J Toxicol Pathol 2002; 15: 123�128

Mini Review

Preneoplastic and Precancerous Lesions in Rodents: Morphologic and Molecular Characteristics

Jerrold M. Ward1

1Veterinary and Tumor Pathology Section, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Frederick, MD 21702�1201, USA

Abstract: Cancer evolves through a sequential process from normal cells in many tissues of humans and animals. Thenatural history of tumor development can be seen histologically and by biochemical and molecular changes. There aretwo common basic pathways for the formation of malignant epithelial tumors; through preneoplastic foci and benigntumors (carcinoma developing in an adenoma) in parenchymal tissue or progression from intraepithelial neoplasia (IN)(atypical hyperplasia, noninvasive carcinoma, carcinoma in situ), a lesion in flat or lining epithelium. In epithelial-liningtissues of humans and rodents (e.g. cervix, mammary gland, prostate, skin), these lesions have been described as IN. Insolid epithelial organs (liver, kidney, endocrine tissues) focal hyperplasia leads to adenomas. Adenomas develop fociof carcinoma, a process that is more common in rodents than in humans. These precancerous lesions in many rodenttissues often have multiple biochemical and molecular lesions which can be similar or different from those found inmalignant tumors. The rodent molecular lesions include mutations in oncogenes (K-ras, H-ras) and tumor suppressorgenes (p53, β-catenin,apc) or loss of heterozygosity (LOH) in tumor suppressor genes of mutant mouse models. Thismanuscript will review specific sequential morphologic and molecular lesions in the histopathogenesis of cancer inseveral rodent tissues. The significance of molecular lesions for diagnosis of rodent lesions will be discussed. (J Toxicol Pathol 2002; 15: 123�128)

Key words: pathology of cancer, molecular lesions, preneoplastic, precancerous

Introduction: The Histopathogenesis of Cancer

Cancer in both humans and rodents develops through asequence of histogenetic and molecular changes or events.Histologically, the sequence of morphological lesions hasbeen well described in humans1 and rodents2�4 in manytissues. The evolution of terminology and criteria forvarious lesions from hyperplasia to carcinoma hasprogressed from initial diagnoses by individual pathologiststo classifications based on consensus diagnoses, and criteriahave been developed by committees of expert pathologistsunder authority of various organizations including the WorldHealth Organization and the United States Society ofToxicologic Pathologists (which included pathologists fromThe United States, Europe, Japan and other countries).While molecular lesions of various types have more recentlybeen found in various stages of human cancer development5,lesions have been more limited to mutations in specificgenes in rodents. This paper will review the morphology,

Received: 2 April 2002, Accepted: 27 April 2002Mailing address: Jerrold M. Ward, Veterinary and Tumor Pathology Section, Center for Cancer Research, National Cancer Institute-FCRDC, Fairview 201, Frederick, MD 21702�1201, USATEL: +1-301-846-1239 FAX: +1-301-846-6389E-mail: ward@mail.ncifcrf.gov

classification, and molecular pathology of preneoplastic andprecancerous lesions in rodents.

Preneoplastic (Adenoma) Pathway (Fig. 1)

The first visible histological lesion in the evolution ofcancer in a specific tissue in rats and mice can be classifiedas a preneoplastic lesion. Preneoplastic lesions have beendescribed as a focus, preneoplastic focus, focus of cellularalteration (liver), focal hyperplasia, hyperplastic focus,atypical hyperplasia, nodular hyperplasia, hyperplasticnodule, atypia, and dysplasia (Table 1). These lesions arecharacteristically observed in parenchymatous epithelialtissues of, for example, lung and liver. Where studied indetail, such as in mouse and rat liver, these lesions appear torepresent clonal expansion of single cells, presumed to begenetically altered. As the small preneoplastic lesions growin size, they become nodular (show compression of adjacenttissue) and have been termed adenomas, hyperplasticnodules, nodular hyperplasia and even well differentiatedcarcinoma. These lesions, however, appear to representbenign tumors since they grow by expansion, are notinvasive remain well differentiated, and do not have themorphologic characteristics of carcinoma2�4. In mouse liverand lung adenomas, in skin papillomas, and, less commonly,

124 Preneoplastic and Precancerous Lesions

in rat liver and lung adenomas, focal areas of atypia,dysplasia or what appears as focal carcinoma can be foundwith increasing frequency within larger adenomas. Focalt rabecular carc inoma can be seen in most mousespontaneous and induced hepatocellular adenomas over 1cm in diameter6. As the carcinomatous areas in the adenomagrow, the tumor becomes characteristically malignant andinvasive and may metastasize to other tissues.

Adenomas may also simply enlarge and never progressto carcinoma. This situation is more common in humanseven though some human cancers may develop withinbenign lesions including colon carcinomas developing frompolyps (polypoid adenomas)1,7.

Precancerous Pathway (Fig. 2)

In humans, many epithelial cancers evolve fromnoninvasive lesions developing within flat lining epithelium.

Focus Adenoma

Carcinoma arising in an adenoma Metastases

Fig. 1. The preneoplastic (adenoma) pathway for mouse liver tumordevelopment.

Table 1. Terms Used for Preneoplastic and PrecancerousLesions in Rodents

Preneoplastic lesions Precancerous lesions(for an adenoma pathway) (for a preinvasive pathway)

Focus Focal hyperplasiaFocal hyperplasia Atypical hyperplasiaHyperplastic focus Nodular hyperplasiaAtypical hyperplasia AtypiaNodular hyperplasia DysplasiaHyperplastic nodule Carcinoma in situAtypia Intraepithelial neoplasiaDysplasia Adenoma/papilloma

These noninvasive lesions have been classified as focalhyperplasia, atypical hyperplasia, atypia, dysplasia,carcinoma in situ, and most recently, intraepithelialneoplasia (IN) (Table 1). Adenomas and papillomas are alsoby definition precancerous lesions. The precancerouspathway reviewed here refers primarily to invasivecarcinoma arising from flat hyperplastic lesions in flatepithelium as opposed to foci in benign tumors in solidtissues. In humans, this pathway is the most commonpathway for cancer development in many tissues includingbreast (DIN), prostate (PIN), stomach, esophagus andpancreas (PANIN)1. In rodents, a precancerous pathway iscommon in chemically induced epithelial tumors (e.g. skin,stomach, prostate) and in the prostate and mammary gland oftransgenic and knockout mice. IN is a relative new conceptfirst used for prostate lesions in humans1. It is a hyperplasticprecancerous/preinvasive state in a specific epithelial liningtissue. IN is morphologically defined for each tissue/species/animal model and proven progression to invasivecarcinoma is required. These preinvasive lesions often havemolecular changes (as described below) similar to ordifferent from those found in invasive carcinomas in thesame tissue.

Molecular Lesions in Preneoplast ic andPrecancerous Lesions

A long list of various types of molecular lesions havebeen found in all stages of human cancer progression (Table2)5. These lesions include chromosomal changes

Intraepithelial Neoplasia

Early Invasive Carcinoma

Late Invasive Carcinoma

Fig. 2. The precancerous pathway for rat colon tumor development.

Ward 125

( instabi l i ty, t ranslocat ions, insert ions, delet ions,amplification), DNA changes (microsatellite instability,genomic instability), specific gene changes (mutations,delet ions, methylat ion, loss of imprint ing, LOH,amplification) and mitochondrial DNA changes.

The most thoroughly studied model has been coloncancer7. In this disease, sequential changes involving at least5 molecular events have been described: 5q loss/APCmutation in colon dysplasia, DNA methylation changes inearly adenoma, K-ras activation in intermediate adenoma,18q LOH in la te adenoma, and p53 muta t ion inadenocarcinoma7. Much less is known about rodent tumorswhere usually only a single mutation or other molecularchange has been found in benign or malignant tumors.

The significance of molecular changes is potentiallyimmense. Depending on the gene in question, even a singlemutation can have large physiological consequences andeach subsequent molecular change can have additive effectson the disruption of normal cell cycle control, growth andtissue properties. The change should result in permanentlesions in genes (leading to abnormal gene expression)which has consequences for function of the gene and itsprotein. Molecular lesions producing a growth advantage toa cell or tissue can result in clonal expansion of the mutatedcell clone and lead to hyperplasia, benign neoplasia,malignant conversion, invasive lesions and metastaticdisease.

Molecular Lesions in Preneoplast ic andPrecancerous Lesions of Rodents

In preneoplastic and precancerous lesions and tumors ofrodents, many induced by chemical carcinogens, mutationshave been found in the ras oncogenes8, and in the tumorsuppressor genes p53, apc9 and β-catenin. Also, loss ofheterozygosity (LOH) is often found in mutant miceheterozygous for p53 , men1 , Rb , or Pten and geneamplification is common in tissues of many types oftransgenic mice. These molecular events may occur as earlychanges in rodent carcinogenesis since they are often found inpreneoplastic or precancerous lesions or benign tumors10�12.With the exception of some genetically engineered mousemodels13,14, a sequence of molecular changes has not yet beendescribed in rodent carcinogenesis as it has been in severalhuman cancers. We may find that, because experimental

Table 2. Classification of Molecular Changes in Preneoplastic,Precancerous Lesions, and Tumors

1. Chromosomal changes (instability, translocations, insertions,deletions, amplification)

2. DNA changes (microsatellite instability, genomic instability)3. Specific gene changes (mutations, deletions, methylation, loss

of imprinting, loss of heterozygosity, amplification)4. Mitochondrial DNA changes

models are less genetically diverse than is the humanpopulation, single molecular lesions may be sufficient forcarcinogenesis in mice and rats.

Various types of mutations in K-ras or H-ras are verycommon in both spontaneous and induced mouse tumorsincluding those in liver15�18, lung10,12,19�21, skin8, bloodvessels22, forestomach19,23, and mammary gland11; rasmutations are also common in rat tumors including those ofthe liver24,25, Zymbal�s gland23, bladder26, lung27, andmammary gland8. However, less information is available onras mutations in the preneoplastic or precancerous lesions ofrodents having been reported only in mouse liverfoci15,18,28,29, mammary hyperplastic alveolar nodules11, andmouse lung hyperplasias10, and in rat aberrant coloniccrypts30,31. The incidence and type of mutations can varies asa function of the chemical carcinogen, the dose and theanimal model system used8,15.

p53 mutations have been found in induced tumors of ratliver9,24,32, lung9, bladder9, and colon9, and in mouse tumorsof the small intestine9, lung12,33, bladder34, and skin9,35,36. Thereported incidences of these mutations are often low in onestudy and high in another. Few preneoplast ic orprecancerous lesions have been shown to have thesemutations consistently although mouse skin papillomasshowed a high frequency of p53 mutations in some studies9

but low incidences were found in rat liver foci32 and otherlesions34.

Recent studies have been performed for determinationof mutations in β-catenin and apc genes. β-cateninmutations have been found in mouse liver37�39, colon40 andbladder tumors41, and in rat aberrant colon crypts42,43 andtumors44,45; further, at least one hepatic preneoplastic focuswas immunoreactive for abundant β-catenin protein38. Apcmutations have been detected in the intestinal tumors of minmice46 and in rat colon tumors23. In mouse Apc mutants,additional Smad4 mutations appeared to contribute toinvasive colon carcinogenesis47.

Abnormal Expression of Antigens as Evidence ofMolecular Lesions

By immunohistochemistry, one can show expression ofproteins in normal tissues, in preneoplastic and precancerouslesions and in both benign and malignant tumors. In the caseof oncogenes, tumor suppressor genes, and some growthfactors, any detectable expression strongly suggestmolecular lesions are present in the tissues; for example,over-expression of TGFα is localized to the cell membranesof preneoplastic and neoplastic liver lesions in TGFαtransgenic mice (Fig. 3). Expression of nuclear p53 intumors is often the result of a more stable mutant proteinimmunoreacting with antibodies to p53 and ras 21immunoreactivity on cell membranes often indicatesmutations in ras oncogenes. Likewise, β-catenin expressionon cell membranes is highly suggestive of a mutation38.

126 Preneoplastic and Precancerous Lesions

Loss of Heterozygosity (LOH) in Mouse andHuman Mutants

LOH is used to describe loss of specific chromosomesor chromosome segments, chromosome deletions,chromosome rearrangements, loss of one parental copy of agene (i.e. allelic loss) and reduction to homozygosity. Lossof one parental copy of a gene frequently occurs in humanswith mutations in tumor suppressor genes such as VHL,MEN1, and p53 since their germline DNA is alreadyheterozygous for the mutated gene5. A homozygous mutantstate has not been described in humans for these genes and isassumed to be embryonic lethal; mice homozygous (-/-)often show embryonic lethality, while heterozygous (+/-)mice develop tumors showing LOH as they age. Thisphenomenon has been found for induced mouse mutants ofp5348�50, Menl51, Pten52�55, apc46,56, Tsc257, and Tsc158. LOHin tumors in wild-type mice or rats have been found in mouseliver tumors59, in mouse thymic lymphomas60 and rat23 andmouse mammary tumors61, often as a result of exposure tochemical carcinogens.

Fig. 3. Gene expression in preneoplastic and precancerous lesionsand tumors shown by immunohis tochemis t ry . Theimmunoreactivity on cell membranes or in nuclei may beindicative of molecular changes.

p53 in papilloma p53 in carcinoma

H-ras p21 in bladder hyperplasia β-catenin in liver focus

�������������

TGRα in liver focus

� ����

Conclusion

Molecular changes are common in tumors and in thepreneoplastic and precancerous lesions of both rodents andhumans. They are usually permanent, but may not beconclusively indicative of early neoplasia, early cancer oreven preneoplastic or precancerous lesions in humans62 andpossibly in rodents. However, the specific molecularchanges described in this review have always been found inpreneoplastic or precancerous lesions and in benign ormalignant tumors of rodents and should be considereddiagnostic of carcinogenic progression for the present time.

Acknowledgments: The author thanks Jon Summers andMarti Welch for graphics, Julie Foley for the β-catenin liversection, Kathleen Breeze and Deborah Devor-Henneman foreditorial assistance, and Jennifer Klabansky and DeborahShores for formatting. Supported, in part, with US federalfunds from the National Cancer Institute under contractNO1-CO-56000 to SAIC Frederick.

References

1. Henson DE and Albores-Saavedra J. Pathology of IncipientNeoplasia. 3rd ed., New York: Oxford University Press, 839,2001.

2. Mohr U. International Classification of Rodent Tumors. TheMouse. Berlin: Springer, 474, 2001.

3. Maronpot RR. Pathology of The Mouse. Vienna, IL: CacheRiver Press, 699, 1999.

4. Ward JM, Mahler JF, Maronpot RR, and Sundberg JP.Pathology of Genetically Engineered Mice. Ames, Iowa:Iowa State University Press, 394, 2000.

5. Mendelsohn J, Howley PM, Israel MA, and Liotta LA. TheMolecular Basis of Cancer, 2nd ed., Philadelphia: WBSaunders Co., 289�312, 2001.

6. Jang JJ, Weghorst CM, Henneman JR, Devor DE, and WardJ M . P ro g re ss i v e a t y p i a i n s p o nt an e o us an d N-nitrosodiethylamine induced hepatocellular adenomas ofC 3H/ HeN Cr mic e ev a lua t ed by ima ge an a ly s i s .Carcinogenesis 1992; 13: 1541�1547.

7. Fearon ER and Gruber SB. Molecular abnormalities in colonand rectal cancer. In: The Molecular Basis of Cancer, 2nded., J Mendelsohn, PM Howley, MA Israel, and LA Liotta(eds), Philadelphia: WB Saunders Co., 289�312, 2001.

8. Sills RC, Borman GA, Neal JE, Hong HL, and DevereuxTR. Mutations in ras genes in experimental tumours ofrodents. In: The Use of Short- and Medium-term Tests forCarcinogens and Data on Genetic Effects in CarcinogenicHazard Evaluation, DB McGregor, JM Rice, and S Venitt(eds), Lyon: IARC Sci Publ No. 146, 55�86, 1999.

9. Perantoni AO and Rice JM. Mutation patterns in non-rasoncogenes and tumour suppressor genes in experimentallyinduced tumours. IARC Sci Publ 1999; 146: 87�122.

10. Leone-Kabler S, Wessner LL, McEntee MF, D�Agostino, JrRB, and Miller MS. Ki-ras mutations are an early event andcorrelate with tumor stage in transplacentally-inducedmurine lung tumors. Carcinogenesis 1997; 18: 1163�1168.

11. Miyamoto S, Sukumar S, Guzman RC, Osborn RC, andNandi S. Transforming c-Ki-ras mutation is a preneoplastic

Ward 127

event in mouse mammary carcinogenesis induced in vitro byN-methyl-N-nitrosourea. Mol Cell Biol 1990; 10: 1593�1599.

12. Horio Y, Chen P, Roth JA, Malkinson AM, and SchrumpDS. K-ras and p53 mutations are early and late events,respectively, in urethane-induced pulmonary carcinogenesisin A/J mice. Mol Carcinog 1996; 17: 217�223.

13. Brodie SG, Xu X, Qiao W, Li WM, Cao L, and Deng CX.Multiple genetic changes are associated with mammarytumorigenesis in Brca1 conditional knockout mice.Oncogene 2001; 20: 7514�7523.

14. Green JE, Shibata MA, Yoshidome K, Liu ML, Jorcyk C,Anver MR, Wiggin ton J , Wi l t rou t R , Sh iba ta E ,Kaczmarczyk S, Wang W, Liu ZY, Calvo A, and CouldreyC. The C3(1)/SV40 T-antigen transgenic mouse model ofmammary cancer: ductal epithelial cell targeting withmultistage progression to carcinoma. Oncogene 2000; 19:1020�1027.

15. Maronpot RR, Fox T, Malarkey DE, and Goldsworthy TL.Mutations in the ras proto-oncogene: clues to etiology andmolecular pathogenesis of mouse liver tumors. Toxicology1995; 101: 125�156.

16. Boivin-Angele S, Lefrancois L, Froment O, Spiethoff A,Bogdanffy MS, Wegener K, Wesch H, Barbin A, Bancel B,Trepo C, Bartsch H, Swenberg J, and Marion MJ. Ras genemutations in vinyl chloride-induced liver tumours arecarcinogen-specific but vary with cell type and specifics. IntJ Cancer 2000; 85: 223�227.

17. Manam S, Storer RD, Prahalada S, Leander KR, KraynakAR, Ledwith BJ, van Zwieten MJ, Bradley MO, and NicholsWW. Activation of the Ha-, Ki-, and N-ras genes inchemically induced liver tumors from CD-1 mice. CancerRes 1992; 52: 3347�3352.

18. Richardson KK, Helvering LM, Copple DM, Rexroat MA,Linville DW, Engelhardt JA, Todd GC, and Richardson FC.Genetic alterations in the 61st codon of the H-ras oncogeneisolated from archival sections of hepatic hyperplasias,adenomas and carcinomas in control groups of B6C3F1mouse bioassay studies conducted from 1979 to 1986.Carcinogenesis 1992; 13: 935�941.

19. Hayashi S-M, Hong H-HL, Toyoda K, Ton T-VT, DevereuxTR, Maronpot RR, Huff J, and Sills RC. High frequency ofras mutations in forestomach and lung tumors of B6C3F1mice exposed to 1-amino-2,4-dibromoanthraquinone for 2years. Toxicol Pathol 2001; 29: 422�429.

20. Malkinson AM. Molecular comparison of human and mousepulmonary adenocarcinomas. Exp Lung Res 1998; 24: 541�555.

21. Miller MS, Leone-Kabler S, Rollins LA, Wessner LL, FanM, Schaeffer DO, McEntee MF, and O�Sullivan MG.Molecular pathogenesis of transplacentally induced mouselung tumors. Exp Lung Res 1998; 24: 557�577.

22. Duddy SK, Gorospe SM, Bleavins MR, and delaIglesia FA.Spontaneous and thiazolidinedione-induced B6C3F1 mousehemangiosarcomas exhibit low ras oncogene mutationfrequencies. Toxicol Appl Pharmacol 1999; 160: 133�140.

23. Nagao M, Ushijima T, Toyota M, Inoue R, and Sugimura T.Genetic changes induced by heterocyclic amines. Mutat Res1997; 376: 161�167.

24. Tamano S, Ward JM, Diwan BA, Keefer LK, Weghorst CM,Calvert RJ, Henneman JR, Ramljak D, and Rice JM.Histogenesis and the role of p53 and K-ras mutations in

hepatocarcinogenesis by glyceryl trinitrtes (nitroglycerin) inmale F344 rats. Carcinogenesis 1996; 17: 2477�2486.

25. Watatani M, Perantoni AO, Reed CD, Enomoto T, WenkML, and Rice JM. Infrequent activation of K-ras, H-ras, andother oncogenes in hepatocellular neoplasms initiated bymethyl(acetoxymethyl)nitrosamine, a methylating agent,and promoted by phenobarbital in F344 rats. Cancer Res1989; 49: 1103�1109.

26. Enomoto T, Ward JM, and Perantoni AO. H-ras activationand ras p21 expression in bladder tumors induced in F344/NCr rats by N-butyl-N-(4-hydroxybutyl)nitrosamine.Carcinogenesis 1990; 11: 2233�2238.

27. Stegelmeier BL, Gillett NA, Rebar AH, and Kelly G. Themolecular progression of plutonium-239-induced rat lungcarcinogenesis: Ki-ras expression and activation. MolCarcinog 1991; 4: 43�51.

28. Buchmann A, Mahr J, Bauer-Hofmann R, and Schwartz M.Mutations at codon 61 of the Ha-ras proto-oncogene inprecancerous liver lesions of the B6C3F1 mouse. MolCarcinog 1989; 2: 121�125.

29. Bauer-Hofmann R, Klimek F, Buchmann A, Muller O,Bannasch P, and Schwarz M. Role of mutations at codon 61of the c-Ha-ras gene during diethylnitrosamine-inducedhepatocarcinogensis in C3H/He mice. Mol Carcinog 1992;6: 60�67.

30. Tachino N, Hayashi R, Liew C, Bailery G, and Dashwood R.Ev id en c e fo r ra s g e n e mu t a t i o n i n 2 -a m i n o -3 -methylimidazo[4,5-f]quinoline-induced colonic aberrantcrypts in the rat. Mol Carcinog 1995; 12: 187�192.

31. Stopera SA, Murphy LC, and Bird RP. Evidence for a rasgene mutation in azoxymethane-induced colonic aberrantcrypts in Sprague-Dawley rats: earliest recognizableprecur sor l es ions o f exper imenta l co lon cancer .Carcinogenesis 1992; 13: 2081�2085.

32. Zhu H, Zhang X, and Chen Y. The expression of mutant(mp53) protein in rat hepatic preneoplastic lesions inducedby diethylnitrosamine (DEN). Zhonghua Zhong Liu Za Zhi1997; 19: 22�24.

33. Hegi ME, Soderkvist P, Foley JF, Schoonhoven R,Swenberg JA, Kari F, Maronpot R, Anderson MW, andWiseman RW. Characterization of p53 mutations inmethylene chloride-induced lung tumors from B6C3F1mice. Carcinogenesis 1993; 14: 803�810.

34. Yamamoto S, Tatematsu M, Yamamoto M, Fukami H, andFukushima S. Clonal analysis of urethelial carcinomas inC3H/HeN↔BALB/c chimeric mice treated with N-butyl-N-(4-hydroxybutyl)nitrosamine. Carcinogenesis 1998; 19:855�860.

35. Queille S, Seite S, Tison S, Medaisko C, Drougard C,Fourtanier A, Sarasin A, and Daya-Grosjean L. p53mutations in cutaneous lesions induced in the hairless mouseby a solar ultraviolet light simulator. Mol Carcinog 1998; 22:167�174.

36. Stenback F, Makinen M, and Jussila T. p53 expression inskin carcinogenesis and its relationship to cell proliferationand tumour growth. Eur J Cancer 1998; 34: 1415�1424.

37. Calvisi DF, Factor VM, Loi R, and Thorgeirsson SS.Activation of β-catenin during hepatocarcinogenesis intransgenic mouse models: Relationship to phenotype andtumor grade. Cancer Res 2001; 61: 2085�2091.

38. Devereux TR, Anna CH, Foley JF, White CM, Sills RC, andBarrett JC. Mutation of β-catenin is an early event in

128 Preneoplastic and Precancerous Lesions

chemically induced mouse hepatocellular carcinogenesis.Oncogene 1999; 18: 4726�4733.

39. Ogawa K, Yamada Y, Kishibe K, Ishizaki K, and TokusashiY. β-Catenin mutations are frequent in hepatocellularca rc i n o ma s b u t a b sen t i n ad en o ma s i n du c ed bydiethylnitrosamine in B6C3F1 mice. Cancer Res 1999; 59:1830�1833.

40. Takahashi M, Nakatsugi S, Sugimura T, and WakabayashiK. Frequent mutations of the beta-catenin gene in mousecolon tumors induced by azoxymethane. Carcinogenesis2000; 21: 1117�1120.

41. Shiina H, Igawa M, Urakami S, Shigeno K, Yoneda T,Terashima M, Deguchi M, Ribeiro-Filho L, and Dahiya R.Alterations of beta- and gamma-catenin in N-butyl-N-(-4-hydroxybutyl) nitrosamine-induced murine bladder cancer.Cancer Res 2001; 61: 7101�7109.

42. Yamada Y, Yoshimi N, Hirose Y, Kawabata K, MatsunagaK, Shimizu M, Hara A, and Mori H. Frequent β-cateningene mutations and accumulations of the protein in theputative preneoplastic lesions lacking macroscopic aberrantcrypt foci appearance, in rat colon carcinogenesis. CancerRes 2000; 60: 3323�3327.

43. Yamada Y, Yoshimi N, Hirose Y, Matsunaga K, KatayamaM, Sakata K, Shimizu M, Kuno T, and Mori H. Sequentialanalysis of morphological and biological properties of β-catenin-accumulated crypts, provable premalignant lesionsi n de p en d en t o f ab e r ran t c ry p t fo c i i n r a t c o l o ncarcinogenesis. Cancer Res 2001; 61: 1874�1878.

44. Koesters R, Hans MA, Benner A, Prosst R, Boehm J, GahlenJ, and Doeberitz MK. Predominant mutation of codon 41 ofthe β-catenin proto-oncogene in rat colon tumors induced by1,2-dimethylhydrazine using a complete carcinogenicprotocol. Carcinogenesis 2001; 22: 1885�1890.

45. Takahashi M, Fukuda K, Sugimura T, and Wakabayashi K.β-Catenin is frequently mutated and demonstrates alteredcellular location in azoxymethane-induced rat colon tumors.Cancer Res 1998; 58: 42�46.

46. Andreassen A, Vikse R., Steffensen IL, Paulsen JE, andAlexander J. Intestinal tumors induced by the foodcarcinogen 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine in multiple intestinal Neoplasia mice havetruncation mutations as well as loss of the wild-type Apc(+)allele. Mutagenesis 2001; 16: 309�315.

47. Seno H, Oshima M, Ishikawa T, Oshima H, Takaku K,Chiba T, Narumiya S, and Taketo MM. Cyclooxygenase 2-and Prostaglandin E2 receptor EP2-dependent angiogenesisin Apc∆716 mouse intestinal polyps. Cancer Res 2002; 62:506�511.

48. French J, Storer RD, and Donehower LA. The nature of theheterozygous Trp53 knockout model for identification ofmutagenic carcinogens. Toxicol Pathol 2001; 29: 24�29.

49. Lawler J, Miao W-M, Duquette M, Bouck N, Bronson RT,and Hynes RO. Thrombospondin-1 gene expression affectssurvival and tumor spectrum of p53-deficient mice. Am JPathol 2001; 159: 1949�1956.

50. Venkatachalam S, Tyner SD, Pickering CR, Boley S, RecioL, French JE, and Donehower LA. Is p53 haploinsufficientfor tumor suppression? Implications for the p53+/- mousemodel in carcinogenicity testing. Toxicol Pathol 2001; 29:147�154.

51. Crabtree JS, Scacheri PC, Ward JM, Garrett-Beal L,Emmert-Buck MR, Edgemon KA, Lorang D, Libutti SK,Chandrasekharappa DC, Marx SJ, Spiegel AM, and CollinsFS. A mouse model of multiple endocrine Neoplasia, type 1,develops multiple endocrine tumors. Proc Natl Acad SciUSA 2001; 98: 1118�1123.

52. Kim MJ, Cardiff RD, Desai N, Banach-Petrosky WA,Parsons R, Shen MM, and Abate-Shen C. Cooperativity ofNkx3.1 and Pten loss of function in a mouse model ofprostate carcinogenesis. Proc Natl Acad Sci USA 2002; 99:2884�2889.

53. Stambolic V, Tsao MS, Macpherson D, Suzuki A, ChapmanWB, and Mak TW. High incidence of breast and endometrialNeoplasia resembling human Cowden syndrome in pten +/-mice. Cancer Res 2000; 60: 3605�3611.

54. Podsypanina K, Ellenson LH, Nemes A, Gu J, Tamura M,Yamada KM, Cordon-Cardo C, Catoretti G, Fisher PE, andParsons R. Mutation of Pten/Mmac1 in mice causesneoplasia in multiple organ systems. Proc Natl Acad SciUSA 1999; 96: 1563�1568.

55. Kwabi-Addo B, Giri D, Schmidt K, Podsypanina K, ParsonsR, Greenberg N, and Ittmann M. Haploindufficiency of thePten tumor suppressor gene promotes prostate cancerprogression. Proc Natl Acad Sci USA 2001; 98: 11563�11568.

56. Haines J, Dunford R, Moody J, Ellender MJ, Cox R, andSilver A. Loss of heterozygosity in spontaneous and X-ray-induced intestinal tumors arising in F1 hybrid min mice:evidence for sequential loss of APC(+) and Dpc4 in tumordevelopment. Genes Chromosomes Cancer 2000; 28: 387�394.

57. Kobayashi T, Minowa O, Kuno J, Mitani H, Hino O, andNoda T. Renal carcinogenesis, hepatic hemangiomatosis,and embryonic lethality caused by a germ-line Tsc2mutation in mice. Cancer Res 1999; 59: 1206�1211.

58. Kobayashi T, Minowa O, Sugitani Y, Takai S, Mitani H,Kobayashi E, Noda T, and Hino O. A germ-line Tsc1mutation causes tumor development and embryonic lethalitythat are similar, but not identical to, those caused by Tsc2mutation in mice. Proc Natl Acad Sci USA 2001; 98: 8762�8767.

59. Davis LM, Caspary WJ, Sakallah SA, Maronpot RR,Wiseman R, Barrett JC, Elliot R, and Hozier JC. Loss ofheterozygosity in spontaneous and chemically inducedtumors of the B6C3F1 mouse. Carcinogenesis 1994; 15:1637�1645.

60. Santos J, Herranz M, Fernandez M, Vaquero C, Lopez P,and Fernandez-Piqueras J. Evidence of a possible epigeneticinactivation mechanism operating on a region of mousechromosome 19 in gamma-radiation-induced thymiclymphomas. Oncogene 2001; 20: 2186�2189.

61. Ritland SR, Rowse GJ, Chang Y, and Gendler SJ. Loss ofheterozygosity analysis in primary mammary tumors andlung metastases of MMTV-MTAg and MMTV-neutransgenic mice. Cancer Res 1997; 57: 3520�3525.

62. Hunt JD, Mera R, Strimas A, Gillespie AT, Ruiz B, CorreaP, and Fontham ETH. KRAS mutations are not predictive forprogression of preneoplastic gastric lesions. CancerEpidemiol Biomarkers Prev 2001; 10: 79�80.