Multifactorial Design of a Supramolecular Peptide Anti-IL-17 Vaccine Toward the Treatment of Psoriasis

Abstract

Current treatments for chronic immune-mediated diseases such as psoriasis, rheumatoid arthritis, or Crohn's disease commonly rely on cytokine neutralization using monoclonal antibodies; however, such approaches have drawbacks. Frequent repeated dosing can lead to the formation of anti-drug antibodies and patient compliance issues, and it is difficult to identify a single antibody that is broadly efficacious across diverse patient populations. As an alternative to monoclonal antibody therapy, anti-cytokine immunization is a potential means for long-term therapeutic control of chronic inflammatory diseases. Here we report a supramolecular peptide-based approach for raising antibodies against IL-17 and demonstrate its efficacy in a murine model of psoriasis. B-cell epitopes from IL-17 were co-assembled with the universal T-cell epitope PADRE using the Q11 self-assembling peptide nanofiber system. These materials, with or without adjuvants, raised antibody responses against IL-17. Exploiting the modularity of the system, multifactorial experimental designs were used to select formulations maximizing titer and avidity. In a mouse model of psoriasis induced by imiquimod, unadjuvanted nanofibers had therapeutic efficacy, which could be enhanced with alum adjuvant but reversed with CpG adjuvant. Measurements of antibody subclass induced by adjuvanted and unadjuvanted formulations revealed strong correlations between therapeutic efficacy and titers of IgG1 (improved efficacy) or IgG2b (worsened efficacy). These findings have important implications for the development of anti-cytokine active immunotherapies and suggest that immune phenotype is an important metric for eliciting therapeutic anti-cytokine antibody responses.

Keywords: active immunotherapy; anti-cytokine; immunoengineering; inflammatory diseases; self-assembly.

Figure 1

Selection of IL-17 epitopes and characterization of IL17.1 and IL17.2 nanofibers. Two peptide epitopes, IL17.1 and IL17.2, were identified within mouse IL17A. (A) Homologous epitopes are shown within the crystal structure of human IL17A. (B–F) Both IL17.1-Q11 and IL17.2-Q11 peptides self-assembled into nanofibers (imaged using negative-stained TEM) alone or when co-assembled with Q11 and PADRE-Q11 (single peptides formulated at 2 mM; co-assembled peptides formulated at 1 mM IL17.1-Q11 or IL17.2-Q11/0.95 mM Q11/0.05 mM PADRE-Q11; all nanofibers diluted 10-fold before application to TEM grids). (G) All formulations had significant β-sheet character by ThT fluorescence (mean ± SD, n = 3, all groups significantly different from all other groups, ***p < 0.001 by one-way ANOVA and Tukey's Multiple Comparisons test).

Figure 2

B cell epitope screening in mice. (A) IL-17.1-Q11 and IL17.2-Q11 peptides were co-assembled into nanofibers with Q11 and PADRE-Q11 and injected subcutaneously without adjuvant at weeks 0, 2.5, and 5. Antibody titers were measured at weeks 2, 4, and 7. Mice immunized with IL17.1 raised significantly higher antibody titers against both immunizing peptide (B) and IL17A protein (C) compared to mice immunized with IL17.2. Analysis of the antibody isotype raised by IL17.1 demonstrated a bias toward IgG1 (D). Statistical comparisons were conducted by Student's t-test (n = 5), *p < 0.05. Data are presented as mean ± SD.

Figure 3

Long-term immunization with IL-17.1 maintained immune memory without an autologous anti-IL17.1 T-cell response. (A) Mice were maintained for 1 year after the final immunization before blood collection and a subsequent boost, with the exception of one mouse that died of unrelated causes at 50 weeks. Final sera were collected at week 58 and mice were sacrificed for analysis of T cell response by ELISpot. (B) Anti-IL17.1 responses had declined by 1 year but rebounded after being recalled (unrelated antigen (OVA) responses included as a negative control). The difference was measurable although not statistically significantly different (ns, p = 0.1411) (C) Splenocytes were left unstimulated or stimulated with IL17.1 peptide, PADRE peptide, or Concanavalin A and demonstrated a significant IL-4 response against PADRE but no significant response against IL17.1. (D) Neither IL17.1 nor PADRE elicited a significant IFN-γ response. n = 4, statistical differences for antibody responses were calculated by paired t-test, while statistical differences for the ELISpot assay were calculated by one-way ANOVA and Tukey's test for multiple comparisons. **p < 0.01, ***p < 0.001. Data are presented as Mean ± SD.

Figure 4

Multifactorial optimization of nanofibers containing IL17.1 B-cell epitopes and PADRE T-cell epitopes. Design of Experiments (DoE) methodology (A,D) was employed to investigate how T-cell epitope and B-cell epitope content influenced anti-IL17 titers (B) and avidity (C). A 2 × 2 multifactorial design with a center point was utilized, using nanofibers with stoichiometrically controlled amounts of B- and T-cell epitope indicated (n = 7). To test the predictive power of the DoE, a formulation not specifically tested in the original design was tested subsequently (Formulation shown by the orange dot in B). (E) Average titers generated by this formulation corresponded closely with the DoE prediction (n = 5, individual mice shown, with mean ± SD indicated).

Figure 5

Adjuvants impact antibody titer and alter IgG subclass. To enhance antibody titers, optimized nanofibers containing IL17.1 B cell epitopes and PADRE T cell epitopes were administered along with the adjuvants shown. CpG (n = 5) and alum (n = 5) enhanced total IgG and IgG1 anti-IL17.1 IgG titers compared to unadjuvanted controls (n = 10), but only CpG adjuvant significantly enhanced anti-IL17.1 IgG2b titers, indicating a more Th1-like phenotype. Data shown represent two separate experiments: Experiment 1: CpG n = 5 and unadjuvanted n = 5, Experiment 2: Alum n = 5 and unadjuvanted n = 5. ***p < 0.001 by one-way ANOVA and Tukey's test for multiple comparisons. Data are presented as means ± SD, with individual mice indicated.

Figure 6

IL17.1 nanofiber immunizations diminished imiquimod-induced psoriasis. Mice were immunized prophylactically at 8 weeks of age and boosted 2.5 weeks afterwards. Five weeks after primary immunization, imiquimod (IMQ) was applied daily for 5 days, and tissues were harvested on the sixth day (A). Groups included unvaccinated mice (B), mice vaccinated with CpG-adjuvanted nanofibers (C), unadjuvanted nanofibers (D), and alum-adjuvanted nanofibers (E), plus mice that were neither vaccinated nor administered imiquimod (F). Epidermal thickness was compared at 8 locations over two sections from each animal (G). Statistical significance was measured by one-way ANOVA and Tukey's Multiple Comparisons post-hoc analysis. *p < 0.05, ***p < 0.001. Data are presented as means ± SD. Data shown represent two separately conducted experiments: Experiment 1: unvaccinated + IMQ n = 5, CpG-adjuvanted nanofibers n = 5, unadjuvanted nanofibers n = 5, unvaccinated without IMQ n = 5; Experiment 2: unvaccinated + IMQ n = 5, alum-adjuvanted nanofibers n = 5, unadjuvanted nanofibers n = 5. See Figure S7 for statistical justification for combining these experiments in the analysis shown in (G).

Figure 7

Correlations between antibody titer, subclass, and psoriasis severity. Increased IgG1 correlates with improved psoriasis (decreased epidermal thickness) (A), while increased IgG2b correlates with exacerbated psoriasis (increased epidermal thickness) (B). A heatmap of epidermal thickness illustrates a target region (green) for reduced epidermal thickening (C). Analysis was carried out by simple linear regression. Data shown represent a combination of two separately conducted experiments: Experiment 1: CpG n = 5 and unadjuvanted n = 5; Experiment 2: alum n = 5 and unadjuvanted n = 5.