Expression and Significance of p53, Rb, p21/waf-1, p16/ink-4a, and PTEN Tumor Suppressors in Canine Melanoma

Vet Pathol 39:458 472 (2002) Expression and Signifiane of p53, Rb, p21/waf-1, p16/ink-4a, and PTEN Tumor Suppressors in Canine Melanoma A. KOENIG, S. R. BIANCO, S. FOSMIRE, J. WOJCIESZYN, AND J. F. MODIANO Department of Small Animal Mediine and Surgery, College of Veterinary Mediine, Texas A&M University, College Station, TX (AK 1 ); Center for Caner Causation and Prevention, AMC Caner Researh Center (SRB, SF, JFM) and University of Colorado Caner Center (JFM), Denver, CO; and IHC Servies, Smithville, TX (JW) Abstrat. The role of tumor suppressor genes in the pathogenesis of anine melanoma is inompletely understood. The genes enoding the tumor suppressors p53, Rb, p21 (waf-1), p16 (ink-4a), and PTEN have been postulated to ontribute to the pathogenesis of melanoma in humans and experimental animal models. To assess whether inativation of these genes similarly ontributes to the origin and progression of anine melanoma, we examined their expression in seven distint anine melanoma ell lines and in 31 retrospetive samples (representing 29 dogs) of spontaneous anine melanoma. Various patterns suggestive of loss of tumor suppressor funtion emerged in these ell lines. The most frequently observed abnormality was loss or signifiant redution of p16 expression in six of seven ell lines and in 21 of 26 tumor samples. Loss or signifiant redution of PTEN expression was seen in four of seven ell lines and in 13 of 27 tumor samples. Although p53 was detetable in all the ell lines and in 24 of 30 tumors, exlusion of p53 from the nulear ompartment was observed in eah of the ell lines and in 18 of 25 tumor samples. These results indiate that loss of funtion of these tumor suppressor proteins is a ommon ourrene that may ontribute to the origin of anine melanoma. In our sample population, abnormalities in the expression or loalization of one or more tumor suppressor proteins ourred with similar frequeny in malignant and benign tumors; thus, additional work is neessary to determine how these proteins may impat disease progression and response to therapy. Key words: Dogs; gene expression; immunoblotting; immunohistology; melanoma; tumor suppressor genes. Melanoma is a ommon tumor in dogs that arises from melanoytes or melanoblasts. It is the most ommon malignany found in the oral avity and on the digits in the dog. 21,65 Dermal melanomas are usually benign, but uveal, digital, and oral melanomas generally are malignant and respond poorly to standard therapies. 46 There are, however, exeptions to these generalities; a few dogs with oral melanoma are ured of the disease, whereas others with dermal melanoma die from metastasis. Thus, shared geneti harateristis may be as or more important than loation. The natural history and pathogenesis of anine melanoma are inompletely understood, but reent work has begun to eluidate the role of growth regulation genes in the development of this disease. 46 Cellular growth and division is a omplex pathway governed by ompeting growth inhibitors and promoters. Protoonogenes enode proteins that promote ell yle progression, proliferation, and survival in normal ells; 1 Present address: Department of Clinial Studies, Shool of Veterinary Mediine, University of Pennsylvania, Philadelphia, PA 19104. tumor suppressor genes do the opposite, enoding proteins that inhibit ellular proliferation and promote ell death. 68 For ells that have undergone malignant transformation, the balane between proliferation and inhibition is upset. For this study, we used reverse transription (RT) polymerase hain reation (PCR) assays, immunoblotting, and immunoytohemistry to examine gene expression and protein aumulation of the p53, p21 (waf-1), p16 (ink-4a), Rb, and PTEN tumor suppressor genes in seven independent ell lines and immunohistohemistry to assess the presene and subellular loation of these proteins in 31 retrospetive tumor samples. Our hypotheses were that abnormalities leading to loss of funtion of these pathways ontribute to the pathogenesis of anine melanoma and that defets in multiple tumor suppressor genes would predit a worse outome. These genes and proteins are members of three interrelated negative growth regulatory pathways. The p53 gene enodes a nulear protein (p53 or TP53) that is integral to maintenane of DNA integrity and is a ornerstone of the DNA repair mahinery. 36 Indution of p53 an result in apoptosis or growth ar- 458

Vet Pathol 39:4, 2002 Tumor Suppressors in Canine Melanoma 459 rest in ells with irreparable DNA damage. 36 The retinoblastoma suseptibility gene RB-1 is a tumor suppressor gene that enodes a protein (Rb) that regulates the transition from the G1 phase to the S phase of the ell yle. 35,69 Rb is initially made in the ytoplasm and is then transported into the nuleus, where it exerts its effets. The ylin-dependent kinase inhibitors (CDKI) p21 (the produt of the waf-1 gene) and p16 (the produt of the ink-4a gene) prevent ell yle progression and promote growth arrest or apoptosis. 25,62,63 PTEN is a gene that enodes a bifuntional phospholipid and tyrosine phosphatase, whih antagonizes growth and survival signals transmitted through the phosphoinositide-3 kinase pathways 4,40,43,66 and may play a role in angiogenesis. 5,29 Redued or absent expression of p16 and PTEN and aumulation of p53 in an inappropriate subellular loation were the most frequent abnormalities identified in this study. Loss of these tumor suppressor pathways seems to be a ommon ourrene that may ontribute to the pathogenesis of anine melanoma. Further studies are neessary to onfirm the prognosti signifiane of these biomarkers in anine melanoma. Materials and Methods Cell lines and normal tissues The melanoma ell lines utilized for this study were derived from primary tumors of dogs with oral melanoma (CML-2, CML-13, JENNY, SCOOTER, TLM-1), a lymph node metastasis of utaneous melanoma (CML-6MC2), and a lung metastasis of oral melanoma (SHADOW) and have been desribed previously. 12,33,56,73 The Cf2Th line, originating from normal anine thymi epithelial ells, was obtained from the Amerian Type Culture Colletion (Rokville, MD). Normal tissue samples obtained from healthy dogs at the time of euthanasia were generously provided by Dr. John Bauer. The proedures for handling and euthanasia of the dogs and for the transfer of samples to this study were approved by the Institutional Animal Care and Use Committee of Texas A&M University. Clinial ases Clinial ases were retrieved from the medial reords database of the Texas Veterinary Medial Center. Every ase that had a histologi diagnosis of melanoma between 1994 and 1997 was reviewed for eligibility; 31 tumor samples with formalin-fixed arhival material that were suitable for immunostaining were utilized. Samples were exluded when 1) bloks did not ontain suffiient material for serial setioning, 2) tumor material was not evident on routine hematoxylin and eosin stained reuts, 3) bloks had evidene of improper or inomplete fixation, suh as loss of ellular detail along plasma membrane and nulear membrane or dereased staining intensity, 4) bloks had evidene of improper or inomplete infiltration of paraffin (based on their gross appearane), and 5) tissues were negative for vimentin ontrol stains, where normal stromal tissues in the blok were evaluated for vimentin expression by immunostaining. The features of some of these ases were reported previously. 33 Some tissue bloks did not yield suffiient tissue to analyze all of the markers inluded in this study. The eligible ases inluded 11 primary oral tumors (one tongue, three lip, seven gingiva), 18 dermal melanomas (three digit, three limb, two eyelid, three trunk, one ear, one srotum, five unspeified), and two lymph node metastasis (one from an oral tumor and one from a digital tumor). Twenty-one tumors were pigmented, and ten were amelanoti. The tumors were ategorized as malignant or benign based on pleomorphism, degree of proliferation, loal invasion, and doumentation of metastasis. Pleomorphism and proliferation were not by themselves onsidered adequate indiators of malignany in the absene of known reurrene, invasiveness, or doumentation of metastasis. Using these riteria, 14 tumors were benign, 11 were identified as malignant, and the behavior of six was not evident from the histology or history. The samples represented 29 dogs and 17 different breeds: four Gordon Setters, four Doberman Pinshers, three Labrador Retrievers, two Miniature Shnauzers, two Boxers, one eah of Giant Shnauzer, Brittany Spaniel, Bihon Frise, Cairn Terrier, Bull Terrier, Weimaraner, Bull Mastiff, Dahshund, Shetland Sheepdog, Golden Retriever, English Pointer, and Greyhound, and two mixed-breed dogs. The dogs ranged in age from 2 to 14 years, with an average age of 11 years and a median age of 12 years. The age of one dog was unknown. Thirteen of the dogs were neutered males, three were intat males, 12 were spayed females, and one was an intat female. Cell ulture Tissue ulture materials were obtained from Nalge Nun (Naperville, IL). Cells were ultured in Dulbeo s modified Eagle medium (DMEM, Gibo BRL, Grand Island, NY) ontaining 10% heat-inativated fetal alf serum (FCS, Hylone, Logan, UT) in a humidified atmosphere ontaining 5% CO 2 at 37 C. All the ell lines grew as monolayer ultures and were maintained by passage when they reahed 90% onfluene. Identifiation of a partial DNA for anine ink-4a The sequene for anine ink-4a was unknown. To obtain a partial sequene of anine ink-4a, we used a set of ommerially available PCR primers (Researh Genetis, Huntsville, AL) designed speifially to examine deletions of ink- 4a in human tumors. The ommerial primers onsist of a onserved sequene within the first intron of ink-4a (sense) and within exon 2 (antisense). Genomi DNA was isolated from normal anine liver using a ommerially available kit (Gentra Systems, Minneapolis, MN) as per the manufaturer s instrutions. PCR was performed using the primers and onditions speified (Researh Genetis). The PCR using the Researh Genetis primers and genomi DNA resulted in amplifiation of a 393-base pair (bp) produt. To onfirm the identity of the putative ink-4a DNA, we isolated a osmid lone from a library (provided by Dr. E. Ostrander, Fred Huthinson Caner Researh Center, Seattle, WA) using a human ink-4a DNA (provided by Dr. D. Beah, Cold Spring Harbor Laboratories, Cold Spring Harbor, NY) as a

460 Koenig, Biano, Fosmire, Wojieszyn, and Modiano Vet Pathol 39:4, 2002 sequene within intron 1 and 243 bp within the oding domain of exon 2 (highlighted). The 243 bp of anine ink-4a exon 2 enompass approximately 56% of the highly onserved sequene that enodes the ankyrin repeats based on the human homologue. Figure 2 shows that the translated partial protein produt onsists of 81 amino aids (of a predited 145, based on the human sequene), with 78/81 identities, one onserved substitution, and two nononserved substitutions. The anine ink-4a fragment also has up to 88% identity at the nuleotide level with porine, murine, rat, and equine ink-4a (also known as CDKN2A) and with the losely related genes ink-4b, ink-4, and ink-4d that enode the p15, p18, and p19 proteins, respetively. The partial anine p16 protein also shares 76 88% identity at the protein level with the porine, murine, rat, equine, and opossum p16 proteins and 55% identity with the platyfish (Xiphophorus maulatus) p16 protein. Preditably, this amino aid sequene is also highly homologous to human and rodent p15; the ankyrin repeats in all the INK-4 proteins are highly onserved. Fig. 1. Double-stranded nuleotide sequene of a partial anine ink-4a DNA region enompassing 393 bp inluding 150 nuleotides of intron 1 sequene and 243 nuleotides of exon 2 sequene. The oding domain for exon 2 is represented by the partial amino aid sequene shown below the nuleotide sequene (highlighted). Nuleotides within the intron represented by an N and highlighted in gray represent bases where the preise sequene is unertain. sreening tool. One positive lone was seleted from a tertiary sreen. PCR amplifiation using the same primers and the osmid as a template produed an idential sequene to that obtained by RT-PCR from anine liver RNA. Figure 1 shows the arrangement of the partial anine ink-4a DNA (GenBank aession No. AF234176), inluding 150 bp of Gene expression Gene expression was analyzed by RT-PCR as previously desribed. 55 RNA was isolated using the RNAWiz kit (Ambion, Austin, TX) as reommended by the manufaturer. Complementary DNA synthesis was aomplished using a kit (Rohe Diagnostis, Chiago, IL) followed by PCR amplifiation with dog-speifi primers. The sequene of the omplete oding domains for the p53 gene is available in GenBank (aession No. AB020761). Amplifiation of p53 mrna was done using a primer pair (sense 5 -CTGGCT AGACGAAGACTCAG-3, antisense 5 -AGGCAGTGCT CGCTTGGTAC-3 ) under onditions desribed previously 38 to produe a 738-bp amplifiation produt representing amino aids 52 296. The amplifiation produts were gel purified and submitted for sequening to the sequening ore faility of the University of Colorado Caner Center. Reproduible sequenes representing amino aids 72 285 were obtained for all ell lines. We previously reported a partial sequene for anine waf-1 (GenBank aession No. AF076469). 55 For this study, we extended the sequene by an additional 25 bases in the 5 (N-terminal) diretion and identified 32 potential polymorphisms, resulting in seven partially onserved nuleotide substitutions and six onserved nuleotide substitutions. Two primer pairs were used for waf-1 amplifiation (pair 1: sense 5 -GATGGA- Fig. 2. Aligned anine and human amino aid sequenes for p16 exon 2. Asterisks represent sequene identity. Conserved substitutions are represented by a dash under the sequene. A blank spae under the sequene represents nononserved substitutions. The regions of the human protein for whih the mathing anine sequene is unknown are highlighted in gray.

Vet Pathol 39:4, 2002 Tumor Suppressors in Canine Melanoma 461 ACTTGGACTTGGGC-3, antisense 5 -GAGTGGTA- GAAATCTTTAAGGCTGG-3 ; pair2:sense5 -GACTGT- GATGCGCTAATGGC-3, antisense 5 -GGGTACAAGA- CAGTGACAGG-3 ) to generate produts of 314 and 259 bases, respetively. For ink-4a, a primer pair (sense 5 - AGCTGCTGCTGCTCCACGG-3, antisense 5 -ACCA- GCGTGTCCAGGAAGCC-3 ) was designed to amplify 103 bp within the anine sequene for exon 2. The anine homologue of PTEN has been loned (GenBank aession No. U92435), and it is 95% idential to the human sequene. Two sets of primer pairs were designed from the published anine sequene (GenBank aession No. U92435). One primerpair(sense5 -ATGACAGCCATCATCAAGGAG-3 ; antisense 5 -CACACACAAATCACAAAAGTC-3 ) was designed to amplify the omplete PTEN oding domains (1,208 bp), and the other (sense 5 -GCTATGGGGTTTCCTG-3, antisense 5 -TAAGGACCAGAGACAAAAAG-3 ) amplified an internal produt of 393 bp that inluded the atalyti domains of the phosphatase. -Atin expression was used to ontrol for integrity of the RNA. The oligonuleotide primers used for amplifiation of -atin were 5 -ATGTTCGA- GACGTTCAACACCCC-3 (sense) and 5 -GCCA- TCTCCTGCTCGAAGTCCAG-3 (antisense), based on the GenBank sequene for Canis familiaris -atin (aession No. AF021873) to produe an amplifiation produt of 318 bp. Normalization of the levels of expression for eah tumor suppressor gene to the levels of ß-atin allowed for semiquantitative assessments of mrna expression. Immunoytohemistry and immunohistohemistry For ell lines, at the end of the ulture period plasti hambers were removed, and slides were rinsed in phosphate buffered saline (ph 7.4), fixed in aetone at 4 C for 5 minutes, and air dried. For arhival tissues, 5- m serial setions from paraffin-embedded bloks were mounted onto positively harged slides (Probe-on, Fisher Sientifi, Pittsburgh, PA). Immunostaining was performed using a modified streptavidin biotin omplex method previously desribed. 55,56 For paraffin-embedded setions, antigen retrieval proedures inluded mirowave heating for 6 minutes in a buffer of 0.1 M sodium itrate (ph 6) for p21, p16, Rb, and PTEN staining and mirowave heating for 6 minutes in Retrieve All (Signet Pathology Systems, Dedham, MA) for p53. Antigen retrieval was not used for staining of the ell lines, exept in the ase of p21, where mirowave heating for 6 minutes in sodium itrate buffer was used to unmask nuleolar staining. Slides were inubated for 1 hour at room temperature with primary antibodies against p53 (antibody CM-1, diluted 20-fold, Signet Pathology Systems), p21 (antibody C-20, 5 g/ml, Santa Cruz Biotehnology, Santa Cruz, CA), Rb (mouse monolonal antibody G3 245, 5 g/ml, Pharmingen, San Diego, CA), p16 (antibody H-156, 0.5 g/ml, Santa Cruz Biotehnology), or PTEN (antibody A2B1, diluted 100-fold, Santa Cruz Biotehnology). Seondary antibodies onsisted of goat anti-rabbit IgG or goat anti-mouse IgG onjugated to biotin (Kirkegaard & Perry Laboratories, Gaithersburg, MD). The presene of the relevant antigens was deteted using streptavidin onjugated to alkaline phosphatase. Following a rinse in Tris buffer, the olor reation was aomplished using the Histomark red kit (Kirkegaard & Perry Laboratories). Negative ontrols were prepared by using irrelevant isotype mathed antibodies in plae of the primary antibodies. Setions obtained from anine tissues (liver, kidney, brain, spleen, gut, skin) served as positive ontrols for the immunostains. To our knowledge, there are no previous reports of immunohistohemial staining for p16 and PTEN in fixed and embedded anine tissues. The onditions and speifiity for staining were optimized in one of our laboratories (J. Wojieszyn) based on those reported by Levine. 37 Expression of p16 in the normal tissues examined was most intense in hepatoytes, with loalization to the ytoplasm and prominent aumulation in Golgi areas. Moderately intense staining also was seen in ell bodies of the gray matter in the brain. Mononulear leukoytes showed faint ytoplasmi staining. Expression of PTEN in the normal tissues examined was prominent in vasular strutures, with the most intense staining in renal glomeruli. Cell ultures and ell lines were graded for intensity and perentage of positive ells. Samples were onsidered negative when no staining was seen above that seen in negative ontrols. Staining that was fine, diffuse, and not always visible at low magnifiation (40 100 ) was onsidered weak. Staining that was puntate or regional but prominent enough to be seen at low magnifiation was onsidered moderate. Staining that was diffuse, prominent, and easily visualized at low magnifiation was onsidered strong. In negative to weak ( to / ) samples, weak staining was visible in 10% of the ells in the tumor. Samples were only onsidered positive when expression of these proteins was learly detetable in melanoyti tumor ells (versus supporting stroma or inflammatory ells) based on the expression of S100a protein, Melan A, or neuron-speifi enolase in serial setions from the tumors. Immunoblotting. Whole ell lysates were made by disrupting ells in a high-salt buffer (300 mm sodium hloride, 50 mm Tris, ph 7.6, 0.5% Triton X-100, 1 mm N-ethylmaleimide, 2 g/ml aprotinin, and 1 g/ml leupeptin). Insoluble material was removed by entrifugation, and protein onentrations of the ell lysates were determined with a protein assay kit (BioRad, Herules, CA). Cellular proteins (20 g) were separated by sodium dodeyl sulfate polyarylamide gel eletrophoresis and transferred to nitroellulose membranes (Hybond, Amersham, Arlington Heights, IL) as previously desribed. 56 The same antibodies against p16 and PTEN used for immunoytohemistry were used for immunoblotting at a 1 : 1,000 dilution. The anti- -atin antibody (mouse monolonal, Sigma Chemial Co., St. Louis, MO) was used at a 1 : 5,000 dilution. Membranes were inubated with primary antibodies for 1 hour at room temperature, followed by seondary anti-rabbit antibody onjugated to horseradish peroxidase (Amersham). Detetion was performed using the enhaned hemiluminesene system (ECL, Amersham) aording to the manufaturer s instrutions. Statistis Statistial analyses were performed at the Biostatistis Core faility of the AMC Caner Researh Center and the University of Colorado Caner Center. Data were analyzed

462 Koenig, Biano, Fosmire, Wojieszyn, and Modiano Vet Pathol 39:4, 2002 Fig. 3. Expression of mrna for eah tumor suppressor gene (p53, waf-1, ink-4a, and PTEN) was examined by RT-PCR in the anine melanoma ell lines CML-2, CML- 6MC2, CML-13, JENNY, SCOOTER, SHADOW, and TLM- 1 using anine-speifi primers. The expression of ß-atin was used as a loading ontrol for the reations and to ensure integrity of the RNA. Normal anine liver was used to evaluate the expression of these genes in a nonneoplasti tissue sample. The predited sizes for the amplifiation produts were 738 bp for p53, 259 bp for waf-1, 103 bp for ink-4a, and 393 bp for PTEN. using the SAS statistial pakage. Correlation among the different variables was determined using Spearman rank orrelation analysis. Results Expression of p53 Alterations in p53 appear to be a late event in the progression of melanoma. Suh alterations an inlude deletion of the gene or mutations that silene the gene, disrupt the DNA binding domains, or alter the stability or subellular loalization of the protein. 26,36 However, many other proteins that interat with p53 affet its stability and subellular loalization, 26,36 thus, overexpression of p53 protein in tumor ells does not neessarily imply mutation or loss of the p53 gene. 10,26 To determine whether the melanoma ells harbored mutant p53 genes, we amplified p53 mrna by RT-PCR. Eah of the ell lines produed amplifiation produts of the predited size (Fig. 3), although the expression of the p53 gene appeared to be redued in JENNY ells. The sequenes enompassing amino aids 77 285 for eah of these produts were idential to the published wild type sequene for the anine p53 gene. The expression of p53 mrna suggested that p53 protein also would be present in these ells. We showed previously that immunoreative p53 protein was undetetable in ells from eah pigmented lesion analyzed in a dog with multiple melanoyti tumors, but p53 was present in normal ells in the tissue setions. 56 To ontinue this line of investigation, we used immunoytohemistry to examine aumulation of p53 protein in the anine melanoma ell lines (Figs. 4, 5). The CM-1 antibody reognizes the wild type as well as various mutant forms of p53 protein, and we previously verified the speifiity of this antibody against anine p53 by immunoblotting. 56 Eah of the melanoma ell lines showed ytoplasmi aumulation of p53, but none of the ell lines had p53 in the nuleus (Table 1, Fig. 4b). Staining in CML-2, CML-13, and SCOOTER ells was faint, diffuse, and uniform. In CML-6MC2 ells, the staining was perinulear, in JENNY ells it was perinulear with a polar to bipolar distribution, in SHADOW ells it was polar to diffuse, and in TLM-1 ells it was faint, perinulear, and polar. Loss or dramati redution of p53 expression (graded as negative or negative to weak) was seen in nine of 30 samples (five benign, three malignant, one of indeterminate behavior, Table 2). Three of these samples (Nos. 8, 10, 11) were from dogs with multiple, utaneous, benign melanoyti tumors, with loss or redution of p53 in every independent tumor examined. Exlusion of p53 from the nulear ompartment ourred in 18 of 25 samples with detetable p53 staining (eight benign, seven malignant, three of indeterminate behavior; Table 2). Figure 5b shows the p53 staining for sample No. 2, where the protein was shown to loalize to the ytoplasmi and nulear ompartment. The two samples (Nos. 4, 8) from dogs with multiple benign melanoyti tumors with detetable p53 expression that were part of this latter group similarly showed loalization of p53 exlusively in the ytoplasm in every independent tumor examined. Expression of p21 The waf-1 gene plays an important role in melanoyte growth and differentiation. 30,55 We previously doumented the loss of p21 expression in a dog with multiple dermal melanoyti tumors 56 and the requirement for loalization of p21 protein to the nulear ompartment for ontat-indued growth arrest (and possibly apoptosis). 55 Our previous work suggests that waf-1 expression is not highly regulated in ultured melanoma ells; thus, qualitative hanges of waf-1 expression would likely be signifiant. We used RT-PCR with two distint sets of primer pairs to amplify produts enompassing approximately 70% of the oding domains of waf-1, inluding the ylin and CDK-inter-

Vet Pathol 39:4, 2002 Tumor Suppressors in Canine Melanoma 463 Fig. 4. Immunoytohemial assessment of the expression and aumulation patterns for p53, p21, Rb, p16, and PTEN in anine melanoma ell lines. Fig. 4a. Negative ontrol staining using an irrelevant primary antibody. Fig. 4b. Staining for p53 in JENNY ells. Fig. 4. Staining for p21 in CML-2 ells. Fig. 4d. Staining for Rb in CML-6MC2 ells. Fig. 4e. Staining for p16 in CML-13 ells. Note the subtle aumulation of stain in the ytoplasmi Golgi regions indiated by the blak arrows. Fig. 4f. Staining for PTEN in TLM-1 ells. Bar 5 m. Fig. 5. Immunohistohemial assessment of the expression and aumulation patterns for p53, p21, Rb, p16, and PTEN in samples from spontaneous ases of anine melanoma. Fig. 5a. Negative ontrol staining using an irrelevant primary antibody. Fig. 5b. Staining for p53 in sample No. 2. Note the fine granular and foal staining in nulei indiated by the white arrowheads. Fig. 5. Staining for p21 in sample No. 21. Note the intense staining in individual nulei indiated by the white arrowheads. Fig. 5d. Staining for Rb sample No. 13. Fig. 5e. Staining for p16 in sample No. 7. Note the inreased intensity of staining in ytoplasmi Golgi regions indiated by the blak arrows. Fig. 5f. Staining for PTEN in sample No. 26. Staining oasionally extended to the nulei in few ells, not shown in this setion. Bar 5 m.

464 Koenig, Biano, Fosmire, Wojieszyn, and Modiano Vet Pathol 39:4, 2002 Table 1 Immunostaining results for the seven anine melanoma ell lines. Cell Line CML-2 CML-6MC2 CML-13 JENNY SCOOTER SHADOW TLM-1 I 2 2 p53 L I 3 3 3 2 2 p21 L /nl /nl /nl /nl /nl /nl /nl I 3 3 3 3 Markers* Rb L n n n n n n I ND 2 / / p16 L ND g nrim g *I intensity: negative; / negative to weak; weak, 2 weak to moderate; 3 moderate; 4 strong; ND not done. L loation: n nuleus; nl nuleoli; ytoplasm; g Golgi zone; n rim nulear rim. I ND 2 2 PTEN L ND ating domains. 55 Eah of the seven ell lines generated amplifiation produts of the predited size and sequene using the C-terminal primer pair, with dereased expression apparent in JENNY, SCOOTER, and SHADOW ells (data not shown). Amplifiation produts obtained using the N-terminal primer pair showed similarly dereased levels of waf-1 expression in the CML-6MC2, CML-13, JENNY, and SHADOW ell lines (Fig. 3). Sequening of the waf-1 amplifiation produts from CML-2, SHADOW, and TLM-1 ell lines revealed no mutations distint from the polymorphisms identified for the wild-type anine gene. The presene of waf-1 mrna in eah ell line suggested these ells would show aumulation of p21 protein. We used immunoytohemistry to examine aumulation of p21 protein in the anine melanoma ell lines. We previously verified the speifiity of the C-20 antibody against anine p21 by immunoblotting. 55,56 Eah of the seven ell lines examined showed aumulation of p21 that was predominantly ytoplasmi and nuleolar (Table 1, Fig. 4). The staining was intense and uniform in the CML-2, CML-13, and JEN- NY ell lines. The SHADOW and TLM-1 ell lines showed granular staining with oasional ells having foal staining in the perinulear or Golgi regions. The CML-6MC2 and SCOOTER ell lines stained faintly for ytoplasmi p21 but still had prominent nuleolar staining. The absene of nulear p21 was preditable in subonfluent ells in the logarithmi growth phase. 55 Loss or dramati redution of p21 expression (graded as negative or negative to weak) was seen in six of 30 samples (three benign, two malignant, one of indeterminate behavior; Table 2). Two of these samples (Nos. 8, 11) were from dogs with multiple, utaneous, benign melanoyti tumors, with loss of p21 in every independent tumor examined. Exlusion of p21 from the nulear ompartment ourred in nine of the 27 samples with detetable p21 staining (six benign, two malignant, one indeterminate, Table 2). One of the three samples from dogs with benign multiple melanoyti tumors with detetable p21 expression (sample No. 10) that was part of this latter group similarly showed loalization of p21 exlusively in the ytoplasm in every independent tumor examined. Figure 5 shows the p21 staining for sample No. 21, where the protein was loalized in the ytoplasmi and nulear ompartments. Expression of Rb Mutations of Rb do not appear to be prevalent in human melanoma, possibly beause mutations of other genes that partiipate in the same growth ontrol pathway our so often as to render Rb expression inonsequential. 8,51,54,59,60,75 The expression of Rb in anine melanoma has not been evaluated extensively. We used immunoytohemistry to evaluate the expression and subellular loalization of Rb in the anine melanoma ell lines. We previously verified the speifiity of the G3 245 antibody against anine Rb by immunoblotting. 47,55 The results show nulear aumulation of Rb in six of the seven ell lines (Table 1, Fig. 4d). Not all the ells from the CML-13 and SCOOTER lines showed Rb staining, although this inonsisteny ould be attributable to the fat that these ells were growing asynhronously. Alternatively, loss or redution of ink-4a in SCOOTER ells may have eliminated the onstraints of Rb expression and loalization. We showed previously that the predited phosphorylation state of Rb in TLM-1 ells orresponded with quiesene (hypophosphorylated) or proliferation (hyperphosphorylated), and overexpression of ylin-dependent kinase inhibitors p21 or p16 in TLM-1 ells and CML-2 ells resulted in hypophosphorylation of Rb with subsequent growth arrest and apoptosis. 47,55 Thus, nulear Rb in these melanoma ell lines probably represented the produt of a funtional wild-type gene. Rb staining in the SHADOW ell line was restrited to faint aumulation in the ytoplasmi and perinu-

Vet Pathol 39:4, 2002 Tumor Suppressors in Canine Melanoma 465 Table 2 Immunostaining results for the 31 arhival anine tumors. Sample No. Loation Malignant* p53 I L p21 I L Markers Rb I L p16 I L PTEN I L 1 Skin N 2 n/ 2 Skin of nek N 2 n/ 2 n/ /g 3 Skin of leg N 2 3 / / 4 Multientri, skin of abdomen N 2 2 n/ ND ND ND ND 5 Skin N 2 2 n/ / 6 Skin of ear N 2 3 n/ / n / 7 Skin N 2 n/ 4 n/ 2 n 2 /g 2 8 Multientri, skin of eyelid N / / n/ / n ND ND ND ND 9 Skin of srotum N / n 2 10 Multientri, skin of leg N 3 2 n/ 3 / 11 Multientri, oral (lip) N n / 12 Oral (lip) N n/ ND ND ND ND 13 Oral (lip) N n/ n/ 2 n ND ND ND ND 14 Digit N 2 2 2 15 Skin Y 3 n/ 2 n/ 2 n/ / 16 Skin of leg Y 4 n/ n / 17 Oral (gingiva) Y 4 n/ n/ 18 Oral (gingiva) Y / / 19 Lymph node met (oral) Y ND ND ND ND ND ND / n/ 20 Oral (tongue) Y 3 n/ n/ / 21 Oral (gingiva) Y 2 2 n/ 2 n/ / 22 Oral (gingiva) Y 2 23 Oral (gingiva) Y / 24 Lymph node met (digit) Y 2 3 n/ 2 / 25 Digit Y 2 n/ n/ / / 26 Skin NA / n/ 2 n/ / /g 2 n/ 27 Skin of eyelid NA 2 n/ n/ ND ND 28 Lumbar skin NA 4 n/ 4 n/ n 3 / 29 Oral (gingiva) NA 2 n/ n/ / 30 Oral (gingiva) NA 2 n/ 2 / /g 31 Digit NA 2 4 n/ *Y yes; N no; NA information not available. I intensity: negative; / negative to weak; weak; 2 weak to moderate; 3 moderate; 4 strong; ND not done. L loation: n nuleus; ytoplasm; g golgi zone. Multientri refers to the availability of samples from more than one tumor that exhibited similar staining. Sample Nos. 12 and 17 originated from the same dog. Sample Nos. 13 and 18 originated from the same dog.

466 Koenig, Biano, Fosmire, Wojieszyn, and Modiano Vet Pathol 39:4, 2002 lear ompartment of few ells, suggesting that this ell line might harbor a mutant Rb gene produt (Table 1). Loss or dramati redution of Rb expression (graded as negative or negative to weak) was seen in nine of 30 samples (four benign, three malignant, two of indeterminate behavior; Table 2). Two of these samples (Nos. 4, 8) were from dogs with multiple, utaneous, benign melanoyti tumors, with loss of Rb in every independent tumor examined. Exlusion of Rb from the nulear ompartment ourred in four of the 24 samples with any detetable Rb staining (three benign, one of indeterminate behavior; Table 2) but was not seen in any of the samples from dogs with multiple benign melanoyti tumors. Figure 5d shows the loalization of Rb staining restrited to the nulear ompartment in sample No. 13. Expression of ink-4a Fig. 6. Aumulation of p16 and PTEN tumor suppressor proteins in anine melanoma ell lines CML-2, CML- 6MC2, CML-13, JENNY, SCOOTER, SHADOW, and TLM- 1 was examined by immunoblotting. Analysis of -atin was used as a loading ontrol. Normal anine Cf2Th ells were used to evaluate the expression of these proteins in a nonneoplasti tissue sample. The sizes for the proteins are 16 kd for p16, 55 kd for PTEN, and 42 kd for -atin. Mutation of ink-4a is one of the most frequent geneti abnormalities haraterized in human tumors 59 and is a feature of 50% of all human melanomas. 16,27,48,50 The importane of this gene in tumorigenesis was onfirmed in mie with a targeted deletion of the INK-4A lous that inludes the ink-4a gene, where the addition of a onditionally ative ras onogene leads to the development of spontaneous melanoma. 7 Three of the seven ell lines (CML-2, CML-6MC2, and JENNY) laked ink-4a mrna, as determined by RT-PCR (Fig. 3). A more quikly migrating (lower) band that orresponded to primer dimers (onfirmed by sequening) was seen in some samples, most often those that had redued or absent ink-4a expression (see lanes for CML-6MC2, SCOOTER, and SHADOW ells, Fig. 3). The levels of expression of the predited ink-4a amplifiation produt in SHADOW ells were slightly higher than those in the liver. An amplifiation produt of the predited size was obtained from the CML-13, SCOOTER, and TLM-1 ell lines (Fig. 3). Immunoblotting showed that a 16 kd protein that was reognized by the anti-p16 antibody and was indistinguishable from the protein found in normal Cf2Th ells was abundant in CML13 ells and was expressed at muh lower levels in JENNY, SCOOTER, SHADOW, and TLM-1 ells (Fig. 6). No p16 was detetable in CML2 or CML-6MC2 ells. Analysis of p16 protein expression by immunoytohemistry similarly showed moderate expression in CML-13 ells (omparable to that seen in Cf2Th ells) and weak staining in TLM-1 ells (Table 1, Fig. 4e). The staining was predominantly ytoplasmi with prominent aumulation in the Golgi regions. Less than 5% of SCOOTER and SHADOW ells had faint staining loalized to the nulear rim or to ytoplasmi granules, respetively. CML-2 and JENNY ells had no detetable p16. Immunoytohemial staining for p16 protein was not performed in CML-6MC2 ells. Loss or dramati redution of p16 expression (graded as negative or negative to weak) was seen in 21 of 26 samples (six benign, 11 malignant, four of indeterminate behavior; Table 2). When present, staining of p16 was generally onfined to the ytoplasm, although intense staining in the Golgi area was seen in four ases (two benign and two indeterminate). The pattern of p16 aumulation is illustrated by the immunostaining for sample No. 7 (Fig. 5e). Expression of PTEN The importane of the PTEN gene in the pathogenesis of multiple aners has reently been a topi of intense interest. 3,4,40,43,66 PTEN is an antagonist of the phosphoinositol-3 kinase (PI3K). 20 PI3K ativation initiates a asade of events that promote growth fator prodution, ell division, and survival (resistane to apoptosis). 18,19 Thus, loss of PTEN an establish an autorine growth loop where there is inreased prodution of growth fators and where ells show exaggerated proliferation in response to these growth fators. We examined PTEN gene expression in the anine melanoma ell lines by RT-PCR. Three of the ell lines (CML-2, CML-13, JENNY) laked PTEN mrna when the internal primer pair was used for amplifiation, and DNA from SHADOW ells generated multiple amplifiation produts that likely represent mutant forms of the protein (Fig. 3), although in some experiments, no amplifiation produts for PTEN mrna were generated from this ell line.

Vet Pathol 39:4, 2002 Tumor Suppressors in Canine Melanoma 467 SHADOW ells appear to be hypodiploid based on DNA ontent analysis (data not shown), suggesting the line may be genetially unstable, and is thus likely to harbor numerous mutations. Eah of these ell lines, exept JENNY, generated amplifiation produts using the primer pair spanning the omplete oding domains (data not shown). The CML-6MC2, SCOOTER, and TLM-1 ell lines produed amplifiation produts of the predited size with eah primer pair, and sequening revealed no disernible mutations in the 393 bp amplifiation produts generated with the internal primer pair. Consistent with the results observed upon examination of gene expression, immunoblotting showed the presene of a 55 kd protein that was reognized by the anti-pten antibody in SCOOTER and TLM-1 ells that was indistinguishable from the protein found in normal Cf2Th ells (Fig. 6). A PTENimmunoreative protein also was present in CML- 6MC2 ells and had a slightly slower eletrophoreti mobility than that seen in the other positive ell lines. PTEN protein was not detetable in CML-2, CML-13, JENNY, or SHADOW ells. Analysis of PTEN protein expression by immunoytohemistry showed moderate, diffuse ytoplasmi staining in TLM-1 ells and SCOOTER ells (Table 1, Fig. 4f). No PTEN staining was evident in CML-2, CML13, JENNY, and SHAD- OW ells. Immunoytohemial staining for PTEN protein was not performed in CML-6MC2 ells. Loss or dramati redution of PTEN expression (graded as negative or negative to weak) was seen in 16 of 27 samples (five benign, 11 malignant, five of indeterminate behavior; Table 2). Although PTEN is a ytoplasmi protein, nulear loalization was seen in two samples (one malignant and one of indeterminate behavior; Table 2). Figure 5f shows the staining for PTEN in sample No. 26. Disussion Melanomas are malignant tumors of melanoytes or melanoblasts. Cultured ells usually resemble the morphology of the parent tumor from whih they were derived; 15,33,49,56 however, ultured melanoma ells are not the same as tumors in vivo. Theoretially, ell lines an mutate and may not possess all the same harateristis of the parent tumor. Variation in environmental onditions may also influene growth and differentiation. Nevertheless, ultured melanomas are a useful preliminary model to study the origin and progression of melanoma. Combined with analysis of primary arhival setions, this approah provides a robust model to define the frequeny with whih the expression of partiular genes or proteins may be lost (or amplified) in these tumors and to investigate possible biologial onsequenes of these abnormalities. In this study, we used RT-PCR and immunoytohemistry to examine tumor suppressor gene and protein expression in seven ultured melanoma ell lines and 31 arhival samples of melanoma from 29 affeted dogs. The gene produts we hose to examine, p53, p21, Rb, p16, and PTEN, have eah been impliated to some extent in the origin or progression of melanoma in humans or other animal models. 1,11,22,28,46,55,56,58,59,70,71 We speifially sought to investigate whether the pathways of growth ontrol, survival, and DNA integrity in whih these proteins partiipate might ontribute to the genesis of anine melanoma. Our working hypotheses were 1) mutations in these pathways are ommon in anine melanoma and 2) mutations in more than one pathway predit a worse outome. The results presented here are onsistent with the first hypothesis. Our results also show that these geneti lesions are ommon in both the benign and malignant forms of anine melanoma, but additional work is neessary to doument their prognosti signifiane. We previously proposed a predisposition of individual dogs to develop multiple melanoyti tumors as a result of a mutation in a melanoyte preursor, based on a shared geneti harateristi in the tumors. 56 This study inludes data from the same dog (sample No. 11) and three additional dogs with multiple, utaneous, benign melanoyti tumors. Two additional dogs were inluded that developed two distint tumors (sample Nos. 12, 17; sample Nos. 13, 18), but in both of these dogs, one of the tumors was benign and one was malignant. Three of these dogs were Gordon Setters, and three of the four Gordon Setters in the study developed more than one melanoma in their lifetime. The results from this study onfirm the presene of a shared phenotype in eah of the tumors from dogs with multiple, utaneous, benign melanoyti tumors, supporting the premise that these tumors may have arisen from a ommon preursor. However, suh shared phenotype was not seen in the dogs that developed more than one tumor with different biologial behavior (i.e., one benign and one malignant), suggesting that different fators may underlie the pathogenesis of these two apparently related tumors. The observation that abnormalities indiative of loss of p16/rb funtion (inluding redued or absent expression of either gene or exlusion of Rb from the nulear ompartment) ourred most frequently in both the ell lines (six of seven) and the linial samples (22 of 26) suggests that inativation of this pathway is a ritial step in the pathogenesis of benign and malignant melanoma in dogs, as it is in humans and other animals. 48,59,61,72 Mutation of ink-4a is a feature of 50% of all human melanomas. 16,27,48,50 In the CML-2 and CML-6MC2 ell lines where gene expression and protein aumulation were undetetable,

468 Koenig, Biano, Fosmire, Wojieszyn, and Modiano Vet Pathol 39:4, 2002 loss of ink-4a expression might have ourred through homozygous deletion of the gene, through mutations of nonoding transriptional regulatory sequenes, or through methylation. 14,48,59,72 Our results suggest that JENNY, SCOOTER, and SHADOW ells may harbor mutations in the ink-4a gene that generate unstable proteins. This instability in SCOOTER and SHADOW ells might be related to rapid degradation due to loalization of the protein to abnormal subellular ompartments. In JENNY ells, the mutation is likely to involve the ankyrin domains in exon 2; no mrna was detetable using primers targeting this region. The p16 protein in TLM-1 ells is also expressed at very low levels, raising the possibility of hemizygous deletion of one ink-4a allele or of a mutation that also produes an unstable protein in these ells. The results from our immunoblotting experiments indiate that it is unlikely that the RT-PCR led to amplifiation of other ink-4 gene family members (i.e., ink-4b, ink-4, or ink-4d). Although the sequene of p16 exon 2 targeted for amplifiation is highly homologous to that of p15 (ink- 4b) from other speies (the sequene for anine ink- 4b is unknown) and the anti-p16 antibody used is reported by the manufaturer to have some degree of ross-reativity with p15, a protein with faster eletrophoreti mobility (smaller moleular weight) was not seen in any ell line. The results from the immunoblotting experiments similarly suggest that it is unlikely that the amplifiation produts from the RT-PCR may represent the ARF (alternative reading frame) gene, whih is enoded in the same lous as ink-4a. 53 Mutations of Rb do not appear to be prevalent in human melanoma, 39 possibly beause mutations of ink- 4a our so often as to render Rb expression inonsequential. 8,51,54,59,60,75 Normal human melanoytes (and other quiesent ells suh as peripheral blood lymphoytes) show little to no detetable Rb by immunostaining, 45,58 but Rb was present in all primary and metastati melanomas in 5 70% of ells and was always loalized to the nuleus. Loss or mutational inativation of both normal Rb alleles leads to loss of Rb expression and an ontribute to the development of malignant melanomas. 58 Exlusion of Rb from the nulear ompartment suh as that seen in the SHADOW ell line and sample Nos. 2, 3, 14, and 30 may represent a novel mehanism of Rb inativation in melanoma and may reflet a mutation of Rb or possibly other genes that ontrol nulear transloation of Rb. Exlusion of p53 from the nulear ompartment ourred almost as frequently as loss of p16/rb. The ytoplasmi aumulation of p53 in the ell lines and tumor samples was not a feature of the antibody or the onditions used or an artifat of tissue ulture; p53 was loalized almost exlusively in the nulear ompartment of Cf2Th ells, 55 and seven tumor samples ontained abundant nulear p53. In human melanomas, mutations of the p53 gene or abnormalities that result in aumulation of p53 protein that is funtionally exluded from the nuleus our frequently, 9,28,31,57,70,71 and melanomas with mutant p53 gene arry a worse prognosis than those without p53 mutations. 17,34 However, numerous events may affet the stability and subellular loalization of p53, 36 and overexpression of p53 protein in tumor ells does not neessarily imply mutation or loss of funtion of the p53 gene. 10 The sequenes for amino aids 77 285 in the p53 mrnas obtained from the ell lines in this study were indistinguishable from the wild type sequene, suggesting that the p53 proteins in these ells were ompetent to bind DNA in a sequene-speifi manner and to undergo homodimerization. More than 90% of point mutations assoiated with loss of funtion of p53 our within the DNA binding and homodimerization domains between amino aids 102 and 292. 36 However, we annot exlude the possibility that the ell lines had mutations of p53 that affeted the mdm-2 interating domain loated between amino aids 13 and 29 or the nulear loalization signal loated near amino aids 305 and 306. 26,36 These domains are responsible, respetively, for targeting p53 for degradation in the proteasome and for allowing transport of p53 to the nuleus. The inativation of p16 in many tumors that ontained ytoplasmi p53 is somewhat paradoxial beause the ARF gene enoded in the same lous stabilizes p53 by targeting mdm-2 for destrution. 41,52,78 Thus, p16, but not ARF, had to be seletively inativated in these tumors (e.g., by methylation), unless mutations of p53 in the mdm-2 binding domain are more ommon in dogs than they are in humans. The overexpression of ytoplasmi p53 also ould reflet a measure of geneti instability (with rapid turnover of the protein in the nulear ompartment) or mutations in other genes that ontrol p53 synthesis and degradation. Loss or redution of PTEN expression also was seen frequently in the ell lines but less so in the tumor samples. The ell line that did not exhibit abnormalities of p16 or Rb (CML-13) showed loss of PTEN expression, suggesting that enhaned survival or growth mediated through growth fator pathways that use PI3K may help to overome the effets of normal p16/rb pathways in anine melanoma. Three ell lines (CML-2, JENNY, and SHADOW) showed onurrent abnormalities of p16 or Rb and PTEN, suggesting that onurrent loss (or redution) of both the p16/rb and PTEN pathways may failitate progression of the disease. At least two mehanisms may aount for loss of PTEN expression. Of the four ell lines that showed mutations of the PTEN gene, three might harbor deletions, insertions, or single nuleotide substitutions

Vet Pathol 39:4, 2002 Tumor Suppressors in Canine Melanoma 469 that preluded amplifiation of PTEN mrna using the internal primer pair, whereas the loss of PTEN gene expression in the other may have been due to aneuploidy or geneti instability. Beause amplifiation produts were obtained from eah ell line using the primer pair designed to amplify the omplete oding domains, it is possible that the internal primer pair may span a hot spot and that mutations in this region destabilize the protein and aount for the lak of PTEN immunostaining in these ells. However, this is not true for every ell line, and additional experiments by us and others suggest that ultured anine tumor ells an express mutant PTEN genes that an be deteted by RT-PCR or immunohistohemistry (E. Dikerson, S. Fosmire, S. Biano, J. Wojieszyn, J. Modiano, and S. Helfand, unpublished results). Therefore, beause mutations that inativate PTEN need not neessarily hinder its expression, immunohistohemial analysis of fresh or arhival tissues may underestimate the frequeny of suh mutations in anine melanoma and other tumors. An important role for p21 in melanoyte growth and differentiation has been doumented. 30,32,44,47,55 The p21 protein is a CDKI that has nonoverlapping funtions with p16, but it also partiipates in ontat-indued growth arrest and ontrols the funtion of stress-ativated protein kinases, polymerases involved in DNA synthesis, proteins that regulate survival and apoptosis, and possibly ell division (ytokinesis). 2,6,23,24,64,67,74 There was no lear orrelation between the presene of p21 protein and the morphology or growth properties of the ell lines. One mehanism through whih p21 expression an be upregulated is through transriptional ativation mediated by p53, 13 and p21 appears to be important in p53-dependent growth arrest in ells that must undergo DNA repair. 67 However, p21 expression an also be indued through pathways independent of p53. 42,76,77 We saw no apparent orrelation between p21 and p53 in the seven ell lines. Loss or redution of p21 was also an infrequent ourrene in the linial samples, but exlusion of p21 from the nulear ompartment was disproportionately seen in utaneous melanomas, suggesting that loss of funtion of this protein may play a role in this form of the disease. In this group of samples, anatomi loation remained the best preditor for tumor behavior. Multiple abnormalities in these proteins were not signifiantly more ommon in malignant lesions than in benign lesions. This lak of signifiane may be due to the limited statistial power afforded by the relatively small sample size. Alternatively, its ould signify that the funtion of these genes is important to prevent malignant transformation of melanoytes, but one transformation takes plae their loss of funtion does not signifiantly impat disease progression. Nevertheless, the results showed distint patterns with shared abnormalities. For example, the absene of funtional p53 was signifiantly orrelated with absene of nulear p21. Similarly, loss or redution of p16 was orrelated with abnormalities of Rb. These results were preditable beause p53 is an important regulator of p21 expression and beause abnormalities in a single member of the p16/rb pathway render other members of the pathway inonsequential. The results from this study show that abnormalities of tumor suppressor genes and proteins are ommon in anine melanomas. These abnormalities likely reflet mutations that may ontribute to the origin of the tumors in vivo. The possibility that malignant tumors aumulate mutations in multiple tumor suppressor gene pathways at an aelerated rate that may influene prognosis merits further investigation in spontaneous anine tumors. Aknowledgements We thank Dr. Steve Dow, Dr. Lauren Wolfe, Dr. Elaine Ostrander, and Dr. David Beah for providing ell lines and reagents, Josy Mayor and Dr. Vijayanagaram Venkatraj for assistane loning anine ink-4a, Dr. Gary Cutter for assistane with statistial analyses, Dr. Roy Levine for information on amplifiation of p53 by RT-PCR and on the performane of antibodies that reognize anine p16 and PTEN, Dr. Erin Dikerson, Dr. Mihelle Ritt, Dr. Kenita Rogers, and Dr. Jennifer Thomas for helpful suggestions, and Dr. Barb Powers for ritial review of the manusript. This work was supported in part by grant 1626 from the AKC Canine Health Foundation and by grants 98PT-16 and 98CA-34 from the Morris Animal Foundation. S. R. Biano was supported in part by a fellowship from the University of Colorado Caner Center. Referenes 1 Albino AP, Vidal MJ, MNutt NS, Shea CR, Prieto VG, Nanus DM, Palmer JM, Hayward NK: Mutation and expression of the p53 gene in human malignant melanoma. Melanoma Res 4:35 45, 1994 2 Barboule N, Chadebeh P, Baldin V, Vidal S, Valette A: Involvement of p21 in mitoti exit after palitaxel treatment in MCF-7 breast adenoarinoma ell line. Onogene 15:2867 2875, 1997 3 Besson A, Robbins SM, Yong VW: PTEN/MMAC1/ TEP1 in signal transdution and tumorigenesis. Eur J Biohem 263:605 611, 1999 4. Cantley LC, Neel BG: New insights into tumor suppression: PTEN suppresses tumor formation by restraining the phosphoinositide 3-kinase/AKT pathway. Pro Natl Aad Si USA 96:4240 4245, 1999 5 Chang HW, Aoki M, Fruman D, Auger KR, Bellaosa A, Tsihlis PN, Cantley LC, Roberts TM, Vogt PK: Transformation of hiken ells by the gene enoding the atalyti subunit of PI 3-kinase. Siene 276:1848 1850, 1997