STUDIES OF RETINOIC ACID SIGNALLING
IN PANCREATIC CANCER
Davendra Segara.
A thesis for the degree of Doctor of Philosophy
Faculty of Medicine University of New South Wales
Cancer Research Program Garvan Institute of Medical Research
February, 2006 PLEASE TYPE THE UNIVERSITY OF NEW SOUTH WALES Thesis/Dissertation Sheet Segara Surname or Family name:
First name: Davendra Other name/s: Ph.D. Abbreviation for degree as given in the University calendar:
School: St-Vincent’s Hospital Clinical School Faculty: Medicine
Title: Studies of Retinoic Acid Signalling in Pancreatic Cancer
Abstract 350 words maximum: (PLEASE TYPE) Pancreatic cancer (PC) is the fourth leading cause of cancer death in Western societies. Despite significant progress in understanding the molecular pathology of PC and its precursor lesion: pancreatic intraepithelial neoplasia (PanIN), there remain no molecules with proven clinical utility. Affymetrix Genechipfi oligonucleotide microarrays were used to interrogate mRNA expression of PC and normal pancreas to identify molecular pathways dysregulated in PC. Analysis of these data identified altered expression of numerous components of the S100 Calcium Binding Protein Family, Retinoic Acid signalling pathway and the HOX transcriptional network in PC compared to normal pancreas. These pathways were assessed using immunohistochemistry (IHC) and in-situ hybridisation (ISH) in a cohort of patients with PC. Increased protein expression, of S100A2, S100A6 and S100P was observed in 43%, 60% and 48% of PC respectively. Expression of S100A2 was associated with a poor outcome (p = 0.009), whilst increased expression of S100A6 (p = 0.0008) and S100P (p = 0.0005) were associated with an improved outcome. Additionally, S100A2 expression was identified as an independent marker of outcome in resected tumours. Aberrant expression of retinoic acid signalling components was demonstrated in PC cell lines using semi-quantitative RT-PCR. ISH demonstrated expression of Retinoic Acid Induced 3 (RAI3), an orphan G protein coupled receptor normally expressed in the fetal lung, in 68% of PC, and this co-segregated with an improved overall survival (p = 0.026).Ectopic protein expression of HOXB2, a transcription factor normally expressed in the developing hindbrain and modulated by retinoic acid, was observed in 15% of early PanIN lesions and 38% of PC specimens. Expression of HOXB2 was associated with non-resectable tumours and was an independent predictor of poor survival in resected tumours. Suppression of HOXB2 protein expression using small interfering RNA, resulted in epithelioid trans-differentiation in the Panc-1 PC cell line, however no alteration in proliferation rates were observed compared to controls. This thesis has shown that transcript profiling and tissue validation has identified potential markers of early diagnosis and outcome in PC. Furthermore, pathways and molecules previously thought to be associated with normal human development have been implicated to play a role in the development and progression of PC. Further analyses of these markers will determine any potential role in future diagnostic and therapeutic strategies.
Declaration relating to disposition of project thesis/dissertation
I hereby grant to the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or in part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all property rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation.
I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstracts International (this is applicable to doctoral theses only).
…………………………………………………………… ……………………………………..……………… ……….……………………...…….… Signature Witness Date
The University recognises that there may be exceptional circumstances requiring restrictions on copying or conditions on use. Requests for restriction for a period of up to 2 years must be made in writing to the Registrar. Requests for a longer period of restriction may be considered in exceptional circumstances if accompanied by a lette r of support from the Supervisor or Head of School. Such requests must be submitted with the thesis/dissertation.
FOR OFFICE USE ONLY Date of completion of requirements for Award:
Registrar and Deputy Principal
THIS SHEET IS TO BE GLUED TO THE INSIDE FRONT COVER OF THE THESIS
N:\FLORENCE\ABSTRACT Certificate of originality
“I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked in UNSW or elsewhere, is explicitly acknowledged in this thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project's design and conception or in style, presentation and linguistic expression is acknowledged.”
......
Davendra Segara
February 2006 TABLE OF CONTENTS:
ACKNOWLEDGEMENTS……………………………………………………………...i ABBBREVIATIONS…………………………………………………………………...iii ABSTRACT…………………………………………………………………………….iv PUBLICATIONS, ABSTRACTS AND PATENT……………………………………...v THESIS OVERVIEW………………………………………………………………….vii
Chapter 1 INTRODUCTION ...... 1 Clinicopathological aspects of Pancreatic Cancer ...... 2 Risk Factors...... 4 Pathology...... 8 Diagnosis and Treatment ...... 9 Non Surgical Therapeutic Options...... 12 Prognostic Factors...... 15 Molecular pathology of pancreatic cancer...... 18 Cell cycle regulation and pancreatic cancer ...... 18 Cell signalling in pancreatic cancer ...... 19 TGF- Signalling in PC...... 20 S100 Calcium Binding Proteins ...... 20 Developmental pathways and pancreatic cancer...... 21 Pancreatic Intra-Epithelial Neoplasia (PanIN)...... 26 Summary...... 27
Chapter 2 METHODS ...... 31 Transcript Profiling...... 32 Organisation of data...... 35 Patient Characteristics...... 38 Immunohistochemistry ...... 39 In-situ Hvbridisation ...... 41 Immunohistochemical and In-situ hybridisation scoring ...... 43 Statistical evaluation ...... 45 General tissue culture ...... 46 Western Blotting ...... 50 Cell proliferation assays ...... 52 Chapter 3 IDENTIFICATION of NOVEL GENES in the DEVELOPMENT and PROGRESSION of PANCREATIC CANCER by TRANSCRIPT PROFILING ...... 53 Introduction ...... 54 Validation ...... 56 Results...... 58 The TGF- signalling pathway ...... 60 Wnt Signalling...... 63 S100 Calcium Binding Proteins ...... 68 Retinoic Acid Signalling ...... 70 Discussion...... 74 Conclusion...... 74
Chapter 4 S100 CALCIUM BINDING PROTEINS and OUTCOME in PANCREATIC CANCER ...... 75 Introduction ...... 76 S100A2, S100A6 and S100P Expression and Outcome in PC...... 80 The PC cohort ...... 80 The Whole Cohort ...... 82 The Resected Cohort ...... 84 Discussion...... 88 Conclusion...... 91
Chapter 5 RETINOIC ACID SIGNALLING IN PC...... 92 Introduction ...... 93 Aberrant Retinoic Acid Signalling in PC ...... 97 Cohort Analysis...... 101 Whole Cohort...... 101 Resected Cohort...... 104 Discussion...... 105 Conclusion...... 106 Chapter 6 THE HOMEOBOX TRANSCRIPTION FACTOR: HOXB2 in PANCREATIC CANCER...... 107 Introduction ...... 108 Cohort characteristics ...... 110 The Whole Cohort ...... 110 Resected Cohort...... 116 HOXB2 Expression In Pancreatic Cancer Cell Lines...... 119 HOXB2 mRNA Expression in PC Cell Lines...... 119 HOXB2 protein expression in PC cell lines...... 119 Knockdown of HOXB2 expression in Panc-1 cells...... 121 Proliferation of Panc-1 cells after decreasing expression of HOXB2...... 125
Chapter 7 SUMMARY, GENERAL DISCUSSION and FUTURE DIRECTIONS...... 129 Key issues in the current management of PC...... 130 Identification of Markers of Early Diagnosis in PC...... 133 Identification of Novel Markers of Prognosis and Response...... 135 Identification of Novel Therapeutic Strategies in PC...... 137 Concluding remarks and future directions ...... 138
REFERENCES ...... 140
APPENDICES ...... 163 i
Acknowledgements
The privilege of undertaking PhD studies after surgical training required a great deal of support to allow a scientific novice to maintain some inertia in the alien environment of the laboratory
There are several extraordinary people whose counsel, have allowed these studies to come to fruition.
Professor Rob Sutherland, the Director of the Cancer Research Program at the Garvan Institute of Medical Research who allowed me to undertake these Ph.D. studies 4 years ago. His support, patience and guidance have provided the stable platform on which these studies could be initiated and completed. Dr Sue Henshall, my supervisor, whose support of this work patience and constant helpful advice even whilst overseas were invaluable. Dr Andrew Biankin my mentor, friend and a modern day renaissance man, his counsel on all things scientific, surgical and practical has been a source of great intellectual and practical support during these studies. There have been many in the Translational group who have made this study a joy, James Kench whose patience and understanding in teaching basic histopathology to a novice has been exemplary. Sarah Eggleton whose tireless efforts and patience in IHC and ISH staining, have allowed smooth progress in these studies. Vanessa Hayes whose instruction on probe design and patience in the face of frequent questions was much appreciated. The “comrades” in Lab 2, some of whom have left, including Kris Rasiah, Rhonda Kwong Larry Kalish, Lisa Horvath, Viola Heinzman Schwarz, Catriona McNeil, Emma Padilla and Jayne Lelliot for academic, practical and social advice during these studies.
Also to a big thank-you to all members of the Cancer group who have exercised polite restraint in answering the many scientifically ignorant and ultimately mirthful enquires during this work. ii
I would like to thank Mr Max Coleman for his enthusiastic support for my research and clinical endeavours. Mr Jim Touli for his support for the project and surgical research in general and The Australian Pancreatic Club whose endeavours in enhancing pancreatic research in Australasia by providing a forum to communicate these findings have now become a yearly and much anticipated occurrence.
There were various funding bodies that have generously supported this research including, The National Health and Medical Research Council, The Royal Australasian College of Surgeons and the St Vincent’s Clinic Foundation.
Of a more personal note, these studies would not be possible without my support team outside the clinical and laboratory environment.
I would like to especially thank, Janet, who during the time of this doctoral thesis has supported me unquestioningly. She has been a constant and unwavering source of, practical, emotional and intellectual input despite her rigorous training schedule over the last 4 years. More importantly her confidence in me, and unconditional love are one of the pillars of that have made this thesis possible.
To my brother Reuben, who keeps on “believing” in his big brother and who has provided a great sounding board during the “trials and tribulations” of the thesis. His brotherly support, advice and understanding, has aided in the completion of this work.
In 1976 my parents decided to leave Malaysia for the unknown possibilities that migration to Australia would bring. In doing so they left supportive family, career advancement and economic stability, because of the restricted educational opportunities available for their two boys. All of my achievements, including this thesis owe much to that selfless decision and the continued sacrifices made by loving parents who endeavour to provide every educational and life opportunity for their children. As a result this thesis is as much an achievement of my parents as it is for me. Pa and Ma, I will forever be in your debt.
Dave iii
Abbreviations ABC avidin-biotin complex CDK cyclin dependent kinase CT computed tomography cDNA complementary DNA cRNA complementary RNA DAB 3,3'-diaminobenzidine DNA deoxyribonucleic acid dsDNA double strand DNA ECL enhanced chemiluminescence EDTA ethylenediaminetetraacetic acid EGF epidermal growth factor ERCP endoscopic retrograde cholangiopancreatography EtBr ethydium bromide EUS endoscopic ultrasound FBC foetal bovine serum FISH fluorescence in-situ hybridisation GPCR G-protein coupled receptor HER human EGF receptor HGF hepatocyte growth factor IHC immunohistochemistry ISH in-situ hybridisation IPMT intraductal papillary mucinous tumour IPMC intraductal papillary mucinous carcinoma kDa kilodaltons MEM minimum essential medium MMP matrix metalloproteinase mRNA messenger ribonucleic acid NHS normal horse serum PanIN pancreatic intraepithelial neoplasia PC pancreatic cancer PCR polymerase chain reaction PBS phosphate buffered saline RA retinoic acid RFLP restriction fragment length polymorphism RTK receptor tyrosine kinase RT-PCR reverse transcriptase PCR SAGE serial analysis of gene expression SDS sodium dodecylsulfate SDS-PAGE SDS-polyacrylamide gel electrophoresis siRNA small interfering RNA TBS tris buffered saline TIMP tissue inhibitor of metalloproteinase UICC International Union Against Cancer uPA urokinase-type plasminogen activator VEGF vascular endothelium growth factor V/v volume per volume WHO world health organization W/v weight per volume iv
Abstract
Pancreatic cancer (PC) is the fourth leading cause of cancer death in Western societies. Despite significant progress in understanding the molecular pathology of PC and its precursor lesion: pancreatic intraepithelial neoplasia (PanIN), there remain no molecules with proven clinical utility. Affymetrix Genechip oligonucleotide microarrays were used to interrogate mRNA expression of PC and normal pancreas to identify molecular pathways dysregulated in PC. Analysis of these data identified altered expression of numerous components of the S100 Calcium Binding Protein Family, Retinoic Acid signalling pathway and the HOX transcriptional network in PC compared to normal pancreas. These pathways were assessed using immunohistochemistry (IHC) and in-situ hybridisation (ISH) in a cohort of patients with PC. Increased protein expression, of S100A2, S100A6 and S100P was observed in 43%, 60% and 48% of PC respectively. Expression of S100A2 was associated with a poor outcome (p = 0.009), whilst increased expression of S100A6 (p = 0.0008) and S100P (p = 0.0005) were associated with an improved outcome. Additionally, S100A2 expression was identified as an independent marker of outcome in resected tumours. Aberrant expression of retinoic acid signalling components was demonstrated in PC cell lines using semi-quantitative RT-PCR. ISH demonstrated expression of Retinoic Acid Induced 3 (RAI3), an orphan G protein coupled receptor normally expressed in the fetal lung, in 68% of PC, and this co-segregated with an improved overall survival (p = 0.026). Ectopic protein expression of HOXB2, a transcription factor normally expressed in the developing hindbrain and modulated by retinoic acid, was observed in 15% of early PanIN lesions and 38% of PC specimens. Expression of HOXB2 was associated with non-resectable tumours and was an independent predictor of poor survival in resected tumours. Suppression of HOXB2 protein expression using small interfering RNA, resulted in epithelioid trans- differentiation in the Panc-1 PC cell line, however no alteration in proliferation rates were observed compared to controls. This thesis has shown that transcript profiling and tissue validation has identified potential markers of early diagnosis and outcome in PC. Furthermore, pathways and molecules previously thought to be associated with normal human development have been implicated to play a role in the development and progression of PC. Further analyses of these markers will determine any potential role in future diagnostic and therapeutic strategies. v
Publications, abstracts and patents arising from work related to this thesis
Manuscripts
Segara D, Biankin AV, Kench JG, Langusch C.C, Dawson A.C, Skalicky DA, Gotley DC, Coleman MJ, Sutherland RL and Henshall SM. Expression of HOXB2, a retinoic acid signalling target in Pancreatic Cancer and PanIN. Clin Cancer Res, 11(9): 3587- 3596 (2005)
Skalicky DA , Kench JG, Segara D, Coleman MJ, Sutherland RL, Henshall SM, Musgrove EA, and Biankin AV. Cyclin E is an independent predictor of outcome in Pancreatic Cancer. Submitted (2005)
Pasca di Magliano M, Biankin AV, Heiser PW, Cano DA, Gutierrez PJA, Deramaudt T, Segara D, Dawson AC, Kench JG, Henshall SM, Sutherland RL, Dlugosz A, Rustgi AK and Hebrok M. Wnt Signalling in Pancreatic Cancer. Submitted (2005)
Sum EY, Segara D, Duscio B, Bath ML, Field AS, Sutherland RL, Lindeman GJ, Visvader JE. Overexpression of LMO4 induces mammary hyperplasia, promotes cell invasion, and is a predictor of poor outcome in breast cancer. Proc Natl Acad Sci U S A. 2005 May 24;102(21):7659-64.
Clancy JL, Henderson MJ, Russell AJ, Anderson DW, Bova RJ, Campbell IG, Choong DY, Macdonald GA, Mann GJ, Nolan T, Brady G, Olopade OI, Woollatt E, Davies MJ, Segara D, Hacker NF, Henshall SM, Sutherland RL, Watts CK. EDD, the human orthologue of the hyperplastic discs tumour suppressor gene, is amplified and overexpressed in cancer. Oncogene. 2003 Aug 7;22(32):5070-81.
Segara D, Skalicky DA, Kench JG, Coleman MJ, Sutherland RL, Henshall SM and Biankin AV. Aberrant expression of S100 protein A2 is associated with poor outcome in pancreatic cancer. In Preparation (2005) vi
Published Abstracts:
Segara D, Biankin A.V. Coleman M.J., Sutherland R.L. and Henshall S.M. A gene based approach to the identification of novel genes in the development and progression of Pancreatic Cancer by transcript profiling. ANZ Journal of Surgery 2003 May supplement: 73; A108.
Dawson AC, Horvath LG, Segara D, Biankin AV, Kench JG, Sutherland RL, Henshall SM Pancreatic Cancer and the Wnt Pathway. ANZ Journal of Surgery 2003 May supplement: 73; A108.
Patent:
Methods of Diagnosis and Prognosis of Pancreatic Cancer Biankin AV, Henshall SM, Segara D and Sutherland RL International Patent Application PCT/AU2004/000194 Filed; 18 February 2004 vii
Thesis overview
This thesis contains 7 chapters. The first chapter is a selective literature review divided into 2 parts: the clinicopathological characteristics of PC, and the molecular pathology of PC with a specific focus on: S100 calcium binding proteins, developmental pathways associated with pancreatic carcinogenesis, including retinoic acid signalling, and the homeobox transcription network. A short introduction to each of the results chapters 3, 4, 5 and 6 is aimed at extending this information in each specific area.
Chapter 2 describes the general materials and methods employed.
Chapter 3 describes investigations aimed at identifying novel candidate genes important in PC through molecular profiling experiments and the analysis of data derived from these experiments and electronic databases.
Chapter 4 introduces the PC cohort, defines its molecular pathology with respect to 3 selected members of the S100 family of calcium binding proteins, S100A2, S100A6 and S100P. This chapter addresses the putative role of S100A2 as a molecular marker of outcome in PC.
Chapter 5 introduces aberrant retinoic acid signalling in PC and examines the aberrant expression of central components of retinoic acid signalling in PC cell lines. The expression of an orphan G protein receptor Retinoic acid Induced 3 (RAI3), is examined in the PC cohort. Furthermore its putative role as a molecular marker of pancreatic resection and potential as a therapeutic target is addressed.
Chapter 6 addresses the molecular pathology of the PC cohort, with respect to the homeobox transcriptional network component, HOXB2, and its putative role as a marker of outcome and response to resection. HOXB2 expression is characterised in a panel of normal and cancer cell lines. In addition, the effects of modulating HOXB2 protein expression on cell morphology and proliferation is investigated in the Panc-1 PC cell line using siRNA. viii
Chapter 7 is a general discussion of the reported findings in the context of the published literature and a critique of the presented work. In addition future directions for the findings of this body of work as it applies to clinical trials is discussed.
Appendices: Appendices 1 to 7 contain detailed protocols for transcript profiling experiments, ISH, upregulated targets identified in the transcript profiling experiments, downregulated targets identified in the transcript profiling experiments, a clinical trial protocol designed for the prospective identification of molecular markers identified in this thesis, reprints of publications and ethical approval documentation. 1
Chapter 1
INTRODUCTION 2
Clinicopathological aspects of Pancreatic Cancer
Epidemiology
In 2003, 670 new cases of PC were diagnosed in NSW, Australia, with a crude incidence rate of 10 per 100 000, a slight preponderance of men over women conferring an overall lifetime risk of 1 in 152 1. This represented the 12th most frequently diagnosed cancer. The incidence in 2003 was slightly more than previous years but has otherwise remained relatively unchanged over the past decade, with a small decreasing trend in the incidence in men countered by a small increase in women. Although it is 12th in incidence, its high mortality rate which almost parallels its incidence makes PC the 4th or 5th highest cause of cancer death in men and in women in most western countries worldwide including Australia. In the U.S.A., the National Cancer Institute has recently identified PC as the 4th highest cause of cancer death in males, behind cancers of the lung, prostate and colon/rectum in that country2. PC also represents the lowest overall survival of all common cancers with a 5-year survival rate of less than 5%. Allowing for differences in age, males were 1.3 times more likely to be diagnosed and die of PC than females. The accuracy of these figures may be questioned as many PC deaths are labelled such without a tissue diagnosis. In addition, those that are labelled "malignant neoplasm, site unspecified", a group of significant size, would likely include a greater proportion of PC than other cancers. On a global scale it is 13th in incidence and 8th as a cause of cancer death. The incidence of PC increases progressively with age with the majority occurring in the seventh and eighth decades of life. Overall survival rates for PC in Australia are comparable with that in other parts of the Western world. The incidence of PC is higher in the Western world than in developing countries, in Jews compared to non-Jews, in Japan, and in the United States there is a slightly greater frequency in African-Americans than whites. 3
Familial Pancreatic Cancer
Recent evidence suggests that an estimated 5–10% of PC may have a familial component 3. Most cancers have a tendency to cluster in families, however the proportion of these cancers that are hereditary varies from cancer to cancer. The relative risk of developing PC is 5.25 if a close relative has PC and 1.86 for those who have a relative with any type of cancer 4. In 150 patients with PC and an equal number of controls, 10 (6.7%) reported a family history of PC compared to (0.7%) of controls 5.In a hospital based case-control study of PC in Ontario and Quebec, the lifetime risk of PC was 4.7% for the first-degree relatives of the PC patients. The risk was 7.2% for relatives of cases diagnosed before age 60, and was 12.3% for relatives of patients with multiple primary cancers (all ages) 5. Thus, overall, 3 to 10% of PC is likely to be causedbyinheritedfactors6, however, that figure may be higher when evidence from other cancer types is considered. Data from the prospectively accrued National Familial Pancreas Tumour Registry (NFPTR) at Johns-Hopkins has suggested that patients with 3 or more affected family members had a 57-fold increased risk of developing PC 7.
Familial Pancreatic Cancer (FPC) is an established disease syndrome, although there is no standardised definition of FPC, it is generally accepted that this term applies to those families having 2 first degree relatives with pancreatic cancer in an absence of an accumulation of other cancers known to be familial 8.
An inherited predisposition to FPC occurs in three defined settings. First in familial cancer syndromes known to be associated with PC, these include Peutz Jaegers Syndrome, Familial Atypical Multiple Mole-Melanoma (FAMMM) syndrome, Hereditary Non-Polyposis Colorectal Carcinoma (HNPCC), Familial Breast Ovarian Cancer, and Familial Adenomatous Polyposis (FAP) syndromes. Secondly, an inherited predisposition to FPC is associated with autosomal dominant hereditary pancreatitis, conferring patients with a relative risk of 100 and a cumulative lifetime risk of 40%. 9. Thirdly a sporadic form where the underlying gene defects are unknown however two or more first degree relatives have pancreatic cancer without fulfilling the criteria for one of the above cancer syndromes 10. Inherited factors may not specifically increase the risk of a particular type of cancer. 4
Although only 2.2% of patients with PC reported a family history of PC in the Japanese cancer registry, 25.9% gave a family history of other cancers 11. Genetic markers associated with these high-risk families have yet to be identified.
Risk Factors
Smoking
Nearly all published reports show that exposure to tobacco products increases the risk of PC, usually with about a 2-fold increased risk, compared to non-smokers. The proportion of PC attributable to smoking is approximately 25% 12. In addition, there is an increased incidence of Pancreatic Intraepithelial Neoplastic (PanIN) lesions, a precursor lesion of PC, within the pancreata of smokers at autopsy 13. A statistical analysis of seven cohort and eight case-control studies in Europe and the U.S.A. conducted by the International Agency for Research on Cancer (IARC) 3 identified a significant correlation between cigarette smoking and the risk of PC, with the risk increasing in relation to the number of cigarettes smoked. In the UK, Doll et al. reported the results of a 40-year study of British physicians 14, finding that annual male mortality rates for PC in non-smokers, ex-smokers and current smokers were 16, 23 and 35 per 100,000 man-years, respectively. Most reports demonstrate a graded dose response, with heavy smokers having a substantially higher risk of PC than light smokers 15,16. Concordantly it takes about 15 years after cessation of cigarette smoking among the heaviest smokers for the risk to fall to a level comparable to that of those that had never smoked, for light smokers, the period required to return to the baseline risk is about 5 years 17. Smoking may be associated with a small proportion of activating K-ras mutations which occur in over 90% of PC and is an early molecular event in the development of PanIN lesions 18. These mutations in PC are almost exclusively found in codon 12 and may be induced by nitrosamines derived from smoke 19. 5
The difference in incidence of PC between men and women may be related to cigarette consumption, as there is no difference in PC incidence amongst non-smokers. Hence there is strong evidence to implicate smoking as a primary risk factor based on these epidemiological and molecular studies.
Chronic Pancreatitis
Individuals with chronic pancreatitis have a substantially increased risk of PC compared to the normal population. Many studies have associated chronic pancreatitis of any aetiology with a 5.7 to 15 times increased risk of PC 20. A large six-country historical cohort of chronic pancreatitis patients identified a cumulative 25-year risk of PC with any form of chronic pancreatitis was around 4% 21. Current evidence suggests that some forms of chronic pancreatitis, particularly hereditary pancreatitis, may play a role in the aetiology of PC 22. Hereditary pancreatitis is an autosomal dominant disorder with a penetrance of 80% and variable expressivity, the causative mutation has been identified in the cationic trypsinogen gene 23.It accounts for 3–6% of all pancreatitis cases, with a well recognised increased risk of PC. A historical cohort of 246 hereditary pancreatitis subjects showed that the cumulative risk of developing PC by the age of 70 years is around 40%. This increases to 75% with the paternal transmission of hereditary pancreatitis 9. Whilst the association of hereditary pancreatitis and PC is clear, the causal relationship between all other forms of chronic pancreatitis and PC has been questioned. In two cohorts the risk of PC was substantially reduced after 10 years 24,25, suggesting a possible common aetiological factor rather than a causal relationship. This argument is strengthened by the fact that much of the histological changes seen in chronic pancreatitis within the pancreas may have been caused by pancreatic duct obstruction due to the invasive nature of PC. 6
Diabetes
Diabetes Mellitus type 2 has been consistently associated with increased risk of PC. In a meta-analysis of 11 case-control and nine cohort studies 26, it was found that the relative risk of PC for diabetics compared to non-diabetics was 2.1. A recent large prospective cohort study (20,475 men and 15,183 women) in the U.S.A., confirmed an independent association between post-load plasma glucose concentration and risk of pancreatic cancer mortality among individuals without self-reported diabetes. This observation indicates that factors associated with abnormal glucose metabolism may play an important role in the aetiology of PC 27. More recently Chari et al 28have observed in a population based cohort of 2122 patients that up to 1% of patients over the age of 50 with new onset diabetes were also diagnosed with PC within 3 years. Diabetes is associated with PC and PanIN lesions are found in greater frequency within the pancreata of diabetics at autopsy 29-31. However, because diabetes is diagnosed not infrequently at or close to the time of diagnosis of PC 32, arguments have been made that the increased prevalence of diabetes observed with PC is a recent onset phenomenon caused by the invading cancer. A case- controlled study performed in New Zealand concluded that diabetes was an epiphenomenon of PC rather than a risk factor for PC 33. The findings of studies to date make it difficult to delineate whether diabetes is a manifestation of PC or a risk factor of this disease.
Diet
Increased caloric content of the diet has been linked to several types of cancer, including PC. In particular the fat content of the diet, which contributes heavily to the overall caloric intake, is a suspected risk factor. In contrast, increased consumption of fruits and vegetables may reduce the risk of PC. Lower serum levels of selenium and of lycopene, a carotenoid present in fruits, were identified in patients who subsequently developed PC 34. Concordantly a protective dose response relationship has been identified between serum levels of folate and pyridoxine (found in fresh fruits and vegetables) and risk of PC 35. It has been suggested that these dietary factors are responsible for the higher incidence of PC in Western countries. The method of food preparation has also been implicated as a risk factor for PC. A high 7 intake of salted, smoked meat, dehydrated food, fried food and refined sugar increases the risk, whereas, food without preservatives or additives, raw food, food prepared by pressure-cooking or with electric or microwave ovens is associated with a lesser risk 36. Although alcohol is a major risk factor for pancreatitis, two major case control studies in the US and in Japan did not identify an increased risk of PC with alcohol intake 37,38. Coffee, tea, cereals and carbohydrates have been implicated as risk factors for PC, but there is insufficient conclusive evidence 39,40.
Associated Risks
Occupational exposure to products of incomplete combustion, some pesticides, asbestos, formaldehyde and styrene have been linked, although not conclusively, with an increased risk of PC 41. Previous gastrectomy and cholecystectomy are associated with an increased risk of developing PC. The increased production of N-nitroso compounds by bacteria following these operations is a potential explanation 42.
In summary, the aetiology of PC is poorly understood. Identified risk factors suggest a role for exposure to carcinogens present in cigarette smoke, diet and certain chemicals. 8
Pathology
Ductal adenocarcinoma of the pancreas is a malignant epithelial tumour composed of structures that show evidence of glandular differentiation. It is the predominant malignancy of the exocrine pancreas, constitutes over 75% of PC, and in general, PC refers to ductal adenocarcinoma in the world literature as it does in this text. Another 10% are variants of ductal adenocarcinoma (mucinous carcinomas, adenosquamous carcinomas, giant cell carcinomas and medullary carcinomas) 43. The majority of PC arise in the head of the pancreas (60%) with the rest arising in the body (13%), tail (5%) or involve the gland diffusely (21%). Macroscopically the neoplasms are firm white- yellow poorly defined masses. Microscopically PC is composed of epithelial cells forming glandular structures of varying size and shape. Depending on differentiation the cells display nuclear hyperchromia, pleomorphism and loss of polarity. The majority of pancreatic ductal adenocarcinomas are moderately differentiated. They elicit a significant desmoplastic response in most cases and hence the epithelial content may only be 50% or less of the tumour mass. This phenomenon also presents difficulties in the study of its molecular pathology using some traditional techniques. Well- differentiated carcinomas are composed of large duct-like structures that represent tubular glands of irregular size and shape. Poorly differentiated carcinomas are composed of neoplastic glands that show little resemblance to pancreatic ducts. Combinations of varying degrees of differentiation are often seen, however, well- differentiated components and poorly differentiated components within the same tumour are uncommon. Well-differentiated tumours can resemble normal pancreatic ducts, however, their architectural relationship to the lobular structure of the pancreas is not maintained. The cell of origin until recently was thought to be the ductal cell, most likely of the smallest ducts of the exocrine ductal acinar complex. Recent evidence suggests that multipotential cells also exist within adult pancreas, resulting in significant developmental plasticity in both endocrine and exocrine cell types 44. One candidate for the cell of origin in the pancreas is the centroacinar cell found at the junction of the acinus and the duct. These cells have been shown to proliferate in PTEN knockout mice 9 which display progressive ductal metaplasia, a small fraction of these mice later develop ductal malignancy 45. An alternative hypothesis is that the pancreatic acinar cells themselves undergo metaplasia/dedifferentiation to cuboidal duct like cells prior to progressing through dysplasia to invasive carcinoma 46.
Diagnosis and Treatment
Despite controversies surrounding the clinical management of PC, variations in diagnostic investigations and modifications in operative treatment, chemotherapy and radiotherapy have done little to alter the natural progression of the disease. The single most significant advance has been due to improvements in perioperative mortality with specialist centres reporting less than a 2% perioperative death rate. The greatest contributors to this improvement are likely to be improved anaesthesia, perioperative care and volume of operations performed in specialist centres. The relative inaccessibility of the pancreas to biopsy presents a significant challenge in diagnosing PC. With only a minority of patients suitable for curative surgery, resection rates vary between centres but in general are of the order of 10 - 20%. Early PC is generally asymptomatic, although patients often report a history of non-specific symptoms. Tumours of the head of the pancreas most commonly present with obstructive jaundice due to their proximity to the common bile duct and consequently are the most commonly resected tumours. Tumours of the distal pancreas, usually found incidentally, are more advanced at presentation. A schematic representation of the National Comprehensive Cancer Network practice guidelines for PC is presented in Figure 1.1. The detection of PC is usually made based on computed tomography (CT) scanning. Endoscopic ultrasound (EUS) is fast becoming a commonly used adjunct in the diagnosis of PC to aid in the determination of resectability. Based on these findings, together with histological confirmation of disease from transduodenal fine needle aspiration biopsy (FNAB), a provisional diagnosis of PC is made and resectability assessed on imaging criteria. Distant metastatic disease and marked local invasion precludes resection. These patients usually undergo percutaneous FNAB to confirm a tissue diagnosis prior to palliative treatment. Positive Neoadjuvant Chemoradiation in trial setting Staging Yes Laparoscopy Negative and Biopsy Adjuvant Candidate for Resectable Laparotomy Resection chemoradiation Neoadjuvant Resectable chemothaerapy No Laparoscopy Options for † Negative for mets Unresectable Biopsy unresectable disease without metastases Head or body § Spiral CT of pancreas Options for with Positive for mets without metastatic contrast jaundice disease Adenocarcinoma confirmed Negative for mets
Unresectable Biopsy Cancer not confirmed Follow + consider repeat biopsy
Other cancer confirmed Treat as appropriate
Clinical Trials preferred Clinical Trials preferred
Good PS Chemoradiation § Good PS Gemcitabine † Options for Options for unresectable disease Gemcitabine metastatic Supportive care without metastases disease Gemcitabine Poor PS Poor PS Gemcitabine supportive care supportive care
Figure 1.1: National Comprehensive Cancer Network (NCCN) practice guidelines for Pancreatic Cancer. Initially resectability is determined based on imaging and then further refined using endoscopic ultrasound and laparoscopy. In some cases non-resectability is determined after laparotomy and trial of dissection. For those tumours that are resectable based on imaging criteria a definitive tissue diagnosis may not be determined until after resection to minimise risk of spread and optimise the chance of cure. Unresectable tumours differ in their ongoing management based on the presence (§) or absence (†) of metastatic disease with chemoradiation reserved for those without evidence of macroscopic metastatic disease. CT - computed tomography, EUS - endoscopic ultrasound, PS - performance status. (Modified from NCCN® Practice Guidelines for Pancreatic Cancer v.1.2005. www.nccn.org 2005). 11
Those that appear resectable on imaging undergo laparoscopy for peritoneal disease and peritoneal fluid cytology and go on to resection if there is no evidence of disseminated disease. Questionable cases undergo EUS and FNAB to further assess resectability.
The initial management of PC is centred on the assessment of resectability and staging of the disease. This study has used the International Union Against Cancer (UICC) staging system, which is presented in Table 1.1.
Table 1.1: UICC Classification of Exocrine PC T: Primary Tumour TX Primary tumour cannot be assessed T0 No evidence of primary tumour T1 Tumour limited to the pancreas T1a Tumour 2cm or less in greatest dimension T1b Tumour more than 2cm in greatest dimension T2 Tumour extends directly into any of the following: duodenum, bile duct, peripancreatic tissue T3 Tumour extends directly into any of the following: stomach, spleen, colon, adjacent large vessels
N: Regional lymph nodes NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N1 Regional lymph node metastasis
M: Distant metastases MX Presence of distant metastasis cannot be assessed M0 No distant metastasis M1 Distant Metastasis
Stage Grouping Stage I T1 N0 M0 T2 N0 M0 Stage II T3 N0 M0 Stage III Any T N1 M0 Stage IV Any T Any N M1
Non-invasive techniques are initially used to assess the tumor stage. Helical CT scanning assesses distant metastatic disease (usually hepatic), which immediately precludes resection in most cases. Endoscopic ultrasound (EUS) is commonly used to give a better assessment of local invasion. Laparoscopy and cytology of peritoneal fluid are sometimes performed prior to laparotomy to detect peritoneal disease providing this is clear of malignant cells, laparotomy proceeds where assessment of the tumor continues. 12
Anatomical considerations play an important role in assessing resectability at this stage particularly local invasion of vascular structures including the portal vein, superior mesenteric artery and superior mesenteric vein. Once it is determined that anatomical clearance can be achieved resection proceeds. It is evident that with this approach to the treatment of PC, the surgeon is presented with a paucity of prognostic factors that are assessable prior to resection. Anatomical considerations determine the resectability of the tumour, which may not be the best method of ascertaining if the patient will benefit from surgery. It is clear that PC needs to be identified earlier in its biological course than it is in current clinical practice and that better determinants of response to operative resection prior to radical surgery are required.
Non Surgical Therapeutic Options
Advanced Disease
The primary goals of treatment for advanced PC are palliation and improved survival, however in this setting comfort-directed measures are paramount. In patients with advanced PC, the current standard of care is single-agent gemcitabine.
Recent studies investigating the use of gemcitabine in combination with radiation therapy, have returned poor results with significant toxicities reported 47-50. Gemcitabine has recently been combined with other chemotherapeutic agents in order to gain potential synergestic benefits, no statistically significant survival advantage observed in patients receiving the combination regimen 51.
Adjuvant Therapy
Chemotherapy and radiotherapy, either alone or in combination have made small differences in outcome, and are still controversial. A standard of 5-fluorouracil (5-FU)– based combined modality chemoradiotherapy has evolved from in-vitro data, animal studies, and a series of human studies. Only one randomized controlled trial assessing adjuvant therapy for pancreatic cancer in the United States has been reported. In 1985, 13 the Gastrointestinal Tumor Study Group (GITSG) initially reported that the median survival of patients undergoing Pancreativc Resection could be prolonged almost 2-fold by postoperative chemoradiation compared to surgery alone 52. A modest benefit was noted with combined chemoradiotherapy versus either chemotherapy alone or radiation therapy alone for those patients with local regionally- advanced and unresectable disease 53,54.
In spite of a growing body of literature supporting the benefit of adjuvant combined modality therapy after resection in patients with pancreatic cancer, adjuvant chemoradiation has not been universally accepted as standard of care. Recently, the European Study Group for Pancreatic Cancer (ESPAC) concluded that there was no survival benefit for adjuvant chemoradiotherapy. In addition, the authors concluded that a potential benefit existed for adjuvant chemotherapy alone after surgical resection and that chemoradiation is unnecessary and perhaps harmful 55. Clarification of the role of chemoradiation will be gained when the results of prospective randomized trials comparing the effects of 5-FU and gemcitabine with or without radiation in those patients who undergo pancreatic resection in the US (RTOG 97-04) and Europe (ESPAC-3), have been analysed.
As the debate regarding the optimal adjuvant therapy for pancreatic cancer continues future studies will be notable for the addition of multiagent chemotherapy to irradiation at the cooperative group level, or by the addition of gemcitabine to the period of chemoirradiation, and by the use of more advanced conformal, 3-dimensional planned irradiation, tailored to patient-specific anatomic and surgical pathologic data.
Neoadjuvant Therapy
Studies assessing neoadjuvant therapy demonstrate that although neoadjuvant chemoradiotherapy can be administered safely, there is no clear survival advantage to this strategy compared with postoperative therapy 56. However, preoperative chemoradiation therapy can convert selected patients with unresectable disease to a resectable status 57-59. 14
Immunotherapy
Immunotherapy offers the potential for modulating tumour activity that can be integrated with surgery, radiation, and chemotherapy. A major advantage of these therapies is their specificity targeting tumour cells rather than the normal cell of origin, thereby minimizing nonspecific toxicities. Immunotherapy is broadly divided into passive and active therapeutic approaches. Passive immunotherapy involves the use of unlabeled or labeled monoclonal antibodies that are specifically raised against tumour antigens. They may be used as diagnostic tools, prognostic indicators, or as primary therapy. A small number of antibodies have undergone testing to assess their suitability to treat PC including antibodies to EGFR including HER2/neu and Vascular Endothelial Growth Factor (VEGF), with poor results.
Active immunotherapy (vaccine therapy) targets specific tumour antigens by inducing antigen-specific B-cell or T-cell–mediated immune responses. Additionaly antigen- specific memory T-cells are generated and are capable of being reactivated if tumour cells expressing the same antigen profile recur. Point mutations in a variety of oncogenes (K-ras) or tumour suppressor genes (p16, p53, BRCA2, DPC4 ) have been described previously, these are associated with different histologically defined precursor lesions, and have made candidate immune targets. Vaccines developed against mutated K-ras have been assessed in several trials, with poor early results 60 Until a panel of pancreatic tumour–specific antigens is discovered and validated, the whole tumour cell, represents the best source of immunogens. This approach uses dessicated or whole tumour cells to stimulate an immune response and has been used in melanoma. A trial using this approach has detected immune responses in 1 of 3 patients with resected PC administered with irradiated allogeneic pancreatic tumour cell lines transfected with GM-CSF in sequence with adjuvant chemoradiation 61. A recent phase 2 trial using an allogenic GM-CSF vaccine has reported 76% 2 year survival in 56 patients, demonstrating that the vaccine approach can induce tumour-directed immune responses. 15
There are many significant challenges that need to be overcome if immune-based therapies are to play a role in the treatment of PC. Although components of the immune system have the capacity to recognize the specific motifs expressed on the tumour cell surface relative to their normal cellular counterparts, these specific targets are yet to be defined in PC. However, with the development of rapid methods for identifying genes that are differentially expressed by tumour cells through molecular profiling strategies, many more candidate targets are expected to be identified that may serve as immunogens for treatment, as well as prevention in the near future.
Prognostic Factors
Indicators of prognosis in PC are inconsistent and have remained controversial. Traditional prognostic factors lack accuracy, as some patients with large tumours and lymph node metastasis survive longer than those with smaller tumours and no metastases. There are few large series of PC assessing the influence of clinicopathologic parameters and treatment of PC on prognosis. Not all studies are congruent in their findings. A summary of the most recent published significant series is presented in Table 1.2.
Most series, report primary tumour size as an independent prognostic factor, but differ with respect to the effect of nodal status, margin involvement, perineural invasion and adjuvant therapy 62-66. The variability in the prognostic influences of these parameters may partly be attributed to differences in the proportions of patients exhibiting these characteristics and selection criteria for resection between different centres. There is also a degree of subjectivity in assessing parameters such as degree of differentiation, microscopic margin status and perineural invasion. The high mortality and rapid demise of patients with PC may also influence statistical analysis. Hence, the inaccuracy of these traditional clinicopathologic prognostic indicators and the difficulties in assessing them pre-resection may result in inappropriate patient selection for operative intervention. Molecular markers may become increasingly important not only as prognostic indicators but for patient selection for treatment in PC, and may serve as surrogate markers of the extent of disease and response to therapy. 16
Table 1.2: Comparison of Prognostic Factors reported in 6 Published Studies
Parameter Sohn Geer Nitecki Kawesha Biankin Segara et al. 65 et al..62 et al. 63 et al. 66 et al. 67 et al * n =616 n =146 n =174 n =157 n =51 n =76
Resection Rate 18 25 - - - (%) Mean Tumour 3.2 4.0 3.1 2 - 5 3.7 3.1 Size (cm) Positive 30 16 16 8 45 47 Margins (%) LN Metastases 72 47 56 45 47 51 (%) Poor 36 32 53 25 25 33 Differentiation (%) PN Invasion -2112-80- (%) 5 yearSurvival 17 24 6.8 - 11 11 (%)
Prognostic Indicators Univariate Multivariate Univariate Multivariate Univariate Multivariate Univariate Multivariate Univariate Multivariate Univariate Multivariate Size XXXXX -NS XXXNS Margins X X NS X -NS XXXX LN Metastases X NS X X X -X-XNSXNS PN Invasion NS NS - ---XX-- Differentiation XXXXNS NS NS NS NS Adjuvant XXNS NS NS NS -- Therapy
X = Statistically significant (p < 0.05) NS = not statistically significant - = not reported * = data derived from the following chapters
Many tumour-associated antigens have been studied in connection with PC, A few including carcinoembryonic antigen (CEA), pancreatic anti-oncofetal antigen, tissue polypeptide antigen, cancer antigen (CA) 125, and CA 19-9 are used variably in the clinical setting. CA 19-9, a sialylated Lewis A blood group antigen, is expressed and shed by pancreatic and hepatobiliary tissue as well as in many malignancies; thus, it is not tumour specific. The degree of increase in CA 19-9 levels may be useful in differentiating PC from pancreatitis 68. A post-operative decrease in serial CA 19-9 levels has been found to correlate with survival of PC patients after resection 69,70. Recent work has identified several biological targets, which show promise as markers of outcome and resectability in PC. Expresssion of MMP7 a matrix metalloproteinase which has been identified in 57% of PC 71 and was reported to be an independent 17 marker of prognosis in PC 72. More recently S100A6 (calicyclin) expression was identified as an independent marker of poor prognosis 73. No prospective trials have been performed on pre-operative biopsy material or pancreatic secretions to assess the potential utility of these targets in a clinical setting.
Further discussion of prognostic factors in PC and their relationship with markers identified in transcript profiling experiments is presented later in this thesis.
In summary, PC presents late in its clinical course and presents specific difficulties in management due to issues of organ accessibility, limited preoperative assessability and lack of response to antineoplastic therapies. A better understanding of the molecular pathology underlying the development and progression of PC may lead to improved patient selection for current therapies, earlier diagnosis and potentially, in the longer term, targeted therapies and chemopreventative strategies. 18
Molecular pathology of pancreatic cancer
Pancreatic cancer is thought to develop through the aberrant function of oncogenes and tumour suppressor genes, as well as abnormalities in growth factors and their receptors, which affect the downstream signal transduction pathways involved in the control of growth and differentiation 74. Aberrant expression of oncogenes is thought to initiate and promote the process of carcinogenesis whilst tumour suppressor genes act to inhibit this process.
Cell cycle regulation and pancreatic cancer
Aberrations of cell cycle regulatory molecules have been observed to occur in cancers of many organs, including PC. Overexpression of Cyclin D1 has been identified in 50% of PC and in over 50% of precursor lesions suggesting a role in the progression from normal ductal epithelium to PC 75. Increased Cyclin E expression was observed in 39 of 118 PC and was identified as a strong independent predictor of poor outcome (data submitted for publication) 76. Overexpression of the cyclin dependent kinase inhibitor (CDKI), p21WAF1/CIP1 is found to be an early event in the development of PanIN 75, and observed in 43% of PC 77. However, there is no association between p21WAF1/CIP1 overexpression and outcome in PC 75. Loss of expression of the CDKI, p27KIP1 is present in up to 67% of PC and was associated with advanced tumour stage in one study 78. Skalicky et al demonstrated low p27KIP1 expression in 41 of 111 PC, this did not correlate with clinico-pathological parameters or outcome. In addition, neither expression of p27KIP1 nor concomitant coexpression of p27KIP1 with cyclin E, cosegregated with outcome in PC 76. Decreased expression of p16INK4A, a CDKI o fthe INK4 family, was associated with highly dysplastic Intraductal Papillary Mucinous Tumours, (a putatative precursor lesion of PC) 79. Loss of p16INK4A expression occurs in 50-90% of PC due to promoter hypermethylation and/or homozygous deletion 80. No relationship with outcome has been reported. 19
Inactivation of p53 usually through point mutations occurs in PanIN-3 lesions and has been consistently reported in 50 - 75% of pancreatic carcinomas 6,81.
Cell signalling in pancreatic cancer
Components of the cell cycle regulatory machinery are themselves under the control of regulatory molecules within the cell. Receptors, commonly on the cell surface are capable of responding to extracellular signals by binding ligands. Two main categories of cell surface receptors have been implicated in cancer, receptor tyrosine kinases (RTK) and G-protein coupled receptors (GPCR).
Receptor tyrosine kinases
Receptor tyrosine kinases (RTK), dimerise/heterodimerise and undergo autophosphorylation of tyrosine residues within their intracellular domains upon ligand binding. The Epidermal Growth Factor Receptor (EGFR) family is one of the best characterised growth factor receptor families. Overexpression of EGFR occurs in 30% to 50% of PC 82 and overexpression of EGF and TGF occur in 12% to 46% and in 50% to 95% of PC respectively 83.
G- protein coupled receptors
G protein receptors constitute the largest known group of membrane bound signal transduction molecules. All share common structural features and including; seven membrane spanning domains, an extracellular domain capable of selectively binding growth factors and other molecules, a cytoplasmic tail, and three intracellular and extracellular loops that connect the transmembrane domains 84. Aberrations in GPCR's have not been extensively studied in PC although aberrations of downstream components of GPCR signalling, notably activating K-ras mutations, occur frequently in PC. 20
TGF- Signalling in PC
The TGF- family of growth factors is involved in cellular differentiation and inhibition of epithelial cell growth. TGF- ligands are known to enhance mesenchymal-derived cell proliferation, stimulate angiogenesis, modulate the composition of the extracellular matrix and exert immunosuppressive effects 85. Homozygous deletion or mutation of DPC4 (MADH4), the gene encoding Smad4 occurs in 55% of pancreatic ductal adenocarcinomas 86. Loss of Smad4 is also associated with improved outcome and a better response to resection in PC 67.
S100 Calcium Binding Proteins
S100 proteins are a family of low-molecular weight, calcium-binding proteins that initiate a number of cellular processes such as cell division, motility, secretion, protein synthesis, and membrane permeability by mediating the second messenger role of calcium. Twenty-one different S100 proteins have been identified and cloned 87,these are characterized by two distinct EF-hand motifs displaying different affinities for Ca2+. The N-terminus contains helix-loop-helix domains which are imperfect calcium binding motifs referred to as a half EF-hand domain, whereas the C-terminal portion has a more conserved EF hand domain conferring a 100 times greater binding affininty for calcium. S100 proteins are found in the cytoplasm as preassembled dimers, upon calcium binding a conformational change occurs in each individual monomer exposing a target protein recognition site. Once in the calcium activated state each monomer can interact with one target protein 88. In this state S100 proteins act as calcium sensors and play a role in protein phosphorylation 89, enzyme activation via phosphorylation 90, inflammation 91, cellular contraction and relaxation by regulating cytoskeletal constituents including actin and myosin 92 and nuclear transcription93 promoting cell proliferation and differentiation. In addition to a purely intracellular role some S100 proteins are excreted from the cell and potentially play a role in tissue organization and remodeling by modulating the proliferation of astrocytes and glial cells 94 and act as a chemoattractant to leukocytes 95. 21
The association of S100 proteins with cancer originated from the observation that chromosomal re-arrangements associated with tumour development occurred at 1q2, the site of an evolutionary conserved gene cluster of S100 genes 96. Aberrant S100 protein expression has been identified in many tumours, including those of glial, ovarian, breast stomach and pancreatic origin and was first used to identify metatastaic deposits of malignant melanoma 97.
Developmental pathways and pancreatic cancer.
The cellular processes associated with normal development including cell proliferation, cell differentiation, cell migration, angiogenesis and cell senescence are recapitulated in an uncontrolled manner in neoplasia. Recent studies have shown that molecular mechanisms important in vertebrate pancreas development are also important in the evolution of PC. Notch signalling regulates exocrine lineage selection in the developing pancreas and appears to maintain a pool of undifferentiated cells in the mature pancreas. Reactivation of Notch signalling is a feature of PC, and occurs early in the development of PanIN 98. Similarly, Hedgehog signalling, important in early endodermal specification, is also a feature of PC and early PanIN 99.
Notch and Hedgehog signalling in pancreas development
The mammalian pancreas is comprised of endocrine and exocrine lineages, both derived from a common progenitor pool in foregut endoderm.100,101 For both the exocrine and endocrine lineages, a hierarchy of nuclear transcripton factors is required for normal specification, epithelial expansion, and subsequent differentiation. In the developing embryo the parahox transcription factor, pdx1 is initially expressed in a broad domain of foregut endoderm extending from the caudal stomach to the rostral duodenum, spanning the endodermal elements destined to form the ventral and dorsal pancreatic buds.102 Within the cranial and caudal aspects of this domain, Sonic Hedgehog (Shh) expression suppresses a pancreatic fate, allowing normal specification of stomach and duodenum. In the nascent pancreatic bud, specific signals from overlying notochord and aorta result in down-regulation of Shh expression 103 and the realisation of a pancreatic fate 104 In ventral endoderm, up-regulation of Shh represses the pancreatic 22 developmental program in the nascent hepatic bud, while more caudal endoderm immediately adjacent to the anterior intestinal portio escapes the influence and pursues a pancreatic fate (Figure 1.2). In addition to the Shh-regulated influence of pdx1, normal pancreatic development also requires the early activity of the helix-loop-helix transcription factor, ptf1a-p48 100. Ptf1a-p48 is the pancreas-specific component of the multiprotein-PTF1 transcription factor complex required for normal exocrine differentiation.
Retinoic Acid Signalling
Retinoids regulate many fundamental cellular processes, including embryogenesis, cell growth, differentiation, and apoptosis 105. These compounds exert significant preventive and therapeutic effects against some human cancers. During vertebrate development, retinoic acid (RA) signalling is important for the correct patterning of embryonic structures 106. Retinoids exert their biologic effects through 2 families of nuclear receptors, retinoic acid receptors alpha, beta and gamma (RAR ,RAR and RAR ) and retinoid X receptors alpha, beta and gamma (RXR ,RXR and RXR ), which belong to the superfamily of steroid/thyroid hormone nuclear receptors. Retinoic acid receptor-alpha is homologous to the receptors for steroid hormones, thyroid hormones, and vitamin D3. Thus, the molecular mechanisms of the effect of vitamin A on embryonic development, differentiation and tumour cell growth may be similar to those described for other members of this nuclear receptor family 107. Reactivation of developmental pathways, specifically those that determine exocrine cell lineage, have been implicated previously in the early development of PC 108 and other pathways that determine duct cell versus acinar cell differentiation involving RA signalling, may also be important in this process. Endodermal expression of pdx-1 (a homeobox-containing transcription factor essential for pancreatic development), is induced by RA 109 marking a pluripotent population of cells that give rise to all cell types in the pancreas. Figure 1.2: Pancreas development. Repression of hedgehog expression in foregut endoderm specifies nascent pancreas. Overlapping domains of HlxB9, Pdx1 and Ptf1a-p48 are essential for pancreas development. The range of the pdx1 domain is regulated by retinoic acid signalling. Active Notch signalling regulates ptf1a-p48 and ngn3 function restricting terminal differentiation of pancreatic precursor cells. 24
RA signalling also regulates pancreas exocrine lineage selection at later stages of development, and treatment with RA analogues can effect a shift from an acinar to a ductal phenotype through epithelial-mesenchymal interactions primarily mediated through RAR- 110. Such a shift from an exocrine to a predominantly ductal phenotype is characteristic of mouse models of PC development. Forced expression of cellular retinoic acid binding protein (CRABP1), a mediator of RA signalling, in transgenic mice results in the development of poorly differentiated PC 111, further supporting a role of aberrant RA signalling in PC evolution. RAR- expression is thought to determine sensitivity of some pancreatic carcinoma cell lines to retinoid-mediated growth inhibition 112 and retinoic acid can induce apoptosis through RAR- mediated alterations in Bcl-2 expression 105. A number of RA responsive genes have been associated with PC and PanIN: MUC4 mucin is overexpressed in a significant proportion of PC 113 and PanIN 114 and can be induced through RAR- activation 115, similarly MMP9 is expressed in PC 116 and is upregulated by RA treatment 117 as is uPAR, HB-EGF and p21WAF1/CIP1 118. Id-1, which antagonizes basic helix loop helix proteins, inhibits differentiation, and can enhance cell proliferation is also overexpressed in PanIN lesions 119 and is RA responsive 120.In addition, pancreatic stellate cells, which are essential for the development of fibrosis associated with chronic pancreatitis and PC, store retinoids in fat droplets and in turn can have their function altered with RA analogue treatment in vitro 117. Recent evidence has shown that RA can exert its effects through the regulation of HOX gene expression 121.
The Homeobox Transcription Network
Homeobox genes are transcription factors associated with multiple aspects of cell function and development. The homeobox is a highly conserved 183 bp DNA sequence coding for a 61 amino acid homeodomain 122. This region binds DNA elements that primarily contain a TAAT core motif 123. Accordingly haemodomain containing proteins act as transcriptional regulators acting as both activators and repressors of DNA transcription. Class 1 haemodomain or HOX genes are structurally and functionally analogous to the homeotic complex (HOM-C) of Drosophillia, from which the haemodomain was first described in 1978 124. Human HOX genes consist of 39 25 genes arranged in 4 clusters (hox loci) HOXA, HOXB, HOXC and HOXD localised on chromosomes 7,17,12 and 2, respectively 125. Mammalian development requires a complex interaction between components of the HOX transcription network, with HOX gene expression associated with the development of tissue and organs from the hindbrain to the tail 126. HOX genes at the 3’ end of the HOX loci are expressed early in development and control the development of anterior regions of the organism, whilst the progressively more 5’ genes are expressed later in development and control the development of more posterior organs. Hence 3’ HOX genes in groups 1-4 (cervical) primarily control brachiofacial development and the embryonic hindbrain. Central HOX genes in groups 5-8 are involved in the development of thoraco-abdominal regions of the body. The 5’ genes in groups 9-13 control the development of the lumbosacral region of the organism. Aberrant HOX gene expression has been demonstrated in the development of some leukaemias 127. Misexpression and translocation leading to fusion protein expression has been identified in myeloid leukaemia 128,129 and lymphomas 130. Ectopic expression of HOX genes has been implicated in the development of solid tumours including: primary renal cancer 131, colon cancer 132, small cell lung cancers 133, ovarian carcinoma 134 , cervical cancer 135 and breast carcinoma 136,137. While the downstream effector genes that are activated by the HOX gene network are not well known, they include genes that encode for extracellular matrix components, angiogenic factors and growth factors all important in embryonal development and in tumour development and invasion. HOXD3 overexpression has been associated with the regulation of - 3 integrin and urokinase plasminogen activator 138. Vascular endothelial growth factor (VEGF) and angiopoietin-2 was up-regulated in SkBr3 cell lines overexpressing HOXB7, implicating HOXB7 as a key factor in the angiogenic process 139. HOXB7, is a secreted protein and its overexpression is associated with the induction of basic fibroblast growth factor (bFGF) in breast cancer and ovarian cell lines 134,140. bFGF is a key extracellular molecule involved in the regulation of cell proliferation, differentiation, migration and survival. Loss of expression of HOXA5 is associated with loss of expression of p53 141 and the progesterone receptor in breast carcinoma cell lines 142. 26
The complex temporal and spatial expression of HOX genes during development referred to as colinearity is critical to their regulatory function. It has been postulated that in each organ the HOX gene network manifests specific expression patterns which correspond to the developed cellular phenotype during development 137,143. Although associated with neoplasia, it is unclear whether carcinogenesis is an intrinsic function of upregulated HOX gene expression alone. HOX genes upregulated in cancer are normally expressed during development in the undifferentiated cell whilst HOX genes that are downregulated in cancer are expressed in the adult or differentiated tissue 123. Further evidence to suggest an important role of the HOX transcriptional network in development and carcinogenesis is the potential link to retinoic acid signalling. HOX genes have been implicated in breast epithelial cell development through regulation by retinoic acid 144. Hoxb2, the zebrafish orthologue of human HOXB2 is overexpressed with retinoic acid treatment, this is associated with abnormal morphogenesis of the midbrain and pharyngeal malformation 145.
Pancreatic Intra-Epithelial Neoplasia (PanIN)
Progression models describing cyto-architechtural changes to cells as they progress from a normal to a neoplastic phenotype have been well described in breast and colon cancer. Evidence for an association between changes in pancreatic ductal epithelium and the development of PC date back to observations made in the 1920’s 146. In the 1950’s autopsy studies identified hyperplastic duct lesions in 41% of pancreata containing carcinoma, whilst only 9% of non malignant pancreata contained hyperplastic duct lesions, interestingly these lesions were observed in 28% of pancreata from patients with diabetes mellitus 29. Case reports have associated the presence of these early lesions in the resection margin to recurrence of PC, many years after initial surgical resection 147. In addition, experimentally induced pancreatic cancer in the Syrian golden hamster model and in genetically engineered mouse models of PC is associated with hyperplastic and dysplastic duct lesions that demonstrate morphological features similar to the duct lesions observed in the pancreata of humans that contain a carcinoma 148. 27
There is now compelling histopathological and molecular evidence to support the evolution of PC, from normal cuboidal epithelium through a series of non-invasive duct lesions called pancreatic intraepithelial neoplasia (PanIN) 149. Early duct lesions designated PanIN-1A (tall columnar cells with some crowding) and PanIN-1B (tall columnar cells with increased crowding and papillary projections and a stromal core), show minimal cytological and architectural atypia, and are associated with activating K- ras mutations, shortened telomeres and overexpress p21WAF1/CIP1 75. PanIN-2 lesions exhibit mild to moderate cytological and architectural atypia (crowding of columnar cells with papillary projections associated with nuclear atypia) and are associated with loss of p16INK4A expression and cyclin D1 overexpression 75. PanIN-3 exhibits significant cytological and architectural atypia (atypical ductal hyperplasia), manifests p53 mutations and loss of DPC4/Smad4 expression 75. These molecular aberrations increase in frequency with advancing PanIN lesions through to invasive cancer.
Summary
Over the last half decade there has been a wealth of data generated in understanding the development of the various phenotypes associated with the normal pancreas and the progression to PC. Numerous data suggest the existence of self-renewing, pluripotent pancreatic stem cells, but their molecular characteristics and differentiation pathways remain to be fully elucidated.
The high mortality rate of PC and poor response to conventional therapies including surgery, neoadjuvant and adjuvant chemotherapies, suggests that any advance that leads to the development of novel diagnostic and therapeutic strategies based on our increasing understanding of the molecular pathology of the disease are likely to greatly improve the outcome of patients whose current prognosis is dismal. For improvements in outcome to be made in PC, three key issues need to be addressed in the short term.
First, PC is usually advanced at presentation with signs and symptoms that are usually non-specific accounting for the low proportion of patients suitable for resection. Understanding the molecular pathology and subsequent molecular characterisation of its 28 precursor lesions may provide scope for earlier identification of PC and subsequently improved outcome with current therapies. An understanding of precursor lesions in breast and colon cancer has significantly altered the management of those diseases. PC is thought to develop through a series of hyperplastic and dysplastic duct lesions termed pancreatic intraepithelial neoplasia (PanIN). PC can also arise from intraductal papillary mucinous tumours (IPMT). Assessment of the molecular pathology of PC is made difficult, by issues of accessibility to the organ for biopsy. Hence the determination of the natural history and malignant potential of precursor lesions such as PanIN and IPMT has not been as well developed as the study of corresponding lesions in other cancers. Currently these lesions can only be assessed following pancreatic resection and consequently these studies must be retrospective. However, it is envisaged that once a better understanding of the molecular aberrations associated with the development and progression of PC becomes available, there will be sufficient motivation and collaborative capacity to make advances that will provide a means of sampling or imaging the pancreas in vivo, leading to the development of early diagnosis, screening and chemoprevention strategies.
Second, resection remains the only therapeutic intervention that affects prognosis in this disease. However, selection criteria for resection are currently based solely, on anatomical criteria, allowing for the surgical resection of the tumour. A proportion of patients that do not undergo resection survive for > 1 year indicating a biologically less aggressive tumour, similarly some patients who undergo resection for PC, have shortened survival after resection indicating a biologically more aggressive tumour. Clearly, identification of molecular markers that co-segregate with response to operative resection will allow physicians to more adequately select patients for pancreatectomy which in 2006 still has significant mortality and morbidity.
Third, there remain no rationally designed, molecularly targeted therapies for PC treatment, as have been implemented for non-Hodgkins lymphoma 150 and breast cancer 78 in the form of specific monoclonal antibodies that target tumour-specific cell surface receptors 151 and tyrosine kinase inhibitors that inhibit growth factor receptor signalling in the case of gastrointestinal stromal tumours 152. This highlights the need for the 29 development of innovative therapies that are designed through the discovery of novel therapeutic targets.
The goal of this thesis is to address the questions posed above in the following hypothesis:
Transcript profiling of pancreatic cancer will allow for the identification of genes of relevance to the pathophysiology of pancreatic cancer (PC) and that these genetic aberrations may be utilised to improve early diagnosis, future treatment and outcome of the disease.
The hypothesis will be addressed by the following specific aims:
1. Using Affymetrix GeneChip® oligonucleotide arrays (HGU-133 A & B) to identify genes and molecular pathways that are aberrantly expressed in PC compared to normal pancreas and other normal organs
2. Assessing the prevalence of selected aberrantly expressed genes and altered molecular pathways identified in aim 1 in PC, using archived pancreatic specimens and established pancreatic cancer cell lines; specifically: Components of the S100 family of calcium binding proteins; S100A2, S100A6, and S100P. The Retinoic Acid Receptors; RAR,RAR,RAR,RXR,RXR,RXR. The Cellular Retinoid Binding Protein; CRBP1 Downstream Retinoic pathway effector molecules; the G protein Coupled receptor RAI3 and the retinoic acid induced developmental protein HOXB2.
3. Dileneateing the relationships between the genetic aberrations investigated in aim 2 and clinicopathological parameters of response to therapy and outcome 30
4. Assessing the abberant expression of the developmental protein, HOXB2 in the development of PC through its precursor lesions, Pancreatic Intraepithelial Neoplasia (PanIN).
5. Investigating the role and function of the abberant expression of HOXB2 in PC identified and validated through aims 1 to 3. in pancreatic cancer cell lines. By determining the denovo expression of this gene in PC cell lines. And determining the effect of altered expression of HOXB2 in cell growth. 31
Chapter 2
METHODS 32
Transcript Profiling
Tissue acquisition
Ethics approval was obtained from the same 3 teaching hospitals in Sydney (St- Vincents Hospital, Westmead Hospital, Concord Hospital) and 1 teaching hospital in Brisbane (Princes Alexandria Hospital) for the acquisition of fresh pancreatic tissue from pancreatectomy specimens. For each specimen an effort was made to collect samples of tumour, exocrine pancreas as far removed from the tumour as possible, duodenal mucosa and gastric mucosa. Not all such samples were available due to the type of procedure performed and the size of the tumour (small tumours were not available for research purposes). Multiple samples of approximately 500mg each were excised with a clean scalpel, wrapped in foil and snap frozen in liquid nitrogen and subsequently stored at minus 80oC in the cryogenic facility at the Garvan Institute of Medical Research.
RNA extraction and Study Design
RNA extraction was performed with the help of Dr A.V Biankin. Portions of the frozen samples (100 to 250 mg) were homogenised with a polytron tissue homogeniser in Trizol® Reagent (Invitrogen, Life Technologies, Carlsbad, CA) and total RNA was extracted using the Trizol method as per the manufacturer's protocol and purified using RNAqueousTM glass fibre filters (Ambion, Inc., Austin, TX). A small piece of tissue adjacent to the sample to be homogenised, was removed and fixed in 10% Neutral Buffered Formalin for histological analysis. For samples to proceed to cRNA synthesis the folowing criteria were required:
1. The histological section of the tissue fated for homogenisation was representative of either exocrine pancreas or pancreatic cancer. 2. Total RNA of sufficient quantity (> 2 g) and quality (260/280 ratio on spectrophotometry > 1.6 3. Clear 28S and 18S bands on 1% agarose gel electrophoresis) were present . 33
The planned study design was based on a total of 18 specimens. A total of 12 cancers, six each of moderately and poorly differentiated carcinoma and 6 samples of "normal" exocrine pancreas from the same patients whose cancers were also in the study (matched) to minimise inter-individual genetic variations. As the cell of origin of PC is still not clearly defined, a comparison of moderately differentiated cancers with poorly differentiated cancers would also allow for some cancer specificity. Twenty-three samples satisfied the above criteria and were carried forward to cRNA synthesis. These included 16 cancer samples and 7 samples of matching exocrine pancreas. cRNA Synthesis cRNA was prepared with the kind help of Dr A.V. Biankin, through a single round of reverse transcriptase polymerase chain reaction (RT-PCR) with a starting amount of 5 g of total RNA with Superscript II (Invitrogen, Life Technologies, Carlsbad, CA) followed by second strand synthesis to create double stranded cDNA. After purification biotinylated cDNA was transcribed using a T7 polymerase with the Enzo BioArrayTM, HighYieldTM, RNA Transcript Labelling Kit (Enzo Diagnostics, Inc., Farmingdale, NY) and purified with the RNeasy® Protect Kit (QIAGEN, Inc., Valencia, CA) (detailed protocol in appendix, as per Baugh et al. 153, Figure 2.1). Quality and quantity of cRNA was assessed spectrophotometrically and on agarose gel electrophoresis. A minimum of 10 g of cRNA was required for hybridisation to Affymetrix Genechip® oligonucleotide HG-U133 arrays. cRNA was fragmented and hybridisation cocktails were prepared as per the Affymetrix protocol and quality assured on Affymetrix Test3 arrays prior to hybridisation to HG-U133A arrays. 1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Figure 2.1: Eukaryotic target labelling for Affymetrix Genechip® oligonucleotide microarrays. 35
Organisation of data
A relational database was constructed in conjunction with Dr A.V Biankin, using FileMaker Pro® (FileMaker, Inc., San Francisco, CA) to facilitate multiple queries of transcript profile data obtained from the above experiments and public domain data available electronically from the internet The database incorporated:
1. Transcript profiles of PC and Normal Pancreas from the experiments performed in this study (Absolute Values). 2. Mathematical algorithms programmed within the database to generate fold change comparisons between subsets of data. 3. Linear statistical analyses generated using the Affymetrix Data Mining Tool Software (Affymetrix Inc. San Francisco, CA), which included T-test and Mann-Whitney U test data for various comparisons between Normal Pancreas and PC (also separating moderately differentiated and poorly differentiated carcinomas). 4. Clustering analyses with self-organising maps using GeneCluster 2 (Whitehead Institute, MIT, Cancer Genomics Department of Computational Biology http://www- genome.wi.mit.edu/cancer/software/genecluster2/gc2.html) and hierarchical clustering using dchip software (Wong Lab, Dept. of Biostatistics, Harvard School of Public Health, Cambridge MA, http://www.biostat.harvard.edu/complab/dchip/). 5. SAGE libraries of cultured normal pancreatic duct cells (HX and H126), PC cell lines Panc1, Capan1, Capan2, HS766T) and human PC (96_6252, 91_16113) and normal tissue from other organs (normal endothelium, prostate, breast ducts, ovary, CNS, spinal cord, lung, heart) was downloaded (http://www.genetics.pitt.edu/sage/). Analysis of relevant libraries was also performed using xProfiler software for comparison of gene expression between groups including normal duct cultures versus PC and cell lines, normal duct cultures versus cell lines alone and normal tissues versus PC (http://www.ncbi.nlm.nih.gov/SAGE/index.cgi?cmd=expsetup). 6. Retrieval of Genbank, Unigene, Locuslink, OMIM, SwissProt, PubMed link and other relevant publicly available data, in particular NCBI repositories, was facilitated by web based search and retrieve interactive query software available at Affymetrix (http://www.affymetrix.com/analysis/index.affx) and GeneCruiser, a similar but 36 complementary program (Whitehead, MIT, Boston MA; http://www- genome.wi.mit.edu/cancer/genecruiser/src/main.jsp). Both allowed retrieval of data based on affymetrix HG-U133 probeset identification strings and hence facilitated linking to other web based tools which required other gene identification codes such as Genbank ID, Unigene Clusters and SwissProt accession numbers. 7. Links were established to automate NCBI and other database searches directly from the database where appropriate data could be imported into the central PC database. 8. GenMAPP software (Gladstone Institutes UCSF, San Francisco, CA, http://www.GenMAPP.org/default.html) is designed to incorporate transcript profile data into maps of known pathways including those involved in carcinogenesis and allowed for the rapid construction of interactive molecular pathway maps which incorporated and presented transcript profile comparisons between experimental groups for molecules within a given pathway of interest.
Figure 2.2 shows the main page of the database. The database can be interrogated with multiple component queries to select genes that satisfied desired criteria. For example, a search could be performed for those genes that had a 5-fold or greater increase in expression in PC compared to normal, with a T-test p value of < 0.005, with less than 5 SAGE tags identified in HX and H126 normal duct epithelium cell lines or other normal tissues and which followed a particular cluster progression on self-organising map analysis. The found set of genes could then be sorted based on fold change and then narrowed further based on their perceived functions as reported in Unigene clusters and further investigated through the links to NCBI data repositories and other databases. The advantage of establishing such a database is that it allows the relatively straightforward manipulation of data, it also contains relevant information about each gene, with rapid linking to web based databases. It also links all relevant gene codes such as Genbank ID and Unigene clusters with the affymetrix probeset ID. As web based analysis tools are not standardised as to the format that data are presented, datasets containing the necessary parameters are easily exported from the database as required. Common Name FileMaker Analysis Chart of Scores
Raw Scores Internal Database }Data
Direct Links to xProfiler Sites SAGE Analysis }
SOM Pattern
Figure 2.2: the main page of the database illustrating the interface used for the interrogating the data based on several analyses results with links to web based applications and databases. 38
Patient Characteristics
A cohort of 348 patients with the diagnosis of pancreatic ductal adenocarcinoma was identified from Westmead Hospital, St Vincent's Hospital Campus, Royal Prince Alfred Hospital and Concord Repatriation Hospital in Sydney, Australia. Electronic and manual database searches were performed in the Departments of Anatomical Pathology and Medical Records to identify patients treated for PC within each institution. Archival formalin-fixed, paraffin-embedded tissue from 129 pancreata that were resected or biopsied between January 1972 and July 1999 was available. Human Research Ethics approval for data collection and tissue use was granted by the St. Vincent's Hosptal, University of NSW and the Royal Prince Alfred Hospital Ethics Committees, as well as specific approval from the Westmead Hospital and Concord Hospital Ethics Committees (relevant documentation can be found in Appendix 7).
Clinicopathological parameters including sex, age, preoperative assessment of disease stage and type of operative procedure were gathered retrospectively from individual patient clinical records. Pathological parameters of disease stage (tumour size, location, involvement of surrounding structures and lymph node status) were obtained from the original pathology report. In addition to the original pathology reports, microscopic findings (tumour type, degree of differentiation and perineural invasion) were reassessed by Dr J.G. Kench a histopathologist affiliated with the Pancreatic Cancer Research Group at the Garvan Institute of Medical Research. Outcome data reporting date and cause of death was obtained on a return to notifier basis from the Cancer Council NSW.These data were collated and entered into a custom designed, password protected, relational database programmed in FileMaker Pro® (FileMaker, Inc. San Francisco, CA).
Haematoxylin and eosin stained sections of issue blocks were examined to select those of relevance to the study including those that contained PC, PanIN lesions, and normal pancreas. 39
Tissue Arrays
Archival formalin-fixed, paraffin-embedded tissue from all the 129 pancreata that were resected or biopsied were used to construct seven PC tissue arrays, using a Beecher tissue array maker (Beecher Instruments, Silver Spring, MD). Each array block contained 55 x 1.6mm cores of tissue. Orientation was maintained using a piece of liver in one corner. These blocks were then cut and mounted onto superfrost plus slides for staining.
Immunohistochemistry
Haematoxylin and eosin staining and immunohistochemistry were performed on 4 μm serial sections of paraffin-embedded, formalin-fixed tissue mounted on SuperFrost slides (Menzel-Glaser, Germany). For immunohistochemistry, sections were deparaffinised in xylene and rehydrated through a series of alcohols. Immunohistochemistry was performed with the kind help of Mr D.A. Skalicky and Ms S. Eggleton.
S100 Calcium Binding Proteins
S100A2: Pancreatic tissue microarrays were dewaxed and rehydrated before unmasking was achieved using target retrieval solution (EDTA and citrate: DAKO Corporation, Carpenteria, CA) in a waterbath, for 30 min. Endogenous peroxidase activity was quenched with 3% hydrogen peroxide in methanol. Sections were incubated for 60 min with 1:100 anti S100A2 (Clone DAK-S100A2/1) monoclonal mouse antibody (DAKO Corporation, Carpenteria, CA). A labelled polymer horse radish peroxidase anti mouse detection system was used according to the manufacturer's instructions (Envision + anti mouse; Dakocytomation; DAKO Corporation, Carpenteria, CA). 3,3'-diaminobenzidine was used as a substrate. Counterstaining was performed with Mayer’s hematoxylin (DAKO Corporation Carpenteria, CA). S100A6: Pancreatic tissue microarrays were dewaxed and rehydrated before unmasking was achieved using target retrieval solution (EDTA and citrate: DAKO Corporation, Carpenteria, CA) in a waterbath, for 30 min. Using a DAKO autostainer endogenous 40 peroxidase activity was quenched with 3% hydrogen peroxide in methanol. Non- specific binding of secondary antibody was blocked by incubation with a serum free protein block (DAKO Corporation, Carpenteria, CA). Sections were incubated for 30 min with 1:2000 anti S100A6 (Clone A5115) polyclonal rabbit antibody (DAKO Corporation, Carpenteria, CA). A labelled polymer horse radish peroxidase anti rabbit secondary antibody detection system was used according to the manufacturer's instructions (Envision + anti rabbit; Dakocytomation; DAKO Corporation, Carpenteria, CA). 3,3'-diaminobenzidine was used as a substrate. Counterstaining was performed with Mayer’s hematoxylin (DAKO Corporation Carpenteria, CA). S100P: Pancreatic tissue microarrays were dewaxed and rehydrated before unmasking was achieved using a protein kinase solution for seven minutes (PK7: DAKO Corporation, Carpenteria, CA). Endogenous peroxidase activity was quenched with 3% hydrogen peroxide in methanol. Sections were incubated for 30 min with 1:150 anti S100P (Clone 16) monoclonal mouse antibody (BD Transduction Laboratories/ BD Biosciences, San Jose, CA). A labelled polymer horse radish peroxidase anti mouse secondary antibody detection system was used according to the manufacturer's instructions (Envision + anti mouse; Dakocytomation; DAKO Corporation, Carpenteria, CA). 3,3'-diaminobenzidine was used as a substrate. Counterstaining was performed with Mayer’s hematoxylin (DAKO Corporation Carpenteria, CA).
Homeobox Transcription Network
HOXB2: Pancreatic tissue microarrays were dewaxed and rehydrated before unmasking in target retrieval solution (EDTA and citrate, DAKO Corporation, Carpenteria, CA) in a microwave for 30 min. Using a DAKO autostainer, endogenous peroxidase activity was quenched in 3% hydrogen peroxide in methanol, followed by avidin/biotin and serum free protein blocks (DAKO Corporation, Carpenteria, CA). Sections were incubated for 30 min with 1:200 anti-HOXB2 (P-20) antibody (Santa Cruz Biotechnology, Santa Cruz, CA). A streptavidin-biotin peroxidase detection system was used according to the manufacturer's instructions (LSAB label + link kit; DAKO Corporation, Carpenteria, CA) with 3,3'-diaminobenzidine as a substrate. Counterstaining was performed with Mayer’s hematoxylin.. 41
Formalin fixed, paraffin embedded cell lines, where the status of the genes examined had been determined by other authors, were used as positive and negative controls (Table 2.1).
IN-SITU HYBRIDISATION
Preparation of in situ probe (riboprobe)
Preparation of in-situ probe was performed with th ekind help of Mr D.A Skalicky. cDNA was prepared using the Expand Reverse Transcriptase System (Roche Diagnostics Mannheim, Germany). This was followed by a polymerase chain reaction (PCR) was performed using the Expand High Fidelity PCR system (Roche Diagnostics Mannheim, Germany) with 5 μL of cDNA and the RAI3 insitu probe forward and reverse primer as shown below.
PROBE FORWARD REVERSE RAI3 5’-TTA AGT GGG AGT CTC AGG CA-3’ 5’-GAG GCA GCA CTA GAG AGA TGA-3’
Electrophoresis of the PCR product was performed on a 1% agarose gel. Once its presence was confirmed, the PCR product was purified using Roche HighPure RNA Isolation Kit (Roche Diagnostics Mannheim, Germany), according to the manufacturers instructions. A template for the in vitro transcription reaction was constructed by ligation of the T7 Promoter to 25ng of the PCR product using a T4 DNA ligase and the T7 Promoter Adapter in the Lig’nScribe Kit (Ambion, Austin Tx, USA). Amplification of the IVT templates was achieved using the RAI3 in situ primers, T7 adapter primers (provided in the Lig’nScribe kit) using the Expand High Fidelity PCR system (Roche diagnostics Mannheim, Germany) and 2 μL of the template material. Two reactions were performed, the antisense template was created using the RAI3 in situ forward primer, whilst the sense template was created using the RAI3 in situ reverse primer. 5μLof each PCR of the resultant PCR products were visualised on a 1% agarose gel with a 100bp marker (Promega, Madison, WI, USA). The IVT templates were sequenced 42
(Australian Genome Research Facility, Brisbane QLD) and the IVT reaction performed using T7 RNA polymerase reaction followed by DNase1 digestion using the DIG RNA Labelling Mix (Roche Diagnostics, Mannheim, Germany) as per manufacturers instructions. The IVT reaction was purified using ethanol precipitation in the presence of LiCl and the resultant pellet dissolved in RNase inhibitor and stored at -70 C.
Quantification of in-situ probe
The RAI3 in situ probe was quantified by serial dilution of the probe and controlled DIG-labelled RNA (Roche Diagnostics, Mannheim, Germany) and spotted onto a Transblot Transfer Nitrocellulose Membrane (Bio-Rad Laboratories, Hercules, CA,USA). The membrane then heated in a microwave at the highest setting for 2 min. This was rinsed in maleic acid buffer (MAB; 100mM Maleic acid, 150mM NaCl, pH 7.5), blocked for 1 min in Blocking Solution (5% skim milk powder in MAB). Anti- digoxigenin-AP Fab fragments (Roche Diagnostics, Mannheim, Germany) ta a final dilution of 1:1000 was incubated for 30 minutes and rinsed twice in MAB for 5 min. The membrane was developed in 0.1 M tris(hydroxymethyl)-aminomethane (Tris) pH
9.5, 0.05 M MgCl2 and 0.25 M NaCl with NBT/BCIP (Roche Diagnostics, Mannheim, Germany) diluted 1:50. The membrane was allowed to develop in the dark until the last spot in the control was visible and then rinsed in water and allowed to dry. The amount of probe in ng/μL was the ng amount in the control (of the same intensity as the last visible dilution of the probe) multiplied by the dilution factor of the last visible dilution of the probe.
In- Situ Hybridisation.
In situ hybridisation was performed with the help of Ms S.A Eggleton, using the Discovery automated machine (Ventana Medical Systems, Tuscon, Az, USA), the RiboMap Kit for tissue preparation and the BlueMap Kit for detection (Ventana Medical Systems, Tuscon, Az, USA). All reagents were purchased from Ventana Medical Systems, Tuscon, Az, USA. Breifly the slides were pre-treated with Riboprep for 30 min at 37 C, RiboClear for 10 min at 37 C, Mild Cell Conditioning Solution #2 43 for 2 min and Protease III for 2 min at 37 C. Hybridisation was performed by wet application of 10ng of in-situ probe in Liquid Cover Slip, denaturation for 10 min at 70 C, and hybridisation for 8 h at 60 C. Two washes were performed using Ribowash for 6 min at 75 C and post-treatment fixation performed for 6 min at 37 C using Ribofix. Signal detection was achieved using a monoclonal anti-digoxin biotin conjugate antibody (Sigma, St Louis, MO, USA) for 30 min at 37 C with the substrate provided by the BlueMap Detection Kit for 6h. A detailed protocol is enclosed in Appendix 2. Slides were counterstained using Nuclear Fast Red (DAKO, Carpenteria, CA, USA), rinsed, dehydrated in increasing concentrations of alcohol and slidebrite (SASCO, Albany, GA, USA). The coverslip was placed using Britemount (IMEB, San Marcos, CA, USA)
Immunohistochemical and In-situ hybridisation scoring
Up to three separate samples of pancreas were examined per patient. Staining was assessed by at least two independent observers blinded to patient outcome, including a specialist histopathologist Dr J.G Kench. Standardization of scoring was achieved by comparison of scores between observers, and by conferencing, where any discrepancies were resolved by consensus. Scores were given as a percentage of nuclei staining positive within a representative area of each tumour and PanIN lesion. The criteria to achieve a positive score for each of the antigens studied were based on the following:
S100 proteins
Scores were given as a percentage of cells with positive cytoplasmic staining within the representative area of the tissue microarray core and the absolute intensity of cytoplasmic staining on a scale of 0 to 3 (0 representing no staining 1,representing heterogenous cytoplasmic staining, 2, representing homogenous cytoplasmic staining and 3 representing intense homogenous cytoplasmic staining). The criteria to achieve a positive score for each of the antigens studied were based on the following criteria: 44
Protein Cytoplasmic Staining Nuclear Intensity S100A2 Homogeneous; >30% Intensity >1 S100A6 Homogeneous; >10% Intensity >1 S100P Homogeneous; >20% Intensity >1
S100A6 and S100P scoring were based on the system described by Vilmachandran et al 73. A 10% cut-off was placed on S100A6 nuclear scoring because those cancers with less than 10 % cells staining were due to small foci of positive cells indistinguishable from artifact. Hence to avoid potential false positives, a cut-off of 10 % was used. S100A2 scoring was determined according to the means of S100A2 nuclear staining and % of nuclei stained in the whole cohort.
Stroma and adjacent pancreatic tissue served as internal positive controls. In every case the intensity of staining within the cancer was compared to these controls.
HOXB2
Scores were given as a percentage of nuclei staining positive within the representative area of the tissue microarray core and the absolute intensity of nuclear staining on a scale of 0 to 3 (0 representing no staining 1,representing, heterogenous nuclear staining, 2, representing homogenous nuclear staining and 3 representing intense homogenous nuclear staining). The criteria to achieve a positive score for each of the antigens studied were based on the following criteria: HOX B2 >20% of nuclei having homogeneous nuclear staining and HOX B2 nuclear intensity being >1. A score of 0 or 1 was defined as HOXB2 negative .
RAI3 ISH
Cells were defined as positive when the antisense strand identified the presence of staining in either nuclei or cytoplasm. Sense strand was used as a negative control. 45
Statistical evaluation
Survival analysis was performed using Kaplan-Meier survival for univariate analysis and the Cox proportional hazards model for multivariate analysis using Statview 5.0 Software (Abacus Systems, Berkeley, CA). A p value of < 0.05 was accepted as statistically significant. Those factors that were prognostic on univariate analysis were then assessed in a multivariate, Cox proportional hazard’s model to identify those variables of parameters that were independently prognostic and those that were the result of confounding. Data displayed includes Hazard’s ratio, 95% confidence intervals and p value. This analysis was performed sequentially on all patients who had available tissue (n = 128) and then on a subgroup of patients that underwent operative resection (n = 76).
Statistical evaluation of molecular aberration in PC, normal pancreas and PanIN lesions was performed using chi square analysis and Fisher's exact test using Statview 5.0 Software (Abacus Systems, Berkeley, CA). A p value of < 0.05 was accepted as statistically significant. 46
General Tissue Culture
Established human cell lines were maintained by Ms Gillian Lerbach at the Garvan
Institute of Medical Research, and cultured in a humidified atmosphere of 5% CO2 and 95% air at 37°C. Cell lines used and media are listed in the following table.
Cell line Source Incubation solution
FaDu ATCC (Rockville MA, USA) Emem + Na Pyruvate +NeAA +10% FBS 1 Detroit 562 ATCC Emem + Na Pyruvate +NeAA +10% FBS MCF 7 EG& G Mason Research Institute RPMI 1640 medium +10%FBS 2 (Worcester, MA, USA) MCF10A ATCC MCDB 170 medium +Bovine Pituatary Extract BT-20 ATCC RPMI 1640 medium +10%FBS 2 SkBr-3 ATCC RPMI 1640 medium +10%FBS 2 MDA-MB- ATCC RPMI 1640 medium +10%FBS 2 330 184 (A kind gift from Martha Stampfer, MCDB 170 medium +Bovine Pituatary University of California) Extract ZR-751 ATCC RPMI 1640 medium +10%FBS 2 HLT-15 ATCC RPMI 1640 medium +10%FBS 2 HOSE 6-3 (A kind gift from Dr George Tsao, Medium 199:MCDB105(50:50) +10%FBS University of Hong Kong) Lovo ATCC Ham’s F12K medium + 10% FBS CaOV3 ATCC DMEM + 10%FBS LnCAP ATCC RPMI 1640 medium +10%FBS 2 Du-145 ATCC RPMI 1640 medium +10%FBS 2 HPAC ATCC DEMEM/Ham’s F12 (50:50) + HEPES (N- [2-hydroxyethyl]piperizine-N-[2- ethanesulfonic acid], pH 7.2, 15mM) + insulin (2μg/ml) + transferrin (5μg/ml) + hydrocortisone (40 μg/ml) + EGF (10ng/ml) +5%FBS CAPAN-2 ATCC McCoy’s 5A + 10% FBS Panc-1 ATCC DMEM + 1mM Na Pyruvate + 10% FBS Bx-PC3 ATCC RPMI 1640 medium +10%FBS MiaPaCa-3 ATCC DMEM + 1mM Na Pyruvate + 10% FBS +2.5% Horse Serum AsPC ATCC RPMI 1640 +20%FBS + 1mM Na Pryuvate HMEC 219-4 MCDB 170 medium +Bovine Pituatary Extract HEC-1-A ATCC McCoy’s 5A medium+ 10% FBS
1 EMEM supplemented with 2 mM L-glutamine, 0.1 mM non-essential amino acids, 1 mM sodium pyruvate, 2 RPMI medium contains HEPES (N-[2-hydroxyethyl]piperizine-N-[2-ethanesulfonic acid], pH 7.2, 20mM),sodium bicarbonate (14mM), L-glutamine (6mM) 47
All cell culture medium were obtained from Invitrogen (VIC, Australia) with the exception of FBS (Foetal Bovine Serum) and Horse serum (CSL Diagnostics, N.S.W, Australia) and penicillin (Pharmacia/Pfizer, NY, USA). Tissue culture flasks were obtained from Corning (N.S.W, Australia).
RNA extraction.
Cell line RNA extraction, was performed by Gillian Lerbach at the Garvan Institute of Medical Research. Total RNA was extracted from the following human cell lines using the RNeasy Maxi Kit (Qiagen); according to manufacturers instructions: breast cancer cell lines MDA-MB-468, BT-20, MCF-7, SK-Br-3, ZR-75-1, normal breast epithelium 184, HMEC-219-4, MCF-10A; normal ovary HOSE 6-3; ovarian cancer CaOV-3; normal uterus HEC-1-A LA; head and neck cancer FaDu, Detroit; prostate cancer PC-3, DU-145, Ln CaP; normal colon LoVo; colon cancer HCT-15 and pancreatic cancer Panc-1, Capan, HPAC, MiaPaCa-2, Bx-PC3, AsPC-1.
Protein Extraction
Cells were lysed as follows: cell monolayers were washed twice with ice-cold phosphate-buffered saline and then scraped into ice-cold “normal“ lysis buffer (0.5% deoxycholate, 150 mM NaCl, 1% sodium pyrophosphate, 50 mM Tris pH 8.0, 0.1% SDS, 10% glycerol, 5 mM EDTA, 20 mM NaF, 10 μg/ml apoprotinin, 10 μg/ml leupeptin, 1 mM phenylmethylsufonyl fluoride, 200 μM sodium orthovanadate) or Sherr lysis buffer (50 mM HEPES (pH 7.5), 1 mM dithiothreitol, 150 mM NaCl, 10% (v/v) glycerol, 0.1% Tween-20, 1 mM EDTA, 2.5 mM EGTA, 10 mM b- glycerophosphate, 10 g/ml aprotinin, 10 g/ml leupeptin, 1 mM phenylmethylsulfonyl fluoride, 0.1 mM sodium orthovanadate, 1 mM NaF). Cell suspensions in lysis buffer were incubated for five minutes on ice, and the cellular debris was cleared by centrifugation (13 000 rpm, five minutes, 4°C). The cleared lysates were stored at -80°C 48
Semi-Quantitative Polymerase Chain Reactions
HOXB2
Polymerase Chain Reaction was performed using the Roche Titan One Tube RT-PCR System (Roche Diagnostics, Mannheim, Germany) according to manufacturers instructions.
PROBE FORWARD REVERSE HOXB2 5’-ATC CGC CAC GGT TCC TCC TC-3’ 5’-GTG GCC GGC ACA GGT ACT TA-3’
Breifly the probe primer described above was diluted with sterile water to a concentration of 100 pmol/μL. 10 μL of PCR grade DNTP was combined with 1ng of downstream and upstream primers respectively. 100pg of template RNA, 2.5 μLof DTT solution 100mM and 10 Units of RNase Inhibitor. This was admixed with a solution of 1μL of TITAN enzyme mix, 10 μL of 5x RT-PCR buffer and 11 μLof sterile water. The sample was placed in a thermocycler equilibrated at 50 C for 30 min. The template was denatured at 94 C for 2 min. Followed by 10 cycles of denaturation at 94Cfor10s,annealingat58Cfor30sandelongationat68Cfor45s.25cyclesof denaturation at 94C for 10s, annealing at 58 C for 30 s and elongation at 68C for 45s with cycle elongation of 5s for each subsequent cycle. The product was then visualised on a 1% agarose gel 49
Retinoic Acid Signalling Components
Polymerase Chain Reaction was performed using the Roche Titan One Tube RT-PCR System (Roche Diagnostics, Mannheim, Germany) according to manufacturers instructions.
PROBE FORWARD REVERSE RAR 5'- AAA CAA GAA GAA GAA GGA GGT GCC -3' 5'- CCA GAG GTC AAT GTC CAG AGA GAC -3'
RAR 5'- TCC AGA AGT GCT TTG AAG TGG G -3' 5'- GTG AAC ACA AGG TCA GTC AGA GGA C - 3' RAR 5'- ACC GCG ACA AAA ACT GTA TCA TC -3' 5'- CCT TCA CCT CTT TCT TCT TCT TGT TC -3'
RXR 5'- AAG GAC CGG AAC GAG AAT GA -3' 5'- ATC CTC TCC ACC GGC ATG T -3' RXR 5'- CTC TGG ATG ATC AGG TCA TAT TGC T-3' 5'- GCC ATC TCG AAC ATC AAT GGA -3'
RXR 5'- GGG AAG CTG TGC AAG AAG AAA -3' 5'- TGG TAG CAC ATT CTG CCT CAC T -3' CRABP-1 5'- TTG TGG CCA AAC TGG CTC CA -3' 5'- ACA CTG GAG CTT GTC TCC GT -3'
Breifly all probe primer described above were diluted with sterile water to a concentration of 100 pmol/μL. 10 μL of PCR grade DNTP was combined with 1 ng of downstream and upstream primers respectively. 100 pg of template RNA, 2.5 μLof DTT solution 100mM and 10 Units of RNase Inhibitor. This was admixed with a solution of 1μL of TITAN enzyme mix, 10 μL of 5x RT-PCR buffer and 11 μLof sterile water. The sample was placed in a thermocycler equilibrated at 50 C for 30 min. The template was denatured at 94 C for 2 min. Followed by 10 cycles of denaturation at 94 C for 10s, annealing at 56 C for 30 s and elongation at 68 C for 45 s. The reaction was completed with 40 cycles of denaturation at 94C for 10s, annealing at 56 C for 30 s and elongation at 68C for 45s with cycle elongation of 5s for each subsequent cycle. Samples (5 μL) were taken at cycle 25, cycle 30 and cycle 40. The resultant products were then visualised on a 1% agarose gel 50
Western Blotting
For Western blot analysis, protein quantitation was carried out using the Bio-Rad (CA, USA) Protein Assay kit according to manufacturer’s instructions. Following normalization of protein concentration, SDS sample buffer (final concentrations: 0.0625 MTrispH6.8,5%(v/v) -mercaptoethanol, 3% (w/v) SDS, 10% (v/v glycerol) was added and the lysates were heated to 95°C for 3 min, protein was separated by 10% polyacrylamide gel and wet transferred to nitrocellulose membranes (Millipore, N.S.W. Australia). Non-specific binding was blocked in 10% (w/v) skim milk powder in TBS/TWEEN 20 (10mM Tris pH 7.4, 150mM NaCl, 0.1% (v/v) TWEEN 20). Membranes were incubated (2 h at room temperature or overnight at 4°C) with the following primary antibodies diluted in TBS/BSA solution (10mM Tris pH 7.4, 150mM NaCl, 5% w/v BSA, 0.02% azide) directed against HOXB2 (p20) (Santa Cruz Biotechnology, Santa Cruz, CA) 1:1000 and actin (AC-15; Sigma, St. Louis, MO) 1:20000. Membranes were washed extensively in TBS/TWEEN solution then incubated for 1 h at room temperature with horseradish peroxidase-conjugated anti-goat or anti- rabbit secondary antibody (Santa Cruz Biotechnology, Santa Cruz, California) in 5% (w/v) skim milk powder in TBS/TWEEN. Membranes were washed extensively in TBS/TWEEN solution before the proteins were visualized using the enhanced chemoluminescence (ECL) detection system (Perkin Elmer, MA, USA) on Xray film. 51
SiRNA transfection
Panc –1 cells were plated onto a 6 well plate in the following concentration of 1 x 105 cells to a volume of 1.5 ml. This was allowed to incubate overnight and cells were confirmed to be 50-70% confluent prior to transfection. Commercially available SiRNA was used to effect downregulation.
SiRNA SiGENOME duplex #4 D-011695-04, HOXB2
Sense Sequence C.A.A.G.G.A.G.U.C.G.A.C.A.U.U.A.A.U.U.U.U. Antisense Sequence 5’-P.A.A.U.U.A.A.U.G.U.C.G.A.C.U.C.C.U.U.G.U.U.
10 μL of 20nmol/μL SiRNA SiGENOME duplex #4 D-011695-04, HOXB2 (Dharmacon, Lafayette, CO) was admixed to 240μL OptiMEM (Invitrogen, Rockville MA) for 5 mins. This was then added to a solution containing 5 μL of lipofectamine 2000 (Invitrogen) and 245μL of OptiMEM (Invitrogen). This was allowed to incubate for 20 mins prior to transfection. 500 μL was added to each well giving a final concentration of siRNA of 200pmol. Media was replenished 24 hours after transfection. Protein was sampled according to the protocol described below 72 hours after transfection to confirm adequate knock down of HOXB2. Transfections were also performed using, no siRNA and a RISC-free SiRNA (Dharmacon) as a negative control and a positive control SiRNA to Laminin A/C (Dharmacon). 52
Cell proliferation assays
Proliferation rate, was initially assessed by manual counting of cells on repeated days. Clones were plated out at a density of 1.0x106 cells per 25 cm2 flask. Cell numbers were estimated in triplicate using a haematocytometer and this was repeated daily until the cells became confluent. Proliferation rates were calculated using the equation: Doubling time = ln 2/k, where k is the constant of the slope of the exponential growth phase.
Proliferation rate was also assessed with the Cell Titer 96® kit (Promega) with 3-(4,5- dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium with an electron coupling agent, phenazine ethosulfate, to estimate total cell number. Each cell line/clone was plated onto 96-well plates at variable cell densities, (with a minimum of 6 replicates per plating density) to a total volume of 100 L/well. On days 1-7, 20 L/well of Cell Titer 96® assay solution was added to the plates which were 0 subsequently incubated for 4 h at 37 C in a humidified 5% CO2 atmosphere. Plates were then directly assayed at an absorbance of 490 nm using an ELISA plate reader. Background absorbance was corrected for by subtracting the average 490 nm absorbance for wells containing medium alone from all other absorbance values. These experiments were duplicated for each transfection. 53
Chapter 3
IDENTIFICATION of NOVEL GENES in the DEVELOPMENT and PROGRESSION of PANCREATIC CANCER by TRANSCRIPT PROFILING 54
Introduction
The prognosis for any patient diagnosed with PC is poor. Despite improvements in therapeutics, the 5-year survival for PC remains less than 5%. PC presents as a clinically advanced disease, as a result only 10-20% of patients are suitable for surgical treatment at the time of presentation 154. For the majority of patients who are unsuitable for operative resection, non-operative approaches to the treatment of PC have met with little success and achieved only limited improvement in the overall mortality of the disease. There remains no rationally designed, molecularly targeted therapies for PC treatment, as have been implemented for non-Hodgkin’s lymphoma 150 and breast cancer 78 in the form of manipulation of receptors, specific monoclonal antibodies that target tumour-specific cell surface receptors 151 and tyrosine kinase inhibitors that inhibit growth factor receptor signalling in the case of gastrointestinal stromal tumours 152. This highlights the need for the identification of novel markers of PC outcome and therapeutic responsiveness, and the development of innovative therapies that are designed based on molecular targets. This chapter deals with the identification of novel candidate genes of relevance to PC by transcript profiling using oligonucleotide microarrays.
RNA-based gene expression profiling techniques are now well established in the study of disease, in particular cancer. Identification of differentially expressed genes whilst interrogating global gene expression allows for the more accurate and timely identification of genes potentially important in cancer. Similar approaches in controlled experiments using cell lines and animal models also facilitates the focus of ongoing studies into the functional characterisation of genes and molecular pathways which regulate cell processes. Several techniques of interrogating global gene expression are currently in use: serial analysis of gene expression (SAGE), cDNA (spotted) microarrays and oligonucleotide microarrays. SAGE is a technique that maps sequences of recombined linear cDNA that has been produced from mRNA extracted from tissue to identify sequences 10 bases in length that are unique to particular genes (SAGE Tags). cDNA microarrays are produced by "spotting" microscopic quantities of unique cDNA's onto a glass slide to which labelled complementary cDNA is prepared from sample RNA and hybridised to the array. The intensity is representative of the mRNA 55 expression for a particular gene. Recent technological advancement has lead to the commercial availability of oligonucleotide microarrays, which allow the researcher to interrogate larger numbers of genes. A technique similar to electronic microchip preparation activates a silicon wafer in a grid pattern and oligonucleotide sequences, usually 25 bases in length, are constructed directly on the surface of the silicon membrane. Both matched and mismatched sequences for specific genes allow for measurement of expression after hybridisation of biotinylated cRNA, which is produced, from mRNA via cDNA.
The Affymetrix Genechip® oligonucleotide microarray, HG-U133 is composed of 2 arrays (U113A and U133B) and covers 44 929 transcripts which in turn represent 33 000 of the best characterised human genes. Sequences used in the design of the array were selected from:
• GenBank (http://www.ncbi.nlm.nih.gov/), • dbEST (http://www.ncbi.nlm.nih.gov/dbEST/) • RefSeq (http://www.ncbi.nlm.nih.gov/LocusLink/refseq.html). • Sequence clusters were from build 133 (April 20, 2001) of UniGene (http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=unigene) and refined by analysis and comparison with a number of other publicly available databases.
In this study transcript profiles of 12 specimens of PC were compared to 6 specimens of morphological normal pancreas, which were taken as distant as possible from the tumour contained within the pancreata resected for PC (1 specimen of PC was not included in the final analysis due to the quality of the extracted RNA not meeting the recommended criteria). Seventeen samples of pancreatic tissue satisfied the Affymetrix recommended criteria for hybridisation as assessed by Affymetrix Genechip® Test arrays. The background staining was below 150 units, with comparable scaling factors and 3' to 5' ratios for transcripts of GAPDH within acceptable limits. All samples were hybridised to Affymetrix Genechip® HG-U133 A and B oligonucleotide microarrays. 56
Validation
A global analysis of differential gene expression using the data generated from the Affymetrix Genechip® HG-U133 array was initially performed to determine if the acquired data demonstrated expected differential expression of gene transcripts based on current knowledge of the cellular composition of normal pancreas. Initial analysis using linear statistical methods, self-organising maps and hierarchical clustering (http://www.biostat.harvard.edu/complab/dchip/) of the data, demonstrated significant expression of pancreatic exocrine-specific genes in samples of exocrine pancreas (Elastase, Lipase etc.). One analysis filtering for genes exhibiting the largest differential expression (mean/standard deviation) between normal pancreatic tissue and PC is presented in Figure 3.1, which demonstrates upregulation of exocrine-specific genes in samples of normal exocrine pancreas. Genes upregulated in PC relative to normal are also demonstrated and constitute genes involved in the development of fibrous tissue eg. collagen; markers of ductal epithelium also expressed at high levels in PC, eg. keratin 19; immune response related genes, eg. interleukin-8; genes known to be upregulated in PC, eg. prostate stem cell antigen (PSCA) and potential novel upregulated genes in PC relevant to this study. Clustering analysis using self-organising maps (Genecluster 2 software, http://www- genome.wi.mit.edu/cancer/software/genecluster2/gc2.html) and principal components analysis (http://www.jmp.com/homepage.shtml) also revealed a similar profile. These findings validate the technique as a reliable representation of the relative amplitudes of differential mRNA expression.
The custom designed relational database described in Chapter 2 incorporating experimental data, other relevant data (eg. SAGE) and web based resources (eg: Ontogeny, Gene annotations, SwissProt descriptions) was interrogated for individual genes and pathways to identify genes potentially involved in PC for further study. Queries of the database were made based on the degree of differential expression between normal pancreas and PC, statistical significance (t-test and Mann-Whitney U test), the number of SAGE tags present in PC, normal pancreatic duct cell cultures (HX, H126) and tissues from other organs. Increased Expression Diminished Expression
Contamination of PC sample with Normal Normal Pancreas Pancreatic Cancer Exocrine Pancreas
Exocrine Pancreas Specific Genes
Fibrosis Related Genes Immune Response Associated Genes Pancreatic Duct Epithelium Specific Genes Potential Novel Genes Relevant to PC (upregulated)
Figure 3.1: Hierachical Clustering of transcript expression levels in Pancreatic Cancer and Normal Pancreas for 218 unique genes filtered based on magnitude of differential expression from the original complement of 22 283 genes interrogated by the Affymetrix Genechip® HG-U133A oligonucleotide microarray. 58
Results
The HG-U133 PC transcript profile database was used to identify genes that were differentially expressed in PC compared to normal pancreas. Using a criteria of a fold change of <0.5 or >2.0, and a p value on paired t test and/or Mann-Whitney U test of < 0.05. Nine hundred and fifty four genes were identified to be overexpressed (fold change >2 between PC and normal pancreas, (Appendix 3), of these 269 (28%) genes were overexpressed in PC compared to normal pancreas, with a fold change of > 5 (Table 3.1). Eight hundred and thirty three genes were downregulated (fold change <0.5 between PC and normal pancreas) (Appendix 4) of which 75 genes were significantly downregulated, with a fold change < 0.2 between PC and normal pancreas. In order to investigate the relationship between the various grades of PC and normal pancreas, a comparison was made between the genes expressed at higher and lower levels in PC compared to normal pancreas in each group of poorly and moderately differentiated PC. Comparing moderately differentiated PC with normal pancreas 933 genes (97.8%) were upregulated and 703 genes (84.4%) were downregulated. When comparing poorly differentiated PC with normal pancreas 813 genes (87.1%) were upregulated, whilst 710 genes (85.2%) were downregulated. In addition, 283 (29.7%) genes were upregulated and 298 (35.8%) genes downregulated when poorly differentiated PC was compared with moderately differentiated PC.
Initial queries of the database revealed upregulation of genes that have been reported to be important in PC based on traditional techniques (eg. cyclin D1, uPAR) and those published recently using other techniques of global gene expression analysis. Many genes that were upregulated in these published studies were also apparent in the experimental data generated in this thesis. Genes identified as overexpressed using SAGE such as keratin 18, keratin 19, S100 Calcium binding protein A10 and retinoic acid induced 3 155 were also upregulated in this study. 59
Other genes identified by SAGE and validated as overexpressed in PC using ISH or IHC (Prostate stem cell antigen 156, mesothelin 157) were upregulated in this study as were others found to be upregulated using cDNA microarrays in PC cell lines (Annexin 1 158), FNA epithelial enriched PC (wnt-5A, TIMP-1 159), and microdiseected PC (MMP-7, SOD1 160). A recent publication of global gene expression in PC using oligonucleotide microarrays (Affymetrix HG-U95) 161 reported upregulation of genes that were also evident in this study (CD44, Claudin 1, Fibronectin 1, Annexin 1, MMP- 7, uPA, uPAR, retinoic acid induced 3, S100 calcium binding protein P, TIMP-1), some of which were confirmed as overexpressed in the epithelial component of PC in a subsequent publication using IHC (heat shock protein 47) 162. These studies support the integrity of the data generated through the approach used in this study.
Table 3.1: Distribution of genes differentially expressed in pancreatic adenocarcinoma (moderate and poor differentiation) compared with normal pancreas.
Genes Genes overexpressed in underexpressed in PC PC >2 <0.5 >5 <0.2 All Cancer versus All Normal 954 833 Pancreata 269 28.2% 75 9%
Moderately Differentiated PC 933 97.7% 703 84.4% versus Normal Pancreata
Poorly Differentiated PC 813 85.2% 710 85.2% versus Normal Pancreata
Poorly Differentiated PC 283 29.7% 298 35.8% versus Moderately Differentiated PC
A full list of genes is presented in appendices 3 and 4 60
Whilst others have analysed transcript profile data by investigating candidate genes identified from a priority list of genes compiled from linear analyses as described above, we chose to harness the relational nature of the database and examine altered gene expression in PC compared to normal pancreas in known molecular pathways and functionally related molecules, in order to identify pathways and/or molecular families with significant components aberrantly expressed in PC compared to normal pancreas. Hence, the data were examined using pre-assembled pathway maps contained within GenMAPP software (http://www.GenMAPP.org/default.html). These were then incorporated in the database and a refined query identified aberrant expression in multiple components of known molecular pathways, in PC compared to normal pancreas. Pathways assessed included cell cycle related genes, MAP kinase cascade, the TGF- signalling pathway, the wnt signalling pathway and the retinoic acid signalling pathway. These pathways contained components that had the largest number of aberrantly expressed genes in PC compared to normal pancreas and some are presented here.
The TGF- signalling pathway
The TGF- growth inhibitory pathway utilises Smad proteins 1 to 8. Characterisation of Smad functions has segregated them into 3 groups: receptor-regulated Smads (Smad1, 2, 3, 5 and 8), the common partner Smad (Smad4) and inhibitory Smads (Smad6 and 7) 163. Ligand induced TGF- receptor activation results in the formation of heterodimeric complexes of Smad2/Smad3 with Smad4 leading to the translocation of these complexes into the nucleus where Smad4 activates transcription of cell cycle inhibitory factors p21WAF1/CIP1 164,165 and p15INK4B 166. Aberrant TGF- signalling is described in PC predominantly as a result of loss of DPC4/Smad4 expression. 61
The transcript profile data (Figure 3.2) identified increased expression of several key Smad molecules including Smad3 (Fold Change = 2.3), Smad7 (Fold Change = 2.7) and Smad4 (Fold Change = 2.4). Homozygous deletion or mutation of DPC4 (MADH4),the gene encoding Smad4, has been reported in 55% of pancreatic ductal adenocarcinomas 86. Loss of Smad4 expression co-segregates with resectability and is associated with improved survival in PC 67. In addition, Smad interacting protein 1 (Sip-1) (Fold Change = 2.3) is associated with the downregulation of E-cadherin in epithelial cancers 167 which increases the metastatic potential of PC cell lines. Furthermore, Sip-1 upregulates matrix metalloproteinases (MMP), several of which were upregulated in the transcript profile data presented here. MMP’s increase tumour invasiveness in hepatocellular carcinoma 168. Recent evidence suggests Sip-1 associated repression of E-cadherin was associated with poor prognosis in breast cancer 169.
Aberrant expression of extracellular signal-regulated kinases (ERK1 and ERK2) occurs in 50% of PC and is associated with expression of K-ras 170. Increased expression of ERK1 (Fold Change = 2.5) and subsequent activation of the ERK pathway protects PC cell lines from apoptosis and increases their progression through the cell cycle 171. Bone Morphologic Protein 2 (BMP-2) (Fold Change = 5.6) was investigated in PC and both BMP2 and its receptor were overexpressed in PC compared to normal pancreas, BMP2 expression correlated with a shorter post-operative survival in PC. Additionally BMP2 stimulated the growth of PC cell lines, suggesting that BMP2 has the capacity to act as a mitogen and plays an important role in the development of PC 172.AML2andAML3 are transcriptional factors associated with the development of acute myeloid leukaemia 173, the forced expression of AML3 (Runx2) in breast and prostate cancer cell lines is associated with the upregulation of matrix metalloproteinases and a more invasive cellular phenotype. 174. The heterodimeric fibronectin receptor integrin alpha (v) beta6 (Fold Change = 4.1) is associated with cell migration, invasion and protease expression in oral squamous carcinoma 175, with increased expression associated with lymph node positive gastric cancer 176. Ligands
β Integrin α V β 6 4.1 Accessory receptors TGF LAP 0.6 β TGF Activin Follistatin Thrombospondin 1.3 β - glycan BMP 5.6 Noggin Endoglin
1.6 1.6 TβR-II TβR-II TβR-1 1.3 SARA 0.5 Osteopontin 1.9 TβR-1 1.3 Repressors SIP 1 2.3} RSmads AML3 2.6
Smad 2 AML2 2.2 FKBP12 1.3 Smad 1 0.8 Smad 5 1.5 Smad 3 2.3 BAMBI Wnt Ligands β Catenin 1.7 TCF1 1.2 Smad 8 Co-Factors 2.4 LIF 1.7 JAK-1 1.1 STAT3 0.6 Smad 6 0.5 Smad 4 TNF α JNK2 0.6 JUN TFE3 0.9 } Smad 7 2.7 FAST 2.6
IFN STAT 1 0.5 TGF β Smad 2 Co-Smads 0.8 2.5 BMP 5.8 Smad 1 ERK 1 P300 1.1 Co Factor CBP 1.2 TNF α NF kappa β 0.6 R-Smads
TGIF 1.9 PAI-1 1.7 EGF 0.2 RAS SNO-N 1.8 C-FOS SKI 1.4
Figure 3.2: A customized GenMAPP of TGF-β Signalling components; with statistically significant relative expression levels in PC compared to normal pancreas denoted beside the gene. Relative expression levels are represented as fold change with those with statistically significant upregulation marked red, those with significant downregulation marked blue and those that statistically did not fulfil the criteria of fold change >2 or < 0.5 are marked green. Those components statistically unchanged remain clear. 63
Signal transducers and activators of transcription (STAT-1) proteins (Fold Change = 2.1) translocate to the nucleus where they bind to DNA consensus site and upregulate the transcription of target proteins. Interferon induces STAT levels in pancreatic carcinoid cells. Additionally STAT1 and retinoic acid receptors have been implicated as secondary effectors in the induction of transcription of MUC4 by interferon and retinoic acid 177. Interestingly MUC4 is also upregulated in the presented transcript profiling data.
The incorporation of transcript profile data with relational databases permitting a visualisation of lesions in known molecular pathways in PC identified aberrant expression of several components of TGF- signalling. Although components have been associated with cancers of the pancreas and an increased invasive phenotype in other cancer, further investigation of TGF- signalling was not pursued in this thesis.
Wnt Signalling
Activation of wnt signalling results in the stabilisation of -catenin, which is then able to translocate into the nucleus and initiate transcription of genes involved in cell growth and proliferation. Wnt signalling is initiated by secreted wnt proteins which bind to a series of seven pass transmembrane cell surface receptors encoded by frizzled genes. Ligand binding of the frizzled receptors leads to phosphorylation of dishevelled (DVL) which in turn forms a complex with Axin and glycogen synthase kinase 3 (GSK3 ) to prevent phosphorylation of -catenin 178. Phosphorylation of -catenin is essential for ubiquitination and proteosome degradation. Lack of phosphorylation leads to the stabilisation of -catenin, nuclear translocation and subsequent inactivation of transcriptional repressors, allowing for the increased transcription of genes involved in growth and proliferation, including cyclin D1/D2/D3 and PPAR 178. Analysis of mRNA expression of components of the wnt signalling pathway demonstrated aberrant expression of a significant number of molecules. Figure 3.3 illustrates a GenMAPP representing aberrant expression of wnt signalling components in PC. Aberrantly expressed components of the wnt signalling pathway genes, in mammalian species, are presented in Table 3.2. Inhibitors of wnt signalling
sFRP1 0.6 sFRP3 2.9 sFRP5 0.3 DKK-1 4.1 sFRP2 ? sFRP4 6.7 WIF-1 0.4 DKK-4 1.8
? Frizzled Ligands
Frizzled 1 1.7 wnt-1 1.3 wnt-3a 2.1 wnt-6 3.9 wnt-10B 0.8 Frizzled 2 4.2 wnt-2 3.1 wnt-4 3.1 wnt-7A 1.4 wnt-11 2.6 Frizzled 3 1.1 wnt-2B 0.7 wnt-5A 2.4 wnt-7B 1.3 wnt-15 ? Frizzled 4 0.7 wnt-3 2.1 wnt-5B 2.6 wnt-8B 2.5 wnt-16 1.4 Frizzled 5 1.3 Frizzled 6 1.3 LRP 5 1.2 Frizzled 7 2.1 LRP 6 1.3 Frizzled 8 0.9 Frizzled 9 2.0 Frizzled 10 0.9 Cytoskeleton
PKC alpha 1.9 Dsh 1 0.6 PKC beta ? Rho A ? JNK 2 0.6 Dsh 2 0.8 PKC gamma 1.3 Rac-1 2.0 JNK 3 1.0 Dsh 3 1.3 PKC delta 0.7 PKC epsilon 0.4 PKC iota 1.3 PP2A su E 1.1 PKC eta 0.8 GBP/FRAT 0.8 PP2A su G 2.0 Apoptosis PKC mu 2.2 PKC theta 0.6 IDAX 0.3 PKC zeta 0.6 Axin 1.4
Casein Kinase E 0.7
GSK3β 2.4 APC 1.0 P
P Recruitment of unknown β-catenin 2.0 β-catenin conjugating enzyme Skp-1 2.2 β-trcp 1.6
Nucleus Ub TC1/LEF 1.6 Ub Ub Ub DPC4/Smad4 1.0 Ub β-catenin SOX-3 0.8 Ub cyclin D1 2.4 Ub ICAT 0.5 Ub cyclin D2 2.4 Ub Ub cyclin D3 1.1 PPARδ 1.1 fra-1 2.8 MMP-7 4.2 fibronectin 12 c-myc 0.3 MMP-26 0.7 retinoic acid receptor γ 2.2 c-jun ? Id-2 1.0 26S Proteosome COX-2 4.0 uPA-R 13.5 WISP 3.0 Degradation
Figure 3.3: Alterations in expression in PC compared to normal pancreas of genes involved in wnt signalling. Red denotes upregulation of ≥ 2, blue - downregulation ≤ 0.5, green - fold change of 0.51 to 1.9 and white - no data. Values adjacent to each represented gene is the fold change of mRNA expression obtained from HG-U133 Affymetrix array data for PC relative to normal pancreas. 65
Table 3.2: Expression of transcripts in PC compared to normal pancreas Note upregulation of wnt receptors, wnt ligands and -Catenin itself as well as downregulation of some inhibitors of wnt signalling (IDAX and ICAT), together with genes, the transcription of which is mediated by -Catenin, suggests activation of wnt signalling in PC.
Unigene Fold ttest Affymetrix Probe Cluster Gene Change P value 204451_at Hs.94234 frizzled (Drosophila) homolog 1 (FZD1) 1.7 0.047 210220_at Hs.81217 frizzled (Drosophila) homolog 2 (FZD2) 4.2 0.019 219683_at Hs.40735 frizzled (Drosophila) homolog 3 (FZD3) 1.1 0.777 218665_at Hs.19545 frizzled (Drosophila) homolog 4 (FZD4) 0.7 0.166 206136_at Hs.152251 frizzled (Drosophila) homolog 5 (FZD5) 1.3 0.569 203987_at Hs.114218 frizzled (Drosophila) homolog 6 (FZD6) 1.3 0.207 203705_s_at Hs.173859 frizzled (Drosophila) homolog 7 (FZD7) 2.1 0.004 216587_s_at frizzled (Drosophila) homolog 8 (FZD8) 0.9 0.701 207639_at Hs.158335 frizzled (Drosophila) homolog 9 (FZD9) 2 0.101 219764_at Hs.31664 frizzled (Drosophila) homolog 10 (FZD10) 0.9 0.829 209468_at Hs.6347 Lipoprotein Receptor Related Protein 5 (LRP5) 1.2 0.682 34697_at Hs.23672 Lipoprotein Receptor Related Protein 6 (LRP6) 1.3 0.507 208570_at Hs.248164 WNT-1 1.3 0.599 205648_at Hs.89791 WNT-2 3.1 0.005 206458_s_at Hs.258575 WNT-2B 0.7 0.314 221455_s_at Hs.224667 WNT-3 2.1 0.199 208606_s_at Hs.302428 WNT-4 3.1 0.138 213425_at Hs.152213 WNT-5A 2.4 0.023 221029_s_at Hs.306051 WNT-5B 2.6 0.007 222086_s_at Hs.29764 WNT-6 3.9 0.332 217681_at Hs.65905 WNT-7A 1.3 0.597 207612_at Hs.137595 WNT-8B 2.5 0.303 206213_at Hs.91985 WNT-10B 0.8 0.493 206737_at Hs.108219 WNT-11 2.6 0.128 221113_s_at Hs.272375 WNT-16 1.4 0.406 204602_at Hs.40499 dickkopf (Xenopus laevis) homolog 1 (DKK1) 4.1 0.057 219908_at Hs.211869 dickkopf (Xenopus laevis) homolog 2 (DKK2) 3.1 0.069 202196_s_at Hs.4909 dickkopf (Xenopus laevis) homolog 3 (DKK3) 4.3 <0.0001 206619_at Hs.159311 dickkopf (Xenopus laevis) homolog 4 (DKK4) 1.8 0.321 202035_s_at Hs.7306 secreted frizzled-related protein 1 (sFRP1) 0.6 0.149 223122_s_at Hs.31386 secreted frizzled-related protein 2 (sFRP2) 6.5 0.007 203697_at Hs.153684 secreted frizzled-related protein 3 (sFRP3) 2.9 0.004 204052_s_at Hs.105700 secreted frizzled-related protein 4 (sFRP4) 6.7 0.042 207468_s_at Hs.279565 secreted frizzled-related protein 5 (sFRP5) 0.3 <0.0001 214633_at Hs.348820 SOX 3 0.8 0.738 204712_at Hs.284122 Wnt Inhibitory Factor 1 (WIF-1) 0.4 0.048 201533_at Hs.171271 -Catenin (CTNNB1) 2 0.154 202332_at Hs.79658 casein kinase 1, epsilon (CSNK1E) 0.7 0.184 203230_at Hs.74375 dishevelled 1 (homologous to Drosophila dsh) (DVL1) 0.6 <0.0001 218759_at Hs.118640 dishevelled 2 (homologous to Drosophila dsh) (DVL2) 0.8 0.597 201908_at Hs.174044 dishevelled 3 (homologous to Drosophila dsh) (DVL3) 1.3 0.059 66
219889_at Hs.126057 GSK-3 binding protein FRAT1 0.6 0.013 209864_at Hs.140720 GSK-3 binding protein FRAT2 1 0.948 203338_at Hs.173328 protein phosphatase 2A, epsilon 1.1 0.594 214083_at Hs.171734 protein phosphatase 2A, gamma 2 0.098 210511_at Hs.73888 transcription factor 1 (TCF1) 0.3 0.032 210776_x_at Hs.101047 transcription factor 3 (TCF3) 3.1 0.005 203753_at Hs.326198 transcription factor 4 (TCF4) 3.9 0.03 221558_s_at Hs.44865 lymphoid enhancer factor 1 (LEF1) 9.3 0.001 203526_s_at Hs.75081 Adenomatos Polyposis Coli (APC) 1 0.916 212849_at Hs.184434 Axin 1.4 0.185 204901_at Hs.226434 beta-transducin repeat containing protein (b-trcp) 1.4 0.223 220277_at Hs.118569 IDAX (inhibition of the Dvl and Axin complex) 0.3 0.032 209945_s_at Hs.78802 glycogen synthase kinase 3 beta (GSK-3 ) 2.4 0.01 203081_at Hs.99816 ICAT 0.5 0.011 207974_s_at Hs.227950 S-phase kinase-associated protein 1A (p19A) (SKP1A) 1.6 0.045 204489_s_at Hs.169610 CD44 antigen (CD44) 1.7 0.068 218182_s_at Hs.7327 claudin 1 1.2 0.609 204748_at Hs.196384 Cyclo-oxygenase 2 (COX-2) 4 0.007 214019_at Hs.82932 cyclin D1 2.4 0.305 200952_s_at Hs.75586 cyclin D2 2.4 0.242 201700_at Hs.83173 cyclin D3 1.1 0.653 210495_x_at Hs.287820 fibronectin 1 12 <0.0001 204948_s_at Hs.9914 follistatin 0.5 0.009 201566_x_at Hs.180919 Id-2 1 0.896 204259_at Hs.2256 matrix metalloproteinase 7 (MMP7) 4.2 0.014 220541_at Hs.204732 matrix metalloproteinase 26 (MMP26) 0.7 0.221 208044_s_at Hs.106415 peroxisome proliferative activated receptor,delta 1.1 0.791 204188_s_at Hs.1497 retinoic acid receptor gamma (RAR ) 2.2 0.059 211924_s_at Hs.179657 Urokinase Plasminogen Activator Receptor (uPA-R) 13.5 0.037 202431_s_at Hs.79070 c-myc 0.3 0.015 206796_at Hs.194680 WISP1 3.3 0.009 205792_at Hs.194679 WISP2 0.7 0.561 210861_s_at Hs.194678 WISP 3 (LIBC) 4.6 0.037 206923_at Hs.169449 protein kinase C, alpha (PRKCA), 1.9 0.103 206270_at Hs.2890 protein kinase C, gamma (PRKCG) 1.3 0.649 202545_at Hs.155342 protein kinase C, delta (PRKCD) 0.7 0.194 206248_at Hs.211592 protein kinase C, epsilon (PRKCE) 0.4 0.013 206099_at Hs.315366 protein kinase C, eta (PRKCH) 0.8 0.461 209677_at Hs.1904 protein kinase C iota isoform (PRKCI) 1.3 0.42 205880_at Hs.2891 protein kinase C, mu (PRKCM) 2.2 0.046 210039_s_at Hs.211593 protein kinase C, theta 0.6 0.054 202178_at Hs.78793 protein kinase C, zeta (PRKCZ), 0.6 0.05 208640_at Hs.173737 rho family, small GTP binding protein Rac-1 2 0.005 67
Recent publications assessing global mRNA expression in PC also report upregulation of wnt signalling components and products of -catenin-mediated transcription. A publication using the Affymetrix platform in PC (HG-U95 microarrays) reported upregulation of wnt target genes CD44, Claudin 1, Fibronectin 1, MMP7 and uPAR 161 , which is consistent with this study. The most convincing data to suggest augmented wnt signalling is the upregulation of transcripts known to be mediated by stabilisation and nuclear localisation of -catenin due to wnt activation. Many of these were upregulated in this study including those that have been identified to be relevant to PC in previous publications including uPAR, cyclo-oxygenase 2 and cyclin D1.
Wnt signalling was not investigated further in this body of work, however validation of aberrant wnt signalling components in PC were, investigated by other members of the laboratory. Unpublished data complements the identified role of altered wnt signalling in PC. Loss of sFRP4 membranous expression was identified in 60% of cancers and associated with a poor outcome. In addition studies of -catenin subcellular localisation and cyclin D1 expression suggests activation of the canonical wnt signalling pathway. Specifically, loss of -catenin cytoplasmic expression correlated with increased - catenin nuclear expression and increased cyclin D1 expression. Similarly loss of - catenin cytoplasmic expression was associated with a poor outcome 179. Although intermediate components of wnt signalling interact through complex formation and phosphorylation making mRNA expression less relevant to their possible functional states in PC, wnt ligands, frizzled receptors and most importantly the products of - catenin mediated transcription are supportive of augmented wnt signalling in PC. 68
S100 Calcium Binding Proteins
Expression profile data demonstrated differential expression of multiple members of the S100 family of calcium binding proteins in PC compared to normal pancreas (Figure 3.4). There was a greater than five fold upregulation of S100A2, S100A4, S100A6, S100A10, S100A11 and S100P in PC compared to normal pancreas. Further interrogation of the data identified that S100A2 (Fold Change = 172), S100A6 (Fold Change = 42) and S100P (Fold Change = 17.2) were upregulated >10 times in PC compared to normal pancreas. S100 genes encode small (9-12 kD) calcium-binding proteins, with 30-50% homology within the group. Thirteen S100 genes map closely together to chromosome 1q21 (except S100P, which is located on 4p16). It is thought that S100 proteins serve as calcium sensor proteins that, upon activation, regulate the function and/or subcellular distribution of specific target proteins. Despite the overwhelming volume of data supporting the role of S100 calcium binding proteins in the pathogenesis of cancer presented in chapter 1 there have been few studies in the literature relating expression of S100 calcium binding protein to markers of clinical progression and outcome.
Chapter 4 validates the expression profile data presented in this chapter and further investigates the association of S100A2, S100A6 and S100P expression in PC and their association with clinicopathological markers of outcome. S100A2 FC=42
S100A1 S100-P FC = 172
S100A3
S100A4 FC = 6.7
S100A5
S100A6 FC = 17.2
S100A7
S100A8
S100A9
S100A10 FC = 5.9
S100A11 FC = 9.3
S100A12 FC=0.3
S100A13 FC=1.9
S100A14 FC = 2.6
S100B FC = 0.2
Figure 3.4: Alterations in expression in PC compared to normal pancreas for members of the S100 calcium binding protein family. Red denotes upregulation of ≥ 2, blue - downregulation ≤ 0.5, lime - fold change of 0.5 to 1.9 and clear representing no significant fold change. Values adjacent to each represented gene is the fold change of mRNA expression obtained from HG-U133 Affymetrix array data for PC relative to normal pancreas. 70
Retinoic Acid Signalling
Analyses of the transcript profile data identified aberrant expression of a significant number of components of RA signalling (Figure 3.5, Table 3.3). Retinoic Acid Receptor alpha (RAR- ) and Retinoic Acid Receptor gamma (RAR- ) were upregulated 2.9-fold and 2.2-fold respectively in PC compared to normal pancreas. Expression of a substantial number of known RA responsive genes was also altered in PC, consistent with dysregulated RA signalling activity, primarily demonstrating upregulation of genes downstream of RAR- . A substantial number of genes regulated by RA and known to be highly expressed in PC and PanIN from other studies, were also upregulated: S100 calcium binding protein P (S100P) 180 (152-fold), MUC4 mucin 114 (24.6-fold), matrix metalloproteinase 9 (MMP9) 116 (2.0-fold), Id-1 119 (2.3-fold), urokinase plasminogen activator receptor (uPAR) 181 (13.5-fold) and Heparin-binding EGF-like growth factor (HB-EGF) 182 (2.5-fold; Table 3.3).
Other genes, yet to be characterized in PC, but thought to be regulated by RA were also aberrantly expressed including a retinoic acid induced G-protein coupled receptor 3 (RAI3) (26.3–fold), retinoic acid receptor responders RARRES 1 (16.5–fold) and 3 (3.8–fold) (Table 3.3).
Retinoic acid induced gene-1 (RAIG1), also known as RAI3, encodes a 357-amino acid protein with a calculated molecular mass of 40,256 Da. It contains 7 predicted transmembrane domains, which is a signature motif of the G protein-coupled receptor superfamily, and a potential N-linked glycosylation site. The levels of RAIG1 mRNA in different cancer cells vary greatly with no correlation between the expression levels and the type of cancer. RAIG1 is expressed in several normal human tissues, with the highest expression levels found in fetal and adult lung. Retinoid Receptors hRADH 4.7 RALDH2 NC RAR α 2.9 RBP4 NC RAR β NC RA synthesis sRABP 2.6 RAR γ 2.2 Extracellular RA RXR α 0.4 carrier proteins RXR β NC CYP26 NC Cellular Retinoic Acid RXR γ NC Binding Proteins RA degradation
CRABP1 0.2 Cellular Retinoid Binding Proteins CRABP2 2.6 CRALBP NC
CRBP1 0.2 CEBPβ 2.1 CRBP2 NC CRBP3 NC CRBP4 NC
GPCR3 26 TNFAIP2 2.9 RARRES1 17 FOSL1 2.8 krox20 3.2 HOXB6 14 HOXB5 2.7 MUC 4 24 IFT5 4.2 HOXC6 2.6 MMP9 2.0 RARRES3 3.8 IL-8 12 Id-1 2.3 HOXD3 3.4 STRA6 1.9 HOXB1 0.4 HOXB2 6.7 S100 P 152 HOXC9 3.2 HOXD8 0.6 p21 NC HOXA5 3.0 RARG-1 0.6 ? uPAR 13 TGM2 2.9 EVX1 0.5 ZNF145/PLZF 0.3 HB-EGF 2.5 laminin-β1 2.2 RARRES2 0.1
Expression induced by RA and Genes containing a RARE and Genes modulated by RA and known to be upregulated in PC induceable by RA associated with development
Figure 3.5: A customized GenMAPP of RA signalling components: with statistically significant relative expression levels in PC compared to normal pancreas denoted beside the gene. Relative expression levels are represented as fold change with those with statistically significant upregulation marked red, those with significant downregulation marked blue and those marked in green represent fold changes of between 0.5 and 1.9, for: retinoic acid receptors; cellular retinoic acid binding proteins (CRABP); downstream targets of RA signalling previously described to be aberrantly expressed in PC and PanIN; downstream targets of RA signalling containing a retinoic response element (RARE) expression in PC not previously investigated; downstream targets modulated by RA associated with embryonal development. Abbreviations not mentioned in text: RBP4, retinoid binding protein 4; CRALBP, cellular retinaldehyde binding protein; RALDH, retinaldehyde dehydrogenase. 72
Table 3.3: Components of RA signalling with differential expression between PC and normal pancreas on Affymetrix U133 microarrays.
Unigene Fold p Probeset Cluster Gene Name change value 204351_at Hs.2962 S100 Calcium Binding Protein P 152 0.001 203108_at Hs.194691 GPCR (retinoic acid induced 3) 26.3 0.007 217109_at Hs.198267 MUC4 24.6 0.001 206392_s_at Hs.82547 retinoic acid receptor responder 1 (RARRES1) 16.5 0.004 205366_s_at Hs.98428 HOXB6 14.4 0.009 211924_s_at Hs.179657 urokinase plasminogen activator receptor (uPAR) 13.5 0.002 202859_x_at Hs.624 Interleukin 8 (IL8) 12.5 0.001 205453_at Hs.2733 HOXB2 6.7 0.001 219799_s_at Hs.179608 retinol dehydrogenase homolog (hRADH) 4.7 0.016 203596_s_at Hs.27610 retinoic acid- and interferon-inducible protein (IFT5) 4.2 0.010 204070_at Hs.17466 retinoic acid receptor responder 3 (RARRES3) 3.8 0.005 228601_at Hs.93574 HOXD3 3.4 0.024 205249_at Hs.1359 Krox20 (EGR2) 3.2 0.008 231936_at Hs.40408 HOXC9 3.2 0.009 213844_at Hs.37034 HOXA5 3.0 0.035 203749_s_at Hs.250505 retinoic acid receptor, alpha (RAR- ) 2.9 0.007 201042_at Hs.512708 TGM2 (transglutaminase 2) 2.9 0.001 202510_s_at Hs.101382 TNFAIP2 (tumor necrosis factor, alpha-induced protein 2) 2.9 0.004 204420_at Hs.283565 FOSL1 (FOS-like antigen-1) 2.8 0.005 205601_s_at Hs.22554 HOXB5 2.7 0.003 206858_s_at Hs.820 HOXC6 2.6 0.016 202575_at Hs183650 cellular retinoic acid-binding protein 2 (CRABP2) 2.6 0.018 38037_at Hs.799 heparin-binding EGF-like growth factor (HB-EGF) 2.5 0.009 2214782_at novel gene similar to retinaldehyde-binding protein (sRABP) 2.4 0.036 208937_s_at Hs.75424 Id-1 (Inhibitor of DNA binding 1) 2.3 0.039 201505_at Hs.82124 laminin 1 2.2 0.026 204118_s_at Hs.1497 retinoic acid receptor, gamma (RAR- ) 2.2 0.049 212501_at Hs.99029 CEBP (CCAAT enhancer binding protein ) 2.1 0.005 203936_s_at Hs.151738 MMP9 (matrix metalloproteinase 9) 2.0 0.008 221701_s_at Hs.24553 STRA6 1.9 0.030 202449_s_at Hs.20084 retinoid X receptor, alpha (RXR- ) 0.4 0.005 231906_at Hs.301963 HOXD8 0.6 0.010 202882_x_at Hs.106346 retinoic acid repressible protein (RARG-1) 0.6 0.045 207914_x_at Hs.336963 even-skipped homeo box 1 (EVX1) 0.5 0.016 208224_at Hs.99992 HOXB1 0.4 0.010 205883_at Hs.37096 zinc finger protein 145 (ZNF145, PLZF) 0.3 0.002 203423_at Hs.101850 cellular retinoic acid-binding protein 1 (CRABP1) 0.2 0.001 209496_at Hs.37682 retinoic acid receptor responder 2 (RARRES2) 0.1 0.001 73
Although the mechanism by which retinoic acid signalling imparts its effects on cellular function is not clear, the transcript profile data presented here suggest that it may be important in PC.
Data presented in Chapter 5 validates aberrant expression of RA signalling components in PC, in addition, the expression of RAI3 is investigated using in-situ hybridisation (ISH).
The reactivation of developmental pathways, specifically those that determine exocrine cell lineage, have been implicated in the early development of PC and pathways that determine duct cell versus acinar cell differentiation including RA and Notch signalling 44,183 may be important in the development of PC. Retinoic acid receptors are also implicated in exocrine pancreatic development 184 by regulating exocrine lineage selection favouring ductal rather than acinar differentiation 110. Studies of hindbrain development have provided the greatest insights into the mechanism of RA signalling. RA dependent lineage restriction in rhombomeres 3 and 5 is marked by krox20 and HOXB2 expression. RA, through an as yet unknown mechanism that may involve CEBP 185 results in increased krox20 expression, which in turn increases HOXB2 expression by directly binding promoter elements of HOXB2 186. krox20 also suppresses HOXB1 expression. The expression profile data presented in this thesis is consistent with this mechanism of HOXB2 regulation (Figure 3.5). In addition, the promyelocytic leukaemia (PML) fusion protein PLZF-RARA also regulates HOXB2 expression by directly binding the promoter region of HOXB2, and is thought to be important in PML development and resistance to RA therapy 187.
Chapter 6 presents data validating aberrant HOXB2 expression in PC and PanIN, examines relationships between HOXB2 expression and clinopathological markers of outcome in PC, and begins to assess to role of HOXB2 in PC cell lines. 74
Discussion
A caveat in the interpretation of data from expression profiling experiments is that no conclusions can be drawn with regard to the cellular events that are orchestrated through protein interactions. However, clues as to the potential relevance of molecules to cancer and pathways involved in carcinogenesis may be identified through global analysis of mRNA expression. The aberrant expression of several components of upregulation of receptors, their ligands, and the downstream responders of the retinoic acid signalling suggests that there is alteration in its functional state in PC compared to normal pancreas and this warrants further investigation.
Specimens that were transcript profiled in this study were heterogeneous. Normal pancreas contains exocrine, endocrine, and stromal elements as well as ductal epithelium. Similarly PC is composed of epithelium, stroma and possibly some contaminating exocrine pancreatic tissue as well as a variable infiltrate of immune response cells. Thus validation of these findings to determine if they are localised to the epithelial or other components of PC is important in future validation studies. IHC for both sFRP4 and -catenin showed that they were localised to cancer epithelium, and that both were differentially expressed between PC and normal pancreatic ducts.
Conclusion
The data presented in this chapter suggests that components of TGF- signalling, wnt signalling, S100 calcium binding protein family and Retinoic Acid signalling, are aberrantly expressed in PC compared to normal pancreas and may play a role in PC. The S100 calcium binding protein family and RA signalling pathway were selected for further investigation in this thesis as several components of the S100 calcium binding protein family are also regulated by retinoic acid. Studies are presented in the following chapters, which begin to examine and define the potential importance of S100 proteins and RA signalling in PC. 75
Chapter 4
S100 CALCIUM BINDING PROTEINS and OUTCOME in PANCREATIC CANCER 76
Introduction
S100 proteins are a family of low-molecular weight, calcium-binding proteins that initiate cellular processes including cell division, motility, secretion, protein synthesis, and membrane permeability, mediating the second messenger role of calcium. This protein family is characterized by the EF-hand structural motifs 188.
The association of S100 proteins with cancer development originated from the association that an evolutionary conserved gene cluster of S100 genes was found on human chromosome 1q21 where several gene rearrangements during tumor development have been observed 96. S100 protein status was first used to identify metastatic deposits of malignant melanoma 97, aberrant S100 protein expression has been identified in many tumors, including those of glial, ovarian, breast, stomach and pancreatic origin. In addition to a purely intracellular role S100 calcium binding proteins, are excreted from the cell, making them strong potential targets for novel serum based diagnostic strategies.
Overexpression of several members of the S100 family of calcium binding proteins including S100A4 155,189,190, S100A6 180,191 , S100A10 192, and S100P 159,161,191-193 have been identified in global gene expression studies, confirmatory RNA and/or protein analyses. A recent study incorporating a comparison between PC and chronic pancreatitis into its gene profiling strategy identified genes differentially expressed by neoplastic epithelium, validating increased expression of S100 calcium binding protein family members in neoplastic epithelium 191.
Transcript profiling experiments described in Chapter 3 identified differential expression of multiple members of the S100 family of calcium binding proteins, in PC compared to normal pancreas (Figure 4.1). Fold Change PC vs Normal Pancreas 10 xrse 0tmsmr nP oprdt omlpancreas. normal to and compared S100A2 PC S100P. in pancreas. more normal times to 10 compared > PC expressed in S100P times >5 up-regulated : S100A10 pancreas protein and normal binding S100A4 to calcium compared S100 PC in of members expression family differential of representation Graphical 4.1: Figure 150 200 -10 10 20 30 40 50 0 10Epeso nPnrai acrV omlPancreas Normal Vs Cancer Pancreatic in Expression S100 oe 10,S0A,S0A,S100A11, S100A6, S100A2, S100P, Note: 78
A greater than five fold upregulation of S100A2, S100A4, S100A6, S100A10, S100A11 and S100P was observed in PC compared to normal pancreas. Further interrogation of the data identified that S100A2 (Fold Change = 172), S100A6 (Fold Change = 42) and S100P (Fold Change = 17.2) were upregulated >10 times in PC compared to normal pancreas. Other members of the S100 calcium binding protein family did not display significant aberrant expression in PC compared to normal pancreas.
There is mounting evidence implicating the role for S100 protein expression in carcinogenesis, although a precise mechanism is yet to be defined. S100A2 protein has been identified as a potential tumor suppressor in tumor-derived mammary epithelial cells 194. S100A2 expression modulates p53 activity in breast and head and neck cancers 195, increased S100A2 expression is associated with the development of a metastatic phenotype in oesophageal cancer 196 and is an independent predictor of outcome in a histological subset of lung cancer 197. S100A2 expression is also upregulated in papillary and follicular carcinomas of the thyroid 198, and gastric cancer 199.
The overexpression of S100A4 also (also known as mts1 pEL-98, 18A2, p9Ka, CAPL, calvasculin, Fspl) has been associated with poor prognosis and increased metastatic potential in breast cancer 200,201 . In addition to these data an increase in S100A4 expression has been correlated with a worse prognosis in colorectal, gallbladder, esophageal, gastric, hepatocellular and bladder cancers Using immunohistochemistry Rosty et al 189 demonstrated protein overexpression of S100A4 in 93% of invasive PC and 17 % of high grade PanIN lesions. Furthermore significant differential expression of S100A4 was demonstrated using RT-PCR between short-term ductal pancreatic cultures and established pancreatic cancer cell lines 155. S100A4 together with 6 other genes that were overexpressed in the neoplastic cells of PC and not expressed in the normal pancreatic ductal epithelium was found to be hypomethylated in a significant proportion of PC cell lines and primary pancreatic carcinomas. Suggesting that gene hypomethylation of S100A4 may be a frequent epigenetic event and is associated with its overexpression in PC. 79
S100A6, also known as calcyclin, was first observed to be up-regulated in proliferating fibroblasts stimulated by growth factor 202. It is implicated in the secretion of insulin from pancreatic endocrine cells 203 and together with S100P, is involved in a novel ubiquitination pathway regulating the degredation of -Catenin 204. Aberrant expression has been observed in colorectal cancer, hepatocellular carcinoma and malignant melanoma 205-207. Transient expression of NF-kappaB (p65) increases the expression of S100A6 mRNA in a human hepatoblastoma cell line by increasing S100A6 promoter activity 208. Loss of S100A6 protein expression is common in prostate cancer development and may occur at an early stage 209, this loss of expression is due to promoter hyper-methylation 210.
S100P, a 95-amino-acid member of the S100 family, was first purified from placenta 211. S100P is overexpressed in many diseases including the inflammatory bowel diseases 212 and some human cancers. S100P expression is associated with resistance to chemotherapy in colon cancer cell lines 213, cellular immortalisation in breast cancer cell lines 214 and increased androgen sensitivity in prostate cancer 215. S100P expression is also observed in IPMT a putative precursor lesion of PC suggesting a role in the development of PC 193.
Given the compelling evidence from cDNA microarray, SAGE library scanning and tissue oligonucleotide microarray data demonstrating differential expression of S100 calcium binding proteins in PC compared to normal pancreas and their potential importance in the neoplastic process of PC, the protein expression of S100A2, S100A6 and S100P was examined in a cohort of patients with PC. 80
S100A2, S100A6 and S100P Expression and Outcome in PC.
The PC cohort
A subset of the PC cohort at the Garvan Institute of Medical Research was used for this study (Table 4.1). The available tissue consisted of 124 patients with the diagnosis of PC (71 males and 53 females). The average age at diagnosis was 64.1 years (median 66.5, range 34 - 84). Of 124 patients for whom tissue was available 75 were from pancreatic resections, 43 intraoperative incision biopsies and 6 post mortem specimens, which were included in analysis as it is part of the natural history of the disease. Median follow-up for the cohort was 7.9 months (range 0 to 117 months). Eight patients were alive at the census date (September 21, 2002). Of the 124 patients, 110 (89%) died of PC and 2 (1.5%) of other causes (peri-operative deaths), 4 were lost to follow-up. The overall median survival and disease specific survival was 7.6 months. Overall disease specific 1-year survival was 26% with a 5-year survival of 6%. Of 124 tumours with available tissue, the majority of tumours were either moderately differentiated (65) or poorly differentiated (47), with only 11 well differentiated cancers. Twenty seven samples were UICC 216 Stage 1, 12 Stage 2, 69 Stage 3 and 13 Stage 4, 3 samples had insufficient pathological and clinical information to make an accurate assessment of stage.
Data were analysed in two groups:
1) As a whole cohort; which included all patients with the diagnosis of PC, these included those that underwent pancreatic resection, those that underwent operative or diagnostic biopsy and those who had PC diagnosed at post mortem. 2) As a sub-group of those patients who had pancreatic resections. 81
Table 4.1 Clinicopathological, S100A2, S100A6 and S100P molecular, and outcome data for All Patients Within the Cohort.
Parameter Whole n= 124 Resected n=75 Cohort Cohort Median pvalue Median Survival pvalue No. (%) Survival No. (%) (months) (logrank) (logrank) (month s )
Sex Female 53 (43) 31 (41) Male 71 (57) 44 (59)
Age at diagnosis (years) Mean 64.1 62.3 Median 66.5 65.0 Range 34.4 – 83.8 34.4 – 82.6
Treatment Resection 75 (60) 11.2 Operative Biopsy 49 (40) 4.1 < 0.0001
Outcome Follow-up (months) 0 - 117 0.2 - 117 Median 7.9 11.0 30 day mortality 2 (2) 2 (3) Death from PC 110 (89) 62 (83) Death from other cause 2 (2) 2 (3) Alive 8 (6) 8 (11) Lost to follow-up 4 (3) 3 (4)
Stage 121 I 27 (22) II 12 (10) 13.6* III 69 (56) IV 13 (10) 6.9 0.0001
Differentiation 123 Well 11 (9) 7 (9) Moderate 65(52) 9.3** 44 (58) 12.2 Poor 47 (38) 5.6 0.009 24 (33) 9.1 0.086
Tumor size 20mm 15 (20) 17.1 >20mm 60 (80) 9.9 0.04
Margins Clear 40 (53) 14.5 Involved 35 (47) 8.6 0.002
Lymph node status§ Negative 34 (46) 13.7 Positive 39 (51) 9.7 0.04
S100 A2 cytoplasmic expression Low (10%) 85 (57) 9.1 53 (72) 12.2 High 30 (43) 4.8 0.009 21 (28) 7.5 0.003
S100 A6 cytoplasmic intensity¶ Low 56 (40) 4.7 17 (23) 11.2 High 67 (60) 9.7 0.0008 57 (77) 10.8 0.61
S100 P expression Low 65 (52) 5.0 20 (27) 9.9 High 59 (48) 10.5 0.0005 55 (73) 12.2 0.46
* Stage I and II tumours versus stage III and IV ** Well and Moderately differentiated tumours grouped together for analysis § Lymph node status was only available in 73 patients in the resected cohort S100 A2 expression was only available in 115 patients in the whole cohort and 74 patients in the resected cohort ¶ S100 A6 expression was only available in 123 patients for the whole cohort and 74 patients in the resected cohort 82
The Whole Cohort
S100A2 expression (Figure 4.2) was identified in 30 of 124 (43%) tumours in the whole cohort and S100A6 and S100P expression was identified in 67 (60%) and 59 (47.5%) of 113 and 123 tumours respectively. Kaplan-Meier analysis identified that operative resection of the tumour was associated with longer survival than operative biopsy (median survival 11.2, and 4.1 months respectively: logrank p<0.0001, Figure 4.3A). Patients with Stage 1 and 2 tumours (lymph node negative) survived significantly longer (median survival 13.6 months) than those with Stage 3 and 4 tumours who had a median survival of 6.9 months (logrank p = 0.0001, Figure 4.3B). Those with poorly differentiated tumours had a shorter survival (median 5.6 months) than those with well or moderately differentiated tumours (9.3 months) (logrank p=0.009, Figure 4.3C). S100A2 overexpression in the whole cohort was associated with a poor outcome on univariate analysis (median survival 4.8 months and 9.1 months logrank, p = 0.009: Figure 4.3D). Increased S100A6 and S100P cytoplasmic expression (Figure 4.2) was associated with an improved outcome in the whole cohort. (median survival benefit of 5.0 months, logrank p = 0.0008 and 5.5 months, logrank p = 0.0005 respectively: Figures 4.3E and 4.3F)
Univariate analysis identified known clinicopathological factors in PC demonstrating that pancreatic resection, low U.I.C.C. stage and well/moderately differentiated tumours were associated with an improved outcome. In addition, high S100A2 and low S100A6/S100P expression was associated with a poor outcome respectively (Table 4.1).
Multivariate analysis using the Cox proportional hazards model for those factors that demonstrated a significant effect on survival on univariate analysis, identified S100A2 expression and resection as the only independent prognostic factors when modelled together with S100A6 status, S100P status, UICC stage and degree of differentiation (Table 4.2A-E). A B
C D
E F
Figure 4.2: S100 Protein expression in Pancreatic Ductal Adenocarcinoma: A. low S100A2 expression, B. high S100A2 expression in pancreatic cancer, C. low S100A6 expression, D. high S100A6 expression in PC, H. low S100P expression and I. high S100P expression and in PC. A. Operative Treatment B. Stage 1 p < 0.0001 1 p=0.0001 Median Survival: 11 . 2 Vs 4 . 1 months .8 .8 Median Survival: 13 . 6 Vs 6 . 9 months n=124 n = 121
.6 .6
.4 .4 CUMULATIVE SURVIVAL CUMULATIVE SURVIVAL .2 .2 STAGEI/II RESECTED
0 BIOPSY 0 STAGE III / IV
0 20 40 60 80 100 120 0 20 40 60 80 100 120 MONTHS MONTHS C. Degree of Differentiation D. S100A2 Expression
1 p = 0.009 1 p = 0.009 Median Survival: 9 . 1 Vs 4 . 8 months .8 Median Survival: 9 . 3 Vs 5 . 6 months n= 123 .8 n = 115
.6 .6
.4 .4 CUMULATIVE SURVIVAL CUMULATIVE SURVIVAL .2 .2 WELL / MODERATE LOW S100A2
0 POOR 0 HIGH S100A2
0 20 40 60 80 100 120 0 20 40 60 80 100 120 MONTHS MONTHS E. S100A6 Expression F. S100P Expression 1 p = 0.0008 1 p = 0.0005 .8 MedianSurvival:9.7 Vs4 .7 months .8 Median Survival: 10 . 5 Vs 5 . 0 months n=123 n=124 .6 .6
HIGH S100A6 .4 .4 LOW S100A6 CUMULATIVE SURVIVAL CUMULATIVE SURVIVAL .2 .2 HIGH S100A2
0 0 LOW S100A2
0 20 40 60 80 100 120 0 20 40 60 80 100 120
MONTHS MONTHS
Figure 4.3: Kaplan-Meier survival curves for the whole cohort of 124 patients : (A) Operative Treatment; (B) Stage; (C) Degree of Differentiation; (D) S100A2 cytoplasmic expression; (E) S100A6 cytoplasmic expression and (F) S100P expression. 85
Table 4.2: Multivariate analysis for clinicopathological parameters and S100A2, S100A6 and S100P expression in the whole and resected cohorts of PC.
Variable Hazard Ratio p-value (95% confidence interval)
A. Whole cohort (n =112) S100A2 expression 2.44 (1.49 – 4.00) 0.0004 S100A6 expression 0.67 (0.37 – 1.20) 0.1795 S100P expression 0.77 (0.47 – 1.27) 0.3065 Operative resection 0.43 (0.24 – 0.78) 0.0051 Stage III/IV vs I/II 1.61 (0.94 – 2.75) 0.0805 Differentiation 0.70 (0.45 – 1.10) 0.1257
B. Whole Cohort (n=113) S100A2 expression 2.43 (1.48 – 3.98) 0.0004 S100A6 expression 0.59 (0.35 – 0.99) 0.0477 Operative resection 0.44 (0.24 – 0.79) 0.0055 Stage III/IV vs I/II 1.69 (1.00 – 2.86) 0.0497 Differentiation 0.68 (0.44 – 1.06) 0.6790
C. Whole Cohort (n=113) S100A2 expression 2.46 (1.51 – 4.02) 0.0003 S100A6 expression 0.64 (0.38 – 1.06) 0.0841 Operative resection 0.43 (0.24 – 0.76) 0.0040 Stage III/IV vs I/II 1.81 (1.08 – 3.04) 0.0240
D. Whole Cohort (n=114) S100A2 expression 2.30 (1.43 – 3.72) 0.0006 Operative resection 0.31 (0.19 – 0.51) <0.0001 Stage III/IV vs I/II 1.67 (0.99 – 2.80) 0.0511
E. Whole Cohort (n=115) S100A2 expression 2.23 (1.38 – 3.59) 0.0011 Operative resection 0.25 (0.16 – 0.40) <0.0001
F. Resected cohort (n = 73) S100A2 expression 2.41 (1.30 – 4.37) 0.0046 S100A6 expression 0.68 (0.33 – 1.41) 0.3012 S100P expression 0.77 (0.37 – 1.60) 0.4777 Tumour size > 20 mm 1.91 (0.95 – 3.86) 0.0696 Margin involvement 1.84 (0.97 – 3.50) 0.0622 Lymph node involvement 1.20 (0.62 – 2.31) 0.5861
G. Resected cohort (n = 74) S100A2 expression 2.29 (1.29 – 4.06) 0.0049 Tumour size > 20mm 1.69 (0.86 – 3.32) 0.1257 Margin involvement 1.93 (1.11 – 3.35) 0.0207
A: Multivariate analysis involving all significant factors associated with outcome identified using univariate analysis in the whole cohort. Note S100A2 and operative resection remain independent prognostic markers. B-D: Stepwise removal of redundant variables, B: S100P removed from model; C: Differentiation removed from model; D: S100A6 removed from model; E: the resolved model of A-D. with U.I.C.C. stage removed, note S100A2 and operative resection remain the only independent markers of outcome. F; Multivariate analysis involving all significant factors associated with outcome identified using univariate analysis in the resected cohort. Note S100A2 is the only independent prognostic marker; G: is the resolved model of F: eliminating redundant variables including S100A6, S100P, and lymph node involvement. Note S100A2 and Margin involvement are independent prognostic factors. 86
The Resected Cohort
Of the 118 patients that underwent laparotomy (Table 4.1), 75 proceeded to pancreatic resection (44 males, median age 65, range 34 - 83). Median follow-up was 11.0 months (0.2 - 117 months) with a median disease specific survival of 10.1 months, 1-year survival of 48.6% and 5-year survival of 11%. The only patients still living in the cohort (n = 8) underwent resection. The majority (65 patients) died of PC, 2 died of other non-PC causes and 3 were lost to follow-up. The 30-day mortality for resection was 2 (3%). Thirty-nine (51%) of resected pancreata had lymph nodes free of tumour. The mean tumour size was 31 mm (range 8 to 80 mm). Resection margins were microscopically free of tumour in 40 cases (53%). Poorly differentiated tumours occurred in 24 patients (33%), with 7 well-differentiated (9%) and 44 moderately differentiated (58%) tumours. Patients whose resected pancreata showed margins clear of microscopic disease had a survival advantage (median survival 14.5 months), compared to those with tumour at the resection margin (median survival 8.6 months) (p = 0.002) as did those who had tumours 20 mm (median survival 17.1 vs 9.9 months, p = 0.04). Patients whose resected pancreata showed no tumour involvement of lymph nodes survived longer (median survival 13.7 vs 9.7 months, p = 0.04). S100A2 expression was present in 21 of 74 resected pancreata (28%); this was associated with a poor outcome (median survival 7.5 vs 12.2 months, p = 0.003). Expression of S100A6 and S100P was observed in 77% and 73 % of resected pancreata respectively and was not associated with outcome as did, degree of differentiation of the tumour (Fig 4.4, A-F).
Multivariate analyses identified S100A2 expression as the only independent prognostic factor when modelled against S100A6 status, S100P expression, lymph node involvement, tumour size >20 mm and positive surgical margins. Table 4.3F shows the multivariate model for resected tumours prior to resolution of the final model shown in Table 4.3G, where S100A2 expression and involved surgical margins are maintained as independent prognostic factors. Hence tumour size > 20mm, lymph node involvement, involved surgical margins and high S100A2 expression all conferred a poor outcome on univariate analysis with S100A2 expression and involved surgical margins identified as independent prognostic factors on multivariate analysis. Resected Cohort
A. Resection Margin B. Tumor Size 1 p=0.002 1 p=0.04 Median Survival: 14 . 5 Vs 8 . 6 months Median Survival: 17 . 1 Vs 9 . 9 months .8 n=75 .8 n=75
.6 .6
.4 .4 ≤ 20 mm CUMULATIVE SURVIVAL CUMULATIVE SURVIVAL .2 .2 CLEAR
0 INVOLVED 0 >20mm
0 20 40 60 80 100 120 0 20 40 60 80 100 120 MONTHS MONTHS C. Lymph Node Invasion D. S100A2 Expression
1 1 p = 0.04 p = 0.003 Median Survival: 13 . 7 Vs 9 . 7 months Median Survival: 12. 2 Vs7.5 months .8 n=73 .8 n=74
.6 .6
.4 .4 CUMULATIVE SURVIVAL CUMULATIVE SURVIVAL
.2 .2 NODE NEGATIVE LOW S100A2
0 NODE POSITIVE 0 HIGH S100A2
0 20 40 60 80 100 120 0 20 40 60 80 100 120
MONTHS MONTHS E. S100A6 Expression F. S100P Expression 1 p = 0.61 1 p=0.46 Median Survival: 11. 2 Vs 10 . 8 months .8 .8 Median Survival: 12. 2 Vs 9 . 9 months n=74 n=75
.6 .6
HIGH S100A6 HIGH S100P .4 .4 LOW S100A6 LOW S100P CUMULATIVE SURVIVAL CUMULATIVE SURVIVAL .2 .2
0 0
0 20 40 60 80 100 120 0 20 40 60 80 100 120 MONTHS MONTHS
Figure 4.4: Kaplan-Meier survival curves for 75 patients that underwent pancreatic resection: (A) Resection status; (B) Tumor size; (C) Lymph node involvement; (D) S100A2 cytoplasmic expression; (E) S100A6 cytoplasmic expression and (F) S100P expression. 88
Discussion
Aberrant expression of S100 proteins S100A2, S100A6 and S100P occurs in a significant proportion of PC. This chapter identifies for the first time an association between high S100A2 expression in PC and poor outcome. In addition, S100A2 expression was the most influential prognostic factor on multivariate analysis compared to known and established clinicopathologic parameters of outcome in all patients including a sub-group of patients who underwent pancreatic resection. Increased expression of S100A6 and S100P was associated with an improved outcome in these patients. However, differential S100A6 and S100P expression did not co-segregate with outcome in a cohort of patients who underwent pancreatic resection.
S100A2
S100A2 protein has been identified as a potential tumor suppressor in tumor-derived mammary epithelial cells. In addition, loss of S100A2 mRNA was progressively decreased in the transition from normal breast epithelium to cancer in clinical specimens 194. Downregulation of S100A2 expression has also been described in breast, prostate and lung carcinomas 217-219. S100A2 expression is decreased in high grade tumors of the larynx compared to lower grade tumors with expression associated with improved relapse free and overall survival 220, leading to the hypothesis that S100A2 is a tumor suppressor and downregulation of the gene, which occurs via promoter hypermethylation, plays a significant role in carcinogenesis 221.
Recent evidence suggests that S100A2 may also act as an oncogene. Increased expression has been identified in Non Small Cell Lung Carcinoma (NSCLC) and in preinvasive bronchial lesions, with no identifiable somatic mutations 222, contrary to the belief that S100A2 is a tumor suppressor. Positive S100A2 expression is associated with a significantly lower overall survival and disease-specific survival rate at 5 years after surgery in patients with NSCLC, in addition, S100A2 expression was identified as a better predictor for disease-specific survival than other clinical and histological variables on multivariate analysis in these patients 197 . 89
In oesophageal cancer S100A2 overexpression is associated with a trend toward preferentially developing lymph node and distant metastases 196.
Insights into the proliferative potential of S100A2 may be gained by examining its association with squamous cell proliferation in the dermis. S100A2 is located within the epidermal differentiation complex and has been associated with squamous cell differentiation 223 and in regenerative hyperplasia of keratinocytes in active psoriasis 224. These proliferative effects are associated with the activation of EGF receptor tyrosine kinases, in particular cErbB2 which selectively stimulates transcription of S100A2 225. S100A2 also interacts with the tumour suppressor, p53. S100A2 translocation to the nucleus inactivates the C terminus of p53 in a calcium-dependent manner, resulting in the inability of p53 to bind to the promoter regions of its target genes, thus repressing the transcriptional activity of p53 195.
Although the biological function of S100A2 in carcinogenesis has not been elucidated, data from this cohort suggest a potential role in PC. The relationship between S100A2 and resection status, tumour size and lymph node involvement on multivariate analysis suggests that it may be a surrogate marker of advanced disease and hence confers a poor prognosis for those patients undergoing resection. This creates scope for the use of S100A2 expression as a potential marker of aggressive disease.
S100A6
S100A6, also known as calcyclin, has been extensively studied in systems other than cancer. The majority of work has centered on the function of S100A6 in neural cells. Tonini et al. describes normal expression of S100A6 in the G1 phase of the cell cycle. In addition S100A6 mRNA is increased after retinoic acid induced neuroblastoma cell differentiation, acting as a regulator of cell-cycle progression in these cells 226. Aberrant retinoic acid signalling may play a significant role in pancreatic carcinogenesis 227,and aberrant S100A6 expression may be a consequence of aberrant RA signalling in PC. 90
Increased nuclear expression of S100A6 was recently reported to be a significant independent predictor of poor outcome in a cohort of 60 resected pancreata 73.The study presented in this chapter, contains a larger number of samples and did not identify S100A6 as an independent predictor of survival. In addition, the supposition that upregulation of S100A6 is associated with a biologically aggressive cancer is not absolute. Recent evidence in osteosarcoma identifies an association between increased expression of S100A6 and a trend toward decreased metastasis, with overexpression of S100A6 associated with decreased motility and anchorage independent growth, suggesting that loss of expression correlates with a metastatic phenotype 228. The data presented above identifies a potential role of S100A6 in PC; however, further validation using independent cohorts is necessary before significant conclusions can be drawn concerning its role in PC.
S100P
S100P expression correlates with resistance to chemotherapy in colon cancer cell lines 213, cellular immortalization in breast cancer cell lines 214 and increased androgen sensitivity in prostate cancer 215. cDNA microarray, SAGE library scanning, and tissue microarray immunohistochemistry data suggesting that differential expression of S100P is exclusive to neoplastic pancreatic tissue, increases its specificity as a potential diagnostic marker 180. Further evidence supporting a key role of S100P in PC is the observation of increased S100P expression in PanIN 229 and IPMT 193, both precursor lesions of PC, suggesting that S100P is upregulated at an early stage of the disease. Recent functional studies have identified S100P expression and secretion in PC cell lines. Gene transfer or extracellular addition of S100P increased cancer cell growth, motility, invasion, and survival in vitro and tumor growth and metastasis in vivo 230 . Although not identified as an independent predictor of survival in this study, further elucidation of the biological function of S100P and correlation of these findings in an independent cohort is encouraged as S100P possesses qualities making it a potential diagnostic marker (high expression in PC, tissue specificity, association with precursor lesions and secretion in bile and pancreatic juice). 91
Conclusion
In summary, a significant proportion of PC have aberrant expression of S100 proteins S100A2, S100A6 and S100P. Increased expression of S100A2 was associated with a poor outcome, with high S100A2 expression identified as an independent prognostic factor in PC. Increased S100A6 and S100P expression was associated with an improved outcome, however, these were associated with resectability and hence not identified as independent prognostic factors. In patients who underwent pancreatectomy, high S100A2 expression was a superior predictor of survival than clinicopathological parameters including tumour size, margin involvement and lymph node status. Hence S100A2 possesses potential clinical utility as a marker of prognosis in a disease where there are few useful markers of prognosis. 92
Chapter 5
RETINOIC ACID SIGNALLING in PANCREATIC CANCER. 93
Introduction
The genome wide screen using Affymetrix Genechip® Microarrays described in Chapter 3 identified aberrant expression in a large number of Retinoic Acid (RA) signalling components in PC compared to normal pancreas (Figure 5.1). Differential expression was identified in several key retinoid receptors, cellular retinoid and retinoic acid binding proteins, as well as molecules involved in the transduction of the RA signal. Expression of downstream molecules regulated by RA signalling, known to be aberrantly expressed in PC were also identified. These include MUC4 mucin 231 114, MMP9 116 232,uPAR,HB-EGF,p21WAF1/CIP1 233 and Id-1 119 120, all of which are known to be aberrantly expressed in PC and identified using techniques such as IHC, ELISA and Western Blotting in several studies.
Retinol is not synthesised endogenously by humans and is ingested as pre-formed retinol and related esters from animal tissue or as pro-vitamin A carotenoids. Retinyl esters are hydrolysed to retinol before absorption by enterocytes in the small intestine whilst carotenoids are directly absorbed by the enterocyte and partially converted to retinol. Enterocyte metabolism converts cellular retinol to retinoid esters, which are then excreted into the circulation bound to chylomicrons. These are primarily absorbed by the liver where the retinyl esters are hydrolysed to retinol and bound to retinol binding protein (RBP) in the hepatocyte, this complex is then transferred and stored in hepatic stellate cells 234. Retinol-RBP complexes are secreted into the plasma where the complexes are taken up by receptors in the target cell. After cellular uptake retinol can be oxidised to retinal and then to retinoic acid (the major active metabolite) via retinaldehyde dehydrogenase and alcohol dehydrogenase. In addition, retinol and the oxidised products of retinol are bound to tissue specific intracellular binding proteins, Cellular Retinol Binding Protein (CRBP) and Cellular Retinoic Acid Binding Proteins (CRABP), which are involved in cellular transport of retinol and RA 234. RBP4 NC A B sRABP 2.6 CRBP1 0.2 RAR α 2.9 CRBP2 NC Extracellular RA RAR β NC CRBP3 NC carrier proteins Retinoid RAR γ 2.2 CRBP4 NC Receptors RXR α 0.4 CRABP1 0.2 RXR β NC Cellular Retinoid C CRABP2 2.6 RXR γ NC Binding Proteins CRALBP NC
Cellular Retinoic Acid TNFAIP2 2.9 Binding Proteins E RAI3 26 D RARRES1 17 FOSL1 2.8 HOXB6 14 HOXB5 2.7 hRADH 4.7 MUC 4 24 IFT5 4.2 HOXC6 2.6 RALDH2 NC MMP9 2.0 RARRES3 3.8 IL-8 12 RA synthesis Id-1 2.3 HOXD3 3.4 STRA6 1.9 S100 P 152 HOXC9 3.2 HOXD8 0.6 HOXA5 3.0 RARG-1 CYP26 NC p21 NC 0.6 uPAR 13 TGM2 2.9 EVX1 0.5 RA degradation HB-EGF 2.5 laminin-β1 2.2 RARRES2 0.1
Expression induced by RA and Expression induced by RA and known to be upregulated in PC upregulated in current study
Figure 5.1: GenMAPP of RA signalling components with statistically significant relative expression levels in PC compared to normal pancreas. Relative expression levels are represented as average fold change with those with statistically significant upregulation marked red, those with statistically significant downregulation marked blue (± < 0.05) and no change (NC) marked green for: A. retinoid receptors; B. cellular retinoid binding proteins (CRBP), C. cellular retinoic acid binding proteins (CRABP); D. downstream targets of RA signaling previously described to be aberrantly expressed in PC and PanIN; E. Novel downstream targets aberrantly expressed in PC. 95
The retinoid signal is then transduced by two families of nuclear transcription factors: RARs and RXRs, each with 3 subtypes: , and . These receptors are members of the nuclear receptor superfamily, which in the presence of ligand heterodimerise to activate the transcription of target genes through RA response elements (Figure 5.2).
Compounds that mimic the biological function of retinoic acid are referred to as retinoids, which are natural and synthetic derivatives of Vitamin A. Retinoids, have diverse biological functions in normal physiology. Complete lack of retinoids is seen in Vitamin A Deprivation syndrome (VAD) and RAR double mutant mice 235. The VAD phenotype includes opthalmic abnormalities, congenital defects of the cardiovascular, respiratory, digestive and genitourinary systems. Retinoids are the precursors to 11-cis retinal, a visual pigment, which is normally taken up by photoreceptor cells where it plays a critical role in light and dark vision as the visual pigment chromophore. Deficiency in retinoids leads to degeneration of these photoreceptor cells. Retinoids regulate the immune system by controlling the proliferation and function of regulatory and structural cells and modulating the expression of immunoregulators. RA induces Major Histocompatibility Complexes (MHC) surface molecules in human embryonal carcinoma cells 236 , promotes the proliferation of murine T lymphocytes 237 and inhibits activation induced T cell apoptosis 238. During vertebrate development RA signalling is important for the correct patterning of embryonic structures through the regulation of cellular processes including cell growth, differentiation and apoptosis 105 106. In mature organs RA signalling regulates and maintains cellular differentiation 239. As in many cancers, including PC, cells appear to adopt a less differentiated phenotype; hence aberrant RA signalling is a potentially important contributor to the carcinogenic process.
Retinoids have been successfully used in the treatment of acute promyelocytic leukaemia (APL) 240. Although the use of retinoids in the treatment of other cancers has been disappointing, it is becoming apparent, that resistance to treatment by retinoids in these cancers may be due to aberrations in molecules involved in transducing the retinoid signal. Figure 5.2: Retinoids are transported extracellularly by extracellular retinoid binding proteins (ERBP), stored interacellularly bound to cellular retinoid binding proteins (CRBP), and are then metabolised by Alcohol Dehydrogenase and subsequently Retinaldehyde Dehydrogenase to RA which is bound to cellular retinoic acid binding proteins (CRABP) intracellularly, and extra-cellularly to retinoic acid binding proteins (ERABP). The retinoid signal is transduced by two families of nuclear transcription factors: RARs and RXRs. These receptors are members of the nuclear receptor superfamily, which in the presence of ligand heterodimerise to activate the transcription of target genes through RA response elements (RARE). 97
Molecular aberrations in receptors and transducers of the RA signal in particular CRBP1 and RAR2, have been associated with cancers of the breast, lung, prostate, oesophagus and skin 240. Despite the compelling evidence suggesting aberrant retinoid signalling in the literature and from transcript profile and cohort studies associating aberrant RA signalling in PC development and substantial interest in retinoid action in other cancers, little work has been done in the case of PC. Treatment with retinoids induces a moderate anti-proliferative and apoptotic effect in some PC cell lines. However, beyond this little is known about RA signalling in PC.
This chapter examines a potential role of aberrant retinoid signalling in PC, in order to investigate the potential for novel targeted therapies, early diagnostic and chemoprevention strategies involving retinoid signalling in PC.
Aberrant Retinoic Acid Signalling in PC
Transcript profiling identified aberrant expression of a large number of RA signalling components in PC compared to normal pancreas. Differential expression was identified in RA synthesis proteins, extracellular RA carrier proteins, Retinoid receptors, Cellular Retinoid Binding Proteins and downstream genes known to be aberrantly expressed in PC, as well as novel genes regulated by RA but not studied in PC. Initially semiquantitative PCR was performed to validate aberrant expression of key retinoic acid receptors (RAR,RAR,RAR,RXR,RXR, RXR) and CRBP1 a key retinoid binding protein in PC cell lines.
Six PC cell lines (AsPc-1, Mia-PaCa-2, Panc-1, CAPAN-2, HPAC, BxPC-3) and two breast cell lines MCF-7 known to have low/absent expression of CRBP1 and MDA-MB- 468 known to express CRBP1 (positive control) were assessed. Heterogenous expression of RA signalling components was identified in PC cell lines (Figure 5.3). PC Cell Lines
MIA-PaCa-2 PANC-1 HPAC BxPC-3 MCF-7 FaDu MDAMB-468
AsPC-1 Capan-2
RARα
RARβ
RARγ
RXRα
RXRβ
RXRγ
CRBP1
No Expression Low Intermediate High
Figure 5.3: Semiquantitative RT-PCR demonstrating heterogeneous expression of retinoid receptors. 99
Hence, validation, using RT-PCR in PC cell lines, demonstrated differential expression of RA receptors, with downregulation of RAR and RXR and upregulation of RAR. Four of 6 PC cell lines demonstrated loss of CRBP1 expression (Figure 5.3), with low expression noted in Panc-1 and intermediate expression noted in BxPC-3. This recapitulates data found in microdissected early PanIN, where a 3 fold downregulation of CRBP1 was demonstrated by RT-PCR 241. Suggesting that aberrant RA signalling may be an early event in the development of PanIN.
As aberrations in signalling pathways important in pancreas development such as Notch and Hedgehog are important early events in the development of PanIN and hence PC, RA signalling was investigated further. Specific components of RA signalling were investigated based on their potential utility as diagnostic and therapeutic targets. These are:
1. Retinoic Acid Induced 3 (RAI3): an orphan G-protein coupled cell surface receptor normally expressed in the fetal lung and found to be upregulated 152 fold in PC compared to normal pancreas. RAI3 is not expressed in normal tissues and is a potential novel therapeutic target for antibody-mediated therapy.
2. Homeobox B2 (HOXB2): a developmental transcriptional factor normally associated with hindbrain development and not normally expressed in the pancreas but identified to be upregulated 26 fold in PC compared to normal pancreas, and not known to be expressed in any endodermal derivatives. This will be further evaluated in Chapter 6. 100
The Expression of the G-Protein Coupled Receptor Retinoic Acid Induced-3 (RAI3) in PC
Retinoic Acid Induced 3 (RAI3, RAIG1, GPCR5A) is a member of a family of putative G-protein-coupled receptors (GPCR), including Retinoic Acid Induced GPCR 2 and 3 (RAIG-2 and RAIG-3) (HGMW-approved symbols GPCR5B and GPCR5C, respectively) 242. These proteins have been identified as type 3 GPCR family members, which include metabotropic glutamate receptors, GABA (B) receptors, calcium-sensing receptors, and pheromone receptors. However, these RAIGs have short extracellular amino-terminal domains compared with other members of this family, which have very long amino-terminal domains 242. All three RAIG genes can be induced rapidly by all-trans retinoic acid (ATRA), suggesting that these genes may be direct targets for transcriptional regulation by retinoic acid 242,243.
The induction of RAI3 by RA is a rapid event in several cell lines 242. This implies that the RAI3 gene promoter may contain cis-acting elements that interact directly with retinoic acid receptors. Several methods, including reporter activation and site-directed- mutagenesis, have identified a novel functional RARE in the proximal promoter of RAI3. This RARE is a direct repeat of two core motifs separated by a 5-bp spacer (DR5) like a typical RARE 244. RAR’s can form heterodimers with RXR’s, and RXR’s can also form homodimers. Such dimers can bind to retinoic acid response elements (RAREs) in the regulatory regions of target genes, thereby activating or repressing transcription, depending on the presence of ligands and binding of corepressors or coactivators245-247. The rapid induction of RAI3 by RA 243 further suggests it is a direct transcriptional target of RA action.
RAIG-1 (RAI3) is found predominantly in the lung, RAIG-2 is expressed predominantly in human brain, and RAIG-3 is ubiquitously expressed in human peripheral tissue 242,248. Although RAI3 is expressed predominantly in normal fetal and adult lung, it is expressed at low levels in other tissues, making RAI3 a potential 101 diagnostic and therapeutic target in PC. RAI3 expression in a cohort of 116 patients was investigated using in-situ hybridization.
Cohort Analysis
The cohort consisted of 116 patients with the diagnosis of PC and was a subset of the PC cohort presented in Chapter 4. Although there is slight variation in the number of patients reported the relationships between clinicopathological markers and outcome remain unchanged further detailed analysis is presented in Table 5.1.
Whole Cohort
RAI3 mRNA expression was identified in 80 (69%) of 116 tumours representing the whole cohort (Figure 5.4). RAI3 mRNA expression in the whole cohort was associated with an improved outcome (median survival 8.9 months and 4.4 months logrank, p = 0.026: Figure 5.5). Multivariate analysis using the Cox proportional hazards model for those factors that demonstrated a significant effect on survival on univariate analysis, identified resection as the only independent prognostic factor when modelled together with RAI3 status and UICCstage(Table5.2A&B). 102
Table 5.1 Clinicopathological, RAI3 molecular and outcome data for Patients within the Cohort.
Whole Resected Parameter Cohort n= 116 Cohort n=75 Median pvalue Median pvalue No. (%) Survival No. (%) Survival (months) (logrank) (months) (logrank) Sex Female 47 (40) 31 (41) Male 69 (60) 44 (59)
Age at diagnosis (years) Mean 64.3 62.3 Median 66.5 65.0 Range 34.4 – 83.8 34.4 – 82.6
Treatment Resection 75 (65) 10.8 Operative Biopsy 41 (35) 3.4 < 0.0001
Outcome Follow-up (months) 0 - 117 0.2 - 117 Median 7.7 11.0 30 day mortality 2 (2) 2 (3) Death from PC 102 (89) 62 (83) Death from other cause 2 (2) 2 (3) Alive 8 (6) 8 (11) Lost to follow-up 4 (3) 3 (4)
Stage 114 I 27 (24) II 10 (9) 12.2* III 64 (56) IV 12 (11) 6.8 0.0006
Differentiation 116 Well 10 (9) 7 (9) Moderate 64(55) 8.8** 44 (58) 12.2 Poor 42 (36) 6.0 0.061 24 (33) 9.1 0.086
Tumor size 20mm 15 (20) 17.1 >20mm 60 (80) 9.9 0.04
Margins Clear 40 (53) 14.5 Involved 35 (47) 8.6 0.002
Lymph node status§ Negative 34 (46) 13.7 Positive 39 (51) 9.7 0.04
RAI3 expression Low 36 (31) 4.4 10 (11) 10.8 High 80 (69) 8.9 0.026 65 (89) 10.8 0.98
* Stage I and II tumours versus stage III and IV ** Well and Moderately differentiated tumours grouped together for analysis § Lymph node status was only available in 73 patients in the resected cohort A B
C D
Figure 5.4: Photomicrographs of RAI3 ISH staining (x 200) A: Negative RAI3 expression in PC using sense ISH probe (negative control) B: Positive RAI3 expression in same duct using antisense ISH probe C: Negative RAI3 expression in PC D: Strongly positive RAI3 expression in PC .
p = 0.026 1 Median Survival:8.9 Vs4.4 months n=116 .8 .6 RAI3+ve
.4 CUMULATIVE SURVIVAL .2
0
0 20 40 60 80 100 120 MONTHS Figure 5.5: Kaplan-Meier survival curve for RAI3 ISH expression in the whole cohort identifying survival benefit for RAI3 +ve tumors. 104
Table 5.2: Multivariate analysis for clinicopathological parameters associated with outcome and RAI3 expression in the whole cohort of PC.
Variable Hazard Ratio (95% confidence p-value interval)
A. Whole cohort (n =114) RAI3 expression 1.19 (0.69 – 2.04) 0.52 Operative resection 0.24 (0.13 – 0.44) <0.0001 Stage III/IV vs I/II 1.62 (0.97 – 2.70) 0.063
B. Whole cohort (n =114) Operative resection 0.27 (0.16 – 0.45) <0.0001 Stage III/IV vs I/II 1.65 (0.99 – 2.74) 0.053
Note: Because resection is a strong independent prognostic factor RAI3 expression and stage were not identified as independent prognostic factors (A). Resection is the only independent prognostic factor when the model is resolved (B)
Resected Cohort
RAI3 expression was present in 65 patients (87%) because such a large portion of the resected cohort expressed RAI3 there was no statistically significant difference in outcome identified when stratified by RAI3 expression status (logrank p =0.98). Further analysis identified that RAI3 expression was associated with resection (2 p value < 0.0001). There was no correlation identified between RAI3 status and stage, differentiation, margin status and lymph node involvement in either the whole and the resected cohort. 105
Discussion
In summary, the data presented in this chapter suggests a potential role of aberrant RA signalling in pancreatic carcinogenesis.
Aberrant expression of the 2 families of retinoid receptors and a key cellular retinoid transport protein was identified in PC cell lines, which may modulate the effect of the retinoid signal on downstream effector molecules. This provided the impetus to investigate specific components of RA signalling based on their potential utility as diagnostic, prognostic or therapeutic targets.
TranscriptprofiledataidentifiedupregulationofRAI3inPCcomparedtonormal pancreas (Fold Change = 26; p value = 0.007). In-situ hybridisation identified overexpression of RAI3 mRNA in 68% of 116 PC, which was associated with an improved survival (log-rank p = 0.026). Overexpression of RAI3 was associated with resectability with 89% of resected pancreata overexpressing RAI3. Fifty to 75% of PC contain mutations of p53 249. Recent data demonstrate that p53 binds to the promoter region of RAI3 causing repression of transcription. In most cell lines with mutant p53 RAI3 is upregulated and conversely is present only in low levels when functional p53 is present. Furthermore, ectopic expression of RAI3 leads to anchorage independent growth, and small RNA-mediated depletion of RAI3 in AsPC3 PC cells induces morphological changes 249. There is almost undetectable expression of RAI3 in normal organs except the lung 250. This specificity makes this cell surface receptor, which is overexpressed in a high proportion of PC, an ideal target for the development of novel diagnostic and therapeutic strategies for PC. As a diagnostic target, passive immunity may be used to design monoclonal antibodies to RAI3 allowing for the use of immunodiagnostics (radioimmunoscintography) as has been used in breast and head and neck cancers to improve the early detection of PC. Therapeutically, active immunity against a RAI3 positive cancer using a peptide based approach, may lead to the development of antineoplastic agent tagged antibodies directed against PC cells. 106
Conclusion
Although the mechanisms by which aberrant RA signalling described above, may contribute to carcinogenesis remain to be fully elucidated, a better understanding of these mechanisms may allow for the development of novel therapeutic and diagnostic strategies in PC, providing scope for the potential use of retinoids in the treatment of PC. 107
Chapter 6
THE HOMEOBOX TRANSCRIPTION FACTOR: HOXB2 in PANCREATIC CANCER 108
Introduction
Homeobox genes are transcription factors with established roles in development and cell function. Mammalian development requires a complex interaction of HOX gene networks, with HOX gene expression commencing during gastrulation and collectively controlling the identity of various regions along the body axis from the hindbrain to the tail 126,251. Aberrant expression of HOX genes has been implicated in the development of solid tumours including renal carcinoma 131, colon cancer 132, ovarian carcinoma 134 and breast carcinoma 136,137.
Evidence to suggest a potentially important role for HOXB2 in PC was observed in the expression profile data, which recapitulated known developmental mechanisms associated with the developing hindbrain. Retinoic acid activation of krox20 decreases the expression of HOXB1 with a complementary increase in the expression of HOXB2 thus determining rhombomere identity in the developing hindbrain (Figure 6.1). HOXB2 was upregulated 6.7 times in PC compared to normal pancreas. Furthermore, a variant fusion protein thought to be the prime carcinogenic stimulus in a subset of acute promelocytic leukaemias (APL) binds to the promoter region of HOXB2 inducing its transcription.
The expression of HOXB2 was examined in both PC and PanIN and its relationship to patient outcome and clinicopathological parameters was assessed in a cohort of patients with pancreatic ductal adenocarcinoma. In addition, the expression of HOXB2 in PC cell lines was also investigated, which demonstrated that all PC cell lines expressed HOXB2 mRNA and protein, and that HOXB2 ‘knockdown’ using siRNA resulted in an altered phenotype, but did not alter proliferation. Extracellular RA carrier proteins RA synthesis Retinoid RBP4 NC Receptors hRADH 4.7 sRABP 2.6 RALDH2 NC RAR α 2.9 RAR β NC RAR γ 2.2 RA degradation RXR α 0.4 RXR β NC CYP26 NC Cellular Retinoic Acid RXR γ NC Binding Proteins
CRABP1 0.2 CRABP2 2.6 CRALBP NC
CEBPβ 2.1
Cellular Retinoid Binding Proteins
CRBP1 0.2 CRBP2 NC krox20 3.2 CRBP3 NC CRBP4 NC
HOXB1 0.4 HOXB2 6.7
?
ZNF145/PLZF 0.3
Genes modulated by RA and associated with development
Figure 6.1: A customized GenMAPP of RA signalling components: with statistically significant relative expression levels in PC compared to normal pancreas denoted beside the gene. Relative expression levels are represented as fold change with those with statistically significant upregulation marked red ,those with significant downregulation marked blue and those with no significant change marked green for: retinoic acid receptors; cellular retinoic acid binding proteins (CRABP); and downstream targets modulated by RA associated with embryonal hindbrain development. krox20 stimulates HOXB2 transcription whilst downregulating the expression of HOXB1. Abbreviations not mentioned in text: RBP4, retinoid binding protein 4; CRALBP, cellular retinaldehyde binding protein; RALDH, retinaldehyde dehydrogenase. 110
Cohort characteristics
The cohort consisted of 128 patients with the diagnosis of PC and consists of the cohort presented in Chapter 4. Although there is slight variation in the number of patients reported, the relationships between clinicopathological markers and outcome remain unchanged further detailed analysis is presented in Table 6.1.
The Whole Cohort
HOXB2 expression defined as homogeneous nuclear staining in >20% of cells were identified in 48 (37.5%) of 128 tumours (Figure 6.2). HOXB2 expression was identified in the normal ducts of 2 (11%) of 18 pancreata validating the transcript profile data (2 p= 0.027). HOXB2 expression was detected in 1 of 24 (4%) PanIN-1A lesions, 3 of 20 (15%) PanIN-1B, 3 of 10 (30%) PanIN-2 and 1 of 4 (25%) PanIN-3 lesions, suggesting, that HOXB2 expression occurs in PanIN and may play a role in the evolution of PanIN.
HOXB2 expression was associated with a poor prognosis (median survival 5.0 vs 9.5 months; logrank p = < 0.0001) (Figure 6.3A). Operative resection did not benefit those patients whose tumours overexpressed HOXB2 (logrank p = 0.37 Fig 6.3B), but was beneficial to those patients who did not express HOXB2 (median survival of 14 months versus 3.7 months, logrank p < 0.0001 Fig 6.3C). Survival for patients with tumours that were HOXB2 negative and who underwent resection was significantly longer than survival in all other groups 14 months versus 4.3 months (logrank p < 0.0001 Fig 6.3D). Hence in this cohort lack of HOXB2 co-segregated with operative resectability. Importantly, only those who were HOXB2 negative benefited from operative resection. 111
Table 6.1: Clinicopathological, HOXB2 molecular, and outcome data for All Patients Within the Cohort.
Parameter Whole n= 128 Resected n=76 Cohort Cohort Median pvalue Median Survival pvalue No. (%) Survival No. (%) (months) (logrank) (logrank) (month s )
Sex Female 56 (44) 31 (41) Male 72 (56) 45 (59)
Age at diagnosis (years) Mean 63.8 61.0 Median 66.5 65.0 Range 34.0 – 86.0 34.0 – 83.0
Treatment Resection 76 (59) 11.0 Operative Biopsy 46 (36) 3.9 < 0.0001
Outcome Follow-up (months) 0 - 117 0.2 - 117 Median 7.6 11.0 30 day mortality 2 (2) 2 (3) Death from PC 114 (89) 63 (83) Death from other cause 2 (2) 2 (3) Alive 8 (6) 8 (11) Lost to follow-up 4 (3) 3 (4)
Stage 127 I 27 (21) II 13 (10) 13.7* III 70 (55) IV 17 (13) 6.4 < 0.0001
Differentiation 127 Well 11 (9) 7 (9) Moderate 68 (53) 8.9** 44 (58) 12.2 Poor 48 (38) 5.0 0.00152 25 (33) 8.6 0.00582
Tumor size 20mm 15 (20) 17.1 >20mm 61 (80) 9.7 0.0375
Margins Clear 40 (53) 14.5 Involved 36 (47) 8.5 0.0014
Lymph node status§ Negative 35 (46) 13.8 Positive 39 (51) 9.2 0.0235
HOXB2 nuclear expression Positive 48 (38) 5.0 16 (21) 6.8 Negative 80 (72) 9.9 < 0.0001 60 (79) 14.0 < 0.0001
* Stage I and II tumours versus stage III and IV ** Well and Moderately differentiated tumours grouped together for analysis § Lymph node status was only available in 74 patients in the resected cohort A B
C D
E F
Figure 6.2: Photomicrographs of HOXB2 nuclear staining (x 200) A: Negative HOXB2 expression in normal pancreatic duct B: Positive nuclear staining in PanIN-1B lesion C: Positive nuclear staining in PanIN-3 lesion. D: Positive nuclear staining in PC E: Negative nuclear staining in PanIN-1A lesion F: Negative nuclear staining in PC A. HOXB2 Nuclear Expression B. HOXB2 Positive p < 0.0001 p = 0.3697 1 1 Median Survival: 9 . 9 Vs 5 .0 months Median Survival: 6 . 7 Vs 4 .4 months
.8 n=128 .8 RESECTION n=48
.6 .6 BIOPSY
.4 .4 CUMULATIVE SURVIVAL CUMULATIVE SURVIVAL .2 .2 HOXB2 -ve
0 HOXB2 +ve 0
0 20 40 60 80 100 120 0 5 10 15 20 25 30 MONTHS MONTHS
C. HOXB2 Negative D. Stratification of HOXB2 expression with resection. p < 0.0001 1 p < 0.0001 1 Median Survival: 14 . 0 Vs 3 .7 months Median Survival: 14 . 0 Vs 4 .3 months
.8 n=80 .8 n = 128
.6 .6
.4 .4 CUMULATIVE SURVIVAL .2 CUMULATIVE SURVIVAL .2 HOXB2 NEGATIVE with RESECTION RESECTION
0 BIOPSY 0
0 20 40 60 80 100 120 0 20 40 60 80 100 120 MONTHS MONTHS
E. HOXB2 Nuclear Expression (resected cohort)
1 p < 0.0001 Median Survival: 14 . 0 Vs 6 .75 months .8 n=76 .6
.4
CUMULATIVE SURVIVAL .2 HOXB2 -ve
0 HOXB2 +ve
0 20 40 60 80 100 120 MONTHS
Figure 6.3: Kaplan-Meier survival curve for the whole cohort: identifying (A) HOXB2 nuclear expression significantly associated with outcome. Kaplan- Meier survival curves reflecting the effect of resection on prognosis in subgroups: (B) HOXB2 positive; (C) HOXB2 negative; (D) all patients stratified for HOXB2 status and resection. These illustrate that cancers that express HOXB2 do not show any survival advantage when resected however those cancers that do not express HOXB2 have a significantly improved outcome when resected (C and D). Kaplan-Meier survival curve for the resected cohort, identifying (E) HOXB2 nuclear expression significantly associated with outcome. 114
Table 6.2 Univariate Analyses assessing clinicopathological and molecular parameters with outcome in the Whole cohort
Hazard’s Variable n Ratio 95% CI p HOXB2 Expression 128 2.65 1.76 – 3.98 < 0.0001 Resection 128 0.30 0.20 – 0.45 < 0.0001 Stage III/IV v Stage I/II 127 2.59 1.63 – 4.10 < 0.0001 Poor v Well/Moderate 128 1.62 1.09 – 2.41 0.017
Multivariate analysis using the Cox proportional hazards model for those factors that demonstrated a significant effect on survival on univariate analysis (Table 6.2), identified resection and stage as the only independent prognostic factors when modelled together with degree of differentiation and HOXB2 status (Table 6.3A & B).
When HOXB2 expression was modelled with molecules previously examined in this thesis with the whole cohort of patients, HOXB2, S100A6 and S100A2 remained independent prognostic factors (Table 6.3 J), Resolution of the model with step-wise deletion of markers not shown to be independent predictors of outcome in multivariate models examined in Chapters 4 and 5, identified HOXB2, S100A2 and resection as independent prognostic outcome in the whole cohort (Table 6.3 K). 115
Table 6.3: Multivariate Analyses for Clinicopathologic Parameters and HOXB2 Expression for all patients in the cohort.
Variable Hazard Ratio p-value (95% confidence interval)
A. Whole cohort (n =127) Stage III/IV vs I/II 2.21 (1.37 – 3.57) 0.0012 Resection 0.44 (0.26 – 0.74) 0.0019 HOXB2 Expression 1.61 (0.97 – 2.67) 0.0659 Differentiation 1.31 (0.88 – 1.96) 0.1877 B. Whole cohort (n =127) Resection 0.33 (0.22 – 0.51) < 0.0001 Stage III/IV vs I/II 2.17 (1.34 – 3.51) 0.0016 Differentiation 1.28 (0.85 – 1.91) 0.2350 C. Whole cohort (n =127) Resection 0.33 (0.22 – 0.50) < 0.0001 Stage III/IV vs I/II 2.3(1.42–3.67) 0.0007 D. Whole cohort (n =128) Resection 0.31 (0.21 – 0.47) < 0.0001 Differentiation 1.45 (0.97 – 2.16) 0.0690 E. Whole cohort (n =127) Stage III/IV vs I/II 2.41 (1.51 – 3.86) 0.002 Differentiation 1.33 (0.89 – 2.00) 0.163 F. Whole cohort (n =128) Differentiation 1.60 (1.08 – 2.38) 0.0204 HOXB2 Expression 2.63 (1.75 – 3.95) < 0.0001 G. Whole cohort (n =127) Stage III/IV vs I/II 2.44 (1.53 – 3.90) 0.0002 HOXB2 Expression 2.50 (1.65 – 3.79) < 0.0001 H. Whole cohort (n =127) HOXB2 Expression 2.54 (1.67 – 3.85) < 0.0001 Stage III/IV vs I/II 2.29 (1.42 – 3.67) 0.0006 Differentiation 1.39 (0.93 – 2.08) 0.1051 I. Whole cohort (n =128) Resection 0.39 (0.23 - 0.65) 0.0004 HOXB2 Expression 1.52 (0.91 – 2.54) 0.1124
J. Whole cohort (n =106) S100A2 Expression 3.02 (1.75 – 5.21) <0.0001 S100A6 Expression 0.48 (0.27 – 0.86) 0.0134 S100P Expression 0.96 (0.56 – 1.66) 0.8981 RAI3 Expression 1.06 (0.58 – 1.91) 0.8537 HOXB2 Expression 2.63 (1.54 – 4.49) 0.0004 Resection 0.44 (0.23 – 0.85) 0.0146 Stage III/IV vs I/II 1.64 (0.96 – 2.80) 0.0698 Differentiation 0.84 (0.53 – 1.33) 0.4588
K. Whole Cohort (n=114) S100A2 Expression 2.15 (1.41 – 4.85) 0.0017 HOXB2 Expression 2.25 (1.30 – 3.90) 0.0037 Resection 0.31 (0.17 – 0.54) <0.0001
Note: Degree of differentiation and HOXB2 expression were not independent prognostic factors when modelled with resection and stage (A). Tumour stage was independent of resection (A, B and C). Differentiation was not independent of stage or resection when modelled together (B) or when modelled with resection and stage separately (D and E). HOXB2 remained an independent prognostic marker when modelled against stage and degree of differentiation separately and together (F, G and H). Because resection was the strongest independent prognostic factor in this cohort HOXB2 expression was not identified as an independent prognostic factor when modelled against resection (Table 6.3, H). 116
Resected Cohort
HOXB2 nuclear expression was observed in 16 of 76 (21%) resected pancreata. Its relationship to other parameters is shown in Table 6.4, there is a relationship between HOXB2 nuclear staining and lymph node involvement (2 p = 0.009).
Table 6.4. Contingency Tables Relating HOXB2 expression to Clinicopathologic Parameters associated with outcome in Resected Pancreata.
HOXB2 HOXB2 p Characteristic Positive Negative (2) Sex Male 8 37 0.39 Female 8 23 Tumour Size 20 mm 2 13 0.41 <20mm 14 47 Margin Clear 7 33 0.42 Involved 9 27 Nodal Status Positive 13 26 0.009 Negative 3 32 Differentiation Well/Moderate 9 18 0.30 Poor 7 42 Diagnosis Date 1972-1989 0 6 0.19 1990-2001 16 54
Note: HOXB2 expression was associated with node positive disease (2 p value = 0.009) and may be a surrogate marker for lymph node involvement in PC.
HOXB2 nuclear expression was associated with a poor prognosis (median survival 6.7 vs 14 months, logrank p<0.0001) (Figure 6.3E). 117
Table 6.5. Univariate analyses assessing clinicopathological and molecular parameters with outcome in the resected cohort
Hazard’s Variable Ratio 95% CI p HOXB2 Expression 3.34 1.77 – 6.32 0.0002 Margins: Involved v Clear 2.33 1.37 – 3.99 0.0019 Lymph Nodes: Involved v Clear 1.85 1.08 – 3.18 0.0258 Tumour >20mm v Tumour < 20mm 1.95 1.03 – 3.71 0.0410 Differentiation Poor v Well/Moderate 1.66 0.98 – 2.84 0.0612
Multivariate analyses of those parameters found to be prognostic on univariate analysis (Table 6.5) are shown in Table 6.6. HOXB2 nuclear expression was an independent prognostic factor on multivariate analysis. HOXB2 overexpression and involved surgical margin were independent prognostic factors when modelled against lymph node involvement and tumour size individually (Table 6.6 A & B). HOXB2 nuclear expression maintains independence as a prognostic factor when modelled against various combinations of, involved surgical margin, lymph node involvement, tumour size, and tumour differentiation (Table 6.6 C – F).
When HOXB2 expression was modelled with molecules previously examined in this thesis with the resected cohort of patients, HOXB2 and S100A2 remained independent prognostic factors (Table 6.6 G), Resolution of the model with step-wise deletion of markers not shown to be independent predictors of outcome in multivariate models examined in Chapters 4 and 5, revealed that HOXB2, S100A2 and surgical margin involvement as independent prognostic outcome (Table 6.6 H). 118
Table 6.6: Multivariate analysis for clinicopathological parameters and HOXB2 nuclear expression in resected pancreata.
Variable Hazard Ratio p-value (95% confidence interval) A. Resected Cohort (n=76) HOXB2 Expression 2.90 (1.51 – 5.57) 0.0014 Margin Involvement 1.89 (1.02 – 3.48 0.0428 Lymph Node Involvement 1.30 (0.71 – 2.40 0.3981 B. Resected Cohort (n=76) HOXB2 Expression 2.82 (1.48 – 5.40) 0.0017 Margin Involvement 2.04 (1.17 – 3.53) 0.0115 Tumour Size > 20mm 1.48 (0.75 – 2.90) 0.2567 C. Resected Cohort (n=76) HOXB2 Expression 2.69 (1.39 – 5.20) 0.0032 Margin Involvement 1.75 (0.94 – 3.25) 0.0777 Lymph Node Involvement 1.34 (0.73 – 2.46) 0.3525 Tumour Size > 20mm 1.49 (0.76 – 2.94) 0.2474 D. Resected Cohort (n=76) HOXB2 Expression 3.26 (1.66 – 6.40) 0.0006 Margin Involvement 1.72 (0.93 – 3.18) 0.0818 Lymph Node Involvement 1.24 (0.68 – 2.27) 0.4867 Differentiation 1.51 (0.85 – 2.70) 0.1639 E. Resected Cohort (n=76) HOXB2 Expression 3.14 (1.61 – 6.14) 0.0008 Margin Involvement 1.80 (1.02 – 3.19) 0.0434 Tumour Size > 20mm 1.48 (0.75 – 2.89) 0.1564 Differentiation 1.51 (0.85 – 2.68) 0.2567 F. Resected Cohort (n=76) HOXB2 Expression 3.01 (1.52 – 5.96) 0.0015 Margin Involvement 1.60 (0.86 – 2.97) 0.1391 Lymph Node Involvement 1.28 (0.70 – 2.34) 0.4262 Tumour Size > 20mm 1.48 (0.75 – 2.92) 0.2539 Differentiation 1.50 (0.84 – 2.68) 0.1683 G. Resected Cohort (n = 70) S100A2 Expression 2.61 (1.30 – 4.37) 0.0023 S100A6 Expression 0.72 (0.34 – 1.53) 0.3987 S100P Expression 0.96 (0.45 – 2.02) 0.9091 RAI3 Expression 1.17 (0.49 – 2.81) 0.7236 HOXB2 Expression 2.91 (1.39 – 6.07) 0.0044 Tumor size > 20 mm 1.48 (0.72 – 3.06) 0.2906 Margin Involvement 1.82 (0.94 – 3.52) 0.0739 Lymph Node Involvement 1.16 (0.60 – 2.25) 0.6542
H. Resected Cohort (n = 74) S100A2 Expression 2.71 (1.50 – 4.90) 0.0009 HOXB2 Expression 3.75 (1.92 – 7.31) 0.0001 Margin Involvement 2.06 (1.20 – 3.54) 0.0091
Note. HOXB2 expression and margin involvement remains an independent prognostic marker in the resolved multivariate model when modelled against tumour size and lymph node involvement either individually or together (A, B and C), HOXB2 expression continues to maintain an independent status when modelled against various combinations of margin involvement, tumour size, lymph node involvement and differentiation either separately or together (D, E and F). Inclusion of all molecules examine in previous chapters in the multivariate model identified S100A2 and HOXB2 as independent markers of outcome, together with margin involvement when the model is resolved with the removal of factors that did not maintain independence in previous multivariate analyses (G and H). 119
HOXB2 Expression In Pancreatic Cancer Cell Lines.
Although HOXB2 expression has been investigated in breast cancer cell lines and embryonal neural cell lines 137 there are no data available on HOXB2 expression in normal pancreas and PC cell lines. Before studies could be conducted to elucidate the function of HOXB2 in cell lines endogenous expression of HOXB2 in these cell lines was investigated. In addition, western blot analysis with the HOXB2 antibody used for IHC allowed for the assessment of antibody specificity.
HOXB2 mRNA Expression in PC Cell Lines
HOXB2 expression was assessed using Reverse Transcriptase Polymerase Chain Reaction (RT-PCR) in head and neck cancer (Detroit, FaDu), breast cancer (MCF-7, MCF10A, BT-20, SkBr-3, MDA-MB-330, ZR-75-1 and 184), prostate cancer (LnCaP, Du-145), colon cancer (HLT-15), normal ovary (HOSE 6-3), ovarian cancer (LoVo, CaOV3) and PC (HPAC, CAPAN-2, Panc-1, Bx PC3, MiaPaCa-3, AsPC) cell lines. Differential expression of HOXB2 was demonstrated between these cancer cell lines (Figure 6.4). Expression of HOXB2 was seen in LnCAP, HLT-15, ZR-75-1, 129-4, 184 and HOSE-3 cells. All 6 PC cell lines demonstrated expression of HOXB2, providing further evidence to suggest a role for HOXB2 in PC.
HOXB2 protein expression in PC cell lines.
The protein expression of HOXB2 was investigated in a subset of cell lines described above based on the RNA expression of HOXB2. These include the head and neck cancer cell line Detroit and the breast cancer cell line MCF-7 used as putative negative controls and the PC cell lines HPAC, CAPAN-2, Mia-PaCa3, ASPC, Panc-1 and Bx- PC3. Differential protein expression was confirmed (Figure 6.4), with no expression of HOXB2 identified in Detroit and MCF-7 cell lines. Expression of HOXB2 was identified in all pancreatic cell lines examined with the highest level identified in the Panc-1 cell line. The Panc-1 cell line was selected for studies involving downregulation of expression of HOXB2 using siRNA. HOXB2 HOXB2 + sc - 17165 P peptide
B. 1. 2.
B-330
HPAC LnCAP AsPc Du - 145
MCF - 10A
BT-20
CAPAN -2 184 Panc - 1 SKBR-3 Ca OV3 HEC -1 -A HOSE 6-3
FaDu
MCF-7 Bx-Pc-3 LoVo ZR - 751 Mia-PaCa -2
Detroit HCT-15
M HMEC- - 219 - 4
A.
t
i
tro
e
Bx-Pc -3
AsPc MCF-7
D
HPAC
Mia-PaCa 2
Panc - 1
CAPAN-2
C. HOXB2 HOXB2 + s c-17165 P peptid e β − Actin 38kDA
A A
etroit etroit
HPAC HPAC
aCa -2
D D
PaCa -2 P
a- a-
MCF-10 MCF-10
Mi Mi
Figure 6.4: A: Semiquantitative RT-PCR for HOXB2 in cell lines; with Western Blot of selected cell lines; Note correlation of RT-PCR data with expression on Western Blotting: B: Immnohistochemistry for HOXB2 in serial sections of PC assessing antibody specificity using peptide absorbance, 1; no peptide added, 2; with blocking peptide sc-17165 P added. C: Western Blot using anti- HOXB2 antibody demonstrating loss of HOXB2 detection with incubation of antibody with blocking peptide sc-17165 P, cell lines used, were MCF10A, breast cancer cell line positive for HOXB2 on RT-PCR, MiaPaCa-2 and HPAC pancreatic cancer cell lines positive for HOXB2 on RT-PCR and Detroit a head and neck cancer cell line weakly positive for HOXB2 on RT-PCR. Note loss of signal post absorption with blocking peptide sc-17165 P at 38 kDa 121
Knockdown of HOXB2 expression in Panc-1 cells
In 1998 researchers observed that specific gene expression can be silenced by injecting homologous long double stranded RNA (dsRNA) into C.elegans 252. This dsRNA is cleaved by the RNase Dicer to create small interfering RNA (siRNA) 253. The resultant siRNA is incorporated into a multiprotein RNA inducing silencing complex (RISC) where the siRNA duplex is unwound and the antisense sequence allowed to target the complex to the corresponding homologous mRNA. The mRNA is cleaved at a site 10 nt from the 5’ end of the siRNA sequence 254, which targets the mRNA for complete degradation in the cytoplasm and results in decreased protein expression 255. In the non-diseased state, siRNA are produced by cells to degrade dsRNA produced by transposons, viruses and endogenous genes 256. Introduction of synthetic siRNA into the human cell did not induce global inhibition of mRNA translation and resulted in a viable cell 257, leading to the use of siRNA transfection as a method for depleting human cells of specific mRNA species (Figure 6.5).
Commercially prepared SiRNA for HOXB2 as described in Chapter 2 was acquired for use in these experiments. The design of the sense and antisense strands conform to most but not all of the following rules which resulted in a 21 nt double stranded RNAs that contain 19 base pairs homologous to the target sequence with 2nt 3’ overhangs 254:
1. 30%-52% GC content. 2. Three or more A/Us at positions 15-19 (sense) 3. The absence of internal repeats or hairpins as measured by a Tm < 20 degrees C. 4. A at position 19 (sense) 5. A at position 3 (sense) 6. U at position 10 (sense) 7. No G/C at position 19 (sense) 8. No G at position 13 (sense) 122
Using the transfection reagent Lipofectamine, 90% transfection efficiency was demonstrated using a cy-3 labelled SiRNA RISC Free (Non Functioning SiRNA) within 24 hours of transfection (Figure 6.6) in Panc-1 cells.
Transfection of Panc-1 cells with HOXB2 siRNA as described in Chapter 2 produced viable cells that continued to proliferate 7 days post transfection. Furthermore a significant reduction in HOXB2 protein expression was observed 72 hours post transfection. (Figure 6.7). This reduction of HOXB2 protein expression was maintained for 168 hours post transfection, during which time cells were actively proliferating.
These were compared with two cell lines; Panc-1 cells that were not transfected with siRNA (control) and, Panc-1 cells transfected with a commercially acquired non functioning siRNA designed not to be incorporated into the RNA inducing silencing complex (RISC free siRNA control), to show that reduction in HOXB2 protein is due to siRNA and not the possible toxic effects of transfection reagents and siRNA to the Panc-1 cell line. dsRNA DICER
P siRNA P
RISC P
Cleavage
mRNA
Figure 6.5: Creation and action of siRNA's. Endogenous small RNAs are created from dsRNA by the action of DICER. SiRNA is then incorporated into the RISC complex which then binds and cleaves homologous target mRNA, which is then degraded in the cytoplasm.
Figure 6.6: Photomicrograph of Panc-1 cells transfected with Flourescent labelled RISC Free (non functioning) siRNA (x100) Note the presence of flourescence in more than 90% of cells in the field, indicating successful transfection using Lipofectamine. Hours Post- Transfection with SiRNA
24 hours 72 hours 120 hours 168 hours Figure 6.7: : Photomicrographs (x100) of control Panc 1 cells, Cells treated with RISC free (non-functioning) siRNA, and Panc-1 Panc-1 cell line (control) cells transfected with HOXB2 siRNA taken 1,3,5,and 7 days after transfection. Note intracellular lipid vesicles A: suggesting transfection of SiRNA and B: epitheloid change noted from day 3 in HOXB2 SiRNA (non-functioning) A transfected cells. Note corresponding Western RISC Free si RNA Blots using anti-HOXB2 antibody demonstrating presence of HOXB2 in the Panc-1 control and RISC free siRNA transfection cell lines, with marked reduction HOXB2 siR NA of HOXB2 expression in HOXB2 siRNA B transfected cells. A
Panc-1 (co ntrol) RISC Free (non-functioning) HOXB2 siR NA siRNA Hours Post Transfecti on 24 72 96 168 24 72 96 168 24 72 96 168
β Actin 125
Morphological Changes observed in Panc-1 Cells after decreasing expression of HOXB2
Repeated experiments identified a demonstrable decrease of HOXB2 protein expression within 24 hours of transfection with cell viability maintained. This ‘knockdown’ of HOXB2 expression was maintained for 7 days post transfection. A dramatic phenotypic alteration was demonstrated, from a fibroblastoid spindle shaped cell found in clumps observed in the Panc-1 (control) and RISC free (non-functioning) siRNA transfected Panc-1 cells. HOXB2 siRNA transfected cells, demonstrated a decrease in HOXB2 protein expression developed a more epithelioid, polygonal shaped cell which tended to be found individually and not clumped in groups (Figure 6.7), compared to RISC free siRNA control transfected cells and Panc-1 control cells, which maintained a fibroblastoid spindle shape.
Proliferation of Panc-1 cells after decreasing expression of HOXB2.
Repeated assays using MTS assay methods identified no difference in proliferation rates between Panc-1 control cells (no transfection), RISC free (non-functioning) siRNA transfected Panc-1 cells and Panc-1 cells transfected with siRNA to HOXB2 (Figure 6.8), over the course of seven days. Proliferation Assay PAnc-1 #3
10
Panc-1 control cell line 1 HOXB2 siRNA RISC free (non-functioning) siRNA
0.1 Day 1 Day2 Day 3 Day 4 Day 5 Time
Figure 6.8: Proliferation curves derived from MTS assays: performed over 5 days using Panc-1 cells (control), RISC Free (non-functioning) siRNA transfected cells and HOXB2 SiRNA transfected Panc-1 cells. Note no change in the gradient of the curves suggesting that "knockdown " of HOXB2 does not affect proliferation in Panc 1 cells 127
Discussion
Expression profiling and RT-PCR of components of RA signalling suggest a potential role in PC. The RA responsive homeodomain transcription factor, HOXB2, was ectopically expressed in a significant proportion of PC. HOXB2, which is not normally expressed in the pancreas at any stage during development or adult life, was expressed in 38% of PC and was present in a significant proportion of PanIN lesions. Ectopic HOXB2 expression was associated with non-resectable tumours and was an independent prognostic factor in resected tumours when modelled with known clinicopathological prognostic factors. In addition, only those patients that were HOXB2 negative obtained a survival advantage with operative resection. Expression of HOXB2 was identified in multiple cancer cell lines including colon cancer, prostate cancer and in normal ovary. All six PC cell lines examined in this study expressed HOXB2 at variable levels. Western blot analysis confirmed protein expression of HOXB2 in all six PC cell lines, including Panc-1. Functional studies using small interfering RNA to HOXB2 demonstrated a decrease in HOXB2 expression in Panc-1 cell lines. This was associated with epithelioid trans- differentiation, which was maintained 1 week after transfection. However, no differences could be elucidated in proliferation rates when HOXB2 expression was decreased in Panc-1 cells compared to control. Although the functional role of HOXB2 is yet to be fully elucidated in PC, expression of HOXB2 is associated with poor prognosis and poor response to resection in PC. The advantage of HOXB2 expression as a prognostic indicator is that it is potentially assessable preoperatively or during staging laparoscopy, whilst all known reliable prognostic indicators such as tumour size, resection margins and lymph node status can only be determined post resection. Pre-operative HOXB2 status may be used in preoperative patient assessment to determine suitability for surgery. A clinical trial assessing HOXB2 for a role in selecting patients for resection has been designed using European Organisation for Research and Treatment of Cancer (EORTC) protocols and has been included in Appendix 5. Briefly the initial role of the study will be to examine if there is a correlation between assessment of HOXB2 nuclear staining in pre-operative PC biopsy samples and HOXB2 nuclear staining in resected pancreata, 128 with subsequent assessment of outcome stratified by HOXB2 expression in pre- operative biopsies. The HOX family of transcription factors have been implicated in the development of many cancers, most recently the development of breast and ovarian cancers. While the downstream effector genes that are activated by the HOX gene network are not well described, they include genes that encode for extracellular matrix components, angiogenic factors and growth factors, which are important in embryonal development and in tumour development and invasion. Although little is known concerning HOXB2 in cancer, there is significant evidence implicating other HOX genes in carcinogenesis. This study provides evidence implicating a potential role of HOXB2 in PC development and progression.
In conclusion, ectopic expression of HOXB2 in PC is a frequent occurrence, an event which manifests in the development of PanIN in a proportion of cases, and is possibly a consequence of aberrant RA signalling. Assessment of HOXB2 expression may provide an alternative method for determining the suitability for resection and the prognosis of patients with PC. Further study to determine the effects of ectopic HOXB2 expression and other components of the HOX transcriptional network, its relationship to RA signalling and clinical utility in pancreatic adenocarcinoma is required. 129
Chapter 7
SUMMARY, GENERAL DISCUSSION and FUTURE DIRECTIONS 130
Key issues in the current management of PC
Pancreatic cancer (PC) is the fourth leading cause of cancer death in men and women in Western societies with a 5-year survival rate of less than 5%. In 2005 PC accounted for over 200,000 deaths worldwide. The mortality of this disease has remained unchanged for the last 30 years. PC continues to present at an advanced stage, and as a result, only 10-20% of patients are suitable for surgical treatment at the time of presentation. Chemotherapeutic agents and radiotherapy have met with little if any success. Thus the treatment and outcome of PC, apart from improvement in peri-operative care, has not changed for more than 3 decades.
This thesis attempts to address several current issues in PC presented in Chapter 1:
First, only a small proportion of patients presenting with PC are suitable for resection because the disease is usually advanced at the time of presentation. Early detection of cancers using data from the molecular characterisation of PC, may lead to the development of early diagnostic strategies, which will increase resection rates and improve outcome in this disease.
Second, as resection remains one of the few therapeutic interventions that affect prognosis in PC, improved selection criteria for resection, which is currently based on anatomical issues, is required to identify patients who are more likely to benefit from this invasive procedure and its attendant morbidities.
Third, there remain no rationally designed, molecularly targeted therapies for PC, such as specific monoclonal antibodies that target specific cell surface receptors, which are currently used successfully in the treatment of other cancers. The molecular characterisation of PC may lead to the development of innovative therapies that are designed using novel therapeutic molecular targets. 131
The data presented in this thesis contributes to our knowledge of the clinical behaviour and the molecular pathology of PC and begins to address some of the key clinical issues presented above. These data have the potential to contribute to clinical strategies that may be developed to improve early diagnosis, aid in patient selection for surgery and improve outcome in patients who develop PC in the future.
Expression profile data identified multiple genes and pathways potentially relevant to the development and progression of PC. Detailed analysis of differential mRNA expression for over 45,000 genes identified 1,787 genes that were aberrantly regulated in PC compared to normal pancreas. Resolution of these lists of genes identified molecular pathways that may play a pivotal role in PC. These included TGF signalling, wnt signalling, the S100 family of calcium binding proteins and RA signalling. The S100 calcium binding protein family and the RA signalling pathway were selected for further investigation.
A substantial number of S100 calcium binding protein family members were differentially expressed in PC compared to normal pancreas, with more than 6 members having a greater than 5-fold upregulation in PC compared to normal pancreas. These included S100A2, S100A6 and S100P.
Furthermore, aberrant expression of components of the RA signalling pathway were identified, including retinoid receptors, cellular retinoid transport proteins and downstream effector molecules regulated by RA signalling. These include genes already known to be important in PC and PanIN as well as novel candidates.
Components of the S100 calcium binding family and RA signalling were selected for validation in a PC cohort and assessed for potential utility as:
1) Markers of early diagnosis, 2) Markers of prognosis and response to therapy 3) Targets for novel therapeutic strategies. 132
Increased expression of S100A2 was associated with a poor outcome in PC. Conversely increased expression of S100P and S100A6 was associated with an improved outcome in the whole cohort. The relationship between S100A2 and resection status, tumour size and lymph node involvement on multivariate analysis suggests that it may be a surrogate marker of advanced disease and hence confer a poor response for those patients undergoing resection.
Semi quantitative and quantitative PCR demonstrated differential expression of several key retinoic acid receptors in six pancreatic cancer cell lines. These included, the global downregulation of RAR , RXR andupregulationofRAR and loss of expression of a key cellular retinoid transport molecule, CRBP1 in four of the pancreatic cancer cell lines, indicating aberrant RA signalling may be associated with the development and progression of PC. Selected effector molecules modulated by RA signalling were further investigated.
Retinoic Acid Induced 3 (RAI3) a retinoic acid responsive cell surface receptor of the GPCR superfamily was upregulated in 68% of PC using ISH, and was associated with improvedoutcome.Furthermore,RAI3isonlyexpressedatsignificantlevelsinadultand fetal lung, with minimal expression in other normal tissue including pancreas, making it a suitable target for the development of novel antibody mediated diagnostic and therapeutic modalities.
HOXB2 a putative downstream target of RA, associated with the normal development of the brainstem and mandibular region, was overexpressed in PC and was associated with a poor outcome. In addition, siRNA knockdown of HOXB2 in Panc-1 cells was associated with a cellular morphological change, however no difference in proliferation was identified. Although the functional role of HOXB2 is yet to be fully elucidated, the relationship between HOXB2, resection status, tumour size and lymph node involvement on multivariate analysis suggests that it may be a surrogate marker of advanced disease and hence confer a poor response for those patients undergoing resection. In addition, only those patients whose tumours did not express HOXB2 had a 133 survival benefit from operative resection. Hence assessment of HOXB2 expression has a potential role as a determinant of patient response to pancreatic resection.
The Identification of Markers of Early Diagnosis in PC.
A better understanding of precursor lesions of PC will lead to novel early detection and screening strategies. An ideal marker of early detection falls into 2 distinct categories. First, those targets that identify a pancreatic tumour mass including occult metastases. Second, those targets that identify precursor lesions of PC that have significant neoplastic potential. Data presented in this thesis have identified molecular markers of early diagnosis, which begin to address the requirements presented above.
In the presence of a mass lesion S100A2 expression was identified in 43% of tumour samples, S100A6 and S100P expression was identified in 60% and 47% of PC respectively. HOXB2 expression was observed in 38% of PC and was associated with a poor outcome.
An ideal in-situ diagnostic marker using current detection modalities would be significantly upregulated in PC and advanced PanIN lesions and be localised to the cell surface. Future development of novel detection modalities may make the molecules listed above more amenable to the development of detection strategies. However, these markers may currently find utility with diagnostic techniques that employ tissue sampling.
RAI3 provides perhaps the best utility as a marker of early diagnosis of PC. It is an orphan G protein coupled receptor localised in the cell membrane and observed in 68% of PC. In addition, RAI3 possesses specificity in relation to PC. RAI3 is normally expressed in fetal and adult lung tissue and not normally expressed in normal adult tissue including, the pancreas. In addition, the localisation of RAI3 to the cell membrane may allow for a diagnostic strategy using a radiolabelled antibody to RAI3 and requiring minor modification in radio-immunoscintographic techniques used for the diagnosis of other solid tumours. 134
These methods do not require the preoperative sampling of pancreatic tissue further increasing the utility of RAI3 as a marker of early diagnosis.
Whilst all the markers presented may provide for the early identification of PC, further confirmation using independent cohorts of PC is necessary and encouraged. In addition, particularly for those markers requiring pancreatic tissue for diagnosis, including S100 proteins and HOXB2 expression, a correlative case control study examining the relationship between preoperative assessment of expression with assessment of expression in the resected pancreas is required to further determine the specificity of these markers in PC. A protocol has already been designed incorporating intra-operative biopsy with rapid immunohistochemistry to prospectively correlate the expression of these markers preoperatively with their expression in the resected pancreatic specimen, and response to resection (Appendix 5).
Although RAI3 expression presently shows the most utility as an early diagnostic marker in PC, assessment of RAI3 protein expression, using antibody developed to the extracellular component of the RAI3 protein, in a large independent cohort of PC is still required to define the sensitivity and specificity of RAI3 expression in PC. If this confirms the findings presented in this thesis then further development of humanised antibodies is encouraged to further develop this target as a marker of early diagnosis.
Of all molecular markers examined in this thesis only HOXB2 expression was assessed in precursor lesions of PC. HOXB2 expression was detected in 1 of 24 (4%) PanIN-1A lesions, 3 of 20 (15%) Pan IN-1B, 3 of 10 (30%) PanIN 2 and 1 of 4 (25%) PanIN-3 lesions, suggesting HOXB2 expression is an early event in the development of PanIN.
Although data about the progression of PanIN to PC remains fragmentary, it is unknown what proportion of PanIN lesions progress to invasive carcinoma or if there is reversibility in the process. The literature suggests that PanIN-2 and PanIN-3 represent committed lesions with observations of progression from PanIN-2 to PC in 17 months to 10 years 258. 135
The Identification of Novel Markers of Prognosis and Response
Although resection remains the only therapeutic intervention that alters prognosis in this disease, selection criteria for resection are still based on anatomical issues relating to the safe resection of the tumour. The identification of patients with tumours whose molecular signature suggests a biologically benign phenotype is imperative, as these patients may have an improved survival with surgical resection.
Increased expression of S100A2 was associated with a poor outcome and high S100A6 and S100P expression was associated with an improved outcome. In patients who underwent pancreatectomy, S100A2 expression was identified as a superior predictor of survival than other, well-known clinicopathological parameters including tumour size, margin involvement and lymph node status. S100A6 and S100P expression have decreased utility given that they were not identified as independent factors of prognosis in PC, however, they may still provide a role in predicting outcome when included in a panel of molecular markers. The expression status of these and other genes can be incorporated to allow assessment of groups of genes for optimal determination of prognosis as has been done recently for breast cancer 259.Thisanalysismayidentifya group of genes that could be trailed prospectively as markers of response to operative resection of PC and allow better selection of patients for a procedure that carries significant morbidity and mortality. Further investigation is required in assessing S100A2, S100A6 and S100P expression status and outcome in a large, independent cohort of PC to confirm the observations presented in this thesis.
Increased HOXB2 expression was identified in 38% of PC; furthermore HOXB2 was the most influential independent prognostic factor in patients who underwent pancreatectomy, indicating that HOXB2 is a strong predictor of prognosis. In addition, the data presented suggest that HOXB2 is a candidate marker of therapeutic response in PC. HOXB2 overexpression is associated with surgical non-resectability, observed in 61.5% of unresectable tumours compared to 21% of resected tumours ( 2 p< 0.0001). 136
In addition, operative resection was only of benefit to those patients whose tumours did not express HOXB2 when compared to those patients whose PC expressed HOXB2 (logrank p < 0.0001). Survival for patients with tumours that were HOXB2 negative and who underwent resection was significantly longer than survival in all other groups 14 months versus 4.3 months (logrank p < 0.0001). This suggests that preoperative HOXB2 status may be used to identify those patients that would most benefit from surgical resection.
The advantage of HOXB2 expression as a prognostic indicator and an indicator of response to resection, is that it is potentially assessable preoperatively or during staging laparoscopy, as currently all known prognostic indicators such as tumour size, resection margins, and lymph node status can only be determined post resection.
Although HOXB2 shows promise as a marker of prognosis and response to resection, further confirmation is required in other independent cohorts of PC to clearly establish its association with outcome and response to resection. The cytoplasmic and nuclear subcellular localisation of HOXB2 requires pancreatic tissue to make an accurate assessment of HOXB2 expression preoperatively, further assessment in relating HOXB2 expression in preoperative biopsy specimen to HOXB2 expression in the resected pancreas is encouraged to confirm the clinical utility of this molecule as a marker of prognosis and response to therapy. 137
The Identification of Novel Therapeutic Strategies in PC
Retinoids influence many fundamental cellular processes, including embryogenesis, cell growth, differentiation, and apoptosis. These compounds also exert significant preventive and therapeutic effects against some human cancers.
Retinoic acid induced 3 (RAI3) is an orphan receptor of the GPCR family of receptors. This receptor was upregulated 24 fold in our samples of PC compared to normal pancreas on microarray analysis. It is also overexpressed in 68% of 118 PC samples using ISH. Recent data demonstrate that p53 binds to the promoter region of RAI3 causing repression of transcription. In most cell lines with mutant p53 RAI3 is upregulated, and conversely is present only in low levels when functional p53 is present. Furthermore, ectopic expression of RAI3 in 293 cells leads to anchorage independent growth, and siRNA-mediated depletion of RAI3 in AsPC3 PC cells induces morphological changes 249. In addition, there is almost undetectable expression in all other normal organs except the lung 250. Hence this cell surface receptor, which is overexpressed in a high proportion of PC with oncogenic characteristics, is an ideal target for the development of an antibody-mediated therapy. Polyclonal antibodies are currently being used to determine protein expression in a cohort of PC prior to the generation of humanized monoclonal antibodies for assessment as therapy in PC. 138
Concluding remarks and future directions
Although data from this thesis makes significant inroads in the understanding of molecular events associated with the development and progression of PC, further study is required before this knowledge can be applied to clinical practice.
Identification of novel genes of relevance, using transcript profiling technology, will facilitate such studies and the validation of data generated with this technique using the assembled cohort of PC and its precursor lesions PanIN and IPMT will focus future studies on genes not only of clinicopathological relevance to PC, but those that may be involved in early PC development.
The development of markers of disease outcome, response to therapy and early diagnosis remain hampered because of the technical difficulties of preoperatively assessing the pancreas. Until a sufficiently safe and reliable technique for imaging neoplastic lesions or sampling the intact pancreas is developed knowledge about the specificity and sensitivity of these targets can only be obtained through the study of surgically removed pancreata, most of which already contain a cancer. Although individually, these molecules may lack sufficient sensitivity or specificity to be assessed as a prognostic test in a clinical trial at this stage, the incorporation of a panel of molecules in novel diagnostic strategies may optimise overall sensitivity and sensitivity. It is hoped that once suitable candidates are identified that prospective trials utilizing minimally invasive intraoperative biopsy with rapid immunohistochemistry or cytopathology, may lead to better selection of patients for radical resection based on their predicted response determined by the expression status of a gene, or panel of genes. 139
The identification of aberrant Retinoic Acid signalling in PC, creates an avenue for the development of novel therapeutic strategies in PC, either through the modification of the retinoid signal using retinoids or using antibody based delivery systems to deliver therapeutic “payloads” to cells expressing the identified epitope.
These identified novel targets and pathways may form the basis of future experiments to determine their role in PC development and their potential utility as markers of early diagnosis, prognosis and suitability as targets for novel therapeutic and chemoprevention strategies. 140
REFERENCES
1. Tracey EA RD, Chen S, Chen W: Cancer in New South Wales: Incidence and Mortality 2003, (ed 1), Cancer Institute of New South Wales, 2005 2. Institute NC: Snapshot of Pancreatic Cancer, National Cancer Institute, 2005 3. Michaud DS: Epidemiology of pancreatic cancer. Minerva Chir 59:99-111, 2004 4. Falk RT, Pickle LW, Fontham ET, et al: Life-style risk factors for pancreatic cancer in Louisiana: A cas control study. Am J Epidemiol 128:324-336, 1988 5. Ghadirian P, Liu G, Gallinger S, et al: Risk of pancreatic cancer among individuals with a family history of cancer of the pancreas. International Journal of Cancer 97:807-10, 2002 6. Hruban RH, Yeo CJ, Kern SE: Pancreatic Cancer, in Vogelstein B, Kinzler KW (eds): The Genetic Basis of Human Cancer, McGraw-Hill, 1996, pp 603-613 7. Tersmette AC, Petersen GM, Offerhaus GJ, et al: Increased risk of incident pancreatic cancer among first-degree relatives of patients with familial pancreatic cancer. Clin Cancer Res 7:738-44, 2001 8. Yamaguchi K, Kinoshita H, Hokazono K, et al: Familial pancreatic cancer: report of one Japanese family. J Hepatobiliary Pancreat Surg 11:434-7, 2004 9. Lowenfels AB, Maisonneuve P, DiMagno EP, et al: Hereditary pancreatitis and the risk of pancreatic cancer. International Hereditary Pancreatitis Study Group. J Natl Cancer Inst 89:442-6, 1997 10. Bartsch DK: Familial pancreatic cancer. Br J Surg 90:386-7, 2003 11. Yamamoto M, Ohashi O, Saitoh Y: Japan Pancreatic Cancer Registry: current status. Pancreas 16:238-42, 1998 12. Fuchs CS, Colditz GA, Stampfer MJ, et al: A prospective study of cigarette smoking and the risk of pancreatic cancer. Arch Intern Med 156:2255-60, 1996 13. Fraumeni JF: Cancers of the pancreas and biliary tract: epidemiological considerations. Cancer Res 35:3437-3446, 1975 14. Doll R, Peto R, Hall E, et al: Mortality in relation to consumption of alcohol: 13 years' observations on male British doctors. Bmj 309:911-8, 1994 141
15. Silverman DT, Dunn JA, Hoover RN, et al: Cigarette smoking and pancreas cancer: a case-control study based on direct interviews. J Natl Cancer Inst 86:1510-6, 1994 16. Silverman DT: Risk factors for pancreatic cancer: a case-control study based on direct interviews. Teratog Carcinog Mutagen 21:7-25, 2001 17. Qiu D, Kurosawa M, Lin Y, et al: Overview of the epidemiology of pancreatic cancer focusing on the JACC Study. J Epidemiol 15 Suppl 2:S157-67, 2005 18. Hruban RH, Van Mansfield ADM, Offerhaus JGA, et al: K-ras oncogene activation in adenocarcinoma of the human pancreas: A study of 82 carcinomas using a combination of mutant-enriched polymerase chain reaction and allele- specific oligonucleotide hybridization. Am J Pathol 143, 1993 19. Topal MD: DNA repair, oncogenes and carcinogenesis. Carcinogenesis 9:691-6, 1988 20. Bansal P, Sonnenberg A: Pancreatitis is a risk factor for pancreatic cancer [see comments]. Gastroenterology 109:247-51, 1995 21. Lowenfels AB, Maisonneuve P, Cavallini G, et al: Pancreatitis and the risk of pancreatic cancer. International Pancreatitis Study Group. N Engl J Med 328:1433-7, 1993 22. Whitcomb DC, Applebaum S, Martin SP: Hereditary pancreatitis and pancreatic carcinoma. Ann N Y Acad Sci 880:201-9, 1999 23. Whitcomb DC, Gorry MC, Preston RA, et al: Hereditary pancreatitis is caused by a mutation in the cationic trypsinogen gene. Nat Genet 14:141-5, 1996 24. Ekbom A, McLaughlin JK, Nyren O, et al: Pancreatitis and pancreatic cancer:a population based study. J Natl Cancer Inst 86:625-627, 1994 25. Karlson BM, Ekbom A, Josefsson S, et al: The risk of pancreatic cancer following pancreatitis: an association due to confounding? Gastroenterology 113:587-92, 1997 26. Everhart J, Wright D: Diabetes mellitus as a risk factor for pancreatic cancer. A meta-analysis. Jama 273:1605-9, 1995 27. Gapstur SM, Gann PH, Lowe W, et al: Abnormal glucose metabolism and pancreatic cancer mortality. Jama 283:2552-8, 2000 142
28. Chari ST, Leibson CL, Rabe KG, et al: Probability of pancreatic cancer following diabetes: a population-based study. Gastroenterology 129:504-11, 2005 29. Sommers SC, Murphy SA, Warren S: Pancreatic Duct Hyperplasia and Cancer. Gastroenterology 27:629-640, 1954 30. Cubilla AL, Fitzgerald PJ: Morphological lesions associated with human primary invasive nonendocrine pancreatic cancer. Cancer Res 36:2690-2698, 1976 31. Kozuka S, Sassa R, Taki T, et al: Relation of pancreatic duct hyperplasia to carcinoma. Cancer 43:1418-1428, 1979 32. Gold EB: Epidemiology of and risk factors for pancreatic cancer. Surg Clin North Am 75:819-43, 1995 33. Frye JN, Inder WJ, Dobbs BR, et al: Pancreatic cancer and diabetes: is there a relationship? A case-controlled study. Aust N Z J Surg 70:722-4, 2000 34. Burney PG, Comstock GW, Morris JS: Serological precursors of cancer serum micronutrients and the subsequent risk of pancreatic cancer. Am J Clin Nutr 49:895-900, 1989 35. Stoltzenberg-Solomon RZ, Albanez D, Nieto FJ, et al: Pancreatic cancer risk and nutrition-related methyl-group availability indicators in male smokers. J Natl Cancer Inst 91:535-541, 1999 36. Ghadirian P, Baillargeon J, Simard A, et al: Food habits and pancreatic cancer: acase control study of the Francophone community in Montreal, Canada. Cancer Epidemiol Biomarkers Prev 4:895-899, 1995 37. Michaud DS, Giovannucci E, Willett WC, et al: Coffee and alcohol consumption and the risk of pancreatic cancer in two prospective United States cohorts. Cancer Epidemiol Biomarkers Prev 10:429-37, 2001 38. Lin Y, Tamakoshi A, Kawamura T, et al: Risk of pancreatic cancer in relation to alcohol drinking, coffee consumption and medical history: findings from the Japan collaborative cohort study for evaluation of cancer risk. Int J Cancer 99:742-6, 2002 39. Ahlgren JD: Epidemiology and risk factors in pancreatic cancer. Semin Oncol 23:241-250, 1996 143
40. Gold EB, Goldin SB: Epidemiology and risk factors for pancreatic cancer. Surg Clin North Am 7:67-91, 1998 41. Li D: Molecular Epidemiology, in Evans DB, Pisters PW, Abbruzzese JL (eds): Pandreatic Cancer. M. D. Andersen solid tulor oncology series. New York, Springer, 2002, pp 3-14 42. Hecht SS: Approaches to cancer prevention based on understanding of N- nitrosoamine carcinogenesis. proc Soc Exp Biol Med 94:1253-1260, 1997 43. Morohoshi T, Held G, Kloppel G: Exocrine pancreatic tumors and their histologic classification. A study based on 167 autopsies and 97 surgical cases. Histopathology:645-661, 1983 44. Miyamoto Y, Maitra A, Zechner U, et al: Notch signaling regulates a metaplasia/neoplasia sequence in mouse and human pancreas. Nature Medicine Submitted, 2002 45. Stanger BZ, Stiles B, Lauwers GY, et al: Pten constrains centroacinar cell expansion and malignant transformation in the pancreas. Cancer Cell 8:185-95, 2005 46. Meszoely IM, Means AL, Scoggins CR, et al: Developmental aspects of early pancreatic cancer. Cancer Journal 7:242-50, 2001 47. Blackstock AW, Bernard SA, Richards F, et al: Phase I trial of twice-weekly gemcitabine and concurrent radiation in patients with advanced pancreatic cancer. J Clin Oncol 17:2208-12, 1999 48. McGinn CJ, Zalupski MM: Combined-modality therapy in pancreatic cancer: current status and future directions. Cancer J 7:338-48, 2001 49. Crane CH, Janjan NA, Evans DB, et al: Toxicity and efficacy of concurrent gemcitabine and radiotherapy for locally advanced pancreatic cancer. Int J Pancreatol 29:9-18, 2001 50. Wolff RA, Evans DB, Gravel DM, et al: Phase I trial of gemcitabine combined with radiation for the treatment of locally advanced pancreatic adenocarcinoma. Clin Cancer Res 7:2246-53, 2001 144
51. Berlin JD, Catalano P, Thomas JP, et al: Phase III study of gemcitabine in combination with fluorouracil versus gemcitabine alone in patients with advanced pancreatic carcinoma: Eastern Cooperative Oncology Group Trial E2297. J Clin Oncol 20:3270-5, 2002 52. Kalser MH, Ellenberg SS: Pancreatic cancer. Adjuvant combined radiation and chemotherapy following curative resection. Arch Surg 120:899-903, 1985 53. Further evidence of effective adjuvant combined radiation and chemotherapy following curative resection of pancreatic cancer. Gastrointestinal Tumor Study Group. Cancer 59:2006-10, 1987 54. Treatment of locally unresectable carcinoma of the pancreas: comparison of combined-modality therapy (chemotherapy plus radiotherapy) to chemotherapy alone. Gastrointestinal Tumor Study Group. J Natl Cancer Inst 80:751-5, 1988 55. Neoptolemos JP, Stocken DD, Friess H, et al: A randomized trial of chemoradiotherapy and chemotherapy after resection of pancreatic cancer. N Engl J Med 350:1200-10, 2004 56. Spitz FR, Abbruzzese JL, Lee JE, et al: Preoperative and postoperative chemoradiation strategies in patients treated with pancreaticoduodenectomy for adenocarcinoma of the pancreas. J Clin Oncol 15:928-37, 1997 57. Hoffman JP, Weese JL, Solin LJ, et al: A pilot study of preoperative chemoradiation for patients with localized adenocarcinoma of the pancreas. Am J Surg 169:71-7; discussion 77-8, 1995 58. Evans DB, Rich TA, Byrd DR, et al: Preoperative chemoradiation and pancreaticoduodenectomy for adenocarcinoma of the pancreas. Arch Surg 127:1335-9, 1992 59. Hoffman JP, Lipsitz S, Pisansky T, et al: Phase II trial of preoperative radiation therapy and chemotherapy for patients with localized, resectable adenocarcinoma of the pancreas: an Eastern Cooperative Oncology Group Study. J Clin Oncol 16:317-23, 1998 60. Gjertsen MK, Buanes T, Rosseland AR, et al: Intradermal ras peptide vaccination with granulocyte-macrophage colony-stimulating factor as adjuvant: Clinical and immunological responses in patients with pancreatic adenocarcinoma. Int J Cancer 92:441-50, 2001 145
61. Jaffee EM, Hruban RH, Biedrzycki B, et al: Novel allogeneic granulocyte- macrophage colony-stimulating factor-secreting tumor vaccine for pancreatic cancer: a phase I trial of safety and immune activation. Journal of Clinical Oncology 19:145-56, 2001 62. Geer RJ, Brennan MF: Prognostic indicators for survival after resection of pancreatic adenocarcinoma. Am J Surg 165:68-73, 1993 63. Nitecki SS, Sarr MG, Colby TV, et al: Long-term survival after resection for ductal adenocarcinoma of the pancreas. Is it really improving? Ann Surg 221:59-66, 1995 64. Fortner JG, Klimstra DS, Senie RT, et al: Tumor size is the primary prognosticator for pancreatic cancer after regional pancreatectomy. Ann. Surg. 223:147-53, 1996 65. Sohn TA, Yeo CJ, Cameron JL, et al: Resected adenocarcinoma of the pancreas- 616 patients: results, outcomes, and prognostic indicators. J Gastrointest Surg 4:567-79, 2000 66. Kawesha A, Ghaneh P, Andren-Sandberg A, et al: K-ras oncogene subtype mutations are associated with survival but not expression of p53, p16(INK4A), p21(WAF-1), cyclin D1, erbB-2 and erbB-3 in resected pancreatic ductal adenocarcinoma. Int. J. Cancer 89:469-74, 2000 67. Biankin AV, Morey AL, Lee C-S, et al: DPC4/Smad4 expression and outcome in pancreatic ductal adenocarcinoma. J Clin Oncol In Press, 2002 68. Safi F, Roscher R, Bittner R, et al: High sensitivity and specificity of CA 19-9 for pancreatic carcinoma in comparison to chronic pancreatitis. Serological and immunohistochemical findings. Pancreas 2:398-403, 1987 69. Glenn J, Steinberg WM, Kurtzman SH, et al: Evaluation of the utility of a radioimmunoassay for serum CA 19-9 levels in patients before and after treatment of carcinoma of the pancreas. J Clin Oncol 6:462-8, 1988 70. Halm U, Schumann T, Schiefke I, et al: Decrease of CA 19-9 during chemotherapy with gemcitabine predicts survival time in patients with advanced pancreatic cancer. Br J Cancer 82:1013-6, 2000 146
71. Yamamoto H, Itoh F, Iku S, et al: Expression of matrix metalloproteinases and tissue inhibitors of metalloproteinases in human pancreatic adenocarcinomas: clinicopathologic and prognostic significance of matrilysin expression. J Clin Oncol 19:1118-27, 2001 72. Jones LE, Humphreys MJ, Campbell F, et al: Comprehensive analysis of matrix metalloproteinase and tissue inhibitor expression in pancreatic cancer: increased expression of matrix metalloproteinase-7 predicts poor survival. Clin Cancer Res 10:2832-45, 2004 73. Vimalachandran D, Greenhalf W, Thompson C, et al: High nuclear S100A6 (Calcyclin) is significantly associated with poor survival in pancreatic cancer patients. Cancer Res 65:3218-25, 2005 74. Li D, Xie K, Wolff R, et al: Pancreatic cancer. Lancet 363:1049-57, 2004 75. Biankin AV, Kench JG, Morey AL, et al: Overexpression of p21WAF1/CIP1 is an early event in the development of pancreatic intraepithelial neoplasia. Cancer Res 61:8830-8837, 2001 76. Skalicky DA, Kench J, Segara D, et al: Cyclin E Expression and Outcome in Pancreatic Ductal Adenocarcinoma. submitted for publication 77. DiGiuseppe JA, Redston MS, Yeo CJ, et al: p53-independent expression of the cyclin-dependent kinase inhibitor p21 in pancreatic carcinoma. American Journal of Pathology 147:884-888, 1995 78. Izumi Y, Xu L, di Tomaso E, et al: Tumour biology: herceptin acts as an anti- angiogenic cocktail. Nature 416:279-80, 2002 79. Biankin AV, Biankin SA, Kench J, et al: Aberrant 16INK4A, p53 and Smad4 expression in intraductal papillary mucinous tumours of the pancreas predicts risk of invasive ductal adenocarcinoma. Gut 50:861-868, 2002 80. Hruban RH, Iacobuzio-Donahue C, Wilentz RE, et al: Molecular pathology of pancreatic cancer. Cancer J 7:251-8, 2001 81. Moore PS, Beghelli S, Zamboni G, et al: Genetic abnormalities in pancreatic cancer. Mol Cancer 2:7, 2003 147
82. Korc M, Chandrasekar B, Yamanaka Y, et al: Overexpression of the epidermal growth factor receptor in human pancreatic cancer is associated with concomitant increases in the levels of epidermal growth factor and transforming growth factor alpha. J Clin Invest 90:1352-1360, 1992 83. Sirivatanauksorn V, Sirivatanauksorn Y, Lemoine NR: Molecular pattern of ductal pancreatic cancer. Langenbecks Archives of Surgery 383:105-15, 1998 84. Bockaert J, Pin JP: Molecular tinkering of G protein-coupled receptors: an evolutionary success. Embo J 18:1723-9, 1999 85. Friess H, Berberat PO, Schilling M, et al: Pancreatic cancer: the potential clinical relevance of alterations in growth factors and their receptors. J Mol Med 74:35-42, 1996 86. Rozenblum E, Schutte M, Goggins M, et al: Tumor-suppressive pathways in pancreatic carcinoma. Cancer Res 57:1731-1734, 1997 87. Eckert RL, Broome AM, Ruse M, et al: S100 proteins in the epidermis. J Invest Dermatol 123:23-33, 2004 88. Donato R: S100: a multigenic family of calcium-modulated proteins of the EF- hand type with intracellular and extracellular functional roles. Int J Biochem Cell Biol 33:637-68, 2001 89. Kriajevska M, Tarabykina S, Bronstein I, et al: Metastasis-associated Mts1 (S100A4) protein modulates protein kinase C phosphorylation of the heavy chain of nonmuscle myosin. J Biol Chem 273:9852-6, 1998 90. Heierhorst J, Kobe B, Feil SC, et al: Ca2+/S100 regulation of giant protein kinases. Nature 380:636-9, 1996 91. Kerkhoff C, Klempt M, Sorg C: Novel insights into structure and function of MRP8 (S100A8) and MRP14 (S100A9). Biochim Biophys Acta 1448:200-11, 1998 92. Sorci G, Agneletti AL, Bianchi R, et al: Association of S100B with intermediate filaments and microtubules in glial cells. Biochim Biophys Acta 1448:277-89, 1998 93. Onions J, Hermann S, Grundstrom T: Basic helix-loop-helix protein sequences determining differential inhibition by calmodulin and S-100 proteins. J Biol Chem 272:23930-7, 1997 148
94. Iwasaki Y, Shiojima T, Kinoshita M: S100 beta prevents the death of motor neurons in newborn rats after sciatic nerve section. J Neurol Sci 151:7-12, 1997 95. Xu K, Geczy CL: IFN-gamma and TNF regulate macrophage expression of the chemotactic S100 protein S100A8. J Immunol 164:4916-23, 2000 96. Hsieh HL, Schafer BW, Sasaki N, et al: Expression analysis of S100 proteins and RAGE in human tumors using tissue microarrays. Biochem Biophys Res Commun 307:375-81, 2003 97. Cochran AJ, Lu HF, Li PX, et al: S-100 protein remains a practical marker for melanocytic and other tumours. Melanoma Res 3:325-30, 1993 98. Miyamoto Y, Maitra A, Ghosh B, et al: Notch mediates TGF alpha-induced changes in epithelial differentiation during pancreatic tumorigenesis. Cancer Cell 3:565-576, 2003 99. Thayer SP, Di Magliano MP, Heiser PW, et al: Hedgehog is an early and late mediator of pancreatic cancer tumorigenesis. Nature 425:851-6, 2003 100. Kawaguchi Y, Cooper B, Gannon M, et al: The role of the transcriptional regulator Ptf1a in converting intestinal to pancreatic progenitors. Nat Genet 32:128-34, 2002 101. Kim SK, MacDonald RJ: Signaling and transcriptional control of pancreatic organogenesis. Curr Opin Genet Dev 12:540-7, 2002 102. Edlund H: Transcribing pancreas. Diabetes 47:1817-23, 1998 103. Kim SK, Hebrok M, Melton DA: Notochord to endoderm signaling is required for pancreas development. Development 124:4243-52, 1997 104. Apelqvist A, Ahlgren U, Edlund H: Sonic hedgehog directs specialised mesoderm differentiation in the intestine and pancreas. Curr Biol 7:801-4, 1997 105. Pettersson F, Dalgleish AG, Bissonnette RP, et al: Retinoids cause apoptosis in pancreatic cancer cells via activation of RAR-gamma and altered expression of Bcl-2/Bax. Br J Cancer 87:555-61, 2002 106. Oosterveen T, Niederreither K, Dolle P, et al: Retinoids regulate the anterior expression boundaries of 5' Hoxb genes in posterior hindbrain. Embo J 22:262- 9, 2003 107. Petkovich M, Brand NJ, Krust A, et al: A human retinoic acid receptor which belongs to the family of nuclear receptors. Nature 330:444-50, 1987 149
108. Miyamoto Y, Maitra A, Ghosh B, et al: Notch mediates TGF alpha-induced changes in epithelial differentiation during pancreatic tumorigenesis. Cancer Cell 3:565-76, 2003 109. Tulachan SS, Doi R, Kawaguchi Y, et al: All-trans retinoic acid induces differentiation of ducts and endocrine cells by mesenchymal/epithelial interactions in embryonic pancreas. Diabetes 52:76-84, 2003 110. Kobayashi H, Spilde TL, Bhatia AM, et al: Retinoid signaling controls mouse pancreatic exocrine lineage selection through epithelial-mesenchymal interactions. Gastroenterology 123:1331-40, 2002 111. Giguere V: Retinoic acid receptors and cellular retinoid binding proteins: complex interplay in retinoid signaling. Endocr Rev 15:61-79, 1994 112. Kaiser A, Wolf-Breitinger M, Albers A, et al: Retinoic acid receptor gamma1 expression determines retinoid sensitivity in pancreatic carcinoma cells. Gastroenterology 115:967-77, 1998 113. Balague C, Audie JP, Porchet N, et al: In situ hybridization shows distinct patterns of mucin gene expression in normal, benign, and malignant pancreas tissues. Gastroenterology 109:953-64, 1995 114. Swartz MJ, Batra SK, Varshney GC, et al: MUC4 expression increases progressively in pancreatic intraepithelial neoplasia. Am J Clin Pathol 117:791- 6, 2002 115. Choudhury A, Singh RK, Moniaux N, et al: Retinoic acid-dependent transforming growth factor-beta 2-mediated induction of MUC4 mucin expression in human pancreatic tumor cells follows retinoic acid receptor-alpha signaling pathway. J Biol Chem 275:33929-36, 2000 116. Gong YL, Xu GM, Huang WD, et al: Expression of matrix metalloproteinases and the tissue inhibitors of metalloproteinases and their local invasiveness and metastasis in Chinese human pancreatic cancer. Journal of Surgical Oncology 73:95-9, 2000 117. Jaster R, Brock P, Sparmann G, et al: Inhibition of pancreatic stellate cell activation by the hydroxymethylglutaryl coenzyme A reductase inhibitor lovastatin. Biochem Pharmacol 65:1295-303, 2003 150
118. Liu TX, Zhang JW, Tao J, et al: Gene expression networks underlying retinoic acid-induced differentiation of acute promyelocytic leukemia cells. Blood 96:1496-504, 2000 119. Maruyama H, Kleeff J, Wildi S, et al: Id-1 and Id-2 are overexpressed in pancreatic cancer and in dysplastic lesions in chronic pancreatitis. Am. J Pathol. 155:815-822, 1999 120. Ma Y, Koza-Taylor PH, DiMattia DA, et al: Microarray analysis uncovers retinoid targets in human bronchial epithelial cells. Oncogene 22:4924-32, 2003 121. Oosterveen T, van Vliet P, Deschamps J, et al: The direct context of a hox retinoic acid response element is crucial for its activity. J Biol Chem 278:24103- 7, 2003 122. Duboule D, Morata G: Colinearity and functional hierarchy among genes of the homeotic complexes. Trends Genet 10:358-64, 1994 123. Abate-Shen C: Deregulated homeobox gene expression in cancer: cause or consequence? Nat Rev Cancer 2:777-85, 2002 124. Lewis EB: A gene complex controlling segmentation in Drosophila. Nature 276:565-70, 1978 125. Apiou F, Flagiello D, Cillo C, et al: Fine mapping of human HOX gene clusters. Cytogenet Cell Genet 73:114-5, 1996 126. Graham A, Papalopulu N, Krumlauf R: The murine and Drosophila homeobox gene complexes have common features of organization and expression. Cell 57:367-78, 1989 127. Celetti A, Barba P, Cillo C, et al: Characteristic patterns of HOX gene expression in different types of human leukemia. Int J Cancer 53:237-44, 1993 128. Nakamura T, Largaespada DA, Lee MP, et al: Fusion of the nucleoporin gene NUP98 to HOXA9 by the chromosome translocation t(7;11)(p15;p15) in human myeloid leukaemia. Nat Genet 12:154-8, 1996 129. Nakamura T, Largaespada DA, Shaughnessy JD, Jr., et al: Cooperative activation of Hoxa and Pbx1-related genes in murine myeloid leukaemias. Nat Genet 12:149-53, 1996 151
130. Iida S, Rao PH, Ueda R, et al: Chromosomal rearrangement of the PAX-5 locus in lymphoplasmacytic lymphoma with t(9;14)(p13;q32). Leuk Lymphoma 34:25-33, 1999 131. Cillo C, Barba P, Freschi G, et al: HOX gene expression in normal and neoplastic human kidney. Int J Cancer 51:892-7, 1992 132. De Vita G, Barba P, Odartchenko N, et al: Expression of homeobox-containing genes in primary and metastatic colorectal cancer. Eur J Cancer 29A:887-93, 1993 133. Tiberio C, Barba P, Magli MC, et al: HOX gene expression in human small-cell lung cancers xenografted into nude mice. Int J Cancer 58:608-15, 1994 134. Naora H, Yang YQ, Montz FJ, et al: A serologically identified tumor antigen encoded by a homeobox gene promotes growth of ovarian epithelial cells. Proc Natl Acad Sci U S A 98:4060-5, 2001 135. Hung YC, Ueda M, Terai Y, et al: Homeobox gene expression and mutation in cervical carcinoma cells. Cancer Sci 94:437-41, 2003 136. Lewis MT: Homeobox genes in mammary gland development and neoplasia. Breast Cancer Res 2:158-69, 2000 137. Cantile M, Pettinato G, Procino A, et al: In vivo expression of the whole HOX gene network in human breast cancer. Eur J Cancer 39:257-64, 2003 138. Boudreau N, Andrews C, Srebrow A, et al: Induction of the angiogenic phenotype by Hox D3. J Cell Biol 139:257-64, 1997 139. Care A, Felicetti F, Meccia E, et al: HOXB7: a key factor for tumor-associated angiogenic switch. Cancer Res 61:6532-9, 2001 140. Care A, Silvani A, Meccia E, et al: HOXB7 constitutively activates basic fibroblast growth factor in melanomas. Mol Cell Biol 16:4842-51, 1996 141. Raman V, Martensen SA, Reisman D, et al: Compromised HOXA5 function can limit p53 expression in human breast tumours. Nature 405:974-8, 2000 142. Raman V, Tamori A, Vali M, et al: HOXA5 regulates expression of the progesterone receptor. J Biol Chem 275:26551-5, 2000 143. Cillo C, Cantile M, Faiella A, et al: Homeobox genes in normal and malignant cells. J Cell Physiol 188:161-9, 2001 152
144. Chariot A, Moreau L, Senterre G, et al: Retinoic acid induces three newly cloned HOXA1 transcripts in MCF7 breast cancer cells. Biochem Biophys Res Commun 215:713-20, 1995 145. Yan YL, Jowett T, Postlethwait JH: Ectopic expression of hoxb2 after retinoic acid treatment or mRNA injection: disruption of hindbrain and craniofacial morphogenesis in zebrafish embryos. Dev Dyn 213:370-85, 1998 146. Nakamura N: Untersuchungen uber das Pankreas bei Foten, Heugeborenen, Kindern und im Pubertatsalter. Virchows Arch. f. path. Anat. 253:286-349, 1924 147. Brat DJ, Lillemoe KD, Yeo CJ, et al: Progression of pancreatic intraductal neoplasias to infiltrating adenocarcinoma of the pancreas. Am J Surg Pathol 22:163-169, 1998 148. Chang KW, Laconi S, Mangold KA, et al: Multiple genetic alterations in hamster pancreatic ductal adenocarcinomas. Cancer Res 55:2560-8, 1995 149. Hruban RH, Takaori K, Klimstra DS, et al: An illustrated consensus on the classification of pancreatic intraepithelial neoplasia and intraductal papillary mucinous neoplasms. Am J Surg Pathol 28:977-87, 2004 150. Chinn P, Braslawsky G, White C, et al: Antibody therapy of non-Hodgkin's B- cell lymphoma. Cancer Immunol Immunother 52:257-80, 2003 151. Slamon DJ, Leyland-Jones B, Shak S, et al: Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med 344:783-92., 2001 152. Kitamura Y, Hirota S, Nishida T: Gastrointestinal stromal tumors (GIST): A model for molecule-based diagnosis and treatment of solid tumors. Cancer Sci 94:315-20, 2003 153. Baugh LR, Hill AA, Brown EL, et al: Quantitative analysis of mRNA amplification by in vitro transcription. Nucleic Acids Research 29:e29, 2001 154. Yeo CJ, Cameron JL, Sohn TA, et al: Six hundred fifty consecutive pancreaticoduodenectomies in the 1990s: pathology, complications, and outcomes. Ann Surg 226:248-57; discussion 257-60, 1997 155. Ryu B, Jones J, Blades NJ, et al: Relationships and differentially expressed genes among pancreatic cancers examined by large-scale serial analysis of gene expression. Cancer Res 62:819-26, 2002 153
156. Argani P, Rosty C, Reiter RE, et al: Discovery of new markers of cancer through serial analysis of gene expression: prostate stem cell antigen is overexpressed in pancreatic adenocarcinoma. Cancer Res 61:4320-4, 2001 157. Argani P, Iacobuzio-Donahue C, Ryu B, et al: Mesothelin is overexpressed in the vast majority of ductal adenocarcinomas of the pancreas: identification of a new pancreatic cancer marker by serial analysis of gene expression (SAGE). Clinical Cancer Research 7:3862-8, 2001 158. Han H, Bearss DJ, Browne LW, et al: Identification of differentially expressed genes in pancreatic cancer cells using cDNA microarray. Cancer Res 62:2890-6, 2002 159. Crnogorac-Jurcevic T, Efthimiou E, Capelli P, et al: Gene expression profiles of pancreatic cancer and stromal desmoplasia. Oncogene 20:7437-46, 2001 160. Crnogorac-Jurcevic T, Efthimiou E, Nielsen T, et al: Expression profiling of microdissected pancreatic adenocarcinoma. Oncogene 21:4587-4594, 2002 161. Iacobuzio-Donahue CA, Maitra A, Shen-Ong GL, et al: Discovery of novel tumor markers of pancreatic cancer using global gene expression technology. Am J Pathol 160:1239-49., 2002 162. Maitra A, Iacobuzio-Donahue C, Rahman A, et al: Immunohistochemical validation of a novel epithelial and a novel stromal marker of pancreatic ductal adenocarcinoma identified by global expression microarrays: sea urchin fascin homolog and heat shock protein 47. Am J Clin Pathol 118:52-9, 2002 163. Piek E, Heldin CH, Ten Dijke P: Specificity, diversity and regulation in TGF- beta superfamily signaling. FASEB J 13:2105-2124, 1999 164. Grau AM, Zhang L, Wang W, et al: Induction of p21waf1 expression and growth inhibition by transforming growth factor beta involve the tumor suppressor gene DPC4 in human pancreatic adenocarcinoma cells. Cancer Res 57:3929-3934, 1997 165. Hunt KK, Fleming JB, Abramian A, et al: Overexpression of the tumour suppressor gene Smad4/DPC4 induces p21 expression and growth inhibition in human carcinoma cells. Cancer Res 58:5656-5661, 1998 154
166. Villanueva A, Garcia C, Paules AB, et al: Disruption of the antiproliferative TGF-b signaling pathways in human pancreatic cancer cells. Oncogene 17:1969- 1978, 1998 167. Comijn J, Berx G, Vermassen P, et al: The two-handed E box binding zinc finger protein SIP1 downregulates E-cadherin and induces invasion. Mol Cell 7:1267-78, 2001 168. Miyoshi A, Kitajima Y, Sumi K, et al: Snail and SIP1 increase cancer invasion by upregulating MMP family in hepatocellular carcinoma cells. Br J Cancer 90:1265-73, 2004 169. Elloul S, Elstrand MB, Nesland JM, et al: Snail, Slug, and Smad-interacting protein 1 as novel parameters of disease aggressiveness in metastatic ovarian and breast carcinoma. Cancer 103:1631-43, 2005 170. Yip-Schneider MT, Lin A, Marshall MS: Pancreatic tumor cells with mutant K- ras suppress ERK activity by MEK-dependent induction of MAP kinase phosphatase-2. Biochem Biophys Res Commun 280:992-7, 2001 171. Boucher MJ, Morisset J, Vachon PH, et al: MEK/ERK signaling pathway regulates the expression of Bcl-2, Bcl-X(L), and Mcl-1 and promotes survival of human pancreatic cancer cells. J Cell Biochem 79:355-69, 2000 172. Kleeff J, Maruyama H, Ishiwata T, et al: Bone morphogenetic protein 2 exerts diverse effects on cell growth in vitro and is expressed in human pancreatic cancer in vivo. Gastroenterology 116:1202-16, 1999 173. Blyth K, Terry A, Mackay N, et al: Runx2: a novel oncogenic effector revealed by in vivo complementation and retroviral tagging. Oncogene 20:295-302, 2001 174. Pratap J, Javed A, Languino LR, et al: The Runx2 osteogenic transcription factor regulates matrix metalloproteinase 9 in bone metastatic cancer cells and controls cell invasion. Mol Cell Biol 25:8581-91, 2005 175. Thomas GJ, Hart IR, Speight PM, et al: Binding of TGF-beta1 latency- associated peptide (LAP) to alpha(v)beta6 integrin modulates behaviour of squamous carcinoma cells. Br J Cancer 87:859-67, 2002 176. Kawashima A, Tsugawa S, Boku A, et al: Expression of alphav integrin family in gastric carcinomas: increased alphavbeta6 is associated with lymph node metastasis. Pathol Res Pract 199:57-64, 2003 155
177. Andrianifahanana M, Agrawal A, Singh AP, et al: Synergistic induction of the MUC4 mucin gene by interferon-gamma and retinoic acid in human pancreatic tumour cells involves a reprogramming of signalling pathways. Oncogene 24:6143-54, 2005 178. Polakis P: Wnt signaling and cancer. Genes Dev 14:1837-51., 2000 179. Pasca di Magliano M BA, Heiser PW, Cano DA, Gutierrez PJA, Deramaudt T, Segara D, Dawson AC, Kench JG, Henshall SM, Sutherland RL, Dlugosz A, Rustgi AK and Hebrok M.: Wnt Signalling in Pancreatic Cancer. submitted for publication, 2005 180. Crnogorac-Jurcevic T, Missiaglia E, Blaveri E, et al: Molecular alterations in pancreatic carcinoma: expression profiling shows that dysregulated expression of S100 genes is highly prevalent. J Pathol 201:63-74, 2003 181. Cantero D, Friess H, Deflorin J, et al: Enhanced expression of urokinase plasminogen activator and its receptor in pancreatic carcinoma. Br J Cancer 75:388-95, 1997 182. Ito Y, Higashiyama S, Takeda T, et al: Expression of heparin-binding epidermal growth factor-like growth factor in pancreatic adenocarcinoma. Int J Pancreatol 29:47-52, 2001 183. Lammert E, Brown J, Melton DA: Notch gene expression during pancreatic organogenesis. Mech Dev 94:199-203., 2000 184. Kadison A, Kim J, Maldonado T, et al: Retinoid signaling directs secondary lineage selection in pancreatic organogenesis. Journal of Pediatric Surgery 36:1150-6, 2001 185. Duprez E, Wagner K, Koch H, et al: C/EBPbeta: a major PML-RARA- responsive gene in retinoic acid-induced differentiation of APL cells. Embo J 22:5806-16, 2003 186. Gavalas A: ArRAnging the hindbrain. TRENDS in Neurosciences 25:62-64, 2002 187. Ivins S, Pemberton K, Guidez F, et al: Regulation of Hoxb2 by APL-associated PLZF protein. Oncogene 22:3685-97, 2003 188. Nakayama S, Kretsinger RH: Evolution of the EF-hand family of proteins. Annu Rev Biophys Biomol Struct 23:473-507, 1994 156
189. Rosty C, Ueki T, Argani P, et al: Overexpression of S100A4 in pancreatic ductal adenocarcinomas is associated with poor differentiation and DNA hypomethylation. Am J Pathol 160:45-50, 2002 190. Sato N, Maitra A, Fukushima N, et al: Frequent hypomethylation of multiple genes overexpressed in pancreatic ductal adenocarcinoma. Cancer Res 63:4158- 66, 2003 191. Logsdon CD, Simeone DM, Binkley C, et al: Molecular profiling of pancreatic adenocarcinoma and chronic pancreatitis identifies multiple genes differentially regulated in pancreatic cancer. Cancer Res 63:2649-57, 2003 192. Iacobuzio-Donahue CA, Maitra A, Olsen M, et al: Exploration of global gene expression patterns in pancreatic adenocarcinoma using cDNA microarrays. Am J Pathol 162:1151-62, 2003 193. Terris B, Blaveri E, Crnogorac-Jurcevic T, et al: Characterization of gene expression profiles in intraductal papillary-mucinous tumors of the pancreas. Am J Pathol 160:1745-54., 2002 194. Liu D, Rudland PS, Sibson DR, et al: Expression of calcium-binding protein S100A2 in breast lesions. Br J Cancer 83:1473-9, 2000 195. Mueller A, Schafer BW, Ferrari S, et al: The calcium-binding protein S100A2 interacts with p53 and modulates its transcriptional activity. J Biol Chem 280:29186-93, 2005 196. Imazawa M, Hibi K, Fujitake S, et al: S100A2 overexpression is frequently observed in esophageal squamous cell carcinoma. Anticancer Res 25:1247-50, 2005 197. Wang H, Zhang Z, Li R, et al: Overexpression of S100A2 protein as a prognostic marker for patients with stage I non small cell lung cancer. Int J Cancer 116:285-90, 2005 198. Ito Y, Yoshida H, Tomoda C, et al: Expression of S100A2 and S100A6 in thyroid carcinomas. Histopathology 46:569-75, 2005 199. El-Rifai W, Moskaluk CA, Abdrabbo MK, et al: Gastric cancers overexpress S100A calcium-binding proteins. Cancer Res 62:6823-6, 2002 157
200. Rudland PS, Platt-Higgins A, Renshaw C, et al: Prognostic significance of the metastasis-inducing protein S100A4 (p9Ka) in human breast cancer. Cancer Res 60:1595-603, 2000 201. Lee WY, Su WC, Lin PW, et al: Expression of S100A4 and Met: potential predictors for metastasis and survival in early-stage breast cancer. Oncology 66:429-38, 2004 202. Breen EC, Tang K: Calcyclin (S100A6) regulates pulmonary fibroblast proliferation, morphology, and cytoskeletal organization in vitro. J Cell Biochem 88:848-54, 2003 203. Okazaki K, Niki I, Iino S, et al: A role of calcyclin, a Ca(2+)-binding protein, on the Ca(2+)-dependent insulin release from the pancreatic beta cell. J Biol Chem 269:6149-52, 1994 204. Filipek A, Jastrzebska B, Nowotny M, et al: CacyBP/SIP, a calcyclin and Siah- 1-interacting protein, binds EF-hand proteins of the S100 family. J Biol Chem 277:28848-52, 2002 205. Komatsu K, Murata K, Kameyama M, et al: Expression of S100A6 and S100A4 in matched samples of human colorectal mucosa, primary colorectal adenocarcinomas and liver metastases. Oncology 63:192-200, 2002 206. Kim J, Kim J, Yoon S, et al: S100A6 protein as a marker for differential diagnosis of cholangiocarcinoma from hepatocellular carcinoma. Hepatol Res 23:274, 2002 207. Maelandsmo GM, Florenes VA, Mellingsaeter T, et al: Differential expression patterns of S100A2, S100A4 and S100A6 during progression of human malignant melanoma. Int J Cancer 74:464-9, 1997 208. Joo JH, Kim JW, Lee Y, et al: Involvement of NF-kappaB in the regulation of S100A6 gene expression in human hepatoblastoma cell line HepG2. Biochem Biophys Res Commun 307:274-80, 2003 209. Rehman I, Cross SS, Azzouzi AR, et al: S100A6 (Calcyclin) is a prostate basal cell marker absent in prostate cancer and its precursors. Br J Cancer 91:739-44, 2004 158
210. Rehman I, Cross SS, Catto JW, et al: Promoter hyper-methylation of calcium binding proteins S100A6 and S100A2 in human prostate cancer. Prostate 65:322-30, 2005 211. Becker T, Gerke V, Kube E, et al: S100P, a novel Ca(2+)-binding protein from human placenta. cDNA cloning, recombinant protein expression and Ca2+ binding properties. Eur J Biochem 207:541-7, 1992 212. Lawrance IC, Fiocchi C, Chakravarti S: Ulcerative colitis and Crohn's disease: distinctive gene expression profiles and novel susceptibility candidate genes. Hum Mol Genet 10:445-56, 2001 213. Bertram J, Palfner K, Hiddemann W, et al: Elevated expression of S100P, CAPL and MAGE 3 in doxorubicin-resistant cell lines: comparison of mRNA differential display reverse transcription-polymerase chain reaction and subtractive suppressive hybridization for the analysis of differential gene expression. Anticancer Drugs 9:311-7, 1998 214. Guerreiro Da Silva ID, Hu YF, Russo IH, et al: S100P calcium-binding protein overexpression is associated with immortalization of human breast epithelial cells in vitro and early stages of breast cancer development in vivo. Int J Oncol 16:231-40, 2000 215. Averboukh L, Liang P, Kantoff PW, et al: Regulation of S100P expression by androgen. Prostate 29:350-5, 1996 216. Sobin LH, Wittekind C: TNM Classification of malignant tumours, UICC International Union Against Cancer (ed Fifth). New York, Wiley-Liss, 1997 217. Ilg EC, Schafer BW, Heizmann CW: Expression pattern of S100 calcium- binding proteins in human tumors. Int J Cancer 68:325-32, 1996 218. Gupta S, Hussain T, MacLennan GT, et al: Differential expression of S100A2 and S100A4 during progression of human prostate adenocarcinoma. J Clin Oncol 21:106-12, 2003 219. Feng G, Xu X, Youssef EM, et al: Diminished expression of S100A2, a putative tumor suppressor, at early stage of human lung carcinogenesis. Cancer Res 61:7999-8004, 2001 159
220. Lauriola L, Michetti F, Maggiano N, et al: Prognostic significance of the Ca(2+) binding protein S100A2 in laryngeal squamous-cell carcinoma. Int J Cancer 89:345-9, 2000 221. Wicki R, Franz C, Scholl FA, et al: Repression of the candidate tumor suppressor gene S100A2 in breast cancer is mediated by site-specific hypermethylation. Cell Calcium 22:243-54, 1997 222. Smith SL, Gugger M, Hoban P, et al: S100A2 is strongly expressed in airway basal cells, preneoplastic bronchial lesions and primary non-small cell lung carcinomas. Br J Cancer 91:1515-24, 2004 223. Shrestha P, Muramatsu Y, Kudeken W, et al: Localization of Ca(2+)-binding S100 proteins in epithelial tumours of the skin. Virchows Arch 432:53-9, 1998 224. Xia L, Stoll SW, Liebert M, et al: CaN19 expression in benign and malignant hyperplasias of the skin and oral mucosa: evidence for a role in regenerative differentiation. Cancer Res 57:3055-62, 1997 225. Stoll SW, Zhao X, Elder JT: EGF stimulates transcription of CaN19 (S100A2) in HaCaT keratinocytes. J Invest Dermatol 111:1092-7, 1998 226. Tonini GP, Casalaro A, Cara A, et al: Inducible expression of calcyclin, a gene with strong homology to S-100 protein, during neuroblastoma cell differentiation and its prevalent expression in Schwann-like cell lines. Cancer Res 51:1733-7, 1991 227. Segara D, Biankin AV, Kench JG, et al: Expression of HOXB2, a retinoic acid signaling target in pancreatic cancer and pancreatic intraepithelial neoplasia. Clin Cancer Res 11:3587-96, 2005 228. Luu HH, Zhou L, Haydon RC, et al: Increased expression of S100A6 is associated with decreased metastasis and inhibition of cell migration and anchorage independent growth in human osteosarcoma. Cancer Lett 229:135-48, 2005 229. Dowen SE, Crnogorac-Jurcevic T, Gangeswaran R, et al: Expression of S100P and its novel binding partner S100PBPR in early pancreatic cancer. Am J Pathol 166:81-92, 2005 230. Arumugam T, Simeone DM, Van Golen K, et al: S100P promotes pancreatic cancer growth, survival, and invasion. Clin Cancer Res 11:5356-64, 2005 160
231. Balague C, Gambus G, Carrato C, et al: Altered expression of MUC2, MUC4, and MUC5 mucin genes in pancreas tissues and cancer cell lines. Gastroenterology 106:1054-61, 1994 232. Jaster R, Hilgendorf I, Fitzner B, et al: Regulation of pancreatic stellate cell function in vitro: biological and molecular effects of all-trans retinoic acid. Biochem Pharmacol 66:633-41, 2003 233. Liu T-X, Zhang J-W, Tao J, et al: Gene expression networks underlying retinoic acid-induced differentiation of acute promyelocytic leukemia cells. Blood 96:1496-1504, 2000 234. Blomhoff R, Green MH, Berg T, et al: Transport and storage of vitamin A. Science 250:399-404, 1990 235. Clagett-Dame M, DeLuca HF: The role of vitamin A in mammalian reproduction and embryonic development. Annu Rev Nutr 22:347-81, 2002 236. Segars JH, Nagata T, Bours V, et al: Retinoic acid induction of major histocompatibility complex class I genes in NTera-2 embryonal carcinoma cells involves induction of NF-kappa B (p50-p65) and retinoic acid receptor beta- retinoid X receptor beta heterodimers. Mol Cell Biol 13:6157-69, 1993 237. Friedman A, Halevy O, Schrift M, et al: Retinoic acid promotes proliferation and induces expression of retinoic acid receptor-alpha gene in murine T lymphocytes. Cell Immunol 152:240-8, 1993 238. Yang Y, Minucci S, Ozato K, et al: Efficient inhibition of activation-induced Fas ligand up-regulation and T cell apoptosis by retinoids requires occupancy of both retinoid X receptors and retinoic acid receptors. J Biol Chem 270:18672-7, 1995 239. Evans TR, Kaye SB: Retinoids: present role and future potential. Br J Cancer 80:1-8, 1999 240. Soprano DR, Qin P, Soprano KJ: Retinoic acid receptors and cancers. Annu Rev Nutr 24:201-21, 2004 241. Prasad NB, Biankin AV, Fukushima N, et al: Gene expression profiles in pancreatic intraepithelial neoplasia reflect the effects of Hedgehog signaling on pancreatic ductal epithelial cells. Cancer Res 65:1619-26, 2005 161
242. Robbins MJ, Michalovich D, Hill J, et al: Molecular cloning and characterization of two novel retinoic acid-inducible orphan G-protein-coupled receptors (GPRC5B and GPRC5C). Genomics 67:8-18, 2000 243. Cheng Y. LDaLR: Proc Am Assoc Cancer Res 40:308, 1999 244. Tao Q, Cheng Y, Clifford J, et al: Characterization of the murine orphan G- protein-coupled receptor gene Rai3 and its regulation by retinoic acid. Genomics 83:270-80, 2004 245. Chambon P: A decade of molecular biology of retinoic acid receptors. Faseb J 10:940-54, 1996 246. Altucci L, Gronemeyer H: The promise of retinoids to fight against cancer. Nat Rev Cancer 1:181-93, 2001 247. Mangelsdorf DJ, Evans RM: The RXR heterodimers and orphan receptors. Cell 83:841-50, 1995 248. Cheng Y, Lotan R: Molecular cloning and characterization of a novel retinoic acid-inducible gene that encodes a putative G protein-coupled receptor. J Biol Chem 273:35008-15, 1998 249. Wu Q, Ding W, Mirza A, et al: Integrative Genomics Revealed RAI3 Is a Cell Growth-promoting Gene and a Novel P53 Transcriptional Target. J Biol Chem 280:12935-43, 2005 250. Nagahata T, Sato T, Tomura A, et al: Identification of RAI3 as a therapeutic target for breast cancer. Endocr Relat Cancer 12:65-73, 2005 251. Cillo C, Faiella A, Cantile M, et al: Homeobox genes and cancer. Exp Cell Res 248:1-9, 1999 252. Fire A, Xu S, Montgomery MK, et al: Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature 391:806-11, 1998 253. Bernstein E, Denli AM, Hannon GJ: The rest is silence. Rna 7:1509-21, 2001 254. Elbashir SM, Martinez J, Patkaniowska A, et al: Functional anatomy of siRNAs for mediating efficient RNAi in Drosophila melanogaster embryo lysate. Embo J 20:6877-88, 2001 255. Hutvagner G, Zamore PD: A microRNA in a multiple-turnover RNAi enzyme complex. Science 297:2056-60, 2002 162
256. Lecellier CH, Voinnet O: RNA silencing: no mercy for viruses? Immunol Rev 198:285-303, 2004 257. Elbashir SM, Harborth J, Lendeckel W, et al: Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature 411:494-8, 2001 258. Brat DJ, Lillemoe KD, Yeo CJ, et al: Progression of pancreatic intraductal neoplasias to infiltrating adenocarcinoma of the pancreas. Am J Surg Pathol 22:163-9, 1998 259. Jacquemier J, Ginestier C, Rougemont J, et al: Protein expression profiling identifies subclasses of breast cancer and predicts prognosis. Cancer Res 65:767- 79, 2005 163
Appendices
1. MICROARRAY TARGET PREPARATION
2. DETAILED ISH PROTOCOL
3. LIST OF UPREGULATED GENES IN PC COMPARED TO NORMAL PANCREAS IDENTIFIED USING TRANSCRIPT PROFILING
4. LIST OF UPREGULATED GENES IN PC COMPARED TO NORMAL PANCREAS IDENTIFIED USING TRANSCRIPT PROFILING
5. TRIAL PROTOCOL
6. PUBLICATIONS
7. ETHICAL APPROVAL DOCUMENTATION 1. Microarray Target Preparation APPENDIX 1: MICROARRAY TARGET PREPARATION 1. Production of fragmented cRNA
First Strand cDNA synthesis reagents:
• T7-(T)24 Primer is ordered from geneworks • Superscript II reverse transcriptase 2000U Gibco Life Sciences Order No. 18064-022 (you get the DTT and First strand buffer with this) • dNTP's Promega 100mM 400ul for each CatNo: U120A, U121A, U122A and U123A
Second Strand cDNA synthesis:
• DNA polymerase I, E.coli 250U (5U/ul) Roche/Boehringer Order No.642711 • E.coli DNA Ligase, 100U (10U/ul) Invitrogen Order No. 18052-019 • RNase H 25U (1U/ul) Roche Order No. 786349 • 2nd stand buffer - 5X Second strand buffer 500UL Gibco the cat no. is 10812-014 • dNTP's Promega product 100mM 400ul for each Cat No U120A, U121A, U122A and U123A • T4 DNA polymerase 100U (1U/ul) Roche Order No. 1004786 • Spin Phase Lock tubes (dscDNA clean up) Eppendorf Order Number 32005.101 • T4 Gene32 Amersham-USB Part no. 70029Z (must be made more concentrated) • Rnase Inhibitor Geneworks 2500U 40U/ul cat no. P.18A • Microcon Millipore Cat no 42421 ( used to concentrate T4 Gene32 before use in cRNA prep) • Glycogen 20mg Roche Order No. 901393
IVT reagents: • Bioarray High Yield Transcript labelling Kit 10 reactions Enzo 900182 • IVT clean up RNAeasy miniKit 50 tubes Qiagen Order No. 74104 Small scale method for preparing cRNA
Method based on protocol by [email protected], Nucleic Acids Research 2001, vol 29, no 5 Baugh et al
(A) Isolating Sample RNA • Use glass filter purified RNA • Run 1-2μl RNA sample on 1% EtBr TBE agarose gel to confirm RNA is not degraded (use RNA loading buffer which contains formalin and heat 70oC 5min to denature). Run gel at ~100V (low) for 40-60min. Also quantify RNA on spectrophotometer. • Aliquot 500ng sample RNA into a 0.2ml PCR tube. If the volume is greater then 5ul dry the RNA down to a 3.5ul volume using pre- measured 0.2ml vials with varying volumes of water in them. If the volume of the 500ng is less then 3.5ul make it up with RNAse free
H20.
(B) First strand cDNA synthesis μ 1. Add 1 l 5uM HPLC T7-(T)24 primer (100ng) to the 500ng sample RNA (section A) (GGC CAG TGA ATT GTA ATA CGA CTC ACT ATA GGG AGG CGG-
(dT)24) 2. Mix, quick spin if needed 3. Heat at 70oC in a PCR machine for 4min 4. Place straight on compacted ice to cool to 4 oC. 5. On ice mix the following cDNA reagent mix if two samples make up 2.25X then aliquot in each tube: 1X
5x 1st strand buffer 2μl 0.1M DTT (fresh) 1μl 10mM dNTPs (10mM each dATP, dCTP, dGTP, dNTP) 0.5μl T4gp32 (2.4mg/ml) 1μl Rnase Inhibitor (40U/μl) 0.5μl Superscript II (200U/μl) 0.5μl
Mix and quick spin if needed
6. Add 5.5μl of cDNA reagent mix to the RNA/Primer tube, vortex and incubate at 42oC, 1 hour 7. Heat inactivate at 65 oC for 15 minutes (Note: take the cDNA tube out of PCR machine whilst it reaches 65 oC, then place it back in the PCR machine), place on ice. 8. Proceed to second strand cDNA synthesis (C) Second Strand cDNA synthesis 1. Ice all reagents and first strand cDNA tubes (from section B) 2. Make up the second strand reagent mix if have two samples make a master mix of 1X
H2O 43.2ul 5x 2nd strand buffer 15ul 10mM dNTPs (10mM each dATP, dCTP, dGTP, dNTP) 1.5ul DNA polymerase I (5U/ul) 4ul E.coli DNA ligase (10U/ul) 0.5ul RnaseH (1U/ul) 1ul
Mix and spin if needed
3. Add 65μl of second strand reagent mix to the first strand cDNA tube and mix by pipetting. Incubate at 16oC for 2 hours. 4. Add 10μl (10U total) T4 DNA polymerase, incubate 16oC for 15min (to finish off ends properly) 5. Heat inactivate by taking tube out of PCR machine, increase temperature to 70 oC, then place tube back in PCR machine for 10 minutes. Place tube on compacted ice to cool. NOTE: Can store at –80oC if required
(D) Clean up of dsDNA 1. Spin Phase-Lock tubes at maximum speed for 30sec (so gel comes down to bottom of tubes) 2. Add all the cDNA reaction (75μl) 3. Add equal volume buffered saturated phenol (or phenol/chloroform) in fume hood 4. Vortex lightly (white ppt) 5. Spin at maximum speed ~13000rpm, 5 min at 4oC 6. Transfer upper phase to new tube (~75μl) 7. Prepare BioGel P-6 Microspin column. Shake Biogel to get rid of air bubbles then snap off bottom of column and remove lid so liquid can run out. Spin at 3500rpm for 2 min. Place colomn into a fresh RNAse free tube. 8. Transfer aqueous phase from the Spin-Phase lock tube to the Bio gel P-6 and spin at 3500rpm for 4min collecting 75ul in the fresh tube. 9. Precipitate the dscDNA by adding 2.5V 100% Ethanol and 1ul glycogen (20mg/ml) leave at -20 oC for 2 hours (or O/N). 10. Spin at 13000rpm for 20 min. Remove S/N from pellet. 11. Wash 1x with 70% ethanol and spin down at 13,000rpm for 5min. Remove S/N and pulse spin and remove residual ethanol. 12. Air dry or use RNA desicator for 2-3min then take pellet up in 22ul of Rnase free water. (E) In vitro transcription
1. To 22μl ds cDNA add: 4μl 10x HY reaction buffer 4μl 10x biotin-labelled ribonucleotide 4μl 10x DTT 4μl 10x RNase inhibitor mix 2μl 20x T7 RNA polymerase Total volume 40μl 2. Carefully mix and spin liquid down and transfer to a 0.2ml tube. 3. Incubate for 9 hours at 37oC in PCR machine. This can be done O/N. 4. Spin down liquid
(F) IVT clean up 1. To IVT reaction tube add: μ 160 lDEPCH2O 700μl RLT buffer (need to add 10μl ME per ml RLT) 2. Mix 3. Add 500μl 100% ethanol 4. Transfer 700μl to two separate spin columns (RNeasy mini column) 5. Spin at maximum speed 13,000rpm for 15sec-1min RT 6. Transfer columns to new collection tubes 7. Add 500μl RPE buffer 8. Spin at 13,000rpm for 30sec 9. Add 500μl RPE buffer 10. Spin at 13,000rpm for 2min 11. Remove flow through and spin again to get rid of RPE on the column surface. 12. Transfer spin columns to new collection tube μ 13. Add 50 lDEPCH2O to membrane of spin column 14. Let soak for 4min 15. Spin at 13,000rpm, 1min 16. Repeat steps 12 to 14 using first the first elution as the second eluate solution i.e instead of adding 50ul of fresh water to column for the second elution use the first elution liquid, now in bottom of tube containing cRNA, to place back on column and run through again. 17. Take 2x1μl samples – one will be run on a gel and the other to measure cRNA concentration using the spectrophotmeter. Store all samples at –80oCO/N
(G) Quantitate cRNA μ μ 1. Take 1 l cRNA sample up in 75 lH2O (1:75 concentration) and take another tube with 75μl of dilution water as a blank 2. Determine spectrophotometer reading – require 260/280 >1.8 3. Need 10-15μg starting material to carry out affymetrix. 4. Fragment all cRNA – require a final concentration of 1μg/μl (H) Fragmentation of cRNA μ μ μ 1. Take all cRNA up in H2O to get a concentration of 1.25 g/ l(eg19.2 gIVT+ μ 16 lH2O) and add 5xfragmentation buffer to get a 1X concentration. 2. Mix 3. Heat at 95oC for 35min (no longer) 4. Take 1μl sample to run on gel (to check cRNA is completely fragmented) and freeze the rest at –80oC
Target Hybridisation (As per Affymetrix Genechip Analysis Manual)
(A) Preparation of hybridisation cocktail
1. Mix the following reagents:
Component Standard Array Final Concentration Fragmented cRNA 15 μg0.05μg/μL
Control Oligonucleotide B2 5 μL50pM (3nM)
20X Eukaryotic Hybridisation 15 μL 1.5, 5, 25 and 100pM Controls (bioB, bioC, bioD, cre) respectively
Herring Sperm DNA (10mg/mL) 3 μL0.1mg/mL
Acetylated BSA (50mg/mL) 3 μL0.5mg/mL
2x Hybridisation Buffer 150 μL1x
H2O to final volume of 300μL
Final Volume 300μL
Note: It is important that frozen stocks of 20x Genechip Eukaryptic Hybridisation Control are heated to 65oC for 5 minutes to completely resuspent the cRNA before aliquotting.
2. Equilibrate probe to room temperature before use. 3. Heat the hybridisation cocktail to 99oC for 5 minutes in a heat block. 4. Meanwhile wet the array with appropriate volume of 1x hybridisation buffer and incubate in hybridisation oven at 45oC and 60 revolutions per minute for 10 minutes. 5. Transfer hybridisation cocktail to 45oC heat block for 5 mins. 6. Spin hynridisation cocktail at maximum for 5 mins to remove insoluble components. 7. Remove buffer from array and fill with appropriate volume of hybridisation cocktail avoiding insoluble matter at the bottom of the tube. 8. Hybridise for 16 hours at 45oC and 60 revolutions per minute in hybridisation oven. (B) Probe Array Staining with antibody amplification (Buffers as per Affymetrix Protocol).
1. Prior to end of hybridisation input experimental data on Affymetrix MicroArray Suite Software package (used for Fluidics Station and Laser Reader. 2. Prepare SAPE staining solution:
Components Volume Final Concentration 2X MES Staining Buffer 600μL1X 50mg/mL Acetylated BSA 48μL 2 mg/mL 1mg/mL Streptavidin Phycoerythrin (SAPE) 12μL10μg/mL μ DI H2O 540 L- Total 1200μL
Mix well and divide into 2 aliquots of 600μL each to be used for stains 1 and 3. 3. Prepare Antibody Solution:
Components Volume Final Concentration 2X MES Staining Buffer 300μL1X 50mg/mL Acetylated BSA 24μL 2 mg/mL 10mg/mL Normal Goat IgG 6μL 0.1mg/mL 0.5 mg/mL biotinylated antibody 3.6μL3μg/mL μ DI H2O 266.4 L- Total 600μL
4. Remove hybridisation cocktail. 5. Fill with non-stringent wash buffer A. 6. Commence appropriate fluidics protocol in Microarray Suite. 7. At termination of fluidics protocols scan array using scanner as directed by Microarray Suite. 2. ISH Protocol
3. Up-regulated Genes in PC Upregulated transcripts
Affy Code UniGene Ace GenBank Acession Name Fold Change t test p value MW U test p value 204351_at Hs.2962 gb:NM_005980.1 S100 calcium-binding protein P 152.4 0.008 0.001 205044_at Hs.70725 gb:NM_014211.1 gamma-aminobutyric acid (GABA) A receptor, pi 38.2 0.042 0.001 226930_at Hs.297939 gb:AI345957_RC cathepsin B 37.7 0.002 0.001 210809_s_at Hs.136348 gb:D13665.1 osteoblast specific factor 2 (fasciclin I-like) 35 0.001 0.001 203256_at Hs.2877 gb:NM_001793.1 cadherin 3, type 1, P-cadherin (placental) 29.3 0.001 0.001 201650_at Hs.182265 gb:NM_002276.1 keratin 19 27.2 0.023 0.035 203108_at Hs.194691 gb:NM_003979.2 retinoic acid induced 3 26.3 0.007 0.012 211959_at Hs.103391 gb:AW007532_RC Human insulin-like growth factor binding protein 5 (IGFBP5) mRNA 25.5 0.011 0.002 210511_s_at Hs.727 gb:M13436.1 inhibin, beta A (activin A, activin AB alpha polypeptide) 24.5 0.008 0.012 223631_s_at Hs.145362 gb:AF213678.1 immortalization-upregulated protein 24.3 0.006 0.007 223278_at Hs.323733 gb:M86849.2 gap junction protein, beta 2, 26kD (connexin 26) 22.4 0.001 0.001 218960_at Hs.63325 gb:NM_016425.1 transmembrane protease, serine 4 22.1 0.004 0.002 223586_at Hs.222024 gb:AF256215.1 transcription factor BMAL2 21.8 0 0.001 204855_at Hs.55279 gb:NM_002639.1 serine (or cysteine) proteinase inhibitor, clade B (ovalbumin), member 5 21.6 0.021 0.002 202404_s_at Hs.179573 gb:NM_000089.1 collagen, type I, alpha 2 21.3 0.008 0.001 211161_s_at Hs.119571 gb:AF130082.1 collagen, type III, alpha 1 (Ehlers-Danlos syndrome type IV, autosomal dominant) 20.8 0.041 0.001 205081_at Hs.17409 gb:NM_001311.1 cysteine-rich protein 1 (intestinal) 20.5 0.018 0.004 207156_at Hs.51011 gb:NM_021064.1 H2A histone family, member P 20.5 0.038 0.007 209360_s_at Hs.129914 gb:D43968.1 runt-related transcription factor 1 (acute myeloid leukemia 1; aml1 oncogene) 20 0.001 0.001 205700_at Hs.11958 gb:NM_003725.1 oxidative 3 alpha hydroxysteroid dehydrogenase; retinol dehydrogenase; 3-hydroxysteroid epimerase 19.7 0.008 0.002 210095_s_at Hs.77326 gb:M31159.1 insulin-like growth factor binding protein 3 19.4 0.019 0.001 202465_at Hs.202097 gb:NM_002593.2 procollagen C-endopeptidase enhancer 19.1 0.001 0.003 205234_at Hs.23590 gb:NM_004696.1 solute carrier family 16 (monocarboxylic acid transporters), member 4 18.7 0.009 0.002 222592_s_at Hs.11638 gb:AW173691_RC long-chain fatty acid coenzyme A ligase 5 18.1 0.04 0.005 227725_at Hs.105352 gb:Y11339.2 GalNAc alpha-2, 6-sialyltransferase I, long form 17.6 0.035 0.004 217755_at Hs.109706 gb:NM_016185.1 hematological and neurological expressed 1 17.6 0.031 0.002 203878_s_at Hs.155324 gb:NM_005940.2 matrix metalloproteinase 11 (stromelysin 3) 16.4 0.021 0.005 204320_at Hs.82772 gb:NM_001854.1 collagen, type XI, alpha 1 16.3 0.01 0.004 206091_at Hs.278461 gb:NM_002381.2 matrilin 3 16.3 0.032 0.004 224694_at Hs.8966 gb:AF279145.1 tumor endothelial marker 8 16 0.002 0.001 205009_at Hs.1406 gb:NM_003225.1 trefoil factor 1 (breast cancer, estrogen-inducible sequence expressed in) 15.5 0.022 0.005 203757_s_at Hs.73848 gb:BC005008.1 carcinoembryonic antigen-related cell adhesion molecule 6 (non-specific cross reacting antigen) 15.4 0.018 0.002 217728_at Hs.275243 gb:NM_014624.2 S100 calcium-binding protein A6 (calcyclin) 15.2 0.004 0.001 205366_s_at Hs.98428 gb:NM_018952.1 homeo box B6 14.4 0.009 0.001 222565_s_at Hs.143460 gb:BF978541 protein kinase C, nu 14.3 0.007 0.001 203234_at Hs.77573 gb:NM_003364.1 uridine phosphorylase 14.2 0 0.004 204619_s_at Hs.81800 gb:BF590263_RC chondroitin sulfate proteoglycan 2 (versican) 14.1 0.001 0.001 203789_s_at Hs.171921 gb:NM_006379.1 sema domain, immunoglobulin domain (Ig), short basic domain, secreted, (semaphorin) 3C 13.8 0 0.001 204698_at Hs.183487 gb:NM_002201.2 interferon stimulated gene (20kD) 13.6 0.015 0.003 216834_at Hs.75256 gb:S59049.1 regulator of G-protein signalling 1 13.6 0.014 0.035 212354_at Hs.70823 gb:BE500977_RC KIAA1077 protein 13.5 0.002 0.001 212531_at Hs.204238 gb:NM_005564.1 lipocalin 2 (oncogene 24p3) 13.5 0.031 0.009 211924_s_at gb:AY029180.1 soluble urokinase plasminogen activator receptorprecursor 13.5 0.037 0.002 221729_at Hs.82985 gb:AL575735_RC collagen, type V, alpha 2 13.1 0.009 0.001 205927_s_at Hs.1355 gb:NM_001910.1 cathepsin E 12.6 0.011 0.001 208515_at Hs.182432 gb:NM_003521.1 H2B histone family, member E 12.5 0.033 0.014 202859_x_at Hs.624 gb:NM_000584.1 interleukin 8 12.5 0.01 0.001 202898_at Hs.158287 gb:NM_014654.1 KIAA0468 gene product 12.2 0.029 0.004 212489_at Hs.146428 gb:AI983428_RC collagen, type V, alpha 1 12.1 0.001 0.001 209792_s_at Hs.69423 gb:BC002710.1 kallikrein 10 12.1 0.046 0.016 209373_at Hs.185055 gb:BC003179.1 BENE protein 12 0.035 0.016 210495_x_at Hs.287820 gb:AF130095.1 fibronectin 1 12 0 0.001 214385_s_at Hs.102482 gb:AI521646_RC mucin 5, subtype B, tracheobronchial 12 0.028 0.007 209125_at Hs.111758 gb:J00269.1 keratin 6A 11.9 0.007 0.004 211430_s_at Hs.300697 gb:M87789.1 immunoglobulin heavy constant gamma 3 (G3m marker) 11.7 0.008 0.003 204734_at Hs.80342 gb:NM_002275.1 keratin 15 11.7 0.009 0.004 223229_at Hs.5199 gb:AB032931.1 HSPC150 protein similar to ubiquitin-conjugating enzyme 11.5 0.004 0.003 209803_s_at Hs.154036 gb:AF001294.1 tumor suppressing subtransferable candidate 3 11.4 0.003 0.004 217428_s_at Hs.179729 gb:X98568 collagen, type X, alpha 1 (Schmid metaphyseal chondrodysplasia) 11.3 0 0.001
Page 1 of 17 Upregulated transcripts
Affy Code UniGene Ace GenBank Acession Name Fold Change t test p value MW U test p value 206023_at Hs.2841 gb:NM_006681.1 neuromedin U 11.3 0.027 0.002 218644_at Hs.39957 gb:NM_016445.1 pleckstrin 2 (mouse) homolog 11.2 0.003 0.001 203240_at Hs.111732 gb:NM_003890.1 Fc fragment of IgG binding protein 11.1 0.02 0.007 206067_s_at Hs.1145 gb:NM_024426.1 Wilms tumor 1 11 0.009 0.012 213790_at Hs.8850 gb:W46291_RC a disintegrin and metalloproteinase domain 12 (meltrin alpha) 10.9 0.004 0.004 202497_x_at Hs.7594 gb:AI631159_RC solute carrier family 2 (facilitated glucose transporter), member 3 10.9 0.003 0.002 222484_s_at Hs.24395 gb:AF144103.1 small inducible cytokine subfamily B (Cys-X-Cys), member 14 (BRAK) 10.7 0.008 0.007 218468_s_at Hs.40098 gb:AF154054.1 cysteine knot superfamily 1, BMP antagonist 1 10.6 0.008 0.003 213160_at Hs.17211 gb:D86964.1 dedicator of cyto-kinesis 2 10.1 0.001 0.003 212768_s_at Hs.273321 gb:AL390736 differentially expressed in hematopoietic lineages 10.1 0.008 0.002 202267_at Hs.54451 gb:NM_005562.1 laminin, gamma 2 (nicein (100kD), kalinin (105kD), BM600 (100kD), Herlitz junctional epidermolysis bullosa)) 9.9 0.022 0.007 202934_at Hs.198427 gb:AI761561_RC hexokinase 2 9.7 0.003 0.009 221730_at Hs.82985 gb:NM_000393.1 collagen, type V, alpha 2 9.6 0.002 0.001 201645_at Hs.289114 gb:NM_002160.1 hexabrachion (tenascin C, cytotactin) 9.6 0 0.001 204885_s_at Hs.155981 gb:NM_005823.2 mesothelin 9.6 0.039 0.007 205421_at Hs.81086 gb:NM_021977.1 solute carrier family 22 (extraneuronal monoamine transporter), member 3 9.6 0.037 0.027 203083_at Hs.108623 gb:NM_003247.1 thrombospondin 2 9.6 0.001 0.007 207172_s_at Hs.75929 gb:NM_001797.1 cadherin 11, type 2, OB-cadherin (osteoblast) 9.5 0.001 0.001 207714_s_at Hs.241579 gb:NM_004353.1 serine (or cysteine) proteinase inhibitor, clade H (heat shock protein 47), member 1 9.5 0.022 0.035 221558_s_at Hs.44865 gb:AF288571.1 lymphoid enhancer binding factor-1 9.3 0.001 0.002 206560_s_at Hs.279651 gb:NM_006533.1 melanoma inhibitory activity 9.3 0.036 0.007 204825_at Hs.184339 gb:NM_014791.1 KIAA0175 gene product 9.1 0.005 0.003 202307_s_at Hs.158164 gb:NM_000593.2 ATP-binding cassette, sub-family B (MDRTAP), member 2 9 0.001 0.003 228754_at Hs.296770 gb:BG150485_RC KIAA1719 protein 9 0.001 0.001 203726_s_at Hs.83450 gb:NM_000227.1 laminin, alpha 3 (nicein (150kD), kalinin (165kD), BM600 (150kD), epilegrin) 9 0.013 0.044 219478_at Hs.36688 gb:NM_021197.1 WAP four-disulfide core domain 1 9 0.005 0.001 209260_at Hs.184510 gb:BC000329.1 stratifin 8.9 0.039 0.005 205997_at Hs.174030 gb:NM_021778.1 a disintegrin and metalloproteinase domain 28 8.8 0.02 0.012 202310_s_at Hs.172928 gb:K01228.1 collagen, type I, alpha 1 8.8 0.001 0.003 203213_at Hs.184572 gb:AL524035_RC cell division cycle 2, G1 to S and G2 to M 8.6 0.013 0.003 202437_s_at Hs.154654 gb:NM_000104.2 cytochrome P450, subfamily I (dioxin-inducible), polypeptide 1 (glaucoma 3, primary infantile) 8.6 0.002 0.003 219508_at Hs.194710 gb:NM_004751.1 glucosaminyl (N-acetyl) transferase 3, mucin type 8.6 0.048 0.001 218166_s_at Hs.20509 gb:NM_016578.2 HBV pX associated protein-8 8.6 0.008 0.001 217875_s_at Hs.83883 gb:NM_020182.1 transmembrane, prostate androgen induced RNA 8.6 0.001 0.005 228923_at Hs.183418 gb:AW166825_RC cell division cycle 2-like 1 (PITSLRE proteins) 8.5 0.008 0.016 209955_s_at Hs.418 gb:U76833.1 fibroblast activation protein, alpha 8.5 0 0.007 211402_x_at Hs.278599 gb:AF004291.1 nuclear receptor subfamily 6, group A, member 1 8.5 0.017 0.005 225275_at Hs.10283 gb:AA053711_RC RNA binding motif protein 8B 8.4 0.009 0.001 212457_at Hs.274184 gb:AL161985.1 transcription factor binding to IGHM enhancer 3 8.4 0.003 0.005 202555_s_at Hs.211582 gb:NM_005965.1 myosin, light polypeptide kinase 8.3 0.003 0.002 209546_s_at Hs.114309 gb:AF323540.1 apolipoprotein L 8.1 0.004 0.012 202450_s_at Hs.83942 gb:NM_000396.1 cathepsin K (pycnodysostosis) 8.1 0.003 0.001 201474_s_at Hs.265829 gb:NM_002204.1 integrin, alpha 3 (antigen CD49C, alpha 3 subunit of VLA-3 receptor) 8.1 0.009 0.002 212236_x_at Hs.2785 gb:Z19574 keratin 17 8.1 0.021 0.002 228491_at Hs.182265 gb:AW662246_RC keratin 19 8.1 0.008 0.001 229689_s_at Hs.170290 gb:AV721789 discs, large (Drosophila) homolog 5 7.6 0.002 0.003 201798_s_at Hs.234680 gb:NM_013451.1 fer-1 (C.elegans)-like 3 (myoferlin) 7.6 0.001 0.001 205844_at Hs.12114 gb:NM_004666.1 vanin 1 7.6 0.013 0.009 217296_at Hs.258612 gb:AF135564.1_RC killer cell immunoglobulin-like receptor, two domains, short cytoplasmic tail, 4 7.5 0.031 0.001 203417_at Hs.83551 gb:NM_017459.1 microfibrillar-associated protein 2 7.5 0.009 0.005 220066_at Hs.135201 gb:NM_022162.1 NOD2 protein 7.5 0.002 0.002 204636_at Hs.117938 gb:NM_000494.1 collagen, type XVII, alpha 1 7.4 0.041 0.021 212413_at Hs.90998 gb:D50918.1 KIAA0128 protein; septin 2 7.4 0.032 0.009 218755_at Hs.73625 gb:NM_005733.1 RAB6 interacting, kinesin-like (rabkinesin6) 7.4 0.013 0.004 200665_s_at Hs.111779 gb:NM_003118.1 secreted protein, acidic, cysteine-rich (osteonectin) 7.4 0.002 0.007 218002_s_at Hs.24395 gb:NM_004887.1 small inducible cytokine subfamily B (Cys-X-Cys), member 14 (BRAK) 7.4 0.019 0.012 211833_s_at Hs.159428 gb:U19599.1 BCL2-associated X protein 7.3 0.024 0.004 205466_s_at Hs.40968 gb:NM_005114.1 heparan sulfate (glucosamine) 3-O-sulfotransferase 1 7.3 0.027 0.016 201621_at Hs.76307 gb:NM_005380.1 neuroblastoma, suppression of tumorigenicity 1 7.3 0 0.002
Page 2 of 17 Upregulated transcripts
Affy Code UniGene Ace GenBank Acession Name Fold Change t test p value MW U test p value 204286_s_at Hs.96 gb:NM_021127.1 phorbol-12-myristate-13-acetate-induced protein 1 7.3 0.006 0.005 226864_at Hs.75209 gb:BF245954 protein kinase (cAMP-dependent, catalytic) inhibitor alpha 7.3 0.007 0.004 205890_s_at Hs.44532 gb:NM_006398.1 diubiquitin 7.2 0.026 0.001 202499_s_at Hs.7594 gb:NM_006931.1 solute carrier family 2 (facilitated glucose transporter), member 3 7.2 0.004 0.021 218308_at Hs.104019 gb:NM_006342.1 transforming, acidic coiled-coil containing protein 3 7.2 0.028 0.001 219134_at Hs.57958 gb:NM_022159.1 EGF-TM7-latrophilin-related protein 7.1 0.014 0.009 214574_x_at Hs.88411 gb:NM_007161.1 lymphocyte antigen 117 7.1 0.001 0.002 202869_at Hs.82396 gb:NM_016816.1 2,5-oligoadenylate synthetase 1 (40-46 kD) 7 0.002 0.004 202086_at Hs.76391 gb:NM_002462.1 myxovirus (influenza) resistance 1, homolog of murine (interferon-inducible protein p78) 7 0.005 0.001 217764_s_at Hs.223025 gb:AF183421.1 RAB31, member RAS oncogene family 7 0.005 0.001 206391_at Hs.82547 gb:NM_002888.1 retinoic acid receptor responder (tazarotene induced) 1 7 0 0.002 210665_at Hs.170279 gb:AF021834.1 tissue factor pathway inhibitor (lipoprotein-associated coagulation inhibitor) 7 0.026 0.004 200660_at Hs.256290 gb:NM_005620.1 S100 calcium-binding protein A11 (calgizzarin) 6.9 0.001 0.001 209720_s_at Hs.227948 gb:BC005224.1 serine (or cysteine) proteinase inhibitor, clade B (ovalbumin), member 3 6.9 0.032 0.012 209596_at Hs.72157 gb:AF245505.1 DKFZP564I1922 protein 6.8 0 0.001 204415_at Hs.265827 gb:NM_022873.1 interferon, alpha-inducible protein (clone IFI-6-16) 6.8 0.022 0.003 202237_at Hs.76669 gb:NM_006169.1 nicotinamide N-methyltransferase 6.8 0.002 0.005 204678_s_at Hs.79351 gb:U90065.1 potassium channel, subfamily K, member 1 (TWIK-1) 6.8 0.01 0.027 208131_s_at gb:NM_000961.1 prostaglandin I2 (prostacyclin) synthase 6.8 0.004 0.002 209950_s_at Hs.103665 gb:BC004300.1 villin-like 6.8 0.025 0.005 204254_s_at Hs.2062 gb:NM_000376.1 vitamin D (1,25- dihydroxyvitamin D3) receptor 6.8 0.006 0.005 209498_at Hs.50964 gb:X16354.1 carcinoembryonic antigen-related cell adhesion molecule 1 (biliary glycoprotein) 6.7 0.029 0.007 206994_at Hs.56319 gb:NM_001899.1 cystatin S 6.7 0 0.001 219479_at Hs.44970 gb:NM_024089.1 endoplasmic reticulum resident protein 58; hypothetical protein MGC5302 6.7 0.003 0.004 205453_at Hs.2733 gb:NM_002145.1 homeo box B2 6.7 0.001 0.001 214677_x_at Hs.289110 gb:X57812.1 immunoglobulin lambda joining 3 6.7 0.005 0.005 204052_s_at Hs.105700 gb:NM_003014.2 secreted frizzled-related protein 4 6.7 0.042 0.035 209900_s_at Hs.75231 gb:AL162079.1 solute carrier family 16 (monocarboxylic acid transporters), member 1 6.7 0 0.005 209813_x_at Hs.112259 gb:M16768.1 T cell receptor gamma locus 6.7 0.007 0.007 222449_at Hs.83883 gb:AL035541 transmembrane, prostate androgen induced RNA 6.7 0.013 0.016 205922_at Hs.121102 gb:NM_004665.1 vanin 2 6.7 0.005 0.007 205379_at Hs.154510 gb:NM_001236.2 carbonyl reductase 3 6.6 0.002 0.004 208727_s_at Hs.146409 gb:BC002711.1 cell division cycle 42 (GTP-binding protein, 25kD) 6.6 0.002 0.003 203821_at Hs.799 gb:NM_001945.1 diphtheria toxin receptor (heparin-binding epidermal growth factor-like growth factor) 6.6 0 0.002 204421_s_at Hs.284244 gb:M27968.1 fibroblast growth factor 2 (basic) 6.6 0.003 0.002 204879_at Hs.135150 gb:NM_006474.1 lung type-I cell membrane-associated glycoprotein 6.6 0.012 0.007 205499_at Hs.126782 gb:NM_014467.1 sushi-repeat protein 6.6 0.001 0.007 206662_at Hs.28988 gb:NM_002064.1 glutaredoxin (thioltransferase) 6.5 0.001 0.001 201655_s_at Hs.211573 gb:M85289.1 heparan sulfate proteoglycan 2 (perlecan) 6.5 0 0.001 203148_s_at Hs.179703 gb:NM_014788.1 KIAA0129 gene product 6.5 0.007 0.007 223122_s_at Hs.31386 gb:AF311912.1 secreted frizzled-related protein 2 6.5 0.007 0.035 225544_at Hs.267182 gb:AI806338_RC T-box 3 (ulnar mammary syndrome) 6.5 0 0.003 210656_at Hs.151461 gb:AF099032.1 embryonic ectoderm development 6.4 0.007 0.002 214031_s_at Hs.23881 gb:AI920979_RC keratin 7 6.4 0.016 0.007 209270_at Hs.75517 gb:L25541.1 laminin, beta 3 (nicein (125kD), kalinin (140kD), BM600 (125kD)) 6.4 0.015 0.007 207266_x_at Hs.241567 gb:NM_016837.1 RNA binding motif, single stranded interacting protein 1 6.4 0.002 0.035 212764_at Hs.232068 gb:AI806174_RC transcription factor 8 (represses interleukin 2 expression) 6.4 0.001 0.004 206925_at Hs.170180 gb:NM_005668.1 sialyltransferase 8 (alpha-2, 8-polysialytransferase) D 6.3 0.047 0.005 209901_x_at Hs.76364 gb:U19713.1 allograft inflammatory factor 1 6.2 0.012 0.007 227566_at Hs.288433 gb:AW085558_RC neurotrimin 6.2 0.001 0.001 219594_at Hs.239208 gb:NM_016533.1 ninjurin 2 6.2 0.015 0.003 212022_s_at Hs.80976 gb:BF001806_RC antigen identified by monoclonal antibody Ki-67 6.1 0.009 0.005 217028_at Hs.89414 gb:AJ224869 chemokine (C-X-C motif), receptor 4 (fusin) 6.1 0.003 0.003 213125_at Hs.43658 gb:AW007573_RC DKFZP586L151 protein 6.1 0 0.001 205376_at Hs.153687 gb:NM_003866.1 inositol polyphosphate-4-phosphatase, type II, 105kD 6.1 0.017 0.012 223344_s_at Hs.11090 gb:AB026043.1 membrane-spanning 4-domains, subfamily A, member 7 6.1 0.012 0.007 206082_at Hs.1845 gb:NM_006674.1 MHC class I region ORF 6.1 0.003 0.007 212667_at Hs.111779 gb:AL575922_RC secreted protein, acidic, cysteine-rich (osteonectin) 6.1 0.009 0.002 231353_at Hs.81008 gb:R77414_RC filamin B, beta (actin-binding protein-278) 6 0.008 0.005
Page 3 of 17 Upregulated transcripts
Affy Code UniGene Ace GenBank Acession Name Fold Change t test p value MW U test p value 202828_s_at Hs.2399 gb:NM_004995.2 matrix metalloproteinase 14 (membrane-inserted) 6 0.001 0.001 206482_at Hs.51133 gb:NM_005975.1 PTK6 protein tyrosine kinase 6 6 0.016 0.014 225701_at Hs.159578 gb:AK024431.1 AT-hook transcription factor AKNA 5.9 0.01 0.012 234315_at Hs.154276 gb:AF317902.1 BTB and CNC homology 1, basic leucine zipper transcription factor 1 5.9 0.011 0.021 205199_at Hs.63287 gb:NM_001216.1 carbonic anhydrase IX 5.9 0.02 0.007 202870_s_at Hs.82906 gb:NM_001255.1 CDC20 (cell division cycle 20, S. cerevisiae, homolog) 5.9 0.01 0.021 203358_s_at Hs.77256 gb:NM_004456.1 enhancer of zeste (Drosophila) homolog 2 5.9 0.004 0.003 202766_s_at Hs.750 gb:NM_000138.1 fibrillin 1 (Marfan syndrome) 5.9 0.001 0.005 205483_s_at Hs.833 gb:NM_005101.1 interferon-stimulated protein, 15 kDa 5.9 0.014 0.004 209351_at Hs.117729 gb:BC002690.1 keratin 14 (epidermolysis bullosa simplex, Dowling-Meara, Koebner) 5.9 0.036 0.031 204641_at Hs.153704 gb:NM_002497.1 NIMA (never in mitosis gene a)-related kinase 2 5.9 0.013 0.016 200872_at Hs.119301 gb:NM_002966.1 S100 calcium-binding protein A10 (annexin II ligand, calpactin I, light polypeptide (p11)) 5.9 0.001 0.001 219229_at Hs.14805 gb:NM_013272.2 solute carrier family 21 (organic anion transporter), member 11 5.9 0 0.003 201666_at Hs.5831 gb:NM_003254.1 tissue inhibitor of metalloproteinase 1 (erythroid potentiating activity, collagenase inhibitor) 5.9 0.006 0.009 210643_at Hs.115770 gb:AF053712.1 tumor necrosis factor (ligand) superfamily, member 11 5.9 0.001 0.009 205534_at Hs.34073 gb:NM_002589.1 BH-protocadherin (brain-heart) 5.8 0.003 0.007 220356_at Hs.62794 gb:NM_006587.1 corin 5.8 0.01 0.008 201925_s_at Hs.1369 gb:NM_000574.1 decay accelerating factor for complement (CD55, Cromer blood group system) 5.8 0.004 0.005 203840_at Hs.158205 gb:NM_003666.1 basic leucine zipper nuclear factor 1 (JEM-1) 5.7 0.008 0.004 210607_at Hs.428 gb:U03858.1 fms-related tyrosine kinase 3 ligand 5.7 0.029 0.008 207191_s_at Hs.102171 gb:NM_005545.1 immunoglobulin superfamily containing leucine-rich repeat 5.7 0.005 0.001 213230_at Hs.78358 gb:AI422335_RC paraneoplastic antigen 5.7 0.001 0.002 208510_s_at Hs.100724 gb:NM_015869.1 peroxisome proliferative activated receptor, gamma 5.7 0.007 0.016 209453_at Hs.170222 gb:M81768.1 solute carrier family 9 (sodiumhydrogen exchanger), isoform 1 (antiporter, Na+H+, amiloride sensitive) 5.7 0.003 0.001 215235_at Hs.77196 gb:AL110273.1 spectrin, alpha, non-erythrocytic 1 (alpha-fodrin) 5.7 0 0.001 214066_x_at Hs.256747 gb:AA565715_RC sperm associated antigen 8 5.7 0.005 0.007 201792_at Hs.118397 gb:NM_001129.2 AE-binding protein 1 5.6 0 0.001 226618_at Hs.75893 gb:AW572911_RC ankyrin 3, node of Ranvier (ankyrin G) 5.6 0.036 0.012 204416_x_at Hs.268571 gb:NM_001645.2 apolipoprotein C-I 5.6 0.008 0.009 204446_s_at Hs.89499 gb:NM_000698.1 arachidonate 5-lipoxygenase 5.6 0.004 0.001 216942_s_at Hs.75626 gb:D28586.1 CD58 antigen, (lymphocyte function-associated antigen 3) 5.6 0.001 0.009 209396_s_at Hs.75184 gb:M80927.1 chitinase 3-like 1 (cartilage glycoprotein-39) 5.6 0.003 0.005 213110_s_at Hs.169825 gb:AW052179_RC collagen, type IV, alpha 5 (Alport syndrome) 5.6 0.011 0.012 203903_s_at Hs.31720 gb:NM_014799.1 hephaestin 5.6 0.005 0.001 205403_at Hs.25333 gb:NM_004633.1 interleukin 1 receptor, type II 5.6 0.014 0.006 206584_at Hs.69328 gb:NM_015364.1 MD-2 protein 5.6 0.048 0.021 207651_at Hs.159545 gb:NM_013308.1 platelet activating receptor homolog 5.6 0.036 0.021 208893_s_at Hs.180383 gb:BC005047.1 dual specificity phosphatase 6 5.5 0.038 0.021 217889_s_at Hs.31297 gb:NM_024843.1 duodenal cytochrome b 5.5 0.005 0.004 202422_s_at Hs.81452 gb:NM_022977.1 fatty-acid-Coenzyme A ligase, long-chain 4 5.5 0.001 0.002 201141_at Hs.82226 gb:NM_002510.1 glycoprotein (transmembrane) nmb 5.5 0.005 0.009 205270_s_at Hs.2488 gb:NM_005565.2 lymphocyte cytosolic protein 2 (SH2 domain-containing leukocyte protein of 76kD) 5.5 0.013 0.004 212647_at Hs.9651 gb:NM_006270.1 related RAS viral (r-ras) oncogene homolog 5.5 0.024 0.027 230652_at Hs.77183 gb:AI760277_RC v-raf murine sarcoma 3611 viral oncogene homolog 1 5.5 0.01 0.003 214580_x_at Hs.321223 gb:AL569511_RC keratin 6B 5.4 0.031 0.021 212713_at Hs.296049 gb:R72286_RC microfibrillar-associated protein 4 5.4 0.01 0.016 208995_s_at Hs.77965 gb:U40763.1 peptidyl-prolyl isomerase G (cyclophilin G) 5.4 0.007 0.004 204083_s_at Hs.300772 gb:NM_003289.1 tropomyosin 2 (beta) 5.4 0.006 0.012 215125_s_at Hs.2056 gb:AV691323 UDP glycosyltransferase 1 family, polypeptide A9 5.4 0 0.003 203559_s_at Hs.75741 gb:NM_001091.1 amiloride binding protein 1 (amine oxidase (copper-containing)) 5.3 0.033 0.007 201438_at Hs.80988 gb:NM_004369.1 collagen, type VI, alpha 3 5.3 0.002 0.004 209138_x_at Hs.181125 gb:M87790.1 immunoglobulin lambda locus 5.3 0.003 0.009 219884_at Hs.103137 gb:NM_014368.1 LIM homeobox protein 6 5.3 0.005 0.001 205407_at Hs.29640 gb:NM_021111.1 reversion-inducing-cysteine-rich protein with kazal motifs 5.3 0.007 0.012 210612_s_at Hs.61289 gb:AF318616.1 synaptojanin 2 5.3 0.029 0.012 205513_at Hs.2012 gb:NM_001062.1 transcobalamin I (vitamin B12 binding protein, R binder family) 5.3 0.036 0.007 204137_at Hs.15791 gb:NM_003272.1 transmembrane 7 superfamily member 1 (upregulated in kidney) 5.3 0.001 0.005 219725_at Hs.44234 gb:NM_018965.1 triggering receptor expressed on myeloid cells 2 5.3 0.008 0.007 205146_x_at Hs.17528 gb:NM_004886.1 amyloid beta (A4) precursor protein-binding, family A, member 3 (X11-like 2) 5.2 0.032 0.007
Page 4 of 17 Upregulated transcripts
Affy Code UniGene Ace GenBank Acession Name Fold Change t test p value MW U test p value 218900_at Hs.175043 gb:NM_020184.1 ancient conserved domain protein 4 5.2 0.005 0.027 204345_at Hs.26208 gb:NM_001856.1 collagen, type XVI, alpha 1 5.2 0.007 0.016 209893_s_at Hs.2173 gb:M58596.1 fucosyltransferase 4 (alpha (1,3) fucosyltransferase, myeloid-specific) 5.2 0.016 0.044 222968_at Hs.109798 gb:NM_016947.1_RC G8 protein 5.2 0.009 0.005 216331_at Hs.74369 gb:AK022548.1 integrin, alpha 7 5.2 0.027 0.021 213680_at Hs.111758 gb:AI831452_RC keratin 6A 5.2 0.002 0.005 210139_s_at Hs.103724 gb:L03203.1 peripheral myelin protein 22 5.2 0.002 0.016 206359_at Hs.296176 gb:BG035761 STAT induced STAT inhibitor 3 5.2 0.003 0.004 219682_s_at Hs.332150 gb:NM_016569.1 TBX3-iso protein 5.2 0.002 0.005 201012_at Hs.78225 gb:NM_000700.1 annexin A1 5.1 0.003 0.035 201261_x_at Hs.821 gb:BC002416.1 biglycan 5.1 0.001 0.001 211368_s_at Hs.2490 gb:U13700.1 caspase 1, apoptosis-related cysteine protease (interleukin 1, beta, convertase) 5.1 0.023 0.003 205308_at Hs.118821 gb:NM_016010.1 CGI-62 protein 5.1 0.004 0.007 222108_at Hs.121520 gb:AC004010 Human BAC clone GS1-99H8 5.1 0.017 0.009 214660_at Hs.116774 gb:X68742.1 integrin, alpha 1 5.1 0.015 0.008 204709_s_at Hs.270845 gb:NM_004856.3 kinesin-like 5 (mitotic kinesin-like protein 1) 5.1 0.004 0.003 201105_at Hs.227751 gb:NM_002305.2 lectin, galactoside-binding, soluble, 1 (galectin 1) 5.1 0.014 0.044 205668_at Hs.153563 gb:NM_002349.1 lymphocyte antigen 75 5.1 0.003 0.016 224225_s_at Hs.272398 gb:AF218365.1 transcription factor ets 5.1 0.002 0.003 209335_at Hs.76152 gb:AI281593_RC decorin 5 0.046 0.035 203490_at Hs.151139 gb:NM_001421.1 E74-like factor 4 (ets domain transcription factor) 5 0.008 0.001 204162_at Hs.58169 gb:NM_006101.1 highly expressed in cancer, rich in leucine heptad repeats 5 0.001 0.003 221756_at Hs.26670 gb:AL540260_RC Human PAC clone RP3-515N1 from 22q11.2-q22 5 0 0.002 205227_at Hs.173880 gb:NM_002182.1 interleukin 1 receptor accessory protein 5 0.017 0.005 203565_s_at Hs.82380 gb:NM_002431.1 menage a trois 1 (CAK assembly factor) 5 0.003 0.007 204766_s_at Hs.388 gb:NM_002452.1 nudix (nucleoside diphosphate linked moiety X)-type motif 1 5 0.03 0.005 204068_at Hs.166684 gb:NM_006281.1 serinethreonine kinase 3 (Ste20, yeast homolog) 5 0 0.001 208112_x_at Hs.155119 gb:NM_006795.1 EH domain containing 1 4.9 0.005 0.001 205107_s_at Hs.42331 gb:NM_005227.1 ephrin-A4 4.9 0.024 0.009 201497_x_at Hs.78344 gb:NM_022844.1 myosin, heavy polypeptide 11, smooth muscle 4.9 0.027 0.012 219773_at Hs.93847 gb:NM_016931.1 NADPH oxidase 4 4.9 0.003 0.005 200906_s_at Hs.194431 gb:AK025843.1 palladin 4.9 0.017 0.005 201954_at Hs.11538 gb:NM_005720.1 actin related protein 23 complex, subunit 1A (41 kD) 4.8 0.004 0.021 205239_at Hs.270833 gb:NM_001657.1 amphiregulin (schwannoma-derived growth factor) 4.8 0.001 0.002 204714_s_at Hs.30054 gb:NM_000130.2 coagulation factor V (proaccelerin, labile factor) 4.8 0.001 0.002 202668_at Hs.30942 gb:BF001670_RC ephrin-B2 4.8 0.004 0.001 231031_at Hs.164568 gb:AI761573_RC fibroblast growth factor 7 (keratinocyte growth factor) 4.8 0.021 0.035 210663_s_at Hs.169139 gb:BC000879.1 kynureninase (L-kynurenine hydrolase) 4.8 0.002 0.001 200923_at Hs.79339 gb:NM_005567.2 lectin, galactoside-binding, soluble, 3 binding protein (galectin 6 binding protein) 4.8 0.009 0.007 203708_at Hs.188 gb:NM_002600.1 phosphodiesterase 4B, cAMP-specific (dunce (Drosophila)-homolog phosphodiesterase E4) 4.8 0.008 0.007 205479_s_at Hs.77274 gb:NM_002658.1 plasminogen activator, urokinase 4.8 0.032 0.004 224021_at Hs.251687 gb:AF146592.1 retinitis pigmentosa 1 (autosomal dominant) 4.8 0.04 0.016 209954_x_at Hs.289105 gb:AF343880.1 synovial sarcoma, X breakpoint 2 4.8 0.025 0.005 209573_s_at Hs.153498 gb:AW008505_RC chromosome 18 open reading frame 1 4.7 0.005 0.002 203418_at Hs.85137 gb:NM_001237.1 cyclin A2 4.7 0.01 0.01 223490_s_at Hs.177677 gb:AF281132.1 exosome component Rrp40 4.7 0.034 0.035 202581_at Hs.274402 gb:NM_005346.2 heat shock 70kD protein 1B 4.7 0.024 0.004 204041_at Hs.82163 gb:NM_000898.1 monoamine oxidase B 4.7 0.007 0.035 207238_s_at Hs.170121 gb:NM_002838.1 protein tyrosine phosphatase, receptor type, C 4.7 0.006 0.005 219799_s_at Hs.179608 gb:NM_005771.1 retinol dehydrogenase homolog 4.7 0.04 0.016 219090_at Hs.12321 gb:NM_020689.2 sodium calcium exchanger 4.7 0.008 0.003 206025_s_at Hs.29352 gb:AW188198_RC tumor necrosis factor, alpha-induced protein 6 4.7 0.002 0.007 213934_s_at Hs.22182 gb:AL567808_RC zinc finger protein 23 (KOX 16) 4.7 0.002 0.005 214255_at Hs.44697 gb:N35112_RC ATPase, Class V, type 10C 4.6 0.011 0.005 202391_at Hs.79516 gb:NM_006317.1 brain abundant, membrane attached signal protein 1 4.6 0.001 0.005 201697_s_at Hs.77462 gb:NM_001379.1 DNA (cytosine-5-)-methyltransferase 1 4.6 0 0.002 219432_at Hs.274446 gb:NM_014556.1 Ellis van Creveld syndrome 4.6 0.005 0.007 201313_at Hs.146580 gb:NM_001975.1 enolase 2, (gamma, neuronal) 4.6 0.028 0.007 201088_at Hs.159557 gb:NM_002266.1 karyopherin alpha 2 (RAG cohort 1, importin alpha 1) 4.6 0.001 0.002
Page 5 of 17 Upregulated transcripts
Affy Code UniGene Ace GenBank Acession Name Fold Change t test p value MW U test p value 222848_at Hs.192843 gb:BC005400.1 leucine zipper protein FKSG14 4.6 0.025 0.007 212873_at Hs.196914 gb:BE349017_RC minor histocompatibility antigen HA-1 4.6 0 0.001 231867_at Hs.173560 gb:AB032953.1 odd Ozten-m homolog 2 (Drosophila, mouse) 4.6 0.013 0.007 227486_at Hs.153952 gb:AI086864_RC 5 nucleotidase (CD73) 4.5 0.012 0.012 231711_at Hs.1852 gb:BF592752_RC acid phosphatase, prostate 4.5 0.014 0.01 205738_s_at Hs.49881 gb:NM_004102.2 fatty acid binding protein 3, muscle and heart (mammary-derived growth inhibitor) 4.5 0.007 0.004 206334_at Hs.159177 gb:NM_004190.1 lipase, gastric 4.5 0.028 0.007 212885_at Hs.201676 gb:AL545921_RC M-phase phosphoprotein 10 (U3 small nucleolar ribonucleoprotein) 4.5 0.004 0.007 226673_at Hs.26054 gb:AW665063_RC novel SH2-containing protein 3 4.5 0.001 0.002 201037_at Hs.99910 gb:NM_002627.1 phosphofructokinase, platelet 4.5 0.013 0.027 218643_s_at Hs.39733 gb:NM_014171.1 postsynaptic protein CRIPT 4.5 0.013 0.012 227928_at Hs.2182 gb:AI224977_RC pro-melanin-concentrating hormone 4.5 0.01 0.005 219210_s_at Hs.321245 gb:NM_016530.1 RAB-8b protein 4.5 0.003 0.005 209051_s_at Hs.106185 gb:AF295773.1 ral guanine nucleotide dissociation stimulator 4.5 0.009 0.027 227998_at Hs.738 gb:AA045184_RC ribosomal protein L14 4.5 0.008 0.007 210135_s_at Hs.55967 gb:AF022654.1 short stature homeobox 2 4.5 0.019 0.027 200974_at Hs.195851 gb:NM_001613.1 actin, alpha 2, smooth muscle, aorta 4.4 0 0.001 209524_at Hs.127842 gb:AK001280.1 CGI-142 4.4 0.001 0.009 221058_s_at Hs.15159 gb:NM_016326.2 chemokine-like factor, alternatively spliced 4.4 0.003 0.003 201431_s_at Hs.74566 gb:NM_001387.1 dihydropyrimidinase-like 3 4.4 0 0.005 212942_s_at Hs.50081 gb:AB033025.1 KIAA1199 protein 4.4 0.014 0.007 218729_at Hs.109276 gb:NM_020169.1 latexin protein 4.4 0.004 0.009 230134_s_at Hs.112227 gb:BF972355 membrane-associated nucleic acid binding protein 4.4 0.01 0.003 207202_s_at Hs.118138 gb:NM_003889.2 nuclear receptor subfamily 1, group I, member 2 4.4 0.032 0.035 204669_s_at Hs.30524 gb:NM_007219.2 ring finger protein 24 4.4 0.01 0.005 227480_at Hs.131819 gb:Z92546 Sushi domain (SCR repeat) containing 4.4 0.016 0.027 211796_s_at Hs.303157 gb:AF043179.1 T cell receptor beta locus 4.4 0.022 0.044 208998_at Hs.80658 gb:U94592.1 uncoupling protein 2 (mitochondrial, proton carrier) 4.4 0.001 0.002 203741_s_at Hs.172199 gb:NM_001114.1 adenylate cyclase 7 4.3 0.001 0.004 205715_at Hs.169998 gb:NM_004334.1 bone marrow stromal cell antigen 1 4.3 0.001 0.001 209083_at Hs.109606 gb:U34690.1 coronin, actin-binding protein, 1A 4.3 0.001 0.009 202196_s_at Hs.4909 gb:NM_013253.1 dickkopf (Xenopus laevis) homolog 3 4.3 0 0.001 207850_at Hs.89690 gb:NM_002090.1 GRO3 oncogene 4.3 0.02 0.005 200799_at Hs.8997 gb:NM_005345.3 heat shock 70kD protein 1A 4.3 0.029 0.012 221813_at Hs.62767 gb:AI129395_RC KIAA1332 protein 4.3 0.004 0.002 224937_at Hs.300591 gb:BF311866 KIAA1436 protein 4.3 0 0.001 225929_s_at Hs.17767 gb:AA233374_RC KIAA1554 protein 4.3 0.005 0.027 215446_s_at Hs.102267 gb:L16895 lysyl oxidase 4.3 0.003 0.012 201069_at Hs.111301 gb:NM_004530.1 matrix metalloproteinase 2 (gelatinase A, 72kD gelatinase, 72kD type IV collagenase) 4.3 0.004 0.012 210873_x_at Hs.226307 gb:U03891.2 phorbolin (similar to apolipoprotein B mRNA editing protein) 4.3 0.049 0.027 202620_s_at Hs.41270 gb:NM_000935.1 procollagen-lysine, 2-oxoglutarate 5-dioxygenase (lysine hydroxylase) 2 4.3 0.005 0.001 214404_x_at Hs.79414 gb:AI435670 prostate epithelium-specific Ets transcription factor 4.3 0.001 0.009 224940_s_at Hs.250655 gb:BF107618 prothymosin, alpha (gene sequence 28) 4.3 0.007 0.005 230405_at Hs.41587 gb:AI143416_RC RAD50 (S. cerevisiae) homolog 4.3 0.006 0.016 227850_x_at Hs.29759 gb:AW084544_RC RNA POLYMERASE I AND TRANSCRIPT RELEASE FACTOR 4.3 0.003 0.009 203186_s_at Hs.81256 gb:NM_002961.2 S100 calcium-binding protein A4 (calcium protein, calvasculin, metastasin, murine placental homolog) 4.3 0.015 0.007 209723_at Hs.104879 gb:BC002538.1 serine (or cysteine) proteinase inhibitor, clade B (ovalbumin), member 9 4.3 0.001 0.001 202205_at Hs.93183 gb:NM_003370.1 vasodilator-stimulated phosphoprotein 4.3 0.003 0.007 204908_s_at Hs.31210 gb:NM_005178.1 B-cell CLLlymphoma 3 4.2 0 0.001 200755_s_at Hs.7753 gb:BF939365_RC calumenin 4.2 0.013 0.027 209172_s_at Hs.77204 gb:U30872.1 centromere protein F (350400kD, mitosin) 4.2 0.023 0.007 211434_s_at Hs.302043 gb:AF015524.1 chemokine (C-C motif) receptor-like 2 4.2 0.022 0.027 205076_s_at Hs.166066 gb:NM_006697.1 cisplatin resistance associated 4.2 0.002 0.001 206224_at Hs.123114 gb:NM_001898.1 cystatin SN 4.2 0.01 0.031 210220_at Hs.81217 gb:L37882.1 frizzled (Drosophila) homolog 2 4.2 0.019 0.012 226554_at Hs.104640 gb:AW445134_RC HIV-1 inducer of short transcripts binding protein 4.2 0.012 0.004 226966_at Hs.33104 gb:BF108696_RC Huntingtin interacting protein C 4.2 0.018 0.012 212621_at Hs.14912 gb:AB006624.1 KIAA0286 protein 4.2 0.001 0.005 212792_at Hs.11217 gb:AB020684.1 KIAA0877 protein 4.2 0.006 0.012
Page 6 of 17 Upregulated transcripts
Affy Code UniGene Ace GenBank Acession Name Fold Change t test p value MW U test p value 220911_s_at Hs.288348 gb:NM_025081.1 KIAA1305 protein 4.2 0.005 0.009 222937_s_at Hs.231958 gb:AF219624.1 matrix metalloproteinase 28 4.2 0.002 0.002 218224_at Hs.194709 gb:NM_006029.2 paraneoplastic antigen MA1 4.2 0 0.001 203596_s_at Hs.27610 gb:NM_012420.1 retinoic acid- and interferon-inducible protein (58kD) 4.2 0.01 0.014 202856_s_at Hs.85838 gb:NM_004207.1 solute carrier family 16 (monocarboxylic acid transporters), member 3 4.2 0.003 0.021 201746_at Hs.1846 gb:NM_000546.2 tumor protein p53 (Li-Fraumeni syndrome) 4.2 0.005 0.009 206638_at Hs.2507 gb:NM_000867.1 5-hydroxytryptamine (serotonin) receptor 2B 4.1 0.024 0.021 222608_s_at Hs.62180 gb:AK023208.1 anillin (Drosophila Scraps homolog), actin binding protein 4.1 0.004 0.002 209708_at Hs.6909 gb:AY007239.1 DKFZP564G202 protein 4.1 0.003 0.012 224252_s_at Hs.294135 gb:AF177940.1 FXYD domain-containing ion transport regulator 5 4.1 0.011 0.005 218847_at Hs.30299 gb:NM_006548.1 IGF-II mRNA-binding protein 2 4.1 0 0.001 208083_s_at Hs.123125 gb:NM_000888.3 integrin, beta 6 4.1 0.006 0.027 212396_s_at Hs.154797 gb:AI143233_RC KIAA0090 protein 4.1 0.021 0.027 225978_at Hs.236463 gb:AW409794_RC KIAA1238 protein 4.1 0.001 0.012 202998_s_at Hs.83354 gb:NM_002318.1 lysyl oxidase-like 2 4.1 0.015 0.007 205415_s_at Hs.66521 gb:AI888099_RC Machado-Joseph disease (spinocerebellar ataxia 3, olivopontocerebellar ataxia 3, autosomal dominant, ataxin 3) 4.1 0.001 0.003 204706_at Hs.25156 gb:NM_019892.1 phosphatidylinositol (4,5) bisphosphate 5-phosphatase homolog; phosphatidylinositol polyphosphate 5-phosphatase type IV 4.1 0.002 0.001 218828_at Hs.103382 gb:NM_020360.1 phospholipid scramblase 3 4.1 0.001 0.001 211012_s_at Hs.89633 gb:BC000080.1 promyelocytic leukemia 4.1 0 0.004 205091_x_at Hs.235069 gb:NM_002907.1 RecQ protein-like (DNA helicase Q1-like) 4.1 0.012 0.012 201250_s_at Hs.169902 gb:NM_006516.1 solute carrier family 2 (facilitated glucose transporter), member 1 4.1 0.002 0.002 204653_at Hs.18387 gb:BF343007 transcription factor AP-2 alpha (activating enhancer-binding protein 2 alpha) 4.1 0.001 0.001 202664_at Hs.24143 gb:AW058622_RC Wiskott-Aldrich syndrome protein interacting protein 4.1 0.002 0.035 202207_at Hs.111554 gb:BG435404 ADP-ribosylation factor-like 7 4 0.009 0.012 210538_s_at Hs.127799 gb:U37546.1 baculoviral IAP repeat-containing 3 4 0.001 0.016 201170_s_at Hs.171825 gb:NM_003670.1 basic helix-loop-helix domain containing, class B, 2 4 0.015 0.027 205965_at Hs.41691 gb:NM_006399.1 basic leucine zipper transcription factor, ATF-like 4 0.006 0.012 203440_at Hs.161 gb:M34064.1 cadherin 2, type 1, N-cadherin (neuronal) 4 0.011 0.016 225646_at Hs.10029 gb:AI246687_RC cathepsin C 4 0.002 0.002 211964_at Hs.75617 gb:X05610.1 collagen, type IV, alpha 2 4 0 0.001 226751_at Hs.26358 gb:AW193693_RC DKFZP566K1924 protein 4 0.002 0.012 204057_at Hs.14453 gb:AI073984_RC interferon consensus sequence binding protein 1 4 0 0.005 214453_s_at Hs.82316 gb:NM_006417.1 interferon-induced, hepatitis C-associated microtubular aggregate protein (44kD) 4 0.013 0.007 213066_at Hs.26951 gb:AB002373.2 KIAA0375 gene product 4 0.001 0.007 225847_at Hs.22941 gb:AB037784.1 KIAA1363 protein 4 0.001 0.007 229554_at Hs.79914 gb:AI141861_RC lumican 4 0.018 0.035 228846_at Hs.109012 gb:AW071793_RC MAX dimerization protein 4 0.007 0.012 205101_at Hs.3076 gb:NM_000246.1 MHC class II transactivator 4 0.008 0.021 204748_at Hs.196384 gb:NM_000963.1 prostaglandin-endoperoxide synthase 2 (prostaglandin GH synthase and cyclooxygenase) 4 0.007 0.035 209879_at Hs.79283 gb:AI741056_RC selectin P ligand 4 0 0.001 226907_at Hs.192822 gb:N32557_RC serologically defined breast cancer antigen NY-BR-81 4 0.024 0.027 202691_at Hs.86948 gb:NM_006938.1 small nuclear ribonucleoprotein D1 polypeptide (16kD) 4 0.012 0.007 215926_x_at Hs.113265 gb:AK023513.1 small nuclear RNA activating complex, polypeptide 4, 190kD 4 0.005 0.021 218638_s_at Hs.288126 gb:NM_012445.1 spondin 2, extracellular matrix protein 4 0 0.001 201328_at Hs.85146 gb:AL575509_RC v-ets avian erythroblastosis virus E26 oncogene homolog 2 4 0 0.001 213906_at Hs.300592 gb:AW592266_RC v-myb avian myeloblastosis viral oncogene homolog-like 1 4 0.013 0.027 219386_s_at Hs.20450 gb:NM_020125.1 BCM-like membrane protein precursor 3.9 0.011 0.009 217188_s_at Hs.15106 gb:AC007182 chromosome 14 open reading frame 1 3.9 0.007 0.035 203469_s_at Hs.77313 gb:NM_003674.1 cyclin-dependent kinase (CDC2-like) 10 3.9 0.011 0.016 225785_at Hs.279789 gb:BG112359 histone deacetylase 3 3.9 0.003 0.005 224307_x_at Hs.49169 gb:AF213259.1 KIAA1634 protein 3.9 0.001 0.005 232558_at Hs.69954 gb:AL133595.1 laminin, gamma 3 3.9 0.01 0.044 204458_at Hs.6770 gb:AL110209.1 LCAT-like lysophospholipase 3.9 0.04 0.021 209894_at Hs.226627 gb:U50748.1 leptin receptor 3.9 0.025 0.035 210410_s_at Hs.112193 gb:AF034759.1 mutS (E. coli) homolog 5 3.9 0.003 0.003 204310_s_at Hs.78518 gb:NM_003995.2 natriuretic peptide receptor Bguanylate cyclase B (atrionatriuretic peptide receptor B) 3.9 0.016 0.009 207677_s_at Hs.196352 gb:NM_013416.1 neutrophil cytosolic factor 4 (40kD) 3.9 0.036 0.035 202007_at Hs.62041 gb:BF940043_RC nidogen (enactin) 3.9 0.01 0.039 220330_s_at Hs.24633 gb:NM_022136.1 SAM domain, SH3 domain and nuclear localisation signals, 1 3.9 0.008 0.027
Page 7 of 17 Upregulated transcripts
Affy Code UniGene Ace GenBank Acession Name Fold Change t test p value MW U test p value 200832_s_at Hs.119597 gb:AB032261.1 stearoyl-CoA desaturase (delta-9-desaturase) 3.9 0.021 0.035 216985_s_at Hs.82240 gb:AJ002077.1 syntaxin 3A 3.9 0.021 0.035 225655_at Hs.108106 gb:AK025578.1 transcription factor 3.9 0.024 0.035 229265_at Hs.2969 gb:AA927480_RC v-ski avian sarcoma viral oncogene homolog 3.9 0.001 0.003 208636_at Hs.119000 gb:AI082078_RC actinin, alpha 1 3.8 0 0.001 219087_at Hs.10760 gb:NM_017680.1 asporin (LRR class 1) 3.8 0.007 0.012 202504_at Hs.82237 gb:NM_012101.1 ataxia-telangiectasia group D-associated protein 3.8 0.015 0.044 209530_at Hs.250712 gb:U07139.1 calcium channel, voltage-dependent, beta 3 subunit 3.8 0 0.001 224314_s_at Hs.6523 gb:AF277174.1 chromosome 1 open reading frame 12 3.8 0.015 0.044 219482_at Hs.50748 gb:NM_017438.1 chromosome 21 open reading frame 18 3.8 0.007 0.021 227609_at Hs.180669 gb:AA633203_RC conserved gene amplified in osteosarcoma 3.8 0.007 0.001 226676_at Hs.26799 gb:AK021452.1 DKFZP564D0764 protein 3.8 0.008 0.012 203886_s_at Hs.198862 gb:NM_001998.1 fibulin 2 3.8 0.015 0.035 202949_s_at Hs.8302 gb:NM_001450.1 four and a half LIM domains 2 3.8 0 0.001 211067_s_at gb:BC006454.1 growth arrest-specific 7 3.8 0 0.003 202748_at Hs.171862 gb:NM_004120.2 guanylate binding protein 2, interferon-inducible 3.8 0 0.001 203089_s_at Hs.115721 gb:NM_013247.1 HtrA-like serine protease 3.8 0.013 0.027 215946_x_at Hs.296552 gb:AL022324 immunoglobulin lambda-like polypeptide 3 3.8 0.004 0.012 206295_at Hs.83077 gb:NM_001562.1 interleukin 18 (interferon-gamma-inducing factor) 3.8 0.006 0.012 225189_s_at Hs.42656 gb:AA194149_RC KIAA1681 protein 3.8 0.012 0.035 214175_x_at Hs.79691 gb:AI254547 LIM domain protein 3.8 0.019 0.003 205455_at Hs.2942 gb:NM_002447.1 macrophage stimulating 1 receptor (c-met-related tyrosine kinase) 3.8 0.005 0.021 205395_s_at Hs.20555 gb:NM_005590.1 meiotic recombination (S. cerevisiae) 11 homolog A 3.8 0.001 0.005 213847_at Hs.37044 gb:NM_006262.1 peripherin 3.8 0.015 0.002 203402_at Hs.298184 gb:AL520102_RC potassium voltage-gated channel, shaker-related subfamily, beta member 2 3.8 0.004 0.004 219654_at Hs.114062 gb:NM_014241.1 protein tyrosine phosphatase-like (proline instead of catalytic arginine), member a 3.8 0.009 0.007 220955_x_at Hs.94769 gb:NM_016277.1 RAB23, member RAS oncogene family 3.8 0.044 0.012 200750_s_at Hs.10842 gb:AF054183.1 RAN, member RAS oncogene family 3.8 0 0.001 221601_s_at Hs.58831 gb:AI084226_RC regulator of Fas-induced apoptosis 3.8 0.006 0.012 203687_at Hs.80420 gb:NM_002996.1 small inducible cytokine subfamily D (Cys-X3-Cys), member 1 (fractalkine, neurotactin) 3.8 0.033 0.035 208851_s_at Hs.125359 gb:AL161958.1 Thy-1 cell surface antigen 3.8 0 0.001 209118_s_at Hs.272897 gb:AF141347.1 Tubulin, alpha, brain-specific 3.8 0 0.001 202625_at Hs.80887 gb:AI356412_RC v-yes-1 Yamaguchi sarcoma viral related oncogene homolog 3.8 0 0.002 205181_at Hs.96448 gb:NM_006299.1 zinc finger protein 193 3.8 0.002 0.005 200625_s_at Hs.104125 gb:NM_006367.2 adenylyl cyclase-associated protein 3.7 0.004 0.012 209935_at Hs.106778 gb:AF225981.1 ATPase, Ca++ transporting, type 2C, member 1 3.7 0.002 0.004 205709_s_at Hs.152981 gb:NM_001263.1 CDP-diacylglycerol synthase (phosphatidate cytidylyltransferase) 1 3.7 0.005 0.01 210052_s_at Hs.9329 gb:AF098158.1 chromosome 20 open reading frame 1 3.7 0.004 0.012 204136_at Hs.1640 gb:NM_000094.1 collagen, type VII, alpha 1 (epidermolysis bullosa, dystrophic, dominant and recessive) 3.7 0.007 0.009 209474_s_at Hs.205353 gb:U87967.1 ectonucleoside triphosphate diphosphohydrolase 1 3.7 0.004 0.016 205659_at Hs.116753 gb:NM_014707.1 histone deacetylase 7B 3.7 0.004 0.009 213603_s_at Hs.301175 gb:BE138888_RC HSPC022 protein 3.7 0 0.001 205718_at Hs.1741 gb:NM_000889.1 integrin, beta 7 3.7 0.009 0.027 205422_s_at Hs.82582 gb:NM_004791.1 integrin, beta-like 1 (with EGF-like repeat domains) 3.7 0.026 0.035 204882_at Hs.1528 gb:NM_014882.1 KIAA0053 gene product 3.7 0.01 0.009 214684_at Hs.182280 gb:X63381.1 MADS box transcription enhancer factor 2, polypeptide A (myocyte enhancer factor 2A) 3.7 0 0.002 210004_at Hs.77729 gb:AF035776.1 oxidised low density lipoprotein (lectin-like) receptor 1 3.7 0.002 0.002 204198_s_at Hs.170019 gb:AA541630_RC runt-related transcription factor 3 3.7 0.001 0.004 209154_at Hs.12956 gb:AF234997.1 Tax interaction protein 1 3.7 0.001 0.005 213811_x_at Hs.101047 gb:AW062341_RC transcription factor 3 (E2A immunoglobulin enhancer binding factors E12E47) 3.7 0.001 0.012 203410_at Hs.77770 gb:NM_006803.1 adaptor-related protein complex 3, mu 2 subunit 3.6 0.003 0.027 204347_at Hs.274691 gb:AI653169_RC adenylate kinase 3 3.6 0.009 0.027 202685_s_at Hs.83341 gb:AI467916_RC AXL receptor tyrosine kinase 3.6 0.002 0.004 204378_at Hs.129057 gb:NM_003657.1 breast carcinoma amplified sequence 1 3.6 0.004 0.007 201458_s_at Hs.40323 gb:NM_004725.1 BUB3 (budding uninhibited by benzimidazoles 3, yeast) homolog 3.6 0 0.004 209619_at Hs.84298 gb:K01144.1 CD74 antigen (invariant polypeptide of major histocompatibility complex, class II antigen-associated) 3.6 0.001 0.012 203561_at Hs.78864 gb:NM_021642.1 Fc fragment of IgG, low affinity IIa, receptor for (CD32) 3.6 0.024 0.005 202489_s_at Hs.301350 gb:BC005238.1 FXYD domain-containing ion transport regulator 3 3.6 0.009 0.044 205240_at Hs.278338 gb:NM_013296.1 LGN protein 3.6 0.026 0.027
Page 8 of 17 Upregulated transcripts
Affy Code UniGene Ace GenBank Acession Name Fold Change t test p value MW U test p value 221552_at Hs.20220 gb:BC001698.1 lipase protein 3.6 0.002 0.007 226289_at Hs.278672 gb:BE551139_RC membrane component, chromosome 11, surface marker 1 3.6 0.004 0.016 202864_s_at Hs.77617 gb:NM_003113.1 nuclear antigen Sp100 3.6 0.012 0.044 214179_s_at Hs.83469 gb:H93013_RC nuclear factor (erythroid-derived 2)-like 1 3.6 0.004 0.005 213028_at Hs.282441 gb:AI887378_RC nuclear factor related to kappa B binding protein 3.6 0.001 0.007 206288_at Hs.766 gb:NM_005023.1 protein geranylgeranyltransferase type I, beta subunit 3.6 0.002 0.012 206407_s_at Hs.11383 gb:NM_005408.1 small inducible cytokine subfamily A (Cys-Cys), member 13 3.6 0.017 0.009 218854_at Hs.58636 gb:NM_013352.1 squamous cell carcinoma antigen recognized by T cell 3.6 0.016 0.007 210882_s_at Hs.259802 gb:U04811.1 trophinin 3.6 0.018 0.021 227907_at Hs.170263 gb:BF060782_RC tumor protein p53-binding protein, 1 3.6 0.001 0.004 224184_s_at Hs.153961 gb:AY027658.1 ARP1 (actin-related protein 1, yeast) homolog A (centractin alpha) 3.5 0.004 0.007 211890_x_at Hs.40300 gb:AF127765.3 calpain 3, (p94) 3.5 0.037 0.035 211980_at Hs.119129 gb:AI922605_RC collagen, type IV, alpha 1 3.5 0.002 0.012 208488_s_at Hs.193716 gb:NM_000651.1 complement component (3b4b) receptor 1, including Knops blood group system 3.5 0.023 0.016 213923_at Hs.155218 gb:AW005535_RC E1B-55kDa-associated protein 5 3.5 0 0.001 219612_s_at Hs.75431 gb:NM_000509.3 fibrinogen, gamma polypeptide 3.5 0.017 0.027 205505_at Hs.159642 gb:NM_001490.1 glucosaminyl (N-acetyl) transferase 1, core 2 (beta-1,6-N-acetylglucosaminyltransferase) 3.5 0.008 0.007 219179_at Hs.48950 gb:NM_016651.2 heptacellular carcinoma novel gene-3 protein 3.5 0.003 0.012 213910_at Hs.119206 gb:AW770896_RC insulin-like growth factor binding protein 7 3.5 0.003 0.009 201163_s_at Hs.119206 gb:NM_001553.1 insulin-like growth factor binding protein 7 3.5 0.018 0.016 208436_s_at Hs.166120 gb:NM_004030.1 interferon regulatory factor 7 3.5 0.008 0.044 200791_s_at Hs.1742 gb:NM_003870.1 IQ motif containing GTPase activating protein 1 3.5 0.001 0.009 213204_at Hs.117177 gb:AB014608.1 KIAA0708 protein 3.5 0.027 0.021 203044_at Hs.110488 gb:NM_014918.1 KIAA0990 protein 3.5 0 0.002 227189_at Hs.285714 gb:AB046819.1 KIAA1599 protein 3.5 0.003 0.007 209211_at Hs.84728 gb:AF132818.1 Kruppel-like factor 5 (intestinal) 3.5 0.008 0.021 205330_at Hs.268515 gb:NM_002430.1 meningioma (disrupted in balanced translocation) 1 3.5 0.003 0.009 201318_s_at Hs.233936 gb:NM_006471.1 myosin, light polypeptide, regulatory, non-sarcomeric (20kD) 3.5 0.001 0.007 209121_x_at Hs.288869 gb:M64497.1 nuclear receptor subfamily 2, group F, member 2 3.5 0.006 0.027 201310_s_at Hs.142827 gb:NM_004772.1 P311 protein 3.5 0.002 0.009 203131_at Hs.74615 gb:NM_006206.1 platelet-derived growth factor receptor, alpha polypeptide 3.5 0.022 0.016 205372_at Hs.14968 gb:NM_002655.1 pleiomorphic adenoma gene 1 3.5 0.013 0.021 215354_s_at Hs.274149 gb:BC002875.1 proline and glutamic acid rich nuclear protein 3.5 0.029 0.007 204279_at Hs.9280 gb:NM_002800.1 proteasome (prosome, macropain) subunit, beta type, 9 (large multifunctional protease 2) 3.5 0.004 0.007 202964_s_at Hs.166891 gb:NM_000449.1 regulatory factor X, 5 (influences HLA class II expression) 3.5 0.024 0.007 205943_at Hs.183671 gb:NM_005651.1 tryptophan 2,3-dioxygenase 3.5 0.002 0.004 203476_at Hs.82128 gb:NM_006670.1 5T4 oncofetal trophoblast glycoprotein 3.4 0 0.005 202820_at Hs.170087 gb:NM_001621.2 aryl hydrocarbon receptor 3.4 0 0.002 226292_at Hs.6133 gb:BF195709_RC calpain 5 3.4 0.013 0.012 200021_at Hs.180370 gb:NM_005507.1 cofilin 1 (non-muscle) 3.4 0.008 0.009 201201_at Hs.695 gb:NM_000100.1 cystatin B (stefin B) 3.4 0.001 0.001 212358_at Hs.7357 gb:AL117468.1 DKFZP586N1922 protein 3.4 0.044 0.004 204464_s_at Hs.76252 gb:NM_001957.1 endothelin receptor type A 3.4 0.005 0.007 208788_at Hs.250175 gb:AL136939.1 homolog of yeast long chain polyunsaturated fatty acid elongation enzyme 2 3.4 0.003 0.027 201904_s_at Hs.147189 gb:BF031714 HYA22 protein 3.4 0.035 0.035 201601_x_at Hs.146360 gb:NM_003641.1 interferon induced transmembrane protein 1 (9-27) 3.4 0.002 0.004 206332_s_at Hs.155530 gb:NM_005531.1 interferon, gamma-inducible protein 16 3.4 0.001 0.001 200650_s_at Hs.2795 gb:NM_005566.1 lactate dehydrogenase A 3.4 0.003 0.002 206343_s_at Hs.172816 gb:NM_013959.1 neuregulin 1 3.4 0.041 0.021 202052_s_at Hs.15165 gb:NM_015577.1 novel retinal pigment epithelial gene 3.4 0.002 0.027 203554_x_at Hs.252587 gb:NM_004219.2 pituitary tumor-transforming 1 3.4 0 0.001 225871_at Hs.118258 gb:BF680588 prostate cancer associated protein 1 3.4 0.006 0.035 221044_s_at Hs.125300 gb:NM_021616.1 ring finger protein 21, interferon-responsive 3.4 0.001 0.003 213194_at Hs.301198 gb:BF059159_RC roundabout (axon guidance receptor, Drosophila) homolog 1 3.4 0.003 0.044 203021_at Hs.251754 gb:NM_003064.1 secretory leukocyte protease inhibitor (antileukoproteinase) 3.4 0.028 0.012 203455_s_at Hs.28491 gb:NM_002970.1 spermidinespermine N1-acetyltransferase 3.4 0 0.001 230313_at Hs.274701 gb:AA524412_RC thymidine kinase 2, mitochondrial 3.4 0.005 0.005 218918_at Hs.8910 gb:NM_020379.1 1,2-alpha-mannosidase IC 3.3 0.01 0.035 202464_s_at Hs.195471 gb:NM_004566.1 6-phosphofructo-2-kinasefructose-2,6-biphosphatase 3 3.3 0.003 0.016
Page 9 of 17 Upregulated transcripts
Affy Code UniGene Ace GenBank Acession Name Fold Change t test p value MW U test p value 203008_x_at Hs.153884 gb:NM_005783.1 ATP binding protein associated with cell differentiation 3.3 0.001 0.009 221530_s_at Hs.33829 gb:BE857425_RC bHLH protein DEC2 3.3 0.005 0.035 213857_s_at Hs.82685 gb:BG230614_RC CD47 antigen (Rh-related antigen, integrin-associated signal transducer) 3.3 0 0.001 203100_s_at Hs.16081 gb:NM_004824.1 chromodomain protein, Y chromosome-like 3.3 0.001 0.006 219947_at Hs.115515 gb:NM_016184.1 C-type (calcium dependent, carbohydrate-recognition domain) lectin, superfamily member 6 3.3 0.003 0.012 208896_at Hs.100555 gb:X98743.1 DEADH (Asp-Glu-Ala-AspHis) box polypeptide 18 (Myc-regulated) 3.3 0 0.003 218567_x_at Hs.22880 gb:NM_005700.1 dipeptidylpeptidase III 3.3 0.002 0.009 213199_at Hs.6285 gb:AL080220.1 DKFZP586P0123 protein 3.3 0 0.009 202968_s_at Hs.173135 gb:Y09216.1 dual-specificity tyrosine-(Y)-phosphorylation regulated kinase 2 3.3 0 0.005 214088_s_at Hs.169238 gb:AW080549_RC fucosyltransferase 3 (galactoside 3(4)-L-fucosyltransferase, Lewis blood group included) 3.3 0.009 0.021 214510_at Hs.188859 gb:NM_005293.1 G protein-coupled receptor 20 3.3 0.018 0.016 201667_at Hs.74471 gb:NM_000165.2 gap junction protein, alpha 1, 43kD (connexin 43) 3.3 0.004 0.035 224209_s_at Hs.239147 gb:AF019638.1 guanine deaminase 3.3 0.023 0.007 226142_at Hs.154762 gb:AV682252 HIV-1 rev binding protein 2 3.3 0.008 0.044 213537_at Hs.914 gb:AI128225_RC Human mRNA for SB classII histocompatibility antigen alpha-chain 3.3 0 0.001 211990_at Hs.914 gb:M27487.1 Human mRNA for SB classII histocompatibility antigen alpha-chain 3.3 0.018 0.021 212390_at Hs.129928 gb:AB007923.1 KIAA0477 gene product 3.3 0.002 0.007 204589_at Hs.200598 gb:NM_014840.1 KIAA0537 gene product 3.3 0.001 0.016 217118_s_at Hs.13255 gb:AK025608.1 KIAA0930 protein 3.3 0 0.002 203766_s_at Hs.79386 gb:NM_012134.1 leiomodin 1 (smooth muscle) 3.3 0.013 0.016 204670_x_at Hs.308026 gb:NM_002125.1 major histocompatibility complex, class II, DR beta 5 3.3 0.003 0.007 210210_at Hs.287832 gb:AF181660.1 myelin protein zero-like 1 3.3 0.039 0.016 218589_at Hs.189999 gb:NM_005767.1 purinergic receptor (family A group 5) 3.3 0.001 0.001 203567_s_at Hs.59545 gb:AU157590_RC ring finger protein 15 3.3 0.002 0.007 204655_at Hs.241392 gb:NM_002985.1 small inducible cytokine A5 (RANTES) 3.3 0.01 0.001 219511_s_at Hs.24948 gb:NM_005460.1 synuclein, alpha interacting protein (synphilin) 3.3 0.001 0.007 219728_at Hs.84665 gb:NM_006790.1 titin immunoglobulin domain protein (myotilin) 3.3 0.038 0.009 223502_s_at Hs.270737 gb:AF134715.1 tumor necrosis factor (ligand) superfamily, member 13b 3.3 0.002 0.005 208743_s_at Hs.279920 gb:BC001359.1 tyrosine 3-monooxygenasetryptophan 5-monooxygenase activation protein, beta polypeptide 3.3 0.009 0.012 206796_at Hs.194680 gb:NM_003882.1 WNT1 inducible signaling pathway protein 1 3.3 0.009 0.007 214454_at Hs.120330 gb:NM_014244.1 a disintegrin-like and metalloprotease (reprolysin type) with thrombospondin type 1 motif, 2 3.2 0.006 0.021 211241_at Hs.166072 gb:M62895.1 annexin A2 pseudogene 2 3.2 0.001 0.004 211758_x_at gb:BC005968.1 ATP binding protein associated with celldifferentiation 3.2 0 0.005 219161_s_at Hs.15159 gb:NM_016951.2 chemokine-like factor, alternatively spliced 3.2 0.003 0.004 203979_at Hs.82568 gb:NM_000784.1 cytochrome P450, subfamily XXVIIA (steroid 27-hydroxylase, cerebrotendinous xanthomatosis), polypeptide 1 3.2 0.033 0.044 214581_x_at Hs.159651 gb:BE568134 death receptor 6 3.2 0.016 0.027 215537_x_at Hs.247362 gb:AJ012008 dimethylarginine dimethylaminohydrolase 2 3.2 0.016 0.035 205249_at Hs.1395 gb:NM_000399.2 early growth response 2 (Krox-20 (Drosophila) homolog) 3.2 0.008 0.035 203925_at Hs.89709 gb:NM_002061.1 glutamate-cysteine ligase, modifier subunit 3.2 0.002 0.004 202478_at Hs.155418 gb:NM_021643.1 GS3955 protein 3.2 0.004 0.035 205349_at Hs.73797 gb:NM_002068.1 guanine nucleotide binding protein (G protein), alpha 15 (Gq class) 3.2 0.007 0.021 211675_s_at gb:AF054589.1 HIC protein isoform p32; HIC protein isoform p40 3.2 0.007 0.021 205579_at Hs.1570 gb:NM_000861.2 histamine receptor H1 3.2 0.007 0.007 231936_at Hs.40408 gb:AK000445.1 homeo box C9 3.2 0.009 0.007 203233_at Hs.75545 gb:NM_000418.1 interleukin 4 receptor 3.2 0.002 0.016 207719_x_at Hs.25132 gb:NM_014812.1 KIAA0470 gene product 3.2 0.001 0.004 213118_at Hs.153293 gb:AL136821.1 KIAA0701 protein 3.2 0.007 0.021 225731_at Hs.28783 gb:BF196876_RC KIAA1223 protein 3.2 0.01 0.005 205928_at Hs.142150 gb:NM_005815.1 Kruppel-type zinc finger (C2H2) 3.2 0.012 0.024 222830_at Hs.321264 gb:BE566136 LBP protein 32 3.2 0.027 0.044 200644_at Hs.75061 gb:NM_023009.1 macrophage myristoylated alanine-rich C kinase substrate 3.2 0.005 0.027 211499_s_at Hs.57732 gb:U92268.1 mitogen-activated protein kinase 11 3.2 0.017 0.012 216602_s_at Hs.23111 gb:AD000092 phenylalanine-tRNA synthetase-like 3.2 0.044 0.012 204939_s_at Hs.85050 gb:NM_002667.1 phospholamban 3.2 0.019 0.035 206214_at Hs.93304 gb:NM_005084.1 phospholipase A2, group VII (platelet-activating factor acetylhydrolase, plasma) 3.2 0.001 0.007 205228_at Hs.20938 gb:NM_002898.1 RNA binding motif, single stranded interacting protein 2 3.2 0.025 0.027 204404_at Hs.110736 gb:NM_001046.1 solute carrier family 12 (sodiumpotassiumchloride transporters), member 2 3.2 0.013 0.003 218404_at Hs.106260 gb:NM_013322.1 sorting nexin 10 3.2 0.011 0.009 222749_at Hs.21431 gb:AF159447.1 suppressor of fused 3.2 0.002 0.021
Page 10 of 17 Upregulated transcripts
Affy Code UniGene Ace GenBank Acession Name Fold Change t test p value MW U test p value 224863_at Hs.5437 gb:BF477658_RC Tax1 (human T-cell leukemia virus type I) binding protein 1 3.2 0.017 0.039 201506_at Hs.118787 gb:NM_000358.1 transforming growth factor, beta-induced, 68kD 3.2 0.044 0.035 223077_at Hs.22826 gb:AW576360_RC tropomodulin 3 (ubiquitous) 3.2 0.001 0.001 201090_x_at Hs.278242 gb:NM_006082.1 tubulin, alpha, ubiquitous 3.2 0 0.001 205020_s_at Hs.201672 gb:NM_005738.1 ADP-ribosylation factor-like 4 3.1 0.003 0.021 213592_at Hs.9305 gb:X89271.1 angiotensin receptor-like 1 3.1 0.003 0.014 205263_at Hs.193516 gb:AF082283.1 B-cell CLLlymphoma 10 3.1 0.007 0.003 220988_s_at gb:NM_030945.1 complement-c1q tumor necrosis factor-relatedprotein; likely ortholog of mouse CORS26 (collagenousrepeat-containing sequ 3.1 0.011 0.021 206414_s_at Hs.12802 gb:NM_003887.1 development and differentiation enhancing factor 2 3.1 0 0.001 201468_s_at Hs.80706 gb:NM_000903.1 diaphorase (NADHNADPH) (cytochrome b-5 reductase) 3.1 0.005 0.001 202609_at Hs.2132 gb:NM_004447.1 epidermal growth factor receptor pathway substrate 8 3.1 0.003 0.035 208782_at Hs.296267 gb:BC000055.1 follistatin-like 1 3.1 0.007 0.003 205582_s_at Hs.1675 gb:NM_004121.1 gamma-glutamyltransferase-like activity 1 3.1 0.009 0.005 213416_at Hs.40034 gb:BG532690 integrin, alpha 4 (antigen CD49D, alpha 4 subunit of VLA-4 receptor) 3.1 0.004 0.005 201125_s_at Hs.149846 gb:NM_002213.1 integrin, beta 5 3.1 0.001 0.002 212975_at Hs.18166 gb:AB020677.2 KIAA0870 protein 3.1 0.002 0.012 229245_at Hs.241161 gb:AA535361_RC KIAA0969 protein 3.1 0.002 0.009 223393_s_at Hs.278436 gb:AL136805.1 KIAA1474 protein 3.1 0.002 0.007 225188_at Hs.42656 gb:AA194149_RC KIAA1681 protein 3.1 0 0.005 201720_s_at Hs.79356 gb:AI589086_RC Lysosomal-associated multispanning membrane protein-5 3.1 0.001 0.001 211700_s_at gb:AF349719.1 magphinin alpha 3.1 0.023 0.035 221875_x_at Hs.110309 gb:AW514210_RC major histocompatibility complex, class I, F 3.1 0.005 0.016 212012_at Hs.118893 gb:BF342851 Melanoma associated gene 3.1 0 0.001 203414_at Hs.79889 gb:NM_012329.1 monocyte to macrophage differentiation-associated 3.1 0 0.002 200738_s_at Hs.78771 gb:NM_000291.1 phosphoglycerate kinase 1 3.1 0.006 0.005 214435_x_at Hs.288757 gb:NM_005402.1 v-ral simian leukemia viral oncogene homolog A (ras related) 3.1 0.001 0.002 219614_s_at Hs.107854 gb:NM_020208.1 X transporter protein 3 3.1 0.004 0.012 209765_at Hs.278679 gb:Y13786.2 a disintegrin and metalloproteinase domain 19 (meltrin beta) 3 0.003 0.004 219099_at Hs.24792 gb:NM_020375.1 chromosome 12 open reading frame 5 3 0.001 0.001 208747_s_at Hs.169756 gb:M18767.1 complement component 1, s subcomponent 3 0.004 0.005 202141_s_at Hs.75193 gb:BC003090.1 COP9 homolog 3 0.003 0.001 222409_at Hs.17377 gb:AL162070.1 coronin, actin-binding protein, 1C 3 0.007 0.004 223454_at Hs.82407 gb:AF275260.1 CXC chemokine ligand 16 3 0.001 0.001 234863_x_at Hs.272027 gb:AK026197.1 F-box only protein 5 3 0.002 0.009 202270_at Hs.62661 gb:NM_002053.1 guanylate binding protein 1, interferon-inducible, 67kD 3 0.002 0.005 214328_s_at Hs.289088 gb:R01140_RC heat shock 90kD protein 1, alpha 3 0.011 0.035 203284_s_at Hs.169939 gb:AW151887_RC heparan sulfate 2-O-sulfotransferase 3 0.002 0.009 200697_at Hs.118625 gb:NM_000188.1 hexokinase 1 3 0 0.007 212646_at Hs.79123 gb:D42043.1 KIAA0084 protein 3 0.003 0.007 212538_at Hs.8021 gb:AL576253_RC KIAA1058 protein 3 0.003 0.016 227230_s_at Hs.205293 gb:BE855799_RC KIAA1211 protein 3 0.009 0.044 213932_x_at Hs.181244 gb:AI923492_RC major histocompatibility complex, class I, A 3 0.005 0.016 200600_at Hs.170328 gb:NM_002444.1 moesin 3 0.026 0.027 204078_at Hs.207251 gb:NM_006455.1 nucleolar autoantigen (55kD) similar to rat synaptonemal complex protein 3 0.002 0.016 212724_at Hs.6838 gb:BG054844_RC ras homolog gene family, member E 3 0.012 0.021 208540_x_at Hs.247697 gb:NM_021039.1 S100 calcium-binding protein A14 (calgizzarin) 3 0 0.001 224811_at Hs.5724 gb:BF112093_RC sclerostin 3 0.004 0.018 220054_at Hs.98309 gb:NM_016584.1 SGRF protein, Interleukin 23 p19 subunit 3 0.039 0.035 209370_s_at Hs.167679 gb:BE502377_RC SH3-domain binding protein 2 3 0 0.001 202363_at Hs.93029 gb:AF231124.1 sparcosteonectin, cwcv and kazal-like domains proteoglycan (testican) 3 0.011 0.009 215108_x_at Hs.110826 gb:U80736.1 trinucleotide repeat containing 9 3 0.006 0.016 219700_at Hs.125036 gb:NM_020405.1 tumor endothelial marker 7 precursor 3 0 0.001 220998_s_at gb:NM_030930.1 unc93 (C.elegans) homolog B 3 0.007 0.012 206219_s_at Hs.116237 gb:NM_005428.2 vav 1 oncogene 3 0 0.005 223709_s_at Hs.121540 gb:AY009400.1 wingless-type MMTV integration site family, member 10A 3 0.011 0.016 203585_at Hs.16622 gb:NM_007150.1 zinc finger protein 185 (LIM domain) 3 0.001 0.001 219540_at Hs.145498 gb:AU150728_RC zinc finger protein 267 3 0.001 0.002 227915_at Hs.182416 gb:AI872284_RC ankyrin repeat-containing protein ASB-2 2.9 0.048 0.035 204820_s_at Hs.167741 gb:NM_006994.2 butyrophilin, subfamily 3, member A3 2.9 0.002 0.035
Page 11 of 17 Upregulated transcripts
Affy Code UniGene Ace GenBank Acession Name Fold Change t test p value MW U test p value 204118_at Hs.901 gb:NM_001778.1 CD48 antigen (B-cell membrane protein) 2.9 0.004 0.009 213151_s_at Hs.184326 gb:AU157515_RC CDC10 (cell division cycle 10, S. cerevisiae, homolog) 2.9 0.009 0.009 212476_at Hs.24340 gb:D26069.1 centaurin beta2 2.9 0.002 0.005 223451_s_at Hs.15159 gb:AF096895.2 chemokine-like factor, alternatively spliced 2.9 0.006 0.005 220088_at Hs.2161 gb:NM_001736.1 complement component 5 receptor 1 (C5a ligand) 2.9 0.012 0.035 215000_s_at Hs.103419 gb:AL117593.1 fasciculation and elongation protein zeta 2 (zygin II) 2.9 0.004 0.007 203697_at Hs.153684 gb:U91903.1 frizzled-related protein 2.9 0.004 0.004 216651_s_at Hs.170808 gb:X69936.1 glutamate decarboxylase 2 (pancreatic islets and brain, 65kD) 2.9 0.048 0.016 219403_s_at Hs.44227 gb:NM_006665.1 heparanase 2.9 0.017 0.027 212262_at Hs.15020 gb:AA149639_RC homolog of mouse quaking QKI (KH domain RNA binding protein) 2.9 0.003 0.009 203914_x_at Hs.77348 gb:NM_000860.1 hydroxyprostaglandin dehydrogenase 15-(NAD) 2.9 0.002 0.002 200989_at Hs.197540 gb:NM_001530.1 hypoxia-inducible factor 1, alpha subunit (basic helix-loop-helix transcription factor) 2.9 0.003 0.021 205786_s_at Hs.172631 gb:NM_000632.2 integrin, alpha M (complement component receptor 3, alpha; also known as CD11b (p170), macrophage antigen alpha poly 2.9 0.006 0.021 212837_at Hs.82324 gb:D63877.1 KIAA0157 protein 2.9 0 0.003 202916_s_at Hs.5737 gb:NM_014864.1 KIAA0475 gene product 2.9 0.001 0.005 201776_s_at Hs.62515 gb:AK001487.1 KIAA0494 gene product 2.9 0 0.004 209627_s_at Hs.197955 gb:AY008372.1 KIAA0704 protein 2.9 0.005 0.007 227688_at Hs.65366 gb:AK022128.1 KIAA1495 protein 2.9 0.003 0.027 203637_s_at Hs.27695 gb:NM_000381.1 midline 1 (OpitzBBB syndrome) 2.9 0.013 0.035 219032_x_at Hs.279926 gb:NM_014322.1 opsin 3 (encephalopsin) 2.9 0.002 0.001 223854_at Hs.232500 gb:AF131761.1 protocadherin beta 10 2.9 0.038 0.035 209684_at Hs.62349 gb:AL136924.1 ras association (RalGDSAF-6) domain containing protein JC265 2.9 0.004 0.007 203749_s_at Hs.250505 gb:AI806984_RC retinoic acid receptor, alpha 2.9 0.007 0.009 201288_at Hs.83656 gb:NM_001175.1 Rho GDP dissociation inhibitor (GDI) beta 2.9 0.003 0.035 205542_at Hs.61635 gb:NM_012449.1 six transmembrane epithelial antigen of the prostate 2.9 0.002 0.009 206055_s_at Hs.80506 gb:NM_003090.1 small nuclear ribonucleoprotein polypeptide A 2.9 0.001 0.001 209146_at Hs.239926 gb:AV704962 sterol-C4-methyl oxidase-like 2.9 0.001 0.004 223686_at Hs.58715 gb:AB028138.1 thiamine pyrophosphokinase 2.9 0.018 0.044 201042_at Hs.8265 gb:AL031651 transglutaminase 2 (C polypeptide, protein-glutamine-gamma-glutamyltransferase) 2.9 0 0.001 202510_s_at Hs.101382 gb:NM_006291.1 tumor necrosis factor, alpha-induced protein 2 2.9 0.004 0.027 219684_at Hs.43388 gb:NM_022147.1 28kD interferon responsive protein 2.8 0.013 0.044 212186_at Hs.7232 gb:BE855983_RC acetyl-Coenzyme A carboxylase alpha 2.8 0.006 0.016 203935_at Hs.150402 gb:NM_001105.2 activin A receptor, type I 2.8 0.004 0.007 228323_at Hs.283099 gb:BF248364 AF15q14 protein 2.8 0.001 0.004 209939_x_at Hs.195175 gb:AF005775.1 CASP8 and FADD-like apoptosis regulator 2.8 0 0.001 224414_s_at gb:AF356193.1 caspase recruitment domain protein 6 2.8 0.003 0.021 201560_at Hs.25035 gb:NM_013943.1 chloride intracellular channel 4 2.8 0.004 0.003 213428_s_at Hs.108885 gb:AA292373_RC collagen, type VI, alpha 1 2.8 0.002 0.003 208159_x_at Hs.27424 gb:NM_004399.1 DEADH (Asp-Glu-Ala-AspHis) box polypeptide 11 (S.cerevisiae CHL1-like helicase) 2.8 0.02 0.044 217901_at Hs.2631 gb:BF031829 desmoglein 2 2.8 0.027 0.027 209457_at Hs.2128 gb:U16996.1 dual specificity phosphatase 5 2.8 0.006 0.012 217892_s_at Hs.10706 gb:NM_016357.1 epithelial protein lost in neoplasm beta 2.8 0.001 0.001 204420_at Hs.283565 gb:BG251266 FOS-like antigen-1 2.8 0.005 0.016 201415_at Hs.82327 gb:NM_000178.1 glutathione synthetase 2.8 0.001 0.003 215155_at Hs.166299 gb:J04178.1 Human abnormal beta-hexosaminidase alpha chain (HEXA) mRNA, partial cds 2.8 0.014 0.014 211991_s_at Hs.914 gb:M27487.1 Human mRNA for SB classII histocompatibility antigen alpha-chain 2.8 0.007 0.027 203820_s_at Hs.79440 gb:NM_006547.1 IGF-II mRNA-binding protein 3 2.8 0.024 0.035 211075_s_at gb:Z25521.1 integrin associated protein 2.8 0.002 0.005 202503_s_at Hs.81892 gb:NM_014736.1 KIAA0101 gene product 2.8 0.017 0.035 205223_at Hs.155987 gb:NM_014662.1 KIAA0645 gene product 2.8 0.01 0.009 203907_s_at Hs.4764 gb:NM_014869.1 KIAA0763 gene product 2.8 0.008 0.012 226716_at Hs.104417 gb:AB033031.1 KIAA1205 protein 2.8 0.021 0.027 231909_x_at Hs.71109 gb:AB033055.1 KIAA1229 protein 2.8 0.032 0.044 225056_at Hs.18760 gb:AB037810.1 KIAA1389 protein 2.8 0.011 0.035 219666_at Hs.17914 gb:NM_022349.1 membrane-spanning 4-domains, subfamily A, member6 2.8 0.019 0.035 203553_s_at Hs.246970 gb:NM_006575.1 mitogen-activated protein kinase kinase kinase kinase 5 2.8 0.005 0.016 209791_at Hs.33455 gb:AL049569 peptidyl arginine deiminase, type II 2.8 0.002 0.004 211084_x_at gb:Z25429.1 protein-serinethreonine kinase 2.8 0.033 0.035 221127_s_at Hs.278503 gb:NM_006394.1 regulated in glioma 2.8 0.008 0.021
Page 12 of 17 Upregulated transcripts
Affy Code UniGene Ace GenBank Acession Name Fold Change t test p value MW U test p value 210117_at Hs.153057 gb:AF311312.1 sperm associated antigen 1 2.8 0.009 0.027 201416_at Hs.83484 gb:BG528420 SRY (sex determining region Y)-box 4 2.8 0.002 0.005 202720_at Hs.165986 gb:NM_015641.1 testin 2.8 0 0.001 201147_s_at Hs.245188 gb:BF347089 tissue inhibitor of metalloproteinase 3 (Sorsby fundus dystrophy, pseudoinflammatory) 2.8 0.006 0.044 211058_x_at gb:BC006379.1 tubulin alpha 1 2.8 0 0.001 229555_at Hs.55968 gb:AI633503_RC UDP-N-acetyl-alpha-D-galactosamine:polypeptide N-acetylgalactosaminyltransferase 5 (GalNAc-T5) 2.8 0.018 0.035 205051_s_at Hs.81665 gb:NM_000222.1 v-kit Hardy-Zuckerman 4 feline sarcoma viral oncogene homolog 2.8 0.005 0.016 204453_at Hs.9450 gb:NM_003428.1 zinc finger protein 84 (HPF2) 2.8 0.014 0.004 228705_at Hs.182485 gb:BF197540_RC actinin, alpha 4 2.7 0.021 0.027 202666_s_at Hs.274350 gb:NM_004301.1 BAF53 2.7 0.002 0.012 228931_at Hs.20159 gb:AW628685_RC CGI-92 protein 2.7 0.026 0.035 214710_s_at Hs.23960 gb:BE407516 cyclin B1 2.7 0.003 0.012 221586_s_at Hs.2331 gb:U15642.1 E2F transcription factor 5, p130-binding 2.7 0.019 0.021 203780_at Hs.116651 gb:AF275945.1 epithelial V-like antigen 1 2.7 0.002 0.012 222843_at Hs.137516 gb:AK023411.1 fidgetin-like 1 2.7 0.012 0.027 221020_s_at gb:NM_030780.1 folate transportercarrier 2.7 0.001 0.009 203706_s_at Hs.173859 gb:NM_003507.1 frizzled (Drosophila) homolog 7 2.7 0 0.001 216973_s_at Hs.819 gb:S49765.1 homeo box B7 2.7 0 0.001 202771_at Hs.79077 gb:NM_014745.1 KIAA0233 gene product 2.7 0.014 0.044 213107_at Hs.170204 gb:R59093_RC KIAA0551 protein 2.7 0.003 0.003 224791_at Hs.10669 gb:AW513835_RC KIAA1249 protein 2.7 0.003 0.012 224790_at Hs.10669 gb:AI023398_RC KIAA1249 protein 2.7 0.039 0.027 226284_at Hs.24106 gb:BF111616_RC KIAA1483 protein 2.7 0.001 0.001 204682_at Hs.83337 gb:NM_000428.1 latent transforming growth factor beta binding protein 2 2.7 0.027 0.044 204790_at Hs.100602 gb:NM_005904.1 MAD (mothers against decapentaplegic, Drosophila) homolog 7 2.7 0.001 0.007 210224_at Hs.101840 gb:AF031469.1 major histocompatibility complex, class I-like sequence 2.7 0.001 0.007 205859_at Hs.184018 gb:NM_004271.1 MD-1, RP105-associated 2.7 0.003 0.007 220014_at Hs.157461 gb:NM_016644.1 mesenchymal stem cell protein DSC54 2.7 0.02 0.035 226101_at Hs.24379 gb:AI093546_RC MUM2 protein 2.7 0.01 0.021 226209_at Hs.59745 gb:AI129346_RC NADH dehydrogenase (ubiquinone) flavoprotein 3 (10kD) 2.7 0.025 0.035 215073_s_at Hs.288869 gb:AL554245 nuclear receptor subfamily 2, group F, member 2 2.7 0.036 0.035 208690_s_at Hs.75807 gb:BC000915.1 PDZ and LIM domain 1 (elfin) 2.7 0 0.001 212221_x_at Hs.303154 gb:AV703259 popeye protein 3 2.7 0.003 0.004 221840_at Hs.31137 gb:AA775177_RC protein tyrosine phosphatase, receptor type, E 2.7 0.003 0.012 214697_s_at Hs.145078 gb:AW190873_RC regulator of differentiation (in S. pombe) 1 2.7 0.003 0.012 213513_x_at Hs.252280 gb:BG034239 Rho guanine nucleotide exchange factor (GEF) 1 2.7 0 0.003 214853_s_at Hs.81972 gb:AI091079_RC SHC (Src homology 2 domain-containing) transforming protein 1 2.7 0 0.001 213168_at Hs.44450 gb:AU145005_RC Sp3 transcription factor 2.7 0 0.002 202043_s_at Hs.89718 gb:NM_004595.1 spermine synthase 2.7 0.001 0.002 209436_at Hs.5378 gb:AB018305.1 spondin 1, (f-spondin) extracellular matrix protein 2.7 0.023 0.027 202338_at Hs.105097 gb:NM_003258.1 thymidine kinase 1, soluble 2.7 0.028 0.027 204033_at Hs.6566 gb:NM_004237.1 thyroid hormone receptor interactor 13 2.7 0.007 0.016 215111_s_at Hs.114360 gb:AK027071.1 transforming growth factor beta-stimulated protein TSC-22 2.7 0.004 0.027 216623_x_at Hs.110826 gb:AK025084.1 trinucleotide repeat containing 9 2.7 0.008 0.021 221485_at Hs.107526 gb:AL035683 UDP-Gal:betaGlcNAc beta 1,4- galactosyltransferase, polypeptide 5 2.7 0.002 0.016 228785_at Hs.59757 gb:AA121673_RC zinc finger protein 281 2.7 0.038 0.007 209122_at Hs.3416 gb:BC005127.1 adipose differentiation-related protein 2.6 0.004 0.007 209621_s_at Hs.135281 gb:AF002280.1 alpha-actinin-2-associated LIM protein 2.6 0.006 0.016 218230_at Hs.301064 gb:AL044651_RC arfaptin 1 2.6 0.007 0.016 202686_s_at Hs.83341 gb:NM_021913.1 AXL receptor tyrosine kinase 2.6 0.017 0.044 221059_s_at Hs.157439 gb:NM_021615.1 carbohydrate (N-acetylglucosamine 6-O) sulfotransferase 6 2.6 0.016 0.039 213182_x_at Hs.106070 gb:R78668_RC cyclin-dependent kinase inhibitor 1C (p57, Kip2) 2.6 0.016 0.012 209250_at Hs.185973 gb:BC000961.2 degenerative spermatocyte (homolog Drosophila; lipid desaturase) 2.6 0 0.001 201061_s_at Hs.160483 gb:M81635.1 erythrocyte membrane protein band 7.2 (stomatin) 2.6 0 0.001 219079_at Hs.5741 gb:NM_016230.1 flavohemoprotein b5+b5R 2.6 0.003 0.007 228238_at Hs.289721 gb:AW105301_RC growth arrest specific transcript 5 2.6 0.03 0.005 208744_x_at Hs.36927 gb:BG403660 heat shock 105kD 2.6 0.04 0.027 210184_at Hs.51077 gb:M81695.1 integrin, alpha X (antigen CD11C (p150), alpha polypeptide) 2.6 0.008 0.031 209762_x_at Hs.38125 gb:AF280094.1 interferon-induced protein 75, 52kD 2.6 0 0.001
Page 13 of 17 Upregulated transcripts
Affy Code UniGene Ace GenBank Acession Name Fold Change t test p value MW U test p value 212657_s_at Hs.81134 gb:U65590 interleukin 1 receptor antagonist 2.6 0.001 0.004 212709_at Hs.22559 gb:D83781.1 KIAA0197 protein 2.6 0.005 0.009 208999_at Hs.80712 gb:D86957.1 KIAA0202 protein 2.6 0.006 0.012 213261_at Hs.16950 gb:AA035414_RC KIAA0342 gene product 2.6 0 0.007 209983_s_at Hs.124085 gb:AB035266.1 KIAA0921 protein 2.6 0.016 0.044 227231_at Hs.205293 gb:AI991996_RC KIAA1211 protein 2.6 0.002 0.007 234733_s_at Hs.62717 gb:AK001672.1 KIAA1596 protein 2.6 0.041 0.035 202202_s_at Hs.78672 gb:NM_002290.2 laminin, alpha 4 2.6 0 0.007 201153_s_at Hs.28578 gb:NM_021038.1 muscleblind (Drosophila)-like 2.6 0.002 0.012 201604_s_at Hs.16533 gb:NM_002480.1 myosin phosphatase, target subunit 1 2.6 0.011 0.012 201876_at Hs.169857 gb:NM_000305.1 paraoxonase 2 2.6 0.001 0.016 218319_at Hs.7886 gb:NM_020651.2 pellino (Drosophila) homolog 1 2.6 0.007 0.016 204517_at Hs.110364 gb:BE962749_RC peptidylprolyl isomerase C (cyclophilin C) 2.6 0.018 0.005 201274_at Hs.76913 gb:NM_002790.1 proteasome (prosome, macropain) subunit, alpha type, 5 2.6 0.031 0.035 202988_s_at Hs.75256 gb:NM_002922.1 regulator of G-protein signalling 1 2.6 0.006 0.012 230318_at Hs.297681 gb:T62088_RC serine (or cysteine) proteinase inhibitor, clade A (alpha-1 antiproteinase, antitrypsin), member 1 2.6 0.001 0.003 204542_at Hs.288215 gb:NM_006456.1 sialyltransferase 2.6 0.021 0.044 211744_s_at gb:BC005930.1 Similar to CD58 antigen, (lymphocytefunction-associated antigen 3) 2.6 0.005 0.012 227516_at Hs.288883 gb:AI655996_RC splicing factor 3a, subunit 1, 120kD 2.6 0.011 0.035 202644_s_at Hs.211600 gb:NM_006290.1 tumor necrosis factor, alpha-induced protein 3 2.6 0.004 0.005 225406_at Hs.247302 gb:AA195009_RC twisted gastrulation 2.6 0.001 0.002 204122_at Hs.9963 gb:NM_003332.1 TYRO protein tyrosine kinase binding protein 2.6 0 0.005 203343_at Hs.28309 gb:NM_003359.1 UDP-glucose dehydrogenase 2.6 0 0.001 221029_s_at gb:NM_030775.1 WNT5b protein 2.6 0.007 0.027 207753_at Hs.287374 gb:NM_020657.1 zinc finger protein 304 2.6 0.025 0.027 222838_at Hs.132906 gb:AL121985 19A24 protein 2.5 0.005 0.021 226694_at Hs.42322 gb:BG540494 A kinase (PRKA) anchor protein 2 2.5 0.003 0.012 205996_s_at Hs.171811 gb:NM_013411.1 adenylate kinase 2 2.5 0.005 0.009 203312_x_at Hs.89474 gb:NM_001663.2 ADP-ribosylation factor 6 2.5 0 0.001 212645_x_at Hs.80426 gb:AL566299_RC brain and reproductive organ-expressed (TNFRSF1A modulator) 2.5 0.002 0.005 201419_at Hs.106674 gb:NM_004656.1 BRCA1 associated protein-1 (ubiquitin carboxy-terminal hydrolase) 2.5 0.01 0.044 230204_at Hs.2799 gb:AU144114_RC cartilage linking protein 1 2.5 0.036 0.016 201743_at Hs.75627 gb:NM_000591.1 CD14 antigen 2.5 0.005 0.009 225719_s_at Hs.237924 gb:BG497783 CGI-69 protein 2.5 0.004 0.044 204971_at Hs.2621 gb:NM_005213.1 cystatin A (stefin A) 2.5 0.006 0.021 218435_at Hs.279884 gb:NM_013238.1 DNAJ domain-containing 2.5 0.001 0.003 218338_at Hs.305985 gb:NM_004426.1 early development regulator 1 (homolog of polyhomeotic 1) 2.5 0.022 0.035 225078_at Hs.111334 gb:AV686514 ferritin, light polypeptide 2.5 0.006 0.035 203988_s_at Hs.118722 gb:NM_004480.1 fucosyltransferase 8 (alpha (1,6) fucosyltransferase) 2.5 0.022 0.016 213237_at Hs.129061 gb:AI652058_RC Human Chromosome 16 BAC clone CIT987SK-A-101F10 2.5 0.042 0.035 202803_s_at Hs.83968 gb:NM_000211.1 integrin, beta 2 (antigen CD18 (p95), lymphocyte function-associated antigen 1; macrophage antigen 1 (mac-1) beta subun 2.5 0.002 0.012 201315_x_at Hs.174195 gb:NM_006435.1 interferon induced transmembrane protein 2 (1-8D) 2.5 0.001 0.016 221922_at Hs.93121 gb:AW195581_RC KIAA0761 protein 2.5 0.037 0.035 224796_at Hs.10669 gb:W03103_RC KIAA1249 protein 2.5 0.007 0.004 217933_s_at Hs.182579 gb:NM_015907.1 leucine aminopeptidase 2.5 0.005 0.027 211135_x_at Hs.105928 gb:AF009644.1 leukocyte immunoglobulin-like receptor, subfamily B (with TM and ITIM domains), member 3 2.5 0.003 0.006 204994_at Hs.926 gb:NM_002463.1 myxovirus (influenza) resistance 2, homolog of murine 2.5 0.005 0.035 218708_at Hs.24563 gb:NM_013248.1 NTF2-related export protein 1 2.5 0.001 0.007 201599_at Hs.75485 gb:NM_000274.1 ornithine aminotransferase (gyrate atrophy) 2.5 0.002 0.014 227307_at Hs.96908 gb:AL565381_RC p53-induced protein 2.5 0.034 0.035 208680_at Hs.180909 gb:L19184.1 peroxiredoxin 1 2.5 0.01 0.021 216883_x_at Hs.48291 gb:AJ001626.1 phosphodiesterase 6D, cGMP-specific, rod, delta 2.5 0.039 0.035 203879_at Hs.162808 gb:U86453.1 phosphoinositide-3-kinase, catalytic, delta polypeptide 2.5 0.005 0.021 204484_at Hs.132463 gb:NM_002646.1 phosphoinositide-3-kinase, class 2, beta polypeptide 2.5 0.006 0.021 217996_at Hs.82101 gb:AA576961_RC pleckstrin homology-like domain, family A, member 1 2.5 0.035 0.021 202429_s_at Hs.272458 gb:AL353950.1 protein phosphatase 3 (formerly 2B), catalytic subunit, alpha isoform (calcineurin A alpha) 2.5 0.015 0.001 223150_s_at Hs.25524 gb:AF290614.1 protein tyrosine phosphatase, non-receptor type 23 2.5 0.001 0.004 200887_s_at Hs.21486 gb:NM_007315.1 signal transducer and activator of transcription 1, 91kD 2.5 0.005 0.016 218757_s_at Hs.103832 gb:NM_023010.1 similar to yeast Upf3, variant B 2.5 0.003 0.016
Page 14 of 17 Upregulated transcripts
Affy Code UniGene Ace GenBank Acession Name Fold Change t test p value MW U test p value 202565_s_at Hs.154567 gb:NM_003174.2 supervillin 2.5 0 0.002 201999_s_at Hs.266940 gb:NM_006519.1 t-complex-associated-testis-expressed 1-like 1 2.5 0 0.001 217733_s_at Hs.76293 gb:NM_021103.1 thymosin, beta 10 2.5 0.007 0.009 212190_at Hs.21858 gb:AL541302 trinucleotide repeat containing 3 2.5 0.026 0.027 212481_s_at Hs.250641 gb:AI214061_RC tropomyosin 4 2.5 0.016 0.044 201336_at Hs.66708 gb:BC003570.1 vesicle-associated membrane protein 3 (cellubrevin) 2.5 0.001 0.009 203739_at Hs.155040 gb:NM_006526.1 zinc finger protein 217 2.5 0.023 0.012 221727_at Hs.74861 gb:AA456973_RC activated RNA polymerase II transcription cofactor 4 2.4 0.006 0.004 200653_s_at Hs.177656 gb:M27319.1 calmodulin 1 (phosphorylase kinase, delta) 2.4 0.011 0.009 203989_x_at Hs.128087 gb:NM_001992.2 coagulation factor II (thrombin) receptor 2.4 0.009 0.016 209473_at Hs.205353 gb:AV717590 ectonucleoside triphosphate diphosphohydrolase 1 2.4 0.006 0.027 201579_at Hs.166994 gb:NM_005245.1 FAT tumor suppressor (Drosophila) homolog 2.4 0.025 0.012 209787_s_at Hs.236774 gb:BC001282.1 high-mobility group (nonhistone chromosomal) protein 17-like 3 2.4 0 0.007 202351_at Hs.295726 gb:AI093579_RC integrin, alpha V (vitronectin receptor, alpha polypeptide, antigen CD51) 2.4 0.014 0.035 203882_at Hs.1706 gb:NM_006084.1 interferon-stimulated transcription factor 3, gamma (48kD) 2.4 0.001 0.004 204116_at Hs.84 gb:NM_000206.1 interleukin 2 receptor, gamma (severe combined immunodeficiency) 2.4 0.014 0.021 212149_at Hs.84087 gb:AW470003_RC KIAA0143 protein 2.4 0.015 0.021 212585_at Hs.109694 gb:BF970829 KIAA1451 protein 2.4 0.018 0.012 232950_s_at Hs.272759 gb:AB040890.1 KIAA1457 protein 2.4 0.007 0.021 202729_s_at Hs.241257 gb:NM_000627.1 latent transforming growth factor beta binding protein 1 2.4 0.005 0.021 204890_s_at Hs.1765 gb:U07236.1 lymphocyte-specific protein tyrosine kinase 2.4 0.007 0.027 201930_at Hs.155462 gb:NM_005915.2 minichromosome maintenance deficient (mis5, S. pombe) 6 2.4 0 0.001 200864_s_at Hs.75618 gb:NM_004663.1 RAB11A, member RAS oncogene family 2.4 0.007 0.007 207419_s_at Hs.173466 gb:NM_002872.2 ras-related C3 botulinum toxin substrate 2 (rho family, small GTP binding protein Rac2) 2.4 0.043 0.035 227003_at Hs.166563 gb:BG150494_RC replication factor C (activator 1) 1 (145kD) 2.4 0.004 0.009 204433_s_at Hs.48513 gb:U28164.1 spermatogenesis associated 2 2.4 0.046 0.039 203508_at Hs.256278 gb:NM_001066.1 tumor necrosis factor receptor superfamily, member 1B 2.4 0.004 0.021 202603_at Hs.172028 gb:N51370_RC a disintegrin and metalloproteinase domain 10 2.3 0.004 0.01 207988_s_at Hs.83583 gb:NM_005731.1 actin related protein 23 complex, subunit 2 (34 kD) 2.3 0 0.001 232617_at Hs.181301 gb:AK024855.1 cathepsin S 2.3 0.003 0.021 211919_s_at gb:AF348491.1 chemokine receptor CXCR4 2.3 0.002 0.035 204980_at Hs.50722 gb:NM_004898.1 clock (mouse) homolog 2.3 0.012 0.027 214428_x_at Hs.170250 gb:K02403.1 complement component 4A 2.3 0.04 0.027 221474_at Hs.180224 gb:U26162.1 death-associated protein 6 2.3 0.02 0.044 201938_at Hs.3436 gb:NM_004642.1 deleted in oral cancer (mouse, homolog) 1 2.3 0.003 0.016 212698_s_at Hs.79844 gb:BF966021 DKFZP564M1416 protein 2.3 0.003 0.009 201324_at Hs.79368 gb:NM_001423.1 epithelial membrane protein 1 2.3 0.008 0.035 219778_at Hs.106309 gb:NM_012082.2 Friend of GATA2 2.3 0.019 0.035 207168_s_at Hs.75258 gb:NM_004893.1 H2A histone family, member Y 2.3 0.006 0.021 217918_at Hs.100002 gb:NM_014183.1 HSPC162 protein 2.3 0.011 0.002 209648_x_at Hs.169836 gb:AL136896.1 KIAA0671 gene product 2.3 0.015 0.035 212405_s_at Hs.19469 gb:AK001172.1 KIAA0859 protein 2.3 0.006 0.016 224909_s_at Hs.109315 gb:BF308645 KIAA1415 protein 2.3 0.006 0.035 210605_s_at Hs.3745 gb:BC003610.1 milk fat globule-EGF factor 8 protein 2.3 0.036 0.018 203072_at Hs.82251 gb:NM_004998.1 myosin IC 2.3 0.004 0.021 204702_s_at Hs.22900 gb:NM_004289.3 nuclear factor (erythroid-derived 2)-like 3 2.3 0.009 0.027 202725_at Hs.171880 gb:NM_000937.1 polymerase (RNA) II (DNA directed) polypeptide A (220kD) 2.3 0.001 0.002 202165_at Hs.267819 gb:BF966540_RC protein phosphatase 1, regulatory (inhibitor) subunit 2 2.3 0 0.003 217841_s_at Hs.63304 gb:NM_016147.1 protein phosphatase methylesterase-1 2.3 0.004 0.012 201745_at Hs.82643 gb:NM_002822.1 protein tyrosine kinase 9 2.3 0.001 0.007 219151_s_at Hs.145409 gb:NM_007081.1 RAB, member of RAS oncogene family-like 2B 2.3 0.018 0.035 225585_at Hs.301746 gb:AI963476_RC RAP2A, member of RAS oncogene family 2.3 0.005 0.009 201266_at Hs.13046 gb:NM_003330.1 thioredoxin reductase 1 2.3 0.003 0.035 209153_s_at Hs.101047 gb:M31523.1 transcription factor 3 (E2A immunoglobulin enhancer binding factors E12E47) 2.3 0.003 0.012 209135_at Hs.283664 gb:AF289489.1 aspartate beta-hydroxylase 2.2 0.021 0.035 208908_s_at Hs.279607 gb:AF327443.1 calpastatin 2.2 0.031 0.044 200935_at Hs.16488 gb:NM_004343.2 calreticulin 2.2 0.025 0.044 208374_s_at Hs.184270 gb:NM_006135.1 capping protein (actin filament) muscle Z-line, alpha 1 2.2 0 0.001 204172_at Hs.89866 gb:NM_000097.1 coproporphyrinogen oxidase (coproporphyria, harderoporphyria) 2.2 0.017 0.012
Page 15 of 17 Upregulated transcripts
Affy Code UniGene Ace GenBank Acession Name Fold Change t test p value MW U test p value 202370_s_at Hs.179881 gb:NM_001755.1 core-binding factor, beta subunit 2.2 0.001 0.007 209015_s_at Hs.181195 gb:BC002446.1 DnaJ (Hsp40) homolog, subfamily B, member 6 2.2 0.01 0.021 209365_s_at Hs.81071 gb:U65932.1 extracellular matrix protein 1 2.2 0.015 0.044 208988_at Hs.219614 gb:BE675843_RC f-box and leucine-rich repeat protein 11 2.2 0.004 0.027 214511_x_at Hs.77424 gb:L03419.1 Fc fragment of IgG, high affinity Ia, receptor for (CD64) 2.2 0.008 0.027 221510_s_at Hs.239189 gb:AF158555.1 glutaminase 2.2 0.012 0.016 217718_s_at Hs.182238 gb:NM_014052.1 GW128 protein 2.2 0.001 0.035 203336_s_at Hs.173274 gb:AL548363 integrin cytoplasmic domain-associated protein 1 2.2 0.025 0.027 226185_at Hs.52620 gb:AK026697.1 integrin, beta 8 2.2 0.012 0.021 211762_s_at gb:BC005978.1 karyopherin alpha 2 (RAG cohort 1, importinalpha 1) 2.2 0 0.004 212573_at Hs.167115 gb:AF131747.1 KIAA0830 protein 2.2 0.001 0.005 212765_at Hs.23585 gb:AB029001.1 KIAA1078 protein 2.2 0.023 0.044 227493_s_at Hs.173042 gb:AI863484_RC KIAA1143 protein 2.2 0.007 0.012 226489_at Hs.173392 gb:BG177562 KIAA1145 protein 2.2 0.003 0.009 201505_at Hs.82124 gb:NM_002291.1 laminin, beta 1 2.2 0.026 0.044 208306_x_at Hs.293934 gb:NM_021983.2 major histocompatibility complex, class II, DR beta 4 2.2 0.02 0.044 200978_at Hs.75375 gb:NM_005917.1 malate dehydrogenase 1, NAD (soluble) 2.2 0.009 0.012 209332_s_at Hs.42712 gb:BC003525.1 MAX protein 2.2 0 0.001 217975_at Hs.15984 gb:NM_016303.1 pp21 homolog 2.2 0.012 0.044 217739_s_at Hs.239138 gb:NM_005746.1 pre-B-cell colony-enhancing factor 2.2 0.004 0.007 225975_at Hs.97266 gb:AW189885_RC protocadherin 18 2.2 0.005 0.009 208070_s_at Hs.115521 gb:NM_002912.1 REV3 (yeast homolog)-like, catalytic subunit of DNA polymerase zeta 2.2 0.041 0.044 213566_at Hs.23262 gb:NM_005615.1 ribonuclease, RNase A family, k6 2.2 0.019 0.044 225039_at Hs.125845 gb:AV699857_RC ribulose-5-phosphate-3-epimerase 2.2 0.018 0.035 205241_at Hs.278431 gb:NM_005138.1 SCO (cytochrome oxidase deficient, yeast) homolog 2 2.2 0.009 0.027 204362_at Hs.52644 gb:NM_003930.1 SKAP55 homologue 2.2 0.023 0.044 205644_s_at Hs.77496 gb:NM_003096.1 small nuclear ribonucleoprotein polypeptide G 2.2 0.007 0.021 213664_at Hs.91139 gb:AW235061_RC solute carrier family 1 (neuronalepithelial high affinity glutamate transporter, system Xag), member 1 2.2 0.002 0.004 204588_s_at Hs.194693 gb:NM_003982.1 solute carrier family 7 (cationic amino acid transporter, y+ system), member 7 2.2 0.004 0.016 201110_s_at Hs.87409 gb:NM_003246.1 thrombospondin 1 2.2 0.046 0.044 225643_at Hs.75968 gb:AI261542_RC thymosin, beta 4, X chromosome 2.2 0.001 0.002 201821_s_at Hs.20716 gb:BC004439.1 translocase of inner mitochondrial membrane 17 (yeast) homolog A 2.2 0.004 0.005 220177_s_at Hs.298241 gb:NM_024022.1 Transmembrane protease, serine 3 2.2 0.007 0.021 223675_s_at Hs.24135 gb:AF216644.1 transmembrane protein vezatin; hypothetical protein DKFZp761C241 2.2 0.006 0.016 212242_at Hs.75318 gb:AL565074_RC tubulin, alpha 1 (testis specific) 2.2 0.008 0.021 211986_at Hs.301417 gb:BG287862 AHNAK nucleoprotein (desmoyokin) 2.1 0.039 0.035 230875_s_at Hs.29189 gb:AW068936_RC ATPase, Class VI, type 11A 2.1 0.006 0.016 206562_s_at Hs.283738 gb:NM_001892.1 casein kinase 1, alpha 1 2.1 0.012 0.044 203416_at Hs.82212 gb:NM_000560.1 CD53 antigen 2.1 0.011 0.016 218983_at Hs.98571 gb:NM_016546.1 complement C1r-like proteinase precursor, 2.1 0.011 0.009 203187_at Hs.82295 gb:NM_001380.1 dedicator of cyto-kinesis 1 2.1 0.014 0.012 209486_at Hs.322901 gb:BC004546.1 disrupter of silencing 10 2.1 0.047 0.044 222154_s_at Hs.5297 gb:AK002064.1 DKFZP564A2416 protein 2.1 0.007 0.021 225734_at Hs.289074 gb:AW294765_RC F-box only protein 22 2.1 0.01 0.027 204232_at Hs.743 gb:NM_004106.1 Fc fragment of IgE, high affinity I, receptor for; gamma polypeptide 2.1 0.011 0.021 201209_at Hs.88556 gb:NM_004964.2 histone deacetylase 1 2.1 0.02 0.035 218728_s_at Hs.108854 gb:NM_014184.1 HSPC163 protein 2.1 0.013 0.027 209297_at Hs.66392 gb:AF114488.1 intersectin 1 (SH3 domain protein) 2.1 0.009 0.016 220358_at Hs.62919 gb:NM_018664.1 Jun dimerization protein p21SNFT 2.1 0.005 0.012 203764_at Hs.77695 gb:NM_014750.1 KIAA0008 gene product 2.1 0.036 0.035 212950_at Hs.22039 gb:BF941499_RC KIAA0758 protein 2.1 0.019 0.012 200784_s_at Hs.89137 gb:BF304759 low density lipoprotein-related protein 1 (alpha-2-macroglobulin receptor) 2.1 0.023 0.027 203932_at Hs.1162 gb:NM_002118.1 major histocompatibility complex, class II, DM beta 2.1 0.008 0.007 211063_s_at gb:BC006403.1 NCK adaptor protein 1 2.1 0 0.001 202430_s_at Hs.198282 gb:NM_021105.1 phospholipid scramblase 1 2.1 0.002 0.012 200815_s_at Hs.77318 gb:L13386.1 platelet-activating factor acetylhydrolase, isoform Ib, alpha subunit (45kD) 2.1 0.03 0.035 205361_s_at Hs.91161 gb:AI718295_RC prefoldin 4 2.1 0.012 0.016 207543_s_at Hs.76768 gb:NM_000917.1 procollagen-proline, 2-oxoglutarate 4-dioxygenase (proline 4-hydroxylase), alpha polypeptide I 2.1 0.042 0.044 201063_at Hs.167791 gb:NM_002901.1 reticulocalbin 1, EF-hand calcium binding domain 2.1 0.008 0.027
Page 16 of 17 Upregulated transcripts
Affy Code UniGene Ace GenBank Acession Name Fold Change t test p value MW U test p value 218117_at Hs.279919 gb:NM_014248.1 ring-box 1 2.1 0.002 0.007 201542_at Hs.110796 gb:AY008268.1 SAR1 protein 2.1 0.01 0.027 225243_s_at Hs.4007 gb:AB046821.1 Sarcolemmal-associated protein 2.1 0.018 0.044 202951_at Hs.8724 gb:BE048506_RC serine threonine protein kinase 2.1 0.003 0.007 226068_at Hs.74101 gb:BF593625_RC spleen tyrosine kinase 2.1 0.005 0.007 226693_at Hs.469 gb:AW172779_RC succinate dehydrogenase complex, subunit A, flavoprotein (Fp) 2.1 0.004 0.004 202369_s_at Hs.153954 gb:NM_012288.1 TRAM-like protein 2.1 0.015 0.044 202687_s_at Hs.83429 gb:U57059.1 tumor necrosis factor (ligand) superfamily, member 10 2.1 0.014 0.021 207426_s_at Hs.181097 gb:NM_003326.1 tumor necrosis factor (ligand) superfamily, member 4 (tax-transcriptionally activated glycoprotein 1, 34kD) 2.1 0.016 0.021 227853_at Hs.279860 gb:AW024350_RC tumor protein, translationally-controlled 1 2.1 0.009 0.027
Page 17 of 17 4. Down-regulated Genes in PC Downregulated transcripts
Affymetrix code Unigene Code GenBank Code Name Fold Change t test p value MW U P value 219664_s_at Hs.15898 gb:NM_020664.1 2,4-dienoyl CoA reductase 2, peroxisomal 0.4 0.004 0.004 219076_s_at Hs.49912 gb:NM_018663.1 22kDa peroxisomal membrane protein-like 0.4 0.007 0.003 203557_s_at Hs.3192 gb:NM_000281.1 6-pyruvoyl-tetrahydropterin synthasedimerization cofactor of hepatocyte nuclear factor 1 alpha (TCF1) 0.4 0.018 0.007 211513_s_at Hs.67896 gb:AF172449.1 7-60 protein 0.2 0 0.001 201674_s_at Hs.78921 gb:BC000729.1 A kinase (PRKA) anchor protein 1 0.4 0.009 0.007 200837_at Hs.291904 gb:NM_005745.3 accessory proteins BAP31BAP29 0.4 0.024 0.016 205412_at Hs.37 gb:NM_000019.1 acetyl-Coenzyme A acetyltransferase 1 (acetoacetyl Coenzyme A thiolase) 0.2 0 0.001 213768_s_at Hs.1619 gb:AW950513 achaete-scute complex (Drosophila) homolog-like 1 0.4 0.01 0.016 202767_at Hs.75589 gb:NM_001610.1 acid phosphatase 2, lysosomal 0.4 0 0.001 221462_x_at Hs.250770 gb:NM_017509.1 ACO for serine protease homologue 0.3 0.038 0.004 223874_at Hs.12887 gb:AB039791.1 actin-related protein 3-beta 0.4 0.026 0.012 200779_at Hs.181243 gb:NM_001675.1 activating transcription factor 4 (tax-responsive enhancer element B67) 0.4 0 0.001 206069_s_at Hs.1209 gb:NM_001608.1 acyl-Coenzyme A dehydrogenase, long chain 0.3 0.004 0.003 200710_at Hs.82208 gb:NM_000018.1 acyl-Coenzyme A dehydrogenase, very long chain 0.4 0 0.001 214164_x_at Hs.5344 gb:BF752277 adaptor-related protein complex 1, gamma 1 subunit 0.3 0.008 0.005 214726_x_at Hs.183706 gb:AL556041 adducin 1 (alpha) 0.4 0.007 0.005 202211_at Hs.13014 gb:BC005122.1 ADP-ribosylation factor GTPase activating protein 1 0.4 0.013 0.002 206128_at Hs.123022 gb:AI264306_RC adrenergic, alpha-2C-, receptor 0.4 0.006 0.009 210326_at Hs.144567 gb:D13368.1 alanine-glyoxylate aminotransferase 0.2 0.046 0.001 202888_s_at Hs.1239 gb:NM_001150.1 alanyl (membrane) aminopeptidase 0.2 0 0.004 212224_at Hs.76392 gb:NM_000689.1 aldehyde dehydrogenase 1 family, member A1 0.3 0.01 0.007 205083_at Hs.174151 gb:NM_001159.2 aldehyde oxidase 1 0.1 0.006 0.001 206469_x_at Hs.284236 gb:NM_012067.1 aldo-keto reductase family 7, member A3 (aflatoxin aldehyde reductase) 0.3 0.017 0.005 218203_at Hs.227933 gb:NM_013338.2 Alg5, S. cerevisiae, homolog of 0.3 0.001 0.001 218636_s_at Hs.279881 gb:NM_016219.1 alpha 1,2-mannosidase 0.3 0.003 0.007 205477_s_at Hs.76177 gb:NM_001633.1 alpha-1-microglobulinbikunin precursor 0.2 0.029 0.009 209309_at Hs.71 gb:D90427.1 alpha-2-glycoprotein 1, zinc 0.1 0.004 0.001 218908_at Hs.298351 gb:NM_024083.1 alveolar soft part sarcoma chromosome region, candidate 1 0.3 0 0.001 200714_x_at Hs.76228 gb:NM_006812.1 amplified in osteosarcoma 0.3 0.003 0.001 208498_s_at Hs.274376 gb:NM_004038.1 amylase, alpha 1A; salivary 0.1 0.001 0.001 209462_at Hs.74565 gb:U48437.1 amyloid beta (A4) precursor-like protein 1 0.2 0.019 0.012 208722_s_at Hs.7101 gb:BC001081.1 anaphase-promoting complex subunit 5 0.4 0.001 0.001 232810_at Hs.107528 gb:AK001347.1 androgen induced protein 0.4 0.013 0.009 211689_s_at gb:AF270487.1 androgen-regulated serine protease TMPRSS2precursor 0.4 0.001 0.016 224339_s_at gb:AB056476.1 angiopoietin-related protein 1 0.3 0.001 0.007 210085_s_at Hs.279928 gb:AF230929.1 annexin A9 0.2 0.004 0.001 218140_x_at Hs.12152 gb:NM_021203.1 APMCF1 protein 0.3 0 0.001 207542_s_at Hs.74602 gb:NM_000385.2 aquaporin 1 (channel-forming integral protein, 28kD) 0.4 0.023 0.007 203747_at Hs.234642 gb:NM_004925.2 aquaporin 3 0.3 0.036 0.021 204288_s_at Hs.278626 gb:NM_021069.1 ArgAbl-interacting protein ArgBP2 0.3 0.026 0.007 208260_at Hs.1372 gb:NM_000707.2 arginine vasopressin receptor 1B 0.4 0.014 0.005 202655_at Hs.75412 gb:NM_006010.1 arginine-rich, mutated in early stage tumors 0.4 0 0.002 205784_x_at Hs.171900 gb:NM_001670.1 armadillo repeat gene deletes in velocardiofacial syndrome 0.4 0.023 0.016 202024_at Hs.165439 gb:NM_004317.1 arsA (bacterial) arsenite transporter, ATP-binding, homolog 1 0.4 0.001 0.007 210237_at Hs.194689 gb:AF120274.1 artemin 0.4 0.022 0.009 205969_at Hs.587 gb:NM_001086.1 arylacetamide deacetylase (esterase) 0.4 0.027 0.035 204443_at Hs.88251 gb:NM_000487.3 arylsulfatase A 0.3 0 0.001 205894_at Hs.74131 gb:NM_000047.1 arylsulfatase E (chondrodysplasia punctata 1) 0.2 0.001 0.001 205047_s_at Hs.75692 gb:NM_001673.1 asparagine synthetase 0.3 0.003 0.001 216064_s_at Hs.207776 gb:W27131_RC aspartylglucosaminidase 0.4 0.002 0.004 207552_at Hs.89399 gb:NM_005176.3 ATP synthase, H+ transporting, mitochondrial F0 complex, subunit c (subunit 9), isoform 2 0.4 0.034 0.021 203926_x_at Hs.89761 gb:NM_001687.1 ATP synthase, H+ transporting, mitochondrial F1 complex, delta subunit 0.3 0 0.001 207522_s_at Hs.5541 gb:NM_005173.1 ATPase, Ca++ transporting, ubiquitous 0.2 0 0.001 205095_s_at Hs.267871 gb:NM_005177.1 ATPase, H+ transporting, lysosomal (vacuolar proton pump) non-catalytic accessory protein 1A (110116kD) 0.3 0.003 0.001 207139_at Hs.36992 gb:NM_000704.1 ATPase, H+K+ exchanging, alpha polypeptide 0.2 0.004 0.005 220948_s_at Hs.76549 gb:NM_000701.1 ATPase, Na+K+ transporting, alpha 1 polypeptide 0.3 0 0.001 211852_s_at Hs.194019 gb:AF106861.1 attractin 0.3 0 0.001 234727_at Hs.248046 gb:Z83801.1_RC axonemal dynein heavy chain 7 0.4 0.006 0.012 216878_x_at Hs.100015 gb:X83412.1_RC B1 for mucin 0.4 0.041 0.027
Page 1 of 15 Downregulated transcripts
Affymetrix code Unigene Code GenBank Code Name Fold Change t test p value MW U P value 207832_at Hs.7936 gb:NM_017451.1 BAI1-associated protein 2 0.3 0.045 0.009 217461_x_at Hs.181966 gb:M90355 basic transcription factor 3, like 2 0.4 0.02 0.016 208677_s_at Hs.74631 gb:AL550657 basigin (OK blood group) 0.3 0.001 0.003 200920_s_at Hs.77054 gb:AL535380 B-cell translocation gene 1, anti-proliferative 0.4 0.005 0.003 201849_at Hs.79428 gb:NM_004052.2 BCL2adenovirus E1B 19kD-interacting protein 3 0.1 0.008 0.001 219326_s_at Hs.48730 gb:NM_006577.2 beta-1,3-N-acetylglucosaminyltransferase bGnT-2 0.4 0.001 0.009 224335_s_at gb:AB050436.1 beta-site APP cleaving enzyme I-476 0.2 0 0.001 217904_s_at Hs.49349 gb:NM_012104.1 beta-site APP-cleaving enzyme 0.1 0 0.001 204167_at Hs.78885 gb:NM_000060.1 biotinidase 0.4 0 0.003 205750_at Hs.7298 gb:NM_004332.1 biphenyl hydrolase-like (serine hydrolase; breast epithelial mucin-associated antigen) 0.4 0.046 0.012 207866_at Hs.99948 gb:NM_001720.1 bone morphogenetic protein 8 (osteogenic protein 2) 0.4 0.036 0.009 204230_s_at Hs.6535 gb:NM_020309.1 brain-specific Na-dependent inorganic phosphate cotransporter 0.4 0.005 0.021 214452_at Hs.157205 gb:NM_005504.1 branched chain aminotransferase 1, cytosolic 0.3 0.011 0.003 203576_at Hs.101408 gb:NM_001190.1 branched chain aminotransferase 2, mitochondrial 0.3 0.006 0.001 202331_at Hs.78950 gb:NM_000709.1 branched chain keto acid dehydrogenase E1, alpha polypeptide (maple syrup urine disease) 0.4 0.001 0.002 205531_s_at Hs.325443 gb:NM_013267.1 breast cell glutaminase 0.2 0 0.001 222737_s_at Hs.279762 gb:AI674162_RC bromodomain-containing 7 0.4 0 0.001 201236_s_at Hs.75462 gb:NM_006763.1 BTG family, member 2 0.1 0.003 0.001 211009_s_at Hs.86276 gb:AF159567.1 C2H2 (Kruppel-type) zinc finger protein 0.4 0 0.001 233950_at Hs.151301 gb:AK000873.1 Ca2+-dependent activator protein for secretion 0.4 0.011 0.016 206280_at Hs.57691 gb:NM_004934.1 cadherin 18, type 2 0.4 0.007 0.007 232846_s_at Hs.244343 gb:AL122081.1 cadherin related 23 0.3 0.004 0.004 207745_at Hs.278984 gb:NM_016366.1 calcium binding protein 2 0.3 0.018 0.012 208377_s_at Hs.139263 gb:NM_005183.1 calcium channel, voltage-dependent, alpha 1F subunit 0.1 0.003 0.001 219896_at Hs.148680 gb:NM_015722.2 calcyon; D1 dopamine receptor-interacting protein 0.4 0.012 0.007 226811_at Hs.182278 gb:AL046017_RC calmodulin 2 (phosphorylase kinase, delta) 0.2 0.004 0.001 208852_s_at Hs.155560 gb:AI761759_RC calnexin 0.4 0.002 0.009 203357_s_at Hs.7145 gb:NM_014296.1 calpain 7 0.4 0.002 0.005 206208_at Hs.89485 gb:NM_000717.2 carbonic anhydrase IV 0.4 0.006 0.007 203963_at Hs.5338 gb:NM_001218.2 carbonic anhydrase XII 0.1 0.011 0.001 219464_at Hs.235168 gb:NM_012113.1 carbonic anhydrase XIV 0.3 0.009 0.005 205910_s_at Hs.99918 gb:NM_001807.1 carboxyl ester lipase (bile salt-stimulated lipase) 0.1 0 0.001 205615_at Hs.2879 gb:NM_001868.1 carboxypeptidase A1 (pancreatic) 0.1 0.001 0.001 206212_at Hs.89717 gb:NM_001869.1 carboxypeptidase A2 (pancreatic) 0.1 0 0.001 205509_at Hs.180884 gb:NM_001871.1 carboxypeptidase B1 (tissue) 0.1 0.001 0.001 209522_s_at Hs.12068 gb:BC000723.1 carnitine acetyltransferase 0.2 0.02 0.001 204263_s_at Hs.274336 gb:M58581.1 carnitine palmitoyltransferase II 0.4 0.014 0.016 212075_s_at Hs.155140 gb:AI161318_RC casein kinase 2, alpha 1 polypeptide 0.4 0.016 0.009 203575_at Hs.82201 gb:NM_001896.1 casein kinase 2, alpha prime polypeptide 0.4 0.003 0.003 210955_at Hs.5353 gb:U86214.1 caspase 10, apoptosis-related cysteine protease 0.4 0.004 0.005 205373_at Hs.150917 gb:NM_004389.1 catenin (cadherin-associated protein), alpha 2 0.3 0 0.003 203657_s_at Hs.11590 gb:NM_003793.2 cathepsin F 0.4 0.002 0.004 214450_at Hs.87450 gb:NM_001335.1 cathepsin W (lymphopain) 0.3 0.005 0.003 209357_at Hs.82071 gb:AF109161.1 Cbpp300-interacting transactivator, with GluAsp-rich carboxy-terminal domain, 2 0.4 0.018 0.016 204039_at Hs.76171 gb:NM_004364.1 CCAATenhancer binding protein (CEBP), alpha 0.4 0.001 0.016 211141_s_at Hs.108300 gb:AF180474.1 CCR4-NOT transcription complex, subunit 3 0.2 0 0.001 204581_at Hs.171763 gb:NM_001771.1 CD22 antigen 0.3 0.001 0.005 209772_s_at Hs.286124 gb:X69397.1 CD24 antigen (small cell lung carcinoma cluster 4 antigen) 0.2 0 0.001 206680_at Hs.52002 gb:NM_005894.1 CD5 antigen-like (scavenger receptor cysteine rich family) 0.4 0.026 0.007 200663_at Hs.76294 gb:NM_001780.1 CD63 antigen (melanoma 1 antigen) 0.4 0.003 0.001 211192_s_at Hs.137548 gb:AF054818.1 CD84 antigen (leukocyte antigen) 0.3 0.014 0.009 202140_s_at Hs.73987 gb:NM_003992.1 CDC-like kinase 3 0.4 0.003 0.005 215913_s_at Hs.107056 gb:AK023668.1 CED-6 protein 0.4 0.025 0.016 221295_at Hs.249129 gb:NM_001279.1 cell death-inducing DFFA-like effector a 0.1 0 0.001 212540_at Hs.76932 gb:BG476661 cell division cycle 34 0.4 0.028 0.021 219300_s_at Hs.106552 gb:AB020675.1 cell recognition molecule Caspr2 0.3 0.011 0.012 217770_at Hs.84038 gb:NM_015937.1 CGI-06 protein 0.4 0 0.001 218034_at Hs.84344 gb:NM_016068.1 CGI-135 protein 0.4 0 0.002 227841_at Hs.9204 gb:BG260181 CGI-14 protein 0.4 0.001 0.005
Page 2 of 15 Downregulated transcripts
Affymetrix code Unigene Code GenBank Code Name Fold Change t test p value MW U P value 218052_s_at Hs.9275 gb:NM_020410.1 CGI-152 protein 0.4 0.001 0.001 217950_at Hs.7236 gb:NM_015953.1 CGI-25 protein 0.2 0 0.001 223696_at Hs.19978 gb:BC003660.1 CGI-30 protein 0.3 0.009 0.009 218765_at Hs.20102 gb:NM_015996.1 CGI-40 protein 0.3 0 0.001 228872_at Hs.124831 gb:AA843160_RC CGI-67 protein 0.4 0 0.007 228924_s_at Hs.184325 gb:AA491236_RC CGI-76 protein 0.4 0.005 0.004 229237_s_at Hs.211773 gb:N21195_RC checkpoint suppressor 1 0.2 0.002 0.001 210381_s_at Hs.203 gb:BC000740.1 cholecystokinin B receptor 0.2 0.004 0.003 231994_at Hs.131668 gb:AJ272267.1 choline dehydrogenase 0.2 0.009 0.003 204266_s_at Hs.77221 gb:NM_001277.1 choline kinase 0.4 0 0.003 204233_s_at Hs.77221 gb:AI991328_RC choline kinase 0.3 0 0.001 214596_at Hs.7138 gb:T15991_RC cholinergic receptor, muscarinic 3 0.3 0.003 0.005 211587_x_at Hs.89605 gb:M37981.1 cholinergic receptor, nicotinic, alpha polypeptide 3 0.4 0.017 0.007 206736_x_at Hs.10734 gb:L35901.1 cholinergic receptor, nicotinic, alpha polypeptide 4 0.4 0.003 0.009 206703_at Hs.89739 gb:NM_000747.1 cholinergic receptor, nicotinic, beta polypeptide 1 (muscle) 0.4 0.003 0.004 206297_at Hs.8709 gb:NM_007272.1 chymotrypsin C (caldecrin) 0.1 0.001 0.001 205971_s_at Hs.74502 gb:NM_001906.1 chymotrypsinogen B1 0.1 0.001 0.001 205328_at Hs.26126 gb:NM_006984.1 claudin 10 0.3 0 0.001 210689_at Hs.266416 gb:AF314090.1 claudin 14 0.4 0.018 0.016 203954_x_at Hs.25640 gb:NM_001306.1 claudin 3 0.4 0.024 0.021 208791_at Hs.75106 gb:M25915.1 clusterin (complement lysis inhibitor, SP-40,40) 0.2 0.001 0.001 214787_at Hs.135202 gb:BE268538 c-myc promoter-binding protein 0.4 0.009 0.003 205229_s_at Hs.21016 gb:AA669336_RC coagulation factor C (Limulus polyphemus) homology (cochlin) 0.1 0.004 0.001 205756_s_at Hs.79345 gb:NM_000132.2 coagulation factor VIII, procoagulant component (hemophilia A) 0.2 0.032 0.001 206610_s_at Hs.1430 gb:NM_000128.2 coagulation factor XI (plasma thromboplastin antecedent) 0.2 0.001 0.002 204427_s_at Hs.323378 gb:NM_006815.1 coated vesicle membrane protein 0.3 0.003 0.003 200811_at Hs.119475 gb:NM_001280.1 cold inducible RNA-binding protein 0.2 0 0.001 228634_s_at Hs.1139 gb:BF195718_RC cold shock domain protein A 0.4 0.004 0.005 206131_at Hs.1340 gb:NM_001832.2 colipase, pancreatic 0.1 0 0.001 229932_at Hs.284394 gb:AV723710 complement component 3 0.4 0 0.001 205500_at Hs.1281 gb:NM_001735.1 complement component 5 0.1 0.005 0.001 210168_at Hs.1282 gb:J05064.1 complement component 6 0.4 0.009 0.009 231336_at Hs.7130 gb:AI703256_RC copine IV 0.4 0.049 0.044 208056_s_at Hs.110099 gb:NM_005187.2 core-binding factor, runt domain, alpha subunit 2; translocated to, 3 0.3 0 0.001 212695_at Hs.7278 gb:AB014558.1 cryptochrome 2 (photolyase-like) 0.3 0.033 0.007 206967_at Hs.279906 gb:NM_001240.1 cyclin T1 0.3 0.002 0.009 214638_s_at Hs.155478 gb:AV681875 cyclin T2 0.4 0.007 0.009 204247_s_at Hs.166071 gb:NM_004935.1 cyclin-dependent kinase 5 0.3 0.007 0.005 212816_s_at Hs.84152 gb:BE613178_RC cystathionine-beta-synthase 0.1 0.006 0.001 213681_at Hs.173884 gb:AW512817_RC cysteine and histidine rich protein 0.4 0.001 0.002 210764_s_at Hs.8867 gb:AF003114.1 cysteine-rich, angiogenic inducer, 61 0.4 0.042 0.021 205043_at Hs.663 gb:NM_000492.2 cystic fibrosis transmembrane conductance regulator, ATP-binding cassette (sub-family C, member 7) 0.4 0.004 0.009 207843_x_at Hs.83834 gb:NM_001914.1 cytochrome b-5 0.4 0 0.004 209163_at Hs.153028 gb:AL514271_RC cytochrome b-561 0.3 0 0.002 220331_at Hs.25121 gb:NM_006668.1 cytochrome P450, subfamily 46 (cholesterol 24-hydroxylase) 0.3 0 0.005 214320_x_at Hs.183584 gb:T67741_RC cytochrome P450, subfamily IIA (phenobarbital-inducible), polypeptide 6 0.4 0.024 0.009 201571_s_at Hs.76894 gb:AI656493_RC dCMP deaminase 0.4 0.014 0.021 202929_s_at Hs.180015 gb:NM_001355.2 D-dopachrome tautomerase 0.3 0.015 0.004 201095_at Hs.75189 gb:NM_004394.1 death-associated protein 0.4 0 0.002 209183_s_at Hs.93675 gb:AL136653.1 decidual protein induced by progesterone 0.2 0.017 0.003 210397_at Hs.32949 gb:U73945.1 defensin, beta 1 0.2 0.007 0.004 218819_at Hs.58570 gb:NM_012141.1 deleted in cancer 1; RNA helicase HDBDICE1 0.4 0.006 0.009 222898_s_at Hs.127792 gb:BE350882_RC delta (Drosophila)-like 3 0.4 0 0.003 223525_at Hs.3736 gb:AB036931.1 delta-like 4 homolog (Drosophila) 0.4 0.045 0.016 209560_s_at Hs.169228 gb:U15979.1 delta-like homolog (Drosophila) 0.3 0.01 0.001 210165_at Hs.113221 gb:M55983.1 deoxyribonuclease I 0.1 0.023 0.004 213068_at Hs.80552 gb:AI146848_RC dermatopontin 0.3 0.001 0.009 201021_s_at Hs.82306 gb:BF697964 destrin (actin depolymerizing factor) 0.4 0.033 0.016 207556_s_at Hs.89981 gb:NM_003646.1 diacylglycerol kinase, zeta (104kD) 0.4 0.01 0.007
Page 3 of 15 Downregulated transcripts
Affymetrix code Unigene Code GenBank Code Name Fold Change t test p value MW U P value 215003_at Hs.106311 gb:AA921844_RC DiGeorge syndrome gene A 0.2 0.004 0.001 205369_x_at Hs.139410 gb:J03208.1 dihydrolipoamide branched chain transacylase 0.3 0.042 0.018 205983_at Hs.109 gb:NM_004413.1 dipeptidase 1 (renal) 0.1 0.003 0.001 220611_at Hs.8108 gb:NM_021080.1 disabled (Drosophila) homolog 1 0.1 0.04 0.004 201907_x_at Hs.174044 gb:U49262.1 dishevelled 3 (homologous to Drosophila dsh) 0.4 0.01 0.016 227722_at Hs.107187 gb:AW043594_RC divalent cation tolerant protein CUTA 0.4 0.027 0.001 226606_s_at Hs.35389 gb:AI860690_RC DKFZP434C0935 protein 0.4 0.036 0.021 212886_at Hs.209100 gb:AL080169.1 DKFZP434C171 protein 0.4 0.009 0.007 224742_at Hs.236522 gb:BF570412_RC DKFZP434P106 protein 0.4 0.001 0.003 212018_s_at Hs.85963 gb:AK000822.1 DKFZP564M182 protein 0.4 0.003 0.004 207170_s_at Hs.75884 gb:NM_015416.1 DKFZP586A011 protein 0.4 0 0.001 207761_s_at Hs.288771 gb:NM_014033.1 DKFZP586A0522 protein 0.3 0.001 0.004 221790_s_at Hs.184482 gb:AL545035_RC DKFZP586D0624 protein 0.4 0.033 0.004 213861_s_at Hs.49378 gb:N67741_RC DKFZP586D0919 protein 0.4 0.001 0.003 214156_at Hs.26970 gb:AL050090.1 DKFZP586F1018 protein 0.4 0.018 0.021 202500_at Hs.77768 gb:NM_006736.1 DnaJ (Hsp40) homolog, subfamily B, member 2 0.4 0.006 0.005 202843_at Hs.6790 gb:NM_012328.1 DnaJ (Hsp40) homolog, subfamily B, member 9 0.4 0.028 0.004 206782_s_at Hs.172847 gb:NM_005528.1 DnaJ (Hsp40) homolog, subfamily C, member 4 0.3 0.009 0.007 228622_s_at Hs.172847 gb:AW071239_RC DnaJ (Hsp40) homolog, subfamily C, member 4 0.2 0.006 0.001 208675_s_at Hs.34789 gb:D29643.1 dolichyl-diphosphooligosaccharide-protein glycosyltransferase 0.2 0.004 0.001 219373_at Hs.110477 gb:NM_018973.1 dolichyl-phosphate mannosyltransferase polypeptide 3 0.4 0.018 0.007 204522_at Hs.153299 gb:NM_005510.1 DOM-3 (C. elegans) homolog Z 0.3 0.001 0.001 206590_x_at Hs.73893 gb:NM_000795.1 dopamine receptor D2 0.4 0.002 0.004 221689_s_at Hs.66493 gb:AB035745.1 Down syndrome critical region gene 5 0.4 0.016 0.003 230073_at Hs.16697 gb:AI052744_RC down-regulator of transcription 1, TBP-binding (negative cofactor 2) 0.4 0.018 0.016 201041_s_at Hs.171695 gb:NM_004417.2 dual specificity phosphatase 1 0.4 0.011 0.021 211541_s_at Hs.75842 gb:U52373.1 dual-specificity tyrosine-(Y)-phosphorylation regulated kinase 1A 0.3 0.004 0.004 208430_s_at Hs.54435 gb:NM_001390.1 dystrobrevin, alpha 0.4 0.046 0.012 203881_s_at Hs.169470 gb:NM_004010.1 dystrophin (muscular dystrophy, Duchenne and Becker types), includes DXS142, DXS164, DXS206, DXS230, 0.3 0.023 0.005 201622_at Hs.79093 gb:NM_014390.1 EBNA-2 co-activator (100kD) 0.4 0.002 0.002 218225_at Hs.22199 gb:NM_016581.1 ECSIT 0.3 0.003 0.001 206191_at Hs.47042 gb:NM_001248.1 ectonucleoside triphosphate diphosphohydrolase 3 0.4 0.011 0.004 234946_at Hs.12330 gb:AL035252 ectonucleoside triphosphate diphosphohydrolase 6 (putative function) 0.4 0.002 0.005 205066_s_at Hs.11951 gb:NM_006208.1 ectonucleotide pyrophosphatasephosphodiesterase 1 0.3 0.012 0.009 221141_x_at Hs.279953 gb:NM_013333.1 EH domain-binding mitotic phosphoprotein 0.4 0.02 0.016 220161_s_at Hs.267997 gb:NM_019114.1 EHM2 gene 0.1 0.001 0.001 206446_s_at Hs.21 gb:NM_001971.1 elastase 1, pancreatic 0.1 0.001 0.001 210080_x_at Hs.181289 gb:D00306.1 elastase 3, pancreatic (protease E) 0.1 0.001 0.001 206151_x_at Hs.183864 gb:NM_007352.1 elastase 3B 0.1 0.001 0.001 234421_s_at Hs.274446 gb:AK025394.1 Ellis van Creveld syndrome 0.4 0.021 0.012 214446_at Hs.173334 gb:NM_012081.1 ELL-RELATED RNA POLYMERASE II, ELONGATION FACTOR 0.3 0.012 0.001 218436_at Hs.297875 gb:NM_022464.1 endoplasmic reticulum chaperone SIL1, homolog of yeast 0.4 0.004 0.004 200805_at Hs.75864 gb:NM_006816.1 endoplasmic reticulum glycoprotein 0.4 0.004 0.003 201216_at Hs.75841 gb:NM_006817.2 endoplasmic reticulum lumenal protein 0.2 0 0.001 220012_at Hs.150763 gb:NM_019891.1 endoplasmic reticulum oxidoreductin 1-Lbeta 0.1 0 0.001 221487_s_at Hs.111680 gb:AF157510.1 endosulfine alpha 0.4 0 0.002 206722_s_at Hs.122575 gb:NM_004720.3 endothelial differentiation, lysophosphatidic acid G-protein-coupled receptor, 4 0.3 0.001 0.001 209059_s_at Hs.174050 gb:AB002282.1 endothelial differentiation-related factor 1 0.4 0 0.004 204718_at Hs.3796 gb:NM_004445.1 EphB6 0.4 0.002 0.002 206254_at Hs.2230 gb:NM_001963.2 epidermal growth factor (beta-urogastrone) 0.2 0 0.001 222112_at Hs.147176 gb:AV710549 epidermal growth factor receptor substrate EPS15R 0.2 0.049 0.024 202017_at Hs.89649 gb:NM_000120.2 epoxide hydrolase 1, microsomal (xenobiotic) 0.1 0.044 0.001 209368_at Hs.113 gb:AF233336.1 epoxide hydrolase 2, cytoplasmic 0.1 0.003 0.001 210825_s_at Hs.160483 gb:AF130103.1_RC erythrocyte membrane protein band 7.2 (stomatin) 0.4 0.024 0.009 217254_s_at Hs.2303 gb:AF053356 erythropoietin 0.3 0.005 0.004 211120_x_at Hs.103504 gb:AB006590.1 estrogen receptor 2 (ER beta) 0.3 0.006 0.005 220275_at Hs.114648 gb:NM_022034.1 estrogen regulated gene 1 0.1 0 0.001 207981_s_at Hs.151017 gb:NM_001438.1 estrogen-related receptor gamma 0.4 0.002 0.007 216375_s_at Hs.43697 gb:X76184.1 ets variant gene 5 (ets-related molecule) 0.3 0.002 0.005
Page 4 of 15 Downregulated transcripts
Affymetrix code Unigene Code GenBank Code Name Fold Change t test p value MW U P value 204892_x_at Hs.181165 gb:NM_001402.1 eukaryotic translation elongation factor 1 alpha 1 0.4 0.001 0.005 200705_s_at Hs.275959 gb:NM_001959.1 eukaryotic translation elongation factor 1 beta 2 0.4 0 0.001 203113_s_at Hs.223241 gb:NM_001960.1 eukaryotic translation elongation factor 1 delta (guanine nucleotide exchange protein) 0.2 0 0.001 200689_x_at Hs.2186 gb:NM_001404.1 eukaryotic translation elongation factor 1 gamma 0.3 0 0.001 204102_s_at Hs.75309 gb:NM_001961.1 eukaryotic translation elongation factor 2 0.3 0.005 0.001 218696_at Hs.102506 gb:NM_004836.1 eukaryotic translation initiation factor 2-alpha kinase 3 0.4 0.019 0.003 200596_s_at Hs.198899 gb:BE614908 eukaryotic translation initiation factor 3, subunit 10 (theta, 150170kD) 0.3 0.001 0.001 208887_at Hs.28081 gb:BC000733.1 eukaryotic translation initiation factor 3, subunit 4 (delta, 44kD) 0.4 0 0.001 226014_at Hs.7811 gb:BF115977_RC eukaryotic translation initiation factor 3, subunit 5 (epsilon, 47kD) 0.4 0.029 0.012 200647_x_at Hs.4835 gb:NM_003752.2 eukaryotic translation initiation factor 3, subunit 8 (110kD) 0.4 0.013 0.009 211937_at Hs.93379 gb:NM_001417.1 eukaryotic translation initiation factor 4B 0.4 0 0.001 221539_at Hs.71819 gb:AB044548.1 eukaryotic translation initiation factor 4E binding protein 1 0.4 0 0.001 213753_x_at Hs.119140 gb:BF541557_RC eukaryotic translation initiation factor 5A 0.4 0.025 0.016 213540_at Hs.288354 gb:AL031228 FabG (beta-ketoacyl-acyl-carrier-protein reductase, E coli) like 0.2 0 0.001 204231_s_at Hs.326190 gb:NM_001441.1 fatty acid amide hydrolase 0.4 0.003 0.009 219305_x_at Hs.132753 gb:NM_012168.1 F-box only protein 2 0.3 0 0.001 225591_at Hs.81001 gb:AA749085_RC F-box only protein 25 0.4 0.001 0.001 205305_at Hs.107 gb:NM_004467.1 fibrinogen-like 1 0.1 0.006 0.001 208417_at Hs.166015 gb:NM_020996.1 fibroblast growth factor 6 0.2 0.014 0.004 207822_at Hs.748 gb:NM_023107.1 fibroblast growth factor receptor 1 (fms-related tyrosine kinase 2, Pfeiffer syndrome) 0.2 0.002 0.001 204379_s_at Hs.1420 gb:NM_000142.2 fibroblast growth factor receptor 3 (achondroplasia, thanatophoric dwarfism) 0.3 0.001 0.002 219117_s_at Hs.24048 gb:NM_016594.1 FK506 binding protein precursor 0.1 0 0.001 203391_at Hs.227729 gb:NM_004470.1 FK506-binding protein 2 (13kD) 0.2 0.001 0.001 219187_at Hs.99134 gb:NM_022110.1 FK506-binding protein like 0.4 0.004 0.004 212025_s_at Hs.83849 gb:BG421186 flightless I (Drosophila) homolog 0.4 0 0.004 206741_at Hs.113019 gb:NM_015931.1 fls485 0.3 0.002 0.003 204131_s_at Hs.14845 gb:N25732_RC forkhead box O3A 0.4 0.034 0.016 220604_x_at Hs.36218 gb:NM_006657.1 formiminotransferase cyclodeaminase 0.3 0.029 0.007 205208_at Hs.9520 gb:NM_012190.1 formyltetrahydrofolate dehydrogenase 0.3 0.008 0.003 206109_at Hs.69747 gb:NM_000148.1 fucosyltransferase 1 (galactoside 2-alpha-L-fucosyltransferase, Bombay phenotype included) 0.2 0.003 0.001 208505_s_at Hs.46328 gb:NM_000511.1 fucosyltransferase 2 (secretor status included) 0.4 0.038 0.007 205674_x_at Hs.19520 gb:NM_001680.2 FXYD domain-containing ion transport regulator 2 0.2 0.003 0.002 218205_s_at Hs.261828 gb:NM_017572.1 G protein-coupled receptor kinase 7 0.4 0 0.001 219327_s_at Hs.58014 gb:NM_022036.1 G protein-coupled receptor, family C, group 5, member C 0.3 0.003 0.001 221342_at Hs.247879 gb:NM_025260.1 G6B protein 0.3 0.012 0.004 204973_at Hs.2679 gb:NM_000166.1 gap junction protein, beta 1, 32kD (connexin 32, Charcot-Marie-Tooth neuropathy, X-linked) 0.4 0.002 0.012 207929_at Hs.73883 gb:NM_005314.1 gastrin-releasing peptide receptor 0.4 0.005 0.005 218350_s_at Hs.234896 gb:NM_015895.1 geminin 0.2 0.001 0.001 224657_at Hs.11169 gb:AL034417 Gene 33Mig-6 0.3 0.004 0.004 202354_s_at Hs.68257 gb:AW190445_RC general transcription factor IIF, polypeptide 1 (74kD subunit) 0.2 0.014 0.007 201338_x_at Hs.75113 gb:NM_002097.1 general transcription factor IIIA 0.4 0.001 0.001 206921_at Hs.169363 gb:NM_001499.1 GLE1 (yeast homolog)-like, RNA export mediator 0.4 0.025 0.007 225001_at Hs.8036 gb:AI744658_RC glioblastoma overexpressed 0.2 0.003 0.001 217807_s_at Hs.2237 gb:NM_015710.1 glioma tumor suppressor candidate region gene 2 0.4 0.001 0.001 232167_at Hs.9475 gb:BE675356_RC glucose transporter protein 10 0.4 0.021 0.016 202830_s_at Hs.26655 gb:NM_001467.1 glucose-6-phosphatase, transport (glucose-6-phosphate) protein 1 0.3 0 0.001 209093_s_at Hs.282997 gb:K02920.1 glucosidase, beta; acid (includes glucosylceramidase) 0.4 0 0.002 230125_at Hs.183868 gb:AA767279_RC glucuronidase, beta 0.4 0.048 0.009 203157_s_at Hs.239189 gb:AB020645.1 glutaminase 0.4 0.012 0.012 208369_s_at Hs.184141 gb:NM_013976.1 glutaryl-Coenzyme A dehydrogenase 0.3 0.013 0.001 201348_at Hs.172153 gb:NM_002084.2 glutathione peroxidase 3 (plasma) 0.4 0.02 0.016 203924_at Hs.89552 gb:NM_000846.1 glutathione S-transferase A2 0.1 0.016 0.001 203815_at Hs.77490 gb:NM_000853.1 glutathione S-transferase theta 1 0.4 0.032 0.016 209531_at Hs.26403 gb:BC001453.1 glutathione transferase zeta 1 (maleylacetoacetate isomerase) 0.4 0.011 0.012 230958_s_at Hs.12482 gb:BE670797_RC glyceronephosphate O-acyltransferase 0.4 0.001 0.009 203178_at Hs.75335 gb:NM_001482.1 glycine amidinotransferase (L-arginine:glycine amidinotransferase) 0.1 0.001 0.001 202210_x_at Hs.118890 gb:NM_019884.1 glycogen synthase kinase 3 alpha 0.4 0.041 0.012 202755_s_at Hs.2699 gb:AI354864_RC glypican 1 0.2 0.023 0.004 214730_s_at Hs.78979 gb:AK025457.1 Golgi apparatus protein 1 0.3 0 0.003
Page 5 of 15 Downregulated transcripts
Affymetrix code Unigene Code GenBank Code Name Fold Change t test p value MW U P value 210425_x_at Hs.182982 gb:AF164622.1 golgin-67 0.1 0.006 0.001 208519_x_at Hs.129715 gb:NM_001501.1 gonadotropin-releasing hormone 2 0.4 0.006 0.012 223373_s_at Hs.25206 gb:AF306567.1 group XII secreted phospholipase A2 0.3 0.004 0.003 209305_s_at Hs.110571 gb:AF078077.1 growth arrest and DNA-damage-inducible, beta 0.3 0.032 0.009 205848_at Hs.129818 gb:NM_005256.1 growth arrest-specific 2 0.2 0.006 0.001 221136_at Hs.279463 gb:NM_016204.1 growth differentiation factor 2 0.2 0.003 0.003 220053_at Hs.86232 gb:NM_020634.1 growth differentiation factor 3 0.4 0.036 0.012 206614_at Hs.1573 gb:NM_000557.2 growth differentiation factor 5 (cartilage-derived morphogenetic protein-1) 0.3 0.021 0.007 209410_s_at Hs.81875 gb:AF000017.1 growth factor receptor-bound protein 10 0.3 0 0.001 206204_at Hs.83070 gb:NM_004490.1 growth factor receptor-bound protein 14 0.3 0 0.001 208068_x_at Hs.115352 gb:NM_022562.1 growth hormone 1 0.3 0.036 0.012 213467_at Hs.99034 gb:BF511718_RC GTP-binding protein Rho7 0.3 0.048 0.007 205354_at Hs.81131 gb:NM_000156.3 guanidinoacetate N-methyltransferase 0.1 0.007 0.001 200780_x_at Hs.273385 gb:NM_000516.2 guanine nucleotide binding protein (G protein), alpha stimulating activity polypeptide 1 0.4 0 0.001 222005_s_at Hs.179915 gb:AL538966 guanine nucleotide binding protein (G protein), gamma 3 0.4 0 0.005 233953_at Hs.233363 gb:AF110003.1 guanylate cyclase activator 1C 0.2 0.006 0.003 223928_s_at Hs.233363 gb:AF110002.1 guanylate cyclase activator 1C 0.1 0 0.001 216293_at Hs.285688 gb:X81636.1_RC H.sapiens clathrin light chain a gene 0.3 0.025 0.009 234880_x_at Hs.247935 gb:X63338 H.sapiens HB2B gene for high sulfur keratin 0.4 0 0.001 200806_s_at Hs.79037 gb:BE256479 heat shock 60kD protein 1 (chaperonin) 0.4 0.007 0.009 210338_s_at Hs.180414 gb:AB034951.1 heat shock 70kD protein 8 0.4 0.001 0.009 201391_at Hs.182366 gb:NM_016292.1 heat shock protein 75 0.4 0 0.001 213756_s_at Hs.1499 gb:AI393937 heat shock transcription factor 1 0.4 0.015 0.007 215933_s_at Hs.118651 gb:Z21533.1 hematopoietically expressed homeobox 0.3 0 0.001 204753_s_at Hs.250692 gb:AI810712_RC hepatic leukemia factor 0.4 0.001 0.005 228463_at Hs.36137 gb:R99562_RC hepatocyte nuclear factor 3, gamma 0.3 0 0.001 204934_s_at Hs.823 gb:NM_002151.1 hepsin (transmembrane protease, serine 1) 0.3 0.005 0.001 221162_at Hs.285026 gb:NM_005712.1 HERV-H LTR-associating 1 0.2 0.003 0.001 227744_s_at Hs.303627 gb:AW005670_RC heterogeneous nuclear ribonucleoprotein D (AU-rich element RNA-binding protein 1, 37kD) 0.3 0.048 0.027 213619_at Hs.245710 gb:AV753392 heterogeneous nuclear ribonucleoprotein H1 (H) 0.4 0.013 0.012 200643_at Hs.177516 gb:NM_005336.1 high density lipoprotein binding protein (vigilin) 0.3 0 0.001 209192_x_at Hs.6364 gb:BC000166.2 HIV-1 Tat interactive protein, 60 kDa 0.4 0.002 0.005 212081_x_at Hs.25911 gb:AF129756 HLA-B associated transcript-2 0.4 0.023 0.007 208224_at Hs.99992 gb:NM_002144.1 homeo box B1 0.4 0.01 0.007 217080_s_at Hs.93564 gb:Y19026.1 Homer, neuronal immediate early gene, 2 0.3 0.008 0.002 217168_s_at Hs.146393 gb:AF217990.1 homocysteine-inducible, endoplasmic reticulum stress-inducible, ubiquitin-like domain member 1 0.3 0.003 0.001 205221_at Hs.15113 gb:NM_000187.1 homogentisate 1,2-dioxygenase (homogentisate oxidase) 0.4 0.003 0.005 217906_at Hs.20597 gb:NM_014315.1 host cell factor homolog 0.4 0.001 0.001 234460_at Hs.283101 gb:AF283901.1_RC HRPAP20 short form 0.3 0.023 0.007 201145_at Hs.15318 gb:NM_006118.2 HS1 binding protein 0.4 0 0.002 218026_at Hs.16059 gb:NM_014019.1 HSPC009 protein 0.3 0.001 0.001 217719_at Hs.119503 gb:NM_016091.1 HSPC025 0.3 0 0.001 217927_at Hs.11125 gb:NM_014041.1 HSPC033 protein 0.3 0.005 0.001 223207_x_at Hs.297214 gb:AF285119.1 HSPC141 protein 0.4 0.001 0.004 226462_at Hs.99291 gb:AW134979_RC HSPC156 protein 0.3 0.01 0.003 220994_s_at Hs.99291 gb:NM_014178.1 HSPC156 protein 0.3 0.001 0.003 219865_at Hs.279842 gb:NM_014179.1 HSPC157 protein 0.1 0 0.001 226229_s_at Hs.30026 gb:BF510732_RC HSPC182 protein 0.3 0 0.001 220597_s_at Hs.306208 gb:NM_018694.1 HSVI binding protein 0.4 0.003 0.001 215395_x_at Hs.302180 gb:U66061 Human germline T-cell receptor beta chain 0.1 0.002 0.001 209898_x_at Hs.330549 gb:U61167.1 Human SH3 domain-containing protein SH3P18 mRNA, complete cds 0.4 0.002 0.003 212246_at Hs.84775 gb:BE880828 Human transposon-like element mRNA 0.2 0 0.001 209559_at Hs.96731 gb:AB013384.1 huntingtin interacting protein-1-related 0.3 0.002 0.002 202105_at Hs.3631 gb:NM_001551.1 immunoglobulin (CD79A) binding protein 1 0.4 0 0.001 215420_at Hs.69351 gb:BE869172 Indian hedgehog (Drosophila) homolog 0.4 0.037 0.016 213544_at Hs.107153 gb:AI186701_RC inhibitor of growth family, member 1-like 0.4 0.032 0.027 211323_s_at Hs.198443 gb:L38019.1 inositol 1,4,5-triphosphate receptor, type 1 0.4 0.023 0.027 203126_at Hs.5753 gb:NM_014214.1 inositol(myo)-1(or 4)-monophosphatase 2 0.1 0.001 0.001 206598_at Hs.89832 gb:NM_000207.1 insulin 0.4 0.01 0.035
Page 6 of 15 Downregulated transcripts
Affymetrix code Unigene Code GenBank Code Name Fold Change t test p value MW U P value 201627_s_at Hs.56205 gb:NM_005542.1 insulin induced gene 1 0.3 0.01 0.001 213792_s_at Hs.89695 gb:AA485908_RC insulin receptor 0.2 0.014 0.004 201392_s_at Hs.76473 gb:BG031974 insulin-like growth factor 2 receptor 0.4 0.003 0.005 202718_at Hs.162 gb:NM_000597.1 insulin-like growth factor binding protein 2 (36kD) 0.1 0.001 0.001 208837_at Hs.179516 gb:BC000027.1 integral type I protein 0.2 0 0.001 205785_at Hs.172631 gb:J03925.1_RC integrin, alpha M 0.4 0.04 0.009 202147_s_at Hs.7879 gb:NM_001550.1 interferon-related developmental regulator 1 0.2 0.002 0.001 220056_at Hs.110915 gb:NM_021258.1 interleukin 22 receptor 0.2 0 0.001 206148_at Hs.172689 gb:NM_002183.1 interleukin 3 receptor, alpha (low affinity) 0.4 0.025 0.005 210045_at Hs.5337 gb:AU151428_RC isocitrate dehydrogenase 2 (NADP+), mitochondrial 0.4 0.029 0.012 216958_s_at Hs.77510 gb:AK022777.1 isovaleryl Coenzyme A dehydrogenase 0.4 0.031 0.021 220782_x_at Hs.159679 gb:NM_019598.1 kallikrein 12 0.3 0.002 0.001 210339_s_at Hs.181350 gb:BC005196.1 kallikrein 2, prostatic 0.1 0 0.001 206541_at Hs.1901 gb:NM_000892.2 kallikrein B, plasma (Fletcher factor) 1 0.4 0.02 0.016 203904_x_at Hs.323949 gb:NM_002231.2 kangai 1 0.3 0 0.002 211955_at Hs.113503 gb:NM_002271.1 karyopherin (importin) beta 3 0.4 0.004 0.005 202057_at Hs.169149 gb:BC002374.1 karyopherin alpha 1 (importin alpha 5) 0.3 0.002 0.004 212103_at Hs.301553 gb:BG403834 karyopherin alpha 6 (importin alpha 7) 0.4 0.048 0.035 203163_at Hs.275675 gb:NM_005886.1 katanin p80 (WD40-containing) subunit B 1 0.2 0.002 0.001 213287_s_at Hs.99936 gb:X14487 keratin 10 (epidermolytic hyperkeratosis; keratosis palmaris et plantaris) 0.4 0.011 0.007 212110_at Hs.89868 gb:D31887.1 KIAA0062 protein 0.3 0.001 0.001 202753_at Hs.23488 gb:NM_014814.1 KIAA0107 gene product 0.2 0.001 0.001 200616_s_at Hs.181418 gb:BC000371.1 KIAA0152 gene product 0.4 0.002 0.004 212310_at Hs.241552 gb:D87742.1 KIAA0268 protein 0.4 0 0.004 204757_s_at Hs.26899 gb:R61539_RC KIAA0285 gene product 0.3 0 0.001 231221_at Hs.23263 gb:AI553936_RC KIAA0350 protein 0.3 0.002 0.001 205689_at Hs.31438 gb:NM_014801.1 KIAA0435 gene product 0.4 0.001 0.005 212326_at Hs.194737 gb:AB007922.2 KIAA0453 protein 0.3 0.001 0.001 203959_s_at Hs.4236 gb:NM_014870.1 KIAA0478 gene product 0.3 0.004 0.003 201777_s_at Hs.62515 gb:BC002525.1 KIAA0494 gene product 0.4 0.003 0.005 213157_s_at Hs.16032 gb:BF115148_RC KIAA0523 protein 0.4 0.003 0.003 204074_s_at Hs.200595 gb:AI936976_RC KIAA0562 gene product 0.3 0.008 0.003 208959_s_at Hs.154023 gb:BC005374.1 KIAA0573 protein 0.4 0 0.001 205631_at Hs.77724 gb:NM_014749.1 KIAA0586 gene product 0.4 0.005 0.002 212778_at Hs.37656 gb:AL583340_RC KIAA0602 protein 0.3 0 0.001 206728_at Hs.129801 gb:NM_014693.1 KIAA0604 gene product 0.3 0.001 0.001 213842_x_at Hs.295112 gb:AK021688.1 KIAA0618 gene product 0.4 0.025 0.009 215407_s_at Hs.30898 gb:AK024064.1 KIAA0634 protein 0.3 0.005 0.005 212776_s_at Hs.6654 gb:AI978623_RC KIAA0657 protein 0.4 0.005 0.016 215151_at Hs.137579 gb:AB014594.2 KIAA0694 gene product 0.4 0.041 0.016 209244_s_at Hs.139648 gb:BE885926 KIAA0706 gene product 0.4 0.002 0.003 213099_at Hs.7285 gb:AB018302.1 KIAA0759 protein 0.2 0 0.001 214961_at Hs.22201 gb:AI818409_RC KIAA0774 protein 0.1 0.033 0.004 212495_at Hs.301011 gb:BE256900 KIAA0876 protein 0.4 0.017 0.021 202976_s_at Hs.188006 gb:NM_014899.1 KIAA0878 protein 0.2 0.004 0.003 209760_at Hs.37892 gb:AL136932.1 KIAA0922 protein 0.3 0.003 0.001 205594_at Hs.190386 gb:NM_014897.1 KIAA0924 protein 0.3 0.002 0.012 204903_x_at Hs.272586 gb:AL080168.1 KIAA0943 protein 0.3 0 0.001 203641_s_at Hs.300855 gb:BF002844_RC KIAA0977 protein 0.4 0 0.005 213111_at Hs.158135 gb:AB023198.1 KIAA0981 protein 0.4 0.028 0.012 213912_at Hs.11912 gb:AW134976_RC KIAA0984 protein 0.4 0.006 0.005 213280_at Hs.301552 gb:AK000478.1 KIAA1039 protein 0.3 0.002 0.002 204251_s_at Hs.18624 gb:NM_014956.1 KIAA1052 protein 0.3 0 0.001 208914_at Hs.155546 gb:BE646414_RC KIAA1080 protein; Golgi-associated, gamma-adaptin ear containing, ARF-binding protein 2 0.3 0 0.001 224676_at Hs.301226 gb:AI472339_RC KIAA1085 protein 0.3 0 0.001 215267_s_at Hs.143026 gb:AI127885_RC KIAA1087 protein 0.3 0 0.003 232059_at Hs.233023 gb:AI433419_RC KIAA1132 protein 0.4 0.045 0.035 224924_at Hs.131728 gb:BE205790_RC KIAA1140 protein 0.4 0.011 0.007 221868_at Hs.102657 gb:AB032981.1 KIAA1155 protein 0.1 0 0.001
Page 7 of 15 Downregulated transcripts
Affymetrix code Unigene Code GenBank Code Name Fold Change t test p value MW U P value 223831_x_at Hs.33122 gb:BC004442.1 KIAA1160 protein 0.3 0 0.001 226718_at Hs.12264 gb:AA001423_RC KIAA1163 protein 0.4 0.003 0.012 225726_s_at Hs.284157 gb:AB033026.1 KIAA1200 protein 0.4 0.013 0.003 212095_s_at Hs.7946 gb:BE552421_RC KIAA1288 protein 0.3 0.026 0.009 226248_s_at Hs.104696 gb:AI565067_RC KIAA1324 protein 0.1 0 0.001 232358_at Hs.186928 gb:AB037749.1 KIAA1328 protein 0.4 0.018 0.012 234324_at Hs.170162 gb:AK000699.1_RC KIAA1357 protein 0.4 0.018 0.027 228984_at Hs.32156 gb:AB037815.1 KIAA1394 protein 0.4 0.002 0.002 230112_at Hs.66159 gb:AB037820.1 KIAA1399 protein 0.2 0.001 0.001 229550_at Hs.267150 gb:AB037830.1 KIAA1409 protein 0.4 0.001 0.005 226492_at Hs.191098 gb:AL036088_RC KIAA1479 protein 0.4 0.032 0.021 228598_at Hs.91625 gb:AL538781_RC KIAA1492 protein 0.3 0.004 0.007 223365_at Hs.107382 gb:BC004463.1 KIAA1517 protein 0.4 0.049 0.021 229628_s_at Hs.113426 gb:AI831514_RC KIAA1529 protein 0.4 0.027 0.016 211389_x_at Hs.274601 gb:U73396.1 killer cell immunoglobulin-like receptor, three domains, long cytoplasmic tail, 1 0.4 0.04 0.012 209661_at Hs.23131 gb:BC001211.1 kinesin family member C3 0.4 0.028 0.009 208128_x_at gb:NM_030615.1 kinesin-like 3 isoform 1 0.4 0.012 0.005 212699_at Hs.7934 gb:BE222801_RC Kruppel-like factor 4 (gut) 0.3 0 0.001 201030_x_at Hs.234489 gb:NM_002300.1 lactate dehydrogenase B 0.4 0.002 0.001 213801_x_at Hs.181357 gb:AW304232_RC laminin receptor 1 (67kD, ribosomal protein SA) 0.4 0 0.001 229050_s_at Hs.82124 gb:AL533103 laminin, beta 1 0.4 0.008 0.007 211019_s_at Hs.93199 gb:D63807.1 lanosterol synthase (2,3-oxidosqualene-lanosterol cyclase) 0.3 0.014 0.003 208450_at Hs.113987 gb:NM_006498.1 lectin, galactoside-binding, soluble, 2 (galectin 2) 0.1 0.008 0.001 206268_at Hs.278239 gb:NM_020997.1 left-right determination, factor B 0.2 0.031 0.003 211562_s_at Hs.79386 gb:BC001755.1 leiomodin 1 (smooth muscle) 0.4 0.031 0.035 232018_at Hs.172407 gb:AI701895_RC LENG1 protein 0.4 0.032 0.035 220519_s_at Hs.162754 gb:NM_030657.1 lens intrinsic membrane protein 2 (19kD) 0.4 0.009 0.003 219019_at Hs.123136 gb:NM_018494.1 leucine rich repeat and death domain containing protein 0.1 0.002 0.001 204381_at Hs.143641 gb:NM_002333.1 low density lipoprotein receptor-related protein 3 0.4 0.001 0.003 217045_x_at Hs.194721 gb:AL136967 lymphocyte antigen 95 (activating NK-receptor ; NK-p44) 0.3 0.001 0.001 204674_at Hs.40202 gb:NM_006152.1 lymphoid-restricted membrane protein 0.3 0 0.001 201552_at Hs.150101 gb:NM_005561.2 lysosomal-associated membrane protein 1 0.4 0.001 0.002 205614_x_at Hs.278657 gb:NM_020998.1 macrophage stimulating 1 (hepatocyte growth factor-like) 0.3 0.006 0.001 212347_x_at Hs.102402 gb:AA831438_RC Mad4 homolog 0.3 0.003 0.001 203003_at Hs.111243 gb:AL530331_RC MADS box transcription enhancer factor 2, polypeptide D (myocyte enhancer factor 2D) 0.4 0.025 0.004 204663_at Hs.2838 gb:NM_006680.1 malic enzyme 3, NADP(+)-dependent, mitochondrial 0.4 0.015 0.012 218869_at Hs.150748 gb:NM_012213.1 malonyl-CoA decarboxylase 0.3 0 0.001 205105_at Hs.32965 gb:NM_002372.1 mannosidase, alpha, class 2A, member 1 0.3 0.01 0.003 209467_s_at Hs.5591 gb:BC002755.1 MAP kinase-interacting serinethreonine kinase 1 0.2 0.002 0.001 207123_s_at Hs.278489 gb:NM_003833.2 matrilin 4 0.3 0.001 0.002 225854_x_at Hs.279009 gb:W73718_RC matrix Gla protein 0.4 0 0.004 221953_s_at Hs.3743 gb:W45551_RC matrix metalloproteinase 24 (membrane-inserted) 0.4 0.002 0.007 219543_at Hs.16341 gb:NM_022129.1 MAWD binding protein 0.3 0 0.003 234897_s_at Hs.241587 gb:AF129756 megakaryocyte-enhanced gene transcript 1 protein 0.3 0.007 0.005 211596_s_at gb:AB050468.1 membrane glycoprotein LIG-1 0.2 0.006 0.001 203524_s_at Hs.74097 gb:NM_021126.1 mercaptopyruvate sulfurtransferase 0.4 0.001 0.005 212859_x_at Hs.74170 gb:BF217861 metallothionein 1E (functional) 0.3 0.002 0.002 217165_x_at Hs.203936 gb:M10943 metallothionein 1F (functional) 0.2 0.007 0.001 204745_x_at Hs.173451 gb:NM_005950.1 metallothionein 1G 0.2 0.006 0.001 206461_x_at Hs.2667 gb:NM_005951.1 metallothionein 1H 0.3 0.007 0.003 204326_x_at Hs.94360 gb:NM_002450.1 metallothionein 1L 0.3 0.028 0.005 208581_x_at Hs.278462 gb:NM_005952.1 metallothionein 1X 0.4 0.018 0.007 212185_x_at Hs.118786 gb:NM_005953.1 metallothionein 2A 0.4 0.028 0.012 205970_at Hs.73133 gb:NM_005954.1 metallothionein 3 (growth inhibitory factor (neurotrophic)) 0.4 0.007 0.005 203443_at Hs.173043 gb:AB012922 metastasis-associated 1-like 1 0.3 0.047 0.016 205813_s_at Hs.323715 gb:NM_000429.1 methionine adenosyltransferase I, alpha 0.4 0.014 0.012 200769_s_at Hs.77502 gb:NM_005911.1 methionine adenosyltransferase II, alpha 0.2 0 0.001 208595_s_at Hs.6211 gb:NM_015845.1 methyl-CpG binding domain protein 1 0.3 0.008 0.001 218440_at Hs.47649 gb:NM_020166.2 methylcrotonoyl-Coenzyme A carboxylase 1 (alpha) 0.4 0.005 0.001
Page 8 of 15 Downregulated transcripts
Affymetrix code Unigene Code GenBank Code Name Fold Change t test p value MW U P value 221588_x_at Hs.293970 gb:AI640855_RC methylmalonate-semialdehyde dehydrogenase 0.4 0.002 0.003 204397_at Hs.24178 gb:AF103939.1_RC microtubule-associated protein like echinoderm EMAP 0.4 0 0.007 214458_at Hs.152701 gb:AF230877.1 microtubule-interacting protein that associates with TRAF3; DKFZP434F124 protein 0.3 0.013 0.001 215916_at Hs.112028 gb:AL157418.1_RC MisshapenNIK-related kinase 0.3 0 0.001 213897_s_at Hs.3254 gb:AI832239_RC mitochondrial ribosomal protein L23 0.4 0.001 0.007 231274_s_at Hs.300496 gb:R92925_RC mitochondrial solute carrier 0.4 0.029 0.021 219822_at Hs.80683 gb:NM_004294.1 mitochondrial translational release factor 1 0.4 0 0.003 213177_at Hs.88500 gb:AB028989.1 mitogen-activated protein kinase 8 interacting protein 3 0.4 0.041 0.039 213490_s_at Hs.72241 gb:AI762811 mitogen-activated protein kinase kinase 2 0.4 0.003 0.009 209951_s_at Hs.110299 gb:AW007458_RC mitogen-activated protein kinase kinase 7 0.3 0.011 0.005 202788_at Hs.227789 gb:NM_004635.1 mitogen-activated protein kinase-activated protein kinase 3 0.4 0 0.002 207847_s_at Hs.89603 gb:NM_002456.1 mucin 1, transmembrane 0.4 0.035 0.012 212064_x_at Hs.7647 gb:AI471665_RC MYC-associated zinc finger protein (purine-binding transcription factor) 0.4 0.003 0.001 218966_at Hs.111782 gb:NM_018728.1 myosin 5C 0.4 0 0.001 227707_at Hs.20072 gb:AW027183_RC myosin regulatory light chain interacting protein 0.4 0.001 0.009 209742_s_at Hs.75535 gb:AF020768.1 myosin, light polypeptide 2, regulatory, cardiac, slow 0.4 0.007 0.002 205144_at Hs.170482 gb:L03785.1_RC myosin, light polypeptide 5, regulatory 0.4 0.001 0.005 204360_s_at Hs.50727 gb:NM_000263.1 N-acetylglucosaminidase, alpha- (Sanfilippo disease IIIB) 0.3 0.003 0.003 226959_at Hs.8087 gb:AL137430.1 NAG-5 protein 0.4 0.002 0.007 220307_at Hs.157872 gb:NM_016382.1 natural killer cell receptor 2B4 0.2 0.022 0.005 207279_s_at Hs.278999 gb:NM_016365.1 nebulette 0.4 0.019 0.021 211677_x_at gb:AF062733.2 nectin-like protein 1 0.3 0.015 0.009 220937_s_at Hs.3972 gb:NM_014403.1 NeuAc-alpha-2,3-Gal-beta-1,3-GalNAc-alpha-2, 6-sialyltransferase alpha2,6-sialyltransferase 0.4 0.006 0.021 208230_s_at Hs.172816 gb:NM_013960.1 neuregulin 1 0.2 0.001 0.001 200981_x_at Hs.113368 gb:NM_016592.1 neuroendocrine secretory protein 55 0.4 0.002 0.001 204105_s_at Hs.7912 gb:NM_005010.1 neuronal cell adhesion molecule 0.4 0.001 0.009 206699_x_at Hs.79564 gb:NM_002517.1 neuronal PAS domain protein 1 0.4 0.013 0.035 213479_at Hs.3281 gb:U26662.1 neuronal pentraxin II 0.4 0.014 0.016 204424_s_at Hs.301914 gb:AL050152.1 neuronal specific transcription factor DAT1 0.2 0.013 0.001 231739_at Hs.247993 gb:NM_030651.1 NG5 protein 0.3 0.019 0.007 217966_s_at Hs.48778 gb:NM_022083.1 niban protein 0.3 0.011 0.003 209459_s_at Hs.283675 gb:AF237813.1 NPD009 protein 0.1 0.002 0.001 230103_at Hs.184771 gb:BF515002_RC nuclear factor IC (CCAAT-binding transcription factor) 0.4 0.001 0.009 229207_x_at Hs.144904 gb:AI951724_RC nuclear receptor co-repressor 1 0.3 0 0.002 206410_at Hs.11930 gb:NM_021969.1 nuclear receptor subfamily 0, group B, member 2 0.3 0.004 0.007 211143_x_at Hs.1119 gb:D49728.1 nuclear receptor subfamily 4, group A, member 1 0.4 0.019 0.009 208337_s_at Hs.183123 gb:NM_003822.1 nuclear receptor subfamily 5, group A, member 2 0.3 0 0.003 203675_at Hs.3164 gb:NM_005013.1 nucleobindin 2 0.1 0 0.001 211949_s_at Hs.75337 gb:AI355279_RC nucleolar phosphoprotein p130 0.4 0.013 0.016 200874_s_at Hs.296585 gb:BE796327 nucleolar protein (KKED repeat) 0.4 0.015 0.007 229505_at Hs.78103 gb:AA939154_RC nucleosome assembly protein 1-like 4 0.4 0 0.004 204880_at Hs.1384 gb:NM_002412.1 O-6-methylguanine-DNA methyltransferase 0.3 0 0.001 221409_at Hs.283806 gb:NM_019897.1 olfactory receptor, family 2, subfamily S, member 2 0.4 0.009 0.012 234770_at Hs.278905 gb:X81445 olfactory receptor, family 51, subfamily A, member 1 pseudogene 0.4 0.011 0.012 217499_x_at Hs.332964 gb:AW874308_RC olfactory receptor, family 7, subfamily E, member 38 pseudogene 0.4 0.001 0.001 207563_s_at Hs.100293 gb:U77413.1 O-linked N-acetylglucosamine 0.4 0.001 0.001 215952_s_at Hs.125078 gb:AF090094.1 ornithine decarboxylase antizyme 1 0.3 0.001 0.002 200825_s_at Hs.277704 gb:NM_006389.2 oxygen regulated protein (150kD) 0.4 0.001 0.002 214607_at Hs.152663 gb:AW085556_RC p21 (CDKN1A)-activated kinase 3 0.4 0.005 0.005 223920_s_at Hs.160953 gb:AB045831.1 p53-regulated apoptosis-inducing protein 1 0.3 0.022 0.007 209230_s_at Hs.8603 gb:AF135266.1 p8 protein (candidate of metastasis 1) 0.1 0.002 0.001 205112_at Hs.6733 gb:NM_016341.1 pancreas-enriched phospholipase C 0.3 0 0.002 205912_at Hs.102876 gb:NM_000936.1 pancreatic lipase 0.1 0.001 0.001 211766_s_at gb:BC005989.1 pancreatic lipase-related protein 2 0.1 0.025 0.001 223216_x_at Hs.27410 gb:BC001237.1 papillomavirus regulatory factor PRF-1 0.4 0.009 0.016 202109_at Hs.75139 gb:NM_012402.1 partner of RAC1 (arfaptin 2) 0.4 0.003 0.009 227332_at Hs.102497 gb:BF511170_RC paxillin 0.2 0 0.001 202290_at Hs.278426 gb:NM_014891.1 PDGFA associated protein 1 0.3 0.013 0.005 205380_at Hs.15456 gb:NM_002614.1 PDZ domain containing 1 0.3 0 0.003
Page 9 of 15 Downregulated transcripts
Affymetrix code Unigene Code GenBank Code Name Fold Change t test p value MW U P value 219132_at Hs.44038 gb:NM_021255.1 pellino (Drosophila) homolog 2 0.3 0.002 0.001 214986_x_at Hs.93523 gb:U37221.1 peptidylprolyl isomerase (cyclophilin)-like 2 0.4 0.037 0.012 200967_at Hs.699 gb:NM_000942.1 peptidylprolyl isomerase B (cyclophilin B) 0.4 0.001 0.004 228469_at Hs.143482 gb:BF431902_RC peptidylprolyl isomerase D (cyclophilin D) 0.4 0 0.003 201490_s_at Hs.173125 gb:NM_005729.1 peptidylprolyl isomerase F (cyclophilin F) 0.4 0 0.002 202861_at Hs.68398 gb:NM_002616.1 period (Drosophila) homolog 1 0.2 0.03 0.003 205420_at Hs.79993 gb:NM_000288.1 peroxisomal biogenesis factor 7 0.4 0.019 0.009 218025_s_at Hs.15250 gb:NM_006117.1 peroxisomal D3,D2-enoyl-CoA isomerase 0.2 0.003 0.001 218021_at Hs.6318 gb:NM_021004.1 peroxisomal short-chain alcohol dehydrogenase 0.4 0 0.004 206352_s_at Hs.247220 gb:AB013818.1 peroxisome biogenesis factor 10 0.2 0.002 0.001 203244_at Hs.158084 gb:NM_000319.1 peroxisome receptor 1 0.4 0.003 0.005 202212_at Hs.13501 gb:NM_014303.1 pescadillo (zebrafish) homolog 1, containing BRCT domain 0.2 0.002 0.001 210617_at Hs.72874 gb:U87284.1 phosphate regulating gene with homologies to endopeptidases on the X chromosome 0.4 0.007 0.016 213651_at Hs.21492 gb:AI935720_RC phosphatidylinositol (4,5) bisphosphate 5-phosphatase, A 0.4 0.003 0.003 214151_s_at Hs.247118 gb:AU144243_RC phosphatidylinositol glycan, class B 0.4 0 0.003 219048_at Hs.108787 gb:NM_012327.1 phosphatidylinositol glycan, class N 0.3 0.002 0.001 210060_at Hs.1857 gb:M36476.1 phosphodiesterase 6G, cGMP-specific, rod, gamma 0.2 0.012 0.002 201397_at Hs.3343 gb:NM_006623.1 phosphoglycerate dehydrogenase 0.3 0 0.001 206311_s_at Hs.992 gb:NM_000928.1 phospholipase A2, group IB (pancreas) 0.1 0.001 0.001 222256_s_at Hs.198161 gb:AK000550.1 phospholipase A2, group IVB (cytosolic) 0.3 0.015 0.005 212680_x_at Hs.100623 gb:BE305165_RC phospholipase C, beta 3, neighbor pseudogene 0.4 0.002 0.007 205125_at Hs.80776 gb:NM_006225.1 phospholipase C, delta 1 0.4 0.011 0.004 202075_s_at Hs.283007 gb:NM_006227.1 phospholipid transfer protein 0.2 0.003 0.001 230352_at Hs.2910 gb:AI392908_RC phosphoribosyl pyrophosphate synthetase 2 0.4 0.021 0.009 220892_s_at Hs.286049 gb:NM_021154.1 phosphoserine aminotransferase 0.3 0 0.001 228972_at Hs.79709 gb:AI028602_RC phosphotidylinositol transfer protein 0.3 0.003 0.003 220683_at Hs.272405 gb:NM_015725.1 photoreceptor outer segment all-trans retinol dehydrogenase 0.4 0.013 0.012 203335_at Hs.172887 gb:NM_006214.1 phytanoyl-CoA hydroxylase (Refsum disease) 0.4 0.004 0.001 212914_at Hs.152151 gb:AV648364 plakophilin 4 0.3 0.004 0.002 228726_at Hs.234392 gb:AW512196_RC platelet-activating factor acetylhydrolase 2 (40kD) 0.4 0.011 0.005 205463_s_at Hs.37040 gb:NM_002607.1 platelet-derived growth factor alpha polypeptide 0.4 0.002 0.001 213030_s_at Hs.300622 gb:AI688418_RC plexin A2 0.2 0.014 0.001 217225_x_at Hs.227823 gb:AL512687.1 pM5 protein 0.3 0 0.001 201064_s_at Hs.169900 gb:NM_003819.2 poly(A)-binding protein, cytoplasmic 4 (inducible form) 0.3 0 0.001 228854_at Hs.117176 gb:AI492388_RC poly(A)-binding protein, nuclear 1 0.3 0.046 0.027 224152_s_at Hs.44143 gb:AF225872.1 polybromo 1 0.4 0.018 0.016 214527_s_at Hs.30570 gb:AB041836.1 polyglutamine binding protein 1 0.4 0.003 0.012 203422_at Hs.99890 gb:NM_002691.1 polymerase (DNA directed), delta 1, catalytic subunit (125kD) 0.4 0.001 0.007 201115_at Hs.74598 gb:NM_006230.1 polymerase (DNA directed), delta 2, regulatory subunit (50kD) 0.4 0.006 0.004 213887_s_at Hs.24301 gb:AI554759 polymerase (RNA) II (DNA directed) polypeptide E (25kD) 0.4 0.002 0.007 212782_x_at Hs.80475 gb:BG335629 polymerase (RNA) II (DNA directed) polypeptide J (13.3kD) 0.4 0.014 0.009 211730_s_at gb:BC005903.1 polymerase (RNA) II (DNA directed) polypeptide L(7.6kD) 0.4 0.006 0.002 212016_s_at Hs.172550 gb:AA679988_RC polypyrimidine tract binding protein (heterogeneous nuclear ribonucleoprotein I) 0.4 0.048 0.016 205952_at Hs.24040 gb:NM_002246.1 potassium channel, subfamily K, member 3 (TASK-1) 0.2 0 0.001 211806_s_at Hs.17287 gb:D87291.1 potassium inwardly-rectifying channel, subfamily J, member 15 0.4 0 0.001 219564_at Hs.50151 gb:NM_018658.1 potassium inwardly-rectifying channel, subfamily J, member 16 0.4 0.004 0.007 211304_x_at Hs.193044 gb:D50134.1 potassium inwardly-rectifying channel, subfamily J, member 5 0.4 0.005 0.009 205304_s_at Hs.102308 gb:NM_004982.1 potassium inwardly-rectifying channel, subfamily J, member 8 0.3 0.025 0.009 211301_at Hs.184889 gb:AF120491.1 potassium voltage-gated channel, Shal-related subfamily, member 3 0.3 0.007 0.004 210263_at Hs.23735 gb:AF029780.1 potassium voltage-gated channel, subfamily F, member 1 0.3 0 0.001 223726_at Hs.64064 gb:AB022696.1 potassium voltage-gated channel, subfamily H (eag-related), member 3 0.3 0 0.001 216329_at Hs.2815 gb:Z21967.1_RC POU domain, class 6, transcription factor 1 0.3 0.02 0.005 221009_s_at Hs.9613 gb:NM_016109.1 PPAR(gamma) angiopoietin related protein 0.3 0.019 0.003 220928_s_at Hs.302022 gb:NM_022114.1 PR domain containing 16 0.4 0.007 0.007 205253_at Hs.155691 gb:NM_002585.1 pre-B-cell leukemia transcription factor 1 0.3 0.049 0.012 205360_at Hs.91161 gb:AI718295_RC prefoldin 4 0.3 0.006 0.005 207132_x_at Hs.288856 gb:NM_002624.1 prefoldin 5 0.3 0 0.001 219475_at Hs.31773 gb:NM_013370.1 pregnancy-induced growth inhibitor 0.2 0.001 0.001 204262_s_at Hs.25363 gb:NM_000447.1 presenilin 2 (Alzheimer disease 4) 0.3 0.011 0.006
Page 10 of 15 Downregulated transcripts
Affymetrix code Unigene Code GenBank Code Name Fold Change t test p value MW U P value 220858_at Hs.278940 gb:NM_014133.1 PRO0618 protein 0.1 0.004 0.001 220612_at Hs.8345 gb:NM_014135.1 PRO0641 protein 0.4 0.001 0.009 211218_at Hs.156243 gb:AF118078.1 PRO1848 protein 0.4 0.038 0.027 207743_at Hs.278926 gb:NM_014104.1 PRO1880 protein 0.4 0.014 0.005 200656_s_at Hs.75655 gb:NM_000918.1 procollagen-proline, 2-oxoglutarate 4-dioxygenase 0.1 0.002 0.001 202730_s_at Hs.296251 gb:NM_014456.1 programmed cell death 4 0.3 0 0.001 219275_at Hs.166468 gb:NM_004708.1 programmed cell death 5 0.4 0 0.001 209385_s_at Hs.301959 gb:AL136616.1 proline synthetase co-transcribed (bacterial homolog) 0.3 0.002 0.005 211223_at Hs.158301 gb:AF076215.1 prophet of Pit1, paired-like homeodomain transcription factor 0.3 0.005 0.007 210834_s_at Hs.170917 gb:D38299.1 prostaglandin E receptor 3 (subtype EP3) 0.4 0.018 0.027 204897_at Hs.199248 gb:AA897516_RC prostaglandin E receptor 4 (subtype EP4) 0.2 0.009 0.002 204394_at Hs.18910 gb:NM_003627.1 prostate cancer overexpressed gene 1 0.1 0 0.001 205353_s_at Hs.80423 gb:NM_002567.1 prostatic binding protein 0.4 0.027 0.016 205869_at Hs.241395 gb:NM_002769.1 protease, serine, 1 (trypsin 1) 0.1 0.001 0.001 208165_s_at Hs.274407 gb:NM_005865.1 protease, serine, 16 (thymus) 0.4 0.01 0.016 205402_x_at Hs.241561 gb:NM_002770.1 protease, serine, 2 (trypsin 2) 0.1 0.002 0.001 207463_x_at Hs.278310 gb:NM_002771.1 protease, serine, 3 (trypsin 3) 0.1 0 0.001 213421_x_at Hs.58247 gb:AW007273_RC protease, serine, 4 (trypsin 4, brain) 0.1 0.001 0.001 206691_s_at Hs.66581 gb:NM_006849.1 protein disulfide isomerase 0.1 0.01 0.001 208658_at Hs.93659 gb:BC000425.1 protein disulfide isomerase related protein (calcium-binding protein, intestinal-related) 0.4 0.005 0.004 207668_x_at Hs.182429 gb:NM_005742.1 protein disulfide isomerase-related protein 0.4 0 0.001 201651_s_at Hs.18842 gb:NM_007229.1 protein kinase C and casein kinase substrate in neurons 2 0.2 0.001 0.001 200707_at Hs.1432 gb:NM_002743.1 protein kinase C substrate 80K-H 0.3 0.02 0.003 206248_at Hs.211592 gb:NM_005400.1 protein kinase C, epsilon 0.4 0.013 0.012 204843_s_at Hs.289075 gb:NM_004157.1 protein kinase, cAMP-dependent, regulatory, type II, alpha 0.4 0.037 0.027 203680_at Hs.77439 gb:NM_002736.1 protein kinase, cAMP-dependent, regulatory, type II, beta 0.4 0.002 0.007 200913_at Hs.17883 gb:NM_002707.1 protein phosphatase 1G (formerly 2C), magnesium-dependent, gamma isoform 0.3 0.001 0.003 202883_s_at Hs.108705 gb:T79584_RC protein phosphatase 2 (formerly 2A), regulatory subunit A (PR 65), beta isoform 0.3 0.002 0.002 228025_s_at Hs.279909 gb:AW955803_RC protein phosphatase 2 (formerly 2A), regulatory subunit B (PR 52), gamma isoform 0.4 0.022 0.007 203133_at Hs.77028 gb:NM_006808.1 protein translocation complex beta 0.3 0.002 0.001 206574_s_at Hs.43666 gb:NM_007079.1 protein tyrosine phosphatase type IVA, member 3 0.3 0.016 0.005 203029_s_at Hs.74624 gb:NM_002847.1 protein tyrosine phosphatase, receptor type, N polypeptide 2 0.2 0.001 0.001 208034_s_at Hs.1011 gb:NM_003891.1 protein Z, vitamin K-dependent plasma glycoprotein 0.3 0.016 0.004 207341_at Hs.928 gb:NM_002777.2 proteinase 3 (serine proteinase, neutrophil, Wegener granulomatosis autoantigen) 0.2 0.025 0.012 234029_at Hs.283802 gb:AF152529.1_RC protocadherin gamma subfamily B, 8 pseudogene 0.4 0.017 0.009 228443_s_at Hs.111988 gb:AI127452_RC PRSET domain containing protein 07 0.4 0 0.001 217780_at Hs.108969 gb:NM_016145.1 PTD008 protein 0.4 0.004 0.001 210401_at Hs.41735 gb:U45448.1 purinergic receptor P2X, ligand-gated ion channel, 1 0.1 0 0.001 224069_x_at Hs.258580 gb:AF260426.1 purinergic receptor P2X, ligand-gated ion channel, 2 0.4 0 0.007 204088_at Hs.321709 gb:NM_002560.1 purinergic receptor P2X, ligand-gated ion channel, 4 0.4 0.002 0.007 218608_at Hs.128866 gb:NM_022089.1 putative ATPase 0.2 0.002 0.003 205960_at Hs.299221 gb:NM_002612.1 pyruvate dehydrogenase kinase, isoenzyme 4 0.1 0.017 0.001 209123_at Hs.75438 gb:BC000576.1 quinoid dihydropteridine reductase 0.4 0.034 0.016 203136_at Hs.11417 gb:NM_006423.1 Rab acceptor 1 (prenylated) 0.4 0 0.003 208730_x_at Hs.78305 gb:AA535244_RC RAB2, member RAS oncogene family 0.3 0.001 0.002 219562_at Hs.3797 gb:NM_014353.1 RAB26, member RAS oncogene family 0.1 0.001 0.001 210127_at Hs.277445 gb:BC002510.1 RAB6B, member RAS oncogene family 0.4 0.031 0.027 221614_s_at Hs.198551 gb:BC005153.1 rabphilin 3A-like (without C2 domains) 0.4 0 0.005 212127_at Hs.183800 gb:BE379408 Ran GTPase activating protein 1 0.4 0.001 0.004 214487_s_at Hs.239527 gb:NM_002886.1 RAP2B, member of RAS oncogene family 0.3 0.022 0.009 204916_at Hs.32989 gb:NM_005855.1 receptor (calcitonin) activity modifying protein 1 0.4 0 0.001 205326_at Hs.25691 gb:NM_005856.1 receptor (calcitonin) activity modifying protein 3 0.4 0.002 0.002 221377_s_at Hs.248217 gb:NM_014276.1 recombining binding protein suppressor of hairless-like (Drosophila) 0.1 0.001 0.001 209752_at Hs.1032 gb:AF172331.1 regenerating islet-derived 1 alpha (pancreatic stone protein, pancreatic thread protein) 0.1 0.001 0.001 205886_at Hs.4158 gb:NM_006507.1 regenerating islet-derived 1 beta (pancreatic stone protein, pancreatic thread protein) 0.2 0.025 0.012 210751_s_at Hs.77854 gb:D31815.1 regucalcin (senescence marker protein-30) 0.4 0.005 0.004 202388_at Hs.78944 gb:NM_002923.1 regulator of G-protein signalling 2, 24kD 0.3 0.041 0.016 206518_s_at Hs.117149 gb:NM_003835.1 regulator of G-protein signalling 9 0.4 0.017 0.012 208021_s_at gb:NM_002913.1 replication factor C (activator 1) 1 (145kD) 0.4 0 0.001
Page 11 of 15 Downregulated transcripts
Affymetrix code Unigene Code GenBank Code Name Fold Change t test p value MW U P value 209496_at Hs.37682 gb:BC000069.1 retinoic acid receptor responder (tazarotene induced) 2 0.1 0.001 0.001 203423_at Hs.101850 gb:NM_002899.2 retinol-binding protein 1, cellular 0.2 0.008 0.001 216317_x_at Hs.278994 gb:X63095.1 Rhesus blood group, CcEe antigens 0.4 0.003 0.021 206455_s_at Hs.247565 gb:NM_000539.2 rhodopsin (opsin 2, rod pigment) (retinitis pigmentosa 4, autosomal dominant) 0.3 0.046 0.012 217983_s_at Hs.8297 gb:NM_003730.2 ribonuclease 6 precursor 0.3 0.023 0.005 201476_s_at Hs.2934 gb:AI692974_RC ribonucleotide reductase M1 polypeptide 0.4 0.001 0.005 200725_x_at Hs.29797 gb:NM_006013.1 ribosomal protein L10 0.4 0 0.001 200036_s_at Hs.252574 gb:NM_007104.2 ribosomal protein L10a 0.4 0.004 0.002 200010_at Hs.179943 gb:NM_000975.1 ribosomal protein L11 0.4 0.008 0.005 200088_x_at Hs.182979 gb:AK026491.1 ribosomal protein L12 0.4 0 0.002 212933_x_at Hs.180842 gb:AA961748_RC ribosomal protein L13 0.3 0 0.001 200716_x_at Hs.119122 gb:NM_012423.1 ribosomal protein L13a 0.3 0 0.001 200074_s_at Hs.738 gb:U16738.1 ribosomal protein L14 0.4 0.003 0.001 200038_s_at Hs.82202 gb:NM_000985.1 ribosomal protein L17 0.4 0 0.001 200022_at Hs.75458 gb:NM_000979.1 ribosomal protein L18 0.3 0 0.001 200869_at Hs.163593 gb:NM_000980.1 ribosomal protein L18a 0.3 0 0.001 200029_at Hs.252723 gb:NM_000981.1 ribosomal protein L19 0.4 0.003 0.001 227755_at Hs.184108 gb:AA042983_RC ribosomal protein L21 0.2 0.005 0.002 203034_s_at Hs.76064 gb:NM_000990.1 ribosomal protein L27a 0.4 0 0.001 200823_x_at Hs.183698 gb:NM_000992.1 ribosomal protein L29 0.2 0 0.001 201217_x_at Hs.119598 gb:NM_000967.1 ribosomal protein L3 0.3 0 0.001 200002_at Hs.182825 gb:NM_007209.1 ribosomal protein L35 0.3 0 0.001 219762_s_at Hs.300759 gb:NM_015414.1 ribosomal protein L36 0.4 0.03 0.003 201154_x_at Hs.286 gb:NM_000968.1 ribosomal protein L4 0.3 0 0.001 200937_s_at Hs.180946 gb:NM_000969.1 ribosomal protein L5 0.4 0 0.001 200034_s_at Hs.174131 gb:NM_000970.1 ribosomal protein L6 0.4 0 0.001 212042_x_at Hs.153 gb:BG389744 ribosomal protein L7 0.4 0 0.001 217740_x_at Hs.99858 gb:NM_000972.1 ribosomal protein L7a 0.3 0 0.001 200936_at Hs.178551 gb:NM_000973.1 ribosomal protein L8 0.3 0 0.001 200817_x_at Hs.76230 gb:NM_001014.1 ribosomal protein S10 0.4 0.001 0.001 200031_s_at Hs.182740 gb:NM_001015.1 ribosomal protein S11 0.4 0 0.001 213377_x_at Hs.285405 gb:AI799007_RC ribosomal protein S12 0.4 0 0.001 208645_s_at Hs.244621 gb:AF116710.1 ribosomal protein S14 0.3 0 0.001 200819_s_at Hs.133230 gb:NM_001018.1 ribosomal protein S15 0.4 0 0.001 201665_x_at Hs.5174 gb:NM_001021.1 ribosomal protein S17 0.4 0 0.001 201049_s_at Hs.275865 gb:NM_022551.1 ribosomal protein S18 0.4 0 0.001 203107_x_at Hs.182426 gb:NM_002952.1 ribosomal protein S2 0.3 0 0.001 230199_at Hs.539 gb:BF438512_RC ribosomal protein S29 0.3 0.026 0.003 201257_x_at Hs.77039 gb:NM_001006.1 ribosomal protein S3A 0.4 0 0.001 200933_x_at Hs.108124 gb:NM_001007.1 ribosomal protein S4, X-linked 0.4 0 0.001 200024_at Hs.76194 gb:NM_001009.1 ribosomal protein S5 0.3 0.001 0.001 201254_x_at Hs.241507 gb:NM_001010.1 ribosomal protein S6 0.3 0 0.001 200082_s_at Hs.301547 gb:AI805587_RC ribosomal protein S7 0.3 0 0.001 217747_s_at Hs.180920 gb:NM_001013.1 ribosomal protein S9 0.2 0 0.001 200909_s_at Hs.119500 gb:NM_001004.1 ribosomal protein, large P2 0.4 0.003 0.005 201033_x_at Hs.73742 gb:NM_001002.1 ribosomal protein, large, P0 0.4 0 0.001 201206_s_at Hs.98614 gb:NM_004587.1 ribosome binding protein 1 (dog 180kD homolog) 0.1 0.001 0.001 201394_s_at Hs.201675 gb:U23946.1 RNA binding motif protein 5 0.3 0 0.001 228030_at Hs.173993 gb:AI041522_RC RNA binding motif protein 6 0.3 0.002 0.004 213629_x_at Hs.8765 gb:BF246115 RNA helicase-related protein 0.2 0.003 0.001 218258_at Hs.106127 gb:NM_015972.1 RNA polymerase I 16 kDa subunit 0.4 0.047 0.003 200006_at Hs.10958 gb:NM_007262.1 RNA-binding protein regulatory subunit 0.4 0.002 0.004 216976_s_at Hs.79350 gb:X96588.1 RYK receptor-like tyrosine kinase 0.4 0 0.002 205942_s_at Hs.181345 gb:NM_005622.1 SA (rat hypertension-associated) homolog 0.3 0 0.003 200903_s_at Hs.172673 gb:NM_000687.1 S-adenosylhomocysteine hydrolase 0.4 0.01 0.005 217716_s_at Hs.306079 gb:NM_013336.1 sec61 homolog 0.2 0.001 0.001 203317_at Hs.110121 gb:NM_012455.1 SEC7 homolog 0.4 0 0.001 205697_at Hs.116428 gb:NM_006998.1 secretagogin 0.2 0.001 0.002 207468_s_at Hs.279565 gb:NM_003015.2 secreted frizzled-related protein 5 0.3 0 0.003
Page 12 of 15 Downregulated transcripts
Affymetrix code Unigene Code GenBank Code Name Fold Change t test p value MW U P value 202062_s_at Hs.181300 gb:NM_005065.1 sel-1 (suppressor of lin-12, C.elegans)-like 0.2 0 0.001 217977_at Hs.279623 gb:NM_016332.1 selenoprotein X, 1 0.1 0 0.001 218122_s_at Hs.3355 gb:NM_021627.1 sentrin-specific protease 0.4 0.011 0.007 203458_at Hs.301540 gb:AI951454_RC sepiapterin reductase (7,8-dihydrobiopterin:NADP+ oxidoreductase) 0.4 0.001 0.002 201224_s_at Hs.18192 gb:AU147713_RC SerArg-related nuclear matrix protein (plenty of prolines 101-like) 0.4 0.002 0.005 209443_at Hs.76353 gb:J02639.1 serine (or cysteine) proteinase inhibitor, clade A (alpha-1 antiproteinase, antitrypsin), member 5 0.2 0.006 0.001 206325_at Hs.1305 gb:NM_001756.2 serine (or cysteine) proteinase inhibitor, clade A (alpha-1 antiproteinase, antitrypsin), member 6 0.4 0.031 0.012 211474_s_at Hs.41072 gb:BC004948.1 serine (or cysteine) proteinase inhibitor, clade B (ovalbumin), member 6 0.4 0.007 0.004 205352_at Hs.78589 gb:NM_005025.1 serine (or cysteine) proteinase inhibitor, clade I (neuroserpin), member 1 0.2 0.008 0.001 206239_s_at Hs.181286 gb:NM_003122.1 serine protease inhibitor, Kazal type 1 0.1 0 0.001 223715_at Hs.170819 gb:AF020089.1 serinethreonine kinase 29 0.2 0.007 0.001 227217_at Hs.132792 gb:AI637586_RC serologically defined colon cancer antigen 43 0.2 0 0.001 201739_at Hs.296323 gb:NM_005627.1 serumglucocorticoid regulated kinase 0.2 0.005 0.001 202475_at Hs.9234 gb:NM_006326.1 seven transmembrane domain protein 0.4 0.004 0.001 222477_s_at Hs.10071 gb:BC005176.1 seven transmembrane protein TM7SF3 0.4 0.002 0.009 221216_s_at Hs.57475 gb:NM_012236.1 sex comb on midleg homolog 1 0.4 0.002 0.004 211210_x_at Hs.151544 gb:AF100539.1 SH2 domain protein 1A, Duncans disease (lymphoproliferative syndrome) 0.4 0.016 0.009 204979_s_at Hs.47438 gb:NM_007341.1 SH3 domain binding glutamic acid-rich protein 0.3 0.041 0.007 208322_s_at Hs.301698 gb:NM_003033.1 sialyltransferase 4A (beta-galactosidase alpha-2,3-sialytransferase) 0.4 0.047 0.027 200918_s_at Hs.75730 gb:NM_003139.1 signal recognition particle receptor (docking protein) 0.2 0 0.001 200652_at Hs.74564 gb:NM_003145.2 signal sequence receptor, beta (translocon-associated protein beta) 0.3 0 0.001 201004_at Hs.102135 gb:NM_006280.1 signal sequence receptor, delta (translocon-associated protein delta) 0.1 0 0.001 217790_s_at Hs.28707 gb:NM_007107.1 signal sequence receptor, gamma (translocon-associated protein gamma) 0.4 0.008 0.005 202781_s_at Hs.178347 gb:AI806031_RC SKIP for skeletal muscle and kidney enriched inositol phosphatase 0.4 0 0.001 208221_s_at Hs.133466 gb:NM_003061.1 slit (Drosophila) homolog 1 0.2 0.001 0.001 223539_s_at Hs.32567 gb:AF073518.1 small EDRK-rich factor 1A (telomeric) 0.4 0 0.001 206042_x_at Hs.58606 gb:NM_022804.1 SNRPN upstream reading frame 0.3 0.001 0.001 208389_s_at Hs.380 gb:NM_004171.1 solute carrier family 1 (glial high affinity glutamate transporter), member 2 0.3 0 0.001 209610_s_at Hs.323878 gb:BF340083 solute carrier family 1 (glutamateneutral amino acid transporter), member 4 0.4 0.004 0.003 203125_x_at Hs.57435 gb:AF046997.1 solute carrier family 11 (proton-coupled divalent metal ion transporters), member 2 0.4 0.003 0.007 208354_s_at Hs.158462 gb:NM_000339.1 solute carrier family 12 (sodiumchloride transporters), member 3 0.2 0.028 0.004 207051_at Hs.128827 gb:NM_005495.1 solute carrier family 17 (sodium phosphate), member 4 0.4 0.004 0.009 229239_x_at Hs.235782 gb:AW574753_RC solute carrier family 21 (organic anion transporter), member 12 0.4 0.032 0.021 220554_at Hs.251395 gb:NM_006672.1 solute carrier family 22 (organic anion transporter), member 7 0.4 0.013 0.021 221298_s_at Hs.266223 gb:NM_004254.1 solute carrier family 22 (organic anion transporter), member 8 0.3 0.001 0.003 212085_at Hs.164280 gb:AA916851_RC solute carrier family 25 (mitochondrial carrier; adenine nucleotide translocator), member 6 0.2 0 0.002 218653_at Hs.78457 gb:NM_014252.1 solute carrier family 25 (mitochondrial carrier; ornithine transporter) member 15 0.2 0.003 0.001 205799_s_at Hs.239106 gb:M95548.1 solute carrier family 3 0.3 0.001 0.004 208177_at Hs.936 gb:NM_003052.1 solute carrier family 34 (sodium phosphate), member 1 0.4 0.005 0.007 203908_at Hs.5462 gb:NM_003759.1 solute carrier family 4, sodium bicarbonate cotransporter, member 4 0.2 0 0.002 226904_at Hs.187958 gb:AI820043_RC solute carrier family 6 (neurotransmitter transporter, creatine), member 8 0.2 0.023 0.004 202104_s_at Hs.296847 gb:NM_003119.1 spastic paraplegia 7, paraplegin (pure and complicated autosomal recessive) 0.4 0 0.003 229952_at Hs.47431 gb:AI936724_RC spectrin, beta, erythrocytic (includes spherocytosis, clinical type I) 0.4 0.002 0.009 228246_s_at Hs.107164 gb:AA772306_RC spectrin, beta, non-erythrocytic 1 0.2 0.001 0.003 219888_at Hs.123159 gb:NM_003116.1 sperm associated antigen 4 0.2 0.001 0.001 201516_at Hs.76244 gb:NM_003132.1 spermidine synthase 0.4 0.02 0.001 221032_s_at gb:NM_030770.1 spinesin 0.3 0.042 0.035 209964_s_at Hs.108447 gb:AF032105.1 spinocerebellar ataxia 7 (olivopontocerebellar atrophy with retinal degeneration) 0.3 0.019 0.001 232392_at Hs.167460 gb:BE927772 splicing factor, arginineserine-rich 3 0.3 0.031 0.012 225430_at Hs.77608 gb:AA541697_RC splicing factor, arginineserine-rich 9 0.4 0.004 0.004 221519_at Hs.24307 gb:AF281859.1 split handfoot malformation (ectrodactyly) type 3 0.4 0 0.001 203439_s_at Hs.155223 gb:BC000658.1 stanniocalcin 2 0.3 0.001 0.001 231038_s_at Hs.108689 gb:AI309438_RC sterol regulatory element binding transcription factor 2 0.4 0 0.005 233049_x_at Hs.25197 gb:AF217968.1 STIP1 homology and U-Box containing protein 1 0.4 0.002 0.007 217934_x_at Hs.25197 gb:NM_005861.1 STIP1 homology and U-Box containing protein 1 0.3 0 0.001 200970_s_at Hs.76698 gb:AL136807.1 stress-associated endoplasmic reticulum protein 1; ribosome associated membrane protein 4 0.3 0 0.001 207983_s_at Hs.8217 gb:NM_006603.1 stromal antigen 2 0.4 0.043 0.021 209687_at Hs.237356 gb:U19495.1 stromal cell-derived factor 1 0.3 0.001 0.007 218681_s_at Hs.303116 gb:NM_022044.1 stromal cell-derived factor 2-like 1 0.3 0.001 0.001
Page 13 of 15 Downregulated transcripts
Affymetrix code Unigene Code GenBank Code Name Fold Change t test p value MW U P value 229820_at Hs.469 gb:BF509179_RC succinate dehydrogenase complex, subunit A, flavoprotein (Fp) 0.4 0.033 0.021 216906_at Hs.56937 gb:U20428.1 suppression of tumorigenicity 14 (colon carcinoma, matriptase, epithin) 0.3 0.024 0.009 224839_s_at Hs.79265 gb:BG328998 suppression of tumorigenicity 5 0.1 0.004 0.001 207871_s_at Hs.5814 gb:NM_018412.2 suppression of tumorigenicity 7 0.3 0.007 0.004 214387_x_at Hs.1074 gb:AA633841_RC surfactant, pulmonary-associated protein C 0.3 0.031 0.001 214199_at Hs.253495 gb:NM_003019.1 surfactant, pulmonary-associated protein D 0.3 0.038 0.027 208793_x_at Hs.78202 gb:AI744900_RC SWISNF related, matrix associated, actin dependent regulator of chromatin, subfamily a, member 4 0.3 0.001 0.005 201827_at Hs.250581 gb:AF113019.1 SWISNF related, matrix associated, actin dependent regulator of chromatin, subfamily d, member 2 0.3 0 0.001 210613_s_at Hs.6139 gb:BC000731.1 synaptogyrin 1 0.3 0 0.001 212154_at Hs.1501 gb:AI380298_RC syndecan 2 (heparan sulfate proteoglycan 1, cell surface-associated, fibroglycan) 0.3 0 0.001 208584_at Hs.279002 gb:NM_016432.1 synoretin 0.4 0.035 0.004 221499_s_at Hs.102178 gb:AK026970.1 syntaxin 16 0.4 0.015 0.004 216470_x_at Hs.303157 gb:AF009664 T cell receptor beta locus 0.1 0.001 0.001 229565_x_at Hs.267182 gb:N29712_RC T-box 3 (ulnar mammary syndrome) 0.4 0.046 0.009 221624_at Hs.144519 gb:AF195821.1 T-cell leukemialymphoma 6 0.4 0.041 0.027 204158_s_at Hs.46465 gb:NM_006019.1 T-cell, immune regulator 1 0.3 0 0.001 210323_at Hs.127111 gb:AB033823.1 tektin 2 (testicular) 0.4 0.038 0.005 217413_s_at Hs.169886 gb:X71923.1 tenascin XB 0.3 0.025 0.009 221429_x_at gb:NM_031274.1 testis expressed sequence 13A 0.2 0.001 0.001 218486_at Hs.12229 gb:AA149594_RC TGFB inducible early growth response 2 0.4 0 0.001 201923_at Hs.83383 gb:NM_006406.1 thioredoxin peroxidase (antioxidant enzyme) 0.1 0 0.001 207555_s_at Hs.89887 gb:U27325.1 thromboxane A2 receptor 0.3 0 0.001 204100_at Hs.724 gb:NM_003250.1 thyroid hormone receptor, alpha (avian erythroblastic leukemia viral (v-erb-a) oncogene homolog) 0.4 0.049 0.016 208195_at Hs.172004 gb:NM_003319.1 titin 0.4 0.003 0.007 226870_at Hs.111110 gb:AI924025_RC titin-cap (telethonin) 0.3 0 0.001 217930_s_at Hs.25413 gb:NM_019009.1 TOLLIP protein 0.4 0.003 0.005 214299_at Hs.91175 gb:AI676092_RC topoisomerase (DNA) III alpha 0.3 0.028 0.007 212758_s_at Hs.232068 gb:AI373166_RC transcription factor 8 (represses interleukin 2 expression) 0.4 0.006 0.021 202704_at Hs.178137 gb:AA675892_RC transducer of ERBB2, 1 0.3 0.014 0.005 228834_at Hs.178137 gb:BF240286 transducer of ERBB2, 1 0.2 0.028 0.005 208700_s_at Hs.89643 gb:L12711.1 transketolase (Wernicke-Korsakoff syndrome) 0.3 0.001 0.001 218188_s_at Hs.23410 gb:NM_012458.1 translocase of inner mitochondrial membrane 13 (yeast) homolog B 0.4 0 0.004 210130_s_at Hs.31130 gb:AF096304.1 transmembrane 7 superfamily member 2 0.3 0.025 0.005 205102_at Hs.318545 gb:NM_005656.1 transmembrane protease, serine 2 0.3 0 0.001 200929_at Hs.74137 gb:NM_006827.1 transmembrane trafficking protein 0.2 0 0.001 221270_s_at gb:NM_031209.1 tRNA-guanine transglycosylase 0.2 0 0.001 218766_s_at Hs.227274 gb:NM_015836.1 tryptophanyl tRNA synthetase 2 (mitochondrial) 0.4 0.017 0.021 216520_s_at Hs.279860 gb:AF072098 tumor protein, translationally-controlled 1 0.4 0 0.001 223458_at Hs.6314 gb:BC000567.1 type I transmembrane receptor (seizure-related protein) 0.4 0.003 0.007 204079_at Hs.26350 gb:NM_003595.1 tyrosylprotein sulfotransferase 2 0.1 0.001 0.001 221700_s_at gb:AF348700.1 ubiquitin A-52 residue ribosomal protein fusionproduct 1 0.4 0.001 0.001 218837_s_at Hs.19196 gb:NM_015983.1 ubiquitin-conjugating enzyme HBUCE1 0.4 0.009 0.009 220757_s_at Hs.11081 gb:NM_025241.1 UBX domain-containing 2 0.4 0.004 0.003 229451_at Hs.301062 gb:AW294162_RC UDP-N-acetyl-alpha-D-galactosamine:polypeptide N-acetylgalactosaminyltransferase 9 (GalNAc-T9) 0.4 0.001 0.003 219869_s_at Hs.284205 gb:NM_022154.1 up-regulated by BCG-CWS 0.2 0 0.001 213023_at Hs.251967 gb:NM_007124.1 utrophin (homologous to dystrophin) 0.4 0.009 0.004 227770_at Hs.234839 gb:AI949559_RC vacuolar sorting protein 4 0.3 0.003 0.007 218807_at Hs.267659 gb:NM_006113.2 vav 3 oncogene 0.4 0.025 0.005 202226_s_at Hs.306088 gb:NM_016823.1 v-crk avian sarcoma virus CT10 oncogene homolog 0.4 0.002 0.007 209822_s_at Hs.73729 gb:L22431.1 very low density lipoprotein receptor 0.3 0.003 0.001 201832_s_at Hs.325948 gb:NM_003715.1 vesicle docking protein p115 0.4 0 0.001 201710_at Hs.179718 gb:NM_002466.1 v-myb avian myeloblastosis viral oncogene homolog-like 2 0.2 0 0.002 202431_s_at Hs.79070 gb:NM_002467.1 v-myc avian myelocytomatosis viral oncogene homolog 0.3 0.015 0.012 214058_at Hs.92137 gb:M19720 v-myc avian myelocytomatosis viral oncogene homolog 1, lung carcinoma derived 0.3 0.014 0.005 206036_s_at Hs.44313 gb:NM_002908.1 v-rel avian reticuloendotheliosis viral oncogene homolog 0.4 0.03 0.044 222138_s_at Hs.12142 gb:AF158978.1 WD repeat domain 13 0.4 0.001 0.002 200670_at Hs.149923 gb:NM_005080.1 X-box binding protein 1 0.2 0.002 0.001 224367_at gb:AF251053.1 X-linked protein 0.4 0.025 0.012 208453_s_at Hs.284202 gb:NM_006523.1 X-prolyl aminopeptidase (aminopeptidase P)-like 0.4 0 0.002
Page 14 of 15 Downregulated transcripts
Affymetrix code Unigene Code GenBank Code Name Fold Change t test p value MW U P value 214713_at Hs.159471 gb:AI703162_RC ZAP3 protein 0.4 0.002 0.009 234716_at Hs.41154 gb:U79264.1 Zic family member 1 (odd-paired Drosophila homolog) 0.3 0.004 0.001 204175_at Hs.102419 gb:NM_015871.1 zinc finger protein 0.4 0 0.002 222407_s_at Hs.15220 gb:AI493587_RC zinc finger protein 106 0.3 0.027 0.021 205883_at Hs.37096 gb:NM_006006.1 zinc finger protein 145 (Kruppel-like, expressed in promyelocytic leukemia) 0.3 0.012 0.002 218006_s_at Hs.108642 gb:NM_006963.1 zinc finger protein 22 (KOX 15) 0.3 0.004 0.002 202452_at Hs.29285 gb:AI991574_RC ZYG homolog 0.4 0.021 0.005
Page 15 of 15 5. Trial Protocol An assessment of HOXB2 overexpression in pancreatic cancer Version : 01/12//05
TITLE: A prospective evaluation of the correlation of HOXB2 overexpression in preoperative pancreatic biopsies compared to operative specimens in resected pancreatic cancer
Coordinating Center: Garvan Institute of Medical Research 384 Victoria Street Darlinghurst N.S.W. 2010 Sydney Australia
Principal Investigator: Dr Davendra Segara Garvan Institute of Medical Research Phone: +612 9295 8332 e-mail: [email protected]
Co-Investigators: Dr S.M.Henshall Garvan Institute of Medical Research
Dr A.V. Biankin Garvan Institute of Medical Research
Dr J.G. Kench Institute of Clinical Pathology and Medical Research, Westmead Hospital
Dr A.L. Morey Department of Anatomical Pathology St-Vincent’s Hospital
Associate Professor C.S. Lee Department of Anatomical Pathology Royal Prince Alfred Hospital
Dr M.J. Coleman Department of Surgery St-Vincent’s Hospital
Professor R.L. Sutherland Garvan Institute of Medical Research
Protocol Version: 01/12/05 An assessment of HOXB2 overexpression in pancreatic cancer Version : 01/12//05
Synopsis:
Background & Rationale: Pancreatic Cancer (PC) is the fourth leading cause of cancer death in Western societies. While resection confers a significant survival advantage in a minority of patients, the overall 5-year survival for PC remains at <5%. Thus, there is a critical need for improved therapeutic strategies in the management of PC. Recent novel data has identified that ectopic nuclear expression of HOXB2 is associated with a poor prognosis for all patients with PC and is the most influential independent predictor of survival in patients who underwent resection. Although pancreatic resection offers the best chance of cure in patients with PC, it is a procedure, which carries significant morbidity and mortality. The development of a reliable preoperative assessment of HOXB2 status would be an important addition to a physician’s limited diagnostic armamentarium in this disease and may be used, together with current preoperative clinicopathological parameters of disease outcome, to determine a patient’s suitability for operative resection.
Aims: To investigate the correlation between HOXB2 staining in preoperative biopsy derived pancreatic cancer specimens and resected pancreatic cancer specimens.
Hypothesis:
Primary: HOXB2 nuclear expression in pancreatic cancer biopsies correlates strongly with HOXB2 expression in pancreatic resection specimens.
Secondary: HOXB2 nuclear expression in pancreatic cancer biopsies is a predictor of survival.
Study Design: Multi-centre prospective cohort study
Population and Setting:
Target Population: - All patients who undergo pancreatic resection for PC in 5 collaborating teaching hospitals in Sydney Australia.
Inclusion Criteria: - Patients who have undergone preoperative pancreatic biopsy. - Confirmed histological or cytological diagnosis of PC. - Informed consent.
Exclusion criteria: - None. An assessment of HOXB2 overexpression in pancreatic cancer Version : 01/12//05
Interventions:
Tissue biopsies, from specimens retrieved, as part of the normal pre-operative assessment of PC, (endoscopic retrograde pancreatico-cholangiogram, radiological guided biopsy, endoscopic ultrasound or laproscopic assessment of the pancreas) will be assessed for nuclear expression of HOXB2 using a commercially available antibody for HOXB2 and correlated with expression of HOXB2 in the resected pancreas.
Outcomes and Measures:
Primary outcome: - Positive Predictive Value of the HOXB2 status of preoperative biopsy specimens compared to resected specimen
Secondary outcome: - Evaluate the specificity and sensitivity of HOXB2 staining in PC biopsy samples compared to resected specimen. -Evaluate the relationship between HOXB2 nuclear overexpression and survival time - Evaluate the relationship between HOXB2 nuclear overexpression and survival time adjusting for confounding of known clinico-pathological factors associated with survival such as tumor size, tumor differentiation, lymph node involvement and margin ionvolvement.
Study Procedures: With informed consent, patients with a preoperative diagnosis of PC will be recruited from 5 collaborating teaching hospitals in Sydney, Australia. Only those patients who have undergone successful pancreatic resection will have both, archived formalin fixed paraffin biopsy and resection samples collected and quarantined until the computed sample size is accrued. Samples will be stained for HOXB2 and nuclear expression will be assessed using two observers Standardisation of scoring will be achieved by comparison of scores, where any discrepancies are resolved by consensus.
Statistical Considerations: Sample size = 100 (positive predictive value of 0.9 (CI 0.74-0.97)) as determined using an exact confidence interval formula. Correlations will be determined using a kappa score of reliability. An assessment of HOXB2 overexpression in pancreatic cancer Version : 01/12//05
Feasibility; Our multidisciplinary team combines expertise in basic cancer biology, biomarkers, bioinformatics and histopathology. In addition our laboratory has a fully equipped, custom designed histopathology facility. As this study requires the assessment of 100 patients from 5 surgical centres treating PC, over 2 years, and each collaborating centre currently treats approximately 10 patients / year with pancreatic resection, successful accrual of patients will be feasible in the time period proposed for this study. An assessment of HOXB2 overexpression in pancreatic cancer Version : 01/12//05
1. Background and Rationale
Pancreatic cancer (PC) is the fifth leading cause of cancer death in Western societies with a 5-year survival rate of less than 10% 1. PC presents at an advanced stage, and as a result, only 10-20% of patients are suitable for surgical treatment at the time of presentation2. Clinical management of these patients is complicated by inconsistencies in the influence of conventional clinicopathological variables on outcome suggesting that some of these parameters lack accuracy. In addition, preoperative assessment of some parameters such as lymph node metastases is difficult. Whereas in other cancers assessment of aberrations in gene expression that cosegregate with therapeutic response and outcome are being adopted routinely to increase predictive power (e.g. ER and HER2/neu in breast cancer), there remain no molecular markers of clinical utility in PC. Recent novel data has identified that ectopic nuclear expression of HOXB2 is associated with a poor prognosis for all patients with PC and is the most influential independent predictor of survival in patients who underwent resection3. Although pancreatic resection offers the best chance of cure in patients with PC, it is a procedure, which carries significant morbidity and mortality. The development of a reliable preoperative assessment of HOXB2 status would be an important addition to a physician’s limited diagnostic armamentarium in this disease and may be used together with current preoperative clinicopathological parameters of disease outcome to determine a patient’s suitability for operative resection.
2 Objectives of the trial
2.1 General objectives
The aim of the proposed study is to investigate the correlation between HOXB2 staining in preoperative biopsy derived pancreatic cancer specimens and resected pancreatic cancer specimens.
2.2 End-points
a) Primary outcome: - Correlation of the HOXB2 nuclear staining intensity of preoperative biopsy specimens compared to resected specimen. b) Secondary outcomes: - Evaluate the specificity and sensitivity of HOXB2 staining in PC biopsy samples compared to resected specimen. - To evaluate the relationship between HOXB2 nuclear overexpression in resected PC and outcome. - To evaluate the relationship between HOXB2 nuclear overexpression and traditional clinicopathological parameters of outcome (margin involvement, tumor size, lymph node involvement and degree of differentiation). An assessment of HOXB2 overexpression in pancreatic cancer Version : 01/12//05 3 Patient Selection Criteria
All patients who undergo pancreatic resection for PC in 5 collaborating teaching hospitals (St-Vincents Public Hospital, St-Vincent’s Private Hospital, Royal Prince Alfred Hospital, Westmead Hospital and Concord Hospital), in Sydney Australia.
3.1 Inclusion Criteria: - Confirmed histological or cytological diagnosis of PC -. Patients who have undergone preoperative pancreatic biopsy. -.Absence of any psychological, familial, sociological or geographical condition potentially hampering compliance with the study protocol and follow-up schedule; those conditions should be discussed with the patient before registration in the trial. - Before patient registration, informed consent must be given according to national/local regulations 3.2 Exclusion Criteria: - Patient chooses to withdraw from study.
4. Trial Design
• With appropriate ethics approval and informed consent, all patients with a preoperative diagnosis of PC will be recruited from 5 collaborating teaching hospitals (St-Vincent’s Public Hospital, St-Vincent’s Private Hospital, Royal Prince Alfred Hospital, Westmead Hospital and Concord Hospital), in Sydney, Australia. • Only those patients who have undergone successful pancreatic resection will have both, archived formalin fixed paraffin biopsy and resection samples collected and stored at our tissue storage facility until the computed sample size is accrued (see statistical considerations). • Patient demographic and clinico-pathological data will be collected at the time of recruitment into the study. Six monthly follow-up will occur by: o Audit of treating physicians notes. o Query of the New South Wales Cancer Council Database for reported patient death on a “return to notifier” basis • Samples will be assessed for HOXB2 nuclear expression with an established immunohistochemical protocol3 using a commercially available antibody to HOXB2. • HOXB2 nuclear overexpression will be assessed using two blinded independent observers, one of which is a histopathologist with experience in PC and immunohistochemistry. Standardisation of scoring will be achieved by comparison of scores, and by conferencing where any discrepancies will be resolved by consensus. An assessment of HOXB2 overexpression in pancreatic cancer Version : 01/12//05 5 Interventions
The proposed intervention involves the collection of tissue sampling biopsies (fine needle aspirate biopsies or core biopsies), from specimens retrieved, as part of the normal pre-operative assessment of PC, (endoscopic retrograde pancreatico-cholangiogram, radiological guided biopsy, endoscopic ultrasound or laproscopic assessment of the pancreas). These will be assessed for nuclear expression of HOXB2 using a commercially available polyclonal antibody for HOXB2 and correlated with expression of HOXB2 in the resected pancreatic cancer specimen. Scores are given as the percentage of nuclei staining positive within the representative area of the tissue core and the absolute intensity of nuclear staining on a scale of 0 to 3 (0 representing no staining 1, slight/weak heterogenous nuclear staining, 2, strong homogenous nuclear staining and 3, intense homogenous nuclear staining). The criteria to achieve a positive score were: HOX B2 nuclear intensity >1 in >20% of nuclei.
6 Clinical evaluation, and follow-up
6.1 Before intervention: All patients must have a confirmed histlogical or cytological diagnosis of pancreatic cancer All patients are in the process of being evaluated for pancreatic resection for treatment of pancreatic cancer. Informed consent is given. Demographic data is collected by trial co-ordinator, this data includes: • Patient Name • Patient age • Patient Address and contact details • Local Medical Officer contact details • Treating Surgeon Contact details
Pathological data is collected, this data includes: • Type of biopsy performed (fine needle aspirate or core biopsy) Mode of assessment : ERCP radiological guided biopsy, endoscopic ultrasound laproscopic assessment of the pancreas • Site of biopsy • Histopathological / Cytological result: • Pathologist performing assessment. An assessment of HOXB2 overexpression in pancreatic cancer Version : 01/12//05
6.2 After Resection: The following data will be collected: Pathological Data: • Site of Tumor (head, body, or tail of the pancreas) • Histological Diagnosis • Size of tumor (mm) • Type of tumor differentiation (well, moderate or poor) • Margin Involvement (presence of cancer cells at the resected surgical margin) • Draining Lymph Node Involvement (present, absent)
6.3 Follow-up: Six monthly follow-up will occur by :
Audit of treating physicians notes. This includes all clinicians treating the patient Local Medical Officer, Surgeon, Medical Oncologist, Radiation Oncologist Palliative Care Physician Query of the New South Wales Cancer Council Database for reported patient death on a “return to notifier” basis Follow-up endpoint is death defined as primarily or secondarily attributable to pancreatic cancer.
7 Criteria of evaluation
HOXB2 nuclear intensity in both the biopsy and resected specimen: Scores are given as the percentage of nuclei staining positive within the representative area of the tissue core and the absolute intensity of nuclear staining on a scale of 0 to 3 (0 representing no staining 1, slight/weak heterogenous nuclear staining, 2, strong homogenous nuclear staining and 3, intense homogenous nuclear staining). The criteria to achieve a positive score were: HOX B2 nuclear intensity >1 in >20% of nuclei. An assessment of HOXB2 overexpression in pancreatic cancer Version : 01/12//05
8 Statistical considerations
8.1 Statistical design
8.1.1 Sample size
The computed sample size required for the proposed study is 100 patients with resected PC, which would provide a positive predictive value of 0.9 (CI 0.74-0.97) using the exact confidence interval formula. As this study requires the assessment of 100 patients from 5 surgical centres treating PC, over 2 years, and each collaborating centre currently treats approximately 10 patients / year with pancreatic resection, successful accrual of patients will be feasible in the time period proposed for this study.
8.1.2 Randomization and stratifications
No Randomisation and or Stratification is required for this study.
8.2 Analysis
Correlations between HOXB2 nuclear intensity of resected PC and preoperative biopsy specimen, will be determined using a kappa (k) score of reliability. Where a k of greater or equal to 0.8 indicates acceptable reliability4.
Kaplan-Meier models will used initially to assess the relationship between HOXB2 nuclear staining, from the biopsy specimens and survival.
Cox Proportional Hazard’s Models will be used to explore the relationshipbetween HOXB2 nuclear staining and survival after adjusting for confounding factors such as margin involvement, tumor size, lymph node involvement and degree of differentiation.
Univariate and multivariate analyses will be generated using Statview 5.0 Software (Abacus Systems, Berkeley, CA). A p value of < 0.05 was accepted as statistically significant. Those factors that were prognostic on univariate analysis will be assessed in a multivariable model to identify factors that were independently prognostic and those that were the result of confounding.
8.3 Interim Analyses
No interim analysis will be carried out during this trial. An assessment of HOXB2 overexpression in pancreatic cancer Version : 01/12//05 9 Investigator authorization procedure
Investigators will be authorized to register or randomize patients in this trial only when they have returned to the Data Center:
• a commitment statement / study acknowledgment form, indicating that they will fully comply with the protocol, to include an estimate of their yearly accrual and if any conflict of interest may arrive due to their participation in the trial, • a copy of the letter of acceptance of the protocol by their local ethics committee, • a signed conflict of interest disclosure form: this document will be required only if a possible conflict is declared by the commitment form. • and, if the following documents are not yet available at the Data Center: • their updated Curriculum Vitae, • the list of the normal ranges, in their own institution, of all laboratory data required by the protocol, • the list of their staff members authorized to sign case report forms, with a sample of each authorized signature. • The new investigator will be added to the “authorization list”, and will be allowed to register/randomize patients in the trial as soon as • all the above mentioned documents are available at the Data Center • all applicable national health authorities requirements are fulfilled • Patients registration from centers not (yet) included on the authorization list will not be accepted.
10 Quality assurance
10.1 Control of data consistency Data forms will be entered in the database of the Data Center by a double data entry procedure. Computerized and manual consistency checks will be performed on newly entered forms; queries will be issued in case of inconsistencies. Consistent forms will be validated by the Data Manager to be entered on the master database. Inconsistent forms will be kept "on-hold" until resolution of the inconsistencies. An assessment of HOXB2 overexpression in pancreatic cancer Version : 01/12//05 11 Ethical considerations
11.1 Patient protection The responsible investigator will ensure that this study is conducted in agreement with either the Declaration of Helsinki (Tokyo, Venice, Hong Kong and Somerset West amendments) or the laws and regulations of the country, whichever provides the greatest protection of the patient. The protocol has been written, and the study will be conducted according to the ICH Harmonized Tripartite Guideline for Good Clinical Practice (ref: http://www.ifpma.org/pdfifpma/e6.pdf).
The protocol will by approved by the Local, Regional Ethics Committees.
11.2 Subject identification The name of the patient will be asked for and recorded at the Data Center. A sequential identification number will be automatically attributed to each patient registered in the trial. This number will identify the patient and must be included on all case report forms. In order to avoid identification errors, patients initials (maximum of 4 letters), date of birth and local chart number (if available) will also be reported on the case report forms.
11.3 Informed consent All patients will be informed of the aims of the study, the possible adverse events, the procedures and possible hazards to which he/she will be exposed, and the mechanism of treatment allocation. They will be informed as to the strict confidentiality of their patient data, but that their medical records may be reviewed for trial purposes by authorized individuals other than their treating physician. An example of a patient informed consent statement is given as an appendix to this protocol. It will be emphasized that the participation is voluntary and that the patient is allowed to refuse further participation in the protocol whenever he/she wants. This will not prejudice the patient’s subsequent care. Documented informed consent must be obtained for all patients included in the study before they are registered at the Data Center. This is done in accordance with the national and local regulatory requirements. The informed consent procedure must conform to the ICH guidelines on Good Clinical Practice. This implies that “the written informed consent form should be signed and personally dated by the patient or by the patient’s legally acceptable representative”. An assessment of HOXB2 overexpression in pancreatic cancer Version : 01/12//05
12 Administrative responsibilities
12.1 The study coordinator
The Study Coordinator (in cooperation with the Data Center) will be responsible for writing the protocol, reviewing all case report forms and documenting his/her review on evaluation forms, discussing the contents of the reports with the Data Manager and the Statistician, and for publishing the study results. He will also generally be responsible for answering all clinical questions concerning eligibility, treatment, and the evaluation of the patients.
Study Co-ordinator:
Pancreatic Research Group
Address: 384 Victoria Street Darlinghurst N.S.W. 2010 Sydney, Australia
Tel: + 612 9295 8320 Fax: + 612 9295 8321 e-mail: [email protected]
12.2 The Data Center The Data Center will be responsible for reviewing the protocol, collecting case report forms, controlling the quality of the reported data, and generating reports and analyses in cooperation with the Study Coordinator. All methodological questions should be addressed to the Data Center.
DATA CENTER Garvan Institute of Medical Research Translational Research Group 384 Victoria Street Darlinghurst N.S.W. 2010 Sydney Australia
Registration of patients: Tel +612 9295 8320 An assessment of HOXB2 overexpression in pancreatic cancer Version : 01/12//05 13 Publication Policy
The final publication of the trial results will be written by the Study Coordinator on the basis of the final analysis performed at the Data Center. A draft manuscript will be submitted by the study coordinator to the Data Center for review no later than six months after receiving the Data Center report. After revision by the Data Center and other co-authors the manuscript will be sent to a major scientific journal.
Authors of the manuscript will include at least the Study Coordinator, the investigators who have included more than 5% of the eligible patients in the trial (by order of inclusion), and members of the Data Center team who have contributed to the trial.
All publications, abstracts or presentations including data from the present trial will be submitted for review to the Data Center prior to submission. All manuscripts will include an appropriate acknowledgment section, mentioning all investigators who have contributed to the trial, as well as supporting bodies (NH&MRC, Royal Australasian College of Surgeons and St- Vincent’s Clinic Foundation)
The Study Coordinator and the Data Center must approve all publications, abstracts and presentations based on patients included in this study. This is applicable to any individual patient registered in the trial, or any subgroup of the trial patients. Such a publication cannot include any analysis of any of the study end-points unless the final results of the trial have already been published by the Study Coordinator. An assessment of HOXB2 overexpression in pancreatic cancer Version : 01/12//05 14 References
1 Jemal A, Murray T, Samuels A, Ghafoor A, Ward E, Thun MJ. Cancer statistics, 2003. CA Cancer J Clin 2003;53:5-26. 2 Yeo CJ, Cameron JL, Sohn TA, Lillemoe KD, Pitt HA, Talamini MA, Hruban RH, Ord SE, Sauter PK, Coleman J, Zahurak ML, Grochow LB, Abrams RA. Six hundred fifty consecutive pancreaticoduodenectomies in the 1990s: pathology, complications, and outcomes. Ann Surg 1997;226:248-57; discussion 257-60. 3 Segara D, Biankin A.V. Kench J.G., Dawson A.C., Skalicky D. Coleman M., Sutherland R.L., and Henshall S.M. Expression of HOXB2 is an early event in the development of pancreatic cancer and is associated with a poor prognosis. (submitted for publication) 4 Landis, J.R. and Koch, G.G. The measurement of observer agreement for categorical data. Biometrics 33:159-174, 1977 An assessment of HOXB2 overexpression in pancreatic cancer Version : 01/12//05
PARTICIPANT INFORMATION SHEET
A prospective evaluation of the correlation of HOXB2 overexpression in preoperative pancreatic biopsies compared to operative specimens in resected pancreatic cancer
This is a clinical trial (a type of research study). Clinical trials include only patients who choose to take part. Please take your time to make your decision. Discuss it with your friends and family.
You are being asked to take part in this study because you have been diagnosed pancreatic cancer.
WHYISTHISSTUDYBEINGDONE?
Currently there are no tests available which help doctors decide will do better after an operation for your pancreatic cancer.
Recent studies have identified that a test called HOXB2 may predict which patients do better after an operation for pancreatic cancer.
Presently we can only use this test after you have had your operation.
The purpose of this study is to see if a test performed on your pancreas before your operation gives the same results as a test performed on your pancreas after your operation An assessment of HOXB2 overexpression in pancreatic cancer Version : 01/12//05
WHAT IS INVOLVED IN THE STUDY?
If you decide to participate, a portion of the samples taken during the diagnosis and treatment of your condition will be taken and stored by our institution.
A test for HOXB2 will be done on a piece of your pancreas taken before your operation and compared with a test for HOXB2 done on a piece of your pancreas taken as part of your operation for pancreatic cancer
WHAT ARE THE RISKS OF THE STUDY?
Participation will not involve any extra risk or discomfort to you it means that you give us permission to take some of the sample that has already taken as part of the standard medical investigation and treatment, for our research.
WHAT ARE THE BENEFITS OF TAKING PART IN THE STUDY?
If you agree to take part in this study, there may not be direct medical benefit to you. However we hope the information learned from this study will benefit other patients with pancreatic cancer in the future.
WHAT ABOUT CONFIDENTIALITY?
If you give us your permission by signing the Participant Consent Form, we plan to discuss the results at conferences and publish the results in medical journals. In any publication or discussion, information will be provided in such a way that you cannot be identified.
Efforts will be made to keep your personal information confidential. Only investigators involved in this study will have access to your personal information We cannot guarantee absolute confidentiality. Your personal information may only be disclosed if required by law.
WHAT ARE THE COSTS?
As the specimens that we are asking permission to acquire, store and test are portions of specimens taken as part of the normal diagnostic and treatment procedures for your condition, there will be no additional costs incurred for your treatment.
You will receive no payment for taking part in this study.
WHAT ARE MY RIGHTS AS A PARTICIPANT?
Taking part in this study is voluntary. You may choose not to take part or may leave the study at any time. Your decision whether or not to participate will not prejudice your future relations with The Garvan Institute of Medical Research or xxx Hospital Leaving the study will not result in any penalty or loss of benefits to which you are otherwise entitled. An assessment of HOXB2 overexpression in pancreatic cancer Version : 01/12//05
WHOM DO I CALL IF I HAVE QUESTIONS OR PROBLEMS?
For questions about the study or a research-related injury, contact the researcher
Dr Davendra Segara Phone: +612 9295 8320
This project has been xxxx Area Health Service approved by the Human Ethics Research Committee For questions about your rights as a research participant, contact the xxx Hospital Patient Representative Phone: 12345678
You will get a copy of this form. You can also request a copy of the protocol (full study plan).
SIGNATURE
I agree to take part in this study.
Participant ______Date ______An assessment of HOXB2 overexpression in pancreatic cancer Version : 01/12//05
RESEARCH STUDY: A prospective evaluation of the correlation of HOXB2 overexpression in preoperative pancreatic biopsies compared to operative specimens in resected pancreatic cancer
1 CONSENT FORM
I, ………………………………………………………………………….[name] of……………………………………………………………………………[address] have been invited to participate in the above named research study and have discussed the study with………………………………………………[Name of Informant and Position]
• I acknowledge that I have received and read the Participant Information Sheet, which includes the aims of this study, the procedures involved in this study, including any inconvenience, risk, discomfort or side effects, and of their implications.
• I understand that my participation in this study is entirely voluntary and that I can withdraw at any stage. If I withdraw, this decision will not affect in any way, my future treatment or my relationship with my doctors and other members of my health care team.
• I also understand that the information relating to my participation in the study is strictly confidential.
• I understand that the research project will be carried out according to the principles in the National Health & Medical Research Council Statement on Human Experimentation.
• I understand that if I have any questions relating to my participation in this research study, I may contact Dr Segara on (02) 9295 8320, who will be happy to discuss them with me.
• I understand that if I have any questions about my rights as a research subject or on any other administrative matters, I may contact the xxxxx Hospital Patient Representative, , Telephone No xxxxx xxxxxx.
I hereby freely agree to participate in this research study.
I also state that I have/have not participated in any other research project in the past 3 months. If I have, the details are as follows:
Name (Print):……………………………………………………………………………………….
Signature:……………………………………………………………….Date:…………
Name of Witness (Print):………………………………………………………………………….
Signature of Witness:……………………………………………….Date:……………………. An assessment of HOXB2 overexpression in pancreatic cancer Version : 01/12//05
ADDENDUM CONSENT:
I agree to allow the use of blood samples, other body fluids and tissues obtained during testing, operative procedures, or other standard medical practices for further research purposes. a. I agree to the use of my specimens for research and teaching purposes related to Pancreatic Cancer
Yes No b. I agree to be recontacted in the future to discuss whether I will give permission for my specimens to be used for genetic research.
Yes No c. I agree to allow my specimens to be used for research unrelated to Pancreatic Cancer
Yes No
Name (Print):……………………………………………………………………………………….
Signature:……………………………………………………………….Date:…………
Name of Witness (Print):………………………………………………………………………….
Signature of Witness:……………………………………………….Date:……………………. 6. Publications Cancer Prevention
Expression of HOXB2, a Retinoic Acid SignalingTarget in Pancreatic Cancer and Pancreatic Intraepithelial Neoplasia Davendra Segara,1Andrew V. Biankin,1, 2 James G. Kench,1, 3 Catherine C. Langusch,1Amanda C. Dawson,1 David A. Skalicky,1David C. Gotley,4 Maxwell J. Coleman,2 RobertL.Sutherland,1and Susan M. Henshall1
Abstract Purpose: Despite significant progress in understanding the molecular pathology of pancreatic cancer and its precursor lesion: pancreatic intraepithelial neoplasia (PanIN), there remain no molecules with proven clinical utility as prognostic or therapeutic markers. Here, we used oligo- nucleotide microarrays to interrogate mRNA expression of pancreatic cancer tissue and normal pancreas to identify novel molecular pathways dysregulated in the development and progression of pancreatic cancer. Experimental Design: RNA was hybridized to Affymetrix Genechip HG-U133 oligonucleotide microarrays. A relational database integrating data from publicly available resources was created to identify candidate genes potentially relevant to pancreatic cancer. The protein expression of one candidate, homeobox B2 (HOXB2), in PanIN and pancreatic cancer was assessed using immunohistochemistry. Results: We identified aberrant expression of several components of the retinoic acid (RA) signaling pathway (RARa,MUC4,Id-1,MMP9,uPAR,HB-EGF,HOXB6,andHOXB2),manyof which are known to be aberrantly expressed in pancreatic cancer and PanIN. HOXB2, a down- stream target of RA, was up-regulated 6.7-fold in pancreatic cancer compared with normal pan- creas. Immunohistochemistry revealed ectopic expression of HOXB2 in15% of early PanINlesions and 48 of 128 (38%) pancreatic cancer specimens. Expression of HOXB2 was associated with nonresectable tumors and was an independent predictor of poor survival in resected tumors. Conclusions: We identified aberrant expression of RA signaling components in pancreatic cancer, including HOXB2, which was expressed in a proportion of PanIN lesions. Ectopic expression of HOXB2 was associated with a poor prognosis for all patients with pancreatic cancer and was an independent predictor of survival in patients who underwent resection.
Pancreatic cancer is the fifth leading cause of cancer death that some of these variables lack accuracy. In addition, in Western societies with a 5-year survival rate of <10% (1). preoperative assessment of some variables such as lymph node Pancreatic cancer presents at an advanced stage; thus, only metastases is difficult. Whereas in other cancers assessment of 10% to 20% of patients are suitable for surgical treatment at aberrations in gene expression that cosegregate with therapeu- the time of presentation (1). Clinical management of these tic response and outcome are being adopted routinely to patients is complicated by inconsistencies in the influence of increase predictive power (e.g., ER and HER-2/neu in breast conventional clinicopathologic variables on outcome suggesting cancer), there remain no molecular markers of clinical utility in pancreatic cancer. This highlights the need for the identification of novel regulatory pathways important in Authors’ Affiliations: 1Cancer Research Program, Garvan Institute of Medical pancreatic cancer that may also have diagnostic, therapeutic Research and 2Division of Surgery, St. Vincent’s Hospital, Darlinghurst, Sydney, and prognostic utility. New South Wales, Australia; 3Institute of Clinical Pathology and Medical Research, There is now compelling histopathologic and molecular 4 Westmead Hospital, Westmead, New South Wales, Australia; and University of evidence to support the evolution of pancreatic cancer through Queensland, Department of Surgery, Princess Alexandra Hospital, Woolloongabba, Queensland, Australia a series of noninvasive duct lesions called pancreatic intra- Received 9/5/04; revised 2/1/05; accepted 2/10/05. epithelial neoplasia (PanIN; refs. 2, 3). Early duct lesions Grant support: Royal Australasian College of Surgeons, National Health and designated PanIN-1A and PanIN-1B show minimal cytologic Medical Research Council of Australia, St. Vincent’s Clinic Foundation Sydney, and architectural atypia and are associated with activating K-ras Cancer Council New South Wales, R.T. Hall Trust, and Prostate Cancer Foundation mutations (4), shortened telomeres (5), and overexpress of Australia (S.M. Henshall). WAF1/CIP1 The costs of publication of this article were defrayed in part by the payment of page p21 (6). PanIN-2 lesions exhibit mild to moderate charges. This article must therefore be hereby marked advertisement in accordance cytologic and architectural atypia and are associated with with 18 U.S.C. Section 1734 solely to indicate this fact. loss of p16INK4A expression (7) and cyclin D1 overexpression Note: D. Segara and A. Biankin contributed equally to this work. (6). PanIN-3 exhibits significant cytologic and architectural Requests for reprints: Robert L. Sutherland, Cancer Research Program Garvan atypia, manifests p53 mutations (8), and loss of DPC4/ Institute of Medical Research 384 Victoria Street, Darlinghurst. New South Wales 2010, Australia. Phone: 61-2-9295-8322; Fax: 61-2-9295-8321; E-mail: Smad4 expression (6). These molecular aberrations increase [email protected]. in frequency with advancing PanIN lesions through to invasive F 2005 American Association for Cancer Research. cancer.
www.aacrjournals.org3587 Clin Cancer Res 2005;11(9) May 1, 2005 Cancer Prevention
During vertebrate development, retinoic acid (RA) signaling clinicopathologic data. Multiple samples of pancreatic tissue of f500 is important for the correct patterning of embryonic structures mg were excised intraoperatively from 12 patients, undergoing (9). Endodermal expression of pdx-1 (a homeobox-containing pancreatic resection for pancreatic cancer, immediately snap frozen in j transcription factor essential for pancreatic development) is liquid nitrogen and stored at 80 C, before RNA extraction. Total RNA was isolated from 12 pancreatic cancer specimens and six macroscopi- induced by RA (10) and marks a pluripotent population of cells cally and microscopically normal appearing pancreas from the same that give rise to all cell types in the pancreas. RA signaling patients (matched). Biotinylated cRNA for Affymetrix Genechip hybrid- regulates pancreas exocrine lineage selection, and treatment ization was prepared through a single round of reverse transcription with RA analogues can effect a shift from an acinar to a ductal with Superscript II (Life Technologies, Rockville, MD) followed by se- phenotype through epithelial-mesenchymal interactions (11). cond strand synthesis to create double stranded cDNA. After purification Such a shift from an exocrine to a predominantly ductal the cDNA was transcribed and labeled using a T7 polymerase (Enzo phenotype is characteristic of mouse models of pancreatic Technologies, New York, NY) and purified (27). Hybridization cocktails cancer development. In addition, pancreatic stellate cells, which were prepared as per the Affymetrix protocol (Affymetrix, Santa Clara, are essential for the development of fibrosis associated with CA) and quality assured on Affymetrix Test3 arrays, before hybridization chronic pancreatitis and pancreatic cancer, store retinoids in fat to HG-U133A and B oligonucleotide microarrays. Data analysis. A relational database was constructed using File- droplets, and in turn can have their function altered with RA in vitro Maker Pro (FileMaker, Inc., San Francisco, CA) to facilitate multiple analogue treatment (12). The retinoid signal is queries of gene expression data generated from the above experiments transduced by two families of nuclear transcription factors: and public domain data available electronically from the Internet. The RA receptors (RAR) and retinoid X receptors, that are members database incorporated (a) transcript profiles of pancreatic cancer and of the nuclear receptor superfamily, which in the presence of normal pancreas from the experiments done in this study (absolute ligand heterodimerize to activate the transcription of target values); (b) mathematical algorithms programmed within the database genes through RA response elements (13). Although few RA to generate fold change comparisons between the average expression response elements have been identified, one of the mechanisms across all samples of pancreatic cancer to the average in normal pancreas; by which retinoids exert their effects is thought to be through (c) linear statistical analyses generated using the Affymetrix Data Mining t U regulation of HOX gene expression (9, 14). Tool Software (MAS 5.0), which included test and Mann-Whitney Homeobox genes are transcription factors with established test data for comparisons between normal pancreas and pancreatic cancer and (d) interactive molecular pathway maps were generated using roles in development and cell function. The homeobox is a GenMAPP software (Gladstone Institutes UCSF, San Francisco, CA, highly conserved 183-bp DNA sequence coding for a 61-amino- http://www.GenMAPP.org/default.html), designed to incorporate tran- acid domain, the homeodomain (15). This region binds DNA script profile data into maps of known pathways including those elements, primarily those that contain a TAAT core motif involved in carcinogenesis and development. Data files using Swissprot (16). Accordingly, homeodomain containing proteins act as identification numbers were uploaded into the program, and various both activators and repressors of transcription. Human class 1 pathway maps available as part of the package were used to model homeobox genes called HOX genes consist of 39 genes numerous pathways. An existing RA signaling GenMAPP was modified arranged in four clusters HOXA, HOXB, HOXC, and HOXD to include all molecules thought to be regulated by RA signaling and is localized on chromosomes 7, 17, 12, and 2, respectively (17). presented in Fig. 1. Statistical data were generated using the t test and U Mammalian development requires a complex interaction of Mann-Whitney tests to compare the average expression across samples HOX gene networks, with HOX gene expression commencing of pancreatic cancer to the average expression of all samples of normal pancreas for the GenMAPP that is presented. during gastrulation and collectively controlling the identity of Patient cohort. We identified a cohort of 128 patients with a diag- various regions along the body axis from the hindbrain to the nosis of pancreatic adenocarcinoma that underwent pancreatic resection HOX tail (18, 19). Aberrant expression of genes has been or biopsy between January 1972 and November 2001 with available implicated in the development of solid tumors including renal archived tissue. This cohort represents a subset of a previously described carcinoma (20), colon cancer (21), ovarian carcinoma (22), group of 348 patients (28). Archival formalin-fixed, paraffin-embedded and breast carcinoma (23, 24). Given the emerging importance tissue from all 128 pancreata that were resected or biopsied were used to of developmental pathways in pancreatic cancer such as Notch construct seven pancreatic cancer tissue arrays, which contained up to (25) and sonic hedgehog (26), the role of RA signaling in early 55 1.6 mm cores per slide. Conventional sections of 26 cases of normal pancreas development and evidence of RA signaling and pancreas from areas distal to the pancreatic cancer were used to assess homeobox gene network dysregulation in carcinogenesis, we gene expression in benign ductal epithelial cells and PanIN lesions. For this cohort, the average age at diagnosis was 63.8 years present data suggesting that aberrant RA signaling may be (median, 66.5; range, 34-86; Table 1). Of the 128 patients, 76 were important in pancreatic cancer. Based on these data, we from pancreatic resections, 46 intraoperative incision biopsies, and HOXB2 assessed , a RA-responsive gene, and show that ectopic 6 postmortem specimens. Median follow-up for the cohort was expression of HOXB2 occurs in a significant proportion of 7.6 months (range, 0-117 months). Eight patients were alive at the pancreatic cancer, is detectable in a proportion of PanIN, and is census date (September 21, 2002). Median disease-specific survival associated with a poor prognosis, supporting a potential role of was 7.25 months. For the resected group of 76 patients, 39 (51%) had HOXB2 in the biological behavior of some pancreatic cancer. lymph node metastasis (Table 1). The mean tumor size was 31 mm. Resection margins were microscopically free of tumor in 40 patients (53%). Poorly differentiated tumors occurred in 25 patients (33%). Materials and Methods Median follow-up was 11.0 months with a median disease-specific survival of 10.1 months, 1-year survival of 48.6%, and 5-year survival RNA preparation and transcript profiling. Ethical approval was of 11%. The 30-day mortality for resection was 2 (3%). obtained from five teaching hospitals (The Princess Alexandra Hospital, Immunohistochemistry. Pancreatic tissue microarrays were cut at Brisbane, Australia, Westmead Hospital, Concord Hospital, Royal Prince 4 Am, deparaffinized, and rehydrated before unmasking in target Alfred Hospital, and St. Vincent’s Hospital Campus in Sydney, Australia) retrieval solution (EDTA and citrate, DAKO Co., Carpinteria, CA) in a for the acquisition of fresh and archival tissue and recording of microwave for 30 minutes. Using a DAKO autostainer, endogenous
Clin Cancer Res 2005;11(9) May 1, 2005 3588 www.aacrjournals.org HOXB2 and Pancreatic Cancer
Fig. 1. A customized GenMAPP of RA signaling components with statistically significant relative expression levels in pancreatic cancer (PC) compared with normal pancreas. Relative expression levels are represented as average fold change with those with statistically significant up-regulation marked red, those with statistically significant down-regulation marked blue (a < 0.05 on t test and/or Mann-Whitney U test) and no change (NC)markedgreenfor(A)RARs,(B) cellular RA-binding proteins (CRABP), (C) downstream targets of RA signaling previously described to be aberrantly expressed in pancreatic cancer and PanIN, (D) downstream targets of RA signaling aberrantly expressed in this study, (E) krox20-mediated regulation of HOXB2 and HOXB1expression in normal hindbrain development. Ps presented inTable 2. Abbreviations not mentioned in text: RBP4, retinoid binding protein 4; CRALBP, cellular retinaldehyde binding protein; RALDH, retinaldehyde dehydrogenase; hRADH, retinol dehydrogenase homologue. peroxidase activity was quenched in 3% hydrogen peroxide in methanol intensity of nuclear staining on a scale of 0 to 3 (0, no staining; 1, followed by avidin/biotin and serum-free protein blocks (DAKO). slight/weak heterogenous nuclear staining; 2, strong homogenous Sections were incubated for 30 minutes with 1:200 anti-HOXB2 (P-20) nuclear staining; and 3, intense homogenous nuclear staining). The goat polyclonal antibody (Santa Cruz Biotechnology, Santa Cruz, CA). A criteria to achieve a positive score were HOX B2 nuclear intensity of >1 streptavidin-biotin peroxidase detection system was used according in >20% of nuclei. to the manufacturer’s instructions (LSAB label + link kit; DAKO) with Statistical analysis. Kaplan-Meier and the Cox proportional hazards 3,3V-diaminobenzidine as a substrate. Counterstaining was done with model were used for univariate and multivariate analysis using Statview Mayer’s hematoxylin. HOXB2-positive breast cancer was used as a 5.0 Software (Abacus Systems, Berkeley, CA). P < 0.05 was accepted as positive control (24), whilst ovary was used as a negative control. statistically significant. Those factors that were prognostic on univariate Antibody specificity was confirmed using blocking peptide sc-17165 P analysis were assessed in a multivariable model to identify factors that (Santa Cruz Biotechnology), which abrogated nuclear staining for were independently prognostic and those that were the result of immunohistochemistry and eliminated a specific band on Western confounding. This analysis was done sequentially on all patients who blotting using pancreatic cancer cell lines. In addition, mRNA expres- had available tissue (n = 128) and on a subgroup of patients who sion in cell lines using reverse transcription-PCR correlated with protein underwent operative resection (n = 76). expression on Western blotting. Immunohistochemical scoring. Up to four separate samples of pancreas were examined per patient. Staining was assessed by two Results blinded independent observers (D.S. and J.G.K.). Standardization of scoring was achieved by comparison of scores between observers and by Transcript profiling data analysis. Whereas previous tran- conferencing, where any discrepancies were resolved by consensus. script profiling studies have been limited to identifying single Scores were given as the percentage of nuclei staining positive within genes aberrantly expressed in pancreatic cancer (29, 30), we the representative area of the tissue microarray core and the absolute employed a strategy that used GenMAPP software to identify
www.aacrjournals.org3589 Clin Cancer Res 2005;11(9) May 1, 2005 Cancer Prevention
Ta b l e 1. Clinicopathologic and outcome data for all patients in the cohort
Whole cohort Median Resected Median Variable no. (%) survival (mo) P (log-rank) cohort no. (%) survival (mo) P (log-rank)
Sex 128 76 Male 72 (56) 45 (59) Female 56 (44) 31(41) Age (y) 128 76 Mean 63.8 61 Median 66.5 65 Range 34-86 34-83 Treatment 128 Resection 76 (59) 11 Operative biopsy 46 (36) 3.9 <0.0001 No operative intervention 6 (5) Outcome 128 76 Follow-up (mo) 0-117 0.2-117 Median 7.6 11 30-d mortality 2(3) Death from pancreatic cancer 114(89) 63 (83) Death from other cause 2 (2) 2 (3) Alive 8 (6) 8 (11) Lost to follow-up 4 (3) 3 (4) Stage 127 I27(21) II 13 (10) 13.7 III 70 (55) IV 17 (13) 6.4 <0.0001 Differentiation 127 Well 11 (9) 7 (9) Moderate 68 (53) 8.9 44 (58) 12.2 Poor 48 (38) 5 0.0152 25 (33) 8.6 0.0582 Tumor size (mm) V20 15 (20) 17.1 >20 61 (80) 9.7 0.0375 Margins Clear 40 (53) 14.5 Involved 36 (47) 8.5 0.0014 Lymph node status Positive 39 (51) 9.2 Negative 35 (46) 13.8 0.0235 HOXB2 expression 128 76 Positive 48 (38) 5 16 (21) 6.75 Negative 80 (72) 9.9 <0.0001 60 (79) 14.0 <0.0001 molecular pathways in which a significant proportion of genes cancer and PanIN from other studies, were also up-regulated: showed aberrant expression. Using this approach, we confirmed S100 calcium binding protein P (S100P; ref. 31; 152-fold), aberrations in molecular pathways known to be important in MUC4 mucin (ref. 32; 24.6-fold), matrix metalloproteinase 9 pancreatic cancer (transforming growth factor–h signaling, cell (MMP9; ref. 33; 2.0-fold), Id-1 (ref. 34; 2.3-fold), urokinase cycle regulation, and apoptosis; data not shown). In addition, plasminogen activator receptor (uPAR; ref. 35; 13.5-fold), and we identified aberrant expression of a significant number of heparin-binding epidermal growth factor-like growth factor components of RA signaling (Table 2; Fig. 1). RAR-a and RAR-g (HB-EGF; ref. 36; 2.5-fold; Table 2). Other genes, yet to be were up-regulated 2.9- and 2.2-fold, respectively, in pancreatic characterized in pancreatic cancer but thought to be regulated by cancer compared with normal pancreas. Expression of a RA were also aberrantly expressed, including a RA-induced G- substantial number of known RA-responsive genes was also protein–coupled receptor (26.3-fold) and RAR responders altered in pancreatic cancer, consistent with dysregulated RA RARRES 1 (16.5-fold) and RARRES 3 (3.8-fold; Table 2). signaling activity, primarily demonstrating up-regulation of Studies of hindbrain development have provided the greatest genes downstream of RAR-a. A substantial number of genes insights into the mechanism of RA signaling. RA-dependent regulated by RA and known to be highly expressed in pancreatic lineage restriction in rhombomeres 3 and 5 is marked by
Clin Cancer Res 2005;11(9) May 1, 2005 3590 www.aacrjournals.org HOXB2 and Pancreatic Cancer krox20 and HOXB2 expression. RA, through an as yet HOXB2 expression in pancreatic cancer and pancreatic intra- unknown mechanism that may involve CEBPh (37), results epithelial neoplasia. Representative examples of HOXB2 in increased krox20 expression, which in turn increases immunostaining are shown in Fig. 2. Nuclear expression was HOXB2 expression by directly binding promoter elements of identified in 48 of 128 cancers (38%). When HOXB2 HOXB2 (38). krox20 also suppresses HOXB1 expression. The expression was present within the tumor, >80% of the nuclei expression profile in the present study is consistent with stained positively. HOXB2 expression was detected in the activity of this pathway of HOXB2 regulation (Fig. 1). In histologically normal pancreatic ducts of 2 of 26 (8%) patients, addition, the variant promyelocytic leukemia fusion protein in 1 of 24 (4%) PanIN-1A lesions, 3 of 20 (15%) PanIN-1B, 3 of PLZF-RARA also regulates HOXB2 expression through a 10 (30%) PanIN-2, and 1 of 4 (25%) PanIN-3 lesions, showing similar mechanism and is thought to be important in that HOXB2 expression occurs in PanIN and may play a role in promyelocytic leukemia development and resistance to RA the evolution of PanIN. therapy (39). For these reasons, HOXB2,apreviously HOXB2 expression in the whole cohort was associated with a uncharacterized gene in pancreatic cancer, which showed a poor outcome (median survival, 5 versus 9.9 months; log-rank 6.7-fold increase (P < 0.001) compared with normal pancreas, P < 0.0001; Fig. 3A). In addition, operative resection (P < was selected for further study. 0.0001), low-stage (P < 0.0001), and non–poorly differentiated
Ta b l e 2 . Components of RA signaling with differential expression between pancreatic cancer and normal pancreas on Affymetrix U133 microarrays
Probe set Unigene cluster Gene name Fold change P 204351_at Hs.2962 S10 0 (calcium-binding protein P) 152 0.001 203108_at Hs.194691 GPCR (RA-induced 3) 26.3 0.007 217109_at Hs.198267 MUC4 24.6 0.001 206392_s_at Hs.82547 RAR responder1 (RARRES1)16.50.004 205366_s_at Hs.98428 HOXB6 14.4 0.009 211924_s_at Hs.179657 UPAR 13.5 0.002 202859_x_at Hs.624 Interleukin 8 (IL8 ) 12.5 0.001 205453_at Hs.2733 HOXB2 6.7 0.001 219799_s_at Hs.179608 Retinol dehydrogenase homologue (hRADH)4.70.016 203596_s_at Hs.27610 RA- and IFN-inducible protein (IFT5) 4.2 0.010 204070_at Hs.17466 RAR responder 3 (RARRES 3) 3.8 0.005 228601_at Hs.93574 HOXD3 3.4 0.024 205249_at Hs.1359 Krox20 (EGR2) 3.2 0.008 231936_at Hs.40408 HOXC9 3.2 0.009 213844_at Hs.37034 HOXA5 3.0 0.035 203749_s_at Hs.250505 RAR-a 2.9 0.007 201042_at Hs.512708 TGM2 2.9 0.001 202510_s_at Hs.101382 TNFAIP2 2.9 0.004 204420_at Hs.283565 FOSL1 (FOS-like antigen-1) 2.8 0.005 205601_s_at Hs.22554 HOXB5 2.7 0.003 206858_s_at Hs.820 HOXC6 2.6 0.016 202575_at Hs183650 Cellular RA-binding protein 2 (CRABP2) 2.6 0.018 38037_at Hs.799 HB-EGF 2.5 0.009 2214782_at Novel gene similar to retinaldehyde-binding 2.4 0.036 protein (sRABP) 208937_s_at Hs.75424 Id-1 (inhibitor of DNA binding1) 2.3 0.039 201505_at Hs.82124 Laminin b1 2.2 0.026 204118_s_at Hs.1497 RAR-c 2.2 0.049 212501_at Hs.99029 CEBP b (CCAATenhancer-binding protein h)2.10.005 203936_s_at Hs.151738 MMP9 2.0 0.008 221701_s_at Hs.24553 STRA6 1. 9 0.0 3 0 202449_s_at Hs.20084 Retinoid X receptor a (RXR-a)0.40.005 231906_at Hs.301963 HOXD8 0.6 0.010 202882_x_at Hs.106346 RA-repressible protein (RARG-1) 0.6 0.045 207914_x_at Hs.336963 Even-skipped homeobox1 (EVX1) 0.5 0.016 208224_at Hs.99992 HOXB1 0.4 0.010 205883_at Hs.37096 Zinc finger protein145 (ZNF145, PLZF) 0.3 0.002 203423_at Hs.101850 Cellular RA-binding protein1 (CRABP1) 0.2 0.001 209496_at Hs.37682 RAR responder 2 (RARRES 2)0.10.001
www.aacrjournals.org3591 Clin Cancer Res 2005;11(9) May 1, 2005 Cancer Prevention
Fig. 2. Photomicrographs of HOXB2 nuclear staining (magnification, 200). A, negative HOXB2 expression in normal pancreatic duct. B, positive nuclear staining in PanIN-1Blesion. C,positivenuclear staining in PanIN-3 lesion. D,positive nuclear staining in pancreatic cancer. E, negative nuclear staining in PanIN-1A lesion. F, negative nuclear staining in pancreatic cancer. G, HOXB2 staining in pancreatic cancer without blocking peptide. H, serial section of (G) stained with anti-HOXB2 antibody after incubation with blocking peptide sc-17165 P.
Clin Cancer Res 2005;11(9) May 1, 2005 3592 www.aacrjournals.org HOXB2 and Pancreatic Cancer tumors (P = 0.0152) were associated with significantly 14.0 versus 3.7 months; log-rank P < 0.0001; Fig. 3C). improved survival using Kaplan-Meier analysis. However, Survival for patients with tumors that were HOXB2 negative multivariate analysis identified resection and stage as the only and who underwent resection was significantly longer than independent prognostic factors when modeled together with survival in all other groups (14 versus 4.3 months; log-rank degree of differentiation and HOXB2 status (Table 3A). Whereas P < 0.0001; Fig. 3D). Hence, in this cohort, lack of HOXB2 HOXB2 expression was identified in 32 of 52 (62%) unresected expression cosegregated with operative resectability, with only tumors, it was present in only 16 of 76 (21%) resected those who were HOXB2 negative having a survival advantage pancreatic cancers. Hence, HOXB2 expression was associated from operative resection. with nonresectable tumors (m2; P < 0.0001) and consequently Survival analysis of patients that underwent operative was not an independent prognostic factor. Operative resection resection identified decreased survival associated with HOXB2 did not benefit those patients whose tumors expressed nuclear expression (median survival, 6.75 versus 14.0 months; HOXB2 (log-rank P = 0.37; Fig. 3B) but was beneficial to log-rank P < 0.0001; Fig. 3E). Kaplan-Meier analyses identified those patients who did not express HOXB2 (median survival, clear margin status (P = 0.0014), tumor size of V20 mm
Fig. 3. Kaplan-Meier survival curves for (A) HOXB2 nuclear expression in the whole cohort. Effect of resection on prognosis in the following subgroups: (B)HOXB2 positive, (C) HOXB2 negative, (D)all patients stratified for HOXB2 status and resection. Kaplan-Meier survival curves for 76 patients who underwent surgical resection: (E) HOXB2 nuclear expression, (F) margin status, (G) tumor size, and (H) lymph node involvement.
www.aacrjournals.org3593 Clin Cancer Res 2005;11(9) May 1, 2005 Cancer Prevention
Ta b l e 3 . Multivariate analysis for clinicopathologic variables and HOXB2 expression in the whole cohort and resected pancreatic cancer
Variable Hazards ratio (95% confidence interval) P A.Whole cohort (n = 127) Stage III/IV versus I/II 2.30 (1.44-3.69) 0.005 Resection 0.43 (0.26-0.72) 0.0013 HOXB2 expression 1.56 (0.94-2.57) 0.085 B. Resected (n = 74) HOXB2 expression 2.90 (1.51-5.57) 0.0014 Margin involvement 1.89(1.02-3.48) 0.0428 Lymph node involvement 1.30 (0.71-2.40) 0.3981 C. Resected (n = 76) HOXB2 expression 2.82 (1.48-5.40) 0.0017 Margin involvement 2.04 (1.17-3.53) 0.0115 Tu m o r s i z e , >20 mm 1.48 (0.75-2.90) 0.2567 D. Resected (n = 74) HOXB2 expression 2.69 (1.39-5.20) 0.0032 Margin involvement 1.75 (0.94-3.25) 0.0777 Lymph node involvement 1.34(0.73-2.46) 0.3525 Tu m o r s i z e , >20 mm 1.49 (0.76-2.94) 0.2474
(P = 0.0375), and no lymph node involvement (P = 0.0235) as pancreatic fate (41). RARs also regulate exocrine pancreatic being associated with a survival advantage (Fig. 3F-H). Degree development at later stages (42), by modulating lineage of differentiation was not associated with a survival advantage selection favoring ductal rather than acinar differentiation (P = 0.0582). HOXB2 expression and involved surgical margins primarily through RAR-a (10, 11). Reactivation of develop- were independent prognostic factors when modeled against all mental pathways, specifically those that regulate exocrine cell combinations of HOXB2 expression, involved surgical margin, lineage, have been implicated previously in the early develop- lymph node involvement, and tumor size in the subgroup of ment of pancreatic cancer (25) and other pathways that patients who underwent surgical resection (Table 3B-D). determine duct cell versus acinar cell differentiation involving RA signaling, may also be important. In addition, forced Discussion expression of cellular retinol binding protein (CRABP1), a mediator of RA signaling, in transgenic mice results in the Expression profiling identified differential expression of a development of poorly differentiated pancreatic cancer (43), significant number of RA signaling pathway components and further supporting a role of aberrant RA signaling in pancreatic downstream responders in pancreatic cancer compared with cancer evolution. The transcript profile data presented here normal pancreas. These included some genes that are known to suggests the RA signaling pathway has a role in pancreatic be associated with pancreatic cancer: MUC4, MMP9, Id-1, cancer, specifically, a number of RA-responsive genes known to uPAR, HB-EGF, and S100P, as well as novel candidates. be important in pancreatic cancer and PanIN development Although there is substantial evidence implicating aberrant were aberrantly expressed in this study: MUC4 mucin is retinoid signaling in carcinogenesis (e.g., acute promyelocytic overexpressed in a significant proportion of pancreatic cancer leukemia; ref. 40), the mechanisms by which retinoid target (44) and PanIN (32) and can be induced through RAR-a genes exert these effects remains to be elucidated. Here we activation (45); similarly, MMP9 is expressed in pancreatic present evidence, implicating RA signaling in pancreatic cancer cancer (33) and is up-regulated by RA treatment (12) as is WAF1/CIP1 and show that a RA-responsive homeodomain transcription uPAR, HB-EGF, and p21 (46). Id-1, which antago- factor, HOXB2, is ectopically expressed in a significant nizes basic helix loop helix proteins, inhibits differentiation proportion of pancreatic cancer, with a profound association and can enhance cell proliferation is overexpressed in PanIN with tumor progression. HOXB2, which is not normally lesions (34) and is also RA responsive (47). In the present expressed in the pancreas at any stage during development or study, HOXB2 expression was also detected in PanIN lesions. adult life, was expressed in 38% of pancreatic cancers and RA seems to exert its effect on exocrine lineage selection seemed to occur during the development of a proportion of towards a ductal phenotype during development through PanIN. Ectopic HOXB2 expression was associated with non- laminin-h1 (11), which was up-regulated in the present study. resectable tumors and was an independent prognostic factor in As was the putative tumor suppressor gene RARRES 3 (48). resected tumors when modeled with known clinicopathologic Stra6, a gene whose function is yet to be determined responds prognostic factors. In addition, only those patients that were to RA, is up-regulated in colon cancer (49) and was up- HOXB2 negative obtained a survival advantage with operative regulated in this study as was transglutaminase 2 (TGM2) and resection. tumor necrosis factor a–induced protein 2 (TNFAIP2), both Numerous lines of evidence from separate studies suggest also RA-responsive genes (47). There is clear evidence the importance of individual RA signaling components in supporting RA regulation of HOXB2 expression from studies pancreas development and pancreatic cancer evolution. RA of hindbrain patterning and branchial arch development (38); regulates early instructive signals from lateral plate mesoderm however, the mechanism by which RA signaling imparts its that is essential for specification of endoderm towards a effects on the HOX network and cellular function is poorly
Clin Cancer Res 2005;11(9) May 1, 2005 3594 www.aacrjournals.org HOXB2 and Pancreatic Cancer understood. There were, however, some inconsistencies in the efficacy as a marker of prognosis in pancreatic cancer, as ectopic data where downstream targets of RA were down-regulated expression of a protein is a more reliable indicator than loss of such as HOXD8. Presumably, other mechanisms can also expression. Although pancreatic resection offers the best chance regulate the expression levels of these transcripts other than of cure in patients with pancreatic cancer, it is a procedure which through RA signaling alone. Validation of these data with carries significant morbidity and mortality. The development of manipulation of RA signaling is required to further investigate a reliable preoperative assessment of HOXB2 status would be an a putative functional role in pancreatic cancer; however, the important addition to a physician’s limited diagnostic arma- strength of data from the literature and evidence presented mentarium in this disease and may be used, together with here makes a strong case for an important role in pancreatic current clinicopathologic variables of disease outcome, to cancer. determine a patient’s suitability for operative resection. Multivariate analysis identified HOXB2 expression as an In conclusion, gene expression profiling of pancreatic cancer independent predictor of survival in the subgroup of patients has suggested that RA signaling is a potentially important that underwent pancreatic resection. Although HOXB2 expres- regulatory pathway in pancreatic cancer evolution. Ectopic sion was not identified as an independent predictor of survival expression of HOXB2 in pancreatic cancer is a frequent in the whole cohort due to its association with resection, lack of occurrence, an event which manifests in the development of HOXB2 expression combined with surgical resection conferred a PanIN in a proportion of cases, and is possibly a consequence of significant survival advantage. Because all known prognostic aberrant RA signaling. Current prognostic factors for pancreatic indicators in pancreatic cancer, such as tumor size, resection cancer remain poorly defined, depend upon examination of the margins, and lymph node status can only be determined post resected pancreas, and cannot be accurately determined preop- resection, HOXB2 expression has potential utility as a prognos- eratively. Assessment of HOXB2 expression may provide an tic indicator in pancreatic cancer, especially because it seems to alternative method for determining the suitability for resection have a profound independent influence on survival, with the and the prognosis of patients with pancreatic cancer. Further advantage that it can be assessed using biopsy techniques study to determine the effects of ectopic HOXB2 expression and without resection. We have previously identified that loss of other components of the HOX transcriptional network, its DPC4/Smad4 expression is associated with poor outcome in relationship to RA signaling, and clinical utility in pancreatic pancreatic cancer (28). However, HOXB2 may have more adenocarcinoma is required.
References 1. Yeo CJ, Cameron JL, Sohn TA, et al. Six hundred selection through epithelial-mesenchymal interac- 24. Cantile M, Pettinato G, Procino A, Feliciello I, fifty consecutive pancreaticoduodenectomies in the tions. Gastroenterology 2002;123:1331 ^ 40. CindoloL,CilloC.In vivo expression of the whole 1990s: pathology, complications, and outcomes. 12. Jaster R, Hilgendorf I, Fitzner B, et al. Regulation of HOX gene network in human breast cancer. Eur J Ann Surg 1997;226:248 ^ 57; discussion 257 ^ 60. pancreatic stellate cell function in vitro: biological and Cancer 2003;39:257 ^ 64. 2. Hruban RH, Adsay NV, Albores-Saavedra J, et al. molecular effects of all-trans retinoic acid. Biochem 25. Miyamoto Y, Maitra A, Ghosh B, et al. Notch medi- Pancreatic intraepithelial neoplasia: a new nomencla- Pharmacol 2003;66:633^41. ates TGF a-induced changes in epithelial differentia- ture and classification system for pancreatic duct 13. Mangelsdorf DJ, Evans RM. The RXR heterodimers tion during pancreatic tumorigenesis. Cancer Cell lesions. AmJ Surg Pathol 2001;25:579 ^ 86. and orphan receptors. Cell 1995;83:841 ^ 50. 2003;3:565^76. 3. Hruban RH, Takaori K, Klimstra DS, et al. An illus- 14. OosterveenT, Niederreither K, Dolle P, Chambon P, 26.Thayer SP, di Magliano MP, Heiser PW, et al. Hedge- trated consensus on the classification of pancreatic Meijlink F, Deschamps J. Retinoids regulate the anteri- hog is an early and late mediator of pancreatic cancer intraepithelial neoplasia (PanIN) and intraductal or expression boundaries of 5VHoxb genes in posterior tumorigenesis. Nature 2003;425:851 ^ 6. papillary mucinous neoplasms (IPMNs). Am J Surg hindbrain. EMBO J 2003;22:262 ^ 9. 27. Baugh LR, Hill AA, Brown EL, Hunter CP. Pathol 2004;28:977^87. 15. Duboule D, Morata G. Colinearity and functional Quantitative analysis of mRNA amplification by 4. Moskaluk CA, Hruban RH, Kern SE. p16 and K-ras hierarchy among genes of the homeotic complexes. in vitro transcription. Nucleic Acids Res 2001; gene mutations in the intraductal precursors of human Trends Genet 1994;10:358^64. 29:e29. pancreatic adenocarcinoma. Cancer Res 1997;57: 16. Abate-Shen C. Deregulated homeobox gene ex- 28. Biankin AV, Morey AL, Lee C-S, et al. DPC4/ 2140 ^ 3. pression in cancer: cause or consequence? Nat Rev Smad4 expression and outcome in pancreatic 5. van Heek NT, Meeker AK, Kern SE, et al. Telomere Cancer 2002;2:777 ^ 85. ductal adenocarcinoma. J Clin Oncol 2002;20: shortening is nearly universal in pancreatic intraepithe- 17. ApiouF,FlagielloD,CilloC,MalfoyB,PouponMF, 4531 ^ 42. lial neoplasia. Am J Pathol 2002;161:1541 ^ 7. Dutrillaux B. Fine mapping of human HOX gene clus- 29. Iacobuzio-Donahue CA, Ashfaq R, Maitra A, et al. 6. Biankin AV, Kench JG, Morey AL, et al. Overexpres- ters. Cytogenet Cell Genet 1996;73:114 ^ 5. Highly expressed genes in pancreatic ductal adeno- sion of p21WAF1/CIP1is an early event in the develop- 18. Graham A, Papalopulu N, Krumlauf R. The murine carcinoma: a comprehensive characterization and ment of pancreatic intraepithelial neoplasia. Cancer and Drosophila homeobox gene complexes have comparison of the transcript profiles obtained from Res 2001;61:8830 ^ 7. common features of organization and expression. Cell three major technologies. Cancer Res 2003;63: 7. Wilentz RE, Geradts J, Maynard R, et al. Inactivation 1989;57:367^78. 8614^ 22. of the p16 (INK4A) tumor-suppressor gene in pan- 19. Cillo C, Faiella A, Cantile M, Boncinelli E. Homeobox 30. Logsdon CD, Simeone DM, Binkley C, et al. Mole- creatic duct lesions: loss of intranuclear expression. genes and cancer. Exp Cell Res 1999;248:1 ^ 9. cular profiling of pancreatic adenocarcinoma and Cancer Res 1998;58:4740 ^ 4. 20. Cillo C, Barba P, Freschi G, Bucciarelli G, Magli MC, chronic pancreatitis identifies multiple genes differen- 8. DiGiuseppe JA, Hruban RH, Goodman SN, et al. Boncinelli E. HOX gene expression in normal and neo- tially regulated in pancreatic cancer. Cancer Res Overexpression of p53 protein in adenocarcinoma of plastic human kidney. IntJCancer1992;51:892 ^ 7. 2003;63:2649 ^ 57. the pancreas. AmJ Clin Pathol 1994;101:684 ^8. 21. De Vita G, Barba P, Odartchenko N, et al. Expres- 31. Crnogorac-JurcevicT, Missiaglia E, Blaveri E, et al. 9. Oosterveen T, van Vliet P, Deschamps J, Meijlink F. sion of homeobox-containing genes in primary and Molecular alterations in pancreatic carcinoma: expres- The direct context of a hox retinoic acid response ele- metastatic colorectal cancer. Eur J Cancer 1993;29A: sion profiling shows that dysregulated expression of ment is crucial for its activity. J Biol Chem 2003;278: 887 ^ 93. S100 genes is highly prevalent. J Pathol 2003;201: 24103 ^ 7. 22. Naora H, YangYQ, Montz FJ, Seidman JD, Kurman 63 ^74. 10. TulachanSS,DoiR,KawaguchiY,etal.All-trans RJ, Roden RB. A serologically identified tumor antigen 32. Swartz MJ, Batra SK, Varshney GC, et al. MUC4 retinoic acid induces differentiation of ducts and encoded by a homeobox gene promotes growth of expression increases progressively in pancreatic endocrine cells by mesenchymal/epithelial inter- ovarian epithelial cells. Proc Natl Acad Sci U S A intraepithelial neoplasia. Am J Clin Pathol 2002;117: actions in embryonic pancreas. Diabetes 2003;52: 2001;98:4060^5. 791 ^ 6. 76 ^ 84. 23. Lewis MT. Homeobox genes in mammary gland de- 33. GongYL, Xu GM, Huang WD, Chen LB. Expression 11. Kobayashi H, Spilde TL, Bhatia AM, et al. Retinoid velopment and neoplasia. Breast Cancer Res 2000;2: of matrix metalloproteinases and the tissue inhibitors signaling controls mouse pancreatic exocrine lineage 15 8 ^ 6 9. of metalloproteinases and their local invasiveness
www.aacrjournals.org3595 Clin Cancer Res 2005;11(9) May 1, 2005 Cancer Prevention
and metastasis in Chinese human pancreatic cancer. Krumlauf R, Zelent A. Regulation of Hoxb2 by 45. Choudhury A, Singh RK, Moniaux N, El-Metwally J Surg Oncol 2000;73:95 ^ 9. APL-associated PLZF protein. Oncogene 2003;22: TH, Aubert JP, Batra SK. Retinoic acid-dependent 34. Maruyama H, Kleeff J, Wildi S, et al. Id-1 and Id-2 3685^97. transforming growth factor-h 2-mediated induction are overexpressed in pancreatic cancer and in dys- 40. Melnick A, Licht JD. Deconstructing a disease: of MUC4 mucin expression in human pancreatic tu- plastic lesions in chronic pancreatitis. Am J Pathol RARa, its fusion partners, and their role in the patho- mor cells follows retinoic acid receptor-a signaling 1999;155:815^ 22. genesis of acute promyelocytic leukemia. Blood1999; pathway. J Biol Chem 2000;275:33929 ^ 36. 35. Cantero D, Friess H, Deflorin J, et al. Enhanced ex- 93:3167 ^ 215. 46. Liu T-X, Zhang J-W, Tao J, et al. Gene expression pression of urokinase plasminogen activator and its re- 41. Kumar M, Jordan N, Melton DA, Grapin-Botton A. networks underlying retinoic acid-induced differentia- ceptor in pancreatic carcinoma. Br J Cancer 1997; Signals from lateral plate mesoderm instruct endo- tion of acute promyelocytic leukemia cells. Blood 75:388 ^ 95. derm towards a pancreatic fate. Dev Biol 2003;259: 2000;96:1496^504. 36. Ito Y, Higashiyama S, Takeda T, Yamamoto Y, 109 ^22. 47. Ma Y, Koza-Taylor PH, DiMattia DA, et al. Micro- Wakasa KI, Matsuura N. Expression of heparin- 42. Kadison A, Kim J, Maldonado T, et al. Retinoid array analysis uncovers retinoid targets in human binding epidermal growth factor-like growth factor signaling directs secondary lineage selection in bronchial epithelial cells. Oncogene 2003;22: in pancreatic adenocarcinoma. Int J Gastrointest pancreatic organogenesis. J Pediatr Surg 2001;36: 4924 ^32. Cancer 2001;29:47 ^52. 115 0 ^ 6 . 48. DiSepio D, Ghosn C, Eckert RL, et al. Identification 37. Duprez E, Wagner K, Koch H, Tenen DG. C/EBPb: a 43. Giguere V. Retinoic acid receptors and cellular reti- and characterization of a retinoid-induced class II major PML-RARA responsive gene in retinoic acid- noid binding proteins: complex interplay in retinoid tumor suppressor/growth regulatory gene. Proc Natl induced differentiation of APL cells. EMBO J 2003; signaling. Endocr Rev 1994;15:61 ^ 79. Acad Sci U S A1998;95:14811^5. 22:5806 ^ 16. 44. Balague C, Gambus G, Carrato C, et al. Altered ex- 49. Szeto W, Jiang W,Tice DA, et al. Overexpression of 38. Gavalas A. ArRAnging the hindbrain. Trends Neuro- pression of MUC2, MUC4, and MUC5 mucin genes in the retinoic acid-responsive gene Stra6 in human sci 2002;25:62 ^ 4. pancreas tissues and cancer cell lines. Gastroentero- cancers and its synergistic induction by Wnt-1 and 39. Ivins S, Pemberton K, Guidez F, Howell L, logy1994;106:1054^61. retinoic acid. Cancer Res 2001;61:4197 ^ 205.
Clin Cancer Res 2005;11(9) May 1, 2005 3596 www.aacrjournals.org CYCLIN E EXPRESSION AND OUTCOME IN PANCREATIC DUCTAL
ADENOCARCINOMA
DA Skalicky1, JG Kench1,3,DSegara1,MJColeman2,RLSutherland1,SMHenshall1, EA Musgrove1 and AV Biankin1,2,4
1Cancer Research Program, Garvan Institute of Medical Research and 2Division of Surgery, St. Vincent's Clinic, Sydney, NSW 2010 AUSTRALIA, 3Institute of Clinical Pathology and Medical Research, Westmead Hospital, Sydney, NSW 2140. 4Division of Surgery, Bankstown-Lidcombe Hospital, Eldridge Road, Bankstown, Sydney, NSW 2200 AUSTRALIA.
Keywords: Cyclin E, pancreatic cancer, prognosis, outcome, p27 Research Article: Early Detection and Diagnosis
This paper reports for the first time that cyclin E expression is a strong independent prognostic factor in pancreatic cancer, making cyclin E a potentially clinically useful prognostic marker in PC
Corresponding author: Andrew V. Biankin Cancer Research Program, Garvan Institute of Medical Research 384 Victoria St., Darlinghurst, NSW 2010. AUSTRALIA Tel: 61 2 9295 8330 Fax: 61 2 9295 8321 Email: [email protected]
Abstract
The association of high Cyclin E expression with poor outcome in some cancers, in particular breast cancer, suggests that it may play an important role in tumor biology. Since the influence of cyclin E expression on outcome is yet to be examined in pancreatic cancer (PC), we assessed the relationship between the expression of cyclin E, p27Kip1 and survival in a large cohort of pancreatic cancer patients with long term follow-up. Expression of cyclin E and p27Kip1 was assessed by immunohistochemistry using tissue microarrays of tumor samples from 118 patients with pancreatic ductal adenocarcinoma (75 resections and 43 biopsies). High cyclin E expression (>10% positive nuclei) was identified in 39 of 118 (33%) patients. This was associated with poor prognosis on univariate analysis in the whole cohort (p=0.005), as well as in the subgroup of 75 patients who underwent operative resection (p=0.04). On multivariate analysis high cyclin E expression was an independent predictor of poor survival in both the entire cohort (p=0.005) and the resected subgroup (p=0.03), and was superior to all tested clinicopathological factors (tumor size, lymph node metastases, differentiation, margin involvement and perineural invasion) as a marker of survival. Low p27Kip1 expression (<5% positive nuclei) was present in 41 of 111 (37%) patients, but was not associated with survival, and co-expression of p27Kip1 did not influence the association of high cyclin E expression with poor survival. High cyclin E expression is a strong independent predictor of poor outcome in patients with pancreatic cancer. Thus, if these data are confirmed in independent cohorts, measurement of cyclin E may add significant prognostic information to the currently used clinicopathological parameters and hence have potential clinical utility in the management of this disease. 2
Introduction
Pancreatic cancer (PC) is the fourth leading cause of cancer death in men and women in Western societies, with a 5- year survival rate of less than 10% 1. PC presents at an advanced stage, and as a result, only 10-20% of patients are suitable for surgical treatment at the time of presentation 1. Clinical management of these patients is complicated by inconsistencies in the influence of conventional clinicopathological variables on outcome suggesting that some of these parameters lack accuracy. In addition, preoperative assessment of some parameters such as lymph node metastases is difficult. Whereas in other cancers assessment of aberrations in gene expression that cosegregate with therapeutic response and outcome is being adopted to increase predictive power (e.g. ER and HER2/neu,and potentially cyclin E in breast cancer), there remain no molecular markers of proven clinical utility in PC. This highlights the need to identify novel molecular markers of prognosis relevant to PC that may also have diagnostic and therapeutic utility.
Dysregulation of the normal cell cycle machinery is integral to the neoplastic process and there is now compelling evidence that the development and progression of most human cancers is associated with a high frequency of 2 abnormalities in the retinoblastoma (RB) pathway that controls G1 to S phase progression . The elements in the RB pathway that have been implicated in cancer include RB itself, a tumor suppressor that in its underphosphorylated state represses the transcription of genes necessary for cell cycle progression, the cyclin-dependent kinases (CDKs) that phosphorylate RB during G1 phase, and the cyclins and CDK inhibitors that regulate CDK activity. Cdk4 and Cdk6 are activated by the D-type cyclins, cyclins D1, D2 and D3, and initially phosphorylate RB during G1. Subsequent RB phosphorylation by cyclin E-Cdk2 then relieves RB inhibition of cell cycle progression and allows initiation of DNA synthesis. Recent studies have identified functions of cyclin E in addition to Cdk2 activation and RB phosphorylation that may contribute to tumorigenesis; these include initiation of DNA replication, genomic instability and centrosome amplification 3. The CDKs are regulated at multiple levels, including regulation of the abundance of the activating cyclins and two families of endogenous low molecular weight inhibitors. The INK4 family of CDK inhibitors includes p16INK4A and specifically targets cyclin D-associated CDKs. The CIP/KIP family includes p27Kip1 and p21WAF1/Cip1, which preferentially inhibit the activity of cyclin E-Cdk2 complexes. Many common cancers display deletion or mutation of RB, overexpression of cyclins D1 or E, or reduced expression of p16INK4A or p27Kip1 2. In pancreatic ductal adenocarcinomas, alterations in cell cycle regulators (cyclins D1 and D3, p16INK4A, p21WAF1/CIP1 and p27Kip1) have been reported at high frequencies 4, are apparent in the precursor lesions of PC: pancreatic intraepithelial neoplasia (PanIN) 5-8 and intraductal papillary mucinous neoplasms (IPMN) 9, 10, but do not cosegregate with survival. Despite the association between aberrant cyclin E expression and outcome in other cancers 11-13, in particular the strong association of high cyclin E expression and poor survival in breast cancer 12, 13, the influence of cyclin E on outcome in PC has not been addressed.
Here we report a strong association between high cyclin E expression and poor outcome in PC. Furthermore, high cyclin E expression was an independent prognostic factor in resected PC that was a better indicator of patient outcome than traditional clinicopathologic parameters.
Methods
Tumor Samples and Patient Population We identified a cohort of patients with a diagnosis of pancreatic ductal adenocarcinoma that underwent pancreatic resection or biopsy at Westmead Hospital, Concord Hospital, Royal Prince Alfred Hospital or the St Vincent's Hospital Campus in Sydney, Australia between 1972 and 2001. Multicenter ethical approval for data collection and tissue use was granted by the St Vincent’s Hospital/ University of NSW, and Royal Prince Alfred Hospital Ethics Committees, as well as specific approval from the Westmead Hospital and Concord Hospital Ethics Committees. Cyclin E and p27Kip1 expression were evaluated in 118 patients. The clinicopathological and survival data presented in Table 1 are confined to the whole cohort of 118 patients who had either pancreatic resection or biopsy, and a subgroup of 75 patients treated with operative resection. Eighty-two of this cohort were included in a previously described group of 139 patients examining prognostic factors in PC 14, while the remaining 36 have been accrued since July 1999. 3
Clinical parameters including sex, age at diagnosis, preoperative assessment of disease state and type of operative procedure were gathered retrospectively from patient records. Pathological findings including tumor size, UICC T stage and lymph node status were obtained from the pathologists' original report. In addition, microscopic findings (tumor type, degree of differentiation, perineural invasion and margin status) were independently reassessed by a pathologist (J.G.K.). Date and cause of death was obtained on a return-to-notifier basis from The Cancer Council, NSW.
Immunohistochemistry Seven pancreatic cancer tissue arrays consisting of 2 mm diameter tissue core biopsies were constructed from the archival tissue. Four- m sections of the tissue arrays were dewaxed in xylene and rehydrated through graded alcohol concentrations. Antigen retrieval was achieved using EDTA buffer solution (pH 8.0) at high pressure in a microwave for 30 minutes. Slides were then transferred to a DAKO autostainer (DAKO Corporation, Carpinteria, CA) and endogenous peroxidase activity was quenched by 3% hydrogen peroxide treatment before incubation with mouse monoclonal antibody against cyclin E (clone 13A3, Novocastra, Newcastle-upon-Tyne, United Kingdom) or p27Kip1 (clone 57, Transduction Laboratories, Lexington, KY). For evaluation of cyclin E protein expression, MDA- MB-157 human breast cancer cells (which display amplification and overexpression of cyclin E 15) and normal placenta were used as positive controls, and normal testis as a negative tissue control. Epithelial nuclei at the bases of normal duodenal crypts stained strongly, providing internal positive controls within the tissue arrays. IgG2a- negative mouse serum was used as a technical control to demonstrate antibody specificity. The positive control for Kip1 p27 staining was normal skeletal muscle, while normal spleen was used as a negative control. IgG1-negative mouse serum served as a technical control to demonstrate antibody specificity. The primary antibody was visualised using the DAKO Envision+ secondary detection system followed by color development using 3,3-diaminobenzidine (DAKO Corporation, Carpinteria, CA). Sections were counterstained using hematoxylin.
Up to 4 separate cores of pancreatic cancer were examined per patient. Staining was assessed by two independent blinded observers for each case (D.A.S. and J.G.K.). Standardization of scoring was achieved by comparison of scores between observers and by multiviewer microscope conferencing, where any discrepancies were resolved by consensus. The following criteria were used to dichotomize each antigen: cyclin E expression was considered high if >10% of nuclei were positive; p27Kip1 expression was considered high if 5% of nuclei were stained. This threshold for cyclin E was between the mean and the median of the distribution of scores. For p27Kip1 published studies in pancreatic cancer have used either >1% or >5% 16, 17; we chose 5% as the threshold since this was closer to the median score for this cohort. When more than 1 core was present for a particular cancer, the highest scoring core was used in the analysis.
Statistical Analysis The association between cyclin E or p27Kip1 protein expression and clinicopathological parameters including clinical stage, degree of differentiation, operative resection, perineural invasion, tumor size, lymph node invasion and surgical margin involvement were evaluated using nonparametric statistical tests (Kruskal-Wallis or Mann- Whitney).
Both univariate and multivariate analyses were performed to assess the association of cyclin E and survival in relation to covariates using Statview 5.0 Software (Abacus Systems, Berkeley, CA). A P value of <0.05 was considered to be statistically significant. The outcome variables were assessed as time-to-event, which was defined as the difference between the time of diagnosis and the time of death.
For univariate analysis, event-survival curves were constructed using the Kaplan-Meier method for each cyclin E category. The differences in survival times between categories were compared using the two-tailed log-rank statistic. In addition, survival curves for the coexpression of cyclin E with p27Kip1 were plotted. Cox proportional hazards models were used to estimate hazard ratio (and its 95% CI) associated with each risk factor and covariate. Those factors that were prognostic on univariate analysis were then assessed in multivariable models to identify factors that were independently prognostic. 4 Results
Cohort characteristics The cohort consisted of 118 patients with a histological diagnosis of pancreatic ductal adenocarcinoma (64 males and 54 females). Clinicopathological and survival data for these patients are presented in Table 1. The mean age at diagnosis was 64.8 years (range: 34.4-89.7). Tissue was available from 75 patients who had undergone pancreatic resections and 43 who had intraoperative biopsies. A large proportion of the cancers were either moderately (66/118, 56%) or poorly (43/118, 37%) differentiated with only 9 (7%) well differentiated tumors. Well and moderately differentiated tumors were grouped together for analysis. The majority of patients had advanced stage disease (23% UICC Stage I, 9% stage II, 59% Stage III and 9% Stage IV). Stage I and II tumors were grouped together for analysis and compared to stage III and IV tumors which were also grouped together.
The median follow-up for the cohort was 8.5 months (range: 0-117.4). Of the 118 patients, seven patients were alive at the census date (September 21st, 2002), 105 had died from pancreatic cancer, 2 due to other causes (post-operative deaths), and 4 were lost to follow-up. The overall median survival was 8.5 months and the disease-specific survival was 8.7 months. The overall disease-specific 1-year survival rate was 35% while only one patient survived longer than five years. The actuarial 5 year survival rate for patients who underwent resection was 10%. On univariate analysis for the whole cohort, those patients that underwent pancreatectomy survived longer (logrank, P < 0.0001) as did those with a lower clinical stage (logrank P < 0.0001) and those with non-poorly differentiated tumors (logrank P = 0.008; Table 1, Figure 1A-C). For those patients who underwent pancreatectomy, those with a tumor size of less than 20 mm survived longer (logrank P = 0.04), as did those with surgical margin clear of tumor (logrank P = 0.002) and those without lymph node involvement (logrank P = 0.02; Table 1, Figure 2A-C). However, neither degree of differentiation (logrank P = 0.053) nor perineural invasion (logrank P =0.11) influenced survival.
Cyclin E expression Nuclear cyclin E immunostaining was identified in 101 of 118 patients; the remainder had no detectable staining (Figure 3A-B). Beyond a threshold of 10%, high expression of cyclin E was detected in 39 of 118 (33%) of the whole cohort and 24 of 75 (32%) of the resected cohort. On univariate survival analysis, high cyclin E expression was associated with shortened survival in the whole cohort (logrank P = 0.005; HR = 1.8, 95% CI 1.2-2.7) with a median survival of 6.4 months for patients with high cyclin E expression, compared with 9.8 months for those with low cyclin E expression. Similarly, high cyclin E expression was associated with reduced survival in the patients who underwent resection (median survival 8.5 months vs 14.2 months, logrank P = 0.03; HR = 1.8, 95% CI 1.1-3.1, Table 1, Figure 1D, 2D).
Cyclin E did not cosegregate with any clinicopathological parameters assessed (data not shown). We next performed multivariate analysis on the whole and resected cohorts including all clinicopathological parameters which were significant predictors of survival on univariate analysis. Cyclin E overexpression was an independent marker of prognosis in both the whole cohort and the resected subgroup (Table 2). Other independent predictors of poor outcome in the whole cohort were poor differentiation, advanced clinical stage (stages III or IV versus stages I or II) and operative biopsy versus resection. In the resected subgroup, in addition to high cyclin E expression, tumor size >20 mm and positive surgical margins were independent markers of poor prognosis. In Table 2, grouping A shows the resolved model for the whole cohort, while B shows the multivariate model for resected tumors prior to resolution to the final model shown in Table 2C.
Coexpression of cyclin E and p27Kip1 Since cyclin E functionally interacts with p27Kip1 and this interaction appears to be prognostically significant in breast cancer 11-13, we evaluated the influence of cyclin E and p27Kip1 coexpression on outcome. Low p27Kip1 expression alone, i.e. <5% of nuclei positive for p27Kip1, did not correlate with any clinical or pathological parameters, nor was it associated with survival in either the whole cohort (P = 0.42) or in those patients that underwent resection (P = 0.38). Differential expression of p27Kip1 did not significantly alter the association of cyclin E with survival in patients with high cyclin E expression (logrank P = 0.73) or low cyclin E expression (logrank P= 0.93). 5 Discussion
In this study we present the first report concerning cyclin E expression and outcome in patients with PC. High cyclin E expression was associated with a poor prognosis on univariate analysis, and was the most influential prognostic factor on multivariate analysis compared to clinicopathologic parameters in patients that underwent pancreatectomy. These findings suggest a potentially important role for cyclin E in the clinical behavior of PC supporting similar data seen in breast cancer where high cyclin E expression is a powerful predictor of outcome 12, 13. Hence cyclin E expression is a potentially clinically useful prognostic marker for PC creating scope for the application in PC of novel therapeutic strategies being developed that target aberrant cyclin E.
Numerous studies have assessed the prognostic value of clinicopathologic parameters in PC 18-22. Only involved surgical margins and large tumor size are consistently associated with poor patient outcome 14. These factors are either difficult to identify or indeterminable preoperatively. The prognostic significance of molecular markers in PC, including growth factors and their receptors, cell cycle regulators, oncogenes, tumor suppressor genes, apoptotic factors, angiogenic and stromal factors have been assessed 23. Although some are associated with outcome, with the exception of DPC4/Smad4 14 and HOXB2 24 they are not independent of clinicopathologic factors as determinants of prognosis, thus limiting their clinical utility. Given the importance of cell cycle deregulation in carcinogenesis, and that aberrant expression of any other cell cycle regulatory molecules important in PC (cyclin D1, p53, p16INK4A, p21WAF1/CIP1 and p27Kip1) have not been associated with outcome to date (reviewed in 23 and our own unpublished data), the association of high cyclin E expression and poor prognosis in PC presented here suggests a particular importance for cyclin E deregulation in PC.
Oncogenic effects of cyclin E are usually attributed to its ability to promote cell cycle progression through phosphorylation of RB but recent data provide increasing evidence for induction of genetic instability as a likely contributor 3. Cyclin E overexpression in cultured cells leads to chromosome loss, apparently via generalised chromosomal instability 25. Potential mechanisms for this effect include interference with the assembly of a pre- replication complex at origins of replication 26, and centrosome amplification, which can contribute to aneuploidy by induction of mitotic spindle defects 27. Aneuploidy has been reported as a poor prognostic factor in some studies 28, and it would be of interest to examine the relationship between aneuploidy and cyclin E expression.
Recent work by some groups has concentrated on the role of low molecular weight (LMW) forms of cyclin E in cancer. It appears that truncated LMW forms of cyclin E are highly prevalent in, and may be unique to cancer 29. These LMW forms appear to be resistant to p21WAF1/CIP1 and p27Kip1 inhibition 30, and like normal cyclin E can promote cell cycle progression through mechanisms independent of RB phosphorylation 3, 31. Truncated LMW forms lack the C-terminus of the molecule, and attempts at detection through the use of antibodies directed at this region may explain some conflicting studies concerning the association of high cyclin E expression and prognosis in breast cancer and other cancers such as colon cancer 3. Antibodies directed at the N-terminus or the full length molecule (as used in this study) are thought to detect all forms of cyclin E 13. The lack of influence of p27Kip1 expression on prognosis either alone or stratified by cyclin E expression in PC does not recapitulate that seen in breast cancer and suggests that Cdk2-independent mechanisms may contribute to the effects of cyclin E in PC. More studies are required to elucidate the mechanisms of enhanced cyclin E activity in cancer development and progression, particularly in PC.
Only 2 previous studies have examined cyclin E expression in PC. Schraml et al. identified expression of cyclin E in 1ofonly8PCsamples32. However, another study demonstrated high expression of cyclin E in 75% of 38 PC samples, and it was also detectable in 25% of PanIN-3 lesions 7. These and the present study used the same antibody generated to the full length cyclin E protein and the differences in the proportion of PC expressing high levels of cyclin E is likely a reflection of sample variability and different scoring thresholds to define high expression levels rather than lack of detection of LMW forms of cyclin E.
In summary, we have demonstrated that high cyclin E expression is an independent prognostic factor in PC. In patients who underwent pancreatectomy, high cyclin E expression was a superior predictor of survival than clinicopathological parameters including tumor size, margin involvement and lymph node status. Hence cyclin E is a potentially clinically useful prognostic marker in PC suggesting a simple, well described test that can be implemented in the routine pathology laboratory in a disease where there are few useful markers of prognosis. If confirmed in other independent cohorts, these data create scope for the potential application of novel therapies targeting aberrant cyclin E, a marker of poor prognosis cancers, in PC. 6 Acknowledgements:
This work was supported by the National Health and Medical Research Council of Australia, The Cancer Council NSW, the St Vincent’s Clinic Foundation, the Royal Australasian College of Surgeons and the Prostate Cancer Foundation of Australia and the R. T. Hall Trust. AVB, SMH and EAM are supported by Fellowships from the Cancer Institute NSW. We also thank Dr. Catherine Langusch for her help with the maintenance of the Garvan Institute of Medical Research Pancreatic Cancer tissue bank and clinical database.
References
1. Yeo CJ, Cameron JL, Sohn TA, Lillemoe KD, Pitt HA, Talamini MA, Hruban RH, Ord SE, Sauter PK, Coleman J, Zahurak ML, Grochow LB, et al. Six hundred fifty consecutive pancreaticoduodenectomies in the 1990s: pathology, complications, and outcomes. Annals of Surgery 1997;226(3):248-57; discussion 57-60. 2. Malumbres M, Barbacid M. To cycle or not to cycle: a critical decision in cancer. Nat Rev Cancer 2001;1(3):222-31. 3. Hwang HC, Clurman BE. Cyclin E in normal and neoplastic cell cycles. Oncogene 2005;24(17):2776-86. 4. Rozenblum E, Schutte M, Goggins M, Hahn SA, Panzer S, Zahurak M, Goodman SN, Sohn TA, Hruban RH, Yeo CJ, Kern SE. Tumor-suppressive pathways in pancreatic carcinoma. Cancer Res 1997;57:1731-34. 5. Wilentz RE, Geradts J, Maynard R, Offerhaus GJA, Kang M, Goggins M, Yeo CJ, Kern SE, Hruban RH. Inactivation of the p16 (INK4A) tumor-suppressor gene in pancreatic duct lesions: loss of intranuclear expression. Cancer Res 1998;58:4740-44. 6. Biankin AV, Kench JG, Morey AL, Lee C-S, Biankin SA, Head DR, Hugh TB, Henshall SM, Sutherland RL. Overexpression of p21WAF1/CIP1 is an early event in the development of pancreatic intraepithelial neoplasia. Cancer Res 2001;61:8830-37. 7. Al-Aynati MM, Radulovich N, Ho J, Tsao MS. Overexpression of G1-S cyclins and cyclin-dependent kinases during multistage human pancreatic duct cell carcinogenesis. Clin Cancer Res 2004;10(19):6598-605. 8. Hruban RH, Takaori K, Klimstra DS, Adsay NV, Albores-Saavedra J, Biankin AV, Biankin SA, Compton C, Fukushima N, Furukawa T, Goggins M, Kato Y, et al. An illustrated consensus on the classification of pancreatic intraepithelial neoplasia and intraductal papillary mucinous neoplasms. Am J Surg Pathol 2004;28(8):977-87. 9. Biankin AV, Biankin SA, Kench JG, Morey AL, Lee C-S, Head DR, Eckstein RP, Hugh TB, Henshall SM, Sutherland RL. Aberrant p16INK4A and DPC4/Smad4 expression in intraductal papillary mucinus tumours of the pancreas is associated with invasive ductal adenocarcinoma. Gut 2002;50:861-68. 10. Biankin AV, Kench JG, Biankin SA, Lee CS, Morey AL, Dijkman FP, Coleman MJ, Sutherland RL, Henshall SM. Pancreatic intraepithelial neoplasia in association with intraductal papillary mucinous neoplasms of the pancreas: implications for disease progression and recurrence. Am J Surg Pathol 2004;28(9):1184-92. 11. Takahashi S, Kamata Y, Tamo W, Koyanagi M, Hatanaka R, Yamada Y, Tsushima T, Takaya S, Fukuda I. Relationship between postoperative recurrence and expression of cyclin E, p27, and Ki-67 in non-small cell lung cancer without lymph node metastases. Int J Clin Oncol 2002;7(6):349-55. 12. Porter PL, Malone KE, Heagerty PJ, Alexander GM, Gatti LA, Firpo EJ, Daling JR, Roberts JM. Expression of cell-cycle regulators p27Kip1 and cyclin E, alone and in combination, correlate with survival in young breast cancer patients. Nat Med 1997;3(2):222-5. 13. Keyomarsi K, Tucker SL, Buchholz TA, Callister M, Ding Y, Hortobagyi GN, Bedrosian I, Knickerbocker C, Toyofuku W, Lowe M, Herliczek TW, Bacus SS. Cyclin E and survival in patients with breast cancer. N Engl J Med 2002;347(20):1566-75. 14. Biankin AV, Morey AL, Lee C-S, Kench JG, Biankin SA, Head DR, Hugh TB, Sutherland RL, Henshall SM. DPC4/Smad4 expression and outcome in pancreatic ductal adenocarcinoma. J Clin Oncol 2002;20:4531-42. 15. Buckley MF, Sweeney KJ, Hamilton JA, Sini RL, Manning DL, Nicholson RI, deFazio A, Watts CK, Musgrove EA, Sutherland RL. Expression and amplification of cyclin genes in human breast cancer. Oncogene 1993;8(8):2127-33. 16. Feakins RM, Ghaffar AH. p27 Kip1 expression is reduced in pancreatic carcinoma but has limited prognostic value. Human Pathology. 2003;34(4):385-90. 17. Lu CD, Morita S, Ishibashi T, Hara H, Isozaki H, Tanigawa N. Loss of p27Kip1 expression independently predicts poor prognosis for patients with resectable pancreatic adenocarcinoma. Cancer. 1999;85(6):1250-60. 18. Geer RJ, Brennan MF. Prognostic indicators for survival after resection of pancreatic adenocarcinoma. Am J Surg 1993;165:68- 73. 19. Nitecki SS, Sarr MG, Colby TV, van Heerden JA. Long-term survival after resection for ductal adenocarcinoma of the pancreas. Is it really improving? Ann Surg 1995;221(1):59-66. 20. Gudjonsson B. Treatment and survival in 13,560 patients with pancreatic cancer, and incidence of the disease, in the west Midlands: an epidemiological study [letter; comment]. British Journal of Surgery 1996;83(6):874. 21. Kawesha A, Ghaneh P, Andren-Sandberg A, Ograed D, Skar R, Dawiskiba S, Evans JD, Campbell F, Lemoine N, Neoptolemos JP. K-ras oncogene subtype mutations are associated with survival but not expression of p53, p16(INK4A), p21(WAF-1), cyclin D1, erbB-2 and erbB-3 in resected pancreatic ductal adenocarcinoma. Int. J. Cancer 2000;89(6):469-74. 22. Sohn TA, Yeo CJ, Cameron JL, Koniaris L, Kaushal S, Abrams RA, Sauter PK, Coleman J, Hruban RH, Lillemoe KD. Resected adenocarcinoma of the pancreas-616 patients: results, outcomes, and prognostic indicators. J Gastrointest Surg 2000;4(6):567-79. 23. Ghaneh P, Kawesha A, Evans JD, Neoptolemos JP. Molecular prognostic markers in pancreatic cancer. J Hepato-Biliary- Pancreatic Surg 2002;9(1):1-11. 24. Segara D, Biankin AV, Kench JG, Langusch CC, Dawson AC, Skalicky DA, Gotley DC, Coleman MJ, Sutherland RL, Henshall SM. Expression of HOXB2, a retinoic acid signaling target in pancreatic cancer and pancreatic intraepithelial neoplasia. Clin Cancer Res 2005;11(9):3587-96. 25. Spruck CH, Won KA, Reed SI. Deregulated cyclin E induces chromosome instability. Nature 1999;401(6750):297-300. 26. Ekholm-Reed S, Mendez J, Tedesco D, Zetterberg A, Stillman B, Reed SI. Deregulation of cyclin E in human cells interferes with prereplication complex assembly. J Cell Biol 2004;165(6):789-800. 27. Matsumoto Y, Maller JL. A centrosomal localization signal in cyclin E required for Cdk2-independent S phase entry. Science 2004;306(5697):885-8. 28. Yeo CJ, Cameron JL, Lillemoe KD, Sitzmann JV, Hruban RH, Goodman SN, Dooley WC, Coleman J, Pitt HA. Pancreaticoduodenectomy for cancer of the head of the pancreas. 201 patients. Annals of Surgery 1995;221(6):721-31; discussion 31-3. 29. Akli S, Keyomarsi K. Low-molecular-weight cyclin E: the missing link between biology and clinical outcome. Breast Cancer Res 2004;6(5):188-91. 30. Wingate H, Zhang N, McGarhen MJ, Bedrosian I, Harper JW, Keyomarsi K. The Tumor-specific Hyperactive Forms of Cyclin E Are Resistant to Inhibition by p21 and p27. J Biol Chem 2005;280(15):15148-57. 7 31. Akli S, Zheng PJ, Multani AS, Wingate HF, Pathak S, Zhang N, Tucker SL, Chang S, Keyomarsi K. Tumor-specific low molecular weight forms of cyclin E induce genomic instability and resistance to p21, p27, and antiestrogens in breast cancer. Cancer Res 2004;64(9):3198-208. 32. Schraml P, Bucher C, Bissig H, Nocito A, Haas P, Wilber K, Seelig S, Kononen J, Mihatsch MJ, Dirnhofer S, Sauter G. Cyclin E overexpression and amplification in human tumours. Journal of Pathology. 2003;200(3):375-82. 8 Table 1: Clinicopathologic and outcome data for all patients in the cohort. Parameter Whole Resected Cohort n= 118 Cohort n=75 Median p value Median p value No. (%) Survival No. (%) Survival (months) (logrank) (months) (logrank) Sex Female 54 (46) 31 (41) Male 64 (54) 44 (59)
Age at diagnosis (years) Mean 64.8 62.3 Median 66.7 65.0 Range 34.4 – 89.7 34.4 – 82.6
Treatment Resection 75 (64) 11.2 Operative Biopsy 43 (36) 4.7 < 0.0001
Outcome Follow-up (months) 0 - 117 0.2 - 117 Median 8.5 11.9 30 day mortality 2 (2) 2 (3) Death from PC 103 (87) 60 (80) Death from other cause 2 (2) 2 (3) Alive 7 (6) 7 (9) Lost to follow-up 4 (3) 4 (5)
Stage I 27 (23) II 11 (9) 14.0* III 69 (59) IV 11 (9) 7.2 < 0.0001
Differentiation Well 9(7) 6(8) Moderate 66 (56) 9.5** 44 (59) 12.9 Poor 43 (37) 5.8 0.008 25 (33) 9.6 0.053
Tumor size 20mm 15 (20) 9.9 >20mm 60 (80) 17.1 0.04
Margins Clear 40 (53) 14.5 Involved 35 (47) 8.5 0.002
Lymph node status§ Negative 38 (51) 14.0 Positive 36 (49) 9.5 0.02
Perineural invasion Negative 31 (41) 11.0 Positive 44 (59) 11.2 0.11
Cyclin E expression Low (10% nuclear) 79 (67) 9.8 51 (68) 14.2 High (>10% nuclear) 39 (33) 6.4 0.005 24 (32) 8.5 0.03 p27Kip1 expression¶ Low (<5% nuclear) 41 (37) 9.7 25 (35) 14.8 High (5% nuclear) 70 (63) 7.4 0.42 47 (65) 10.1 0.38 * Well and Moderately differentiated tumors grouped together for analysis ** Stage I and II tumors versus stage III and IV § Lymph node status was only available in 74 patients in the resected cohort ¶ p27Kip1 expression was only available in 111 patients for the whole cohort and 72 patients in the resected cohort 9 Table 2: Multivariate analysis for clinicopathological parameters and cyclin E expression in the whole and resected cohorts of pancreatic cancer.
Variable Hazard Ratio p-value (95% confidence interval)
A. Whole cohort (n = 118) High cyclin E expression 1.71 (1.12 – 2.63) 0.0128 Operative resection 2.74 (1.79 – 4.19) <0.0001 Stage III/IV vs I/II 1.95 (1.91 – 3.19) 0.0079 Poor differentiation 1.56 (1.04 – 2.35) 0.0335
B. Resected cohort (n = 74) High cyclin E expression 2.39 (1.30 – 4.37) 0.0048 Tumor size > 20 mm 2.22 (1.07 – 4.58) 0.0315 Margin involvement 1.92 (1.03 – 3.56) 0.0388 Lymph node involvement 1.18 (0.62 – 2.22) 0.6069
C. Resected cohort (n = 75) High cyclin E expression 2.48 (1.39 – 4.41) 0.0021 Tumor size > 20mm 2.26 (1.09 – 4.68) 0.0276 Margin involvement 2.09 (1.19 – 3.66) 0.0104
C. is the resolved model of B. eliminating redundant variables
Figure Legends:
Figure 1: Kaplan-Meier survival curves for the whole cohort.
Figure 2: Kaplan-Meier survival curves for patients who underwent pancreatectomy..
Figure 3: Pancreatic ductal adenocarcinoma: A. Low cyclin E expression, B. High cyclin E expression, C. Cyclin E expression in Placenta, D. High p27Kip1 expression and E. low 27Kip1 expression in pancreatic cancer, F. 27Kip1 expression in smooth muscle. A. Operative Treatment B. Stage 1 p < 0.0001 1 p<0.0001 Median Survival: 11 . 2 Vs 4 . 7 months .8 .8 Median Survival: 14 . 0 Vs 7 . 2 months n=118 n = 118
.6 .6
.4 .4 CUMULATIVE SURVIVAL CUMULATIVE SURVIVAL .2 .2 STAGEI/II RESECTED
0 BIOPSY 0 STAGE III / IV
0 20 40 60 80 100 120 0 20 40 60 80 100 120 MONTHS MONTHS C. Degree of Differentiation D. Cyclin E Expression
1 p = 0.008 1 p = 0.005 Median Survival: 9 . 8 Vs 6 . 4 months .8 Median Survival: 9 . 5 Vs 5 . 8 months n= 118 .8 n = 118
.6 .6
.4 .4 CUMULATIVE SURVIVAL
.2 CUMULATIVE SURVIVAL .2 LOW CYCLIN E WELL / MODERATE
0 POOR 0 HIGH CYCLIN E
0 20 40 60 80 100 120 0 20 40 60 80 100 120 MONTHS MONTHS E. p27Kip1 Expression 1 p = 0.46 n=111 .8
.6 LOW p27
.4 HIGH p27
CUMULATIVE SURVIVAL .2
0
0 20 40 60 80 100 120 MONTHS
Figure 1 Resected Cohort
A. Resection Margin B. Tumor Size 1 p = 0.002 1 p = 0.04 Median Survival: 14 . 5 Vs 8 . 5 months Median Survival: 17 . 1 Vs 9 . 9 months .8 n=75 .8 n=75
.6 .6
.4 .4 ≤20 mm CUMULATIVE SURVIVAL CUMULATIVE SURVIVAL .2 .2 CLEAR
0 INVOLVED 0 >20mm
0 20 40 60 80 100 120 0 20 40 60 80 100 120 MONTHS MONTHS C. Lymph Node Invasion D. Cyclin E Expression 1 1 p = 0.02 p = 0.03 Median Survival: 14 . 0 Vs 9 . 5 months Median Survival: 14 . 2 Vs 8 . 5 months .8 .8 n=74 n=75
.6 .6
.4 .4 CUMULATIVE SURVIVAL
LOW CYCLIN E .2 NODE NEGATIVE CUMULATIVE SURVIVAL .2
HIGH CYCLIN E 0 NODE POSITIVE 0
0 20 40 60 80 100 120 0 20 40 60 80 100 120 MONTHS MONTHS E. p27Kip1 Expression
1 p = 0.38 n=72 .8
.6
LOW p27 .4 HIGH p27
CUMULATIVE SURVIVAL .2
0
0 20 40 60 80 100 120 MONTHS
Figure 2 Figure 3 Overexpression of LMO4 induces mammary hyperplasia, promotes cell invasion, and is a predictor of poor outcome in breast cancer
Eleanor Y. M. Sum*, Davendra Segara†, Belinda Duscio*, Mary L. Bath*, Andrew S. Field‡, Robert L. Sutherland†, Geoffrey J. Lindeman*, and Jane E. Visvader*§
*The Walter and Eliza Hall Institute of Medical Research and Bone Marrow Research Laboratories, 1G Royal Parade, Parkville VIC 3050, Australia; †Garvan Institute of Medical Research, 384 Victoria Street, Darlinghurst NSW 2010, Australia; and ‡Department of Anatomical Pathology, St. Vincent’s Hospital, Darlinghurst NSW 2010, Australia
Communicated by Suzanne Cory, The Walter and Eliza Hall Institute of Medical Research, Melbourne, Victoria, Australia, April 12, 2005 (received for review November 23, 2004) The zinc finger protein LMO4 is overexpressed in a high proportion in governing cell proliferation. We have shown that LMO4 forms of breast carcinomas. Here, we report that overexpression of a a complex with the corepressor CtIP and BRCA1 in breast epi- mouse mammary tumor virus (MMTV)-Lmo4 transgene in the thelial cells and that LMO4 can repress BRCA1-mediated tran- mouse mammary gland elicits hyperplasia and mammary intraepi- scriptional activation (15). thelial neoplasia or adenosquamous carcinoma in two transgenic To further explore a role for LMO4 in breast oncogenesis, we strains with a tumor latency of 13–18 months. To investigate generated transgenic mice expressing Lmo4 under the control of cellular processes controlled by LMO4 and those that may be the mouse mammary tumor virus (MMTV) promoter and used deregulated during oncogenesis, we used RNA interference. RNA interference (RNAi) to investigate the cellular processes Down-regulation of LMO4 expression reduced proliferation of affected by LMO4 in breast cancer cells. Mammary hyperplasia, human breast cancer cells and increased differentiation of mouse mammary intraepithelial neoplasia (MIN), and͞or adenosquamous mammary epithelial cells. Furthermore, small-interfering-RNA- carcinoma were observed in two MMTV-Lmo4 transgenic strains, transfected breast cancer cells (MDA-MB-231) had a reduced ca- demonstrating that overexpression of LMO4 contributes to mam- pacity to migrate and invade an extracellular matrix. Conversely, mary tumorigenesis. RNAi revealed that down-regulating LMO4 overexpression of LMO4 in noninvasive, immortalized human expression substantially reduces the proliferation, migration, and MCF10A cells promoted cell motility and invasion. Significantly, in invasion of breast cancer cells. These findings indicate that LMO4 a cohort of 159 primary breast cancers, high nuclear levels of LMO4 may induce tumorigenesis by perturbation of cellular proliferation were an independent predictor of death from breast cancer. and motility. Further, the observation that high levels of nuclear Together, these findings suggest that deregulation of LMO4 in LMO4 correlate with poor patient outcome suggests that LMO4 breast epithelium contributes directly to breast neoplasia by al- may provide an additional marker in breast cancer. tering the rate of cellular proliferation and promoting cell invasion. Materials and Methods LIM domain ͉ oncogene ͉ proliferation Generation and Analysis of Transgenic Mice. The rabbit globin intron (RG-IVS2) was subcloned as a HindIII–EcoRI fragment into MO4 belongs to the LIM-only (LMO) subclass of LIM domain HindIII–EcoRI sites of the MMTV LTR expression plasmid (16) Lproteins that are defined by two tandem zinc finger domains (1). to generate the MMTV-RG-SV40 vector to provide a longer One of the central functions of the LIM domain is to mediate intronic sequence for mRNA stability. A HindIII–EcoRV fragment protein–protein interactions, allowing LMO proteins to act as spanning the mouse Lmo4 coding region was cloned into MMTV- RG-SV40 to generate MMTV-Lmo4. A 6-kb MMTV-Lmo4 frag- MEDICAL SCIENCES adaptors for the assembly of multiprotein complexes. LMO pro- ͞ teins have critical roles in normal development. LMO2 is essential ment was microinjected into C57BL 6 fertilized mouse oocytes, and several transgenic founder mice were identified. Four trans- for embryonic hematopoiesis and angiogenesis (2, 3), whereas mice ͞ lacking both LMO1 and LMO3 die shortly after birth from un- genic lines on a BALB c background (more than six generations) were further examined. Transgenic mice were identified by PCR known causes (4). Targeted deletion of Lmo4 (4, 5) results in Ј perinatal lethality accompanied by failure of neural tube closure analysis using the following SV40-specific primers: 5 -CTCTA- GAGGATCTTTGTGAAGG-3Ј (forward) and 5Ј-GGACAAAC- and homeotic transformations in the rib cage and cervical verte- CACAACTAGAATGC-3Ј (reverse). Total RNA was isolated brae, suggesting that Lmo4 functions as a modulator of Hox protein from mammary tissue of transgenic mice by using TRIzol function. (GIBCO͞BRL) according to the manufacturer’s instructions, and Deregulation of LMO expression can lead to oncogenesis. LMO1 Northern blot analysis was performed by using 15 g of total RNA and LMO2 were originally discovered by their association with (12). For histology, tissue was fixed in 10% (wt͞vol) buffered recurrent translocations in T cell acute lymphocytic leukemia and formalin, dehydrated, and embedded in paraffin, and sections were subsequently were shown to act as T cell oncogenes in transgenic stained with hematoxylin and eosin. models (1, 6–9). Remarkably, the LMO2 gene was ectopically activated by retroviral integration in severe combined immunode- Cell Lines and Cell Transduction. BT-549, MCF-7, and MDA-MB-231 ficient patients who developed T cell leukemia after gene therapy cells were maintained in RPMI medium 1640 (GIBCO͞BRL) (10). LMO4 was initially identified in an expression screen by using containing 10% FCS. HC11 cells were maintained as described serum from a breast cancer patient (11). Moreover, deregulated expression of LMO4 has been demonstrated in a significant pro- portion of breast carcinomas (12) and, more recently, in cancers of Abbreviations: siRNA, small interfering RNA; LMO, LIM-only; RNAi, RNA interference; the oral cavity (13). In breast cancer, LMO4 expression appears to MMTV, mouse mammary tumor virus; MIN, mammary intraepithelial neoplasia; HR, hazard be inversely correlated with estrogen receptor ␣ (ER␣) expression ratio; CI, confidence interval; ER, estrogen receptor; PR, progesterone receptor. (14). LMO4 can act as a negative regulator of mammary epithelial §To whom correspondence should be addressed. E-mail: [email protected]. differentiation in vitro (12), suggesting that LMO4 may have a role © 2005 by The National Academy of Sciences of the USA www.pnas.org͞cgi͞doi͞10.1073͞pnas.0502990102 PNAS ͉ May 24, 2005 ͉ vol. 102 ͉ no. 21 ͉ 7659–7664 (17). MCF10A(EcoR) breast epithelial cells were kindly provided cohort of 194 patients with a diagnosis of invasive ductal breast by J. Brugge and D. Lynch (Harvard Medical School, Boston) (18). carcinoma between May 1984 and March 2001. Median follow up Full-length LMO4 cDNA carrying a 5Ј-FLAG tag was subcloned was 49.5 months (range, 1.3–135). Archival formalin-fixed, paraffin- into the retroviral expression vector pBabe-puro (19), and stable embedded pathology specimens were used to construct eight breast pools of MCF10A(EcoR) transductants were generated (18) by carcinoma tissue microarrays (TMAs), which contained up to 45 ϫ using 2 g͞ml puromycin. MCF10A(EcoR) cells infected with 1.0-mm cores per slide. The pathology of each core on the TMAs empty pBabe vector provided a control. was reviewed by a specialist breast pathologist (A.S.F.) before immunohistochemistry. RNAi. Cells plated at a density of 0.5 to 1 ϫ 105 cells per well in six-well plates were transfected with 200 pmol of LMO4-specific and Immunohistochemistry and Scoring. Sections were subjected to tar- control small interfering RNA (siRNA) oligoduplexes (Dharma- get retrieval solution (DAKO) at 100°C for 20 min with a DAKO con Research, Layfayette, CO) by using oligofectamine reagent autostainer. Endogenous peroxidase activity was quenched with (Invitrogen) according to the manufacturer’s protocols. The se- 3% hydrogen peroxide in methanol, followed by avidin͞biotin and quences of each oligoduplex were as follows: siRNA276, 5Ј- serum-free protein blocks (DAKO). Sections were incubated for 30 GUCGAUUCCUGCGAGUGAAdTdT-3Ј; siRNA376, 5Ј-GAU- min with anti-LMO4 mAb (20F8) at 7 g͞ml, followed by biotin- CGGUUUCACUACAUCAdTdT-3Ј; and siRNA276mut, 5Ј- ylated rabbit anti-rat IgG (DAKO). A streptavidin-biotin peroxi- GUCCAUUUCUCGGCGUUAAdTdT-3Ј. The siRNA276mut dase detection system was used with 3,3Ј-diaminobenzidine as oligoduplex is based on the sequence of siRNA276 and contains six substrate (DAKO). LMO4 immunostaining was assessed indepen- transversions (bold and underlined). dently by two observers (D.S. and A.S.F.). Scores were given as the percentage of carcinoma cell nuclei staining positive, with an Cell Proliferation and Differentiation Assays. Proliferation of RNAi- absolute intensity on a scale of 0–4 (0, none; 1, pale; 2, mild; 3, treated cells was analyzed by using the Cell Titer 96 AQueous One strong homogenous; and 4, intense). The following criteria were solution cell-proliferation assay (Promega) according to the man- used to achieve a positive score for LMO4 overexpression: nuclear ufacturer’s instructions. Cells were replated into a 96-well flat- intensity, Ͼ2inϾ50% of nuclei. bottom plate in triplicate wells at a density of 2,500 cells per well. The assay was performed on day 0 to control for cell density and Statistical Analysis. Kaplan–Meier and the Cox proportional- on subsequent days 1–5. For the differentiation assay, RNAi- hazards model were used for univariate and multivariate analysis treated HC11 cells were grown to confluency for 1–2 days and then with STATVIEW 5.0 software (Abacus Systems, Berkeley, CA). Death starved of EGF overnight before the addition of the lactogenic from breast cancer was the end point. The factors that were stimulus [10Ϫ6 M dexamethasone͞5 g/ml insulin͞5 g/ml ovine prognostic on univariate analysis were assessed in a multivariable prolactin (17), kindly provided by C. Parlow (National Institute of model to identify factors that were independently prognostic. This Diabetes and Digestive and Kidney Diseases)]. After 48 h, cells analysis was performed sequentially on all patients who had avail- were harvested for RNA extraction by using TRIzol; cDNA syn- able tissue (n ϭ 159). thesis and PCR were performed by using primers for -casein and Hprt as described (12). Results LMO4 Induces Mammary Hyperplasia and MIN in MMTV-Lmo4 Trans- Transwell Migration and Invasion Assays. In vitro migration and genic Mice. To assess the role of Lmo4 as a potential oncogene in invasion assays were performed by using 24-well, 8-m pore the mammary gland, we generated transgenic mice expressing this transwell inserts (Becton Dickinson). Cells were first resuspended gene under the control of the MMTV long-terminal repeat. The in Matrigel (Becton Dickinson) for invasion assays. We seeded 106 transgene included rabbit -globin and SV40 intronic sequences to (migration) or 5 ϫ 106 (invasion) cells in 200 l of serum-free augment mRNA stability, as well as a polyadenylation sequence growth medium in the upper chamber, and 600 l of medium with (Fig. 1A). Analysis of four independent transgenic strains revealed or without chemoattractant (2% FCS or 20 ng͞ml EGF) was added the expected Ϸ2.6-kb transgene transcript in mammary tissue. to the lower chamber. Cells were incubated at 37°C for 5 (migra- Moderate transgene expression was detected during midpregnancy, tion) or 16 (invasion) h, then fixed in 10% (wt͞vol) buffered- with the highest levels noted during late pregnancy, when the formalin and stained with 0.5 g͞ml DAPI (Sigma). Cells on the MMTV promoter is most active. Western blot analysis confirmed upper surface were removed with a cotton swab, and migrated cells increased Lmo4 expression during pregnancy and lactation (data on the underside were counted (average of 10 fields at a magnifi- not shown). The two strains (strains 34 and 36) that expressed cation of ϫ40 per transwell). highest levels of the transgene were selected for further study (Fig. 1B). The level of transgene-derived Lmo4 mRNA was estimated to Western Blotting. Protein extracts were generated by lysing cells in be Ն3-fold higher than the endogenous level by Northern blot Triton X-100 lysis buffer (12). Proteins were separated by SDS͞ analysis (data not shown). Neither strain expressed the transgene at PAGE; transferred to polyvinylidene difluoride membranes (Mil- appreciable levels in nonpregnant nulliparous mammary glands lipore); and probed with rat ␣-LMO4 20F8 (20), ␣-FLAG (Sigma), (Fig. 1B), compatible with the hormone-responsive nature of the or ␣-tubulin Ab (Sigma). MMTV-LTR (21). Because luminal cells in the mammary glands of both virgin and pregnant mice already express abundant Lmo4 (12, Immunofluorescence. Cells were plated on coverslips in six-well 20), it was anticipated that high levels of transgene-derived Lmo4, plates, fixed in 4% paraformaldehyde, permeabilized in 0.1% expressed with altered kinetics, would be necessary for the induc- Triton X100, and blocked in 5% normal goat serum. Costaining for tion of tumors. LMO4 and focal adhesion complexes was performed by incubating No overt phenotypic abnormalities were detected in the with anti-LMO4 (5 g͞ml) (20) and anti-vinculin (1 g͞ml; Sigma) mammary glands of nulliparous mice or mice during their first mAbs. Staining was detected with secondary Alexa Fluor 488 and pregnancy by whole-mount and histological analyses. Further- Alexa Fluor 594 Abs (Molecular Probes). Cells were mounted in more, transgenic mice were capable of lactation and their glands fluorescent mounting medium (DAKO) and visualized by confocal were histologically indistinguishable from wild-type mice (data microscopy (TCS.NT.SP2, Leica, Deerfield, IL). not shown). Given that highest Lmo4 expression in transgenic animals was achieved during pregnancy, we assessed the capacity Patient Cohort. After receiving ethical approval from the St. Vin- of the Lmo4 transgene to promote mammary tumorigenesis in cent’s Hospital Campus Human Ethics Committee, we identified a two strains (strains 34 and 36), both on a BALB͞c background,
7660 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0502990102 Sum et al. Fig. 1. Overexpression of Lmo4 in transgenic mice leads to mammary hyperplasia and tumors. (A) Schematic representation of the MMTV-Lmo4 transgene. Lmo4 cDNA was cloned into the MMTV-RG-SV40 vector. (B) Northern blot analysis of total RNA (15 g) from the mammary glands of transgenic strains 34 and 36. Filters were hybridized with a transgene-specific SV40 probe followed by a glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNA probe. dP, day pregnant. (C–E) Hematoxylin and eosin sections of mammary glands and lung from strain 34, showing high grade MIN (C) and acinar hyperplasia (D), and alveolar bronchocarcinoma (E). (F–H) Hematoxylin and eosin sections from strain 36 showing mammary glands from multiparous mice. (F) Littermate control. (G) Diffuse acinar hyperplasia. (H) Adenosquamous carcinoma. [Original magnification: ϫ400 (C–E and H) and ϫ200 (F and G).] after three pregnancies. For strain 34, 3 of 13 transgenic mice Mammary glands from two other mice displayed diffuse acinar monitored for 14–18 months exhibited frank tumor nodules, hyperplasia (Fig. 1D) and one of these females developed a which proved to be MIN (Fig. 1C), as defined by the Annapolis bronchoalveolar carcinoma of the lung at 14 months (Fig. 1E). guidelines (22). The MIN lesions were multilayered and exhib- For strain 36, four of five multiparous female mice developed ited nuclear pleomorphism and increased mitotic figures, with multifocal acinar hyperplasia (Fig. 1G), with a mean latency of two lesions also showing evidence of squamous metaplasia. 16 months (range, 14–20) and one of these mice also exhibited MEDICAL SCIENCES
Fig. 2. Down-regulation of LMO4 expression in breast epithelial cells by RNAi inhibits proliferation and augments differentiation. (A) Western blot analysis of protein lysates from MCF-7 and BT-549 breast cancer cells transiently transfected with LMO4-specific siRNA276 and siRNA376, compared with control siRNA and mock-transfected cells, 2 and 6 days after transfection by using ␣-LMO4 (20F8) mAb. Anti-tubulin was used to verify protein loading. (B) The proliferation rate of siRNA-transfected MCF-7 and BT-549 cells, and mock-transfected cells, was determined from days 0–5, in three independent experiments. Error bars indicate SEM; n ϭ 3. (C) Cell cycle defect in LMO4-deficient MCF7 cells. siRNA and mock-transfected cells were fixed in ice-cold 70% ethanol for 24 h before staining with 0.5 g͞ml propidium iodide (PI). FACScan analysis revealed an increase in G1͞G0 cells and a decrease in S-phase cells in LMO4-deficient MCF-7 cells, compared with control cells, in three independent experiments. Error bars indicate SEM; n ϭ 3. *, P Ͻ 0.05. (D) Down-regulation of LMO4 expression in HC11 cells transiently transfected with siRNA376 (or a control siRNA) was confirmed by Western blot analysis of cells harvested 3 days after transfection by using ␣-LMO4 (20F8) mAb. Anti-tubulin immunoblotting provided a control. (E) RT-PCR analysis was performed by using total RNA derived from RNAi-treated HC11 cells that were stimulated with (ϩ) prolactin, insulin, and dexamethasone or unstimulated (Ϫ) for 48 h. -casein and Hprt were used as markers of differentiation and loading, respectively. At least three independent experiments were performed.
Sum et al. PNAS ͉ May 24, 2005 ͉ vol. 102 ͉ no. 21 ͉ 7661 Fig. 4. LMO4 overexpression augments mammary epithelial cell migration and invasion. (A) The motility of MCF10A(EcoR) cells, stably transduced with a FLAG-LMO4 retroviral construct or vector alone (pBabe-puro), was deter- mined by counting the number of migrated cells in 10 random fields from each of four experiments. Error bars indicate SEM; n ϭ 4. (B) The number of MCF10A(EcoR) cells expressing either FLAG-LMO4 or vector alone (pBabe- puro), capable of invading through Matrigel, was determined in four exper- iments, as described for A. Error bars indicate SEM; n ϭ 4. (C) Western blot analysis confirmed the expression of FLAG-LMO4 in the transductants. Lysates from cells expressing FLAG-LMO4 or empty vector were subjected to SDS͞ PAGE and immunoblotted with mouse ␣-FLAG mAb. Anti-tubulin was used to control for protein loading.
LMO4 expression. Two breast cancer cell lines, MCF-7 and BT-549, were transiently transfected with siRNA oligoduplexes. LMO4- specific siRNAs (siRNA276 and siRNA376) effectively suppressed LMO4 expression in both cell lines, whereas a control mutant siRNA had no effect, nor did mock transfection (Fig. 2A). LMO4 Fig. 3. Reduced LMO4 expression impedes the migration and invasion of expression was down-regulated rapidly, with low levels evident at breast cancer cells. (A) Western blot analysis of protein lysates from MDA-MB- day 1 after transfection, consistent with the short half-life (Ͻ 2h) 231 breast cancer cells transfected with siRNA276, siRNA376, control siRNA, or of this protein (data not shown). A reduction in LMO4 protein mock-transfected cells, 2 and 6 days after transfection by using ␣-LMO4 (20F8) levels was observed up to day 6 after transfection in both cell lines mAb. Anti-tubulin provided a control. (B) Focal adhesions in MDA-MB-231 with siRNA376, although this was not the case for siRNA276 in cells transfected with siRNA376 or a control siRNA were visualized by indirect BT-549 cells (Fig. 2A). The LMO4-specific siRNAs substantially immunofluorescence by using ␣-vinculin mAb. Costaining with ␣-LMO4 (20F8) reduced the proliferation of both MCF-7 and BT-549 breast cancer mAb confirmed the reduction in LMO4 expression in siRNA376-transfected cells, whereas the control siRNAs had no effect (Fig. 2B). No cell nuclei. [Scale bars, 20 (i and ii)and8(iii and iv) m.] (C) The number of increase in apoptosis was observed upon transfection with LMO4- migrating MDA-MB-231 cells transfected with either LMO4-specific siRNA276 specific siRNAs (data not shown). FACScan analysis of transfected or siRNA376, or a mutant siRNA, was determined by counting 10 random fields MCF-7 cells revealed that down-regulation of LMO4 expression led in each of three independent experiments. Error bars indicate SEM; n ϭ 3. (D) ͞ The number of MDA-MB-231 cells, transfected with siRNA276, siRNA376, or a to cell cycle changes. An increase in the proportion of G1 G0 cells and a corresponding decrease in S-phase cells was evident in three mutant siRNA, capable of invading Matrigel was determined for three inde- Ͻ pendent experiments, as in C. Error bars indicate SEM; n ϭ 3. independent experiments with a P value of 0.05, whereas the control siRNA had no effect (Fig. 2C). Thus, LMO4 appears to promote S-phase entry of breast epithelial cells, providing evidence an adenosquamous carcinoma (Fig. 1H) at 18 months. Hyper- for a role in cell proliferation. plasia was also observed in a third transgenic strain. In contrast, To examine the effect of LMO4-specific RNAi on mammary mammary glands from 13 nontransgenic littermate mice that differentiation, we used mouse mammary epithelial cells as there is underwent three pregnancies did not develop hyperplasia or no suitable human counterpart line to study differentiation. HC11 tumors during a similar follow-up period of 14–19 months (Fig. cells have the ability to differentiate into milk-producing cells upon 1F). In addition to these findings, lymphoproliferative disease or treatment with a lactogenic stimulus (17). Down-regulation of lymphomas were observed in four transgenic mice and one LMO4 expression mediated by siRNA376 (Fig. 2D) was found to wild-type mouse, presumably reflecting MMTV promoter activ- augment -casein mRNA levels compared with cells transfected ity in lymphoid cells (21) and aging of the animals. Although with the control siRNA (Fig. 2E). These data are consistent with Lmo4 is not a potent oncoprotein, these data demonstrate that the notion that LMO4 maintains the proliferative rather than it can contribute to mammary tumorigenesis. differentiative state of mammary epithelial cells.
Down-Regulation of LMO4 Expression by siRNA Inhibits Proliferation Reduced LMO4 Expression Impedes the Migration and Invasion of of Breast Epithelial Cells. To determine whether LMO4 regulates the Breast Cancer Cells. To examine the role of LMO4 in cell motility proliferation of breast cancer cells, we used RNAi to down-regulate and invasion, we used RNAi to down-regulate LMO4 expression in
7662 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0502990102 Sum et al. Table 1. Univariate analysis of clinicopathological parameters Parameter Cohort, % HR (95% CI) P value
Tumor size (Ͼ20 mm) 42 3.83 (1.74–8.43) 0.0008 Axillary lymph node involvement 54 5.40 (2.06–14.18) 0.0006 High grade 52 7.45 (2.59–21.42) 0.0002 ER-positive 71 0.19 (0.09–0.41) Ͻ0.0001 PR-positive 66 0.29 (0.14–0.60) 0.0009 Her2-positive 15 2.40 (1.01–5.66) 0.047 LMO4-positive 43 2.38 (1.12–5.04) 0.023
Although the cohort comprised 194 patients, it was only possible to score 159 samples for LMO4 because of tissue loss from the tissue microarrays. Therefore, all analyses were on these 159 samples. ER and PR status were not available for two samples.
or overexpression of LMO4 markedly affects the motility of mam- mary epithelial cells and their ability to invade an extracellular matrix in vitro.
LMO4 Overexpression in Breast Cancer Correlates with Poor Clinical Outcome. We have reported in a small series of 60 cases that LMO4 protein is overexpressed in Ͼ50% of primary breast cancers (12). To further investigate the relationship between LMO4 expression Fig. 5. Overexpression of LMO4 in breast cancer is a predictor of poor clinical and clinicopathological features of the disease, we performed outcome. (A) LMO4 immunostaining was performed on tissue microarrays immunohistochemical analysis on tissue from a cohort of 194 containing archival breast tumor specimens using ␣-LMO4 (20F8) mAb. Rep- primary breast cancer patients of known clinical outcome. resentative images of tumor specimens displaying low (Left) and high (Right) LMO4 overexpression, as defined in Materials and Methods, was levels of LMO4 expression are shown. (B) Kaplan–Meier curves for overall apparent in 68 of 159 (42.7%) ductal carcinomas from which tissue survival (OS) of 159 breast cancer patients according to LMO4 nuclear staining. was available (Fig. 5A). In a univariate Cox proportional-hazards High LMO4 expression was significantly associated with decreased OS in regression analysis, high nuclear LMO4 expression was significantly univariate (P ϭ 0.023) analysis. associated with decreased patient survival (P ϭ 0.023), as were other well established markers of outcome [i.e., tumor size; tumor MDA-MB-231 cells, an invasive breast cancer cell line. Transfection grade; axillary lymph node status; and ER, progesterone receptor with either of the LMO4-specific siRNAs profoundly reduced (PR), and HER2 status] (Table 1). In a multivariate analysis LMO4 expression in MDA-MB-231 cells, whereas the control incorporating these parameters, only axillary lymph node status siRNA did not (Fig. 3A). Transfection of MDA-MB-231 cells with [hazard ratio (HR), 4.02; 95% confidence interval (CI), 1.4–11.5; siRNA276 and siRNA376 resulted in a 2.5- and 3.8-fold decrease P ϭ 0.009] and LMO4 nuclear overexpression (HR, 2.27; 95% CI, in cell motility, using transwell migration assays in the presence of 1.01–5.11; P ϭ 0.048) were independent predictors of death from 2% FCS, compared with that for control cells (Fig. 3C). Next, we breast cancer (Table 2). Kaplan–Meier analysis confirmed a sig- examined invasion of these cells through an extracellular matrix nificant relationship between LMO4 overexpression and patient (Matrigel) in a modified Boyden chamber, which is an assay that has survival (Fig. 5B). Interestingly, most (15͞23) tumors overexpress- been shown to correlate with in vivo metastatic potential (23). ing HER2, an adverse prognostic marker, also overexpressed siRNA276 resulted in a 1.8-fold decrease in the invasion of MDA- LMO4 (Fisher’s exact test; P ϭ 0.0247). However, HER2-positive MEDICAL SCIENCES MB-231 cells, whereas siRNA376 elicited a 2.6-fold decrease (Fig. tumors only represented 23% of LMO4-positive tumors (64 3D). These findings indicate that LMO4 regulates the motility and samples). invasiveness of breast cancer cells. To determine whether LMO4 influences focal adhesions, we assessed vinculin distribution in Discussion siRNA-transfected MDA-MB-231 cells by indirect immunofluo- Here, we provide evidence that LMO4 is an important regulator of rescence. Focal adhesions appeared to be less prominent in MDA- breast epithelial cell proliferation and invasion and that it can act MB-231 cells transfected with siRNA376, compared with cells as a mammary oncoprotein in a transgenic mouse model. Enforced transfected with a control siRNA (Fig. 3B). Costaining with anti- expression of Lmo4 in the mammary glands of transgenic mice LMO4 mAb (20F8) confirmed a substantial decrease in LMO4 recapitulates the high levels of Lmo4 that occur in Ϸ50% of primary levels in the nuclei of siRNA376-transfected cells (Fig. 3B). breast cancers. The development of mammary hyperplasia and MIN in MMTV-Lmo4 transgenic mice supports the notion that LMO4 Overexpression Stimulates Mammary Epithelial Cell Migration and Invasion. To further investigate the role of LMO4 in cell motility and invasion, we transduced human breast epithelial MCF-10A Table 2. Multivariate analysis of clinicopathological parameters cells harboring the murine ecotropic receptor, MCF-10A(EcoR), Parameter HR (95% CI) P value with a LMO4-expressing retrovirus. MCF10A(EcoR) cells express- Tumor size (Ͼ20 mm) 1.51 (0.59–3.83) 0.389 ing FLAG-LMO4 showed a 2.9-fold increase in migration when Axillary lymph node-positive 4.02 (1.40–11.5) 0.009 EGF was used as the chemoattractant, compared with cells infected High grade 3.03 (0.96–9.59) 0.059 with a control virus (Fig. 4A). Moreover, a 2.1-fold increase in the ER-positive 0.28 (0.07–1.12) 0.072 number of invasive cells was observed for MCF10A(EcoR) cells PR-positive 1.16 (0.31–4.38) 0.822 transduced with pBabe-FLAG-LMO4, relative to control cells (Fig. HER2-positive 1.54 (0.61–3.89) 0.367 4B). Expression of FLAG-LMO4 was confirmed by Western blot LMO4-positive 2.27 (1.01–5.11) 0.048 analysis using ␣-FLAG Ab (Fig. 4C). Thus, either down-regulation
Sum et al. PNAS ͉ May 24, 2005 ͉ vol. 102 ͉ no. 21 ͉ 7663 Lmo4 has a causative role in the pathogenesis of breast cancer. the undifferentiated state and that the absolute levels of LMO4 may Diffuse acinar hyperplasia was observed in 15% and 80% of mice govern the rate of proliferation. Interestingly, the dosage of the in transgenic strains 34 and 36, respectively. MIN or adenosqua- Drosophila LMO gene Beadex has been established to be critical for mous carcinoma arose in Ϸ20% of mice (4 of 18 mice) with long wing formation, indicating that the stoichiometry of LMO- latency, indicating that Lmo4 is not a potent oncogene. Neverthe- containing complexes is an important parameter (28). Similarly, the less, D-type cyclins (D1 and D2) and cyclin E also induce carcino- stoichiometry and composition of LMO4-multimeric complexes in mas in transgenic mice with long latency and͞or a low incidence normal breast epithelial cells is likely to be perturbed by even small (24–26). For example, tumors arise in 10% and 19% of cyclin E (24) changes in protein concentration, resulting in profound effects on and cyclin D2 (25) transgenic mice, respectively. Interestingly, when cell homeostasis. Lmo4 was fused to the repressor engrailed in a transgenic model, no LMO4 has a role in regulating the motility and invasiveness of hyperplasia or tumors were apparent (26). The RNAi studies breast epithelial cells. We have demonstrated that increased LMO4 reported here suggest that overexpression of Lmo4 contributes to levels promote migration and invasion, whereas a reduction in breast oncogenesis by deregulating proliferation and increasing cell LMO4 expression is inhibitory to these processes. Although there motility. was no apparent change in the actin cytoskeleton of siRNA- LMO4 appears to be necessary but not sufficient for mammary transfected MDA-MB-231 cells (data not shown), down-regulation tumorigenesis, similar to that observed for several transgenic of LMO4 levels led to a consistent decrease in the number of focal oncogene models. The latency of mammary lesions in Lmo4 adhesions. Increased cell invasion has been linked to an increase in transgenic mice was 13–18 months, indicating that additional mu- focal adhesions. However, there are examples in which fewer focal tations were required. It is particularly relevant that other Lmo adhesions have been observed in more invasive cells, such as cells family members elicit tumorigenesis with long latency. Overexpres- overexpressing LIM-kinase (29). These findings underscore the sion of the Lmo1 or Lmo2 gene in transgenic mice induces clonal complexity of the process of cell migration. T-cell leukemia after several months. All tumors in Lmo1 trans- The influence of LMO4 on cell proliferation, migration, and genic animals have an immature phenotype, and Lmo2 transgenic invasion has implications for the role of this protein in breast cancer mice exhibit a marked accumulation of immature double negative progression. Indeed, high nuclear expression of LMO4 in primary T cells, indicating that a block in differentiation has occurred in invasive breast carcinomas was significantly associated with patient these mice (1, 8). survival. Although there was a significant association between LMO4 functions as a positive regulator of cellular proliferation LMO4 and HER2 overexpression, Ϸ75% of patients with high and a negative regulator of differentiation. Down-regulation of LMO4 levels were not HER2-positive. These data raise the possi- LMO4 expression in breast cancer cell lines by RNAi led to bility that LMO4 may provide an additional marker in breast cancer markedly reduced proliferation, accompanied by a partial G0͞G1 and that it represents a potential target for therapeutics. Defining block. Therefore, we examined expression of cell cycle regulators, the molecular pathways that are regulated by LMO4 is an essential including cyclins D1 and E, retinoblastoma protein (pRb), p21, and step toward dissecting the mechanisms through which LMO4 p27 in these cells. Although changes were evident in the case of influences the development and progression of breast carcinomas. cyclin D1 and p27, they were not always consistent. Overall, the data indicate that LMO4 is unlikely to regulate the cell-cycle machinery. We thank M. Buchanan for invaluable help with pathology; O. Bernard, Rather, deregulation of LMO4 levels may lead to an uncoupling of E. Musgrove, R. Anderson, and C. Restall for helpful discussions; P. proliferation and differentiation. Consistent with this proposal, Crea for access to breast cancer specimens; and M. Lam, S. Mihajlovic, E. Tsui, and F. Feleppa for expert assistance with confocal microscopy RNAi-mediated knock-down of Lmo4 expression in HC11 cells and histology. This work was supported by the Victorian Breast Cancer augmented differentiation. Conversely, overexpression of LMO4 in Research Consortium, and the National Health and Medical Research mammary epithelial or neuroblastoma cells blocked their matura- Council. R.L.S. is supported by the National Health and Medical tion (12, 27). These data, in combination with RNAi-mediated Research Council, The Cancer Council New South Wales, and the R. T. inhibition of proliferation, imply that LMO4 functions to maintain Hall Trust.
1. Rabbitts, T. H. (1998) Genes Dev. 12, 2651–2657. 15. Sum, E. Y., Peng, B., Yu, X., Chen, J., Byrne, J., Lindeman, G. J. & Visvader, J. E. 2. Yamada, Y., Warren, A. J., Dobson, C., Forster, A., Pannell, R. & Rabbitts, T. H. (2002) J. Biol. Chem. 277, 7849–7856. (1998) Proc. Natl. Acad. Sci. USA 95, 3890–3895. 16. Stewart, T. A., Pattengale, P. K. & Leder, P. (1984) Cell 38, 627–637. 3. Yamada, Y., Pannell, R., Forster, A. & Rabbitts, T. H. (2000) Proc. Natl. Acad. Sci. 17. Taverna, D., Groner, B. & Hynes, N. E. (1991) Cell Growth Differ. 2, 145–154. USA 97, 320–324. 18. Debnath, J., Muthuswamy, S. K. & Brugge, J. S. (2003) Methods 30, 256–268. 4. Tse, E., Smith, A. J., Hunt, S., Lavenir, I., Forster, A., Warren, A. J., Grutz, G., Foroni, 19. Morgenstern, J. P. & Land, H. (1990) Nucleic Acids Res. 18, 3587–3596. 24, L., Carlton, M. B., Colledge, W. H., et al. (2004) Mol. Cell. Biol. 2063–2073. 20. Sum, E. Y., O’Reilly, L. A., Jonas, N., Lindeman, G. J. & Visvader, J. E. (2005) 5. Hahm, K., Sum, E. Y., Fujiwara, Y., Lindeman, G. J., Visvader, J. E. & Orkin, S. H. J. Histochem. Cytochem. 53, 475–486. 24, (2004) Mol. Cell. Biol. 2074–2082. 21. Leder, A., Pattengale, P. K., Kuo, A., Stewart, T. A. & Leder, P. (1986) Cell 45, 6. Fisch, P., Boehm, T., Lavenir, I., Larson, T., Arno, J., Forster, A. & Rabbitts, T. H. 485–495. 7, (1992) Oncogene 2389–2397. 22. Cardiff, R. D., Anver, M. R., Gusterson, B. A., Hennighausen, L., Jensen, R. A., 7. Larson, R. C., Fisch, P., Larson, T. A., Lavenir, I., Langford, T., King, G. & Rabbitts, Merino, M. J., Rehm, S., Russo, J., Tavassoli, F. A., Wakefield, L. M., et al. (2000) T. H. (1994) Oncogene 9, 3675–3681. Oncogene 19, 968–988. 8. McGuire, E. A., Rintoul, C. E., Sclar, G. M. & Korsmeyer, S. J. (1992) Mol. Cell. Biol. 23. Albini, A., Iwamoto, Y., Kleinman, H. K., Martin, G. R., Aaronson, S. A., Kozlowski, 12, 4186–4196. J. M. & McEwan, R. N. (1987) Cancer Res. 47, 3239–3245. 9. Neale, G. A., Rehg, J. E. & Goorha, R. M. (1995) Blood 86, 3060–3071. 24. Bortner, D. M. & Rosenberg, M. P. (1997) Mol. Cell. Biol. 17, 453–459. 10. Hacein-Bey-Abina, S., Von Kalle, C., Schmidt, M., McCormack, M. P., Wulffraat, N., 25. Kong, G., Chua, S. S., Yijun, Y., Kittrell, F., Moraes, R. C., Medina, D. & Said, T. K. Leboulch, P., Lim, A., Osborne, C. S., Pawliuk, R., Morillon, E., et al. (2003) Science 21, 302, 415–419. (2002) Oncogene 7214–7225. 11. Racevskis, J., Dill, A., Sparano, J. A. & Ruan, H. (1999) Biochim. Biophys. Acta 1445, 26. Wang, N., Kudryavtseva, E., Chen, I., McCormick, J., Sugihara, T. M., Ruiz, R. & 23, 148–153. Andersen, B. (2004) Oncogene 1507–1513. 12. Visvader, J. E., Venter, D., Hahm, K., Santamaria, M., Sum, E. Y. M., O’Reilly, L., 27. Wang, T. C., Cardiff, R. D., Zukerberg, L., Lees, E., Arnold, A. & Schmidt, E. V. White, D., Williams, R., Armes, J. & Lindeman, G. J. (2001) Proc. Natl. Acad. Sci. (1994) Nature 369, 669–671. USA 98, 14452–14457. 28. Vu, D., Marin, P., Walzer, C., Cathieni, M. M., Bianchi, E. N., Saidji, F., Leuba, G., 13. Mizunuma, H., Miyazawa, J., Sanada, K. & Imai, K. (2003) Br. J. Cancer 88, Bouras, C. & Savioz, A. (2003) Brain Res. Mol. Brain Res. 115, 93–103. 1543–1548. 29. Matthews, J. M. & Visvader, J. E. (2003) EMBO Rep. 4, 1132–1137. 14. Gruvberger, S., Ringner, M., Chen, Y., Panavally, S., Saal, L. H., Borg, A., Ferno, M., 30. Yoshioka, K., Foletta, V., Bernard, O. & Itoh, K. (2003) Proc. Natl. Acad. Sci. USA Peterson, C. & Meltzer, P. S. (2001) Cancer Res. 61, 5979–5984. 100, 7247–7252.
7664 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0502990102 Sum et al. Oncogene (2003) 22, 5070–5081 & 2003 Nature Publishing Group All rights reserved 0950-9232/03 $25.00 www.nature.com/onc
EDD, the human orthologue of the hyperplastic discs tumour suppressor gene, is amplified and overexpressed in cancer
Jennifer L Clancy1, Michelle J Henderson1, Amanda J Russell1, David W Anderson2, Ronaldo J Bova1, Ian G Campbell3, David YH Choong3, Graeme A Macdonald4, Graham J Mann5, Tania Nolan6, Ged Brady6, Olufunmilayo I Olopade7, Erica Woollatt8, Michael J Davies1, Davendra Segara1, Neville F Hacker9, Susan M Henshall1, Robert L Sutherland1 and Colin KW Watts*1
1Cancer Research Program, Garvan Institute of Medical Research, 384 Victoria St., Darlinghurst, NSW 2010, Australia; 2Laboratory of Molecular Genetics, NIDCD/NIH, Rockville, MD, USA; 3Peter MacCallum Cancer Institute, Melbourne, Victoria, Australia; 4Population and Clinical Sciences Division, Queensland Institute of Medical Research and Department of Medicine, The University of Queensland, Brisbane, Australia; 5Westmead Institute of Cancer Research, University of Sydney at Westmead Millennium Institute, Westmead, Australia; 6School of Biological Sciences, University of Manchester, Manchester, UK; 7Center for Medical Genetics, University of Chicago Medical Center, Chicago, IL, USA; 8Center for Medical Genetics, Women’s and Children’s Hospital, Adelaide, Australia; 9Gynaecological Cancer Centre, Royal Hospital for Women, Randwick, NSW 2031, Australia
EDD (E3 isolated by differential display), located at Keywords: allelic imbalance; amplification; human neo- chromosome 8q22.3, is the human orthologue of the plasms; gene expression; mutation Drosophila melanogaster tumour suppressor gene ‘hyper- plastic discs’ and encodes a HECT domain E3 ubiquitin protein-ligase. To investigate the possible involvement of EDD in human cancer, several cancers from diverse tissue sites were analysed for allelic gain or loss (allelic Introduction imbalance, AI) at the EDD locus using an EDD-specific microsatellite, CEDD, and other polymorphic microsa- Previously, we identified EDD (E3 isolated by Differ- tellites mapped in the vicinity of the 8q22.3 locus. Of 143 ential Display) (Callaghan et al., 1998), the human cancers studied, 38 had AI at CEDD (42% of 90 orthologue of the Drosophila melanogaster ‘hyperplastic informative cases). In 14 of these cases, discrete regions discs’ gene (hyd) (Mansfield et al., 1994). Both genes of imbalance encompassing 8q22.3 were present, while the encode large proteins of approximately 300 kDa that remainder had more extensive 8q aberrations. AI of contain conserved carboxy terminus HECT domains, CEDD was most frequent in ovarian cancer (22/47 indicating they function as E3 ubiquitin protein ligases informative cases, 47%), particularly in the serous (Huibregtse et al., 1995; Schwarz et al., 1998). Mutagen- subtype (16/22, 73%), but was rare in benign and esis studies (Mansfield et al., 1994) define an apparent borderline ovarian tumours. AI was also common in tumour suppressor role for hyd in Drosophila by breast cancer (31%), hepatocellular carcinoma (46%), demonstrating its critical requirement in control of cell squamous cell carcinoma of the tongue (50%) and proliferation during development. Some mutations metastatic melanoma (18%). AI is likely to represent result in imaginal disc hyperplasia and adult sterility amplification of the EDD gene locus rather than loss of due to germ cell defects, while the null hyd phenotype is heterozygosity, as quantitative RT–PCR and immunohis- lethality in the pupal or larval stages. Defects in tochemistry showed that EDD mRNA and protein are hedgehog signalling may contribute to these phenotypes frequently overexpressed in breast and ovarian cancers, (Lee et al., 2002). ONCOGENOMICS while among breast cancer cell lines EDD overexpression These properties of hyd suggest the potential involve- and increased gene copy number were correlated. These ment of EDD in regulating mammalian cell prolifera- results demonstrate that AI at the EDD locus is common tion and hence in human tumorigenesis or tumour in a diversity of carcinomas and that the EDD gene is progression. This possibility is reinforced by an implied frequently overexpressed in breast and ovarian cancer, role for EDD in DNA damage signalling. A recent study implying a potential role in cancer progression. found that topoisomerase IIb binding protein (TopBP1) Oncogene (2003) 22, 5070–5081. doi:10.1038/sj.onc.1206775 is ubiquitinylated and targeted for proteosomal degra- dation by EDD (Honda et al., 2002). TopBP1 coloca- lizes with BRCA1 at stalled replication forks, evidence *Correspondence: CKW Watts, Cancer Research Program, Garvan that TopBP1 functions in the DNA damage response Institute of Medical Research, 384 Victoria Street, Darlinghurst, NSW 2010, Australia; E-mail: [email protected] (Makiniemi et al., 2001). EDD also interacts in a DNA Received 3 September 2002; revised 17 April 2003; accepted 29 April damage-dependent manner with the calcium- and 2003 integrin-binding protein (CIB) (Henderson et al., Abnormalities of the EDD gene in human cancer JL Clancy et al 5071 2002). This protein in turn interacts with DNA- addition, as can be seen from the following data, 30–40% dependent protein kinase, which also functions in of those cancers that have chromosomal aberrations at DNA damage pathways (Wu and Lieber, 1997). the CEDD microsatellite show no involvement of one or EDD has been localized by fluorescence in situ more microsatellites at the telomeric end of the region hybridization (FISH) to chromosome band 8q22.3 studied. This indicates that AI at the EDD locus often (Callaghan et al., 1998) and recent genomic sequencing occurs independently of more extensive aberrations in shows that the gene has a centromeric orientation with these cancers (i.e. loss or gain of the whole chromosome its first exon placed at 8q23 (Genbank Accession arm or coamplification with the MYC proto-oncogene). AP002852). Many studies have reported chromosomal aberrations on 8q but used sparse and imprecisely Ovarian cancer mapped markers. Only recently have more detailed studies of 8q been conducted. In genome-wide com- The ovarian tumour set comprised several cancer parative genomic hybridization (CGH) and microsatel- subtypes (predominantly mucinous, endometrioid and lite scans, gene amplification at 8q is commonly serous), borderline tumours and nonmalignant benign observed in a wide variety of cancers (van Dekken tumours. Within the carcinomas, 48% (34/71) displayed et al., 1997; Knuutila et al., 1998) and has been AI at one or more markers in the region (Figure 2). In attributed to increases in 8q copy number or to marked contrast, benign and borderline tumours ex- amplification of the MYC oncogene (8q24.12) and hibited only 1/23 and 1/5 cases of AI, respectively, surrounding regions (Wong et al., 1999; Yokota et al., involving several microsatellites but not CEDD or 1999). However, there is evidence in many tumour types DS8257 (Table 1). Owing to the low frequency of for other foci of amplification and several genes have involvement of 8q22.3–8q23.3 in benign and borderline been identified as potential 8q oncogenes (Balleine et al., tumours, they were omitted from the following analysis. 2000; Nupponen et al., 2000). Several cancers had AI across all informative mar- As a first step to test whether EDD might have a role kers, consistent with large chromosomal aberrations, for in human cancer, microsatellite allelotyping was used to example, cases 63, 114 and 154. However, of interest are determine whether chromosomal aberrations within those cancers where AI is present at the EDD locus but 8q22.3–8q24.13, a region spanning the EDD locus, are not at the more telomeric markers (e.g. cases 14, 22, 32 common in a range of human cancers. These data and 211; Figure 2). This suggests that an important provided evidence for frequent allelic imbalance (AI) at region of chromosomal aberration is located close to the the EDD locus which in breast cancer cell lines was EDD locus. Indeed a central finding of this study is the associated with mRNA overexpression and amplifica- observation in serous cancers that CEDD was the tion of the EDD gene. The subsequent demonstration of microsatellite most commonly affected by AI (16/22, protein overexpression in a significant proportion of 73% of informative cases). This very high frequency of breast and ovarian cancers provided supporting evi- involvement differs significantly from that for the more o dence for the common amplification and overexpression telomeric markers D8S545 (6/18, 33%, P 0.01) or o of the EDD gene in cancer. D8S85 (5/14, 36%, P 0.01). In fact, when serous cancers are compared to other ovarian cancers, CEDD is the only microsatellite for which AI differs signifi- cantly between the two groups (P ¼ 0.0018). However, Results no correlation was apparent between AI at any locus and tumour grade and stage, either within the entire set Microsatellite analysis of cancers or the serous subtype (data not shown). The frequency of AI surrounding the EDD locus (8q22. 3) was investigated in malignant ovarian tumours, Hepatocellular carcinoma, squamous cell carcinoma of breast cancers, hepatocellular carcinomas, metastatic the tongue and metastatic melanoma melanomas and squamous cell carcinomas of the anterior tongue using nine microsatellites shown in In all three cancer types, AI was common (Table 1) with Figure 1. As CEDD (CA repeat within EDD) and 74, 33 and 47% of cancers displaying AI of at least one 586F18b are located within introns of the EDD gene, it microsatellite respectively. can be assumed that chromosomal loss or gain involving Of 19 hepatocellular carcinomas studied, four had these alleles is equivalent to loss or gain of an EDD regions of imbalance that included CEDD but did not allele. extend continuously telomeric to 8q22.3 (Figure 1a). Within the complete cancer set, the frequency of AI in Similarly, analysis of 17 metastatic melanomas (Table 1) the narrow region of 8q under study was considerable: identified three cancers where the region of AI did not 60% (83/139 informative cases) had AI at one or more involve markers telomeric to 8q22.3 (example shown in markers. The individual frequencies (cases with AI/ Figure 1). Of the 12 squamous cell tongue carcinomas informative cases) were D8S326 47/103 (46%), CEDD analysed four had AI involving 8q22.3 (CEDD and/or 38/90 (42%), D8S257 29/96 (30%), D8S300 27/69 D8S326), and in three of these cancers the region of (39%), D8S545 28/87 (32%) and D8S85 24/90 (27%) imbalance did not extend to the most telomeric markers (Table 1). Notably, CEDD and the neighbouring 8q22.3 D8S85 and D8S545 (example shown in Figure 1). Thus, marker D8S326 had the greatest frequency of AI. In in these three cancer types, the chromosomal region at
Oncogene Abnormalities of the EDD gene in human cancer JL Clancy et al 5072
Figure 1 (a) Allelic imbalance in several cancer types showing AI at the EDD locus (8q22.3) but not extending continuously to 8q24. Key: allelic imbalance, probable allelic imbalance, J heterozygote, & uninformative homozygote, gap denotes no data available. Positions of polymorphic microsatellites were identified from the Genome Data Base (GDB) (http://gdbwww.gdb.org/). Microsatellites CEDD and 586F18b are encoded in introns of EDD. The EDD gene is located at 8q22.3 (Callaghan et al., 1998) and MYC at 8q24.12 (GDB). (b) Microsatellite analysis of allelic imbalance from normal hepatocellular tissue and hepatocellular carcinoma from patient 8 (a). Allelic imbalance was seen for microsatellites D8S326, CEDD, D8S545 and D8S85. Arrows indicate significant increase (430%) in proportion of allele indicated
Figure 2 Allelic imbalance in ovarian cancers on chromosome 8q22.3–23.3. Cancers are grouped according to histopathology. Overlap indicates mixed histology. ‘Other’ comprises four adenomas and a germ cell tumour (case 124). Key: allelic imbalance, J heterozygote, & uninformative homozygote, gap denotes no data available
Oncogene Abnormalities of the EDD gene in human cancer JL Clancy et al 5073 Table 1 Allelic status of polymorphic microsatellites at 8q22.3–24.1 in five tumour types1 D8S326 CEDD D8S257 D8S300 D8S545 D8S85 MYC.PCR3 D8S198
Ovarian–malignant 22/55 (40) 22/46 (48) 17/41 (41) 12/31 (37) 13/45 (29) 13/39 (33) Serous 12/25 (48) 16/22 (73) 11/19 (58) 7/14 (50) 6/18 (33) 5/14 (36) Endometrioid 7/17 (41) 4/11 (36) 4/11 (40) 2/9 (22) 5/16 (31) 5/14 (36) Mucinous 1/7 (14) 1/8 (13) 0/5 (0) 1/3 (33) 1/6 (16) 1/5 (20) Other2 2/6 (33) 1/5 (33) 2/6 (33) 2/5 (40) 1/5 (16) 2/6 (33) Ovarian–benign 0/17 (0) 0/10 (0) 0/12 (0) 0/9 (0) 1/16 (6) 1/9 (11) Ovarian–borderline 1/4 (25) 0/3 (0) 0/3 (0) 0/2 (0) 0/3 (0) 1/4 (25) Breast 7/11 (64) 6/16 (38)3 4/20 (20) 3/8 (38) 5/9 (56) 3/14 (21) 5/12 (42) 4/13 (31) Hepatocellular 7/14 (50) 6/13 (46) 2/15 (13) 8/15 (53) 6/16 (38) 4/13 (31) Melanoma 7/16 (44) 2/11 (18) 6/13 (46) 4/15 (27) 4/13 (31) 4/15 (27) Tongue 4/7 (57) 2/4 (50) 0/7 (0) N/A 0/4 (0) 0/9 (0) Total4 47/103 (46) 38/90 (42) 29/96 (30) 27/69 (39) 28/87 (32) 24/90 (27)
1Results are expressed as cases of allelic imbalance/informative cases with percentages in bold in parentheses; 2includes adenocarcinoma, germ cell tumours and tumours of mixed histology; 3includes CEDD and 586F18b AI; 4excludes benign and borderline ovarian tumours N/A: not available
the EDD locus is commonly and often specifically locus (p53 ribonucleotide reductase (p53R2) and carbo- aberrant. nic anhydrase II (CA II)) in 18 breast cancer cell lines. The corresponding DNA copy numbers for the three genes were determined by quantitative PCR. For EDD, Breast cancer these values were consistent with those determined by For microsatellite analysis of breast cancer DNA, three FISH analysis which was performed in four cell lines. additional microsatellites were introduced to provide There was a strong trend for EDD to be overexpressed more information about AI at the EDD locus (586F18b) at the mRNA level when EDD was amplified at the and at the telomeric region of 8q around MYC DNA level (Figure 3b). Almost all cell lines where EDD (8q24.12). (D8S198 and MYC-PCR.3). As with the was overexpressed had amplified EDD at the genomic other cancer types studied, AI was common on 8q with level (7/8), whereas EDD was rarely overexpressed when 16/24 (67%) breast cancers displaying AI at one or more the gene copy number was two or less. This was also true markers (Table 1). CEDD or 586F18b were involved in for the expression of p53R2, the gene adjacent to EDD 6/16 (38%) of informative cases. Commonly, AI at chromosome position 8q23.1. All cell lines where involved MYC-PCR.3 or D8S198, consistent with the p53R2 was overexpressed had increased p53R2 gene frequent amplification of MYC in breast cancer, but in copy number (Figure 3b). The relative expression of the majority of cancers AI was not continuous from EDD and p53R2 was significantly correlated (R2 ¼ 0.62, 8q22.3 to 8q24 (examples shown in Figure 1). These Po0.0001) suggesting a similar mechanism of over- data support previous suggestions that there are foci for expression for both genes. In contrast, the expression of AI on chromosome 8q in breast cancer distinct from and CA II, located at 8q22, showed no relationship with centromeric to MYC (Nesbit et al., 1999; Nupponen DNA copy number (data not shown), with 15/18 cell et al., 1999, 2000). lines having less than 10% of the expression levels of CA II mRNA of 184 normal breast epithelial cells. EDD mRNA expression in breast cancers EDD protein expression in breast and ovarian cancers EDD mRNA expression levels were determined by quantitative RT–PCR in 41 breast cancers, and in EDD protein levels were determined by immunohisto- matched normal breast tissue controls for 14 of these chemistry in nine specimens of normal breast tissue, 46 cases (Figure 3a). Although the majority of cancers breast carcinomas and 94 serous ovarian carcinomas. expressed EDD mRNA within the normal range, a Positive controls of wild-type EDD embryonic mouse significant number 11/41 (27%) had higher expression. neural tissue (Figure 4b) and WT-30 cells (not shown) Elevated expression was even more apparent when demonstrated intense nuclear staining, while no expres- cancers were compared to their matched normal tissue sion was seen in EDD null embryonic mouse neural controls, such that 6/14 (43%) cancers had a 4four-fold tissue (Figure 4a). Of nine normal breast samples, four increase in EDD expression, including one cancer with a had no detectable expression and five had low-level 159-fold increase (Figure 3a, inset). expression of EDD (Figure 4c). Of 46 breast carcino- mas, all expressed EDD and demonstrated either low- EDD, p53R2 and CA II mRNA expression and genomic intensity (37%) or high-intensity (63%) nuclear staining copy number in breast cancer cell lines (Figure 4d). Among the 94 serous ovarian carcinomas, 2% failed to express EDD while 59% had low Using quantitative RT–PCR, mRNA expression levels expression (Figure 4e) and 39% had high EDD were determined for EDD and two genes near the same expression (Figure 4f).
Oncogene Abnormalities of the EDD gene in human cancer JL Clancy et al 5074
Figure 3 mRNA expression of EDD in breast cancers, normal breast tissue and breast epithelial cell lines. (a) Relative mRNA expression of EDD in 41 breast cancers and 14 normal breast samples determined by quantitative RT–PCR. Horizontal line depicts upper limit of normal tissue expression. Inset: Comparison of EDD mRNA expression in 14 breast cancer samples and matched normal breast tissue determined by quantitative RT–PCR. (b) Expression of EDD mRNA in breast epithelial cell lines determined by quantitative RT–PCR. Gene copy number (indicated above bars) was determined by quantitative PCR. EDD genomic copy number was determined in MDA-MB-436, MDA-MB-468, BT-20 and BT-483 by FISH (indicated in brackets). 184 is a normal breast epithelial cell line while all others are breast cancer cell lines (Sutherland et al., 1999)
Mutation analysis of EDD in cell lines and tumours variations in EDD mRNA that resulted in a change to the translated amino-acid sequence. These putative To assess the frequency of mutations in the EDD gene in missense mutations were confirmed in the genomic cancer, the complete coding region of the EDD mRNA sequence. The amino-acid changes were not in regions of (8397 nucleotides) was sequenced using cDNA derived the EDD protein with any obvious functional motifs from 26 breast, ovarian and prostate cancer cell lines. and only the His4Asn change in the SK-Br-3 line alters Three normal breast epithelial cell lines were also amino-acid polarity. In the absence of matched normal sequenced. A list of the sequence variants identified is DNA from the individuals from which the cell lines were shown in Table 2. Only 2/25 cancer cell lines had derived, it is not possible to determine whether these
Oncogene Abnormalities of the EDD gene in human cancer JL Clancy et al 5075
Figure 4 EDD protein expression in breast and ovarian cancers. Tissues were stained with a polyclonal EDD antibody and counterstained with haematoxylin. (a) Negative control, neural tissue from an EDD-null mouse embryo (neural epithelium (NE), neural mesenchyme (NM)). (b) Positive control, neural tissue from wild-type mouse embryo. (c) Normal human breast duct. (d) Human breast carcinoma with high EDD expression. (e) Human serous ovarian cancer with low EDD expression. (f) Human serous ovarian cancer with high EDD expression
Table 2 Summary of EDD sequence variants detected in 26 normal and cancer cell lines Nucleotide postion1 Codon Base change Predicted aa2 change Cell line(s)
4753 1584 C-A His-Asn SK-BR-3 6279 2093 A-G Asn-Ser IGROV-1 886–902 296–300 Splice variant VLLLPL removed 26/26 cell lines 7689 A-C No change Hs 578T 4055 C-T No change BT-20, Human mRNA3 4390 A-G No change Human mRNA3 4556 A-G No change T-47D, MDA-MB-134 3956 A-G No change 9/26 cell lines 7634 C-A No change 6/26 cell lines
Base changes in bold confirmed at the genomic level 1Nucleotide numbering starts at the A of the initiation codon; 2aa, amino acid; 3David Anderson (unpublished data) represent true somatic mutations or rare polymorph- found either in RNA derived from normal cell lines or isms. Few polymorphisms or silent substitutions were tissues, or in multiple cell lines. A splice variant was also observed. Of the six conservative sequence variants, at observed in all cell lines. The variant differs from the least five are likely to be polymorphisms as they are full-length EDD mRNA by deletion of 18 bp of
Oncogene Abnormalities of the EDD gene in human cancer JL Clancy et al 5076 sequence from nt 884–901, removing the amino-acid target oncogene(s). Multiplication or duplication of sequence VLLLPL. Although this motif is not located in the entire 8q arm is also a common feature of many any of the putative functional domains of EDD cancers, especially breast cancer (Kallioniemi et al., (Henderson et al., 2002), removal of this sequence 1994; Mark et al., 1997; Sisley et al., 1997), sometimes does have the potential to disrupt protein structure and accompanied by 8p loss (isochromosome 8q) (van might alter enzymatic function or localization of the Dekken et al., 1997; Yokota et al., 1999). While the protein. majority of 8q amplicons encompass 8q24 where a likely SSCP analysis of 29 ovarian cancers displaying AI target gene for amplification is the myc oncogene (Figure 2), 37 breast cancers and 29 colon cancers (8q24.12) (Yokota et al., 1999; Forozan et al., 2000; confirmed the low frequency of mutation of the EDD Savelyeva and Schwab, 2001), CGH and microsatellite gene, although only eight exons of EDD (covering studies suggest that there are multiple amplicons on 8q approximately 13% of the coding sequence) were at 8q21, 8q22–23 and 8q24 (Kallioniemi et al., 1994; studied. These exons encode a nuclear localization Muleris et al., 1994; El-Naggar et al., 1997; Fejzo et al., signal (nt 1482–1598), a zinc-finger motif (nt 3573– 1998; Nupponen et al., 1999; Forozan et al., 2000; Seute 3812) and the majority of the HECT domain (nt 7602– et al., 2001). 8339) (Henderson et al., 2002). No mutations were Evidence that the region of 8q containing EDD is a found in the coding regions or splice junction sites of specific focus of chromosomal abnormality is particu- any cancer (data not shown). larly clear for serous carcinoma of the ovary in which the frequency of AI at the EDD gene-specific micro- satellite CEDD was almost twice that of the most telomeric microsatellite examined. Ovarian carcinomas Discussion commonly show a high level of chromosomal rearrange- ment in comparison with other tumour types. This is EDD is the human orthologue of the D. melanogaster subtype specific, with serous tumours displaying the ‘hyperplastic discs’ gene, which mutagenesis studies have greatest changes, endometrioid intermediate and shown has a critical role in control of cell proliferation mucinous cancers the least (Pieretti et al., 1995), as during Drosophila development (Mansfield et al., 1994), was the case for AI at 8q in our study. In one study, the defining hyd as a tumour suppressor gene. This property invasive potential of the serous subtype has been suggests that the EDD gene might play a role in human correlated with changes to chromosome 8 (Diebold cancer. Our interest in this possibility was strengthened et al., 1996), although no correlation between clinical by recent evidence that EDD interacts with one or more data (including grade and stage) and AI at 8q was found proteins in the DNA damage response (Henderson et al., in our study. 2002; Honda et al., 2002). As an initial approach to Amplification is also common on 8q in hepatocellular evaluate EDD’s involvement in cancer, we used micro- carcinoma. Frequencies of AI (interpreted as LOH) vary satellite allelotyping to examine a narrow region of from 26 to 77% (Nagai et al., 1997; Piao et al., 1998), chromosome 8q that encompasses the locus for EDD while CGH showed 8q gain in 100% of hepatocellular (8q22.3) and identified this as a significant and specific carcinoma arising from noncirrhotic liver (Wong et al., area of allelic imbalance in ovarian cancer, hepatocel- 1999). In addition to frequent amplification at MYC,a lular carcinoma, squamous cell carcinoma of the tongue, discrete 8q22 amplicon was found in 3/30 cases of breast cancer and metastatic melanoma. This was hepatocellular carcinoma (Fujiwara et al., 1993), but particularly evident in serous carcinoma of the ovary, little is known about the precise extent of these where AI of the EDD locus occurred in 73% of cancers. aberrations. Similarly, AI at 8q is common but poorly Overexpression of either EDD mRNA or protein was defined in head and neck cancer (Nawroz et al., 1994; observed in a significant proportion of breast and Hermsen et al., 1997) and melanoma (Sisley et al., 1997; ovarian cancers and breast cancer cell lines, suggesting Parrella et al., 1999). Although we only studied small the possibility that EDD may be a target for amplifica- numbers of these cancer types, we also confirmed the tion in this region of 8q. presence of common AI at 8q, and found that this often Involvement of the 8q chromosomal arm is common occurred in discrete regions at or adjacent to the EDD in a wide variety of cancer types. Although several locus. microsatellite analyses have interpreted the occurrence Given that EDD is commonly amplified in several of AI of various 8q markers as evidence for loss of cancer types and this often occurs within discrete heterozygosity (LOH) (Iwabuchi et al., 1995; El-Naggar amplicons that have telomeric boundaries that are et al., 1997; Miyazaki et al., 1997; Dahiya et al., 1998), located within 8q23, aberrations at the EDD gene locus amplification is a more likely explanation as CGH cannot merely be ascribed to the presence of an has only rarely detected 8q losses (Knuutila et al., extensive MYC containing amplicon or to 8q duplica- 1999). This is consistent with our quantitative tion. Even though this study showed that AI at EDD PCR analysis, which demonstrates common increases and MYC occurred at approximately the same fre- in EDD and p53R2 gene copy number in cancer cell quency (38 and 42%, respectively), AI was only rarely lines. Such chromosomal gains often indicate regions continuous between these two loci, consistent with other of amplification, arising during tumour progression, studies where amplification was often discrete and and which are presumed to include specific discontinuous along 8q (Tsuneizumi et al., 2001).
Oncogene Abnormalities of the EDD gene in human cancer JL Clancy et al 5077 A likely consequence of EDD amplification is the 1994), consistent with a tumour suppressor gene frequent overexpression of mRNA that we have function, other mutations result in retarded growth of observed in breast carcinomas. Certainly, in breast the imaginal discs, suggesting that HYD might cancer cell lines there is a strong trend for over- alternatively act as a positive regulator of cell expression of EDD mRNA to occur in cells where proliferation. Lee et al. (2002) report that hyd null cells EDD is amplified in the genome. Conversely, diploid in the eye induce overgrowth of surrounding wild-type cell lines rarely overexpressed EDD. Given that tissue but fail to proliferate themselves, an effect that amplicons contain several genes with increased mRNA appears to be mediated through activation of hedgehog expression (Orntoft et al., 2002), it was therefore not and decapentaplegic signalling. Finally, in contrast to surprising to find another overexpressed gene (p53R2) in our data, a recent study found that EDD missense the proximity of EDD on 8q23. The involvement of mutations were relatively common in microsatellite- p53R2 in the DNA damage response (Yamaguchi et al., unstable gastric and colorectal cancers, that is, cancers 2001) also makes it an interesting candidate oncogene having deficient DNA mismatch repair (Mori et al., on 8q. In contrast, expression of carbonic anhydrase II 2002). In conclusion, there is mounting evidence (8q22) showed no correlation with gene copy number, from several studies that aberrations in EDD gene indicating that only certain genes within this amplicon expression and presumably function are a common are regulated by this mechanism. Moreover, CA II feature in many cancer types. Further studies are mRNA was significantly reduced in most cancer cell currently underway to gain a deeper understanding of lines compared to cell lines derived from the normal the biological role of EDD and its potential involvement breast epithelium (data not shown). This is consistent in tumorigenesis. with reports of reduced CA II expression in colorectal tumours (Nogradi, 1998). Consistent with the above data on gene amplification and mRNA overexpression, EDD protein is highly Materials and methods expressed in many breast carcinomas. Indeed, as is also the case for ovarian serous carcinomas, this appears to Tumours and DNA extraction occur more frequently than can be accounted for by Ovarian tumour tissue and matched blood or normal gene amplification, suggesting that additional mechan- ovarian tissue adjacent to cancer were obtained from 98 isms result in EDD protein upregulation. Similar patients and DNA extracted as described previously observations have been made for cyclin D1 in breast (Obata et al., 1998). Metastatic melanoma tissue and cancers, where protein overexpression is significantly matched blood or normal skin tissue were obtained more common than gene amplification (Buckley et al., from 20 patients and DNA isolated as reported previously 1993). Clearly, given that EDD is found only at low to (Indsto et al., 1998). DNA was extracted from matched very low levels in normal tissues, an important issue is to normal liver and hepatocellular carcinoma tissue samples determine the functional consequences of EDD over- microdissected from 19 primary liver tumours (Macdonald expression in cancer. et al., 1998). For AI analysis, paraffin-embedded breast cancer tissues and normal blood were obtained from 24 Despite the frequency of gross aberrations at the patients. Tumour location was determined by haemotoxylin EDD locus and the common overexpression of the gene, and eosin staining and cells were microdissected from 4–5 sequencing of the large EDD mRNA (8397 nt) in a adjacent sections. DNA was extracted in lysis buffer (0.45% search for mutations revealed only 2/25 cancer cell lines Tween 20, 5 mg/ml proteinase K, 0.25% BSA) at 551C for 8 h, with single nucleotide changes that result in amino-acid then boiled for 10 min. DNA was extracted from blood using a changes. These alterations might represent rare poly- Puregene DNA isolation kit (Gentra Systems, Minneapolis, morphisms rather than mutations, as none of the MN, USA). changes were clearly disruptive and occurred in regions For RT–PCR, breast cancer samples were collected at the of the protein without any predicted functional sig- time of surgery. Normal breast tissue (based on histological nificance. A splicing variant was found but this was examination) was removed from unaffected regions of the excised cancer. present along with the unspliced transcript in every Paraffin-embedded archival samples from 12 squamous cell normal and cancer cell type examined. Similarly, SSCP carcinoma of the anterior tongue and matched lymph nodes analysis of a limited number of exons covering 13% of were from a series described by Bova et al. (1999). Areas of the coding region, and which code for putative func- malignant squamous cells were identified by a pathologist from tional domains, found no evidence of mutations in a haemotoxylin and eosin stained slides and were microdissected series of breast, ovarian and colon cancers. This from unstained adjacent 10 mm sections under a light micro- sequence conservation may point to a critical require- scope. Normal tissue was gained from uninvolved lymph nodes ment for this gene in cell function, a conclusion and/or microdissected from areas of normal cells surrounding the squamous cell carcinoma. Samples were digested in 250 ml reinforced by the hyd knockout lethality (Mansfield 1 et al., 1994) and by our recent finding that targeted of proteinase K (2 mg/ml) for 5 days at 37 C with constant agitation. The digest was then extracted once with phenol, deletion of EDD in mice results in embryonic lethality once with chloroform, and DNA precipitated in ethanol accompanied by a generalized block in cell proliferation overnight and redissolved in 30 ml of TE buffer (‘microdis- (Saunders et al., submitted). However, complex roles for sected DNA’). these genes are likely. Although some hyd mutations The study proceeded following approval from the St result in imaginal disc hyperplasia (Mansfield et al., Vincents Hospital Human Research Ethics Committee.
Oncogene Abnormalities of the EDD gene in human cancer JL Clancy et al 5078 Isolation and cloning of EDD-linked microsatellites CEDD and hold. Amplification of the CEDD microsatellite from micro- 586F18b dissected DNA required a 30 s annealing step, a 1 min extension step, and 35 cycles. Amplification of the micro- A human P1 genomic library (Incyte Genomics, CA, USA) satellite D8S300 was performed as follows: 12 min hotstart was screened with two cDNA probes covering 4 kb of the 951C; 35 cycles of (1) 941C 30 s, (2) 601C 30 s and (3) 721C60s; EDD coding sequence (Callaghan et al., 1998). DNA from 5 min 721C; 41C hold. No PCR products could be obtained positive clones was extracted, digested with HincII, separated from DNA recovered from paraffin-embedded tongue carci- by gel electrophoresis and rescreened by Southern blotting noma with D8S300 primers, presumably because of DNA using an oligonucleotide probe (CA) . Positive genomic DNA 15 shearing. fragments were cloned into pBluescript and sequenced and the Duplicate fluorescent products were separated on an ABI resulting data identified a novel dinucleotide repeat micro- 377 DNA Sequencer (Applied Biosystems) and analysed using satellite CEDD. Unique primers were designed around the Genescan and Genotyper software (Applied Biosystems). microsatellite to give a product of approximately 220 bp; 0 Ambiguous results were resolved on separate gels. The CEDD forward 5 -TACCCTGCAGTAAAATCTCACATG- 0 0 abundance of each heterozygous allele of a microsatellite was TACTCCC-3 , CEDD reverse 5 -AGAATCGCTTGAACC- 0 measured as a fluorescence peak area, and relative abundance TAGTAGGTGAAGGTG-3 . Subsequently, EDD genomic was then calculated. AI was defined by a reproducible sequence became available (Genbank Accession: AC021004) reduction in relative abundance of an allele between normal and CEDD was located in an intron between bases 2616 and and cancer DNA. A 30% reduction threshold was used as 2617 of the EDD coding sequence. The minimum size of the previously described for the cohort of melanoma samples EDD gene is 100 kb, consisting of at least 46 exons. Within the (Indsto et al., 1998), although for some samples, AI was same genomic clone, another EDD-specific dinucleotide repeat considered to be probable when a 20–30% reduction was microsatellite was identified and named 586F18b. Unique reproducibly found. A 30% threshold was also used for breast primers were designed around this microsatellite to give a 0 and hepatocellular cancers. A difference of 50% or more was product of approximately 200 bp; forward 5 -GCT AGG GAA 0 0 considered to represent AI in squamous cell carcinomas of the CCA AAC TGC CAG -3 , reverse 5 -TGC AAA ATA ACA 0 tongue due to a lower level of normal cellular DNA ATA GCTTTG CTT AG-3 . 586F18b is located in an intron contamination in these samples. between bases 6631 and 6632 of the EDD coding sequence. CEDD and 586F18b were 43 and 55% heterozygous, respectively, in a collection of DNA from 20 human cell lines FISH (data not shown). Two P1 plasmids encoding approximately 100 kb of EDD genomic sequence used for FISH were nick-translated with biotin-14-dATP and hybridized at a final concentration of Microsatellite analysis of allelic imbalance 20 ng/ml to metaphases from the breast cancer cell lines MDA- Allelic imbalance is defined as the relative gain or loss of one MB-436, SKBR-3, BT-20 and BT-483. The single copy FISH allele when genomic DNA from tumour and matching normal method was modified from that previously described (Callen tissue is compared. AI is indicative of either gene amplification et al., 1990) in that chromosomes were stained before analysis or LOH of an allele and (presumably) the surrounding with both propidium iodide (as counterstain) and DAPI (for chromosomal region. The frequency and distribution of AI chromosome identification). Images of metaphase prepara- on 8q22.3–24.13 was determined by microsatellite analysis, tions were captured by a CCD camera using the ChromoScan which allows comparison of alleles (microsatellites) that image collection and enhancement system (Applied Imaging, consist of variable numbers of short repetitive sequences, that Newcastle, UK). is, CEDD, 586F18b and seven other dinucleotide and tetranucleotide repeat polymorphic markers mapped in this Quantitative RT–PCR – breast tissue region (Genome Database, www.gdb.org/) (Figure 1). CEDD, RNA was extracted from samples using Trizol reagent (Life D8S326, D8S257, D8S300, D8S545 and D8S85 were used for Technologies, Paisley Scotland, UK). Reverse transcription analysis of ovarian cancers, hepatocellular carcinoma, meta- was performed, followed by quantitative PCR using a static melanoma and squamous cell carcinoma of the tongue. Taqman-based methodology as previously described (Al- The additional markers 586F18b, MYC-PCR.3 and D8S198 Taher et al., 2000). PCR primers for EDD were designed were used for analysis of breast cancers. Primer pairs were within 300 bases of the polyA addition site. The sequences obtained from Research Genetics (Huntsville, AL, USA) and of EDD-specific primers were: forward EDD-407F GCTA- Pacific Oligos (Lismore, Australia). Ovarian DNA was GTCACCAACTTCTGGGTCTAA; reverse EDD-490R analysed by radioisotope-based methods as previously de- CAGCAAAAAGATAAATCACAGTGTAAATT; fluores- scribed (Obata et al., 1998). For DNA from other tissues, in cent probe EDD-433 T FAM-CCCAGCCAAAGATGA- each PCR reaction the forward primer was labelled with either CAGCAGAACAAC-TAMRA. Results are expressed as 6-FAM or TET fluorescent label. Reactions were performed copies of EDD cDNA/3 109 per sample, corrected for using 40–60 ng of DNA from hepatocellular carcinoma and m relative expression of the housekeeping gene translation metastatic melanoma, 1 l of breast cancer extracted DNA or initiation factor 2B (IF2B). 2–3 ml of microdissected DNA from squamous cell carcinoma of the tongue. The PCR reaction components were as follows: 10 mm Tris-HCl, pH 8.3; 50 mm KCl; 66 mm dNTP mix; 1.5 mm Quantitative PCR and RT–PCR – breast cancer cell lines 7 2 MgCl2; 5 pmol of both forward and reverse primer and 0.8 U Cells (approximately 5 10 ) harvested from duplicate 150 cm of Amplitaq Gold DNA polymerase (Applied Biosystems, flasks were pooled and RNA was extracted using the RNeasy Sydney, Australia). PCR conditions for microsatellites RNA extraction kit (Qiagen, Melbourne, Australia). cDNA D8S326, D8S257, D8S545, D8S85, D8S198 and MYC- was made as described below (see EDD mRNA sequence PCR.3 were: 12 min hotstart 951C; 35 cycles (40 for micro- analysis) from the normal breast epithelial cell line 184 and the dissected DNA) of (1) 941C 15 s, (2) 601C 15 s (661C for breast cancer cell lines MDA-MB-468, MDA-MB-436, MDA- CEDD, 521C for 586F18b) and (3) 721C 30 s; 5 min 721C; 41C MB-231, MDA-MB-453, MDA-MB-175, MDA-MB-361,
Oncogene Abnormalities of the EDD gene in human cancer JL Clancy et al 5079 MDA-MB-157, BT-20, T-47D, BT-549, BT-483, MCF-7, BT- Single-stranded conformation polymorphism analysis 474, SK-Br-3, ZR-75-1, Hs 578 T and MDA-MB-330. All PCR Single-stranded conformation polymorphism analysis (SSCP) reactions were run on a LightCycler using LightCycler analysis was performed as previously described (Campbell software 3.5 (Roche Diagnostics, Sydney, Australia). Primers et al., 1994) on eight exons of EDD-encoding regions of for: EDD, forward TTAGGCTTTTGGTAAATGGCTGCG, potential functional significance, that is, the HECT domain, reverse TGAGGGCATAGGCTGGAATCCTTC; carbonic nuclear localization signal and zinc-finger motif. Primers were anhydrase II, forward CCACCCCTCCTCTTCTGGAATG, designed within intronic sequence to generate PCR products of reverse GCTTTGATTTGCCTGTTCTTCAGTG; p53 ribo- 200–300 bp. nucleotase reductase (p53R2), forward GTGACTTTGCTTGC- CTGATGTTC, reverse TCTGTGGTTTCTGCCATAACTGC; GAPDH, forward GACATCAAGAAGGTGGTGAA, EDD immunohistochemistry reverse TGTCATACCAGGAAATGAGC. All primer pairs Immunohistochemistry (IHC) was performed on paraffin- spanned at least one intron, and products were visualized on embedded, formalin-fixed breast (nine normal breast and 46 an agarose gel to confirm product size. Relative expression of breast cancers) and ovarian tissues (94 ovarian cancers). all genes was corrected for cDNA concentration using Paraffin-embedded embryonic neural tissues from wild-type GAPDH expression. and EDDD/D (knockout) mice (Saunders et al., submitted) were DNA was extracted from the above cell lines using a used as positive and negative controls, respectively. Paraffin- DNA extraction kit (Stratagene, Sydney, Australia). PCR embedded cell pellets from the WT-30 cell line, which reactions to determine genomic copy number utilized the overexpresses EDD (Henderson et al., 2002), were used as an reverse primer from above (unless mentioned). Primers for: additional positive control. EDD, forward CATTGCTGACCCTATCCCTGTGTTG, For EDD IHC, sections were dewaxed and rehydrated reverse TAGCCCGTGAAATCCTCCCATCTC; CA II, before unmasking in target retrieval solution (high pH: DAKO forward ACCCGCCTCATGCCTCAGCCTTAC; p53R2, Corporation, Carpenteria, CA, USA) in a waterbath, at 1001C forward TGTCAGCCTTGAGTACCTCCAGGG; beclin, for- for 30 min. ward TAGGTTTGGGGTGAGAGTGG, reverse AGTCTG- Using a DAKO autostainer, endogenous peroxidase activity TGGGCAGCAAGG. All products were visualized on an was quenched in 3% hydrogen peroxide in methanol, agarose gel to confirm product size. Relative DNA copy endogenous avidin and biotin were blocked with an avidin number was corrected using known beclin gene copy number biotin block (DAKO Corporation) and nonspecific binding of determined by FISH (Aita et al., 1999). secondary antibody was blocked by incubation with a serum- free protein block (DAKO Corporation). Sections were EDD mRNA sequence analysis incubated for 30 min with 1 : 150 anti-EDD (M19) antibody (Santa Cruz Biotechnology, Santa Cruz, CA, USA) and Total RNA was extracted from the following human cell lines subsequently for 15 min with 1 : 200 biotinylated horse anti- using the RNeasy Maxi Kit (Qiagen): breast cancer MDA- goat (Vector Laboratories, Burlingame, CA, USA). A MB-468, MDA-MB-436, MDA-MB-231, MDA-MB-453, streptavidin–biotin peroxidase detection system was used MDA-MB-175, MDA-MB-361, MDA-MB-134, MDA-MB- according to the manufacturer’s instructions (Vectastain Elite 157, BT-20, T-47D, BT-549, BT-483, MCF-7, BT-474, SK-Br- kit; Vector Laboratories) with 3,3’-diaminobenzidine as 3, ZR-75-1, Hs 578T; normal breast epithelium HBL-100 (SV- substrate. Counterstaining was performed with Mayer’s 40 transformed), 184, HMEC 219-4; ovarian cancer OVCAR- haematoxylin (DAKO Corporation). 3, OVCA-420, IGROV-1, SKOV3, A2780; prostate cancer PC- The degree of staining was assessed by two separate 3, DU-145, LnCaP. cDNA was synthesized using the Expand observers, and discrepancies were resolved by conferencing. Reverse Transcriptase System (Boehringer Mannheim, Syd- EDD expression scores were determined by combining the ney, Australia). In all, 2 mg of total RNA and 2 ml oligo dT percentage of cells expressing EDD (range of 1–4 for 1–100% (Boehringer Mannheim) was made up to 22 ml with water, cells stained) and intensity of staining (range of 0–3). A heated at 651C for 10 min and cooled on ice followed by combined score of 0 was called no expression, a score in the addition of 8 ml Expand Reverse Transcriptase buffer (5 ; range of 1–5 was low expression and a score of 6 and 7 was m m m 250 m Tris-HCl, 200 m KCl, 25 m MgCl2, 2.5 % Tween 20 called high expression. (v/v), pH 8.3), 1 mm each dATP, dCTP, dGTP and dTTP, 10 mm DTT and 2 ml (100 U) Expand Reverse Transcriptase. The reverse transcriptase reaction was then performed at 421C for 45–60 min. A total of 10 PCR products were designed to Abbreviations cover the entire 8.5 kb coding region of the EDD gene. PCR AI, allelic imbalance; CA II, carbonic anhydrase II; CEDD, reactions included 3 ml of the reverse transcriptase reaction, CA repeat within EDD; CGH, comparative genomic hybridi- 10 pmol of each primer, 1.75 U Expand High Fidelity DNA zation; EDD, E3 isolated by differential display; FISH, fluorescence in situ hybridization; IHC, immunohistochemis- polymerase (Boehringer) and 1.5 mm MgCl2. Amplification was carried out using the following protocol: 2 min denature at try; LOH, loss of heterozygosity; MI, microsatellite instability; 941C; 10 cycles of 30 s denature, 30 s annealing and 60 s p53R2, p53 ribonucleotide reductase; SSCP, single-stranded extension at 721C; 24 cycles of 30 s denature, 30 s annealing conformational polymorphism. and 60 s extension with a 5 s increase per cycle; 5 min extension. Annealing temperatures depended on the primers Acknowledgements used (primer sequences available on request). PCR products We thank Gillian Lehrbach, Saskia IJ Ellenbroek, Darren were purified using the QIAquick PCR purification system Saunders, Samantha Hird, and other members of the Cancer (Qiagen) and quantitated by visualization on a gel. Sequencing Research Program for technical assistance and helpful discus- reactions were performed at the Australian Genome Research sions and Nigel Bundred at the University of Manchester for Facility (Brisbane, Australia) and sequence analysis and the supply of clinical materials. assembly were performed using Editview and Autoassembler This work was supported by the National Health and software, respectively (Applied Biosystems). Medical Research Council of Australia (NHMRC), the US
Oncogene Abnormalities of the EDD gene in human cancer JL Clancy et al 5080 Army Medical Research and Materiel Command Breast Cancer the Association for International Cancer Research, the Gynae- Research Program (Grants DAMD17-98-1-8335 and cological Oncology (GO) Fund of the Royal Hospital for DAMD17-00-1-253), The Cancer Council New South Wales, Women Foundation Sydney, Australia and the RT Hall Trust.
References
Aita VM, Liang XH, Murty VV, Pincus DL, Yu W, Cayanis J, Tarkkanen M, Varis A, Wasenius VM, Wolf M and Zhu E, Kalachikov S, Gilliam TC and Levine B. (1999). Y. (1999). Am. J. Pathol., 155, 683–694. Genomics, 59, 59–65. Knuutila S, Bjorkqvist AM, Autio K, Tarkkanen M, Wolf M, Al-Taher A, Bashein A, Nolan T, Hollingsworth M and Brady Monni O, Szymanska J, Larramendy ML, Tapper J, Pere H, G. (2000). Yeast, 17, 201–210. El-Rifai W, Hemmer S, Wasenius VM, Vidgren V and Zhu Balleine RL, Fejzo MS, Sathasivam P, Basset P, Clarke CL Y. (1998). Am. J. Pathol., 152, 1107–1123. and Byrne JA. (2000). Genes Chromosomes Cancer, 29, Lee JD, Amanai K, Shearn A and Treisman JE. (2002). 48–57. Development, 129, 5697–5706. Bova RJ, Quinn DI, Nankervis JS, Cole IE, Sheridan BF, Macdonald GA, Greenson JK, Saito K, Cherian SP, Jensen MJ, Morgan GJ, Hughes CJ and Sutherland RL. Appelman HD and Boland CR. (1998). Hepatology, 28, (1999). Clin. Cancer Res., 5, 2810–2819. 90–97. Buckley MF, Sweeney KJE, Hamilton JA, Sini RL, Makiniemi M, Hillukkala T, Tuusa J, Reini K, Vaara M, Manning DL, Nicholson RI, deFazio A, Watts CK, Huang D, Pospiech H, Majuri I, Westerling T, Makela TP Musgrove EA and Sutherland RL. (1993). Oncogene, 8, and Syvaoja JE. (2001). J. Biol. Chem., 276, 30399–30406. 2127–2133. Mansfield E, Hersperger E, Biggs J and Shearn A. (1994). Dev. Callaghan MJ, Russell AJ, Woollatt E, Sutherland GR, Biol., 165, 507–526. Sutherland RL and Watts CKW. (1998). Oncogene, 17, Mark HF, Taylor W, Afify A, Riera D, Rausch M, Huth A, 3479–3491. Gray Y, Santoro K and Bland KI. (1997). Pathobiology, 65, Callen D, Baker E, Eyre HJ, Chernos JE, Bell JA and 184–189. Sutherland GR. (1990). Ann. Genet., 33, 219–221. Miyazaki S, Imatani A, Ballard L, Marchetti A, Buttitta F, Campbell IG, Nicolai HM, Foulkes WD, Senger G, Stamp Albertsen H, Nevanlinna HA, Gallahan D and Callahan R. GW, Allan G, Boyer C, Jones K, Bast RC and Solomon E. (1997). Genomics, 46, 155–158. (1994). Hum. Mol. Genet., 3, 589–594. Mori Y, Sato F, Selaru FM, Olaru A, Perry K, Kimos MC, Dahiya R, Perinchery G, Deng G and Lee C. (1998). Int. J. Tamura G, Matsubara N, Wang S, Xu Y, Yin J, Zou TT, Oncol., 12, 811–816. Leggett B, Young J, Nukiwa T, Stine OC, Abraham JM, Diebold J, Deisenhofer I, Baretton GB, Blasenbreu S, Suchy Shibata D and Meltzer SJ. (2002). Cancer Res., 62, B, Schneiderbanger K, Meier W, Haas CJ and Lohrs U. 3641–3645. (1996). Lab. Invest., 75, 473–485. Muleris M, Almeida A, Gerbault-Seureau M, Malfoy B and El-Naggar AK, Hurr K, Kagan J, Gillenwater A, Callender D, Dutrillaux B. (1994). Genes Chromosomes Cancer, 10, Luna MA and Batsakis JG. (1997). Am. J. Surg. Pathol., 21, 160–170. 691–697. Nagai H, Pineau P, Tiollais P, Buendia MA and Dejean A. Fejzo MS, Godfrey T, Chen C, Waldman F and Gray JW. (1997). Oncogene, 14, 2927–2933. (1998). Genes Chromosomes Cancer, 22, 105–113. Nawroz H, van der Riet P, Hruban RH, Koch W, Ruppert JM Forozan F, Mahlamaki EH, Monni O, Chen Y, Veldman R, and Sidransky D. (1994). Cancer Res., 54, 11152–11155. Jiang Y, Gooden GC, Ethier SP, Kallioniemi A and Nesbit CE, Tersak JM and Prochownik EV. (1999). Oncogene, Kallioniemi OP. (2000). Cancer Res., 60, 4519–4525. 18, 3004–3016. Fujiwara Y, Monden M, Mori T, Nakamura Y and Emi M. Nogradi A. (1998). Am. J. Pathol., 153, 1–4. (1993). Cancer Res., 53, 857–860. Nupponen NN, Isola J and Visakorpi T. (2000). Genes Henderson MJ, Russell AJ, Hird S, Munoz M, Clancy JL, Chromosomes Cancer, 28, 203–210. Lehrbach GM, Calanni ST, Jans DA, Sutherland RL and Nupponen NN, Porkka K, Kakkola L, Tanner M, Persson K, Watts CK. (2002). J. Biol. Chem., 277, 26468–26478. Borg A, Isola J and Visakorpi T. (1999). Am. J. Pathol., 154, Hermsen MA, Joenje H, Arwert F, Braakhuis BJ, Baak JP, 1777–1783. Westerveld A and Slater R. (1997). Oral Oncol., 33, 414–418. Obata K, Morland SJ, Watson RH, Hitchcock A, Chenevix- Honda Y, Tojo M, Matsuzaki K, Anan T, Matsumoto M, Trench G, Thomas EJ and Campbell IG. (1998). Cancer Ando M, Saya H and Nakao M. (2002). J. Biol. Chem., 277, Res., 58, 2095–2097. 3599–3605. Orntoft TF, Thykjaer T, Waldman FM, Wolf H and Celis JE. Huibregtse JM, Scheffner M, Beaudenon S and Howley PM. (2002). Mol. Cell. Proteomics, 1, 37–45. (1995). Proc. Natl. Acad. Sci. USA, 92, 2563–2567. Parrella P, Sidransky D and Merbs SL. (1999). Cancer Res., Indsto JO, Holland EA, Kefford RF and Mann GJ. (1998). 59, 3032–3037. Cancer Genet. Cytogenet., 100, 68–71. Piao Z, Park C, Park J and Kim H. (1998). Int. J. Cancer, 75, Iwabuchi H, Sakamoto M, Sakunaga H, Ma YY, Carcangiu 29–33. ML, Pinkel D, Yang-Feng TL and Gray JW. (1995). Cancer Pieretti M, Cavalieri C, Conway PS, Gallion HH, Powell DE Res., 55, 6172–6180. and Turker MS. (1995). Int. J. Cancer, 64, 434–440. Kallioniemi A, Kallioniemi OP, Piper J, Tanner M, Savelyeva L and Schwab M. (2001). Cancer Lett., 167, Stokke T, Chen L, Smith HS, Pinkel D, Gray JW 115–123. and Waldman FM. (1994). Proc. Natl. Acad. Sci. USA, 91, Schwarz SE, Rosa JL and Scheffner M. (1998). J. Biol. Chem., 2156–2160. 273, 12148–12154. Knuutila S, Aalto Y, Autio K, Bjorkqvist AM, El-Rifai W, Seute A, Sinn HP, Schlenk RF, Emig R, Wallwiener D, Hemmer S, Huhta, T, Kettunen E, Kiuru-Kuhlefelt S, Grischke EM, Hohaus S, Dohner H, Haas R and Bentz M. Larramendy ML, Lushnikova T, Monni O, Pere H, Tapper (2001). Int. J. Cancer, 93, 80–84.
Oncogene Abnormalities of the EDD gene in human cancer JL Clancy et al 5081 Sisley K, Rennie IG, Parsons MA, Jacques R, Hammond DW, van Dekken H, Rosenberg C, Krijtenburg PJ and Alers JC. Bell SM, Potter AM and Rees RC. (1997). Genes Chromo- (1997). Histochem. Cell Biol., 108, 419–430. somes Cancer, 19, 22–28. Wong N, Lai P, Lee SW, Fan S, Pang E, Liew CT, Sheng Z, Sutherland RL, Watts CKW, Lee CSL and Musgrove AM. Lau JW and Johnson PJ. (1999). Am. J. Pathol., 154, 37–43. (1999). Human Cell Culture Vol. II, Masters JRW, Palsson B Wu X and Lieber MR. (1997). Mutat. Res., 385, 13–20. (eds). Kluwer Academic Publishers: Great Britain, pp Yamaguchi T, Matsuda K, Sagiya Y, Iwadate M, Fujino MA, 79–106. Nakamura Y and Arakawa H. (2001). Cancer Res., 61, Tsuneizumi M, Emi M, Nagai H, Harada H, Sakamoto G, 8256–8262. Kasumi F, Inoue S, Kazui T and Nakamura Y. (2001). Clin. Yokota T, Yoshimoto M, Akiyama F, Sakamoto G, Kasumi Cancer Res., 7, 3526–3532. F, Nakamura Y and Emi M. (1999). Cancer Lett., 139, 7–13.
Oncogene 7. Ethics Documentation