Papillomaviral DNA and increased p16CDKN2A protein are frequently present within feline cutaneous squamous cell carcinomas in ultraviolet-protected skin

Squamous cell carcinomas (SCCs) are common feline skin tumours. While exposure to ultraviolet (UV) light causes some SCCs, a subset develop in UV-protected skin. In cats, papillomaviruses (PVs) cause viral plaques and Bowenoid in situ carcinomas (BISCs). As both may progress to SCC, it was hypothesized that SCCs in UV-protected skin may represent neoplastic transformation of a PV-induced lesion. To investigate this hypothesis, PCR was used to amplify PV DNA from 25 UV-protected and 45 UV-exposed SCCs. Oncogenic human PVs cause neoplasia by mechanisms that also increase p16(CDKN2A) protein (p16). As increased p16 is present in feline viral plaques and BISCs, immunohistochemistry was used to detect p16 within the SCCs. Papillomaviral DNA was amplified from 76% of UV-protected SCCs, but only 42% of UV-exposed SCCs. Increased p16 was present in 84% of UV-protected SCCs, but only 40% of UV-exposed SCCs. The more frequent detection of PV DNA and increased p16 within UV-protected SCCs supports the hypothesis that some develop from a PV-induced plaque or BISC. Felis domesticus PV-2 is thought to cause viral plaques and BISCs. This PV was detected most frequently within the UV-protected SCCs, supporting development from a PV-induced lesion. Increased p16 and PV DNA were less frequent within UV-exposed SCCs, presumably because these developed from actinic keratosis rather than a PV-induced lesion. The results support the hypothesis that some feline cutaneous SCCs are caused by PV infection and suggest that PVs may cause neoplasia by mechanisms that also increase p16.     

Amplification of papillomaviral DNA sequences from a high proportion of feline cutaneous in situ and invasive squamous cell carcinomas using a nested polymerase chain reaction

Squamous cell carcinomas (SCCs) are common skin tumours of cats. Previous studies have suggested that papillomaviral (PV) DNA is detectible within some feline SCCs. A PV DNA sequence has been previously amplified from five feline bowenoid in situ carcinomas (BISCs). Primers specific for this sequence were used in a nested polymerase chain reaction to compare PV detection rates in SCCs to rates within non-SCC skin lesions. Papillomaviral DNA was amplified from 20 of 20 BISC, 17 of 20 invasive SCC and 3 of 17 non-SCC controls. The rate of PV amplification from feline cutaneous SCCs was significantly higher than from non-SCC lesions. These results confirm that feline cutaneous SCCs are associated with PV infection. In humans, there is evidence that PVs promote SCC development within sun-exposed skin. The demonstrated association between PVs and feline cutaneous SCCs suggests, but does not prove, that PVs may also promote feline SCC development. If PVs are oncogenic in cats, prevention of PV infection may reduce feline cutaneous SCC development. To the authors' knowledge, this is the first time that PV DNA has been amplified from a non-SCC sample of feline skin.     

Evaluation of feline oral squamous cell carcinomas for p16 protein immunoreactivity and the presence of papillomaviral DNA

Oral squamous cell carcinomas (OSCCs) develop commonly in cats. While the cause of the feline neoplasms is unknown, a quarter of human OSCCs are caused by papillomavirus (PV) infection. As PV DNA has been previously detected in a feline OSCC, it was hypothesised that PV infection could be a significant cause of feline OSCCs. Human OSCCs that are caused by PVs contain increased p16^CDKN2A protein (p16), which can be detected using immunohistochemistry. In cats, increased p16 immunoreactivity has been reported within PV-associated skin lesions. This study evaluated p16 immunoreactivity within 30 feline OSCCs. Additionally, PCR was used to amplify PV DNA from the OSCCs. Increased p16 immunoreactivity was present within 2 OSCCs. However, as PV DNA was not amplified from any OSCC in this study, it cannot be confirmed that the increased p16 was caused by PV infection. Therefore, these results do not support the hypothesis that PVs are a significant cause of OSCCs in cats. Loss of p16 expression is considered an important process in the development of human non-PV-induced OSCCs. In contrast, loss of p16 immunoreactivity was only present in 2 feline OSCCs. This suggests that human and feline OSCCs develop due to different molecular mechanisms.     

Detection of novel papillomaviruses in canine mucosal, cutaneous and in situ squamous cell carcinomas

Papillomavirus (PV) DNA is frequently uncovered in samples of human skin squamous cell carcinomas (SCC). However, the role of these viruses in the development of such cancers in canine species remains controversial. While approximately 100 human PVs are known, only one single canine oral PV (COPV) has been identified and studied extensively. Therefore, we applied a narrow-range polymerase chain reaction (PCR) suitable for the detection of classical canine and feline PVs, as well as a broad-range PCR, which has been used for the detection of various novel PVs in humans, in order to analyse 42 paraffin-embedded samples, representing three different forms of canine SCCs. Ten samples of skin tissues with various non-neoplastic conditions served as controls. While none of the negative controls reacted positively, PV DNA was discovered in 21% of the tested SCC samples. Interestingly, the classical COPV was amplified from only one sample, while the other positive cases were associated with a variety of thus far unknown PVs. This study suggests that a fraction of canine SCC is infected with PVs and that a genetic variety of canine PVs exists. Therefore, these results will facilitate the future study of the role of PVs in the development of canine skin cancers.     

Infrequent detection of papillomaviral DNA within canine cutaneous squamous cell carcinomas, haemangiosarcomas and healthy skin on the ventrum of dogs

Background – Canine squamous cell carcinomas (SCCs) most frequently develop on the ventral abdomen and are thought to be caused by ultraviolet (UV) light. Papillomaviruses (PVs) have been associated with cutaneous SCCs in multiple species, including dogs. Hypothesis – That PVs act as cofactors in canine UV‐induced SCCs. Animals – The study was performed on skin from the ventrum of 60 dogs. These samples included 20 SCCs, 20 haemangiosarcomas and 20 samples of clinically normal skin. Two canine viral plaques were included as positive controls for PV. Methods – PCR was used to amplify PV DNA from all samples. Primers used included two sets of consensus primers and two sets of primers that were designed specifically to amplify PV DNA sequences detected in the viral plaques. Results – The MY09/11 consensus primers amplified PV DNA from both viral plaques. One plaque contained a DNA sequence (CfPV‐JM) that had been previously reported from a dog with multiple cutaneous SCCs. The other plaque contained a previously unreported PV DNA sequence. No PV DNA was amplified by either consensus primer from any of the ventrum skin samples. Primers designed specifically to amplify the CfPV‐JM sequence amplified DNA from one SCC, but no other sample. No PV DNA was amplified using the other specific PCR primer set. Conclusions and clinical importance – These results do not support a significant role for PVs in SCC development from the ventrum of dogs. However, they contribute another PV sequence to the list of PVs that have been associated with viral plaque development in dogs.     

Detection of human papillomavirus DNA in feline premalignant and invasive squamous cell carcinoma

Squamous cell carcinoma (SCC) is the most common malignant cutaneous and oral neoplasm of cats. Papillomavirus (PV) DNA has been identified in a proportion of feline Bowenoid in situ carcinomas (BISCs), cutaneous SCCs and a single oral SCC, but its exact role in the pathogenesis remains unknown. In humans, it has been suggested that ultraviolet (UV) light and human PV (HPV) may act as cofactors in cutaneous SCC carcinogenesis. Little is known about the influence of UV light on PV prevalence in feline cutaneous lesions, including actinic keratosis (AK). Additionally, PV prevalence in noncutaneous feline lesions, including oral SCC, is largely not known. This study aimed to determine the presence of PV in 84 cats with premalignant and invasive SCC from cutaneous and noncutaneous sites using polymerase chain reaction and to investigate an association with UV light. Papillomaviral DNA was amplified from two of 12 cases of AK, seven of 22 BISCs, nine of 39 cutaneous SCCs and two of 35 non‐cutaneous SCCs. Of the PV DNA sequenced, 50% was most similar to HPV of the genus Betapapillomavirus, while the other 50% was most similar to Felis domesticus PV type 2. Exposure to UV was not associated with an increase in PV for cutaneous SCC. The results of this study suggest that in the cat, HPV DNA may be detectible within a higher percentage of squamous lesions than previously demonstrated, UV exposure may not be a confounder for PV presence, and noncutaneous lesions may have a low prevalence of PV.     

Amplification of three different papillomaviral DNA sequences from a cat with viral plaques

A 3‐year‐old cat from New Zealand developed three small raised non‐ulcerated plaques on the face. Serology detected antibodies against feline immunodeficiency virus (FIV). Histology of the plaque revealed epidermal hyperplasia with keratinocytes either distended with large blue‐grey cytoplasmic bodies or with shrunken nuclei surrounded by a clear halo. Papillomavirus (PV) antigen was detected immunohistochemically and feline viral plaque was diagnosed. Swabs were taken of both lesional and non‐lesional skin, and polymerase chain reactions were used to detect PV DNA. Three different PV DNA sequences were amplified, one from a Felis domesticus PV type 1 (FdPV‐1) previously amplified from a feline viral plaque, a second (FdPV‐JM) previously amplified from feline cutaneous squamous cell carcinomas, and a third FdPV‐MY that was not reported previously. All three sequences were amplified from swabs of both lesional and non‐lesional skin. These results extend the geographical range of FdPV‐1 outside North America and also demonstrate the ability of FdPV‐1 to asymptomatically infect feline skin. However, the detection of multiple PV sequences within both lesional and non‐lesional samples makes it difficult to determine whether or not any of the PVs caused feline viral plaque development in this cat. This is the first time PV DNA has been detected in a feline skin swab sample. Additionally, it is the first report of multiple PVs being detected in a single sample from a cat. This may suggest that FIV infection predisposes cats to cutaneous PV infection.     

Detection of papillomaviral DNA sequences in a feline oral squamous cell carcinoma

Oral squamous cell carcinomas (OSCCs) are common and often fatal feline neoplasms. Factors that predispose to neoplasm development in cats are poorly defined. Around 25% of human OSCCs are caused by papillomaviruses (PVs). To determine if PVs are associated with OSCCs in cats, three sets of consensus primers were used to evaluate 20 feline OSCCs and 20 non-neoplastic feline oral lesions for the presence of PV DNA. Papillomaviral sequences were detected within one OSCC, but no non-neoplastic lesion. Sequencing of the amplified DNA revealed a previously unreported PV that was most similar to human PV type 76. This is the first time PV DNA has been amplified from the oral cavity of a cat. However, while these results suggest that feline gingival epithelial cells can be infected by PVs, they do not support a causal association between viral infection and the development of feline OSCCs.     

Frequent detection of papillomavirus DNA in clinically normal skin of cats infected and noninfected with feline immunodeficiency virus

Feline cutaneous squamous cell carcinomas (SCCs) often contain felis domesticus papillomavirus type 2 (FdPV‐2) DNA. While this may suggest FdPV‐2 causes feline SCC development, the proportion of cats that are asymptomatically infected by this PV is unknown. Infection by feline immunodeficiency virus (FIV) is associated with high rates of cutaneous SCC development, possibly due to increased PV infection. This study examines the frequency of cutaneous asymptomatic FdPV‐2 infections in cats and compares the rate of FdPV‐2 infection in 22 FIV‐positive cats with that in 22 FIV‐negative cats. FdPV‐2 sequences were detected in 39% of skin swabs. One or both swabs contained FdPV‐2 DNA from 52% of the cats. FIV status, age or sex of the cat did not significantly influence FdPV‐2 infection. Cats that shared a household with a PV‐infected cat could remain uninfected suggesting infection depends more on host factors than exposure to the PV. These results indicate that asymptomatic FdPV‐2 infections are common in cats, but do not provide evidence that FdPV‐2 causes feline SCC development.     

Detection of novel papillomaviruslike sequences in paraffin-embedded specimens of invasive and in situ squamous cell carcinomas from cats

OBJECTIVE: To detect and partially characterize papillomavirus (PV) DNA in squamous cell carcinoma (SCC) tumor specimens from cats. SAMPLE POPULATION: 54 formalin-fixed paraffinembedded skin biopsy specimens were examined. Specimens originated from Bowenoid in situ SCC (BISC; n = 21), invasive SCC (22), and skin affected by miscellaneous nonneoplastic conditions (11). PROCEDURES: Samples from each tissue block underwent DNA extraction after deparaffinization, and PCR assays were performed. Two sets of primers derived from PV E1 were used. The first set of primers was designed for the narrow-range PCR assay and was able to generate amplification products of feline PV (FePV), canine oral PV, or closely related PVs. The second set of primers was selected for the broad-range PCR assay because of its ability to amplify DNA from 64 human PVs. Sequence analysis of each amplified DNA was performed. RESULTS: 1 of the 21 specimens of BISC was positive for PV DNA on the basis of narrow-range PCR assay results, whereas all the other specimens (BISC, invasive SCC, and controls) had negative results for PV DNA. In contrast, 5 of 21 BISC specimens and 4 of 22 invasive SCC specimens were positive for PV DNA on the basis of broad-range PCR assay results. Sequence analysis revealed that only 1 specimen was infected by a virus closely related to classic FePV. In the 8 other specimens positive for PV DNA, DNA of unknown PVs was uncovered. CONCLUSIONS AND CLINICAL RELEVANCE: Bowenoid in situ SCC and invasive SCC of cats may be associated with PVs of genetic diversity.     

Immunohistochemical analysis of cyclin A, cyclin D1 and P53 in mammary tumors, squamous cell carcinomas and basal cell tumors of dogs and cats

The involvement of cyclin A, cyclin D1 and p53 proteins in canine and feline tumorigenesis was analyzed immunohistochemically. In the present study, a total of 176 cases were examined, among which there were 108 canine cases (75 mammary lesions, 16 squamous cell carcinomas and 17 basal cell tumors) and 68 feline cases (43 mammary lesions, 20 squamous cell carcinomas and 5 basal cell tumors). Speckled nuclear staining for cyclin A was observed in 19/38 (50%) canine malignant mammary tumors and 18/37 (48.6%) feline mammary carcinomas, while this was not seen in benign mammary tumors of either dogs or cats. Marked intense nuclear cyclin A staining was seen in 7/16 (43.8%) canine squamous cell carcinomas and 18/20 (90.0%) feline squamous cell carcinomas. Only 3/17 (17.6%) canine basal cell tumors showed slight and scattered staining for cyclin A. Expression of cyclin D1 was very rare in both canine and feline tumors. Nuclear staining of p53 was found in 7/37 (18.9%) feline mammary carcinomas. Intense immunoreactivity for p53 was found in 6/16 (37.5%) canine squamous cell carcinomas and 8/20 (40%) feline squamous cell carcinomas. These results suggest that cyclin A may have a role in the proliferation of canine malignant mammary tumors, feline mammary carcinomas and squamous cell carcinomas of dogs and cats, and p53 may associate with the tumorigenesis of feline mammary carcinomas and squamous cell carcinomas of dogs and cats.     

Feline papillomas and papillomaviruses

Papillomaviruses (PVs) are highly species- and site-specific pathogens of stratified squamous epithelium. Although PV infections in the various Felidae are rarely reported, we identified productive infections in six cat species. PV-induced proliferative skin or mucous membrane lesions were confirmed by immunohistochemical screening for papillomavirus-specific capsid antigens. Seven monoclonal antibodies, each of which reacts with an immunodominant antigenic determinant of the bovine papillomavirus L1 gene product, revealed that feline PV capsid epitopes were conserved to various degrees. This battery of monoclonal antibodies established differential expression patterns among cutaneous and oral PVs of snow leopards and domestic cats, suggesting that they represent distinct viruses. Clinically, the lesions in all species and anatomic sites were locally extensive and frequently multiple. Histologically, the areas of epidermal hyperplasia were flat with a similarity to benign tumors induced by cutaneotropic, carcinogenic PVs in immunosuppressed human patients. Limited restriction endonuclease analyses of viral genomic DNA confirmed the variability among three viral genomes recovered from available frozen tissue. Because most previous PV isolates have been species specific, these studies suggest that at least eight different cat papillomaviruses infect the oral cavity (tentative designations: Asian lion, Panthera leo, P1PV; snow leopard, Panthera uncia, PuPV-1; bobcat, Felis rufus, FrPV; Florida panther, Felis concolor, FcPV; clouded leopard, Neofelis nebulosa, NnPV; and domestic cat, Felis domesticus, FdPV-2) or skin (domestic cat, F. domesticus, FdPV-1; and snow leopard, P. uncia, PuPV-2).     

Detection of canine oral papillomavirus-DNA in canine oral squamous cell carcinomas and p53 overexpressing skin papillomas of the dog using the polymerase chain reaction and non-radioactive in situ hybridization

Nineteen cutaneous and mucocutaneous papillomas, as well as 29 oral and 25 non-oral squamous cell carcinomas of dogs were analyzed immunohistologically for the presence of papillomavirus (PV)-antigens. Canine oral papillomavirus (COPV)-DNA was detected in formalin-fixed, paraffin-embedded tissues by polymerase chain reaction (PCR) and non-radioactive in situ hybridization (ISH). Furthermore, the expression of the tumor suppressor protein p53 was investigated. PV-antigens were detectable in more than 50% of the oral and cutaneous papillomas, while no PV-antigens could be demonstrated in venereal papillomas. One squamous cell carcinoma was PV-antigen positive. Only two cutaneous papillomas of the head showed a strong p53-specific immunostaining, while overexpressed p53 was detectable in approximately 35% of all squamous cell carcinomas. It was possible to amplify fragments of the E6, E7 and L1 gene by polymerase chain reaction (PCR) from five of eight oral and from five of eight cutaneous papillomas as well as from three oral squamous cell carcinomas. Nine of 10 papillomas showed a strong nucleus-associated hybridization signal typical for COPV-DNA. In three squamous cell carcinomas COPV-DNA was located in nests of the epithelial tumor cells surrounding ‘horn pearls' or disseminated in the carcinoma tissue. These observations support the view that COPV may also induce non-oral papillomas in the dog and confirm the opinion that a progression of viral papillomas into carcinomas in dogs may occur.     

p53 protein expression in conjunctival squamous cell carcinomas of domestic animals

The expression of p53 protein was investigated in eight formalin-fixed, paraffin-embedded conjunctival squamous cell carcinomas of five horses and one cow, dog and cat each by an immunohistochemical procedure in order to evaluate protein overexpression. Anti-human p53 protein mouse monoclonal antibodies known to be cross-reactive with p53 protein of the animal species examined were used. Positive p53 nuclear immunostaining was detected in five equine, one bovine and one feline cases. Conversely, no p53 immunostaining was found in the only canine case examined. These results demonstrate a frequent p53 overexpression in conjunctival squamous cell carcinoma that could be related to UV-induced mutations of the p53 tumor suppressor gene.     

EcPV2 DNA in equine genital squamous cell carcinomas and normal genital mucosa

Squamous cell carcinoma (SCC) represents the most common genital malignant tumor in horses. Similar to humans, papillomaviruses (PVs) have been proposed as etiological agents and recently Equine papillomavirus type 2 (EcPV2) has been identified in a subset of genital SCCs. The goals of this study were (1) to determine the prevalence of EcPV2 DNA in tissue samples from equine genital SCCs, penile intraepithelial neoplasia (PIN) and penile papillomas, using EcPV2-specific PCR, (2) to examine the prevalence of latent EcPV2 infection in healthy genital mucosa and (3) to determine genetic variability within EcPV2 and to disentangle phylogenetic relationships of EcPV2 among PVs. EcPV2 DNA was detected in all but one penile SCC (15/16), in all PIN lesions (8/8) and penile papillomas (4/4). Additionally, EcPV2 DNA was demonstrated in one of two metastasized lymph nodes, one contact metastasis in the mouth, two vaginal and one anal lesion. In healthy horses, EcPV2 DNA was detected in 10% (4/39) of penile swabs but in none of vulvovaginal swabs (0/20). This study confirms the presence of EcPV2 DNA in equine genital SCCs and shows its involvement in anal lesions, a lymph node and contact metastases. Latent EcPV2 presence was also shown in normal male genital mucosa. We found that different EcPV2 variants cocirculate among horses and that EcPV2 is related to the Delta+Zeta PVs and is only a very distant relative of high-risk human PVs causing genital cancer. Thus, similar viral tropism and similar malignant outcome of the infection do not imply close evolutionary relationship.     

Development of multiple pigmented viral plaques and squamous cell carcinomas in a dog infected by a novel papillomavirus

Canine viral plaques are uncommon skin lesions that are induced by papillomaviruses (PVs). Plaques are usually of little clinical significance in dogs, although they have been reported rarely to progress to squamous cell carcinoma (SCC). Here is described a 7‐year‐old mixed‐breed dog that developed numerous darkly pigmented plaques up to 8 cm in diameter. Multiple ulcerated nodular masses were visible within plaques on the ventrum and axilla. The dog showed no clinical evidence of immunodeficiency and appeared otherwise healthy. Over the next 2 years, five surgeries were performed to remove 23 ulcerated masses that ranged in size from 2 to 5 cm in diameter. Five masses were submitted for histology, and all were SCCs. Each was surrounded by epidermis that contained histological features consistent with those described in canine plaques. Suggestive of a PV aetiology, massive numbers of large keratohyaline granules were present throughout the thickened epidermis. Additionally, koilocytes were focally present, and one sample contained a band of keratinocytes within the superficial epidermis that contained pale cytoplasm and marginated chromatin. From two samples, DNA sequences from a previously unreported PV were amplified, and immunohistochemistry confirmed the presence of PV antigen in both. The PV DNA sequences were most similar to those of canine PVs previously associated with plaque formation. The plaques observed in this case were unusual owing to their rapid growth, large size and frequent malignant transformation. It is unknown whether this unusual behaviour was due to the specific PV detected in this case or to host factors within the dog.     

Expression of p63 normal canine skin and primary cutaneous glandular carcinomas

p63, a recently identified homologue of the p53 protein, is expressed consistently in basal cells of several human multilayered epithelia. In this study, expression of p63 was determined in 31 primary cutaneous glandular carcinomas, including sebaceous, perianal (hepatoid) gland, apocrine and ceruminous carcinomas, as well as their adjacent normal skin. Similar to humans, p63 is a reliable marker for basal and myoepithelial cells in canine epidermis, cutaneous appendages and malignant apocrine and ceruminous gland neoplasms. In sebaceous carcinomas, not only basal cells, but also some sebocytes, showed nuclear staining for p63. Most mature epithelial cells in perianal gland carcinomas exhibited strong p63 expression. Based on these findings, basal/myoepithelial cells could be involved in the oncogenesis of these tumours and p63 might be used as a diagnostic marker in these lesions.     

Novel betapapillomavirus associated with hand and foot papillomas in a cynomolgus macaque

Betapapillomavirus is a genus of papillomaviruses (PVs) commonly found in human skin and associated with both benign and malignant skin lesions. Only 2 previous beta-PVs have been fully characterized in nonhuman species. This report describes a novel beta-PV, named Macaca fascicularis PV type 2 (MfPV2), isolated from exophytic skin papillomas on the hands and feet of a 2-year-old male cynomolgus monkey (M. fascicularis). On histology the papillomas were composed of diffusely thickened epidermis with superficial foci of cytomegaly, cytoplasmic pallor, marginalized chromatin, and rare eosinophilic intranuclear inclusion bodies. Positive immunostaining for p16 and the proliferation marker Ki67 was present multifocally within affected epidermis, most prominently within basal-type cells. Complete sequence identity (100%) was noted between PV genomes fully sequenced from hand and foot lesions. The MfPV2 genome was 7632 base pairs in length and included putative open reading frames (ORFs) for E1, E2, E4, E6, E7, L1, and L2 genes, similar to other PVs. The closest relatives to MfPV2 based on the L1 ORF sequence were all beta-PVs. These included human PV (HPV) 9, HPV115, HPV76, HPV75, and MfPV1 (60-70% pairwise identity for all), the latter of which was also isolated from hand and foot papillomas in a cynomolgus macaque. Phylogenetic analysis placed MfPV2 in a new species group (beta-6), distinct from HPVs (beta-1 to beta-5) and MfPV1 (beta-1). These findings characterize a new nonhuman beta-PV and provide additional support for the idea that tissue tropism among ancestral primate PVs developed prior to divergence of certain Old World primate lineages.     

Mutational analysis of tumor suppressor gene p53 in feline vaccine site-associated sarcomas

OBJECTIVES: To investigate the role of tumor suppressor gene p53 mutation in feline vaccine site-associated sarcoma (VSS) development and to evaluate the relationship between p53 nucleotide sequence and protein expression. SAMPLE POPULATION: Formalin-fixed paraffin-embedded tissues of 8 feline VSS with dark p53 immunostaining (high p53 expression) and 13 feline VSS with faint or no staining (normal p53 expression). PROCEDURE: DNA was extracted from neoplastic and normal tissue from each paraffin block. The following 3 regions of the p53 gene were amplified by polymerase chain reaction: 379 base pair (bp) region of exon 5, intron 5, and exon 6, 108 bp region of exon 7, and 140 bp region of exon 8. Amplified p53 products were sequenced and compared with published feline p53. The p53 mutations identified were correlated with p53 mutations predicted by immunostaining. RESULTS: Neoplastic cells of 5 of 8 (62.5%) VSS that had high p53 expression harbored single missense mutations within the p53 gene regions examined.The p53 gene mutations were not detected in the 13 tumors with normal p53 immunostaining. Nonneoplastic tissues adjacent to all 21 VSS lacked mutations of these p53 gene regions. CONCLUSIONS: The p53 gene mutations were restricted to neoplastic tissue and, therefore, were unlikely to predispose to VSS. However, p53 mutations may have contributed to cancer progression in 5 of the 21 VSS. There was very good (kappa quotient = 0.67 with a confidence limit of 0.3 to 1.0), although not complete, agreement between prediction of mutation by p53 immunostaining and identification of mutations by sequencing of key p53 gene regions.