6–8 weeks old BALB/c. H-2^d, C57BL/6J H-2^b, FVB/J H-2^q and CBA/J H-2^k were obtained from breeding facilities of National Institute of Immunology, New Delhi. Each experimental group consisted of 4–6 animals.

MDP analog (MA729) was the product of CDRI (Central Drug Research Institute).

1. V3-gp41 – combination of V3 loop (308–322) and gp41 (593–604) sequence bridged with two glycine motifs.

(LGLWGCSGKLIC GG KRIHIQGPGRAFVT)

------gp41------   ------V3 loop------

2. gp41 (728–753) – (DRPEGIEEEGGERDRDRSIRLVNGSL).

3. p17 (114–136) – (KKAQQAAADTGNRGNSSQVSQNY).

4. p24 (287–307) – (QGPKEPFRDYVDRFYKTLLRA)

All the peptides were synthesized using t-Boc chemistry on PAM resin by the method of Merrifield 14. The resulting peptides were purified using HPLC and characterized by amino acid composition and homogeneity of the peptides were assessed by various analytical techniques. All the peptides were found to be > 95% pure.

Liposome's were prepared according to our reported protocol 15. Briefly multilamellar vesicles (MLV's) were prepared composed of phosphatidyl choline, cholesterol and phosphatidyl glycerol taken in molar ratio of 7:4:1. MLV's were probe sonicated to small unilamellar vesicles (SUV's). Peptide or peptide with adjuvant at 5 mg/ml was freeze dried with SUV's. After rehydration and suspending in PBS, liposomes were pelleted at 100,000 × g. The amount of peptide antigen entrapped within liposomes was measured using I¹²⁵ peptide. The percent efficiency was found to be in the range of 25–30%.

ISCOMs were prepared according to our reported protocol 15. Peptides were palmitified before entrapment into ISCOMs using the reagent N-Palmitoyloxy succinimide 16. ISCOMs containing peptide antigens were pelleted by centrifugation on sucrose gradient. The amount of peptide incorporated in the ISCOMs was estimated using colorimetric assay 17. The efficiency of incorporation was found to be 40–50%.

The size, shape and morphology of the preparations were studied by TEM studies.

V3-gp41, gp41, p17 and p24 peptide(s) alone adsorbed on alum or incorporated in liposomes (with or without adjuvant MA729) or in ISCOMs were used as immunogens. Each peptide was studied in all the four groups. Dose of 25 μg of peptide adsorbed on alum and 5 μg in liposomes was standardized to study the humoral response in all the four haplotypes. For every dose, four mice were used in a group for every haplotype studied. Whenever MA729 was included in the formulation, the amount of adjuvant i.e. 50 μg (optimal dose) was kept constant with 5 μg of all the peptide formulations. Mice were immunized in the footpad on days 0, 21, 35 and 42. Bleeds were collected on days 27(Ist), 42(IInd) and 60^th (IIIrd). The third bleed was taken without a booster dose. Amounts of individual peptides in cocktail mixture followed identical immunization schedule as described above.

Peptide specific antibodies were detected at fixed antisera dilution (1:100) as well as by measuring the peak (end point) titers. The peak titers were expressed and assessed as the reciprocal of sera dilution which gave an absorbance > 0.2 (i.e. ± 4 standard deviations from the mean values of pre immune sera).

IgG subclasses were determined using the isotyping kit. Plates were coated overnight with 100 ng/well of peptides. Usual ELISA procedure was followed using goat anti-mouse IgG subclass monoclonal antibodies (Sigma isotyping kit) at 1:1000 dilutions for IgG1, IgG2a, IgG2b or IgG3. The respective peptide mice antiserum was used at 1:100 dilutions. Pre-immune sera were used as negative control.

The binding affinity of antibodies of different antigenic formulations were determined by measuring the K_D (dissociation constant) value using the reported protocol 18 in high and low responder strains of mice. Peptide antigen at various molar concentrations (1 × 10^-7 M to 1 × 10^-10 M) was incubated at a fixed dilution 1:500 of the respective peptide-antiserum. The unbound antibodies were then determined using an indirect ELISA. Scatchard and Klotz mathematical equation was used to calculate the K_D value. A_O / (A_O-A) = 1 + K_D / a_O, A_O = Absorbance without antigen, A = Absorbance with free antigen. a_O= Free antigen concentration. The equation permits the determination of dissociation constant even if specific antibody concentration is not known. The equation is of a straight line and the slope determines the K_D value.

Human immunodeficiency virus (HIV) plaque assay was done with CD4⁺ expressing MT-2 cell line as described 19. This assay permits or works with anti-envelope antibodies and not with the core protein (p17 and p24) antibodies. All sera were heat inactivated before use. Virus and sera to be tested were diluted in 50% assay medium (85 % DMEM with 15% fetal bovine serum, 2 μg of polybrene) and 50% normal human plasma pools (NHPP). MT-2 cells were added to each well and incubated for seven days at 37°C in 5% CO₂ and stained with propidium iodide. Representative peptide antisera that showed peak titers in a given haplotype with a given formulation were used for the HIV microplaque assay. The neutralizing titer of each serum sample was expressed as the inverse of the greatest dilution of serum, which resulted in 50% inhibition in the average plaque count relative to the count in the control wells. The percent inhibition at each dilution was calculated using the mathematical equation was expressed as percent control.

Peptide binding assays was done to check the crossreactivity of antiserum with different Clades or isolates of HIV. ELISA plates were coated with avidin subsequently blocked and then was incubated with 50 ng/well biotinylated peptides. Four fold-diluted antiserum was used from 1:100 to 1:6000. Titer of the antisera was defined as reciprocal of the dilution, which gave a mean absorbance of 0.5.

To study the antigen induced T cell stimulation and to optimize the day of maximal T cell response V3-gp41 peptide adsorbed on alum was used to immunize mice of haplotype H-2^d. 25 μg of peptide adsorbed on alum was immunized on day 0, boosted with half of the primary dose on 8^th day and mice were sacrificed separately on 12^th and 15^th day. Separate groups of mice also received a booster on 12^th day followed by sacrifice on 15^th and 20^th day.

V3-gp41, gp41, p17 and p24 peptides were separately adsorbed on alum and immunized in four different mice of haplotypes H-2^d, H-2^b, H-2^k and H-2^q with 25 μg of peptide dose in primary immunization and 12.5 μg for subsequent boosters. A cocktail peptide approach was also studied in a high responder H-2^b and a low responder H-2^k with 25 μg of peptide equivalent immunized in alum, with half the primary dose for booster immunization.

Mice spleen cells obtained after immunization with respective peptides in alum were in vitro stimulated with peptide alone or peptide entrapped in liposomes (with or without MA729) or peptide entrapped in ISCOMs as studied for four different strains. In another set of experiment V3-gp41 peptide was immunized along with MA729 and the spleen cells obtained were in vitro stimulated with peptide alone, peptide entrapped in liposomes (with or without MA729) or in ISCOMs. Moreover to look for the mitogenic activity of MA729, mice were primed with the adjuvant and the spleen cells were in vitro stimulated with MA729 (alone or in liposomes) and with V3-gp41 peptide alone or entrapped along with MA729 in liposomes. T cell proliferation was also studied using cocktail mixture of peptides adsorbed in alum and in vitro stimulated with respective peptides alone or entrapped in liposomes (with or without MA729) or in ISCOMs. Cells were cultured in triplicate and Conconavalin A 5 μg/ml was used as a positive control. Negative control consisted of unstimulated cells containing the media alone. Cells were cultured for 72h in a humidified chamber with at 37°C, 5% CO₂. 50 μl culture supernatants were collected for estimation of IL-2, IL-4 & IFN-γ. The cells in each well were pulsed with 0.5 μCi of ³H (6.7 Ci/m mole) and incubated for 18 hrs. Thymidine incorporation was measured in a liquid scintillation counter (Wallac B-counter, counting efficiency 65%).

Stimulation index was calculated as:

The culture supernatants collected during antigen induced T-cell proliferation assay were used to measure cytokine levels (IL-2, IL-4 & IFN-γ). Culture supernatants were taken from wells pulsed with the optimal antigen concentration, which showed the maximal stimulation index, as assayed for different peptides. The culture supernatants obtained were centrifuged at 5,000 rpm for 20 min and filtered through 0.22 μM Millipore membranes and were quantitated using duo set ELISA kits (Genzyme Corp., USA) as per the manufacturer's instructions.

Cytokines IFN-γ, IL-2 and IL-4 were measured in the culture supernatants obtained after priming with cocktail mixture in alum and in vitro stimulation with respective peptides alone or entrapped in liposomes (with or without MA729) or in ISCOMs.

Levels of statistical significance (P) values for the difference in total IgG (in the sera raised against different antigen preparations) were performed by Non-Parametric Kruskal-Walli's one-way analysis of variance. For calculating the P value in respect of difference in isotypes, Non-Parametric Friedman's two way ANOVA with multiple range tests between the different formulations and Kruskal-Walli's one-way ANOVA between different IgG subclasses were applied. Levels of statistical significance between antibody levels in the first, second, third and fourth (bleeds except otherwise stated) were determined by Wilcoxson's signed rank test. The difference between the T cell proliferation and cytokine levels of peptide alone, peptide in liposomes (with or without adjuvant) and in ISCOMs were also analyzed using Non-Parametric Kruskal-Walli's one way analysis of variance by ranks.

TEM studies revealed that the size of the ISCOMs and liposomes to be in the range of 40 nm and 300 nm respectively in diameter.

The amount of peptide antigen required to produce optimum immune response was standardized using V3-gp41 peptide and was found to be 25 μg, for first immunization and 12.5 μg for subsequent boosters for peptide adsorbed on alum (data not shown). For liposome and ISCOM preparations, the optimum dose that showed peak antibody levels was found to be 5 μg. Adjuvant combination of 5:50 μg dose of peptide: MA729 elicited highest antibody response, in a dose kinetic study. The above doses were fixed for subsequent immunizations with other peptide antigens in different haplotypes.

Sera obtained from of mice of different genetic background with synthetic peptides in different formulations were assayed for peptide specific antibodies. The results show that all the peptides in alum were immunogenic in one or more of the strains of mice depending upon the nature of the formulation used. V3-gp41 and gp41peptides produced high titer antibodies and the peak titers sustained even in third bleed without booster immunization while the antibody titers for p17 and p24 peptides were moderate in all the strains (Fig. 1A,1B,1C,1D,1E,1F,1G,1H,1I,1J,1K,1L,1M,1N,1O,1P).

Highest sustained antibody titers (>60 days) were observed in all the strains with peptide entrapped alongwith MA729 in liposomes and were statistically significant (P < 0.05) as compared to the other formulations.

For V3-gp41 peptide, the antibody titres obtained for mice bearing the haplotype H2^b and H2^d were higher as compared to mice bearing the haplotype H-2^k and H-2^q. Peptide: MA729 in liposomes showed the highest antibody titres in all haplotypes studied (Fig. 1A,1B,1C,1D). It is interesting to note that the peptide itself showed comparable peak titres of 1/102,400 in both liposome and ISCOM preparation except in secondary bleed of mice bearing the haplotype H-2 ^d&k with peak titers of 1/64,000. Secondary immune response for V3-gp41 peptide in liposomes (with or without adjuvant) generated similar peak titer of 1/128,000 in mice bearing the haplotypes H-2^b. Comparable peak titers were observed for V3-gp41peptide entrapped in liposomes and ISCOMs except in the third bleed of mice bearing the haplotype H-2^d&k. The antibody response in liposomes was 2–4 folds higher than the ISCOM based preparation. Although, liposome and ISCOM preparations showed slight strain-to-strain variation in antibody titers, the difference however was statistically insignificant.

Peak antibody response with gp41 peptide was significantly higher (p < 0.01) in mice bearing the haplotype H-2^d&k as compared to haplotype H-2^b&q irrespective of the nature of the formulation used (Fig 1E,1F,1G,1H). The antibody response in liposome preparation for gp41 peptide together with adjuvant was highly significant (P < 0.01) with peak titers of 1/102,400 for all the haplotypes especially in the third bleed, when compared to alum preparation with peak titers of 1/32,000. The titers were significantly higher in liposomes and ISCOM preparation for all the haplotypes and in all the bleeds as compared to alum based formulation. Although all the haplotypes responded equally well, mice bearing haplotypes H-2^b and H-2^k showed the highest and lowest antibody titers respectively among the four haplotypes studied.

P17 and p24 peptide in alum showed peak titers of 1/6,400 and 1/3,200 respectively (Fig. 1I,1J,1K,1L,1M,1N,1O,1P). All the p17 peptide formulations were immunogenic, but mice bearing the haplotype H-2^q showed maximum peak titers of 1/51,200 followed by haplotypes H-2^k (1/32,000), H-2^b and H-2^d(1/16,000) (Fig 1I,1J,1K,1L). High antibody titers were obtained when the p17 antigen was entrapped in liposomes (1/32,000) or ISCOMs (1/51,200). Addition of adjuvant MA729, improved the immunogenicity of the formulations with antibody levels statistically significant (P < 0.01) for p17 peptide in all the haplotypes. For p24 peptide, mice bearing the haplotypes H-2^k,b&d showed peak titers of 1/25,600 in liposomes containing adjuvant, but mice bearing the haplotype H-2^q failed to show similar response (Fig 1M,1N,1O,1P). Unlike V3-gp41 and gp41 peptide, the antibody response for p17 and p24 peptide was significantly lower and the pattern of antibody response obtained for different delivery systems was more or less same. Peptide (5 μg) in liposome (with or without adjuvant) and ISCOMs elicited antibody response, which was 3–4 fold higher in magnitude as compared to alum, based preparations.

Humoral response to mixture of peptides containing equivalent of V3-gp41, gp41, p17 and p24 in different formulations were studied in high (H-2^d) and low responder strain (H-2^k) of mice (Fig. 2A,2B,2C,2D). When peptide specific antibodies were measured, delivery in alum showed lower antibody response as compared to other adjuvant formulations. For V3-gp41 peptide booster response was observed for all the formulations and was maintained till third bleed (Fig. 2A). gp41 peptide in liposome (with or without adjuvant) or in ISCOMs did not induce significant boosting effect in both the haplotypes however the levels of antibodies were maintained for all the formulations till the third bleed (Fig. 2B). p17 peptide showed booster response for both liposomes (with or without MA729) and ISCOMs formulations in both the haplotypes studied (Fig. 2C). Irrespective of the strain and the formulation, V3-gp41 specific antibodies were found to be predominant followed by p17, gp41 and p24 anti-peptide antibodies.

Very low levels of anti p24 antibodies were observed for all the formulations in both the haplotypes. The levels of antibody for peptide entrapped in liposomes (with or without adjuvant) and in ISCOMs were statistically significant (P < 0.05) as compared to antibody response in alum formation.

Mice immunized with peptide or peptide formulations predominantly generated IgG2a/IgG2b isotypes and the levels were comparable in all the bleeds and strains. The inclusion of adjuvant in liposomes formulation showed higher levels of IgG2a/2b subclass when compared to other formulations (Table 1).