Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The signal transducer and activator of transcription 6 (Stat6) is an obligatory transcription factor in interleukin (IL)-4 and IL-13 signaling (1). IL-4 and IL-13 signal via their common IL-4 receptor (IL-4R)  chain (5) and play a critical role in the induction of germline C transcripts and immunoglobulin (Ig)E isotype switching in B cells (8, 9). Stat6 is therefore a key effector in allergic disease and may be central to the pathophysiology of asthma. T-helper (Th)2 cells respond to antigen presentation by releasing IL-4, which, in turn, results in Th2 cell expansion and the production of further Th2 cell cytokines (IL-13, IL-5, and IL-4), thus amplifying the initiation cascade (10- 12). The pivotal role of IL-4 is highlighted by IL-4-neutralizing antibodies that selectively inhibit the expansion of Th2 cells (13). IL-13 is capable of inducing characteristics associated with the allergic asthmatic phenotype, including goblet-cell metaplasia, increased airway mucus production, airway hyperresponsiveness, raised bronchoalveolar lavage eosinophil number, and elevated total serum IgE (14, 15). Moreover, the importance of the IL-4/ IL-13 pathway for IgE synthesis has been demonstrated in IL-4 and IL-13 knockout mice that fail to develop a Th2 response after nematode infection (16, 17) and are deficient in IgE synthesis (17, 18).

IL-4 signaling is mediated through two types of heterodimer receptor complexes (7, 19). In hematopoietic cells IL-4 signals through the type I receptor, which is comprised of the high-affinity IL-4 binding IL-4R  chain, and the IL-2R  common (c) chain, also shared by IL-7R, IL-9R, and IL-15R (19). In nonhematopoietic cells, IL-4 signals via the type II receptor complex, which is comprised of the IL-4R  chain and the low-affinity binding IL-13R  chain and is able to mediate both IL-4 and IL-13 signaling. The IL-4R  and IL-2R c subunits are associated with Jak1 and Jak3 tyrosine kinases, respectively (22). In nonhematopoietic cells Jak1 and Jak2 are activated by IL-4; however, the IL-13R -specific Jak has not yet been identified. Upon stimulation with IL-4 in hematopoietic cells, the IL-4R  chain is phosphorylated by Jak3 and binds Stat6 via interaction between its phosphotyrosine motif and the Src homology 2 domain of Stat6. Stat6 itself is subsequently phosphorylated by Jak1 and homodimerises. The dimeric form translocates to the nucleus where it binds to IL-4 response elements, initiating transcription of several genes, including IgE (22).

Stat6 is essential for IgE switch recombination (3, 4) and contributes to the differentiation of naive T cells to the Th2 subtype and the transcription of IL-4 and the IL-4 receptor (3, 23). This is exemplified by Stat6 / knockout mice that are unable to mount an IgE response following challenge with Nippostrongylus brasiliensis or to direct the differentiation of naive T cells toward the Th2 phenotype (1). Further, Stat6-deficient mice do not develop bronchial hyperresponsiveness or secrete large amounts of mucus when compared with wild-type mice after antigen sensitization/challenge (26).

Inhibition of Stat6 would be expected to attenuate the allergic response. It therefore represents an attractive but challenging target for conventional drug discovery strategies. The antisense conceptselective inhibition of protein synthesis through hybridization of an oligonucleotide to its complementary sequence in messenger RNA (mRNA) of the target proteinwas first proposed by Zamecnik and Stephenson in 1978 (27, 28). This approach has subsequently found broad application as an experimental tool and for generating putative therapeutic molecules (29).

The present study describes the identification and characterization of antisense oligonucleotides to Stat6 as an alternative approach to interrupting IL-4 and IL-13 signaling. Homologous human/murine oligonucleotides were selected to facilitate future testing of human-relevant antisense oligonucleotides in in vivo murine models of atopy. Further, special consideration was given to the design of the oligonucleotide in order to avoid immune stimulation (32) and nonspecific effects (35, 36). Second-generation, 2'-O-(2-methoxy)- ethyl (2'-MoE)-modified oligonucleotides (37) were used and a ribonuclease (RNase) H-dependent antisense mechanism of action was demonstrated. The functional consequence of antisense downregulation of Stat6 on IL-4-mediated germline C transcription was assessed in vitro.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Oligonucleotide Design and Synthesis

Human and murine Stat6 mRNA sequences (accession numbers U16031 and L47650, respectively) were obtained from Genbank and mRNA secondary structures were predicted using the foldRNA program (Wisconsin Package Version 9.1; Genetics Computer Group [GCG], Madison, WI). The coding regions (human = 166-2,710, murine = 280-2,793) were aligned and homologous regions identified. Oligonucleotides were designed according to the following principles: (1) 20-base antisense sequence, (2) complete human-mouse homology, (3) complementary to at least five bases that are not hybridized in the predicted secondary structure, (4) contain less than four contiguous guanine or thymine bases, and (5) minimal cytosine-guanine motifs. The identified sequences (see Table 1) were synthesized as phosphorothioate (PS) 1a-9a (Oswell, Southampton, UK); 2'-MoE gapmer 1b-9b, 9b1mm-9b4mm; or 2'-MoE fully modified oligonucleotides 1c, 7c, and 9c (Dr. François Natt; Novartis, Basel, Switzerland).

Cell Culture

Cell culture reagents were supplied by GIBCO BRL (Paisley, UK). A549 human lung epithelial cells were cultured in Dulbecco's modified Eagle's medium supplemented with 15% fetal calf serum (FCS) at 37°C in a humidified atmosphere of 10% CO₂. L929 murine lung fibroblast cells were cultured in minimum essential medium supplemented with 10% FCS at 37°C in a humidified atmosphere of 5% CO₂. DND39 cells were grown in RPMI 1640 plus glutamax, supplemented with 10% FCS, 20 µM -mercaptoethanol at 37°C in a humidified atmosphere of 5% CO₂.

Cationic Lipid-Mediated Transfection

Active oligonucleotides were selected on the basis of their ability to downregulate Stat6 mRNA and protein in the A549 (human) and L929 (murine) adherent cell lines using Lipofectin- or LipofectAmine-mediated transfection, respectively. The culture medium was aspirated and the cells were washed with phosphate-buffered saline (PBS), then transfected with a transfection cocktail (3 µl transfection reagent per milliliter of OptiMEM per 100 nM oligonucleotide) with or without test oligonucleotide. After 4 h incubation at 37°C, the transfection cocktail was replaced with culture medium and the cells were maintained under their normal culture conditions. Cells were harvested for mRNA or protein analysis at the defined time intervals. The cells were transfected once (t = 0 h) for time points of 24 h and twice (t = 0 then t = 24 h) for time points greater than 24 h.

Northern Blot Analysis

Cells were harvested into Triazol reagent (GIBCO BRL) and total RNA was isolated according to the manufacturer's instructions. RNA was separated by electrophoresis on a formaldehyde, 1.2% agarose gel; transferred onto a nylon membrane (Hybond N; Amersham [Buckinghamshire, UK] Ltd.) and hybridized to [^[32]P]-labeled human-Stat6, mouse-Stat6, germline C transcript, or glyceraldehyde-3-phosphate dehydrogenase (GAPDH) complementary DNA probes. The probes were prepared by the random primer method (Rediprime; Amersham). After high-stringency washes the membrane was exposed to a phosphor screen for 16 h before analysis on a PhosphorImager (Storm 840; Molecular Dynamics, Buckinghamshire, UK). mRNA levels were quantitated using Quantanalysis software (Molecular Dynamics) and normalized by reference to GAPDH.

Western Blot Analysis

Cells were harvested into protein sample loading buffer (10% glycerol, 4% sodium dodecyl sulfate [SDS], 200 mM dithiothreitol, 100 mM Tris-HCl [pH 6.8], and 0.2% bromophenol blue) and boiled for 15 min before loading on a 10% SDS-polyacrylamide gel electrophoresis gel; separation was performed at 20 V/cm for 1 h. The separated proteins were transferred onto a polyvinylidene difluoride membrane (Millipore, Hertfordshire, UK). Stat6, Stat1 (cross-reactive with 1 and 1), Stat2, Stat3, Stat5b (cross-reactive with 5a and 5b), and extracellular signal-regulated kinase (Erk)-2 proteins were probed with specific antibodies (Autogen Bioclear, Wiltshire, UK; Transduction Laboratories, Lexington, KY) and visualized by enhanced chemifluorescence (Vistra ECF, Amersham; Storm 840, Molecular Dynamics). Stat6 protein was quantitated relative to Erk-2 using Quantanalysis software (Molecular Dynamics).

Electroporation Transfection

The ability of oligonucleotide 9b to exert a functional effect on the expression of the germline C transcript was determined in the nonadherent DND39 cell line. An electroporation transfection protocol was developed because cationic lipid-mediated transfection was unsuccessful. DND39 cells (1 × 10⁷) were resuspended in 400 µl of culture medium in the presence and absence of test oligonucleotide. Cells were chilled on ice for 15 min and electroporated in a 4-mm gap cuvette using a Gene pulser II electroporator (Bio-Rad, Hertfordshire, UK) set at 275 V, 50 ohms, and 1,000 µfarads. Electroporated cells were transferred to 10 ml of culture medium in 100-mm dishes (Corning Costar, Buckinghamshire, UK) and maintained at 37°C for 24 h under normal culture conditions. The cells were resuspended, 8 ml was removed for RNA analysis, and 2 ml was diluted to 5 ml of medium and cultured for a further 24 h before being harvested for protein analysis.

IL-4 Stimulation

A recombinant Chinese hamster ovary/human IL-4-rich supernatant was prepared from CHO-huIL4 (KI9) cells. At 24 h after electroporation, in the presence or absence of oligonucleotide 9b or 9b4mm, 5 × 10⁶ DND39 cells were washed in PBS and resuspended in 5 ml of medium (RPMI 1640, 10% FetalClone I, 2 mM glutamine, 50 µM -mercaptoethanol, 50 µg/ml gentamycin, and 5% IL-4 supernatant fraction). The cells were harvested 24 h after stimulation and analyzed by Northern blot for human Stat6, germline C transcript, or GAPDH mRNAs, and by Western blot for Stat6 or Erk-2 proteins.

Germline C Transcript Probe Generation

mRNA was isolated from IL-4-stimulated DND39 cells and reverse transcribed. A germline C transcript probe was amplified using polymerase chain reaction (PCR) with forward primer AGGCTCCACTGCCCGGCACAGAAAT (epsilon I exon: accession number X56797) and reverse primer TGAAGTCCCTGGAGCAGACTGGGGG (epsilon constant region: accession number J00222) resulting in a 448-base pair fragment. This fragment was purified (QIAquick gel extraction kit; Qiagen, West Sussex, UK) and its identity confirmed by DNA sequence analysis using an ABI 310 automated sequencer. PCR-amplified product was radiolabeled as described previously for Northern blot analysis.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

In Vitro Activity Studies

Effects of Stat6 antisense oligonucleotides on Stat6 mRNA and protein in A549 and L929 cell lines. The limited gene-walk with antisense oligonucleotides complementary to homologous human and murine Stat6 mRNA identified a number of oligonucleotides capable of downregulating Stat6 mRNA and protein (Figure 1). Oligonucleotide activity was dependent upon the choice of chemical modification (compare Figures 1A with 1B or 1C with 1D) and demonstrated species selectivity (compare Figures 1A with 1C or 1B with 1D). A consistent relationship between Stat6 mRNA levels and Stat6 protein expression was not observed. Taking 60% protein downregulation as a cutoff point for active sequences, 2'-MoE-gapmer oligonucleotides had a greater hit rate and higher antisense activity than did the 2'-deoxy phosphorothioate sequences, and were of more interest for subsequent studies, especially because they are associated with fewer nonspecific effects. Of the 2'-MoE-gapmer sequences, oligonucleotide 9b, targeting homologous sequences in the coding region of human and mouse Stat6 mRNA, demonstrated a high level of activity in both species and was chosen for further studies.

View larger version (29K): [in this window] [in a new window]   Figure 1.   In vitro activity studies. A549 and L929 cells were treated with oligonucleotides (400 nM) complementary to human and murine Stat6, using cationic lipid-mediated transfection (n = 4). Cells were harvested for mRNA and protein analysis at 24 and 48 h, respectively. (A) A549 cells transfected with oligonucleotides 1a-9a containing PS backbone; (B) A549 cells transfected with oligonucleotides 1b-9b containing 2'-MoE-gapmer backbone; (C) L929 cells transfected with oligonucleotides 1a-9a containing PS backbone; (D) L929 cells transfected with oligonucleotides 1b-9b containing 2'-MoE-gapmer backbone. Stat6 mRNA (filled bars) and protein (open bars) expression levels were measured by Northern blot and Western blot analyses.

Dose response. Oligonucleotide 9b was shown to downregulate Stat6 mRNA and protein in A549 cells in a dose-dependent manner. Oligonucleotide 9b was used at concentrations ranging from 100 to 400 nM for Northern analysis and at concentrations ranging from 50 to 500 nM for Western analysis (see Figure 2). The concentrations that produced 50% inhibition of effect (IC₅₀) were 168 and 215 nM for its effect on Stat6 mRNA and protein, respectively.

View larger version (16K): [in this window] [in a new window]   Figure 2.   Dose response of oligonucleotide 9b for the reduction of Stat6 mRNA and protein in A549 cells. A549 cells were treated with the indicated concentrations of oligonucleotide 9b (2'-MoE gapmer) using cationic-mediated transfection (n = 4). (A) Stat6 mRNA and (B) Stat6 protein expression were determined at 24 and 48 h after transfection, respectively, by Northern and Western blot analyses.

Time-course analysis. Maximum downregulation of Stat6 mRNA and protein by oligonucleotide 9b was seen from 8 to 24 h and 24 to 48 h, respectively, in both A549 and L929 cells after a single oligonucleotide treatment (Figure 3). A second treatment at 24 h resulted in further reduction of Stat6 mRNA levels and maintained the downregulation of Stat6 protein for 72 h. In all cases, Stat6 mRNA depletion preceded the reduction in Stat6 protein.

View larger version (14K): [in this window] [in a new window]   Figure 3.   Time-course effect of oligonucleotide 9b for the reduction of Stat6 mRNA and protein in A549 and L929 cells. (A) A549 and (B) L929 cells were treated with oligonucleotide 9b (2'-MoE gapmer, 400 nM) using cationic lipid-mediated transfection (n = 4). Stat6 mRNA (filled circles) and Stat6 protein (open circles) expression were determined by Northern and Western blotting. Samples harvested at 48 h were treated either once (48[1]), at t = 0 h, or twice (48[2]), at t = 0 and t = 24 h, with antisense oligonucleotide. Samples harvested at 72 h were treated at t = 0 and 24 h.

RNase H-Dependent Activity and Oligonucleotide Specificity

Antisense mechanism. The oligonucleotides exerted their antisense effects through an RNase H-mediated mechanism in A549 and L929 cells (Figures 4A and 4B, respectively). 2'-MoE-gapmer oligonucleotides 1b, 7b, and 9b, with a central eight-base stretch of PS 2'-deoxy nucleotides capable of supporting RNase H activity, decreased both human and murine Stat6 mRNA. In contrast, fully modified 2'-MoE sequence analogues 1c, 7c, and 9c, which do not support RNase H, were inactive. Stat6 mRNA downregulation by oligonucleotides 1b, 7b, and 9b was followed by Stat6 protein inhibition. Downregulation of Stat6 protein after treatment with oligonucleotide 9c was observed in A549 cells in the absence of Stat6 mRNA inhibition. A significant increase in Stat6 mRNA expression was seen in L929 cells treated with oligonucleotides 1c and 7c together with a 20% decrease in Stat6 protein.

View larger version (22K): [in this window] [in a new window]   Figure 4.   Antisense mechanism. The antisense mechanism was investigated by comparing the activity of antisense oligonucleotides containing the 2'-MoE-gapmer modification with the full 2'-MoE modification. (A) A549 and (B) L929 cells were transfected (n = 4) with 2'-MoE-gapmer oligonucleotides 1b, 7b, or 9b (400 nM) or full 2'-MoE oligonucleotides 1c, 7c, or 9c (400 nM). Stat6 mRNA (filled bars) and protein (open bars) were analyzed 24 and 48 h after transfection by Northern and Western blotting, respectively.

Oligonucleotide specificity: mismatch analysis. Oligonucleotide specificity was demonstrated at the Stat6 protein level using a series of oligonucleotides in which successive nucleotide base substitutions were made in the 2'-MoE wings (Table 1). To examine the sequence specificity of oligonucleotide 9b on Stat6 protein expression, A549 cells were treated with either oligonucleotide 9b or mismatched oligonucleotides 9b1mm-9b4mm over a concentration range from 100 to 500 nM. None of the mismatched oligonucleotides were as active as oligonucleotide 9b. Further, the degree of protein downregulation decreased with each additional base substitution up to four mismatches, at which point antisense activity was completely abrogated (Figure 5).

View larger version (19K): [in this window] [in a new window]   Figure 5.   Oligonucleotide specificity: mismatch analysis. Sequence specificity of oligonucleotide 9b was investigated using a series of oligonucleotides containing 0 to 4 mismatches (0 mismatches, filled circles; 1 mismatch, open circles; 2 mismatches, filled triangles; 3 mismatches, open triangles; 4 mismatches, filled squares) in the 2'-MoE wings (Table 1). A549 cells were treated with the indicated oligonucleotides at increasing concentrations (100 to 500 nM) at t = 0 and 24 h (n = 4). Stat6 protein expression was measured by Western blot analysis at 48 h after transfection.

Oligonucleotide specificity: cross-reactivity studies. The activity of oligonucleotide 9b in A549 and DND39 cells was highly specific to Stat6 (Figure 6). Western blot analysis of A549 cells treated with oligonucleotide 9b, using antibodies to Stat1, Stat1, Stat2, Stat3, Stat5a, Stat5b, and Stat6, demonstrated > 80% downregulation of Stat6 protein without affecting other members of the closely related Stat family. Similarly, oligonucleotide 9b demonstrated high specificity in DND39 cells, with > 50% reduction in Stat6 protein. Erk-2 was used for normalization purposes.

View larger version (64K): [in this window] [in a new window]   Figure 6.   Oligonucleotide specificity: cross-reactivity studies. Sequence specificity of oligonucleotide 9b was investigated by Western blot analysis using antibodies to Stat6 and the highly related Stat family members Stat1, Stat1, Stat2, Stat3, Stat5a, and Stat5b. A549 and DND39 cells were treated with oligonucleotide 9b (400 nM and 10 µM, by cationic lipid and electroporation, respectively). C: no oligonucleotide control, T: oligonucleotide-treated. Stat protein expression was measured by Western blot analysis at 48 h after transfection. Erk-2 was used to normalize for sample loading.

Functional Effect of Stat6 Protein Downregulation

Dose response of oligonucleotide 9b in DND39 cells. Dose-dependent downregulation of Stat6 mRNA and protein was demonstrated in DND39 cells treated with oligonucleotide 9b (Figure 7). Transfection was performed via electroporation because cationic lipid treatment was ineffective in these cells. Stat6 mRNA and protein expression was investigated after treatment with oligonucleotide 9b over a concentration range of 1 to 10 µM. The IC₅₀ for both mRNA and protein was calculated to be ~ 1 µM.

View larger version (14K): [in this window] [in a new window]   Figure 7.   Dose response of oligonucleotide 9b for the reduction of Stat6 mRNA and protein in DND39 cells. DND39 cells were treated with oligonucleotide 9b at the concentrations indicated using electroporation (n = 4). Stat6 mRNA (filled bars) and protein (open bars) were analyzed 24 and 48 h after transfection by Northern and Western blotting, respectively.

Effect of oligonucleotide 9b on the expression of germline C transcript. Stat6 antisense oligonucleotide 9b inhibited germline C transcription in DND39 cells. IL-4 stimulation of DND39 cells upregulated the 1.78-kb germline C mRNA transcript (Figure 8b). Treatment of DND39 cells with oligonucleotide 9b before IL-4 stimulation downregulated Stat6 mRNA (Figure 8B) and protein (Figure 8C), with concurrent inhibition of germline C transcript synthesis (Figure 8B). A larger, 3.57-kb transcript (Figure 8B), also inhibited by oligonucleotide 9b treatment, was observed by Northern analysis using the germline C transcript probe. This unidentified transcript has previously been reported (38) but has not been characterized. Oligonucleotide 9b4mm, an analogue of oligonucleotide 9b containing four mismatches in the 2'-MoE modified wings, failed to downregulate Stat6 mRNA or protein and was unable to inhibit germline C transcript synthesis after IL-4 treatment (Figure 9B).

View larger version (30K): [in this window] [in a new window]   Figure 8.   Effect of oligonucleotide 9b on the expression of germline C transcript. DND39 cells were treated with or without oligonucleotide 9b (10 µM) for 24 or 48 h, followed by a 24-h stimulation with human recombinant IL-4. (A) Treatment regime. (B) Northern blot analysis: Stat6 transcript (3.1 kb) was measured before IL-4 stimulation, followed by the germline C transcript (1.78 kb) 24 h after IL-4 stimulation. An uncharacterized transcript (3.57 kb) was also observed using the germline C transcript probe. (C) Western blot analysis of Stat6 (97 kD) was determined before IL-4 stimulation. GAPDH (1.7 kb) and Erk-2 (42 kD) were used to normalize for sample loading.

View larger version (29K): [in this window] [in a new window]   Figure 9.   Effect of mismatch oligonucleotide 9b4mm on the expression of germline C transcript. DND39 cells were treated with or without oligonucleotide 9b or 9b4mm (10 µM), as indicated, for 24 h followed by a 24-h stimulation with human recombinant IL-4. (A) Treatment regime. (B) Northern blot analysis: Stat6 transcript (3.1 kb) was measured before IL-4 stimulation, followed by the germline C transcript (1.78 kb) 24 h after IL-4 stimulation. (C) Western blot analysis of Stat6 (97 kD) was determined before IL-4 stimulation. GAPDH (1.7 kb) and Erk-2 (42 kD) were used to normalize for sample loading.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Cytokines IL-4 and IL-13 are responsible for Th2 cell proliferation and the induction of IgE synthesis in B cells and consequently play a key role in the pathophysiology of diseases such as asthma. Antigen presentation to Th2 cells results in T-cell proliferation with the release of a host of cytokines including IL-4 and IL-13, which promote IgE synthesis, via activation of the transcription factor Stat6. Further exposure to antigen results in crosslinking of IgE on the surface of mast cells and basophils, triggering the release of inflammatory mediators and thus resulting in tissue damage and subsequent tissue remodeling observed in allergic asthma. Because isotype switching is the mechanism by which B lymphocytes differentiate into IgE-producing plasma cells, this process represents an attractive target for drug intervention. Isotype switching occurs due to a DNA rearrangement initiated by two successive signaling processes. First, IL-4/IL-13 stimulation leads to activation of the IgE germline promoter (GLP) via the transcription factor Stat6. Activation of the IgE GLP results in the synthesis of a germline C transcript, a "sterile" transcript, which, although not translated into protein, is essential for IgE synthesis. Subsequent interaction of the CD40 receptor on B cells with the CD40 ligand present on activated T lymphocytes signals switch recombination of the Ig heavy-chain locus necessary for IgE synthesis. It has been reported that although CD40 ligation contributes to deletional recombination, it is not a requirement for this molecular event inasmuch as Epstein-Barr virus infection and lipopolysaccharide stimulation have been shown to substitute for CD40 signaling (39). Further, downregulation of the activity of the IgE GLP by transforming growth factor-, a natural inhibitor of germline C transcription, results in diminished IgE production (42). On the basis of this knowledge we adopted an antisense approach to selectively downregulate the synthesis of Stat6, an essential transcription factor in IL-4/IL-13-induced IgE synthesis.

First-generation PS 2'-deoxynucleotides have been used extensively in antisense studies due to their enhanced resistance to cellular nucleases and their significantly extended half-life compared with the naturally occurring phosphodiester form. However, reports suggesting that some PS oligonucleotides may exert their biologic effects through non-antisense mechanisms (35), together with their association with immune stimulation (32), have given rise to the development of a second generation of oligonucleotides with modifications at the 2'-sugar position (37, 43). These modifications give significant increases in duplex stability and nuclease resistance, resulting in a class of antisense oligonucleotides that are substantially more effective at inhibiting gene expression than are the widely used PS modified sequences (37, 44, 45). 2'-MoE- modified oligonucleotides exhibit higher-affinity binding over their PS sequences but do not support RNase H- dependent degradation of the target mRNA. To overcome the problems of PS nonspecificity and the inability of 2'-MoE oligonucleotides to support RNase H activity, chimeric oligonucleotides with reduced PS content have been used. 2'-MoE-gapmer sequences are chimeras in which the central portion of the molecule contains phosphorothioate 2'-deoxynucleotides, to support RNase H degradation, and the outer regions or wings are composed of phosphodiester 2'-MoE-modified nucleotides.

Antisense oligonucleotides targeted to homologous regions of human and murine Stat6 mRNA were designed following principles (1) to (5) described in MATERIALS AND METHODS. Oligonucleotides directed to regions of human/ murine homology were chosen to allow future in vivo testing of active compounds from this study. In vitro activity studies were performed in human A549 and murine L929 cell lines, using 400 nM of each oligonucleotide for both PS (oligonucleotides 1a-9a) and 2'-MoE-gapmer modifications (oligonucleotides 1b-9b). Antisense sequences tested in this study showed a differential ability to downregulate Stat6 mRNA and protein; the activity of each sequence was both species-specific and dependent on the chemical modification used. 2'-MoE gapmer-modified sequences demonstrated the greatest level of activity when compared with the 2'-deoxy PS sequences, with 67% of the sequences producing  60% protein downregulation in both human and murine cells. This confirms earlier reports of increased antisense activity using 2'-MoE-modified sequences (37, 45). Oligonucleotide 9b demonstrated high levels of antisense activity for mRNA and protein in both A549 and L929 cells, and was chosen for further in vitro profiling.

Antisense oligonucleotides have been described as exogenous regulators of mRNA metabolism that exert their effects either by promoting enzyme-mediated mRNA degradation by nucleases (e.g., RNase H) or by steric interference with one or more of the essential steps in the intermediary metabolism of the mRNA or its utilization by the cell (e.g., initiation of translation) (36). The antisense mechanism of oligonucleotides in this study was investigated using 2'-MoE-gapmer (oligonucleotides 1b, 7b, and 9b) and fully modified 2'-MoE (oligonucleotides 1c, 7c, and 9c) sequences. The antisense activity of oligonucleotides 1b, 7b, and 9b at the mRNA level was completely ablated using oligonucleotides 1c, 7c, and 9c in both A549 and L929 cells, demonstrating that the 2'-MoE-gapmer oligonucleotides act by an RNase H-dependent mechanism. Additionally, treatment of A549 cells with oligonucleotide 9c resulted in inhibition of Stat6 protein synthesis while Stat6 mRNA levels remained unchanged. This indicates that oligonucleotide 9c downregulates protein synthesis by a non-RNase H-dependent mechanism, probably by steric interference of one or more of the events involved in the mRNA translation process. Interestingly, a significant increase in Stat6 mRNA expression was observed in L929 cells treated with oligonucleotides 1c and 7c. A previous study in which antisense oligonucleotides increased the abundance of cytosolic target transcript (46) suggested a decrease in the rate at which the transcript is normally degraded. The lack of corresponding protein upregulation by oligonucleotides 1c and 7c suggests that their hybridization also interferes with protein synthesis. In fact, these sequences show approximately 20% downregulation of Stat6 protein.

Compelling evidence to support a sequence-specific mechanism of action is demonstrated in A549 cells by rank order oligonucleotide potency on the basis of activity of a series of mismatched oligonucleotide sequences. Base substitutions were placed in the high-affinity binding 2'-MoE wing regions of oligonucleotide 9b. Progressive reduction of antisense activity was observed with each additional base substitution, and complete inhibition of activity was achieved with four mismatches. These data extend previous observations using mismatched PS oligonucleotides in A549 cells (47) in which five mismatches were demonstrated to completely abolish antisense activity at the mRNA level. Further, positioning of the mismatches in the 2'-MoE wings, which provide high-affinity binding and increased duplex stability, demonstrates the stringent selectivity that can be achieved via antisense technology.

Positive demonstration of antisense specificity in A549 and DND39 cells was also achieved using Western blot analysis of closely related family members of Stat6. Western blot analysis of Stat1, Stat1, Stat2, Stat3, Stat5a, Stat5b, and Stat6 proteins demonstrated no cross-reactivity of oligonucleotide 9b with any of these highly related sequences. Stat4 protein levels were too low for detection.

In previous studies, B cells from Stat6 knockout (Stat6 /) mice were shown to have a significantly impaired serum IgE response when immunized with conventional T-dependent antigens (1, 3, 23). Further, studies using splenic B cells from Stat6-deficient mice have demonstrated that Stat6 is an essential transcription factor in the IL-4-mediated activation of germline C transcription, a prerequisite to IgE synthesis (4). Additionally, expression of a truncated human Stat6 mutant, lacking the transactivation domain, in a human Burkitt lymphoma cell line shows partial inhibition of germline C transcription due to a dominant-negative effect (48). The present study shows that antisense inhibition of Stat6 gives rise to downregulation of germline C transcription in Burkitt lymphoma DND39 cells. DND39 cells produce a germline C transcript after IL-4 stimulation (38, 49) and were therefore considered a suitable model to examine the functional effect of inhibiting Stat6 using antisense oligonucleotides. Oligonucleotide 9b shows sequence-specific, dose-dependent downregulation of Stat6 mRNA and protein in DND39 cells. The higher IC₅₀ values for oligonucleotide 9b observed in DND39 cells compared with A549 cells probably reflects a less efficient oligonucleotide uptake rather than reduced antisense activity. Further, induction of germline C transcription after IL-4 stimulation of oligonucleotide 9b-treated DND39 cells has been shown to be dependent on the presence of Stat6. A larger, uncharacterized, 3.57-kb transcript, also upregulated by IL-4 stimulation, was observed by Northern blot analysis using the germline C transcript probe. Oligonucleotide 9b treatment of DND39 cells demonstrates that Stat6 is also essential for the synthesis of this larger, uncharacterized transcript; however, its role in IgE synthesis remains to be elucidated. These data exemplify the potential value of antisense technology as an experimental tool for drug target validation because the inhibition of gene expression can be achieved more rapidly and at a relatively lower cost using antisense compared with producing knockout mice.

This work describes the discovery and characterization of oligonucleotides, which are capable of downregulating Stat6 through an RNase H-dependent antisense mechanism in a highly specific manner. In particular, it demonstrates (1) the ability to successfully target homologous regions of target mRNA across species, (2) the specificity that can be achieved using second-generation oligonucleotides, and (3) that antisense downregulation of Stat6 results in functional effects, thereby providing further validation of its involvement in IgE synthesis in human cells. For these reasons, second-generation antisense oligonucleotides represent powerful tools for concept validation in vitro. Moreover, the encouraging pharmacologic effects reported for first-generation PS oligodeoxynucleotides in animal models and efficacy in clinical trials suggest the second-generation oligonucleotides may also be promising therapeutic candidates (50). Because Stat6 represents an attractive but technically challenging drug discovery target, antisense oligonucleotides are being considered as an alternative approach to low molecular-weight compounds for inhibiting the IL-4/IL-13 signaling pathway. To this end, the in vivo activities of selected Stat6 antisense oligonucleotides, identified and characterized in this study, are currently being investigated.