Dysregulation of retinoid biosynthesis and metabolism during colon carcinogenesis

The research work presented in this dissertation describes the dysregulation of the retinoid biosynthetic and metabolic pathways accompanying colon carcinogenesis. Colon cancer remains the third leading cause of cancer deaths in the United States. A better understanding of the molecular mechanisms underlying colon tumorigenesis is needed to develop more effective therapies against colon cancer. Chapter 1 is an introduction on colon cancer and retinoids. The relationship between normal intestinal development and progression of colon cancer, including a background on adenomatous polyposis coli (APC) and several colorectal cancer syndromes are presented in the first half of Chapter 1. The latter part is a discussion of retinoid storage, biosynthesis and metabolism. It also summarizes what has been reported about the role of retinoids in normal colon function and their therapeutic effects on various human tumors. Retinol dehydrogenase like (DHRS9) is a colon-specific retinol dehydrogenase that is down-regulated in colon carcinomas. Chapter 2 describes studies that were undertaken to confirm the retinol dehydrogenase activity of DHRS9, demonstrate impairment of the retinoid biosynthetic capabilities of colon carcinoma cells and investigate the possible control of retinoic acid biosynthesis by APC through regulation of DHRS9 expression. A brief description of the retinol dehydrogenase activity of three zebrafish retinol dehydrogenases---rdh1, rdh1l and rdh5, is presented in Chapter 3. A discussion on the regulation of retinoid biosynthesis by APC and the importance of retinoic acid in zebrafish eye and gut development is also included. This is followed by Chapter 4, which describes significant differences in the dysregulation of the retinoid biosynthetic pathway in human colon carcinomas compared to the Apc[min/+ mouse, an established animal model system for studying colon cancer. Chapter 5 shows that the dysregulation of retinoid biosynthesis at multiple points along the pathway is a common occurrence in human colon carcinomas and colon carcinoma cell lines. Studies on cellular retinol binding protein II (CRBPII), its importance in normal intestinal development and retinol uptake are presented as well. Chapter 5 also discusses a broader role for APC in regulating intestinal differentiation by controlling several retinoid biosynthetic components. Conclusions and implications of presented findings are discussed in Chapter 6.

DYSREGULATION OF RETINOID BIOSYNTHESIS AND METABOLISM DURING COLON CARCINOGENESIS by Imelda Tumanan Sandoval A dissertation submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Medicinal Chemistry The University of Utah May 2006 Copyright © Imelda Tumanan Sandoval 2006 All Rights Reserved THE UNIVERSITY OF UTAH GRADUATE SCHOOL SUPERVISORY COMMITTEE APPROVAL of a dissertation submitted by In1elda Tumanan Sandoval This dissertation has been read by each member of the following supervisory committee and by majority vote has been found to be satisfactory_ Dale I r I Co-chair: Chris M. Ireland ~C.l~"o.../~ _ ~ es A. McCloskey §/? ~ ~~~~ Eric W. Schmidt THE UNIVERSITY OF UTAH GRADUATE SCHOOL FINAL READING APPROVAL To the Graduate Council of the University of Utah: I have read the dissertation of Imelda Tumanan Sandoval in its final form and have found that (1) its forn1at, citations, and bibliographic style are consistent and acceptable; (2) its illustrative materials including figures, tables, and charts are in place; and (3) the final manuscript is satisfactory to the supervisory committee and is ready for submission to The Graduate School. ~!t LOft; , I D~ Date Chair: Supervisory Committee Approved for the Major Department Chair Approved for the Graduate Council ABSTRACT The research work presented in this dissertation describes the dysregulation of the retinoid biosynthetic and metabolic pathways accompanying colon carcinogenesis. Colon cancer remains the third leading cause of cancer deaths in the United States. A better understanding of the molecular mechanisms underlying colon tumOIigenesis is needed to develop more effecti ve therapies against colon cancer. Chapter 1 is an introduction on colon cancer and retinoids. The relationship between normal intestinal development and progression of colon cancer, including a background on adenomatous polyposis coli (APC) and several colorectal cancer syndormes are presented in the first half of Chapter 1. The latter pati is a discussion of retinoid storage, biosynthesis and metabolism. It also summarizes what has been reported about the role of retinoids in normal colon function and their therapeutic effects on various human tumors. Retinol dehydrogenase like (DHRS9) is a colon-specific retinol dehydrogenase that is down-regulated in colon carcinomas. Chapter 2 describes studies that were undertaken to confirm the retinol dehydrogenase activity of DHRS9, demonstrate impairment of the retinoid biosynthetic capabilities of colon carcinoma cells and investigate the possible control of retinoic acid biosynthesis by APC through regulation of DHRS9 expression. A brief description of the retinol dehydrogenase activity of three zebrafish retinol dehydrogenases - rdhl, rdhll and rdh5, is presented in Chapter 3. A discussion on the regulation of retinoid biosynthesis by APC and the importance of retinoic acid in zebrafish eye and gut development is also included. This is followed by Chapter 4, which describes significant differences in the dysregulation of the retinoid biosynthetic pathway I,n human co Ion ca'rCIn omas compare d to t he A pc 1//;/1/+ mouse, an esta bl'I S he d anI' ma I mo de I system for studying colon cancer. Chapter 5 shows that the dysregulation of retinoid biosynthesis at multiple points along the pathway is a common occun-ence in human colon carcinomas and colon carcinoma cell lines. Studies on cellular retinol binding protein II (CRBPII), its importance in normal intestinal development and retinol uptake are presented as well. Chapter 5 also discusses a broader role for APC in regulating intestinal differentiation by controlling several retinoid biosynthetic components. Conclusions and implications of presented findings are discussed in Chapter 6. v To my mother, Luisa, who is the most beautiful person I know To my husband, Richard, for always believing in me TABLE OF CONTENTS AB S'I'RA CT ....................................................................................................................... i v LIST OF TABLES ............................................................................................................. ix LIsrr OF FIG lTRES ............................................................................................................ x LIST OF ABBREVIATIONS ........................................................................................... xii NOMENCLATURE ..................................................................................................... xviii ACKNOWLEDGMENTS ............................................................................................... xix Chapter 1 INTRODUCTION ....................................................................................... 1 1.1 Colon cancer .............................................................................. 1 1.2 Retinoids .................................................................................. 14 1.3 Project Rationale ...................................................................... 24 1.4 References ................................................................................ 26 2 CHARACTERIZATION OF A NOVEL COLON-SPECIFIC RETINOL DEHYDROGENASE ............................................................. 33 2.1 Introduction .............................................................................. 33 2.2 Methodology ............................................................................ 35 2.3 Results ...................................................................................... 39 2.4 Discussion ................................................................................ 44 2.5 References ............................................................................... 47 3 STUDIES CHARACTERIZING ZEBRAFISH RETINOL DEHYDROGENASES .............................................................................. 52 3.1 Introduction .............................................................................. 52 3.2 Methodology ............................................................................ 54 3.3 Results ...................................................................................... 55 3.4 Discussion ................................................................................ 56 3.5 References ............................................................................... 59 4 INVESTIGATION OF THE RETINOID BIOSYNTHETIC PATHWAY IN A]JC lllilll+ MOUSE ................................................................................... 61 4.1 Introduction .............................................................................. 61 4.2 Methodology ............................................................................ 62 4.3 Results ...................................................................................... 65 4.4 Discussion ................................................................................ 68 4.5 References ................................................................................ 72 5 CRBPII IS DOWN-REGULATED IN HUMAN COLON CARCINOMAS AND ESSENTIAL FOR INTESTINAL DIFFERENTIATION IN ZEBRAFISH .............................................................................................. 74 5.1 Introduction .............................................................................. 74 5.2 Methodology ............................................................................ 76 5.3 Results ...................................................................................... 82 5.4 Discussion ................................................................................ 92 5.5 Contributors ............................................................................ 95 5.6 References ................................................................................ 96 6 CONCLUSIONS ...................................................................................... 100 APPENDIX: SUPPLEMENTARY DATA ........................................................ 102 LIST OF PUBLICATIONS ................................................................................ .1 05 Vlli LIST OF TABLES Table page 4.1 Primers and annealing temperatures used for RT -PCR of mouse tissue samples ....................................................................................................... 64 5.1 Primers and annealing temperatures used for RT-PCR of human colon tissue samples and cell lines ...................................................................... 78 5.2 Primers and annealing temperatures used for RT -PCR of mouse and zebrafish tissue samples ............................................................................. 79 LIST OF FIGURES Figure page 1.1 The normal architecture of the colon .......................................................... .4 1.2 Schematic representation of the APC protein .............................................. 6 1.3 The function of APC as a tumor suppressor ................................................ 9 1.4 Genetic changes accompanying colon cancer progression ........................ 11 1.5 Natural and synthetic retinoids ................................................................. .15 1.6 Retinoid biosynthesis and metabolism ....................................................... 18 1.7 Retinoids in clinical studies ....................................................................... 22 1.8 Retinoic acid metabolism blocking agents ................................................ 25 2.1 DHRS9 and RDH5 are down-regulated in colon adenomas and carcinomas ................................................................................................. 40 2.2 Down-regulation of DHRS9 and RDH5 in colon carcinoma cell lines is accompanied by poor conversion of retinol to RA ................................... .42 2.3 DHRS9 metabolizes retinol to RA in colon tumor cells ........................... .43 2.4 Re-introduction of APC induced DHRS9 expression and improved RA biosynthesis ................................................................................................ 45 3.1 rdh1, rdh11 and rdh5 convert retinol into retinoic acid in HCT116 cells ... 57 4.1 Expression level of alcohol dehydrogenases in ApClllilll + mouse ................ 66 4.2 Expression level of short chain alcohol dehydrogenases in ApC'llill /+ mouse ........................................................................................... 67 4.3 E xpressl.o n 1e ve 1 0 f retm. a 1 de h yd rogenases I. n A pcm ill/+ mouse ................ .. 69 4.4 E xpressJ.o n ]e veI o·f re'tm ol. d nuc I ear receptors m. A pc1 11111/+ mouse ............ .7 0 4. 5 E xpressl.o n Ie ve I 0 f~ C yp 26A l'm A pcl Ilill/+ mouse ....................................... 71 5.1 Retinoid genes are dysregulated in colon adenomas and carcinomas ....... 84 5.2 Retinoid genes are dysregulated in colon carcinoma cell lines ................. 85 5.3 RBP2a is down-regulated in APC-deficient model systems ..................... 87 5.4 ape mutant zebrafish show impaired retinol uptake .................................. 88 5.5 rbp2a is involved in retinol uptake in zebrafish ........................................ 90 5.6 rbp2a morphants exhibit an RA-deficient phenotype ................................ 91 5.7 rbp2a is important for normal intestinal development in zebrafish ........... 93 A.l HPLC profi Ie of retinoid standards .......................................................... 103 A.2 Phylogenetic analysis of the murine retinol dehydrogenase enzyme family ..................................................................................................... 104 Xl ~-gal p,Ci p,M p,mol 4-hpr A A ADH ALDH AML AOM APC APL Asef ATCC ATRA bcox BHT BMP LIST OF ABBREVIATIONS ~-galactosidase microCurie micromolar micromole N-(4-hydroxyphenyl) retinamide adenoma alanine alcohol dehydrogenase aldehyde dehydrogenase acute myelogenous leukemia azoxymethane adenomatous polyposis coli acute promyelocytic leukemia APC-stimulated guanine nucleotide exchange factor American Type Culture Collection all-trans retinoic acid ~,~-carotene-15, 15' -oxygenase butylated hydroxy toluene bone morphogenetic protein BSA C - terminus C CAT eDNA CHRPE CO2 CRABP CRAD CRALBP CRBP D DHRS9 DMSO DNA dNTP EBI EGFR F FAP FDA flo bovine serunl albumin cat"box y termin us Celsius chloramphenicol acetyltransferase complementary deoxyribonucleic acid congenital hypertrophy of the retinal pigment epithelium carbon dioxide cellular retinoic acid binding protein cis-retinol/androgen dehdyrogenase cellular retinal binding protein cellular retinol binding protein aspm1ic acid dehydrogenase/reductase member 9 dimethyl sulfoxide deoxyribonucleic acid deoxyribonucleotide triphosphate end-binding protein 1 endothelial growth factor receptor phenylalanine familial adenolnatous polyposis Food and Drug Administration flotte lotte Xlll G GAPDH GSK3p h H HC] HDAC hDLG HMEC HMG HNPCC hpf HPLC HSD iFABP JPS K Kap3 kDa L LEF LRAT glycine glyceraldehyde 3-phosphatase dehydrogenase glycogen synthase-3p kinase hour histidine hydrochloric acid histone deacetylase human discs large protein human mammary epithelial cell high mobility group hereditary non-polyposis colorectal cancer hours post fertilization high performance liquid chromatography h ydrox ysteroi d deh ydro genase intestinal fatty acid binding protein juvenile polyposis syndrome lysine kinesin associated protein 3 kiloDalton leucine lymphoid enhancer factor lecithin :retinol acyl transferase XIV MCR MeOH MgC}z min MIN MMR MO mRNA MUT N terminus N N NADP nls nof P PBS PCR pie PJS PP2A PTP-BL mutation cluster region methanol magnesium chloride minute multiple intestinal neoplasia mismatch repair morphant messenger RNA mutant amino terminus normal asparagine nicotine amide dinucleotide phosphate neckless no fin proline phosphate-buffered saline polymerase chain reaction pie bald Peutz-Jeughers syndrome phosphatase 2a protein tyrosine phosphatase-basophil-like xv RA RALDH RAMBA RAR RBP RDH RDHL RNA RPE rpm rRNA RT-PCR RXR s S SABA SAMP SD SDR slj SMAD SNP retinoic acid retinaldehyde dehydrogenase retinoic acid metabolism blocking agent retinoic acid receptor retinol binding protein retinol dehydrogenase retinol dehydrogenase like ribonucleic acid retinal pigment epitheliun1 revolutions per minute ribosomal RNA real ti me-PCR retinoid X receptor second serine suberoylanilide hydroxamic acid serine-alanine-methionine-proline standard deviation short chain alcohol dehydrogenase/reductase slim jim Sma- and Mad- related protein single nucleotide polymorphism XVI sst TBE TCF TOF TTNPB UPAR VAD VEOFR WT y straight shot tracheobronchial epithelial T cell factor transforming growth factor 4-(E-2-(5,6,7 ,8-tetrahydro-5,5,8,8-tetramethyl-2-naphthalenyl)-1- propenyl) benzoic acid urokinase plasminogen activator receptor vitamin A deficiency vascular endothelial growth factor receptor wild type tyrosine XVll NOMENCLATURE GENE human gene Gene mouse gene gene zebrafish gene GENE human gene product Gene mouse gene product gene zebrafish gene product ACKNOWLEDGMENTS First of all, I would like to express my sincerest gratitude to my advisors, David Jones and Chris Ireland, for their support, guidance and patience. Dave and Chris are not only brilliant scientists but also excellent teachers who are generous with their time, knowledge and kind words. I would also like to thank my thesis committee - James McCloskey, Glenn Prestwich and Eric Schmidt for providing additional guidance and support. Our meetings have always been encouraging. I am thankful to the Jones lab and the Ireland lab, both past and present members, for their friendship and words of encouragement. I would like to especially acknowledge those who not only contributed to this project, but who also shared their expertise as well - Cicely, Dawne, Lindsay, Eryn, Betsy, Stephanie, Lincoln, Rohan, Camille, Anokha, and Tim. I would like to recognize the Core Facilities at the University of Utah for their excellent service, especially Donna and Talea at the HCI Animal Resource Facility for taking good care of my mice and always being so helpful. In addition, I would like thank the Huntsman Cancer Institute for providing a very stimulating working environment. I would also like to thank the National Institutes of Health, American Cancer Society and Huntsman Cancer Foundation for funding. I am grateful to Gisela Concepcion, for showing me the passion and the determination to succeed in science. My experience in her lab has helped me tremendously in graduate school. I am also indebted to Manny and Lita Imperial, Pete and Pitang Raneses, Chris Ireland and Mary Kay Harper who have been our parents away from honle. I will never forget their kindness and generosity. I would like to give my heartfelt appreciation to my ever growing family for their incredible support and constant prayers. My nephews and nieces - Mica, Trina, Mark, Ap-ap, An-an, 1sak and Francis for inspiring me to a be good role model for them; my brothers and sisters - Tony and Susan, Alan and Tess, Josie, Archie and Susan, Manny, Adora and Chris, their stories made me feel that home is much closer than I thought; my parents - Luisa and Jose for their unconditional love and support, thanks for letting me fol1ow my dream and my heart; and of course Richard for always making me laugh. Thanks for making this ride as fun and exciting as it can possibly be © Finally, I would like to give thanks to the Lord Almighty, I am so humbled by all His blessings. And to everyone and anyone who has helped me get here, in one way or another, maraming maraming salamat. xx CHAPTER 1 INTRODUCTION 1.1 Colon Cancer Colorectal cancer is the third leading cause of cancer death ]n the United States. Approximately 150,000 new cases will be diagnosed and 60,000 deaths will be reported in 2005 (1). Survival rate is high when the cancer is detected at an early stage. When the neoplasia is localized to the colon and colon wall, survival rate is 900/0 (1,2). However, when the cancer cells have spread to nearby tissues and metastasized to organs such as the liver and lung, the rate of survival is dramatically reduced to 10% (1,2). Unfortunately, the majority of colon tumors are detected at later stages due to lack of better screening procedures (2). At present, the best way to check for presence of cancerous polyps is to undergo a colonoscopy; it involves inserting a flexible, lighted tube through the rectum to visually inspect the entire length of the colon and rectum, a procedure that many patients find invasive and uncomfortable. There are three major treatment options for persons afflicted with colon cancer: surgery, radiation therapy, and chemotherapy (2). Surgery, the removal of cancerous tissue in an operation, is the most common treatment for colon cancer and is most effective when the cancer is at its early stages. Radiation therapy and chemotherapy, treatments that involve high energy rays and drugs, respectively, are often used after surgery to eliminate any remaining cancer cells, and are particularly helpful when the 2 cancer has become metastatic. Commonly used chemotherapeutic drugs include fluorouracil, leucovorin, oxaliplatin and irinotecan (3). The Food and Drug Administration (FDA) has also recently approved the use of the monoclonal antibodies, Avastin (bevacizumab) and Erbitux (cetuximab), to treat advanced colorectal cancer (4). Bevacizumab binds to the vascular endothelial growth factor while cetuximab binds to the epidermal growth factor receptor, resulting in reduced cellular proliferation and vascular growth (4,5). However, both drugs can cause seIious side effects including gastrointestinal perforation, hemonhage, highllow blood pressure and difficulty in breathing (4). While colon cancer incidence and mortality is in a decline, a more effective screening and therapy that would detect and treat the disease at its earliest stages remain an unmet need (1). Therefore, a better understanding of the mechanisms underlying colon tumorigenesis as well as improved diagnosis and treatment are needed to further minimize mOltality and morbidity from colorectal cancer. 1.1.1 Normal Intestinal Development The intestinal tract is tasked with absorbing and digesting nutrients from ingested food and eliminating materials that are toxic and unusable by the body (6,7). The small intestine is the primary site for absorption and digestion, aided by the presence of pancreatic enzymes and bile, while the colon is where the final phase of digestion occurs - the fonnation and storage of feces, which involves recovery of water and electrolytes and addition of mucus and bacteria (6). The intestinal gut tube is organized into four layers: the outermost loose connecti ve tissue layer or mesenchyme, which provides structural support and growth signals to 3 overlying epithelium, followed by a double layer of smooth muscle that imparts motility on the intestinal tube, a connective tissue layer containing blood and lymphatic vessels and finally, the innermost epithelial cell layer that lines the lumen and where nutrient absorption takes place (Fig. 1.1) (6,8). This single layer of epithelial cells have microvilli at their apical surface facing the lumen to allow them to digest and absorb nutrients while keeping indigestible materials and associated microflora inside the lumen (6,9). In the colon, normal epithelium is fil1ed with invaginations called crypts, where cell proliferation, migration, differentiation and apoptosis are tightly regulated (Fig. 1.1) (9- 13). Stem cells, which proliferate at the base of the crypts, differentiate into one of the four mature cell types as they migrate up the crypt axis: colonocytes, enteroendocrine cells, goblet cells and paneth cells (6,8,10-12,14). The absorptive epithelial cells, or colonocytes, make up the majority of the cells in the intestinal epithelium. They line the gut tube, are highly polarized, and are responsible for taking up nutrients from the lumen and carrying them to the underlying connective tissue (6,7). The enteroendocrine cells are scattered throughout the epithelial cell layer~ they release hormones that regulate the release of digestive enzymes such as bile and bicarbonate-rich fluids from the pancreas and liver into the intestines (6,7). Goblet cells are spread among colonocytes as well, and they secrete lubrication that not only facilitates the movement of intestinal contents but also protects the gut tube from shear stress and toxins (6,7,9). Paneth cells do not nligrate upward and are mostly found at the bottom of the crypts where they primarily function as a protection for the proliferating stem cells by producing antimicrobial factors such as cryptidin, defensins and lysozymes to control the microbial content of the intestine (6,7,9,12,13,15). Shedding Growth arrest and Diffe renti ati on Clonal proliferation Connective tissue Lumen Colonocytes, Goblet cells, Enteroendocrine cells Stem cells, P aneth cells MUSCle~ 4 FIGURE 1.1 The normal architecture of the colon. The intestinal epitheliUlTI is organized into crypts and supported by the underlying mesenchyme consisting of connective tissue and muscle layers. At the bottom of the crypts are proliferating stem cells, which differentiate into four mature cell types: paneth cells, enteroendocrine cells, goblet cells and colonocytes. Differentiation takes place as the cells move upward, followed by apoptosis at the top of the crypt where cells are shed into the intestinal lumen. 5 Once the differentiated cells reach the top of the crypt they are eliminated by apoptosis and are sloughed off into the lumen 00-12). This cycle of cell proliferation, differentiation, and apoptosis in the intestinal epithelium is a continuous process brought about by the constant exposure of the colon to toxins and stressful environment. It also helps in maintaining the normal architecture of the colonic crypt (11). The highly proliferative progenitor stem cells give rise to daughter cells that undergo only three to four cell di visions before they terminally differentiate into one of the mature cell lineages (6,9). Once matured, these cells have 3-5 days to move up the crypt before becoming senescent (6,16). When this delicate balance between cell proliferation and cell death is disrupted, it can give rise to the fOlmation of colon adenomas and carcinomas. Uncontrolled cell division leads to the expansion of the proliferative zone until it encompasses the whole crypt, resulting in the formation of abenant crypts, which appear larger and thicker than normal, tend to aggregate in clusters, and are believed to be the precursors of adenomatous colon polyps (10,17). This alteration of cell proliferation in the colon crypts has been reported to be one of the earliest events in colorectal tumorigenesis and is believed to be accompanied by a corresponding decrease in the differentiative and apoptotic activities within the crypt (10,11,17). 1.1.2 Adenomatous Polyposis Coli (APC) The dysregulation of cellular proliferation in neoplastic colonocytes has been linked to mutations in the adenomatous polyposis coli (APe) gene. APC is a multifunctional, multi domain protein consisting of 2843 amino acids with a molecular mass of about 300 kDa (Fig. 1.2) (18-20). It spans 21 exons, with ex on 15 comprising more than 750/0 of the 6 6 57 453 767 1020 11701265 2035 2200 2400 2559 2771 ! ! ! ! c,i\,:(;i'n~~f l:l t >.:.:''i Armadillo 15-aa 20-aa SAMP Repeats Basic POZ Repeats Repeats Repeats Domain Domain ASEF p-catenin ~-cateninl AXIN Microtubule EB1 HLDG GSK3 P ."" ""."""",,,, ,_'t" ~' ,,',' < ~ , , ~ ·l~~,,!!.llilltii.,: \l'1>!!flillil1li\i\ ~I~!~~ J~~~'~\l~~: f • ~ ~ ,~l~ '" ~ • '" II. :Ifi·, , FIGURE 1.2. Schematic representation of the APe protein. APC is a multi domain protein (1st and 2nd rows) that is reported to interact with several binding partners (3rd row) implicating APC in a variety of cellular functions (4th row). Codon nurnbers are shown at the top of the figure (6, 19, 21, 35). 7 APC coding sequence (18,21). APC is expressed within the normal colonic epithelium, specifically at the basolateral membrane in colonocytes, with expression beconling more distinct as the cells move up the crypt (21-23). It is present both in the nucleus and cytoplasm, and in spite of its size, it is a highly mobile protein, shuttling in and out of the nucleus, and also to the cell periphery (21,22,24). The N-terminus of the APC protein is well-conserved among different species, has a high degree of structural organization, and contains nuclear export signals required to transport APC inside and outside of the nucleus (6,22). It also includes an oligomerization domain containing heptad repeats that can form coiled-coil domains and allow APC to fOlm homodimers, and an armadillo region that has been reported to interact with several proteins including phosphatase 2a (PP2A), APC-stimulated guanine nucleotide exchange factor (Asef), and the kinesin linker protein, Kap3 (6,21-23). The central region of the APC protein contains the domains responsible for interacting with the Wnt signaling pathway proteins. This stretch of the APC molecule contains three 1 and seven 20-amino acid repeats that both bind to ~-catenin, with the latter being regulated by phosphorylation and marks ~-catenin for degradation (6,21,22). Also scattered among the amino acid repeats are SAMP repeats that are involved in axin binding (6,21). The C-terminal of APC is the least conserved region between species and lacks the well-ordered structure found at the N-terminus (6). It contains a basic domain enriched in lysine and arginine amino acids, with an unusually high percentage of proline residues as well, that constitutes the major microtubule-binding site of APC (6,21). In addition, the C-terminal fragment also binds to EE 1, a small microtubule end binding protein, and holds the PDZ binding motif that allows the interaction of APC with HDLG, 8 the human homologue of the Drosophila discs large tumor suppressor gene, and the protein tyrosine phosphatase, PTP-BL (6,21-23). The best characterized function of APC is its negative regulation of ~-catenin, a cytoplasmic protein that is originally reported to interact with E-cadherin, a-catenin, y­catenin and participate in cell adhesion (6,24-26). Recently, ~-catenin has been found to be a key effector of the Wnt signaling pathway (Fig. 1.3) (27,28). Fully functional wild type APC forms a complex mainly with axin, PP2A, and glycogen synthase-3~ kinase (GSK3P) that sequentially phosphorylates ~-catenin at highly conserved serine and threonine residues near its amino terminus, targeting ~-catenin for ubiquitin-mediated proteolysis and thus controlling the amount of p-catenin available for transcriptional activation (6,7,21,22,29-31). However, in cells harboring APC mutations or in the presence of a Wnt signal, ~-catenin accumulates in the cytosol, translocates to the nucleus, and interacts with high mobility group (HMG)-box transcription factors, T cell factor (TCF) and lymphoid enhancer factor (LEF). This results in the upregulation of proliferative oncogenes like c-myc and cyclin Dl, genes that control cell cycle progression and can promote neoplastic growth (32,33). Other important targets of P­catenin/ TCF complex include ephrins and their receptors, caspases, c-jun, fra-1, matrylysin, fibronectin, CD44 and uPAR, proteins that are involved in cell-cell interaction, apoptosis, cellular proliferation and invasive growth (6,26,34). Recent studies have elucidated additional roles for APC. It has been implicated in maintaining actin cytoskeletal integrity and ensuring the upward migration of cells in colonic crypts (6,19,22). APC has also been shown to playa role in maintaining fidelity of chromosomal segregation by binding and stabilizing microtubules and contributing to Neighboring cell Cytoskeleton -. • ... • l3-catenin degradation Neighboring cell 9 FIGURE 1.3. The function of APe as a tumor suppressor. Wild type APC forms a complex with axin and GSK3-~ to target ~-catenin for degradation by proteosomes (left). When APC is mutated, cytoplasmic f3-catenin level increases, enabling it to move into the nucleus to tUI11 on the expression of target genes (right)( adapted with pennissioll from Macnlillan Publishers Ltd: [NatureJ(31 ) copyright 2005). 10 mitotic spindle fonnation and function during mitosis (6,19,35,36). Last, it has been proposed that APC can function as a nuclear-cytoplasmic shuttling protein, as supported by the presence of several nuclear export signals located at the amino and central regions of the APC protein (6,22). It is widely accepted that colon cancer develops from normal epithelium to a metastatic carcinoma through a multistep progression of malignant transformation by acquiring inactivating and activating mutations in tumor suppressor genes and oncogenes, respectively (23,37,38). In a model proposed by Fearon and Vogelstein to elucidate genetic events underlying colon cancer development (Fig. 1.4), mutations in the APe gene are sufficient to initiate colon carcinogenesis (38). This is then followed by a progression from adenoma to carcinoma, a process accompanied by sequential mutations of multiple genes including K-ras, p53 and mismatch repair genes (MMR) , as well as epigenetic alterations such as CpG methylation that confer growth and survival advantage to neoplastic cells (23,37,38). Indeed, APe mutations are found in 80% of sporadic colorectal tumors and in all cases of familial adenomatous polyposis (FAP), an inherited predisposition to colorectal cancer (20,23,37). It is also present in the ear1iest neoplastic lesions that can be examined, the abenant crypt foci, further confirming that loss of APC function is an early event in colon tumorigenesis (7,9,37). About 95% of APe mutations found in patients are nonsense or frameshift mutations that result in the expression of a truncated APC protein lacking the C-terminal fragment (6,18,20,39). A study on the distribution of APe mutations reveals that most of the defects fall in the first half of the APe gene, and mutations beyond codon 1600 do not have a severe effect on protein function (20). Also, about 60% of somatic APe mutations 11 APC K-RAS DCC/DPC4IJV18? p53 other Changes? Normal I Dysplastic__ Early I ·lntermediate I· Late I I . EPithelium.J.,.... ACF ~ Adenoma ..1.,.. Adenoma ..1.,.. Adenoma ..1.,.. Carcinoma ..L,. Metastasis @O.fiCiency FIGURE 1.4. Genetic changes accompanying colon cancer progression. Colorectal carcinogenesis is initiated by mutations in APe. This is followed by a progression from adenoma to carcinoma accompanied by genomic instability and accumulated mutations in several oncogenes and tumor suppressor genes (adapted from reference 23, copyright (] 996), with permission from Elsevier). 12 are contained within the mutation cluster region (MCR), spanning codons 1286-1513, which contains the binding sites for axin and all but one ~-catenin binding site (18,20,21). Loss of APC function affects a variety of cellular processes including cytoskeletal regulation and chromosomal segregation. However, the regulation of Wnt/~-catenin­mediated gene activation seems to be the primary tumor suppressor function of APC in cancer. When functioning normally, Wnt genes are involved in controlling cell fate and positioning along the crypt axis, but when constitutively activated, the Wnt/~-catenin pathway leads to uncontrolled proliferation and transformation of colonocytes in the intestinal crypts that upsets the balance between cell proliferation, differentiation and cell death (9,15,31,40,41). However, this being said, several groups have shown that other cellular functions are affected when APC is inactivated that are independent of ~-catenin. It is reported that APC knock down in murine gut epithelium led to dramatic changes in tissue organization particularly, a lack of migration and differentiation (16). Tumors with fJ-catenin mutations are less aggressive and smaller in size compared to tumors arising from APC mutations and mutations in fJ-catenin are found only in about 250/0 of colorectal carcinomas (6,20,35). Furthermore, nuclear accumulation of ~-catenin is not consistent in human colorectal tumors and does not always associate with areas of high proliferative activity (6,34). These suggest that loss of ~-catenin does not equal loss of APC and support the idea that APC must play an additional role that is important for normal intestinal development. 1.1.3 Co]orectal Cancer Syndromes Familial adenomatous polyposis (FAP) is an autosomal dominant colorectal cancer syndrome that was first decribed by Lockhart-Mummery in 1925 (20). It is caused by 13 germJine mutations in the APe gene on chromosome 5q21, as determined by linkage analysis (42). FAP accounts for <1 % of all colorectal cancers and is characterized by the presence of hundreds to thousands of benign polyps in the colon and rectum, which could progress to adenomas if left untreated (20,39,42). Other phenotypic features connected with FAP include gastrointestinal tumors, desmoid tumors and CHRPE, congenital hypertrophy of the retinal pigment epithelium, a condition that does not affect sight and has no malignant potential (18,21,42). Two other variants of FAP exists: Gardner's syndrome, described by the presence of osteomas, odontomas, epidermoid cysts, fibromas, or demoid tumors in addition to colonic adenomas, and attenuated familial polyposis, characterized by the appearance of less than 100 adenomas in the colon and lesions in the upper gastrointestinal tract at a later age than classical F AP (39,42). Hereditary nonpolyposis colorectal cancer (HNPCC) is also an autosomal dominant syndrome that predisposes the afflicted individual to multiple primary tumors such as the endometrium, stomach, ovaries, and brain, without the formation of numerous intestinal polyps (7,39). It accounts for 3-5°1£) of all colorectal cancers and is characterized by mutations in MMR genes responsible for maintaining fidelity of DNA replication, thus, causing microsatellite instability that leads to neoplastic transformation (7,39,42). Patients afflicted with HNPCC have a 70-80% lifetime risk of developing colorectal cancer (42). Peutz-Jeughers Syndrome (PJS) and Juvenile Polyposis Syndromes (JPS) are examples of hamartomatous polyposis syndromes, which are typified by an overgrowth of normal tissues that involves mostly mesenchymal components (7). In addition to the presence of hamartomatous polyps in small intestine, PJS is also presented with 14 pigmentation of the peri-oral region, hands and feet (18). About half of PJS cases are caused by mutations in the LKBl gene (39). Indi viduals with JPS are also predisposed to hamm10matous polyps, mostly in the stomach and colorectum (18). This disorder is associated with mutations in SMAD-4 and BMP receptor lA genes, both of which affect the TGF-p pathway (7,39). Both syndromes are associated with an increased lifetime risk of intestinal and nonintestinal malignancies (7). 1.2 Retinoids Vitamin A and its derivatives, both synthetic and natural, are collectively known as retinoids. Structurally, retinoids have three distinct features: a substituted cyclohexenyl ring, a polar end group, and a conjugated polyene chain that makes them susceptible to photodegradation, isomerization and oxidation (Fig. 1.5) (43). Retinoids have essential roles in vertebrate growth and development, supporting vision, embryonic development and epithelial differentiation (44,45). However, all-trans RA (ATRA), the major signaling retinoid in the body, is also a very potent teratogen. Because of this, retinoic acid levels have to be tightly regulated, since too much or too little RA can result in severe malformations during embryogenesis (45). Biological responses to retinoids are mediated by the activation of specific retinoic acid (RA) receptors - retinoic acid receptors, RARu, RAR~, RARy, or retinoid X receptors, RXRu, RXR~, RXRy (45-48). These receptors belong to the nuclear hormone receptor superfamily and function primarily as ligand activated transcription factors, either as homodimers (e.g., RXR-RXR) or heterodimers (e.g., RAR-RXR), that control gene expression by binding to RA response elements in the promoter region of target genes. All-trans RA only binds to RARs, while 9-cis RA activates both RARs and A. II B. III O~~IH N H All-trans retinol Vitamin A All-trans retinoic acid 9-cis retinoic acid TTNPB (E)-4-[2-(5,6, 7,8-tetrahydro-5,5, 8,8-tetramethyl-2-naphthalenyl) propen-1-yl]benzoic acid 4-HPR N-( 4-hydroxyphenyl) retinam ide CD495 4-(5,6,7,8-tetrahydro-5,5,8,8- tetramethylnaphthol[2,3-b] furan-2-yl)benzoic acid 15 FIGURE 1.5. Natural and synthetic retinoids. (A) Natural derivatives of vitamin A include all-trans RA and 9-cis RA. Retinoids are composed of a substituted cyclohexenyl ring (I), a conjugated aliphatic chain (II) and a polar functional end group (III). (B) Synthetic derivatives of vitamin A exhibit various structural modifications to the basic retinoid skeleton such as cyclization of the polyene side chain (1TNPB), conversion of the carboxylic acid to an amide (4-HPR) and addition of a benzofuran moiety (CD495) (49). 16 RXRs. Recently, however, several other proteins have been found to interact with the retinoid nuclear receptors. In particular, RXRs are reported to be quite promiscuolls heterodimerization partners for various nuclear receptors such as the vitamin D3 receptor, the thyroid hormone receptor, the peroxisonlal proliferator-activator receptor (a and y) and some orphan receptors (44,50). These retinoid binding partners help establish other signaling pathways that contribute to RA-mediated biological effects. 1.2.1 Retinoid Storage and Transport Because retinoids are highly hydrophobic, they are often complexed with a binding protein to improve their solubility in a polar milieu. Retinol binding protein (RBP) is responsible for transporting retinol in the plasma to its target tissues while cellular retinoid binding proteins (CRBP) are retinoid-specific, cytosolic binding proteins assigned with the intracellular transport of retinoids (43). Three types of CRBPs have been identified so far. The cellular retinol binding protein (CRBP) binds to both all-trans isonlers of retinol and retinal while the cellular retinoic acid binding protein (CRABP) has a higher affinity to all-trans RA (43). The cellular retinal binding protein (CRALBP), which is present in the RPE, retina and brain, specifically binds to both II-cis retinol and II-cis retinal (43). Aside from carrying the retinoids within the cell, CRBPs have been postulated to perform several functions that include protecting their ligands from nonspecific oxidation and regulating retinoid metabolism by controlling the free concentration of retinoids in the cell (43,44,51). CRBPs have also been proposed to be responsible for presenting their ligands to the proper enzyme for metabolism (43,44,51). 17 CRBPs also facilitate the intracellular storage of retinol. Retinol is stored as retinyl esters, mainly as retinyl paln1itate and stearate, in a reaction catalyzed by lecithin:retinol acyltransferase (LRAT) (52,53). LRAT is a merrlbrane-bound enzyme, localized in the endoplasmic reticulum that transfers an acyl group from lecithin to generate retinyl esters (54). It has been reported that LRA T has a high affinity to retinol bound to CRBPs (55). It has also been demonstrated, through mouse CrbpI knockout studies, that loss of CrbpI gene function resulted in reduced levels of stored retinol in the liver, the primary site of vitamin A storage (52,55,56). Retinol esterification is a reversible process, retinyl esters can be hydrolyzed to retinol which can be re-esterified, absorbed into the blood or lnetabolized to RA for gene transctiption (57,58). 1.2.2 Retinoid Biosynthesis and Metabolism In addition to the nuclear hormone receptors, retinoid responsiveness is regulated by the availability of the retinoid ligands (Fig. 1.6). Cells acquire retinoids as dietary vitamin A from animal sources and as provitamin carotenoids from plant sources (45,51,55). Retinol is absorbed from circulation, converted to either retinyl esters by LRA T for storage or converted to all-trans RA. Retinol is not as potent as RA in terms of transcriptional regulation. Thus, it has to be metabolized to various RAs to induce biological response to retinoids. Conversion of retinol to RA is a two-step process. Retinol is first oxidized to the aldehyde, retinal, and this reaction is catalyzed by the cytosolic medium-chain alcohol dehydrogenases/reductases (ADH) or by the microsomal short chain alcohol dehydrogenases/reductases (SDR) (59,60). Retinal is then rapidly oxidized to RA by the aldehyde dehydrogenases (ALDH) (59,60). It is reported that the first step, which is reversible, is the rate determining step in retinol metabolism, and Q~ ·(:I·carot.n~ ROL ~OH LRAT- 1 ADH, SDR p-carotene 15, 15'· dioxygenase ~ RAt' 1 ALDH ~OH RA P450 4-0H·~·· ' > ,; ,\;~~"~:' ''', OH :. - /";""""" ". ' 4-oxo:RA ~OH o 18 Retinyl Ester Gene Transcription FIGURE 1.6. Retinoid biosynthesis and metabolism. Retinol is either esterified for storage or n1etabolized to RA for gene activation through a two-step oxidation with retinaldehyde as intermediate. ~-carotene is converted to RA through a similar mechanism that involves a central double bond cleavage. RA is further metabolized to more polar compounds by cytochrome P450 enzymes. 19 therefore, the step most likely to be precisely regulated by the cell (45,47,59,61-63). Carotenoid metabolism proceeds through a similar pathway. ~-carotene is first converted to retinal by ~-carotene-15,15' -dioxygenase by cleavage at its central double bond. Retinal is then further transformed to either retinol for storage or retinoic acid for activation of RA responsive genes (64). All-trans RA can be isomerized to 13-cis and 9-cis RA isomers and metabolized to various forms including 5,6-epoxy-RA and 3,4-didehydro-RA (65). However, the major pathway involved in clearing retinoic acid is through an irreversible, cytochrome P450- catalyzed oxidation (62). Retinoic acid is oxidized to more polar metabolites such as 4- hydroxy-RA, 4-oxo-RA and 18-hydroxy-RA which are then conjugated with glucoronic acid for excretion (51). Several CYP isoforms have been implicated in RA metabolism, but it is widely accepted that CYP26A 1 is the most specific ATRA 4-hydroxylase and is the major contributing enzyme to this metabolic process (63,65-67). 1.2.3 Retinoids and Cancer Dysregulation of retinoid nuclear receptors and biosynthetic genes have been reported in various human tumors. Acute promyelocytic leukemia (APL) is characterized by a genetic translocation that results in the formation of dominant-negative fusion proteins containing RARa, and is thought to be the underlying cause of the disease (46). RARa is also epigenetically silenced in MCF7, a human breast carcinoma cell line (44). RAR~ is reduced or unexpressed either through promoter hypermethylation, mutation or deletion in many tumors and cancer cells including breast, lung, head and neck squamous cell carcinoma, oral tissue, cervix, ovary, stomach and prostate while low expression level of 20 RXRs is observed in squamous cell carcinoma, non-small cell lung cancer and thyroid cancer. (44,46,48,68-71). Recent studies have shown that defects in RA biosynthesis are also present in human cancers. Reduced level of endogeous RA is observed in oral premalignant lesions, which is attributed to increased RA metabolism (72). LRAT protein is undetectable in renal cancer cell lines, which is paralleled by negligible retinyl ester synthesis (73). Decreased retinyl ester levels are also repOlted in a number of carcinoma cell lines of the oral cavity, kidney, skin, breast and prostate, including human melanoma cultures exhibiting a fibroblastoid morphology (73-75). Retinol conversion is reported to be impaired in breast cancer cells compared to normal cells and aldehyde dehydrogenase 6, an enzyme that catalyzes the conversion of retinal to RA, is shown to be absent in MCF-7, a breast cancer cell line that has been proven to be defective in RA biosynthesis (76,77). Triano et al. also revealed that class I ADH is dramatically reduced in invasive breast cancers (78). FUlthermore, the cellular retinol binding protein I (CRBPI) is epigenetically inactivated in several cancer cell lines and prin1ary tumors such as lymphoma, leukemia, gastric, breast and liver (79-82). Retinoids have also been implicated in maintaining normal intestinal function and in the development of colorectal cancers. Studies have shown that vitamin A deficiency leads to abnormalities in the colon such as decreased mucus production, hypersecretion, reduced absorpti ve function and expansion of the crypt proliferati ve compartment (83- 88). Retinoids have also been reported to be effective in preventing azoxymethane (AOM)-induced colon tumors in rats, In addition to their ability to inhibit cell proliferation, induce cell differentiation and apoptosis, increase cell adhesion and 21 suppress invasiveness in colon cancer cell lines (89-97). More recently, it has been demonstrated that knock down of zebrafish retinol dehydrogenases, rdhl (zRDlIB) and rdhll (zRDHA), resulted in a vitamin A-deficient phenotype with defects in intestinal differentiation and development, all of which were rescued with exogenous RA treatment (98). Finally, dysregulation in the retinoid signaling and biosynthetic pathway have also been observed in colorectal tumors. Carotenoids, which are vitamin A precursors, are reduced in colorectal adenomas (99). Retinoid nuclear receptors, predominantly RAR~, are absent in colon tumors and cell lines and are reported to be due largely to hypermethylation of CpG islands in the promoter region (82,91,93,100). A similar hypermethylation-associated gene inactivation has been rep0l1ed for CRBPI in colon tumors (82). Jette et ai. revealed that absence of retinoid responsive genes in colon tumors is paralleled by the absence of RA biosynthetic enzymes, RDHL and RDH5 (101). The same study also showed that both enzymes have low expression in colon tunlor cell lines and this is accompanied by the inability of cancer cells to efficiently convert retinol to RA (101). 1.2.4 Retinoids in Cancer Chemoprevention and Treatment The therapeutic property of retinoids is attributed to its ability to inhibit cellular proliferation and restore the differentiation and apoptotic activities of tumor cells (45,47,51,91,102). This is particularly true in APL, where treatment with all-trans RA induces the differentiation of leukemic promyelocytes (45-47), Several retinoids are cunently in clinical trials for therapeutic evaluation (Fig. 1.7). All-trans RA (tretinoin), 9-cis RA (alitretinoin, pam'etin) and I3-cis RA (isotretinoin) are being investigated for OOH ~ OOH LDG1069 4-[1-(3,5,5,8,8- pentamethyltetralin-2-yl) ethenyl]benzoic acid Adapalene 6-[3[( 1-adamantyl)-4- methoxyphenyl]-2-naphthoic acid NIK-333 (2E, 4E, 6E, 1 0E)-3,7, 11,15- tetramethyl-2,4,6,1 0,14- hexadecapentaenoic acid 22 FIGURE 1.7. Retinoids in clinical studies. Several retinoids have shown remarkable acti vity against various human tumors in the clinic, including the RXR agonist LDG 1 069 and the atypical retinoids adapalene and NIK-333 (45,103). 23 antitumor activity against a variety of human cancers including thyroid, prostate, breast, skin, renal-cell and squamous cell carcinoma (45). The RXR agonist, LDG 1069 (targretin, bexacarotene), shows promising activity against cutaneous T-cell lymphoma and non-small-cell-lung cancer (45), Atypical retinoids, which can bind and activate RARs, also exhibit potent antitumor activity in the clinic. 4-hydroxyphenylretinamide (4- HPR) is in phase III for treating bladder cancer while acyclic retinoid NIK-333 and adapalene are effective in preventing hepatocellular carcinoma and cervical intraepithelial neoplasia, respectively (45). Retinoids, in combination with other anticancer regimens, have also been proven effective in various cancer malignancies. A1l-trans RA or 13-cis RA, together with interferon-a-2a, have been used in treating cancers such as squamous cell carcinoma and metastatic renal cell carcinoma (50,51,104). Histone deacetylase (HDAC) inhibitors alleviate RA resistance in AML, possibly through reversing transcriptional silencing that blocks retinoid action, and restore responsiveness to retinoid therapy (45). All-trans RA also sensitizes tumor cells to the cytotoxic agents cisplatin, etoposide and bleomycin (104). Retinoids have also been used extensively as chemopreventive agents against a diverse set of human cancers including skin, head and neck, breast, lung and liver (45,47,51,104). Early studies indicated an inverse relationship between dietary intake of vitamin A or ~-carotene and cancer incidence (105). Recently, it has been shown that high vitamin A consumption resulted in cancer-risk reduction for larygeal, lung, esophageal and tongue carcinoma (106), Retinoid treatment has also been successful in 24 treating preneoplastic lesions, such as oral leukoplakia, actinic keratosis and cervical dysplasia (45,50,104). There are, however, disadvantages associated with retinoid therapy. Chronic administration of high concentrations of RA can lead to toxicity characterized by anorexia, weight loss, bone and joint pain (51). Because of its teratogenic effect, retinoid use is limited in women of childbearing age (51). Also, increased RA metabolism can lead to resistance to retinoid therapy by diminishing the amount of RA available to cells. Indeed, retinoic acid metabolism blocking agents (RAMBAs), have been shown to enhance the antitumor effects of retinoids (Fig 1.8). Liarazole, a retinoic acid 4- hydroxylase inhibitor, improved the ability of retinoids to inhibit cell proliferation in breast cancer and human glioblastoma cells (51). More potent novel RAMBAs have been synthesized recently and Patel et al. have demonstrated that VN/14-1 and RA, administered together, inhibited cell proliferation of MCF7 more effecti vely than either RA or VN/14-1 alone (66). The same group also showed that retinoids and the HDAC inhibitor, suberoylanilide hydroxamic acid (SARA), have a synergistic effect on human LNCaP prostate cancer cell proliferation (107). 1.3 Project Rationale Gene expression profile analysis of nOlmal colon and tumor tissues revealed that retinoid response genes are missing in tumor samples (101). This is paralleled by the absence of RA biosynthetic genes - RDI-I5, a known retinol dehydrogenase and DHRS9, a novel, colon-specific retinol dehydrogenase homolog (101). Although a number of studies have implicated retinoids in normal colonocyte function and in the development of colon neoplasms, at present, little is known about the role of retinoids in N l!.) N H J CI Liarozole OH VN/14-1 4-(±)-1 H-imidazol-1-yl)­( E)-retinoic acid 25 FIGURE 1.8. Retinoic acid metabolism blocking agents. Inhibitors of the P450 enzymes that metabolize retinoic acid, such as liarozole and VN/14-1, have been shown to be an effective strategy in augmenting the antitumor effects of RA and minimizing resistance to RA treatment (66,108). 26 colonocyte differentiation and its regulation in the colon (83,85,90,94,96). This study aims to demonstrate that colon carcinogenesis is accompanied by dysregulation of retinoid biosynthesis and to identify retinoid genes that are essential for normal intestinal development. 1.4 References 1. Jemal, A., Munay, T., Ward, E., Samuels, A., Tiwari, R. C., Ghafoor, A., Feuer, E. J., and Thun, M. J. (2005) CA Cancer.1. Clin. 55, 10-30 2. www.cancer.org. 3. www.oncologychannel.com. 4. www.fda.gov. 5. www.centerwatch.com. 6. Nathke,1. S. (2004) Aruul. Rev. Cell Dev. BioI. 20,337-366 7. Sancho, E., BatIe, E., and Clevers, H. (2004) Annu. Rev. Cell. Dev. BioI. 20,695- 723 8. Gregorieff, A., Pinto, D., Begthel, H., Destree, 0., Kielman, M., and Clevers, H. (2005) Gastroenterology 129, 626-638 9. Radtke, F., and Clevers, H. (2005) Science 307, 1904-1909 10. Roncucci, L., Pedroni, M., Vaccina, F., Benatti, P., Marzona, L., and Pol, A. D. (2000) Cell Prolif. 33, 1-18 11. Shanmugathasan, M., and Jothy, S. (2000) Pathol. Int. 50, 273-279 12. Booth, C., and Potten, C. S. (2000).1. Clin. Invest. 105, 1493-1499 13. Andreu, P., Col not, S., Godard, C., Gad, S., Chafey, P., Niwa-Kawakita, M., Laurent-Puig, P., Kahn, A., Robine, S., Penet, C., and Romagnolo, B. (2005) Development 132, 1443-1451 14. Clatw0l1hy, J. P., and Subramanian, V. (2001) Mech. Dev. 101,3-9 15. Es, J. H. v., Jay, P., Gregorieff, A., Gijn, M. E. V., Jonkheer, S., Hatzis, P., Thiele, A., BOll1, M. V. d., Begthel, H., Brabletz, T., Taketo, M. M., and Clevers, H. (2005) Nat. Cell Bio!. 7,381-386 27 16. Sansom, O. J., Reed, K. R., Hayes, A. J., Ireland, H., Brinkmann, H., Newton, 1. P., Batlle, E., Simon-Assmann, P., Clevers, H., Nathke, I. S., Clarke, A. S., and Winton, D. J. (2004) Genes Dev. 18, 1385-1390 17. Cheng, L., and Lai, M. D. (2003) World 1. Gastroenterol. 9,2642-2649 18. Feall1head, N. S., Wilding, J. L., and Bodlner, W. F. (2002) Br. Med. Bull. 64,27- 43 19. Mimori-Kiyosue, Y., and Tsukita, S. (2001) 1. Cell Biol. 154, 1105-1109 20. Bodmer, W. (1999) Cytogenet. Cell Genet. 86,99-104 21. Featllhead, N. S., Britton, M. P., and Bodmer, W. F. (2001) Hum. Mol. Genet. 10, 721-733 22. Narayan, S., and Roy, D. (2003) Mol. Cancer 2,41 23. Kinzler, K. W., and Vogelstein, B. (1996) Cell 87, 159-170 24. Bienz, M. (2003) Curro Biol. 13, R215-R216 25. Li, H., Pamukcu, R., and Thompson, W. J. (2002) Cancer Biol. Ther. 1,621-625 26. Hao, X. P., Pretlow, T. G., Rao, J. S., and Pretlow, T. P. (2001) Cancer Res. 61, 8085-8088 27. Behrens, J., Kries, J. P. V., Kuhl, M., Bruhn, L., Wedlich, D., Grosschedl, R., and Birchmeier, W. (1996) Nature 382,638-642 28. Molenaar, M., Watering, M. v. d., Oosterwegel, M., Peterson-Maduro, J., Godsave, S., Korinek, V., Roose, J., Destree, 0., and Clevers, H. (1996) Cell 86, 391-399 29. Mariadason, J. M., Arango, D., Comer, G. A., Aranes, M. J., Hotchkiss, K. A., Yang, W., and Augenlicht, L. H. (2002) Cancer Res. 62,4791-4804 30. Spink, K. Flidman, S. G., and Weis, W. 1. (2001) EMBO 1. 20,6203-6212 31. Reya, T., and Clevers, H. (2005) Nature 434, 843-850 32. Korinek, V., Barker, N., Morin, P. J., Wichen, D. V., Weger, R. d., Kinzler, K. W., Vogelstein, B., and Clevers, H. (1997) Science 275, 1784-1787 33. Morin, P. J., Sparks, A. B., Korinek, V., Barker, N., Clevers, H., Vogelstein, B., and Kinzler, K. W. (1997) Science 275, 1787 .. 1790 28 34. Brabletz, T., Jung, A., Reu, S., Porzner, M., Hlubek, F., Kunz-Schughart, L., Knuechel, R., and Kirchner, T. (2001) Proc. Natl. Acad. Sci. USA 98, 10356- 10361 35. Fodde, R. (2003) Nat. Cell Biol. 5, 190-192 36. Kaplan, K. B., Burds, A. A., Swedlow, J. R., Bekir, S. S., SorgeI', P. K., and Nathke,1. S. (2001) Nat. Cell BioI. 3,429-432 37. Houlston, R. S. (2001) Mol. Pathol. 54,206-214 38. Fearon, E. R., and Vogelstein, B. (1990) Cell 61, 759-767 39. Chapelle, A. d. (2004) Nat. Rev. Cancer 4,769-780 40. Pinto, D., Gregorieff, A., Begthel, H., and Clevers, H. (2003) Genes Dev. 17, 1709-1713 41. Wetering, M. v. d., Sancho, E., Verweij, C., Lau, W. d., Oving, 1., Hurlstone, A., Hom, K. v. d., BatIle, E., Coudreuse, D., Haramis, A.-P., Tjon-Pon-Fong, M., Moerer, P., Born, M. v. d., Soete, G., Pals, S., Eilers, M., Medema, R., and Clevers, H. (2002) Cell II!, 241-250 42. Rowley, P. T. (2005) Annu. Rev. Med. 56, 539-554 43. Noy, N. (1999). In: Nau, H., and Blaner, W. S. (eds). Retinoids: The Biochemical and Molecular Basis of Vitamin A and Retinoid Action, Springer-Verlag, Berlin 44. Niles, R. M. (2004) Mutat. Res. 555, 81-96 45. Altucci, L., and Gronemeyer, H. (2001) Nat. Rev. Cancer 1,181-193 46. Soprano, D. R., Qin, P., and Soprano, K. J. (2004) Annu. Rev. Nutr. 24,201-221 47. Hansen, L. A., Sigman, C. C., Andreola, F., Ross, S. A., Kelloff, G. J., and Luca, L. M. D. (2000) Carcinogenesis 21,1271-1279 48. Widschwendter, M., Berger, J., Hermann, M., Muller, H. M., Amberger, A., Zeschnigk, M., Widschwendter, A., Abendstein, B., Zeimet, A. G., Daxenbichler, G., and Marth, C. (2000) 1. Natl. Cancer Inst. 92, 826-832 49. Delescluse, C., Cavey, M. T., Martin, B., Bernard, B. A., Reichert, U., Maignan, J., Darmon, M., and B.Shroot. (1991) Mol. Phannacol. 40, 556-562 50. Freelnantle, S. J., Spinella, M. J., and Dmitrovsky, E. (2003) Oncogene 22,7305- 7315 51. Miller, W. H. (1998) Cancer 83, 1471-1482 29 52. HaITison, E. H. (2005) Annu. Rev. Nutr. 25,5.1-5.18 53. Ruiz, A., Winston, A., Lim, Y.-H., Gilbert, B. A., Rando, R. R., and Bok, D. (1999) 1. Bioi. Chem. 1999, 3834-3841 54. Ross, A. C., and Zolfaghari, R. (2004) 1. Nutr. 134, 269S-275S 55. Vogel, S., Gamble, M. V., and Blaner, W. S. (1999). In: Nau, H., and Blaner, W. S. (eds). Retinoids: The Biochemical and Molecular Basis of Vitamin A and Retinoid Function, Springer-Verlag, Berlin 56. Ghyselinck, N. B., Bavik, C., Sapin, V., Mark, M., Bonnier, D., Hindelang, C., Dierich, A., B.Nilsson, C., Hakansson, H., Sauvant, P., Azais-Braesco, V., Frasson, M., Picaud, S., and Chambon, P. (1999) EMBO 1. 18, 4903-4914 57. Ong, D. E., Newcomer, M. E., and Chytil, F. (1994). In: Sporn, M. B., Roberts, A. B., and Goodman, D. S. (eds). The Retinoids: Biology, Chelnistry and Medicine, 2nd Ed., Raven Press Ltd., New York 58. Yost, R. W., Hamson, E. H., and Ross, A. C. (1988) J. Bioi. Chenl. 263, 18693- 18701 59. DIven, S. M., Gundersen, T. E., Weedon, M. S., Landaas, V. 0., Sakhi, A. K., Fromm, S. H., Geronimo, B. A., Moskaug, J. 0., and Blomhoff, R. (2000) Dev. BioI. 220, 379-391 60. Duester, G. (2000) Eur. 1. Biochem. 267, 4315-4324 61. Soref, C. M., Di, Y.-P., Hayden, L., Zhao, Y. H., Satre, M. A., and Wu, R. (2001) 1. BioI. Chent. 276,24194-24202 62. Chen, H., Fantel, A. G., and Juchau, M. R. (2000) Drug Metab. Dispos. 28, 1051- 1057 63. White, J. A., Guo, Y.-D., Baetz, K., Beckett-Jones, B., Bonasoro, J., Hsu, K. E., Dilworth, F. 1., Jones, G., and Petkovich, M. (1996) J. BioI. Chel11. 271, 29922- 29927 64. Nagao, A. (2004) 1. Nutr. 134, 237S-240S 65. Blaner, W. S., Piantedosi, R., Skyes, A., and Vogel, S. (1999). In: Nau, H., and Blaner, W. S. (eds). Retinoids: The Biochemical and Molecular Basis of Vitamin A and Retinoid Action, Springer-Verlag, Berlin 66. Patel, J. B., Huynh, C. K., Handratta, V. D., Gediya, L. K., Broadie, A. M. H., Goloubeva, O. G., Clement, O. 0., Nanne, I. P., Soprano, D. R., and Njar, V. C. O. (2004) 1. Med. Chem. 47, 6716-6729 30 67. Dobbs-McAuliffe, B., Zhao, Q., and Linney, E. (2004) Mech. Dev. 121,339-350 68. Haugen, B. R., Larson, L. L., Pugazhenthi, U., Hays, W. R., Klopper, J. P., Kramer, C. A., and Sharma, V. (2004) 1. Clin. Endocrinol. Metab. 89,272-280 69. Brabender, J., Danenberg, K. D., Metzger, R., Schneider, P. M., Lord, R. V., Groshen, S., Tsao-Wei, D. D., Park, J., Salonga, D., Holscher, A. H., and Danenberg, P. V. (2002) Clin. Cancer Res. 8,438-443 70. Li, M., Indra, A. K., Warot, X., Brocard, J., Messaddeq, N., Kato, S., Metzger, D., and Chambon, P. (2000) Nature 407,633-636 71. Arapshian, A., Kuppumbatti, Y. S., and Mira-y-Lopez, R. (2000) Oncogene 19, 4066-4070 72. Lotan, R. (1980) Biochim. Biophys. Acta. 605, 33-91 73. Guo, X., Nanus, D. M., Ruiz, A., Rando, R. R., Bok, D., and Gudas, L. J. (2001) Cancer Res. 61, 2774-2781 74. Boorjian, S., Tickoo, S. K., Mongan, N. P., Yu, H., Bok, D., Rando, R. R., Nanus, D. M., Schen, D. S., and Gudas, L. J. (2004) Clin. Cancer Res. 10,3429-3437 75. Simmons, D. P., Andreola, F., and Luca, L. M. D. (2002) Carcinogenesis 23, 1821-1830 76. Rexer, B. N., Zheng, W. L., and Ong, D. E. (2001) Cancer Res. 61,7065-7070 77. Mira-Y-Lopez, R., Zheng, W. L., Kuppumbatti, Y. S., Rexer, B., ling, Y., and Ong, D. E. (2000) 1. Cell. Physiol. 185, 302-309 78. Triano, E. A., Slusher, L. B., Atkins, T. A., Beneski, l. T., Gestl, S. A., Zolfaghari, R., Polavarapu, R., Frauen h offer, E., and Weisz, J. (2003) Cancer Re~ 63, 3092-3100 79. Kwong, J., Lo, K.-W., Chow, L. S.-N., To, K.-F., Choy, K.-W., Chan, F. L., Mok, S. C., and Huang, D. P. (2005) Neoplasia 7,67-74 80. Arapshian, A., Bertran, S., Kuppumbatti, Y. S., Nakajo, S., and Mira-y-Lopez, R. (2004) Mol. Cancer 3, 13 81. Cvetkovic, D., Williams, S. l., and Hamilton, T. C. (2003) Clin. Cancer Res. 9, 1013-1020 82. Esteller, M., Guo, M., Moreno, V., Peinado, M. A., Capella, G., Galm, 0., Baylin, S. B., and Herman, l. G. (2002) Cancer Res. 62, 5902-5905 83. Nzegwu, H. C., and Levin, R. l. (1992) Gut 33,794-800 31 84. Zile, M., Bunge, C., and Deluca, H. F. (1977) 1. NutI'. 107, 552-560 85. Narisawa, T., Reddy, B. S., Wong, C.-Q., and Weisburger, J. H. (1976) Cancer Re~ 36, 1379-1383 86. Zile, M., and Deluca, H. F. (1970) Arch. Biochem. Biophys. 140, 210-214 87. Perumal, A. S., Lakshmanan, M. R., and Cama, H. R. (1968) Biochim. Biophys. Acta. 170, 399-408 88. Perumal, A. S., Samy, T. S., Lakshmanan, M. R., Jungalwala, F. B., and Rao, P. B. (1966) Biochim. Biophys. Acta. 124,95-100 89. Adachi, Y., Itoh, F., Yamamoto, H., 1ku, S., Matsuno, K., Arimura, Y., and Imai, K. (200 1) Tumour BioI. 22, 247-253 90. Wargovich, M. J., Jimenez, A., McKee, K., Steele, V. E., Velasco, M., Woods, J., Price, R., Gray, K., and Kelloff, G. J. (2000) Carcinogenesis 21, 1149-1155 91. Lee, M.-O., Han, S.-Y., Jiang, S., Park, 1. H., and Kim, S. J. (2000) Biochem. Pharnlacol. 59, 485-496 92. Nicke, B., Kaiser, A., Widenmann, B., Riecken, E.-O., and Rosewicz, S. (1999) Biochem. Biophys. Res. Commun. 261, 572-577 93. Nicke, B., Riecken, E.-O., and Rosewicz, S. (1999) Gut 45,51-57 94. Zheng, Y., Kramer, P. M., Lubet, R. A., Steele, V. E., Kelloff, G. J., and Pereira, M. A. (1999) Carcinogenesis 20,255-260 95. Reynolds, S., Rajagopal, S., and S.Chakrabarty. (1998) Cancer Lett. 134, 53-60 96. Zheng, Y., Kramer, P. M., Olson, G., Lubet, R. A., Steele, V. E., Kelloff, G. J., and Pereira, M. A. (1997) Carcinogenesis 18, 2119-2125 97. Niles, R. M., Wilhelm, S. A., Thomas, P., and Zamcheck, N. (1988) Cancer Invest. 6, 39-45 98. Nadauld, L. D., Sandoval, 1. T., Chidester, S., Yost, H. J., and Jones, D. A. (2004) 1. BioI. Chem. 279,51581-51589 99. Muhlhofer, A., Buhler-Ritter, B., Frank, J., Zoller, W. G., Merkle, P., A.Bosse, Heinrich, F., and Biesalski, H. K. (2003) CUn. Nutr. 22, 65-70 100. Youssef, E., Estecio, M., and Issa, J. (2004) Cancer BioI. Ther. 3, 82-86 101. Jette, C., Peterson, P. W., Sandoval, I. T., Manos, E. J., Hadley, E., Ireland, C. M., and Jones,D. A. (2004) 1. Biol. Chem. 279,34397-34405 32 102. Shah, S., Pishvaian, M. J., Easwaran, V., Brown, P. H., and Byers, S. W. (2002) J. BioI. Chem. 28, 25313-25322 103. Kagawa, M., Sano, T., Ishibashi, N., Hashimoto, M., Okuno, M., Moriwaki, H., Suzuki, R., Kohno, H., and Tanaka, T. (2004) Carcinogenesis 25, 979-985 104. Evans, T. R. J., and Kaye, S. B. (1999) Br. J. Cancer 80, 1-8 105. Fujimaki, Y. (1926) J. Cancer Res. 10,469-477 106. Lippman, S. M., Lee, J. S., Lotan, R., and Hong, K. W. (1990) Head Neck 12, 5- 20 107. Gediya, L. K., Chopra, P., Purushottamachar, P., Maheshwari, N., and Njar, V. C. O. (2005) J. Med. Chem. 48,5047-5051 108. Yee, S. W., Jamo, L., Gonlaa, M. S., Elford, C., Ooi, L., Coogan, M. P., McClelland, R., Nicholson, R. 1., Evans, B. A. J., Brancale, A., and Simons, C. (2005) J. M ed. Chem. 48, 7123-7131 CHAPTER 2 CHARACTERIZATION OF A NOVEL COLON-SPECIFIC RETINOL DEHYDROGENASE 2.1 Introduction Retinoids are signaling molecules that play a role in various biological processes including vision, embryonic development and cellular differentiation (1,2). Retinoid metabolism is a two-step oxidation process that involves the conversion of retinol to the active metabolite retinoic acid (RA), with retinal as an intermediate. Oxidation of retinol to retinal is reversible and rate-limiting. It is catalyzed by two families of alcohol dehydrogenases - the microsomal short chain alcohol dehydrogenases (SDR) and the cytosolic medium chain alcohol dehydrogenases (ADH) (2-8). Both classes of enzymes exhibit wide substrate specificities that include ethanol and hydroxysteroids (3). The second step, conversion of retinal to RA, is mediated by the aldehyde dehydrogenases (3,8). Once synthesized, RA binds and activates the nuclear retinoid receptors to initiate transcription of target genes (2,4,9,10). Several enzymes have been implicated in maintaining vitamin A function in humans (3). ADHl, ADH2, ADH4, RODH4, RDH5 and retSDRl have all been reported to metabolize retinol. Numerous studies have been undertaken to determine which enzymes are essential for retinol metabolism however, experiments with knockout mice revealed a redundancy in retinol dehydrogenase function. Adhrl -, Adh3-1 -, Adh4-1 - and RdhS-I - 34 knockout mice are all viable, suggesting that other retinol dehydrogenases compensate for the loss of gene function and metabolize enough retinol to ensure animal survi val under a standard diet (11-14). Under vitamin A deficiency (VAD), Adh3 and Adh4 null mutant mice exhibit increased postnatal mortality and reduced RA biosynthesis, implying that these genes are important for maintaining sufficient RA levels in the body (11,14). RALDHl, RALDH2 and RALDH3 ilTeversibly oxidize retinal to RA. Raldhl and Raldh3 are mainly involved in RA biosynthesis in the developing eye, affecting the patterning of the dorsoventral axis of the eye (IS). However, Raldh2 gene knockout in mice led to severe embryonic developmental defects which were rescued by exogenous RA, proving that Raldh2 is the primary ALDH that provides RA during development (16,17). This is further supported by studies done on zebrafish, where knocking out raldh2 resulted in a neckless phenotype that is consistent with RA deficiency (18). Microarray analysis of norma] and tumor colon tissues revealed that two retinoid biosynthetic enzymes are missing in the tumor samples, RDHS and DHRS9, a novel, colon-specific retinol dehydrogenase (19). DHRS9 shares 4S% identity to RDHS, a known retinol dehydrogenase, and phylogenetic analysis of various SDR gene family menlbers revealed that DHRS9 is most closely related to the retinoid-associated SDR subgroup (S). It also contains amino acid motifs that are highly conserved in the SDR family of enzymes - the co-factor binding site OXXXOXO (residues 36-42), the active site YXXXK (residues 176-180), LXNNAO (residues 109-1114) and PO (residues 206-207) (3,S). Three other groups have identified and characterized human DHRS9 in other tissues. Soref et al. cloned DHRS9 (referred to as hRDH-TBE) from primary human 35 tracheobronchial epithelial (TBE) cells by microanay and demonstrated that its expression is specific to the airway epithelial cells and is capable of metabolizing retinol to RA in COS cells overexpressing hRDH-TBE (5). Markova et al. identified DHRS9 (refened to as hRoDH-E2) in the epidermis and epidermal keratinocytes by PCR using degenerate primers derived from a conserved region of the SDR enzyme family (20). They supported the previous findings by Soref et al. that DHRS9 functions as a retinol dehydrogenase by showing that NHEK cells overexpressing DHRS9 produced more retinal compared to control cells and were able to activate a retinoic acid-responsive CAT reporter construct (20). They further proved that DHRS9 is a microsomal retinol dehydrogenase that prefers NADP as co-factor (20). Chetyrkin et al. also cloned and purified recombinant DHRS9 (referred to as 3a-HSD) from liver tissue (21). However, they reported that in vitro, DHRS9 was 100 times more active as a 3a-hydroxysteroid dehydrogenase than as a retinol dehydrogenase, preferring to metabolize allopregnanolone and 3a-androstanediol to dihydroprogesterone and dihydrotestosterone, respectively (21). Furthermore, they demonstrated that DHRS9 was an integral membrane protein with a cytosolic orientation (21). Because of its tissue specific distribution, it is proposed that DHRS9 is the retinol dehydrogenase in the colon that catalyzes retinol oxidation to retinal. Studies presented in this chapter aim to determine the enzymatic activity of DHRS9 in the colon and evaluate the retinoid biosynthetic capabilities of colon tumor cells. 2.2 Methodology Quantitative RT-PCR. Total RNA was isolated from 10 matched sets of frozen normal and adenoma or carcinoma tissue samples using an RNeasy kit (Qiagen). cDNA was 36 synthesized from 2 ~g of total RNA using Superscript III (Invitrogen). Real-time PCR was performed using the Roche Light Cycler instrument and software, version 3.5 (Roche Diagnostics). Intron-spanning primers (DHRS9: forward, 5' -TGGAAACTTGGCAGCC AGAA-3'; reverse, 5'-CCAGAGACCTTTCTCCCCAA-3', RDH5: forward, 5'- ITCTC TGACAGCCTGAGGCG-3'; reverse, 5'-TGCGCTGITGCAITTTCAGG-3', CYCLIN Dl: forward, 5'- AGCTCCTGTGCTGCGAAGTG -3'; reverse, 5'-TCCTCGCAGACCT CCAGCAT-3', 18S rRNA: forward, 5'- GGTGAAAITCITGGACCGGC-3', reverse, 5' -GACTITGGTITCCCGGAAGC-3') were designed to amplify 200-bp products to exclude amplification from genomic DNA. PCR was performed in duplicate with a master ll1ix consisting of cDNA template, buffer (500 mM Tris pH 8.3, 2.5 mglmL BSA, 30 mM MgClz), dNTPs (2 mM), TaqStart antibody (Clontech), Biolase DNA polymerase (Bioline), gene-specific forward and reverse primers (10 ~M) and SYBR Green I (Molecular Probes) under the following conditions: 1 cycle at 95° C for 10 min followed by 35 cycles of amplification at 95° C for 1 s, 5-second annealing at 54° C (DHRS9), 58° C (RDH5), 60° C (CYCLIN D) and 53° C (I8S rRNA), 72° C for 8 s, 84° C for 0 s followed by 1 cycle of melting at 72° C for 0 s, 96° C for 0 s followed by 1 cycle of cooling at 40° C for 60 s. A template-free negative control was included in each experiment. The copy number was measured by comparing gene amplification with the amplification of standard samples containing 103 to 107 copies of the gene. The relative expression level of each gene was normalized to 18S rRNA to account for variations in the amount of input cDNA. 37 Cell culture. CaC02, Col0205, DLD-l, HCTl16, HT29, RKO and SW480 cells were cultured as recommended by the American Type Culture Collection (A TCC). HT29 APC-inducible and LacZ-inducible cells were graciously provided by Dr. Bert Vogelstein (Johns Hopkins Oncology Center, Baltimore, MD). Northern blot analysis. Total RNA was isolated from colon tumor cell lines using Trizol (Invitrogen) followed by messenger RNA preparation using Oligotex mRN A kit (Qiagen). Samples containing 1Y1RNA were fractionated through a formaldehyde-agarose gel and transferred onto Hybond-N nylon membranes (Amersham Biosciences). Radiolabeled probes were made with the Rediprime II labeling system (Amersham Biosciences) supplemented with [a)2P]-dCTP (NEN Life Science). Hybridizations with 32P-labeled DHRS9, RDH5 or glyceraldehyde-3-phosphate dehydrogenase (GAPDH) probes were carried out using ULTRAhyb-Oligo hybridization buffer (Ambion) as recommended by the manufacturer. Plasmids. Open reading frames of DlIRS9, RDH5 and RALDH2 were cloned into a pcDNA 3.1II-IisC vector (Invitrogen) between BamHI and EcoRI sites. Mutant RDH5, S73F, and DHRS9 constructs, K184A and D290H, were made using the QuikChange Site-Directed Mutagenesis kit (Stratagene). Transient transfections. Lipofectamine 2000 (Invitrogen) was used to transfect HCT 116 cells according to the manufacturer's protocol. Cells were seeded at a density of 40,000 cells/well in 6-well plates and transfected the following day. Transfections were performed using 3.0 J.!g of his-tagged DNA or empty vector for 24 h. Retinol treatment and extraction. Cells were treated with 100 nmol all-trans retinol (Sigma) at 80-900/0 confluence for 12 h or for 4 h for transfected cells. Medium was 38 removed and cells were scraped into Ix PBS for protein quantification. For HT -29 inducible cell lines, cells were treated with 100 nmol all-trans retinol for 4 h following a 24 h induction with 100 ~M zinc chloride. Medium was removed and cells were saved in Trizol (Invitrogen) for RNA quantification. Following addition of 100 nmol internal standard, 4-(E-2-(S,6,7 ,8-tetrah ydro-S,S,8,8-tetrameth yl-2-naphthaleny I)-I-propenyl) benzoic acid (TTNPB, Sigma), the medium was acidified with 6N HCI (0.03 X volume) and extracted twice with an equal volume of hexanes containing 0.1 mg/ml butylated hydroxytoJuene. The resulting solution was mixed vigorously and spun down at II,SOO rpm for 20 min. The organic layer was saved, dried down under nitrogen, and stored at - 20° C for HPLC analysis. HPLC analysis. Dried extracts were reconstituted in 100 JlI 1: 1 DMSO/MeOH and analyzed by reversed-phase HPLC using an Agilent 1100 Series with a diode-array detector (Agilent Technologies). The retinoids were eluted on a Luna C18 analytical column (4.6 x 2S0 mm, SI-lm particle size, Phenomenex) with a gradient starting at 80% acetonitrile/ 20% 0.1 M ammonium acetate, pH S.O, to 100% acetonitrile in 40 min, with a flow rate of I.S ml/min at 3S0 nm. Identification and quantification of extracted retinoids were done by comparing elution times and peak areas with matching retinoid standards. Extracted retinoic acid (RA) from control cells was subtracted from RA from treated samples. Reported RA values were normalized to both TTNPB and sample protein or RNA content. Statistical analyses were performed using GraphPad Prism (v 4.03). 39 2.3 Results DfIRS9 and RDH5 are down-regulated in colon carcinoma and carcinoma cell lines - Microarray analysis of norma] and carcinoma colon tissues revealed that the absence of RA responsi ve genes is paralleled by the down-regulation of two RA biosynthetic emzymes, RDH5 and DHRS9 (19). However, due to limited microarray data for DHRS9, the expression level of both DHRS9 and RDH5 were evaluated in ten additional matching normal and tumor colon tissues by RT-PCR. All tumor samples used in the experiment have elevated CYCLIN D expression relative to normal tissues, a positive marker for identifying colon carcinomas (data not shown) (22,23). Figure 2.1 shows that RDH5 expression is reduced in 6 of 10 carcinoma samples while DHRS9 has a more robust down-regulation, it is decreased by at least a factor of 2 in 9 of 10 samples examined. Colon tumor cell lines were also analyzed by Northern blot, and Figure 2.2A reveals that both DHRS9 and RDH5 expression are barely detectable to completely absent in the carcinoma cell lines compared to normal colon. Colon tumor cells have impaired RA biosynthesis - To investigate whether the absence of DHRS9 and RDH5 in colon carcinoma cells is accompanied by a defect in RA biosynthesis, cells were treated with retinol, extracted with acidic hexane and analyzed by HPLC. Retinoic acid was chosen the endpoint of the experiment due to difficulties in extracting retinal, and the amount of extracted RA was normalized to both internal standard, TTNBP, and total protein content (24,25). Colon tumor cells were compared to normal human mammary epithelial cells (HMEC) since there are currently no primary colon epithelial cells in culture and HMEC has RDH5 levels similar to normal colon c o o (.) <.OO~CON<'oI'-<'oI'-~ o -~I'-.~CO ~CO~NN(N'f)N('Nf)lN-!)N<,'oN CO C'O E s- o c ..o..... -1 -2 -3 -4 a.> -5 :;>:::; -6 C'O a.> -7 s-a.> -8 en C'O -9 a.> s- -10 (.) ~ -10j -0 -40 o -70 LL . RDH5 <.OO~CON<'oI'-<'oI'-~ I'-COCO~N('f)('f)l!)<'oCO ~~~NNNNNNN ~r .. , , DHRS9 40 FIGURE 2.1. DHRS9 and RDH5 are down-regulated in colon adenomas and carcinomas. Quantitative RT-PR was performed on 10 matching human normal and adenoma (sample 267) or carcinoma (all other samples except 267) colon tissues for RDH5 and DHRS9. Gene expression level for each sample was normalized to 18S rRNA. Data are plotted as fold decrease in expression level in tumor sample relative to its matching normal colon. Values shown represent the mean ± S.D. 41 (data not shown). Figure 2.2B shows that all colon carcinoma cell lines have impaired RA biosynthesis when compared to HMEC. DHRS9 functions as a retinol dehydrogenase in colon tUl1zor cells - DHRS9 has been reported to be a bona fide retinol dehydrogenase in tracheobronchial epithelial cells and in epidermal keratinocytes (5,20). However, Chetyrkin et al. showed that in vitro, DHRS9 is more active as a 3a-HSD (21). To determine the prevalent enzymatic activity of DHRS9 in the colon, HCT 116 cells transfected with DHRS9 or RDH5 were treated with retinol, and the extracted retinoids were analyzed as described previously. The retinol dehydrogenase activity of DHRS9 was compared with RDH5, a known retinol dehydrogenase. Figure 2.3 shows that cells overexpressing DHRS9 or RDH5 made significantly more RA than control cells. Co-expression of RALDH2 with DHRS9 or RDH5, which catalyzes the oxidation of retinal to RA, further improved RA biosynthesis by at least 2-fold (Fig. 2.3). To confirm if the increase in RA levels is due to enhanced DHRS9 expression, mutant Df1RS9 and RDH5 constructs were generated. The DHRS9 mutation K184A is located at the active site while D290f1 is a reported single nucleotide polymorphism (SNP) for DHRS9 (26). The RDH5 mutant, S73F, is commonly found in patients afflicted with fundus albipunctatus and has been demonstrated to have diminished retinol dehydrogenase activity (27). HPLC analysis of retinol-treated HeT 116 cells overexpressing the mutant constructs revealed a reduction in RA biosynthesis by at least a factor of 3 when compared to cells transfected with wild type DI-JRS9 and RDH5 (Fig. 2.3). A. B. DHRS9 RDH5 GAPDH co 0.. 0 20 t: r- x o c E::t ..C...D... ~ e a: c.. 0) o E E'"'­c.. 16 co E '- o c N I ocoo U () W ~ I L.() o N o (5 U C\J LO I 0 o C\I () 0 n1 0 () () --I o .....J o r- I o .....J o co ---- I-U I co ~ () I CJ) N l­I o ::::c: a: 42 FIGURE 2.2. Down-regulation of DHRS9 and RDH5 in colon carcinoma cell lines is accompanied by poor conversion of retinol to RA. (A) Northern blot analysis of pooled normal colon and tumor cell lines for DHRS9 (top panel), RDH5 (middle panel) and GAPDH (bottom panel) for loading control (B) HMEC and colon tumor cell lines were treated with 100 nmol retinol for 12 h and analyzed for RA biosynthesis by HPLC. The amount of extracted RA is expressed as pmol RA normalized to TTNPB and total protein content. Values shown represent the mean ± S.D. -.. c "a:; Ca..O.c.l. C. z elf- Ef- -c.(. 0 ~ E ::s 0 E c.. 600 ...J :c c c::: ::r: o (f) N Cl c::: + ...J c::: + L() 43 FIGURE 2.3. DHRS9 metabolizes retinol to RA in colon tumor cells. RCT 116 cells overexpressing wild type DHRS9 (S9), RDH5 (5), RALDH2 (R) (black bars), mutant DHRS9 and RDH5 constructs (K184A and D290H, S73F, respectively, gray bars) were treated with 100 nmol retinol for 4 h and analyzed for RA biosynthesis by HPLC. The amount of extracted RA is expressed as pmol RA normalized to TTNPB and total protein content. Values shown represent the mean ± S.D. Statistical significance (*, ., ., A) was established using Student's t test (p < 0.05 with 95% confidence). 44 APC regulates DliRS9 expression and possibly RA biosynthesis - APC is a tumor suppressor gene that is mutated early in colon carcinogenesis (28-32). To determine the possibility of APC regulating RDH5 and DHRS9 expression, HT29 APC and ~-gal inducible cell lines were utilized. Upon induction with zinc chloride, these cells will express either wild type APC or ~-gal as control. Previous studies with the HT29- inducible cells showed that DHRS9, but not RDH5, is re-expressed when APC is induced (19). It has also been repOlted that both APC and DHRS9 share a similar pattern of induction, expression is highest after 24 h of zinc chloride treatment and starts to decrease by 48 h (19). To find out if induction of DHRS9 expression coincides with increased RA biosynthesis, HT29 APC-inducible cells were exposed to retinol after treatment with zinc chloride. HPLC analysis of retinoid extracts revealed a statistically significant increase in RA levels in APC-induced HT29 cells compared to un-induced control cells (Fig. 2.4). No significant RA biosynthesis was observed for ~-gal cells, with or without zinc chloride treatment. 2.4. Discussion Cellular response to retinoid action is governed not only by the retinoid nuclear receptors, but also by the availability of RA, the active ligand. Several groups have reported defects in RA biosynthesis in human tumors. Mira-y-Lopez et al. demonstrated that retinol conversion is in1paired in breast cancer cells compared to normal cells (33). Rexer et al. also showed that ALDH6 is absent in MCF-7, a breast tumor cell line that has been proven to be defective in RA biosynthesis, while Triano et ai. revealed that class I ADHs are significantly downregulated in invasive breast cancers (34,35). Reduced levels of endogenous RA in oral premalignant lesions have also been reported (36). 45 co 750- 0z.. * t::« oz 500- EO: :::to) ::c~ 0: 250- * 0 E Q. 0 ... .... ZnCI2 - + - + FIGURE 2.4. Re-introduction of APe induced DHRS9 expression and improved RA biosynthesis. HT29 APC (black bars) and ~-gal (gray bars) inducible cells were treated with 100 nmol retinol for 4 h after a 24 h treatment with zinc chloride or vehicle. RA biosynthesis was evaluated by HPLC. The amount of extracted RA is expressed as pmo} RA normalized to TTNPB and total RNA content. Values shown represent the mean ± S.D for two replicate experiments. Statistical significance (*) was established using Student's t test (p < 0.05 with 95% confidence). 46 Two RA biosynthetic enzymes, DHRS9 and RDH5, have been shown to be downregulated in colon tun10rs compared to normal tissue by microanay (19). Multi­tissue Northel11 blot analysis revealed that DHRS9 is robustly expressed in normal colon, with minimal expression in the spleen, placenta and lung while RDH5 is mostly expressed in the kidney and liver (19). Results presented in this chapter confirm that both DHRS9 and RDH5 are downregulated in colon carcinomas and tumor cell lines. Further evaluation of the RA biosynthetic capabilities of the colon tumor cells revealed a poor conversion of retinol to RA. Thus, investigation as to whether DHRS9 is the retinol dehydrogenase responsible for catalyzing retinol oxidation in the colon was undertaken. HPLC analysis of retinoid extracts from transfected HCTl16 cells treated with retinol revealed that cells overexpressing either DI-IRS9 or RDH5 made more RA than control cells, supporting the previous findings by Soref et al. and Markova et al. This was further confirmed by demonstrating that cells transfected with mutant DHRS9 and RDH5 showed reduced RA levels compared to cells overexpressing wild type enzymes. Co-transfection of RALDH2 with DHRS9 or RDH5 enhanced RA biosynthesis, suggesting that the second step in retinol metabolism, the conversion of retinal to RA, might be impaired in colon tumor cells as well. Adenomatous polyposis coli (APe) is a tumor suppressor gene that is mutated in 80% of sporadic colorectal tumors and in all cases of familial adenomatous polyposis (FAP), an inherited predisposition to colon cancer (30-32). The best characterized function of APC is its negative regulation of ~-catenin, a known effector of the Wnt signaling pathway (37,38). Wild type APC targets ~-catenin for ubiquitin-mediated proteolysis however, when APC is mutated, ~-catenin accumulates and translocates into the nucleus, 47 activating oncogenes such as MYC and CYCLIN Dl (29,39-41). While these genes can explain how cellular proliferation is promoted in the colon, few of the cun-ent APC/~­catenin target genes can account for loss of colonocyte differentiation, an event that is postulated to occur early in colon carcinogenesis (42,43), To explore the possibility that APC also regulates cellular differentiation in the colon by controlling RA biosynthesis, HT29 APC and ~-gal inducible cells were utilized. The results show that induction of APC leads to restoration of DHRS9 expression and this was accompanied by a cOITesponding increase in RA biosynthesis (19). It was further demonstrated that this APC regulated-RA biosynthesis is p-catenin independent and is mediated by the homeobox transcription factor CDX2 using reporter assays (19). These findings suggest that APC might regulate a separate pathway that controls the RA-mediated program of differentiation in the colon, distinct from the ~-catenin/Wnt signaling pathway. Retinoids have been implicated in maintaining normal colon function. Vitamin A­deficient rats displayed abnormalities in the colon such as decreased mucus production, hypersecretion, reduced absorptive function and expansion of the proliferative zone in colonic crypts (44-49). Retinoids have also been shown to be effective in preventing azoxyn1ethane (AOM)-induced colon tumors in rats, inhibiting cell proliferation and inducing cell proliferation and apoptosis in colon cancer cell lines (50-58). This work supports the impOltance of retinoids in normal intestinal development and the restoration of retinoid activity as a potential therapy for colon cancer. 2.5 References 1. Niles, R. M. (2004) Mutat. Res. 555, 81-96 2. Altucci, L., and Gronemeyer, H. (2001) Nat. Rev. Cancer 1, 181-193 48 3. Duester, G. (2000) Eur. 1. Biochem. 267,4315-4324 4. Hansen, L. A., Sigman, C. C., Andreola, F., Ross, S. A., Kel1off, G. J., and Luca, L. M. D. (2000) Carcinogenesis 21, 1271-1279 5. Soref, C. M., Di, Y.-P., Hayden, L., Zhao, Y. H., Satre, M. A., and Wu, R. (2001) 1. Bioi. Chem. 276, 24194-24202 6. White, J. A., Guo, Y.-D., Baetz, K., Beckett-Jones, B., Bonasoro, J., Hsu, K. E., Dilworth, F. 1., Jones, G., and Petkovich, M. (1996) 1. BioI. Chem. 271, 29922- 29927 7. Chen, H., Fantel, A. G., and Juchau, M. R. (2000) Drug Metab. Dispos. 28, 1051- 1057 8. UIven, S. M., Gundersen, T. E., Weedon, M. S., Landaas, V. 0., Sakhi, A. K., Fromm, S. H., Geronimo, B. A., Moskaug, J. 0., and Blomhoff, R. (2000) Dev. BioI. 220, 379-391 9. Soprano, D. R., Qin, P., and Soprano, K. J. (2004) Annu. Rev. Nutr. 24,201-221 10. Widschwendter, M., Berger, 1., Hermann, M., Muller, H. M., Amberger, A., Zeschnigk, M., Widschwendter, A., Abendstein, B., Zeimet, A. G., Daxenbichler, G., and Marth, C. (2000) 1. Natl. Cancer Inst. 92,826-832 11. Deltour, L., Foglio, M. H., and Duester, G. (1999) Dev. Genet. 25,1-10 12. Deltour, L., Foglio, M. H., and Duester, G. (1999) 1. BioI. Chenl. 274, 16796- 16801 13. Driessen, C. A. G. G., Winkens, H. J., Hoffmann, K., Kuhlmann, L. D., Janssen, B. P. M., Vugt, A. H. M. V., Hooser, 1. P. V., Wieringa, B. ,Deutman, A. F., and Palczewski, K. (2000) Mol. Cell. BioI. 20,4275-4287 14. Molotkov, A., Fan, X., Deltour, L., Foglio, M. H., Martras, S., Farres, J., Pares, X., and Duester, G. (2002) PNAS 99,5337-5342 15. Duester, G., Mic, F. A., and Molotkov, A. (2003) Chem. BioI. Interact. 143-144, 201-210 16. Niedeneither, K., Subbarayan, V., Dolle, P., and Chambon, P. (1999) Nat. Genet. 21,444-448 17. Mic, F. A., Haselbeck, R. J., Cuenca, A. 129, 2271-2282 and Duester, G. (2002) Developnwnt 18. Begemann, G., Schilling, T. F., Rauch, G.-J., Geisler, R., and Ingham, P. W. (2001) Development 128, 3081-3094 49 19. Jette, C., Peterson, P. W., Sandoval, 1. T., Manos, E. J., Hadley, E., Ireland, C. M., and Jones, D. A. (2004) 1. Bio!. Chem. 279,34397-34405 20. Markova, N. G., Pinkas-Sarafova, A., Karaman-Jurukovska, N., Jurukovski, V., and M.Simon. (2003) Mol. Genet. Metab. 78, 119-135 21. Chetyrkin, S. V., Belyaeva, O. V., Gough, W. H., and Kedishvilli, N. Y. (2001) 1. Bio!. Chent. 276, 22278-22286 22. Arber, N., Hibshoosh, H., Moss, S. F., Sutter, T., Zhang, Y., Begg, M., Wang, S., Weinstein, 1. B., and Holt, P. R. (1996) Gastroenterology 110, 669-674 23. Bartkova, J., Lukas, J., Strauss, M., and Bartek, J. (1994) Int. 1. Cancer 58, 568- 573 24. Groenendjik, G. W., Grip, W. J. d., and Daemen, F. J. (1980) Biochim. Biophys. Acta. 617, 430-438 25. Groenendjik, G. W., Grip, W. J. d., and Daemen, F. J. (1979) Anal. Biochem. 99, 304-310 26. www.ncbi.nih.gov. 27. Liden, M., Romert, A., Tryggvason, K., Persson, B., and Eriksson, U. (2001) 1. Bio!. Chem. 276, 49251-49257 28. Radtke, F., and Clevers, H. (2005) Science 307, 1904-1909 29. Sancho, E., BatIe, E., and Clevers, H. (2004) Annu. Rev. Cell. Dev. Biol. 20, 695- 723 30. HouIston, R. S. (2001) Mol. PatlIol. 54,206-214 31. Bodmer, W. (1999) Cytogenet. Cell Genet. 86,99-104 32. Kinzler, K. W., and Vogelstein, B. (1996) Cell 87, 159-170 33. Mira-Y-Lopez, R., Zheng, W. L., Kuppumbatti, Y. S., Rexer, B., Jing, Y., and Ong, D. E. (2000) 1. Cell. Physiol. 185, 302-309 34. Triano, E. A., Slusher, L. B., Atkins, T. A., Beneski, J. T., Gestl, S. A., Zolfaghari, R., Polavarapu, R., Frauenhoffer, E., and Weisz, J. (2003) Cancer Res. 63,3092-3100 35. Rexer, B. N., Zheng, W. L., and Ong, D. E. (2001) Cancer Res. 61, 7065-7070 36. Lotan, R. (1980) Biochim. Biophys. Acta. 605,33-91 50 37. Molenaar, M., Watering, M. v. d., Oosterwegel, M., Peterson-Maduro, J., Godsave, S., Korinek, V., Roose, J., Destree, 0., and Clevers, H. (1996) Cell 86, 391-399 38. Behrens, J., Kries, J. P. V., Kuhl, M., Bruhn, L., Wedlich, D., Grosschedl, R., and Birchmeier, W. (1996) Nature 382, 638-642 39. Nathke,1. S. (2004) Annu. Rev. Cell Dev. Bio!. 20, 337-366 40. Korinek, V., Barker, N., Morin, P. J., Wichen, D. V., Weger, R. d., Kinzler, K. W., Vogelstein, B., and Clevers, H. (1997) Science 275, 1784-1787 41. Morin, P. J., Sparks, A. B., Korinek, V., Barker, N., Clevers, H., Vogelstein, B., and Kinzler, K. W. (1997) Science 275,1787-1790 42. Shanmugathasan, M., and Jothy, S. (2000) Pathol. Int. 50,273-279 43. Roncucci, L., Pedroni, M., Vaccina, F., Benatti, P., Marzona, L., and Pol, A. D. (2000) Cell Prolif. 33, 1-18 44. Nzegwu, H. C., and Levin, R. J. (1992) Gut 33,794-800 45. Zile, M., Bunge, C., and Deluca, H. F. (1977) J. Nutr. 107, 552-560 46. Narisawa, T., Reddy, B. S., Wong, C.-Q., and Weisburger, J. H. (1976) Cancer Res. 36, 1379-1383 47. Zile, M., and Deluca, H. F. (1970) Arch. Biochem. Biophys. 140,210-214 48. Perumal, A. S., Lakshmanan, M. R., and Cama, H. R. (1968) Biochim. Biophys. Acta. 170, 399-408 49. Perumal, A. S., Samy, T. S., Lakshmanan, M. R., Jungalwala, F. B., and Rao, P. B. (1966) Biochim. Biophys. Acta. 124,95-100 50. Adachi, Y., Itoh, F., Yamamoto, H., Iku, S., Matsuno, K., Arimura, Y., and Imai, K. (2001) Tumour Bio!. 22,247-253 51. Wargovich, M. J., Jimenez, A., McKee, K., Steele, V. E., Velasco, M., Woods, J., Price, R., Gray, K., and Kel1off, G. J. (2000) Carcinogenesis 21, 1149-1155 52. Lee, M.-O., Han, S.-Y., Jiang, S., Park, J. H., and Kim, S. J. (2000) Biochem. Pharmacal. 59,485-496 53. Nicke, B., Kaiser, A., Widennlann, B., Riecken, E.-O., and Rosewicz, S. (1999) Biochem. Biophys. Res. Commun. 261,572-577 54. Nicke, B., Riecken, E.-O., and Rosewicz, S. (1999) Gut 45,51-57 51 55. Zheng, Y., Kramer, P. M., Lubet, R. A., Steele, V. E., Kelloff, O. J., and Pereira, M. A. (1999) Carcinogenesis 20, 255-260 56. Reynolds, S., Rajagopal, S., and S.Chakrabarty. (1998) Cancer Lett. 134,53-60 57. Zheng, Y., Kramer, P. M., Olson, G., Lubet, R. A., Steele, V. E., Ke]]off, O. 1., and Pereira, M. A. (1997) Carcinogenesis 18,2119-2125 58. Niles, R. M., Wilhelm, S. A., Thomas, P., and Zamcheck, N. (1988) Cancer Invest. 6, 39-45 CHAPTER 3 STlTDIES CHARACTERIZING ZEBRAFISH RETINOL DEHYDROGENASES 3.1 Introduction Dania reria, commonly called zebrafish, has been used extensively in studying developmental biology and human disease. Its high fecundity, external fertilization, rapid developmental cycle and similarity to mammals, both morphologically and physiologically, make it an excellent experimental organism (1-4). In addition, the transparency of the embryos, which allows the observation of biological changes without having to dissect or sacrifice the organism, and the ease of maintaining a large number of zebrafish within a relatively small space further contribute to its appeal as an animal model system (1-4). The role of retinoic acid in zebrafish embryogenesis has been investigated quite considerably and such studies have led to the identification of phenotypes associated with lack of retinoic acid. Through the use of retinoid receptor antagonists, it has been demonstrated that loss of RA signaling impaired the development of both pancreas and liver (5). Inhibiting RA biosynthesis also disrupted fin, ventral retina and photoreceptor development and led to the formation of cardiac edema (6-9). Furthermore, zebrafish mutants carrying retinoid gene mutations have been described. The neckless (nls) zebrafish mutant possesses an inactivating mutation in raldh2, an enzyme that oxidizes 53 retinaldehyde to the active RA (10). It exhibits phenotypes that are consistent with vitamin A deficiency such as absence of pectoral fins, defects in branchial musculature, craniofacial skeleton and caI1ilage development (10). nls also phenocopies zebrafish treated with inhibitors of retinoid signaling and biosynthesis (11). Another raldh2 mutant, no~fin (no/), also shows loss of pectoral fins, deformed jaw and branchial arches, in addition to pericardial edema and impaired spinal cord development (12). Finally, targeted gene disruption of ~,~-carotene-15,15' -oxygenase, a key enzyme in provitamin A conversion, resulted in a bcox morphant that displays craniofacial skeleton malfollnation, defects in pectoral fin and retina development (13,14). The zebrafish has also beconle a good model organism for studying vertebrate intestinal development. Characterization of the zebrafish gut tube morphogenesis and architecture, using histological and immunohistochemical analyses of the different developmental stages of the zebrafish, revealed several similarities with mammalian intestinal development. Conlparable to the mammalian intestine, the zebrafish gut is a self-renewing tissue system composed of an elongated tube filled with epithelial folds and lined with polarized epithelial cells, with a simple muscle layer underlying the epithelium (15-17). However, in contrast to higher vertebrates, the zebrafish gut is stomachless, does not have a division into small and large intestines, and paneth cells are missing - only absorptive enterocytes, goblet cells and enteroendocrine cells have been identified, with a f0U11h epithelial cell type characterized by a prominent supranuclear vacuole which stores pinocytosed luminal contents (15-17). Several mutants that display intestinal differentiation defects have also been identified. slimjinl (slj), flotte lotte (flo), piebald (pie), m750, no relief and straight shot( sst) all appear to have an undifferentiated 54 gut as illustrated by the absence of intestinal folds, polarized epithelial cells and differentiated gut cell types, including epithelial detachment and degeneration (l 17). Previous studies on APC-inducible HT29 colon carcinoma cells revealed that re­expression of wild type APC induced expression of DHRS9 (refened to as RDHL) and was accompanied by increased retinoic acid biosynthesis (18). In an attempt to further confirm this genetic relationship between APC and DHRS9, zebrafish homologues of DHRS9 were identified. Studies presented in this chapter demonstrate the retinol dehydrogenase activity of three novel zebrafish rdhs - rdhl, rdhll and rdh5. A discussion on the possible regulation of retinoic acid biosynthesis, and ultimately, zebrafish intestinal and retinal differentiation by APC, is also included. 3.2 Methodology Cell Culture. HCT 116 colorectal carcinoma cell line was cultured as recommended by the American Type Culture Collection (ATCC). Transient transfections. Lipofectamine 2000 (Invitrogen) was used to transfect RCT 116 cells according to the manufacturer's protocol. Cells were seeded at a density of 40,000 cells/well in 6-well plates and transfected the following day. Transfections were performed using 3.0 J,lg of V5-tagged rdhl, rdhll, rdh5 or empty vector for 24 h. Retinol treatl11ent and extraction. RCT 116 cells were treated wi th 100 nmol all-trans retinol (Sigma) at 80-90% confluence for 18 h (rdh 1, rdh 11) or 24 h (rdh5). Medi urn was removed and cells were scraped into Ix PBS for protein quantification. Following addition of 100 nmol internal standard, 4-(E-2-(5,6,7,8-tetrahydro-5,5,8,8-tetramethyl-2- naphthaJenyl)-I-propenyl) benzoic acid (TTNPB, Sigma), the medium was acidified with 6N HCl (0.03 X volume) and extracted twice with an equal volume of hexanes containing 55 0.1 mg/ml butylated hydroxy toluene (BHT). The resulting solution was mixed vigorously and spun down at 11,500 rpm for 20 min. The organic layer was saved, dried down under nitrogen, and stored at -200 C for HPLC analysis. HPLC analysis. Dlied extracts were reconstituted in 100 J.d 1: 1 DMSO/MeOH and analyzed by reversed-phase HPLC using an Agilent 1100 Series with a diode-array detector (Agilent Technologies). The retinoids were eluted on a Luna C18 analytical column (4.6 x 250 mm, 5 J-lm particle size, Phenomenex) with a gradient starting at 80% acetonitrile/ 20% 0.1 M ammonium acetate, pH 5.0, to 1000/0 acetonitrile in 40 min, with a flow rate of 1.5 ml/min at 350 nm. Identification and quantification of extracted retinoids were done by comparing elution times and peak areas with matching retinoid standards. Extracted retinoic acid (RA) was nonnalized to both ITNPB and sample protein content. Statistical analyses were performed using OraphPad Prism (v 4.03). 3.3 Results Studies on APC-inducible HT29 colon carcinoma cells revealed that re-introduction of wild type APC led to the induction of DHRS9 expression which was paralleled by an increase in retinoic acid biosynthesis (18). To further confirm this genetic relationship between APC and DHRS9, the zebrafish model system was utilized. Sequence analysis of DHRS9 led to the identification of two novel zebrafish homologues, rdhl and rdhll (19). rdhl and rdh11 share 47 and 45.5% sequence identity to DHRS9, respectively. However, tissue distribution analysis of both enzymes in adult fish revealed that rdh 1 is colon-specific, whereas rdh 11 is more ubiquitously expressed (19). To detern1ine whether the zebrafish rdh enzymes exhibit retinol dehydrogenase activity, V5-tagged rdh1 and rdhll constructs were made and overexpressed in HCTl16 cells. 56 Transfected cells were incubated with retinol for 18 hand retinoids were extracted with acidified hexanes. HPLC analysis of the retinoid extracts showed that HCT116 cells overexpressing rdhl or rdhll produced at least 4-fold more RA than control cells, thereby confirming their enzymatic activity towards retinol (Fig. 3.1A). Further analysis of zebrafish retinol dehydrogenases revealed a third novel rdh enzyme, rdh5 (20). In contrast to rdh 1 and rdhll, rdh5 is predominantly expressed in the eyes (20). To confirm the retinol dehydrogenase activity of rdh5, retinoid extracts from retinol-treated HCT116 cells transfected with V5-tagged rdh5 were analyzed by HPLC. Figure 3.1B shows that HCT116 cells overexpressing rdh5 made significantly more RA than control cells. 3.4 Discussion Three novel zebrafish retinol dehydrogenases were identified from phylogenetic analysis of the retinol dehydrogenase family members. rdhl and rdhll appear to be involved in RA biosynthesis in the gut whereas rdh5 may be important for maintaining RA levels in the zebrafish eye (19-21). HPLC analysis of cells overexpressing rdhl, rdh 11 or rdh5 resulted in increased RA biosynthesis, confirming their retinol dehydrogenase activity. Because of the homology of rdhl and rdhll to DHRS9, further studies were undertaken to verify previous results demonstrating a possible role for APC in controlling retinoic acid biosynthesis (18). Using gene-specific morpholino oligonucleotides, knock­down of rdh 1 and rdhll expression in zebrafish resulted in phenotypes recapitulating vitamin A deficiency. rdhl and rdhll morphants lack pectoral fins and jaw, have impaired pancreatic development, in addition to intestinal differentiation defects 57 A. --s:: ..O...J, em 0.0- 400 OlZ Et: 300 <-- -0 _a: :: :E:l 200 0 E 100 c.. B. Vector rdh1 rdh11 -r-:: 8900000000 * '.Q...3., 70000 000 c-..0z. 60000 E~ 50000 -- (5 40000 C2 E 30000 _ :::l 0 20000 Ec.. 10000 0 Vector rdh5 FIGURE 3.1. rdh1, rdhll and rdh5 convert retinol into retinoic acid in HCT116 cells. (A) HCT116 cells overexpressing empty vector, V5-tagged rdhl, rdhlI or (B) rdh5 were treated with 100 nmol retinol for 18 h or 24 h and analyzed for RA biosynthesis by HPLC. The amount of extracted RA is expressed as pmol RA normalized to TfNPB and total protein content. Values shown represent the mean ± S.D. Statistical significance (*) was established using Student's t test (p < 0.05 with 95% confidence). 58 (19,21). Interestingly, similar phenotypes were observed for ape morphants, which are consistent with the phenotypes reported by HurIstone et al. for the ape mutant zebrafish (22). rdh} , rdhll and ape morphants were all rescued with exogenous RA (19,21). These findings establish a role for APC in regulating RA levels by controlling the retinol dehydrogenases important for retinoid biosynthesis and also implicate RA in zebrafish intestinal development (19,21). One of the phenotypes accompanying FAP syndrome in humans is CHRPE, a congenital hypel1rophy of the retinal pigmented epithelium that does not affect sight and has no malignant potential (23-25). The identification of another novel zebrafish retinol dehydrogenase, rdh5, that is predominantly expressed in the eye, led to additional studies on ocular abnormalities in ape mutant and rdhS morphant zebrafish. Phylogenetic analysis revealed that rdh5 is the zebrafish ortholog of RDH5 and examination of cells transfected with rdh5 by HPLC established it as a bona fide retinol dehydrogenase (20). Knock-down of rdh5 expression in zebrafish resulted in eye developmental defects including retinal colobomas similar to that found in human FAP patients, and ventral retina abnormalities (20,26). The same phenotype was observed in the eyes of ape mutant zebrafish and pm1ial rescue of the retinal defects was observed with RA treatment (20). These observations support a model similar to that proposed for intestinal development, wherein APe regulates retinal differentiation by controlling rdh5, the retinol dehydrogenase that is essential for n1aintaining normal retinoid levels in the eye. APe is a tumor suppressor gene that is proposed to primarily function as a negati ve regulator of ~-catenin levels (25,27,28). The apc, rdhl, rdll and rdh5 knock-down studies in zebrafish reinforce a new role for APC in regulating the retinoid-induced program of 59 cellular differentiation both in the intestine and eyes. Because of the similarity between mammals and zebrafish, it may be that this novel function of APC in the zebrafish is conserved in humans as we11, a possibility that further emphasizes the potential of restoring retinoid biosynthesis and signaling as an effective therapeutic for colon cancer. 3.5 References 1. Dodd, A., Curtis, P. M., Williams, L. C., and Love, D. R. (2000) Hum. Mol. Genet. 9, 2443-2449 2. Pichler, F. B., Laurenson, S., Williams, L. C., Dodd, A., Copp, B. R., and Love, D. R. (2003) Nat. Biotechnol. 21, 879-883 3. Stem, H. M., and Zon, L. 1. (2003) Nat. Rev. Cancer 3, 1-7 4. Zon, L. 1., and Peterson, R. T. (2005) Nat. Rev. Drug Discov. 4,35-44 5. Stafford, D., and Prince, V. E. (2002) Curro Biol. 12, 1215-1220 6. Arn1strong, N. M.-., McCaffery, P., Gilbert, W., Dowling, J. E., and Drager, U. C. (1994) Proc. Natl. Acad. Sci. 91, 7286-7290 7. Hyatt, G. A., Schmitt, E. A., Fadool, J. M., and Dowling, J. E. (1996) Proc. Natl. Acad. Sci. 93, 13298-13303 8. Vandersea, M. W., Fleming, P., McCarthy, R. A., and Smith, D. G. (1998) Dev. Genes Evol. 208, 61-68 9. Perz-Edwards, A., Hardison, N. L., and Linney, E. (2001) Dev. Bioi. 229,89-101 10. Begemann, G., Schilling, T. F., Rauch, G.-J., Geisler, R., and Ingham, P. W. (2001) Development 128, 3081-3094 11. Begemann, G., Marx, M., Mebus, K., Meyer, A., and Bastmeyer, M. (2004) Dev. BioI. 271, 119-129 12. Grandel, H., Lun, K., Rauch, G.-1., Rhinn, M., Piotrowski, T., Houart, C., Sordino, P., Kuchler, A. M., Schulte-Merker, S., Geisler, R., Holder, N., Wilson, S. W., and Brand, M. (2002) Development 129,2851-2865 13. Biehlmaier, 0., Lampert, J. M., Lintig, 1. v., and Kohler, K. (2005) Eur. 1. Neurosci. 21, 59-68 60 14. Lanlpert, J. M., Holzschuh, J., Hessel, S., Driever, W., Vogt, K., and Lintig, J. v. (2003) Development 130, 2173-2186 15. Ng, A. N. Y., Jong-Curtain, T. A., Mawdsley, D. J., White, S. J., Shin, J., Appel, B., Dong, P. D. S., Stainier, D. Y. R., and Heath, 1. K. (2005) Dev. Biol. 286, 114- 135 16. Pack, M., Solnica-Krezel, L., Malicki, J., Neuhauss, S. C. F., Schier, A. F., Stemple, D. L., Driever, W., and Fishman, M. C. (1996) Developl11ent 123, 321- 328 17. Wallace, K. N., Akhter, S., Smith, E. M., Lorent, K., and Pack, M. (2005) Mech. Dev. 122, 157-173 18. Jette, C., Peterson, P. W., Sandoval, I. T., Manos, E. J., Hadley, E., Ireland, C. M., and Jones, D. A. (2004) J. Biol. Chem. 279, 34397-34405 19. Nadauld, L. D., Sandoval, 1. T., Chidester, S., Yost, H. J., and Jones, D. A. (2004) J. Biol. Chem. 279,51581-51589 20. Nadauld, L. D., Chidester, S., Shelton, D. N., Sandoval, 1. T., Peterson, P. W., Manos, J., Ireland, C. M., Yost, H. J., and Jones, D. A. (2005) J. Biol. Chem. Submitted 21. Nadauld, L. D., Shelton, D. N., Chidester, S., Yost, H. J., and Jones, D. A. (2005) J. Biol. Chem. 280, 30490-30495 22. Hurlstone, A. F. L., Haramis, A.-P. G., Wienholds, E., Begthel, H., Korving, J., Eeden, F. v., Cuppen, E., Zivkovic, D., Plasterk, R. H. A., and Clevers, H. (2003) Nature 425, 633-637 23. Rowley, P. T. (2005) Annu. Rev. Med. 56, 539-554 24. Feamhead, N. S., Wilding, 1. L., and Bodmer, W. F. (2002) Br. Med. Bull. 64,27- 43 25. Fearnhead, N. S., Britton, M. P., and Bodmer, W. F. (2001) Hum. Mol. Genet. 10, 721-733 26. Kermane, A., Tachfouti, S., Moussaif, H. Oplztalmol. 292,59-64 and Mohcine, Z. (2004) Bull. Soc. 27. Sancho, E., BatIe, E., and Clevers, H. (2004) Annu. Rev. Cell. Dev. Biol. 20,695- 723 28. Nathke,1. S. (2004) Annu. Rev. Cell Dev. Bio!. 20,337-366 CIIAPTER4 INVESTIGATION OF TIlE RETINOID BIOSYNTHETIC PA THW A Y IN ApCIllin/ + MOUSE 4.1 Introduction Adenomatous polyposis coli (APe) is a tumor suppressor gene that is mutated in approximately 80% of sporadic colorectal cancers 0-4). It has been proposed that loss of APC function is sufficient to initiate colon carcinogenesis and this is supported by studies showing that mutations in APe are present in aberrant crypt foci, the earliest neoplastic lesion that can be examined (l,4-6). It is also the initiating mutation in familial adenomatous polyposis (PAP), an inherited predisposition to colon cancer that is characterized by the appearance of hundreds to thousands of benign polyps in both colon and rectum, which can progress to adenomas if left untreated (1-3,7-9). It is therefore important to develop animal model systems that can recapitulate the phenotype associated with APe mutation to further elucidate the molecular mechanisms accompanying colon tumor progression. The C57BU6Jlllill /+ mouse is a strain derived from C57BL/6J that was established during an ethylnitrosourea-induced mutagenesis experiment (10). C57BL/6Jltlin /+ animals CatTY a mutant gene, multiple intestinal neoplasia (Min), that is dominantly expressed, fully penetrant, and predisposes the afflicted animals to develop numerous adenomas throughout the intestinal tract, mostly in the small intestine (10,11). Linkage analysis 62 showed that murine Ape gene is tightly linked to the Min locus and sequence analysis of Ape from Min mice revealed a nonsense mutation converting codon 850 from a leucine (ITO) to a premature stop codon (TAO), resulting in a truncated Apc protein of about 95 kDa (3,9). Because this defect is the same initiating mutation found in human FAP syndrome as well as in majority of sporadic colorectal cancers, C57BL/6J-Apclllilll + has become an established animal model systen1 for studying intestinal tumorigenesis. Homozygous Min mutation is embryonic lethal (1-3,9,12). The ApCltlilll + lineage is propagated by mating wild type female with ApCllli/l/+ male, since intestinal polyps intelfere with pregnancy (13). And while ApClllilll + animals rarely survive beyond 17 weeks, this model system has been utilized in several studies to reveal genes involved in colon carcinogenesis and demonstrate the effect of known carcinogens and chemotherapeutic drugs to polyp incidence and tumor progression (14-18). The dysregulation of the retinoid biosynthetic pathway in ApClllilll + animals was investigated by evaluating the expression level of murine retinoid genes in adenomatous polyp and normal tissues (19). A similar experiment has been performed using human tissue samples, and data from this investigation would further validate the use of ApClllitll + mice in studying colon tumorigenesis, in addition to characterizing any similarities between small intestinal and colonic polyps (20). 4.2 Methodology Mouse breeding and tissue harvest. C57BU6J-+/+ female and C57BL/6J-Apclllilll + male mice were purchased from The Jackson Laboratory (Bar Harbor, ME), The animals were kept and bred at the Huntsman Cancer Institute Animal Facility in a room with a 12 h light/dark cycle under controlled temperature (220 C) and humidity (300/0). The 63 offspring were genotyped using a PCR protocol recommended by The Jackson Laboratory to differentiate wild type from ApClllill/ + pups (13). Water and feed were given ad libitum. The pups were fed with a high fat Capecchi Diet 3080 (Harlan Teklad) after weaning to promote formation of intestinal polyps. By 12-16 weeks of age, the animals were killed by CO2 asphyxiation. The abdomen was cut open and the small intestine, from just below the stomach to caecum, and colon, from caecum to rectum, were harvested. The intestines were cut longitudinally, washed with ice-cold IX phosphate-buffered saline (PBS) then examined visually for presence of polyps. Tumors from the small intestine and colon were counted and excised. Tissue samples were stored in RNAlater (Qiagen) at - 80° C to avoid degradation. Quantitative RT-PCR. Total RNA was isolated from normal tissue harvested from C57BU6J-+I+ animals and polyps from C57BL/6J-Apcmilll + animals using an RNeasy kit (Qiagen). cDNA was synthesized from 2 Jlg of total RNA using Superscript III (Invitrogen). Quantitative RT-PCR was performed using the Roche Light Cycler instrument and software, version 3.5 (Roche Applied Bioscience), Primer sets were designed to amplify 200-bp products. Table 4.1 lists all the primers and annealing temperatures used in the experiment. PCR was performed in duplicate using the LightCycler FastStart DNA MasterPLUS SYBR Green I kit (Roche Applied Bioscience) under the following conditions: 1 cycle at 95° C for 10 min followed by 35 cycles of amplification at 95° C for 10 s, Tanncaling for 5 s, 72° C for 10 s followed by 1 cycle of melting at 95° C for 0 s, 65° C for 15 s, 95° C for o s followed by 1 cycle at 40° C for 30 s. A template-free negative control was included Table 4.1. Primers and annealing temperatures used for RT-PCR of mouse tissue samples Gene Primers (5' - 3') Ta,oC I8S F- acgtctgccctaacaactttc 52 R - gcctc gaaagagtcctgtatt Adhl F- gttcgaattaagatggtggcc 52 R - atcctgcattctccacactga Adh2 F- gtgtgtaggtctttctgccat 52 R - agtcaagggagtaatccacac Adh4 F-taaaactgccaaggtcacccc 54 R- tgatgggcttggtagagtcct Crad2 F- acatctgacaggctggagaca 54 R- cccaacaggttcacatccagt Crad-L F- tgttgggaacagaggactctg 54 R - ccaagatgctggagac gttga Dhrs9 F- aaaggcttccgtgtcattgct 52 R- acgccaagaacaccagcatta Rdh5 F- ttctctgacagcctgaggcgg 56 R- cggcgctgtactcgaagataa Rdh6 F- catgttgggaacagaggactc 54 R- atgacactggagacgttgacc Retsdrl F- gcatgagagtcaggtttccca 54 R - catacaggtgtaagtcccc ga Raldhi F- ccgacttggacattgctgttg 54 R - cttgtcaatctgagggccttg Raldh2 F- gaagtaacctgaagagagtga 54 R- cctttccacacttcttttcac Raldh3 F- ctcatcaaagaggtcgggttc 56 R - ttcttgcctcctagctccagt Rara F- tgagggctgtaagggcttctt 54 R - gcttgggtgcctctttcttct Rar~ F- gaaagcccaccaggaaacctt 54 R- tgatctggtctgcgatggtca Rary F- ctctgtggagaccgaatgga 54 R - ccatcttcagggttatagccc Rxra F- gctcaccaaatgaccctgtta 52 R- gagaatcccatctttcacagc Rxr~ F- gaaaagggaggc ggttcagga 54 R - tcactgggtcatttgggctgc RXll' F- tgtttaacccagatgccaagg 52 R - gagcttgaagaagaagaggtg Cyp26Al F- gataaaagcaagggcttact 50 R- aaataacattccagcccttgg 64 65 in each experiment. The copy number was measured by comparing gene amplification with the amplification of standard samples containing 103 to 107 copies of the gene. The relative expression level of each gene was normalized to 18S rRNA to account for variations in the amount of input cDNA. 4.3 Results }\':>e tl•n ol'd genes are {.I y sregu l ated 'li Z a de nomatous po1 y ps fir om A pcl IIilll+ mice - Th e expression level of murine retinoid genes, including retinol dehydrogenases (Rdh), retinal dehydrogenases (Raldh), retinoid receptors (Rars, Rxrs) and Cyp26Al was investigated in ApCl1lill/ + polyps and wild type nOlmal tissue (19). RT-PCR of adenomatous polyps from the colon and small intestine revealed downregulation of several alcohol dehydrogenases and short chain alcohol dehydrogenases as shown in Figures 4.1 and 4.2. Adh2 is downregulated in small intestinal adenomas while Adh4 expression is decreased in small intestinal adenomas but increased in colon adenomas (Fig. 4.1), Cis-retinol/ androgen dehydrogenases (Crad) are short chain alcohol dehydrogenases that have been identified only in the mouse and are more active towards hydroxysteroids than retinol (19). Crad-2 is downregulated in small intestinal adenomas but upregulated in colon polyps, whereas no difference in expression level was detected for Crad-L in both tissues examined (Fig. 4.2). Retsdrl expression is diminished in adenoma samples while Rdh5 and Rdh6 transcript levels showed corresponding decrease and increase in expression only in colon adenomas (Fig. 4.2). Dhrs9 is the murine homolog of DHRS9 based on phylogenetic analysis (A.2). Gene expression analysis of Dhrs9 in ApCltlilll + mice revealed that it is upregulated in the adenoma samples compared to normal tissue 66 Adh1 Small Intestine Colon ceon 900 ... en SOD • 800 co ,... 800 • "',Q....l 700 •• '"Q .l 700 • 6:00 ,... -. -. 6:00 ","" L.. 500 L.. SOD .. Q.l Q.l .0 - .0 E 400 ... E 400 -----.-.- ....... :l 300 --.;-- I. :l 300 Z ... z --..!:- >c... 200 ----4--.-.. >. 200 • .... 100 • • i..t. A c.. 0 '100 .- 0 •••• C) ...... C) () 0 N A N A Adh2 * Small Intestine Colon en 120 en 25 co 110 ,... • co '"Q .l 100 "'Q.l 20 - ... ,... 90 ,... -. BO -. .... 15 Q.l 70 (j) • .0 60 .0 E 50 E 10 :l ... :l ---- ~ Z 40 • Z :..: .. >. 30 ......... ... ~ >. c.. 20 I. ~ c.. • "- 0 I •• 0 C) -10 ...... C) n 0 N A N A Adh4 * Small Intestine * Colon en 10.0 • en 12 ,c.o.. ... ,c.o.. 11 10 • "'Q.l ... 7.5 "'Q.l $ ,... .. ,... .. -.Q.....l ---- -.Q.....l 18 Mt.~ .0 5.0 • ... .0 S .. - ~ E • ... E ;.. ~ 5 :l :l 4 --..... .. z -.. L .... z •• >. 2.5 .". >. 3 • "- c.. c.. 2 •• ... C0) C0) 1 0.0 0 N A N A FIGURE 4.1. Expression level of alcohol dehydrogenases in ApC1llilll + nlouse. Quantitative RT-PCR was performed on total RNA from normal (N, _) (colon, 11 = 11; small intestine, 11 = 10) and adenoma tissue CA, .. ) (colon, n == 12; small intestine, n = 13) harvested from wild type and ApCmill/ + aninlals, respectively, for alcohol dehydrogenases. Data were plotted as gene copy number normalized to 18S rRNA. Values shown represent the mean for two replicates. Statistical significance (*) was established using the Mann-Whitney test (p < 0.05 with 95% confidence). 67 Crad-2 Crad-L * Small Intestine * Colon Small Intestine Colon (c(o) 1750 (c(o) 200 • c0o0 250 • 0co0 400 • 1500 "'Q) 150 "'Q) 200 "'Q) ... '" 300 •• .l • Q) -..... 1250 -..... • 150 • .. .. "Q-) 1D "- ----L- a> a> 200 .8100 ... ..0 ....... ..0 ..0 .. E 760 E • E 100 .. E ::l ::l •• • ::l ....... ::l Z 500 Z 60 ~• • z 00 .- Z -\00 ••• -rr- >- -.:r ." >- ..... >- II .- • • ......... >- ... I- 0- 250 0- • 0- I.- ........ 0- -.- 0 II • -«* 0 0 ... 0 --- ! ... 0 .. 0 0 0 0 0 (l N A N A N A N A Retsdr1 Dhrs9 * Small Intestine * Colon * Small Intestine * Colon (() 550 (() 400 00 110 • 00200 .. co 500 co co II .c..o.. 100 II "'Q) 450 "'Q) IOQ) 90 ~150 400 ..... 300 -..... 80 ... ... 350 - "- 70 "- a> "- Q) Q) .... 300 Q) ..0 .. ..0 200 .E.0 6500 ... ~1oo II ... E 250 • E ... --"-- ::l 200 • ::l ::l 40 II ~ ::l ... Z .............. Z Z Z .- >- 150 ~ >- 100 >- 30 ... ~50 0- 100 II. • 0- ,.,.. It. ...... 0- 20 -.- ...... f ... IIr1'j..• ...... 0 50 • ... 0 0 0 ... ..... .... -t?t 0 10 .,.,.,. 0 0 ... ... 0 0 ... 0 0 0 N A N A N A N A Rdh5 Rdh6 Small Intestine * Colon Small Intestine * Colon (() 35 00 50 II (() 65 (() 130 .. co .c..o.. co GO ... co 120 "'Q) 30 • IOQ) 55 ..... 110 40 "'Q) ..... 50 II IOQ) 100 25 ..... ..... 45 - - -..... 90 a> • "- 30 a> 40 - "- 80 20 • Q) ..0 • ... ..0 35 - ..... Q) 70 •• ..0 ..0 E • E II E 30 • • E 60 ... ::l 15 ... ::l 20 II 25 ... 50 ... Z ... Z ::l • --4.-- ::l 10 ..!..a.!- Z 20 Z 40 >- ... ~ 10 >- 15 • •• ...... i >- 30 • 0- .. 0 0 0- 10 II 0- 20 i!;' ~ • ... •• 0 0 o 5 .... 0 o 0 0 10 ~ ...... 0 0 N A N A N A N A FI GlTRE 4.2. Expression level of short chain alcohol dehydrogenases in Apclllilll + mouse. Quantitative RT-PCR was performed on total RNA from normal (N, _) (colon, 11 = 11; small intestine, n = 10) and adenoma tissue (A, A) (colon, 11 = 12; small intestine, 11 = 13) harvested from wild type and ApClllilll+ animals, respectively, for short chain alcohol dehydrogenases. Data were plotted as gene copy number normalized to 18S rRNA. Values shown represent the mean for two replicates. Statistical significance (*) was established using the Mann-Whitney test (p < 0.05 with 950/0 confidence). 68 (Fig. 4.2). Retinal dehydrogenases, which catalyze the in-eversible oxidation of retinal to the active metabolite, retinoic acid (RA), are also downregulated in Apclllilll + animals (19). Raldh 1 expression in upregulated in small intestinal adenomas but is remarkably reduced in colon polyps (Fig. 4.3). Raldh3, on the other hand, showed approximately a 6-fold increase in transcript level in colon adenomas (Fig. 4.3). The expression level of retinoid receptors was also analyzed to determine any defect in RA signaling in ApClllill/ + mice. RanI, Rar~ are both significantly upregulated in colon adenomas while Rxr~ expression is reduced in small intestinal adenomas but increased in colon polyps (Fig. 4.4). Finally, the transcript level of Cyp26A1 was evaluated in adenomatous polyps to investigate abnormal metabolism of RA. Figure 4.5 shows that Cyp26A1 expression is decreased by a factor of 2 in colon adenomas compared to normal tissue. 4.4 Discussion The ApClllilll + animal model system shares the initiating APe mutation that is responsible for the human FAP syndrome and a greater part of sporadic colorectal cases (1-3). In addition, ApClllill/ + animals spontaneously develop adenomatous polyps in the intestinal tract (10). Because of these, the ApClllilll + mouse is regarded as an excellent model system for studying colon tumorigenesis and indeed, numerous groups have utilized it to determine gene expression changes accompanying colon tumor development and identify factors that affect polyp formation and progression (14-18). However, one major difference between murine and human FAP is that ApClllilll + mice acquire polyps mostly in the small intestine while patients afflicted with PAP develop polyps in the 69 Raldh1 * Small Intestine * Colon 400 750 0,c.o0. ;. ,c0.o0. • lOCI) 300 lOCI) ,.. • - I. ,.. 500 - "- "- CI) • CI) .c .c E .....;.... E :::l :::l 250 --....... z • z ••• >- :. -'-fA >- .... Q. Q. 0 *1. 0 .... 0 0 0 N A N A Raldh2 Small Intestine Colon 4 • 00 .t- oo ,c.o. ,c.o. • lO,C.I.) lO,C.I.) • - - I.'" "CI-) Q; .c ..... .c • E • E :::l • :::l ..,... z -..:r- z ~ >- >- •• ··"'t • Q. Q. 0 "'. 0 ! •• .... 0 0 0 .'" N A N A Raldh3 Small Intestine * Colon 00 00 350 ,c.o. .. ,c.o. 300 ... lO,C.I.) lO,C.I.) 2.50 ..... "- "- 200 ... CI) CI) .c .c E E :::l .- ... :::l .. Z >- •• ." z> - ".. Q. • ............... Q. .!. 0 ... 0 0 0 N A N A FIGURE 4.3. Expression level of retinal dehydrogenases in ApC11lilll + mouse. Quantitative RT-PCR was pelformed on total RNA from normal (N, .) (colon, 11 = 11; small intestine, 11 = 10) and adenoma tissue (A, A) (colon, 11 = 12; small intestine, 11 = 13) harvested from wild type and ApClllilll + animals, respectively, for retinal dehydrogenases. Data were plotted as gene copy nUInber normalized to 18S rRNA. Values shown represent the mean for two replicates. Statistical significance (*) was established using the Mann-Whitney test (p < 0.05 with 95% confidence). 70 Rara Rxra Small Intestine * Colon Small Intestine Colon e,0.n.0. 111000 - 0en0 112300 • en 200 en 200 ';-110 00 - 00 ... It> 90 ,... (\) 80 (\) 100 .,....... 70 :,..:. 9800 It>(\) 150 • It>( \) 150 aa ... ,... 4i ... ". a. I. ..... -.- 60 (\) 70 '- ~ 100 .0 50 .. .0 60 (\) 100 ..... ... E E 50 • ~ .0 ----;r ...... :::J 40 • • ... ..... ~ 40 E ........ E Z 3G --. ~ .:.-.- ..... :::J :::J ~ >0- 20 .-. >0- 30 Z 50 •• ... ...... Z 50 • ... e.. • ... .. e.. 20 >0- .' >0- .... ... 0 10 o 10 e.. I. e.. U 0 Il/10.,'''' ,\ U 0 0 ... 0 U U 0 i I N A N A N A N A Rarp Rxrp Small Intestine * Colon * Small Intestine * Colon en 35 en 40 en 110 • en 75 • ,0..0. 30 00 ... ,0..0. 100 ,0..0. 90 It>(\) 25 • It>(\) 30 ... ... It>(\) 80 "'(\) ,... ,........ ,... ..... 70 ~50 20 '- .. ..'.-.. • ... 4i 6.0 ('\-) .: . .(0\) .8 20 .0 50 .0 E 15 ~ E E §25 .A.u.- • ... :::J 40 .,.. :::J ... -.:-:- i ...... 10 :::J Z ~ .... " Z 10 - f .... Z>0 - 3200 -'. ~ Z>0 - M >0- S " U0e. . .' ... i. ... U>e0. - . 0 ... Ue0. . 100 • Ue0. . 0 ... N A N A N A N A Rary Rxry Small Intestine Colon Small Intestine Colon en 4 en 50 en 70 en 100 00 00 ... 00 • ,... ,... ,... 60 ,0..0. • It>( \) "'(\) 40 ". . "'(\) co(\) ,... • ,... 50 ,... 75 ..... :: :10 • ..... ..... ... '- • 4i 40 '- (\) (\) .- " (\) .0 ..c .0 .0 50 E § 20 • E 30 • E • . :::J :::J 20 t;~ ... :::J Z Z -.-- .... Z rfYtttt' Z 25 v. ...· :!T e>0..- ~10 • .... e>0..- 10 - ~",-,.. >e.-. • 0 0 .... .. 0 0 u U l\ .. U 0 U 0 N A N A N A N A FI GURE 4.4. Expression level of retinoid nuclear receptors in ApC11lilll + mouse. Quantitative RT-PCR was performed on total RNA from normal (N, _) (colon, 11 = 1]; small intestine, 11 = 10) and adenoma tissue (A, ~) (colon, 11 = 12; small intestine, 11 = ]3) harvested from wild type and ApClllill/ + animals, respectively, for retinoic acid receptors (Rar) and retinoid X receptors (Rxr). Data were plotted as gene copy number normalized to 18S rRNA. Values shown represent the mean for two replicates. Statistical significance (*) was established using the Mann-Whitney test (p < 0.05 with 95% confidence). 71 Cyp26A1 Small I ntesti ne * Colon 5 0.9 en A en 0.8 ,0..0. 4 • ...... ... ,0..0. 0.7 to ...... to .,.C.....1..). • ...... ,C..1.) O.S ......... 3 -. ....... ... a.. a.. C1) C1) 0.5 .0 • .0 .- E E 0.4 z:::J 2 z:::J 0.3 •• • ... ...... >- >- A. ... c.. c.. 0.2 - 0 1 A. 0 -.