OVA-induced allergic diarrhea occurs as a consequence of over-expression of Th1 inhibitory IL-12p40 monomers and homodimers in the large intestine, establishing a dominant Th2-type environment. In this study, we demonstrate that intranasally administered murine IL-12p70 naked DNA expression plasmids resulted in the synthesis of corresponding cytokine in the large intestinal CD11c+ dendritic cells, leading to the inhibition of Ag-specific Th2-type response for the prevention of allergic diarrhea and the suppression of clinical symptoms including OVA-specific IgE Ab synthesis. The nasal IL-12p70 DNA treatment proved effective even after the establishment of allergic diarrhea. These results suggest that the mucosal administration of naked IL-12p70 DNA plasmid should be considered as a possible preventive and therapeutic treatment for Th2 cell-mediated food allergic diseases in the intestinal tract.
Allergic diseases of the respiratory and gastrointestinal tracts, including pollinosis, asthma, and allergic diarrhea, respectively, are generally characterized by hyperresponsiveness stemming from the accumulation of eosinophils, mast cells, and mononuclear cells in mucosal tissues along with an elevation of Ag-specific IgE. This hyperreaction is caused by a defect in the immunological balance between Th1 and Th2 lymphocytes, leading to an increase in Th2 cells. The Th2 cytokines IL-4, IL-5, and IL-13 have been shown to be key mediators of the pathology of these allergic diseases (1, 2, 3). In our previous study, large intestinal IL-4-producing Th2 cells induced by systemic priming followed by oral challenge with OVA were shown to play a major pathologic role in the development of allergic diarrhea (4). Furthermore, mast cells, IgE and FcεR were also shown to contribute to the induction of this experimental Th2 intestinal allergy (5). Intestinal allergic inflammation was also induced in the small intestine by the eosinophils, eotaxin, and IL-5 associated with a pathologic Th2 environment in mice treated with OVA in enteric-coated beads (6). Our most recent study elucidated the molecular and cellular pathologic mechanisms underlying Th2 cell-mediated intestinal allergic disease by showing that locally produced IL-12p40 contributes to the generation of a dominant Th2-type environment in the large intestine of mice with allergic diarrhea (7).
Dendritic cell (DC)3-derived heterodimeric structure, a p35 and a p40 subunit of IL-12 (IL-12p70), is known to promote Th1 cell differentiation and to stimulate the production of IFN-γ from Th1 cells and NK cells (8, 9, 10). Production of IL-12 by activated macrophages and DCs also results in the secretion of monomeric and homodimeric IL-12p40 that antagonize IL-12p70 bioactivity via binding to the β1 subunit of the IL-12R (11, 12, 13). IL-12p40 transgenic mice have been shown to be more susceptible to the malaria infection due to the reduced Th1 responses (14). These findings, together with our previous observation that large intestinal IL-12p40 contributes to the generation of Th2 cell-mediated allergic disease, suggest that the mucosal IL-12 balance between p70 and p40 might be a key regulator for the Th1 and Th2 networks associated with the mucosal immune compartment (7). Because it is located on chromosome 5q with its allergy-associated cytokine gene cluster of IL-4, IL-5, and IL-13 (15), the IL-12p40 gene in particular can be considered as one of the potent regulators of Th2 cell induction.
Although IL-12p40 can shift the delicate balance existing between the Th1 and Th2 networks more toward the Th2 environment, IL-12p70 is a well-known Th1 inducer with potential for use in treating Th2-dominant diseases such as asthma (16). IL-12 has also been shown to inhibit IL-4 production in bulk cultures of peripheral blood leukocytes from allergic patients and thus to markedly suppress IL-4-mediated IgE production (17, 18). In addition, IL-12 prevents the differentiation of bone marrow cells into eosinophils in a murine model of asthma (19). The i.p. delivery of IL-12 has been shown to inhibit murine Ag-induced eosinophilic inflammation airway hyperresponsiveness and IgE production (20). In a clinical study of allergic asthma, s.c. administration of rIL-12 resulted in a significant decrease in the number of circulating eosinophils, however only minor effects were seen on histamine-associated airway hyperresponsiveness (16). These studies collectively suggest that IL-12 may be an effective suppressor of allergen-induced airway hyperreaction and of atopic asthma-associated inflammation.
Gene therapy using IL-12 may also have a role to play in the development of vaccines administered by mucosal or systemic routes and in disease modification (21, 22). Gene therapy has recently been used both experimentally and clinically as a tool for developing new vaccines or for halting the progression of immunological diseases. Nasal TGF-β DNA plasmid has been shown to be effective in mitigating the severity of ocular and intestinal inflammatory diseases (23, 24) The administration of Fms-like tyrosine kinase-3 (Flt3) ligand DNA has also been shown to increase the number of activated lymphoid DCs and to thereby induce Ag-specific mucosal and systemic Ab responses (25). Although the mucosal delivery of naked DNA specific for regulatory molecules has been shown to be effective for the modulation of immune responses, the fate of the mucosally delivered naked DNA and its efficacy in inducing and regulating immune responses at distant sites remains unknown. In the current study, we demonstrate that nasal IL-12p70 DNA administration results in the expression of the corresponding protein in large intestinal DCs, which promote the shift to a Th1-type cell response at the disease site for the prevention and treatment of Th2-mediated allergic diarrhea.
Materials and Methods
BALB/c mice were purchased from CLEA Japan. All mice were 6–7 wk of age at the beginning of the individual experiments.
The plasmid pORF9-IL-12p70 consists of the pORF9 multiple cloning site (pORF9-mcs) vector plus the full-length recombinant murine IL-12p40 and IL-12p35 cDNA gene (InvivoGen). The pORF is an expression vector containing the hybrid elongation factor 1α/human T cell leukemia virus promoter and the ampicillin-resistant gene. The plasmid DNA was purified using the EndoFree Plasmid Mega kit (Qiagen). pORF-mcs empty plasmid was used as control vector. For use in a time kinetics study, pIRES2-EGFP plasmid was purchased from BD Biosciences. Allophycocyanin anti-CD11b (M1/70, rat IgG2b), allophycocyanin anti-CD11c (HL3, hamster IgG), and allophycocyanin anti-B220 (RA3-6B2, rat IgG2a) were purchased from BD Pharmingen. Anti-IL-12p70 Ab was obtained from R&D Systems and was biotinylated for immunohistochemical analysis. Streptavidin-PE and Alexa Fluor 660-donkey anti-goat IgG were obtained from Molecular Probes. Anti-LYVE (lymphatic vessel endothelial hyaluronan receptor 1) (26) was obtained from Santa Cruz Biotechnology.
Induction of allergic diarrhea
For the induction of allergic diarrhea, we used a previously established protocol (4, 7). Briefly, on the first day of the experiment (day 0), mice were primed by s.c. injection of 1 mg of OVA in CFA (Difco). One week after the systemic priming (day 7), mice were repeatedly challenged with 50 mg of OVA by oral route three times per week for several weeks (4). Within 1–2 h after the tenth administration with OVA, the mice were sacrificed and analyzed.
Nasal IL-12p70 DNA treatment
In vivo nasal treatment was performed as previously described (25). Briefly, BALB/c mice were nasally administered with 50 μg of purified IL-12p70 plasmid or pORF vector (with empty plasmid used as a control) twice a week for the duration of the experiment. Nasal DNA treatment was started at the time of oral challenge to examine preventive effects. In experiments designed to elucidate the therapeutic potential of nasal IL-12 DNA, mice with ongoing allergic diarrhea were treated with nasal IL-12 DNA.
ELISA for OVA-specific IgE Abs in serum
To assess OVA-specific IgE Ab levels in serum, a sandwich ELISA system was adopted as described earlier (4). End-point titers of OVA-specific IgE Abs were expressed as the reciprocal log2 of the last dilution that showed a level of absorbance 0.1 higher than that of the sera of nonimmune background mice. Total IgE was also analyzed by a sandwich ELISA system that used anti-mouse IgE (R35-72; BD Pharmingen) to capture mAb and biotin anti-mouse IgE (R35-92; BD Pharmingen) for use as the detection mAb (4).
Isolation of mononuclear cells and the Ag-specific Ig production assay
To isolate mononuclear cells from small and large intestines, we used an enzymatic dissociation method (4, 7). Mononuclear cells were incubated in 96-well nitrocellulose plates (Millititer HA; Millipore) precoated with OVA (1 mg/ml) for 4 h at 37°C with 5% CO2 in air. OVA-specific Ab-forming cells (AFCs) were detected by addition of peroxidase-labeled anti-mouse IgM, IgG, IgG1, IgG2a, IgG2b, and IgA Abs (Southern Biotechnology Associates) and visualized by the reaction of 3-amino-9-ethylcarbozole (Moss). OVA-specific AFCs were automatically counted by using the KS ELISPOT compact (Carl Zeiss).
Ag-induced cytokine analysis
For the detection of IL-4- and IFN-γ-producing CD4+ T cells, intracellular FACS analysis was performed (27). Briefly, mononuclear cells isolated from large intestine or spleen of empty vector-treated allergic mice, IL-12p70 DNA-treated mice, or healthy control mice were incubated with OVA (1 mg/ml) at 37°C for 4 days and were then reacted with 10 μg/ml monensin for the last 5 h of culture. The cells were washed with PBS (pH 7.2) after removal of dead cells using Ficoll gradient separation and fixed with 4% paraformaldehyde in PBS for 20 min at room temperature. Fixed cells were washed, suspended in staining buffer (PBS containing 2% FCS and 0.01% NaN3), and incubated with 1 μg/ml PE anti-IL-4 mAb and FITC anti-IFN-γ (BD Pharmingen) in staining buffer containing 0.1% saponin (Sigma-Aldrich) in the presence of 5 μg/ml anti-FcγRII/III (2.4G2; BD Pharmingen) for 30 min at 4°C. After washing three times with staining buffer containing saponin, the cells were stained with 1 μg/ml allophycocyanin anti-CD4 mAb (BD Pharmingen) in staining buffer without saponin for 20 min at 4°C. Following a final washing, the cells were analyzed with a FACSCalibur (BD Biosciences) using the CellQuest software.
Immunoprecipitation and Western blot analyses
For the detection of IL-12p70 heterodimers comprised of IL-12p40 and IL-12p35, large intestinal tissue extracts were prepared as previously described with minor modifications (7). Large intestines were removed, minced in cold PBS with protease inhibitor (Complete Mini; Roche Diagnostics), homogenized and incubated to allow cytokine release from the tissue. After centrifugation and the measurement of their protein concentrations, intestinal tissue extracts were precleared with protein G-Sepharose beads (Pharmacia Biotech), incubated with anti-IL-12p40, and mixed with protein G-Sepharose beads. The beads were washed and then subjected to SDS-PAGE under nonreducing conditions. After electrophoresis, proteins were transferred to a polyvinylidene difluoride microporous membrane (Immobilon; Millipore) and the membrane was reacted with biotinylated anti-IL-12p70 (48110.11, rat IgG1; R&D Systems) and then incubated with a biotin-streptavidin complex (ABC-AP kit; Vector Laboratories). The signal was visualized using a nitroblue tetrazolium/5-bromo-4-chloro-3-indolyl phosphate substrate kit (Bio-Rad). The anti-IL-12p70 mAb has no cross-reactivity with recombinant murine IL-12p40.
Following extensive washing, large intestines were fixed in 4% paraformaldehyde-PBS and treated with sucrose gradient, before freezing in OCT embedding medium as previously described with minor modifications (7). For IL-12p70 immunostaining, cryosections were subjected to Ag retrieval using 10 mM citric buffer (pH 6.0) for 5 min at 98°C. Slides were then blocked with anti-FcγRII/III (2.4G2; BD Pharmingen) and incubated with biotin anti-IL-12p70 and goat anti-LYVE for 16 h at 4°C. The sections were then treated with Alexa Fluor 660 anti-goat IgG (Molecular Probes) and Streptavidin-PE (Molecular Probes). For surface marker staining, serial sections were incubated with allophycocyanin anti-CD11b (M1/70; BD Pharmingen) or allophycocyanin anti-CD11c (HL3; BD Pharmingen). Counter staining was performed using 4′,6′-diamidino-2-phenylindole (DAPI; Sigma-Aldrich).
In vivo monitoring of protein expression following nasal GFP-DNA administration
For assessing the time kinetics of the protein expression induced by nasal DNA, 50 μg of GFP plasmid was nasally administered to normal BALB/c mice. Then, mice were subjected to in vivo imaging system (IVIS) analysis for 20 s using the Xenogen IVIS CCD camera system at 0, 3, 6, 9, 12, 18, and 24 h after nasal administration (28). After ventral images were taken, the mice were sacrificed and various tissues removed for ex vivo image analysis using the IVIS. The image data of GFP expression was analyzed by LivingImage software (Xenogen) and the GFP intensity from selected areas was quantified.
Nasal GFP-DNA administration and sample analysis
For the time kinetics study of corresponding protein expression following nasal GFP DNA treatment, mice were sacrificed at the indicated time and then analyzed for GFP expression by FACS or subjected to immunohistochemical analysis. Mononuclear cells from nasal passages, spleen, cervical lymph node (CLN), submandibular glands, and mesenteric lymph node (MLN) were isolated, fixed in 4% paraformaldehyde, stained for cell surface markers such as B220, CD4, CD11b, and CD11c and then analyzed for GFP expression using the FACSCalibur. Nasopharynx-associated lymphoid tissue (NALT) and large intestine were fixed in 4% paraformaldehyde-PBS and treated with sucrose gradient, before being frozen in OCT-embedding medium. For lymphoid vessel staining using anti-LYVE (Santa Cruz Biotechnology) (26), cryosections were subjected to Ag retrieval using 10 mM citric buffer at pH 6.0 for 8 min at 98°C. Slides were then blocked with anti-FcγRII/III (2.4G2; BD Pharmingen), incubated with goat anti-LYVE or control goat IgG, and then treated with Alexa Fluor 660 anti-goat IgG. Counter staining was performed using DAPI (Sigma-Aldrich), and immunohistochemical analysis was performed using confocal laser scanning microscopy (Leica).
Analysis of large intestinal cells
As soon as mice developed diarrhea after the final oral OVA challenge, they were sacrificed and cells were isolated from the large intestine as described earlier. Cytospin slides were prepared and then stained with Diff-Quik (Sysmex) for the morphological analysis of differential cell populations including mononuclear cells, eosinophils, basophils, and neutrophils.
Statistical analyses were performed by the two-sample nonparametric Welch test with a significance level of p < 0.01 or p < 0.05 for IgE levels. Values for cytokine synthesis and IgA, IgG, and IgG1 AFCs in the samples between IL-12p70 and control plasmid-treated mice were analyzed by using Student’s t test at values of p < 0.01.
Nasal IL-12p70 DNA prevents allergic diarrhea
In a previous study, we showed that large intestinal IL-12p40 led to a pathologic Th2-dominant environment conducive to the development of OVA-induced allergic diarrhea (7). Because IL-12p70 is one of the most effective Th1-inducing cytokines (8), we sought to examine whether artificially introduced IL-12p70 DNA could prevent the development of allergic diarrhea by up-regulation of IL-12p70 expression of the disease site. Mice nasally treated with the IL-12p70 plasmid did not develop allergic diarrhea, whereas mice given the empty vector developed severe disease as that observed in OVA-induced diarrhea mice (Fig. 1,A). Thus, mice receiving the empty vector served as a control for the remainder of the investigation. In addition, elevated OVA-specific IgE Abs were detected in the serum of diarrhea-afflicted mice treated with the control vector DNA, whereas the mice nasally treated with IL-12p70 DNA showed only low OVA-specific IgE Abs (Fig. 1,B). Furthermore, nasal treatment of IL-12p70 DNA reduced the level of total serum IgE Abs (Fig. 1 C). These results demonstrate that nasal IL-12p70 plasmid prevented the development of OVA-induced allergic diarrhea and reduced both Ag-specific and total IgE responses. Based on these findings, we hypothesized that nasal treatment with IL-12p70 DNA down-regulates the Th2 environment and thus inhibits the development of allergic diarrhea.
Reduction of large intestinal IgA and IgG production by nasal administration of IL-12p70 DNA
To assess nasal IL-12p70 DNA effects on Th1/Th2 responses, we next examined OVA-specific IgM, IgG, IgG1, IgG2a, IgG2b, and IgA Ab responses in large intestinal lamina propria mononuclear cells using the ELISPOT assay. We found that the numbers of Th2-associated OVA-specific IgA and IgG1 AFCs in large intestinal lamina propria mononuclear cells were significantly reduced in the mice treated nasally with IL-12p70 DNA when compared with those of mice treated with the empty vector DNA (Fig. 2,A, left). As we reported earlier (4), OVA-specific IgG2a, IgG2b (data not shown), and IgM AFCs were not detected in the large intestine of all groups of mice examined. In spleen, no significant differences were detected in the numbers of OVA-specific IgA, IgG, and IgG1 AFCs between IL-12p70-treated and control mice (Fig. 2 A, right). As we have shown previously (4), Ag-specific Ab responses are not induced in the small intestine of mice suffering from allergic diarrhea. Thus, OVA-specific IgA, IgG, and IgG1 Ab responses were not detected in the small intestine of either IL-12p70 DNA-treated or control vector-treated mice (data not shown). Although the exact role of OVA-specific IgA and IgG Ab responses in the large intestine of mice with allergic diarrhea still needs to be elucidated, these findings further support the notion that nasal administration of IL-12p70 DNA can inhibit the generation of locally enhanced Th2 cell-mediated OVA-specific Ab responses in mice suffering from allergic diarrhea.
Suppression of large intestinal Th2 cytokine by nasal IL-12p70 DNA
To directly confirm decreased Th2-type responses in the large intestine after nasal IL-12p70 treatment, we next examined Ag-induced cytokine production of the large intestinal CD4+ T cells using intracellular FACS analysis. Nasal IL-12p70 DNA treatment decreased the number of IL-4-producing cells in large intestinal CD4+ T cells when compared with the empty vector-treated mice (Fig. 2,B, left). No difference in the frequency of IFN-γ-producing CD4+ Th cells was seen between the mice treated with nasal IL-12p70 DNA and control vector DNA (Fig. 2,B, left). In contrast, when splenic CD4+ Th cells were examined, no major changes were observed between IL-12p70 DNA-treated and control vector-treated mice (Fig. 2,B, right). However, nasal IL-12p70 treatment did reduce the number of IL-4-producing cells and increase the number of IFN-γ-producing cells in splenic CD4+ Th cells (Fig. 2 B, right). A similar pattern of changes was observed when culture supernatants from the various groups of IL-12p70 DNA- and empty vector-treated mice were examined (data not shown).
Nasal GFP DNA resulted in protein expression in NALT, spleen, and intestine
To directly confirm that the nasal administration of DNA resulted in protein expression in the large intestine, we initially used GFP plasmid DNA for the visualization of corresponding protein expression in vivo. When we used the IVIS to analyze the GFP distribution in vivo following nasal administration of GFP plasmid DNA, the intense fluorescence expression was observed in areas associated with the nasopharynx and CLN as of 3 h after nasal administration (Fig. 3,A). A high intensity of GFP expression was observed in the intestinal region 9 h after nasal administration (Fig. 3,A). In addition, we analyzed ex vivo levels of GFP expression in various tissues isolated from mice that had been given nasal GFP plasmid. The spleen and lymph nodes expressed significant GFP beginning at 3 h after administration and continuing up to 18 h (Fig. 3 B). In mice administered with the GFP gene, GFP expression was particularly strong in CLN. We also analyzed GFP expression levels in the lung, liver, and nasal cavity of mice given nasal IL-12p70 DNA; however, levels of GFP expression were below levels of detection (data not shown). Our effort to analyze GFP expression in the small and large intestine was hampered by autofluorescence, making it difficult to obtain reproducible data.
Next, we assessed whether nasal GFP plasmid would lead to the induction of corresponding protein expression in NALT because the tissue has been shown to be a primary site for the initial uptake of nasal Ags (29). GFP+ cells were detected under the epithelium of NALT, but not until 12 h after nasal administration (Fig. 4,A). These GFP+ cells were examined by costaining them with red fluorescence-conjugated anti-CD11c Ab (Fig. 4 A, right). These data suggest that nasal administration with GFP plasmid resulted in the expression of the corresponding protein in NALT DCs. Because GFP expression had already been noted in other tissues distant from NALT, it can be assumed that some of the administered gene might rapidly pass through NALT without protein expression in the lymphoid tissue.
To chronologically assess the GFP-expressed DCs located in various tissues after nasal GFP plasmid, mice were sacrificed at predetermined time points (0, 1, 3, 6, and 12 h and 1, 2, 3, 4, 5, and 7 days) and mononuclear cells from various lymphoid tissues were analyzed for the presence of GFP+ DCs at the single cell level. When mononuclear cells from nasal passages, spleen, CLN, MLN, and submandibular glands were examined, we detected some GFP+ cells in the DC fraction. However, the frequency was not high enough to meet the minimum limits of reliable FACS analysis. For example, we were able to detect GFP+ CD11c+ cells in MLN at the frequency of ∼0.1%. In contrast, it was possible to detect sufficient numbers of GFP-expressing DCs in spleen due to the quantity of naturally occurring CD11c+ cells. Thus, the levels of CD11c+ cells expressing GFP were detected in spleen as early as 3 h, peaked at 12–24 h, and gradually decreased beginning 96 h after nasal administration (Fig. 4 B). In contrast, CD11b, B220, and CD4+ cells did not express GFP in spleen (data not shown).
To directly confirm GFP expression in intestinal tissues, we next analyzed large intestines for fluorescence expression using immunochemical analysis. GFP+ cells appeared in large intestines from 6 h to 4 days after nasal administration of GFP plasmid (Fig. 4,C). High magnification analysis revealed that GFP+ cells were located in nearby lymphoid vessels (Fig. 4 C, right). Most GFP+ cells in the large intestines were CD11c+ (data not shown). These findings directly demonstrate that nasal delivery of naked DNA resulted in the expression of the corresponding protein in a distant mucosal compartment (e.g., large intestine).
Nasal IL-12p70 DNA led to the protein expression in large intestine
To directly demonstrate that nasal IL-12p70 DNA administration induces IL-12p70 expression at the site of disease development, we used immunohistochemical analysis to assess the cytokine expression in the large intestine of mice nasally treated with IL-12p70 DNA. When large intestines were examined, IL-12p70-producing cells were located in nearby lymph vessels (Fig. 5,A). In contrast, large intestinal cells of mice treated with vector DNA only did not contain any cells expressing IL-12p70 (Fig. 5,A). Further analysis of the IL-12p70 expressed by large intestinal mononuclear cells revealed that these cells were costained with green fluorescence-coupled anti-CD11c (Fig. 5,C). These data suggest that nasal administration of IL-12p70 DNA resulted in IL-12p70 protein expression in large intestinal DCs. An identical finding was also observed using nasal GFP DNA treatment (Fig. 4). Furthermore, to finally confirm that IL-12p70 is expressed in mice treated nasally with IL-12p70 DNA, we examined IL-12p70 protein expression by Western blot analysis. Extracts of large intestine from mice nasally treated with IL-12p70 DNA expressed a high intensity band corresponding to IL-12 (Fig. 5 B). However, only a faint band corresponding to residual IL-12p70 was detected in large intestinal extracts from mice treated with the empty vector. These results indicate that nasal IL-12p70 DNA treatment leads to IL-12p70 expression in large intestinal cells including DCs, accounting for the inhibition of pathologic Th2 reactions and thus for the prevention of allergic diarrhea.
Nasal IL-12p70 DNA administration cures ongoing OVA-induced allergic diarrhea
Once we had established that nasal IL-12p70 DNA prevented allergic diarrhea, we examined whether nasal IL-12p70 DNA could alter the disease condition in mice with existing allergic diarrhea. Severe allergic diarrhea was first induced in systemically primed mice by 10 oral challenges with OVA. When mice were given nasal IL-12p70 DNA, the diarrhea was completely cured after three doses (Fig. 6, A and B). As one might expect based on our previous findings (4), a large number of basophils and eosinophils were seen in the large intestine of mice suffering from allergic diarrhea (Fig. 6,C, top). The nasal treatment with IL-12p70 DNA removed these basophils and eosinophils from the large intestine (Fig. 6,C, bottom). Furthermore, the large intestinal hyperresponse of Th2-associated IgA, IgG, and IgG1 AFCs were also significantly decreased by the nasal IL-12p70 DNA treatment (Fig. 6 D). Thus, nasal administration with IL-12p70 DNA proved to be effective at inhibiting ongoing large intestinal allergic reaction.
Food allergies in humans are caused by hypersensitivity to food allergens and can result in severe diarrhea (30). The OVA-induced Th2 cell-mediated allergic diarrhea model, which involves systemic priming followed by oral challenge, is a useful and adequate model for the investigation of mechanisms that result in food allergies (4, 7). Our most recent study showed that local accumulations of IL-12p40 are a major triggering factor for the creation of an aberrant Th2 environment in the large intestine, one which is conducive to the induction of allergic diarrhea (7). In the current study, our findings demonstrate that nasal IL-12p70 DNA treatment prevents the development of IL-12p40-mediated OVA-induced Th2-dominant allergic disease by inducing IL-12p70 production in the large intestinal tract. Not surprisingly, the inhibition of pathologic Th2 cell responses was seen in the large intestine of mice nasally treated with IL-12p70 DNA. Based on the results obtained by in vivo image and immunohistochemical analyses, we suggested that nasal deposition of naked IL-12p70 DNA resulted in the expression of the corresponding protein at distant sites including large intestine. This finding was further substantiated by results showing that nasal administration of naked GFP plasmid DNA resulted in the expression of GFP-positive DCs but not of macrophages and B cells in the intestinal tract, NALT, and spleen. Although the efficacy of the nasal administration of naked DNA was previously reported using Flt3 ligand and TGF-β DNA plasmid (23, 25), our study is the first to show that expression of nasal DNA occurs in distant mucosal compartments at the single cell level (e.g., IL-12p70-producing large intestinal DCs) and to link that expression to the prevention and treatment of intestinal allergic diseases. Taken together, these data show that noninvasive nasal administration with naked DNA plasmid results in the expression of the corresponding protein in both the mucosal and systemic lymphoid tissues and so may offer a new avenue of therapy for mucosa-associated diseases. Specifically, nasal administration of naked IL-12p70 DNA should be given consideration as a new tool in preventing and treating allergic diseases associated with the distant mucosal tissues.
The IL-12 protein treatment has been shown to be effective in inhibiting allergic asthma-associated airway hyperreaction and in blocking the associated eosinophil accumulation in the lung and the elevation of allergen-specific IgE (19, 20, 31). Oral IL-12 treatment inhibited peanut allergic reactions by reducing the release of histamine and peanut-specific serum IgE and IgG1 levels (32). These results suggest that IL-12 might be a useful immunotherapeutic agent for the control of respiratory mucosa-associated allergic diseases because these clinical symptoms have been shown to result from the development of aberrant Th2-type responses (4, 7). It is also reported that the IL-12 gene therapy using cationic liposome or virus as the DNA delivery vehicle is effective in controlling allergic asthma (22, 33, 34). Thus, the i.v. injection of IL-12p70 DNA plasmid mixed with liposome achieved high protein expression in the lung (35, 36). Systemic IL-12 gene therapy resulted in the down-regulation of airway inflammation by suppressing the secretion of eotaxin in the lung tissue (33). When the effects of systemic IL-12p70 DNA and protein administration were compared, the half-life of the circulating IL-12p70 protein produced was much longer after gene transfer than after IL-12 protein injection, and no negative side effects were seen (21). Other groups have reported that local intratracheal or nasal administration of the IL-12p70 gene prevented the development of respiratory allergic disease (22, 34). These elegant studies demonstrated the efficacy of the recombinant vaccinia virus vectors designed to deliver the IL-12-encoding gene to respiratory tissues for the treatment of allergic airway disease (34). The vaccinia virus vector system allowed the restricted expression of IL-12p70 locally in the airway but not systemically. Our present study further demonstrates the attractiveness of mucosal gene therapy for the prevention and treatment of mucosa-associated immunological diseases. The current study adds a new dimension to the efficacy of the nasal delivery of IL-12p70-specific DNA by showing that it allows the expression of the corresponding protein at distant mucosal sites, presumably via APCs, namely DCs. Considering possible application of our findings to clinical setting, one must realize a fact that murine IL-12p40 homodimer has been shown to bind to IL-12R with high affinity (37), whereas human IL-12p40 homodimer possesses a somewhat weaker binding affinity than mouse IL-12p40 (13).
We could not detect the elevation of IL-12p70 levels in serum (data not shown), but the presence of IL-12p70-positive DCs were noted in the large intestine and spleen. This finding strongly suggests a possibility that the selective expression of IL-12p70 by DCs may lead to the effective delivery of the protein perhaps made in limited quantifies to Ag-specific CD4+ T cells via the cognate cell-to-cell interaction, and thus obviating elevated serum levels to elicit therapeutic response as was seen in serum isolated from nasally treated mice. This result is of specific significant in light of well-documented systemic toxicities associated with parentally administrated IL-12. Thus, it is well known that systemic injection of IL-12p70 protein shows efficacy in suppressing tumors in both mice and humans (38, 39, 40). However, the results of the initial human clinical trials using systemic treatment with rIL-12 protein were discouraging due to dose-dependent toxicity (41, 42). To overcome the obstacle posed by such toxicity, the IL-12p70 DNA was substituted for the IL-12 protein in a murine study focused on the suppression of tumors (43). The study showed that the IL-12 gene therapy proved to be as efficient as the IL-12p70 protein therapy, while inducing far fewer toxic side effects such as weight loss, splenomegaly, and elevated IFN-γ levels in serum. Similarly, the mice treated nasally with IL-12p70 in the murine intestinal allergy model showed no such side effects in this study (data not shown). It has been shown that systemic administration of IL-12p70 plasmid DNA had therapeutic effects against murine tumors (44, 45, 46). The antitumor activity induced by administration of naked IL-12p70 DNA was associated with the augmentation of tumor-specific CTLs (46) or the prevention of tumor angiogenesis (45). In terms of clinical applications for the control of human diseases including cancer and allergies, IL-12p70 DNA treatment may be better than IL-12p70 protein treatment in that it avoids unnecessary toxic side effects while remaining fully capable of controlling disease.
In this study, we were able to show the dynamic chronologic expression and migration of the nasally introduced naked DNA from at the site of introduction to the distant large intestinal tract. Nasal deposition of the naked IL-12p70 or GFP DNA resulted in the expression of the gene in NALT DCs. Subsequently, DCs specifically expressing the corresponding protein were also noted in CLN, spleen, and the large intestine. Using the IVIS, we noted steady accumulation of green fluorescence in the region associated with the nasal cavity, including the CLN and distant spleen following the nasal naked GFP DNA. Another recent report used IVIS to show that nasal administration of streptococci transfected with a bioluminescent gene induced bioluminescent signals in nasal tissues, particularly in NALT, and that these signals were observed very early and peaked 1 h following nasal treatment (28). It was also demonstrated that M cells located in the NALT epithelium were the primary entry point for streptococci invasion and that luminescence was subsequently observed in the systemic tissues such as the spleen and lymph nodes of the nasally treated mice (28). To this end, our results also showed the early expression of IL-12p70 or GFP gene-expressing DCs in NALT after nasal delivery, though the level of expression was low. In another report, i.v. and intratracheal IFN-γ gene deliveries were examined in mice with allergen-induced airway hyperresponsiveness (47). Although both routes of gene delivery resulted in the expression of IFN-γ, the former route was much more effective in inducing protein expression in the lung.
In our study, nasal deposition of expression plasmid DNA resulted in more protein expression in spleen and intestine than lung. The expression of IL-12p70 in both spleen and intestine may account for the effectiveness of the nasal IL-12p70 treatment. We thus demonstrated previously that the systemic priming in spleen was essential for the development of large intestinal allergic diarrhea following oral challenge (4). Furthermore, it was also shown that systemically primed splenic OVA-specific CD4+ T cells preferentially migrated into the large intestine (48). These findings suggest that some of the pathologic Ag-specific CD4+ T cells originate from the spleen and thus the expression of IL-12p70 in the spleen may provide an opportunity to alter Th2 dominance. Of course, the high expression of IL-12p70 at the site of Th2 hyperresponse will lead to the inhibition of IL-12p40-mediated pathologic Ag-specific CD4+ T cell induction in the large intestine. Although our current findings suggest that nasal gene therapy with IL-12p70 is effective in the prevention of IL-12p40-mediated and Th2 cell-mediated allergic diarrhea, one cannot exclude a possibility that swallowed DNA due to possible spill over of nasally administrated IL-12p70 to esophagus may contribute the expression of the corresponding protein in the intestine. Thus, our ongoing experiments also aim to assess the efficacy of orally administered IL-12p70 DNA in controlling gastrointestinal allergic diseases.
Inasmuch as our recent and separate study showed that the formation of IL-12p40 over IL-12p70 is a major pathologic factor in the creation of the dominant Th2 environment in the large intestine, which is the most conducive to the development of allergic diarrhea (7), an obvious approach was to use the IL-12p35 gene in an attempt to alter the dominant p40 to p70 formation in mice suffering from intestinal allergies. For IL-12p70 protein expression, it is necessary to have the same cell expression in both IL-12p40 and IL-12p35 (10). Thus, it was logical to also test the feasibility of IL-12p35 DNA nasal administration. When the p35 DNA was nasally administered, it did not prevent the development of allergic diarrhea (data not shown). Furthermore, we were unable to find any increase in IL-12p70 expression in any of tissues tested in the p35-treated mice. Although we cannot explain why the p35 treatment failed, it is possible that most of the large intestinal DCs obtained from allergic diarrhea-afflicted mice were already programmed to form the homodimeric p40 and thus cannot be altered even via the exogenous deposition of the p35 gene. In contrast, the delivery of the intact p70 gene may effectively induce naive DCs to express the heterodimeric gene and subsequently to synthesize protein. Furthermore, the introduction of the IL-12p70 gene forcefully induces protein expression even by DCs already committed to the expression of p40 homodimers.
We thank Drs. J. R. McGhee and P. L. Perera for kind and helpful suggestions and T. Enokida for encouragement. We also thank Drs. C. Kai and K. Kohara, Animal Research Center, Institute of Medical Science, University of Tokyo (Tokyo, Japan), and Drs. S. Watanabe and A. Namiki, SC Bioscience (Tokyo, Japan), for help in using IVIS analysis.
The authors have no financial conflict of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
This work was supported in part by the Core Research for Evolutional Science and Technology (CREST) Program, Japan Science and Technology Corporation and a Grant-in-Aid from the Ministry of Education, Science, Sports and Culture and the Ministry of Health and Welfare of Japan. It is also supported by U.S. Public Health Service Grants DK 44240 and DE 12242.
Abbreviations used in this paper: DC, dendritic cell; Flt3, Fms-like tyrosine kinase-3; AFC, Ab-forming cell; DAPI, 4′,6′-diamidino-2-phenylindole; NALT, nasopharynx-associated lymphoid tissue; CLN, cervical lymph node; MLN, mesenteric lymph node; IVIS, in vivo imaging system.