Alimentary Pharmacology & Therapeutics
A new concept in colonic drug targeting: a combined
pH-responsive and bacterially-triggered drug delivery technology
V. C. IBEKWE*, M. K. KHELA , D. F. EVANS  & A. W. BASIT*
*Department of Pharmaceutics,
The School of Pharmacy, University
of London, London, UK;  The Wingate
Institute, Queen Mary College,
University of London, London, UK
Correspondence to:
A. W. Basit, Department of
Pharmaceutics, The School of
Pharmacy, University of London,
29-39 Brunswick Square, London,
WC1N 1AX, UK.
E-mail: abdul.basit@pharmacy.ac.uk
Publication data
Submitted 13 June 2008
First decision 4 July 2008
Resubmitted 15 July 2008
Accepted 16 July 2008
Epub Accepted Article 19 July 2008
ª 2008 The Authors
Journal compilation ª 2008 Blackwell Publishing Ltd
doi:10.1111/j.1365-2036.2008.03810.x
SUMMARY
Background
Current approaches to colonic drug delivery exploit one of two main
physiological characteristics: the pH change or increase in bacterial
numbers along the gastrointestinal tract. Here, we describe a new con-
cept in targeted delivery, which combines these triggers to improve
colonic delivery.
Aim
To assess the in-vivo targeting performance of a novel colonic delivery
coating comprising a mixture of pH-responsive enteric polymer (Eudra-
git S) and biodegradable polysaccharide (resistant starch) in a single
layer matrix film.
Methods
Tablets (radio-labelled) were film-coated with the dual-mechanism coat-
ing and administered in a three-way crossover study to eight healthy
volunteers (i) without food, (ii) with breakfast or (iii) 30 min before
breakfast. The site of intestinal disintegration was assessed using
gamma scintigraphy.
Results
The coated tablets were able to resist breakdown in the stomach and
small intestine. Consistent disintegration of the dosage form was seen at
the ileocaecal junction ⁄ large intestine. The site of disintegration
remained unaffected by feeding.
Conclusions
The dual-mechanism (pH ⁄ bacterial) coating provides colon-specificity.
Each trigger mechanism has the capacity to act as a failsafe, ensuring
appropriate targeting in the gastrointestinal tract. This platform technol-
ogy has potential for systemic applications or the treatment of local dis-
orders of the large intestine, such as inflammatory bowel disease.
Aliment Pharmacol Ther 28, 911–916
911
912 V . C . I B E K W E et al.
INTRODUCTION
Colon-specific drug delivery is essential for the treat-
ment of inflammatory conditions affecting the colon,
such as ulcerative colitis and Crohn’s disease.1–3 It is
an approach that could be extended for the treatment
of colonic cancers,4 vaccination5 or used for systemic
delivery of drug molecules, peptides or proteins.6 Here,
we describe a new concept in oral colonic delivery,
which exploits two physiological parameters to ensure
site-specific targeting.
Current approaches to colonic drug delivery mainly
utilise one of two intestinal characteristics: the pH
change along the gastrointestinal tract or the increase
in bacterial numbers in the distal gut.7 The former
approach uses pH responsive polymer coatings, such
as Eudragit S (dissolves >pH 7), to effect release from
a formulation as the intestinal pH increases distally to
a maximum at the ileocaecal junction.8, 9 Despite con-
siderable variability in the site of release, this
pH-responsive approach to colonic delivery has been
embraced clinically, with several established products
on the market and more having been introduced
recently for the treatment of inflammatory bowel dis-
eases. These include Asacol (Procter & Gamble, Mason,
OH, USA and Tillotts, Ziefen, Switzerland), Mesavant
(Shire, Basingstoke, UK), Lialda (Shire, Wayne, PA,
USA) each of which incorporates mesalazine (mesala-
mine), and Budenofalk (Dr Falk GmbH, Freiburg,
Germany), which delivers the corticosteroid budesonide.
The bacterial approach to colonic targeting exploits
the fact that there are around one hundred billion cfu ⁄ ml
of bacteria in the colon10 of many hundreds of different
species.11, 12 The colonic microbiota are said to have a
metabolic potential equal to or greater than that of the
liver13and their metabolic and fermentative capabilities
can be exploited to effect drug release. Examples of
molecules which utilise the microbiotia for drug delivery
are sulfasalazine, balsalaside and olsalazine, which are
pro-drugs of mesalazine and are cleaved by colonic bac-
teria to release the active moiety. This is highly drug-
specific, which is a limitation of the approach and so
systems have been developed, which could theoretically
be applied to any drug. These are coatings or matrices
based on polysaccharides.7, 14 Some polysaccharides can
resist pancreatic digestion, but are degraded by the colo-
nic bacteria making them suitable for the purposes of
colonic drug delivery. The so-called ‘resistant starch’ is
one such polysaccharide, which has been successfully
developed into a colon-specific coating (COLAL,
Alizyme, Cambridge, UK), which can be applied to dos-
age forms such as tablets, capsules or granules.15, 16 This
is independent of pH and time and can prevent drug
release in the stomach and small intestine, only allowing
drug release upon entry to the colon upon exposure to
the increased levels of colonic bacteria. In COLAL, ethyl-
cellulose is used to control the swelling of starch and
in doing so produces a slow, sustained release of drug
once in the colon. This system has been investigated
with a variety of drugs and is in late stage clinical trials.
To overcome the limitations associated with the cur-
rent marketed products for inflammatory bowel disease,
a simple, robust and reliable carrier for colonic delivery
was developed. This coating system combines the bac-
terial and pH mediated approaches. These independent
and complementary release mechanisms should over-
come any limitations associated with single-trigger sys-
tems and improve site-specificity. This was achieved by
combining a pH-sensitive polymer with resistant starch
and using this mixture as a film coating matrix, which
was applied to tablets. This technology was tested in
healthy volunteers to assess the site of disintegration
using gamma scintigraphic techniques. Further to this,
the effect of feeding status on dosage form transit and
disintegration was assessed.
MATERIALS AND METHODS
Preparation, coating and radio-labelling of
tablets
Model tablet cores (8 mm diameter, 200 mg weight),
which were suitable for film coating, were prepared.17
A coating mixture was prepared by mixing Eurylon VII
[high amylose maize starch (Roquette, Lestrem, France)]
aqueous dispersion and Eudragit S ethanolic solution
(Evonik, Darmstadt, Germany) (ratio 3:7 solids). Addi-
tional components were included to aid film formation
[triethyl citrate (Lancaster Synthesis, Lancashire, UK)
and glyceryl monostearate (Huls AG, Witten,
Germany)]. The mixture was spray coated onto the
tablet cores using a Strea-1 bottom spray fluidized bed
coater (Aeromatic AG, Bubendorf, Switzerland). The
radiolabelling procedure was carried out by complexing
technetium-99m (99mTc) to diethylenediaminepenta-
acetic acid (Amersham, Hammersmith Hospital,
London).17, 18 Lactose was dissolved in this complex
and a 1 mm diameter hole was drilled into the coated
tablet surface and the radiolabelled lactose used to fill
the hole (to an activity of 4 MBq) and the hole sealed
ª 2008 The Authors, Aliment Pharmacol Ther 28, 911–916
Journal compilation ª 2008 Blackwell Publishing Ltd
 A N E W P H ⁄ B A C T E R I A L T R I G G E R E D C O N C E P T F O R C O L O N I C D R U G D E L I V E R Y 913

with bone cement (DeTrey1 Zinc, DentsPly GmbH, Con-
stance, Germany). The radiolabelling procedure was
then validated by methods described previously.17, 18
Study protocol
The study protocol and radioactivity administration were
approved by the Committee on Ethics of Human Research
of the East London and City Health Authority and the
Administration of Radioactive Substances Advisory Com-
mittee (Department of Health). The study followed the
tenets of the Declaration of Helsinki (1964). Eight healthy
adult male volunteers, aged 22–34 years, took part in the
study with treatment order randomized and 7 day wash-
out periods observed. After an overnight fast, the volun-
teers were administered a radiolabelled coated tablet with
150 mL of water under three different feeding regimes:
(i) Fasted
(ii) Fed [after a standard breakfast: 30 g cornflakes;
100 mL semi-skimmed milk; two slices of toasted brown
bread; 5 g margarine; and 150 mL orange juice (392 kcal)]
(iii) Prefeed (30 min before the standard breakfast).
A standard lunch was provided at 4 h postdose, after
which water and other non-alcoholic drinks were
freely available. Dinner was provided at 9 h.
A sealed point source of 0.5 MBq 99mTc was taped to the
most lateral part of the lower costal margin to be used for
correction of posture between imaging and as an anatomi-
cal reference marker. Images were acquired using a single
head camera (GE Maxicamera 400AC) with energy
windows set at 126–150 KeV. Images were acquired over a
1-min period, at (at most) 10 min intervals for up to 12 h.
The acquired images for each volunteer were replayed on
a computer and processed using Nucmed software
(MicasX, Farnborough, UK). The images were analysed
and the gastric emptying, upper small intestinal transit,
ileocaecal junction residence, caecal arrival and tablet
disintegration times were derived. As the images were not
continuous, the time of the various events were taken as
the mean of the two time points either side of the event.
Statistical analysis
The median values for the transit and disintegration
times were calculated from the assumption that the tab-
let transit and disintegrations were Bernoulli random
events, as described by Podczeck et al.19 The disintegra-
tion times, ileocaecal junction residence times, gastric
emptying times, upper small intestinal transit times and
caecal arrival times were analysed for differences
between fed, fasted and prefeed doses using one-way
ANOVA followed by post hoc analysis using Tukey’s test.
Significance was taken where P < 0.05. Analysis was
carried out using SPSS vs. 15.0.
RESULTS
The transit times and site of disintegration for the
dual pH- and bacterially-triggered coated tablets in
eight healthy volunteers are shown in Tables 1–3.

Table 1. Transit and disintegration times (min) for tablets coated with the dual pH- and bacterially-triggered coating (fasted
dose)

Ileocaecal
Gastric Upper small junction
emptying intestinal residence Caecal arrival Time of Site of
Volunteer time transit time time time disintegration disintegration

1 58 163 16 – 237 Ileocaecal junction

2 26 152 10 188 200 Ascending colon
3 30 130 30 190 201 Ascending colon
4 58 170 40 268 292 Ascending colon
5 94 189 10 293 465 Transverse colon
6 * * * * * *
7 13 157 77 247 274 Ascending colon
8 74 141 123 338 614 Ascending colon
Median 37 (20–51) 155 (138–165) 23 (19–49) 247 (189–281) 355 (201–355)
(interquartile range)

* Subject absent from study day.

ª 2008 The Authors, Aliment Pharmacol Ther 28, 911–916

Journal compilation ª 2008 Blackwell Publishing Ltd
914 V . C . I B E K W E et al.

Table 2. Transit and disintegration times (min) for tablets coated with the dual pH- and bacterially-triggered coating (fed
dose)

Ileocaecal
Gastric Upper small junction
emptying intestinal residence Caecal arrival Time of Site of
Volunteer time transit time time time disintegration disintegration

1 101 117 10 228 240 Ascending colon

2 20 269 19 308 316 Ascending colon
3 224 110 46 380 510 Ascending colon
4 30 173 163 366 415 Ascending colon
5 138 162 60 360 555 Ascending colon
6 219 71 95 385 523 Ascending colon
7 110 155 61 326 465 Splenic flexure
8 217 116 89 422 455 Ascending colon
Median 110 (30–217) 117 (110–162) 60 (19–89) 343 (288–370) 435 (297–476)
(interquartile range)

The median gastric emptying time was 37 min (inter-
quartile range 20–51) in the fasted state and 110 min
(interquartile range 30–217) in the fed state. In the
prefeed state, the median gastric emptying time
occurred at 35 min (interquartile range 20–63).
Statistical analysis failed to find any significant
differences in gastric emptying times between the
feeding regimens (P > 0.05). In one case, the tablet
failed to empty from the stomach for the 12 h of
the study (Subject 8, prefeed dose). The remaining
tablets spent statistically similar amounts of time in
the small intestine (P > 0.05) in each feeding
regimen.
In all cases, the tablets were able to resist break-
down in the upper gastrointestinal tract and this
included the tablet that remained in the stomach. This
was seen under all feeding conditions. Disintegration
of the dosage form was consistently seen at the ileo-
caecal junction or in the large intestine (excluding the
tablet that did not empty from the stomach). There
was no discernable relationship between the feeding
regimen and the site or time of disintegration.

Table 3. Transit and disintegration times (min) for tablets coated with the dual pH- and bacterially-triggered coating (pre-
feed dose)

Ileocaecal
Gastric Upper small junction Caecal
emptying intestinal residence arrival Time of Site of
Volunteer time transit time time time disintegration disintegration

1 18 178 48 - 244 Ileocaecal junction

2 45 85 209 - 339 Ileocaecal junction
3 118 115 117 - 350 Ileocaecal junction
4 33 162 105 300 390 Hepatic flexure
5 240 133 26 399 678 Splenic flexure
6 37 228 107 372 523 Ascending colon
7 20 137 23 180 244 Ascending colon
8 * * * * *
Median (interquartile range) 35 (20–63) 135 (108–166) 77 (25–110) 300 (180–322) 345 (213–423)

* Tablet did not empty from the stomach.

ª 2008 The Authors, Aliment Pharmacol Ther 28, 911–916

Journal compilation ª 2008 Blackwell Publishing Ltd
 A N E W P H ⁄ B A C T E R I A L T R I G G E R E D C O N C E P T F O R C O L O N I C D R U G D E L I V E R Y 915
DISCUSSION
Gastric emptying times in the fasted state were all
under 2 h being under the influence of phase III con-
tractions of the migrating myoelectric complex (MMC),
which cycles every 90–120 min.20 The presence of
food (fed dose) interrupts the MMC and accounts for
the delayed emptying seen with this dose. The prefeed
dose (tablet given 30 min before food) was introduced
to expand upon the traditional fed and fasted states
and to provide a broader scope of dosing scenarios.
Highly variable gastric emptying times were observed
in this prefeed state and in one subject (subject 8) the
tablet did not empty from the stomach. This anoma-
lous result was useful in that it was able to confirm
the ability of the film coating to resist breakdown for
up to 12 h in the stomach (the study was stopped after
this time period). This means that the risk of dose
dumping is low with this novel delivery system, which
has a special significance when the new high-dose
products for ulcerative colitis are considered.21
To be considered colon-specific, a dosage form must
resist breakdown in the upper gastrointestinal tract.
These dual-mechanism coated tablets were able to
avoid disintegration in the upper gastrointestinal tract
by virtue of the Eudragit S component, which is insol-
uble at pH < 7. The gastric and proximal small intesti-
nal pH remains below this neutral value.8 The Eudragit
S has a further function; it controls the swelling of
starch on passage through the gut. The starch compo-
nent of the coating gives another form of upper gas-
trointestinal resistance, specifically resistance to
digestion by mammalian amylase enzymes secreted by
the pancreas. There are three main types of starch
(easily digestible starch, partially resistant starch and
resistant starch) and only if starch is in its resistant
form is it able to avoid mammalian pancreatic amylase
digestion. The presence of resistant starch in this dos-
age form was confirmed by these results; even in the
fed state where amylase secretion rates increase from
5–250 U ⁄ min22 (interdigestive) to 500–2000 U ⁄ min22
(postprandial), resistance to pancreatic digestion was
seen.
The second prerequisite of colon-specific delivery
is the ability to disintegrate upon reaching the large
intestine. In all cases, disintegration of the dosage
form was seen at the ileocaecal junction or in the
large intestine. This is in contrast to studies on Asa-
col which, having a Eudragit S coating, have passed
through the gut intact or incompletely disintegrated
ª 2008 The Authors, Aliment Pharmacol Ther 28, 911–916
Journal compilation ª 2008 Blackwell Publishing Ltd
in some patients.23, 24 It has been speculated that a
failure to disintegrate in these studies may have
been because of a lower intestinal pH in ulcerative
colitis patients. However, other studies have con-
firmed these results in healthy subjects using
Eudragit S as a tablet or granule coating.17, 18, 25
One such investigation by our group17 used the
same protocol, same healthy volunteers and same
tablet size as described in the present study. This
study using Eudragit S coated tablets showed that
out of 24 doses, only 17 disintegrated at the appro-
priate site, with the other seven tablets failing to
disintegrate.17 In contrast, in the present study we
describe no instances of the Eudragit S ⁄ starch coated
tablets passing through the gut intact. The success of
this system is attributable to the role of starch,
which adds an extra element of site-specificity. The
starch, which is not digestible by mammalian pan-
creatic amylase, is readily digestible by colonic
bacterial enzymes. Colonic bacterial enzymes (specifi-
cally amylases) are able to digest resistant starch
because they are much more efficient than mamma-
lian pancreatic enzymes.26 Thus, even if the Eudragit
S component of the film remains intact, the highly
active colonic bacterial enzymes will still digest the
starch component allowing dosage form disintegra-
tion. As over fifty percent of bacteria in the colon
produce amylase enzymes, they are a ubiquitous
presence and available for starch digestion.26 In this
system, the starch provides a back up or fail safe
for dosage form disintegration. Thus, in spite of low
fluid in the distal gut27, 28 and variable intestinal pH
values,9 combining the pH and bacteria-sensitive
approaches in a matrix film produces a successful
new delivery concept for colonic targeting.
CONCLUSIONS
Here, we report a novel dual-triggered colonic deliv-
ery system with improved site-specificity over the
pH-responsive systems currently used for ulcerative
colitis. The dual pH- and bacterially-triggered coating
was applied to tablets and dosed a total of 23 times
in healthy volunteers under three different feeding
conditions. On one occasion, the tablet did not empty
from the stomach during the study period, but of
those 22 tablets that emptied, disintegration occurred
at the ileo-caecal junction or in the large intestine
confirming successful targeting with this system.
The independent triggers of a bacterially-triggered
916 V . C . I B E K W E et al.

component within a pH-responsive polymer are effec-

tive, complementary and act as failsafe mechanisms
for each other. This platform technology could theo-
retically be adapted to any drug and used for a vari-
ety of disease states. The clinical potential for
inflammatory bowel disease is obvious with a low
risk of dose dumping and low risk of the tablets
passing intact and for new systemic applications for
colonic delivery.
ACKNOWLEDGEMENTS
Declaration of personal interests: A. W. Basit has
served as a consultant for Alizyme, and has received
research funding from Alizyme, Evonik, Teva and Til-
lotts. Declaration of funding interests: This study was
funded in full by The School of Pharmacy, University
of London. Writing support was provided by E. L.
McConnell of The School of Pharmacy.

REFERENCES
1 Frieri G, Pimpo MT, Andreoli A, et al.
Prevention of post-operative recurrence
of Crohn’s disease requires adequate
mucosal concentration of mesalazine. Ali-
ment Pharmacol Ther 1999; 13: 577–82.
2 Klotz U, Schwab M. Topical delivery of
therapeutic agents in the treatment of
inflammatory bowel disease. Adv Drug
Deliv Rev 2005; 57: 267–79.
3 Cohen RD. Review article: evolutionary
advances in the delivery of aminosalicy-
lates for the treatment of ulcerative coli-
tis. Aliment Pharmacol Ther 2006; 24:
465–74.
4 Minko T. Drug targeting to the colon with
lectins and neoglycoconjugates. Adv Drug
Deliv Rev 2004; 56: 491–509.
5 McConnell EL, Basit AW, Murdan S. Colo-
nic antigen administration induces signif-
icantly higher humoral levels of colonic
and vaginal IgA, and serum IgG com-
pared to oral administration. Vaccine
2008; 26: 639–46.
6 Mackay M, Phillips J, Hastewell J. Peptide
drug delivery: colonic and rectal absorp-
tion. Adv Drug Deliv Rev 2004 1997; 28:
253–73.
7 Basit AW. Advances in colonic drug
delivery. Drugs 2005; 65: 1991–2007.
8 Evans DF, Pye G, Bramley R, Clark AG,
Dyson TJ, Hardcastle JD. Measurement of
gastrointestinal pH profiles in normal
ambulant human subjects. Gut 1988; 29:
1035–41.
9 Fallingborg J, Christensen LA, Ingeman-
Nielsen M, Jacobsen BA, Abildgaard K,
Rasmussen HH. pH-profile and regional
transit times of the normal gut measured
by a radiotelemetry device. Aliment Phar-
macol Ther 1989; 3: 605–13.
10 Simon GL, Gorbach SL. The human intes-
tinal microflora. Dig Dis Sci 1986; 31(9
Suppl): 147S–62S.
11 Eckburg PB, Bik EM, Bernstein CN, et al.
Diversity of the human intestinal micro-
bial flora. Science 2005; 308: 1635–8.
12 Sousa T, Paterson R, Moore V, Carlsson A,
Abrahamsson B, Basit AW. The gastroin-
testinal microbiota as a site for the bio-
transformation of drugs. Int J Pharm 2008;
doi: 10.1016/j.ijpharm.2008.07.009.
13 Scheline RR. Metabolism of foreign com-
pounds by gastrointestinal micro-
organisms. Pharmacol Rev 1973; 25:
451–523.
14 McConnell EL, Murdan S, Basit AW. An
investigation into the digestion of chito-
san (noncrosslinked and crosslinked) by
human colonic bacteria. J Pharm Sci
2008; 97: 3820–9.
15 Cummings JH, Milojevic S, Harding M,
et al. In vivo studies of amylose- and
ethylcellulose-coated [13C] glucose micro-
spheres as a model for drug delivery to
the colon. J Control Rel 1996; 40: 123–
31.
16 Tuleu C, Basit AW, Waddington WA, Ell
PJ, Newton JM. Colonic delivery of 4-am-
inosalicylic acid using amylose-ethylcel-
lulose-coated hydroxypropylmethylcellu-
lose capsules. Aliment Pharmacol Ther
2002; 16: 1771–9.
17 Ibekwe VC, Fadda HM, McConnell EL,
Khela MK, Evans DF, Basit AW. Interplay
between intestinal pH, transit time and
feed status on the in vivo performance of
pH responsive ileo-colonic release sys-
tems. Pharm Res 2008; 25: 1828–35.
18 Ibekwe VC, Liu F, Fadda HM, et al. An
investigation into the in vivo performance
variability of pH responsive polymers for
ileo-colonic drug delivery using gamma
scintigraphy in humans. J Pharm Sci
2006; 95: 2760–6.
19 Podczeck F, Course NJ, Newton JM.
Determination of the gastric emptying
of solid dosage forms using gamma-
scintigraphy: a problem of image timing
and mathematical analysis. Eur J Nucl
Med 1999; 26: 373–8.
20 Code CF, Marlett JA. The interdigestive
myo-electric complex of the stomach and
small bowel of dogs. J Physiol 1975; 246:
289–309.
21 Baumgart DC, Sandborn WJ. Inflamma-
tory bowel disease: clinical aspects and
established and evolving therapies. Lancet
2007; 369: 1641–57.
22 Keller J, Layer P. Human pancreatic exo-
crine response to nutrients in health and
disease. Gut 2005; 54 (Suppl 6): vi1–28.
23 Safdi AV. Determination of mesalazine in
whole or partial mesalamine delayed-
release tablets recovered from fecal
samples of healthy volunteers. Am J
Gastroenterol 2005; 100: S159.
24 Sinha A, Ball BJ, Connor AL, Nightingale
J, Wilding IR. Intestinal performance of
two mesalamine formulations in patients
with active ulcerative colitis as assessed
by gamma scintigraphy. Pract Gastro-
enterol 2003; 27: 56–69.
25 McConnell EL, Short MD, Basit AW. An
in vivo comparison of intestinal pH and
bacteria as physiological trigger mecha-
nisms for colonic targeting in man.
J Control Rel 2008; 97: 3820–9.
26 Macfarlane GT, Englyst HN. Starch utili-
zation by the human large intestinal
microflora. J Appl Bacteriol 1986; 60:
195–201.
27 Gotch F, Nadell J, Edelman IS. Gastro-
intestinal water and electrolytes. IV The
equilibration of deuterium oxide (D2O) in
gastrointestinal contents and the propor-
tion of total body water (T.B.W) in the
gastrointestinal tract. J Clin Invest 1957;
36: 289–96.
28 Cummings JH, Banwell JG, Segal I,
Coleman N, Englyst HN, Macfarlane GT.
The amount and composition of large
bowel contents in man. Gastroenterology
1990; 98: A408.

ª 2008 The Authors, Aliment Pharmacol Ther 28, 911–916

Journal compilation ª 2008 Blackwell Publishing Ltd