TLR4 and TLR7/8 Adjuvant Combinations Generate Different Vaccine Antigen-Specific Immune Outcomes in Minipigs when Administered via the ID or IN Routes

Abstract

The induction of high levels of systemic and mucosal humoral immunity is a key goal for many prophylactic vaccines. However, adjuvant strategies developed in mice have often performed poorly in the clinic. Due to their closer similarity to humans, minipigs may provide a more accurate picture of adjuvant performance. Based on their complementary signalling pathways, we assessed humoral immune responses to model antigens after co-administration with the toll-like receptor 4 (TLR4) stimulator glucopyranosyl lipid adjuvant (GLA-AF) or the TLR7/8 agonist resiquimod (R848) (alone and in combination) via the intradermal (ID), intranasal (IN) or combined routes in the Gottingen minipig animal model. Surprisingly, we discovered that while GLA-AF additively enhanced the adjuvant effect of R848 when injected ID, it abrogated the adjuvant activity of R848 after IN inoculation. We then performed a route comparison study using a CN54 gp140 HIV Envelope model antigen adjuvanted with R848 + GLA-AF (ID) or R848 alone (IN). Animals receiving priming inoculations via one route were then boosted by the alternate route. Although differences were observed in the priming phase (IN or ID), responses converged upon boosting by the alternative route with no observable impact resultant from the order of administration (ID/IN vs IN/ID). Specific IgG responses were measured at a distal mucosal site (vaginal), although there was no evidence of mucosal linkage as these closely reflected serum antibody levels. These data indicate that the complex in vivo cross-talk between innate pathways are likely tissue specific and cannot be predicted by simple in vitro models.

Fig 1. GLA-AF enhances R848 induced humoral responses after ID injection.

A) The TLR4 agonist GLA-AF (20 μg) increased the TLR7 (R848–50 μg) adjuvanted ESAT-6 (recombinant M. tuberculosis early secreted antigenic target-6 kDa) antigen-specific IgG responses four-fold after a single prime and one boost ID vaccination (* p = 0.0335; Week 3 and p = 0.0338; Week 7). B) The combination of GLA-AF and R848 had no effect on serum IgA responses over R848 alone. C) The TLR4 agonist GLA-AF significantly ablated the R848 induced β-Gal (Beta galactosidase) antigen-specific IgG responses after IN injection (* p = 0.0261; Week 7). D) The combination of GLA-AF and R848 as well as R848 alone poorly augmented antigen-specific β-Gal serum IgA responses.

Fig 2. Combinations of R848 and/or GLA-AF administered by each route significantly enhance HIV Env gp140 antigen-specific serum immunoglobulin responses.

A) Animals that received three priming inoculation via the IN route followed by two ID boost injections (Group A) exhibited significantly augmented serum IgG responses compared to animals that received the unadjuvanted vaccine (*p = 0.0167; Week 13). B) Animals that received three priming inoculations via the ID route followed by two IN boost injections (Group B) exhibited significantly augmented serum IgG responses compared to animals that received the unadjuvanted vaccine (*p = 0.0167; Week 4). C) Group A animals generated antigen-specific IgA in response to IN inoculation which was boosted by the first but not the second ID injection. Unadjuvanted IN administration failed to elicit detectable responses and no boosting or anamnestic response was observed upon ID vaccination. D) Group B animals demonstrated low level IgA generation in response to adjuvanted ID injections and this was boosted by subsequent IN inoculations (*p = 0.037; week 13).

Fig 3. Combinations of R848 and/or GLA-AF administered by each route significantly enhance HIV Env gp140 antigen-specific nasal mucosal immunoglobulin responses.

A) Group A (3 × IN + 2 × ID) animals had undetectable antigen-specific nasal IgG until after the first ID vaccination While slightly elevated these responses were essentially indistinguishable from animals that received the unadjuvanted vaccine. B) Group B (3 × ID + 2 × IN) animals demonstrated higher nasal mucosal IgG compared to animals that received the unadjuvanted vaccine (*p = 0.0167; Week 15). C) IN inoculation elicited antigen-specific IgA that was boosted intranasally after ID vaccinations. D) ID vaccinations did not elicit any detectable antigen-specific IgA in the nasal cavity. Subsequent IN inoculation generated low levels of specific IgA.

Fig 4. Combinations of R848 and/or GLA-AF administered by each route significantly enhance HIV Env gp140 antigen-specific vaginal mucosal immunoglobulin responses.

A) Group A (3 × IN + 2 × ID) animals exhibited augmented vaginal mucosal IgG levels compared to animals that received the unadjuvanted vaccine (*p = 0.0167; week 14). B) Group B animals had higher vaginal mucosal IgG compared to animals that received the unadjuvanted vaccine (**p = 0.0069; week 4 and *p = 0.0247; week 9). C) IN inoculations elicited antigen-specific IgA that was boosted by a second and third IN application and by the first ID vaccination. D) ID vaccinations failed to elicit detectable vaginal IgA, subsequent IN inoculations generated very low levels of antigen-specific IgA.

Fig 5. Spearman Rank correlations comparing serum, nasal and vaginal mucosal IgG.

A strong positive correlation was found at week 9 and 12 between the amounts of antigen-specific antibody present at each site but this relationship was lost by the end of the vaccination regimen as the nasal compartment exhibited increased levels of specific IgG while vaccine-specific antibody decreased in the serum and vaginal vault. By week 15, 3 weeks after the final vaccinations, the IgG present in the serum and vaginal samples still correlate strongly while that in the nasal compartment exhibits little or no significant relationship.

Fig 6. The antigen-specific polyclonal sera from the different regimens exhibited differing avidity indices throughout the course of the vaccinations.

Low avidity = <25%, medium avidity = 25–50% and high avidity = >50%. After two priming vaccinations (week 35) the sera from animals inoculated via the IN route had low avidity binding to the vaccine antigen while animals that received ID injections had medium–high avidity gp140-antigen reactive sera. Optimally adjuvanted ID injections were able to continue to enhance avidity while the IN inoculation failed to enhance or even maintain the previous high level avidity elicited by prior ID injections.

Fig 7. Sera from each group of animals were tested in a functional HIV neutralising TZM-bl assay.

The serum dilution that was able to achieve half maximal inhibition of HIV infectivity was higher in Group B (3 × ID + 2 × IN) for both a closely sequence matched Clade C virus CN54 (A) and a more sequence divergent Clade C HIV (B). All animals from this group achieved neutralisation compared to only 4 of the 7 animals in Group A (3 × IN + 2 × ID) (C).