The priming, boosting, and restoration of memory cytotoxic CD8+ T lymphocytes by vaccination or immunotherapy in vivo is an area of active research. Particularly, nucleic acid–based compounds have attracted attention due to their ability to elicit strong Ag-specific CTL responses as a vaccine adjuvant. Nucleic acid–based compounds have been shown to act as anticancer monotherapeutic agents even without coadministration of cancer Ag(s); however, so far they have lacked efficacy in clinical trials. We recently developed a second-generation TLR9 agonist, a humanized CpG DNA (K3) complexed with schizophyllan (SPG), K3-SPG, a nonagonistic Dectin-1 ligand. K3-SPG was previously shown to act as a potent monoimmunotherapeutic agent against established tumors in mice in vivo. In this study we extend the monoimmunotherapeutic potential of K3-SPG to a nonhuman primate model. K3-SPG activated monkey plasmacytoid dendritic cells to produce both IFN-α and IL-12/23 p40 in vitro and in vivo. A single injection s.c. or i.v. with K3-SPG significantly increased the frequencies of activated memory CD8+ T cells in circulation, including Ag-specific memory CTLs, in cynomolgus macaques. This increase did not occur in macaques injected with free CpG K3 or polyinosinic-polycytidylic acid. Injection of 2 mg K3-SPG induced mild systemic inflammation, however, levels of proinflammatory serum cytokines and circulating neutrophil influx were lower than those induced by the same dose of polyinosinic-polycytidylic acid. Therefore, even in the absence of specific Ags, we show that K3-SPG has potent Ag-specific memory CTL response–boosting capabilities, highlighting its potential as a monoimmunotherapeutic agent for chronic infectious diseases and cancer.

In the past decade, there has been increased attention on the use of natural ligands or synthetic agonists for well-defined pattern-recognition receptors (PRRs), including TLRs, on professional APC such as dendritic cells (DCs), macrophages, and monocytes (1, 2) as adjuvants, either alone or with various formulations, to elicit strong CTL responses.

Circulating DCs can be divided into two subsets: plasmacytoid DCs (pDCs) and conventional DCs (cDCs). It is well known that pDCs express mainly TLR7 and 9 and produce type I IFN upon stimulation and that cDCs express several TLRs (3). Synthetic oligodeoxynucleotides containing CpG (CpG ODN) are recognized by TLR9, and are classified into at least four groups on the basis of differences in their structures and immunostimulatory effects: A/D, B/K, C, and P type (4). The B/K type, which encodes multiple CpG motifs on a phosphorothioate backbone, is being developed as an adjuvant against infectious diseases including hepatitis B (57), malaria (811), influenza (12), and cancer (1317). The most widely studied B/K type CpG-ODN TLR9 agonist is CpG7909 (also known as CpG2006). However, a phase III clinical trial for patients with non–small-cell lung cancer evaluating CpG7909 with or without gemcitabine/cisplatin failed to show significant efficacy (18). This may be because B/K type CpG ODN is unable to stimulate adequate amounts of type I or type II IFN in vivo. To overcome these limitations, from B/K type CpG ODN (K3) and a nonagonistic Dectin-1 ligand schizophyllan (SPG) we developed K3-SPG. K3-SPG induces the robust production of type I and type II IFN and Th1 type cytokines in human PBMCs and acts as a strong adjuvant for influenza split vaccine in mice and nonhuman primates (NHPs) to elicit Ag-specific Abs (19). K3-SPG can also stimulate strong Ag-specific CTL responses and induce the extensive production of anticancer cytokines, such as type I IFN and IL-12, to function as an anticancer monotherapy agent in mice (20, 21). However, type I IFN, but not IL-12, is required for IFN-γ production by human PBMCs and for the proliferation of CMV-specific CD8+ T cells in vitro (20). Nevertheless, it remains unclear whether humanized CpG DNA molecules can induce IL-12 or what the effect of IL-12 on optimal CTL responses might be in vivo. Because the innate immune system in humans is more similar to that of NHPs than to that of mice, we took advantage of the NHP model to evaluate the potency and safety of K3-SPG as an anticancer monotherapy agent in vivo.

In this study, we used cynomolgus macaques to address the following questions to estimate the potential of K3-SPG as an immunotherapeutic agent: 1) whether K3-SPG could induce type I IFN and IL-12 in vivo; 2) whether a single injection of K3-SPG could stimulate and reactivate Ag-specific memory CTLs; and 3) whether K3-SPG induces aggressive inflammatory responses.

K3 (5ʹ-ATC GAC TCT CGA GCG TTC TC-3ʹ) was synthesized by Gene Design (Osaka, Japan). Polyinosinic-polycytidylic acid [poly(I:C)] and SPG were purchased from InvivoGen (CA). The method of K3-SPG formulation was previously described (19).

Male cynomolgus macaques (Macaca fascicularis) ranging from 4 to 8 y were maintained and used in this study. All animals were maintained in accordance with the Guide for the Care and Use of Laboratory Animals, National Institutes of Health Report no. 85-23 (Department of Health and Human Services, Bethesda, MD, 1985), and were housed in the Tsukuba Primate Research Center of the National Institute of Biomedical Innovation after full protocol approval from the Institutional Animal Care.

In total, 19 male cynomolgus macaques (Macaca fascicularis) ranging from 4 to 8 y of age were divided into six groups (n = 3 per group) for either s.c. or i.v. injection with 2 mg of K3, K3-SPG, or poly(I:C). (i.e., three animals received i.v. injections and three animals received s.c. injections of each adjuvant). Sample collection, and PBMC and plasma isolation from blood samples containing EDTA by Ficoll-Paque PLUS (GE Healthcare, Buckinghamshire, U.K.) were performed as previously described (22). Either fresh or frozen PBMCs were in FBS containing 10% DMSO and stored in liquid-phase nitrogen until analysis. Complete blood counts were performed using an automated hematology analyzer (Sysmex K-4500; Sysmex, Hyogo, Japan). The pie charts in Fig. 4B and Supplemental Fig. 4A were created using the ggplot2 package in R (23). The mean of cell counts was used for visualization.

For analysis of APCs, cells were analyzed using a modified LSRII flow cytometer (BD Biosciences, MA) as described previously (22, 24). Briefly, PBMCs were incubated in 1 ml of RPMI 1640 medium supplemented with 10% FBS plus penicillin/streptomycin in the absence or presence of each adjuvant for 1 h. After that, monensin (0.7 μg/ml; BD Biosciences) and brefeldin A (1 μg/ml; BD Biosciences) were added and incubated for 13 h for APC analysis and 6 h for T cell analysis.

Next, the cells were surface stained with the following: CD3-APC-Cy7 (SP34-2), CD4-BV605 (L200), CD8-APC-H7 (SK1), CD20-APC-H7 (L27), CD95-PE-Cy5 (DX2), CD107a-BV421 (H4A3), CD123-PE-Cy7 (7G3), and IFN-γ–Alexa Fluor 700 (B27) (all from BD Biosciences); CCR7-PerCP-Cy5.5 (G043H7), CD4-BV570 (OKT-4), CD8-BV510 (SK1), CD11c-PE-Cy5/PE (3.9), CD14-BV605/Alexa Fluor 488 (M5E2), CD16-BV510 (3G8), CD69-PE and PE-Cy7 (FN50), CD83-APC (HB15e), and CD137-PE-Cy7 (4B4-1) (all from BioLegend, CA); CD28-ECD (CD28.2), (all from Beckman Coulter, CA); HLA-DR–TRPE (TU36) (Life Technologies, CA). LIVE/DEAD Fixable Violet and Blue Dead Cell Stain Kit (Life Technologies) were used to exclude dead cells from the analysis.

Following permeabilization (Cytofix/Cytoperm kit; BD Biosciences), cells were stained for IFN-α–FITC (LT27:295) (Miltenyi Biotec, Gladbach, Germany), IFN-γ–Alexa Fluor 700 (B27) (BD Biosciences), and IL-12/23 p40-eFluor450 (C8.6) (eBioscience, CA). For analysis of T cells, frozen PBMCs were thawed and then incubated in the absence or presence of hepatitis B surface (HBs) overlapping peptides [PepMix hepatitis B virus (HBV); JPT Peptide Technologies, Berlin, Germany] in a final concentration of 2 μg/ml for each peptide with αCD107a. After 6 h, cells were surface stained for αCD3, αCD4, αCD8, αCD28, αCD69, and αCD95. Following permeabilization, cells were stained for αCD137 and αIFN-γ. The total memory CD8+ cells were CD3+CD8+CD95+ cells. All Abs for flow cytometry analyses are antihuman Abs cross-reacting to NHPs. Between 3 × 105 and 1 × 106 events were collected in each case. Data were analyzed using FlowJo software version 9.8.2 (Tree Star, OR).

For in vitro stimulation, the culture supernatants of fresh PBMCs were collected after 24 h of stimulation with each adjuvant (final concentration was 10 μg/ml). The cytokines and chemokines in culture supernatants from in vitro stimulation or serum from in vivo administration were measured using ELISA kits for monkey IFN-α2 (PBL Assay Science, NJ) and human IL-12/23 p40 (Mabtech, Nacka Strand, Sweden). We also performed a ProcartaPlex NHP immunoassay (IFN-γ, IL-1β, IL-1RA, IL-6, MCP-1, and TNF-α; eBioscience) and Milliplex (IL-6, IL-1RΑ, IL-1β, TNF-α, and MCP-1; Merck Millipore, Darmstadt, Germany) on a Bio-Plex 200 device (Bio-Rad, CA).

Each subset was sorted from PBMCs using a FACSAria II system (BD Biosciences) based on the gating strategy shown in Supplemental Fig. 1B. mRNA was isolated using an Absolutely RNA Nanoprep kit (Agilent Technologies, CA). cDNA was synthesized from purified mRNA using a PrimeScript RT reagent kit (Takara, Siga, Japan) and 100 ng of cDNA was used as a template for real-time PCR for each reaction. Real-time PCR was performed on an LC480 (Roche, Mannheim, Germany) system with Premix Ex Taq (Takara). The following primers and probes were used: TLR3 (Hs00152933_m1), TLR9 (Hs00370913_s1), IPS-1 (Rh02790526_m1), MDA5 (Hs00223420_m1), RIG-I (Hs01061436_m1) for PRR expression and CCL3 (Rh02788104_gH), CXCL11 (Rh02621763_m1), IL12A(Rh02621733_m1), IRF7 (Rh02839174_m1), ISG15 (Rh02915441_g1), IFIT1 (Rh00929909_m1), IFNA1 (Hs00256882_s1), IFNA2 (Rh02902794_s1), MX1 (Rh02842279_m1), TNFA (Rh02789784_m1), IL6 (Rh02621719_u1), IL1B(Rh02621711_m1), IL23A(Rh02872166_m1), and IFNB1(Rh02913347_s1) for gene expression ex vivo. Eukaryotic 18S rRNA (Hs99999901_s1) was used as a housekeeping gene. The analyses for PRR expression were performed based on the 2−Δcycle threshold method and for ex vivo gene expression were performed based on 2−ΔΔcycle threshold. The heatmaps in Fig. 2 were created using ggplot2 (23) and the RColorBrewer package in R (25). The log-transformed medians of mRNA expression levels were used for visualization.

Experimental variables were analyzed using a two-tailed unpaired t-test with Welch correction, a two-tailed paired t-test, Mann–Whitney U test, one-way ANOVA, and a two-tailed Wilcoxon matched-pairs signed-rank test. A p value < 0.05 was considered significant. GraphPad Prism statistical analysis software (Version 6; GraphPad Software, CA) was used throughout.

To investigate the effect of humanized nanoparticulate CpG, namely K3-SPG, in NHPs, we first measured the production of type I and type II IFN, and IL-12 from monkey PBMCs following in vitro stimulation with K3-SPG. When monkey PBMCs were stimulated by 10 μg/ml of naked CpG ODN (K3) or a nanoparticulate CpG complexed with β-glucan (K3-SPG), compared with either the no-stimulation control or 10 μg/ml of dsRNA TLR3 ligand, poly(I:C), the level of IFN-α2 secreted in culture supernatant at 24 h poststimulation was significantly upregulated by both K3 and K3-SPG, but not poly(I:C) (Fig. 1A). K3-SPG stimulated even higher levels of IFN-α2 production than K3 stimulation (p = 0.0078). The level of IL-12/23 p40 in culture supernatant from monkey PBMCs was significantly increased by K3 or K3-SPG compared with the no stimulation, although their levels were comparable (Fig. 1A). Additionally, K3-SPG showed significantly higher levels of IFN-γ production than those induced by K3 or poly(I:C). When PBMCs were stimulated by poly(I:C) in vitro, the production of IL-6 and MCP-1 was significantly higher than in control nonstimulation conditions (Supplemental Fig. 1A). IL-23 p19 could not be detected by any stimulation (data not shown). Thus, although all of the immunostimulatory molecules, K3, K3-SPG, and poly(I:C), are potent activators of innate immune responses in monkey PBMCs in vitro, their cytokine profiles appear to be distinct from each other, both qualitatively and quantitatively.

FIGURE 1.

K3-SPG, humanized CpG DNA complexed with β-glucan provokes innate immune activation of cynomolgus macaques in vitro and in vivo. (A) Production of IFN-α2, IL-12/23 p40, and IFN-γ from monkey PBMCs (n = 8) at 24 h after 10 μg/ml of each nucleic acid–based adjuvant stimulation in vitro. Solid lines represent the median and p values were calculated using a two-tailed Wilcoxon matched-pairs signed-rank test for comparison with normal, poly(I:C), or K3. *p < 0.05, **p < 0.01. (B) The frequency of IFN-α and IL-12/23 p40–positive cells from each subset by nucleic acid–based adjuvant stimulation in vitro after 14 h (IFN-α: n = 7, IL-12/23 p40: n = 5). Data were calculated following background subtraction of the nonstimulation condition. The p values were calculated using a two-tailed Wilcoxon matched-pairs signed-rank test for comparison with monocytes, cDCs, or pDCs. *p < 0.05. (C) Kinetics of IFN-α2 and IL-12/23 p40 production in serum after injection of nucleic acid–based adjuvants in vivo in male 4–8 y old cynomolgus macaques (n = 5), as measured by ELISA. Data show means ± SEM. The p values were calculated using a Mann–Whitney U test at the indicated time points for comparison with poly(I:C). *p < 0.05.

FIGURE 1.

K3-SPG, humanized CpG DNA complexed with β-glucan provokes innate immune activation of cynomolgus macaques in vitro and in vivo. (A) Production of IFN-α2, IL-12/23 p40, and IFN-γ from monkey PBMCs (n = 8) at 24 h after 10 μg/ml of each nucleic acid–based adjuvant stimulation in vitro. Solid lines represent the median and p values were calculated using a two-tailed Wilcoxon matched-pairs signed-rank test for comparison with normal, poly(I:C), or K3. *p < 0.05, **p < 0.01. (B) The frequency of IFN-α and IL-12/23 p40–positive cells from each subset by nucleic acid–based adjuvant stimulation in vitro after 14 h (IFN-α: n = 7, IL-12/23 p40: n = 5). Data were calculated following background subtraction of the nonstimulation condition. The p values were calculated using a two-tailed Wilcoxon matched-pairs signed-rank test for comparison with monocytes, cDCs, or pDCs. *p < 0.05. (C) Kinetics of IFN-α2 and IL-12/23 p40 production in serum after injection of nucleic acid–based adjuvants in vivo in male 4–8 y old cynomolgus macaques (n = 5), as measured by ELISA. Data show means ± SEM. The p values were calculated using a Mann–Whitney U test at the indicated time points for comparison with poly(I:C). *p < 0.05.

Close modal

We therefore sought to determine which cell subset is responsible for the production of type I IFN and IL-12/23 p40 in vitro. We next measured the mRNA expression of each PRR to the nucleic acid–based adjuvants used in this study [TLR3, TLR9, RIG-I, MDA5, and IPS-1 on sorted monocytes (HLA-DR+ CD14+), cDCs, (HLA-DR+ CD14 CD11c+ CD123), and pDCs (HLA-DR+ CD14 CD11c CD123+) (Supplemental Fig. 1B) from PBMCs]. However, there were no significant differences in the PRRs between each sorted subset (Supplemental Fig. 1C).

Therefore, we used flow cytometry analysis to detect intracellular cytokine production. We stimulated monkey PBMCs with poly(I:C), K3, or K3-SPG for 14 h and analyzed cytokine production of the two DCs and monocytes (Supplemental Fig. 2). After subtracting the background values from the nonstimulation control, the frequency of IFN-α–positive cells from pDCs following K3 and K3-SPG stimulation was higher than that from cDCs and monocytes. The frequency of IL-12/23 p40–positive cells from pDCs following K3-SPG stimulation tended to be higher than that from cDCs and monocytes although there was no statistical significance (Fig. 1B). Our results clearly suggest that K3-SPG stimulation in vitro induced the production of type I IFN and possibly IL-12/23 p40, mainly by pDCs.

To confirm these in vitro data, we measured the production of IFN-α2 and IL-12/23 p40 after in vivo s.c. or i.v. administration of each nucleic acid–based adjuvant in adult cynomolgus macaques (4–8 y old). As shown in Fig. 1C, we observed that the level of IFN-α2 in serum 3 h after injection of K3 and K3-SPG was significantly higher than following poly(I:C) injection at 3 h postinjection. The levels of IL-12/23 p40 following K3-SPG but not K3 injection were significantly higher than poly(I:C) injection at 1 d postinjection. We could not detect IFN-γ production in serum for all monkeys.

Taken together, these results indicate that K3-SPG is a strong inducer of type I IFN and IL-12/23 p40 in vivo even in NHPs.

To assess the molecular signatures of the systemic immune responses mediated by three adjuvants, K3, K3-SPG, and poly(I:C), in cynomolgus macaques, we sorted monocytes, cDCs, and pDCs from the injected monkey PBMCs ex vivo according to the sorting panel shown in Supplemental Fig. 1B, and analyzed the levels of mRNA expression in each subset by quantitative RT-PCR. We selected tnfa, il6, and il1b as markers of proinflammatory cytokines (26); il23a and il12a as markers of Th1/17 cytokines (27); mx1, isg15, irf7, ifit1, ifnb1, ifna1, and ifna2 as markers of IFN-related genes (28); and ccl3 and cxcl11 as markers of chemokines (29). First, as shown in Fig. 2, the gene signatures of nanoparticulate K3-SPG at various routes, times, and cell types display three characteristics distinct from that of K3, soluble TLR9 ligand, and poly(I:C), a TLR3 agonist. One is that K3-SPG–specific mRNA upregulation for most IFN-related genes, cytokines, and chemokines was evident in the pDC fraction at 8 h after s.c. injection, and at 3 h (and slightly less at 8 h) after i.v. injection. It is of note that not only IFN-related genes such as Ifna2, but also Il12a and Il23a, were upregulated specifically on pDCs by K3-SPG (Fig. 2). The second characteristic of K3-SPG is that K3-SPG specifically stimulated cDC expression of mx1 (3 and 8 h), and ifti1 on cDCs by the s.c., but not i.v. injections (Fig. 2). The third is that poly(I:C) displayed distinct signatures specifically activating monocytes and cDCs upregulating IFN-related genes such as mx1 and Ifit1 by s.c., but not i.v. injection, whereas poly(I:C) also activated pDCs to upregulate Il-6 and Il-1b by both s.c and i.v. injection (Fig. 2). Thus, these data obtained by ex vivo gene signatures of each subsets of APCs after s.c. and i.v. injections of these three adjuvants demonstrate that they target the distinct cell types to activate different genes at the various time points, indicating that their qualitative differences can be attributed to the target cell types and their spatiotemporal regulations of intra- and intercellular signaling pathways.

FIGURE 2.

Alteration of gene expression in each subset after injection of nucleic acid–based adjuvants ex vivo. Heatmap is shown for ex vivo gene signatures of sorted monocytes, cDCs, and pDCs from PBMCs after s.c. or i.v. injection, obtained by each adjuvant injection as indicated (n = 3 for each adjuvant injection). The each column represents name of the adjuvant injected (left, 3 h postinjection; right, 8 h postinjection), and the horizontal row represents each gene selected as typical proinflammatory cytokines, IFN stimulated genes, and chemokines (summarized as cytokines, IFN-related genes, and chemokines, respectively). Warm color indicates a high expression level as indicated.

FIGURE 2.

Alteration of gene expression in each subset after injection of nucleic acid–based adjuvants ex vivo. Heatmap is shown for ex vivo gene signatures of sorted monocytes, cDCs, and pDCs from PBMCs after s.c. or i.v. injection, obtained by each adjuvant injection as indicated (n = 3 for each adjuvant injection). The each column represents name of the adjuvant injected (left, 3 h postinjection; right, 8 h postinjection), and the horizontal row represents each gene selected as typical proinflammatory cytokines, IFN stimulated genes, and chemokines (summarized as cytokines, IFN-related genes, and chemokines, respectively). Warm color indicates a high expression level as indicated.

Close modal

Previously, we reported that K3-SPG can induce potent CTL responses and anticancer effects without Ag in mice (1921); however, the effects of K3-SPG on the immune system of NHPs remain unknown. To confirm the potency of K3-SPG in NHPs, we first verified the activation of total memory CD8+ T cells by a single injection of 2 mg of K3, K3-SPG, or poly(I:C) in vivo (Fig. 3A). The frequency of CD69+ memory CD8+ T cells, representing cells recently activated by in vivo injection of K3-SPG, was significantly increased by either s.c. or i.v. injection at 1 and 3 d postinjection but not by K3 or poly(I:C) (Fig. 3B, Supplemental Fig. 3A, 3B). These data strongly suggest that the activation of total memory CD8+ T cells can be induced by both s.c. and i.v. injection of K3-SPG, but not K3 or poly(I:C).

FIGURE 3.

Activation and induction of CTLs following single injection of nucleic acid–based adjuvants without Ag in vivo. (A) Adjuvant injection protocol in NHPs. Cynomolgus macaques were s.c. or i.v. injected with 2 mg of K3, K3-SPG, or poly(I:C). Blood samples were collected preinjection and 1 and 3 d after injection, and activation of memory CD8+ T cells was analyzed by flow cytometry. (B) Frequency of CD69+ memory CD8+ T cells at 0 h and at 1 and 3 d after injection of nucleic acid–based adjuvants (n = 6). Open circles are the s.c. injection group and closed circles are the i.v. injection group. Solid lines represent the median and p values were calculated using a two-tailed Wilcoxon matched-pairs signed-rank test for comparison with preinjection. *p < 0.05. (C) Adjuvant injection protocol in vaccinated NHPs. Cynomolgus macaques were i.m. vaccinated with 10 μg of HBs recombinant proteins with alum in a final volume of 500 μl at 0 and 4 wk. At 20 wk postvaccination, animals were s.c. injected with 2 mg of K3 or K3-SPG. Blood samples were collected preinjection and at 3 d after injection. (D) Frequencies of total HBs-specific IFN-γ–, CD137–, and/or CD107a–positive cells in memory CD8+ T cells before and after injection of K3 or K3-SPG (n = 5). The monkey PBMCs were stimulated by the HBs overlapping peptides for 6 h and intracellular staining analyses were performed by flow cytometry. The p values were calculated using a two-tailed paired t test for comparison with preinjection. *p < 0.05.

FIGURE 3.

Activation and induction of CTLs following single injection of nucleic acid–based adjuvants without Ag in vivo. (A) Adjuvant injection protocol in NHPs. Cynomolgus macaques were s.c. or i.v. injected with 2 mg of K3, K3-SPG, or poly(I:C). Blood samples were collected preinjection and 1 and 3 d after injection, and activation of memory CD8+ T cells was analyzed by flow cytometry. (B) Frequency of CD69+ memory CD8+ T cells at 0 h and at 1 and 3 d after injection of nucleic acid–based adjuvants (n = 6). Open circles are the s.c. injection group and closed circles are the i.v. injection group. Solid lines represent the median and p values were calculated using a two-tailed Wilcoxon matched-pairs signed-rank test for comparison with preinjection. *p < 0.05. (C) Adjuvant injection protocol in vaccinated NHPs. Cynomolgus macaques were i.m. vaccinated with 10 μg of HBs recombinant proteins with alum in a final volume of 500 μl at 0 and 4 wk. At 20 wk postvaccination, animals were s.c. injected with 2 mg of K3 or K3-SPG. Blood samples were collected preinjection and at 3 d after injection. (D) Frequencies of total HBs-specific IFN-γ–, CD137–, and/or CD107a–positive cells in memory CD8+ T cells before and after injection of K3 or K3-SPG (n = 5). The monkey PBMCs were stimulated by the HBs overlapping peptides for 6 h and intracellular staining analyses were performed by flow cytometry. The p values were calculated using a two-tailed paired t test for comparison with preinjection. *p < 0.05.

Close modal

Next, we evaluated the efficacy of K3-SPG for reactivation of Ag-specific memory CTLs in vivo. Cynomolgus macaques were immunized with recombinant protein of HBs Ag plus alum adjuvant, the most commonly used vaccine adjuvant. At 20 wk postvaccination, animals received s.c. injection of K3 or K3-SPG (Fig. 3C). Because poly(I:C) did not induce a robust activation of total memory CD8+ T cells (Fig. 3B), we focused on TLR9 agonists. Then 3 d after the adjuvant injection, the PBMCs were collected and stimulated with the overlapping HBs peptides ex vivo for 6 h, after which the frequencies of all the HBs-specific CTLs, based on cell positivity for IFN-γ, CD137 (also known as 4-1BB), and/or CD107a, were measured by flow cytometry (Supplemental Fig. 3C). As a result, the frequency of total HBs-specific CTLs was significantly increased compared with preinjection following injection of K3-SPG but not K3 (p = 0.0224, Fig. 3D).

Taken together, these results suggest that K3-SPG can reactivate and boost Ag-specific memory CTLs in NHPs by injection of adjuvant alone without the addition of any Ag.

To assess the early effect of systemic immune responses to poly(I:C), K3, and K3-SPG, we injected poly(I:C), K3, or K3-SPG i.v. or s.c. in cynomolgus macaques. S.c. poly(I:C) injection induced a persistent increase in neutrophils until 1 d postinjection, which was significantly higher than with K3 or K3-SPG (Fig. 4). The number of WBCs, including neutrophils and some monocytes, eosinophils, and basophils in the blood, was significantly increased by poly(I:C) but not by K3 or K3-SPG at 8 h postinjection (Fig. 4A). Next, by i.v. injection, we observed a more rapid and massive increase of WBCs, especially neutrophils, by poly(I:C) injection as early as 3 h postinjection, compared with K3 or K3-SPG (Supplemental Fig. 4A). S.c. injection of poly(I:C) reduced the number of lymphocytes, although there was no significant difference compared with with K3 or K3-SPG injection. The kinetics of the elevated frequencies of neutrophils coincided with a decrease in the frequencies of other cellular subsets throughout the time course (Fig. 4B). Taken together, the injections of K3-SPG and K3 increased similar numbers of neutrophils and reduced similar numbers of lymphocytes, which were both significantly lower than with poly(I:C) injection. It is of note that the kinetics of neutrophil expansion differed slightly between the s.c. and i.v. injection route.

FIGURE 4.

Lymphocyte, monocyte, and neutrophil dynamics after injection of poly(I:C), K3 or K3-SPG in cynomolgus macaques. (A) Kinetics of the number of WBCs, neutrophils, monocytes, eosinophils, and basophils, and lymphocytes were assessed by hematologic analysis, respectively. Cynomolgus macaques were injected s.c. with 2 mg of K3, K3-SPG, or poly(I:C) (n = 3). Blood samples were collected preinjection and at each indicated time point after injection. Data show means ± SEM. The p values were calculated using a two-tailed unpaired t test with Welch correction for comparison with poly(I:C). *p < 0.05, **p < 0.01, ***p < 0.001. (B) Temporal change in the ratio of cell populations is shown as pie charts. Radius of each pie represents the number of cells per microliter of blood as indicated, and the area of pie represents the mean ratio of lymphocytes, monocytes, eosinophils, and basophils, and neutrophils (colored in white, black, and orange, respectively) for s.c. injection.

FIGURE 4.

Lymphocyte, monocyte, and neutrophil dynamics after injection of poly(I:C), K3 or K3-SPG in cynomolgus macaques. (A) Kinetics of the number of WBCs, neutrophils, monocytes, eosinophils, and basophils, and lymphocytes were assessed by hematologic analysis, respectively. Cynomolgus macaques were injected s.c. with 2 mg of K3, K3-SPG, or poly(I:C) (n = 3). Blood samples were collected preinjection and at each indicated time point after injection. Data show means ± SEM. The p values were calculated using a two-tailed unpaired t test with Welch correction for comparison with poly(I:C). *p < 0.05, **p < 0.01, ***p < 0.001. (B) Temporal change in the ratio of cell populations is shown as pie charts. Radius of each pie represents the number of cells per microliter of blood as indicated, and the area of pie represents the mean ratio of lymphocytes, monocytes, eosinophils, and basophils, and neutrophils (colored in white, black, and orange, respectively) for s.c. injection.

Close modal

Furthermore, we investigated the activation status of APC subsets after injection of nucleic acid–based adjuvants ex vivo. We examined and compared the activation and expansion of monocytes following injection of the three different nucleic acid-based adjuvants. We found that by focusing specific monocyte-fraction, where monocytes can be separated into three subsets based on their differential expression of CD14 and CD16: CD14+CD16 (classical monocytes), CD14+CD16+ (intermediate monocytes), and CD14dimCD16++ (nonclassical monocytes). Previously, studies reported that the inflammatory status in vivo could be determined by the phenotype and activation of these monocytes, especially focusing on the subset of intermediate monocytes in PBMCs (30, 31). Based on these studies, we assessed the frequencies of monocyte subsets to estimate the inflammatory state after nucleic acid–based adjuvant injections. Following s.c. or i.v. injection of poly(I:C), the frequency of intermediate monocytes but not classical and nonclassical monocytes was significantly increased at 1 d postinjection compared with the preinjection frequency (Fig. 5A, 5B). This transient expansion of intermediate monocytes declined to baseline levels at 7 d postinjection (Fig. 5B). In contrast, injections of K3 or K3-SPG did not affect the frequency of intermediate monocytes at any time point. The s.c. or i.v. injection of poly(I:C) induced a more notable expansion of intermediate monocytes at 1 d postinjection (Fig. 5B). There was no change in the frequency of classical and nonclassical monocytes by s.c. or i.v. injection of nucleic acid–based adjuvants (data not shown). Therefore, we examined the activation status of intermediate monocytes for each nucleic acid–based adjuvant injection at 1 d postinjection, based on the expression of CCR7 and CD83, which are typical activation markers of APCs. We observed a higher expression of CCR7 and CD83 by s.c. or i.v. injection of poly(I:C) compared with K3 or K3-SPG (Fig. 5C, 5D). We also checked the expression of CCR7 and CD83 on pDCs at 1 d postinjection. The expression of these markers on pDCs was significantly higher compared with those at preinjection by s.c. or i.v. injection of K3-SPG (Fig. 5E, 5F). In contrast, injections of K3 or poly(I:C) did not affect the expression of CCR7 and CD83 on pDCs at 1 d postinjection. These data clearly demonstrate that poly(I:C) and K3-SPG target distinct cell types, activate PRRs differentially, and induce a variety of effector functions in vivo, thereby leading to variable, but distinct, therapeutic applications as well as safety profiles.

FIGURE 5.

Distinct activation of intermediate monocytes and pDC in vivo by poly(I:C) and K3-SPG, respectively, in cynomolgus macaques. (A) Representative flow cytometry data showing three distinct subsets within total monocytes by CD14 and CD16 staining [CD14+CD16 (classical monocytes), CD14+CD16+ (intermediate monocytes), and CD14dimCD16++ (nonclassical monocytes)]. (B) Pooled data of the frequencies of intermediate monocytes in HLA-DR+ cells were assessed at the indicated time points after nucleic acid–based adjuvant s.c. or i.v. injection (n = 6). (C) Representative histograms of surface CCR7 and CD83 are shown in intermediate monocytes at day 1 postinjection. (D) Pooled data of the mean fluorescent intensity from flow cytometry analysis of the activation markers on intermediate monocytes at preinjection and day 1 postinjection (n = 6). (E) Representative histograms of surface CCR7 and CD83 are shown in pDCs at day 1 postinjection. (F) Pooled data of the mean fluorescent intensity from flow cytometry analysis of the activation markers on pDCs at preinjection and day 1 postinjection (n = 6). Data show means ± SEM. The p values were calculated using a two-tailed Wilcoxon matched-pairs signed-rank test for comparison with preinjection. *p < 0.05.

FIGURE 5.

Distinct activation of intermediate monocytes and pDC in vivo by poly(I:C) and K3-SPG, respectively, in cynomolgus macaques. (A) Representative flow cytometry data showing three distinct subsets within total monocytes by CD14 and CD16 staining [CD14+CD16 (classical monocytes), CD14+CD16+ (intermediate monocytes), and CD14dimCD16++ (nonclassical monocytes)]. (B) Pooled data of the frequencies of intermediate monocytes in HLA-DR+ cells were assessed at the indicated time points after nucleic acid–based adjuvant s.c. or i.v. injection (n = 6). (C) Representative histograms of surface CCR7 and CD83 are shown in intermediate monocytes at day 1 postinjection. (D) Pooled data of the mean fluorescent intensity from flow cytometry analysis of the activation markers on intermediate monocytes at preinjection and day 1 postinjection (n = 6). (E) Representative histograms of surface CCR7 and CD83 are shown in pDCs at day 1 postinjection. (F) Pooled data of the mean fluorescent intensity from flow cytometry analysis of the activation markers on pDCs at preinjection and day 1 postinjection (n = 6). Data show means ± SEM. The p values were calculated using a two-tailed Wilcoxon matched-pairs signed-rank test for comparison with preinjection. *p < 0.05.

Close modal

Nucleic acid–based adjuvants that target the canonical innate immune receptors, including TLR3, TLR7/8 and TLR9, have potential for development for efficacious vaccines for certain infectious diseases and cancer because of their ability to induce strong Ag-specific CTL responses. However, in the case of systemic administration, there have been safety concerns regarding their reactogenicity, as well as systemic inflammatory responses that perturb their translation into clinical practice except TLR9 agonists (3237). Among the numerous efforts to improve their potency and safety by modifying their formulation and delivery systems, we recently developed a second-generation humanized CpG DNA, namely K3-SPG. K3-SPG is a nanoparticulate CpG ODN wrapped in the nonagonistic Dectin-1 ligand, SPG. K3-SPG should be superior to the current K3 CpG DNA, which is under clinical development (38), in terms of its ability to induce type I IFN production, and its adjuvant and monoimmunotherapeutic activities against cancer (1921). However, discrepancies in the expression of PRRs, including TLR9, between mice and humans or NHPs limit the translation of any findings into clinical use (39). Therefore, this study investigated the in vivo efficacy of the nucleic acid–based adjuvants K3, K3-SPG, and poly(I:C) in NHPs.

We investigated the effect of K3-SPG injection on memory CD8+ T cells. At 1 and 3 d after injection, the frequency of activated memory CD8+ T cells was significantly higher than at baseline. Furthermore, we investigated the effect of K3-SPG following injection of adjuvant alone without any additional Ag to recall or reactivate Ag-specific CTL responses. For this purpose, we selected HBV vaccination as a model because most individuals vaccinated with it generate mainly long-term HBs-specific Ab responses, and HBs-specific memory CTL responses to a lesser extent (40). We hypothesized that if we could boost HBs-specific memory CTL responses by injection with adjuvant alone, this would be better at silencing HBV replication and disease progression during postvaccination breakthrough infections (41). Hence, we vaccinated cynomolgus macaques with HBs Ag together with alum adjuvant according to current HBV vaccine regimens. As expected, the magnitude of HBs-specific CTL responses was small after vaccination. However, after injecting the NHPs with K3-SPG, their HBs-specific memory CD8+ T cells increased significantly over the preinjection levels. This suggests that K3-SPG is a superior monotherapy agent and can boost Ag-specific memory CTLs in NHPs, even without additional Ag. To our knowledge, this is the first report of the injection of nucleic acid–based adjuvant alone (without Ag) inducing in vivo Ag-specific CTL responses in NHPs. Although prolonged CTL responses against HBV might represent a risk factor (42) for inducing liver damage, and in addition to the importance of HBV-specific CTLs in controlling HBV, we suggest that the effect of K3-SPG injection was transient based on our data, as shown in Fig. 1C. It might be possible to avoid the risk of liver damage by injecting K3-SPG alone, and we propose that injection of K3-SPG may represent a novel strategy for eliminating HBV-infected cells in HBV-infected patients.

In terms of the potent ability of K3-SPG to induce CTLs as a vaccine adjuvant and/or an anticancer agent, we hypothesized that the induction of type I IFN and IL-12 production is important in the NHP model based on our previous study (20). Stimulation of PBMCs by K3-SPG in vitro significantly induced the production of IFN-α2 compared with K3 and poly(I:C) and that of IL-12/23 p40 compared with no stimulation. The source of IFN-α and IL-12/23 p40 was mainly pDCs, as confirmed by flow cytometry analysis, although the results may not be conclusive through this data alone due to the limited numbers of pDCs. Furthermore, only the injection of K3-SPG upregulated the expression of CCR7 and CD83 on pDCs compared with preinjection (Fig. 5F). Recently, it was reported that CpG ODN induced the production of IL-12 p40 from the CD123+ population, especially from CD33+CD123+ pre-DC subsets, in human PBMCs in vitro (43). Therefore, because we used CD123 as a marker of monkey pDCs, our pDC populations might contain pre-DC subsets.

In contrast, we observed similar profiles between the in vivo injection of K3-SPG and K3 injection in terms of induction of IFN-α2 in serum (Fig. 1C). However, we think that the production of IFN-α2 in the draining lymph nodes might be a key point to boost Ag-specific memory CTL responses, because most memory CTLs could be located in lymphoid tissues in the memory phase. Because we previously reported that K3-SPG but not K3 can accumulate and activate macrophages and DCs in the draining lymph nodes of mice (19), the different distribution of K3 and K3-SPG in vivo may lead to later CTL responses after in vivo injection. The levels of IL-12/23 p40 in serum were significantly increased by injection of K3-SPG but not K3 compared with injection of poly(I:C) (Fig. 1C). As far as we know, this is the first report to suggest that TLR9 agonist can induce the production of IL-12/23 p40 in vivo in NHPs or humans. Although we could not detect the production of IL-12 p70 in serum (data not shown), the production of IL-12/23 p40 by K3-SPG injection may affect CTL responses. Furthermore, K3-SPG showed higher levels of IFN-γ production compared with K3 stimulation in vitro although we could not detect IFN-γ production in vivo (Fig. 1A). To explore the contribution of IFN-γ in vivo in NHPs, further investigation will be needed.

In these animals, the upregulation of IFN-related gene mRNA expression by injection of K3-SPG was mainly observed in pDCs but also in cDCs and monocytes. This could be because K3-SPG stimulates pDCs first to produce type I IFNs, which are recognized by IFN receptors on monocytes and cDCs and mediate autocrine or paracrine effects to induce IFN-related gene expression. IL-12/23 p40 was also mainly produced by pDCs in vitro and the upregulation of il12a, which is a gene of IL-12 p35, was observed in pDCs after K3-SPG injection in vivo. These results strongly suggest that K3-SPG is more effective at inducing type I IFNs and Th1 cytokines than K3, even in NHPs.

In contrast, we observed that K3-SPG did not induce extensive inflammatory responses compared with poly(I:C) following s.c. or i.v. injection. Poly(I:C) is an adjuvant that evokes robust inflammation, and its administration and derivatives cause high temperatures in NHPs and in human clinical trials, possibly caused by inflammation (33, 34, 44). We observed that the number of neutrophils and the frequency of intermediate monocytes rapidly and transiently increased following injection of poly(I:C), although injection of K3 or K3-SPG did not significantly change the number of cells in the blood (Fig. 4). In a clinical study of poly(I:C) injection, an increase in the absolute number of granulocytes paralleled fever development (45). Although we did not measure body temperature in cynomolgus macaques, our data also suggest that neutrophils may promote fever following the injection of poly(I:C).

It is also clear that the levels of IL-6 and IL-1RΑ in serum were significantly increased after injection of poly(I:C) but not K3 or K3-SPG (Supplemental Fig. 4B). The transcripts of il-6 mRNA were mainly expressed in monocytes, cDCs, and pDCs following injection of poly(I:C) (Fig. 2). Monocytes and cDCs express MDA5, RIG-I, and IPS-1 and can recognize poly(I:C). pDCs express lower levels of these receptors, and the upregulation of il6 mRNA might be indirectly induced by a secondary mechanism. Taken together, we assume that K3-SPG may overcome safety concerns in humans regardless of the injection route.

In conclusion, to our knowledge this study is the first to indicate that a humanized nanoparticulate TLR9 agonist, K3-SPG, might represent a novel and potent immunostimulator and especially an inducer of Ag-specific CTLs without Ag in NHPs. K3-SPG is therefore a potential candidate for human clinical trials of next-generation vaccine adjuvants and as a novel monotherapeutic agent for cancer and infectious disease.

We thank all members of Prof. Ishii’s laboratory, especially Dr. E. Kuroda for helpful discussions and M. Kawatsu for assistance. We thank all members of Prof. Yasutomi’s laboratory.

This work was supported by grants from the Japan Agency for Medical Research and Development.

The online version of this article contains supplemental material.

Abbreviations used in this article:

cDC

conventional DC

CpG ODN

oligodeoxynucleotide containing CpG

DC

dendritic cell

HBs

hepatitis B surface

HBV

hepatitis B virus

NHP

nonhuman primate

pDC

plasmacytoid DC

poly(I:C)

polyinosinic-polycytidylic acid

PRR

pattern-recognition receptor

SPG

schizophyllan.

1
Steinman
,
R. M.
2008
.
Dendritic cells in vivo: a key target for a new vaccine science.
Immunity
29
:
319
324
.
2
Kawai
,
T.
,
S.
Akira
.
2010
.
The role of pattern-recognition receptors in innate immunity: update on Toll-like receptors.
Nat. Immunol.
11
:
373
384
.
3
Pulendran
,
B.
2004
.
Modulating vaccine responses with dendritic cells and Toll-like receptors.
Immunol. Rev.
199
:
227
250
.
4
Vollmer
,
J.
,
A. M.
Krieg
.
2009
.
Immunotherapeutic applications of CpG oligodeoxynucleotide TLR9 agonists.
Adv. Drug Deliv. Rev.
61
:
195
204
.
5
Halperin
,
S. A.
,
B.
Ward
,
C.
Cooper
,
G.
Predy
,
F.
Diaz-Mitoma
,
M.
Dionne
,
J.
Embree
,
A.
McGeer
,
P.
Zickler
,
K. H.
Moltz
, et al
.
2012
.
Comparison of safety and immunogenicity of two doses of investigational hepatitis B virus surface antigen co-administered with an immunostimulatory phosphorothioate oligodeoxyribonucleotide and three doses of a licensed hepatitis B vaccine in healthy adults 18-55 years of age.
Vaccine
30
:
2556
2563
.
6
Sablan
,
B. P.
,
D. J.
Kim
,
N. G.
Barzaga
,
W. C.
Chow
,
M.
Cho
,
S. H.
Ahn
,
S. G.
Hwang
,
J. H.
Lee
,
H.
Namini
,
W. L.
Heyward
.
2012
.
Demonstration of safety and enhanced seroprotection against hepatitis B with investigational HBsAg-1018 ISS vaccine compared to a licensed hepatitis B vaccine.
Vaccine
30
:
2689
2696
.
7
Cooper
,
C. L.
,
H. L.
Davis
,
M. L.
Morris
,
S. M.
Efler
,
M. A.
Adhami
,
A. M.
Krieg
,
D. W.
Cameron
,
J.
Heathcote
.
2004
.
CPG 7909, an immunostimulatory TLR9 agonist oligodeoxynucleotide, as adjuvant to Engerix-B HBV vaccine in healthy adults: a double-blind phase I/II study.
J. Clin. Immunol.
24
:
693
701
.
8
Crompton
,
P. D.
,
M.
Mircetic
,
G.
Weiss
,
A.
Baughman
,
C. Y.
Huang
,
D. J.
Topham
,
J. J.
Treanor
,
I.
Sanz
,
F. E.
Lee
,
A. P.
Durbin
, et al
.
2009
.
The TLR9 ligand CpG promotes the acquisition of Plasmodium falciparum-specific memory B cells in malaria-naive individuals.
J. Immunol.
182
:
3318
3326
.
9
Ellis
,
R. D.
,
G. E.
Mullen
,
M.
Pierce
,
L. B.
Martin
,
K.
Miura
,
M. P.
Fay
,
C. A.
Long
,
D.
Shaffer
,
A.
Saul
,
L. H.
Miller
,
A. P.
Durbin
.
2009
.
A phase 1 study of the blood-stage malaria vaccine candidate AMA1-C1/Alhydrogel with CPG 7909, using two different formulations and dosing intervals.
Vaccine
27
:
4104
4109
.
10
Duncan
,
C. J.
,
S. H.
Sheehy
,
K. J.
Ewer
,
A. D.
Douglas
,
K. A.
Collins
,
F. D.
Halstead
,
S. C.
Elias
,
P. J.
Lillie
,
K.
Rausch
,
J.
Aebig
, et al
.
2011
.
Impact on malaria parasite multiplication rates in infected volunteers of the protein-in-adjuvant vaccine AMA1-C1/Alhydrogel+CPG 7909.
PLoS One
6
:
e22271
.
11
Ellis
,
R. D.
,
Y.
Wu
,
L. B.
Martin
,
D.
Shaffer
,
K.
Miura
,
J.
Aebig
,
A.
Orcutt
,
K.
Rausch
,
D.
Zhu
,
A.
Mogensen
, et al
.
2012
.
Phase 1 study in malaria naïve adults of BSAM2/Alhydrogel®+CPG 7909, a blood stage vaccine against P. falciparum malaria.
PLoS One
7
:
e46094
.
12
Cooper
,
C. L.
,
H. L.
Davis
,
M. L.
Morris
,
S. M.
Efler
,
A. M.
Krieg
,
Y.
Li
,
C.
Laframboise
,
M. J.
Al Adhami
,
Y.
Khaliq
,
I.
Seguin
,
D. W.
Cameron
.
2004
.
Safety and immunogenicity of CPG 7909 injection as an adjuvant to Fluarix influenza vaccine.
Vaccine
22
:
3136
3143
.
13
Speiser
,
D. E.
,
D.
Liénard
,
N.
Rufer
,
V.
Rubio-Godoy
,
D.
Rimoldi
,
F.
Lejeune
,
A. M.
Krieg
,
J. C.
Cerottini
,
P.
Romero
.
2005
.
Rapid and strong human CD8+ T cell responses to vaccination with peptide, IFA, and CpG oligodeoxynucleotide 7909.
J. Clin. Invest.
115
:
739
746
.
14
Baumgaertner
,
P.
,
C.
Jandus
,
J. P.
Rivals
,
L.
Derré
,
T.
Lövgren
,
L.
Baitsch
,
P.
Guillaume
,
I. F.
Luescher
,
G.
Berthod
,
M.
Matter
, et al
.
2012
.
Vaccination-induced functional competence of circulating human tumor-specific CD8 T-cells.
Int. J. Cancer
130
:
2607
2617
.
15
Tarhini
,
A. A.
,
S.
Leng
,
S. J.
Moschos
,
Y.
Yin
,
C.
Sander
,
Y.
Lin
,
W. E.
Gooding
,
J. M.
Kirkwood
.
2012
.
Safety and immunogenicity of vaccination with MART-1 (26-35, 27L), gp100 (209-217, 210M), and tyrosinase (368-376, 370D) in adjuvant with PF-3512676 and GM-CSF in metastatic melanoma.
J. Immunother.
35
:
359
366
.
16
Valmori
,
D.
,
N. E.
Souleimanian
,
V.
Tosello
,
N.
Bhardwaj
,
S.
Adams
,
D.
O’Neill
,
A.
Pavlick
,
J. B.
Escalon
,
C. M.
Cruz
,
A.
Angiulli
, et al
.
2007
.
Vaccination with NY-ESO-1 protein and CpG in Montanide induces integrated antibody/Th1 responses and CD8 T cells through cross-priming.
Proc. Natl. Acad. Sci. USA
104
:
8947
8952
.
17
Karbach
,
J.
,
A.
Neumann
,
A.
Atmaca
,
C.
Wahle
,
K.
Brand
,
L.
von Boehmer
,
A.
Knuth
,
A.
Bender
,
G.
Ritter
,
L. J.
Old
,
E.
Jager
.
2011
.
Efficient in vivo priming by vaccination with recombinant NY-ESO-1 protein and CpG in antigen naive prostate cancer patients.
Clin. Cancer Res.
17
:
861
870
.
18
Manegold
,
C.
,
N.
van Zandwijk
,
A.
Szczesna
,
P.
Zatloukal
,
J. S.
Au
,
M.
Blasinska-Morawiec
,
P.
Serwatowski
,
M.
Krzakowski
,
J.
Jassem
,
E. H.
Tan
, et al
.
2012
.
A phase III randomized study of gemcitabine and cisplatin with or without PF-3512676 (TLR9 agonist) as first-line treatment of advanced non-small-cell lung cancer.
Ann. Oncol.
23
:
72
77
.
19
Kobiyama
,
K.
,
T.
Aoshi
,
H.
Narita
,
E.
Kuroda
,
M.
Hayashi
,
K.
Tetsutani
,
S.
Koyama
,
S.
Mochizuki
,
K.
Sakurai
,
Y.
Katakai
, et al
.
2014
.
Nonagonistic Dectin-1 ligand transforms CpG into a multitask nanoparticulate TLR9 agonist.
Proc. Natl. Acad. Sci. USA
111
:
3086
3091
.
20
Kobiyama
,
K.
,
B.
Temizoz
,
T.
Kanuma
,
K.
Ozasa
,
M.
Momota
,
T.
Yamamoto
,
T.
Aoshi
,
E.
Kuroda
,
K. J.
Ishii
.
2016
.
Species-dependent role of type I IFNs and IL-12 in the CTL response induced by humanized CpG complexed with β-glucan.
Eur. J. Immunol.
46
:
1142
1151
.
21
Kitahata
,
Y.
,
T.
Kanuma
,
M.
Hayashi
,
N.
Kobayashi
,
K.
Ozasa
,
T.
Kusakabe
,
B.
Temizoz
,
E.
Kuroda
,
H.
Yamaue
,
C.
Coban
, et al
.
2016
.
Circulating nano-particulate TLR9 agonist scouts out tumor microenvironment to release immunogenic dead tumor cells.
Oncotarget
7
:
48860
48869
.
22
Yamamoto
,
T.
,
R. M.
Lynch
,
R.
Gautam
,
R.
Matus-Nicodemos
,
S. D.
Schmidt
,
K. L.
Boswell
,
S.
Darko
,
P.
Wong
,
Z.
Sheng
,
C.
Petrovas
, et al
.
2015
.
Quality and quantity of TFH cells are critical for broad antibody development in SHIVAD8 infection.
Sci. Transl. Med.
7
:
298ra120
.
23
Wickham
,
H.
2009
.
ggplot2: Elegant Graphics for Data Analysis.
Springer
,
Berlin
.
24
Yamamoto
,
T.
,
M. J.
Johnson
,
D. A.
Price
,
D. I.
Wolinsky
,
J. R.
Almeida
,
C.
Petrovas
,
M.
Nason
,
W. W.
Yeh
,
L.
Shen
,
M.
Roederer
, et al
.
2012
.
Virus inhibition activity of effector memory CD8(+) T cells determines simian immunodeficiency virus load in vaccinated monkeys after vaccine breakthrough infection.
J. Virol.
86
:
5877
5884
.
25
R Development Core Team. 2008. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Available at: http://www.R-project.org. Accessed: June 21, 2016.
26
Tak
,
P. P.
,
G. S.
Firestein
.
2001
.
NF-kappaB: a key role in inflammatory diseases.
J. Clin. Invest.
107
:
7
11
.
27
Vignali
,
D. A.
,
V. K.
Kuchroo
.
2012
.
IL-12 family cytokines: immunological playmakers.
Nat. Immunol.
13
:
722
728
.
28
Schneider
,
W. M.
,
M. D.
Chevillotte
,
C. M.
Rice
.
2014
.
Interferon-stimulated genes: a complex web of host defenses.
Annu. Rev. Immunol.
32
:
513
545
.
29
Sokol
,
C. L.
,
A. D.
Luster
.
2015
.
The chemokine system in innate immunity.
Cold Spring Harb. Perspect. Biol.
7
:
a016303
.
30
Ingersoll
,
M. A.
,
A. M.
Platt
,
S.
Potteaux
,
G. J.
Randolph
.
2011
.
Monocyte trafficking in acute and chronic inflammation.
Trends Immunol.
32
:
470
477
.
31
Kwissa
,
M.
,
H. I.
Nakaya
,
H.
Oluoch
,
B.
Pulendran
.
2012
.
Distinct TLR adjuvants differentially stimulate systemic and local innate immune responses in nonhuman primates.
Blood
119
:
2044
2055
.
32
Steinhagen
,
F.
,
T.
Kinjo
,
C.
Bode
,
D. M.
Klinman
.
2011
.
TLR-based immune adjuvants.
Vaccine
29
:
3341
3355
.
33
Champney
,
K. J.
,
D. P.
Levine
,
H. B.
Levy
,
A. M.
Lerner
.
1979
.
Modified polyriboinosinic-polyribocytidylic acid complex: sustained interferonemia and its physiological associates in humans.
Infect. Immun.
25
:
831
837
.
34
Levine
,
A. S.
,
M.
Sivulich
,
P. H.
Wiernik
,
H. B.
Levy
.
1979
.
Initial clinical trials in cancer patients of polyriboinosinic-polyribocytidylic acid stabilized with poly-L-lysine, in carboxymethylcellulose [poly(ICLC)], a highly effective interferon inducer.
Cancer Res.
39
:
1645
1650
.
35
Savage
,
P.
,
V.
Horton
,
J.
Moore
,
M.
Owens
,
P.
Witt
,
M. E.
Gore
.
1996
.
A phase I clinical trial of imiquimod, an oral interferon inducer, administered daily.
Br. J. Cancer
74
:
1482
1486
.
36
Goldstein
,
D.
,
P.
Hertzog
,
E.
Tomkinson
,
D.
Couldwell
,
S.
McCarville
,
S.
Parrish
,
P.
Cunningham
,
M.
Newell
,
M.
Owens
,
D. A.
Cooper
.
1998
.
Administration of imiquimod, an interferon inducer, in asymptomatic human immunodeficiency virus-infected persons to determine safety and biologic response modification.
J. Infect. Dis.
178
:
858
861
.
37
Krieg
,
A. M.
,
S. M.
Efler
,
M.
Wittpoth
,
M. J.
Al Adhami
,
H. L.
Davis
.
2004
.
Induction of systemic TH1-like innate immunity in normal volunteers following subcutaneous but not intravenous administration of CPG 7909, a synthetic B-class CpG oligodeoxynucleotide TLR9 agonist.
J. Immunother.
27
:
460
471
.
38
Tougan
,
T.
,
T.
Aoshi
,
C.
Coban
,
Y.
Katakai
,
C.
Kai
,
Y.
Yasutomi
,
K. J.
Ishii
,
T.
Horii
.
2013
.
TLR9 adjuvants enhance immunogenicity and protective efficacy of the SE36/AHG malaria vaccine in nonhuman primate models.
Hum. Vaccin. Immunother
.
9
:
283
290
.
39
Coffman
,
R. L.
,
A.
Sher
,
R. A.
Seder
.
2010
.
Vaccine adjuvants: putting innate immunity to work.
Immunity
33
:
492
503
.
40
Schumann
,
A.
,
M.
Fiedler
,
U.
Dahmen
,
H.
Grosse-Wilde
,
M.
Roggendorf
,
M.
Lindemann
.
2007
.
Cellular and humoral immune response to a third generation hepatitis B vaccine.
J. Viral Hepat.
14
:
592
598
.
41
Bertoletti
,
A.
,
C.
Ferrari
.
2016
.
Adaptive immunity in HBV infection.
J. Hepatol.
64
(
1
Suppl.
):
S71
S83
.
42
Guidotti
,
L. G.
,
F. V.
Chisari
.
2006
.
Immunobiology and pathogenesis of viral hepatitis.
Annu. Rev. Pathol.
1
:
23
61
.
43
See
,
P.
,
C. A.
Dutertre
,
J.
Chen
,
P.
Günther
,
N.
McGovern
,
S. E.
Irac
,
M.
Gunawan
,
M.
Beyer
,
K.
Händler
,
K.
Duan
, et al
.
2017
.
Mapping the human DC lineage through the integration of high-dimensional techniques.
Science
356
:
eaag3009
.
44
Stephen
,
E. L.
,
D. E.
Hilmas
,
J. A.
Mangiafico
,
H. B.
Levy
.
1977
.
Swine influenza virus vaccine: potentiation of antibody responses in rhesus monkeys.
Science
197
:
1289
1290
.
45
Cornell
,
C. J.
 Jr.
,
K. A.
Smith
,
G. G.
Cornwell
III
,
G. P.
Burke
,
O. R.
McIntyre
.
1976
.
Ssytemic effects of intravenous polyriboinosinic-polyribocytidylic acid in man.
J. Natl. Cancer Inst.
57
:
1211
1216
.

K.K. and K.J.I. have a patent pending for K3-SPG. Y.M. is an employee of Nippon Shinyaku Co., Ltd. The other authors have no financial conflicts of interest.

Supplementary data