lsevier.com/locate/jconrel

Journal of Controlled Release

Enhancement of bronchial octreotide absorption by chitosan and N-trimethyl

chitosan shows linear in vitro/in vivo correlation

Bogdan I. Florea a,b,*, Maya Thanou c, Hans E. Junginger a, Gerrit Borchard a,d

a Division of Pharmaceutical Technology, Leiden/Amsterdam Center for Drug Research, P.O. Box 9502, 2300 RA Leiden, The Netherlandsb Department of Bio-organic Synthesis, Leiden Institute of Chemistry, P.O. Box 9502, 2300 RA Leiden, The Netherlands

c Genetic Therapies Centre, Department of Chemistry, Flowers Building, Imperial College London, London, SW7 2AZ, UKd Enzon Pharmaceuticals, Inc., 20 Kingsbridge Road, Piscataway, NJ, 08854, USA

Received 20 June 2005; accepted 4 October 2005

Available online 2 November 2005

Abstract

Chitosan is a biocompatible polysaccharide of natural origin that can act as a permeation enhancer. In this study, we used an integral in vitro/in

vivo correlation approach to: a) investigate polysaccharide-mediated absorption kinetics of the peptide drug octreotide across mammalian airway

epithelium, b) assess formulation toxicity, c) correlate the mechanism of permeation enhancement. The 20% and 60% N-trimethylated chitosan

derivatives (TMC20 and TMC60) were synthesized by alkaline methylation using chitosan as starting material. Octreotide was administered in

control buffers or in 1.5% (w/v) gel-phase formulations of pH 5.5 for chitosan and pH 7.4 for TMCs. In vitro, reconstituted Calu-3 cell monolayers

were used for trans-epithelial electrical resistance (TEER), transport and cytotoxicity assays. Intratracheal instillation in rats was used to determine

octreotide kinetics and formulation toxicity in vivo.

Chitosan, TMC20 and TMC60 decreased TEER significantly and enhanced octreotide permeation in vitro by 21-, 16- and 30-fold. In vivo,

sustained release properties of the formulations were observed and the bio-availability was enhanced by 2.4-, 2.5- and 3.9-fold, respectively.

Interestingly, we found a linear in vitro/in vivo correlation between calculated absorption rates (R2=0.93), suggesting that the permeation

enhancement by polysaccharides, both in vitro and in vivo, proceeds via an analogous mechanism. Cell viability and histology studies showed that

the TMCs are safer than chitosan and that Calu-3 cell monolayers are a valuable model for predicting paracellular transport kinetics in airway

epithelia. Additionally, cationic polysaccharides are promising enhancers for peptide drug absorption with prospect for sustained-release

formulations.

D 2005 Elsevier B.V. All rights reserved.

Keywords: Chitosan; N-Trimethyl chitosan (TMC); Octreotide; Calu-3; Pulmonary delivery; In vitro/in vivo correlation

1. Introduction

Peptide and protein drugs are hydrophilic macromolecules

of great therapeutic value. Regrettably, they permeate poorly

across epithelial barriers and are commonly administered by

distressing parenteral injections that lower the patient com-

pliancy of the therapy. Delivery of peptide drugs by

inhalation might be a more patient compliant alternative [1].

Relevant bio-availability can be achieved by efficient target-

ing of the alveolar region of the lung with mono-dispersed

0168-3659/$ - see front matter D 2005 Elsevier B.V. All rights reserved.

doi:10.1016/j.jconrel.2005.10.001

* Corresponding author. Department of Bio-organic Synthesis, Leiden

Institute of Chemistry, P.O. Box 9502, 2300 RA Leiden, The Netherlands.

Tel.: +31 71 527 4776; fax: +31 71 527 4349.

E-mail address: b.florea@chem.leidenuniv.nl (B.I. Florea).

particle formulations with aerodynamic diameters of 2–3 Am[2]. However, scintigraphy data show that 30% to 50% of the

inhaled dose impacts at airway bifurcations [3] and is

subsequently excluded from systemic absorption by muco-

ciliary clearance. We hypothesized that this loss in dose

efficacy in the upper airways might be compensated by

peptide drug co-administration with polysaccharide materials

that can safely augment peptide drug permeation across the

bronchial barrier. To test this hypothesis we used an integral

in vitro/in vivo study approach to (i) investigate the

pulmonary delivery of octreotide, (ii) explore the use of

Calu-3 cell monolayers as in vitro model for passive

permeation across the bronchial epithelium, (iii) assess the

bronchial permeation and toxic effects of cationic polysac-

charides in gel-phase formulations, (iv) correlate rates of

110 (2006) 353 – 361

www.e

B.I. Florea et al. / Journal of Controlled Release 110 (2006) 353–361354

absorption and (v) compare the mechanism of permeation in

vitro and in vivo.

In this study, the octapeptidic somatostatin analog octreotide

(see Fig. 1a) was used as model peptide drug. Octreotide is

imperative for the therapy of acromegaly, carcinoid syndrome

and endocrine tumours of the GI tract [4], is metabolically

stable and is 7000 times more potent than the parent drug

somatostatin [5]. Despite its stability, the oral bio-availability

of octreotide in humans is negligible (0.3%, Ref. [6]) and its

pulmonary absorption is also expected to be low given the

reported 1% pulmonary bio-availability for somatostatin [7].

Importantly, in a previous study in rats and pigs [8,9], we

significantly enhanced the oral uptake of octreotide by co-

administration with chitosan and its derivates.

Chitosan or polyh(1Y4)d-glucose-2-amine is a linear

polysaccharide of natural origin [10], shows bioadhesive and

biocompatible properties and is a popular permeation enhancer

[11]. However, chitosan has the disadvantage of poor solubility

at physiological pH due to its relatively low pKa value (5.5). To

overcome this problem, N-trimethylated chitosan derivates

(TMC, see Fig. 1c) were synthesized [12]. The TMCs have

discrete degrees of substitution (20% and 60%), permanent

cationic charge densities and superior solubility at neutral pH.

Noticeably, several studies have shown that chitosan and TMC

can enhance the paracellular permeation of hydrophilic

compounds via a largely unknown mechanism that involves

transient and reversible modulation of tight-junctional com-

plexes between adjacent epithelial cells [13].

In functional epithelial tissues, the paracellular route is

regulated mainly by tight-junctional complexes, which confer a

virtually impermeable barrier to peptide drugs [14]. In vitro, the

tightness of this intercellular barrier can be determined by

measuring the transepithelial electrical resistance (TEER) and

the paracellular apparent permeability (Papp). The effects of the

polysaccharides on the TEER, the Papp of octreotide and cell

viability were studied in reconstituted, air–fluid interface

grown Calu-3 cell monolayers. The human airway epithelium

cell line Calu-3 was used because reconstituted monolayers

Fig. 1. Chemical structures of octreotide (a), chitosan (b) and N-trimethylated

chitosan (TMC, c) derivative. TMCs with 20% or 60% degree of trimethylation

were named TMC20 and TMC60, respectively.

show features of the normal airway epithelium such as tight

junctions, high TEER values and mucous excretions [15]. In

vivo, octreotide formulations were administered by intratra-

cheal instillation to ensure dose deposition in the bronchial

region of anaesthetized male Wistar rats. An important question

is whether the polysaccharides act via a similar mechanism

both in vitro and in vivo. This question can be addressed via a

correlation plot of the in vitro versus the in vivo absorption

rates. A linear correlation suggests, at least, an analogy in the

mechanism between in vitro and in vivo permeation enhance-

ment.

Here, we report the potential use of chitosan, TMC20 and

TMC60 as pulmonary absorption enhancers and demonstrate

the in vitro/in vivo correlation in mechanism of polysaccha-

ride-induced permeation for the therapeutic macromolecule

octreotide across Calu-3 cell monolayers and rat airway

epithelium, respectively.

2. Materials and methods

2.1. Materials

Octreotide acetate (Fig. 1a) and the radioimmunoassay

(RIA) kit for determination of octreotide in plasma were kindly

provided by Dr. P. Marbach (Novartis Pharma AG, Basel,

Switzerland). Chitosan of 200–400 Am particle size (Seacure

244, 93% N-de-acetylated, viscosity 40 mPa s, Mw 100–500

kDa) was a gift from Pronova AS (Drammen, Norway). Hank’s

Balanced Salt Solution (HBSS, pH 7.4) and Dulbecco’s

Modified Eagle Medium (DMEM, 4.5 g/l glucose, pH 7.4)

were purchased from Gibco BRL (Basel, Switzerland).

Characterised foetal calf serum (FCS) was from Hyclone

(Perbio Science, Helsingborg, Sweden). N-[2-hydroxyethyl]pi-

perazine-N V-[2-ethanesulfonic acid] (HEPES), (3-[4,5-

dimethylthiazol-2-yl]-2,5-diphenyltetrazolium) bromide

(MTT) assay and all other chemicals of analytical grade were

obtained from Sigma-Aldrich Chemie (Zwijndrecht, The

Netherlands).

2.2. Methods

2.2.1. TMC synthesis

N-Trimethylated chitosan chloride (TMC) was synthesised

from chitosan (Fig. 1b) via a two-step methylation procedure

described by Sieval et al. [12]. After the first methylation step,

TMC (Fig. 1c) with a degree of substitution (DS) of 20% was

isolated and named TMC20. TMC20 underwent a second

methylation step and yielded TMC with a DS of 60% that was

called TMC60. The degree of substitution was determined by1H NMR on a 600 MHz spectrometer (Bruker, Fallanden,

Switzerland).

2.2.2. Cell culture

Calu-3 cells (# HTB-55) were purchased from the American

Type Culture Collection (ATCC, Rockville, MD, USA) at

passage number (PN) 19. Calu-3 cells were seeded at 105 cells/

cm2 on collagen-coated permeable supports with pore size of

B.I. Florea et al. / Journal of Controlled Release 110 (2006) 353–361 355

0.4 Am called Transwells\ (Corning Costar, Schiphol-Rijk,

The Netherlands). The cells were cultured at an air–fluid

interface (i.e. apical culture medium removed 1 day after

seeding) at 37 -C in a 90% humidified incubator with 5% CO2.

The experiments were performed in 18 days old, differentiated

and polarised Calu-3 cells of PN 20 to PN 30.

2.2.3. TEER studies

Calu-3 cells were cultured on small size Transwell inserts

(area 0.33 cm2) for 18 days at air–fluid interface. The TEER

was measured with a Millicell\-ERD apparatus equipped with

chopstick electrodes (Millipore Corp., Bedford, MA, USA).

Chitosan was aseptically dissolved at 1.5% (w/v) in HBSS,

buffered with 30 mM HEPES (HBSS/HEPES), of pH 5.5,

while 1.5% (w/v) TMC20 and TMC60 were dissolved in

HBSS/HEPES of pH 7.4. The polymer formulations were

allowed to swell overnight. Two hours prior to the TEER

studies, the culture medium was removed and Calu-3 cells

were equilibrated in 1 ml HBSS/HEPES pH 7.4 in the

basolateral chamber and 200 Al HBSS/HEPES, pH 5.5 or pH

7.4, in the apical compartment. At t =0, the apical medium was

replaced by 200 Al chitosan, TMC20 or TMC60 formulations

or by 200 Al HBSS/HEPES of pH 5.5 or 7.4 that served as

controls. The TEER was measured at t=120 and 60 before

administration and t =0, 30, 60, 90, 120, 150, 210 and 240 min

after administration. At the end of the experiment, the cells

were washed 3 times with HBSS/HEPES, changed to culture

medium and kept 16 h in the incubator to determine the

recovery of the monolayer integrity.

The TEER value for every monolayer at t =0 min was

normalised to 100% to exclude inter-group differences caused

by differences in initial TEER value between monolayers. Each

data point shows the meanT standard deviation (SD) of six

experiments with triplicate determination.

2.2.4. Octreotide transport studies

Calu-3 cells were cultured on large size Transwell inserts

(area 4.71 cm2) for 18 days at air–fluid interface. Octreotide

(Mw 1004 Da) was dissolved in HBSS/HEPES of pH 5.5 or

7.4 and spiked to an end concentration of 190 AM in 1.5% (w/

v) chitosan and TMC formulations in HBSS/HEPES. Two

hours prior to the transport studies, the culture medium was

removed and Calu-3 cells were equilibrated in 2 ml HBSS/

HEPES pH 7.4 in the basolateral chamber and 1 ml HBSS/

HEPES, pH 5.5 or pH 7.4, in the apical compartment. At t=0,

the apical medium was replaced by 1 ml chitosan, TMC20 or

TMC60 formulations containing octreotide or by 1 ml

octreotide solutions in HBSS/HEPES of pH 5.5 or 7.4 that

served as controls. Samples of 200 Al were withdrawn from the

basolateral chamber at t =0, 15, 30, 45, 60, 90, 120, 150, 180,

210 and 240 min and replaced by fresh HBSS/HEBES pH 7.4.

The octreotide concentration in the samples was assayed by

isocratic HPLC analysis method on a Spectra Physics P200

system (Eindhoven, The Netherlands). A reversed phase

(Applikon, Schiedam, The Netherlands) C18 column (10

cm�3.0 mm, particle size 5 Am) was used as stationary

phase. The mobile phase was aqueous 0.1 M ammonium

acetate solution of pH 8.2 and acetonitrile (67 :33 v/v). At a

flow-rate of 1 ml/min, using a 100 Al injection loop and UV-

detection at 218 nm, the retention time (tret) of octreotide was

4.5 min and the detection limit was 60 ng/ml.

The apical to basolateral transport of octreotide across Calu-

3 cells is expressed as the percentage transported (%) or

apparent permeability (cm/s). Calculation of the apparent

permeability (Papp) has the advantage of being independent

of experimental design, surface area, time of experiment and

drug concentration. Papp is calculated using Eq. (1):

Papp ¼ k IVR=AI60: ð1Þ

The volume in the receiver chamber VR (ml) was 2 ml and the

surface area A was the filter size (4.71 cm2). The transport rate

constant k (min�1) was determined by linear regression from

the linear part of the Cumulative-Fraction-Absorbed (FAcum)

versus time. FAcum was calculated by a summation of the

amount of octreotide transported across the cell monolayer in

time (mi) divided by the total amount applied in the donor

chamber as shown in Eq. (2):

FAcum ¼ ~mi=mtot: ð2Þ

Transport enhancement ratios (ER) were calculated from the

apparent permeability values according to Eq. (3):

ER ¼ Papp;polymer=Papp;control: ð3Þ

Data are presented as meansTSD for at least six experi-

ments with triplicate determination. Comparison tests were

performed by using an unpaired Student’s t-test (two tailed).

P <0.05 was considered significant.

2.2.5. Cell viability assay

Viability of confluent, 18 days old, Calu-3 cell monolayers

was assessed by an MTT colorimetric assay in 96-well plates.

Prior to the 1 h MTT (5 mg/ml in HBSS/HEPES) treatment, the

Calu-3 cells were exposed for 3 h to chitosan, TMC20 and

TMC60 at the same end concentrations used for the TEER and

transport experiments. HBSS/HEPES of pH 5.5 and 7.4 served

as controls. After lysis of the cells in 0.01% NaOH/1% SDS (v/

v), the absorbance was measured at 590 nm in a Bio-Rad 96-

well plate reader (Alphen a/d Rijn, The Netherlands). Values of

8 measurements were normalised to 100% for exposure to cell

culture medium (DMEM/FCS). Data are presented as

meansTSD for three experiments and statistical significance

was assessed using Student’s t-test with P <0.05 taken as

significant.

2.2.6. In vivo studies

The animal studies were approved and performed according

to the guidelines of the Ethical Committee of the Leiden

University. Male Wistar rats of an average body weight of 318

g (Charles River, Maastricht, The Netherlands) were anaes-

thetized with Hypnorm\ (1.5 ml/kg) and Dormicum\ (500 Ag/kg) and the body temperature was maintained at 37 -C using

thermostated mattresses. A tracheal incision was made between

the 3rd and 4th cartilage ring below the thyroid gland and a

flexible canula (<, 1.1 mm) was inserted until the principal

Fig. 2. Effect of cationic polysaccharides on air– fluid interface grown and

differentiated Calu-3 cell monolayers in vitro. TEER measurements of controls

and 1.5% gel-phase polysaccharide formulations are expressed as percentage o

the value at t =0 min, indicated by the arrow (a). Cumulative apical to

basolateral transport from 1 ml of 190 AM octreotide in control and gel-phase

formulations (b).

B.I. Florea et al. / Journal of Controlled Release 110 (2006) 353–361356

bifurcation. In order to prevent bronchial obstruction, the

torsos were kept in an upright position by suspending the

animals by their incisor teeth. Octreotide was dissolved to an

end concentration of 0.97 mM in 0.9% saline containing

1.5% chitosan (pH 5.5), TMC20 (pH 7.4) or TMC60 (pH

7.4). From these solutions, 200 Al were instilled into the

trachea of the animals, the canula was removed and the

tracheal incision was closed with minute glue (Pattex, Henkel,

Duesseldorf, Germany). Solutions of octreotide in 0.9% saline

of pH 5.5 and 7.4 were instilled as controls. Every

formulation was tested in a group of 6 rats. Blood samples

of 200 Al were taken at t=0, 30, 60, 90, 120, 180 and 240

min by tail incision according to Fluttert et al. [16]. A group

of 6 animals received a 200 Al i.v. bolus containing 190 AMoctreotide in 0.9% saline in order to determine the pharma-

cokinetics (absolute bio-availability) of octreotide. Plasma

was collected by cold centrifugation for 10 min at 10,000 g

and kept at �20 -C. Octreotide concentrations in plasma

were determined by RIA analysis [4].

2.2.7. Histo-pathology assessment

After the experiment, the animals were sacrificed, the lungs

were excised, carefully inflated with air, fixed in 3.7%

paraformaldehyde in 0.9% NaCl overnight and stored in 70%

EtOH. Fixed lungs were embedded in paraffin; slices of 3 Amthickness were sectioned and stained with hematoxylin/eosin.

Histological evaluation was performed by light microscopy on

a Zeiss Axioskop microscope equipped with a MC80 camera

(Carl Zeiss BV, Weesp, The Netherlands). Photographs were

taken using the 40� magnification objective.

2.2.8. Data analysis and statistics

The in vivo experiments were performed in six groups of 6

rats (n =6). Each data point represents the meansTSDoctreotide plasma concentration from 6 animals. The values

for AUC and clearance (Cl) after i.v. administration of

octreotide were determined using the WinNonlin Professional

version 3.1 (Scientific Consultants, Inc., Lexington, KY, USA)

software assuming a two-compartment i.v. bolus, no lagtime,

1st order elimination model (PK Model 7, 1/(Y^^2). The areas

under the plasma curve (AUC) for intratracheal absorption

were estimated using the linear trapezoidal rule. The AUC

values were tested for statistical significance by two-factor

ANOVA with a P <0.05 taken as significant. Absolute bio-

availability values after intratracheal administration were

calculated according to Eq. (4):

F ¼ AUCit � Divð Þ= AUCiv � Ditð ÞÞð � 100 ð4Þ

in which F is the absolute bio-availability and D is the

administered dose. Enhancement ratios (ER) were calculated

from the AUC values in analogy to Eq. (3).

The in vitro/in vivo correlation of octreotide absorption data

was performed by plotting the Papp values from transport

studies across Calu-3 cells versus the rate of absorption in vivo.

The rate of octreotide absorption (R0) after intratracheal

instillation was calculated assuming the total clearance (Cl)

after i.v. administration as representative and considering the

plasma concentration-time profiles analog to infusion at

constant rate, according to Eq. (5):

R0 ¼ Css � Cl: ð5Þ

The steady state octreotide concentration (Css) at t =30 min

was used for all formulations. The values of the means were

correlated by linear regression using the Microsoft Excel 97 SR

software.

3. Results

3.1. Chitosan and TMC reduce TEER in vitro

We monitored the TEER of differentiated, air – fluid

interface cultured Calu-3 cell monolayers in the presence and

absence of polysaccharide formulations (Fig. 2a). All formula-

tions were isotonic with osmolality values of 300 mOsm. After

an initial increase in TEER, due to changing from culture

medium to HBSS/HEBES (t =�120 min), the TEER returned

to normal values (840T25 V cm2) during the next hour. The

TEER of Calu-3 monolayers was not affected by the apical pH

of the controls. Apical TMC20 induced an immediate decrease

in TEER by 35% followed by a gradual increase to a constant

value of 10% below the initial value. Chitosan decreased TEER

by 50% and leveled off at 40%, while TMC60 induced a 65%

decrease that equilibrated around 50% below the initial value.

After the experiment, the monolayers were rinsed and allowed

to recover overnight in culture medium in the incubator. The

TEER values recovered to some 90% of the initial (t =0)

f

Table 2

In vitro permeation parameters of octreotide and cell viability in Calu-3 cells

Formulation Papp 10�7

(cm/s)

ER Cell viability

(% culture medium)

TMC60 (pH 7.4) 2.42T0.39 P <0.01 30T4.5 102T12

Chitosan (pH 5.5) 2.53T0.38 P <0.01# 21T1.7 68T7#

TMC20 (pH 7.4) 1.24T0.40 P <0.01 16T3.3 103T14

Control (pH 5.5) 0.12T0.03 P <0.01 1 89T11

Control (pH 7.4) 0.08T0.02 n.s. 1 96T17

Apparent permeability coefficient ( Papp) and enhancement ratio (ER) values

for octreotide acetate permeation across air– fluid interface grown Calu-3 cell

monolayers, calculated from data in Fig. 2b (n =6; meansTSD). Also presented

viability assay of 18 days old, confluent Calu-3 cells exposed to culture

medium (100% viability), controls and polysaccharide gel formulations (n =8);

n.s. not significant; #compared to control pH 5.5, all other statistical values are

compared to control pH 7.4.

B.I. Florea et al. / Journal of Controlled Release 110 (2006) 353–361 357

values, as shown in Table 1, indicating that the effect of the

polymers on the TEER is reversible.

3.2. Chitosan and TMC enhance octreotide transport in vitro

Transport studies with octreotide were performed to

determine the functionality of chitosan, TMC20 and TMC60

as permeation enhancers across the airway epithelium in vitro.

Gel-phase formulations containing octreotide were translucent

indicating that octreotide and cationic polysaccharides do not

aggregate in solution. The profiles of the cumulative transport

of octreotide across Calu-3 monolayers in the apical to

basolateral direction are presented in Fig. 2b. In the control

groups, the transport of octreotide was low, confirming the

TEER data that Calu-3 cells form tight monolayers. TMC20

facilitated the cumulative permeation of 1.5% of the dose from

the apical to the basolateral compartment during 4 h. Chitosan

and TMC60 increased the octreotide permeation up to 3% of

the applied dose in the time course of the experiment. Overall,

the cumulative transport profiles showed a linear increase with

time, which followed a zero order kinetic profile. We have

previously performed 14C-mannitol permeation studies across

Calu-3 cell monolayers that showed similar zero order kinetic

profiles to octreotide (data not shown) and higher permeation

values because of the lower molecular weight of mannitol

compared to octreotide. Such zero order absorption profiles

across Calu-3 monolayers suggest that octreotide transport

occurs via passive (paracellular) diffusion after tight junction

modulation by the polysaccharide formulations.

From the transport data, we calculated the apparent

permeability (Papp) and enhancement ratio (ER), which are

the in vitro permeation parameters of octreotide across Calu-3

monolayers. Table 2 shows that, although in the control solution

the octreotide transport was negligible, at pH 5.5 there was a

slight but significant effect on the Papp compared to the control

at physiological pH. Chitosan, TMC20 and TMC60 also

increased the Papp significantly when compared to control(s)

(P <0.05). The ER values for the controls were set at 1 and were

calculated for each enhancer respective to the pH of the

formulation. This calculation revealed that although chitosan

induced a Papp value of a magnitude comparable to TMC60, the

ER for chitosan is significantly lower (P <0.05) than for

TMC60 when corrected for pH effects. These results show that

the cationic polysaccharides are able to facilitate the transport of

the hydrophilic compound octreotide by modulating the

paracellular integrity of impermeable Calu-3 cell monolayers.

Table 1

TEER recovery of Calu-3 cell monolayers

Formulation TEER t240 (%) TEER tend (%)

TMC60 (pH 7.4) 52T5 88T4

Chitosan (pH 5.5) 61T3 93T3TMC20 (pH 7.4) 89T2 100T2

Control (pH 5.5) 100T5 100T5

Control (pH 7.4) 100T3 100T3

TEER values measured at t =240 min and after washing of the monolayer

followed by 16 h recovery in culture medium at 37 -C, 5% CO2 in the

incubator.

3.3. Chitosan and TMC toxicity in vitro

The MTT assay uses a tetrazolium salt that is oxidised by

mitochondrial dehydrogenases, in living cells only, yielding a

dark blue formazane product that can be quantified photo-

spectrometrically. Damaged or dead cells show reduced or no

dehydrogenase activity. The controls at pH 7.4 and 5.5 (see

Table 2) showed no significant decrease in dehydrogenase

activity. Chitosan diminished the cell viability by 30%. The

high viscosity and bioadhesiveness of the chitosan formulation

might have caused some loss of cells during the washing steps,

although slight toxic effects cannot be excluded. TMC20 and

TMC60 showed a slight increase in dehydrogenase activity.

Possibly, cells sensing the reduced intercellular adhesion might

react by accelerating their transcription (protein production)

and metabolic rates in order to compensate for the loss in

adhesion. This study indicates that the effects of TMC20 and

TMC60 on the TEER decrease and the enhancement of

paracellular octreotide transport are not caused by membrane

damaging and cytotoxicity effects.

3.4. Chitosan and TMC enhance bronchial octreotide absorp-

tion in vivo

The permeation of octreotide and the local effects of the

polysaccharide formulations at pH 5.5 and 7.4 on the

pulmonary tissues were studied by intratracheal instillation in

vivo in rats (Fig. 3). Per group of 6 rats, the animals received an

Fig. 3. Plasma profiles of octreotide absorption from control and polysaccharide

formulations after intratracheal instillation in rats. Each data point represents

the meansTSD of octreotide plasma concentration from 6 animals.

Table 3

In vivo absorption parameters of octreotide after intravenous or intratracheal administration in rats

Formulation AUC (ng min/ml) F (%) ER Css (ng/ml) R0 (ng/s)

i.v. 4241T402 n.d. 100 n.d. n.d. n.d.

Chitosan (pH 5.5) 4455T576 P <0.01# 4.2T0.5 2.4T0.5 24.7T2.6 0.95T0.10

TMC60 (pH 7.4) 4086T944 P <0.01 3.9T0.9 3.9T1.0 18.4T5.2 0.71T0.20TMC20 (pH 7.4) 2678T960 P <0.01 2.5T0.9 2.5T0.8 13.7T5.5 0.53T0.21

Control (pH 5.5) 1860T194 P <0.01* 1.8T0.2 1 7.7T0.6 0.29T0.02

Control (pH 7.4) 1071T130 n.s. 1.0T0.1 1 4.3T1.3 0.17T0.05

Pharmacokinetic data enhancement ration of octreotide absorption after i.v. bolus and intratracheal administration of control solutions and polysaccharide

formulations (n =6; meansTSD). F is absolute bio-availability, ER is enhancement ratio, Css is steady state concentration, R0 is absorption rate in vivo. Significance

levels were calculated respective to the pH controls using two-factor ANOVAwith P <0.05 taken as significant; n.d. not determined; n.s. not significant; #compared

to control pH 5.5; *compared to control pH 7.4, all other statistical values are compared to control pH 7.4.

B.I. Florea et al. / Journal of Controlled Release 110 (2006) 353–361358

intratracheal dose of 3 mg/kg octreotide in either 0.9% saline of

pH 5.5 or 7.4 as control or in 1.5% (w/v) polysaccharide gel-

phase formulations. Octreotide plasma levels reached steady

state within 30 min, which were maintained throughout the 4-

h experiment. Apparently, octreotide absorption from the

airways follows zero-order kinetics with permeation across

the bronchial epithelium as rate limiting step.

A group of 6 animals received an i.v. bolus dose of 25 Agoctreotide. The pharmacokinetic parameters after i.v. and

intratracheal administrations were determined as described in

the Methods section and are summarised in Table 3. The i.v.

parameters shown in Table 4 matched previously published

values [9]. In the control situations, the AUC at pH 5.5 was

higher (P <0.05) than at pH 7.4 but the bio-availability (F)

remained low (1.8% and 1.0%, respectively). Chitosan induced

a higher bio-availability than TMC20 and comparable to

TMC60. However, when corrected for pH, the ER of chitosan

was significantly lower (P <0.05) than TMC60 and not

different from TMC20.

3.5. Chitosan and TMC toxicity in vivo

All animals survived the i.v. and intratracheal treatment

indicating that no acute toxic effects were induced by the

formulations within the time course of the experiment. After

the in vivo absorption studies, the lungs of the animals were

excised and prepared for histopathological examination by

light microscopy (Fig. 4). Rat lungs from the i.v. group

represent the normal healthy situation (Fig. 4a). Instillation of

saline of pH 5.5 shows a slight irritation in the bronchial region

(Fig. 4b) and no adverse effects were detected for physiological

saline (Fig. 4c). Formulations containing TMC20 induced a

certain inflammatory response because sites of neutrophil

infiltrations were detected around the bronchial regions (Fig.

Table 4

Intravenous parameters of octreotide in rats

Parameter AverageTSE

Body weight (g) 318T20Vd (ml) 65T8

Cl (ml/min kg) 2.3T0.1

t1/2 elim (min) 55T3

Octreotide was administered at a dose of 25 Ag/rat. Data are presented as

meansTSE of 6 rats. Vd volume of distribution, Cl clearance, t1/2 elim

elimination half-life.

4d). TMC60 showed only some mild irritation (Fig. 4e). By

contrast, chitosan caused a strong inflammatory response with

high levels of neutrophil infiltration and severe structural

damage of the pulmonary tissue (Fig. 4f). These histopatho-

logical results suggest that TMC20 and TMC60 are safer for

use as permeation enhancers in vivo in the lung than chitosan.

However, it cannot be excluded that the deleterious effects seen

with chitosan are a result of hypoxia caused by physical

obstruction of bronchioli because of the high viscosity of the

gel-phase formulation.

3.6. In vitro/in vivo correlation

The result of the correlation study between the in vitro

permeation and in vivo rates of octreotide absorption is

presented in Fig. 5. The apparent permeability (Papp) and rate

of absorption (R0) values are listed in Tables 2 and 3,

Fig. 4. Representative photographs for histo-pathological evaluation of ra

lungs after in vivo experiments with octreotide by i.v. administration (a)

intratracheal instillation of control solutions, 0.9% saline pH 5.5 (b) or 7.4 (c)

1.5% TMC20, pH 7.4 (d); 1.5% TMC60, pH 7.4 (e) or chitosan 1.5%, pH 5.5

(f). Arrows indicate regions with increased neutrophil accumulation or areas

that sustained structural damage to the lung tissue.

t

;

;

Fig. 5. Correlation plot of octreotide apparent permeability ( Papp) in vitro in

Calu-3 cells and rate of absorption (R0) in vivo after intratracheal instillation in

rats, with and without co-administration of polysaccharide formulations. Papp

and R0 values are listed in Tables 1 and 2.

B.I. Florea et al. / Journal of Controlled Release 110 (2006) 353–361 359

respectively. The R0 value for every formulation was deter-

mined by assuming an ocreotide systemic clearance of 2.3T0.1(ml/min kg) and considering the octreotide plasma levels after

instillation analogue to infusion at constant rate. The plot of

Papp vs. R0 showed a strong in vitro/in vivo correlation

between values of the means and linear regression gave the

equation: y =0.25x +0.21 with R2=0.93 (Fig. 5). The strong

linearity between the Calu-3 cell model and intratracheal

administration indicates a good correlation in paracellular

absorption of octreotide in vitro and in vivo in the presence

of polysaccharides as permeation enhancers.

4. Discussion

Chitosan (Fig. 1a) is a bioadhesive, cationic polysaccharide

that has found broad applications in biomedical research [3,11].

Lately, the synthesis of N-trimethylated chitosan (TMC)

derivatives has overcome the solubility problem of chitosan

at physiological pH (see Ref. [12]). In vitro and in vivo studies

have shown that chitosan [11] and TMC [13,17] are able to

transiently modulate the paracellular permeability of intestinal

and nasal epithelia enhancing the absorption of small and

macromolecular compounds. Our data (Fig. 2a) show that the

cationic polysaccharides, applied as gel-phase formulation, are

also active in Calu-3 cells of bronchial epithelium origin. The

polysaccharides induced a strong initial decrease in TEER of

the Calu-3 monolayers, as seen previously in Caco-2 intestinal

cell models [18]. The fact that paracellular permeability

modulation has no deleterious effects on the cells is illustrated

by the recovery studies where the TEER of all monolayers

returned to 90% to 100% of the initial values 16 h after the

experiment. These data are in agreement with data from Caco-2

cells [19]. The strong decrease in TEER induced by TMC60

suggests that pH, solubility and cationic charge density are

important factors for the modulation of the paracellular barrier.

From our previous oral delivery studies in rats [9,13] we

expected that octreotide would permeate only passively across

the bronchial epithelial barrier via the paracellular route. The

control formulations in Fig. 2b and Table 1 indeed show that

octreotide permeates poorly across the tight Calu-3 cell

monolayers. Functional evidence that modulation of intercel-

lular contacts can facilitate the paracellular permeation of

octreotide is shown in Fig. 2b. The apical to basolateral

permeation of octreotide was enhanced and followed zero-

order kinetic profiles, indicating that the polysaccharides can

functionally modulate the structural integrity of tight junctions

[20,21]. The same zero-order kinetic profiles were found

during studies with 14C-mannitol (data not shown), which is

not cell permeable, so we have strong indications that the

polysaccharides facilitate passive paracellular permeation of

octreotide through the widened tight junctions. TMC60 showed

the highest enhancement ratio, which might be a direct effect of

the strong decrease in TEER induced by the polymer.

Loss of cell viability might also account for the decrease in

TEER, enhanced permeation of octreotide and should be

thoroughly studied. For this reason we have performed a MTT

viability study in Calu-3 cells presented in Table 1. Chitosan

decreased the cell viability, while TMC20 and TMC60 seemed

to slightly increase the cell viability or accelerate the rate of

MTT turnover. We can only speculate that this increase in MTT

turnover might be caused by intensified metabolic activity in

the Calu-3 cells, possibly activation of redundant systems,

trying to compensate for the loss of intercellular adhesion due

to the interaction with the cationic polysaccharides. Toxicity

studies like ciliary beat frequency (CBF) in excised nasal

tissues from guinea pigs [22] and embryonic chicken trachea

[23] showed that chitosan and TMC slightly decrease the CBF

by 10%. Taken together, these in vitro data suggest that

TMC20 and TMC60 can be considered as non-toxic perme-

ation enhancers in bronchial epithelium cells, whereas chitosan

at pH 5.5 might induce mild cytotoxic effects.

In vivo studies showed that, in the case of controls, the

systemic bio-availability of octreotide after intratracheal

instillation is low and slightly pH dependent (Fig. 3). This

was expected because somatostatin shows low absorption from

the airways (F 1.0T0.1%; Ref. [7]) and octreotide permeates

poorly across the intestinal epithelium (F 0.3%; Ref. [6]). The

steady state octreotide plasma concentrations suggest the

prospective for sustained delivery from the conductive airways,

however enhancement of uptake by polysaccharides is neces-

sary for achieving therapeutically relevant plasma concentra-

tions. In accordance with the in vitro transport studies, chitosan

displayed the highest AUC and F compared to TMC20 and

TMC60, but the ER was similar to TMC20 when pH effects

were taken under consideration. These data suggest that a

lower pH of the formulation enhances octreotide permeation,

however it might also diminish the cell viability (Table 1).

Chitosan, TMC20 and TMC60 were able to significantly

enhance the octreotide permeation by 2.4-, 2.5- and 3.9-fold,

respectively (Table 2). The difference in permeation enhance-

ment between TMC20 and TMC60 might lay in the stronger

cationic charge density of TMC60 that makes it better soluble

at neutral pH and apparently more effective in modulating the

tight-junctional barrier. Although bio-availabilities of around

4% of the administered dose are low, the results of this proof of

principle study are promising.

All animals survived the treatment during the in vivo

studies. However, histological evaluation revealed that chitosan

triggered unexpectedly severe neutrophil infiltration and

B.I. Florea et al. / Journal of Controlled Release 110 (2006) 353–361360

structural damage in the lung parenchyma, compared with the

other formulations. It is unlikely that this effect is caused by

systemic absorption of the polysaccharides; hence their high

molecular weight (100–200 kDa) and hydrophilicity make

them virtually impermeable across the bronchial mucosa. It is

also unlikely that pH caused this effect since the control at pH

5.5 showed only minor irritation. However, we cannot exclude

that the more viscous chitosan formulation physically

obstructed the bronchioles and caused local asphyxiation of

the lung parenchyma tissues that might explain the severity of

the adverse effects compared to the TMCs. Chitosan also

induced a decrease in Calu-3 cell viability in vitro that might

have been caused by a combination of high formulation

viscosity and strong bioadhesion to the cells leading to

detachment and loss of cells during the washing procedure.

Some studies reported evidence of toxicity such as chitosan

adhered to Caco-2 cells causing plasma-membrane perturba-

tions [21] and chitosan powder formulations induced produc-

tion of immune stimulatory IL-6 and IL-8 cytokines in Calu-3

cells [24]. Furthermore, chitosan nanoparticles elicited a

stronger mucosal adjuvant effect [25] compared to TMC

microparticles [26]. It can be speculated that chitosan particles

of several nanometers in size can be endocytosed or penetrate

the epithelium paracellularly and trigger the immune system.

The mechanism of permeation enhancement induced by the

cationic polysaccharides is still mostly unknown. At cellular

level, immunohistochemical staining and cell fractionation

studies [20,21] showed that upon treatment of Caco-2 cell

monolayers with chitosan, the occludin and ZO-1 proteins

translocated from the membrane fraction to the cytoplasmatic

pool. This event seemed to be accompanied by rearrangements

of polymeric F-actin in the cytoplasm. Translocation of

occludin, ZO-1 and ZO-2 was also observed during neutrophil

transmigration across endothelial cells [27] suggesting that this

process might play a role in a relevant physiological process.

However, electron microscopy [21,27] showed that this protein

translocation did not alter TJ morphology indicating that the TJ

integrity is not lost but rather delicately modulated by the

neutrophils as well as by chitosan. Another intriguing question

is whether chitosan induces the same mechanism of permeation

both in vitro and in vivo. To assess this paradigm we plotted

the in vitro apparent permeabilities vs. the in vivo intratracheal

rates of absorption [28], and observed a good linear correlation

(R2=0.93) between these values for octreotide. This good

correlation suggests that polysaccharides may sustain similar

mechanism of permeation enhancement in vitro as in vivo,

based on transient modulation of tight-junctional complexes.

5. Conclusion

In conclusion, this study emphasizes the potential of Calu-3

cell monolayers to predict the (passive) pulmonary permeation

of peptide drugs in vivo. We demonstrated the potential of

cationic polysaccharide derivatives (TMC) to safely enhance

peptide drug permeation across the bronchial barrier in vitro

and in vivo and suggest a prospect for sustained pulmonary

drug delivery that can increase therapeutic benefit and lower

the costs. The pH, cationic charge density and viscosity of the

formulations proved to be key factors for optimizing formula-

tions for pulmonary delivery. This implies that hydrophilic,

polysaccharide scaffolds bearing chemically inert cationic

charge densities present a novel paradigm for design of safe

permeation enhancers. Further research should be conducted to

improve the bio-availability by formulation strategies, investi-

gate the cellular dynamics of TJ components and to identify the

molecular targets that lead to opening of the paracellular route

upon interaction with polysaccharides.

Acknowledgements

The authors would like to thank C.A. Jansma, E. Vasbinder,

P. Roemele and the Department of Pathology of the Leiden

University Medical Center for technical assistance, Dr. P.

Marbach for the kind donation of materials, H. Maas for data

modelling, M. Jansen for the critical comments, Dr. Saal van

Zwanenbergstichting and The Dutch Foundation for Pharma-

cological Sciences for financial support.

References

[1] R.F. Service, Drug delivery takes a deep breath, Science 277 (1997)

1199–1200.

[2] J.S. Patton, Breathing life into protein drugs, Nat. Biotechnol. 16 (1998)

141–143.

[3] S.S. Davis, Delivery of peptide and non-peptide drugs through the

respiratory tract, Pharm. Sci. Technol. Today (PSTT) 2 (1999) 450–456.

[4] P. Marbach, M. Neufeld, J. Pless, Clinical applications of somatostatin

analogs, Adv. Exp. Med. Biol. 188 (1985) 339–353.

[5] P. Marbach, W. Bauer, D. Bodmer, U. Briner, C. Bruns, A. Kay, I.

Lancranjan, J. Pless, F. Raulf, R. Robison, J. Sharkey, T. Soranno, B.

Stolz, P. Vit, G. Weckbecker, Discovery and development of somato-

statin agonists, in: R.T. Borchardt (Ed.), Integration of Pharmaceutical

Discovery and Development: Case Studies, Plenum Press, New York,

1998, pp. 183–209.

[6] E. Kohler, M. Duberow-Drewe, J. Drewe, G. Ribes, M.M. Loubatieres-

Mariani, N. Mazer, K. Gyr, C. Beglinger, Absorption of an aqueous

solution of a new synthetic somatostatin analogue administered to man by

gavage, Eur. J. Clin. Pharmacol. 33 (1987) 167–171.

[7] J.S. Patton, Pulmonary delivery of drugs for bone disorders, Adv. Drug

Deliv. Rev. 42 (2000) 239–248.

[8] M. Thanou, J.C. Verhoef, J.H. Verheijden, H.E. Junginger, Intestinal

absorption of octreotide using trimethyl chitosan chloride: studies in pigs,

Pharm. Res. 18 (2001) 823–828.

[9] M. Thanou, J.C. Verhoef, P. Marbach, H.E. Junginger, Intestinal

absorption of octreotide: N-trimethyl chitosan chloride (TMC) ameliorates

the permeability and absorption properties of the somatostatin analogue in

vitro and in vivo, J. Pharm. Sci. 89 (2000) 951–957.

[10] K.M. Varum, M.W. Anthonsen, H. Grasdalen, O. Smidsrod, 13C-n.m.r.

studies of the acetylation sequences in partially N-deacetylated chitins

(chitosans), Carbohydr. Res. 217 (1991) 19–27.

[11] L. Illum, Chitosan and its use as a pharmaceutical excipient, Pharm. Res.

15 (1998) 1326–1331.

[12] A.B. Sieval, M. Thanou, A.F. Kotze, J.E. Verhoef, J. Brussee, H.E.

Junginger, Preparation and NMR characterization of highly substituted N-

trimethyl chitosan chloride, Carbohydr. Polym. 36 (1998) 157–165.

[13] M. Thanou, J.C. Verhoef, H.E. Junginger, Oral drug absorption enhance-

ment by chitosan and its derivatives, Adv. Drug Deliv. Rev. 52 (2001)

117–126.

[14] B. Forbes, Human airway epithelial cell lines for in vitro drug transport

and metabolism studies, Pharm. Sci. Technol. Today (PSTT) 3 (2000)

18–27.

B.I. Florea et al. / Journal of Controlled Release 110 (2006) 353–361 361

[15] B.I. Florea, M.L. Cassara, H.E. Junginger, G. Borchard, Drug transport

and metabolism characteristics of the human airway epithelial cell line

Calu-3, J. Control. Release 87 (2003) 131–138.

[16] M. Fluttert, S. Dalm, M.S. Oitzl, A refined method for sequential blood

sampling by tail incision in rats, Lab. Anim. 34 (2000) 372–378.

[17] A.F. Kotze, M.M. Thanou, H.L. Luessen, A.G. de Boer, J.C. Verhoef,

H.E. Junginger, Enhancement of paracellular drug transport with highly

quaternized N-trimethyl chitosan chloride in neutral environments: in vitro

evaluation in intestinal epithelial cells (Caco-2), J. Pharm. Sci. 88 (1999)

253–257.

[18] P. Artursson, T. Lindmark, S.S. Davis, L. Illum, Effect of chitosan on the

permeability of monolayers of intestinal epithelial cells (Caco-2), Pharm.

Res. 11 (1994) 1358–1361.

[19] M. Thanou, B.I. Florea, M.W. Langemeyer, J.C. Verhoef, H.E. Junginger,

N-trimethylated chitosan chloride (TMC) improves the intestinal perme-

ation of the peptide drug buserelin in vitro (Caco-2 cells) and in vivo

(rats), Pharm. Res. 17 (2000) 27–31.

[20] J. Smith, E. Wood, M. Dornish, Effect of chitosan on epithelial cell tight

junctions, Pharm. Res. 21 (2004) 43–49.

[21] V. Dodane, K.M. Amin, J.R. Merwin, Effect of chitosan on epithelial

permeability and structure, Int. J. Pharm. 182 (1999) 21–32.

[22] T.J. Aspden, L. Illum, Æ. Skaugrud, The effect of chronic nasal

application of chitosan solutions on cilia beat frequency in guinea pigs,

Int. J. Pharm. 153 (1997) 137–146.

[23] M.M. Thanou, J.C. Verhoef, S.G. Romeijn, J.F. Nagelkerke, F.W. Merkus,

H.E. Junginger, Effects of N-trimethyl chitosan chloride, a novel

absorption enhancer, on Caco-2 intestinal epithelia and the ciliary beat

frequency of chicken embryo trachea, Int. J. Pharm. 185 (1999) 73–82.

[24] C. Witschi, R.J. Mrsny, In vitro evaluation of microparticles and polymer

gels for use as nasal platforms for protein delivery, Pharm. Res. 16 (1999)

382–390.

[25] K. Roy, H.Q. Mao, S.K. Huang, K.W. Leong, Oral gene delivery with

chitosan-DNA nanoparticles generates immunologic protection in a

murine model of peanut allergy, Nat. Med. 5 (1999) 387–391.

[26] I. van der Lubben, J.C. Verhoef, G. Borchard, H.E. Junginger, Chitosan

and its derivatives in mucosal drug and vaccine delivery, Eur. J. Pharm.

Sci. 14 (2001) 201–207.

[27] A.R. Burns, R.A. Bowden, S.D. MacDonell, D.C. Walker, T.O.

Odebunmi, E.M. Donnachie, S.I. Simon, M.L. Entman, C.W. Smith,

Analysis of tight junctions during neutrophil transendothelial migration, J.

Cell Sci. 113 (2000) 45–57.

[28] N.R. Mathias, J. Timoszyk, P.I. Stetsko, J.R. Megill, R.L. Smith, D.A.

Wall, Permeability characteristics of Calu-3 human bronchial epithelial

cells: in vitro– in vivo correlation to predict lung absorption in rats, J.

Drug Target. 10 (2002) 31–40.