Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Identification of novel genetic alterations in non-small cell lung cancer progression Siwoski, Arek 2002

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

Item Metadata


831-ubc_2002-0246.pdf [ 13.04MB ]
JSON: 831-1.0090286.json
JSON-LD: 831-1.0090286-ld.json
RDF/XML (Pretty): 831-1.0090286-rdf.xml
RDF/JSON: 831-1.0090286-rdf.json
Turtle: 831-1.0090286-turtle.txt
N-Triples: 831-1.0090286-rdf-ntriples.txt
Original Record: 831-1.0090286-source.json
Full Text

Full Text

IDENTIFICATION OF NOVEL GENETIC ALTERATIONS IN NON-SMALL C E L L LUNG CANCER PROGRESSION By A R E K SIWOSKI B.Sc , The University of British Columbia, 1999 A THESIS SUBMITTED IN PARTIAL F U L F I L L M E N T OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE F A C U L T Y OF G R A D U A T E STUDIES Department of Pathology and Laboratory Medicine We accept this thesis as conforming to the required standard THE UNiVERSITY^OF BRITISH C O L U M B I A December 2001 © Arek Siwoski, 2001 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada DE-6 (2/88) Abstract To improve the survival rate in lung cancer, novel molecular targets that facilitate early detection and intervention are required. This thesis aims at screening multiple loci from distinct stages of lung cancer using genome scanning methods to identify frequently occurring genetic alterations that could potentially reveal novel regions of chromosomal instability. Due to the tissue heterogeneity in biopsies, tissue microdissection is necessary to isolate specific populations of cells from formalin-fixed, paraffin-embedded tissues representing progression stages of lung cancer. Since some of the commonly used histological stains such as Hematoxylin & Eosin have been shown to degrade D N A due to their acidic nature, a panel of histological stains have been tested for minimal degradative effect on DNA. To work with the minute quantities of D N A recovered from premalignant lesions, we have adapted a Randomly Amplified Polymorphic D N A -Polymerase Chain Reaction (RAPD-PCR) assay to assess the quality of D N A extracted from the microdissected specimens, as well as a Southern hydridization assay to assess D N A quantity. D N A samples have been extracted from normal epithelial, hyperplastic, dysplastic, CIS and invasive carcinoma cells originated from 41 patients at various bronchial sites. D N A yield varied from 20ng to 2000ng depending on the amount of cells available in each lesion. Typical molecular techniques such as loss of heterozygosity (LOH) assay, D N A microarrays and fluorescent in situ hybridization (FISH) require large amounts of D N A to detect genomic changes. Using RAPD-PCR as a multi-loci fingerprinting technique, it is possible to screen for genetic alterations using minute D N A samples from microdissected specimens. RAPD-PCR involves simultaneous amplification of genomic D N A at multiple loci using short oligonucleotide primers. ii Gains and losses of PCR signals that occur in multiple patients identify regions potentially harboring candidate tumour suppressor genes or oncogenes. Our results show that twelve patients showed a gain of PCR signal in carcinoma in situ (CIS) tissue compared to normal tissue. By cloning, sequencing, and mapping the fragments, the recurring gains in PCR signal in multiple patients corresponded to regions at 8q24.3 (6/17), lq23.3 (2/18), 5q33.1 (2/16), and 7q36.2 (2/15). Microsatellite markers were chosen surrounding all four regions, however, due to minimal amounts of patient D N A , only regions 8q24.3 and 7q36.2 were verified for allelic imbalance in multiple patients using L O H . The groundwork has been laid for others to define the boundaries of these unstable regions as both regions contain many potential oncogenes and tumor suppressor genes. iii TABLE OF CONTENTS Abstract i i Table of Contents iv List of Tables vi List of Figures vii Acknowledgements viii Chapter 1: Introduction 1.1 Background 1 1.2 Genes Implicated in NSCLC 3 1.2.1 Potential Oncogenes 3 1.2.2 Potential TSGs 5 1.3 Chromosomal Regions Implicated in Premalignant N S C L C 7 1.4 Epigenetic Changes in Cancer Tumourigenesis 8 1.5 Smoking and Lung Cancer 9 1.6 Current Methods for Genome-Wide Scanning of Alterations 10 1.6.1 Comparative Genomic Hybridization 10 1.6.2 Loss of Heterozygosity 12 1.7 Archival Tissue Used in Search of Novel Markers 14 1.8 Tissue Heterogeneity and Microdissection 15 1.9 Visualization of Morphology within Biopsies 16 1.10 Use of RAPD-PCR to Search for Novel Genetic Alterations.... 17 1.11 Objectives and Hypothesis 22 Chapter 2: Materials and Methods 2.1 Tissue Microdissection and D N A Extraction 23 2.2 D N A Quantification 23 2.3 Gene-Specific PCR 24 2.4 RAPD-PCR 25 2.5 Cloning 25 2.6 Localization of Clones to a Chromosomal Position 27 2.7 Choosing Microsatellite Markers 27 2.8 Loss of Heterozygosity Assay 28 Chapter 3: Results 3.1 Determining a Patient Pool 29 3.2 Determining a Method for Obtaining Pure Cell Populations 30 3.2.1 Manual Microdissection 31 3.2.2 Laser Capture Microdissection 33 3.3 Effects of Histological Stains on D N A Quality 33 3.4 Developing an Assay for Assessing D N A Quality of Archival Samples 37 3.4.1 Assessment of D N A Quality Using Gene-Specific PCR 38 3.4.2 Assessment of D N A Quality Using RAPD-PCR 38 iv 3.4.3 Screening Biopsies for D N A Degradation Using RAPD-PCR 40 3.5 D N A Quantification of Microdissected Archival Lung Biopsies Using Southern Blot Hybridization 40 3.6 Screening Oligonucleotides for Increased Genome Scanning Density 42 3.7 Scanning Microdissected Archival Lung Biopsies for Genetic Alterations Using RAPD-PCR 44 3.8 Cloning of Genetic Alterations Potentially Associated with Lung Cancer Progression 46 3.9 Localizing Sequenced Products to Chromosomal Locations.. ..53 3.10 Identification of L O H Markers Specific to Cloned Regions.. ..53 3.11 Loss of Heterozygosity Analysis 57 3.12 Previously Discovered Genes at 8q24.3 and 7q36.2 60 Chapter 4: Discussion 4.1 Summary of Results 63 4.2 Harvesting D N A from Multiple Grades of Tumourigenesis in Multiple Patients 63 4.3 RAPD-PCR, A Method of Genome-Wide Scanning 65 4.4 Importance of Recurrences 67 4.5 Two Candidate Regions at 8q24.3 and 7q36.2 68 4.5.1 Genes Implicated at 8q24.3 69 4.5.2 Genes Implicated at 7q36.2 70 4.6 Implications of 8q24.3 and 7q36.2 in Epithelial Carcinomas.. .70 4.7 Future Directions 71 Conclusion 74 References 76 Appendix 1. Microdissection Protocol and Histological Stains 83 Appendix 2. Patient Profiles and Microdissected Cases 84 Appendix 3. Cloning Procedures 89 LIST OF TABLES Table 1: Relevant Histological Grades in Non-Small Cell Lung Cancer 30 Table 2: RAPD-PCR Primers Screened for Increased Scanning Density 43 Table 3: Fifteen Different Band Type Alterations 47 Table 4: Microsatellite Markers Chosen for 8q24.3, 7q36.2, 5q33.1 and lq23.3 Regions 56 Table 5: Analysis of 7q36.2 L O H Results Using ImageQuant 5.0 Quantitation Software 61 vi LIST OF FIGURES Figure 1: C G H Analysis of the Chromosomal Imbalances in 11 Neuroendocrine and 11 NSCLC Carcinomas 11 Figure 2: Loss of Heterozygosity (LOH) 13 Figure 3: Randomly Amplified Polymorphic D N A (RAPD) - PCR 19 Figure 4: Manual Microdissection of Lung Biopsies 32 Figure 5: Assessing the Degradative Effect of Hematoxylin and Methyl Green on D N A Quality 35 Figure 6: Determining Staining Ability of Histological Stains 36 Figure 7: D N A Quality Assessment of Seven Microdissected Archival Oral Tumour Pairs 39 Figure 8: Southern Blot Quantitation Assay of Microdissected Lung Biopsies... .41 Figure 9: Screening Primer Pairs for Increased Scanning Density 45 Figure 10: Recurring Alterations in Multiple Patients 48 Figure 11: Reamplification and Purification in Cloning Procedure 50 Figure 12: Colony PCR and Colony Fingerprint in Cloning Procedure 51 Figure 13: D N A Sequence of Clone B / M 52 Figure 14: Localization of Sequenced Product to Chromosomes 54 Figure 15: Identification of L O H Markers at Chromosomal Regions 8q24.3 and7q36.2 55 Figure 16: Contigs of Overlapping BACs 58 Figure 17: Loss of Heterozygosity (LOH) Analysis 59 Figure 18: Known Genes at 8q24.3 and 7q36.2 62 vii Acknowledgements I would like to my supervisor Dr. Wan Lam for his great ideas, scientific enthusiasm and guidance and members of the laboratory especially Krista Cleveland, Cathie Garnis and Chad Malloff for their technical advice and assistance. I would like to thank Dr. Juergen Vielkind for all his guidance. I would like to acknowledge Dr. Stephen Lam for providing us with lung biopsies, Chris Dawe for microdissection assistance, and Dr. Calum MacAulay for innovative ideas. Finally, I would like to thank my beautiful wife, Jenny for all her support, encouragement and laughter. I would also like to thank my family for their constant optimism and support throughout the years. viii Chapter 1: Introduction 1.1 Backeround As cancer is becoming the major cause of death in the civilized world, lung cancer mortality rate is increasing dramatically. In the United States, it was estimated that 171,600 new cases and 158,900 deaths were due to lung cancer in 1998 (Tseng, 1999). The most effective treatment for lung cancer is surgical resection however, present screening techniques result in 65% of patients presenting with advanced stage disease at time of diagnosis (Esteller, 1999). After surgical resection, many patients will succumb to recurrent lung cancer at distal sites. Despite some advances in treatment, the two-year survival rate of lung cancer patients without complete tumour resection was estimated at 10% (Wiest, 1997). As with other potential metastatic cancers, patients with early stage lung cancer, where the abnormal cells are localized and small in volume, show the best response to therapies and present the greatest survival rate compared to patients with the advanced-stage disease. Therefore, improvements in early detection techniques through the identification of prognostic markers in conjunction with the identification of a gene or group of genes linked to lung cancer predisposition are needed to reduce the mortality rate. Lung cancer can be divided into two main types: small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC). SCLC's account for -25% of all lung cancer cases whereas the remainder of cases are adenocarcinoma, squamous cell carcinoma or large cell carcinoma, subtypes of NSCLC's . Adenocarcinomas account for 40% of all diagnosed lung tumours. Although there are four discrete histological types, it is well accepted that in many cases, an individual tumour can present features of more than one 1 type. It has been suggested that all histological types arise from similar bronchial epithelium differentiation pathway. Embryologically, lung development begins with the out-pouching of epithelial cells from the foregut endoderm. Fetal lung cellular differentiation and branching morphogenesis gives rise to the characteristic respiratory epithelium. The pseudostratified epithelium of the bronchus is composed of basal, secretory and ciliated cells whereas the alveolar epithelium consists of type I and II alveolar cells (Burkitt, 1993). Presently, a pluripotent stem cell giving rise to the epithelial cells of the lung has not been identified. Lung cancer development has been suggested to involve the accumulation of genetic mutations in the bronchial epithelium resulting in a progression through a series of morphologically distinct premalignant changes leading to invasive cancer. The progressive histological stages include basal cell hyperplasia, squamous cell metaplasia, mild, moderate and severe dysplasias, carcinoma in situ and invasive cancer. As this is a multistep process, it involves the mutation of several entities such as oncogenes, tumour suppressor genes and D N A repair genes. According to Knudson's two-hit theory of cancer progression, tumour suppressor genes are inactivated by a recessive mutation in one allele and loss of the other allele that retained wild-type function. Since both copies of the gene must be inactivated for a phenotypic change, the first copy is inactivated by point mutation, methylation or small deletion whereas the second copy is inactivated by large deletions. Regions of frequent chromosomal deletions represent potential tumour suppressor gene presence. Similarly, the mutagenic theory of cancer development, as introduced by Vogelstein and Fearon, encompasses four main ideas: 1) activation of 2 oncogenes coupled with the inactivation of tumour suppressor genes gives rise to tumours. 2) the formation of a malignant tumour requires mutations in at least four or five genes. 3) a tumour's biologic properties are representative of the total accumulation of genetic changes and 4) tumour suppressor genes that are mutated can cause a phenotypic effect in a heterozygous state. By identifying the early genetic mutations, potential prognostic markers may be discovered. 1.2. Genes Implicated In NSCLC 1.2.1 Potential oncogenes Potential oncogenes at multiple chromosomal locations have been identified using cytogenetic and allelotyping assays. These oncogenes include the Ras gene, M Y C gene and Bcl-2 gene. The Ras gene family composed of K-Ras, H-Ras and N-Ras encode a Ras protein primarily involved in growth signal transduction to the cell nucleus. In its active signaling state, the Ras protein binds guanosine triphosphate (GTP) and in its inactive state, the Ras protein binds guanosine diphosphate (GDP). The Ras protein has an intrinsic GTPase activity causing hydrolysis of GTP to GDP. Mutations in the Ras gene cause Ras protein to lose its capability to hydrolyze GTP resulting in the permanent active state of the Ras protein and continued growth signal transduction. Common mutations in the Ras gene are single point mutations at codons 12, 13, or 61 occurring in 20-30% of lung adenocarcinomas and 15-20% of NSCLCs (Gazdar, 1994; Sugio, 1994; Sekido,1998). K-Ras mutations seem to account for 90% of Ras mutations in lung adenocarcinomas (Sekido,1998). Many studies report that K-Ras mutations predict a 3 poor prognosis in both early and late stage NSCLC and suggest its use as a clinical prognostic marker (Rosell,1993). The M Y C gene family composed of M Y C L , M Y C N , and M Y C encode for a transcription factor activated by the RAS signal transduction cascade. The M Y C proteins recognize a consensus sequence C A C G T G resulting in activation of genes involved in normal cell growth and proliferation. It is implied that there is direct activation of genes involved in D N A synthesis, R N A metabolism and cell-cycle progression. Common aberrations of M Y C genes include gene amplifications and transcriptional dysregulations resulting in overexpression of protein product. Previous studies show M Y C amplifications in 18% of SCLC tumours and 8% of NSCLC tumours (Richardson, 1993). Comparative genomic hybridization (CGH) (described in 1.6.1) has detected gene amplifications ranging from 20 to 115 copies per cell. M Y C genes localize to three distinct chromosomal regions: M Y C L (lp32), M Y C N (2p25) and M Y C (8q24). Previous studies show that 31% and 20% of SCLC and NSCLC cell lines, respectively have M Y C amplifications (Richardson, 1993). It is reported that M Y C amplifications correlate with decreased survival as cell lines derived from metastatic lesions demonstrate increased frequencies of amplifications compared to primary tumours (Richardson, 1993). Apoptotic pathways have been considered as possible areas of gene regulation involved in lung cancer development. Bcl-2, a protein involved in normal apoptosis, is believed to protect cells from the natural programmed cell death enabling tumour cells to escape apoptotic death. Several reports show that the Bcl-2 protein expression is greater in SCLCs than in NSCLCs. It has also been reported that 75-90% of SCLCs express increased levels of Bcl-2 protein (Jiang, 1995). Acting as an antagonist to Bcl-2, B A X , a 4 tumour suppressor gene candidate, encodes a protein involved in promoting apoptosis. The ratio of BAX:Bcl-2 is believed to indicate a cell's susceptibility to apoptosis. 1.2.2 Potential Tumour Suppressor Genes Cytogenetic and allelotyping studies have elucidated potential tumour suppressor genes at multiple chromosomal locations. Mutated tumour suppressor genes such as p\6mK4, RB, p53 and FHIT have been characterized at 9p21, 13ql4, 17pl3 and 3pl4, respectively. Various mutations including deletions, insertions and splice mutations at chromosome 9p21 in many lung cancers indicated that there was a possible causative gene, pl6MK4. The p l 6 I N t c 4 protein is responsible for cell cycle modulation of the RB pathway by inhibiting CDK4:cyclin D l kinase activity. Abnormalities at p l 6 I N K 4 are frequent in N S C L C where homozygous deletions and point mutations have been reported in 10-40% of NSCLCs (Okami, 1997). Gene inactivation of pl6mK4 by hypermethylation of the 5' CpG island leading to disruption of the RB pathway has been reported in 30-70% of NSCLCs (Hamada, 2000; Sanchez-Cespedes, 2001). It has also been reported that p l 6 I N K 4 mutations can be correlated with tumour progression as the mutation frequency increases in metastatic lesions compared to primary lesions (Sekido, 1998). Initially discovered in childhood retinoblastomas, the retinoblastoma gene, RB, was localized to chromosome 13ql4. The RB gene encodes a nuclear phosphoprotein involved in the pl6-cyclin D1-CDK4-RB pathway responsible for the transition of the cell cycle from G l to S. In its inactive state, the RB protein binds the transcription factor E2F-1 responsible for the G l to S transition, inhibiting S phase entry. When 5 phosphorylated by the cyclin dependent kinases, the RB protein loses its binding ability allowing S phase entry. The RB gene has also been linked to the repression of apoptosis and induction of cellular differentiation. Common aberrations of the RB gene include truncating deletions, nonsense mutations and splicing mutations. Studies report an abnormal RB protein in 90% of SCLCs and 15-30% of NSCLCs (Tamura, 1997). It was also shown that germline carriers of an RB mutation were 15 times more likely to die from lung cancer. It was also observed that lack of RB expression was correlated with poor prognosis in NSCLCs (Dosaka-Akita, 1997; Onuki, 1999). The p53 gene localized at chromosome 17pl3 encodes a transcription factor primarily involved in D N A damage response. The function of the p53 protein is to bind sites of D N A damage and activates a cascade of genes involved in arresting the cell cycle and apoptosis. It is believed that p53 function is critical in the death of genetically damaged cells preventing the evolution of cancer cells. Common alterations of the p53 gene include deletions, missense and nonsense mutations and splicing abnormalities of exons 5-8. Allelic loss of the p53 gene is frequent in both SCLC and N S C L C and mutational inactivation of the remaining allele occurs in 75-100% of SCLCs and 50% of N S C L C (Ahrendt, 2000; Onuki, 1999). Abnormal p53 protein expression has been shown in 40-70% of SCLCs and 40-60% of NSCLCs (Ahrendt, 1999). However, data has shown that the use of p53 expression as a prognostic marker for survival is not correlative (Dosaka-Akita, 1997). Using cytogenetic and allelotyping analyses, it was determined that there is a deletion of one copy of the chromosome 3p in 90% of SCLCs and 80% of NSCLCs (Wistuba, 2000). This allelic loss occurred at both premalignant and malignant stages of 6 lung cancer. Using allelotyping, three distinct regions were identified at 3p25-26, 3p21.3-22 and 3pl4-centomere suggesting that there are at least three tumour suppressor genes at 3p. Further, these regions were confirmed by homozygous deletions of these areas seen in several lung cancer cell lines. At 3pl4.2, a potential tumour suppressor gene, FHIT, was proposed based on frequent loss of heterozygosity in N S C L C and homozygous deletions in several lung cancer cell lines. In 80% of smokers, the FHIT gene showed allelic imbalance. Studies on FHIT expression have shown that 49% of N S C L C specimens exhibited significant staining decrease for FHIT using immunohistochemistry assays. Although many potential oncogenes and tumour suppressor genes have been identified, no particular gene has been supported as characteristic of N S C L C progression. There is a need to discover a gene(s) specific to an increased risk of N S C L C development and progression in order to develop therapeutic applications and impact the lung cancer mortality rate. 1.3 Chromosomal Regions Implicated in Premalignant Cases of NSCLC Since cancer progression involves many accumulating genetic aberrations, it has become important to study the premalignant stages of tumourigenesis to identify abnormalities that act as gatekeepers to further progression. Studies report that premalignant lesions express genetic abnormalities found in cancer cells such as RAS upregulation, M Y C overexpression, and allelic losses at the p53 locus. A previous allelotyping study of microdissected premalignant cells showed that the earliest change involved loss of the 3p allele followed by allele losses at 9p, 17p, 5q, and RAS mutations 7 suggesting that the earliest change at 3p may harbour a potential gate of lung cancer pathogenesis (Gazdar, 1994; Wiest, 1997). Another study suggests that K R A S mutations occurring in hyperplastic lesions may serve as potential gates for pathogenesis (Sugio, 1994). Interestingly, it has been shown that the allelic losses occurring in the spatially distinct premalignant lesions such as the loss of 3p, 9p and 17p, are identical to the allelic losses in the primary tumour. This evidence supports the field cancerization theory of cancer development whereby either identical clones spread throughout the lung or there exists an inherited predisposition of specific allelic loss. It is clear that mutations identified in premalignant stages have the most effective potentials as prognostic markers. 1.4 Epigentic Changes in Cancer Tumourisenesis The accumulation of epigenetic changes such as D N A methylation is believed to be involved in cancer tumourigenesis. Gene expression is regulated at the 5' end of genes by the D N A methylation of the fifth carbon position of cytosine residues within the CpG islands. D N A methylation is a native process that is crucial in aging, embryonic development, and gene imprinting. Methylation of D N A is also critical in the regulation of housekeeping genes and tissue specific gene silencing. It is believed that the gene silencing is either caused by the inability of native transcription factors to bind to D N A sequences that have been methylated or the ability of repressive transcription factors to bind to methylated CpG islands or the altering of chromatin structure via methylation converting the D N A into an inactive form (Esteller, 1999). M Y C expression has been shown to be affected by methylation as the M Y C binding sites contain CpG islands and 8 several repressive transcription factors such as MeCPl and MeCP2 have been identified (Momparler, 2000). Promoter hypermethylation of genes involved in cell cycle control, transcription factor regulation, and cell-cell interactions is believed to play a critical in tumourigenesis. Studies show that in NSCLCs, gene silencing of the tumour suppressor gene, p ^ 1 ^ 4 , is caused by hypermethylation (Sekido, 1998). However, unlike the classic two hit hypothesis for disruption of a gene, methylation is more subtle as the loss of gene expression is related to methylation density. Since epigenetic modifications involve transcriptional inactivation of alleles without nucleotide sequence alterations, the changes are undetectable by genome-wide fingerprinting techniques such as RAPD-PCR or C G H . 1.5 Smokins and Luns Cancer Unlike other malignancies, the majority of lung cancer prevalence is largely attributed to an environmental factor such as cigarette smoking. Tobacco smoke is known to contain thousands of substances including carcinogens and tumour promoters. Carcinogenesis of the lung involves the formation of covalent D N A adducts causing D N A mutation and misreplication. One potent carcinogen, benzopyrene diol epoxide has been shown to bind preferentially to regions of the p53 gene. Ahrendt et al report that 53% of 105 N S C L C patients who were smokers had p53 mutations. Cigarette smoking has also been associated with an increased incidence of K-ras mutations and chromosomal loss at the FHIT gene locus. It was reported that allelic losses and gains are present more often at chromosome arms 3p, 6q, 9p, 16p, 17p, and 19p in smokers than non-smokers (Sanchez-Cespedes, 2001). It is estimated that -90% of all newly 9 diagnosed lung cancer cases will occur in patients with a prior history of smoking. When compared to non-smokers, heavy smokers have a 15 to 25 fold greater risk of dying from lung cancer (Ahrendt, 2000). There is some evidence of familial inheritance of lung cancer susceptibility genes in a Mendelian co-dominant fashion. It has been reported that there is a 6.1 fold increase in risk among relatives. Unfortunately, no gene has yet been identified. 1.6 Current Methods for Genome-wide Scanning of Alterations 1.6.1 Comparative Genomic Hybridization Comparative genomic hybridization (CGH) is a genome-wide scanning assay that identifies gene copy number differences at various chromosomal regions. The technique is based on the hybridization of two differently labeled whole genomic D N A probes, tumour D N A and normal DNA, to normal metaphase chromosome spreads. Probes are visualized by fluorochromes and signal intensites are measured and reported as either increased or decreased D N A content at a specific chromosomal region. C G H analysis has identified regions of chromosomal imbalances as gains, losses and amplifications of D N A in various solid tumours. Taguchi et al. reported D N A amplification detected by C G H at 7pl2-13, 8q24, 9p24, and 10q22 in a NSCLC cell line. Michelland et al. (1999) reported observing more gains than losses in N S C L C as detected by C G H . The chromosomal gains included lp, 2p, 7q, 8q, 15q, 17q, 19p, 19q, 20p, 21q, and 22q. Chromosomal losses were observed at 3p, 9p, lOp, 13q and 17p. Figure 1 shows a comprehensive study of 11 neuroendocrine (NE) tumours and 11 N S C L C tumours and the chromosomal imbalances seen in these patients. However, C G H has a limited capacity in detection. C G H is unable to provide a precise chromosomal location as 10 II Simoom am 2 m x m a X IESD ^ CO B. r 3 - 'IHIMU1 0 0 <riBirnMMlJ <N am) 2] o <N H O D D G S i S i i aramnm \o a ^ -fi as w w § 0 •r u « rs O to a .c 0 *c 3 &> * H O CO n <D a> , C o -2 l > GO «s s = (J ITT O = <D G . -73 as o > O CO CD <D co r 3 o a o •1—1 <D C <D ft* s <D <D 5-1 aS o 3 I w O co u « u 4=1 O " C O ofi co is .2 ta 03 co a CO 2 5 5 « 3 S 3 aJ J2 S 3 co - C O g o ^ - J u C/3 s x O u S-i /—s CD * H CD o3 co <D • 1—1 CO g co \ £ <D oJ CO CD j3 g CD <D O & CD < CD s o CO O g g fc! 3 o ^ O CD CD *eib 5 3 .S O CO CO resolution of genetic changes is limited to a chromosome band, at best, because the detection involves microscope-based visualization of chromosome spreads. Also, signal intensities involving a two-fold change representing a loss of a chromosomal region are difficult to discern using fluorophores. 1.6.2 Loss of Heterozygosity (LOH) Loss of Heterozygosity (LOH) is a PCR based strategy for detecting regions of chromosomal instability using polymorphic microsatellites, nucleotide repeats, distributed throughout the genome. These polymorphic sites are generated by the differences in D N A sequence between maternal and paternal chromosomes at many genetic loci. The PCR reaction amplifies these polymorphic sites and they are represented as two maternal and paternal alleles on a polyacrylamide gel (Figure 2). The two alleles are different in their base pair composition attributed to the polymorphism at the loci. By comparing normal D N A versus tumour DNA, one can determine allelic imbalance at a specific loci. The imbalance can arise either as a deletion of the normal allele or a duplication of the mutant allele. In any event, the allelic imbalance is represented as a signal intensity ratio change between the two alleles in the normal D N A versus the ratio of the two alleles in the tumour DNA. Genetic alteration may be viewed as allelic imbalance. Nonrandom distribution of allelic imbalance is due to the presence of both oncogenes and tumour suppressor genes within chromosomal arms. Microsatellites have been effective in allelotyping due to their abundance, hypervariability, high rate of polymorphism and genomic distribution. Girard et al. report 12 C 2 HQ P * © ft «5* & W " CD • • « CD P H a © I 6 8 1 3 CTQ 03 ft ft p i & O CD S CD c o V ! o ft "3 3 i O o X o v I f o he) •§3 2-pt o v ; ? 3 2 03 Ha 3 CO 09 o c o CD O 5 " 8. 3 O CO i-t • O > £ S ft ® IS o §•* 3 3 cr 03 CO CD 3 o •1 3 I I I I I I I I I I I H c 3 o e IT tr £ 4 U I I I ft I I I 1 1 -tr tr I H d so 00 55 13 p w C Q CTQ ^ o n O CD M o CO O B P 13 C/5 I o o 1—1. « Z B 3 _ cr (IQ p 2 p & O I' g ^ n 8 O a s. ff o CD o o n £ B a j=t o ^ o B cr. P CD o o g . a o <*> i-t • o > S CD CD o # B B cr P B 2 CD g C/a CD O 25 o •1 3 SO H c 3 o c i:i I I I ft • ; i 1 11 1 1 1 2 H in -a AO 00 13 allelic imbalance at lp, 3p, 4p, 4q, 5q, 6q, 8p, 9p, lOp, lOq, 13q, 15q, 17p, 18q and 19p (Shivapurkar, 1999; Wu, 1998; Wistuba, 1999; Nomoto, 2000). Zhou et al demonstrated that L O H could be used to determine tumour behaviour as increased allelic imbalance correlated with high frequency of tumour invasiveness and reduced patient survival. Like C G H , L O H has many limitations that hamper its ability to screen the entire genome in search of novel regions of chromosomal instability. Firstly, each microsatellite marker is informative at only one locus making it inefficient in cases where limited microdissected D N A is available. Secondly, although microsatellite markers are distributed throughout the genome, some chromosomal regions do not contain markers preventing L O H analysis. Thirdly, using known polymorphic markers discovered in other cancers prevents discovery of novel regions of chromosomal instability. 1.7 Archival Tissue Use in Search of Novel Markers Archived paraffin-embedded tissues (PET) are a vast resource of material that is being increasingly used for retrospective studies. Since cancers develop through the accumulation of somatic genetic changes that accompany histopathological progression from normal epithelium to invasive neoplasia, it is important to have tissue material from many patients and many histopathological grades to link an initial gene mutation critical to increased risk of progression thereby identifying prognostic markers. Unfortunately, D N A quality from PET is influenced by several factors: fixative type, fixation time, and storage time. As it is believed that fixation is the primary source of D N A degradation in PET, Greer et al. (1994) evaluated a broad set of 14 fixatives and fixation time to discover conditions that are the least degradative. Acetone or 10% neutral-formalin were observed to be the best fixatives with minimum fixation times. Further, acetone was observed to be best fixative for long-term storage. As many PET samples are degraded, determination of D N A quality and quantity is important. Researchers generally assess D N A quality using gene specific amplification of varied size fragments. Assays of this type tend to waste non-renewable D N A as multiple reactions are necessary to determine the extent of D N A degradation. D N A quantitation of minute amounts of D N A is important in any study using archived paraffin-embedded tissues. Researchers use a variety of methods to measure nucleic acid concentrations including absorbance at 260nm (A260) (Sambrook, 1989), fluorescent nucleic acid stains such as PicoGreen (Molecular Probes, Eugene, OR) (Serth, 2000) and Southern blot hybridization (Sambrook, 1989). Nucleic acid quantitation using A260 measurements is disadvantageous due to the assay's sensitivity limits, contamination interference by proteins and inability to distinguish D N A from RNA. PicoGreen quantitation of D N A is more sensitive and efficient but the technique is still influenced by contamination. Southern blot hybridization is efficient at quantitating D N A and is not influenced by contamination as the quantitation is dependent on DNA: D N A hybridization. 1.8 Tissue Heterogeneity and Microdissection Pure cell populations represent the new focus of molecular analyses as native tissue environments have a high level of heterogeneity that affect cell specific studies. Unlike in the past when tissue studies involved crushing fresh tissue and assaying the 15 extracted molecules, isolating pure cell populations is more complicated as tissue is composed of a large number of various cell types organized in a three dimensional structure. For example, a lung biopsy of carcinoma may contain many cells including endothelial cells and fibroblasts in the stroma, normal epithelium, mild to severe dysplastic cells, premalignant carcinoma in situ lesions and clusters of invasive carcinoma cells. In any study, i f the genetic focus is harvesting invasive carcinoma cells, this subpopulation may only represent a small fraction of the total tissue mass. Conducting genetic studies on one population of cells using tissue material that is contaminated by 80% with other cell subpopulations results in the masking of many observations. In the past, pure cell populations were studied by culturing specific cells from fresh tissue. Although this afforded the researcher with pure cells, it hampered native gene expression studies as cultured cell gene expression was influenced by the culture environment. In order to overcome the many problems surrounding tissue heterogeneity, microdissection was developed as a technique to separate individual cell populations. Microdissection involves the manual guidance of a needle to scrape cells of interest from a thin tissue section. With the advent of technology, there are many new devices such as the Laser Capture Microdissection (LCM) system that fuse the technique with a laser-assisted platform to make microdissection less operator dependent (Simone,1998). 1.9 Visualization of Morphology within Biopsies The identification of benign, premalignant and malignant lesions is important in a progressive comparison of tumourigenesis. As most target lesions appear as small foci of cells, microdissection is important to overcome tissue heterogeneity. 16 Microdissection requires the precise identification of specific cellular features by histological stains such as haematoxylin and eosin (H&E) to be successful. Haematoxylin and eosin staining is the most common technique used in routine pathology. The basic dye, haematoxylin, stains acidic structures blue whereas the acidic dye, eosin, stains basic structures pink. Nuclei, ribosomes and rough endoplasmic reticulum have a high content of nucleic acids so the structures stain blue. Cytoplasmic proteins are basic therefore the cytoplasm stains pink. Although the stains distinguish morphology, there is a concern about the effects of the stains on tissue sample contents. Using an assay for D N A amplification by PCR, Murase et al report that several stains including hematoxylin have an adverse effect. It was proposed that the stains interfere with proteinase K digestion by binding D N A or influence the divalent cation concentration critical to Taq polymerase activity (Murase, 2000). A similar study confirmed that haematoxylin is detrimental to tissue for PCR amplification and that two stains, methyl green and nuclear fast red do not interfere with D N A amplification while maintaining their ability to distinguish chromatin within the nucleus (Burton, 1998). However, it is unclear i f the integrity of D N A influences D N A amplifications. 1.10 Use of RAPD-PCR to Search for Novel Genetic Alterations Although locus-by-locus allelotyping techniques such as L O H have been effective in discovering critical genetic events in cancer tumourigenesis, a technique that would screen multiple loci would be more appropriate for a genome scanning study. Developed by Williams et al., Randomly Amplified Polymorphic D N A (RAPD)-PCR (commonly 17 known as Arbitrarily Primed PCR (AP-PCR)) is a D N A fingerprinting technique that involves the PCR amplification of random fragments of genomic D N A with short oligonucleotides of arbitrary sequence (Micheli, 1994). The PCR amplification can be visualized as two short primers, generally 10 nucleotides in length, that anneal to many complementary sites under low stringency throughout the genome and providing that the primers are within a few hundred base pairs of each other and on opposite strands, amplification occurs (Figure 3). When amplification does not occur, the absence of product is either due to mutation, insertion or deletion of the D N A region. Polymorphisms between individuals are detected as D N A fragments that are amplified from one individual but not from another resulting in differences in the pattern of D N A fragments. A large number of polymorphic markers can be obtained by simply changing the primer sequence without previous sequence knowledge of any loci. RAPD-PCR is a very powerful technique used for a variety of purposes such as gene mapping, population and pedigree analysis, phylogenetic studies and the identification of bacterial strains. For example, RAPD was used to identify markers linked to the Pseudomonas resistance gene in the tomato (Martin, 1991), perform single-tree genetic linkage mapping in conifers (Tulsieram, 1992) and the estimation of outcrossing rates in Dastica glomerata (Fritsch, 1992). RAPD-PCR simultaneously reveals about 50 arbitrarily sampled sequences per reaction when displayed on a denaturing polyacrylamide gel, revealing mutations that accumulated either somatically or differences in the relative abundances of corresponding sequences (Welsh, 1995). RAPD-PCR sensitivity is related to the types 18 61 CD CJ of genetic alterations. There are three ways that mutations affect RAPD-PCR fingerprinting: altering the ability of the primer to anneal, altering the distance between the two primers and altering the relative amplification targets. Mutations that affect primer annealing result in the gain or loss of a band or a change in signal intensity. Single base mutations in the primer-binding site are more critical in shorter primers and in the 3' end of the primer. Deletions and insertions that alter the distance between primer binding sites result in the mobility shift of products or the loss of amplification, however, the detection of single base mutations is not possible (Williams, 1993). Ploidy is detectable as the relative band intensity is representative of corresponding template sequence. For example in heterozygotes, loss of a region would result in a loss of band as the only allele is lost. In homozygotes, a loss of a region would result in a 50% change in band intensity as one allele is lost. RAPD is a powerful tool for the detection of somatic genetic alterations during tumourigenesis. RAPD is applied to the detection of qualitative (structural) and quantitative (aneuploid) genetic alterations (Welsh, 1995). R A P D has had applications in cancer research because of its semi-quantitative ability to detect copy number changes from chromosomal aberrations such as gains and losses of D N A sequence. The differences in RAPD band intensity between normal and tumour D N A provide an estimation of tumour aneuploidy. For example, Kohno et al. discovered a homozygous deletion on chromosome 2 in a lung cancer cell line. Losses and gains of D N A sequence can be identified because of the linkage to oncogenes and tumour suppressor genes. Ong et al. (1998) examined genomic instability in N S C L C and SCLC using several RAPD primers. It was found that genomic instability was 20 detected in 19/20 normal/tumour pairs. De Juan et al. (1999) analyzed 65 N S C L C patients in search of genomic alterations and found that 64% of all alterations were gains (increased band intensity) rather than losses (lowered or no band intensity). Kawakami et al.(1998) fingerprinted 44 NSCLCs using AP-PCR to find a loss of band intensity in 15 tumours that corresponded to the long arm of chromosome 10; the loss was confirmed with allelic imbalance analysis. De Juan et al. (1999) screened 65 N S C L C patients using AP-PCR to find that 45% of the sample set contained an amplified D N A fragment at 6pl2. Anami et al. (2000) adapted the AP-PCR technique to fingerprint methanol-fixed lung cancer tissues and discovered D N A sequence losses at chromosome 7 and 22 in 41.7% of adenocarcinomas and 84.6% of small-cell carcinomas, respectively. Chromosome 22ql3.3 losses were confirmed with microsatellite markers in 11 of 13 small cell carcinomas. They also reported D N A amplifications at chromosome 1, 8, 13 in 40% of adenocarcinomas and chromosome 2 in 63.3% of squamous-cell carcinomas. Yamada et al. (2000) used AP-PCR to detect chromosomal imbalances in 13 SCLCs and discovered gains and losses corresponding to chromosomes 1, 7, 8, 16 and chromosomes 2, 10 and 22, respectively. The most common band intensity gain corresponded to a D N A amplification at 8q24. Amplification of the C - M Y C gene was confirmed in five of eight tumours analyzed, suggesting that another potential oncogene was amplified in the region. RAPD-PCR is an effective means of determining chromsomal alterations. 21 1.11 Objectives and Hypothesis The first objective of this thesis was to deduce suitable conditions for the microdissection of formalin-fixed, paraffin-embedded archival lung tissue. The rationale was to identify a histological stain conducive to maintaining D N A integrity while enabling morphological distinctions within the biopsy. The second objective was to isolate D N A from progressive stages of lung carcinoma in order to conduct a stage-specific study. In collaboration with Dr. Stephen Lam at the BC Cancer Agency, a large cohort of biopsies from a large number of patients was obtained and microdissected. The third objective was to identify genetic alterations using a genome scanning technique known as RAPD-PCR. Hundreds of loci were analyzed using this technique. The fourth objective was to characterize genetic alterations by cloning and sequencing in hope of identifying critical genetic events in lung tumourigenesis. The fifth objective was to localize sequenced products to chromosomal regions and the sixth objective was to verify these regions for chromosomal instability. These objectives were created in order to support the two-fold hypothesis: 1) RAPD-PCR will identify genetic alterations in NSCLC by simultaneously screening multiple loci from specific grades of archival, paraffin-embedded lung tissue and 2) evaluation of alterations will reveal chromosomal regions of instability. 22 Chapter 2: Materials and Methods 2.1 Tissue Microdissection and DNA Extraction A total of 172 paraffin-embedded archived lung samples were provided by Dr. S. Lam. This set included 33 normal, 16 hyperplasia, 24 mild to severe dysplasias, 74 carcinoma in situ (CIS) and 25 invasive carcinomas. In all, 41 patients were represented. The patient cohort consisted of 27 men, 14 woman, 14 current smokers, 21 ex-smokers, and 4 non-smokers with a mean age of 69 and a mean smoking history of 58 pack years. Smoking history was not available for two patients. The rationale for the patient cohort was to include all available lung biopsies. A l l samples were histologically graded by Dr. J. LeRiche. Serial sections were cut from each paraffin block and stained with Methyl Green. Every seventh section was stained with hematoxylin and eosin and evaluated by the pathologist. Manual microdissection was performed by C. Dawe and myself. Tissue was digested in a digestion buffer (lOmM Tris, ImM EDTA pH 8.0, 0.5% sodium dodecyl sulfate (SDS) and 50mM NaCl) at 55°C for 72 hours with a fresh 20ug proteinase K spiking every 24hours. D N A was extracted twice with phenol/chloroform/isoamyl alcohol followed by precipitation with ethanol using lOpg glycogen as a D N A carrier. D N A material was extracted from blood in exactly the same manner once the cellular components were pelleted. 2.2 DNA Quantification A 5% aliquot of unqualified D N A sample was run out on a 1% agarose gel along with known quantities of DNA, denatured and transferred to HyBond nylon membrane 23 (Amersham Pharmacia Biotech). The membrane was dried at 80°C for 30min and probed with 200 nanograms (ng) total genomic lung DNA. The probe D N A was labelled using a Random Priming Kit (Gibco B R L Inc.) as follows: 200ng denatured DNA, 14mM of each dCTP, dGTP and dTTP, I X Random Priming Buffer (0.67M HEPES, 0.17M Tris-HC1, 17mM M g C l 2 , 33mM 2-mercaptoethanol, 1.33 mg/ml BSA, 18 OD 2 6o units/ml oligodeoxyribonucleotide primers, pH 6.8), 25 uGi a 3 2 P ATP. The probed membrane was washed in 2X SSC (5mins), 2X SSC, 1% SDS ( 3 x 1 5 mins) and 0.2X SSC, 0.1% SDS (2 x 30mins). The membrane was exposed on Kodak X - O M A T A R autoradiography film and on a phosphoimager cassette for further analysis. The cassette was analyzed by the STORM phosphoimager and using ImageQuant 5.0 software (Molecular Dynamics), the D N A concentration was determined. 2.3 Gene-specific PCR Two sizes of fragments were amplified from the human glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene (GenBank accession no. J04038) in 25 ul reaction mixtures containing: I X PCR Buffer (50mM KC1, lOmM Tris-HCl pH 9, 0.1% Triton X-100), 2 m M M g C l 2 , 0.2 m M deoxyribose trinucleotides (dNTPs), 20 pmoles of each G A P D H primer, 2.5 Units Taq D N A Polymerase and 2 ng of DNA. The PCR primers were: GAP1 (5' - A A C C T G C C A A A T A T G A T G A C A T C A-3'); GAP2 (5'-G T C G T T G A G G G C A A T G C C A - 3 ' ) ; and GAP3 (GTCTTACTCCTTGGAGGCCATGA-3'). GAP1 and 2 produce a 163 bp fragment while GAP1 and 3 produce a 363 bp fragment (see Results, Fig.7A). PCR conditions were as follows: Denaturation at 94°C for 2 min, followed by 35 cycles of amplification (94°C for 45 sec, 60°C for 45 sec and 24 72°C for 2 min) and a final extension (72°C for 10 min) in a PTC-100 Thermal Cycler (MJ Research). 2.4 RAPD-PCR Each RAPD-PCR fingerprint consisted of multiple patients of various histological grades analyzed with a favorable primer pair. A l l oligonucleotide primers were purchased from Alpha D N A Inc. In each reaction, 20 picomoles of each primer were end-labelled with 2uCi of adenosine 5' [y-32 P] ATP (6000 Ci/mmol; Amersham Pharmacia Biotech), at 37°C for 1 hr and 65°C for 5 min. Each PCR reaction contained 2 ng of D N A , 200uM each dNTP, lOmM Tris-Cl, pH 8.3, 50 m M KC1, 2mM M g C l 2 , 0.001% gelatin, and 2.5 units of Recombinant Taq D N A polymerase in a final volume of lOul. A l l reactions were amplified for 45 cycles (94°C for lmin, 35°C for lmin, 72°C for 2min) in a thermal cycler (MJ Research PTC-100). A l l reaction products were run on 6% non-denaturing polyacrylamide gels at 800V for 3 hrs. Dried gels were exposed on Kodak X - O M A T A R autoradiography film. In each experiment, positive and negative controls consisted of D N A extracted from a frozen lung tissue and a reaction without D N A template, respectively. A l l noticeable PCR signal changes that recurred in many patients were further analyzed. 2.5 Cloning Fragments of interest were cut out of the dried polyacrylamide gel after overlaying the exposed film against the dried gel with orientation markers. D N A was eluted into 20ul d H 2 0 by boiling for 10 minutes. Ten percent of eluted material was amplified using primers matching the original RAPD-PCR primer sequence but containing a Bam HI or PST1 recognition sequence. As the RAPD-PCR fragments were 25 created using two possible primers, it was important to determine the terminal sequence of the products. Therefore, the reamplification reactions consisted of three possible primer combinations: primer 1 with itself, primer 1 and primer 2, and primer 2 with itself. Aliquots of PCR products were resolved on 1% agarose gels to determine which restriction enzyme was needed to prepare the product for insertion into a plasmid vector. PCR reactions contained I X Mg-free PCR buffer (50mM KC1, lOmM Tris-HCl pH 9, 0.1% Triton X-100), 0.5mM dNTPs, 2 units recombinant Taq polymerase, 1.9 m M M g C l 2 and 10 picomoles of each primer. A l l reactions were performed using the M J cycler. 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 by the appropriate enzyme (Bam HI or Pst 1) and purified by excising from a 5% polyacrylamide gel. D N A was eluted from the excised gel piece using dHiO at room temperature. The insert was quantified and combined with cut and dephosphorylated P S K + plasmid for subsequent ligation at 16°C overnight. Transformations involved combining ligation with heat shock competent DH5ct E.coli. cells (Bergmans, 1981). Cells were plated on L B Amp + plates containing IPTG and X -Gal and grown at 37°C overnight. Growing colonies containing an insert were white while colonies without an insert were blue. Ten white colonies (as a minimum) were subjected to colony PCR to confirm that the appropriate size insert was present in the plasmid. Colony PCR contained I X Mg-free PCR buffer (buffer (50mM KC1, lOmM Tris-HCl pH 9, 0.1% Triton X-100), 0.25mM dNTPs, 2 units recombinant Taq polymerase, 5 picomoles each of M l 3 -47 and 26 -48 primers (New England Biolabs) and 2mM M g C l 2 . Colony PCR reaction involves an initial denaturation at 94°C for 3 mins followed by 30 cycles at 94°C for 45 sees, 40°C for 45 sees and 72°C for 45 sees. Colonies that had appropriate size inserts were subjected to colony fingerprinting to determine that the colonies had the same sequence. Colony fingerprinting reaction contains I X FP nucleotide stock (0.5mM dNTPs, 0.5mM ddGTP), I X FP buffer (lOmM Tris pH 8, 50mM KC1, 1.5mM MgCl 2 ) , 0.5 pmol 3 2 P ATP labelled T7 primer, 2.5 units recombinant Taq polymerase, and lug RNase A . The T7 primer radiolabelling reaction contains 0.34 pmol y P dATP, 1 X PNK buffer, and 0.5 pmol T7 primer and involves incubation at 37°C for 1 hour followed by 65°C for 5mins. Colony fingerprinting reaction involved an initial denaturation at 94°C for 3 mins followed by 30 cycles at 94°C for 1 min, 40°C for 1 min and 72°C for 1 min. (Krishnan, 1991). Plasmid preparation was performed on one representative colony (see Appendix for detailed protocol). The prepared plasmid was sequenced using the A B I Prism at the Nucleic Acids Protein Service (NAPS) Unit at the University of British Columbia (see Appendix for detailed protocol. 2.6 Localization of clones to a chromosomal position Sequenced clones were matched to the partially sequenced human genome using NCBI standard nucleotide-nucleotide B L A S T (blastn) at [www.ncbi.nlm.nih.gov1. Matched results confirmed chromosome locations for each clone. 2.7 Choosing Microsatellite Markers In order to test a cloned region for allelic imbalance, microsatellite markers must be determined for each chromosomal region. Using the University of Southern 27 California genome database [] to identify microsatellite markers surrounding the sequenced clone region and the Genome Database [http://www.gdb.orgl to identify primer sequences and frequency of heterozygosity, microsatellite primers were chosen and ordered from Alpha D N A Inc. 2.8 Loss of Heterozygosity (LOH) Assay One primer (2ng) from each L O H primer pair was end-labelled with 20u,Ci y 3 2 P ATP using T4 polynucleotide kinase (PNK) at 37°C for 1 hour followed by 65°C for 5 mins. PCR amplification involved 4 ng genomic DNA, 2 ng labeled primer, 20 ng of each unlabelled primer, 0.5mM of each dATP, dCTP, dGTP, 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 35 cycles consisting of denaturation at 95°C for 30 sec, annealing at 50-60°C for 60 sec, and extension at 72°C for 60 sec. PCR products were separated on 6% urea-formamide-polyacrylamide gels run at constant power (50W) for 2.5 hours. The dried gels were exposed on Kodak X - O M A T A R autoradiography film. The gels were analyzed by determining whether there was a signal intensity change for the two alleles seen in the normal D N A compared to tumour D N A . A signal intensity decrease of 50% was classified as an allelic loss. A l l samples showing allelic loss were subjected to repeat analysis after a second independent amplification. 28 Chapter 3: Results 3.1 Determining a Patient Pool In order to determine prognostic markers for lung cancer, a disease that affects hundreds of thousands of individuals each year and develops over many decades, it is important to establish an unbiased, patient set. Ideally for a complete genetic study, the patient set should include all relevant information including age, sex, smoking history, recurrences and treatment. The patient pool consisted of non-small cell lung carcinomas (NSCLCs) collected during fluorescence bronchoscopy by Dr. Stephen Lam (chair of the lung tumour group at the B C Cancer Agency) using Lung Imaging Fluorescence Endoscope (LIFE) system (Xillix Technologies Corp.). A l l biopsies were classified by a lung pathologist, Dr. Jean LeRiche, according to a histological grading system (Table 1). Biopsies were fixed in formalin, paraffin-embedded, and mounted on glass slides by Chris Dawe. Staining was performed as described in "Materials and Methods", (see Appendix for staining protocol). Metaplasia is characterized as a maturation step where cells appear to be changing from one histological cell type to another. As seen in Table 1, the two possible grades for each early-stage pathological class are distinguished by the presence of absence of metaplasia (eg., 3.1 or 3.2, 4.1 or 4.2, 5.1 or 5.2 and 5.3 or 5.4). Cells that appear squamous are arranged in a flattened epithelial appearance whereas glandular appearance exhibits cells in a gland-like fashion. The two possible grades seen in the progressed 29 stages distinguish between squamous or glandular appearance (eg., 6.1 or 6.2, 7.1 or 7.2 and 8.1 or 8.2). Table 1: Relevant Histological Grades in N S C L C Pathological Class Cellular Features Grade Normal Pseudostratified ciliated columnar epithelium 1.0 Hyperplasia Increase in the number of normal-appearing basal cells with normal ciliated or mucin secreting cells 3.1,3.2 Mi ld Dysplasia Stratified epithelium, hyperchromatism, slight increase in nuclear:cytoplasmic ratio, limited to lower 1/3 of epithelium 4.1,4.2 Moderate Dysplasia Stratified epithelium, hyperchromatism, larger increase in nuclear: cytoplasmic ratio, mitotic figures seen, limited to lower 2/3 of epithelium 5.1,5.2 Severe Dysplasia More pronounced changes than seen in moderate dysplasia 5.3, 5.4 Carcinoma in situ Multi-layered epithelium with malignant cells but with an intact basement membrane 6.1,6.2 Microinvasive Cells have breached the basement membrane into the subepithelial layer; confined within layer above cartilage 7.1,7.2 Invasive Carcinoma Multi-layered epithelium with malignant like cells invading the underlying stroma 8.1,8.2 3.2 Determining a Method for Obtaining Pure Cell Populations Lung tissue is composed of various types of cells including epithelial, stromal, and inflammatory cell populations. Previous studies use gross tissue samples for genetic analyses risking the masking of specific genetic events by heterogenous tissue. In order to define stage specific markers for tumour progression, tissue heterogeneity had to be addressed. Microdissection was considered to be the only method for procuring specific cell populations. Two microdissection techniques were considered: manual microdissection and laser capture microdissection (LCM). 30 3.2.1 Manual Microdissection Guided by many previously published studies using manual microdissection, an inverted Nikon microscope fitted with an X - Y - Z micromanipulator was made available by Dr. Calum MacAulay. Although many reports mentioned the use of a needle attached to a micromanipulator, we decided to practice the technique using our own manual dexterity. After several hours of practice and experimentation with various objects and needle sizes, it was determined that a 30 gauge needle was sufficient to dissect individual cells. Manual microdissection was attempted using various solutions such as water, buffer and alcohol. The effect of the solution was to provide a moist medium to prevent the static repulsion of the tiny tissue flakes from the needle. It was determined that the 'wetting' of the slide with 80% EtOH prior to immediate dissection was most appropriate. In fact, not only did the alcohol provide moisture, as the alcohol dried on the slide it would cause the scraped tissue to adhere to the needle allowing safe transfer to the digestion vial. Figure 4 shows an example of two tissue biopsies before and after microdissection. In figure 4A, hyperplastic cells are present in three distinct areas in the epithelium. Figure 4C shows the biopsy after the removal of the hyperplastic area using manual microdissection. It is clear that the basement membrane is still intact confirming the lack of contamination by nonepithelial cells. In figure 4B, there is an invasive tumor within a deep layer of stroma. The CIS cells are microdissected and the biopsy post-dissection is seen in figure 4D. 31 Figure 4: Manual Microdissection of Lung Biopsies. A . Lung biopsy (before microdissection) with hyperplastic regions. B. Lung biopsy with invasive tumor. C , D. Lung tissue remaining after microdissection. Intact basement membrane in C. indicates no contamination by nonepithelial cells. 32 3.2.2 Laser Capture Microdissection (LCM) Since L C M was promoted to be much more advantageous than manual microdissection, the use of PixCell II (Arcturus), kindly provided by Dr. Juergen Vielkind (Cancer Endocrinology, BCCA) was considered in this study. L C M is based on the fusion of cells of interest to a plastic polymer film via melting using an operator-guided laser. Several trial experiments were preformed to compare the efficiency of the technique versus manual microdissection specific to these small tissue biopsies. It was concluded that in order to maximize D N A quantity recovered from each biopsy, the tissue material was to be freely digested in a buffer as opposed to being digested while adhering to a polymer film. Experiments to optimize D N A yield were not performed since manual microdissection was adequate for obtaining pure cell populations. 3.3 The Effects of Histological Stains on DNA Quality D N A quality remains an important issue in many genetic studies. Lung biopsies used in this study are classified as archival material because they have been formalin-fixed and paraffin-embedded immediately after excision from the patient. A common fixative, formalin has been known to cause the nicking of D N A and subsequent degradation. Despite these poor preservation techniques, the present concern is to maintain the level of degradation without further destruction of DNA. Although it is important to visualize cell morphology within these biopsies, common histological stains such as H & E have a low pH and could be detrimental to D N A quality. To examine the effects of histological stains on DNA, a time trial was conducted where cultured Bronchial Epithelial Cells (BEC) were stained with Hematoxylin, a component of H & E , for various time lengths similar to common biopsy staining 33 practices. D N A was extracted and an aliquot was run on a 1% gel. (Figure 5A) The lengths of exposure to Hematoxylin stain at its natural pH 2.59 were 1, 10, 30, 90 and 270 seconds. It is clearly shown that with increased time exposure, there is a prominent degradation of the high molecular weight genomic band seen in the unstained BEC control with increased accumulation of smaller degraded D N A material. A similar time trial was conducted using Methyl Green stain, pH 4.18, on BEC (Figure 5B1). The high molecular weight genomic band was present in the unstained BEC control. After staining the BEC cells for 1 sec, there is noticeable degradation of the D N A ; D N A degradation accumulates with increasing staining lengths. Believing that the degradation was solely caused by the low pH, we decided to neutralize the Methyl Green stain to pH 6.0 and studied the effects on D N A quality (Figure 5B2). The previous experiment was conducted using the buffered stain and the D N A was analyzed for D N A quality. The modified stain did not affect D N A quality as all staining results showed the presence of the high molecular weight genomic band without accumulated degradation. As most histological stains are used at an acidic pH, it was important to examine whether the acidity is necessary for proper stain binding by comparing histological stains at various pH's. Three stains, H & E , Nuclear Fast Red and Methyl Green at pH 2, 4, and 6 were used to stain serial sections of a lung biopsy (Figure 6). The stained sections were examined by Dr. LeRiche for the ability to distinguish varying morphological species. Staining the biopsy with H & E (pH 2.59), the nuclei appeared blue and the stroma appeared pink. With increasing pH, the H & E stain was unable to distinguish nuclei from stroma becoming a homogenous stain. Nuclear Fast Red (pH 2.17) staining of the biopsy allowed cells to be classified according to morphology but with increasing pH, the stain 34 a> -w " " - n fa C <N .S CO CD c o O c o a J (D T3 53: H9 NFR pH 2.17 — •• ' i i ' s • ' MG pH 4.18 NFR pH 4.0 l i t e ~ . ; N>:;« . (i, . 4..' -J , 1 -— u i . MG pH 6.0 NFR pH 6.0 Figure 6: Determining Staining Ability of Histological Stains. Serial sections of lung biopsy stained with Hematoxylin & Eosin (H&E), Nuclear Fast Red (NFR) and Methyl Green (MG) at various pH's to assess effect on staining ability. M G pH 6.0 was the optimal stain. 36 became indiscriminate between cell types. Staining the biopsy with Methyl Green at its natural and buffered pH showed similar staining pattern allowing for morphological characteristics to be distinguished. Since Methyl Green at the buffered pH appeared to retain its staining ability and was not degradative to DNA, it was chosen for use with all microdissectable sections. 3.4 Developing an Assay for Assessing DNA Quality of Microdissected Archival Samples Archival specimens vary considerably in D N A quality due to handling and storage techniques. Some D N A samples are highly fragmented or damaged by preservation procedures and require special handling in reactions, for example, the use of larger quantities of D N A in a PCR based assay, while other samples seem to be better preserved, readily yielding amplified products of larger size. Since specimen size is often a limiting factor in studying non-renewable, archival material, the ability to assess D N A quality, with a minimal amount of material, would allow the sorting of such specimens for specific analytical use. RAPD-PCR enables consistent genetic fingerprints from archival, microdissected specimens by the random amplification of multiple loci throughout the genome producing an unbiased representation of D N A quality Using seven arbitrarily chosen oral squamous cell carcinomas provided by the Oral Biopsy Service of British Columbia, a trial experiment was designed to determine the applicability of RAPD-PCR as an alternative to gene specific PCR to determine the D N A quality of a sample prior to microdissection and further analyses. 37 3.4.1 Assessment of DNA Using Gene Specific PCR D N A extracted from seven formalin-fixed, paraffin-embedded oral biopsy pairs (14 samples in total) was assessed by the amplification of two different size fragments, 163 bp and 363 bp, from the human glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene (Figure 7A). Figure 7B summarizes the results. Although all samples generated a 163 bp fragments only 8 yielded a 363 bp PCR product. The latter included samples from cases 402, 446, 448 and 451. In all cases, D N A extracted from normal cells in connective tissue (C) and tumour cells (T) from each specimen showed the same PCR response suggesting that the PCR limitation was a result of specimen preservation. 3.4.2 Assessment of DNA Quality Usine RAPD-PCR The same quantity of D N A was sufficient to generate RAPD profiles from the above samples (Fig. 7C). The size range of RAPD products varied between sample pairs, again with both members of the same sample pair giving a similar response. Three of the sample pairs (446, 448 and 451) had fingerprint patterns resembling the profile generated from high quality D N A extracted from cryopreserved tissue up to 500 bp. A fourth (402) gave reproducible PCR products up to 350 bp. A l l 4 of these sample pairs gave a 363 bp G A P D H fragment. In contrast, 405 gave a reliable profile up to 200 bp with less amplification of larger sized fragments (lighter bands) and variation between C and T. Specimens 414 and 416 gave even poorer profiles with fragments restricted to <150 bp. Increasing D N A concentration did not improve the quality of the profiles. This explains the absence of a G A P D H 363 bp fragment for these 3 samples as the D N A appears to be highly fragmented. 38 6£ oo + H U H U 5 u i—( -* u m H o ^ U o ^ U A A O H & XI X) -Q o o O O o o O O m "«t m £ 1 CO OH CO OH CO fc n-ON in in -jr 3 2fc a , o o R A P D is an alternative manner of assessing D N A quality and it is advantageous compared to gene specific PCR assessment because it gives a clear representation of D N A quality in one reaction limiting the waste of precious microdissected D N A . As clearly shown, the same assessment requires a minimum of two reactions using gene specific PCR. 3.4.3 Screening Biopsies for DNA Degradation Using RAPD-PCR Prior to microdissection, tissue material was scraped and digested from an unstained section of every biopsy to prescreen the D N A quality. D N A material was extracted and 1/20 of the D N A material was used in a RAPD-PCR reaction. Any biopsies unable to amplify D N A fragments greater than 200 bps were excluded from the study. A l l 180 microdissected cases used in this thesis consist of good quality D N A material as the early pre-selection excluded some biopsies. 3.5 DNA Quantification of Microdissected Archival Lung Biopsies Using Southern Blot Hybridization Many techniques exist for the determination of D N A quantity in specimens. Quantitative D N A stains such as PicoGreen and spectrophotometer absorbance readings (A260/A280) tend to consume large amounts of D N A during quantitation and are influenced by residual nucleic acids and proteins. A Southern blot hybridization assay was proposed as a quantitation method because it was based on the hybridization of D N A to D N A on a membrane; it was more representative of D N A than absorbance readings. A 5% aliquot of each microdissected sample was quantitated against known D N A quantities using a southern blot hybridization assay (Figure 8A). The probed membrane 40 A . Manually Microdissected Lung Biopsies (0.5 ul) D N A Standards (ng) A B C D E F G H I J K L M N O P 2 5 10 20 ImageQuant 5.0, Molecular Dynamics Nanograms of DNA Figure 8: Southern Blot Quantitation Assay of Microdissected Lung Biopsies. A . Five percent of each microdissected sample was probed with D N A and quantitated by comparing signal intensities of unknown quantities of material with signal intensities of known quantities of D N A using ImageQuant. A-P represent microdissected samples, known D N A quantities are 2, 5, 10, 20ngs. B.Graph represents a linear relationship between signal intensity and known quantity of DNA. 41 was analyzed using ImageQuant 5.0 software (Molecular Dynamics) and D N A quantity was determined. The quantitation software expresses a numerical reading for radioactive signal present for each area of a lane. Numerical readings were also assigned to the signal emitted by the known quantities of D N A (2, 5, 10, 20ngs). To relate each numerical reading for an unknown quantity to the numerical reading of known quantity, the unknown quantity numerical reading was expressed as a factor of each known quantity numerical reading. The factor was multiplied by the known D N A quantity and an average D N A quantity was determined. In most cases, the relationship between numerical reading to known D N A quantities was linear (Figure 8B). D N A amount from all microdissected cases was quantitated in this manner. Example: Unknown D N A Quantity Reading X Known D N A Quantity = Unknown D N A Quantity Known D N A Quantity Reading 48671.81 (Unknown Reading) X2ng=1 .82ng 53553.78 (2ng Reading) 3.6 Screening Oligonucleotides for Increased Amplification Density Short oligonucleotides (10 base pairs) match variably throughout the human genome depending on D N A sequence. To achieve proper sampling of the human genome, it was expected that the primer set should amplify > 50 regions. Sequence of thirty-six oligonucleotides (Table 2) was randomly designed with the criteria that each oligonucleotide must have a G:C content similar to the human genome, i.e. 60%. 42 Table 2: RAPD-PCR primers screened for increased scanning density RAPD-PCR Primers Sequence (5' to 3') % GC Content Alpha-1 (9mer) C A G G C T A T C 55.56% Alpha-2 (9mer) G A A T C A G T G 44.44% Alpha-3 (9mer) A A C T G G C A T 44.44% Alpha-4 (9mer) T C C A G C T A G 55.56% Alpha-5 (9mer) T T C A G T C A G 44.44% Alpha-6 (9mer) A A C T G C A T C 44.44% Alpha-7 (8mer) A C G T T G A C 50% Alpha-8 (8mer) A A C T G C C A 50% Alpha-9 (8mer) T T C A G T C A 37.50% Alpha-10 (8mer) A A C G T T C G 50% Alpha-11 (9mer) C G G C A G C A T 66% Alpha-12 (9mer) G G C T A C C A G 66% Alpha-13 (9mer) C G C T T G G A C 66% Alpha-14 (9mer) C G C A T C C A G 66% Alpha-15 (9mer) G C C A T G G C A 66% Alpha-16 (9mer) C T C A G G C A C 66% Alpha-17 G A C G C C G C T T 70% Alpha-18 A C G T C A G G C T 60% Alpha-19 GGCTACTGCT 60% Alpha-20 C G C C T A A T G C 60% Alpha-21 AGGCTCTCGT 60% Alpha-22 C T A G G A T C C G 60% Alpha-23 CGCCTATGGT 60% Alpha-24 G G C A T A C C G T 60% Alpha-25 C A C G G T A A G G 60% Alpha-26 A T T C C G G A C C 60% Alpha 27 CGTATCGCCT 60% Alpha 28 G C A T A C G G C A 60% Alpha 29 A C T G C A C G T C 60% Alpha 30 T C G G A T C C A G 60% Alpha 31 C T A G C A C G C T 60% Alpha 32 A C T G A C C G C T 60% Alpha 33 C G T A C G G C T G 70% Alpha 34 C A C G T G A C G T 60% Alpha 35 C A T G G C T C A G 60% Alpha 36 C C T A G A C C G A 60% 43 Each primer was randomly combined with another in a RAPD-PCR reaction (see Materials & Methods) to screen for primer pairs that target many loci throughout the genome. The expectation was to find a primer pair that amplified > 50 bands in one reaction. Figure 9 is an example of a primer screen gel. Not all primer combinations succeeded in amplifying any PCR products. The primer combination Alpha 31/32 was chosen as the primary set for conducting RAPD-PCR on the lung biopsies. 3.7 Scanning Microdissected Archival Lung Biopsies for Genetic Alterations Using RAPD-PCR In order to identify genetic alterations such as chromosome amplifications, translocations or deletions associated with the development of lung cancer, it is important to screen a patient set that is complete with both histologically normal tissue and progressive stages leading up to tumour development. Using RAPD-PCR, 172 microdissected samples were analyzed. The samples were comprised of 41 different patient sets (see Appendix for patient details). Each patient set analyzed consisted of at least one histologically normal tissue (or blood equivalent) and one carcinoma in-situ (CIS). These two grades were the focus of the analysis. In many cases, there were multiple sites dissected from each patient corresponding to different sources of CIS or tumour. The resolved PCR products on polyacrylamide gels were analyzed in search of PCR products that did not exist throughout a patient set. Interesting alterations were identified and the gels were analyzed to determine i f the same change occurred throughout other patient 44 A B C D E F G H I J K L M N O P Q R S T U V W X 700 bp 400 bp 300 bp I N 200 bp A-27/29 B-27/30 C-27/31 D-27/32 E-27/33 F-27/34 G-27/35 H-27/36 1-28/29 J-28/30 K-28/31 L-28/32 M-28/33 N-30/36 0-31/32* P-31/33 Q-31/34 R-31/35 S-31/36 T-32/33 U-32/34 V-32/35 W-32/36 X-33/34 t * ? * a a • a m I Figure 9: Screening Primer Pairs for Increased Scanning Density. Thirty-six randomly designed 10 bp primers were tested in random pairs to assess the number of SMAL-PCR fragments amplified. Pair 31/32 was chosen for the study because of the increased number of products. 45 sets. As expected, more alterations were seen in the latter stages of progression. Fifteen different recurring band types (alterations) were identified from the entire patient pool. Table 3 categorizes all fifteen alterations according to size of alteration, number of recurrences in different patients, loss (PCR product present in the normal tissue only) or gain (PCR product present in the progressed stage), broad or tight band, faint or strong signal, singlet or doublet and chromosome localization. The number in parentheses denotes the number of patients exhibiting the polymorphism. Figure 10 shows four recurring alterations that mapped to distinct chromosomal regions. These alterations were identified on different gels therefore the band types do not appear perfectly matched across a gel. Alteration A is a 550bp gain in CIS that was seen in two of sixteen patients exhibiting the polymorphism. Alteration D is a 150bp gain in CIS that was seen in two of eighteen patients exhibiting the polymorphism. Alteration N is a 330bp gain in CIS that was seen in two of fifteen patients exhibiting the polymorphism. Alteration B / M is a 380bp gain in CIS that was seen in six of seventeen patients exhibiting the polymorphism. This alteration was also seen in biopsies from two different sites derived within one patient. Alteration B / M was assigned both letter designations because the alterations were initially identified as two separate band types that turned out to be the same alteration after further review. 3.8. Clonins of Genetic Alterations Potentially Associated with Luns Cancer Progression Cloning enables the identification of PCR products by the insertion of foreign DNA, by means of a vector, into bacterial cells allowing increased production of the desired D N A sequence. 46 Name Size (kb) # of Recurrences Loss/Gain Broad/Tight Faint/Strong Doublet/Singlet Chromosome A1 550 2(16) loss tight faint doublet 5q33.1 A2 550 loss tight faint doublet 5q33.1 B1 380 4(17) gain broad faint singlet 8q24.3 B2 380 gain broad faint singlet 8q24.3 B3 380 gain broad faint singlet 8q24.3 B4 380 gain broad faint singlet 8q24.3 C1 340 4 gain tight strong singlet Reamp Failed C2 340 gain tight strong singlet Reamp Failed C3 340 gain tight faint singlet Reamp Failed C4 340 gain tight faint singlet Reamp Failed D1 150 2(18) gain broad strong singlet 1q23.3 D2 150 gain broad strong singlet 1q23.3 E1 140 2 gain broad faint singlet No Match E2 140 gain broad faint singlet No Match F1 260 3 gain tight strong singlet No Match F2 260 gain tight strong singlet No Match F3 260 gain tight strong singlet No Match G1 240 3 gain tight faint singlet Ligation Failed G2 240 gain tight strong singlet Ligation Failed G3 240 gain tight strong singlet Ligation Failed H1 220 loss tight strong singlet Reamp Failed H2 220 loss tight strong singlet Reamp Failed H3 220 3 loss tight faint singlet Reamp Failed 11 650 2 loss tight strong singlet No Match 12 650 loss tight strong singlet No Match J1 340 2 loss tight strong singlet No Match J2 340 loss tight faint singlet No Match K1 410 3 gain tight faint doublet No Match K2 410 gain tight faint doublet No Match K3 410 gain tight faint doublet No Match L1 550 2 loss tight faint singlet Ligation Failed L2 550 loss tight faint singlet Ligation Failed M1 380 3(17) gain tight faint singlet 8q24.3 M2 380 gain tight faint singlet 8q24.3 M3 380 gain tight strong singlet 8q24.3 N1 330 2(15) gain tight strong singlet 7q36.2 N2 330 gain tight strong singlet 7q36.2 P1 130 2 loss broad strong singlet Repeat Element P2 130 loss broad strong singlet Repeat Element Table 3: Fifteen Different Band Type Alterations. A l l alterations were cloned; four alterations (A, B / M , D, N) mapped to distinct chromosomal regions, 5q33.1, 8q24.3, lq23.3, 7q36.2, respectively. Number in parentheses indicates number of cases exhibiting polymorphism. 47 T N N N N N N N N T 57 58 B 27 28 29 30 32 B 61 62 Alteration D (150 bp gain) Alteration N (330 bp gain) N T T N N T N T N N N T T T T T N T N N T N N 24 25 26 B , 27 28 B . 29 30 B 35 36 37 39 40 44 45 46 B B 47 48 B Alterations B /M (380 bp gain) Alteration A (550 bp gain) Figure 10: Recurring Alterations in Multiple Patients. Fifteen different band types were cloned; four band types mapped to four distinct chromosomal regions. Alteration A occurred in 2/16 patients, alteration D occurred in 2/18 patients, alteration N occurred in 2/15 patients and alteration B / M occurred in 7/17 patients. Numbers represent samples, B represents blood, 48 Fifteen band types were cloned as outlined in the Materials and Methods (See Table 3 for details). Alterations B / M are used in this cloning example to demonstrate the procedure. Figure 11A shows the reamplification of the original PCR products ( B l , B2, B3, B4, M l , M2, M3) using modified SMAL-PCR primers with restriction sites within their sequence. The new PCR products contain BamHl and PST1 sites. After cutting the PCR product with the enzymes, the product was precipitated and visualized on a polyacrylamide gel. Figure 1 IB shows the well-resolved PCR products on a polyacrylamide gel. Due to the increased resolution ability of the polyacrylamide, this gel shows many more PCR products that appeared as a single product on the agarose gel. The PCR product matching the appropriate size was excised, in this case, 380 bp. The product was inserted into a P S K + vector containing a p-galactosidase reporter gene and an ampicillin resistance gene. A l l colonies chosen for colony PCR were white because the inserted D N A disrupted the P-galactosidase gene. Figure 12A shows a colony PCR used to demonstrate that the colonies still contain the appropriate size insert. Colonies from plates containing the B and M clones were used in the reaction. In some cases, there is no product in the lane or the product size does not match the expected size. A l l colonies confirmed by colony PCR were tested by colony fingerprint (G-track sequencing) to determine i f all contained the same sequence (See Appendix). Figure 12B shows a colony fingerprint of B and M clones. It is clear that all colonies contain the same sequence. One colony from each set of similar sequenced colonies was grown for plasmid purification (see Appendix). The purified plasmid was sequenced at Nucleic Acids Protein Service (NAPS) Unit. Figure 13 shows the sequence of the original B insert. 49 _ M _ J » _ J j 2 _ B _ 3 B4 M l M2 M3 A . H H H H H H I 4 380bp M B l B2 B3 B4 M l M2 M3 4 380bp Figure 11: Ream pi idea lion and Purification in Cloning Procedure. A. Reamplification of eluted DNA representing alterations B and M . B. Acrylamide purification of cut inserts prior to ligation in vector. B l - 4 represent four recurring alterations at 380 bp size; M l - 3 represent three recurring alterations at 380 bp size. 50 B l B2 M l M2 A . j a r mm ** • u l < U K M M I I U U M M I I * t i t t l *» Figure 12: Colony PCR and Colony Fingerprint in Cloning Procedure. A . Colony PCR: Transformed colonies were verified to contain specific size insert. B. Colony Fingerprint: Transformed colonies containing same size inserts were verified to contain the same sequence by G-track sequencing. B l - 2 and M l - 2 represent transformed colonies with D N A insert B and M , respectively. 51 CD CD P3 9 13 5 o 03 Q §2 So 52 CD »-b o 52 3.9. Localizing Sequenced Products to Chromosomal Locations There are many chromosomal regions throughout the human genome proposed to be involved in lung cancer development. Many of these regions were identified by gross mapping using FISH or radiation hybrids (Yasuda, 1996). A l l cloned sequences (Table 3) were matched against the high throughput genome sequences (htgs) available through NCBI. Figure 14A shows alteration B matching to RP11-637F16 whereas figure 14B shows alteration N matching to RP11-26106. Both alterations match their sequence with the bacterial artificial chromosomes (BACs) at high similarity. The same sequences were matched against the Human Genome Project Program at the University of Southern California (http ://genome.ucsc. edu). Alteration B / M matches 8q24.3. Alterations N , D, and A match 7q36.2, lq22.3, and 5q33.1, respectively. 3.10 Identification of LOH Markers Specific to Cloned Regions There are many microsatellite markers scattered throughout the human genome. However, many of these di- and tri-nucleotide repeats are not polymorphic therefore their use is not informative in a L O H study. Using the human genome project at the University of Southern California ( potential L O H markers were chosen specific to all four sequenced regions. Figure 15A shows the sequenced region of 7q36.2 matched against 53 Sequences producing s i g n i f i c a n t alignments: (bits ) Value giI 165Q6447I qh I AC02 1 63 6 • 6 I Homo sapiens chromosome 8 clone . . . 458 e-12 6 gll9256419|qta|AC02 2 41S.4IAC02241B Homo sapiens chromosome 5... 46 0.019 q i l 9 2 56328|qbl ACQ 12 63 1 • 4 I ACQ 12 63 1 Homo sapiens chromosome S... 46 0.019 q i | 6Q41753 IqrjI ACQ118Q5• 1 I ACQ11805 Homo sapiens chromosome 2... 44 0.074 B . Score E Sequences produ c i n g s i g n i f i c a n t alignments: ( b i t s ) Value qiI777Q4 4 5I qbI ACQ16222 .41 ACQ16222 Homo sa p i e n s c l o n e RP11-2. . . 2 IP 3e-52 aiI 15383794 IqbIACQ935Q5•11 Homo sapi e n s chromosome S cl o n e ... 42 O.IS qiI 152 OS569 IqbIAC093311•II Homo sapi e n s chromosome S c l o n e ... 42 0.15 q i l 151934Q1IqtaIACQ93 2 67•1 I Homo sa p i e n s chromosome S c l o n e ... 42 O.IS Figure 14. Localization of Sequenced Product to Chromosomes. Sequenced products were matched against the high throughput genome sequences database at NCBI. Matches were scored based on degree of similarity. A) Alteration B / M highly matched (redline) to a chromosome 8 clone RP11-637F16. B) Alteration N highly matched (red line) clone RP11-26106. Both clones are sequenced BACs that were mapped using the Fingerprinting Contigs (FPC) program at Washington University. 54 Sequences producing s i g n i f i c a n t alignments: Score E (bi t s ) Value g \ttl.6447lqblAC°21fi.^.S| Homo sapiens chromosome 8 clone 4 5 8 e 12,5 g i 925641 9| qb|AC0224ia.4|ACn :>74 18 Homo sapiens chromosome 5 ~ g i 9256328!^ I AC012631.4I , r n 1 7 ^ , Homo sapiens chromosome I g i | 6041753 I crb I ACOHRnc; 11 ACOllRn.s Homo sapiens chr 46 0.019 46 0.019 omosome 2... 4 4 0 . 0 7 4 Sequences p r o d u c i n g s i g n i r l e a n t alignments: T^bits) Value gi|777Q445Igb|AC016222.41ACQ16222 Homo sapiens c l o n e R P l i - 2 gij15383794,gb|ACQ93 5 0 5. 1| Homo sapi e n s chromosome 5 c l o n e ''' gi|152Q8569|qb|ACQ933n. 1| Homo sapi e n s chromosome S c l o n e gi|151934Ql|qb|AC093267.1| Homo sapi e n s chromosome S c l o n e " Figure 14. Localization of Sequenced Product to Chromosomes. Sequenced products were matched against the high throughput genome sequences database at NCBI. Matches were scored based on degree of similarity. A) Alteration B / M highly matched (redline) to a chromosome 8 clone RP11-637F16. B) Alteration N highly matched (red line) clone RP11-26106. Both clones are sequenced BACs that were mapped using the Fingerprinting Contigs (FPC) program at Washington University. 54 A . D8S1741-D8S452-D8S345-Base Position Chromosome Bands fiFMBeilXGl RFMB9B7ZD5 SHGC-85939 RH46926 RH48482 RH36664 RH25746 RH26828 SHGC-145999 D8S452 RH62296 SHGC-152729 SHGC-149396 - • D8S345 YourSeq 1445689961 1459999991 Chromosome Bands Localized by FISH Mapping Clones 8q24.3 STS Markers on Genetic (blue) and Radiation Hybrid (black) Maps I Your Sequence from BUTT Search I B . D7S798-D7S2462-Chromosome Bands flFM295Vf)3 SUSS1332 SHGC-68566 SMSS2859 D7S1951 SHGC-142785 SHGC-144192 SHGC-155777 SHGC-13291 WI-643 CHLC.GflTR2C88 SHGC-192439 SHGC-U9374 CHLCGCT3B94 SHGC-149543 D7S1495 SHGC-146281 CHLC.GfiTfll21Dll D7S1829 CHLC.GHTfl48Ee5 SHGC-83996 SHGC-142687 SHGC-78134 D7S2825 >• flFMft342YC9 YourSeq Chromosome Bands Localized by FISH Mapping Clones STS Markers on Genetic (blue) and Radiation Hybrid (black) Maps I I I Your Sequence from BLflT Search UCSC Genome Browser ( Figure 15. Identification of LOH markers at chromsomal regions 8q24.3 and 7q36.2. Polymorphic microsatellite markers were picked from the Human Genome Project at UCSC. that mapped very close to the cloned sequenced regions. A. Region 8q24.3 with three markers: D8S1741, D8S452 and D8S345. B. Region 7q36.2 with two markers: D7S798 and D7S2462. Vertical lines represent relative positions of sequence and markers. 55 D8S1741-D8S452-D8S345-Base Position Chromosome Bands, • RFMB811XG1 HFMB087ZD5 SHGC-85839 RH46926 RH48482 RH36664 RH25746 RH26828 SHGC-145988 D8S452 RH62296 SHGC-152728 SHGC-149396 D8S345 YourSeq STS Markers on; Genetic : 1 4 4 5 8 9 6 9 8 1 , , . Chromosome I Bands Localized by FISH Mapping Clones 8q24.3 ' •; Your Cblue) and Radiation Hybrid i Sequence from BLAT Search i I ! I i ill I f I I I I I I (black) 1.1 Maps B . D7S798-D7S2462-Chromosome Bands flFM285Vfi3 SNSS1332 SHGC-68566 SMSS2659 D7S1951 SHGC-142785 SHGC-144182 SHGC-15S777 SHGC-13291 NI-643 CHLCGftTfl2C88 SHGC-182489 SHGC-118374 CHLC.GCT3B84 SHGC-149543 D7S1495 SHGC-146281 CHLC.GHTni21Dll D7S1829 CHLC,GflTn48E85 SHGC-83996 SHGC-142687 SHGC-78134 D7S2825 • flFMH342YC9 YourSeq I STS Chromosoine Bands Localized by FISH Mapping Clones M M — a — (black) Maps Markers oh Genetic I I I M i l I ! (blue) ! ! I Your j I and Radiation Hybrid Sequence from BLflT Search Hi i i i i I I I i I I I I !' UCSC Genome Browser ( Figure 15. Identification of LOH markers at chromsomal regions 8q24.3 and 7q36.2. Polymorphic microsatellite markers were picked from the Human Genome Project at UCSC. that mapped very close to the cloned sequenced regions. A . Region 8q24.3 with three markers: D8S1741, D8S452 and D8S345. B. Region 7q36.2 with two markers: D7S798 and D7S2462. Vertical lines represent relative positions of sequence and markers. 55 the UCSC data with the sequence and surrounding L O H markers. Figure 15B shows the sequenced region matching with 8q24.3 surrounded by L O H markers. Minimum product size and heterozygosity information for each marker was obtained at ( Table 4 lists all microsatellite markers ordered from Alpha D N A Inc. and temperature tested. The table categorizes the markers according to chromosome regions targeted, size of product, optimal temperature, and level of heterozygosity within the human population. Table 4: Microsatellite Markers Chosen for 8q24.3, 7q36.2, 5q33.1 and lq23.3 Primer Chromosome Product Annealing Heterozygosity Size(bp) Temp. (°C) D1S2768F 1q23.3 198 Only 50 61% D1S2768R 1q23.3 198 Only 50 61% D1S2844F 1q23.3 177 60>55>50 82% D1S2844R 1q23.3 177 60>55>50 82% D9S1864F 1q23.3 270 Only 50 69% D9S1864R 1q23.3 270 Only 50 69% D5S673F 5q33.1 260-288 60>55>50 82% D5S673R 5q33.1 260-288 60>55>50 82% D7S798F 7q36.2 200-218 60>55>50 79% D7S798R 7q36.2 200-218 60>55>50 79% D7S2462F 7q36.2 127 60>55>50 85% D7S2462R 7q36.2 127 60>55>50 85% D7S483F 7q36.2 166-188 60>55>50 83% D7S483R 7q36.2 166-188 60>55>50 83% D8S1727F 8q24.3 257 50»55 76% D8S1727R 8q24.3 257 50»55 76% D8S1741F 8q24.3 298 60>55>50 60% D8S1741R 8q24.3 298 60>55>50 60% D8S1743F 8q24.3 171 Failed 82% D8S1743R 8q24.3 171 Failed 82% D8S345F 8q24.3 291-296 50»55 50% D8S345R 8q24.3 291-296 50»55 50% D8S452F 8q24.3 218 60>55>50 50% 56 D8S452R D8S508F D8S508R D8S534F D8S534R 8q24.3 8q24.3 8q24.3 8q24.3 8q24.3 218 60>55>50 230 60>55>50 230 60>55>50 176-210 55>50>60 176-210 55>50>60 50% 55% 55% 80% 80% L O H marker positions were verified by matching the sequence using NCBI and mapping the sequences using the Fingerprinting Contigs (FPC) Program at Washington University. Figure 16 shows two contigs of overlapping BACs with the sequence at 7q36.2 and 8q24.3 identified along with verified markers. Confirmation of precise marker positions was necessary using B A C contigs because marker positions using UCSC data are based on linkage studies. 3.11. Loss ofHeterozygosity Analysis Allelic imbalance is a characteristic method in assessing chromosomal instability on a locus-specific basis. Many studies have identified regions of chromosomal instability in lung cancer using microsatellite markers distributed throughout the human genome. Due to D N A quantity constraints of microdissected lung samples, only two chromosomal regions, 8q24.3 and 7q36.2 were subjected to L O H analysis. Six patients were screened for allelic imbalance at 8q24.3 using D8S452, a microsatellite marker. Allelic imbalance is assessed as a difference in the PCR signal ratio of the two paternal and maternal alleles within the normal compared to the tumour DNA. Two patients showed allelic imbalance (Al), three patients were retention (R) and one patient was not informative (NI) (Figure 17A). 57 A . D8S1741 D8S345' N02B4H1B. D8S452 (RAPD Fragment) D7S798 RAPD Fragment Figure 16. Contigs of Overlapping BACs. Microsatellite markers were mapped to overlapping BACs for both A) 8q24.3 and B) 7q36.2 using the Fingerprinting Contigs (FPC) Program at Washington University. Cloned region (RAPD fragment) was also localized to a B A C . 58 A. D8S1741 -—N05Z6P07— ' W570U4 - ' - -N0311H16+ D8S345' B . D7S798 N0364L19 N11B1N22W D8S452 (RAPD Fragment) D7S2462-"TO55i?coi \ " N0175L13 ' '\jQlB2NI17 R$5£4I15 N071SL0B^ F054ZG22 N0395OO« I746E12W FO5S4P0s\" N137SC19W -• NOQ1SA194-M23DOD \ N 0 3 4 9 C W ) -'- \ M214QP1B N0464 C23 N| ' NO: N05: •N0748O03-N1433012W N037OI22-~N1196J02w~ "7uw52G20t-N0561E01 N0103E18 aEOflw N0M0P04t N05S1D01 TT0535L02-"7387BK02b -F0497K1S N026SB08-" 0^026106+" N0*21 EOS N077«B15 '— RAPD Fragment Figure 16. Contigs of Overlapping BACs. Microsatellite markers were mapped to overlapping BACs for both A) 8q24.3 and B) 7q36.2 using the Fingerprinting Contigs (FPC) Program at Washington University. Cloned region (RAPD fragment) was also localized to a B A C . 58 A . C T C T C T C T C T C T 27 28 30 29 48 47 B 19 B 133 B 100 IIIII IIIIII NI Al Al R R R B . C T C T C T C T C T C T C T C T C T 30 29 61 62 B 133 B 100 151 153 161 87 45 44 121 120 81 77 | l | I l l i l H NI Al R NI R Al R NI Al C T C T C T C T C T C T C T C T C T C T C T 113 2 64 63 54 55 93 59 68 69 85 82 92 19 43 41 94 22 TT5 53 109TFl . I U U U I - ' U I I I I U M I " R R Al R Al R NI NI R R NI Figure 17: Loss of Heterozygosity (LOH) Analysis. A . Six patients were screened for allelic imbalance at 8q24.3 using D8S452. Two patients showed A l , three patients exhibited R and one patient was NI. B. Twenty N S C L C tumor/normal pairs were screened for allelic imbalance at 7q36.2 using D7S798. Five patients showed allelic imbalance (Al), nine patients exhibited retention(R) and six patients were non-informative(NI). • Allelic Imbalance • Retention 59 A . C T C T C T C T C T C T 27 28 30 29 48 47 B 19 B 133 B 100 B . C T C T C T C T C T C T C T C T C T 30 29 61 62 B 133 B 100 151 153 161 87 45 44 121 120 81 77 <^ JD ff, ^ W iff ,!'( W NI Al R NI R Al R NI Al C T C T C T C T C T C T C T C T C T C T C T Figure 17: Loss of Heterozygosity (LOH) Analysis. A . Six patients were screened for allelic imbalance at 8q24.3 using D8S452. Two patients showed A l , three patients exhibited R and one patient was NI. B . Twenty NSCLC tumor/normal pairs were screened for allelic imbalance at 7q36.2 using D7S798. Five patients showed allelic imbalance (Al), nine patients exhibited retention(R) and six patients were non-informative(NI). • Allelic Imbalance • Retention Twenty patients were screened for allelic imbalance at 7q36.2 using D7S798. Five patients showed A l , nine patients showed R and six patients were NI at the locus (Figure 17B). To confirm the visual assessment of allelic imbalance, PCR signals were quantitated using a phosphorscreen and ImageQuant 5.0 software (Table 5). Quantitated signal was interpreted by calculating the ratio of lower allele signal to upper allele signal. The ratio was calculated and compared for both C and T. Ratio differences equal to or greater than 50% were considered as significant. 3.12 Previously Discovered Genes at 8q24.3 and 7q36.2 Many oncogenes such as ras and c-myc and tumour suppressor genes such as p53 and p i 6 have been implicated as potential markers of NSCLC; however, no one gene has been able to act as a great predictor of NSCLC. Novel genes need to be discovered as early indicators of N S C L C progression to assist in NSCLC disease management. Using the UCSC database, known genes were identified within 2 Mb of each cloned region. Figure 18A shows seven genes in the 8q24.3 region including LOC51059, K C N K 9 , MOST-1, MGC4737, CHRAC1, PTK2 (FAK) and FLJ12193. Figure 18B shows four genes in the 7q36.2 region including FLJ21634, M L L 3 , X R C C 2 and ARP3BETA. Unfortunately, most of these genes are not characterized. 60 Name Signal Intensity Ratio of Lower Allele/Upper Allele LOH Designation 61L 1141 1.32 61U 866 Allelic Imbalance 62L 1873 3.04 62U 616 BL 1487 1.40 BU 1064 Retention 133L 1483 1.38 133U 1073 151L 251974 1.99 151U 126660 Retention 153L 143799 .1.48 153U 97394 161L 274999 2.07 161U 132823 Allelic Imbalance 87L 135375 1.04 87U 130032 45L 258388 1.81 45U 143125 Retentbn 44L 225255 1.84 44U 122355 81L 154866 1.31 81U 117912 Allelic Imbalance 77L 246970 2.00 77U 123510 113L 271113 0.96 113U 283364 Retention 2L 199117 0.89 2U 222937 64L 91254 0.55 64U 166443 Retention 63L 106656 0.54 63U 196431 54L 88629 0.45 54U 195988 Allelic Imbalance 55L 171238 0.74 55U 229867 93L 201621 0.66 93U 304562 Retention 59L 255519 0.82 59U 313312 68L 42451 0.22 68U 194474 Allellic Imbalance 69L 97030 0.38 "69U 254075 85L 82725 0.45 85U 182550 Retention 82L 102110 0.53 82U 192157 94L 268687 0.84 94U 320623 Retention 22L 258564 0.87 22U 297515 115L 318397 0.87 115U 366043 Retention 53L 258142 0.88 53U 293722 Table 5: Analysis of 7q36.2 L O H Results Using ImageQuant 5.0 Quantitation Software. Band signal intensities were determined and ratio of lower allele to upper allele were calculated. Ratios between tumor/normal within each patient were compared to designate either allelic imbalance or retention. A minimum of 50% change was required to designate allelic imbalance. 61 A . i43eeeeee| 1449888891 1458888881 1468899991 Chromosome B a n d s L o c a l i z e d toy F I S H M a p p i n g C l o n e s 8024.3 Y o u r S e q u e n c e f r o m BLf lT s e a r c h Known G e n e s ( f r o m R e f S e q ) H B a s e P o s i t i o n Chromosome B a n d s v o u r s e a L Q C 5 1 8 5 9 KCNK9 MOST-1 MGC4737 CHRf lCl P T K 2 F L J 1 2 1 9 3 LOC51059=hypothetical protein K C N K 9 = potassium channel MOST-l= unknown protein MGC4737 = hypothetical protein CHRAC1= chromatin accessibility gene PTK2 = protein tyrosine kinase 2 (FAK) gene FLJ12193 = hypothetical protein B . B a s e P o s i t i o n Chrom osome B a n d s v o u r s e a F L J S 1 6 3 4 MLL3 XRCC2 flRP3BETfl i5789eeeel i5759eeee| 15889eeee| issseeees Chromosome B a n d s L o c a l i z e d toy F I S H M a p p i n g C l o n e s V o u r S e q u e n c e f r o m BLf lT S e a r c h Known C e n e s ( f r o m R e f S e c O FLJ21634= Unknown MLL3=Myeloid/Lymphoid Leukemia XRCC2 = X-Ray Damage Repair Gene ARP3BETA=Actin Related Gene Figure 18: Known Genes at 8q24.3 and 7q36.2. A) Seven genes are known within 2Mb radius of sequenced product. Some potential oncogenes/TSGs include CHRAC1 and PTK2 (FAK). B) Four known genes localize to within 2Mb radius of sequenced product. Potential oncogenes/TSGs are M L L 3 , XRCC2 and ARP3BETA. Many of these genes are not characterized and expression data is not available. 62 Chapter 4: Discussion 4.1 Summary of Results The focus of this thesis was the identification of novel genetic alterations in microdissected N S C L C tumours and their precursors on a genome-wide scale. To discover grade specific alterations, it was necessary to isolate specific pure cell populations from surrounding cells using manual microdissection. Due to the small nature of lung biopsies, the quantity of D N A extracted was minimal as the biopsies contained few numbers of cells. As most other genome-wide analyses such as L O H require larger amounts of DNA, RAPD-PCR was used to perform high resolution scanning of the N S C L C sample set. Using one specific primer pair, fifteen different recurring alterations were identified among 41 patients. Four recurring alterations mapped to distinct chromosomal regions at 8q24.3, 7q36.2, 5q33.1 and lq23.3. Due to the limited amount of microdissected patient DNA, verification of chromosomal instability was performed only at 8q24.3 and 7q36.2 using L O H analysis. A detailed physical map and a tiling set of human B A C clones were deduced for each region to facilitate future definition of minimal regions of alteration and gene identification. 4.2 Harvesting DNA from Multiple Grades ofTumourisenesis in Multiple Patients Lung biopsies are composed of many different cell populations. To overcome the problem with tissue heterogeneity, 172 NSCLC biopsies composed of various stages including normal, hyperplasia, different degrees of dysplasia, CIS and invasive tumour were manually microdissected. Prior to the microdissection, Dr. Jean LeRiche classified sections according to progressive stages and identified cell populations representing 63 various grades to guarantee that the correct cells were extracted. Contamination by unwanted cells was limited to approximately 10% of the total pool of cells. This level of contamination was tolerable as only changes that were recurring were targeted for subsequent cloning, sequencing and chromosomal localization. As very few cells were extracted, our concern was the D N A quantity. Total D N A yield from most grades was approximately 120 ng. As each RAPD reaction in this study only used 2ng of quantified DNA, most microdissected samples would support 60 separate analyses. Using RAPD-PCR, primer pair 31/32 generated about 50 fragments, targeting 50 separate genetic loci. Assuming that other appropriate primer pairs would target as many loci, the D N A material would support screening of approximately 3000 separate loci. Assuming that these loci are equally distributed throughout the human genome, the genome-wide analysis is sampling loci at less than 1 Mb intervals. Compared to LOH, that requires a minimum of 5ng per reaction to screen one locus, RAPD-PCR was an efficient method to use on this non-renewable resource. Pathologists use various histological stains to discern morphologically distinct cells. However, reports suggest that certain stains are detrimental to D N A quality. Unfortunately, the cause of the degradation is not known. We have shown that increasing exposure of cultured epithelial cells to hematoxylin, a component of H & E , causes the degradation of DNA. It was also shown that methyl green, at its natural pH, degrades DNA. We believe that the low pH is the main cause of the increased degradation. We believe that an acidic environment would cause degradation by the hydrolysis of purine bases in the D N A structure. As an alternative, it was suggested that we localized specific cells on a stained section and microdissect from unstained serial sections. This 64 alternative would be possible when concentrating on large areas of tumour but in lung biopsies, cells of interest were usually present in a focal manner and there was a high level of cell diversity between serial sections. To overcome this problem, we decided to buffer the various histological stains to a neutral pH and to test their ability to distinguish cellular features. It was clear that the only functioning stain after buffering was methyl green. 4.3 RAPD-PCR, A Method of Genome-Wide Scanning RAPD-PCR is a D N A fingerprinting technique that involves the PCR amplification of random fragments of genomic D N A with short oligonucleotides of arbitrary sequence. The main advantage of this high throughput technique is the ability to screen multiple loci simultaneously using minute quantities of D N A from archival specimens to detect gene copy number changes without the knowledge of the D N A sequence target. Apart from the efficiency benefits due to the high-resolution capacity of RAPD-PCR, the discovery of novel genes is facilitated as RAPD-PCR promotes the quick identification of alterations through cloning. At best, other mapping techniques such as C G H resolve increased or decreased copy number changes to chromosomal bands whereas RAPD-PCR alterations are generally in the 100 to 800 bp size range. With the increased completion of the human genome sequence, simple database searches are now replacing experimental chromosomal localization through techniques such as fluorescence in situ hybridization (FISH) or humamrodent monochromosome cell hybrid hybridization (Yasuda, 1996). 65 A major concern of discovering alterations with RAPD-PCR is the inability of the technique to distinguish between chromosomal gains and losses. In the analysis of the RAPD-PCR fingerprinting gels, alterations are classified as either gains or losses of PCR fragments in the progressive disease stages compared to the normal stages. The technique does not predict the underlying mechanism of the alteration. A gain is characterized as a chromosomal amplification where a specific loci experiences an increased copy number in that region. However, a gain can also be characterized as a chromosomal deletion where new primer binding sites have been created via a deletion or a deletion has brought two primers binding sites close enough together to form a product. A RAPD-PCR loss is characterized as a deletion of a chromosomal region eliminating primer-binding sites. At the same time, a loss can be characterized as insertion of chromosomal material thereby separating two primer-binding sites far enough to prevent PCR amplification. Other infrequent chromosomal aberrations such as translocations or inversion are difficult to distinguish as these changes would appear as either gains or losses depending on the break point. Another major concern of this genome-wide scanning technique is the potential missing of key genes as the technique is unable to detect single base mutations (i.e. point mutations). Point mutations in certain genes such as K-ras and pl6 have been the cause of phenotypic changes. However, it is understood that gene disruption occurs by many different mechanisms in many different patients therefore the same mutational event is unlikely to occur in every patient. Although a gene may be disrupted in one patient by a point mutation, another patient's gene inactivation may be caused by a different mechanisms such as a large deletion. The hope is that the genes involved in signaling 66 tumour progression would be detected in a large cohort as clustering of locus-specific instability by various mechanisms including larger chromosomal amplifications or deletions. 4.4 Importance of Recurrences Every individual is composed of a human D N A sequence that is based on the maternal and paternal chromosomes. Although the majority of the D N A sequences in a population is similar, there are regions where D N A sequence is varied. RAPD-PCR targets many different loci with each primer pair but the loci targeted are not always represented in every patient. A l l four cloned regions stemmed from recurring alterations. Region 8q24.3 was cloned from an alteration that was denoted as a gain of a PCR fragment in the latter stages of progression, specifically CIS. The gain was evident in 6 patients out of 17 patients that exhibited the polymorphism. Region 7q36.2 was cloned from an alteration that was a gain in CIS present in 2 of 15 patients that carried the polymorphism. Region lq23.3 was cloned from an alteration that was a gain in CIS in 2 of 18 patients whereas 5q33.1 was derived from a gain seen in 2 of 16 patients. Many mechanisms exist that cause the disruption of a gene's native function. These events include single base mutations, deletions, amplifications, translocations and trans versions. Recurrences that occur at the frequencies seen in this thesis, 6/17, 2/15, 2/18, 2/16, are considered significant as the RAPD-PCR technique can only detect a fraction of the gene altering mechanisms. It has been suggested to use more primer pairs to screen for different loci. Sooner or later, by screening more primers, the regions of chromosomal instability will begin to cluster at specific regions. False positive results 67 are tolerated as only recurring alterations are further analyzed and false negative results are negligible as other primer pairs, in a comprehensive study, would detect the missed loci. 4.5 Two Candidate Regions at 8q24.3 and 7q36.2 After extensive L O H analysis, two novel candidate regions at 8q24.3 and 7q36.2 were identified that exhibited allelic imbalance. Previous reports have suggested these regions as areas of chromosomal instability. Michelland et al.(1999) report consistent patterns of chromosomal alterations in high-grade neuroendocrine (NE) carcinomas and N S C L C using C G H analysis. They observed amplifications in 6 of 11 N E carcinomas involving a minimal region of 8q23-telomere. The same group also reported amplifications in 2 of 11 N E carcinomas involving 7q31 - telomere and deletions in 2 of 11 N E carcinomas involving 7q21-telomere. Virmani et al.(1998) report allelic loss at 7q31.1-q31.2 in 2 of 19 lung cell lines, 1 of 6 SCLCs and 1 of 13 NSCLCs using L O H . Yamada et al.(2000) reported moderate gains of chromosomes 8q and 7q in 8 of 13 and 5 of 13 SCLCs, respectively. At 8q24.3, high frequency of allelic imbalance was seen in 40% (2/5) of CIS. Similarly, allelic imbalance was seen in 36% (5/14) of CIS at 7q36.2. One can suggest that both these regions experience an early change in the development of N S C L C . To verify these regions as risk predictors, one would need to analyze dysplastic lesions for chromosomal instability at these regions. In comparison, it has been shown that certain regions of allelic imbalance have been effective as risk predictors of N S C L C progression. For example, allelic losses at lp32-ter have been correlated with late stage N S C L C and 68 metastasis risk (Chizhikov, 2001). Zhou et al.(2000) report that microsatellite instability at either 3pl4 or 10q24 was associated with shortened disease-specific survival. With verification, both 8q24.3 and 7q36.2 may act as a new region of predictive risk. In close proximity to the cloned region at 8q24.3 exists a known oncogene, c-myc, located at 8q24.12, 12 Mb centromeric to the cloned region. C-myc has been associated with many different cancers including breast, colon and lung. Yamada et al. (2000) tested for c-myc amplification in the 8 SCLC tumours expressing gains in the 8q region and found that five tumours had associated c-myc amplification. These results are consistent with those shown by Little et al. (1983) and Okazaki et al.(1996) demonstrating a several fold increase in copy number of the c-myc gene in primary SCLC. However, three tumours did not show c-myc amplification suggesting that there were other genes in the regions over-represented in SCLC. 4.5.1 Genes Implicated at 8q24.3 Using data from the Human Genome Project at UCSC, we determined that there were seven known genes within a 2 Mb radius of the cloned sequence. These genes included LOC51059, K C N K 9 (Kim, 2000), MOST-1, MGC4737, C H R A C 1 , PTK2 (Fiedorek, 1995) and FLJ12193. Some information is available on these genes but many have undetermined functions and tissue specificity. The K C N K 9 gene encodes a protein that acts as a tandem pore domain acid-sensitive potassium channel while the CHRAC1 gene encodes for a chromatin accessibility complex. An interesting observation is that one of the genes, the PTK2 (FAK) gene encodes for a protein tyrosine kinase 2 believed to be an early step in intracellular signaling transduction pathways. The phosphorylation 69 of PTK2 is reported to be triggered by the interactions of integrins with various extracellular matrix adhesion molecules and by neuropeptide growth factors. PTK2 is expressed in most tissues types including T and B lymphocytes but has been shown to be most abundant in brain tissue. The potential role of PTK2 is believed to be oncogenic transformations resulting in increased kinase activity. Unfortunately, expression data is not available for these genes but possible experiments are described in the future directions 4.5.2 Genes Implicated at 7q36.2 Using data from the Human Genome Project at UCSC, we determined that there are four previously discovered genes within a 2 Mb radius of the cloned sequence. These genes include M L L 3 (Tan, 2001), XRCC2 (Thacker, 1995), FLJ21634 and ARP3BETA (Jay, 2000). The M L L 3 (myeloid lymphoid leukemia 3) gene encodes a protein associated with leukemia and developmental defects. It is suggested that the protein is involved in transcriptional regulation and studies have shown that M L L 3 deletions are associated with hematological neoplasias. The XRCC2 (X-ray repair cross complementing protein 2) gene is believed to be involved in the control of the rearrangement process leading to repair of D N A double strand breaks. Both FLJ21634 and ARP3BETA have functions not remotely relevant to tumourigenesis. Unfortunately, expression data is not available for these genes but possible experiments are described in the future directions. 70 4.6 Implications of 8q24.3 and 7q36.2 in Epithelial Carcinomas As epithelial cancers are categorized based on similar progressive characteristics, it is of great interest to compare our candidate region findings with observations seen in other epithelial cancers. Weist et al. (1997) reported that allelic loss was seen at 3p and 9p in lung, head and neck and oral cavity carcinomas. Saha et al. (2001) report identification of a phosphatase gene, PRL-3, at 8q24.3 that is associated with colorectal cancer metastasis. Kurdistani et al. (1998) report the downregulation of the RTP/rit42 gene may be involved in the development of a malignant tumour phenotype. In a separate study in our laboratory, RAPD-PCR was used to identify a region at 7q22 to be linked to oral epithelial carcinogenesis (personal communication). Multiple loci at 7q have been implicated in many cancers such as oral cancer (Wang, 1998), breast cancer (Zeng, 1999) and myeloid neoplasms (Tosi, 1999). Recently, Zenklusen et al (2001) identified ST7 as a highly conserved tumour-suppressor gene on chromosome 7q31. It is clear that both these regions are potential hot-spots of chromosomal instability that need to be further investigated. 4.7 Future Directions Four distinct chromosomal regions were identified by the analysis of RAPD-PCR fingerprints of progressive stages of NSCLC. These regions include 8q24.3, 7q36.2, 5q33.1, and lq23.3. Verification was only performed on 8q24.3 and 7q36.2 due to D N A quantity constraints. However, with increased microdissected D N A sources, verification of the other two regions will be performed. Increasing D N A quantity will also allow 71 fine-mapping of all four regions by defining boundaries for chromosomal instability and the discovery of the oncogenes or tumour suppressor genes in these regions. As potential oncogenes or tumour suppressor genes are discovered, there are various ways to perform expression studies on these novel genes in NSCLC. In order to analyze expression, one needs to study R N A levels. Northern blot hybridization is an obvious choice for RNA analysis but is only appropriate when high R N A concentrations are extracted from frozen biopsies. A n alternative is Real-Time PCR, a highly sensitive quantitative assay designed for analyzing gene expression from minute amounts of RNA. Also, RT-PCR has capabilities to screen a large number of genes and to process a large number of samples. Its sensitive and high-throughput approach makes it optimal for the rapid expression analysis of genes discovered by RAPD-PCR. Expression analysis using RT-PCR is optimal but it requires the availability of fresh frozen material. In many cases, tissue material is only available as formalin-fixed archival biopsies, which does not contain intact R N A due to fixation and lengthy storage. A n alternative, in cases where R N A is not available, immunohistochemistry allows the analysis of gene expression using antibodies. The only drawback is that antibodies are not produced for every discovered gene, especially novel genes and all antibodies are not able to hybridize to archival biopsy sections. With 8q24.3 and 7q36.2 regions verified, the next goal would involve the definition of boundaries of the two alterations. It is understood that the identified region could be altered by many different mutational mechanisms therefore it would be necessary to identify the region directly involved in these events. The easiest way to set boundaries would involve the picking of microsatellite markers on either side of the 72 cloned region at increasing distances. Boundaries would be set when experimental L O H analysis would show retention in a certain number of patients on both sides of the cloned region using distal markers. However, some disadvantages of this fine-mapping technique exist. Firstly, since the microdissected material is non-renewable, the sample material would not be able to support the hundreds of analyses required to define the boundaries. Secondly, polymorphic markers do not exist at regular intervals throughout the human genome preventing the ability to move at small intervals away from the region of interest. Finally, the definition would be very time-consuming and labor intensive. A n alternative to fine-mapping using LOH analysis is the use of a new technology consisting of an array of overlapping bacterial artificial chromosomes (BACs) (Snijders, 2001). The B A C array is composed of sequence verified BACs containing approximately 100 kilobases each, organized in an overlapping fashion from one chromosome end to the other, fixed on a glass support. The technology involves the comparison of fluorescent signal representing hybridization of normal D N A to the signal representing hybridization of tumour D N A to the B A C array. Differences in signal intensities represent gene copy number differences. The rationale is that by using a tiled B A C contig for a specific region, one can identify a small region of a chromosome that has a gene copy number change represented by an altered signal on a B A C . More importantly, the identification of the region would only require small quantities of DNA, as little as lOng and represent information equivalent to hundreds of L O H analyses. Fine-mapping of the four regions would be more efficient using the B A C array. 73 Conclusion In conducting a genome-wide search of novel genetic alterations in archival grade progressive N S C L C biopsies using RAPD-PCR, many experimental goals were identified and achieved. As lung biopsies are composed of many varied cell populations, a method for separating specific cells representing different histological grades had to be developed. Manual microdissection was optimized for the extraction of cells from surrounding cell populations. Although archival tissues present a vast resource of material, the D N A quality is varied between biopsies and is, in some cases, degraded. Suitable staining conditions had to be developed to allow for the visualization of morphologically distinct cells while maintaining D N A integrity. After experimenting with many histological stains including H&E, a stain known as Methyl Green, at a buffered pH 6.0, was chosen as an optimal stain. To prevent the microdissection of archival tissues with initially degraded DNA, unstained biopsies were screened for good D N A integrity using RAPD-PCR. Good quality D N A material extracted from these small N S C L C biopsies was very limiting. A method had to be developed to accurately measure D N A concentrations of extracted material while minimizing loss of D N A used to quantitate. Southern blot hybridization was chosen as an efficient method for assessing D N A quantities. After the microdissection of 180 biopsies representing 41 different patients, RAPD-PCR was conducted to identify novel alterations recurring in multiple patients. RAPD-PCR was able to screen multiple loci from minute archival specimens in an unbiased fashion. Fulfilling the first hypothesis, RAPD-PCR was able to identify fifteen recurring alterations and four alterations, specific to CIS, were mapped to distinct chromsomal regions: 8q24.3, 7q36.2, 5q33.1 and lq23.3. Due to patient D N A 74 quantity constraints, two regions 8q24.3 and 7q36.2 were chosen for further investigations. As polymorphic markers are not available in many regions, only a few markers were temperature-tested for verification of chromsomal instability in these regions. Fulfilling the second hypothesis, two of five patients showed allelic imbalance at 8q24.3 and five of twelve showed allelic imbalance at 7q36.2. Many possible genes exist in these two regions that could act as potential oncogenes or tumour suppressor genes however, further studies are necessary to define the expression patterns of these genes. 75 References Ahrendt SA, Chow JT, X u L H , Yang SC, Eisenberger CF, Esteller M , Herman, JG, Wu L, Decker PA, Jen J, Sidransky D. (1999). Molecular detection of tumor cells in bronchoalveolar lavage fluid from patients with early stage lung cancer. Journal of the National Cancer Institute 91(4): 332-339. Ahrendt SA, Decker PA, Doffek K, Wang B, X u L, Demeure MJ , Jen J, Sidransky D. (2000). Microsatellite instability at selected tetranucleotide repeats is associated with p53 mutations in non-small cell lung cancer. Cancer Research 60: 2488-2491. Ahrendt SA, Chow JT, Yang SC, Wu L, Zhang MJ , Jen J, Sidransky D. (2000). Alcohol consumption and cigarette smoking increase the frequency of p53 mutation in non-small cell lung cancer. Cancer Research 60: 3155-3159. Anami Y , Takeuchi T, Mase K, Yasuda J, Hiroshashi S, Perucho M , Noguchi M . (2000). Amplotyping of microdissected, methanol-fixed lung carcinoma by arbitrarily primed polymerase chain reaction. International Journal of Cancer 89: 19-25. Bergmans HE, van Die IM, Hoekstra WP. (1981). Transformation in Escherichia coli: stages in the process. Journal of Bacteriology 146(2):564-70. Burkitt, H.G., B. Young and J.W. Heath. (1993). Wheater's Functional Histology. Churchill Livingstone, New York. Burton MP, Schneider BG, Brown R, Escamilla-Ponce N , Gulley M L . (1998). Comparison of histologic stains for use in PCR analysis of microdissected, paraffin-embedded tissues. Biotechniques 24:86-92. Chizhikov V , Zborovskaya I, Laktionov K, Delektorskaya V , Polotskii B, Tatosyan A , Gasparian A . (2001). Two consistently deleted regions within chromosome lp32-pter in human non-small cell lung cancer. Molecular Carcinogenesis 30: 151-158. De Juan C, Iniesta P, Cruces J, Sanchez A , Massa MJ , Gonzalez-Quevedo R, Torres A J , Balibrea JL, Benito M . (1999). D N A amplification on chromosome 6pl2 in non small cell lung cancer detected by arbitrarily primed polymerase chain reaction. International Journal of Cancer 84: 334-349. De Juan C, Iniesta P, Vega FJ, Peinado M A , Fernandez C, Caldes T, Massa M J , Lopez JA, Sanchez A , Torres AJ , Balibrea JL, Benito M . (1998). Prognostic value of genomic damage in non-small-cell lung cancer. British Journal of Cancer 77 (11): 1971-1977. Dosaka-Akita H, Hu S, Fujino M , Harada M , Kinoshita I, X u H, Kuzumaki N , Kawakami Y , Benedict W. (1997). Altered retinoblastoma protein expression in nonsmall cell lung cancer: its synergistic effects with altered ras and p53 protein status on prognosis. Cancer 79:1329-1337. 76 Esteller M , Sanchez-Cespedes M , Rosell R, Sidransky D, Baylin SB, Herman JG. (1999). Detection of aberrant promoter hypermethylation of tumor suppressor genes in serum D N A from non-small cell lung cancer patients. Cancer Research 59: 67-70. Fearon ER, Vogelstein B. (1990). A genetic model for colorectal tumorigenesis. Cell 61: 759-767. . Fiedorek FT Jr, Kay ES. (1995). Mapping of the focal adhesion kinase (Fadk) gene to mouse chromosome 15 and human chromosome 8. Mammalian Genome 6:123-6. Fritsch P, Riesenberg L H . (1992). High outcrossing rates maintain male and hermaphrodite individuals in populations of the flowering plant Dastica glomerata. Nature 359:633-636. Gazdar A F , Bader S, Hung J, Kishimoto Y , Sekido Y , Sugio K, Virmani A , Fleming J, Carbone DP, Minna JD. (1994). Molecular genetic changes found in human lung cancer and its precursor lesions. Cold Spring Harbor Symposia on Quantitative Biology 59: 565-571. Girard L, Zochbauer-Muller S, Virmani A K , Gazdar A F , Minna JD. (2000). Genome-wide allelotyping of lung cancer identifies new regions of allelic loss, differences between small cell lung cancer and non-small cell lung cancer, and loci clustering. Cancer Research 60: 4894-4906. Greer C, Wheeler C, Michele Manos M . (1994). Sample Preparation and PCR Amplification from Paraffin-Embedded Tissues. PCR Methods and Applications: SI 13-122. Hamada K , Kohno T, Takahashi M , Yamazaki M , Tashiro H , Sugawara C, Ohwada S, Sekido Y , Minna JD, Yokota J. (2000). Two regions of homozygous deletion clusters at chromosome band 9p21 in human lung cancer. Genes Chromosomes Cancer 27: 308-318. Jay P, Berge-Lefranc JL, Massacrier A , Roessler E, Wallis D, Muenke M , Gastaldi M , Taviaux S, Cau P, Berta P. (2000). ARP3beta, the gene encoding a new human actin-related protein, is alternatively spliced and predominantly expressed in brain neuronal cells. European Journal of Biochemistry 267:2921-8. Jiang S, Sato Y . , Kuwao S, Kameya T. (1995). Expression of bcl-2 oncogene protein is prevalent in small cell lung carcinoma. Journal of Pathology 177:135-138. Kawakami K, Yasuda J, Shiraishi M , Kayama T, Doi K , Perucho M , Sekiya T. (1998). Detection of D N A abnormalities by arbitrarily primed PCR fingerprinting: allelic losses in chromosome lOq in lung cancers. Biochemical and Biophysical Research Communications 251: 153-157. 77 Kim Y , Bang H , Kim D. (2000). TASK-3, a new member of the tandem pore K(+) channel family. Journal of Biological Chemistry 275:9340-7. Kohno T, Kawanishi M , Matsuda S, Ichikawa H, Takada M , Ohki M , Yamamoto T, Yokota J. (1998). Homozygous deletion and frequent allelic loss of the 21ql 1.1 -q21.2 region including the A N A gene in human lung carcinoma. Genes Chromosomes Cancer 21:236-243. Krishnan BR, Blakesley RW, Berg DE (1991). Linear amplification D N A sequencing directly from single phage plaques and bacterial colonies. Nucleic Acids Research 19(5): 1153. Kurdistani SK, Arizti P, Reimer CL, Sugrue M M , Aaronson SA, Lee SW. (1998). Inhibition of tumor cell growth by RTP/rit42 and its responsiveness to p53 and D N A damage. Cancer Research 58(19):4439-44. Little CD, Nau M M , Carney DN, Gazdar AF, Minna JD. (1983). Amplification and expression of the c-myc oncogene in human lung cancer cell lines. Nature 306 (5939): 194-6. Martin GB, Williams JG, Tanksley SD. (1991). Rapid identification of markers linked to a Pseudomonas resistance gene in tomato by using random primers and near-isogenic lines. Proceedings of the National Academy of Sciences (U S A). 15;88:2336-40. Michelland S, Gazzeri S, Brambilla E, Robert-Nicoud M . (1999). Comparison of chromosomal imbalances in neuroendocrine and non-small-cell lung carcinomas. Cancer Genetic Cytogenetic 114: 22-30. Micheli MR, Bova R, Calissano P, D'Ambrosio E. (1993). Randomly amplified polymorphic D N A fingerprinting using combinations of oligonucleotide primers. BioTechniques 15(3):388-389. Momparler R, Bovenzi V . (2000). D N A methylation and cancer. Journal of Cellular Physiology 183: 145-154. Murase T, Inagaki H, Eimoto T. (2000). Influence of Histochemical and Immunohistochemical stains on polymerase chain reaction. Modern Pathology 13(2): 147-151. Nomoto S, Haruki N , Tatematsu Y , Konishi H, Mitsudomi T, Takahashi T, Takahashi T. (2000). Frequent allelic imbalance suggests involvement of a tumor suppressor gene at lp36 in the pathogenesis of human lung cancers. Genes Chromosomes Cancer 28: 342-346. 78 Okami K , Cairns P, Westra W H , Linn JF, Ahrendt SA, Wu L, Sidransky D, Jen J. (1997). Detailed deletion mapping at chromosome 9p21 in non-small cell lung cancer by microsatellite analysis and fluorescence in situ hybridization. International Journal of Cancer 74: 588-592. Okazaki T, Takita J, Kohno T, Handa H, Yokota J. (1996). Detection of amplified genomic sequences in human small-cell lung carcinoma cells by arbitrarily primed-PCR genomic fingerprinting. Human Genetics 98(3):253-8. 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(1): 233-235. Onuki N , Wistuba II, Travis WD, Virmani A K , Yashima K, Brambilla E, Hasleton P, Gazdar A F . (1999). Genetic changes in the spectrum of neuroendocrine lung tumors. Cancer 85(3): 600-607. Richardson G, Johnson B. (1993). The biology of lung cancer. Seminars in Oncology 20:105-127. Rosell R, L i S, Skacel Z, Mate JL, Maestre J, Canela M , Tolosa E, Armengol P, Barnadas A , Ariza A . (1993). Prognostic impact of mutated K-ras gene in surgically resected non-small cell lung cancer patients. Oncogene 8:2407-2412. Saha S, Bardelli A , Buckhaults P, Velculescu V E , Rago C, Croix BS, Romans K E , Choti M A , Lengauer C, Kinzler K W , Vogelstein B. (2001). A phosphatase associated with metastasis of colorectal cancer. Science 294(5545): 1343-6. Sambrook J, Fritsch EF, Maniatis T. (1989). Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press. Sanchez-Cespedes M , Ahrendt SA, Piantadosi S, Rosell R, Monzo M , Wu L, Westra W H , Yang SC, Jen J, Sidransky D. (2001). Chromosomal alterations in lung adenocarinoma from smokers and nonsmokers. Cancer Research 61: 1309-1313. Sanchez-Cespedes M , Decker PA, Doffek K M , Esteller M , Westra W H , Alawi W A , Herman JG, Demeure MJ , Sidransky D, Ahrendt SA. (2001). Increased loss of chromosome 9p21 but not pi6 inactivation in primary non-small cell lung cancer from smokers. Cancer Research 61: 2092-2096. Sekido Y , Fong K M , Minna JD. (1998). Progress in understanding the molecular pathogenesis of human lung cancer. Biochimica et Biophysica Acta 1378: F21-F59. 79 Serth J, Kuczyk M A , Paeslack U , Lichtinghagen R, Jonas U . (2000). Quantitation of D N A extracted after micropreparation of cells from frozen and formalin-fixed tissue sections. American Journal of Pathology 156:1189-96. Simone N L , Bonner RF, Gillespie JW, Emmert-Buck MR, Liotta L A . (1998). Laser-capture microdissection: opening the microscopic frontier to molecular analysis. Reviews 14(7): 272-276. Shivapurkar N , Virmani A K , Wistuba II, Milchgrub S, Mackay B, Minna JD, Gazdar A F . (1999). Deletions of chromosome 4 at multiple sites are frequent in malignant mesothelioma and small cell lung carcinoma. Clinical Cancer Research 5: 17-23. Snijders A M , Nowak N , Segraves R, Blackwood S, Brown N , Conroy J, Hamilton G, Hindle A K , Huey B, Kimura K, Law S, Myambo K, Palmer J, Ylstra B, Yue JP, Gray JW, Jain A N , Pinkel D, Albertson DG. (2001). Assembly of microarrays for genome-wide measurement of D N A copy number. Nature Genetics 29(3):263-4. Sugio K , Kishimoto Y , Virmani A K , Hung JY, Gazdar AF. (1994). K-ras mutations are a relatively late event in the pathogenesis of lung carcinomas. Cancer Research 54: 5811-5815. Taguchi T, Cheng GZ, Bell DW, Balsara B, Liu Z, Siegfried JM, Testa JR. (1997). Combined chromosome microdissection and comparative genomic hybridization detect multiple sites of amplification D N A in a human lung carcinoma cell line. Genes Chromosomes Cancer 20C2): 208-12. Tamura K , Zhang X , Murakami Y , Hirohashi S, X u HJ, Hu SX, Benedict WF, Sekiya T. (1997). Deletion of three distinct regions on chromosome 13q in human non-small cell lung cancer. International Journal of Cancer 74: 45-49. Tan Y C , Chow VT. (2001). Novel human H A L R (MLL3) gene encodes a protein homologous to A L R and to ALL-1 involved in leukemia, and maps to chromosome 7q36 associated with leukemia and developmental defects. Cancer Detection Prevention 25:454-69. Thacker J, Tambini CE, Simpson PJ, Tsui L C , Scherer SW. (1995). Localization to chromosome 7q36.1 of the human XRCC2 gene, determining sensitivity to D N A -damaging agents. Human Molecular Genetics 4:113-20. Tosi S, Scherer SW, Giudici G, Czepulkowski B, Biondi A , Kearney L. (1999). Delineation of multiple deleted regions in 7q in myeloid disorders Genes Chromosomes Cancer 25(4):384-92. Tseng JE, Kemp B L , Khuri FR, Kurie JM, Lee JS, Zhou X , Liu D, Hong W K , Mao L i . (1999). Loss of FHIT is frequent in Stage 1 non-small cell lung cancer and in the lungs of chronic smokers. Cancer Research 59: 4798-4803. 80 Tulsieram L K , Glaubitz JC, Kiss G, Carlson JE. (1992). Single tree genetic linkage mapping in conifers using haploid D N A from megagametophytes. Biotechnology 10:686-90. Virmani A K , Fong K M , Kodagoda D, Mclntire D, Hung J, Tonk V, Minna JD, Gazdar A F . (1998). Allelotyping demonstrates common and distinct patterns of chromosomal loss in human lung cancer types. Genes, Chromosomes & Cancer 21: 308-319. Wang X L , Uzawa K , Miyakawa A , Shiiba M , Watanabe T, Sato T, Miya T, Yokoe H , Tanzawa H. (1998). Localization of a tumour-suppressor gene associated with human oral cancer on 7q31.1. International Journal of Cancer 75(5):671-4. Weist JS, Franklin W A , Drabkin H, Gemmill R, Sidransky D, Anderson M W . (1997). Genetic markers for early detection of lung cancer and outcome measures for response to chemoprevention. Journal of Cellular Biochemistry Supplements 28/29: 64-73. Welsh J, Rampino N , McClelland M , Perucho M . (1995). Nucleic acid fingerprinting by PCR-based methods: applications to problems in aging and mutagenesis. Mutation Research 338: 215-229. Williams JGK, Kubelik AR, Livak KJ , Rafalski JA, Tingey SV. (1990). D N A polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucleic Acids Research 18(22): 6531-6535. Wistuba II, Behrens C, Virmani A K , Mele G, Milchgrub S, Girard L , Fondon III JW, Garner HR, McKay B, Latif F, Lerman MI, Lam S, Gazdar A F , Minna JD. (2000). High resolution chromosome 3p allelotyping of human lung cancer and preneoplastic/ preinvasive bronchial epithelium reveals multiple, discontinuous sites of 3p allele loss and three regions of frequent breakpoints. Cancer Research 60: 1949-1960. Wistuba II, Behrens C, Virmani A K , Milchgrub S, Syed S, Lam S, Mackay B, Minna JD, Gazdar A F . (1999). Allelic losses at chromosome 8p21-23 are early and frequent events in the pathogenesis of lung cancer, Cancer Research 59: 1973-1979. Wu X , Zhao Y , Kemp B L , Amos CI, Siciliano MJ, Spitz MR. (1998). Chromosome 5 aberrations and genetic predisposition to lung cancer.. International Journal of Cancer 79: 490-493. Yamada T, Kohno T, Nawaro JM, Ohwada S, Perucho M , Yokota J. (2000). Frequent chromosome 8q gains in human small cell lung carcinoma detected by arbitrarily primed-PCR genomic fingerprinting. Cancer Genetics Cytogenetic 120: 11-17. 81 Yasuda J, Navarro JM, Malkhosyan S, Velazquez A , Arribas R, Sekiya T, Perucho M . (1996). Chromosomal assigment of human D N A fingerprint sequences by simultaneous hybridixation to arbitrarily primed PCR products from human/rodent monochromosome cell hybrids. Genomics 34: 1-8. Zeng WR, Watson P, Lin J, Jothy S, Lidereau R, Park M , Nepveu A . (1999) 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 18(11):2015-21. Zenklusen JC, Conti CJ, Green ED. (2001). Mutational and functional analyses reveal that ST7 is a highly conserved tumor-suppressor gene on human chromosome 7q31. Nature Genetics 27(4):392-8. Zhou X , Kemp B L , Khuri FR, Liu D, Lee JJ, Wu W, Hong W K , Mao L. (2000). Prognostic implication of microsatellite alteration profiles in early-stage non-small cell lung cancer. Clinical Cancer Research 6:559-565. 82 APPENDIX 1. MICRODISSECTION PROTOCOL 1. Using outlined borders of lesion or area, all H&E slides are photographed and used as guides to corresponding areas on Methyl Green slides. 2. Methyl Green slides are mounted on a dissection microscope, the diagnostic areas identified, hand scraped using a 30G needle and collected into a buffer solution. 3. "Before" and "after" photos are taken on one section per slide as a record. STAINING PROTOCOL Hematoxylin and Eosin (H&E) 1. Xylene soak for 10 minutes to remove wax from sections 2. Hydrate to water through graded Ethanol 3. Rinse in water 4. Stain in Hematoxylin for 5 minutes 5. Rinse in water 6. Differentiate in 1% Acid Alcohol, 10 dips 7. Rinse in water 8. Blue in Lithium Carbonate (Saturated) 3 minutes 9. Rinse in water 10. Stain in Eosin 10 dips 11. Dehydrate through graded Ethanol 12. Clear in Xylene and coverslip Methyl Green Staining Reagent Preparation: -Methyl Green Stock Solution (2%) -2.0 grams of Methyl Green dissolved in 100 ml d H20, mixed with 100 ml Chloroform. Shake and discard Chloroform layer. Repeat until violet (Crystal Violet) contamination is gone. Adjust pH to 6.0 83 Staining Procedure: 1. Dewax slides in Xylene 2. Hydrate through graded Ethanols to water 3. Rinse in dH 20 4. 10 seconds in 0.2% Methyl Green, pH 6.0 5. Rinse in dH 20 6. Rinse in 2 changes of 100% Ethanol and airdry APPENDIX 2. PATIENT PROFILES Patient SEX BIRTHDATE SUB GROUP PACK YEARS 1 F 8/17/28 CURRENT-SMOKER 42 2 F 3/18/28 CURRENT-SMOKER 96 3 F 7/19/38 CURRENT-SMOKER 23 4 F 10/18/32 CURRENT-SMOKER 36 5 F 1/18/37 CURRENT-SMOKER 50 6 M 1/24/38 CURRENT-SMOKER 66 7 M 1/14/23 CURRENT-SMOKER 67.5 8 M 7/10/40 CURRENT-SMOKER 35 9 M 7/17/13 CURRENT-SMOKER 64 10 M 1/10/23 CURRENT-SMOKER 61 11 M 8/21/30 CURRENT-SMOKER 70.83 12 M 5/21/27 CURRENT-SMOKER 70.5 13 M 11/7/46 CURRENT-SMOKER 34 14 M 10/7/38 CURRENT-SMOKER 31.25 15 F 2/5/38 EX-SMOKER 50 16 F 6/23/28 EX-SMOKER 45 17 F 10/6/18 EX-SMOKER 108 18 F 9/9/22 EX-SMOKER 75 19 M 10/14/18 EX-SMOKER 32 20 M 6/11 /44 EX-SMOKER 47 21 M 1/23/24 EX-SMOKER 100 22 M 1/19/37 EX-SMOKER 80 23 M 5/16/08 EX-SMOKER 10 24 M 7/12/31 EX-SMOKER 48 25 M 3/19/19 EX-SMOKER 75 26 M 6/15/29 EX-SMOKER 123 27 M 11/14/32 EX-SMOKER 64 28 M 7/13/38 EX-SMOKER 107.3 29 M 3/26/15 EX-SMOKER 43 30 M 10/20/43 EX-SMOKER 30 84 31 M 7/8/21 EX-SMOKER 47.25 32 M 2/11/20 EX-SMOKER 50 33 M 10/6/46 EX-SMOKER 72 34 M 11/2/32 EX-SMOKER 33 35 M 5/13/34 EX-SMOKER 45 36 F 8/3/52 NON-SMOKER 0 37 F 8/27/30 NON-SMOKER 0 38 F 1/29/30 NON-SMOKER 0 39 F 10/22/58 NON-SMOKER 0 40 F 6/7/60 ? ? 41 M ? ? ? MICRODISSECTED CASES Patient Microdissected Case Histological Grade DNA Concentration (ng/ul) Number 1 45 3.1 173.7 44 8.2 38.9 2 113 1 6.5 B 1 3.5 1 6.1 52.5 3 6.1 43.4 4 6.1 82 5 6.1 46.1 6 6.1 42.2 7 6.1 41.6 8 6.1 44.3 9 6.1 6.4 10 6.1 19.3 2 7.1 40.9 3 B 1 2.6 24 1 4.8 154 5.4 14.5 25 6.1 42.5 26 6.1 17.1 89 6.1 36 90 6.1 16.8 4 B 1 8.9 27 2.2 10 28 6.1 17.3 88 6.2 40.7 163 9.4 188.2 5 138 3.2 4 136 3.2 4.1 137 4.2 2.6 135 8.1 27.6 6 114 1 5.6 46 8.1 256.5 7 B 1 3.1 48 3.1 22.9 85 47 8.1 107.2 8 151 1 54.9 150 5.2 12.4 149 6.1 10.4 152 8.1 21.7 153 8.1 106.5 9 86 1 14.6 B 1 30.1 161 1 47.3 87 6.1 10.8 10 B 1 110.7 11 1 13.5 13 6.1 9.1 12 6.2 5.1 11 50 1 19.7 49 8.5 40.5 51 8.5 42.6 12 115 1 7 52 8.2 100.1 53 8.2 260.8 13 B 1 1.4 64 3.1 27.8 63 8.1 39.6 14 B 1 35.4 54 3.1 21.1 55 8.3 117.7 15 B 1 112.8 35 1 8.2 36 6.1 9.2 37 6.1 10.4 38 6.1 13.9 39 6.1 6.9 40 6.1 7.7 16 116 1 6.6 56 8.5 101.2 17 B 1 12.2 58 1 12 57 8.1 220 18 B 1 3.2 93 3.1 41.7 59 8.4 135.7 60 8.4 X 19 B 1 8.6 17 1 8.4 16 3.1 30 14 6.1 78 15 6.1 33.2 18 6.1 80.8 20 61 1 21.1 62 8.1 190.9 21 B 1 3.8 86 68 1 9.3 160 3.1 12.8 69 6.1 30.1 71 6.1 78.2 70 7.1 30.6 22 80 3.1 X 79 6.2 X 23 B 1 2.4 85 3.1 15.1 82 6.1 46.2 84 6.1 68.8 83 8.1 48 24 105 1 1.8 B 1 15.2 102 6.1 14.2 103 6.1 5 104 6.1 1.8 25 B 1 14 98 3.1 7 95 6.1 11.2 99 6.1 29.4 100 6.1 51.4 101 6.1 37.1 96 7.1 11.9 97 8.1 37.1 26 B 1 120 155 5.2 26.5 156 6.1 21.1 157 6.1 18.9 158 6.1 49.9 27 132 1 30.5 126 6.1 10.2 127 6.1 9.9 128 6.1 5.8 129 6.1 7 130 6.1 6.5 131 6.1 24.7 28 109 1 3.6 139 4.2 2.6 141 4.2 3 140 5.2 2.4 142 5.2 5.4 107 5.4 6.3 106 6.1 3.8 108 6.1 9.9 110 6.1 10.2 112 6.1 21.9 111 8.1 24.1 29 B 1 13.3 134 3.1 112.7 133 7.1 48.5 87 30 124 1 2.3 122 6 4.4 123 6 3 125 8.3 52.5 31 B 1 4.6 119 1 5.9 118 6.1 25.5 32 B 1 35.8 43 1 7.4 171 4.1 5.2 173 4.2 7.4 166 4.2 30.4 170 4.2 18.5 174 4.2 10.7 175 4.2 10.9 176 4.2 20.3 169 5.1 11.5 172 5.1 14.8 167 5.2 15 165 5.4 38.5 164 5.4 16.4 168 5.4 12.2 177 6.1 31.9 178 6.1 75 179 6.1 53.1 180 6.1 20.9 41 7.1 10.3 42 7.1 20.9 33 B 1 11.6 81 1 14.4 74 6.1 19.9 76 6.1 17.4 77 6.1 52.6 78 6.1 X 75 8.1 47.6 34 147 1 20.2 B 1 2.5 148 5.4 4.2 143 6.1 29.3 144 6.1 29.3 145 6.1 66.7 146 6.1 14.1 35 B 1 7.8 30 1 8.5 32 1 13.7 29 8.2 20.1 31 8.2 X 36 B 1 159.7 92 3.2 83.7 162 3.2 178.8 19 6.1 56.5 20 6.1 48.1 21 6.1 62.4 37 B 1 4.9 23 1 12.6 22 6.1 15.3 94 9.4 21 38 73 1 9.2 117 1 3.6 72 6.1 28.3 39 121 1 12.4 120 8.1 32.2 40 66 3.1 9 67 5.4 10.8 65 6.1 13 41 B 1 13.5 34 6.1 16.4 33 7.1 10.5 91 9.4 24.6 APPENDIX 3. CLONING PROCEDURES Preparation of Competent E.coli Cells by CaCl? Treatment. Preparation 1. Pre-cool centrifuge (the one in Ling lab) to 4 degrees Celsius 2. Pre-cool 0.1M CaC12 solution to 4 degrees Celsius Growing cells 3. Grow cells to 0.5 units at OD600 4. Divide into 50 ml aliquots Harvesting Cells 5. Chill cells on ice (10 min) 6. Spin cells at 3K rpm for 10 min, keep cold! 7. Pour off medium and drain (on ice for 30 sec), pipet off medium Washing and resuspending cells 8. Add 0.8 ml of ice cold 0.1M CaC12 and resuspend cells by pipetting (use filtered tip) 9. Add 15 ml of cold 0.1M CaC12, invert to mix 10. Spin cells at 3K rpm for 10 min, keep cold! 89 11. Pour off medium and drain (on ice for 30 sec), pipet off medium 12. Add 0.8 ml of ice cold 0.1M CaC12 and resuspend cells by pipetting (use filtered tip) 13. Add 1.2 ml of 0.1 M CaC12 and store cells in cold room 14. Cells are useful for two days G-Track Sequencing Similar to standard sequencing except it involves the use of only one dideoxynucleotide, ddGTP. (see Colony Fingerprinting for protocol). Plasmid Preparation Protocol 1. From the plate of bacteria take one colony and inoculate 2mL LB broth containing the required antibiotic 2. swirl flask and place 1.5mL culture into 1.7mL microcentrifuge tube 3. spin lmin. 14000rpm 4. remove supernatant by dumping back into culture stock (which we put bleach into to kill): remove the last remaining bit with a pipett 5. resuspend fully in lOOuL TE (lOmM tris, ImM EDTA) 6. Add 200uL alkaline lysis buffer (1%SDS, 0.2MNaOH) 7. Allow the samples to sit for lOmin FROM HERE EVERYTHING MUST BE ON ICE 8. addl50uLKOAc(3M,pH5.5) 9. shake 10. place on ice for 15 min 11. spin samples in cold room (4 degrees) 14000rpm for lOmin 12. place the supernatant into new tubes using 200uL pipett (toss the old ones) 13. add 2.5X volume ethanol (100%) 14. place samples in -20 for 30 min 15. spin samples in 4 degrees for 15min 14000 rpm 16. remove ethanol and airdry 90 17. add lOOuL TE (make sure to resuspend pellet completely: do not want to see flakes- vortex and spin down) 18. add luLRnase 19. sit for 30 min 20. add equal volumes of phenol/ shake/let sit for 5min/ spin for 5min to separate 21. take off aq. (top) layer (cut pipette tip): keep phenol tube for disposal 22. "guess and check" vol. with pipette (approx. 80uL) 23. add 1/10 vol NaOAc and 2.5X vol. Ethanol 24. place in -20 for 15 min. 25. spin in 4 degrees for 15min 14000 26. remove supernatant (if smells of phenol wash with 70% ethanol) -get EtOH from freezer -put in slowly 200uL as to not disrupt pellet -let sit for 5-10min -spin down and remove supernatant -resuspend in TE for freezing 27. freeze the sample in TE TE: dissolves DNA KOAc: ppt protein and cell debris Phenol: protein goes into phenol and DNA goes into aqueous layer (top) Sequencing Protocol 1. Big Dye reaction: Big Dye Terminator 4ul Primer 1.6 uM (T7) 2ul (final cone. = 3.2 pmol) Plasmid 200ng ddH20 adjust lOul 2. PCR cycling: 94°C 2 min 96°C 10 sec ^ 50°C 5 sec I 28 cycles 60°C 3 min J 4°C 3. Pelleting: a. Add all of the Big Dye Rxn (10 ul) to a tube with: 2ul 3M sodium acetate, pH 4.6 50ul 95% EtOH b. Vortex. Incubate at room temperature, 15 min for long products, 1 hr for short products, to precipitate the extension products. c. Centrifuge for 20 min at max speed d. Remove the supernatant with pipet and discard. 91 e. Rinse the pellet with 250ul of 70% EtOH. Vortex. f. Spin for 5 min. Remove the supernatant and discard. Air-dry the pellet to evaporate all of EtOH. There should be a tiny pink pellet. 92 


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