Introduction

Top Abstract Introduction Material and Methods Results Discussion References

Eosinophils are terminally differentiated leukocytes which originate from myeloid precursors in the bone marrow but are not present in abundance in the peripheral blood under normal conditions. Although eosinophils appear to play a protective role during the host-defense response to parasitic infections, accumulating evidence implicates this cell in the pathophysiology of a number of allergic diseases, such as asthma, allergic rhinitis, and atopic dermatitis. Indeed, blood, bronchoalveloar lavage, and tissue eosinophilia is a characteristic abnormality in asthma, and eosinophil-derived proteins contribute to specific pathologic features of this disease (1, 2). Although several cytokines and chemokines have been shown to affect the maturation, tissue infiltration, degranulation, and survival of eosinophils, increasing evidence indicates a unique role for interleukin (IL)-5 in the regulation of this selective airway eosinophilia (3). Significantly elevated levels of IL-5 have been found in the peripheral blood and lung tissue compartment both in patients with asthma and in experimental models of asthmatic inflammation (4), and inhalation of IL-5 by asthmatics causes bronchial hyperreactivity and sputum eosinophilia (8).

The unique role of IL-5 in eosinophil production, activation, and localization is further supported by findings that mice overexpressing IL-5 develop a long-lasting and selective eosinophilia (9), whereas IL-5-deficient mice are unable to produce increased numbers of eosinophils in response to specific antigens (10, 11). Moreover, inhibition of IL-5 by neutralizing antibodies prevents eosinophil differentiation and infiltration of mature cells into inflamed tissues after antigen exposure (12, 13). In vitro, IL-5 not only induces the differentiation and expansion of eosinophil precursors in the bone marrow but also regulates many of the specialized function of mature eosinophils, such as adhesion, priming of both degranulation and chemotaxis, and the promotion of cell survival (14).

The effects of IL-5 are mediated through a heterodimeric receptor complex composed of a cytokine-binding, ligand-specific IL-5 receptor  chain (IL-5R) and a non-ligand binding, high-affinity receptor-forming and signal-transducing  subunit ( common) shared with the IL-3 and granulocyte macrophate colony-stimulating factor (GM-CSF) receptors (17). Both IL-3 and GM-CSF stimulate various lineages of hematopoietic cells (18), whereas IL-5, due to the restricted expression of its receptor on eosinophils and hematopoietically related basophils, is mainly an eosinophil lineage-specific factor (3, 18, 19). This is further supported by the finding that mice deficient in the IL-5R subunit required for binding of IL-5 are unable to respond to parasitic infections with increased numbers of eosinophils (20).

Together, these observations suggest that the development of drugs to neutralize the effect of IL-5 on eosinophil might represent a novel therapeutic approach in allergic diseases such as asthma. Indeed, clinical studies using neutralizing anti-IL-5 antibodies showed encouraging results in terms of effects on blood and tissue eosinophil numbers after allergen challenge (21). The purpose of the present study, therefore, was to examine whether a strategy developed at inhibiting expression of the IL-5R chain using antisense oligonucleotides (ASOs) could be of benefit in preventing IL-5-induced increases in bone-marrow eosinophil progenitors as well as tissue eosinophil accumulation.

	Material and Methods

Top Abstract Introduction Material and Methods Results Discussion References

Materials

All culture reagents were from Life Technologies (Paisley, UK). O-phenylenediamine, phorbol 12-myristate 13-acetate, 30% hydrogen peroxide, and bovine serum albumin (BSA) were obtained from Sigma Chemical Co. (Poole, UK). IL-5 was from R&D Systems (Abingdon, UK). Diff-Quik stain was from Baxter Dade AG (Düdingen, Switzerland).

BCL-1 Cell Cultures

The BCL-1 B lymphoma cell line was purchased from ATCC (Rockville, MD). BCL-1 cells were cultured in RPMI 1640 medium (RPMI) supplemented with 10% heat-inactivated fetal bovine serum (FBS) (Sigma Chemical Co., St. Louis, MO), 10 mM N-2-hydroxyethylpiperazine-N'-tetraacetic acid (pH 7.2), 50 µM 2-Mercaptoethanol, 2 mM L-glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin (GIBCO, Grand Island, NY).

ASO Synthesis and In Vitro Cell Transfection

2'-O-Methoxyethylribose-modified phosphorothioate oligonucleotides were synthesized on an automated DNA synthesizer (Applied Biosystems model 380B), as described previously (22). The chimeric oligonucleotides contain 2'-O-methoxyethyl (MOE)- modified residues flanking a 2'-deoxynucleotide/phosphorothioate region (gap) that supports ribonuclease (RNase) H activation (23). Oligonucleotides were analyzed by capillary gel electrophoresis and judged to be at least 85% full-length material. BCL-1 cells (1 × 10⁷ cells in phosphate-buffered saline [PBS]) were transfected with oligonucleotides by electroporation at 200 V, 1,000 µF, using a BTX Electro Cell Manipulator 600 (Genetronics, San Diego, CA).

Northern Blot Analysis

Total cellular RNA was isolated using the RNeasy kit (Qiagen, Santa Clarita, CA). Northern blotting was performed as previously described (22), using a complementary DNA probe prepared from BCL-1 cell RNA by standard reverse transcriptase/ polymerase chain reaction followed by a nested primer reaction. Signals were quantitated using a Molecular Dynamics PhosphorImager (Molecular Dynamics, Sunnyvale, CA).

Riboprobe Design and RNAse Protection Assay

RNAse protection experiments were conducted using Riboquant kits according to the manufacturer's instructions (Pharmingen, San Diego, CA). A custom riboprobe was designed to protect messenger RNA (mRNA) sequence corresponding to the distal half of exon 6, all of exons 7 and 8, and the proximal half of exon 9, and was purchased from Pharmingen. Signals were quantitated using a Molecular Dynamics PhosphorImager.

Western Blot Analysis

Western blotting was performed as described previously (22). Whole-cell extracts were prepared from BCL-1 cells and equal amounts of total protein separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis using 8% gels. Antibody to mouse IL-5R was purchased from Santa Cruz Biotechnology (Santa Cruz, CA) and used at 1:1,000 dilution for Western blotting.

Fetal Liver- and Bone Marrow-Derived Cultures

C57BL/6 mice (females, 18 to 25 g, 6 to 8 wk of age) were obtained from Harlan (Bicester, UK). Timed pregnancies were terminated at midgestation (embryonic day [E]16 to E17). Embryos were harvested. Single-cell suspensions of fetal liver were generated in 0.5 ml of RPMI supplemented with 10% FBS by five repeated passages of the minced livers through a 22-gauge needle and three subsequent passages through a 25-gauge needle. Clumps were removed by passage through 70-µm cell nylon mesh. Fetal liver single-cell suspensions were cultured at 10⁶cells/ml in RPMI supplemented with 2 mM glutamine, 10% FBS, 50 µM -mercaptoethanol, 100 IU/ml penicillin, and 100 µg/ml streptomycin in the presence of recombinant murine IL-5 (0.1 to 5 ng/ml) and oligonucleotides present in the culture medium from Day 0 (5 to 20 µM) for various periods of time. Cells were analyzed every second day for 12 d.

Excised femurs, cut at the epiphyses, were flushed with 0.5 ml RPMI. Single-cell bone-marrow suspension was obtained by gentle passage of the marrow through a 25-gauge needle into RPMI supplemented with 10% FBS. Cells were cultured in the media described for fetal liver cultures.

IL-5-Induced and Ragweed-Induced Peritonitis Models

For the IL-5-induced peritonitis model, Balb/c mice (females, 18 to 25 g, 6 to 8 wk of age) obtained from Harlan were treated daily for 7 d with 15 mg/kg intravenously of antisense, mismatch oligonucleotide, or vehicle (saline) together with daily dosing of IL-5 at 1 µg/mouse intraperitoneally for the last 5 d. At 24 h after the last dosing, animals were killed, peritoneal lavages were performed with 3 ml of PBS/0.4% BSA, and blood and femurs were collected. Total cell numbers, leukocyte differential counts, and eosinophil counts were performed as described later.

For the ragweed-induced peritonitis model, Balb/c mice (females, 18 to 25 g, 6 to 8 wk of age) obtained from Harlan were immunized subcutaneously with short ragweed antigen (Ag) extract (1/10,000 dilution; Greer Laboratories, Lenoir, NC) in 0.2 ml of saline containing penicillin/streptomycin (50 U/ml and 5 µg/ml) on Days 1 and 8. Sham-immunized mice received two injections of saline alone. At 7 d after the last immunization (Day 15), animals were challenged by intraperitoneal (i.p.) injection of 0.2 ml ragweed Ag extract. Saline- or ragweed-immunized control groups received an i.p. injection of 0.2 ml of saline. Oligonucleotides dissolved in saline were injected intravenously in the tail vein by bolus infusion at the indicated doses daily over a 3-d period starting the day before immunization (Day 0) until Day 2, daily from Days 7 to 9, and daily from Days 14 to 16. At 48 h after the Ag provocation (Day 17), animals were killed, peritoneal lavages were performed with 3 ml of PBS/0.4%BSA, and blood and femurs were collected. Total cell numbers, leukocyte differential counts, and eosinophil counts were performed as described later.

Determination of Total Cells, Leukocyte Differentials, and Eosinophil Counts

Cell viability was determined by trypan blue exclusion; cell numbers were determined by hemocytometry. Cytologic examination of peritoneal lavage cells and blood cells were done after cytocentrifugation and staining with Diff Quick. The relative proportions of the various leukocyte subpopulations were determined by a cell differential count of 400 cells. For the specific eosinophil quantitation in the bone marrow of treated mice and in fetal liver- and bone marrow-derived cultures, cytocentrifuge preparations were made and the eosinophil phenotype was assessed both by the morphologic features and by eosinophil peroxidase staining with O-phenylenediamine and counterstaining with thiazine.

Statistical Analysis

Statistical analysis was performed using the unpaired two-way Student's t test for all comparisons between two groups (24). Differences associated with probability values of P < 0.05 were considered significant.

	Results

Top Abstract Introduction Material and Methods Results Discussion References

Selection and Characterization of IL-5R-Specific ASOs

To identify an ASO against the murine IL-5R chain, we used as an in vitro screening model the murine B cell leukemia cell line BCL-1, which is known to express IL-5R- chain mRNA and protein and to proliferate in response to human and murine IL-5. A lead 20-mer ASO complementary to a sequence within the 3'-untranslated region (UTR) of the murine IL-5R gene was identified (Table 1). It is a chimeric oligonucleotide containing a uniform phosphorothioate backbone and a stretch of 10 2'-deoxy residues in the center of the molecule which supports RNAse H cleavage flanked by five bases at each of the 5' and 3' ends that are MOE-modified and thus convey greater resistance to exonuclease activities and higher affinity for hybridization to RNA (23, 25). As shown in Figure 1, the IL-5R-specific ASO was found to dose-dependently decrease IL-5R-chain mRNA expression in BCL-1 cells as measured by Northern blot. To further analyze the specificity of the IL-5R ASO, increasing numbers of base mismatches were introduced into the selected oligonucleotide and the effect on IL-5R mRNA expression was analyzed. The ASO-induced suppression of IL-5R mRNA was gradually lost by changing some of the bases within the specific sequence, with a complete loss of the inhibitory effect using a sequence containing three to five base mismatches (Table 1). Similar results were obtained by using a nonrelated ASO sequence specific for protein kinase (PK) C-, which again showed no inhibition of the expression of IL-5R mRNA (Figure 2a).

View larger version (53K): [in this window] [in a new window]   Figure 1.   In vitro characterization of IL-5R-chain ASO in murine BCL-1 cells. BCL-1 cells were transfected with antisense or control mismatch oligonucleotides by electroporation as described in MATERIALS AND METHODS, and RNA was isolated 24 h later. (a) Phosphorimage analysis of Northern blots of IL-5R- chain mRNA and control glyceraldehyde-3-phosphate dehydrogenase (G3PDH) levels after treatment with IL-5R ASO at different concentrations. (b-d) Analysis of IL-5R mRNA expression after treatment with one- (b), three- (c), and five- (d) base mismatch oligonucleotides. Loss of inhibitory activity coinciding with increasing number of base mismatches indicates a hybridization-based antisense mechanism of action.

View larger version (46K): [in this window] [in a new window]   Figure 2.   Effect of IL-5R ASO on IL-5R mRNA and protein expression of BCL-1 cells. (a) Northern blot of IL-5R mRNA from BCL-1 cells treated with IL-5R ASO or a nonrelevant control ASO directed against PKC-. The PKC- ASO was of the same length and chemical design as the IL-5R oligonucleotide. Cells were treated with oligonucleotide and RNA was isolated as noted in Figure 1a. (b) Kinetic analysis of IL-5R mRNA and protein expression as measured by Northern and Western blotting. RNA data shown are normalized to G3PDH levels. Protein data represent determinations performed on equal amounts of loaded protein (n = 2 per time point). (c) RPA analysis of BCL-1 cells after treatment with IL-5R ASO (10 µM) over 72 h. The top two bands represent membrane and soluble IL-5R mRNA, respectively. As expected, due to the targeting of sequence common to both the membrane and soluble isoforms, equal potency was observed for each of these transcripts. No effects were seen on expression of other constitutively expressed related cytokine receptor mRNA species.

To further analyze the selectivity of the IL-5R chain- specific ASO, we compared the effect of this oligonucleotide on the mRNA expression of other cytokines receptors using an RNAse protection assay (RPA). As shown in Figure 2c, the IL-5R-specific ASO selectively suppressed the expression of both the membrane and soluble forms of the IL-5R chain but did not affect the expression of other cytokine receptors, including the closely related  chain of the GM-CSF receptor.

Next, the onset and duration of IL-5R-chain mRNA and protein inhibition by specific ASOs were analyzed. For this purpose, cells were incubated with a concentration of 10 µM of the IL-5R-specific ASO and the cells were analyzed for their expression of IL-5R mRNA and protein at various time points over a 72-h period using RPA and Northern and Western blots. As shown in Figures 2b and 2c, the murine IL-5R-specific ASO almost completely inhibited the expression of IL-5R-specific mRNA within 20 h. This inhibition of the IL-5R mRNA was observed for up to 48 h, and after this time point the cells started to re-express new mRNA for this cytokine receptor. This inhibition is antisense-specific, as no effect was seen on expression of other constitutively expressed related cytokine receptor mRNA species. As expected, the analysis of the expression of the IL-5R-chain protein revealed a similar inhibition profile with a slight delay compared with the expression of the mRNA, reaching maximal levels after 48 h.

Effect of IL-5R ASO on IL-5-Induced Eosinopoiesis In Vitro

The results obtained so far clearly demonstrated that an IL-5R chain-specific ASO is able to selectively downregulate the expression of IL-5R chain mRNA and protein in a murine B-cell leukemia cell line. The next obvious question, therefore, was to analyze whether this oligonucleotide also inhibits IL-5-mediated effects on normal cells. To answer this question, we established an in vitro system to investigate the effect of the IL-5R-chain ASO on IL-5-induced differentiation and maturation of eosinophils from fetal liver or bone-marrow cells. As shown in Figure 3a, using single-cell suspensions of fetal livers, IL-5 dose-dependently and selectively stimulated the growth of eosinophil precursors and the differentiation into mature eosinophils. Optimal growth was achieved at a concentration of 2.5 ng/ml of IL-5 (Figure 3a). The kinetic analysis of IL-5-induced eosinopoiesis conducted over 12 d revealed that the total eosinophil number present in the culture increased in a time-dependent manner, reaching a plateau at Day 8 (Figure 3a). At this time most eosinophils present in the culture displayed morphologic characteristics of mature eosinophils, with very few eosinophil precursors still present (data not shown). Almost identical results were obtained when investigating IL-5-induced eosinopoiesis from bone marrow-derived cells (data not shown).

View larger version (21K): [in this window] [in a new window]   Figure 3.   Effect of IL-5R ASO on the IL-5-induced eosinopoesis from fetal liver-derived cells. (a) Time course of IL-5 (filled triangles, 0.1; filled squares, 0.5; filled circles, 2.5 ng/ml) effect on fetal liver-derived cultures. Total eosinophil numbers were characterized and quantitated as described in MATERIALS AND METHODS in cultures derived from four fetal livers. Data are expressed as means ± standard error of the mean (SEM). (b) Representative experiment of a kinetic analysis of IL-5-induced eosinopoiesis from fetal liver cells in the presence of various concentrations of IL-5R antisense (open squares) or mismatch (filled circles) oligonucleotides. (c) Dose-dependent inhibition of IL-5-induced eosinopoiesis of fetal liver cells in the presence of IL-5R antisense (open bars) or mismatch (filled bars) oligonucleotides after 10 d (mean values ± SEM from three independent experiments). Significant differences: *P < 0.05, **P < 0.01.

To analyze the effect of IL-5R ASO, single-cell suspensions of fetal liver cells were grown with IL-5 in the presence or absence of IL-5R antisense or mismatch oligonucleotides in the culture medium from the start of the culture. In the absence of IL-5, the oligonucleotides had no endogenous effect on the cell differentiation and maturation into eosinophils (data not shown). Figure 3b shows results from a representative experiment where the effects of IL-5R ASO and mismatch control were analyzed at different time points during the IL-5-induced eosinophil maturation of fetal liver cells. A significant and dose-dependent inhibition at all time points was observed with the IL-5R ASO. Maximal inhibition of eosinophil maturation was reached at Day 12. Although to a considerably lower degree, the mismatch control oligonucleotides also produced some inhibitory effects, especially at high concentrations. The results in Figure 3c summarize the data obtained from multiple experiments at Day 10. Significant reductions of the numbers of eosinophils were found after treating the cells with 10 or 20 µM of the IL-5R ASO, whereas only minimal effects of the mismatch controls were observed. Together, the data obtained so far clearly demonstrate the ability of IL-5R ASO to selectively reduce mRNA and proteins for the IL-5R chain with functional consequences for IL-5 responding cells in vitro.

Effect of IL-5R ASO Treatment on IL-5-Induced Eosinophilia In Vivo

To analyze the potency of the oligonucleotides to reduce eosinophilia in vivo, administration of antisense, mismatch, and vehicle was performed daily intravenously for 7 d together with daily dosing of recombinant murine IL-5 (intraperitoneally) for the last 5 d. The i.p. administration of IL-5 over 5 d induced a moderate but significant blood and peritoneal eosinophilia in these mice (Figure 4) without significantly affecting the numbers of the other leukocyte subpopulations. Treatment with the IL-5R ASO resulted in a reduction of both the relative and absolute numbers of eosinophils present in blood and peritoneal lavages without significantly changing the number of other leukocyte subsets. (Blood total eosinophil counts, in millions: 0.107 ± 0.050, PBS; 0.139 ± 0.022, IL-5; 0.089 ± 0.027, IL-5 + antisense; 0.143 ± 0.044, IL-5 + mismatch. Peritoneal lavage total eosinophil counts, in millions: 0.159 ± 0.049, PBS; 0.545 ± 0.141, IL-5; 0.275 ± 0.094, IL-5 + antisense; 0.638 ± 0.180, IL-5 + mismatch.) Moreover, administration of the IL-5R ASO also significantly reduced the number of eosinophils and eosinophil precursors in the bone marrow as determined by eosinophil peroxidase staining (Figure 4c). No nonspecific effect related to oligonucleotide administration was observed, inasmuch as no change in cell numbers occured upon administration of the mismatch control oligonucleotide in the peritoneal cavity and blood. Thus, the observation that the control oligonucleotide containing five base mismatches had no effect on the development of eosinophils after IL-5 treatment indicates that the effect observed with the antisense is sequence-specific and consistent with the antisense mode of action found in the in vitro experiments.

View larger version (26K): [in this window] [in a new window]   Figure 4.   Reduction of eosinophilia in an IL-5-induced peritonitis model. Balb/c mice were injected intravenously daily with antisense, mismatch and vehicle for 7 d together with daily dosing of IL-5 (intraperitoneally) for the last 5 d. Peritoneal lavage, blood, and bone marrow were collected 24 h after the last injection. Differential cell counts in blood (a) and peritoneal washout (b) from mice treated with PBS (open bars), IL-5 vehicle (filled bars), IL-5 mismatch (MS) (hatched bars), and IL-5 antisense (AS) (shaded bars). (c) Eosinophil cell counts (percentage of total cells) in the bone marrow ( filled bars), blood (shaded bars), and peritoneal lavage (hatched bars) of treated mice. Values are means ± SEM from six to eight mice per group. Significantly different from the corresponding control group: *P < 0.05, **P < 0.01.

IL-5R ASOs Suppress the Development of Eosinophilia in a Ragweed-Induced Allergic Peritonitis Model In Vivo

To confirm the inhibition seen with IL-5R ASO in the direct IL-5-induced eosinophilia in vivo, the effects of IL-5R ASO treatment were analyzed in an Ag (ragweed)-induced allergic peritonitis model. As shown in Figure 5, the i.p. administration of ragweed Ags in actively sensitized mice induced a significant bone-marrow, blood, and peritoneal lavage eosinophilia. Intravenous treatment of these mice with an IL-5R-specific ASO at the time of ragweed sensitization and challenge resulted in a significant and dose-dependent reduction in the number of eosinophils recovered in bone marrow, blood, and peritoneal lavages. An almost-100% inhibition of the ragweed-induced eosinophilia in bone marrow and blood was found after the treatment with 20 mg/kg of the IL-5R ASO, whereas the same concentration of a mismatch control oligonucleotide resulted in no significant reduction of the number of eosinophils. The inhibition of eosinophils present in the peritoneal cavity was also significantly reduced but to a lower degree suggesting that the active infiltration of eosinophils into the inflamed tissue is less affected by IL-5-dependent processes. Similar to the results observed after IL-5 treatment (Figure 4), neither the administration of the IL-5R antisense nor the mismatch control affected the numbers of the other leukocyte subpopulations (data not shown).

View larger version (42K): [in this window] [in a new window]   Figure 5.   Reduction of eosinophilia mediated by an IL-5R ASO in a ragweed-induced peritonitis model. Balb/c mice were immunized with ragweed Ag extract on Days 1 and 8 and challenged on Day 15. Oligonucleotides were given daily over a 3-d period starting the day before immunization (Day 0) until Day 2, from Days 7 to 9, and from Days 14 to 16. Peritoneal lavage, blood, and bone marrow were collected 48 h after the ragweed challenge. Eosinophil cell counts (percentage of total cells) in the bone marrow (filled bars), blood (shaded bars), and peritoneal lavage (hatched bars) of treated mice. Values are means ± SEM from six to eight mice per group. Significantly different from the corresponding control group: *P < 0.05, **P < 0.01.

	Discussion

Top Abstract Introduction Material and Methods Results Discussion References

Considerable progress has been made in the development of modified ASOs that are complementary to the mRNA encoding a specific protein and can be used therapeutically to inhibit its synthesis. This approach has successfully been applied to target, for example, the synthesis of A1 adenosine receptors in an asthma model in rabbits, which resulted in beneficial therapeutic effects (26). A similar strategy could therefore be used to block the synthesis of IL-5 or its receptor. Indeed, IL-5-specific ASOs were identified which selectively suppressed the production of IL-5 both in vitro and in vivo (27). The data reported in this study represent an alternative approach targeting the cytokine-specific  subunit of the IL-5 receptor.

ASOs that support RNAse H-mediated degradation of murine IL-5R-chain mRNA were developed and used in an in vitro model system of IL-5-mediated effector functions. More specifically, we could demonstrate that the IL-5R ASO selectively reduced IL-5R mRNA and protein expression and effectively inhibited IL-5-induced eosinopoiesis. The loss of specific inhibition of target mRNA and protein when using mismatch oligonucleotides with increasing numbers of mismatches confirmed the sequence-specific effect and the antisense mechanism of action. Moreover, no other cytokine receptor gene products, including the closely related GM-CSF receptor  chain, were downregulated in response to the IL-5R chain-specific ASO.

The position of the oligonucleotide sequence in the 3'UTR and the experimental in vitro data demonstrated that the IL-5R chain-specific ASO inhibited both the membrane and soluble forms of the receptor, which are produced by differentially spliced transcripts and mediate or antagonize IL-5 functions, respectively. The soluble IL-5R subunit binds IL-5 with only slightly reduced affinity compared with the membrane form (28). Moreover, it was shown to act as a selective IL-5 antagonist in vitro and to suppress Ag-induced eosinophil infiltration into the bronchial lumen in vivo (29). Thus, as a therapeutic principle, it would be preferential to block only the membrane form of the IL-5R chain without affecting the expression of the soluble form, and experiments are in progress to develop membrane-specific IL-5R ASOs.

Our in vivo experiments in the IL-5-induced and the ragweed-induced peritonitis models, known to be associated with a selective and IL-5-dependent tissue accumulation of eosinophils (30), demonstrated that oligonucleotide treatment efficiently suppressed the accumulation of eosinophils. Intravenous administration of IL-5R ASO significantly reduced the number of eosinophils in bone marrow and blood and, to a lesser extent, in the peritoneal cavity. This finding is consistent with the concept that IL-5 is more important for the differentiation of eosinophilic precursors and for mobilizing mature eosinophils from the bone marrow than for their recruitment into the inflamed tissues, and further supports animal studies reporting the potential of IL-5 to both mobilize a bone-marrow pool of eosinophils and increase eosinophil differentiation from bone-marrow precursors (30, 31). Indeed, in vivo distribution studies using similar oligonucleotides have shown an appreciable bone-marrow uptake after intravenous administration in rats and, combined with the improved nuclease resistance of the MOE chemistry used in the 5' and 3' ends of the oligonucleotides, these data suggest that the site of action is likely to be the bone marrow (32). Moreover, specific upregulation of IL-5R subunit expression on bone marrow-derived CD34 + cells as well as increased eosinophil progenitor cell numbers in the bone marrow have been reported in patients with mild asthma upon local lung allergen challenge (24, 33). This suggests that down-regulating the IL-5R expression may be critical in controlling the contribution of the bone marrow to produce more eosinophil-committed cells as well as reducing eosinophil trafficking into the inflammed tissue. Indeed, in the two in vivo models used in this study there was an increase of bone-marrow eosinophil progenitors, indicating that the increased eosinophilia involves an expansion of the relevant bone-marrow stem-cell population, which is probably the main oligonucleotide target.

In both in vivo models, treatment with the IL-5R-specific ASOs resulted only in a limited reduction of eosinophilia in the peritoneal cavity. This is most likely due to the fact that chemokines, in particular eotaxin, play a more important role than does IL-5 in recruiting eosinophils into the inflamed tissues. Evidence for a maximal expression level of the eotaxin receptor CCR3 on mature peripheral blood eosinophils was reported (34), further supporting a possible role for eotaxin in controling cell trafficking from the blood into the inflamed tissue. The eosinophilia seen in the peritoneal lavages in our two models may therefore result from a synergy between eotaxin and IL-5 to mobilize eosinophil from a pre-existing blood pool and thereby may partly hide the overall eosinophilia-suppressing effect of the IL-5R ASO within tissues. However, at a later time point or in a more chronic stituation, the potent reduction in bone-marrow expansion of eosinophil progenitors and in blood eosinophilia seen upon IL-5R ASO treatment could subsequently lead to a reduction of eosinophil recruited in the inflamed tissue. Indeed, upon ovalbumin challenge in a mouse model of airway allergen challenge, airway eosinophilia was present earlier than the bone-marrow responses (35) as a result of rapid recruitment of eosinophils from pre-existing pools.

Another potential explanation for the oligonucleotide- mediated reduction of eosinophilia is the induction of a T helper (Th) 1 immune response by immunostimulatory DNA sequences that may counteract an ongoing Th2 response characteristic of allergic inflammation. However, the IL-5R ASO does not contain any known immunostimulatory CpG motifs, and treatment with the IL-5R ASO or the mismatch controls did not result in splenomegaly, a characteristic pathologic feature observed in mice treated with CpG-containing oligonucleotides (data not shown).

In conclusion, the development of drugs to neutralize the effect of IL-5 still represents a novel therapeutic approach to control the characteristic eosinophilia in allergic diseases. The ideal IL-5 antagonist would be a low molecular weight synthetic molecule that presents fewer problems in drug production, administration, cost, and stability than do peptides or protein therapeutics. However, due to the difficulties in identifying a low molecular weight antagonist of protein-protein interaction, most investigators have focused on the development of protein or peptide therapeutics. Both preclinical and clinical studies using monoclonal antibodies have shown encouraging results in terms of effect on blood and tissue eosinophil numbers after allergen challenge. The data reported in this study correspond well with the IL-5 antibody studies and suggest that an antisense approach may represent another option for therapeutic intervention in IL-5-mediated diseases.