Despite remarkable progresses in vaccinology, therapeutic cancer vaccines have not achieved their full potential. We previously showed that an excessively long duration of Ag presentation critically reduced the quantity and quality of vaccination-induced T cell responses and subsequent antitumor efficacy. In this study, using a murine model and tumor cell lines, we studied l-tyrosine amino acid–based microparticles as a peptide vaccine adjuvant with a short-term Ag depot function for the induction of tumor-specific T cells. l-Tyrosine microparticles did not induce dendritic cell maturation, and their adjuvant activity was not mediated by inflammasome activation. Instead, prolonged Ag presentation in vivo translated into increased numbers and antitumor activity of vaccination-induced CD8+ T cells. Indeed, prolonging Ag presentation by repeated injection of peptide in saline resulted in an increase in T cell numbers similar to that observed after vaccination with peptide/l-tyrosine microparticles. Our results show that the duration of Ag presentation is critical for optimal induction of antitumor T cells, and can be manipulated through vaccine formulation.

Immunotherapy is a potent modality in the treatment of several cancers, thanks to the major success of immune checkpoint blockade therapy with anti-CTLA4 and anti-PD1/PD-L1 mAbs. Immune checkpoint blockade potentiates pre-existing tumor-specific T cell responses to mediate tumor destruction (1). However, many tumors induce insufficient spontaneous T cell responses, a limitation that can potentially be overcome by anticancer vaccination. Unfortunately this approach has yet to deliver robust therapeutic efficacy (2, 3). With recent advances in the personalized identification of tumor Ags (i.e., neoepitopes derived from mutated gene products) (4) and better understandings of vaccine adjuvants (i.e., delivery systems and immunopotentiators), new avenues are open for more potent therapeutic cancer vaccines (5). For example, Gubin et al. (6) showed in a murine model that immunization using tumor neoantigens (peptides) was as effective as checkpoint blockade. Recently, personal neoantigen vaccines were demonstrated to be safe and effective in treating patients with high-risk melanoma (7, 8). Exciting results from these studies provide a strong rationale for cancer vaccine as a stand-alone treatment or in combination with checkpoint blockade or other immunotherapies.

We have previously shown that for peptide-based cancer vaccines, the choice of Ag delivery system can affect the ensuing antitumor immune response (9). Long-lived water-in-oil emulsions of IFA, which greatly prolong Ag presentation time through a long-lived Ag depot function, diminished therapeutic efficacy when used as adjuvant for short antigenic peptides. Specifically, tumor-specific CD8 T cells became sequestered at the persisting, Ag-rich vaccination site, where they underwent apoptosis without reaching the tumor. Although a very short-lived, water-based formulation showed no T cell sequestration and consequently improved antitumor activity, T cell responses were weaker, possibly because Ag was cleared too rapidly to allow maximal T cell priming. We therefore hypothesized that Ag delivery systems can be created to extend Ag presentation time sufficiently long to allow induction of an optimal T cell response, but not so long as to induce T cell sequestration at the vaccination site. In this study, we report that microparticles consisting of the poorly soluble amino acid l-tyrosine are a promising peptide Ag delivery system for the induction of potent antitumor immune responses. Mechanistically, l-tyrosine functioned as a short-lived depot, extending the Ag presentation time during a critical window for optimal T cell priming. Interestingly, this effect could be largely mimicked by repeated injections of peptide in saline, thus suggesting a simple strategy for increasing the potency of peptide-based anticancer vaccines. Overall, our results point to duration of Ag presentation as a critical factor in vaccine-induced T cell priming, which can be controlled by proper choice of vaccine adjuvant.

The synthetic, high-affinity H-2Db–restricted heteroclitic mouse gp10025–33 peptide (KVPRNQDWL) and H-2Kb–restricted chicken OVA-I257–264 peptide (SIINFEKL) were purchased from Peptides International (Louisville, KY) at a purity >95%. Optima-grade acetonitrile, methanol, and water were purchased from Thermo Fisher Scientific (Waltham, MA). Mass spectrometry–grade formic acid (Fluka; 98%) was purchased from Sigma-Aldrich (St. Louis, MO). 1× PBS was purchased from Mediatech (Manassas, VA). Sodium hydroxide (molecular biology grade) and hydrochloric acid (36.5% v/v) were purchased from Fisher Scientific (Fair Lawn, NJ). Sodium chloride, sodium phosphate monobasic, sodium phosphate dibasic, and l-tyrosine (cell culture grade) were purchased from Sigma-Aldrich. A mouse cytokine/chemokine Milliplex kit (catalog no. MXMCY70KPX25MGBK) was purchased from EMD Millipore (Burlington, MA). OVA-I dextramer H2-Kb was purchased from Immudex (Fairfax, VA). Aluminum hydroxide gel, referred to as alum, was purchased from Invitrogen (Carlsbad, California).

All mouse protocols were reviewed and approved by the Institutional Animal Care and Use Committee at the University of Texas MD Anderson Cancer Center. Pmel-1 TCR transgenic mice on C57BL/6 background (The Jackson Laboratory, Bar Harbor, ME) were crossed with CD90.1 congenic mice to yield pmel-1+/+CD90.1+/+ mice (hereafter referred to as pmel-1 mice). C57BL/6 mice were purchased from Charles River Laboratories (Wilmington, MA). B16-F10 melanoma-bearing mice were established by injecting 300,000 B16 cells in a volume of 0.1 ml s.c. Tumor-bearing mice received treatments on day 6 after tumor injection when the average tumor size was ∼30 mm2. ASC knockout mice were a gift from Dr. T.-D. Kanneganti at St. Jude Children’s Research Hospital.

B16-F10 melanoma cell line, from American Type Culture Collection (Manassas, VA), was cultured in complete medium including RPMI 1640 with 10% FBS, 100 μg/ml streptomycin, and 100 μg/ml penicillin (Life Technologies, Carlsbad, CA).

Peptides in saline and IFA were prepared as previously described (9). Preparation of peptide with l-tyrosine was adapted from a protocol for grass pollen/l-tyrosine described by Wheeler et al. (10). A peptide solution was prepared by dissolving 4 mg of peptide in 2 ml of 1× PBS. After dissolution, 0.667 ml of strong PBS (0.83 M Na2HPO4, 0.25 M NaH2PO4, 0.137 M NaCl) was added to the solution. Next, 0.667 ml each of 3.2 M sodium hydroxide and 1.3 M l-tyrosine in 3.9 M hydrochloric acid were added simultaneously and mixed, resulting in a final solution volume of 4 ml. The suspension was centrifuged and supernatant was discarded. The remaining peptide/l-tyrosine pellet was dissolved in PBS to make up a volume of 4 ml and ready for injection. Final peptide concentration was ∼0.25 mg/ml. Each mouse received 50 μg of peptide in 200 μl of vaccine (100 μl at two vaccination sites). For the quantification of tumor-specific CD8+ T cells, pmel-1 splenocytes were i.v. transferred to C57BL/6 mice on the same day with vaccination. A mixture of anti-CD40 mAb (clone FGK4.5/FGK45; Bio X Cell, West Lebanon, NH), IL-2 (TECIN; Hoffman LaRoche), and imiquimod 5% cream (Fougera, Melville, NY) (covax), was given on the same day with peptide vaccination. An anti-CD40 dose (50 μg s.c.), IL-2 (100,000 IU, once on day 0 and twice on days 1 and 2 i.p.), and imiquimod cream (50 mg) were applied topically on the vaccine site.

Unless specified, each mouse received ∼80,000 naive pmel-1 CD8 T cells from pmel-1 donor mouse via i.v. tail vein injection. For in vivo Ag detection experiments, pmel-1 CD8 T cells were purified using a CD8 T cell enrichment kit (Stemcell Technologies, Vancouver, BC, Canada) and then labeled with CFSE as described elsewhere (11). Each mouse received 2 × 106 CFSE-labeled pmel-1 CD8 T cells i.v.

gp100-Specific CD8 T cell responses of mice receiving pmel-1 T cells were detected by basing on congenic Thy1.1 (CD90.1). Endogenous gp100- and OVA-I–specific CD8 T cell responses were detected by IFN-γ and OVA-I dextramer, respectively, using flow cytometry.

Mice were tail bled on the indicated days. Extracellular staining was performed using FACS buffer containing 2% FBS. Intracellular cytokine staining was performed using the Cytofix/Cytoperm kit from BD Biosciences (San Jose, CA) based on the manufacturer’s recommendations. Granzyme B staining was done without stimulation whereas IFN-γ staining was done after 4 h of stimulation with 1 μM gp10025–33 peptide. Abs were either purchased from eBioscience or BD Biosciences: CD8a (clone 56-6.7), CD4 (GK1.5), CD90.1 (HIS51), IFN-γ (XMG1.2), TNF-α (MP6-XT22), granzyme B (NGZB), CD19 (eBio1D3), CD3e (145-2C11), NK1.1 (PK136), CD44 (IM7), B220 (RA3-6B2), CD11b (M1/70), CD11c (N418), F4/80 (BM8), CD62L (MEL-14), CD27 (A7R34) MHC class II (M5/114.15.2), CD40 (HM40-3), CD86 (GL-1), Ly6G (1A8), Ly6C (AL-21).

On days 1, 2, 3, and 7 postvaccination, skins at the vaccine site were depilated, weighted, mechanically disrupted in ice-cold PBS (1 ml per sample), and centrifuged for supernatant collection. The cytokines/chemokines in the supernatant were measured using a Milliplex mouse cytokine/chemokine panel (Millipore) according to the manufacturer’s instructions. Fluorescence signal was measured on a Luminex 100/200 system, and data were analyzed using Excel software. Final cytokine/chemokine readouts were normalized by sample weight.

After the peptide/l-tyrosine coprecipitation (as described in the vaccination section), the final volumes of the supernatant and crystal fractions were determined to be 2.85 and 1.15 ml, respectively. The individual fractions were stored at 4°C until analysis. Peptide stock (2.49 mg/ml) and intermediate (100 μg/ml) solutions were prepared in water and were stored at 4°C until analysis. The intermediate solution was used to prepare calibration standards at 50.0, 25.0, 10.0, 2.00, and 1.00 μg/ml concentrations in water. Prior to sample processing, the peptide-loaded particle and supernatant fractions were warmed to room temperature. The peptide-loaded l-tyrosine particles contained in the crystal fraction were dissolved by an addition of 4 ml of formic acid followed by gentle vortex mixing. Once the particles were completely dissolved, an additional 1.88-ml aliquot of water was added to the sample to increase the final sample volume to 7.00 ml. In prior to analysis, three individual sample dilutions were prepared at 10×, 50×, and 100× in water.

Sample analysis was performed on an Agilent 1290 Infinity binary ultra-HPLC system coupled to an Agilent 6460 tandem mass spectrometer. Mobile phase A (MPA) and mobile phase B (MPB) used for this study were 0.1% formic acid in water and 0.1% formic acid in acetonitrile, respectively. The chromatographic column used was an Agilent Zorbax RRHD Eclipse Plus C18 (2.1 × 50 mm, 1.8 μm; dead volume, ∼0.12 ml; dead time, ∼0.60 min at 0.200 ml/min). The column was heated to 40°C, and the chromatographic flow rate was 0.200 ml/min. The gradient elution program was set as follows: dwell at initial conditions of 90:10 MPA/MPB for 1.5 min postinjection; ramp to 20:80 MPA/MPB at 4.0 min postinjection; ramp to 0:100 MPA/MPB at 5.0 min postinjection; and ramp back to initial conditions (90:10 MPA/MPB) at 5.5 min postinjection until the gradient stops at 6.5 min postinjection. The overall cycle time for a single injection was ∼7 min. The mass spectrometer acquisition source parameters were as follows: source: Agilent Jet Stream ESI source; gas temperature, 275°C; gas flow, 6 l/min; nebulizer, 40 ψ; sheath gas temperature, 325°C; sheath gas flow, 9 l/min; capillary voltage, +3750 V; nozzle voltage. 0 V. The molecule-specific acquisition parameters were as follows: precursor to product transition, m/z 490.n2 to m/z 848.4; MS1 and MS2 were set to unit resolution; dwell time, 250 ms; fragmentor voltage, 100 V; collision energy voltage, 10 V; cell acceleration voltage, 7 V; and the source polarity was set to positive mode.

All results are expressed as mean ± SEM (group size was at least three mice per group). Tumor challenge experiments had group size of 10 mice per group. Data were analyzed using an unpaired t test where p < 0.05 is considered as significant. Survival experiments used a log-rank (Mantel–Cox) test for survival analysis. All experiments were repeated at least once with comparable results.

l-Tyrosine is an amino acid with poor water solubility and the capacity to adsorb macromolecular grass and tree allergens during pH change-induced flash-precipitation (Supplemental Fig. 1A, 1C) (10). In vivo, injected l-tyrosine microparticles dissolve over a period of several days, and l-tyrosine–formulated macromolecular grass and tree pollen allergens have been used in human allergen desensitization vaccines (10, 12). In this study, we tested whether l-tyrosine could also be used as a short-term slow release formulation for short (9 aa) gp10025–33 (hereafter referred to as gp100) antigenic peptide. After coprecipitation of l-tyrosine and gp100 peptide (hereafter referred to as gp100/l-tyrosine), we determined the gp100 peptide content in the resulting microparticles (Supplemental Table I). Approximately 25% of initial gp100 peptide input was reproducibly retained in the gp100/l-tyrosine formulation. We previously showed that covax (a molecularly defined adjuvant consisting of agonistic anti-CD40 Ab, the TLR7 agonist, imiquimod, and IL-2) could remarkably improve vaccination-induced CD8+ T cell responses (9). We therefore vaccinated mice with gp100/l-tyrosine and covax and measured the resulting gp100-specific pmel-1 T cell response. gp100/l-Tyrosine induced superior T cell numbers in peripheral blood compared with the equivalent dose (50 μg per mouse) of gp100 in saline formulation (Fig. 1A). Functionally, l-tyrosine induced similar IFN-γ but significantly less granzyme B production by pmel-1 T cells (Fig. 1B, 1C). Three hundred days later, gp100/l-tyrosine–vaccinated mice contained significantly more memory T cells than did mice vaccinated with gp100/saline. T cells in both groups displayed a central memory (CD62LhiCD127hi) phenotype (Fig. 1D). Noticeably, peptide formulated in either the d- or l- optical isomer of tyrosine induced similar T cell responses, suggesting the ability of l-tyrosine to form microparticles was more important for its vaccine adjuvant activity than its possible pharmacological activity (Supplemental Fig. 2A). In the absence of covax, gp100/l-tyrosine induced a very modest pmel-1 T cell response (Supplemental Fig. 2B), confirming the importance of including specific APC-activating (TLR7 agonist and agonistic anti-CD40 mAb) and T cell survival signals (IL-2) for inducing a robust T cell response after vaccination with synthetic peptides.

FIGURE 1.

l-Tyrosine is a potent vaccine adjuvant for the induction of CD8+ T cell responses. (A) Pmel-1 T cell level as a percentage of CD8+ T cells in the blood of mice at different time points. (B) IFN-γ and (C) granzyme B production by pmel-1 T cells. (D) Memory phenotype of pmel-1 T cells on day 300 postvaccination. Mice received 8 × 105 pmel-1 T cells and indicated treatments on day 0. Peptide dose was 50 μg per mouse. Data are shown as the mean ± SEM. Statistical differences between the two groups were determined by Student t test. n = 5 mice per group per experiment. Data are representative of three independent experiments. *p < 0.05. ns, not significant.

FIGURE 1.

l-Tyrosine is a potent vaccine adjuvant for the induction of CD8+ T cell responses. (A) Pmel-1 T cell level as a percentage of CD8+ T cells in the blood of mice at different time points. (B) IFN-γ and (C) granzyme B production by pmel-1 T cells. (D) Memory phenotype of pmel-1 T cells on day 300 postvaccination. Mice received 8 × 105 pmel-1 T cells and indicated treatments on day 0. Peptide dose was 50 μg per mouse. Data are shown as the mean ± SEM. Statistical differences between the two groups were determined by Student t test. n = 5 mice per group per experiment. Data are representative of three independent experiments. *p < 0.05. ns, not significant.

Close modal

Successful vaccination requires the induction of local inflammation, resulting in recruitment and activation of APC such as dendritic cells (DC). To understand the contribution of these processes in gp100/l-tyrosine–induced T cell immunity, we characterized chemokine production and leukocyte recruitment at the cutaneous vaccination site. l-Tyrosine microparticles did not induce upregulation of costimulatory molecules such as CD86 and CD40 on DC (Fig. 2A, 2B). Instead, l-tyrosine induced particularly high expression of the granulocyte/neutrophil attractant CXCL1, as well as G-CSF, which was accompanied by increased recruitment of neutrophils to the vaccination site (Fig. 2C–F). To determine whether this neutrophil recruitment contributed to superior CD8 T cell priming by l-tyrosine microparticles, we depleted neutrophils with an Ly6G-specific Ab. We found no difference in gp100-specific CD8 T cell responses between Ly6G-depleted and control groups (Fig. 3), suggesting that although the particulate nature of l-tyrosine vaccine triggered the influx of neutrophils, these did not contribute to the enhanced T cell priming.

FIGURE 2.

Local inflammatory response to l-tyrosine–based vaccination. l-Tyrosine did not induced CD86 and CD40 upregulation on DC at vaccine-draining LN (VdLN) (A and B) but induced chemoattractant CXCL1 and G-CSF at vaccine site (C and D). (E and F) l-Tyrosine induced massive neutrophil (CD11b+, MHC class II, F4/80, Ly6G+) infiltration to vaccine site. Statistical differences between the two groups were determined by the unpaired two-tailed t test. n = 3–4 mice per group per experiment. Data are representative of two independent experiments (mean ± SEM). *p < 0.05, **p < 0.01. ns, not significant.

FIGURE 2.

Local inflammatory response to l-tyrosine–based vaccination. l-Tyrosine did not induced CD86 and CD40 upregulation on DC at vaccine-draining LN (VdLN) (A and B) but induced chemoattractant CXCL1 and G-CSF at vaccine site (C and D). (E and F) l-Tyrosine induced massive neutrophil (CD11b+, MHC class II, F4/80, Ly6G+) infiltration to vaccine site. Statistical differences between the two groups were determined by the unpaired two-tailed t test. n = 3–4 mice per group per experiment. Data are representative of two independent experiments (mean ± SEM). *p < 0.05, **p < 0.01. ns, not significant.

Close modal
FIGURE 3.

Neutrophils do not contribute to T cell priming after l-tyrosine vaccine. Anti-Ly6G Ab was given 2 d prior to vaccination and every 3 d afterward. (A) Pmel-1 T cell levels in the blood after indicated treatments are shown. The depletion of Ly6G population was confirmed in peripheral blood (B) and the skin at vaccine sites (C) on day 4 after vaccination. Mice received 8 × 105 pmel-1 T cells and indicated treatments on day 0. n = 5 mice per group. Data are representative of two independent experiments.

FIGURE 3.

Neutrophils do not contribute to T cell priming after l-tyrosine vaccine. Anti-Ly6G Ab was given 2 d prior to vaccination and every 3 d afterward. (A) Pmel-1 T cell levels in the blood after indicated treatments are shown. The depletion of Ly6G population was confirmed in peripheral blood (B) and the skin at vaccine sites (C) on day 4 after vaccination. Mice received 8 × 105 pmel-1 T cells and indicated treatments on day 0. n = 5 mice per group. Data are representative of two independent experiments.

Close modal

Because particulate materials have previously been shown to exert vaccine adjuvant activity through inflammation mediated by activation of the inflammasome (13, 14), and the recruitment of neutrophils is a strong indicator of inflammation (15), we determined whether the adjuvant activity of l-tyrosine was mediated by the inflammasome. We measured gp100-specific pmel-1 T cell responses after l-tyrosine vaccination in wild-type and genetically inflammasome-deficient ASC knockout mice. We found no difference in T cell response among wild-type and genetically inflammasome-deficient ASC knockout mice (Fig. 4A). In addition, vaccination with gp100 peptide mixed with “empty” l-tyrosine microparticles (preparation shown in Supplemental Fig. 1B) did not improve T cell response, as opposed to vaccination with coprecipitated gp100 peptide and l-tyrosine (Fig. 4B). Thus, the vaccine adjuvant activity of l-tyrosine required coprecipitation of peptide and l-tyrosine into mixed microparticles, and l-tyrosine microparticles did not demonstrate any vaccine adjuvant activity even when coinjected with free peptide Ag, as would be expected if the microparticles induced local inflammation. Taken together, the activity of l-tyrosine did not appear to be mediated by inflammasome activation or direct activation of APC.

FIGURE 4.

l-Tyrosine adjuvant activity does not require inflammasome activation. (A) No difference in T cell responses between wild-type (WT) and ASC knockout (KO) groups was found, regardless of vaccine formulations. (B) l-Tyrosine adjuvant activity requires peptide coprecipitation. Mice received 8 × 105 pmel-1 T cells and indicated treatments on day 0. n = 5 mice per group. Data are representative of three independent experiments.

FIGURE 4.

l-Tyrosine adjuvant activity does not require inflammasome activation. (A) No difference in T cell responses between wild-type (WT) and ASC knockout (KO) groups was found, regardless of vaccine formulations. (B) l-Tyrosine adjuvant activity requires peptide coprecipitation. Mice received 8 × 105 pmel-1 T cells and indicated treatments on day 0. n = 5 mice per group. Data are representative of three independent experiments.

Close modal

Because l-tyrosine was only active when coprecipitated with peptide, we hypothesized that l-tyrosine extends the duration of peptide presentation to naive CD8+ T cells compared with free peptide delivered in saline. To test this hypothesis, we vaccinated mice with gp100/l-tyrosine, gp100/saline, or gp100/IFA (all with covax) and then transferred CFSE-labeled, naive gp100-specific pmel-1 CD8 T cells at different time points after vaccination. IFA formulation with peptide was included because it has been well documented to extend Ag presentation time over a long period of time and served as a positive control (9, 16, 17). The productive presentation of gp100 peptide Ag in vivo was detected by measuring the proliferation of naive gp100-specific pmel-1 T cells, as indicated by CFSE dilution, at 72 h after their adoptive transfer. We found that gp100/l-tyrosine extended the duration of gp100 peptide presentation beyond gp100/saline by ∼3–4 d, but not to the extent caused by gp100/IFA, which was still potently presented after 98 d, in line with our previous observations (Fig. 5, Supplemental Fig. 3) (9). Thus, l-tyrosine microparticles functioned as a peptide vaccine formulation that caused an intermediate duration of Ag presentation to T cells in vivo.

FIGURE 5.

l-Tyrosine formulation extends the duration of Ag presentation. (A) Schematic of the experimental design. All mice were treated as indicated on day 0. (B) At indicated time points, 2 × 106 CFSE-labeled pmel-1 CD8+ T cells were transferred to hosts. Seventy-two hours after transfer, vaccine-draining LN were harvested and CFSE dilution of pmel-1 T cells was measured by flow cytometry. n = 3–5 mice per group. Data are representative of two independent experiments.

FIGURE 5.

l-Tyrosine formulation extends the duration of Ag presentation. (A) Schematic of the experimental design. All mice were treated as indicated on day 0. (B) At indicated time points, 2 × 106 CFSE-labeled pmel-1 CD8+ T cells were transferred to hosts. Seventy-two hours after transfer, vaccine-draining LN were harvested and CFSE dilution of pmel-1 T cells was measured by flow cytometry. n = 3–5 mice per group. Data are representative of two independent experiments.

Close modal

Because the extended duration of Ag presentation by l-tyrosine formulation correlated with its ability to induce a superior T cell response, we determined whether a similar result could be attained by extending Ag presentation through repeated injections of unformulated, soluble peptide. Indeed, T cell levels were very comparable between mice receiving one dose of gp100/l-tyrosine and three doses of gp100/saline on 3 consecutive days (Fig. 6A), whereas cytokine (IFN-γ/TNF-α) production was reduced (Fig. 6B). This may explain why tumor rejection after repeated peptide injection was improved compared with single gp100/saline injection, but still not as efficiently as gp100/l-tyrosine (Fig. 6C).

FIGURE 6.

Antitumor efficacy of different vaccine formulations regimens. Mice received 8 × 105 pmel-1 T cells and indicated treatments on day 0. (A) Pmel-1 T cell response after different vaccine formulations were followed and (B) their function at day 7 postvaccination is shown (mean ± SEM). (C) Antitumor efficacy of corresponding groups of the same experiment. n = 10–20 mice per group. Peptide dose was 50 μg per mouse. Data were pooled from two independent experiments. Differences in survival among groups were compared using a log-rank test. **p < 0.01. n/a, not available. ns, not significant.

FIGURE 6.

Antitumor efficacy of different vaccine formulations regimens. Mice received 8 × 105 pmel-1 T cells and indicated treatments on day 0. (A) Pmel-1 T cell response after different vaccine formulations were followed and (B) their function at day 7 postvaccination is shown (mean ± SEM). (C) Antitumor efficacy of corresponding groups of the same experiment. n = 10–20 mice per group. Peptide dose was 50 μg per mouse. Data were pooled from two independent experiments. Differences in survival among groups were compared using a log-rank test. **p < 0.01. n/a, not available. ns, not significant.

Close modal

To rule out the possibility that our findings applied uniquely to our specific model system based on gp100 Ag and gp100-specific, transgenic T cells, we examined endogenous T cell responses to gp100 and OVA-I257–264 (SIINFEKL) peptides. When testing endogenous T cell responses to gp100, we observed a very low response after one vaccination, likely due to the reported very low T cell precursor frequency to this self-antigen (18). However, after a two-booster vaccination, mice receiving gp100/l-tyrosine showed a dramatically stronger gp100-specific CD8+ T cell response than did mice receiving gp100/saline (Fig. 7A). We then tested endogenous T cell responses to the unrelated non-self Ag, OVA-I, and observed a strong CD8+ T cell response after two vaccinations, with OVA-I/l-tyrosine and especially three successive daily OVA-I/saline vaccinations giving a clearly enhanced T cell response. Overall, these results demonstrate that approaches that prolong Ag presentation in vivo deserve further investigation in the development of human cancer vaccines.

FIGURE 7.

T cell responses from the endogenous repertoire. (A) Endogenous gp100-specific T cell responses after saline and l-tyrosine vaccines, detected by IFN-γ+ CD8 T cells. (B) Endogenous OVA-I–specific T cell responses after saline and l-Tyrosine vaccines, detected by OVA-I dextramer H2-Kb+ CD8 T cells. Boosters were given as indicated times, with same dose and formulations as for priming. n = 3–5 mice per group.

FIGURE 7.

T cell responses from the endogenous repertoire. (A) Endogenous gp100-specific T cell responses after saline and l-tyrosine vaccines, detected by IFN-γ+ CD8 T cells. (B) Endogenous OVA-I–specific T cell responses after saline and l-Tyrosine vaccines, detected by OVA-I dextramer H2-Kb+ CD8 T cells. Boosters were given as indicated times, with same dose and formulations as for priming. n = 3–5 mice per group.

Close modal

Cancer vaccines are widely explored as a means to induce tumor-specific T cells, but thus far clinical success has been limited. One reason is a profound lack of sufficiently potent vaccine adjuvants available for the potentiation of cancer vaccines aimed at inducing robust tumor-specific T cell responses. Vaccine adjuvants can be broadly categorized into two groups: Ag delivery systems and immunopotentiators (1921).

Most Ag delivery systems influence the availability of Ag in vivo, either by protecting Ag from rapid clearance (e.g., by proteases) and/or delivering the Ag to lymph nodes (LN), either directly or indirectly (by targeting tissue APC that traffic to LN). Importantly, Ag delivery systems can also serve as carriers for immunopotentiators) as well as function as innate immune activators themselves. For example, IFA and other water-in-oil emulsion adjuvants (Montanide oil series; SEPPIC) can both retain and slowly release the Ag at the vaccination site and cause local inflammation (16). Alum, which has been used for >80 y, is the most widely used vaccine adjuvant for human and veterinary vaccines (20). The Ag depot effect has generally been accepted as a major mechanism of action of alum until recently. Hutchison et al. (22) showed that Ab production induced by alum remained intact even when the injection site was surgically removed shortly (2 h) after vaccination. This observation suggested that adjuvant activity of alum is likely due to its immunopotentiating activity. When we tested gp100 peptide in alum, we did not observe any effect on T cell response (Supplemental Fig. 2C), possibly indicating that alum does not function as a depot for gp100 peptide, and/or that its known activation of the inflammasome was irrelevant due to the presence of our potent combination of proinflammatory mediators (covax). Indeed, alum was shown to activate the NALP3 inflammasome in DC (23). Inflammasome activation results in production of proinflammatory cytokines such as IL-1β and IL-18 (24, 25). In contrast, the adjuvant activity of l-tyrosine microparticles did not require inflammasome activation, but could be largely mimicked by extending Ag presentation through repeated Ag injection, suggesting that the depot function of l-tyrosine was a major determinant of its adjuvant activity. Micro- and nanoparticles that can be generated in well-defined structure, size, and shape, and they offer several advantages in vaccine design. Particles can be loaded with Ag and different molecular immunopotentiators (e.g., STING agonist cdGMP, TLR9 agonist CpG) and target desired cell types (such as APC) or tissues (such as LN or tumors) (2630). Size, shape, half-life, surface charge, hydrophobicity, material choice, as well as the type of physicochemical interaction with peptide all contribute to adjuvanticity of particles. In terms of size, particles ≥200 nm are trapped by local APC, which eventually migrate to draining LN. Particles of 20–200 nm drain passively through LN where they will be taken up by LN-resident DC (17, 24). Particles <10 nm drain through blood capillaries (30). A 20- to 200-nm size range is ideal for DC, whereas ≥500 nm preferentially target macrophages (31). Microparticles/nanoparticles can activate innate immune components, depending on their materials. Poly(dl-lactic-coglycolide) and polystyrene particles are inflammasome activators enhancing IL-1β production by DC in a phagocytosis-dependent manner (32), whereas carbon nanotubes activate the complement system and subsequent inflammatory responses via binding to the complement factor C1q (33). The lipid layer of liposomes can be positively charged (cationic liposome) to promote interaction with cell membranes (which is negatively charged) (34). The persistence of particle liposomes in vivo can be extended when they are coated with polyethylene glycol or other biocompatible polymers (35). Besides the half-life of Ag carriers, Ag–carrier interaction is another factor controlling Ag persistence. Generally, encapsulation and chemical conjugation provide stronger interaction between Ag and carrier than adsorption. which is basically a charge or hydrophobic interaction (30). Virus-like particles (VLP), formed by viral structural proteins, are distinct from other Ag delivery systems due to their unique formation. Repetitive antigenic epitopes displayed on VLP cross-link B cell receptors, leading to humoral responses, whereas their ability to target to DC (and be cross-presented on MHC class I) is responsible for cellular response (3638). Target Ags can be genetically fused with structural proteins in nonenveloped VLP or integrated in the outer surface (derived from host cell membrane) in enveloped VLP (39).

Immunopotentiators induce costimulatory signals on APC to increase immune responses. In the present study we used a combination of three defined immunopotentiators: IL-2, CD40 agonist mAb, and imiquimod, collectively called covax. This combination of multiple specific immunological signals, similar to those induced by a viral infection, induces a potent and functional T cell response to suppress established tumors. Versions of covax that include one or two of these three components are dramatically less powerful (data not shown). Indeed, experimental single-agent vaccine adjuvants such as IFA, GM-CSF, or polyinosinic-polycytidylic acid have met with limited success in clinical trials of vaccines against established tumors.

l-Tyrosine microparticle adjuvant has been used clinically in pollen allergy vaccines and its safety has been extensively documented (12). Formulated with pollen allergens, it promotes Th1 (particularly IgG2a) Ab responses and therefore desensitizes allergic reactions (40). In comparison, the same allergens formulated with alum induce Th2 responses. Although the exact nature of the interaction between l-tyrosine particles and peptide Ags remains to be thoroughly investigated, we speculate that peptide Ag is adsorbed to l-tyrosine by hydrophobic interaction. Our preliminary observations suggest that peptides with higher grand average hydropathy scores may adsorb more efficiently to l-tyrosine particles (H. Khong and W.W. Overwijk, unpublished observations). As previously mentioned, the interaction between Ag and other delivery systems can be manipulated through chemical conjugation, encapsulation, or charge interaction. It would be interesting to compare the activity of such other Ag delivery platforms with l-tyrosine. In particular, it would be interesting to examine the relationship between the duration of Ag presentation of different Ag formulations with the magnitude and quality of the T cell responses they induce. However, besides a certain duration of Ag presentation, each Ag delivery system will have additional, unique attributes, including the ability to activate the inflammasome and complement systems to attract influx of specific immune cells, particle size, stability, and propensity for in vivo dispersal, surface charge, and resulting uptake by specific (immune) cell types. This would make it impossible to isolate the impact of duration of Ag presentation on T cell immunity from a study of multiple different vaccine adjuvants. To more directly probe the impact of extended duration of Ag presentation on T cell response, we therefore administered peptide in saline for a varying number of days, keeping all other variables constant, and found that duration of Ag presentation was a powerful driver of T cell immunity. This demonstrated that prolonging the availability of Ag by repeated administration directly induced a more powerful T cell response. It is important to recognize that different vaccine adjuvants and formulations have many important effects on T cell immunity beyond controlling the duration of Ag presentation. It is also quite possible that individual peptides may benefit most from different vaccine adjuvants, depending on the unique physicochemical attributes of the peptide. In our studies, gp100-specificic immunity was strongly enhanced by formulating the gp100 peptide in l-tyrosine as well as by repeated peptide administration in saline, whereas OVA-specific immunity was only modestly enhanced by formulation in l-tyrosine and benefited from more repeated administration in saline. It will therefore be interesting to determine, for individual peptides, the relative potency and safety of the wide variety of currently approved and experimental vaccine formulations and adjuvants, including l-tyrosine microparticles and repeated peptide administration in saline, to gain a deeper understanding of their relative utility in clinical vaccine applications.

The contribution of Ag exposure time to the expansion and differentiation of T cells has been previously recognized. Initial studies suggested that a very brief Ag stimulation (∼2 h) was enough for CD8 T cells to enter autonomous clonal expansion (41, 42). Subsequent studies showed that longer antigenic stimulation (20–64 h) was required for CD8 T cells to acquire full effector function and memory differentiation after expansion (4345). In vitro settings used in these studies, however, did not truly mimic conditions of T cell priming. To overcome such shortcomings, independent groups used Listeria monocytogenes infection followed by antibiotic treatment as an in vivo model to study the role of Ag presentation time to CD8 T cell response. Mercado et al. (46) showed that the magnitude and kinetics of Ag-specific T cell responses were only determined within the first 24 h of L. monocytogenes infection (44, 45). Williams and Bevan (47) found that reducing infection time diminished memory differentiation but not expansion of CD8 T cells. Importantly, when L. monocytogenes was increased to high dose (10-fold), reduction of infection time had minimal effect on memory CD8 T cell differentiation. The authors proposed that the initial dose of Ag and costimulatory signals/cytokines dictated the differentiation of effector to memory T cells. However, another possibility was that high-dose L. monocytogenes infection resulted in more Ag that required longer time for complete clearance after antibiotic administration. In fact, manipulating bacterial clearance through antibiotics was a caveat of these studies. Although antibiotics block the production of new protein Ag, previously produced Ag requires time to be completely cleared and meanwhile can still prime T cells. Also, antibiotics curtail pathogen-induced innate immune activation by nucleic acids and TLR ligands. To circumvent these issues, Prlic et al. (48) developed a model using peptide-loaded diphtheria receptor expressing DC to finely tune Ag presentation time. The authors found that duration of TCR stimulus controlled the magnitude but not functionality of CD8 T cells response. Blair et al. (49) altered the Ag availability to T cells by employing mAb that blocks Ag/MHC from engaging TCR of Ag-specific T cells. With this elegant model, they reached similar conclusions as Prlic et al. However, in all of these cases, the natural duration of Ag presentation was shortened, reducing the resulting T cell response level. In a setting of anticancer vaccination, the goal would be to induce a T cell response of the greatest magnitude and antitumor function. We therefore employed an opposite tactic where Ag presentation was not limited but extended beyond its “natural” duration (i.e., the ubiquitously employed single injection of peptide Ag), either by formulation with l-tyrosine or through repeated injection of peptide in saline. Using this approach, we confirmed the critical contribution of Ag exposure time for the activation and differentiation of CD8 T cells, and demonstrate that artificially extending the duration of Ag presentation results in superior T cell number and consequent antitumor activity. Importantly, repeated injection of the free peptide model circumvents shortcomings of using different formulations to study contribution of Ag presentation to T cell responses for the reasons mentioned above. Theoretically, the magnitude of T cell response by prolonged Ag stimulation may be controlled by the number of Ag-specific T cells recruited to clonal expansion (50). Nonetheless, Ag-specific T cells were recruited to clonal expansion very efficiently in our vaccine settings, with virtually no naive T cells remaining after 5 d, regardless of vaccine formulation (data not shown). This finding is consistent with previous reports, where naive T cell precursor recruitment was also highly efficient and nearly complete (41, 48, 51). Therefore, we conclude that the increased magnitude of CD8 T cell response either after l-tyrosine or by repeated injection is caused by increased Ag-driven clonal expansion rather than enhanced recruitment of Ag-specific naive T cells into proliferation.

Intriguingly, although repeated injection of peptide and l-tyrosine formulation induced similar tumor-specific T cell levels, the resulting antitumor efficacy of l-tyrosine was more potent. This discrepancy could imply that l-tyrosine has other, undiscovered impacts on T cell responses besides the extended Ag presentation effect we described in the present study. In fact, meta- and ortho-tyrosine have been shown to mediate a form of concomitant tumor resistance, a phenomenon where primary tumor suppresses the growth of distant secondary tumors (52). l-Tyrosine used in our vaccine, however, was para-l-tyrosine, which was described as not mediating concomitant immunity. It appears more likely that three daily injections of peptide in saline and one injection of peptide in l-tyrosine microparticles do not produce identical kinetics of Ag release and presentation by APC, resulting in similar quantity but not quality of the ensuing T cells response, as we observed. This indicates that there may be more effective regimens of repeated Ag injection, for example a gradual increase of Ag dose followed by a gradual decrease over time, more closely mimicking the kinetics of Ag production during a viral infection; these permutations remain to be explored. From a clinical perspective, repeated Ag injection in saline is attractive because it simplifies the preparation process but it will complicate the administration because patients will have to remain hospitalized for multiple days. In contrast, formulating Ag in an appropriate depot adjuvant will complicate vaccine preparation but reduce the patients’ hospital stays. The dose and frequency of Ag administration may depend on the kinetics of Ag presentation and are likely to be unique for each peptide, based on its solubility, length, susceptibility to proteases and peptidases, and affinity for the MHC class I binding groove, which protects bound peptides from degradation. T cells with a low-affinity TCR are less likely to undergo prolonged expansion than are high-affinity T cell clones (53, 54). Still, in formulation with l-tyrosine, the kinetics of Ag presentation of an individual peptide will depend on the efficiency to (dis)associate with l-tyrosine microparticles, and how long the peptide will persist once released in vivo. Future refinements in peptide vaccine adjuvants may include formulations that are relatively insensitive to the physicochemical nature of the antigenic peptide, allowing for standardized peptide incorporation efficiency. This will facilitate the development of multipeptide vaccines, reducing the chance of tumor escape through Ag loss (5457). Given recent progress in the design of therapeutic vaccines, including tumor Ag selection, Ag delivery platform, and immunopotentiators, the development of effective therapeutic vaccine for cancer is accelerating. The duration of Ag presentation is an important parameter that can be controlled through vaccine formulation to drive more effective anticancer T cell responses for the therapy of patients with cancer.

We thank Dr. Thirumala-Devi Kanneganti from St. Jude Children’s Research Hospital for providing the ASC knockout mice, as well as Dr. Louis M. Vence and Yi Xiaohui for expertise and technical support on cytokine/chemokine assays.

This work was supported by Cancer Prevention Research Institute of Texas Individual Investigator Research Award RP140522 (to W.W.O.), a Vietnam Education Foundation fellowship (to H.K.), National Cancer Institute T32 Training Grant CA009599 (to D.T.M.), Cancer Prevention Research Institute of Texas Training Grant RP140106 (to C.S.I.), National Cancer Institute Cancer Center Support Grant P30CA16672 to the Flow Cytometry Core, Cancer Prevention Research Institute of Texas Grant RP130397, and National Institutes of Health High-End Instrument Grant 1S10OD012304-01 to the Proteomics and Metabolomics Core Facility of the University of Texas MD Anderson Cancer Center.

The online version of this article contains supplemental material.

Abbreviations used in this article:

     
  • alum

    aluminum hydroxide gel

  •  
  • covax

    a mixture of anti-CD40 mAb, IL-2, and imiquimod 5% cream

  •  
  • DC

    dendritic cell

  •  
  • LN

    lymph node

  •  
  • MPA

    mobile phase A

  •  
  • MPB

    mobile phase B

  •  
  • VLP

    virus-like particle.

1
Topalian
,
S. L.
,
J. M.
Taube
,
R. A.
Anders
,
D. M.
Pardoll
.
2016
.
Mechanism-driven biomarkers to guide immune checkpoint blockade in cancer therapy.
Nat. Rev. Cancer
16
:
275
287
.
2
Rosenberg
,
S. A.
,
J. C.
Yang
,
N. P.
Restifo
.
2004
.
Cancer immunotherapy: moving beyond current vaccines.
Nat. Med.
10
:
909
915
.
3
Klebanoff
,
C. A.
,
N.
Acquavella
,
Z.
Yu
,
N. P.
Restifo
.
2011
.
Therapeutic cancer vaccines: are we there yet?
Immunol. Rev.
239
:
27
44
.
4
Schumacher
,
T. N.
,
R. D.
Schreiber
.
2015
.
Neoantigens in cancer immunotherapy.
Science
348
:
69
74
.
5
van der Burg
,
S. H.
,
R.
Arens
,
F.
Ossendorp
,
T.
van Hall
,
C. J. M.
Melief
.
2016
.
Vaccines for established cancer: overcoming the challenges posed by immune evasion.
Nat. Rev. Cancer
16
:
219
233
.
6
Gubin
,
M. M.
,
X.
Zhang
,
H.
Schuster
,
E.
Caron
,
J. P.
Ward
,
T.
Noguchi
,
Y.
Ivanova
,
J.
Hundal
,
C. D.
Arthur
,
W.-J.
Krebber
, et al
.
2014
.
Checkpoint blockade cancer immunotherapy targets tumour-specific mutant antigens.
Nature
515
:
577
581
.
7
Ott
,
P. A.
,
Z.
Hu
,
D. B.
Keskin
,
S. A.
Shukla
,
J.
Sun
,
D. J.
Bozym
,
W.
Zhang
,
A.
Luoma
,
A.
Giobbie-Hurder
,
L.
Peter
, et al
.
2017
.
An immunogenic personal neoantigen vaccine for patients with melanoma.
Nature
547
:
217
221
.
8
Sahin
,
U.
,
E.
Derhovanessian
,
M.
Miller
,
B.-P.
Kloke
,
P.
Simon
,
M.
Löwer
,
V.
Bukur
,
A. D.
Tadmor
,
U.
Luxemburger
,
B.
Schrörs
, et al
.
2017
.
Personalized RNA mutanome vaccines mobilize poly-specific therapeutic immunity against cancer.
Nature
547
:
222
226
.
9
Hailemichael
,
Y.
,
Z.
Dai
,
N.
Jaffarzad
,
Y.
Ye
,
M. A.
Medina
,
X.-F.
Huang
,
S. M.
Dorta-Estremera
,
N. R.
Greeley
,
G.
Nitti
,
W.
Peng
, et al
.
2013
.
Persistent antigen at vaccination sites induces tumor-specific CD8+ T cell sequestration, dysfunction and deletion.
Nat. Med.
19
:
465
472
.
10
Wheeler
,
A. W.
,
D. M.
Moran
,
B. E.
Robins
,
A.
Driscoll
.
1982
.
L-Tyrosine as an immunological adjuvant.
Int. Arch. Allergy Appl. Immunol.
69
:
113
119
.
11
Lyons
,
A. B.
2000
.
Analysing cell division in vivo and in vitro using flow cytometric measurement of CFSE dye dilution.
J. Immunol. Methods
243
:
147
154
.
12
Baldrick
,
P.
,
D.
Richardson
,
A. W.
Wheeler
.
2002
.
Review of L-tyrosine confirming its safe human use as an adjuvant.
J. Appl. Toxicol.
22
:
333
344
.
13
Ben-Sasson
,
S. Z.
,
A.
Hogg
,
J.
Hu-Li
,
P.
Wingfield
,
X.
Chen
,
M.
Crank
,
S.
Caucheteux
,
M.
Ratner-Hurevich
,
J. A.
Berzofsky
,
R.
Nir-Paz
,
W. E.
Paul
.
2013
.
IL-1 enhances expansion, effector function, tissue localization, and memory response of antigen-specific CD8 T cells.
J. Exp. Med.
210
:
491
502
.
14
Hornung
,
V.
,
F.
Bauernfeind
,
A.
Halle
,
E. O.
Samstad
,
H.
Kono
,
K. L.
Rock
,
K. A.
Fitzgerald
,
E.
Latz
.
2008
.
Silica crystals and aluminum salts activate the NALP3 inflammasome through phagosomal destabilization.
Nat. Immunol.
9
:
847
856
.
15
Dinarello
,
C. A.
,
A.
Simon
,
J. W. M.
van der Meer
.
2012
.
Treating inflammation by blocking interleukin-1 in a broad spectrum of diseases.
Nat. Rev. Drug Discov.
11
:
633
652
.
16
Stills
H. F.
 Jr
.
2005
.
Adjuvants and antibody production: dispelling the myths associated with Freund’s complete and other adjuvants.
ILAR J.
46
:
280
293
.
17
Bijker
,
M. S.
,
S. J. F.
van den Eeden
,
K. L.
Franken
,
C. J. M.
Melief
,
S. H.
van der Burg
,
R.
Offringa
.
2008
.
Superior induction of anti-tumor CTL immunity by extended peptide vaccines involves prolonged, DC-focused antigen presentation.
Eur. J. Immunol.
38
:
1033
1042
.
18
Rizzuto
,
G. A.
,
T.
Merghoub
,
D.
Hirschhorn-Cymerman
,
C.
Liu
,
A. M.
Lesokhin
,
D.
Sahawneh
,
H.
Zhong
,
K. S.
Panageas
,
M.-A.
Perales
,
G.
Altan-Bonnet
, et al
.
2009
.
Self-antigen-specific CD8+ T cell precursor frequency determines the quality of the antitumor immune response.
J. Exp. Med.
206
:
849
866
.
19
Khong
,
H.
,
W. W.
Overwijk
.
2016
.
Adjuvants for peptide-based cancer vaccines.
J. Immunother. Cancer
4
:
56
.
20
Cox
,
J. C.
,
A. R.
Coulter
.
1997
.
Adjuvants—a classification and review of their modes of action.
Vaccine
15
:
248
256
.
21
Lindblad
E. B.
2004
.
Aluminium adjuvants—in retrospect and prospect.
Vaccine
22
:
3658
3668
.
22
Hutchison
,
S.
,
R. A.
Benson
,
V. B.
Gibson
,
A. H.
Pollock
,
P.
Garside
,
J. M.
Brewer
.
2012
.
Antigen depot is not required for alum adjuvanticity.
FASEB J.
26
:
1272
1279
.
23
Eisenbarth
,
S. C.
,
O. R.
Colegio
,
W.
O’Connor
,
F. S.
Sutterwala
,
R. A.
Flavell
.
2008
.
Crucial role for the Nalp3 inflammasome in the immunostimulatory properties of aluminium adjuvants.
Nature
453
:
1122
1126
.
24
Lamkanfi
,
M.
,
V. M.
Dixit
.
2014
.
Mechanisms and functions of inflammasomes.
Cell
157
:
1013
1022
.
25
Seubert
,
A.
,
E.
Monaci
,
M.
Pizza
,
D. T.
O’Hagan
,
A.
Wack
.
2008
.
The adjuvants aluminum hydroxide and MF59 induce monocyte and granulocyte chemoattractants and enhance monocyte differentiation toward dendritic cells.
J. Immunol.
180
:
5402
5412
.
26
Liu
,
H.
,
K. D.
Moynihan
,
Y.
Zheng
,
G. L.
Szeto
,
A. V.
Li
,
B.
Huang
,
D. S.
Van Egeren
,
C.
Park
,
D. J.
Irvine
.
2014
.
Structure-based programming of lymph-node targeting in molecular vaccines.
Nature
507
:
519
522
.
27
Hanson
,
M. C.
,
M. P.
Crespo
,
W.
Abraham
,
K. D.
Moynihan
,
G. L.
Szeto
,
S. H.
Chen
,
M. B.
Melo
,
S.
Mueller
,
D. J.
Irvine
.
2015
.
Nanoparticulate STING agonists are potent lymph node-targeted vaccine adjuvants.
J. Clin. Invest.
125
:
2532
2546
.
28
Mehta
,
N. K.
,
K. D.
Moynihan
,
D. J.
Irvine
.
2015
.
Engineering new approaches to cancer vaccines.
Cancer Immunol. Res.
3
:
836
843
.
29
Wang
,
A. Z.
,
R.
Langer
,
O. C.
Farokhzad
.
2012
.
Nanoparticle delivery of cancer drugs.
Annu. Rev. Med.
63
:
185
198
.
30
Zhao
,
L.
,
A.
Seth
,
N.
Wibowo
,
C.-X.
Zhao
,
N.
Mitter
,
C.
Yu
,
A. P. J.
Middelberg
.
2014
.
Nanoparticle vaccines.
Vaccine
32
:
327
337
.
31
Xiang
,
S. D.
,
A.
Scholzen
,
G.
Minigo
,
C.
David
,
V.
Apostolopoulos
,
P. L.
Mottram
,
M.
Plebanski
.
2006
.
Pathogen recognition and development of particulate vaccines: does size matter?
Methods
40
:
1
9
.
32
Sharp
,
F. A.
,
D.
Ruane
,
B.
Claass
,
E.
Creagh
,
J.
Harris
,
P.
Malyala
,
M.
Singh
,
D. T.
O’Hagan
,
V.
Pétrilli
,
J.
Tschopp
, et al
.
2009
.
Uptake of particulate vaccine adjuvants by dendritic cells activates the NALP3 inflammasome.
Proc. Natl. Acad. Sci. USA
106
:
870
875
.
33
Salvador-Morales
,
C.
,
E.
Flahaut
,
E.
Sim
,
J.
Sloan
,
M. L. H.
Green
,
R. B.
Sim
.
2006
.
Complement activation and protein adsorption by carbon nanotubes.
Mol. Immunol.
43
:
193
201
.
34
Schwendener
,
R. A.
2014
.
Liposomes as vaccine delivery systems: a review of the recent advances.
Ther. Adv. Vaccines
2
:
159
182
.
35
Milla
,
P.
,
F.
Dosio
,
L.
Cattel
.
2012
.
PEGylation of proteins and liposomes: a powerful and flexible strategy to improve the drug delivery.
Curr. Drug Metab.
13
:
105
119
.
36
Li
,
K.
,
A.
Peers-Adams
,
S. J.
Win
,
S.
Scullion
,
M.
Wilson
,
V. L.
Young
,
P.
Jennings
,
V. K.
Ward
,
M. A.
Baird
,
S. L.
Young
.
2013
.
Antigen incorporated in virus-like particles is delivered to specific dendritic cell subsets that induce an effective antitumor immune response in vivo.
J. Immunother.
36
:
11
19
.
37
Win
,
S. J.
,
V. K.
Ward
,
P. R.
Dunbar
,
S. L.
Young
,
M. A.
Baird
.
2011
.
Cross-presentation of epitopes on virus-like particles via the MHC I receptor recycling pathway.
Immunol. Cell Biol.
89
:
681
688
.
38
Chackerian
,
B.
2007
.
Virus-like particles: flexible platforms for vaccine development.
Expert Rev. Vaccines
6
:
381
390
.
39
Kushnir
,
N.
,
S. J.
Streatfield
,
V.
Yusibov
.
2012
.
Virus-like particles as a highly efficient vaccine platform: diversity of targets and production systems and advances in clinical development.
Vaccine
31
:
58
83
.
40
Wheeler
,
A. W.
,
S. R.
Woroniecki
.
2001
.
Immunological adjuvants in allergy vaccines: past, present future.
Allergol. Int.
50
:
295
301
.
41
Kaech
,
S. M.
,
R.
Ahmed
.
2001
.
Memory CD8+ T cell differentiation: initial antigen encounter triggers a developmental program in naïve cells.
Nat. Immunol.
2
:
415
422
.
42
van Stipdonk
,
M. J.
,
E. E.
Lemmens
,
S. P.
Schoenberger
.
2001
.
Naïve CTLs require a single brief period of antigenic stimulation for clonal expansion and differentiation.
Nat. Immunol.
2
:
423
429
.
43
Gett
,
A. V.
,
F.
Sallusto
,
A.
Lanzavecchia
,
J.
Geginat
.
2003
.
T cell fitness determined by signal strength.
Nat. Immunol.
4
:
355
360
.
44
van Stipdonk
,
M. J. B.
,
G.
Hardenberg
,
M. S.
Bijker
,
E. E.
Lemmens
,
N. M.
Droin
,
D. R.
Green
,
S. P.
Schoenberger
.
2003
.
Dynamic programming of CD8+ T lymphocyte responses.
Nat. Immunol.
4
:
361
365
.
45
Curtsinger
,
J. M.
,
C. M.
Johnson
,
M. F.
Mescher
.
2003
.
CD8 T cell clonal expansion and development of effector function require prolonged exposure to antigen, costimulation, and signal 3 cytokine.
J. Immunol.
171
:
5165
5171
.
46
Mercado
,
R.
,
S.
Vijh
,
S. E.
Allen
,
K.
Kerksiek
,
I. M.
Pilip
,
E. G.
Pamer
.
2000
.
Early programming of T cell populations responding to bacterial infection.
J. Immunol.
165
:
6833
6839
.
47
Williams
,
M. A.
,
M. J.
Bevan
.
2004
.
Shortening the infectious period does not alter expansion of CD8 T cells but diminishes their capacity to differentiate into memory cells.
J. Immunol.
173
:
6694
6702
.
48
Prlic
,
M.
,
G.
Hernandez-Hoyos
,
M. J.
Bevan
.
2006
.
Duration of the initial TCR stimulus controls the magnitude but not functionality of the CD8+ T cell response.
J. Exp. Med.
203
:
2135
2143
.
49
Blair
,
D. A.
,
D. L.
Turner
,
T. O.
Bose
,
Q.-M.
Pham
,
K. R.
Bouchard
,
K. J.
Williams
,
J. P.
McAleer
,
L. S.
Cauley
,
A. T.
Vella
,
L.
Lefrançois
.
2011
.
Duration of antigen availability influences the expansion and memory differentiation of T cells.
J. Immunol.
187
:
2310
2321
.
50
Au-Yeung
,
B. B.
,
J.
Zikherman
,
J. L.
Mueller
,
J. F.
Ashouri
,
M.
Matloubian
,
D. A.
Cheng
,
Y.
Chen
,
K. M.
Shokat
,
A.
Weiss
.
2014
.
A sharp T-cell antigen receptor signaling threshold for T-cell proliferation.
Proc. Natl. Acad. Sci. USA
111
:
E3679
E3688
.
51
van Heijst
,
J. W. J.
,
C.
Gerlach
,
E.
Swart
,
D.
Sie
,
C.
Nunes-Alves
,
R. M.
Kerkhoven
,
R.
Arens
,
M.
Correia-Neves
,
K.
Schepers
,
T. N. M.
Schumacher
.
2009
.
Recruitment of antigen-specific CD8+ T cells in response to infection is markedly efficient.
Science
325
:
1265
1269
.
52
Ruggiero
,
R. A.
,
J.
Bruzzo
,
P.
Chiarella
,
P.
di Gianni
,
M. A.
Isturiz
,
S.
Linskens
,
N.
Speziale
,
R. P.
Meiss
,
O. D.
Bustuoabad
,
C. D.
Pasqualini
.
2011
.
Tyrosine isomers mediate the classical phenomenon of concomitant tumor resistance.
Cancer Res.
71
:
7113
7124
.
53
Zehn
,
D.
,
S. Y.
Lee
,
M. J.
Bevan
.
2009
.
Complete but curtailed T-cell response to very low-affinity antigen.
Nature
458
:
211
214
.
54
Gronski
,
M. A.
,
J. M.
Boulter
,
D.
Moskophidis
,
L. T.
Nguyen
,
K.
Holmberg
,
A. R.
Elford
,
E. K.
Deenick
,
H. O.
Kim
,
J. M.
Penninger
,
B.
Odermatt
, et al
.
2004
.
TCR affinity and negative regulation limit autoimmunity.
Nat. Med.
10
:
1234
1239
.
55
Walter
,
S.
,
T.
Weinschenk
,
A.
Stenzl
,
R.
Zdrojowy
,
A.
Pluzanska
,
C.
Szczylik
,
M.
Staehler
,
W.
Brugger
,
P.-Y.
Dietrich
,
R.
Mendrzyk
, et al
.
2012
.
Multipeptide immune response to cancer vaccine IMA901 after single-dose cyclophosphamide associates with longer patient survival.
Nat. Med.
18
:
1254
1261
.
56
Bai
,
X.-F.
,
J.
Liu
,
O.
Li
,
P.
Zheng
,
Y.
Liu
.
2003
.
Antigenic drift as a mechanism for tumor evasion of destruction by cytolytic T lymphocytes.
J. Clin. Invest.
111
:
1487
1496
.
57
Melero
,
I.
,
G.
Gaudernack
,
W.
Gerritsen
,
C.
Huber
,
G.
Parmiani
,
S.
Scholl
,
N.
Thatcher
,
J.
Wagstaff
,
C.
Zielinski
,
I.
Faulkner
,
H.
Mellstedt
.
2014
.
Therapeutic vaccines for cancer: an overview of clinical trials.
Nat. Rev. Clin. Oncol.
11
:
509
524
.

The authors have no financial conflicts of interest.

Supplementary data