A Small Molecule RIG-I Agonist Serves as an Adjuvant to Induce Broad Multifaceted Influenza Virus Vaccine Immunity

Abstract

Retinoic acid-inducible gene I (RIG-I) is essential for activating host cell innate immunity to regulate the immune response against many RNA viruses. We previously identified that a small molecule compound, KIN1148, led to the activation of IFN regulatory factor 3 (IRF3) and served to enhance protection against influenza A virus (IAV) A/California/04/2009 infection. We have now determined direct binding of KIN1148 to RIG-I to drive expression of IFN regulatory factor 3 and NF-κB target genes, including specific immunomodulatory cytokines and chemokines. Intriguingly, KIN1148 does not lead to ATPase activity or compete with ATP for binding but activates RIG-I to induce antiviral gene expression programs distinct from type I IFN treatment. When administered in combination with a vaccine against IAV, KIN1148 induces both neutralizing Ab and IAV-specific T cell responses compared with vaccination alone, which induces comparatively poor responses. This robust KIN1148-adjuvanted immune response protects mice from lethal A/California/04/2009 and H5N1 IAV challenge. Importantly, KIN1148 also augments human CD8+ T cell activation. Thus, we have identified a small molecule RIG-I agonist that serves as an effective adjuvant in inducing noncanonical RIG-I activation for induction of innate immune programs that enhance adaptive immune protection of antiviral vaccination.

Figure 1.. KIN1148 is a small molecule IRF3 and NFκB agonist that binds to and induces RIG-I signaling activation.

(a) Chemical structure of parent compound KIN1000 and its medicinal chemistry optimized lead KIN1148. (b) Western blot analysis of the phosphorylation state of IRF3 and NFκB at time points up to 12 hrs after HEK293 cells were treated with KIN1148 at 10 or 20μM. Included as comparison are cells treated with 0.5% DMSO (vehicle control) and cells infected with SeV at 40 hemagglutinin units (HAU)/mL. (c) qRT-PCR analysis of Ifit1 and Il6 in WT and RIG-I^−/− MEFs at 18hrs after KIN1148 treatment as compared to respective DMSO controls. Fold induction over DMSO control from 3 independent experiments shown. Error bars represent standard deviation. (d) Western blot detection of input and pulldown products that associate with either biotin-fluorescein (F) or biotin-KIN1148 (K) beads using recombinant proteins generated using a rabbit reticulocyte lysate in vitro transcription and translation system and detected using antibodies specific for the respective epitope tags. (e) BIOMOL green colorimetric analysis of free phosphate levels as an indication of RIG-I ATP hydrolysis activity. Purified recombinant RIG-I protein and ATP were incubated with DMSO or increasing concentrations of KIN1148 in the absence and presence of polyU/UC or xRNA. Each point shows the average concentration of free phosphates in each reaction from at least three independent experiments with error bars showing standard deviation. (f) Western blot detection of endogenous RIG-I and associating signaling cofactors in the HEK293 whole cell lysate (input) and amongst the pulldown products of biotin-KIN1148 beads (K) or biotin-fluorescein bead alone (F). MAVS* indicates higher exposure blot. (g) Western blot detection of various Flag-tagged RIG-I constructs generated by in vitro transcription and translation for interaction with biotin-fluorescein or biotin-KIN1148. Asterisks indicate expected location of band of interest across upper and lower blots in g.

Figure 2.. KIN1148 treatment induces a gene expression profile that shares common and unique properties to that achieved with LPS, SeV-infection and polyU/UC RNA transfection.

Microarray analysis of THP-1 cells treated with KIN1000 or KIN1148 at concentrations up to 20 μM, 0.5 μg/mL LPS or 100 IU/mL IFNβ as compared to DMSO control. For a RIG-I specific gene expression signature, polyU/UC PAMP RNA-transfected cells were compared to those transfected with xRNA whereas SeV-infected cells were compared to mock-infected cells. Differential gene expression is defined as at least a 2-fold change in expression with a Benjamini-Hochberg corrected p-value <0.01 as compared to the respective negative controls. (a) Heatmap showing all genes differentially expressed in at least one treatment with hierarchical clustering and classification by the most highly enriched gene ontology (GO) biological process. (b) Venn diagram showing numbers of differentially expressed genes, up (red) and down (blue), that are shared and unique among cells stimulated with KIN1148, polyU/UC or SeV. (c-d) Analysis showing global gene expression profile with Gene Ontology clustering of enriched genes. Heatmap of genes whose expression is perturbed by compound treatment and designated by gene clustering as (c) antigen presentation or (d) cytokines.

Figure 3.. KIN1148 adjuvants influenza virus vaccination induce humoral responses and confers protection against H1N1 and H5N1 infection.

(a-c) C57BL/6J mice (n=10 mice per group except mock where n=5) were immunized intramuscularly with (a) A/VN/1203/04 H5N1 6+2 SV (H5-SV) or (b) A/Cal/04/09 H1N1 SV (H1-SV) in combination with PBS (IAV-SV + PBS, grey circles), blank liposome (IAV-SV + Vehicle, blue circles), KIN1148 formulated in liposome (IAV-SV + KIN1148, purple circles) or PBS alone (black circles). On day 30 following vaccination mice were challenged intranasally with 5x LD₅₀ of the homologous virus used for vaccination and monitored for survival (top), clinical illness (middle), and weight loss (bottom). (c) Mice were immunized and challenged as described in a using H1-SV. On day 5 p.i., lungs were harvested, and virus titers were determined by plaque assay. n=2(mock)-8 mice/group with data representative of two independent experiments. (d) C57BL/6J mice were immunized intramuscularly PBS alone (PBS, black circles), H5-SV in combination with blank liposome (H5-SV + Vehicle, blue circles), or H5-SV with KIN1148 formulated in liposomes (H5-SV + KIN1148, purple circles). Mice were boosted on day 14 and serum was collected 5 days later for analysis of A/VN/1203/04-specific antibodies. n=6–8 mice/group with data from two pooled experiments shown. (e) C57BL/6J mice were immunized as described in d. Mice were administered a homologous boost on day 14 and serum was collected on day 19. The ability of antibodies in the serum to prevent hemagglutination of homologous (A/VN/1203/04 H5N1 6+2) or heterologous (A/Egypt/N03072/10 H5 7+1, A/Barn Swallow/HK/D10-1161/2010 7+1) recombinant avian influenza viruses was determined. n=10–13 mice/group with data from two pooled experiments shown.

Figure 4.. KIN1148 enhances cellular immune responses following H5N1 vaccination.

C57BL/6J mice were immunized intramuscularly with PBS (black circles), H5-SV in combination with blank liposome (H5-SV + Vehicle) or H5-SV with KIN1148 formulated in liposomes (H5-SV + KIN1148, purple circles). Mice were administered a homologous boost on day 14 and lung-draining lymph nodes and spleen were collected on day 19. (a) The frequency and number of PNA+ GC B cells in dLN was determined by flow cytometry. (b) IAV-specific CD4⁺ T cell responses were assessed by flow cytometry. (c) IAV-specific CD8⁺ T cell responses were assessed by flow cytometry. n=6–8 mice/group with data from two pooled experiments shown.

Figure 5.. KIN1148 promotes DC maturation and human T cell proliferation.

(a) Flow cytometry analysis of maturation markers expressed on the cell surface of human monocytic DCs after 18hr treatment with DMSO, LPS or KIN1148. Histograms (a, top) of CD80, CD83, and CD86 on human monocytic DCs from a representative experiment. Graphs (a, bottom) showing the expression of each maturation marker from at least 6 different healthy donors, expressed as fold change in mean fluorescence intensity (MFI) as compared to DMSO control. (b) PBMCs from healthy HLA-A0201⁺ donors were pulsed with Melan-A/MART-1 peptide in combination with DMSO or KIN1148 in the presence of IFNγ for 24 hrs and then removed from the treatments to be co-cultured with autologous T cells in the presence of IL-2. Cells were collected 11 days later and analyzed by flow cytometry for CD8⁺ T cells that stain positive with the corresponding Melan A/MART-1 tetramer. Density plot from a representative experiment showing CD8 and MART-1 tetramer staining and gating strategy (b, top). Graphs (b, bottom) showing the frequency and numbers of MART-1-tetramer⁺ CD8⁺ T cells from 7 donors.