NOD2/RIG-I Activating Inarigivir Adjuvant Enhances the Efficacy of BCG Vaccine Against Tuberculosis in Mice

Abstract

Tuberculosis (TB) caused by Mycobacterium tuberculosis (MTB) kills about 1.5 million people each year and the widely used Bacille Calmette-Guérin (BCG) vaccine provides a partial protection against TB in children and adults. Because BCG vaccine evades lysosomal fusion in antigen presenting cells (APCs), leading to an inefficient production of peptides and antigen presentation required to activate CD4 T cells, we sought to boost its efficacy using novel agonists of RIG-I and NOD2 as adjuvants. We recently reported that the dinucleotide SB 9200 (Inarigivir) derived from our small molecule nucleic acid hybrid (SMNH)^® platform, activated RIG-I and NOD2 receptors and exhibited a broad-spectrum antiviral activity against hepatitis B and C, Norovirus, RSV, influenza and parainfluenza. Inarigivir increased the ability of BCG-infected mouse APCs to secrete elevated levels of IL-12, TNF-α, and IFN-β, and Caspase-1 dependent IL-1β cytokine. Inarigivir also increased the ability of macrophages to kill MTB in a Caspase-1-, and autophagy-dependent manner. Furthermore, Inarigivir led to a Capsase-1 and NOD2- dependent increase in the ability of BCG-infected APCs to present an Ag85B-p25 epitope to CD4 T cells in vitro. Consistent with an increase in immunogenicity of adjuvant treated APCs, the Inarigivir-BCG vaccine combination induced robust protection against tuberculosis in a mouse model of MTB infection, decreasing the lung burden of MTB by 1-log10 more than that afforded by BCG vaccine alone. The Inarigivir-BCG combination was also more efficacious than a muramyl-dipeptide-BCG vaccine combination against tuberculosis in mice, generating better memory T cell responses supporting its novel adjuvant potential for the BCG vaccine.

Keywords: BCG vaccine; NOD2; RIG-I; adjuvants; antigen presentation; inarigivir; inflammasome; tuberculosis.

Figure 1

Binding of SB 9000 (Inarigivir) to retinoic acid-inducible gene I (RIG-I). (A) Sandwich ELISA method: Binding of SB 9000 to RIG-I by using SB 9000 derivatized with biotin at the 5′-OH group of the dinucleotide. ELISA microplate coated with anti-DDK antibody at 1:200 dilution in coating buffer (0.1M carbonate, pH 9.6), followed by 1 µg/ml DDK-tagged RIG-I recombinant protein. Biotinylated SB 9000 was then added as indicated and the SB 9000 bound to RIG-I was detected using HRP-conjugated Streptavidin followed by development with TMB substrate solution and absorbance measurement at 450nm (mean ± SD; n = 2). (B) Surface Plasmon Resonance Method: Human RIG-I (hRIG-I) was generated in mammalian cells using Expi293F cells and DDX58 - HaloTag(R) human ORF in pFN21A. The protein was then purified using HaloTag^® Mammalian Protein Purification System. SPR was measured by using various concentrations of biotinylated SB 9000 dissolved in water manually printed onto PlexArray Nanocapture Sensor Chip at 40% humidity. The signal changes after binding and washing (in AU) were recorded as the assay value. Selected protein-grafted regions in the SPR images were analyzed, and the average reflectivity variations of the chosen areas were plotted as a function of time. Real-time binding signals were recorded and analyzed by Data Analysis Module. Kinetic analysis was performed using BIA evaluation 4.1 software. Data are representative of 2 independent experiments, carried out in duplicates and values are expressed as mean ± SD.

Figure 2

Small molecule nucleotide hybrids (SMNHs, aka. SB series) enhance antigen presentation by BCG vaccine infected macrophages and increase Th1 cytokine secretion. C57Bl/6 mouse bone marrow derived MΦs and dendritic cells (DCs) (aka. APCs) were treated with SMNHs (SB series) at indicated doses, followed by infection with Mycobacterium bovis Bacille Calmette Guerin (BCG) for 4 h (MOI = 1). Mycobacterial Antigen-85B (Ag85B) derived p25 peptide specific BB7 CD4 T cells were overlaid and supernatants collected after 18 h were tested for IL-2 using sandwich ELISA. (A) Initial screening of SB series of SMNHs including Inarigivir (NOD2/RIG-I agonist), SB-2 (TLR-7/9 agonist), and SB-3 and SB-4 (dinucleotides related to Inarigivir). *p values vs. BCG alone. (B, C) Comparison of Inarigivir and its resynthesized isomers Rp-Inarigivir and Sp-Inarigivir (10 µM) with the LPS agonist of TLR-4 (1 µg/ml) and related SB compounds during antigen presentation. *p values for groups compared. (D) Structures of Rp-Inarigivir and Sp-Inarigivir. (E) MΦs and DCs were treated or not with Rp-Inarigivir and Sp-Inarigivir (10 µM), followed by BCG infection. Supernatants collected at 18 h tested for Th1 cytokine levels using sandwich ELISA. *p values for treated vs. BCG alone (*<0.01; **< 0.009; p values, 1-way ANOVA with Tukey’s posttest). Data are representative of two independent experiments carried out in duplicates and values are expressed as mean ± SD.

Figure 3

Inarigivir activates NOD2/RIG-I receptors enhancing Capsase-1 dependent IL-1β secretion and NOD2 dependent antigen presentation. HEK cells were used for assay of receptor activation (A) and wild type C57Bl/6 (B–F) or NOD2^−/− (G–H) mouse derived APCs were left either untreated or treated with Inarigivir, Rp-Inarigivir and Sp-Inarigivir and SB-2 or SB-3 (10 µM), followed by inhibitors as indicated, and where indicated, a 4-h infection with BCG (MOI = 1). Supernatants were tested for cytokines at 18 h or APCs overlaid with BB7 CD4 T cells for antigen presentation assay to measure IL-2. (A) Induction of NF-kB activity by Inarigivir (mix of Rp-Inarigivir and Sp-Inarigivir) through NOD2 and RIG-I activation. HEK-293 cells were transfected with pcDNA, NOD2, RIG-I, and NF-kB-luciferase. The cells were then incubated with Inarigivir (10 µM). Following 12 h incubation, Luciferase activity was measured and presented as mean ± S.D. from three independent experiments. (B) Rp-Inarigivir and Sp-Inarigivir combinations with BCG increase IL-1β secretion which is decreased after Caspase-1 blockade using Z-YVAD-fmk (50 µM). (C–E) IFN-β, TNF-α, and IL-12 cytokines are not affected by the Caspase-I blockade, although Inarigivir, SB2 and SB3 enhance IFN-β levels. (F) Caspase-1 blockade decreases antigen presentation. (G, H) Rp-Inarigivir and Sp-Inarigivir enhance antigen presentation by BCG infected wt-APCs, which is decreased in NOD2^−/− APCs. (* < 0.01 ** < 0.006 p values using 1-way ANOVA with Tukey’s posttest). Data are representative of 2 independent experiments carried out in duplicates and values are expressed as mean ± SD.

Figure 4

Inarigivir enhances the bactericidal function of mouse and human THP-1 macrophages against M. tuberculosis. Wt-C57Bl/6 mouse derived MΦs and phorbol myristyl acetate activated THP-1 human MΦs were treated with Rp-Inarigivir and Sp-Inarigivir or Inarigivir (10 µM) followed by infection with M. tuberculosis and incubated maintaining >90% viability of MΦs (MOI = 1). On day three, lysates were plated for colony counts (CFUs) of MTB. (A) Rp-Inarigivir and Sp-Inarigivir enhance the bactericidal function mouse derived MΦs, which is decreased by prior incubation of MΦs with either Z-YVAD-fmk inhibitor of Caspase-1 (50 µM) or 3-methyladenine (100 µM), which inhibits autophagy through PI-3 Kinase. (B) Rp-Inarigivir and Sp-Inarigivir enhance the bactericidal function THP-1 MΦs. (**p < 0.007; 1-way ANOVA with Tukey’s posttest). Data are representative of 2 independent experiments carried out in duplicates and values are expressed as mean ± SD.

Figure 5

Inarigivir enhances the efficacy of BCG vaccine against aerosol induced tuberculosis of mice and induces an expansion of memory T cells. (A) C57Bl/6 mice were immunized subcutaneously with one dose of BCG (1 x 10⁶ CFU) alone or mixed with 25 µg per dose of Inarigivir or indicated SB compounds. Four weeks later, mice were aerosol challenged with of ~200 CFUs of M. tuberculosis (Erdman strain) in the lungs of mice and bacterial counts (CFUs) of lungs and spleens at indicated times were determined by plating their homogenates on 7H11 agar. Spleen derived T cells were analyzed using flow cytometry for memory markers. (B) Inarigivir-BCG vaccine combination is better than BCG combinations with SB-2, SB3 and SB4 in reducing the MTB burden of lungs and spleens of mice (10 mice per group; 5 mice each from two experiments; p values; 2-way ANOVA with Dunette’s post hoc test). Individual value of Log10 CFUs of organs for each mouse from 2 independent experiments are shown. (C) Histogram analysis of spleen derived T cells (n = 3 per group per experiment and represents the results of 2 independent experiments) for T cells using flow cytometry at the time of sacrifice. (D) Inarigivir-BCG vaccine combination induces higher levels of memory precursor-effector T cells (MPECs; CD62L⁺ CCR7⁺ CD127⁺ CD44^+/−) and comparable short lived-effector T cells (SLECs; CD62L⁻ CCR7⁻CD127⁺CD44^+/−) (* < 0.005, p values, t test). Data are representative of 2 independent experiments carried out in duplicates (after pooling 3 mice from per group, per experiment) and values are expressed as mean ± SD.

Figure 6

Rp-Inarigivir and Sp-Inarigivir are better adjuvants for BCG vaccine compared to muramyl dipeptide (MDP) in mice. (A) C57Bl/6 mice were vaccinated with BCG alone, BCG mixed with Rp-Inarigivir, Sp-Inarigivir (10⁶ CFU of BCG mixed with 25 µg/dose of adjuvants given once, s.c.) or MDP (25 µg/dose) followed by an aerosol challenge with M. tuberculosis (~200 CFUs/mouse) for evaluation of protection as in Fig. 4A. Rp-Inarigivir and Sp-Inarigivir combinations with BCG are more effective than BCG alone or BCG combination with MDP in reducing the growth of MTB in the lungs (5 mice per group per experiment; p values, 2-way ANOVA with Dunette’s post hoc test). Individual value of Log10 CFUs of organs for each mouse from 2 independent experiments are shown (10 mice per group; 5 mice each from two experiments). (B, C) Inarigivir and Sp-Inarigivir combinations with BCG induce better levels of cytokine positive T cells in the spleens (panel B) and comparable levels of BCG antigen specific CD8 T cell tetramers specific for ESAT6, MTB32, Psts, TB10.4 antigens (3 mice per group per experiment; * < 0.01 p values vs. BCG + MDP groups; 1-way ANOVA. Data are representative of 2 independent experiments carried out in duplicates (after pooling 3 mice per group per experiment) and values are expressed as mean ± SD. (D) Inarigivir and Sp-Inarigivir combinations with BCG induce better levels of MPECs in the spleens of vaccinated mice (3 mice per group; * < 0.01 p values; 1-way ANOVA). Data are representative of 2 independent experiments carried out in duplicates (after pooling 3 mice per group per experiment) and values are expressed as mean ± SD.

Figure 7

Inarigivir and Sp-Inarigivir combinations with BCG induce better T cell responses allowing mice to mount a better recall response against re-challenge with tuberculosis. (A) C57Bl/6 mice were vaccinated with Inarigivir and Sp-Inarigivir combinations with BCG and infected with MTB as in Figure 4 for protection. As indicated, a separate group of challenged mice were cleared of vaccine and MTB organisms using isoniazid (25 mg/kg oral daily) and rifampin (10 mg/kg oral daily) for 3 weeks followed by resting for 1 week and re-challenged with an aerosol dose of virulent MTB (~200 CFUs per mouse). Four weeks later, protection was evaluated through CFU counts of lungs and spleens (10 mice per group; 5 mice each from two experiments) and day 120 lung T cell analysis (3 mice per group). (B, C) Rp-Inarigivir and Sp-Inarigivir combinations with BCG protect better than BCG both after primary and secondary rechallenge (p values, 2-way ANOVA with Dunette’s post hoc test). Individual value of Log10 CFUs of organs for each mouse from 2 independent experiments groups are shown. (D) Rp-Inarigivir combination with BCG increases the numbers of ESAT6⁺ IFN-γ⁺ and TB10.4⁺ IFN-γ⁺ CD8 T cells. (* p < 0.01, t test). Data are representative of two independent experiments carried out in duplicates (3 mice pools per group per experiment) and values are expressed as mean ± SD.