The coding sequence for the HIV-1 p24 antigen was amplified from the pHXB2 plasmid (GenBank Accession Number K03455) (18) using the following primers: sense 5′-CGATG^∧CTAGCATGCAGAACATCCAGGGGCA-3′and antisense 5′-GTG^∧AATTCTCAAGCCAAAACTCTTGCCTTA-3′. The amplified fragment was inserted in frame into the pET28b expression vector (Novagen) between the NheI and EcoRI restriction sites (underlined). BL21(DE3) strain E. coli (Invitrogen) were transformed with the expression vector, and after it was carried out screening for positive clones by restriction analysis and DNA sequencing. Induction of the recombinant protein was achieved by addition of isopropyl-β-d-thiogalactopyranoside (IPTG) to the culture for 4 h at 37°C under aeration. The recombinant p24 with an N-terminal histidine tag was purified using a HisTrap™ HP column (GE Healthcare Bio-Sciences Corp). The protein concentration was determined using the BCA assay (Pierce), and purity was monitored in 15% polyacrylamide gel electrophoresis (SDS-PAGE). Endotoxin was removed after successive washing steps with 1% Triton X-114 (19). Endotoxin levels of the preparations, determined by Chromogenic Limulus Amebocyte Lysate assay (LAL) (Cambrex Biosciences), were equal to 0.1 EU of endotoxin per 1 μg of purified protein.

Construction of the pML19, pML21, and pLTK63 plasmids encoding the LT1 (GenBank Accession Number GI 408994), LT2, and LTK63 toxins, respectively, was carried out as previously described (9) and using data available at GenBank for LT1 and showed by Imamura and coworkers (17) and Pizza and coworkers (12). We constructed LTK63, in which serine at position 63 of the A subunit was replaced by lysine, at the background of LT1. The LT2 natural variant has five polymorphic sites in the A subunit (S190L, G196D, K213E, S224T, and N238D) and one in the B subunit (T75A) compared to LT1 (9, 16). Toxin purification by affinity chromatography on immobilized d-galactose columns (Pierce) was described previously (9). Endotoxin levels of the preparations determined by LAL assay were equal or lower to 0.2 EU of endotoxin per 1 μg of purified protein.

Proteins separated on 15% polyacrylamide gels were stained with Coomassie Blue R250 or transferred to nitrocellulose membranes. Immunoblots for the detection of the p24 protein were carried out using standard procedures with anti-HIV p24 rabbit serum (NIH AIDS catalog no. 4250) diluted in 1% bovine serum albumin (BSA) in PBS (block solution).

Determination of the intracellular adenosine cAMP levels was performed using the Y-1 adrenal tumor cell line, as previously described (20). The GM1-ELISA was carried out using standard procedures (21). The binding of LT to mouse Y-1 cells was determined using 1 μg of the tested toxin at 4°C for 1 h in DMEM medium. Treated cells were washed with PBS containing 1% fetal bovine serum and incubated with mouse anti-LT serum. After 30 min, the cells were stained with a FITC-conjugated anti-mouse IgG antibody (Invitrogen) and analyzed by flow cytometry (FACS Calibur, Becton Dickinson, Mountain View, CA, USA). The data were analyzed using the FlowJo program (Tree Star).

The specificity of the antibodies raised in mice immunized with the recombinant HIV-1 p24 protein was determined with A3.01 cells transfected with the pHXB2 plasmid encoding the complete genome of HIV-1 (18). Transfection with the pHXB2 plasmid encoding the complete HIV-1 genome was carried out with polyethylenimine (PEI) (1 mg/mL) (Polysciences) for 30 min at 37°C in a 5% CO₂ atmosphere. After 7–8 days, transfected cells were harvested, added to 96-well plates (Nunc), washed twice with PBS, and fixed with methanol at room temperature. The plates were then incubated with blocking solution (0.2% Tween-20, 3% skim milk in PBS) for 30 min at 37°C before being probed with primary antibodies for subsequent immunofluorescence labeling. The plates were exposed to primary antibodies (sera derived from mice immunized with the recombinant p24 protein plus Complete Freund’s Adjuvant or PBS) diluted to 1:500 for 1 h at 37°C. After two washings with PBS, the plates were incubated with FITC-labeled goat anti-mouse IgG (Invitrogen) plus 1 μg/mL DAPI (Invitrogen) for 1 h at 37°C. Cells were washed twice with PBS, dried at room temperature, and visualized with an immunofluorescence inverted microscope (Nikon) equipped with a digital camera.

BALB/c mice were inoculated with intradermal (i.d.) or subcutaneous (s.c.) injections of 1 μg of LT1, LT2, or LTK63, and the inguinal lymph nodes (ILNs) were removed 24 h after the immunization. Pooled ILNs cells were harvested after tissue homogenization with a glass homogenizer and extensive washing with 2% fetal bovine serum (FBS)/PBS. Cells were plated at a concentration of 5 × 10⁶ cells per well and stained with anti-CD11c and anti-I-A^d (MHC-II), plus anti-H-2K^d (MHC-I), anti-CCR7, anti-CD80, anti-CD86, or anti-CD40 conjugated antibodies (BD Biosciences). Labeled cells were suspended in PBS containing 2% FBS and analyzed (1.5 × 10⁶ events/sample) in a cell cytometer (FACSCanto, Becton, Dickinson and Company). The data were analyzed using FlowJo software (Tree Star) gated on the CD11c⁺/MHC-II⁺ cells. Results were measured in pooled ILNs cells harvested from mice subjected to the different treatments (n = 5 per group). All samples were tested in duplicates, and each tested condition was repeated independently twice.

Specific pathogen-free BALB/c mice (8–15 weeks old) were supplied by the Isogenic Mouse Breeding Facility of the Department of Parasitology, Institute of Biomedical Sciences, University of São Paulo (USP). All animal handling was carried out in accordance with the principles of the Brazilian code for the use of laboratory animals and was approved by the committee on the ethical use of laboratory animals at the Institute of Biomedical Sciences, USP. Mice were immunized with 20 or 100 μL aliquots administered by i.d. or s.c. injections, respectively, on the dorsal surface on days 0, 14, 28, and 42. For the i.d. injections, the mice were partially shaved on the dorsum 1 day before injection. Purified p24 protein was injected at a dose of 20 μg/animal admixed with 1 μg of LT1, LT2, or LTK63. Control groups received sterile PBS or non-adjuvanted p24 protein. Serum samples were collected 1 day before each immunization for ELISA assays. Animals were sacrificed 2 weeks after the last immunization for the determination of the CD4⁺ and CD8⁺ T cell responses and the in vivo CTL activity. The experiments were performed independently three or two times, respectively with n = 5 animals per immunization group as indicated on respective legend.

BALB/c mice were i.d. inoculated with a single dose of 1 μg of LTs per animal/20 μL at previously shaved dorsal caudal area. Local skin reactions were inspected regularly for redness, edema, and swelling for 10 days after injection. The diameter of swelling was measured with a caliper. The experiments were performed independently twice with n = 5 animals per immunization group.

p24-specific ELISA was carried out with 96-well Maxisorp plates (Nunc International) coated with purified p24 protein (5 μg/mL) diluted in PBS at pH 7.5 and incubated overnight at 4°C. Serum samples from immunized mice, anti-HIV p24 rabbit serum (NIH AIDS catalog no. 4250) or serum from an HIV-infected individual (NIH AIDS catalog no. 192) was serially diluted (twofold) in blocking solution and added to the wells for 1 h at 37°C. Reactions were performed with diluted horseradish peroxidase-conjugated rabbit anti-mouse immunoglobulin (total IgG, IgG1, or IgG2a) (Sigma and Southern Biotech), anti-human IgG antibody (Southern Biotech), or rabbit anti-IgG antibody.

Secreted cytokines were measured using the ELISA BD OptEIA™ kit (BD Biosciences) with the supernatants of cultured spleen cells harvested from immunized mice. The cells were stimulated with a synthetic peptide (GenScript) corresponding to the I-A^d CD4⁺ T cell-specific (AMQMLKETINEEAAE) epitope of the HIV-1 p24 protein at a final concentration of 5 μg/mL for 48 h at 37°C in a 5% CO₂ atmosphere. After stimulation, the supernatants were collected for determination of the IFN-γ and IL-5 concentrations. All samples were tested in duplicate, and each immunization experiment was repeated independently twice.

The IFNγ-ELISPOT was performed essentially as described earlier (22). Spleen cells collected from immunized animals were diluted in complete media, plated at a concentration of 5 × 10⁵ cells per well, and incubated for 24 h at 37°C in a 5% CO₂ atmosphere in medium with and without the synthetic peptide corresponding to the major H-2K^d CD8⁺ T cell-specific (AMQMLKETI) p24 epitope at a final concentration of 2.5 μg/mL. All samples were tested in duplicate, and each immunization experiment was repeated independently three times.

The in vivo CTL responses were determined as previously described (22). Briefly, splenocytes collected from naive BALB/c mice were stained with 0.5 μM (CFSE^low) or 5 μM (CFSE^high) carboxyfluorescein diacetate succinimidyl ester (CFSE) (Invitrogen Molecular Probes). The CFSE^high cells were pulsed for 40 min at 37°C with 1 μM of the synthetic p24 CD8⁺ T cell-specific peptide. Subsequently, aliquots containing equal amounts of CFSE^low and CFSE^high cells (2 × 10⁷ cells) were intravenously injected into mice previously immunized with the different vaccine formulations and naïve control mice. Twenty hours later, the numbers of CFSE^low and CFSE^high cells were determined in the spleens of the recipient mice using flow cytometry. The percentage of specific cell lysis was determined for each animal using the formula 100 − [(% CFSE^high in immunized mice/% CFSE^low in immunized mice)/(% CFSE^high in naïve mice/% CFSE^low in naïve mice)] × 100%. All samples were tested in duplicate, and each immunization experiment was repeated independently twice.

One-way ANOVA followed by Tukey’s HSD test was used to compare the differences between the mean values of the immunization groups. Differences with p ≤ 0.05 were considered statistically significant. The statistical analyses were performed with the animals belonging to each immunization group (n = 4–5) with independent experiments.

A soluble recombinant form of the HIV-1 p24 protein was generated in bacterial cells purified by nickel affinity chromatography (Figure 1A). The recombinant protein retained the antigenicity of the native viral protein, as demonstrated in Western blots and ELISAs developed with reference serum samples collected from a rabbit immunized with purified p24 protein and an HIV-infected subject (Figures 1B,D). The recombinant protein p24 were at least 95% pure (Figure 1C) and contained 0.1 EU of endotoxin per microgram of purified protein. Additionally, anti-p24 antibodies raised in mice immunized with the recombinant p24 recognized the native virus protein expressed by A3.01 cells transfected with a plasmid vector encoding all virus proteins (Figure 1E).

The three LT derivatives (LT1, LT2, LTK63) were expressed and purified from recombinant E. coli strains. The recombinant toxins displayed similar electrophoretic motility, binding to purified GM1 receptor and Y-1 cells (Figures 2A,C,D). In addition, LT1 and LT2 displayed similar enzymatic activity, as demonstrated by the increase in intracellular cAMP in treated Y-1 cells (Figure 2B). Mice inoculated by the i.d. route with LT1 and LT2 developed intense local inflammatory reactions measured by edema formation and induration at the inoculation site as early as 24 h post administration and persisted for more than 10 days. In contrast, mice inoculated with LTK63 induced a slight swelling reaction at the injection site detected at 48 h after inoculation and vanished 8 days later (Figure 2E). Similar reactions were observed in mice inoculated via the s.c. route (data not shown).

We also tested the in vivo effects of the LTs on the behavior of CD11c⁺/MHC-II⁺ dendritic cells (DCs) after administration via the s.c. and i.d. routes. As shown in Figure 3, 24 h after a single i.d. administration of the tested LTs, the numbers of DCs at the ILNs were significantly higher among LT-treated mice than PBS-treated animals. Mice inoculated with the non-toxic LTK63 showed a lower number of CD11c⁺/MHC-II⁺ cells in the draining lymph nodes than mice inoculated with LT1 or LT2 (Figure 3A). Similar results were observed in mice immunized via the s.c. route (data not shown). However, no significant difference in the expression of CCR7, a marker associated with DC migratory behavior (23), was observed in mice treated with the LTs (Figure 3B). We also measured the in vivo activation of CD11c⁺/MHC-II⁺ cells in ILNs after inoculation of the different LTs. Although no significant increase in the expression of MHC-I and CD40 (Figures 3C,D) was detected in mice treated with the LT derivatives, CD11c⁺/MHC-II⁺ cells harvested from mice inoculated with LT1 or LT2 showed an enhanced expression of CD80 and CD86 compared to the control group (Figures 3E,F). Similar results were also observed in mice inoculated via the s.c. route (data not shown).

Titers of p24-specific serum IgG were increased in mice immunized via s.c. and i.d. routes with the recombinant p24 protein admixed with LT1, LT2, or LTK63. No statistically significant differences were observed in the anti-p24 antibody titers in mice treated with LT1 or LT2 following either i.d. or s.c. immunization (Figures 4A,B). However, lower anti-p24 IgG titers were detected in mice immunized with i.d. injections of LTK63 compared with mice immunized with LT1 or LT2 (Figure 4A). Analysis of the IgG subclass responses elicited in LT-vaccinated mice showed that, after four vaccine doses, mixed IgG1/IgG2a response was observed among animals regardless of the route of immunization. However, higher IgG1/IgG2a ratios were detected in mice immunized via s.c. route compared to mice immunized via the i.d. route (Figures 4C,D).

Immune response evaluations of cytokine profiles showed that splenic CD4⁺ T cells of mice immunized with i.d. injections of the LT derivatives secreted similar levels of IFN-γ and undetectable amounts of IL-5 (Figure 4E). Nonetheless, following s.c. vaccination, mice immunized with LTK63 showed higher levels of IFN-γ secretion by CD4⁺ T cells than mice treated with LT1 or LT2, whereas differences in the IL-5 levels were not observed between any of the immunization groups (Figure 4F). IL-4 was not detected in the culture supernatants of splenic cells from mice vaccinated via either the i.d. or the s.c. route (data not shown). Additionally, all LT forms increased IFN-γ levels compared with p24 administered alone, following either i.d. or s.c. immunization. These results, together with the IgG subclass findings, show that LT derivatives modulate the anti-p24 response toward a more biased Th2 profile when administered via the s.c. route compared to mice immunized via the i.d. route.

Adjuvant effects on CD8⁺ T cells were measured in spleen cells harvested 2 weeks after the last immunization dose. IFNγ-ELISPOT results showed that mice immunized with the LT derivatives induced strong activation of p24-specific CD8⁺ T cells, whereas no significant response was detected in mice immunized with non-adjuvanted p24. Mice immunized with LT2 by the i.d. route showed statistically higher CD8⁺ T cell responses compared to mice immunized with LT1 or LTK63 (Figure 5A). In contrast, mice immunized with LT2 via the s.c. route showed similar CD8⁺ T cell responses compared to mice immunized with LT1 (Figure 5B). The results also showed that the LT derivatives enhanced in vivo CTL responses after administration either i.d. or s.c. immunization. Nonetheless, mice immunized with LTK63 mounted significantly lower in vivo CTL responses than mice immunized with LT1 and LT2 (Figures 5C,D).