sustained inhibition of il-6 and il-8 expression by decoy odn to nf-κb delivered through respirable...

THE JOURNAL OF GENE MEDICINE R E S E A R C H A R T I C L EJ Gene Med 2011; 13: 200–208.Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/jgm.1546

Sustained inhibition of IL-6 and IL-8 expressionby decoy ODN to NF-κB delivered throughrespirable large porous particles inLPS-stimulated cystic fibrosis bronchial cells

Daniela De Stefano1†

Francesca Ungaro2†

Concetta Giovino2

Antonio Polimeno1

Fabiana Quaglia2

Rosa Carnuccio1∗

1Department of ExperimentalPharmacology, School ofBiotechnological Sciences, Universityof Naples Federico II, Naples, Italy2Department of Pharmaceutical andToxicological Chemistry, School ofPharmacy, University of NaplesFederico II, Naples, Italy

∗Correspondence to:Rosa Carnuccio, Department ofExperimental Pharmacology,University of Naples Federico II,Via D. Montesano 49, 80131Naples, ItalyE-mail: [email protected]

†Both investigators contributedequally and should beconsidered as senior authors

Received : 12 November 2010Revised: 18 January 2011Accepted: 27 January 2011

AbstractBackground Cystic fibrosis (CF) is caused by mutations in the cysticfibrosis transmembrane conductance regulator (CFTR) gene. Neutrophil-dominated inflammation and chronic bacterial infection are still consideredthe primary cause of bronchioectasis, respiratory failure and consequentdeath in CF patients. Activation of nuclear factor (NF)-κB is responsible foroverproduction of cytokines, such as interleukin (IL)-6 and IL-8, in airwaysof CF patients. Thus, decoy oligodeoxynucleotides against NF-κB (dec-ODN)may limit lung inflammation in CF. In the present study, we studied theeffects of dec-ODN delivered through biodegradable and respirable poly(D,L-lactide-co-glycolide) large porous particles (LPP) on IL-6 and IL-8 mRNAexpression as well as NF-κB/DNA binding activity in cystic fibrosis cellsstimulated with lipopolysaccharide (LPS) from Pseudomonas aeruginosa.

Methods dec-ODN LPP were prepared by a modified double emulsiontechnique and characterized in terms of size, morphology, tapped densityand dec-ODN loading. Human epithelial bronchial IB3-1 (CFTR-mutated) aswell as S9 (CFTR-corrected) were stimulated with LPS from P. aeruginosa for24 and 72 h in the absence or presence of naked dec-ODN or dec-ODN LPP.

Results Stimulation of cells with LPS from P. aeruginosa caused an increaseof IL-6 and IL-8 mRNA levels, which were significantly inhibited by dec-ODN LPP at 24 and 72 h, whereas naked dec-ODN inhibited those only at24 h. Similar effects were exhibited by dec-ODN LPP or naked dec-ODN onNF-κB/DNA binding activity.

Conclusions Our observations indicate that respirable biodegradabledec-ODN LPP may represent a promising strategy for inhibiting NF-κBtranscriptional activity and related gene expression and, thus, reduce lungchronic inflammation in CF patients.Copyright 2011 John Wiley & Sons, Ltd.

Keywords cystic fibrosis; decoy oligodeoxynucleotide; pulmonary delivery;dry powders; nuclear factor-κB; P. aeruginosa LPS

Introduction

Cystic fibrosis (CF) is the most common autosomal recessive and lethaldisease in Caucasian populations [1]. It is caused by mutations in the cystic

Copyright 2011 John Wiley & Sons, Ltd.

NF-κB decoy ODN-loaded microparticles in cystic fibrosis cells 201

fibrosis transmembrane conductance regulator (CFTR)gene encoding for a cAMP-regulated chloride channelglycoprotein, expressed in several epithelial cells [1].Defective CFTR reduces the channel function, leading toan altered fluid and electrolyte composition of airwaysecretions, that can predispose CF patients to abnormalmicrobial colonization and resultant inflammation [2].Neutrophil-dominated endobronchial inflammation andchronic bacterial infection are still considered to bethe primary cause of bronchioectasis, respiratory failureand consequent death in patients affected by CF [3].Pseudomonas aeruginosa, an opportunistic pathogen,chronically infects the lung and triggers via Toll-likereceptors a signaling pathway leading to hyper-activationof nuclear factor (NF)-κB and over-expression of a numberof pro-inflammatory cytokines, such as interleukin (IL)-6and chemoattractant IL-8 [4]. Bronchoalveolar fluid fromCF patients contains increased levels of pro-inflammatorycytokines and neutrophils in all the stages of thedisease [5–7]. A growing body of evidence indicatesthat lung inflammation is an event occurring prior toinfection in CF patients [8,9]. The over-expression of pro-inflammatory mediators in the airways of a 24-week-oldCF fetus has been reported [10]. Thus, mutant CFTR mayitself contribute to defective regulation of inflammatoryresponse in the lung, even in the absence of pathogens [9].The most common mutation (�F508) causes misfoldingof CFTR that, when retained in the endoplasmicreticulum, may activate the unfolded protein responseand contribute to an excessive production of NF-κB-dependent inflammatory mediators [11]. Furthermore,persistent inflammation and cytokine prolonged secretionhave been also attributed to the IκB/NF-κB deregulationpathway in CF [12]. Thus, NF-κB blockade may becrucial for limiting lung chronic inflammation in CF.NF-κB has been proposed as a candidate target forthe development of anti-inflammatory therapies [13].Several studies have reported new approaches aimingto block NF-κB transcriptional activity, among which thedecoy oligodeoxynucleotide (ODN) strategy is consideredto be very promising [14]. It has been demonstratedthat synthetic double-stranded ODN such as decoy cis-elements block NF-κB binding to the promoter regionsof its target genes, resulting in the inhibition of genetransactivation both in vitro [15] and in vivo [16], evenin humans [14].

ODN treatment through inhalation has been recentlyproposed as an alternative and more convenient ther-apeutic option for chronic lung diseases (e.g. chronicobstructive pulmonary disease, cystic fibrosis) comparedto systemic administration [17–19]. Nonetheless, toachieve the desired therapeutic effect specifically engi-neered inhalable systems are needed [17,20–21]. Afirst attempt to administer a decoy ODN to NF-κB vialipid-based carrier in the lung to target resident alve-olar macrophages has been reported [22]. Despite animproved pharmacological profile, colloidal dispersionsstill suffer from the short-term duration of the thera-peutic effect, the need to deliver the formulation via

time-consuming nebulizers, and a waste of drug in themouth/throat [23]. To allow a more efficient lung depo-sition as well as a prolonged duration of the therapeuticeffect, specifically designed dry powders could provemuch more attractive to develop as inhalable ODN thera-peutics [24,25]. Among the formulation options, particlesbased on poly(lactic-co-glycolic acid) (PLGA) could be apromising candidate as a result of their ability to entrapmacromolecules, protect them from in vivo degradationand sustain their release in the body [26]. Nevertheless,PLGA microspheres have to be especially engineered tobe large, as well as ‘light’ and porous, so as to fulfil therequirements of inhalation and to achieve aerodynamicproperties that are suitable for the widespread and homo-geneous lung deposition required in the treatment of localdiseases [27–29].

In the present study, we investigated the effects ofa decoy ODN to NF-κB (dec-ODN) delivered throughrespirable PLGA-based large porous particles (LPP) onNF-κB/DNA binding activity and IL-6 and IL-8 expressionin CF human epithelial bronchial cells stimulated withlipopolysaccharide (LPS) from P. aeruginosa for 24 and72 h.

Materials and methods

Chemicals

Phosphorothioate oligodeoxynucleotide synthesis wasperformed by Tib Molbiol (Roche Diagnostics, Italy)in accordance to our instructions. PLGA (50:50)(Resomer RG 504 H; Mw 41.9 kDa; inherent viscos-ity 0.5 dl/g) was purchased from Boehringer (Ingel-heim, Germany). Polysorbate 80, polyvinylalcohol (PVA;Mowiol 40–88), NaN3, 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC), ammonium bicarbonate and LPSfrom P. aeruginosa (serotype 10) were obtained fromSigma-Aldrich (Milan, Italy). Analytical grade sodiumchloride, potassium chloride, sodium phosphate dibasicanhydrous, sodium bicarbonate and methylene chloridewere supplied by Carlo Erba (Rodano, Italy). 32P-γ -ATP was from Perkin-Elmer (Milan, Italy). Anti-p50and anti-p65 were from Santa Cruz (Milan, Italy).D,L-dithiothreitol, pepstatin A, leupeptin, benzamidineand phenylmethylsulfonilfluoride were from Applichem(Darmstadt, Germany). Fetal bovine serum, glutamine,penicillin, streptomycin, Hepes, sodium pyruvate andphosphate-buffered saline (PBS) were from BioWhittaker(Caravaggio, BG, Italy). LHC-8 medium, TRIzol, dNTPs,oligo (dT)12–18 primer and MMLV-Reverse Transcrip-tase Taq Polymerase and Syber Safe were obtained fromInvitrogen (Milan, Italy).

Transcription factor decoy ODN

A plain double-stranded phosphorotioate ODN decoy toNF-κB (dec-ODN) was prepared by annealing of sense and

Copyright 2011 John Wiley & Sons, Ltd. J Gene Med 2011; 13: 200–208.DOI: 10.1002/jgm

202 D. De Stefano et al.

Figure 1. Properties of dec-ODN LPP. (A) Scanning electron microscopy images of dec-ODN LPP and (B) in vitro release profiles ofdec-ODN in phosphate buffer at pH 7.2 and 37 ◦C. Data in (B) are expressed as the mean ± SEM of two experiments performed intriplicate (n = 6)

Table 1. Overall properties of dec-ODN LPP (n = 6)

Characteristic Value

Volume mean diameter (µm ± SEM)a 31.5 ± 1.7Dec-ODN actual loading (nmol ± SEM)b 0.112 ± 0.005Dec-ODN encapsulation efficiency (% ± SEM) 80.3 ± 3.6Tapped density (g/ml ± SEM)c 0.035 ± 0.015MMADt (µm ± SEM)d 5.5 ± 0.9

aMean geometric diameter as determined by laser diffraction.bNanomoles of dec-ODN per mg of LPP.cTapped density was evaluated according to Eu. Ph.dMMADt was evaluated according to Ungaro et al. [29].

antisense oligodeoxynucleotides in vitro in filtered watersolution as described previously [30].

Preparation and characterizationof gas-foamed large porous particlescontaining dec-ODN

PLGA-based large porous particles containing dec-ODN(dec-ODN LPP) were prepared by a double emulsion-solvent evaporation technique assisted by gas-foamingas described previously [31]. Briefly, 0.25 ml of watercontaining (NH4)HCO3 (10% w/v) and dec-ODN werepoured into 2.5 ml of methylene chloride containingPLGA (15% w/v) in the presence of DPPC (0.1% w/v) ashelper lipid excipient. Emulsification was achieved usinga high-speed homogenizer (Ystral equipped with a tool6G; Heidolph, Kelheim, Germany) operating at 755 g for3 min. The primary emulsion was rapidly added to 25 mlof aqueous PVA (1% w/v) and homogenized at 676 g (tool10F) for 2 min to produce the multiple emulsion. Solventevaporation and subsequent particle hardening wasachieved under magnetic stirring at room temperature.Subsequently, particles were collected, washed threetimes with distilled water by centrifugation (HettichZentrifugen, Universal 16R, Tuttlingen, Germany) andfrozen in liquid nitrogen. Samples were then dried for

36 h by a Modulyo freeze-drier (Edwards, Crawley, UK)operating at 0.01 atmosphere and −60 ◦C. Each batch wasprepared in triplicate at dec-ODN theoretical loading of0.14 nmol/mg LPP. Blank LPP without dec-ODN (blankLPP) and LPP containing scramble dec-ODN (scrambledec-ODN LPP) were prepared as control for in vitro cellculture studies.

Particle shape and morphology were analyzed byscanning electron microscopy (Leica S440; Leica, Wetzlar,Germany). The samples were stuck on a metal stuband coated with gold under vacuum for 90–120 s.The mean geometric diameter and size distribution ofthe particles were determined by laser light scattering(Coulter LS 100Q, Beckman Coulter S.p.a., Milano, Italy)on a dispersion of freeze-dried microspheres in 0.2%w/v aqueous PVA. Particle size is expressed as volumemean diameter (± SEM) of values collected from threedifferent batches. Powder tapped density and flowability(Carr’s Index) were evaluated according to EuropeanPharmacopoeia 6 (Eu. Ph.), whereas theoretical massmean aerodynamic diameter (MMADt) was estimated aspreviously reported [31].

The amount of dec-ODN encapsulated within dec-ODNLPP was determined by a solvent extraction method.Briefly, 3 mg of dec-ODN LPP were dissolved into 1.2 mlof CH2Cl2 and dec-ODN was extracted into 1.2 ml ofwater. The suspension was centrifuged (2236 g roomtemperature, 15 min) and the supernatant analyzed fordec-ODN content by spectrophotometric analysis usinga Shimadzu 1204 apparatus (Shimadzu, Milan, Italy)562 nm. The linearity of the response was verified overthe concentration range 0.07–1.3 nmol/ml (r2 > 0.99).Results are expressed as encapsulation efficiency (ratio ofactual and theoretical dec-ODN loading × 100) ± SEM ofvalues collected from three different batches.

In vitro release of dec-ODN from LPP was monitoredby membrane dialysis in PBS containing 0.05% w/vsodium azide as preserving agent at pH 7.2 and 37 ◦C,as previously described [31]. Briefly, a known amountof dec-ODN LPP (20 mg) was suspended in 0.35 ml of



Figure 2. (A) Representative EMSA shows the effect of dec-ODN released from LPP or naked dec-ODN decoy on NF-κB/DNA bindingactivity in LPS-stimulated-IB3-1 cells at 24 and 72 h. Densitometric data are expressed as the mean ± SEM of three separateexperiments. < 0.01, < 0.001 versus unstimulated cells; ∗p < 0.05, ∗∗∗p < 0.001 versus LPS alone. (B) Characterization ofNF-κB/DNA complex was performed on nuclear extracts from LPS-stimulated IB3-1 cells at 24 h. Data are (A) and (B) are from asingle experiment and are representative of three separate experiments

PBS and placed in a dialysis membrane bag (MWCO:50000 Da, Spectra/Por, Spectrum Laboratories Inc.,Rancho Dominguez, CA, USA). The sample was droppedinto 2.5 ml of PBS (sink condition), and kept at 37 ◦C.At scheduled time intervals, 1.5 ml of external mediumwas withdrawn and replaced by the same amount offresh PBS. The withdrawn medium was analyzed for dec-ODN content by spectrophotometric analysis as describedabove. Experiments were run in triplicate for eachtime point of release kinetics. Results are reported aspercentage of dec-ODN released over time (± SEM) ofvalues collected from three different batches.

Cell culture

Human epithelial bronchial IB3-1 (with �F508 CFTRmutation) as well as S9 (CFTR-corrected) cells werepurchased by LGC Standards (Sesto San Giovanni,Milan, Italy). Cells were cultured at 37 ◦C in humidified5%CO2/95% air in LHC-8 medium with 5% fetal bovineserum. Petri dishes as well as multiwell were pre-coatedwith albumin 1 mg/ml, collagen 3 mg/ml and fibronectin1 mg/ml. The cells were plated in 48 culture wells ata density of 125 × 104 cells/ml per well or in culturedishes (diameter 10 cm) at a density of 5 × 106 cells/mlper dish and allowed to adhere for 2 h. Thereafter, themedium was replaced with fresh medium, and cells werestimulated with LPS (10 µg/ml) from P. aeruginosa for 24and 72 h in the absence or presence of naked dec-ODN(0.5 µM), scramble dec-ODN LPP (0.5 µM), dec-ODNLPP (0.5 µM) and blank LPP (0.5 µM). Cell viabilitywas determined using the 3-(4,5-dimethylthiazol-2yl)-2,5-diphenyl-2H-tetrazolium bromide conversion assay asdescribed previously [30].

Cytosolic and nuclear extracts

Cytosolic and nuclear extracts from cells were preparedas described previously with some modifications [30].Protein concentration was determined by a Bio-Rad(Milan, Italy) protein assay kit.

Electrophoretic mobility shift assay(EMSA)

Double-stranded oligonucleotides containing NF-κBrecognition sequences were end-labeled with 32P-γ -ATPand EMSA was performed as previously described [30].

Reverse transcription-polymerasechain reaction (RT-PCR)

Total RNA extraction from cultured cells was performedby using TRIzol (Invitrogen) as described previously[30]. The levels of IL-6 and IL-8 mRNA were evalu-ated by using PCR amplification of reverse-transcribedmRNA. The housekeeping gene GAPDH was used asan internal control. Five micrograms of total RNAwere reverse-transcribed into cDNA by using oligo(dT)12–18 primer (Invitrogen) and MMLV-Reverse Tran-scriptase (Invitrogen). One microlitre of cDNA wasamplified by PCR using Taq Polymerase (Invitrogen)in accordance with the manufacturer’s instructions. Theprimers were: IL-6 5′-TGACAAACAAATTCGGTACATCC-3′ (forward) and 5′-ATCTGAGGTGCCCATGCTAC-3′

(reverse); per IL-8 5′-TGCCAAGGAGTGCTAAAG-3′ (for-ward) and 5′-TCTCAGCCCTCTTCAAAA-3′ (reverse); perGAPDH 5′-CGCTGAGTACGTGGAG-3′ (forward) and 5′-GAGGAGTGGGTCTCGCTGTT-3′ (reverse). PCR reaction



Figure 3. (A) Representative EMSA shows the effect of dec-ODN released from LPP or naked dec-ODN decoy on NF-κB/DNA bindingactivity in LPS-stimulated S9 cells at 24 and 72 h. Data in (A) are from a single experiment and are representative of three separateexperiments. Densitometric data in are expressed as the mean ± SEM of three separate experiments. < 0.01, < 0.001versus unstimulated cells; ∗p < 0.05, ∗∗∗p < 0.001 versus LPS alone. (B) Characterization of NF-κB/DNA complex was performed onnuclear extracts from LPS-stimulated S9 cells at 24 h. Data in (A) and (B) are from a single experiment and are representative ofthree separate experiments

products were run on 1% agarose gel and visualized bySyber Safe staining.

Statistical analysis

Results are expressed as the mean ± SEM of nexperiments. Statistical significance was calculated byone-way analysis of variance and Bonferroni-correctedp-values for a multiple comparison test. p < 0.05 wasconsidered statistically significant.

Results

Preparation and characterizationof dec-ODN LPP

Biodegradable PLGA-based LPP containing dec-ODN (dec-ODN LPP) were prepared with good yields by the doubleemulsion technique employing ammonium bicarbonateand DPPC as aid excipients. A homogeneous population ofspherical particles with regular and uniformly distributedsurface pores was obtained (Figure 1A). The overall prop-erties of the developed dec-ODN LPP are reported inTable 1. Despite particle porosity and high solubility ofdec-ODN in aqueous media, the adopted formulation con-ditions allowed its very efficient entrapment in LPP. Thehigh porosity of dec-ODN LPP resulted in a very lowtapped density of approximately 0.04 g/ml and a Carr’sindex of approximately 4%, suggesting excellent powderflowability according to Eu. Ph. A preliminary estimationof MMADt gave a value lower than 6 µm, confirming goodpotential for widespread deposition of dec-ODN LPP inthe lung.

The results of the in vitro release studies are reportedin Figure 1B as a percentage of dec-ODN released fromLPP over time. The formulation displayed a typical releaseprofile, characterized by an early dec-ODN burst (8 ± 1%of the total content of dec-ODN into LPP released after6 h) followed by a sustained release of dec-ODN from LPP.

Effects of dec-ODN LPP on NF-κB/DNAbinding activity

The ability of dec-ODN LPP to inhibit NF-κB/DNA bindingactivity in nuclear protein extracts from IB3-1 and S9cells stimulated with LPS from P. aeruginosa for 24 and72 h was investigated. Dec-ODN LPP (0.5 µM), nakeddec-ODN (0.5 µM), scramble dec-ODN LPP (0.5 µM) orblank LPP were added to the cells 10 min before LPSchallenge, whether alone or in combination. The results,as reported in Figure 2A, demonstrate that a basallevel of NF-κB/DNA-binding activity was detected inunstimulated IB3-1 cells. Conversely, a retarded bandwas clearly observed following LPS challenge. Treatmentof IB3-1 cells with dec-ODN LPP significantly reduced NF-κB/DNA binding activity at 24 and 72 h. By contrast,naked dec-ODN exhibited these effects only at 24 h.Scramble dec-ODN LPP (0.5 µM) and blank LPP hadno effects. Comparable results were observed in S9 cells(Figure 3A). The composition of the NF-κB complex wasdetermined by competition and supershift experimentsin nuclear extracts from LPS-stimulated IB3-1 and S9cells. The specificity of the NF-κB/DNA binding complexwas demonstrated by the complete displacement of NF-κB/DNA binding in the presence of a 50-fold molarexcess of unlabeled NF-κB probe in the competitionreaction. By contrast, a 50-fold molar excess of unlabeled



Figure 4. Representative RT-PCR shows IL-6 (A) and IL-8 (B) mRNA levels induced by LPS in IB3-1 cells at 24 and 72 h. GAPDHmRNA levels are reported as a control. Data are from a single experiment and are representative of three separate experiments.Densitometric data are expressed as the mean ± SEM of three experiments. < 0.001 versus unstimulated cells; ∗∗∗p < 0.001versus LPS

mutated NF-κB probe or Sp-1 oligonucleotide had noeffect on DNA-binding activity. The composition of theNF-κB complex activated by LPS was determined by usingspecific antibodies against p50 and p65 subunits of NF-κB. The anti-p50 and anti-p65 antibodies clearly gaverise to a characteristic supershift of the retarded complex,suggesting that the NF-κB complex contained p50 andp65 heterodimers (Figure 2B). Comparable results wereobserved in S9 cells (Figure 3B).

Effects of dec-ODN LPP on IL-6 and IL-8mRNA expression

The effects of dec-ODN LPP on LPS-induced IL-6 and IL-8mRNA expression in IB3-1 and S9 cells were evaluated.The stimulation of IB3-1 with LPS (10 µg/ml) for 24and 72 h induced a significant increase either of IL-6and IL-8 mRNA levels compared to unstimulated cells.

Treatment of IB3-1 cells with dec-ODN LPP (0.5 µM)reduced, in a significant manner, either IL-6 and IL-8mRNA levels (by 74.12 ± 1.16% and 68.01 ± 0.67%, at24 h, respectively; by 83.74 ± 0.27% and 82.79 ± 0.94%,at 72 h, respectively; n = 3). By contrast, naked dec-ODN (0.5 µM) exhibited these effects only at 24 h (by50.13 ± 0.81% and 39.51 ± 0.55%, respectively; n = 3).Blank LPP and scramble dec-ODN LPP (0.5 µM) had noeffect (Figure 4). Comparable results were observed in S9cells (Figure 5).

Discussion

A current paradigm indicates that mutant CFTR, even inthe absence of pathogen infections in the airways of CFpatients, may itself contribute to persistent NF-κB acti-vation driving the deregulated inflammatory response



Figure 5. Representative RT-PCR shows IL-6 (A) and IL-8 (B) mRNA levels induced by LPS in S9 cells at 24 and 72 h. GAPDHmRNA levels are reported as a control. Data are from a single experiment and are representative of three separate experiments.Densitometric data are expressed as the mean ± SEM of three experiments. < 0.001 versus unstimulated cells; ∗∗∗p < 0.001versus LPS

[32–34]. Thus, NF-κB blockade by decoy ODN maylimit the progression of lung chronic inflammation inCF. The therapeutic potential of ODN decoy to NF-κB hasbeen studied in several chronic inflammatory associated-disorders [13]. An open-label phase I/II clinical trial hasshown that transfection of ODN decoy to NF-κB preventsrestenosis after coronary percutaneous intervention [14].To date, there have been no studies of controlling lungchronic inflammation by ODN decoy to NF-κB in in vivoCF models. The only evidence reports that transfection ofCF bronchial cells with a decoy ODN mimicking the NF-κB sequences and complexed with cationic lipids inhibitsP. aeruginosa-dependent expression of IL-8 [35]. Morerecently, a first attempt to administer a decoy ODN toNF-κB via lipid-based carrier in the lung to target residentalveolar macrophages has been reported [22]. However,translation of this approach in humans would run intothe need for frequent administrations by bulky, expensive

and time-consuming nebulizers, sometimes resulting inpoor lung deposition of the carrier [23]. In view of noveland more effective anti-inflammatory therapies for CF, weare currently developing inhalable powders for sustaineddelivery of a decoy ODN to NF-κB (dec-ODN) in the lung.

We have previously demonstrated that decoy ODNslowly released from PLGA microspheres in RAW 264.7macrophages stimulated with LPS for 24, 48 and 72 hreached higher cytoplasmic concentrations compared tonaked ODN [30]. In consideration of the application ofthis delivery system in lung chronic inflammation, PLGAmicrospheres have been specifically engineered by aid ofan effervescent salt and a lipid helper excipient [31]. Sometechnological crucial aspects were taken into account,such as bulk properties (e.g. morphology, size, flowability,aerodynamic diameter), as well as dec-ODN encapsulationefficiency and the in vitro release profile. Indeed, amass density lower than 0.4 g/cm3, the excellent flow



properties (flowability according to Eu. Ph.), a MMADt

ideal for widespread lung deposition and a geomet-ric diameter higher than 10 µm suitable for escapingmacrophage uptake, as well as the sustained releaseof dec-ODN, suggest that developed LPP have a greatpotential for inhalation therapy with ODN in CF patients.Subsequently, we investigated the effects of dec-ODNloaded LPP on NF-κB/DNA binding activity as well as IL-6and IL-8 mRNA levels in �F508 CFTR-mutated humanbronchial epithelial cells stimulated with LPS from P.aeruginosa for 24 and 72 h. Our findings show that LPSchallenge caused an increase of NF-κB/DNA binding activ-ity, which was significantly inhibited by dec-ODN loadedLPP at 24 and 72 h. This inhibitory effect on NF-κB/DNAbinding activity was correlated with decreased IL-6 andIL-8 mRNA levels. By contrast, naked dec-ODN exhibitedthese effects only at 24 h. The mechanisms by which LPPsustain the inhibitory effect of ODN on NF-κB/DNA bind-ing activity for a more prolonged time compared to nakeddec-ODN are still unclear. The slow release of dec-ODNfrom particles should occur extracellularly because LPPdeveloped here cannot be internalized in cells. There-fore, LPP may protect dec-ODN from nuclease attackin cultured cells and play a crucial role in determiningthe long-lasting inhibitory effect on NF-κB/DNA bindingactivity compared to naked dec-ODN. To our knowledge,this is the first time that highly porous PLGA particlesare shown to be as effective as their nonporous counter-part, allowing dec-ODN to exert a prolonged inhibition ofNF-κB/DNA binding activity in CF cells. Furthermore, itshould also be emphasized that the amount of naked dec-ODN administered to cells was equal to the total amountof dec-ODN loaded into LPP. However, only a limited

fraction of dec-ODN is expected to be released from dec-ODN LPP up to 72 h. It has been reported that dec-ODNenters into living cells by a endocytic process [36]. Thus,we hypothesize that dec-ODN, slowly released from LPP,does not saturate this process. By contrast, naked dec-ODN could result in the saturation of the endocytosisprocess exposing the excess of dec-ODN to esonucleases.These results are in agreement with previous findingsshowing the enhanced inhibition of NF-κB/DNA bindingactivity by decoy ODN slowly released from nonporousPLGA microspheres compared to naked ODN [30].

In conclusion, our observations, showing that dec-ODNLPP are able to inhibit NF-κB/DNA binding activity andIL-6 and IL-8 expression in CF human bronchial cellsat lower concentrations and for a more prolonged timecompared to naked dec-ODN, suggest that the developedLPP may increase dec-ODN biological stability and activ-ity. Although the molecular mechanisms underlying thepersistent inflammatory response in CF have not beenyet elucidated, respirable dec-ODN LPP may represent apromising strategy for inhibiting NF-κB-dependent geneexpression, and thus reduce chronic lung inflammation inCF patients.

Acknowledgements

The financial support of ‘Fondazione per la Ricerca sulla FibrosiCistica-Onlus/Delegazione di Torino’ (FFC#5/2007) is gratefullyacknowledged. The authors declare that there are no conflicts ofinterest. This is an original research article for which the ethicalbackground and any institutional or national ethical committeehave been approved.

References

1. Rowe SM, Miller S, Sorscher EJ. Cys-tic fibrosis. N Engl J Med 2005; 352:1992–2001.

2. Boncoeur E, Roque T, Bonvin E, et al.Cystic fibrosis transmembrane conduc-tance regulator controls lung protea-somal degradation and nuclear factor-kappaB activity in conditions of oxida-tive stress. Am J Pathol 2008; 172:1184–1194.

3. Downey DG, Bell SC, Elborn JS. Neu-trophils in cystic fibrosis. Thorax 2009;64: 81–88.

4. Greene CM, Carroll TP, Smith SG, et al.TLR-induced inflammation in cysticfibrosis and non-cystic fibrosis airwayepithelial cells. J Immunol 2005; 174:1638–1646.

5. Bonfield TL, Panuska JR, Konstan MW,et al. Inflammatory cytokines in cysticfibrosis lungs. Am J Respir Crit Care Med1995; 152: 2111–2118.

6. Khan TZ, Wagener JS, Bost T, Mar-tinez J, Accurso FJ, Riches DW. Earlypulmonary inflammation in infants with

cystic fibrosis. Am J Respir Crit Care Med1995; 151: 1075–1082.

7. Noah TL, Wortman IA, Hu PC,Leigh MW, Boucher RC. Cytokine pro-duction by cultured human bronchialepithelial cells infected with areplication-deficient adenoviral genetransfer vector or wild-type adenovirustype 5. Am J Respir Cell Mol Biol 1996;14: 417–424.

8. Weber AJ, Soong G, Bryan R, Saba S,Prince A. Activation of NF-kappaB inairway epithelial cells is dependent onCFTR trafficking and Cl–channel func-tion. Am J Physiol Lung Cell Mol Physiol2001; 281: L71–L78.

9. Verhaeghe C, Remouchamps C, Hen-nuy B, et al. Role of IKK and ERK path-ways in intrinsic inflammation of cys-tic fibrosis airways. Biochem Pharmacol2007; 73: 1982–1994.

10. Verhaeghe C, Delbecque K, de Leval L,Oury C, Bours V. Early inflammation inthe airways of a cystic fibrosis foetus.J Cyst Fibros 2007; 6: 304–308.

11. Rottner M, Freyssinet JM, Mar-tinez MC. Mechanisms of the noxiousinflammatory cycle in cystic fibrosis.Respir Res 2009; 10: 23–34.

12. Venkatakrishnan A, Stecenko AA, KingG, et al. Exaggerated activation ofnuclear factor-kappaB and alteredIkappaB-beta processing in cystic fibro-sis bronchial epithelial cells. Am J RespirCell Mol Biol 2000; 23: 396–403.

13. Courtois G, Gilmore TD. Mutations inthe NF-kappaB signaling pathway:implications for human disease. Onco-gene 2006; 25: 6831–6843.

14. Egashira K, Suzuki J, Ito H, Aoki M,Isobe M, Morishita R. Long-term fol-low up of initial clinical cases withNF-kappaB decoy oligodeoxynucleotidetransfection at the site of coronary stent-ing. J Gene Med 2008; 10: 805–809.

15. Viedt C, Dechend R, Fei J, Hansch GM,Kreuzer J, Orth SR. MCP-1 inducesinflammatory activation of human tubu-lar epithelial cells: involvement of the



transcription factors, nuclear factor-kappaB and activating protein-1. J AmSoc Nephrol 2002; 13: 1534–1547.

16. De Stefano D, De Rosa G, Maiuri MC,et al. Oligonucleotide decoy to NF-kappaB slowly released from PLGAmicrospheres reduces chronic inflam-mation in rat. Pharmacol Res 2009; 60:33–40.

17. Densmore CL. Advances in noninvasivepulmonary gene therapy. Curr DrugDeliv 2006; 3: 55–63.

18. Ziady AG, Davis PB. Current prospectsfor gene therapy of cystic fibrosis. CurrOpin Pharmacol 2006; 6: 515–521.

19. Seguin RM, Ferrari N. Emergingoligonucleotide therapies for asthmaand chronic obstructive pulmonary dis-ease. Expert Opin Investig Drugs 2009;18: 1505–1517.

20. Griesenbach U, Cassady RL, Cain RJ, duBois RM, Geddes DM, Alton EW. Cyto-plasmic deposition of NFkappaB decoyoligonucleotides is insufficient to inhibitbleomycin-induced pulmonary inflam-mation. Gene Ther 2002; 9: 1109–1115.

21. Griesenbach U, Scheid P, Hillery E,de Martin R, Huang L, Geddes DM,et al. Anti-inflammatory gene therapydirected at the airway epithelium. GeneTher 2000; 7: 306–313.

22. Wijagkanalan W, Kawakami S, HiguchiY, Yamashita F, Hashida M. Intratra-cheally instilled mannosylated cationicliposome/NFkappaB decoy complexesfor effective prevention of LPS-inducedlung inflammation. J Control Release2011; 149: 42–50.

23. Rogueda PG, Traini D. The nanoscale inpulmonary delivery. Part 2: formulationplatforms. Expert Opin Drug Deliv 2007;4: 607–620.

24. Cryan SA. Carrier-based strategies fortargeting protein and peptide drugs tothe lungs. AAPS J 2005; 7: E20–E41.

25. Salama R, Traini D, Chan HK,Young PM. Recent advances in con-trolled release pulmonary therapy. CurrDrug Deliv 2009; 6: 404–414.

26. Mundargi RC, Babu VR, RangaswamyV, et al. Nano/micro technologies fordelivering macromolecular therapeuticsusing poly (D,L-lactide-co-glycolide)and its derivatives. J Control Release2008; 125: 193–209.

27. Edwards DA, Ben-Jebria A, Langer R.Recent advances in pulmonary drugdelivery using large, porous inhaledparticles. J Appl Physiol 1998; 85:379–385.

28. Ungaro F, d’Emmanuele di Villabi-anca R, Giovino C, et al. Insulin-loadedPLGA/cyclodextrin large porous parti-cles with improved aerosolization prop-erties: in vivo deposition and hypogly-caemic activity after delivery to ratlungs. J Control Release 2009; 135:25–34.

29. Ungaro F, De Rosa G, Miro A, QuagliaF, La Rotonda MI. Cyclodextrins in theproduction of large porous particles:development of dry powders for the sus-tained release of insulin to the lungs. EurJ Pharm Sci 2006; 28: 423–432.

30. De Rosa G, Maiuri MC, Ungaro F, et al.Enhanced intracellular uptake and inhi-bition of NF-kappaB activation by decoyoligonucleotide released from PLGAmicrospheres. J Gene Med 2005; 7:771–781.

31. Ungaro F, Giovino C, Coletta C, Sor-rentino R, Miro A, Quaglia F. Engineer-ing gas-foamed large porous particlesfor efficient local delivery of macro-molecules to the lung. Eur J Pharm Sci2010; 41: 60–70.

32. DiMango E, Ratner AJ, Bryan R,Tabibi S, Prince A. Activation of NF-kappaB by adherent Pseudomonasaeruginosa in normal and cystic fibrosisrespiratory epithelial cells. J Clin Invest1998; 101: 2598–2605.

33. Joseph T, Look D, Ferkol T. NF-kappaBactivation and sustained IL-8 geneexpression in primary cultures of cys-tic fibrosis airway epithelial cells stim-ulated with Pseudomonas aeruginosa.Am J Physiol Lung Cell Mol Physiol 2005;288: L471–L479.

34. Machen TE. Innate immune responsein CF airway epithelia: hyperinflamma-tory? Am J Physiol Cell Physiol 2006;291: C218–C230.

35. Bezzerri V, Borgatti M, Nicolis E, et al.Transcription factor oligodeoxynu-cleotides to NF-kappaB inhibit transcrip-tion of IL-8 in bronchial cells. Am JRespir Cell Mol Biol 2008; 39: 86–96.

36. Loke SL, Stein CA, Zhang XH, et al.Characterization of oligonucleotidetransport into living cells. Proc Natl AcadSci USA 1989; 86: 3474–3478.


sustained inhibition of il-6 and il-8 expression by decoy odn to nf-κb delivered through respirable...

Documents