Search

Cell-mediated immunity plays a key role in the regression of papillomavirus-induced warts and intra-epithelial lesions but the target antigens that induce this response are not clear. Canine oral papillomavirus (COPV) infection of the oral cavity in dogs is a well-characterized model of mucosal papillomavirus infection that permits analysis of the immune events during the infectious cycle. In this study we show that during the COPV infectious cycle, systemic T cell responses to peptides of several early proteins particularly the E2 protein, as assayed by delayed type hypersensitivity, lymphoproliferation and IFN-gamma ELISPOT, can be detected. The maximal response occurs in a narrow time window that coincides with maximal viral DNA replication and wart regression: thereafter, systemic T cell responses to early proteins decline quite rapidly. Vaccination using particle-mediated immunotherapeutic delivery (PMID) of codon-modified COPV E2 and E1 genes induces strong antigen-specific cell-mediated immune responses in the vaccinated animals. These data show that therapeutic immunization by PMID with codon-modified E2 is completely effective, that to E1 is partially protective, that this correlates with the intensity of antigen-specific cell-mediated immune responses and, further, they emphasize the importance of these responses and the route of immunization in the generation of protective immunity.

Replication-deficient adenoviral (rAd5) vaccines containing codon-optimized E1, E2, E4, and E7 genes of canine oral papillomavirus (COPV) were tested singly or in combination to determine which vaccines could protect against mucosal challenge with COPV. In three studies, groups of 4-6 beagle dogs were immunized subcutaneously (s.c.) with 10(11) rAd5 at 8-10 weeks and 4-6 weeks prior to challenge with infectious COPV particles at multiple oral mucosal sites. Control dogs were immunized with equivalent doses of rAd5 expressing human papillomavirus (HPV) type 16 L1 (rAd5-HPV-16 L1). In the first study, complete protection from COPV-induced papillomas was achieved by immunization with rAd5 vaccine combinations expressing either E1 + E2 or E1 + E2 + E4 + E7; whereas two of six dogs immunized with rAd5-E4 + rAd5-E7 and six of six rAd5-HPV16-L1-immunized control dogs developed oral papillomas. In two subsequent studies, rAd5-E1 and rAd5-E2 vaccines were tested singly or in combination to assess levels of protective immunity to COPV challenge. Subcutaneous immunization with either one or two doses of rAd5 expressing the COPV E1 and E2 genes could protect > 90% of challenged dogs from wart formation. In contrast, all eight dogs immunized with rAd5-HPV-16 L1 developed papillomas at multiple sites. Protection was accompanied by significant IFN-gamma responses to COPV E1 and E2 peptides. Partial protection was conferred by two immunizations with either rAd5-E1 (6 of 9 protected) or rAd5-E2 (8 of 9 protected). These data indicate that rAd5 expressing papillomavirus E1 and E2 proteins can induce strong protective responses even in outbred populations under practical immunization conditions.

Following challenge with COPV (canine oral papillomavirus), DNA plasmids encoding COPV L1, E1 or E2 protein were delivered into oral mucosal and cutaneous sites in beagles using particle-mediated immunotherapeutic delivery (PMID). Two weeks post-challenge, a priming dose of 8 microg DNA was delivered followed by a booster dose after a further two weeks. A group of control dogs were vaccinated using plasmid DNA encoding Hepatitis B virus surface (HBVs) gene. All of the control animals developed warts at the vast majority of sites (94%). All of the animals given wild type L1, E1, or E2 developed warts at most sites (88%, 75%, and 88%, respectively). The animals given codon optimised E2 however, were protected from wart growth with only one tiny lesion seen on a single animal that persisted for only a few days. The E1 codon optimised group was also significantly protected with a far lower number of smaller warts (48%) that persisted for a shorter duration. These data suggest that therapeutic immunisation by PMID with papillomavirus early genes is effective and emphasizes the importance of antigen load in the generation of protective responses to papillomavirus proteins.

DNA plasmids encoding the open reading frames of canine oral papillomavirus (COPV) nonstructural early genes E1, E2, or E7 protein were delivered into both oral mucosal and cutaneous epithelial sites in beagle dogs using particle-mediated immunotherapeutic delivery (PMID) technology. Control dogs were vaccinated with plasmid encoding either hepatitis B virus surface antigen (HBVs) or COPV L1. Using a prophylactic immunisation protocol, a priming dose of plasmid DNA was followed by a booster dose 6 weeks later. Four weeks after boost, all dogs were challenged with infectious COPV particles. Following viral challenge, as shown previously (M. A. Stanley et al., 2001, Vaccine 19, 2783-2792), mucosal papillomas developed in the negative-control HBVs vaccinated dogs, but all animals in the COPV L1 group were fully protected from disease development. In the early gene-vaccinated groups five of six in the E1-vaccinated dogs, two of six in E2-vaccinated dogs, and three of six in the E7-vaccinated beagles developed oral papillomas. Compared to the HBVs negative-control group the oral papillomas that did develop in the early-gene vaccinated beagles were significantly smaller, shorter in duration, and fewer in number. Taken together the disease burden was markedly reduced and this was statistically significant. In a second experiment one group of animals was vaccinated with plasmid encoding the wild-type COPV E1 gene, and a separate group was vaccinated with plasmid encoding a synthetic codon-optimised COPV E1 gene sequence. None of the codon-optimised E1-vaccinated animals developed papillomas at any challenge site. However, all animals vaccinated with wild-type E1 had papillomas. These data suggest that immunisation by PMID with papillomavirus early genes can significantly impact upon subsequent disease development and that full protection can be achieved using improved vectors encoding codon-optimised gene sequences perhaps emphasizing the importance of antigen load in the generation of protective responses to papillomavirus proteins., (Copyright 2002 Elsevier Science (USA))

We studied experimental canine oral papillomavirus (COPV) infection by in situ hybridization and immunohistochemistry of weekly biopsies. After 4 weeks, viral DNA in rete ridges suggested a keratinocyte stem cell target. Abundant viral DNA was seen in E4-positive cells only. E4 was predominantly cytoplasmic but also nuclear, being concentrated in the nucleoli during wart formation. Infected cells spread laterally along the basal layer and into the parabasal layers, accompanied by E7 transcription and increased mitoses. Most of the lower epithelium was positive for viral DNA, but, in mature warts, higher levels of E4 expression and genome amplification occurred in only sporadic superficial cells. L1 expression was late and in only a subset of E4-positive cells. During regression, viral DNA was less abundant in deep epithelial layers, suggesting downregulation of replication prior to replacement of infected cells from beneath. Detection of viral DNA in post-regression tissue indicated latent infection., (Copyright 2001 Academic Press.)

Canine oral papillomavirus (COPV) infection is used in vaccine development against mucosal papillomaviruses. The predictable, spontaneous regression of the papillomas makes this an attractive system for analysis of cellular immunity. Immunohistochemical analysis of the timing and phenotype of immune cell infiltration revealed a marked influx of leukocytes during wart regression, including abundant CD4+ and CD8+ cells, with CD4+ cells being most numerous. Comparison of these findings, and those of immunohistochemistry using TCRalphabeta-, TCRgammadelta-, CD1a-, CD1c-, CD11a-, CD11b-, CD11c-, CD18-, CD21-, and CD49d-specific monoclonal antibodies, with previously published work in the human, ox, and rabbit models revealed important differences between these systems. Unlike bovine papillomavirus lesions, those of COPV do not have a significant gamma/delta T-cell infiltrate. Furthermore, COPV lesions had numerous CD4+ cells, unlike cottontail rabbit papillomavirus lesions. The lymphocyte infiltrate in the dog resembled that in human papillomavirus lesions, indicating that COPV is an appropriate model for human papillomavirus immunity., (Copyright 2001 Academic Press.)

Papillomaviruses occasionally cause severe, nonregressing or recurrent infections in their human and animal hosts. The mechanisms underlying these atypical infections are not known. Canine oral papillomavirus (COPV) typically regresses spontaneously and is an important model of mucosal human papillomavirus infections. A severe, naturally occurring, nonregressing COPV infection provided an opportunity to investigate some aspects of viral pathogenicity and host immunity. In this case, the papillomas proved refractory to surgical and medical treatments, including autogenous vaccination and vaccination with capsid (L1) virus-like particles. High levels of induced anti-L1 antibodies appeared to have no effect on the infection. The papillomas spread to oesophageal mucosa, perioral haired skin, and remote cutaneous sites. Isolation of COPV from the animal and sequencing of several regions of the viral genome showed no differences to the COPV prototype. Experimental infection of beagle dogs with this viral isolate resulted in the uncomplicated development and regression of oral warts within the usual period, indicating that the virus was not an unusual pathogenic variant. These findings support the hypothesis that the recurrent lesions seen in some human papillomavirus infections, such as recurrent laryngeal papillomatosis, are associated with specific defects in host immunity rather than variations in viral pathogenicity., (Copyright 1999 Academic Press.)