In vitro and ex vivo intestinal tissue models to measure mucoadhesion of poly (methacrylate) and N-trimethylated chitosan polymers

Article abstract of DOI:10.1007/s11095-004-9007-1

Purpose. The adhesion of a range of polymers based on poly(2- (dimethylamino-ethyl) methacrylate (pDMAEMA) was assessed using human mucus-secreting and non mucus-secreting intestinal cell monolayers, HT29-MTX-E12 (E12) and HT29 monolayers, as well as excised non-everted intestinal sacs from rats. Differentiation of mucoadhesion from bioadhesion was achieved by pre-treatment with the mucolytic agent, N-acetyl cysteine (NAC). Adherence of pDMAEMA polymers was compared to that obtained with the mucoadhesive, N-trimethylated chitosan (TMC). Methods. The quantity of adherent coumarin 343-conjugated polymers to HT29, E12, and intestinal sacs was measured by fluorescence. Confocal laser scanning microscopy (CLSM), light microscopy, and fluorescent microscopy were used to provide direct evidence. Measurements of transepithelial electrical resistance (TEER), permeability to FITC-dextran 4000 (FD-4), and the release of lactate dehydrogenase (LDH) were used to assess potential cytotoxicity of polymers. Results. Adherence of unquaternized and of 10%, 24%, and 32% methyl iodide-quaternized pDMAEMA polymers was measured in E12, HT29, and sacs. All pDMAEMA polymers showed significantly higher levels of adhesion to mucus (mucoadhesion) than to epithelium (bioadhesion). Colocalization of pDMAEMA with mucus was confirmed in E12 by microscopy. TMC showed equally high levels of mucoadhesion as unquaternized and 24% quaternized pDMAEMA, but displayed higher levels of bioadhesion. pDMAEMA-based polymers demonstrated lower levels of adherence to E12 and rat sacs in the presence of NAC, whereas adherence of TMC was unchanged. pDMAEMA significantly decreased the permeability of FD-4 across E12 monolayers and sacs and was less cytotoxic in E12 than in HT29. In contrast, TMC increased the permeability of FD-4 across E12 and sacs and was less cytotoxic in E12 than in HT29. Conclusions. Human mucus-producing E12 monolayers can be used to assess polymer mucoadhesion and give similar data to isolated rat intestinal sacs. pDMAEMA displayed similar levels of mucoadhesion and lower levels of bioadhesion than a chitosan derivative and it was not cytotoxic. pDMAEMA decreased FD-4 flux in the presence of mucus, whereas TMC increased it. The combination of mucus and methacrylate polymers appears to increase barrier function of the apical membrane.

Full text of DOI:10.1007/s11095-004-9007-1

Pharmaceutical Research, Vol. 22, No. 1, January 2005 (© 2005)
DOI: 10.1007/s11095-004-9007-1
Research Paper
In Vitro and ex Vivo Intestinal Tissue Models to Measure Mucoadhesion of
Poly (Methacrylate) and N-Trimethylated Chitosan Polymers
Simon Keely,1,2 Atvinder Rullay,3 Carolyn Wilson,1 Adrian Carmichael,3,5 Steve Carrington,1
Anthony Corfield,4 David M. Haddleton,3 and David J. Brayden1,2,6
Received July 20, 2004; accepted September 16, 2004
Purpose. The adhesion of a range of polymers based on poly(2-(dimethylamino-ethyl) methacrylate
(pDMAEMA) was assessed using human mucus-secreting and non mucus-secreting intestinal cell mono-
layers, HT29-MTX-E12 (E12) and HT29 monolayers, as well as excised non-everted intestinal sacs from
rats. Differentiation of mucoadhesion from bioadhesion was achieved by pre-treatment with the mu-
colytic agent, N-acetyl cysteine (NAC). Adherence of pDMAEMA polymers was compared to that
obtained with the mucoadhesive, N-trimethylated chitosan (TMC).
Methods. The quantity of adherent coumarin 343-conjugated polymers to HT29, E12, and intestinal sacs
was measured by fluorescence. Confocal laser scanning microscopy (CLSM), light microscopy, and
fluorescent microscopy were used to provide direct evidence. Measurements of transepithelial electrical
resistance (TEER), permeability to FITC-dextran 4000 (FD-4), and the release of lactate dehydrogenase
(LDH) were used to assess potential cytotoxicity of polymers.
Results. Adherence of unquaternized and of 10%, 24%, and 32% methyl iodide-quaternized
pDMAEMA polymers was measured in E12, HT29, and sacs. All pDMAEMA polymers showed sig-
nificantly higher levels of adhesion to mucus (mucoadhesion) than to epithelium (bioadhesion). Co-
localization of pDMAEMA with mucus was confirmed in E12 by microscopy. TMC showed equally high
levels of mucoadhesion as unquaternized and 24% quaternized pDMAEMA, but displayed higher levels
of bioadhesion. pDMAEMA-based polymers demonstrated lower levels of adherence to E12 and rat
sacs in the presence of NAC, whereas adherence of TMC was unchanged. pDMAEMA significantly
decreased the permeability of FD-4 across E12 monolayers and sacs and was less cytotoxic in E12 than
in HT29. In contrast, TMC increased the permeability of FD-4 across E12 and sacs and was less cytotoxic
in E12 than in HT29.
Conclusions. Human mucus–producing E12 monolayers can be used to assess polymer mucoadhesion
and give similar data to isolated rat intestinal sacs. pDMAEMA displayed similar levels of mucoadhe-
sion and lower levels of bioadhesion than a chitosan derivative and it was not cytotoxic. pDMAEMA
decreased FD-4 flux in the presence of mucus, whereas TMC increased it. The combination of mucus
and methacrylate polymers appears to increase barrier function of the apical membrane.
KEY WORDS: chitosan; HT29 monolayers; living radical polymerization; methacrylate polymers;
mucoadhesion.
INTRODUCTION
1 Faculty of Veterinary Medicine, University College, Dublin, Ire-
land.
The use of polymers as bioadhesives (adhering to epithe-
lium) and mucoadhesives (adhering to mucus) offer signifi-
cant potential for oral drug delivery. Cargos formulated with
mucoadhesive polymers may increase gastrointestinal tract
residence time leading to improved oral drug bioavailability,
as the formulation should allow more opportunity to contact
the epithelium (1,2). Adhesive polymeric systems may also be
useful for topically coating the damaged intestinal wall in
inflammatory bowel disease (3), or for facilitating healing in
the oral cavity (4). For these applications, polymers need to
be inert, nonabsorbable, stable, and easy to process. A re-
quirement for any adhesive polymer is quantifiable and re-
producible adhesion to tissues and/or supramucosal gels.
Changes in indices of intestinal barrier structure and function
2 Conway Institute of Biomedical and Biomolecular Research, Uni-
versity College, Dublin, Ireland.
3 Department of Chemistry, University of Warwick, Coventry, UK.
4 Mucin Research Group, Clinical Science at South Bristol, Royal
Infirmary, Bristol, UK.
5 Current address: Warwick Effect Polymers, Vanguard Centre, Sci-
ence Park, Coventry, UK.
6 To whom correspondence should be addressed. (e-mail: david.
brayden@ucd.ie)
ABBREVIATIONS: CLSM, confocal laser scanning microscopy;
FD-4, fluoresceinisothiocyanate-dextran (MW 4 kDa); Mn, average
molecular weight number; Papp, apparent permeability coefficient;
PDi, polydispersity index (molecular weight distribution);
pDMAEMA, poly(2-(dimethylamino-ethyl) methacrylate; TEER,
transepithelial electrical resistance; TMC, N-trimethylated chitosan.
0724-8741/05/0100-0038/0 © 2005 Springer Science+Business Media, Inc.
38

Models to Measure Mucoadhesion of Poly (Methacrylate) and N-Trimethylated Chitosan Polymers
39
may be caused by adhesion of some polymers. Examples in- to tissue in physiologic buffer. The surface charge on
clude the opening of tight junctions by chitosan (5) and re- pDMAEMA can be readily altered by varying the quater-
duction of bacterial access to the epithelial surface by poly- nization of the ammonium groups, a process implicated in
ethylene glycol (6).
antimicrobial activity of the polymer in vitro (25). In addition,
There are few reliable intestinal models that adequately pDMAEMA polymers possess functional side groups in their
provide a mucus gel lining in the gastrointestinal tract to dimethylamino ethyl moieties, and these may be suitable can-
which adherence can be measured. Most are based on flow- didates for glycoprotein interaction with sialic acid residues
through systems using isolated mucus-covered mucosae and on mucins (26). pDMAEMA can be synthesized in high yield
examples include rat stomach mucosa (7), porcine tissue mu- and purity by the novel and simple process of living radical
cosa (8) and perfused rat intestinal loops (9). Furthermore, polymerization (27) and is therefore an appropriate polymer
attempts to rank polymer adhesion to these preparations type to assess in adhesion studies. For comparative purposes,
maintained under sub-optimal physiologic conditions rely we used N-trimethylated chitosan (TMC) as a known positive
largely on qualitative tensiometer measurements of the force control for adhesion as it is also soluble at pH 7.4 (11).
required to separate formulations coated onto solid matrices
Our aims were to compare mucus-covered E12 monolay-
(10,11). Recently, there has been an attempt to measure par- ers and non-everted rat intestinal sacs as tissue models to
ticulate adhesion to isolated thawed pig oesophagi through quantitatively assess adhesion of fluorescently labeled
use of a dynamic test system with image analysis (12).
pDMAEMA analogues. In addition, light, fluorescent, and
Caco-2 and HT29 monolayers are used as in vitro cell confocal laser scanning microscopy were used to visualize
culture models of the human intestinal epithelium. Caco-2 polymer binding. Cytotoxic potential and effects on barrier
cells spontaneously differentiate into monolayers of polarized function of pDMAEMA and TMC were also assessed.
enterocytes, connected by tight junctions, but they lack mu-
cus-secreting goblet cells (13). In contrast, HT29 monolayers
expresses mucin-secreting mature goblet cells in the presence
of methotrexate (MTX) selection pressure (14). There are
issues for drug delivery studies arising from the variable levels
of mucus production and inter-passage inconsistency in early
sub-clones (15). Attempts have also been made to create mu-
cus-producing monolayers by co-culturing Caco-2 in specific
ratios with the HT29 sub-clones, HT29GlucH (16), and
HT29-MTX (17). These mixed co-cultures are, however, dif-
ficult to maintain as growth rates of the mixed cell lines vary.
This leads to the production of a mucus gel layer that may not
be stable and reproducible through successive passages.
Stable sub-clones of HT29-MTX have, however, recently
been isolated and partially characterized (18). Monolayers of
one such subclone, HT29-MTX-E12 (E12), elaborate a 150
␮m mucous gel layer that corresponds to in vivo measure-
ments of human small intestine. The overlying mucus acts as
a significant barrier to the permeation of lipophilic drugs, and
also to the uptake of hydrophobic nanoparticles (19). In ad-
dition, it expresses the mucins MUC1 and MUC2, which are
found in the small intestine and are implicated in host-
pathogen relationships (18). Taken together, E12 could be an
appropriate in vitro model for assessing polymer mucoadhe-
sion.
In studies of mucoadhesive polymers, traditional rat
everted intestinal sacs are un-economic due to the large
amount of polymer required in the bath. While morphologi-
cally intact immediately after inversion, the epithelial surface
quickly degrades (20). One aim of the current study was to
assess the use of non-everted sacs for measuring polymer ad-
herence. We used a rapid dissection method followed by the
maintenance of sac viability in a high-nutrient medium (21).
In this study E12 and rat intestinal non-everted sacs were
compared in the presence and absence of a mucolytic agent,
N-acetyl cysteine (NAC) (16,22), as it allows differentiation
between mucoadhesion and bioadhesion.
MATERIALS AND METHODS
Materials
All tissue culture reagents were from Gibco (Biosci-
ences, Dublin, Ireland). FITC-Dextran (MW 4400) was from
Sigma (Sigma-Aldrich, Dublin, Ireland). Tissue culture ma-
terials and plates were from Corning Costar (Fannin Health-
care, Dublin, Ireland) and chitosan was obtained from Fluka
(Fluka Chemicals, Dorset, UK). Ethidium homodimer-1 was
obtained from Molecular Probes (Molecular Probes Europe
BV, Leiden, The Netherlands). The Titramax 1000 shaking
incubator was from Heidolph Instruments (Germany).
Cell Culture
E12 Cells (passage 50-57) were a generous gift from Pro-
fessor Per Artursson, Uppsala University, Sweden. HT29
cells were obtained from ATCC (passage 121-128). Cells were
grown and subcultured as previously described (18). Both
lines were maintained with Dulbecco’s Modified Eagles Me-
dium (DMEM) with 10% fetal calf serum (FCS), 1% non-
essential amino acids and 1% L-glutamine at 5% CO2, 95%
O2 at 37°C.
For adhesion and transport studies, HT29 and E12 were
subcultured at a density of 2 × 104 cells/filter on 12 mm
Transwell polycarbonate membrane inserts (Corning Costar,
cat. no. 3401). Monolayers were fed on both sides every two
days and differentiated over 21 days. Barrier formation of
monolayers was examined by measuring the transepithelial
electrical resistance (TEER) of cell monolayers before and
after adhesion experiments using an EndOhm electrode sys-
tem with background correction made for unseeded filters.
Monolayers with TEER values below 50 ⍀ cm2 were ex-
cluded from experiments.
Lactate Dehydrogenase Cytotoxicity Assay
The hydrophilic cationic methacrylate-based polymer,
poly(2-(dimethylamino-ethyl) methacrylate (pDMAEMA),
Lactate dehydrogenase (LDH) activity was measured in
appears to have potential as a delivery vehicle for gene de- culture supernatants at 0, 30, and 60 min following polymer
livery (23) and tumor targeting (24). The polymer is water exposure to cells grown on Transwell membranes, using a
soluble at neutral pH and is therefore suitable for adhesion cytotoxicity assay kit (Roche Diagnostics, East Sussex, UK).

40
Keely et al.
LDH concentrations were expressed as percentage LDH re-
lease relative to treatment with the detergent Triton-X 100
and the percentage cytotoxicity calculated (28).
Polymer Synthesis
Preparation of Coumarin Initiator
Hydroquinone (110.11 g, 1.0 mol), tetrahydrofuran (800
ml), and triethylamine (15.3 ml, 0.11 M) were placed in a 1 L
round-bottom flask equipped with a pressure equilibrating
dropping funnel. 2-Bromoisobutyryl bromide (12.4 ml, 0.1 M)
and tetrahydrofuran (87.6 ml) were then placed into the drop-
ping funnel. The 2-bromoisobutyryl bromide solution was
added dropwise over 2 h with stirring which formed triethyl-
ammonium bromide. Upon complete addition, the reaction
was stirred for 2 h. The triethylammonium bromide was re-
moved by filtration, and the solvent was removed under re-
duced pressure. This resulted in a brown white crystalline
mixture that was placed into a round bottom flask with chlo-
roform (600 ml) and stirred overnight to separate the un-
reacted excess hydroquinone. The unreacted hydroquinone
was removed from the solution by filtration and the filtrate
was concentrated under reduced pressure to give a brown
crystalline compound. The resulting brown crystalline solid
was purified by column chromatography eluting with 100%
dichloromethane followed by 100% methanol to give a pale
brown crystalline solid, hydroxyl initiator (Fig. 1A), yield ס
18.5 g (71%).
Coumarin 343 (1 g, 3.5 mM, Exciton Inc., Ohio, USA),
hydroxyl initiator (0.91g, 3.5 mM), N-(3-dimethylaminopro-
pyl)-N’-ethylcarbodiimide hydrochloride (1.3g, 7.0 mM), and
4-(dimethylamino)pyridine (0.043 g, 0.35 mM) were then dis-
solved in anhydrous dichloromethane (25 ml) under nitrogen
and stirred at room temperature for 72 h. The resultant so-
lution was washed with distilled water (2 × 25 ml), 0.5 M HCl
(4 × 25 ml), saturated NaHCO3 (4 × 25 ml), distilled water
(2 × 25 ml), and then dried over MgSO4. Removal of the
solvent in vacuo gave an orange product that was purified by
column chromatography on silica gel using 9:1 dichlorometh-
ane/ethyl acetate, followed by 9:1 dichloromethane/diethyl
ether to give the coumarin 343 initiator (Fig. 1B) in 80%
yield. This method was adapted from Ref. 29.
Preparation of Coumarin 343-Initiated
pDMAEMA Polymers
The methacrylate-based polymers were quaternized de-
rivatives of 2-(dimethylamino- ethyl) methacrylate
(DMAEMA) monomer, which were labeled with the fluores-
cent probe, coumarin 343. Coumarin fluoresces with an exci-
tation wavelength of 400 nm and an emission wavelength of
490 nm. Coumarin 343 initiator 2 (0.50 g, 0.95 mM), Cu(I)Br
(0.136 g, 0.95 mM, 1 eq), DMAEMA (14.2 g, 0.090 M), tolu-
ene (30 ml), and a magnetic follower were placed in an oven-
dried Schlenk tube. The resulting solution was deoxygenated
via three freeze-pump-thaw cycles and, following de-gassing,
N-propyl-2-pyridylmethanimine (0.28 g, 1.9 mM) was added.
The reaction was placed in a thermostatically controlled oil
bath at 90°C for 4 h. The polymer solution was diluted with
toluene (100 ml) and filtered through a column of basic alu-
mina to remove the catalyst. The column was washed with
Fig. 1. Structures of compounds. (A) Hydroxyl initiator; (B) couma-
rin 343 initiator; (C) coumarin 343-initiated pDMAEMA; (D) quat-
ernized coumarin 343-initiated pDMAEMA; where Q10x/(x + y) ס
0.1; (E) structure of TMC; (F) coumarin 343-labeled TMC.
to rotary evaporation to remove the solvent. The polymer was
purified twice by precipitation using dichoromethane/
petroleum ether to give coumarin 343-initiated pDMAEMA
(Mn 12,300, PDi of 1.08, Fig. 1C).
Quaternization of Coumarin 343- Initiated pDMAEMA
Coumarin 343-initiated pDMAEMA was quaternized to
toluene (2 × 100 ml) and the combined filtrate was subjected yield 10% (Q10), 24% (Q24), and 32% (Q32) of the repeat

Models to Measure Mucoadhesion of Poly (Methacrylate) and N-Trimethylated Chitosan Polymers
41
units. In a typical quaternization, coumarin 343 initiated Adhesion Assay: Monolayers
pDMAEMA (5g, 0.41 mM, Mn 12,300, PDi 1.08) was dis-
HT29 and E12 monolayers were rinsed with HEPES-
buffered-DMEM and equilibrated in the same medium at
37°C for 60 min. In the some instances, monolayers were
incubated with the mucolytic agent, N-acetyl cysteine (10mM,
NAC) for 15 min prior to polymer addition. To remove NAC,
monolayers were rinsed gently with medium before use. Test
polymer (0.5 ml) was added to the apical side of the mono-
layer at two concentrations (0.1 mg/ml or 0.01 mg/ml). Mono-
layers were incubated at 37°C in a Titramax 1000 shaking
incubator for 60 min at 100 rpm. Polymer concentrations were
measured in 50 ␮l samples taken both sides of the monolayers
at 0 and 60 min. Following aspiration of the polymer-loaded
donor side, the monolayers were then washed three times in
incubation buffer, homogenized, and lysed in 2% SDS and
EDTA (50 mM) at pH 8.0. All samples were assayed on an LS
50B Luminescence Fluorimeter as described above and data
expressed as ␮g polymer/cm2. Negligible amounts of signal
were detected in the washes.
solved in tetrahydrofuran (50 ml) and methyl iodide (0.43g, 3
mM) was added and the solution left stirring for 24 h at am-
bient temperature. The solvent was removed and the polymer
dried under vacuum. The degree of quaternization, as deter-
1
mined by H NMR, was 10% of the available amine groups.
This gives a polymer with a molecular mass of 13,400 (Fig. 1D).
Trimethylated Chitosan Chloride
TMC was synthesized as previously described, (30). In
brief, 1-methyl-2-pyrolidinone (80 ml) was added to a flask
containing low-viscosity heat-dissolved chitosan (2 g) and so-
dium iodide (4.8 g). On addition of NaOH (aq.) (11 ml, 15%
w/v) and iodomethane (11 ml), the reaction mixture was con-
densed for 60 min. The mixture was removed from the oil
bath and ethanol (200 ml) was added to precipitate the chi-
tosan derivative. The suspension was centrifuged for 10 min
at 4000 × g, the supernatant decanted off and the resulting
solid was sequentially washed with ethanol and diethyl ether.
The product was isolated by centrifugation. The isolated solid
was added to 80 ml of 1-methyl-2-pyrolidinone and heated to
dissolve. NaI (4.8 g), NaOH (aq) (11 ml, 15% w/v) and io-
domethane (11 ml) were added and the mixture condensed
for 30 min. Additional iodomethane (2 ml) and NaOH pellets
(2 g) were added and stirring continued for 1 h. The product
was washed and isolated as described above. It was dissolved
in 10% (w/v) aqueous NaCl (80 ml) and stirred for 30 min.
The product was precipitated with ethanol and isolated by
centrifugation to yield N-trimethylated chitosan chloride
(TMC, Fig. 1E). The molecular weight of TMC is 85,000 rela-
tive to poly(ethylene) glycol narrow molecular weight stan-
dards, measured using 2 × 30 cm PL aquagel columns (Poly-
mer Laboratories) at 1 ml/min with differential refractive in-
dex detection.
Adhesion Assay: Tissue
Rats were starved overnight before euthanasia by cervi-
cal dislocation. The intestine was removed after a midline
incision, and the jejunum rapidly removed and flushed with
oxygenated medium. 6 sacs, each 5 cm long, were cut from the
isolated jejunum. Sacs were placed in oxygenated TC199 me-
dium at 37°C, according to the method of Barthe et al. (21).
The sacs were tied tightly at one end with silk suture and a
small animal vascular catheter (Data Sciences International
Physiocath 277-1-002) was tied in to the other end. A 1 ml
syringe with a sterile 26 gauge micro lance was fixed to the
catheter. In some instances, intestinal sacs were pretreated
with 10 mM NAC for 15 min, which was flushed out with 20
ml of medium. Sacs were then filled with 0.5 ml polymer
solution (1 mg/ml) via the catheter. Each sac was placed in a
separate sealed 50 ml flask containing 15 ml of oxygenated
TC-199 medium on a shaking water bath for 30 min at 37°C.
Duplicate 50 ␮l samples of incubation medium were removed
from the bath after 30 min to assess leakage. Sacs were then
removed from the bath and the internal contents recovered
using a fresh 1 ml volume syringe. Following aspiration of the
polymer-loaded donor compartment, sacs were then washed
sequentially four times with a total of 5 ml of medium and the
washes collected for assay. Samples were adjusted to pH 7.4
by addition of sodium citrate (10 mM) and assayed by fluo-
rescence as described above. Adhesion to sacs was calculated
by subtraction and expressed as ␮g polymer/cm2. Polymer
mass balance was present in all monolayer and sac experi-
ments (data not shown). None of the washes contained sig-
nificant concentrations of polymers.
Coumarin-Labeled Trimethylated Chitosan Chloride
A mixture of TMC chloride salt (200 mg) in water (20
ml) was stirred at ambient temperature for 25 min. Coumarin
343 (2 mg, 1% w/w) was added followed by dimethylamino-
pyridine (DMAP) (1 crystal). Ethyl carbodiimide hydrochlo-
ride (EDCI) (6.7 mg) was added and the mixture was stirred
at ambient temperature. After 26 h, the reaction mixture was
freeze-dried. The solid was dissolved in 25% ethanol in water
(40 ml), transferred to a dialysis bag (molecular weight cutoff
12–14,000) and dialyzed against 25% ethanol in water (4 L).
The dialysis was undertaken for 7 days, with increasing etha-
nol concentration to 60% (to remove un-reacted coumarin
343, and then back to 100% water. The contents of the dialy-
sis bag were freeze-dried to yield coumarin 343-labeled TMC
FD-4 Transport Studies
Transport studies were carried out on monolayers and
(Fig. 1F). Coumarin-labeled TMC has 0.17 molecules of cou- sacs with FITC-dextran 4.4 kDa (FD-4) in order to assess
marin per TMC chain (i.e., about 1 for every 6 TMC mol- epithelial functional integrity. FD-4 permeability across
ecules). This was derived from the UV absorbance of couma- monolayers was examined according to previous reports (31).
rin in the visible range calibrated against free coumarin in In brief, 0.5 ml of FD-4 (250 ␮g/ml) was added to the apical
aqueous solution.
side of the monolayers. FD-4 flux to the basolateral side was
For adhesion assays, polymers were dissolved in HEPES- measured over 120 min. Samples (0.1 ml) were collected from
buffered DMEM (monolayers) and TC-199 medium (sacs), the basolateral side every 15 min and subsequently replaced
respectively. There was no alteration of fluorescent signal in with fresh transport medium. FD-4 was assayed fluorimetri-
these media.
cally at excitation wavelength of 490 nm and an emission of

42
Keely et al.
wavelength of 525 nm. In studies using non-everted intestinal neutral red (1%) for 5 min at room temperature and washed
sacs, 0.5 ml of FD-4 (1.0 mg/ml) was injected into each sac in a water bath with flowing water. They were viewed by light
lumen as described above. Samples (50 ␮l) were collected and fluorescent microscopy using a Nikon Eclipse E400 Fluo-
from the bath every 15 min and replaced with fresh medium. rescent Microscope with Nikon Digital Camera (model
After 120 min, the sacs were cut open and the contents DXM1200). Images were captured using Nikon ACT-1 (ver-
sampled. The apparent permeability (Papp) for FD-4 was sion 2.0) imaging software, using alternate transmitted light
calculated from the following equation: Papp (cm/s) ס and fluorescent modes. A field comparison between light and
(dQ/dt)/(A · Co), where dQ/dt is the transport rate (mol/s), A fluorescent images allowed the visualization of fluorochrome-
is the surface area of the monolayer or sac (cm2), and Co is the labeled polymers in the mucus gel layer.
initial concentration in the donor compartment (mol/ml).
Statistics
Statistical analysis for significant differences between
groups was calculated using two-tailed student’s unpaired t
tests (GraphPad Instat, version 3.05) unless indicated. p val-
ues of <0.05 were considered significant. Values in Figs. 2–5
and Table 1 were expressed as mean SEM for each group.
Confocal Scanning Laser Microscopy
Polycarbonate membranes with attached monolayers
were exposed to 0.1% sodium dodecyl sulfate for 2 min at
room temperature to render cell membranes permeable.
Membranes were then removed from the insert scaffold and
stained with ethidium homodimer-1 (EthD1) (Molecular
Probes, Leiden, The Netherlands, cat. no. L-3224). The mem-
branes were placed on glass slides, surrounded by an adhesive
compound and covered in 0.2 ml of DMEM-HEPES medium.
Glass cover slips were attached, sealed and examined using
confocal scanning laser microscopy (CLSM; Leica TCS SL).
The excitation/emission wavelengths were 495 nm/635 nm,
respectively.
RESULTS
Adhesion Assays
Adhesion of pDMAEMA Derivatives to E12 Monolayers
pDMAEMA, quaternized derivatives (10%, 24%, and
32% levels) and TMC were assessed for their adhesion to E12
and HT29 (Fig. 2). Because E12 monolayers were covered
with a mucus layer and HT29 were not, we designate adher-
ence of polymers to these monolayers as “mucoadhesion” and
“bioadhesion,” respectively. The data showed that all four
Light and Fluorescent Microscopy
In order to view the mucus gel layer on frozen sections, pDMAEMA derivatives adhered significantly to E12 at both
E12 monolayers were washed in DMEM-HEPES medium, concentrations used (mucoadhesion) and, in each case, they
sandwiched between thin strips of chicken liver, and snap- displayed significantly increased mucoadhesion over that seen
frozen in liquid nitrogen. The tissue was embedded in optimal in HT29 at the matched concentration (bioadhesion). There
cutting temperature (OCT) compound, and 20 ␮m cross- was no evidence that the coumarin dissociated from the poly-
sections were cut with a cryo-microtome and transferred to mer since no fluorescent signal could be detected on the ba-
glass slides for staining. To visualize the mucus gel layer, sec- solateral side of any monolayer exposed to the fluorescent
tions were stained for 10 min at room temperature with alcian polymers. Furthermore, stability of pDMAEMA formula-
blue (1%) in distilled water, adjusted to pH 2.5 with glacial tions were assessed over three weeks in aquous buffer and no
acetic acid (3%). Sections were then washed in a water bath diminution of fluorescent signals were detected (results not
with flowing water. Sections were then counterstained with shown).
Fig. 2. pDMAEMA adhesion to E12 and HT29 monolayers at four levels of quaternization. TMC was
used as a positive control. ᮀ E12 (50 ␮g/ml); HT29 (50 ␮g/ml); E12 (5 ␮g/ml); ᭿ HT29 (5 ␮g/ml).
n ס 12 each group. **p < 0.01; * <0.05, comparing adhesion to E12 to HT29 at two concentrations.
Mean SEM.

Models to Measure Mucoadhesion of Poly (Methacrylate) and N-Trimethylated Chitosan Polymers
43
Levels of adhesion were significantly increased for Comparison of Polymer Adherence to E12 and
each polymer using a 10-fold higher concentration in each Intestinal Sacs
type of culture. In the cases of unquaternized and 24% quat-
The adhesion of three polymers (pDMAEMA, 24%
ernized pDMAEMA, they were significantly more mucoad- quaternized pDMAEMA and TMC) to monolayers and in-
hesive than TMC at the 5 ␮g/ml concentration, while all testinal sacs was compared in the presence and absence of the
pDMAEMA polymers displayed similar degrees of bioadhe- mucolytic, NAC. Significantly reduced levels of mucoadhe-
sion to TMC at this concentration, but relatively less bioad- sion of both pDMAEMA polymers were present in NAC-
hesion at the higher concentration. On the other hand, TMC treated E12 compared to untreated E12, values being close to
bound equally well to both E12 and HT29 at both concentra- those levels detected in HT29. TMC, in contrast displayed
tions, suggesting that the presence of a mucus layer had no similar high levels of adherence to all three monolayer groups
impact on its binding. There was no evidence that the degree (Fig. 3A). In sacs, the same pattern of adhesion was apparent
of quaternization of pDMAEMA influenced adhesion to ei- (Fig. 3B), namely, a significant reduction in adherence of both
ther E12 or HT29. Overall, this data suggested that the pres- pDMAEMA polymers in the presence of NAC. In contrast,
ence of a mucus layer contributed significantly to the adhe- TMC adhered equally well to sacs in the presence and ab-
sion of pDMAEMA and its analogues. These polymers retain sence of NAC. pDMAEMA was more adherent to sacs than
a significant capacity to bind non-mucus covered epithelia at TMC in the presence of mucus, but was less adherent in
levels within the range of TMC.
its absence (p < 0.05 in each case). There was no evidence
Fig. 3. (A) The effect of NAC on polymer adherence to monolayers. ᮀ E12; NAC-E12; ᭿ HT-29. 50
␮g/ml of each polymer was used. n ס 12 for each group. (B) Effect of NAC on polymer adherence to
sacs. ᮀ Sac; ᭿ NAC-Sac. 500 ␮g/ml of each polymer were used. In (A) and (B), NAC (10 mM) was
present for 15 min in advance of polymer before wash-out. (n ס 8 for each group). **p < 0.01, *<0.05.
Mean SEM.

44
Keely et al.
that quaternization had any positive effect on adherence to the higher concentration. In contrast, no statistical difference
either model. Overall, this data suggests that adhesion of in the Papp of FD-4 was seen in HT29 exposed to either
pDMAEMA to both E12 and intestinal sacs is indeed related concentration of the polymer (Fig. 4A), and therefore a physi-
in large part to the presence of a mucus gel layer and that this cal interaction between pDMAEMA and the paracellular
class of polymer is therefore predominantly mucoadhesive. marker appeared unlikely. E12 monolayers pretreated with
TMC does not require mucus in order to be adherent to both NAC and then exposed to pDMAEMA also had a statistically
cell and tissue models and binds equally well regardless.
reduced Papp for FD-4, but the reduction was far less than in
the case of pDMAEMA in the absence of NAC (Fig. 5).
Similar to E12, pretreatment of sacs with pDMAEMA (1
mg/ml) also resulted in a significant decrease in the Papp for
FD-4 (Fig. 4B). In contrast to pDMAEMA, following expo-
sure of sacs to TMC (0.1 mg/ml) there was a significant in-
crease in the Papp of FD-4 across sacs (Fig. 4b), whereas in
E12 the Papp increase was almost significant (Fig. 5). The
Papp of FD-4 across E12 exposed to TMC was, on the other
hand, significantly greater than that obtained in monolayers
treated with pDMAEMA or NAC-pDMAEMA. NAC itself
did not impact on FD-4 flux across either E12 or sacs (data
not shown). Overall, the FD-4 Papp translated to 0.14%
transport across E12; this was decreased to 0.09% with
pDMAEMA and increased to 0.19% with TMC. Basal FD-4 flux
Polymer Effects on FD-4 Transport and TEER
The effect of pDMAEMA and TMC on TEER values
and permeability to FD-4 was investigated in monolayers and
sacs. Over ten passages, E12 monolayers had a mean TEER
of 168 29 ⍀ cm2 (n ס 72), whereas HT29 monolayers had
a statistically lower mean TEER of 83 11 ⍀ cm2 (n ס 72),
p < 0.01. Basal flux of FD-4 from the apical to the basolateral
side of E12 was almost 50% lower than in HT29, data which
suggested an inverse relationship of FD-4 flux and TEER.
Following exposure of E12 to pDMAEMA (0.01 mg/ml and
0.1 mg/ml) for 60 min, the Papp of FD-4 across E12 was
significantly reduced at each concentration, and more so at
Fig. 4. (A) Papp of FD-4 across E12 and HT29 monolayers exposed to DMAEMA (0.01 and 0.1 mg/ml,
120 min). ᭿ HT29 and ᮀ E12. + p < 0.05, E12 vs. HT29; **p < 0.01, polymer-treated E12 vs. untreated
E12, unpaired t tests. n ס 6, each group. (B) Papp of FD-4 across pDMAEMA and TMC-treated sacs
(1 mg/ ml, 120 min). ᭿ Matched untreated controls vs. pDMAEMA, **p < 0.01; Matched untreated
controls vs. ᮀ TMC; *p < 0.05, n ס 5, each group. Mean SEM.

Models to Measure Mucoadhesion of Poly (Methacrylate) and N-Trimethylated Chitosan Polymers
45
Fig. 5. Papp of FD-4 across E12 and NAC-treated E12 monolayers treated with pDMAEMA and TMC
(0.1 mg/ml in each case). **p < 0.001, *p < 0.05, compared to control flux. One-way ANOVA. n ס 6,
each group. Mean SEM.
across sacs was 0.011%; this was decreased to 0.004% with Microscopy
pDMAEMA and increased to 0.019% with TMC. TEER val-
E12 and HT29 monolayers stained with EthD1 were ex-
amined by CLSM following exposure to pDMAEMA conju-
gated to coumarin. The mucus gel layer overlying E12 mono-
layer was calculated to be 100–160 ␮m thick (n ס 3). Floures-
cently labeled pDMAEMA bound to HT29 in an irregular
punctate manner. In contrast, a more continuous layer of
polymer was seen over the E12 monolayer (Fig. 6). In cor-
roboration with the adherence studies, more fluorescent
pDMAEMA was visualized attaching to E12 than to HT29.
By contrasting light and fluorescent images of the same E12
sections it was possible to demonstrate that the majority of
the fluorochrome-tagged polymer co-localized with the mu-
cus gel layer in preference to the apical membranes of the
epithelia (Fig. 7).
ues of E12 and HT29 were unchanged in the presence of
either pDMAEMA or TMC at concentrations up to 0.1 mg/
ml (data not shown); this is indirect evidence however, that
monolayer viability was in large part retained by monolayers
following polymer exposure.
Cytotoxicity of pDMAEMA and TMC
The release of LDH was measured following exposure of
monolayers to pDMAEMA and TMC in order to assess any
in vitro cytotoxic effects that might be induced by these two
polymers. After 60 min incubation with pDMAEMA (2 mg/
ml), LDH release was approximately 8% for E12 and 12% for
HT29 (Table I). LDH release was significantly higher in
HT29 exposed to pDMAEMA at all concentrations above 0.1
mg/ml in comparison to E12. Cytotoxicity was exacerbated in
HT29, presumably because it lacks a mucus gel layer. LDH
DISCUSSION
Studies on the adhesion of polymers to mucosal surfaces
release increased in a concentration-dependent relationship have mainly used tensiometry as a method of choice. This
to pDMAEMA concentration. At 0.1 mg/ml and 0.01 mg/ml method allows qualitative measurements of the detachment
concentrations of pDMAEMA, LDH release from E12 and work required to separate polymer formulations from an un-
HT29 monolayers was very low. This suggests that derlying substrate or tissue (e.g., Refs. 10, 11, 32, 33). It does
pDMAEMA does not cause significant damage to either not however, allow quantitative assessment of the adherence
monolayer type at the concentrations used for adherence. of soluble polymers to fresh viable tissue under physiologic
TMC showed similar results to pDMAEMA, with LDH re- conditions in which epithelial barrier function can be exam-
lease ranging from <1% at 0.01 mg/ml to ∼10% release at 2 ined in parallel. Neither does the method take into account
mg/ml across both monolayer types. Cytotoxicity was again the presence or impact of overlying mucus. Notable targets
significantly higher in HT29 than in E12 at the two higher for tensile tests have included isolated thawed bovine duode-
concentrations of TMC. Similar to pDMAEMA, TMC did nal (32), porcine esophageal (34), and rat intestinal tissue
not cause significant damage to either monolayer type at the (35). Recently, atomic force microscopy has been used to
concentrations used for adherence. Neither were there any quantitatively measure and image polymer interaction with
differences in LDH release between TMC and pDMAEMA mucin (36), although this technique has yet to be applied to
on either E12 or HT29 monolayers exposed to matched con- mucus-producing epithelia. In addition, mucoadhesion of mi-
centrations. In comparison to a positive control, E12 and croparticles comprising fluorescent polycarbophil to isolated
HT29 monolayers exposed to low concentrations of Triton- fresh porcine intestinal mucosa has recently been assessed in
x-100 detergent (0.01%) elaborated maximal LDH release a flow-through system using phosphate buffer (37). Our aim,
over the 60 min period.
therefore, was to assess the suitability of human intestinal E12

46
Keely et al.
Table I. LDH Release from Monolayers Following Exposure to
pDMAEMA and TMC
[pDMAEMA]
(mg/ml)
E12 (%)
HT29 (%)
2
1
0.1
0.01
7.8 1.0
3.2 0.8
1.3 0.2
0.9 0.2
11.6 1.8a
5.4 1.0b
3.2 0.8b
1.8 0.5
[TMC] (mg/ml)
2
1
0.1
7.1 0.8
3.4 0.5
1.2 0.3
0.7 0.1
100 0.1
0.9 1.0a
5.3 0.8b
2.1 0.5
1.6 0.5
100 0.1
0.01
Triton-x-100 (0.01%)
a p < 0.01, bp < 0.05, E12 versus HT29 at matched concentrations
(n ס 6). Mean SEM.
elaborates no mucus and is an appropriate control to assess
polymer adherence to the intestinal epithelium. It was also
anticipated that data achieved in HT29 would be similar to
that demonstrated in E12 exposed to a mucolytic, NAC
(16,22). We undertook adhesion experiments with an estab-
lished adhesive hydrophilic chitosan-derived polymer (TMC)
and a group of hydrophilic poly (methacrylate) polymers.
Chitosan itself is a known cationic mucoadhesive, but low
solubility limits its potential for mucosal drug delivery at
physiologic pH values. Trimethylation of chitosan eliminates
these solubility issues (30,38,39). Poly(2-(dimethylamino-
ethyl) methacrylate (pDMAEMA) is a synthetic polymer that
is finding use in gene delivery (23) and cancer targeting stud-
ies (24), and was synthesized using the polymerization tech-
nique of living radical polymerization (27). The resulting
pDMAEMA polymers have tertiary amino side chains that
can be conjugated to peptides, stabilizers, and enhancers and
may offer potential in non-injected delivery, similar to that
recently demonstrated for thiolated poly (acrylic acid)-insulin
conjugates (40).
Fig. 6. Confocal images of monolayers stained with EthD1 and ex-
posed to pDMAEMA conjugated to coumarin. (A) pDMAEMA (0.1
mg/ml, green) covers gel layer of E12 but is not associated with
epithelia (red). (B) pDMAEMA attaches in punctate fashion to
HT29 epithelia. Bar ס 10 ␮m.
monolayers for measuring mucoadhesion of a selection of
novel and established soluble fluorescent polymers and to
compare data against that obtained in fresh intestinal non-
everted sacs from rats. As yet, to our knowledge no stable
human mucus-gel forming intestinal epithelial cell line has
been used to study polymer mucoadhesion. Non-everted sacs
are a useful system to study in parallel, as they offer a rapid
and economical method of studying mucoadhesion and tissue
viability is maximized in high nutrient medium (21).
The E12 cell line has been partially characterized, but the
similarity between its mucus gel and that of the human small
intestine in vivo remains largely unexplored, although at-
tempts have been made, for example, to measure human mu-
Fig. 7. (A) Light micrograph of alcian blue/neutral red staining
of E12 exposed to pDMAEMA-coumarin conjugate. X: Mucus
gel layer; Y: Cell monolayer. (B) Fluorescent micrograph of
pDMAEMA-coumarin conjugate on same fixed section. Broken
white lines denote cell monolayer outline. Co-localization of
pDMAEMA with mucus layer is evident by superimposing micro-
cin gene expression profiles (18). The parental line, HT29, graph (B) on (A). Bars ס 25 ␮m.

Models to Measure Mucoadhesion of Poly (Methacrylate) and N-Trimethylated Chitosan Polymers
47
When exposed to pDMAEMA polymers (especially the lower in E12 than HT29 even in the absence of the polymer,
0 and 24% quaternized forms), significant adhesion levels although it is not possible to decipher whether this is due to
were obtained in E12 and these were comparable to those the presence of mucus and/or the higher basal TEER in E12,
achieved in sacs. While pDMAEMA polymers were shown to or to a combination of both. It is worth noting the that the
be as mucoadhesive as TMC, it did not, however, have the FD-4 Papp values were overall far higher in HT29 and E12
same level of adherence as TMC for epithelial membranes per than in tighter epithelia such as Caco-2 and thus reflecting the
se, especially at high concentrations. This difference was evi- leakiness of these tissues. Data from E12 (18) and mucus-
dent from NAC-treated E12, HT29 and sac experiments. It covered co-cultures (16) suggests that mucus should act as an
was, in addition, clearly visible in the CLSM images of impediment to the flux of hydrophobic drugs but not to hy-
pDMAEMA binding predominantly to the mucus layer of drophilic agents such as dextrans. While no significant change
E12 rather than to the epithelium of HT29. TMC was both in TEER was shown in E12 in the presence of pDMAEMA
muco- and bioadhesive to the same extent and it is likely this despite the decrease in FD-4 flux, it is worth noting that para-
polymer should retain some mucosal adherence even in the cellular flux and TEER are not always inversely correlated,
presence of high mucus turnover.
especially in leaky epithelia such as the small intestine where
While it has been shown that electrostatic interaction is a transcellular resistors contribute significantly to overall resis-
major factor in the adhesion of cationic chitosan to purified tance (48). The fact that basal FD-4 across both E12 and sacs
anionic pig mucin (36), charge-based interactions of polymers was detectable confirmed that these are low resistance epi-
with mucus remains poorly understood. The main adhesive thelial models with similar levels of hydrophilic paracellular
interactions between adhesive polymers and mucus occur via marker fluxes to those obtained in human jejunum (49) and in
covalent bonding, hydrogen bonding and electrostatic and hy- recently developed low resistance intestinal cell lines (50).
drophobic interactions or a combination of these (41). A sig-
TMC had opposite effects to pDMAEMA on FD-4 flux
nificant contribution to these interactions is made by the in E12 and sacs. It increased epithelial permeability to the
negative charge carried by the terminal sialyl- and ester sul- paracellular flux marker. This data was in keeping with the
fate moieties on mucin glycans (42). A decrease in mucoad- well known tight junction opening effects of chitosan and
hesion with increasing quaternization of TMC has previously TMC (5,43,44). This property of chitosan(s) may be helpful in
been shown, although both quaternized and unquaternized enhancing drug delivery across mucosal surfaces such as the
TMC were overall still less mucoadhesive than chitosan (11). cornea (51,52). The capacity of pDMAEMA to reduce mu-
Furthermore, it appears that increased quaternization of cosal permeability may be especially advantageous for topical
TMC leads to an associated increase in polymer absorption- localized delivery where a reduction in systemic drug absorp-
enhancing properties (43,44), perhaps indicating that the re- tion may be desired. Barrier enhancement is a desirable prop-
maining unassociated free polymer has increased access to the erty in adhesive compounds, perhaps offering potential for
plasma membrane and tight junction proteins. In the case of treatment of inflammatory bowel disease where epithelia and
pDMAEMA, quaternization should theoretically alter the ef- lamnia propria immunocytes are exposed to damaging patho-
fect of electrostatic interactions between the polymer and gly- gens (53).
coproteins due to the increased charged groups on the poly-
It is known that injected cationic polymers may be cyto-
mer (45). Of the four levels of quaternization tested however, toxic (54). Recent data demonstrated that pDMAEMA has a
both the unquaternized and 24% quaternized pDMAEMA cytotoxic effect on Jurkat and U937 lymphocyte cell lines
showed the highest levels of mucoadhesion to E12 and sacs. (55). In contrast, our data showed that LDH release from
Both 10% and 32% quaternized pDMAEMA, however, both types of human intestinal epithelial monolayers exposed
showed low levels of mucoadhesion. This strongly suggests to high concentrations of pDMAEMA (and TMC) was low.
that, unlike chitosan and TMC, pDMAEMA binding to mu- Furthermore, because LDH release was less in E12 than in
cus in monolayers or sacs was not strongly influenced by elec- NAC-treated E12 or HT29 exposed to the polymer, this sug-
trostatic charges arising from fully charged species. This does gests that the mucus gel layer overlying E12 cells provides
not rule out the overall importance of subtle hydrogen bond- additional cytoprotection. The confocal micrographs further
ing in mucoadhesion. Other conjugation strategies are likely demonstrated that there was little direct interaction between
to further increase pDMAEMA mucoadhesion. A similar pDMAEMA and E12 epithelial cell membranes, above which
type of polymer, poly(acrylic) acid, was thiolated via cysteine mucus is interposed, and this may explain why it caused no
conjugation and a significant increase in mucoadhesion was cytotoxicity compared to HT29. Moreover, mucus gels also
measured, due to the formation of disulphide bridges with decreased the efficiency of plasmid transfection in an intesti-
mucosal glycoproteins (46).
nal epithelial co-culture model (56), so it is not unlikely that
The adhesion of unquaternized pDMAEMA to E12 the mucus gel layer may limit access of pDMAEMA to the
monolayers was shown to significantly reduce the apparent epithelial surface of E12. pDMAEMA is unlikely to be ab-
permeability of FD-4, whereas HT29 monolayers treated with sorbed and should be excreted in faeces, thus reducing po-
pDMAEMA showed no decrease in permeability. A signifi- tential systemic toxicity of this non-biodegradable polymer.
cant decrease in FD-4 flux also occurred across intestinal sacs Overall, our data suggest that the assumption that
that had been treated with pDMAEMA compared to un- pDMAEMA may be cytotoxic is premature, since its dispo-
treated sacs. The reduction of flux therefore appears to be sition depends on the surface characteristics and type of tissue
dependent on the presence of a combination of the mucosal target and as to whether the polymer is internalized.
mucus gel and pDMAEMA. Another mucoadhesive polymer,
One of the potential applications of muco-integrated
carbopol 934P increases the viscosity of mucins (47) and per- polymer therapeutics is in the area of topical anti-microbial
haps this mechanism might explain, at least in part, the mu- drug delivery. Studies have shown that poly(ethylene) glycol
cosal protection by pDMAEMA. The FD-4 Papp was also delivered by lavage to the mucus lining of the intestine pre-

48
Keely et al.
thal sepsis due to intestinal Pseudomonas aeruginosa. Gastroen-
terology 126:488–498 (2004).
7. S. Higo, K. Ori, H. Takeuchi, H. Yamamoto, T. Hino, and Y.
Kawashima. A novel evaluation method of gastric mucoadhesive
property in vitro and the mucoadhesive mechanism of tetracy-
cline-sucralfate acidic complex for the eradication of Helicobacter
pylori. Pharm. Res. 21:413–419 (2004).
8. M. Bogataj, T. Vovk, M. Kerec, A. Dimnik, I. Grabnar, and A.
Mrhar. The correlation between zeta potential and mucoadhe-
sion strength on pig vesical mucosa. Biol. Pharm. Bull. 26:743–
746 (2003).
9. C.-M. Lehr, J. A. Bouwstra, J. J. Tukker, and H. E. Junginger.
Intestinal transit of bioadhesive microspheres in an in situ loop in
the rat -A comparative study with copolymers and blends based
on poly(acrylic acid). J. Control. Rel. 13:51–62 (1990).
10. D. Quintanar-Guerrero, R. Villalobos-Garcia, E. Alvarez-Colin,
and J. M. Cornejo-Bravo. In vitro evaluation of the bioadhesive
properties of hydrophobic polybasic gels containing N,N-
dimethylaminoethyl methacrylate-co-methyl methacrylate. Bio-
materials 22:957–961 (2001).
vented death in mucosally-compromised mice infected with
Pseudomonas aeruginosa (6). Chitin and chitosan have also
been reported to protect mice against challenge with a range
of bacteria (57), while ethyl cellulose appears to prevent bac-
terial translocation in the liver (58) Recently, quaternized
pDMAEMA was directly polymerized onto filter paper and
was shown to have antimicrobial action against Bacillus sub-
tilis and Escherichia coli (25). It seems, therefore, that anti-
bacterial action is present inherently in a number of common
polymers, but that the mechanisms remain largely undeci-
phered. The lack of a reproducible reductionist model should
expand investigations into the cellular basis of these effects.
E12 may prove a suitable model system for ascertaining the
role of human intestinal mucus in pathogen-epithelial inter-
actions and for screening polymers in order to prevent access
to the apical membrane. Initial experiments, for example,
suggest that cytotoxicity of cholera toxin can be reduced in
E12 but not HT29 in the presence of pDMAEMA (Keely and
Brayden, unpublished data).
11. D. Snyman, J. H. Hamman, and A. F. Kotze. Evaluation of the
mucoadhesive properties of N-trimethyl chitosan chloride. Drug
Dev. Ind. Pharm. 29:61–69 (2003).
12. S. Kockisch, G. D. Rees, S. A. Young, J. Tsibouklis, and J. D.
Smart. In situ evaluation of drug-loaded microspheres on a mu-
cosal surface under dynamic test conditions. Int. J. Pharm. 276:
51–58 (2004).
13. I. J. Hidalgo, T. J. Raub, and R. T. Borchardt. Characterisation of
the human colon carcinoma cell line (Caco-2) as a model system
for intestinal epithelial permeability. Gastroenterology 96:736–
749 (1989).
14. T. Lesuffleur, A. Barbat, E. Dussaulx, and A. Zweibaum. Growth
adaptation to methotrexate of HT-29 human colon carcinoma
cells is associated with their ability to differentiate into columnar
absorptive and mucus-secreting cells. Cancer Res. 50:6334–6343
(1990).
15. A. Wikman. J, Karlsson, I. Carlstedt, and P. Artursson. A drug
absorption model based on the mucus layer producing human
intestinal goblet cell line HT29-H. Pharm. Res. 10:843–852
(1993).
In summary, this study has demonstrated that a mucus-
producing sub-clone of HT29 displays adhesive properties
similar to an established method, intestinal sacs. It may have
useful application in screening mucoadhesives under physi-
ologic conditions. A group of methacrylate-based polymers,
pDMAEMA, were made by a novel method of living radical
polymerization and was shown to be as mucoadhesive as a
soluble chitosan derivative (TMC), and to co-localize in the
mucus gel. Furthermore, low level cytoxicity of pDMAEMA
and TMC in monolayers was reduced still further in presence
of mucus. Finally, this cationic polymer could be differenti-
ated mechanistically from TMC not only on the basis of ad-
hesion, but also by its capability to reduce flux of a paracel-
lular marker in the presence of mucus in the two systems.
16. C. Meaney and C. O’Driscoll. Mucus as a barrier to the perme-
ability of hydrophilic and lipophilic compounds in the absence
and presence of sodium taurocholate micellar systems using cell
culture models. Eur. J. Pharm. Sci. 8:167–175 (1999).
17. E. Walter, S. Janich, B. J. Roessler, J. M. Hilfinger, and G. L.
Amidon. HT29-MTX/Caco-2 cocultures as an in vitro model for
the intestinal epithelium: in vitro-in vivo correlation with perme-
ability data from rats and humans. J. Pharm. Sci. 85:1070–1076
(1996).
18. I. Behrens, P. Stenberg, P. Artursson, and T. Kissel. Transport of
lipophilic drug molecules in a new mucus-secreting cell culture
model based on HT29-MTX cells. Pharm. Res. 18:1138–1145
(2001).
ACKNOWLEDGMENTS
We would like to thank Professor Per Artursson (Uni-
versity of Uppsala, Sweden) for the kind gift of E12 cells. We
also thank Eamonn Fitzpatrick and Anne Cullen at Univer-
sity College Dublin for technical assistance with histology and
CLSM, respectively. Funding for this work was provided by
Genzyme Corporation (MA, USA). We thank Thomas
Neenan of Genzyme for initiating the collaboration.
19. I. Behrens, A. I. Pena, M. J. Alonso, and T. Kissel. Comparative
uptake studies of bioadhesive and non-bioadhesive nanoparticles
in human intestinal cell lines and rats: the effect of mucus on
particle adsorption and transport. Pharm. Res. 19:1185–1193
(2002).
20. R. R. Levine, W. F. McNary, P. J. Kornguth, and R. LeBlanc.
Histological reevaluation of everted gut technique for studying
intestinal absorption. Eur. J. Pharmacol. 9:211–219 (1970).
21. L. Barthe, J. F. Woodley, S. Kenworthy, and G. Houin. An im-
proved everted gut sac as a simple and accurate technique to
measure paracellular transport across the small intestine. Eur. J.
Drug Metab. Pharmacokinet. 23:313–323 (1998).
REFERENCES
1. J. M. Gu, J. R. Robinson, and S. H. Leung. Binding of acrylic
polymers to mucin /epithelial surfaces: structure-property rela-
tionships. Crit. Rev. Ther. Drug Carrier Syst. 5:21–67 (1988).
2. A. Bernkop-Schnurch and G. Walker. Multifunctional matrices
for oral peptide delivery. Crit. Rev. Drug Carrier Syst. 18:459–501
(2001).
3. M. K. Chourasia and S. K. Jain. Pharmaceutical approaches to
colon targeted delivery systems. J. Pharm. Pharm. Sci. 6:33–66
(2003).
4. D. S. Jones, M. S. Lawlor, and A. D. Woolfson. Rheological and
mucoadhesive characterisation of polymeric systems composed
of poly(methylvinylether-co-maleic anhydride) and poly(vinyl-
pyrrolidone), designed as platforms for topical drug delivery. J.
Pharm. Sci. 92:995–1007 (2003).
5. J. Smith, E. Wood, and M. Dornish. Effect of chitosan on epi-
thelial cell tight junctions. Pharm. Res. 21:43–49 (2004).
22. A. L. Scheffner. The reduction in vitro in viscosity of mucopro-
tein solutions by a new mucolytic agent, N-acetyl-L-cysteine.
Ann. N. Y. Acad. Sci. 106:298–310 (1963).
23. P. van de Wetering, J. Y. Cherng, H. Talsma, D. J. Crommelin,
and W. E. Hennink. 2-(Dimethylamino)ethyl methacrylate based
(co)polymers as gene transfer agents. J. Control. Rel. 53:145–153
(1998).
6. L. Wu, O. Zaborina, A. Zaborin, E. B. Chang, M. Musch, C. 24. F. J. Verbaan, C. Oussoren, C. J. Snel, D. J. Crommelin, W. E.
Holbrook, J. Shapiro, J. R. Turner, G. Wu, K. Y. Lee, and J. C.
Alverdy. High-molecular-weight polyethylene glycol prevents le-
Hennink, and G. Storm. Steric stabilization of poly(2-
(dimethylamino)ethyl methacrylate-based polyplexes mediates

Models to Measure Mucoadhesion of Poly (Methacrylate) and N-Trimethylated Chitosan Polymers
49
prolonged circulation and tumor targeting in mice. J. Gene Med. 42. A. P. Corfield, D. Carroll, N. Myerscough, and C. S. Probert.
6:64–75 (2004).
Mucins in the gastrointestinal tract in health and disease. Front.
Biosci. 6:D1321–D1357 (2001).
43. J. H. Hamman, C. M. Schultz, and A. F. Kotze. N-trimethyl chi-
tosan chloride: optimum degree of quaternization for drug ab-
sorption enhancement across epithelial cells. Drug Dev. Ind.
Pharm. 29:161–172 (2003).
44. C. Jonker, J. H. Hamman, and A. F. Kotze. Intestinal paracellular
permeation enhancement with quaternised chitosan: in situ and in
vitro evaluation. Int. J. Pharm. 238:205–213 (2002).
25. S. B. Lee, R. R. Koepsel, S. W. Morley, K. Matyjaszewski, Y. Sun,
and A. J. Russell. Permanent, nonleaching antibacterial surfaces.
1. Synthesis by atomic transfer radical polymerization. Biomac-
romolecules 5:877–882 (2004).
26. A. P. Corfield, N. Myerscough, R. Longman, P. Sylvester, S. Arul,
and M. Pignatelli. Mucins and mucosal protection in the gastro-
intestinal tract: new prospects for mucins in the pathology of
gastrointestinal disease. Gut 47:589–594 (2000).
27. D. M. Haddleton, M. C. Crossman, B. H. Dana, D. J. Duncalf, A.
M. Heming, D. Kukulj, and A. J. Shooter. Atom transfer poly-
merization of methyl methacrylate mediated by alkylpyridyl-
methanimine type ligands, copper(I) bromide, and alkyl halides
in hydrocarbon solution. Macromolecules 32:2110–2119 (1999).
28. C. Korzeniewski and D. M. Callewaert. An enzyme-release assay
for natural cytotoxicity. J. Immunol. Methods 64:313–320 (1983).
29. S. L. Gilat, A. Adronov, and J. M. J. Fréchet. Modular approach
to the accelerated convergent growth of laser dye-labeled
poly(aryl ether) dendrimers using a novel hypermonomer. J. Org.
Chem. 64:7474–7484 (1999).
30. A. B. Sieval, M. Thanou, A. F. Kotze, J. C. Verhoef, J. Brussee,
and H. E. Junginger. Preparation and NMR characterization of
highly substituted N-trimethyl chitosan chloride. Carbohydrate
Polymers 36:157–165 (1998).
31. M. Leonard, E. Creed, D. Brayden, and A. W. Baird. Evaluation
of the Caco-2 monolayer as a model epithelium for iontophoretic
transport. Pharm. Res. 17:1181–1188 (2000).
45. R. K. Willits and W. M. Saltzman. Synthetic polymers alter the
structure of cervical mucus. Biomaterials 22:445–452 (2001).
46. V. M. Leitner, M. K. Marschutz, and A. Bernkop-Schnurch. Mu-
coadhesive and cohesive properties of poly (acrylic) acid-cysteine
conjugates with regard to their molecular mass. Eur. J. Pharm.
Sci. 18:89–96 (2003).
47. S. N. E. Foster, J. P. Pearson, D. A. Hutton, A. Allen, and P. W.
Dettmar. Interaction of polyacrylates with porcine pepsin and the
gastric mucus barrier: a mechanism for mucosal protection. Clin.
Sci. 87:719–726 (1994).
48. M. S. Balda, J. A. Whitney, C. Flores, S. Gonzalez, M. Cereijido,
and K. Matter. Functional dissociation of paracellular permeabil-
ity and transepithelial electrical resistance and disruption of the
apical-basolateral intramembrane diffusion barrier by expression
of a mutant tight junction membrane protein. J. Cell Biol. 134:
1031–1049 (1996).
49. H. Lennernas, S. Nylander, and A. L. Ungell. Jejunal permeabil-
ity: a comparison between the Ussing chamber technique and
single pass perfusion in humans. Pharm. Res. 14:667–671 (1997).
50. S. Tavelin, V. Milovic, G. Ocklind, S. Olsson, and P. Artursson. A
conditionally immortalized cell line for studies of intestinal drug
transport. J. Pharm. Exp. Ther. 290:1212–1221 (1999).
32. D. Acilli, G. Menghi, G. Bonacucina, P. Di Martino, and G. F.
Palmieri. Mucoadhesion dependence of pharmaceutical polymers
on mucosa characteristics. Eur. J. Pharm. Sci. 22:225–234 (2004).
33. S. E. Harding. Mucoadhesive interactions. Biochem. Soc. Trans.
31:1036–1041 (2003).
34. H. K. Batchelor, D. Banning, P. W. Dettmar, F. C. Hampson, I.
G. Jolliffe, and D. Q. Craig. An in vitro mucosal model for pre-
diction of the bioadhesion of alginate solutions to the oesopha-
gus. Int. J. Pharm. 238:123–132 (2002).
35. J. Shimoda, H. Onishi, and Y. Machida. Bioadhesive character-
istics of chitosan microspheres to the mucosa of rat small intes-
tine. Drug Dev. Ind. Pharm. 27:567–576 (2001).
36. M. P. Deacon, S. McGurk, C. J. Roberts, P. M. Williams, S. J.
Tendler, M. C. Davies, S. S. Davis, and S. E. Harding. Atomic
force microscopy of gastric mucin and chitosan mucoadhesive
systems. Biochem. J. 348:557–563 (2000).
37. A. Krauland and A. Bernkop-Schnurch. Thiomers: development
and in vitro evaluation of peroral microparticulate peptide deliv-
ery system. Eur. J. Pharm. Biopharm. 57:181–187 (2004).
38. A. F. Kotze, H. L. Luessen, B. J. de Leeuw, B. G. de Boer, J. C.
Verhoef, and H. E. Junginger. N-trimethyl chitosan chloride as a
potential absorption enhancer across mucosal surfaces: in vitro
evaluation in intestinal epithelial cells (Caco-2). Pharm. Res. 14:
1197–1202 (1997).
39. A. F. Kotze, H. L. Luessen, A. G. de Boer, J. C. Verhoef, and H.
E. Junginger. Chitosan for enhanced intestinal permeability:
Prospects for derivatives soluble in neutral and basic environ-
ments. Eur. J. Pharm. Sci. 7:145–151 (1999).
40. P. Calceti, S. Salmaso, G. Walker, and A. Bernkop-Schnurch.
Development and in vivo evaluation of an oral insulin-PEG de-
livery system. Eur. J. Pharm. Sci. 22:315–323 (2004).
41. S. E. Harding, S. S. Davis, M. P. Deacon, and I. Fiebrig. Biopoly-
mer mucoadhesives. Biotechnol. Genet. Eng. Rev. 16:41–86
(1999).
51. I. P. Kaur and R. Smitha. Penetration enhancers and ocular bio-
adhesives: two new avenues for ophthalmic drug delivery. Drug
Dev. Ind. Pharm. 28:353–369 (2002).
52. G. Di Colo, Y. Zambito, S. Burgalassi, A. Serafini, and M. F.
Saettone. Effect of chitosan on in vitro release and ocular deliv-
ery of ofloxacin from erodible inserts based on poly(ethylene
oxide). Int. J. Pharm. 248:115–122 (2002).
53. J. P. Kraehenbuhl, E. Pringault, and M. R. Neutra. Review ar-
ticle: Intestinal epithelia and barrier functions. Aliment. Pharma-
col. Ther. 11:3–9 (1997).
54. S. Rosengren and K. E. Arfors. Polycations induce microvascular
leakage of macromolecules in hamster cheek pouch. Inflamma-
tion 15:159–172 (1991).
55. R. A. Jones, M. H. Poniris, and M. R. Wilson. pDMAEMA is
internalised by endocytosis but does not physically disrupt endo-
somes. J. Control. Rel. 96:379–391 (2004).
56. S. A. Cryan and C. M. O’Driscoll. Mechanistic studies on non-
viral gene delivery to the intestine using in vitro differentiated
cell culture models and an in vivo rat intestinal loop. Pharm. Res.
20:569–575 (2003).
57. Y. Okawa, M. Kobayashi, S. Suzuki, and M. Suzuki. Comparative
study of protective effects of chitin, chitosan, and N-acetyl chito-
hexaose against Pseudomonas aeruginosa and Listeria monocy-
togenes infections in mice. Biol. Pharm. Bull. 26:902–904 (2003).
58. X. Wang, R. Andersson, V. Soltesz, W. Guo, and S. Bengmark.
Water-soluble ethylhydroxyethyl cellulose prevents bacterial
translocation induced by major liver resection in the rat. Ann.
Surg. 217:155–167 (1993).