MW” n! . . 7 ' (9H ‘ ‘7 I}! {1:37, .; 41?,fifg’é'xfiiégifiix 5 217,-, “4?”!l-I’fifihgfr‘ié’mi’."fi‘r hififwf.f . 1" if", 5’ ' ’5’ 555432? / 51515515555552: ~55; "1;? ..l‘ 1:7],- 4"! '37:?! is"; Vii-(flfu, £551.. '- ‘j/ 4‘1? 2:2 large ‘sz @V- :{J if}, 2/1 19"!" 4.14 fi/" )4)!” ‘v I 'f ‘1‘?! f, at." 3;, 3.22914“ if L". i ‘i’l‘ay Ff?! l 4: r I? to ,t ,w I‘A‘H‘V'J {‘1’ . r 6:";va ‘1” ‘55 - ”‘1’; I; 74:55] 'Vl'fjs-v'vv' J' wifh‘u‘ ‘ ‘7 . 1‘), If) ($5.? .451. 13‘" n‘.~.",«'.-‘5.'-- .5; '11: M5. i:,"('1l "H ““‘V.~I.‘-. 4 V’Jyfiwu. “Iv-.443: . ’4 if; v l 5”; v 5:1;4'“ 15's.??? ~‘..., ’.Y'qu.y/{,- .5"? J5}: .{ 1‘, i: 7- ' 0V1:l "f— . ‘ “ ”Eiréfl ‘1‘?’ I5;- y_«1‘.l Q» .'. v. ) 'I h 4 ' ,q- .r «way If ‘ 1.". w" 4’ 1"“ 4. “4‘” Egg ‘54""4' ., , I‘QW" ‘ ‘ ‘ ’4 1.1" .351, .,‘ V ' i“ ”'2"; ‘ {\"wq, 5 -- 55.5... _ ' 4 NJ.‘ ' 3. “46 33353-1}. I ’ J “Va-:1? v ‘I “M54.“ 1'. 'l ‘, 1 l“ '4 1-150}. .I . In.” . ‘4’ N‘- s 24/ ' :N' ‘Vr . - I3}, .1" {Th 5 [113,5 "3.2/1 4 ' :45; , fI'I“ ’_ 4‘}. 1-,.“- 1 "‘0':1”““21'lwl‘1’r/«v- ] . J“ I) ‘ 15%.? 144:. 4-, r... ' . .‘l‘l.’ ”Jay”; 12-:- (441;)? ‘ - I‘ l".' .‘.‘ J 4 .- . . .‘ 1:... , ' . ‘ ‘ 4'” “I‘ ‘ l I'LNU’Hr‘I‘c‘H-H' :" " '7 A 511’5‘: [4“«1‘ 1A! .4' .4M' INH’? 1- 4143,“... ‘ 4l‘v".1°‘"ln 5 3: 4‘ 1/4544!- i-I4I’(t.‘\,‘/~ w4\.1‘lo 7....4v4., ‘1‘: (4‘)” ’ ”C‘flu‘lfi, V ,4‘ “I u 1‘ In." "4}! V 4M“; “‘4'?” all?!) I" <1 II'J: :l“ £69.)” “‘41 (”‘31: , ”I 10129;"! I. ". 1 "VI‘ #4:]...‘5. A "225‘” {1‘ "' ’ cu 'g'lllz‘. ’thi I 411.71% $4471th 'IJ‘I 1‘ 1‘.‘ 1/ P1 ‘0 4 15241.44 .1. ' ““l' ”’1'". j! ‘14 ll. '74.”! I', J; 4 "(3‘ ‘11:: ‘A’(' 4‘1, < :;£-' .911. My“ .2355 fa. .55 «, 1;. «151... fit. {4' M ‘ ifé‘%:¢m:u.‘ 4/’4$:;4..,444‘,4 ' ‘5 1w. . . ‘1‘. 4 y" ’6‘... :1}; i 51 ‘j‘fl‘l‘f‘: .. A 5.51:5. 1 MI} 4 rig-U)! J 1:12.4‘ I: . ' i . “hi—”$45” ' 9 ",4 J I r/("4’;%"} V ' "24;? ’1' (3:4 37-5. "’35.; ”in“ ,2/‘2 fig, n” 3’ (fab if: Idle va 1.. :gyjea 5-5 -- .1 ~47, (’1 ”g? ‘11, 3‘4ij X31“ ‘ .11)“ V I'l- c'\(' 4 ftgff .9525"? 31‘?” 3125:? ‘5’ m 7%.”; .1};er l:.’ 1‘7.‘ I? ‘/ ‘ 1‘ \--’.£:‘51 J'Iqfl.cg?$c§51'rw "114.85.251. {11.2" i 1%.“...d‘ési. Vlfli‘lfl 6:4?! .22: «533:5 1‘2, ‘14:: o 4 5‘3: 1: V'W‘ :35)“.- v. «j. .4. Y. «'-‘ 1 V l I I” '«V 1 [v- at.” ,-‘ 1.1 ( 11’)"; £1; ”1".) -I' 1’ 1.31 .11 451,5 £21? 29 - . 33:5?! 5% I “52,. ., .M «1152 "V “' '4‘ ;,1V'4;."., VT“! "I 11 a a ‘1‘! (‘fll‘fil‘dgt'fldV 4‘ ‘H '4‘ :fi 3* .. ‘1 1x3?" 4/1")? J . . 5 1". m! at Y) 1‘15... 4"};"33; 5";434'3’233‘ «#5 ‘ ’5 ‘ . . 'l‘ 5.» $8.5“: ”II/"1‘ ‘n V: :L .4; .» ‘ ix {‘141‘fi4‘. 111.4"’d U 4““, {51 5.5. :1‘1 I ’13. :6}; I.“ 64' 4' V ' “11,. 4.:‘V‘; I( :14; 61:] I! I 4 233(- .31.”. 7.3:“ v...‘ .\ (In, I?“ 1;? ”Silt...- "I.“ 5 31‘ Q 3/ 3" f! 55.5% I. «V V? ’7' ‘Je‘u v.1 , . ,1 .}. $5.555 . ”."=.?.«;» ~ 551$: 7:3": " V . N)‘ ‘41.? 15} «I! IV ‘ '44:]; .‘ IIO'I "7’91”“ . W 11:22:“.- (”4:4 19:, f1}, 5'? .t?‘ ‘6; I 5. ., 5325’ {’59:} ”tag-3!, '1‘ (34-45%; .9 4' I t 414/ V . - ( 5‘4“ :‘ Jump (“V 19.15145 12*.- M‘." if": -' 11.5; ,V~. ""MW: '1‘";- .‘ I'. ‘ it: II J! h 1- ". A??? j}; . ‘ 'tl‘q ”i4 . . .‘1’3 'u g ‘5. ”31’? 3.5V; ." 4 '2‘3 ' .. wt‘j. .3": :i...“(- "a f.‘ "1;, .C p I . 1,. 5" ‘ (KT “filia’j. 11’.» V“~ ‘31. N ‘57." £3.57 '1‘” 4' 51.3515 5 1’4" 1"“ 1f“?- 5; 1‘ A .5”; 4.44 , 1 L . '1'. 4 . 0“", y‘r?fl.f .. . “12%;:- ”1:541:11: '55:..1/135(,..f, 9/4} .1; . ‘ ‘ 1:9]? ‘,4 mega; (5‘92, [‘5‘]. 6‘.” 1‘<:.‘ '.""‘:‘/ IV"; ,1 '0“! 1. 'r‘j/{IIJI4./f. 'p"“$ .. . ‘ 1'1??? {Afr/3.1.7 5;“: -u1‘*“|"‘.'n "v (:V' x?4 " W5‘VI;‘;; («‘5 75 in) Afr" J! V ‘23:”: :fi’x. l (:41; 4; {It‘ll-k “'4 4:); . l 4. ‘ t!“1‘j‘:‘1/4'“}; / a, :1} 4 “f A" y. 45,..- . ,,.;.; 41,5 f ’5’5334‘5: :. . {211:}: 4/ - “1,7141 . . 2/?!" - u.- 1'11": QLIEDB 3£>}L:L ‘ ml" MICHIGAN STATE UNIVERSITY Ll lllll!"“lilliifliflllll E t l H 1; ill“: 3 1293 00540 5703 LIBRARY Michigan State University ll! This is to certify that the dissertation entitled BIOCHEMICAL AND BIOLOGICAL ANALYSIS OF HUMAN FIBROBLASTS TRANSFORMED WITH N-ras QNGQGBNES DANIEL MICHAEL WILSON has been accepted towards fulfillment of the requirements for A Ph. D . degree in Biochemistry/ Environmental Toxicology (z 1 Major professor Date Mb‘DL/t {0ng MS U is an Affirmative Action/Equal Opportunity Institution 0-12771 MSU LlBRARlES m RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. BIOCHEMICAL AND BIOLOGICAL ANALYSIS OF HUMAN FIBROBLASTS TRANSFORMED WITH N-BAS ONCOGENES by Daniel Michael Wilson A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biochemistry and Center for Environmental Toxicology 1988 ABSTRACT BIOCHEMICAL AND BIOLOGICAL ANALYSIS OF HUMAN FIBROBLASTS TRANSFORNED HITH N-BAS ONCOGENES by Daniel H. Hilson N-Lag, a member of the La; gene family, has been isolated as a transforming gene from many human tumors, including fibrosarcomas. To examine what role expression of an N-rg; oncogene plays in the etiology of human tumors, I transfected plasmids containing N-ng oncogenes into early passage diploid human fibroblast and into an infinite lifespan human fibroblast cell line, MSU-l. Plasmid pSV N- 33; contains the n_e_g gene, coding for Geneticin resistance, and a human leukemia cell line-derived N-rgg oncogene activated by a transversion in codon 12. The cellular promoter for N-ras was replaced by a viral long terminal repeat which initiates transcription of both the N-rgs gene and the neg gene. Plasmid pNR-MGl contains the ggg gene and a human fibrosarcoma-derived N-rgg oncogene which is activated by a mutation in codon 61. Transcription of the N-ng oncogene is initiated from its cellular promoter, and transcription of the geg gene is initiated from a simian virus-40 promoter. Transfection of pSV N-y;a_§ into diploid human fibroblasts yielded Geneticin resistant colonies, at a frequency of l x 10'6 cells transfected per 10 ug of plasmid DNA, of which 70% were morphologically transformed. In the absence of drug selection, dense multilayered groups of cells, foci, were formed at approximately the same frequency. The transformed cells had highly anaplastic morphologies, with many multinucleated and irregularly shaped cells, but the majority either reverted to a normal morphology or senesced prematurely. However, progeny cells from two of the Geneticin resistant colonies maintained their transformed morphology for the duration of a normal lifespan in culture. Immunoprecipitation analysis indicated that high amounts of N-rgg oncogene encoded protein were produced. These cell lines also exhibited anchorage independence, a characteristic of many tumor- derived cells. However, they did not form tumors when injected into athymic mice. Transfection of early passage diploid human fibroblasts with pNR- MGl yielded many Geneticin resistant colonies. However it was not clear that any of these were morphologically transformed. In the absence of drug selection, foci were observed, but their morphologies were not markedly abnormal. Cells from these foci yielded populations which soon reverted to a normal morphology, exhibited normal growth characteristics, and were not tumorigenic. Immunoprecipitation analysis indicated that the N-ng oncogene encoded proteins were not expressed. Transfection of pSV N-rgg into infinite lifespan MSU-l cells yielded foci of' morphologically transformed cells which maintained their' morphologies, were anchorage independent, growth factor independent, and malignant. DNA-DNA hybridization analysis indicated that the transfected oncogene was present in the tumor-derived cells, and inlnunoprecipitation analysis showed that high amounts of N-Lag oncogene encoded protein were present in all of the focus-derived cell lines, and in all but one of the tumor-derived cell lines. To my parents, David and Patricia Wilson; and my wife, Louise Hemond-Hilson ii ACKNOHLEDGHENT I would like to thank first and foremost Dr. Justin McCormick, my graduate thesis advisor, for the countless hours he devoted towards directing my research. In addition I owe a special appreciation to Dr. Veronica Maher for serving on my thesis committee and for all of her time and effort. I wish to thank Drs. Clifford Nelsch, Zachary Burton, and Robert Hausinger, members of my graduate thesis committee, for their advice and invaluable time. I also wish to express gratitude to Dr. Dennis Fry for his helpful advice on many technical matters, and Lonnie Milan and Stephen Dietrich for their excellent technical assistance and the sacrifice of their free-time. In addition I would like to thank Karen Hawes and Bob Kronick for technical assistance. I also owe thanks to numerous members of the Carcinogenesis laboratory whose encouragement and friendship supported me during my time here. They include Jia Ling Yang, Peter Hurlin, Tony Fox, Dajun Yang, John Dilberger, Helen Palmer, Bob Schilz, Toshiyuki Noriumura, Kenji Sato, Nitae Bhattacharyya, Tohru Tsujimura, Carol Patterson, and Kay Robinson. TABLE OF CONTENTS Page LIST OF TABLES .................................................... vii LIST OF FIGURES ................................................... viii ABBREVIATIONS ..................................................... x INTRODUCTION ...................................................... 1 CHAPTER 1. LITERATURE REVIEW A. Studies implicating environmental agents in carcinogenesis 8 1. Cancer frequencies and epidemiology ................... 8 2. Treatment of laboratory animals with suspected carcinogenic agents ................................. 10 3. Treatment of mammalian cells in culture with suspected carcinogens ............................... 12 4. In vitro transformation of human cells ................ l4 5. Relationship between mutation and cancer .............. 16 B. Involvement of the 3;; genes in carcinogenesis ............ 16 1. Evolution of RNA tumor virus :9; genes and identification of homologous cellular genes ......... 16 2. Isolation and molecular characterization of cellular [gs genes ............................... 18 3. Mechanisms of activation of cellular ng genes ........ 23 4. Studies of multistepped carcinogenesis by transfection or infection of Lg; oncogenes into mammalian cells.. 24 5. Transformation of human cells in culture by Egg oncogenes ....................................... 27 6. Activation of fig; genes by treatment with known carcinogens ................................... 29 C. Biochemical properties of Lg; proteins .................... 32 1. Structure and biochemical properties of the {Ag genes and their products ........................ 32 2. The role of [3; proteins in transmembrane signaling... 37 References ................................................ 41 iv CHAPTER CHAPTER II STABLE TRANSFORMATION OF DIPLOID HUMAN FIBROBLASTS FIBROBLASTS BY A TRANSFECTED N-RAS ONCOGENE ........... 56 Abstract .................................................. 57 Introduction .............................................. 59 Materials and Methods Cells and Culture medium ................................ 61 Plasmids ................................................ 61 DNA transfection and Geneticin selection ................ 61 Focus assay ............................................. 62 Anchorage independence .................................. 62 Lifespan analysis and assay for tumorigenicity .......... 62 Immunoprecipitation of p21 [3; .......................... 63 Results Morphological transformation by N-rgg ................... 65 Transformation of human fibroblasts to focus formation.. 68 Biological characterization of the cells from morphologically transformed colonies .................. 71 Evidence of expression of N-ng oncogene ................ 72 Other parameters related to transformation .............. 75 Discussion ................................................ 78 Acknowledgements .......................................... 82 References ................................................ 83 III MALIGNANT TRANSFORMATION OF INFINITE LIFESPAN, HUMAN FIBROBLAST CELL LINE MSU-I BY A TRANSFECTED N-RAS ONCOGENE ............................................. 88 Abstract .................................................. 89 Introduction .............................................. 90 Materials and Methods Cells ................................................... 92 Growth factors .......................................... 92 Culture medium .......................................... 92 Plasmids ................................................ 93 DNA transfection and selection for focus formation ...... 93 Anchorage independence .................................. 94 Assay for tumorigenicity ................................ 94 Cytogenetic analysis .................................... 94 DNA hybridization analysis .............................. 95 Immunoprecipitation of p21 rgg, .......................... 96 Results Transformation of MSU-l cells to focus formation by transfection of an N-ggg oncogene ..................... 98 Anchorage independence of MSU-l-N-rgg cell strains ...... 98 Growth factor requirements of MSU-l-N-[gg cells ......... 102 Tumorigenicity of the MSU-l-N-ggg transformed cell strains .......................................... 107 Cytogenetic analysis .................................... 111 Expression of the N-rgg oncogene ........................ 112 Evidence of the presence of N-rgg oncogene DNA sequences ......................................... 115 V Discussion ................................................ 119 Acknowledgements .......................................... 123 References ................................................ 124 APPENDICIES Appendix A. Plasmids used in transfection experiments. Appendix 8. Discussion of controls used in transfection experiments. vi LIST OF TABLES Table Page Chapter III I. Anchorage independence of focus-derived MSU-l-N-ng cell strains .................................................. 101 2. Tumorigenicity of focus-derived MSU-l-N:Lg§ cell strains ...... 108 vii LIST OF FIGURES Figure page Chapter II 1. Morphology of normal fibroblasts (A) and pSV N-ng transformed fibroblasts (b) ................................... 67 2. Foci of human fibroblasts following transfection with pSV N-rgg ..................................................... 70 3. Reconstruction assay for focus formation ...................... 74 4. Immunoprecipitation evidence of N-rgg oncogene expression ..... 77 Chapter III I. Photomicrograph of crystal violet stained human fibroblasts. (A) Focus induced by transfection of infinite lifespan MSU-l cells with pSV N-ng. (B) MSU-l cells ........................ 100 Growth factor independence of N-[gg-transformed cells ......... 104 Cell number of MSU-l cells and N—rgg-transformed cell strains as a function of Ca++ concentration and response to growth factors ............................................. 106 Photomicrograph of formalin fixed tissue sections of subcutaneous tumors formed by N-rgs-transformed cell strains ....................................................... 110 Immunoprecipitation evidence of N-ng oncogene expression ..... 114 Southern hybridization analysis for N-ggg specific gene fragments in N-rgg transformed human fibroblasts .............. 117 viii c-H-rgs DAG EF-Tu EGF EJ-H-rgg ElA G proteins H-ras 1P3 K-ras LTR MSV N-ras p53 p21 PIP2 PDGF PKC ABBREVIAIIONS cellular Harvey 3;; gene diacylglycerol elongation factor-Tu epidermal growth factor EJ bladder carcinoma-derived Harvey ras oncogene adenovirus early region 1A gene guanine nucleotide binding proteins :1; gene homologous to transforming gene of Harvey murine sarcoma virus inositol-1,2,3-triphosphate 3;; gene homologous to transforming gene of Kirsten murine sarcoma virus long terminal repeat murine sarcoma virus gene homologous to transforming gene of avian myelocytomatosis virus cellular Lg; gene first isolated from neuroblastoma common tumor antigen of 53,000 daltons 21,000 dalton Lg; protein phosphatidyl inositol, 4-5-bisphosphate platelet-derived growth factor protein kinase-C ix SHE SV40 T24-H-rgg v-mxs v-H-rgs v-K-ng Syrian hamster embryo simian virus 40 T24-bladder carcinoma-derived H—gg; oncogene myg oncogene of avian myelocytomatosis viru £1; gene from Harvey murine sarcoma virus [as gene from Kirsten murine sarcoma virus INTRODUCTION Chemical and physical agents have for years been implicated in the etiology of many cancers, and in most cases these agents have induced mutations and in vitro transformation of mammalian cells with similar kinetics (Barrett and Ts’o, I978, Chan and Little, 1978, Landolph, 1985), and have interacted with DNA (Miller, 1978). However, there were few clues as to the nature of the specific genes or regions of DNA which, when altered via mutation, were responsible for conferring transformed properties on normal cells. The advent of many new techniques in molecular biology during the past two decades has dramatically enhanced research on the mechanisms by which cancer occurs in man. These techniques include the ability to isolate transforming genes from tumor-derived cells, to clone individual genes into various vectors, to sequence genes, to dissect DNA with restriction endonucleases, to probe DNA for the presence of specific genes by DNA-DNA hybridization, and to analyze the expression of individual genes by DNA-RNA hybridization analysis and other methods. Several key findings have oriented much of the recent research being conducted on the origins of cancer. Included in these findings is the fact that the transforming genes of RNA tumor viruses of the [as gene family are homologous to mammalian genes and were actually derived from them (Ellis et al. 1980). Another is the recognition that 2 transfection of human tumor-derived DNA into an indicator cell line causes the cells which incorporate and express a transfected [1; gene to form multilayered groups of morphologically altered cells, a characteristic of tumor—derived cells known as focus formation (Der et al., 1982, Parada et al., 1982, and Santos et al., 1982). This rgg gene was later shown by DNA sequence analysis to be altered by only a single point mutation which caused substitution of one amino acid for another in the encoded protein (Tabin et al., 1982; Reddy et al., 1982; Taparowsky et al., 1982). flag genes have since been identified as transforming genes in many human tumor-derived cells (Suarez, et al., 1987), and have been activated in model systems by carcinogen treatment of cloned normal La; genes (Marshall et al., 1984), by treatment of cultured mammalian cells (Sukumar et al., 1984), and by treatment of animals (Balmain and Pragnel, 1983). In addition to the rgs gene family, many other transforming retroviral genes have been shown to have homologous mammalian counterparts. One pivotal finding was that the transforming gene of simian sarcoma virus encoded a protein which was nearly identical to a mitogenic peptide known as platelet-derived growth factor, which is synthesized by some mammalian cells (Robbins et al., 1983). This discovery implied that one mechanism by which human tumor-derived cells escape the constraints which limit normal cellular division is by the synthesis of growth factors which drive replication. Genes which encode other growth factors have since been implicated as transforming genes of various human tumors (Salomon and Perroteau, 1986). In addition, genes for growth factor receptors (Ullrich et al., 1984), as well as genes which encode some of the proteins thought to couple 3 growth factor binding to DNA replication (Persons et al., 1988), have been identified as possible transforming genes. The portrait which has evolved from these studies is one of a tumor cell which replicates autonomously by circumventing at some point the complex network of biochemical signals which regulate cell division. Rapid advances in cancer research have been made by manipulation of the suspected transforming oncogenes. Studies on their role have most commonly involved, first using any of various techniques to put them into a nontumor-derived cell, and then assaying the phenotype of the cell. Studies of this nature have most often employed rodent fibroblasts as the recipient cells. Such model systems have provided insight into the multistepped nature of carcinogenesis (Land et al., 1983), and have provided critical information on the biochemical action of the oncogene products. The ultimate aim of most cancer research, and of the bioassays which are used to assess the carcinogenicity of a particular agent, is to protect human life. However, the majority of basic cancer research and most short-term assays, have used either animals or nonhuman cells. Although carcinogenicity tests which use animals are irreplaceable for some purposes, and much valid information has probably come from cell culture assays using nonhuman cells, the use of only these systems to analyze the risk to human life would seem unacceptable if more test systems using human cells, such as model transformation systems using cultured fibroblasts, could be devised. This thesis was undertaken (1) to investigate the multistepped process of carcinogenesis, by transfecting cloned oncogenes into diploid human fibroblasts in culture in attempts to obtain cells 4 capable of forming malignant tumors in athymic mice; (2) to determine the biological and biochemical characteristics associated with transformation of diploid human fibroblasts with an N-ng oncogene; (3) to determine the biological and biochemical characteristics of transformed cells obtained by transfection of an N—Lag oncogene into indefinite lifespan human fibroblast cell line MSU-l. Chapter I of the thesis reviews the literature that covers the relationship between environmental agents and carcinogenesis, including discussions of epidemiology, animal models for cancer, studies on transformation of cells in culture, and the relationship between mutations and cancer. Also discussed in Chapter I is literature dealing with the involvement of fig; genes in carcinogenesis and with the biochemical properties of rgs gene products. Chapter 11 consists of a manuscript to be submitted to the journal Cgrgigggggggig. This manuscript details transfections of N-[gg oncogenes into finite lifespan diploid human fibroblasts and characterization of the two stable transformants obtained. I carried out all the studies described in the manuscript and am the principal author on the paper. Dennis G. Fry, Ph.D., a senior research associate in the Carcinogenesis Laboratory, assisted with the research by conducting related pilot studies and by providing training in the molecular biology techniques. Chapter III consists of a manuscript to be submitted to the journal of n l ar i l . This manuscript reports the results of transfections of an N-rgg oncogene into infinite lifespan human fibroblast cell line MSU-l, which was derived in this laboratory. I carried out the major studies included in the manuscript. Dajun Yang, 5 M.D., carried out the cytogenetic analysis and John Dillberger, D.V.M., conducted the pathology analysis of the tumors. REFERENCES Balmain, A. and Pragnell, 1.8. (1983). Mouse skin carcinomas induced in vivo by chemical carcinogens have a transforming Harvey-[gs oncogene. Nature 303,72-74. Barrett, J.C. and Ts’o, P.O. (1978). Relationship between somatic mutation and neoplastic transformation. Proc. Natl. Acad. Sci. USA 15,3297-3301. Chan, G.L. and Little, J.B. (1978). Induction of ouabain-resistant mutations in C3H 10T /2 mouse cells by ultraviolet light. Proc. Natl. Acad. Sci. USA 15,3363-3366. Der. C.J., Krontiris, T.G. and Cooper G.M. (1982). Transforming genes of human bladder and lung carcinoma cell lines are homologous to the rag genes of Harvey and Kirsten sarcoma viruses. Proc. Natl. Acad. Sci. USA 12,3637-3640. Ellis, R.W., DeFeo, D., Maryak, J.M., Young, H.A., Shih, T.Y., Chang, E.H., Lowy, R.R. and Scolnick, E.M. (1980). Dual evolutionary origin for the rat genetic sequences of Harvey murine sarcoma virus. J. Virol. 35,408-420. Land, H., Parada, L.F. and Weinberg, R.A. (1983). Tumorigenic conversion of primary embryo fibroblasts requires at least two cooperating oncogenes. Nature 391,596-602. Landolph, J.R. (1985). Mechanisms of chemically induced multistep neoplastic transformation in C3H 10T1/2 cells. Carcinogenesis 10,211- 223. Marshall, D.J., Vousden, K.H. and Phillips, D.H. (1984). Activation of c-Ha-Lag-l proto-oncogene by in vitro modification with a chemical carcinogen, benzo(a)pyrene diol-epoxide. Nature 319,587-589. Miller, E.C. (1978). Some current perspectives on chemical carcinogenesis in humans and experimental animals: Presidential address. Cancer Res. 38,1479-1496. Parada, L.F., Tabin, C.J., Shih, C. and Weinberg, R.A. (1982). Human EJ bladder carcinoma oncogene is homologue of Harvey sarcoma virus rgg gene. Nature 221,474-478. Persons, D.A., Wilkison, W.O., Bell, R.M. and Finn, O.J. (1988). Altered growth regulation and enhanced tumorigenicity of NIH 3T3 fibroblasts transfected with protein kinase C-1 cDNA. Cell 52,447-458. Reddy, E.P., Reynolds, R.K., Santos, E. and Barbacid, M. (1982). A point mutation is responsible for the acquisition of transforming properties by the T24 human bladder carcinoma oncogene. Nature 390,149-152. Robbins, K.C., Antoniades, H.N., Devare, S.G., Hunkapiller, M. W. and Aaronson, S.A. (1983). Structural and immunological similarities between simian sarcoma virus gene product(s) and human platelet-derived growth factor. Nature 305,605-608. Salomon, 0.5. and Perroteau, I. (1986). Growth factors in cancer and their relationship to oncogenes. Cancer Investig. 4(1)43-60. Santos, E., Tronick, S.R., Aaronson, S.A., Pulciani, S. and Barbacid, M. (1982). T24 human bladder carcinoma oncogene is an activated form of the normal human homologue of BALB- and Harvey-MSV transforming genes. Nature 228,343-347. Suarez, H.G., Nardeux, P.C., Andeol, Y. and Sarasin, A. (1987). Multiple activated oncogenes in human tumors. Oncogene Res. 1,201-207. Sukumar, 5., Notario, V., Martin-Zanca, D. and Barbacid, M. (1983). Induction of mammary carcinomas in rats by nitrosomethylurea involves malignant activation of H-rgg-l locus by single point mutations. Nature 396,658-661. Tabin, C.J., Bradlery, S.N., Bargmann, C.I., Weinberg, R.A., Papageorge, A.G., Scolnick, E.H., Dhar, R., Lowy, D.R. and Chang, E.H. (1982). Mechanism of activation of a human oncogene. Nature 309,143- 149. Taparowsky, E., Suard, Y., Fasano, 0., Shimizu, K., Goldfarb, M. and Nigler, M. (1982). Activation of the T24 bladder carcinoma transforming gene is linked to a single amino acid change. Nature 390,762-765. Ullrich, A., Coussens, L., Hayflick, J.S., Dull, t.J., Gray, A., Tam, A.W., Lee, J., Yarden, Y., Libermann, T.A., Schlessinger, J., Downward, J., Mayes, E.L.V., Whittle, N., Waterfield, M.D. ‘and Seeburg, P.H. (1984). CHAPTER I LITERATURE REVIEW A. Studies implicating environmental agents in carcinogenesis. Research into the causes and cures for cancer has evolved into an in-depth effort at understanding the myriad of metabolic processes, and genetic and epigenetic events that interact during carcinogenesis, and also trying to sort out those phenomena that are critical from those that are not. The desired outcome of this intense effort is a pronounced reduction in the number of cancer deaths through combined improvements in prevention, early detection, and intervention in the course of the disease. Such improvements may well affect the quality of the lifestyle and environment of human beings as well as other aspects of human life. 1. Cancer frequencies and epidemiology. Cancer currently accounts for approximately 22% of all deaths in the United States (Silverberg and Lubera, 1987). Epidemiological reports were originally the only data available to determine if certain agents had the potential to cause cancer in humans. The first of two hallmark studies in the 1700’s related nasal cancer to the use of tobacco snuff (referred to in Redmond, 1970). Shortly thereafter Pott showed a correlation between scrotal cancer in chimney sweeps and exposure to products in coal tar (Pott, reprinted 1963). Some epidemiological data provide overwhelming evidence that certain forms of cancer are caused by a particular environmental agent. That skin cancer is often caused by excessive exposure to ultraviolet light from the sun has been well documented (Emmett, 1973). A very 9 strong correlation has also been shown between tobacco smoking and lung cancer, both for males (US Public Health Service, 1964), and more recently for females (US Public Health Service, 1978). Epidemiological data implicating certain agents as being carcinogenic has often been sufficient to prompt interventive measures by regulatory agencies. Pott’s early observations yielded guidelines on hygiene for chimney sweeps (Butlin, 1892). The correlation between tobacco smoking and lung cancer was convincing enough to lead the US surgeon general to place restrictions on advertising of tobacco products and to require written warnings on their packages, to impel legislatures to regulate smoking in some comunities, and to promote initiation of anti-smoking laws in some businesses. Quite often the results of epidemiological studies suggest that a correlation exists between a given type of cancer and a particular factor, but do not convincingly demonstrate a cause and effect relationship. Although epidemiological research is still widely carried out, there are inherent difficulties with such studies. These problems include improper diagnosis of the type of cancer, false estimates of cancer incidence, unreliability of data that must be drawn from the memory of the persons interviewed concerning lifestyle and prior exposures to carcinogenic agents, and trying to relate recent cancer occurrence to carcinogen exposures that may have occurred many years previously. Although epidemiological studies provide invaluable information concerning the cause of some cancers, the lack of mechanistic information and the inherent problems with such studies, combined with many attractive features of laboratory experiments, led to the creation 10 of model systems utilizing whole animals, cell culture, and other in vitro assays, that might yield new insight into the causes of cancer. 2. Treatment of laboratory animals with suspected carcinogenic agents. Reports that exposures to some agents were associated with unusually high incidences of certain cancers prompted some early workers to administer suspected carcinogens to laboratory animals. Fisher (1906) showed that application of the dye scarlet red to rabbits caused tumor-like growths on the exposed areas. The pivotal findings that dermal application of coal tar to rabbit ears (Yamagiwa and Ichikawa, I915 - referred to in Haddow, 1947) and that extracts of residues from the pyrolysis of hydrocarbons (Kennaway, 1925) caused skin cancer in experimental animals, fostered an intense effort to identify the carcinogenic agents present in such material. Polycyclic aromatic compounds, such as benzo(a)pyrene (Cook et al., 1933) and derivatives of benz(a)anthracene (Kenneway and Hieger, 1930) that were formed during combustion of hydrocarbons, were implicated as the causative agents. Many classes of carcinogens, both organic and inorganic, naturally occurring and man-made, were shown to cause cancer in treated animals. Early investigations, such as those of Berenblum (1940), indicated that carcinogenesis was a multistepped process. He showed that repeated application of croton oil to skin that had previously been initiated with as low as 0.05% benzo(a)pyrene served to "promote” the formation of a cancerous lesion at the site of application. Initiation is regarded as a permanent change that can occur after the single application of a subthreshold dose of a carcinogenic agent that is 11 insufficient by itself to cause cancer (Boutwell, 1974). Promotion is broadly defined as a reversible process whereby a tumor is elicited by application of a promoting agent subsequent to administration of an initiating agent (Boutwell, 1974). This might occur by growth enhancement of previously initiated cells into larger populations with a proportionally larger probability of undergoing a subsequent initiation-like event. Later studies went on to show that promotion could be subcategorized into at least two discrete steps (Slaga et al., 1980). According to Scribner and Suss (1978), evidence that initiating and promoting events actually take place in humans is shown by the strong probability of developing bronchiogenic carcinoma in asbestos workers who also smoke tobacco, and by the decline with time of the risk of developing lung cancer in people who have quit smoking. Experiments utilizing live animals are commonly carried out to test for toxicity, mutagenicity, and teratogenic and carcinogenic effects of compounds, and are often required by law to screen new products before market approval. Although there are some obvious advantages in using live animals, including weighing the effects of uptake, metabolism, and excretion, and generation of dose response data, there are also disadvantages. These include host specificity, expense, latency periods that are often long, sacrifice of many animals, and a poor ability to resolve questions of the mechanisms involved. Similar to the need for epidemiological studies, tests utilizing live animals are also essential to determine some aspects of the toxicity and carcinogenicity of various agents. However, various in vitro strategies for the study of carcinogenesis, including a battery of short term tests to determine the mutagenic potential of agents, 12 treatment of cultured cells to determine the capacity of different agents to cause changes characteristic of tumor-derived cells (transformation), and elaborate molecular biology techniques (see below) have been developed. 3. Treatment of mammalian cells in culture with suspected carcinogens. There have proved to be many advantages of using cultured mammalian cells in mutation and transformation assays. Some of the advantages include a relatively short duration between treatment time and endpoint, ease of growing many cell types, cost effectiveness of the experiments, and the capability' of ascertaining "mechanisms”, etc. The cell type most widely utilized for in vitro tissue culture experiments is the rodent fibroblast because early attempts to culture these cells were very successful. In addition, well characterized assay systems were developed based on the use of such cells, which enabled the identification of' mutagenic or carcinogenic agents by virtue of their causing discernable effects. Berwald and Sachs (1963) made the initial discovery that cells could be transformed in culture by chemical carcinogens when they showed that polycylic aromatic hydrocarbons could cause alterations in the colony growth pattern of primary Syrian hamster embryo fibroblasts in culture. These cells are diploid and have a finite lifespan in culture (Meyer, 1983). Much work has been done to further elaborate the best assay conditions for in vitro transformation assays using these cells. DiPaolo et al. (1969, 1971) refined the assay, showed it was sensitive to many carcinogens, and showed that some of the transformants were tumorigenic. Pienta et al. (1977) developed the 13 assay further by utilizing cryopreserved primary SHE fibroblasts for target cells and feeder layers. Since SHE cells have a finite lifespan and do not spontaneously transform in culture, Barrett’s group has extensively employed them to study the process of neoplastic progression after carcinogen treatment (Kai and Barrett, 1986, Oshimura et al., 1985). The first mouse fibroblast cell line that was developed for transformation studies was derived from the ventral prostate of a C3H mouse (Chen and Heidelberger, 1969a). This cell line was selected because it exhibited density dependent contact inhibition of growth and . was therefore suitable for focus assays of the transforming potential of a number of suspected carcinogens (Chen and Heidelberger, 1969b, 1969c). Subsequently, a highly contact inhibited cell line designated C3H10TI/2Cl8, was established from a C3H mouse embryo (Reznikoff et al., 1973a), and has since been extensively employed to assay the capacity of various agents to cause focus formation (Reznikoff et al., I973b; Benedict et al., 1977; Landolf and Heidelberger, 1979). The ability to form a multilayered group of cells called a focus is one characteristic of transformed fibroblasts which distinguishes them from normal fibroblasts that grow to a confluent single layer of cells in tissue culture dishes. A focus assay involves treating normal cells at low density with an agent suspected of causing cancer, and then allowing the cells to grow to confluence. They are stained after (an appropriate expression interval and any foci present are evaluated. The types of foci observed in the C3H10T1/2 system have been categorized into three classes, Types I - 111. Type I foci stain lightly and are not considered significant, Type II foci stain darkly 14 because of multiple dense cell layers, but have smooth edges, and Type III foci also stain darkly because of multiple dense cell layers, but have irregular edges resulting from the criss-cross orientation of the cells (Landolf, 1985b). This assay is deemed a relatively valid predictor of carcinogenic potential because, when the focal derived populations of cells are harvested, expanded, and injected into immunosuppressed C3H mice, 50% of those derived from Type II foci and 80% of those derived from Type III foci form tumors (Landolf, I985b). 4. In Vitro Transformation of Human Cells. Since evidence strongly suggests that most human cancers are caused by exposure to environmental carcinogens, cultured human cells would, in principle, be the most logical choice for use in developing in vitro screening assays for carcinogens, and for studying the mechanisms by which transformation occurs. Although the majority of human tumors are of epithelial origin, human epithelial derived cells have proved to be difficult to maintain in culture and so reproducible methods to culture such cells have only recently been established. Therefore the majority of transformation experiments with human cells, as with all mammalian cells, have utilized cultured fibroblasts. Normal diploid human fibroblasts do not, as a rule, spontaneously transform in culture to either indefinite lifespan cell lines or malignant cells. This stability makes them ideal candidates for use in studying the process of carcinogen-induced neoplastic transformation. There are numerous reports of partial transformation of human fibroblasts in culture following carcinogen treatment. The end-points most often assayed are growth in soft agar or the acquisition of 15 infinite lifespan in culture. Namba et al. (1978, 1981) have successfully transformed human fibroblasts to infinite lifespan following repeated treatment with either 4-nitroquinaline-1-oxide or X- rays. Many workers have used growth in soft agar to assay for transformation of human cells after treatment with suspected carcinogens. For example, Milo and DiPaolo (1978) used a variety of carcinogens to induce anchorage independent growth in human fibroblasts. Similar studies have been done by Landolf’s group using carcinogenic metals (Biedermann and Landolph, 1987), and by McCormick and Maher and their coworkers using radiation and chemical carcinogens (Silinskas et al., 1981, Wang et al., 1986). There are several reports that human fibroblasts can be malignantly transformed after exposure to carcinogens. However, results of a recent study have indicated that most of these reports are invalid. For example, the most widely known study on the malignant transformation of human fibroblasts is that. of' Kakunaga (1978) who reported the in vitro transformation of a cell line designated KD by exposure to 4-nitroquinoline-l-oxide or N-methyl-N’~ nitro-N-nitrosoguanidine. Careful analysis of the reported transformants has revealed that the transformed cells were not derived from KD cells, but were most likely derived from a contaminating human fibrosarcoma cell line. Other reports of transformation of human fibroblasts did not convincingly demonstrate that the tumors were, in fact, malignant (Borek, 1980; Min et al., 1980; Ming et al., 1986). 16 5. Relationship Between Mutation and Cancer. As noted above, most known human cancers are thought to be caused by environmental carcinogens. The combined results of many experiments using both human and nonhuman cells suggest that these agents exert their carcinogenic effects by interacting with and altering the genome. Many chemical carcinogens are either electrophilic or can be metabolically activated to electrophilic species that bind covalently to DNA to form "adducts" (Miller, 1981). Such adducts can subsequently be converted to mutations, i.e., permanent changes in DNA. Quite often the mutation induction curve for a carcinogen parallels its dose-response curve for transformation of cells treated in culture (Barrett and Ts’o, 1978; Landolph, 1985a). In recent years many workers have concentrated on identifying particular genetic sequences that, when modified by carcinogen treatment, result in genes being able to cause changes in cells that are characteristic of transformed cells. One of the major accomplishments of modern cancer research has been finding that the genetic material in certain tumor viruses that is capable of causing cancer in animals is homologous to normal endogenous human genes, i.e., protooncogenes (see below). 8. Involvement of the gas genes in carcinogenesis. 1. Evolution of RNA tumor virus ,[g§_ genes and identification of homologous cellular genes. Four murine sarcoma viruses (MSV) of the 'rgs" (rgt sarcoma) family of RNA tumor viruses have been identified. Harvey-MSV was isolated from a BALB/c mouse after it had been inoculated with a virus that was obtained by infecting rats with Maloney murine leukemia virus (Harvey, 17 1964). Kirsten-MSV was obtained from rats that had been inoculated with an extract from a C3H mouse lymphoma (Kirsten and Mayer, 1967). BALB-MSV was isolated from a BALB/c mouse hemangiosarcoma that was obtained after injection of a filtrate from a BALB/c mouse chloroleukemia (Anderson et al., 1981). Rasheed-MSV was isolated after cocultivation of a chemically transformed rat tumor cell line with spontaneously transformed Sprague-Dawley rat embryo cells that were releasing a slow-transforming endogenous ecotropic type C virus, a virus which normally occurs as a stably integrated provirus, but which when it is induced to replicate, does so best in rat cells (Rasheed et al., 1978). Scolnick and Parks (1974) showed that Harvey-MSV contained genetic sequences of Maloney murine leukemia virus origin and also sequences that had been transduced from a rat genome during evolution. Scolnick et al. (1973) showed that Kirsten-MSV was also a recombinant virus which had evolved by transduction of rat DNA into Kirsten murine leukemia virus. Harvey-MSV and Kirsten-MSV are oncogenic to rats and were shown to transform in vitro a commonly utilized mouse fibroblast cell line developed at the National Institutes of Health (NIH/3T3). Transformation of NIH/3T3 cells following transfection of subgenomic clones of these viruses, followed by hybridization of the transfected transforming gene with rat-specific probes, indicated that the transforming 21 kd protein was the product of a rat gene (Chang et al., 1980). De Feo et al. (1981) used sequences derived from the transforming gene of Harvey-MSV to probe rat genomic DNA. Two related genes homologous to the viral probe were identified, isolated, ligated to a Harvey-MSV long terminal repeat (LTR), and found to induce 18 transformation when transfected into NIH/3T3 fibroblasts. When Chang et al. (1982) probed human genomic DNA with the transforming sequences from Harvey-MSV and Kirsten-MSV, a total of four homologous genes were identified. These cellular genes which are often referred to as protooncogenes, were named c—Ha-rggl and 2 and c—Ki-rgsl and 2. Later, c-Ha-rggl and c-Ki-rggz were shown to be functional genes, and c-Ha-rgsz and c-Ki-rgsl were shown to be processed pseudogenes, genetic sequences which does not actually code for a functional protein (DeFeo et al., 1981; McGrath et al., 1983). 2. Isolation and molecular characterization of cellular Lg; oncogenes. A test that has been widely used to detect activated oncogenes in malignant tissue is the NIH/3T3 transfection assay. NIH/3T3 cells are efficient recipients of transfected DNA, and form readily detectable foci with activated as and other oncogenes (Lemoine, 1987). The focus-derived cells form malignant tumors when injected into athymic mice, whereas nonfocus—derived NIH/3T3 cells usually do not form tumors in athymic mice. To isolate activated oncogenes, tumor cell- derived genomic DNA from a species other than mouse is transfected into NIH/3T3 cells which are allowed to grow to confluence and assayed for focus formation. DNA from focal derived populations can then be used in subsequent rounds of transfection of NIH/3T3 cells, to enrich for the transforming gene of interest. Libraries can then made of genomic DNA from the transformed NIH/3T3 cells and probed for highly repetitive sequences characteristic of the donor species. Using the above protocol, Shih and Weinberg (1982) cloned a transforming gene from the human EJ bladder carcinoma line, and Goldfarb et al. (1982) cloned a transforming gene from the human T24 l9 bladder carcinoma cell line. Each group found that when transfected into NIH/3T3 cells, their cloned gene caused transformation. It was subsequently realized that these two cell lines were derived from the same original tumor (Shimizu et al., 1983). Der et al. (1982), Parada et al. (1982), and Santos et al. (1982), independently found that the transforming gene isolated in the above manner from the human EJ bladder carcinoma cell line was homologous to the transforming gene of Harvey-MSV. Shortly thereafter, the genetic alteration that activated the H-rgs oncogene in the EJ cell line was identified by first constructing hybrid vectors between cloned transforming and nontransforming alleles, and then determining the nucleotide sequence of a short restriction fragment that was sufficient to confer transforming ability when ligated in place of its homologous counterpart into the normal c-H-rgs gene (Tabin et al., 1982, Reddy et al., 1982, and Taparowsky et al., 1982). These groups found that a point mutation which caused a base substitution in codon twelve of the c-H-rgs protooncogene, was responsible for the transforming activity of the 1;; sequence. Point mutations in several critical codons of :35 genes which cause the transfected DNA to cause focus formation in NIH/3T3 cells are now often referred to as ”activation” of a fig; gene. In addition to the above findings, there are many other examples of H- ras activations that have been detected in human tumor-cell DNA that were capable of causing focus formation in NIH/3T3 cells. For example Yuasa et al. (1983) found that c-H—rgg was activated by a single base substitution in codon 61 in cells from lung carcinoma cell line H5242. Since the only protooncogene activating mutations in m genes known at the time were codon 12 mutations in the c-H-rgs, gene, 20 Feinberg et al. (1983) analyzed 29 human tumor DNA samples by restriction enzyme analysis to detect a given point mutation at codon 12. None of the 29 samples analyzed had the specific point mutation assayed for, so the authors concluded that previous results obtained with the human EJ bladder carcinoma cell line provided a good model for human carcinogenesis, but were not representative of the majority of human cancers. Subsequent to the findings of Feinberg’s group, many other point mutations have been found that are capable of activating cellular Lg; genes. Most often, when a human oncogene has been identified by virtue of its causing focus formation in NIH/3T3 cells, it has been found to be a cellular Kirsten as gene (c-K-rgs). For example, Der (1982) first reported that a cellular gene homologous to the transforming gene of Kirsten-MSV was identified in DNA from the human lung carcinoma cell line LX-I by virtue of its ability to cause focus formation in NIH/3T3 cells. Capon et al. (1983) then reported that the human lung carcinoma cell line Calu-l has an activated c-K-Lasz gene which resulted from substitution of cysteine for glycine at the amino acid specified for by codon 12, and that the human colon carcinoma cell line SW480 is also activated at codon 12 of c-K—rgsz, but the product of this gene has a valine in place of glycine. Shimizu et al. (1983) found that DNA from the human neuroblastoma cell line SK—N-SH also induced foci on NIH/3T3 cells. When the human transforming gene was cloned by this group, it was found to be a third cellular member of the 1;; gene family distantly related to Harvey and Kirsten murine sarcoma viruses, but not detected by ‘viral derived probes to their transforming genes. Shimuzu et al. (1983a) named this 21 newly identified gene N-rgg, and established its orientation of transcription and approximate exon coding regions. Marshall et al. (1982) had reported that DNA from the human fibrosarcoma cell line HT1080 and the human rhabdomyosarcoma cell line R0 was capable of causing focus formation in transfected NIH/3T3 cells. Soon after Shimuzu’s report, Hall et al. (1983) reported that N-rg; had also been identified as the transforming gene of the HT1080 and RD cell lines, and also of the human promyelocytic leukemia cell line HL60. The activating mutation of the HT1080 N-_r;_a_§_ oncogene was localized by testing the ability of chimeric molecules between the cloned N-ggs gene from normal human fibroblasts and the HT1080 fibrosarcoma cell line to transform transfected NIH/3T3 cells, and then was sequenced (Brown et al., 1983). The only difference between the normal and activated gene was shown to be a C to A transversion altering the 6lst amino acid from glutamine in the normal allele to lysine in the HT1080 gene. The identical activating base substitution was also identified in the cell line SK-N-SH (Taparowski et al., 1983). In a similar fashion the N-rgs oncogene of the RD cell line was shown to be activated by an A to T transversion which caused substitution of histidine for glutamine at codon 61 (Chardin et al., 1985). In humans, N-rgs has been shown to be activated predominantly in hematopoietic. malignancies (805 et al., 1987, Needleman et al., 1986). Souyri and Fleissner (1983) found that N—rgg was an activated oncogene in the three human T-cell leukemia cell lines RPMI 8402, CCRF-HSBZ, and p-12. To investigate whether the presence of an activated N-rgg oncogene correlated with the clinical course of acute myeloblastic leukemia, Gambke et al. (1984) evaluated its occurrence in bone marrow cells 22 from a patient at the outbreak of the disease. An active N-rgs oncogene was detected in the bone marrow cells during the acute phase of the disease. However, an N-m oncogene was not present in non- affected fibroblasts of the same individual. This N-rgg oncogene was later found to be activated by a point mutation in codon 12 that caused substitution of aspartic acid for glycine (Marshall, 1985). Bos et al. (1985) found activating mutations in codon 13 of peripheral blood cells in four out of five patients with acute myeloid leukemia. Peripheral blood cells from one patient tested in remission no longer had a detectable N-Las mutation. Senn et al. (1988) analyzed DNA isolated from the cells of blood or bone-marrow samples from 18 patients with acute non-lymphocytic leukemia and 14 patients with acute lymphocytic leukemia. The only N-rgs mutations that were found were in cellular DNA derived blood and bone marrow cells from five acute non- lymphocytic leukemia patients. In a follow-up study on three affected patients, the two individuals in remission no longer had detectable levels of a mutant N-Lag gene in DNA isolated from peripheral blood cells, whereas the other affected patient exhibited detectable levels of a mutant N-rgg gene (codon 13) in DNA from peripheral blood cells. Hirai et al. (1987) analyzed bone marrow cells from eight patients with myelodysplastic syndrome, a disease which occasionally progresses to frank leukemia, and found three of those patients had marrow cells containing an active N-rg; oncogene with a point mutation in codon 13. These investigators monitored the progression of the disease in all eight individuals and found that the three patients who had an activated N-Lgs gene later developed acute myeloid leukemia, whereas 23 the other five patients exhibited no progression of the disease within one year following diagnosis. 3. Mechanisms of activation of cellular Lg; genes. figs oncogenes were detected in many early analyses of human tumor cells by virtue of their ability to cause focus formation in transfected NIH/3T3 cells. The transforming potential of the isolated [gs genes was attributed soley to various point mutations that caused single amino acid substitutions in the gene product. But there was certainly precedence for investigating other possible mechanisms of Lg; gene activation, since DeFeo et al. (1982) had found that induction of high expression of the normal c-H-r_a_s_ gene was capable of causing transformation in transfected NIH/3T3 cells. Tahara et al. (1986) found that expression of c-H-ng was higher in metastatic nodules of patients with gastric carcinomas than in the primary tumors, and that those patients that had the highest expression of c-H-Las had the poorest prognosis. One possible mechanism by which higher than normal expression of a gene might occur is by amplification of that gene. There are numerous examples of' amplification of’ rg§_ genes in DNA obtained from tumor cells. The cellular K-rgg gene was amplified 5- fold in the colon carcinoma cell line SW480 (Capon et al., 1983; McCoy et al., 1983), 40-fold in DNA from a primary bladder tumor (Fujita et al., 1985), and 10-fold in the giant cell carcinoma cell line Lu65 (Alitalo et al., 1984; Taya et al., 1985). The cellular H-rgs gene was amplified up to 20-fold in cell line SK—2 derived from a malignant melanoma patient (Sekiya et al., 1985), and 10 fold in cells from a primary bladder carcinoma (Hayashi et al., 1983). 24 Another proposed mechanism by which cellular Lag genes might be activated is through alterations in the methylation status of genomic cytosine residues. The restriction endonucleases 119311 and [11131 do not cleave the sequence 5’-C-G-3’ if the cytosine is methylated, whereas Mggl will cleave that sequence. By digesting DNA purified from human cancers and adjacent normal tissue with these enzymes, followed by hybridization with labeled K-Lag and H115; probes, it was ascertained that there were marked decreases in the methylation status of some tumor DNAs compared to DNA obtained from nontumor-derived cells of the same patients (Feinberg and Vogelstein, 1983). Another possible mechanism of gas oncogene activation was suggested by Cichutek and Duesberg (1986) who showed that in many human tumor DNAs a noncoding exon of c-H-r_a_s_ is altered. 4. Studies of multistepped carcinogenesis by transfection or infection of m oncogenes into manalian cells. Many studies on the multistepped nature of carcinogenesis have involved 13; oncogenes. Some workers have transfected cloned as oncogenes into cultured normal cells in attempts to create cancer cells in vitro. Most of these attempts have required that more than just an activated [gs gene be transfected, or that the recipient cells already possess some attributes of a tumor-derived cell, such as an extension of the normally limited lifespan usually observed once the cells are cultured. Land et al. (1983) showed that transfection of finite lifespan, rat embryo fibroblasts with an activated H-rgs oncogene resulted in morphological transformation, but with limited potential to grow in soft agar. The ability to grow in semisolid 25 medium such as soft agar is an attribute of tumor-derived cells and is often used as an assay to predict the malignant potential of a cell. Cotransfection of the T24-H13; oncogene and either the v-myg or polyoma virus large-T antigen oncogenes into finite lifespan rat embryo fibroblasts resulted in the formation of transformed foci. When the cells from the foci were isolated, expanded in culture, and injected into athymic mice, they gave rise to tumors. The tumors induced by cells containing the cotransfected mg and myg oncogenes were benign and stopped growing before ever becoming sufficiently large to kill the mouse, whereas the tumors induced by the cotransfected rag and polyoma virus large-T antigen oncogenes formed a malignant tumor that killed the mouse. When infinite lifespan Rat-l cells were cotransfected with T24-H-Las and v-myc oncogenes, foci composed of morphologically transformed cells were formed. When isolated, expanded in culture, and injected into nude mice they formed invasive malignant tumors. Newbold and Overell (1983) transfected the c-H-ras oncogene into either normal finite lifespan Syrian hamster fibroblasts, as well as four infinite lifespan hamster fibroblast cell lines. Following transfection, transformed foci were observed against a confluent monolayer for each of the infinite lifespan cell lines, but no foci were evident in the normal diploid cell line. Ruley (1983) found that the adenovirus early region-1A gene could cooperate with either the T24-Wm oncogene or the polyoma virus middle-T gene to cause focus formation in finite lifespan baby rat kidney cells transfected in culture, whereas neither the T24-H41; or polyoma virus middle-T genes caused any noticeable effects on their 26 own. Ruley et al. (1986) later showed that the EIA gene cooperated with T24-H-rgg by a mechanism other than merely facilitating the acquisition of an infinite lifespan. This is because the T24-H-rgg was not sufficient by itself to transform an infinite lifespan rat embryo fibroblast cell line, but did so when co-transfected with the EIA gene. Parada et al. (1984) found that the gene coding for the p53-tumor antigen was also capable of cooperating with the EJ-H-rgs oncogene in the tumorigenic transformation of transfected rat embryo fibroblasts. p53 is a protein that is expressed in high levels in many different tumor-derived cells. Barrett’s group found that transfection of Syrian hamster embryo cells with the v-H-r_a_s oncogene caused morphological transformation, but that the cells were not immortal and senesced shortly after being isolated (Thomassen et al., 1985). However cotransfection of v-H-Lgs and the v—myc oncogene yielded transformed cell populations that formed progressively growing tumors after being injected into nude mice. On subsequent karyotypic analysis all tumors that were formed showed a consistent loss of chromosome 15, suggesting that loss of a possible suppressor gene was also necessary for malignant transformation (Oshimura et al., 1985). Support for this concept was given by Geiser et al. (1986) who showed that fusion of normal human fibroblasts with EJ-bladder carcinoma-derived cells caused suppression of tumorigenicity, even though expression of the c-Ha-ras oncogene was maintained. Contrary to the reports that the activation of multiple protooncogenes are necessary for complete transformation of cells in culture, Spandidos and Wilkie (1984) found that transfection of early 27 passage Chinese hamster fibroblasts or rat embryo fibroblasts with rgs genes cloned into high expression vectors was capable of causing either imortality or malignant transformation, depending on the as gene used. When cells were transfected with the normal c—H-rgs gene which had been cloned into the Homer plasmid, a plasmid designed to give high expression, clones with a normal morphology were obtained after drug selection. When these clones were isolated and propagated in culture some of the resulting populations acquired immortality. When rodent fibroblasts were transfected with the T24-H-[gg oncogene which had been cloned into the Homer plasmid, morphologically transformed clones were obtained that gave rise to tumorigenic populations of cells. When these two H-rgs genes were cloned into low expression vectors and then transfected into rodent fibroblasts, no transformation was observed under any circumstances. Pozzatti et al., (1986), also reported that malignant cells could be generated by transfection of early passage rat embryo fibroblasts with the cloned T24-H-rgs oncogene. 5. Transformation of human cells in culture by :3; oncogenes. Various workers have utilized human cells as recipients for exogenous rgs oncogenes, with limited success. Newbold et al. (1983) reported that finite lifespan human fibroblasts could not be transformed by transfection of the cloned T24-Wm oncogene. Similarly, Sager et al. (1983) reported that diploid human fibroblasts were resistant to transformation after transfection of the EJ-H-La; oncogene, even though the transfected DNA was shown by Southern hybridization analysis to be incorporated intact into the genome. Sager (1986) later showed that even though the mutant form of the H-ras 28 oncogene product was being expressed in the transfected human cells, they still were not transformed. Sutherland et al. (1985) found that the ability to grow in soft agar was conferred on normal human fibroblasts that were transfected with the cloned T24-Wm oncogene. However these investigators did not analyze the recipient cells to ensure that the transfected plasmid had been integrated within the genome and was actually being expressed. Feramisco et al. (1984) microinjected high levels of mutant H-rgs protein into both rodent and human fibroblasts. Although the rodent fibroblasts exhibited a transformed morphology until the transient effects of the microinjected protein subsided, the human fibroblasts remained normal in morphology. Similarly, Marshall et al. (1983) observed no morphological alterations in normal human fibroblasts that had been transfected with the activated N-rgg, oncogene *which was isolated from the HT1080 human fibrosarcoma cell line and cloned into the plasmid pSV2ggg. Recently Hurlin et al. (1987) working in this laboratory, observed morphological transformation, focus formation, and anchorage independent growth of diploid human fibroblasts by the transfected T24-Mia; oncogene which was cloned into the Homer high expression plasmid. Their results contradict those of Spandidos (1986) who reported that, in his hands, transfection of human fibroblasts with this construct had failed to cause detectable alterations. The Homer vector was designed to be a high expression vector, and Hurlin et al. showed that the levels of' mutant Hmn .3; ..n a a - to 8 s. to a — - 2-»... ‘ 8.. ~~ . _ .5 I 3.2.. ’ 2:-.. .e. 38 . . = 22.... ’3 APPENDIX B Discussion of controls used in transfection experiments. The phenotypes of the human fibroblasts which are transformed after transfection with pSV N-gas are most likely the result of a highly expressed N-gas oncogene, and are not just the effects of the Maloney leukemia virus long terminal repeats. The best evidence for this is that transfection of human fibroblasts with the H-gas oncogene cloned into the Homer plasmid, a high expression vector also containing LTRs, is capable of causing transformation, whereas the Homer plasmid lacking the inserted oncogene does not cause transformation (P. Hurlin, personal conlnunication). In addition, Hurlin found that the non- mutated cellular H-gas gene, when cloned into the Homer plasmid, does not cause transformation of transfected human fibroblasts. It. is unlikely therefore, that transfection of human fibroblasts with the cellular N-gas gene cloned into the ZIP plasmid would cause transformation of human fibroblasts. Transfection of human fibroblasts with N-gas oncogenes having other activation mutations, or with other gas oncogenes such as H-gas and K- gas would most likely cause transformation if the genes were cloned into vectors capable of giving high expression in human fibroblasts. Evidence for this is that I found the HT1080 N-gas oncogene which is mutated at codon 61 causes morphological transformation of human fibroblasts when cloned into a vector with two LTRs, and Hurlin has found that the H-ms gene cloned into the Homer plasmid can cause transformation of human fibroblasts. The presence of a different LTR could have caused quite different results since the activity of LTRs varies depending on the origin of the LTR and the species of donar of the cells that the LTR has been introduced into. There is strong evidence for the cause and effect relationship of transformation of human fibroblasts by a transfected N-gas oncogene. When pSV N-gas is transfected into diploid human fibroblasts and these cells are then selected for resistance to Geneticin, most of the drug resistant colonies are morphologically transformed. If the transfected cells are not selected for drug resistance, but allowed to grow to confluence, the frequency of focus formation is approximately the same as the frequency of morphologically transformed Geneticin resistant colonies obtained in dishes containing cells that were selected for Geneticin resistance. When expression of the transfected N-gas oncogene is evaluated, it is high in those cells which remain transformed, but undetectable in cells which reverted to a normal phenotype after showing initial morphological transformation. The control cells, either diploid human fibroblasts transfected with pSV2neo which does not contain an oncogene, or immortal MSU-l cells, grow to only a limited extent in soft agar, and do not form foci or tumors. Since the amount of growth of normal human fibroblats in soft agar can be regulated by the amount of serum which is supplied to the agar-medium, the fact that the control cells grow poorly is the result of titering the serum concentration to a point where normal fibroblasts have only a low background growth. This allows us to measure the increased capacity for growth in agar exhibited by the human fibroblasts that were transformed by N-gas. Normal human fibroblats do not form foci and have not been induced to do so by the addition of high concentrations of serum to the culture medium. Yet human fibroblasts transformed by the Simian sarcoma virus oncogene, coding for a mitogenic peptide related to platelet-derived growth factor, form distinct foci. Therefore the phenotype of focus formation appears to be attributable to enhanced mitosis in the focus forming cells in relation to the background cells. Simple addition of more serum to the culture medium would not cause focus formation in the control cells because all of the cells should respond equally well, and the background of confluent cells should only become more dense. The phenotype of tumorigenicity is very complex and not well understood. The control cells may or may not remain viable at the site of injection without giving rise to a tumor. However, when the cells are first injected a small nodule is present at the site of injection which is not apparent by 10 days post-injection. Therefore, the majority of the control cells have probably been removed by the immune system of the mouse by this time, or else they did not have the capacity to remain viable in the mouse.