In humans, induction of protective immunity to smallpox by immunization with vaccinia virus depends on the delivery of virus to the epidermis by scarification and is characterized by development of an epidermal “pox” reaction. Indeed, vaccination by intradermal, subcutaneous and intramuscular strategies have all been shown to result in lower neutralizing antibody titers and a reduced vaccinia-specific cytotoxic T lymphocyte response than vaccination by scarification (McClain et al., 1997
). From early in the history of smallpox vaccination, it has been recognized that certain patient groups are at risk from complications of the scarification technique (Lofquist et al., 2003
). Vaccinia necrosum, or progressive vaccinia, is seen in severely immunocompromised patients and leads to severe morbidity and death from overwhelming infection. Eczema vaccinatum, which occurs in patients with active or quiescent atopic dermatitis, reflects an inability of the atopic host to control the spread of virus from the inoculation site and can result in substantial tissue injury or death. Generalized vaccinia infection, which occurs in patients without atopic dermatitis, is not typically lethal but can also cause significant injury. Due to concern regarding these complications of vaccine use, and the apparent elimination of smallpox infection worldwide, widespread vaccination with vaccinia virus was halted in the United States in 1972 and worldwide in the years to follow. Recent concerns about the possible reemergence of smallpox, or its deliberate release as a bioterror agent, has led to renewed interest in vaccination for the armed forces, front-line health care personnel, and, perhaps, the general population. At the same time, a relative increase over the past 30 years in the percent of the world population with atopic dermatitis, and immune deficiency related to organ transplantation, chemotherapy treatments, and HIV disease has spurred interest in understanding the biology of vaccinia immune responses and driven efforts to design safer vaccination protocols.
The basis for the severe adverse effects of vaccinia scarification seen in atopic patients has not been defined. Vaccination with live virus inherently involves a “race” between virus growth and effective immune response. In the case of inoculation by scarification, the balance and interaction of innate and acquired immune mechanisms in the skin are crucial. Complications can result from an inability to respond to a specific infection or an imbalance that delays control of viral growth. Large DNA viruses, including poxviruses and herpesviruses, commonly encode homologues of cytokines, chemokines, and their receptors as a strategy to evade the host immune response (Alcami, 2003
). Genes expressed by vaccinia virus encode proteins that inhibit complement, bind IL-1 and IL-18, inhibit CC chemokines, and block IFN function, among others (Alcami and Smith, 1992
; Alcami et al., 1998
; Kotwal et al., 1990
; Moss and Shisler, 2001
; Smith et al., 2000
; Symons et al., 1995
). Together, these viral products alter the evolution of inflammatory signals and limit the recruitment of leukocytes to the site of infection, providing the virus with a crucial advantage in the race with the immune response. In the normal host, APCs carrying viral antigen migrate from the site of infection through afferent lymphatics to the draining lymph nodes where the act to prime antigen-specific T cells. Vaccinia-specific CD8+ T cells activated in skin draining lymph nodes become imprinted with a skin-homing phenotype and will home preferentially to the inoculation site (Liu et al., 2006
), where they can lyse infected target cells promote resolution of the infection (Robert and Kupper, 1999
). The dose of vaccinia virus used for scarification has been empirically optimized such that the immune system of a normal host usually “wins,” leading to control of virus growth and long-term immunity. In AD patients, however, the immune response is altered or inhibited such that vaccinia virus replicates faster than the immune system responds, leading to greater local tissue injury and a risk for dissemination.
Atopic skin, therefore, appears to provide a specialized environment that promotes vaccinia virus growth and/ or inhibits anti-viral immunity. Evidence for both functions has been described. Activation and growth of cytotoxic anti-viral T cells, for example, are associated with Th1-type cytokines (e.g., IL-12 and IFNγ). Atopic skin is characterized, however, by reduced expression of Th1 cytokines and increased expression of Th2 cytokines (e.g., IL-4 and IL-13) (Hamid et al., 1994
). Vaccinia virus infection of human keratinocytes has also been shown to induce expression of Th2 cytokines and other immunoregulatory factors, including transforming growth factor beta, interleukin-10 (IL-10), and IL-13, suggesting that vaccinia may skew local cytokine production against the generation of a protective Th1 and cytotoxic T cell responses (Liu et al., 2005
). Innate immune mechanisms are also altered in atopic hosts. LL-37, as an example, is a cathelicidin antimicrobial peptide produced by mammalian skin that exhibits antiviral activity against purified vaccinia virus. Recent studies have demonstrated that atopic skin has decreased LL-37 expression and supports increased replication of vaccinia virus compared with normal or psoriasis skin (Howell et al., 2006
). IL-4 and IL-13 treatment of vaccinia-infected keratinocytes, in turn, enhanced replication of vaccinia virus while down-regulating LL-37 expression. These and other alterations in innate and adaptive immune responses seen in the skin of atopic individuals may promote local growth of vaccinia virus and restrict or delay the expansion or function of Th1 and Tc1 cells that form the basis for the anti-viral response.
As noted previously, the balance of IL-1 agonists and antagonists plays a central role in the regulation of cutaneous immune responses and the development of inflammatory skin diseases (Groves et al., 1995
; Groves et al., 1996
). In patients with atopic skin disease (AD), for example, the ratio of IL-1ra to IL-1α in stratum corneum samples from uninvolved skin of the face, trunk and extremities is significantly increased due to a decrease in IL-1α and an increase in IL-1ra production (Terui et al., 1998
). IL-1β mRNA is increased at baseline in lesional skin from patients with AD, and has been shown to rise in response to patch testing with house dust mite antigen (Jeong et al., 2003
; Junghans et al., 1998
). It is therefore, perhaps, not surprising to find inhibitors of IL-1 function among the immune modulators produced by vaccinia. The vaccinia virus gene B15R, for example, encodes a soluble IL-1 receptor, which binds soluble IL-1β and inhibits functional responses. Experimental deletion of B15R from vaccinia virus accelerates the appearance of symptoms of illness and mortality in mice infected intranasally (Alcami and Smith, 1992
). Myxoma virus, a poxvirus that infects rabbits, encodes a protein named M13L-PYD that blocks production of the interleukin-1 family cytokines by inhibiting the activation of the cytoplasmic “inflammasome.” Knockout viruses that do not express this protein are markedly attenuated, showing decreased viral dissemination and enhanced inflammatory responses at sites of infection (Johnston et al., 2005
). These studies support the impression that over-expression of IL-1 could enhance the innate immune response, inhibit vaccinia replication, and thus diminish the adaptive immune response (Alcami and Smith, 1992
; Johnston et al., 2005
; Spriggs et al., 1992
). A positive protective role for IL-1 in vaccinia virus infection is also reported by several groups. Modified vaccinia virus Ankara (MVA), which is a isolate that lacks the soluble IL-1β receptor gene as well as other immune modulating genes, induces a better CD8+ T cell memory response and confers higher levels of protection against subsequent lethal respiratory with wild-type vaccinia virus (Staib et al., 2005
). Similarly, the viral gene B13R (SPI-2) encodes a serpin homolog that inhibits caspase-1, and thus prevents the maturation of IL-1 β from inactive precursor to active cytokine. Although deletion of B13R diminishes virulence, this recombinant virus still elicited potent humoral, T cell helper, and cytotoxic T cell immune response in the mice, revealing that attenuation did not implicitly reduce immunogenicity (Legrand et al., 2004
In this study, we used a K14/IL-1α transgenic mouse model to examine the role of cutaneous IL-1α as a modulator of immune response to vaccinia virus inoculated by scarification. As shown, K14/IL-1α mice scarified with 2 × 106 PFU of vaccinia mounted an earlier and stronger T-cell immune response and an enhanced humoral immune response compared with WT control animals. Furthermore, K14/IL-1α mice recruited T cells and APC more rapidly to the site of inoculation, and displayed more rapid maturation of DC arriving at the draining lymph nodes. In addition to the effects on adaptive immunity, K14/IL-1α mice also displayed a stronger innate immune response than WT mice. When the dose used for scarification was decreased to 5,000 PFU, WT mice still developed pox lesions and produced significant vaccinia-specific antibody and cell-mediated responses. K14/IL-1α transgenic mice, however, showed much lower levels of virus present in the site of inoculation at 7 days and more than half did not develop a local pox reaction or subsequently show evidence of a memory immune response. We hypothesize that low dose vaccinia virus inoculated by scarification in WT mice is able to suppress the local innate immune response sufficiently to allow local growth of virus, leading ultimately to strong activation of the adaptive immune response and eventual control of the infection. Innate immune function is sufficiently enhanced in K14/ IL-1α mice to clear the virus before there can be growth sufficient to create a pox and/or activate an adaptive immune response. High-dose innoculation, in contrast, overwhelms the protective effect of excess IL-1 in the transgenic animals, resulting in pox lesion formation and strong immune responses in both WT and IL-1 transgenic mice.
In summary, the observation that epidermal IL-1α enhances T cell and antibody responses to vaccinia virus suggests that IL-1α could be used as an adjuvant in humans, increasing the effectiveness of vaccinia virus inoculation in at-risk populations and potentially supporting the use of lower inoculation doses or attenuated virus strains. It is clear, however, that there is an optimal balance between innate and adaptive immune function that must be considered in any such strategy, and that development of a pox lesion may still be the most suitable indicator of successful, protective immunization.