Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Molecular genetic characterization of pediatric malignant fibrous histiocytoma Palmer, Jessica Lynn 1997

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

Item Metadata

Download

Media
831-ubc_1997-0583.pdf [ 12.21MB ]
Metadata
JSON: 831-1.0099200.json
JSON-LD: 831-1.0099200-ld.json
RDF/XML (Pretty): 831-1.0099200-rdf.xml
RDF/JSON: 831-1.0099200-rdf.json
Turtle: 831-1.0099200-turtle.txt
N-Triples: 831-1.0099200-rdf-ntriples.txt
Original Record: 831-1.0099200-source.json
Full Text
831-1.0099200-fulltext.txt
Citation
831-1.0099200.ris

Full Text

MOLECULAR GENETIC CHARACTERIZATION OF PEDIATRIC MALIGNANT FIBROUS HISTIOCYTOMA by JESSICA LYNN PALMER B.Sc, The University of British Columbia, 1993 A THESIS SUBMITTED IN P A R T I A L F U L F I L L M E N T O F T H E REQUIREMENTS FOR T H E D E G R E E OF M A S T E R S O F SCIENCE i n T H E F A C U L T Y OF G R A D U A T E STUDIES (Department of Pathology) We accept this thesis as conforming to the required standard T H E UNIVERSITY O F BRITISH C O L U M B I A 1997 © Jessica Lynn Palmer, 1997 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 of 7 The University of British Columbia Vancouver, Canada DE-6 (2/88) 11 A B S T R A C T Cytogenetic and molecular genetic studies were performed on a pleiomorphic sarcoma diagnosed as malignant fibrous histiocytoma (MFH), removed from the heart of a 15 year old girl. M F H is the most common soft tissue sarcoma in adults, however, it is extremely rare in the pediatric population. Accurate pathologic classification of M F H is difficult because these primitive sarcomas lack recognizeable features of differentiation using conventional pathological techniques. Recently, analysis of chromosomal alterations in human malignancies has revealed that recurring genetic changes are often closely associated with specific subtypes of tumours, and can be useful for diagnostic and prognostic classification. To date, there are no cytogenetic alterations considered to be diagnostic for M F H . Cytogenetic analysis of the present case revealed a complex karyotype with a near-tetraploid chromosome complement and with several alterations previously reported to occur non-randomly in adult M F H , including abnormalities of chromosomal band 19pl3. Since 19pl3 aberrations are relatively common in M F H and result in 19p+ derivative chromosomes that are associated with a high relapse rate, we chose to further characterize chromosome 19 abnormalities in M F H . To investigate whether there is a chromosomal deletion in the derivative 19 chromosomes, loss of heterozygosity (LOH) studies were performed by microsatellite analysis, revealing no allelic loss (i.e. no gross deletions). Fluorescence in situ hybridization (FISH) was used to more accurately define the chromosome 19 aberrations. Using a chromosome 19 painting probe, three derivative 19 chromosomes and two normal 19 chromosomes were identified in this M F H . Dual-coloured FISH with 19q and 19p specific probes indicated that two of the extra pieces of chromosome 19 originated from 19q, and the other from 19p. Furthermore, using chromosome 4 and 19 I l l painting probes, two t(4;19) translocation derivatives were identified, which we believe represent the two 19q derivatives. Moreover, the tumor demonstrated homogeneously staining regions (HSRs) and double minutes (dmins) suggestive of gene amplification. Genes localized to chromosomal bands 12ql3-14, including the putative proto-oncogenes MDM2, CDK4, SAS, CHOP, and GLI, are frequently amplified and overexpressed in adult M F H . We therefore screened the present case for amplification of these genes. Southern and Northern blot analysis demonstrated co-amplification of MDM2, CDK4, SAS, and CHOP. Not only do the results in this study provide cytogenetic and molecular genetic evidence that pediatric and adult M F H are histogenetically related entities, but it may also provide valuable information regarding the identification of potential tumour markers in pediatric and adult M F H . i v T A B L E O F C O N T E N T S Page A B S T R A C T ii T A B L E O F C O N T E N T S i v LIST O F T A B L E S v i i LIST O F FIGURES v i i i LIST O F A B B R E V I A T I O N S xi A C K N O W L E D G E M E N T S xiii Chapter I I N T R O D U C T I O N 1 A . Soft Tissue Sarcomas 1 1. Introduction 1 2. Malignant Fibrous Histiocytoma (MFH) 2 3. M F H in the Pediatric Population 6 B. General Concepts in Cancer Genetics 6 1. Introduction 6 2. Chromosomal Aberrations in Cancer 9 3. Genetic Aberrations in Cancer 11 3.1. Loss of Tumour Suppressor Genes 11 3.2. Activation of Proto-oncogenes 13 3.2.1. Gene Amplification 13 3.2.2. Gene Rearrangements and Mutation 16 3.3. Chimaeric Genes 16 4. Cell Cycle Regulation and Growth 19 4.1. The Cell Cycle 20 4.2. Cell Cycle Checkpoints 20 4.3. Cell Cycle Regulators and Oncogenesis 22 p53 and M D M 2 23 Cycl in-CDK Complexes 23 C D K Inhibitors 25 C. Genetic Diagnosis in Pediatric Soft Tissue Tumours 26 1. Diagnostic Difficulties of Pediatric Tumours 26 2. Cytogenetic and Molecular Genetic Anaylses 27 2.1. Cytogenetic Analyses 27 2.2. Molecular Cytogenetic Analyses 30 2.3. Molecular Genetic Analyses 32 D. Thesis Objectives 34 E. References 35 V Page Chapter II M A T E R I A L S A N D M E T H O D S 43 A . Introduction 43 1. Case Report 43 2. Pathology 43 3. Cytogenetics 44 B. Tissue Culture Techniques and Cytogenetic Analysis 47 C. Isolation of D N A and R N A 48 D. Southern and Northern Blot Analysis 49 E. Chromosome 19 PCR-Based Microsatellite Analysis 50 F. Slide Preparation for Fluorescence in situ Hybridizatyion (FISH) 51 G. Whole Chromosome FISH: Chromosome 19 Painting Probe 53 H . Dual-Coloured FISH: 19ql3.1-Specific D N A Probe and 19p Cosmid Probes 54 I. Dual-Coloured FISH: Chromosomes 4 and 19 Painting Probes 56 J. Chromosome G-Banding and D API/PI Staining 57 K. References 57 Chapter III A M P L I F I C A T I O N O F M U L T I P L E G E N E S F R O M T H E 12ql3-14 C H R O M O S O M A L R E G I O N IN PEDIATRIC M F H 60 A . Introduction 60 B. Results 61 1. M D M 2 Gene Amplification and Overexpression 61 2. CDK4, SAS, C H O P , and GLI Amplification and Overexpression 62 3. Comparative Genomic Hybridization (CGH) 67 C. Discussion 71 D. References 78 Chapter IV I N V E S T I G A T I O N O F P U T A T I V E T U M O U R SUPPRESSOR G E N E I N V O L V E M E N T O N C H R O M O S O M E 19 IN PEDIATRIC M F H 82 A . Introduction 82 B. Results 84 1. Chromosome 19 L O H Analysis 84 2. INK4d: A Candidate Gene 88 C. Discussion 90 D. References 96 vi Page Chapter V F L U O R E S C E N C E IN SITU HYBRIDIZATION T O C H A R A C T E R I Z E C H R O M O S O M E 19 A B N O R M A L I T I E S IN PEDIATRIC M F H 100 A . Introduction 100 B. Results 101 1. Whole Chromosome FISH Using a Chromosome 19 Painting Probe 101 2. Chromosome Banding of C O A T A S O M E 19 Hybridized Metaphases 103 3. Dual-coloured FISH Using a 19ql3.1-Specific D N A Probe and 19p Cosmid Probes 107 4. Dual-Coloured Whole Chromosome FISH Using Chromosome 4 and 19 Painting Probes 115 C. Discussion 122 D. References 138 Chapter VI S U M M A R Y A N D C O N C L U S I O N S 143 A . Amplification of Multiple Genes From Chromosomal Region 12ql3-14 144 B. Chromosome 19 Microsatellite Analysis Reveals No L O H 144 C. FISH Identifies Two t(4;19) Derivative Chromosomes 145 D. General Comments 146 E. References 147 LIST O F T A B L E S Table 1. Proto-oncogene Amplification in Human Tumours Table 2. Characteristic Genetic Abnormalities of Soft Tissue Tumours Table 3. Characteristic Cytogenetic Aberrations in Malignant Bone and Soft Tissue Tumours Table 4. 12ql3-14 Gene Amplification and Expression in a Pediatric M F H (BCCH-Sn) Table 5. Chromosome 19 Microsatellite Analysis to Assess L O H in B C C H - S n vn Page 15 18 29 68 85 v i i i LIST O F FIGURES Page Figure 1. Schematic Presentation of the Development of Some Soft 3 Tissue Tumours in Relation to Age Figure 2. Approach to the Molecular Characterization of Recurring 8 Chromosomal Abnormalities in Cancer Figure 3. Schematic Presentation of the Common Structural 10 Chromosomal Changes Seen in Tumours Figure 4. The Somatic Cell Cycle 21 Figure 5. Molecular Regulators of the Cell Cycle 24 Figure 6. Histopathology of a Pediatric M F H (BCCH-Sn) 45 Figure 7. Representative G-banded Karyotype of a Pediatric M F H (BCCH-Sn) 46 Figure 8. Southern Blot Analyses of Genes From the 12ql3-14 63-64 Amplicon in Pediatric Sarcomas Figure 9. Northern Blot Analyses of Genes From the 12ql3-14 65-66 Amplicon in Pediatric Sarcomas Figure 10. The 12ql3-14 Amplicon in Pediatric M F H (BCCH-Sn) 69 Figure 11. B C C H - S n Comparative Genomic Hybridization (CGH) Profile 70 Analysis Figure 12. Amplification Units in 11 M F H Previously Shown to Have 74 M D M 2 Amplification Figure 13. Chromosome 19 Polymorphic Genetic Markers Used to 86 Investigate L O H in Pediatric M F H With Chromosome 19 Abnormalities Figure 14. Representative Autoradiographs of Two Microsatellite 87 Markers Used to Assess L O H in B C C H - S n ix Page Figure 15. Southern Blot Analysis of p l9 iNK4d in Pediatric Sarcomas 89 Figure 16. Whole Chromosome FISH Analysis With C O A T A S O M E 19 102 Total Chromosome Probe on a Metaphase Spread From Normal Human Fibroblasts Figure 17. Whole Chromosome FISH Analysis With C O A T A S O M E 19 104 Total Chromosome Probe on a Metaphase Spread From B C C H - S n (Short-Term Culture) Figure 18. Whole Chromosome FISH Analysis With C O A T A S O M E 19 105 Total Chromosome Probe on Two Metaphase Spreads From B C C H - S n (Long-Term Culture) Figure 19. B C C H - S n Metaphase Probed With C O A T A S O M E 19 and 106 Stained With D API /PI Figure 20. FISH Analysis With a 19ql3.1-Specific D N A Probe on 108-Metaphase Spreads and Interphase Cells From Normal 109 Human Lymphocytes Figure 21. Dual-Coloured FISH Hybridization of the 19ql3.1-Specific 110 D N A Probe and the 19p Cosmid Probes to a Normal Chromosome 19 Compared to a Chromosome 19 Idiogram Figure 22. Dual-Coloured FISH Analysis With the 19ql3.1-Specific D N A 112 Probe and the 19p Cosmid Probes on a Metaphase Spread and Interphase Cells From Normal Human Lymphocytes Figure 23. Dual-Coloured FISH Analysis with the 19ql3.1-Specific D N A 113 Probe and the 19p Cosmid Probes on Interphase Cells From B C C H - S n Figure 24. Dual-Coloured FISH Analysis with the 19ql3.1-Specific D N A 114 Probe and the 19p Cosmid Probes on a Partial Metaphase From B C C H - S n Figure 25. Dual-Coloured Whole Chromosome FISH Analysis With 117-C O A T A S O M E 19 Total Chromosome Probe and a 118 Chromosome 4 Painting Probe on a Metaphase Spread From Normal Human Lymphocytes Page Figure 26 Dual-Coloured Whole Chromosome FISH Analysis With 119-C O A T A S O M E 19 Total Chromosome Probe and a 120 Chromosome 4 Painting Probe on Metaphase Spreads From B C C H - S n Figure 27. Putative Series of Genetic Events in B C C H - S n Resulting in 125 Three Aberrant Chromosome 19 Derivatives and Two Normal 19 Chromosomes LIST O F A B B R E V I A T I O N S add a d d i t i o n a l c h r o m o s o m a l m a t e r i a l B A C bacter iophage a r t i f i c ia l ch romosome C A P c h r o m o s o m e a r m p a i n t i n g probes C D K cyc l in -dependen t k inase C D K I cyc l in -dependen t k inase inh ib i to r s C G H compara t ive genomic h y b r i d i z a t i o n D A P I 4' , 6 -d i amid ino -2 -pheny l i ndo l e d i h y d r o c h l o r i d e hydra te d e l d e l e t i o n of c h r o m o s o m a l m a t e r i a l D E P C d ie thylpyrocarbonate der d e r i v a t i v e c h r o m o s o m e D F S P de rmatof ib rosa rcoma protuberans D M S O d i m e t h y l s u l p h o x i d e d m i n d o u b l e m i n u t e c h r o m o s o m e s D N A deoxyr ibonuc le ic ac id E D T A ethylenediaminetetra-acetate E M C extraskeletal m y x o i d chondrosa rcoma F B S fetal b o v i n e s e r u m F I S H f luorescence in situ h y b r i d i z a t i o n F I T C f luoresce in iso th iocyanate H S R h o m o g e n e o u s l y s t a in ing r e g i o n I D S R C T i n t r a - a b d o m i n a l desmoplas t i c s m a l l r o u n d ce l l t u m o u r Ig i m m u n o g l o b u l i n ITS i n s u l i n - t r a n s f e r r i n - s o d i u m seleni te L O H loss of heterozygosi ty m a r m a r k e r c h r o m o s o m e M F H m a l i g n a n t f ib rous h i s t i o c y t o m a M M S P m a l i g n a n t m e l a n o m a of soft parts m R N A messenger r i bonuc le i c ac id O N B olfactory neurob las toma P B D phosphate buffered detergent xii P C N A proliferating cell nuclear antigen PCR polymerase chain reaction P A C phage artificial chromosome PI propidium iodide p P N E T peripheral primitive neuroectodermal tumour Rb retinoblastoma RFLP restriction fragment length polymorphism R M S rhabdomysarcoma R N A ribonucleic acid R T - P C R reverse transcriptase polymerase chain reaction SSC sodium chloride/sodium citrate T E M E D N,N,N^NMetramethylethylenediamine T M 4 S F transmembrane 4 superfamily protein T S G tumour suppressor gene W C P whole chromosome painting probe Y A C yeast artificial chromosome X l l l A C K N O W L E D G M E N T S I would very much like to thank my supervisor Dr. Poul Sorensen for his guidance and scientific expertise, without whose inspiration this thesis would not have been possible. From the Sorensen laboratory, I wish to thank Jerian L im, Stevan Knezevich, and Beth Lawlor for their technical assistance and their friendship. I am extremely grateful to those in the laboratory of Dr. Dagmar Kalousek, especially Irene Barrett, who helped extensively with the FISH studies. I also want to express my appreciation to Sharon Masui and Joan Mathers from B.C. Children's Hospital for their contributions to this thesis, and to the members of my supervisory committee, Dr. Greg Bondy, Dr. Doug Horsman, and Dr. Keith Humphries, for their continued support and interest. Finally, I thank my family for their encouragement and patience, especially my husband Brent, who have enabled me to see this thesis to completion. CHAPTER I 1 I N T R O D U C T I O N A . SOFT-TISSUE S A R C O M A S 1. Introduction Embryologically, soft tissue originates predominately from mesoderm, with a minor contribution from neuroectoderm (1). Soft tissue sarcomas are a highly heterogeneous group of tumours that are classified histogenetically according to the adult tissue they resemble (2). For example, lipomas and liposarcomas are tumours that have recapitulated normal fatty tissue; rhabdomyosarcomas exhibit varying degrees of differentiation towards skeletal muscle; pPNETs contain cells of neural origin; and malignant ectomesenchymoma exhibits both myogenic and neural differentiation (1, 3). The majority of soft tissue sarcomas develop in the extremities, chest wall, mediastinum, and retroperitoneum; however, they may occur anywhere in the body (2). Although these neoplasms may occur at any age, some subtypes occur more frequently within a certain age population, as seen in Figure 1 (1). Compared with carcinomas and other neoplasms, soft tissue tumours are relatively rare, accounting for approximately 0.8-1% of all cancers and 2% of all cancer deaths (2). In the classification of soft tissue tumours described by Enzinger and Weiss, 15 different categories are described which include greater than 30 groups with approximately 200 subtypes (2). These tumours are usually divided into benign and malignant forms within the various histogenetic groups. Benign tumours more closely resemble normal tissue, as they possess limited potential for deregulated growth and have a low rate of local recurrence. Alternatively, malignant tumours 2 (sarcomas) have a high capacity for autonomous growth and are capable of recurrence and distant metastasis. The different sarcomas have variable likelihoods of metastasizing. Malignant fibrous histiocytoma (MFH), for example, is an aggressive sarcoma which metastasizes frequently (1). Consequently, a statement describing the degree of differentiation or the histological grade is critical to qualify the term sarcoma. The qualitative terms well-differentiated and poorly differentiated are used to describe the relative morphologic maturity of the lesion with respect to the corresponding normal tissue. By applying a set of histological criteria, histological grade is assigned to the tumour as a means of quantitating the degree of differentiation. Well-differentiated sarcomas are generally low-grade lesions and poorly differentiated sarcomas are high-grade lesions. However, when borderline cases present, it can be a challenge to assess their malignant potential. Cytogenetic findings in these tumours may contribute significantly to their differential classification and diagnosis. 2. Malignant Fibrous Histiocytoma (MFH) Malignant fibrous histiocytoma (MFH) is the most commonly diagnosed soft-tissue sarcoma of late adult life (2). The peak incidence is in the sixth to seventh decades, and it is rarely seen before 20 years of age (4, 5). The majority of cases occur in men, predominantly in the lower extremities (4). Since its first description in the early 1960s (6, 7), M F H has been a controversial entity because the histogenesis remains uncertain (4, 5, 8). Histiocytic, fibroblastic, and primitive mesenchymal origins of M F H have been proposed (4, 9-11). Weiss and Enzinger described it as "a primitive and pleomorphic sarcoma showing partial fibroblastic and histiocytic differentiation" (4). Recently, it was reported that the concept of M F H differentiation ("histogenesis") should show M F H as a "primitive, embryonal 3 P e a k A g e I n c i d e n c e o f S o m e S a r c o m a s Malignant Fibrous Histiocytoma Malignant Schwannoma Lfoosarcoma Fibrosarcoma Synovial Sarcoma Rhabdomyosarcoma 20 30 40 50 60 70 Age in Years Figure 1. Schematic presentation of the development of some soft tissue tumours in relation to age. From Sandberg and Bridge, 1994 (1). 4 fibrosarcoma" (5). Regardless of its precise histogenesis, both fibroblastic and histiocytic cells are found in the tumour in varying proportions, thus accounting in part for its wide morphological spectrum. Several M F H subtypes have been defined, including storiform-pleomorphic, myxoid, giant cell, inflammatory, and angiomatoid (2). The storiform-pleomorphic subtype is the most common form, accounting for greater than two-thirds of all cases (12). Classically, this M F H subtype which is considered the prototype for M F H (4), is composed of spindled (fibroblastic) and rounded (histiocytic) cells arranged in a storiform pattern, accompanied by pleomorphic giant cells and inflammatory cells (Figure 6, Chapter 2). Due to the primitive appearance of this malignancy and its highly variable morphologic pattern, M F H remains a significant diagnostic problem. Neither the clinical presentation nor gross appearance distinguish it from other types of sarcoma (4), and electron microscopy has a limited role, as there are no ultrastructural features which are specific for M F H (12). Consequently, the lack of consensus criteria for M F H has traditionally led to this diagnosis being assigned to a variety of potentially unrelated pleomorphic spindle cell sarcomas, such as pleomorphic rhabdomyosarcoma and liposarcoma (2, 4). Since M F H are usually high-grade sarcomas which will metastasize in approximately one-half of patients (4, 12), recognition of a given sarcoma as M F H is critical so that the appropriate aggressive treatment can be performed. Furthermore, morphological distinction of the different subtypes of M F H is also critical due to differences in biologic behaviour and prognosis. Differences are most significant for myxoid M F H (metastatic rate 23%; 5-year survival rate 66%) versus storiform-pleomorphic M F H (metastatic rate 42%; 5-year survival rate 30%) (5). Some reported prognostic factors in all subtypes of M F H include tumour size, location (particularly depth), mitotic activity, histological grade, D N A ploidy, and cytogenetic findings (1,13,14). 5 Comparatively few cytogenetic studies of M F H have been reported (1). As seen with many sarcomas, M F H often shows very complex karyotypes at diagnosis, but few recurrent nonrandom abnormalities have been identified (1, 13-15). Mandahl and coworkers described telomeric associations, ring chromosomes, and dicentric chromosomes in 11 (44%) of 25 M F H s studied (15). Similar structural aberrations have been reported in M F H by other investigators (14, 16-19). The most commonly reported chromosomal alterations are structural rearrangements of chromosomal bands l q l l , 3pl2, l l p l l , and 19pl3 (1, 14, 15). Currently, the genes affected by these alterations remain unknown. Genetic alterations of 19pl3, resulting in a nonrandom 19p+ marker chromosome, was reported in 9 (41%) of 22 (13) and in 19 (28%) of 69 M F H studied (14), and was associated with an increased recurrence rate (13, 14). This 19p+ marker chromosome has shown to be of prognostic value in M F H , and although the biological significance of 19p+ aberrations is presently unclear, it is believed to be biologically important (14). Nevertheless, there are currently no cytogenetic alterations considered to be diagnostic for M F H . Homogeneously staining regions (HSRs) and double minute chromosomes (dmins), which are suggestive of gene amplification, have been detected in M F H (19, 20). Gene amplification, particularly from the 12ql3-14 chromosomal region, has also been demonstrated in M F H (20 - 22). Proto-oncogenes identified in the 12ql3-15 amplicon include MDM2, CDK4, SAS, CHOP, and GLI (21, 23 - 27). Amplification of these genes has previously been reported in human sarcomas, and the frequency of involvement strongly suggests that their overexpression might contribute to tumourigenesis (21, 23 - 27). Presently, it is unknown which of these amplified genes, if any, is required for oncogenesis in M F H , or if there are as yet unidentified gene amplifications in these regions that contribute to tumourigenesis. 6 3. M F H in the Pediatric Population Despite its high frequency among adult sarcomas, M F H is rare in pediatric patients, comprising 2-6% of all pediatric sarcomas (2, 18, 28, 29). It is reported that only 3-10% of patients included in major reviews of M F H are under 20 years of age (4, 5, 30, 31). Histiocytic tumours in children were first described in the literature in the 1960s (9), and although several cases of pediatric M F H have been reported (18, 20, 28 - 33), it has been questioned whether this entity exists in the pediatric population. Due to its lack of differentiating features, previously, there has been diagnostic confusion of M F H with fibromatoses and more common pediatric soft tissue sarcomas such as rhabdomyosarcoma (30). Molecular characterization leading to the identification of tumour-specific markers may help to establish whether M F H exists as a distinct histogenetic entity in the pediatric population. Molecular genetic and cytogenetic findings indicating similarities between pediatric and adult M F H have been demonstrated (20), thus providing preliminary evidence that the childhood and adult forms of this pleomorphic sarcoma are histogenetically related. Such studies may have important implications for the treatment of this disease in the pediatric age group. B. G E N E R A L C O N C E P T S IN C A N C E R GENETICS 1. Introduction Cancer is a genetic disorder - although sometimes associated with inherited predisposing mutations, it is always associated with somatically acquired genetic changes. As early as 1900, it was proposed that these chromosomal changes result in neoplasia (34). However, the cytogenetic techniques available at that time were not able to identify consistent chromosomal changes, and this possibility was not 7 initially accepted. Since then, much improved cytogenetic and molecular genetic techniques have identified recurring and highly consistent chromosomal aberrations in various human neoplasms. The identification and subsequent molecular characterization of these tumour-specific markers has begun to influence clinical management, as it has led to novel approaches in the diagnosis, staging, and treatment of malignancies. Furthermore, since genetic changes are central to the initiation and progression of neoplasms, critical information regarding the mechanism of disease can be obtained through their identification and subsequent molecular analysis (Figure 2). It has been well established that most human tumours are monoclonal in origin and that tumourigenesis is a multi-step process resulting from an accumulation of genetic damage (35-37). The incidence of human cancer with age implies that an average of 6 - 7 events, with as many as 10 - 15 changes, over a span of 20 - 40 years may be necessary for transformation (36, 38). Although there is not a uniform basis for cancer cell development, two general mechanisms account for most of the genetic contributions to cancer: activation and suppression of gene activity. Mutations of two major classes of tumour-associated genes, oncogenes and tumour suppressor genes (TSGs), have been implicated in the tumourigenic process of many types of human neoplasms. Genes whose activation exert a dominant effect on cellular function and can cause malignant transformation are called oncogenes. The non-activated native forms of such genes are referred to as proto-oncogenes, and these do not become oncogenic until mutated or aberrantly activated. For example, genes involved in normal cell growth and development such as growth factors, growth factor receptors, and transcription factors, are classified as proto-oncogenes. O n the other hand, genes that when inactivated contribute to malignancy and act in a recessive manner, are called tumour 8 cytogenetic analysis recurring specific chromosomal alterations 1 molecular characterization of the aberrant genes / I \ new modalities new modalities insights into for diagnosis for determining mechanisms prognosis and of malignant treatment transformation Figure 2. Approach to the molecular characterization of recurring chromosomal abnormalities in cancer. 9 suppressor genes (TSGs) . N o r m a l l y , T S G s regulate the rate of ce l l d i v i s i o n or entry of cells in to a d i f ferent ia t ion pa thway . There is l i t t le ev idence that specif ic t u m o u r suppressor genes or oncogenes have an exc lus ive role i n tumour igenes i s ; rather, i t is l i k e l y that a m a l i g n a n t pheno t ype results f r o m an a c c u m u l a t i o n of m u l t i p l e genetic a l terat ions. It has also been h y p o t h e s i z e d that specific combina t ions of genetic aberrat ions occur i n v a r i o u s t u m o u r s to act i n coopera t ion w i t h each other. In h u m a n neu rob l a s toma , for example , i t has been suggested that d i s ta l de le t ion o n c h r o m o s o m e l p a n d a m p l i f i c a t i o n of the MYCN proto-oncogene, may be related (39, 40). 2. C h r o m o s o m a l A b e r r a t i o n s i n Cancer Cytogene t ic alterations i n neoplast ic cells can be d i v i d e d in to three classes. P r i m a r y aberrat ions are be l i eved to be of pa thogenic impor tance , and m a y represent the sole change a n d characterize a cer ta in t u m o u r type (41). Fo r example , the t ( l l ; 2 2 ) a n d t(21;22) t ranslocat ions i n E w i n g s a r c o m a / p e r i p h e r a l p r i m i t i v e neu roec tode rma l t u m o u r s ( p P N E T s ) (42, 43) and the t(12;16) t rans loca t ion i n m y x o i d l i p o s a r c o m a (44) are associated w i t h these ma l ignanc ies . D u r i n g t u m o u r p rogress ion , secondary aberrat ions, w h i c h m a y s h o w a n o n - r a n d o m pat tern, occur i n a d d i t i o n to the p r i m a r y changes. A d d i t i o n a l l y , 'cytogenet ic noise ' , a b a c k g r o u n d of m u l t i p l e , n o n - c l o n a l a n d s eeming ly r a n d o m al terat ions, can mask the p r i m a r y a n d secondary changes. Consequent ly , i n sarcomas w i t h c o m p l e x ka ryo types , such as osteosarcoma, r h a b d o m y o s a r c o m a , and ma l ignan t f ibrous h i s t iocy toma , it is more d i f f i cu l t to de termine the s ignif icant p r i m a r y and secondary changes (45). C o n c l u s i o n s can be reached, however , b y c o m p i l i n g cytogenet ic i n f o r m a t i o n f r o m a large n u m b e r of cases, a n d aberrat ions f rom p r i m a r y t umour s can be c o m p a r e d w i t h l o c a l recurrences a n d metastases f r o m the same t u m o u r to es tabl i sh p rogres s ion -10 CD T 3 C « S-: 3 o »-< •£ t>jo 54 o * ° C » bD £ w ^ 2 -o o -35 £ C C O • 2 >-i tL, O S3 /— o > QJ QJ ^ J-i 3 ^ O to e c £ cn at u 1/5 T 1 n w + ; 01 v c 1 i?.g hT) KS 5-1 11 related abnormalities. A l l chromosomes can be a target of clonal structural abnormalities (1). As depicted in Figure 3, the three most common structural chromosome changes seen in tumours are translocations, deletions, and inversions (1). In a reciprocal translocation, two chromosomes exchange material, usually resulting in a chimaeric gene fusion due to the in-frame juxtaposition of open reading frames of genes (or parts of genes) that are not normally contiguous. A deletion can involve submicroscopic segments or large chromosomal segments, and can occur anywhere in a chromosome. Deletions often result in loss of heterozygosity (LOH) if there are polymorphic loci in the regions. As seen with translocations, inversions lead to the juxtaposition of genes that are not normally contiguous. Inversions can be paracentric (when the break takes place in the same chromosomal arm) or pericentric (when one break occurs in each arm). 3. Genetic Aberrations in Cancer As a result of these chromosomal aberrations, various genetic changes can occur that ultimately cause the activation or suppression of gene activity. Major classes of genetic changes associated with malignancy include loss of tumour suppressor genes (TSGs), activation of proto-oncogenes, and production of novel proteins due to gene fusions. 3.1. Loss of Tumour Suppressor Genes The events resulting in the inactivation or loss of a tumour suppressor gene can range from a point mutation to complete loss of a chromosome. Regardless of the mechanism, since TSGs negatively regulate cell proliferation, both alleles of a T S G must be mutated or lost in order for cancer to develop (i.e. this is a "two-hit" model). Most often, mutations of the two alleles arise through different 12 mechanisms; and in heredity forms of cancer such as retinoblastoma, one allele is already mutated in the germline. A n individual with a germline mutation would be at a significantly higher risk for cancer development since only one additional mutation is required to result in a homozygous mutation. This can account for familial predisposition to malignancy, and such tumours often develop earlier than their sporadic equivalents and tend to be multifocal. In the general population, however, homozygous loss of a TSG happens in two stages, beginning with a normal cell. Since two rare events are required, tumour development is much rarer and will most likely be a unifocal clonal growth. To date, numerous tumour suppressor genes have been isolated and characterized, including the Wilms' tumour gene (WT1) on l i p , the retinoblastoma gene (RBI) on 13q, the p53 gene on 17p, the neurofibromatosis type 1 gene (NF1) on 17q (Reviewed in 46), and the BRCA1 gene on 17q in breast and ovarian cancers (47). Studies of allelic loss in human tumours have provided a useful tool for localizing and cloning such tumour suppressor genes (48). Since segmental D N A deletions near known or putative TSGs are often detected in tumours in which these genes play a role, mapping of regions of L O H can help identify putative tumour suppressor loci involved in oncogenesis (49). L O H studies provide an alternative and more powerful approach to identify loss of chromosomal material than before, when karyotyping alone was responsible for the identification of many of the chromosomal losses observed in human neoplasms (46). Advancements have also been made in the methods for assaying L O H . Previous genome scans for L O H relied on restriction fragment length polymorphisms (RFLPs), which are normal heritable variations in the restriction patterns of human genomic D N A . Now, the use of microsatellite markers in a polymerase chain reaction (PCR)-based analysis allows for a much higher power of discrimination. Approximately 50,000 -13 100,000 microsatelhte loci are believed to be distributed throughout the genome, each consisting of mono-, di-, tri-, terra-, or pentanucleotide repeats. PCR amplification using primer pairs that flank these polymorphic genetic markers results in differently sized PCR-products, which when separated on polyacrylamide gels and visualized (e.g. by autoradiography), result in unique banding patterns. Due to the high degree of polymorphism at microsatelhte loci, examination of these D N A sequences provides an enormous degree of specificity. Microsatellite analysis is ~ 3 times more informative than RFLPs, and can render 90% of chromosome arms informative for L O H (50). As a result, it is possible with this technique to obtain almost a complete description of L O H status in individual tumours. 3.2. Activation of Proto-Oncogenes As a result of gene activation, transcription levels of proto-oncogenes may be elevated, transcript stability may be increased, or the biological activity of the protein product may be increased. Genetic changes associated with malignancy that can result in proto-oncogene activation include gene amplification, gene rearrangement, and gene mutation. 3.2.1. Gene Amplification A n increase in gene copy number by D N A sequence amplification, which often leads to an increase in gene expression, occurs frequently in tumour cells (51). Cytogenetic evidence of gene amplification, however, is uncommon in primary tumour cell culture. Since a gene or group of genes can be replicated tens or hundreds of times, this is a potent way in which gene expression may be increased in the cell (52). By far, the most frequent target of gene amplification in vivo are proto-oncogenes (51). In fact, the oncogenic potential of some proto-oncogenes (e.g. MYCN, L-myc, and GLI) was discovered through their amplification in human tumours (51). Although amplified proto-oncogenes in tumours generally belong to 14 one of three families {erbB, ras, myc) or to the l l q l 3 locus, recent studies have demonstrated the involvement of other genes, such as MDM2 and SAS at the 12ql3-14 locus in a high proportion of human sarcomas (21, 23, 24, 52). MDM2, for example, is reported to be amplified in over a third of human sarcomas (23). High levels of M D M 2 , which binds the tumour suppressor p53, is believed to cause tumour growth by functional inactivation of p53 and subsequent deregulation of p53 growth control (23). Table 1 lists examples of proto-oncogene amplification in human tumours (51, 53). Cytogenetically, the amplified sequences appear as double minute chromosomes (dmins) and homogeneously staining regions (HSRs). dmins are small, paired, spherical structures and HSRs are chromosome segments of increased length that stain uniformly with banding methods, dmins are the most common form of amplified sequences in tumours; they are acentric and segregate passively at cell division. In contrast, HSRs, are integrated into a chromosome and often arise in primary tumours and when cells with dmin amplified genes are cultured in vivo. At the molecular level, Southern blot analysis and PCR can be used to detect gene amplification. Such molecular techniques are presently being used to determine MYCN copy number as a prognostic indicator in neuroblastoma. MYCN is amplified in -25% of all primary neuroblastoma tumours, and is associated with advanced stages of disease and a poor prognosis (54). Gene amplification is often seen more frequently in advanced-stage disease (51). The frequency of proto-oncogene amplification in other human tumours, such as erb-B2 and c-MYC in breast cancer, suggests that they might be used more frequently as additional prognostic markers in clinical oncology. Note that gene dosage may also be increased by aneuploidy, (i.e. increased chromosome number), although to a much less degree than with gene amplification. 15 Table 1. Proto-oncogene Amplification in Human Tumours Proto-oncogene Tumour Type erbB family: c-erbBl c-erbB2 ras family: N-ras myc family: c-myc h-myc MYCN Hql3 locus: Other: int-2/HST-l c-myb GLI MDM2/SAS Glioblastoma; squamous carcinomas Breast cancer; adenocarcinomas; ovarian cancer Lung carcinomas; head and neck squamous cell carcinomas Carcinomas of the lung; breast cancer Small-cell lung cancers (SCLC) Neuroblastomas; retinoblastomas; brain tumours; small-cell lung cancer Breast cancers; esophageal carcinomas; bladder carcinomas Breast cancer; head and neck carcinomas Gastrointestinal carcinomas; nonlymphoid leukemias; breast cancers Gliomas Sarcomas 16 3.2.2. Gene Rearrangements and Gene Mutation Gene rearrangements may result from translocations (which can lead to gene deletions at the breakpoint site) or inversions that may be grossly detectable by standard cytogenetic methods or they may be submicroscopic. Such gene rearrangements may have different consequences. One possibility is that a proto-oncogene is activated as a result of juxtapositioning to other genes that are subject to active transcription (i.e. a gene for a T-cell receptor or an immunoglobulin gene which are highly expressed in certain cells). T-cell receptor and immunoglobulin genes are often involved in chromosomal alterations because rearrangement to create active antigen-receptor genes normally occurs. The c-MYC gene translocation in Burkitt's lymphoma is an example where a proto-oncogene is juxtaposed to a highly expressed immunoglobulin gene by fusion with joining or diversity segments, thereby causing its activation (55). Alternatively, a portion of a gene that is responsible for regulation of transcription, m R N A stability, or protein activity could be altered through gene rearrangement. To screen tumours for gross rearrangements, Southern analysis and fluorescence in situ hybridization (FISH) can be performed. Several types of gene mutations, such as intragenic deletions and base-pair substitutions, may result in activated gene expression. As seen with gene rearrangement, the transcription rate, m R N A stability, or biological activity of the protein can be affected by such a mutation. Direct sequencing of PCR amplified D N A , along with several other techniques can be used to detect point mutations. 3.3. Chimaeric Genes Distinct translocations in leukemias and solid tumours result in either the activation of proto-oncogenes, as previously discussed, or in the creation of tumour-specific chimaeric proteins. The genes involved in these mechanisms often involve 17 transcription factors, thus indicating that altered transcription plays a significant role in oncogenesis (34). The chromosomal breaks can occur within a gene on each partner chromosome, so that the 5' portion of the gene from one chromosome fuses to the 3'portion of the gene from the other chromosome to result in a chimaeric or fusion gene. This fusion gene encodes a chimaeric protein with oncogenic properties. Genes involved in chromosomal translocations have been cloned and characterized in a number of different tumour types (56). For example, in chronic myelogenous leukemia the Philadelphia chromosome, t(9;22), results in the fusion of the BCR (22qll) and c-ABL (9q34) genes (57). Recently, investigation of solid tumour translocations has focused on sarcomas (Reviewed in 58). The genetic structural consequences of translocations in several sarcomas have now been determined, including liposarcoma, Ewing sarcoma / p P N E T , synovial sarcoma, alveolar rhabdomyosarcoma, intra-abdominal desmoplastic small round cell tumour (IDSRCT), malignant melanoma of soft parts (MMSP), extraskeletal myxoid chondrosarcoma (EMC), and dermatofibrosarcoma protuberans (DFSP) (Reviewed in 58, 59-61) and the resulting chimaeric proteins are known (Table 2). In Ewing sarcoma and related tumours, for example, the t(ll;22)(q24;ql2) translocation fuses the 5' portion of the EWS gene from band 22ql2 to the 3' portion of the FLI-1 gene from band Hq24 to result in the formation of chimaeric EWS/FLI-1 m R N A s and proteins, which appear to act as aberrant transcription factors (42). EWS is a ubiquitously expressed gene of unknown function, and FLI-1 is a member of the ETS gene family of transcription factors (42). Variant translocations which involve a common gene and a pair of specific chromosomal loci have been reported in several human malignancies, including Ewing sarcoma/pPNET and rhabdomyosarcoma (43,62). In Ewing sarcoma and related tumours, it has been shown that in addition to the t(ll;22) translocation 18 Table 2. Characteristic Genetic Abnormalities of Soft TissueTumours Chromosome Molecular Neoplasm Abnormality Target Myxoid liposarcoma t(12;16)(ql3.3;pll.2) CHOP-FUS Synovial sarcoma t(X;18)(pll.2;qll.2) SYT-SSX SYT-SSX2 Alveolar rhabdomyosarcoma t(2;13)(q37;ql4) PAX3-FKHR t(l;13)(P36;ql4)* PAX7-FKHR Ewing sarcoma/pPNET t(ll;22)(q24;ql2) EWS-FLI1 t(21;22)(q22;ql2)* EWS-ERG I D S R C T t(ll;22)(pl3;ql2) EWS-WF1 E M C t(9;22)(q22-31;ql2) EWS-TEC M M S P t(12;22)(ql3;ql2) EWS-ATF1 D F S P t(17;22)(q22;ql3) PDGFp-COLlAl Neuroblastoma del(lp) MYCN amp. pPNET - peripheral primitive neuroectodermal tumour IDSRCT - intra-abdominal desmoplastic small round cell tumour EMC - extraskeletal myxoid chondrosarcoma MMSP - malignant melanoma of soft parts DFSP - dermatofibrosarcoma protuberans * variant translocation 19 found in -85% of these tumours, a variant t(21;22) translocation is found in approximately half of the 15% that do not show t(ll;22) (10). As a result of this alteration, the EWS gene is fused to ERG (located on band 21q22), which encodes another ETS family transcription factor (43). Consequently, an E W S / E R G hybrid protein is expressed. Since two distinct chromosomal aberrations can occur in the same tumour type, it is believed that the resulting chimaeric products from different genomic rearrangements can act in the same or in different oncogenic pathways in phenotypically similar tumours (43). This suggests that there may be redundancy in oncogenesis at both the genetic and biological levels. 4. Cell Cycle Regulation and Growth In higher eukaryotes, somatic cell growth and division is normally strictly regulated. Cancer cells are an exception, arising as genetic variants that have lost their regular growth control. When a cell becomes malignant, three types of properties distinguish it from a normal cell: immortalization, transformation, and the ability to metastasize (36). Immortalization describes the characteristic of indefinite potential for cell division, transformation is the feature that describes the failure to heed normal growth constraints, and metastasis describes the ability of the cancer cell to move away from the tissue of origin and establish a new colony at a distant site. Due to deregulated cell growth and division in neoplastic cells, an imbalance between the growth of new cells and the death of senescent cells occurs, resulting in tumour formation. Various clinical observations, such as different growth rates occurring in different areas of the same tumour and pathologically indistinguishable tumours growing at significantly different rates in different individuals, suggests that environmental factors influence tumour growth 20 regulation, in addition to the self-regulation of the tumour cells (63). Knowledge of the molecular mechanisms that mediate this regulation has emphasized the significance of the cell cycle as a framework within which to understand these events. 4.1. The Cell Cycle The somatic cell cycle is defined as the period between two mitotic divisions, with the M phase (mitotic) being the period of actual cellular division. This period begins with nuclear division and ends with cytoplasmic division to result in two identical daughter cells, each containing a diploid (2n) complement of chromosomes. However, nuclear division in the absence of cellular division occurs in some tumours, as seen by hyperdiploidy in many childhood tumours (63). The period from the end of one mitosis to the start of the next is called interphase, and is divided into the GI, S, and G2 phases, as depicted in Figure 4. After mitosis the cells enter GI phase, during which R N A and proteins are synthesized, but D N A is not replicated. The cells then move from GI phase into S phase, where D N A replication occurs. After D N A synthesis is complete, the cells enter G2 phase before beginning mitosis. Variations in the duration of GI are largely responsible for the differences in cell cycle duration that characterize neoplastic cells (36). The length of GI is adjusted in response to growth conditions, and it is during this phase that normal cells may become quiescent and undergo growth arrest (GO phase) or continue through the cell cycle. 4.2. Cell Cycle Checkpoints In the current model of cell cycle control, transitions between different cell cycle states are regulated at checkpoints (64), which prevent the cycle from proceeding until a particular condition has been satisfied, for example, adequate nutritional state of the medium and adequate mass of the cell. The two major 2 1 Figure 4. The somatic cell cycle: interphase is divided into the G I , S, and G2 phases; mitosis marks the boundary between one cell cycle and next. Cells may withdraw from the cycle into GO or reenter GI from it. From Lewin, 1994 (36). 22 control points in the somatic cell cycle are at G l / S and G 2 / M . One of the most important checkpoints is S T A R T in late G l (also known as the restriction point in mammalian cells), at which point the cell is committed to another round of replication. It is at the G l restriction point when both positive and negative extracellular signals that regulate cell division are integrated into the cell cycle. These checkpoints are regulated by a family of protein kinases, the cyclin-dependent kinases (CDKs), and their obligate activating partners, the cyclins. Specifically, the cyclins D and E , and CDK2, CDK4, and CDK6 control the G l to S phase transition (65). C D K inhibitors (CDKIs), on the other hand, block the cell cycle by inhibiting the kinase activities of these complexes. The CDKI, pl6, for example, inhibits the kinases activities of CDK4 and CDK6, thus preventing transit between G l and S and suppressing cellular proliferation (66). In oncogenic cells, many checkpoints are deregulated as a result of alterations in these cyclin-CDK complexes or CDKIs (64, 67). This could be due to altered expression of positive regulators (e.g. cyclins) or loss of negative regulators (e.g. CDKIs). The deregulation of S T A R T in particular, may permit cell growth and division to become insensitive to external signals and it may allow the cells to replicate unrepaired D N A mutations, thereby enabling the accumulation of genetic changes which further contribute to the tumour phenotype. A n understanding of the molecular events involved in regulation at the S T A R T and other checkpoints, has obvious implications for improving our understanding of mechanisms that lead to unregulated cellular growth. 4.3. Cell Cycle Regulators and Oncogenesis Knowledge regarding the number of recognized connections between tumourigenesis and cell cycle regulators has increased significantly over the past few years. Several proto-oncogenes are also known to be associated with the cell 23 cycle. For example, proto-oncogenes such as myc are involved in the G O / G I transition by cooperating with cyclin-Dl in transforming cells. Myc induces cyclin-D l when inducibly expressed in quiescent fibroblasts (68), thus providing a link between mitogenic signalling pathways and GI progression. Considerable efforts have been made to better understand the regulation of tumour suppressor proteins, such as p53 and the retinoblastoma protein (Rb), during the cell cycle. As well, the role that cyclins, CDKs, and CDKIs play in the cell cycle and oncogenesis is now much more thoroughly defined. The following sections will describe some of these cell cycle regulators in greater detail and their interactions are depicted in Figure 5. p53 and M D M 2 . The p53 gene encodes a protein which binds to specific nucleotide sequences (p53-responsive elements) and acts as a transcription factor that activates or represses a variety of genes, some of which are involved in growth regulation, including cell cycle arrest and apoptosis after D N A damage (69, 70, 71). p53 mutations are commonly found in carcinomas and sarcomas (23, 69) and result in aberrant p53 proteins that fail to act as a transcription factor at a p53-responsive D N A element (20). This loss of function that contributes to cancerous growth, coincides with the fact that p53 is a tumour suppressor gene. A p53 regulator is M D M 2 , a protein which can complex with p53 to inhibit p53-mediated transcriptional activation (23, 72). MDM2 is amplified in greater than one-third of human sarcomas (23), and the data suggests that the overexpression of M D M 2 in these tumours inactivates the wild-type p53 transcription factor activity (23, 73, 74). Several studies suggest that the p53 and M D M 2 genes function along the same tumour-suppression pathway, and may therefore be inactivated by alterations to either gene (23, 71, 72). The tumourigenic activity of high-level M D M 2 expression is also thought to derive from the binding and functional inactivation of Rb by M D M 2 24 Growth signals G l CDK4 <+> + M Y C • cyclin DI (+) (+) DNA damage • p53 (e.g. y radiation) i ± ) MDM2 (-) C D K I -(e.g. p21, pl6, pl9) H Rb-E2F Rb-P + E2F (unbound) ( + ) Transcription of S phase genes (e.g. D N A poi) Figure 5. Molecular regulators of the cell cycle. Rb is a critical negative regulator of the cell cycle. Rb regulates the E2F transcription factor (TF) which induces transcription of genes required for cell cycle initiation (i.e. S phase). Only when Rb is inactivated (i.e. phosphorylated or mutated) can these genes be expressed. Unphosphorylated Rb binds to the transcription-activating domain of E2F, thereby converting it to a transcription-suppressing TF . Phosphorylation of Rb by cyclin/ C D K s results in loss of E2F binding, thus restoring the transcription-activating function. (+) indicates positive regulation; (-) indicates negative regulation; Rb-P indicates phosphorylated Rb. 2 5 (75). Cycl in-CDK Complexes. The cyclin-CDK complexes most closely associated with S T A R T regulation in mammalian cells are the D-type cyclins and their C D K partners, primarily CDK4. The D-type cyclins (DI, D2, and D3) activate CDK4, and in some cases CDK6, to drive the cell through START. There is considerable evidence suggesting that the amplification and overexpression of C D K 4 may promote neoplastic cell growth (52). Amplification of the CDK4 gene and alterations in other cyclin/ C D K family members have been reported to play critical roles in the development of some tumours (52, 67, 76). Cycl in-Dl , for example, is strongly implicated in oncogenesis (67). Cyclin DI , encoded by the CCND1 gene which has been identified as the PRAD1 proto-oncogene and is the most likely candidate for the BCL1 proto-oncogene, has been shown to be overexpressed and altered in numerous human cancers (Reviewed in 67). It is believed that D-type cyclins are fundamental to cell cycle regulation of the tumour suppressor protein, Rb. Rb is underphosphorylated during G l phase and requires phosphorylation in order for the cell to enter S phase. Several observations in vitro have led to the proposal that the exclusive role of cyclin-Dl is to phosphorylate (i.e. inactivate) the Rb protein so that cells can enter S phase and replicate their D N A (64). The D-type cyclins and Rb also play a significant role in the switch between proliferation and differentiation. C D K Inhibitors. The inactivation of CDKIs, which potentially act as tumour suppressors, has also been implicated in tumourigenesis. CDKIs fall into two families, the INK4 and the CIP/KIP. The INK4 family is composed of a group of four structurally related proteins, p l6 iNK4a r p i5 iNK4b / p i8 iNK4c / and the newest member, p l9 lNK4d which inhibit CDK4 and CDK6 by competitively binding with cyclin D (77). Frequent biallelic deletions and mutations of lNK4a, have been found in several human malignancies, suggesting it can function as a tumour suppressor 2 6 (77). It is believed that p21Ci P i /WAFi / a CDKI from the CIP/KIP family, plays a role in cellular transformation. p21Cipl/WAFl binds to and inhibits a variety of cyclin-CDK complexes, and it also has shown to bind to proliferating cell nuclear antigen (PCNA), a component of the D N A replication machinery (78). Furthermore, the p21Cipl/WAFi gene is a target of p53, and appears to be essential to the p53-mediated arrest of the cell cycle in GI in response to D N A damage (79). C. G E N E T I C DIAGNOSIS IN SOFT TISSUE T U M O U R S 1. Diagnostic Difficulties of Pediatric Tumours Although the spectrum of pediatric neoplasms is more limited in variety than adult cancer, it is more challenging diagnostically (56). The greatest difficulty in childhood cancer diagnosis relates to accurate identification of often similar-appearing tumours, and recognition of tumour subgroups within a larger group. Although a handful of tumours account for the majority of pediatric malignancies, there can be considerable variety within a diagnostic group. Also, since histopathologic groups can be directly related to prognosis within a tumour type (56), accurate classification and the appropriate therapy is critical. For example, within kidney tumours, rhabdoid tumour and clear cell sarcoma are both more biologically aggressive than other tumours such as Wilms' tumours and consequently have poor prognoses (56). The development of effective treatment regimens tailored to specific tumour types has further demanded accurate diagnosis of even the most primitive types of pediatric malignancies. Therefore, in view of the possibility of successful therapy, the problem of undifferentiated or primitive tumour diagnosis in children is particularly concerning. Despite the significance of primitive tumours in childhood, however, 27 most pathologic and clinical studies of these entities have been performed on phenotypically similar tumours in the adult population with extrapolation of results to the pediatric population (29). As a result, there is a general lack of understanding of childhood forms of these malignancies in clinical presentation, staging, biologic behaviour, and response to treatment. 2. Cytogenetic and Molecular Genetic Analyses in Soft Tissue Tumours Traditionally, the diagnosis of malignancy has relied upon histologic and morphologic criteria using conventional pathological methods, such as electron microscopy, light microscopy, and immunohistochemistry. Some types of solid tumours, for example the small round cell tumours of childhood (e.g. neuroblastoma, rhabdomyosarcoma, lymphoma, and Ewing sarcoma), may be difficult to diagnose using these methods because of an ambiguous appearance or lack of specific differentiation. Like small round cell tumours of childhood, it can also be difficult to discriminate between different types of spindle cell tumours (e.g. M F H , synovial sarcoma, myxoid liposarcoma, and extraskeletal myxoid chondrosarcoma). However, recent advances in cytogenetic and molecular genetic techniques have led to the ability to identify specific chromosomal abnormalities and the resulting molecular alterations that are associated with morphologically and clinically distinct subsets of tumours. Detection of these recurring aberrations in a tumour may provide a definitive diagnosis, as well as information of prognostic significance. 2.1. Cytogenetic Analyses Cytogenetic analyses allows for the rapid assessment of chromosomal genetic aberrations in solid tumours (1) which includes whole chromosome gains and 28 losses, and deletions and rearrangements that involve large chromosomal regions. Although these methods permit identification of many of the critical alterations in solid tumours, they do not reveal the oncogenetic events that arise from point mutations and microdeletions. Such aberrations are often only apparent using molecular detection methods. However, since molecular methods (e.g. nucleic acid hybridization and sequencing) must be targeted to specific loci within the genome for efficient analysis, prior knowledge of putative genetically altered sites is required. Cytogenetic analyses, on the other hand, is based on the morphologic recognition of chromosome banding abnormalities across the genome, and therefore, does not require foreknowledge of potential genetic hotspots. The low-resolution genetic overview that cytogenetic analyses provides makes it a particularly useful approach in tumours that have not been extensively characterized at the genetic level. The cytogenetic examination of solid tumours has proven to be diagnostically and prognostically important through the identification of characteristic chromosome gains, deletions, and translocations (1, 58, 80). Characteristic cytogenetic aberrations common in malignant bone and soft-tissue tumours are listed in Table 3. Leukemias have the longest history of analysis for chromosome alterations due to its relative accessibility of tumour cells for culture, and many tumour-specific chromosomal markers have been identified in leukemias, for example the Philadelphia chromosome, t(9;22)(q34;qll), in chronic myelogenous leukemia (57), t(15;17)(q21;q21) in acute promyelocyte leukemia (57), t(8;21)(q22;q22) in acute myelogenous leukemia (57), and t(l;19)(q23;pl3) in pre-B acute lymphocytic leukemia (81). Although most solid tumours possess acquired nonrandom chromosomal aberrations, the study of chromosomes in solid tumours has been technically more challenging than in hematopoietic malignancies, and less is T a b l e 3. C h a r a c t e r i s t i c C y t o g e n e t i c A b e r r a t i o n s i n M a l i g n a n t B o n e a n d Sof t T i s s u e T u m o u r s . F r o m D r a c o p o l i , 1994 (53) . T u m o r type Cytogenetic aberration Frequency Clear ce l l sarcoma t ( 1 2 ; 2 2 ) ( q l 3 ; q l 2 ) 9 0 % Dermatof ib rosarcoma protuberans R ing chromosome 17 9 0 % Endomet r ia l st romal sarcoma t ( 7 ; 1 7 ) ( p l 5 ; q l l ) U n k n o w n E w i n g ' s sarcoma/peripheral P N E T t ( l l ; 2 2 ) ( q 2 4 ; q l l . 2 - 1 2 ) 9 5 % F i b r o s a r c o m a — i n f a n t i l e + 1 1 , + 1 7 , + 2 0 9 0 % In t raabdomina l desmoplastic smal l t ( l l ; 2 2 ) ( p l 3 ; q l 2 ) 7 5 % round cel l tumor Le iomyosarcoma l p - 7 5 % L i p o s a r c o m a — m y x o i d t ( 1 2 ; 1 6 ) ( q l 3 ; p l l ) 7 5 % L i p o s a r c o m a — p l e o m o r p h i c C o m p l e x0 9 0 % L iposarcoma—we l l -d i f f e ren t ia ted R ing chromosome 12 8 0 % Ma l ignan t f ibrous h is t iocytoma Complex" 9 0 % Ma l ignan t nerve sheath tumor Complex" 9 0 % Meso the l ioma l p - , 3 p - - 2 2 8 0 % Neurob las toma (good prognosis) Hyperd ip lo id , no l p - 9 0 % Neurob las toma (poor prognosis) 1 P - 9 0 % Osteosarcoma l p - , 6 q - , 9 p - 1 3 q - 9 0 % Rhabdomyosarcoma—a lveo la r t (2 ;13) (q35-37;q l4 ) 8 0 % R h a b d o m y o s a r c o m a — e m b r y o n a l +2q , +20 8 0 % Synov ia l sarcoma t ( X ; 1 8 ) ( p l l ; q l l ) 9 5 % "Indicates the consistent finding of extremely complex karyotypes containing multiple numerical and structural chromosome aberrations. 30 currently known about tumour-specific rearrangements in solid tumours than in hematological malignancies (57). Specifically, it has been difficult to induce the neoplastic cells of many solid tumours to divide in vitro, and the resulting metaphase chromosomes were often of poor quality. Over the past decade, however, advances in cell culture, preparation of cells for chromosomal analysis, and advances in methods for detecting chromosomal aberrations, such as FISH, have enabled significant progress in solid tumour cytogenetics, particularly in bone and soft tissue tumours (1). Cytogenetic information has traditionally been obtained by directly analysing chromosomes arrested in metaphase that exhibit characteristic banding patterns due to various staining methods. However, due to the presence of complex and subtle chromosomal rearrangements, cytogenetic analysis of human solid tumours by conventional banding techniques is often technically difficult, even when performed by highly experienced cytogeneticists (82). In such circumstances, it may be necessary to apply additional molecular cytogenetic and molecular methods to uncover the origin of these uninterpretable chromosomes. 2.2 Molecular Cytogenetic Analyses FISH is a relatively new method of D N A hybridization that has proven to be a sensitive and efficient method for identifying subtle chromosomal rearrangements that are not detectable with standard karyotypic analysis. Numerous studies have demonstrated reliable identification of chromosomal alterations by FISH (1, 83, 84). With this technique, the detection of specific D N A sequences occurs when target D N A (nuclear D N A of interphase cells or the D N A of metaphase chromosomes) hybridizes chromosome-specific D N A probes, most commonly labeled by enzymatic incorporation of biotin- or digoxigenin-labeled nucleotides. Detection of probe hybridization is achieved through a series of 31 treatments involving fluorescently-labeled detection reagents and visualization using a fluorescence microscope. A variety of probe-labeling schemes are now available, allowing simultaneous detection of two or more sequences in the same nucleus. Depending on the application, various types of probes can be used, such as region-, band-, or gene-specific probes which are valuable for the detection of structural abnormalities and centromeric probes to reveal aneuploidy and entire chromosome loss. Additionally, chromosome-specific D N A libraries have been developed which label entire chromosomes. These whole chromosome paints are excellent for identifying unknown, subtle rearrangements. Other uses of FISH include gene mapping, detection of gene amplification, and characterization and detection of translocation breakpoints. For example, in order to identify a specific translocation, cosmid clones that contain D N A from the flanking regions of the chromosomal breakpoints or Y A C s (yeast artificial chromosomes) that contain a D N A fragment which spans the specific translocation breakpoint can be used as FISH probes. For example, to screen a tumour for the t(ll;22) translocation characteristically found in Ewing sarcoma, cosmid clones from the flanking regions of the chromosome 11 and chromosome 22 breakpoints can be used to perform FISH (84,85). Not only do the applications of FISH continue to expand, but novel techniques are emerging. For example, comparative genomic hybridization (CGH) is a recently developed molecular cytogenetic method which can detect and map relative D N A sequence copy number between genomes (86). By comparing D N A from malignant and normal cells, a copy number karyotype can be generated for a tumour, thus identifying regions of gain or loss of D N A . With this method, differentially labeled tumour D N A and normal genomic reference D N A are hybridized simultaneously to normal metaphase spreads and are detected with two 32 different fluorochromes. Areas of genomic imbalances, such as amplification, duplications, or deletions, are seen as changes in the ratio of the intensities of the two fluorochromes along the target chromosomes (86) . 2.3. Molecular Genetic Analyses Presently, a central focus of cancer research is the study of the genetic changes involved in oncogenesis. Developments in molecular genetic analysis have now made it both practical and clinically beneficial to use tumour D N A or R N A to provide diagnostic, prognostic, and histogenetic information. Not only are molecular methods of analysis more specific than cytogenetic methods, but the molecular analysis of tumours do not require cultured cells and can detect small numbers of malignant cells in samples. Therefore, these techniques can be particularly helpful when cytogenetic methods are not possible due to lack of dividing cells and metaphase chromosomes. The first application of molecular genetic techniques for diagnosis involved immunoglobulin and T-cell receptor gene rearrangement studies by Southern blot analysis in hematopoietic malignancies (56) . For several years, chromosomal translocations in hematological malignancies have been useful for isolating genes involved in the development of leukemias (87) . Identification of the precise chromosomal bands involved in a specific cytogenetic change can guide studies in molecular approaches towards establishment of the precise genes affected. Recently, further analysis of the karyotypic findings in sarcomas has led to the identification of genes involved in recurring translocations and the resulting gene fusions of some sarcomas. Cloning of these aberrant genes and gene fusions enables D N A probes to be developed for various diagnostic tests, including Southern blot analysis, RT-PCR, PCR, and FISH. Southern blot analysis can be used to detect several types of genetic 33 alterations, including chromosomal translocations, gene amplification, deletions, and even single base pair mutations. The altered D N A structure of tumour cells can be identified as changes from the Southern hybridization pattern found in D N A of normal tissues. This method is very sensitive to low-copy number amplification and provides a means of quantitating amplification. Use of this method to detect MYCN amplification in neuroblastoma, for example, has become a routine clinical prognostic tool in many institutions (57). Similar hybridization analyses, called Northern blot analyses, are used to examine R N A and to determine expression levels of various genes in malignant and normal tissues. Other sensitive and specific molecular diagnostic tests have now been developed to detect many gene fusions found in pediatric soft tissue sarcomas (88). For example, in pPNETs, the unique fusion transcripts EWS/FLI-1 (42) and E W S / E R G (43) that result from the t(ll;22) and t(21;22) translocations, respectively, can be detected using the reverse transcriptase polymerase chain reaction (RT-PCR) (89). With this method, the fusion gene is identified by in vitro amplification of the m R N A transcribed from the fusion gene. Tumour cytogenetics and molecular genetics has become an important and sometimes decisive diagnostic modality in childhood tumour diagnosis. The emergence of these new diagnostic and prognostic parameters in clinical oncology is adding a new dimension to solid tumour diagnosis. Furthermore, these methods are helping to elucidate the nature of the effects resulting from these genetic alterations, thereby allowing us to appreciate the carcinogenetic processes which lead to tumour development. Ultimately, this will allow one to design therapies that are specifically tailored to the unique molecular events in different tumours and tumour-specific markers can be used as specific targets for therapy, for example, immunotherapy and anti-sense oligonucleotide treatment approaches. As further 34 technical advances are made, solid tumour genetic analysis will increasingly play a role in clinical management. D. THESIS OBJECTIVES Malignant fibrous histiocytoma (MFH) and other sarcomas are difficult tumours to diagnose and classify due to their primitive histologic appearances. Accurate diagnosis is of critical importance, as therapeutic strategies are individually tailored to specific tumour classes. M F H is the most common soft tissue sarcoma in adults but is considered a rare entity in the pediatric population. However, using conventional pathologic techniques, diagnostic criteria for M F H are ill-defined and the true incidence of this tumour in childhood remains unclear. Recent advances in the cytogenetic identification of chromosomal abnormalities in some human tumours and their molecular genetic characterization, has resulted in their use as sensitive and specific tumour markers which have valuable diagnostic and prognostic applications. To date, however, there are no known tumour-specific markers for M F H , although cytogenetic studies of M F H in adults have revealed recurring abnormalities of chromosomal bands l q l l , 3pl2, l l p l l , and 19pl3 (1, 13-20). The 19pl3 alterations frequently result in a 19p+ marker chromosome which has been correlated with an increased relapse rate (13, 14). Additionally, cytogenetic indices of gene amplification, namely double minute chromosomes (dmins) and homogeneously staining regions (HSRs), have been identified in some M F H (19, 20). The genes affected by these alterations are presently unknown. Ultimately, however, identification of the affected genes and determination of the genetic basis for tumour formation, wil l improve our understanding of the phenotypic processes that are involved in malignant transformation. Furthermore, the isolation of 3 5 tumour-specific markers will allow for more accurate diagnoses of both pediatric and adult M F H . The overall objective of my work, then, was to characterize molecular genetic abnormalities in pediatric M F H in order to identify similarities between pediatric and adult M F H , and to gain insights into putative tumour-specific alterations in M F H . The specific objectives were to: 1. Characterize gene amplification in a pediatric M F H (BCCH-Sn). 2. Characterize chromosome 19 aberrations in a pediatric M F H (BCCH-Sn). 2.1. Assess chromosome 19 L O H to determine whether chromosome 19 abnormalities represent a deletion of chromosome 19 material. 2.2. Characterize chromosome 19 abnormalities using fluorescence in situ hybridization (FISH). E. R E F E R E N C E S 1. Sandberg A A , Bridge JA (1994): The cytogenetics of Bone and Soft Tissue Tumors. R.G. Landes Company, Texas. 2. Enzinger F M , Weiss SW (1988) Soft Tissue Tumors. 2nd ed. St. Louis: C .V. Mosby. 3. Sorensen PHB, Shimada H , L iu XF, L i m JF, Thomas G , Triche TJ (1995): Biphenotypic sarcomas with myogenic and neural differentiation express the Ewing's sarcoma EWS/FLI1 fusion gene. Cancer Res 55:1385-1392. 4. Weiss SW, Enzinger F M (1978) Malignant fibrous histiocytoma: A n analysis of 200 cases. Cancer 41:2250-2266. 5. Meister P (1995): Malignant fibrous histiocytoma: A "fibrohistiocytic" or primitive, fibroblastic sarcoma. In: Current Topics in Pathology V. 89, eds. Harms D , Schmidt D, Springer-Verlag, Germany. 6. Ozzello L , Stout A P , Murray M R (1963): Cultural characteristics of malignant histiocytomas and fibrous xanthomas. Cancer 16:331-344. 36 7. O'Brien JE, Stout A P (1964): Malignant fibrous xanthomas. Cancer 17:1445-1458. 8. Roholl PJM, Prinsen I, Rademakers L P H M , Hsu S M , Van Unnik J A M (1991) Two cell lines with epithelial cell-like characteristics established from malignant fibrous histiocytomas. Cancer 68:1963-1972. 9. Kauffman SL, Stout A P (1961): Histiocytic tumours (fibrous xanthoma and histiocytoma) in children. Cancer 14:469-482. 10. Fu YS, Gabbiani G, Kaye GI, Lattes R (1975): Mlignant soft tissue tumours of probable histiocytic origin (malignant fibrous histiocytomas): general considerations and electron microscopic and tissue culture studies. Cancer 35:176-198. 11. Wood GS, Beckstead JH, Turner RR (1986): Malignant fibrous histiocytoma cells resemble firoblasts. A m J Surg Pathol 10:323-324. 12. Weiss SW (1982): Malignant fibrous histiocytoma. A m J Surg Pathol 6:773-784. 13. Rydholm A , Mandahl N , Heim S, Kreicbergs A , Willen H , Mitelman F (1990): Malignant fibrous histiocytomas with a 19p+ marker chromosome have increased relapse rate. Genes Chromosom Cancer 2:296-299. 14. Choong P F M , Mandahl N , Mertens F, Willen H , Alvegard T, Kreicbergs A , Mitelman F, Rydholm A (1996): 19p+ marker chromosome correlates with relapse in malignant fibrous histiocytoma. Genes Chromosom Cancer 16: 88-93. 15. Mandahl N , He im S, Willen H , Rydholm A , Eneroth M , Nilbert M , Kreicbergs A , Mitelman F (1989): Characteristic karyotypic anomalies identify subtypes of malignant fibrous hisitiocytoma. Genes Chromosom Cancer 1:9-14. 16. Mandahl N , Heim S, Kristoffersson U , Mitelman F, Rooser B, Rydholm A , Willen H (1985): Telomeric association in a malignant fibrous histiocytoma. H u m Genet 71:321-342. 17. Mandahl N , He im S, Arheden K, Rydholm A , Willen H , Mitelman F (1988): Rings, dicentrics and telomeric associations in histiocytomas. Cancer Genet Cytogenet 30:23-33. 18. Biegel JA, Perilongo G, Rorke LB, Parmiter A H , Emanuel BS (1992) Malignant fibrous histiocytoma of the brain in a six-year-old girl. Genes Chromosom Cancer 4:309-313. 37 19. Szymanska J, Tarkkanen M , Wiklund T, Virolainen M , Blomqvist C , Asko-Selijavaara S, Tukiainen E, Elomaa I , Knuutila S (1995): A cytogenetic study of malignant fibrous histiocytoma. 85:91-96. 20. Palmer JL, Masui S, Pritchard S, Kalousek D K , Sorensen PHB (1997): Cytogenetic and molecular genetic analysis of a pediatric pleomorphic sarcoma reveals similarities to adult malignant fibrous histiocytoma. Cancer Genet Cytogenet 95:141-147. 21. Smith S H , Weiss SW, Jankowski SA, Coccia M A , Meltzer PS (1992): SAS Amplification in Soft Tissue Sarcomas. Cancer Res 52:3746-3749. 22. Nilbert M , Mandahl N , Aman P, Rydholm A , Mitelman F (1994): No Rearrangements of the CHOP gene in malignant fibrous histiocytoma. Cancer Genet Cytogenet 72:155-156. 23. Oliner JD, Kinzler K W , Meltzer PS, George D L , Vogelstein B (1992): Amplification of a gene encoding a p53-associated protein in human sarcomas. Nature 358:80-83. 24. Khatib Z A , Matsushime H , Valentine M , Shapiro D N , Sherr CJ, Look A T (1993): Coamplification of the CDK4 gene with MDM2 and GLI in human sarcomas. Cancer Res 53:5535-5541. 25. Roberts W M , Douglas E C , Peiper SC, Houghton PJ, Look A T (1989): Amplification of the GLI gene in childhood sarcomas. Cancer Res 49:5407-5413. 26. Forus A , Florence V A , Maelandsmo G M , Meltzer PS, Fodstad O, Myklebost O (1993): Mapping of amplification units in the ql3-14 region of chromosome 12 in human sarcomas: some amplica do not include M D M 2 . Cell Growth Diff 4: 1065-1070. 27. Ladanyi M , Lewis R, Jhanwar SC, Gerald W, Huvos A G , Healey J H (1995): MDM2 and CDK4 gene amplification in Ewing sarcoma. J Path 175:211-217. 28. Raney RB, Allen A , Handler SD, Uri A , Littman P (1986): Malignant fibrous histiocytoma of soft tissue in childhood. Cancer 57:2198-2201. 29. Dil lon P, Maurer H , Jenkins J, Krummel T, Parham D, Webber B, Salzberg A (1992): A prospective study of nonrhabdomyosarcoma soft tissue sarcomas in the pediatric age group. J Pediatr Surg 27:214-245. 30. Cole C H , Magee JF, Gianoulis M , Rogers PCJ (1993): Malignant fibrous histiocytoma in childhood. Cancer 71:4077-4083. 38 31. Enjoji M , Hashimoto H , Tsuneyoshi M , Iwazaki H (1980): Malignant fibrous histiocytoma: a clinicopathological study.of 130 cases. Acta Pathol Jpn 30:727-741. 32. Kearney M M , Soule E H , Ivins JC (1980) Malignant fibrous histiocytoma: A retrospective study of 167 cases. Cancer 45:167-178. 33. K i m O H , Lee K Y (1994): Malignant fibrous histiocytoma of primary omental origin in an infant. Pediatr Radiol 24:285-287. 34. Rabbitts T H (1994): Chromosomal translocations in human cancer. Nature 372:143-149. 35. Therman Eeva, Susman M (1993): Human Chromosomes Structure, Behaviour, and Effects, ed. 3, Springer-Verlag, New York. 36. Lewin B (1994): Oncogenes: gene expression and cancer. In: Genes V . Oxford University Press, New York. 37. Orndal C, Rydholm A , Willen H , Mitelman F, Mandahl N (1994): Cytogenetic intratumor heterogeneity in soft tissue tumors. Cancer Genet Cytogenet 78: 127-137. 38. Marx J (1989): Many gene changes found in cancer. Science 246:1386-1388. Bieche I, Champeme M H , Lidereau R (1994): A tumor suppressor gene on chromosome lp32-pter controls the amplification of M Y C family genes in breast cancer. Cancer Res 54:4274-4276. 39. Bieche I, Chambeme M H , Lidereau R (1994): A tumor suppressor gene on chromosome lp32-pter controls the amplification of M Y C family genes in breast cancer. Cancer Res 54:4274-4276. 40. Caron H , Peter M , van Sluis P, Speleman F, de Kraker J, Laureys G, Michon J, Brugieres L , Voute PA, Westerveld A (1995): Evidence for two tumour suppressor loci on chromosomal bands lp35-36 involved in neuroblastoma: one probably imprinted, another associated with N-myc amplification. H u m M o l Genet 4:535-539. 41. He im S, Mitelman F (1987): Cancer Cytogenetics. Alan R. Liss, New York. 42. Delattre O, Zucman-Plougastel B, Desmaze C, Melot T, Peter M , Kovar H , Joubert I, de Jong P, Rouleau G (1992): Gene fusion with an E T S - D N A -binding domain caused by chromosome translocation in human tumors. Nature 359:162-165. 39 43. Sorensen PHB, Lessnick SL, Lopez-Terrada D , L iu XF, Triche TJ, Denny C T (1994): A second Ewing's sarcoma translocation, t(21;22), fuses the EWS gene to another ETS-family transcription factor, E R G . Nat Genet 6.T46-151. 44. A m a n P, Ron D, Mandhal N (1992): Rearrangement of the transcription factor gene C H O P in myxoid liposarcomas with t(12;16)(ql3;pll). Genes Chromosom Cancer 5:278-285. 45. Mitelman F (1991): Catalog of Chromosome Aberrations in Cancer, 4th ed. Wiley-Liss, New York. 46. Seizinger BR, Klinger P, Junien C, Nakamura Y, Le Beau M , Cavenee W, Emanuel B, Ponder B, Naylor S, Mitelman F, Louis D , Menon A , Newsham I, Decker J, Kaelbling M , Henry I, von Deimling A (1991): Report of the committee on chromosome and gene loss in human neoplasia. Cytogenet Cell Genet 58:1080-1096. 47. M i k i Y, Swensen J, Shattuck-Eidens D (1994): A strong candidiate for the breast and ovarian suseptibility gene BRCA1. Science 266:66. 48. Ritland SR, Ganju V , Jenkins RB (1995): Region-specific loss of heterozygosity on chromosome 19 is related to the morphologic type of human glioma. Genes Chromosom Cancer 12:277-282. 49. Marshall CJ (1991): Tumor suppressor genes. Cell 64:313-326. 50. Thrash-Bingham C A , Greenberg RE, Howard S, Bruzel A , Bremer M , Goll A , Salazar H , Freed JJ, Tartof K D (1995): Comprehensive allelotyping of human renal cell carcinomas using microsatellite D N A probes. Proc Natl Acad Sci U S A 92:2854-2858. 51. Brison O (1993): Gene amplification and tumor progression. Biochimica et Biophysica Acta 115:25-41. 52. Reifenberger G, Reifenberger J, Ichimura K, Meltzer PS, Collins V P (1994): Amplification of multiple genes from chromosomal region 12ql3-14 in human malignant gliomas: preliminary mapping of the amplicons shows preferential involvement of CDK4, SAS, M D M 2 . Cancer Res 54:4299-4303. 53. Dracopoli H C , Haines JL, Korf BR, moir DT, Morton C C , Seidman C E , Siedman JG, Smith DR, eds (1994): Cancer Genetics. In: Current Protocols in Human Genetics Vol.2, John Wiley and Sons, U.S .A. 54. Brodeur G M , Seeger RC, Schwab M , Varmus H E , Bishop J M (1984): Amplification of N-myc in untreated human neuroblastomas correlates with advanced disease stage. Science 224:1121-1124. 40 55. Dalla-Favera R, Bregni rvL Erikson J, Patterson D, GalloRC, Croce C M (1982): Human c-myc oncogene is located on the region of chromosome 8 that is translocated in Burkitt lymphoma cells. Proc natl Acad Sci U S A 79:7824. 56. Triche TJ (1993): Pathology of Pediatric Malignancies. In: Principles and Practice of Pediatric Oncology, vol 2. Pizzo PA, Poplack D G eds. J.B. Lippincott, Philadelphia. 57. Rowley J (1990): Recurring chromosomal abnormalities in leukemia and lymphoma. Sem Hematol 27:122-236. 58. Sorensen PHB, Triche TJ (1996): Gene fusions encoding chimaeric transcription factors in solid tumours. Sem Cancer Biology 7:3-14. 59. Pedeutour F, Simon M P , Minoletti F, Barcelo G, Terrier-Lacombe MJ, Combemale P, Sozzi G, Ayraud N , Turc-Carel C (1996): Translocation t(17;22)(q22;ql3) in dermatofibrosarcoma protuberans: a new tumor-associated chromosome rearrangement. Cytogenet Cell Genet 72:171-174. 60. Kleinfinger P, Labele Y, Melot T, Thomas G, Delattre O, Aureas A (1996): Localization of T E C to 9q22.3-q31 by fluorescence in situ hybridization. Annales de Genetique 39:233-235. 61. Simon M P , Pedeutour F, Sirvent N , Grosgeorge J, Minsletti F, Coindre JM, Terrier-Lacombe MJ, Mandahl N , Craven RD, Blin N , Sozzi G , Turc-Carel C, O'Brien KP, Kedra D, Fransson K, Guilbaud C, Dumanski JP (1997): Deregulation of the platelet-derived growth factor (3-chain via fusion with collagen gene COL1A1 in dermatofibrosarcomas protuberans and giant-cell fibroblastoma. Nat Genet 15:95-98. 62. Davis RJ, D'Crux C M , Lovell M A , Biegel JA, Barr F G (1994): Fusion of PAX7 to F K H R by the variant t(l;13)(p36;ql4) translocations in alveolar rhabdomyo-sarcoma. Cancer Res 54:2869-2872. 63. Israel M A (1993): Cancer Cell Biology. In: Principles and Practice of Pediatric Oncology, vol 2. Pizzo PA, Poplack D G eds. J.B. Lippincott, Philadelphia. 64. Hunter T, Pines J (1994): Cyclins and cancer II: cyclin D and C D K inhibitors come of age. Cell 79:573-582. 65. Sherr CJ (1993): Mammalian G l cyclins. Cell 73:1059-1065. 66. Serrano M , Hannon GJ, Beach D (1993): A new regulatory motif in cell-cycle control causing specific inhibitionof cycl inD/CDK4. Nature 366:704-707. 41 67. Xiong Y, Zhang H , Beach D (1993) Subunit rearrangement of the cyclin-dependent kinases is associated with cellular transformation. Gene Dev 7:1572-1583. 68. Daksis JL L u RY, Facchini L M , Martin W W , Penn LJZ (1994): Myc induces cyclin DI expression in the absence of de novo protein systhesis and links mitogen-stimulated signal transduction to the cell cycle. Oncogen 69. Levine AJ, Momand J, Finlay C A (1991): The p53 tumour suppressor gene. Nature 351:453-456. 70. Zambetti GP, Bargonetti J, Walker K, Prives C, Levine AJ (1992): Wild-type p53 mediates positive regulation of gene expression through a specific D N A sequence element. Genes Dev 6:1143-1152. 71. Cordon-Cardo C, Latres E, Drobnjak M , Oliva MR, Pollack D, Woodruff JM, Marechal V , Chen J, Brennan M F , Levine AJ (1994): Molecular abnormalities of mdm2 and p53 genes in adult soft tissue sarcomas. Cancer Res 54:794-799. 72. Momand J, Zambetti GP, Olson D C , George D, Levine AJ (1992): The M D M 2 oncogene product forms a complex with the p53 protein and inhibits p53-mediated transactivation. Cell 69:1237-1245. 73. Ladanyi M , Cha C, Lewis R, Jhanwar SC, Huvos A G , Healey J H (1993): MDM2 gene amplification in metastatic osteosarcoma. Cancer Res 53:16-18. 74. Leach FS, Tokino T, Meltzer P, Burrell M , Oliner JD, Smith S, Hi l l D E , Sidransky D, Kinzler K W , Vogelstein B (1993): p53 mutation and mdm2 amplification in human soft tissue sarcomas. Cancer Res 53:2231-2234. 75. Xiao ZX, Chen J, Levine AJ, Modjtahedi N , Xing J, Sellers WR, Livingston D M (1995): Interaction between the retinoblastoma protein and the oncoprotein M D M 2 . Nat 375:694-698. 76. Schmidt E E , Ichimura K, Reifenberger G , Collins P (1994) CDKN2 (pl6/MTSl) gene deletion or CDK4 amplification occurs in the majority of glioblastomas. Cancer Res 54:6321-6324. 77. Okuda T, Hirai H , Valentibe V A , Shurtleff SA, K i d d VJ, Lahti JM, Sherr CJ, Downing JR (1995) Molecular cloning, expression pattern and chromosomal localization of human lNK4d, an inhibitor of cyclin-D dependent kinases. Genomics 29:623-630. 78. Waga S, Hannon GJ, Beach D, Stillman B (1994): The p21 cyclin-dependent kinase inhibitor directlycontrols D N A replication via interaction with P C N A . Nature 369:574-578. 42 79. Xiong Y, Hannon GJ, Zhang H , Casso D, Kobayashi R, Beach D (1993): p21 is a universal inhibitor of cyclin kinases. Nature 366:701-704. 80. Rowley JD (1994): Cytogenetic and molecular analysis of pediatric neoplasms. Ped Pathol 14:167-176. 81. Hunger SP, Galili N , Carroll AJ, Crist W M , Link MP, Cleary M L (1991): The t(l;19)(q23;pl3) results in consistent fusion of E2A and PBX1 coding sequences in acute lyphoblastic leukemias. Blood 77:687-693. 82. Le Beau M M (1993): Detecting genetic changes in human tumour cells: have scientists "Gone Fishing?". Blood 81:1979-1983. 83. Cremer T, Lichter P, Borden J, Ward D C , Manuelidis L (1988): Detection of chromosome aberrations in metaphase and interphase tumour cells by in situ hybridization using chromosome-specific library probes. Exp Cell Res 176:199-220. 84. Giovannini M , Selleri L , Biegel JA, Scotlandi K, EManuel BS, EVans G A (1992) : Interphase cytogenetics for the detection of the t(ll;22)(q24;ql2) in small round cell tumours. J Cl in Invest 90:1911-1918. 85. Desmaze C, Zucman J, Delattre O, Thomas G, Aurias A (1992): Unicolor and bicolor in situ hyridization in the diagnosis of P N E T and related tumors. Genes Chrom Cancer 5:30-34. 86. Kallioniemi A , Kallioniemi O, Sudar D , Rutovitz D, Gray JW, Waldmah F, Pinkel D (1992): Comparative genomic hybridization for molecular cytogenetic analysis of solid tumours. Science 258:818-821. 87. Sandberg A A (1993): Chromosome changes in leukemia and cancer and their molecular limning. In: The Causes and Consequences of Chromosomal Aberrations, Kirsch IR, ed. Boca Raton: C R C Press, 141-163. 88. Barr F G , Chatten J, D'Cruz C M , Wilson A E , Nauta L E , Nycum L M , Biegel JA, Womer RB (1995): Molecular assays for chromosomal translocations in the diagnosis of pediatric soft tissue sarcomas. J A M A 273:553-557. 89. Sorensen P H , L i u FX, Delattre O, Rowland JM, Biggs C A , Thomas G , Triche TJ (1993) : Reverse transcriptase PCR amplification of EWS/FL1-1 fusion transcriptase as a diagnostic test for peripheral primitive neuroectodermal tumors of childhood. Diagn M o l Pathol 2:146-151. 43 C H A P T E R II M A T E R I A L S A N D M E T H O D S A . I N T R O D U C T I O N 1. Case Report The patient was a 15-year-old girl who presented with a 5 week history of anterior chest pain and increasing dyspnea and cough. Family history was significant in that her only brother had died 4 months earlier, at age 18, of clear cell sarcoma of the foot, originally diagnosed 12 years earlier. Maternal grandmother had breast cancer and paternal grandfather died of carcinoma of the colon. Physical examination revealed dyspnea, a right sided pleural rub, and increased second heart sounds. C T scan revealed a large mass in the right atrium, and echocardiography showed that it prolapsed through the mitral valve. There was no evidence of metastatic lesions. The tumor was successfully removed while the patient was supported on cardiopulmonary bypass. It arose from the left atrial appendage and extended into the right superior and inferior pulmonary veins. The patient was subsequently treated with chemotherapy for 12 months, consisting of monthly pulses of Vincristine, Adriamycin, and Cyclophosphamide alternating with Ifosfamide and VP-16. Three years post diagnosis, at the age of 19, the patient was diagnosed with osteosarcoma of the tibia. 2. Pathology The gross specimen consisted of an 8x4x3 cm pear-shaped mass with a short thick stalk. Histologic examination revealed a highly pleomorphic tumor with a 4 4 marked monocytic inflammatory infiltrate in the background (Figure 6). Tumor cells ranged from variably sized histiocytoid cells to highly malignant spindle cells. Mitotic activity ranging up to 5 mitotic figures per 10 high power fields was observed, and atypical mitoses were readily apparent. The tumor cells stained for muscle-specific actin and neuron-specific enolase, and focally for desmin (data not shown). Ultrastructural analysis showed a mixture of fibroblastic, myofibroblastic, histiocytic, and inflammatory cells (data not shown). Based on these findings, a diagnosis of storiform-pleomorphic M F H was made, which was confirmed in consultation with the Armed Forces Institute of Pathology, Washington, D C . 3. Cytogenetics Cytogenetic analysis revealed a hypertriploid tumor possessing multiple structural and numerical aberrations. The complete karyotype was as follows: 77-82, X, del(X)(qllql3), add(l)(p21), +del(l)(pl2), add(2)(pll), +add(2)(pl3), add(3)(qll), +add(3)(qll), der(?)t(4;?)(qll;?), der(?)t(4;?)(ql3;?), add(5)(pl5), +add(5)(P15)x2, +add(5)(pll), +add(8)(P23), +add(8)(p22), -9, add(10)(pl5), add(10)(pl3), del ( l lXpl l ) , del(ll)(q23), +del(12)(ql3), -13, +14, -15, add(15)(pll), -16, -17, -18, -18, add(19)(pl3), +add(19)(pll), +20, +20, -21, -22, +8mar, -1-2 dmin, inc [cpl5]. A representative karyotype is shown in Figure 7. Curved arrow indicates deletion of H p ; straight arrow indicates chromosomes 19 with additional material (?); arrow head indicates homogeneously staining region (HSR) on long marker chromosome; white arrowheads indicate putative derivative 19 chromosomes to be discussed in Chapter 5. Using standard cytogenetic methods, there was no evidence of chromosome 19 abnormalities in the recurring lesion. 45 F i g u r e 6. Histopathology of a pediatric M F H (BCCH-Sn). Low power photo-micrograph of primary tumor showing storiform arrangement of malignant spindle cells in an inflammatory cell background (A). High power appearance of malignant histiocytoid cells with pleomorphic vesicular nuclei and several atypical mitoses (B). 46 Figure 7. Representative G-banded karyotype of a pediatric M F H (BCCH-Sn). Cytogenetic analysis of cultured tumour revealed a hypertriploid karyotype with numerous structural abnormalities of chromosomes X, 1, 2, 3, 4, 5, 8, 10, 11, 12, 15, and 19 and marker chromosomes. Curved arrow, deletion of l i p ; straight arrow, chromosomes 19 with additional material (?); black arrowhead, HSR on a long marker chromosome; white arrowheads, putative derivative 19 chromosomes. 47 B. TISSUE C U L T U R E T E C H N I Q U E S A N D C Y T O G E N E T I C A N A L Y S I S Tumour tissue for cytogenetic and molecular studies was collected from the patient at the time of surgery. The short-term tissue cultures and cytogenetic analysis of the tumour were performed by the cytogenetics laboratory at B.C. Children's Hospital according to established protocols (1). Briefly, excised tumour tissue was minced in collagenase (200 units/ml, Sigma) and incubated for 2 hours. Washed cells were then incubated in 60 mm plastic petri dishes in RPMI 1640 medium with L-glutamine (GIBCO BRL) supplemented with 20% fetal bovine serum (FBS, Sigma), 5% antibiotic-antimycotic solution (GIBCO BRL), and Insulin-Transferrin-Sodium Selenite (ITS, 5 ng /u l , Sigma) and maintained in this medium at 37°C in a 5% CO2 incubator. Short-term cultures used for cytogenetic analysis were arrested in metaphase with Colcemid (1 n g / m l final concentration, GIBCO BRL) for 4 hours prior to harvesting in situ after ~ 7 days in culture. The cells were fixed in a 3:1 solution of methanol to acetic acid and applied to glass slides. G -banding techniques were used to stain metaphases previously fixed, dried, and treated overnight at 60 °C on the glass slides (1). Chromosome abnormalities were classified according to the International System for Human Cytogenetic Nomenclature (ISCN 1995) (2). The remaining cells were used to establish long-term cultures of the tumour (BCCH-Sn) which were maintained in a 37°C incubator in RPMI 1640 (HEPES buffered with L-glutamine) supplemented with 20% FBS, 5% antimycotic-antibiotic solution, and ITS (10 ng/ul) . Cells were harvested at various passages and were used for cytogenetic analysis (some cells stored at -20°C in 3:1 fixative solution for greater than a year) and as a source of D N A and R N A . Cells were also frozen in a 10% D M S O solution (Fisher) and stored in liquid nitrogen until being thawed at a later date. 48 Over a ten month period, great efforts were made to establish a cell line (BCCH-Sn), so that we would have a steady source of material for continued cytogenetic analysis. Methods included: LipofectAMINE (Life Technologies) transfection of pRSV-T/ t , a plasmid containing the SV40 large and small T antigens for cell immortalization (plasmid supplied by Dr. D. Mager, Terry Fox Laboratory, Vancouver, B.C.), tumour cell intramusculature injection into SCID mice, alterations of the growth medium from HEPES buffered RPMI 1640 to non-HEPES buffered RPMI 1640 (GIBCO BRL) and to F10 nutrient mixture (GIBCO BRL), alterations of the concentration of FBS and ITS supplement in the growth medium, and replacement of FBS with human serum (Sigma). A l l methods, however, were unsuccessful in establishing the pediatric M F H cell line (BCCH-Sn). This is a common problem with M F H and other solid tumours, and few M F H cell lines have been established to facilitate investigation of the characteristics of M F H (3-10). C. I S O L A T I O N O F D N A A N D R N A B C C H - S n was used as a source of R N A for Northern blot analysis and B C C H -Sn and primary tumour tissue were used as sources of D N A for Southern analysis. Several other previously established pediatric sarcoma cell lines were used as controls, including rhabdomyosarcoma (RMS) cell lines C T R (11), Birch (established at St. Jude's Hospital, Memphis, T N , Piper S, unpublished data), and Rh-18 (12); olfactory neuroblastoma (ONB) cell lines J-FEN and TC-268 (13), p P N E T cell lines TC-71 (14) and TC-32 (12); and malignant ectomesenchymoma cell lines TC-547 (15) and TC-206 (14). D N A was isolated from the above cell lines using standard methods and an Applied Biosystems D N A Extractor, Model 340A (16). Total R N A was extracted using the acid guanidinium thiocyanate phenol/ chloroform 49 method (17). D. S O U T H E R N A N D N O R T H E R N B L O T A N A L Y S I S For Southern analysis, 10 ug of genomic D N A from each cell line was digested with HindRl and electrophoresis was performed using 0.8% agarose gels. D N A from B C C H - S n , Rh-18, and CTR were also digested with BaraHI, Bglll, EcoRI, EcoRV, Hindi, Hpal, Kpnl, and PstI for additional Southern analyses. The D N A was transferred onto nylon membrane filters (Hybond) by capillary blotting and fixed to the membrane by U V crosslinking using a U V Stratalinker 1800. For Northern analysis, 25 pg of total cellular R N A from each cell line was electrophoresed in a 1.2% formaldehyde-agarose gel, and transferred to Hybond filters and fixed as described above. To inhibit RNase activity, all solutions for Northern blotting were prepared using sterile distilled water that was treated with diethylpyrocarbonate (DEPC). The probes used for Southern and Northern analysis included a 400-base pair human p l9 iNK4d c D N A fragment (donated by Dr. J. Downing, St. Jude Children's Research Hospital, Memphis, TN) , a 438-base pair human M D M 2 c D N A product (Clontech, Palo Alto, CA) , a 500-base pair CDK4 c D N A fragment, cSAS14, an 800-base pair SAS c D N A fragment, a 640-base pair CHOP c D N A fragment, and a 1.55-kilobase GLI c D N A probe (the latter four were donated by Dr. P. Meltzer, Laboratory of Cancer Genetics, National Institutes of Health, Bethesda, MD). Probes were radiolabeled with [a-32p]dCTP (50 uCi) using random primer extension (Oligo Labeling Kit, Pharmacia Biotech Inc.) followed by nick-column purification (Pharmacia Biotech Inc.). The D N A and R N A membranes were hybridized overnight with the radiolabeled probes and the membranes were washed and 50 autoradiographed at -70°C using standard protocols (16). To control for equal loading of D N A and for aneuploidy of chromosome 12, the Southern blots were stripped and reprobed with the chromosome 12 probe, D12S2 (American Type Culture Collection 57181) and the chromosome 19 probe, E2A (donated by Dr. A . Goldfarb). To control for equal loading of R N A , the Northern blots were stripped and probed with a p-Actin D N A probe. E. C H R O M O S O M E 19 PCR-BASED M I C R O S A T E L L I T E A N A L Y S I S Using eighteen primer pairs for polymorphic markers spanning the length of chromosome 19, PCR was performed on D N A extracted from the primary tumour and from D N A extracted from peripheral blood lymphocytes of the patient and the patient's mother and father. The loci of the primer pairs were D19S177, D19S199, D19S209, D19S216, D19S218, D19S220, D19S221, D19S223, D19S226, D19S228, D19S247, D19S394, D19S411, D19S424, D19S425, D19S427, EPO, and EPOR (MapPairs, Research Genetics, Huntsville, A L ) (18-22). With the exception of D19S218 (0.61) and D19S228 (0.64), primers with a maximal heterozygosity greater than 0.70 were chosen. The majority of primers used mapped to 19pl3, our area of primary interest; however, loci from other regions of chromosome 19 were investigated based on previous studies of other tumour types which assessed chromosome 19 L O H (23, 24). The PCR conditions per 15 ul volume reaction were as follows: 0.5 ul forward primer (20 uM), 0.5 ul reverse primer (20 uM), 1.5 ul 10X T a q / D N A polymerase buffer (500 m M K C L , 100 m M Tris -HCL p H 8.3, 15 m M MgC h , Perkin Elmer), 2 ul dNTP mix (0.010 m M each d A T P , dTTP, dGTP, dCTP, Pharmacia), 2 uCi [a-33P]dATP at 0.1 C i / u l , and 0.1 ul Taq D N A polymerase (5 U / ul, Perkin Elmer). 50 ng of D N A was added to each tube and the volume was adjusted to 15 ul with sterile distilled water. PCR 51 amplification was performed on a Perkin Elmer PCR 9600 Thermocycler at 96°C for 5 minutes, followed by 35 cycles consisting of 40 seconds at 94°C, 30 seconds at 55°C, and then a final elongation for 10 minutes at 72°C. After amplification, 5 \i\ of a formamide loading dye (98 ml deionized formamide, 2 ml 10 m M E D T A , 0.025 g xylene cyanol, 0.05 g bromophenol blue) was added to each reaction and heated at 95°C for 5 minutes, followed by quenching on ice. Samples were loaded (7 |LX1 per lane) and electrophoretically separated on a 6% polyacrylamide sequencing gel (37.5 grams urea, 11.25 ml 40% Aerylamide / Bis Solution 19:1 (BIO-RAD), 7.5 ml 10X Tris-borate/ E D T A , 28.5 ml distilled H 2 0 , 500 | i l 10% Ammonium Persulfate (BIO-RAD), and 50 ul T E M E D (BIO-RAD), prepared using a Model S2 Sequencing Gel Electrophoresis System (GIBCO BRL). The gel was run at 80 watts either for 2.5 hours (150-200 base pair products) or for 3.5 hours (> 200 base pair products). The gel was dried on a gel drier (Labconco) for 1.5 hours and exposed to XAR-5 film (Kodak) at room temperature for 16 - 48 hours. F. SLIDE P R E P A R A T I O N FOR F L U O R E S C E N C E IN SITU H Y B R I D I Z A T I O N (FISH) Cell suspensions of normal cultured lymphocytes, fibroblasts, and B C C H - S n tumour cells (short and long-term cultures) were processed according to standard cytogenetic procedures (1) and stored at -20°C in methanol/acetic acid fixative (3:1) until use. Metaphase chromosome spreads and interphase nuclei were prepared on glass microscope slides (1). Best results were achieved when the slides were used within 2 days of preparation. Although reasonable hybridization was achieved when FISH was performed on slides that were greater than one year old (stored at -20°C) with whole chromosome painting probes and a region-specific probe, 52 successful hybridiation with a second region-specific probe was more difficult on this older material. Briefly, the cells were washed three times in methanol/ acetic acid fixative (3:1), resuspended in fresh fixative, dropped onto glass slides, and air dried for a minimum of 4 hours. The chromosomes were aged (to make them less sensitive to over-denaturation) by pretreating the slide in 2X SSC, p H 7.0 at 37°C for 15 minutes. This was followed by dehydration in 70%, 80%, 90%, and 100% ethanol at room temperature for 2 minutes each. After air drying, slides were denatured in 70% deionized formamide/2X SSC, p H 7.0 at 70°C for 2-5 minutes, followed immediately by dehydration in a series of 2 minute ice cold ethanol washes (70%, 80%, 90%, and 100%). After air drying, the slide was ready for probe application. Without an established cell line (BCCH-Sn), there was a limited supply of fixed cell suspensions containing metaphase chromosomes for FISH. This prompted us to make touch preparations using primary tumour tissue and normal fetal lung tissue as a control (both frozen at -70°C) for interphase analysis. First, the tissue was cut with a scalpel (to get a fresh exposed surface) and was lightly touched onto silanized glass slides (FISHER-Super frost plus). After air drying, the slides were fixed in 100% methanol for 5 minutes and 3:1 methanol/acetic acid for 5 minutes. Slides were dehydrated in a room temperature ethanol series and then pretreated with 0.03% trypsin (Difco) in 40 ml phophate buffered saline (PBS) for 5 - 8 seconds (tumour tissue) or for 3 - 5 seconds (normal tissue). Two 5 minute PBS washes followed the trypsin treatment. The slides were then fixed in 3.7% formaldehyde for 10 minutes and washed three times for 5 minutes in PBS. This was followed by incubation in 50% formamide/2X SSC (pH 7.0) for 15 minutes at room temperature. Denaturation and dehydration proceeded from this point as described above. Additionally, due to the limited number of B C C H - S n cytogenetic preparations available for FISH, slides with metaphases previously used for FISH were 53 rehybridized with different D N A probes (25). Prior to the second denaturation step, coverslips were soaked off the slides in IX phosphate buffered detergent (PBD). Unfortunately, rehybridization of the slides previously hybridized with C O A T A S O M E 19 were unsuccessful upon the second FISH attempt. This was mostly likely because the first slide denaturation step (5 minutes), in addition to the second denaturation (2 minutes), resulted in over-denaturation of the chromosomes and no hybridization of the probe. To increase the ability to repeatedly analyse the same metaphase chromosomes, slide denaturation was decreased to 2-3 minutes for all subsequent FISH experiments. G. W H O L E C H R O M O S O M E FISH: C H R O M O S O M E 19 P A I N T I N G PROBE Whole chromosome FISH was performed using the digoxigenin-labeled C O A T A S O M E 19 total chromosome probe (Oncor, Gaithersburg, MD) . Slides were prepared as previously described and C O A T A S O M E 19 was prepared as follows. After pre-warming at 37°C for 5 minutes, 10 ul of probe per slide was aliquoted, denatured at 70°C for 10 minutes, and then allowed to preanneal at 37°C for 45-60 minutes. 10 ul of C O A T A S O M E 19 was applied to the denatured slide, followed by a 22 x 22 m m glass coverslip. After sealing with rubber cement, the slides were incubated overnight at 37°C in a humid chamber. The post-hybridization wash to remove the background associated with nonspecific hybridization consisted of 15 minutes in 50% formamide/2X SSC at 72°C without agitation, followed by 2 minutes in IX PBD at room temperature. Since the probes we used were not directly labeled with a fluorochrome, in order to visualize the specifically hybridized probe with fluorescence microscopy, it was necessary to apply a fluorescent-labeled antibody to the slide. Specifically, 60 ul of 54 anti-digoxigenin-FTTC, Fab fragments (20 ug/ ml, Boehringer Mannheim) and a plastic coverslip were applied to the slides and incubated for 15-20 minutes at 37°C. The slides were rinsed 3 times in IX PBD at room temperature for 2 minutes each with intermittent agitation. The cells were counterstained with 20 ul of antifade solution (10 m g / m l p-phenylenediamine dihydrochloride in 90% glycerol p H 8.0) containing 0.3 ug/ml propidium iodide (PI), and a 22 x 50 mm coverslip was applied to the slides. Hybridization signals were visualized using 400x, 600x, and lOOOx objectives on a Nikon Optiphot episcopic-flourescence microscope (attachment EF-D) with an Omega dual band filter set (Omega Optical Inc.). Photographs were taken using a Nikon FX-35A camera and Nikon HFX-II adaptor. H . D U A L - C O L O U R E D FISH: C H R O M O S O M E 19ql3.1-SPECIFIC D N A PROBE A N D 19p COSMID PROBES Dual-coloured FISH was performed using a digoxigenin-labeled chromosome 19ql3.1-specific D N A probe (Oncor) and 3 biotin-labeled 19p cosmid probes (19474, 18382, and 16575) (13) spaced at ~ 5 megabase intervals from the telomere of human chromosome 19p (generously supplied by Lawrence Livermore National Laboratory, Livermore, CA) . The 19p cosmid probes were labeled with biotin-14-dATP (Gibco-BRL) by nick translation (Nick Translation Kit, Gibco-BRL) according to standard protocols. For each 19p cosmid, a 10 ul hybridization mixture was prepared per slide containing 1.2 ul biotin-labeled 19p cosmid (~ 20 ng), 8 ul hybridization mix (50% formamide, 2X SSC, 10% dextran sulfate), and 0.96 pi sonicated human placental D N A (10 ug/[il). The three cosmid probe mixtures were combined as a "19p cocktail probe" in order to obtain a larger signal size that was more easily detectable. The resulting 30 ul mixture was denatured for 5 minutes at 70°C and incubated at 37°C 55 for 30 minutes to preanneal. The chromosome 19ql3.1-specific D N A probe was prewarmed at 37°C for 5 minutes and 10 ul was applied to the slide simultaneously with the 19p hybridization mixture. Coverslips (22 x 40 mm) were applied to slides, sealed with rubber cement, and incubated overnight at 37°C in a humid chamber. The optimal post-hybridization stringency for the two different probe cocktails was determined to be 15 minutes in 50% formamide/2X SSC at 43°C with intermittent agitation, followed by 8 minutes in 2X SSC at 37°C and 2 minutes in IX PBD at room temperature. The biotin-labeled 19p cosmid probes were detected with FITC (green) and the digoxigenin-labeled 19ql3.1 probe was detected with rhodamine (red). Two-coloured detection was achieved with the following series of antibody applications. 60 ul FITC-avidin D N (5 p.g/ml, Vector Laboratories) and 60 pl mouse-monoclonal anti-digoxigenin (8.2 pg/ml, Sigma) per slide were premixed and chilled on ice for 5 minutes. This antibody mixture was applied to the slides and incubated at 37°C in a humid chamber for 30 minutes. Slides were washed 3 times in IX PBD at room temperature for 2 minutes per wash with intermittent agitation. The next four antibody layers were applied individually in the following order: 60 ul anti-mouse-Ig-digoxigenin F(ab')2 fragments (0.8 pg /ml , Boehringer Mannheim), 60 ul anti-digoxigenin-rhodamine, Fab fragments (2 ug/ml, Boehringer Mannheim), 60 ul biotinylated anti-avidin D (5 ug/ml, Vector), and 60 ul FITC-avidin D N (5 ug/ml). After each antibody application, slides were incubated at 37°C in a humid chamber for 15-30 minutes, followed by 3 room temperature washes (2 minutes) in IX PBD. Amplification of the FITC signal was necessary; therefore, the biotinylated anti-avidin D and FITC-avidin D N antibody layers were repeated. Cells were counterstained with 12 pl antifade solution containing 0.01 ug/ul 4',6-diamidino-2-phenylindole dihydrochloride hydrate (DAPI), and a 22 x 40 m m glass coverslip was applied. Using a 40x objective and a lOOx oil immersion objective, 56 slides were viewed through a Zeiss Axioplan Universal microscope equipped with epifluorescence. A triple band pass filter for D A P I / F I T C / Texas Red (Chroma Technology Corp.) allowed for dual-colour detection. The microscope was also equipped with a C O H U High Performance camera run by PSI Scientific Systems software (League City, TX). Images were converted to TIFF files for printing. I. D U A L - C O L O U R E D FISH: C H R O M O S O M E 4 A N D 19 P A I N T I N G PROBES Dual-coloured FISH was performed on BCCH-Sn slides (made kindly by Chris Salski, B.C. Cancer Agency), using the digoxigenin-labeled C O A T A S O M E 19 total chromosome probe (Oncor) and a biotin-labeled chromosome 4-specific paint (Cambio, Cambridge, M A ) . Note that the slides made for this FISH procedure were composed of the last B C C H - S n cell suspensions. Both probes were prepared as previously described for C O A T A S O M E 19, and 10 pi of C O A T A S O M E 19 was applied to each slide simultaneously with 15 ul of the chromosome 4-specific paint. A 22 x 40 m m coverslip was applied, sealed with rubber cement, and incubated at 37°C overnight in a humid chamber. After removing the coverslips using 2X SSC at 37°C, the post-hybridization washes consisted of two 5 minute incubations in 50% formamide/2X SSC at 43°C with intermittent agitation, followed by two 5 minute incubations in 0.1X SSC at 43°C, and 2 minutes in IX PBD at room temperature. The biotin-labeled C O A T A S O M E 19 was detected with FITC (green) and the digoxigenin-labeled chromosome 4 painting probe was detected with rhodamine (red). Dual-coloured detection was achieved as described previously, except that the PBD washes between the antibody layers were performed at 37°C instead of at room temperature and no amplification of the FITC signal was required with the chromosome 4 paint. 57 J. C H R O M O S O M E G - B A N D I N G A N D DAPI /PI STAINING Attempts were made to G-band the C O A T A S O M E 19 hybridized chromosomes using a Wright-Giemsa/ Borate buffer method (14) with several modifications, including varying reagent concentrations and incubation times. Despite these efforts, banding was unsuccessful using this method. DAPI /PI staining was also attempted, which provided a minor improvement in chromosome band resolution. The optimal concentrations were determined to be 2 ul DAPI (200 ng/ ul) and 2 ul PI (20 ng/ul) in 200 ul antifade, of which 10 ul were applied to the previously hybridized slides. K. R E F E R E N C E S 1. Mandahl N (1991): Methods in solid tumour cytogenetics. In: Human Cytogenetics, A Practical Approach. Vol. II Malignancy and acquired abnormalities, 2nd Ed . Rooney D E , Czepulkowski R H , eds. Oxford University Press, New York, pp. 155-187. 2. ISCN (1995): A n International System for Human Cytogenetic Nomenclature. Mitelman F, ed. Karger S, Basel. 3. Iwasaki H , Kikuti M , Taki M , Enjoji M (1982): Benign and malignant fibrous histiocytoma of the soft tissues. Cancer 50:520-530. 4. Shirasuna K, Sugiyama M , Miyazaki T (1985): Establishment and characteriz-ation of neoplastic cells from a malignant fibrous histiocytoma. Cancer 55: 2521-2532. 5. Roholl PJM, Kleyne J, Von Blokland M , Spies PL (1986): Characterization of two cell lines, derived from two malignant fibrous histiocytomas. J Pathol 150:103-112. 6. Kanzaki T, Kitajima S, Suzumori K (1991): Biological behavior of clone cells of human malignant fibrous histiocytoma in vivo and in vitro. Cancer Res 51:2133-2137. 58 7. Roholl PJM, Prinsen I, Rademakers L P H M , Hsu SM, Van Unnik A M (1991): Two cell lines with epithelial cell-like characteristics established from malignant fibrous histiocytomas. Cancer 68:1963-1972. 8. Yokota T, Ishizaki R, Ohishi H , Ohami H (1995): Establishment and Characterization of a fibrous histiocytoma cell line (NMSG 10) derived from a malignant fibrous histiocytoma. Path Res Pract 191:1231-1238. 9. Yonemoto T, Takenouchi T, Tokita H , Tatezaki S, Mucaida N , Mikata A , Moriya H (1995): Establishment and characterization of a human malignant fibrous histiocytoma cell line. Cl in Orthopaed Rel Res 320:159-167. 10. Mugneret F, Lizard S, Aurias A , Turc-Carel C (1988): Chromosomes in Ewing's sarcoma. II. Nonrandom Additional Changes, trisomy 8 and der(16)t(l:16). Cancer Genet Cytogenet 32:239-245. 11. Whang-Peng J, Knutsen T, Theil K, Horowitz M E , Triche TJ (1992): Cytogenetic studies in subgroups of rhabdomyosarcoma. Genes Chromosom Cancer 5:299-310. 12. Houghton JA, Houghton PJ, Webber BL, (1982): Growth and characterization of childhood rhabdomyosarcomas as xenografts. J Natl Cancer Inst 68:437-443. 13. Sorensen PHB, W u JK, Berean K W , Donn W, Frierson H F , Reynolds CP, Lopez-Terrada D , Triche TJ (1996): Olfactory neuroblastoma is a peripheral primitive neuroectodermal tumor related to Ewing sarcoma. Proc Natl Acad Sci 9:1038-1043. 14. Whang-Peng J, Knutsen T, Thiel K, Horowitz M E , Triche TJ (1986): Cytogenetic characterization of selected small round cell tumors of childhood. Cancer Genet Cytogenet 21:185-208. 15. Sorensen PHB, Shimada H , Liu XF, L i m JF, Thomas G, Triche TJ (1995): Biphenotypic sarcomas with myogenic and neural differentiation express the Ewing's sarcoma EWS/FLU fusion gene. Cancer Res 55:1385-1392. 16. Sambrook J, Frilsch EF, Maniatis T (1989): Molecular cloning: A Laboratory Manual. Cold Spring Harbour, New York. 17. Chomczynski P, Sacchi N (1987): Single-step method of R N A isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Analytical Biochem 162:156-159. 59 18. Amfo K, Neyns B, Teugels E , Lissens W, Bourgain C, De Sutter P, Vandamme B, Vamos E , De Greve J (1995): Frequent deletion of chromosome 19 and a rare rearrangement of 19pl3.3 involving the insulin receptor gene in human ovarian cancer. Oncogene 11:351-358. 19. von Deimling A , Bender B, Jahnke R, Waha A , Kraus J, Albrecht S, Wellenreuther R, Fabbender F, Nagel J, Menon A G , Louis D N , Lenartz D , Schramm J, Wiestler O D (1994): Loci associated with progression in astrocytomas: a candidate on chromosome 19q. Cancer Res 54:1397-1401. > 20. Mandahl N , He im S, Willen H , Rydholm A , Eneroth M , Nilbert M , Kreicbergs A , Mitelman F (1989): Characteristic karyotypic anomalies identify subtypes of malignant fibrous hisitiocytoma. Genes Chromosom Cancer 1:9-14. 21. Choong P F M , Mandahl N , Mertens F, Willen H , Alvegard T, Kreicbergs A , Mitelman F, Rydholm A (1996): 19p+ marker chromosome correlates with relapse in malignant fibrous histiocytoma. Genes Chromosom Cancer 16: 88-93. 22. Guan K L , Jenkins CW, L i Y, O'Keefe C L , Noh S, W u X, Zariwala M , Matera A G , Xiong Y (1996): Isolation and characterization of p l9 lNK4d a pl6-related inhibitor specific to CDK6 and CDK4. Mol Biol Cell 7:57 -70. 23. Wang MR, Perissel B, Malet P (1995): Rehybridization on metaphases studied previously by FISH. Cancer Genet Cytogenet 85:58-60. 24. Brandiff BF, Gordon L A , Fertitta, Olsen AS, Christensen M , Ashworth L K , Nelson D O , Carrano A V , Mohrenweiser H W (1994): Human Chromosome 19p: A Fluorescence in Situ Hybridization Map with Genomic Distance Estimates for 79 Intervals Spanning 20 Mb. Genomics 23:582-592. 25. Cannizzaro L A , Emanual BS (1984): A n improved method for G-banding chromosomes after in situ hybridization. Cytogenet Cell Genet 38:308-309. 60 C H A P T E R III A M P L I F I C A T I O N OF M U L T I P L E G E N E S F R O M C H R O M O S O M A L R E G I O N 12ql3-14 IN PEDIATRIC M F H A . I N T R O D U C T I O N Homogeneously staining regions (HSRs) and double minutes (dmins), cytogenetic indices of gene amplification, were detected in the primary tumour in the clinical workup of the case (see case history in Materials and Methods). Gene amplification, particularly from the 12ql3-14 chromosomal region, has been demonstrated previously in adult human sarcomas, including M F H (1-3). In addition to amplification, rearrangements of this region have been reported to occur frequently in soft tissue tumours (3-5), strongly suggesting that this chromosomal region contains gene(s) that might be important in the evolution of these tumours (6). This information prompted an investigation to determine whether there was amplification and overexpression of genes from 12ql3-14 in the presented pediatric M F H . The five markers from the 12ql3-14 region that were studied were all genes known to be rearranged or amplified in different tumour types, including MDM2, CDK4, SAS, CHOP, and GLI (1-2, 7-9). The chromosome 12 probe D12S2, located on the same chromosome as the tested genes but mapping to a different region, was used as a single gene copy control. This enabled D N A gene amplification to be distinguised from chromosomal aneuploidy which is commonly seen in tumour cells (10). Southern and Northern blot analysis were the principle methods used to study amplification and increased expression, respectively, of these five 12ql3-14 region genes. Additionally, the tumour genomic material was analyzed for unbalanced genetic alterations using comparative genomic hybridization (CGH). 61 This new technique has shown great utility for the detection and chromosome localization of copy number changes, such as gene amplification, in human solid tumours (13). In the presented pediatric M F H , gene amplification studies were initiated by looking for increased copy numbers of the MDM2 gene through Southern analysis. This gene was chosen first since it is reported that 36% of adult human sarcomas (7) and 20 - 32% of adult M F H (7, 11-12) have MDM2 amplification. Following this analysis, additional 12ql3-14 genes were analyzed. B. R E S U L T S 1. MDM2 Gene Amplification and Overexpression Genomic D N A from B C C H - S n and control pediatric sarcoma cell lines were digested with Hindlll and subjected to Southern blot analysis using a human M D M 2 c D N A probe (Figure 8A). MDM2 amplification was detected in B C C H - S n (approximately 30 - 40 fold) and in Rh-18 (>50 fold), an alveolar rhabdomyosarcoma (RMS) cell line which had previously exhibited MDM2 amplification (unpublished data, D . Lopez-Terrada and P. Sorensen). Amplification was not detected in a variety of pediatric sarcoma cell lines, including embryonal RMS, olfactory neuroblastoma (ONB), pPNETs , nor in malignant ectomesenchymoma cell lines (data not shown). The blot was stripped and rehybridized with the chromosome 12 probe, D12S2, which demonstrated that the observed amplification was not a result of chromosome 12 aneuploidy or unequal D N A loading. To confirm increased m R N A expression of MDM2 in B C C H - S n , total R N A from B C C H - S n and 5 pediatric sarcoma cell lines was subjected to Northern blot analysis using the same human M D M 2 c D N A probe (Figure 9). B C C H - S n and Rh-18 both exhibited strong overexpression of the 5.5 kb M D M 2 transcript, while no overexpression was detected in the cell lines which lacked gene amplification by 62 Southern blot analysis. A p-actin probe was used to control for equal loading of R N A samples. Direct sequencing of the TP53 gene (exons 5-8) revealed normal TP53 sequences in the primary tumour with MDM2 amplification (performed in the laboratory of Dr. D . Horsman , British Columbia Cancer Agency). These results are consistent with those of previous studies which suggest that MDM2 amplification occurs only in tumours expressing wild type p53 (7,11). 2. CDK4. SAS. CHOP, and GLI Amplification and Overexpression Having confirmed amplification and overexpression of the M D M 2 gene from 12ql3-14 in the pediatric M F H , the amplicon was further characterized by studying several different loci mapped to this region, including CDK4, SAS, CHOP, and GLI. The initial Southern blot described above and a second blot containing the same cell lines were then screened for amplification of these additional genes from the 12ql3-14 region (Figure 8A,B). Amplification of all four genes was seen in B C C H - S n and in Rh-18, but not in the other primitive sarcomas tested. The degrees of gene amplification in B C C H - S n were variable, with the greatest amplification seen for CDK4 (estimated to be 30-40 fold, approximately the same as M D M 2 ) . SAS gene amplification was estimated to be 20-30 fold and CHOP and GLI were estimated to be amplified 5-10 fold. The second blot was also stripped and reprobed with D12S2 to control for chromosome 12 aneuploidy and equal D N A loading. HmdIII-digested BCCH-Sn D N A probed with CDK4 revealed three signals (Figure 8B, lane 5), whereas the seven other tumour samples had only one. To investigate this further, additional Southern blots were made using genomic D N A from BCCH-Sn , Rh-18 and CTR digested again with Hindlll, plus eight other restriction enzymes (BamHI, Bglll, EcoRI, EcoRV, Hindi, Hpal, Kpnl, and Pstl). 63 F i g u r e 8. S o u t h e r n b l o t a n a l y s e s o f genes f r o m t h e 1 2 q l 3 - 1 4 a m p l i c o n i n p e d i a t r i c s a r c o m a s . ( A ) G e n o m i c D N A f r o m 8 p e d i a t r i c s a r c o m a c e l l l i n e s w a s d i g e s t e d w i t h Hindlll a n d h y b r i d i z e d w i t h t h e f o l l o w i n g c D N A p r o b e s : MDM2, CHOP, GLI, a n d D12S2, a c h r o m o s o m e 1 2 p p r o b e t o c o n t r o l f o r t u m o r c e l l a n e u p l o i d y . (B) G e n o m i c D N A f r o m 8 p e d i a t r i c s a r c o m a ce l l l i n e s w a s d i g e s t e d w i t h Hindlll a n d h y b r i d i z e d w i t h t h e f o l l o w i n g c D N A p r o b e s : CDK4, SAS, a n d D12S2. L a n e s 1-2, e m b r y o n a l R M S c e l l l i n e s C T R a n d B i r c h ; 3-4, O N B ce l l l i n e s J - F E N a n d T C - 2 6 8 ; 5, p e d i a t r i c M F H c e l l l i n e B C C H - S n ; 6, a l v e o l a r R M S ce l l l i n e R h - 1 8 ; 7-8, p P N E T c e l l l i n e s T C - 3 2 a n d T C - 7 1 . (see p a g e 64) 65 Figure 9. Northern analyses of genes from the 12ql3-14 amplicon in pediatric sarcomas. (A) Total R N A from 5 pediatric sarcoma cell lines was hybridized with the following c D N A probes: MDM2, CDK4, GLI, and (3-Actin, to confirm equal loading of samples. (B) Total R N A from 5 pediatric sarcoma cell lines was hybridized with the following c D N A probes: SAS, CHOP, and (3-Actin. Lanes 1-2, embryonal RMS cell lines CTR and Birch; 3, pediatric M F H cell line B C C H - S n ; 4, alveolar RMS cell line Rh-18; 5, p P N E T cell line TC-32. (see page 66) 66 67 Although the extra bands were confirmed a second time with Hindlll, additional bands were not observed with the other enzymes (negative data not shown). The additional bands seen in Figure 8B, are therefore not likely due to CDK4 gene rearrangement, and may represent heterogeneity in amplicons containing the CDK4 gene or a D N A polymorphism. To investigate increased expression of the additional 12ql3-14 genes in B C C H -Sn, Northern analysis was performed using the same human c D N A probes, CDK4, SAS, CHOP, and GLI. (Figure 9). BCCH-Sn and Rh-18 both exhibited strong overexpression of a 1.5 kb CDK4 transcript and the 1.8 kb SAS transcript, while no overexpression was detected in the cell lines which did not show gene amplification by Southern blot analysis. Overexpression of a 2.7 kb C H O P transcript was also detected in B C C H - S n , but not in Rh-18 nor in the other pediatric sarcomas. Although Southern analysis revealed GLI gene amplification in B C C H - S n and Rh-18, Northern analysis revealed no significant increase in the 4.0 kb GL7 m R N A transcript. Table 4 summarizes the results from the 12ql3-14 region gene amplification and overexpression studies that were performed on the presented pediatric M F H and Figure 10 depicts the 12ql3-14 amplicon. 3. Comparative Genomic Hybridization (CGH) Comparative genomic hybridization (CGH) was performed in the laboratory of Dr. Jeremy Squire (Hospital for Sick Children, Toronto, Ontario) according to the methods described by Kallioniemi et al., 1992 (13), using D N A from the primary M F H tumour tissue. Amplification of the 12q region on both chromosome 12 copies was apparent in the C G H metaphase (data not shown), and this 12ql3-14 amplification is well represented in the C G H profiles (Figure 11). Gains and high-level amplification of 12q sequences are frequently found by C G H analysis in 68 Table 4 . 12ql3-14 gene amplification and expression in BCCH-Sn. Gene Amplification Overexpression M D M 2 + + C D K 4 + + SAS + + C H O P + + GLI + — 69 Chromosome 12ql3-14 Telomere Centromere MDM2 CDK4 SAS CHOP GLI q!3-14 q!3-14 q!3-14 ql3.3 ql3 > 30 copies 20-30 5-10 Figure 10. The 12ql3-14 amplicon in a BCCH-Sn, including the estimated gene dosages (amplification levels by comparison with cell lines containing single gene copy levels). The tested loci from 12ql3-14 are ordered according to their chromosomal location from the most telomeric locus ( M D M 2 ) to the most centromeric locus (GLI). 70 Figure 11. B C C H - S n comparative genomic hybridization (CGH) profile analysis demonstrating amplification of the 12ql3-14 region (data provided by Dr. J. Squire). 71 various sarcomas (14-15). Therefore, C G H confirms our Southern analysis data on 12ql3-14 amplification on the pediatric M F H case. C. DISCUSSION Gene amplification leading to the overexpression of specific proto-oncogenes is an important mechanism in the development and progression of human solid tumours (1, 2). Amplification and overexpression of the 12ql3-14 region proto-oncogenes, MDM2, CDK4, SAS, CHOP, and GLI, have previously been reported in various human sarcomas, including M F H (1, 2, 7-9), and the frequency of involvement strongly suggests that their overexpression might contribute to tumour evolution. Presently, it is unclear which of these amplified genes, or others from this region, is required for oncogenesis. In order to define the target(s) of the amplification event, one must first identify the genes that are included in the 12ql3-14 amplicon and understand the function of the resulting gene products in non-malignant cells. With this base knowledge, one can then apply information acquired through amplification studies, such as the levels of amplified genes and the patterns of co-amplified genes to gain further insight into their role in tumour development and progression. However, elucidation of amplified genes is not a trivial endeavour. One of the greatest problems is that amplicons in human tumour cells can span greater than a megabase in length and the stretch of D N A may contain several expressed genes (16). Consequently, it is of great importance to determine which of the amplified genes functionally contribute to oncogenesis and which are only passenger genes of the amplicon. Alternatively, there are yet unidentified amplified target genes in the amplicon that are the true implicators in tumour progression. A second factor to contend with is the variability that is seen between amplicons: amplification levels 72 from case to case and from gene to gene can be different, discontinuous amplicons have been detected, and the size of amplicons can vary between different cases (3,9,17). This inconsistency contributes to the difficulty in trying to establish co-amplification patterns, and as a result, to date there does not appear to be a consistent pattern of genes included in the 12ql3-14 amplicon amongst the various sarcomas (3). Previous studies report MDM2, CDK4, and SAS to be amplified more often than other 12ql3-14 region genes, and that these three genes are frequently co-amplified (3,7,9,18-21). Considering the physical map of the 12ql3-14 which links MDM2, CDK4, and SAS, it is not surprising that they are often co-amplified. Similarly, the present case revealed co-amplification of MDM2, CDK4, and SAS. Based on the current 12ql3-14 amplification data in the literature, it is believed that MDM2, CDK4, and/or SAS are the most probable target genes for many sarcomas with amplification of this region (22). Therefore, the putative roles of MDM2, CDK4, and SAS, three proto-oncogenes from 12ql3-14 that are considered to be "the genes of potential pathogenetic importance for bone and soft tissue tumour development in humans" (17), will now be further explored. MDM2, reported to be amplified in over a third of human sarcomas (7), is a likely proto-oncogene selected for by the 12ql3-14 amplification. It is hypothesized that the tumourigenic activity of high level MDM2 expression is based on functional inactivation of p53, resulting in deregulated p53 growth control (7). M D M 2 has also been shown to interact functionally with the retinoblastoma protein (pRB) and inhibit pRB growth regulatory function (23). There is considerable evidence supporting the tumourigenic significance of MDM2 in the 12ql3-14 amplicon. For example, MDM2 amplification levels are reported to be significantly higher than the other genes in the amplicon (3) and it is often the only amplified 73 gene from the 12ql3-14 region (3,17). Nilbert and co-workers (1995) characterized the 12ql3-14 amplicon in 11 M F H that had previously been shown to have MDM2 amplification (17). In five of these tumour samples MDM2 was the only amplified gene of the 14 markers screened for in the 12ql3-14 region (Figure 12). This might argue that either MDM2 or an unknown gene in its proximity (most likely between SAS and MDM2) is the target of the 12ql3-14 amplification in M F H (17). There are several reports of sarcomas with 12ql3-14 amplicons which do not include MDM2 (2,3,13,20,24) suggesting that other genes in this region may be the relevant targets amongst various histopathologic types of soft tissue tumours. There is considerable evidence suggesting that the amplification and overexpression of CDK4 may promote neoplastic cell growth (3). In eukaryotic cells, CDK4 and the cyclin protein, cyclin DI , play a critical role in cell cycle regulation (3). It has been proposed that high levels of the cyclin D I / CDK4 complex phosphorylates and functionally inactivates pRB, thus promoting transit from GI into S phase of the cell cycle (25). Amplification of the CDK4 gene and alterations in other c y c l i n / C D K family members have been reported to play critical roles in the development of some tumours (20,26). Schmidt and coworkers (1994) reported that at least 85% of human glioblastomas and 50% of anaplastic astrocytomas had an aberration in either the CDK4 gene or in its inhibitor, CDKN2 (pl6/MTSl) (20). SAS (sarcoma amplified sequence), a new member of the transmembrane 4 superfamily (TM4SF) proteins (27), is also a potential target gene in the 12ql3-14 amplicon. Overexpression of SAS has been reported in several tumour cell lines and it is speculated that overexpression of this gene may alter cell growth (27). Although the function of the TM4SF proteins is not known, current data suggests a possible role in signal transduction and growth control (27). SAS gene amplification and overexpression was originally identified in M F H (1,27). Since M F H is the most 74 1 2 3 4 5 6 7 8 9 10 11 GLI - - - - - - - - - -C H O P - - - - - - - - - - -SAS - - + + - - + - - + + C D K 4 - - + + - - + - - + + M D M 2 + + + + + + + + + + + RAP1B - - - - + - + - - + + D12S8 - - + - + - + - - ~ + PTPP - - - - + D12S7 - - - - - - - - - - -Figure 12. Amplification units in 11 MFH previously shown to have MDM2 amplification (17). + represents amplified markers; — represents non-amplified markers. 75 common sarcoma of adults and serves as a general prototype of adult high grade sarcomas (28), an understanding of the role of SAS overexpression may provide valuable information regarding oncogenic mechanisms in many types of sarcomas. A study by Smith and coworkers reported SAS amplification to be the most frequently observed change in M F H at the genetic level; they studied 22 M F H , 7 (32%) of which exhibited SAS amplification (1). Although another group comments that the frequent amplification of SAS in M F H could be due to coamplification with the more relevant target gene, MDM2 (22). Before a link can be established between SAS amplification and malignant transformation, the function of SAS must be determined. As previously mentioned, it is believed that CHOP and GLI are less likely to drive the 12ql3-14 amplicon than are MDM2, CDK4, and SAS. CHOP is a D N A damage inducible gene that encodes a protein homologous to the G / E B P family of transcription factors (29). It has been shown to be involved in the oncogenic F U S / C H O P gene fusion resulting from the t(12;16) translocation in myxoid liposarcomas (MLS) (24), and CHOP is reported to be amplified in some sarcomas, although less frequently than MDM2 and SAS (9,24). The GLI gene, which was originally identified in human gliomas, encodes a potential transcriptional regulatory protein with five zinc finger D N A binding motifs (30). Although amplified in other sarcomas (2,8), it is reported to lie outside the usual amplification domain in M F H , and it has been suggested that the amplification boundary in M F H lies between SAS and GLI (31). There have been no reports in the literature of M F H with either CHOP or GLI amplification: Forus and coworkers tested 20 M F H with 12ql3-14 amplification and found no CHOP amplification (24); Nilbert and coworkers who tested 11 M F H with M D M 2 amplification found no CHOP or GLI amplification (17). Although both CHOP and GLI amplification were detected in the 76 present M F H , their amplification levels were considerably lower than the other genes, and it is likely that they were co-amplified due to their proximity to the more relevant target gene(s), MDM2, CDK4, and SAS (Figure 7). CHOP is located 250 kilobases towards the centromere from SAS (32) and GLI is 55 kilobases towards the centromere from CHOP (19). Considering the variability of amplicons and the close proximity of these genes, it is not unreasonable to observe an extended amplicon. Although GLI was included in this extended amplicon, amplification did not correspond with increased expression of m R N A , as seen with the other four amplified genes. GLI may not have been overexpressed despite the elevated gene copy number if the full genomic sequence was not included in the amplicon. There is also the possibility that there is a small deletion or rearrangement in the gene in the amplicon (undetected through Southern analysis) that is preventing GLI transcription. Regardless, the key is that there is not an abnormal level of m R N A being produced to potentially alter normal cell regulation. Therefore, it is reasonable to conclude that GLI is a passenger gene in this particular amplicon and it is not pathologically significant in this M F H , as reported in other studies (17,31). As shown above, elevated gene copy numbers do not necessarily correlate with elevated m R N A levels. Even though gene amplification would be expected to be selected for through increased expression of the corresponding protein(s), this should be confirmed by studying expression levels. So, in reports which study gene amplification but do not analyze expression levels, it is unclear whether the observed gene amplifications are involved in oncogenic growth. Another consideration is that the absence of gene amplification does not preclude the possibility of gene overexpression, and therefore, even unamplified genes should be studied for expression levels. As a result of limited gene expression studies, the significance of particular genes might be underestimated. 7 7 By characterizing genes in the 12ql3-14 amplicon in human sarcomas, we hope to improve our understanding of the oncogenic processes of these tumours, which may ultimately be of diagnostic and prognostic importance. As discussed, the variability amongst 12ql3-14 amplicons has created difficulties in establishing characteristic patterns of amplified genes that could be used as a diagnostic aid amongst the different tumour types. Many more experiments wil l need to be performed in order to determine more consistent and specific amplification patterns that can be used for diagnostic purposes. Although not yet fully defined, a possible relationship between SAS amplification and clinical behaviour in soft tissue tumours has been suggested: tumours with SAS amplification are more likely to be large and deeply situated, i.e. more clinically aggressive (1). Currently, in clinical oncology, amplification of specific genes in many tumours are significant as prognostic markers. In neuroblastoma, for example, it has been clearly demonstrated that amplification of the MYCN gene correlates with advanced stage disease (32). Additional clinical studies are required, however, to ascertain a link between SAS amplification, stage of disease, and prognosis. Amplicon size, as a time dependent phenomenon, may also provide useful information regarding progression and aggressiveness. In other words, the longer and/or faster a tumour has been growing, the more opportunity there has been for genetic alterations to have occurred, possibly giving rise to a larger amplicon. Therefore, determining the extent of the amplicon, even if the relevant target genes are unknown, may be beneficial to the clinician. Although we are continuing to make advancements in our understanding of the 12ql3-14 amplicon, considerably more information is required in order to confirm the true target(s) of gene amplification and to comprehend the role of this amplification and subsequent overexpression in tumour progression. To 78 understand the significance of the overexpression, it is critical to know the function of the involved proteins in the cell. Therefore, it is of primary importance to further analyze the functions of 12ql3-14 genes that are presently unclear (e.g. SAS). By determining the new properties that an amplification event confers to the cell, insight may be gained which links the poorly understood relationship between the genetic and phenotypic alterations seen in tumour cells (10). Additionally, it will be interesting to determine the mechanism(s) by which these amplicons arise, since it is still unknown how they emerge. Of particular interest is the possibility that genetic alterations of other regions may control the emergence of these gene amplifications, or vice versa. It has been proposed that in neuroblastoma and breast cancer, amplification of MYC family genes are related to allelic losses of the distal arm of chromosome l p (33,34). It is therefore of considerable interest to determine which, if any, of the recurring karyotypic abnormalities ( l q l l , 3pl2, l l p l l , and 19pl3) in M F H correlate with amplification of genes from the 12ql3-14 region. The 12ql3-14, remarkably rich in potential oncogenes (31), is proving to be of great importance in our efforts to understand the molecular genetics of many types of cancer. D. R E F E R E N C E S 1. Smith S H , Weiss SW, Jankowski SA, Coccia M A , Meltzer PS (1992): SAS Amplification in Soft Tissue Sarcomas. Cancer Res 52:3746-3749. 2. Khatib Z A , Matsushime H , Valentine M , Shapiro D N , Sherr CJ, Look A T (1993): Coamplification of the CDK4 gene with MDM2 and GLI in human sarcomas. Cancer Res 53:5535-5541. 3. Reifenberger G , Reifenberger J, Ichimura K, Meltzer PS, Collins V P (1994): Amplification of multiple genes from chromosomal region 12ql3-14 in human malignant gliomas: preliminary mapping of the amplicons shows preferential involvement of CDK4, SAS, M D M 2 . Cancer Res 54:4299-4303. 79 4. Turc-Carel C, Limon J, Dal Cin P, Rao U , Karakousis C, Sandberg A A (1986): Cytogenetic studies of adipose tissue tumours. Recurrent reciprocal translocation t(12;16)(ql3-l) in myxoid liposarcomas. Cancer Genet Cytogenet 23:291-299. 5. Mandahl N , Heim S, Johansson B, Bennet K , Mertens F, Olsson G , Rooser B, Rydholm A , Willen H , Mitelman F (1987): Lipomas have characteristic structural chromosomal rearrangements of 12ql3-14. Int J Cancer 39:685-688. 6. Van de Ven WJM, Schoenmakers E F P M , Wanschura S, Kazmierczak B, Kools PFJ, Geurts JMW, Bartnitzke S, Van den Berghe H , Bullerdiek J (1995): Molecular characerization of M A R , a multiple aberration region on human chromosome segment 12ql3-ql5 implicated in various solid tumours. Genes Chromosom Cancer 12:296-303. 7. Oliner JD, Kinzler K W , Meltzer PS, George D L , Vogelstein B (1992): Amplif i -cation of a gene encoding a p53-associated protein in human sarcomas. Nature 358:80-83. 8. Roberts W M , Douglas E C , Peiper SC, Houghton PJ, Look A T (1989): Amplif i-cation of the GLI gene in childhood sarcomas. Cancer Res 49:5407-5413. 9. Forus A , Florenes V A , Maelandsmo G M , Meltzer PS, Fodstad O, Myklebost O (1993): Mapping of amplification units in the ql3-14 region of chromosome 12 in human sarcomas: some amplica do not include M D M 2 . Cell Growth Diff 4:1065-1070. 10. Brison O (1993): Gene amplification and tumour progression. Biochimica et Biophysica Acta 115:25-41. 11. Leach FS, Tokino T, Meltzer P, Burrell M , Oliner JD, Smith S, Hi l l D E , Sidransky D, Kinzler K W , Vogelstein B (1993): p53 mutation and M D M 2 amplification in human soft tissue sarcomas. Cancer Res 53:2231-2234. 12. Reid A H , Tsai M M , Venzon DJ, Wright CF, Lack E E , O'Leary TJ (1996): M D M 2 amplification, P53 mutation, and accumulation of the p53 gene product in malignant fibrous histiocytoma. Diagn Molec Path 5:65-73. 13. Kallioniemi A , Kallioniemi O, Sudar D , Rutovitz D , Gray JW, Waldman F, Pinkel D (1992): Comparative genomic hybridization for molecular cytogenetic analysis of solid tumours. Science 258:818-821. 80 14. Suijkerbuijk RF, Olde Weghuis D E M , Van Den Berg M , Pedeutour F, Forus A , myklebost O, Glier C, Turc-Carel C, van Kessel A G (1994): Comparative genomic hybridization as a tool to define two distinct chromosome 12-derived amplification units in well-differentiated liposarcomas. Genes Chromosom Cancer 9:292-295. 15. Szymanska J, Mandahl N , Mertens F, Tarkkanen M , Karaharju E , Knuutila S (1996): Ring chromosomes in parosteal osteosarcoma contain sequences from 12ql3-15: A combined cytogenetic comparative genomic hybridization study. Genes Chromosom Cancer 16:31-34. 16. Schneider SS, Hiemstra JL, Zehnbauer B A , Taillon-Miller P, LePaslier D L , Vogelstein B, Brodeur G M (1992): Isolation and structural analysis of a 1.2-megabase N-rayc amplicon from a human neuroblastoma. M o l Cell Biol 12: 5563-5570. 17. Nilbert M , Rydholm A , Mitelman F, Meltzer PS, Mandahl N (1995): Characterization of the 12ql3-15 amplicon in soft tissue tumours. Cancer Genet Cytogenet 83:32-36. 18. Ladanyi M , Cha C, Lewis R, Jhanwar SC, Huvos A G , Healey J H (1993): MDM2 gene amplification in metastatic osteosarcoma. Cancer Res 53:16-18. 19. Forus A , Kools PFJ, Schoenmakers E F P M , Van de Ven W J M , Myklebost O (1994): A long range restriction map spanning the myxoid liposarcoma breakpoint in the ql3-14 region of human chromosome 12. H u m Genet 94:259-264. 20. Schmidt E E , Ichimura K, Reifenberger G , Collins P (1994) CDKN2 (pl6/MTSl) gene deletion or CDK4 amplification occurs in the majority of glioblastomas. Cancer Res 54:6321-6324. 21. Pedeutour F, Suijkerbuijk RF, Forus A , Van Gaal J, V a n de Klundert W, Coindre J -M, Nicoloc G, Collin F, Van Haelst U , Huffermann K, Turc-Carel C (1994): Complex composition and coamplification of SAS and M D M 2 in ring and giant marker chromosomes in well-differentiated liposarcoma. Genes Chrom Cancer 10:85-94. 22. Forus A , Maelandsmo G M , Fodstad O, Myklebost O (1992): High correlation between amplification of either A 2 M R (LRP) or GLI and M D M 2 suggests that the genes are part of the same amplification unit in 12ql3-ql4. Cytogenet Cell Genet 61:56-57. 23. Xiao ZX, Chen J, Levine AJ, Modjtahedi N , Xing J, Sellers WR, Livingston D M (1995): Interaction between the retinoblastoma protein and the oncoprotein M D M 2 . Nature 375:694-698. 81 24. Forus A , Florenes V A Maelandsmo G M , Fodstad O, Myklebost O (1994): The proto-oncogene C H O P / GADD153, involved in growth arrest and D N A damage response, is amplified in a subset of human sarcomas. Cancer Genet Cytogenet 78:165-171. 25. Serrano M , Hannon GJ, Beach D (1993) A new regulatory motif in cell-cycle control causing specific inhibition of cycl inD/CDK4. Nature 366:704-707. 26. Xiong Y, Zhang H , Beach D (1993) Subunit rearrangement of the cyclin-dependent kinases is associated with cellular transformation. Gene Dev 7:1572-1583. 27. Jankowski SA, Mitchell DS, Smith S H , Trent JM, Meltzer PS (1994): SAS, a gene amplified in human sarcomas, encodes a new member of the transmembrane 4 superfamily of proteins. Oncogene 9:1205-1211. 28. Weiss SW, Enzinger F M (1978): Malignant fibrous histiocytoma: and analysis of 200 cases. Cancer 41:2250-2266. 29. Ron D, habener JF (1992): C H O P , a novel developmentally regulated nuclear protein that dimerizes with transcription factor C / E B P and L A P and function as a dominant-negative inhibitor of gene transcription. Genes Dev 6:439-453. 30. Ruppert JM, Kinzler K W , Wong AJ, Bigner S H , Kao FT, Law M L , Seuanez H N , O'Brien SJ, Volgelstein B (1988): The GLI-Kruppel family of human genes. Mol Cell Biol 8:3104-3113. 31. Gemmil R M , Paulien S, Cougherty C M , Bolin R, Liao M , Jankowski SA, Sandberg A A , Meltzer PS (1992): A physical map for 12ql3.3 links together several human disease-related genes [ M D M 2 , SAS, and GLI] and includes the myxoid liposarcoma t(12;16) breakpoint region. Cytogenet Cell Genet 61:257. 32. Brodeur G M , Seeger RC, Schwab M , Varmus H E , Bishop J M (1984): Amplification of N-rayc in untreated human neuroblastomas correlates with advanced disease stage. Science 224:1121-1124. 33. Caron H , Peter M , van Sluis P, Speleman F, de Kraker J, Laureys G, Michon J, Brugieres L , Voute PA, Westerveld A (1995): Evidence for two tumour suppressor loci on chromosomal bands lp35-36 involved in neuroblastoma: one probably imprinted, another associated with N-rayc amplification. H u m M o l Genet 4:535-539. 34. Bieche I, Champeme M H , Lidereau R (1994) A tumour suppressor gene on chromosome lp32-pter controls the amplification of M Y C family genes in breast cancer. Cancer Res 54:4274-4276. 82 C H A P T E R IV I N V E S T I G A T I O N O F P U T A T I V E T U M O U R SUPPRESSOR G E N E I N V O L V E M E N T O N C H R O M O S O M E 19 IN PEDIATRIC M F H A . I N T R O D U C T I O N Inactivation or loss of a tumour suppressor gene (TSG) is thought to be one of the most important mechanisms of oncogenesis in humans (1). A n increasing number of TSGs have been isolated and characterized, for example the Wilms' tumour gene (WT1) on l i p , the retinoblastoma gene (RBI) on 13q, and the TP53 gene on 17p (1). D N A deletions near putative or known TSGs have been detected in tumours in which these genes play a role (2-4). Originally these deletions were detected by karyotyping, however, now a more powerful approach exists which assesses tumour allelic loss (or loss of heterozygosity, L O H ) , and is based on the detection of highly polymorphic D N A regions in the patient's constitutional D N A compared to the tumour D N A (2). Studies of allelic loss in human tumours have provided a useful tool for localizing and cloning such TSGs (5). However, to date, there have been no reports describing allelic loss in M F H and no putative TSGs have yet been identified and implicated in the pathogenesis of this disease. As previously reported in adult M F H , abnormalities of chromosome 19 were observed in the present case. It was initially proposed that alterations of 19pl3 in M F H may represent a deletion of material from chromosome 19, potentially resulting in loss of one or more TSGs from this region. One of the derivative 19 chromosomes seen in the present case was add(19)(pl3), which suggests a reciprocal translocation with possible loss of chromosome 19 material at the breakpoint. 83 L O H studies were initiated by assaying chromosome 19 for allelic deletions. Since many tumours exhibit L O H for polymorphic markers in regions containing TSGs, mapping regions of L O H can help identify putative tumour suppressor loci involved in oncogenesis (2, 6). L O H occurs when constitutional D N A is heterozygous for a particular locus and there is complete or partial loss of one allele in the tumour D N A compared to the constitutional D N A . Although this method can reliably identify single allelic loss, it is unsuitable to detect homozygous deletions (7). Assaying for L O H can be done using a PCR-based technique which amplifies highly polymorphic genetic markers mapped to particular chromosomal regions. In order to investigate potential L O H on chromosome 19 in the pediatric M F H we PCR-amplified several chromosome 19 polymorphic loci from both constitutional and tumour D N A (7-11). A second method, the candidate gene approach, was also employed to investigate putative T S G involvement. Using this approach, putative target TSGs known to map to the chromosomal region of interest are identified and analysed for potential alterations using various methods. A n excellent candidate gene for a pathogenetic change common to the 19p+ aberrations found in M F H is INK4d, a new member of the INK4 gene family which maps to 19pl3.2 (11-13). Since the human INK4d gene encodes a protein (pl9lNK4d) which functions as an inhibitor of the cyclin-D dependent kinases, CDK4 and CDK6 (11-13), it is conceivable that alterations of INK4d may result in deregulated cell cycle control leading to pathogenesis. To investigate putative INK4d gene involvement, the tumour D N A was assayed for L O H using microsatelhte markers specifically mapping to 19pl3.2. Southern blot analysis using a p l 9 t N K 4 d c D N A probe allowed INK4d genomic D N A levels in the pediatric M F H to be compared to other pediatric sarcomas, as well as to look for INK4d gene rearrangement and large deletions. Finally, Northern blot 84 analysis was performed to assess INK4d m R N A expression. B. R E S U L T S 1. Chromosome 19 L O H Analysis Tumour D N A was extracted from the frozen primary tissue and normal D N A (peripheral blood lymphocytes) was obtained from the patient, the patient's mother, and the patient's father. 18 highly polymorphic markers from chromosome 19 were chosen for L O H analysis based on their chromosomal locations and on their high degree of heterozygosity (7-11). Next, the four D N A samples (tumour D N A , normal patient D N A , maternal D N A , and paternal D N A ) were assayed for L O H by PCR with primer pairs that flank these polymorphic markers. 14 of the 18 markers tested were "informative" and are listed in Table 5. A marker is considered to be "informative" when the constitutional D N A of the patient is heterozygous at a particular locus and it can be used to determine whether allelic loss has occurred in the tumour D N A , ( D N A from the mother and father are used to ascertain patient heterozygosity). Figure 13 depicts chromosome 19 and the approximate locations of the polymorphic markers evaluated (note that 9 of these markers are from 19pl3 and 5 are specifically from 19pl3.2, the INK4d locus) (11). As shown in Table 5, of the 14 "informative" markers tested, none revealed L O H in the tumour D N A . In other words, the two alleles present in the normal D N A were also present in the tumour D N A . PCR amplification was repeated 3 to 4 times for each marker, showing reproducibility of the results. Representative autoradiographs demonstrating no L O H on chromosome 19 are illustrated in Figure 14. Additionally, there were no significant differences in allelic intensity in the tumour D N A compared to the normal D N A , suggesting that allelic imbalance is not 85 Table 5 . Chromosome 19 Microsatellite Analysis to Assess L O H in BCCH-Sn Marker Location Heterozygosity LOH 19p: D19S209 19pl3.3 0.78 No D19S177 19pl3.3 0.82 No D19S424 19pl3.3 0.79 No D19S221 19pl3.2 0.87 No D19S226 19pl3.2 0.86 No D19S411 19pl3.2 0.76 No E P O 19pl3.2 0.74 No E P O R 19pl3.2 0.87 No D19S199 19pl3.1 0.83 No 19q: D19S425 19ql2-13.1 0.87 No D19S220 19ql3.2 0.85 No D19S228 19ql3.2 0.64 No D19S223 19ql3.2 0.82 No D19S218 19ql3.4 0.60 No 19p 19q 13.3 13.1 13.1 13.3 -D19S209 D19S177 -D19S424 -D19S221 D19S226 D19S411 EPOR EPO -D19S199 D19S425 -D19S220 -D19S228 -D19S223 -D19S218 Figure 13. Informative chromosome 19 polymorphic genetic markers used to investigate L O H in pediatric M F H with chromosome 19 abnormalities. ON ON H 00 H CN ON _ O SH ON OH o N O o * u , 0 o> en ' X u u a; pa +J 5 < £ 2 S O o n 0 B HV 01 n 3 C J 2 O too i — 3 .9 o 0 CD • T H H-> Of OH OJ H-> (fl H H-> CO s O 6 OJ T 3 JH QJ £ 1 3 QJ H r ^ H SH o H-H cfl QJ _> H-» (d + J QJ CO QJ IH OH QJ Pti OJ H-> cn 0 O £ OJ 13 OJ HC o •a < z p < Q ON O O 1—1 (N CO OJ H 3 60 -t-» • QJ 5H o c QJ H-> CC OH 0 S H—* £ *J CC OH 1 pa QJ H—* ca OH «S OH i -o o >H QJ o OH< 88 present in the 19p region. This is apparent in Figure 14, where the two alleles in the tumour D N A are of similar intensities to the two alleles in the normal D N A . It should be noted that these studies do not rule out L O H resulting from small deletions of regions between the polymorphic loci tested in these studies. 2. INK4d: A Candidate Gene Genomic D N A from B C C H - S n and control pediatric sarcoma cell lines were digested with Hindlll and subjected to Southern blot analysis using a human p!9iNK4d c D N A probe to determine if there were rearrangements or obvious deletions involving this gene in the case study (Figure 15). As seen in Figure 15, a single hybridizing restriction fragment (~9 kb) of comparable intensities was detected in all pediatric sarcomas tested, including the pediatric M F H (lane 5). This suggests that no significant deletion of coding regions of the INK4d gene has occurred. It also indicates that there has been no INK4d gene rearrangement, (i.e. no gene fusion involving INK4d) at the 19pl3 breakpoint. The blot was stripped and reprobed with a chromosome 19p probe, E2A, to control for chromosome 19 aneuploidy. To investigate m R N A expression of INK4d in B C C H - S n and other pediatric sarcomas, total R N A from B C C H - S n and 5 pediatric sarcoma cell lines was subjected to Northern analysis using the same p l9 lNK4d c D N A probe. The 1.4 kb pl9lNK4d m R N A transcript was expressed at low levels in all tumours. In particular, no underexpression of the p l9iNK4d transcript in B C C H - S n was apparent compared to the other pediatric sarcomas. < oo LO CO CN ON .a ^ ON in in 90 C. DISCUSSION In order to identify putative TSGs involved in the pathogenesis of M F H with 19p abnormalities, putative regions of L O H on chromosome 19 were investigated. However, no allelic deletions on chromosome 19 were detected in the present case. Although this data does not give us information regarding the site of a possible TSG, it does indicate that there has not been a large deletion of 19pl3. It remains possible, however, that loss of a T S G from chromosome 19 is still involved in this case, but not detected by our assays. The rationale for this approach is based on the "two-hit" model of Knudson and Strong (14), and maintains that TSGs regulate cell proliferation in a negative manner, and therefore, both alleles of the gene must be mutated or lost for malignant transformation to occur. While the first allele may be altered in a variety of ways, the second allele is usually inactivated as a result of a point mutation, deletion, or chromosome loss (15). Depending on the nature of the second allelic mutation, L O H may or may not occur. Large deletions, chromosome loss, and mitotic recombination (which comprise approximately half of the secondary events), leading to inactivation of a TSG, do result in L O H (16, 17). However, point mutations and submicroscopic deletions do not result in detectable L O H using this method. Therefore, L O H assessment can underestimate the extent to which T S G loss from a particular chromosome region may be involved in tumours and subsequent analysis may be required. Additionally, even though allelic deletions on chromosome 19 were not detected in the present pediatric case, we cannot formally exclude the possibility that L O H is not occurring in other M F H with 19pl3 alterations due to the genetic variability seen in different tumours. Another factor to consider when interpreting L O H studies is the possibility that D N A from normal cells might contaminate tumour D N A , potentially obscuring detection of L O H in the tumour tissue. Thrash-Bingham and coworkers 91 found that even with 20% normal D N A contamination, L O H could readily be detected (15). Various studies in the literature have reported the finding of L O H when the intensity of the allele in the tumour D N A was less than 50% of that in the corresponding normal tissue D N A (6, 15, 18). In such instances, the observed partial loss was attributed to either contaminating D N A or to tumour heterogeneity (i.e. gains or losses of alleles in some, but not all tumour cells). In our L O H assessment, contaminating normal D N A did not appear to be a significant concern since with all 14 polymorphic markers the alleles in the tumour D N A were approximately equal in intensity to the alleles in the normal D N A . Changes in chromosome ploidy can also be detected by microsatellite PCR. It has been shown in some tumours that PCR amplification of particular microsatellite sequences consistently yield fragments that differ in intensity, such that a band representing one allele is significantly darker than the other compared to that of normal D N A (15). These intensity differences, which may be due to chromosome gains or nonclonal chromosome losses can be further distinguished by performing FISH on the tumour cells (15). Differences in allelic intensity were not detected at any of the chromosome 19 microsatellite markers studied in the tumour D N A . This would suggest that there is an equal representation of chromosome 19 material from each parent in the tumour D N A . However, as wil l be apparent in Chapter 5, when FISH was performed on the tumour cells, information regarding chromosome 19 aneuploidy was revealed which would imply an allelic imbalance of 19p. It may not have been possible to perceive differences in the allelic intensity of the chromosome 19p microsatellite markers because the difference in intensity between one and two copies was too subtle to detect. Although our results did not indicate a chromosome imbalance by PCR microsatellite analysis, previous data and cytogenetic observations suggest that scanning the genome by this method can 92 reveal aneuploid chromosomes in tumour cells (15). Therefore, since both L O H and allelic imbalance information can be obtained with this method, it may have diagnostic and prognostic applications. Determining chromosome imbalance using this approach would be particularly beneficial for solid tumour analysis since metaphase chromosomes are often not readily available. As an alternative approach to screening genomic material for L O H in order to isolate regions housing putative TSGs, the INK4d gene was studied, as it was believed to be a good candidate for a progression-associated gene for tumours possessing chromosome 19p abnormalities. Previously, it has also been suggested that the insulin receptor gene (INSR), which has homologies to SRC family members and is mapped to 19pl3.2-13.3 (19), is the best candidate for a pathogenetic change common to the 19p+ aberrations (8). Rearrangements involving 19pl3 have been described in approximately half of all ovarian carcinomas (20). It is believed that some of these 19p+ marker chromosomes, reported to occur frequently in high grade human ovarian cancers, result from a deletion that involves a rearrangement of the INSR locus (9). However, involvement of the INSR gene in M F H with 19p alterations has not been reported and no other genes have yet been implicated in M F H pathogenesis. INK4d, the newly discovered 1NK4 gene member, was believed to be an interesting and possibly significant target gene for these chromosome 19 abnormalities for several reasons. Not only does INK4d map to 19pl3.2 (11), the chromosomal region frequently altered in M F H (8), but functional inactivation of this gene has the potential to result in tumour formation through cell cycle deregulation. p l9 iNK4d and the other INK4 proteins ( p l 6 i N K 4 3 / p i5iNK4b and pl8lNK4c) all interact physically with CDK4 and CDK6 to inhibit their cyclin D -dependent kinase activities. The rationale for evolving four INK4 genes remains a 93 mystery. Since all INK4 proteins are similar on a biochemical level, it is not certain how each contributes to C D K 4 and CDK6 regulation in vivo (21). In vitro, pl6iNK4a/ pl8lNK4c/ and pl9lNK4d are equally effective at inhibiting the activities of holoenzymes formed between the D-type cyclins and their catalytic partners (21). Their specificities in vivo, however, may be dependent on their binding to higher order holoenzyme complexes (22). It is possible that the INK4 proteins are regulated in a cell lineage-specific manner and that they respond differentially to anti-proliferative stimuli. For example, it is known that pl5lNK4b is specifically induced in human keratinocytes by TGF-p (23). The current model is that INK4 genes are independently regulated, and that functional loss of individual members may result in selective deficits in cellular responses to antimitotic signals, thus contributing to tumourigenesis (13). Previous studies have shown that INK4a (also referred to as multiple tumour suppressor-1, MTS1 and CDKN2), which maps to chromosome 9p21, frequently undergoes biallelic deletions and/or mutations in several human malignancies (24, 25), indicating that INK4a functions as a tumour suppressor (13). Miller and co-workers reported no alterations of INK4a or INK4b in M F H (26). It has not been determined whether the other INK4 genes, including INK4d, function in a similar tumour suppressing manner. Alterations to INK4d in human malignancies have not yet been extensively studied. Although INK4d has been analysed in a few hematological malignancies, to date there are no reports where INK4d has been analysed in human solid tumours. Okuda and co-workers analysed the genomic status of INK4d in 38 acute lymphoblastic leukemia (ALL) primary tumours and cell lines possessing a variant (I;19)(q23;pl3) translocation, and shown to have either normal INK4a/INK4b loci or homozygous or hemizygous loss of these genes (27). They found no evidence of allelic loss or genomic rearrangement, suggesting that 94 INK4d does not undergo frequent allelic loss in A L L . In the present study of pediatric M F H , also containing 19pl3 alterations, no large allelic deletions, rearrangements, or gene fusions involving INK4d gene were observed. Furthermore, genomic D N A levels of this gene were comparable to the other pediatric sarcomas which did not have 19p abnormalities. However, in order to fully assess the putative involvement of INK4d, one must analyse the INK4d sequence to exclude the possibility of inactivating mutations within either regulatory or coding sequences of INK4d. Since Northern analysis suggests low level expression of INK4d transcription, a mutation or a deletion in the promoter region is not inhibiting transcription. Although the observed transcript was approximately the correct size (1.4 kb), it is possible that a point mutation in the stop codon, which did not significantly alter the size of the transcript, resulted in a non-functional transcript. Alternatively, a transcript may arise with a mutation in the coding region which results in the substitution of an incorrect amino acid during translation. Depending on the site of the alteration and the specific amino acid substitution, the function of the protein may or may not be affected. For example, if the amino acid change prevents the protein from folding into its correct three-dimensional conformation, its function may be compromised. In addition to mutations and deletions at the genomic level, recent findings suggest that gene-specific methylation is another way to "suppress the suppressor" (28). Genes can be transcriptionally modulated by alteration of the methylation pattern of upstream or 5' regions known as C p G islands (29). Tumour cell lines with a methylated pl6 C p G island (and no pl6 gene mutation or deletion) were examined by Merlo and coworkers (28). Not only did they find that the pl6 gene was transcriptionally repressed when methylated, but they also showed that demethylation of the pl6 C p G island restored pl6 expression (28). It is possible that 95 INK4d and the other INK4 genes can be suppressed in a similar manner. As discussed above, there are several mechanisms which can alter the INK4d gene, thereby disrupting normal INK4d gene expression. Since it may not be possible to identify the alterations at the D N A level, information can be gained by assessing the expression of the gene. In the present study, the investigation of INK4d expression in a solid tumour with 19p abnormalities is reported for the first time. To date, little is reported regarding INK4d expression in various tumour groups. Previous studies on INK4d expression have focused on the determination of differential expression in normal tissues and of cell cycle expression levels (6, 8, 18). INK4d is ubiquitously expressed as a 1.4 kb transcript and can be readily detected in most tissues (21). Considering that p l9 has demonstrated potent C D K 4 inhibitory activity in vivo, provoking cell cycle arrest (21), theoretically, an inactivating mutation in the regulatory or coding region of INK4d can prevent cell cycle arrest. In the present study, there did not appear to be any abnormalities in the expression of this gene. Low levels of the expected 1.4 kb INK4d transcript were detected in the pediatric M F H and in the other pediatric sarcomas. Since the appropriately sized transcript was seen, it appears that there were no significant mutations in the regulatory sequences of this gene. It can also be inferred from this data that the INK4d gene is not likely involved at the 19pl3 breakpoint in a reciprocal translocation. If this was the case, then one might expect to see an altered transcript as a result of a fusion of INK4d with a gene from the reciprocal chromosome. Although the principle objective in the L O H studies was to identify chromosomal deletions or rearrangements that might indicate a site of an altered T S G , it warrants mention that such structural chromosomal abnormalities can also result in the activation of oncogenes (9). There are several proto-oncogenes which map to 19pl3, including JUNB, JUND, and INSR (30). Many of the microsatellite 96 markers that were used in this L O H analysis of chromosome 19 are near these genes. Even though our allelic analysis did not reveal L O H in this region, there may still be alterations of these genes. For example, a balanced translocation involving a proto-oncogene will not be detected by L O H analysis. In summary, no large deletions or rearrangements of 19p were detected in pediatric M F H with abnormalities of this region. Thus, no potential sites housing a TSG were identified. Similarly, the putative TSG, INK4d, revealed no large genomic aberrations and normal INK4d m R N A transcripts. Therefore, our data is not consistent with 19pl3 alterations in M F H representing deletions of material from chromosome 19 as originally hypothesised. However, due to limited tumour material it was not possible to investigate putative submicroscopic deletions in this region. Furthermore, considering tumour heterogeneity, it wil l be necessary to assess additional cases of M F H with 19p abnormalities. D. R E F E R E N C E S 1. Seizinger BR, Klinger HP, Junien C, Nakamura Y, Le Beau M , Cavenee W, Emmanuel B, Ponder B, Naylor S, Mitelman F, Louis D , Menon A , Newsham I, Decker J, Kaelbling M , Henry I, Deimlin A V (1991): Report of the committee on chromosome and gene loss in human neoplasis. Cytogenet Cell Genet 58: 1080-1096. 2. Marshall CJ (1991): Tumor suppressor genes. Cell 64:313-326. 3. Fujiwara Y, Emi M , Ohata H , Kato Y, Nakajima T, Mor i T, Nakamura Y (1993) : Evidence for the presence of two tumor suppressor genes on chromosome 8p for colorectal cancer. Cancer Res 53:1172-1174. 4. Hampton G M , Penny L A Baergen R N , Larson A , Brewer C, Liao S, Busby -Earle R M C , Williams A W R , Steel C M , Bird C C , Stanbridge EJ, Evans G A (1994) : Loss of heterozygosity in cervical carcinoma: Subchromosomal localization of a putative tumour gene to chromosome Ilq22-q24. Proc Natl Acad Sci USA 91:6953-6957. 97 5. Ritland SR, Ganju V , Jenkins RB (1995): Region-specific loss of heterozygosity on chromosome 19 is related to the morphologic type of human glioma. Genes Chromosom Cancer 12:277-282. 6. Raskind W H , Conrad E U , Matsushita M (1996): Frequent loss of heterozygosity for markers on chromosome arm lOq in chondrosarcomas. Genes Chromosom Cancer 16:138-143. 7. von Deimling A , Bender B, Jahnke R, Waha A , Kraus J, Albrecht S, Wellenreuther R, Fabbender F, Nagel J, Menon A G , Louis D N , Lenartz D , Schramm J, Wiestler O D (1994): Loci associated with progression in astrocytomas: a candidate on chromosome 19q. Cancer Res 54:1397-1401. 8. Mandahl N , He im S, Willen H , Rydholm A , Eneroth M , Nilbert M , Kreicbergs A , Mitelman F (1989): Characteristic karyotypic anomalies identify subtypes of malignant fibrous hisitiocytoma. Genes Chromosom Cancer 1:9-14. 9. Amfo K, Neyns B, Teugels E , Lissens W, Bourgain C, De Sutter P, Vandamme B, Vamos E , De Greve J (1995): Frequent deletion of chromosome 19 and a rare rearrangement of 19pl3.3 involving the insulin receptor gene in human ovarian cancer. Oncogene 11:351-358. 10. Choong P F M , Mandahl N , Mertens F, Willen H , Alvegard T, Kreicbergs A , Mitelman F, Rydholm A (1996): 19p+ marker chromosome correlates with relapse in malignant fibrous histiocytoma. Genes Chromosom Cancer 16: 88-93. 11. Guan K L , Jenkins CW, L i Y, O'Keefe C L , Noh S, W u X, Zariwala M , Matera A G , Xiong Y (1996): Isolation and characterization of p l9 iNK4d, a pl6-related inhibitor specific to CDK6 and CDK4. Mol Biol Cell 7:57 -70. 12. Chan F K M , Zhang J, Cheng L , Shapiro D , Winoto A (1995): Identification ofj Human and Mouse pl9, a novel CDK4 and C D K 6 inhibitor with homology to pl6ink4 M o l Cell Biol 15:2682-2688. 13. Okuda T, Hirai H , Valentine V A , Shurtleff SA, K i d d VJ, Lahit JM, Sherr CJ, Downing JR (1995): Molecular cloning, expression pattern, and chromosomal localization of human C D K N 2 D / INK4d, an inhibitor of cyclin D-dependent kinases. Genomics 29:623-630. 14. Knudsen A G , Strong L C (1971): Mutation and cancer: Statistical study of retinoblastoma. Proc Natl Acad Sci U S A 68: 820-823. 9 8 15. Thrash-Bingham C A , Greenberg RE, Howard S, Bruzel A , Bremer M , Goll A , Salazar H , Freed JJ, Tartof K D (1995): Comprehensive allelotyping of human renal cell carcinomas using microsatellite D N A probes. Proc Natl Acad Sci U S A 92:2854-2858. 16. Cavenee W K , Dryja TP, Phillips RA, Benedict WF, Godbout R, Gallie BL, Murphree BL, Strong A L , White RL (1983): Expression of recessive alleles by chromosomal mechanisms in retinoblastoma. Nature 305:779-784. 17. Gnarra JR, Tory K, Weng Y, Schmidt L , Wei M H , L i H , Latif F, L iu S, Chen F, D u h F M (1994): Mutations of the V H L T S G in renal carcinoma. Nat Genet 7:85-90. 18. Bieche I, Champeme M , Lidereau R (1994): A tumor suppressor gene on chromosome lp32-pter controls the amplification of M Y C family genes in breast cancer. Cancer Res 54:4272-4276. 19. Yang-Feng T L , Han H , Chen K (1993): Allelic loss in ovarian cancer . Int J Cancer 54:546-551. 20. A m a n P, Pejovic T, Wennborg A , Heim S, Mitelman F (1993): Mapping of the 19pl3 breakpoint in an ovarian carcinoma between the INSR and TCF3 loci. Gene Chromosom Cancer 8:134-136. 21. Hirai H , Martine FR, Kato J, Ashmun RA, Sherr CJ (1995): Novel INK4 proteins, p l9 and pl8, are specific inhibitors of the cyclin D-dependent kinases CDK4 and CDK6. Mol Cel Biol 15:2672-2681. 22. Xiong Y, Hannon GJ, Zhang H , Casso D, Kobayashi R, Beach D (1992): p21 is a universal inhibitor of cyclin kinases. Nature (London) 366:701-704. 23. Hannon GJ, Beach D (1994): pl5iNK4b i s a potential effector of cell cycle arrest mediated by TGF-p. Nature 371:257-261. 24. Kamb A , Gruis N A , Weaver-Feldhaus J, L i u Q, Harshman K, Tavtigian SV, Stockert E , Day RS, Johnson BE, Skolnick M H (1994): A cell cycle regulator potentially involved in genesis of many tumor types. Science 264:436-440. 25. Nobori T, Mirua K, W u DJ, Lois A , Takabayashi K, Carson D A (1994): Deletions of the cyclin-dependent kinase-4 inhibitor gene in multiple human cancers. Nature 368:753-756. 26. Miller C W , Aslo A , Campbell MJ, Kawamata N , Lampkin B C , Koeffler H P (1996): Alterations of the pl5, pl6, and pl8 genes in osteosarcoma. Cancer Genet Cytogenet 86:136-142. 99 27. Okuda T, SHurtleff SA, Valentine M B , Raimondi SC, Head DR, Behm F, Curcio-Brint A M , L iu Q, Pui C - H , Sherr CJ, Beach D, Look A T , Downing JR (1995): Frequent deletion otpl6iNK4a/ MTS1 and pl5INK4b/MTS2 in pediatric acute lymphoblastic leukemia. Blood 85:2321-2330. 28. Merlo A , Herman JG, Lee DJ, Gabrielson E, Burger PC, Baylin SB, Sidransky D (1995): 5' C p G island methylation is associated with transcriptional silencing of the tumour suppressor p l 6 / C D K N 2 / M T S 1 in human cancers. Nat Med 1:686-692. 29. Little M , Wainwright B (1995): Methylation and pl6: suppressing the suppressor. Nature Med 1:633-634. 30. Ashworth L, Batzer M A , Brandiff B, Branscomb E , de Jong P, Garcia E G , Games JA, Gordon L A , Lamerdin JE, Lennon G, Mohrenweiser H , Olsen AS, Slezak T, Carrano A V (1995): A n integrated metric physical map of human chromosome 19. Nature Genetics 11:422-427. 100 C H A P T E R V F L U O R E S C E N C E IN SITU H Y B R I D I Z A T I O N (FISH) C H A R A C T E R I Z A T I O N O F C H R O M O S O M E 19 A B N O R M A L I T I E S IN PEDIATRIC M F H A . I N T R O D U C T I O N Using conventional cytogenetic methods, accurate identification of small marker chromosomes and subtle rearrangements has traditionally been a challenge (1). With the development of fluorescence in situ hybridization (FISH), however, numerical and structural abnormalities can be detected more precisely and with increased sensitivity (2). As a result, FISH has proven to be a powerful tool for characterizing the complex chromosomal aberrations frequently seen in human solid tumours (3-7). As described in Chapter 2, cytogenetic analysis of the the presented pediatric M F H (BCCH-Sn), revealed a hypertriploid tumor with multiple structural and numerical aberrations, including rearrangements to chromosome 19 (see tumour karyotype in Chapter 2, Figure 7). Although chromosome 19 abnormalities are reported to occur non-randomly in M F H , these alterations have not yet been well characterized. Therefore, FISH was used in the present study to further characterize the identified chromosome 19 abnormalities in B C C H - S n . Although these were originally described as 19p abnormalities based on standard chromosome banding methods [+add(19)(pl3), +add(19)(pll)], the concern is that G-banding cannot accurately determine whether these are der(19p) or der(19q). Considering the complex and subtle nature of chromosomal rearrangements in solid tumours and the complexity of the karyotype seen in BCCH-Sn , FISH has the potential to shed 101 considerable light on chromosome 19 aberrations in M F H . Though the additional material described at 19p is suggestive of a translocation, the reciprocal chromosome(s) partners could not be determined with standard banding techniques. Clarification of structural abnormalities, such as translocations, can be best achieved by applying chromosome-specific D N A libraries, commonly referred to as whole chromosome painting probes (WCPs). Therefore, to investigate possible translocations involving chromosome 19 and the reciprocal chromosome(s), we first performed whole chromosome FISH with a chromosome 19 W C P on metaphase chromosomes from B C C H - S n . However, since chromosome paints can only identify the origin of the chromosomal material, region specific probes are required to sublocalize the chromosomal breakpoint. So, in order to determine whether the additional chromosome 19 material identified with a chromosome 19 W C P originated from the long arm (q) or the short arm (p) of the chromosome, dual-coloured FISH with differentially labelled 19q and 19p specific probes was performed on tumour metaphase spreads and interphase nuclei. Finally, in order to identify the reciprocal chromosome(s) involved in the chromosome 19 translocation, dual-coloured whole chromosome FISH was performed on B C C H - S n metaphase spreads using a chromosome 19 W C P with differentially labeled WCPs representing candidate partner chromosomes. B. R E S U L T S 1. Whole Chromosome FISH Using a Chromosome 19 Painting Probe Initially, the digoxigenin-labeled chromosome 19 W C P ( C O A T A S O M E 19) was applied to human metaphase spreads from normal fibroblasts (Figure 16). C O A T A S O M E 19 fluoresces yellow (FITC) and the chromosomes were 102 Figure 16. Whole chromosome FISH analysis with C O A T A S O M E 19 total chromosome probe on a metaphase spread from normal human fibroblasts, final magnification 600x. Normal chromosomes 19 fluoresce yellow (FITC); chromosomes are counterstained with PI. 103 counterstained orange with propidium iodide (PI). Non-specific hybridization did not occur and the two normal chromosomes 19 were intensely visualized. Having established specific hybridization of C O A T A S O M E 19 to normal metaphase chromosomes, the painting probe was applied to slides containing B C C H - S n metaphase spreads from short- and long- term cultures. Figure 17 demonstrates C O A T A S O M E 19 hybridization to a representative metaphase from a short-term culture of BCCH-Sn . Figure 18 demonstrates C O A T A S O M E 19 hybridization to two representative metaphases from long-term cultures of B C C H - S n . As illustrated, two normal chromosome 19 signals were routinely seen, as well as three abnormal chromosome 19 signals on the ends of three unidentified chromosomes. This pattern of five C O A T A S O M E 19 signals was consistently detected in all metaphases (-10 metaphases) from both short- and long-term cultures. Not only did whole chromosome FISH with C O A T A S O M E 19 confirm the initial cytogenetic diagnosis which stated that the cultured tumour possessed numerous abnormalities of chromosome 19, but it also provided us with more detailed information regarding these alterations. Specifically, three additional translocated pieces of chromosome 19 material are present in this pediatric M F H in addition to two normal copies of chromosome 19. 2. Chromosome Banding of C O A T A S O M E 19 Hybridized Metaphases In an effort to identify the reciprocal chromosome(s) involved in the chromosome 19 translocations, attempts were made to destain and G-band the B C C H - S n slides that had been hybridized with C O A T A S O M E 19. This method, however, proved to be too severe for the chromosomes which had previously been through a harsh denaturation process. Chromosome integrity was not maintained and banding was therefore not possible. A second attempt was made to band the 104 Figure 17. Whole chromosome FISH analysis with C O A T A S O M E 19 total chromosome probe on a metaphase spread from B C C H - S n (short-term culture), final magnification 400x (A) and 600x (B). Three derivative 19 chromosomes in addition to two normal chromosome 19s are apparent. 105 A B Figure 18. Whole chromosome FISH analysis with C O A T A S O M E 19 total chromosome probe on two metaphase spreads from B C C H - S n (long-term culture), final magnification 400x (A, B). Three derivative 19 chromosomes in addition to two normal chromosome 19s are apparent. 106 A B Figure 19. A B C C H - S n metaphase probed with C O A T A S O M E 19 is shown, final magnification 600x (A), prior to staining with D A P I / P I to obtain chromosome banding, final magnification 600x (B). Arrows in the D A P I / P I stained metaphase indicate chromosome 19 material. 107 chromosomes on the remaining C O A T A S O M E 19 slides using a D A P I / P I solution. Figure 19 demonstrates a B C C H - S n metaphase initially hybridized with C O A T A S O M E 19 (Figure 18B) and then stained with DAPI /PI . Although this resulted in a minor improvement in chromosome band resolution, it was not adequate for definitive identification of the reciprocal chromosomes. 3. Dual-Coloured FISH Using a Digoxigenin-Labeled 19ql3.1-Specific D N A Probe and Biotin-Labeled 19p Cosmid Probes Since whole chromosome FISH does not reveal information regarding the origin of the extra chromosome 19 material (i.e. whether it originated from 19q or 19p), additional FISH analysis was required. In order to distinguish between chromosome 19q and 19p chromatin, dual-coloured FISH was performed with a digoxigenin-labeled 19ql3.1 DNA-specific cosmid probe and biotin-labeled 19p cosmid probes. Initially, the digoxigenin-labeled 19ql3.1 probe and the biotin-labeled 19p cosmid probes were applied individually to metaphase spreads and interphase nuclei from normal peripheral blood lymphocytes (Figure 20). Comparison of Figure 20A with 20B allows one to appreciate the differences when the identical digoxigenin-labeled 19ql3.1-specific probe is detected with FITC (yellow) against PI counterstained chromosomes (orange) (Figure 20A) versus detection with rhodamine (red) against DAPI counterstained chromosomes (blue) (Figure 20B). In Figure 20C, the biotin-labeled 19p cosmid probes fluoresce green (FITC), against DAPI counterstained chromosomes. Since sequence-specific probes have a unique target, and therefore a smaller signal, two chromatid signals are often seen. Next, using the 19q and 19p probes described above, dual-colour FISH was performed on metaphase spreads from normal peripheral blood lymphocytes. The 108 Figure 20 . FISH analysis with a digoxigenin-labeled 19ql3.1-specific D N A probe detected with FITC and counterstained with PI (A); the same probe detected with rhodamine and counterstained with DAPI (B); and with biotin-labeled 19p cosmid probes (C) on metaphase spreads and interphase nuclei from normal human lymphocytes, final magnification lOOOx. 109 no F i g u r e 21. D u a l - c o l o u r e d F I S H h y b r i d i z a t i o n of the 19ql3.1-specif ic D N A probe (red) and the 19p c o s m i d probes (green) to a n o r m a l ch romosome 19. C o m p a r e d to the i d i o g r a m of ch romosome 19, the 19ql3.1 signals (red) i l lus t ra te specif ic h y b r i d i z a t i o n to sub-band 19ql3.1 a n d the 19pl3.3 signals (green) i l lustrate specific h y b r i d i z a t i o n to sub-band 19pl3.3. I l l idiogram of chromosome 19 in Figure 21 allows one to appreciate localization of the 19p probe to the telomeric region of 19pl3.3 and localization of the 19q probe to 19ql3.1. As illustrated in Figure 22A, when these probes are applied to a normal chromosome spread, we see two 19q signals (red) and two 19p signals (green). These signals can also be detected on normal interphase nuclei (Figure 22B). Having established specific hybridization and dual-colour detection on normal cells, the probes were applied to remaining B C C H - S n slides in parallel with normal control slides. Despite continued success with the controls, it proved to be a considerable challenge to hybridize the two probes, particularly the 19p cosmids, to the tumour material. The Oncor 19ql3.1-specific D N A probe hybridized much more readily to B C C H - S n than the 19p cosmid probes, which had not been prepared commercially. This can be appreciated in Figure 23, which represent the results obtained when both 19q and 19p probes were applied to BCCH-Sn interphase cells. Four 19ql3.1 signals were consistently detected. However, 19p signals were not seen in B C C H - S n interphase nuclei. Note that 19p signals were detected in the corresponding control slides, suggesting that the hybridization problem was not due to an inadequate probe or FISH procedure, but rather due to an inadequate D N A target for the probe to bind to (i.e. non-optimal quality D N A ) . Nevertheless, these data provided us with novel information regarding the origin of the additional chromosome 19 material that was previously detected in BCCH-Sn . We reasoned that two of the four 19ql3.1 signals observed in the B C C H - S n interphase nuclei represent the two normal 19q arms, and that the other two 19ql3.1 signals represent two of the three extra pieces of chromosome 19 material. Based on this data, we hypothesized that the third extra piece of chromosome 19 originated from 19p. Evidence supporting this hypothesis emerged when successful hybridization of the 19q and 19p probes occurred on a B C C H - S n metaphase (Figure 24). Not only is specific hybridization of 112 A B Figure 2 2 . Dual-coloured FISH analysis with the 19ql3.1-specific D N A probe (red) and the 19p cosmid probes (green) on a metaphase spread from normal human lymphocytes (A) and on an interphase nuclei from normal human lymphocytes (B), final magnification lOOOx. 113 Figure 23. Dual-coloured FISH analysis with the 19ql3.1-specific D N A probe and the 19p cosmid probes on interphase nuclei from BCCH-Sn , final magnification lOOOx. As can be appreciated by the presence of red signals only (19ql3.1), the 19p cosmids did not hybridize to the tumour nuclei. Four 19ql3.1 signals can be seen in all interphase nuclei presented. 114 Figure 24. Dual-coloured FISH analysis with the 19ql3.1-specific D N A probe (red) and the 19p cosmid probes (green) on a metaphase from BCCH-Sn . This partial metaphase demonstrates 19ql3.1 material translocated to the ends of two derivative chromosomes (red) and 19pl3.3 material translocated to the end of a third derivative chromosome (green). 115 the 19ql3.1 probe to the ends of two derivative chromosomes illustrated, but specific hybridization of the 19p cosmid probes to the end of a third derivative chromosome is demonstrated. This provided concrete evidence that two of the extra chromosome 19 pieces originated from 19q and one originated from 19p. Furthermore, this is consistent with the C O A T A S O M E 19 FISH results, which identified three additional pieces of chromosome 19 on three derivative chromosomes (Figures 17 and 18). Unfortunately, lack of usable metaphases prevented further 19p/19q experiments. Although we did make several attempts to reuse the previously hybridized slides (as described in Chapter 2), successful hybridization of the 19p probe was not achieved on these slides. After performing FISH two to three times on each slide, the chromosomes became overdenatured and were no longer suitable for FISH. Additionally, "touch preparations" of frozen primary tumour tissue were tried as an additional source of material for FISH. Dual-coloured FISH with the 19q and 19p probes was attempted on these interphase cells, as well as on normal kidney and lung tissues; however, hybridization and amplification of the 19p cosmid probes was unsuccessful on both the control cells and the tumour cells (negative data not shown). 4. Dual-Coloured Whole Chromosome FISH Using Chromosome 4 and 19 Painting Probes Since it was not possible to identify the reciprocal chromosome(s) involved in the chromosome 19 translocation with banding methods, it was hoped that information could be gained using dual-coloured whole chromosome FISH using probes for other chromosomes. The difficulty with this approach lay in choosing the correct painting probe to use with C O A T A S O M E 19, based on our limited 116 knowledge of the reciprocal chromosome(s). With an ample supply of metaphase spreads, one could apply various WCPs with C O A T A S O M E 19, until the paint representing the reciprocal chromosome was found. In the present case, however, we possessed a very small supply of metaphases which would quickly be depleted after several W C P applications. In order to choose the second probe, the C O A T A S O M E 19 derivative chromosomes (Figures 17 and 18) was compared with G-banded chromosomes in the original karyotype, looking for resemblances in chromosome length and the position of the centromere. Based on similar morphological features and banding features previously obtained with D A P I / P I staining (Figure 19), it was felt that the material adjacent to the 19 derivatives most closely resembled chromosome 4. Therefore, a chromosome 4 W C P was chosen to be used for dual-coloured FISH with the chromosome 19 W C P . The white arrowheads in Figure 7 (Chapter 2) highlight the derivative 19 chromosome candidates from the original karyotype that were believed to contain chromosome 4 material. First, the biotin-labeled chromosome 4 W C P and the digoxigenin-labeled chromosome 19 W C P were hybridized to normal peripheral blood lymphocytes. Specific hybridization of these two chromosome paints to their respective chromosomes is evident in the metaphase spread in Figure 25A and B where the red signals represent chromosome 19 and the green signals represent chromosome 4. Since the chromosomes in interphase cells are despiralized threads (8), the signals in most interphase cells are quite diffuse and it is difficult to distinguish the individual signals (Figure 25A). As seen in Figure 25C, the use of WCPs has enabled us to visualize chromosomes in a prophase cell, the stage of mitosis when chromosomes first become visible as long thin threads that gradually shorten and thicken (8). The WCPs were then applied to the remaining two B C C H - S n slides. 117 B Figure 25. Dual-coloured whole chromosome FISH analysis with C O A T A S O M E 19 total chromosome probe (red) and a chromosome 4 painting probe (green) on a metaphase spread (and interphase cells) from normal human lymphocytes, final magnification 400x (A) and lOOOx (B); and on a prophase cell (center) from normal human lymphocytes, final magnification lOOOx (C). 118 Figure 25 C. Legend on previous page. 119 A B Figure 26. Dual-coloured whole chromosome FISH analysis with C O A T A S O M E 19 total chromosome probe (red) and a chromosome 4 painting probe (green) on metaphase spreads from BCCH-Sn, final magnification lOOOx (A-C). Arrows indicate two t(4;19) translocations. 120 Figure 26 C. Legend on previous page. 121 Successful hybridization occurred to the three metaphases that were present on the slides. A l l three metaphases demonstrated consistent abnormalities involving chromosomes 4 and 19 (Figure 26A-C), although hybridization to the normal chromosomes 4 and 19 was distinctive in only one of the three metaphases (Figure 26A). This is not surprising considering that the chromosomes spread poorly in these metaphases, resulting in overlapping chromosomes. Figure 26A, however, clearly illustrates hybridization to one normal chromosome 4 (large green signal in the lower left of the metaphase) and two normal chromosome 19s (the two red signals in the lower left of the metaphase). Multiple abnormal signals can also be seen in Figure 26A: there are three extra pieces of chromosome 19 (which is consistent with the data previously documented in Figures 17 and 18), and the chromosome 4 W C P revealed several abnormal pieces of chromosome 4. As indicated by the arrows, two similar translocation derivatives involving chromosome 4 and chromosome 19, which we will refer to as t(4;19), can be clearly identified in this metaphase by the juxtapositioning of the red and green signals (Figure 26A). Chromosome 4 material was not found adjacent to the third abnormal piece of chromosome 19. A similar pattern of two t(4;19) derivatives was seen in the other two metaphases (Figures 26B, C). In Figures 26A & C a third unidentified chromosomal segment was seen on the telomeric end of the chromosome 4 material in the rearranged chromosomes, indicating an additional chromosomal rearrangement. This extra material is not clearly evident in Figure 26B, likely due to the poor presentation of the metaphase chromosomes. Using the available probes on the slides that demonstrated the t(4;19) translocations, attempts were made to identify which regions of chromosome 19 were involved by performing dual-coloured FISH with the digoxigenin-labeled 19ql3.1 specific probe and the biotin-labeled chromosome 4 WCP. The problem 122 with hybridizing these two types of probes together is that the sequence-specific probes require very low stringency post-hybridization conditions and the chromosome paints require very high stringency washes. Although several post-hybridization conditions were tried, as expected it was not possible to achieve conditions that retained the D N A specific probe while sufficiently eliminating the non-specific signals resulting from the repetitive regions in the painting probes. Furthermore, since WCPs result in large, diffuse signals, even if hybridization of the specific-DNA probe did occur, its signal would undoubtedly have been masked by the WCP. C. DISCUSSION As demonstrated, FISH can be a powerful tool for enhancing the resolution of solid tumour cytogenetic analysis. In this assessment of chromosome 19 abnormalities in a pediatric M F H (BCCH-Sn), whole chromosome painting allowed aneuploidy and subtle translocations to be strikingly visualized. Specifically, two normal 19 chromosomes and three derivative 19 chromosomes were identified. Two of these extra pieces of chromosome 19 appear to include 19ql3.1 material, and the other includes 19pl3.3 material. Two of these chromosome 19 pieces are involved in a translocation with chromosome 4, t(4;19). Additional chromosomal material was observed at the telomeric end of chromosome 4 in the t(4;19) chromosomes; however, its origin was not further investigated since characterization of the chromosome 19 breakpoint was the primary concern. We do not have definitive evidence that demonstrates whether the chromosome 19 breakpoint occurred in 19q or 19p (i.e. whether chromosome 19q or 19p material is juxtaposed to chromosome 4). However, based on the following 125 19q der (4) 4 19q is 19 3-way rearrangement 19p der (19) der(?) 2n—**4n \ r 4 I 19 normal der(4) t(4;19) x2-der(19) t(19;?) x2-der(?) t(4;?) x2-chromosome 4 x2-chromosome 19 x2-•* rearrangement with-a third chromosome (may occur before or after duplication) - chromosome loss duplication, rearrangements chromosome loss C H R O M O S O M E S D E T E C T E D BY FISH • der(4) t(4;19;?) x2 >- der(19) t(?;19) x l several unidentified derivatives with chromosome 4 material - • chromosome 4 x l chromosome 19 x2 Figure 27. Putative series of genetic events that may have occurred in B C C H - S n to result in two der(4) t(4;19) involving 19q and which have undergone further rearrangements to result in two t(4;19;?) derivatives; one der(19) t(19;?); and two normal chromosomes 19. Since the t(4;19) breakpoints in these arrangements are not presently known in B C C H - S n , the 4q35 and 19ql3 breakpoints previously described in sarcomas with t(4;19) translocations (9-11) have been used as examples for the sake of simplification. 126 breakpoints in B C C H - S n are not known, the 4q35 and 19ql3 breakpoints previously-described in sarcomas with t(4;19) translocations (9-11) will be used for the sake of simplicity. One explanation then is that a three-way translocation occurred between chromosome 19, chromosome 4, and a third unidentified chromosome (?). As depicted in Figure 27, this would result in three derivative chromosomes: der(4) t(4;19)(q35;ql3), der(19) t(19;?)(ql3;?), and der(?) t(4;?)(q35;?). There would also be one normal chromosome 4, 19, and (?). Since ultimately there are two t(4;19) and two normal chromosomes 19 based on FISH results, chromosome duplication may have occurred. Although there may have been individual chromosome duplication events, it is more likely that the entire genome was duplicated to become tetraploidy (4n), and then individual chromosome loss occurred. Based on our data, there would then have been chromosomal loss of one 19p derivative, i.e. der (19), and one normal chromosome 4. Based on the FISH images obtained with chromosome 4 and 19 WCPs, it appears that an additional rearrangement has occurred in the chromosome 4 region of der(4) to result in the addition of unidentified chromosomal material that is juxtaposed to chromosome 4, i.e. der(4) t(4;19;?). If this rearrangement occurred prior to duplication, then the two der(4) would contain the same chromosomal material. However, if these were random rearrangements after duplication, they likely represent different chromosomes. Although this additional material cannot be clearly seen in all of the t(4;19) chromosomes in Figure 26, the material is likely present but is not visible due to the poor spreading of the metaphase chromosomes. Furthermore, all three of the aberrant 19 chromosomes identified with the chromosome 19 W C P and with the chromosome 19 specific probes (Figures 17, 18 & 24) reveal long chromosomes, supporting the presence of additional material that is juxtaposed to the small piece of chromosome 4 in der(4). Subsequent chromosome 4 rearrangements must also 127 have occurred to result in the multiple chromosome 4 signals detected when the chromosome 4 W C P was applied to tumour metaphases (note multiple green signals in Figures 26A-C). Other than a t(15;19)(ql2;pll) translocation described in a myxoid M F H (14), no other chromosome 19p translocations have been documented in M F H . Although there have been many reports suggesting translocations involving chromosome 19 as manifested by the presence of additional material at chromosome 19p (15,16) and 19q (14,17), the additional material juxtaposed to chromosome 19 in these derivatives has not yet been identified. This is likely due in part to the methods used to ascertain the derivative chromosomes; in other words, chromosome banding techniques rather than FISH. Mandahl and coworkers performed standard cytogenetic analyses of 25 M F H , 8 of which were reported to have "material of unknown origin" translocated to 19pl3 (15). The 19p+ material in 3 of the 8 cases was reported to "look alike", suggesting a possible non-random translocation, and had a banding pattern "reminiscent " of the distal part of 6q or 13q (15). Presently, t(4;19) has been identified as a candidate for a tumour-specific marker in M F H . Although this result must be strengthened by a greater number of cases, the t(4;19) may represent a marker chromosome that is specific for M F H and can be used as a diagnostic marker, in addition to providing valuable information regarding tumourigenesis and the metastatic potential of M F H . Additional research is required in order to fully characterize the t(4;19) alteration and to understand its putative oncogenic role in M F H . Firstly, the incidence of this translocation must be assessed in a larger series of primary M F H tumours, so that it can be determined whether this alteration is a non-random occurrence in M F H . If t(4,T9) is found to be non-random, it will be important to specify the particular M F H subtype in which this alteration is found. Just as some 128 tumour subtypes are characterized by a specific cytogenetic abnormality (4,18), the t(4;19) may prove to be specific for one of the histologic subtypes of M F H . As demonstrated with BCCH-Sn , a rapid and sensitive method to screen other M F H for this translocation is dual-coloured FISH using chromosome 4 and 19 WCPs on tumour metaphase chromosomes. Currently, our laboratory is screening additional M F H with chromosome 19 aberrations (generously supplied by N . Mandahl, Sweden) for the t(4;19), using dual-coloured FISH with chromosome 4 and 19 WCPs. Considering the complexity of many M F H karyotypes, it is important to screen all M F H with abnormal karyotypes, not only tumours with chromosome 19 abnormalities detected with chromosome banding methods. With this powerful technique we may find that chromosome 19 alterations in M F H are a more common occurrence than previously thought. Furthermore, efforts must be made to identify chromosomes other than chromosome 4 that are rearranged with chromosome 19, including the reciprocal chromosome in the third derivative 19 marker in B C C H - S n which remains to be characterized. A technique involving chromosome microdissection that wil l be described in more detail later, can be used to create region-specific probes to identify these unknown chromosomal segments (3). It will also be important to characterize all chromosome 19 derivatives that are identified, since at the present time we do not know how many oncogenically significant derivatives exist. Since variant translocations involving a common chromosome are frequently observed in one tumour type (18-20), it is possible that the same gene with the chromosome 19 breakpoint (e.g. 19pl3 or 19ql3) is involved in variant translocations in M F H . Recurring chromosomal changes can provide critical clues as to the location of the genes involved in the genesis of specific tumours (19). Therefore, if the t(4;19) or another similar recurring translocation is found, the next step is to characterize 129 the breakpoint region so that the genes involved in the rearrangement can be identified and isolated. It is critical to identify the participating genes for several reasons. Firstly, knowledge of the genes altered at specific genetic sites and resulting fusion genes and transcripts can lead to the generation of molecular probes which may form the basis of sensitive and specific diagnostic tests for the particular class of tumours. Furthermore, characterizing the specific genetic alterations at these sites will allow the molecular basis of the translocation to be elucidated, potentially yielding valuable information regarding the oncogenesis of M F H and other neoplasms. Although in our analysis whole chromosome painting was an excellent method for identifying the unknown and subtle translocation, t(4;19), WCPs have the limitation of recognizing an entire chromosome rather than identifying each chromosome arm or particular bands of the chromosome (21). Hence, WCPs do not provide the regional sublocalization information which is necessary to clone specific breakpoints in chromosomal translocations (3). Therefore, additional methods that are capable of isolating the specific chromosomal region and defining the breakpoint regions must be performed. Although it was not possible to perform such methods on B C C H - S n due to a lack of metaphase chromosomes, these methods can be used for future analysis of t(4;19) and other recurring translocations found in M F H . Two different approaches that could be taken to further characterize a translocation breakpoint, both of which employ the FISH technique, wil l now be discussed. The first method involves the basic FISH technique and the use of several different kinds of probes, including chromosome arm painting probes (CAPs) and region-specific probes. Following the use of WCPs, for example, recently developed CAPs can be used to identify the specific chromosomal arms involved in the rearrangement (21). If CAPs had been available at the time of our study, once it was 130 determined that chromosomes 19 and 4 were involved in a translocation, dual-coloured FISH using differently labelled chromosome 19 and 4 CAPs could have been applied in various combinations until the rearranged arms were identified. For example, one of the combinations would be to apply a biotin-labeled 19p C A P with a digoxigenin-labeled 4q C A P . This technique, which has comparable hybridization intensity and specificity to WCPs, would have been particularly helpful in our analysis of BCCH-Sn , considering the non-optimal quality of the metaphases which did not permit good hybridization of the smaller DNA-specific probes that were used to identify the rearranged chromosomal arms. Once the participating chromosomal arms are determined, appropriate region-specific probes can be applied to further isolate the breakpoint region. Although this step is dependent on the availability of region-specific probes, now with the development of cosmid, Y A C , B A C , P A C , and P l libraries which contain defined chromosomal sequences of varying sizes and can be used as FISH probes, probe availability is no longer such an obstacle. In fact, as one of the goals of the human genome project is to develop integrated genetic and physical chromosomal maps which include ordered sets of cosmid clones (22), the cloning of genes involved in oncogenic rearrangements will be greatly facilitated. The 19p cosmid probes that were used for dual-coloured FISH in our study, for example, actually represent three cosmids that span the 19pl3.3 region of chromosome 19 (22). For the purposes of our initial analysis, these cosmids were applied to the tumour metaphase chromosomes simultaneously as a "19p cocktail probe", to determine whether the 19p arm was involved in any of the three rearranged chromosomes. Ultimately, however, if further characterization of 19p is required, individual application of these cosmids can provide much more detailed information about a breakpoint site and putative gene involvement. Although it was difficult to successfully hybridize the 19p 131 cosmids to the tumour metaphases, one derivative 19p chromosome was identified (Figure 24). Since the cosmid probes used to obtain this signal span from the 19pl3.3 telomere to the centromeric end of the 19pl3.3 band, it can be specified that this chromosome 19 derivative contains 19pl3.3 material. A n alternative strategy for accurately defining a translocation breakpoint or for characterizing any unidentified chromosomal region, involves the generation of region specific probes by chromosome microdissection. This is an increasingly important method for assessing chromosome rearrangements in many human cancers (3, 23, 24). This powerful technique combines the accuracy of chromosome microdissection with the specificity of FISH to result in the generation of region-specific probes and marker-specific chromosome paints (3). The development of such probes will be important for routine diagnostic testing and for monitoring patients during the course of disease. Briefly, metaphase spreads from the sample are prepared on glass slides and the chromosomal region of interest (e.g. the translocation breakpoint) is identified by banding. Next, the chromosomal material is microdissected using capillary pipette needles under a dissecting microscope. After the desired number of copies are dissected (10-20 are standard, although as few as 6 copies have been utilized), the D N A is amplified by the polymerase chain reaction (PCR) with degenerate dinucleotide primers. The PCR fragments are then fluorescently labeled with a secondary PCR reaction or by nick translation (3). This region-specific probe can then be hybridized to normal metaphase chromosomes to provide subregional information about the chromosomal regions involved in the breakpoint. The localization information obtained can be used to select cosmid clones or Y A C s that are labelled for FISH and contain chromosomal sequences from the isolated region. When FISH is performed on tumour metaphases, if the D N A -specific probe is not involved in the breakpoint region, then two hybridization 132 s ignals w i l l be seen, represent ing the t w o n o r m a l c h r o m o s o m a l reg ions . H o w e v e r , i f the p robe contains the D N A f o u n d at the t r ans loca t ion b reakpo in t , t h e n three s ignals w i l l be detected, one f r o m the n o r m a l c h r o m o s o m a l r e g i o n a n d t w o f r o m the rea r ranged D N A w h i c h has been sp l i t b y the t rans loca t ion a n d is n o w located o n t w o d e r i v a t i v e ch romosomes . A n o t h e r n e w technique pe rmi t s the d i rec t i den t i f i c a t i on of region-speci f ic genes, w h i l e e l i m i n a t i n g the need for c o s m i d or Y A C pools , b y c o m b i n i n g chromosome mic rod i s sec t i on a n d c D N A h y b r i d se lect ion (25) . Once the specif ic breakpoin ts of a t rans locat ion have been ass igned , pu ta t ive gene i n v o l v e m e n t can be inves t iga ted . O n e w o u l d first l o o k at cand ida te genes i n the b reakpo in t area that encode t ransc r ip t ion factors a n d other po ten t i a l pro to-oncogenes, s ince the p a t h o p h y s i o l o g i c outcome of these rearrangements are u s u a l l y r egu la ted b y t ranscr ip t iona l de regu la t ion or s t ruc tura l a l tera t ion of proto-oncogenes (20). T h i s is based o n the two p r i n c i p l e outcomes of t ranslocat ions: proto-oncogene ac t iva t ion a n d gene fusions. F i r s t ly , h i g h l y expressed genes such as the T -ce l l receptor ( T C R ) or i m m u n o g l o b u l i n genes m a y come to l ie near a pro to-oncogene , r e su l t i ng i n its ac t iva t ion (26). A t y p i c a l example w h e r e a t rans loca t ion induces p ro to-oncogene ac t iva t ion is i n B u r k i t t ' s l y m p h o m a . In this B - c e l l m a l i g n a n c y the t r ans loca t ion t(8;14)(q24;q32) juxtaposes i m m u n o g l o b u l i n h e a v y - c h a i n ( H ) genes to the c - M Y C oncogene, thus ac t iva t ing c - M Y C (27). A l t e r n a t i v e l y , a fus ion gene e n c o d i n g a ch imaer i c t r ansc r ip t ion factor is created w h e n the breaks o c c u r i n g w i t h i n genes o n each ch romosome i n c l u d e at least one gene e n c o d i n g a t r a n s c r i p t i o n factor (26) . It i s n o w apparent that the m a i n consequence of c h r o m o s o m a l t rans loca t ions i n s o l i d t u m o u r s m a y be the fo rma t ion of gene fusions e n c o d i n g c h i m a e r i c t r ansc r ip t ion factors that deregulate k e y target genes (26). The s t ruc tura l ou tcome of severa l s o l i d t u m o u r t rans locat ions a n d the i r r e su l t i ng t r a n s c r i p t i o n factors have 133 now been determined (Table 3, Chapter 1). The cloning of additional translocation breakpoints in the future is likely to uncover new genes and gene fusions of interest in M F H and other solid tumours. Since at the present stage of our investigation there is minimal information about the specific chromosomal regions that are involved in t(4;19), one can look at reports of translocations involving chromosome 4 and 19 in other neoplasms, for example, the t(4;19)(q35;ql3.1) translocations reported in other soft tissue tumours (9-11). Although no putative gene involvement was discussed in these translocations, they may provide us with valuable information regarding putative chromosome 4 and 19 breakpoints in M F H . It is possible that the chromosome 19 breakpoint in BCCH-Sn is the same 19ql3.1 breakpoint described in the above cases. As more tumours are studied, more reports of overlapping cytogenetic findings are arising, as seen in the above cases where identical translocations were observed in both a mesenchymal chondrosarcoma and an embryonal RMS (11). This information can provide insight regarding the location of genes involved in the genesis of these tumours. For example, presently, there are no known proto-oncogenes localized to 19ql3.1 (28). However, considering the increasing reports of its involvement in chromosomal translocations in various neoplasms (9, 10, 12, 13, 28-31), it is highly possible that a gene (or genes) in this region will be identified that contributes to oncogenesis. Previously reported M F H cytogenetic information can also be looked at in order to gain insight into candidate gene involvement. Based solely on the M F H cytogenetic information in the literature, the 19pl3 band is the most logical site to look for an altered gene. Not only do genetic alterations of 19pl3 in M F H result in a non-random 19p+ marker chromosome, but 19p+ is believed to be biologically and clinically significant (32, 33). To date, there have been no reports describing the 134 molecular analysis of potentially aberrant 19pl3 genes in M F H and no chimaeric genes involving 19p have been described in solid tumours. There are, however, several characteristic translocations in haematopoietic malignancies that involve 19pl3 (26), for example, t(l;19)(q23;pl3.3) in pre-B-cell acute lymphoblastic leukemias (ALL) (34). This translocation results in a fusion gene located on the der(19) chromosome, which consists of the 5' sequences of E2A at 19pl3 and the 3' sequences of PBX. The fusion gene consists of the transcription activation domain of the transcription factor E2A fused to the DNA-binding homeodomain of the transcription factor PBX (34), thus enabling target genes that bind the PBX homeodomain in the fusion protein to be activated and tumour development initiated. It is interesting that both the t(l;19) translocation in pre-B-cell A L L and the 19p+ marker chromosome in M F H have been associated with a poor prognosis (19, 33). Possibly, a similar E2A fusion product is formed with another gene in a subset of M F H with 19pl3 rearrangements. Another 19pl3 candidate gene, once believed to be "the best candidate for a pathogenetic change common to the frequently encountered 19p+ aberrations" in M F H , is the insulin receptor gene (INSR) (15) which maps to 19pl3.3 (22). The insulin receptor is a protein tyrosine kinase (PTK) receptor, similar to the epidermal growth factor (EGF) receptor and platelet-derived growth factor (PDGF) receptor (35). Activation of these growth factor receptors by extracellular ligands initiates the signal transduction cascade by phosphorylating tyrosine residues of target proteins, thereby activating various pathways important in signal transduction. Since the targets at the end of the pathway include nuclear transcription factors, which can significantly change the pattern of gene expression, constitutive activation of the kinase activity at any point in a signal transduction cascade (eg. at an altered P T K receptor) may result in an oncogenic phenotype. Rearrangement of the INSR gene 135 has been documented in some high grade ovarian tumours (36), and it has been suggested that 19p+ markers and 19pl3.3 breaks frequently described in ovarian tumours, might also be occurring in the INSR gene (36). Similarly, abnormalities of INSR may form the basis of 19p alterations in M F H . Although alterations to chromosome 19q in M F H are not reported as frequently as alterations to 19p, several 19q abnormalities and unspecified chromosome 19 derivatives have been documented (14-17, 37-39). These include two reports where unidentified additional material was detected at 19q, suggesting 19q translocations (14), and a derivative 19 interpreted to result from an unbalanced translocation to 19q (17). Other documented 19q alterations include ring chromosomes, r(19)(pl3ql3) (15, 31, 38) and telomeric associations of 19ql3 with chromosomes 9, 17, and 20 (31, 37). Considering that previous M F H cytogenetic reports have been based on chromosome banding methods, which left many unidentified marker chromosomes, it is highly possible that subtle chromosome 19 aberrations, including alterations to 19q and the t(4;19) translocation, were present but undetected. This was demonstrated in B C C H - S n , where prior to the use of FISH, no 19q abnormalities were reported. This is a particularly interesting finding, since it may suggest a more significant role for 19q aberrations in M F H than previously believed. Therefore, our data suggests that 19q aberrations may occur more frequently than currently reported and that the use of FISH to characterize complex M F H karyotypes in the future may likely reveal additional rearrangements to 19q and other chromosomes. The only specific site on 19q that chromosomal breaks have been reported in M F H is at band 19ql3.3 (16). A candidate gene which maps to 19ql3.3 is the proto-oncogene r-ras (28). The ras family of genes encode an important group of membrane-associated GTP-binding proteins which play a central role in 136 transmitting the signal from receptor tyrosine kinases in the signal transduction pathway (35). The Ras protein bound to G D P is inactive, while Ras carrying G T P is active and acts on its target molecule to stimulate cells into S phase of the cell cycle. Oncogenic Ras remains constitutively in the active GTP-bound form. Mutations to various ras oncogenes are estimated to occur in 15-30% of human tumours, and mutated Ras proteins promote constitutive activation of its downstream effector pathways (40). In regards to chromosome 4 abnormalities in M F H , chromosome 4 is one of the most frequently observed numerical gains in abnormal M F H karyotypes (14, 29, 37). However, to date, no structural alterations of chromosome 4 in M F H have been cited in the literature (14, 17, 30, 31, 37), which again may be due in part to the fact that subtle chromosome 4 rearrangements were previously undetectable without the use of FISH. Furthermore, there are very few reports of altered chromosome 4 genes in other human malignancies (9-13, 26, 41). Alterations at band 4q35 have * been described in five different solid tumours (12, 13, 16, 31, 39), in T-lymphoma a t(4;16)(q26;pl3) translocation is described which involves the IL-2 gene on chromosome 4q26 (26), and a non-random t(4;ll)(q21;q23) translocation is reported in pro-B-cell acute lymphoblastic leukemia (ALL). In the original B C C H - S n karyotype two chromosome 4 alterations were described, rea(4p) and rea(4)(pl4). As indicated by the white arrowheads in Figure 7 (Chapter 2), it was later proposed that these two aberrant 4 chromosomes were possibly involved in rearrangements with chromosome 19 material. After performing dual-coloured FISH, two t(4;19) translocations were identified, however, in addition to several extra pieces of rearranged chromosomal 4 material. Therefore, based on our data, it cannot conclusively be said that 4p is involved in t(4;19), although it is possible. Upon identification of one or more characteristic translocations in M F H , 137 specific diagnostic tests can be developed. As discussed previously, cosmid or Y A C clones that contain region-specific D N A from characteristic translocation breakpoints, can be used as a diagnostic tool for assessing putative M F H . With this technique, the specific rearrangement can be identified in tumour interphase nuclei or metaphase chromosomes by screening for split cosmids or Y A C s . Another diagnostic method involves a molecular technique called reverse transcriptase polymerase chain reaction (RT-PCR). If the t(4;19) or another translocation is found to encode a specific chimaeric gene product, detection of the resulting aberrant transcript by RT-PCR could form the basis of a sensitive and specific diagnostic test for M F H . This method is presently used for the detection of the EWS/FLI (42) and EWS/ERG (20) fusion transcripts in Ewing sarcoma (43), which result from t(ll;22) and t(21;22) translocations, respectively. Before concluding, it is important to address the issue of the validity of our results since the majority of the material used for FISH analysis was derived from long-term cultures of the tumour (BCCH-Sn). The concern with long-term cultures is that chromosome changes in tumour cells in vitro over long periods are often numerous and heterogeneous within a cell population. Although this can raise uncertainty as to whether the alterations seen are truly representative of the original tumour, this was not an issue in our study for the following reasons. It has been reported that structural changes seen in long-term cultures can be considered to be already present in vivo in the tumour when the same alteration was also observed in short-term cultures (44). Not only were the same structural chromosome 19 abnormalities detected in short- and long-term cultures using C O A T A S O M E 19, but originally these structural aberrations involving chromosome 19 were reported by cytogenetic analyses from short-term cultures (chromosome banding). Identical numerical chromosome 19 abnormalities were also consistently 138 detected in short- and long-term cultures with C O A T A S O M E 19 (i.e. three additional pieces of chromosome 19). Although the FISH findings were not entirely consistent with the initial karyotype which identified only two aberrant chromosomes 19, it is not surprising that a further small piece of rearranged chromosome 19 was unidentifiable, given the previously discussed inability of standard cytogenetic techniques to identify all complex and subtle alterations. In fact, this study emphasizes how useful FISH is for accurately analysing complex chromosomal rearrangements. Furthermore, our results correlate with cytogenetic reports of M F H which describe structural abnormalities involving chromosome 19 to occur non randomly in M F H (15, 31, 32). Therefore, it is concluded that the chromosome 19 abnormalities observed in the long-term cultures are representative of the original tumour and are not random alterations resulting from long-term growth in vitro. The potential that FISH has for enhancing the resolution of solid tumour cytogenetic analysis has been realized by the cytogeneticist and the pathologist alike. The development of disease-specific probes based on tumour-specific genetic alterations, especially for diagnostically difficult tumours, is revolutionizing the field of clinical oncology. Although additional analysis is required before we identify an M F H tumour-specific marker with diagnostic capabilities, the door has now been opened which may lead directly to new pertinent molecular analyses, providing further insight into the molecular characterization of chromosome 19 abnormalities in M F H . D. R E F E R E N C E S 1. Le Beau M M (1993): Detecting genetic changes in human tumour cells: have scientists "Gone Fishing?". Blood 81:1979-1983. 139 2. Verma RS, Babu A (1995): Human Chromosomes Principles and Techniques, ed. 2, McGraw-Hil l , New York. 3. Meltzer PS, Guan XY, Burgess A , Trent JM (1992): Rapid generation of region specific probes by chromosome microdissection and their application. Nat Genet 1:24-28. 4. Dal Cin P, Kools P, Sciot R, Wever ID, Van Damme B, V a n de Ven W, Van Den Berghe H (1993): Cytogenetic and fluorescence in situ hybridization investigation of ring chromosomes characterizing a specific pathologic sub-group of adipose tissue tumors. Cancer Genet Cytogenet 68:85-90. 5. Sullivan BA, Leana-Cox J, Schwartz S (1993): Clarification of subtle reciprocal rearrangements using fluorescence in situ hybridization. A m J Med Genet 47: 223-230. 6. Amler L C , Corvi R, Praml C, Savelyeva L , Le Paslier D , Schwab M (1995): A Reciprocal translocation (1;15) (36.2;q24) in a neuroblastoma cell line is accompanied by D N A duplication and may signal the site of a putative tumor suppressor-gene. Oncogen 10:1095-1101. 7. Mezzelani A , Sozzi G , Pierotti M A , Pilotti S (1996): Rapid differential diagnosis of myxoid liposarcoma by fluorescence in situ hybridization on cytological preparations. J Clin Pathol 49:M308-M309. 8. Therman Eeva, Susman M (1993): Human Chromosomes Structure, Behavior, and Effects, ed. 3, Springer-Veriag, New York. 9. Roberts P, Browne CF, Lewis IJ, Bailey G C , Spicer RD, Williams J, Batcup G (1992): 12ql3 abnormality in rhabdomyosarcoma - a nonrandom occurrence? Cancer Genet Cytogenet 66:43-46. 10. Urumov IJ, Manolova Y (1992): Cytogenetic analysis of an embryonal rhabdomyosarcoma cell line. Cancer Genet Cytogenet 61:214-215. 11. Richkind K E , Romansky SG, Finklestein JZ (1996): t(4;19)(q35;ql3.1): A recurrent change in primitive mesenchymal tumours? Cancer Genet Cytogenet 87:71-74. 12. Gladstone B, Parik P M , Balsara B, Kadam PR, Rao SR, Hair C N , Jambekar N A , Adavani S H (1993): Rhabdomyosarcoma: A cytogenetically interesting case report. Cancer Genet Cytogenet 66:43-46. 13. Hirabayashi Y, Yoshida M A , Ikeuchi T, Ishida T, Kojima T, Higaki S, Machinami R, Tonomura A (1992): Chromosomal rearrangement at 12ql3 in two cases of chondrosarcomas. Cancer Genet Cyotgenet 60:35-40. 140 14. Orndal C, Rydholm A , Willen H , Mitelman F, Mandahl N (1994): Cytogenetic intratumor heterogeneity in soft tissue tumors. Cancer Genet Cytogenet 78: 127-137. 15. Mandahl N , He im S, Willen H , Rydholm A , Eneroth M , Nilbert M , Kreicbergs A , Mitelman F (1989): Characteristic karyotypic anomalies identify subtypes of malignant fibrous hisitiocytoma. Genes Chromosom Cancer 1:9-14. 16. Sandberg A A , Bridge JA (1994): The cytogenetics of Bone and Soft Tissue Tumors. R.G. Landes Company, Texas. 17. Biegel JA, Perilongo G, Rorke LB, Parmiter A H , Emanuel BS (1992): Malignant fibrous histiocytoma of the brain in a six-year-old girl. Genes Chromosom Cancer 4:309-313. 18. Davis RJ, D'Crux C M , Lovell M A , Biegel JA, Barr F G (1994): Fusion of PAX7 to F K H R by the variant t(l;13)(p36;ql4) translocations in alveolar rhabdomyo-sarcoma. Cancer Res 54:2869-2872. 19. Rowley JD (1994): Cytogenetic and molecular analysis of pediatric neoplasms. Ped Pathol 14:167-176. 20. Sorensen PHB, Lessnick SL, Lopez-Terrada D , Liu XF, Triche TJ, Denny C T (1994): A second Ewing's sarcoma translocation, t(21;22), fuses the EWS gene to another ETS-family transcription factor, E R G . Nat Genet 6:146-151. 21. Guan XY, Zhang H , Bittner M , Jiang Y, Meltzer P, Trent J (1996): Chromosome arm painting probes. Nat Genet 12:10-11. 22. Brandiff BF, Gordon L A , Fertitta A , Olsen AS, Christensen M , Ashworth L K , Nelson D O , Carrano A V , Mohrenweiser H W (1994): Human chromosome 19p: A fluorescence in situ hybridization map with genomic distance estimates for 79 intervals spanning 20 Mb. Genomics 23:582-591. 23. Guan XY, Trent JM, Meltzer PS (1993): Generation of band-specific painting probes from a single microdissected chromosome. H u m Molec Genet 2:1117-1121. 24. Jonveaux P, Le Coniat M , Derre J, Flexor M A , Daniel M T , Berger R (1996): Chromosome microdissection in leukemia: a powerful tool for the analysis of complex chromosomal rearrangements. Genes Chromosom Cancer 15:26-33. 25. Su Y A , Trent JM, Guan XY, Meltzer PS (1994): Direct isolation of genes encoded within a homogeneously staining region by chromosome micro-dissection. Proc Natl Acad Sci U S A 91:9121-9125. 141 26. Rabbitts T H (1994): Chromosomal translocations in human cancer. Nature 372:143-149. 27. Dalla-Favera R, Bregni M , Erikson J, Patterson D, GalloRC, Croce C M (1982): Human c-myc oncogene is located on the region of chromosome 8 that is translocated in Burkitt lymphoma cells. Proc natl Acad Sci U S A 79:7824. 28. Ashworth L K , Batzer M A , Brandriff F, Branscomb E, de Jong P, Garcia E, Games JA, Gordon L A , Lamerdin JE, Lennon G, Mohremveiser H , Olsen AS, Slezak T, Carrano A V (1995): A n integrated metric physical ap of human chromosome 19. Nat Genet 11:422-427. 29. Bridge JA, Sanger W G , Shaffer B, Neff JR (1987): Cytogenetic findings in malignant fibrous histiocytoma. Cancer Genet Cytogenet 29:97-102. 30. Aspberg F, Mertens F, Bauer H C F , Lindholm J, Mitelman F, Mandahl N (1995): Near-haploidy in two malignant fibrous histiocytomas. Cancer Genet Cytogenet 79:119-122. 31. Szymanska J, Tarkkanen M , Wiklund T, Virolainen M , Blomqvist C , Asko-Selijavaara S, Tukiainen E, Elomaa I, Knuutila S (1995): A cytogenetic study of malignant fibrous histiocytoma. 85:91-96. 32. Rydholm A , Mandahl N , Heim S, Kreicbergs A , Willen H , Mitelman F (1990): Malignant fibrous histiocytomas with a 19p+ marker chromosome have increased relapse rate. Genes Chromosom Cancer 2:296-299. 33. Choong P F M , Mandahl N , Mertens F, Willen H , Alvegard T, Kreicbergs A , Mitelman F, Rydholm A (1996): 19p+ marker chromosome correlates with relapse in malignant fibrous histiocytoma. Genes Chromosom Cancer 16: 88-93. 34. Hunger SP, Galili N , Carroll AJ, Crist W M , Link MP, Cleary M L (1991): The t(l;19)(q23;pl3) results in consistent fusion of E2A and PBX1 coding sequences in acute lyphoblastic leukemias. Blood 77:687-693. 35. Lewin B (1994): Oncogenes: gene expression and cancer. In: Genes V , Oxford University Press, New York. 36. A m f o K, Neyns B, Teugels E , Lissens W, Bourgain C, De Sutter P, Vandamme B, Vamos E , De Greve J (1995): Frequent deletion of chromosome 19 and a rare rearrangement of 19pl3.3 involving the insulin receptor gene in human ovarian cancer. Oncogene 11:351-358. 142 37. Mandahl N , He im S, Kristoffersson U , Mitelman F, Rooser B, Rydholm A , Willen H (1985): Telomeric association in a malignant fibrous histiocytoma. H u m Genet 71:321-324. 38. Mandahl N , He im S, Areden K, Rydholm A , Willen H , Mitelman F (1988): Rings, dicentrics and telomeric association in histiocytomas. Cancer Genet Cytogenet 30:23-33. 39. Sreekantaiah C, Rao U N M , Karakousis CP, Sandberg A A (1992): Cytogenetic findings in a malignant fibrous histiocytoma of the gallbladder. Cancer Genet Cytogenet 59:30-34. 40. Bohle R M , Brettreich S, Repp R, Borkhardt A , Kosmehl H , Altmannsberger H M (1996): Single somatic ras gene point mutation in soft tissue malignant fibrous histiocytomas. A m J Path 148:731-738. 41. Morrissey J, Tkachuk D C , Milatovich A , Francke U , Link M , Cleary M L (1993): A serine/proline-rich protein is fused to H R X in t(4;ll) acute leukemias. Blood 81:1124. 42. Delattre O, Zucman-Plougastel B, Desmaze C, Melot T, Peter M , Kovar H , Joubert I, de Jong P, Rouleau G (1992): Gene fusion with an E T S - D N A -binding domain caused by chromosome translocation in human tumors. Nature 359:162-165. 43. Sorensen PHB, L iu XF, Delattre O, Rowland JM, biggs C A , Thomas G , Triche TJ (1993): Reverse transcriptase PCR amplification of EWS/FLI-1 fusion transcripts as a diagnostic test for peripheral primitive neuroectodermal tumors of childhood. Diag Mol Pathol 2:147-157. 44. Mugneret F, Lizard S, Aurias A , Turc-Carel C (1988): Chromosomes in Ewing's sarcoma. II. Nonrandom Additional Changes, trisomy 8 and der(16)t(l:16). Cancer Genet Cytogenet 32:239-245. 143 CHAPTER VI S U M M A R Y A N D C O N C L U S I O N S In the broadest sense, cancer is a result of genetic mutations. Cancer cells are unique from their nonmalignant counterparts because within their D N A they harbour gene amplifications, point mutations, deletions, and gene rearrangements, each of which can alter normal cellular growth. Such mutations can now be characterized by current methods of D N A analysis. These changes result in chromosomal alterations, some of which can only be appreciated by analysis of the fine structure of the chromosome by molecular means, and others that generate morphologically unique structures that can be identified cytogenetically. Regardless, it is now clear that certain recurring chromosomal aberrations lead to specific genetic aberrations that are unique for different neoplasms. Subsequent identification and characterization of these aberrant chromosomes and altered genes can provide novel diagnostic and prognostic information, as well as basic insights into the processes of oncogenesis. Furthermore, unsuspected similarities between tumour types may arise, as well as the identification and characterization of new entities. The identification of tumour-specific markers will be particularly useful for difficult to diagnose tumours such as M F H , which currently has ill-defined diagnostic criteria. However, despite being the most common sarcoma in adulthood, M F H is not yet well characterized at the cytogenetic or molecular levels. Therefore, invaluable knowledge can be gained through the genetic characterization of M F H . Furthermore, the lack of distinguishing features in M F H have made it a difficult diagnosis in pediatric cases where it occurs much less frequently and its 144 existence as an entity is often controversial. Cytogenetic and molecular genetic methods have therefore been used to characterize a pediatric M F H , to compare the pediatric and adult forms and to further characterize abnormalities which may prove to be unique to M F H . A . A M P L I F I C A T I O N O F M U L T I P L E G E N E S F R O M C H R O M O S O M A L REGION 12ql3-14 Cytogenetic analysis of a pediatric M F H (BCCH-Sn) revealed homogeneously staining regions (HSRs) and double minute chromosomes (dmins), cytogenetic indices of gene amplification. Therefore, the case was screened for amplification of genes localized to chromosomal bands 12ql3-14, including the putative proto-oncogenes, MDM2, CDK4, SAS, CHOP, and GLL which are frequently amplified and overexpressed in adult M F H . Southern and Northern blot analysis confirmed co-amplification of MDM2, CDK4, SAS, and CHOP. The frequency of involvement and co-amplification of M D M 2 , CDK4, and SAS in human sarcomas, strongly suggests that their overexpression might contribute to tumour development and progression (1-3). Such co-amplification studies of the 12ql3-14 amplicon have not been previously reported in pediatric M F H . These results suggest that the co-amplification of genes in the 12ql3-14 regions is similar in pediatric and adult M F H . C H R O M O S O M E 19 M I C R O S A T E L L I T E ANALYSIS R E V E A L S N O L O H B. Cytogenetic studies of M F H in adults have revealed recurring abnormalities of chromosomal band 19pl3 (4), which have been correlated with an increased relapse rate (5,6). To date, however, the genes affected by these alterations remain 145 unknown. Since chromosome 19p abnormalities were also identified in the pediatric M F H (BCCH-Sn), the possibility that chromosome 19 abnormalities in M F H represent a deletion of material from chromosome 19, possibly leading to loss of a tumour suppressor gene, was investigated. Chromosome 19 allelic loss was therefore studied in a pediatric M F H (BCCH-Sn). Using microsatellite PCR analysis, 18 polymorphic markers spanning chromosome 19 were screened for L O H . None of the 14 informative markers (9 from 19p; 5 from 19q) revealed L O H in the tumour D N A . This data indicates that there has not been a gross deletion of material from the regions tested; however, it remains possible that a point mutation or submicroscopic deletion exists that causes the inactivation of a tumour suppressor gene. A candidate gene potentially altered by the recurring 19pl3 alterations in M F H is INK4d, a C D K inihibtor which maps to 19pl3 (7). Southern blot analysis revealed no large deletions or genomic rearrangements of INK4d in B C C H - S n , and Northern blot analysis revealed similar expression levels of B C C H - S n to other pediatric sarcomas without 19pl3 abnormalities. C. FISH IDENTIFIES T W O t(4:19) D E R I V A T I V E C H R O M O S O M E S The chromosome 19 abnormalities in B C C H - S n were originally identified by standard cytogenetic banding methods; however, due to the complexity of the alterations and the limitations of banding techniques, the aberrant chromosomes could not be fully defined. Therefore, FISH, which has proven to be a powerful tool for characterizing complex and subtle chromosomal alterations (8), was chosen as a method to further characterize these chromosome 19 aberrations. Whole chromosome FISH consistently revealed five chromosome 19 signals, representing three chromosome 19 derivatives and two normal copies of chromosome 19. Dual-146 coloured FISH with 19p and 19q unique probes indicated that two of the 19 derivatives contained 19ql3.1 material and the third contained 19pl3.3 material. Finally, dual-coloured whole chromosome FISH identified chromosome 4 as the reciprocal chromosome in two of the three derivative 19 chromosomes, i.e. two t(4;19) derivative chromosomes were present in the case. Although the chromosome 19 breakpoint has not been determined, based on our data it is hypothesized that t(4;19) contains a breakpoint that lies between the proximal region of 19ql3.1 and the centromere. Furthermore, although this is the first report of a t(4;19) in an M F H , there are other soft tissue tumours reported to have translocations involving chromosome 4 and 19q (9-11). It is also hypothesized that the 19q chromosomal material is involved in the translocation with chromosome 4 rather than der(19p). Since two 19q derivatives and one 19p derivative were identified in B C C H - S n , at least one of the translocations must contain 19q material. A n d because it is more likely that this 19q translocation underwent duplication rather than a second translocation occurring with 19p and chromosome 4, it is proposed that both t(4;19) translocations involve 19q material and originated from a single rearrangement with chromosomes 4 and 19. D. G E N E R A L C O M M E N T S The presented cytogenetic and molecular genetic findings in a pediatric M F H are significant to our overall understanding and classification of M F H . Firstly, cytogenetic and molecular genetic similarities have been demonstrated which provide further evidence for a nosologic link between pediatric and adult forms of M F H . With recent advances in genetic analysis, it is likely that more pediatric cases of M F H will be evaluated on a genetic level in the future, thereby, revealing 147 additional information regarding childhood forms of M F H . Ideally, this will lead to more accurate diagnoses of pediatric M F H and the ascertainment of its true incidence in this age group. With the identification of t(4;19) derivatives and 12ql3-14 amplification in pediatric M F H , this work contributes to the characterization of chromosome 19 aberrations in M F H and the pursuit of an MFH-specific marker. However, in order to identify a tumour-specific marker for M F H , considerably more cases must be evaluated at both the cytogenetic and molecular levels. When truly homogeneous groups of patients can be compared based on their genetic abnormalities, improvements in therapy and patient outcome should become apparent. Ultimately, the ability to identify genetic alterations at the molecular level will revolutionize the clinical evaluation of patients with cancer. E. R E F E R E N C E S 1. Oliner JD, Kinzler K W , Meltzer PS, George D L , Vogelstein B (1992): Amplification of a gene encoding a p53-associated protein in human sarcomas. Nature 358:80-83. 2. Ladanyi M , Cha C, Lewis R, Jhanwar SC, Huvos A G , Healey J H (1993): MDM2 gene amplification in metastatic osteosarcoma. Cancer Res 53:16-18. 3. Schmidt E E , Ichimura K, Reifenberger G, Collins P (1994): CDKN2 (p!6/MTSl) gene deletion or CDK4 amplification occurs in the majority of glioblastomas. Cancer Res 54:6321-6324. 4. Mandahl N , He im S, Willen H , Rydholm A , Eneroth M , Nilbert M , Kreicbergs A , Mitelman F (1989): Characteristic karyotypic anomalies identify subtypes of malignant fibrous hisitiocytoma. Genes Chromosom Cancer 1:9-14. 5. Rydholm A , Mandahl N , Heim S, Kreicbergs A , Willen H , Mitelman F (1990): Malignant fibrous histiocytomas with a 19p+ marker chromosome have increased relapse rate. Genes Chromosom Cancer 2:296-299. 148 6. Choong P F M , Mandahl N , Mertens F, Willen PL Alvegard T, Kreicbergs A , Mitelman F, Rydholm A (1996): 19p+ marker chromosome correlates with relapse in malignant fibrous histiocytoma. Genes Chromosom Cancer 16: 88-93. 7. Okuda T, Hirai H , Valentine V A , Shurtleff SA, Kidd VJ, Lahit JM, Sherr CJ, Downing JR (1995): Molecular cloning, expression pattern, and chromosomal localization of human C D K N 2 D /INK4d, an inhibitor of cyclin D-dependent kinases. Genomics 29:623-630. 8. Sullivan B A , Leana-Cox J, Schwartz S (1993): Clarification of subtle reciprocal rearrangements using fluorescence in situ hybridization. A m J Med Genet 47: 223-230. 9. Roberts P, Browne CF, Lewis IJ, Bailey G C , Spicer RD, Williams J, Batcup G (1992): 12ql3 abnormality in rhabdomyosarcoma - a nonrandom occurrence? Cancer Genet Cytogenet 66:43-46. 10. Urumov IJ, Manolova Y (1992): Cytogenetic analysis of an embryonal rhabdomyosarcoma cell line. Cancer Genet Cytogenet 61:214-215. 11. Richkind K E , Romansky SG, Finklestein JZ (1996): t(4;19)(q35;ql3.1): A recurrent change in primitive mesenchymal tumours? Cancer Genet Cytogenet 87:71-74. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

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"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0099200/manifest

Comment

Related Items