;._.,. , ‘ ' ’ n-.. . - fl :- Iv N’V'f'flvl; ....,..N I’:r.... N..\ A .l.. .5. '1'". ..A «[450! v "1:: :‘S 31.”. u... , ._ 5.1. m - ”1"." n v ."flu {1.4L I; , i- . A .nA- ”:1". r A., ,A. . . nu v'l m r. "I 3 "Hiram-1'7.”W 0 . n; ' 1 ~rr' 9" I-v~.;.--. ; t .A”.,, > 'glo.;:'..' a..‘,.' «2-. . .. . 9 ma. -v v. ..,, 'vraA .....,., '-:~¢..«<.. v3.- ,-..,.. , ”NM ; v' v I . .....;.‘!:' a ....3 ,4“ ”an!” x '1", ‘ ”.43. o...:'..‘...A,A'.A ,H...“ v.17: r. L, ,., .q -| ‘ ." ’“VV’PIJM fl "mm... --« ,5; I": - ,:.,..,,; . .M v... .- fl "5 "I; '1 ¢.._,',," ' _v . Zulu-N, _,., V ' ’ f' 'o.’ ‘ rift w~ "l"~ ’Ah . ' ”Or-\ln: ' u“ r Haw rm»... . -. ‘ Av .— " ‘ . " -' ’ t" .."¢.' r v N... . Li}; LA 2‘ .A ‘ . I "rim my..- ,. 17-1-5. ;. . " '::V a l 1-}. .473?! £34" ‘9” ow». :oo.':, '0‘. I. ,,. u" . ' .4... . 'vu "( 5:: ,, ‘2"- ‘t' 5:: _ I'r . » ,A., “a". _ ‘ 3;; .A‘ is: v: k t .S‘ . 3'..'r,'::* :2 ’~' _ '3" A R, IVSER SITY U8 8m IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII III II IZIIIIIIIIII 3 1293 01055 This is to certify that the dissertation entitled Hypomethylation of raf and Ha-ras in Mouse Liver Following Cell Proliferation and in Mouse Liver Tumors presented by Jean S. Ray has been accepted towards fulfillment of the requirements for Ph. D. degree m Pharmacology and Tox1cology Major professor Date :1” 0 /?F’Z/ MS U is an Affirmative Action/Equal Opportunity Institution 0-12771 LIBRARY I Michigan State University PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or More data duo. DATE DUE DATE DUE DATE DUE L__L_J I-CJE] fist—J , I___II::II___I I I I—TI J MSU In An Affirmative Action/Equal Opportunlty Institution HYPOMETHYLATION OF raf AND Ha-ras IN HOUSE LIVER FOLLOWING CELL PROLIFERATION AND IN HOUSE LIVER TUMORS BY Jean Sara Ray A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Pharmacology and Toxicology 1992 ABSTRACT HYPOMETHYLATION OF raf AND Ha-ras IN HOUSE LIVER FOLLOWING CELL PROLIFERATION AND IN MOUSE LIVER TUMORS BY Jean Sara Ray. The liver tumor-prone B6C3F1 male (C57BL/6 female x C3H/He male) mouse, the more liver tumor-prone paternal strain (C3H/He), and the relatively resistant maternal strain (C57BL/6), plus the liver tumor-prone CD-l male mouse were employed for this study. The objective was to test the hypothesis that hypomethylation of DNA might lead to the aberrant expression of proto-oncogenes which, along with mutation, appears to play a role in carcinogenesis. This was accomplished by examination of the methylation status and mRNA levels of raf and Ha-ras in the nascent liver, following cell proliferation, and in liver tumors. Rat liver tumors induced by the injection.of“viral-Ha-ras infected.rat liver epithelial cells were also examined. Methylation status of the proto- oncogenes rat and Ha-ras was determined by digestion of DNA with. methylation sensitive restriction endonucleases and Southern blot analysis. Messenger RNA and p21 ras protein levels were assessed by Northern and Western blotting, respectively. The raf gene was relatively hypomethylated in the tumor- prone C3H/He relative to the tumor—resistant C57BL/6. The B6C3F1 appears to inherit a raf allele from the C57BL/6 that is methylated, at the external cytosine in a 5'-CCGG-3' sequence, while the corresponding allele inherited from the C3H/He is not.methy1ated at this site. ‘With regard to the 5'- CCGG-3' site of interest in raf, the CD-l exhibited a level of methylation similar to the 36C3F1 strain. Hypomethylation of raf was found in BGC3F1 and C57BL/6 following partial hepatectomy, and in the BGC3F1, but not the C57BL/6, following phenobarbital administration (75 mg/kg/day) for 14 days. Ha- ras and rat were found to be hypomethylated in mouse liver tumors. Elevated levels of raf mRNA were found in phenobarbital-induced tumors, but not spontaneous tumors, while increased Ha-ras mRNA levels were found in both. The studies reported here support the notion that hypomethylation of DNA, an epigenetic change that is expected to result from threshold-exhibiting events, is involved in the multistep process underlying carcinogenesis. iii This dissertation is dedicated to my husband, John Bergsma. My graduate studies would not have been possible without his help and support. iv ACKNOWLEDGMENTS I am grateful to Dr. Jay Goodman for his guidance throughout this project. I would like to thank my guidance committee, Dr» James Bus, Dr. James.Trosko, Dr; James Bennett, Dr. Richard Schwartz, and especially Dr. Clifford Welsch for their assistance and support. Stephanie Bell provided excellent technical assistance during major portions of this project. Financial support in the form of a fellowship from The Upjohn Company is gratefully acknowledged. The friendship of my fellow graduate students, especially Roseann Vorce, Sandy Hewett, Cindy Hoorn, and Jennifer Pavlock has made the years spent on this project more pleasant. I thank Dr. Susan Barman for moral and nutritional support. Finally, I thank my family: my father for teaching me that I could do anything, my mother for being a good role model, and my husband and children for brightening my life. TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES LIST OF ABBREVIATIONS in in in in INTRODUCTION 1. An Overview of Multistep Carcinogenesis 2. The Role of Proto-oncogenes Carcinogenesis A. The role of Ha-ras in carcinogenesis B. The role of raf in carcinogenesis C. Signal transduction in hepatocytes 3. Carcinogenesis in Mouse Liver A. The B6C3F1d mouse B. The CD-ld mouse 4. The Involvement of Cell Proliferation Carcinogenesis . A. The role of cell proliferation carcinogenesis B. Partial hepatectomy and cell proliferation C. Phenobarbital and cell proliferation 5. The Role of S-Methylcytosine in Mediating Gene Expression A. The role of 5-methylcytosine differentiation and aging B. S-Methylcytosine as a regulator of gene expression C. The inheritance of methylation status 1) Somatic 2) Through the germ line 3) Partial methylation 4) Imprinting D. DNA. hypomethylation. associated. with carcinogen treatment 6. Hypothesis and Objectives vi 12 12 13 14 16 16 18 21 21 22 26 27 28 30 MATERIALS AND METHODS 1. 6. 7. 10. 11. 12. 13. 14. 15. Animals: Maintenance and Treatment A. Mice B. Cells C. Phenobarbital administration D. Partial hepatectomy Isolation of DNA Restriction. Endonuclease ‘Digestion, .Agarose Gel Electrophoresis, and Southern Transfer A. Restriction endonuclease digestion B. Agarose gel electrophoresis C. Southern transfer Labeling of Probe with :”P by the Random Primers Method A. Probes B. Labeling Hybridization of DNA Affixed to Hybond Membranes Autoradiography RNA Isolation Simultaneous Isolation of DNA and RNA A. DNA isolation B. RNA isolation Assessment of RNA Integrity by Electrophoresis Agarose Gel Electrophoresis of RNA and Northern Transfer A. Agarose gel electrophoresis B. Northern transfer Hybridization of RNA Affixed to Gene Screen Plus 5' End-labeling of 28S Oligonucleotide Hybridization of RNA Affixed to Gene Screen Plus with 288 Oligonucleotide Image Analysis of Autoradiographs A. Southern blot analysis. B. Northern blot analysis Protein Isolation vii 33 33 33 34 34 34 35 36 36 37 39 40 4O 43 44 45 46 47 48 48 49 49 49 50 51 51 52 53 53 54 54 16. 17. RESULTS 1. Polyacrylamide Gel Electrophoresis and Western Blotting A. Polyacrylamide gel electrophoresis B. Western blotting Detection of p21 by immunoblotting Methylation Status and mRNA Levels of Ha-ras in WB cell-induced Rat Liver Tumors A. Assessment of methylation status of Ha- ras in WB cells B. Assessment of methylation status offlHa- ras in F344 rats injected with WB ' m cells C. Assessment of Ha-ras mRNA and ras p21 protein levels Ha-ras and raf in the Nascent Liver of B6C3Fld, C3H/Hed, and C57BL/69 Mice A. RFLP screening B. Assessment of mRNA in nascent liver Ha-ras and raf Following Phenobarbital Administration or Partial Hepatectomy in BGCBFld, C3H/Hed, and C57BL/69 Mice A. Assessment of raf methylation status following partial hepatectomy B. Methylation status of Ha-ras 7 days after partial hepatectomy C. Methylation status of raf following 14 day phenobarbital treatment D. Methylation status of Ha-ras following phenobarbital administration. E. Assessment of raf and Ha-ras mRNA levels Ha-ras and raf in Phenobarbital-induced and Spontaneous B6C3Fld Liver Tumors A. Methylation status of raf B. Methylation status C. Assessment of raf and Ha-ras mRNA levels Ha-ras and raf in the Nascent Liver of the CD- 1d Mouse A. Methylation status of raf B. Methylation status of Ha-ras C. RFLP analysis of raf and Ha-ras in CD-ld viii 54 54 55 56 57 57 57 6O 64 64 73 81 81 88 97 102 105 105 105 111 118 123 123 123 127 6. Ha-ras and raf in Phenobarbital-induced CD-ld Liver Tumors A. Methylation status of raf B. Methylation status of Ha-ras C. Ha-ras and raf mRNA levels DISCUSSION 1. Factors Influencing the Regulation of raf Expression 2 . Correlation between Hypomethylation, Increased Expression and Tumorigenicity 3. Differences in the Mechanisms Involved in PB-induced and Spontaneous Tumorigenesis SUMMARY AND CONCLUSIONS LIST OF REFERENCES ix 127 127 134 137 141 141 147 149 154 157 LIST or TABLES Table 1. Methylation Status and mRNA Levels of raf and Ha-ras Following Cell Proliferation 89 Table 2. Methylation Status and mRNA Levels of raf and Ha-ras in BGCBFld and CD-1d Liver Tumors 113 LIST OF FIGURES Figure 1. Hepatocyte signal transduction pathway Figure 2. Illustration of the inheritance of alleles Figure 3. Illustration of the inheritance of methylation status Figure 4. Restriction endonucleases and recognition sites Figure 5. Schematic representation of probes Figure 6. The methylation status of Ha-ras: WB cells; MspI and HpaII digests Figure 7. The methylation status of Ha-ras: NE cells; HhaI digests Figure 8. The methylation status of Ha-ras: F344 rat liver and lung tumors; MspI and HpaII digests Figure 9. The methylation status of Ha-ras: F344 rat liver tumors; HhaI digests Figure 10. Ha-ras mRNA and p21: F344 rat liver tumors Figure 11. Restriction fragment analysis of raf: BSC3F1d, C3H/Hed, and C57BL/69; MspI Figure 12. MspI and HpaII double digests of C57BL/69 DNA; raf Figure 13. Restriction fragment analysis of raf: B6C3F1d, C3H/Hed, and C57BL/69; Tan, StuI, HindIII, ECORI Figure 14. The methylation status of the CCGG site in exon 12 of raf: B6C3Fld, C3H/Hed, andC57BL/69. Figure 15. Restriction fragment analysis of ‘Ha-ras: B6C3Fld, C3H/Hed' and C57BL/69: Tan, StuI, HindIII, EcoRI xi 23 24 38 41 58 61 62 65 66 68 7O 71 76 Figure 16. Restriction fragment analysis of raf and.Ha- ras: B6C3F1d, C3H/Hed, and C57BL/69; HhaI and XhoI Figure 17. Ha-ras and raf mRNA levels: B6C3Fld, C3H/Hed, and C57BL/69 Figure 18. The methylation status ‘of raf: B6C3Flo' 7 days after PH; MspI and HpaII Figure 19. The methylation status of raf: C3H/Heo‘ 7 days after PH; MspI and HpaII Figure 20. The methylation status of raf: C57BL/6Q 7 days after PH; MspI and HpaII Figure 21. Relative changes in raf band intensities following PH Figure 22. The methylation status of Ha-ras: B6C3Fld 7 days after PH; MspI and HpaII Figure 23. The methylation status of Ha-ras: C3H/Hed 7 days after PH; MspI and HpaII Figure 24. The methylation status of Ha-ras: C57BL/69 7 days after PH; MspI and HpaII Figure 25. Relative changes in Ha-ras band intensities following PH Figure 26. The methylation status of raf: B6C3F1d following 500 ppm PB for 14 days; MspI and HpaII Figure 27. Relative changes in raf band intensities following 500 ppm PB: B6C3Fld Figure 28. The methylation status of raf: C3H/Hed after 14 days of PB treatment; MspI and HpaII Figure 29. The methylation status of raf: C57BL/69 following 14 days of PB treatment; MspI and HpaII Figure 30. The methylation status of Ha-ras: B6C3F10‘ after 14 days of PB treatment; MspI and HpaII Figure 31. The methylation status of Ha-ras: C3H/Hed following 14 days of PB treatment; MspI and HpaII Figure 32. The methylation status of Ha-ras: C57BL/69 following 14 days of PB treatment; MspI and HpaII xii 78 80 82 83 84 86 90 92 93 95 98 100 103 104 106 107 108 Figure 33. The methylation status of raf: B6C3Fld PB- induced tumors and age-matched controls; MspI and HpaII Figure 34. The methylation status of raf: B6C3Fld spontaneous tumors; MspI and HpaII Figure 35. The methylation status of CCGG site in exon 12 of the raf gene in PB-induced and spontaneous B6C3Fld tumors Figure 36. The methylation status of Ha-ras: B6C3F1d’ PB-induced tumors and age-matched controls; MspI and HpaII Figure 37. The methylation status of Ha-ras: B6C3Flo' spontaneous tumors; MspI and HpaII Figure 38. Ha-ras and raf mRNA: B6C3Fld PB-induced and spontaneous tumors and age-matched controls Figure 39. Relative amounts of raf and Ha-ras mRNA in B6C3F1d PB-induced and spontaneous tumors Figure 40. The methylation status of raf: CD-ld; MspI and HpaII Figure 41. Relative intensity of the 6.7 kb band in the C57BL/6Q, B6C3F1d, and CD-ld raf gene Figure 42. The methylation status of Ha-ras: CD-lo‘; MspI and HpaII Figure 43. Restriction fragment analysis of raf and Ha- ras: CD-1d; Tan Figure 44. The methylation status of raf: CD-1d PB- induced tumors and age-matched controls; MspI and HpaII Figure 45. The methylation status of the CCGG site in exon 12 of the raf gene in CD-1d liver, PB-induced CD-ld tumors and age-matched controls Figure 46. The methylation status of Ha-ras: CD-1d PB- induced tumors and age-matched controls; MspI and HpaII Figure 47. The methylation status of Ha-ras: CD-1d PB— induced tumors and age-matched controls; MspI and HpaII xiii 109 112 114 116 117 119 121 124 125 126 128 130 132 135 136 Figure 48. Ha-ras and raf mRNA: CD-ld PB-induced tumors and age-matched controls 138 Figure 49. Relative levels of raf and Ha-ras mRNA in CD-ld PB-induced and spontaneous tumors 139 xiv 5MeC DNA EGF EGF-R LTR mRNA PB PH PKC PLC RFLP RNA rRNA TGF-a TGF-fi TPA LIST OF ABBREVIATIONS S-methylcytosine base pairs deoxyribonucleic acid epidermal growth factor epidermal growth factor receptor kilobases long terminal repeat sequence messenger ribonucleic acid phenobarbital partial hepatectomy protein kinase C phospholipase Cy restriction fragment length polymorphism ribonucleic acid ribosomal ribonucleic acid transforming growth factor-a transforming growth factor-8 12-O-tetradecanoylphorbol-13-acetate INTRODUCTION 1. An Overview of Multistep Carcinogenesis The development of cancer is a complex, multistep process. Carcinogenesis can be thought of as the progression of cells from a normal to an abnormal phenotype. At least 3 events must occur in the classical 3 stage carcinogenesis model: a genetic change for initiation, promotion by selective proliferation of initiated cells, and one or more additional genetic changes during progression (reviewed in Pitot and Sirca, 1980; Farber, 1984; Boyd and Barrett, 1990; Weinstein, 1988). Initiation, promotion and progression are not necessarily discreet steps; some chemical compounds can function as one or more steps in the carcinogenic pathway. Diethylnitrosamine, for example, when applied repeatedly, will result in the development of skin tumors in C57BL/69 mice without the application of additional chemicals (Reiners, et a1., 1984). 2. The Role of Proto-oncogenes in Carcinogenesis Proto-oncogenes were first discovered as the cellular homologs of the oncogenes present in transforming retroviruses. Proto-oncogenes are cellular genes which are generally involved in functions such as proliferation and 1 2 differentiation (reviewed in Bishop, 1991) . They can be categorized by the cellular compartment in which their protein product functions: transmembrane (e.g. erbB) , cytoplasmic (e.g. the ras family and raf), or nuclear (e.g. fos and jun). Alternatively, proto-oncogenes can be categorized by biochemical function. There are proto-oncogenes whose protein products function as protein tyrosine kinases (e.g. src, erbB) , protein serine threonine kinases (e.g. raf, mos), GTPases (e.g. the ras family), and transcription factors (e.g. fos and jun which, together, function as the transcription factor AP-l and enhance transcription of genes containing AP-l enhancer sites). Proto-oncogenes are termed activated oncogenes when they develop transforming ability by changes in gene expression or mutations in the protein products which remove the normal controls on their biochemical functions. While activation of a single oncogene is not generally capable of causing tumors by itself, two or more oncogenes can act together to transform cells. Generally, oncogenes which code for cytoplasmic components of signal transduction pathways (e.g. ras) complement oncogenes which code for nuclear proteins involved in cell proliferation (e.g. myc) (Land, et a1., 1983). A. The role of Ha-ras in carcinogenesis Activation of Ha-ras by point mutations, primarily in codon 61, has been implicated in liver tumors in the B6C3F1 mouse and the transforming capabilities and tumorigenicity of these DNA sequences (Reynolds, et a1. , 1987; Stowers, et a1. , 3 1987; Wiseman, et a1., 1986; Fox, et al., 1990). In the human Ha-ras gene, the mutation responsible for conferring transforming ability is most often in codon 12 (Taparowsky, et a1., 1982; Tabin, et al., 1982; Reddy, 1982). The T24 ras gene from a human bladder cell carcinoma contains a mutation at codon 12, and, when transfected into an immortal human epithelial cell line, caused transformation and the transfected cells were tumorigenic (Hurlin, et a1., 1989). Increased expression of Ha-ras has also been implicated in the transforming abilities of transfected DNA (Chang, et a1., 1982; Pulciani, et a1., 1985; Rimoldi, et al., 1991; Huber and Cordingley, 1988; Seyama, et a1., 1988). Chang and coworkers (1982) found that induction of high levels of c-Ha-ras mRNA by dexamethasone treatment of cells transfected with LTR-complexed c-Ha-ras DNA resulted in transformation. Increased expression of normal Ha-ras mRNA and. p21 due to transfection ‘with large amounts of DNA containing normal c-Ha-ras was shown by Pulciani and coworkers, (1985) to result in a transformed phenotype in NIH 3T3 cells; these cells were also capable of forming tumors when injected into NIH Swiss mice. 5-aza-2'deoxycytidine treatment of nontumorigenic interferon-induced revertants of NIH 3T3 cells transfected with LTR-activated cellular-(c-)Ha-ras resulted in increased expression of Ha-ras mRNA and exhibition of a transformed phenotype (Rimoldi, et a1., 1991). When injected into nude mice these cells gave rise to malignant and metastatic tumors. 4 Thus, it is apparent that activation of Ha-ras by point mutation and increased expression of the Ha-ras gene are each capable of causing a transformed phenotype, and these two mechanisms might act synergistically or additively to produce a qualitatively more transformed phenotype. Huber and Cordingley (1988) transfected rat epithelial cells with viral-(v-)Ha-ras (which is capable of transforming NIH 3T3 cells and contains mutations in codons 12 and 59) under the transcriptional control of the corticosteroid-inducible long terminal repeat (LTR) of the murine mammary tumor virus (MMTV). Low levels of v-Ha-ras mRNA were correlated with a transformed phenotype. However, the induction of expression by dexamethasone of this mutated ras gene caused the cells to exhibit more profound transformed characteristics such as increasingly bizarre cell morphology and larger colony formation than non-induced transfected cells. Many rodent liver tumors do not have detectible mutations in the Ha-ras gene. In recent studies, mutated Ha-ras was detected in 63-64% of spontaneous B6C3Fld liver tumors (Fox, et a1., 1990; Dragani, et a1., 1991) and 29% of spontaneous C3H/He tumors (Rumsby, et a1., 1991). Benzidine- induced liver tumors contained mutated Ha-ras at a rate of 59% in. B6C3Fld (Fox, et (a1., 1990) while. diethylnitrosamine induction resulted in mutated Ha-ras in 41% of C3H/He mice (Rumsby, et a1., 1991). It is reasonable to suspect that one or more other oncogenes might also be involved in carcinogenesis. S B. The role of raf in carcinogenesis The raf proto-oncogene is a good candidate for involvement in carcinogenesis because the raf gene product (Raf-1) plays a jpivotal role in -signal transduction. by reception of signals from membrane-associated growth factor receptors and transmission of these signals to nuclear targets (Rapp, 1991). Viral raf was originally characterized as the transforming oncogene present in the 3611-MSV (murine sarcoma virus) (Rapp, et al., 1983a). The v-raf protein is a fusion protein consisting of the amino-terminal 384 amino acids of the viral gag gene and.the carboxy-terminal 323 amino acids of mouse c-raf (Rapp, et al., 1983a; Rapp, et a1., 1983b). When transfected into NIH-3T3 cells, 3611-MSV DNA caused transformation and.these cells were tumorigenic when injected into newborn NFS/N mice. Transfection of immortalized human bladder epithelial cells with 3611-MSV DNA also resulted in malignant transformation (Skouv, et al., 1989). There are three known members of the raf proto- oncogene family. A-raf-l and B-raf are expressed in urogenital and brain tissue, respectively, while c-raf-l is expressed in all tissues (Storm, et a1., 1990). The raf oncogene is related by sequence homology to the tyrosine kinase class of oncogenes (v-src, v-fps, v-abl, and v-fes), but does not demonstrate any tyrosine kinase activity (Mark, et a1., 1984). Raf-1 is a serine/threonine kinase which is activated by phosphorylation (Moelling, et al., 1984). The 1nechanism.by which Raf-1 transmits its nuclear signal appears 6 to be mediated through the heterodimeric (composed of the protein products of the c-fos and c-jun proto-oncogenes) transcription factor AP-1 (Wasylyk, et al., 1989). Thus, activation of Raf-1 results in increased transcription of genes with AP-l enhancer elements. The cellular raf proto-oncogenes have transforming potential. For example, loss or substitution of the regulatory amino terminal region of raf genes results in transformation of cells (Ikawa, et al., 1988). Activated c-raf genes (i.e. capable of causing transformation in cells transfected with the tumor DNA) from human and rodent tumors are frequently found to be fused to other genes such that the amino-terminal portion of the c-raf gene is missing (Ishikawa, et a1., 1986; Stanton et a1., 1987; Heidecker, et a1., 1990). Increased expression of raf genes has been implicated in tumorigenesis. Cells cultured from tumors induced by injection of v-raf infected rat liver epithelial cells exhibited an increased level of v-raf mRNA relative to the injected cells (Worland, et a1., 1990). Elevated levels of raf mRNA have been found in PB-induced rat liver tumors (Beer, et a1., 1988).Increased expression of raf proto-oncogenes has been implicated in the transforming ability of raf genes by others, such as Beck, et a1. (1987) who showed that the normal A-raf—l gene was capable of transforming NIH 3T3 cells when over-expressed in a viral vector. 7 C. Signal transduction in hepatocytes Normal cell replication depends on the transduction of an external signal to the nucleus. In many transformed cells, however, cell proliferation. is uncoupled from the signal transduction pathway resulting in growth factor independence. The following discussion will focus on the events occurring after binding of the hepatocyte growth-stimulatory factors epidermal growth factor (EGF) (Richman, et a1., 1976) or ‘transforming’ growth factor-a (TGF-a) (Fausto and Mead, 1989) to the epidermal growth factor receptor (EGF-R) (reviewed in Velu, 1990) and the potential roles of ras p21 and Raf-1 proteins in hepatocyte signal transduction (Figure 1). EGF and TGF-a are endogenous ligands for the EGF-R; EGF-R is activated by ligand binding and functions as a tyrosine kinase. Autophosphorylation of the EGF-R on tyrosine or serine phosphorylation by protein kinase C (PKC) leads to internalization and degradation of EGF and EGF-R. The EGF-R tyrosine kinase also phosphorylates other cellular proteins, including phospholipase Cy (PLC) which then becomes activated. The activation of PLC is thought to be mediated through a G-protein. The p21 protein product of the ras family of proto-oncogenes (Ha-ras, Ki-ras, and N-ras) acting at the cytoplasmic face of the cellular membrane, binds and hydrolyzes GTP. Thus, p21 appears to act in a manner analogous to the alpha subunit of the G-proteins in signal transduction (Sigal, et a1., 1988) and might be involved in GROWTH FACTOR e.g. EGF 1 GROWTH FACTOR RECEPTOR g TYROSINE KINASE Raf-1 (4:; ras p21 4 V PLC SERINE THREONINE KINASE '\'=._ , ‘1! PKC->PKC* Raf-1* NUCLEAR EFFECTS e.g ACTIVATION OF fos, jun Figure 1. Hepatocyte signal transduction pathway. * denotes activated form of kinase. 9 the activation of PLC. PLC catalyzes the hydrolysis of phosphatidylinositol 4,5 bisphosphate into inositol 1,4,5 triphosphate, which functions to release intracellular stores of Ca”, and diacylglycerol, which activates protein kinase C (PKC) . Phorbol ester tumor promoters [e.g. 12-0- tetradecanoylphorbol-13-acetate (TPA)] can substitute for diacylglycerol and activate PKC (Nishizuka, et a1., 1986; Blumberg, 1991). The serine/threonine kinase function of Raf-1 can be activated by phosphorylation on a tyrosine residue (Carroll, et a1. , 1990; Morrison, et a1. , 1989) . However, phosphorylation in response to EGF has been demonstrated to occur on serine, not tyrosine residues (App, et a1., 1991). The most probable explanation for this apparent discrepancy is the presence of an intermediate serine/threonine kinase in the pathway between the EGF-R and Raf-1. In other words, stimulation of the EGF-R results in the activation of a serine/threonine kinase whose target is Raf-1. One candidate for the this intermediate kinase is PKC. T cell antigen mediated Raf-1 activation is dependent on phosphorylation by PKC (Siegel, et a1., 1990). Also, phorbol 12-myristate 13- acetate was shown to increase the activity of the raf protein through its action on PKC (Morrison, et a1. , 1988) . However, EGF stimulates serine phosphorylation of Raf-1 in cells which have been depleted of PKC (App, et a1. , 1991) , suggesting the presence of an as-yet-unidentified serine/threonine kinase. Another possibility for this intermediate serine/threonine 10 protein kinase is mitogen activated protein kinase (MAP-1 kinase) , as suggested by the finding that MAP-1 kinase can phosphorylate Raf-1 in vitro and MAP-1 kinase can be activated by tyrosine and threonine phosphorylation (Anderson, et a1., 1991). 3. Carcinogenesis in Mouse Liver A. The B6C3F1d mouse The hybrid B6C3F1o‘ mouse, along with the parental strains, C3H/Hed and C57BL/69, is a good model to study the molecular mechanisms involved in the conversion of phenotypically normal cells to cancer cells. The 86C3Flo‘ spontaneously develops liver tumors at a rate of approximately 30% (Becker, 1982, Buchmann, et a1., 1991). This is intermediate between the spontaneous liver tumor incidences of its parents, the C3H/Heo’ at approximately 60% and the C57BL/6t1> with a negligible rate (Buchmann, et a1. , 1991) which suggests a heritable factor influencing spontaneous tumor development. The susceptibility of the different strains (only regarding the sexes used in this study) to chemically-induced liver tumors is comparable to the incidence of spontaneous tumor development; B6C3F1d and C3H/Hed are very sensitive to the chemical induction of liver tumors in response to both mutagens and many nonmutagenic compounds such as trichloroethylene (Clayson, 1987; Maronpot, -et a1., 1987; Ashby and Tennant, 1988) while the C57BL/69 is relatively resistant (Buchmann, et a1., 1991). The B6C3F19 develops 11 fewer spontaneous liver tumors than the B6C3Fld, but is equally, or even more sensitive to the chemical induction of liver tumors. Clearly, there are one or more characteristics of the C3H/He which cause it to be more prone to liver tumor development than the C57BL/6. In histologically normal primary hepatocyte cultures from C3H/He mice, twice as many cells are synthesizing DNA (measured as 3H-thymidine incorporation) as in hepatocytes derived from C57BL/6 mice (Hanigan, et a1., 1988). Primary' C3H/HeNJcl hepatocyte cultures spontaneously develop multiple colonies, while C57BL/6NJcl hepatocyte cultures only rarely develop colonies (Lee, et a1. , 1989) . In a study of diethylnitrosamine-induced preneoplastic lesions in the livers of C3H/HeN<->C57BL/6N chimeric mice, it was found that relative to C57BL/6N cells, more (5x) C3H/HeN hepatocytes developed lesions and altered C3H/HeN cells exhibited a more malignant phenotype (Lee, et a1. , 1991) . There are intrinsic differences between the cells of the two strains which have the potential to influence susceptibility to liver tumors. One difference that has been found is that C3H/He, but not C57BL/6 mice possess the Hcs (hepatocarcinogen sensitivity) gene (Drinkwater and Ginsler, 1986). Another identified difference is a variation in methylation status. In comparison to the C3H/Heo' and B6C3F1d, the C57BL/69 possesses at least one additional methylated site (5'-"°CCGG-3') in the Ha-ras gene in the nascent liver (Vorce and Goodman, 1989a). 12 B. The CD-lo' mouse The CD-1o’ mouse is also a good model in which to study mouse liver carcinogenesis. The CD-ld develops spontaneous liver tumors at a rate similar to that of the B6C3F1d (Hoffmann LaRoche Laboratories, unpublished data) and is not related to either the C3H/He or C57BL/6. The CD-l HaM/ICR mouse stock is a non-inbred subcolony of the minimally inbred (0.5% per generation) Icr:Ha(ICR) mouse stock (Eaton, et a1., 1980) which was developed from inbred Swiss mice (Lynch, 1969). 4. The Involvement of Cell Proliferation in Carcinogenesis A. The role of cell proliferation in carcinogenesis Cell proliferation plays a major role in carcinogenesis (Ames and Gold, 1990; Cohen and Ellwein, 1990) . Epidemiologically, chronic mitogenic stimuli have been implicated in numerous human cancers (e.g. papilloma.virus in cervical cancer, alcohol in liver cancer, cigarette smoke in lung cancer) (Preston-Martin, et a1. , 1990) . Carcinogen bioassays involve the administration of chemicals at the maximum tolerated dose. This high dose can cause cell death and thus act as a stimulus for mitogenesis, in a manner similar to chronic wounding and.wound repair. ,Alternatively, some noncytotoxic compounds, such as PB, stimulate cell proliferation without causing cell death. Under conditions in which an increased number of cells are undergoing mitogenesis at the same time, there is a higher probability that errors 13 will occur. Increased demand for DNA repair enzymes, maintenance methylase, DNA binding proteins, and cofactors might alter DNA primary, secondary, or tertiary structure which might in turn alter gene function or expression and contribute to carcinogenesis. B. Partial hepatectomy and cell proliferation Partial hepatectomy (PH; surgical removal of 2/3 of the liver volume) stimulates regeneration of liver cells until the original liver mass is attained. In hepatocytes from B6C3Fld mice, 48 hours after PH, DNA synthesis is increased 14-fold over hepatocytes from an intact liver (Itze, et a1., 1973). The peak DNA synthesis occurs at approximately 30-36 hours post-PH in B6C3Fld mice (Itze, et a1., 1975), approximately 40 hours post-PH in B6AFld and 46 hours post-PH in B6AF19 mice (Chernozemski and Warwick, 1970) . Diurnal rhythm influences timing of DNA synthesis even after a major stimulatory event such as PH. Therefore, the time of day at which a PH is performed will influence the time of peak DNA synthesis. Both raf and Ha-ras mRNA levels increase following PH. In the rat [peak DNA synthesis following PH in the rat is 24 hrs (Goyette, et a1., 1983)], raf mRNA levels peak at 3-5x basal levels 24 hours after PH and return to baseline by 72 hours (Silverman, et a1., 1989). Ha-ras mRNA levels are maximal (2-3x basal levels) between 18-36 hours after PH in the rat and have returned to normal levels by 72 hours (Goyette, et a1., 1983; Silverman, et a1., 1989). 14 C. Phenobarbital and cell proliferation Phenobarbital (PB) is a liver tumor promoter in rats (Peraino, et a1., 1980; Pereira, et al., 1986a; Pitot, et a1., 1987) and.mice, including the B6C3F1.(Pereira, et a1., 1986b; Klaunig, et a1., 1987; Klaunig, et a1., 1988), and C3H but not the C57BL (Lee, et a1., 1989). Furthermore, PB (500 ppm in drinking water for 18 months) alone results in 100% liver tumor incidence in B6C3F1d‘and C3H/fled, but.not.C57BL/69, mice (Becker, 1982). In aging (12 month old) C3H mice, administration of 500 ppm PB in the drinking water for 12, 24, or 36 weeks was found to increase the size and number of foci and neoplasms arising from alleged spontaneously initiated hepatocytes (Ward, et a1., 1988). In the presence of 2 mM PB, there was a ten-fold increase in the number of proliferative hepatocyte colonies which developed from initiated [by administration of methyl(acetoxymethyl)nitrosamine or benzo[a]pyrene-7,8-diol-9,10-epoxide(anti)]ratcellsrelative to control (not initiated, but treated with PB) cells (Kaufmann, et a1., 1986; Kaufmann, et a1., 1988). PB administration results in a dose dependent increase in DNA synthesis as measured by' 3H-thymidine incorporation in DNA in primary hepatocyte cultures obtained from rats [Edwards and Lucas, 1985 (2 mM PB for 46 hours); Sawada, et a1., 1987 (maximum increase with 1 mM PB for 44 hours); Yusof and Edwards, 1990 (2 mM PB; maximum DNA synthesis at approximately 25 hours after addition of PB) ] , in hepatocytes obtained from PB-treated rats [Eckl, et a1., 1988 15 (0.1% PB in the drinking water for up to 30 days), in vivo in rats [Busser and Lntz, 1987 (1 dose of 0.1 mmole/kg PB 24 hours prior tatsfi-thymidine incorporation)], and in C57BLd, C3Ho‘, B6C3Fld, and C386Flo‘ mice [Lin,. et al. , 1989 (500 ppm PB in the drinking 'water for’ 4 days - all strains, or 28 days - C57BL/69 and C3B6Fld) ] . In B6C3Flo‘ mice, 500 ppm, but not 20 ppm, PB in the drinking water stimulates DNA synthesis; this effect peaks at 2 weeks, then declines (Weghorst and Klaunig, 1989; Siglin, et a1., 1991; Klaunig, et a1.; 1991; J.E. Klaunig, unpublished data). Chronic PB treatment (more than one month) in vivo has an inhibitory effect on hepatocyte proliferation (Eckl, et a1., 1988). PB has additional effects on cell physiology. PB treatment results in increased unscheduled DNA synthesis (i.e. DNA repair) following initiation in cultured rat hepatocytes (Althaus, et a1., 1986). PB reduceS' gap junctional communication in a dose dependent manner (Ruch, et a1., 1987; Klaunig and Ruch, 1987; Klaunig, et a1., 1990) which might contribute to unregulated cell proliferation. Hepatocytes isolated from PB pre-treated rats (0.1% PB in the drinking water for 2 weeks) showed a 15-fold increase in clonogenicity relative to hepatocytes from non-PB treated rats when injected into the fat pad of recipient rats (Jirtle and Michalopoulos, 1986). The inhibitory effect of physiological concentrations of extracellular calcium is eliminated during the first month of PB treatment (Eckl, et a1., 1988). Hepatocyte EGF-R mRNA (Hseih, et a1., 1988 - 0.05% PB for 16 days) or number (Eckl, 16 et a1., 1988 - 0.1% PB for 2 or 8 weeks) is decreased in PB treated rats. 5. The Role of 5-Methylcytosine in Mediating Gene Expression A. The role of 5-methylcytosine in differentiation and aging S-Methylcytosine (5MeC) is the only naturally occurring methylated base in mammalian DNA. 5MeC plays an important role in development and differentiation (reviewed in Michalowsky'and.Jones, 1989b; Cedar and Razin, 1990; Razin and Cedar, 1991) . DNA in sperm is heavily methylated, while oocyte DNA is unmethylated. The early gamete has an intermediate methylation pattern, suggesting a mixture of methylated and unmethylated DNA. By the time the blastocyst stage is reached, the DNA is unmethylated. In the embryo, the genome is generally heavily methylated, with genes becoming hypomethylated only in the tissues in which they are expressed. Thus, as cells differentiate into specific tissues, the DNA undergoes active demethylation to activate genes and de novo methylation to inactivate genes resulting in the appropriate phenotypic expression. Undifferentiated cells (e.g. transformed cells) can be stimulated to differentiate by treatment with a hypomethylating agent such as 5-azacytidine (Creusot, et a1., 1982; Jones and Taylor, 1980). Conversely, DNA which is unmethylated in normal cells might become methylated as the cell dedifferentiates during transformation. Jones and 17 coworkers (1990) found that the normally unmethylated CpG island upstream of the MyoDl gene is methylated in immortalized cell lines. There is a decrease in 5MeC content in the genome in general in aged animals (Wilson, et a1., 1987b; Singal, et a1. , 1987) . Aging and subsequent hypomethylation of DNA might predispose animals to carcinogenesis. There is also a decrease in overall 5MeC content with time in vitro as primary cell cultures senesce and the rate of loss correlates with the lifespan of a particular cell culture (Wilson and Jones, 1983a) . Specific genes have been shown to become hypomethylated during aging. Mouse intracysternal A particle genes have lost both 5'-C"°CGG-3' and 5'-"°CCGG-3' sites, seen as increased digestion by both MspI and HpaII in DNA from 24- 26 month old C57BL/6d mice (Mays-Hoopes, et a1., 1983). The major mouse long interspersed sequence was found to have lost approximately 8% of its 5'-C"°CGG-3' sites in C57BL/6J by the age of 27 months (Mays-Hoopes, et a1., 1986). While the genome, in general, becomes hypomethylated with aging, some specific genes have been found to be hypermethylated in older animals. The 5' spacer region and external transcribed spacer of 18S and 288 rRNA genes in liver, brain, and spleen are more methylated in 18 month old than 6 month old CBA/Ca mice (Swisshelm, et a1. , 1990) . This hypermethylation is accompanied by a decrease in detectable activity of rRNA genes in some chromosomes which is restored by treatment with 5-azacytidine. While the methylation status 18 of the 5'-flanking region, first exon and first intron of the mouse c-fos gene remains constant from the age of one month, a site in exon 2 is partially methylated at one month and a progressively greater fraction of cells have this site in its methylated. state in older' mice (Uehara, et a1., 1989). Relative to 2 month old mice, the c-myc gene in 26 month old mice was found to be hypermethylated in liver DNA at sites 3' to the first intron (Ono, et a1., 1986; Ono, et a1., 1989). Following digestion of DNA with MspI, additional large fragments are detected, indicating that the hypermethylation of the myc gene is due to an increase in 5'-mCCGG-3' sites. This hypermethylation of myc was correlated with decreased levels of myc mRNA in the liver. B. 5-Methylcytosine as a regulator of gene expression Modification of DNA by methylation of cytosine is one mechanism postulated to be involved in the regulation of gene expression. The methylation status of a gene has been found to inversely correlate with its expression. In general, quiescent genes are hypermethylated while genes which are actively being expressed or have the potential to be expressed are :relatively' hypomethylated (Doerfler, 1983; Riggs [and Jones, 1983; Jones, 1986; Cedar, 1988) . Hypomethylation might facilitate the binding of transcription factors, thereby resulting in increased expression of the gene. The presence of cytosine rather than 5MeC results in a local destabilization of the double helix structure (Murchie and Lilley, 1989). The a1globin and MyoDl genes in C3H 10T1/2 19 cells treated with 5-aza-2'deoxycytidine were hypomethylated and the chromatin was more accessible to nuclease digestion than in control cells (Michalowsky and Jones, 1989a), suggesting a more open chromatin structure. 5MeC can also negatively influence gene expression. The binding of some specific transcription factors has been found to be physically hindered by 5MeC (Watt and Molloy, 1988; Comb and Goodman, 1990; Antequera, et a1., 1989; Boyes and Bird; 1991). For example, Boyes and Bird, (1991) found that methylation of the murine myeloproliferative sarcoma virus promoter region had a direct inhibitory effect on transcription. Additionally, hemimethylated (i.e. 5MeC present in only one strand) sites were found to partially inhibit binding of the MLTF transcription factor in the adenovirus major late promoter region (Watt and Molloy, 1988) . Another mechanism by which 5MeC interferes with transcription is by binding proteins (e.g. the methyl CpG binding protein) which inhibit transcription (Meehan, et a1., 1989; Levine, et a1., 1991; Boyes and Bird, 1991). A.minimum number of 5MeC in a gene is required.for'binding the inhibitory proteins and the binding affinity of the inhibitory methyl CpG binding protein is increased in the presence of additional 5MeC (Boyes and Bird, 1991). Thus either the absolute number of methylated sites in a gene or the methylation status of a critical site involved in the binding of transcription factors can influence expression of a gene. 20 The methylation status of sites in the promoter and 5' flanking region of the gene is critical (Bird, 1986; Langner, et a1., 1984), but the methylation status of sites in the interior region of a gene has also been found to influence expression (Dizik, et a1., 1991; Langner, et a1., 1984). Langner and coworkers (1984) found that in vitro methylation of three 5'-CCGG-3' sites in the promoter and 5' flanking region of the human adenovirus type 2 E2a gene resulted in transcriptional inactivity. In vitro methylation of 11 5'-CCGG-3' sites in the coding region of the gene resulted in a decreased level of transcription (Langner, et a1., 1984). Therefore, a decrease in methylated sites in the interior of a gene might allow increased transcription. The 5'-CCGG-3' sites in the CG rich 5' flanking region of c-myc are unmethylated in normal rat liver DNA but the regions including the second and third exons are heavily methylated (Dizik, et a1., 1991). Following 1-4 weeks of a methyl-deficient diet, hypomethylation of the coding and 3' flanking regions was detected and c-myc mRNA levels were increased. More than 90% of 5MeC occur in the sequence 5'-CG-3' (Riggs and Jones, 1983). However, 5MeC at other sites can also influence gene expression. Kong and coworkers (1991) found a decrease in methylation of the external, but not the internal, cytosine of 5'-CCGG-3' sites after treatment of cells with 5-azacytidine, a potent DNA maintenance methylase inhibitor, which resulted in the induction of expression of a previously quiescent gene. 21 C. The inheritance of methylation status 1) Somatic The methylation status (i.e. number and position of 5MeC) of a gene is an epigenetic characteristic which is heritable through multiple rounds of cell division (Holliday, 1987; Holliday, 1990). There is a difference between maintenance methylation which results in the methylation of’hemimethylated sites following DNA replication and the de novo methylation which takes place during differentiation. During DNA replication, an unmethylated daughter strand is synthesized using a parental strand as a template. CpG sites which were fully methylated (i.e. 5MeC on each strand) are now hemimethylated. DNA maintenance methylase recognizes this hemimethylated site and catalyzes the transfer of a methyl group from the cofactor S-adenosyl-methionine to the unmethylated cytosine (Wigler, et a1., 1981). Fully unmethylated sites are poor substrates for DNA maintenance methylase. If there is some interference in the activity of the DNA maintenance methylase, such as inhibition of the enzyme, or decreased availability of the cofactor and the DNA replicates prior to methylation, a fully unmethylated site could result. Because some regions of chromatin are less accessible than.others to DNA repair enzymes (Topal, 1988), it is reasonable to suspect that some hemimethylated sites might also be relatively inaccessible to maintenance methylase. 22 This scenario also has the potential to result in heritably unmethylated sites following DNA replication. 2) Through the germ line Both the base sequence of.a gene and secondary modifications of DNA such as methylation are heritable through the germ line. Alleles are variants in DNA base sequence at a single genetic locus. When the difference in base sequence results in a change in the cleavage pattern of restriction endonucleases, restriction fragment length polymorphisms (RFLP) are detectible with Southern blot analysis. The inheritance of alleles as detected by the presence of RFLP is diagrammed in Figure 2. A variation in methylation at a genetic locus is not synonymous with different alleles because the base sequences at that locus might be identical. However, differences in methylation status which affect cleavage by methylation sensitive restriction endonucleases will also be seen as RFLP. An example of the inheritance of methylation status through the germ line is diagrammed in Figure 3. The methylation pattern at a specific genetic locus was found to be inherited in a Mendelian fashion (Silva and White, 1988; Ghazi, et a1., 1990; Chandler, et a1., 1987) Silva and White (1988) examined the methylation status of MspI sites in ten loci in related individuals. The methylation status of a particular allele was found to persist through at least 3 generations regardless of the sex of the parent through which the allele was transmitted. Differential methylation of 23 I) STRAIN A d STRAIN B 9 (homozygous) (homozygous) l l l 1 l " I—llkb kb 52 5"— 3-— 3 — 3 L = restriction site ' -- = probe binding site Figure 2. Illustration of the inheritance of alleles. I) The physical maps of a hypothetical genetic locus in two homozygous strains of mice are shown with recognition sites of a restriction endonuclease (_L) . The Southern blot showing the restriction pattern of the DNA probed for this gene is drawn below the maps. II) The physical map of the parental alleles in the F generation is shown along with a Southern blot illustrating the restriction pattern. 24 Figure 3. Illustration of the inheritance of methylation status. I) The physical maps of a hypothetical genetic locus in two homozygous strains of mice are shown. The restriction endonuclease used to generate the map is methylation sensitive, i.e. does not cleave when .the recognition site contains a 5MeC (.L = unmethylated site, I = methylated site). A Southern blot showing the restriction pattern of the DNA probed for this gene is depicted below the map for each strain. II) The physical map of the genetic locus inherited from each parent in the F1 generation is pictured along with the Southern blot showing the methylation sensitive restriction pattern. III) Southern blots illustrating the restriction patterns of Y x 2 F1 DNA in which one or more methylated sites have become unmethylated. 25 I) STRAIN Y 6 STRAIN Z 9 (homozygous) (homozygous) a b c d e a'b'c'd'e' 9!?9! 9!!9! ’ Hlkb 1gb 1m 3..— 2-— II) Y x z = F1 a b c d e 39 j!??! 3_ a'b'c'd'e' 2- ._._. 9!!9! III) Q 1512 Kb 3' -w- 2—* 2-— 1—— 1— LOSS OF Me LOSS OF Me LOSS OF Me AT SITE b AT SITE c' AT SITE b' 26 parental Ha-ras alleles was found in tissues (Ghazi, et al.1990) and in an immortal cell line (Chandler, et al.1987). 3) Partial methylation .A difference in the methylation status at a genetic locus between cells in a tissue (e.g. hepatocytes) is referred to as partial methylation at a site or mosaicism of methylation status (McGowan, et a1., 1989). Partial methylation is generally expressed as the percent of the specific loci which are methylated in a given cell population. Differential methylation of the two parental alleles (i.e. the maternal allele is methylated and the paternal allele is not) would result in a methylation value at that site of 50%. In a gene which is 25% methylated at a site, one allele out of every four would be methylated at that site (e.g. the maternal allele in every second cell is methylated and the paternal alleles are all unmethylated) . While conclusive-determination of methylation status of a particular site requires sequencing, partial methylation might be indicated by differing intensities of bands on a Southern blot of a DNA sample digested with a methylation sensitive restriction endonuclease. Decreased intensity of one band relative to the other bands in the same lane is not diagnostic of partial methylation, but partial methylation at a site, resulting in partial digestion by a methylation sensitive restriction endonuclease, is one possible explanation for decreased intensity of a band. 27 Partial methylation of a site is heritable both somatically and through the germ line. The methylation pattern of a CpG site 420 bp upstream of the mouse adenine phosphoribosyltransferase gene is stably maintained between different individuals (i.e. it is 44% methylated in liver DNA of CBA mice) and is not altered with aging (Turker, et a1., 1989) . The degree of methylation at partially methylated sites has been found to influence the level of expression of that gene. MCGowan, et a1. (1989) studied the methylation status of the 1acZ' transgene in neural tube tissue of transgenic mice. The lacz transgene was found to be in a highly methylated, unmethylated or intermediately methylated state in individual transgenic mice. The level of lacz gene expression was inversely correlated with the methylation state of the gene. Mice exhibiting an intermediate methylation pattern of the transgene lacz were found to express an intermediate amount of lacz. This was interpreted as indicating a mosaic pattern of expressing and nonexpressing cells. 4) Imprinting Imprinting refers to epigenetic modifications such as the relative hypermethylation of one parentally- derived allele that result in differential expression of genetic material, depending on the parent of origin. Imprinting by hypermethylation is thought to be responsible for the silencing of the inactive paternally—derived X chromosome. Genomic imprinting has been implicated in some 28 cancers such as familial retinoblastoma and osteosarcoma (Hall, 1990; Sapienza, 1991). In accordance with the theory that.methylation plays a role in imprinting, the methylation status of the inserted gene (transgene) in the offspring of transgenic mice has been found to be dependent on the sex of the parent through which the transgene was transmitted (Sapienza, et a1. , 1989; Swain, et a1., 1987; Reik, et a1., 1987; Sapienza, et a1., 1987). In most of the studies, the transgene was not expressed regardless of its methylation status, but transgenes transmitted through the female parent were consistently hypermethylated relative to the same transgene at the same locus transmitted through the male parent. However, if the c-myc and the Rous sarcoma virus LTR transgene was inherited from the male, it was hypomethylated and expressed while it was not expressed if inherited (relatively hypermethylated) from the female (Swain, et a1., 1987). D. DNA hypomethylation associated with carcinogen treatment A major factor in multi-step carcinogenesis is the activation of previously quiescent genes, such as a-fetoprotein, which is expressed in some preneoplastic foci, but not in normal liver. Hypomethylation has the potential to alter the expression of genes. The Ha-ras proto-oncogene in B6C3Fld mouse liver tumors induced by benzidine, chloroform, or phenobarbital, or in spontaneous mouse liver tumors is hypomethylated relative to surrounding nontumorous tissue 29 (Vorce and Goodman, 1989a; Vorce and Goodman, 1989b) and Ha-ras mRNA levels are elevated in benzidine-induced tumors (Vorce and Goodman, 1989a). The level of DNA methylation is decreased in many cancer cells and tumor tissues (Jones and Buckley, 1990). This may be due to the hypomethylation seen during aging, indeed, cancer is more prevalent in aged individuals. Alternatively, hypomethylation can be a result of chemical exposure. Treatment of normal dividing human bronchial epithelial cells with chemical carcinogens with diverse mechanisms of action was shown by Wilson, et al. (1987a) to significantly decrease the total 5MeC content of cellular DNA. Chemicals which form DNA adducts such as acrolein (Cox, et a1., 1988) and antibenzo[a]pyrenediol epoxide (Ruchirawat, et a1., 1984), have been found to interfere with the activity of DNA maintenance methylase. The DNA adduct complexes might physically interfere with the enzyme's access to ‘the Ihemimethylated sites or some chemicals, such as N-methyl-N-nitro-N-nitrosoguanidine or ethylnitrosourea form adducts on the enzyme itself (Wilson and Jones, 1983b). When incorporated into DNA, the cytidine analog 5-azacytidine binds and inhibits DNA maintenance methylase (Santi, et a1., 1984). Ethionine, a nongenotoxic carcinogen which is an antimetobolite of methionine, interferes with methylase function.by competing with the cofactor S-adenosyl-methionine (Shivapurkar, et a1., 1984). There is‘also the possibility of active demethylation resulting in the occurrence of 30 demethylated sites on a gene (Gjerset and Martin, 1982; Razin, et a1., 1986). This action would be due to one or more demethylating enzymes, would not be a function of DNA maintenance methylase, and could occur in the absence of cell proliferation. 5MeC might also play a role in mutagenesis. Mutations are a major factor in carcinogenesis. Both initiation and progression frequently involve alterations in DNA base sequence. Methylated cytosines might protect DNA from mutations that occur during DNA replication by directing repair of a mismatched base to the newly synthesized unmethylated strand, thus preserving the parental sequence (Hare and Taylor, 1989). Conversely, the presence of 5MeC might increase mutations because spontaneous deamination of 5MeC results in a change from cytosine to thymine (Rideout, et a1., 1990). This possibility is supported by the fact that actual CpG dinucleotide content in most of the‘genome is lower than predicted statistically (Schorderet and Gartler, 1990). In contrast, GC-rich islands are generally unmethylated (Bird, 1986) and might thereby be protected against mutations. 6. Hypothesis and Objectives The hypothesis underlying this study is that specific proto-oncogenes are hypomethylated in mouse liver tumors and that this hypomethylation results in elevated levels of the proto-oncogene mRNA. Hypomethylation is one mechanism by which transcription from particular genes might be 31 facillitated. Hypomethylation of proto-oncogenes has been found previously in tumors and increased expression of proto- oncogenes has been implicated in rodent liver tumors. The specific proto-oncogenes examined in this study were Ha-ras and raf because of their role in signal transduction and involvement in rodent liver tumors. The B6C3Flo', its parental strains, the C3H/Heo‘ and the C57BL/69, and the CD-lo‘ were chosen as the models in which to study alterations in raf and Ha-ras methylation and expression because of the differing susceptibilities to liver tumor development of each strain. My first aim was to determine if there are intrinsic differences in the primary structure or methylation patterns of raf and Ha-ras between the C3H/Heo' and C57BL/69. The methylation status of the genes was analyzed by digestion with methylation sensitive restriction endonucleases, and the genes were screened for the presence of RFLPs by digestion with several other restriction endonucleases to detect differences in base sequence. The same parameters were examined in the B6C3Flo‘ to determine if any differences found were hereditary. Additionally, raf and Ha-ras methylation and RFLP patterns were examined in the unrelated CD-1d strain. Secondly, the effect of cell proliferation on the maintenance of methylation patterns of raf and Ha-ras was examined. PH or PB (500 ppm in the drinking water for 14 days) were used in separate groups of B6C3F1d, C3H/Hed, and C57BL/69 mice to increase the liver cell growth fraction. The methylation status and mRNA levels of raf and Ha-ras from each 32 of these groups was measured to examine the relationship between changes in methylation status and expression of the genes. The third aim of this study was to examine the relationship between tumorigenicity, hypomethylation, and mRNA levels. The methylation status of v-Ha-ras, Ha-ras mRNA and ras p21 levels were examined in rat liver tumors induced by injection of v-Ha—ras—infected rat liver epithelial cells (WBflrm cells). Although the presence of the viral Ha-ras DNA might be the major factor governing the tumorigenicity of the transfected cells, it was important to determine the relationship between the methylation status of this DNA and its expression. The methylation status and mRNA levels of raf and Ha-ras in B6C3Fld and CD41d liver tumors was analyzed to determine if there was a relationship between hypomethylation of proto-oncogenes, expression, and tumor formation in vivo. Finally, to determine if PB-induced liver tumors were due to an acceleration in the development of spontaneous tumors or if PB administration resulted the clonal expansion of a different subset of cells, the methylation status and mRNA levels of raf and Ha-ras in PB-induced B6C3F1d liver tumors were compared with those of spontaneous B6C3Fld' liver tumors. MATERIALS AND METHODS 1. Animals: Maintenance and Treatment A. Mice Four strains of mice were used in these studies: the B6C3F1, C3H/He, C57BL/6, and CD-1. Young adult B6C3F1d, C3H/Heo' and C57BL/69 mice (19-20 g) were obtained from Charles River Laboratories (Portage, MI). Male mice at 19-20 g were generally 4-5 weeks old, while female mice were approximately one to two weeks older. The mice were housed in a university laboratory animal care facility at constant temperature (70°F) and humidity (35-40%) with a reverse phase 12 hour light/dark cycle and allowed food and water ad libitum.. Prior to any treatment or sacrifice, mice were acclimated to the environment for at least 6 days. All mice were euthanized by cervical dislocation. Following sacrifice, the entire liver was placed in liquid nitrogen, then stored at -80°C until use. The PB-induced B6C3Flo' and CD—ld liver tumors were provided by from Hoffman LaRoche Laboratories (Nutley, NJ). To induce the tumors, mice were treated with 1000 ppm PB (150 mg/kg/day) in the diet for 24 months. Spontaneous B6C3Fld liver tumors, age-matched B6C3Fld and CD—1d liver tissue and young CD-1d liver tissue were also provided by Hoffman LaRoche Laboratories . 33 34 B. Cells One study involved cultured cells. WBHt cells are immortal rat liver epithelial cells. Cells were infected with the viral vector (pZip) containing the neomycin marker alone Ha-ras) (WB"m) or with pZip containing v-Ha-ras DNA (WB . .All WB cells were provided by Dr. James Trosko. Tumors were induced Ha- ras cells in Fischer 344 (F344) rats by injection of 1 x 10‘5 WB into the portal vein. Rats were sacrificed after 3 weeks. Liver and lung tissue from an untreated rat, liver and lung 0 o H ' tumors and nontumorous liver tissue from WB ' "3 -treated rats, and.cells cultured from F344 liver tumors were provided by Dr. James Klaunig. C. Phenobarbital administration In the 14 day PB administration studies, the PB treated mice (B6C3Fld, C3H/Bed, or C57BL/69) were given 0.002% (3 mg/kg/day) or 0.05% (75 mg/kg/day) (w/v) PB (free acid, Sigma Chemical Co., St. Louis, MO) in the drinking water. PB solutions were made fresh weekly and replaced twice weekly. Control animals were given distilled water. D. Partial hepatectomy PH were performed on B6C3F1d, C3H/Hed, or C57BL/69 mice under ether anesthesia, modified from the procedure described by Higgins and Anderson (1931). A 1-1.5 cm ventral midline incision was made in the anesthetized mouse and the left and middle liver lobes were exteriorized. The lobes were ligated with 4-0 silk and resected. Removed liver was immediately placed in liquid nitrogen, broken into smaller 35 pieces and then stored at -80°C until use. The incision was closed.with 4-0 silk in two layers. The animals were allowed to recover several hours at approximately 80°F by placing the cage about 24" under an incandescent light before returning the mice to the laboratory animal care facility. Viability following surgery ranged from <50% of the C57BL/69 to nearly 100% of the B6C3Flo‘. Mice which were moribund or visibly jaundiced 7 days after PH were not used in these studies. 2. Isolation of DNA DNA was isolated from frozen tissue or frozen tissue culture cells by a modification of the method of Strauss (1987). Approximately 100-300 mg of liver tissue was ground in liquid nitrogen, then homogenized in 5 ml of 100 mM sodium chloride (NaCl), 10 mM Tris (pH 8.0), 25 mM ethylenediaminetetraacetic acid (EDTA, pH 8.0), 0.5% sodium dodecyl sulfate (SDS) with 200 ug/ml Proteinase K (Boehringer Mannheim Biochemicals; BMB, Indianapolis, IN; Proteinase K is prepared as a 10 mg/ml solution in 50 mM Tris, 1 mM CaCLfl. Digestion proceeded overnight at 50°C. DNA was washed with an equal volume of a mixture of 50% equilibrated phenol [phenol is prepared by saturating solid redistilled phenol (BMB) with water, adding 0.1% 8-hydroxyquinoline as an antioxidant, and equilibrating with 100 mM Tris buffer(pH 8.0) so that the pH of the phenol was >7.6], and 50% chloroform (with 3% isoamyl alcohol). The layers were separated by centrifugation at 10,000 rpm (Sorvall centrifuge with SA-600 rotor), 10 minutes 36 at 25°C. The aqueous layer was removed and the DNA precipitated with 1/2 volume of 7.5 M ammonium acetate (pH 7.5), and.2 volumes of 95% ethanol. The DNA was rinsed in 70% ethanol and dissolved in 5 ml TE buffer [10 mM Tris (pH 8.0) , 1 mM EDTA (pH 8.0)]. The samples were then digested with 400 ug/ml RNase A (Sigma Chemical Company; preboiled 10 minutes) in the presence of 0.01% SDS for 1 hour at 37°C, followed by 200 ug/ml Proteinase K for 1 hour at SONS. The DNA was washed in phenol and chloroform and precipitated as above. DNA isolation from tissue culture cells was essentially the same, but smaller volumes were used and centrifugation was in a Brinkmann microfuge. After dissolution in TE buffer, the absorbances at 260 and 280 nm (Gilford Response spectrophotometer) were used to determine concentration (1 Amp = 50 pg DNA) and assess purity as the ratio of absorbance at 260nm to that at 280nm. Generally, acceptable Azoo/Azao ratios were between 1.7 and 2.0, however DNA from some tumor samples had lower ratios. 3. Restriction Endonuclease Digestion, Agarose Gel Electrophoresis, and Southern Transfer A. Restriction endonuclease digestion Digestion of DNA with restriction endonucleases and electrophoresis was performed essentially as described by Vorce and Goodman (1987) . Ten microgram aliquots of DNA were digested to completion with 5 units/pg MspI, HpaII, StuI, 10101, or Tan; or 3/units ug HhaI, HindIII, or EcoRI. See 37 Figure 4 for restriction sites. All enzymes were purchased from Bethesda Research Laboratories (BRL; Gaithersburg, MD) or BMB and were supplied with the appropriate 10X concentrated reaction buffer. The DNA, buffer, .1/2 the enzyme, and distilled water to 57.5 ul were mixed thoroughly and incubated at 37°C for 30 minutes, then the remaining enzyme was added, the reaction mixture was mixed thoroughly and incubated for an additional 90 minutes at 379C. Tan digestions were carried out at 65W3. For estimation of fragment size, 1 ul lambda phage HindIII fragments (BRL) diluted to 60 pl with water was prepared. A 5X marker dye (15% Ficoll 400, 0.125% bromphenol blue, 0.125% xylene cyanole ff, 5x TBE) was added to each sample and the samples were heat-treated at 65°C for 10 minutes. B. Agarose gel electrophoresis Electrophoresis was carried out in a model HO/Hl horizontal electrophoresis apparatus (BRL). Gels were prepared as 0.9% or 1.1% agarose (BRL) in 1X TBE by heating in a microwave until boiling, then stirring until approximately 50°C, pouring, and inserting the well former. After solidification, the well former was removed and the gel was submerged in 1x TBE (89 mM Tris, pH 8.3, 89 mM boric acid, 2.5 mM EDTA) in the electrophoresis apparatus. The samples were loaded using a Pipetman (Rainin Instrument Co.). Electrophoresis at 50 V proceeded for 16 hours or until the dye front was approximately 1 inch from the end of the gel. 38 ENZYME REnggégION REFERENCES EcoRI G/AATTC’ Rubin and Modrich, 1980 HhaI GC*G/C Bird and Southern, 1977 I Hind III A/AGC'TT Smith and Marley, 1980 HpaII C/ C*GG WaalMVLinjnk aanndd 5:13;:111'9717978 ; * Sneider, 1980; MspI c:/CGG van der Ploeg and Flavell, 1980 StuI AGG/C'C'T Shimotsu, et a1. , 1980 Tan T/CGA Streek, 1980 10101 C/TC'GAG Gingeras, et a1., 1977 Figure 4. Restriction endonucleases and recognition sites. The :restriction. endonucleases ‘used in 'these studies. are listed, along with their recognition sites. the enzyme is inhibited by the presence of 5MeC, indicates those inhibitory sites. If cleavage by an * 39 The gel was removed from the apparatus, stained in 0.5 ug/ml ethidium bromide (Oncor, Gaithersburg, MD) for 15 minutes, then destained in glass distilled water for 15 minutes on a LabLine orbital shaker. The DNA was visualized on a UV light box (Fotodyne, New Berlin, WI) and photographed using a Polaroid MP-3 camera with a Kodak Wrattan No. 9 filter, an aperture of 6.8, exposure time of 1 second, and Polaroid 667 black and white film. A fluorescent ruler was photographed alongside the lane containing the lambda HindIII fragments. A graph of the fragment sizes on the ordinate of semilog paper as a function of the distance each fragment traveled was used to estimated the size of sample fragments. C. Southern transfer The gel was soaked in 0.25 N HCl for 10 minutes to fragment the DNA by depurination and facilitate the transfer of high molecular weight DNA. The DNA was: denatured by soaking the gel in 0.6 M NaCl, 0.4 M NaOH for 30 minutes, then neutralized by soaking in 1.5 M NaCl, 0.5 M Tris (pH 7.5) for 30 minutes. Constant agitation was provided during all gel soaks by an Orbit Shaker (Lab Line). A piece of 3 MM paper (Whatman) served as a wick for the transfer buffer, 20XISSC (3 M NaCl, 1 M sodium citrate, pH 7.0). The gel was inverted onto the wick and a Hybond (Amersham, Arlington Heights, IL) nylon membrane, trimmed to size, was placed on the gel. Four pieces of trimmed 3 MM paper, wet in 20X SSC, were placed on the membrane and bubbles were removed by rolling a pipet gently over the stack. Trimmed blotting pads and paper towels 40 were added and a smooth plate and 500 g weight were placed on top. ‘Wet pads were removed as needed and the entire apparatus was covered with plastic wrap overnight. Capillary transfer of DNA to the membrane was allowed to proceed for approximately 22 hours. After transfer, the membrane was placed face up on the damp 3 MM paper in a Stratlinker (Stratagene, LaJolla, CA) and irradiated with 120,000 ujoules of ultraviolet light, rinsed briefly in 5X SSC and baked for 2 hours in an 80°C vacuum oven. Membranes were stored in a hybridization bag at room temperature until use. 4. Labeling of Probe with 32P by the Random Primers Method A. Probes Oncogene probes were purchased from Oncor (Gaithersburg, MD). The raf probe is 290 base pairs long and consists of the StuI digestion fragment of v-raf that extends from the interior of exon 11 to the interior of exon 14. This is homologous to the region from base #1295 (a StuI site) to base #1564 (not a StuI site in human DNA) in the human raf gene (Bonner, et a1., 1986; Rapp, et a1., 1988; Devereux, et a1., 1984) (Figure 5A). The Ha-ras probe is the SstI/PstI digestion fragment of v-Ha-ras and is approximately 730 base pairs long. It includes the entire coding region and about 100 base pairs both 5' and 3' to the coding region (Dhar, et a1., 1982) (Figure 5B). 41 Figure 5. Schematic representation of probes. A) raf. The area of the human raf gene which is homologous to the viral probe is shown between the 2 arrows labeled with P. The sizes are for the human raf gene; the size of the introns in the mouse raf gene is not known. The arrow labeled with an M points to an MspI site known to be in exon 12 of human raf. B) Ha-ras. The SstI/PstI digestion fragment of the Harvey Murine Sarcoma Virus, used as the probe, is shown between the arrows labeled with P. The region which codes for the p21 protein is indicated by the dotted line. 42 a) P 3 H fl 5 P l .. i .. . O ‘ I. v- 87 hp 177 hp 46 hp 120 bp EXON EXON EXON EXON 11 12 18 14 b) P P l .. l [-9192'91‘993--I r 43 B. Labeling Probes were labeled using a random primer DNA labeling kit (BMB) which is a modification of the method originally described by Feinberg and‘Vogelstein (1983; 1984). Template DNA was diluted to 25 ng/10 pl in TE buffer and stored at -20°C in 25 ng aliquots. 25 ng of the template DNA was denatured by boiling 10 minutes, then immediately immersing in an ice water bath. The buffer containing assorted hexanucleotides (the random primers), dATP, dGTP, dTTP and asdeCTP (3000 Ci/mmol, New England Nuclear) were added to the template DNA. The large Klenow fragment of DNA polymerase was added to extend the primers and the reaction was carried out at 37°C for 30-60 minutes. The reaction was stopped by the addition of 1/5 volume 0.1 M EDTA (pH 8.0) and the probe was diluted to 50 pl with STE [10 mM.Tris (pH 8.0), 1 mM EDTA (pH 8.0), 100 mM NaCl]. Unincorporated nucleotides were removed by passage through a BioSpin 30 column (BioRad) spun at 1,100 x g 4 minutes in a GLC-l centrifuge. The specific activity of the labeled probes was determined by averaging the counts per minute (cpm) of two 2 pl aliquots counted in Safety-Solve in a Packard Tricarb 460C scintillation counter. The labeled probes were stored at - 20°C until use and were discarded when one half-life of the 32P (2 weeks) had elapsed. 5. Hybridization of DNA Affixed to Hybond Membranes A modification of the method described by Seldon (1987) was used for hybridization of DNA affixed to nylon membranes. 44 Ten. ml of pre-hybridization. buffer [5x SSC, 25 mM KPO, (pH 7.4), 0.5% dextran sulfate, 1% SDS, 50% deionized formamide, 5X Denhardt's solution (0.1% Ficoll 400, 0.1% polyvinylpyrrolidone, 0.1% bovine serum albumin)] was added to the dry membranes and distributed throughout the bag, then the bag was sealed and placed in a 32°C water bath for 4 hours. Ha—ras (2 x 107 cpm) or raf (1 x 107 cpm) 32P-labeled probe, diluted with 500 pl STE, and 1 g (final concentration - 100 mg/ml) sheared, sonicated salmon sperm DNA (Sigma Chemical Co.), was boiled 10 minutes, then immersed in ice water. The prehybridization buffer was completely removed from the bag by rolling a pipet over the open bag. Ten ml of hybridization buffer (the same as pre-hybridization buffer without Denhardt's solution) was quickly poured into the bag. The probe mixture was added to the hybridization buffer by pasteur pipette and rapidly dispersed over the membrane. The bag was sealed, then extensively massaged to distribute the probe. Hybridization was carried out in a 32°C water bath for 20 hours with the Ha-ras probe and 16 hours for the raf probe. Following hybridization, the membranes were rinsed in 2X SSC, 0.1% SDS for 5 minutes, then washed, with agitation, in 2x SSC, 0.1% SDS for 15 minutes at room.temperature, 0.5X.SSC, 0.1% SDS 15 minutes at room temperature, 0.1x SSC, 0.1% SDS for 15 minutes at room temperature, and 0.1x SSC, 1.0% SDS for 30 minutes at 37°C. Membranes hybridized with the Ha-ras probe were washed an additional 10 minutes at room temperature in 0.1): SSC, 1.0% SDS. The membranes were then blotted on 45 3 MM paper and wrapped in Handi-Wrap (Dow Chemical Co.) for autoradiography. The radiolabeled probe was removed from the membrane following autoradiography by pouring a 78-80°C solution of 10mM Tris (ph 8.0), ImM EDTA (pH 8.0), 1% SDS on the membrane and agitating for 30 minutes, then repeating once. 6. Autoradiography The plastic-wrapped membrane was placed between 2 Cronex Lightening Plus intensifying screens inside a Kodak cardboard exposure cassette. Kodak X-OMAT AR5 film was added between the membrane and the upper intensifying screen (in complete darkness), the cassette was wrapped in foil, and clamped between 2 clipboards. The film was exposed to the membrane for 1-7 days at -80°C. The developing process was carried out in complete darkness. The film was removed from the cassettes, placed in a film holder, immersed in Kodak GBX developer for 4 minutes (at 729E), rinsed briefly in running water, and immersed in Kodak.GBX fixer for 4 minutes. The developed film was rinsed thoroughly in running water and dried for 45 minutes in a drying cabinet. 7. RNA Isolation RNA isolation requires the use of RNase-free materials. Glassware was baked at 400°C for 4 hours, glass distilled water and buffers were treated with 0.1% diethylpyrocarbonate 46 (DEPC) and autoclaved, plasticware was assumed to be RNase- free if new. RNA was isolated by a modification of the method described by Chomczynski and Sacchi (1987). A frozen liver sample weighing from 50 to 300 mg was placed in 3 ml of 4 M guanidinium isothiocyanate, 25 mM sodium citrate (pH 7.0) , 0.5% sarcosyl with 21.6 pl B-mercaptoethanol and immediately homogenized in a Polytron tissue grinder (Brinkmann) . The solution was poured into a new 15 ml Corning polypropylene tube. One-tenth volume of DEPC-treated, autoclaved 2 M sodium acetate (pH 4.0) was added and the solution was mixed gently. One volume of water-saturated (not equilibrated) phenol (with 0.1% 8-hydroxyquinoline) was added and the solution mixed gently, then one-third volume of chloroform (containing 3% isoamyl alcohol) was added. The solution was shaken vigorously for 10 seconds, transferred to a baked 15 ml Corex tube and placed on ice 15 minutes. The layers were separated by centrifugation at 8,100 x g (7,500 rpm in Sorvall centrifuge) 20 minutes at 0°C. The aqueous layer was removed by transfer pipet to another baked 15 ml Corex tube and the RNA was precipitated with 1 volume n—propanol at -20°C at least.1.houru The RNA.was pelleted by centrifugation.at 8,100 x g (7,500 rpm) for 20 min at 0°C. The pellet was redissolved in 300 pl of the same guanidinium isothiocyanate-containing denaturing solution used initially, transferred to a new eppendorf tube, and the RNA was reprecipitated with 1 volume of n-propanol at -20°C for at least 1 hour. The RNA was again pelleted by centrifugation at 10,000 rpm for 10 minutes in an 47 Eppendorf centrifuge (Brinkmann) in a cold room. The pellet was rinsed in 75% ethanol and recentrifuged. After air drying, the pellet was dissolved in 100 pl of DEPC-treated 0.05% SDS at 65°C with frequent vortexing, then placed on ice for 10 minutes. Insoluble salts were removed by centrifugation in the Eppendorf centrifuge at 10,000 rpm 5 minutes. The supernatent containing the RNA was transferred to a new tube and concentration and purity were determined spectrophotometrically. Concentration was determined using the constant value of 1 A260 = 40 pg of RNA/ml. The “zoo/A280 ratios in TE buffer routinely approximated 2.0. RNA was stored for future use at —80°C. 8. Simultaneous Isolation of DNA and RNA When less than 200 mg of liver tissue was available (e.g. some tumor samples), DNA and RNA were isolated simultaneously by a modification of the method of Chirgwin, et a1. (1979). Frozen liver tissue was pulverized in liquid nitrogen with a mortar and pestle, then stirred rapidly into 4 M guanidine isothiocyanate, 0.5% sarcosyl, 1 M sodium citrate (pH 7.0), 0.5% fi-mercaptoethanol until homogeneous. This was layered over a 5.7 M/3 M CsCl [0.1 M EDTA (pH 8.0), DEPC-treated, and autoclaved) gradient and centrifuged in a Beckman ultracentrifuge in a SW41 rotor at 29,000 rpm for 22 hours. A. DNA isolation Following centrifugation, the guanidium layer was discarded. Using a transfer pipette, the viscous DNA 48 containing layer was removed from just below the 5.7 M/3 M CsCl interface and measured. One thousandth volume of RNase A (50 mg/ml in TE buffer, preboiled 10 minutes) was added to the DNA solution and the mixture was dialyzed against a solution of 10 mM Tris (pH 7.5), 1 mM EDTA (pH 7.5), 0.1% SDS for 1 hour at room temperature. One hundredth volume of Proteinase K (10 mg/ml in TE buffer) was added, the dialysis solution replaced, and.the DNAIwas dialyzed.another'houru The dialysis solution was then replaced with 10 mM Tris (pH 7.5), 1 mM EDTA. Following overnight dialysis at.43C, the DNA was washed with phenol and chloroform, precipitated, and dissolved as in section 2. B. RNA isolation After removal of the DNA layer, the tube*was rapidly emptied and inverted to drain. The portion of the tube containing the RNA.pellet.was cut and the upper portion of the tube discarded to avoid RNase contamination. The RNA was dissolved.in.DEPC-treateduwater’andmwashed.twice'with.an.equal volume of 4:1 chloroform:butanol and brief centrifugation at 14,000 rpm in a Brinkmann microfuge. The top, RNA containing, layer was removed and precipitated with an equal volume of n- propanol at -20°C for at least 1 hour, centrifuged, washed, and dissolved in 0.5% SDS as in Section 7. 9. Assessment of RNA Integrity by Electrophoresis Assessment of RNA was performed using either extra samples prepared at the time an electrophoresis for Northern 49 transfer was done (10 pg RNA/25 pl total volume) or 3.5 pl RNA (independent of concentration) denatured in 50% formamide, 0.5x MOPS (pH 7.0), 2.2 M formaldehyde. One pl of 10,000X ethidium bromide (Oncor) was added to, each sample and the samples were heated at 60°C for 15 minutes then placed on ice. One-fifth volume of 5 x RNA loading buffer (50% glycerol, 1 mM EDTA, 0.4% bromphenol blue, 0.4% xylene cyanol ff) was added to each sample and 7 pl of each sample was loaded on a 1% agarose, 0.66 M formaldehyde, 1X MOPS (pH 7.0) gel in a minigel apparatus (BRL). The running buffer was 1 X MOPS and electrophoresis was carried out at 80 V for approximately 1 hour. The gel was photographed as described in section 3B. The RNA was determined to be intact if 2 major bands were seen and if the band most proximal to the wells (the 28S ribosomal RNA band) was most intense. 10. Agarose Gel Electrophoresis of RNA and Northern Transfer A. Agarose gel electrophoresis RNA was diluted in DEPC-treated water to 10 pg in a total volume of 4.5 pl and denatured in 50% formamide, 0.5 X MOPS (pH 7.0) , 2.2 M formaldehyde at 60°C for 15 minutes, then placed on ice. One-fifth volume of 5 X RNA loading buffer (50% glycerol, 1 mM EDTA, 0.4% bromphenol blue, 0.4% xylene cyanol ff) was added. RNA ladder (BRL) was treated as one sample. The samples were loaded on a 1% agarose, 1x MOPS, 0.66M formaldehyde gel and electrophoresis was carried out in 1X MOPS at 80 V for 6-7 hours or until the dye front was 50 approximately 2/3 of the way between the wells and the end of the gel. The lane containing the RNA ladder was cut from the gel and stained with 1:1000 of 10,000X ethidium bromide for 20 minutes and destained in water 1-2 hours). During the staining and destaining procedure, sodium, not fluorescent, lights were used in the room to decrease uptake of ethidium bromide by the formaldehyde-containing gel and subsequent autofluorescence of the gel. The RNA ladder was photographed as in section 3B using 2A and 23A filters and a f-stop of 11. B. Northern transfer Prior to northern transfer, the gel was rinsed 5 times in DEPC-treated water. Gene Screen Plus (NEN) was cut to size, wet in water, then soaked in 10x SSPE [1.5 M NaCl, 0.1 M NaHzPO“ 0.01 M EDTA (pH 7.4)] for 15 minutes. The Northern transfer was set up exactly as in section 3C, except the transfer buffer was 10X SSPE. After transfer, the membrane was UV-crosslinked and baked as described in section 3C. If not hybridized immediately, the membrane was stored in a hybridization bag at -20°C. 11. Hybridization of RNA Affixed to Gene Screen Plus Northern blots were prehybridized in 10 ml of 5X SSPE (pH 7.4), 50% deionized formamide, 2.5x Denhardt's solution, 10% dextran sulfate and 1% SDS at 42°C for 2-4 hours. Radiolabeled probe (1 x 107cpm/10 ml hybridization buffer) and 1 g sheared salmon sperm DNA were denatured by boiling add immersing in an ice water bath, then added to the 51 hybridization bag. After hybridization at 42°C for 16-18 hours, the membranes were washed with agitation twice in 2X SSPE for 15 minutes at room temperature, twice in 2X SSPE, 2% SDS at 65°C for 45 minutes, and twicein 0.1X SSPE at room temperature for 15 minutes. Autoradiography was as described in Section 6; exposure was for 3-14 days. The membranes were stripped of the radiolabeled probe by agitating the membrane in boiling 0.1% SDS 30 minutes. 12. 5' End-labeling of 28S Oligonucleotide A 25-mer DNA oligonucleotide specific for 28S rRNA (5'- .AAC GAT CAG ACT AGT GGT ATT TCA CC-3'; Barbu and Dautry, 1989) was synthesized by the Michigan State University Macromolecular Synthesis Facility. The oligonucleotide was diluted and the concentration determined spectrophotometrically. Based on the extinction coefficients of the dNTPs in this oligonucleotide, 1 A26°=29.4 pg/ml. The oligonucleotide was divided into aliquots of 20 pg/tube, dried in a Speed-vac (Savant) and stored dessicated. For use 20 pg was diluted with 100 pl sterile water. A 5' DNA terminus labeling system (BRL) was used to label the oligonucleotide. 1 pl (200 ng) of oligonucleotide was mixed with 12 pl of water, incubated at 65°C for 15 minutes, and removed to an ice-water bath. 5 pl forward reaction kinase buffer, 5 pl 7321, ATP (3000 Ci/mmole, NEN) and 10 units T4 polynucleotidezkinase were added. The solution was mixed gently, centrifuged briefly, and incubated at 37°C for 20 minutes. An additional 52 10 units of T4 polynucleotide kinase was added and the mixture was incubated for another' 20 minutes. The volume was increased to 100 pl with STE buffer and unincorporated nucleotides were removed by centrifugation through a BioSpin 6 column (BioRad). The specific activity was determined by scintillation counting. 13. Hybridization of RNA Affixed to Gene Screen Plus with 288 Oligonucleotide Northern blots were examined for levels of 28S rRNA.as an internal control to correct for possible differences in the amount of RNA loaded in each lane. The membrane ‘was prehybridized. ‘with 200x: denatured 'unlabeled 28S oligonucleotide in 5x SSPE, 10X Denhardt's solution and 0.5% SDS at 48°C at least 15 minutes. Labeled probe (6.35 x 10° cpm/10 ml hybridization buffer) and 200x 'unlabeled 28S oligonucleotide were denatured by boiling and immersing in an ice water bath, then added to the hybridization bag. Following hybridization at 48°C for 16-18 hours, the membranes were rinsed in 2x SSPE, 0.1% Na pyrophosphate, 0.1% SDS for 5 minutes at room temperature, then washed in 2x SSPE, 0.1% Na pyrophosphate, 0.1% SDS in a 42°C incubator for 30 minutes. Autoradiography as described in Section 6 was accomplished in 1-24 hours. 53 14. Image Analysis of Autoradiographs A. Southern blot analysis. For analysis of alterations in methylation status of raf the ratio of the intensity of the band at 6.7 kb to the sum of the intensities of the bands at 5.1 and 3.0 kb in the MspI digests was determined using a BioQuant MEG IV VCMTE image analyzer (R&M Biometrics, Inc., Nashville, TN). The intensity (average density within a constant sized region multiplied by the number of pixels above background) was measured for each band of interest. An individual background correction was made for each band by averaging the density of the area above and below the band and subtracting that number from the band density. The ratios of the intensities of the bands rather than the absolute value for the 6.7 kb band was used to minimize misinterpretations that could occur if there were differences in the amount of DNA loaded per lane. Alterations in the methylation status of Ha-ras was examined similarly by measuring the intensity of the 2.6 kb band in HpaII digests and comparing that with the intensity of the major band at <2.0 kb in the same lane. Because ratios were determined within one lane, comparisons could be made between samples on different autoradiographs. One-way analysis of variance *was. employed, for' statistical analysis and significance was determined by the Newman-Keuls test (ps0. 05) . B. Northern blot analysis Analysis of Northerns blots was accomplished by comparison of the intensity of the hybridization of the raf or 54 Ha-ras probe to the intensity of the hybridization of the 28S oligonucleotide. Because ratios were determined from bands on different autoradiographs, comparisons could only be made between samples on the same autoradiograph. 15. Protein Isolation Approximately 50 mg of frozen liver tissue was added to 1 ml Triton lysis buffer [150 mM NaCl, 5 mM ugcaz, 50 mM Tris (pH 7.4), 1% Triton X-100] containing 8.8 pl 2% phenylmethyl sulfonyl fluoride (PMSF; dissolved in ethanol) and 10 pl aprotinin, homogenized immediately, and centrifuged at 16,000 x g, 15 minutes at 4°C in an Eppendorf microfuge. The supernatent was transferred to a new tube. Protein concentration was determined spectrophotometrically with a Bradford protein assay using bovine serum albumin to generate a standard curve (Bradford, 1976). 16. Polyacrylamide Gel Electrophoresis and Western Blotting A. Polyacrylamide gel electrophoresis A 12% acrylamide, 0.5% bis-acrylamide gel was prepared in 375 mM Tris (pH 8.8) with 0.1% SDS and 0.1% ammonium persulfate. 12 pl N,N,N',N'- tetramethylethylenediamine (TEMED) (for a 30 ml gel) was added as a catalyst to polymerize the gel (modified from Laemmli, 1970). The gel was poured in.a Hoeffer (San Francisco, CA) SE 650 vertical gel apparatus and allowed to polymerize about one hour. A 3.75% stacking gel (3.75% acrylamide, 0.1% bis- 55 acrylamide, 83 mM Tris (pH 6.8), 0.1% SDS, 0.1% ammonium persulfate with 6 pl TEMED (in a total volume of 6 ml) was layered over the 12% gel and allowed to polymerize overnight with the well-former in place. Protein (150 pg) was diluted to 20 pl, mixed with one volume of 2X sample buffer [125 mM Tris (pH 6.8), 4% SDS, 20% glycerol, 10% fi-mercaptoethanol, 0.01% bromphenol blue], boiled for 3 minutes and loaded on the gel. Prestained protein markers (20 pl) were included as one sample to monitor the transfer and 20 pl unstained low'molecular weight protein standards (BioRad; diluted 1:10) were included for estimation of protein size. The samples were stacked using 20 mA (constant current) for about 30 minutes, then resolved at 40 mA for 3-4 hours. The electrophoresis buffer was 25 mM Tris (pH 8.3), 192 mM glycine, 0.1% SDS. B. Western blotting Following electrophoresis, the stacking gel was removed and the resolving gel sandwiched against a sheet of nitrocellulose between 2 sheets of filter paper, 2 Scotch- Brite pads and, 2 plastic supports. The proteins were transferred to the nitrocellulose electrophoretically overnight at 30 volts overnight in 25 mM'Tris (pH 8.3), 192 mM glycine, 20% methanol (w/v) as described by Towbin, et a1., (1979). After'the transfer“was complete, the lanes containing the low molecular weight protein standards were removed and visualized by India Ink staining in a solution of 0.1% India 56 Ink, 0.3% Tween 20 in phosphate-buffered saline several hours to overnight. 17. Detection of p21 by immunoblotting The nitrocellulose was incubated in 10 ml of incubation buffer [1X TTBS (100 mM Tris (pH 7.5), 0.9% NaCl, 0.1% Tween 20), 5% non-fat dry milk] per lane in a sealable sandwich bag for 1 hour on a Red Rocker mixer (Hoeffer). Fresh incubation buffer containing a 1:500 dilution of the monoclonal antibody Ab-l (Oncogene Science, Manhasset, NY; from the clone Y13-259, Furth, et a1., 1987; Sigal, et a1., 1987) was added. Following overnight incubation, the membrane was washed in TTBS, placed.in a new*bag with fresh incubation.buffer and.the second. antibody' (biotinylated. anti-mouse IgG; ‘Vector Laboratories, Burlingame, CA), incubated overnight, then washed in TTBS. Then the membrane was incubated with the Vectastain reagent, consisting of a biotin-avidin-horseradish peroxidase complex (Vector), for 1 hour, and washed with TTBS. The diagnostic color change was catalyzed by incubation with a solution of diaminobenzidine, hydrogen peroxide, and NiCl (Vector). The color change is temporary (about 2 weeks if exposed to light), so the membranes were photographed to preserve the data. RESULTS 1. Methylation Status and mRNA Levels of Ha-ras in WBm'm cell-induced Rat Liver Tumors A. Assessment of methylation status of Ha-ras in WB cells The methylation status of Ha-ras in DNA from WB'", m, and WB’WF“ (cell lines described in the Methods section) WB was examined by MspI, HpaII, and HhaI digestion and hybridization with the 32P-labeled v-Ha-ras probe. In both the WB"t and the WBmm cells, the DNA is largely undigested by HpaII (Figure 6, lanes 1-4), i.e. the majority of 5'-CCGG-3' sites are methylated on the internal cytosine (5'-C'°CGG-3 ') . In DNA from WB""'°° cells, additional bands are detectible in the MspI digest at 280 and approximately 250 base pairs (Figure 6, lane 5) indicating the presence of foreign DNA (i.e. the virally- introduced v-Ha-ras). These bands are also present in the HpaII digest of the same DNA; additionally, a band at 380 bp which was present in MspI digests of WB"t and WE” cells is now Ha-ras detectible in the HpaII digest of WB cells (Figure 6, lane 6). This indicates that the introduced Ha-ras DNA is hypomethylated since it is cleaved by both MspI and HpaII. A further indication of hypomethylation of the v-Ha-ras DNA is the detection of additional bands (1.9 and 1.1 kb) in Ha-ras 57 58 Figure 6. The methylation status of Ha-Fas: WB cells; MspI and HpaII digests. DNA from WB" , WB"°°, WB ' "'8 cells and cells culturegrfarom liver tumors from F344 rats which were injected with WB cells (WBK1, WBK2, and WBK3 cells) was digested with 5U/pg MspI (M) or HpaII (H), fractionated through a 1.1% agarose gfil, and analyzed by Southern blotting, hybridization with the P-labeled v-Ha-ras probe, and autoradiography. The arrows point to the additional bands seen in the HpaII digests of the Ha-ras infected cells and the cells cultunre'ced from tumors. Lanes 1 $139.82: WB cells; lanes 3 and 4: WB cells; lanes 5 and 6: WB cells; lanes 7 and 8: WBKl cells; lanes 9 and 10: WBKZ cells; lanes 11 and 12: WBK3 cells. vav monou- N no 00 NN 1.353— - 59 1.072‘ 0.872' 0.603": 1? 0.310* 0 271‘ 60 Na-ras from WB cells digested with HhaI compared to WB"t cells (Figure 7, lanes 1 and 2). Cells cultured from liver tumors in WB"°'"°-injected rats (WBK1, WBK2, and WBK3 cell lines) have the same HpaII (Figure 6, lanes 7-12) and HhaI (Figure 7, lanes 3-5) digestion patterns as the original injected cells, suggesting that the tumors did indeed evolve from the WB"°'"° cells. B. Assessment of methylation status of Ha-ras in F344 lie-ras cells rats injected with WB Figure 8 shows the results of MspI and HpaII digestion of DNA from the liver and lung of an untreated F344 rat and liver and lung tumor tissue and surrounding nontumorous liver tissue from one F344 rat given approximately Ha-ras 1 x 107 WB cells via portal vein injection 3 weeks previously. The Ha-ras gene is poorly digested by HpaII in thenormal F344 rat liver (lanes 1 and 2) and lung DNA (lanes 7 and 8) and in DNA from the liver tissue surrounding the tumor in the treated rat (lanes 3 and 4) indicating that most 5'-CCGG-3 ' sites are methylated on the internal cytosine. The DNA isolated from liver (lanes 5 and 6) and lung tumors (lanes 9 and 10) has a relatively hypomethylated Ha-ras gene indicated by the presence of an additional band in the HpaII "B'PBS digests at 380 bp as seen in the WB cells. HhaI digestion of DNA from liver tumors from 2 rats injected with WBflrm cells results in the detection of additional bands at approximately 1.9 and 1.1 kb, also indicating hypomethylation of Ha-ras in the tumor tissue relative to Ha-ras in untreated 61 Figure 7. The methylation B’status of Ha-ras: WB cells; HhaI digests. DNA from WB" WB , WBK1,WBK2,WBK3 cells was digested with 3U/pg HhaI, fractionated through a 0.9% gel and analyzed as in Figure 6. The arrows point to the additional bands seen in the Ha-ras infected cells and Hthe cells cultured from tumors. Lane 1: WB" cells; lane 2: WBH' "3 cells; lane 3: WBKl cells; lane 4: WBKZ cells; lane 5: WBK3 cells. 62 Figure 8. The methylation status of Ha-ras: F344 rat liver and lung tumors; MspI and HpaII digests. DNA from liver and lung tissue from an untreated F344 rat, liver" and lung tumors from.a F344 rat which had received 1 x 10 8cells via the portal vein 3 weeks previously and nontumorous liver tissue surrounding the liver tumor was analyzed as in Figure 6. Lanes 1 and 2: normal liver; lanes 3 and 4: nontumorous liver surrounding a tumor; lanes 5 and 6: liver tumor; lanes 7 and 8: normal lung; lanes 9 and 10: lung tumor. 63 1.353 “ 0.603 0.310 0.271 . a... 64 F344 rat liver and liver tissue surrounding tumor tissue (Figure 9). C. Assessment of Ha-ras mRNA and ras p21 protein levels Because hypomethylation of a gene has the potential to increase mRNA. expression of that gene and increased expression of Ha-ras has been implicated in tumorigenesis, we examined Ha-ras mRNA levels in tumors from two F344 rats Ha-ras injected with WB cells (Figure 10a). The amount of Ha-ras mRNA in the adjacent nontumor tissue was similar to that seen in normal liver tissue from an untreated rat. The level of Ha-ras mRNA in both the tumors is greatly increased relative to nontumor tissue (Figure 10A). The level of p21 protein from a liver tumor was increased relative to control liver tissue (Figure 10B) . Therefore there is a relationship between hypomethylated v-Ha-ras DNA, increased expression of Ha-ras (both mRNA and p21) and tumorigenicity. 2. Ha-ras and raf in the Nascent Liver of B6C3Fld, C3H/Hed, and C57BL/69 Mice A. RFLP screening DNA from B6C3F1o‘, C3H/Hed, and C57BL/69 mice was screened for RFLP. MspI digestion detected an RFLP (Figure 11); bands of 6.7, 5.1, 3.0, and 2.3 kb were seen in ref- probed C57BL/69 and B6C3Fld DNA, while the 6.7 kb band is not seen in C3H/neg DNA, To determine if the MspI RFLP‘was due to a change in base sequence or to a difference in methylation status, C57BL/69 DNA was digested with MspI and HpaII 65 2.3- 2.0- Figure 9. The methylation status of Ha-ras: F344 rat liver tumors; HhaI digests. DNA from liver tissue from an untreated F3fignrat, liver tumors from 2 rats which had received 1 x 10 WE cells via the portal vein 3 weeks previously’ and nontumorous liver tissue adjacent to a liver tumor was analyzed as in Figure 7. Lane 1: normal liver; lane 2: nontumorous liver adjacent to a tumor; lanes 3 and 4: liver tumors from 2 rats. 66 Figure 10. Ha-ras mRNA and p21: F344 rat liver tumors. A) Ha-ras mRNA levels. RNA from liver tissue from an untreate F3 a{Irmram liver tumors from 2 rats which had received 1 x 10 WB cells via the portal vein 3 weeks previously and nontumorous liver tissue adjacent to a liver tumor fias analyzed by Northern blot analysis, hybridization with the P- labeled v-Ha-ras probe, and autoradiography. Lane 1: normal liver; lane 2: nontumorous liver tissue adjacent to a tumor; lanes 3 and 4: liver tumors from 2 rats. B) p21 levels. Total protein from liver tissue from an untreatedasrat and from a liver tumor induced by injection of WB cells was fractionated by size through a polyacrylamide gel, transferred to nitrocellulose, and analyzed by immunoblotting with the Y13-259 ras p21 antibody. Lane 1: normal liver; lane 2: liver tumor . a) b) 67 68 C3H H8 (:5 BL ,____ Figure 11. Restriction fragment analysis of raf: B6C3Fld, C3H/Had, and C57BL/69; MspI. DNA from C3H/Had, C57BL/69, and B6C3Flo' was odigested with 5U/pg MspI (M) or HpaII (H) for 2 hours at 37 C, fractionated through a 0. 9% agarose gel and analyzed by Southern blotting, hybridization with the labeled raf probe, and autoradiography. The arrows point to the 6.7 kb band seen in C57BL/69 and B6C3F1o‘, but not C3H/lied DNA. Lanes 1 and 2: B6C3Flo’; lanes 3 and 4: C3H/Hed; lanes 5 and 6: C57BL/69. 69 sequentially (Figure 12). The 6.7 kb band is not seen in the DNA digested with both MspI and HpaII (Figure 12, lane 3). Cleavage by MspI is inhibited by the presence of a methyl group on the external cytosine of its recognition sequence (Sneider, 1980; van der Ploeg and Flavell, 1980; Waalwijk and Flavell, 1978) , while HpaII will cleave 5'-"‘CCGG—3', but not 5'-C"°CGG-3' sites (Mann and Smith, 1977). This indicates the presence of a 5'-MCCGG-3' site within the fragment which is detected as a 6.7 kb band in MspI-digested, raf—probed C57BL/69 DNA. In the C3H/Hed, this site is not methylated since it is digested by MspI. Thus the raf gene in the tumor- prone C3H/Hed is hypomethylated relative to the tumor- resistant C57BL/69. An RFLP was detected in Tan digested DNA (Figure 13). Hybridization with the raf probe results in 3.0 and 2.0 kb bands in C3H/Hed DNA and 2.5 and 2.0 kb bands in C57BL/69 DNA. 3.0, 2.5, and 2.0 kb bands are detected in B6C3Fld DNA (Figure 13) . Therefore, the raf gene in the 2 parental strains differs in base sequence and the B6C3Fld inherits one allele from each parent. RFLPs of the raf gene were not detected with EcoRI, HindIII, or StuI (Figure 13). The published sequence of c-raf-l was examined for recognition sites of the enzymes used in this study using the Map subprogram of the GCG sequence analysis program on the Vax computer system (Devereux, et a1., 1984). There is an MspI recognition site (5'-CCGG-3') in exon 12 approximately 0.8 kb 70 Figure 12. MspI and HpaII double digests of C57BL/69 DNA; raf. 10 pg of DNA was digested with 5U/pg MspI or HpaII in the appropriate buffer [MspI reaction buffer is 50 mM Tris (pH 8.0), 10 mM MgC12; HpaII reaction obuffer is 20 mM Tris (pH 7.4) , 10 mM MgC12] for 2 hours at 37 C. For the double digest, 10 pg of DNA was digested with HpaII in the buffer supplied by the manufacturer for 2 hours, then the buffer was adjusted to the proper concentrations [Tris (pH 8.0) added to a final concentration of 50 mM] for digestion with MspI which was carried out for 2 hours. Analysis was performed as in Figure 11. Arrows point to the 6.7, 5.1, 3.0, and 2.3 kb bands in the MspI digests. Lane 1: DNA digested with MspI in manufacturer's recommended buffer; lane 2: DNA digested with MspI in buffer recommended for HpaII, adjusted to 50 mM Tris; lane 3: DNA digested with HpaII, then MspI; lane 4: DNA digested with HpaII. 71 Figure 13. Restriction fragment analysis of raf: B6C3Fld’, C3H/Hed, and C57BL/69; Tan, StuI, HindIII, EcoRI. DNA from B6C3F1d (lane 1), C3H/Hed (lane 2), and C57BL/6Q (lane 3) was digested with 5U/pg Tan (2 hours at 65°C) , 5U/pg StuI, 3U/pg HindIII, or 3U/pg EcoRI and analyzed as in Figure 11. The arrows point to the major bands. 72 Lg 23.1— 9.4- 6.6- 4.4- 5 2.3- g g 2.0— __ ’1: 335 fig; 1=B6C3F1&; 2=c3H/Hea; 1=B6C3Fld; 2=c3H/Hea; 5 '3=C57BL/62 3=cs7BL/62 ' RAF: HINDIII RAF: EcoRI 1 2 3 1 2 3 I, , . L9 2 - 23.1- . 9.4- ‘- 6.6--% 4.4-” 2.3—. > i: 2.0- . l=86 3F1d; 2=c3H/Hee; .45 . 4 3=c57BL/62 1=86c3Fla- 2=c3H/Hea; 3=cé7BL/62 73 from the StuI site at the 5' boundary of the area homologous to the probe (Figure 5A). StuI digestion of B6C3Flo‘ DNA results in a 3.2 kb band when hybridized with the raf probe (Figure 14, lanes 1—6). The length, of the human c-raf fragment homologous to the probe is approximately 6.2 kb (Rapp, et a1., 1988). Therefore there are differences in human and mouse c-raf intron size. The major bands detectable in a StuI and MspI double digest of B6C3F1d DNA are 2.4 and 0.8 kb (Figure 14, lanes 7-9), while double digestion with StuI and HpaII does not result in cleavage of the 3.2 kb fragment (Figure 14, lanes 10-12). This indicates that the 5'-CCGG-3' site in exon 12 is methylated on the internal cytosine (i.e., 5'-C"°CGG-3') since it is digested by MspI but not HpaII. Aside from the RFLP previously detected in the MspI digests of the 3 mice strains, seen as a 15 kb band in the C57BL/69, but not the B6C3Fld or C3H/Hed (Vorce and Goodman, 1989a), no Ha-ras RFLPs were detected with EcoRI, HindIII, StuI, or Tan (Figure 15). The methylation sensitive enzymes HhaI and XhoI did not.detect any RFLPs in either raf or Ha—ras in any of the 3 mice strains (Figure 16). B. Assessment of mRNA in nascent liver Northern blots of RNA from each strain were probed for raf and.Ha-ras mRNA and 288 rRNA» The levels of each were approximately equal between the strains (Figure 17). Thus, despite the relative hypomethylation of raf and Ha-ras of 74 Figure 14. 'The methylation status of the CCGG site in exon 12 Of raf: B6C3Fld, C3H/fled, andC57BL/69. DNA from B6C3Flo’ (lanes 1,4,7,10), C3H/Hed'(lanes 2,5,8,11) and.C57BL/69 (lanes 3,6,9,12) was digested with StuI (lanes 1-6; lanes 1-3 were digested in the buffer supplied by the manufacturer, lanes 4-6 were digested in the buffer supplied with MspI adjusted to be optimum for StuI digestion), MspI then StuI (lanes 7-9), or HpaII then StuI (lanes 10-12). Double digests were performed with MspI or HpaII first in the recommended buffer, then the buffer was adjusted to 10 mM Tris (pH 8.0), 100 mM NaCl, 5 mM MgCl., 1'mM fi-mercaptoethanol for digestion with StuI. Southern lots were analyzed as in Figure 11. The arrows point to the 3.2 kb bands in StuI and StuI/HpaII double digests and to 2.4 and 0.8 kb bands in StuI/MspI double digests. 75 N1 .. m. Hebm\ssem= m w m Haum\Hamz noo.o Inw.o Iwo.~ Imm.H In.mm 76 Figure 15. Restriction fragment analysis of Ha-ras: B6C3Flo', C3H/Heo' and C57BL/69; Tan, StuI, HindIII, EcoRI. The Southern bloés pictured in Figure 13 were stripped, hybridized with the P-labeled v-Ha-ras probe, and analyzed by autoradiography. The arrows point to the major bands. HA-RAS: TAaI _.1 2 3 ._'.-: 1=B6C3Fld; 2=C3H/Hed; 3=C57BL/62 HA-RAS: HINDIII 1 2 3 L_ 23.1- 9.4- 4— 6.6- 4.4- ’- 4— 4— 2.3_ j-L‘: I” . 6'0" 2’— 1=B6C3F1d; 2=C3H/He&; 3=C57BL/62 77 HA-RAS: STuI 2.3- 2.0— ‘ ._ . _, 1=86C3F1&; 2=C3H7Hee; 3=C57BL/69 2.3- 2.0- ’3' 1=35c3F1a5 2=c3H/Hee; 3=C57BL/62 78 Figure 16. Restriction fragment analysis of raf and Ha-ras: B6C3Flo‘, C3H/Hed, and C57BL/6Q; HhaI and XhoI. DNA from B6C3Fld (lane 1), C3H/Hed (lane 2), and C57BL/69 (lane 3) was digested with 3U/pg HhaI, or 5U/pg XhoI and analyzed as in Figure 11 (raf) or Figure 15 (Ha-ras). The arrows point to the major bands. RAF: 2HHAI avgii_ 9. 4- 6.6- 4.4— 2.3- 2.0— 1=86C3Fld; 2=C3H/Hed; 3=C57BLI69 RAF: XHoI 1 2 3 :r-fi‘wFfi ”5:; 2.3-: 2.0- 1=BGC3Fld; 2=C3H/Hed; 3=C57BL/62 HA-RAs: HHAI 1=B6C3Fld; 2=C3H/Hea; 3=C57BLI69 HA-RAS: XHoI 1 2 3 Kg . 23.1— 9.4- 6.6- 4.4- 2.3— 2.0- 1=86C3F1d; 2=C3H/Hed; 3=C57BL/69 80 4.7- Figure 17. Ha-ras and raf mRNA levels: B6C3Fld, C3H/Hed, and C57BL/69. RNA isolated from B6C3Fld (lane 1), C3H/Hed (lane 2), or C57BL/69 (lane 3) was analyzed by Northern blotting, hybridization with the P-labeled raf, Ha-ras, or 288 probes (membranes were stripped between subsequent probings), and autoradiography. The size of the RNA calculated as function of the distance traveled relative to an RNA ladder is given for each gene. 81 C3H/Hed' mice, there is no corresponding increase in mRNA levels. 3. Ha-ras and raf Following Phenobarbital Administration or Partial Hepatectomy in B6C3F1d, C3H/Hed, and C57BL/69 Mice A. Assessment of raf methylation status following partial hepatectomy To determine the effect of hepatocyte proliferation on the methylation status of raf in the liver of B6C3Flo‘, C3H/Had, and C57BL/69 mice, a 2/3 surgical PH was performed, the ‘mice were allowed to recover for 7 days and then sacrificed. Probing MspI-digested DNA with the 3“:P-labeled raf probe results in the detection of 6.7, 5.1, 3.0, and 2.3 kb bands in B6C3F1d (Figure 18) and C57BL/69 (Figure 20), and 5.1, 3.0, and 2.3 kb bands in the C3H/Bed (Figure 19). Figure 21 shows results of densitometric analysis of Figures 18, 19, 20. Because the relative intensities of the 5.1 and 3.0 kb bands varies between the 3 strains, the sum of the 2 bands was used for comparisons where possible. The 6.7 kb band is greatly diminished relative to the other major bands (5.1 and 3.0 kb) following PH of B6C3Fld mice (Figure 18, 21A). The C3H/Hed lacks the 6.7 kb band which is altered in the B6C3Fld, however, the 5.1 kb band is decreased relative to the 3.0 kb band following PH (Figure 19, Figure 21B). Like the 86C3F1d, the C57BL/69 shows decreased intensity of the 6.7 kb band relative to the sum of the 5.1 and 3.0 kb bands following PH 82 PRE POST PRE POST PRE POST PH PH PH PH PH PH 11 12 Figure 18. The methylation status of raf: B6C3Fld 7 days after PH; MspI and HpaII. Pre-PH refers to the liver sample removed at the time of PH, while post-PH refers to the liver sample removed 7 days after PH. The arrows indicate the bands seen following analysis as in Figure 11. Lanes 1 and 2, 5 and 6, and 9 and 10: DNA from 3 mice prior to PH; lanes 3 and 4, 7 and 8, and 11 and 12: DNA from the same 3 mice 7 days after PH. 83 PRE POST PRE POST PRE POST PH PH PH PH PH PH 1 2 3 4 s 6 11 12 523.1 '-9.4 -6.6 Figure 19. The methylation status of raf: C3H/Hed' 7 days after PH; MspI and HpaII. Pre-PH refers to the liver sample removed at the time of PH, while post-PH refers to the liver sample removed 7 days after PH. DNA.was analyzed as in Figure 11. Arrows point to the 5.1 and 3.0 kb bands in the MspI digests. Lanes 1 and 2, 5 and 6, and 9 and 10: DNA removed from 3 mice at the time of PH; lanes 3 and 4, 7 and 8, and 11 and 12: DNA removed from the same 3 mice 7 days after PH. 84 Figure 20. The methylation status of raf: C57BL/6Q 7 days after PH; MspI and HpaII. DNA was analyzed as in Figure 11. Arrows point to the 6.7, 5.1, and 3.0 kb bands in the MspI digests. Lanes 1 and 2, 5 and 6, 9 and 10, and 13 and 14: DNA removed from.4 mice at the time of PH; lanes 3 and 4, 7 and 8, 11 and 12, and 15 and 16: DNA removed from the same 4 mice 7 days after PH. 85 o.~1 m.N1 ¢.v1 o.m1 ¢.m1 H.m~1 fl ca mH :a hmom In mam :m hmom In man In fine; In man In hmom In men H. 86 Figure 21. Relative changes in raf band intensities following PH. A) B6C3Fld. The ratio of the intensity of the 6.7 kb band.to the sum.of the intensities of the 5.1 and 3.0 kb bands in the MspI digests on the autoradiograph pictured in Figure 18 was determined as described in the Methods section. The numbers along the x-axis refer to the individual mice as shown in Figure 18. The actual ratios determined by densitometric analysis are given in parentheses. 1 corresponds to lanes 1-4 (pre-PH: 0.0817, post-PH: 0.0091); 2 corresponds to lanes 5-8 (pre-PH: 0.0774, post-PH: 0.0023); and 3 corresponds to lanes 9-12 (pre-PH: 0.1678, post-PH: 0.0362). B) C3H/Hed. The ratio of the intensity of the 5.1 kb band to the intensity of the 3.0 kb band in the MspI digests on the autoradiograph pictured in Figure 19 was determined. 1 corresponds to lanes 1-4 (pre-PH: 0.191, post-PH: 0.117); 2 corresponds to lanes 5-8 (pre-PH: 0..326, post-PH: 0..137); and 3 corresponds to lanes 9-12 (pre-PH: 0..474, post-PH: 0..049). C) C57BL/6Q. The ratio of the intensity of the 6.7 kb band to the sum of the intensities of the 5.1 and 3.0 kb bands in the MspI digests on the autoradiograph.pictured in Figure 20 was determined. The background.in lane 11 interfered.with.image analysis, so lanes 9-12 are not included. 1 corresponds to lanes 1-4 (pre-PH: 0.015, post-PH: 0.014); 2 corresponds to lanes 5-8 (pre-PH: 0.020, post-PH: 0.005); and 3 corresponds to lanes 13-16 (pre-PH: 0.027, post-PH:0.008). b) e) RATIO OF BANDS RATIO OF BANDS RATIO OF BANDS 0.10 0.10 [:1 PHE-PH m POST-PH 0.1 8 0.00 0.04 0.00 0.00 0.40 [:1 PHE-PH W POST-PH 0.00 0.10 0.10 0.00 I: PRE-PH 0.03 m POST-PH 0.0 1 87 88 (Figure 20, 21C). The loss (or decrease) of a band detected in MspI digested DNA indicates the loss of methylation at a 5'-"°CCGG-3' site in some, but not all of the cells, i.e. there is mosaicism of methylation between ' liver cells. These results, summarized in Table 1, indicate that, in the 3 strains examined, raf becomes hypomethylated following cell proliferation stimulated by PH relative to the nascent liver. B. Methylation status of Ha-ras 7 days after partial hepatectomy To determine the methylation status of Ha-ras following PH, the blots pictured in Figure 18, 19, and.20 were stripped of the raf probe and hybridized with the 52P-labeled v-Ha-ras probe. The MspI and HpaII digestion patterns are similar in all 3 strains (B6C3F1d--Figure 22, C3H/Hed--Figure 23, C57BL/6Q--Figure 24). An approximately 15 kb band is detectible in the MspI digests of C57BL/69 DNA but not the other DNA as was found previously (Vorce and Goodman, 1989a); the intensity of this band is not altered.by treatments There is an increase in intensity of the 2.6 kb band in HpaII digests relative to the other, smaller (<2.0 kb) major band present in the HpaII digests in 2/3 of the B6C3F10' mice (Figure 22, 25A), 3/3 of the C3H/Hed mice (Figure 23, 25B), and 2/4 of the C57BL/69 mice (Figure 24, 25C) following PH. A 5'-C"°CGG-3 ' site which was present in the largely undigested DNA seen at the top of the lane has become unmethylated and is now cleaved by HpaII, resulting in increased intensity of the band at 2.6 kb. Clearly, this site is unmethylated in some 89 .mamaom menu on» :w damn mom no H0>0H as» ou 0>Huoamu coswaumuop ouoz.4zma manna: no man no mHo>oa n on» Ca confiuomop m0 mamaamsm ofiuumfioufimsoc an poseauouop mos soauoaasuoaomhn 0>Hvaaom .uoumz codawumep co>eooou mamawsm Houusoo .sowuoom mposuoa ~ .mowuwuomm ou mowed when «H you “mums mcaxsfiup may a“ >mp\mx\nm as me no coaumuumficwapm on» on mnouou mm .mm no use» on» us Hmaacm 0500 on» son“ co>oaou demand on» no yes» on nonmmaoo me maeamm unlumon on» no msumum caeumaasuoa one .ooauwuomm ou HORHQ name b an n\~ 0 on muoumu mm ! i _. oxo mxo e\o o\o mm ooxqmpmo exo «\o e\~ n\~ mm oqumnmo e\o o\o w\o m\o mm 60m\mno «\o «\o n\n e\e mm bmm\mno mxo m\o m\o m\n me hammoom m\o «\m mxm n\n mm bannomm . qm>mq czms Anm>ea «zms magnum ~umu manna: 2H men 2H so so mmammozH mmma czma_pcm usumum soaunaanuoz .H manna 90 Figure 22. The methylation status of Ha-ras: B6C3F1d 7 days after PH; MspI and HpaII. The blot pictured in Figure 18 was stripped of the raf probe and analyzed as in Figure 15. The arrow points to the band at 2.6 kb in the HpaII digests. Lanes 1 and 2, 5 and 6, and 9 and 10: DNA from 3 mice at the time of PH; lanes 3 and 4, 7 and 8, and 11 and 12: DNA from the same 3 mice 7 days after PH. 91 POST POST-PRE PH ,PRE POST PRE PH PH PH PH 9 10 11 12 PH _ l 3 2 glu— 65- 4A- 203' 200— 92 PRE POST PRE POST PRE POST PH PH PH PH PH PH 11 12 %, ‘1 ‘h' [{gm ',-’3 .l‘, In": .11 '1 xv 55;“. '.° .1. ,\. 3151' '. 13”,}. 4 '1 ’1 f Figure 23. The methylation status of Ha-ras: C3H/Had 7 days after PH; MspI and HpaII. The blot pictured in Figure 19 was stripped of the raf probe and analyzed as in Figure 15 (Lanes 1-12 are pictured). The arrow points to the 2.6 kb band in the HpaII digests. Lanes 1 and 2, 5 and 6, 9 and 10: DNA from 3 mice at the time of PH; lanes 3 and 4, 7 and 8, and 11 and 12: DNA from the same 3 mice 7 days after PH. 93 Figure 24. The methylation status of Ha-ras: C57BL/6Q 7 days after PH; MspI and HpaII. The blot pictured in Figure 20 was stripped of the raf probe and analyzed as in Figure 15. The arrow points to the 2.6 kb band in the HpaII digests. Lanes 1 and 2, 5 and 6, 9 and 10, and 13 and 14: DNA from 4 mice at the time of PH; lanes 3 and 4, 7 and 8, 11 and 12, and 15 and 16: DNA from the same 4 mice 7 days after PH. POST PH PRE POST PRE POST PRE PH PH POST PRE PH 10 11 12 13 14 15 16 PH PH PH PH 9 94 tam?“ :- r'.-"vfl ’ '- .11.? . é . I .. “ I. . . 1 I \ , f i: , "0 I' ’ . . a “ ’ . . t 4 . -' 0‘ u! ' “f 7' {delum~ ' .'.. ' ‘:_ t...’ .. 95 Figure 25. Relative changes in Ha-ras band intensities following PH. The ratio of the intensity of the 2.6 kb band to the intensity of the smaller major band in the HpaII digests on the autoradiographs pictured in Figures 22, 23, and 24 was determined as in the Methods section. The numbers along the x-axis refer to the individual mice as shown in the figures. The actual ratios determined by densitometric analysis are given in parentheses. A) 1 corresponds to Figure 22, lanes 1-4 (pre-PH: 0.063, post-PH: 0.096); 2 corresponds to lanes 5-8 (pre-PH: 0.115, post-PH: 0.495); 3 corresponds to lanes 9-12 (pre-PH: 0.295, post-PH: 0.479). B) C3H/He0'. 1 corresponds to Figure 23, lanes 1-4 (pre-PH: 0.016, post-PH: 0.079); 2 corresponds to lanes 5-8 (pre-PH: 0.061, post-PH: 0.183); and 3 corresponds to lanes 9-12 (pre-PH: 0.191, post-PH: 0.460). C) C57BL/6Q. 1 corresponds to Figure 24, lanes 1—4 (pre-PH: 0.040, post-PH: 0.042); 2 corresponds to lanes 5-8 (pre-PH: 0.058, post-PH: 0.295); 3 corresponds to lanes 9-12 (pre-PH: 0.046, post-PH: 0.087); and 4 corresponds to lanes 13-16 (pre-PH: 0.085, post-PH: 0.077). 96 IIIIIIIIIIIIIIII 'I'IIIIIIIIIIIII I'IIIIIIIIIIIIII I'llIIIIIIIIIII' IIIIIII'IIIIIIII IIIIIIIIIIIIIIII ‘I-‘~Il"‘l‘l‘l‘ IIIIIIIIIIIIIIIIIIIII I- IIIIIIIIIIIIIIIIIII III-IIIIIIIIIIIIIIIII III--- Ilr IIIIIIIIIIIIIIIIIII 0.6 0 ) 8 04° ’Dm-PH WW 0 1. 0 0.30 0.20 ‘ oe0H 0s» 0» 0>wu0H0u c0swsh0u0u 0H03_0A n .cowuo00 uposu0a 0:» ca 60930000 00 0.343050 owuu0aouems0p .an 605.5906 003 sowumfinsugnogs 0>wuud0m ~ 113 .msusoa 0m How u0wp 0:» Ga >00\mx\nm we owe no soaunuumfiswaum 0:» an p0oso:« 0H0; muoana . l . 1 1 1.1. 1 I11 1 1 -1 1 1. .1. . . . mxm mxe m\m m\e mm bauoo m\H m\m m\¢ mzoz bamnoom oH\H .m\s oamda mm bamnoom 04 1 , 11-1 5. 11.1. ,1 0208 080180 0208 080 08018: no ~unu mo 2H 0000003 2H egozH 23959502890 23939090200»: $224009. .muossa 00>Ha bauoo one hammoom a“ 08010: one may no 0H0>0A «208 can msumum cowumaasuwz .n manna 114 Figure 35. The methylation status of CCGG site in exon 12 of the raf gene in PB-induced and spontaneous B6C3F10‘ tumors. DNA from age-matched control liver (lanes 1-3) , PB-induced adenoma (lanes 4-6) , PB-induced carcinoma (lanes 7-9) , spontaneous adenoma (lanes 10-12), or spontaneous carcinoma (lanes 13-15) was digested with StuI (lanes 1,4,7,10,13), MspI then StuI (lanes 2,5,8,11,14), or HpaII then StuI (lanes 3,6,9,12,15). Double digests were performed as in Figure 14. DNA was analyzed as in Figure 11. The arrows point to the 3.2 kb bands in StuI and StuI/HpaII double digests and to 2.4 and 0.8 kb bands in StuI/MspI double digests. 115 oo.o1 mm.on. mo.~1,. mm.H1 ma ea mm NH HH ea <202Hum and B6C3Flo‘ mice (Figure 11). This indicates that the CD—1o’ also possesses the 5'-"°CCGG-3' site which is present in the C57BL/69 and the B6C3F1d, but not the C3H/Bed. There is some variation in the intensity of the 6.7 kb band between individuals which may be because the CD-l is a non-inbred mouse stock. As determined by image analysis, the ratio of the intensity of the 6.7 kb band to the sum of the intensities of the other 2 bands (5.1 and 3.0 kb) for the CD-1o‘ is most similar to that of the 86C3Flo‘ (Figure 41). This suggests that the CD-ld may have differential methylation of raf alleles similar to the B6C3Fld, mosaicism of methylation status of the raf gene between hepatocytes or the fragment responsible for the 6.7 kb band may bind the probe less avidly due to a lower sequence homology. B. Methylation status of Ha-ras The 5'-"’CCGG—3' site in Ha-ras, responsible for the approximately 15 kb band in MspI digests probed with Ha-ras in C57BL/69 but not C3H/lied (Vorce and Goodman, 1989a) , is detectible in 4/5 of the young adult CD-lo‘ (Figure 42) . Otherwise, the MspI and HpaII digests of CD-lo‘ DNA, probed for Ha-ras, appear identical to the BGC3F16, C3H/fled, and C57BL/69. 124 1 2 3 4 s 6 7 3 9 10 11 12 a I n t I ,. 1.: ' .-.-' ‘2: ‘ ‘9-4 - .4. ' W. -6.6 ~00 . a. '- -.4,, Figure 40. The methylation status of raf: CD-lo‘; MspI and HpaII. DNA was analyzed as in Figure 11. Arrows point to the 6.7, 5.1, and 3.0 kb bands seen in the MspI digests. Lanes 1 and 2, 3 and 4, 5 and 6, 7 and 8, and 9 and 10: DNA from 5 CD- 10‘ mice. 125 0.20 (g I [:1 C57BL/6 z 0.15 -- 1 < m u. -_ o 0.10 9 l- -- < 0.05 D: Figure 41. Relative intensity of the 6.7 kb band in the C57BL/69, 86C3Fld, and CD-ld raf gene. The raf gene from 6 C57BL/69 (not shown), 6 Bscamd (not shown), and 5 CD-ld (Figure 40, lanes 1,3,5,7,9) was analyzed. The ratio of the intensity of the 6.7 kb band to the sum of the intensities of the 5.1 and 3.0 kb bands in the MspI digests was determined as described in the methods section. * indicates a significant difference in the ratios of the B6C3F1d and CD-ld' bands relative to that of the C57BL/69. There is no difference between the ratios of the B6C3Fld and CD-ld bands. 126 Figure 42. The methylation status of Ha-ras: CD-ld; MspI and HpaII. DNA from the same mice as pictured in Figure 40, lanes 1-10 was analyzed as in Figure 30. 127 C. RFLP analysis of raf and Ha-ras in CD-ld CD-1d DNA contains the same Tan restriction sites as C57BL/69 DNA in the region of the raf gene detected by the probe (Figure 43A). No RFLPs in the TagI digested DNA were detected in Ha-ras in the CD-ld compared to 86C3F1d, C3H/Bed, or C57BL/6Q (Figure 438) . These data imply that rat and Ha-ras in the CD-1d are more similar to the C57BL/69 and the B6C3Fld than the C3H/Red. No RFLPs were detected between the 4 strains with EcoRI, HindIII, StuI, HhaI, or XhoI (data not shown). 6. Ha-ras and raf in Phenobarbital-induced CD-lo' Liver Tumors A. Methylation status of raf In 26 month old CD-ld mice, the raf gene is more methylated than it is in 2 month old mice; this is seen as additional bands in the MspI digests above the major 5.1 kb band in 5 out of 6 of the age-matched controls (Figure 44A, lanes 1,3,5; Figure 448, lanes 3,5). To rule out incomplete digestion, this DNA was restricted again with 50% more MspI which was added in.3 aliquots with identical results (data not shown). MspI from the same lot was also shown to digest C57BL/6Q liver DNA to completion under the same conditions (data not shown). The bands larger than 5.1 kb were either not detected or were greatly diminished in 4/5 PB-induced CD— 16 liver tumors (Figure 44A, lanes 7,9,11; Figure 448, lane 7, Table 2), indicating loss of SLJ”CCGG-3' sites in the tumor 128 Figure 43. Restriction fragment analysis of raf and Ha-ras: CD-lo‘; Tan. A) raf. DNA from B6C3Flo‘ (lane 1), C3H/Bed (lane 2) , C57BL/69 (lane 3), or CD-ld’ (lanes 4-9) was digested with 5U/ug Tan at 65°C and analyzed as in Figure 11. Arrows point to the major bands. B) Ha-ras. The same blot pictured in panel A was stripped of the raf probe and analyzed as in Figure 15. The arrows point to the major digestion products. 129 CD-l CD-l 7 57BL C 3H _2 C 1 B6C3F1 b k b) 6.6- 2 u ......»- 4.4- 21.31....- irate? ., .45. fl! 0.. .63..-- 2.3— 29 a 2.0- 130 Figure 44. The methylation status of raf: CD-ld PB-induced tumors and age-matched controls; MspI and HpaII. DNA was analyzed as in Figure 11. The arrow points to the major 5.1 kb band in MspI-digested DNA. Note the multiple bands detectible above the 5.1 kb band in the age-matched control tissue. A) Lanes 1-6: DNA from 3 age-matched controls; lanes 7-10: DNA from PB-induced liver tumors from 2 mice. B) Lanes 1-6: DNA from 3 different age-matched controls; lanes 7-12: DNA from PB-induced liver tumors from 3 mice. 131 TUMORS CONTROLS . 212 a) 2?? 222-2... ,. a 2 ”m-&EH-'3:: A. _m N M TUMORS 9 III .... 2:... nw—--!§. -_ CONTROLS oc---.r?. b) _ 3 ...-2.2: .2.- _ 1. 22.322... . J 22 ii 41"! 132 Figure 45. The methylation status of the CCGG site in exon 12 of the raf gene in CD-ld liver, PB-induced CD-ld tumors and age-matched controls. DNA from young adult CD-ld liver (lanes 1-3), age-matched control liver (lanes 4-9), and PB-induced liver tumors (lanes 10-15) was digested with StuI (lanes 1,4,7,10,13), MspI then StuI (lanes 2,5,8,11,14), or HpaII then StuI (lanes 3,6,9,12,15). Double digests were performed as in Figure 14. Southern blots were analyzed as in Figure 11. The major digestion products are noted by arrows. 133 I ch.o ...-r. II? 1 a .mw.... Tao." .. . -23 ‘7‘. s . .. .. . .. 1 Hmw I¢.e .. _ ...”... ... ...: .. “...... - 4.2 1 2-2 wk” 1~.mN ma e2 m~ Na a“ a“ a a m a m e m N u mzozzh mgczpzou chzhzcu amu=o2H14 mice should be compared for ability to methylate 5'-CCGG-3' sites both in naked DNA and in intact chromatin. 2 . Correlation between Hypomethylation, Increased Expression and Tumorigenicity The hypomethylation of raf seen in the B6C3Flo‘ after a 14 day treatment with PB was found to be accompanied by increased amounts of raf mRNA in PB-induced tumors. Both raf and Ha-ras were hypomethylated in PB-induced B6C3Fld‘ (raf--10/10, Ha-ras--7/8) and CD-ld (raf--4/5, Ha-ras--5/5) tumors. PB-induced tumors in BGCBFld and CD-ld mice also had elevated levels of raf (BGC3Fld--7/10, CD-ld--4/5) and Ha-ras (BGCBFld--8/10, CD-ld--5/5) mRNA. There might have been loss of additional methylated sites which were not detectable by MspI or HpaII digestion. The MspI site in exon 12 remained methylated on the internal cytosine in the B6C3F10‘ and methylated on both the internal and external cytosines in the CD—lo‘, indicating that the loss of methylation was neither 148 generalized nor random. .Alternatively, other changes such as relaxation of the chromatin structure or alterations in base sequence might have occurred to allow increased transcription. Loss of additional methylated sites occurred in the Ha-ras gene which might be directly responsible for the increased expression, or other changes might also have been involved. Ha-ras Ha-ras The exogenous v-Ha-ras in WB cells and WB cell- induced F344 rat liver tumors was hypomethylated relative to c-Ha-ras. The hypomethylation of Ha-ras in tumors was correlated with increased expression which was seen as elevated levels of both Ha-ras mRNA and p21 ras protein. Because transformation of cells transfected with mutated Ha-ras is dependent on the use of hypomethylated Ha-ras DNA (Borrello, et a1., 1987), it follows that a major factor in Ha-ras the tumorigenicity of WB cells was the hypomethylated state of v-Ha-ras and subsequent increase in p21 ras protein levels. Because hypomethylation has the potential to facilitate expression of genes, and there is a correlation between hypomethylation and expression in tumor tissue, it appears that the hypomethylation of these genes could play a role in carcinogenesis. Since hypomethylation seen after a 14 day administration of PB did not affect gene expression, hypomethylation alone is not sufficient alone to increase gene expression. Other, as-yet-unidentified alterations, occurred in the cells with hypomethylated proto-oncogenes during the progression to actual tumor formation. 149 3. Differences in.the Mechanisms Involved in PB-induced and Spontaneous Tumorigenesis PB is not simply accelerating the development of spontaneous tumors, but is selecting a different population of cells. In spontaneous B6C3Fld liver tumors, hypomethylation of raf was seen in 4/5 tumors, but raf mRNA levels were elevated in only one of the tumors. The methylation status and mRNA levels of Ha-ras in spontaneous tumors were similar to those in PB-induced tumors. Because there is a difference in the phenotype of spontaneous vs. PB-induced tumor cells, it is likely that the two tumor types arose via different mechanisms. In addition to the findings in this study involving differential levels of raf expression, others have found differences in the activation of the Ha-ras gene in PB-induced vs. spontaneous tumors (Fox, et a1., 1990; Rumsby, et a1., 1991). Others have also detected intrinsic differences in the response of C3H/Heo‘ and CS7BL/69 to PB promotion. C3H/He hepatocytes exhibit enhanced responsiveness to PB alone or to PB administration as a promoter following initiation by other chemicals. Primary C3H/HeNJcl hepatocyte cultures developed more (5x) colonies in the presence of 1.5 mM’ PB than C57BL/6NJcl hepatocytes which did not respond to PB administration with enhanced colony formation (Lee, et a1., 1989b). Lee and coworkers (1989a) found that the number of enzyme altered (glucose-é-phosphatase negative) foci was increased. in both C3H/HeN and CS7BL/6N’ mice which. were 150 initiated with one injection of diethylnitrosamine 20 hours after a PH and then fed 500 ppm PB in the diet for 20 weeks vs. mice that did not receive PB. However, the foci were >100x larger in CBH/HeN mice than in C57BL/6N mice. There has also been found to be interstrain variation in the response to TPA promotion of skin tumors. CD-1 and C3H mice are responsive, but CS7BL/6 mice are resistant to TPA promotion of dimethylbenz(a)anthracene-induced skin lesions (DiGiovanni, et a1., 1988). Multiple applications of TPA (8 bi-weekly treatments) resulted in a 3 fold increase in epidermal thickness in SENCAR mice (which were developed by breeding 00-1 to STS mice for 8 generations and selecting those offspring which were most sensitive to phorbol ester promotion. for’ breeding), but only a slight increase in epidermal thickness in C57BL/6 mice (DiGiovanni, et a1., 1991). In other words, TPA induces epidermal cell proliferation to a much greater extent in SENCAR mice than C57BL/6 mice. However, C57BL/6 mice respond to benzoyl peroxide promotion of dimethylbenz(a)anthracene-induced skin tumors (Reiners, et a1., 1984; DiGiovanni, et a1., 1991) suggesting that the signal transduction pathway mediated by PKC (and therefore activated by TPA) functions differently in C57BL/6 epidermal cells than in the other strains studied. It might be possible to extrapolate the theory that the epidermal cell PKC pathway is altered in C57BL/69 relative to the other strains to hepatocytes. TPA's effects on hepatocytes in vivo cannot be determined because TPA is too 151 toxic to be administered internally. It is also not known if TPA and diacylglycerol, the endogenous activator, activate PKC through the same mechanism, although that is generally assumed to be the case. There are multiple isozymes of PKC and they might be present in different amounts in different cell types (e.g., epidermal vs. endothelial) or they might be activated by different mechanisms. Therefore, the theory that the PKC pathway in C57BL/69 hepatocytes is different than in the other strains relies on several assumptions and. would require extensive additional testing for proof. PB has an effect on the signal transduction pathway involving EGF, PKC, ras p21 and Raf-1 (Figure 1). Chronic PB administration. such. as 'that. used. in. promotion. protocols decreases the response of hepatocytes in vivo to the stimulatory effects of PH (Barbason, et a1., 1983) and the response of hepatocytes in vitro to EGF (Eckl, et a1., 1988). Hepatocytes from chronically treated animals have decreased numbers of EGF-R (Eckl, et a1., 1988; Meyer and.Jirtle, 1989), decreased levels of EGF-R mRNA (Hsieh, et a1., 1988), do not respond to TPA treatment with activation and membrane translocation of PKC (Brockenbrough, et a1., 1991), and have elevated levels of transforming growth factor 31 (TGF-fil) (Meyer and J irtle, 1991) . TGF-fil is produced by liver nonparenchymal cells, binds to a specific receptor on hepatocytes, and inhibits hepatocyte proliferation (Braun, et a1., 1988). Each of these effects of chronic PB treatment would result in decreased responsiveness to cell proliferative 152 stimuli in normal hepatocytes. Therefore, the population of cells which participate in PB-induced tumorigenesis must be able to overcome the inhibitory effects of chronic PB administration. Cells with activated raf or Ha-ras have the potential to overcome PB's growth inhibition. Hampton and coworkers (1990) showed that tumor derived liver cells (cultured from tumors resulting from injection of v-raf transfected rat liver epithelial cells into athymic nude mice) could grow in medium without supplemental EGF, while the cells used to induce the tumors (transfected with v-raf) would not grow under the same conditions. Increased levels of normal or mutated Ha-ras mRNA have been found to stimulate increased expression of EGF-R mRNA which results in increased numbers of EGF-R on the cell surface (Theodorescu, et a1., 1990; Theodorescu, et a1., 1991). The EGF-R gene contains an AP-l enhancer site about 200 bp downstream of the transcription start site (Theodorescu, et a1., 1991). Elevated levels of ras p21 protein or activation of Raf-1 have been implicated in the enhanced expression of genes with AP-l enhancer sites. The presence of either activated raf or Ha-ras also has the potential to interfere with the inhibitory effects of elevated TGF-fll in hepatocytes due to chronic PB treatment. Following transfection of rat liver epithelial cells with v-raf, transformed cells are relatively resistant to the growth inhibitory effects of TGF-fil (Huggett, et a1., 1990). Some of the transformed cells have a decreased number of 153 TGF-fil receptors, while others have normal numbers of receptors, but post-receptor signalling pathways are perturbed and cell proliferation is not inhibited by TGF-fil. v-Ha-ras or T24 ras transfected rat liver epithelial cells are also resistant to the growth inhibitory effects of TGF-fi (Houck, et a1., 1989). These studies examined the effects of oncogenes which were activated by mutation, but since increased expression of the corresponding proto-oncogenes can transform cells, it is likely that oncogenes activated by deregulation of expression might act similarly. SUMMARY AND CONCLUSIONS Four major findings resulted from this study. First, there are intrinsic differences in the raf gene between the C3H/Hed and C57BL/69 mice which increase the potential for aberrant expression of the raf gene in the C3H/Hed. Therefore, the C3H/Had mouse might be one step further along the multistep pathway involved in carcinogenesis than the C57BL/69 due to hypomethylation of the raf proto-oncogene in the nascent liver. The B6C3Flo‘ inherits a methylated raf allele from the C57BL/69 and an unmethylated allele form the C3H/Hed. This gives credence to the involvement of hypomethylated raf in carcinogenesis since the spontaneous liver tumor rate of the BGC3Fld is intermediate between that of the C3H/Had and the C57BL/69. BGC3Fld and CD—ld’mice have similar methylation patterns of raf and similar incidences of spontaneous liver tumor development. Second, methylation at the external cytosine of one or more 5'-CCGG-3' sites appears to be more important than sites which are methylated on the internal cytosine in the regulation of expression of the raf gene. The 5'-mbCGG-3' site which is present in the C57BL/69 but not the C3H/fled is apparently difficult to maintain. Methylation at this site is decreased following cell proliferation induced by PH in the 154 155 BGCBFld and CS7BL/6Q and after 14 days of 500 ppm. PB administration in the CS7BL/69. Third, there is a correlation between hypomethylation and increased expression of raf and Ha-ras in PB-induced tumors. Additionally, there is a correlation between hypomethylation of v-Ha-ras, levels of ras:mRNA.and p21, and.tumorigenicity'of cells transfected with v-Ha-ras (WEN-r” cells). While hypomethylation alone is not sufficient for increased expression, this suggests that increased expression of raf and/or Ha-ras genes, subsequent to hypomethylation and additional as-yet-unidentified cellular changes, is a mechanism involved in carcinogenesis. The fourth major finding confirms the findings of others that PB administration does not simply accelerate the development of spontaneous liver tumors. While both raf and Ha-ras are hypomethylated in spontaneous BGC3F10’ liver tumors, elevated levels of raf mRNA are not detected in the majority of the tumors, indicating a difference in.the phenotype of the cells involved in spontaneous vs. PB-induced.tumorsu ‘While it is possible that increased raf mRNA levels are a consequence of PB administration and not causal in tumorigenesis, it appears that PB administration results in selective expansion of cells with either increased potential for raf expression or actual elevated levels of raf mRNA. Based on these results, it can be concluded that hypomethylation of proto-oncogenes plays a role in the potential of mice to develop liver tumors. There is an 156 intrinsic inability of the C3H/Hed mouse to methylate S'J“CCGG-3' sites in the raf gene relative to the C57BL/69 mouse. A defect in this pathway might be one of the mechanisms underlying the increased propensity of C3H/Hed and BGCBFlo‘ mice to develop tumors. Increased raf expression allows cells to overcome PB's growth inhibitory effects and thus appears to be a major factor in PB-induced mouse liver carcinogenesis. LIST OF REFERENCES LIST OF REFERENCES Althaus, F.R., Eichenberger, R., and Pitot, H.C. (1986). Tumor-promoting barbiturates act on DNA repair of cultured hepatocytes. Mutat. Res. 173:147-152. Ames, B.N. and.Gold, L.S. (1990). Chemical carcinogenesis: too many rodent carcinogens. Proc. Natl. Acad. Sci. USA 87:7772-7776. Anderson, N.G., Li, P., Marsden, L.A., Williams, N., Roberts, T.M., and Sturgill, T.W. (1991). Raf-1 is a potential substrate for mitogen-activated protein kinase in vivo. Biochem. J. 2773573-576. Antequera, R., Boyes, J., and Bird, A. (1990). High levels of de novo methylation and altered chromatin structure a CpG islands in cell lines. Cell 62:503-514. App, H., Hazan, R., Zilberstein, A., Ullrich, A., Schlessinger, J., and Rapp, U.R. (1991) . Epidermal growth factor (EGF) stimulates association and kinase activity of Raf-1 with the EGF receptor. Mol. Cell. Biol. 11:913- 919. Ashby, J. and Tennant, R.W. (1988). Chemical structure, Salmonella.mutagenicity and extent of carcinogenicity as indicators of genotoxic carcinogenesis among 222 chemicals tested in rodents by the U.S. NCI/NTP. Mutat. Res. 204817-115. Barbason, H., Rassenfosse, C., and Betz, E.H. (1983). Promotion mechanism of phenobarbital and partial hepatectomy in DENA hepatocarcinogenesis effect. Br. J. Cancer 47:517-525. Barbu, V. and Dautry, F. (1989). Northern blot normalization with a 285 rRNA oligonucleotide probe. Nucl. Acid. Res. 17:7115. Beck, T.W., Huleihel, Mg, Gunnell, H., Bonner; T.I., and Rapp, U.R. (1987) . The complete coding sequence of the human A- raf-l oncogene and transforming activity of a human A-raf carrying retrovirus. Nucl. Acid. Res. 15:595-609. 157 158 Becker, F.F. (1982). Morphological classification of mouse liver tumors based on biological characteristics. Cancer Research 42:3918-3923. Becker, P.B., Siegfried, R., and Schutz, G. (1987). Genomic footprinting reveals cell type-specific DNA binding of ubiquitous factors. Cell 51:435-443. Beenken, S.W., Karsenty, G., Raycroft, L., and Lozano, G. (1991). An intron binding protein is required for transformation ability of p53. Nucl. Acid. Res. 19:4747- 4752. Beer, D.G., Neveu, M.J., Paul, D.L., Rapp, U.R., and Pitot, H.C. (1988). Expression of the c-raf protooncogene, y- glutamyltranspeptidase, and gap junction protein in rat liver neoplasms. Cancer Res. 48:1610-1617. Bird, A.P. (1986). CpG-rich islands and the function of DNA methylation. Nature (London) 321:209-213. Bird, A.P. and Southern, E.M. (1977). Use of restriction enzymes to study eukaryotic DNA methylation: I. the methylation pattern in ribosomal DNA from Xenopus laevis. J. M01. Biol. 118827-47. Bishop, J.M. (1991). Molecular themes in oncogenesis. Cell 64:235-248. Blumberg, P.M. (1991). Complexities of the protein kinase C pathway. Mol. Carcinogenesis 4:339-344. Bonner, T.I., Oppermann, H., Seeburg, P., Kerby, S.B., Gunnell, M.A., Young, A.C., and Rapp, U.R. (1986). The complete coding sequence of the human raf oncogene and the corresponding structure of the c-raf-l gene. Nucl. Acid. Res. 14:1009-1015. Borrello, M.G., Pierottti, M.A., Bongarzone, I., Donghi, R., Mondellini, P., and Della Porta, G. (1987). DNA methylation affecting the transforming activity of the human Ha-ras oncogene. Cancer Res. 47:75-79. Boyd, J.A. and Barrett, J.C. (1990). Genetic and cellular basis of multistep carcinogenesis. Pharmac. Ther. 46:469- 486. Boyes, J. and Bird, A. (1991). DNA methylation inhibits transcription indirectly via a methyl-CpG binding protein. Cell 64:1123-1143. 159 Bradford, M.M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 728248-254. Braun, L., Mead, J.E., Panzica, M., Mikumo, R., Bell, G.I. and Fausto, N. (1988) . Transforming growth factor 6 mRNA increases during liver regeneration: a possible paracrine mechanism of growth regulation. Proc. Natl. Acad. Sci. USA 8581539-1543. Brinster, R.L., Allen, J.M., Behringer, R.R., Gelinas, R.E., and Palmiter, R. D. (1988) . Introns increase transcriptional efficiency in transgenic mice. Proc. Natl. Acad. Sci. USA 858836-840. Brockenbrough, J.S., Meyer, S.A., Li, C., and Jirtle, R.L. (1991). Reversible and phorbol ester-specific defect of protein kinase C translocation in hepatocytes isolated from phenobarbital-treated.rats. Cancer'Res. 51:130-136. Buchmann, A., Bauer-Hofmann, R., Mahr, J., Drinkwater, N.R., Luz, A. , and Schwarz, M. (1991) . Mutational activation of the c-Ha-ras gene in liver tumors of different rodent strains: correlation with susceptibility to hepatocarcinogenesis. Proc. Natl. Acad. Sci. USA 88:911- 915. Busser, M.-T. and Lutz, W.K. (1987). Stimulation of DNA synthesis in rat and mouse liver by various tumor promoters. Carcinogenesis 8:1433-1437. Carroll, M.P., Clark-Lewis, I., Rapp, U.R. and May, W.S. (1990) . Interleukin-3 and granulocyte-macrophage colony- stimulating factor mediate rapid phosphorylation and activation of cytosolic c-raf. J. Biol. Chem. 265:19812- 19817. Cedar, H. (1988). DNA methylation and gene activity. Cell 5383-4. Cedar, H. and Razin, A. (1990). DNA methylation and development. Biochim. Biophys. Acta 1049:1-8. Chandler, L.A., Ghazi, H., Jones, P.A., Boukamp, P., and Fusenig, N.E. (1987) . Allele-specific methylation of the human c-Ha-ras-l gene. Cell 50:711-717. Chang, E.H., Furth, M.E., Scolnick, E.H., and Lowy, D.R. (1982) . Tumorigenic transformation of mammalian cells induced by a normal human gene homologous to the oncogene of Harvey murine sarcoma virus. Nature (London) 297:479- 483. 160 Chernozemski, I.N. and. ‘Warwick, G.P. (1970). Liver regeneration and induction of hepatomas in BéAFl mice by urethan. Cancer Res. 30:2685-2690. Chirgwin, J.M., Przyble, E., MacDonald, R.J., and Rutter, W.J. (1979) . Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry 1885294-5299. ' Chomczynski, P. and Sacchi, N. (1987). Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol- chloroform extraction. Analytical Biochemistry 162:156- 159. Clayson, 0.8. (1987). The need for biological risk assessment in reaching decisions about carcinogens. Mutat. Res. 1858243-269. Cohen, S.M. and Ellwein, L.B. (1990). Cell proliferation in carcinogenesis. Science 249:1007-1011. Comb, M. and Goodman, H.M. (1990). CpG methylation inhibits proenkephalin gene expression and binding of the transcription factor AP-2. Nucl. Acid. Res. 18:3975-3982. Cox, R., Goorha, S., and Irving, C.C. (1988). Inhibition of DNA methylase activity by acrolein. Carcinogenesis 9:463- 465. Creusot, F., Acs, G., and Christman, J.K. (1982). Inhibition of DNA methyltransferase and induction of Friend erythroleukemia cell differentiation by 5-azacytidine and S-aza-2'-deoxycytidine. J. Biol. Chem. 257:2041-2048. Devereux, J., Haeberli, P., and Smithies, O. (1984). A comprehensive set of sequence analysis programs for the VAX. Nucl. Acid. Res. 12:387-395. Dhar, R., Ellis, R.W., Shih, T.Y., Oroszlan, S., Shapiro, B., Maizel, J., Lowy, D., and.Scolnick, E. (1982). Nucleotide sequence of the p21 transforming protein of Harvey murine sarcoma virus. Science 217:934-937. DiGiovanni, J., Naito, M., and Chenicek, K.J. (1988). Genetic factors controlling susceptibility to skin tumor promotion in mice. In Langenbach, R., Elmore, E., and Barrett, J.C. (eds.) Tumor Promoters: Biological Approaches for Mechanistic Studies and Assay Systems. Raven Press, New York, pp.51-69. 161 DiGiovanni, J., Walker, S.C., Beltran, L., Naito, M., and Eastin, Jr., W.C. (1991). Evidence for a common genetic pathway controlling susceptibility to mouse skin tumor promotion by diverse classes of promoting agents. Cancer Res. 5181398-1405. Dizik, M., Christman, J.K. and Wainfain, E. (1991). Alterations in expression and methylation of specific genes in livers of rats fed a cancer promoting methyl- deficient diet. Carcinogenesis 12:1307-1312. Doerfler, W. (1983). DNA methylation and gene activity. Ann. Rev. Biochem. 52:93-124. Dragani, T.A., Manenti, G., Colombo, B.M., Falvella, F.S., Gariboldi, M., Pierotti, M.A., Della Porta, G. (1991). Incidence of mutations at codon 61 of the Ha-ras gene in liver tumors of mice genetically susceptible and resistant to hepatocarcinogenesis. Oncogene 6:333-338. Drinkwater, N.R. and Ginsler, J. (1986). Genetic control of hepatocarcinogenesis in C57BL/6J and C3H/He inbred.mice. Carcinogenesis 7:1701-1707. Eaton, G.J., Johnson, F.N., Custer, R.P., and Crane, A.R. (1980). The Icr:Ha(ICR) mouse: a current account of breeding, mutations, diseases and mortality. Lab. An. 14:17-24. Eckl, P.M., Meyer, S.A., Whitcombe, W.R., and Jirtle, R.L. (1988) . Phenobarbital reduces EGF receptors and the ability of physiological concentrations of calcium to suppress hepatocyte proliferation. Carcinogenesis 8:479- 483. Edwards, A.M. and Lucas, C.M. (1985). Phenobarbital and some other liver tumor promoters stimulate DNA synthesis in cultured rat hepatocytes. Biochem. Biophys. Res. Commun. 131:103-108. Farber, E. (1984) . The multistep nature of cancer development. Cancer Res. 44:4217-4223. Fausto, N. and Mead, J.E. (1989). Regulation of liver growth: proto-oncogenes and transforming growth factors. Lab. Invest. 60:4-13. Feinberg, A.P. and Vogelstein, B. (1983). A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 132:6-13. 162 Feinberg, A.P. and Vogelstein, B. (1984). Addendum: A technique for radiolabeling DNA.restriction.endonuclease fragments to high specific activity. Anal. Biochem. 1378266-267. Fox, T.R., Schumann, A.M., Watanabe, P.G., Yano, B.L., Maher, V.M., and.McCormick, J.J. (1990). Mutational analysis of the H-ras oncogene in spontaneous C57BL/6 x C3H/He mouse liver tumors and tumors induced with genotoxic and nongenotoxic hepatocarcinogens. Cancer Res. 50:4014-4019 . Furth, .M.E., Aldrich, T.H., and. Cordon-Cardo C. (1987). Expression of ras proto-oncogene proteins in normal human tissue. Oncogene 1:47-58. Ghazi, H., Magewu,.A.N., Gonzales, F., and.Jones, P.A. (1990). Changes in the allelic methylation.patterns of c-Ha-ras- 1, insulin and retinoblastoma genes in human development. Dev. 1990 Suppl.:115-123. Gingeras, T.R., Myers, P.A., Olson, J.A., Hanberg, F.A., and Roberts, R.J . (1977) . A new specific endonuclease present in xanthomonas holcicola, Kanthomonas papavericola, and Brevibacterium luteum. J. Mol. Biol. 118:113-122. Gjerset, R.A., and Martin, Jr., D.W. (1982). Presence of’a DNA demethylating activity in the nucleus of murine erythroleukemic cells. J. Biol. Chem. 257:8581=8583. Goyette, M., Petropoulos, C.J., Shank, P.R., and Fausto, N. (1983). Expression of a cellular oncogene during liver regeneration. Science 219:510-512. Hall, J. (1990). Genomic imprinting: review and relevance to human diseases. Am. J. Hum. Genet. 46:857-783. Hampton, L.L., Worland, P.J., Yu, B., Thorgeirsson, S.S., and Huggett, A.C. (1990). Expression of growth related.genes during tumor progression in v-raf-transformed rat liver epithelial cells. Cancer Res. 50:7460-7467. Hanigan, M.H., Kemp, C.J., Ginsler, J.J., and Drinkwater, N.R. (1988). Rapid growth of preneoplastic lesions in hepatocarcinogen-sensitive C3H/HeJ male mice relative to C57BL/6J male mice. Carcinogenesis 9:885-890. Hare, J.R. and Taylor, J.H. (1989). Methylation in Eucaryotes influences the repair of G/T and A/C basepair mismatches. Cell Biophysics 15:29-40. 163 Heidecker, G., Huleihel, M., Cleveland, J.Iu, Kolch, W., Beck, T.W., Lloyd, P., Pawson, T., and Rapp, U.R. (1990). Mutational activation of c-raf-l and definition of the minimal transforming sequence. Mol. Cell. Biol. 10:2503- 2512. Higgins, G.M. and Anderson, R.M. (1931). Experimental pathology of the liver. I. restoration of the liver of the white rat following partial surgical removal. Arch. Path. 128186-202. Hinds, P., Finlay, C., and Levine, A.J. (1989). Mutation is required to activate the p53 gene for cooperation with the ras oncogene and transformation. J. Virol. 63:739- 746. Holliday, R. (1987). The inheritance of epigenetic defects. Science 238:163-170. Holliday, R. (1990). Mechanisms for the control of gene activity during development. Biol. Rev. 65:431-471. Houck, R.A., Michalopoulos, G.K., and Strom, S.C. (1989). Introduction of a Ha-ras oncogene into rat liver epithelial cells and parenchymal hepatocytes confers resistance to the growth inhibitory effects of TGF-fl. Oncogene 4:19-25. Hsieh, L.L., Peraino, C., and Weinstein, I.B. (1988). Expression of endogenous retrovirus-like sequences and cellular oncogenes during phenobarbital treatment and regeneration in rat liver. Cancer Res. 48:265-269. Huber, B.E. and. Cordingley, M.G. (1988). Expression. and phenotypic alterations caused by an inducible transforming ras oncogene introduced into rat liver epithelial cells. Oncogene 3:245-256. Huggett, A.C., Hampton, L.L., Ford, C.P., Wirth, P.J., and Thorgeirsson, 8.8. (1990). Altered responsiveness of rat liver epithelial cells to transforming growth factor pl following their transformation with v-raf. Cancer Res. 50 8 7468-7475 . Hurlin, P.J., Maher, V.M., and McCormick, J.J. (1989). Malignant transformation of human fibroblasts caused by expression of a transfected T24 HRAS oncogene. Proc. Natl. Acad. Sci. USA 86:187-191. Ikawa, S., Fukui, M., Ueyama, Y., Tamaoki, N., Yamaoto, T., and Toyoshima, K. (1988). B-raf, a new member of the raf family, is activated by DNA rearrangement. Mol. Cell. Biol. 882651-2654. 164 Ishikawa, F., Takaku, F., Hayashi, H., Nagao, M., and Sugimura, T. (1986) . Activation of c-raf during transfection of hepatocellular carcinoma DNA. Proc. Natl. Acad. Sci. USA 8383209-3212. Itze, L., ‘Vesselinovitch, S.D., and. Rao, K.V.N. (1973). Estimation of the rate of DNA in newborn, regenerating, and intact mouse liver. Physiol. Bohemoslov. 22:457. Itze, L., ‘Vesselinovitch, S.D., and. Rao, K.V.N. (1975). Macromolecular turnover in mice (C57BL x C3H) liver, thymus and kidney after partial hepatectomy. Physiol. Bohemoslov. 24:9-22. Jirtle, R.L. and Michalopoulos, G. (1986). Enhancement of the clonability of adult parenchymal hepatocytes with the liver tumor promoter phenobarbital. Carcinogenesis 781813-1817. Jones, P.A. (1986). DNA methylation and cancer. Cancer Res. 46:461-466. Jones, P.A. and Buckley, J.D. (1990). The role of DNA methylation in cancer. Adv. Cancer Res. 54:1-23. Jones, P.A. and Taylor, S.M. (1980) . Cellular differentiation, cytidine analogs and DNA methylation. Cell 20:85-93. Jones, P.A., Wolkowicz, M.J., Rideout, III, W.M., Gonzales, F.A., Marziasz, C.M., Coetze, G.A., and Tapscott, S.J. (1990). De novo methylation of the MyoDl CpG island during the establishment of immortal cell lines. Proc. Natl. Acad. Sci. 8786117-6121. Kaufmann, W.R., Ririe, D.G., and Kaufman, D.G. (1988). Phenobarbital-dependent proliferation of putative initiated rat hepatocytes. Carcinogenesis 9:779-782. Kaufmann, W.K. , Tsao, M.-S. , and Novicki, D.L. (1986) . In vitro colonization ability appears soon after initiation of hepatocarcinogenesis in the rat. Carcinogenesis 7:669- 671. Klaunig, J.E. and.Ruch, R.J. (1987). Role of cyclic AMP in the inhibition of mouse hepatocyte intercellular communication by liver tumor promoters. Toxicol. Appl. Pharmacol. 91:159-170. Klaunig, J.E., Ruch, R.J., and Weghorst, C.M. (1990). Comparative effects of phenobarbital, DDT, and lindane on mouse hepatocyte gap j unctional intercel lular communication. Toxicol. Appl. Pharmacol. 102:553-563. 165 Klaunig, J.E., Weghorst, C.M., and Pereira, M.A. (1987). Effect of the age of B6C3F1 mice on phenobarbital promotion of diethylnitrosamine-initiated liver tumors. Toxicol. Appl. Pharmacol. 90:79-85. Klaunig, J.E. , Weghorst, C.M. , and Pereira, M.A. (1988) . Effect of phenobarbital on diethylnitrosamine and dimethylnitrosamine induced hepatocellular tumors in male B6C3F1 mice. Cancer Lett. 428133-139. Klaunig, J.E., Siglin, J.C., Schafer, L.D., Hartnett, J.A., Weghorst, C.M., Olson, M.J., Hampton, J.A. (1991). Correlation between species and tissue sensitivity to chemical carcinogenesis in rodents and the induction of DNA synthesis. In Butterworth, B.E., Slaga, T.J., Farland, W., McClain, M. (eds). Chemically Induced Cell Proliferation, Wiley-Liss, Inc., New York, NY, pp. 185- 194. Kong, X.-B., Tong, W.P., and Chou, T.-C. (1991). Induction of deoxycytidine kinase be 5-azacytidine in an HL-60 cell line resistant to arabinosylcytosine. Mol. Pharmacol. 398250-257. Laemmli, U.K. (1970). Cleavage of the structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227:680-685. Land, H., Parada, L.F., andflWeinberg, R.A. (1983). Tumorigenic conversion of primary embryo fibroblasts requires at least two cooperating oncogenes. Nature (Lond.) 304:596- 602. - Langner, K.-D., Vardimon, L., Renz, D., and Doerfler, W. (1984). DNA methylation of three 5' C-C-G-G 3' sites in the promoter and 5' region inactivate the E2a gene of adenovirus type 2. Proc. Natl. Acad. Sci. USA 81:2950- 2954. Lee, G.-H., Nomura, K., Kanda, H., Kusakabe, M., Yoshiki, A., Sakakura, R., and Kitagawa, T. (1991). Strain specific sensitivity to diethylnitrosamine-induced carcinogenesis is maintained in hepatocytes of C3H/HeN<->C57BL/6N chimeric mice. Cancer Res. 51:3257-3260. Lee, G.-H., Nomura, K., and Kitagawa, T. (1989a). Comparative study’ of‘ diethylnitrosamine-initiated two-stage hepatocarcinogenesis in C3H, C57BL and BALB mice promoted by variousfihepatopromoters. Carcinogenesis 10:2227-2230. 166 Lee, G.-H., Sawada, N., Mochizuki, Y., Nomura, R., and Kitagawa, T. (1989b) . Immortal epithelial cells of normal C3H mouse liver in culture: possible precursor population for spontaneous hepatocellular carcinoma. Cancer Res. 498403-409. Levine, A., Cantoni, G.L., and.Razin, A. (1991). Inhibition of promoter activity by methylation: possible involvement of protein mediators. Proc. Natl. Acad. Sci. USA 88:6515- 6518. Lin, E.L.C., Klaunig, J.E., Mattox, J.K., Weghorst, C.M., McFarland, E.H. and Pereira, M.A. (1989). Comparison of the effects of acute and subacute treatment of phenobarbital in different strains of mice. Cancer Lett. 48843-51. Lozano, G. and.Levine,.A.J. (1991). Tissue specific‘expression of p53 in transgenic mice is regulated by intron sequences. Mol. Carcinogenesis 4:3-9. Lynch, C.J. (1969). The so-called Swiss mouse. Lab. An. Care 19:214-220. Mann, M.B. and Smith, H.O. (1977). Specificity of Hpa II and HaeIII DNA methylases. Nucl. Acid. Res. 4:4211-4221. Mark, G.E. and Rapp, U.R. (1984). Primary structure of v-raf: relatedness to the src family of oncogenes. Science 224:285-289. Maronpot, R.R., Haseman, J.K., Boorman, G.A., Eustis, S.E., Rao, G.N., and.Huff, J.E. (1987). Liver lesions in B6C3F1 mice: the National Toxicology Program, experience and position. Arch. Toxicol. Suppl. 10:10-26. IMays-Hoopes, L.L., Brown, A., and. Huang, R.C.C. (1983). Methylation and rearrangement of mouse intracisternal A particle genes in development, aging, and myeloma. Mol. Cell. Biol. 3:1371-1380. Mays-Hoopes, L.L., Chao, W., Butcher, H.C., and Huang, R.C.C. (1985). Decreased methylation of the major mouse long interspersed repeated DNA during aging and in myeloma cells. Dev. Genetics. 7:65-73. McGowan, R., Campbell, R., Peterson, A., and Sapienza, C. (1989). Cellular mosaicism in the methylation and expression of hemizygous loci in the mouse. Gene. Dev. 381669-1676. 167 Meehan, R.R., Lewis, J.D., McKay, S., Kleiner, E.L., and Bird, A.P. (1989). Identification of a mammalian protein that binds specifically to DNA containing methylated Cst. Cell 588499-507. Meyer, S.A. and Jirtle, R.L. (1991). Liver tumor promotion: effect of phenobarbital on EGF and protein kinase C signal transduction and transforming growth factor-31 expression. Dig. Dis. Sci. 36:659-668. Meyer, S.A. and Jirtle, R.L. (1989). Phenobarbital decreases hepatocyte EGF receptor expression independent of protein kinase C activation. Biochem. Biophys. Res. Comm. 158:652-659. Michalowsky, L.A. and Jones, P.A. (1989a). Gene structure and transcription in mouse cells with extensively demethylated DNA. Mol. Cell. Biol. 9:885-892. Michalowsky, L.A. and Jones, P.A. (1989b) . DNA methylation and differentiation. Env. Health Perspect. 80:189-197. Moelling, K., Heimann, B., Beimling, P., Rapp, U.R., and Sander, T. (1984) . Serine- and threonine-specific protein kinase activities of purified gag-mil and gag-raf proteins. Nature (London) 312:558-561. Morrison, D.K., Kaplan, D.R., Escobedo, J.A., Rapp, U.R., Roberts, T.M., and Williams, L.T. (1989). Direct activation of the serine/threonine kinase activity of Raf-1 through tyrosine phosphorylation by the PDGF 8- receptor. Cell 58:649-657. Morrison, D.K., Kaplan, D.R., Rapp, U., and Roberts, T.M. (1988). Signal transduction from membrane to cytoplasm: growth factors and. membrane-bound oncogene products increase Raf-1 phosphorylation and associated protein kinase activity. Proc. Natl. Acad. Sci. USA 85:8855-8859. Munzel, P.A., Pfohl-Leskowicz, A., Rohrdanz, E., Keith, G., Dirheimer, G. , and Bock, K.W. (1991) . Site-specific hypomethylation of c-myc protooncogene in liver nodules and inhibition of DNA methylation by N-nitrosomorpholine. Biochem. Pharmacol. 42:365-371. Murchie, A.I.H. and Lilley, D.M.J. (1989). Base methylation and local DNA helix stability: effect on the kinetics of cruciform extrusion. J. Mol. Biol. 205:593-602 Nishizuka, Y. (1986). Studies and perspectives of protein kinase C. Science 233:305-312. 168 Ono, T., Tawa, R., Shinya, K., Hirose, S., and Okaka, S. (1986) . Methylation of the c-myc gene changes during aging process of mice. Biochem. Biophys. Res. Comm. 13981299-1304. Ono, T., Takahashi, N., and Okada, S. (1989). Age-associated changes in DNA methylation and mRNA level of the c-myc gene in spleen and liver of mice. Mutat. Res. 219:39-50. Palmiter, R.D., Sandgre, E.P., Avarbock, M.R., Allen, D.D., and Brinster, R.L. (1991) . Heterologous introns can enhance expression of transgenes in mice. Proc. Natl. Acad. Sci. USA 888478-482. Peraino, C., Staffeldt, E.F., Haugen, D.A., Lombard, Ins., Stevens, F.J., and.Fry, R.J.M. (1980). Effects of varying the dietary concentration of phenobarbital on its enhancement of 2-acety1aminofluorene-induced hepatic tumors. Cancer Res. 40:3268-3273. Pereira, M.A., Herren-Freund, S.L., and Long, R.E. (1986a). Dose-response relationship'ofjphenobarbital promotion of diethylnitrosamine initiated.tumors in rat liver. Cancer Lett. 328305-311. Pereira, M.A., Klaunig, J.E., Herren-Freund, S.L., and Ruch, R.J. (1986b) . Effect of phenobarbital on the development of liver tumors in juvenile and adult mice. J. Natl. Cancer Res. 778449-452. Pitot, H.C. and Sirca, A.E. (1980). The stages of initiation and promotion in hepatocarcinogenesis. Biochim. Biophys. Acta 605:191-215. Pitot, H.C., Goldsworthy, T.L., Moran, S., Kennan, W., Glauert, M.P., Maronpot, R.R., and Campbell, H.A. (1987) . A method of quantitate the relative initiating and promoting potencies of hepatocarcinogenic agents in their dose-response relationships to altered hepatic foci. Carcinogenesis 8:1491-1499. Preston-Martin, S., Pike, M.C., Ross, R.R., Jones, P.A., and Henderson, B.E. (1990). Increased cell division as a cause of human cancer. Cancer Res. 50:7415-7421. Pulciani, S., Santos, E., Long, L.K., Sorrentino, V., and Barbacid, M. (1985) . ras Gene amplification and malignant transformation. Molec. Cell. Biol. 5:2836-2841. Rapp, U.R. (1991). Role of Raf-1 serine/threonine protein kinase in growth factor signal transduction. Oncogene 68495-500. 169 Rapp, U.R., Cleveland, J.L., Bonner, T.I., and Storm, S.M. (1988). The raf oncogenes. In Reddy, E.P., Skalka, A.M., and Curran, T. (eds.) The Oncogene Handbook, Elsevier Science Publishers B.V. (Biomedical Division), pp 213- 253. Rapp, U.R., Goldsborough, M.D., Mark, G.E., Bonner, T.I., Groffen, J., Reynolds, Jr., F.H., and Stephenson, J.R. (1983a). Structure and biological activity of v-raf, a unique oncogene transduced by a retrovirus. Proc. Natl. Acad. SCi. USA 8084218-4222. Rapp, U.R., Reynolds, F.H., and Stephenson, J.R. (1983b). New mammalian transforming retrovirus: demonstration of a polyprotein gene product. J. Virol. 45:914-924. Razin, A. and Cedar, H. (1991). DNA methylation and gene expression. Microbiol. Rev. 55:451-458. Razin, A., Szyf, M., Kafri, T., Roll, M., Gilof, H., Scarpa, S., Carlotti, D., and Cantoni, G.L. (1986). Replacement of 5-methylcytosine by cytosine: a possible mechanism for transient DNA demethylation during differentiation. Proc. Natl. Acad. Sci. USA 8382827-2831. 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 (London) 300:149-152. Reik, W., Collick, A., Norris, M.L., Barton, S.C., and Surani, M.A. (1987) . Genomic imprinting determines methylation of parental alleles in transgenic mice. Nature (London) 3288248-251. Reiners, Jr., J.J., Nesnow, S., and.Slaga, T.J. (1984). Murine susceptibility to two-stage skin carcinogenesis is influenced by the agent used for promotion. Carcinogenesis 5:301-307. Reynolds, S.H., Stowers, S.J., Patterson, R.M., Maronpot, R.R., Aaronson, S.A., and Anderson, M.W. (1987). Activated oncogenes in B6C3F1 mouse liver tumors: implications for risk assessment. Science 237:1309-1316. Richman, R.A., Claus, T.H., Pilkis, S.J., and Friedman, D.L. (1976). Hormonal stimulation of DNA synthesis in.primary cultures of rat hepatocytes. Proc. Natl. Acad. Sci. USA 7383589-3593. Rideout, W.M. III, Coetzee, G.A., Olumi, A.E., Jones, P.A. (1990). 5-methylcytosine as an endogenous mutagen in the human LDL receptor and p53 genes. Science 249:1288-1290. 170 Riggs, A.D. and Jones, P.A. (1983). 5-Methylcytosine, gene regulation and cancer. Adv. Cancer Res. 40:1-30. Rimoldi, D., Srikantan, V., Wilson, V.L., Bassin, R.H., and Samid, D. (1991) . Increased sensitivity of nontumorigenic fibroblasts expressing ras or myc oncogenes to malignant transformation induced by 5-aza-2'-deoxycytidine. Cancer Res. 518324-330. Rubin, R.A. and Modrich, P. (1980). Purification and properties of EcoRI endonuclease. Methods Enzymol. 65:96- 103. Ruch, R.J., Klaunig, J.E., and Pereira, M.A. (1987). Inhibition of intercellular communication between mouse hepatocytes by tumor promoters. Toxicol. Appl. Pharmacol. 878111-120. Ruchirawat, M., Becker, F.F., and. Lapeyre, J-N. (1984). Mechanism of rat liver DNA.methyltransferase interaction with anti-benzo[a]pyrenediol epoxide modified DNA templates. Biochemistry 23:5426-5432. Rumsby, P.C., Barrass, N.C., Phillimore, H.E., and.Evans, J.G. (1991). Analysis of the Ha-ras oncogene in C3H/He mouse liver tumours derived spontaneously or induced with diethylnitrosamine or phenobarbitone. Carcinogenesis 1282331-2336. Sanford, J.P., Clark, H.J., Chapman, V.M., and Rossant, J. (1987). Differences in DNA methylation during oogenesis and spermatogenesis and their persistence during early embryogenesis in the mouse. Genes Dev. 1:1039-1046. Santi, D.V., Norment, A., and Garrett, G.E. (1984). Covalent bond formation between a DNA-cytosine methyltransferase and DNA containing 5-azacytosine. Proc. Natl. Acad. Sci. USA 8186993-6997. Sapienza, C. (1991). Genome imprinting and carcinogenesis. Biochim. Biophys. Acta 1072:51-61. Sapienza, C., Paquette, J., Tran, T.H., and Peterson, A. (1989). Epigenetic and genetic factors affect transgene methylation imprinting. Dev. 107:165-168. Sapienza, C., Peterson, A.C., Rossant, J., and Balling, R. (1987). Degree of methylation of transgenes is dependent on gamete of origin. Nature (London) 328:251-254. 171 Sawada, N., Staecker, J.L., and.Pitot, H.C. (1987). Effects of tumor-promoting agents 12-O-tetradecanoylphorbol-13- acetate and phenobarbital on DNA synthesis of rat hepatocytes in primary culture. Cancer Res. 47:5665-5671. Schorderet, D.F. and Gartler, S.M. (1990). Absence of methylation at HpaII sites in three human genomic tRNA sequences. Nucl. Acid. Res. 18:6965-6969. Seldon, R.F. (1987). Analysis of DNA sequences by blotting and hybridization. In Ausubel, P.M., Brent, R., Kinston, R.E., Moore, D.D., Seidman, .J.G., Smith, J.A., and Struhl, K. (eds.) Current Protocols in Molecular Biology, John Wiley and Sons, New York, pp 2.2.1-2.2.3. Seyama, T., Godwin, A.K., DiPietro, M., Winokur, T.S., Lebovitz, R.M. and Lieberman, M.W. (1988). In vitro and in vivo regulation of liver epithelial cells carrying a metallothionein-ras T24 fusion gene. Molec. Carcinogenesis 1:89-95. Shimotsu, H., Takahashi, H., and.Saito, H. (1980). A.new site- specific endonuclease StuI from. Streptomyces tubercidicus. Gene 11:219-225. Shivapurkar, S., ‘Wilson, 2M.J., and. Poirer, L.A. (1984). Hypomethylation of DNA in ethionine-fed rats. Carcinogenesis 5:989-992. Siegel, J.N., Klausner, R.D., Rapp, U.R., and Samelson, L.E. (1990). T cell antigen receptor engagement stimulates c- raf phosphorylation and induces c-raf-associated kinase activity via a protein kinase C-dependent pathway. J. Biol. Chem. 265818472-18480. Sigal, I. S., D'Alonzo, J.S., Ahern, J.D., Marshall, M.S., Smith, G.M., Scolnick, E.M., and Gibbs, J.E. (1988). The ras oncogene protein as a G-protein. Advances in Second Messenger and Phosphoprotein Research 21:193-200. Sigal, I.S., Gibbs, J.E., D'Alonzo, J.S., and Scolnick, E.H. (1986). Identification of effector residues an neutralizing epitope of Ha-ras-encoded p21. Proc. Natl. Acad. Sci. USA 8384725-4729. Siglin, J.C., Weghorst, C.M., and Klaunig, J.E. (1991). Role of hepatocyte proliferation in a-hexachlorocyclohexane and phenobarbital tumor promotion in B6C3F1 mice. In Butterworth, B.E., Slaga, T.J., Farland, W., and McClain, M. (eds). Chemically Induced Cell Proliferation, Wiley- Liss, Inc., New York, NY, pp.407-416. 172 Silva, A.J. and White, R. (1988). Inheritance of allelic blueprints for methylation patterns. Cell 54:145-152. Silverman, J.A., Zurlo, J., Whtson, M.A., and Yager, J.D. (1989). Expression of c-raf-l and A-raf-l during regeneration of rat liver following surgical partial hepatectomy. Mol. Carcinogenesis 2:63-67. Singhal, R.P., Mays-Hoopes, L.L., and Eichhorn, G.L. (1987). DNA methylation in aging of mice. Mech. Ageing Dev. 418199-210. Skouv, J. Ottensen, S., Mark, G., and Autrup, H. (1989). Malignant transformation of human bladder epithelial cells by DNA transfection with the v-raf oncogene. Mol. Carcinogenesis 2:59-62. Smith, H.O. and Marley, G.M. (1980). Purification and properties of Hind II and Hind III endonucleases from Haemophilus influenzae. Methods Enzymol. 65:104-108. Sneider, T.W. (1980) . The 5'-cytosine in CCGG is methylated in two eukaryotic DNAs and MspI is sensitive to methylation at this site. Nucl. Acid. Res. 8:3829-3840. Stanton, V.P., Jr., and Cooper, G.M. (1987). Activation of human raf transforming genes by deletion of normal amino- terminal coding sequences. Mol. Cell. Biol. 7:1171-1179. Storm, S.M., Cleveland, J.L., and Rapp, U.R. (1990). Expression of raf family proto oncogenes in normal mouse tissues. Oncogene 5:345-351. Stowers, S.J., Maronpot, R.R., Reynolds, S.H., and Anderson, M.W. (1987). The role of oncogenes in chemical carcinogenesis. Environ. Health Perspect. 75:81-86. Strauss, W.M. (1987). Preparation of genomic DNA from mammalian tissue. In Ausubel, F.M., Brent, R., Kinston, R.E., Moore, D.D., Seidman, J.G., Smith, J.A., and Struhl, K. (eds). Current Protocols in Molecular Biology, John Wiley and Sons, New York, pp 2.2.1-2.2.3. Streeck, R.E. (1980) . Single-strand and double-strand cleavage at half-modified and fully modified recognition sties for the restriction nucleases Sau3A and Tan. Gene 12:267- 275. Swain, J.L., Stewart, T.A. and Leder, P. (1987). Parental legacy determines methylation and expression of an autosomal transgene: a molecular mechanism for parental imprinting. Cell 50:719-727. 173 Swisshelm, K., Disteche, D.M., Thorvaldsen, J., Nelson, A., and Salk, D. (1990). Age-related increase in methylation of ribosomal genes and inactivation of chromosome- specific rRNA gene clusters in mouse. Mutat. Res. 2378131-146. Tabin, C.J., Bradley, S.M., 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 (London) 300:143-149. Taparowsky, E., Suard, Y., Fasano, 0., Shimizu, K., Goldfarb, M., and‘Wigler, M. (1982). Activation of the T24 bladder carcinoma transforming gene is linked to a single amino acid change. Nature (London) 300:762-765. Theodorescu, D., Cornil, I., Sheenhan, C., Man, M.S., and Kerbel, R.S. (1991). Ha-ras induction of the invasive phenotype results in up-regulation of epidermal growth factor receptors and altered responsiveness to epidermal growth factor in human papillary transitional cell carcinoma cells. Cancer Res. 51:4486-4491. Theodorescu, D., Fernandez, B., Cornil, I., and Kerbel, R.S. (1990). Overexpression of normal and mutated forms of c- Ha-ras induce orthotopic bladder invasion in a human superficial transitional cell carcinoma. Proc. Natl. Acad. Sci. USA 8789047-9051. Topal, M.D. (1988) . DNA repair, oncogenes, and carcinogenesis. Carcinogenesis 9:691-696. - Towbin, H. , Staehelin, T. , and Gordon, J. (1979) . Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA 76:4350-4354. Turker, M.S., Swisshelm, K., Smith, A.C., and Martin, G.M. (1989). A partial methylation profile for a CpG site is stably maintained in mammalian tissues and cultured cell lines. J. Biol. Chem. 264811632-11636. Uehara, Y., Ono, T., Kurishita, A., Kokuryu, H., and.Okada, S. (1989). Age-dependent and tissue-specific changes of DNA methylation within and around the c-fos gene in mice. Oncogene 481023-1028. van der Ploeg, L.K.T. and Flavell, R.A. (1980). DNA methylation in the human *gamma-delta-beta-globin locus in erythroid and nonerythroid tissues. Cell 19:947-958. 174 Velu, T.J. (1990). Structure, function and transforming potential of the epidermal growth factor. Mol. Cell. Endocrinol. 70:205-216. Vorce, R.L. and Goodman, J.I. (1987). Investigation of parameters associated with activity of the Kirsten-ras, Harvey-ras, and. myc oncogenes. in. normal rat liver. Toxicol. Appl. Pharmacol. 90:86-95. Vorce, R.L. and Goodman, J.I. (1989a). Altered methylation of ras oncogenes in benzidine-induced B6C3Fl mouse liver tumors. Toxicol. Appl. Pharmacol. 100:398-410. Vorce, R.L. and Goodman, J.I. (1989b). Hypomethylation of ras oncogenes in chemically induced and spontaneous B6C3F1 mouse liver tumors. J. Molec. Toxicol. 2:99-116. Waalwijk, C. and Flavell, R.A. (1978). MspI, an isoschizomer of HpaII which cleaves both unmethylated and methylated HpaII sites. Nucl. Acid. Res. :3231-3236. Ward, J.M., Lynch, P., and Riggs, C., (1988). Rapid development of hepatocellular neoplasms in aging male C3H/HeNCr mice given phenobarbital. Cancer Lett. 39:9-18. Wasylyk, C., Wasylyk, B., Heidecker, G., Huleheil, M., and Rapp, U.R. (1989). Expression of raf oncogenes activates the PEAl transcription factor motif. Mol. Cell. Biol. 982247-2250. Watt, F., and. Molloy, P.L. (1988). Cytosine 'methylation prevents binding to DNA of a HeLa cell transcription factor required for optimal expression of the adenovirus major late promoter. Gene. Dev. 2:1136-1143. Weghorst, C.M. and Klaunig, J.E. (1989). Phenobarbital promotion in diethylnitrosamine-initiated infant B6C3F1 mice: influence of gender. Carcinogenesis 10:609-612. Weih, F., Nitsch, D., Reik, W., Schutz, G., and Becker, P.B. (1991). Analysis of CpG methylation and genomic footprinting at the tyrosine aminotransferase gene: DNA methylation alone is not sufficient to prevent protein binding in vivo. EMBO J. 10:2559-2567. Weinberg, ‘R.A. (1989). Oncogenes, antioncogenes, and. the molecular bases of multistep carcinogenesis. Cancer'Res. 49:3713-3721. Weinstein, I.B. (1988) . The origins of human cancer: molecular mechanisms of carcinogenesis and their implications for cancer prevention and treatment--twenty-seventh G~H.A. Clowes memorial award lecture. Cancer Res. 48:4135-4143. 175 Wigler, M., Leny, D., and Perucho, M. (1981). The somatic replication of DNA methylation. Cell 24:33-40. Wilson, V. and Jones, P. (1983a). DNA methylation decreases during aging but not in immortal cells. Science 220:1055- 1057. Wilson, V.L. and Jones, P.A. (1983b). Inhibition of DNA methylation by chemical carcinogens in vitro. Cell 328239-246. Wilson, V.L., Smith, R.A., Longoria, J., Liotta, M.A., Harper, C.M., and Harris, C.C. (1987a) . Chemical carcinogen- induced decreases in genomic 5-methyldeoxycytidine content of normal human bronchial epithelial cells. Proc. Natl. Acad. Sci. USA 8483298-3301. Wilson, V.L., Smith, R.A., Ma, S., and Cutler, R.G. (1987b). Genomic 5-methylcytosine decreases with age. J. Biol. Chem. 26289948-9951. Wiseman, R.W., Stowers, S.J., Miller, E.C., Anderson, M.W., and Miller, J.A. (1986). Activating mutations of the c- Ha-ras protooncogene in chemically induced hepatomas of the :male B6C3Fl :mouse. Proc. Natl. Acad. Sci. USA 8385825-5829. Worland, P.J., Hampton, L.L., Thorgeirsson, S.S., and Huggett, A.C. (1990). Development of an in vitro model of tumor progression using v-raf and v-raf/v-myc transformed rat liver epithelial cells: correlation of tumorigenicity with the downregulation of specific proteins. Mol. Carcinogenesis 3:20-29. Yusof, Y.A.M. and Edwards, A.M. (1990). Stimulation of DNA synthesis in primary rat hepatocyte cultures by liver tumor promoters: interactions with other growth factors. Carcinogenesis 11:761-770. 31293010550253