European Journal of Pharmacology, 195 (1991) 251-259

251
0 1991 Elsevier Science Publishers B.V. 0014-2999/91/$03.50
ADONIS 001429999100246R

RJP 51789

A 55, a thiazolylami W
properties in the rat

Alessandro Grassi ‘, Joachim Ippen ‘, Maurizio Bruno ’ and Giinter Thomas ’

’ Department of Pharmacology, Bayer Italia, Via delle Groane 126. 20024 Garbagnate M.se (MI). Ztab and ’ Zentrale Forschung, Bayer AG,
Leverkusen. F.R. G.

Received 4 July 1990, revised MS received 4 January1991, accepted 8 January1991

The antiulcer activity of BAY P 14551 a thiazolylaminobenzimidazole derivative, was evaluated in different experimental ulcer
models and its antiulcer activity was compared to that of different reference drugs. ‘The overall activity of the compound was equal
to or more potent than that of reference antiulcer drugs, such as pirenzepine, cimetidine and carbenoxolone, but it was not as
potent as tioprostil. The EDso values (expressed as pmol/kg p.o.) were 68 (confidence limits: 51-91) for indomethacin-induced
ulcers, 21 (confidence limits: 13-31) for stress-induced ulcers and 1260 Pmol/kg p.o. (confidence limits: 412-3800) for ulcers
induced by absolute ethanol. The compound had no activity against cysteamine-induced duodenal ulcers and lost its cytoprotective
activity in adrenalectomised rats. Since inhibition of gastric acid secretion was seen, if at all, only with the higher doses, the
gastro-protective action of BAY P 1455 seemed not to be due to an antisecretory effect, but more likely to a gastroprotective action
as hypothesised for prostaglandins.

Antiulcer drugs; Ulcer models (experimental); Indomethacin; Stress; Cysteamine; Ethanol; F’rostaglandins; Gastric acid secretion

1. Introduction 1984) suggested that the definition of cytoprotection

Peptic ulcer disease can occur in any portion of the
gastrointestinal tract, which is bathed by acid and
pepsin, and is defined as a condition in which the
integrity of the gastroduodenal mucosa is compromised.
Peptic ulcer is thought to result from an imbalance
between aggressive (acid and pepsin) and defensive
factors (mucosal resistance) (Grossman, 1979).
The concept of cytoprotection was developed by
Robert and coworkers, based on the macroscopic ob-
servation that rat stomachs resisted the action of necro-
tizing agents when treated with prostaglandins. Cyto-
protection was originally defined as ‘the property of
many prostaglandins to protect the mucosae of the
stomach from becoming inflamed and necrotic, when
this mucosae is exposed to noxious agents’ (Robert et
al., 1979). Subsequent observations demonstrated that
prostaglandins protected the gastric cells localised deep
in the mucosa but not those at the surface of the
epithelium (Lacy and Ito, 1982; Robert et al., 1985).
The failure of prostaglandins to protect superficial epi-
thelial cells against the action of ethanol (Guth et al.,
had to be reconsidered (Robert et al., 1984). Accord-
ingly, the term ‘gastric cytoprotection’ should refer to
the property of certain agents to protect gastric mucosal
tissue located under the superficial epithelium from
becoming haemorrhagic and necrotic after exposure to
noxious agents (Robert, 1984). Prostaglandin-dependent
protection is observed at doses that do not inhibit
gastric acid secretion.
In a large random screening program carried out
with rats, we found that BAY P 1455 (2-(2’-benzimi-
dazolyl)-amino-4-methyl-thiazol), as shown in fig. 1,
behaved as a strong antiulcer and gastroprotective agent.
This paper describes the properties of this compound
and compares its activity with that of some of the best
Bay P 1455
2-(2’-Benzimidazolyl)-amino-4-methyl-thiazol

Correspondence to: A. Grassi. Department of Pharmacology, Bayer Fig. 1. Chemical structure of BAY P 1455 (2-(2’-benzimidazolyl)-
Italia, Via delle Groane 126, 20024 Garbagnate M.se (MI). Italy. amino-4.methyl-thiazol).
 er and antisecretory drugs. including the
etidine. the sntimuscarinic pireuzepine.
K csrbenosolone. and the alcohol pros-
~~~~~~~~~di~~
E B analogue rioprostil.
In-house-bred Sprague-Dawley rats of both sexes
he animals were kept under standard con-
ditions. received normal rodent chow (Dott. Piccioni,
Gestate, (MI) Italy) and had free access to water
throughout the experimental procedures.
2.2. s1dm1nces
B_4Y P 1455 (molecular weight 230.31) was synthe-
sised by Dr. J. Ippen. cimetidine was obtained from
SigurtS (Italy). pirenzepine dihydrochloride and in-
domethacin were purchased from Prodotti Gianni
(Italy). carbenoxolone sodium from Searle. rioprostil
from Ortho. omeprazole from Hassle. cysteamine-hy-
drochloride from BDH Italia, histamine-hydrochloride
and carbamoylcholine-hydrochloride from Merck and
pentagastrin from ICI.
2.3. I,ldornetltacirt-i?ldllcedgastric ulcers
Male Sprague-Dawley rats (150-200 g) were fasted 18
h before the experiments. Indomethacin was suspended
in 1% arabic gum and administered i.p. (15 mg/kg;
according to Lee et al., 1971). Test compounds were
given orally at the same time. Stomachs were examined
5 h after treatment as follows: the animals were killed
and the stomachs were removed, opened along the
smaller curvature, gently rinsed under tap water and
examined-The severity of gastric lesions was calculated
by counting the number of lesions present in each
stomach according to the following severity score: 0 =
normal stomach: 1 = punctiform redness: 2 = lesions of
less than 3 mm in any direction; 3 = lesions more than
3 mm long in any direction.
The ulcer index for each stomach was obtained by
summing the individual lesions of a given score class
and by multiplying the sum by the respective severity
score. Drug effects were evaluated by considering the
percent reduction in the ulcer index of treated animals
compared to that of untreated animals. Stomachs with
an ulcer index of less than 5 were considered protected,
unless two ulcers larger than 1 mm or 1 ulcer larger
than 3 mm were observed.
.?.4. Stress-induced gustric ulcers
Female Sprague-Dawley rats (200-220 g) were fasted
18 h before the experiments. Ulcers were induced by
restraint and water immersion (Senay and Levine, 1967)
as follows. Test compounds were given 60 min before
the animals were placed in plastic cylinders with an
internal diameter of 4.5 cm and with walls perforated
with several holes of 0.7 cm diameter. The rats were
then immersed in cold water (18°C) up to the
manubrium for 2 h. Two hours after the stress, the
stomachs were removed, examined, and scored as de-
scribed above.
2.5. Ethunol-induced gastric ulcers
Female Sprague-Dawley rats (180-210 g) were fasted
18 h before the experiments. Each rat was given 1 ml of
absolute ethanol orally to induce gastric lesions. Test
compounds were given orally 30 min before ethanol
administration. The animals were killed and the
stomachs removed after 1 h, opened along the greater
curvature. gently rinsed under tap water and examined.
The lesions looked like extended bloody streaks cover-
ing about 60% of the corpus surface, as described by
Robert et al. (1979).
Stomachs without lesions or with only one puncti-
form lesion were considered protected. The percentage
of protected animals from each group was calculated.
2.6. C’steamine-induced duodenal ulcers
Female Sprague-Dawley rats (180-200 g) received
normal animal chow throughout the experiments.
Cysteamine was dissolved in saline and administered
S.C. at a dose of 425 mg/kg (Robert et al., 1974). Test
compounds were given orally at the same time. The
duodenum was examined 24 h after cysteamine admin-
istration. A IO-cm segment of duodenum, starting from
the pylorus, was rernoved, opened longitodinally, gently
rinsed under tap water and examined. Cysteamine ad-
ministration induced a single big ulcer in the duodenum.
The severity of this ulceration was evaluated by using
the following arbitrary score: 0 = normal duodenum;
1 = superficial erosions; 2 = deep ulcer and damage to
the muscularis mucosae; 3 = perforated ulcer or ulcer in
animals that died 5 h after cysteamine application. Data
from animals that died within S h after cysteamine
treatment were discarded.
2.7. Studies with adrenalectomised rats
Female Sprague-Dawley rats weighing 180-210 g were
adrenalectomised under ether anaesthesia as described
by Szabo et al. (1983). One week after adrenalectomy

the rats received a drinking solution of 0.9% NaCl for
the duration of the experiment.
Gastric lesions were induced in surgically treated
animals and in control animals by ethanol treatment, as
described above. The number and size of lesions were
similar in both groups. Gastric lesions were evaluated as
described above.
2.8. Determination ofserum corticosterone ievels
Corticosterone levels were measured in the serum of
adrenalectomised and non-adrenalectomised rats. After
the animals had been killed, blood was collected,
cent~fuged, and the sera was frozen. ~orti~osterone was
determined by the fluorimetric assay (Guillemin et al.,
1959).
2.9. Studies with rats with a ligated pylorus
Female Sprague-Dawley rats (180-200 g) were fasted
for 24 h. A mid-line incision was made under ether
anaesthesia and the pylorus was ligated according to the
method of Shay et al. (1954). Drugs were administered
p.o. 60 min before pylorus ligation. Animals were killed
4 h after pylorus ligation, the stomachs were removed,
opened and the gastric content was collected and
centrifuged. The volume of supernatant was measured,
and the acid ~n~entration was estimated by titration at
pH 7.0 with NaOH 0.05 N, using an automatic titrator
(Radiometer).
2.10. Gastric acid secretion in perfused rat stomach
Female rats were fasted 18 h before the experiments
and anaesthetised with ethyl urethane (1 g/kg i.p.).
After a median epigastric laparotomy and a median cut
in the neck had been made, the oesophagus was cannu-
lated, using a cannula with an external diameter of 1.5
mm, as described by Lai (1964). In experiments in
which gastric acid secretion had to be stimulated by an
agonist, the jugular vein was also cannulated with can-
nulas of 1 mm external diameter for the infusion of
secretagogues. A cannula (external diameter 2.5 mm)
was introduced into the stomach through the pylorus,
by means of an incision in the duodenum. A second
cannula (external diameter 1.5 mm) was introduced into
the duodenal loop via the same duodenal incision for
intraduodenal (i.d.) drug application. The stomach was
cleaned by flushing 30 ml of physiological saline solu-
tion (37’C) through the oesophageal cannula. The
stomach was perfused with physiological saline at a rate
of 0.5 ml/min through the same cannula. After 30 min,
15-min samples of perfusate were collected via the
py!oric cannula for acid determination. Each sample
was titrated to pH = 7 with 0.005 N NaOH with an
automatic titrator (Radiometer). After samples had been
253
collected for 30 min under basal conditions, gastric
secretion was stimulated by infusing different secreta-
gogues (1 ml/h for 2.5 h) throuf& the cannula inserted
in the jugular vein. Sixty minutes after the infusion of
the stimulant. the compound or vehicle solution, ad-
justed to pH 5.1, was injected through the duodenal
cannula. To evaluate changes in gastric acid secretion.
acid output in 15 min was measured before and after
drug application and expressed as PEq H’.
2. f 1. statistical anaiys~s
In experiments where total gastro-protection (T-P.)
was evaluated, the EDs, values were calculated with the
statistical test of Litchfield and Wilcoxon (1949). In
experiments where our arbitrary score was used, the
EDso values (i.e. the dose that reduced the ulcerogenic
effect of the irritants by 50%) were obtained by extrapo-
lation from a dose-response curve calculated by vari-
ance analysis. The confidence limits of variance were
calculated according to formula no. 672, as published in
Documenta Geigy (1960).
All values shown in tables express the means f S.E.M.
Unpaired Student’s t-tests were used to evaluate the
significance of experiments results. Data were consid-
ered significantly different when the P value was less
than 0.05.
3. Results
3. I. Indomethacin-induced gastric ulcers
The mean ulcer index in the control group was
between 37 and 43. BAY P 1455, tested from 20 to 220
pmol/kg p.o., reduced the severity of indomethacin-in-
duced lesions and showed a clear dose-response rela-
tionship (fig. 2). The threshold dose was 43 pmol/kg
p.o. The highest dose of BAY P 1455 (220 pmol/kg
po.) provided some rats with complete protection from
ulcer formation.
Cimetidine, which was used as the reference com-
pound for H, blockers, and rioprostil, which was used
as the reference compound for prostaglandins, were also
dose dependently active. The threshold dose of cimeti-
dine was 40 pmol/kg p.o. One fifth of the animals that
received the highest dose (396 pmol/kg p.o.) were com-
pletely protected from ulcer formation. The tbreshold
dose of rioprostil was 0.14 pmol/kg p.o., and tbe num-
ber of animals per group that were protected from ulcer
formation was proportional to the dose used. The EDso
values and respective confidence limits were 68 (51-91)
pmol/kg for BAY P 1455, 96 (65-132) pmol/kg for
cimetidine and 0.3 (0.2-0.4) pmol/kg for rioprostil.
Pirenzepine, which was used as the antimuscatinic refer-
ence drug, and ~arbenoxolone, used as the cytotrop~c
 Mxroscopic score

20 -- ~-~

m----m Rioprostil
0-0 Bay P 1455
151
A-A Pirenzepine

I I
I
701
T
\
54 ---xl
‘\
r

a+ ___-(--- -,
0 ~~~~_~_~~~~_____~_~ I
‘C-1
1..
2 10 100 1000 lE-1 1 10 100 1000
Log doses (uMol,‘kg P.o.~ Log doses (uMol/kg p.o.)
Fig. 2. Dos+response curves for BAY P 1455 (20. 43, 87 and 220 Fig. 3. Dose-response curves for BAY P 1455 (4, 13, 49 and 130
pm&/kg p.0.). cimetidine (40. 119 and 3% pmol/kg p.0.) and pmol/kg p.0.). pirenzepine (7, 24, 71 and 236 pmol/kg p.o.) and
rioprostil(O.14. 0.3. 0.6. 1.1 and 2.3 pmol/kg p-0.) to prevent experi- rioprostil (0.6. 1.1. 2.3 and 3.4 pmol/kg p.o.) against stress-induced
mental gastric ulcer induced by indomethacin (15 mg/kg i.p.) in rats. gastric ulcers. Each point represents the meanfS.E.M. score values.
Twenty animals were used per dose on four different days. The data Twenty animals were used per dose in four different experiments. The
far the cimetidine group were pooled from two experiments per- mean ulcer index of the control rats was between 11 and 16.
formed at different times. Each point represents the mean f S.E.M. of
scores. The mean ulcer index of the control rats was between 37
and 43.
3.2. Stress-induced gastric ulcers

The ulcer index in the control group was 13.5, with

reference drug, did not show a dose-dependent activity
in this ulcer model. Pirenzepine significantly (P < 0.01)
reduced the gastric lesions by 5060% at all doses tested
(table 1). Carbenoxolone demonstrated weak activity
when tested from 5 to 53 pmol/kg p.o. Only the lowest
dose had a statistically significant effect (P c 0.05); no
activity was observed with the highest dose (table 1).
Since some drug remained in the stomach after the
highest dose, we hypothesised that this might have
negatively influenced the availability of the drug.
limits of 11 and 16. When tested from 4 to 130 pmol/kg
p.o. BAY P 1455 exhibited appreciable antiulcer activity
with a clear dose-response relationship (fig. 3). The
EDS, values and their confidence limits were 21 (13-31)
pmol/kg p.o. for BAY P 1455, 15 (7-25) for pirenzepine
tested from 7 to 236 pmol/kg p.o., and 0.98 (0.73-1.22)
for rioprostil tested from 0.6 to 3.4 pmol/kg p.o.
The three compounds protected the rat stomachs in a
dose-dependent manner when the percentage of animals
without ulcers was used as a second parameter (fig. 4).
The other reference compounds showed only weak,

TABLE 1 Percent of protected stomachs

Mean scores + S.E.M. of gastric lesions and percent inhibition versus

controls of pirenxepine and carhenoxolone against indomethacin-in-
duced gastric ulcers.

Compound N. of Dose Score %

rats (pmol/kg p.o.) (meanIS.E.M.) inhibition 60
Pirenzepine 70 0 48.2 f 2.0
20 30 22.6k2.1 h 53.2
20 60 24.2 f 3.0 ’ 49.8
20 118 22.7 f 2.0 ’ 53.0
20 236 20.4* 1.8 b 57.8
20 470 18.9k2.2 b 60.9

Carbenoxolone 20 0 22.7 + 4.5 0

4 13 43 130 7 24 71 236 0.6 1.1 2.3 3.4
20 5 12.5 f 2.3a 44.9
20 18 12.6 f 3.4 44.5 uMol/kg p.o.
20 53 17.9 f 3.3 21.4
20 175 31.4*3.3 - 38.3 Fig. 4. Percentage of animals with no gastric ulcer in the stress-in-
duced ulcer model after administration of BAY P 1455 (4, 13. 43 and
a P < 0.05 and ’ P < 0.01 are sigrrrficantly different from contrcl val- 130 pmol/kg p.0.). pirenzepine (7, 24, 71 and 236 f.tmol/kg p.0.) and
UW.
rioprostil (0.6. 1.1.2.3 and 3.4 pmol/kg p.0.).
 255

TABLE 2 TABLE 3
Mean scores f S.E.M. of gastric lesions and percent inhibition versus Mean scores* S.E.M. of gastric lesions of alI compounds test4
controls of cimetidine and carbenoxolone against stress-induced gastric against Wteamine-inducedduodenal ulcers.The EDso v&e could &
ulcers. calculated for pirenzepine only.
Compound N. of Dose Score % Compound N. Dose Score % ED,
rats (wnol/ (mean f inhibition of (pmol/ (mean* inhibition (conf.
kg P.o.) S.E.M.) rats kg p.o.) S.E.M.) lim)
Cimelidine 25 0 11.1 f1.5 Bay P 1455 25 0 3.2 +0.2
20 119 11.9k2.7 - 7.0 25 109 3.5 +0.2 - 10.1
20 400 5.5 *0.7 b 51.0 22 217 3.2 +0.3 -2.1
20 1189 6.4kO.9 b 42.4 18 434 2.8 +0.4 12.1
Carbenoxolone 25 0 11 868 2.8 +0.3 10.8
12.9k2.9
25 53 7.6kl.O a 41.0 Cimetidine 20 0 3.0 +I0.2
24 175 9.4* 1.3 27.2 16 198 3.5 + 0.4 - 16.7
25 526 lO.O& i.8 22.0 14 396 2.7 f 0.5 9.5
a P < 0.05 and h P < 0.01 are significandy different from control val- 12 793 2.9kO.4 2.8
ues. 13 1585 2.6 +0.3 12.8

Pirenzepine 33 0 3.2 f 0.3

non-dose-dependent actions in this ulcer model: 38 60 2.2*0.2 b 31.8 100
cimetidine significantly reduced (P < 0.01) the ulcer in- 37 118 1.8kO.2 b 45.1 (67-l 36)
39 236 0.8 i 0.2 b 74.7
dex after the two highest doses (51 and 42%, table 2), 40 470 0.4+0.1 h 89.2
whereas carbenoxolone had a significant effect (P <
Carbenoxolone 19 0 4.4kO.2
0.05) only at the intermediate dose of 53 pmol/kg p.o.
10 175 3.8 +0.6 14.0
20 526 4.7 * 0.2 - 5.2
3.3. Ethanol-induced gastric ulcers 17 1750 3.7 io.4 16.2

The protective action of BAY P 1455 was less pro- Rioprostil 22 0 2.7kO.3
19 1.1 1.8kO.3 = 34.4
nounced in this ulcer model. The compound exhibited a
18 2.3 1.4kO.3 b 47.0
rather flat dose-response relationship in the range from 19 3.4 1.6kO.3 b 42.1
130 to 4340 pmol/kg p.o. (fig. 5). The EDS,, calculated 21 6.8 0.7kO.2 b 74.0
according to the Litchfield and Wilcoxon’s method a P -=z0.05 and b P -Z0.01 are significantly different from control val-
(1949), was 1260 (confidence limits: 412-3800) pmol/kg. ues.
In contrast, the reference compounds pirenzepine,
carbenoxolone and rioprostil exhibited a clear dose-re-
sponse relationship. Cimetidine was not active in this
experimental model in doses up to 3960 pmol/kg p.o_ ble activity in this ulcer model (table 3). Only pirenze-
The respective EDso values (confidence limits) of pine (from 60 to 470 pmol/kg p.o.) exhibited a dose-de-
pirenzepine, carbenoxolone and rioprostil were 990 pendent protective action, with an EDso of 100 (confi-
(674-1450), 482 (319-725) and 0.51 (0.28-0.90) pmol/kg dence limits: 67-136) pmol/kg, while good protection,
p.o. The potency of BAY P 1455 was similar to that of but not completely dose-related, was observed with
carbenoxolone and pirenzepine but less pronounced rioprostil (table 3).
than that of rioprostil.
3.5. Antiulcer activity in adrenalectomised rats
3.4. Cysteamine-induced duodenal ulcers
Since corticosteroids exert a permissive role on the
Like cimetidine and carbenoxolone, BAY P 1455 cytoprotective effect of prostaglandins, as described by
(from 109 to 868 pmol/kg p.o.) did not show apprecia- Szabo et al. (1983), the protective action of BAY P 1455

TABLE 4
Mean blood corlicosterone levels (pg/ml) f S.E.M. after administration of lest compounds lo normal (NOR) and adrenslectomised (ADR) rats.

Compound Dose N. of NOR rats N. of ADR rats %

(IrmoWg P.o.) rats (mean f S.E.M.) rats (mean* S.E.M.) reduction
Controls 0 30 1.02fO.W 28 n.l2+O.Ol 88
Bay P 1455 1300 10 1.24 f 0.C9 9 0.17f0.03 86
Pirenzepine 707 10 0.78 f 0.05 10 0.07 f 0.01 91
Carbenoxolone 526 10 1.00*0.07 8 0.18&0.03 82
Rioprostil 0.8 10 1.04kO.08 10 0.13kO.02 88
 To?c acid o&put (uEq ;i+)

:L100-- ..____-. ~~_..

e -. e@ay P ?355 E--0Cimetidine
A--~APirenzepile *- * Orneprazole
8CO..

0 +~_-- ,.. ~~~ _~~i-.~-.--T-.-._~_ _1

__-
:
.- .“.

.
.^,.-.
i I ._
?E-’ t
10 ‘00 1000
[; ;;c<s ..‘.‘,: ,,_: ? : Log doses (uLlol,‘kg p.0.)

Fig. 5. Ckw-re~~~nsr tune< for B.AY P 1455 (130. 434. 1300 and Fig. 7. Dose-response curves for BAY P 1455 (130. 434 and 1300
4340 amoI;kg p.0.). pirenzepine (154. 707 and 1410 gmol/kg p.0.). pmol/kg). cimetidinc (39. 119. 396 and 1189 pmol/kg), pirenzepine
csrbenoxokme (175. 330. 526 and 1750 pmol/kg p.0.) and rioprostt! (0.71. 2.4. 7.1 and 23.6 pmol/kg) and rioprostil (0.3. 3 and 30
(0.W. 0.08. 0.3 and 08 pmol.,‘kg po.) against ethano!-induced gastric pmol/kg) administered orally to pylorus-ligated rats. The data repre-
ukxrs. Twenty animals per group \\ere used in four experiments. srnt the mean+S.E.M. FEq of H ’ secreted in 4 h under basal
conditions.

(13 ~mol/kg p.o.). pirenzepine (707 pmol/kg p.1~).

carhcnosolone (526 ,ttmol/kg p.0.) and rioprostil 10.8 ties in adrenalectomised animals (fig. 6). Since pirenze-
~mol/kg po.) was studied in adrenalectomised rats in pine exhibited only weak protection in normal rats, we
\vhich gastric lesions had been induced by the oral could not determine whether the absence of a protective
ad~nistration of absolute ethanol. as described under effect after adrenalectomy was due to the surgical pro-
1.5. All compounds were tested at a dose near the EDso cedure or not.
values determined in the previous experiments. In order
to establish vvhether the adrenalectomy had been per-
TABLE 5
formed correctly. we assayed the levels of corticosterone
in rat blood (table 4): these were found to be reduced Effect of the antiulcer drugs on the meanfS.E.M. total acid output
(TAO) in :he in situ perfused stomach model under basal conditions.
by about 909 in comparison to the levels found in
normal rats. BAY P 1455. carbenoxolone and rioprostil. Compound N. of Dose TAO(pEqH+) %
\vbich under normal conditions prevented gastric lesions rats (wol/ (mean f S.E.M.) inhibition
by 60. 70 and 60%. showed no gastroprotective proper- kg p.o-)
BAY P 1455 14 0 29.8k4.3 -
8 868 20.9 rt 3.7 29.0
8 1300 21.9+ 2.3 26.0

Cimetidine 46 0 26.8k1.8
20 12 23.2 + 2.6 13.4
20 40 20.6? 2.2 a 22.9
19 119 17.4k2.1 h 34.9
20 400 13.9k1.8 h 48.1

Pirenzepine 24 0 24.9 + 2.9

16 70 17.4* 1.8 ’ 30.1
-.. -
16 236 20.0 + 2.6 19.6
16 700 14.9*1.4h 40.1

Omeprazole 8 0 21.9*3.4
9 3 20.8 + 2.8 4.8
8 9 12.4k1.3 a 43.6
9 30 9.3*0.5 h 57.6

Rioprostil 8 0 27.4k4.2
Fig. 6. Gastroprotective activity (percentage of protected stomachs) of 8 0.04 28.3 +4.1 -3.5
BAY P 1455 (1300 pmol/kg p.o.). pirenzepine (707 amol/kg p.0.). 8 0.14 16.3k4.4 40.4
carbenoxolone (526 pmol/kg p.0.) and rioprostil (0.8 pmol/kg p.0.) 8 0.56 12.2+1.6 h 55.3
against gastric u!cers induced by absolute ethanol in normal and a P < 0.05 and h P < 0.01 are significantly different from control val-
adrenalectomised rats. Eight rats were used for each group. ues.
 257

3.6. Gastric acid secretion in pylorus-ligated rats
BAY P 1455 tested at 130, 434 and 1300 pmol/kg
p.o. dose dependently reduced the total gastric acid
output in this classical experimental model (fig. 7). The
EDSUwas 970 (confidence limits: 618-2290) pmol/kg.
The volume of gastric juice secreted was not reduced in
a 4-h period.
The other compounds used as reference were also
active in a dose-dependent manner, not only by reduc-
ing the total acid output but also by reducing the
volume of gastric juice secreted. Omeprazole was the
most active compound, with an ED,, of 3 pmol/kg p.o.
(confidence limits: 0.9-lo), while the EDso values of
pirenzepine and cimetidine were 5.4 (confidence limits:
2.3-12.9) and 179 (confidence limits: 102-295) pmol/kg
p.0.. respectively.
3.7. Gastric acid secretion in perfused rat stomachs
The effect of BAY P 1455 on basal and stimulated
gastric acid secretion in the perfused rat stomach model
was compared with rhat of cimetidine. pirenzepine,
rioprostil and omeprazole. Gastric acid secretion was
stimulated with either histamine. pentagastrin, or
carbamoylcholine.
secretion in rats this effect was not dose-dependent
(table 5). Pirenzepine, when tested from 70 to 700
qWkg id., had the same effect.
As expected. gastric secretion was reduced in a dose-
dependent manner by all three reference compounds.
Cimetidine was tested i.d. from 12 to 400 pmol/kg_
omeprazole from 3 to 30 pmol/kg. and rioprostil from
0.04 to 0.56 pmol/kg.
3.7.2. Stimulated gastric acid secretion
The thiazolylaminobenzimidazole derivative BAY P
1455, tested at 220 and 868 pmol/kg i.d.. did not show
appreciable activity in reducing gastric acid secretion
stimulated by histamine (table 6). In contrast. cimeti-
dine, rioprostil and omeprazole reduced the stimulated
gastric acid secretion in a dose-dependent manner. The
dose-response curves for cimetidine and omeprazole
were steeper than the dose-response curve for rioprostil.
Pirenzepine was completely inactive in this experimen-
tal model when tested in doses up to 2400 Pmol/kg i.d.
BAY P 1455 also did not influence gastric acid secretion
stimulated by pentagastrin and carbamoylcholine.
4. Discussion

3.7.1. Basal gastric acid secretion The aim of the study was to verify the antiulcer
BAY P 1455 tested at 868 and 1300 pmol/kg i.d. activity of BAY P 1455. as demonstrated in primary
reduced slightly, but significantly, basal gastric acid screening p:sts, and to clarify its mode of action.
The benzimidazole derivative, BAY P 1455. had clear
TABLE 6
antiulcer effects in all the experimental models tested,
with the exception of the cysteamine duodenal ulcer
Effect of the antiulcer drugs tested on the mean+S.E.M. total acid
output (TAO) in the in situ perfused stomach model after histamine model. Thus, the compound had gastroprotective prop-
stimulation (6 mg/kg per min i.v.). erties, since duodenal ulcers induced by cysteamine are
mainly the consequence of gastric hypersecretion (Szako
Compound N. OC Dose TAO (FEq H+ ) %
rats (pmol/ (mean + S.E.M.) inhibition et al., 1979; Kirkegaard et al., 1980; Boesby et al..
kg P.o.) 1983). The protective activity of BAY P 1455 was
BAY P 1455 16 0 219k22
clear-cut and dose-dependent in the gastric ulcer models
8 220 166+39 24.2 where the ulcerogenic agents produced ulcers not only
8 868 176&29 19.5 by causing hypersecretion. but also by reducing defen-
Cimetidine 22 0 304 f 31
sive mechanisms. Overall, the compound proved to be a
8 4 226 f 43 25.7 better gastroprotective agent than the reference com-
8 12 208+18 31.6 pounds used, with the exception of the prostaglandin
8 40 217+28 a 28.6 analogue rioprostil.
x 120 131+15h 56.9
As shown in table 7, BAY P 1455 was more active in
Omeprazole 0 176 f 26 the stress-induced ulcer model than in the other experi-
3 148&18 15.7 mental models (EDso values around 20 pmol/kg P.0.).
9 53k 6h 69.9
However, its protective activity against indomethacin-
30 26k 4h 85.2
90 23+ 6h 87.2
induced ulcers was also noteworky (ED,, of 68 pmol/kg
p.0.). In the stress-induced ulcer model, BAY P I455
Rioprostil 8 0 177123 was at least as active as the antimuscarinic drug
8 0.14 137511 22.8
0.28 128f15 27.8
pirenzepine (fig. 3), while the H, blocker cimetidine and
8
8 0.56 124k 16 30.0 the cytotrophic compound carbenoxolone were com-
8 1.13 109*15 ‘I 98.5 pletely inactive. This might indicate that histamine-in-
a P < 0.05 and h P < 0.01 are significantly different from control G
duced gastric secretion does not play an important role
ues. in this ulcer model.
 Aknowledgements
, t&tea wtth thetr conhdcnce hmit?s of the antiulcer drugs tested
tn dtffe,mnt e\pertmetnaf rat models of ulcers. nd. = EDS,, not dcfina- The authors wish to express their gratitude to G. Boldorini and R.
blc. Paglia for their excellent technical assistance.

Q’om~+wnd Imiomethaiin Sirens Ahsolute Cysteamine

(&krn0l,‘& p.o.1 OJmol ethanol (fimof
kg p.0.) (pmol,/ kg P-0.)
kg P.0.) References

RI> P lS55 6S 21 1260 n.d.

(13-31) (41338.w Boesby. S.. WK. Man. R. Mendez-Diaz and J. Spencer. 1983, Effect
(51~91)
bllS9 n.d. n.d. of cysteamine on gastroduodenal mucosal histamine in rat, Gut
Ctmettdme 96
24. 935.
(65-132)
990 100 Documenta Geigy Wissenchaftliche Tabellen, 1960, 6 Auflage. for-
Arenzcpine No dose-resp. 15
(7-25) (674-l 450) (67-136) mula n. 672. p. 170.8.
r&tionship
Grossman, M.T.. 1979. in: Cecil Textbook of Medicine, 4th ed., eds.
CarhertkW-
> 526 482 r.d. P.B. Beenson. W. McDermott and J.B. Wyngaarden (W.B.
OIORRC >I35
(319-7251 Saunders. Philadelphia) p. 1502.
No dose-resp. Guillemin. R.. G.W. Clayton. H.S. Lipscomb and J.D. Smith. 1959.
Riaprostil 0.3 i 0.51
(0.2-0.4) (0.7-1.2) (0.2%0.90) relationship Fluorimetric measurement of rat plasma and adrenal corti-
costerone concentration. J. Lab. Clin. Med. 53. 830.
Guth. P.H.. G. Paulsen and H. Nagata, 1984. Histologic and microcir-
culatory changes in alcohol-induced gastric lesions in the rat:
Effect of prostaglandin cytoprotection, Gastroenterology 87. 1083.
Johnson. F.R.. 1968. The cytological approach to the evaluation of
in the indometbacin-induced ulcer model, BAY P drug action on gastric mucosh. in: Carbenoxolone Sodium, eds.
J.M. Robson and F.M. Sullivan (Butterworths, London) p. 93.
id55 was but 2.5 times as active as cimetidine (fig. 2).
Kirkegaard. P.. S.S. Poulsen. F.B. Loud, C. Halse and J. Christiansen.
Tire observation that cimetidine was effective in this 1980. Cysteamine-induced duodenal ulcer and acid secretion in the
model indicates that inhibition of acid secretion might rat. Stand. J. Gastroenterol. 15, 621.
improve gastric tolerance to non-steroideal anti-in- Kusterer. K. and S. Szabo. 1987, Gatric mucosal protection by
flammatory drugs. acetazolamide derivatives: role of carbonic anbydrase and sulf-
hydryls. European J. Pharmacol. 141, 7.
In the ethanol-induced ulcer model. where the lesions
Lacy. E.R. and S. Ito, 1982, Microscopic analysis of ethanol damage
were more severe. carbenoxolone, which is thought to to rat gastric mucosa after treatment with a prostaglandin, Gastro-
act by increasing mucus secretion and by accelerating enterology 83. 619.
mucosal repair mechanisms (Johnson. 1968). and BAY Lai, K.S.. 1964, Studies on gastrin. Gut 5, 327.
P 1455 had equal EDs, values. However, the doses Lee. H.Y.. K.W. Mollison and W.D. Cheng, 1971, The effect of
needed were rather high. and the effects were less clear- anti-ulcer agents on indomethacin-induced gastric ulceration in
the rat, Arch. fnt. Pharmacodyn. 191. 370.
cut than those observed in the other ulcer models. Litchfield. J.T. and F. Wilcoxon. 1949, A simplified method of
Furthermore. we feel that our method of ulcer evalua- evaluating dose-effect experiments, J. Pharmacol. Exp. Ther. 96.
tion may not have allowed us to detect more subtle 99.
differences. Robert, A., 1984. Role of endogenous and exogenous prostaglandins
The loss of cytoprotective activity in adren- in mucosal protection, in: Mechanism of Mucosal Protection in
the Upper Gastrointestinal Tract, eds. A. Allen. G. Flemstrijm, A.
alectomised rats suggested that corticosteroids may have
Gamer, W. Silen and L.S. Tumberg (Raven Press, New York) p.
a permissive role in the protective effects of BAY P 377.
1455. carbenoxolone and rioprostil. The permissive role Robert, A.. C. Lancaster. J.P. Davis, S.O. Field and J.E. Nezamis,
of corticosteroids in the efficacy of prostaglandins was 1984. Distinction between antiulcer effect and cytoprotection,
first described by Szabo et al. (1983). Stand. J. Gastroenterol. 101 (Suppl. 19). 69.
Robert, A., C. Lancaster, J.P. Davis. S.O. Field, A.J. Wickrema Sinha
The results obtained in the experiments in which
and B.A. Thornburgh. 1985, Cytoprotection by prostaglsndin oc-
gastric acid secretion was investigated showed that, like curs in spite of penetretion of absolute ethanol into the gastric
prostaglandins (Robert et al., 1979), BAY P 1455 re- mucosa, Gastroenterology 88. 328.
duced the total acid output at doses much higher than Robert. A., J.E. Nezamis, C. Lancaster and J.N. Badalamenti. 1974,
the gastroprotective doses. Cysteamine-induced duodenal ulcers: a new model to test anti-
ulcer agents, Digestion 11, 199.
In conclusion, BAY P 1455 is a gastroprotective
Robert. A.. J.E. Nezamis. C. Lancaster and A.J. Hanchar. 1979,
compound that was more active than other standard Cytoprotection by prostaglandins in rats. Prevention of gastric
antiulcer drugs in many experimental ulcer models; necrosis produced by alcohol, HCI, NaOH, hypertonic NaCl, and
however, it was less potent than the synthetic pros- thermal injury. Castroenterology 77,433.
taghindin rioprostil. We think that the sulphur in the Senay. E.C. and R.J. Levine, 1967, Synergism between cold and
restraint for rapid production of stress ulcers in rats, Proc. Sot.
thi=ole ring may be implicated in the gastroprotective
Exp. Biol. Med. 124. 1221.
action of the compound, as suggested by Szabo et al. Shay. H.. D.C. Sun and M. Gruenstein, 1954, A quantitative sponta-
(1981) and Kusterer and Szabo (1987). neous gastric secretion in the rat, Ctstroenterology 26, 906.
Sxabo. S., G.T. Gallagher. H.C. Homer, P.W. Frankel. R.H. Under-
wood, S.J. Konturek, T. Brzozowski and S. Trier, 1983. Role of
adrenal cortex in gastric mucosal protection by prostaglandins,
sulfhydryls, and cimetidine in the rat. Gastroenterology 85. 1384.
Sxabo. S., L.R. Haith and E.S. Reynolds, 1979, Pathogenesis of
duodenal ulceration produced by cysteamine or propionittite. In-
Buenos of vagotomy. sympathectomy, histamine depletion. H-2
receptor antagonists and hormones, Dig. Dis. Sci. 24. 471.
Smbo, S.. J.S. Trier and P.W. Frankel. 1981. Sulfhydryl compounds
may mediate gastric cytoprotection. Science 214,200.