Peptides,Vol. 13, pp. 373-376, 1992
0196-9781/92 $5.00 + .00 Copyright © 1992PergamonPressLtd.
Printedin the USA.
Vasopressin, Corticosterone Levels, and Gastric Ulcers During Food-Restriction Stress H E L E N M. M U R P H Y
A N D C Y R I L L A H. W I D E M A N
Departments of Psychology and Biology, John Carroll University, Cleveland, O H 44118 Received 11 M a r c h 1991 MURPHY, H. M. AND C. H. WIDEMAN. Vasopressin, corticosteronelevels, and gastric ulcers duringfood-restriction stress. PEPTIDES 13(2) 373-376, 1992.--Corticosterone levelsand ulcerswere compared in vasopressin-containing(LE) and vasopressindeficient (DI) rats under ad lib and food-restricted conditions. In the ad lib situation, DI and LE rats had similar corticosterone levels and no ulcers. After 1 day of food restriction, the corticosterone levels were elevated in DI and LE rats, with a significantly higher level in LE rats. No ulcers were present in either strain. After 2 days of food restriction, the corticosterone levels were similar in DI and LE rats. The level in DI rats was comparable to that of the preceding day, but the level in LE animals dropped significantlyfrom the previous day. Significantulceration was evident in DI rats, but absent in LE rats. Following 3 days of food restriction,the corticosteronelevelin LE rats had returned to the ad lib level,whereas, for DI rats, an elevatedlevelwas maintained. There were no ulcers in LE rats, but they were present in DI rats. Thus LE and DI rats responded differentlyto the stress of food restriction. The mechanism underlying the response is most likely related to changes in the hypothalamic-pituitary-adrenocortical axis and its reaction to stress. Vasopressin Brattlebororat Food-restriction stress Hypothalamic-pituitary-adrenocorticalaxis
Ulcers
FOR many years, scientists interested in studying the reaction of the body to stress have utilized Selye's General Adaptation Syndrome (G.A.S.) as background material to examine the physiological responses of the organism (12). The G.A.S. encompasses three stages: 1) the alarm reaction characterized by a "generalized call to arms of the defensive forces in the organism" (e.g., increased levels of glucocorticoids and gastric ulcer formation); 2) the stage of adaptation or resistance; and 3) the stage of exhaustion. One of the bodily responses to stress is activation of the hypothalamic-pituitary-adrenocorticalaxis with the release ofcorticotropin-releasing factor (CRF), adrenocorticotropic hormone (ACTH), and glucocorticoids (e.g., corticosterone), respectively (22). Studies strongly suggest that vasopressin (VP) has a role to play in the stress response (5,9-11). It would appear, then, that a lack of VP should influence the responsiveness of an organism to stress. A genetic mutant of the Long-Evans (LE) strain of rat lacks the ability to produce hypothalamic VP and is known as the homozygous Brattleboro rat (15). Brattleboro rats exhibit hereditary diabetes insipidus and are referred to as DI animals (16). These animals show an altered response to stress, when compared to control animals, due to hormonal and physiological changes. Stress studies, centering on the hypothalamic-pituitaryadrenocortical axis in DI rats, demonstrated that plasma corticosterone levels were significantly lower in DI rats under certain types of stress when compared to control animals (1,3,21). A second index of responsiveness to stress is gastric ulcer formation (12,13). We have demonstrated that DI animals are more susceptible to stress ulcer formation than LE animals under conditions of restraint and restraint plus intermittent shock (18).
Corticosterone
We have also shown that DI animals develop more gastric ulcers than LE animals in the activity-stress ulcer paradigm (4,19). The activity-stress ulcer paradigm was devised in 1973 and demonstrated that when rats are housed in running-wheel activity cages with 1-hour daily access to food, they exhibit excessive running behavior which is followed by death and the presence of large gastric ulcers (6). The investigators maintained that the two factors of restricted feeding and running activity are necessary components to the production of ulcers and that neither factor alone can be considered as a sufficient etiological variable. Surprisingly, in our activity-wheel studies, DI animals, maintained in the home cage and restricted to 1 hour of food each day and no running activity (one type of control group in the studies), died and exhibited large gastric ulcers. The LE animals subjected to the same conditions survived and showed no significant stomach pathology. These data indicate that, in DI rats, food restriction alone is a sufficient etiological variable capable of inducing gastric ulcers. This observation, we believed, was intimately associated with the absence of VP. A follow-up study involving VP replacement in DI rats confirmed this hypothesis (20). The present experiment was concerned with the role of VP in the modulation of the hypothalamic-pituitary-adrenocortical system during food-restriction stress, as evidenced by variations in circulating levels of corticosterone in the blood. To this end, corticosterone levels were measured in vasopressin-containing (LE) and vasopressin-deficient (DI) rats under ad lib and foodrestricted conditions. In addition to determining corticosterone levels, gastric ulceration was recorded as an index of the severity of the stress upon the animals.
373
374
MURPHY AND WIDEMAN METHOD
The subjects utilized were 20 male LE rats and 20 male homozygous DI rats obtained from Harlan-Sprague-Dawley, Inc. As in previous studies (4,18-20), all animals were 5 weeks of age and weighed between 80-100 grams at the beginning of the experiment. Animals were housed in individual home cages with the room temperature maintained at 22-24°C with a 12-hour light period alternating with a 12-hour dark period. The habituation period, lasting from days 1 through 7, was characterized by ad lib access to food and water. The experimental period extended through days 8, 9, or 10. During this time, the rats were given l-hour access to food each day, except on day 8, when animals were food restricted for 24 hours. The feeding period occurred during the second hour of the light cycle. Throughout the experimental period, the animals had ad lib access to water. The animals were divided into four groups: Group 1 consisted of five LE and five DI rats that were not food restricted and were sacrificed at the end of the habituation period (control group); groups 2, 3, and 4 consisted of five LE and five DI rats that were sacrificed after 1, 2, and 3 days of food restriction, respectively. Body weight, food intake, and water intake were measured daily. Animals were sacrificed by decapitation, which occurred 15 hours after feeding for the food-restricted groups. The time of sacrifice was 5 hours into the dark cycle. Trunk blood was collected in heparinized tubes and the plasma was analyzed for corticosterone levels by RIA (Endocrine Sciences, CA). The assay utilized has been validated by comparison to assays with purification as well as with standard linearity and recovery checks. In addition, the stomach was removed from each animal and examined for gastric pathology. Although the type of pathology is properly referred to as gastric erosions, the literature in acute stress-related upper gastrointestinal tract bleeding commonly refers to this phenomenon as "stress ulcers" (2). Glandular ulcers were defined as acute gastric lesions that were bright red or reddish brown in color. They were either irregular and parallel to the folds of the gastric mucosa, or were circumscribed and focal. Lesion areas were measured by an experimenter who was unaware of the experimental treatment of the animals. At the conclusion of the experiment a 2 × 2 ANOVA was utilized for analysis of plasma corticosterone levels, ulcer length, percentage loss of body weight, and food intake during the experimental period. Protected t-tests (Fisher LSD) were employed in post hoc testing. An alpha level of 0.05 was used in interpretation of statistical significance for the protected t-tests.
60
+~
T
50
~Ol
40
I---ILE
v
,0 o -I-,' o
o
20
10~-l0
0
3 Days of Food Restr'iction
FIG. I. Mean plasma corticosterone levels in ad lib and food-restricted conditions.
Figure 2 shows the mean length of ulcers for the four groups of DI and LE animals. There was a main effect for strain, F(1,32) = 8.14, p < 0.01; there was a main effect for days, F(3,32) = 3.35, p < 0.05; and there was an interaction, F(3,32) = 3.32, p < 0.05. In the ad lib situation, no animals exhibited stomach pathology. After 1 day of food restriction, no LE and two DI animals exhibited pathology. Each of the two DI animals had one small lesion. After 2 days of food restriction, one LE rat had a small ulcer and all five DI rats had developed glandular ulcers. After 3 days of food restriction, no LE but all five DI rats had ulcers. Thus, over the 3-day period of food restriction, the DI animals developed increasingly more gastric ulcers, while the LE animals remained free of significant stomach pathology. Body weight, food intake, and water intake were monitored throughout the experiment. At the time of arrival at our laboratory, there was no significant difference in body weight between DI and LE animals. By the end of the habituation period, LE rats weighed significantly more than DI rats. Because of the different weights in the two strains before the introduction of food restriction, body weight loss during the experimental period was calculated on the basis of percentage of body weight loss from the last day of habituation. Table 1 summarizes the mean values for the percentage of body weight loss for the 3 experimental days. There was a main effect for strain, F(1,24) = 25.15, p < 0.001; there was a main effect for days, F(2,24) = 69.31, p < 0.001; and there was an interaction, F(2,24) = 7.38, p < 0.01. Following 3 days of food restriction, DI animals had lost a significantly greater percentage of body weight than did LE animals.
RESULTS Figure 1 shows the mean corticosterone levels for the four groups of DI and LE animals. There was no main effect for strain, F(1,32) = 0.20, NS; there was a main effect for days, F(3,32) = 8.97, p < 0.001; and there was an interaction, F(3,32) = 3 88, p < 0.05. In the ad lib (day 0--last day of habituation) situation, DI and LE animals had similar corticosterone levels. After 1 day of food restriction, the corticosterone levels were elevated in both DI and LE rats, with a significantly higher level in LE rats. After 2 days of food restriction, the corticosterone levels were similar in DI and LE animals. The level in DI rats was comparable to their level of the preceding day, but the level in LE animals dropped significantly from the previous day, Following 3 days of food restriction, the corticosterone level in LE rats had returned to the ad lib level, whereas, for DI rats, an elevated level was maintained.
12
lo
@oi FILE
0 ~ 0 , 0 , __, 0 , ~ _ _ 0 , O 1 2 3 Days of Food Restriction
FIG. 2. Mean total ulcer length in ad lib and food-restricted conditions.
VP, STRESS, CORTICOSTERONE, AND ULCERS
375
TABLE 1 MEAN % BODY WEIGHT LOSS FROM LAST DAY OF HABITUATION (+_SEM) Daysof Food Restriction
DI animals LE animals
1
2
3
-20.5% ±0.8 - 19.3% ±0.8
-26.9% ±1.0 -24.1% +0.8
-36.3%* _+1.3 -27.6% _ 1.4
* p < 0.05.
Thus a large decrease in body weight in DI animals paralleled their increase in stomach pathology. With respect to food intake, no significant differences were noted between DI and LE animals in the food-restricted period, F(1,24) = 0.85, NS. Water intake values were similar to those reported in a previous study (20). Because of the condition of diabetes insipidus, water consumption was significantly higher for DI rats than LE rats during both the habituation and experimental periods. Both strains dropped in their water intake during the experimental period compared to the habituation period. DISCUSSION Selye has described the General Adaptation Syndrome as the reaction of the body to a stressor (12). A variety of bodily responses are coordinated during the "alarm reaction." The biological responses that Selye described as part of the alarm reaction included: hypophysial-adrenocortical stimulation, release of catecholamines, and the "morphologic triad" of adrenal hypertrophy, thymolytic atrophy, and gastroduodenal ulcers. The alarm reaction may be followed by what Selye described as the "stage of resistance" in which the organism adapts to the stressor. The manifestations of this second stage were quite different from, and in many instances the exact opposite of, those which characterized the alarm reaction. On the last day of the nonstressed period (day 0), the vasopressin-containing LE animals and vasopressin-deficient DI animals had similar corticosterone levels and showed no stomach pathology. These findings are in agreement with other researchers who have compared basal corticosterone levels in DI and control rats (1,21). Following 1 day of food-restriction stress, the LE as well as the DI animals exhibited a marked alarm reaction, as evidenced by dramatic increases in corticosterone levels. The LE animals, however, produced a significantly higher level of corticosterone than the DI animals. Both hypothalamic VP and CRF, working synergistically upon the release of ACTH in the anterior pituitary, are likely to be responsible for this sudden burst of corticosterone production in the LE animals. A deficiency of VP in the DI animals probably caused the blunted corticosterone response observed. The smaller increase in plasma corticosterone level in DI rats compared to control animals has also been reported by other investigators examining the effect of various stressors on corticosterone levels (3,21). Even though two DI rats had one small lesion each, no significant difference in stomach pathology was evident between the two strains on this day. Following the second day of food-restriction stress, there was a significant decrease in the corticosterone level of LE animals
from the previous day. The DI animals, however, maintained the same corticosterone level as on day 1. There was no significant difference between LE and DI animals in corticosterone levels. Apparently, the LE animals were moving into the resistance stage, whereas the DI animals remained in the alarm reaction. Upon examination of the stomachs of the LE animals, no significant ulceration was evident. Significant ulceration in the stomachs of DI animals was apparent. At the end of the third day of stress, marked differences were found in LE and DI animals. The corticosterone level of the LE animals had returned to the normal, unstressed level and no stomach pathology was evident. Obviously, the LE animals were in the stage of resistance. The corticosterone level in the DI animals remained elevated and even more stomach ulceration was apparent. It appears that the DI animals were still deeply entrenched in the alarm reaction. The observed phenomena may be explained on the basis of certain endocrinological mechanisms. On the last day of habituation (day 0), the levels of corticosterone in DI and LE rats were not significantly different. Although the DI animals do not release hypothalamic VP into the hypophysial portal blood, it has been shown that the normal plasma concentrations of ACTH in this mutant strain are maintained by CRF-41 and possibly other factors which facilitate ACTH release (7,14). In the LE animals, the basal level of corticosterone may be due primarily to the action of the VP-deficient CRF neurosecretory cells, which may maintain a baseline level of ACTH synthesis and release (17). Following the first day of food-restriction stress in the LE animals, VP may be released in large quantities from the parvocellular VP-containing CRF neurosecretory cells, which may be more adapted to rapid modulation of ACTH output (17). Thus the level of corticosterone in these rats increased dramatically. Since DI animals are deficient in VP, only CRF levels increased, resulting in elevated, but blunted, levels of corticosterone. The food-restriction stress continued through days 2 and 3, and probably resulted in increased levels of VP and CRF in the LE animal. It has been observed that a dose-dependent inhibition of portal immunoreactive CRF concentration results with increasing intracerebroventricular VP doses (8). This possibly could explain the progressive decrease in corticosterone levels observed on days 2 and 3 in the LE animals. However, this phenomenon does not exclude the negative feedback of corticosterone itself in the hypothalamic and/or pituitary region. Following days 2 and 3 of food-restriction stress, the DI animals showed no significant alteration in corticosterone levels from day 1. The inhibitory effect on CRF of increasing levels of VP mentioned previously was not effective in these animals, since they were deficient in VP. This possibly could account for maintenance of elevated levels of corticosterone through day 3 of stress. In addition, it has been proposed that the glucocorticoid-negative feedback system may have a higher set point in DI rats as compared with the LE animals (14). It is interesting to note that the DI animals, with a deficiency of VP and increased corticosterone levels maintained over the 3-day stress period, developed marked gastric ulceration. This is in direct contrast with the proposal of Parr, which holds that food restriction alone is not a sufficient etiological variable to induce gastric ulcers (6). In conclusion, vasopressin-containing and vasopressin-deficient rats respond differently to the stress of food restriction. The mechanism underlying the response is most likely related to the interaction between VP and the hypothalamic-pituitaryadrenocortical axis and its reaction to stress.
376
MURPHY AND WIDEMAN
REFERENCES 1. Arimura, A.; Saito, T.; Bowers, C. Y.; SchaUy, A. V. Pituitary-adrenal activation in rats with hereditary hypothalamic diabetes insipidus. Acta Endocrinol. 54:155-165; 1967. 2. Hernandez, D. E.; Glavin, G. B., eds. Neurobiology of stress ulcers. New York: New York Academy of Sciences; 1990. 3. McCann, S. M.; Antunes-Rodrigues, J.; Nallar, R.; Valtin, H. Pituitary-adrenal function in the absence of vasopressin. Endocrinology 79:1058-1064; 1966. 4. Murphy, H. M.; Wideman, C. H. Self-starvation and activity stress in Brattleboro rats. Physiol. Psychol. 11:209-213; 1983. 5. Ono, N.; Bedran de Castro, J.; Khorram, O.; McCann, S. Role of arginine vasopressin in control of ACTH and LH release during stress. Life Sci. 36:1779-1786; 1985. 6. Parr, W. P.; Hauser, V. P. Activity and food-restriction effects on gastric glandular lesions in the rat: The activity-stress ulcer. Bull. Psychon. Soc, 2:213-214; 1973. 7. Perlmutter, A. F.; Kokas, L. A.; Loeser, B.; Saffran, M. Corticotrophin releasing factor activity in the Brattleboro rat median eminence. Ann. NY Acad. Sci. 394:587-590; 1982. 8. Plotsky, P. M. Hypophyseotropic regulation of adenohypophyseal adrenocorticotropin secretion. Fed. Proc. 44:207-213; 1985. 9. Rivier, C.; Rivier, J.; Mormede, P.; Vale, W. Studies of the nature of the interaction between vasopressin and corticotropic-releasing factor on adrenocorticotropin release in the rat. Endocrinology 115: 882-886; 1984. 10. Rivier, C.; Vale, W. Modulation of stress-induced ACTH release by corticotropic-releasing factor, catecholamines and vasopressin. Nature 305:325-327; 1983. I I. Rivier, C.; Vale, W. Interaction ofcorticotropin-releasing factor and arginine-vasopressin on adrenocorticotropin secretion in vivo. Endocrinology 113:939-942; 1983.
12. Selye, H. The stress of life (revised edition). New York: McGrawHill Book Co.; 1978. 13. Szabo, S.; Glavin, G. B. Hans Selye and the concept of biological stress: Ulcer pathogenesis as a historical paradigm. Ann. NY Acad. Sci. 597:14-16; 1990. 14. Tannahill, L. A.; Dow, R. C.; Fairhall, K. M.; Robinson, I. C. A. F.; Fink, G. Comparison ofadrenocorticotropin control in Brattleboro, Long-Evans, and Wistar rats. Neuroendocrinology 48:650-657; 1988. 15. Valtin, H.; Sawyer, W. H.; Sokol, H. W. Neurohypophysial principles in rats homozygous and heterozygous for hypothalamic diabetes insipidus (Brattleboro strain). Endocrinology 77:701-706; 1965. 16. Valtin, H.; Schroeder, H. A. Familial hypothalamic diabetes insipidus in rats (Brattleboro strain). Am. J. Physiol. 206:425-430; 1964. 17. Whitnall, M. H. Stress selectively activates the vasopressin-containing subset of corticotropin-releasing hormone neurons. Neuroendocrinology 50:702-707; 1989. 18. Wideman, C. H.; Murphy, H. M. The effects of restraint and restraint plus intermittent shock on ulcer formation in Brattleboro rats. Physiol. Psychol. 11:78-80; 1983. 19. Wideman, C. H.; Murphy, H. M. Effects of vasopressin deficiency, age, and stress on stomach ulcer induction in rats. Peptides 6(Suppl. 1):63-67; 1985. 20. Wideman, C. H.; Murphy, H. M. Effects ofvasopressin replacement during food-restriction stress. Peptides 12:285-288; 1991. 21. Wiley, K. M.; Pearlmutter, A. F.; Miller, R. E. Decreased adrenal sensitivity to ACTH in the vasopressin-deficient Brattleboro rat. Neuroendocrinology 14:257-270; 1974. 22. Williams, J. D. Textbook of endocrinology (7th ed.) Philadelphia: W. B. Saunders Co.; 1985.
0196-9781/92 $5.00 + .00 Copyright © 1992PergamonPressLtd.
Printedin the USA.
Vasopressin, Corticosterone Levels, and Gastric Ulcers During Food-Restriction Stress H E L E N M. M U R P H Y
A N D C Y R I L L A H. W I D E M A N
Departments of Psychology and Biology, John Carroll University, Cleveland, O H 44118 Received 11 M a r c h 1991 MURPHY, H. M. AND C. H. WIDEMAN. Vasopressin, corticosteronelevels, and gastric ulcers duringfood-restriction stress. PEPTIDES 13(2) 373-376, 1992.--Corticosterone levelsand ulcerswere compared in vasopressin-containing(LE) and vasopressindeficient (DI) rats under ad lib and food-restricted conditions. In the ad lib situation, DI and LE rats had similar corticosterone levels and no ulcers. After 1 day of food restriction, the corticosterone levels were elevated in DI and LE rats, with a significantly higher level in LE rats. No ulcers were present in either strain. After 2 days of food restriction, the corticosterone levels were similar in DI and LE rats. The level in DI rats was comparable to that of the preceding day, but the level in LE animals dropped significantlyfrom the previous day. Significantulceration was evident in DI rats, but absent in LE rats. Following 3 days of food restriction,the corticosteronelevelin LE rats had returned to the ad lib level,whereas, for DI rats, an elevatedlevelwas maintained. There were no ulcers in LE rats, but they were present in DI rats. Thus LE and DI rats responded differentlyto the stress of food restriction. The mechanism underlying the response is most likely related to changes in the hypothalamic-pituitary-adrenocortical axis and its reaction to stress. Vasopressin Brattlebororat Food-restriction stress Hypothalamic-pituitary-adrenocorticalaxis
Ulcers
FOR many years, scientists interested in studying the reaction of the body to stress have utilized Selye's General Adaptation Syndrome (G.A.S.) as background material to examine the physiological responses of the organism (12). The G.A.S. encompasses three stages: 1) the alarm reaction characterized by a "generalized call to arms of the defensive forces in the organism" (e.g., increased levels of glucocorticoids and gastric ulcer formation); 2) the stage of adaptation or resistance; and 3) the stage of exhaustion. One of the bodily responses to stress is activation of the hypothalamic-pituitary-adrenocorticalaxis with the release ofcorticotropin-releasing factor (CRF), adrenocorticotropic hormone (ACTH), and glucocorticoids (e.g., corticosterone), respectively (22). Studies strongly suggest that vasopressin (VP) has a role to play in the stress response (5,9-11). It would appear, then, that a lack of VP should influence the responsiveness of an organism to stress. A genetic mutant of the Long-Evans (LE) strain of rat lacks the ability to produce hypothalamic VP and is known as the homozygous Brattleboro rat (15). Brattleboro rats exhibit hereditary diabetes insipidus and are referred to as DI animals (16). These animals show an altered response to stress, when compared to control animals, due to hormonal and physiological changes. Stress studies, centering on the hypothalamic-pituitaryadrenocortical axis in DI rats, demonstrated that plasma corticosterone levels were significantly lower in DI rats under certain types of stress when compared to control animals (1,3,21). A second index of responsiveness to stress is gastric ulcer formation (12,13). We have demonstrated that DI animals are more susceptible to stress ulcer formation than LE animals under conditions of restraint and restraint plus intermittent shock (18).
Corticosterone
We have also shown that DI animals develop more gastric ulcers than LE animals in the activity-stress ulcer paradigm (4,19). The activity-stress ulcer paradigm was devised in 1973 and demonstrated that when rats are housed in running-wheel activity cages with 1-hour daily access to food, they exhibit excessive running behavior which is followed by death and the presence of large gastric ulcers (6). The investigators maintained that the two factors of restricted feeding and running activity are necessary components to the production of ulcers and that neither factor alone can be considered as a sufficient etiological variable. Surprisingly, in our activity-wheel studies, DI animals, maintained in the home cage and restricted to 1 hour of food each day and no running activity (one type of control group in the studies), died and exhibited large gastric ulcers. The LE animals subjected to the same conditions survived and showed no significant stomach pathology. These data indicate that, in DI rats, food restriction alone is a sufficient etiological variable capable of inducing gastric ulcers. This observation, we believed, was intimately associated with the absence of VP. A follow-up study involving VP replacement in DI rats confirmed this hypothesis (20). The present experiment was concerned with the role of VP in the modulation of the hypothalamic-pituitary-adrenocortical system during food-restriction stress, as evidenced by variations in circulating levels of corticosterone in the blood. To this end, corticosterone levels were measured in vasopressin-containing (LE) and vasopressin-deficient (DI) rats under ad lib and foodrestricted conditions. In addition to determining corticosterone levels, gastric ulceration was recorded as an index of the severity of the stress upon the animals.
373
374
MURPHY AND WIDEMAN METHOD
The subjects utilized were 20 male LE rats and 20 male homozygous DI rats obtained from Harlan-Sprague-Dawley, Inc. As in previous studies (4,18-20), all animals were 5 weeks of age and weighed between 80-100 grams at the beginning of the experiment. Animals were housed in individual home cages with the room temperature maintained at 22-24°C with a 12-hour light period alternating with a 12-hour dark period. The habituation period, lasting from days 1 through 7, was characterized by ad lib access to food and water. The experimental period extended through days 8, 9, or 10. During this time, the rats were given l-hour access to food each day, except on day 8, when animals were food restricted for 24 hours. The feeding period occurred during the second hour of the light cycle. Throughout the experimental period, the animals had ad lib access to water. The animals were divided into four groups: Group 1 consisted of five LE and five DI rats that were not food restricted and were sacrificed at the end of the habituation period (control group); groups 2, 3, and 4 consisted of five LE and five DI rats that were sacrificed after 1, 2, and 3 days of food restriction, respectively. Body weight, food intake, and water intake were measured daily. Animals were sacrificed by decapitation, which occurred 15 hours after feeding for the food-restricted groups. The time of sacrifice was 5 hours into the dark cycle. Trunk blood was collected in heparinized tubes and the plasma was analyzed for corticosterone levels by RIA (Endocrine Sciences, CA). The assay utilized has been validated by comparison to assays with purification as well as with standard linearity and recovery checks. In addition, the stomach was removed from each animal and examined for gastric pathology. Although the type of pathology is properly referred to as gastric erosions, the literature in acute stress-related upper gastrointestinal tract bleeding commonly refers to this phenomenon as "stress ulcers" (2). Glandular ulcers were defined as acute gastric lesions that were bright red or reddish brown in color. They were either irregular and parallel to the folds of the gastric mucosa, or were circumscribed and focal. Lesion areas were measured by an experimenter who was unaware of the experimental treatment of the animals. At the conclusion of the experiment a 2 × 2 ANOVA was utilized for analysis of plasma corticosterone levels, ulcer length, percentage loss of body weight, and food intake during the experimental period. Protected t-tests (Fisher LSD) were employed in post hoc testing. An alpha level of 0.05 was used in interpretation of statistical significance for the protected t-tests.
60
+~
T
50
~Ol
40
I---ILE
v
,0 o -I-,' o
o
20
10~-l0
0
3 Days of Food Restr'iction
FIG. I. Mean plasma corticosterone levels in ad lib and food-restricted conditions.
Figure 2 shows the mean length of ulcers for the four groups of DI and LE animals. There was a main effect for strain, F(1,32) = 8.14, p < 0.01; there was a main effect for days, F(3,32) = 3.35, p < 0.05; and there was an interaction, F(3,32) = 3.32, p < 0.05. In the ad lib situation, no animals exhibited stomach pathology. After 1 day of food restriction, no LE and two DI animals exhibited pathology. Each of the two DI animals had one small lesion. After 2 days of food restriction, one LE rat had a small ulcer and all five DI rats had developed glandular ulcers. After 3 days of food restriction, no LE but all five DI rats had ulcers. Thus, over the 3-day period of food restriction, the DI animals developed increasingly more gastric ulcers, while the LE animals remained free of significant stomach pathology. Body weight, food intake, and water intake were monitored throughout the experiment. At the time of arrival at our laboratory, there was no significant difference in body weight between DI and LE animals. By the end of the habituation period, LE rats weighed significantly more than DI rats. Because of the different weights in the two strains before the introduction of food restriction, body weight loss during the experimental period was calculated on the basis of percentage of body weight loss from the last day of habituation. Table 1 summarizes the mean values for the percentage of body weight loss for the 3 experimental days. There was a main effect for strain, F(1,24) = 25.15, p < 0.001; there was a main effect for days, F(2,24) = 69.31, p < 0.001; and there was an interaction, F(2,24) = 7.38, p < 0.01. Following 3 days of food restriction, DI animals had lost a significantly greater percentage of body weight than did LE animals.
RESULTS Figure 1 shows the mean corticosterone levels for the four groups of DI and LE animals. There was no main effect for strain, F(1,32) = 0.20, NS; there was a main effect for days, F(3,32) = 8.97, p < 0.001; and there was an interaction, F(3,32) = 3 88, p < 0.05. In the ad lib (day 0--last day of habituation) situation, DI and LE animals had similar corticosterone levels. After 1 day of food restriction, the corticosterone levels were elevated in both DI and LE rats, with a significantly higher level in LE rats. After 2 days of food restriction, the corticosterone levels were similar in DI and LE animals. The level in DI rats was comparable to their level of the preceding day, but the level in LE animals dropped significantly from the previous day, Following 3 days of food restriction, the corticosterone level in LE rats had returned to the ad lib level, whereas, for DI rats, an elevated level was maintained.
12
lo
@oi FILE
0 ~ 0 , 0 , __, 0 , ~ _ _ 0 , O 1 2 3 Days of Food Restriction
FIG. 2. Mean total ulcer length in ad lib and food-restricted conditions.
VP, STRESS, CORTICOSTERONE, AND ULCERS
375
TABLE 1 MEAN % BODY WEIGHT LOSS FROM LAST DAY OF HABITUATION (+_SEM) Daysof Food Restriction
DI animals LE animals
1
2
3
-20.5% ±0.8 - 19.3% ±0.8
-26.9% ±1.0 -24.1% +0.8
-36.3%* _+1.3 -27.6% _ 1.4
* p < 0.05.
Thus a large decrease in body weight in DI animals paralleled their increase in stomach pathology. With respect to food intake, no significant differences were noted between DI and LE animals in the food-restricted period, F(1,24) = 0.85, NS. Water intake values were similar to those reported in a previous study (20). Because of the condition of diabetes insipidus, water consumption was significantly higher for DI rats than LE rats during both the habituation and experimental periods. Both strains dropped in their water intake during the experimental period compared to the habituation period. DISCUSSION Selye has described the General Adaptation Syndrome as the reaction of the body to a stressor (12). A variety of bodily responses are coordinated during the "alarm reaction." The biological responses that Selye described as part of the alarm reaction included: hypophysial-adrenocortical stimulation, release of catecholamines, and the "morphologic triad" of adrenal hypertrophy, thymolytic atrophy, and gastroduodenal ulcers. The alarm reaction may be followed by what Selye described as the "stage of resistance" in which the organism adapts to the stressor. The manifestations of this second stage were quite different from, and in many instances the exact opposite of, those which characterized the alarm reaction. On the last day of the nonstressed period (day 0), the vasopressin-containing LE animals and vasopressin-deficient DI animals had similar corticosterone levels and showed no stomach pathology. These findings are in agreement with other researchers who have compared basal corticosterone levels in DI and control rats (1,21). Following 1 day of food-restriction stress, the LE as well as the DI animals exhibited a marked alarm reaction, as evidenced by dramatic increases in corticosterone levels. The LE animals, however, produced a significantly higher level of corticosterone than the DI animals. Both hypothalamic VP and CRF, working synergistically upon the release of ACTH in the anterior pituitary, are likely to be responsible for this sudden burst of corticosterone production in the LE animals. A deficiency of VP in the DI animals probably caused the blunted corticosterone response observed. The smaller increase in plasma corticosterone level in DI rats compared to control animals has also been reported by other investigators examining the effect of various stressors on corticosterone levels (3,21). Even though two DI rats had one small lesion each, no significant difference in stomach pathology was evident between the two strains on this day. Following the second day of food-restriction stress, there was a significant decrease in the corticosterone level of LE animals
from the previous day. The DI animals, however, maintained the same corticosterone level as on day 1. There was no significant difference between LE and DI animals in corticosterone levels. Apparently, the LE animals were moving into the resistance stage, whereas the DI animals remained in the alarm reaction. Upon examination of the stomachs of the LE animals, no significant ulceration was evident. Significant ulceration in the stomachs of DI animals was apparent. At the end of the third day of stress, marked differences were found in LE and DI animals. The corticosterone level of the LE animals had returned to the normal, unstressed level and no stomach pathology was evident. Obviously, the LE animals were in the stage of resistance. The corticosterone level in the DI animals remained elevated and even more stomach ulceration was apparent. It appears that the DI animals were still deeply entrenched in the alarm reaction. The observed phenomena may be explained on the basis of certain endocrinological mechanisms. On the last day of habituation (day 0), the levels of corticosterone in DI and LE rats were not significantly different. Although the DI animals do not release hypothalamic VP into the hypophysial portal blood, it has been shown that the normal plasma concentrations of ACTH in this mutant strain are maintained by CRF-41 and possibly other factors which facilitate ACTH release (7,14). In the LE animals, the basal level of corticosterone may be due primarily to the action of the VP-deficient CRF neurosecretory cells, which may maintain a baseline level of ACTH synthesis and release (17). Following the first day of food-restriction stress in the LE animals, VP may be released in large quantities from the parvocellular VP-containing CRF neurosecretory cells, which may be more adapted to rapid modulation of ACTH output (17). Thus the level of corticosterone in these rats increased dramatically. Since DI animals are deficient in VP, only CRF levels increased, resulting in elevated, but blunted, levels of corticosterone. The food-restriction stress continued through days 2 and 3, and probably resulted in increased levels of VP and CRF in the LE animal. It has been observed that a dose-dependent inhibition of portal immunoreactive CRF concentration results with increasing intracerebroventricular VP doses (8). This possibly could explain the progressive decrease in corticosterone levels observed on days 2 and 3 in the LE animals. However, this phenomenon does not exclude the negative feedback of corticosterone itself in the hypothalamic and/or pituitary region. Following days 2 and 3 of food-restriction stress, the DI animals showed no significant alteration in corticosterone levels from day 1. The inhibitory effect on CRF of increasing levels of VP mentioned previously was not effective in these animals, since they were deficient in VP. This possibly could account for maintenance of elevated levels of corticosterone through day 3 of stress. In addition, it has been proposed that the glucocorticoid-negative feedback system may have a higher set point in DI rats as compared with the LE animals (14). It is interesting to note that the DI animals, with a deficiency of VP and increased corticosterone levels maintained over the 3-day stress period, developed marked gastric ulceration. This is in direct contrast with the proposal of Parr, which holds that food restriction alone is not a sufficient etiological variable to induce gastric ulcers (6). In conclusion, vasopressin-containing and vasopressin-deficient rats respond differently to the stress of food restriction. The mechanism underlying the response is most likely related to the interaction between VP and the hypothalamic-pituitaryadrenocortical axis and its reaction to stress.
376
MURPHY AND WIDEMAN
REFERENCES 1. Arimura, A.; Saito, T.; Bowers, C. Y.; SchaUy, A. V. Pituitary-adrenal activation in rats with hereditary hypothalamic diabetes insipidus. Acta Endocrinol. 54:155-165; 1967. 2. Hernandez, D. E.; Glavin, G. B., eds. Neurobiology of stress ulcers. New York: New York Academy of Sciences; 1990. 3. McCann, S. M.; Antunes-Rodrigues, J.; Nallar, R.; Valtin, H. Pituitary-adrenal function in the absence of vasopressin. Endocrinology 79:1058-1064; 1966. 4. Murphy, H. M.; Wideman, C. H. Self-starvation and activity stress in Brattleboro rats. Physiol. Psychol. 11:209-213; 1983. 5. Ono, N.; Bedran de Castro, J.; Khorram, O.; McCann, S. Role of arginine vasopressin in control of ACTH and LH release during stress. Life Sci. 36:1779-1786; 1985. 6. Parr, W. P.; Hauser, V. P. Activity and food-restriction effects on gastric glandular lesions in the rat: The activity-stress ulcer. Bull. Psychon. Soc, 2:213-214; 1973. 7. Perlmutter, A. F.; Kokas, L. A.; Loeser, B.; Saffran, M. Corticotrophin releasing factor activity in the Brattleboro rat median eminence. Ann. NY Acad. Sci. 394:587-590; 1982. 8. Plotsky, P. M. Hypophyseotropic regulation of adenohypophyseal adrenocorticotropin secretion. Fed. Proc. 44:207-213; 1985. 9. Rivier, C.; Rivier, J.; Mormede, P.; Vale, W. Studies of the nature of the interaction between vasopressin and corticotropic-releasing factor on adrenocorticotropin release in the rat. Endocrinology 115: 882-886; 1984. 10. Rivier, C.; Vale, W. Modulation of stress-induced ACTH release by corticotropic-releasing factor, catecholamines and vasopressin. Nature 305:325-327; 1983. I I. Rivier, C.; Vale, W. Interaction ofcorticotropin-releasing factor and arginine-vasopressin on adrenocorticotropin secretion in vivo. Endocrinology 113:939-942; 1983.
12. Selye, H. The stress of life (revised edition). New York: McGrawHill Book Co.; 1978. 13. Szabo, S.; Glavin, G. B. Hans Selye and the concept of biological stress: Ulcer pathogenesis as a historical paradigm. Ann. NY Acad. Sci. 597:14-16; 1990. 14. Tannahill, L. A.; Dow, R. C.; Fairhall, K. M.; Robinson, I. C. A. F.; Fink, G. Comparison ofadrenocorticotropin control in Brattleboro, Long-Evans, and Wistar rats. Neuroendocrinology 48:650-657; 1988. 15. Valtin, H.; Sawyer, W. H.; Sokol, H. W. Neurohypophysial principles in rats homozygous and heterozygous for hypothalamic diabetes insipidus (Brattleboro strain). Endocrinology 77:701-706; 1965. 16. Valtin, H.; Schroeder, H. A. Familial hypothalamic diabetes insipidus in rats (Brattleboro strain). Am. J. Physiol. 206:425-430; 1964. 17. Whitnall, M. H. Stress selectively activates the vasopressin-containing subset of corticotropin-releasing hormone neurons. Neuroendocrinology 50:702-707; 1989. 18. Wideman, C. H.; Murphy, H. M. The effects of restraint and restraint plus intermittent shock on ulcer formation in Brattleboro rats. Physiol. Psychol. 11:78-80; 1983. 19. Wideman, C. H.; Murphy, H. M. Effects of vasopressin deficiency, age, and stress on stomach ulcer induction in rats. Peptides 6(Suppl. 1):63-67; 1985. 20. Wideman, C. H.; Murphy, H. M. Effects ofvasopressin replacement during food-restriction stress. Peptides 12:285-288; 1991. 21. Wiley, K. M.; Pearlmutter, A. F.; Miller, R. E. Decreased adrenal sensitivity to ACTH in the vasopressin-deficient Brattleboro rat. Neuroendocrinology 14:257-270; 1974. 22. Williams, J. D. Textbook of endocrinology (7th ed.) Philadelphia: W. B. Saunders Co.; 1985.
