INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Antisense oligonucleotides (ASONs) are a promising new class of therapeutics, designed to specifically inhibit the expression of a disease gene, rather than inhibiting the activity of the preformed protein already participating in the disease process (1- 4). ASONs are generally short (14 to 25 mer), single-stranded oligonucleotides that hybridize to their target messenger RNAs (mRNAs) with high specificity and avidity through Watson-Crick base pairing (5, 6). Current therapeutic applications of ASONs include the attenuation of discordantly expressed genes in cancer (7, 8), restenosis after vascular surgery (9), viral infections (10), inflammatory bowel disease (13), hypertension (14), and most recently, asthma (15, 16).

Delivery of ASONs to their intended target tissue has been a problem for most therapeutic applications of these compounds described to date. Recently, in vivo efficacy after the pulmonary administration of a respirable antisense oligodeoxynucleotide (RASON) has been reported from our laboratory. EPI-2010, a phosphorothioate RASON targeting the adenosine A₁ receptor, completely blocked adenosine-induced bronchoconstriction in an allergic rabbit model of human asthma and significantly attenuated both allergen-induced airway obstruction and bronchial hyperresponsiveness (15). This occurred in a dose-dependent and gene-specific fashion, because only the adenosine A₁ receptor, and neither the adenosine A₂ nor bradykinin B₂ receptors, were attenuated. These results provide strong evidence that RASONs are readily absorbed in the lung, are dispersed throughout lung tissues, and specifically attenuate the function of discordantly overexpressed genes (15).

The present study was undertaken to establish the absorption, tissue distribution, metabolism, and excretion of EPI-2010 after aerosol delivery into the rabbit lung. Auxiliary studies to determine if inhaled EPI-2010 had any effects upon a variety of cardiovascular parameters were also performed.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Synthesis of EPI-2010

Unlabeled EPI-2010 drug compound for preclinical pharmacologic studies was synthesized by Boston BioSystems (Bedford, MA). The compound was purified to 97.3% full-length product as measured by capillary gel electrophoresis.

Synthesis of [³⁵S]-labeled EPI-2010

[³⁵S]-labeled EPI-2010 was synthesized by TriLink BioTechnologies, Inc. (San Diego, CA). Two moles of [³⁵S] were added per mole of EPI-2010 near the 5' end of the molecule. It was purified to greater than 95% by reverse-phase high-performance liquid chromatography (HPLC) (Beckman Ultrasphere C-18 column; Irvine, CA; Buffer A = 50 mM triethylamine acetate pH 7.0, Buffer B = acetonitrile; flow rate = 1 ml/ min; column temperature = 50° C). HPLC-purified product was further assessed for purity by 15% polyacrylamide gel electrophoresis (PAGE). Final product contained a total of 1.2 mCi (183.1 OD₂₆₀ units); a specific activity 1.363 Ci/mmoles using an extinction coefficient of 207.9 OD₂₆₀/ µmol; molecular weight 6,954.1. Base composition and sequence were as follows: 5'-GAT[³⁵S] G[³⁵S]GA GGG CGG CAT GGC GGG-3'.

Study Design and Procedures

Five groups of four New Zealand white pasteurella-free normal rabbits (two males and two females each) 14 to 16 wk old (Charles River, Wilmington, MA), weighing between 2.9 and 3.1 kg, were used in the present study. Radioactive EPI-2010 was reconstituted in sterile, injectable grade saline to a concentration of 1.0 µCi (5 µg)/µl. It was dispersed into 50-µCi aliquots and stored at 70° C until use. For each animal, 6,000 µg of unlabeled EPI-2010 was dissolved into 2.45 ml of sterile injectable saline and combined with an aliquot of radiolabeled compound for a final concentration of 6,250 µg/2.5 ml (50 µCi/ animal).

Rabbits were anesthetized and relaxed with 1.5 ml of a mixture of ketamine hydrochloride (35 mg/kg) and acepromazine maleate (1.5 mg/ kg) administered intramuscularly. After induction of anesthesia, the rabbits were laid supine on a soft-molded animal board in a comfortable position. Each animal was then intubated with a 4.0 mm intermediate high-low Murphy 1 endotracheal tube (Mallinkrodt Inc., Glens Falls, NY), as previously described (17). The endotracheal tube was connected to a modified PARI LC Star Nebulizer and PARI compressor air source (PARI Respiratory Equipment, Inc., Midlothian, VA). The nebulizer was modified by replacing the mouthpiece with a section of polypropylene tubing of equal length and diameter, fitted with a connector for the endotracheal tube. The nebulizer exhausted into a 1-L vapor reservoir that was connected to an activated charcoal/Drierite filter to trap exhausted/bypassed radioactivity. The filter was mesh size 8, creating no detectable backpressure to the nebulizer. The rabbits breathed without mechanical assistance from a nebulized atmosphere containing 5,000 µg of EPI-2010 (specific activity 0.055 Ci/mmol) for a period of 10 min. The residual volume of the drug in the nebulizer was measured (approximately 500 µl dead volume in the PARI LC Star). Two 10-µl aliquots, representing 25 µg of EPI-2010, were stored at 70° C until they were counted with the corresponding animal samples as a reference standard. The animals were housed in individual stainless steel metabolism cages designed to permit the separate collection of urine and feces (Allentown Caging Equipment Company, Allentown, NJ). The rabbits were allowed food and water ad libitum. Each group of four rabbits was killed by intravenous injection of Somlethol (24 mg/kg; J.A. Webster, Inc., Sterling, MA) at 0, 6, 24, 48, and 72 h after drug inhalation. For each group, urine, feces, and selected tissues were collected and assessed for distribution of [³⁵S]-EPI-2010.

Sample Collection

Plasma. Blood (2 to 3 ml) was collected at 0, 6, 24, 48, and 72 h from the central ear artery of the rabbits in a heparinized tube. Plasma was separated and stored at 70° C until analysis.

Urine and feces. Total urine and fecal samples were quantitatively collected in 0-6, 6-24, 24-48, and 48-72 h time intervals. Urine samples were collected in amber plastic containers over ice packs and preserved using Stabilur tablets (R. P. Cargille Laboratories, Inc., Cedar Grove, NJ). All samples were stored at 70° C until analysis.

Tissue. Immediately after killing the animals, the following tissues were collected for analysis: esophagus, trachea, lung, heart, liver, and kidney. The collected tissues were cleaned, rinsed with saline, blotted dry, and weighed.

Radioactive Sample Analysis

Plasma. Plasma samples were mixed by vortexing and duplicate aliquots (0.1 ml) were transferred to scintillation vials. Ultima-Gold scintillation cocktail (6 ml; Packard Instrument Co., Meriden, CT) was added and samples were analyzed for radioactivity for 5 min.

Urine. Each urine sample was thoroughly mixed and duplicate aliquots (0.1 ml) were processed and analyzed in the same manner described previously.

Tissues. The weighed tissues were homogenized with a polytron into a 20% solution with sterile saline. Immediately thereafter, 0.5-ml aliquots (100 mg tissue) in duplicate were solubilized with 1 ml of a mixture of Soluene-350/isopropyl alcohol (1:1, vol/vol) and incubated at 60° C for 30 min with constant shaking. After 30 min, the samples were cooled to room temperature and 0.5 ml of 30% H₂O₂ was added to each sample with constant shaking to decolorize the samples. The samples were analyzed for radioactivity after the addition of 15 ml of Ultima-Gold scintillation cocktail.

Feces. Fecal samples were processed and analyzed in the same manner as described previously for the tissue samples.

Tissue Extraction and 3' Postlabeling

Tissue homogenates, plasma, and urine (200 µl) were digested with 1 mg proteinase K in 0.5% sodium dodecyl sulfate (SDS), 10 mM NaCl, 20 mM Tris (pH 7.6), 10 mM ethylenediaminetetraacetic acid (EDTA) for 2 h at 60° C. The digests were extracted with phenol then chloroform before they were precipitated with glycogen, sodium acetate, and ethanol. Each precipitate was washed with 70% ethanol before being dried and resuspended in 30 µl of water. Samples (4 µl) were 3'-labeled with 20 units of terminal deoxynucleotidyl transferase and 2 µCi of [-³²P]-deoxyadenosine 5'-triphosphate in a buffer of 100 mM cacodylate (pH 6.8), 1 mM CoCl₂, and 0.1 mM dithiothreitol (DTT) at 37° C for 30 min. All samples were treated with ribonuclease (RNase) A followed by ethanol precipitation before PAGE analysis (20% acrylamide, 7 M urea) and autoradiography (18).

Autoradiography of Lung Sections

[³⁵S]-EPI-2010 (specific activity 0.055 Ci/mmol) was administered in one additional New Zealand white rabbit weighing 3.0 kg as previously described. The animal was killed at 6 h after inhalation of the drug, and lungs were flash-frozen in isopentane cooled to 30° C on dry ice for 30 s and stored at 70° C until processed. The frozen lung was subdivided before sectioning in a cryostat. Unfixed, 20-µ-thick sections were mounted on slides and exposed to Bio Max film (Eastman Kodak, Rochester, NY) for 14 d. The film was developed with Dektol and fixed with Kodak fixer. Film images were then photographed.

Measurement of Cardiovascular Function

Cardiomax-II procedure. Twenty-one New Zealand white rabbits were administered aerosolized saline, low- or high-dose EPI-2010 (2 mg and 20 mg, respectively), or mismatch control (5'-GTAGGTGGCGGGCAAGGCGGG-3') by endotracheal tube (n = 7, 4, 5, and 5, respectively). We have previously demonstrated that this dosing regimen resulted in the inhibition of adenosine- and allergen-induced asthmatic responses in allergic rabbits (15). Saline vehicle, 0.5, or 5.0 mg of EPI-2010, or 5.0 mg of mismatch control was nebulized for 10 min twice a day for 2 d (total dose EPI-2010 nebulized per animal, 2.0 or 20.0 mg). Sixteen to 18 h after the last dose of EPI-2010, animals were anesthetized using ketamine and xylazine (80 mg/ml, 20 mg/ml, respectively), and the cardiovascular functions of the rabbits were measured as described earlier (19).

Data and Statistical Analysis

All samples were analyzed for 5 min in a scintillation counter (Model LS3801; Beckman Instruments, Inc., Irvine, CA). Scintillation counting data counts per minute (cpm) were converted to disintegrations per minute (dpm) using an instrument stored quench curve. Tissue, plasma, urine, and fecal dpms were corrected for either weight or volume to represent the total dpms for the indicated organ, entire animal plasma volume, total urine or total fecal output over the indicated time period, respectively. The dpms were then converted to microgram equivalents with the dpms derived from the reference standards collected for each animal.

All data are expressed as mean ± SEM. Data were analyzed with SSPS software (SigmaStat version 2.03) using one-way analysis of variance (ANOVA) and multiple comparison (Bonferroni t test) when appropriate. In all cases, p values less than 0.05 were considered statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Kinetics of EPI-2010 in Tissue

The concentration of EPI-2010-derived radioactivity was determined in various organs at 0, 6, 24, 48, and 72 h after aerosol delivery via nebulizer attached to an endotracheal tube. Figure 1 illustrates the distribution profile of EPI-2010 in various tissues at these times. Based on radioactivity, approximately 1.4% of the total amount of the aerosolized EPI-2010 was delivered into the lung of the rabbits by passive respiration. The total concentrations of the drug in the lung at 0, 6, 24, 48, and 72 h were 64.0 ± 1.5, 67.0 ± 4.4, 32.0 ± 3.7, 23.4 ± 1.4, and 2.1 ± 0.5 µg equivalent/organ, respectively. The elimination half-life (t_1/2) for EPI-2010 in the lung was calculated to be 30 h (Figure 1, insert). Approximately 3% of the delivered drug remained in the lung after 72 h of administration. A small amount of radioactivity was also detected in other organs. Extrapulmonary tissue concentrations peaked at 6 h. The [³⁵S]-radioactivity found in the heart, liver, and kidney combined (derived from EPI-2010 and its derivatives) was only 6.9% of the total radioactivity associated with the administered dose. The [³⁵S] radioactivity found in the heart ranged from 0.18 ± 0.02 µg equivalents to 0.03 ± 0.02 µg equivalent of EPI-2010-derived radioactivity at 0 and 72 h, respectively. Because the organs were not perfused, the residual low radioactivity in the organs may be from the blood.

View larger version (26K): [in this window] [in a new window]   Figure 1.   Distribution profile for EPI-2010 in tissue. The tissue concentration of EPI-2010-derived radioactivity is based on µg equivalent/organ and is the mean of four rabbits (two male and two female) for each tissue. Inset is a linear regression graph of the lung concentrations of EPI-2010 over 72 h. The graph indicates that the half-life (T1/2) for EPI-2010-derived radioactivity in lung tissue is 30 h.

Kinetics of EPI-2010 in Plasma

A small amount of inhaled EPI-2010 was detected in plasma at every time point. Approximately 3.0 ± 0.7, 5.3 ± 0.3, 10.3 ± 1.6, 8.8 ± 1.3, and 3.2 ± 2.0 µg equivalent/total plasma volume were present in the rabbit at 0, 6, 24, 48, and 72 h, respectively (Figure 2). With inhalation, maximal concentration achieved in plasma was 10.3 µg equivalent/total plasma volume at 24 h after drug administration, representing less than 15.0% of the radioactivity associated with the administered dose.

View larger version (30K): [in this window] [in a new window]   Figure 2.   Plasma profile for EPI-2010. Plasma concentrations are the mean values from all samples collected at each time point throughout the study. The plasma concentrations are adjusted to represent the entire plasma volume based on the animal's weight and a mean plasma volume of 38.8 ml/kg body weight.

Kinetics of EPI-2010 in Urine and Feces

The majority of the drug (67.5%) was eliminated by 72 h. As shown in Figure 3, urinary excretion represented the major pathway of elimination of inhaled EPI-2010. The concentrations of drug in the total urine collected at 6, 24, 48, and 72 h were 4.7 ± 1.3, 13.4 ± 1.0, 17.9 ± 2.1, and 12.2 ± 4.6 µg equivalent, respectively. Approximately 60% of the delivered dose was excreted in urine over a period of 72 h. Excretion in feces represented a minor route of RASON elimination with 7% eliminated over 72 h. In both cases, excretion of EPI-2010 was time-dependent, being minimum during the first 6 h, peaking at 48 h, and then declining.

View larger version (24K): [in this window] [in a new window]   Figure 3.   Elimination profile for EPI-2010. The urine and fecal concentrations are mean values calculated for the total output from all rabbits collected during the following time points: 0-6, 6-24, 24-48, and 48-72 h.

Evaluation of EPI-2010 Integrity by 3' Postlabeling

Analysis of 3'-postlabeled EPI-2010 revealed intact EPI-2010 present only in the lung from 0 through 72 h. Some metabolites of EPI-2010 were also detected in the lung at 24, 48, and 72 h. Neither intact EPI-2010 nor its degradation products were detected in any other tissue or samples using this method (data not shown).

Autoradiography of Lung Section

As seen in Figure 4, autoradiographic analysis of the lung demonstrated a clear deposition of [³⁵S]-EPI-2010 throughout the large and small airways of normal rabbit lung. The distribution throughout the upper right and left lobes was of a similar intensity whereas delivery to the lower right lobe was uniform but less intense (data not shown).

View larger version (84K): [in this window] [in a new window]   Figure 4.   Autoradiography of lung sections after inhalation of [35S]- EPI-2010. Sections from the left lobe (A) and right upper lobe (B) demonstrate that the RASON was deposited in the large and small airways of the rabbit lung. Lung section panels each represent an area 3.4 × 5.2 cm.

Effect of Aerosolized EPI-2010 on Cardiovascular Functions

After both low (0.5 mg twice daily for 2 d) and high (5.0 mg twice daily for 2 d) aerosol delivery of EPI-2010 into the lung of the rabbits, no significant changes in cardiac output, stroke volume, mean arterial pressure, heart rate, total peripheral resistance, and cardiac contractility were observed (Figure 5).

View larger version (33K): [in this window] [in a new window]   Figure 5.   Cardiovascular safety profile of different doses of inhaled EPI-2010 in rabbits. The effects of EPI-2010 were evaluated on cardiac output (CO, ml/min), stroke volume (SV, µl/beat), mean arterial pressure (MAP, mm Hg), heart rate (HR, beats/min), total peripheral resistance (TPR, mm Hg/min/100 g), and cardiac contractility (dP/dT). No significant difference was found between saline and the treatment groups using one-way ANOVA and multiple comparison (Bonferroni t test) when appropriate.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The data presented in our earlier work showed that the adenosine A₁ receptor could be effectively targeted in the lung by the RASON approach (15). In the present studies, we demonstrate that RASONs have a fairly circumscribed delivery to the lung, at least when administered in low dose. Radiolabeled EPI-2010 delivered by nebulization produced uniform deposition of drug within the lung. There was little extrapulmonary tissue distribution with peak tissue concentrations, as measured by radioactivity, remaining less than 7% of the administered dose. This is in contrast to intravenous administration performed in other species where the highest concentrations of ASONs were found in liver and kidney (18, 20, 21). The postlabeling studies add support to this finding, because intact EPI-2010 was only detectable in lung tissues. In the case of intravenous or intratracheal administration of ASONs, intact drug was found in several tissues including kidney, liver, spleen, lymph nodes, lung, and plasma (18, 20). It should be noted that the delivered dosage of EPI-2010 for this study was in the effective therapeutic range and much lower than those used in most of the cited studies. These data support the finding that low doses of RASONs are effectively confined to the lungs or that they leave in very low amounts or in a degraded or modified form.

The plasma profile for pulmonary administration differs from the typical biphasic profile (rapid distribution phase followed by a prolonged elimination phase) seen with intravenous administration (21, 22). Similar results were seen with low-dose intratracheal administration in that pulmonary delivery resulted in a slower distribution phase with a much lower systemic availability, as compared with intravenous routes, followed by a prolonged elimination phase (22). In general, the excretion pathway of aerosolized EPI-2010 in the rabbits was very similar to other routes of administration reported for phosphorothioate ASONs in mice, rat, and monkey (18, 20, 23).

Results from cardiovascular studies provide evidence that inhaled EPI-2010 does not escape the lung in amounts sufficient to compromise the cardiovascular system. If this lack of extrapulmonary deposition of inhaled RASONs translates to reduced systemic toxicity, RASONs may provide an entirely new, safe and effective approach to the treatment of airway diseases. Antisense oligonucleotides are a promising new class of therapeutics that offers great potential when problems of their delivery to target tissues can be overcome. Local delivery of antisense oligonucleotides to the target tissue appears to offer significant advantages over systemic delivery, for reasons which may include (1) unavoidable drug dilution during systemic delivery versus high local concentrations during local delivery (22); (2) complexing with serum proteins during systemic delivery that prevents cellular uptake; (3) high dose requirement for systemic delivery which equates to greater potential for toxicity (24, 25), versus efficacy at much lower doses for local delivery and consequent reduced potential for toxicity. Fomivirsen, an antisense oligonucleotide targeting drug-resistant cytomegalovirus retinitis, has received approval from the Food and Drug Administration. Fomivirsen, it is important to note, is designed for local delivery, by direct injection into the vitreous humor of the eye (12).

As reported earlier (15), EPI-2010 reduced sensitivity to applied adenosine in a dose-dependent manner over the dose range of 0.2 mg total dose (provocative concentration of adenosine causing a 50% reduction in FEV₁ [PC₅₀], 8.32 ± 7.2 mg/ ml), 2.0 mg total dose (PC₅₀ adenosine, 14.0 ± 2.7 mg/ml), and 20.0 mg total dose (PC₅₀ adenosine, 19.5 ± 0.34 mg/ml) as compared with mismatch control, 20 mg total dose (PC₅₀ adenosine, 3.25 ± 3.4 mg/ml). Based on the total doses of 2 and 20 mg and the lung deposition reported here, the effective delivered pulmonary dose for EPI-2010 can be calculated to fall between 10 and 100 µg/kg. This is dramatically lower than for most ASONs reported in the literature. Therefore, the low dose efficacy, which may be related to local delivery or the presence of surfactants in the airways, makes the RASON approach to the therapy of lung diseases very appealing. These factors may permit the development of respiratory therapeutics with much improved toxicity profiles as compared with conventional drugs.