INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Keywords: common  chain; antisense oligonucleotides; eosinophilia; hyperresponsiveness; rats

The cytokines interleukin (IL)-3, IL-5, and granulocyte macrophage colony-stimulating factor (GM-CSF) are important regulators of hematopoiesis and of inflammation. All three cytokines are involved in eosinophil, mast cell, and macrophage activation and survival (1-3). There is accumulating evidence that implicates these cytokines in the pathophysiology of a number of allergic diseases, such as asthma, allergic rhinitis, and atopic dermatitis (4, 5). These cytokines promote airway eosinophilia, and prime and activate eosinophils to release proinflammatory cytoplasmic granule products, lipid mediators, and cytokines that are thought to contribute to the tissue damage, remodeling, and hyperresponsiveness of the asthmatic airways (6). Until now, inhibitors of IL-5 have received most of the initial attention in allergic inflammation. Indeed, several strategies have been developed to block the effects of IL-5 on eosinophils. These include mainly neutralizing anti-IL-5 antibodies (Ab) (7) and antisense oligonucleotides directed against IL-5 and the alpha subunit of its receptor (8, 9). However, recent studies performed in rats and in humans suggest that inhibition of IL-5 alone is not sufficient to control all the inflammatory and physiologic changes that are encountered in asthma (7, 10). Because both GM-CSF and IL-3 are also released at sites of allergic inflammation (4, 11), it may be that inhibition of all three mediators would result in a more important inhibition of atopic inflammation.

IL-3, IL-5, and GM-CSF stimulate eosinophils and other cells by binding to cell surface receptors that comprise a ligand-recruiting  chain, which is specific for each cytokine, and a signal-transducing  subunit (_c), which is shared by all three cytokines (12, 13). The  chains can bind their ligands with low affinity, whereas the _c chain does not bind ligand itself but, when complexed with an  chain, forms a high-affinity signaling-competent receptor (14). The engagement of _c leads to the activation of JAK-2, STAT-5, and other signaling molecules (15), culminating in the full plethora of cellular activities commonly associated with IL-3, IL-5, GM-CSF stimulation (16).

The importance of the _c for IL-3, IL-5, and GM-CSF function was demonstrated by gene-targeting studies. Eosinophil numbers were reduced in the _c mutant mice, a phenomenon accompanied by the lack of an eosinophilic response to parasites, and IL-5 and GM-CSF failed to stimulate colony formation in clonal cultures of bone marrow (BM) cells (17). Monoclonal antibody (mAb), BION-1, raised against the isolated membrane proximal domain of _c blocked eosinophil production, survival, and activation stimulated by IL-5 as well as by GM-CSF and IL-3 on purified human eosinophils (18). Furthermore, a Lyn-binding peptide inhibitor associated with the common _c blocked eosinophil differentiation, survival, and airway eosinophilic inflammation in a murine model of eosinophilia (15). Together, these studies clearly suggest that _c is important in allergic inflammation with the added advantage of potentially allowing antagonism of all three proinflammatory cytokines.

Antisense oligodeoxynucleotides (AS-ODN) are now commonly used as a selective strategy to inhibit the expression of a variety of genes (19). The lung represents an ideal organ to target with AS-ODN, given that its large surface area and the surfactant that is present could facilitate the cellular absorption of ODNs. Aerosol or intratracheal administration could avoid the potential toxicity of systemic administration. We have assessed whether a topical antisense oligonucleotide approach could inhibit common _c of IL-3, IL-5, and GM-CSF receptors within the lungs. Phosphorothioate antisense ODNs that support ribonuclease H-mediated degradation of the common _c messenger RNA (mRNA) were tested in the ovalbumin (OVA)-sensitized and challenged Brown Norway (BN) rat model. We report that a rat common _c antisense oligonucleotide reduces IL-3, IL-5, GM-CSF mRNA and protein expression and effectively inhibits antigen-induced eosinophil infiltration and airway hyperresponsiveness in vivo in a manner that is consistent with an antisense mechanism of action.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Synthesis and Design of ODNs

On the basis of the reported rat IL-3 receptor  subunit mRNA sequence (20), a phosphorothioate antisense ODN, corresponding to nucleotide sequence 143-161 of the coding region, a sense and a mismatch ODN (Table 1) were designed and synthesized.

Rat Bone Marrow Cell Culture

In vitro liquid culture was performed as described elsewhere (21). OVA-sensitized BN rats were killed under general anesthesia and femurs were removed and flushed with saline. The BM cells (5 × 10⁵ cells/ml) were suspended in serum-free RPMI alone (control) or with antisense 143 (AS143), sense 143 (SS143), or mismatch 143 (MM143) ODNs (10 µM) for eight hours.

Animals, Sensitization, and Study Protocol

Active sensitization of male BN rats was performed by subcutaneous injection of 1 ml of saline containing 1 mg of chicken egg OVA and 3.5 mg of aluminum hydroxide gel.

To measure the inflammatory response after antigen challenge, eight sensitized rats received a 500 µg intraperitoneal injection of AS143-ODN the evening before the experiments, followed by a 200 µg intratracheal injection of AS143 before OVA challenge, and eight hours later bronchoalveolar lavage (BAL) was performed as previously described (22). Another group of rats received a single dose of AS143 (200 µg, n = 8) or MM143 (200 µg, n = 8) intratracheally before OVA challenge, and 15 hours later BAL was performed. Each group had its own control animals (n = 8) that received saline instead of AS143.

The assessment of the effects of _c inhibition on airway responsiveness to leukotriene D₄ (LTD₄) was performed in separate groups of rats. Sensitized rats received a single dose of AS143 (50, 100, and 200 µg) or MM143 (200 µg) intratracheally before OVA challenge, and 15 hours later LTD₄ challenge was performed. A control group of rats was treated with saline. The equipment and methodology for measuring pulmonary resistance (RL) was as previously described (22). Rats were exposed to incremental doses of LTD₄ (50 ng to 1,000 ng in 50 µl of saline) intratracheally until RL underwent a doubling from the baseline value.

At the completion of the study, all animals were exsanguinated and the lungs were prepared as previously described (23).

RNA Preparation and Semiquantitative Reverse Transcriptase/Polymerase Chain Reaction

Total RNA from frozen biopsies or harvested cells was isolated as previously described (23). Polymerase chain reaction was performed using specific primers (Table 2). Preliminary experiments determined the optimal number of cycles for each primer, which were as follows: glyceraldehyde-3- phosphate dehydrogenase (G3PDH, a housekeeping gene) 21 cycles; common  chain, 25 cycles; IL-5R, 35 cycles.

Immunohistochemistry for Eosinophils and Common _c

Immunocytochemistry was performed as previously described (23) using anti-human major basic protein (MBP; BMK-13; Biodesign International, SACO, ME) mAb or anti-IL-3/IL-5/GM-CSF receptors  chain polyclonal Ab (Santa Cruz Biotechnology, Inc., Santa Cruz, CA).

Statistical Analysis

Differences between groups were determined by analysis of variance (ANOVA) (Tukey-Kramer multiple comparisons test) and the Student's t test. For inhomogenous variances, Mann-Whitney test and Kruskal-Wallis nonparametric ANOVA were used. All results are expressed as mean values for the group of animals analyzed ± SEM. Statistical significance was claimed when p values were less than 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In Vitro Characterization of a Rat Common _c Antisense Oligonucleotide

To identify an antisense oligonucleotide against rat _c, we used an in vitro screening method on rat BM cells. IL-3, IL-5, and GM-CSF stimulate various lineages of hematopoietic cells (24) and their receptors are expressed on BM cells. A lead 18-mer AS-ODN complementary to a 5'-coding region of IL-3R mRNA was identified. This ODN was phosphorothioate modified to increase its resistance to nuclease degradation. As shown in Figure 1, the IL-3R-specific AS-ODN was found to decrease IL-3R mRNA expression in BM cells treated for eight hours with AS143 (10 µM). This inhibition is specific to the antisense ODN and not due to RNA destruction or to a loss of cell viability as demonstrated by the presence of the 450-bp product corresponding to G3PDH mRNA. To further analyze the specificity of the IL-3R AS-ODN, mismatches were introduced into the selected ODN and the effect on _c mRNA expression was analyzed. The AS143-MM completely lost its inhibitory effect on the expression of _c in BM cells. The sense ODN did not have any effect on _c mRNA expression as well. Thus, this inhibition was antisense-specific, as no effect was seen on expression of _c mRNA with AS143-MM or sense ODN.

View larger version (49K): [in this window] [in a new window]   Figure 1.   In vitro characterization of the common c AS-ODN in rat bone marrow (BM) cells. BM cells were incubated in serum-free RPMI alone (control) or in the presence of AS-143 (10 µM), sense ODN (SS, 10 µM), or mismatch ODN (MM, 10 µM) for 8 h. Total RNA was extracted from rat BM cells, and the presence of c mRNA expression was assessed by RT-PCR of the complementary DNA (cDNA). G3PDH was used as a housekeeping gene.

Reduction of Lung _c mRNA and Protein Expression after Antisense Oligonucleotide Treatment

Using semiquantitative RT-PCR, we assessed the expression of _c mRNA in RNA extracted from the lungs of AS-ODN, mismatch-ODN-treated and control rats. _c mRNA was expressed significantly less in the lungs of AS-treated rats (Figure 2A; **p < 0.003) eight hours after challenge than in mismatch or control rats. There was no difference in the expression of _c mRNA expression between the lungs of mismatch-treated and control rats (Figure 2B). We also found expression of _c mRNA in the lungs of control OVA-sensitized, but nonchallenged rats. Expression of _c mRNA in control nonchallenged rats did not differ from control challenged animals (87% ± 12% as a percentage of G3PDH density in control challenged rats, n = 8 versus 89% ± 10% as a percentage of G3PDH density in control nonchallenged rats, n = 6). Because several lines of evidence implicate IL-5 as the central cytokine for producing tissue eosinophilia, it was of interest to determine whether AS treatment would affect the IL-5R mRNA expression. The AS treatment had no effect on  chain mRNA expression in both groups of animals (59% ± 24% as a percentage of G3PDH density in controls versus 53% ± 16% as a percentage of G3PDH density in AS143-treated rats, n = 8 rats per group).

View larger version (45K): [in this window] [in a new window]   View larger version (18K): [in this window] [in a new window]   Figure 2.   Semiquantitative RT-PCR assessment of c mRNA expression in the lungs of control, AS, and mismatch (MM)-treated BN rats. (A) Autoradiographs demonstrate the expression of c mRNA in control, AS143, and MM-treated rats. G3PDH was used for standardization. (B) Semiquantitative comparison of integrated density values of the radioactive signal of specific bands for c mRNA as a percent of the G3PDH-specific signal from the lungs of control, AS143, and MM-treated BN rats. AS143 significantly inhibited the expression of lung c mRNA when compared with control and mismatch-treated rats (**p < 0.003).

To determine whether the decreased RNA levels also affected _c translation, we then studied protein expression by immunostaining for the common _c. Airways from AS-treated rats showed significantly decreased numbers of _c-immunoreactive cells in the subepithelial region (Figures 3A and 3B, AS-treated rats: 3.15 ± 1 positive cells/mm basal membrane, n = 8; control rats: 19.04 ± 4.4 positive cells/mm basal membrane, n = 8; **p < 0.003).

View larger version (117K): [in this window] [in a new window]   View larger version (12K): [in this window] [in a new window]   Figure 3.   Immunohistochemical analysis of c-positive cells in lung tissue sections. (A) Expression and cellular distribution of c protein in lung tissue sections from AS143-treated (a, b) and control (c, d ) rats after antigen challenge. Sections shown are from a representative experiment viewed at ×20 (a, c) and ×40 (b, d ) magnifications. (B) The numbers of c-positive cells per mm of basal membrane (BM) in AS-treated and control rats. Individual data points are shown along with group mean values (horizontal lines). The differences between AS-treated and control rats are significant (**p < 0.003).

Effect of _c AS-ODN on Antigen-induced Eosinophilia

To analyze the effect of the ODN on cellular recruitment in vivo, intratracheal administration of AS or mismatch ODNs was performed 14 days after OVA sensitization and before antigen challenge in BN rats. Lung eosinophils were measured in BAL performed eight hours after OVA challenge and were found to be reduced from 0.5 ± 0.08 × 10⁶ cells in OVA-sensitized and challenged control rats to 0.17 ± 0.03 × 10⁶ cells (**p < 0.003) in OVA-sensitized and challenged rats that received AS-ODN (500 µg intraperitoneally + 200 µg intratracheally; Figure 4A). No significant effect was found on the total number of other leukocyte subpopulations that was retrieved in BAL.

View larger version (24K): [in this window] [in a new window]   View larger version (27K): [in this window] [in a new window]   View larger version (26K): [in this window] [in a new window]   Figure 4.   Effect of AS143-ODN or mismatch-ODN treatment on antigen-induced cellular influx. OVA-sensitized BN rats were treated with saline (control rats; hatched bars), antisense or mismatch ODNs (empty bars) and 10 min after were challenged with 200 µg of OVA (in 50 µl of normal saline solution). Differential cell counts from the BAL (8 h after challenge) of (A) BN rats treated with 500 µg intraperitoneally + 200 µg intratracheally of AS-143 ODN, (B) BN rats treated with a single dose of 200 µg intratracheally of AS-143 ODN (BAL was performed 15 h later). (C ) BN rats treated with a mismatch ODN (200 µg, BAL was performed 15 h later). *p < 0.01; **p < 0.003.

Experiments were also performed on rats that received a single dose (200 µg intratracheally) of AS-ODN 10 minutes before OVA challenge. A significant inhibition of lung eosinophilia occurred also in BAL 15 hours after OVA challenge (from 1.6 ± 0.38 × 10⁶ eosinophils in control OVA-challenged rats to 0.47 ± 0.1 × 10⁶ cells in AS-treated rats; *p < 0.01; Figure 4B). Treatment with AS143 resulted in a reduction of both the relative and total numbers of eosinophils present in the lungs after antigen challenge without any significant effect on the total number of other leukocyte subpopulations. Administration of the same amount of mismatch ODN before OVA challenge did not affect lung BAL eosinophils when compared with controls, indicating that the effect of the AS-ODN is sequence-specific and consistent with an antisense mode of action (Figure 4C).

To assess whether the changes that we found in BAL were also found in lung tissue, we performed immunohistochemical analyses on lung sections that were obtained from tissue fixed immediately after BAL. Immunohistochemical analyses of lung sections from control and AS-treated rats with a mAb against eosinophil granule MBP revealed distinct differences between two groups of rats as well. Lungs from control rats contained significantly more eosinophils eight hours after OVA challenge than AS-treated rats, as shown in Figure 5 (AS-treated rats: 6.9 ± 0.7 positive cells/mm basal membrane, n = 8, versus control rats: 17.7 ± 1.2 positive cells/mm basal membrane, n = 8; **p < 0.001).

View larger version (11K): [in this window] [in a new window]   Figure 5.   Immunohistochemical analysis of MBP-positive cells in lung tissue sections of AS143-ODN-treated and control rats. Results are expressed as the number of positive cells per mm of basal membrane (BM). Means for each group are indicated with horizontal lines. The differences in the numbers of MBP-positive cells between AS143-treated and control animals are significant (**p < 0.001).

Effect of _c AS-ODN on Antigen-induced Airway Hyperresponsiveness

Allergen-induced cellular influx into the lungs has been correlated with airway hyperresponsiveness in humans (25), although this correlation is not always found in animal models of asthma (26). It is thus possible that blocking the cellular influx into the lungs affected the increased airway responsiveness that occurs in BN rats after antigen challenge. To determine whether AS143-ODN could affect airway responses to LTD₄ (27), we investigated the effects of a single dose of AS143-ODN (200 µg) on the cumulative dose-response curve to LTD₄, 15 hours after OVA challenge (Figure 6A). The baseline value for RL was first measured, followed by the responses to intratracheal instillation of increasing amounts of LTD₄. As shown in Figure 6A, pretreatment with AS143 at the time of OVA challenge increased the effective concentration causing a doubling in pulmonary resistance to LTD₄ (EC₂₀₀ LTD₄) from 113 ± 31 ng in control, OVA-sensitized and challenged rats to 328 ± 49 ng in AS-treated, OVA-sensitized and challenged rats (**p < 0.001). The AS143-MM ODN at a concentration of 200 µg had no effect on LTD₄ responsiveness (106 ± 14 ng in mismatch-treated rats). The airway hyperresponsiveness to LTD₄ was also tested in sensitized rats that received 50 µg or 100 µg of AS143-treatment, 10 minutes before OVA challenge, and the results show that AS143 reduced airway hyperresponsiveness to LTD₄ in a dose- dependent manner (Figure 6B). Collectively, these results show that IL-3, IL-5, and GM-CSF contribute to airway hyperresponsiveness in the BN rat and that antisense blocking of _c in vivo results in a significant reduction of airway hyperresponsiveness to LTD₄.

View larger version (14K): [in this window] [in a new window]   View larger version (12K): [in this window] [in a new window]   Figure 6.   Effect of the c AS-ODN on antigen-induced airway hyperresponsiveness. (A) On Day 14 postsensitization, BN rats were anesthetized, were either nonchallenged or received saline intratracheally, AS143-ODN intratracheally, or AS143 mismatch intratracheally and challenged with OVA intratracheally; 15 h later, they were reintubated and baseline RL was measured. Rats were exposed to incremental doses of LTD4 until RL underwent a doubling from the baseline value. **p < 0.001 in AS143-ODN-treated versus control animals or in AS143-ODN- versus mismatch-treated rats. (B) Rats (n = 6 to 9 per group, circles) were treated with different concentrations of AS143-ODN (50 µg, 100 µg, and 200 µg) intratracheally before antigen challenge, and LTD4 challenge was performed 15 h after OVA challenge; squares are results for control animals.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In this study we demonstrate that specific blockade of the common _c of IL-3, IL-5, GM-CSF receptors by AS-ODN is possible with topical antisense technology and leads to a decrease in lung eosinophilia and airway hyperresponsiveness that occurs after antigen challenge. There thus seems to be an important role for _c in modulation of the in vivo biologic activities of IL-3, IL-5, and GM-CSF. There is also a potential therapeutic utility of _c AS-ODN as a novel molecular approach for the treatment of patients with asthma and allergic disorders.

IL-3, IL-5, and GM-CSF are produced by activated T cells and mast cells. These cytokines play important roles in hematopoiesis and in immune and inflammatory reactions (28). Experiments aimed at investigating the in vivo function of IL-3, IL-5, GM-CSF using a gene knockout approach suggest that these cytokines are central for the accelerated production of granulocytes, macrophages, and eosinophils in inflammation and infection (29, 30). Both IL-3 and GM-CSF stimulate various lineages of hematopoietic cells (24), whereas IL-5 is mainly an eosinophil lineage-specific factor (24, 31). Of the three cytokines that stimulate cells through _c, IL-5 has received most of the initial attention because of its role in causing eosinophil accumulation at the site of allergic inflammation. The importance of IL-5 has been inferred, in part, from studies describing increased IL-5 mRNA expression and increased amounts of this cytokine in BAL fluid of asthmatic patients after allergen inhalation challenge (32). Recent studies also showed that an anti-IL-5 mAb completely blocked eosinophilic airway infiltration in guinea-pig, mouse, and monkey models of asthma (33-35). These observations have led to great interest in the effects of anti-IL-5 mAbs in human asthma. A recent clinical trial performed in patients with asthma has shown that although an IL-5 blocking mAb prevented blood and lung eosinophilia, it had little effect on the late airway asthmatic response or on airway hyperresponsiveness in these patients (7). This study was predicated on the hypothesis that eosinophils were the cause of the symptoms and physiologic changes that were monitored in the studies. It was not designed to assess whether inhibition of IL-5 release or its effects at the time of antigen challenge affected the late airway response. We have recently assessed what effect administration of IL-5 would have on the airways of BN rats (10). Although IL-5 increased airway responsiveness 20 hours after intratracheal administration, it had no effect on the late airway response after antigen challenge. These results suggest that targeting IL-5 alone in patients with asthma is not sufficient to decrease the symptoms of asthma. Because IL-3 and GM-CSF are also increased in the airways of patients with asthma, strategies such as the concomitant antagonism of all three proinflammatory cytokines may more profoundly downregulate allergic inflammation and airway hyperresponsiveness.

In this respect, targeting the common _c of IL-3, IL-5, and GM-CSF receptors is of interest because it is a specific approach, since only cells expressing the receptor will be affected, and even with overproduction of ligands, cell responses could thus be inhibited. Inhibition of the common _c of the human IL-3, IL-5, and GM-CSF receptors with mAbs was able to decrease eosinophil colony formation by IL-3, IL-5, and GM-CSF in vitro (18). We have found that AS-ODNs against the common _c of IL-3, IL-5, and GM-CSF receptors significantly inhibited proliferation of the erythroleukemia cell line TF-1 and eosinophil survival in vitro whether the culture media contained IL-3, IL-5, or GM-CSF (36). Adachi and coworkers reported that a Lyn-binding peptide inhibitor inhibited eosinophil influx into antigen-challenged mice (15). This inhibitor does not affect the JAK-STAT pathway which is important in IL-3/IL-5/GM-CSF signaling (37, 38). We assessed whether inhibition of expression of the common chain of the IL-3/IL-5/GM-CSF receptors was possible in vivo and the effect of this inhibition on modulation of eosinophilic inflammation and airway hyperresponsiveness.

Experiments in OVA-sensitized and challenged BN rats demonstrate that pretreatment with oligonucleotides suppressed eosinophil recruitment within the airways. These results are consistent with several previous reports implicating the _c gene in eosinophil recruitment. In _c gene knockout experiments, _c null mice were shown to have a major reduction in eosinophil numbers in the peripheral blood and BM (17). Blood analysis from parasite-infected _c-deficient mutants showed no detectable eosinophilia in these animals (29). The absence of inhibition of eosinophil recruitment into the airways with a mismatch AS-ODN confirmed the specificity of the antisense ODN.

Because eosinophils (25) or _c, or both, may be involved in airway responsiveness, we examined whether _c inhibition affected the increased airway responsiveness to LTD₄ after antigen challenge. A single dose of AS-ODN directed against _c significantly decreased the airway responsiveness to LTD₄ that occurs after OVA challenge, and this inhibition was dose-dependent. Interestingly, experiments performed in guinea pigs have found that a high dose of anti-IL-5 mAb was able to block the antigen-induced airway hyperresponsiveness, whereas a lower dose blocked only pulmonary eosinophilia (33). Furthermore, only high and repeatedly introduced doses of AS-ODN directed against IL-5 caused a significant inhibition of antigen-mediated late-phase airway hyperresponsiveness in a murine model of asthma (9). These results emphasize the redundancy of immune responses and suggest that it may be more important to target several inflammatory mediators (_c for example) in order to affect all the changes that are encountered in allergy or asthma.

Topical administration of AS-ODN (intratracheally) results in effective cellular uptake into target cells and substantially reduces the required antisense dosage and potential toxicity when compared with systemic administration of AS-ODN. This is important especially when dealing with _c, which is a gene involved in hematopoiesis and for which significant toxicity has been shown after systemic _c neutralization (29). Although _c AS-ODN did inhibit the mRNA and protein expression of _c within the airways, complete abrogation of the gene was not observed in our system (_c mRNA expression was reduced by 60% after treatment with AS-ODN). It seems that partial inhibition of _c function in vivo is sufficient to affect the allergic inflammatory reaction within the airways and thus potentially avoid the side effects that are encountered with complete abrogation of _c.

In summary, we have shown that topical treatment of rats with a _c antisense oligonucleotide significantly inhibits gene expression and results in decreased eosinophil infiltration into the airways and a pronounced reduction in airway hyperresponsiveness. These data show an important role for _c in the modulation of allergic diseases.