Department of Internal Medicine and Therapeutics, Osaka University
Graduate School of Medicine, Osaka, Japan (S.T., W.-H.S., M.T., N.K.,
A.K., Y.K., S.Y., M.K., H.M., Y.S., M.H.); and Department of Clinical
Laboratory Science, Osaka University Faculty of Medicine, Suita, Japan
(S.K.)
Proton pump inhibitors (PPIs) are antiulcer agents that have both
gastric antisecretory and mucosal protective actions. The mechanisms of PPI-induced gastric mucosal protection are not known. The
present study was designed to examine the mechanism for
lansoprazole-induced gastric mucosal protection in rats. Rats were
given 0.5, 5, and 50 mg/kg/day lansoprazole alone or both lansoprazole
(50 mg/kg/day) and a specific gastrin receptor antagonist
3R-1-(2,2-diethoxyethyl)-((4-methylphenyl)amino-carbonyl methyl)-3-((4-methylphenyl)ureidoindoline-2- one)
(AG-041R) (3, 10, and 30 mg/kg/day) for 14 days. Serum gastrin
concentrations were measured. The expression of cyclooxygenases (COX-1
and COX-2) in the gastric mucosa was analyzed using Western blotting
and immunohistochemical staining. Another series of rats was used to
examine the 1) levels of prostaglandin (PG) E2 in gastric
mucosa, 2) influences of the drugs on gastric damage caused by absolute ethanol, and 3) effects of a COX-2-specific inhibitor on
PGE2 in the gastric mucosa and the mucosal protection
afforded by lansoprazole. Lansoprazole dose dependently increased the
serum gastrin concentration and enhanced the mucosal expression of
COX-2 but not that of COX-1. Lansoprazole increased gastric mucosal
PGE2 and reduced gastric damage caused by ethanol.
Concomitant administration of AG-041R abolished the
lansoprazole-induced COX-2 expression, and increased mucosal
PGE2 and mucosal protection. A specific COX-2 inhibitor blocked the lansoprazole-induced increase in mucosal PGE2
and mucosal protection. Activation of gastrin receptors by endogenous gastrin has a pivotal role in the effects of lansoprazole on COX-2 up-regulation and mucosal protection in the rat stomach.
 |
Introduction |
Proton
pump inhibitors (PPIs) are potent antiulcer agents that have both
gastric antisecretory and mucosal protective actions (Ruwart et al.,
1984
; Okabe et al., 1986; Holm, 1988; Bergmann et al., 1992; Kawano et
al., 1992; Fukuda et al., 1995; Murakami et al., 1996; Blandizzi et
al., 1999). The antisecretory action is due to their inhibitory effects
on the H+,K+-ATPase (proton
pump) in parietal cells (Satoh et al., 1989; Nagaya et al., 1990). The
mechanisms of PPI-induced gastric mucosal protection are not known.
Lansoprazole is a PPI that is widely used to treat peptic ulcers in
humans. A previous study reported that the protective effects of
lansoprazole were inhibited by pretreatment with indomethacin in
gastric lesion models induced by ethanol or ethanol-hydrochloric acid,
suggesting that endogenous prostaglandin (PG) synthesis might account
for the gastroprotective effects of lansoprazole (Blandizzi et al.,
1999). Another study suggested that lansoprazole protects gastric
mucosa from ethanol- and acidified taurocholate-induced damage in a
dose-dependent manner, with ID50 values of 8.5 mg/kg p.o. (ethanol) and 4.1 mg/kg p.o. (acidified taurocholate). The protective effect of lansoprazole was suppressed by functional ablation
of capsaicin-sensitive sensory neurons, or prior administration of
indomethacin or a selective inhibitor of nitric oxide synthesis. These findings suggest that endogenous PGs, together with
capsaicin-sensitive sensory neurons and nitric oxide, mediate
PPI-induced gastric mucosal protection (Murakami et al., 1996).
PGs have an important role in gastric mucosal defense (Robert et al.,
1983; Arakawa et al., 1990). Synthesis of PGs is governed by PG
endoperoxide synthase, or cyclooxygenase (COX: EC 1.14.99.1), which
consists of two isoforms (Kujubu et al., 1991; Xie et al., 1991). The
constitutive isoform (COX-1) is dominantly expressed in platelets,
prostate, and stomach. The mitogen-inducible isoform (COX-2) is
minimally expressed in normal stomach (Seibert et al., 1994; Kargman et
al., 1996). COX-2 expression is enhanced, however, in gastric
epithelial cells after growth stimulation in vitro and in gastric
epithelium after acid-induced damage in vivo (Tsuji et al., 1996;
Sawaoka et al., 1997). We recently demonstrated that COX-2 protein is
overexpressed during the healing of gastric lesions and a
COX-2-specific inhibitor delays healing in the rat, suggesting an
important role for this isozyme in gastric ulcer healing (Sun et al.,
2000). Therefore, the present study examined the effects of 14-day
administration of a PPI, lansoprazole, on gastric mucosal expression of
COX-1 and COX-2 and on gastric mucosal protection in rats. In addition,
long-term acid suppression might result in elevated gastrin, an
important humoral factor in stomach. Gastrin is reported to have
fundamental role in stimulating gastric acid secretion. Furthermore,
gastrin has protective action on gastric mucosa against ethanol-induced
injury (Mercer et al., 1997) and induces growth-promoting effects on
diversity of target cells (Yassin, 1999). Various mechanisms, including
endocrine, paracrine, and autocrine, have been proposed for gastrin's
actions. The mitogenic effects of gastrin are mediated by specific cell surface receptors activated after gastrin binding. The functionally defined receptors for gastrin include cholecystokinin A receptor, which
is discriminating for sulfated CCK8;
gastrin/cholecystokinin B (CCKB) receptor, which binds
gastrin17 sulfated and nonsulfated CCK8 with nearly equal affinities;
cholecystokinin C, which is a low-affinity gastrin binding protein; and
novel, high-affinity receptors selective for amidated gastrin,
processing intermediates of gastrin, or both (Yassin, 1999). We also
examined the influences of a gastrin/CCKb receptor antagonist (Ding et
al., 1997; Chiba et al., 1998; Fukui et al., 1998; Hakanson et al.,
1999) on gastric mucosal expression of cyclooxygenase and gastric
mucosal protection.
| |
Materials and Methods |
Animals and Agents.
Specific pathogen-free male
Sprague-Dawley rats (Nippon SLC, Shizuoka, Japan), aged 6 weeks and
weighing approximately 150 g, were fed with standard pellet chow
and tap water ad libitum. The rats were food-deprived for 24 h but
allowed free access to water before sacrifice. All of the experiments
were performed according to the Guidelines of the Institutional
Committee on Experimental Animals.
Lansoprazole was a gift from Takeda Chemical Industries Co., Ltd.
(Osaka, Japan). AG-041R, a potent and specific gastrin receptor antagonist (Ding et al., 1997; Chiba et al., 1998; Fukui et al., 1998;
Hakanson et al., 1999), was a gift from Chugai Pharmaceutical Co., Ltd.
(Tokyo, Japan). NS-398, a specific inhibitor of COX-2 (Futaki et al.,
1993a,b; Masferrer et al., 1994; Tsuji et al., 1996), was purchased
from Cayman Chemicals (Ann Arbor, MI). The agents were suspended in
0.5% carboxymethylcellulose (CMC; Sigma-Aldrich, St. Louis, MO). The
other agents were purchased from Nacalai Tesque Co. (Kyoto, Japan),
unless stated otherwise.
Influences of Lansoprazole and AG-041R on Gastric Mucosa:
Administration of Test Drugs.
The rats were divided into eight
groups. Seven animals per group were used for assay for serum gastrin.
In these seven animals, four were used for Western analysis and three
were used for immunohistochemical study. The other seven animals per
group were used for assessment of gastric mucosal injury. Six animals
per group were used for prostaglandin assay. The first group received
vehicle, 0.5% CMC (1 ml/kg), via a gastric tube once a day for 14 days. Groups 2, 3, and 4 were treated with lansoprazole (0.5, 5, and 50 mg/kg). Groups 5, 6, and 7 were treated with both lansoprazole (50 mg/kg) and the gastrin receptor antagonist AG-041R (3, 10, and 30 mg/kg). The eighth group was administered with AG-041R (30 mg/kg) for 14 days.
Radioimmunoassay of Serum Gastrin Levels.
Ten hours after
the last administration of the above-mentioned agents, the rats
(n = 7) were anesthetized with sevoflurane and blood
was drawn by a cardiac puncture. The blood was centrifuged at
2500g for 10 min, and serum was collected and stored at
20°C until gastrin determination was performed in duplicated manner (Gastrin-RIA kit II; Dainabot, Tokyo, Japan).
Expression of COX-1 and COX-2 in Rat Gastric Mucosa.
After
the blood samples were collected, the stomach was harvested and opened
along the greater curvature. The oxyntic mucosa was scraped with glass
slides, and immediately frozen in liquid nitrogen and stored at
80°C for Western blot analysis of COX-1 and COX-2 expression.
The gastric mucosal samples were homogenized in phosphate-buffered
saline containing 1% Nonidet P-40, 0.5% sodium deoxycholate, and
0.1% SDS. Protein concentration of the homogenate was measured using
protein assay reagent (BCA kit; Pierce Chemical, Rockford, IL). The
tissue homogenates, 100 µg protein/lane, were electrophoresed in a
10% SDS-polyacrylamide gel, and transferred onto polyvinylidene difluoride membranes (Immobilon; Millipore Corporation, Bedford, MA)
using a semidry transfer cell (Bio-Rad, Hercules, CA). The blots were
pretreated with Tris-buffered saline containing 5% nonfat dry milk,
1% albumin, and 0.1% Tween 20, and then incubated overnight at 4°C
with antibodies for COX-1 and COX-2 (Santa Cruz Biotechnology, Inc.,
Santa Cruz, CA). Filters were washed three times and incubated with a
horseradish peroxidase-conjugated secondary antibody against goat IgG
(DAKO, Glostrup, Denmark), developed using a commercial enhanced
chemiluminescence system (Amersham Biosciences UK, Ltd., Little
Chalfont, Buckinghamshire, UK), and exposed to films (Hyperfilm;
Amersham Biosciences UK, Ltd.). Expression of COX-1 and COX-2 was
semiquantified using a densitometric scanner. The density of the
proteins is expressed as percentage of the control treated with CMC only.
For immunohistochemical analysis, gastric specimens were taken from two
groups of rats (n = 3/group) treated with CMC or 50 mg/kg lansoprazole and fixed in 10% buffered formalin. Sections (4 µm in thickness) cut from paraffin-embedded tissues were
deparaffinized. Sections were then microwaved in citrate buffer, pH
6.1, at 95°C, for 10 min for antigen retrieval. Endogenous peroxidase
activity was quenched by incubation in 3.0%
H2O2 in methanol for 30 min. Nonspecific binding was blocked with 3% normal rabbit serum in phosphate-buffered saline, and the tissues were incubated with primary
goat polyclonal antibodies against COX-1 and COX-2 (1:200 dilution)
overnight at 4°C in 1.5% normal rabbit serum. The sections were
stained according to the avidin-biotin complex method using a
commercial kit (Vectastain kit; Vector Laboratories, Burlingame, CA)
and visualized using 3,3'-diaminobenzidine (DAB) (Vectastain DAB kit;
Vector Laboratories). The specimens were subsequently counterstained
with hematoxylin. Other sections were incubated with the antibodies
preabsorbed with pure blocking COX-1 and COX-2 peptides (Santa Cruz
Biotechnology, Inc.) and then treated with the secondary antibody and
stained using the avidin-biotin complex-DAB method.
Measurement of Prostaglandin E2 Levels in Gastric
Mucosa.
Further experiments were conducted to determine whether
lansoprazole and the gastrin receptor antagonist influence
PGE2 synthesis in rat gastric mucosa. Ten hours
after the final administration of each reagent, all rats, six per
group, were anesthetized with sevoflurane, intragastrically
administered saline containing 100 µM indomethacin and 10 mM EDTA to
block excess PG production in gastric mucosa, and immediately
laparotomized. The stomach was harvested and the oxyntic mucosa was
scraped with glass slides and immediately frozen in liquid nitrogen.
The tissue was weighed and homogenized at 4°C in cold ethanol, and
the homogenate was acidified to pH 4 using diluted HCl and centrifuged.
PGE2 in the supernatant was purified using
C18 solid phase extraction cartridges (Sep-Pak; Waters,
Millford, MA) and eluted with ethyl acetate containing 1% methanol.
PGE2 was solidified using a rotary evaporator, reconstituted in a buffer, and measured by enzyme immunoassay (Cayman
Chemicals). PGE2 levels in the gastric mucosa
were expressed as picograms of PGE2 per gram of
wet tissue.
Effects of Lansoprazole and AG-041R on Ethanol-Induced Gastric
Mucosal Injury.
A separate experiment was designed to determine
whether 14-day administration of lansoprazole and/or the gastrin
receptor antagonist could prevent or attenuate absolute ethanol-induced gastric mucosal injury. Ten hours after the final administration of
lansoprazole and/or AG-041R, all rats (n = 7/group)
were administered with 1 ml of absolute ethanol through an orogastric
tube. One hour later, the rats were sacrificed and laparotomized. The
stomach was harvested, opened along to the greater curvature, extended on a plastic board, and photographed. The areas of macroscopic hemorrhages and erosions were assessed by planimetry. The ulcer index
was expressed as a percentage of the lesion area to the total gastric
glandular area.
For histological assessment, the gastric corpus wall was fixed in
phosphate-buffered formalin, sectioned, and paraffin-embedded. Semithin
sections were deparaffinized, stained with hematoxylin and eosin, and
examined under a light microscope by a pathologist without knowledge of
to which group the specimen belonged. The specimens were coded and
assessed according to the criteria of Whittle et al. (1990). In brief,
a 1-cm length of each histological section was assessed for epithelial
cell damage (a score of 1); glandular disruption, vasocongestion, or
edema in the upper mucosa (a score of 2); hemorrhagic damage in the
mid- to lower mucosa (a score of 3); and deep necrosis and ulceration
(a score of 4). Each section was evaluated on a cumulative basis to
give the histological score, the maximum score thus being 10.
Influences of NS-398 on PGE2 Synthesis in Gastric
Mucosa and Mucosal Protection Afforded by Lansoprazole.
To examine
whether a specific COX-2 inhibitor influences prostaglandin synthesis
in gastric mucosa, rats were treated orally with 10 mg/kg NS-398
10 h after the final administration of 0.5% CMC or 50 mg/kg
lansoprazole. Controls received the corresponding vehicle. One hour
later, the gastric mucosa were harvested from six animals per group and
homogenized for determination of PGE2 levels as
described above.
To investigate whether the gastric mucosal protective action of
lansoprazole is related to endogenous PGs produced by COX-2, NS-398 (10 mg/kg) was administered orally 1 h before the administration of
absolute ethanol. The corresponding vehicle was administered to
controls. One hour later, the rats, seven animals per group, were
sacrificed, and the ulcer index and histological score were assessed as
described above.
Statistical Analyses.
Data were shown as mean ± S.E.M.
from more than six rats per group and were then analyzed by analysis of
variance with Dunnett's multiple comparison test. A probability value
of less than 0.05 was considered statistically significant.
| |
Results |
Serum Gastrin Levels.
The fasting serum gastrin level was
higher in the lansoprazole groups than in the CMC-treated control
group. Lansoprazole increased the serum gastrin level in a
dose-dependent manner. Serum gastrin was higher in rats given 30 mg/kg
AG-041R than in the control group, possibly due to the inhibition of
gastric acid secretion by the agent. Concomitant administration of
lansoprazole and AG-041R significantly increased the serum gastrin
levels compared with control. There is no statistical significance on
serum gastrin between the lansoprazole 50 mg/kg group in the absence of
AG-401R and in the presence of lansoprazole 50 and 3 mg/kg AG-401R
group (Fig. 1).
View larger version (40K):
[in this window]
[in a new window]
|
Fig. 1.
Influences of lansoprazole and AG-041R on serum
gastrin in rats fasted for 24 h. *, p < 0.05; **, p < 0.01; and ***,
p < 0.005 versus the CMC-treated control group.
Data are shown as mean + S.E.M. (n = 7/group).
AG-041R (30 mg/kg) increased serum gastrin level compared with the
vehicle-treated control, suggesting that inhibition of gastric acid
secretion by the presence of the gastrin receptor antagonist also
increases serum gastrin levels. This could explain why serum gastin
levels are maintained in the presence of lansoprazole and AG-401R.
There is no statistical significance on serum gastrin between the
lansoprazole 50 mg/kg group in the absence of AG-401R and in the
presence of lansoprazole 50 and 3 mg/kg AG-401R group.
|
|
Expression of COX-1 and COX-2 in Rat Gastric Mucosa.
In the
control rats treated with 0.5% CMC, the gastric mucosa expressed a
small amount of COX-2 and strongly expressed COX-1. In the groups
treated with lansoprazole (0.5, 5, and 50 mg/kg) for 14 days, the
expression of COX-1 did not change significantly compared with that of
the controls. The expression of COX-2 was enhanced in the animals after
daily administration of lansoprazole for 14 days. The level of COX-2
immunoreactivity was increased by lansoprazole. This increase in COX-2
expression was reduced almost back to the control level by concomitant
administration of AG-041R. There was a dose-dependent reduction in
COX-2 expression with increasing dose of AG-401R. COX-1 expression was
not affected by the gastrin receptor antagonist (Fig.
2).
View larger version (45K):
[in this window]
[in a new window]
|
Fig. 2.
Expression of COX-1 and COX-2 in rat gastric mucosa
by Western blot analysis. A, representative Western blot data for COX-1
and COX-2. Separate blots were used for the analysis of COX-1 and COX-2
expression. The lane marked "0" is the positive control for COX-1
and COX-2 proteins, respectively. B, densitometric data. Density of the
proteins is expressed as percentage of the control treated with CMC
only. Data are shown as mean + S.E.M., n = 4/group.
*, p < 0.05 versus CMC-treated control. The
COX-2 expression in gastric oxyntic mucosa increased after 14-day
treatment with lansoprazole (0.5-50 mg/kg). There is a dose-dependent
reduction in COX-2 expression with increasing dose of AG-401R.
|
|
Localization of COX-1 and COX-2 in gastric mucosa was determined by
immunohistochemical staining (Figs. 3
and 4). Strong COX-1 immunoreactivity was found mainly
in parietal cells and other gastric epithelial cells in both the
control (Fig. 3A) and lansoprazole-treated rats (Figs. 3B and 4A), as
described previously (Jackson et al., 2000). COX-1 was also expressed
in less extent in other types of cells in submucosal areas (Fig. 3A).
The microscopic study also confirmed that the treatment with
lansoprazole did not affect expression of COX-1 in rats. This
immunostaining disappeared with the pretreatment of antibody with the
blocking peptide, indicating that the immunoreactivity was specific to
COX-1.
View larger version (143K):
[in this window]
[in a new window]
|
Fig. 3.
Localization of COX-1 (top) and COX-2 (bottom)
immunoreactivities in the CMC-treated control rats (A and D) and the
lansoprazole-treated rats (B and E). Strong COX-1 immunoreactivity
locates mainly in parietal cells and other gastric epithelial cells in
both the control (Fig. 3A) and lansoprazole-treated rats (Figs. 3B).
COX-1 is also expressed in less extent in other types of cells in
submucosal areas (Fig. 3A). Lansoprazole does not affect expression of
COX-1 in rats. Immunostaining for COX-2 is minimal or negligible in
control rats treated with 14-day administration of CMC (Fig. 3D). After
14-day administration of lansoprazole, however, COX-2 immunoreactivity
is strongly detected from the neck to the bottom of the gastric glands
(Fig. 3E). The immunoreactivity disappears after preincubation with
blocking peptides, suggesting that the immunoreactivity is specific to
the antigens (C and F). Original magnification, 100×.
|
|
View larger version (127K):
[in this window]
[in a new window]
|
Fig. 4.
Enlarged pictures of COX-1 (A) and COX-2 (B)
immunoreactivities in gastric mucosa in a rat treated with
lansoprazole. COX-1 immunoreactivity is found mainly in parietal cells
and other gastric epithelial cells in lansoprazole-treated rats (Figs.
4A). After 14-day administration of lansoprazole, however, COX-2
immunoreactivity is strongly detected from the neck to the bottom of
the gastric glands (Figs. 4B).
|
|
Immunostaining for COX-2 was minimal or negligible in control rats
treated with 14-day administration of CMC (Fig. 3D). After 14-d
administration of lansoprazole, however, COX-2 immunoreactivity was
strongly detected from the neck to the bottom of the gastric glands
(Figs. 3E and 4B). When the antibody was preincubated with the blocking
COX-2 peptide and applied to the sections, no immunoreactive signals
appeared (Fig. 3F).
PGE2 Levels in Gastric Mucosa.
Gastric mucosal
PGE2 level was higher in the lansoprazole groups
than in the CMC-treated control group. Lansoprazole increased gastric
mucosal PGE2 levels in a dose-dependent manner.
Concomitant administration of AG-041R dose dependently suppressed
lansoprazole-induced increase in gastric mucosal
PGE2 (Fig. 5).
View larger version (37K):
[in this window]
[in a new window]
|
Fig. 5.
Influences of lansoprazole and AG-041R on mucosal
PGE2 in rats. *, p < 0.05;
***, p < 0.005 versus the control group. ++,
p < 0.02; +++, p < 0.005 versus the group treated with 50 mg/kg lansoprazole. Data are shown as
mean + S.E.M. (n = 6/group). Gastric mucosal
PGE2 level is higher in the lansoprazole groups than in the
CMC-treated control group. Lansoprazole increases gastric mucosal
PGE2 levels in a dose-dependent manner. Concomitant
administration of AG-041R dose dependently suppresses
lansoprazole-induced increase in gastric mucosal
PGE2.
|
|
Effects of Lansoprazole and AG-041R on Ethanol-Induced Gastric
Mucosal Injury.
In the rats given 0.5% CMC, significant
hemorrhagic streaks and erosions developed mainly in the gastric corpus
mucosa 1 h after the intragastric administration of absolute
ethanol. In the groups administered lansoprazole (0.5, 5, and 50 mg/kg), however, the ulcer index was significantly lower than in the
control group treated with CMC. On the other hand, there were no
significant differences in the ulcer index between the group
administered CMC and the group administered AG-041R. The ulcer index
was significantly larger in the group administered both lansoprazole
and AG-041R than in that of lansoprazole alone (Fig.
6).
View larger version (32K):
[in this window]
[in a new window]
|
Fig. 6.
Influences of lansoprazole and AG-041R on macroscopic
gastric damage induced by ethanol. *, p < 0.05;
**, p < 0.01; and ***,
p < 0.005 versus the control group. ++,
p < 0.02 versus the group treated with 50 mg/kg
lansoprazole. Data are shown as mean + S.E.M. (n = 7/group). In the groups preadministered lansoprazole (0.5, 5, and 50 mg/kg), ethanol-induced gastric damage is significantly smaller than in
the control group pretreated with CMC. On the other hand, there are no
significant differences in the ulcer index between the group
preadministered CMC and the group preadministered AG-041R. The ulcer
index is significantly larger in the group preadministered both
lansoprazole and AG-041R than in that of lansoprazole alone
|
|
Histological assessment demonstrated that hemorrhagic damage to mid- to
lower mucosa was occasionally associated with deep necrosis in rats
treated with CMC and then absolute ethanol. In contrast, in the groups
treated with lansoprazole and then ethanol, the gastric mucosal injury
was restricted to glandular disruption and damage within the upper part
of the mucosa. These results indicated that lansoprazole protects
gastric mucosa from ethanol-induced injury. On the other hand, there
were no significant differences in the histological scores between the
group administered CMC and the group administered AG-041R. The
histological score was significantly larger, however, in the group
given both lansoprazole and AG-041R than in that of lansoprazole alone
(Fig. 7). Consequently, treatment with a
gastrin receptor antagonist inhibited gastric mucosal protection by
lansoprazole.
View larger version (37K):
[in this window]
[in a new window]
|
Fig. 7.
Influences of lansoprazole and AG-041R on
histological gastric damage induced by ethanol. *,
p < 0.05; **, p < 0.01;
and ***, p < 0.005 versus the control group.
+, p < 0.05; ++, p < 0.02 versus the group treated with 50 mg/kg lansoprazole. Data are shown as
mean + S.E.M. (n = 7/group). The 14-day
administration with lansoprazole (0.5, 5, and 50 mg/kg) significantly
reduced microscopic damage after ethanol administration compared with
the control group pretreated with CMC. On the other hand, there are no
significant differences in the ulcer index between the group
preadministered CMC and the group preadministered AG-041R. The ulcer
index is significantly larger in the group preadministered both
lansoprazole and AG-041R than in that of lansoprazole alone.
|
|
Effects of NS-398 on PGE2 Synthesis in Gastric Mucosa
and Mucosal Protection Afforded by Lansoprazole.
Treatment with
NS-398 significantly reduced lansoprazole-induced mucosal
PGE2 formation (Fig.
8). There were no significant differences
in the ulcer index between the group administered CMC and the group
given CMC followed by administration of NS-398. Therefore, NS-398 did
not aggravate the ethanol-induced macroscopic injury of the stomach.
The ulcer index was significantly larger in the group administered
lansoprazole followed by NS-398 than in that of lansoprazole followed
by CMC (Fig. 9A). There were no
significant differences in the histological score between the group
administered CMC and the group administered CMC followed by NS-398.
Therefore, NS-398 did not aggravate the ethanol-induced histological
damage of stomach. The histological score was significantly larger in
the group administered lansoprazole followed by NS-398 than in the
group administered lansoprazole followed by CMC (Fig. 9B).
Consequently, treatment with NS-398 inhibited gastric mucosal protection by lansoprazole.
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 8.
Influences of lansoprazole and NS-398 on gastric
mucosal PGE2. **, p < 0.01 versus
the control treated only with CMC. ++, p < 0.02 versus the rats treated with lansoprazole for 14 days and then CMC.
Data are shown as mean + S.E.M. (n = 7/group).
There is no significant difference on gastric mucosal PGE2
between the rats treated with CMC and NS-398 and the rats treated with
lansoprazole and NS-398.
|
|
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 9.
Influences of lansoprazole and NS-398 on macroscopic
(A) and microscopic (B) damage in rat stomach induced by ethanol.
**, p < 0.01 versus the control treated only
with CMC. ++, p < 0.02 versus the rats treated
with lansoprazole for 14 day and then CMC. Data are shown as mean + S.E.M. (n = 7/group). There are no significant
differences on gastric mucosal damage between the rats treated with CMC
and NS-398 and the rats treated with lansoprazole and NS-398.
|
|
| |
Discussion |
The present study demonstrated that lansoprazole protected gastric
mucosa from ethanol-induced injury in rats. PGE2,
one of the dominant PGs in gastric mucosa and an endogenous mediator of
gastric mucosal protection, increased after 14-day administration of
lansoprazole. Lansoprazole also increased serum gastrin levels in rats,
consistent with clinical findings of increased levels of serum gastrin
in humans during PPI therapy, possibly because of the marked decrease
in gastric acid secretion.
Cytoprotective effects have been demonstrated by PPIs, including
timoprazole, lansoprazole, and others (Ruwart et al., 1984; Okabe et
al., 1986; Holm, 1988; Bergmann et al., 1992; Kawano et al., 1992;
Fukuda et al., 1995; Murakami et al., 1996; Blandizzi et al., 1999).
Studies suggest that PGs are involved in the mucosal protection
afforded by lansoprazole (Murakami et al., 1996; Blandizzi et al.,
1999). However, the mechanism responsible for lansoprazole-induced gastric mucosal protection has not yet been determined. Furthermore, the consequences of a long-term administration of lansoprazole in
gastric mucosal PG production have not yet been investigated. The
present study demonstrated that repeated administration with lansoprazole for 14 days significantly enhanced gastric mucosal PGE2.
In the present study, 14-day administration of lansoprazole was also
associated with a significant increase in fasting serum gastrin
compared with vehicle. Indeed, several studies have suggested that
long-term administration of lansoprazole elevates serum gastrin (Muller
et al., 1989; Brunner et al., 1995). Consequently, AG-041R, the
gastrin-specific receptor antagonist (Ding et al., 1997; Chiba et al.,
1998; Fukui et al., 1998; Hakanson et al., 1999), was used to clarify
the involvement of gastrin and its receptor on mucosal protection
induced by 14-day administration of lansoprazole. No direct
interactions have been reported between lansoprazole and AG-041R (Ding
et al., 1997; Chiba et al., 1998; Fukui et al., 1998; Hakanson et al.,
1999). The study clearly indicated that AG-041R suppresses the increase
in gastric mucosal PGE2 after long-term
administration with lansoprazole, suggesting an important role for
gastrin and its receptors in gastric mucosal PG production induced by
lansoprazole. Furthermore, AG-04R also abolished lansoprazole-induced gastric mucosal protection against ethanol. Thus, lansoprazole might
induce serum gastrin, stimulate gastric mucosal production of
PGE2, and participate in gastric mucosal
protection in rats. The effects of PPIs on COX-2 induction in organs
other than stomach are not yet available in the literature. However,
the present data suggest that the effects of PPIs on COX-2 in stomach
are mediated by gastrin. Thus, PPIs may induce COX-2 only in upper gastrointestinal tract that has cells expressing receptors for gastrin.
The present study also examined the source of PGs in rat gastric mucosa
after 14-day administration of lansoprazole. Because COX is the
rate-limiting enzyme for PG production, the influence of lansoprazole
on the expression of the two isoforms of COX was examined. An initial
study by Seibert et al. (1997) suggested that COX-1 mRNA was readily
detectable in all normal tissue examined, especially in stomach and
others, but that levels of COX-2 mRNA were substantially lower, with
the exception of the brain. Several studies, including the present
study, have shown that unstimulated gastric mucosa express a lower
level of COX-2, whereas gastric mucosal epithelium expresses this
isoform after various stimuli (Sawaoka et al., 1997; Sun et al., 2000).
In the present study, lansoprazole enhanced the expression of
mitogen-inducible COX-2, but not that of constitutive COX-1 in rat
gastric mucosa. AG-041R abolished lansoprazole-induced COX-2
expression, but not that of COX-1 in rat gastric mucosa. These findings
indicate that lansoprazole induces gastric mucosal production of PGs by
enhancing expression of COX-2. Furthermore, the lansoprazole-induced
increase in gastric mucosal PGE2 was blocked by
NS-398, a specific COX-2 inhibitor. NS-398 also attenuated mucosal
protection induced by lansoprazole. Taken together, lansoprazole
induces COX-2-dependent mucosal PG production and mucosal protection in
the rat stomach mediated by endogenous gastrin and gastrin receptors.
Gastrin receptors are localized at parietal cells, mast cells, and
enterochromaffin-like cells in rat corpus mucosa. On the other hand,
immunohistochemical evaluation of COX-2 indicated that COX-2 is
up-regulated not only in parietal cells but also in mucosal neck cells.
Although the precise mechanism for gastrin-induced COX-2 in the stomach
is not known, one source of PG production could be parietal cells that
express gastrin receptors. Gastrin up-regulates heparin-binding epidermal growth factor (EGF)-like growth factor in RGM1 gastric epithelial cells (Kobayashi et al., 1996) transfected with the gene for
gastrin receptor (Miyazaki et al., 1999). Several ligands for EGF
receptors such as EGF and transforming growth factor-
up-regulate
COX-2 in several gastrointestinal cell lines, including RGM1 in vitro
(Sawaoka et al., 1997; 1999). Therefore, future studies should
investigate whether gastrin up-regulates COX-2 via stimulating
expression of one of the EGF-receptor ligands in rat stomach in vivo.
PPIs, including lansoprazole, are used in patients with various
gastroduodenal disorders. PPIs are acid-labile, and clinical formulae
for PPIs are encapsulated granules or coated tablets. The formulation
of lansoprazole used in the present study was purified powder, neither
encapsulated nor coated against intragastric degradation. In addition,
the lowest dose of lansoprazole used in the study of lansoprazole was
0.5 mg/kg, which is equivalent to 30 mg in subjects weighing 60 kg.
Nevertheless, the results clearly indicated that, at doses ranging from
0.5 to 50 mg/kg, lansoprazole efficiently increases serum gastrin,
enhances COX-2 expression and PG production in gastric mucosa, and
protects gastric mucosa. These findings might be relevant to the
long-term effects of the drug in human subjects. It has been suggested
that omeprazole, another PPI, accelerates the healing of gastric ulcers
via an increase in gastrin secretion possibly mediated by the trophic actions of the peptide (Ito et al., 1994). However, there are differences between PPIs in general and synthetic PGs with respect to
their mucosal protective actions. For example, PGs protect gastric
mucosa from necrotizing stimuli at doses that do not inhibit gastric
acid secretion. On the other hand, PPIs inhibit gastric acid secretion
more potently than synthetic prostaglandins. Therefore, gastric
protective actions of PPIs have been attributed to their potent
inhibitory effect on acid secretion (Bergmann et al., 1992; Blandizzi
et al., 1999). However, in the present study, the long-term administration of lansoprazole protected gastric mucosa from absolute ethanol, one of the necrotizing stimuli with the increase in endogenous PGE2. The clinical implications of
lansoprazole-induced gastric mucosal protection remain to be investigated.
Localization of COX-1 and COX-2 in rat gastric mucosa should be
discussed. Jackson et al. (2000) reported that COX-1 immunoreactivity is expressed in parietal cells and other components in gastric mucosa
in human subjects. These findings were confirmed in the present study.
In contrast, the long-term treatment with lansoprazole enhanced the
expression of COX-2 in gastric epithelial cells that did not
necessarily express gastrin/CCKb receptors. These findings strongly
suggest that gastrin indirectly enhances the expression of COX-2 in
gastric mucosa. Recently, it was reported that gastrin up-regulates
growth factors such as heparin-binding EGF-like growth factor in
epithelial cells expressing gastrin/CCKb receptors (Miyazaki et al.,
1999; Kinoshita and Ishihara, 2000). Expression of COX-2 in gastric
epithelium in response to growth factors and mitogens has been well
documented (Sawaoka et al., 1997, 1999). Consequently, long-term
treatment with lansoprazole may enhance COX-2 expression and exert
cytoprotective activity via gastrin-dependent transactivation of a
growth factor and its receptor. Precise mechanisms for
gastrin-dependent COX-2 up-regulation remain to be investigated in
future studies.
In conclusion, the present results demonstrate that intragastric
administration with lansoprazole increases serum gastrin levels,
induces COX-2 expression in gastric mucosa, elevates gastric mucosal
PGE2, and protects gastric mucosa from
necrotizing agents in rats. Treatment with a specific gastrin receptor
antagonist or a specific COX-2 inhibitor abolishes the
lansoprazole-induced increase in gastric mucosal
PGE2 and protection of gastric mucosa from acute
injury. In this experimental rat ulcer model, the mucoprotective effects of lansoprazole depend on COX-2, which is induced by
lansoprazole. Thus, activation of gastrin receptors by endogenous
gastrin has a pivotal role in the effects of lansoprazole on COX-2
up-regulation and PGE2-mediated gastric mucosal protection.
PPI, proton pump inhibitor;
PG, prostaglandin;
COX, cyclooxygenase;
CMC, carboxymethylcellulose;
DAB, 3,3'-diaminobenzidine;
EGF, epidermal growth factor;
NS-398, N-[2-(cyclohexyloxyl)-4-nitrophenyl]-methane
sulfonamide;
AG-041R, 3R-1-(2,2-diethoxyethyl)
amino-carbonyl methyl)-3-((4-methylphenyl)ureido-indoline-2-one).
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Top
Abstract
Introduction
