Synthesis and Toxicological Evaluation of a Chitosan- l -Leucine Conjugate for Pulmonary Drug Delivery Applications

Synthesis and Toxicological Evaluation of a Chitosan‑L‑LeucineConjugate for Pulmonary Drug Delivery ApplicationsMohammad D. A. Muhsin,†,‡ Graeme George,† Kenneth Beagley,† Vito Ferro,§ Charles Armitage,†

and Nazrul Islam*,†,‡

†Institute of Health and Biomedical Innovation, Queensland University of Technology, 60 Musk Avenue, Kelvin Grove, Brisbane,Queensland 4059, Australia‡Pharmacy Discipline, Faculty of Health, Queensland University of Technology, Brisbane, Queensland 4000, Australia§School of Chemistry and Molecular Biosciences, The University of Queensland, Brisbane, Queensland 4072, Australia

*S Supporting Information

ABSTRACT: Herein are reported the synthesis of a conjugate of chitosanwith L-leucine, the preparation of nanoparticles from both chitosan and theconjugate for use in pulmonary drug delivery, and the in vitro evaluation oftoxicity and inflammatory effects of both the polymers and theirnanoparticles on the bronchial epithelial cell line, BEAS-2B. Thenanoparticles, successfully prepared both from chitosan and the conjugate,had a diameter in the range of 10−30 nm. The polymers and theirnanoparticles were tested for their effects on cell viability by MTT assay, ontrans-epithelial permeability by using sodium fluorescein as a fluid phasemarker, and on IL-8 secretion by ELISA. The conjugate nanoparticles had alow overall toxicity (IC50 = 2 mg/mL following 48 h exposure; no inductionof IL-8 release at 0.5 mg/mL concentration), suggesting that they may besafe for pulmonary drug delivery applications.

■ INTRODUCTION

Chitosan is a linear copolymer of β(1→4)-linked D-glucos-amine with occasional N-acetyl-D-glucosamine residues. Due toits natural abundance,1,2 biocompatibility,3 and biodegrad-ability,4 chitosan has drawn wide interest in diverse fieldsincluding medicine, pharmacy, and biotechnology.5,6 Onedeterrent to the exploitation of the polymer to its full potentialis its poor solubility and processability, and thus there has beenmuch interest in the improvement of its functional applicabilityby chemical modification.7,8

For many years, there has been a keen interest in theconjugation of amino acids to chitosan for potential utilizationin various fields including adsorption of heavy metals,9 removalof low-density lipoproteins,10 and immobilization of lipases.11

Of late, enhanced dispersibility of particles from a chitosan-based dry powder inhaler (DPI) upon addition of the freeamino acid, L-leucine to the formulation has been reported.12,13

This effect was attributed to the migration of L-leucine to thesurface of chitosan particles during the drying phase owing toits surfactant-like properties giving the particles a pittedappearance and in turn reducing interparticle cohesion.14 Ithas also been shown for L-leucine and a leucine tripeptide that,at the particle-air interface, the hydrophobic side-chains of thesespecies are projected outward.15 In the current study, thechemical conjugation of chitosan with L-leucine was inves-tigated to determine whether this effect on the dispersibility ofchitosan from a DPI could be maintained. Considering the

structural features of L-leucine and the previously reportedorientation of the hydrophobic part of L-leucine (added to aparticle formulation) at the particle−air interface, it washypothesized that, upon conjugation, the L-leucine substituentwill be oriented around the chitosan backbone with itshydrophobic part projected outward, resulting in a reducedinterparticle interaction. It was also hypothesized that theconjugation of L-leucine to chitosan would enhance thedispersibility of the particles from a DPI. Although the physicaladdition of L-leucine to chitosan has been shown to improve itsdispersibility, the poor solubility of chitosan still remains amajor obstacle to its utilization. Thus, a second objectivebehind chemical conjugation of L-leucine to chitosan was toimprove its solubility by creating an amphiphilic environmentaround the chitosan backbone.The respiratory system is an established drug delivery route

for lung diseases such as asthma and chronic obstructivepulmonary disease16,17 and is of considerable interest for drugdelivery to the systemic circulation.18 However, it is importantto ensure the safety of any formulation intended for inhalation.Previous studies on chitosan and its micro- or nanoparticles onpulmonary epithelial cell lines such as Calu-3 and A-549indicate their low toxicity.19,20 Trimethylchitosan chloride

Received: June 11, 2014Revised: September 4, 2014Published: September 5, 2014

Article

pubs.acs.org/Biomac

© 2014 American Chemical Society 3596 dx.doi.org/10.1021/bm5008635 | Biomacromolecules 2014, 15, 3596−3607

pubs.acs.org/Biomac

(TMC)a water-soluble derivative of chitosanhas, however,been reported to be more toxic than the precursor chitosan.21,22

The safety of respiratory formulations can best be evaluatedin vitro by a combination of toxicological tests on respiratoryepithelial cell lines. Cell viability assays, e.g., the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT)assay, are widely used to evaluate the safety of inhaledmaterials.23,24 Another indicator of adverse effects of inhaledagents on epithelial cell layers is the change in the permeabilityof the epithelial barrier. This can be assessed by measuring thepermeability of a fluid phase marker such as sodium fluorescein(Na Flu) across a confluent cell monolayer.20 Anotherimportant concern is to rule out any inflammatory responseof the respiratory epithelium to the inhaled product. This canbe accomplished by monitoring the secretion of chemokinessuch as interleukin 8 (IL-8) by enzyme-linked immunosorbentassay (ELISA).25 IL-8 is a chemokine whose secretion is oftenassociated with initiation of inflammatory processes in lungtissue.26

The respiratory epithelial cell models, used in previoustoxicological investigations of chitosan and TMC (e.g., Calu-3and A-549), are derived from human lung tumor. BEAS-2B27 isa nonmalignant cell line derived from normal human bronchialepithelial cells immortalized by a Simian Virus (SV) 40/adenovirus-12 hybrid virus.28 It has been used widely as an invitro bronchial epithelial cell model.29 However, no studieshave so far been reported to investigate the toxicity of chitosanor any of its derivatives using this cell line.This study describes the synthesis of an L-leucine conjugate

of chitosan and its subsequent use for fabrication ofnanoparticles suitable for inhalation as a DPI formulation.The L-leucine conjugate was characterized by Fourier transforminfrared spectroscopy (FT-IR), 1H, 13C and 2D 1H−13Cgradient-enhanced heteronuclear single quantum correlation(ge-HSQC) nuclear magnetic resonance (NMR) spectroscopy,elemental analysis, and X-ray photoelectron spectroscopy(XPS). The safety of chitosan, its L-leucine conjugate, andtheir nanoparticles for pulmonary delivery was evaluated interms of three well-recognized toxicity indicators (cell viability,trans-epithelial permeability and IL-8 secretion) using theBEAS-2B cell line as an in vitro model.

■ MATERIALS AND METHODSMaterials. Low molecular weight chitosan (degree of deacetylation

(DDA), 92%; molecular weight (MW), 50−190 kDa), aceticanhydride, trityl chloride, aqueous hydrazine hydrate (50−60%),Boc-L-leucine succinimide (Boc-Leu-OSu), 4 M HCl in 1, 4-dioxane,N,N-dimethylformamide (DMF), pyridine, chloroform-d (CDCl3, 99.8atom % D), pyridine-d5 (99.5 atom % D, 0.03% v/v tetramethylsilane,TMS), deuterium oxide (D2O, 99.99 atom % D), MTT, dimethylsulfoxide (DMSO, spectrophotometric grade) and Na Flu wereobtained from Sigma-Aldrich Pty Ltd. (Australia). Diethyl ether, glacialacetic acid, hexane and petroleum ether (40−60) were purchased fromMerck Pty. Ltd. (Australia). RPMI 1640 (Gibco), fetal bovine serum(Lonza), L-glutamine (Gibco), penicillin-streptomycin (Gibco), and0.5% trypsin-ethylenediaminetetracetic acid (EDTA) (Gibco) weresupplied by Life Technologies Australia Pty Ltd. Phthalic anhydridewas obtained from Merck Schuchardt (Germany), methanol andparaffin oil heavy 68 (viscosity: 66.0−70.0 cST @ 40 °C) were fromChem-Supply (Australia), ethanol was from Recochem (Australia),span 80 was from PCCA (Australia), phosphate-buffered saline (PBS)tablets (pH 7.3 ± 0.2 when dissolved in prescribed volume of H2O)were from Oxoid Ltd. (England), and the human IL-8 ELISA MAXDeluxe kit was from Biolegend (USA). The bronchial epithelial cellline, BEAS-2B was a gift from Prof. Philip Hansbro (School of

Biomedical Sciences and Pharmacy, The University of Newcastle,Australia). Pyridine was stored over 4 Å molecular sieves to preventmoisture absorption. Glacial acetic acid was diluted to a 2% aqueoussolution for dissolution of chitosan. Trypsin-EDTA was diluted to0.01% in PBS prior to use for splitting cells. All other reagents wereused as received without any further purification or dilution.

Synthesis. N-Phthaloylchitosan (2). N-Phthaloylchitosan (2) wassynthesized according to the method of Holappa et al.30 with somemodifications. Phthalic anhydride (2.43 g, 16.4 mM) was added to adispersion of chitosan (1) obtained by overnight stirring of 1 (1 g, 5.48mM free −NH2 group) in DMF (20 mL) containing 5% (v/v) water.The mixture was heated with stirring at 130 °C for 8 h, cooled to roomtemperature, and poured into ice water (500 mL). The precipitate wascollected by filtration, washed with copious amounts of methanol, anddried to give 1.56 g (95%) of a pale tan powder. Degree of substitution(DS): 0.91. IR (ATR): ν 3700−3100 (O−H stretch overlapping N−Hstretch), 2980−2830 (C−H stretch, pyranose), 1774, 1703 (COstretch, imide), 1612 (amide I, N-acetyl), 1547 (N−H bendoverlapping amide II), 1468 (asymmetric C−H bend in CH2), 1385(CC, phth), 1200−800 (C−O stretch, pyranose), 718 cm−1 (arom,phth). Calcd: C, 53.73; H, 5.00; N, 4.63. Found: C, 51.25; H, 4.87; N,4.44.

N-Phthaloyl-3,6-di-O-acetylchitosan (2a). N-Phthaloyl-3,6-di-O-acetylchitosan (2a) was synthesized following the method describedby Nishimura et al.31 with some modifications. Acetic anhydride (10mL, 105.8 mmol) was added to a suspension of 2 (100 mg, 0.33mmol) in pyridine (20 mL) and heated with stirring at 130 °Covernight. The resulting homogeneous mixture was cooled to roomtemperature and precipitated in ice−water (60 mL). The precipitatewas washed successively with ethanol and ether and dried to give 124mg of 2a (97%). DS: 1.97. IR (ATR): ν 3700−3100 (O−H stretchoverlapping N−H stretch), 2980−2830 (C−H stretch, pyranose),1777 (CO stretch, imide), 1743 (CO stretch, O-acetyl ester),1711 (CO stretch, imide), 1613 (amide I, N-acetyl), 1543 (N−Hbend overlapping amide II), 1469 (asymmetric C−H bend in CH2),1430 (asymmetric C−H bend in CH3), 1385 (CC, phth), 1371(symmetric C−H bend in CH3), 1219 (C−O stretch, O-acetyl),1200−800 (C−O stretch, pyranose), 722 cm−1 (arom, phth). 1HNMR (CDCl3): δ 1.63−2.30 (O- and N-acetyl), 2.80−4.90(pyranose), 5.18 (bs), 5.51 (bd, J = 8.4 Hz, H1), 7.60−7.80 (Ar).13C NMR (CDCl3): δ 20.4−20.8 (O- and N-acetyl), 55.3 (C-2), 62.1(C-6), 70.2 (C-3), 72.4 (C-5), 75.3 (C-4), 97.2 (C-1), 124.0, 131.3,134.6 (Ar), 167.7−170.2 ppm (CO). Calcd: C, 54.47; H, 4.96; N,3.63. Found: C, 54.48; H, 4.88; N, 3.66.

N-Phthaloyl-6-O-tritylchitosan (3). N-Phthaloyl-6-O-tritylchitosan(3) was synthesized from N-phthaloylchitosan (2) according to themethod of Zhang et al.32 with some modifications. A suspension of 2(1 g, 3.32 mmol free −OH) in pyridine (47 mL) was treated with tritylchloride (9.23 g, 33.1 mmol) and stirred with heating at reflux (∼115°C) for 24 h under argon. After cooling to room temperature, thereaction mixture was poured into EtOH. The precipitate was collectedby filtration and washed successively with EtOH and Et2O. The yieldof the product was 1.45 g (80%). DS: 1.06. IR (ATR): ν 3700−3100(O−H stretch overlapping N−H stretch), 3100−3000 (C−H, trityl),2980−2830 (C−H, pyranose and CH3), 1776, 1712 (CO, imide),1611 (amide I, N-acetyl), 1591 (N−H bend overlapping amide II),1490 (CC, trityl), 1468 (asymmetric C−H bend in CH2), 1448(CC, trityl), 1384 (CC, phth), 1200−800 (C−O, pyranose), 764,746 (arom, trityl), 719 (arom, phth), 699 cm−1 (arom, trityl). Calcd:C, 71.72; H, 5.38; N, 2.57. Found: C, 71.91; H, 5.27; N, 2.50.

N-Phthaloyl-3-O-acetyl-6-O-tritylchitosan (3a). N-Phthaloyl-3-O-acetyl-6-O-tritylchitosan (3a) was synthesized following the sameprocedure as described above for 2a. Acetic anhydride (10 mL, 105.8mM) was added to a suspension of 3 (100 mg, 0.18 mM) in pyridine(20 mL) and heated overnight at reflux (130 °C). The resultingsolution was cooled to room temperature and precipitated by pouringinto ice−water (60 mL). The precipitate was collected by filtration,washed successively with ethanol and ether, and dried to give a yield of90 mg. This was 83% of the theoretical yield assuming that 100%substitution has taken place. The microanalysis results were not

Biomacromolecules Article

dx.doi.org/10.1021/bm5008635 | Biomacromolecules 2014, 15, 3596−36073597

satisfactory and so could not be used for calculating the actual DS. IR(ATR): ν 3700−3100 (O−H stretch overlapping N−H stretch),3100−3000 (C−H, trityl), 2980−2830 (C−H stretch, pyranose andCH3), 1777 (CO stretch, imide), 1745 (CO stretch, O-acetylester), 1714 (CO stretch, imide), 1612 (amide I, N-acetyl), 1595(N−H bend), 1547 (amide II, N-acetyl), 1529, 1491 (CC, trityl),1468 (asymmetric C−H bend in CH2), 1449 (CC, trityl), 1385(CC, phth), 1220 (C−O stretch, O-acetyl), 1200−800 (C−Ostretch, pyranose), 765, 747 (arom, trityl), 721 (arom, phth), 704 cm−1

(arom, trityl). 1H NMR (CDCl3): δ 1.63−2.30 (O- and N-acetyl),3.00−5.00 (pyranose), 5.19 (bs), 5.50 (bs), 6.89−7.80 ((Ar). 13CNMR (CDCl3): δ 20.4−20.8 (O- and N-acetyl), 55.3 (C-2), 62.1 (C-6), 70.2 (C-3), 72.4 (C-5), 75.3 (C-4), 97.2 (C-1), 123.9−147.0 (Ar),167.8−170.3 (CO,). Calcd: C, 70.68; H, 5.33; N, 2.39. Found: C,61.65; H, 4.89; N, 3.05.6-O-Tritylchitosan (4). Deprotection of the phthaloyl group from 3

to produce 6-O-tritylchitosan (4) was performed according to theliterature31,32 with some modifications. A suspension of N-phthaloyl-6-O-trityl-chitosan 3 (1 g, 1.84 mmol −NH2 group equivalent) inhydrazine hydrate (50−60%, 50 mL) was stirred with heating at reflux(∼110 °C) for 18 h under an argon atmosphere. The reaction mixturewas cooled to room temperature and poured into water (500 mL).The precipitate was filtered off and washed with water (2 × 500 mL)and finally with ethanol and ether. The yield of the product was 0.75 g(96%). Microanalytical data was unsatisfactory to calculate the actualdegree of dephthaloylation. However, FT-IR and 1H NMR spectraindicated complete removal of the phthaloyl group. IR (ATR): ν3700−3100 (O−H stretch overlapping N−H stretch), 3100−3000(C−H, trityl), 2980−2830 (C−H, pyranose and CH3), 1661 (amide I,N-acetyl), 1596 (amide II overlapping N−H bend), 1490, 1448 (CC, trityl), 1374 (symmetric C−H bend in CH3), 1318 (amide III),1200−800 (C−O, pyranose), 763, 746, 697 cm−1 (arom, trityl). 1HNMR (pyridine-d5): δ 1.67−2.13 (CH3, N-acetyl), 3.00−5.79(pyranose), 7.28−7.80 (Ar). Calcd: C, 71.13; H, 6.44; N, 3.30.Found: C, 71.14; H, 6.13; N, 3.54.N-(Boc-L-leucine)-6-O-tritylchitosan- (5). A solution of Boc-Leu-

OSu (1.16 g, 3.53 mmol) in pyridine (15 mL) was added dropwise to4 (500 mg, 1.18 mmol −NH2 group equivalent) at 0 °C over a periodof 1 h under an argon atmosphere, and the reaction mixture was thenstirred for an additional 3 h at 0 °C. An additional 773 mg (2.35mmol) of Boc-Leu-OSu in pyridine (10 mL) was then added dropwiseto the reaction mixture and the reaction was continued at roomtemperature under argon for a further 21 h until a clear solution wasobtained. Finally, the product was precipitated by pouring the reactionmixture into diethyl ether (250 mL). The precipitate was washed withdiethyl ether (× 3) and dried to give 485 mg (70%) of (5) as a solid.DS: 0.74. IR (ATR): ν 3700−3100 (O−H stretch overlapping N−Hstretch), 3100−3000 (C−H, trityl), 2980−2830 (C−H, pyranose andBoc-Leucine), 1683 (amide I, N-acetyl, N-L-leucine, and N-carbamate),1597 (amide II overlapping N−H bend), 1491, 1448 (CC, trityl),1390 and 1367 (Boc t-butyl and L-leucine isopropyl), 1319 (amideIII), 1200−800 (C−O, pyranose), 764, 747, 702 cm−1 (arom, trityl).1H NMR: δ 0.90 (H-δ1, leu), (H-γ, leu overlapped with t-butyl, Boc),1.44 (t-butyl, Boc overlapped with CH3, N-acetyl at 1.66 ppm), 1.92(H-β, leu overlapped with H-δ1), 2.15 (CH3, N-acetyl), 2.65 (H-δ2,leu), 3.00−5.5 (pyranose protons overlapped with H-α, leu at 4.51),7.10−8.00 (Ar). 13C NMR (CDCl3): δ 20.7−23.8 (CH3, N-acetyl),25.3, 26.4 (C-δ1&2, leu), 28.9 (t-butyl, Boc), 30.3 (C- γ, leu), 43.0 (C-β,leu), 54.5 (C-α, leu), 56.3 (C-2), 58.4 (C-6), 72.6 (C-3), 75.8 (C-5),77.4 (C-4), 79.2 (C(CH3)3), 87.1 (C-Ph3), 97.4, 101.0, 102.0, 104.5,104.8 (C-1), 127.8, 128.7, 129.9, 145.0 (Ar), 156.9 (CO,carbamate), 173.6 (CO, N-leu), 176.3 ppm (CO, N-acetyl).Calcd: C, 68.23; H, 7.25; N, 4.33. Found: C, 67.98; H, 6.78; N, 4.14.Chitosan-N-L-leucine·HCl (6). 4 M HCl in 1,4-dioxane (4 mL) was

added dropwise to 5 (200 mg, 343 μmol) under an argon atmosphereand stirred for 15 min at this temperature at 0 °C. Stirring wascontinued at room temperature under argon for 24 h. The reactionmixture was poured into diethyl ether (40 mL), the precipitate filteredoff, washed with diethyl ether (×3) and dried to give the product,chitosan-N-L-Leucine·HCl 6 (80 mg, 66%) as a solid. The micro-

anlaytical data indicated that 91% of the protecting Boc and triytlgroups had been removed. IR (ATR): ν 3700−3100 (O−H stretchoverlapping N−H stretch), 2980−2830 (C−H, pyranose & L-leucine),1672 and 1628 (amide I, N-acetyl overlapping N-L-leucine), 1558 and1512 (amide II overlapping symmetric N−H bend in NH3

+), 1390 and1372 (isopropyl, leu), 1322 (amide III), 1200−800 cm−1 (C−O,pyranose). 1H NMR: δ 0.98 (H-δ1&2, leu), 1.78 (H-δ2 and H-β of leuwith some overlapping with H-δ1 and H-γ), 2.07 (N-acetyl and H-δ1,leu), 2.80 (H-γ, leu), 3.19 (H-2, pyranose), 3.30−4.34 (H-2 to H-6,pyranose overlapped with H-α and H-β of leu), 4.50−5.54 ppm (H-1,pyranose), 13C NMR (CDCl3): δ 20.7 (CH3, N-acetyl), 21.7 and 23.5(C-δ1&2, leu), 25.0 (C-γ, leu), 40.0 (C-β, leu), 52.0 (C-α, leu), 55.6 (C-2), 59.8 (C-6), 69.8−76.1 (C-3, C-5, C-4), 97.3−101.1 (C-1), 170.8(CO, N-leu), 174.4 ppm (CO, N-acetyl). Calcd: C, 43.81; H,7.62; N, 8.40. Found: C, 36.84; H, 7.14; N, 6.93. XPS Analysis (Atom%): Chitosan: O, 28.59; C, 63.02; N, 7.34, Cl 1.05. L-leucine: O, 19.40;C, 69.44; N, 11.55. Chitosan-N-L-leucine·HCl: O, 25.68; C, 60.99; N,9.31; F, 0.64; Cl, 3.38.

Preparation of Chitosan and Chitosan-L-leucine ConjugateNanoparticles. Chitosan and conjugate nanoparticles were preparedby a water-in-oil (W/O) emulsion-solvent evaporation techniquemodified from that reported previously for fabrication of chitosanmicroparticles.33,34 For fabricating chitosan nanoparticles, 2.5 mL of a2% solution of chitosan in 2% acetic acid was added dropwise at 60 °Cto 100 g heavy paraffin oil containing 1 mL span 80 and homogenizedinto a W/O emulsion at 11,000 rpm. The emulsion was stirredovernight at 5000 rpm and 60 °C using an IKA Eurostar Digitaloverhead stirrer (IKA Works (Asia), Inc.) to remove the internal waterphase and harden the droplets into nanoparticles. The nanoparticleswere then separated from the oil phase by centrifugation overnight at5000g using an Allegra X15R benchtop centrifuge (Beckman-Coulter,Inc. USA), washed three times with hexane to remove excess oil,filtered and dried under vacuum at 60 °C. For making chitosan-L-leucine conjugate nanoparticles, the same protocol was followedexcept that no acetic acid was added for preparing the solution of theconjugate in water.

The yields of nanoparticles were quantified as the percentage ofanticipated yields. The nanoparticles obtained were weighed and thepercentage yield was calculated using the following formula:production yield (%) = (weight of the dried particle/sum of the dryweight of starting material) × 100.35 Measurement was done intriplicate (n = 3).

Characterization. FT-IR spectra of the chitosan, conjugates, andintermediates were recorded on a Nicolet 5700 FT-IR attenuated totalreflectance (ATR) spectrophotometer (Thermo Electron Corporation,Madison, Wisconsin, USA) equipped with a Smart Endurancediamond internal reflection element under dry air at ambienttemperature (25 °C). The spectra were collected by using standardspectral collection techniques and the software Omnic 32 over thewavenumber range 4000−650 cm−1 using 64 scans at a resolution of 4cm−1.

1H, 13C and 2D 1H−13C ge-HSQC NMR spectra were recorded ona Bruker AVANCE 400 high-resolution spectrometer (Bruker BioSpinGmbH, Germany) using a quad nucleus probe (QNP; tunable for 1H,19F, 31P, 13C) under a static magnetic field of 400 and 100 MHz,respectively, at 298 K. Spectra of 16384 data points were recordedover a spectral width of 12.0144 ppm using the accumulation of 32transients and an exponential line broadening (LB) of 0.30 Hz wasapplied prior to Fourier transformation for 1H NMR. 13C NMRspectra were recorded over a spectral window of 219.8552 ppm with16384 data points and 6144 transients using proton decoupling. Priorto Fourier transformation, 2.00 Hz LB was applied. The 1H−13C ge-HSQC experiments were performed in phase-sensitive mode usingEcho/Antiecho-time proportional phase incrementation (TPPI)gradient selection. The data matrix was 256 × 2048 with spectralwidths of 4807.69 Hz for proton and 18 115.94 Hz for carbon. Theevolution time was set to 1/(8JCH) = 0.89 ms. Four gradient pulseswere applied along the z axis with ratios of 80, 20.1, 11 and −5,respectively, and a duration of 1 ms. Eighteen transients wereaccumulated for each increment. A square sine window function was



applied in both dimensions prior to Fourier transformation. CDCl3,pyridine-d5, and D2O were the solvents used for preparing NMRsolutions of the intermediates and the final product according to theirsolubility. A concentration of approximately 1 mM was used forpreparing the NMR samples. After adding the sample to the solvent ina vial, it was stirred for several hours on a magnetic stirrer tothoroughly mix and dissolve it, and finally centrifuged to settle downany undissolved solids. Approximately 0.6 mL of the sample solutionwas transferred to a 5 mm NMR tube, which was inserted in themagnet to perform the experiment. Chemical shifts were recorded inparts per million (ppm) and were referenced to solvent peaks (at 7.24and 77.23 ppm, for example, for 1H and 13C NMR in CDCl3).Elemental analyses were performed by the Microanalytical Service,

School of Chemistry and Molecular Biosciences, the University ofQueensland. The samples were extensively (ca. 24 h) dried in vacuo at60 °C. The C/N ratio determined with elemental analysis was used toevaluate the DS of the intermediates and the final product.The XPS analysis of chitosan, L-leucine and chitosan-N-L-leucine·

HCl was performed with a Kratos Axis Ultra Spectrophotometer(Kratos Analytical, Manchester, UK) equipped with monochromatizedaluminum X-ray source (powered at 10 mA and 15 kV), 165 mmradius hemispherical analyzer and 8-channel electron multiplier(channeltron) detection system. Samples were mounted onto stainlesssteel sample holders using double-sided adhesive tape and the spectrawere collected using an analysis area of 700 × 300 μm. Low-resolutionwide scans (survey spectrum) in the binding energy scale (0−1200 eVwith 1.0 eV steps) were collected at a constant analyzer pass energy of160 eV to identify the elements present. High-resolution narrow scans(multiplex spectra) were collected at 20 eV with 0.05 eV steps for O1s, N 1s, C 1s, and Cl 2p to identify their chemical status. The datawere processed using Casa XPS software Version 2.3.14. Bindingenergies (BE) of the various elements were referenced to the C 1s lineat 285.0 eV.The scanning electron microscopy (SEM) imaging of chitosan and

chitosan-L-leucine conjugate nanoparticles was done on an ultrahighresolution scanning electron microscope, JEOL 7001F (JEOLAustralasia Pty Ltd., Frenchs Forest, Australia). A tiny drop (5 μL)of nanoparticle suspension in hexane (100 μg/mL) was placed oncarbon adhesive tape adhered onto a properly labeled aluminum stub.The specimen stubs were then sputter coated with a thin layer of goldwith a Leica EM SCD005 sputter coater (Leica Microsystems, NorthRyde, Australia) at 30 mA for 1.5 min using an argon gas purge. Thespecimens were examined with the SEM under high vacuum using aspot size of 10 with an accelerating voltage of 20.0 kV and a workingdistance of 10 mm. Several photomicrographs (secondary electronimages) of the samples were recorded at different magnifications.The transmission electron microscopy (TEM) analysis was

conducted using a high resolution transmission electron microscope,JEOL 1010 (JEOL Australasia Pty Ltd., Frenchs Forest, Australia). Foranalysis, each nanoparticle sample was suspended in ethanol at aconcentration of 0.05−0.1 mg/mL. A tiny drop (2 μL) of thesuspension was placed on strong carbon film on a 400-mesh coppergrid and allowed to sit for 1−2 min until it became nearly dry. Thenthe sample was negatively stained with a drop (2 μL) of 1% uranylacetate solution (AJAX Univar, Ajax Finechem Pty Ltd., Australia) foranother 2−3 min, and any excess uranyl acetate was removed withfilter paper. The specimens were then examined on the TEM machineunder high vacuum using an accelerating voltage of 100 kV and a spotsize of 6. The images of samples were taken at different magnifications.The size distribution and polydispersity index (PDI) of chitosan and

chitosan-L-leucine conjugate nanoparticles were determined bydynamic light scattering using a ZetaSizer Nano S instrument(Malvern Instruments Ltd., UK) equipped with a 4 mW He−Nelaser operating at a wavelength of 633 nm. The analysis was performedat 25 °C with a detection angle of 173°. For analysis, the particles werediluted to an appropriate concentration (ca. 0.1%) with heavy mineraloil containing 1% span 80 as the dispersant and placed in a glasscuvette with square aperture (Sarstedt AG & Co., Numbrecht,Germany). The measurement was done in triplicate and averaged. Thereported size is the Z-average (or “cumulants mean”) for the particle

hydrodynamic diameters calculated from the intensity of scatteredlight, and the PDI is a width parameter, which was also calculated fromthe cumulants analysis of the signal intensity. Data acquisition andanalysis was done by Malvern Zetasizer Software version 6.32. Theviscosity and refractive index values of the dispersant, heavy paraffin oilwith 1% span 80 (used for data analysis) were determined to be 72.7cps (at 25 °C) and 1.482, respectively, by a Cannon-Fenske RoutineViscometer (Poulten, Self-e and Lee, Ltd. England), and an AbbeRefractometer (Atago Co., Ltd., Tokyo, Japan), respectively.

Cell Culture. For in vitro toxicological and inflammatory evaluationof chitosan, its L-leucine conjugate and their nanoparticles, thebronchial epithelial cell line, BEAS-2B (ATCC# CRL-9609) was used.Cell cultures were grown using 75 cm2

flasks in a humidified incubatorin an atmosphere of 5% CO2/95% air at 37 °C. The same conditionswere used for cell cultivation in subsequent experiments. The cellculture medium (CCM) used consisted of RPMI 1640 supplementedwith 10% heat-inactivated fetal bovine serum, 2 mM L-glutamine, 100μg/mL penicillin, and 100 μg/mL streptomycin sulfate.

In Vitro Evaluation of Cytotoxicity by MTT Assay. The MTTassay was performed according to previous reports36 with somemodifications. BEAS-2B cells were seeded at 5 × 104 cells/well in a 96-well plate and grown overnight at 37 °C. Cells were then incubatedwith fresh medium containing chitosan or the conjugate or theirnanoparticles at a range of concentrations (viz., 0.125, 0.25, 0.375,0.50, 1.0, 2.0, 4.0, 8.0, 12.0, and 16.0 mg/mL). After 12, 24, or 48 h, 50μL of MTT solution (0.5 mg/mL in PBS) was added to each well. Thecells were incubated for an additional 4 h, and the medium wasremoved. The formazan crystals generated by the cells were solubilizedwith 100 μL DMSO and the absorbance of each well was measured byan xMark Microplate Spectrophotometer (Bio-Rad Laboratories,USA) at 550 nm. Relative cell viability was determined from theequation: viability (%) = (absorbance of test−absorbance of blank)/(absorbance of untreated cells−absorbance of blank) × 100. Theconcentration causing 50% inhibition of cell growth (IC50) wasdetermined by regression analysis using the points on the steepportion of the concentration effect curve according to Adamson et al.37

Effect on Na Flu Transport across BEAS-2B Cell Monolayers.The effect of different concentrations of chitosan, the conjugate andtheir nanoparticles on the permeability of polarized BEAS-2B cellmonolayers was investigated using Na Flu as the permeability markeraccording to the method described by Grenha et al.19 with somemodifications. 1 ×105 BEAS-2B cells in a volume of 200 μL mediumwere seeded onto collagen-coated 0.4 μm polytetrafluoroethylene(PTFE) Transwell inserts (Corning Incorporated, USA) and grown in24-well plates. Transepithelial electrical resistance (TEER) wasmeasured daily using an EVOM electrode (Millipore Australia PtyLtd.). Fresh media were applied every second day. After 5 days whenthe TEER reached a stable maximum value, 200 μL of CCMcontaining 0.2 mg/mL of Na Flu was added to the apical chamber incombination with test substances at increasing concentrations (viz., 0,0.5, 1, 2, and 4 mg/mL). Cells were then incubated at 37 °C andbasolateral samples taken at 0.5, 1, 2, 4, 6, 12, 24, and 48 h. Passivetransport of Na Flu across epithelial layers was determined from astandard curve using a POLARstar OPTIMA fluorescent spectropho-tometer (BMG Labtech Inc., USA). The apparent permeabilitycoefficient (Papp) of Na Flu was determined from the equation: Papp(cm/s) = (the cumulative amount of Na Flu transported over time) ×(1/(transwell surface area (cm2) × the initial apical concentration ofNa Flu (mg/mL).

Immunocytochemistry. BEAS-2B cells were seeded at 5 × 104

cells/0.4 μm transwell inserts and grown in 24 well transwell chambersfor 6 days at 37 °C in 5% CO2, changing media every second day. Onday six, cells were washed twice in PBS, and then fixed with ice cold100% MeOH for five min at 4 °C. Cells were washed twice with PBS,and then blocked in 10% FCS in PBS + Tween 20 (PBST) (0.5% v/v)for 1 h at room temperature. Cells were then probed with primaryantibody, rabbit antizona occludins (ZO)-1 antibody (Invitrogen,AUS) in blocking buffer (1:1000) for 1 h at room temperature. Cellswere washed three times with PBST and secondary antibody, goatantirabbit IgG-AlexaFluor 488 (Invitrogen, AUS) (1:2000), and 4,6-



diamidino-2-phenylindole (DAPI) (125 ng/mL) for 1 h at roomtemperature in the dark. Inserts were washed three times with PBST,and then the transwell membrane was removed with a scalpel blade,placed on a slide and mounted with a coverslip in Prolong Gold(Invitrogen, AUS) overnight at room temperature in the dark. ZO-1expression and DNA was observed using a TCS SP5 X confocalmicroscope (Leica Microsystems, AUS).Effect on IL-8 Secretion. The proinflammatory effect of the test

materials was evaluated by measuring IL-8 secretion by BEAS-2B cellsaccording to the method reported38,39 with some modifications.Briefly, 5 × 104 BEAS-2B cells were seeded into a 96-well plate andgrown at 37 °C until 80−90% confluent. Fresh media containingincreasing concentrations of test substances (0, 0.5, 1, 2, and 4 mg/mL) were applied. After 24 h at 37 °C, supernatants were collectedand secreted IL-8 quantified using an IL-8 ELISA MAX kit (BiolegendInc., USA) as per manufacturer’s instructions.40

Statistical Analysis. All the pharmaceutical and cell lineexperiments were performed in triplicate, and the results are expressed

as mean ± standard deviation (SD). The statistical significance of thedifferences was assessed by one-way and two-way analysis of variance(ANOVA) with posthoc (Tukey-HSD) analysis; differences with a p-value of 0.05 or less were considered significant.

■ RESULTS AND DISCUSSION

Conjugation of L-Leucine with Chitosan. The freeamino group at C-2 of chitosan 1 was selected as the preferredsite for chemoselective conjugation of L-leucine. For controlledfunctionalization at C-2 and to confer sufficient organo-solubility,30,41 the C-6 position was required to be protectedas the trityl ether. Since chitosan per se is not sufficientlyorganosoluble to allow its reaction with trityl chloride, it wasfirst reacted with phthalic anhydride to give N-phthaloylchito-san 2.31,42 After tritylation at C-6, the phthaloyl group wasremoved by treatment with aqueous hydrazine to give 6-O-

Figure 1. FT-IR spectra of (A) chitosan, (B) N-phthaloyl-chitosan (2), (C) N-phthaloyl-3,6-di-O-acetylchitosan (2a), (D) N-phthaloyl-6-O-tritylchitosan (3), (E) N-phthaloyl-3-O-acetyl-6-O-tritylchitosan (3a), (F) 6-O-tritylchitosan (4), (G) N-(Boc-L-leucine)-6-O-tritylchtiosan (5), and(H) chitosan-N-L-leucine·HCl (6).



tritylchitosan 4. L-Leucine was conjugated to the amino groupat C-2 by reaction with Boc-leu-OSu to give N-(Boc-L-leucine)-6-O-trityl-chitosan 5. Finally, both trityl and Boc groups weredeprotected simultaneously by treatment with 4 M HCl indioxane to give the target conjugate, chitosan-N-L-leucine 6 asthe hydrochloride salt. The intermediates and the final product,chitosan-N-L-leucine·HCl 6 were characterized by a combina-tion of FT-IR, 1H and 13C NMR spectroscopy, and elementalanalysis. Compounds 5 and 6 were further characterized by 2D1H−13C ge-HSQC spectroscopy.The FT-IR spectra of the unreacted chitosan, the

intermediates, and the final product 6 in the synthetic pathwayare presented in Figure 1. In accordance with previouslypublished data,30,32,43 the spectra showed characteristicabsorptions for N-phthaloyl-chitosan 2 (phthaloyl CC1703 and 1774 cm−1, phthaloyl aromatic 718 cm−1), N-phthaloyl-6-O-trityl-chitosan 3 (trityl CC 1448 and 1490cm−1, trityl aromatics 699, 746, and 764 cm−1 in addition tophthaloyl signals in place), N-(Boc-L-leucine)-6-O-trityl-chito-san 5 (intensified amide-I signal at 1683 cm−1, Boc t-butyl/L-leucine isopropyl at 1367 and 1390 cm−1 in addition to tritylsignals in place) and chitosan-N-L-leucine·HCl 6 (amide-I 1628and 1672 cm−1, L-leucine isopropyl 1372 and 1390 cm−1). TheIR spectrum of 6-O-trityl-chitosan 4 confirmed the completeremoval of the phthaloyl group (disappearance of the signals at1712 and 1776 cm−1). The spectrum of chitosan-N-L-leucine·HCl confirmed the removal of the trityl group (disappearanceof signals at 702, 747, and 764 cm−1), and a reduction in theintensity of the doublet at 1372 and 1390 cm−1 suggested thatthe Boc group had been removed with the signal arising from L-leucine isopropyl only.The 1H and 13C NMR spectra of the final product 6 are

shown in Figure 2. The 1H and 13C NMR spectra of theintermediates and their derivatives and the 2D 1H−13C ge-HSQC spectra of the final two products 5 and 6 are alsoprovided in the Supporting Information (Figures S-1 to S-4).The solubility of 2 and 3 in common NMR solvents was notsufficient to obtain useful 1H and 13C NMR spectra and so,following the strategy of Nishimura et al.,44 they wereacetylated (Scheme 1B) to improve their solubility for NMRcharacterization. The NMR spectra of 2 and 3 in DMSO-d6 andpyridine-d5 have been reported previously.45,46 This differencein solubility probably arises from differences in the molecularweight and DDA of the chitosan used as the starting material.The 1H NMR spectrum of 4 in pyridine-d5 was in accord withthe IR data, confirming the selective removal of the phthaloylgroups. However, a 13C NMR spectrum of the compoundcould not be obtained due to poor solubility. The 1H and 13CNMR spectra of 5 in pyridine-d5 showed signals at δH 1.44 ppmand δC 28.9 ppm, respectively, for Boc t-butyl in addition tosignals for trityl aromatic, pyranose, and L-leucine protons andcarbons at their expected chemical shifts, thereby confirmingconjugation of Boc-L-leucine to 4. NMR signals were alsoobserved for CO of carbamate (156.9 ppm) and amide (L-leucine and N-acetyl) at 173.6 and 176.3 ppm, respectively, aswell as for the quaternary carbons of Boc and trityl groups at79.2 and 87.1 ppm, respectively. The final product chitosan-N-L-leucine·HCl 6 showed adequate solubility in water to givemore intense proton and carbon NMR signals in solution inD2O; however, while the 1H NMR signals of the compoundshowed considerable overlap, many carbon signals, especiallythose from pyranose carbons, were doubled up. Apparently, thiswas due to the presence of three different kinds of monomeric

units in the product, viz., unsubstituted, N-acetyl-substituted,and L-leucine-substituted glucosamine residues. This kind ofmultiplicity of carbon peaks in chitosan derivatives has alsobeen reported previously.47 As Boc t-butyl and L-leucineisopropyl had IR absorptions at the same positions (a doubletat around 1366 and 1390 cm−1), the NMR spectra wereimportant to definitively ascertain the conjugation of L-leucineand ultimate deprotection of the Boc group. The 2D 1H−13Cge-HSQC spectra of 5 and 6 showed clear cross peakscorrelating the carbon signals to proton signals (Figures S-3and S-4). The 2D data of 5 was helpful in confirming thechemical shifts of some carbons that did not give very well-defined peaks in the 1-D 13C NMR spectrum (viz. pyranosecarbons). The 2D data were also useful in assigning the signalsof different protons in the 1H NMR spectrum that showed a lotof overlap. One limitation of the 2D spectrum of thiscompound was the absence of peaks corresponding to theanomeric carbon, but the presence of corresponding peaks inthe 2D spectrum of 6 indicated that their absence from thespectrum of the former was simply due to lack of solubility.Similarly to 5, the 2D spectrum of 6 was also able to resolve theoverlappings of 1H NMR signals.The final product 6 was also subjected to XPS analysis (see

Supporting Information, Figure S-6). The high resolution XPSscan of N (Figure S-5A) showed a reduction in the intensity ofNH signal from 5.94% (in 1) to 0.68% (in 6) and elevation inthe intensity of N+ (from 0.78% to 2.99%) and Namide (from0.72% to 5.14%), as expected. Moreover, a high-resolution scanof C (Figure S-5 B) also showed the expected elevation in theC−C carbon signal from 12.39% (in 1) to 19.14% (in 6). The

Figure 2. 1H (A) and 13C (B) NMR spectra of chitosan-N-L-Leucine·HCl 6 (in D2O).



elemental analysis of most intermediates and the final product 6provided satisfactory results (see Materials and Methods).Chitosan and Conjugate Nanoparticles. Nanoparticles

of chitosan and the conjugate were prepared by a W/Oemulsion solvent evaporation technique. Being freely soluble inwater, in contrast to chitosan, no acidification was needed fordissolving the conjugate when preparing the conjugatenanoparticles. To our knowledge, this is the first report forpreparing particles of chitosan or any of its derivatives innanoscale by this technique using paraffin oil as the externalphase. The morphologies of nanoparticles were examined bySEM, and almost all particles appeared to be oblate, rather thanperfect spheres (Figure 3). The particles showed a tendency toaggregate, although there appeared to be some isolated singleparticles as well. In addition, we studied the morphologicalproperties of nanoparticles using TEM, which further revealedthat those nanoparticles are nonspherical with unsmoothsurfaces (Figure 4). The SEM analysis suggested that thenanoparticles had diameters predominantly in the range of 10−30 nm (Figure 3); however, SEM is not an appropriate methodfor determining the particle size.Using a zetasizer, particle size and size distributions of

nanoparticles were analyzed, and both the chitosan andconjugate nanoparticles showed a unimodal size distributionwith average diameters (Z-averages) of 134.4 ± 2.8 and 153.2± 1.2 nm, respectively (Figure 5A,B) and the correspondingpolydispersity indices were 0.58 ± 0.26 and 0.35 ± 0.08,respectively. The Z-average of conjugate nanoparticles was

larger than that of chitosan nanoparticles. The larger size ofparticles observed by te zetasizer probably suggests thepresence of nanoagglomerates arising from cohesion of verysmall individual particles. This is a common phenomenon withmicro- and nanosized particles, which results from high surfacearea and associated surface free energy of particles in this sizerange. Similar observations have been reported previously forvarious polymeric and nonpolymeric microparticles.48,49 Theaverage particle sizes measured by the zetasizer were muchlarger compared with the sizes of the particles observed bySEM. The actual reason behind this is unclear; however, thesmaller sizes revealed by the SEM could be due to shrinkageduring dehydration under the high vacuum conditions, thus

Scheme 1. (A) Synthetic Pathway for Conjugation of L-Leucine to Chitosan (1) and (B) Acetylation of N-Phthaloylchitosan (2)and N-phthaloyl-6-O-tritylchitosan (3)

Figure 3. Scanning electron micrographs of (A) chitosan and (B)conjugate nanoparticles. [Note: Dimension bar is shown under eachmicrograph.]



underestimating nanoparticle size. Therefore, the particle sizesmeasured by the hydrodynamic technique (zetasizer) are largerand more accurate. It is noteworthy that the principles of thetwo instruments are different in determining particle sizes andSEM has limited efficiency in determining particle sizeaccurately.Toxicological Evaluation of Chitosan, the Conjugate,

and Their Nanoparticles. The polymers were assessed, bothin neat and particulate forms, for their safety for pulmonarydelivery using cytotoxicity, trans-epithelial permeability and IL-8 secretion as the indicators of their toxicity and inflammatoryeffects on the bronchial epithelial cell line, BEAS-2B.The MTT assay assessed effects of increasing concentrations

(0.125 to 16 mg/mL) of chitosan, the conjugate, and theirnanoparticles on cell viability over three time periods, viz., 12,24, and 48 h. The 4 highest concentrations at which most of thetest samples showed a percent survival above 50% and beyondwhich it was found to fall abruptly (viz. 0.5, 1, 2, and 4 mg/mL), were chosen to perform the subsequent sodiumfluorescein transport and IL-8 secretion studies.The results of the MTT assay (Figure 6) and IL-8 induction

study (Figure 7) indicated that particulate forms of thepolymers were more cytotoxic and inflammatory than neatpolymers, with the conjugate nanoparticles appearing to be themost cytotoxic and inflammatory. However, the IC50 value ofthe conjugate nanoparticles following the longest exposuretested (48 h) was determined to be 2 mg/mL. Althoughchitosan is biocompatible, due to its broad molecular weightrange and heterogeneity, it produces different toxicologicalprofiles in varieties of cell lines. In addition, the toxicity ofchitosan nanoparticles in different cell lines is concentrationdependent.50 Therefore, actual toxicity profiles (low or highlevel) of chitosan nanoparticles prepared by different method-ologies are largely unclear. The toxicity levels of pharmaceut-

icals have been reported to be categorized as IC50 < 1 mg/L -very high toxicity; IC50 = 1−10 mg/L - high toxicity; and IC50>100 mg/L - low toxicity.51 Therefore, the outcomes of ourstudies (i.e., 2 mg/mL) belong to the low level of toxicity.Drugs administered by inhalation (e.g., beta agonists andcorticosteroids) are generally potent with a dose size notexceeding a few hundred micrograms. Therefore, the conjugatenanoparticles could be considered promising as a matrix forcontrolled release delivery through the pulmonary route. Theincreased toxicity of conjugate and its nanoparticles appears tobe reasonable if compared with TMC, another water-solublederivative that has been reported to be more toxic thanchitosan.22 This has been attributed to its increased solubilitythat allows a better interaction with anionic cell surfacecomponents. Previous literature suggests that nondegradableadditives, such as surfactants (e.g., (Pluronic P-105) orpolymers (e.g., poly(vinyl alcohol)),52 or residues from organicsolvents53 used in the emulsion method may contribute to thetoxicity of the prepared particles. The increased toxicity of theparticulate forms compared to the neat polymers observed inthe this study might be related to any remaining organicchemicals like span 80 or hexane used in the preparation andpurification of the nanoparticles. Nanoparticles have also beenreported to be internalized by cellular uptake into pulmonaryepithelial cells.54 This kind of possible uptake of thenanoparticles might have also contributed to the toxicity byallowing the particles to interact with subcellular organelles inaddition to the cell surface. The IL-8 release caused byconjugate nanoparticles at concentrations of 1 and 2 mg/mLwas 3783 ± 738 and 6062 ± 533 pg/mL, respectively, whichwere 5- to 8-fold higher than the basal level of IL-8 release (768± 133 pg/mL) and so raises some concern about their safetyfor respiratory delivery. However, there was no significantinduction of IL-8 at a concentration of 0.5 mg/mL. Previousliterature reported a wide range of constitutive IL-8 secretionfrom as low as 274 pg/mL20 to as high as 1000−10 000 pg/mL55 at different time-points of incubation from variouspulmonary epithelial cell lines. So, it is difficult to predict whatcould be the biologically relevant concentration of IL-8secretion for initiating pro-inflammatory effect; however,considering that there was no significant IL-8 inductioncompared with the control at a conjugate nanoparticleconcentration of 0.5 mg/mL, they could be consideredadequate for respiratory delivery of potent drugs with smalldose sizes.The trans-epithelial permeability study was performed using

sodium fluorescein as a fluid phase marker (Figures 8 and S-6).The results indicated that chitosan and its nanoparticles do notcause any significant change in permeability. On the other

Figure 4. Transmission electron micrographs (TEM), (A) chitosanand (B) conjugate nanoparticles. [Note: Dimension bar is shown undereach micrograph.]

Figure 5. Particle size (diameter, d (nm)) distribution of (A) chitosan and (B) conjugate nanoparticles (n = 3).



hand, the conjugate showed a mild but significant (p < 0.05)increase in permeability (estimated mean cumulative Na Flu

transports of 0.011458, 0.011842, and 0.011617 mg at 0.5, 1,and 2 mg/mL concentrations vs 0.010738 mg at 0 mg/mLconcentration (control) as per ANOVA), while the conjugatenanoparticles caused an overt reduction in permeability.Increased permeability with the conjugate is consistent withreports on the permeability enhancing effect of TMC.56,57 It isworth noting that cellular death causing gaps in the epitheliallayer might have contributed to the increased Na Flupermeability. Further studies are warranted to investigate thepermeability effect of the highest nanoparticle concentration.Using immunocytochemistry, the ZO-1 staining (Figure S-7)confirmed the presence of tight junctions under the cultureconditions used, consistent with previous studies.58 Thus, webelieve that the culture conditions used were appropriate todetermine transport across the cell monolayer. The reductionin the permeability shown by the conjugate nanoparticles doesnot fit with the reports on the micro-/ nanoparticles of TMCthat have been found to induce permeation enhancementsimilar to neat TMC.59,60 Further studies are required tounderstand this unexpected effect produced by the conjugatenanoparticles. Except for neat chitosan, the highest dose of 4mg/mL applied in the experiments showed a different trend incausing increase/decrease in permeability compared with thelower doses. This may be related to the high level of toxicity/cell death encountered at this dose; further studies arewarranted to explore the exact mechanism involved.

Figure 6. Effect of chitosan (A), the conjugate (C), and their nanoparticles (B and D) on the viability of BEAS-2B cell line. The % survival of BEAS-2B cells was determined by MTT assay after 12, 24 and 48 h exposure to increasing concentrations (viz. 0.125, 0.25, 0.375, 0.50, 1.0, 2.0, 4.0, 8.0,12.0, and 16.0 mg/mL) of suspension/solution of chitosan, the conjugate and their nanoparticles in cell culture medium. * = The concentration atwhich no significant difference was observed in % survival with control. Results presented as % survival relative to controls and expressed as the mean± SD (n = 3). p ≤ 0.05 was considered significant.

Figure 7. Effect of chitosan, the conjugate, and their nanoparticles onIL-8 secretion by BEAS-2B cells. The IL-8 levels in culturesupernatants were determined by ELISA 24 h after exposure of thecells with increasing concentrations (0, 0.5, 1, 2, and 4 mg/mL) of thetest samples suspended/dissolved in the cell culture medium. The IL-8levels were normalized to the number of cells by dividing the measuredvalues by the % viability of cells estimated by MTT assay. Resultsexpressed as the mean ± SD (n = 3) and p ≤ 0.05 was consideredsignificant.



■ CONCLUSIONSL-Leucine has been successfully conjugated to chitosan, giving anovel compound with improved solubility and thus paving theway for better utilization of chitosan in drug deliveryapplications. The work also established a method based onemulsion-solvent evaporation technique for preparing nano-particles of both chitosan and its conjugate suitable fordeposition to the deeper regions of the respiratory airways.The conjugate and its nanoparticles appeared to be relativelymore toxic and inflammatory than chitosan and its nano-particles; however, the level of toxicity (IC50 2 mg/mLfollowing 48 h exposure) and inflammatory effect (no inductionof IL-8 at 0.5 mg/mL) can still enable its utilization forpulmonary drug delivery unless the dose size of the drug to beincorporated is high. The leucine-conjugated nanoparticlesrequire further investigation to evaluate appropriate toxicitylevels at wider concentration ranges using different cell lines atdifferent exposure times. More conclusive inferences can,however, be drawn only after in vivo trials in animal models.

■ ASSOCIATED CONTENT*S Supporting Information1H and 13C NMR spectra of synthetic intermediates and thefinal product 6; 2D 1H−13C ge-HSQC data of the final twoproducts 5 and 6 in the synthetic pathway; XPS multiplexspectra of chitosan 1, L-leucine, and chitosan-N-L-leucine·HCl 6

for nitrogen and carbon and comparison of the effect ofchitosan, the conjugate, and their nanoparticles on Na Flutransport across the BEAS-2B cell monolayer; and confocalimages of ZO-1 expression in BEAS-2B cells grown on 0.4 μmtranswell inserts for 6 days. This material is available free ofcharge via the Internet at http://pubs.acs.org.

■ AUTHOR INFORMATION

Corresponding Author*Telephone: +61 7 3138 1899. Fax: +61 7 3138 1534. E-mail:[email protected].

Author ContributionsThe manuscript was written through contributions of allauthors. All authors have given approval to the final version ofthe manuscript.

NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTS

We thank Professor Prof Philip Hansbro, School of BiomedicalSciences and Pharmacy, The University of Newcastle, Australia,for providing the BEAS-2B cell line. This research wassupported by an International Postgraduate Research Scholar-ship (IPRS) to M.D.A.M. granted by the Australian Govern-ment through Queensland University of Technology.

Figure 8. Effect of chitosan (A), the conjugate (C), and their nanoparticles (B and D) on Na Flu transport across the BEAS-2B cell monolayer.Cumulative amounts of Na Flu transported to the basolateral chambers of transwells containing polarized cell monolayers were determined byspectrofluorometry at 0.5, 1, 2 4, 6, 12, 24, and 48 h time-points after apical treatment with increasing concentrations (viz., 0, 0.5, 1, 2, and 4 mg/mL)of suspension/solution of chitosan, the conjugate, and their nanoparticles in cell culture medium containing Na Flu (0.2 mg/mL). Results expressedas the mean ± SD (n = 3) and p ≤ 0.05 was considered significant.



http://pubs.acs.org

mailto:[email protected]

■ ABBREVIATIONSATCC, American type culture collection; ATR, attenuated totalreflectance; Boc-Leu-OSu, t-butyloxycarbonyl-L-leucine-succini-mide; CCM, cell culture medium; DDA, degree of N-deacetylation; DMF, N,N-dimethylformamide; DMSO, dimeth-yl sulfoxide; DPI, dry powder inhaler; DS, degree ofsubstitution; EDTA, ethylenediaminetetraacetic acid; ELISA,enzyme-linked immunosorbent assay; FT-IR, Fourier transforminfrared; IC50, median inhibitory concentration; ge-HSQC,gradient-enhanced heteronuclear single quantum correlation;IL-8, inerleukin-8; LB, line broadening; MTT, 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide; MW,molecular weight; Na Flu, sodium fluorescein; NMR, nuclearmagnetic resonance; PBS, phosphate-buffered saline; PDI,polydispesity index; SEM, scanning electron microscopy;SV40, Simian virus 40; TEER, trans-epithelial electricalresistance; TMC, trimethlychitosan chloride; XPS, X-rayphotoelectron spectroscopy

■ REFERENCES(1) Shahidi, F.; Arachchi, J. K. V.; Jeon, Y.-J. Trends Food Sci. Technol.1999, 10, 37−51.(2) Synowiecki, J.; Al-Khateeb, N. A. Crit. Rev. Food Sci. Nutr. 2003,43, 145−71.(3) VandeVord, P. J.; Matthew, H. W. T.; DeSilva, S. P.; Mayton, L.;Wu, B.; Wooley, P. H. J. Biomed. Mater. Res. 2001, 59, 585−90.(4) Onishi, H.; Machida, Y. Biomaterials 1999, 20, 175−182.(5) Denkbas, E. B.; Ottenbrite, R. M. J. Bioact. Compat. Polym. 2006,21, 351−368.(6) Krajewska, B. Enzyme Microb. Technol. 2004, 35, 126−139.(7) Bobu, E.; Nicu, R.; Lupei, M.; Ciolacu, F.; Desbrieres, J. Cell.Chem. Technol. 2011, 45, 619−625.(8) Mourya, V. K.; Inamdar, N. N.; Tiwari, A. Adv. Mater. Lett. 2010,1, 11−33.(9) Oshita, K.; Takayanagi, T.; Oshima, M.; Motomizu, S. Anal. Sci.2007, 23, 1431−1434.(10) Fu, G.; Yu, H.; Yuan, Z.; Liu, B.; Shen, B.; He, B. Artif. Cells,Blood Substitutes, Biotechnol. 2004, 32, 303−313.(11) Yi, S.-S.; Noh, J.-M.; Lee, Y.-S. J. Mol. Catal. B: Enzym. 2009, 57,123−129.(12) Learoyd, T. P.; Burrows, J. L.; French, E.; Seville, P. C. Eur. J.Pharm. Biopharm. 2008, 68, 224−234.(13) Learoyd, T. P.; Burrows, J. L.; French, E.; Seville, P. C. Int. J.Pharm. 2009, 372, 97−104.(14) Rabbani, N. R.; Seville, P. C. J. Controlled Release 2005, 110,130−140.(15) Raula, J.; Thielmann, F.; Naderi, M.; Lehto, V.-P.; Kauppinen, E.I. Int. J. Pharm. 2010, 385, 79−85.(16) Mason, R. J.; Broaddus, V. C.; Martin, T. R.; Jr., King, T. E.;Schraufnagel, D. E.; Murray, J. F.; Nadel, J. A., Ed.; Murray & Nadel’sTextbook of Respiratory Medicine, 5th ed.; Saunders Elsevier,Philadelphia, PA, 2010; p 2400.(17) Newhouse, M. T.; Dolovich, M. B. N. Engl. J. Med. 1986, 315,870−4.(18) Okamoto, H.; Shiraki, K.; Yasuda, R.; Danjo, K.; Watanabe, Y. J.Controlled Release 2011, 150, 187−195.(19) Grenha, A.; Grainger, C. I.; Dailey, L. A.; Seijo, B.; Martin, G. P.;Remunan-Lopez, C.; Forbes, B. Eur. J. Pharm. Sci. 2007, 31, 73−84.(20) Sivadas, N.; O’Rourke, D.; Tobin, A.; Buckley, V.; Ramtoola, Z.;Kelly, J. G.; Hickey, A. J.; Cryan, S.-A. Int. J. Pharm. 2008, 358, 159−167.(21) Fischer, D.; Li, Y.; Ahlemeyer, B.; Krieglstein, J.; Kissel, T.Biomaterials 2003, 24, 1121−1131.(22) Mao, S.; Shuai, X.; Unger, F.; Wittmar, M.; Xie, X.; Kissel, T.Biomaterials 2005, 26, 6343−6356.(23) Salama, R. O.; Traini, D.; Chan, H.-K.; Sung, A.; Ammit, A. J.;Young, P. M. J. Pharm. Sci. 2009, 98, 2709−2717.

(24) Zhang, W. F.; Zhou, H. Y.; Chen, X. G.; Tang, S. H.; Zhang, J. J.J. Mater. Sci.: Mater. Med. 2009, 20, 1321−1330.(25) Saedisomeolia, A.; Wood, L. G.; Garg, M. L.; Gibson, P. G.;Wark, P. A. B. Br. J. Nutr. 2009, 101, 533−540.(26) Drumm, K.; Attia, D. I.; Kannt, S.; Micke, P.; Buhl, R.; Kienast,K. Respiration 2000, 67, 291−297.(27) Reddel, R. R.; Ke, Y.; Gerwin, B. I.; McMenamin, M. G.;Lechner, J. F.; Su, R. T.; Brash, D. E.; Park, J. B.; Rhim, J. S.; Harris, C.C. Cancer Res. 1988, 48, 1904−9.(28) Graness, A.; Chwieralski, C. E.; Reinhold, D.; Thim, L.;Hoffmann, W. J. Biol. Chem. 2002, 277, 18440−18446.(29) Wong, C. K.; Li, M. L. Y.; Wang, C. B.; Ip, W. K.; Tian, Y. P.;Lam, C. W. K. Int. Immunol. 2006, 18, 1327−1335.(30) Holappa, J.; Nevalainen, T.; Savolainen, J.; Soininen, P.; Elomaa,M.; Safin, R.; Suvanto, S.; Pakkanen, T.; Masson, M.; Loftsson, T.;Jaervinen, T. Macromolecules 2004, 37, 2784−2789.(31) Nishimura, S.; Kohgo, O.; Kurita, K.; Kuzuhara, H. Macro-molecules 1991, 24 (17), 4745−8.(32) Zhang, F.; Bernet, B.; Bonnet, V.; Dangles, O.; Sarabia, F.;Vasella, A. Helv. Chim. Acta 2008, 91, 608−618.(33) Abd El-Hameed, M. D.; Kellaway, I. W. Eur. J. Pharm. Biopharm.1997, 44, 53−60.(34) Onishi, H.; Oosegi, T.; Machida, Y.; McGinity, J. W. Drug Dev.Ind. Pharm. 2005, 31, 597−605.(35) Dandge, B. H.; Dehghan, M. H. G. Int. J. ChemTech Res. 2009, 1,1036−1042.(36) Manca, M.-L.; Mourtas, S.; Dracopoulos, V.; Fadda, A. M.;Antimisiaris, S. G. Colloids Surf., B 2008, 62, 220−231.(37) Adamson, P. C.; Balis, F. M.; Arndt, C. A.; Holcenberg, J. S.;Narang, P. K.; Murphy, R. F.; Gillespie, A. J.; Poplack, D. G. CancerRes. 1991, 51, 6079−83.(38) Fujisawa, T.; Kato, Y.; Atsuta, J.; Terada, A.; Iguchi, K.; Kamiya,H.; Yamada, H.; Nakajima, T.; Miyamasu, M.; Hirai, K. J. Allergy Clin.Immunol. 2000, 105, 126−133.(39) Schulz, C.; Farkas, L.; Wolf, K.; Kraetzel, K.; Eissner, G.; Pfeifer,M. Scand. J. Immunol. 2002, 56, 294−302.(40) Human IL-8 ELISA MAX Deluxe Kit. Manufacturer’s Protocol;BioLegend, Inc., San Diego, USA (http://www.biolegend.com/media_assets/pro_detail/datasheets/431504.pdf), 2011.(41) Holappa, J.; Nevalainen, T.; Soininen, P.; Elomaa, M.; Safin, R.;Masson, M.; Jaervinen, T. Biomacromolecules 2005, 6, 858−863.(42) Kurita, K.; Ichikawa, H.; Ishizeki, S.; Fujisaki, H.; Iwakura, Y.Makromol. Chem. 1982, 183, 1161−9.(43) Stefanescu, C.; Daly, W.; Negulescu, I. Polym. Prepr. (Am. Chem.Soc., Div. Polym. Chem.) 2008, 49, 793−794.(44) Nishimura, S.; Kohgo, O.; Kurita, K.; Vittavatvong, C.;Kuzuhara, H. Chem. Lett. 1990, 243−6.(45) Kurita, K.; Ikeda, H.; Shimojoh, M.; Yang, J. Polym. J. (Tokyo,Jpn.) 2007, 39, 945−952.(46) Yu, H.; Wang, W.; Chen, X.; Deng, C.; Jing, X. Biopolymers2006, 83, 233−242.(47) Heras, A.; Rodriguez, N. M.; Ramos, V. M.; Agullo, E.Carbohydr. Polym. 2000, 44, 1−8.(48) Li, H.-Y.; Birchall, J. Pharm. Res. 2006, 23, 941−950.(49) Steckel, H.; Rasenack, N.; Villax, P.; Muller, B. W. Int. J. Pharm.2003, 258, 65−75.(50) Hu, Y.-L.; Qi, W.; Han, F.; Shao, J.-Z.; Gao, J.-Q. Int. J.Nanomed. 2011, 6, 3351−3359.(51) Besse, J. P.; Garric, J., In Pharmaceuticals in the Environment:Current Knowledge and Need Assessment to Reduce Presence and Impact,Roig, B., Ed.; IWA Publishing: London, 2010; pp 137−168.(52) Hong, Y.; Li, Y.; Yin, Y.; Li, D.; Zou, G. J. Aerosol Sci. 2008, 39(6), 525−536.(53) de Villiers, M. M.; Aramwit, P.; Kwon, G. S. Biotechnol. Pharm.Aspects 2009, 659.(54) Lai, Y.; Chiang, P.-C.; Blom, J. D.; Li, N.; Shevlin, K.; Brayman,T. G.; Hu, Y.; Selbo, J. G.; Hu, L. G. Nanoscale Res. Lett. 2008, 3, 321−329.(55) Witschi, C.; Mrsny, R. J. Pharm. Res. 1999, 16, 382−390.



http://www.biolegend.com/media_assets/pro_detail/datasheets/431504.pdf

http://www.biolegend.com/media_assets/pro_detail/datasheets/431504.pdf

(56) Thanou, M.; Florea, B. I.; Langemeyer, M. W. E.; Verhoef, J. C.;Junginger, H. E. Pharm. Res. 2000, 17, 27−31.(57) van der Merwe, S. M.; Verhoef, J. C.; Verheijden, J. H. M.;Kotze, A. F.; Junginger, H. E. Eur. J. Pharm. Biopharm. 2004, 58, 225−235.(58) Heijink, I. H.; Brandenburg, S. M.; Noordhoek, J. A.; Postma, D.S.; Slebos, D. J.; van Oosterhout, A. J. M. Eur. Respir. J. 2010, 35, 894−903.(59) Coco, R.; Plapied, L.; Pourcelle, V.; Jerome, C.; Brayden, D. J.;Schneider, Y.-J.; Preat, V. Int. J. Pharm. 2013, 440, 3−12.(60) Sandri, G.; Bonferoni, M. C.; Rossi, S.; Ferrari, F.; Boselli, C.;Caramella, C. AAPS PharmSciTech 2010, 11, 362−371.



Documents

Synthesis and Toxicological Evaluation of a Chitosan- l -Leucine Conjugate for Pulmonary Drug Delivery Applications