| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of
*
Molecular and Cellular Pharmacology and
Pharmacology, Isis Pharmaceuticals, Carlsbad, CA 92008; and
Novartis-Horsham Research Centre, Horsham, United Kingdom
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although several cytokines and chemokines have been demonstrated to
affect eosinophil maturation, tissue infiltration, degranulation, and
survival, IL-5 appears to be critical for producing a specific tissue
eosinophilia. It has been demonstrated that mRNA for IL-5 is
up-regulated within the airways of both atopic and nonatopic asthmatic
patients and that IL-5 expression inversely correlates with pulmonary
function (8, 9). In addition, IL-5 expression directly
correlates with the number of eosinophils detected in asthmatic airways
(10) and is decreased after corticosteroid treatment
(11). IL-5 has been shown to be selectively elevated in
helminth-infected patients with eosinophilia (12) and a
role for IL-5 in human eosinophilia has also been suggested in both
idiopathic hypereosinophilic syndrome (13) and
eosinophilia-myalgia syndrome (14). Several lines of
evidence also implicate IL-5 as the central cytokine for producing
eosinophilia in mice: 1) inhibition of IL-5 responses with neutralizing
Ab prevents blood and tissue eosinophilia in response to helminth
infection or exposure to Ag (15, 16); 2) transgenic mice
overexpressing IL-5 develop a selective eosinophilia (17);
3) IL-5-deficient mice do not respond with increased numbers of
eosinophils on Ag challenge (18, 19); and 4) mice
deficient in the IL-5 receptor 
subunit required for binding IL-5
are unable to respond to parasitic infections with increased numbers of
eosinophils (20). Collectively, these observations suggest
that inhibition of IL-5 expression would result in impaired eosinophil
infiltration into the asthmatic lung, thus effectively reducing the
damage produced by eosinophilic components and thereby improving the
clinical symptoms.
As an alternative to searching for small molecule inhibitors that prevent the interaction of IL-5 with its receptor, we used an antisense oligonucleotide approach to inhibit IL-5 gene expression. Antisense oligonucleotides that support RNase H-mediated degradation of murine IL-5 mRNA were developed using an in vitro model system of inducible IL-5 expression. After characterization in vitro, a lead oligonucleotide was tested in the OVA lung challenge and OVA-induced peritonitis mouse asthma models as well as in mouse airway hyperresponsiveness studies. Here we report the development and characterization of a murine IL-5 antisense oligonucleotide that reduces IL-5 protein expression and effectively inhibits Ag-induced eosinophil infiltration and airway responses in vivo in a manner that is consistent with an antisense mechanism of action.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The murine thymoma cell line EL-4 was obtained from the American Type Culture Collection (Manassas, VA) and was maintained and stimulated to express IL-5 as previously described (21).
Oligonucleotide synthesis and cell transfection
2'-O-Methoxyethylribose
(MOE)2-modified
phosphorothioate oligonucleotides were synthesized and purified as
described previously (22). Chimeric oligonucleotides
containing 2'-O-methoxyethyl-modified residues flanking a
2'-deoxynucleotide/phosphorothioate region that supports RNase H
activation were used in all experiments (22, 23, 24).
Sequences of the oligonucleotides used are shown in Table I
. EL-4 cells (1 x
107 cells in PBS) were transfected with
oligonucleotides by electroporation, at 200 V, 1000 µF, 13 ohms using
a BTX Electro Cell Manipulator 600 (Genetronics, San Diego,
CA).
|
Male BALB/c-ByJ mice were immunized i.p. with 20 µg OVA in aluminum hydroxide adjuvant, reimmunized 2 weeks later, and challenged with 2% OVA aerosol for 1 min on days 20, 21, and 23. Oligonucleotides were dissolved in saline and injected daily i.v. in the tail vein by bolus infusion at the indicated doses from 8 days after Ag sensitization (day 8) through day 21. For the allergic peritonitis model, male BALB/c mice were immunized s.c. with 20 µg OVA as above on days 0 and 7 and challenged i.p. with 10 µg OVA on day 14. Oligonucleotides were injected i.v. in the tail vein as above either from day -2 through day 14 or from day -2 through day 5, or as indicated.
ELISA for IL-5
Murine IL-5 was measured using a sandwich ELISA (Endogen, Woburn, MA).
Cell differentials
After euthanasia, tracheal cannulation or peritoneal lavage was performed, and saline washes were collected for processing cell counts. Eosinophils were enumerated based on morphology and staining characteristics and expressed as percentages of total cells at the Allergic Disease Center, Creighton University (Omaha, NE), or as total eosinophils.
Determination of airway physiology
Pulmonary function was monitored using a Buxco BioSystem Plethysmograph (Buxco, Troy, NY) at the Allergic Disease Center, Creighton University, and expressed as enhanced pause (Penh), which correlates to measured airway resistance (25). After the last challenge with aerosolized OVA, pulmonary function recordings were performed for 30 min to examine the early phase allergic response. For the late phase reaction, recordings were subsequently performed hourly from 2 to 9 h after OVA challenge. As a positive control, dexamethasone was administered i.p. (2.5 mg/kg) 1 day before the sensitization, again 2 h before the challenge, and 18 h after the challenge.
Statistics
Analysis of dose-response curves and airway physiological data was performed using ANOVA with a post hoc Fisher’s protected least significant difference test. Values for all measurements are expressed as the mean ± SEM. Statistical analysis was conducted using SuperANOVA (Abacus Concepts, Berkeley, CA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To identify an antisense oligonucleotide against murine IL-5, we utilized the EL-4 thymoma, which has been shown to express IL-5 mRNA and protein after treatment with phorbol ester and cyclic AMP-elevating agents (21) as an in vitro screening model. An active 20-base oligonucleotide complementary to sequence within the 3'-untranslated region of the murine IL-5 mRNA was identified. This oligonucleotide is a chimeric molecule containing a uniform phosphorothioate backbone and a stretch of ten 2'-deoxy residues in the center of the molecule which supports RNase H-mediated cleavage. The 2'-deoxy core is 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 (22, 23, 24). The cytosines within the sequence were methylated at the 5-position to reduce the immunostimulatory potential sometimes observed with oligonucleotides in rodent studies.
The IL-5 antisense oligonucleotide decreased PMA plus
dibutyryl-cAMP-stimulated IL-5 mRNA expression after electroporation
into EL-4 cells in a dose-dependent manner, whereas the same sequence
carrying five base mismatches was without effect (Fig. 1A), as determined by Northern
blot. To further confirm an antisense mechanism of IL-5 mRNA reduction,
the active sequence was tested for potency in reducing IL-5 mRNA while
containing one, three, or five base mismatches. Activity was
compromised after incorporation of a single base mismatch (Fig. 1B) and was completed abrogated by mismatching three or five
bases in the sequence, indicating the importance of RNA hybridization
for this effect.
|
A), whereas the five-base
mismatch control had no effect. Antisense-mediated reduction of IL-5
protein expression was long-lived, with potent suppression of the
response apparent up to 72 h after a single treatment (Fig. 2B).
|
Administration of the IL-5 antisense oligonucleotide i.v. in
BALB/c mice beginning 8 days after Ag sensitization and continuing
daily through day 21, with i.p. Ag boost occurring on day 14 and
repeated aerosol lung challenges on day 20, 21, and 23, resulted in
potent inhibition of lung eosinophilia (Table II), similar to that produced by
dexamethasone. A control oligonucleotide that contained five base
mismatches did not show significant activity in this model, indicating
that this effect is sequence specific and consistent with an antisense
mode of action. This effect was maintained up to 17 days after the last
dose of oligonucleotide, with lung eosinophils reduced from 41 ±
4.5% of the total BAL cells in OVA-sensitized and
-challenged mice to 27 + 2.0% in OVA near-scrambled control
oligonucleotide-treated mice, whereas OVA IL-5 antisense treatment
resulted in 7.6 ± 1.3% (p 
0.01)
eosinophils, and OVA dexamethasone treatment i.p. at 2.5 mg/kg 6 h
before challenge reduced eosinophils to 7.5 ± 1.5%
(p 0.01).
|
). The total eosinophil count in
peritoneal washout fluid was significantly reduced in IL-5
antisense-treated mice compared with mismatch oligonucleotide-treated
animals. Eosinophil peroxidase levels measured by ELISA were also
correspondingly reduced in antisense but not mismatch control
oligonucleotide-treated animals (data not shown). Thus, inhibition of
Ag triggered eosinophilia has been observed in two separate in vivo
inflammatory models.
|
The IL-5 antisense oligonucleotide does not contain any known immunostimulatory sequence motifs. One potential explanation for oligonucleotide-mediated reduction of eosinophilia is the induction of a Th1 immune response that may counteract an ongoing Th2 cytokine profile characteristic of allergic inflammatory states (26). To test for potential immunostimulatory activities of this oligonucleotide in the experimental models used in this study, BALB/c mice were dosed i.v. with either a CpG-containing oligonucleotide, the IL-5 oligonucleotide, or the mismatch control oligonucleotide and cytokine gene expression and spleen weight subsequently examined. After four consecutive daily doses of the IL-5 oligonucleotide at 10 mg/kg, no change in Th1 or Th2 cytokine gene expression was observed using a RNase protection assay (RPA), whereas the CpG-containing oligonucleotide induced a predominantly Th1 cytokine gene response; dosing throughout the immunization schedule (daily, day -2 to day 14) with the IL-5 oligonucleotide did not result in splenomegaly, whereas the CpG oligonucleotide almost tripled spleen weight (data not shown). In other experiments, the IL-5 antisense oligonucleotide was further shown not to affect mRNA levels of IL-10 or IL-3 (data not shown), suggesting that targeting was highly specific.
IL-5 protein knock-down in vivo
We next examined whether IL-5 mRNA and protein levels were
decreased in vivo in antisense-treated animals. IL-5 mRNA levels were
evaluated either by RPA of whole spleen RNA or by quantitative RT-PCR
analysis of poly(A)-purified mRNA from splenic T cells isolated using
Thy1.2 Ab-conjugated magnetic beads. In both instances, although IL-5
mRNA levels were decreased in animals dosed with the IL-5 antisense
relative to OVA or to either mismatch or control
oligonucleotide-treated animals (data not shown), the low level of IL-5
mRNA expression complicated analyses. Reductions on the order of
50–60% compared with the five-base mismatch control were observed by
quantitative RT-PCR analysis, and 
30% by RPA; however, the inherent
variability of the data due to the low abundance of IL-5 mRNA in
recovered splenocytes resulted in lack of statistical significance.
Analysis of lymph node or lung tissue samples by RPA yielded similar
results.
Reduction of IL-5 protein levels was also assessed in antisense-treated
animals, by using an IL-5-specific sandwich ELISA. In these studies,
IL-5 protein was measured in peritoneal washout fluid from OVA-primed
and -challenged mice. IL-5 protein levels were significantly inhibited
after i.v. administration of IL-5 antisense oligonucleotide from day
-2 through day 5, or from day 0 through day 7 (Fig. 4). This effect occurred to a lesser
degree after IL-5 antisense oligonucleotide dosing on days -2 and -1
only. Additional experiments showed that this response was dose related
(data not shown). Collectively, these data suggest that antisense
targeting in vivo results in measurable reduction of IL-5 protein
levels and that the observed inhibition of eosinophilia in vivo is due
to decreased IL-5 expression.
|
Allergen-induced pulmonary eosinophilia has been correlated with
late phase airway hyperresponsiveness in humans (27, 28),
although this correlation is less robust in Ag-induced animal models of
asthma (29, 30). Therefore, blocking the influx of
eosinophils into the lung may be expected to result in amelioration of
the late phase response. We thus examined whether IL-5 antisense
oligonucleotide-mediated inhibition of OVA-induced lung eosinophilia
was correlated to improvement in allergic airway physiological
performance using a whole body plethysomography mouse model. With the
use of BALB/c mice and the same schedule and route of antisense
oligonucleotide administration as the eosinophilia experiments
presented in Table II
, airway hyperresponsiveness was evaluated after
OVA rechallenge. Penh was recorded during both
the early phase (up to 30 min after Ag challenge) and the late phase
(from 2 until 9 h after OVA challenge) airway responses (Fig. 5A). Although there were no
significant effects of IL-5 oligonucleotide treatment on the early
phase response, as expected, a dose-related decrease in the late phase
allergic airway response was observed (Fig. 5B). Significant
decreases in late phase hyperreactivity of the airways were noted in
IL-5 antisense oligonucleotide-treated mice at both 10- and 20-mg/kg
doses compared with the mismatch control-treated and the saline-treated
groups (Fig. 5, C and D), suggesting that IL-5 is
a contributing factor to the late phase allergic airway response in
this model.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although IL-5 antisense did inhibit the late phase airway response to
OVA rechallenge in a dose-related manner, complete abrogation of this
response was not observed (Fig. 5). The initiation of an IL-5-dependent
or -independent response in BALB/c mice has been suggested to depend on
the sensitization protocol utilized (32). Specifically,
s.c. sensitization as performed by Corry et al. (31)
resulted in IL-5-independent airway responses while intraperitoneal or
airway sensitization yielded hyperreactivity responses with an
IL-5-dependent component (32, 35, 39). In addition, a
recent study evaluating respiratory delivery of anti-IL-5 in BALB/c
mice also showed an IL-5-dependent effect on airway hyperreactivity
after i.p. sensitization and aeroallergen challenge (40).
Thus, available data indicate that an IL-5-dependent allergic airway
response can be measured in BALB/c mice under conditions of
sensitization similar to those used in this study and suggest that
multiple pathways of airway hyperreactivity can occur in BALB/c mice,
some of which do not involve IL-5. Earlier studies have found similar
induction of airway hyperreactivity in methacholine-treated
anesthetized and conscious mice, utilizing either airway constriction
measurements in cannulated mice or whole body plethysmography,
respectively (32). However, the relative importance of
various mechanisms of airway hyperreactivity may also differ in these
two experimental systems and therefore contribute to discordant
observations on the role of IL-5 in murine models of asthma. The
contribution of non-IL-5-mediated airway hyperreactivity in our model
may account for the lack of complete suppression of airway
hyperresponsiveness, despite dramatic suppression of lung
eosinophilia. In light of current knowledge on the differential
manifestation of airway hyperresponsiveness in BALB/c mice, the lack of
a correlation between eosinophilia and airway reactivity in some asthma
patients is perhaps not surprising and suggests that additional
pathways distinct from IL-5 may also control airway responsiveness in
some instances in humans.
Our in vivo target reduction experiments suggest that oligonucleotide
treatment is most efficacious when administered early in the
immunization period (at or before Ag sensitization) and for more than
two consecutive daily doses (at 20 mg/kg; Fig. 4). In vivo distribution
studies have shown that appreciable bone marrow uptake occurs after
i.v. administration of oligonucleotide in rats (41, 42)
and, combined with the improved nuclease resistance of the 2'-MOE
chemistry used in the 5'- and 3'-ends of the oligonucleotide (23, 24), these data suggest that one site of action is likely to be
the bone marrow. Recent evaluation of oligonucleotide uptake in various
lymphoid tissues of mice by capillary gel electrophoresis analysis of
i.v. injected oligonucleotide has shown that phosphorothioate and
2'-MOE-modified oligonucleotides accumulated chiefly in lymph node and
bone marrow and that the 2'-MOE chemistry used in the experiments
presented herein was recovered from these tissues largely intact
(K. J. Myers and M. J. Graham, unpublished observations).
Intranasal administration of the IL-5 oligonucleotide in mice at 24 and
2 h before challenge failed to reduce eosinophil infiltration in
the OVA-induced lung model (data not shown). The failure of intranasal
administration of the oligonucleotide to reduce OVA-induced lung
eosinophilia may be due to poor oligonucleotide uptake by mature T
cells in the lung, given that studies have shown poor staining of
mature peripheral T cells with fluorescently labeled oligonucleotides
(43, 44).
In vivo application of antisense technology for inhibition of genes linked to asthma pathology has been limited. In addition to adenosine A1 receptor targeting in a rabbit model of asthma (38), Molet et al. have described the lack of activity of an IL-5 antisense oligonucleotide on the late phase allergic response in adoptive transfer experiments (45). The disagreement between the observations made in this study and ours, we believe, most likely represent differences in the site(s) of action of the antisense in vivo vs T cell transfection ex vivo. In particular, the lack of effect following intranasal administration of the IL-5 oligonucleotide in mice in our study suggests that it may be important to target IL-5-producing cells in vivo that are perhaps at a more immature stage than those recovered from lymph node after Ag sensitization. Thus, an IL-5 antisense strategy that targets the bone marrow tissue directly may be more effective than an approach aimed only at peripheral T cells. Further, the antisense treatment schedule may be important for in vivo activity as well, because we observed less potent activity when administering the antisense compound after the Ag challenge in our immunization model (unpublished observation).
In humans, IL-5 is thought to be specific for promoting eosinophil and basophilic maturation (46) and eosinophil survival and chemotaxis (47). Although discovered as a B cell growth and differentiation factor in mice, its actions on human B cells have been controversial but appear to be considerably more limited. Human IL-5 has no activity in standard human B cell growth factor assays (48) and appears to be costimulatory only with restricted mitogens or with the addition of IL-2 or IL-4 (49, 50, 51). These data suggest that selectively blocking the function of human IL-5 in vivo would predominantly affect eosinophils and avoid the side effects that are currently encountered with the broad acting antiinflammatory steroids.
In summary, we have shown that treatment of mice with an IL-5 antisense oligonucleotide results in decreased eosinophil infiltration into the lung and peritoneal cavity and improved airway function in mouse models of asthma. These pharmacological effects correlated with a significant reduction of IL-5 protein levels in vivo, supporting an antisense mechanism of action. Importantly, IL-5 blockade with an antisense oligonucleotide imparted significant protection from late phase airway hyperresponsiveness as measured in a whole body plethysmography model, suggesting a connection between IL-5-mediated eosinophilia and reduced airway performance in this setting. Moreover, few therapeutics currently available for asthma characteristically ameliorate the late phase reaction; thus, it is important to further characterize the dosing schedule required for this effect and the efficacy of other routes of administration and to determine the duration of action of this oligonucleotide-mediated effect.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Abbreviations used in this paper: 5 Me-C, 5-methylcytosine; 2'-MOE, 2'-O-methoxyethylribose; Penh, enhanced pause; RPA, RNase protection assay; UTR, untranslated region.
Received for publication November 22, 1999. Accepted for publication March 6, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
cytokine gene expression. Am. Rev. Respir. Dis. 148:401.[Medline]
-deficient mice. Immunity 4:483.[Medline]
This article has been cited by other articles:
|
|
 |
|
  R. S. Peebles Jr, K. Hashimoto, J. R. Sheller, M. L. Moore, J. D. Morrow, S. Ji, J. A. Elias, K. Goleniewska, J. O'Neal, D. B. Mitchell, et al. Allergen-Induced Airway Hyperresponsiveness Mediated by Cyclooxygenase Inhibition Is Not Dependent on 5-Lipoxygenase or IL-5, but Is IL-13 Dependent J. Immunol., December 15, 2005; 175(12): 8253 - 8259. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. C. Dorman, R. Sehmi, G. M. Gauvreau, R. M. Watson, R. Foley, G. L. Jones, J. A. Denburg, M. D. Inman, and P. M. O'Byrne Kinetics of Bone Marrow Eosinophilopoiesis and Associated Cytokines after Allergen Inhalation Am. J. Respir. Crit. Care Med., March 1, 2004; 169(5): 565 - 572. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. K. Farraj, J. R. Harkema, T.-R. Jan, and N. E. Kaminski Immune Responses in the Lung and Local Lymph Node of A/J Mice to Intranasal Sensitization and Challenge with Adjuvant-Free Ovalbumin Toxicol Pathol, June 1, 2003; 31(4): 432 - 447. [Abstract] [PDF]
|
|
|
|
|
|
V. del Pozo, M. Rojo, M. L. Rubio, I. Cortegano, B. Cardaba, S. Gallardo, M. Ortega, E. Civantos, E. Lopez, C. Martin-Mosquero, et al. Gene Therapy with Galectin-3 Inhibits Bronchial Obstruction and Inflammation in Antigen-challenged Rats through Interleukin-5 Gene Downregulation Am. J. Respir. Crit. Care Med., September 1, 2002; 166(5): 732 - 737. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. H. Mohamed, A. I. Abdelhamid, Y. C. G. Lee, K. B. Lane, B. Conner, M. Hawthorne, and R. W. Light Pleural Fluid Levels of Interleukin-5 and Eosinophils Are Closely Correlated* Chest, August 1, 2002; 122(2): 576 - 580. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. Allakhverdi, M. Allam, and P. M. Renzi Inhibition of Antigen-induced Eosinophilia and Airway Hyperresponsiveness by Antisense Oligonucleotides Directed against the Common beta Chain of IL-3, IL-5, GM-CSF Receptors in a Rat Model of Allergic Asthma Am. J. Respir. Crit. Care Med., April 1, 2002; 165(7): 1015 - 1021. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. M. Fleming, H. He, A. Ciota, D. Perkins, and P. W. Finn Administration of Pentoxifylline During Allergen Sensitization Dissociates Pulmonary Allergic Inflammation from Airway Hyperresponsiveness J. Immunol., August 1, 2001; 167(3): 1703 - 1711. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.C. Kips, K.G. Tournoy, and R.A. Pauwels New anti-asthma therapies: suppression of the effect of interleukin (IL)-4 and IL-5 Eur. Respir. J., March 1, 2001; 17(3): 499 - 506. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Lach-Trifilieff, R. A. McKay, B. P. Monia, J. G. Karras, and C. Walker In Vitro and In Vivo Inhibition of Interleukin (IL)-5-Mediated Eosinopoiesis by Murine IL-5R{alpha} Antisense Oligonucleotide Am. J. Respir. Cell Mol. Biol., February 1, 2001; 24(2): 116 - 122. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, No oligonucleotide control; 
, IL-5
antisense; 
, mismatch control. *, Statistical significance
compared with mismatch control, p 
