It has become increasingly clear that the early cellular and molecular events after infection not only are important in the regulation of the early infection but also strongly influence the nature of the acquired immune response and the ultimate outcome of the disease. While the critical role of the early events in infection has been documented for some intracellular organisms, there is little information available on the nature of the early events in the genital tract following infection with chlamydiae. Interestingly, during our experiments in which we observed the appearance of IFN-γ-producing cells concomitantly with resolution of infection, we also noted a large number of IFN-γ-producing cells appearing as early as 7 days after infection. Since preliminary experiments suggested that this early IFN-γ response was not mediated by CD4 cells, we hypothesized that the cellular source of the IFN-γ was NK cells.
Initially, we confirmed that IFN-γ, as measured by RT-PCR, was up-regulated beginning as early as 12 to 18 h after intravaginal infection with chlamydiae and that high levels were attained by 36 h after infection. When the NK response was assessed in the genital tract by the YAC-1 cytotoxicity assay, an increase in NK cells in the genital tract was also observed as early as 12 to 24 h after infection, with the peak number being seen 48 to 96 h after infection. Thus, these data demonstrated a strong temporal association between the appearance of NK cells in the genital tract and the up-regulation of the local IFN-γ response. An increase in NK activity was also observed in the ILN and spleen, although it was delayed by several days in comparison to the genital tract. The fact that the killing of the YAC-1 cells was indeed caused by NK cells was confirmed by the abrogation of the cytotoxic response by genital tract and ILN mononuclear cells by the in vitro treatment with rabbit anti-asialo-GM1 and complement but not with anti-CD3 and complement. A similar reduction in YAC-1 cytotoxicity was seen when mice were treated in vivo with anti-asialo-GM1.
In order to confirm that the IFN-γ response was dependent upon the presence of NK cells, MoPn-infected mice were depleted of NK cells by in vitro and in vivo treatment with anti-asialo-GM1. In both experiments, when the number of cells producing IFN-γ was quantified from the ILN by the ELISPOT assay, a significant reduction in number was seen in the groups treated with anti-asialo-GM1. In vitro treatment of ILN cells with anti-CD3 did not alter the number of cells producing IFN-γ. Finally, when the transcription of IFN-γ was determined in genital tract lymphocytes from mice treated with anti-asialo-GM1, a marked decrease in transcripts was also seen. Thus, these data strongly indicate that NK cells trafficking to or within the genital tract as a result of chlamydial genital infection are responsible for the production of IFN-γ early in the infection course.
The appearance of NK cells following chlamydial respiratory infection has been previously documented by Williams et al. (30
) when they observed an increase in cytotoxicity for YAC-1 cells by spleen cells 5 days after intranasal inoculation with MoPn. The finding that IFN-γ production by NK cells occurs early in the infection is not surprising in that similar results have been observed in mice infected with MoPn in the respiratory tract (30
) and in other murine models of infection (5
). Williams et al. (30
) also reported that IFN-γ could be detected in lung homogenates of immunologically intact mice as well as the lung homogenates of nude and SCID mice. Treatment of SCID mice with anti-asialo-GM1 was able to reduce the amount of IFN-γ recovered. In murine listeriosis, subcutaneous inoculation of a sublethal dose of Listeria monocytogenes
induced the early appearance of IFN-γ-producing NK cells in the draining lymph nodes (5
). The peak level of NK response occurred at 24 h after infection. Scharton and Scott (24
) also found that NK cells were the source of IFN-γ early in infection of mice with Leishmania major
Of importance in the current study is the observation that NK cells appear locally very quickly in response to a relatively localized infection in the genital tract and may thus act as a first line of defense with respect to the production of IFN-γ. However, recent studies of MoPn infection in mice deficient in IFN-γ receptors or IFN-γ did not show any increase in the number of organisms early in the infection course but rather demonstrated longer duration of infections in comparison to immunologically intact controls (4
). These data coupled with the data in our study would suggest that the primary role of the early IFN-γ production by NK cells is to down-regulate the Th2 response, thereby allowing expression of a strong Th1 response which has been shown to be essential for resolution of the infection in the murine model. However, unfortunately for the host, a number of studies have also suggested that pathologic changes may be associated with the development of a CMI response (15
). Thus, the role of NK cells may have both positive and negative consequences.
Finally, since the data presented here strongly indicate that NK cells are responsible for the production of IFN-γ, one would expect that depletion of NK cells would modify the course of MoPn genital infection. Indeed, when mice were depleted of NK cells by treatment with anti-asialo-GM1, the infection was prolonged compared with that in the controls. In addition, assessment of the IgG subclass response to chlamydial antigen indicated a significant increase in IgG1. In contrast, IgG2a was the dominant antibody in untreated mice. Therefore, these data support those of previous studies in the murine model that demonstrated an important role for CD4 Th1 cells in the resolution of chlamydial genital infection (9
). It was apparent that depletion of NK cells effected an up-regulation of the Th2 response. With an increase in Th2 cells, and thus, an increase in IL-4 and IL-10 production, one would anticipate a down-regulation of the Th1 response. Although the Th2 response was increased, the Th1 response was probably not totally abrogated, based on the presence of IgG2a. Nevertheless, the data indicate that functional NK cells are necessary for optimal clearance of the infection.
The data presented in this study provide new information on the regulation of the Th1 response to chlamydial genital infection and demonstrate an important role of NK cells similar to what has been described for other infectious systems. They further support the significance of key events occurring in the first 96 h of the infection; i.e., the nature of the acquired immune response which develops is dependent upon these early events. Clearly, a critical event for the initiation of the immune response and the eventual outcome of the infection and disease is the initial cytokine and chemokine profile which is elicited upon infection of the host cell by chlamydiae or the exposure of certain cells to bacterial products. In this regard, Rasmussen et al. (22
) have recently demonstrated that infection of endocervical cells by C. trachomatis
can evoke the production of both IL-1 and IL-8 which play important roles in the initiation of the inflammatory response. Moreover, Ingalls et al. (10
) have shown that chlamydial lipopolysaccharide can elicit the production of TNF-α, which is also intimately involved in a variety of roles in the inflammatory response. Regardless of the stimulus for NK targeting of the genital tract infection, NK cells do indeed play a significant role in the reduction of the level of infection early in the disease and contribute to the development of the protective immune response.