in vivo. Normalized data values for amplicon-directed Gag expression in cultured murine bone marrow dendritic cells (bmDC; data from Figure 2E) are plotted against the magnitude of the Gag-specific T cell response in vivo, as measured by intracellular cytokine staining for IFN-γ in CD8+ cells (data from Figure 6B) or IL-2 in CD4+ cells (data from Figure 6C). A correlation analysis was performed initially, which yielded a correlation coefficient (Pearson r value) of 0.982 (panel A; p <0.0001) or 0.961 (panel B; p=0.0001). The data were then fitted to a linear regression model, which yielded the plots shown; the R² (goodness of fit) values for these plots were 0.965 (panel A) and 0.923 (panel B).

HSV-1 amplicon constructs encoding HIV-1 Gag from seven different promoters (Figure 1) were inoculated into the tail base of BALB/c mice (4 mice/group), at a dose of 5 x 10⁵ transducing units. The injection site was marked, and 24 h later, the injection site was recovered with a tissue punch. A tissue lysate was prepared, and subjected to Bradford assay in order to determine the protein concentration. A fixed amount of lysate was then used in a HIV-1 p24 ELISA; the results are expressed as ng of HIV-1 p24 protein per ml of reaction volume, where 1 ml of the reaction contained 2.5 μg of tissue extract. The data shown represent mean values of triplicate measurements; bars denote the standard deviation of those values. The results shown are representative of two independent experiments that yielded very similar data.

Seven helper virus-free HSV-1 amplicon vectors encoding HIV-1 Gag were developed, each with a different promoter. Promoters that were studied included: the major human cytomegalovirus immediate-early promoter [CMV], and several hybrid promoters derived from this element - including a CMV promoter linked to the CMV Intron A sequence [CMV-I], a modified CMV promoter containing regulatory sequences from the human T-cell leukemia virus type-1 LTR [CMV-R] and a composite CMV/chicken ß-actin promoter [CAG], as well as two retroviral promoters (the Rous sarcoma virus [RSV] and the myeloproliferative sarcoma virus [MPSV] long terminal repeat [LTR]). The Herpes Simplex Virus Type 1 immediate early 4/5 promoter [HSVie] was also used. Each of the indicated Gag expression cassettes was then inserted into the polylinker region of the parental amplicon plasmid.

BALB/c mice (3/group) were inoculated in the tail base with 5 x 10⁵ transducing units of untreated or UV-inactivated amplicon particles containing the (CMV-R) driven Gag insert (Figure 1). Animals were sacrificed 10 days later, and splenocytes were harvested. Gag-specific T cell responses were then assayed by intracellular cytokine staining for IFNγ (CD8⁺ T cells) following stimulation of cells using HIV-1 Gag peptide pools. The data represent mean values of triplicate measurements; bars denote the standard deviation of those values. Statistical analysis was performed by a nonparametric one-way ANOVA test (GraphPad Prism). A statistically significant reduction in the Gag-specific CD8⁺ T cell response was observed in mice inoculated with UV-treated Gag-encoding amplicons (CMV-R, UV-treated), versus mice immunized with untreated Gag-encoding amplicons (CMV-R) (*, p < 0.01). The data values for mice immunized with UV-treated Gag-encoding amplicons were not statistically different from those for mice immunized with a negative control amplicon (HSV-lacZ; p > 0.05).

HSV-1 amplicon constructs encoding HIV-1 Gag from seven different promoters (Figure 1) were inoculated in the tail base of BALB/c mice (4 mice/group), at a dose of 5 x 10⁵ transducing units. Animals were sacrificed 10 days later, and splenocytes were harvested. Gag-specific T cell responses were then assayed by (A) staining with a H-2K^d AMQMLKETI tetramer and by intracellular cytokine staining for (B) IFNγ (CD8⁺ T cells) and for (C) IL-2 (CD4⁺ T cells) following stimulation of cells using HIV-1 Gag peptide pools. The data represent mean values of triplicate measurements; bars denote the standard deviation of those values. Statistical analysis was performed by a nonparametric one-way ANOVA test (GraphPad Prism). Statistically significant T cell responses, compared to mice immunized with irrelevant amplicon particles (HSV-lacZ), are denoted by asterisks, as follows: *, p<0.05; **, p<0.01; ***, p<0.001. Results shown are representative of two independent experiments that yielded very similar data.

BALB/c mice (3/group) were inoculated in the tail base with 5 x 10⁵ transducing units of UV-inactivated amplicon particles containing the (CMV-R) driven Gag insert (Figure 1). Groups of mice were then sacrificed at timed intervals following amplicon inoculation (5 min, as well as 3, 6 and 24 h), and the injection site was recovered using a tissue punch. A tissue lysate was prepared, and protein content in each lysate was determined by Bradford assay. A fixed amount of lysate was then used in a HIV-1 p24 ELISA; the results are expressed as ng of HIV-1 p24 protein per ml of reaction volume, where 1 ml of the reaction contained 2.5 μg of tissue extract. The data shown represent mean values of triplicate measurements, and bars denote the standard deviation of those values; the sensitivity cutoff of the Gag ELISA is 10 pg/ml, and samples in which the measured Gag level was below this were assigned a value of 10 pg/ml. A nonlinear curve for a single-phase exponential decay was fitted to the data using GraphPad Prism (and is shown in the Figure). The R2 value (goodness of fit) for this curve was 0.88, and the in vivo half-life of the preformed Gag antigen present within the HSV-1 amplicon inoculum was estimated at 2.84 hours (with 95% confidence intervals ranging from 1.73 to 7.94 hours).

Murine bone marrow-derived dendritic cells (bmDC; panels A and E), NIH 3T3 cells (a murine fibroblast cell line; panels B and F), XB2 cells (a murine keratinocyte cell line; panels C and G) and EMT6 cells (a murine mammary epithelial cell line; panels D and H) were transduced with HSV-1 amplicon constructs encoding HIV-1 Gag from seven different promoters (as indicated in Figure 1) at an MOI of 1. Twenty-four hours later, cell lysates were prepared, and protein content within the lysates was determined by Bradford assay. A fixed amount of each lysate was then loaded onto a SDS/PAGE gel for subsequent immunoblot analysis with a Gag-specific monoclonal antibody (panels A–D), or used in a HIV-1 p24 Gag ELISA (panels E–H). For the immunoblot assays shown in panels A–D, the cell lysates were also probed with an anti-α-tubulin antibody (as a loading control). The p24 values in panels E–H represent mean values of triplicate measurements; bars denote the standard deviation of those values. The results shown (panels A–H) are representative of two independent experiments that yielded very similar data. Note that, due to the limited number of wells available on the SDS/PAGE gels, the lacZ control sample on all of the immunoblots was run on a different gel than the other samples; this is denoted by the vertical line on each of the four immunoblot panels. Results (panels E–H) are expressed as ng of HIV-1 p24 protein per ml of reaction volume, where 1 ml of the reaction contained 2.5 μg of cell extract.