University of Southern California Dissertations and Theses

Cyclooxygenase-2 as a candidate biomarker for progression, prognosis and chemoprevention in various tumor types

(USC Thesis Other)

Cyclooxygenase-2 as a candidate biomarker for progression, prognosis and chemoprevention in various tumor types

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content CYCLOOXYGENASE-2 AS A CANDIDATE BIOMARKER FOR PROG RESSIO N, PROGNOSIS AND CH EM OPREVENTION IN VARIO US TU M O R TYPES by Ji Min Yochim A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (BIO CHEM ISTRY AND MOLECULAR BIOLOGY) May 2003 Copyright 2003 Ji Min Yochim Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UMI N um ber: 3103985 Copyright 2003 by Yochim, Ji Min All rights reserved. ® UMI UMI Microform 3103985 Copyright 2003 by ProQuest Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code. ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UNIVERSITY OF SOUTHERN CALIFORNIA THE GRADUATE SCHOOL UNIVERSITY PARK LOS ANGELES, CALIFORNIA 90089-1695 This dissertation, written by J i H tn Y ochim under the direction o f h £ jr_ dissertation committee, and approved by a ll its members, has been presented to and accepted by the D irector o f Graduate and Professional Programs, in p artial fulfillm ent o f the requirements fo r the degree o f DOCTOR OF PHILOSOPHY Director Date May 1 6 , 2003______ Dissertation Cdfhmittee Chair Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. DEDICATION Dedicated to God, who has blessed and protected me in every way during my time at school, and to my husband, who has always been with me, giving me tremendous support and praying for me. “ When you pass through the waters, I will be with you; and when you pass through the rivers, they will not sweep over you. When you walk through the fire, you will not be burned; the flames will not set you ablaze.” Isaiah 43:2 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ACKNOWLEDGEMENTS I would like to thank my mentor, Dr. Peter Danenberg, for his guidance, support, patience, and encouragement during my tenure in the Ph.D. program. He has been an exceptional scientist. I would also like to thank Dr. Jan Brabender and Dr. Sylke Schneider for their encouragement and for their help in medical information, Kathy Danenberg and Response Genetics Inc. personnel for their technical advice and help, and Dr. Susan Groshen, W ei Ye, and Jesse Lin for their help with statistical analysis. In addition, I would like to thank my committee members, Dr. Robert Haile and Dr. French Anderson for their advice, guidance, concerns, understanding, and time. Finally, I would like to thank all my friends who gave me advice, showed their support, and prayed for me. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS D E D IC A T IO N .......................................................................................................................ii AC K N O W LED G EM EN TS................................................................................................iii LIST OF T A B L E S ............................................................................................................viii LIST OF F IG U R E S ............................................................................................................ix A B S T R A C T ........................................................................................................................xii CHAPTER I ..........................................................................................................................1 IN T R O D U C T IO N ........................................................................................................... 1 1-1. COX Isoforms..................................................................................................1 I-2. Regulation of COX-2 Expression..............................................................2 I-3. Evidence of COX-2 Involvement in Tum origenesis............................6 I-4. Mechanisms of COX-2 Contribution to Tumorigenesis......................7 I-5. COX-2 Inhibitors and Cancer..................................................................... 9 I-6. COX-2 mRNA Expression and Real-time R T -P C R ...........................10 CHAPTER I I ...................................................................................................................... 12 M ETHODS: REAL-TIME QUANTITATIVE R T -P C R .......................................12 iv Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 11-1. Principle of Real-time P C R ......................................................................12 II-2. Instrumentation and Detection Chemistry...........................................16 II-3. Quantification............................................................................................... 17 II-4. Designing Primers and Probes..............................................................21 CHAPTER I I I .....................................................................................................................23 CYCLO O XYG ENASE-2 (COX-2) GENE EXPRESSION IS UP- REGULATED DURING THE PROGRESSION OF ESOPHAGEAL C A R C IN O M A ................................................................................................................23 III-1. Introduction................................................................................................. 23 III-2. Materials and M ethods........................................................................... 27 Patients and Sam ples...................................................................................27 mRNA Isolation and cDNA Synthesis...................................................... 29 Real-Time Polymerase Chain Reaction (PCR) Quantification 30 Statistical Analysis..........................................................................................31 HI-3. Results..........................................................................................................34 III-4. Discussion.................................................................................................. 47 CHAPTER IV .....................................................................................................................53 v Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CYCLO O XYG ENASE-2 (COX-2) GENE EXPRESSION IS ASSO CIATED W ITH SURVIVAL IN CURATIVELY RESECTED NON-SMALL CELL LUNG C A NC ER ........................................................................................................... 53 IV-1. Introduction................................................................................................ 53 IV-2. Materials and Methods........................................................................... 55 Patients and Specimens...............................................................................55 mRNA Isolation and cDNA Synthesis.......................................................56 Real-Time Polymerase Chain Reaction (PCR) Quantification 57 Statistical Analysis..........................................................................................57 IV-3. Results.........................................................................................................59 IV-4. Discussion.................................................................................................. 64 CHAPTER V ......................................................................................................................69 A M OLECULAR-EPIDEM IOLOGICAL ANALYSIS OF NSAIDS, COX-1 AND COX-2 GENE EXPRESSION, AND RISK OF COLORECTAL A D E N O M A S .................................................................................................................69 V-1. Introduction................................................................................................. 69 V-2. Materials and Methods.............................................................................72 Description of Parent Study.........................................................................72 Microdissection of Tissues.......................................................................... 74 vi Reproduced with permission of the copyright owner. Further reproduction prohibited w ithout permission. mRNA Isolation and cDNA Synthesis...................................................... 76 Real-Time Polymerase Chain Reaction (PCR) Quantification 76 Statistical Analysis..........................................................................................76 V-3. Results..........................................................................................................79 V -4. Discussion....................................................................................................86 CHAPTER V I .....................................................................................................................89 R E F E R E N C E S ............................................................................................................ 89 vii Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES Table 3.1. TaqM an® PCR primers and probes for COX-2 and (3-actin, used in the esophageal adenocarcinoma study................................. 33 Table 3.2. COX-2 mRNA expression in tissues from patients with esophageal adenocarcinoma (EA group), Barrett’s esophagus (BE group), and patients without a history of adenocarcinoma, Barrett’s esophagus, or gastroesophageal reflux disease (Cont group).......................................................................35 Table 4.1. Associations between clinicopathological variables and COX- 2 expression status..................................................................................... 61 Table 4.2. Survival in NSCLC based on clinical and molecular parameters..................................................................................................... 62 Table 4.3. Cox’s proportional hazard regression models................................... 65 Table 5.1. TaqMan® PCR primers and probes used for the colorectal adenoma study..............................................................................................78 Table 5.2. Age, gender, polyp size, polyp location and number of polyps from participants with colorectal adenom as........................................ 82 Table 5.3. COX-1, COX-2 and the COX-2/COX-1 ratio by NSAID use..........83 Table 5.4. Odds ratios of patients with polyps, compared to controls who had no history of polyps and were free of polyps at time of sigmoidoscopy...............................................................................................85 viii Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF FIGURES Figure 1.1. Biosynthesis of prostaglandins.................................................................3 Figure 1.2. Regulatory elements in human COX-2 promoter................................4 Figure 2.1. Amplification plots generated by the ABI PRISM 7700 Sequence Detection System [TaqMan] from Applied Biosystems. (A) Amplification plot: the change in normalized reporter signal (ARn) versus cycle number. (B) Amplification plot: the same samples shown in (A). A logarithmic plot of ARn versus cycle number.......................................................................... 14 Figure 2.2. Standard curve plotting log starting copy number versus Ct (threshold cycle). The diamonds (♦ ) represent standard sample data and the round dots represent unknown gene data................................................................................................................... 15 Figure 2.3. Schematic diagram for TaqMan system ............................................. 18 Figure 3.1. Multistage progression of Barrett’s esophagus to esophageal adenocarcinom a.......................................................................................... 24 Figure 3.2. Schematic diagram for the esophageal adenocarcinoma study, using fresh tissues...........................................................................32 Figure 3.3a. Box and whisker plots of relative COX-2 mRNA expression levels in normal and intermediate tissues from patients with Barrett’s esophagus.....................................................................................36 Figure 3.3b. Paired data showing relative COX-2 mRNA expression levels from intermediate and matching normal esophageal tissues from patients with Barrett’s esophagus. Mean values are shown as diamonds ( ♦ ) ...................................................................... 37 Figure 3.4a. Box and whisker plots of relative COX-2 mRNA expression levels from normal(N), intermediate(l), and adenocarcinoma(A) tissues from patients with esophageal adenocarcinoma........................................................................................... 39 ix Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3.4b. Matched data showing relative COX-2 mRNA expression levels from normal, intermediate, and adenocarcinoma tissues from patients with esophageal adenocarcinoma. Mean values are shown as diamonds ( ♦ ) ............................................40 Figure 3.5a. Box and whisker plots of relative COX-2 mRNA expression levels from normal squamous epithelial tissues from 1) a control group, 2) patients with Barrett’s esophagus, and 3) patients with esophageal adenocarcinoma..........................................41 Figure 3.5b. Scatter plot of relative COX-2 mRNA expression levels from normal squamous epithelial cells taken from 1) a control group, 2) patients with Barrett’s esophagus, and 3) patients with esophageal adenocarcinoma. Mean values are shown as a horizontal bar....................................................................................... 42 Figure 3.6. Box and whisker plots of relative COX-2 mRNA expression levels for 1) dysplastic(D) tissues from patients with esophageal adenocarcinoma, and 2) metaplastic(M) tissues from patients with Barrett’s esophagus................................................. 44 Figure 3.7. Box and whisker plots of relative COX-2 mRNA expression levels for 1) adenocarcinoma(A) tissues from patients with esophageal adenocarcinoma, and 2) metaplastic(M) tissues from patients with Barrett’s esophagus................................................. 45 Figure 3.8. Box and whisker plots of relative COX-2 mRNA expression levels for dysplastic (D) and adenocarcinoma(A) tissues from patients with esophageal adenocarcinoma..........................................46 Figure 4.1. Schematic diagram for the lung cancer study using fresh tissues..............................................................................................................58 Figure 4.2. COX-2 mRNA expression (the ratio between CO X-2 and the internal reference gene p-actin) in specimens of primary non small cell lung cancer. High COX-2 mRNA levels contained ratios g re ate r than 0.6; low levels contained ratios less than or equal to 0.6................................................................................................63 Figure 5.1. Multi-step progression of colon cancer................................................ 70 x Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 5.2. Stained FFPE adenoma specimens before and after L C M 75 Figure 5.3. Schematic diagram for the colorectal adenoma study, using paraffin-embedded tissues........................................................................77 xi Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ABSTRACT Epidemiological studies have shown that prolonged use of non steroidal anti-inflammatory drugs (NSAIDs) including aspirin reduces the risk of esophageal, gastric, lung, and colorectal cancer, and other studies have suggested that NSAIDs can reduce tumor formation, progression, and metastasis, and inhibit angiogenesis that is essential for tumorigenesis. The key target of NSAIDs is cyclooxygenase (COX). COX is an enzyme that converts arachidonic acid (AA) to prostaglandins (PGs). It has two isoforms: COX-1 and COX-2. COX-1 is constitutively expressed in most cell types, but COX-2 expression is inducible by various stimuli. Recently, up-regulated COX-2 expression has been reported in several human carcinomas, increasing the attention to a potential role for COX-2 in tumorigenesis. Furthermore, COX-2 has been demonstrated to inhibit apoptosis, induce angiogenesis, and modulate the immunological environment through regulation of the cytokine balance. All of these factors, when combined, strongly implicate COX-2 involvement in tumor development. The inducible nature of the COX-2 gene and the up-regulation of COX-2 in tumors strongly suggests that it is the precise amount of CO X-2 in tissues that may be an important determinant of tumor biology, influencing such facto rs as tu m o r aggressiveness, m etastatic potential, drug response and thus ultimately, patients’ prognosis. Furthermore, the amount of COX-2 up-regulation may also be within defined limits at various stages of xii Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. tumorigenesis, thus providing a means of molecular pathology to aid in identifying those stages. While in the past quantitative measurements of molecular factors in patients’ tissues were very difficult, recent advances in quantitative PCR technology (e.g., the Taqman®) have made it feasible to perform rapid, real-time monitoring of PCR reactions with a high-throughput capacity. In this project, through the measurement of COX-2 gene expression levels, it was demonstrated that 1) up-regulated CO X-2 gene expression might be an early event in the development of esophageal adenocarcinoma, 2) COX-2 gene expression can be a good prognostic factor for lung cancer, and 3) NSAIDs have a preventive effect in colorectal adenocarcinoma by reducing COX-2 gene expression. XIII Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER I INTRODUCTION The importance of Cyclooxygenase (COX) -2 as a promising pharmacological target for the prevention and treatment of many human cancers is increasing. COX, known as prostaglandin synthase, catalyzes sequential synthesis of prostaglandin (PG) G 2 and PGH2 from arachidonic acid (AA) by virtue of intrinsic cyclooxygenase and peroxidase activities (Figure 1.1). PGH2 is then converted to other eicosanoids, including PGs, thromboxane (Tx), and prostacyclin by specific isomerases. Cyclooxygenase-derived PGs mediate many important physiological processes, including hemostasis, platelet aggregation, kidney and gastric function, reproduction, immunity, etc. (Vane and Botting 1997; Vane, Bakhle etal. 1998). 1-1. CO X Isoforms There are two isoforms of COX: COX-1 and COX-2. Although they have similar enzymatic activities and share 60% identity at the amino acid level, the COX-1 and COX-2 isoforms are encoded by two unique genes located on different chromosomes (Smith, Garavito et al. 1996), have distinct biological properties, and different expression patterns (Hla and Neilson 1 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1992; Feng, Sun et al. 1993; Kennedy, Chan et al. 1993; Smith, Meade et al. 1994; Williams and DuBois 1996). COX-1 is constitutively expressed in most mammalian tissues and is thought to carry out “housekeeping” functions such as cytoprotection of the gastric mucosa, regulation of renal blood flow, and control of platelet aggregation. In contrast, the CO X-2 protein and mRNA are normally undetectable in most tissues, but can be rapidly induced by inflammation, ovulation, and other stimuli, such as growth factors, cytokines, and mitogens (Hamasaki, Kitzler et al. 1993; Jones, Carlton et al. 1993; DuBois, Tsujii et al. 1994; Hempel, Monick et al. 1994; Prescott and White 1996). Increased expression of COX-2 has been reported in multiple inflammatory diseases, including rheumatoid arthritis, Crohn’s disease, ulcerative colitis and Helicobacter pylori infectious gastritis (Eberhart, Coffey et al. 1994; Kargman, O'Neill et al. 1995; Sano, Kawahito et al. 1995; Kang, Freire-Moar et al. 1996; Sawaoka, Kawano et al. 1998; Singer, Kawka et al. 1998). I-2. Regulation of COX-2 Expression The promoter of COX-2 (Figure 1.2) contains a TAJA box and multiple transcription factor binding sites such as nuclear factor-icB (NF-kB), the n uclea r fa cto r fo r interleukin 6 (NF-IL6; C/EBPp) and the cyclic AM P response element (CRE) binding protein (Yamamoto, Arakawa et al. 1995). The sites proximal to the transcription start site have been reported to be 2 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 1.1. Biosynthesis of prostaglandins Arachidonic Acid COX-1 /COX-2 I Cyclooxygenase activity Prostaglandin G2 COX-1 /COX-2 I Peroxidase activity Prostaglandin H2 MDA PGD TxA PGI Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Figure 1.2. Regulatory elements in human COX-2 promoter - 270/-265 5 ’ -4 4 5 M 2 7 | -223/-214 NF-kB Sp1 NF-k B Start 132/-124 - 55/-50 - 58/-53 | 31/-25 NF/IL6 ATF/CRE E-box TATA Box differentially responsive to various stimuli, including tumor necrosis factor-a (TNF-a), cytokines (IL-1(3, IL-10, etc.), v-src oncogene, lipopolysaccharide (LPS), and platelet-derived growth factor (PDGF). Several studies have shown that the expression of COX-2 is regulated at the transcriptional and post-transcriptional levels (Yamamoto, Arakawa et al. 1995; Madrid, W ang et al. 2000; Shao, Sheng et al. 2000; Sheng, Shao et al. 2001; Monick, Robeff et al. 2002). Several signaling pathways have been reported in COX-2 transcriptional regulation. V-src (Xie and Herschman 1995) and PDGF (Xie and Herschman 1996) act by way of a Ras/MEKK-1/JNK signal transduction pathway to activate c-Jun, and by way of a Ras/Raf-1/ERK pathway to increase AP-1 activity. AP-1 mediates induction of c-Fos and co- activation of c-Jun. This, in turn, stimulates CRE and induces expression of COX-2. The NF-k B and NF-IL6 sites are required for the induction of COX-2 expression by TN F-a through the PI-3K/Akt signal transduction pathway (Yamamoto, Arakawa et al. 1995; Ozes, Mayo et al. 1999). COX-2 is also regulated at the post-transcriptional level. The stability of COX-2 m RNA is positively regulated by Ras (Sheng, Shao et al. 2000) and p38 (Matsuura, Sakaue et al. 1999). Additionally, the COX-2 mRNA 3’ untranslated region (3’-UTR) contains an AU-rich sequence element (ARE) which negatively regulates the stability of COX-2 mRNA (Dixon, Kaplan et al. 2000). 5 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. During tumorigenesis, COX-2 expression can increase as a consequence of two types of effects: 1) promoter activation, and 2) enhanced mRNA stability. The COX-2 promoter may be activated by growth factors, oncogenes, and other stimuli. Or, COX-2 mRNA may be stabilized by Ras and/or the p38 signal transduction pathway, as well as by the mutation of ARE-binding proteins which normally negatively regulate transcript stability. I-3. Evidence of COX-2 Involvement in Tumorigenesis Recent studies have suggested that COX-2 plays a role in human tumorigenesis. For example, overexpression of COX-2, but not COX-1, has been reported in colon cancer (Eberhart, Coffey et al. 1994; Kargman, O'Neill et al. 1995; Sano, Kawahito et al. 1995), breast cancer (Hwang, Scollard et al. 1998), gastric cancer (Ristimaki, Honkanen et al. 1997), esophageal cancer (Wilson, Fu et al. 1998; Zimmermann, Sarbia et al. 1999), lung cancer (Wolff, Saukkonen et al. 1998; Hida, Yatabe et al. 1998b), liver cancer (Koga, Sakisaka et al. 1999), and pancreatic cancer (Okami, Yamamoto et al. 1999; Tucker, Dannenberg et al. 1999). Williams et al. (2000) found strong evidence in support of a cause-effect relationship between overexpression of COX-2 and tumorigenesis. Tumor growth was markedly attenuated in COX- 2'7 ' mice, but not in either COX-1'7 ' or wild-type mice. Tumor growth was also markedly attenuated by a selective COX-2 inhibitor (Williams, Tsujii et al. 6 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2000). In addition, Liu et al. (2001) reported that COX-2 expression was sufficient enough to induce tumorigenesis in transgenic mice by a genetic gain-of-function approach. Multiparous female transgenic mice with overexpression of COX-2 from mouse mammary tumor virus (MM TV), but not virgin female non-transgenic mice, exhibited a high incidence of focal mammary gland hyperplasia, dysplasia, and transformation into metastatic tumors (Liu, Chang et al. 2001). These studies provide strong evidence in support of a role for COX-2 in tumor development. I-4. Mechanisms of COX-2 Contribution to Tumorigenesis COX-2, a prostaglandin synthase, could contribute to tumorigenesis through multiple mechanisms. One possible mechanism is through COX-2 overexpression-induced PG synthesis. High PG levels have been reported in colorectal carcinoma (Huang, Stalina et al. 1998; Cianchi, Cortesini et al. 2001). Additionally, several studies demonstrated that increased PG levels play a role in carcinogenesis. Sheng and colleagues showed that induced PGE 2 enhances the proliferation and motility of colorectal carcinoma cells (Sheng, Shao et al. 2001). Huang et al. (1998) also demonstrated that lung tumor-derived PGE2 plays an important role in promoting macrophage IL-10 induction and overproduction at the tum or site, w h ile sim u lta n e o u sly inhibiting macrophage IL-12 production. IL-10 is involved in tumor-mediated immunosuppression (Kim, Modlin et al. 1995; Qin, Noffz et al. 1997), while 7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. IL-12 plays a crucial role in tumor immunity (Bianchi, Grohmann et al. 1996; Colombo, Vagliani et al. 1996). Therefore, increased PG levels modulated macrophage-derived cytokines in the tumor environment, and contributed to tumorigenesis (Huang, Stolina et al. 1998; Stolina, Sharma et al. 2000). COX-2 mediated PGs also promote tumor growth by inducing neoangiogenesis (Masferrer, Leahy et al. 2000) and suppressing apoptosis through up-regulation of Bcl-2 (Sheng, Shao et al. 1998). Recently, COX-2 mediated tumor angiogenesis has been reported (Tsujii, Kawano et al. 1998; Masferrer, Leahy et al. 2000; Williams, Tsujii et al. 2000; Cianchi, Cortesini et al. 2001). Angiogenesis promotes tumor growth by supplying oxygen and nutrients. Tsujii and colleagues explored the role of COX-2 in angiogenesis using the coculture model system of endothelial cells and colon carcinoma cells. They found that COX-2 modulates production of the angiogenic factors vascular endothelial growth factor (VEGF) and basic fibroblast growth factor (bFGF) in colon cancer cells. They also showed that selective COX-2 inhibitors reduce endothelial tubule formation (Tsujii, Kawano et al. 1998), preventing angiogenesis. Williams et al. (2000) demonstrated that vascular density is reduced by 30% in tumors of CO X-2'7 ' mice compared with tumors in wild-type mice, and that VEG F production is reduced by 94% in fibroblasts from COX-2'7 ' mice compared with wild-type mice (Williams, Tsujii et al. 2000). These findings demonstrate 8 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. that COX-2 contributes to tumor vascularization by inducing angiogenic factors. Other studies have demonstrated that up-regulated COX-2 promotes an anti-apoptotic pathway. Inhibition of apoptosis is essential for tumor development by preventing death of abnormal cells. COX-2 induced tumor tissue to express elevated levels of the anti-apoptotic protein Bcl-2 and reduced levels of the apoptotic proteins Bax and B cI-x l, indicating COX-2 contributes to tumorigenesis by resisting apoptosis (Liu, Chang et al. 2001). Additional studies demonstrate that overexpressed COX-2 in rat intestinal epithelial cells increased Bcl-2 protein expression and resisted butyrate- induced apoptosis. Inhibition was reversed, however, by application of a selective COX-2 inhibitor (Tsujii and DuBois 1995). I-5. COX-2 Inhibitors and Cancer Several pharmacological studies support COX-2 as a promising therapeutic target for prevention and treatment of many human cancers, given the ability of COX-2 inhibitors to reverse the effects of up-regulated COX-2 expression; evidence includes the following: Compared with normal controls, regular users of NSAIDs had 40-50% lower mortality rates from colorectal and esophageal cancers (Greenberg, Baron et al. 1993; Thun, Namboodiri et al. 1993). 9 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Sulindac, a selective COX-2 inhibitor, significantly reduced adenoma size and number in patients with familial adenomatous polyposis (FAP) (Giardiello, Hamilton et al. 1993). Selective COX-2 inhibitors suppress tumor cell growth (Sheng, Shao et al. 1997). I-6. CO X-2 mRNA Expression and Real-time RT-PCR Many studies have investigated the translational level of CO X-2 gene expression in various tumor types, but the transcriptional level of COX-2 gene expression has not been well studied. In this study, real-time quantitative polymerase chain reaction (PCR) involving a fluorescence-based detection method (ABI PRISM 7700 Sequence Detection System [TaqMan], Applied Biosystems, Foster City, CA) was used to quantitate CO X-2 mRNA expression. Other methods commonly used to study gene expression include Northern blot, in-situ hybridization, RNase protection, nuclear run-on transcription, cDNA array, and conventional reverse transcription polymerase chain reaction (Gibson, Heid et al. 1996). Unlike these conventional methods, real-time quantitative RT-PCR is more sensitive and accurate of the quantification methods, and allows a high throughput, is capable of utilizing very small sample volumes and can detect very low levels of gene expression, that are not possible with any of the conventional methods. 10 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Moreover, real-time PCR does not require intensive, laborious radioactive labeling, use of polyacrylamide gels, or detection by phosphorimaging, thereby decreasing risks associated with PCR-product carryover contamination and usage of hazardous chemicals. Utilizing real-time RT-PCR, this study provides a better understanding of the mechanism of action of COX-2, and of the potential prognostic role of COX-2 mRNA expression in various tumor types and pre-cancerous diseases including lung cancer, esophageal cancer and adenomatous polyps. 11 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER II METHODS: REAL-TIME QUANTITATIVE RT-PCR 11-1. Principle of Real-time PCR Real-time PCR, a new technique for quantifying PCR products, was recently developed. This technique utilizes Taq polymerase 5 ’ to 3’ exonuclease activity with dual-labeled oligonucleotide probes. The probes have a reporter dye (e.g. FAM [6-carboxyfluorescein]) at one end, and a quencher dye (e.g. TAMRA [6-carboxytetramethylrhodamine]) at the other end. W hen the probe is intact, the fluorescent emission of the reporter dye is absorbed by the quencher dye through fluorescent energy transfer. During the extension phase of the PCR cycle, Taq polymerase cleaves an internally labeled non-extendable probe, yielding the reporter and quencher complex. Exonuclease degradation of the probe frees the reporter (FAM) from the quencher (TAMRA), resulting in an increase in reporter fluorescence emission. During each cycle, the process repeats unhindered by the exponential accumulation of PCR product. Reactions are characterized by the point when amplification of PCR product is first detected, rather than by the amount of PCR product that accumulates after a fixed number of cycles (end-point conventional PCR). A sequence detector measures real time 12 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. fluorescent emission specific for PCR target amplification. The detector then transfers emission data directly to computer software in a Macintosh computer. The computer program calculates fluorescent emission (ARn) by subtracting PCR product emission from baseline emission. The fluorescent emission thus represents the amount of probe degraded during PCR amplification. The computer software uses the data to construct amplification plots (Figure 2.1), graphing ARn values versus cycle number. In analysis, the baseline is first determined from data obtained in cycle 3 to n-2, where n is the first cycle in which PCR product is detected (Figure 2.1). For example, if PCR product is first detected in cycle 17, the baseline will be based on emissions from cycles 3 to 15. Then an arbitrary threshold is selected, based on the variability of the baseline. The threshold may be manually manipulated for each individual experiment as necessary. After the threshold is determined, the threshold cycle (Ct) of each sample is determined automatically by the computer program. The threshold cycle (Ct) is the fractional cycle number at which the fluorescence emitted by the probe cleavage product passes over the determined threshold. The target gene is quantified based on Ct, using a standard curve (Figure 2.2) to determine the starting copy number. Quantity is also calculated (based on Ct) after the result is exported to a Microsoft Excel spreadsheet. The reported quantities can be used as a quantitative measurement of the input target (Giulietti, Overbergh et al. 2001). 13 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 2.1. Amplification plots generated by the ABI PRISM 7700 Sequence Detection System [TaqMan] from Applied Biosystems. (A) Amplification plot: the change in normalized reporter signal (ARn) versus cycle number. (B) Amplification plot: the same samples shown in (A). A logarithmic plot of ARn versus cycle number. A A m p lific a tio n Exponential Phase \ > ^Threshold ^ Plateau effect 23 30 9 3 4 0 C y c le -T h r e s h o ld C y c le C a lc u la tio n — — T h re s h o ld — — ■ ■ ■ Use Threshold: j ,0Q6 I [ Suggest j ]*c Stddev start Calculations: C t S td Dev F A M - H2 19.138 0.001 F A M - H3 21.579 0.001 F A M - H4 17.953 0.001 m F A M - HS 20.252 0.001 j * F A M - H6 22.905 0.001 ▼ Baseline - S a m p le s □ FAM - H4 □ FAM - H5 □ FAM - H6 □ FAM - H7 □ FAM - H0 □ FAM - H9 □ FAM - H10 □ F A M -H 1 1 □ FAM - H12 □ □ Viewer: | ARn (B.. R e p o rte r: I FAM ‘ I P lease S e t th e T h resh ald V a lu e an A ll R e p o rte r L a y e rs . Click OK tn C ontinue. A Rn = A Rn+ - A Rn' A R n +: The fluorescence emission of the product at each time point A Rn-: The fluorescence emission baseline B A m p lific a tio n •Samples 0 FAM - H I 0 FAM - H2 0 FAM - H3 4 .0 0 0 - 3 .0 0 0 - e ■ 4 A R n 2.000 - Threshold 1.000 - 0 .0 0 0 Baseline - 1.000 - 0 2 4 6 8 10 18 20 24 20 30 34 38 40 44 48 SO 14 C y c le — T h re s h o ld C y c le C a lc u la tio n T h re s h o ld • .001 Use Thresho Id: I .19 I f5uggest | M ult. * S tddev: f 10.0 I * [ Omit Threshold: f 2.0 r— Baseline — — S ta r t:) 5 J ] S top:[ 1 Update Calculations j C t S td D ev FA M - H I 21.526 0.001 = FA M - H2 23.981 0.001 FA M - H3 26.259 0.001 A ▼ 14 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 2.2. Standard curve plotting log starting copy number versus Ct (threshold cycle). The diamonds (♦) represent standard sample data and the round dots represent unknown gene data. Standard Curre 40.00 Unknowns 50.00 Standards 25.00 S lo p e : Y - Intercept: C o r r e la tio n Coetff: 15.00 1 0 .0 0 1 0 * 2 1 0 * 3 10* 5 - 3.398 1 .000 Starting Quantity 15 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Only when the target sequence is complementary to the specially designed, dual-labeled probe and matching primer set, is the increase in fluorescence signal detected and amplified during PCR. Because of the high specificity of this method, nonspecific amplification is not detected. In addition, since real time PCR collects data during amplification, unlike end-point conventional PCR, it is more sensitive than other conventional PCR methods. 11-2. Instrumentation and Detection Chemistry The instrument used for real-time quantitative RT-PCR is the ABI PRISM 7700 Sequence Detection System (SDS) from Applied Biosystems (Foster City, CA), designated as TaqMan (Gibson, Heid et al. 1996; Heid, Stevens et al. 1996). It has a built-in thermal cycler and a laser that is pointed at each of the 96 sample-wells via fiber optical cables. The laser induces fluorescent emission from the probe products. A charge-coupled- device (CCD) camera in the sequence detector collects the fluorescent emission from each sample and automatically analyzes the data. The reaction conditions are programmed in a Macintosh linked to the sequence detector. PCR amplifications are performed in a total volume of 25pl, containing 600 nM of each primer, 200 nM probe, 2.5 U AmpliTaq Gold Polymerase, 200 pM each of dATP, dCTP, dGTP, 400 pM dUTP, 5.5 mM MgCI2, and 1 * TaqMan Buffer A with reference dye (all reagents from Applied Biosystems). PCR amplifications are always performed in duplicate 16 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. or triplicate wells. Cycling conditions are usually 50°C for 10 seconds and 95°C for 10 minutes, followed by 46 cycles at 95°C for 15 seconds and 60°C for 1 minute. The Sequence Detection Software (SDS) program came along with the 7700 system and calculates the Ct number for each sample. One run has a capacity of 96 wells and takes approximately 2 hours to complete 46 cycles. Possible genomic DNA background is excluded during calculation by amplifying matching non-reverse transcribed materials— the no-RT controls (Bieche, Olivi et al. 1998). Three oligonucleotides are used: a forward primer, a probe, and a reverse primer (Figure 2.3). All of them are specific for the target and are able to bind to it. The TaqMan assay uses a probe technology that utilizes the 5’ to 3’ exonuclease activity of Taq polymerase that is very essential in this assay. II-3. Quantification Initially, calibrators need to be established to normalize the results from a real-time quantitative RT-PCR with the actual samples. Commercially available purified RNAs (e.g. total RNA liver (#735017), total RNA lung (#735019), total RNA colon (#735263), Stratagene, La Jolla, CA) were reverse transcribed to cDNAs that were used as calibrators. Three point serial dilutions (5-fold difference/point) of the calibrator cDNAs were prepared 17 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 2.3. Schematic diagram for TaqMan system TaqMan Schematic Polymerization Forward Primer 5’ 3’ 5’ PROBE Reverse Primer Strand Displacement P Is T 5’ 3’ 5’ Fluorescence Cleavage 3’ 5 0 1 Polymerization s s Completed 5’ ■ 3’ ^ —■ 5’ ■ ' i Applied Biosystems instrument manual. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. with a primer and probe set specific for the gene of interest, as well as with the primer and probe set specific for the reference gene (an endogenous housekeeping gene, (3-actin). A portion of the non-reverse transcribed RNA solution was usually used as a no-RT control. Since these RNAs are 100% commercially purified, the DNA amplification of each calibrator was thought to be zero. In establishing calibrators, the serial dilutions were usually run in triplicate. Prepared samples were loaded in an optical 96-well plate (Applied biosystems, Foster City, CA), placed into the thermal cycler of TaqMan, and real-time quantitative RT-PCR was performed. Upon completion, the SDS program calculates ARn using the equation ARn = ARn+ - ARn' (ARn+: the fluorescence emission of the product at each time point; ARn': the fluorescence emission baseline). This value represents probe degradation during PCR. The computer software constructs amplification plots (Figure 2.1) using fluorescent emission data (ARn) collected during PCR amplification. The ARn values are plotted against the cycle number in which they occurred. The baseline, Ct, interest gene quantity, and starting copy number were determined as described in the Principle of Real-time PCR section of this chapter. Next, the ratio of interest gene (e.g. COX-2) quantity to reference gene (e.g. p-action) quantity was calculated based on the following equation: 19 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A Q (interest gene cDNA) = Q (interest gene cDNA) - Q (interest gene no-RT control) A Q ((3-actin cDNA) = Q (p-actin cDNA) - Q (f3-actin no-RT control) A Q (interest gene cDNA) Ratio (Relative gene expression) = --------------------------------------------------------- A Q (P-actin cDNA) After calculation, each dilution’s triplicate data points were averaged and compared with the same calibrator samples to normalize the calibrator values. Usually the variation between these points was less than 5%. After the initial normalization values were established, PCR reactions with the actual samples were finally performed for the interest gene, reference gene, no-RT controls, and three different calibrators. Each cDNA sample was run in triplicate to test reproducibility, and calibrators were run at three dilutions each. Prepared samples were loaded in an optical 96-well plate (Applied Biosystems, Foster City, CA). After completion of PCR, the calculations were performed as described above, and compared with the normalization values obtained earlier, to calculate the optimal multiplier. This number was then multiplied by the sample values obtained during real-time PCR to generate the relative gene expression levels used in our study. 20 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 11-4. Designing Primers and Probes The sequences for the gene of interest (e.g. COX-2) were obtained from NCBI (http://www.ncbi.nlm.nih.gov/Genebank). Then, using a BLAST search query (http://www.ncbi.nlm.nih.gov/BLAST), regions that may be homologous to other genes and/or pseudogene regions were also sought out. If found, those regions were excluded to reduce the potential for amplification of non-specific products. The searched sequences were exported to the computer program, Primer Express, which identified many appropriate primer and probe sets. To select the most desirable primer and probe set, some restrictions were applied. The melting temperature (Tm) of the primers and probes, and the amplicon length were very important. The computer program gave the following default parameters: the range of Tm for the primers was 58-60 °C, Tm of the probe was at least 10°C higher (68-70°C) to allow annealing to the target sequence during PCR extension (PCR extension was performed at 60°C). The probe did not contain guanidine (G) at the 5 ’ end, because that base slightly quenches the reporter signal, even after probe cleavage. Additionally, the probe contained more cytosine (C) than G throughout its sequence, and if this were not the case, the anti-sense probe could have been used instead. The selected amplicon lengths were as short as possible (between 50-150bp; but for the cDNA from paraffin-embedded samples, less than 100bp). Lastly, the primers did not contain more than 3 G ’s or C’s in the 21 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. five nucleotides at the 3’end. Using these restrictions, the most desirable primer and probe sets were identified. 22 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER III CYCLOOXYGENASE-2 (COX-2) GENE EXPRESSION IS UP- REGULATED DURING THE PROGRESSION OF ESOPHAGEAL CARCINOMA 111-1. Introduction Recent studies reported that Barrett’s esophagus occurs in approximately 400 out of 100,000 Americans (Cameron 1994), and that both Barrett’s esophagus and Barrett’s esophagus associated adenocarcinoma incidences are increasing at a rapid rate in western countries. This is especially true for American men since the 1970s. In the 1970s, adenocarcinoma represented just 16% of all esophageal cancer in white men in the United States. By the mid-1980s, it increased to 33%, and by 1990 it approached 50% (Pera, Cameron et al. 1993; Blot and McLaughlin 1999). Esophageal adenocarcinoma has become one of the most malignant human diseases worldwide. The expected 5-year survival rate for this malignancy is poor because patients are only rarely diagnosed at an early, more potentially curable stage (Greenlee, Murray et al. 2000). Barrett’s esophagus is a condition where the normal squamous esophageal epithelium is replaced 23 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3.1. Multistage progression of Barrett’s esophagus to esophageal adenocarcinom a Normal esophagus with white squamous epithelium Barrett’s esophagus with salmon pink epithelium (metaplasia composed of columnar epithelium with goblet cells) Barrett’s with low grade dysplasia Barrett’s with high grade dysplasia Esophageal adenocarcinoma 24 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. with a metaplastic columnar epithelium due to chronic gastroesophageal reflux disease (GERD). It is a pre-malignant condition, with a 30- to 125-fold increased risk of developing esophageal cancer, compared to the general population. The development of esophageal adenocarcinoma is thought to arise from Barrett’s esophagus through a multistage progression (Figure 3.1) in which Barrett’s metaplasia subsequently progresses to low-grade dysplasia (LGD), high-grade dysplasia (HGD), and eventually to esophageal adenocarcinoma (Zhuang, Vortmeyer et al. 1996; Barrett, Sanchez et al. 1999). Since a small proportion of patients with Barrett’s esophagus develop adenocarcinoma, it may not be cost-effective to do routine endoscopic surveillance of all patients with the disease. In addition to expenses associated with Barrett’s esophagus, consistency in assigning a diagnosis and a stage is hampered by the subjective limitations of medicine. Since pathologists all have their own way of reading slides, and staining is often inconsistent, reader biases are unavoidable (Smith, Maxwell-Armstrong et al. 1999; Dixon 2000; Skacel, Petras et al. 2000). Molecular approaches are urgently needed to investigate diagnostic markers and allow accurate identification of individuals at greater risk of progression to adenocarcinoma. Recently, the importance of COX-2 in gastrointestinal carcinogenesis was discovered, and overexpression of COX-2 has been reported in 25 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. esophageal adenocarcinoma (Wilson, Fu et al. 1998; Zimmermann, Sarbia et al. 1999). Several epidemiological studies have shown that intake of NSAIDs, well-known COX inhibitors, is associated with reduced risk of colorectal cancer (Giardiello, Hamilton et al. 1993; Greenberg, Baron et al. 1993; Giovannucci, Egan et al. 1995; Thun 1996). Other studies have shown NSAIDs also decreased risk of developing esophageal cancer by up to 90% (Thun, Namboodiri et al. 1993; Funkhouser and Sharp 1995). Recently, COX-2 involvement in the progression of esophageal adenocarcinoma has been studied. However, technical insensitivities of semi-quantitative conventional PCR methods, Western blot analysis, and immunohistochemical analysis have resulted in a failure to detect COX-2 mRNA expression in normal esophageal tissues and CO X-2 protein expression in metaplastic Barrett’s esophageal tissues (Wilson, Fu et al. 1998). They have also been unable to produce consistent results. Shirvani et al. (2000), and Morris et al. (2001) demonstrated COX-2 involvement in the progression from Barrett’s metaplasia to Barrett’s associated esophageal adenocarcinoma (Shirvani, Ouatu-Lascar et al. 2000; Morris, Armstrong et al. 2001). However, due to limitations in the designs of both studies, the results do not reflect progression through different stages in the adenocarcinoma sequence. Instead, they represent snapshots at a particular stage of progression in individual patients. All of the samples from different stages 26 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. came from different patients. Consequently, the role of CO X-2 in the development of adenocarcinoma and, the progression of disease has remained unclear. In this study we collected samples taken from individuals with different stages of disease progression. This enabled us to study the pattern of COX- 2 gene expression in progressive disease, from normal to Barrett’s disease to esophageal adenocarcinoma. W e measured COX-2 m RNA expression using a very sensitive method, real-time quantitative RT-PCR, and assessed the prevalence and timing of COX-2 activation in the adenocarcinoma sequence. Our findings invite further exploration of the role COX-2 plays in this disease, and add information that may translate into the use of CO X-2 in the management of patients with this disease. 111-2. Materials and Methods Patients and Samples One hundred and eight tissue samples were endoscopically and surgically excised from 19 (39%) patients with Barrett's esophagus without esophageal adenocarcinoma (BE group), from 20 (41% ) patients with Barrett’s associated esophageal adenocarcinoma (EA group), and from 10 (20% ) patients with no symptomatic, endoscopic, or histopathological evidence of Barrett's esophagus or chronic gastroesophageal reflux disease (control group). After excision, they were immediately frozen in liquid 27 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. nitrogen. From the total of 49 patients, 33 (67%) were male and 16 (33% ) were female, with a median age of 61.8 years (range 24 to 77 years). Endoscopic biopsies were obtained according to a protocol that required biopsies to be taken at 2 cm intervals from each quadrant (anterior, posterior, right, and left lateral positions) spanning the visible length of Barrett's mucosa, and an additional biopsy from the normal-appearing squamous mucosa of the esophagus. Biopsy specimens of the normal esophagus were taken at least 4 cm proximal to the macroscopically abnormal epithelium. Part of the specimen or an adjacent specimen was fixed in formalin and paraffin for histopathological examination. Specimens were classified as metaplastic if only metaplasia was present. Specimens were classified as dysplastic if either low-grade dysplasia (LGD) or high-grade dysplasia (HGD) was present. Dysplastic tissues were not divided into LGD or HGD groups because areas of LGD and HGD were commonly present in the same specimen. Using these criteria, tissues were placed into one of three categories and analyzed for COX-2 mRNA expression. The first category, from patients in the BE group, consisted of tissues with metaplasia (n = 16), dysplasia (n = 3), and matching normal squamous tissue (n = 19). The second category, from patients in the EA group, consisted of tissues with adenocarcinoma (n = 20), metaplasia (n = 5), dysplasia (n = 15), and matching normal squamous tissues (n = 20). The third category consisted of our control group, and therefore only had 28 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. normal squamous esophageal tissues. The total number of specimens was 108. Metaplasia and dysplasia were placed into an intermediate histopathological category, since there were only 3 dysplastic tissues from the BE group, and just 5 metaplastic tissues from the EA group. mRNA Isolation and cDNA Synthesis Messenger RNA was isolated from biopsy tissues using a single-step guanidinium thiocyanate method (QuickPrep™ Micro mRNA Purification Kit, Amersham Pharmacia Biotech Inc., Piscataway, NJ). First, biopsy tissues were snap frozen in liquid nitrogen and pulverized with a mortar and pestle. Next, the pulverized tissue samples were homogenized using a guanidinium thiocyanate solution provided in the QuickPrep™ Micro mRNA Purification Kit. Messenger RNA isolation and purification was then carried out according to the manufacturer’s instructions. The isolated mRNA was then dissolved in 50 pi of DEPC-treated water. For cDNA synthesis, 20 pi 5x Moloney murine leukemia virus (MMLV) buffer (containing 250 mM Tris-HCI [pH 8.3], 375 mM KCI, and 15 mM MgCI2; Life Technologies, Gaithersburg, MD), 10 j l x I dithiothreitol (DTT; 100 mM; Life Technologies, Gaithersburg, MD), 10 pi dNTP (10 mM each of dATP, dCTP, dGTP, dTTP; Amersham Pharmacia Biotech, Piscataway, NJ), 0.5 pi pd(N)6 random hexamers (50 O.D. dissolved in 550 pi of 10 mM Tris- HCI [pH7.5], and 1 mM EDTA; Amersham Pharmacia Biotech, Piscataway, 29 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. NJ), 2.5 jal bovine serum albumin (BSA; 3 mg/ml in 10 mM Tris-HCI [pH 7.5]; Amersham Pharmacia Biotech, Piscataway, NJ), 2.5 (a l RNAguard ribonuclease inhibitor (5x1000 units; Amersham Pharmacia Biotech, Piscataway, NJ), and 5 pi M M LV reverse transcriptase (200 U/ptl; Life Technologies, Gaithersburg, MD) were added to produce a total volume of 50.5 pi. Complementary DNA synthesis was carried out after combining 40 pi of the reaction mix with 40 pi of the mRNA sample. Reverse transcription (RT) was then performed in a thermocycler using the following temperature program for one cycle: Step 1 8 min at 26°C Step 2 45 min at 45°C Step 3 5 min at 95°C After the RT reaction, the cDNA was stored at 4°C if polymerase chain reaction (Kennedy, Chan et al. 1993) was not subsequently performed . Real-Time Polymerase Chain Reaction (PCR) Quantification Complementary DNA was quantified using a fluorescence based real time detection method (ABI PRISM 7700 Sequence Detection System [TaqMan®], Applied Biosystems, Foster City, CA) as previously described (Gibson, Heid et al. 1996; Heid, Stevens et al. 1996). The PCR reaction mixture consisted of 600 nM primers, 200 nM probe, 2.5 U AmpliTaq Gold Polymerase, 200 pM dATP, dCTP, dGTP, 400 30 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. jjM dUTP, 5.5 mM MgC^, and 1 * TaqMan Buffer A containing a reference dye. These reagents combined for a final volume of 25 |jl (all reagents from Applied Biosystems). Cycling conditions were 50°C for 10 seconds and 95°C for 10 minutes, followed by 46 cycles at 95°C for 15 seconds and 60°C for 1 minute. Genomic DNA background was excluded by amplifying a no- RT control. Colon, liver, and lung RNAs (all from Stratagene, La Jolla, CA) were used as calibrators on each plate. TaqMan® analyses yielded values that were expressed as ratios between two absolute measurements (gene of interest [COX-2]/internal reference gene [(3-actin]). Schematic overall methodology is illustrated in Figure 3.2. The primer and probe sequences used for TaqMan® PCR are listed in Table 3.1. Statistical Analysis Separate analyses of variance (ANOVA) or t tests were performed to compare COX-2 mRNA expression levels in different tissues (adenocarcinoma, Barrett’s esophagus, and normal squamous esophageal tissues) and from different disease status (Cont, BE and EA group), th e analyses were based on log-transformed data to reduce the heteroscedasticity, i.e. the association between the means and the standard deviations. 31 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3.2. Schematic diagram for the esophageal adenocarcinoma study, using fresh tissues. Tissue RNA Isolation RNA I Reverse Transcription cDNA I PCR with Taqman * Data Analysis I0‘ t Thresholi 10‘ -1 0 2 4 6 8 10 14 n i« 2 0 24 28 30 34 36 40 44 49 S O Use Threshold:! 008 [ ISuggest I M ull * Stddev: | tQ.Q |* | TO P I | OmH Threshold: | 2.Q | f~—Baseline ■ ■ ■ | Ct Std Dev FAM - H2 19.136 OOOI FAM - H3 21.379 0 001 FAM - H4 17.933 0.001 a | S tjrt | 3 | Slop | 13 || FAM - H3 2D 232 OOOI - 1 UDfljte Calculations I □ FAM • H4 ■ □ FAM - H5 ■ □ FAM ■ M I I □ FAM - H7 I I □ FAM • H8 ■ Q FAM - H9 ■ □ f a m - h io ■ □ FAM - M11 ■ □ FAM - H12 V iev«r: | a Rn (8... » | Please Set the Threshold Value on All Reporter Layers. Click Q IC te Continue. 32 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.1. TaqMan® PCR primers and probes for COX-2 and p-actin, used in the esophageal adenocarcinoma study. Prim er/Probe Sequences COX-2 Forward primer 5’-GCTCAAACATGATGTTTGCATTC-3’ COX-2 Reverse primer 5’-GCTGGCCCTCGCTTATGA-3’ COX-2 TaqMan probe 6FAM5’-TGCCCAGCACTTCACGCATCAGTT-3’TAMRA p-actin Forward primer 5’-TGAGCGCGGCTACAGCTT-3’ P-actin Reverse primer 5’-TCCTTAAT-GTCACGCACGATTT-3’ P-actin TaqMan probe 6FAM5’-ACCACCACG-GCCGAGCG G-3’TAMRA 33 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. W hen different tissues were compared in the same disease status group, subjects were controlled in the analysis. Means and 95% confidence intervals (CIs) were calculated based on the log-transformed data and then transformed back to the original scale. The least significant d ifferen ce (L S D ) method was used for the pair-wise comparisons, once the overall F- test was significant at the 0.05-level. A p value of < 0.05 was considered statistically significant. 1 1 1-3. Results COX-2 mRNA expression was detectable by quantitative real-time PCR (ABI PRISM 7700 Sequence Detection System [TaqMan®], Applied Biosystems, Foster City, CA) in all 108 (100% ) specimens. COX-2 mRNA expression levels in intermediate tissues were higher than in matching normal squamous tissues in the BE group (n = 19). The mean COX-2 mRNA expression in intermediate tissues was 0.88 (95% Cl: 0.52 to 1.51 in the range from 0.22 to 7.57), compared with 0.16 (95% Cl: 0.09 to 0.27 in the range from 0.04 to 0.70) in normal esophageal tissues (P = 0.0001, Figures 3.3a and b, and Table 3.2). In the group of patients with Barrett’s associated adenocarcinoma (EA group, n = 20), COX-2 mRNA expression levels in cancer tissues were higher than in matching intermediate and normal tissues. The mean COX-2 mRNA expression level was 3.27 (95% Cl: 1.83 to 5.86 in the range from 34 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.2. COX-2 mRNA expression in tissues from patients with esophageal adenocarcinoma (EA group), Barrett’s esophagus (BE group), and patients without a history of adenocarcinoma, Barrett’s esophagus, or gastroesophageal reflux disease (Cont group). Pathology n COX-2 Gene Expression Mean Range (95% Cl) P value EA Group 20 P < 0.0001 Adenocarcinoma 3.27 0.43-62.54(1.83-5.86) Intermediate 1.22 0.10-29.01 (0.68-2.18) Normal 0.39 0.04-1.57 (0.22-0.70) BE Group 19 P = 0.0001 Intermediate 0.88 0.22-7.57 (0.52-1.51) Normal 0.16 0.04-0.70 (0.09-0.27) Cont Group 10 Normal 0.19 0.60-1.63 (0.09-0.39) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3.3a. Box and whisker plots of relative COX-2 mRNA expression levels in normal and intermediate tissues from patients with Barrett’s esophagus. Mean values are shown as a horizontal bar inside each box. ‘ Outlier C o '</) < /> 2> CL X Q) E C M ■ X o o a> > > ♦ - » o o ' -2 N = P=0.0001 *0.70 T mean=0.16 (0.04-0.70) (95% Cl: 0.09 - 0.27) 19 Normal *7.57 mean=0.88 (0.22-7.57) (95% Cl: 0 .5 2 - 1.51) 19 Intermediate Barrett’s esophagus without adenocarcinoma 36 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Relative COX-2 m R N A expression Figure 3.3b. Paired data showing relative COX-2 mRNA expression levels from intermediate and matching normal esophageal tissues from patients with Barrett’s esophagus. Mean values are shown as diamonds (♦). (0.22-7.57) P = 0.0001 / / / (0.04-0.70) / mean=0.16 mean=0.88 L ^ --------------------------------------■ Normal______________________ Intermediate Barrett’s esophagus without adenocarcinoma Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 0.43 to 62.54) in adenocarcinoma (A) tissues, compared with 1.22 (95% Cl: 0.68 to 2.18 in the range from 0.10 to 29.01) in intermediate (I) tissues, and 0.39 (95% Cl: 0.22 to 0.70 in the range from 0.04 to 1.57) in tissues with normal (N) esophagus (P < 0.0001, Figure 3.4a and b, and Table 3.2). Figure 3.4a and b show that the mean level of COX-2 mRNA expression was significantly higher in tissues with intermediate progression and adenocarcinoma, compared to matching normal esophageal tissues (N versus I: P = 0.008; N versus A: P < 0.0001). Our result also showed that the mean level of COX-2 mRNA expression was significantly higher in tissues with adenocarinoma compared to matching intermediate tissues (A versus I: P = 0.02). W e also compared COX-2 mRNA expression levels in normal squamous specimens from each of the three disease status groups. Mean COX-2 mRNA expression levels were different in normal histological specimens from each group (P = 0.019, Figure 3.5 a and b). Interestingly, the mean COX-2 mRNA expression level in histologically normal tissues was higher from patients in the EA group (mean = 0.39, 95% Cl: 0.25-0.61 in the range from 0.04 to 1.57, n = 20) than from patients in the BE group (mean = 0.16, 95% Cl: 0.10-0.25 in the range from 0.04 to 0.70, n = 19) and the control group (mean = 0.19, 95% Cl: 0.10-0.34 in the range from 0.06 to 1.63, n = 10). COX-2 expression levels in normal tissues were also compared between different groups (N[Cont] versus N[BE]: P=0.63 38 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3.4a. Box and whisker plots of relative COX-2 mRNA expression levels from normal(N), intermediate(l), and adenocarcinoma(A) tissues from patients with esophageal adenocarcinoma. Mean values are shown as a horizontal bar inside each box. *Outlier C o '5 5 (0 £ o . x < D a : E C M I X o o < D > JO C D O ' 20 15 10 - 5 _ N = P < 0.0001 N vs I: P = 0.008 I vs A: P = 0.02 A vs N: P < 0.0001 mean=0.39 (0.04-1.57) (95% Cl: 0.22 ~ 0.70) *29.01 mean:=1.22 (0.10-29.01) (95% Cl: 0.68- 2.18) *62.54 *15.44 =3.27 T (0.43-62.54) (95% Cl: 1.83 - 5.86) 20 Normal (N) 20 Intermediate (I) 20 Adenocarcinoma (A) Barrett’s associated esophageal adenocarcinoma 39 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Relative COX-2 m R N A expression Figure 3.4b. Matched data showing relative COX-2 mRNA expression levels from normal, intermediate, and adenocarcinoma tissues from patients with esophageal adenocarcinoma. Mean values are shown as diamonds (♦). 30 20 P < 0.0001 Nvs I: P= 0.008 I vs A: P=0.02 A vs N: P<0.0001 (0.43-62.54) / (0.10-29.01 )\ X v 10 (0.04-1.57) mean=0.39 rnean=1.22 mean=3.27 Normal (N)________ Intermediate (I) Adenocarcinoma (A) Barrett’s associated esophageal adenocarcinoma 40 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3.5a. Box and whisker plots of relative COX-2 mRNA expression levels from normal squamous epithelial tissues from 1) a control group, 2) patients with Barrett’s esophagus, and 3) patients with esophageal adenocarcinoma. Mean values are shown as a horizontal bar inside each box. *Outlier C o "S > (/> S > Q . X < D o r E CM I X o o < D > iS < D o ' 2.0i 1.5- 1.0 - .5- 0 .0- -.5, N P=0.019 N(Cont)vsN(BE): P=0.63 N(BE)vsN(EA):P=0.007 N(Cont)vsN(EA): P=0.069 0.70 1.63 mean=0.19 mean=0.16 mean=0.39 (0.06-1.63) (95% Cl: 0.10-0.36) 10 N(Cont group) (0.04-0.70) (0.04-1.57) (95% Cl: 0.10- 0.25) (95% Cl: 0.25 - 0.61) 19 N (BE group) 20 N(EA group) Normal esophagus 41 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3.5b. Scatter plot of relative COX-2 mRNA expression levels from normal squamous epithelial cells taken from 1) a control group, 2) patients with Barrett’s esophagus, and 3) patients with esophageal adenocarcinoma. Mean values are shown as a horizontal bar. C o " t o to 2> Q. X 0) E Csl I X o o o > J 2 o £ 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 P = 0.019 N(Cont)vsN(BE): P=0.63 N(BE)vsN(EA):P=0.007 N(Cont)vsN(EA): P=0.069 (0.04-0.70) (0.06-1.63) ♦ ♦ mean=0.19 T * ♦♦» mean=0.16 10 N (Cont) 19 N (BE) (0.04-1.57) mean=0.39 T T 20 N (EA) Normal esophagus 42 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. [not significant]; N[BE] versus N[EA]: P=0.007; N[Cont] versus N[EA]: P=0.069 [not significant], Figure 3.5 a and b). In that comparison, only COX- 2 mRNA expression levels of normal epithelial cells from the BE group were shown to be significantly different from those of the EA group. Further analysis was performed to search for other differences in COX-2 mRNA expression during the progression from Barrett’s disease to esophageal adenocarcinoma. W e compared COX-2 mRNA expression levels of metaplastic tissues from the BE group (M[BE], mean = 0.80, 95% Cl: 0.46 to 1.39 in the range from 0.22 to 7.57, n=16) with dysplastic tissues from the EA group (D[EA], mean = 1.89, 95% Cl: 0.78 to 4.56 in the range from 0.13 to 29.01, n=15). The difference between M[BE] and D[EA] did not reach the level of significance (P=0.083, Figure 3.6). M[BE] was also compared with adenocarcinoma tissues from the EA groups (A[EA], mean = 3.27, 95% Cl: 1.90 to 5.63 in the range from 0.43 to 62.54, n=20). A[EA] had a significantly higher level of COX-2 expression than M[BE] (P<0.001, Figure 3.7). Finally, D[EA] (mean = 1.89, 95% Cl: 0.78 to 4.56, n=15) was compared with A[EA] (mean = 2.56, 95% Cl: 1.25 to 5.26 in the range from 0.43 to 15.44, n=15) only for patients who had dysplastic tissues, controlled for subjects. There was no significant difference between these two tissues (P=0.53, Figure 3.8). 43 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3.6. Box and whisker plots of relative COX-2 mRNA expression levels for 1) dysplastic(D) tissues from patients with esophageal adenocarcinoma, and 2) metaplastic(M) tissues from patients with Barrett’s esophagus. Mean values are shown as a horizontal bar inside each box. *Outlier C O '</) < /> £ Q . X 0 < z a : E C M ■ X o o 0 > + ■ » 0 0 O ' 12 10 8' N = P = 0.083 *7.60 mean = 0.80 (0.22-7.57) (95% Cl: 0 .4 6 - 1.39) 16 M (BE) *29.01 mean = 1.89 (0.13-29.01) (95% Cl: 0.78 - 4.56) 15 D (EA) 44 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3.7. Box and whisker plots of relative COX-2 mRNA expression levels for 1) adenocarcinoma(A) tissues from patients with esophageal adenocarcinoma, and 2) metaplastic(M) tissues from patients with Barrett’s esophagus. Mean values are shown as a horizontal bar inside each box. ‘ Outlier C o '</> it) £ Q . X d) £ E CM ■ X o o 0) > ■ Q > C £ 14 12 10 8 6 4 2 0 -2 N = P < 0.001 *7.60 mean = 0.80 (0.22-7.57) (95% Cl: 0 .4 6 - 1.39) 16 M (BE) *62.54 *15 44 mean = 3.27 (0.43- (95% Cl: 1 52.54) .90 - 5.63) 20 A (EA) 45 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3.8. Box and whisker plots of relative COX-2 mRNA expression levels for dysplastic (D) and adenocarcinoma(A) tissues from patients with esophageal adenocarcinoma. Mean values are shown as a horizontal bar inside each box. ‘ Outlier C O “ < 7 > (/> £ a x 0 < z c c E CM ■ X o o 0 > ■ +-• 0 0 £ 14' 12 - 10 ' 8- 6- 4- 2- 0- - 2 , N = P = 0.53 ‘ 29.01 *16.93 mean =1.89 (0.13-29.01) (95% Cl: 0.78 ~ 4.56) 15 D (EA) *15.44 *12.46 mean=2.56 (0.43-15.44) (95% Cl: 1.25 ~ 5.26) ~ 1 5 A (EA ) 46 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 111-4. Discussion The presence of Barrett’s esophagus is the most significant risk factor for the development of esophageal adenocarcinoma (Lagergren, Bergstrom et al. 1999). However, mechanisms through which it develops into esophageal adenocarcinoma are not fully understood. There is substantial evidence that progression to adenocarcinoma is associated with up- regulated COX-2 gene expression. Shirvani et al. (2000) showed potent COX-2 induction by both acid and bile acids and up-regulated COX-2 expression in Barrett’s dysplasia and esophageal adenocarcinoma (Shirvani, Ouatu-Lascar et al. 2000). Morris et al. (2001) also showed the increase of COX-2 expression in Barrett’s dysplasia and adenocarcinoma with immunohistochemical analysis (Shirvani, Ouatu-Lascar et al. 2000; Morris, Armstrong et al. 2001). Our results demonstrate that COX-2 mRNA is up-regulated during the progression of esophageal adenocarcinoma, suggesting that COX-2 is correlated with the development of Barrett’s associated esophageal adenocarcinoma. Figures 3.3 b and 3.4 b demonstrate the pattern of COX-2 expression at the individual level, in the progression of Barrett’s disease. Figures 3.6 (a and b) compare up-regulated COX-2 gene expression in different stages during the progression to Barrett’s associated esophageal adenocarcinoma, confirming increased induction of COX-2 protein expression sequentially along the adenocarcinoma sequence, as shown by 47 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Shirvani et al. (2000) and Morris et al. (2001) (Shirvani, Ouatu-Lascar et al. 2000; Morris, Armstrong et al. 2001). As shown in Figure 3.4b, individual COX-2 mRNA expression levels increased as cells become increasingly dysplastic, as measured in tissues from patients in the EA group. This result suggests that the induction of COX-2 expression may be an early event in tumor development. However, 6 of the 20 patients (30%) whose COX-2 expression levels were up-regulated from normal to intermediate disease, had reduced levels in adenocarcinoma tissues. Buskens et al. (2002) showed that patients with Barrett’s associated esophageal adenocarcinoma and low COX-2 protein expression (measured by immunohistochemical analysis) have better survival rates and less risk of developing distant metastases and local recurrences, compared to those with high COX-2 protein expression (Buskens, Van Rees et al. 2002). The follow- up information for patients in the EA group is not currently available, but those patients who have lower COX-2 mRNA expression in adenocarcinoma tissues may fit into the category of low COX-2, as described by Buskens et al. (2002). When we compared normal esophageal tissues among members of the control group with those in the Barrett’s esophagus and Barrett’s associated esophageal adenocarcinoma groups, our results demonstrated that COX-2 mRNA is constitutively expressed in normal esophageal tissues, and that the mean COX-2 mRNA expression is higher in normal esophageal 48 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. tissues from the patients with esophageal adenocarcinoma than it is in those from the other groups (P=0.019, Figure 3.5a and b). There are three possible explanations for this: 1) Field effects, changes in COX-2 expression of normal tissues adjacent to adenocarcinoma cells, may be associated with the location of the biopsy taken from cancer patients. Thus, even though the tissues appeared normal by histology, nearby cancer tissues may have induced expression of COX-2. 2) CO X-2 expression in normal esophageal tissues from the EA group may have been induced by the long exposure to gastroesophageal reflux. Since patients with cancer have greater exposure to reflux disease than those without cancer, it is possible that stomach and bile acids, put into contact with the mucosa through reflux, stimulated COX- 2 expression in normal tissues. The study of Shirvani et al. (2000) reported that both acid and bile salts induced COX-2 expression (Zhang, Subbaramaiah et al. 1998; Shirvani, Ouatu-Lascar et al. 2000). 3) Induction of COX-2 expression in normal esophageal tissues from patients with cancer caused the progression of esophageal a denocarcinom a, indicating th a t COX-2 overexpression is an early event in the normal esophageal squamous cell transformation process of Barrett’s associated esophageal adenocarcinoma. Several studies 49 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. demonstrated that COX-2 contributes to tumorigenesis by 1) decreasing apoptosis through induction of anti-apoptotic proto-oncogene Bcl-2 expression (Tsujii and DuBois 1995; Souza, Shewmake et al. 2000; Liu, Chang et al. 2001), 2) inducing production of several angiogenetic factors, such as vascular endothelial growth factor (VEGF) and basic fibroblastic growth factor (b-FGF) (Tsujii, Kawano et al. 1998), and 3) enhancing prostaglandin synthesis, which stimulates an cellular proliferation and motility (Sheng, Shao et al. 2001). Differences in COX-2 mRNA expression in normal esophageal tissues between patients in the BE group and patients in the EA group were statistically significant (P= 0.007). However, when we compared COX-2 mRNA expression between members of the BE group and those in the control group, we found statistically insignificant differences in COX-2 mRNA expression (P=0.63, Figure 3.5 a and b). This result implies that CO X-2 can be used as a potential factor to predict the risk of disease progression from Barrett’s disease to Barrett’s associated esophageal adenocarcinoma. Our statistical analysis also showed that the differences in COX-2 mRNA expression between the control group and EA group were not statistically significant (P=0.069, Figure 3.5 a and b), but this may be due to small sample sizes. There is a trend toward higher mRNA expression of COX-2 in normal esophageal tissues from members of the EA group. Therefore, to determine whether COX-2 gene expression is an early determinant in the 50 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. progression from normal to Barrett’s disease, further investigation with larger sample sizes should be conducted. When metaplastic tissues from the BE group, and dysplastic and adenocarcinoma tissues from the EA group were compared, as shown in Figure 3.6, 3.7 and 3.8, the difference in mean COX-2 gene expression between Barrett’s metaplasia and esophageal adenocarcinoma was statistically significant (P< 0.001), but we did not observe significant differences in mean COX-2 gene expression between Barrett’s metaplasia and Barrett’s dysplasia (P=0.083, Figure 3.6), and between Barrett’s dysplasia and esophageal adenocarcinoma (P=0.53, Figure 3.8). There were tendencies toward higher gene expression in dysplastic tissues than in metaplastic tissues, but again it is likely that the number of samples was too small to reach significance. The result suggested an increase in mean COX- 2 gene expression with advancement through the Barrett’s metaplasia- dysplasia-adenocarcinoma sequence. In conclusion, our findings demonstrate that COX-2 mRNA expression is correlated with the progression of Barrett’s associated esophageal adenocarcinoma, and therefore suggest that COX-2 may be a potential therapeutic target in this disease. Ferrario et al. (2002) reported that COX-2 inhibitors were effective in combination with conventional chemotherapy in the condition of induced COX-2 expression (Ferrario, Von Tiehl et al. 2002). Therefore, COX-2 inhibitors, as an adjuvant to chemotherapy, may give 51 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. benefits to patients with esophageal adenocarcinoma and high COX-2 expression levels. Furthermore, the use of COX-2 inhibitors as tumor regressive agents may help reduce or even reverse disease progression in patients with dysplasia. This claim is supported by several studies which reported that COX-2 inhibitors, as potential cancer chemoprevention drugs, decreased the risk of development esophageal cancer (Thun, Namboodiri et al. 1993; Funkhouser and Sharp 1995), and reduced the size and number of rectal adenomatous polyps in patients with familial adenomatous polyposis (FAP) (Giardiello, Hamilton et al. 1993). As demonstrated in this study COX- 2 may also be a useful factor in determining the prognosis during the progression from Barrett’s disease to esophageal adenocarcinoma. 52 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER IV CYCLOOXYGENASE-2 (COX-2) GENE EXPRESSION IS ASSOCIATED WITH SURVIVAL IN CURATIVELY RESECTED NON-SMALL CELL LUNG CANCER. IV-1. Introduction Lung cancer is the most common cause of cancer-related deaths in the United States, accounting for more deaths than prostate, breast, and colorectal cancer combined. Each year approximately 170,000 new cases of lung cancer are diagnosed in the US (Greenlee, Murray et al. 2000). Lung cancers can be classified into two types: small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC). Non-small cell lung cancer is more common (75% of lung cancers) than small cell lung cancer, and is divided into three main subtypes— squamous cell carcinoma, adenocarcinoma, and large cell carcinoma— based on histological features. Lung cancer can be diagnosed by chest radiography, bronchoscopy, needle biopsy, and other techniques. However, this disease is commonly diagnosed at an advanced stage, at which time it may not be curable. Late diagnosis results from difficulties in recognizing the specific signs and symptoms of lung cancer. 53 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The symptoms most often present in patients with lung cancer are usually the same as those found in people who smoke, or who have other respiratory disorders, including upper respiratory tract infections. Radical surgery offers the only chance for cure in patients with non-small cell lung cancer (NSCLC) stage I to stage Ilia. Despite continuous improvements in the detection and treatment of lung cancer in the past two decades, the 5-year survival rate remains less than 15% (Ginsberg 1997). To improve the outcome of patients with NSCLC, the development of a prognostic classification based on molecular alterations will be crucial. Such a classification could provide additional diagnostic tools and, eventually, more effective therapeutic options. As with colorectal and esophageal cancer, recent studies have demonstrated that aspirin may also reduce the incidence of lung cancer (Schreinemachers and Everson 1994). Although overexpression of COX-2 has been reported in lung cancer (Wolff, Saukkonen et al. 1998; Hida, Yatabe et al. 1998b; Achiwa, Yatabe et al. 1999; Ochiai, Oguri et al. 1999; Marrogi, Travis et al. 2000), the role of COX-2 protein expression in NSCLC remains controversial. While Achiwa et al. (1999) reported an association between COX-2 overexpression and survival in patients with stage I adenocarcinoma of the lung (Achiwa, Yatabe et al. 1999), Marrogi et al. (2000) was unable to detect any association between COX-2 expression and clinical outcome in patients with NSCLC (Marrogi, Travis et al. 2000). Prior 54 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. to this investigation, no studies concerning the role of COX-2 mRNA expression in NSCLC have been reported. To determine the relevance of COX-2 mRNA expression in NSCLC, and its role in predicting prognosis, we performed quantitative real-time reverse transcription polymerase chain reaction (RT-PCR; TaqMan) (Gibson, Heid et al. 1996; Heid, Stevens et al. 1996) on surgically excised tumor specimens from 89 patients with curatively resected NSCLC. IV-2. Materials and Methods Patients and Specimens Tumor specimens from 89 patients with NSCLC, available from a previous prospective clinical trial of 103 consecutive patients (Schneider, Praeuer et al. 2000), were included in this study. There were 67 (75% ) men and 22 (25% ) women, with a median age of 64 years (range 34-83 ). Forty- one (46% ) patients were diagnosed with squamous cell carcinoma, 33 (37% ) with adenocarcinoma, and 15 (17% ) with large cell carcinoma. The primary tumors were graded histopathologically as well-differentiated (G1, one patient), moderately-differentiated (G2, 19 patients), and poorly-differentiated (G3, 69 patients). Tumor staging was performed according to the International Union Against Cancer (UICC) TNM classification scheme (Mountain 1997): 44 patients (50%) had stage I tumors, 18 (20% ) had stage II tumors, and 27 (30%) had stage Ilia tumors. All 89 patients underwent 55 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. thoracic surgery. All tumors were radically removed (RO resection) by lobectomy (n = 57), bilobectomy (n = 11), pneumonectomy (n = 11), and extended pneumonectomy (n = 10) including mediastinal lymphadenectomy. Patients with histopathological stage Ilia tumors additionally received postoperative radiotherapy. Informed consent was obtained from each patient. The median follow-up time was 85.9 months (range 63 -1 0 5 ), and no patient was lost to follow-up. Tissues used in gene expression analysis were obtained during surgery immediately after lung resection, and before starting mediastinal lymphadenectomy. They were immediately frozen in liquid nitrogen and stored at -8 0 °C . Six micrometer frozen sections were taken from blocks of tumor tissue, and starting with the first section, every fifth section was routinely stained with hematoxylin and eosin and evaluated histopathologically. Sections were pooled for analysis from areas estimated to have at least 75% malignant cells. mRNA Isolation and cDNA Synthesis Messenger RNA (mRNA) was isolated using a single-step guanidinium thiocyanate method (QuickPrep™ Micro mRNA Purification Kit, Amersham Pharmacia Biotech Inc., Piscataway, NJ), as described for mRNA isolation of esophageal tissues in chapter three. Complementary DNA was also synthesized as described in the synthesis of cDNA in the esophageal adenocarcinoma study (chapter three), 56 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. using 80 jlxI final volumes in the RT reaction. A thermocycler was also used with the same settings as described in the esophageal adenocarcinoma study (Chapter three). Following reverse transcription, cDNA was stored at 4°C if polymerase chain reaction was not subsequently performed. Real-Time Polymerase Chain Reaction (PCR) Quantification Complementary DNA was quantified as described in Chapter two, using a dual-labeled fluorogenic oligonucleotide probe. Schematic overall methodology is illustrated in Figure 4.1. The primers and probe sequences used for PCR are the same as those shown in Table 3.1. Statistical Analysis The chi-square (x2) test was used to analyze the association between categorical clinicopathological data and COX-2 expression status. Flazard ratios were used to calculate the relative risks of death, given patient disease profiles. These calculations were based on the Pike estimate, with the use of the observed and expected number of events as calculated in the log-rank test statistic (Pike 1972). The maximal chi-square method of Miller and Siegmund (Miller 1982) and Flalpern (Flalpern 1982) was adapted to determine which expression value best segregated patients into poor and good prognosis subgroups (based on their probability of survival), with the log-rank test and the stratified log-rank test used to measure the strength of the grouping. To determine a p-value that would be interpreted as a 57 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 4.1. Schem atic diagram for the lung cancer study using fresh tissues Tissue I RNA Isolation I RNA I Reverse Transcription i cDNA i PCR with Taqman Data Analysis 10* 1 Threshol 10*-1 0 2 4 6 6 10 14 IS 20 24 28 30 34 38 40 44 40 50 — S a m p les • □ FAM • H4 ■ □ FAM-HS ■ □ FAM - H6 I I □ FAN - H7 I I □ FAH - H9 ■ Q FAM - H9 ■ □ FAN - H1Q ■ □ FAN - H11 ■ □ FAN - H12 - Use Threshold: I 008~] [Suggest I C t e td d » v NuH *S1d<t»v| 10 0 1*1 .001 1 Omit Threshold:! 2.0 1 FAM - MS 51.579 0 001 FAM - H4 17 953 O.OQ1 ■ | S tart:| 3 | Stop :| 15 | FAM - HS 20.232 □ 001 * 1 Update Calculations 1 -- o o m-- V U v e r I A fin (B. i I Reporter: [ FAM 1 3 PIB3SB Sot the Threshold Valua on All Reporter Layers. Click Q IC to Continue. 58 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. measure of the level of confidence in the association based on the maximal chi-square analysis, 1,000 bootstrap-like simulations were run. They were used to estimate the distribution of the maximal chi-square statistics with the null hypothesis “no association” (Halpern 1982), in order to obtain an appropriate p-value. Multivariate analysis was performed under the Cox proportional hazards regression model. The level of significance was set to P < 0.05. All reported p-values were based on two-sided tests. IV-3. Results COX-2 mRNA expression was detected during quantitative real-time PCR (TaqMan) in all 89 (100% ) tumor specimens. The median COX-2 mRNA expression, expressed as a ratio to the internal reference gene, p- actin, was 0.85 (range 0.02-15.77). The median follow-up time was 85.9 months for the 89 patients analyzed in this study, and the median survival time was 59.7 months (range 3.8-105.3). The maximal chi-square statistic of Miller and Siegmund (Miller 1982) and Halpern (Halpern 1982) was adapted to determine which COX-2 mRNA expression level best segregated patients into one of two prognosis subgroups. This method found that segregation based on CO X-2 mRNA levels was best achieved with a relative expression level of 0.6. Based on this criterion, 47 (53%) patients were placed into the good prognosis subgroup (low COX-2 expression status), and 42 (47%) patients were placed 59 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. into the poor prognosis subgroup (high COX-2 expression status). Table 4.1 shows associations between clinicopathological data and CO X-2 expression status. No statistically significant associations were detectable. Median and five-year survival rates, depending on various clinical variables, and COX-2 expression status, are summarized in Table 4.2. The median survival for patients with a high COX-2 mRNA expression status was 31.1 months (95% confidence interval 8.72-53.48), whereas the median survival for those with a low COX-2 mRNA expression status was not reached (P = 0.0032, log-rank test) after five years. The relevant survival curves are presented in Figure 4.2, and show five-year survival rates of 31.2 (95% Cl: 23.5 to 38.9) for patients with high COX-2 gene expression levels, and 61.7 (95% Cl: 54.6 to 68.8) for those with low COX-2 gene expression levels. Furthermore, by stratifying patients by stage, CO X-2 mRNA expression was shown to be an even stronger predictor of survival for patients with curatively resected NSCLC (P = 0.0002, stratified log-rank test). Next, the importance of COX-2 as a prognostic factor was investigated using the Cox proportional hazards model analysis. Two logistic regression models were tested. Model A included the parameters age, gender, histopathological type, UICC TNM tumor stage, and COX-2 expression status. Model B included the pT and pN categories instead of histopathological tumor stage. Significant independent prognostic factors were shown to be UICC TNM tumor stage (P < .001) and CO X-2 (P < .001) 60 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4.1. Associations between clinicopathological variables and COX-2 expression status Variable COX-2 Low High P Sex Male 38 29 n.s. Female 9 13 Age >65 25 22 n.s. <65 22 20 pT Category pT1 11 9 n.s. pT2 27 29 pT3 9 4 pN Category pNO 27 22 n.s. pN1 15 9 pN2 5 11 UICC Stage I 24 20 n.s. II 9 9 Ilia 14 13 Histopathology Squamous Cell Carcinoma 26 15 n.s. Adenocarcinoma 14 19 Large Cell Carcinoma 7 8 Grading Well differentiated 1 0 n.s. Moderately differentiated 11 8 Poorly differentiated 35 34 Abbreviations: n.s. (not significant); 61 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4.2. Survival in NSCLC based on clinical and molecular param eters Parameter n 5-Year Survival Probability Median Survival (Months) 95% Cl P value UICC stage 1 44 0.68 ± 0.07 n.r. . <0.0001 II 18 0.50 ±0.11 39.37 - Ilia 27 0.11 ±0.06 20.40 ±5.14 10.32 -30.48 PT pTi 20 0.65 ±0.11 n.r. . 0.0155 pT2 56 0.46 ± 0.07 59.73 - pTs 13 0.23 ±0.12 26.67 ±6.09 14.73-38.61 pN pN0 49 0.67 + 0.07 n.r. _ <0.0001 pNi 24 0.37 ±0.09 33.97 + 10.09 14.20-53.74 pN2 16 0 16.70 ±4.15 8.56 - 24.84 Histopathology Squamous Cell Ca 41 0.54 + 0.08 n.r. n.s. Adenocarcinoma 33 0.39 ± 0.09 45.47 + 8.92 27.98-62.69 Large Cell Ca 15 0.48 ±0.15 n.r. - COX-2 Overall Low 47 0.62 ± 0.07 n.r. 0.0032 High 42 0.31 ±0.08 31.10 ± 11.42 8.72 - 54.48 Stratified by Stage I Low 24 0.88 ±0.07 n.r. 0.0002** High II Low 20 0.51 ±0.13 n.r. - 9 0.66 + 0.16 n.r. - High 9 0.33 ±0.16 27.67 ±8.41 11.19-44.15 Ilia Low 14 0.21 +0.11 29.00 ±8.79 11.77-46.23 High 13 0 12.60 + 3.65 5.44-19.76 Abbreviations: n.r. (not reached); - (cannot be calculated); Cl 95% (95% confidence interval); n.s. (not significant); n (number of patients); log-rank test based on median survival; "stratified by stage. 62 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Estimated probability o f survival Figure 4.2. COX-2 mRNA expression (the ratio between COX-2 and the internal reference gene p-actin) in specimens of primary non-small cell lung cancer. High COX-2 mRNA levels contained ratios greater than 0.6; low levels contained ratios less than or equal to 0.6. P=0.0032 Low COX-2 expression (n=42) High COX-2expression (n=47) 0.0 0 20 60 80 100 120 40 Survival in months 63 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. expression status in model A, and the pN categories (P < .001) and COX-2 expression status (P = .013) in model B. Table 4.3 shows the statistically significant parameters in the two regression models. IV-4. Discussion To test whether the level of COX-2 expression is of prognostic importance in NSCLC, we used quantitative real-time RT-PCR (TaqMan) to analyze COX-2 mRNA expression in tumor specimens of 89 patients with curatively resected NSCLC. Expression of COX-2 was detectable in all (100% ) specimens analyzed, although, the intra-tumoral content of COX-2 mRNA expression varied over a 788-fold range. This observation of seemingly variable amounts of mRNA implies heterogeneity of expression patterns within individual tumor cells and may reflect the biologic behavior of these tumors. Overexpression of COX-2 has been reported in human NSCLC (Wolff, Saukkonen et al. 1998; Ochiai, Oguri et al. 1999; Marrogi, Travis et al. 2000), and previous studies have shown that COX-2 overexpression may alter the biologic behavior of tumor cells in a number of ways (Tsujii and DuBois 1995; DuBois, Shao et al. 1996; Tsujii, Kawano et al. 1997). Tsujii et al. (1997) showed that constitutive expression of CO X-2 could result in phenotypic changes that enhance the metastatic potential of colorectal cancer cells, leading to an increased invasiveness 64 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4.3. Cox’s proportional hazard regression models Model Parameter Odds Ratio 95% Cl P value A Stage <0.001 l/llla 0.15 0.07-0.29 <0.001 ll/llla 0.30 0.14-0.67 0.003 COX-2 0.34 0.18-0.62 <0.001 B pN <0.001 pNo/pNi 0.14 0.06-0.30 <0.001 pN0 /pN2 0.33 0.15-0.74 0.006 COX-2 0.44 0.23-0.84 0.013 Abbreviations: Cl 95% (confidence interval for odds ratio); Parameter section: e.g. stage l/llla means stage I compared to stage Ilia. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (Tsujii, Kawano et al. 1997). Analogous to colorectal carcinomas, Hida et al. (1998b) reported that a marked increase in COX-2 immunoreactivity in NSCLC adenocarcinoma is associated with tumor-invasive lesions and lymph node metastases, suggesting that increased COX-2 expression may be associated with an invasive and more aggressive phenotype in this disease (Hida, Yatabe et al. 1998b). Indeed, Achiwa et al. (1999) was the first to show the prognostic significance of elevated CO X-2 expression in human cancer. Although he found no association between elevated COX-2 protein expression and clinical outcome in stage I to III adenocarcinomas of the lung, he did correlate high COX-2 gene expression with decreased survival in a subgroup of adenocarcinoma patients with primarily stage I disease (Achiwa, Yatabe et al. 1999). In contrast to our findings, Marrogi et al. (2000) recently reported no association between COX-2 protein expression and clinical outcome in patients with NSCLC (Marrogi, Travis et al. 2000). There are three possible explanations for these discordant findings: First, Marrogi et al. investigated COX-2 expression at the protein level, whereas we examined COX-2 expression at the mRNA level. Second, immunohistochemical methods are semiquantitative compared with the real time quantitative RT-PCR technique used in our study. Third, the study population analyzed by Marrogi et al. (2000) consisted of patients with stage I to IV NSCLC, whereas only patients with curatively resected NSCLC stage I to Ilia were included in our study (Marrogi, Travis et al. 2000). 66 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The most striking finding in our study is the association between high COX-2 mRNA expression levels and reduced survival in patients with curatively resected NSCLC. This finding may be clinically important, particularly because our study population represents a large number of patients with different histopathological subtypes of lung cancer, in all of whom curative surgery was achieved. Our observation adds another step toward the development of a molecular classification of NSCLC and suggests that quantification of COX-2 mRNA expression might help identify NSCLC patients 1) at greater risk for early disease recurrence, and 2) who may benefit from additional therapy to control their disease. Previous studies have shown that treatment with NSAIDs decreases cell growth and induces apoptosis in NSCLC in vitro (Hida, Leyton et al. 1998a; Hida, Kozaki et al. 2000). Interestingly, the responsiveness to selective COX-2 inhibitors was influenced by the degree of COX-2 expression (Hida, Leyton et al. 1998a). Greater COX-2 expression resulted in increased tumor responsiveness to COX-2 inhibition. Schreinemachers et al. (1994) also reported that aspirin was associated with reduced lung cancer incidence in men in an epidemiological study (Schreinemachers and Everson 1994). The availability of selective COX-2 inhibitors, combined with the findings of other investigators, gives our results additional importance because increased COX-2 expression might represent a direct therapeutic 67 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. target in patients with NSCLC. Further studies are required to investigate disease regression and chemopreventive effects of CO X-2 inhibitors in patients with NSCLC in the early stages (I and II), as well as the effect of chemotherapy combined with COX-2 inhibitors on lung cancer patients diagnosed with stage III disease and above. Other studies have demonstrated COX-2 involvement in tumor invasion (Hida, Yatabe et al. 1998b) and tumor-associated immune suppression (Huang, Stolina et al. 1998) in lung cancer. This is significant because lung cancer is known to metastasize to brain, liver, bones, and adrenal glands. Therefore, additional investigation of the effect of COX-2 inhibitors on lung cancer metastasis is necessary to assess their potential benefit to patients with NSCLC. These studies will be needed to gain more insight into the potential clinical usefulness of this therapeutic approach. 6 8 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER V A MOLECULAR-EPIDEMIOLOGICAL ANALYSIS OF NSAIDS, COX-1 AND COX-2 GENE EXPRESSION, AND RISK OF COLORECTAL ADENOMAS V-1. Introduction Colorectal cancer is expected to be the second most common cause of cancer mortality in the United States (Greenlee, Murray et al. 2000). There is histological evidence suggesting that the majority of colonic adenocarcinomas arise from adenomatous polyps (Morson 1974; Saltz 2002), thus identifying them as precursors to colorectal cancer. When an adenoma increases in size, becomes more villous, or contains more high- grade dysplasia, it increases its risk of progressing to adenocarcinoma. However, some carcinomas develop exceptionally from small, highly dysplastic flat adenomas (Owen 1996; Saltz 2002) (Figure 5.1). The detection of early-stage disease at diagnosis is associated with significantly greater survival compared with late detection. Currently, 50% of patients diagnosed with colorectal cancer have metastases at diagnosis, and it has been reported that current conventional chemotherapies only bring about 69 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 5.1. Multi-step progression of colon cancer. caem0^ i$ r Normal Epithelium Myperprollferatwe Epithelium Aberrant Crypt Foci Smail Adenoma Large Adenoma Colon Carcinoma M etastasis Rex et al. 2002 JPSM 23:S41-S50 70 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. marginal survival benefit. Early detection and excision of polyps can prevent the development of colorectal adenocarcinoma (Dunlop 1997). For this reason, research has focused on developing strategies for early detection based on molecular alterations and prevention of early disease. Accumulating evidence suggests that aspirin and other non-steroidal anti-inflammatory drugs (NSAIDs) are associated with a reduced risk of colorectal cancer and adenomas, and can reduce the number and size of polyps in patients with familial adenomatous polyposis (FAP) (Baron 1995; Smalley and DuBois 1997; Taketo 1998b). One of the primary pharmacological properties of NSAIDs is the inhibition COX-1 and COX-2 enzymatic activities (Taketo 1998a). This halts the initial step in prostaglandin synthesis, including the production of PGE2, which is elevated in colorectal tumors (Taketo 1998b; Willis and Colburn 2002). It has, therefore, been hypothesized that the protective effect of NSAIDs is mediated, at least in part, by a COX-related mechanism (Taketo 1998a; Taketo 1998b). Furthermore, experimental results suggest that COX-2 is induced in polyp tissues in mice at a very early stage of development, and plays a key role in polyp formation. Studies also report that inactivation of COX-2 suppresses polyp growth (Oshima, Dinchuk et al. 1996). In a study of human colorectal adenomas, variable expression of COX-1 was detected in all nine adenomas studied and COX-2 expression was detected in eight of nine adenomas (Chappie, Scott et al. 2002). 71 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Epidemiological studies of NSAIDs and COX expression are warranted; however, large-scale studies were precluded in the past because of an inability to measure the transcriptional level of gene expression from paraffin-embedded tissues that are normally available in epidemiological studies. A recent development in our lab has enabled us to isolate RNA of sufficient quality and yield to conduct studies of transcriptional level (see Materials and Methods section). Therefore, in this study we have investigated the pattern of COX-2 gene expression from the patients with colorectal adenomas, some of whom took NSAIDs, and others who did not. W e reported the results of an epidemiological analysis to address any possible effect of NSAIDs on the patients with colorectal adenomas. V-2. Materials and Methods Description of Parent Study Tumor blocks of paraffin-embedded tissue and data on selected risk factors were obtained from a case-control study of colorectal adenomatous polyps. A detailed description of this study can be found in a previous publication (Haile, Witte et al. 1997). In brief, cases and controls were selected from subjects who underwent a sigmoidoscopy in either of the two Southern California Kaiser Permanente Medical Centers (Bellflower and Sunset) during the period from January 1, 1991 to August 25, 1993. Cases were diagnosed for the first time with one or more histologically confirmed 72 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. colorectal adenomas. Controls had no history of polyps and were free of polyps at the sigmoidoscopy and were individually matched to cases by gender, age (within 5-year category), date of sigmoidoscopy (within 3-month category) and Kaiser Permanente Center. Cases and controls were 1) 50-74 years of age at the time of recruitment, 2) residents of Los Angeles or Orange County, and 3) fluent in English. The exclusion criteria were 1) having a history of invasive cancer, inflammatory bowel disease or familial polyposis, 2) having symptoms that suggested gastrointestinal disease, and 3) having physical or mental disability that would preclude an interview. During the accrual period, 628 cases and 689 controls who were potentially eligible were identified. Of these, 70 cases and 94 controls refused interview, and we were unable to contact 29 cases and 32 controls. Thus, interview data for 529 cases and 563 controls were obtained. The response rate (number interviewed/number eligible) was 84% among cases and 82% among controls. If the control initially matched to a case was not interviewed, a replacement control was identified. Among interviewed subjects, the indications for sigmoidoscopy were routine for 45% of cases and 44% of controls, referred due to specific minor symptoms for 16% of cases and 13% of controls and were not given for 39% of cases and 43% of controls. The average depth of penetration of the flexible sigmoidoscope was 55 cm for cases (standard deviation [SD] 11 cm) and 59 cm for controls (SD 5 cm). Fifteen cases had carcinoma in situ in 73 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. addition to adenomas. The size and number of polyps was indicated on a study form completed by the sigmoidoscopist. Participants provided data on smoking, therapeutic drug use, physical activity, height, weight, family history of cancer, and other factors during a 45 minute in-person interview that was administered in the subjects’ homes on average five months after sigmoidoscopy. The interviewer remained unaware of the participants’ case and control status for 70% of cases and 87% of controls. The study was approved by the University of Southern California and the Kaiser Permanente Institutional Review Boards. All subjects signed informed consents. Microdissection of Tissues All patient samples were received as paraffin-embedded tissue blocks. Due to specimen heterogeneity, it was essential to separate the areas of adenomatous polyps by microdissection, using a laser-capture microdissection instrument (PALM® MicroBeam [Positioning and Ablation with Laser Microbeams], Thornwood, NY). Each block was sectioned. One sectioned slide from each sample was stained with hematoxylin and eosin (H&E), and used for classification and guidance of microdissections. The rest of the slides from each sample were then stained with nuclear fast red for microdissections and RNA extraction. Figure 5.2 shows FFPE specimens before and after microdissection. 74 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 5.2. Stained FFPE adenoma specimens before and after LCM H&E staining AFTER 75 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. mRNA Isolation and cDNA Synthesis RNA isolation from paraffin-embedded specimens was done according to a proprietary procedure (US patent number 6,248,535). The isolated RNA was then dissolved in 50 pi of DEPC-treated water. Complementary DNA was synthesized as described in the synthesis of cDNA in the esophageal adenocarcinoma study, using 50 pi final volumes in the RT reaction. A thermocycler was also used with the same settings as described in the esophageal adenocarcinoma study (chapter three). Following reverse transcription, cDNA was stored at 4°C if polymerase chain reaction was not subsequently performed. Real-Time Polymerase Chain Reaction (PCR) Quantification Quantitation of cDNA was done using a fluorescence based real-time detection method (ABI PRISM 7900 Sequence Detection System [TaqMan®], Applied Biosystems, Foster City, CA) as described in chapter two. TaqM an® analyses yielded values that were expressed as ratios between two absolute measurements (gene of interest/internal reference gene). Schematic overall methodology is illustrated in Figure 5.3. The primer and probe sequences used for TaqMan® PCR are listed in Table 5.1. Statistical Analysis 76 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 5.3. Schem atic diagram for the colorectal adenoma study, using paraffin-em bedded tissues. FFPE tumor specimen micro-dissection Platea Exponential P T h resh o ld 28 30 — S a m p le * ----------------------- n f AM - H4 ■ □ PAM - H5 ■ □ FAM - H6 □ □ FAM - H7 □ □ FAM - H9 ■ □ FAM - H9 ■ □ FAM -H10 ■ B □ FAM - H11 ■ □ FAM - H12 ■ Use Threshold: I .006 I I Suacwst I c t S td Dev M u ll a Stdd»v:[ to o l * l .001 1 FAM - H2 19.138 0 .C0 I Omit Threshold:] 2.0 1 FAM - H3 21.579 0.001 1 S ta rt.| 3 | stop ) 15 | FAM - H5 20.252 0.001 * I Update Calculations 1 F AM H6 YUver: | &ftrt (B... i | Reporter' | FAM Z D P le a s e S e t th e T h re s h o ld V a lu e an A ll R e p o r te r L a y e rs . C lick QIC to C o n tin u e . ► RNA Isolation* I RNA Reverse Transcription I cDNA I PCR with TaqMan® Data Analysis 77 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 5.1. TaqM an® PCR primers and probes used for the colorectal adenom a study. Primer/Probe Sequences COX-1 Forward primer 5’- ATGATGGGCCTGCTGTTGA-3’ COX-1 Reverse primer 5’-CTCACCATGCCAAACC AG AA-3’ COX-1 TaqMan probe 6FAM 5’-CTGGCCTCAGCACTCTGGAATGACAA-3TAMRA COX-2 Forward primer 5 '-G CT CAAAC AT GAT GTTT G C ATTC-3’ COX-2 Reverse primer 5’-GCTGGCCCTCGCTTATGA-3’ COX-2 TaqMan probe 6FAM 5’-T GCCCAGCACTT CACGCAT CAGTT-3’TAM RA p-actin Forward primer 5’-TGAGCGCGGCTACAGCTT-3’ p-actin Reverse primer 5’-TCCTTAAT-GTCACGCACGAI I I -3’ p-actin TaqMan probe 6FAM5’-ACCACCACG-GCCGAGCG G-3TAMRA Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The p-values for differences in the mean value of CO X expression by selected subgroups were calculated with a Student’s t-test (Table 5.2). The odds ratios for the effects of NSAID variables were calculated with unconditional logistic regression, in order to make use of all available data on COX expression (Table 5.3). Due to the relatively small sample size for regression analyses, we were unable to run either a conditional regression or an unconditional regression with the matching factors as covariates; however, in all previous data analyses from this study, we have obtained very similar results from conditional and unconditional regression analyses; the matching factors do not make a difference (Haile, Witte et al. 1997). W henever a subject had more than one adenoma with a value for COX-1 or COX-2 expression, we simply took the average of those values so we could assign one value for each gene to that subject, to facilitate the planned analyses. V-3. Results Sections from 659 adenomas, derived from 349 cases, were considered for gene expression analyses. Among the samples, we were able to analyze 373 adenomas (56% ) from 172 cases. This success rate of 56% is comparable to the success rate of 57% (108 of 186 samples) reported by Einspahr (Einspahr, Krouse et al. 2002 AACR meeting abstract) in a similar study of colorectal adenomas sent to our lab for analysis. The 79 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. remaining 286 adenomas derived from 177 cases could not be analyzed primarily because of problems associated with the use of Bouin’s solution as the fixative or because the original adenoma was too small. The main variable of interest in this study, NSAID use (proportion of subjects who reported taking at least three pills per week), was not significantly different between the 172 cases on whom we had gene expression results, and the 177 on whom we could not obtain gene expression results. Among the 172 cases 107 cases were used for analysis in this study due to the reproducibility of data. The COX-2 and COX-1 gene expression values of those 65 cases were detectable but weren’t reproducible. Therefore, those samples were excluded from this study (again, the frequency of NSAID use was not substantially different in this group compared to the remaining cases). W e first investigated whether the levels of COX-1 and COX-2 expression, and the COX-2/COX-1 ratio, were associated with selected study variables. W e observed no significant or substantial differences in the levels of COX-1 or COX-2 expression, or the COX-2/COX-1 ratio between male and female subjects (e.g. P = 0.382 for the COX-2/COX-1 ratio), younger (age < 69 years) versus older (age > 69 years) subjects (P = 0.758 for the ratio), subjects with one versus greater than one adenoma (P = 0.67 for the ratio), subjects whose adenoma was in the rectum versus the colon (P = 0.966 for the ratio), or subjects whose adenoma was less than 10 mm versus 80 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. greater than or equal to 10 mm (P = 0.99 for the ratio) (Table 5.2). These results are very similar to the results reported by Einspahr for levels of COX- 2 expression (Einspahr, Krouse et al. 2002 AACR abstract; Einspahr, Krouse et al. 2002 AACR meeting abstract). There was no consistent pattern of association between COX-1 and the NSAID variables we studied, perhaps consistent with the belief that COX- 1 is constitutively expressed and not generally inducible (Table 5.3). Levels of COX-2 expression, however, appeared to be associated with NSAID use. For example, the mean expression value for COX-2 was 0.875 for subjects who reported taking at least three pills per week compared to a corresponding value of 1.263 for subjects who reported either no use of NSAIDs or taking less than three pills per week, and the mean expression value of COX-2 was 0.622 for subjects who took NSAIDs for greater than five years, suggesting that use of NSAIDs is associated with a relatively lower level of COX-2 gene expression. W e also calculated the COX-2/COX-1 ratio because this ratio has been used in several studies (Fujita, Matsui et al. 1998; Ohno, Yoshinaga et al. 2001) as a "COX-2 index" because COX-1 gene expression is expressed constitutively in both normal and colonic mucosa, and in tumor samples (Eberhart, Coffey et al. 1994; DuBois, Radhika et al. 1996). Thus, COX-1 provides a convenient reference gene for variations in CO X-2 and other genes of the prostaglandin synthesis pathway. The COX-2/COX-1 gene 81 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Table 5.2. Age, gender, polyp size, polyp location and number of polyps from participants with colorectal adenomas Age (Years) Gender Polyp size Polyp location No. of Polyps < 69 >6 9 Female Male < 10mm > 1 0 mm Rectal Left colon 1 > 2 n 85 22 38 69 59 48 29 78 84 23 Average COX- 2/COX-1 Ratio 2.39 (3.79) 2.19 (2.41) 2.86 (5.27) 2.07 (2.06) 2.35 (4.49) 2.35 (1.85) 2.37 (2.53) 2.35 (2.35) 2.43 (3.92) 2.08 (1.54) P-Value 0.758 0.382 0.999 0.966 0.483 Note: Standard deviations included in parentheses. cx» K > Table 5.3. COX-1, COX-2 and the C 0X -2/C 0X -1 ratio by NSAID use Averaae Cox-2/Cox-1 (Ratio) Average COX-2/COX-1 ratio Average COX-1 A verage CO X-2 n Mean Std. Dev. Mean Std. Dev. Mean Std. Dev. NSAID Use Reference Group 85 2.524 3.906 0.701 0.768 1.263 1.531 User (> 3 pills / week) 22 1.686 1.312 1.043 1.962 0.875 0.531 NSAID Duration Reference Group 85 2.524 3.906 0.701 0.768 1.263 1.531 < 2 years 10 1.706 1.318 1.458 2.874 0.994 0.555 2-5 years 8 1.872 1.434 0.585 0.394 0.888 0.676 > 5 years 3 1.341 1.588 0.753 0.367 0.622 0.376 ***NSAID user: subjects had taken NSAIDs during the year prior to their sigmoidoscopy (> 3 pills / week) ***NSAID duration: sum of years taking NSAIDs (including aspirin, prescription, & non prescription medications) 83 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. expression ratio was used to show association of COX-2 expression with tumor size and depth of invasion in colorectal cancer (Fujita, Matsui et al. 1998), and with depth of invasion in gastric carcinoma (Ohno, Yoshinaga et al. 2001). As shown in Table 5.3, we observed similar associations for NSAID use and the COX-2/COX-1 ratio, but not COX-1 expression. For example, the mean ratio among subjects who reported taking at least three NSAID pills per week was 1.686 compared to a mean of 2.524 in our reference group (non-users and those who used fewer than three NSAID pills per week), suggesting that increased use of NSAIDs is associated with reduced expression of COX-2 relative to COX-1. This also suggests that subjects who use NSAIDs for more than five years have a lower COX- 2/COX-1 ratio (1.341). Compared to the controls who had no history of polyps and were free of polyps at the time of sigmoidoscopy, the odds ratio for NSAID user (> 3 pills per week) with adenomas was 0.536 (95% Confidence Interval (Cl) = 0.325-0.885), and the odds ratio was 0.218 (95% Cl = 0.067-0.711) for subjects who took NSAIDs for more than five years, showing the protective effect of NSAID use. Several studies reported that tobacco is positively associated with the risk of colorectal adenoma and colorectal cancer (Saltz 2002). Therefore, we investigated whether there would be any correlation between smoking and the risk of adenoma, and our results showed as a risk factor (Table 5.4). 84 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Table 5.4. Odds ratios of patients with polyps, compared to controls who had no history of polyps and were free of polyps at time of sigmoidoscopy. n O dds Ratio All Subjects 95% Cl NSAID Use R eference G roup 85 1.000 User (> 3 Pills / Week) 22 0.536 0.325-0.885 NSAID Duration Reference Group 85 1.000 < 2 Years 10 0.764 0.375- 1.554 2-5 Years 8 0.680 0.312- 1.483 > 5 Years 3 0.218 0.067-0.711 Smoking Status No (Never) 37 1.000 Yes 70 1.410 0.915- 2.172 Smoking Status N ever 37 1.000 Current Smoker 26 2.677 1.509- 4.749 Ex-smoker 44 1.102 0.688- 1.765 *** NSAID user: subjects had taken NSAIDs during the year prior to their sigmoidoscopy (> 3 pills / week) *** NSAID duration: sum o f years taking NSAIDs (including aspirin, prescription, & non-prescription medications) one subject is missing *** Cl: confidence interval 00 cn V-4. Discussion By all of the measures of NSAID use that we investigated, it appeared that NSAID use was consistently associated with lower CO X-2 gene expression and a lower ratio of COX-2 relative to C O X -1. It is known that NSAIDs inhibit COX-1 and COX-2 enzyme activities (Taketo 1998a), and our findings explored whether the use of NSAIDs would also be associated with reduced gene expression of COX-2. The regulation of COX-2 expression by NSAIDs including aspirin has not been well known. It was recently reported that NSAIDs up-regulated COX-2 gene expression in small intestines of rats when they were administered orally by esophageal intubation into rats (Tanaka, Hase et al. 2002). Our results are, however, not consistent with this finding. The COX-2 promoter contains a TATA box and multiple transcription factor binding sites, such as nuclear factor-KB (NF-k B), the nuclear factor for interleukin 6 (NF-IL6; C/EBPp) and the cyclic AM P response element (CRE) binding protein (Yamamoto, Arakawa et al. 1995). COX-2 expression has also been demonstrated to be induced by various stimuli, such as tumor necrosis factor-a (TNF-a), cytokines (IL-1 p, IL-10, etc.), v-src oncogene, lipopolysaccharide (LPS), and platelet-derived growth factor (PDGF), through various signaling pathways (Yamamoto, Arakawa et al. 1995; Madrid, Wang et al. 2000; Shao, Sheng et al. 2000; Sheng, Shao et al. 2001; Monick, Robeffet al. 2002). 86 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Recent studies showed that aspirin dose-dependently suppressed the promoter activities of NF-k B, AP-1, and cyclic AMP response element (CRE), which are also the elements of the COX-2 promoter, indicating that aspirin may inhibit COX-2 gene expression by inactivating those elements in the COX-2 promoter (Frantz and O'Neill 1995). Another study showed that NS- 398 and indomethacin (NSAID) inhibited TN F-a or LPS-induced COX-2 gene expression through inactivation of NF-k B. Interestingly, in the absence of other COX-2 inducers (TNF-a and LPS in this study) NSAIDs induced COX-2 expression with the MAPKs or NF-KB-independent pathways (Paik, Ju et al. 2000). In summary, NSAIDs may play a role in reducing CO X-2 gene expression stimulated by tumor-associated inducers, including T N F -a and LPS, by inhibiting the signaling pathways (e.g. NF-k B pathway) activated by the inducers. In contrast, NSAIDs may also be associated with increasing COX-2 expression through other signal transduction pathways. These other pathways may be different from the signaling pathways stimulated by COX-2 inducers, in the absence of COX-2 inducers. Flowever, in the physiological environment promoting development of adenomatous polyps it is more likely that CO X-2 expression is induced by the various stimuli. Thus, our finding is concurrent with the hypothesis that NSAIDs are associated with the suppression of COX-2 induced by the various stimuli. To evaluate the effects of NSAIDs on adenomatous polyps, further studies are needed to investigate whether COX-2 expression in 87 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. adenomatous polyps is up-regulated. Other studies are also needed to elucidate whether COX-2 inhibitors reduce the incidence of adenomas through a COX-2 dependent signal transduction pathway, by suppressing COX-2 expression or through a COX-2 independent signal transduction pathway, since several studies showed both possibilities responsible for the preventive effect of NSAIDs in colorectal cancer (Chan 2002; Rex 2002). Studies are also needed to determine whether specific COX-2 inhibitors are more effective than nonspecific COX inhibitors, and whether COX-2 inhibitors provide benefits to patients with 1) sporadic polyps, and 2) hereditary non polyposis colorectal cancer (HNPCC), or 3) to patients with high expression of COX-2. These studies can provide better understanding of the adenoma- carcinoma sequence, and molecular-epidemiological studies such as the one we report here will be helpful to establish the mechanism of the protective effects of NSAIDs as well. Understanding the mechanism of action for the protective effect, and identifying subgroups of subjects most likely to benefit from this protective effect, can help doctors offer the best quality and most appropriate treatment to the patients in cancerous and pre-cancerous states, using chemotherapeutic and/or chemopreventive agents. 8 8 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER VI REFERENCES Achiwa, H., Y. Yatabe, et al. (1999). "Prognostic significance of elevated cyclooxygenase 2 expression in primary, resected lung adenocarcinomas." Clin Cancer Res 5(5): 1001-5. Baron, J. A. (1995). "Aspirin and cancer." Prev Med 24(2): 121-4. Barrett, M. T., C. A. Sanchez, et al. (1999). "Evolution of neoplastic cell lineages in Barrett oesophagus." Nat Genet 22(1): 106-9. Bianchi, R., U. Grohmann, et al. (1996). "IL-12 is both required and sufficient for initiating T cell reactivity to a class l-restricted tumor peptide (P815AB) following transfer of P815AB-pulsed dendritic cells." J Immunol 157(4): 1589- 97. Bieche, I., M. Olivi, et al. (1998). "Novel approach to quantitative polymerase chain reaction using real-time detection: application to the detection of gene amplification in breast cancer." Int J Cancer 78(5): 661-6. Blot, W . J. and J. K. McLaughlin (1999). "The changing epidemiology of esophageal cancer." Semin Oncol 26(5 Suppl 15): 2-8. Buskens, C. J., B. P. Van Rees, et al. (2002). "Prognostic significance of elevated cyclooxygenase 2 expression in patients with adenocarcinoma of the esophagus." Gastroenterology 122(7): 1800-7. Cameron, A. J. (1994). "Epidemiology of Barrett's esophagus." Gastroenterol Clin Biol 18(1 Pt2): D3-4. Chan, T. A. (2002). "Nonsteroidal anti-inflammatory drugs, apoptosis, and colon-cancer chemoprevention." Lancet Oncol 3(3): 166-74. Chappie, K. S., N. Scott, et al. (2002). "Analysis of cyclooxygenase expression in human colorectal adenomas." Pis Colon Rectum 45(10): 1316- 24. 89 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Cianchi, F., C. Cortesini, et al. (2001). "Up-regulation of cyclooxygenase 2 gene expression correlates with tumor angiogenesis in human colorectal cancer." Gastroenterology 121(6): 1339-47. Colombo, M. P., M. Vagliani, et al. (1996). "Amount of interleukin 12 available at the tumor site is critical for tumor regression." Cancer Res 56(11): 2531-4. Dixon, D. A., C. D. Kaplan, et al. (2000). "Post-transcriptional control of cyclooxygenase-2 gene expression. The role of the 3'-untranslated region." J BioLChem 275(16): 11750-7. Dixon, M. F. (2000). "Patterns of inflammation linked to ulcer disease." Baillieres Best Pract Res Clin Gastroenterol 14(1): 27-40. DuBois, R. N., A. Radhika, et al. (1996). "Increased cyclooxygenase-2 levels in carcinogen-induced rat colonic tumors." Gastroenterology 110(4): 1259-62. DuBois, R. N., J. Shao, et al. (1996). "G1 delay in cells overexpressing prostaglandin endoperoxide synthase-2." Cancer Res 56(4): 733-7. DuBois, R. N., M. Tsujii, et al. (1994). "Cloning and characterization of a growth factor-inducible cyclooxygenase gene from rat intestinal epithelial cells." Am J Physiol 266(5 Pt 1): G822-7. Dunlop, M. G. (1997). "Colorectal cancer." Bmj 314(7098): 1882-5. Eberhart, C. E., R. J. Coffey, et al. (1994). "Up-regulation of cyclooxygenase 2 gene expression in human colorectal adenomas and adenocarcinomas." Gastroenterology 107(4): 1183-8. Einspahr, J., R. Krouse, et al. (2002 AACR abstract). "Association Between COX-2 Expression, Colorectal Adenoma Characteristics, and Adenoma Recurrence." AACR 2002 A bstract #616. Einspahr, J., R. Krouse, et al. (2002 AACR meeting abstract). "Association Between COX-2 Expression, Colorectal Adenoma Characteristics, and Adenoma Recurrence." AACR meeting 2002 A bstract #616. 90 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Feng, L., W. Sun, et al. (1993). "Cloning two isoforms of rat cyclooxygenase: differential regulation of their expression." Arch Biochem Biophys 307(2): 361-8. Ferrario, A., K. Von Tiehl, et al. (2002). "Cyclooxygenase-2 inhibitor treatment enhances photodynamic therapy-mediated tumor response." Cancer Res 62(14): 3956-61. Frantz, B. and E. A. O'Neill (1995). "The effect of sodium salicylate and aspirin on NF-kappa B." Science 270(5244): 2017-9. Fujita, T., M. Matsui, et al. (1998). "Size- and invasion-dependent increase in cyclooxygenase 2 levels in human colorectal carcinomas." Cancer Res 58(21): 4823-6. Funkhouser, E. M. and G. B. Sharp (1995). "Aspirin and reduced risk of esophageal carcinoma." Cancer 76(7): 1116-9. Giardiello, F. M., S. R. Hamilton, et al. (1993). "Treatment of colonic and rectal adenomas with sulindac in familial adenomatous polyposis." N Engl J Med 328(18): 1313-6. Gibson, U. E., C. A. Heid, et al. (1996). "A novel method for real time quantitative RT-PCR." Genome Res 6(10): 995-1001. Ginsberg, R. J., Vokes, E.E., Raben, A. (1997). "Non-small cell lung cancer." Philadelphia:Lipincott-Raven(DeVita V T Jr, Hellmann S. Rosenberg SA. eds. Cancer: principles in practice of oncology) 5th ed: 858-910. Giovannucci, E., K. M. Egan, et al. (1995). "Aspirin and the risk of colorectal cancer in women." N Engl J Med 333(10): 609-14. Giulietti, A., L. Overbergh, et al. (2001). "An overview of real-time quantitative PCR: applications to quantify cytokine gene expression." Methods 25(4): 386-401. Greenberg, E. R., J. A. Baron, et al. (1993). "Reduced risk of large-bowel adenomas among aspirin users. The Polyp Prevention Study Group." J Natl Cancer Inst 85(11): 912-6. 91 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Greenlee, R. T., T. Murray, et al. (2000). "Cancer statistics, 2000." CA Cancer J Clin 50(1): 7-33. Haile, R. W ., J. S. Witte, et al. (1997). "A sigmoidoscopy-based case-control study of polyps: macronutrients, fiber and meat consumption." Int J Cancer 73(4): 497-502. Halpern, J. (1982). "Maximally selected chi-square statistics for small samples." Biometrics 38: 1017-1023. Hamasaki, Y., J. Kitzler, et al. (1993). "Phorbol ester and epidermal growth factor enhance the expression of two inducible prostaglandin H synthase genes in rat tracheal epithelial cells." Arch Biochem Biophvs 304(1): 226-34. Heid, C. A., J. Stevens, et al. (1996). "Real time quantitative PCR." Genome Res 6(10): 986-94. Hempel, S. L., M. M. Monick, et al. (1994). "Lipopolysaccharide induces prostaglandin H synthase-2 protein and mRNA in human alveolar macrophages and blood monocytes." J Clin Invest 93(1): 391-6. Hida, T., K. Kozaki, et al. (2000). "Cyclooxygenase-2 inhibitor induces apoptosis and enhances cytotoxicity of various anticancer agents in non small cell lung cancer cell lines." Clin Cancer Res 6(5): 2006-11. Hida, T., J. Leyton, et al. (1998a). "Non-small cell lung cancer cycloxygenase activity and proliferation are inhibited by non-steroidal antiinflammatory drugs." Anticancer Res 18(2A): 775-82. Hida, T., Y. Yatabe, et al. (1998b). "Increased expression of cyclooxygenase 2 occurs frequently in human lung cancers, specifically in adenocarcinomas." Cancer Res 58(17): 3761-4. Hla, T. and K. Neilson (1992). "Human cyclooxygenase-2 cDNA." Proc Natl Acad Sci U S A 89(16): 7384-8. Huang, M., M. Stolina, et al. (1998). "Non-small cell lung cancer cyclooxygenase-2-dependent regulation of cytokine balance in lymphocytes and macrophages: up-regulation of interleukin 10 and down-regulation of interleukin 12 production." Cancer Res 58(6): 1208-16. 92 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Hwang, D., D. Scollard, et al. (1998). "Expression of cyclooxygenase-1 and cyclooxygenase-2 in human breast cancer." J Natl Cancer Inst 90(6): 455-60. Jones, D. A., D. P. Carlton, et al. (1993). "Molecular cloning of human prostaglandin endoperoxide synthase type II and demonstration of expression in response to cytokines." J Biol Chem 268(12): 9049-54. Kang, R. Y., J. Freire-Moar, et al. (1996). "Expression of cyclooxygenase-2 in human and an animal model of rheumatoid arthritis." Br J Rheumatol 35(8): 711-8. Kargman, S. L., G. P. O'Neill, et al. (1995). "Expression of prostaglandin G/H synthase-1 and -2 protein in human colon cancer." Cancer Res 55(12): 2556- 9. Kennedy, B. P., C. C. Chan, et al. (1993). "Cloning and expression of rat prostaglandin endoperoxide synthase (cyclooxygenase)-2 cDNA." Biochem Biophys Res Commun 197(2): 494-500. Kim, J., R. L. Modlin, et al. (1995). "IL-10 production in cutaneous basal and squamous cell carcinomas. A mechanism for evading the local T cell immune response." J Immunol 155(4): 2240-7. Koga, H., S. Sakisaka, et al. (1999). "Expression of cyclooxygenase-2 in human hepatocellular carcinoma: relevance to tumor dedifferentiation." Hepatoloav 29(3): 688-96. Lagergren, J., R. Bergstrom, et al. (1999). "Symptomatic gastroesophageal reflux as a risk factor for esophageal adenocarcinoma." N Engl J Med 340(11): 825-31. Liu, C. H., S. H. Chang, et al. (2001). "Overexpression of cyclooxygenase-2 is sufficient to induce tumorigenesis in transgenic mice." J Biol Chem 276(21): 18563-9. Madrid, L. V., C. Y. Wang, et al. (2000). "Akt suppresses apoptosis by stimulating the transactivation potential of the RelA/p65 subunit of NF- kappaB." Mol Cell Biol 20f5): 1626-38. 93 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Marrogi, A. J., W. D. Travis, et al. (2000). "Nitric oxide synthase, cyclooxygenase 2, and vascular endothelial growth factor in the angiogenesis of non-small cell lung carcinoma." Clin Cancer Res 6(12): 4739-44. Masferrer, J. L., K. M. Leahy, et al. (2000). "Antiangiogenic and antitumor activities of cyclooxygenase-2 inhibitors." Cancer Res 60(5): 1306-11. Matsuura, H., M. Sakaue, et al. (1999). "Regulation of cyclooxygenase-2 by interferon gamma and transforming growth factor alpha in normal human epidermal keratinocytes and squamous carcinoma cells. Role of mitogen- activated protein kinases." J Biol Chem 274(41): 29138-48. Miller, R., Siegmund, D. (1982). " Maximally selected chi-square statistics." Biometrics 38: 1011-1016. Monick, M. M., P. K. Robeff, et al. (2002). "Phosphatidylinositol 3-kinase activity negatively regulates stability of cyclooxygenase 2 mRNA." J Biol Chem 277(36): 32992-3000. Morris, C. D., G. R. Armstrong, et al. (2001). "Cyclooxygenase-2 expression in the Barrett's metaplasia-dysplasia-adenocarcinoma sequence." Am J Gastroenterol 96(4): 990-6. Morson, B. (1974). "President's address. The polyp-cancer sequence in the large bowel." Proc R Soc Med 67(6): 451-7. Mountain, C. F. (1997). "Revisions in the International System for Staging Lung Cancer." Chest 111(6): 1710-7. Ochiai, M., T. Oguri, et al. (1999). "Cyclooxygenase-2 (CO X-2) mRNA expression levels in normal lung tissues and non-small cell lung cancers." Jpn J Cancer Res 90(12): 1338-43. Ohno, R., K. Yoshinaga, et al. (2001). "Depth of invasion parallels increased cyclooxygenase-2 levels in patients with gastric carcinoma." Cancer 91(10): 1876-81. Okami, J., H. Yamamoto, et al. (1999). "Overexpression of cyclooxygenase-2 in carcinoma of the pancreas." Clin Cancer Res 5(8): 2018-24. 94 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Oshima, M., J. E. Dinchuk, et al. (1996). "Suppression of intestinal polyposis in Ape delta716 knockout mice by inhibition of cyclooxygenase 2 (COX-2)." Cell 87(5): 803-9. Owen, D. (1996). "Large intestine and anus." 10th ed.: 1741-78. Ozes, O. N., L. D. Mayo, et al. (1999). "NF-kappaB activation by tumour necrosis factor requires the Akt serine-threonine kinase." Nature 401(6748): 82-5. Paik, J. H., J. H. Ju, et al. (2000). "Two opposing effects of non-steroidal anti inflammatory drugs on the expression of the inducible cyclooxygenase. Mediation through different signaling pathways." J Biol Chem 275(36): 28173-9. Pera, M., A. J. Cameron, et al. (1993). "Increasing incidence of adenocarcinoma of the esophagus and esophagogastric junction." Gastroenterology 104(2): 510-3. Pike, M. C. (1972). "Asymptomatically efficient rank invariant procedures." J R Stat Soc Series A 135: 201-203. Prescott, S. M. and R. L. White (1996). "Self-promotion? Intimate connections between APC and prostaglandin H synthase-2." Cell 87(5): 783- 6. Qin, Z., G. Noffz, et al. (1997). "Interleukin-10 prevents dendritic cell accumulation and vaccination with granulocyte-macrophage colony- stimulating factor gene-modified tumor cells." J Immunol 159(2): 770-6. Rex, D. K. (2002). "Screening for colon cancer and evaluation of chemoprevention with coxibs." J Pain Symptom Manage 23(4 Suppl): S41-50. Ristimaki, A., N. Honkanen, et al. (1997). "Expression of cyclooxygenase-2 in human gastric carcinoma." Cancer Res 57(7): 1276-80. Saltz, L. (2002). "Colorectal Cancer:Multimodality Management." Sano, H., Y. Kawahito, et al. (1995). "Expression of cyclooxygenase-1 and -2 in human colorectal cancer." Cancer Res 55(17): 3785-9. 95 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Sawaoka, H., S. Kawano, et al. (1998). "Helicobacter pylori infection induces cyclooxygenase-2 expression in human gastric mucosa." Prostaglandins Leukot Essent Fatty Acids 59(5): 313-6. Schneider, P. M., H. W. Praeuer, et al. (2000). "Multiple molecular marker testing (p53, C-Ki-ras, c-erbB-2) improves estimation of prognosis in potentially curative resected non-small cell lung cancer." Br J Cancer 83(4): 473-9. Schreinemachers, D. M. and R. B. Everson (1994). "Aspirin use and lung, colon, and breast cancer incidence in a prospective study." Epidemiology 5(2): 138-46. Shao, J., H. Sheng, etal. (2000). "Regulation of constitutive cyclooxygenase- 2 expression in colon carcinoma cells." J Biol Chem 275(43): 33951-6. Sheng, H., J. Shao, et al. (2000). "Transforming growth factor-betal enhances Ha-ras-induced expression of cyclooxygenase-2 in intestinal epithelial cells via stabilization of mRNA." J Biol Chem 275(9): 6628-35. Sheng, H., J. Shao, et al. (2001). "K-Ras-mediated increase in cyclooxygenase 2 mRNA stability involves activation of the protein kinase B1." Cancer Res 61(6): 2670-5. Sheng, H., J. Shao, et al. (1997). "Inhibition of human colon cancer cell growth by selective inhibition of cyclooxygenase-2." J Clin Invest 99(9): 2254-9. Sheng, H., J. Shao, et al. (1998). "Modulation of apoptosis and Bcl-2 expression by prostaglandin E2 in human colon cancer cells." Cancer Res 58(2): 362-6. Sheng, H., J. Shao, et al. (2001). "Prostaglandin E2 increases growth and motility of colorectal carcinoma cells." J Biol Chem 276(21): 18075-81. Shirvani, V. N., R. Ouatu-Lascar, et al. (2000). "Cyclooxygenase 2 expression in Barrett's esophagus and adenocarcinoma: Ex vivo induction by bile salts and acid exposure." Gastroenterology 118(3): 487-96. 96 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Singer, II, D. W . Kawka, et al. (1998). "Cyclooxygenase 2 is induced in colonic epithelial cells in inflammatory bowel disease." Gastroenterology 115(2): 297-306. Skacel, M., R. E. Petras, et al. (2000). "The diagnosis of low-grade dysplasia in Barrett's esophagus and its implications for disease progression." Am J Gastroenterol 95(12): 3383-7. Smalley, W. E. and R. N. DuBois (1997). "Colorectal cancer and nonsteroidal anti-inflammatory drugs." Adv Pharmacol 39: 1-20. Smith, A. M., C. A. Maxwell-Armstrong, et al. (1999). "Surveillance for Barrett's oesophagus in the UK." Br J Sura 86(2): 276-80. Smith, W . L., R. M. Garavito, et al. (1996). "Prostaglandin endoperoxide H synthases (cyclooxygenases)-l and -2." J Biol Chem 271(52): 33157-60. Smith, W. L., E. A. Meade, et al. (1994). "Pharmacology of prostaglandin endoperoxide synthase isozymes-1 and -2." Ann N Y Acad Sci 714: 136-42. Souza, R. F., K. Shewmake, et al. (2000). "Selective inhibition of cyclooxygenase-2 suppresses growth and induces apoptosis in human esophageal adenocarcinoma cells." Cancer Res 60(20): 5767-72. Stolina, M., S. Sharma, et al. (2000). "Specific inhibition of cyclooxygenase 2 restores antitumor reactivity by altering the balance of IL-10 and IL-12 synthesis." J Immunol 164(1): 361-70. Taketo, M. M. (1998a). "Cyclooxygenase-2 inhibitors in tumorigenesis (part I)." J Natl Cancer Inst 90(20): 1529-36. Taketo, M. M. (1998b). "Cyclooxygenase-2 inhibitors in tumorigenesis (Part ID." J Natl Cancer Inst 90(21): 1609-20. Tanaka, A., S. Hase, et al. (2002). "Up-regulation of cyclooxygenase-2 by inhibition of cyclooxygenase-1: a key to nonsteroidal anti-inflammatory drug- induced intestinal damage." J Pharmacol Exp Ther 300(3): 754-61. Thun, M. J. (1996). "NSAID use and decreased risk of gastrointestinal cancers." Gastroenterol Clin North Am 25(2): 333-48. 97 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Thun, M. J., M. M. Namboodiri, et al. (1993). "Aspirin use and risk of fatal cancer." Cancer Res 53(6): 1322-7. Tsujii, M. and R. N. DuBois (1995). "Alterations in cellular adhesion and apoptosis in epithelial cells overexpressing prostaglandin endoperoxide synthase 2." Cejl 83(3): 493-501. Tsujii, M., S. Kawano, et al. (1997). "Cyclooxygenase-2 expression in human colon cancer cells increases metastatic potential." Proc Natl Acad Sci U S A 94(7): 3336-40. Tsujii, M., S. Kawano, et al. (1998). "Cyclooxygenase regulates angiogenesis induced by colon cancer cells." Cell 93(5): 705-16. Tucker, O. N., A. J. Dannenberg, et al. (1999). "Cyclooxygenase-2 expression is up-regulated in human pancreatic cancer." Cancer Res 59(5): 987-90. Vane, J. R., Y. S. Bakhle, et al. (1998). Cyclooxygenases 1 and 2. Annu Rev Pharmacol Toxicol. 38: 97-120. Vane, J. R. and R. M. Botting (1997). Mechanism of action of aspirin-like drugs. Semin Arthritis Rheum. 26: 2-10. Williams, C. S. and R. N. DuBois (1996). "Prostaglandin endoperoxide synthase: why two isoforms?" Am J Physiol 270(3 Pt 1): G393-400. Williams, C. S., M. Tsujii, et al. (2000). "Host cyclooxygenase-2 modulates carcinoma growth." J Clin Invest 105(11): 1589-94. Willis, D. B. and N. H. Colburn (2002). "Molecular targets for cancer prevention: a meeting review of the third American Cancer Society-Schilling Research Conference." Cancer Epidemiol Biomarkers Prev 11(10 Pt 1): 972- 8. Wilson, K. T., S. Fu, et al. (1998). "Increased expression of inducible nitric oxide synthase and cyclooxygenase-2 in Barrett's esophagus and associated adenocarcinomas." Cancer Res 58(14): 2929-34. Wolff, H., K. Saukkonen, et al. (1998). "Expression of cyclooxygenase-2 in human lung carcinoma." Cancer Res 58(22): 4997-5001. 98 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Xie, W . and H. R. Herschman (1995). "v-src induces prostaglandin synthase 2 gene expression by activation of the c-Jun N-terminal kinase and the c-Jun transcription factor." J Biol Chem 270(46): 27622-8. Xie, W . and H. R. Herschman (1996). "Transcriptional regulation of prostaglandin synthase 2 gene expression by platelet-derived growth factor and serum." J Biol Chem 271(49): 31742-8. Yamamoto, K., T. Arakawa, et al. (1995). "Transcriptional roles of nuclear factor kappa B and nuclear factor-interleukin-6 in the tumor necrosis factor alpha-dependent induction of cyclooxygenase-2 in MC3T3-E1 cells." J Biol Chem 270(52): 31315-20. Zhang, F., K. Subbaramaiah, et al. (1998). "Dihydroxy bile acids activate the transcription of cyclooxygenase-2." J Biol Chem 273(4): 2424-8. Zhuang, Z., A. O. Vortmeyer, et al. (1996). "Barrett's esophagus: metaplastic cells with loss of heterozygosity at the APC gene locus are clonal precursors to invasive adenocarcinoma." Cancer Res 56(9): 1961-4. Zimmermann, K. C., M. Sarbia, et al. (1999). "Cyclooxygenase-2 expression in human esophageal carcinoma." Cancer Res 59(1): 198-204. 99 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

EPR analysis of soluble and membrane -bound annexin A2

PDF

Development of novel adenovirus-retroelement vectors and their applications beyond gene therapy

PDF

An E2F-independent mechanism in transcriptional control of the hamster histone H3.2 gene expression

PDF

A novel candidate ovarian cancer and melanoma tumor suppressor gene on chromosome 6q24.3

PDF

Development of HA -pseudotyped retroviral vectors for cell -specific gene delivery

PDF

Biochemical factors determining tumor response to 5,10-dideazatetrahydrofolate, a folate antimetabolite inhibitory to de novo purine biosynthesis

PDF

Characterization of target cell entry by murine leukemia viruses

PDF

Identifying hematopoietic stem cell proliferation genes for bone marrow transplantation

PDF

Characterizing the function of murine epididymal secretory protein 1 (ME1) in hematopoietic stem cells

PDF

Characterization of hematopoietic stem cell capabilities and properties

PDF

Characterization of zebularine: A novel inhibitor of DNA methylation with clinical potential

PDF

Interactions between vesicular stomatitis virus G protein and the cytoskeletal factor protein 4.1

PDF

Investigation of the role of epigenetic modification of DNA and chromatin in aberrant gene silencing in cancer cells

PDF

Improving retroviral vectors for in utero gene therapy of severe combined immune deficiencies

PDF

Different roles of p160 coregulators in myogenesis

PDF

CYP17 polymorphism and risk for colorectal adenomas

PDF

Inhibition of cancer invasion and metastasis: Mechanistic analysis of contortrostatin function at the molecular and cellular levels

PDF

A coactivator complex among GRIP1, CARM1, and TIF1alpha contributes to gene activation directed by androgen receptor

PDF

Identification of the biochemical pathways affected by the anticancer agents Motexafin Gadolinium and Sapphyrin through gene expression profiling

PDF

A tale of two tails: Role of histone methyltransferases in transcriptional regulation

Asset Metadata

Creator Yochim, Ji Min (author)

Core Title Cyclooxygenase-2 as a candidate biomarker for progression, prognosis and chemoprevention in various tumor types

School Graduate School

Degree Doctor of Philosophy

Degree Program Biochemistry and Molecular Biology

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag biology, biostatistics,biology, genetics,biology, molecular,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Danenberg, Peter (committee chair), Anderson, W. French (committee member), Haile, Robert W. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c16-282805

Unique identifier UC11334997

Identifier 3103985.pdf (filename),usctheses-c16-282805 (legacy record id)

Legacy Identifier 3103985.pdf

Dmrecord 282805

Document Type Dissertation

Rights Yochim, Ji Min

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA