Recombinant vaccinia viruses were generated using a pSC11 vector (a kind gift from Dr. J.C. Yewdell, NIAID, Bethesda, MD, USA). These were comprised of a minigene corresponding to a region from the N terminal to the CD8 epitope sequence (39~46 amino acids [aa]) that was cloned into the NotI site of a pSC11 vector to generate rVV-H60₃₉. To generate rVV-H60₃₉+HY-Dby₆₀₈, a minigene containing sequences corresponding to the CD4 epitope of HY-Dby (608~622 aa) with an additional Met at the N-terminal was cloned into the rVV-H60₃₉ vector. Next, 143B cells were transfected with the plasmid described above and TK recombinants were isolated by plaque assay (25). Virus stock was produced in CV-1 cells and the viral titer was determined using BSC-1 cells.

C57BL/6 (B6) mice and H60 congenic mice (B6.CH60) were maintained under specific pathogen free (SPF) conditions at the Center for Animal Resource Development, Seoul National University College of Medicine. CD40-deficient mouse was purchased from the Jackson Laboratory (Bar Harbor, ME, USA) and maintained under SPF conditions at the Center for Animal Resource Development Seoul National University College of Medicine.

Female B6 mice were infected either by intraperitoneal (i.p.) injections with different titers of recombinant vaccinia virus or immunized by i.p. injection with H60 congenic splenocytes (2×10⁷). The immunization studies were reviewed and approved by the Institutional Animal Care and Use Committee (IACUC), Seoul National University.

B6 mice immunized either with rVVs expressing H60 or H60 congenic splenocytes were periodically bled retro-orbitally. Peripheral blood lymphocytes (PBLs) were prepared by red blood cell (RBC) lysis and stained with the following: phycoerythrin (PE)-conjugated H60 tetramer, allophycocyanin (APC)-conjugated anti-CD8 (53-6.7; eBioscience, San Diego, CA, USA), and fluorescein isothiocyanate (FITC)-conjugated anti-CD11a (M174; eBioscience) monoclonal antibodies (mAbs). The PBLs were then placed in fluorescence-activated cell sorter (FACS) buffer (1×phosphate-buffered saline [PBS] with 0.1% bovine calf serum, 0.05% sodium azide) at 4℃ for 30 min. After being washed with FACS buffer, cells were analyzed using a FACSCalibur flow cytometer equipped with Cell Quest software (BD Pharmingen, San Diego, CA, USA).

In order to virally express the H60 epitope, two recombinant vaccinia viruses were constructed (rVV-H60₃₉ and rVV-H60₃₉+HY-Dby₆₀₈). They expressed the H60 determinant under the control of the viral p7.5 early/late promoter and were constructed by cloning the corresponding sequences into a modified pSC11 vector (25). rVV-H60₃₉ expressed H60 from the N-terminal leader peptide to the CD8 epitope sequences (H60_39-46), and rVV-H60₃₉+HY-Dby₆₀₈ expressed the HY-Dby CD4 epitope sequence as well as the H60 epitope (Fig. 1A). Following infection with rVV-H60₃₉, it was expected that CD8 T cells reactive to H60 would arise with help from CD4 T cells which would respond to the CD4 epitopes derived from intrinsic viral proteins (28). Infection with rVV-H60₃₉+HY-Dby₆₀₈ was expected to generate HY-Dby-specific CD4 T cells that would help the H60-reactive CD8 T cells. This was performed in order to identify the specificity of CD4 T cells activated following infection.

We compared the ability of the two different types of rVVs expressing H60 to induce specific CD8 T cell responses. Female B6 mice were i.p. injected with varying doses of the virus (from 0.8×10⁶ PFU to 8×10⁶ PFU). We ascertained the number of H60-specific CD8 T cells in the blood of injected mice by staining PBLs with an H60-tetramer and subsequently performing the flow cytometry analysis. The results from the flow cytometry analysis showed that there were more H60 tetramer-binding CD8 T cells in the blood of B6 mice infected with rVV-H60₃₉+HY-Dby₆₀₈ than those infected with rVV-H60₃₉ (Fig. 1B). We found between 7~13% of H60 tetramer-binding CD8 T cells in the blood of mice infected with 8×10⁵ to 3.2×10⁶ PFU of rVV-H60₃₉+HY-Dby₆₀₈. However, the frequencies ranged between 2~4% of peripheral blood CD8 T cells in mice infected with rVV-H60₃₉. Infection with a higher titer (8×10⁶ PFU) of rVV-H60₃₉ did not increase the number of H60 tetramer-binding CD8 T cells in the blood. These results confirm that infection with both types of virus can induce the CD8 T cell response specific to a virally-originated H60 determinant and demonstrate that CD4 T cells are activated by recognition of virally-expressed HY-Dby CD4 determinant. Despite the fact that the vaccinia virus itself expresses several CD4 target epitopes that originate from the intrinsic viral proteins (28), the results suggest that the HY-Dby CD4 determinant expressed on the recombinant virus might be effective in providing help, perhaps due to the expression of the CD4 determinant as a processed oligopeptide.

We subsequently investigated the kinetics of the immune response induced by infection with rVV-H60₃₉+HY-Dby₆₀₈. Eight female B6 mice were i.p. injected with rVV-H60₃₉+HY-Dby₆₀₈ (1×10⁶ PFU) and periodic retro-orbital blood sample were obtained. The PBLs were stained with the H60-tetramer and subjected to flow cytometry analysis. In order to ascertain whether the infection could generate an H60-specific CD8 T cell response at a comparable level to that induced by H60 from cellular sources, female B6 mice were i.p. immunized with splenocytes from H60 congenic mice. The resulting frequency plot of H60-tetramer-binding cells in peripheral blood CD8 T cells (shown against time) resulted in a kinetic curve similar between the mice that were infected with rVV-H60₃₉+HY-Dby₆₀₈ and were immunized with H60 congenic splenocytes (Fig. 2). Responses peaked on days 10~14 post immunization and waned afterwards in both cases. Even though the peak values were slightly higher in the mice infected with rVV-H60₃₉+HY-Dby₆₀₈ than those immunized with H60 congenic splenocytes (an average of 13% of peripheral blood CD8 T cells compared to an average of 7.5%, respectively), the peak values fell within a normal range (4). This result demonstrates that the immune response induced after infection with rVVs expressing the H60 epitope induced a CD8 T cell response comprised of an initial burst of cell expansion followed by a contraction phase (10).

As the primary CD8 T cell response which is specific for H60 originating from cellular sources is dependent on CD4 T cell help (4,8), we wanted to ascertain whether CD4 T cell help is also needed to induce primary CD8 T cell response specific for a virally-derived H60 epitope. We induced a primary CD8 T cell response to H60 in female B6 mice that were depleted of CD4 T cells following treatment with GK1.5 (an anti-CD4 mAb-depleting GK1.5). This was done by infecting the mice with rVV-H60₃₉+HY-Dby₆₀₈ and analyzing the PBLs periodically via flow cytometry in order to detect any increase in H60 tetramer-binding CD8 T cells. Mice were treated with Rat IgG and infected with rVV-H60₃₉+HY-Dby₆₀₈ as a control. After staining the PBLs (sampled periodically) from the two groups of infected mice, we found that there were very few H60 tetramer-binding cells in the blood of depleted and infected mice during the examination period (days 4 to 28 post-infection). However, there was a significant expansion of H60 tetramer-binding cells in the non-depleted and infected mice between days 7 to 10 (Fig. 3). Based on these results, we conclude that the presence of CD4 T cells is essential for the induction of primary CD8 T cell expansion in response to an infection with rVV-H60₃₉+HY-Dby₆₀₈. This result is consistent with previous findings which show that CD4 T cells are required for a primary CD8 T cell response to infection with recombinant virus encoding hemagglutinin (HA) (23). The results also indicate that CD4 help is needed for the induction of a CD8 T cell response to H60 originated from a vaccinia virus.

It has been shown that help from CD4 T cells is mediated via a CD40/CD40L interaction between CD4 T cells and APCs. The APCs are thought to stimulate the CD8 T cells by presenting peptide MHC I complexes (5-7). We have previously reported the significant roles of CD40 and CD40L in the induction of the CD8 T cell response after immunization with H60 congenic splenocytes (4,8). As the CD8 T cell response to cellular H60 and viral H60 are both CD4 T cell-dependent, we questioned whether the induction of H60-specific CD8 T cell responses after infection with rVV-H60₃₉+HY-Dby₆₀₈ would be impaired in CD40-deficient mice, as was the induction of CD8 T cell response by immunization with H60 congenic splenocytes. In order to answer this question, we challenged CD40-deficient mice and WT B6 mice with rVV-H60₃₉+HY-Dby₆₀₈ and analyzed PBLs periodically to detect H60-tetramer-binding CD8 T cells in the blood. We also immunized mice with H60 congenic splenocytes as a control to test for help-dependent on the CD40-CD40L interaction. The results from the flow cytometry analysis after staining the PBLs with an H60-tetramer showed that a similar expansion of H60-specific CD8 T cells was induced in the both CD40-deficient and WT B6 mice (Fig. 4). The peak frequencies of the H60-tetramer binding cells were 8~10% of peripheral blood CD8 T cells in the CD40-deficient mice infected with rVV-H60₃₉+HY-Dby₆₀₈ and 7~13% in infected WT B6 mice. This finding is in contrast to results from immunization of CD40-deficient mice with H60 congenic splenocytes, where no H60 tetramer-binding cells were detected in the blood. These results show that even though CD4 help is essential for the induction of a CD8 T cell response specific for virally derived H60, CD40 is not an essential factor for this to occur. This suggests that the CD4 help required for the induction of a primary CD8 T cell response might operate differently, depending on whether the antigenic source is cell-based or viral. It has been reported that CD8 T cell responses induced after infection by adenovirus are dependent on CD4 help (29), but not on CD40 (30). It has been shown that CD70-mediated APCs play a significant role in CD8 T cell responses to the ovalbumin adenoviral antigen, in the absence of CD40 (30). It is therefore possible that CD8 T cells enlist CD4 help only for virally expressed H60, and that this response might be mediated by CD70 expressed on dendritic cells in the absence of any CD40-CD40 interaction.

In summary, we report on the generation of a recombinant vaccinia virus expressing an H60 CD8 epitope, as well as on the fact that CD8 T cells require help to react to virally originated H60. We show that the induction of the primary CD8 T cell response specific for H60 after rVV infection has not been impaired in CD40-deficient mice. The results obtained from this study will further our understanding and characterization of CD8 T cell responses specific for antigens derived from cells compared to those derived from pathogens.