Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Identification of two genetic alterations in oral premalignancies and tumors using a novel fingerprinting… Ishkanian, Adrian Shea 2001

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_2001-0413.pdf [ 13.18MB ]
JSON: 831-1.0090007.json
JSON-LD: 831-1.0090007-ld.json
RDF/XML (Pretty): 831-1.0090007-rdf.xml
RDF/JSON: 831-1.0090007-rdf.json
Turtle: 831-1.0090007-turtle.txt
N-Triples: 831-1.0090007-rdf-ntriples.txt
Original Record: 831-1.0090007-source.json
Full Text

Full Text

IDENTIFICATION OF TWO GENETIC ALTERATIONS IN ORAL PREMALIGNANCIES AND TUMORS USING A NOVEL FINGERPRINTING ASSAY by ADRIAN SHEA ISHKANIAN B.Sc, Cornell University, 1996 SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Pathology and Laboratory Medicine) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA September 2001 © Adrian Shea Ishkanian, 2001 UBC Special Collections - Thesis Authorisation Form Page 1 of 1 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the Un i v e r s i t y of B r i t i s h Columbia, I agree that the Li b r a r y s h a l l make i t f r e e l y a v a i l a b l e f or reference and study. I further agree that permission f or extensive copying of t h i s t h e s i s f or s c h o l a r l y purposes may be granted by the head of my department or by his or her representatives. It i s understood that copying or p u b l i c a t i o n of t h i s t hesis f or f i n a n c i a l gain s h a l l not be allowed without my written permission. s Department The U n i v e r s i t y of B r i t i s h Columbia Vancouver, Canada 9/2/01 11 A B S T R A C T The identification of novel markers causal in oral tumorigenesis is critical for clinical diagnosis and intervention. The study of oral premalignancies is necessary in order to identify these early stage genetic events. As such tissues are typically available as archival formalin-fixed paraffin-embedded biopsies, we have modified the standard PvAPD (Randomly Amplified Polymorphic DNA) fingerprinting technique to utilize this material. This modified assay S M A L (Scanning Minute Archival Lesions) enables reproducible high-density fingerprinting of specific cell populations dissected either manually or through L C M (Laser Capture Microdissection). 235 paraffin-embedded formalin-fixed oral tumor and premalignant biopsies were tested using the S M A L assay. Two novel genetic changes in oral cancer progression stages were identified. These changes were mapped to chromosomes 2p23 and 7q22. Genetic instability at these two loci was verified by L O H (Loss Of Heterozygosity) analysis. 72 oral specimens (38 tumors and 34 dysplasias) were used to define a minimal boundary of allelic imbalance at these loci by LOH. Candidate genes Anaplastic Lymphoma Kinase, Dynein, Phosphatidylinositol 3-Kinase and Fos-Like Antigen 2 have been identified. T A B L E O F C O N T E N T S ABSTRACT ii LIST OF TABLES v LIST OF FIGURES vi ACKNOWLEDGEMENTS viii ABBREVIATIONS ix CHAPTER 1 INTRODUCTION 1 1.1 Background 1 1.2 Molecular strategies for analyzing predictive risk in HNSCC 3 1.3 Chromosomal regions implicated in HNSCC 5 1.4 Chromosomal regions implicated in oralpremalignancies 6 1.5 Genes implicated in HNSCC 8 1.6 Review of current methods for genome-wide scanning 10 1.6.1 Comparative Genomic Hybridization (CGH) 10 1.6.2 Loss of Heterozygosity Analysis (LOH) 11 1.6.3 Limitations of LOH analysis 13 1.7 Rationale for the use of archival specimens in genome-wide scanning for novel markers 14 1.8 Rationale for the development of a novel method capable of high-density genome-wide scanning for novel regions of genetic instability in HNSCC 15 1.9 Objectives of thesis 17 CHAPTER 2 METHODS AND MATERIALS 20 2.1 Oral biopsies: microdissection and DNA extraction 20 2.2 SMAL-PCR amplification 20 2.3 Cloning SMAL fragments 22 2.4 Localization of SMAL PCR clones to a chromosomal position 24 2.5 Loss of Heterozygosity analysis 25 CHAPTER 3 RESULTS 26 3.1 Sample Preparation and Reproducibility of Assay 26 3.1.1 Microdissection and DNA extraction from oral archival biopsies 26 3.1.2 Development of genome-wide SMAL-PCR and reproducibility of method 31 Method Development 31 Reproducibility of SMAL-PCR 36 Application of SMAL-PCR across tissue types 39 3.2 Discovery of a Novel Region Involved in HNSCC Progression on Chromosome 7Q22 41 3.2.1 Identification of a recurring SMAL alteration in oral SCCs 41 3.2.2 Verification of alteration at 7q22-31 45 3.2.3 Two discontiguous regions of AI/LOH at 7q22-31 45 3.2.4 Do these alterations occur in early stages? 47 3.2.5 Definition of boundaries for Al within D7S1812-D7S477 49 3.3 Discovery of a Novel Region Involved in HNSCC Progression on Chromosome 2P23 54 3.3.1 Identification of a recurring SMAL alteration in oral premalignant lesions and SCCs 54 3.3.2 Fine mapping region of genetic alteration by AI/LOH at 2p23 60 3.3.3 Definition of boundaries for Al within D2S2247 and D2S400 62 CHAPTER 4 DISCUSSION 63 4.1 Benefits of SMAL PCR assay for genome-wide scanning 63 4.2 7q22-31 as a novel region of allelic imbalance in HNSCC 65 4.3 Identification of candidate genes in HNSCC at 7q31 66 4.4 Identification of candidate genes in HNSCC at 7q22 67 4.5 Identification of candidate genes in HNSCC at 2p23 68 4.6 Future directions 70 4.7 Bacterial artificial chromosome comparative genomic hybridization (BAC-CGH): The future offine mapping allelic imbalance 71 CHAPTER 5 CONCLUSIONS 74 REFERENCES 75 Appendix I. Complete List of Oral Archival Premalignant Squamous Cell Carcinoma (SCC) Samples (Patient Number and DNA Yield) 83 Appendix II. Procedure for Cloning SMAL Fragments From Polyacrylamide Gels. 87 Appendix III. Complete List of L O H Primer Sequences and Locations. 92 Appendix IV. BAC tiling path on 7q22. 95 Appendix V. BAC tiling path on 7q31.100 Appendix VI. Complete summary of L O H on tumor specimens (T) in region between 7q22-31107 Appendix VII. Complete summary of L O H on dysplasia specimens (D) on region between 7q22-31 108 Appendix VIII. A physical map of the region between 7q22 and 7q31.109 V LIST O F T A B L E S T A B L E 1. PRIMERS USED FOR SMAL-PCR 21 T A B L E 2. QUANTITY OF DNA OBTAINED FROM 235 MICRODISSECTED ORAL PREM A L I G N A N T LESIONS 30 T A B L E 3. NUMBER OF PRODUCTS OBTAINED USING DIFFERENT ARBITRARY PRIMER COMBINATIONS 34 T A B L E 4. REPRODUCIBILITY OF SMAL PRODUCTS BETWEEN TWO INDEPENDENT PCR REACTIONS 38 V I LIST O F F IGURES FIGURE 1. THE ACCUMULATION OF GENETIC CHANGES DETERMINES THE HISTOPATHOLOGICAL PROGRESSION OF HNSCC (CALIFANO, 1996) 4 FIGURE 2. SCHEMATIC OF L O H ANALYSIS 12 FIGURE 3. E X A M P L E OF LASER CAPTURE MICRODISSECTION 16 FIGURE 4. SCHEMATIC OF RAPD-PCR ANALYSIS 18 FIGURE 5. MICRODISSECTION OF SEVERE ORAL DYSPLASIA SPECIMEN 28 FIGURE 6. SUMMARY OF DNA QUANTITY GENERATED FROM 153 ORAL ARCHIVAL SPECIMENS (NOTE 153/235 TOTAL ANALYZED) 29 FIGURE 7. IN-SILICO PREDICTION OF PRODUCTS GENERATED FROM SPECIFIC PRIMER PAIRS AIDS IN PRIMER SELECTION FOR GENOME-WIDE FINGERPRINTING...32 FIGURE 8. 7 ARBITRARY DECANUCLEOTIDE PRIMERS TESTED WITH T H E SMAL-PCR PROTOCOL ALONE AND IN COMBINATION 33 FIGURE 9. SMAL FINGERPRINTING TITRATION OF A SINGLE ARCHIVAL BIOPSY 35 FIGURE 10. FINGERPRINTS OF 4 PATIENTS WITH PREMALIGNANT LESIONS IN WHICH DYSPLASTIC EPITHELIAL CELLS AND CORRESPONDING NORMAL CONNECTIVE TISSUE ARE COMPARED 37 FIGURE 11. SMAL-PCR FINGERPRINTS FROM MULTIPLE ARCHIVAL TUMOR TYPES....40 FIGURE 12. GAIN OF A 150BP FRAGMENT FROM 3/20 ARCHIVAL HNSCC SPECIMENS....42 FIGURE 13. BLAST SEARCH USING CLONED 150BP FRAGMENT LOCALIZES SEQUENCE TO BAC CLONE RG030L05 43 FIGURE 14. BAC TILING PATH CONSTRUCTED AT WASHINGTON UNIVERSITY SEQUENCING CENTRE ON CHROMOSOME 7Q22 44 FIGURE 15. CONFORMATION OF ALLELIC IMBALANCE IN THREE PATIENTS SHOWING SMAL ALTERATION AT 7Q22 46 FIGURE 16. A. FOUR CONTIGS SPANNING REGION BETWEEN 7Q22-31. B. SUBSET OF L O H ANALYSIS ON 38 TUMOR SPECIMENS 48 FIGURE 17. A. FOUR CONTIGS SPANNING REGION BETWEEN 7Q22-31. B. SUBSET OF L O H ANALYSIS ON 34 DYSPLASIA SPECIMENS 50 FIGURE 18. L O H AT D7S2539 AND RETENTION AT D7S1812 AND D7S479 IN TUMOR PATIENTS 112 AND 24 AND DYSPLASIA PATIENT 208 51 FIGURE 19. COMPOSITE DIAGRAM OF THE TWO DISCONTIGUOUS REGIONS BETWEEN 7Q22 AND 7Q31 WITH ALLELIC IMBALANCE 53 FIGURE 20.150BP LOSS IN 4/20 TUMORS AND 1/30 DYSPLASIAS 55. FIGURE 21. FISH LOCALIZATION OF BAC N0175E15 TO 2P23 56 FIGURE 22. A. A SINGLE CONTIG (13214) SPANS REGION OF INTEREST ON CHROMOSOME 2P23. B. L O H ANALYSIS OF 49 ORAL SCC SPECIMENS 57 FIGURE 23. A. A SINGLE CONTIG (13214) SPANS REGION OF INTEREST ON CHROMOSOME 2P23. B. L O H ANALYSIS OF 49 ORAL DYSPLASIA SPECIMENS (D) OF VARYING GRADE 59 FIGURE 24. COMPOSITE DIAGRAM OF REGION ON CHROMOSOME 2P23 SHOWING ALLELIC IMBALANCE 61 V l l l A C K N O W L E D G E M E N T S I would like to thank the Rosin lab for L O H analysis and D N A extraction and Dr. Lewei Zhang for tissue pathology and manual microdissection. Thank you to David Walsh, Neil Klompas, Jennifer Kennet, Chad Malloff, K im Lonergan, and Cathie Garnis for help with colony fingerprinting, plasmid preparation, colony PCR, literature searches and S M A L PCR analysis. I would also like to thank Dr. Juergen Vielkind and Stephanie Ma for the Laser Capture Microdissection. I would further like to acknowledge Jennifer Skaug at the Toronto Hospital for Sick Children in both screening radiation hybrid panels and FISH analysis. Finally, I must thank both my supervisors, Dr. Wan Lam and Dr. Victor Ling and the other members of my examination committee, Dr. Jurgen Vielkind, Dr. Calum MacAulay, Dr. Carolyn Brown Dr. Miriam Rosin and Dr. Decheng Yang for their continual support and guidance throughout this project. ABBREVIATIONS A l Allelic Imbalance AP-PCR Arbitrarily Primed Polymerase Chain Reaction BAC Bacterial Artificial Chromosome HNSCC Head and Neck Squamous Cell Carcinoma L O H Loss of Heterozygosity OMIM Online Mendelian Inheritance of Man PCR Polymerase Chain Reaction RAPD Randomly Amplified Polymorphic DNA SCC Squamous Cell Carcinoma SMAL Scanning Microdissected Archival Lesions SNP Single Nucleotide Polymorphism 1 C H A P T E R 1. I N T R O D U C T I O N 1.1 Background Squamous cell carcinoma of the head and neck (HNSCC) accounts for more than 90% of the malignancies of the upper aerodigestive tract. More than 300,000 new cases of HNSCC cancer are diagnosed each year worldwide, and more than 8,000 deaths occur due to this disease in North America alone. Despite advances in diagnosis and treatment, the 5-year survival rate for HNSCC has remained at approximately 50 percent for the last twenty years, one of the lowest among all cancers (Boring, 1992). This is largely due to primary diagnosis at a late disease stage and the high rate of recurrent or second primary cancer (10-40%). Prognosis for cure decreases after invasion by approximately one half i f tumor spread has occurred. Therefore, successful treatment of HNSCC is largely dependent on early diagnosis (Silverman, 1990). Established tumor-based prognostic factors in HNSCC use clinical attributes to predict treatment strategies and outcome. Such factors include the presence and extent of nodal metastases, tumor stage, tumor site, total tumor volume and tumor size. While each factor does affect survival, all are crude measures of predictive risk. Pathological scoring of this kind is subjective and cannot accurately predict which precancerous lesions may progress with time. Further, in cases of oral leukoplakia (the most common premalignancy in HNSCC), while the degree of the preneoplastic lesion is significant to outcome, many patients with mild dysplasia or no evidence of any dysplasia can be at 2 risk for developing oral cancer. Finally, pathological evaluation of tissue pathology cannot assess tumor heterogeneity. Recently, the focus has shifted towards discovering biological markers correlated to predictive risk. While some markers would correlate strongly to poor prognosis, others may indicate a potential regression of a premalignant lesion and negate an invasive treatment strategy. Assays of such markers could also be used to monitor low risk patients by analyzing tissue scrapings or saliva rather than gross biopsies. Changes in such molecular markers at surrounding tissue sites after resection or chemoprevention in high-risk patients could also be used to assess future recurrence and survival. Finally, such markers could potentially be used to monitor risk in healthy current or former smokers, who are at an elevated risk for developing oral cancer in the future. In all instances molecular strategies represent a method for more accurate, less invasive and earlier prognosis of HNSCC (Quon, 2001, Lippman, 2001). Epithelial carcinogenesis occurs through distinct histological changes. These changes are believed to occur through the accumulation of mutations in specific genes. This causal role of genetic changes in cancer progression was discovered first by Vogelstein and Fearon in their model for the development of colon cancer. In this hallmark paper, four fundamental observations were reported: 1) Tumors arise through the activation of oncogenes coupled with the inactivation of tumor suppressor genes. 2) Mutations in at least four or five genes are required for the formation of a malignant tumor. 3) The total accumulation of changes, rather than their order is responsible for a tumor's biologic properties. 4) Mutant tumor suppressor genes are not recessive, and can 3 exert a phenotypic effect in a heterozygous state. Finally, it was suggested that this model may be applicable to other epithelial neoplasms. 1.2 Molecular strategies for analyzing predictive risk in HNSCC Between 1990 and the present, histopathological progression models have been proposed for many other cancer types. In 1996, Califano et al developed a genetic progression model for squamous cell carcinoma (Figure 1). This group assayed for allelic imbalance at 10 chromosomal loci implicated in a variety of epithelial cancers using microsatellite analysis. By allelotyping at a variety of stages, including benign lesions, hyperplasias, dysplasias, carcinoma in situ and invasive lesions, Califano et al discovered that the number of genetic changes progressively increased at each histopathological step. The development of a fully malignant phenotype, therefore, was most strongly correlated to the overall accumulation of genetic changes. Further, only a subset of chromosomal loci tested showed consistent loss from early preneoplastic lesions to fully invasive cancer, suggesting there was a specific pathway of mutagenic events through which these lesions progressed. The significance of these findings were two-fold. Firstly, Califano et al demonstrated that the discovery of genes involved in critical progression events would rely on alleleotyping the whole spectrum of oral cancer stages, from benign and hyperplastic lesions to fully malignant cancers. Secondly, this study highlighted the need for many additional markers in order to elucidate the pathways through which oral carcinogenesis occurs. 4 Figure 1. Califano et al (1996) demonstrate here that the accumulation of genetic changes determines the histopathological progression of HNSCC. 5 From 1996 to the present, comprehensive analysis of allelic imbalance at known or predicted chromosomal regions has been conducted on HNSCC in an effort to identify new markers correlated to the progression of this disease. Data from 45 Loss of Heterozygosity (LOH) and Comparative Genomic Hybridization (CGH) studies conducted between 1998 and 2001 are summarized in Section 1.3. These will provide some insight on the frequency of genomic alterations at each chromosome arm. 1.3 Chromosomal regions implicated in HNSCC Bockmuhl et al (1998) used C G H to asses genomic integrity of 50 primary head and neck squamous cell carcinomas (see 1.6 for C G H description). Deletions were found on chromosomes lp32,35-26 (23/50), 4q (34/50), 5q21-22 (38/50), 9p21 (38/50), 13q (38/50), 18q and 21q (38/50); and amplifications on 3q (9/50), l l q l 3 (27/50), 17q and 22q (23/50), 16p (21/50), l q and 20q (20/50) and 5p (18/50). Gasparotto et al (1999) found Loss of Heterozygosity (LOH) at 43% of 47 HNSCC's at 10q22-26 (see 1.6 for L O H description). In a study by Gupta et al (1999) on 145 HNSCC's, 69% showed L O H at one or more loci on 13ql4.3-21. Using 7 microsatellite markers, Ishwad et al (1999) found 23/51 (45%) HNSCC tumors and 23/29 (79%) HNSCC cell lines to have L O H within a 5cM region on 8p23. Noting that many previous studies have supported the presence of a novel gene causal in many cancer types at 7q22-31.1, Resto et al (2000) used 45 HNSCC's to confirm this was also a region of allelic imbalance in HNSCC. Wang et al (1998) found 19/34 oral SCC patients to show allelic imbalace (Al) at one or more markers between 7q21.3-qter. Further, frequent L O H was restricted to a l c M 6 interval, suggesting a novel tumor suppressor gene may be located in this region. In a review by Jefferies et al (2001) a high incidence of L O H was observed on 9p21-22 (50%) and 17pl3 (87%). Three discrete regions of allelic loss were also commonly reported on 3p (3pl3-14, 3p21 and 3p24-ter). In a study of 42 HNSCC patients, Lydiatt et al (1998) demonstrated L O H of 9p21 correlates with recurrence of disease. Matsumura et al (2000) used C G H and Fluorescent In Situ Hybridization (FISH) on 17 HNSCC cell lines to map an amplified region to within ~1.7Mb on 22ql 1.2-12. Loss of 18q in HNSCC was evaluated by Pearlstein et al (1998). This study assessed 67 patients and found 40%> showed A l at one or more of 4 loci on 18q. Further, these 40% had statistically significant poorer survival than those with retention at this locus, suggesting a region on this arm may be an important prognostic indicator in HNSCC. Ogawara et al (1998) used L O H to study chromosome 13 deletions at 18 distinct polymorphic loci in 34 oral SCC patients. 67.6% showed allelic imbalance at one or more loci, and a minimal region of deletion was found at 13ql4.3. In an extensive study of allelic imbalance using 52 polymorphic markers at 13 key chromosomal regions previously implicated in HNSCC, Partridge et al (1999) discovered that A l at 3pl4.3-12.1 and 9p21 was found to be the most reliable predictor of patient outcome. This study concluded that a more accurate prognostic system could be developed i f these two markers were used in tandem with conventional staging systems. 1.4 Chromosomal regions implicated in oral premalignancies In a study the following year, Partridge et al (2000) compared 39 dysplastic lesions that later developed a tumor with similar early lesions that did not progress. Using a subset of the original markers found to be most informative in his previous study 7 on tumors (3p, 8p, 9p, 13q), this group found A l at more loci in lesions known to progress than those that did not. A l at two or more of these specific markers was enough for this group to advocate complete excision of all suspicious areas regardless of the pathological stage of the lesion. In two separate L O H studies on premalignant lesions of varying grade, Califano et al (1996;2000) reported allelic imbalance in 30-50% of patients at 5 distinct microsatellite markers on 3p, 40-70% at 4 markers on 9p, 1 lq l3 , 14q24, and 40-60% at 17pl3. Chan et al (2000) reported 3p deletions in 81-100% of nasopharyngeal epithelial cancers and 75% dysplasias (n=113). In a study using C G H to assess 45 low grade (no regional lymph node involvement) HNSCC's , Redon et al (2001) found only three chromosomes were affected significantly (3p loss in 73%, 8q gain in 47% and l l q l 3 gain in 27%). By comparing 116 progressing and nonprogressing oral premalignant lesions at 19 microsatellite loci (3p, 4q, 8p, 9p, l l q , 13q and 17p), Rosin et al (2000) demonstrated that progressing lesions showed dramatically different patterns of A l . Further, loss at one or more loci on 3p and 9p was essential for progression and any additional A l gave a 33-fold increase in relative risk of progression. Rosin concluded that assessing A l at 3p and 9p in early premalignancies might provide additional information on relative risk of progression. Taken together, these findings suggest the most common areas of loss in HNSCC occur on chromosomes 3p, 9p and 17p. Other chromosomes with relatively high incidence of genetic imbalance include chromosome 1, 5, 7, 8, 11, 13 and 18. Regions defined by C G H and L O H in the studies reported above typically define large regions of chromosomal imbalance. However, within a subset of these large regions lie a number of oncogenes and tumor suppressor genes which may be implicated in HSCC progression. 8 1.5 Genes implicated in HNSCC p i 6 (CDKN2A) is found on chromosome 9p21, one of the most widespread regions of allelic imbalance in oral cancer. This gene encodes a 16kDa protein that binds to cyclin-dependent kinase (CDK)4 and 6, inhibiting association with cyclin D l , thereby inhibiting the cell cycle G l /S transition (Serrano, 1993). Somatic mutations of p i 6 occur in 10% of HNSCC and homozygous deletions occur in approximately 50%> of cases (Reed, 1996). Miracca et al found 59% of 47 HNSCC cases to have pi6 inactivation, 53%) of which showed LOH at 9p21 implicating this as the gene involved in HNSCC progression at this chromosomal locus. Unfortunately, correlation between loss of pi6 and clinical outcome has not been demonstrated to date. Tp53 is found on chromosome 17p 13.1, another region showing a high degree of A l . It has been suggested that wildtype p53 may play a role in D N A repair and that expression of mutant forms of p53 may alter cellular resistance to D N A damage. Furthermore, p53 had been thought to function as a cell cycle checkpoint, also suggesting that mutant p53 might change the cellular proliferative response to D N A damage (Lee, 1993). This gene normally functions in maintaining cellular integrity after D N A damage. Mutations in 60%> of HNSCC's have been reported (Hainaut, 1998). Mutations in p53 have been related to a shorter time to recurrence of this disease. Aberrant expression of mutant p53 has also been correlated to early recurrence and/or recurrence of a second primary tumor (Shin, 1996). FHIT is found on chromosome 3pl4.2, one of three regions on 3p that when taken together (pi 3-14, p21, p24-ter) account for the second most frequent region of alteration 9 in HNSCC (45-55%) (Virgilio, 1996). The location of FHIT at 3p as well as its involvement in lung cell carcinomas (Sozzi, 1996) and digestive tract cancers (Ohta, 1996) makes it a candidate tumor suppressor gene in oral cancer. The retinoic acid receptor (RAR) P located at 3p24 has also been implicated in the development of HNSCC (Hu, 1991). Cyclin D\(PRAD 1) is found on chromosome l l q l 3 and has been implicated in 30-50% of HNSCC's . Cyclin D l , like other cyclins, can form a complex with and activate p34(cdc2) protein kinase, thereby regulating progression through the cell cycle (Motokura, 1991). Overexpression of cyclin D l has been correlated with poor clinical outcome and is a potential marker for patient prognosis (Michalides, 1995). Overexpression of the epidermal growth factor receptor (EGFR) (7p 12.3-12.1) has also been strongly correlated to HNSCC (Grandis, 1998). This gene is a member of the receptor protein tyrosine kinase family. EGFR binding to a variety of ligands, including transforming growth factor-alpha (TGF-ct) (2pl3), causes activation of a variety of pathways involving ras/raf-l/MAP kinase, PI-3-kinase and Phospholipase Cy (Tannock, 1998). These pathways are involved in cell growth and proliferation. Overexpression of EGFR has been found in 40-80% of HNSCC (Wen, 1996). Further, in a study with 91 HNSCC patients treated with surgical resection and radiation therapy, Grandis et al (1998) concluded that quantitation of EGFR protein levels in primary head and neck squamous cell carcinomas may be useful in identifying subgroups of patients at high risk of tumor recurrence and in guiding therapy. In one novel approach to therapy, antibodies that bind to EGFR have been shown to block cell proliferation in oral cancers. 10 When used in conjunction with standard treatment regimes including chemotherapy such strategies have increased treatment efficacy (Wheeler, 1999). It is clear that after many years of genomic analysis of HNSCC, there is still a fundamental lack of precise clinical and prognostic markers capable of identifying which oral premalignant lesions are potentially malignant. While it is clear that those lesions that do become malignant will accumulate a variety of genetic changes, many of those changes causal in this progression event have yet to be discovered. 1.6 Review of current methods for genome-wide scanning 1.6.1 Comparative Genomic Hybridization (CGH) In C G H , equal amounts of differently labeled tumor D N A and normal reference D N A are hybridized simultaneously to normal metaphase chromosomes (Kallioniemi, 1993). They are then visualized by different fluorochromes, and the signal intensities are quantitiated separately as gray levels along the single chromosomes. The over and underrepresented D N A segments are determined by computation of ratio images and average ratio profiles. While C G H patterns of chromosomal imbalances may help to define the malignant potential of head and neck squamous cell carcinomas (see Section 1.3,1.4), and may also define novel regions of amplification and deletion, there are certain fundamental limitations to this assay. In particular, because data is collected by visualization of metaphase chromosome spreads through a microscope, resolution of detection of genetic changes is on the order of a chromosome band (5-10 Mb). This 11 provides only a crude measure of the exact location of the alteration of interest. Further, detection of a deletion by C G H relies on the ability to distinguish a two-fold intensity difference between fluorophores under ideal circumstances, representing the loss of one chromosomal arm in the tumor D N A compared to the normal D N A reference. This level of resolution is notoriously difficult by C G H and highlights the obvious need for novel genome-wide methods capable of higher resolution scanning (Forozan, 1997). 1.6.2 Loss of Heterozygosity Analysis (LOH) L O H typically refers to the PCR based strategy that uses simple sequence repeat (SSR) polymorphisms located throughout the genome to search for allelic loss in disease tissue versus normal tissue. Maternal and paternal chromosomes contain differences in sequences at most genetic loci. This technique uses primer sets flanking known regions containing SSR's. These primer sets are unique to specific loci and will amplify fragments from both parental alleles. While there are tens of thousands of SSR's, currently, approximately six thousand of these markers exist as informative microsatellite markers, useful for L O H analysis. This allows information on genetic instability to be derived at virtually any region of the genome to within 1-5 megabases (Murray, 1994; Dib, 1996). These primers are radiolabeled prior to PCR. Following amplification, fragments are run on a denaturing polyacrylamide gel and visualized by autoradiography. Maternal and paternal alleles are resolved as two distinct bands, separated only by their differences in base pair composition and degree of polymorphism. (Figure 2). L O H in somatic tissue can inactivate tumor suppressor genes causal to cancer progression via two 12 Normal cells Abnormal cells Figure 2. Schematic of L O H analysis. L O H can detect large genomic alterations. A . Schematic representation of a single chromosme pair from a normal specimen and an abnormal specimen containing a loss of a chromosomal band. Green arrows indicate region of retention. Red arrows indicate region of loss. B. Schematic of PCR amplified fragments as they are visualized on a denaturing polyacrylamide gel. Red fragments show both alleles retained in normal cells and one allele lost in abnormal cells. Green fragments show both alleles retained in both tissue types. C. Actual gel showing same L O H results asB. 13 general mechanisms: 1) when a wild-type allele is replaced by a duplicated copy of the homologous chromosome region that carries the mutant allele (through mitotic recombination) or 2) when there is a deletion of the normal allele and there remains only the mutant allele. In either case, this event is visualized by differences in signal ratio between the normal tissue and disease tissue of a single patient. L O H in multiple patients at a specific polymorphic chromosomal locus can define regions of genomic instability and has aided in the discovery of novel genes in HNSCC. L O H has been used extensively in the analysis of genomic instability in HNSCC. Because it is a PCR based assay, it is capable of detecting allelic imbalance from just a few hundred cells. It is one of very few assays capable of generating reproducible information from specific microdissected cell populations from formalin-fixed, paraffin embedded archival tissue specimens. As there is a tremendous store of these tissue blocks in most hospital archives worldwide, it provides a cheap and accurate method to asses allelic imbalance at a given locus. 1.6.3 Limitations of LOH analysis L O H is an inappropriate tool for high-throughput discovery of novel markers in HNSCC. Each assay requires microdissection and D N A extraction from a large patient set and generates information at only one single locus, making fine mapping a single region of interest labor intensive. Further, many regions throughout the genome do not contain known markers, making LOH analysis impossible. Finally, these studies are 14 biased towards markers previously discovered in other cancer types. New assays must be employed for the discovery of novel markers correlated to HNSCC development. 1.7 Rationale for the use of archival specimens in genome-wide scanning for novel markers The vast stores of tissue blocks archived in pathology departments and hospital laboratories worldwide represents a tremendous resource of tissue for use in novel gene discovery. Longitudinal studies for identifying genetic alterations common to multiple patients over time can be conducted with available materials without prospective collection. In studying cancer progression, such resources would facilitate the identification of somatic mutations associated with pathohistological stages of development. Molecular analysis of the early stages of disease progression would be contingent upon such archives when the disease tissue is not available in other forms. For example, many cancers are thought to develop through histopathological stages (hyperplasia, mild, moderate and severe dysplasia, carcinoma in situ, and invasive cancer), with this progression driven by the accumulation of specific genetic alterations. Since random genetic changes are often associated with tumors, examining premalignant cells would more easily identify changes critical to disease progression. A major obstacle in identifying stage-associated genetic changes, however, is the heterogeneity of cell type within tumor biopsies, as they commonly include blood, stroma, connective and normal tissue and premalignant tissue of varying grades. As tissue microdissection allows precise isolation of cell cluster from slides, it has been commonly used to reduce this problem of normal cell contamination. Recently, the efficiency of this technique has been significantly increased with the advent of Laser 15 Capture Microdissection (LCM). Further, L C M can be used to isolate smaller and more complex cell clusters than was previously possible with manual microdissection. (Simone, 1998; Howe, 1997; Luo, 1999). An example of L C M shown in Figure 3 was kindly provided by Dr. J. Vielkind. Both the quality and quantity of D N A typically extracted from microdissected cell clusters on archived slides limits its use in standard assays. Gene expression profiling using microarray technology is hampered by the RNA quality in such specimens, as they are not treated by fixation methods designed to preserve RNA. Although emergent technologies such as single nucleotide polymorphism chips (SNP's) have been applied to both gene based and genetic map construction, they are still under development. Their use to date requires a probe concentration far exceeding that which could be generated from archival sources and reproducible data has not been produced from archival material to date (Wang, 1998; Pollack, 1999). 1.8 Rationale for the development of a novel method capable of high-density genome-wide scanning for novel regions ofgenetic instability in HNSCC Although locus-by-locus alleleotyping of microdissected material has been instrumental in deducing the pattern of loss of chromosomal intervals during cancer progression, higher resolution genome scanning could be achieved using the quantity of material by simultaneously monitoring multiple loci. One successful approach is the Randomly Amplified Polymorphic D N A (RAPD) fingerprinting assay (commonly 16 Figure 3. Example of Laser Capture Microdissection. Prostate tumor cells isolated from a heterogeneous tissue section with individual laser shots. A . Tissue section prior to Laser Capture Microdissection. B. Tissue section after a specific cell population of interest is fused to cap and removed. C. Procured cells fixed to transfer film on cap. Courtesy of Dr. Jurgen Vielkind. 17 referred to as Arbitrarily Primed PCR or AP-PCR. This PCR based technique uses short arbitrarily derived primers to amplify dozens of fragments throughout the genome of a given sample. PCR products are then resolved either on agarose or polyacrylamide gels (Williams, 1990; McClelland, 1997) (Figure 4). R A P D has been used extensively in bacterial strain typing and genetic map construction of plants and animals. Recently, this technique has been used successfully to detect novel genetic alterations in frozen tumor specimens (Postlethwait, 1994; Cushwa, 1996; Ohtsuka, 1999; Malkhosyan, 1998; Maeda, 1999; Perucho, 1995; D i l -Afroze, 1998; Ong, 1998). The discovery of the microsatellite mutator phenotype by AP-PCR also demonstrates the value of unbiased genome-wide D N A fingerprinting. In these studies, gains, losses and intensity changes between patients' normal and disease tissue D N A have been observed. These changes have mainly been attributed to amplifications and deletions of genomic material, and some have shown that insertions and chromosomal rearrangements may also be detected. Finally, as this assay utilizes arbitrarily derived primers to generate PCR products, it will provide an unbiased genome-wide scanning for genetic alterations independent of mutational hotspots previously identified. 1.9 Objectives of thesis The first objective of this thesis was to adapt RAPD PCR for genome-wide scanning of microdissected archival formalin fixed paraffin-embedded samples. While the value of this application is evident, there are no 18 Figure 4. Schematic of RAPD PCR Analysis. Red arrows indicate two arbitrarily derived ten base pair primers. In above examples, a single primer pair is being used to amplify multiple fragments in 6 patients containing deletions of different sizes (dotted lines) in a specific region of a chromosome. The fragment generated in that region (indicated in blue) will be amplified in all samples where no deletion is present (all normal samples [N], and 1/6 tumor samples [T]). 19 previous studies addressing the suitability of RAPD for material of limiting quantity and quality. Results below indicate that modification of the standard R A P D PCR procedure (which I have termed S M A L - Scanning Microdissected Archival Lesions) can be readily used for high-density, high-resolution comparative genome scanning of such samples. With a functional assay in place the next goal was to procure a large number of premalignant and malignant paraffin tissue specimens for large-scale analysis. In collaboration with the Provincial Oral Biopsy Service and the Vancouver Hospital & Health Science Centre Archive, a large cohort of specimens was acquired. S M A L - P C R was then applied to analyze hundreds of novel loci across a large patient subset. The second objective of this thesis was to use this assay to identify critical genetic events in oral tumorigenesis by cloning and characterizing genomic alterations found at multiple points in progression of the disease. Novel stage specific genomic alterations found by S M A L PCR must be confirmed through a separate method. Fine mapping these regions by defining the minimal region of alteration among a large patient set aids in novel gene discovery. The third objective of this thesis was to use to conduct a large-scale fine-mapping study of allelic imbalance in any novel region identified. These objectives were developed in order to address the following hypothesis: Simultaneous analysis of multiple genomic targets in minute archival specimens will detect genetic alterations in premalignant cells. Genetic alterations critical to cancer progression will occur in the early stages of tumorigenesis, and will be found at equal or greater frequencies in fully malignant cell populations. 20 C H A P T E R 2. M E T H O D S A N D M A T E R I A L S 2.1 Oral biopsies: microdissection and DNA extraction This study used 235 paraffin-embedded samples from the archive of the Oral Biopsy Service of British Columbia. This sample set included 149 dysplasias of varying degrees and 86 squamous cell carcinomas (SCC). In all cases, the histological diagnosis was confirmed by at least 2 oral pathologists. Serial sections were cut from each biopsy, stained with haematoxylin and eosin, and manually microdissected. Dysplastic or malignant cells were digested with SDS/proteinase K at 48°C and spiked twice a day for 72 hrs with fresh enzyme (Zhang et al., 1997). The D N A was then extracted two times with a phenol-chloroform mixture before ethanol precipitation. D N A was quantified fluorometrically using Picogreen (Molecular Probes). Connective tissue from each specimen was employed as a source of normal D N A (See Appendix I for complete list). 2.2 SMAL-PCR amplification Oligonucleotide primers for S M A L PCR were purchased from Operon Technologies and Alpha D N A Inc. [] []. Primers assessed are listed in Table 1. 20picomoles/reaction/primer were labeled with 2uCi of adenosine 5' [y-32 P] ATP (6000 Ci/mmol; Amersham), at 37°C for 1 hr and 65°C for 5 min. PCR reactions were performed in a lOul volume containing 2 nanograms (ng) 21 Primer Name Primer Sequence 01 5'-CAGGCCCTTC-3' 02 5'-TGCCGAGCTG-3' 03 5 ' -AGTCAGCCAC-3 ' 04 5-AATCGGGCTG-3 ' 05 5'-AGGGGTCTTG-3' 06 5'-GGTCCCTGAC-3' 07 5 ' -GAAACGGGTG-3 ' 08 5-GTGACGTAGG-3 ' 09 5 ' -GGGTAACGCC-3' 10 5 ' -GTGATCGCAG-3' 11 5 '-CAATCGCCGT-3' 12 5 '-TCGGCGATAG-3' 13 5 ' -CAGCACCCAC-3 ' 14 5'-TCTGTGCTGG-3' 15 5 '-TTCCGAACCC-3' 16 5 ' -AGCCAGCGAA-3 ' 17 5'-GACCGCTTGT-3' 18 5'-AGGTG ACCGT-3 ' 19 5'-CAAACGTCGG-3* 20 5'-GTTGCGATCC-3' Primer Name Primer Sequence 21 5'-GTTTCGCTCC-3* 22 5'-TGATCCCTGG-3' 23 5'-CATCCCCCTG-3' 24 5 '-GGACTGGAGT-3' 25 5'-TGCGCCCTTC-3' 26 5'-TGCTCTGCCC-3' 27 5'-GGTGACGC AG-3 ' 28 5 ' -GTCCACACGG-3 ' 29 5'-TGGGGGACTC-3' 30 5 '-CTGCTGGGAC-3' 31 5 ' -GTAGACCCGT-3' 32 5 ' -CCTTGACGCA-3' 33 5'-TTCCCCCGCT-3' 34 5'-TCCGCTCTGG-3' 35 5'-GGAGGGTGTT-3' 36 5'-TTTGCCCGGA-3' 37 5 ' -AGGGAACGAG-3 38 5 ' -CCACAGCAGT-3 ' 39 5*-ACCCCCGAAG-3' 40 5 ' -GGACCCTTAC-3' Table 1. Primers Used for S M A L PCR 22 of DNA, 2.5 units of Recombinant Taq D N A polymerase, 200uM each dNTP, lOmM Tris-Cl, pH 8.3, 50 m M KC1, 2mM M g C l 2 and 0.001% gelatin. A l l reactions were performed in a thermal cycler (MJ Research PTC-100) for 45 cycles of 94°C for lmin, 35°C for lmin and 72°C for 2min. PCR reaction products were run on 5%> or 6% non-denaturing polyacrylamide gels at 800V for 2 1 / 2 or 3 1 / 2 hrs. A l l gels were exposed on Kodak X - O M A T A R autoradiography film. S M A L PCR fingerprints were generated from pairs of tumor/dysplasia and normal D N A from each patient. Bulk D N A extracted from a frozen lung tissue was included in each experiment as a positive control and a mock reaction (no D N A template) as a negative control. Signals that recurred between multiple tumors were chosen for further analysis. 2.3 Cloning SMAL fragments Fragments excised from gels shown in figures 12 and 20 were cut from dried polyacrylamide gels guided by the image from the autoradiography film. D N A was eluted by boiling the excised gel slice in dt^O for 5 minutes. Since two primers were used to generate the S M A L PCR products, the identity of the primers at the termini of the fragment of interest need to be determined. Each fragment was analyzed in three independent PCR reactions using each single primer from the original S M A L PCR and the two primers combined. The eluted D N A (10% of eluted volume) was re-amplified using appropriate primer(s), which was modified to contain the sequence of restriction 23 enzyme recognition sites to facilitate subsequent cloning into a plasmid vector. PCR was performed using 2 pi 10X Mg-free PCR buffer (Promega), 2ul 5mM dNTP, 2 units recombinant Taq polymerase, 1.5ul M g C ^ (25mM) and 10 picomoles of each primer in a final volume of 20ul. A l l reactions were performed using the MJ cycler as noted above. Three cycles of 94°C for lmin, 35°C for lmin and 72°C for 2min were first performed followed by 25 cycles of 94°C for 40 sec, 62°C for 40 sec and 72°C for 40 sec. PCR products were digested with appropriate restriction enzymes and purified by polyacrylamide gel electrophoresis. The purified fragment was ligated into digested and dephosphorylated pNEB193 vector (New England Biolabs) at room temperature overnight. 50% of each ligation was transformed into competent DH5a E. coli by the standard heat shock protocol (Bergmans, 1981). Colony PCR was performed on 10 white colonies per transformation to select for the correct size insert. PCR reactions were performed in a 20ul volume containing 2ul 10X Mg-free PCR buffer (Promega), l u l 5mM dNTP, 2 units recombinant Taq polymerase, 10 picomoles each of M l 3 -47 and —48 primers (New England Biolabs) and 1.6ul MgCb (25mM). Colony fingerprinting was performed on each colony containing correct size insert using a protocol previously described by Rajendra, 1991. Identical fingerprints among the patients of interest were sequenced using the CircumVent Thermal Cycle Dideoxy D N A sequencing kit (New England Biolabs) or by using the A B I Prism system. (See Appendix II for cloning strategy). 24 2.4 Localization of SMAL PCR clones to a chromosomal position A l l clones generated by SMAL-PCR were localized using either of two distinct methods: 1) Standard nucleotide-nucleotide B L A S T [blastn] search (Version 2.0) (Altschul, 1990; Gish, 1993). 2) Screen of the Bacterial Artificial Chromosome (BAC) RPCI-11 library using the S M A L clone (500ng/ul labeled by random priming -Gibco/BRL kit 18187-013) as probe to identify a specific B A C with homologous sequence to the clone. B A C library screening procedure is described at Briefly, the procedure is as follows: Wet filters in minimum volume of "Church Buffer" (5mg/ml Bovine Serum Albumin; .5mM EDTA; 500mM N a H P 0 4 pH 7.2; 7% SDS). Place filters in a hybridization bottle and add 25ml "Church Buffer" and seal. Pre-hybridize lhr at 65°C in hybridization oven with continual rotation. Add labeled probe(s) so that the total cpm of each probe is 1X10 6 to 11X107. Hybridize overnight (16-18 hrs) at 65°C. Blocking of all nonspecific binding of probes is accomplished by adding an equal volume of sonicated human placental D N A (lOmg/ml) to the probe solution and incubating one hour at 65° C before adding the probe to the hybridization bottle. Wash filters with Wash I solution (Wash I: .5mM EDTA; 80mM N a H P 0 4 pH 7.2; 1% SDS; 5mg/ml Bovine Serum Albumin) at 65°C. Repeat wash about 4 to 6 times until most of the non-bound probe is removed Wash filters with wash II solution (Wash II: ImM EDTA; 80mM NaHP0 4 ; .1% SDS) repeatedly at 65° C until the level of cpm (monitored with a Geiger counter) removed remains constant (no 25 longer decreases). Rinse filters in water and wrap individually in plastic wrap. Expose to x-ray film for 2 to 72 hrs at -80°C. The B A C clone identified was localized by Flourescent In Situ Hybridization (FISH) to a chromosomal position. The FISH experiments were performed by Jennifer Skaug at the Toronto Hospital for Sick Kids [ 2.5 Loss of Heterozygosity analysis Microsatellite markers were identified in the chromosome regions of interest. Primer sequences were obtained from the Genome Database []. (See Appendix III for complete list of L O H primers used). Primers were selected based on their chromosomal position and degree of heterozygosity. Prior to amplification, T4 polynucleotide kinase (New England BioLabs) was used to end-label lOOng of one primer from each pair with [y-32P]-ATP (20uCi). PCR amplification was carried out in 5ul reaction volumes containing 5ng of genomic D N A , lng of labeled primer, lOng of each unlabeled primer, 1.5mM each of dATP, dGTP, dCTP, and dTTP, 0.5 units of Taq D N A polymerase (GIBCO, BRL) , and PCR buffer [16.6mM ammonium sulfate, 67mM Tris (pH 8.8), 6.7mM magnesium chloride, lOmM B-mercaptoethanol, 6.7mM EDTA, and 0.9% dimethyl sulfoxide]. PCR amplification was performed for 40 cycles consisting of denaturation at 95°C for 30 sec, annealing at 50-60°C for 60 sec, and extension at 70°C for 60 sec with a final extension at 70°C for 5 min. The PCR products were separated on 8% urea-formamide-polyacrylamide gels and visualized by autoradiography. The gels were coded and L O H was scored without knowledge of sample diagnosis. For informative cases, allelic loss was scored when the signal intensity of one allele was decreased by least 50% in the D N A sample from the 26 lesion when compared to the same allele in the matching connective tissue D N A . A l l samples showing allelic loss were subjected to repeat analysis after a second independent amplification. Raw data was analyzed with the help of Dr. Miriam P. Rosin. C H A P T E R 3: R E S U L T S . 3.1 Sample Preparation and Reproducibility of Assay 3.1.1 Microdissection and DNA extraction from oral archival biopsies Genome scanning of archival microdissected specimens has two limitations: D N A quantity and quality. To determine the amount of D N A recoverable from archival lesions, 235 samples representing various pre-neoplastic and neoplastic stages were microdissected from sections of formalin-fixed paraffin-embedded tissue blocks. Manual or Laser Capture Microdissection and D N A extraction was performed on 149 dysplasias of varying degrees and 86 squamous cell carcinomas (SCC's). Manual Microdissection was performed in the majority of the cases and enabled gross isolation of cell populations of interest (Figure 5). In order to determine i f specific cell populations could be excised from more complex tissue pathology, Laser Capture Microdissection was performed on breast, prostate and oral tissue specimens (Figure 3). D N A yield is summarized in Figure 6. Although considerable heterogeneity of yield is observed among these samples, median values ranged from 100 to 300ng (Table 2). The minimum yields were 6 to 13ng, representing about 1- to 2- thousand cells. Conventional PCR-based approaches require 50 -lOOng of D N A to examine individual loci. Current genome fingerprinting approaches require the same amount of high molecular weight template D N A to analyze 27 5-30 loci. Therefore, a modified fingerprinting approach is necessary for genome scanning of archival specimens. 28 Figure 5. Microdissection of severe oral dysplasia specimen. (A) Section before dissection. (B) Section after dissection. The epithelial part is removed. Connective tissue underlying the sample is collected separately. 29 18 16 14 12 S 10 in B Hyperplasia 0Mild Dysplasia • Moderate Dysplasia HSevere Dysplasia/Carcinoma hSitu • hi I Ml 11 T- CN CM CO CO LO m co co co o> Oi T~ T - CM DNA Yield (ng) Figure 6. Summary of D N A quantity generated from 153 oral archival specimens (note: 153/235 total anlayzed). 30 Stage N Median yield (ng) Minimum yield (ng) Maximum yield (ng) Minimum yield (cells*) Mild dysplasia 75 165 6 3120 900 Moderate dysplasia 55 100 13 2125 1950 Severe dysplasia/c/s 59 255 12 1920 1800 Hyperplasia 66 326 8 5572 1200 All samples 235 192 6 5572 900 *Number of cells estimated from the minimum DNA yield with 7 picograms/cell Table 2. Quantity of D N A obtained from 235 microdissected oral premalignant lesions. 31 3.1.2 Development of genome-wide SMAL-PCR and reproducibility of method Method Development To increase the sensitivity of the assay, PCR parameters were optimized by adjusting reaction components, volume, cycle number and temperature. A panel of arbitrarily chosen decanucleotide primers (see Table 1 Section 2.2) were assayed alone and in combination empirically and in-silico to identify a subset which would generate approximately 50 S M A L products or greater (Figure 7). Figure 8 shows an example of a typical initial screen of seven primers alone and in combination by S M A L PCR. Finally, to increase detection limits, radiolabeled primers were utilized and PCR products were resolved on non-denaturing polyacrylamide gels designed for the desired size range (see Materials and Methods Section 2.2 ). We have determined that by selecting primer combinations that amplify a majority of fragments below 600bp, we can generate the highest number of reproducible fragments per reaction. Primer combinations 4, 10, and 15 in Table 3 have generated the highest number of fragments in this size range. This modified protocol was used to test i f D N A fingerprints could be generated from low quantities of archival DNA. D N A titration experiments showed that consistent fingerprinting patterns could be generated using 0.5ng of archival D N A or greater. (Figure 9). In all subsequent studies, 2ng (~ 300 cell equivalents) were used per reaction to ensure that sampling error was minimized. At these concentrations, dysplasia and 32 Coordinate Coordinate PCR Product (Start) (Finish) Size (in bp) 4704302 4704702 400 4706618 4708219 1601 5644529 5645920 1391 6252216 6253151 935 6403692 6405195 1503 6403881 6405195 1314 9296073 9297562 1489 9620187 9620579 392 10471428 10472116 688 12639002 12639144 142 12856776 12857964 1188 13821857 13821992 1135 14364881 14366419 1538 15201723 15202893 1170 15256047 15257293 1246 18352572 18353300 728 18352572 18353948 1376 20822017 20823657 1640 21427094 21428358 1264 24224001 24225660 1659 27827556 27828021 465 27830822 27830963 1141 31976479 1977326 847 Figure 7. Two arbitrary 8mers were selected arbitrarily. Product number and size predicted from complete sequence of Chromosome 22 (35 million base pairs). Product size range was restricted to 50-2000bp. Any genomic sequence may be interrogated using this program. Primers yielding largest number of products are used for S M A L -PCR analysis on archival specimens. Written by Dr. Steven Jones of the BCCA Genome Sequence Centre. 33 — — o. o o MS o — • - H ( S —I VO o o NO O —' <N es c — s ;s 5 Figure 8. 7 arbitrary decanucleotide primers tested with the S M A L -PCR protocol alone and in combination. Products run at 60 Volts for 1 hour on 1% agarose gel. Primers that generate a large number of fragments are used for scanning archival specimens. 34 Combination Primer 1 Primer 2 Products* 1 ggtccctgac caatcgccgt 32 2 ggtccctgac agccagcgaa 24 3 aggggtcttg gaccgcttgt 42 4 gaccgcttgt aggtgaccgt 56 5 agccagcgaa aggtgaccgt 19 6 aggggtcttg ggtccctgac 43 7 ggtccctgac gtgatcgcag 38 8 gggtaacgcc agccagcgaa 37 9 gggtaacgcc caatcgccgt 29 10 aatcgggctg gaaacgggtg 46 11 caaacgtcgg gttgcgatcc 45 12 cggcagcat ggctaccag 34 13 cggcagcat cgcatccag 31 14 ggctaccag cgcttggac 32 15 cgcttggac cgcatccag 31 16 gaatcagtg aactggcat 50 17 caggctatc gaatcagtg 41 18 gaatcagtg aactgcatc 45 19 aactggcat aactgcatc 34 20 caggctatc tccagctag 21 21 caggctatc ggtccctgac 36 Fragments obtained between 50-600 base pairs. Table 3. Number of Products Obtained Using Different Arbitrary Primer Combinations. 35 DNA (ng) M 4 2 1 .75 .5 .25 .12 .06.03 lOObp Figure 9. SMAL Fingerprint titration of a single archival oral biopsy. [DNA] > lng (100-150 cell equivalents yield consistent S M A L fingerprints. Therefore, -1000 loci can be analyzed using 40ng of archival DNA. This represents genome scanning at approximately 2-3centimorgan intervals. 36 normal connective tissue from the same individual yielded reproducible S M A L fingerprint patterns. However, fingerprints were different between individuals due to D N A polymorphisms (Figure 10). In control experiments, the primers alone (without D N A template) did not produce signals. Reproducibility of SMAL-PCR To determine the reproducibility of S M A L PCR using archival specimens, we compared fingerprints in archival samples from 10 patients in two independent PCR reactions using primer combination 10 (see Table 3 above). A total of 383 products were generated in the first trial and 384 products in the second. 376 of the products were common between trials, while 15 products were unique to one trial. Ten of these 14 fragments were greater than 400 bp (Table 4). Although these experiments show that random products are generated at low frequencies in S M A L PCR reactions, these events are not expected to impact on the utility of comparative S M A L fingerprinting as it is highly unlikely the same random event would occur across multiple patients. By cloning only those alterations which recur across multiple patients and by confirming that the alteration contains the same sequence in each patient, randomly derived PCR artifacts will be excluded. Furthermore, the gain or loss of a specific PCR product can be confirmed in repeated experiments. 37 N D N D N D N D F -300bp 200bp N = Normal archival epithelium D = Dysplasia archival epithelium F = Bulk frozen lung tissue - = No D N A control Figure 10. Fingerprints of 4 patients with premalignant lesions in which dysplastic epithelial cells and corresponding normal connective tissue are compared. Closed arrows indicate two examples of polymorphic loci. 38 Patient Number #of Products 1 2 3 4 5 6 7 8 9 10 Total Trial 1* 40 38 34 35 33 39 41 41 39 43 383 Trial 2* 40 39 33 33 36 41 43 38 39 42 389 Unique Products** 0 1 1 2 3 2 2 3 0 1 15 * Number of products between 50-600 bp generated using primer combination 10 in Table 2. **Number of products unique to one of the SMAL-PCR experiments Table 4. Reproducibility of S M A L products between two independent PCR reactions. 39 Application of SMAL-PCR across tissue types We next determined that our modified S M A L assay was capable of reproducible fingerprinting from multiple archival tumor types. The oral sample was manually microdissected, the prostate and breast samples were obtained by Laser Capture Microdissection and the lung tissue D N A was extracted from bulk frozen tissue. Figure 11 shows very similar D N A fingerprints are generated from all sources of D N A suggesting multiple tissues and extraction techniques will yield reliable fingerprints from S M A L PCR. This figure also demonstrates that L C M derived D N A is suitable for S M A L PCRanlaysis. 40 300bp 200bp lOObp Figure 11: SMAL-PCR fingerprints from multiple archival tumor types. The 5 specimens used in this experiment were derived from different patients. Formalin fixed paraffin embedded cells were manually microdissected from oral (OT), biopsies. Prostate (PT) and breast (BT) sections were microdissected using Laser Capture Microdissection. (L) indicates D N A extracted from a non-microdissected frozen lung tumor biopsy (F) by standard phenol/chloroform procedure. (-) indicates no D N A control lane. 41 3.2 Discovery of a Novel Region Involved in HNSCC Progression on Chromosome 7Q22 3.2.1 Identification of a recurring SMAL alteration in oral SCCs D N A samples extracted from tumor and normal cells microdissected from 86 SCCs were fingerprinted using S M A L and recurring signals between multiple tumors were identified. A gain of signal was observed in 3 tumor/normal pair-wise comparisons (Figure 12). The 150bp fragment was cloned from each patient and sequenced (See Appendix II for schematic of cloning strategy). The 3 sequences were verified to be the same by colony fingerprinting, which compares the G track from a dideoxy chain termination sequencing reaction. NCBI BLAST search against NR Genebank database yielded 1 contiguous sequence match of a 129 bp within B A C RG030L05 (AC005050.2) (Figure 13). This B A C was used to search the HUMANace -Human Genomic Physical Map Tracking Database at the Washington University Genome Sequencing Center ( bin/ace/ctc_choices/ctc.ace). Clone 7R030L05b was selected from this search and found within contig 25099 on chromosome 7 (Figure 14). Multiple sts's from this site mapped to 7q22-31 using Entrez Map Viewer ( 42 24 161 162 N T N T N T Willi*'*! v 'l.Jpil|)IW- $ipppB> Figure 12. Gain of a 150bp fragment from 3/20 archival HNSCC specimens. 43 Distribution of 8 Blast Hits on the Query Sequence Mouse-over to show defline and scores. Click to show alignments C o l o r K e y f o r R l i g n n e n t S c o r e s <40 40-50 t n p s e q _ l I i i i i | . . . i r 0 50 100 Sequences producing significant alignments: ablAC005050.2lAC005050 Homo sapiens clone RG030L05, complet... emblZ81364lHS179D3A Human DNA sequence from PAC 179D3, betw... dbi|AB022218.1JAB022218 Arabidopsis thaliana genomic DNA, c... emb[Yl6045|ATY16045 Arabidopsis thaliana G8B7T7 gene, exons... dbi1AP000154.1|AP000154 Homo sapiens genomic DNA, chromosom... emb[X82863lGGMlC68H4 G.gallus microsatellite DNA (68H4) dbi[AP000083.1|AP000083 Homo sapiens genomic DNA of 8pll.2 ... ab|U32694|U32694 Haemophilus influenzae Rd section 9 of 163... Score E (bits) Value 25& le -66 _11 0. 45 _ 2 £ 1. 8 _ 2 £ 1. 8 _ i i 1. 8 _ M 1. 8 _ 2 £ 1. 8 _ 3 i 1. 8 Figure 13. Blast search using cloned 150bp fragment localizes sequence to Bac clone RG030L05. 44 4 5 3.2.2 Verification of alteration at 7q22-31 In order to verify that this genomic region was altered in the 3 patients with the above S M A L alteration, microsatellite markers D7S1812, D7S2539 and AFM249za5 between 7q22-31 were evaluated for LOH/AI (See Table 1 for primer sequences). Figure 15 shows confirmation of genetic alteration in all 3 patients at D7S1812 and/or D7S2539. 3.2.3 Two discontiguous regions of AI/LOH at 7q22-31 A number of previously described genes have been identified between 7q22-31. MPP11 at 7q31 has been reported by Resto et al (2000) to be amplified in head and neck tumors. They defined a region of allelic imbalance (Al) between D7S3080 and D7S501. These primers are greater than 10Mb telomeric to D7S539. This raises the possibility of 2 regions of A l within 7q22-31. In order to select microsatellite markers between 7q22-31, we first identified 4 ordered contigs in this region using the Washington University Human B A C and Accession Maps (see Appendix TV and V for complete B A C tiling path spanning these regions). We selected a total of 9 primers in the 7q22-31 region from NCI Map Viewer database. However, the relative positions of these microsatellite markers on the contigs had to be verified. Sequence accession numbers for these markers were found. Sequence of all the markers were used in B L A S T to identify corresponding BACs in the HTGS database at NCBI. RPCI11 BACs found were matched against BACs in the Patient Patient Patient 24 161 162 N T N T N T ft n s 1 m — ff. D7S2539 D7S2539 D7S2539 Figure 15. Conformation of allelic imbalance in the three patients showing S M A L alteration at 7q22. 47 Washington University B A C Accession Map database (October 7 freeze; in order to identify BACs that overlap with the ones containing the microsatellite markers. The relative order of these BACs was confirmed on Ensembl database ( Relative distances between sts's within a given contig were confirmed on FPC database. STS's not correlated to BACs by B L A S T were confirmed by PCR verification using BACs found on FPC correlating to these markers. Each marker was determined to have a high degree of polymorphism using the Genome database ( Figurel6A shows the relative position of all markers, BACs and contigs identified in this region. 38 tumors were analyzed using these selected primers. 26 (68%) of samples showed AI /LOH for at least 1 of the 9 primers. 22 of 38 SCC (58%) had A l within D7S1812-D7S618. 8 of 20 (40%) informative SCC had A l at D7S501. 6 (30%) SCC had A l in the first region only but retained D7S501. 2 (10%) had A l at D7S501 only. Finally, 5 (25%) SCC showed A l at both regions, but had areas of retention in between the 2 regions of A l . This establishes the existence of 2 distinct regions of alteration at 7q22-31. Figure 16B highlights the most informative cases and Appendix VI shows all L O H performed for oral SCC. 3.2.4 Do these alterations occur in early stages? In order to determine whether or not this alteration occurs in the early stages of oral carcinogenesis, we analyzed dysplastic lesions from 34 patients (15 mild dys, 5 mod, 48 5 S t 5 - S S S £ ? ? S RG030L05b 5 cn cn x: cn cn if cn cn c n c n cn cn cn ctg14473 ctg14422 ctg7 ctg25099 65T n i n n n n n m 112T n i n n n n n m 117T • n n n 125T n n n n n m 161T n i n n n n n n 162T • mn • n m n • 386T m n n r n n n 399T • i n n n n m • • 402T n mmn n n n n n 416T m i n n n n n n m 439T n i n n n n n n m 464T • n n n n n • 482T • B l i n n 568T n mmn n n m n n Figure 16: A) Four contigs (ctg) spanning region between 7q22-31. Markers used for L O H indicated by "D7S...". RG030L05 is original B A C corresponding to B L A S T with 150bp S M A L clone. B) Subset of L O H analysis on 38 tumor specimens (T). Black boxes indicate allelic imbalance. White boxes indicate retention. Gray boxes indicate uninformative reactions. 49 14 severe/cis; in patients without a history of oral cancer) for LOH/AI at the same 9 loci. Alteration occurred in 20 of 34 cases (59%) on at least 1 of the loci. Eighteen of 34 informative cases (53%) showed an alteration in the first region at D7S1812-D7S618. Eight of 20 informative cases (40%) had LOH/AI at D7S501, in the second region. In 2 of 20 (10%) cases informative at both regions, AI /LOH was restricted to D7S1812-D7S618 and 1 of 20 (5%) cases to D7S501 only. In fact, 6 of 20 cases showed AI /LOH at both regions but with retention of loci between. 10 of 20 (50%) cases had retention of all informative loci in both regions and 1 case had loss of all informative loci all the way across. Figure 17B highlights the most informative cases and Appendix VII shows all L O H performed for oral dysplasia. These data show that there are 2 distinct regions of frequent alterations and that both occur in dysplastic stages of carcinogenesis. 3.2.5 Definition of boundaries for Al within D7S1812-D7S477 In 8 of 72 cases (5of 38 tumors, 112T, 161T, 402T, 24T and 4 of 34 dysplasias, 207D, 270D, 306D, 208D, allelic imbalance was observed at D7S2539 with retention at D7S1812, suggesting that the putative gene is telomeric to D7S1812. In 17 cases (9 dysplasias 207D, 270D, 497D, 55D, 564D, 58D, 8ID, 169D and 208D and 8 SCCs, 65T, 112T, 161T, 416T, 439T, 464T, 482T, and 24T), AI /LOH was observed at D7S2539 and retention at D7S479, suggesting that the putative gene is centromeric to D7S479. These data suggest a minimal region of alteration at D7S1812-D7S479. These markers span 5 overlapping BACs (RPCI11-524H16, RPCI11-350C21, RPCI11-440E5, CTB-57J11, RPCI11-610J and RPCI11-682N22). This represents approximately 500 kb. In 3 cases (24T, 208D and 112T), we observed a AI /LOH at D7S2539 and a retention at D7S1812, 50 _D7S1812 -T-D7S2539 L D7S479 —D7S491 CO LO (fl —D7S1558 —D7S1619 CO to CO CO °\3 _D7S2446 _D7S1530 RG030L05b 5 If n i M ctg14473 ctg14422 ctg7 ctg25099 55D • •nca n n ri n • 58D r a i n n n n n n • 81D • i n n n n n n 169D • i n n n n n • 2070 n i n n n n n • 208D n i n n n n n n 214D • fin n n n n 270D n i n n n n n 282D mn n n n • 306D n n n n n n 474D • nnn • n n • 497D • i n n n n • 564D • mnn n n • n n 298D n mmn n n n • Figure 17. A) Four contigs (ctg) spanning region between 7q22-31. Markers used for L O H indicated by "D7S...". RG030L05 is original B A C corresponding to B L A S T with 150bp S M A L clone. B) Subset of LOH analysis on 34 dysplasia specimens (D). Black boxes indicate allelic imbalance. White boxes indicate retention. Gray boxes indicate uninformative reactions. Patient 112T N T N T N T • * i _ i D7S1812 D7S2539 D7S479 Patient 208D N D N D N D H m 1 ' i i m§ I B 1 | • D7S1812 D7S2539 D7S479 Patient 24T N T N T N T • £ m -,jB< Jjjt J K • H p " • D7S1812 D7S2539 D7S491 Figure 18. L O H at D7S2539 and retention at D7S1812and D7S479 in tumor patients 112 and 24, and dysplasia patient 208. 52 suggesting that the gene lies between D7S1812 and D7S2539 (Figure 18). Since D7S2539 and D7S479 map to the same B A C clone (N0682N22) a telomeric boundary is defined within this B A C , a region that is 15 kB in size (by B L A S T alignment of primers for both microsatellite markers against sequence of the B A C clone). In order to identify candidate genes in these 2 regions, we have constructed a minimum overlapping tiling path of sequenced BACs that spans each of the regions (Figure 19). The contig shown in figure 19B is at 7q31 and contains the original S M A L fragment. The contig shown in figure 19A contains D7S1812 and D7S479, the minimal region of genetic alteration at 7q22. Candidate genes are indicated in both regions. 53 D 7 S 5 2 7 D 7 S 4 7 9 A C 0 2 2 2 6 1 I D 7 S 1 8 1 2 D 7 S 2 5 3 9 D 7 S 4 9 1 I I I I A C 0 0 7 3 1 6 N0095A10 N0781H09 N0524H16 7R057J11a N0682N22 N0094N07 N0525A11 N0656E16 N0727M20 N0440E05 ^10717117 7R300E22C N0327C24 N0350C21 •N0610J01 'N0038A06 III II I I min 97.1 D N C I l UNKNOUN UNKNOUN UNKNOWN 99.0 N032SF22 ^07531^02 M230SF09 N470C07 N0144I13 N0708P17 N0098C06 *N( 7R332P12b N0022N19 •N0569C14 S M A L I 7R030L05b D7S501 AC073420 'N0743A24 N0523G16 •N0593F03 7N0022N19aw 7N0017l10bw N0687D07 7R133P21C 107.5 Mb 110.0 Mb i in Q9NHE7 Q9NX19 SVPL STRSK2 LINK NOUN UNKNOMN UNKNOWN UNKNOUN D 7 S 2 1 0 8 E SBC I - B S C S P I K 3 C G Oe 0301 G PRKAR2B UNKNOUN Figure 19. Composite diagram of the two discontiguous regions between 7q22 and 7q31 with allelic imbalance. A l l B A C s indicated by "NO.. ." or " M . . . " , microsatellite markers by "D7S...", accession numbers by " A C . . . " . Region shown to scale in megabases (Mb). Known genes indicated in red text, unknown genes indicated in black text. S M A L fragment indicated with red text. Red arrow indicates B A C identified through B L A S T search. 54 3.3 Discovery of a Novel Region Involved in HNSCC Progression on Chromosome 2P23 3.3.1 Identification of a recurring SMAL alteration in oral premalignant lesions and SCCs The same set of archival premalignant and tumor specimens used yielded another common region of alteration in multiple patients using primer combination 4/7 (Table 1). Figure 20 shows a 150bp loss that was identified in 1/30 dysplasia specimens (71D) and 4/20 tumor specimens (62T, 114T, 122T, 402T). Each S M A L fragment was cloned and the sequences of these fragments were confirmed to be identical by colony fingerprinting. The single dysplasia clone (7ID) was used to screen a B A C RPCI-11 library. Four positive RPI11 B A C clones were identified: N0041M22, N0175E15, N0270P02 and N0285I12. FISH analysis localized N0175E15 to chromosomal position 2p23 (Figure 21). Micosatellite markers D2S2247, D2S2312, D2S170, D2S390, D2S375 and D2S400 surrounding N0175E15 were identified through Entrez Map Viewer ( In order to verify that this genomic region was altered in the 5 patients showing S M A L alterations, all six microsatellite markers chosen at this location were evaluated for LOH/AI. Figure 22B and 23B show confirmation of genetic alteration in patients 62, 114, 122 and 402 but not 71 (patient numbers shown in bold). Integration of information from the Washington University Human B A C Accession Map Database, University of California Santa Cruz Genome Browser ( (April 2001 version) and the Ensembl Database 55 N T N D r> ST W ~ • ' • zin Figure 20. Open arrow indicates 150 bp loss in 4/20 tumors (patient 62T shown) and 1/30 dysplasias (patient 71D shown). 56 Figure 21. FISH localization of B A C N0175E15 to 2p23. 57 D2S158 32436i D2S2350 32556 D2S2247 AFMb009xf1 33 AFMa305wh9 32111 stSG2766 33208 D2S2312 34600 WI-12589, 33559 D2S400 36F04 D2S170 34929 D2S390 35625 D2S2255 36624 D2S375 38353 27Mb 34Mb 35Mb 36Mb 2T n Contig 13214 n n rn n 43T n n n rn 59T1 n n n n n B2TD" n n n 80T n rn n 90T n n — n n n 110T • • • • 112T n n n n 113T n n n n 114T n rn n 116T n n _ rn n n 118T n n n n n 125T _l n n 199T n 161T n n n n 162T n 174T — n n n 175T n n m n 196T n n n n 200T1 n n n 202T1 n 211T — . 213T m n m n 215T i i n n n 366T0 385T2 n n — 386T n n n n n Figure 22. A) A single contig (ctg.13214) spans region of interest on chromosome 2p23. Markers used for L O H analysis indicated by "D2S...). Red star indicates position of original S M A L clone. B) L O H analysis of 49 oral SCC speciemens (T). Black boxes indicate allelic imbalance. White boxes indicate retention. Gray boxes indicate uninformative reactions. Patient numbers in bold showed original S M A L PCR alterations. 58 D2S2350 32656 •2S158 32436 AFMa305wh9 32111 D2S2247 AFMb009xf1 331BS S1SG2766 33208 •2S2312 34600 Wl-12589 33559 D2S170 349C9 D2S390 35525 D2S2255 36624 D2S375 B53 Contig 13214 B. 1«1T n n 303T n n 397T2 n _ l n n 398T n r n —i r n i i r n 405T? n n n n 414T5 n n n n 416T? 43(1T» n n n n 43?T n n 433T n n n n 434T n n n 435T1 n n n n 436T • n n 43BT. n n n n n 448TP n n n n 447T1 n — 451T1 n n n n 453T1 n n n n 24T • n n n n 309T7 n n 122T4 Figure 22 Continued 5 9 D2S2247 AFMbK)9xf1 D2S158 32436 D 2 S 2 3 5 0 33556 AFMa305wh9i 32111 I stSG2766 33108 D2S2312 34600 WI-12589 33559 D2S400 D2S170 34929 D2S390 35525 D2S225: 36624 D2S375 33353 45D —j I I n n rn 471D2 n 4 7 4 D ? - ? M1 497D • 4 D —i ™ 1 1 1 1 502D n 53D r n n 55D 564D1 n r - i R R D 1 9 n mi — 6 4 D • 66D n n 6 8 D 2 n n 60 r n n 71D n n n n •— 76D • n r n i i i i 81D11 • i n 82 D n n n n B3D1-1 n n 88D n n 92D n n 9 D 210D1-1 n n ?9am-i L_i . 169D — 214D • — n Figure 23. A) A single contig (ctg. 13214) spans region of interest on chromosome 2p23. Markers used for L O H analysis indicated by "D2S...". Red star indicates position of original S M A L clone. B) L O H analysis of 49 oral dysplasia speciemens (D) of varying grade. Black boxes indicate allelic imbalance. White boxes indicate retention. Gray boxes indicate uninformative reactions. Patient number in bold showed an original S M A L PCR alteration. 60 (, contig 13214 was found to contain all six markers at 2p23 chosen for AI /LOH analysis. Relative positions of the six markers chosen in this region were aligned to this contig. Positions were determined either by B L A S T alignment against corresponding B A C s from this tiling path or by PCR. Corresponding RPCI-11 B A C s found by B L A S T were correlated to their matching accession numbers using the Ensembl Database. Each marker was determined to have a high degree of polymorphism according to the Genome DataBase (Figure 24). 3.3.2 Fine mapping region of genetic alteration by AI/LOH at 2p23 50 HNSCC specimens were analyzed using these selected primers. 28 (56%) showed AI /LOH for at least 1/6 primers. 17 (34%) had A l restricted to within D2S2312 and D2S390. The region showing highest frequency of A l was D2S170. 15/37 (40%) patients tested at this locus showed A l (Figure 22B). In order to determine whether or not this alteration occurs in the early stages of oral carcinogenesis, we analyzed dysplastic lesions from 28 patients for LOH/AI at the same 9 loci. 7/28 (25%) showed A l for at least one marker. 10/19 (52%) showed retention for all markers tested (when > 2 markers were tested), and 3/19 (16%) showed A l across the entire region. This data shows 2p23 is a region of A l in premalignant oral specimens (Figure 23B). 6 1 AC009301 AC0I3403 D2S2247 S M A L 1 N0195B17 N01S8I13 N0041M22 N0373D23 D2S2312 D2S170 N0713D19 D2S390 D2S375 D2S400 N0328L16 N0056A01 N0026B05 N0023B13 N0062F14 N04J3M20 N1189A12 N0345G14 N0780J06 N0131B07 N0083M02 N0684O03 N0439P04 N0182E07 N0S17F01 M2024L07 I 1.23.2 P23.1 1 27.62 Mb -« 1 LQ9H8PS LGCKR LNOUEL LNOUEL LF0£L2 L«9H7C0 LQ9H0«1 LHBPRE3 LHPU17 LQ9UFJ9 L094864 L09MSH2 LNOUEL L0.9UHX7 L CAD L EIF2B4 L076004 LNOUEL LKHK L£LC3«fi3 LNRPL33 L NOUEL LNOUEL LNOUEL LNOUEL LQ9NXH2 LNRBP LNOUEL L«99674 LQ9H6D8 LPREB LSNX17 L£LC5B6 LNOUEL LS9Y6C2 LPPN1G L«9V5L7 L NOUEL _ , L NOUEL Ensembl Database LNOUEL L«9BYU2 Figure 24. Composite diagram of region on chromosome 2p23 showing allelelic imbalance. A l l B A C s indicated by "NO.. ." or " M . . . " , microsatellite markers by "D2S...", accession numbers by " A C . . . " . Region shown to scale in megabases (Mb). Known genes indicated in red text, unknown genes indicated in black text. S M A L fragment localized to B A C N0041M22 (indicated with red arrow) by FISH. 62 3.3.3 Definition of boundaries for Al within D2S2247 and D2S400 In 3/78 cases (2/50 rumors, 90T, 402T and 1/28 dysplasias, 58D12) allelic imbalance was observed at D2S170 with centromeric retention at D2S2312. In 5/78 cases (5/50 tumors, 90T, 113T, 116T, 118T, 402T2 and no dysplasias) allelic imbalance was observed at D2S170 with telomeric retention at D2S390. While further testing is needed to confirm this boundary of A l , this preliminary data suggests that a putative gene is telomeric to D2S2312 and centromeric to D2S390. Figure 24 shows all known and predicted genes within this boundary. 63 C H A P T E R 4. DISCUSSION 4.1 Benefits of SMAL PCR assay for genome-wide scanning Genome-wide analysis of tumors and their precursors will identify novel genetic alterations which are critical to cancer progression. However, such analysis requires microdissection of specific cell populations due to tissue heterogeneity. The limited amount of D N A recovered from microdissected cells is insufficient to support current methods for genome-wide searches. This thesis shows that S M A L can be used to achieve high-resolution genome scanning of minute quantities of DNA, even i f the D N A is derived from formalin-fixed paraffin-embedded specimens. We microdissected 235 epithelial premalignant lesions to determine the D N A available from such lesions. The median D N A yield for moderate dysplasia was 100 ng (Table 2). This amount can support approximately 50 PCR analyses, since we have set a 300 cell limit (2ng DNA) to avoid artefacts from sampling error. Using our chosen primers, 2ng D N A generates approximately 50 S M A L PCR fragments. Therefore, by selecting appropriate arbitrary primer combinations, over 2500 chromosomal sites (loci) can be tested for genetic alterations per typical lesion. Because these fragments are expected to be randomly distributed throughout the genome, the genome scanning resolution would be approximately 100 sites per chromosome (< 1 megabase intervals). In contrast, the conventional method of testing for loss of heterozygosity of selected microsatellite markers typically requires greater than 5ng per locus. Therefore, genome-wide S M A L would generate 100-fold more genetic information with the same material. 64 In addition to the high-resolution capacity of S M A L , a further advantage of this approach is the ability to quickly clone alterations, a feature that is crucial to the identification of novel genes. This cannot be easily achieved with other mapping strategies such as the cytogenetic approach, Comparative Genomic Hybridization (see introduction). As the human genome sequence database grows, chromosomal localization by experimental approaches will be replaced by simple database searches. This thesis describes the identification of two candidate regions implicated in the progression of HNSCC using this S M A L PCR fingerprinting strategy. One problem with clones obtained from S M A L PCR is the inability to derive the type of alteration which produced the clone. S M A L fingerprint comparisons merely show the gain or loss of a band in disease tissue D N A versus a normal D N A reference. The mechanism of underlying change is unknown. Because most S M A L fragment gains have been found to be contiguous sequence in normal human DNA, it is likely that the fragment represents an amplification. However, S M A L fingerprinting is also capable of identifying deletions. In this scenario, a deletion would bring two S M A L primer sequences close enough together for PCR amplification. This problem would always be resolved by expression analysis of genes in the region of the clone. A common criticism of fingerprinting assays such as S M A L PCR is their inability to pick up small genetic changes such as point mutations, thereby missing key genes involved in carcinogenesis. Because many different mechanisms lead to the disruption of gene function, however, it is unlikely that a given gene will always be inactivated by the same mutational event. While a given gene may be inactivated by a point mutation in some individuals, others may show genetic instability at this locus by another 65 mechanism. Further, the frequency of genetic alteration by point mutation is unknown. Therefore, it is unlikely that high-density fingerprinting will miss a gene of critical importance to carcinogenesis. 4.2 7q22-31 as a novel region of allelic imbalance in HNSCC After extensive L O H analysis, two distinct regions of A l were defined, one near 7q22 and the other near 7q31. The region on 7q31 has been previously reported by Resto et al (2000) to be frequently lost in HNSCC. This group suggested MPP11 as the putative oncogene in this region. A l at 7q22 has not been reported in the literature previously and is 8443Kbp centromeric to 7q31. Boundaries of allelic loss have been defined in this thesis for 7q22 to be between D7S1812 and D7S479 (approximately 300Kbp). The single locus D7S501 was used to confirm previously published data showing A l at 7q31, and fine mapping has not been accomplished to date (Figures 16-19). At 7q22, a high frequency of loss is seen both in S C C s (58%) and dysplasias (53%). This trend is also seen at 7q31, with A l at 40% of S C C s and 40% of dysplasias. This suggests an important change occurs somewhere in this region early in the development of HNSCC. By comparison, both 3p and 9p have been shown to be useful predictors for risk of progression. Loss at any place in these two regions has been reported to be 68% in SCC and 59% in dysplasia (Rosin, 2000). This data suggests 7q22-31 may provide a new region for testing predictive risk for HNSCC progression. 66 7q alterations have been observed in a variety of cancers and multiple critical loci on this arm have been implicated for myeloid neoplasms (Laing, 1998, Tosi, 1999), breast cancer (Zeng, 1999) and oral cancer (Wang, 1998). Recently, Zenklusen et al (2001) identified ST7 as a highly conserved tumor-suppressor gene on chromosome 7q31. 4.3 Identification of candidate genes in HNSCC at 7q31 In the most recent paper discussing A l at 7q31 in HNSCC, Resto et al presented microsatellite data showing A l spanning from D7S3080 to D7S496 (See Appendix V for physical map of this region). Our data supports a centromeric boundary for the first region at D7S2446 (which lies telomeric to D7S3080), further refining the region of alteration to D7S2446-D7S496. Ensembl annotates 10 known genes within this region and UCSC reports 11. Nine of these genes are present in both databases. FLJ11785 and GTC90 are present at UCSC but not Ensembl. Q9ULK2 is present only in Ensembl. It is important to note that while Resto et al reports MPP11 to be the candidate oncogene near the marker D7S496, MPP11 lies 5Mb downstream of D7S496. This was confirmed by B L A S T alignment of the marker sequence to the B A C containing MPP11 sequence. Therefore, the A l reported to be at 104.1Mb from the centromere, is actually at 109.1Mb. A l at D7S496 reported by Resto et al corresponds to our findings at D7S501, 700Kbp centromeric to this marker. D7S501 corresponds to B A C RG030L05b, the B A C containing our original S M A L clone (See Appendix VIII for detailed physical map of this region). 67 Of the 10 known or predicted genes in this region between D7S2446 and D7S496, Phosphatidylinositol 3-Kinase, Catalytic Gamma Subunit (PIK3CG, p i 10 gamma) (Stoyanov, 1995) is the only gene found altered in cancer. Sasaki et al showed PllO-gamma protein expression is lost in primary colorectal adenocarcinomas from patients and in colon cancer cell lines. Overexpression of wildtype or kinase-dead p i 00-gamma in human colon cancer cell with mutations of the tumor suppressor A P C and p53, or the oncogenes beta-catenin and Ki-ras, suppressed tumorigenesis. Therefore, p i 10-gamma has been shown to block the growth of human colon cancer cells. The 85-kD subunit of PIK (p85 alpha) lacks PI3-kinase activity and acts as an adaptor, coupling p i 10 to activated protein tyrosine kinases. P85 alpha is frequently increased in copy number in ovarian cancers, and this increase has been associated with increased p i 10 expression. Increased expression of p85 has also been strongly correlated to cervical carcinoma. Recently, Weng et al (2001) showed that the tumor suppressor PTEN induces apoptosis and cell cycle arrest through phosphoinositol-3-kinase/Akt-dependent and -independent pathways. These findings suggest p85 alpha is a candidate oncogene (Online Mendelian Inheritance of Man [OMIM] reference #171834). 4.4 Identification of candidate genes in HNSCC at 7q22 While 7q22 alterations have not previously been fine mapped in HNSCC, our data has defined a relatively small boundary of allelic imbalance in this region. Only one gene candidate can be localized to the region between D7S1812 and D7S479 using both Ensembl and UCSC Human Browser databases. Cytoplasmic Dynein Intermediate Chain 68 1 (OMIM #603772) is a large multisubunit microtubule-based motor protein composed of heavy, intermediate, light-intermediate and light chains. The DNCI1 gene contains 17 exons and spans approximately 160 Kbp (Crackower, 1999). The only other gene in this region is Solute Carrier Family 25, Member 13 (SLC25A13; O M I M #603859), which maps immediately centromeric to D7S1812. This gene is a calcium dependent mitochondrial solute transporter with a role in urea cycle function and has not been implicated in carcinogenesis (Kobayashi, 1999). 4.5 Identification of candidate genes in HNSCC at 2p23 While more extensive fine mapping need to be conducted in this region, data from this study shows a boundary of A l defined within D2S2312 and D2S390. Overall, 56% of HNSCC and 25% of dysplasias showed A l at one or more loci tested (Figures 22-23). High frequency A l at 2p23 in premalignant specimens suggests that a genetic alteration in this region occurs in the early stages of HNSCC. 2p23 may therefore provide a new region for predicting risk of progression from premalignant specimens. The region within the defined boundary on 2p23 is gene rich, with 39 known or predicted open reading frames (ORF's) presently in the literature ( Of these 39, however, only 9 are genes of known function. Retinitis Pigmentosa is involved in X-linked form of choroidoretinal degeneration. Fructosuria produces a benign, asymptomatic defect of intermediary metabolism. Prolactin Regulatory Element-Binding Protein, when translocated with TCF3 t(l;19)(q23;pl3.3), is involved in pre-B cell leukemia and acute myelogenous 69 leukemia. Solute Carrier Family 5, Member 6 (SLC5A6) is a sodium-dependent vitamin transporter. Glomerulosclerosis (MPV17) has been shown to be involved in kidney disease. Transcarbamolyase/Dihydroorotase (CAD) encodes a trifunctional protein which is associated with the enzymatic activity of the first 3 enzymes in the 6-step pathway of pyrimidine biosynthesis. PPM1G is a magnesium or manganese dependent protein phosphatase. Fos-like antigen 2 (Fosl2/Fra2) has been implicated as a regulator of cell proliferation, differentiation, apoptosis and transformation. Anaplastic Lymphoma Kinase (ALK) is frequently associated with t(2;5)(p23;q35) translocation, creating a N P M - A L K fusion gene. This translocation product has been implicated in lymphoid neoplasms (see O M I M for gene descriptions and references). Of these nine genes, only A L K and FRA2 fall within D2S390 and D2S2312. This suggests only one or both of these genes within this region may play a crucial role in HNSCC development. Recently, it has been shown that A L K has multiple fusion partners, including N P M , ATIC, TFG, TPM3 and A L T C L . Further, by association with these other genes, the A L K translocation product has recently been seen in a variety of tissues including myofibroblastic tumors, and abdomen, mesentery, liver, bladder, mediastinum, lung and bone tissues of children and young adults (Cheuk, 2001). Finally, A L K / N P M fusion product has been shown to activate phosphatidylinositol 3-Kinase, and its downstream effector, serine/threonine kinase (Akt) (Slupianek, 2001). Deregulation of the PI3K signalling pathway has been shown to disrupt both cell survival and cell death, and has been implicated in a variety of cancers (Toker, 2000). The Fos gene family consists of 4 members: FOS, FOSB, FRA1, and FRA2. These genes encode leucine zipper proteins that can dimerize with proteins of the JUN family, thereby 70 forming the transcription factor complex AP-1. As such, the FOS proteins have been implicated as regulators of cell proliferation, differentiation, and transformation, van Dam et al (2001) recently showed c-Jun:Fos dimers exhibit specific functions during Jun-induced transformation of chicken embryo fibroblasts. Further, enhanced levels of Jun:Fra2 specifically triggered the pathway leading to anchorage-independent growth. 4.6 Future directions Eight additional clones mapping to unique genomic positions have been isolated from S M A L fingerprinting of oral premalignant specimens and tumors. These include: lp24.3 (contig 4865), 2p22 (contig 13214), 5pl3.2 (contig 12808), 5q23.2 (contig 16556), 1 Op 11.21 (contig 132657), lOql 1.21 (contig 13419), 19pl3.11 (contig20904), and 20ql3.13 (contig 35683). In each of these novel regions, fine mapping will define boundaries of allelic imbalance, inside of which putative oncogenes and tumor suppressor genes will be identified. There are a number of suitable methods for analyzing expression of novel candidate genes in HNSCC. As S M A L PCR fingerprinting has rapidly identified numerous candidate regions, one must choose assays capable of rapid analysis of multiple gene candidates from numerous patient specimens. Real-Time PCR offers a rapid and quantitative tool for analyzing RNA. It is a highly sensitive technique, allowing analysis of gene expression from a very small amount of RNA. Further, it is designed to be conducted on a large number of samples, and will permit the rapid screening of a number of genes. This sensitivity and high-throughput capacity make it 71 the obvious choice for analysis over Northern or Dot Blot Hybridization. Further, both hybridization techniques require a much greater concentration of R N A than is currently available from the frozen oral biopsies in our current archives. These benefits make RT-PCR an ideal tool for analyzing the multiple genes identified by S M A L PCR (Freeman, 1999). In many situations, however, fresh frozen material is not available from a large enough set of patients for accurate examination of expression using Real-Time PCR. Often, material is limited to archival biopsy slides remaining from clinical diagnosis. R N A cannot be extracted from such slides as they have been fixed and are often many years old. In these circumstances, immunohistochemistry may be the best choice for analyzing gene expression. Unfortunately, one is limited by this assay to the use of antibodies that are capable of staining archival, formalin fixed slides. A further limitation is that only genes to which antibodies have been produced can be analyzed. 4.7 Bacterial artificial chromosome comparative genomic hybridization (BAC-CGH): The future of fine mapping allelic imbalance Defining boundaries of allelic imbalance by L O H analysis has had great success. It is one of few assays capable of utilizing archival material, and provides information on genetic instability at specific loci throughout the genome. Unfortunately, fine mapping a region of allelic imbalance in this way requires a large set of informative microsatellite primers and a large patient set. Thousands of individual PCR reactions must be run to produce an accurate boundary of A l at a given region. This is time consuming, labour intensive and costly. Further, due to the minute quantities of D N A typically extracted 72 from archival biopsies, often entire specimens will be sacrificed in order to fine map a single region. Currently, a novel method for high-throughput locus by locus allelotyping is being developed tentatively named B A C Array CGH. In this method, total genomic D N A from a tumor and a normal cell population are labelled with different fluorochromes and hybridized to arrayed B A C clones fixed to a solid nylon membrane support. The ratio of the fluorescence intensities on each spot in the array is then proportional to the copy number of the corresponding sequences in the tumor versus its normal counterpart. The B A C s used as template are sequenced and mapped. This allows one to construct a B A C array spanning virtually any two points in the human genome. The major advantage of array C G H over current techniques for analyzing gene copy differences between two samples is its ability to asses allelic imbalance at hundreds to thousands of loci in a single experiment. A single hybridization would be analogous to performing many hundreds of L O H PCR's. This high-throughput capacity would allow the most efficient use of the limited patient D N A extracted from archival specimens. Current reports provide compelling evidence that this technique is capable of detecting two-fold differences in copy number and higher. This would allow for detection of a single chromosomal copy deletion (Pollack, 1999; Bruder, 2001; Albertson, 2000). Finally, in May of this year, Daigo et al (2001) reported an array C G H protocol that is capable of measuring increased copy number using formalin-fixed, paraffin embedded, archival material. This protocol pre-amplifies a minute quantity of archival material by degenerate oligonucleotide primed (DOP)-PCR prior to labelling (Daigo, 2001). B A C 73 Array C G H , therefore, will provide a more efficient method for fine mapping the eight novel regions identified. 74 C H A P T E R 5. C O N C L U S I O N S Data from this thesis can be summarized in the following key points: 1) Median yield of D N A from formalin-fixed paraffin-embedded archival material is between 100-300 nanograms. 2) Using selected ten base pair primer combinations, S M A L PCR provides a high-density and reproducible method of fingerprinting minute quantities of D N A from archival specimens. 3) S M A L PCR is capable of detecting genetic alterations in tumor and premalignant HNSCC archival specimens. 4) Using S M A L PCR, regions on chromosomes 7q22-31 and 2p23 showed genetic instability in oral SCC and premalignant specimens. 5) 7q22-31 showed allelic imbalance by L O H in 68% of tumors (26/38) and 59% of dysplasias (20/34). 6) Two discontiguous regions of allelic imbalance were defined by L O H between 7q22-31: a) between D7S1812 A N D D7S479 (300Kbp) and b) at D7S501. 7) 2p23 showed allelic imbalance by L O H in 56% of tumors (28/50) and 25% of dysplasias (17/28). 8) A boundary of allelic imbalance was defined by L O H to be between D2S2312 and D2S390. 9) Alterations on 7q22-31 and 2p23 are found early in the progression of oral SCC, and at greater frequencies in fully malignant cell populations 10) 7q22-31 and 2p23 may provide new regions for testing predictive risk in HNSCC progression. 75 R E F E R E N C E S . Agochiya M , Brunton V G , Owens DW, Parkinson EK, Paraskeva C, Keith NW, Frame M C . Increased dosage and amplification of the focal adhesion kinase gene in human cancer cells. Oncogene 1999 18 5646-5653. Albertson DG, Ylstra B, Segraves R, Collins C, Dairkee SH, Kowbel D, Kuo W, Gray JW, Pinkel D. Quantitative mapping of amplicon structure by array C G H identifies CYP24 as a candidate oncogene. Nature Genetics vol 25 June 2000. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment search tool. J. Mol. Biol. 215:403-410. 1990. Bergmans HE, van Die IM, Hoekstra WP. Transformation in Escherichia coli: stages in the process. JBacteriol 1981 May;146(2):564-570. Bockmuhl U , Gunter w, Schmidt S, Schwendel A , Jahnke V , Dietel M , Petersen I. Genomic alterations associated with malignancy in head and neck cancer. Head and Neck 1998 March 145-151. Boring CC, Squire TS, Tong T. Cancer Statistics. C A Cancer J Clin. 42: 18-381, 1992. Bruder CEG, Hirvela C, Tapia-Paez I, Fansson I, Segraves R, Hamilton G, Zhang X X , Evans DG, Wallace A J , Baser M E , Zucman-Rossi J, Hergersberg M , Boltshauser E, Papi L, Rouleau GA, Poptodorov G, Jordanova A , Rask-andersen H, Kluwe L, Mautner V , Sainio M , Hung G, Mathiesen T, Moller C, Pulst SM, Harder H, Heiberg A , Honda M , Niimura M , Sigrid S, Blennow E,^  Albertson DG, Pinkel D, Dumanski JP. High resolution deletion analysis of constitutional D N A from neurofibromatosis type 2 (NF2) patients using microarray-CGH. Human Molecular Genetics 2001, Vol.10 No.3 271-282. Califano J, van der Riet P, Westra W, Nawroz H, dayman G, Piantadosi S, Corio R, Lee D, Greenberg B, Koch W, Sidransky D. Genetic progression model for head and neck cancer: implications for field cancerization. Cancer Res 1996 Jun l;56(ll):2488-92. Califano J, Westra W H , Meininger G, Corio R, Koch W M , Sidransky D. Genetic progression and clonal relationship of recurrent premalignant head and neck lesions. Clin Cancer Res 2000 Feb;6(2):347-352. Chan CSA, To K F , Lo K W , Mak KF, Pak W, Chiu B, Tse G M K , Ding M , L i X , Lee JCK, Huang DP. High Frequency of Chromosome 3p deletion in Histologically Normal Nasopharyngeal Epithelia from Southern Chinese. Cancer Research 60, 5365-5370, October 1,2000. Cheuk W, Chan JK. Timely topic: anaplastic lymphoma kinase (ALK) spreads its influence. Pathology 2001 Feb;33(l):7-12. 76 Crackower, M . A. ; Sinasac, D. S.; Xia, J.; Motoyama, J.; Prochazka, M . ; Rommens, J. M . ; Scherer, S. W.; Tsui, L . - C . : Cloning and characterization of two cytoplasmic dynein intermediate chain genes in mouse and human. Genomics 55: 257-267, 1999.) Cushwa WT, Dodds K G , Crawford A M , Medrano JF. (1996). Identification and genetic mapping of random amplified polymorphic D N A (RAPD) markers to the sheep genome. Mamm Genome 7:580-5. Daigo Y , Chin S, Gorringe K L , Bobrow L G , Ponder B A J , Pharoah PDP, Caldas C. Degenerate Oligonucleotide Primer-Polymerase Chain Reaction-Based Array Comparative Genomic Hybridization for Extensive Amplicon Profiling of Breast Cancers. A New Approach for the Molecular Analysis of Paraffin-Embedded Cancer Tissue. American Journal of Pathology 2001, 158:1623-1631. De la Guardia C, Casiano CA, Trinidad-Pinedo J, Baiz A . CENP-F amplification and overexpression in head and neck squamous cell carcinomas. Head and Neck 2001 Feb 104-112. Dib C, Faure S, Fizames C, Samson D, Drouot N , Vignal A , Millasseau P, Marc S, Hazan J, Seboun E, Lathrop M , Gyapay G, Morissette J, Weissenbach J. A comprehensive genetic map of the human genome based on 5,264 microsatellites. Nature 1996 Mar 14;380(6570):152-154. Dil-Afroze Misra A , Sulaiman IM, Sinha S, Sarkar C, Mahapatra A K , Hasnain SE (1998) Genetic alterations in brain tumors identified by RAPD analysis. Gene 206:45-8 Fearon ER, Vogelstein B. A genetic model for colorectal tumorigenesis. Cell 1990 Jun l;61(5):759-67. Fischer SG, Lerman jLS. Length-independent separation of D N A restriction fragments in two-dimensional gel electrophoresis. Cell 16:191-200 1979. Forozan F, Karh R, Kononen J, Kallioniemi A , Kallioniemi O.P. (1997). Genome screening by comparative genomic hybridization. Trends Genet 13:405-9 Freeman W M , Walker SJ, Vrana EK. Quantitative RT-PCR: Pitfalls and Potential. Biotechniques. 26:112-125 Jan 1999. Gasparotto D, Vukosavljevic T, Piccinin S, Barzan L, Sulfaro S, Armellin M , Boiocchi M , Maestro R. Loss of heterozygosity at lOq in tumors of the upper respiratory tract is associated with poor prognosis. Int J Cancer 1999 Aug 20;84(4):432-436. Gish W, States DJ. Identification of protein coding regions by database similarity search. Nature Genetics 1993. 3:266-272. 77 Gollin, SM. Chromosomal alterations in squamous cell carcinomas of the head and neck: window to the biology of disease. Head and Neck 2001 Mar;23(3):238-53. Grandis JR, Melhem MF, Gooding WE, Day R, Hoist V A , Wagener M M . Levels of TGF-alpha and EGFR protein in head and neck squamous cell carcinoma and patient survival. J Natl. Cancer Inst 1998 90:824-32. Zenklusen JC, Conti CJ, Green ED. Mutational and functional analyses reveal that ST7 is a highly conserved tumor-suppressor gene on human chromosome 7q31. Nat Genet 2001 Apr;27(4):392-398. Gupta V K , Schmidt AP, Pashia M E , Sunwoo JB, Scholnick SB. Multiple regions of deletion on chromosome arm 13q in head and neck squamous cell carcinoma. Int J Cancer 84, 453-457 1999. Hainaut P, Hernandez T, Robinson A , Rodriquez-Tome P, Flores T, Hollstain M . IARC database of TP53 gene mutation in human tumours and cell lines: updated compilation, revised formats and new visualization tools. Nucleic Acid Research 1998 26:205-13. Hovig E, Mullaart E, Borresen A L , Uitterlinden A G , Vijg J. Genome scanning of human breast carcinomas using micro-and minisatellite core probes. Genomics 17:66-75 1993. Howe JR, Klimstra DS, Cordon-Cardo C. D N A extraction from paraffin-embedded tissues using a salting-out procedure: a reliable method for PCR amplification of archival material. HistolHistopathol 1997 12:595-601. Hu L, Crowe DL , Rheinwald JG, Chambon P, Gudas LJ . Abnormal expression of retinoic acid receptors and keratin 19 by human oral and epidermal squamous cell carcinoma cell lines. Cancer Research 1991. 51:3972-81. Ionov Y , Peinado M A , Malkhosyan S, Shibata D, Perucho M . Ubiquitous somatic mutations in simple repeated sequences reveal a new mechanism for colonic carcinogenesis. Nature 363:558-561 1993. Ishwad CS, Shuster M , Bockmuhl U , Thakker N , Shah P, Toomes C, Dixon M , Ferrell RE, Gollin S M . Frequent allelic loss and homozygous deletion in chromosome band 8p23 in oral cancer. Int J cancer. 80, 25-31 (1999). Jefferies S, Foulkes WD. Genetic mechanisms in squamous cell carcinoma of the head and neck. Oral Oncology 37 2001 115-126. Kallioniemi OP, Kallioniemi A, Sudar D, Rutovitz D, Gray JW, Waldman F, Pinkel D. Comparative genomic hybridization: a rapid new method for detecting and mapping D N A amplification in tumors. Semin Cancer Biol 1993 Feb;4(l):41-46. 78 Kobayashi, K. ; Sinasac, D. S.; Iijima, M . ; Boright, A . P.; Begum, L. ; Lee, J. R.; Yasuda, T.; Ikeda, S.; Hirano, R.; Terazono, H. ; Crackower, M . A. ; Kondo, I.; Tsui, L.-C. ; Scherer, S. W.; Saheki, T. : The gene mutated in adult-onset type II citrullinaemia encodes a putative mitochondrial carrier protein. Nature Genet. 22: 159-163, 1999). Liang H , Fairman J, Claxton DF, Nowell PC, Green ED, Nagarajan L. Molecular anatomy of chromosome 7q deletions in myeloid neoplasms: evidence for multiple critical loci. Proc Natl Acad Sci USA 1998 Mar 31;95(7):3781-5. Lee, J M , Bernstein A . p53 mutations increase resistance to ionizing radiation. Proc. Nat. Acad. Sci. 90: 5742-5746,1993. Lippman SM, Hong WK. Molecular markers of the risk of oral cancer. The New England Journal of Medicine. April 26, 2001. Vol . 344, No. 17. Luo L, Salunga RC, Guo H, Bittner A , Joy K C , Galinda JE, Xiao H, Rogers K E , Wan JS. Gene expression profiles of laser-captured adjacent neuronal subtypes. Nature Medicine vol 5 No. 1 Jan 1999. Lydiatt W M , Davidson BJ, Schantz SP, Caruana S, Chaganti RSK. 9p21 deletion correlates with recurrence in head and neck cancer. Head and Neck March 1998 113-118. Maeda T, Jikko A , Hiranuma H, Fuchihata H. (1999). Analysis of genomic instability in squamous cell carcinoma of the head and neck using the random amplified polymorphic D N A method. Cancer Lett 138:183-8 Malkhosyan S, Yasuda J, Soto JL, Sekiy T, Yokota J, Perucho M . (1998) Molecular karyotype (amplotype) of metastatic colorectal cancer by unbiased arbitrarily primed PCR D N A fingerprinting. Proc Natl Acad Sci USA 95: 10170-5. Matsumura K , Iritani A , Enomoto S, Torikata C, Matsuyama S, Kurita A , Kurahashi H, Tsuchida N . Defining a common region of D N A amplification at 22ql 1.2-12 in head and neck squamous cell carcinomas by quantitative FISH analysis. Genes, Chromosomes and Cancer29:207-212 2000. Michalides R, van Veelan N , Hart A , Loftus B M , Hilgers FJ, Balm AJ. Overexpression of cyclin D l correlates with recurrence in 47 operable squamous cell carcinomas of the head and neck. Cancer Research. 1995 55:975-8. Miracca E, Kowalski L, Nagai M . High prevalence of pi6 genetic alterations in head and neck tumours. Br J Cancer 1999 81:677-83. Murray JC, Buetow K H , Weber JL, Ludwigsen S, Scherpbier-Heddema T, Manion F, Quillen J, Sheffield V C , Sunden S, Duyk G M , et al. A comprehensive human linkage map with centimorgan density. Cooperative Human Linkage Center (CHLC). Science 1994 Sep 30;265(5181):2049-2054. 79 Motokura T, Bloom T, Kim HG, JuppneRH, Ruderman JV, Kronenberg H M , Arnold A . A novel cyclin encoded by a bell-linked candidate oncogene. Nature 350: 512-515, 1991. Ogawara K, Miyakawa a, Shiiba M , Uzawa K Watanabe T, Wang X L , Sata T, Kubosawa H, Kondo Y , Tanzawa H. Allelic loss of chromosome 13ql4.3 in human oral cancer: correlation with lymph node metastasis. Int J Cancer 79 312-317 1998. Ohtsuka M , Makino S, Yoda K, Wada H, Naruse K , Mitani H , Shima A , Ozato K , Kimura M , Inoko H . (1999). Construction of a linkage map of the medaka (Oryzias latipes) and mapping of the Da mutant locus defective in dorsoventral patterning. Genome Res 9:1277-87. Ohta M , Inoue H , Cotticelli M G , Kastury K , Baffa R, Palazzo J, Siprashvili Z, Mori M , McCu P, Druck T, Croce C M , Huebner K . The FHIT gene, spanning the chromosome 3pl4.2 fragile site and renal carcinoma-associated t(3;8) breakpoint, is abnormal in digestive tract cancers. Cell 84: 587-597, 1996. Ong T M , Song B, Qian HW, Wu ZL , Whong WZ. (1998). Detection of genomic instability in lung cancer tissues by random amplified polymorphic D N A analysis. Carcinogenesis 19:233-5. Partridge M , Emilion G, Falworth M , A'Hern R, Phillips E, Pateromichelakis S, Langdon J. Patient-specific mutation databases for oral cancer. IntJCance: 84 284-292 1999. Partridge M , Pateromichelakis S, Phillips E, Emilion GG, A'Hern RP, Langdon JD. A case-control study confirms that microsatellite assay can identify patients at risk of developing oral squamous cell carcinoma within a field of cancerization. Cancer Research 60, 3893-3898 July 15, 2000. Pearlstein RP, Benninger MS, Carey TE, Zarbo RJ, Torres F X , Rybicki B A , Van Dyke DL. Loss of 18q predicts poor survival of patients with squamous cell carcinoma of the head and neck. Genes, Chromosomes and Cancer 21:333-339 1998. Perucho M , Welsh J, Peinado M A , Ionov Y , McClelland M . (1995). Fingerprinting of D N A and R N A by arbitrarily primed polymerase chain reaction: applications in cancer research. Methods Enzymol 254:275-90. Pollack JR, Perou C M , Alizadeh A A , Eisen M B , Pergamenschikov A , William CF, Jeffrey SS, Botstein D, Brown PO (1999). Genome-wide analysis of D N A copy-number changes using cDNA microarrays. Nat Genet 23:41-6. 80 Postlethwait JH, Johnson S, Midson C N , Talbot WS, Gates M , Ballinger EW, Africa D, Andrews R, Carl T, Eisen JS. (1994). A genetic linkage map for the zebrafish. Science 264:699-703. Quon, H. , Liu, FF, Cummings, BJ. Potential molecular prognostic markers in head and neck squamous cell carcinomas. Head andNeckFeb. 2001 147-159. Rajendra B. Linear amplification D N A sequencing directly from single phage plaques and bacterial colonies. Nucleic Acids Research Vol 19, No 5. 1153). Redon R, Muller D, Caulee K, Wanherdrick K, Abecassis J, du Manior S. A simple specific pattern of chromosomal aberrations at early stages of head and neck squamous cell carcinomas: PIK3CA but not p63 gene as a likely target of 3q26-ter gains. Cancer Research 61,4122-4129 May 15, 2001. Reed A , Califano J, Caims P, Westra W, Jones R, Koch W, et al. High frequency of p i 6 inactivation in head and neck squamous cell carcinoma. Cancer Research 1996 56:3630-3. Resto V A , Caballero OL, Buta MR, Westra WH, Wu L, Westendorf JM, Jen J, Hieter P, Sidransky D. A putative oncogenic role for MPP11 in head and neck squamous cell cancer. Cancer Research 60 5529-5535 October 1, 2000. Rosin MP, Cheng X , Poh C, Lam W L Huang Y , Lovas J, Berean K, Epstein J, Priddy R, Le ND, Zhang L. Use of allelic loss to predict malignant risk for low-grade oral epithelial dysplasia. Clinical Cancer Research 6: 357-362 Feb 2000. Sasaki, T.; Irie-Sasaki, J.; Horie, Y . ; Bachmaier, K. ; Fata, J. E.; L i , M . ; Suzuki, A . ; Bouchard, D.; Ho, A. ; Redston, M . ; Gallinger, S.; Khokha, R.; Mak, T. W.; Hawkins, P. T.; Stephens, L. ; Scherer, S. W.; Tsao, M . ; Penninger, J. M . Colorectal carcinomas in mice lacking the catalytic subunit of PI(3)K-gamma. Nature 406: 897-902, 2000. Serrano M , Hannon G, Beach D. A new regulatory motif in eel cycle control causing specific inhibition of cyclin D/CDK4. Nature 1993 366:704-7. Shin D M , Lee JS, Lippman SM, Lee JJ, Tu ZN, Choi G. p53 expressions: predicting recurrence and second primary tumors in head and neck squamous cell carcinoma J Natl Cancer Inst 1996 88:519-29. Silverman S, Jr. Oral Cancer (3rd ed). Atlanta, GA. Am Cancer Soc. 1990. Simone N L , Vonner RF, Gillespie JW, Emmert-Buck MR, Liotta L A . Laser-capture microdissection: opening the microscopic frontier to molecular analysis. TIGS July 1998 Vol.4 No.7. 81 Slupianek A , Nieborowska-Skorska M , Hoser G, Morrione A, Majewski M , Xue L, Morris SW, Wasik M A , Skorski T. Role of Phosphatidylinositol 3-Kinase-Akt Pathway in Nucleophosmin/Anaplastic Lymphoma Kinase-mediated Lymphomagenesis. Cancer Research 61,2194-2199 Marchl,2001. Sozzi G, Veronese M , Negrin M , Baffa R, Cotticelli M G , Inoue H , Tornielli S, Pilotti S, De Gregorio L, Pastorino U , Pierotti M A , Ohta M , Huebner K , Croce CM.The FHIT gene at 3pl4.2 is abnormal in lung cancer. Cell 85: 17-26, 1996. Stoyanov, B.; Volinia, S.; Hanck, T.; Rubio, I.; Loubtchenkov, M . ; Malek, D.; Stoyanova, S.; Vanhaesebroeck, B.; Dhand, R.; Nurnberg, B.; Gierschik, P.; Seedorf, K. ; Hsuan, J. J.; Waterfield, M . D.; Wetzker, R. Cloning and characterization of a G protein-activated human phosphoinositide-3 kinase. Science 269: 690-693, 1995. Tannock IF, Hi l l RP, editors. The basic science of oncology. New York: McGraw-Hill 1998. Toker A , Newton A. Cellular Signaling: Pivoting around PDK-1. Cell Vol . 103, 185-188. October 13, 2000. Tosi S, Scherer SW, Giudici G, Czepulkowski B, Biondi A , Kearney L. Delineation of multiple deleted regions in 7q in myeloid disorders. Genes Chromosomes Cancer 1999 Aug;25(4):384-92. van Dam H , Castellazzi M . Distinct roles of Jun:Fos and Jun: ATF dimers in oncogenesis. Oncogene 2001 20:2453-2464. Virgilio L, Shuster M , Gollin SM, Veronese M , Ohta M , Huebner K, Croce C M . FHIT gene alterations in head and neck squamous cell carcinomas. Proc. Nat. Acad. Sci. 93: 9770-9775, 1996. Wang DG, Fan JB, Siao CJ, Berno A , Young P, Sapolsky R, Ghandour G, Perkins N , Winchester E, Spencer J, Kruglyak L, Stein L, Hsie L, Topaloglou T, Hubbell E, Robinson E, Mittmann M , Morris MS, Shen N , Kilburn D, Rioux J, Nusbaum C, Rozen S, Hudson TJ, Lander ES. (1998). Large-scale identification, mapping, and genotyping of single-nucleotide polymorphisms in the human genome. Science 280:1077-82. Wang X L , Uzawa K , Miyakawa A , Shiiba M , Watanabe T, Sato T, Miya T, Yokoe H , Tanzawa H . Localization of a tumour-suppressor gene associated with human oral cancer on 7q31.1. IntJ Cancer 75,671-674 1998. Warnakulasuriya S. Lack of molecular markers to predict malignant potential of oral precancer. Journal of Pathology. 2000 190: 407-409. Welsh J, McClelland M . (1990). Fingerprinting genomes using PCR with arbitrary primers. Nucleic Acids Res 18:7213-8 82 Wen QH, Miwa T, Yoshizaki T, Nagayama I, Furukawa M , Nishijima H. Prognostic value of EGFR and TGF-alpha in early laryngeal cancer treated with radiotherapy. Laryngoscope 1996 106:884-888. Weng, LP, Brown, JL, Eng, C. PTEN induces apoptosis and cell cycle arrest through phosphoinositol-3-kinase/Akt-dependent and -independent pathways. Hum. Molec. Genet. 10: 237-242,2001. Wheeler R H , Spencer S, Buchsbaum D, Robert F. Monoclonal antibodies as potentiators of radiotherapy and chemotherapy in the management of head and neck cancer. Current Opinion in Oncology 1999 11:187-90. Williams JG, Kubelik AR, Livak KJ , Rafalski JA, Tingey SV (1990). D N A polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucleic Acids Res 18:6531-5. Zeng WR, Watson P, Lin J, Jothy S, Lidereau R, Park M , Nepveu A . Refined mapping of the region of loss of heterozygosity on the long arm of chromosome 7 in human breast cancer defines the location of a second tumor suppressor gene at 7q22 in the region of the CUTL1 gene. Oncogene 1999 Mar 18;18(11):2015-21. Zhuang Z, Vortmeyer A O . Applications of tissue microdissection in cancer genetics. Cell Vis. 5:43-8. 83 Appendix I. Complete List of Oral Archival Premalignant Squamous Cell Carcinoma (SCC) Samples (Patient Number and DNA Yield) Tube ID ng = [DNA] in nanograms/ul C-normal epithelium D= premalignant lesion Tube ID Ng = [DNA] in nanograms/ul C=normal epithelium T=tumor 108D C=16.7ng D=18.1ng IT C = 8.6ng T = 36ng 126D1-1 D=llng 2T C = 29ng T=50ng 143D1-1 D=25ng 3T C = 8.3ngT = 28.5ng 165D C=19.0ng D=13.1ng 10T1 T = 3.9ng 177D C = 6.2ng D = 9.9ng 43T C = 44ng T = 20.8ng 180D C = 5.2ngD= lOng 59T1 C = 326.5ng T = 90.5ng 184D C=10ngD=10.5ng 61T3-L T3L = 49.94ng 206D C= 74.5ng; D= 42.8ng 62T0 T = 30ng 207D C= 20.6ng D= 46.2ng 65T1 T l = 56.7ng 21D C=28.6ngD=29.7ng 80T T = 8.3ng C= 10.7ng 240D C= 10.3g D= lOng 90T T = 26.9ngC = 22.1ng 243D C = 40ng D=40ng HOT T = 57.2ng C= 131.lng 246D C = 4.5 ng D = 8.6 ng 112T T= 13.0ng 257D C= 19.4 ng D = 30ng 113T C = 14ng T=10.8ng 260D C = 6.3ng D = 10 ng 114T T= 17.6ng C = 5.6ng 26D C=5.2ng D= 14.6ng 115T C = 44ng T = 46ng 26D1-1 C=35.4ng Dl-l=13.4ng 116T T = 35.8ng 26D1-2 Dl-2=27.5ng C=35.4ng 117T1 T = 20ng 270D C=15.4ng D= 11.5ng 118T C= lO.lng T= 11.3ng 274D C=10.8ng D=24.6ng 121T1 C=14.5ng Tl=40ng 27D C=5ngD= 18ng 122T4 T = 31.7ng 282D C=40ng D=20ng 123T1 T = 86.5ng 289D C=14.5ng D=20ng 125T T = 93.5ng C = 3.4ng 292D C=13.6ng D=24ng 160T2 T = 20ng 299D C=19.6ng D=20ng 161T T=80ng,T=50 300D C=20ng D=20ng 162T T=15ng 302D C=20ng D = 20ng 166T C=6.5ng T=15ng 304D C=14.7ng D=10ng 173T2 T= lOng 306D C= lOng D=10ng 174T T = 50.1 ng 316D C=10ng D = 20ng 175T T = 34.1ng 32D1-1 Dl-l=3.56ng C=5.17ng 181T C= 15ng T = 28.8ng 84 340D C=8.5ng D=20ng 185T2 T=16.6ng C=15.9ng 34D 196T T = 67.2ng C = 20.2ng 34D1-1 Dl-l=5.37ng C=5.71ng 199T T = 200ng C = 33.5ng 364D1 C =35ng D=39ng 200T1 T = 20.3ng 367D2 C=3ng D 2 - 9ng 202T1 T= lOng C= 14.9ng. 367T1 T = 20ng 211T T = 89.9ng 368D C=5.15ng D = 20ng 213T T = 5ng 370D C=10ng, D=10ng 215T T = 200ng C=100ng 372D2-2 D2 = 20ng 228T T = 50ng C = 20ng 380D C=10ng, D=10ng 236T1 C= 15ng T= lOng 45D C=17.2ng D=22.3ng 278T2 C = 20ng T2=20ng 497T C = 6.8ng T= 15.5ng 342T C = 20ng 20ng 4D C = 27.1ng D = 20.0ng 366T0 C= 100ngTO= lOng 502T C= lOng T=10ng 385T3 T = 30ng 51D C =14.8ng D= 14.3ng 385T2 C = 30ng T = 30ng 53D C = 58.8ng D = 24.0ng 386T C = 73ng T = 30ng 55D C = 8.9ng D = 10.6ng 391T C = 17ng D = 25ng 564T1 C =20ng T l = 20ng 393T C= 19.7ng T = 25ng 58D1-2 Dl-2 =29ng C = 6.4 ng 3 9 7 X 2 T2 = 20ng 64D C= 19ngD=15.6ng 398T C = 20ng T = 20ng 66D C= 12ng D = 25ng 399T2 T2 = 20ng CI = 12.8ng 68D2 C = 6.8ng D2 = 13.2ng 401T2 T2= 12ng CI = .68ng 6D C = 37.7ng D = 32.9ng 402T2 T2 = 40ng CI =9.9ng 71D C= 14.7ng D = 31.6ng 405T2 T2 = 40ng C = 20ng 72D C = 8.9ng D = 30.4.5ng 406T2 T2 = 3.9ng C = 3.95ng 76D C = 24.5ng D= 15.9ng 414T2 C = 24.8ng T = 20ng 81D1-1 Dl-1 =63.4ng C = 27.9ng 416T2 C = 23.2ng T = 20ng 82D C = 7.2ng D = 14.2ng 430Ta C= 15.24ng T = 23ng 83D1-1 Dl-1 =30.2ng C=128.4ng 432T C = 20ng T = 50ng 88D C = 40.6ng D= 15.6ng 433T C= 11.9ng T = 20ng 92D C = 34.2ng D = 14.0ng 434T Cl=20ng T=20ng 9D C= 12.3 D = 4.7ng 435T1 C = 40ng T = 40ng 238D2 436T C= 19.33ng T = 20ng 298D1-1 Dl-1 = 25ng C = 28ng 439Ta C = 25ng T = 60ng 109D1-1 Dl-1 =8.3ng 446T2 C = 40ng T2 = 40ng 122D3-1 C = 3.3ng D3-1 = 5ng 447T1 C = 31.6ng T l = 5.2ng 12D C = 7.5ng D = 6.3ng 448T1 C = 20ng T = 14.6ng 159D C= 1.2ng D= 1.8ng 449T1 C = 5.2ng T l = 17.9ng 15D C = 6.58ng D=12.3ng 451T2 T2 = 50ng 163D1-1 C = 5.4ng Dl-1 = 14ng 451T1 C = 44ng T l =40ng 164D1-2 Dl-2 = 8.7ng 452T1 C = 20ng T = 19.3ng 167D C=11.6ngD= 11.8ng 452T2 T = 20ng 168D C = 4.6ng D = 5ng 453T1 C = 30ng T = 20ng 169D C= 11.7ng D= 10.8ng 460T2 C= 10.9ng T = 7.6ng 176D D= 18.3ng 474T2 C = 40ng T2= 19.8ng 17D1-1 Dl-1 =3.9ng 482T2 C = 31ng T2 = 40ng 85 186D C = 7.8ng D = 5.9ng 486T C= lOOng T = 50ng 187D C = 6.4ng D = 8.1ng 523T2 T2 = 50ng 188D C = 5ng D = 5ng 529T1 C = 7.1ng T = 7.8ng 189D C = 5ng D = 5ng 537T1 C=l lng T l = .6ng 190D C = 2.7ng D = 5ng 559T3-1 C = 20ng T3-1 = 30ng 191D C = 4.7ng D = 5ng 566T1-2 C = 50ng Tl-2 = 50ng 192D C = 0.77ng D = 3ng 568T1 Cl = 20ng T l = 20ng 195D1-1 Dl-1 = 7.6ng 577T1 Cl =20ng T l = lOng 19D1-2 Dl-2 = 5ng C = 21.3ng 587T1 Cl =5.4ng T l = lOng 201D C = 8.5ng D = 5.6ng 204D1-1 Dl-1 = 30ng C = 29ng 208D C=13ng D=9.8ng 209D C = 27.5ng D= 15.8ng 212D1-1 Dl-1 = 18.5ng 214D C = 29.9ng D=19.4ng 216D2 D2 = 8.5ng 22D1-1 22D1-1 = 8.3ng C=7.4ng 232D C = 5ng D = 36ng 233D C = 19.3ng D = lOng 234D C = 20ng D = 20ng 238D1 C = 50ng C = 23ng 239D2 D2 =lOng 241D C = 2.8ng D = 9.8ng 242D C = 2.4ng D = 9.4ng 248D C = 4.5ng D= lOng 24D2-1 D2-l = 13.9ng 24D3 D3 = LOng 24H7 C = 40ng H7 = 20ng 252D C = 20ng D=10ng 256D C = lOng D = 3.3ng 258D C = lOng D = lOng 25D C=17.2ng D = 22.7ng 277D C = 5ng D = 8.4ng 284D C = 9.4ng D = 20ng 288D C = 8ng Dl = 9.4ng 290D C = 14.5ng D = 8.2ng 291D C = 4.8ng D = 9ng 294D C = 3ng D = 8ng 295D C = 2.9ng D = 4ng 296D C = 5.6ng D = 22ng 297D C = 28.8ng D = 20ng 301D C= ling D= lOng 303D C= 12ng D= lOng 305D C = 9ng D = 14ng 305D2 C= lOng D2 = 20ng 307D C = 4.6ng D = 8.8ng 308D C = 6ng D = 9ng 309D C = 5.8ngD= lOng 86 310D C = 3ng D = lOng 311D C = 6.4ng D=10ng 335D C = 6.7ng D = 6.4ng 339D C = 9.3ng D = 20ng 341D C = 6ng D = Ing 347D C = 2ng D = 5ng 361D C = 4.5ng D= lOng 362D1 C = 6.6ng D = Ing 371D C= 13ngD= lOng 373D C= 10ngT= lOng 374D C = 6ng D = 8ng 379D C = 6.8ng D = 7.6ng 382D C = 4.4ng D = 6.5ng 387D1 C = 8.8ng Dl = lOng 388D C= ling D = 9ng 390D C = 4.15ng D=10.9ng 414D1 C = 21.5ngD= 19ng 418D1A C = 50ng A = 23ng 5D Cl = 9.9ngDl = 7.7ng 60D C= 15.8ngD= 13.6ng 65D D = 2.0ng 78D1-1 Dl-1 = 12.6ng 88D2 C = 6.2ng D = 17ng 88D3 D= 10.9ng 87 OX in O a B te ed c « -a a o -o u S ~ B o OX o O a h i •a ox M I <N H -o ox ,2 '5 o - o tu _ s ce e a. -a 'C •c s 09 s _ © 4) •fi a 03 M o> H ox M > vs u ss o '•3 « h 1 3 fl <y U .2 1 i fl " f l w a a « 0 v V= fl _ e» 2 | A I .•fl £ ft • -fl .= U es e* a <w 1 •-= s 1*8 "3 S o « BR H BB U I a as y 8. Si fl •5x •e o ox fl o ex 3 _ _ ce 0 e £ 5 ox a * a _* I? a u. Pi -2 1 S u a ° 1 o « a 2 p _ <fc .2 -5 fl s Sf a § ex ox g fl * o ce II >• O fl o S O a ? 4> —t —i o o .2 -fl 8 a ex 4> fl <U 3 a o ie Oi 0 c V 3 O" 9i c * te a . <S -U -= te z Q 3 s IT 4» .2 a fl 5T O fl • - 4J a u s e a s o 3 3 v a fl P . te ft 13 < ox <U te 9 fl fl C 13 fl 'fl f^l CT g O 1) O co co O « 88 u o c <u 0> 5 o O a •c N fl 0> In a © u a a SB 4» DC u 3 •« O u a •a <u C a E « 2 -<o o c & T3 c + PQ I-I u s •c OH T 3 fl ft) fi ^ O IT) ->-> o ( £ -S -fl o "H, B § fl <u E & .S — ° I m * OH a) fl C t-i m o H—' fl > O 8 « A £ tU i _ , U 0) h2 «> OH —< c3 ^ -« ° 2 W - ° <D t o U ^ fl O -4-» HD C 0) 8 3 < 2 » co S 2 OH » g CO fl ° S PQ £ 89 o 09 o o a>£ V u •O 88 •3 3-s « « « 3 .2 j5> 3 O O 3 * 3 O « ••C -3 3 S » a 3 «ft _3 "3 ^ -a ja au es — s £ S ° «-a, T3 -3 .gp £ 3 ° 0 c o an W -M « 3 A t» S. se t« — ty 3 O s. 3 ot -3 O M> V « S S 2 2 Is 3 9 3D 2 0£ ?r -3 « * •» * | | T » C/D » s / wo g a •d « S . « s , 3 oc O C5 3 -H \a 1 i -a .H CO 9 0 1/3 as 6fJ • -o > o 09 03 i u & a © a •a 4> •<-< 03 M 13 fi et "3 o 1 >> •-o 03 _>•. "© a <u -c £ o •fa T3 0) a> -OS fl £ OJD OS •s o 09 CJ •-© u Si a a> 09 .5 on £ e cs o ee .2 "o -— « T3 C 03 "OJD fi IM 4> -a oc • fi g cU • - M § « fi *> © 91 T3 C 04 tn •a © -a ii £ a a ii U B cu 3 o* a 1/5 "3 _w "3 a ii 2 c ' « * J a o o cu u -u o u !/) a s a o ii OX a *-*-* a a -ii OX © .2 w 1 & O ps! O f C M d %-> m d "is P-. d PH d • 1-H CO -i—> d +-> o a d PH d . ^  c3 PH r3 5 5 3 O 2 i g <U O r3 *7J • H a •o ii u a <u a cr 4> S3 a a •mm « a a "45 3 s s o ,1M a -a ox 3 C o "3 u 09 a 9 a S -s I S * 2 £ S s i 2 S B 2 « -3 00 u 92 Appendix III. Complete List of L O H Primer Sequences and Locations. Oligo name Sequence 5'-> 3' 7Q REGION D7S1812F (GDB) G G A G G A A G G T G G A T A A T A A T A G G D7S1812R (GDB) A A T GTC A C A A G G GTG C A G A T D7S2539F (GDB) A C C A G T T G A TTC A G C A A T CC D7S2539R (GDB) G G A GGG TAT TCC ATT GTC T G D7S491F(GDB) A G C T C C A A A A C C T A A C T C C A D7S491R(GDB) T C A A A A T T A T T T G A C T T C T T G A T T T D7S479F(GDB) C T G G G T G A C A G A G C A C C A G A D7S479R(GDB) C C T T C T T G A T A G A T A T T A G G G G T T T D7S476F(GDB) C T C A G T T T T A T C A G C T G T G A D7S476R(GDB) TTAGCTTGTGGTCTCTACTC D7S624F(GDB) G A G A G A C C A C A T G G A G A A G D7S624R(GDB) A G C T G G T A C T A T A A C A G T G G UT901F (GDB) alias D7S618 A A G A C C C A G T C T C A A A G A A G UT901R (GDB) alias D7S618 T T T C A G A T G A T G A A A C C G A T G D7S1558F (GDB) A T C C A G CCC T A A G A T A A G D7S1558R (GDB) TTT C A C A T C A G T A C C C A G AFM249za5F (GDB) GGC TGC CTT C A A A A A CTC AFM249za5R (GDB) A G C CTG G C A TGT G G A T D7S1619F (GDB) CGT TCT T A C T A C A C A TCC D7S1619R (GDB) TGT GTG ATT A G T GTG A T T CC D7S477R (GDB) TTT GGG TAT CCC CTG TTC A C D7S477gdbF (GDB) T T G C A C C A C T G T C T C C A G T C AFMa226yelF (GDB) TTT G A G TCT T C A C A G C A G TTG AFMa226yelR (GDB) GGG A G G TTG ATT TCC A C A GT D7S1530F (GDB) GTA GTG A G C C G A GAT TGC A D7S1530R (GDB) CCT A C T A C C T C A T G G C A T GT AFMal26zc lF (GDB) CCT GTA TGG A G G G C A A A C T A AFMal26zc lR (GDB) A A A T A A T G A CTG A G G CTC A A A A C A AFMa305ye9F (GDB) A A G A A G T G C A T T G A G A C T C C AFMa305ye9R (GDB) C C G C C T T A G T A A A A C C C D7S501F (MV/GDB) C A C C G T T G T G A T G G C A G A G D7S501gdbR (GDB) A T T T C T T A C C A G G C A G A C T G C T D7S501F C A C C G T T G T G A T G G C A G A G D7S501R T G A C C T T G T A A G C A T G T G 93 2P REGION D2S1998F (GDB) TGA A A C CCC ATT A A G A A G A G A TC D2S1998R (GDB) GGG A G A A C T TTT C A G TTA A T A TTC C R68763F TCA GTG A A G A C A A T G A A A A C C C R68763R A G G TGC TGT GTT A C G G G A G WI-9798F (GDB) CAT A T C A C C TGT CTT TGT G T A CTG C WI-9798R (GDB) CTA A C A CTT ATT G A G C C C TTT C T G RH66882F G A C ATT TTC A G A A C A ATT G RH66882R A T A A A G A G C A G T A G A GTC CC stSG46797F G A G G G A G G A A G A A C G CTT CT stSG46797R A T G A A C A T G A T G G G A TTC TGT G AFMb009xflF (GDB) TCC A T C TTT TGC GTG C AFMb009xflR (GDB) C C G TGC TCT A T G C C A G stSG2766F CCT G A G CTG C A C A G A A A G A C stSG2766R GTA CCC C A C G A A GTC T C A G T G D2S2350F (AFM350yel) (GDB) C C T T C T G A T G G G G G A G D2S2350R (GDB) TGGCCCTATGTAGTCTCTTT AFMa305wh9F (GDB) C A C TGC GCC T A G CCT C AFMa305wh9R (GDB) GGC GAT TTA T G A A T A A T C CTG C D2S158F (AFN217xh8) (GDB) C C A C C A T A T T C C A G T C T G A G C D2S158R (GDB) A T G A A A C A A C C A A C C T G A G A A T A SHGC-10114F GTG G G A A C T CCT TTG C A T GT SHGC-10114R TGA TGT G A G TCC T A C A T T TGC C D2S2170F (AFMal36wh9) (GDB) T A A G G G T T C A A C A A T G C C T G D2S2170 (GDB) G A T C A T A A G A G T G A A A C T C C G T C AFM242yd8F (GDB) TGG C A G G C A G A G GTT A AFM242yd8R (GDB) GTG C A A A G G CCC A G A G AFMa052yb5F (GDB) G C A CTT GCT C A C CGC T AFMa052yb5R (GDB) T A G CCT TCT GCC T C A T A C A T D2S2312F (RAPD change) (GDB) TCTCACAGGGATTGTTTTTC D2S2312R (GDB) G A T G G T T T T G A A T G C C A C WI-12589F (GDB) A G G G A G TAT GGC ATT TAT T A A C C WI-12589R (GDB) T A G TCT A T G A T G T G A CTC C C C C D2S170 T T G C T C A A T A A T G T C A G G T G D2S170 C G C A T G A G A G G C G T C T D2S390F C C C A T A G G A G G T G C T C A A D2S390R C A T G C T G C A T C C T G G T A D2S375F (GDB) T T T G T A T C A G A G G G A T T G C D2S375R (GDB) TTGAGTTTCTGGGGATTC D2S400F (GDB) A A T G T G A C A A A G C C C A G T G T T A G C D2S400R (GDB) G A T A A T C T C C C T G A G T A T G T G T G C C AFMb313yd5F (D2S2283 RAPD) (GDB) GAT CAT GGC CCC A A T A 94 AFMb313yd5R (GDB) GCC C C A GGT A A C C A C T AFMa239yd5F (GDB) alias D2S 2203 A A G TGC TTA G A A G G G TTC C T G A C AFMa239yd5R (GDB) A T A TGC TCT CTG GTG A C T G T A GGT G AFM303zelF (GDB) alias D2S367 TTC TTT GGT C T A A G G GTC A C AFM303zelR (GDB) AGC TTC TTG TTC A C A GGT GT AFMa340zc9F (GDB) CTT A G A A A C G A A C C A CTG G A AFMa340zc9R (GDB) GTG G A A A T A A G T G C A A C C A T D2S2351 F(GDB) T G A A C T T T G C A G T G A G A A A D2S2351R(GDB) TTTACTGTGTATGTGTGTGTACTCT 10 Q REGION D10S681F TTC TGC CGC G A A G A C CCT T G D10S681R G A A CTT A G C C A G TCT TCC TTC D10S601F GGG GAT G C A A A A C A A A C C TT D10S601R GGC A T A A C A GAT GGC TTT A A T TTT C 5Q REGION D5S683F A G A A G A G G T G G A T T C C C C C D5S683R T C T C C A T A A A T A A A G C C A T G C A D5S2059F A G G G G A C T T T T C A G G A T G C D5S2059R G C T A C C C A C G G G G T T T A T T A T T D5S622F (AFM205zd4 alias) A A C A T T T A A T C C C A A A C T T C T C A D5S622R G C T T G A A A T T C T C T T A T T A C A G A G G 95 M x> 03 fl "58 CO CO <U - f l T3 c .S co a> 2 < GO S is 9 T 3 5 S a <U co fl <u —i co 03 * ^ - O . 03 ™ i CJ *° < o PQ A 3 oflO H •a -a .a m co 2 <u fl; CO .fl CO l-J u CN U . (N fl u a w J H 03 CO 3 13 I O H T 3 o fl°|^ a i g " S § CO < ™ < ffl >>(D -3 Ci_, c3 W3 flo 5•» a M a3 -5 .a a !* O H 'fl3 1) < "5 W 96 97 98 99 100 101 102 103 104 105 106 107 T - CN OJ CM CM CM CM , ^ ^ ^ 1 - K r - r - H H H I - H r - r - r - r - r - r - r - r - f - r - l - r - r - h h r - . - 1 - 1 - 1 - 1 - 1 - r- 1 - . h h h h O W n i f l N C O i f i T - ( N U O C D O ) 0 ^ n-OCOinCDNi - n CD CM CD 0 ) ^ ^ C M ( D ( O C O N C ( O C D i n O ' - T - T - T - T - T - f N ( D ( D N C I O ) 0 ^ - ' - T - S C O < 0 0 0 0 ) 0 ) O l O r - CO(Ot*-ODCOCN<Dr*-'^ • Q Q I I I Q Q I * d d d * d d d d i d d d d d d d d d d d d d d d d d d d QQPPPI d d d d d m d d * d d d D P D P 1 0 8 m Q Q Q Q Q Q Q Q Q Q O I - . f v CM A Q Q Q O Q Q Q Q O -ana to in a> in co Q CM (0 2 o . 00 o CO Q CO l>-Q oo CD <£ cn co if) m T— co Q O! CO CO Q m LO co a O a o CN £ £ S q 8 3 Q IP L O § CO (0 » C 1 Ivr-CO I-Q 3 o> CO co. in a - 8 dddd aa pnniiinin dd ddd aaa dd q q q q q dd addd *dddd id i d dd ! 1 ddddd dd ddd diiiddd PPPP id ddd ddd ddd [c cc l a n I T 109 oo 0\ T l -Q o © U © s ^ ™ P i 2 f- W £ ^ © T J -CZ5 Q m © © U 0 0 S V© 1-H 0 0 »—I Q oo oo © IT) © © u oo o PQ T f o rn rn © •s O G <zi •a s co 2 o r n > , CN CN C QH - r t r- u S> ca £ a •a § CD -t < o -OH ^ 6 £ .a 6 b 3 QH G < -a . co p—i CD fcr cj c • PH r-


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items