a model for chronic quantitative studies of colorectal sensitivity using balloon distension in...

A model for chronic quantitative studies of colorectal

sensitivity using balloon distension in conscious

mice – effects of opioid receptor agonists

M. LARSSON,* S. ARVIDSSON,* C. EKMAN� & A. BAYATI*

*Department of Integrative Pharmacology, Research Area CV & GI, Preclinical R&D, AstraZeneca, Molndal, Sweden

�Department of Biostatistics, AstraZeneca, Molndal, Sweden

Abstract In the current study, colorectal distension

(CRD) was performed in conscious mice, in order to

study visceral (colon) sensitivity. Electrodes were

chronically implanted into the external oblique mus-

cle to obtain the electromyographic (EMG) response to

CRD. CRD was performed using a computerized sys-

tem, which inflated the balloon with air to the desired

pressures. An increasing (10–80 mmHg) and a repea-

ted (12 · 55 mmHg) phasic paradigm with distensions

lasting 10 s and with 5-min intervals were used. The

EMG recordings were linearly correlated to intraco-

lonic pressures between 10 and 80 mmHg, which are

characteristic of the visceromotor response (VMR).

Repeated phasic distensions at 55 mmHg resulted in a

stable VMR in female mice, but an increasing VMR in

male mice. Interestingly, the duration of the VMR was

about 5 s, which is shorter than the actual duration of

the distension. U-69593 and fentanyl (selective j and

l opioid receptor agonists) significantly reduced the

VMR at subcutaneous doses of 0.5 and 0.05 mg kg)1,

respectively. In conclusion, a CRD model for repetitive

quantitative studies of colorectal sensitivity and

evaluation of pharmacological modulation of visceral

sensitivity in conscious mice is presented.

Keywords colon, colorectal distension, mouse, opioid

receptor agonist, visceral sensitivity.

INTRODUCTION

Irritable bowel syndrome (IBS) is one of the most

common functional gastrointestinal disorders seen in

primary and specialists care. The symptoms of IBS

include lower abdominal pain or discomfort, disturbed

defecation (diarrhoea and/or constipation) and bloat-

ing. These symptoms occur in the absence of struc-

tural (e.g. inflammation), biochemical (e.g. lactase

deficiency) or pathophysiological abnormalities that

might otherwise explain these symptoms.1 IBS can

thus be characterized as a chronic gastrointestinal

dysfunction, reflected by altered motility and/or vis-

ceral pain.2

Visceral pain, unlike somatic pain, is usually poorly

localized and in many cases it is difficult to pinpoint

the specific organ that is the cause of pain. In man, the

distension of hollow organs like the colon can easily be

studied, as the referred pain caused by distension can

be monitored verbally by the person’s subjective

sensation. Indeed, conscious hypersensitivity to recto-

sigmoid distensions in IBS patients was already repor-

ted by Ritchie3 and these findings have been confirmed

in a number of studies.4,5 In conscious animals, on the

other hand, colorectal distension (CRD) results in a

series of stereotypic behavioural and autonomic

responses, including passive avoidance, increase in

arterial pressure and heart rate and a visceromotor

response (VMR) in the form of contractions of the

abdominal wall muscles.6–8 The pseudo-affective

responses to CRD have been studied thoroughly in

the rat by many investigators6,8 and have been used

extensively in studies of the mechanisms of visceral

sensitivity.9–12

In order to further elucidate the possible mecha-

nisms involved in visceral hypersensitivity, the use of

mice would offer a potential advantage, for transgenic

mice could be used to establish correlations between

Address for correspondence

Marie Larsson, AstraZeneca Molndal, Department ofIntegrative Pharmacology, Research Area CV & GI,Gastrointestinal Biology, S-431 83 Molndal, Sweden.Tel.: +46 31 776 1531; fax: +46 31 776 3747;e-mail: [email protected]: 17 March 2003Accepted for publication: 10 April 2003

Neurogastroenterol Motil (2003) 15, 371–381

� 2003 Blackwell Publishing Ltd 371

distinct genetic modifications and visceral sensitivity.

Until now, studies of pain mechanisms in mice have

largely been limited to somatic pain and to tonic

visceral pain (i.e. writhing test).13–16 However, the

writhing test is a mixed somatic and visceral pain

model with limited clinical relevance. There are also

ethical concerns about such models, as the noxious

stimulus is long lasting and inescapable in conscious

animals. In contrast, CRD reproduces a natural visceral

stimulus and the onset, magnitude and duration of

CRD can easily be controlled. It would thus be of great

benefit to develop a CRD model for mice.

The aim of the present study was, therefore, to

develop a sensitive CRD model that enables elec-

tromyographic (EMG) registration of the external

abdominal oblique muscle activity and to characterize

the physiologic response to CRD in conscious mice.

In addition, the effects of two opioid receptor agonists

(l and j) were tested in order to study if the model is

useful for evaluation of analgesic effects on the VMR to

CRD.

The present study was first reported in abstract form

at the 9th world congress of pain in Vienna 1999.17

Another study using CRD in j-receptor knockout mice

has also been reported in abstract form at the DDW

congress in San Diego 2000.18

MATERIALS AND METHODS

Animals

A total of 53 female and 20 male mice (C57bl/6J M&B,

Denmark) weighing 22–28 g were used in the studies.

The animals were placed in plastic cages, male mice

alone and female mice up to four animals per cage, and

had free access to water and mice chow (R3, Lactamin

AB, Sweden). The cages were placed in a temperature-

controlled environment (19–23 �C) with humidity of

25–70% and a light/dark cycle of 12 h. All the experi-

ments were approved by the local ethics review

committee on animal experiments in Goteborg,

Sweden.

Surgical preparation

Mice were anaesthetized with a mixture of Dormicum

(midazolam 5 mg mL)1, Roche, Stockholm, Sweden),

Hypnorm (fentanyl citrate 0.315 mg mL)1 and fluani-

sone 10 mg mL)1, Janssen Animal Health, Beerse,

Belgium) and sterile water (1 : 1 : 2) (10 mg kg)1 i.p.).

A round-shaped fistula (Plexiglas, made in-house) with

an outer diameter of 4.0 mm and an oval-shaped

platform for suturing was chronically implanted in

the animals. The fistula, able to contain six connec-

tors, was exteriorized from the peritoneal cavity

through the right abdominal muscle, about 1.5 cm

lateral from the midline incision and between the ribs

and the hind limb, and sutured to the muscle (Ti crom

6.0). The six connectors of the fistula allow the

animals to be fitted with a maximum of two bipolar

and one monopolar bioelectric electrodes (Bioflex

insulated wire AS631, Cooner Wire, USA). The mono-

polar electrode was used as ground and the two bipolar

electrodes can be used to record EMG and ECG

signals. However, in the current study we have only

recorded the EMG signals. Approximately 1 cm of

each of the EMG electrodes was implanted from the

peritoneal side in the left external abdominal oblique

muscle about 2 mm apart and about 2 cm lateral to

the midline incision. Experiments started at the

earliest 7 days after surgery. The animals tolerated

the fistula and the electrodes well and could be used in

different experiments twice a week, for up to

6 months.

CRD device and EMG recordings

The CRD-system is composed of an amplifier (devel-

oped at AstraZeneca R&D Molndal) and a barostat

system. The barostat system, designed to control four

animals simultaneously, is composed of a pressure

control device (Pressure meters and controllers, P-602

CFM-k33, 100 mmHg, Bronkhorst HI-TEC, The Neth-

erlands) and four flow meters (Mass flow meters GFM

171-03, Aalborg Instruments and Controls, INC, New

York, USA). The system relies on the airflow from a

high-pressure reservoir (74 psi, 3400 mmHg), allowing

a pressure increase from 0 to 80 mmHg in 0.2 s. The

barostat, which regulates the inflations of the bal-

loons, is controlled by a computer program (Labview,

National instruments), in which different pressure

paradigms can be created and the pressure in the

balloons can be controlled and kept constant at a

predefined level. Another computer program, with a

data-sampling rate of 200 Hz (Labview, National

instruments) is used to continuously monitor online

and record the EMG activity (amplified 5000· or

10 000· and filtered with a 3-Hz high-pass filter and a

1000-Hz low-pass filter) in the external abdominal

oblique muscle and the balloon pressure for subse-

quent analyses. The pressure measured during disten-

sion reflects intracolonic pressure, although it is

measured with a pressure transducer outside the

animal, as the diameter of the balloon, when inflated,

is greater than the intraluminal diameter of the mouse

colon.

372 � 2003 Blackwell Publishing Ltd

M. Larsson et al. Neurogastroenterology and Motility

Balloons

The balloons were made by pulling a plastic sheet

(polythene) with a thickness of about 25 lm over a

Teflon cylinder with a diameter of 10 mm. This

resulted in a plastic cylinder shaped �balloon� with a

length of 10, 20 or 30 mm, a thickness of 15 lm and a

maximum, non-distensible diameter of 10 mm. The

balloon was fastened (5.0 silk thread) to a Teflon

catheter (o.d. 1.06 mm and i.d. 0.56 mm) with a total

length of about 30 cm. The catheter extended almost

into the bottom of the balloon.

To test if a balloon diameter of 10 mm was wide

enough, in order not to limit the degree of colonic

distension by itself, the balloon was inserted into the

colon of anaesthetized female (C57bl/6J) mice. The

abdomen was opened, the colon localized and the outer

diameter of the colon was measured in situ at different

balloon pressures. A maximal intracolonic pressure of

80 mmHg increased the outer diameter of the colon to

about 5 mm.

Experimental procedure

Before each experiment a balloon (described above) was

inserted into the colon under inhalation of isoflurane

anaesthesia (Forene, Abbott Scandinavia AB, Sweden)

and the catheter was fixed to the base of the tail with

tape to prevent displacement. A cable was connected to

the fistula and the mouse was placed in a specially

designed Bollmann cage (made of a plastic cylinder with

an i.d. of 2.6 cm and a length of 10.0 cm to fit the size of

the animal). The mice had been trained at least three

times before the first experiment to tolerate the Bollman

cage. In the experiments where different compounds

were administered, a polyethylene catheter (PE25) was

placed under the skin behind the neck of the animal for

subcutaneous administration. Connecting the cable

from the fistula to the amplifier and the catheter from

the balloon to the barostat system connected the mice to

the CRD device. The mice (usually four animals at a

time) recovered from the anaesthesia and experiments

started approximately 15 min after the last mouse was

placed in the Bollmann cage.

After experiments, the balloon and the connecting

cable were removed under isoflurane anaesthesia and

the animals returned to their normal cage. Experi-

ments were performed no more than twice a week with

each animal and with at least 2 days of rest between

experiments.

Increasing phasic distensions A first series of experi-

ments were performed in female mice, using a 20-mm-

long balloon positioned 5 mm proximal to rectum, to

evaluate the VMR, the linearity between intracolonic

pressure and VMR, the nociceptive threshold, and the

pressure giving a distinct VMR while avoiding unnec-

essary nociceptive stimulation. Ten-second disten-

sions were performed in triplicate at pressures of

10, 25, 40, 65, and 80 mmHg with 5-min intervals

(Fig. 1A).

Balloon size and position In another series of experi-

ments, the VMR to CRD with different balloon lengths

and positions were evaluated in female mice, using the

increasing phasic paradigm described above. Balloons

with sizes of 10, 20 and 30 mm were tested in different

positions. The 10-mm balloons were placed in the

10 s 5 min

Average 1 Average 4Average 3Average 2

10 s

10

25

1010

25

40

25

656565

40 40

8080 80

5 min

mmHgA. Increasing phasic 10–80 mmHg

B. Isobaric phasic 12 × 55 mmHg

1 1274 983 652 10 11

Figure 1 (A) An increasing phasic disten-sion paradigm starting at 10 mmHg andending at 80 mmHg with 10-s distensionsand 5-min intervals between each disten-sion. Each pressure is repeated three timesto acquire a more accurate value for thecorresponding VMR. (B) An isobaric phasicdistension paradigm at 55 mmHg. Thedistensions are divided in four groups,which enable the administration of threecumulative doses of a compound, i.e. afterdistensions 1–3, 4–6 and 7–9, with dis-tensions 1–3 serving as an internalcontrol.


Volume 15, Number 4, August 2003 Colorectal sensitivity in mice

colon 5 or 25 mm proximal to the rectum, the 20-mm

balloons 5 or 15 mm proximal to the rectum and the

30-mm balloon 5 mm proximal to the rectum (see

Fig. 2).

Isobaric phasic distensions The stability and possi-

bility of sensitization (increase) or adaptation (decrease)

in the VMR to repetitive isobaric distensions was

studied using a repeated phasic paradigm (12 ·55 mmHg) with the 20-mm-long balloon positioned

5 mm proximal to the rectum (Fig. 1B). Each distension

lasted for 10 s and was followed by a 5-min period of

deflation. When analysing the data the 12 distensions

were divided into four groups, distensions 1–3, 4–6, 7–9

and 10–12, respectively.

Comparison between female and male mice In

another series of experiments, the VMR to CRD in

male mice was studied and compared to the VMR in

female mice, using the 20-mm-long balloon placed

5 mm proximal to the rectum. Both the increasing

phasic (10–80 mmHg) and the isobaric phasic

(12 · 55 mmHg) paradigms (described above) were

used. To evaluate if the VMR is stable in repeated sets

of experiments, the increasing phasic paradigm was

repeated in new batches of male and female mice

approximately 6–8 months later.

Compounds

Two opioid receptor agonists, U-69593 [j-opioid recep-

tor agonist (RBI, USA)] and fentanyl [l-opioid receptor

agonist (Leptanal 50 lg mL)1, Janssen Pharmaceutical,

Bersse, Belgium)], were tested to study if the model

could be useful for evaluation of drug effects on the

VMR to CRD. The repeated phasic paradigm

(12 · 55 mmHg) was used to evaluate the effect of

the compounds (with the first three distensions serving

as control). The compounds were administered once

per experiment subcutaneously through a catheter

behind the neck at a volume of 5 mL kg)1, 1 min after

the third distension. Five different doses, 0.2, 0.5,

1.0, 5.0 and 25.0 mg kg)1, of U-69593 [dissolved in

2-hydroxypropyl-b-cyclodextrin (RBI, USA)] and four

different doses, 0.005, 0.025, 0.05 and 0.25 mg kg)1, of

fentanyl (diluted in saline) were used.

Plasma concentrations of U-69593 and fentanyl

Plasma concentrations of the opioid receptor agonists

were determined in separate, non-operated female mice

(C57BL/6). Following subcutaneous administration of

U-69593 (1 or 25 mg kg)1 bw) or fentanyl (0.3 mg kg)1

bw), blood was collected at time points 0, 5, 15, 30, 60

and 120 min after administration (n ¼ 2). The blood

was collected by heart puncture after the animals had

been sacrificed. The plasma concentrations of U-69593

and fentanyl were analysed by liquid chromatography-

mass spectrometry (LC-MS). After solid-phase extrac-

tion on Bond Elut C8, 50 mg (Varian, USA), separation

was performed on Zorbax SB-C8, 3.5 lm, 75 · 4.6 mm

(Rockland Techn. Inc., USA) with a mobile phase

consisting of 38% acetonitrile and 0.1% formic acid in

ammonium acetate (2 mmol L)1) and with a flow rate of

0.75 mL min)1. The compounds were detected by elec-

trospray positive ionization mass spectrometry. The

limit of quantification for U-69593 and fentanyl was 25

and 10 nmol L)1, respectively, using 100 lL plasma.

Data analysis

The extracted EMG raw data were analysed in three

steps using specially designed software (Labview,

National instruments) as follows:

1. When all the 10-s distension pulses in the experi-

ment had been detected by the program, the EMG data

during 10 s prior to and during each distension were

extracted from the original data file.

2. A data reduction was performed on the extracted

raw EMG data as follows: the data were divided into 1-s

periods. For each period (200 sampling points) the mode

was calculated (the most frequent or repetitive value in

the data set) and used as baseline (zero level). From the

EMG data, all the peaks were detected using the

baseline as threshold. The average of the peak ampli-

tudes was calculated for each period. To stabilize the

variance of the population, the data were log trans-

formed, average peak amplitude was calculated for the

chosen period (0–10 s or 0–5 s) and the data were anti-

logged. The resting EMG (basal activity) was calculated

Caecum

Colon

5 mm

15 mm

25 mm

35 mm

Balloon

Figure 2 A schematic illustration of the lengths and positionsof the balloon used when determining optimal CRDparameters.



as the area under the curve (AUC) of the average peak

amplitudes of the data 10 s before each distension. The

VMR was calculated as the AUC of the average peak

amplitudes of the data over the first 0–5 s following the

start of distension (the reason for this is explained in

Results).

3. In all the analyses the mean value of the data was

calculated for three consecutive distensions, i.e. dis-

tensions 1–3, 4–6, 7–9 and 10–12 in the isobaric phasic

paradigm and for each pressure in the increasing phasic

paradigm. The reason for this is to minimize the

influence of eventual movements by the mice in the

Bollman cage, which would give an inaccurate baseline

EMG activity and VMR.

When using the isobaric phasic distension paradigm

(12 · 55 mmHg) an average of the relative values for

pulses 1–3 had to be at least 1.6 to be considered as a

response. If the relative value was below 1.6 the mouse

was excluded from the experiment. This level was

determined by calculating the average of the relative

values for pulses 1–3 for a number of experiments, and

taking two standard deviations as the limit.

In most of the analyses, the average of the VMR

(AUC/s) was divided by the average of the basal

activity (AUC/s) to obtain a relative VMR. (If the ratio

of the VMR to basal activity equals 1, then there is no

response to distension.)

For drug effects the VMR is presented as %VMR

remaining after dose, in which the average of the

relative VMR of pulses 1–3 was defined as 100%. This

normalization of data allows comparison of drug effects

between sexes.

Statistics

The statistical analysis was performed on a log-trans-

formed scale assuming normal distributed random

effects. Depending on the nature of the problem, a

repeated measures ANOVA or ANCOVA model has been

adopted. All comparisons were made within the

framework of the postulated model. A significance

level of 5% was used and no multiplicity adjustment of

P-values has been carried out. All values are expressed

as the mean ± SEM.

RESULTS

EMG response profiles

Figure 3 shows a typical EMG tracing of the abdom-

inal oblique muscle during a 10-s distension of the

mouse colon at a pressure of 55 mmHg. There was a

prompt EMG response (reflecting both an increase in

amplitude and frequency) following the increase in

colonic pressure. However, unlike the maintained

response normally seen in rats, it faded away and

returned to basal levels approximately 5 s after the

start of distension, despite the fact that the balloon

pressure was kept constant during the whole disten-

sion period. This transient EMG response occurred

even when using longer tonic distensions (up to

5 min) at pressures of 25, 40 or 60 mmHg (data not

shown). Abdominal contractions could not be detec-

ted visually, neither during phasic nor during tonic

distension periods.

10 s

Bal

loon

pre

ssur

e (m

mH

g)E

MG

(V

* 1

0000

)

–10

0

10

20

30

40

50

60–6

–4

–2

0

2

4

6

Figure 3 A representative EMG of theabdominal oblique muscle during a 10-sphasic distension of the mouse colon at apressure of 55 mmHg. There was a promptincrease in EMG activity in response tothe increase in pressure (i.e. a VMR).However, unlike the response commonlyseen in the rat, the EMG response fadedaway approximately 4–5 s after the start ofdistension, despite the fact that the bal-loon pressure was kept constant through-out the distension period.



The mechanism behind this attenuation phenom-

enon in response to CRD is currently unknown. Due to

this phenomenon, only the first 5 s of the distension

period were used in the following analysis of the EMG

data.

Effect of balloon length and position on the EMGresponse

The results from experiments with 10-, 20- and 30-mm

balloons revealed that the position of the balloon in

the colon is more important than the size of the

balloon. When the 10- and 20-mm balloons were

positioned in the proximal part of the colon, the

VMR was significantly reduced (P < 0.001) compared

to when each balloon was placed in the distal part of

the colon (Fig. 4). When the balloons were positioned

in the distal part of the colon (5 mm proximal to

rectum), increasing the balloon size from 10 to 20 mm

did not significantly (P ¼ 0.09) increase the slope of

the VMR. However, the 20-mm balloon seemed to give

a more linear VMR than the 10-mm balloon. When the

balloon was increased to 30 mm in length and posi-

tioned in the distal part of the colon, the VMR was not

significantly different at lower pressure levels com-

pared to the VMR using the 20-mm balloon. However,

the VMR was only linear up to a pressure of 65 mmHg

using the 30-mm balloon.

There was a significant (P < 0.05) increase in the

VMR at a pressure of 25 mmHg when placing a 10-, 20-

or 30-mm-long balloon in the distal colon. However,

when the balloons were placed in the proximal colon,

higher pressures were required to induce a response.

Figure 4 also shows that insertion of the balloon into

Figure 4 Effects of various balloon lengthsand positions. The first two numbers inthe textbox are the balloon dimensions(length and width in mm), whereas thethird number represents the position(mm proximal from rectum) of theballoon. All pressure levels eliciting asignificant increase in the VMR abovebaseline activity are indicated. Thresholdlevels are defined as the first levelseliciting a significant increase in the VMRabove baseline activity (n ¼ 5–18).Mean ± SEM; **P < 0.01, *P < 0.05.

0

100

200

300

400

500

600

700

1 2 3 4 5 6 7 8 9 10Time (s)

Cha

nge

in E

MG

act

ivity

com

pare

d to

bas

elin

e (%

)

10 mmHg

25 mmHg

40 mmHg

65 mmHg

80 mmHg

n = 10

Figure 5 Effect of different intracolonicpressures on VMR kinetics in femalemice. The response of the abdominaloblique muscle to different distensionpressures was studied using the increasingphasic paradigm (10–80 mmHg) with a20-mm-long and 10-mm-wide balloon.The percentage change in EMG activitywas calculated by dividing distension-induced EMG activity for each second(during each pressure) with basal EMGactivity (n ¼ 10). Mean ± SEM.



the rectum without connecting it to the barostat

(or balloon insertion itself, data not shown) does not

result in a VMR. The 20-mm balloon positioned 5 mm

inside the rectum was used for further experiments as it

induced the highest VMR, had a low threshold level and

had the best linear correlation to increased pressure.

Pressure response profiles

Distending the colon with an increasing phasic para-

digm [3 · (10, 25, 40, 65 and 80 mmHg)] demonstrated

that the VMR is proportional to the pressure applied,

starting at pressures above 10 mmHg. When calcula-

ting the VMR (percent increase in EMG activity during

the distension compared to the mean basal activity

during 10 s before the distension) for each second of the

distension it is evident that there is a prompt VMR,

lasting about 4 s for pressures above 10 mmHg (Fig. 5).

After the first 4 s only a small increase in the VMR

remains.

Analysing the mean value during the first 5 s of the

VMR for all animals at each pressure level shows that

the VMR is linearly correlated, in both female and

male mice, to the pressure applied (Fig. 6). The linear

regression analysis of these data resulted in a slope of

0.04 (female) and a slope of 0.12 (male), which are

significantly (P < 0.001) different from a line with a

slope of zero. The slope of the regression line for male

mice is significantly higher than the slope of the

regression line for female mice (P < 0.001), whereas the

VMR at 10 mmHg of male and female mice are not

different from each other (P ¼ 0.25), indicating that

C57Bl/6j male mice are more sensitive to CRD at

intracolonic pressures above 10 mmHg than C57Bl/6j

female mice. When the same experiment was repeated

in a new series of mice approximately 6 and 8 months

later (male and female, respectively) it was shown that

the VMR to increasing phasic distensions is consistent

between different series of experiments (Fig. 6).

Isobaric, phasic distensions

Repeated isobaric distensions at 55 mmHg indicated

that the VMR is relatively stable over all the 12

distensions in female mice, whereas male mice had an

increasing VMR (Fig. 7). At the last distension pulses

(10–12), the VMR for male mice had increased by 73%

compared to the VMR recorded during pulses 1–3,

whereas female mice only had a 30% increase. The

slope of a regression line of VMR (increase in VMR/

distension group) for female and male mice was

calculated to be 0.07 and 1.13, respectively. The slope

of the regression line for female mice is not signifi-

Male

0

2

4

6

8

10

12

VM

R

Series 1 (n = 7) Series 2 (n = 7)Series 1 (n = 10) Series 2 (n = 18)

Female

10 25 40 65 80

Balloon pressure (mmHg)

0

2

4

6

8

10

12

10 25 40 65 80

Balloon pressure (mmHg)

VM

R

Figure 6 Relation between intracolonicpressure and VMR in female and malemice using the increasing phasic paradigm(10–80 mmHg) with a 20-mm-long and10-mm-wide balloon. In this paradigmeach pressure was repeated three timesand the mean value of the distension-induced EMG activity was calculatedusing the first 5 s. This was then dividedby the mean basal activity and a relativeVMR was obtained. The same experimentswere repeated 6 and 8 months later(female and male, respectively), series 2.Mean ± SEM.

0

1

2

3

4

5

6

7

8

9

1–3 4–6 7–9 10–12

Distension number

VM

R

Female (n = 12)

Male (n = 7)

Figure 7 Relation between isobaric phasic distensions(12 · 55 mmHg) and VMR in female and male mice using a20-mm-long and 10-mm-wide balloon. The 12 distensionswere divided into four groups, distensions 1–3, 4–6, 7–9 and10–12, respectively. A mean value of the distension-inducedEMG activity was calculated for each group using the first 5 s.This was then divided by the mean basal activity and a rel-ative VMR was obtained. Mean ± SEM.



cantly different from a line with a slope of zero

(P ¼ 0.43), whereas the slope of a regression line of

VMR for male mice is significantly different from a

line with a slope of zero (P < 0.001).

Effects of opioid receptor agonists on VMR

Both fentanyl and U-69593 reached their peak concen-

trations 15 min after subcutaneous administration and

had half-lives of 37 min. Fentanyl significantly reduced

the VMR by 80–100% to noxious CRD at doses of 0.05

and 0.25 mg kg)1 in both female and male mice

compared to control (P < 0.001) (Fig. 8). The VMR in

male mice was also inhibited (by 75%) at

0.025 mg kg)1, a dose having little effect in females.

U-69593 also decreased the VMR in both female and

male mice (Fig. 9). Significant effects (P < 0.01) were

seen already at the dose of 0.5 mg kg)1 in both female

and male mice. In contrast to fentanyl, maximal

inhibitory effects were about 60%, with the exception

of 25 mg kg)1 in male mice where 90% inhibitory

effects were seen. The effects of the compounds were

essentially sustained throughout the experiments.

DISCUSSION

The present study demonstrates that the VMR to CRD

in mice can be detected by monitoring the EMG of the

**

**

A

Control 0.005 0.025 0.05 0.25n = 7 n = 7 n = 7n = 8 n = 9

**

** **

Effect of fentanyl in mice

0

50

100

150

200

250

300

Control 0.005 0.025 0.05 0.25

Dose fentanyl (mg kg–1)

VM

R r

emai

ning

afte

r do

se (

% )

n = 11 n = 14 n = 16n = 6 n = 7

**

**

Effect of fentanyl in female mice

Dose fentanyl (mg kg–1)

B

**

** **

male

0

50

100

150

200

250

300

Figure 8 Effect of fentanyl on the VMR during repeated colonic distensions at 55 mmHg in female (A) and male (B) mice using a20-mm-long and 10-mm-wide balloon. Fentanyl (5 mL kg)1) was administered subcutaneously about 1 min after the thirdinflation. The 12 distensions were divided into two groups, before (distensions 1–3) and after drug administration. The average VMRof distensions 1–3 was set to 100%. Non-treated male mice (B) have increased responses in comparison to distensions 1–3 (forcalculations, see Materials and Methods section). Mean ± SEM; **P < 0.001.

Control 0.2 0.5 1 5

Dose U-69593 (mg kg–1)

n = 11 n = 3 n = 11n = 15n = 12

A

***

*

Effect of U-69593 in female mice

Control 0.5 1 5 25

Dose U-69593 (mg kg–1)

n = 7 n = 9 n = 5 n = 5n = 9

**

**

**

**

B Effect of U-69593 in male mice

0

50

100

150

200

250

300

VM

R r

emai

ning

afte

r do

se (

% )

***

***

**

**

**

0

50

100

150

200

250

300

Figure 9 Effect of U-69593 on the VMR during repeated colonic distensions at 55 mmHg in female (A) and male (B) mice using a20-mm-long and 10-mm-wide balloon. U-69593 (5 mL kg)1) was administered subcutaneously about 1 min after the third inflation.The 20 distensions were divided into two groups, before (distensions 1–3) and after drug administration. The average VMR ofdistensions 1–3 was set to 100%. Non-treated male mice (B) have increased responses in comparison to distensions 1–3 (forcalculations, see Materials and Methods section). Mean ± SEM; **P < 0.001, *P < 0.01.



external oblique muscle. The present study also estab-

lishes that the VMR is a reproducible physiologic

response in conscious mice and hence can be used to

study acute visceral nociception from the colon. This

is in contrast to current models of visceral pain in

mice, such as the writhing test, which has significant

limitations as it suffers from unspecificity and lacks

reproducibility. Indeed, in one study, it was shown that

the typical abdominal contractions seen in the wri-

thing tests were inhibited both by substances consid-

ered to be analgesic and substances considered to be

non-analgesic.19 On the other hand, CRD reproduces

natural visceral stimuli, which is reproducible and

easily controlled in studies of colon sensitivity. It is

also related to human pathology as it has been shown

in several studies that patients suffering from IBS

exhibit a lower sensory threshold to CRD compared to

healthy subjects.3,4,20

CRD models for studies of visceral pain in rats are

well established6,7 and have been extensively used to

explore possible mechanisms causing visceral hyper-

sensitivity.10–12,21 The CRD model described in the

current study could potentially be of great benefit as it

creates the possibility for transgenic mice to be used

in studies of particular mechanisms or receptors

involved in visceral hypersensitivity. In rats, the

distension of the colon results in a series of beha-

vioural and pseudo-affective responses such as passive

avoidance, increase in arterial blood pressure and

heart rate and a VMR in the form of contractions of

the abdominal wall muscles.6,7,22

There have been attempts made to count the number

of abdominal contractions visually during CRD in

mice.23 However, the characteristic abdominal con-

tractions in response to CRD seen in rats were absent.

Indeed, we were not able to detect any abdominal

contractions visually in the current study. This might

be due to the small size of mice (contractions are not as

visible as in rats) or to the sensitivity of the model

(EMG recording is required to detect changes in muscle

activity).

The present study, on the other hand, demonstrates

that the VMR in response to CRD in mice can be

detected by monitoring the EMG of the external

oblique muscle, although the duration of the VMR

never lasted longer than about 5 s despite the

distension time (10 s or 5 min). The reason for this

phenomenon is currently unknown. One possible

explanation could be activation of endogenous pain

inhibitory systems that reduce pain perception in

connection to stressful situations and painful stimuli,

the so-called stress-induced analgesia.24 Another poss-

ible explanation could be that pathways involving the

periaqueductal grey (PAG)-mediated analgesia system

are activated.25–27 Due to the use of constant intraco-

lonic pressure, the only conclusion that can be drawn

is that the attenuation phenomenon does not depend

on relaxation of the colon.

As the VMR did not exceed 5 s, the time to increase

the pressure from 0 to 80 mmHg had to be reduced as

much as possible. Using the small balloons and

catheters necessary for mice, this cannot be accom-

plished with existing conventional balloon distension

techniques, such as utilizing a cylinder pump or

available distension control devices. In order to achieve

inflation of the balloon at an adequate speed, a new

pressure control system (barostat) based on the flow

from a very high pressure reservoir (74 psi, 3400

mmHg) forcing air into the balloon was developed,

instead of using a system where the balloon will attain

the same pressure as in the reservoir.28 To further

decrease the inflation time, a material (Teflon) that has

a low resistance to passage of air, was chosen for the

balloon catheter. Consequently, we were able to reduce

the time to increase the pressure from 0–80 mmHg to

around 200 ms, which was considered satisfactory for

our needs.

One important aspect of a CRD model is the

linearity of the VMR related to different intracolonic

pressures in the rat, as has been demonstrated previ-

ously by many authors.6,7,22 In the current model, it

was shown that the VMR is linearly correlated to the

intracolonic pressure at a pressure range between 10

and 80 mmHg in both female and male mice, although

male mice had a higher VMR at all the distension

pressures used above 10 mmHg. By repeating the same

experiments 6–8 months later in a new series of

animals, it was also shown that the results are

reproducible. Using the isobaric phasic distension

paradigm (12 · 55 mmHg), male mice were shown to

be mechanically sensitized (as well as having a higher

VMR than female mice) to repetitive distensions,

whereas female mice had a stable response throughout

the 12 distensions. Male C57Bl/6J mice thus appear to

be more sensitive to CRD than female C57Bl/6J mice.

A recently published study using other strains of mice

has confirmed the validity and usefulness of the

model.29 However, in this study, female 129S6 mice

were more sensitive to CRD than male 129S6 mice.

Hence, strain-dependent differences seem to exist

between female and male mice.

Treating the animals with either fentanyl or U-69593

in the isobaric phasic paradigm (12 · 55 mmHg) resul-

ted in a significant decrease of the VMR in both

female and male mice. The effect of fentanyl was

dose-dependent, which is in good agreement with



observations in rats.9 In contrast, the effect of U-69593

was not clearly dose-dependent at the doses tested, espe-

cially in male mice. This is in conflict with observations

in rats30 and mice.29 The reason for this is unknown.

The analgesic effect of fentanyl is probably mediated

by a central mechanism, whereas U-69593 most likely

uses both a peripheral and central site of action,

depending on the concentration administered9,21,31,32

(unpublished in-house results).

The current paper describes the development of a

technique using CRD and measurements of the VMR in

conscious mice. Furthermore, the attenuation of the

VMR obtained with both l- and j-opioid receptor

agonists demonstrates the usefulness of the model for

pharmacological studies. The study also presents evi-

dence that controlled distension of the descending colon

in mice is an experimental useful visceral stimulus,

being both reliable and reproducible. These findings will

make it possible to use conscious, transgenic mice in

future studies of mechanisms involved in visceral

sensitivity.

ACKNOWLEDGMENTS

The authors are grateful to the Department of Phar-

macokinetics & Drug Metabolism at AstraZeneca and

especially to Marie Strimfors and Marie Heijer for

analysing the blood concentrations of the test sub-

stances. The Preclinical Technical Support Group at

AstraZeneca is also gratefully acknowledged. Thanks

are also due to Hakan Larsson and Erik Lindstrom for

critically reading the manuscript.

REFERENCES

1 Drossman DA. An integrated approach to the irritablebowel syndrome. Aliment Pharmacol ther 1999; 13: 3–14.

2 Schmulson MJ, Mayer EA. Evolving concepts in irritablebowel syndrome. Gastroenterology 1999; 15: 16–21.

3 Ritchie J. Pain from distension of the pelvic colon byinflating a balloon in the irritable colon syndrome. Gut

1973; 14: 125–32.4 Mayer EA, Gebhart GF. Basic and clinical aspects of vis-

ceral hyperalgesia. Gastroenterology 1994; 107: 271–93.5 Mertz H, Naliboff B, Munakata J et al. Altered rectal per-

ception is a biological marker of patients with irritablebowel syndrome. Gastroenterology 1995; 109: 40–52.

6 Ness TJ, Gebhart GF. Colorectal distension as a noxiousvisceral stimulus: physiologic and pharmacologic charac-terization of pseudaffective reflexes in the rat. Brain Res

1988; 450: 153–69.7 Ness TJ, Randich A, Gebhart GF. Further behavioral evi-

dence that colorectal distension is a �noxious� visceralstimulus in rats. Neurosci Lett 1991; 131: 113–16.

8 Gebhart GF, Ness TJ. Central mechanisms of visceralpain. Can J Physiol Pharmacol 1991; 69: 627–34.

9 Danzebrink RM, Green SA, Gebhart GF. Spinal mu anddelta, but not kappa, opioid-receptor agonists attenuateresponses to noxious colorectal distension in the rat. Pain

1995; 63: 39–47.10 Plourde V, St-Pierre S, Quirion R. Calcitonin gene-related

peptide in viscerosensitive response to colorectal disten-sion in rats. Am J Physiol 1997; 273: G191–G196.

11 Julia V, Morteau O, Bueno L. Involvement of neurokinin 1and 2 receptors in viscerosensitive response to rectal dis-tension in rats. Gastroenterology 1994; 107: 94–102.

12 Morteau O, Julia V, Eeckhout C et al. Influence of 5-HT3receptor antagonists in visceromotor and nociceptiveresponses to rectal distension before and during experimen-talcolitisinrats.FundamClinPharmacol1994;8:553–62.

13 Brown DM, Hughes BO. Practical aspects of strain vari-ation in relation to pharmacological testing. J PharmPharmacol 1962; 14: 399–405.

14 Elmer GI, Pieper JO, Negus SS et al. Genetic variance innociception and its relationship to the potency of mor-phine-induced analgesia in thermal and chemical tests.Pain 1998; 75: 129–40.

15 Mogil JS, Kest B, Sadowski B et al. Differential geneticmediation of sensitivity to morphine in genetic models ofopiate antinociception: influence of nociceptive assay.J Pharmacol Exp ther 1996; 276: 532–44.

16 Simonin F, Valverde O, Smadja C et al. Disruption of thekappa-opioid receptor gene in mice enhances sensitivity tochemical visceral pain, impairs pharmacological actions ofthe selective kappa-agonist U-50,488H and attenuatesmorphine withdrawal. EMBO J 1998; 17: 886–97.

17 Abrahamson MH, Bayati A, Ekman C. A new model forquantitative studies of colorectal sensitivity using balloondistension in mice. In: 9th World Congress on Pain.Vienna: IASP Press, 1999.

18 Abrahamson MH, Bayati A. Involvement of kappa-opioidreceptors in visceral nociception in mice. Gastroenterol-

ogy 2000; 118: A379.19 Hendershot LC, Forsaith J. Antagonism of the frequency of

phenylquinone-induced writhing in the mouse by weakanalgesics and nonanalgesics. J Pharmacol Exp ther 1959;125: 237–40.

20 Schmulson M, Chang L, Naliboff B et al. Correlation ofsymptom criteria with perception thresholds during rec-tosigmoid distension in irritable bowel syndrome patients.Am J Gastroenterol 2000; 95: 152–6.

21 Su X, Sengupta JN, Gebhart GF. Effects of kappa opioidreceptor-selective agonists on responses of pelvic nerveafferents to noxious colorectal distension. J Neurophysiol

1997; 78: 1003–12.22 Gebhart GF, Sengupta JN. Evaluation of visceral pain.

In: Gaginella TS, ed. Handbook of Methods in Gastroin-

testinal Pharmacology. Boca Raton, FL: CRC Press, 1995:359–73.

23 Martinez V, Thakur S, Mogil JS et al. Differential effects ofchemical and mechanical colonic irritation on behavioralpain response to intraperitoneal acetic acid in mice. Pain1999; 81: 179–86.

24 Kelly DD. Stress-induced analgesia. Ann N Y Acad Sci

1986; 467: 1–449.25 Lanteri-Minet M, Isnardon P, de Pommery J et al. Spinal

and hindbrain structures involved in visceroception andvisceronociception as revealed by the expression of Fos,Jun and Krox-24 proteins. Neuroscience 1993; 55: 737–53.



26 Bandler R, Keay KA. Columnar organization in the mid-brain periaqueductal gray and the integration of emotionalexpression. Prog Brain Res 1996; 107: 285–300.

27 Keay KA, Bandler R. Deep and superficial noxious stimu-lation increases Fos-like immunoreactivity in differentregions of the midbrain periaqueductal grey of the rat.Neurosci Lett 1993; 154: 23–6.

28 Anderson RH, Ness TJ, Gebhart GF. A distension controldevice useful for quantitative studies of hollow organsensation. Physiol Behav 1987; 41: 635–8.

29 Kamp EH, Jones RC III, Tillman SR et al. Quantitativeassessment and characterization of visceral nociception

and hyperalgesia in mice. Am J Physiol Gastrointest LiverPhysiol 2003; 284: G434–44.

30 Burton MB, Gebhart GF. Effects of kappa-opioid receptoragonists on responses to colorectal distension in rats withand without acute colonic inflammation. J Pharmacol Expther 1998; 285: 707–15.

31 Gebhart GF, Su X, Joshi S et al. Peripheral opioid modu-lation of visceral pain. Ann NY Acad Sci 2000; 909: 41–50.

32 Sengupta JN, Su X, Gebhart GF. Kappa, but not mu ordelta, opioids attenuate responses to distention of afferentfibers innervating the rat colon. Gastroenterology 1996;111: 968–80.



a model for chronic quantitative studies of colorectal sensitivity using balloon distension in...

Documents