The recent identification of tumor Ags as potential vaccines has prompted the search for efficient adjuvants and delivery systems, especially in the case of peptide-based vaccination protocols. Here, we investigated the adjuvant potential of the recombinant 40-kDa outer membrane protein of Klebsellia pneumoniae (P40) for specific CTL induction. We studied the CTL response induced in HLA-A*0201/Kb transgenic mice immunized with peptides derived from two melanoma-associated differentiation Ags, the HLA-A*0201-restricted decapeptide Melan-A26–35 substituted at position 2 and the Kb-restricted tyrosinase-related protein 2181–188 T cell epitope. We found that both peptides are able to generate a specific CTL response when mixed with the protein in the absence of conventional adjuvant. This CTL response is a function of the amount of P40 used for immunization. Moreover, the CTL response generated against the tyrosinase-related protein 2181–188 peptide in presence of P40 is associated with tumor protection in two different experimental models and is independent of the presence of CD4+ T lymphocytes. Thus, the recombinant bacterial protein P40 functions as a potent immunological adjuvant for specific CTL induction.

Many attempts have been made to develop immune-based approaches for cancer therapy. As tumor cells are poorly immunogenic by themselves, the therapeutic use of tumor cells transduced with genes coding for cytokines or costimulatory molecules to enhance in vivo immunity has been demonstrated (1). However, this approach remains problematic because it involves manipulation of patient tumor cells. The identification of CTL-defined tumor-associated Ags has allowed the development of new strategies for cancer immunotherapy.

Two options are mainly investigated. The first is based on the use of professional APCs. Dendritic cells transduced with adenovirus vectors expressing melanoma associated Ag or pulsed with whole tumor lysates or synthetic tumor peptides have been used in successful vaccination approaches in experimental models (2). Although the use of dendritic cells allows the targeting of Ag to lymphoid tissue and the induction of a strong and efficient T cell response, it involves steps of cell purification and culture. In this regard, cell-free vaccines would be more suitable for clinical purposes.

In the case of peptide-based vaccines, the way of delivering Ag to the immune system is critical. It has been shown that a complex mixture of tumor peptides conjugated to beads can prime a CTL response in healthy individuals (3), and lipid compounds have been reported to have adjuvant qualities (4). Yet, the most common adjuvant of T cell responses used to date is IFA. IFA protects immunizing peptides from rapid degradation and enhances their immunogenicity by activating inflammatory processes and costimulatory factors and mediating the production of cytokines. It has been successfully used in immunotherapy against melanoma involving gp100 peptide immunization (5). Although IFA is not associated with severe side effects, this adjuvant is not commonly used in human vaccination protocols due to its undesirable effects, such as erythema and induration at the injection site (5). These effects are partly due to the nonmetabolized mineral oil contained in IFA, and efforts have been made to reduce the oil concentration and to use metabolizable mineral oils (6). In addition, specific CTL tolerance rather than immunity against immunizing peptide in IFA has been reported in an experimental model (7). For these reasons, alternative potent and safe adjuvants need to be identified.

Outer membrane proteins (Omp)2 complexes extracted from meningococcal membranes have the ability to adsorb to peptides via hydrophobic bonds (8). Therefore, we asked whether the 40-kDa major Omp (referred to as P40 hereafter) of Klebsellia pneumoniae, which contains at least one helper epitope (9) can be used to target antigenic peptides to the immune system and elicit a specific CTL response. Peptides from two melanoma-associated differentiation Ags, Melan-A/MART-1 (Melan-A) (10, 11) and tyrosinase-related protein 2 (TRP-2) (12) were used, namely 1) the HLA-A*0201-restricted decapeptide Melan-A26–35 (EAAGIGILTV) substituted at position 2 (Melan-A26–35 A27L peptide analog), which is more immunogenic than the parental peptide both in vitro in human and in vivo in experimental models (13, 14), and 2) the TRP-2 T cell epitope mapping to residues 181–188 (VYDFFVWL) and presented in association with the mouse Kb class I molecule (12). To address the question, we took advantage of transgenic mice expressing a chimeric MHC class I molecule composed of human HLA-A*0201 α1 and α2 domains and mouse Kb α3, transmembrane and cytoplasmic domains in the C57BL/6 × DBA/2 background (15). We found that Melan-A26–35 A27L or TRP-2181–188 peptides are able to generate a specific CTL response when mixed with P40. This CTL response is a function of the amount of P40 used for immunization. Moreover, the CTL response generated against TRP-2181–188 peptide in the presence of P40 is associated with tumor protection in two different experimental models and is independent of the presence of CD4+ T lymphocytes.

Mouse EL-4 cells transfected with the HLA-A*0201/Kb gene (EL-4.A2/Kb transfectants) (16) were provided by Dr. Linda Sherman (The Scripps Clinic and Research Foundation, La Jolla, CA) and maintained in DMEM supplemented with 1% HEPES, 1% strepto-penicillin, 10% heat-inactivated FCS, and 0.5 mg/ml G418. The B16F10 melanoma cell line (17) was provided by Dr. Lars French (University Hospital, Geneva, Switzerland) and maintained in DMEM supplemented with 1% HEPES, 1% strepto-penicillin, and 10% FCS.

Peptides were synthesized by standard solid phase chemistry on a multiple peptide synthesizer (Applied Biosystems, Foster City, CA) using standard F-moc for transient NH2-terminal protection and analyzed by mass spectrometry. All peptides were >90% pure as indicated by analytical HPLC. Lyophilized peptides were diluted in DMSO and stored at −20°C.

All chemicals and solvents from commercial sources were of analytical grade. Acetonitrile was of HPLC grade. K. pneumoniae P40 was expressed as a recombinant protein in Escherichia coli, purified as previously described (9, 18), and will be referred to as P40 thereafter. Purification of P40 was designed in apyrogen conditions. The endotoxin level determined by the Limulus assay was <0.25 endotoxin unit/mg of P40. P40 batches used were produced according to pharmaceutical quality standards, intended for clinical trials in humans. Binding of peptides Melan-A26–35 A27L and TRP-2181–188 to P40 was performed by mixing the respective peptide and P40 in a 7/1 (w/w) ratio in PBS. The final concentrations of peptide and P40 were 0.1 and 0.7 mg/ml, respectively. The mixtures were incubated for 2 h at room temperature under gentle agitation. The P40/peptide complexes were purified by gel filtration on a Superose 12 column (Amersham Pharmacia Biotech, Saint-Quentin en Yvelines, France). Fractions containing P40 were pooled and freeze-dried before analysis by HPLC (Hewlett-Packard, Les Ulis, France). For analysis, the samples were dissolved in 50–100 μl of 10% formic acid and further diluted with 1 vol of HPLC equilibration buffer. Samples were injected on a Vydac C18 column (Hesperia, CA; 150 × 2.1 mm, 5 μm). Peptides complexed to P40 were eluted with an acetonitrile/trifluoroacetic acid gradient.

Breeding pairs of HLA-A*0201/Kb transgenic mice (line 6) (15) were provided by Harlan Sprague-Dawley (Indianapolis, IN). Different protocols of immunization were used. The protocol of immunization in the presence of IFA has been described previously (19). Briefly, HLA-A*0201/Kb transgenic mice were s.c. preinjected, or not, with an emulsion of IFA at the base of the tail. Three weeks later, these mice were immunized s.c. with 50 μg of Melan-A or TRP-2 peptides emulsified in IFA. In experiments with P40, mice were s.c. immunized with 50 μg of Melan-A or TRP-2 peptides mixed with various doses of P40. Ten days after injection, mice were sacrificed, and lymphocyte suspensions from draining lymph nodes were prepared for in vitro stimulation with the relevant peptides.

Lymph node cells (4–5 × 106) were cultured with 2–5 × 105 irradiated (100 Gy) EL-4 A2/Kb cells, prepulsed with 1 μM of the relevant peptides for 1 h at 37°C, in 24-well cell culture plates in 2 ml of DMEM supplemented with 10 mM HEPES, 50 μM β2-ME, 10% FCS, and EL-4 cell culture supernatant containing 30 U/ml of IL-2. After one or more rounds of weekly stimulation, the cultured cells were tested for cytolytic activity.

The cytolytic activity of specific CTL was determined in a 51Cr release assay. Target cells were labeled with 51Cr for 1 h at 37°C in the presence or the absence of the tested peptides, then washed and coincubated with effector cells at the indicated lymphocyte to target cell ratio in V-bottom 96-well plates in a total volume of 200 μl of DMEM. Chromium release was measured in 100 μl of supernatant harvested after 4–6 h of incubation at 37°C. The percentage of specific lysis was calculated as: % specific lysis = (experimental release − spontaneous release)/(total release − spontaneous release) × 100.

C57BL/6 mice were shaved at the flank and injected s.c. with 2.5 × 103 syngeneic B16F10 melanoma cells. Groups of six mice were immunized s.c. at the base of the tail with 300 μg of P40, 50 μg of TRP-2181–188 peptide, a mixture of both, or 50 μg of TRP-2181–188 peptide emulsified in IFA on the same day or 4 days after tumor injection, then boosted 10 days later. When tumors became apparent, their volumes were measured at regular intervals using an electronic caliper. Mice were generally sacrificed when tumors became necrotic or their volume reached 2500–3000 mm3.

On days −3, −1, and +1 relative to the day of B16F10 injection and immunization with a mixture of 300 μg of P40 and 50 μg of TRP-2181–188 peptide, C57BL/6 mice (seven mice per group) were injected i.p. with mAb GK1.5 (anti-CD4) or H35 (anti-CD8). Depletion of CD4+ or CD8+ T cells was monitored by flow cytometry. When tumors became apparent, their volumes were measured at regular intervals using an electronic caliper. Mice were generally sacrificed when tumors became necrotic or their volume reached 2500–3000 mm3.

TRP/T Ag of SV40 (Tag) mice (20) were immunized with PBS, 300 μg of P40, 50 μg of TRP-2181–188 peptide, or a mixture of 300 μg of P40 and 50 μg of TRP-2181–188 peptide at 3 wk of age, then boosted at 4 and 5 wk (five mice per group). The appearance of macroscopically detectable ocular pathology was monitored. Mice with severe eye pathology (eye rupture and bleeding) were sacrificed. In untreated mice this generally occurred 10 days from the first appearance of visible eye pathology.

It has been previously shown that Omp complexes extracted from meningococcal membranes have the ability to adsorb to peptides via hydrophobic bonds (8). Because P40 forms hydrophobic β barrels (21), we investigated the possibility of using mixtures of peptide and P40, rather than chemical conjugate, to prime CTL responses in vivo, as this would considerably simplify the preparation of the vaccine. For that purpose we biochemically characterized the mixture of P40 and Melan-A26–35 A27L. Fig. 1,A shows the elution pattern for the P40/Melan-A26–35 A27L mixture as visualized by gel filtration chromatography with a Superose 12 column. P40 and Melan-A26–35 A27L were detected only in the first peak eluting at 40.7 min, suggesting an association of P40 and peptide. Analysis of pooled fractions containing P40 by SDS-PAGE (data not shown) and reverse phase HPLC on a C18 column (Fig. 1 B) confirmed the coelution of P40 and Melan-A26–35 A27L. Furthermore, the identity of the C18-purified peptide was confirmed by both amino acid analysis and coelution with Melan-A26–35 A27L peptide added to the sample (data not shown). Similar results have been observed for other hydrophobic peptides (data not shown), suggesting that P40 and hydrophobic peptides form relatively stable complexes.

FIGURE 1.

A, Analysis of P40-Melan-A26–35 A27L complex by gel filtration. P40-Melan-A26–35 A27L complex (500 μl) was injected onto a Superose 12 column at a flow rate of 0.4 ml/min. Absorbance was monitored at 206 nm. B, Analysis of Melan-A26–35 A27L associated with P40 by reverse phase HPLC. After purification by gel filtration and freeze-drying, the P40-Melan-A26–35 A27L complex was solubilized in 10% formic acid before HPLC analysis on a Vydac C18 column (150 × 2.1 mm, 5 μm). Melan-A26–35 A27L was eluted with a water-acetonitrile gradient, using 0.1% trifluoroacetic acid in water (solvent A) and 0.1% trifluoroacetic acid in a mixture of water-acetonitrile (20/80, solvent B), at a flow rate of 0.2 ml/min. A quantity of ∼50 μg of purified complex was injected. The column was initially equilibrated at 20% solvent B. The gradient consisted of 20–50% B for 10 min, 50–80% solvent B for 15 min, 80–100% solvent B for 5 min, and isocratic elution at 100% solvent B for 10 min. Before injection the sample was diluted with 1 vol of HPLC equilibration buffer. Absorbance was monitored at 206 nm.

FIGURE 1.

A, Analysis of P40-Melan-A26–35 A27L complex by gel filtration. P40-Melan-A26–35 A27L complex (500 μl) was injected onto a Superose 12 column at a flow rate of 0.4 ml/min. Absorbance was monitored at 206 nm. B, Analysis of Melan-A26–35 A27L associated with P40 by reverse phase HPLC. After purification by gel filtration and freeze-drying, the P40-Melan-A26–35 A27L complex was solubilized in 10% formic acid before HPLC analysis on a Vydac C18 column (150 × 2.1 mm, 5 μm). Melan-A26–35 A27L was eluted with a water-acetonitrile gradient, using 0.1% trifluoroacetic acid in water (solvent A) and 0.1% trifluoroacetic acid in a mixture of water-acetonitrile (20/80, solvent B), at a flow rate of 0.2 ml/min. A quantity of ∼50 μg of purified complex was injected. The column was initially equilibrated at 20% solvent B. The gradient consisted of 20–50% B for 10 min, 50–80% solvent B for 15 min, 80–100% solvent B for 5 min, and isocratic elution at 100% solvent B for 10 min. Before injection the sample was diluted with 1 vol of HPLC equilibration buffer. Absorbance was monitored at 206 nm.

Close modal

We then tested the ability of Melan-A26–35 A27L and TRP-2181–188 peptides mixed with P40 to initiate a CTL response after a single immunization of HLA-A*0201/Kb transgenic mice. As a positive control of CTL induction, we used the same amount of peptides emulsified in IFA and coinjected with a Th epitope, P30, derived from tetanus toxoid 947–967 (22). As a negative control, mice were injected with the same amount of P40 or P30 Th epitope emulsified in IFA in the absence of CTL peptide. CTL specific for Melan-A26–35 A27L peptide generated in vivo could be detected in vitro after two rounds of stimulation when mice were immunized with peptide either mixed with P40 or emulsified in IFA in presence of P30 Th epitope (Fig. 2). A significant CTL response against TRP-2181–188 peptide was observed after one round of in vitro stimulation in mice specifically immunized against peptide in the presence of P40 (Fig. 2). Lymph node cells from mice immunized with the same amount of P40 or IFA plus P30 in the absence of CTL peptide stimulated in vitro in the same conditions did not grow, but some viable cells were maintained in culture for a mean period of 1–2 wk. No CTL activity directed against Melan-A26–35 A27L or TRP-2181–188 peptides was found in cells remaining in these cultures. This suggests that CTLs specific for Melan-A26–35 A27L and TRP-2181–188 peptides detected in vitro have been primed in vivo during immunization with the respective peptides mixed with P40 and do not result from in vitro primary immunization. No CTL response was detected when the same peptides were coinjected with tetanus toxoid alone (not shown), which indicates that the CTL-eliciting activity of P40 cannot be only ascribed to its properties as a carrier protein. Taken together, these data indicate that P40 acts as an adjuvant of the CTL response directed against tumor-associated CTL epitopes.

FIGURE 2.

Specific CTL responses in HLA-A2/Kb transgenic mice immunized with 50 μg of Melan-A26–35 A27L or TRP-2181–188 peptides mixed with 300 μg of P40 or emulsified in IFA plus P30 helper epitope. Control mice were immunized with 300 μg of P40 or IFA plus P30 helper epitope without CTL peptide. Draining lymph node cells from immunized mice were specifically restimulated once or twice in vitro and tested for their cytolytic activity against A2/Kb-transfected EL-4 cells in the presence (▪) or the absence (□) of specific peptide in a chromium release assay. The culture of lymph node cells derived from mice immunized with 300 μg of P40 in the absence of CTL peptide did not contain enough viable cells after two rounds of stimulation in vitro to be tested.

FIGURE 2.

Specific CTL responses in HLA-A2/Kb transgenic mice immunized with 50 μg of Melan-A26–35 A27L or TRP-2181–188 peptides mixed with 300 μg of P40 or emulsified in IFA plus P30 helper epitope. Control mice were immunized with 300 μg of P40 or IFA plus P30 helper epitope without CTL peptide. Draining lymph node cells from immunized mice were specifically restimulated once or twice in vitro and tested for their cytolytic activity against A2/Kb-transfected EL-4 cells in the presence (▪) or the absence (□) of specific peptide in a chromium release assay. The culture of lymph node cells derived from mice immunized with 300 μg of P40 in the absence of CTL peptide did not contain enough viable cells after two rounds of stimulation in vitro to be tested.

Close modal

Part of the activity of P40 is as a carrier protein. Given the epitope suppression observed with certain protein carriers (23, 24), and because humans are often primed to K. pneumoniae OmpA, presumably following natural exposure (25), we investigated whether a pre-existing immune response against P40 could influence its CTL adjuvant potential. For that purpose, we first preprimed mice with either 50 μg of P40 in PBS or IFA/PBS emulsion and immunized them 3 wk later with 50 μg of Melan-A26–35 A27L mixed with 300 μg of P40 or emulsified in IFA, respectively. Control groups received PBS only for prepriming. A significant CTL response specific for Melan-A26–35 A27L was observed after the second round of in vitro stimulation in each group of mice (Fig. 3). There was no correlation between the intensity of the CTL response and the protocol of immunization, indicating that prepriming of mice with P40 does not negatively influence CTL response induction by peptide mixed with P40. The lack of epitope suppression with P40 was also confirmed by measuring the IgG response to a B cell epitope coupled with P40 (not shown).

FIGURE 3.

Specific CTL responses in HLA-A2/Kb transgenic mice preprimed, or not, with 50 μg of P40 or IFA and immunized with 50 μg of Melan-A26–35 A27L mixed with 300 μg of P40 or emulsified in IFA, respectively. Draining lymph node cells from immunized mice were specifically restimulated once or twice in vitro and tested for their cytolytic activity against A2/Kb transfected EL-4 cells in the presence (▪) or the absence (□) of specific peptide in a chromium release assay.

FIGURE 3.

Specific CTL responses in HLA-A2/Kb transgenic mice preprimed, or not, with 50 μg of P40 or IFA and immunized with 50 μg of Melan-A26–35 A27L mixed with 300 μg of P40 or emulsified in IFA, respectively. Draining lymph node cells from immunized mice were specifically restimulated once or twice in vitro and tested for their cytolytic activity against A2/Kb transfected EL-4 cells in the presence (▪) or the absence (□) of specific peptide in a chromium release assay.

Close modal

To define optimal conditions of immunization with P40, mice were immunized with 50 μg of CTL peptide mixed with different amounts of P40. We first measured the number of draining lymph node cells to evaluate the effect of immunization in vivo. As shown in Table I, the number of draining lymph node cells recovered was positively correlated with the quantity of P40 injected. A 2.7- to 3.6-fold increase in lymph node cell number was observed in HLA-A*0201/Kb mice immunized with Melan-A26–35 A27L peptide plus 30 and 300 μg of P40, respectively, compared with that in naive mice. No significant increase in cell yields was observed in mice immunized with Melan-A26–35 A27L peptide plus 3, 0.3, or 0.03 μg of P40 (Table I). In the same manner, a 3-fold increase in cell yield was found in the draining lymph nodes derived from C57BL/6 mice immunized with 300 μg of P40 and 50 μg of TRP-2181–188 peptide compared with that in naive mice (Table I). However, the numbers of cells measured in the lymph nodes derived from mice injected with 300 μg of P40 plus 50 μg of TRP-2181–188 peptide or with 300 μg of P40 alone were not significantly different. This indicates that P40 contributes to the increase in cell yield found in draining lymph nodes independently of the presence of CTL peptide. The same results were obtained in mice injected with IFA in the presence or the absence of CTL peptide (data not shown). Cytofluorometric analysis of draining lymph node cells from immunized mice (P40 plus peptide or P40 alone) did not reveal any significant modification of the percentages of either CD4+ or CD8+ T cells and B cells or in the percentage of activated CD8+ T cells compared with those in naive mice (data not shown). The second parameter investigated was the specific CTL activity of lymph node cells after in vitro stimulation. Whereas no CTL activity was detected after one in vitro stimulation in any protocol of immunization of HLA-A*0201/Kb transgenic mice with Melan-A26–35 A27L peptide, significant levels of peptide-specific lysis were obtained after the second round of in vitro stimulation of lymph node cells from mice immunized with peptide plus 300 μg of P40 (Fig. 4). No CTL activity was detectable in any other group of mice tested. After the third in vitro stimulation, CTL activity could be detected in all groups of mice. Yet, a lower CTL activity was measured in lymphocytes from mice immunized with peptide mixed with 300 μg of P40; this could be attributed to an overstimulation of T cells during in vivo priming in this group of mice. These results indicate that generation of CTL specific for the Melan-A26–35 A27L peptide is strictly dependent on the dose of P40 used as adjuvant. The first dose of 300 μg is enough to consistently detect a CTL response after two rounds of stimulation in vitro, suggesting that the number of CTL precursors activated in this immunization protocol is higher than the number activated with the lower doses of P40 tested.

Table I.

Number of draining lymph node cells isolated from mice immunized with CTL peptide plus P40

MiceP40 (μg)aCTL Peptide (μg)Draining Lymph Node Cells per Mouse (× 10−6)b
HLA-A2/Kb 300 50 22.0 ± 1.0 
 30 50 16.5 ± 0.5 
 50 9.75 ± 1.25 
 0.3 50 7.0 
 0.03 50 7.0 
   6.0 
C57BL/6 300 50 14.2 ± 5.9 
 30 50 3.9 ± 1.2 
 50 5.9 ± 1.7 
 0.3 50 7.0 ± 3.2 
 300  11.3 ± 4.0 
 30  6.4 ± 2.3 
  5.0 ± 3.2 
 0.3  5.0 ± 3.0 
   4.7 ± 0.2 
MiceP40 (μg)aCTL Peptide (μg)Draining Lymph Node Cells per Mouse (× 10−6)b
HLA-A2/Kb 300 50 22.0 ± 1.0 
 30 50 16.5 ± 0.5 
 50 9.75 ± 1.25 
 0.3 50 7.0 
 0.03 50 7.0 
   6.0 
C57BL/6 300 50 14.2 ± 5.9 
 30 50 3.9 ± 1.2 
 50 5.9 ± 1.7 
 0.3 50 7.0 ± 3.2 
 300  11.3 ± 4.0 
 30  6.4 ± 2.3 
  5.0 ± 3.2 
 0.3  5.0 ± 3.0 
   4.7 ± 0.2 
a

HLA-A2/Kb transgenic mice and C57BL/6 mice were immunized s.c. with 50 μg Melan-A26–35 A27L and TRP-2181–188 peptide, respectively, mixed with doses of P40.

b

Mice were sacrificed 10 days later and draining lymph node cells were counted.

FIGURE 4.

Specific CTL response against Melan-A26–35 A27L in HLA-A2/Kb transgenic mice immunized with 50 μg of peptide mixed with doses (300–0.03 μg) of P40 or PBS. Draining lymph node cells from immunized mice were specifically restimulated one, two, or three times in vitro and tested for cytolytic activity against A2/Kb-transfected EL-4 cells in the presence (▪) or the absence (□) of specific peptide in a chromium release assay.

FIGURE 4.

Specific CTL response against Melan-A26–35 A27L in HLA-A2/Kb transgenic mice immunized with 50 μg of peptide mixed with doses (300–0.03 μg) of P40 or PBS. Draining lymph node cells from immunized mice were specifically restimulated one, two, or three times in vitro and tested for cytolytic activity against A2/Kb-transfected EL-4 cells in the presence (▪) or the absence (□) of specific peptide in a chromium release assay.

Close modal

The B16F10 melanoma cell line is a highly aggressive tumor when implanted into a syngeneic host. We chose the dose of 2500 B16F10 cells/mouse following injection of different numbers of B16F10 cells s.c. into C57BL/6 mice (not shown). Of the different doses tested, injection of 2500 cells yielded nearly 100% tumor take with a reasonably long kinetics of tumor growth (∼3 wk from time of tumor injection to maximal tumor volume and sacrifice). This situation (relatively small tumor cell numbers) is an ideal one for testing immunotherapy protocols, because it presents similarities to that of potential human immunotherapy recipients. Indeed, in these subjects tumor mass is generally surgically removed before treatment, so that only residual numbers of tumor cells remain.

A sharp reduction in the growth kinetics of the tumor was observed in mice injected with B16F10 and simultaneously immunized with a mixture of TRP-2181–188 peptide and P40 (Fig. 5,B) or with TRP-2181–188 peptide emulsified in IFA (Fig. 5,C) compared with those in mice immunized with TRP-2181–188 peptide alone (Fig. 5 A). As expected, no effect on tumor growth was observed in mice immunized with P40 or IFA alone (data not shown). This indicates that P40 in combination with a tumor-derived peptide could be a good candidate for therapeutic anti-tumor vaccination.

FIGURE 5.

Compromised tumor growth in mice immunized with tumor-derived peptide and P40 mixture. C57BL/6 mice were shaved at the flank and injected s.c. with 2.5 × 103 syngeneic B16F10 melanoma cells. Groups of six mice were immunized with 50 μg of TRP-2181–188 peptide alone (A), with a mixture of 50 μg of TRP-2181–188 peptide and 300 μg of P40 (B), or with 50 μg of TRP-2181–188 peptide emulsified in IFA (C) simultaneously with B16F10 injection. In A only five lines are shown because one mouse died at the onset of the experiment. When tumors became apparent, their volumes were measured at regular intervals using an electronic caliper. Mice were generally sacrificed when tumors became necrotic or their volume reached about 3000 mm3. Each line represents tumor volume in an individual mouse. Arrows indicate the day of primary immunization relative to the day of tumor inoculation. Similar results were obtained in several separate experiments. Statistical analysis was performed using Student’s t test (∗, p < 0.05; ∗∗, p < 0.01).

FIGURE 5.

Compromised tumor growth in mice immunized with tumor-derived peptide and P40 mixture. C57BL/6 mice were shaved at the flank and injected s.c. with 2.5 × 103 syngeneic B16F10 melanoma cells. Groups of six mice were immunized with 50 μg of TRP-2181–188 peptide alone (A), with a mixture of 50 μg of TRP-2181–188 peptide and 300 μg of P40 (B), or with 50 μg of TRP-2181–188 peptide emulsified in IFA (C) simultaneously with B16F10 injection. In A only five lines are shown because one mouse died at the onset of the experiment. When tumors became apparent, their volumes were measured at regular intervals using an electronic caliper. Mice were generally sacrificed when tumors became necrotic or their volume reached about 3000 mm3. Each line represents tumor volume in an individual mouse. Arrows indicate the day of primary immunization relative to the day of tumor inoculation. Similar results were obtained in several separate experiments. Statistical analysis was performed using Student’s t test (∗, p < 0.05; ∗∗, p < 0.01).

Close modal

To test whether the delay in tumor growth observed following immunization with a mixture of P40 and TRP-2181–188 peptide was the result of the CTL response elicited by this vaccination protocol, and whether this anti-tumor activity was dependent of CD4+ T cell help, we depleted the CD4 or CD8 T cell subsets in vivo in mice that subsequently received both B16F10 cells and a mixture of P40 and TRP-2181–188 peptide. Subset depletion was ascertained by flow cytometry (not shown). As shown in Fig. 6, depletion of CD8+ T cells resulted in complete abrogation of protection (Fig. 6,B) compared with that in nondepleted mice (Fig. 6,A), which, as expected, exhibited very little tumor growth. Interestingly, CD4+ T cell-depleted mice remained protected after immunization with P40 and TRP-2181–188 peptide (Fig. 6 C). These results suggest that the protective effect of immunization with P40 and TRP-2181–188 peptide is completely mediated by CD8+ (presumably cytotoxic) T cells and is independent of the presence of CD4+ (helper) T cells.

FIGURE 6.

Tumor protection following immunization with P40 and peptide mixture is mediated by CD8+ T cells. On days −3, −1, and +1 relative to the day of B16F10 injection and immunization with a mixture of 300 μg of P40 and 50 μg of TRP-2181–188 peptide, C57BL/6 mice (seven mice per group) were injected i.p. with PBS (A), H35 (anti-CD8; B), and GK1.5 (anti-CD4; C). Depletion of CD4+ or CD8+ T cells was monitored by flow cytometry. When tumors became apparent, their volume was measured at regular intervals using an electronic caliper. Statistical analysis was performed using the Kruskal-Wallis test (p < 0.05; ns, nonsignificant).

FIGURE 6.

Tumor protection following immunization with P40 and peptide mixture is mediated by CD8+ T cells. On days −3, −1, and +1 relative to the day of B16F10 injection and immunization with a mixture of 300 μg of P40 and 50 μg of TRP-2181–188 peptide, C57BL/6 mice (seven mice per group) were injected i.p. with PBS (A), H35 (anti-CD8; B), and GK1.5 (anti-CD4; C). Depletion of CD4+ or CD8+ T cells was monitored by flow cytometry. When tumors became apparent, their volume was measured at regular intervals using an electronic caliper. Statistical analysis was performed using the Kruskal-Wallis test (p < 0.05; ns, nonsignificant).

Close modal

To confirm the therapeutic potential of P40 and TRP-2181–188 peptide immunization, we used a transgenic mouse model of ocular tumor formation resulting from TRP-1 promoter-driven SV40 T Ag expression in the retinal pigmented epithelium (RPE) (20). Transformation of RPE cells in these mice starts as early as the fetal stage and continues after birth until the ocular cavity is completely filled with tumor cells. At this time point (∼6 wk of age), the tumor becomes clearly visible macroscopically. Because TRP-2 is expressed by RPE cells and because TRP/Tag mice were backcrossed several times to C57BL/6, we asked whether immunization of these mice with a mixture of P40 and TRP-2181–188 peptide starting at 3 wk of age (with boosts at 4 and 5 wk) would influence the kinetics of eye tumor development. PBS-injected mice or mice inoculated with either TRP-2181–188 peptide alone or P40 alone rapidly developed visible eye pathology (Fig. 7). In contrast, immunization with a mixture of P40 and the peptide resulted in a significant reduction in the number of eyes presenting visible pathology. Moreover, most P40- plus TRP-2181–188 peptide-treated mice that eventually did develop visible eye symptoms showed relatively mild pathology (not shown). These data show that mice treated with a mixture of P40 and peptide several weeks after the onset of the transformation process are still capable of modulating the course of tumor progression, again illustrating the potential of P40 as an adjuvant for immunotherapy of cancer.

FIGURE 7.

P40 mediates a protective immune response against spontaneous tumor in mice. TRP/Tag mice (five mice per group) were immunized with PBS, 300 μg of P40, 50 μg of TRP-2181–188 peptide, or a mixture of 300 μg of P40 and 50 μg of TRP-2181–188 peptide at 3 wk of age, then boosted at 4 and 5 wk. Mice were observed daily, and the appearance of macroscopically detectable ocular pathology was recorded. Mice with severe eye pathology (eye rupture and bleeding) were euthanized. Similar results were obtained in several separate experiments.

FIGURE 7.

P40 mediates a protective immune response against spontaneous tumor in mice. TRP/Tag mice (five mice per group) were immunized with PBS, 300 μg of P40, 50 μg of TRP-2181–188 peptide, or a mixture of 300 μg of P40 and 50 μg of TRP-2181–188 peptide at 3 wk of age, then boosted at 4 and 5 wk. Mice were observed daily, and the appearance of macroscopically detectable ocular pathology was recorded. Mice with severe eye pathology (eye rupture and bleeding) were euthanized. Similar results were obtained in several separate experiments.

Close modal

The use of soluble peptides derived from tumor-associated Ags as immunogens is attractive for the development of antitumoral vaccines, but requires efficient and safe adjuvant formulation. In the present study we show that the 40-kDa major protein of the outer membrane of K. pneumoniae (P40) behaves as a potent adjuvant of the CTL response directed to peptides derived from melanoma-associated Ags. P40/peptide mixtures are immunogenic, which presents the advantage of a simple vaccination procedure. Moreover, immunization against TRP-2181–188 peptide in the presence of P40 is protective in both induced and spontaneous tumor models. This protection is dependent on CTL response and is independent of CD4+ T lymphocytes.

Different compounds with potent adjuvant ability on immune responses have been described. However, only a few of them are able to initiate a CTL response when mixed with soluble Ags and are often toxic (26, 27). Compared with these adjuvants, P40 has the advantages of being a potent adjuvant of the CTL response to immunizing peptides and of being safe. No toxic or associated side effects have been observed in our experimental models or in a standard toxicological survey in animals (data not shown).

A simple mixture of P40 with the immunizing peptide allows the generation of CTL in vivo. P40 is a highly hydrophobic protein in its native form, reflecting its localization in the outer membrane. We have demonstrated that hydrophobic peptides, such as Melan-A26–35 A27L (Fig. 1) and TRP-2181–188 (not shown) peptides, are able to form stable complexes with the protein P40. For several adjuvants, the Ags are incorporated by means of hydrophobic interactions, and the physical association has been demonstrated to be essential for these adjuvants to exert their immunostimulating capacity. As previously described for proteosomes, which are preparations of Omp of meningococci (8), we hypothesize that P40/peptide complexes are also formed by means of hydrophobic interactions of the peptides with the membrane-spanning domain of the protein P40. Finally, the fact that an exogenous helper epitope (such as TT) is not required in the immunizing mixture might be explained by the presence of at least one CD4 T cell epitope in the amino acid sequence of P40 (9), allowing this protein to act as both an adjuvant and a carrier protein. In this context, the observation that preimmunization with P40 does not compromise subsequent immune responses to P40-associated peptides represents an improvement compared with classic carrier proteins, such as TT, which have been shown to elicit epitope suppression (23, 24).

Whereas the main actions of most adjuvants are to form a depot allowing a slow release of Ag in vivo and to induce a strong inflammatory process, their efficiency for CTL induction more likely depends on their capacity to direct soluble Ag to the cytoplasm and class I processing pathway in APCs (28, 29). Different hypotheses have been proposed. For instance, the adjuvant formulation, composed of squalane, Tween 80, and Pluronic L121 (30), when mixed to Ag may result in complexes that are taken up into endosomes of APCs and from there released into the cytoplasm (31). In that case the type of APCs remains unknown. A possible mechanism by which P40 is able to drive a peptide-specific CTL response is suggested by our observation in vitro that P40 selectively binds to professional APC, including dendritic cells, and is internalized by these cells, leading to their activation (32). Experiments are currently in progress in vivo to study the effect of P40 immunization on APC function, and preliminary data indicate the expansion of a subset of CD11c+ cells in the draining lymph node of mice immunized against CTL peptide in the presence of P40. This is of particular interest because dendritic cells provide an excellent costimulatory environment for CTL priming. In this context it is also interesting to note that the protective effect of P40 is independent of CD4+ T cells. Such an independence has been previously reported in immunization protocols, for instance those using Ag-pulsed dendritic cells (33), heat shock proteins (34), or viral pseudoparticles (35), where dendritic cells were efficiently targeted and activated.

The observation that immunization against peptides derived from melanoma Ags in the presence of P40 is associated with tumor protection in vivo reinforces the potential of P40 as an adjuvant in cancer vaccines. Furthermore, we demonstrated the efficiency of TRP-2181–188 immunization in the presence of P40 on tumor regression in a spontaneous tumor model. These data are particularly relevant in cancer treatment in humans, because they show the anti-tumor effect of CTL primed several weeks after the onset of tumor formation. Together, these data indicate that immunization against melanoma Ag-derived peptides in the presence of P40 is efficient enough to prime CTL responses involved in tumor rejection both after tumor challenge and in a spontaneous tumor situation. This is remarkable, especially because the latter is an intraocular tumor, a type particularly resistant to immune-mediated rejection (36).

The demonstration of the adjuvanticity of the protein P40 on the CTL response directed against melanoma Ag-derived peptides both in vitro in a functional assay and in vivo in two different tumor models has important implications for the development of cancer vaccines. Indeed, the facts that no conventional adjuvant is required and that P40/peptide mixtures are effective make the use of P40 a particularly practical immunotherapy approach.

We thank L. Zanna for excellent technical assistance, and L. Goetsch for assistance in setting up the B16F10 tumor model and for statistical analysis.

2

Abbreviations used in this paper: Omp, outer membrane protein(s); TRP-2, tyrosinase-related protein 2; RPE, retinal pigmented epithelium; Tag, T Ag of SV40.

1
Pardoll, D. M..
1995
. Paracrine cytokine adjuvants in cancer immunotherapy.
Annu. Rev. Immunol.
13
:
399
2
Timmerman, J. M., R. Levy.
1999
. Dendritic cell vaccines for cancer immunotherapy.
Annu. Rev. Med.
50
:
507
3
Protti, M. P., M. A. Imro, A. A. Manfredi, G. Consogno, S. Heltai, C. Arcelloni, M. Bellone, P. Dellabona, G. Casorati, C. Rugarli.
1996
. Particulate naturally processed peptides prime a cytotoxic response against human melanoma in vitro.
Cancer Res.
56
:
1210
4
Schmidt, W., M. Buschle, W. Zauner, H. Kirlappos, K. Mechtler, B. Trska, M. L. Birnstiel.
1997
. Cell-free tumor antigen peptide-based cancer vaccines.
Proc. Natl. Acad. Sci. USA
94
:
3262
5
Rosenberg, S. A., J. C. Yang, D. J. Schwartzentruber, P. Hwu, F. M. Marincola, S. L. Topalian, N. P. Restifo, M. E. Dudley, S. L. Schwarz, P. J. Spiess, et al
1998
. Immunologic and therapeutic evaluation of a synthetic peptide vaccine for the treatment of patients with metastatic melanoma.
Nat. Med.
4
:
321
6
Sheikh, K., P. Rajananthanan, W. J. W. Morrow.
1996
. Immunological adjuvants: mechanisms of action and clinical applications.
Exp. Opin. Invest. Drugs
5
:
1079
7
Toes, R. E., R. J. Blom, R. Offringa, W. M. Kast, C. J. Melief.
1996
. Enhanced tumor outgrowth after peptide vaccination: functional deletion of tumor-specific CTL induced by peptide vaccination can lead to the inability to reject tumors.
J. Immunol.
156
:
3911
8
Lowell, G. H., L. F. Smith, R. C. Seid, W. D. Zollinger.
1988
. Peptides bound to proteosomes via hydrophobic feet become highly immunogenic without adjuvants.
J. Exp. Med.
167
:
658
9
Haeuw, J. F., I. Rauly, L. Zanna, C. Libon, C. Andreoni, T. N. Nguyen, T. Baussant, J. Y. Bonnefoy, A. Beck.
1998
. The recombinant Klebsiella pneumoniae outer membrane protein OmpA has carrier properties for conjugated antigenic peptides.
Eur. J. Biochem.
255
:
446
10
Coulie, P. G., V. Brichard, A. Van Pel, T. Wolfel, J. Schneider, C. Traversari, S. Mattei, E. De Plaen, C. Lurquin, J. P. Szikora, et al
1994
. A new gene coding for a differentiation antigen recognized by autologous cytolytic T lymphocytes on HLA-A2 melanomas.
J. Exp. Med.
180
:
35
11
Kawakami, Y., S. Eliyahu, C. H. Delgado, P. F. Robbins, L. Rivoltini, S. L. Topalian, T. Miki, S. A. Rosenberg.
1994
. Cloning of the gene coding for a shared human melanoma antigen recognized by autologous T cells infiltrating into tumor.
Proc. Natl. Acad. Sci. USA
91
:
3515
12
Bloom, M. B., D. Perry-Lalley, P. F. Robbins, Y. Li, M. el-Gamil, S. A. Rosenberg, J. C. Yang.
1997
. Identification of tyrosinase-related protein 2 as a tumor rejection antigen for the B16 melanoma.
J. Exp. Med.
185
:
453
13
Valmori, D., J. F. Fonteneau, C. M. Lizana, N. Gervois, D. Lienard, D. Rimoldi, V. Jongeneel, F. Jotereau, J. C. Cerottini, P. Romero.
1998
. Enhanced generation of specific tumor-reactive CTL in vitro by selected Melan-A/MART-1 immunodominant peptide analogues.
J. Immunol.
160
:
1750
14
Men, Y., I. Miconnet, D. Valmori, D. Rimoldi, J. C. Cerottini, P. Romero.
1999
. Assessment of immunogenicity of human Melan-A peptide analogues in HLA-A*0201/Kb transgenic mice.
J. Immunol.
162
:
3566
15
Vitiello, A., D. Marchesini, J. Furze, L. A. Sherman, R. W. Chesnut.
1991
. Analysis of the HLA-restricted influenza-specific cytotoxic T lymphocyte response in transgenic mice carrying a chimeric human-mouse class I major histocompatibility complex.
J. Exp. Med.
173
:
1007
16
Irwin, M. J., W. R. Heath, L. A. Sherman.
1989
. Species-restricted interactions between CD8 and the α3 domain of class I influence the magnitude of the xenogeneic response.
J. Exp. Med.
170
:
1091
17
Fidler, I. J..
1973
. Selection of successive tumour lines for metastasis.
Nature (London) New Biol.
242
:
148
18
Nguyen, T. N., P. Samuelson, F. Sterky, C. Merle-Poitte, A. Robert, T. Baussant, J. F. Haeuw, M. Uhlen, H. Binz, S. Stahl.
1998
. Chromosomal sequencing using a PCR-based biotin-capture method allowed isolation of the complete gene for the outer membrane protein A of Klebsiella pneumoniae.
Gene
210
:
93
19
Valmori, D., J. F. Romero, Y. Men, J. L. Maryanski, P. Romero, G. Corradin.
1994
. Induction of a cytotoxic T cell response by co-injection of a T helper peptide and a cytotoxic T lymphocyte peptide in incomplete Freund’s adjuvant (IFA): further enhancement by pre-injection of IFA alone.
Eur. J. Immunol.
24
:
1458
20
Penna, D., A. Schmidt, F. Beermann.
1998
. Tumors of the retinal pigment epithelium metastasize to inguinal lymph nodes and spleen in tyrosinase-related protein 1/SV40 T antigen transgenic mice.
Oncogene
17
:
2601
21
Pautsch, A., G. E. Schulz.
1998
. Structure of the outer membrane protein A transmembrane domain.
Nat. Struct. Biol.
5
:
1013
22
Panina-Bordignon, P., A. Tan, A. Termijtelen, S. Demotz, G. Corradin, A. Lanzavecchia.
1989
. Universally immunogenic T cell epitopes: promiscuous binding to human MHC class II and promiscuous recognition by T cells.
Eur. J. Immunol.
19
:
2237
23
Chabalgoity, J. A., B. Villareal-Ramos, C. M. Khan, S. N. Chatfield, R. D. de Hormaeche, C. E. Hormaeche.
1995
. Influence of preimmunization with tetanus toxoid on immune responses to tetanus toxin fragment C-guest antigen fusions in a Salmonella vaccine carrier.
Infect. Immun.
63
:
2564
24
Kaliyaperumal, A., V. S. Chauhan, G. P. Talwar, R. Raghupathy.
1995
. Carrier-induced epitope-specific regulation and its bypass in a protein-protein conjugate.
Eur. J. Immunol.
25
:
3375
25
Bagley, S. T..
1985
. Habitat association of Klebsiella species.
Infect. Control
6
:
52
26
Takahashi, H., T. Takeshita, B. Morein, S. Putney, R. N. Germain, J. A. Berzofsky.
1990
. Induction of CD8+ cytotoxic T cells by immunization with purified HIV-1 envelope protein in ISCOMs.
Nature
344
:
873
27
Heeg, K., W. Kuon, H. Wagner.
1991
. Vaccination of class I major histocompatibility complex (MHC)-restricted murine CD8+ cytotoxic T lymphocytes towards soluble antigens: immunostimulating-ovalbumin complexes enter the class I MHC-restricted antigen pathway and allow sensitization against the immunodominant peptide.
Eur. J. Immunol.
21
:
1523
28
Moore, M. W., F. R. Carbone, M. J. Bevan.
1988
. Introduction of soluble protein into the class I pathway of antigen processing and presentation.
Cell
54
:
777
29
Carbone, F. R., M. J. Bevan.
1990
. Class I-restricted processing and presentation of exogenous cell-associated antigen in vivo.
J. Exp. Med.
171
:
377
30
Raychaudhuri, S., M. Tonks, F. Carbone, T. Ryskamp, W. J. Morrow, N. Hanna.
1992
. Induction of antigen-specific class I-restricted cytotoxic T cells by soluble proteins in vivo.
Proc. Natl. Acad. Sci. USA
89
:
8308
31
Debrick, J. E., P. A. Campbell, U. D. Staerz.
1991
. Macrophages as accessory cells for class I MHC-restricted immune responses.
J. Immunol.
147
:
2846
32
Jeannin, P., T. Renno, I. Miconuet, L. Goetsch, J. P. Aubry, T. Delneste, N. Herbault, T. Baussaut, G. Magistrelli, C. Soulas, J. Y. Bonneloy.
2000
. OmpA targets dendritic cells, induces their maturation and delivery into the MHC class I presentation pathway.
Nat. Immunol.
1
:
502
33
Larsson, M., D. Messmer, S. Somersan, J. F. Fonteneau, S. M. Donahoe, M. Lee, P. R. Dunbar, V. Cerundolo, I. Julkunen, D. F. Nixon, et al
2000
. Requirement of mature dendritic cells for efficient activation of influenza A-specific memory CD8+ T cells.
J. Immunol.
165
:
1182
34
Moroi, Y., M. Mayhew, J. Trcka, M. H. Hoe, Y. Takechi, F. U. Hartl, J. E. Rothman, A. N. Houghton.
2000
. Induction of cellular immunity by immunization with novel hybrid peptides complexed to heat shock protein 70.
Proc. Natl. Acad. Sci. USA
97
:
3485
35
Sedlik, C., G. Dadaglio, M. F. Saron, E. Deriaud, M. Rojas, S. I. Casal, C. Leclerc.
2000
. In vivo induction of a high-avidity, high-frequency cytotoxic T-lymphocyte response is associated with antiviral protective immunity.
J. Virol.
74
:
5769
36
Niederkorn, J. Y..
1997
. Immunoregulation of intraocular tumours.
Eye
11
:
249