the origins of human cancer: molecular mechanisms of

1988;48:4135-4143. Cancer Res I. Bernard Weinstein Award Lecture

Twenty-seventh G. H. A. Clowes Memorial−−and TreatmentCarcinogenesis and Their Implications for Cancer Prevention The Origins of Human Cancer: Molecular Mechanisms of

Updated version

http://cancerres.aacrjournals.org/content/48/15/4135

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

[email protected] at

To request permission to re-use all or part of this article, contact the AACR Publications

Research. on December 12, 2013. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from




http://cancerres.aacrjournals.org/cgi/alerts

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]




http://cancerres.aacrjournals.org/




[CANCER RESEARCH 48,4135-4)43, August l, 1988J

Special Lecture

The Origins of Human Cancer: Molecular Mechanisms of Carcinogenesis and TheirImplications for Cancer Prevention and Treatmentâ€”Twenty-seventh G. H. A.Clowes Memorial Award Lecture1

I. Bernard WeinsteinComprehensive Cancer Center, Institute of Cancer Research, Department of Medicine, and School of Public Health, Columbia University, New York, New York 10032

Abstract

EpidemiolÃ³gica!studies provide evidence that environmental factors(external agents such as chemicals, radiation, and viruses) play a majorrole in the causation of the majority of human tumors. This is a highlyoptimistic message, since it implies that cancer is largely a preventabledisease. To meet this challenge we must, however, understand the mechanisms of cancer causation at the cellular and molecular levels and, in aparallel effort, develop new laboratory methods that can be used toidentify specific causative agents in humans. The approach must becomprehensive since it is likely that human cancers are due to complexinteractions between multiple factors, including the combined actions ofchemical and viral agents. This paper reviews recent studies from ourlaboratory and studies by other investigators related to these themes.

A major principle in studies on mechanisms of carcinogenesis is thatthe process proceeds through multiple discernible stages, including initiation, promotion, and progression. It is likely that the transition betweenthese stages is driven by different environmental and endogenous factorsand involves different biochemical mechanisms and genetic elements.Several types of chemicals initiate the carcinogenic process by yieldinghighly reactive species that bind covalenti) to cellular DNA. Our grouphas elucidated the details of this process with two groups of compounds,aromatic amines and polycyclic aromatic hydrocarbons, emphasizing howthese agents distort the conformation of DNA and its functions duringDNA replication and transcription. The implications of these findingswith respect to oncogene activation, DNA amplification, gene transposition, and chromosome translocations are discussed. Our studies on tumorpromotion have concentrated on the mechanisms of action of the potenttumor promoter 12-O-tetradecanoylphorbol-l 3-acetate. Studies from several laboratories indicate that this agent and related compounds producetheir effects by activating a specific cellular enzyme, protein kinase C(PKC). This produces a cascade of events which include alterations inthe function of membrane-associated ion channels and receptors, alterations in gene expression and, ultimately, changes in cellular differentiationand proliferation. Recent studies on the isolation and stable overexpres-sion of a cloned DNA sequence that encodes PKC are described. Theresults obtained provide direct evidence that PKC plays a critical role ingrowth control. The possible role of PKC, and other mediators of signaltransduction pathways, in the origin of certain human cancers is alsodiscussed.

Finally, this paper discusses how insights into molecular mechanismsof carcinogenesis provide new approaches to the detection of specificcauses of human cancer, using an approach termed "molecular cancerepidemiology," and also new strategies for cancer prevention and treat

ment.

Introduction

MaimÃ³nides, the famous 12th century Hebrew scholar andphysician, said "honor your teachers because they have broughtyou into the world of the future." My career was molded by

Received 4/11/88; accepted 5/3/88.1Financial support of this research was provided by the National Cancer

Institute, the American Cancer Society, the Alma Toorock Memorial for CancerResearch, and the National Foundation for Cancer Research. Presented at theSeventy-eighth Annual Meeting of the American Association for Cancer Research, May 21, 1987, in Atlanta, GA.

several wonderful teachers, beginning with my parents who, asimmigrants from ( '/.arisi Russia, did not have a formal educa

tion but inspired a dedication to learning in their children.Other inspirational teachers included: William S. Middleton,who stimulated my interest in academic medicine when I was amedical student at the University of Wisconsin; Daniel Laszlo,whose compassion for the cancer patient and clinical insightsstimulated my interest in oncology when I was a MedicalResident at Montefiore Hospital; Nathaniel Berlin, who guidedme in clinical research when I was a Clinical Associate at theNational Cancer Institute; Boris Magasanik, who taught mebiochemistry and how to think at the cellular and molecularlevels when I was a research fellow at MIT: Alfred Gellhorn,who fostered my career in cancer research and continues toprovide a role model as a physician/scientist and humanitarian;and Jacob Furth, who shared with me his insights into tumorbiology and served as a role model for commitment to research,students, and colleagues. I also want to acknowledge the inspiration of several leaders in the field of chemical carcinogenesis,especially Charles Heidelberger, Elizabeth and James Miller,Ross Boutwell, and Harold Rusch. It is curious that all five ofthese individuals are from Wisconsin, and so am I. There mustbe something in the Wisconsin environment that encouragescancer research.

I am also grateful to numerous colleagues and students withwhom I have worked over the past 25 years. In particular, I amindebted to Dezider Grunberger with whom I have had a veryproductive collaboration and warm friendship for many years.I am also grateful to Columbia University and the Columbia-Presbyterian Medical Center, which has provided an intellectually rich environment and marvelous resources for the pursuitof my research. Finally, I am indebted to the National CancerInstitute, the American Cancer Society, and other fundingsources for providing the funding required for our research.

At the beginning of this century the major cause of death inthe United States was infectious diseases. Due to a combinationof basic research and prevention and treatment approaches,death from these diseases has been markedly reduced. Thus, aswe approach the latter decade of the 20th century, cancer hasbecome a leading cause of death in this country and in manyother countries throughout the world (1). The challenge thatcurrently faces cancer researchers is the following: how can wemount effective research and clinical approaches so that withina few decades we can achieve a decline in cancer mortalitysimilar to that achieved during the earlier part of this centuryin the field of infectious diseases? Recently, there has been agood deal of rhetoric about whether or not we are "winning thewar on cancer." This rhetoric, like war itself, can be destructive.

Progress in the field of cancer will not come from a singleshort-sighted approach but rather from a long-term commitment to both fundamental laboratory and clinical research, anda pluralistic approach that includes research on cancer causa-

4135




MOLECULAR ORIGINS OF HUMAN CANCER

tion, prevention, and treatment, as well as skilled and compassionate care of patients with cancer. In fact, what makes thecurrent period of cancer research so exciting is that advancesin molecular genetics, cellular biology, carcinogenesis, virology,immunology, and therapeutics are providing unifying conceptsand methods that bring together investigators from diversefields of cancer research. The following quotation from H. L.Mencken is relevant to the need for a pluralistic approach tothe cancer problem: "For any complex question there is almostcertainly a simple answer that is almost always wrong."

In this paper I will provide an overview of recent advances inour understanding of the molecular and cellular mechanismsby which various chemical agents influence the multistage carcinogenic process, thus resulting in the evolution of malignanttumor cells. I will also provide examples of how advances inthese areas of research are likely to provide new and effectivestrategies for both cancer prevention and treatment. Since thispaper provides a broad overview, the reader is referred to morespecific review articles for detailed references.

Multistage Carcinogenesis: Initiation

The Multistage Process Predicts Multiple Mechanisms andGenes. The development of a fully malignant tumor involvescomplex interactions between several factors, both environmental (i.e., exogenous) and endogenous (i.e., genetic, hormonal,etc.). In addition, carcinogenesis often proceeds through multiple discernible stages (initiation, promotion, progression)(Fig. 1) and the overall process can occupy a major fraction ofthe life span of the individual (2-6). The transitions betweensuccessive stages can be enhanced or inhibited by different typesof agents. Thus, it appears that the individual stages may involvequalitatively different mechanisms at the cellular and geneticlevels. These aspects predict that the establishment and maintenance of a malignant tumor involve multiple cellular genesand multiple types of changes in genomic structure and function.

Action of Initiating Agents. Agents that initiate the carcinogenic process often do so by damaging cellular DNA. Indeedthis has become an axiom in the field of chemical carcinogenesis(5-7). This principle is well illustrated with the chemicalbenzo(a)pyrene, an ubiquitous carcinogen that is representativeof a large number of polycyclic aromatic hydrocarbon carcinogens (5,7,8). The compound is not active as such but undergoesmetabolism by the microsomal cytochrome P-450 monooxy-

INITIATION

V V V V

PROMOTIONSTAGES 1,2....

PROGRESSION

INCREASING MALIGNANCY -

Fig. 1. Schematic representation of multistage carcinogenesis based on studieson mouse skin. Initiation requires only a single exposure to a carcinogen andappears to involve DNA damage. Promotion involves multiple exposure to agentsthat do not damage DNA directly. Promotion can also be subdivided into twophases. Progression involves the conversion of benign to malignant tumors andcan be considered an open-ended process since tumors often continue to increasein their degree of malignancy and heterogeneity. Reprinted from Ref. 6, withpermission.

genÃ¤sesystem to yield the highly reactive metabolite BPDE.2

Several years ago, Dr. Alan Jeffrey in our research group atColumbia University provided evidence that a specific isomerof BPDE preferentially reacts with the 2-amino group of gua-nine residues present in nucleic acids to yield the structureshown in Fig. 2. We also obtained evidence that, when boundto double-stranded DNA, the bulky carcinogen residue lies inthe minor groove of the DNA helix (Fig. 3). This is in contrastto the situation found with the aromatic amine carcinogen N-2-acetylaminofluorene which, following metabolic activation,

Fig. 2. Chemical structure and stereochemistry of the covalent adduct formedby the reaction of BPDE with guanine residues in nucleic acids. Position 10 ofBPDE is linked to the 2-amino group of guanine. Also displayed (large spheres)are the hydroyl residues on positions 7, 8, and 9 of BPDE. For additional detailssee Refs. 5 and 8.

Fig. 3. (A) Computer graphics-generated display of BPDE covalently boundto B DNA with the BPDE residue in the minor groove of the DNA helix. In thismodel the normal conformation of the guanine base to which the BPDE isattached is retained. The 7- and 8-hydroxyl groups and the 9- and 10-hydrogensof BPDE are quasiequatorial. The angle between the plane of the pyrene and theaxis of the DNA is 28 degrees. This results in some slight steric hindrance, basedon ver der Walls radii, which is presumably relieved by slight distortions of thehelix, (li) iY-2-Acetylaminofluorene (AAF) covalently bound to the C-8 positionof guanine in B-DNA; the "base displacement model." The guanine to which theJV-2-acetylaminofluorene is attached has been rotated out of the helix, and the N-2-acetylaminofluorene moiety is inserted into the helix and stacked with the basesabove and below. .-(C. acmi group of JV-2-acetylaminofluorene. The cytosine (C)residue on the opposite strand would overlap with the .V-2-acct>laminofluoreneresidue; therefore it is not displayed. This cytosine probably rotates out of thehelix to accommodate the .V-2-acctylaniinofluorene: its exact conformation is notknown, although there is evidence that this region of DNA is "single-stranded."

Reprinted from Ref. S, with permission.

2The abbreviations used are: BPDE, benzo(a)pyrene-7,8-diol-9,10-epoxide;cDNA, complementary DNA; TPA, 12-O-tetradecanoylphorbol-13-acetate; PKC,protein kinase C; DAG, diacylglycerol; PDGF, platelet-derived growth factor.

4136





becomes bound to the C-8 position of guanine residues andresults in a major distortion of the DNA helix which we termedbase-displacement (Fig. 3) (5, 9). The precise Â«informationalchanges in the structure of DNA produced by the covalentbinding of various aromatic amine carcinogens have been elucidated by my colleague Dr. Dezider Grunberger (9). An important principle that has emerged is that the modification ofspecific sequences in DNA by certain carcinogens can inducethe Z DNA conformation. The modification of DNA by lessbulky carcinogens, i.e., methylating or ethylating agents, causesfunctional disturbances by directly interfering with base pairing,rather than through major changes in DNA conformation.Thus, we are beginning to understand, at the molecular level,the multiple ways in which initiating carcinogens attack DNAand the types of conformational and functional changes in DNAthat result from these modifications (for a detailed review ofthis subject see Refs. 5 and 9).

Critical Genetic Targets: Oncogenes, Tumor SuppressorGenes, Transcriptional Regulatory Sequences. As advances inthe above area of research were proceeding, studies with RNAtumor viruses revealed the existence of a set of tumor-causinggenes, the oncogenes (for review see Refs. 10-13). This was

followed by the finding that a set of related genes, termed"protooncogenes," normally exists in the genomes of higher

organisms and the accumulating evidence that the latter genescode for components of signal transduction pathways that playa role in controlling normal growth and differentiation. Thesefindings led to the concept that initiating carcinogens might actby mutating protooncogenes to produce "activated" oncogenes

and thus lead to abnormalities in growth control and differentiation (Fig. 4). Indeed, there is accumulating evidence thatseveral types of tumors induced in rodents by chemical carcinogens and certain tumors in humans are associated with basesubstitutions at specific sites in the c-ras oncogenes (10-13).As this data base expands, however, it is apparent that there isconsiderable variation between tumors in the same or differenttissues in terms of whether or not they carry a mutated ras geneand the type of mutation seen, even when the tumors wereinduced by the same carcinogen. Furthermore, the types of basesubstitutions seen do not always to correlate with the known

Normal Growth & Differentiation

Carcinogens

VirusRecombination

Aberrant ExpressionMutationTranspositionTranslocation

TCancer

Fig. 4. Schematic diagram indicating three classes of DNA sequences thatmight be critical targets in the action of chemical carcinogens or radiation:oncogenes (or protooncogenes); transcriptional enhancer and promoter DNAsequences (Keg. Seq.); and tumor suppressor genes (suppressor). Specific RNAtumor viruses contain the first two classes of these DNA sequences. Thus tumorformation by certain RNA tumor viruses shares mechanisms common with thatinduced by chemical carcinogens or radiation.

chemistry of carcinogen-DNA modifications, and in some casesit appears that the ras mutation occurred after the initiatingevent. Therefore, I am not convinced that these mutationsrepresent the major mechanism of initiation. Elsewhere (13), Ihave predicted that, given the multiplicity of signal transductionpathways and the overall complexity of growth control anddifferentiation, there probably exist more than 300 protooncogenes in the mammalian genome. As probes for these genesbecome available, I suspect that we will discover numerousother oncogene mutations in carcinogen-induced tumors.

In addition to protooncogenes, I believe that two other classesof DNA sequences that are normally present in the mammaliangenome, tumor suppressor genes and transcriptional regulatorysequences, may also be critical targets during chemical earcinogenesis (Fig. 4) (13). The probable roles of these two classesof DNA elements in carcinogenesis are discussed in greaterdetail below.

Complex Genomic Changes and Inducible Responses to DNADamage. Although the above-mentioned base substitution mutations in c-ras oncogenes are of great interest, they represent

only a limited glimpse of the complex types of DNA changesseen in carcinogen-induced tumors. Tumors can also displaydeletions, chromosome translocations, amplifications, andtranspositions of protooncogenes and other genes (10-13). Inaddition, in certain experimental systems the early steps in thecarcinogenic process occur with a much higher frequency thanwould be expected for random point mutations (14).

For these reasons, our laboratory has been interested in thepossibility that DNA damage in mammalian cells not onlycauses targeted point mutations but also induces the accumulation of factors which produce more complex genomic changes.We have been examining this possibility in a model system inwhich a rat fibroblast cell line (H3) carries an integrated copyof polyoma virus DNA which normally replicates synchronously with the host DNA (15-19). This system is of interestsince if these cells are exposed to either a chemical carcinogenor UV, both of which damage DNA, then the polyoma DNAreplicates asynchronously, producing thousands of copies ofviral DNA. We found that if we treated normal rat cells (thatlack polyoma DNA) with BPDE or UV irradiation and nextfused these cells with untreated H3 cells, then this also inducedpolyoma DNA replication. The latter studies provided evidencethat this induction is due to the formation of a fra/zs-actingfactor in the carcinogen-treated cells. Direct evidence for thisputative factor has been obtained using extracts of UV-irradi-ated cells ( 19). Since this factor can be induced in normal cells,we believe it may play a role in the amplification of specificcellular genes or cause other types of structural changes incellular DNA.

Studies are under way to identify more precisely this factorand its mechanism of action. Other investigators have demonstrated that DNA damaging agents can also induce the replication of SV40 viral DNA and that this effect may also bemediated by a fra/is-acting factor (20-23). We believe that theresults obtained in both the polyoma and SV40 virus DNAsystems may be of interest for several reasons, (a) They mayserve as useful models for understanding mechanisms underlying the amplification of cellular genes during the carcinogenicprocess, (b) They might explain the well known phenomenonthat chemical carcinogens can act synergistically with certainviruses to enhance cell transformation (5). Since certain humantumors may be caused by the combined effects of viruses andenvironmental chemicals, this phenomenon could be of directrelevance to the origin of certain human tumors (5). (c) They

4137





may provide assays for detecting potential environmental carcinogens, thus complementing the current battery of short-term

tests for mutagens.Table 1 summarizes evidence that DNA damage in mam

malian cells not only enhances the replication of certain viralDNAs but also induces other complex responses. The diversityof these responses indicates that the concept of "genomicstress," first elucidated in lower organisms, applies to mam

malian cells (5, 15, 22, 23). Genomic stress responses may playa role not only in carcinogenesis but also in the complex effectsof radiation therapy and chemotherapy on tumors cells. Thus,a better understanding of genomic stress responses may suggestways to enhance the therapeutic effects of radiation and chemotherapy.

Disturbances in Transcriptional Control and Retrotransposons.As illustrated in Fig. 4 it seems likely that carcinogenesisinvolves not only the activation of cellular protooncogenes butalso disturbances in the function of DNA sequences that play arole in controlling gene transcription, including civ-acting promoter and enhancer sequences. With respect to the latter typesof DNA sequences, we have been intrigued with the fact thatthe genomes of higher organisms, including humans, containthousands of copies of retrovirus-like elements (6, 16, 24).These elements are bounded by long terminal repeat sequencesthat contain specific promoter and enhancer sequences thatcontrol gene transcription. Moreover, in lower organisms (notably Drosophila and yeast) similar endogenous retrovirus-likeelements function as mobile elements which can insert into newsites in the cellular genome, thus deleting or activating specificgenes (24). There now exist several examples in which anendogenous mammalian retrovirus-like element, the intracis-ternal A particle sequence, has undergone transposition inmurine tumors (25-28), leading in some cases to activation ofthe c-mos oncogene or causing the constitutive expression ofthe interleukin 3 gene. In yeast and Drosophila the transpositionof similar retrovirus-like elements occurs via the reverse transcription of RN A transcripts of these elements, hence the term"retrotransposition" (24). We have found that murine and rat

tumors that were induced by chemical carcinogens display ahigh level of constitutive expression of specific endogenousretrovirus-like sequences (29-33). The increased expression of

these sequences is seen in both benign and malignant tumorsand is often more striking than alterations in the expression ofthe c-ras or c-myc oncogenes. This phenomenon provides evidence for a fundamental disturbance in gene transcription intumor cells. Studies are in progress to determine whether ornot it predisposes to gene transposition during the process ofcarcinogenesis.

Tumor Suppressor Genes. There is increasing recognition ofthe fact that in addition to genes that can code for componentsof pathways that stimulate cell growth (protooncogenes), thereexists a reciprocal set of genes the products of which inhibitcell growth and/or induce cells to undergo terminal differentiation (for review see Refs. 13, 34, and 35). These so-called"tumor suppressor genes" or "antioncogenes" could inhibit the

evolution and growth of tumor cells. Mutations that delete theirfunction would be expected to enhance tumor formation. Indeed

Table 1 Inducible responses to DNA damage in mammalian cells"

1. Virus reactivation.2. Induction of plasminogen activator, ornithine decarboxylase, metallothionein

I, c-myc, c-fos, CIN I, etc.3. Induction of DNA amplification and virus replication.4. Induction of endogenous retroviruses (? transposition)

Â°For details see Refs. 15-23.

the deletion of such genes appears to be involved in certainhuman tumors including retinoblastoma, Wilm's tumor, lung

cancer, colon cancer, and other cancers (34-37). In view of theability of chemical carcinogens to induce not only point mutations but also gene deletions and various types of chromosomalanomalies, it seems likely that tumor suppressor genes alsorepresent critical targets in the action of chemical carcinogens(Fig. 4). Hopefully, the cloning of tumor suppressor genes willprovide probes for examining this possibility. One approachwhich our laboratory has explored is to isolate from a cDNAlibrary DNA sequences the expression of which is turned offrather than on in cells undergoing a response to a mitogenicstimulus (38). Possible biochemical functions of these suppressor genes are discussed elsewhere (13).

Tumor Promotion

Tumor Promoters Produce Epigenetic Effects. I now want toturn to studies on tumor promotion, emphasizing recent advances in our understanding of the mechanism of action of thepotent tumor promoter TPA. Tumor promoters can be definedas compounds which have very weak or no carcinogenic activitywhen tested alone but markedly enhance tumor yield whenapplied repeatedly following a low or suboptimal dose of acarcinogen (initiator) (3,4, 39,40). Early studies indicated that,in contrast to initiating agents, the phorbol ester tumor promoters do not bind to DNA but instead act by binding tomembrane-associated receptors and thus produce their initialeffects at the epigenetic level. Studies from a number of laboratories, including our own, indicated that TPA induces anumber of phenotypic effects in cell culture systems whichcould be grouped into three categories: mimicry of the transformed phenotype; modulation (either inhibition or induction)of differentiation; and membrane effects (39,40). The discoveryin 1982 by Castagna et al. (41) that TPA activates the enzymePKC and subsequent studies indicating that PKC is the majorcellular receptor for TPA have merged research on tumorpromotion with that on growth factors, signal transduction,and the action of specific oncogenes (13, 42, 43).

Cloning of DNA Sequences That Encode PKC. Because of thecentral role of PKC in signal transduction, growth control, andtumor promotion our laboratory has done extensive studies onthe function of PKC and has also cloned DNA sequences thatencode this enzyme. The enzyme has two domains, a catalyticdomain, containing an ATP-binding site and the region towhich protein substrates bind, and a regulatory domain whichis controlled by allosteric cofactors, i.e., lipid, Ca2+, and DAG,

or lipid and TPA. We hypothesize that the usual function ofthe regulatory domain is to inactivate the enzyme by "closing"

the catalytic site and that the binding of cofactors to theregulatory domain induces a Â«informational change that opensthe catalytic site and thus activates enzyme function (Fig. 5)(13). Consistent with this scheme is evidence that limited pro-teolysis of the enzyme yields a fragment with a molecular weightof about 66,000 which has catalytic activity in the absence oflipid and other allosteric cofactors (42). In addition, there existinhibitors of PKC that appear to act preferentially on theregulatory domain or the catalytic domain of the enzyme (44).A recent paper by House and Kemp (45) is also consistent withthis model. Indeed, the latter authors provide evidence that theregulatory domain contains a peptide sequence that acts as apseudosubstrate inhibitor of the catalytic domain when theenzyme is in the inactive configuration.

Our laboratory (46, 47) and other laboratories (48-52) haverecently reported the isolation of cDNA clones that encode

4138





Fig. 5. Hypothetical model of PKC emphasizing a catalytic domain thatcontains the ATP and peptide substrate-binding sites and a regulatory domainthat binds phosphatidylserine (PS) (or other lipids) and C'a'" or phosphatidylser-ine and TPA. We postulate that binding of phosphatidylserine and Or', or

phosphatidylserine plus TPA, to the regulatory domain induces a conformationalchange in the enzyme that "opens" the catalytic domain, thus activating the

phosphorylation of a protein substrate, e.g., histone HI. Reprinted from Ref. 13,with permission.

1-60 MADPAAGPPPSEGEESTVRFARKGPLRQKNVHEVKNHKFTARFFKQPTFCSHCTDFIWGF

I--I .......

2................... ......... Rl ......... .....................

61-120 GKQGFQCQVCCFWHKRCHEFVTFSCPGADKGPASDDPRSKHKFKIHTYSSPTFCDHCGS

...... I--I ....... I....... I............................ I--I--

13 2 7 7 28 2

121-180 LLYGLIHQGMKCDTCMMNVHKRCVMNVPSLCGTDHTERRGRIYIQAHIDREVLIVWRDA........... I--I ....... I ....... I

13 27 7

181-240 KNLVPMDPNGLSDPYVKLKLIPDPKSESKQKTKTIKCSLNPEWNETFRFQLKESDKDRRLI--S2-I

241-300 SVEIWDWDLTSRNDFMCSLSFGISELQKAGVDCWFKLLSQEEGEYFNVPVPPEESEGNEE

301-360 LRQKFERAKIGQGTKAPEEKTANTISKFDNNGNRDRHKLTDFNFLMVLGKGSFGKVMLSE

361-420 RKGTDELYAVKILKKDWIQDDDVECTMVEKRVLALPGKPPFLTQLHSCFQTMDRLYFVH

421-480 EYVNGGDLMYH1QQVGRFKEPHAVFYAAEIAIGLFFIXÃŒSKGIIYRDLKLDNVMLDSEGHI

481- 540 K1ADFGMCKENIWDGVTTKTFCGTPDYIAPEIIAYQPYCKSVDWWAFGVLLYEMLAGQAP

*** *** *** p2

541-600 FEGEDEDELFQSIMEHNVAYPKSMSKEAVAICKGLMTKHPGKRLGCGPEGERDIKEHAFF

-f+ +

PKC. The results indicate that PKC belongs to a multigenefamily, with at least 6 separate genes that encode distinctisozymes of PKC. The deduced amino acid sequences of theisolated cDNA clones indicate remarkable homologies betweenthe individual forms of PKC and also appreciable homologieswith other protein kinases, especially the cyclic AMP-depen-dent protein kinase (Fig. 6). The carboxyl-terminal portion comprises the catalytic domain, whereas the amino-terminal portionis the regulatory domain. The latter region contains two repeatsof a cysteine-rich consensus sequence found in certain metal-binding proteins and in several DNA-binding proteins. Wethink that this may induce a specific conformation that isrequired for the interaction of this region of the enzyme withphospholipid and DAG, or TPA. TPA, teleocidin, and aplysia-toxin, three chemically diverse skin tumor promoters that arealso potent activators of PKC, share certain common stereo-chemical features (6, 53). We suspect that this allows them tobind to a specific conformation in the cysteine-rich region ofPKC. The development of variant PKCs by site-directed mu-tagenesis of cloned cDNAs may clarify these aspects and thuslead to the rational design of PKC inhibitors.

The existence of multiple genes for PKC, the preservation ofthese multiple forms during evolution, and the accumulatingevidence that they are differentially expressed in different tissues (46-52) suggest that the individual forms may have subtlefunctional differences, but this requires further study.

Overexpression of PKC Alters Growth Control. To betterdefine the function of a specific form of PKC we have insertedthe cDNA sequence for the ÃŸ\form of PKC into a retrovirus-derived vector developed in our laboratory (pMV7) (54), encap-sidated the corresponding RNA into defective murine leukemiavirus particles, and then infected the Rat 6 fibroblast cell linewith these viral particles. This yielded several cell lines thatstably overexpress 20 to 53 times the normal level of PKCenzyme activity (47). These cell lines also have an increase inhigh affinity phorbol ester receptors when compared to controlcells. Phosphorylation studies indicate that they display a

601-660 RYIDVEKLERKEIQPPYKPKARDKRDTSNFDKEFTRQPVELTPTDKLFIMNLDQNEFAGF

+ A*****

661-671 SYTNPEFVINV

Fig. 6. Deduced amino acid sequence of I'M '.,,. The sequence begins with the

first in-frame methionine codon and encodes a 671-amino acid protein with apredicted molecular weight of 76,800. The arrow (Rl) above the amino acidsequence designates a repeating pattern of conserved cysteine residues. Thespacing between conserved cysteines is indicated. *. conserved amino acid residuesfound in other serine/threonine protein kinases. +, consensus ATP-binding sitesequences. S2 and /'- denote peptides which were determined by automatedEdman degradation. For additional details see Refs. 46 and 47.

marked increase in a M, 76,000 32P-labeled protein, reflecting

autophosphorylation of the overproduced PKC, as well as anincrease in other phosphorylated proteins. The cell lines thatoverproduce PKC exhibit a dramatic change in morphologywhen exposed to TPA. Unlike control cells, which becomerefractory to the effects of TPA due to down-regulation of PKC,the overproducer cell lines continue to respond to repeatedTPA treatments. In monolayer culture, these cell lines have ashorter exponential doubling time and grow to a higher saturation density and, when maintained at postconfluence, developsmall, dense foci. In contrast to the control cells, which displaycomplete anchorage dependence in the absence or presence ofTPA, the cell lines that overproduce PKC form small coloniesin soft agar in the absence of TPA and larger and more frequent colonies in the presence of TPA. Thus, overproductionof a single form of PKC is sufficient to confer anchorage-independent growth and other growth abnormalities in ratfibroblasts (47).

Taken together, the above results provide direct evidence thatPKC plays a critical role in normal cellular growth control andthat it mediates several of the cellular effects of the phorbolester tumor promoters. These results also provide evidence thatdisturbances in the activation or down-regulation of PKC maybe critical events in tumor promotion and multistage carcino-genesis. The exaggerated phosphorylation of specific cellularproteins in the cell lines that overproduce PKC may provide a

4139





useful tool for identifying these target proteins and their rolein signal transduction. In recent studies we have found thatthese cells are also much more susceptible to transformation byan activated c-H-ros oncogene than control cells.3 Studies are

in progress to determine whether they are also more susceptibleto malignant transformation by chemical carcinogens.

Isolation of Phorbin, a Gene Induced by PKC Activation. TPAinduces the expression of a number of cellular genes (38-40).

We are also interested in the role of PKC activation in mediating these effects and the underlying molecular mechanisms.To identify specific genes that are regulated by TPA, we prepared a cDNA library using polyadenylate-containing RNAobtained from quiescent C3H 10T'/2 murine fibroblasts 4 h

after treatment with TPA and then screened this library forinduced sequences by differential hybridization. We have isolated and characterized a cDNA clone (TPA-S1) the corresponding mRNA of which is induced up to 20-fold in responseto the treatment of cells with TPA, PDGF, epidermal growthfactor, serum, or diacylglycerol (38). This effect is apparentlyat the level of transcription and does not require de novo proteinsynthesis. We have designated the gene corresponding to TPAS-l "phorbin" (phorbol ester induced). The role of PKC in the

induction of phorbin is supported by the following lines ofevidence: (a) agents that activate PKC, such as TPA, mezerein,PDGF, serum, and l-oleoyl-2-acetylglycerol, also increasephorbin mRNA levels; (b) protein kinase inhibitors with differential effects on PKC activity demonstrate the same relativeinhibitory effects on the induction of phorbin by TPA; (c) down-regulation of PKC activity, by treatment of lOT'/i cells with

TPA for 24 h, results in a loss of responsiveness to phorbininduction by subsequent TPA treatment; and (d) the above-described cell lines that overexpress PKC are extremely sensitive to TPA induction of phorbin RNA and also yield anexaggerated and prolonged response (44).4

Complete sequence analysis of the phorbin cDNA (730 basepairs) predicts a cysteine-rich, secreted protein with a molecularweight of 22,600. The sequence of phorbin exhibits homologywith two previously isolated cDNA sequences, designated"EPA" (erythroid-potentiating activity) and "TIMP" (tissue

inhibitor of metalloproteinase) (38). The significance of thesehomologies and the possible role of phorbin in cell proliferationand tumor promotion are under investigation.

Relevance of the Above Studies to Tumor Promotion in VariousTissues. Although the studies described above are concernedwith the molecular mechanisms of action of TPA, a potenttumor promoter on mouse skin, we think that they have relevance to other types of tumor promoters and to tumor promotion in other tissues and species, including the process ofmultistage carcinogenesis in humans. For example, we havefound that certain bile acids that are implicated in colon carcinogenesis can activate PKC, induce translocation of PKC tothe membrane fraction of cells, and induce phorbin expression(55, 56). Although it is unlikely that all classes of tumorpromoters function by directly activating PKC, it is possiblethat they activate PKC indirectly by enhancing DAG production or by distorting related pathways of signal transduction(Fig. 1). This merits further study, particularly with certainhalogenated organic compounds that have tumor-promotingactivity in rodent liver. In recent studies on patterns of geneexpression during rat liver regeneration and carcinogenesis (33,57), we have discovered that in both systems there is an earlydecrease in the abundance of epidermal growth factor-receptor

3W. Hsiao and 1. B. Weinstein, unpublished data.' Unpublished studies.

mRNA. This may be an important clue to an underlying changein signal transduction associated with hepatocyte proliferation.

Some Major Unresolved Aspects of Multistage Carcinogenesis

Despite the astounding progress that has been made in ourunderstanding of the mechanisms by which initiating carcinogens modify the structure and conformation of DNA, the recentexplosion of information on tumor promotion, PKC and signaltransduction pathways, and the revelation of cellular oncogenesand tumor suppressor genes, we still lack a comprehensiveunderstanding of the cellular and molecular events involved inmultistage carcinogenesis. As discussed earlier in this paper, itis not known with certainty whether carcinogens that act asinitiators produce a heritable event in somatic cells by causingpoint mutations that activate cellular oncogenes, whether theycause deletions in critical growth suppressor genes, whetherthey act on elements that specifically regulate transcription,whether they act by inducing complex changes in the genome(e.g., DNA amplification or transposition), or whether they actby combinations of such mechanisms. The possible role ofalterations in DNA methylation in initiation or in later stagesof carcinogenesis also requires clarification (for review see Ref.58). A further consideration is that although most of the emphasis in studies on chemical carcinogens has focused on nuclear DNA and nuclear genes, we and others have demonstratedthat the mitochondria! DNA of cells is also a major target forcovalent binding by certain carcinogens (59). Whether or notcarcinogen-induced alterations in mitochondrial DNA play acritical role in carcinogenesis remains to be determined.

Although the fields of chemical and viral carcinogenesis haveevolved as two rather separate disciplines, human populationsare often exposed to both types of agents. I think it is likely,therefore, that certain forms of human cancer are due to syn-ergistic interactions between specific viruses, which on theirown have little or no tumorigenic potential, and chemicals thathave initiating (genotoxic) or tumor-promoting activity (5, 6).There are a number of such examples in experimental models(5, 6, 60). I would encourage, therefore, further studies on themechanisms that underly chemical-viral synergy and their possible roles in the causation of specific forms of human cancer.Another unresolved question is how certain initiating agentscan act as "complete" carcinogens, particularly with high or

repeated doses. Some of these chemicals can produce pheno-typic effects that mimic the action of tumor promoters (61).They might, therefore, act as both initiators and promoters(61).

With respect to tumor promotion, I have described excitingprogress in studies on PKC, but we still know very little of how,following PKC activation, signals are carried from the cytoplasm to the nucleus to alter patterns of gene expression.Further studies on the phorbin gene (38) and on various tix-acting regulatory elements and the corresponding activatorproteins that control the transcription of specific genes (62)may clarify these aspects. At the present time we do not knowwhy tumor promoters appear to act preferentially on the initiated cell to cause its clonal expansion. I suspect that the criticaleffects of tumor promoters involve not just clonal expansion ofthe initiated cell but also some type of heritable "imprinting."

Why, for example, is it that on mouse skin (4) and in certaintissue culture systems (60, 63-66), some of the effects of TPAare irreversible, since even after cessation of repeated treatmentswith TPA some skin papillomas do not regress, and transformed foci in cell culture remain stably transformed? There is

4140





also evidence that tumor promotion on mouse skin can besubdivided into two phases (4, 40), but we do not understandat a mechanistic level the differences between these two phases.Some investigators have emphasized that since in certain cellsystems TPA can induce the formation of activated forms ofoxygen and also cause sister chromÃ¢tid exchange, tumor promotion might involve indirect damage to cellular DNA (67). Idoubt that these effects explain the early effects of TPA duringtumor promotion, since these effects are usually reversible, butindirect DNA damage may explain some of the later andirreversible effects of TPA.

We understand even less about the molecular mechanismsresponsible for the process of tumor progression, i.e., the conversion of benign tumors to malignant tumors. Since there isevidence that the process is enhanced by genotoxic agents (68),progression may require further DNA damage. Amplificationof specific DNA sequences and instability of the karyotypeappear to play important roles in tumor progression and thetendency of malignant tumors to display increasing degrees ofautonomy and heterogeneity. Peyton Rous (69) described tumorprogression as "the process whereby tumors go from bad toworse." Obviously the mechanisms that underlie tumor pro

gression require much more intensive study, particularly in viewof the fact that tumor invasion, metastasis, and heterogeneitypresent the major clinical challenges in terms of cancer treatment.

Finally, I would emphasize that some of the current seemingly divergent hypotheses used to explain carcinogenesis neednot be mutually exclusive because of the complex multistagenature of the carcinogenic process, the diversity of causativefactors, and the heterogeneity of tumor cell phenotypes.

Implications of Molecular Carcinogenesis Research with Respectto Cancer Treatment and Prevention

Fig. 7 provides a schematic diagram of a cell showing variouspathways of signal transduction. Several growth factors, e.g.,PDGF, act through membrane-associated receptors to activatetyrosine protein kinases, whereas certain hormones act throughmembrane-associated receptors to increase cellular levels ofcyclic AMP, which then activate the cyclic AMP-dependentprotein kinase. The central role of PKC in signal transductionhas already been discussed in detail. A number of oncogenesact by disturbing specific aspects of these signal transductionpathways. Carcinogenesis can, therefore, be viewed as a "distortion in signal transduction." This concept suggests a variety

of strategies for developing novel drugs that might be usefulboth in cancer chemoprevention and in the therapy of fullyestablished tumors. These strategies are outlined in Table 2.Thus, research on multistage carcinogenesis, growth factors,and protein kinases should provide clinicians with a variety ofnew pharmacological approaches to cancer control.

Our laboratory has been particularly interested in developinginhibitors of PKC, in view of the central role of this enzyme intumor promotion and growth control (47, 70). In principle, onecould design inhibitors that act on either the catalytic or theregulatory (amino-terminal) domain of the enzyme (Fig. 5). Wehave concentrated on the latter approach since we believe thatsuch compounds would have greater specificity in view of theunique structure of this region of PKC. By screening a numberof compounds, we discovered that the antiestrogen tamoxifen(and certain structurally related triphenylethylenes) are ratherpotent inhibitors of PKC and have obtained evidence that theydo so by acting at the regulatory domain (44, 70). Tamoxifen

RECEPTOR

lclustering

endocytosis

0 AdrenergicEtc.

Fig. 7. Schematic diagram of a cell showing various pathways of signaltransduction mediated by membrane-associated receptors. The growth factorspathway applies to epidermal growth factor, PDGF, and insulin. The 0-adrenergicpathways involves the coupling of a specific receptor via a G regulatory protein(Ns) to adenylate cyclase (Ac), and cyclic AMP (cAMP) binding to the regulatorysubunit (R) of protein kinase A (/'A, I). Various agonists can activate phospholi-pase C (l'I. UM'C), presumably via a G protein, leading to the release of DAGand inositol 1,4,5-triphosphate (II',}. DAG activates PKC and inositol 1.4.5triphosphate causes the release of < .r " from the endoplasmic reticulum. This

leads to a cascade of events that alters the functions of membrane-associatedreceptors, ion channels, and cytoplasmic proteins. Signals (undefined) also enterthe nucleus to induce the expression of various genes including c-fos and c mir.Reprinted from Ref. 13, with permission.

Table 2 Strategies for modulating signal transduction

1. Block receptors: peptide analogues, mimetic drugs2. Inhibit protein kinases: PKC, tyrosine protein kinases3. Modulate ion functions: calmodulin inhibitors, ( 'Â¡r' channel blockers, etc.

4. Antagonize the mitogenic program: inhibitors of ornithine decarboxylase, myc,fos, ras, etc.

is of interest because of its efficacy in the treatment of breastcancer. Although it is known to compete with estrogens forbinding to the estrogen receptor, there is reason to believe thatits antitumor effects might also be due to action on othercellular targets (44). Tamoxifen may, therefore, be a usefulprototype for the design of specific inhibitors of PKC. Asstressed earlier, the development of such inhibitors could provide a powerful new approach to both cancer prevention andtreatment.

Finally, let me discuss the implications of studies on themechanism of action of chemical carcinogens with respect toresearch on human cancer causation, emphasizing an approachwhich I and my colleagues call "molecular epidemiology" (71,

72). At the present time there exist three distinct methods fordetecting potential human chemical carcinogens, epidemiolÃ³gica! studies, long-term rodent bioassays, and short-term assays,(e.g., the Ames bacterial mutagenesis assay). Although theshort-term assays are relatively simple and highly sensitive it isdifficult to extrapolate the results obtained to humans. At thesame time, conventional epidemiolÃ³gica!methods have seriouslimitations since they are generally retrospective and are notvery sensitive. Insights into the molecular mechanisms involvedin carcinogenesis provide new laboratory methods which can

4141





be used in epidemiolÃ³gica! studies, to more precisely identifythe causes of human cancers. Since, as I stressed at the beginning of this article, a number of carcinogens act by binding toDNA, one can now use highly sensitive techniques to detectand quantitate the possible presence of specific carcinogensbound to the DNA in human tissues. Toward this end, we andothers have developed monoclonal antibodies and highly sensitive immunoassays that can detect very low levels ofbenzo(a)pyrene and other carcinogens on human DNA (71,72).

Foundry workers are exposed to high concentrations of poly-cyclic aromatic hydrocarbons and they are known to be atincreased risk of developing lung cancer. Our group at Columbia University (73) and others (74) have found that theseworkers have increased levels of polycyclic aromatic hydrocar-bon-DNA adducts, when compared to control subjects. A variety of methods are now available, and others are being developed, which will make it possible to determine whether otherspecific chemicals encountered in the work place, the diet, orother environmental sources cause DNA damage in humans(71, 72). Current research on the roles of protein kinases andaltered gene expression during tumor promotion, which I havebriefly reviewed in this paper, should also provide novel methods for detecting the action of tumor promoters in humans,thus broadening our understanding of human cancer causation.

In summary, as we approach the latter years of the 20thcentury, cancer research has reached an exciting crescendo.Advances in molecular and cellular biology have merged withresearch on growth control and the mechanisms that underliemultistage carcinogenesis. I am confident, therefore, that evenbefore the year 2000 these advances will provide more preciseinformation on the causes of specific human cancers and alsolead to novel and more effective strategies for cancer preventionand treatment.

Acknowledgments

This lecture is dedicated to Drs. Elisabeth Miller and CharlotteFriend, both of whom passed away during the year 1987. These twowomen were inspiring and dedicated leaders in cancer research and hada profound influence on the author's own career. They made outstand

ing contributions, Dr. Elisabeth Miller in the areas of cancer biologyand chemical carcinogenesis and Dr. Charlotte Friend in the areas oftumor virology and cellular differentiation. The author is also indebtedto numerous colleagues, students, and collaborators for the invaluablecontributions they have made to his research. He is grateful to theNational Cancer Institute, the American Cancer Society, the AlmaToorock Memorial for Cancer Research, and the National Foundationfor Cancer Research for providing the financial support required tocarry out this research.

References

1. Doll, F. R. S., and Peto, R. The Causes of Cancer. Cambridge, UnitedKingdom: Oxford University Press, 1981.

2. Foulds. L. The experimental study of tumor progression: a review. CancerRes., 14: 327-339, 1954.

3. Berenblum, I. Sequential aspects of chemical carcinogenesis: skin. In: F. F.Becker (ed.), Cancer: A Comprehensive Treatise, Vol. 1, Ed. 2. New York:Plenum Publishing Corp., 1982.

4. Slaga, T. J., Sivak, A., and Boutwell, R. K. (eds.). Mechanisms of TumorPromotion and Cocarcinogenesis, Vol. 2. New York: Raven Press, 1978.

5. Weinstein, I. B. Current concepts and controversies in chemical carcinogenesis. J. Supramol. Struct. Cell. Biochem., 17:99-120, 1981.

6. Weinstein, I. B., Gattoni-Celli, S., Kirschmeier, P., Lamben, M., Hsiao, W.-I ., Backer, J., and Jeffrey, A. Multistage carcinogenesis involves multiplegenes and multiple mechanisms. In: A. Levine, G. Vande Woude, J. D.Watson, and W. C. Topp (eds.), Cancer Cells 1. The Transformed Phenotype.pp. 229-237. Cold Spring Harbor, NY: Cold Spring Harbor LaboratoryPress, 1984.

10.n-

13-

,4

15

17-

is.

19.

2o.

21.

22.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

Miller, E. Some current perspectives on chemical carcinogenesis in humansand experimental animals: Presidential address. Cancer Res., 38:1479-1496,1978.Jeffrey, A. M. DNA modification by chemical carcinogens. In: D. Grunbergerand S. Goff (eds.), Mechanism of Cellular Transformation by CarcinogenicAgents, pp. 33-72. Elmsford, NY: Pergamon Press, 1987.Grunberger, D., Santella, R. M., Hanau, L. H., and Erlanger, B. F. Stabilization of Z-DNA conformation by chemical carcinogens. In: E. Hubermanand S. H. Harr (eds.), Carcinogenesis, Vol. 10, pp. 465-480. New York:Raven Press, 1985.Bishop, J. M. Viral oncogenes. Cell, 42: 23-28, 1985.Guroff, G. Oncogenes, Genes and Growth Factors. New York: Wiley-Inter-science, 1987.Barbanti, M. Oncogenes and human cancer: cause or consequence? Carcinogenesis (Lond.), 7: 1037-1042, 1986.Weinstein, I. B. Growth factors, oncogenes and multistage carcinogenesis. J.Cell Biochem., 33: 213-224, 1987.Kennedy, A. R. Evidence that the first step leading to carcinogen-inducedmalignant transformation is a high frequency, common event. In:]. C. Barrettand R. W. Tennant (eds.), Carcinogenesis, Vol. 9. New York: Raven Press,1985.Lambert, M. E., (jarrÃ©is,J. E., McDonald, J., and Weinstein, I. B. Induciblecellular responses to DNA damage in mammalian cells. In: Shankel et al.(eds.), Antimutagenesis and Anticarcinogenesis, pp. 291-311. New York:Plenum Publishing Corp., 1986.Weinstein, I. B., Lambert, M. E., Carrels, J. E., Ronai, Z., Hsiao, W.-L.,Dragani, T. A., Hsieh, L. L., and Begemann, M. The role of asynchronousDNA replication and endogenous retrotransposons in multistage carcinogenesis. In: H. Zur Hausen and J. R. Schlehofer (eds.), Accomplishments inOncology, Vol. 2, No. 1, pp. 69-83. Philadelphia: J. B. Lippincott Co., 1987.Lamben, M. E., Pellegrini, S., Gattoni-Celli, S., and Weinstein, I. B. Carcinogen induced asynchronous replication of polyoma DNA is mediated by atrans acting factor. Carcinogenesis (Lond.), 7: 849-852, 1986.Ronai, Z., Lambert, M. E., Johnson, M. D., Okin, E., and Weinstein, I. B.Induction of asynchronous replication of polyoma DNA by ultraviolet irradiation and the effects of various inhibitors. Cancer Res., 47: 4565-4570,1987.Ronai, Z. A., and Weinstein, I. B. Identification of a UV induced franj-actingprotein that stimulates polyoma DNA replication. J. Virol., 62:1057-1060,1988.Lavi, S., Kohn, N., Kleinberger, T., Berko, Y., and Etkin, S. Amplificationof SV40 and cellular genes in SV40 transformed Chinese hamster embryocells treated with chemical carcinogens. In: E. Friedberg and B. Bridges(eds.). Cellular Responses to DNA Damage, pp. 656-670. New York: AlanR. Liss, Inc., 1983.Zur Hausen, H., and Schlehofer, J. R. (eds.). Accomplishments in Oncology,Vol. 2, No. 1. Philadelphia: J. B. Lippincott Co., 1987.Herrlich, P., Angel, P., Rahmsdorf, H. J., Mallick, U., Poting, A., Hieber,L., Lucke-Huhle, C., and Schorpp, M. The mammalian genetic stress response. Adv. Enzyme Regul., 25:485-504, 1986.McDonald, J. F., Strand, D. J., Lamben, M. E., and Weinstein, I. B. Theresponsive genome: evidence and evolutionary implications. In: R. A. Raffand E. C. Raff (eds.), MBL Lectures in Biology, Vol. 8, pp. 239-263. NewYork: Alan R. Liss, Inc., 1987.Baltimore, D. Retroviruses and retrotransposons: the role of reverse transcription in shaping the eukaryotic genome. Cell, 40:481-482, 1985.Ruff, E. L., Feensta, A., Lenders, K., Smith, L., Harvey, R., Horumi, N.,and Shulman, M. Intracisternal A-particle genes as movable elements in themouse genome. Proc. Nati. Acad. Sci. USA, 80: 1992-1996, 1983.Rechavi, G., Givol, D., and Canaani, E. Activation of a cellular oncogene byDNA rearrangement: possible involvement of an IS-like element. Nature(Lond.), 300:607-611, 1982.Gattoni-Celli, S., Hsiao, W.-L., and Weinstein, I. B. Rearranged c-mos locusin a MOPC-21 murine myeloma cell line and its persistence in hybridomas.Nature (Lond.), 306: 795-796, 1983.Ymer, S., Tucker, W. O. J., Sanderson, C. J., Hapel, A. J., and Campbell, I.G. Constitutive synthesis of interleukin-3 by leukaemia cell line WEHI-3Bis due to retroviral insertions near the gene. Nature (Lond.), 371: 255-258,1985.Kirschmeier, P., Gattoni-Celli, S., Dina, D., and Weinstein, I. B. Carcinogenand radiation transformed C3H10TVÃ•cells contain RNA's homologous to

the long terminal repeat sequences of a murine leukemia virus. Proc. Nati.Acad. Sci. USA, 79: 2773-2777, 1982.Hsiao, W.-L., Gattoni-Celli, S., and Weinstein, I. B. Effects of 5-azacytidineon the expression of endogenous retrovirus-related sequences in C3H1 OT'/zcells. J. Virol., 57:1119-1126, 1986.Housey, G. H., Kir hmeier. P., Gane, S. J., Burns, F., Troll, W., andWeinstein, I. B. Expression of long terminal repeat (LTR) sequences incarcinogen-induced murine skin carcinomas. Biochem. Biophys. Res. Commun., 727: 392-398, 1985.Dragani, T. A., Manenti, G., Della Porta, G., Gattoni-Celli, S., and Weinstein, I. B. Factors influencing the expression of endogenous retrovirus-related sequences in the liver of B6C3 mice. Cancer Res., 46: 1915-1919,1986.Hsieh, L. L., Hsiao, W.-L., Peraino, C., Maronpot, R. R., and Weinstein, I.B. Expression of retroviral sequences and oncogenes in rat liver tumorsinduced by diethylnitrosamine. Cancer Res., 47: 3421-3424,1987.

4142





34. Knudson. A. G., Jr. Hereditary cancer, oncogenes and antioncogenes. CancerRes., 45: 1437-1443, 1985.

35. Klein, G. The approaching era of the tumor suppressor genes. Science (Wash.DC), 238: 1539-1545, 1987. 55.

36. Naylor, S. L., Johnson, B. E., Minna, J. D., and Sakaguchi, A. Y. Loss ofheterozygosity of chromosome 3p markers in small cell lung cancer. Nature(Lond.), 239:451 -454, 1987. 56.

37. Bucinici".W. F., Bailey, C. J., Bucinici-,J., Bussey, H. J. R., Ellis, A., Gorman,

P., Lucibello, F. C, Murday, V. A., Rider, S. H., Scambler, P., Sheer, D.,Solomon, E., and Spurr. N. K. Localization of the gene for familial adenom- 57.atous polyposis on chromosome 5. Nature (Lond.), 328:614-616, 1988.

38. Johnson, M. D., Housey, G. M., Kirschmeier, P., and Weinstein, I. B.Molecular cloning of gene sequences regulated by tumor promoters and 58.mitogens through protein kinase C. Mol. Cell. Biol., 7:2821-2829, 1987.

39. Weinstein, I. B., Horowitz, A. D., Fisher, P., Ivanovic, V., Gattoni-Celli, S.,and Kirschmeier, P. Mechanisms of multistage carcinogenesis and their 59.relevance to tumor cell heterogeneity. In: A. H. Owens, D. S. Coffey, and S.B. Baylin (eds.). Tumor Cell Heterogeneity, pp. 261-283. New York: Academic Press, 1982. 60.

40. Diamond, L. Tumor promoters and cell transformation. In: G. Grunbergerand S. Goff (eds.). Mechanisms of Cellular Transformation by CarcinogenicAgents, pp. 73-134. Elmsford, NY: Pergamon Press, 1987. 61.

41. Castagna, M., Takai, Y., Kaibuchi, K., Sano, K., Kikkawa, U., and Nishizuka,Y. Direct activation of calcium-activated phospholipid-dependent proteinkinase by tumor-promoting phorbol esters. J. Biol. Chem., 257: 7847-7851, 62.1982.

42. Nishizuka, T. Perspectives on the role of protein kinase C in stimulus-response coupling. J. Nati. Cancer Inst., 76: 363-370, 1986. 63.

43. Berridge, M. J. Inositol triphosphate and diacylglycerol: two interactingsecond messengers. Annu. Rev. Biochem., 56:159-193, 1987.

44. O'Brian, C. A., Liskamp, R. M., Solomon, D. H., and Weinstein, I. B. 64.

Triphenylethylenes: a new class of protein kinase C inhibitors. J. Nati. CancerInst., 76:1243-1246, 1986.

45. House, C., and Kemp, B. E. Protein kinase C contains a pseudosubstrate 65.prototype in its regulatory domain. Science (Wash. DC), 238: 1726-1728,1987.

46. Housey, G. M., O'Brian, C. A., Johnson, M. D., Kirschmeier, P. T., and 66.

Weinstein, I. B. Isolation of cDNA clones encoding protein kinase C:evidence for a novel protein kinase ("-related gene family. Proc. Nati. Acad.

Sci. USA, 84:1065-1069, 1987. 67.47. Housey, G. M., Johnson, M. D., Hsiao, W.-L., O'Brian, C. A., Murphy, J.

P., Kirschmeier, P., and Weinstein, I. B. Overproduction of protein kinase 68.C causes disordered growth control in rat fibroblasts. Cell, 52: 343-354,1988.

48. Coussens, L., Parker, P. J., Rhee, L., Yang-Feng, T. L., Chen, E., Waterfield,M. D., Francke, I,'.. and Ullrich, A. Multiple, distinct forms of bovine and 69.human protein kinase C suggest diversity in cellular signaling pathways.Science (Wash. DC), 233: 859-866,1986.

49. Knopf, J. L., Lee, M-H., Sultzman, L. A., Kriz, R. W., Loomis, C. R., 70.Hewick, R. M., and Bell, R. M. Cloning and expression of multiple proteinkinase C cDNAs. Cell, 46:491 -502, 1986.

50. Kikkawa, U., Ogita, K., Ono, Y., Asaoka, Y., Shearman, M. S., Tomoko, F.,Ase, K., Sekiguchi, K., Igarashi, K., and Nishizuka, Y. The common structureand activities of four subspecies of rat brain protein kinase C family. FEBS 71.Lett., 233: 212-216, 1987.

51. Ohno, S., Kawasaki, H., Imajoh, S., Suzuki, K., Inagaki, M., Yokohura, H.,Sakohn, T., and Hidaka, H. Tissue-specific expression of three distinct types 72.of rabbit protein kinase C. Nature (Lond.), 325:161-166, 1987.

52. Huang, F. L., Yoshida, Y., Nakabayashi, H., Knopf, J. L., Young, W. S., 73.and Huang, K.-P. Immunochemical identification of protein kinase C iso-zymes as products of discrete genes. Biochem. Biophys. Res. Commun., 149:946-951, 1987. 74.

53. Jeffrey, A. M., and Liskamp, R. M. J. Computer-assisted molecular modelingof tumor promoters: rationale for the activity of phorbol esters, teleocidin B,and aplysiatoxin. Proc. Nati. Acad. Sci. USA, 83: 241-245, 1986.

54. Kirschmeier, P. T., Housey, G. M., Johnson, M. D., Perkins, A. S., and

Weinstein, I. B. Construction and characterization of a retroviral vectordemonstrating efficient expression of cloned cDNA sequences. DNA, 7:219-225, 1988.Fitzer, C. J., O'Brian, C. A., Guillem, J. G., and Weinstein, I. B. The

regulation of protein kinase C by chenodeoxycholate, deoxycholate andseveral related bile acids. Carcinogenesis (Lond.), 8: 217-220, 1987.Guillem, J. G., O'Brian, C. A., Fitzer, C. J., Johnson, M. D., FÃ¶rde,K. A.,

LoGerfo, P., and Weinstein, I. B. Studies on protein kinase C and coloncarcinogenesis. Arch. Surg., 122:1475-1478, 1987.Hsieh, L. L., Peraino, C., and Weinstein, I. B. Expression of endogenousretrovirus-like sequences and cellular oncogenes during phÃ©nobarbitaltreatment and regeneration in rat liver. Cancer Res., 48: 265-269, 1988.Hsiao, W.-L. W., Gattoni-Celli, S., and Weinstein, I. B. Effects of 5-azacytidine on the progressive nature of cell transformation. Mol. Cell. Biol.,5:1800-1803, 1985.Backer, J., and Weinstein, I. B. Interaction of benzo(a)pyrene and its dihy-drodiol epoxide derivative with nuclear and mitochondria! DNA inC3H10TV2 cell cultures. Cancer Res., 42:2764-2769, 1982.Fisher, P. B., Bozzone, J. H., and Weinstein, I. B. Tumor promoters andepidermal growth factor stimulate anchorage-independent growth of adeno-virus transformed rat embryo cells. Cell, 18:695-705, 1979.Ivanovic, V., and Weinstein, I. B. Benzo(o)pyrene and other inducers ofcytochrome Pi-450 inhibit binding of epidermal growth factor to cell surfacereceptors. Carcinogenesis (Lond.), 3: 505-510, 1982.Angel, P., Allegretto, E. A., Okino, S. T., Hattori, K., Boyle, W. J., Hunter,T., and Karin, M. Oncogene jun encodes a sequence specific frans-activatorsimilar to AP-1. Nature (Lond.), 332: 166-171, 1988.Colburn, N. H., Former, B. F., Nelson, K. A., and Yuspa, S. H. Tumorpromoter induces anchorage independence irreversibly. Nature (Lond.), 281:589-591, 1979.Hsiao, W.-L. W., Gattoni-Celli, S., and Weinstein, I. B. Oncogene-inducedtransformation of C3H10T1/: cells is enhanced by tumor promoters. Science(Wash. DC), 226: 552-555, 1984.Hsiao, W.-L. W., Wu, T., and Weinstein, I. B. Oncogene-induced transformation of a rat embryo fibroblast cell line is enhanced by tumor promoters.J. Mol. Cell. Biol., 6:1943-1950,1986.Dotto, G. P., Parada, L. F., and Weinberg, R. A. Specific growth responseof ros-transformed embryo fibroblasts to tumour promoters. Nature (Lond.),318:472-175,1985.( 'erutti. P. Prooxidant states and tumor promotion. Science (Wash. DC),

227: 375-381, 1985.Hennings, H., Shores, R., Wenk, M. L., Spangler, E. F., Tarane, R., andYuspa, S. Malignant conversion of mouse skin tumours is increased bytumour initiators and unaffected by tumour promoters. Nature (Lond.), 304:67-69, 1983.Rous, P., and Kidd, J. G. A comparison of virus-induced rabbit tumors withthe tumors of unknown cause elicited by tarring. J. Exp. Med., 69:339-434,1939.Weinstein, I. B., O'Brian, C. A., Housey, G. M., Johnson, M. D., Kirschmeier, P. T., and Hsiao, W.-L. Studies on the mechanism of action of proteinkinase C and the isolation of molecular clones encoding the enzyme. In: F.F. Becker and T. L. Slaga (eds.), Symposium on Fundamental CancerResearch, pp. 173-186. Austin, TX: University of Texas Press, 1987.Perera, F., and Weinstein, I. B. Molecular epidemiology and carcinogen-DN A adduci detection: new approaches to studies of human cancer causation.J. Chronic Dis., 35: 581-600,1982.Perera, F. P. Molecular cancer epidemiology: a new tool in cancer prevention.J. Nati. Cancer Inst., 78: 887-898, 1987.Perera, F. P., Hemminki, K., Young, T.-L., Brenner, D., Kelly, G., andSantella, R. M. Detection of polycyclic aromatic hydrocarbon-UN A adductsin white blood cells of foundry workers. Cancer Res., 48: 2288-2291, 1988.Harris, C. C., Vahakangas, K., Newman, M. J., Trivers, G. E., Shamsuddin,A., Sinopoli, N., Mann, D. L., and Wright, W. E. Detection of benzo-(a)pyrene dial epoxide-DNA adducts in peripheral blood lymphocytes andantibodies to the adducts in serum from coke oven workers. Proc. Nati. Acad.Sci. USA, Â«2:6672-6676, 1985.

4143




the origins of human cancer: molecular mechanisms of

Documents