Abstract
Feline sarcoids are uncommon dermal neoplasms that are associated with papillomavirus (PV) infection. A single PV type, designated feline sarcoid–associated PV (FeSarPV), was detected in 9 feline sarcoids from North America. As FeSarPV has only been detected within feline sarcoids, the epidemiology of the infection remains unknown. The present study used polymerase chain reaction (PCR) to investigate whether this PV is also present within sarcoids from New Zealand cats. Additionally, as PVs are often host-specific, it was hypothesized that FeSarPV may often asymptomatically infect cats but rarely cause disease. To test this hypothesis, specific PCR primers were designed to investigate the presence of FeSarPV DNA within 120 samples from the skin and mouth of cats without sarcoids. Feline sarcoids from both New Zealand and North America contained FeSarPV DNA sequences. However, FeSarPV DNA was not detected within any non-sarcoid feline sample. To the authors' knowledge, this is the first time that FeSarPV has been reported in a country outside North America. As FeSarPV does not asymptomatically infect cats, feline sarcoids are likely due to cross-species infection. Although the reservoir host of FeSarPV is unknown, the host is present and has contact with cats, in both New Zealand and North America.
Feline sarcoids (also called feline cutaneous fibropapillomas) are uncommon, dermal neoplasms that tend to occur on the philtrum, nares, upper lip, digits, ears, and tail of younger, male cats from rural areas. 9 Lesions histologically consistent with feline sarcoids have been previously reported in North America, New Zealand, England, Sweden, and Australia. 1,10,19 Although feline sarcoids do not contain histologic or immunohistochemical evidence of papillomavirus (PV) infection, PV DNA has been detected using polymerase chain reaction (PCR) and in situ hybridization. 19,21 Sequencing of short sections of PV DNA amplified from 9 North American feline sarcoids suggested the presence of a novel feline sarcoid-associated PV (FeSarPV). 21 As PV DNA has not previously been amplified from a feline sarcoid outside North America, it is unknown whether FeSarPV is also associated with feline sarcoids in other regions. In the present study, PV DNA was amplified from both New Zealand and North American feline sarcoids.
To the authors' knowledge, there have been no studies investigating the presence of FeSarPV within non-sarcoid feline samples, and the epidemiology of infection is unknown. Asymptomatic PV infections are common in many species, 2 and more than half of cats are asymptomatically infected by Feline papillomavirus type 2 (FDPV-2). 15 Most PVs appear to be species-specific, 8 and recent investigations of equine sarcoids, a neoplasm that shares many features with feline sarcoids, 9 revealed that the causative PV is often present on equine skin. 3,16 Therefore, it was hypothesized that FeSarPV, frequently and asymptomatically, infects cats but rarely causes disease. To test this hypothesis, specific primers were designed to detect FeSarPV in samples from the skin and oral cavity of cats without sarcoids.
Feline sarcoids were identified by searching biopsy databases at the Diagnostic Center for Population and Animal Health, Michigan State University; New Zealand Veterinary Pathology Ltd, New Zealand; and Gribbles-Alpha, New Zealand. Once potential cases were identified, histologic sections were examined by one author (JSM). Cases were classified as a feline sarcoid according to histologic criteria previously reported. 9 Briefly, feline sarcoids consisted of a dermal proliferation of spindle cells that extended to overlying markedly hyperplastic epidermis. Long rete pegs of hyperplastic epidermis were present extending into the spindle cell population. Four feline sarcoids were identified within the New Zealand databases and 3 from Michigan. None contained histologic evidence of PV infection.
DNA was extracted from the formalin-fixed, paraffin-embedded, feline sarcoids, as previously described. 13 DNA was also extracted from 36 formalin-fixed, cutaneous samples (18 invasive squamous cell carcinomas, 4 fibrosarcomas, 4 biopsies of allergic dermatitis, 3 biopsies of normal skin, 2 cases of feline leprosy, 2 feline mast cell tumors, 2 apocrine gland cystadenomas, and a follicular cyst) and 16 oral samples (8 cases of plasmacytic stomatitis, 5 eosinophilic granulomas, 2 cases of glossitis, and a case of periodontal disease). Skin and oral swabs were collected from 16 clinically healthy cats with the skin of a further 36 clinically healthy cats also swabbed. All cats were allowed outside access by their owners and were from both urban and rural locations within New Zealand. Swabs were performed as previously described. 14 placed in 1 ml 0.9% NaCl, and stored at −20°C.
The presence of PV DNA within the feline sarcoids was investigated using MY09/MY11 consensus primers, which amplify a 450-bp conserved region of the PV L1 gene. 7 Reaction conditions were as previously reported. 7 DNA extracted from a formalin-fixed, equine sarcoid was used as a positive control, whereas template DNA was not added to the negative control. The MY09/MY11 primers amplified PV DNA from 4 of the 7 sarcoids (57%). Direct sequencing 13 of 1 amplicon revealed a 340-bp sequence (GenBank accession no. FJ977616). This sequence partially overlapped a 102-bp section of PV L1 gene that had been previously amplified from 9 feline sarcoids. 21 Within the 63-bp overlapping section, the 2 sequences were 95.2% similar. The sequence was also compared with known sequences using the basic local alignment search tool (http://www.ncbi.nlm.nih.gov/blast). The 340-bp sequence was 72% similar to Bovine papillomavirus-1 (BPV-1) and Ovine papillomavirus type 2 (OvPV-2), 71% similar to Bovine papillomavirus type 2 (BPV-2) and Western roe deer papillomavirus type 1, and 70% to Deer papillomavirus. Comparison of the 340-bp sequence to Feline papillomavirus type 1 (FDPV-1) and FDPV-2 revealed similarities of 57% and 60%, respectively.
Specific primers were designed to detect the PV DNA sequence amplified by the consensus primers. These primers were designated jmpSA-F (5′-GGAACAAACCTCACAATCAC-3′) and jmpSA-R (5′-CCAGTTCTCTAATACTGAGG-3′). Final concentrations of the reaction products using the jmpSA primers were 1 x PCR buffer, 1.5 mM MgCl2, 200 μM ethylenediamine tetra-acetic acid, 0.25 μM of each primer, 1.25 U DNA polymerase, and 2.5 μl template DNA in a final reaction volume of 30 μl. Amplification conditions were 94°C for 10 min followed by 45 cycles of 94°C for 1.5 min, 60°C for 1.5 min, and 72°C for 1.5 min. The final extension was at 72°C for 5 min. Electrophoresis in a 1% agarose gel containing ethidium bromide was used to detect the 195-bp amplified fragment. Samples of a feline viral plaque that contained FDPV-1 and FDPV-2, 14 an equine sarcoid containing BPV-1, and a canine oral papilloma containing Canine oral papillomavirus were used to evaluate the specificity of the primers. Negative controls consisted of the above reaction conditions without template DNA. Amplicons were sequenced as previously described. 13
The jmpSA primers amplified PV DNA from 6 of the 7 feline sarcoid samples (86%) but none of the 120 non-sarcoid feline samples. All 6 amplicons were between 97% and 99% similar. This small variance is considered to be the result of errors during sequencing rather than the presence of multiple PVs. DNA was not amplified by the jmpSA primers from samples of feline viral plaque, equine sarcoid, or canine oral papilloma.
Two studies have previously used PCR to detect PV DNA in feline sarcoids. 19,21 The 2 studies revealed that PV DNA was detectable in 17 out of 17 (100%) and 9 out of 12 (75%) sarcoids from North American cats. 19,21 In the second study, all 9 sarcoids contained the same PV, designated FeSarPV. 21 In the present study, specific primers amplified FeSarPV DNA from 3 out of 4 (75%) New Zealand and all 3 (100%) North American feline sarcoids. To the authors′ knowledge, this is the first time that FeSarPV has been detected outside North America. The consistent identification of FeSarPV within feline sarcoids from 2 geographically diverse locations supports a causal role of this PV in neoplasm development. Although additional research is required, considering the previous reports of feline sarcoids in Europe, North America, and Australasia, 1,10,19 it appears likely that FeSarPV has a widespread distribution.
The FeSarPV sequence amplified from 6 feline sarcoids in the present study was most similar to BPV-1, OvPV-1, and BPV-2. This is consistent with a previous study that reported a section of PV E1 gene amplified from a feline sarcoid was 75% similar to BPV-1 and 64% similar to BPV-2. 19 Likewise, a 102-bp section of the L1 gene that was amplified from 9 feline sarcoids was 82% similar over 93 bp to OvPV-2 and 82% similar over 62 bp to BPV-2. 21 All PVs contain the early protein E1, a DNA helicase and ATPase that initiates viral replication, 11 and the late protein L1, the main component of the viral capsid. 12 Papillomaviruses are classified by their L1 genes, and greater than 10% variance indicates a different PV type. 8 Although definitive classification of FeSarPV is not possible until the entire L1 gene has been sequenced, the results of the current and previous studies suggest that FeSarPV is a novel PV.
Feline sarcoid–associated PV appears most similar to BPV-1, OvPV-1, and BPV-2, which are all members of the genus Deltapapillomavirus. 8 The deltapapillomaviruses are generally considered to have ruminant reservoir hosts but are often able to infect multiple species. 8 Equine sarcoids have similar histology and clinical behavior to feline sarcoids and are caused by the deltapapillomaviruses BPV-1 and BPV-2. 5,16 Initially, these PVs were considered to be restricted to bovine fibropapillomas and equine sarcoids, indicating that equine sarcoids develop from cross-species infection of horses by a PV from cattle. 4,6,18 However, more sensitive methods detected BPV-1 or BPV-2 DNA in 57% of skin samples from clinically healthy horses 3 and in samples of equine dermatitis. 6,22 Furthermore, analysis of variability within BPV sequences from equine sarcoids and bovine papillomas suggests that equine-adapted BPV strains exist. 16 Considering the similarities between equine and feline sarcoids, it was hypothesized that FeSarPV could be a feline-adapted strain of deltapapillomavirus that commonly infects cats but causes sarcoids only secondary to an additional factor. Trauma has previously been suggested as an important process in both equine and feline sarcoids, 3,21 and more frequent trauma from fighting or hunting could explain the predisposition for sarcoids observed in young male cats from rural areas. The failure to detect FeSarPV DNA within any non-sarcoid sample in the present study does not support this hypothesis. However, as the detection limits of the PCR used in the current study are unknown, asymptomatic infection of cats by FeSarPV cannot be definitively excluded.
The results of the present study are more consistent with cross-species PV infection from a reservoir host as the origin of feline sarcoids. However, as FeSarPV has only ever been detected within feline sarcoids, the reservoir host is unknown. If FeSarPV is a deltapapillomavirus, a ruminant reservoir host may be most likely. Because FeSarPV is present within both New Zealand and North American feline sarcoids, the reservoir host has contact with cats in both locations. Many rural cats in both New Zealand and North America have contact with cattle. Therefore, as has been previously suggested, 19 feline sarcoids could be due to cross-species infection of cats from cattle. However, 15 bovine papillomas and 122 bovine skin swabs were investigated for the presence of PVs using the MY09/MY11 consensus primers. 17 Although DNA from 4 different BPVs was identified in 16 of the bovine samples, none of the DNA sequences were from FeSarPV. 17 Whereas this could suggest that cattle are not the reservoir host of FeSarPV, it is possible that the consensus primers used were not sensitive enough to detect small quantities of FeSarPV within healthy skin. Cats from both countries may also have contact with sheep. Unlike North American cats, New Zealand cats could have contact with red deer but not white-tailed deer.
Both formalin-fixed samples and swabs were used to investigate the presence of FeSarPV in the current study. Previous studies revealed that PVs are more frequently detectible from swabs than formalin-fixed samples, possibly because formalin fixation damages DNA. 14 Indeed, the failure to amplify FeSarPV DNA from one feline sarcoid in the present study was considered most likely because of formalin-induced DNA degradation. However, using DNA extracted from formalin-fixed samples allowed inclusion of inflammatory skin lesions, which were, in horses, more commonly infected by BPV than normal skin. 22 Specific primers against FeSarPV were designed and used in the present study for 2 reasons. First, specific primers have greater affinity for the target DNA and are expected to be more sensitive. This greater sensitivity was demonstrated by the amplification of FeSarPV by the specific primers, but not the consensus primers, from 2 feline sarcoids. Second, asymptomatic PV infection of feline skin is common. 15 Consensus primers may have amplified DNA from PVs other than FeSarPV making subsequent sequencing necessary. Additional sequencing was avoided in the present study by using specific primers.
In conclusion, the detection of a single PV type within all 6 feline sarcoids strengthens the association between FeSarPV and lesion development. Feline sarcoids from both North America and New Zealand contained FeSarPV DNA. To the authors' knowledge, this is the first detection of FeSarPV outside North America and suggests that this PV may have a wide distribution. It was hypothesized that FeSarPV could be a frequent, asymptomatic infection of feline skin that only causes sarcoids after trauma. However, FeSarPV was not detected within any non-sarcoid sample, suggesting that feline sarcoids develop from cross-species infection. The reservoir host of FeSarPV remains unknown, but evidence suggests that the host has contact with domestic cats in both North America and New Zealand. Identification of the reservoir host is important in understanding the epidemiology of this disease.