Adjuvant Activities of Novel Cytokines, Interleukin-23 (IL-23) and IL-27, for Induction of Hepatitis C Virus-Specific Cytotoxic T Lymphocytes in HLA-A*0201 Transgenic Mice

Abstract

Searching the sequence databases has revealed two novel cytokines: interleukin-23 (IL-23) and IL-27. These cytokines are quite similar to, but clearly distinct from IL-12 in their structures and T-cell stimulatory fashions. In contrast to IL-12, however, little is known about the roles of IL-23 and IL-27 in the immune regulation. Previously, we evaluated the prime-boost immunization consisting of priming and the first boosting with the hepatitis C virus (HCV)-core expression plasmid, followed by a second boosting with recombinant adenovirus expressing HCV core for induction of HCV core-specific cytotoxic T lymphocytes (CTLs) in BALB/c mice. The present study demonstrates that HCV-specific CTL induction was greatly enhanced by coinoculation of an IL-12 expression plasmid in the prime-boost immunization, indicating the potent adjuvant activity of IL-12. We investigated whether similar adjuvant effects could be exerted by either IL-23 or IL-27 in a prime-boost immunization with HLA-A*0201 transgenic mice. Coadministration of either an IL-23 or an IL-27 expression plasmid, as well as an IL-12 expression plasmid, in a prime-boost immunization enhanced induction of HCV-specific CTLs and led to dramatic increases in the numbers of gamma interferon (IFN-γ)-producing, HCV-specific CD8⁺ cells. Further, preinjections of IL-12, IL-23, or IL-27 expression plasmids before immunization resulted in great increases in the number of IFN-γ-producing, HCV-specific CD8⁺ cells in response to immunization with recombinant adenovirus. These data revealed that both IL-23 and IL-27, as well as IL-12, are potent adjuvants for epitope-specific CTL induction. The two novel cytokines might offer new prophylactic and therapeutic strategies against infectious pathogens such as HCV.

CD8⁺ cytotoxic T lymphocytes (CTLs) recognize virus-derived peptides in association with major histocompatibility complex class I molecules on the surface of antigen-presenting cells and kill the virus-infected target cells. There are a number of evidences showing that CTLs play a central role in the clearance of pathogenic viruses (67). In case of hepatitis C virus (HCV) infection, vigorous HCV-specific CTL responses existed in the persons resolving acute HCV infection (27). In the experimental model, chimpanzees who cleared HCV generated strong CTL but poor antibody responses, whereas other chimpanzees developing chronic hepatitis generated much weaker CTL response (12). Thus, spontaneous resolution of HCV is likely to be associated with HCV-specific CTLs rather than neutralizing antibodies (12, 18, 27, 58). However, >60% of HCV-infected individuals turn out to have chronic hepatitis (1). Chronic HCV hepatitis eventually progresses to cirrhosis and hepatocellular carcinoma (53). In the chronic stage, HCV-specific CTLs are detectable in both peripheral blood and liver, but the precursor frequency of HCV-specific CTLs is extremely low (27, 50, 51). Therefore, enhancement of HCV-specific CTL induction in HCV-infected individuals should be considered to be a strategy to clear the virus.

DNA vaccination has been proven to be a useful strategy for inducing both humoral and cellular immune responses (19). DNA vaccine safely mimics the effect of live, attenuated virus-based vaccine to generate a long-lasting CTL response. However, the efficiency of DNA vaccine is sometimes quite low and, therefore, several modifications have been attempted in recent years (19). Thus far, the most successful protocol of DNA immunization for CTL induction is likely to be a consecutive immunization involving priming with plasmid DNA and boosting with recombinant virus (2, 24, 35, 43, 54). The rationale behind this strategy is that DNA priming elicits low-level but persistent immunity, followed by strong boosting with virus encoding the same recombinant antigen as the DNA encodes. This regimen of the consecutive immunization has been proven to be efficient for CTL induction by many groups (2, 24, 35, 37, 43, 54). Recently, McConkey et al. (37) have shown that the prime-boost immunization induced high frequencies of antigen-specific T-cell responses to malaria antigen and displayed partial protection in humans.

Interleukin-12 (IL-12) is a heterodimeric proinflammatory cytokine formed by a 35-kDa light chain (p35) and a 40-kDa heavy chain (p40) (57). This cytokine is a dominant factor in the differentiation of T helper type 1 (Th1) cells and plays an essential role in a link between innate and adaptive immunities (60). Recently, two novel polypeptides, p19 (42) and p28 (48), have been identified by searching the databases with a computationally derived profile of IL-6. These factors do not show any biological activity by themselves. However, p19 associates with a p40 subunit of IL-12 to form a biologically active, new heterodimeric cytokine, termed IL-23 (42). The IL-23 receptor (IL-23R) is composed of the IL-12Rβ1 subunit and a novel IL-23R subunit (44). On the other hand, p28 combines with Epstein-Barr virus-induced gene 3 (EBI3), which is an IL-12 p40-related protein, and shapes a new cytokine named IL-27 (48). It has recently been found that the orphan cytokine receptor WSX-1/TCCR, which is homologous to the IL-12Rβ2 subunit (48), and gp130 (47) constitute the IL-27R. In addition to the structural similarities, all three cytokines definitely favor the differentiation of Th1 cells. However, these three cytokines act at different stages during the Th1 development (52). Thus, IL-23 and IL-27 are similar to, but clearly distinct from IL-12. Although the effects of IL-12 in the regulation of immune response are well established (60), little is known about the roles of IL-23 and IL-27.

We previously evaluated the prime-boost immunization consisting of an HCV core-expression plasmid and a recombinant adenovirus expressing HCV core for the induction of HCV core-specific CTLs in BALB/c mice (35). An IL-12 expression plasmid has been shown to be an effective adjuvant for CTL induction in the conventional DNA vaccine by several groups (25, 41, 49, 56, 61). We showed that coadministration of an IL-12 expression plasmid in a prime-boost immunization could enhance HCV-specific CTL induction (35). In the present study, we tested whether adjuvant effects could be exerted by either an IL-23 or an IL-27 expression plasmid, as well as an IL-12 expression plasmid, in a prime-boost immunization using human histocompatibility leukocyte antigen HLA-A*0201 transgenic mice. To circumvent uneven gene expression of the two subunits (32), genetically linked, single-chain IL-12 (scIL-12), scIL-23, and scIL-27 were used in the experiments.

MATERIALS AND METHODS

Mice.

The HHD mice express a transgenic HLA-A*0201 monochain, designated HHD, in which human β₂-microglobulin is covalently linked to a chimeric heavy chain composed of HLA-A*0201 (α1 and α2 domains) and H-2D^b (α3, transmembrane, and cytoplasmic domains) (46). Because the innate H-2D^b and mouse β₂-microglobulin genes have been disrupted by homologous recombination, the only major histocompatibility complex class I molecule on the cell surface, HLA-A*0201, is efficiently utilized by HLA-A*0201-restricted CTLs in HHD mice. Six- to eight-week-old HHD mice were used for all experiments. Mice were housed in appropriate animal care facilities at Saitama Medical School, Saitama, Japan, and handled according to international guidelines for experiments with animals.

Cell lines.

The HHD gene-transfected mouse lymphoma cell line, RMA-HHD (H-2^b), was previously described (46). The cell surface expression of the HHD molecule was confirmed by flow cytometry using the anti-HLA-A2 monoclonal antibody BB7.2 (45) (data not shown). The adenovirus type 5 DNA-transformed, human kidney cell lines, 293 (16), and its derivative, 293T, were obtained from the American Type Culture Collection (Rockville, Md.). RMA-HHD cells were maintained in RPMI 1640 supplemented with 10% fetal calf serum (FCS) and 500 μg of G418 (Sigma, St. Louis, Mo.)/ml. 293 and 293T cells were cultured in Dulbecco modified Eagle medium with 10% FCS.

Synthetic peptides.

Two HCV core-derived, three HCV NS3-derived, and two HCV NS4-derived, HLA-A*0201-restricted peptides were synthesized by Qiagen (Tokyo, Japan) by using Fmoc (9-fluorenylmethoxy carbonyl) chemistry (Table 1). These peptides were desalted and then analyzed by high-performance liquid chromatography.

TABLE 1.

HLA-A*0201-associated, HCV-derived CTL epitopes used in this study

Name	Region	Residue	Sequence	Reference
core-35	HCV-core	35-44	YLLPRRGPRL	4
core-132	HCV-core	132-140	DLMGYIPLV	4
NS3-1073	HCV-NS3	1073-1081	CINGVCWTV	9
NS3-1406	HCV-NS3	1406-1415	KLVALGINAV	9
NS3-1585	HCV-NS3	1585-1593	YLVAYQATV	65
NS4-1666	HCV-NS4A	1666-1674	VLVGGVLAA	55
NS4-1769	HCV-NS4B	1769-1777	HMWNFISGI	55

Plasmid constructs.

The cDNAs encoding the p40 and p35 chains of mouse IL-12 and the EBI3 and p28 chains of mouse IL-27 were generated by reverse transcriptase PCR from total RNA of concanavalin A-activated mouse spleen cells. DNA constructs for scIL-12 and scIL-27 were then prepared as described previously (5). In brief, the p40 and p35 subunits of IL-12, or the EBI3 and p28 subunits of IL-27 were genetically fused with a DNA linker encoding 15 amino acids, (Gly₄Ser)₃ (5), by PCR amplification (scIL-12 sense, 5′-AAGCTTATGTGGGAGCTGGAGAAA-3′; scIL-12 antisense, 5′-CTCGAGTCAGGCGGAGCTCAGATA-3′; scIL-27 sense, 5′-AAGCTTCCTGGTTACACTGAAACA-3′; and scIL-27 antisense, 5′-AGATCTTAGGAATCCCAGGCTGA-3′). The linker sequences were synthesized as sense and antisense oligonucleotides spanning the sequence as reported by Huston et al. (22). The PCR products were subcloned into the pCR-Blunt II-TOPO vector (Invitrogen, Carlsbad, Calif.). After digestion with HindIII and EcoRI for IL-12 or HindIII and BglII for IL-27, the PCR products were cloned into the p3XFLAG-CMV-9 (p3XFLAG) (Sigma) expression vector (p3XFLAG-IL-12 and p3XFLAG-IL-27). This vector encodes the preprotrypsin signal peptide and the FLAG epitope tag sequence (DYKDHDGDYKDHDIDYKDDDDK) upstream of the multiple cloning region and, hence, expresses a secreted N-terminal 3× FLAG fusion protein in mammalian cells. The cDNA encoding the scIL-23 composed of the p40 chain, a (Gly₄Ser)₃ linker, and the p19 chain, was amplified from the scIL-23-immunoglobulin (Ig) fusion protein expression plasmid (5). The following primers were used: sense, 5′-AAGCTTATGTGGGAGCTGGAGAAA-3′; and antisense, 5′-CTCGAGTCAAGCTGTTGGCACTAA-3′. The PCR product was subcloned into the pCR-Blunt II-TOPO vector (Invitrogen). After digestion with HindIII and EcoRI, the scIL-23 cDNA was cloned into p3XFLAG (p3XFLAG-IL-23) as well. The nucleotide sequences of the inserts were confirmed by DNA sequencing.

The plasmids expressing HCV-NS3 and -NS4 proteins were generated as described before (35, 36). Briefly, the two HCV-derived genes corresponding to amino acid residues 1018 to 1657, including the NS3 region, and amino acid residues 1607 to 1972 including the NS4 region, were amplified by PCR from the HCV cDNA clone, pBRTM/HCV1-3011con (26), which was kindly provided by C. M. Rice (Rockefeller University, New York, N.Y.). The primers used here were as follows: NS3 sense, 5′-GCCGGATCCATGGTCTCCAAGGGGTGGAG-3′; antisense, 5′-TCACGTGACGACCTCCAGGTCGGCC-3′; NS4 sense, 5′-GTGGGGATCCATGTGGAAGTGTTTGATCC-3′; and antisense, 5′-TCAGCATGGAGTGGTACACTCCGAGC-3′. The PCR-amplified genes were subcloned into the pCR2.1 vector in the TA cloning kit (Invitrogen), and the nucleotide sequences were verified to be identical to those in pBRTM/HCV1-3011con by DNA sequencing. After digestion with HindIII and NotI for NS3 or with BamHI for NS4, each of the genes was cloned into the multiple cloning site in the eukaryotic expression vector, pCEP4 (Invitrogen; pCEP4-NS3 and pCEP4-NS4, respectively). pCEP4 contains the Epstein-Barr virus origin of replication and the Epstein-Barr nuclear antigen, permitting high-copy extrachromosomal replication, hygromycin B gene as the selectable marker, and cytomegalovirus promoter for gene expression. The pCEP4-core plasmid expressing the HCV core protein was described previously (35).

Purification of plasmid DNA for immunization was performed by ultracentrifugation to equilibrium in cesium chloride-ethidium bromide gradients.

Detection of scIL-12, scIL-23, and scIL-27 fusion proteins.

Each of p3XFLAG, p3XFLAG-IL-12, p3XFLAG-IL-23, and p3XFLAG-IL-27 plasmids was transiently transfected into 293T cells by using FuGENE 6 (Roche Diagnostics Co., Indianapolis, Ind.). Briefly, in an Eppendorf tube, 1.5 μl of FuGENE 6 was diluted in 100 μl of serum-free medium, and then 0.5 μg of each DNA solution was added into the medium containing FuGENE 6. The mixture was incubated for 15 min at room temperature and then added to 4 × 10⁵ 293T cells in a well of a 12-well plate. After 2 days, the supernatant was collected.

Fusion proteins in the supernatants were immunoprecipitated by using the anti-FLAG M2 monoclonal antibody (Sigma). In brief, 2 μg of the M2 antibody was mixed with 20 μl of 50% protein G-Sepharose solution (Pharmacia, Piscataway, N.J.) for 1 h at 4°C and then added into 100 μl of each supernatant. After a 2- h incubation at 4°C, the immunoprecipitates were washed twice with 1 ml of phosphate-buffered saline (PBS). Immunoprecipitated fusion proteins were then released from the protein G by boiling the samples for 3 min in 50 μl of digest buffer composed of 0.01 M Na₂HPO₄, 0.03% sodium dodecyl sulfate (SDS), and 0.1% Triton X-100. After immunoprecipitation, Western blotting of immunoprecipitated proteins was performed. Briefly, the released fusion proteins were separated by electrophoresis on a SDS-12% polyacrylamide gel electrophoresis under reducing conditions and blotted onto a nitrocellulose membrane. The membrane was blocked with PBS containing 5% nonfat milk. The blot was stained with 5 μg of the M2 antibody (Sigma)/ml for 1 h at room temperature, followed by secondary staining with peroxidase-conjugated anti-mouse IgG antibody. The protein bands were developed by the ECL system (Amersham, Piscataway, N.J.).

STAT tyrosine phosphorylation assay.

Mouse IL-12Rβ1 and IL-12Rβ2 cDNAs were previously cloned into an expression vector, pME18S (pME18S-IL-12Rβ1 and pME18S-IL-12Rβ2) (69). Mouse IL-23R, WSX-1/TCCR, and gp130 cDNAs were isolated by reverse transcription-PCR from total RNA of concanavalin A-activated mouse spleen cells and cloned into the expression vector, p3XFLAG-CMV-14 (Sigma) (p3XFLAG-IL-23R, p3XFLAG-WSX-1/TCCR, and p3XFLAG-gp130). For detection of tyrosine phosphorylation (pY) of the signal transducer and activator of transcription 4 (STAT4) triggered by either IL-12 or IL-23, 293T cells were transiently transfected with either pME18S-IL-12Rβ1 plus pME18S-IL-12Rβ2 or pME18S-IL-12Rβ1 plus p3XFLAG-IL-23R, together with the STAT4 expression vector as described by Visconti et al. (63) by using FuGENE 6 (Roche Diagnostics). For detection of pY-STAT1 induced by IL-27, 293T cells were transiently transfected with p3XFLAG-WSX-1/TCCR, and p3XFLAG-gp130 by using FuGENE 6 (Roche Diagnostics). After 36 h, the transfectants were stimulated for 45 min with the culture supernatant of 293T cells transiently transfected with either p3XFLAG-IL-12, p3XFLAG-IL-23, or p3XFLAG-IL-27 at various concentrations of 0.5, 5.0, and 50%. The cells were then subjected to Western blotting with anti-STAT1, anti-STAT4 (Santa Cruz Biotechnology, Santa Cruz, Calif.), anti-pY-STAT1 (Cell Signaling Technology, Inc., Beverly, Mass.), or anti-pY-STAT4 (Zymed Laboratories, Inc., South San Francisco, Calif.) antibody.

Adenoviruses.

The replication-defective recombinant adenoviruses expressing the HCV structural proteins (core, E1, and E2; Adex1SR3ST) and the HCV nonstructural proteins (NS3, NS4, and NS5A; Adex1CA3269) were previously described (33, 62). The wild-type adenovirus lacking the insert (Adex1w) was used as a negative control. Virus was amplified in 293 cells and titers were determined by standard plaque assays on 293 cells. All three adenoviruses were kindly provided by T. Miyamura (National Institute of Health, Tokyo, Japan).

Immunization.

Mice received the prime-boost immunization as described previously (35). Briefly, mice were intramuscularly immunized twice via the tibialis muscles with 100 μg of either pCEP4, pCEP4-core, pCEP4-NS3, or pCEP4-NS4, together with either various amounts (10, 30, and 100 μg) of p3XFLAG, p3XFLAG-IL-12, p3XFLAG-IL-23, p3XFLAG-IL-27, or the mixture of p3XFLAG-IL-12, -IL-23, and -IL-27 dissolved in PBS at 1 mg/ml and then immunized intraperitoneally with 5 × 10⁷ PFU recombinant adenovirus diluted in PBS at 10⁸ PFU/ml (Fig. 1). The immunization interval was 2 weeks.

FIG. 1.

Schedule of the prime-boost immunization. HHD mice received the consecutive immunization consisting of priming and the first boosting with DNA followed by the second boosting with adenovirus. DNA*, 100 μg of either pCEP4-core, pCEP4-NS3, or pCEP4-NS4 together with 10 to 100 μg of either p3XFLAG-IL-12, p3XFLAG-IL-23, or p3XFLAG-IL-27; Virus**, 5 × 10⁷ PFU of either Adex1SR3ST or Adex1CA3269.

CTL assay.

After 2 to 3 weeks after immunization, mice were sacrificed and spleen cells were prepared. Spleen cells were then cultured for 6 to 7 days with equal numbers of irradiated (30 Gy), syngeneic naive spleen cells prepulsed with a 10 μM concentration of an appropriate peptide and used as effector cells. Cytotoxic activities were measured in standard ⁵¹Cr release assays (34). In brief, 10⁶ RMA-HHD cells were pulsed with a 10 μM concentration of an appropriate peptide for 1 h,and then labeled with 100 μCi of Na₂⁵¹CrO₄ for 30 min. After being washed, the labeled target cells were plated in wells of a round-bottom 96-well plate at 5 × 10³ cells/well with or without effector cells at various effector/target (E:T) ratios. After a 4-h incubation at 37°C, supernatant from each well was harvested, and the radioactivity was counted. The results were calculated as the mean of a triplicate assay. The percent specific lysis was calculated according to the following formula: % specific lysis = [(cpm_sample − cpm_spontaneous)/(cpm_maximum − cpm_spontaneous)] × 100, where cpm stands for counts per minute. Spontaneous release represents the radioactivity released by target cells in the absence of effectors, and maximum release represents the radioactivity released by target cells lysed with 5% Triton X-100. At least three mice per group were used in each experiment. Each experiment was repeated at least three times.

Intracellular IFN-γ staining.

Three to five mice per group were immunized as described above. Spleen cells of immunized mice per group were pooled, and were resuspended in RPMI 1640 with 10% FCS. In each well of a 96-well round-bottom plate, 2 × 10⁶ cells were incubated with 1 μl of brefeldin A (GolgiPlug, BD Pharmingen, San Diego, Calif.)/ml for 5 h at 37°C in the presence or absence of an antigenic peptide at a final concentration of 10 or 50 μM. The cells were then washed once with ice-cold PBS containing 1% FCS and 15 mM sodium azide (fluorescence-activated cell-sorting buffer) and incubated for 10 min at 4°C with the rat anti-mouse CD16/CD32 monoclonal antibody (Fc Block; BD Pharmingen) at a concentration of 1 μg/well to block FcR. After incubation, the cell surface was stained with fluorescein isothiocyanate (FITC)-conjugated rat anti-mouse CD8α monoclonal antibody (clone 53-6.7; BD Pharmingen) at a concentration of 0.5 μg/well for 30 min at 4°C. For detection of effector and memory CD8⁺ T cells (21, 66), the cell surface was stained with CyChrome-conjugated anti-mouse CD8α monoclonal antibody (clone 53-6.7), together with either FITC-conjugated rat anti-mouse CD11a antibody (clone 2D7; BD Pharmingen) or FITC-conjugated rat anti-mouse CD62L antibody (clone MEL-14; BD Pharmingen) at a concentration of 0.5 μg/well. After two washes with fluorescence-activated cell-sorting buffer, the cells were fixed and permeabilized by using the Cytofix/Cytoperm kit (BD Pharmingen) and stained with phycoerythrin-conjugated rat anti-mouse gamma interferon (IFN-γ) monoclonal antibody (clone XMG1.2; BD Pharmingen) to detect intracellular IFN-γ according to the manufacturer's instructions. Isotype-matched control antibodies were added to control wells to confirm the specificity of the staining. After removal of unbound antibody by a wash with the 1× Perm/Wash solution in the kit, flow cytometric analyses were performed. Each experiment was repeated at least three times.

ELISPOT assay for IFN-γ-secreting cells.

Detection of IFN-γ secreting cells was performed by using a mouse IFN-γ enzyme-linked immunospot (ELISPOT) set (BD Pharmingen). Each well of a sterile 96-well ImmunoSpot ELISPOT plate (BD Pharmingen) was precoated with 0.5 μg of unlabeled anti-IFN-γ capture antibody (clone R4-6A2; BD Pharmingen)/well at 4°C overnight. Plates were washed and then blocked with RPMI 1640 containing 10% FCS for 2 h at room temperature. Spleen cells from individual, immunized mice were prepared, and red blood cells were removed by treatment with ammonium chloride. After the blocking solution was discarded, effector spleen cells were added into wells at various cell densities (10⁴ to 10⁶ cells/well), along with 10⁶ gamma-irradiated (30 Gy) syngeneic spleen cells. Cells were incubated with 100 U of human recombinant IL-2 (Cetus Corp., Emeryville, Calif.)/ml in the presence or absence of an appropriate peptide at a final concentration of 10 μM at 37°C for 2 days. Cells were then removed by five washes with PBS containing 0.05% Tween 20, followed by the addition of biotinylated anti-mouse IFN-γ detection antibody (clone SMG1.2; BD Pharmingen) at a concentration of 0.5 μg/well. After a 2-h incubation at room temperature, the detection antibody was removed by three washes with PBS containing 0.05% Tween 20. Avidin-horseradish peroxidase was added into the wells. After the unbound avidin-HRP was washed, spots were developed by using freshly prepared substrate buffer (0.3 mg of 3-amino-9-ethyl-carbazole/ml and 0.015% H₂O₂ in 0.1 M sodium acetate).

Statistical analyses.

Statistical analyses were performed with the Student t test. A P value of <0.05 was considered statistically significant.

RESULTS

Expression and activity of the scIL-12, scIL-23, and scIL-27 fusion proteins. Mouse scIL-12, scIL-23, and scIL-27 genes were constructed and cloned into the mammalian expression vector. Each of the plasmids, p3XFLAG-IL-12, -IL-23, and -IL-27, and the empty vector, p3XFLAG, was transiently transfected into 293T cells. The proteins secreted into the culture supernatants were immunoprecipitated with the anti-FLAG monoclonal antibody, M2 (Sigma), since the p3XFLAG vector produces a secreted N-terminal 3XFLAG fusion protein in mammalian cells. The immunoprecipitated fusion proteins were then subjected to Western blot analysis with the M2 antibody. As shown in Fig. 2A, the scIL-12, scIL-23, and scIL-27 fusion proteins were clearly detected in the lanes as single polypeptide bands of 76, 74, and 66 kDa, respectively. In order to examine whether these cytokine fusion proteins are functional, STAT tyrosine phosphorylation assays were carried out (Fig. 2B to D). As shown in Fig. 2C, STAT4 in 293T cells expressing IL-12Rβ1 and IL-23R was phosphorylated on tyrosine in response to scIL-23 but not by either scIL-12 or scIL-27. STAT1 in cells expressing WSX-1/TCCR and gp130 was phosphorylated on tyrosine by the addition of scIL-27 but not by either scIL-12 or scIL-23 (Fig. 2D). Further, the scIL-12 fusion protein induced STAT4 tyrosine phosphorylation in cells expressing IL-12Rβ1 and IL-12Rβ2 (Fig. 2B). Taken together, these data indicate that the scIL-12, scIL-23, and scIL-27 fusion proteins are functional.

FIG. 2.

Expression and activity of the scIL-12, scIL-23, and scIL-27 fusion proteins. (A) p3XFLAG-IL-12 (IL-12), p3XFLAG-IL-23 (IL-23), p3XFLAG-IL-27 (IL-27), and p3XFLAG (vector) were transiently transfected into 293T cells. Proteins secreted in the supernatants were immunoprecipitated with the anti-FLAG antibody. The immunoprecipitated proteins were separated on by SDS-12% polyacrylamide gel electrophoresis and subjected to Western blotting analysis with the anti-FLAG antibody. The positions of protein molecular mass markers (in kilodaltons) are shown in the figure, and the three arrows indicate the bands of the scIL-12, scIL-23, and scIL-27 fusion proteins. (B to D) STAT tyrosine phosphorylation assay. 293T cells were transiently transfected with either pME18S-IL-12Rβ1 plus pME18S-IL-12Rβ1 (B), pME18S-IL-12Rβ1 plus p3XFLAG-IL-23R (C), or p3XFLAG-WSX-1/TCCR plus p3XFLAG-gp130 (D). After 36 h, the cells were stimulated for 45 min with the culture supernatant containing either scIL-12 (IL-12) (B), scIL-23 (IL-23) (C), or scIL-27 (IL-27) (D) at final concentrations of 0, 0.5, 5, and 50%. As negative controls, the cells were stimulated for 45 min with the culture supernatant containing either scIL-23/scIL-27 (B), scIL-12/scIL-27 (C), or scIL-12/scIL-23 (D) at a final concentration of 50%. The cells were then subjected to Western blotting with anti-STAT1 (Total STAT1) (D), anti-STAT4 (Total STAT4) (B and C), anti-pY-STAT1 (pY-STAT1) (D), and anti-pY-STAT4 (pY-STAT4) (B and C) antibodies.

Enhanced induction of HCV-specific CTLs by coadministration of either p3XFLAG-IL-23 or p3XFLAG-IL-27 in the prime-boost immunization.

We then tested whether adjuvant effects could be exerted by these cytokine expression plasmids in the prime-boost immunization for induction of HCV-specific CTLs. As shown in Fig. 1, HHD mice received the prime-boost immunization consisting of priming and the first boosting with 100 μg of either pCEP4-core, pCEP4-NS3, or pCEP4-NS4, together with 100 μg of any one of p3XFLAG, p3XFLAG-IL-12, p3XFLAG-IL-23, and p3XFLAG-IL-27, followed by the second boosting with either Adex1SR3ST for core-specific CTL induction or Adex1CA3269 for NS3 and NS4-specific CTL induction. Spleen cells were then stimulated in vitro with appropriate peptide-pulsed syngeneic spleen cells. One week later, CTL assays were performed at various E:T ratios. In agreement with the previous results (35), the induction of CTLs specific for various HCV-derived epitopes, including core-132 (Fig. 3A), NS3-1073 (Fig. 3B), NS4-1666 (Fig. 3C), and NS4-1769 (Fig. 3D), was greatly enhanced by coinoculation of p3XFLAG-IL-12 in conjunction with the prime-boost immunization in comparison with coinoculation of the empty vector, p3XFLAG. It was also found that coadministration of either p3XFLAG-IL-23 or p3XFLAG-IL-27 in the prime-boost immunization greatly enhanced induction of CTLs specific for these epitopes (Fig. 3) relative to coinoculation of the empty vector. These results clearly demonstrate that both p3XFLAG-IL-23 and p3XFLAG-IL-27, as well as p3XFLAG-IL-12, have potent adjuvant activities to induce high levels of HCV-specific CTLs in the prime-boost immunization. Because it has been reported that an inhibitory effect was observed by inoculation with an IL-12 expression plasmid at a dose as high as 200 μg (61), we tested the adjuvant activities of p3XFLAG-IL-12, p3XFLAG-IL-23, and p3XFLAG-IL-27 at lower doses than 100 μg. However, significantly greater boosting was not observed at the lower doses (Fig. 4A), indicating that there was no inhibitory effect at a dose of 100 μg in our experimental system. Hence, an inoculation dose of 100 μg was used in the following experiments. On the other hand, the combination of p3XFLAG-IL-12, p3XFLAG-IL-23, and p3XFLAG-IL-27 led to a synergistic adjuvant effect in comparison to an equal amount of p3XFLAG-IL-12 (Fig. 4B).

FIG. 3.

Enhancement of HCV-specific CTL induction by coadministration of either p3XFLAG-IL-12, p3XFLAG-IL-23, or p3XFLAG-IL-27 in the prime-boost immunization. Mice were immunized as follows: priming and the first boosting with 100 μg of either pCEP4-core (A), pCEP4-NS3 (B), or pCEP4-NS4 (C and D), together with 100 μg of either p3XFLAG-IL-12 (circles), p3XFLAG-IL-23 (triangles), p3XFLAG-IL-27 (squares), or p3XFLAG (diamonds), followed by the second boosting with 5 × 10⁷ PFU of either Adex1SR3ST (A) or Adex1CA3269 (B to D). The interval between immunizations was 2 weeks. At 2 to 3 weeks after the last immunization, spleen cells were prepared and stimulated in vitro with syngeneic spleen cells pulsed with 10 μM concentrations of core-132 (A), NS3-1073 (B), NS4-1666 (C), or NS4-1769 (D). After 1 week, ⁵¹Cr release assays were performed at various E:T ratios with RMA-HHD cells pulsed with (open symbols) or without (solid symbols) 10 μM concentrations of core-132 (A), NS3-1073 (B), NS4-1666 (C), or NS4-1769 (D) as targets. The data are representative of one of three independent experiments and are shown as the mean ± the standard deviation (SD) of triplicate wells. Similar results were obtained in the three independent experiments.

FIG. 4.

(A) Dose-response curves for the adjuvant plasmids p3XFLAG-IL-12, -IL-23, and -IL-27. Mice were primed and boosted with 100 μg of pCEP4-core, together with various doses (0, 10, 30, and 100 μg) of either p3XFLAG-IL-12 (circles), p3XFLAG-IL-23 (inverted triangles), or p3XFLAG-IL-27 (squares), followed by the second boosting with 5 × 10⁷ PFU of Adex1SR3ST. (B) Synergistic effect of the combination of p3XFLAG-IL-12, -IL-23, and -IL-27. Mice were primed and boosted with 100 μg of pCEP4-core, together with 100 μg of either p3XFLAG, p3XFLAG-IL-12 (IL-12), or the combination of p3XFLAG-IL-12, -IL-23, and -IL-27 (33 μg each) (IL-12/23/27), followed by a second boosting with 5 × 10⁷ PFU of Adex1SR3ST. At 2 to 3 weeks after the last immunization of either the experiment A or B, spleen cells were prepared and stimulated in vitro with syngeneic spleen cells pulsed with 10 μM concentrations of core-132. After 1 week, ⁵¹Cr release assays were performed at an E:T ratio of 50 with RMA-HHD cells pulsed with (open symbols or bars) or without (solid symbols or bars) 10 μM core-132 as targets. The data are shown as the mean ± the SD of triplicate wells.

Quantification of IFN-γ-producing cells in response to HCV-derived peptides by intracellular IFN-γ staining and ELISPOT.

To further address the adjuvant activities of p3XFLAG-IL-12, -IL-23, and -IL-27, we next examined the numbers of HCV-specific CD8⁺ T cells by staining for antigen-induced intracellular IFN-γ in spleen cells of HHD mice immunized by the prime-boost immunization in combination with either p3XFLAG-IL-12, -IL-23, or -IL-27. Since the cells were stimulated in vitro with an appropriate peptide for only 5 h in this assay, the possibility of substantial in vitro expansion of responder cells is precluded (8). After the consecutive immunization involving priming and the first boosting with pCEP4-core, along with p3XFLAG, followed by a second boosting with Adex1SR3ST (pCEP4-core/p3XFLAG ×2 + Adex1SR3ST), ∼6.5% of the CD8⁺ spleen cells produced IFN-γ in response to stimulation with core-132 (Fig. 5D), whereas the frequency of IFN-γ-producing CD8⁺ cells was very low (1.1%) in nonimmunized mice (Fig. 5B). On the other hand, the percentages of IFN-γ-producing CD8⁺ cells in response to core-132 were 23.9, 18.3, and 20.2% in mice immunized with pCEP4-core/p3XFLAG-IL-12 ×2 + Adex1SR3ST (Fig. 5F), pCEP4-core/p3XFLAG-IL-23 ×2 + Adex1SR3ST (Fig. 5H), and pCEP4-core/p3XFLAG-IL-27 ×2 + Adex1SR3ST (Fig. 5J), respectively. Furthermore, similar patterns were also observed in stimulations with NS3-1073. As shown in Fig. 5K to T, coinoculation of either p3XFLAG-IL-12, -IL-23, or -IL-27 in the prime-boost immunization resulted in significant expansions of the IFN-γ-producing CD8⁺ spleen cells in response to NS3-1073. Differences in the increase of the HCV-specific CD8⁺ response in Fig. 5 to NS3-1073 compared to core-132 could be explained by a difference in the immunogenicity between the two epitopes (unpublished data). The IFN-γ-producing CD8⁺ T cells in response to core-132 were further stained with either anti-CD11a or anti-CD62L antibody. The IFN-γ-producing CD8⁺ T cells expressed the CD11a^high/CD62L^low effector phenotype (21, 66) in mice immunized with either pCEP4-core/p3XFLAG-IL-12 ×2 + Adex1SR3ST (99.5%/94.4%), pCEP4-core/p3XFLAG-IL-23 ×2 + Adex1SR3ST (98.7%/95.5%), or pCEP4-core/p3XFLAG-IL-27 ×2 + Adex1SR3ST (99.0%/93.1%).

FIG. 5.

Intracellular IFN-γ staining of HCV-specific CD8⁺ cells in spleen cells of mice that received coinoculation of either p3XFLAG-IL-12, p3XFLAG-IL-23, or p3XFLAG-IL-27 in the prime-boost immunization. HHD mice were immunized as follows: priming and first boosting with 100 μg of either pCEP4-core (C to J) or pCEP4-NS3 (M to T), along with 100 μg of either p3XFLAG (C, D, M, and N), p3XFLAG-IL-12 (E, F, O, and P), p3XFLAG-IL-23 (G, H, Q, and R), or p3XFLAG-IL-27 (I, J, S, and T); followed by second boosting with 5 × 10⁷ PFU Adex1SR3ST (C to J) or Adex1CA3269 (M to T). Nonimmunized mice (naive) (A, B, K, and L) were used as a negative control. At 2 to 3 weeks after the last immunization, spleen cells were prepared and cultured for 5 h in the presence or absence (A, C, E, G, I, K, M, O, Q, and S) of 10 μM core-132 (B, D, F, H, and J), or 50 μM NS3-1073 (L, N, P, R, and T). After stimulation, cells were stained for their surface expression of CD8 (x axis) and their intracellular expression of IFN-γ (y axis). The numbers shown indicate the percentages of CD8⁺ cells that are positive for intracellular IFN-γ. The data shown are representative of three independent experiments. Three to five mice per group were used in each experiment, and the spleen cells of the mice per group were pooled.

To confirm the expansions of IFN-γ-producing cells in response to HCV-derived peptides, the numbers of IFN-γ-producing cells were determined in ELISPOT assays. As shown in Table 2, coinoculation of either p3XFLAG-IL-12, -IL-23, or -IL-27 in the prime-boost immunization augmented the numbers of IFN-γ-secreting cells in response to any of the HCV-derived epitopes tested, including core-35, core-132, NS3-1073, NS3-1406, and NS3-1585.

TABLE 2.

Quantitation of IFN-γ-secreting cells by ELISPOT in spleens of HHD mice immunized by the prime-boost immunization^a

Expt and immunizationb	Mean no. of IFN-γ-secreting cells/106 cells ± SD stimulated with:
	No peptide	core-35	core-132	NS3-1073	NS3-1406	NS3-1585
Expt 1
pCEP4-core/p3XFLAG ×2 + Adex1SR3ST	63 ± 11	127 ± 12	483 ± 122
pCEP4-core/p3XFLAG-IL-12 ×2 + Adex1SR3ST	81 ± 14	281 ± 35*	993 ± 134*
pCEP4-core/p3XFLAG-IL-23 ×2 + Adex1SR3ST	107 ± 42	285 ± 34*	1,116 ± 152*
pCEP4-core/p3XFLAG-IL-27 ×2 + Adex1SR3ST	55 ± 10	303 ± 43**	976 ± 146**
Expt 2
pCEP4-NS3/p3XFLAG ×2 + Adex1CA3269	60 ± 19			117 ± 25	124 ± 40	112 ± 9
pCEP4-NS3/p3XFLAG-IL-12 ×2 + Adex1CA3269	81 ± 16			229 ± 19*	292 ± 35*	228 ± 34†
pCEP4-NS3/p3XFLAG-IL-23 ×2 + Adex1CA3269	61 ± 21			246 ± 15*	296 ± 41*	267 ± 19*
pCEP4-NS3/p3XFLAG-IL-27 ×2 + Adex1CA3269	66 ± 25			234 ± 13*	316 ± 46*	250 ± 51†

^aSpleen cells were prepared from mice immunized by the prime-boost immunization, and the numbers of IFN-γ-secreting cells in response to the indicated peptides were determined by ELISPOT. Data represent results from individual mice and are shown as the mean of six determinations (n = 6). *, P < 0.001; **, P < 0.005; †, P < 0.01 compared to the pCEP4-core/p3XFLAG ×2 + Adex1SR3ST or the pCEP4-NS3/p3XFLAG ×2 + Adex1CA3269 regime in response to each peptide.

^bMice received priming and the first boost with 100 μg of either pCEP4-core or pCEP4-NS3, along with 100 μg of either p3XFLAG, p3XFLAG-IL-12, p3XFLAG-IL-23, or p3XFLAG-IL-27, followed by a second boost with 5 × 10⁷ PFU of recombinant adenovirus (Adex1SR3ST or Adex1CA3269).

In support of the data in CTL assays (Fig. 3), these results indicate that coadministration of either p3XFLAG-IL-12, p3XFLAG-IL-23, or p3XFLAG-IL-27 in the prime-boost immunization leads to dramatic expansions of IFN-γ-producing cells in response to HCV-derived peptides. Taken together, these data have revealed that both IL-23 and IL-27, as well as IL-12, are potent adjuvants for the induction of HCV-specific CTLs.

Preinjections of either p3XFLAG-IL-12, -IL-23, or -IL-27 enhance the efficiency of HCV-specific CTL induction raised by immunization with the recombinant adenovirus.

To obtain control data for Fig. 5, we examined the frequencies of CD8⁺ T cells producing antigen-induced intracellular IFN-γ in spleen cells of mice primed and boosted with the empty vector, pCEP4, together with p3XFLAG-IL-12, p3XFLAG-IL-23, or p3XFLAG-IL-27, followed by a second boosting with Adex1SR3ST. As shown in Fig. 6, the percentages of the IFN-γ-producing CD8⁺ cells were 7.3, 9.0, and 4.5% in mice immunized with pCEP4/p3XFLAG-IL-12 ×2 + Adex1SR3ST (Fig. 6L), pCEP4/p3XFLAG-IL-23 ×2 + Adex1SR3ST (Fig. 6N), and pCEP4/p3XFLAG-IL-27 ×2 + Adex1SR3ST (Fig. 6P), respectively. These values were much higher than that (1.3%) in mice immunized with pCEP4-core/p3XFLAG ×2 + Adex1SR3ST (Fig. 6B), although these values were clearly less than, but comparable to those in mice immunized with pCEP4-core/p3XFLAG-IL-12 ×2 + Adex1SR3ST (18.3%) (Fig. 6D), pCEP4-core/p3XFLAG-IL-23 ×2 + Adex1SR3ST (22.3%) (Fig. 6F), and pCEP4-core/p3XFLAG-IL-27 ×2 + Adex1SR3ST (16.4%) (Fig. 6H), respectively. Furthermore, killing response by core-132-specific CTLs was greatly enhanced by double preinjections of either p3XFLAG-IL-12, -IL-23, or -IL-27 without pCEP4-core but with pCEP4, followed by immunization with Adex1SR3ST (Fig. 7). These results indicate that preinjections of either p3XFLAG-IL-12, -IL-23, or -IL-27 enhance the efficiency of HCV-core-specific CTL induction raised by immunization with Adex1SR3ST.

FIG. 6.

Intracellular IFN-γ staining of HCV core-specific CD8⁺ cells in spleen cells of mice that received preinjections with either p3XFLAG-IL-12, p3XFLAG-IL-23, or p3XFLAG-IL-27, followed by immunization with Adex1SR3ST. HHD mice were primed and boosted with 100 μg of either pCEP4-core (A to H) or pCEP4 (K to P), together with 100 μg of either p3XFLAG (A and B), p3XFLAG-IL-12 (C, D, K, and L), p3XFLAG-IL-23 (E, F, M, and N), or p3XFLAG-IL-27 (G, H, O, and P), followed by the second boosting with 5 × 10⁷ PFU Adex1SR3ST (A to H and K to P). At 2 to 3 weeks after the last immunization, spleen cells were prepared and cultured for 5 h in the presence (B, D, F, H, J, L, N, and P) or absence (A, C, E, G, I, K, M, and O) of 10 μM core-132. After stimulation, cells were stained for their surface expression of CD8 (x axis) and their intracellular expression of IFN-γ (y axis). The numbers shown indicate the percentages of CD8⁺ cells that are positive for intracellular IFN-γ. The data shown are representative of three independent experiments. We analyzed all data of the three experiments and concluded that the difference between panel B and either panel L (P = 0.005), N (P = 0.038), or P (P = 0.003) was statistically significant as determined by applying the Student t test. At least three mice per group were used in each experiment, and spleen cells of mice per group were pooled.

FIG. 7.

Enhancement of CTL induction specific for HCV-core by preinjections of either p3XFLAG-IL-12, p3XFLAG-IL-23, or p3XFLAG-IL-27, followed by immunization with Adex1SR3ST. Mice were preinjected twice with either p3XFLAG-IL-12 (circles), p3XFLAG-IL-23 (inverted triangles), or p3XFLAG-IL-27 (squares), along with pCEP4, followed by immunization with Adex1SR3ST. As a control, mice were primed and boosted with pCEP4-core, together with p3XFLAG, followed by second boosting with Adex1SR3ST (diamonds). The injection doses were 100 μg of plasmid DNA and 5 × 10⁷ PFU of virus, and the interval between immunizations was 2 weeks. After the last immunization, spleen cells were prepared and stimulated in vitro with syngeneic spleen cells pulsed with 10 μM core-132. After 1 week, ⁵¹Cr release assays were performed at various E:T ratios with RMA-HHD cells pulsed with (open symbols) or without (solid symbols) 10 μM core-132 as targets. The data are representative of one of three independent experiments and are shown as the mean ± the SD of triplicate wells. Similar results were obtained in the three independent experiments.

To further confirm these results in Fig. 6, we performed similar experiments accompanied with supplementary controls. In accordance with the data in Fig. 6, the frequency of IFN-γ-producing CD8⁺ cells in mice immunized with pCEP4-core ×2 + Adex1SR3ST (Fig. 8D) was higher than that in mice immunized with p3XFLAG ×2 + Adex1SR3ST (Fig. 8F) but lower than in mice that received double preinjections of either p3XFLAG-IL-12 (Fig. 8H), -IL-23 (Fig. 8J), or -IL-27 (Fig. 8L) without pCEP4, followed by immunization with Adex1SR3ST. In contrast, IFN-γ-secreting CD8⁺ cells in response to core-132 were not practically detected in mice that received double preinjections of either p3XFLAG-IL-12, -IL-23, or -IL-27, followed by injection of Adex1w (Fig. 8M to R). These data confirm that preinjections of either p3XFLAG-IL-12, -IL-23, or -IL-27 enhance induction of HCV-specific CTLs raised by immunization with the recombinant adenovirus. In addition, preinjections of p3XFLAG-IL-27 marginally enhanced induction of peptide-specific IFN-γ-secreting CD8⁺ cells in mice immunized with pCEP4-core (Fig. 8V), as well as preinjections of either p3XFLAG-IL-12 or -IL-23 (data not shown).

FIG. 8.

Intracellular IFN-γ staining of HCV core-specific CD8⁺ cells in spleen cells of mice that received preinjections with either p3XFLAG-IL-12, p3XFLAG-IL-23, or p3XFLAG-IL-27, followed by immunization with either recombinant adenovirus or plasmid DNA. HHD mice were injected with various combinations of 100 μg of plasmid DNA and/or 5 × 10⁷ PFU of recombinant adenovirus as follows: priming and the first boosting with pCEP4-core, followed by the second boosting with Adex1SR3ST (C and D); preinjections twice with either p3XFLAG (E, F, S, and T), p3XFLAG-IL-12 (G, H, M, and N), p3XFLAG-IL-23 (I, J, O, and P), or p3XFLAG-IL-27 (K, L, Q, R, and U-X), followed by immunization with Adex1SR3ST (C-L), Adex1w (M-R), pCEP4-core (S-V), or pCEP4 (W and X). At 2 to 3 weeks after the last immunization, spleen cells were prepared and cultured for 5 h in the presence (B, D, F, H, J, L, N, P, R, T, V, and X) or absence (A, C, E, G, I, K, M, O, Q, S, U, and W) of 10 μM core-132. After in vitro stimulation, the cells were stained for their surface expression of CD8 (x axis) and their intracellular expression of IFN-γ (y axis). The numbers shown indicate the percentages of CD8⁺ cells that are positive for intracellular IFN-γ. The data shown are representative of three independent experiments. We analyzed all data of the three experiments and concluded that the difference between panel D and either panel H (P = 0.002), J (P = 0.015), or L (P = 0.001) was statistically significant as determined from the Student t test. At least three mice per group were used in each experiment, and spleen cells of mice per group were pooled.

Taken together, these data also support the idea that the novel cytokines IL-23 and IL-27, as well as IL-12, have potent adjuvant activities for the induction of HCV-specific CTLs.

DISCUSSION

It was previously demonstrated that p40-deficient mice are more susceptible to infection with Mycobacterium tuberculosis (11) and Cryptococcus neoformans (14) than p35-deficient mice. In addition, p35-deficient mice are quite resistant to Listeria monocytogenes at low doses of infection (7). Because p35-deficient mice do not produce IL-12 but can make IL-23 composed of p40 and p19, these observations suggest that IL-23 plays a crucial role in the clearance of pathogens. Furthermore, mice lacking one subunit of the IL-27 receptor, WSX-1/TCCR, were remarkably susceptible to Leishmania major infection and showed impaired IFN-γ production early in the infection (68). Thus, these data strongly suggest that both IL-23 and IL-27 are crucial for T-cell-mediated immunity against various infectious agents.

In the present study, we have demonstrated that the coadministration of either p3XFLAG-IL-23 or p3XFLAG-IL-27, as well as p3XFLAG-IL-12, in the prime-boost immunization enhanced induction of CTLs specific for various HCV-derived epitopes (Fig. 3) and led to dramatic expansions of IFN-γ-producing cells in response to HCV-derived, CTL epitopes (Fig. 5 and Table 2). We have also shown that preinjections of either p3XFLAG-IL-12, -IL-23, or -IL-27 resulted in great increases in IFN-γ-producing, HCV-specific CD8⁺ cells induced by immunization with the recombinant adenovirus. Since p3XFLAG-IL-12, -IL-23, and -IL-27 produce high amounts of scIL-12, scIL-23, and scIL-27 fusion proteins in mammalian cells, respectively (Fig. 1), all of these data indicate that the secreted scIL-23 and scIL-27, as well as scIL-12 are potent adjuvants for the induction of HCV-specific CTLs. Each of the single-chain cytokines is composed of two appropriate subunits connected by a flexible 15-amino-acid linker. Using the same (5) or a similar (31) peptide linker, the previously described scIL-12 and scIL-23 fusion proteins showed comparable biological activity to those of the native heterodimeric IL-12 and IL-23, respectively. Furthermore, we have shown that the scIL-12, scIL-23, and scIL-27 fusion proteins could induce tyrosine phosphorylation of STAT via IL-12Rβ1/IL-12Rβ2, IL-12Rβ1/IL-23R, and WSX-1/gp130, respectively (Fig. 2B to D). Taken together, the current data have revealed that two novel cytokines, IL-23 and IL-27, possess potent adjuvant activities for induction of antigen-specific CTLs, as well as IL-12. As far as we know, this is the first report of the adjuvant effect of IL-27 on epitope-specific CTL induction and the second one, following the study described by Lo et al. (31), for IL-23. The present data, however, do not show whether these cytokines act directly on CTL or not.

In some sense, it is easily predictable that IL-23 and IL-27 possess adjuvant activities for CTL induction because they are in many aspects closely related to IL-12. IL-12 has been shown to be a potent adjuvant for CTL induction by a number of groups including us (6, 20, 25, 35, 38, 41, 49, 56, 59, 61). First, all three cytokines are heterodimers that consist of related subunits (42, 48). Second, the subunits of their receptors are also closely related (42, 48). In addition to the structural similarity, they display biological homology. All three cytokines are dominant factors in the development of the Th1 phenotype and do not support the Th2 development (52, 60). However, they act on T cells at different stages during the development of Th1-mediated immune response. The activities of IL-23 are preferentially restricted to memory CD4⁺ T cells, whereas IL-12 mainly acts on naive CD4⁺ T cells. In contrast, IL-27 does not affect memory CD4⁺ T cells at all but acts on naive CD4⁺ T cells at an earlier stage than does IL-12 (52, 60). In addition, it is known that IL-27 synergizes with IL-12 to produce IFN-γ in naive T cells (48). Accordingly, it is likely that mechanisms of the enhancement in HCV-specific CTL induction are different between the three cytokines and, therefore, the combined and sequential use of the three cytokines should be more effective for antigen-specific CTL induction than the single use. Actually, it was reported that the combined exposure of IL-12 and IL-23 resulted in additive effects on cytokine production from dendritic cells (5). We also showed that the combination of p3XFLAG-IL-12, p3XFLAG-IL-23, and p3XFLAG-IL-27 led to a synergistic adjuvant effect for induction of HCV-specific CTLs (Fig. 4B).

It is surprising that double preinjections of either p3XFLAG-IL-12, -IL-23, or -IL-27 have potent adjuvant effects on the induction of HCV-specific CTLs raised by immunization with Adex1SR3ST (Fig. 6 to 8), although the increases in HCV-specific CD8⁺ cells in this case were substantially less than in mice that received coinjections of either p3XFLAG-IL-12, -IL-23, or -IL-27 in conjunction with the prime-boost immunization (Fig. 6). It is possible to explain that the persistently secreted, single-chain cytokine was sufficiently left in mice and enhanced the cell-mediated immunity elicited by the recombinant adenovirus. In contrast, double preinjections of a different IL-12 expression plasmid, named pCAGGS-IL-12, did not significantly enhance HCV-specific CTL induction raised by immunization with Adex1SR3ST, although coinjections of pCAGGS-IL-12 in the prime-boost immunization did lead to dramatic enhancement of HCV-specific CTL induction (35). In pCAGGS-IL-12, the gene named IRES (for internal ribosome entry site) of the encephalomyocarditis virus exists between p35 and p40 cDNAs and allows the two subunits to coexpress in cells (35, 40). The genetically linked scIL-12 is presumably more stable than the natively associated, heterodimeric IL-12 produced by pCAGGS-IL-12. In addition, the monomeric nature of the single-chain fusion protein ensures equimolar expression of each subunit, thereby avoiding the formation of inhibitory p40 dimers (32). These differences might be able to explain the discrepancy of the opposite results. If this is the case, an artificial single-chain forms of IL-12, IL-23, or IL-27 might be more suitable as an adjuvant than a native form.

On the other hand, negative effects of these cytokines should be considered. Actually, the first clinical trial of IL-12 resulted in the death of 2 persons and caused severe toxic effects in 15 others (23, 28). It is noteworthy that artificial mutant IL-12 reduced toxicity without reducing the adjuvant activity (20). IL-12 has also been considered to be related to several autoimmune diseases such as multiple sclerosis, diabetes, rheumatoid arthritis, and inflammatory bowel disease (15). Many data on a major role for IL-12 in multiple sclerosis have been obtained from rodent studies of the experimental autoimmune encephalomyelitis. However, most of the experiments were performed primarily through the use of neutralizing antibodies against p40 that affect both IL-12 and IL-23 (10, 29, 30). Recently, Cua et al. (13) compared the susceptibility of mice lacking either IL-12 alone (p35 deficient), IL-23 alone (p19 deficient), or both cytokines (p40 deficient) to experimental autoimmune encephalomyelitis and showed that IL-23 rather than IL-12 is the crucial factor for the disease. Although no side effects of IL-27 have been reported, too much activation by IL-27 might cause toxicity and/or autoimmunity.

In the chronic HCV infection, the precursor frequency of HCV-specific CTLs is quite low (27, 50, 51). Therefore, people believe that this is one of the major reasons why HCV-specific CTLs cannot clear the virus in the chronic stage. In addition to the quantitative aspect, several groups have suggested that the impaired T-cell-mediated immune response may establish chronicity in HCV infection. In fact, it has been reported that HCV-specific CD8⁺ T cells in the chronic HCV infection show less proliferation after antigen stimulation, less production of IFN-γ and TNF-α, and less HCV-specific cytotoxicity (17, 64). Further, Bain et al. (3) demonstrated the impaired stimulatory function of dendritic cells in chronic HCV infections. The mechanisms of the dysfunctional immune response are still unclear (39, 57). However, supplementation of the vaccine with IL-12, IL-23, and/or IL-27 may enhance the normal development of Th1-associated immunity in patients with chronic HCV infection, thereby causing qualitatively and quantitatively improved induction of HCV-specific CTLs. In fact, it has been reported that the coadministration of gene plasmids encoding the Th1-type cytokines (i.e., IL-2, IL-12, IL-15, and IL-18) enhanced Th1 immune responses (38, 56).

In conclusion, we have shown that coadministration of either an IL-23 or an IL-27 expression plasmid, as well as of an IL-12 expression plasmid, in the prime-boost immunization enhanced the induction of HCV-specific CTLs in mice. We have also demonstrated that preinjections of either an IL-12, an IL-23, or an IL-27 expression plasmid promoted HCV-specific CTL induction raised by the recombinant adenovirus. These results have revealed that IL-23 and IL-27, as well as IL-12, possess potent adjuvant effects on induction of epitope-specific CTLs. The two novel cytokines might offer new prophylactic and therapeutic strategies for CTL-based DNA vaccine against infectious pathogens such as HCV.