Differential polarization of immune responses by genetic cotransfer of chemokines changes the protective immunity of DNA vaccine against pseudorabies virus

Abstract

Chemokines play a key role in eliciting adaptive immune responses by selectively attracting the innate cellular components to the site of antigen presentation. To evaluate the effect of the genetic adjuvant of chemokines on the adaptive immune responses induced by a plasmid DNA vaccine expressing glycorotein B (gB) of the pseudorabies virus (PrV), a PrV DNA vaccine was co-inoculated with plasmid DNA expressing certain chemokines including CCL3 (MIP-1α), CCL4 (MIP-1β), CCL5 (RANTES), CXCL8 (MIP-2), and CXCL10 (IP-10). A co-injection of the CCL3 plasmid DNA induced immunity that was biased to the T helper type 2 (Th2) pattern, as judged by the ratio of immunoglobulin G isotypes and the production of interleukin-4 cytokine generated from stimulated immune T cells. However, CCL5 and CXCL10 induced immune responses of the Th1-type, which rendered the recipients more resistant to a virulent virus infection. CXCL8 also showed enhanced humoral and cell-mediated immunity (mixed-type pattern) providing effective protection against a viral challenge. However, there was no change in the immune responses induced by the PrV DNA vaccine in CCL4 recipients. These results suggest that co-injection of a chemokine, in the form of an adjuvant preparation, causes a rebalancing of the immunity, which subsequently affects the protective efficacy against a virulent virus infection.

Introduction

The pseudorabies virus (PrV) is an alphaherpesvirus that causes a fatal disease in swine known as Aujeszky's disease. This disease is one of the most infectious diseases affecting the swine industry.¹ Recently, vaccination with the plasmid DNA encoding the glycoprotein antigens (gB, gC and gD) of PrV has been one of the most significant advances in PrV vaccinology.^2,3 However, despite the significant progress made, the immunogenicity of PrV DNA vaccines requires considerable improvement because of the high mortality from the disease in vaccinated animals.^1,3 The immunization procedure for a DNA vaccine is often made more effective by delivering immunogens, defined as adjuvants, along with the empirical preparations.^4–6 These preparations enhance the immune responses by improving antigen presentation, basically by inducing the local innate immune responses. This in turn increases the expression of the costimulatory molecules and cytokines that stimulate T-cell growth and differentiation and recall antigen-presenting cells to the site of immunization. Therefore, there has been considerable interest in cytokines and chemokines and the role they appear to play in modulating the adaptive immune responses.^6–10 Both types of molecules, which are induced non-specifically upon an infection, are involved in regulating the inflammatory reaction and the subsequent adaptive T helper type 1 (Th1) or Th2 type of T-cell reaction that occur in the draining lymph nodes. Hence, manipulating the expression of these cytokines and chemokines during exposure to infectious agents or to a vaccine represents a valuable approach for achieving optimal protection.

The chemokines represent a large superfamily of proteins with molecular weights between 8000 and 15 000 dalton that possess a wide variety of biological activities. They can be subdivided into four main subclasses, designated C, CC, CXC and CX3C according to the presence and position of the conserved cysteine motifs.^11–14 The principal targets of chemokines are bone marrow-derived cells, and chemokines play pivotal roles in co-ordinating leucocyte navigation because motility is an essential part of their function.^11–14 However, chemokines are not only simple chemotactic factors. They play important roles in various activities including maintaining homeostasis, angiogenesis/angiostasis, cellular differentiation and activation, lymphocyte homing, and influencing the overall Th1/Th2 balance of the immune responses.^11–14 These properties suggest that exogenous chemokines, which are administered as recombinant proteins or in their genetic form in expression vectors, so-called genetic adjuvants, might be effective in amplifying and shaping the immune responses to vaccines.

A few reports have shown the non-specific adjuvant effect of chemokines but those studies that used chemokine DNA administered systemically with the antigen provide conflicting data.^9,10,15,16 This study investigated whether or not during the immunization of plasmid DNA expressing the gB of the PrV, the genetic cotransfer of certain chemokines, including CCL3 (MIP-1α), CCL4 (MIP-1β), CCL5 (RANTES), CXCL8 (murine MIP-2), and CXCL10 (IP-10), can affect the efficiency of subsequent acquired immune responses. The results show that the cotransfer of the chemokine genetic adjuvants affects the nature of the acquired immune responses against a PrV gB DNA vaccine, resulting in a subsequent change in the protective immunity against a virulent virus infection. Therefore, the use of chemokines in an adjuvant preparation has the potential to rebalance the immune response, thereby improving the success rate of a vaccination programme.

Materials and methods

Animals

Female C57BL/6 (H-2b) mice, 5–6 weeks of age, were purchased from KOATECH (Pyeongtaek, Korea). The mice were maintained at the animal facility of Chonbuk National University under standard conditions according to the Institutional Guidelines. All the experiments were performed according to the guidelines of the committee on the Care of Laboratory Animal Resources of the Commission on Life Science, National Research Council.

Cells and viruses

The wild-type PrV Yangsan (YS) strain, which was generously supplied by the National Veterinary Research and Quarantine Service in Korea, was propagated in a porcine kidney cell line, PK-15, using Dulbecco's modified Eagle's medium supplemented with 2·5% fetal bovine serum, penicillin (100 U/ml), and streptomycin (100 U/ml). The cultures were incubated at 37° in a humidified CO₂ incubator. The virus stocks were concentrated, titrated, and stored in aliquots at −80° until needed.

Plasmid DNA preparation

The plasmid DNA encoding glycoprotein B (gB) of PrV under the cytomegalovirus promoter (pCI-PrVgB) is described in detail elsewhere.¹⁷ The plasmids encoding CCL3, CCL4 and CXCL8 were prepared as described previously.¹⁵ Other chemokine-encoding plasmid DNAs, such as CCL5 and CXCL10, were provided from the CytokineBank of Chonbuk National University (Jeonju, Korea). These chemokine-encoding plasmid DNAs were expressed using the same cytomegalovirus promoter of pCI-neo eukaryotic expression vector (Promega, Madison, WI). The chemokine protein expression of the different plasmids was determined by reverse transcription polymerase chain reaction (RT-PCR) and/or dot blot after in vitro transfection into NIH-3T3 cells. The plasmid DNA for immunization was purified by polyethylene glycol precipitation using the method described elsewhere.¹⁵ The cellular proteins were precipitated with 1 volume of 7·5 m ammonium acetate. The supernatant was then precipitated with isopropanol. After polyethylene glycol precipitation, the plasmids were extracted three times with phenol–chloroform and precipitated with pure ethanol. The quality of DNA was checked by electrophoresis on a 1% agarose gel. The concentration of plasmid DNA was measured by using the GeneQuant RNA/DNA calculator (Biochrom, Cambridge, UK). The amount of endotoxin was determined by the Limulus amoebocyte lysate test (< 0·05 EU/μg). The in vivo effects of endotoxin and CpG motif were always addressed in parallel by the administration of a control vector.

Immunization and sample collection

Groups of 5- to 6-week-old female mice (n = 7) were co-immunized intramuscularly (i.m.) with 100 μg pCI-PrVgB plus 200 μg of the plasmid DNA vaccine encoding the chemokines formulated in phosphate-buffered saline (PBS) (pH 7·2). To address the effect of the chemokine-encoded plasmid DNA backbone (e.g. CpG motif), some mice were co-immunized i.m. with 100 μg pCI-PrVgB plus 200 μg control vector, pCI-neo, in parallel. The co-administration of the gene expression cassettes involved mixing the chosen plasmid DNA before administration. Immunization was performed three times at 7-day intervals via both anterior tibialis muscles. The control mice for PrV DNA vaccine were immunized i.m. with 100 µg of pCI-neo alone. Serum samples from the mice were collected on the 7th day by retro-orbital bleeding after each immunization and were stored at −80° until needed.

Quantitative real-time RT-PCR

Quantitative real-time RT-PCR (qRT-PCR) was performed to determine the expression levels of chemokines, CCL3, CCL4, CCL5, CXCL8 and CXCL10 using the Bio-Rad Laboratories MiniOpticon real-time PCR detection system (Bio-Rad, Hercules, CA), after transfection into human fibroblast cells along with either pCI-neo or pCI-PrVgB. The total RNA was extracted from transfected cells using easy-Blue™ (iNtRON Biotech, Daejeon, Korea), according to the manufacturer's instructions. The contaminating plasmid DNA was removed by treatment with RQ1-RNase-free DNase (Sigma, St Louis, MO). Following reverse-transcription of 500 ng total RNA, the resulting cDNAs were used for real-time PCR amplification. The chemokine primers that were used are listed in Table 1. PCR amplification was performed with DyNAmo™ SYBR® Green 2-Step qRT-PCR kit (Fynnzymes, Espoo, Finland) by using initial denaturation (95°, 15 min) then 45 cycles of denaturation (94°, 15 seconds), annealing (55°, 30 seconds) and extension (72°, 30 seconds), followed by an additional extension cycle (72°, 10 min) for all mRNAs. A standard curve was generated by plotting threshold cycle values against serially diluted plasmid DNA encoding the chemokine. The copy number of the experimental samples was determined by interpolating threshold cycle values into the standard curve. All data were analysed using the MJOpticon Monitor™ version 3·1 analysis software.

Table 1.

Sequences of the primers used for quantitative real-time PCR to determine the expression level of chemokines

Chemokines	Direction	Primers (5′→3′)	Product size (bp)	Reference
CCL3	forward	CCAAGTCTTCTCAGCGCCAT	71	18
	reverse	GAATCTTCCGGCTGTAGGAGAAG
CCL4	forward	TTCTGTGCTCCAGGGTTCTC	276	19
	reverse	GAGGAGGCCTCTCCTGAAGT
CCL5	forward	CCCTCACCATCATCCTCACT	218	19
	reverse	CTTCTTCTCTGGGTTGGCAC
CXCL8	forward	ATCCAGAGCTTGAGTGTGACGC	90	20
	reverse	AAGGCAAACTTTTTGACCGCC
CXCL10	forward	AAGTGCTGCCGTCATTTTCT	140	19
	reverse	CATTCTTTTTCATCGTGGCA

ELISA for PrV-specific antibody, IgG, IgG1, and IgG2a

A standard enzyme-linked immunosorbent assay (ELISA) was used to determine the level of PrV-specific antibodies [total immunoglobulin G (IgG), IgG1, and IgG2a] in the serum samples. Briefly, the ELISA plates were coated overnight at 4° with an optimal dilution (0·5–1·0 μg/well) of the semi-purified PrV antigen for the sample wells and with goat anti-mouse IgG/IgG1/IgG2a for the standard wells (Southern Biotechnology Associate Inc., Birmingham, AL). The viral antigen for the coating was prepared by semi-purification of the viral stock by centrifugation at 50 000 g after treating them with 0·5% Triton X-100.²¹ The plates were washed three times with PBS–Tween-20 (PBST) and blocked with 3% non-fat dried milk. The samples and standard immunoglobulin were serially diluted two-fold, loaded onto the plate then incubated for 2 hr at 37°, and then further incubated with goat anti-mouse IgG/IgG1/IgG2a-conjugated horseradish peroxidase for 1 hr. The colour was developed by adding a suitable substrate [11 mg of 2,2-azinobis-3-ethylbenzothiazoline-6-sulphonic acid (ABTS) in 25 ml 0·1 m citric acid/25 ml 0·1 m sodium phosphate/10 μl hydrogen peroxide]. The concentrations of antibodies were determined using an automated ELISA reader and the SOFTmax Pro4·3 program (Spectra MAX340, Molecular Device, Sunnyvale, CA).

Determination of cytokine following in vitro stimulation

Two weeks after the final immunization, the mice were killed to prepare the splenocytes and popliteal lymph node cells. The erythrocytes were depleted by treating the single-cell suspensions with 1·6 m ammonium chloride containing 170 mm Tris–HCl buffer for 5 min at 37°. These cells were used as the responder cells. The enriched antigen-presenting cell (APC) populations, which were obtained using the method described by Eo et al.,¹⁵ were used as the stimulators. Briefly, the splenocytes from the naive female retired breeders were depleted of erythrocytes, and 3 ml containing 10⁷ cells was layered over 2 ml of a metrizamide gradient (Accurate Chemical and Sci., Westbury, NY; analytical grade, 14·5 g added to 100 ml of PBS, pH 7·2). The cells were then centrifuged at 600 g for 10 min and the cell interface was collected. The enriched APC population was pulsed with UV-inactivated PrV at a multiplicity of infection of 5·0 for 3 hr (before inactivation). The cells were then washed and counted. The responder cells and the PrV-pulsed APCs were added to 200 μl RPMI-1640 medium at responder-to-stimulator ratios of 5 : 1, 2·5 : 1 and 1·25 : 1. The culture supernatants were harvested after 3 days of incubation. A similar number of responder cells was stimulated with 5 μg concanavalin A as a polyclonal positive stimulator for 48 hr.

ELISA was used to determine the cytokine levels in the culture supernatants. Onto each ELISA plate, 100 ng/well of either interleukin-2 (IL-2), IL-4 or interferon-γ (IFN-γ) anti-mouse antibody (Pharmingen, San Diego, CA; clone no. JES6-1A12, 11B11 and R4-6A2, respectively) was dispensed, and incubated overnight at 4°. The plates were washed three times with PBST and blocked with 3% non-fat dried milk for 2 hr at 37°. Culture supernatant and recombinant IL-2, IL-4 and IFN-γ proteins (Pharmingen) were serially diluted two-fold and added to the corresponding plates as standards. The plates were subsequently incubated overnight at 4°. Biotinylated IL-2, IL-4 and IFN-γ antibodies (Pharmingen; clone no. JES6-5H4, BVD6-24G2, and XMG1.2, respectively) were then added, and the plates were further incubated for 2 hr at 37°. They were then washed and incubated with peroxidase-conjugated streptavidin (Pharmingen) for 1 hr; their colour was then developed by adding a substrate (ABTS) solution. The concentrations of cytokines were determined using an automated ELISA reader and SOFTmax Pro4·3 by comparison with two concentrations of standard cytokine protein.

Virus challenge experiment

The co-immunized mice were infected intranasally (i.n.) with a virulent PrV Yangsan strain [lethal dose 50% (LD₅₀) 10] 2 weeks after the final immunization. The challenged mice were examined daily to determine the number of dead animals. The challenged mice generally began to elicit clinical signs from 3 to 4 days postchallenge.

Statistical analysis

Statistical analysis was carried out using the paired Student's t-test and χ² test. Values of the group receiving pCI-PrVgB alone were compared with values of other groups given co-delivery of the plasmid DNAs expressing chemokines. P-values < 0·05 were considered significant.

Results

Determination of the expression level of chemokines by qRT-PCR

Before evaluating the in vivo immunomodulatory effect of plasmid DNAs encoding chemokines, we determined the expression level of each chemokine using qRT-PCR after transfecting chemokine-encoding plasmids into human fibroblast cells along with either control vector pCI-neo or PrV DNA vaccine pCI-PrVgB. Twenty-four and 48 hr after transfection, the total RNA, which had been extracted from transfected cells and treated with RNase-free DNase, was reverse-transcribed. The resulting cDNA was then used for real-time PCR amplification using each gene-specific primer. As shown in Fig. 1, the expression level of each chemokine was comparable in the presence and absence of pCI-PrVgB DNA vaccine. Therefore, this indicates that the incorporation of either control vector or PrV DNA vaccine induces no change in the expression levels of chemokines.

Figure 1.

The comparative expression levels of chemokines determined by quantitative real-time RT-PCR. Chemokine DNAs, CCL3, CCL4, CCL5, CXCL8 and CXCL10 were cotransfected into human fibroblast cells along with with either control vector pCI-neo or PrV DNA vaccine pCI-PrVgB. Twenty-four and 48 hr following transfection, total RNAs were extracted from transfected cells and used for quantitative real-time RT-PCR (qRT-PCR). The copy number of the experimental samples was expressed by interpolating threshold cycle values into the standard curve generated with serially diluted plasmid DNA encoding chemokine.

Modulation of systemic antibody responses

The immunomodulatory function of the chemokine genetic adjuvants was evaluated by co-immunizing the mice i.m. on three occasions with the chemokine-encoding plasmid DNA plus the DNA vaccine expressing gB of PrV. Subsequently, the PrV-specific IgG levels in the sera were determined on the 7th day after each immunization, as shown in Fig. 2. No detectable PrV-specific IgG response was induced by the control vector, pCI-neo. In contrast, the plasmid DNA expressing PrV gB (pCI-PrVgB) showed detectable IgG responses after the first administration (Fig. 2a). The IgG level induced by pCI-PrVgB was boosted with a subsequent injection, as shown in Fig. 2(b,c). When the plasmid DNAs expressing chemokines were co-administered, the levels of PrV-specific IgG induced by pCI-PrVgB were modulated according to the encoded chemokines. A co-injection of CCL3 produced a marginal decrease in the PrV-specific IgG responses in the sera, whereas the co-administration of the other chemokines, including CCL5, CXCL8 and CXCL10, significantly enhanced the PrV-specific IgG levels (Fig. 2b,c). In contrast, there was no change in PrV-specific IgG level in groups co-administered pCI-PrVgB and CCL4.

Figure 2.

Serum PrV-specific IgG levels of the animals co-injected with pCI-PrVgB and the chemokine-expressing plasmid DNA. Groups of C57BL/6 (H-2^b) mice were immunized i.m. on three occasions at 7-day intervals. On the 7th day after each immunization, the PrV-specific IgG levels in the sera were determined by ELISA: (a) first injection; (b) second injection; (c) third injection. The circles on the graph indicate the individual serum IgG level and the height of the boxes shows the average of each group.

When the effect of the chemokine DNA on the distribution of the PrV-specific IgG isotype (IgG2a and IgG1) induced by pCI-PrVgB was determined, a co-injection of the chemokine DNA was found to show a differential pattern of IgG isotype IgG2a and IgG1 production (Fig. 3). A co-injection of CCL3 induced a decrease in the level of the PrV-specific IgG2a isotype but there was no change in the production of the IgG1 isotype. In contrast, a co-injection of CCL5 and CXCL10 produced enhanced levels of the IgG2a isotype, whereas the levels of the PrV-specific IgG1 isotype were reduced in groups co-administered CCL5, CXCL8 and CXCL10 (Fig. 3a,b). When the IgG2a : IgG1 isotype ratio was estimated based on the levels of isotypes produced, the co-injection of CCL3 produced a lower IgG2a : IgG1 ratio (Th2-type inducer). However, a co-injection of CCL5, CXCL8 and CXCL10 enhanced the ratio of the isotypes (Th1-type inducer) (Fig. 3c). CCL4 produced no change in the level of isotype production. This suggests that co-injection of chemokine DNA can modulate the production of PrV-specific IgG induced by a PrV DNA vaccine and can induce a different pattern of immunity depending on the chemokine encoded.

Figure 3.

Distribution of the serum PrV-specific IgG isotype (IgG1 and IgG2a) of the animals co-administered with pCI-PrVgB and the chemokine-expressing plasmid DNA. Groups of C57BL/6 (H-2^b) mice were immunized i.m. on three occasions at 7-day intervals. Seven days after the final immunization, the PrV-specific IgG isotypes (a) IgG2a (b) IgG1, and (c) the ratio of IgG2a : IgG1, in the sera were determined by ELISA. The circles on the graph indicate the individual serum IgG level, and the height of boxes shows the average of each group.

Differential production of Th1- and Th2-type cytokines

The effect of co-administering plasmid DNAs expressing chemokines with pCI-PrVgB on the generation of Th1- and Th2-type cytokines produced by the immune T cells stimulated with the PrV antigen was examined. Groups of mice co-immunized with the chemokine DNA plus pCI-PrVgB were killed 2 weeks after the final immunization, and the immune T cells obtained from the spleen and popliteal lymph nodes were then stimulated with the syngeneic APC pulsed with inactivated PrV antigen. There was no IL-2, IL-4 and IFN-γ production from the immune T cells from the mice administered the control vector pCI-neo, as shown in Fig. 4. However, the DNA vaccine pCI-PrVgB induced the detectable production of both Th1- and Th2-type cytokines. In splenocytes, the co-injection of the chemokine-encoded plasmid DNA elicited a significant change in the production of cytokines caused by the PrV DNA vaccine depending on the encoded chemokine, with the exception of IL-2 production (Fig. 4a–c). The co-administration of the plasmid DNA expressing CCL3 and CXCL8 significantly enhanced the production of the Th2-type cytokine, IL-4, whereas CCL5 and CXCL10 increased the level of the Th1-type cytokine, IFN-γ, which was produced from the immune T cells in response to the PrV antigen. In contrast, there was no change in the production of cytokines generated by the immune T cells from the mice co-administered the CCL4 DNA. Interestingly, the co-administration of CXCL8 also enhanced the production of IFN-γ along with IL-4 production. With regard to the cells obtained from popliteal lymph nodes, which may be a major draining site of antigens injected into the anterior tibialis muscle, similar patterns of Th1- and Th2-type cytokine production were observed after in vitro stimulation (Fig. 4d–f). These results are consistent with the production of PrV-specific IgG isotypes in sera, and suggest that CCL3 induces immunity biased toward the Th2 type while CCL5 and CXCL10 drive the nature of the immunity induced by PrV DNA vaccine to the Th1 pattern. Furthermore, a mixed pattern of immunity was observed in the group co-immunized with CXCL8 DNA.

Figure 4.

The profile of cytokine production (IL-2, IL-4, and IFN-γ) from the splenocytes and popliteal lymph node cells of the co-immunized mice after specific restimulation with the inactivated PrV antigen. Two weeks after the final immunization, as indicated in Materials and methods, the responder cells (a, b and c for splenocytes; d, e and f for popliteal lymph node cells) were mixed with the irradiated syngeneic enriched APCs that had been pulsed with the UV-inactivated PrV and then incubated for 3 days. The levels of cytokines in the supernatants of the stimulated T cells were determined by cytokine ELISA. The test was carried out in quadruplicate wells. The height of each box represents the average and standard deviation of three independent experiments. Statistically significant differences compared with pCI-PrVgB recipients: *P < 0·05; **P < 0·01; ***P < 0·001.

Protective immunity against virus challenge

To determine whether the altered nature of immunity afforded by the co-delivery of the chemokine-expressing plasmid DNA with pCI-PrVgB could affect protection against a virulent virus infection, the groups of mice given both the plasmid DNAs expressing chemokines and pCI-PrVgB were challenged i.n. with the virulent PrV Yangsan strain (10 LD₅₀) 2 weeks after the final immunization. The first examination of the anamnestic serum PrV-specific IgG level 3 days after the challenge revealed a similar pattern to the serum PrV-specific IgG levels induced by the co-injection of both chemokine-encoded plasmid DNA and pCI-PrVgB, as shown in Fig. 5. The groups of mice co-administered with either CCL5, CXCL8 or CXCL10 showed a higher level of anamnestic PrV-specific IgG than the group given the PrV DNA vaccine, whereas the lower level of IgG was observed in the group co-administered CCL3 DNA. The level was similar to the group given pCI-PrVgB when co-administered with the plasmid DNA expressing CCL4.

Figure 5.

Anamnestic PrV-specific IgG responses determined at 3 days postchallenge with the virulent virus strain. Two weeks after the final immunization, groups of mice (n = 7) were challenged i.n. with the PrV Yangsan strain (10 LD₅₀), and the sera were then collected 3 days postchallenge. Conventional ELISA was used to determine the levels of the PrV-specific IgG in the sera. The circles on the graph indicate the individual serum IgG level, and the height of boxes shows the average of each group.

Such anamnestic IgG responses were closely related to protection against a virulent virus challenge, as shown in Fig. 6. All the animals immunized with the control vector pCI-neo were killed by the virulent virus infection, whereas one out of seven animals given the DNA vaccine pCI-PrVgB survived (survival rate = 14·3%) 12 days after the challenge. Such a protection rate of the pCI-PrVgB DNA vaccine was altered by co-injecting the mice with the chemokine-expressing plasmid DNAs (Fig. 6a). The co-injection of CXCL10-encoding plasmid DNA produced the most potent protection against a virulent virus challenge. The co-administration of the plasmid DNA expressing either CCL5 or CXCL8 also produced enhanced protection. However, the co-injection of the other chemokines, including CCL3 and CCL4, did not alter the level of protection against the challenge. Furthermore, a co-injection of plasmid DNA expressing CCL5, CXCL8 or CXCL10 caused significantly prolonged survival after the challenge, when the average survival days were determined (Fig. 6b). Therefore, co-delivery of plasmid DNAs expressing chemokines with the PrV DNA vaccine can alter the protective efficacy against a virulent virus infection. In particular, the co-delivery of plasmid DNA expressing CCL5, CXCL8 and CXCL10 (Th1 type and mixed type inducers) induced enhanced protection, whereas plasmid DNA expressing CCL3 and CCL4 (Th2-type inducer and non-modulator, respectively) offered no added protection against a PrV challenge.

Figure 6.

(a) Susceptibility and (b) average day of death of the animals co-immunized with pCI-PrVgB plus plasmid DNA expressing the chemokines against a virulent virus challenge. Two weeks after the final immunization, groups of mice (n = 7) were challenged i.n. with the PrV Yangsan strain (10 LD₅₀). The challenged mice were examined daily for death until 12 days postchallenge. Statistically significant differences compared with the pCI-PrVgB recipients: *P < 0·05; **P < 0·01.

Discussion

Unlike traditional adjuvants, genetic adjuvants delivered concomitantly in the form of a plasmid DNA vector with a vaccine are unable to affect the very early events of immune responses, which include the maturation and migration of APCs from the site of inflammation to the draining lymph nodes. The co-injection of chemokines in an attempt to attract immature APCs can enhance the probability of facilitating an interaction between the APCs and effector cell (e.g. B or T cell) and hence promote more effective antigen presentation, as well as increasing the autocrine and paracrine chemokine levels of cells.^11–14 Furthermore, the capacity of certain chemokines to selectively recruit specific cell subsets can be exploited to bias the immune responses toward the Th1-type rather than the Th2-type pattern.²² In this study, the co-delivery of chemokine-expressing plasmid DNAs altered the nature of the immunity induced by the PrV gB DNA vaccine depending on the chemokine gene encoded, which resulted in different levels of protection against a virulent virus infection.

All the chemokines tested attract innate cellular components, including immature dendritic cells (DCs), natural killer (NK) cells, macrophages and neutrophils, albeit through overlapping receptors. The immature DCs of the innate cellular components are required to initiate an adaptive immune response by the so-called professional APCs.²³ However, the recruitment of immature DCs was not sufficient to increase the PrV DNA vaccine-induced adaptive immune responses because only CCL4 and CCL5 (not CCL3, CXCL8 and CXCL10) had a beneficial effect on immature DCs. CCL3 binds to CCR1 and CCR5 while CCL4 binds to CCR5, and CCL5 binds to CCR1, CCR3 and CCR5.^11–14 Thus, the augmentation of the immune responses is clearly not linked to the receptor specificity of the chemokines but might reflect the delicate balance between the ligation of different receptors. There is evidence suggesting that different subsets of DCs (roughly divided into myeloid, lymphoid and plasmacytoid DCs), which differentially induce the nature of immunity biased to Th1 and/or Th2, are selectively recruited by chemokines.^24,25 The different DC subsets may also differ in their susceptibility to the DNA-vector-delivered maturation signal depending on the expression level of toll-like receptor 9, and exert distinct effects on the activation or even tolerance of naive T cells. Moreover, because cross-presentation of encoded antigen appears to be one of the major mechanisms of DNA vaccine-induced immune responses,²⁶ it is possible that the co-administration of chemokine-expressing plasmid DNAs with DNA vaccine might change the pathway of cross-presentation, resulting in the exhibition of differential immune responses. Therefore, the exact mechanisms of immunomodulatory function induced by genetic co-delivery of chemokine need to be further studied.

Previous reports suggested that CCL3 had an enhancing effect on antibody responses.^27,28 Even genetic co-delivery of CCL3 was found to mount immune responses of a Th1 type, which produced more resistance to herpesvirus vaginal infection.¹⁵ Unexpectedly, such enhancement of the antibody titres and Th1-type biased immunity were not observed using the CCL3 chemokine in this study. This apparent lack of consistency might be the result of antigen encoded in the DNA vaccine and the injection doses. Alternatively, it is possible that the strategy of co-injecting the chemokine and antigen expression might require some time to reach optimal levels. Therefore, injecting chemokine in a time–course study before injecting the antigen could induce more antibody production than injecting with the antigen alone. However, this study also showed that the nature of the immunity was biased to the Th2-type using the CCL3 plasmid DNA, which is consistent with previous reports.^27,28 Unlike the effect of CCL3, the co-immunization of CCL5 previously showed variable antibody responses.^9,10,16,27 CCR5, which is one of the CCL5 receptors, is strongly expressed on monocytes and DCs.^11–14 If monocytes and DCs migrate in response to CCL5 expressed from the plasmid DNA constructs, the enhanced migration of those immune cells might increase the immune responses against an antigen. In reality, several reports along with these results showed that CCL5 enhanced the antigen-specific Th1 and cytotoxic T lymphocyte responses.^16,27 Based on the need to boost the humoral and cellular responses with regard to the development of a vaccine, the co-injection of CCL5 shows promise in being able to induce an effective immune response against infectious agents.

The CXC chemokine, CXCL8, is the second chemokine that enhanced humoral and cellular immune responses and protective immunity against a virulent virus challenge. CXCL8 also enhanced the production of both IFN-γ and IL-4 generated from the stimulated immune T cells (mixed-type response). The modulatory mechanism of CXCL8 remains to be resolved because this chemokine had no modulatory effect on the APCs that do not express CXCR2, a receptor of CXCL8.^11–14 Instead, the mechanism by which it achieved modulation might involve its effects on the function of NK cells.^11–14 Since NK cells, which were presumably activated in the CXCL8 recipients, act as an important source of IFN-γ secretion,²⁹ the early enhanced IFN-γ secretion in the recipients receiving CXCL8 or after a virus challenge could have been derived from such cells, as described by others.^29,30 In addition, the early IFN-γ produced from the NK cells might play an important role in shaping the subsequent nature of the immunity, which is known to occur in vitro.^31,32 An alternative mechanism by which CXCL8 modulates immunity can include its effects on the neutrophil function. Therefore, CXCR2 is expressed on neutrophils and NK cells.^11–14 Moreover, because neutrophils are known to be involved in the control of a herpesvirus infection,^33,34 an enhanced neutrophil function could account for the effective removal of the virus in the CXCL8 recipients. The role of neutrophils as the mediators of antiviral immunity requires further investigation.

An even more surprising finding was that CXCL10 was the most potent inducer of the Th1-type immune responses and provided effective protection against a virulent virus infection. Many studies have demonstrated the ability of CXCL10 to stimulate the directional migration of activated Th1 cells.^35–37 Neutralizing the in vivo activity of CXCL10 altered the Th1/Th2 balance, either by increasing the selective accumulation of the effector T cells at the inflammatory site and/or following the direct effect of CXCL10 on T-cell polarization.³⁸ A previous report and the present results suggest in vivo polarization of immune T cells by CXCL10.³⁹ However, whether or not CXCL10 can directly drive T-cell polarization toward Th1 remains unclear. Also, no enhanced production of IL-2, one of the Th1-type cytokines, in groups co-administered with CCL5, CXCL8 and CXCL10 (Th1-type inducer) requires further research.

Cell-mediated immunity biased toward the Th1-type might be a major player in the protective immunity against a PrV infection.^21,40 Studies on a murine model have shown that both IFN-γ and Th1-type CD4⁺ T cells are important for protecting against a lethal PrV infection.²¹ However, both antibody- and cell-mediated immunity might play a role in protection. In reality, both enhanced humoral and cellular immunity provided effective protection against a virulent PrV challenge.¹⁷ Furthermore, it is clear that antibodies protect against mortality after a PrV challenge.^41,42 These observations support the evidence showing that both antibody- and cell-mediated immunity might play a role in conferring protection against a PrV infection. Considering the substantial role of antibody- and cell-mediated immunity against PrV challenge, CXCL10 DNA appears to be the beneficial modulator in enhancing both immunity, and subsequently provide effective protection against viral infection. At least CXCL10 can be a useful modulator for PrV DNA vaccination, even though injection of doses of CXCL10 DNA to induce optimal immunity remains to be determined. In conclusion, these results suggest that chemokine can be used to engineer the immune responses to a PrV DNA vaccine by favouring the formation of either cellular or humoral protection or of both, hence, shifting the Th profile of the immune response. This study is important for establishing the impact of co-administering chemokines with different immunogens on the immune system. However, it appears that the optimal induction of a desired response in vivo may require a combination of several factors such as chemokines and other cytokine adjuvants to enhance the presentation of an antigen and the helper function and to ultimately direct the immune responses.^43,44 Further immunization protocols may be modified and ‘tailored’ to elicit an immune response of a desired profile that is considered to be the best for fighting pathogens including PrV.

Abbreviations

ABTS

2,2-azinobis-3-ethylbenzothiazolin-6-sulphonic acid

APC

antigen-presenting cell

DCs

dendritic cells

ELISA

enzyme-linked immunosorbent assay

gB

glycoprotein B

IFN

interferon

IgG

immunoglobulin G

IL

interleukin

i.m.

intramuscularly

i.n.

intranasally

LD₅₀

lethal dose for 50%

NK

natural killer

PBS

phosphate-buffered saline

PrV

pseudorabies virus

qRT-PCR

quantitative real-time reverse transcription-polymerase chain reaction

Th1

T helper type 1