Toll-Like Receptor Adaptor Molecules Enhance DNA-Raised Adaptive Immune Responses against Influenza and Tumors through Activation of Innate Immunity

Expression plasmids.cDNA fragments encoding immunogens and TLR adaptor molecules were amplified from parental plasmids by PCR and subcloned into the pGA vector. The FLAG-tagged expression plasmids CMV4-MyD88, CMV4-TRIF, CMV4-TIRAP, CMV4-TOLLIP, CMV4-IRAK1, and CMV4-TRAF6 have been described previously (29). LacZ or EGFP cDNA was amplified from pSV-β-Galactosidase (Promega, Madison, WI) or pEGFP-N1 (BD Biosciences, San Jose, CA), respectively. Influenza virus A/PR/8/34 (H1N1) HA (amino acids 18 to 566) was amplified from pJW4303/H1 and introduced into the pFLAG-CMV9 vector (Sigma, St Louis, MO). cDNA encoding E7₄₉ attached to the carboxy terminus of EGFP was cloned into pGA vector (4). Sequences of cloned PCR products were confirmed with an ABI PRISM Genetic Analyzer (PE Applied Biosystems, Foster City, CA).

Cell transfection and reporter gene assay.Transient transfections were conducted with FuGene 6 (Roche Diagnostics, Indianapolis, IN) according to the manufacturer's protocol. HEK293 cells (3 × 10⁴) were cotransfected with each expression plasmid (CMV4, CMV4-MyD88, CMV4-TRIF, CMV4-TIRAP, CMV4-TOLLIP, CMV4-IRAK1, CMV4-TRAF6, pGA-LacZ, pGA-LacZ-MyD88, or pGA-LacZ-TRIF [200 ng]), the reporter plasmid (p5xNF-κB-luc or pGL3 hIFNβ [25 ng]), and the control Renilla luciferase (LUC) plasmid (pTK-RL [25 ng]; Promega, Madison, WI). Cells were incubated for 48 h after transfection, and LUC activity was measured with the Dual Luciferase Reporter Assay System (Promega) according to the manufacturer's protocol. Alternatively, LacZ activity was measured with a β-Gal Reporter Gene Assay kit (Roche Diagnostics). Finally, firefly LUC activity or LacZ activity of individual cell lysates was normalized to Renilla LUC activity.

Immunization schedule.Eight-week-old female BALB/c and C57BL/6 mice were purchased from SLC (Shizuoka, Japan) and housed in an animal facility under specific-pathogen-free conditions. After anesthetizing with a ketamine/xylazine mixture, the mice (eight mice/group) were injected in the quadriceps muscles with a plasmid solution (1 μg/μl saline, 50 μl/muscle) containing either pGA-LacZ, pGA-LacZ-MyD88, pGA-LacZ-TRIF, pGA-GFP, pGA-GFP-MyD88, pGA-GFP-TRIF, CMV9-HA, CMV9-HA-TRIF, pGA-GFP-E7, or pGA-GFP-E7-TRIF. The injection site was electroporated with a field strength of 30 V/cm (constant) and three pulses of 50 ms each, by using CUY21EDIT (NepaGene, Tokyo, Japan). The booster immunization was given 2 or 4 weeks after the primary immunization. The institutional animal care and welfare committee approved all of the animal experiments, and the mice were treated in accordance with NIH animal care guidelines.

ELISA.Serum antibody (Ab) titers were measured by enzyme-linked immunosorbent assay (ELISA), as described previously (30).

RT-PCR.Splenocytes were harvested 2 weeks after the final immunization. The cells were incubated with 1 μg/ml of H-2^d-restricted LacZ class I peptide (TPHPARIGL), E7 peptide (RAHYNIVTF), or NP peptide (ASNENMDAM) for 24 h at 37°C. Alternatively, inguinal lymph node (LN) cells were harvested 24 or 72 h after the final immunization. Total RNA was isolated as described previously (31). Real-time PCR (RT-PCR) was performed with TaqMan probes (Applied Biosystems, Foster City, CA) specific for IFN-γ, IL-12 p40, IL-18, IL-4, IL-6, IFN-β, TNF-α, IP-10, JE/MCP-1, major histocompatibility complex class I (MHC-I) (D1), MHC-II (Ea), CD40, CD80, CD86, or 18S rRNA and an ABI PRISM 7700 sequence detection system (Applied Biosystems). The relative mRNA expression levels of IFN-γ, IL-12 p40, IL-18, IL-4, IL-6, IFN-β, TNF-α, IP-10, JE/MCP-1, MHC-I (D1), MHC-II (Ea), CD40, CD80, and CD86 in the individual samples were normalized to 18S rRNA levels.

CTL assay.Splenocytes were harvested 2 weeks after the final immunization. The cytotoxic T-lymphocyte (CTL) isolates (effector cells) were prepared after incubation with 1 μg/ml of H-2^d-restricted LacZ class I peptide and 20 U/ml of IL-2 (Sigma) for 4 days at 37°C. P815 cells pulsed with 1 μg/ml of H-2^b-restricted LacZ class I peptide (DAPIYTNV [control peptide]) or H-2^d-restricted LacZ class I peptide were used as the target cells. The target cells (1 × 10⁴) were cultured with increasing numbers of the effector cells for 4 h. The amount of lactate dehydrogenase released from target cells was measured by the Cytotox 96 (Promega) system, and the percentage of specific lysis was calculated according to the manufacturer's protocol.

Influenza challenge.Two weeks after the booster immunization with pGA-GFP, CMV9-HA, or CMV9-HA-TRIF, the mice were challenged intranasally with 1 × 10⁴ PFU (4 50% lethal doses [LD₅₀]) or 5 × 10⁴ PFU (20 LD₅₀) of influenza virus A/PR/8/34 (21). The body weight and the mortality of the challenged mice were monitored for the next 10 days.

HI assay.Sera were collected 2 weeks after the booster immunization with pGA-GFP, CMV9-HA, or CMV9-HA-TRIF. Hemagglutinin inhibition (HI) titers were determined by using influenza virus A/PR/8/34, as described previously (4).

Tumor transplantation.Three days before the primary immunization with pGA-GFP, pGA-GFP-E7, or pGA-GFP-E7-TRIF, the mice were administered subcutaneously 2.5 × 10⁴ cells of TC-1, a mouse lung carcinoma expressing E7 antigen. The size of the local tumor mass was monitored for the next 4 weeks.

Statistical analysis.All experiments were repeated at least twice. Statistical significance was evaluated by Student's t test, a one-way analysis of variance using the Bonferroni method, or nonparametric survival analysis using the Kaplan-Meier method, with a P value of <0.05 considered statistically significant.

Activation of NF-κB and IFN-β promoters by overexpression of TLR adaptor molecules.To identify candidate molecules that might serve as adjuvants for DNA vaccines, a variety of TLR adaptor molecules, including MyD88, TRIF, TIR domain-containing adaptor protein/MyD88 adaptor-like protein (TIRAP/Mal), Toll-interacting protein (TOLLIP), IRAK1, and TRAF6, were tested for the ability to activate the NF-κB and IFN-β promoters of HEK293 cells. NF-κB up-regulates the expression of cytokine, chemokine, and costimulatory molecules central to activation of the innate immune system. Type I IFNs, such as IFN-α and IFN-β, play critical roles in the innate immune response. As shown in Fig. 1A, MyD88 was the most potent inducer of NF-κB activation, while TRIF induced the strongest activation of the IFN-β promoter (P < 0.0001). Thus, these two molecules were selected for further evaluation as candidate genetic adjuvants.

• Open in new tab
• Download powerpoint

FIG. 1.

Characterization of TLR adaptor molecules as activators of the NF-κB and IFN-β promoters. (A) HEK293 cells (4 × 10⁵) were cotransfected with 25 ng of pTK-RL and 25 ng of 5× NF-κB-luc or pGLhIFNβ-luc, plus 250 ng of each indicated expression plasmid for TLR adaptor molecules. Firefly LUC activity was assayed 48 h after transfection and normalized to Renilla LUC activity. Values are means ± standard deviations for three independent experiments. The relative LUC activity of cells transfected with 5× NF-κB-luc plus CMV4-MyD88 was significantly higher than that with 5× NF-κB-luc plus any other plasmid (P < 0.0001). The relative LUC activity of cells transfected with pGLhIFNβ-luc plus CMV4-TRIF was significantly higher than that with pGLhIFNβ-luc plus any other plasmid (P < 0.0001). (B) Schematic diagram of a DNA vaccine expressing a single antigen alone (pGA-LacZ, pGA-GFP, CMV9-HA, and pGA-GFP-E7) or both antigen and adjuvant molecules (pGA-LacZ-MyD88, pGA-GFP-MyD88, pGA-LacZ-TRIF, pGA-GFP-TRIF, CMV9-HA-TRIF, and pGA-GFP-E7-TRIF). (C) HEK293 cells (4 × 10⁵) were transfected as described above. Values are means ± standard deviations for three independent experiments. Significances of differences are as follows: c and d, P < 0.01; e and f, P < 0.001; a and b, P < 0.0001.

Activity of dual-promoter plasmids encoding antigen plus MyD88 or TRIF.To optimize analysis of their adjuvant activity, dual-promoter plasmids were constructed to coexpress LacZ (a surrogate antigen) plus MyD88 or TRIF (Fig. 1B) in a single backbone. Our dual-promoter strategy ensures that the TLR adaptor molecules and an antigen are simultaneously present in the same cells. The ability of these combination plasmids to promote antigen expression was examined in HEK293 cells. Transfection with pGA-LacZ-MyD88 resulted in a 7.8-fold increase in NF-κB-dependent LUC activity compared to control plasmid (pGA-LacZ) (Fig. 1C) (P < 0.0001). Transfection with pGA-LacZ-TRIF resulted in an 11.7-fold increase in IFN-β promoter-dependent LUC activity compared with the control plasmid (Fig. 1C) (P < 0.001). The levels of LacZ expression were similar in all groups (Fig. 1C). These data demonstrate that antigen expression is comparable among the groups, but innate cell signaling is activated in cells transfected with the combination plasmids of LacZ and a TLR adaptor molecule.

The MyD88-expressing DNA vaccine enhances humoral immunity.Mice were immunized and boosted 4 weeks later by intramuscular electroporation (imEPT) with the pGA-LacZ, pGA-LacZ-MyD88, or pGA-LacZ-TRIF DNA vaccine. Four weeks after primary immunization, all three groups developed similarly low immunoglobulin G (IgG) anti-LacZ Ab responses (Fig. 2A). By comparison, animals immunized and boosted with pGA-LacZ-MyD88 generated sevenfold-higher serum IgG anti-LacZ titers at week 6 than did recipients of the control pGA-LacZ or pGA-LacZ-TRIF plasmid (Fig. 2A) (P < 0.01). This improved humoral response was primarily due to the 15-fold-higher IgG2a anti-LacZ response elicited by pGA-LacZ-MyD88 (Fig. 2B) (P < 0.05). These results suggest that overexpression of MyD88 improves induction of Th1-dependent antigen-specific humoral immunity to an antigen coexpressed by the same DNA vaccine.

• Open in new tab
• Download powerpoint

FIG. 2.

MyD88 enhances humoral immunity, whereas TRIF enhances cellular immunity. (A and B) Five mice/group were immunized at 0 and 4 weeks with pGA-LacZ, pGA-LacZ-MyD88, or pGA-LacZ-TRIF by imEPT. Blood was drawn at 4 and 6 weeks, and serum anti-LacZ Ab titers were monitored by ELISA. IgG (panel A), IgG1 (panel B, open bars), and IgG2a (panel B, filled bars) titers were examined by ELISA. (C) Splenocytes were prepared from individual mice and stimulated in vitro with 1 μg/ml of class I (H-2^d) peptide (filled bars) or class I (H-2^b) peptide (open bars) for 24 h. mRNA was extracted and reverse transcribed. Aliquots of the reaction were examined for IFN-γ, IL-4, and 18S rRNA levels. The levels of IFN-γ and IL-4 mRNA expression were normalized to the levels of 18S rRNA and described as relative levels of mRNA expression. The graphs present means ± standard deviations. (D) Effector CTL isolates were prepared as described in Materials and Methods. MHC haplotype-matched (H-2K^d) P815 cells were pulsed with 1 μg/ml of class I (H-2^d) peptide (filled symbols) or class I (H-2^b) peptide (empty symbols) for 1 h and used as target cells. The effector and target cells were mixed at effector-to-target ratios of 6 to 50 and cultured for 4 h, and the lactate dehydrogenase activity of the culture supernatants was assayed. The graph shows the mean percent specific lysis ± standard deviation calculated as described in Materials and Methods. • and ○, pGA-LacZ; ▴ and ▵, pGA-LacZ-MyD88; ▪ and □, pGA-LacZ-TRIF. Significances of differences are as follows: a (B), b (B), a (C), and c (C), P < 0.05; a (A), b (A), b (C), c (D), and f (D), P < 0.01; b (D) and e (D), P < 0.001; a (D), d (D), and g (D), P < 0.0001.

Cellular immune responses modulated by the MyD88 and TRIF genetic adjuvants.The antigen-specific cytokine production by bulk splenocytes from mice 2 weeks post-boost was analyzed. Splenocytes were restimulated in vitro with a LacZ-derived H-2^d-restricted MHC-I peptide, and mRNA levels for IFN-γ and IL-4 were quantified by RT-PCR. Splenocytes from the pGA-LacZ-MyD88- and pGA-LacZ-TRIF-treated groups produced 2.4- and 6.6-fold more IFN-γ mRNA than cells from the control group (Fig. 2C) (P, <0.05 and <0.01, respectively). There was no difference between groups with respect to IL-4 mRNA expression (Fig. 2C). Finally, CTL activity was examined by the lysis of P815 (H-2K^d) cells pulsed with a class I LacZ peptide. CTL isolates from pGA-LacZ-TRIF-vaccinated animals showed the strongest lytic activity (Fig. 2E) (P, <0.001 at all effector-to-target ratios), although those from pGA-LacZ-MyD88-vaccinated mice also exceeded the response of controls (Fig. 2D) (P < 0.01). These results indicate that MyD88 and TRIF can act as genetic adjuvants to enhance the induction of a Th1-dominated cellular immune response to a coexpressed antigen encoded by a DNA vaccine.

TRIF genetic adjuvant increases the number of CD11c⁺ and CD40⁺ cells in draining LNs.The immune response induced by DNA vaccines is dependent on transfected antigen-presenting cells (APCs) reaching the draining LNs (1, 5, 20). A single-cell suspension was prepared from inguinal LNs 3 days after boosting, and cell frequencies were determined by surface staining. Compared with resting LNs, the number of CD11c⁺ dendritic cells (DCs) increased significantly within 72 h of imEPT delivery of plasmid (Table 1) (P < 0.01). The pGA-GFP-TRIF plasmids significantly increased the number of CD40⁺ cells present in draining LNs compared with control plasmid, suggesting that these genetic adjuvants might increase the number of mature B lymphocytes, activated T lymphocytes, and CD40⁺ DCs at that site (Table 1) (P < 0.05).

Effect of TRIF genetic adjuvant on cytokine, chemokine, and cell surface molecule expression in draining LNs.Inguinal LNs were removed 24 and 72 h after booster immunization, and changes in mRNA expression for various cytokines, chemokines, and cell surface molecules were quantified by RT-PCR. The levels of IFN-γ and Th1 cell-promoting cytokines, including IL-12 p40 and IL-18, were significantly up-regulated 72 h after immunization with pGA-GFP-TRIF compared to control groups (Fig. 3) (P < 0.05). Th2 cytokine mRNAs, including IL-4 and IL-6, were also up-regulated at 72 h (P < 0.05). Factors regulating APC function (i.e., IFN-β, TNF-α, IP-10, and JE/MCP-1) and elements associated with APC function (i.e., MHC-II, CD40, CD80, and CD86) were consistently increased 72 h after immunization with pGA-GFP-TRIF compared to control plasmid (Fig. 3) (P < 0.05).

• Open in new tab
• Download powerpoint

FIG. 3.

Activation of cells present in draining LNs after imEPT with the MyD88 or TRIF genetic adjuvant. Twenty-four or 72 h after booster immunization with 50 μg of pGA-GFP or pGA-GFP-TRIF by imEPT, draining LNs (inguinal LNs, 10/group) were removed. Sample mRNA was extracted and reverse transcribed, and then the cDNA aliquots were assessed for expression of IFN-γ, IL-12 p40, IL-18, IL-4, IL-6, IFN-β, TNF-α, IP-10, JE/MCP-1, MHC-I, MHC-II, CD40, CD80, and CD86 by RT-PCR. The graph shows means ± standard deviations for samples prepared from 10 individual LNs. Significances of differences are as follows: a, d, i, j, l, and m, P < 0.05; b, c, e, f, g, h, and k, P < 0.01.

Incorporation of TRIF genetic adjuvant into DNA vaccines targeting influenza and tumors.To explore whether TRIF could contribute to improving the protective effect of a DNA vaccine, vectors targeting the influenza HA antigen and the tumor-associated antigen E7 were constructed (Fig. 1B). Mice primed and boosted with each HA-encoding vaccine were challenged intranasally 2 weeks later with 4 to 20 LD₅₀ of live influenza virus A/PR/8/34. Two outcomes of disease were monitored: weight loss and mortality. Animals immunized with the irrelevant pGA-GFP plasmid uniformly demonstrated severe weight loss and subsequently died (Fig. 4A and B). Vaccination with CVM9-HA protected approximately half of the mice from low-dose challenge but <20% of animals from higher-dose challenge. In contrast, the CMV9-HA-TRIF vaccine protected all mice from low-dose challenge and 70% of mice from high-dose influenza virus challenge (P < 0.05). The HI titer was determined as shown in Fig. 4C. CMV9-HA-TRIF and CMV9-HA raised comparable levels of HI activity, suggesting that the TRIF genetic adjuvant raised a stronger innate and/or cellular immune response, which may confer superior protection in combination with the humoral immune response (Fig. 4C). Mice that had been established with subcutaneous tumors were primed and boosted with each E7-expressing vaccine. Tumor masses grew steadily in mice vaccinated with the irrelevant pGA-GFP plasmid (Fig. 5A). pGA-GFP-E7 treatment partially protected against tumor outgrowth, while pGA-GFP-E7-TRIF progressively protected (Fig. 5A). This observation was in accordance with the levels of E7-specific IFN-γ production from splenocytes of immunized mice (Fig. 5B). These results clearly indicate that the TRIF genetic adjuvant can improve the protective effect of DNA vaccines in infectious and neoplastic disease models, both major targets of current vaccine development.

• Open in new tab
• Download powerpoint

FIG. 4.

TRIF increases the protective activity of an influenza HA-encoding DNA vaccine. Ten mice/group were immunized at 0 and 4 weeks with 2 μg of pGA-GFP, CMV9-HA, or CMV9-HA-TRIF by imEPT. Two weeks later, they were challenged with 4 or 20 LD₅₀ of influenza virus A/PR/8/34. (A) Percent change in body weight compared to day zero. (B) Survival rates monitored for the following 12 days. ○, pGA-GFP; ▴, CMV9-HA; •, CMV9-HA-TRIF. *, P < 0.01. (C) HI titers measured with sera obtained 2 weeks after booster immunizations.

• Open in new tab
• Download powerpoint

FIG. 5.

TRIF increases the protective activity of a tumor-associated antigen E7-encoding DNA vaccine. (A) Three days before vaccination, mice were challenged with 2.5 × 10⁴ TC-1 cells. Ten mice/group were immunized at 0 and 2 weeks with 20 μg of pGA-GFP, pGA-GFP-E7, or pGA-GFP-E7-TRIF by imEPT. Tumor outgrowth was monitored for the following 4 weeks. ○, pGA-GFP; ▴, pGA-GFP-E7; •, pGA-GFP-E7-TRIF. *, P < 0.05. (B) Splenocytes were prepared 2 weeks after the booster immunization and stimulated in vitro with 1 μg/ml of E7 peptide (filled bars) or NP peptide (open bars) for 24 h. The relative IFN-γ mRNA expression levels were measured as described in the legend to Fig. 2. *, P < 0.01. The graphs present means ± standard deviations.