Mousepox, Skin, Mouse

The vaccinia virus growth factor (VGF) gene encodes a polypeptide with amino acid sequence homology to epidermal growth factor (EGF) and transforming growth factor alpha and is present twice, once at each end of the virus genome within the inverted terminal repetition. Recombination procedures were used to replace more than half of both VGF genes with a beta-galactosidase cassette which served as a color indicator for isolating an unconditionally viable VGF- mutant. The VGF- mutant genotype and phenotype were confirmed by Southern blot analysis and assays for functional growth factor. The plaque-forming efficiencies of VGF- and wild-type (WT) viruses were similar in a variety of cell types containing low or high densities of EGF receptors, suggesting a lack of a specific requirement for either VGF or the EGF receptor in the initiation of virus infection. The yield of VGF- virus was similar to that of WT virus in growing BS-C-1 and Swiss 3T3 cells, but lower in resting Swiss 3T3 cells. The greatest differences between VGF- and WT virus occurred in vivo: higher doses of VGF- virus than WT virus were required for intracranial lethality in mice and for production of skin lesions in rabbits. Thus, expression of the VGF gene is important to the virulence of vaccinia virus.

VGF, a polypeptide encoded by vaccinia virus, shares amino acid sequence homology and functional properties with cellular growth factors EGF and TGF-alpha. The availability of a VGF minus (VGF-) virus mutant has enabled us to examine the role of VGF in the replication of virus in vitro and in vivo. Studies in vitro with A431 cells (high EGF receptor density) showed that VGF+ wild-type virus induced the rapid formation of a focus of infection (not a plaque) which could be blocked by a monoclonal antibody to the EGF receptor. In vivo experiments with chicken embryos indicated that VGF+ virus stimulated the growth of ectodermal and entodermal cells of the chorioallantoic membrane. At early times, the majority of proliferating cells contained no detectable virus antigen, indicating that cell growth preceded infection and was a consequence of VGF secretion. Relative to VGF- virus, VGF+ virus produced lesions which contained more proliferating cells, more virus antigen, and increased amounts of infectious progeny. Secretion of VGF thus explains the conundrum of a nontransforming, strongly cytopathic virus inducing a hyperplastic cell response.

Cytotoxic T lymphocytes (CTL) seem to provide the major line of defence against many viruses. CTL effector functions are mediated primarily by cells carrying the CD8 (Ly-2) antigen (CD8+ cells) and are triggered by interactions of the T-cell receptor with an antigenic complex, often termed 'self plus X', composed of viral determinants in association with class I molecules of the major histocompatibility complex (MHC). The mechanism(s) of induction of virus-specific CTL in vivo is poorly understood, but data from in vitro experiments suggest that their generation is strictly dependent on functions provided by CD4+ helper T cells (also referred to as L3T4+; or TH) that respond to antigens in the context of class II (Ia) MHC determinants. The prevailing opinion that induction of most functions of CD8+ cells requires help provided by CD4+ cells has recently been challenged by the observation that CD8+ cells alone can mediate a variety of responses to alloantigens in vitro and in vivo; however, the possibility that CTL to self plus X could be generated in vivo in the absence of TH cells has not been evaluated. We report here that C57BL/6J (B6) and AKR/J mice, when functionally depleted of CD4+ cells by in vivo treatment with the CD4+-specific rat monoclonal antibody GK1.5 (refs 8-14) responded to ectromelia virus infection by developing an optimal in vivo virus-specific CTL response, and subsequently recovered from the disease (mousepox) that was lethal for similarly infected nude mice (CD4-, CD8-).