow to assure a clear analysis and endoscopic pictures of high quality. 6 Longitudinal Tumor Development Using Mouse Endoscopy Despite numerous studies on the etiology of CRC, the mechanisms driving tumor initiation, promotion, and progression remain incompletely understood. Mouse models of CRC have proven to be an indispensable tool for basic research on colon cancer. Furthermore, mouse models allowed the preclinical evaluation of potential drug candidates and treatment regimens. Numerous mouse models of colon cancer have been developed in recent years, including chemically induced CRC models, mutant mouse models, and implantation models. 2,3 The most commonly used model for chemically induced CRC is based on the mutagenic agent AOM, either alone or in combination with DSS. AOM is a carcinogen that is metabolized in the liver and finally by colonic bacteria or intestinal epithelial cells to the unstable methylazoxymethane, whose methylradicales are believed to damage the DNA. The process of tumorigenesis is strongly promoted by DSS, which causes colonic inflammation, probably by direct cytotoxicity to the colonic mucosa or alteration of gut permeability. 7 The combination of AOM and DSS, therefore, has been considered an useful model of colitis-associated colon cancer because it is frequently observed in patients with a history of ulcerative colitis. Mutant mouse cancer models simulate hereditary tumor diseases. The first and most widely used mutant mouse model for colon carcinogenesis is the APC mutant mouse, which carries a mutation in the Apc gene resulting in the spontaneous development of multiple polyps in the small intestine as well as in the colon. As further colon cancer models, xenograft and allograft transplantation of tumor cells offer the possibility to follow the fate of defined human or mouse cancer cells in vivo and allow the preclinical testing of human-specific drug candidates in vivo. Despite these well-established mouse cancer models, in recent years it has been impossible to determine tumor number and size without killing the animals. In humans, endoscopic examination of the colon is the most important method for the discovery of CRC and follow-up care postoperatively. We recently developed an high-resolution endoscopic system for the use in live mice, which not only allows a direct view of colonic tumor development, but also permits longitudinal studies of individual tumors by repeated taking of biopsies at different stages and time points. 5 To demonstrate the usefulness of this in vivo imaging system, colon tumors were induced in mice by intraperitoneal injection of a single dose of AOM followed by 3 cycles of DSS in the drinking water as previously described. 8 For the continuous monitoring of the tumor development, the Coloview high-resolution mouse endoscopic system was used (Figure 1A). Mouse endoscopy allows a detailed and high-quality view of the multiple colonic tumors developing in the AOM/DSS model. It enables the user to count the number of tumors and to distinguish between different tumor sizes and shapes. Tumor burden can be quantified by counting the number of tumors and staging the tumors by size from 1 to 5 (Figure 1B), as previously described in detail. 5,6 In humans chromoendoscopic techniques using dyes like methylene blue have been developed to enhance the contrast of the mucosal surface to detect small, flat lesions. Chromoendoscopy in mice can also be performed using the 4 Figure 1. (A) Pictures of the endoscopic workstation (top) showing in detail the endoscopic instruments [(a) biopsy forceps, (b) endoscope sheath with working channel, (c) injection needle, (d) monitor, (e) camera module, (f) light source and air pump, (g) telescope with camera head] and the anesthetic setup [(h) isoflurane chamber, (i) gas flow tubes, (k) gas mixing chamber]. The endoscope has an outer diameter of 1.9 mm. The air pump provides a continuous air flow necessary for the inflation of the mouse colon. The digital camera is connected to a personal computer to allow recording of endoscopic videos. Instruments, including a biopsy forceps or injection needle, can be introduced through the working channel. (B) Development of tumors monitored by using the mouse endoscopic setup. Different tumor sizes and shapes can be distinguished. Colitis-associated colon cancer was induced by intraperitoneal injection with a single dose of the mutagenic agent AOM followed by 3 cycles of DSS in the drinking water for 1 week and normal drinking water for 2 weeks. For endoscopy mice were anesthetized by a mixture of air and isoflurane using the anesthetic setup and high-resolution colonoscopy was performed, as previously described. 6 Black arrows indicate the tumor. (C) Comparison of normal colonoscopy of a healthy colon (left) and chromoendoscopy (right). Visualization of crypt pattern was achieved by flushing the mouse colon with 500 L of a 1% solution of methylene blue. The arrow indicates an aberrant crypt focus. 1839

4 Coloview system by flushing the mouse colon with methylene blue (Figure 1C). The use of this method makes it possible to detect early flat lesions and aberrant crypt foci, which are not visible without staining (Figure 1C, right). In many studies it is necessary to follow the development of mutations or changes in gene expression profiles in 1 and the same tumor over time. Such an analysis, however, requires access to the tumor and the sampling of tumor material in live mice. Using mouse endoscopy and biopsy forceps easily allows to obtain such samples from the same tumor at different stages and time points in living mice. Accordingly, a flexible forceps can be introduced through the working channel of the endoscope and biopsying may be performed during endoscopy (Figure 2A). Biopsies can be snapped frozen in liquid nitrogen for future analysis. To verify that the obtained biopsies contain neoplastic tissue, biopsy tissue was prepared for histopathologic analysis (Figure 2B). Using tumor samples obtained at different tumor stages during the AOM/DSS model, we could further demonstrate that several genes are regulated in tumor development. Accordingly, the expression of IL-17 and Cyclin D1 was significantly increased is all tumor sizes compared to nontumor tissue (Figure 2C). Strikingly, expression levels were independent of tumor size. Beside biopsy taking during endoscopy, the working channel can be used to directly influence a distinct tumor by applying substances directly into the tumor. Accordingly, a needle attached to a small, flexible tube was moved through the working channel of the endoscope. The needle was introduced directly into the tumor and a small volume of liquid was injected (Figure 2D). Endoscopy-guided injection can be a valuable tool to test therapeutics in preclinical studies. Drug candidates and control substances can be locally delivered into colon tumors within the same mouse, allowing a direct comparison of the efficiency of drug candidates independent of interindividual variability. Furthermore, the procedure enables to inject substances directly into the bowel wall (Figure 2E; Supplementary Video). One possible application is the generation of orthotopic tumor models by injection of human or mouse tumor cells into the colon wall of susceptible mice. Overall, mouse endoscopy together with biopsying offers an adequate method for the longitudinal analysis of experimental colon cancer. High-Resolution Endoscopy for Monitoring of Colitis and Wound Healing In Vivo The described endoscopic procedure is an adequate method for the investigation of colitis and wound healing as well. For the analysis of mucosal damage in the gut, investigators have either used in vitro assay systems using colon epithelial cell lines or used DSS colitis as a model for intestinal wound healing. DSS is a polysaccharide that leads to the destruction of the epithelial layer of the colon. Endoscopically, healthy mice show a normal blood vessel structure and a smooth and translucent mucosal surface (Figure 3A, left). In contrast, mice treated with DSS to induce colitis show multiple signs of mucosal inflammation, as indicated by a thickened bowel wall (less translucent), bleeding, loss of normal blood vessel architecture, fibrin, and diarrhea (Figure 3A). After removal of DSS from the drinking water, epithelial restitution leads to the regeneration of the mucosa. Endoscopically, this regeneration of the mouse colon is indicated by the appearance of a granular colonic surface. Thus, DSS induced experimental colitis has been considered as a wound healing model. However, the interpretation of this model in wound healing analyses is clearly complicated by the concomitant inflammation. Methods for the in vivo analysis of wound healing have been largely restricted to the easily accessible skin. However, high-resolution endoscopy in mice can be used for monitoring of in vivo wound healing in the colon. We and others have previously demonstrated a model that is based on mechanical wounding and is independent of colitis. To generate the wound, a biopsy forceps was inserted into the working channel of the endoscope and introduced into the colon of an anaesthetized mouse. Under visual control, a small piece of mucosa was carefully removed, paying attention not to perforate the colon wall (Figure 3B, left). The process of wound healing was subsequently monitored by daily endoscopies and videos of the procedure were recorded to allow documentation and reevaluation of the wound healing process. Wound repair has been demonstrated to be an event of 4 different, overlapping stages. Shortly after injury, thrombocytes aggregate at the wound bed, generating a fibrin clot to stop the bleeding and to build a first barrier against invading pathogens. The inflammatory phase further removes bacteria and prepares for the next phase by releasing factors that cause the migration and proliferation of cells. The proliferative phase is characterized by epithelialization and wound contraction. Remodelling of the fast but transiently built tissue completes the procedure of wound healing. In agreement with this wound repair model, the formation of a fibrin clot was observed shortly after placing the wound in the in vivo wound healing model described herein (Figure 3B). By performing regular endoscopies, the speed of wound healing can be calculated by comparing the wound diameter at a given time point relative to the initial wound diameter by video analysis (Figure 3B, C, upper left). For further analysis, colon samples including the wound bed can be taken by bioptic sampling and prepared for histological examination by hematoxylin and 1840

5 Figure 2. (A) Stepwise procedure of taking a biopsy from a tumor in a living mouse using a biopsy forceps during mouse colonoscopy. (B) Longitudinal endoscopic and histologic analysis of a single tumor. Colonic tumor development was followed by colonoscopy and biopsies for histopathologic analysis were taken at the indicated time points. Black arrows indicate the tumor. The development of a single tumor over time is shown. (C) Measurement of IL-17 (left) and Cyclin D1 (right) expression in control colonic tissue. DSS-treated inflamed colonic tissue and colon tumors at different stages of tumor development were assessed using real-time polymerase chain reaction. Accordingly, RNA was isolated from tumor samples and real-time polymerase chain reaction was performed. *P.05 compared with control; **P.01 compared with control. Statistical analysis was performed using the 1-sided Student t-test. (D) Injection of a fluidic substance into a tumor during mouse colonoscopy using an injection needle coupled with a syringe. (E) Stepwise procedure of submucosal injection of reagents. 1841

6 Figure 3. An endoscopy-guided, in vivo wound healing model. (A) Representative endoscopic pictures of a healthy mouse and from mice with colitis showing various signs of inflammation, including loose stool, an altered vascular pattern, a bleeding mucosa, and abundant fibrin. (B) Generation of a wound in the mucosal surface by using a biopsy forceps during mouse colonoscopy (left) and its appearance on days 0, 1, 3, and 6. The fresh wound bed usually had a diameter of approximately 800 m. The black arrow indicates the wound bed. (C) The diagram shows an analysis of wound healing. The size of the wound bed was calculated relative to the diameter of the initial wound bed on day 0; histologic examination of the wound bed from a wild type mice on day 2 after the wounding by cryosections (stain: hematoxylin and eosin and myeloperoxidase antibodies; original magnification, left,4 ; right,10 ). eosin staining and immunohistochemical analysis. For instance, an immunohistochemical staining of the wound bed at day 2 indicated the appearance of the inflammatory cells. Specifically, staining for the inflammation marker of active myeloperoxidase showed the attraction of granulocytes to the site of injury (Figure 3C). Macroscopically, wound healing was completed within 6 days after wounding, as determined by in vivo endoscopy. The endoscopic wound healing model described here uses a biopsy forceps to generate a mechanical wound, is independent of inflammation and has no need for additional, chemically induced tissue damage. In conclusion, numerous mouse models of intestinal inflammation and colon cancer have been developed. Mouse endoscopy is a fascinating technique allowing insights into the pathogenesis of these experimental disease models. In addition to monitoring disease development, manipulations including local injection of reagents or the taking of biopsies can be performed. Biopsies can be used for histochemical and molecular biological analysis, thereby allowing longitudinal analysis of experimental disease in 1 and the same mouse at different stages of disease development. Endoscopic techniques are also useful for the development of novel disease models. The endoscopy guided in 1842

7 vivo wound healing model in the colon described here can be useful for the investigation of wound healing independent of colitis. Supplementary Material Note: The first 5 references associated with this article are available below in print. The remaining references accompanying this article are available online only with the electronic version of the article. To access the remaining references, as well as additional online-only data, visit the online version of Gastroenterology at and at doi: /j.gastro References 1. American Cancer Society. Cancer facts and figures Atlanta: American Cancer Society; Taketo MM. Mouse models of gastrointestinal tumors. Cancer Sci 2006;97: Heijstek MW, Kranenburg O, Borel Rinkes IHM. Mouse models of colorectal cancer and liver metastases. Dig Surg 2005;22: Lewis JS, Achilefu S, Garbow JR, et al. Small animal imaging: current technology and perspectives for oncological imaging. Eur J Cancer 2002;38: Becker C, Fantini MC, Witz S, et al. In vivo imaging of colitis and colon cancer development in mice using high resolution chromoendoscopy. Gut 2005;54: Reprint requests Address requests for reprints to: Christoph Becker, PhD, Department of Medicine 1, University Clinic, Friedrich-Alexander University Erlangen-Nuremberg, Erlangen, Germany. christoph.becker@uk-erlangen.de; fax: (0049) Acknowledgments M.F.N. and N. W. share first authorship. Funding The research leading to these results has received funding from the European Community s 7th Framework Programme (FP7/ ) under grant agreement n , acronym GENINCA. 1843

8 1843.e1 Imaging and Advanced Technology GASTROENTEROLOGY Vol. 139, No. 6 References (Online Only) 6. Solomon L, Mansor S, Mallon P, et al. The dextran sulphate sodium (DSS) model of colitis: an overview. Comp Clin Pathol 2010;19: Wirtz S, Neufert C, Weigmann B, et al. Chemically induced mouse models of intestinal inflammation. Nat Protoc 2007;2: Seno H, Miyoshi H, Brown SL, et al. Efficient colonic mucosal wound repair requires Trem2 signaling. Proc Natl Acad Sci U S A 2009;106: Pickert G, Neufert C, Leppkes M, et al. STAT3 links IL-22 signaling in intestinal epithelial cells to mucosal wound healing. J Exp Med 2009;206:

Chromoendoscopy and Endomicroscopy for detecting colonic dysplasia

Chromoendoscopy and Endomicroscopy for detecting colonic dysplasia Ralf Kiesslich I. Medical Department Johannes Gutenberg University Mainz, Germany Cumulative cancer risk in ulcerative colitis 0.5-1.0%