In this study, we used rSV40 vectors to elicit HIV-1NL4-3 gp120-specific cytotoxic lymphocyte and antibody responses. We have observed that these vectors may be administered repeatedly to boost those responses. Further studies also suggested that such responses are durable in vivo. Our results here demonstrate several important strengths of using rSV40 vectors to immunize against lentiviral antigens: Among these are the ability of the vector to be administered multiple times without eliciting neutralizing responses [11–13], and the magnitude of the cytotoxic responses to the vector-encoded lentiviral target antigen. When SV(gp120) was given alone, i.e. without added SV(mIL-15), levels of specific cytotoxicity increased with additional SV(gp120) injections: After 2 injections, ≈ 20% specific lysis was seen, which increased to >70% specific lysis after 7 injections. The potency of rSV40 immunization to elicit cytotoxic immune responses is underscored by the fact that these responses were measured in direct 51Cr-release assays: unselected lymphoid organ populations were added directly to labeled target cells at low E:T ratios, and specific 51Cr release was measured.
Analysis to confirm CD8 expression, or expression of other CTL markers was not performed on the effector cells. However, it is unlikely that these data reflect the cytotoxic activity of NK cells. NK cytolytic activity is non-specific and does not increase with repeated immunization. The patterns of 51Cr release observed in the current studies were extensively controlled to ascertain the antigen-specificity of the cytolysis observed: background lysis of wild type P815 cells was subtracted, as was lysis by lymphocytes from mice immunized with an irrelevant rSV40 vector. We also found that cytolysis increased with increasing numbers of immunizations, which is not a characteristic of NK cell-mediated lysis.
Since a key goal for a vaccine against HIV is to generate immune responses that are durable in vivo, we tested whether cytotoxic lymphocyte activity elicited by SV(gp120) immunization, was detectable one month after inoculation. Thus, cytolytic responses, assayed one month after a second injection, were ≈ 20%, which is comparable to those of splenic cytotoxic cells assayed four days following a third inoculation (data not shown). Further, mice given multiple injections of SV(gp120), then rested for one year, gave ≈ 70% specific lysis when challenged with SV(gp120). Therefore, SV(gp120) administration may thus favor development of cytotoxic lymphocyte memory.
In an attempt to accelerate and to improve upon these specific cytotoxic and particularly cytotoxic memory responses, we co-immunized mice with SV(gp120) and a rSV40 carrying mouse IL-15. IL-15 promotes cytotoxic lymphocyte responses, and in particular, cytotoxic memory responses [23, 24]. The biological effects of IL-15 are less well understood than are those of some of the other immunostimulatory cytokines that have been applied to these types of immunization protocols, such as IL-2, IL-12 and IFN-γ. IL-15 is not a T cell-derived product, but rather appears to be produced by a variety of cells, such as epithelial cells, stromal cells and muscle. It acts on activated T cells, sometimes similarly to IL-2, but it has activities distinct from those of IL-2. IL-15 may play a role in T cell activation in the CNS. It also promotes cytotoxic responses, cytotoxic T cell memory, and natural killer (NK) cell maturation [33, 34].
Accordingly, our analysis of the contribution by IL-15 to cytotoxic responses, focused mainly on the ability of SV(mIL-15) to augment specific cytotoxic responses of spleen cells from animals rested 1 month following immunization. Because quiescent cytotoxic T cells are not strong effectors, we non-specifically activated the splenocytes prior to assay with Con A. Non-specific activation was used to avoid specifically enriching effector cell population for gp120-specific cells in vitro. Furthermore, low effector:target ratios (20:1 and 10:1) were used in these assays. Our immunization protocols tested both simultaneous and staggered administration of rSV40s carrying HIV-1NL4-3 gp120 and murine IL-15.
IL-15 co-immunization dramatically accelerated cytotoxic responses, depending on the immunization regimen used: Animals given SV(mIL-15) alone made no gp120-specific cytolytic responses. Mice receiving 2 treatments with a mixture of SV(gp120) and SV(mIL-15) gave much higher specific lysis, depending on the coadministration regimen, as compared to those receiving SV(gp120) alone (≈ 10% specific lysis). Thus, among mice given staggered injections of SV(mIL-15) and SV(gp120), the order of cytokine administration greatly affected the response: if SV(gp120)was given first, no detectable gp120-specific cytolysis was observed. However, if the cytokine was given first, followed 3 days later by SV(gp120), ≥ 60% specific lysis was seen at both 20:1 and 10:1 effector:target ratios.
Why the order of cytokine administration should affect antigen-specific responses so dramatically is not yet clear. Cytokine given after, or together with antigen, may have insufficient time to augment cytotoxic responses. In addition, Western analysis of IL-15 production showed that IL-15 secretion was not detectable in supernatants beyond 36 hours, but could be stimulated subsequently. Thus, a specific, possibly brief, window for IL-15 expression and secretion may need to be attained, in order for its effects on gp120-specific responses to be detectable. We observed very high levels of specific lysis by these unselected effector populations following just two tandem injections of SV(mIL-15) followed by SV(gp120).
The strong anti-lentiviral cytolytic responses we report were observed in a strain of mouse, BALB/cJ, that generally mounts relatively weak type 1 T cell responses. The finding of >60% cytolysis with two administrations of SV(gp120) + SV(mIL-15), suggests that a strategy similar to that described herein may be helpful in individuals who would generate relatively low cytolytic responses.
Serum antibody levels assayed by CELISA where SV(gp120) was administered alone, multiple times, were detectable after two immunizations, and continued to increase up to week 4 following the third immunization. These responses were not further enhanced by subsequent boosting immunizations. While specific antibody responses against gp120 were detected in all experimental groups following SV(gp120) and SV(mIL-15) co-immunization, IL-15 co-administration did not augment anti-gp120 antibody levels, compared to gp120 alone. This was to be expected, since IL-15 reportedly acts primarily on T cell and NK cell functions, rather than on humoral immune responses.
Our data argue in favor of using IL-15 as an adjuvant for antigen-specific immune responses, particularly cytotoxic lymphocyte responses. We also demonstrate that a single transgene, administered multiple times (>3), may be very effective at eliciting both humoral and cell-mediated responses. These results thus both corroborate and extend our previous observations [13, 14, 32], and suggest that combining rSV40s encoding antigens and immunostimulatory cytokines sequentially in multi-administration regimens may provide high levels of long-lasting immunity against the target antigen.