1. to amplify DNA specific to feline papillomavirus type 1 (FcaPV-1) and canine papillomavirus (CaPV) – the narrow-range amplification;
2. to amplify a broad spectrum of papillomaviruses across multiple species – the broad-range amplification.

• Narrow-range amplification of FcaPV-1 and CaPV DNA yielded positive results in one BISC sample, and negative results in all other lesions.
• Broad-range amplification of PV DNA yielded positive results in five BISCs and four SCC lesions.
• No PV DNA was detected in any of the non-neoplastic controls.

• This study utilised PCR in order to demonstrate an association between cutaneous SCC and PV. However, amplification of DNA within a lesion does not necessarily prove an oncogenic link, merely an association. PV can still be isolated from a high proportion of healthy cats (Thomson et al., 2016; and Munday & Kiupel, 2010). PCR alone cannot differentiate whether the presence of PV in SCC lesions is a contaminant, transient infection or an active carcinogen (Hoggard et al., 2018).
• Formalin fixation of tissues can fracture DNA which may interfere or complicate PCR amplification of PV DNA, especially if only small quantities of PV DNA are present (Little et al., 2006).
• Difficult to control or assess the impact of co-factors and co-carcinogens which may potentially influence the results.
• Small sample size.

• DNA was extracted from the FFPE samples for PCR.
• PCR was used to amplify PV DNA in the FFPE samples using two consensus primer sets that detect a broad range of cutaneous papillomaviruses in both humans and animals.
• Six of the amplicons were sequenced.

• PV DNA was amplified in 11 of the 18 BISC samples.
• PV DNA was not amplified in any of the non-SCC samples.
• In the six amplicons that were sequenced, one strain was homologous with a papillomaviral strain that caused multiple cutaneous SCCs in a human; and the other five were weakly homologous to HPV type 17.

• This study utilised PCR in order to demonstrate an association between feline BISCs and PV. However, amplification of DNA within a lesion does not necessarily prove an oncogenic link, merely an association. PV can still be isolated from a high proportion of healthy cats (Thomson et al., 2016; and Munday & Kiupel, 2010). PCR alone cannot differentiate whether the presence of PV in BISCs is a contaminant, transient infection or an active carcinogen (Hoggard et al., 2018).
• Formalin fixation of tissues can fracture DNA which may interfere or complicate PCR amplification of PV DNA, especially if only small quantities of PV DNA are present (Little et al., 2006).
• Difficult to control or assess the impact of co-factors and co-carcinogens which may potentially influence the results.
• Small sample size.

• SCC FFPE samples were divided into two groups –BISC and invasive SCC by a board-certified pathologist.
• DNA was extracted from the tissue samples and nested PCR was performed with two amplifications – the first one was to amplify a DNA fragment found in a wide range of papillomaviruses across multiple species, including cats. The second one was to amplify a DNA fragment in a papillomavirus that was previously found in five feline BISC lesions in another study (Munday et al., 2007).

• Papillomaviral DNA can still be isolated from a proportion of healthy cats (Thomson et al., 2016; and Munday & Kiupel, 2010). PCR alone cannot differentiate whether the presence of PV in SCC lesions is a contaminant, transient infection or an active carcinogen (Hoggard et al., 2018).
• Formalin fixation of tissues can fracture DNA which may interfere or complicate PCR amplification of PV DNA, especially if only small quantities of PV DNA are present (Little et al., 2006).
• Difficult to control or assess the impact of co-factors and co-carcinogens which may potentially influence the results.
• Small sample size.

• SCC FFPE samples were divided into UV-exposed areas (pinna or nasal planum) and UV-protected areas (other areas of the body).
• DNA was extracted from the FFPE samples for PCR using two amplification reactions: firstly to amplify DNA specific to Felis catus papillomavirus 2 (FcaPV-2); and the second was to amplify papillomaviral DNA from a broad range of species. The negative control utilised DNA from a bovine papilloma.
• Immunohistochemistry (IHC) was performed on the FFPE samples to measure the intensity of immunostaining for p16 in the tissues.

• Presence or absence of FcaPV-2 and PV DNA using PCR.
• p16 levels in the FFPE lesions based on intensity of immunostaining.

• FcaPV-2 DNA was amplified from 12 of the 18 SCCs (all seven UV-protected lesions, and five of 11 UV-exposed lesions), one trichoblastoma and all BISCs and feline viral plaque FFPE tissue samples.
• Increased p16 immunostaining was detected in all BISCs, feline viral plaques and UV-protected SCC lesions. In contrast, little p16 immunostaining was observed in UV-exposed SCC and trichoblastoma lesions.
• p16 immunostaining was significantly less intense in UV-exposed SCC lesions compared to UV-protected SCC lesions (P < 0.001).

• DNA was not extracted from six viral plaques.
• Formalin fixation of tissues can fracture DNA which may interfere or complicate PCR amplification of PV DNA, especially if only small quantities of PV DNA are present (Little et al., 2006).
• The specificity of IHC testing to demonstrate a PV/FcaPV-2 aetiology in cats has not been definitely shown and it is possible that spontaneous mutations in a cell’s DNA, not due to PV infection, can lead to increases in p16 (Munday et al., 2019).
• Difficult to control or assess the impact of co-factors and co-carcinogens which may potentially influence the results.
• Small sample size.

• Tissue samples were divided into two groups:
  1. SCC lesions from UV protected skin (densely haired areas – neck, digit, thigh and face);
  2. non-UV protected skin (sparsely haired areas – eyelid, third eyelid, pinna and nasal planum).
• DNA was extracted from the SCC tissue samples, and conventional PCR was used to amplify papillomaviral DNA (including FcaPV-2 specific DNA and papillomaviral DNA from multiple other species).
• IHC was performed on the SCC tissue samples using a mouse anti-human p16 monoclonal antibody to detect increased p16.

• Presence/absence of FcaPV-2 from the FFPE SCC lesions based on amplification of its DNA using conventional PCR.
• Increased p16 based on immunoreactivity levels in the SCC lesions. Percentage of cells in the samples that exhibited intense cytoplasmic and nuclear immunoreactivity was estimated through the examination of five different x400 fields in each section. SCC lesions were considered to have increased p16 if >90% of the cytoplasm or nuclei of the neoplastic cells contained intense immunoreactivity against p16.

• Papillomaviral DNA was identified and amplified in 38/70 SCC cases (54%).
• FcaPV-2 specific DNA was amplified in 33/70 SCC cases (47%).
• Increased p16 levels were detected in 39/70 SCC cases (56%).
• Papillomaviral DNA was amplified significantly more frequently in SCCs with increased p16 compared to SCCs that did not have increased p16 (P < 0.001). Thus, increased p16 were associated with the presence of papillomaviral DNA in feline cutaneous SCC lesions in this study (P < 0.001).

• No comparisons to normal/non-SCC samples (only SCC tissue samples were included in this study). This would have allowed for a more effective comparison, as it would serve as a negative control. Papillomaviral DNA can also be amplified from the skin of normal, healthy cats (Munday & Kiupel, 2010; and Thomson et al., 2016).
• Determination of increased p16 relied on subjective estimation rather than objective measurement. It relied on the estimation of the percentage of increased immunoreactivity levels against p16 through observation of cytoplasmic and nuclear changes.
• Formalin fixation of tissues can fracture DNA which may interfere or complicate PCR amplification of PV DNA, especially if only small quantities of PV DNA are present (Little et al., 2006).
• The specificity of IHC testing to demonstrate a PV/FcaPV-2 aetiology in cats has not been definitely shown and it is possible that spontaneous mutations in a cell’s DNA, not due to PV infection, can lead to increases in p16 (Munday et al., 2019).
• Difficult to control or assess the impact of co-factors and co-carcinogens which may potentially influence the results.

• SCC samples were separated into three categories – dysplasia, carcinoma in situ and SCC.
• PV DNA was extracted from the FFPE samples for PCR and direct sequencing. A FFPE sample of an oral canine papilloma was used as the positive control.

• PV DNA was amplified in 20/126 SCC samples (16% of the total sample size).
• PV DNA was amplified in 9/39 cutaneous SCC samples.
• PV DNA was amplified in 7/22 BISC samples.
• 50% of the PV DNA sequenced was most similar to the HPV genus betapapillomavirus.
• The other 50% of the PV DNA sequenced was most similar to FcaPV-2.

• This study utilised PCR in order to demonstrate an association between feline SCC lesions and PV. However, amplification of DNA within a lesion does not necessarily prove an oncogenic link, merely an association. PV can still be isolated from a high proportion of healthy cats (Thomson et al., 2016; and Munday & Kiupel, 2010). PCR alone cannot differentiate whether the presence of PV in SCC lesions is a contaminant, transient infection or an active carcinogen (Hoggard et al., 2018).
• Formalin fixation of tissues can fracture DNA which may interfere or complicate PCR amplification of PV DNA, especially if only small quantities of PV DNA are present (Little et al., 2006).
• Difficult to control or assess the impact of co-factors and co-carcinogens which may potentially influence the results.

• SCC tissues were grouped into UV-protected or UV-exposed areas according to their location on their body. UV-protected areas included the thigh, digit, face and neck. UV-exposed areas included the eyelid, third eyelid, pinna and nasal planum.
• DNA was extracted from the FFPE samples for PCR.
• PCR was used to amplify two types of PV DNA: DNA specific to FcaPV-2 and papillomaviral DNA from multiple host species. The positive control involved the use of DNA from a feline viral plaque, and the negative control involved the use of DNA from a bovine fibropapilloma.
• IHC was performed on the FFPE samples to detect retinoblastoma (pRb), p16 and p53.

• Presence or absence of PV DNA in the FFPE tissue samples using PCR amplification.
• Presence or absence of p53 and pRb immunostaining using IHC.
• Normal or increased levels of p16 immunostaining using IHC.

• PV DNA was amplified from all viral plaques, all BISCs, and 15 SCC tissues (50% of SCC lesions contained PV DNA).
• Of the 30 samples that contained PV DNA, 29 of those contained DNA specific to FcaPV-2.
• Reduced pRb immunoreactivity was more frequent in tissues that contained PV DNA compared to tissues without PV DNA (P < 0.001). This included 13 SCCs, 5 viral plaques and eight BISCs.
• Increased p16 levels were detected more frequently in lesions that contained PV DNA compared to lesions that did not contain PV DNA (P < 0.001).
• All lesions that had reduced pRb immunostaining also had a simultaneous increase in p16 immunostaining.
• No significant difference in the frequency of p53 immunostaining and the presence or absence of PV DNA (P = 0.45).
• No significant difference in the frequency of p53 immunostaining between lesions with increased p16 (P = 0.227) or decreased pRb (p = 0.183).
• Presence of PV DNA within feline neoplastic and paraneoplastic cutaneous lesions was associated with reduced pRb immunostaining. Reduced pRb was also associated with increased p16. However, feline PVs do not appear to degrade p53 (no significant association).

• Formalin fixation of tissues can fracture DNA which may interfere or complicate PCR amplification of PV DNA, especially if only small quantities of PV DNA are present (Little et al., 2006).
• The specificity of IHC testing to demonstrate a PV/FcaPV-2 aetiology in cats has not been definitely shown and it is possible that spontaneous mutations in a cell’s DNA, not due to PV infection, can lead to increases in p16 (Munday et al., 2019).
• Difficult to control or assess the impact of co-factors and co-carcinogens which may potentially influence the results.
• Small sample size.

• Western blotting (WB) analysis was performed on the CRFK feline epithelial cells to determine the expression levels of protein kinase B (Akt), and the mitogen-activated protein kinases (MAPK), mitogen-activated protein kinase kinase (MEK), extracellular signal-regulated kinase (ERK). WB analysis would also detect their activated forms – pMEK, pERK and pAkt.
• Ribonucleic acid (RNA) was then extracted from the CRFK feline epithelial cells and reverse transcription (RT) was performed to yield complementary DNA (cDNA).
• The cDNA was then subjected to real-time quantitative PCR (qPCR) on epidermal growth factor receptor (EGFR) transcripts.

• Levels of MEK, ERK, Akt and their activated forms: pMEK, pERK and pAkt – in CRFK cell cultures expressing FcaPV-2 genes.
• Levels of EGFR activation or overexpression in the CRFK cell cultures expressing FcaPV-2 genes.

• Total levels of MEK, ERK and Akt were equal and unincreased amongst all cell lines.
• Expression of FcaPV-2 E6 was associated with increased levels of pMEK, pERK and pAkt levels compared to control cells.
• No significant differences in EGFR gene expression were detected amongst all cell lines. This suggests that FcaPV-2 E6 oncogene causes activation of MAPK and Akt pathways independent of EGFR activation/overexpression.

• This is a completely in vitro study which provides a challenge to extrapolate the results to living cats.
• Whilst the study utilised feline epithelial cells, these cells are independent of the tissue environment that skin epithelial cells would be. As such, the molecular pathways examined in this study may not accurately represent the molecular pathways in skin epithelial cells.
• Difficult to control or assess the impact of co-factors and co-carcinogens which may potentially influence the results.

1. In vitro Part: Cell culture of CRFK feline epithelial cells transfected with FcaPV-2 genes – E6E7, E6 and E7. Ultraviolet-B (UVB)-treated CRFK cells were also utilised in a separate intervention group. CRFK cells were obtained from ATCC cell bank.
2. In vivo Part: Feline tissue biopsies from hospitalised cats at the Veterinary Teaching Hospital, University of Naples; and private veterinary practices in Southern Italy.

In vivo Part: Eight cutaneous SCCs, two non-SCC skin lesions, one oral SCC and 10 normal skin samples.

In vitro Part

• RNA was extracted from the CRFK feline epithelial cells and RT performed to yield cDNA.
• The cDNA was then subjected to real-time quantitative PCR (qPCR) for feline p53, glyceraldehyde 3-phosphate dehydrogenase (GAPDH), Bcl-XI and Cdc2.
• CRFK cells for UVB treatment were irradiated in phosphate-buffered saline (PBS) and harvested after 8 hours.
• WB analysis was then performed utilising antibodies against p53, p21, Caspase 3, Bax, Bak, pRb, Cyclin A, Cdc2, Glutathione-S-Transferase (GST) and beta-actin.
• FcaPV-2 and HPV16 E6 and E7 genes were then cloned in a GST fusion vector for GST pull-down assays. The molecular complexes were then analysed by WB for the detection of p53, pRb and GST tags.

In vivo Part

• DNA was extracted from the tissue biopsies for PCR amplification of a specific fragment of the FcaPV-2 L1 gene.
• RNA was extracted from the tissue biopsies and RT performed to yield cDNA.
• The cDNA was then subjected to PCR to amplify specific fragments of FcaPV-2 E2, E6 and E7 genes, and feline GAPDH.

• Establish an in vitro model of the transformative properties of FcaPV-2 E6 and E7 oncogenes and their impacts on the p53 and pRb pathways, and UVB-induced apoptosis.

In vivo Part

• Expression of E6 and E7 oncogenes in SCC samples.
• Expression of E2 in SCC samples.

In vitro Part

• FcaPV-2 E6 bound to feline p53 based on the GST pull-down assay.
• p53 levels were significantly decreased in the CRFK cells expressing E6E7, E6 and E7 genes.
• Presence of E7 but not E6 was always associated with lower messenger ribonucleic acid (mRNA) levels in the CRFK cell lines.
• Presence of E6 and E7 resulted in lower retinoblastoma protein (pRb) levels, suggesting that E7 degrades pRb. However, E6 exhibited no physical interaction with pRb according to the GST pull-down assay.
• UVB-triggered accumulation of p53 and p21 were consistently inhibited in CRFK cells expressing the FcaPV-2 oncogenes – E6E7, E6 and E7. This suggests a synergistic action of FcaPV-2 with UV exposure in oncogenic transformation.

In vivo Part

• Presence of L1 viral gene DNA was detected in 100% of cutaneous SCC lesions, one oral SCC and 40% of normal skin samples.
• FcaPV-2 was observed to be transcriptionally active in 67% of SCC lesions and 20% of normal skin tissue. This was based on the presence of FcaPV-2 E2, E6 and E7 mRNA detected using reverse transcription polymerase chain reaction (RT-PCR). Expression of E2 genes is indicative that FcaPV-2 is biologically active.
• Circular PV DNA was detected in one SCC lesion, suggesting that the expression of E2 occurs independently of the physical status of the virus.
• p53 mRNA levels were downregulated in most of the SCC lesions compared to normal skin samples based on real-time qPCR.

• Difficult to control or assess the impact of co-factors and co-carcinogens which may potentially influence the results.
• Small sample size.
• Whilst the study utilised feline epithelial cells, these cells are independent of the tissue environment that skin epithelial cells would be. As such, the molecular pathways examined in this study may not accurately represent the molecular pathways in skin epithelial cells.

• The SCC tissue samples were classified according to their location – densely haired areas (neck, trunk, paw and face) and sparsely haired areas (pinna, eyelids and nasal planum).
• RNA and DNA were extracted from each FFPE sample.
• Real-time PCR was performed on the FFPE samples to amplify FcaPV-2 DNA.
• Real-time PCR was performed on the FFPE samples to amplify FcaPV-2 E6 and E7 mRNA.
• IHC was then performed on the FFPE samples to detect p16 immunostaining.

• Presence or absence of FcaPV-2 DNA in the FFPE lesions based on real-time PCR amplification.
• Presence or absence of FcaPV-2 E6 and E7 mRNA oncogenes in the FFPE lesions based on real-time PCR amplification.
• Presence of p16 based on immunostaining levels using IHC.

• The presence of amplifiable DNA was confirmed in 52/60 SCCs, all 10 premalignant lesions and in 8/10 normal skin samples.
• Of those FFPE samples that contained amplifiable DNA – FcaPV-2 DNA was amplified from 36/52 SCCs (69%), 3/8 normal skin samples (38%) and all 10 premalignant lesions (100%).
• FcaPV-2 gene expression was detected in 21/60 SCCs (35%), all premalignant lesions (100%) and none of the normal skin samples (0%). The proportion of positive FcaPV-2 gene expression samples was significantly higher in SCC lesions than normal skin samples (P < 0.05).
• FcaPV-2 E6 and E7 gene expression/presence was detected in 20/60 SCCs (33%).
• p16 immunostaining was observed in 31/60 SCCs (52%), all 10 premalignant lesions (100%) and none of the normal skin samples (0%).
• There was a significant association between FcaPV-2 E6/E7 gene expression and p16 immunostaining in the SCC lesions (P < 0.05).

• Formalin fixation of tissues can fragment DNA which may interfere with or complicate PCR amplification of PV DNA, especially if only small quantities of PV DNA are present (Little et al., 2006).
• The specificity of IHC testing to demonstrate a PV/FcaPV-2 aetiology in cats has not been definitively shown and it is possible that spontaneous mutations in a cell’s DNA, not due to PV infection, can lead to increases in p16 (Munday et al., 2019).
• Difficult to control or assess the impact of co-factors and co-carcinogens which may potentially influence the results.

• FFPE samples were divided into SCC lesion (positive control) and non-SCC lesion (negative control) groups. SCC was confirmed histologically by a board-certified veterinary pathologist.
• Conventional PCR was performed on the FFPE tissues in order to identify the presence of FcaPV-2 in the samples. The positive control utilised gDNA from feline viral plaques, whilst the negative control sample had no template DNA.
• Colorimetric RNAscope in-situ hybridisation (ISH) was then conducted to identify FcaPV-2 E6 and E7 mRNA in the FFPE samples using targeting probes specific for the nucleic acid.

• Presence/absence of FcaPV-2 in the FFPE samples by conventional PCR based on amplification of FcaPV-2 DNA.
• Identification and localisation of FcaPV-2 E6 and E7 mRNA in the FFPE samples using colorimetric RNAscope ISH. Identification and localisation of E6 and E7 mRNA is based on positive hybridisation signals from the ISH test.

• FcaPV-2 DNA was amplified in 9/18 SCC samples (50% of SCC lesions had FcaPV-2).
• Of those nine FcaPV-2 positive SCC samples – FcaPV-2 E6 and E7 mRNA was detected in 5/9 of them. 4/9 of these FcaPV-2 positive SCC samples did not contain E6 and E7 mRNA.
• The other nine did not have either FcaPV-2 infection or presence of E6 and E7 nucleic acid. Nine SCC cases did not have FcaPV-2 DNA or the associated E6 and E7 mRNA.
• None of the 10 non-SCC samples returned any positive result for FcaPV-2 E6 and E7 mRNA.
• FcaPV-2 E6 and E7 mRNA was detected and localised in 5/18 SCC samples (28%).

• RNAscope ISH has the potential to detect both RNA and DNA. The study only used this method to detect E6 and E7 mRNA.
• Formalin fixation of tissues can fragment DNA which may interfere or complicate PCR amplification of PV DNA, especially if only small quantities of PV DNA are present (Little et al., 2006).
• Difficult to control or assess the impact of co-factors and co-carcinogens which may potentially influence the results.
• Small sample size.