Herpes simplex virus type 1 (HSV-1) initially infects epidermal or mucosal tissues where it replicates and enters the cutaneous sensory axons. It is then retrogradely transported to the neurons of dorsal root ganglia where latent infection is established. Sporadic reactivation of latent virus and transmission to peripheral sites leads to asymptomatic viral shedding or recurrent herpes lesions (41, 50). Symptomatic recurrences are dependent on the levels of local immunity at the periphery and are principally controlled by cell-mediated responses via direct T-cell effector function and cytokine release (24,34,37,45). CD4 ϩ T cells are the initial infiltrating cells in the recurrent lesion, followed by CD8 ϩ cells (9). An immediate-early gene product, ICP47, inhibits major histocompatibility complex (MHC) class I-mediated antigen presentation by binding via its N terminus to the transporter-associated protein (TAP) (1,13,18,49,52). This ICP47-TAP complex prevents translocation of the MHC class I-processed peptide complex to the cell surface, which would prevent recognition by CD8 ϩ T cells. However, gamma interferon secreted by CD4 cells reverses the MHC class I downregulation and stimulates MHC class II expression, allowing targeting of HSV-infected epidermal cells by both CD4 ϩ and CD8 ϩ cytotoxic T cells (33). The persistence of HSV-specific CD8 ϩ clones in seropositive individuals, however, suggests their complementary importance to CD4 ϩ T cells in the control of recurrences (35, 36). The early appearance of virus-specific T-cell effectors depends on prompt signaling from antigen-presenting cells, such as dendritic cells (DC), B cells, or macrophages. The strategic positioning of Langerhans cells in the skin and mucosa identifies them as the most likely antigen-presenting DC to have the earliest contact with the incoming virus.DC form a network of bone marrow-derived antigen-presenting cells that are required for the initiation of adaptive immune responses. DC can be found in virtually all peripheral tissues and are characterized by the CD1a ϩ HLA-DR ϩ CD80 ϩ CD86 ϩ phenotype (3, 17). The capacity of DC to stimulate T cells is closely related to their maturation stage (40). Immature DC, exemplified by epidermal Langerhans cells, act as sentinels and are highly specialized at antigen uptake and processing but are poor stimulators of primary immune responses (3,17,43). A variety of immune stimuli induce phenotypic and functional changes to resting DC as they migrate out of the tissues and into secondary lymphoid organs. Mature DC no longer take up and process antigen but upregulate MHC class I and II costimulatory and adhesion molecules, effectively boosting their ability to present processed peptides to antigen-specific T cells. During the process of maturation, DC also express CD83, a maturation marker not found on resting immature DC such as Langerhans cells (54). Terminal maturation of DC is induced upon contact with T cells via CD40 ligation (6). Experimental studies of human Langerhans cells have been hamp...

In early recurrent herpetic lesions, CD4 T lymphocytes are the predominant infiltrating cells, and keratinocytes expressing major histocompatibility complex (MHC) class II antigens, induced by interferon-gamma (IFN-gamma), are the major site of herpes simplex virus (HSV) replication. IFN-gamma pretreatment of human keratinocytes in vitro reduced MHC class I antigen down-regulation by HSV-1 infection and induced expression of HLA-DR that was unaltered by subsequent HSV-1 infection. Incubation of these infected keratinocytes with phosphonoacetic acid (PAA) almost completely inhibited expression of four major HSV glycoproteins, although expression of early proteins was not affected. Weak CD8 T lymphocyte cytotoxicity against IFN-gamma-stimulated, HLA-DR-expressing HSV-1-infected keratinocytes was consistently directed to the immediate early/early proteins (all 9 patients tested) but against late proteins to a lesser degree (4/9 patients). However, CD4 T lymphocyte cytotoxicity was much greater and directed predominantly against late HSV-1 glycoproteins (all 9 subjects tested) in these cells.

The ability of alpha interferon (IFN-␣) and IFN-␥ to inhibit transmission of herpes simplex virus type 1 (HSV-1) from neuronal axon to epidermal cells (ECs), and subsequent spread in these cells was investigated in an in vitro dual-chamber model consisting of human fetal dorsal root ganglia (DRG) innervating autologous skin explants and compared with direct HSV-1 infection of epidermal explants. After axonal transmission from HSV-1-infected DRG neurons, both the number and size of viral cytopathic plaques in ECs was significantly reduced by addition of recombinant IFN-␥ and IFN-␣ to ECs in the outer chamber in a concentrationdependent fashion. Inhibition was maximal when IFNs were added at the same time as the DRG were infected with HSV-1. The mean numbers of plaques were reduced by 52% by IFN-␣, 36% by IFN-␥, and by 62% when IFN-␣ and IFN-␥ were combined, and the mean plaque size was reduced by 64, 43, and 72%, respectively. Similar but less-inhibitory effects of both IFNs were observed after direct infection of EC explants, being maximal when IFNs were added simultaneously or 6 h before HSV-1 infection. These results show that both IFN-␣ and IFN-␥ can interfere with HSV-1 infection after axonal transmission and subsequent spread of HSV-1 in ECs by a direct antiviral effect. Therefore, both IFN-␣ and -␥ could contribute to the control of HSV-1 spread and shedding in a similar fashion in recurrent herpetic lesions.