- Email: [email protected]
Bilateral adrenalectomy worsens gastric mucosal lesions induced by indomethacin in the rat
GASTROENTEROLOGY
1989;97:284-93
Bilateral Adrenalectomy Worsens Gastric Mucosal Lesions Induced by Indomethacin in the Rat Role of Enhanced
Gastric Motility
KOJI TAKEUCHI, HIDEYUKI NISHIWAKI, MEGUMU HIROMICHI NIIDA, and SUSUMU OKABE
OKADA,
Department of Applied Pharmacology, Kyoto, Japan
Misasagi, Yamashina,
Kyoto Pharmaceutical
The mechanism by which bilateral adrenalectomy worsens indomethacin-induced gastric lesions was investigated in rats. In sham-operated rats subcutaneously administered indomethacin produced gastric lesions at doses of 10 mg/kg body wt or greater, in association with lowering of blood glucose levels. In a parallel study, indomethacin induced gastric hypermotility at the same dose levels but had no effect on acid output or mucosal blood flow even at 25 mg/kg body wt. Adrenalectomy (2 wk) itself significantly reduced the blood glucose levels (-50%) and markedly potentiated the ulcerogenic and motility responses caused by indomethacin; the ED,, values dropped to -10 times lower than those in sham-operated rats. Both acid output and mucosal blood flow were significantly reduced by adrenalectomy, but these values were increased after indomethacin treatment (3 mg/kg body wt). The ulcerogenic and motility responses caused by indomethacin were significantly reduced by acute infusion of glucose (25% wtht, 1.2ml/h) intravenously in both sham-operated and adrenalectomized rats, and by subcutaneous administration of hydrocortisone acetate (10 mg/kg body wt for 2 wk) in the latter group. When the motility and the ulcer score were determined in the same animals, a highly significant relationship was found between these two factors in both sham-operated and adrenalectomized rats. These results suggest that (a) the increased gastric motility may be a key element in the pathogenesis of indomethacin-induced lesions and in the mechanism for aggravation of the lesions by adrenalectomy, and (b) abrasion of adrenal glands by inducing hypoglycemia may sensitize the system to indomethacin and increase gastric motility.
University,
ndomethacin, a nonsteroidal antiinflammatory drug, is known to induce damage in the gastrointestinal mucosa of experimental animals and humans. The mechanism by which indomethacin induces gastric lesions is generally considered to involve depletion of endogenous prostaglandins (1,Z). However, recent studies showed the importance of gastric hypermotility in the pathogenetic mechanism of these lesions in response to indomethacin (3-5). Adrenalectomy has been demonstrated to worsen these lesions in rats (6,7), although the detailed mechanisms for aggravation remain unknown. If the motor alterations are causally related to formation of gastric lesions induced by indomethacin, adrenalectomy might induce gastric motility alterations by itself or increase the motility response to indomethacin, thereby worsening gastric lesions. The present study was therefore designed (a) to examine the effects of indomethacin on gastric acid output, mucosal blood flow, motility, and blood glucose level in adrenalectomized and sham-operated rats, and (b) to correlate these effects with the ulcerogenic property of indomethacin.
I
Materials
and
Methods
Male Sprague-Dawley rats (230-250 g) were used. All studies were carried out using 4-8 rats per group. General Group
Protocol
Experiments A consisted
were performed of sham-operated
0 1989 by the American
in three groups. rats, and group B of
Gastroenterological 0016.5085/89/$3.50
Association
ADRENALECTOMY ON INDOMETHACIN-INDUCED LESIONS 285
August 1989
adrenalectomized rats. In group B, bilateral adrenalectomy was performed by a translumbar route under ether anesthesia, and the adrenal glands were removed carefully with forceps. The animals were then fed on normal laboratory diet and 2% NaCl solution after operation. Some animals in group B received hydrocortisone acetate (Sigma Chemicals, St. Louis, MO.) subcutaneously in a dose of 10 mgikg body wt once daily for 2 wk after adrenalectomy (group C). In group A, bilateral body walls of the lumbar portion were incised, the incisions were closed without removing the adrenal glands, and rats were fed on normal diet and tap water. The animals in each group were used in the following experiments 14 days after the above operation, and they were deprived of food but allowed free access to tap water (group A) or 2% NaCl (groups B and C) Although body weight for 18 h before the experiments. was increased to lesser degrees in groups B and C as compared with group A, these animals tolerated the operation. Induction
of, Gastric
Lesions
The animals were given indomethacin [Sigma), suspended in saline with a drop of Tween 80 (Nakarai, Kyoto, Japan), in doses of l-25 mg/kg body wt in group A, in doses of 0.1-25mgikg body wt in group B, and in a dose of 3 mg/kg body wt in group C. In some cases, the effect of glucose infusion on gastric lesions induced by indomethacin was examined in groups A and B. The animals were kept in Bollman cages, and glucose (25% solution, wtiwt) was infused intravenously through the tail vein at a rate of 1.2 ml/h, starting 1 h before indomethacin treatment. Four hours later, the animals were killed, and the stomachs were removed, inflated by injecting 10 ml of 2% formalin, immersed in 2% formalin for 10 min to fix the gastric wall, and opened along the greater curvature. The length of each lesion developed in the glandular mucosa was measured under a dissecting microscope with a square grid (x10), summed per stomach, and used as a lesion index. In each study, the person measuring the lesions did not know the treatment given to the animals. In the above study, the terminal blood glucose levels were also determined in the animals of groups A, B, and C with or without indomethacin treatment. Four hours after indomethacin treatment, blood was collected from the descending aorta, and blood glucose levels were measured chemically using 0-toluidine boric acid reagent (Glucose test kit; Wako, Osaka, Japan). The absorbance was measured at 635 nm on a Hitachi spectrophotometer (model 200-100). In a preliminary study, a calibration curve was made using the standard glucose solution (Wako; 50-500 mgidl) and a linearity was obtained in the above range (r = 0.9999). Determination of glucose was performed in duplicate, and the results were expressed as milligrams per deciliter. Determination Acid Output
of Mucosal
Blood
Flow
and
Gastric mucosal blood flow was measured using the aminopyrine clearance method described by Jacobson
et al. (8)and modified by Kitagawa et al. (9). A constant plasma aminopyrine level was maintained by continuous infusion of aminopyrine (Sigma: 6.6mg/kg body wt h) at a rate of 1.2 ml/h, starting 30 min after a single dose of 30 mgikg body wt given intravenously by a bolus injection. Under ether anesthesia, the abdomen was incised and both the stomach and duodenum were exposed. A polyethylene tube (inside diameter 3 mm) was inserted into the stomach through the pylorus from a small incision made in the duodenum, and the tube was placed by a ligature around the pylorus and withdrawn through the lateral abdominal wall. After recovery from anesthesia, the animals were kept in Bollman cages during the test period. The stomachs were filled with 2 ml of gastric solution of the following compositions: 4.47 g n-( -)-mannitol and 0.45g glycine in 100 ml of distilled water adjusted to pH 3.5 by adding 0.1 N HCl. The gastric solution was changed every 15 min through the fistula for a total period of 3 h. In the present study, the plasma aminopyrine levels 60 min after the onset of aminopyrine infusion and immediately after the end of the experiment were 18.8 t 2.3 and 16.2 ? 1.9 pgiml, respectively (n = 9). The plasma level of aminopyrine at each 15-min interval during the experimental period was thus estimated by interpolating between two measured points and was paired with each aminopyrine determination in the gastric juice. The content of aminopyrine in the plasma and gastric juice was measured according to the method of Brodie and Axelrod (10). Gastric mucosal blood flow was determined every 15 min and expressed as milliliters of blood per 15 min. The collected gastric juice was analyzed for volume and titrated with 0.1 N NaOH to pH 7.0 for titratable acidity. Acid output was calculated by subtracting the acidity of the initial gastric juice (glycine buffered solution) from the titratable value of each final sample, and expressed as microequivalents per 15 min. Approximately 1 h after basal acid output had stabilized, the animals were given indomethacin subcutaneously in a dose of 25 mg/kg body wt (group A) or 3 mg/kg body wt (groups B and C), and the measurement was continued for 2 h thereafter.
Determination
of Gastric
Motility
Gastric motility was measured in the animals of groups A, B, and C, using a balloon according to a previous paper (4). Briefly, under ether anesthesia a balloon (containing -0.8 ml of water) and the support catheter were placed in the glandular stomach through an incision of the forestomach. The animals were then placed in Bollman cages, and the support catheter was connected to a pressure transducer and polygraph device (Nihon Koden, Ibaragi, Japan). Gastric motility was continuously monitored on a Hitachi recorder (model 056; Hitachi, Mito, Japan) as intraluminal pressure recordings. Quantitative analysis of the motility was performed by counting the number of contractions with an amplitude of 15 cm H,O or greater, and by measuring the amplitude of each contraction over a lo-min period, determining the mean of each rat for this period from these values, and then by calculating the mean for each time period from 4-6 rats per group.
286
TAKEUCHI ET AL.
GASTROENTEROLOGY Vol. 97, No. 2
Approximately 1 h after basal motility had stabilized, indomethacin was given subcutaneously in doses of l-25 mg/kg body wt, 0.1-S mg/kg body wt, and 3 mgikg body wt, respectively, to group A, B, and C, and gastric motility was measured for 2 h thereafter. In some cases, the effect of glucose (25% wffwt) on gastric motility was examined in groups A and B. Glucose was infused intravenously through the tail vein at a rate of 1.2 ml/h during a test period, starting 60 min before administration of indomethacin (11). In another experiment, the motility, the lesion index, and the terminal blood glucose were determined in the same animals of groups A and B in response to various doses of indomethacin. In these cases, the motility was recorded for 4 h after indomethacin treatment, and the lesion index and the glucose level were determined as described previously.
Statistics Data are presented as the mean * SE from 4-8 rats per group. Statistical analysis was performed using a two-tailed Dunnett’s multiple comparison test (12), and values of p < 0.05 were regarded as significant. A regression analysis was used to determine a correlation coefficient between two different variates.
Results Development lndomethacin
of Gastric Lesions Induced
by
Subcutaneously administered indomethacin dose-dependently produced lesions in the glandular stomach of sham-operated rats at 210 mg/kg body wt, but at 1 and 3 mg/kg body wt did not cause any damage in the mucosa, the lesion index being 10.9 + 2.1 and 24.0 + 3.9 mm at 10 and 25 mg/kg body wt, respectively (Table 1). The lesions induced by indomethacin were mostly confined to the corpus mucosa along the long axis of the stomach, and there was no damage in either the antrum or duodenum. The ulcerogenic activity of indomethacin, however, was markedly potentiated in adrenalectomized rats, and the lesions were observed in the gastric mucosa after administration of this agent even at 0.3 mg/kg body wt; the lesion index was 9.2 + 0.9, 15.3 2 5.1, and 34.4 ? 7.3 mm at 0.3, 1, and 3 mg/kg body wt, respectively. The severity of gastric lesions (44.2 * 6.3 mm) induced by 25 mglkg body wt of indomethacin in adrenalectomized rats was in the same range as that observed at 3 mg/kg body wt, but was significantly greater when compared with sham-operated animals given 25 mg/kg body wt. In adrenalectomized rats, indomethacin produced lesions not only in the stomach but in the duodenum as well; the incidence was 62.5% at 3 mglkg body wt. Adrenalectomy alone did not cause gross damage in the mucosa of these tissues.
Table
1.
Gastric Ulcerogenic
in
Action of lndomethacin
Sham-Operated and Adrenalectomized Rats Dose (mgikgl
Group Sham-operated rats Saline Indomethacin
Adrenalectomized Saline Indomethacin
Rats (n)
Lesion index (mm)
4
0
014
4 a a
0 3.2 iz 1.6 10.9 + 2.1 24.0 t 3.9
o/4 218 618 ala
5 4 7 7 7 5
0 0 9.2 + 15.3 ? 34.4 ? 44.2 k
o/5 o/4 717 717 717 515
1 3 10 25
a
Incidence
rats 0.1 0.3 1 3 25
0.9" 5.1" 7.3" 6.3"
Values are presented as the mean 2 SE from 4-a rats per group. Indomethacin was given subcutaneously, and the animals were killed 4 h later. ’ Statistically significant from the corresponding group in sham-operated rats, at p < 0.05.
Effects of Hydrocortisone
and Glucose
Hydrocortisone, given subcutaneously for 2 wk in a dose of 10 mg/kg body wt . day, significantly prevented gastric lesions induced by 3 mg/kg body wt of indomethacin in adrenalectomized rats (Table 2). The lesion index was redured from 33.8 rf: 5.2 to being 80.5%. Intrave6.6 ? 1.8 mm, the inhibition nous infusion of glucose (25% wt/wt, 1.2 ml/h) also produced a significant protection against these lesions in both sham-operated and adrenalectomized rats, and the degree of inhibition was almost equivalent to that observed in adrenalectomized rats after hydrocortisone treatment. Blood Glucose Levels Under Various Conditions Blood glucose levels were 134.3 5 6.3 mg/dl in sham-operated rats after 18 h fasting (Table 2). After adrenalectomy the levels were significantly decreased to 66.8 -+ 14.2 mg/dl, the values being about 50% of those observed in sham-operated rats. Subcutaneously administered indomethacin at 25 mg/kg body wt significantly reduced blood glucose levels to 89.3 5 5.7 mg/dl in sham-operated rats, although this agent at 3 mg/kg body wt did not further reduce the already low glucose levels in adrenalectomized rats. Both hydrocortisone treatment (10 mg/kg body wt . day for 2 wk) and intravenous infusion of glucose (25% wt/wt, 1.2 ml/h) significantly restored the reduced glucose levels above the normal values in adrenalectomized rats, and the latter also prevented the decrease of glucose
ADRJXNALECTOMY ON INDOMETHACIN-INDUCED
August
1989
Table
2. Effects of Hydrocortisone Treatment and Glucose Infusion Levels in Rats Treated With Indomethacin Rats Group
Sham-operated rats A. Control B. Indomethacin (25 mg/kg body wt) C. Indomethacin plus glucose infusion (25% wt/wt) Adrenalectomized rats D. Control E. Indomethacin (3 mg/kg body wt) F. Indomethacin plus glucose infusion (25% wtiwt) G. Indomethacin plus hydrocortisone (10 mg/kg body wt. day)
Gastric
on Gastric Lesions and Terminal
lesion
(nl
(mm)
6 6 8
0 18.3 t 2.1 2.1 + 1.2b
8 8 8
0 33.8 ? 5.2 10.0 * 3.1”
8
6.6 f 1.8”
Blood
Inhibition
LESIONS
287
Blood Glucose
glucose
Inhibition
(%)
(mgidl)
(%)
88.5
134.3 2 6.3 89.3 2 5.7’ 165.2 ? 14.gb
33.5 -23.0
70.4
66.8 2 14.2” 84.3 -c 9.3” 184.1 + 31.7”
50.3 37.2 -37.1
80.5
137.7
-
2 8.4’
-2.5
Values are presented as the mean + SE from 6-8 rats per group. Indomethacin was given subcutaneously, and the animals were killed 4 h later. Glucose was infused intravenously at a rate of 1.2 ml/h during a test period, and hydrocortisone was given subcutaneously for 2 wk after adrenalectomy. Statistical significance at p < 0.05: a from group A; b from group B; and ’ from the E.
levels
induced
by indomethacin
in sham-operated
(25 mgikg body wt)
animals.
Effects of Indomethacin on Gastric Acid Output and Mucosal Blood Flow In sham-operated rats, values of acid output calculated from the amount of hydrogen trapped by
were 42.8 + 3.1to 61.1k 11.8pEqll5 min and those of mucosal blood flow were 3.8 + 0.4 to 4.4-+0.5ml/l5 min (Figure 1).These values were not significantly increased or decreased by subcutaneous administration of indomethacin (25 mg/kg body wt), for at least 2 h. Bilateral adrenalectomy by itself significantly reduced both acid output and mucosal blood flow as compared with those in aminopyrine
AWENALEClQlY
ro-
1
0:
**VIE
.:
-wAcI(
1
00.
so-
gZ k;403, E$o5t < 2010-
INDONETNACIN (5 no/ha)
oi?
140
tip Qgloo
1
$6
I,,.......,
Figure
1. Effects of indomethacin (25 mg/kg body wt) on gastric acid output and mucosal blood flow in sham-operated rats, Data in acid output and volume of gastric contents are presented as the mean t SE of values determined every 15 min from 6 rats per group. Data in mucosal blood flow are expressed as percentage of the values observed before administration of indomethacin or saline, and are presented as the mean + SE of values determined every 15 min from 6 rats per group.
TIME (x 15 mln)
Figure
2. Effects of indomethacin (3 mg/kg body wt) on gastric acid output and mucosal blood flow in adrenalectomized rats. Hydrocortisone (10 mg/kg body wt. day) was given subcutaneously for 2 wk after adrenalectomy. Data are presented as the mean + SE of values determined every 15 min from 4-8 rats per group. *Statistically significant from controls (without indomethacin), at p < 0.05.
288
TAKEUCHI
300
1
ET AL.
1WAY
GASTROENTEROLOGY
prevented such time-related decreases in both acid output and mucosal blood flow, and at 1 h after indomethacin treatment the values reached significantly higher levels when compared with control animals given saline alone.
N:5
OPERATDN]
Vol. 97, No. 2
200
ii g e 100 6 I
c
500
z
!i d 5
Effect of Hydrocortisone
ADnfmMEcTouv
400
0
/ \I11 JI
I
/I3
I]
/
0 8
300
8
~YDONETWACW , ..C.,
2 t
I
200
Treatment of the adrenalectomized rats with hydrocortisone (10 mgikg body wt. day) for 2 wk significantly restored the reduced acid output and mucosal blood flow (Figure 2). The values of acid output were 36.5 ? 6.3 to 51.8 & 10.8 pEq/l5 min and remained unaltered before and after administration of indomethacin (3 mg/kg body wt). In these animals, the mucosal blood flow remained in the range of 2.9 2 0.5 to 3.6 + 0.4 ml115 min and was not significantly affected by indomethacin. Effects of Indomethacin
100
TIME (x 1Omh)
Figure
3. Effects of indomethacin on gastric motility in shamoperated (upper panel) and adrenalectomized (lower panel) rats. Data are expressed as percentage of the values observed before administration of indomethacin, and are presented as the mean 2 SE of values determined every 10 min from 5 rats per group.
sham-operated animals, and the values of acid output were in the range of 23.4 5 4.2 to 6.9 ? 1.3 pEq/l5 min, with a marked time-related decrease (Figure 2). Subcutaneously administered indomethacin (3 mg/kg body wt) to adrenalectomized rats
40,
on Gastric Motility
The stomachs in sham-operated rats contracted at a frequency of 16.8 + 3.2110 min with an amplitude of 17.6 2 1.8 cm H,O. Subcutaneous administration of indomethacin to these animals did not significantly affect gastric motility at 1 and 3 mgikg body wt but markedly augmented the amplitude of contractions at 210 mg/kg body wt [Figure 3). The amplitude of contractions started to increase about 20 min after the administration, reached plateau levels (200% of basal values) within 1 h, and remained elevated thereafter. Adrenalectomized rats responded to indomethacin at 20.3 mg!kg body wt by a significant rise in amplitude of contractions, and at 3 mgikg body wt the maximal values were
400
N: 4-8 $1 LESION
_
30,
PI MOTILITY
300
s p w d Y F al
6 ii!
$ 20
200
z 3 i? a ;
d 10
100
:
C
Figure
4. Dose-response curves of the ulcerogenic and the motility effects of indomethacin in sham-operated and adrenalectomized rats. The dose of indomethacin was plotted in a log scale, and the motility response was expressed as percentage of increase against the value observed before indomethacin treatment in each group. Data are taken from Table 1 and Figure 3.
0
0 0.1
0.3
1
DOSE OF INDOMETHACIN
3
10 (Log Scale)
25
mglkg
August
1989
ADRENALECTOMY
ON INDOMETHACIN-INDUCED
LESIONS
289
500-
*P< E r Oh
400-
N
0.05
: 5-8
INDOMETHACIN ( 3 mg1kg.s.c.
-6.0
-3-o
)
30
0 TIME
Figure
60
( min
90
120
1
and glucose on the increased gastric motility induced by indomethacin (3 mg/kg body wt) in 5. Effects of hydrocortisone adrenalectomized rats. Hydrocortisone (10 mgikg body wt day] was given subcutaneously for 2 wk after adrenalectomy, and glucose (25% wtiwt) was infused intravenously during a test period. M, saline plus indomethacin; U, hydrocortisone plus indomethacin; M. saline infusion plus indomethacin; M. glucose infusion plus indomethacin. Data are presented as the mean t SE of values determined every 10 min from 5-6 rats per group. *Statistically significant from controls, at p < 0.05.
about five times greater than basal values. Even at 0.3 mg/kg body wt, indomethacin increased contractile activity to 160% of the basal values; the degrees of enhancement were almost equivalent to those induced by 10 mgikg body wt of indomethacin in sham-operated rats. Adrenalectomy alone did not significantly alter gastric motility in terms of both frequency and amplitude. When the motility and ulcerogenic responses obtained in different rats were plotted against log dose of indomethacin, it was found that adrenalectomy shifted the dose-response curves of indomethacin to the left (lower doses) (Figure 4). In both sham-operated and adrenalectomized rats, the dose-response relationship for the motility effect was similar to that for the ulcerogenic action.
Effect of Hydrocortisone
and
Glucose
Indomethacin at 3 mgikg body wt markedly enhanced gastric motility in adrenalectomized rats, and the amplitude of contractions reached the maximal values (432% 2 84% of basal values) within 70 min after the administration (Figure 5). Hydrocortisone (10 mg/kg body wt day for 2 wk) significantly inhibited the increased motility responses caused by indomethacin, and the amplitude of contractions remained in the same range (78% ? 4% to 128% * 43% of basal values] before and after indomethacin treatment (Figures 5 and 6). Hypermotility caused by
indomethacin was also significantly inhibited by intravenous infusion of glucose (25% wt/wt, 1.2 ml/h). In the animals infused with glucose, spontaneous gastric contractions were potently suppressed and did not respond to later administration of indomethacin by a significant increase in amplitude. The inhibitory effect of glucose on the motility was also seen in sham-operated rats given 25 mg/kg body wt of indomethacin (not shown).
Relationship Between Blood Glucose Level
Lesion,
Motility,
and
To further investigate roles of gastric hypermotility in the genesis of indomethacin-induced gastric lesions, we examined the relationship between lesion, motility, and terminal blood glucose level in the same animals of both sham-operated and adrenalectomized groups. Indomethacin dose-dependently produced gastric lesions in both groups of rats, in association with enhancement of the motility, and the severity of lesions was not significantly different from that observed in the stomach without a balloon (Table 3). When changes in the motility were plotted against the severity of lesion or the terminal blood glucose level, there was a significant relationship between the lesion and the motility in both groups, the correlation coefficient being 0.874
290
TAKEUCHI
ET AL.
GASTROENTEROLOGY
60 ADRENALECTOMl2ED
RATS
A : CONTROL
B: c
4
1
3
HYDROCORTISONE TREATMENT : OLUCOSE TREATMENT
Cm)60
40
t
Vol. 97, No. 2
-
I
C
OLUCOIE
( 26% , I.v. )
E 2
20
$
01
1
I
1
1
-30
-60
1
0
1
30
I
I
60
80
TIME ( nun1 Figure
(Figure
6. Representative figures showing the effect of indomethacin control; B, hydrocortisone treatment; C, glucose infusion. enhanced gastric motility induced by indomethacin.
7). Although
a significant
relationship
(3 mgikg body wt) on gastric motility in adrenalectomized rats. A, Note that both hydrocortisone and glucose significantly inhibited the
was
indomethacin
noted between the lesion and the bIood glucose level in sham-operated rats given various doses of indomethacin, there was no such relation in adrenalectomized rats. The blood glucose level remained in the same range in response to different doses of
gland
Table
Glucose
3. Gastric Lesion, Hypermotility, and Terminal Following lndomethacin Treatment
Indomethacin (mg/kg body wt) Sham-operated 3 10 25 Adrenalectomized 1 3
3 mg/kg body wt), despite in the lesion index.
(1 and
marked difference
a
Discussion
Blood
Rats
Lesion index
(n)
(mm)
Although a relationship between the adrenal and gastric lesions has been postulated in
in Sham-Operated
and Adrenalectomized
Hypermotility (% increase)
Rats
Blood glucose (mgldl)
rats 5 4 5
0.8 2 0.5 13.8 L 4.4 39.4 ? 6.3
26.1 + 1.9 96.5 + 32.2 178.2 t 36.9
116.5 2 6.7 112.3 + 13.2 79.9 ? 10.2
4 4
18.0 k 7.0 41.0 k 9.7
75.8 2 21.9 198.8 2 16.4
36.4 t 7.2 32.5 + 7.5
rats
Values are presented as the mean + SE from 4-5 rats per group. Indomethacin was given subcutaneously, and the animals were killed 4 h later. Hypermotility was expressed as the percent increase of the amplitude observed during 4 h against basal values, and the blood glucose levels were determined 4 h after indomethacin treatment. These measurements were performed in the same animals of sham-operated and adrenalectomized groups in response to various doses of indomethacin.
August
1989
ADRENALECTOMY
8 : 22
0
f : 8.814
, ,’ A
0
A'
,9’
0 I
IX
??
A,
,’
,’
,’
u
0
0 0
.li
i
1;
i8
3.8 48 LE8I88S ml
i8
88
148 = 128 z P I88 z p 88 S
0 A 0
A
88 ’ : -8.882
48 ‘
8
Figure
A
I8
28
88
48
88
8
88
7. Relationship between ulcerogenesis, motility, and terminal blood glucose level in the same animals of sham-operated and adrenalectomized groups. A highly significant relationship was noted between ulcerogenesis and motility in both groups of rats, the correlation being 0.874.
humans (131, the influences of adrenalectomy on experimental lesions in the rat gastric mucosa remains controversial (6,7,14). The present study confirmed the finding by Urushidani et al. (73, who showed potentiation by adrenalectomy of the mucosal ulcerogenic response induced by indomethacin, and for the first time demonstrated that the sensitization of the gastric motor system to indomethacin may be causally related to aggravation of these lesions in adrenalectomized rats. Recent studies showed gastric hypermotility as an important element in the pathogenesis of indomethacin-induced gastric mucosal lesions in rats (35). A deficiency of prostaglandins, if involved, may not be directly associated with formation of gross lesions induced by this agent, as indomethacin at a lower dose (5 mg/kg body wt), although showing >80% inhibition of cyclooxygenase activity, did not produce any damage in the mucosa (4). As shown in the present study, indomethacin at an ulcerogenic dose (25 mg/kg body wt) did not significantly alter either acid output or mucosal blood flow in shamoperated rats. Therefore, it is unlikely that alterations in these factors account for development of
ON INDOMETHACIN-INDUCED
LESIONS
291
the lesions. As expected, indomethacin produced a marked elevation in gastric motility at the ulcerogenie doses in both sham-operated and adrenalectomized rats, and there was a highly significant relationship between the motility and the ulcer score in these animals. The previous study showed that both atropine and a prostaglandin inhibited gastric hypermotility induced by indomethacin and prevented the lesions even in the presence of exogenous acid (3,s). The present study also showed that hydrocortisone treatment or glucose infusion significantly attenuated the enhanced motility and the mucosal ulcerogenic responses caused by this agent in adrenalectomized rats. Thus, these results strongly suggest that the motility alterations may be closely associated with the mucosal ulcerogenesis following administration of indomethacin. How does bilateral adrenalectomy worsen gastric lesions induced by indomethacin? The absence of adrenal glands results in altered function in many parts of the body, and much attention has been focused on the gastrointestinal tract (15). As the medulectomy had no effect on these lesions (73, and because hydrocortisone mitigated the above phenomenon, the worsening effect of adrenalectomy may be largely due to removal of the cortex (steroid secretion). Participation of acid hypersecretion may be excluded in the mechanism for aggravation by adrenalectomy based on the following reasons. First, acid output was significantly decreased in adrenalectomized rats. Second, hydrocortisone treatment significantly restored the reduced acid output in these animals, yet prevented indomethacin-induced lesions. Other studies showed gastric cellular atrophy (16) or insufficient vasomotion of the small vessels (17) following adrenalectomy. The latter may lead to a defective blood supply to the stomach, and supports the present observation that the mucosal blood flow was significantly reduced in the adrenalectomized rats. The severity of gastric lesions induced by aspirin was significantly potentiated in the hypotrophic mucosa of rats fed liquid diets and by inducing hypogastrinemia (18). Thus, these factors may contribute to the aggravation process of the lesions in adrenalectomized rats, probably by stimulating the experimental circumstance for lesions in response to indomethacin. Certainly, as adrenalectomy alone did not induce any damage in the mucosa, a global mucosal atrophy or a defective blood supply by itself cannot explain the occurrence of focal lesions in the stomach following administration of indomethacin. Although the mechanism by which indomethacin induced gastric hypermotility remains unclear, this may be associated with the effect of indomethacin on the central nervous system and involves stimulation
292
TAKEUCHI
GASTROENTEROLOGY
ET AL.
of the vagus nerves, based on previous findings (3,5,19); gastric hypermotility induced by indomethacin was inhibited by both vagotomy and atropine treatment, and the electrical stimulation of the vagus nerves induces hypermotility and mucosal damages, similar to indomethacin. On the other hand, Mersereau and Hinchey (11) postulated that uncoupling in the glycoprivic receptor system may be responsible for the motility effect of nonsteroidal antiinflammatory drugs. They demonstrated that phenylbutazone, by perturbing energy metabolism in specific receptor cells, provoked a vagally mediated, glucosesensitive gastric contractile response in the rat. Indomethacin, like other nonsteroidal antiinflammatory drugs, may act by uncoupling oxidative phosphorylation, and as a result delivers the glycoprivic stimulus. As expected, higher doses of indomethacin produced significant hypoglycemia and hypermotility in sham-operated rats, and these responses were significantly blocked by exogenous glucose infusion. Thus, the motility effect of indomethacin (910 mgikg body wt) in sham-operated rats may involve the same receptor system in its mechanism. The most important aspect of this study is how adrenalectomy sensitized the motility system to indomethacin. The dose-response curve of indomethacin for the motility shifted to the left in adrenalectomized rats; the value of ED,, was about 10 times lower than that in sham-operated rats. As adrenalectomy alone induced a profound hypoglycemia but did not provoke either hypermotility or mucosal lesion, the hypoglycemia induced by adrenalectomy may not affect the cerebral glucoprivation by a critical amount, Nevertheless, the hypoglycemic stimulus appears to be important in the sensitization of the motility system to indomethacin following adrenalectomy, as both hydrocortisone treatment and glucose infusion blocked the motility responses caused by indomethacin and completely reverted the reduced blood glucose levels in these animals. It has been shown that a lower dose of indomethacin (5 mg/kg body wt) potentiated the motility and the ulcerogenic responses induced by insulin or 2deoxy-o-glucose, an agent known to deliver the glycoprivic stimulus (20,21). As indomethacin at doses as low as 1 mg/kg body wt reduces prostaglandin biosynthesis >80% in the central nervous system as well as in the periphery (4,22), it is possible that a prostaglandin deficiency caused by indomethacin may facilitate the excitation of the glycoprivic system in response to hypoglycemia induced by adrenalectomy. In any case, these results suggest that hypoglycemia may be crucial in the mechanism related to the sensitization of the motility system to indomethacin in rats following adrenalectomy.
Vol. 97, No. 2
In summary, the present study together with previous findings (3-5) suggests that gastric motility may play a key role in the pathogenesis of indomethacin-induced gastric lesions in rats, and may be associated with the worsening of the lesions by adrenalectomy. Abrasion of adrenal glands (cortex) by inducing hypoglycemia may sensitize the system to indomethacin and increase gastric motility and mucosal ulcerogenic responses. As insufficiency of adrenocortical activity was seen after severe stress or in association with diseases such as tuberculosis and cancer, treatment with nonsteroidal antiinflammatory drugs in such patients should be applied cautiously to avoid serious damage to the stomach. References 1. Robert A. Prostaglandins and the gastrointestinal tract. In: Johnson LR, Christensen J, Grossman MI, Jacobson ED, Schultz SG, eds. Physiology of the gastrointestinal tract. New York: Raven, 1981:1407-34. 2. Whittle BJR. Temporal relationship between cyclooxygenase inhibition, as measured by prostacyclin biosynthesis, and the gastrointestinal damage induced by indomethacin in the rat. Gastroenterology 1981;80:94-a. 3. Mersereau WA, Hinchey EJ. Prevention of indomethacininduced gastric hypercontractility; a mucosal protective mechanism of prostaglandin E, (abstr). Gastroenterology ~980:78:1221. 4. Takeuchi K, Ueki S. Okabe S. Importance of gastric motility in the pathogenesis of indomethacin-induced gastric lesions in rats. Dig Dis Sci 1986;31:1114-22. K, Okabe S. Gastric motility is an important 5. Ueki S, Takeuchi factor in pathogenesis of indomethacin-induced gastric mucosal lesions in rats. Dig Dis Sci 1988;33:209-16. KP, Gupta MB, Tangri KK. Mechanism of ulcero6. Bhargawa genie activity of indomethacin and oxyphenbutazone. Eur J Pharmacol 1973;22:191-5. 7. Urushidani T, Kasuya Y, Okabe S. The mechanism of aggravation of indomethacin-induced gastric ulcers by adrenalectomy in the rat. Jpn J Pharmacol 1979;29:775-80. a. Jacobson ED, Linford RH, Grossman MI. Gastric secretion in relation to mucosal blood flow studied by a clearance technique. J Clin Invest 1966;45:1-13. 9. Kitagawa H, Fujiwara M, Osumi Y. Effect of water-immersion stress on gastric secretion and mucosal blood flow in rats. Gastroenterology 1979;77:298-302. (pyramidon) in 10. Brodie BB, Axelrod J. The fate of aminopyrine man and methods for the estimation of aminopyrine and its metabolites in biological materials. J Pharmacol Exp Ther 1950;99:171-4. 11. Mersereau WA, Hinchey EJ. Prevention of phenylbutazone ulcer in the rat by glucose: role of a glycoprivic receptor system. Am J Physiol 1982;242:G429-32. 12. Dunnett CW. A multiple comparison procedure for comparing several treatments with a control. Am J Stat Assoc 1955;50:1096-121. 13. Shapiro H, Britt LG. The effect of adrenalectomy on the gastrointestinal tract. A review of the literature. Rev Surg 1972:29:229-39. 14. Diahangujri B, Hemmati M, Abtahi F, Farrokhoslan M. Preventive effects on indomethacin-induced gastric ulceration in the rat of disulfiram, phenoxybenzamine and bile duct liga-
August 1989
ADRENALECTOMY ON INDOMETHACIN-INDUCED LESIONS
tion. In: Gheorghiu J, ed. Experimental ulcer. Cologne: Gehard Witzstrock, 1975:31-5. 15. Cook AR. Role of adrenocortical steroids in the regulation of gastric secretion. Progress in gastroenterology. Gastroenterology 1967;52:272-81. 16. Crean GP. A comparison between the effects of hypophysectomy and adrenalectomy on the gastric mucosa of the rat. Gut 1968;9:343-7. 17. Takeuchi K, Johnson LR. Effect of cell proliferation and loss on aspirin-induced gastric damage in the rat. Am J Physiol 1982;243:G463-8. 18. Ramey ER, Goldstein MS, Levine R. Action of norepinephrine and adrenal cortical steroids on blood pressure and work performance of adrenalectomized dogs. Am J Physiol 1951; 165:450-5. 19. Ohno T, Uramoto H, Ishihara T, Okabe S. Effects of 16,16dimethyl prostaglandin E, on surface epithelial cell damage
293
in the rat stomach induced by vagal nerve stimulation. Jpn J Pharmacol 1987;43:429-39. 20. Mersereau WA, Hinchey EJ. The role of mechanical factors in the genesis of gastric ulcers in the indomethacin-insulin treated rat (abstr). Dig Dis Sci 1985;30:388. 21. Okada M, Niida H, Takeuchi K, Okabe S. Role of prostaglandin deficiency in the pathogenetic mechanism of gastric lesions induced by indomethacin in rats. Dig Dis Sci (in press). 22. Fargeas MJ, Fioramonti J, Bueno L. Prostaglandin E,; a neuromodulator in the central control of gastrointestinal motility and feeding behavior by calcitonin. Science 1984;225:1059-2.
Received January 29, 1988. Accepted February 9, 1989. Address requests for reprints to: Dr. Koji Takeuchi, Department of Applied Pharmacology, Kyoto Pharmaceutical University, Misasagi, Yamashina, Kyoto 607, Japan.
1989;97:284-93
Bilateral Adrenalectomy Worsens Gastric Mucosal Lesions Induced by Indomethacin in the Rat Role of Enhanced
Gastric Motility
KOJI TAKEUCHI, HIDEYUKI NISHIWAKI, MEGUMU HIROMICHI NIIDA, and SUSUMU OKABE
OKADA,
Department of Applied Pharmacology, Kyoto, Japan
Misasagi, Yamashina,
Kyoto Pharmaceutical
The mechanism by which bilateral adrenalectomy worsens indomethacin-induced gastric lesions was investigated in rats. In sham-operated rats subcutaneously administered indomethacin produced gastric lesions at doses of 10 mg/kg body wt or greater, in association with lowering of blood glucose levels. In a parallel study, indomethacin induced gastric hypermotility at the same dose levels but had no effect on acid output or mucosal blood flow even at 25 mg/kg body wt. Adrenalectomy (2 wk) itself significantly reduced the blood glucose levels (-50%) and markedly potentiated the ulcerogenic and motility responses caused by indomethacin; the ED,, values dropped to -10 times lower than those in sham-operated rats. Both acid output and mucosal blood flow were significantly reduced by adrenalectomy, but these values were increased after indomethacin treatment (3 mg/kg body wt). The ulcerogenic and motility responses caused by indomethacin were significantly reduced by acute infusion of glucose (25% wtht, 1.2ml/h) intravenously in both sham-operated and adrenalectomized rats, and by subcutaneous administration of hydrocortisone acetate (10 mg/kg body wt for 2 wk) in the latter group. When the motility and the ulcer score were determined in the same animals, a highly significant relationship was found between these two factors in both sham-operated and adrenalectomized rats. These results suggest that (a) the increased gastric motility may be a key element in the pathogenesis of indomethacin-induced lesions and in the mechanism for aggravation of the lesions by adrenalectomy, and (b) abrasion of adrenal glands by inducing hypoglycemia may sensitize the system to indomethacin and increase gastric motility.
University,
ndomethacin, a nonsteroidal antiinflammatory drug, is known to induce damage in the gastrointestinal mucosa of experimental animals and humans. The mechanism by which indomethacin induces gastric lesions is generally considered to involve depletion of endogenous prostaglandins (1,Z). However, recent studies showed the importance of gastric hypermotility in the pathogenetic mechanism of these lesions in response to indomethacin (3-5). Adrenalectomy has been demonstrated to worsen these lesions in rats (6,7), although the detailed mechanisms for aggravation remain unknown. If the motor alterations are causally related to formation of gastric lesions induced by indomethacin, adrenalectomy might induce gastric motility alterations by itself or increase the motility response to indomethacin, thereby worsening gastric lesions. The present study was therefore designed (a) to examine the effects of indomethacin on gastric acid output, mucosal blood flow, motility, and blood glucose level in adrenalectomized and sham-operated rats, and (b) to correlate these effects with the ulcerogenic property of indomethacin.
I
Materials
and
Methods
Male Sprague-Dawley rats (230-250 g) were used. All studies were carried out using 4-8 rats per group. General Group
Protocol
Experiments A consisted
were performed of sham-operated
0 1989 by the American
in three groups. rats, and group B of
Gastroenterological 0016.5085/89/$3.50
Association
ADRENALECTOMY ON INDOMETHACIN-INDUCED LESIONS 285
August 1989
adrenalectomized rats. In group B, bilateral adrenalectomy was performed by a translumbar route under ether anesthesia, and the adrenal glands were removed carefully with forceps. The animals were then fed on normal laboratory diet and 2% NaCl solution after operation. Some animals in group B received hydrocortisone acetate (Sigma Chemicals, St. Louis, MO.) subcutaneously in a dose of 10 mgikg body wt once daily for 2 wk after adrenalectomy (group C). In group A, bilateral body walls of the lumbar portion were incised, the incisions were closed without removing the adrenal glands, and rats were fed on normal diet and tap water. The animals in each group were used in the following experiments 14 days after the above operation, and they were deprived of food but allowed free access to tap water (group A) or 2% NaCl (groups B and C) Although body weight for 18 h before the experiments. was increased to lesser degrees in groups B and C as compared with group A, these animals tolerated the operation. Induction
of, Gastric
Lesions
The animals were given indomethacin [Sigma), suspended in saline with a drop of Tween 80 (Nakarai, Kyoto, Japan), in doses of l-25 mg/kg body wt in group A, in doses of 0.1-25mgikg body wt in group B, and in a dose of 3 mg/kg body wt in group C. In some cases, the effect of glucose infusion on gastric lesions induced by indomethacin was examined in groups A and B. The animals were kept in Bollman cages, and glucose (25% solution, wtiwt) was infused intravenously through the tail vein at a rate of 1.2 ml/h, starting 1 h before indomethacin treatment. Four hours later, the animals were killed, and the stomachs were removed, inflated by injecting 10 ml of 2% formalin, immersed in 2% formalin for 10 min to fix the gastric wall, and opened along the greater curvature. The length of each lesion developed in the glandular mucosa was measured under a dissecting microscope with a square grid (x10), summed per stomach, and used as a lesion index. In each study, the person measuring the lesions did not know the treatment given to the animals. In the above study, the terminal blood glucose levels were also determined in the animals of groups A, B, and C with or without indomethacin treatment. Four hours after indomethacin treatment, blood was collected from the descending aorta, and blood glucose levels were measured chemically using 0-toluidine boric acid reagent (Glucose test kit; Wako, Osaka, Japan). The absorbance was measured at 635 nm on a Hitachi spectrophotometer (model 200-100). In a preliminary study, a calibration curve was made using the standard glucose solution (Wako; 50-500 mgidl) and a linearity was obtained in the above range (r = 0.9999). Determination of glucose was performed in duplicate, and the results were expressed as milligrams per deciliter. Determination Acid Output
of Mucosal
Blood
Flow
and
Gastric mucosal blood flow was measured using the aminopyrine clearance method described by Jacobson
et al. (8)and modified by Kitagawa et al. (9). A constant plasma aminopyrine level was maintained by continuous infusion of aminopyrine (Sigma: 6.6mg/kg body wt h) at a rate of 1.2 ml/h, starting 30 min after a single dose of 30 mgikg body wt given intravenously by a bolus injection. Under ether anesthesia, the abdomen was incised and both the stomach and duodenum were exposed. A polyethylene tube (inside diameter 3 mm) was inserted into the stomach through the pylorus from a small incision made in the duodenum, and the tube was placed by a ligature around the pylorus and withdrawn through the lateral abdominal wall. After recovery from anesthesia, the animals were kept in Bollman cages during the test period. The stomachs were filled with 2 ml of gastric solution of the following compositions: 4.47 g n-( -)-mannitol and 0.45g glycine in 100 ml of distilled water adjusted to pH 3.5 by adding 0.1 N HCl. The gastric solution was changed every 15 min through the fistula for a total period of 3 h. In the present study, the plasma aminopyrine levels 60 min after the onset of aminopyrine infusion and immediately after the end of the experiment were 18.8 t 2.3 and 16.2 ? 1.9 pgiml, respectively (n = 9). The plasma level of aminopyrine at each 15-min interval during the experimental period was thus estimated by interpolating between two measured points and was paired with each aminopyrine determination in the gastric juice. The content of aminopyrine in the plasma and gastric juice was measured according to the method of Brodie and Axelrod (10). Gastric mucosal blood flow was determined every 15 min and expressed as milliliters of blood per 15 min. The collected gastric juice was analyzed for volume and titrated with 0.1 N NaOH to pH 7.0 for titratable acidity. Acid output was calculated by subtracting the acidity of the initial gastric juice (glycine buffered solution) from the titratable value of each final sample, and expressed as microequivalents per 15 min. Approximately 1 h after basal acid output had stabilized, the animals were given indomethacin subcutaneously in a dose of 25 mg/kg body wt (group A) or 3 mg/kg body wt (groups B and C), and the measurement was continued for 2 h thereafter.
Determination
of Gastric
Motility
Gastric motility was measured in the animals of groups A, B, and C, using a balloon according to a previous paper (4). Briefly, under ether anesthesia a balloon (containing -0.8 ml of water) and the support catheter were placed in the glandular stomach through an incision of the forestomach. The animals were then placed in Bollman cages, and the support catheter was connected to a pressure transducer and polygraph device (Nihon Koden, Ibaragi, Japan). Gastric motility was continuously monitored on a Hitachi recorder (model 056; Hitachi, Mito, Japan) as intraluminal pressure recordings. Quantitative analysis of the motility was performed by counting the number of contractions with an amplitude of 15 cm H,O or greater, and by measuring the amplitude of each contraction over a lo-min period, determining the mean of each rat for this period from these values, and then by calculating the mean for each time period from 4-6 rats per group.
286
TAKEUCHI ET AL.
GASTROENTEROLOGY Vol. 97, No. 2
Approximately 1 h after basal motility had stabilized, indomethacin was given subcutaneously in doses of l-25 mg/kg body wt, 0.1-S mg/kg body wt, and 3 mgikg body wt, respectively, to group A, B, and C, and gastric motility was measured for 2 h thereafter. In some cases, the effect of glucose (25% wffwt) on gastric motility was examined in groups A and B. Glucose was infused intravenously through the tail vein at a rate of 1.2 ml/h during a test period, starting 60 min before administration of indomethacin (11). In another experiment, the motility, the lesion index, and the terminal blood glucose were determined in the same animals of groups A and B in response to various doses of indomethacin. In these cases, the motility was recorded for 4 h after indomethacin treatment, and the lesion index and the glucose level were determined as described previously.
Statistics Data are presented as the mean * SE from 4-8 rats per group. Statistical analysis was performed using a two-tailed Dunnett’s multiple comparison test (12), and values of p < 0.05 were regarded as significant. A regression analysis was used to determine a correlation coefficient between two different variates.
Results Development lndomethacin
of Gastric Lesions Induced
by
Subcutaneously administered indomethacin dose-dependently produced lesions in the glandular stomach of sham-operated rats at 210 mg/kg body wt, but at 1 and 3 mg/kg body wt did not cause any damage in the mucosa, the lesion index being 10.9 + 2.1 and 24.0 + 3.9 mm at 10 and 25 mg/kg body wt, respectively (Table 1). The lesions induced by indomethacin were mostly confined to the corpus mucosa along the long axis of the stomach, and there was no damage in either the antrum or duodenum. The ulcerogenic activity of indomethacin, however, was markedly potentiated in adrenalectomized rats, and the lesions were observed in the gastric mucosa after administration of this agent even at 0.3 mg/kg body wt; the lesion index was 9.2 + 0.9, 15.3 2 5.1, and 34.4 ? 7.3 mm at 0.3, 1, and 3 mg/kg body wt, respectively. The severity of gastric lesions (44.2 * 6.3 mm) induced by 25 mglkg body wt of indomethacin in adrenalectomized rats was in the same range as that observed at 3 mg/kg body wt, but was significantly greater when compared with sham-operated animals given 25 mg/kg body wt. In adrenalectomized rats, indomethacin produced lesions not only in the stomach but in the duodenum as well; the incidence was 62.5% at 3 mglkg body wt. Adrenalectomy alone did not cause gross damage in the mucosa of these tissues.
Table
1.
Gastric Ulcerogenic
in
Action of lndomethacin
Sham-Operated and Adrenalectomized Rats Dose (mgikgl
Group Sham-operated rats Saline Indomethacin
Adrenalectomized Saline Indomethacin
Rats (n)
Lesion index (mm)
4
0
014
4 a a
0 3.2 iz 1.6 10.9 + 2.1 24.0 t 3.9
o/4 218 618 ala
5 4 7 7 7 5
0 0 9.2 + 15.3 ? 34.4 ? 44.2 k
o/5 o/4 717 717 717 515
1 3 10 25
a
Incidence
rats 0.1 0.3 1 3 25
0.9" 5.1" 7.3" 6.3"
Values are presented as the mean 2 SE from 4-a rats per group. Indomethacin was given subcutaneously, and the animals were killed 4 h later. ’ Statistically significant from the corresponding group in sham-operated rats, at p < 0.05.
Effects of Hydrocortisone
and Glucose
Hydrocortisone, given subcutaneously for 2 wk in a dose of 10 mg/kg body wt . day, significantly prevented gastric lesions induced by 3 mg/kg body wt of indomethacin in adrenalectomized rats (Table 2). The lesion index was redured from 33.8 rf: 5.2 to being 80.5%. Intrave6.6 ? 1.8 mm, the inhibition nous infusion of glucose (25% wt/wt, 1.2 ml/h) also produced a significant protection against these lesions in both sham-operated and adrenalectomized rats, and the degree of inhibition was almost equivalent to that observed in adrenalectomized rats after hydrocortisone treatment. Blood Glucose Levels Under Various Conditions Blood glucose levels were 134.3 5 6.3 mg/dl in sham-operated rats after 18 h fasting (Table 2). After adrenalectomy the levels were significantly decreased to 66.8 -+ 14.2 mg/dl, the values being about 50% of those observed in sham-operated rats. Subcutaneously administered indomethacin at 25 mg/kg body wt significantly reduced blood glucose levels to 89.3 5 5.7 mg/dl in sham-operated rats, although this agent at 3 mg/kg body wt did not further reduce the already low glucose levels in adrenalectomized rats. Both hydrocortisone treatment (10 mg/kg body wt . day for 2 wk) and intravenous infusion of glucose (25% wt/wt, 1.2 ml/h) significantly restored the reduced glucose levels above the normal values in adrenalectomized rats, and the latter also prevented the decrease of glucose
ADRJXNALECTOMY ON INDOMETHACIN-INDUCED
August
1989
Table
2. Effects of Hydrocortisone Treatment and Glucose Infusion Levels in Rats Treated With Indomethacin Rats Group
Sham-operated rats A. Control B. Indomethacin (25 mg/kg body wt) C. Indomethacin plus glucose infusion (25% wt/wt) Adrenalectomized rats D. Control E. Indomethacin (3 mg/kg body wt) F. Indomethacin plus glucose infusion (25% wtiwt) G. Indomethacin plus hydrocortisone (10 mg/kg body wt. day)
Gastric
on Gastric Lesions and Terminal
lesion
(nl
(mm)
6 6 8
0 18.3 t 2.1 2.1 + 1.2b
8 8 8
0 33.8 ? 5.2 10.0 * 3.1”
8
6.6 f 1.8”
Blood
Inhibition
LESIONS
287
Blood Glucose
glucose
Inhibition
(%)
(mgidl)
(%)
88.5
134.3 2 6.3 89.3 2 5.7’ 165.2 ? 14.gb
33.5 -23.0
70.4
66.8 2 14.2” 84.3 -c 9.3” 184.1 + 31.7”
50.3 37.2 -37.1
80.5
137.7
-
2 8.4’
-2.5
Values are presented as the mean + SE from 6-8 rats per group. Indomethacin was given subcutaneously, and the animals were killed 4 h later. Glucose was infused intravenously at a rate of 1.2 ml/h during a test period, and hydrocortisone was given subcutaneously for 2 wk after adrenalectomy. Statistical significance at p < 0.05: a from group A; b from group B; and ’ from the E.
levels
induced
by indomethacin
in sham-operated
(25 mgikg body wt)
animals.
Effects of Indomethacin on Gastric Acid Output and Mucosal Blood Flow In sham-operated rats, values of acid output calculated from the amount of hydrogen trapped by
were 42.8 + 3.1to 61.1k 11.8pEqll5 min and those of mucosal blood flow were 3.8 + 0.4 to 4.4-+0.5ml/l5 min (Figure 1).These values were not significantly increased or decreased by subcutaneous administration of indomethacin (25 mg/kg body wt), for at least 2 h. Bilateral adrenalectomy by itself significantly reduced both acid output and mucosal blood flow as compared with those in aminopyrine
AWENALEClQlY
ro-
1
0:
**VIE
.:
-wAcI(
1
00.
so-
gZ k;403, E$o5t < 2010-
INDONETNACIN (5 no/ha)
oi?
140
tip Qgloo
1
$6
I,,.......,
Figure
1. Effects of indomethacin (25 mg/kg body wt) on gastric acid output and mucosal blood flow in sham-operated rats, Data in acid output and volume of gastric contents are presented as the mean t SE of values determined every 15 min from 6 rats per group. Data in mucosal blood flow are expressed as percentage of the values observed before administration of indomethacin or saline, and are presented as the mean + SE of values determined every 15 min from 6 rats per group.
TIME (x 15 mln)
Figure
2. Effects of indomethacin (3 mg/kg body wt) on gastric acid output and mucosal blood flow in adrenalectomized rats. Hydrocortisone (10 mg/kg body wt. day) was given subcutaneously for 2 wk after adrenalectomy. Data are presented as the mean + SE of values determined every 15 min from 4-8 rats per group. *Statistically significant from controls (without indomethacin), at p < 0.05.
288
TAKEUCHI
300
1
ET AL.
1WAY
GASTROENTEROLOGY
prevented such time-related decreases in both acid output and mucosal blood flow, and at 1 h after indomethacin treatment the values reached significantly higher levels when compared with control animals given saline alone.
N:5
OPERATDN]
Vol. 97, No. 2
200
ii g e 100 6 I
c
500
z
!i d 5
Effect of Hydrocortisone
ADnfmMEcTouv
400
0
/ \I11 JI
I
/I3
I]
/
0 8
300
8
~YDONETWACW , ..C.,
2 t
I
200
Treatment of the adrenalectomized rats with hydrocortisone (10 mgikg body wt. day) for 2 wk significantly restored the reduced acid output and mucosal blood flow (Figure 2). The values of acid output were 36.5 ? 6.3 to 51.8 & 10.8 pEq/l5 min and remained unaltered before and after administration of indomethacin (3 mg/kg body wt). In these animals, the mucosal blood flow remained in the range of 2.9 2 0.5 to 3.6 + 0.4 ml115 min and was not significantly affected by indomethacin. Effects of Indomethacin
100
TIME (x 1Omh)
Figure
3. Effects of indomethacin on gastric motility in shamoperated (upper panel) and adrenalectomized (lower panel) rats. Data are expressed as percentage of the values observed before administration of indomethacin, and are presented as the mean 2 SE of values determined every 10 min from 5 rats per group.
sham-operated animals, and the values of acid output were in the range of 23.4 5 4.2 to 6.9 ? 1.3 pEq/l5 min, with a marked time-related decrease (Figure 2). Subcutaneously administered indomethacin (3 mg/kg body wt) to adrenalectomized rats
40,
on Gastric Motility
The stomachs in sham-operated rats contracted at a frequency of 16.8 + 3.2110 min with an amplitude of 17.6 2 1.8 cm H,O. Subcutaneous administration of indomethacin to these animals did not significantly affect gastric motility at 1 and 3 mgikg body wt but markedly augmented the amplitude of contractions at 210 mg/kg body wt [Figure 3). The amplitude of contractions started to increase about 20 min after the administration, reached plateau levels (200% of basal values) within 1 h, and remained elevated thereafter. Adrenalectomized rats responded to indomethacin at 20.3 mg!kg body wt by a significant rise in amplitude of contractions, and at 3 mgikg body wt the maximal values were
400
N: 4-8 $1 LESION
_
30,
PI MOTILITY
300
s p w d Y F al
6 ii!
$ 20
200
z 3 i? a ;
d 10
100
:
C
Figure
4. Dose-response curves of the ulcerogenic and the motility effects of indomethacin in sham-operated and adrenalectomized rats. The dose of indomethacin was plotted in a log scale, and the motility response was expressed as percentage of increase against the value observed before indomethacin treatment in each group. Data are taken from Table 1 and Figure 3.
0
0 0.1
0.3
1
DOSE OF INDOMETHACIN
3
10 (Log Scale)
25
mglkg
August
1989
ADRENALECTOMY
ON INDOMETHACIN-INDUCED
LESIONS
289
500-
*P< E r Oh
400-
N
0.05
: 5-8
INDOMETHACIN ( 3 mg1kg.s.c.
-6.0
-3-o
)
30
0 TIME
Figure
60
( min
90
120
1
and glucose on the increased gastric motility induced by indomethacin (3 mg/kg body wt) in 5. Effects of hydrocortisone adrenalectomized rats. Hydrocortisone (10 mgikg body wt day] was given subcutaneously for 2 wk after adrenalectomy, and glucose (25% wtiwt) was infused intravenously during a test period. M, saline plus indomethacin; U, hydrocortisone plus indomethacin; M. saline infusion plus indomethacin; M. glucose infusion plus indomethacin. Data are presented as the mean t SE of values determined every 10 min from 5-6 rats per group. *Statistically significant from controls, at p < 0.05.
about five times greater than basal values. Even at 0.3 mg/kg body wt, indomethacin increased contractile activity to 160% of the basal values; the degrees of enhancement were almost equivalent to those induced by 10 mgikg body wt of indomethacin in sham-operated rats. Adrenalectomy alone did not significantly alter gastric motility in terms of both frequency and amplitude. When the motility and ulcerogenic responses obtained in different rats were plotted against log dose of indomethacin, it was found that adrenalectomy shifted the dose-response curves of indomethacin to the left (lower doses) (Figure 4). In both sham-operated and adrenalectomized rats, the dose-response relationship for the motility effect was similar to that for the ulcerogenic action.
Effect of Hydrocortisone
and
Glucose
Indomethacin at 3 mgikg body wt markedly enhanced gastric motility in adrenalectomized rats, and the amplitude of contractions reached the maximal values (432% 2 84% of basal values) within 70 min after the administration (Figure 5). Hydrocortisone (10 mg/kg body wt day for 2 wk) significantly inhibited the increased motility responses caused by indomethacin, and the amplitude of contractions remained in the same range (78% ? 4% to 128% * 43% of basal values] before and after indomethacin treatment (Figures 5 and 6). Hypermotility caused by
indomethacin was also significantly inhibited by intravenous infusion of glucose (25% wt/wt, 1.2 ml/h). In the animals infused with glucose, spontaneous gastric contractions were potently suppressed and did not respond to later administration of indomethacin by a significant increase in amplitude. The inhibitory effect of glucose on the motility was also seen in sham-operated rats given 25 mg/kg body wt of indomethacin (not shown).
Relationship Between Blood Glucose Level
Lesion,
Motility,
and
To further investigate roles of gastric hypermotility in the genesis of indomethacin-induced gastric lesions, we examined the relationship between lesion, motility, and terminal blood glucose level in the same animals of both sham-operated and adrenalectomized groups. Indomethacin dose-dependently produced gastric lesions in both groups of rats, in association with enhancement of the motility, and the severity of lesions was not significantly different from that observed in the stomach without a balloon (Table 3). When changes in the motility were plotted against the severity of lesion or the terminal blood glucose level, there was a significant relationship between the lesion and the motility in both groups, the correlation coefficient being 0.874
290
TAKEUCHI
ET AL.
GASTROENTEROLOGY
60 ADRENALECTOMl2ED
RATS
A : CONTROL
B: c
4
1
3
HYDROCORTISONE TREATMENT : OLUCOSE TREATMENT
Cm)60
40
t
Vol. 97, No. 2
-
I
C
OLUCOIE
( 26% , I.v. )
E 2
20
$
01
1
I
1
1
-30
-60
1
0
1
30
I
I
60
80
TIME ( nun1 Figure
(Figure
6. Representative figures showing the effect of indomethacin control; B, hydrocortisone treatment; C, glucose infusion. enhanced gastric motility induced by indomethacin.
7). Although
a significant
relationship
(3 mgikg body wt) on gastric motility in adrenalectomized rats. A, Note that both hydrocortisone and glucose significantly inhibited the
was
indomethacin
noted between the lesion and the bIood glucose level in sham-operated rats given various doses of indomethacin, there was no such relation in adrenalectomized rats. The blood glucose level remained in the same range in response to different doses of
gland
Table
Glucose
3. Gastric Lesion, Hypermotility, and Terminal Following lndomethacin Treatment
Indomethacin (mg/kg body wt) Sham-operated 3 10 25 Adrenalectomized 1 3
3 mg/kg body wt), despite in the lesion index.
(1 and
marked difference
a
Discussion
Blood
Rats
Lesion index
(n)
(mm)
Although a relationship between the adrenal and gastric lesions has been postulated in
in Sham-Operated
and Adrenalectomized
Hypermotility (% increase)
Rats
Blood glucose (mgldl)
rats 5 4 5
0.8 2 0.5 13.8 L 4.4 39.4 ? 6.3
26.1 + 1.9 96.5 + 32.2 178.2 t 36.9
116.5 2 6.7 112.3 + 13.2 79.9 ? 10.2
4 4
18.0 k 7.0 41.0 k 9.7
75.8 2 21.9 198.8 2 16.4
36.4 t 7.2 32.5 + 7.5
rats
Values are presented as the mean + SE from 4-5 rats per group. Indomethacin was given subcutaneously, and the animals were killed 4 h later. Hypermotility was expressed as the percent increase of the amplitude observed during 4 h against basal values, and the blood glucose levels were determined 4 h after indomethacin treatment. These measurements were performed in the same animals of sham-operated and adrenalectomized groups in response to various doses of indomethacin.
August
1989
ADRENALECTOMY
8 : 22
0
f : 8.814
, ,’ A
0
A'
,9’
0 I
IX
??
A,
,’
,’
,’
u
0
0 0
.li
i
1;
i8
3.8 48 LE8I88S ml
i8
88
148 = 128 z P I88 z p 88 S
0 A 0
A
88 ’ : -8.882
48 ‘
8
Figure
A
I8
28
88
48
88
8
88
7. Relationship between ulcerogenesis, motility, and terminal blood glucose level in the same animals of sham-operated and adrenalectomized groups. A highly significant relationship was noted between ulcerogenesis and motility in both groups of rats, the correlation being 0.874.
humans (131, the influences of adrenalectomy on experimental lesions in the rat gastric mucosa remains controversial (6,7,14). The present study confirmed the finding by Urushidani et al. (73, who showed potentiation by adrenalectomy of the mucosal ulcerogenic response induced by indomethacin, and for the first time demonstrated that the sensitization of the gastric motor system to indomethacin may be causally related to aggravation of these lesions in adrenalectomized rats. Recent studies showed gastric hypermotility as an important element in the pathogenesis of indomethacin-induced gastric mucosal lesions in rats (35). A deficiency of prostaglandins, if involved, may not be directly associated with formation of gross lesions induced by this agent, as indomethacin at a lower dose (5 mg/kg body wt), although showing >80% inhibition of cyclooxygenase activity, did not produce any damage in the mucosa (4). As shown in the present study, indomethacin at an ulcerogenic dose (25 mg/kg body wt) did not significantly alter either acid output or mucosal blood flow in shamoperated rats. Therefore, it is unlikely that alterations in these factors account for development of
ON INDOMETHACIN-INDUCED
LESIONS
291
the lesions. As expected, indomethacin produced a marked elevation in gastric motility at the ulcerogenie doses in both sham-operated and adrenalectomized rats, and there was a highly significant relationship between the motility and the ulcer score in these animals. The previous study showed that both atropine and a prostaglandin inhibited gastric hypermotility induced by indomethacin and prevented the lesions even in the presence of exogenous acid (3,s). The present study also showed that hydrocortisone treatment or glucose infusion significantly attenuated the enhanced motility and the mucosal ulcerogenic responses caused by this agent in adrenalectomized rats. Thus, these results strongly suggest that the motility alterations may be closely associated with the mucosal ulcerogenesis following administration of indomethacin. How does bilateral adrenalectomy worsen gastric lesions induced by indomethacin? The absence of adrenal glands results in altered function in many parts of the body, and much attention has been focused on the gastrointestinal tract (15). As the medulectomy had no effect on these lesions (73, and because hydrocortisone mitigated the above phenomenon, the worsening effect of adrenalectomy may be largely due to removal of the cortex (steroid secretion). Participation of acid hypersecretion may be excluded in the mechanism for aggravation by adrenalectomy based on the following reasons. First, acid output was significantly decreased in adrenalectomized rats. Second, hydrocortisone treatment significantly restored the reduced acid output in these animals, yet prevented indomethacin-induced lesions. Other studies showed gastric cellular atrophy (16) or insufficient vasomotion of the small vessels (17) following adrenalectomy. The latter may lead to a defective blood supply to the stomach, and supports the present observation that the mucosal blood flow was significantly reduced in the adrenalectomized rats. The severity of gastric lesions induced by aspirin was significantly potentiated in the hypotrophic mucosa of rats fed liquid diets and by inducing hypogastrinemia (18). Thus, these factors may contribute to the aggravation process of the lesions in adrenalectomized rats, probably by stimulating the experimental circumstance for lesions in response to indomethacin. Certainly, as adrenalectomy alone did not induce any damage in the mucosa, a global mucosal atrophy or a defective blood supply by itself cannot explain the occurrence of focal lesions in the stomach following administration of indomethacin. Although the mechanism by which indomethacin induced gastric hypermotility remains unclear, this may be associated with the effect of indomethacin on the central nervous system and involves stimulation
292
TAKEUCHI
GASTROENTEROLOGY
ET AL.
of the vagus nerves, based on previous findings (3,5,19); gastric hypermotility induced by indomethacin was inhibited by both vagotomy and atropine treatment, and the electrical stimulation of the vagus nerves induces hypermotility and mucosal damages, similar to indomethacin. On the other hand, Mersereau and Hinchey (11) postulated that uncoupling in the glycoprivic receptor system may be responsible for the motility effect of nonsteroidal antiinflammatory drugs. They demonstrated that phenylbutazone, by perturbing energy metabolism in specific receptor cells, provoked a vagally mediated, glucosesensitive gastric contractile response in the rat. Indomethacin, like other nonsteroidal antiinflammatory drugs, may act by uncoupling oxidative phosphorylation, and as a result delivers the glycoprivic stimulus. As expected, higher doses of indomethacin produced significant hypoglycemia and hypermotility in sham-operated rats, and these responses were significantly blocked by exogenous glucose infusion. Thus, the motility effect of indomethacin (910 mgikg body wt) in sham-operated rats may involve the same receptor system in its mechanism. The most important aspect of this study is how adrenalectomy sensitized the motility system to indomethacin. The dose-response curve of indomethacin for the motility shifted to the left in adrenalectomized rats; the value of ED,, was about 10 times lower than that in sham-operated rats. As adrenalectomy alone induced a profound hypoglycemia but did not provoke either hypermotility or mucosal lesion, the hypoglycemia induced by adrenalectomy may not affect the cerebral glucoprivation by a critical amount, Nevertheless, the hypoglycemic stimulus appears to be important in the sensitization of the motility system to indomethacin following adrenalectomy, as both hydrocortisone treatment and glucose infusion blocked the motility responses caused by indomethacin and completely reverted the reduced blood glucose levels in these animals. It has been shown that a lower dose of indomethacin (5 mg/kg body wt) potentiated the motility and the ulcerogenic responses induced by insulin or 2deoxy-o-glucose, an agent known to deliver the glycoprivic stimulus (20,21). As indomethacin at doses as low as 1 mg/kg body wt reduces prostaglandin biosynthesis >80% in the central nervous system as well as in the periphery (4,22), it is possible that a prostaglandin deficiency caused by indomethacin may facilitate the excitation of the glycoprivic system in response to hypoglycemia induced by adrenalectomy. In any case, these results suggest that hypoglycemia may be crucial in the mechanism related to the sensitization of the motility system to indomethacin in rats following adrenalectomy.
Vol. 97, No. 2
In summary, the present study together with previous findings (3-5) suggests that gastric motility may play a key role in the pathogenesis of indomethacin-induced gastric lesions in rats, and may be associated with the worsening of the lesions by adrenalectomy. Abrasion of adrenal glands (cortex) by inducing hypoglycemia may sensitize the system to indomethacin and increase gastric motility and mucosal ulcerogenic responses. As insufficiency of adrenocortical activity was seen after severe stress or in association with diseases such as tuberculosis and cancer, treatment with nonsteroidal antiinflammatory drugs in such patients should be applied cautiously to avoid serious damage to the stomach. References 1. Robert A. Prostaglandins and the gastrointestinal tract. In: Johnson LR, Christensen J, Grossman MI, Jacobson ED, Schultz SG, eds. Physiology of the gastrointestinal tract. New York: Raven, 1981:1407-34. 2. Whittle BJR. Temporal relationship between cyclooxygenase inhibition, as measured by prostacyclin biosynthesis, and the gastrointestinal damage induced by indomethacin in the rat. Gastroenterology 1981;80:94-a. 3. Mersereau WA, Hinchey EJ. Prevention of indomethacininduced gastric hypercontractility; a mucosal protective mechanism of prostaglandin E, (abstr). Gastroenterology ~980:78:1221. 4. Takeuchi K, Ueki S. Okabe S. Importance of gastric motility in the pathogenesis of indomethacin-induced gastric lesions in rats. Dig Dis Sci 1986;31:1114-22. K, Okabe S. Gastric motility is an important 5. Ueki S, Takeuchi factor in pathogenesis of indomethacin-induced gastric mucosal lesions in rats. Dig Dis Sci 1988;33:209-16. KP, Gupta MB, Tangri KK. Mechanism of ulcero6. Bhargawa genie activity of indomethacin and oxyphenbutazone. Eur J Pharmacol 1973;22:191-5. 7. Urushidani T, Kasuya Y, Okabe S. The mechanism of aggravation of indomethacin-induced gastric ulcers by adrenalectomy in the rat. Jpn J Pharmacol 1979;29:775-80. a. Jacobson ED, Linford RH, Grossman MI. Gastric secretion in relation to mucosal blood flow studied by a clearance technique. J Clin Invest 1966;45:1-13. 9. Kitagawa H, Fujiwara M, Osumi Y. Effect of water-immersion stress on gastric secretion and mucosal blood flow in rats. Gastroenterology 1979;77:298-302. (pyramidon) in 10. Brodie BB, Axelrod J. The fate of aminopyrine man and methods for the estimation of aminopyrine and its metabolites in biological materials. J Pharmacol Exp Ther 1950;99:171-4. 11. Mersereau WA, Hinchey EJ. Prevention of phenylbutazone ulcer in the rat by glucose: role of a glycoprivic receptor system. Am J Physiol 1982;242:G429-32. 12. Dunnett CW. A multiple comparison procedure for comparing several treatments with a control. Am J Stat Assoc 1955;50:1096-121. 13. Shapiro H, Britt LG. The effect of adrenalectomy on the gastrointestinal tract. A review of the literature. Rev Surg 1972:29:229-39. 14. Diahangujri B, Hemmati M, Abtahi F, Farrokhoslan M. Preventive effects on indomethacin-induced gastric ulceration in the rat of disulfiram, phenoxybenzamine and bile duct liga-
August 1989
ADRENALECTOMY ON INDOMETHACIN-INDUCED LESIONS
tion. In: Gheorghiu J, ed. Experimental ulcer. Cologne: Gehard Witzstrock, 1975:31-5. 15. Cook AR. Role of adrenocortical steroids in the regulation of gastric secretion. Progress in gastroenterology. Gastroenterology 1967;52:272-81. 16. Crean GP. A comparison between the effects of hypophysectomy and adrenalectomy on the gastric mucosa of the rat. Gut 1968;9:343-7. 17. Takeuchi K, Johnson LR. Effect of cell proliferation and loss on aspirin-induced gastric damage in the rat. Am J Physiol 1982;243:G463-8. 18. Ramey ER, Goldstein MS, Levine R. Action of norepinephrine and adrenal cortical steroids on blood pressure and work performance of adrenalectomized dogs. Am J Physiol 1951; 165:450-5. 19. Ohno T, Uramoto H, Ishihara T, Okabe S. Effects of 16,16dimethyl prostaglandin E, on surface epithelial cell damage
293
in the rat stomach induced by vagal nerve stimulation. Jpn J Pharmacol 1987;43:429-39. 20. Mersereau WA, Hinchey EJ. The role of mechanical factors in the genesis of gastric ulcers in the indomethacin-insulin treated rat (abstr). Dig Dis Sci 1985;30:388. 21. Okada M, Niida H, Takeuchi K, Okabe S. Role of prostaglandin deficiency in the pathogenetic mechanism of gastric lesions induced by indomethacin in rats. Dig Dis Sci (in press). 22. Fargeas MJ, Fioramonti J, Bueno L. Prostaglandin E,; a neuromodulator in the central control of gastrointestinal motility and feeding behavior by calcitonin. Science 1984;225:1059-2.
Received January 29, 1988. Accepted February 9, 1989. Address requests for reprints to: Dr. Koji Takeuchi, Department of Applied Pharmacology, Kyoto Pharmaceutical University, Misasagi, Yamashina, Kyoto 607, Japan.