Thermophobic Trehalose Glycopolymers as Smart C-Type Lectin Receptor Vaccine Adjuvants

Abstract

Herein, this work reports the first synthetic vaccine adjuvants that attenuate potency in response to small, 1-2 °C changes in temperature about their lower critical solution temperature (LCST). Adjuvant additives significantly increase vaccine efficacy. However, adjuvants also cause inflammatory side effects, such as pyrexia, which currently limits their use. To address this, a thermophobic vaccine adjuvant engineered to attenuate potency at temperatures correlating to pyrexia is created. Thermophobic adjuvants are synthesized by combining a rationally designed trehalose glycolipid vaccine adjuvant with thermoresponsive poly-N-isoporpylacrylamide (NIPAM) via reversible addition fragmentation chain transfer (RAFT) polymerization. The resulting thermophobic adjuvants exhibit LCSTs near 37 °C, and self-assembled into nanoparticles with temperature-dependent sizes (90-270 nm). Thermophobic adjuvants activate HEK-mMINCLE and other innate immune cell lines as well as primary mouse bone marrow derived dendritic cells (BMDCs) and bone marrow derived macrophages (BMDMs). Inflammatory cytokine production is attenuated under conditions mimicking pyrexia (above the LCST) relative to homeostasis (37 °C) or below the LCST. This thermophobic behavior correlated with decreased adjuvant R_g is observed by DLS, as well as glycolipid-NIPAM shielding interactions are observed by NOESY-NMR. In vivo, thermophobic adjuvants enhance efficacy of a whole inactivated influenza A/California/04/2009 virus vaccine, by increasing neutralizing antibody titers and CD4⁺ /44⁺ /62L⁺ lung and lymph node central memory T cells, as well as providing better protection from morbidity after viral challenge relative to unadjuvanted control vaccine. Together, these results demonstrate the first adjuvants with potency regulated by temperature. This work envisions that with further investigation, this approach can enhance vaccine efficacy while maintaining safety.

Keywords: C-type lectin receptors; MINCLE; adjuvants; cord factors; glycopolymers; thermoresponsive polymers.

Figure 1

A) This study creates multivalent thermophobic glycolipid vaccine adjuvants that self‐assemble into nanoparticles which decrease in size, and immunogenicity, in response to temperature. B) Combining the established minimally active motif of glycolipid adjuvants and noting the tolerability for aryl substituents at the 6/6’ positions, we created C) a polymerizable glycolipid adjuvant that activates the macrophage‐inducible C‐type lectin receptor (MINCLE) on innate immune cells and could be readily incorporated within a thermoresponsive polymer.

Scheme 1

Synthesis of glycolipid monomer 1, and thermophobic glycopolymer adjuvants (P1 and P2) along with unadjuvanted thermoresponsive glycopolymer P3 synthesized as a negative control. The synthesis began from an established intermediate, 4,6‐O‐(4‐vinylbenzylidene)‐α,α‐trehalose monomer. Preparation of the glycolipid monomer is then accomplished via a Steglich esterification of the 6’ hydroxyl group with behenic acid to yield the polymerizable adjuvant 1. The copolymer is then synthesized via reversible addition‐fragmentation chain‐transfer (RAFT) polymerization, where 1 and poly‐N‐isoporpylacrylamide (NIPAM) are used to create a rand copolymer. Subsequent re‐initiation with AIBN using P1 as a macro‐chain‐transfer agent and additional NIPAM monomer afforded P2 which contained a congruent adjuvant degree of polymerization, with an additional thermoresponsive NIPAM block.

Figure 2

Characterization of thermophobic glycopolymer adjuvant copolymers. A) Polymers (1 mg mL⁻¹ in PBS) were observed to have lower critical solution temperatures (LCSTs) near 37 °C by UV–Vis. Further dilution diminished this effect and instead resulted in B) polymer nanoparticles (100 µg mL⁻¹ in PBS, pH = 7.4) observed by DLS at 25 °C, 35 °C, 37 °C, and 39 °C. For P1 and P2, particle size decreased as temperature increased. This was the opposite trend observed for the control polymer P3. C) NOESY‐NMR of P1 and P2 confirmed interactions between the behenic ester lipid and isopropyl amide on poly‐N‐isoporpylacrylamide (NIPAM) indicating a possible mechanism for the attenuated potency observed at increased temperatures.

Figure 3

A) A selected binding pose of truncated adjuvant glycolipid subunit 1 docked with the MINCLE crystal structure (PDB: 4ZRW) using Autodock Vina suggested preservation of key bridging interactions between Arg182 and Phe198 in the binding groove. B) P1 and P2 activated HEK‐mMINCLE cells comparable to equimolar amounts of trehalose‐6,6’‐dibehenate (TDB) positive control. polymer concentrations indicated are with respect to agonist subunit. Error bars are standard deviations of the mean for experiments performed in triplicate; ^* p < 0.05, ^** p < 0.01 for P1 or P2 compared to negative control (PBS). C) Proinflammatory response of P1 and P2 in a model JAWS II innate immune cell line treated with 100 µg mL⁻¹ of each adjuvant and incubated for 48 h at the indicated temperatures. Cytokines were assessed by cytometric bead array, and error bars are standard deviations of the mean for experiments performed in triplicate; ^* p < 0.05, ^** p < 0.01 for attenuated cytokine production at 39 °C compared to the same adjuvant at 35 °C.

Figure 4

Characterization of thermophobic glycopolymer adjuvants in primary murine immune cells. Polymers or trehalose‐6,6’‐dibehenate (TDB; 100 µg mL⁻¹) were incubated with A) F4/80⁺ bone marrow derived macrophages (BMDMs) or B) CD11c⁺ bone marrow derived dendritic cells (BMDCs) for 48 h at the indicated temperatures (35–39 °C). Secreted inflammatory cytokines were assessed by cytometric bead array. Error bars are standard deviation of the mean for experiments repeated in hexaplet, ^* p < 0.05, ^** p < 0.01, ^*** p < 0.001 for attenuated cytokines produced at 39 °C relative to 37 °C or 35 °C. See Figures S9 and S10 (Supporting Information) for complete cytokine profiles and additional positive control (LPS). C) Linking these observations to the DLS and NOESYdata, we conclude that the thermally induced change in activity could be due to either decreased particle size itself or enhanced lipid‐poly‐N‐isoporpylacrylamide (NIPAM) shielding interactions that result from this transition.

Figure 5

In vivo performance of thermophobic vaccine adjuvants. A) Mice exhibited normal weight gain postvaccination. B) IL‐6 levels measured from mouse sera 24 h after second vaccination. IL‐6 production was increased for P2 and trehalose‐6,6’‐dibehenate (TDB) compared to unvaccinated cohorts. C) Levels of CD4⁺/44⁺/62L⁺ central memory T‐cells found in lung and lymph node samples were increased in the P2 cohort, consistent with D) latent IL‐6 levels measured 7 days after administering booster vaccination and E) increases in central memory T‐cell populations following the full vaccination schedule as well. F) Weight loss of mice post challenge with influenza indicated protective effects conferred by the vaccine adjuvanted with P2 was comparable to that of alum. This was consistent with neutralizing antibodies assessed by G) Hemagglutination (HA) inhibition assay at day 14 and 28 as well as H) neuraminidase (NA) inhibition assay at day 28 postvaccination each of which indicated that HA and NA neutralization in the P2 adjuvanted cohort is comparable to, or better than, the alum adjuvanted cohort. According to the HA inhibition assay, all adjuvanted vaccines performed better than unadjuvanted control by day 28.