- Email: [email protected]
Potentiating interactions between medullary serotonin and thyrotropin-releasing hormone-induced gastric erosions in rats
Neuroscience Letters, 161 (1993)199 202 © 1993 Elsevier Scientific Publishers Ireland Ltd. All rights reserved 0304-3940193l$06.00
199
NSL 09897
Potentiating interactions between medullary serotonin and thyrotropinreleasing hormone-induced gastric erosions in rats M a s a s h i Y o n e d a , Y v e t t e Tach6* CURE~Digestive Disease Centel; VA Wadsworth Medical Center. Building 115, Room 115. Wilshire and Sawlelle Boulevards. Los Angeles, CA ~,~0073. USA. Department of Medicine and Brain Research Institute. UCLA. Los An~,,eles. CA 90073. { 'SA
(Received 3 June 1993: Revised version 12 July 1993: Accepted 14 July 19931 Key words:
5-HT: Thyrotropin-releasing hormone; RX 77368: Gastric mucosal lesion: Central nervous system: Ketanserin: lndomethacin
The central interaction between 5-HT and exogenous and endogenous thyrotropin-releasing hormone (TRH)-induced gastric lesions was investigated in conscious rats. lntracisternal injection (i.c.) of the TRH analog, RX 77368, (2,5 nmolt in indomethacin (2 mg/kg, i.p. l-treated rats produced 1.4 _+0.1% of gastric corpus mucosal lesions which were aggravated by 30 and 208% by simultaneous i.c. injection 5-HT at 10 and 100 nmol, respectively, whereas i.v. 5-HT (100 nmol) had no effect. The 5-HT2~,cantagonist, ketanserin, given i.c. at 10 or 100 nmol reduced by 44 and 76%, respectively, cold restraint stress-induced 3.4 + 0.6% gastric lesions in indomethacin-pretreated rats whereas, when given i.v. (100 nmol), it was inactive. Ketanserin or 5-HT (100 nmol, i.c.) alone did not modit}, the gastric mucosa. The present data suggesl a potentiating interaction between endogenous 5-HT and TRH which has implication in the understanding of medullary mechanisms involved in gastric lesion tk~rmation induced by cold restraint.
Recent pharmacological studies suggest that serotonin (5-HT) interacts with T R H in the dorsal vagal complex (DVC) to potentiate the stimulation o f gastric secretory and m o t o r responses [17]. 5-HT microinjected into the D V C potentiates the stable T R H analog, R X 77368-induced acid secretion t h r o u g h interaction with 5-HT,/w since ketanserin injected i.c. reversed the potentiating effect o f 5-HT in the D V C whereas other 5-HT antagonists had no effect [21]. We previously showed that maximal effective doses o f T R H or R X 77368 injected i.c. induce gastric hemorrhagic lesions in fasted rats resulting from central vagal cholinergic activation [6, 17]. Cold exposure is k n o w n to release central T R H [9] and convergent functional and anatomical evidence indicate a role o f medullary T R H in the formation of gastric erosions in response to cold [13, 17, 20]. The purpose o f the present study was to assess whether medullary 5-HT influences the expression o f gastric d a m a g e induced by cold restraint stress to establish whether the interaction between 5-HT and T R H has physiological relevance. Experiments were conducted in indomethacin-pretreated rats to eliminate the c o n f o u n d "Corresponding author. Bldg. 115, Room 115, VA Wadsworth Medical Center, Wilshire and Sawtelle Blvds., Los Angeles, CA 90073, USA. Fax: (1) (310) 824 6752.
ing gastroprotective action o f gastric P G E , released by central vagal cholinergic activation [22, 23]. Male Sprague Dawley albino rats (220- 250 g: Harlan Laboratories, San Diego, CA) were housed under standard conditions. Rats were deprived o f food for 24 h with free access to water up to 2 h before treatment. In the first experiment, rats were pretreated with indomethacin (2 mg/kg) or vehicle injected i.p. and 30 rain later, rats were anesthetized with ether for i.c. or i.v. injection o f saline or 5-HT (10 or 100 mnol), followed immediately after by i.c. injection o f saline or T R H analog, RX 77368 (2.5 nmol). After awaking from anesthesia, rats were maintained in a single cage under non-stress conditions for 4 h, then sacrificed by CO~ inhalation. Stomachs were removed and examined for mucosal lesions. In the second experiment, rats pretreated with indomethacin (2 mg/kg, ip) or vehicle (ip) were anesthetized 20 min later with ether and injected i.c. or i.v. with saline or the 5-HT antagonist, ketanserin (10 or 100 nmol). Rats were placed into restraint stainless tube cages 10 rain alter i.c. or i.v. injection and maintained at 4°C for 3 h. Immediately after rats were sacrificed by COx and stomachs were removed for the examination o f gastric lesions. Gastric lesions were assessed by macroscopic criteria as previously described [22]. Each stomach was opened along the greater curvature, gently rinsed in saline,
200
pinned open to expose the mucosa. The corpus was photographed using a Polaroid pack film camera and Polaroid Type 108 film (Polaroid Tech. Photo, Cambridge, MA) at 2:1 magnification. To perform the image analysis, the corpus was divided into four areas and measurements were repeated twice. The lesion area (cm2), total area of corpus (cm 2) and percentage corpus mucosa containing lesions were determined by computerized imageanalyser device (MICRO/PDP-1 t, Digital Equipment, Maynard, MA) equipped with imaging boards (Imaging Technology, Woburn, MA). The validation of the imageanalysis technique to measure gastric lesions has been previously reported [14]. The following substances were used: the stable T R H analog, RX 77368 pGlu-His-(3,3"-dimethyl)-proNH2 (Reckitt and Colman, Kingston upon Hill, UK), 5-HT hydrochloride (Research Biochemicals, Natick, MA), the 5-HT2nc receptor antagonist, ketanserin tartrate [5], 3-(2-(4-fluorobenzoyl)-l-piperdinyl)-ethyl)-2,4(1H,3H)quinazolinedione tartrate (Research Biochemicals) and indomethacin (Sigma, St Louis, MO). Before each experiment, the stock solution of RX 77368 (l.5 nmol//A in 0.5% BSA, 0.9% saline) was diluted in saline and injected i.c. in 5 ~tl volume. 5-HT and ketanserin in powder form were freshly dissolved in saline before injection either i.c. in 5/11 or i.v. in 1 ml/kg. Indomethacin was dissolved in 1% N a H C O 3 solution and injected i.p. (1 ml/kg). The choice of doses for i.c. RX 77368, 5-HT and ketanserin was based on previous studies showing maximal gastric damage induced by RX 77368 [6] and a potentiating effect of central injection of 5-HT with RX 77368 on acid secretion and reversal of its effects by ketanserin [21]. Results are expressed as mean + S.E.M. Comparisons between two groups were calculated by Student's t test. Multiple group comparisons were performed by (x3
ANOVA followed by Dunnet's or D u n c a n s contrast. A P < 0.05 was considered statistically significant. In vehicle-pretreated rats, RX 77368 (2.5 nmol) rejected i.c. induced the formation of gastric lesions which was characterized macroscopically by long, dark red vertical lines primarily confined to the proximal corpus mucosa. In terms of total surface area, macroscopic lesions involved 0.8 + 0.1% of the corpus as evaluated 4 h after i.c. peptide injection (Fig. 1). In indomethacin-pretreated rats, gastric lesions induced by i.c. TRH analog were significantly increased (1.4 + 0. 1%; Fig. 1). Saline injected i.c. did not induce gastric lesions in rats pretreated with indomethacin (0 + 0%, n = 6). 5-HT (10 and 100 nmol) injected i.c. enhanced by 30 and 208%, respectively, gastric lesions induced by i.c. T R H analog (2.5 nmol) in indomethacin-pretreated rats (Fig. 1). 5-HT (100 ng) had no effect when injected i.v. under otherwise similar conditions (gastric lesions: RX 77368 plus i.v. saline: 1.9 + 0.3%, n = 12; RX 77368 plus i.v. 5-HT: 1.6 + 0.3%, n = 9, NS). 5-HT (100 mol, i.c.) alone did not influence the gastric mucosa (0.1 + 0.0%, n = 6). Exposure to cold stress for 3 h produced gastric lesions involving 0.9 + 0.4% of the gastric corpus mucosa in vehicle pretreated-rats (Fig. 2). Pretreatment with indomethacin (2 mg/kg, i.p.) significantly aggravated cold stress-induced gastric damage (3.4 + 0.6%, P < 0.01). In indomethacin-pretreated rats, ketanserin (10 and 100 nmol) injected i.c. reduced gastric damage by 44 and 76%, respectively (Fig. 2). The 5-HT antagonist (100 nmol) given i.v. under the same conditions had no effect (gastric lesions: cold plus i.v. saline: 2.3 _+ 0.4%, n = 11: cold plus ketanserin: 2.6 + 0.6%, n--11). Ketanserin (100 nmol) injected i.c, by itself had no effect on the gastric mucosa (data not shown). The T R H analog, RX 77368, injected i.c. at the 2.5
sl
I--3 Vehicle (ip) + RX 7 7 3 6 6 2.5 nrnol, ic 6.0Indomethecin (2 r n g / k g , ip) + RX 7 7 3 6 8 2.5 nmol, ic 5.0-
8
(10
i
g g
4.0-
3.0 t3_
L_
1.0-
O,_
0.0
..J
2.0
2.0-
c
#
4.0-
.3.0-
t_
I--7 Vehicle (ip) + Cold stress (4"C, 3 h) g2~ Indomethacin (2 mg/kg, ip) + Cold stress (4 °C, 3 h)
5.0-
C1 0
t.0 O-
10
,o
10 Serotonin Dose (nmol, ic)
Fig. 1. Effect of i.c. injection of 5-HT on i.c. R X 77368-induced gastric lesions. Each column represents mean + S.E.M. ~P < 0.05 compared with vehicle-pretreatment group. *P < 0.05, "'P < 0.01 compared with indomethacin-pretreated, i.c. R X 77368 control group.
I
10
0.0 100
o
9
_. 10 10
100
Ketanserin Dose (nmol, ic)
Fig. 2. Effect of i.c. injection of ketanserin on cold stress-induced gastric lesions. Each column represents m e a n + S.E.M. #P < 0.05 compared with vehicle pretreatment group. "P < 0:05, *'P < 0.01 compared with indomethacin-pretreated, i.c. saline control group.
201
nmol induced within 4 h gastric hemorrhagic lesions in conscious 24 h fasted rats as previously reported [6]. Peripheral administration of indomethacin given at a nonulcerogenic dose inhibiting prostaglandin synthesis [16] results in a two-fold enhancement of gastric lesions induced by i.c. TRH analog. We previously reported that i.c. injection of RX 77368 stimulates gastric PGE 2 released and that indomethacin blocked the PGE2 response [22, 23]. Taken together these data indicate that the stimulation of endogenous gastric prostaglandins plays a protective role in partly countering the expression of gastric damage induced by central TRH analog. These findings are consistent with the previous demonstration that the enhanced PGE2 release induced by i.c. injection of RX 77368 is biologically active to provide gastric cytoprotection against ethanol [22]. PGE2 protects the gastric mucosa against injury produced by various agents or stress through inhibiting acid and/or increasing mucus and bicarbonate secretion and blood flow [15]. Gastric damage induced by i.c. injection of RX 77368 is blocked by antisecretory drugs [6]. The increased gastric erosions by indomethacin may be related to the enhanced acid secretion since the acid response to i.c. injection RX 77368 is potentiated in indomethacin-pretreated rats [23]. When 5-HT was injected i.c. at 10 and 100 nmol immediately before TRH analog, gastric damage was enhanced by 46 and 209%, respectively, in indomethacinpretreated rats. The 209% enhancement of TRH analoginduced gastric lesions represents a centrally mediated potentiating effect of 5-HT. 5-HT (100 nmol) given i.c. alone did not produce gastric mucosal lesions and injected i.v. did not alter gastric erosions induced by i.c. RX 77368 in indomethacin-treated rats. Drugs delivered into the cisterna magna do not enter into the cerebral ventricle and act mainly in brainstem area [3]. Such a central interaction is most likely to occur in the DVC where high concentrations of TRH and 5-HT receptors have been localized [10, 18]. In addition, previous studies indicate that 5-HT acts in the DVC to potentiate the gastric motor and secretory responses to TRH or RX 77368 [11, 12, 21]. In addition, 5-HT has been shown to increase the spontaneously occurring action potential firing rate in dorsal motor nucleus neurons in rat brain stem slices [19]. Since the stimulation of gastric acid, pepsin and motility contributes to the development of gastric erosions [4, 6], the potentiation by 5-HT of the gastric secretory and motor responses to central TRH analog [11, 12, 21] may result in the observed enhanced gastric damage. Cold exposure produces gastric ulceration in conscious rats which was likewise increased by indomethacin. Gastric lesions induced by cold stress were redticed
dose-dependently by i.c. injection of the selective antagonist 5-HT2/~c, ketanserin [5] in indomethacin-pretreated rats. Ketanserin action represents a centrally mediated effect since the 5-HT antagonist given i.v. at 100 nmol did not influence lesion formation. Several sets of evidence are compatible with the possibility that ketanserin effect results from the blocking of the potentiating interactions between 5-HT and TRH in the DVC. ( 1) We have previously shown that i.c. ketanserin blocked the potentiating effect of 5-HT and RX 77368 in the DVC on the stimulation of acid secretion whereas other 5-HT antagonists acting at the 5-HTIA, 5-HT~B or 5-HT 3 receptors subtypes were inactive [21]. (2) Growing evidence indicates that medullary TRH is involved in mediating gastric damage induced by cold in fasted rats. Cold activates raphe pallidus and obscurus neurons projecting to the DVC and increases the expression of TRH mRNA in these nuclei and neuronal activity in the dorsal motor nucleus of the vagus [1, 20]. In addition. TRH antibody injected into the CSF prevents cold stress-induced acid secretion and gastric erosions [13]. (3) 5-HT and TRH are co-localized in medullary raphe nuclei projecting to the DVC [8] and there is evidence that electrical activation of raphe obscurus increases 5-HT release in the DVC [2] where high levels of 5-HTw receptors are expressed [18]. (4) Stress-induced gastric lesions were shown to increase 5-HT release in pons/medulla [7]. Taken together, these data provide evidence that the corelease of TRH and 5-HT in the DVC induced by cold can both contribute to gastric lesion formation through potentiating interactions. 5-HT can amplify the central vagally mediated gastric erosions induced by TRH analog whereas the 5-HT2/~c antagonist, ketanserin, decreased lesion formation induced by cold restraint in indomethacin-treated rats. These data suggest that endogenous 5-HT plays a role in cold stress-induced gastric lesions, probably through potentiating interactions with TRH in the DVC leading to enhanced vagal stimulation of gastric secretory and motor function and related lesion formation. In addition these studies indicate that gastric lesion formation induced by central vagal stimulation represents the net result of aggressive as well as defensive factors such as prostaglandins. Supported by the NIHDDK, Grants DK-30110 and DK 41301 (Blood Flow and Experimental Ulcer Core) and the NIMH Grant MH-00663. The authors thank J.G. Loyd-Jones (Reckitt & Collman, Kingston-uponHill, UK) for the generous donation of RX 77368. 1 Bonaz, B. and Tach6, Y.. C-Fos expression in specific brain nuclei during cold restraint stress-induced increase in colonic transit and gastric lesions in rats, Gastroenterology, 104 (1993) A43.
202
2 Brodin, E., Linderoth, B., Giony, M., Yamamoto, Y., Gazelius, B.. Millhorn, D.E., Hokfelt, T. and Ungerstedt, U., In vivo release of serotonin in cat dorsal vagal complex and cervical ventral horn by electrical stimulation of the medullary raphe nuclei. Brain Res.. 535 (19901 227 236. 3 Feldberg, W.S., Fifty years on: looking back on some developments in neurohumoral physiology, Liverpool University Press, Liverpool, 1982, 31 pp. 4 Garrick, T., Buack, S. and Bass, R, Gastric motilit? is a major factor in cold restraint-induced lesion formation in rats, Am. J. Physiol., 250 (1986) G191-G199. 5 Glennon, R.A. and Dukat, M., Serotonin receptors and their ligands: a lack of selective agents, Pharmacol. Biochem. Behav., 40 (19911 1009-1017. 6 Goto, Y. and Tach6, Y., Gastric erosions induced by intracisternal thyrotropin-releasing hormone (TRH) in rats, Peptides, 6 (1985) 153 156. 7 Hellhammer, D.H., Hingtgen, J.N., Wade, S.E., Shea, P.A. and Aprison, M.H., Serotonergic changes in specific areas of rat brain associated with activity-stress gastric lesions, Psychosom. Med., 45 (1983) t15 122. 8 Kachidian, E, Poulat, R, Marlier, L. and Privat, A., Immunohistochemical evidence for the coexistence of substance R thyrotropinreleasing hormone, GABA, methionin-enkephalin, and leucinenkephalin in the serotonergic neurons of the caudal raphe nuclei: a dual labeling in the rat, J. Neurosci. Res., 30 (199l) 521 530. 9 Lin, M.T., Wang, ES., Chuang, J., Fan, L.J. and Won, S.J., Cold stress or a pyrogenic substance elevates thyrotropin-releasing hormone levels in the rat hypothalamus and induces thermogenic reactions, Neuroendocrinology, 50 (1989) 177 181. 10 Manaker, S. and Rizio, G., Autoradiographic localization of thyrotropin-releasing hormone and subtance P receptors in the rat dorsal vagal complex, J. Comp. Neurol., 290 (1989) 5t6-526. 11 McCann, M.J., Hermann, G.E. and Rogers, R.C., Dorsal medullary serotonin and gastric motility: enhancement of effects by thyrotropin-releasing hormone, J. Aut. Nerv. Syst., 25 (1988) 3540. t2 McTigue, D.M., Rogers, R.C. and Stephens, R.L., Thyrotropin releasing hormone analogue and serotonin interact within the dorsal vagal complex to augment gastric acid secretion, Neurosci. Letl.. 144 (1992) 61 64.
13 Niida, H., Takeuchi, K. and Okabe, S., Role of thyrotropin-releasing hormone in acid secretory response induced by lowering of body temperature in the rat, Eur. J. Pharmacol., 198 !1991) 137 t42. 14 Robert, A., Leung, F.W. and Guth, EH., Morphological and lun~tional gastric cytoprotection by prostaglandin m rats receiving absolute ethanol orally, Gut, 33 t 1992)444 451 15 Robert, A.. Nezamis, J.E.. Lancaster, C. a~ld Hanchar, A.J. (ktoprotection by prostaglandins m rats. Prevmmon of gastric necrosis produced by alcohol, HCL, NaOH, hypertonic NaC1, and thermal injury, Gastroenterology, 77 (1979) 433 443. 16 Saperas, E., Kauffman, G. and/'ache, Y., Role of central prostaglandin E: in the regulation of gastric acid secretion in the rat, Eur. J. Pharmacol., 209 (1991) 1 7 17 Tache, Y., Yang, H. and Yoneda. M., Vagal regulation of gastric function involves thyrotropin-releasing hormone in the medulary raphe nuclei and dorsal vagal complex, Digestion, 54 (1993) 65 72. 18 Thor, K.B., Blitz-Siebert, A. and Helke, C.J., Autoradiographic localization of 5HT~ binding sites in autonomic areas of the rat dorsomedial medulla oblongata, Synaps, 10 (1992) 217- 227. 19 Travagli, R.A. and Gillis, R.A.. Effects of serotonin alone, and its interaction with thyrotropin-releasing hormone on neurons in rat dorsal motor nucleus of the wtgus, J. Pharmacol. Exp. Ther., in press. 20 Yang, H., Wu, V. and Tachd, Yi, Thyrotropin-releasing hormone (TRH) mRNA expression is increased in medullary raphe nuclei during cold-induced gastric ulcer formation. Gastroenterology, 102 ( 1991 ) A767. 21 Yoneda, M. and Tach6, Y., Serotonin (5-HT) in the dorsal vagat complex (DVC) potentiates gastric acid stimulation induced by TRH analog through 5-HT2/Ic receptor in rats, Gastroenterology, 102 (1992) A768. 22 Yoneda, M. and Tach6, Y., Central thyrotropin-releasing factor analog prevents ethanol-induced gastric damage through prostaglandins in rats, Gastroenterology, 102 (1992) 1568-1574. 23 Yoneda, M. and Tach6, Y., Vagal regulation of gastric prostaglandin E, release by central TRH in rats, Am. J. Physiol., 264 11993) G231-G236.
199
NSL 09897
Potentiating interactions between medullary serotonin and thyrotropinreleasing hormone-induced gastric erosions in rats M a s a s h i Y o n e d a , Y v e t t e Tach6* CURE~Digestive Disease Centel; VA Wadsworth Medical Center. Building 115, Room 115. Wilshire and Sawlelle Boulevards. Los Angeles, CA ~,~0073. USA. Department of Medicine and Brain Research Institute. UCLA. Los An~,,eles. CA 90073. { 'SA
(Received 3 June 1993: Revised version 12 July 1993: Accepted 14 July 19931 Key words:
5-HT: Thyrotropin-releasing hormone; RX 77368: Gastric mucosal lesion: Central nervous system: Ketanserin: lndomethacin
The central interaction between 5-HT and exogenous and endogenous thyrotropin-releasing hormone (TRH)-induced gastric lesions was investigated in conscious rats. lntracisternal injection (i.c.) of the TRH analog, RX 77368, (2,5 nmolt in indomethacin (2 mg/kg, i.p. l-treated rats produced 1.4 _+0.1% of gastric corpus mucosal lesions which were aggravated by 30 and 208% by simultaneous i.c. injection 5-HT at 10 and 100 nmol, respectively, whereas i.v. 5-HT (100 nmol) had no effect. The 5-HT2~,cantagonist, ketanserin, given i.c. at 10 or 100 nmol reduced by 44 and 76%, respectively, cold restraint stress-induced 3.4 + 0.6% gastric lesions in indomethacin-pretreated rats whereas, when given i.v. (100 nmol), it was inactive. Ketanserin or 5-HT (100 nmol, i.c.) alone did not modit}, the gastric mucosa. The present data suggesl a potentiating interaction between endogenous 5-HT and TRH which has implication in the understanding of medullary mechanisms involved in gastric lesion tk~rmation induced by cold restraint.
Recent pharmacological studies suggest that serotonin (5-HT) interacts with T R H in the dorsal vagal complex (DVC) to potentiate the stimulation o f gastric secretory and m o t o r responses [17]. 5-HT microinjected into the D V C potentiates the stable T R H analog, R X 77368-induced acid secretion t h r o u g h interaction with 5-HT,/w since ketanserin injected i.c. reversed the potentiating effect o f 5-HT in the D V C whereas other 5-HT antagonists had no effect [21]. We previously showed that maximal effective doses o f T R H or R X 77368 injected i.c. induce gastric hemorrhagic lesions in fasted rats resulting from central vagal cholinergic activation [6, 17]. Cold exposure is k n o w n to release central T R H [9] and convergent functional and anatomical evidence indicate a role o f medullary T R H in the formation of gastric erosions in response to cold [13, 17, 20]. The purpose o f the present study was to assess whether medullary 5-HT influences the expression o f gastric d a m a g e induced by cold restraint stress to establish whether the interaction between 5-HT and T R H has physiological relevance. Experiments were conducted in indomethacin-pretreated rats to eliminate the c o n f o u n d "Corresponding author. Bldg. 115, Room 115, VA Wadsworth Medical Center, Wilshire and Sawtelle Blvds., Los Angeles, CA 90073, USA. Fax: (1) (310) 824 6752.
ing gastroprotective action o f gastric P G E , released by central vagal cholinergic activation [22, 23]. Male Sprague Dawley albino rats (220- 250 g: Harlan Laboratories, San Diego, CA) were housed under standard conditions. Rats were deprived o f food for 24 h with free access to water up to 2 h before treatment. In the first experiment, rats were pretreated with indomethacin (2 mg/kg) or vehicle injected i.p. and 30 rain later, rats were anesthetized with ether for i.c. or i.v. injection o f saline or 5-HT (10 or 100 mnol), followed immediately after by i.c. injection o f saline or T R H analog, RX 77368 (2.5 nmol). After awaking from anesthesia, rats were maintained in a single cage under non-stress conditions for 4 h, then sacrificed by CO~ inhalation. Stomachs were removed and examined for mucosal lesions. In the second experiment, rats pretreated with indomethacin (2 mg/kg, ip) or vehicle (ip) were anesthetized 20 min later with ether and injected i.c. or i.v. with saline or the 5-HT antagonist, ketanserin (10 or 100 nmol). Rats were placed into restraint stainless tube cages 10 rain alter i.c. or i.v. injection and maintained at 4°C for 3 h. Immediately after rats were sacrificed by COx and stomachs were removed for the examination o f gastric lesions. Gastric lesions were assessed by macroscopic criteria as previously described [22]. Each stomach was opened along the greater curvature, gently rinsed in saline,
200
pinned open to expose the mucosa. The corpus was photographed using a Polaroid pack film camera and Polaroid Type 108 film (Polaroid Tech. Photo, Cambridge, MA) at 2:1 magnification. To perform the image analysis, the corpus was divided into four areas and measurements were repeated twice. The lesion area (cm2), total area of corpus (cm 2) and percentage corpus mucosa containing lesions were determined by computerized imageanalyser device (MICRO/PDP-1 t, Digital Equipment, Maynard, MA) equipped with imaging boards (Imaging Technology, Woburn, MA). The validation of the imageanalysis technique to measure gastric lesions has been previously reported [14]. The following substances were used: the stable T R H analog, RX 77368 pGlu-His-(3,3"-dimethyl)-proNH2 (Reckitt and Colman, Kingston upon Hill, UK), 5-HT hydrochloride (Research Biochemicals, Natick, MA), the 5-HT2nc receptor antagonist, ketanserin tartrate [5], 3-(2-(4-fluorobenzoyl)-l-piperdinyl)-ethyl)-2,4(1H,3H)quinazolinedione tartrate (Research Biochemicals) and indomethacin (Sigma, St Louis, MO). Before each experiment, the stock solution of RX 77368 (l.5 nmol//A in 0.5% BSA, 0.9% saline) was diluted in saline and injected i.c. in 5 ~tl volume. 5-HT and ketanserin in powder form were freshly dissolved in saline before injection either i.c. in 5/11 or i.v. in 1 ml/kg. Indomethacin was dissolved in 1% N a H C O 3 solution and injected i.p. (1 ml/kg). The choice of doses for i.c. RX 77368, 5-HT and ketanserin was based on previous studies showing maximal gastric damage induced by RX 77368 [6] and a potentiating effect of central injection of 5-HT with RX 77368 on acid secretion and reversal of its effects by ketanserin [21]. Results are expressed as mean + S.E.M. Comparisons between two groups were calculated by Student's t test. Multiple group comparisons were performed by (x3
ANOVA followed by Dunnet's or D u n c a n s contrast. A P < 0.05 was considered statistically significant. In vehicle-pretreated rats, RX 77368 (2.5 nmol) rejected i.c. induced the formation of gastric lesions which was characterized macroscopically by long, dark red vertical lines primarily confined to the proximal corpus mucosa. In terms of total surface area, macroscopic lesions involved 0.8 + 0.1% of the corpus as evaluated 4 h after i.c. peptide injection (Fig. 1). In indomethacin-pretreated rats, gastric lesions induced by i.c. TRH analog were significantly increased (1.4 + 0. 1%; Fig. 1). Saline injected i.c. did not induce gastric lesions in rats pretreated with indomethacin (0 + 0%, n = 6). 5-HT (10 and 100 nmol) injected i.c. enhanced by 30 and 208%, respectively, gastric lesions induced by i.c. T R H analog (2.5 nmol) in indomethacin-pretreated rats (Fig. 1). 5-HT (100 ng) had no effect when injected i.v. under otherwise similar conditions (gastric lesions: RX 77368 plus i.v. saline: 1.9 + 0.3%, n = 12; RX 77368 plus i.v. 5-HT: 1.6 + 0.3%, n = 9, NS). 5-HT (100 mol, i.c.) alone did not influence the gastric mucosa (0.1 + 0.0%, n = 6). Exposure to cold stress for 3 h produced gastric lesions involving 0.9 + 0.4% of the gastric corpus mucosa in vehicle pretreated-rats (Fig. 2). Pretreatment with indomethacin (2 mg/kg, i.p.) significantly aggravated cold stress-induced gastric damage (3.4 + 0.6%, P < 0.01). In indomethacin-pretreated rats, ketanserin (10 and 100 nmol) injected i.c. reduced gastric damage by 44 and 76%, respectively (Fig. 2). The 5-HT antagonist (100 nmol) given i.v. under the same conditions had no effect (gastric lesions: cold plus i.v. saline: 2.3 _+ 0.4%, n = 11: cold plus ketanserin: 2.6 + 0.6%, n--11). Ketanserin (100 nmol) injected i.c, by itself had no effect on the gastric mucosa (data not shown). The T R H analog, RX 77368, injected i.c. at the 2.5
sl
I--3 Vehicle (ip) + RX 7 7 3 6 6 2.5 nrnol, ic 6.0Indomethecin (2 r n g / k g , ip) + RX 7 7 3 6 8 2.5 nmol, ic 5.0-
8
(10
i
g g
4.0-
3.0 t3_
L_
1.0-
O,_
0.0
..J
2.0
2.0-
c
#
4.0-
.3.0-
t_
I--7 Vehicle (ip) + Cold stress (4"C, 3 h) g2~ Indomethacin (2 mg/kg, ip) + Cold stress (4 °C, 3 h)
5.0-
C1 0
t.0 O-
10
,o
10 Serotonin Dose (nmol, ic)
Fig. 1. Effect of i.c. injection of 5-HT on i.c. R X 77368-induced gastric lesions. Each column represents mean + S.E.M. ~P < 0.05 compared with vehicle-pretreatment group. *P < 0.05, "'P < 0.01 compared with indomethacin-pretreated, i.c. R X 77368 control group.
I
10
0.0 100
o
9
_. 10 10
100
Ketanserin Dose (nmol, ic)
Fig. 2. Effect of i.c. injection of ketanserin on cold stress-induced gastric lesions. Each column represents m e a n + S.E.M. #P < 0.05 compared with vehicle pretreatment group. "P < 0:05, *'P < 0.01 compared with indomethacin-pretreated, i.c. saline control group.
201
nmol induced within 4 h gastric hemorrhagic lesions in conscious 24 h fasted rats as previously reported [6]. Peripheral administration of indomethacin given at a nonulcerogenic dose inhibiting prostaglandin synthesis [16] results in a two-fold enhancement of gastric lesions induced by i.c. TRH analog. We previously reported that i.c. injection of RX 77368 stimulates gastric PGE 2 released and that indomethacin blocked the PGE2 response [22, 23]. Taken together these data indicate that the stimulation of endogenous gastric prostaglandins plays a protective role in partly countering the expression of gastric damage induced by central TRH analog. These findings are consistent with the previous demonstration that the enhanced PGE2 release induced by i.c. injection of RX 77368 is biologically active to provide gastric cytoprotection against ethanol [22]. PGE2 protects the gastric mucosa against injury produced by various agents or stress through inhibiting acid and/or increasing mucus and bicarbonate secretion and blood flow [15]. Gastric damage induced by i.c. injection of RX 77368 is blocked by antisecretory drugs [6]. The increased gastric erosions by indomethacin may be related to the enhanced acid secretion since the acid response to i.c. injection RX 77368 is potentiated in indomethacin-pretreated rats [23]. When 5-HT was injected i.c. at 10 and 100 nmol immediately before TRH analog, gastric damage was enhanced by 46 and 209%, respectively, in indomethacinpretreated rats. The 209% enhancement of TRH analoginduced gastric lesions represents a centrally mediated potentiating effect of 5-HT. 5-HT (100 nmol) given i.c. alone did not produce gastric mucosal lesions and injected i.v. did not alter gastric erosions induced by i.c. RX 77368 in indomethacin-treated rats. Drugs delivered into the cisterna magna do not enter into the cerebral ventricle and act mainly in brainstem area [3]. Such a central interaction is most likely to occur in the DVC where high concentrations of TRH and 5-HT receptors have been localized [10, 18]. In addition, previous studies indicate that 5-HT acts in the DVC to potentiate the gastric motor and secretory responses to TRH or RX 77368 [11, 12, 21]. In addition, 5-HT has been shown to increase the spontaneously occurring action potential firing rate in dorsal motor nucleus neurons in rat brain stem slices [19]. Since the stimulation of gastric acid, pepsin and motility contributes to the development of gastric erosions [4, 6], the potentiation by 5-HT of the gastric secretory and motor responses to central TRH analog [11, 12, 21] may result in the observed enhanced gastric damage. Cold exposure produces gastric ulceration in conscious rats which was likewise increased by indomethacin. Gastric lesions induced by cold stress were redticed
dose-dependently by i.c. injection of the selective antagonist 5-HT2/~c, ketanserin [5] in indomethacin-pretreated rats. Ketanserin action represents a centrally mediated effect since the 5-HT antagonist given i.v. at 100 nmol did not influence lesion formation. Several sets of evidence are compatible with the possibility that ketanserin effect results from the blocking of the potentiating interactions between 5-HT and TRH in the DVC. ( 1) We have previously shown that i.c. ketanserin blocked the potentiating effect of 5-HT and RX 77368 in the DVC on the stimulation of acid secretion whereas other 5-HT antagonists acting at the 5-HTIA, 5-HT~B or 5-HT 3 receptors subtypes were inactive [21]. (2) Growing evidence indicates that medullary TRH is involved in mediating gastric damage induced by cold in fasted rats. Cold activates raphe pallidus and obscurus neurons projecting to the DVC and increases the expression of TRH mRNA in these nuclei and neuronal activity in the dorsal motor nucleus of the vagus [1, 20]. In addition. TRH antibody injected into the CSF prevents cold stress-induced acid secretion and gastric erosions [13]. (3) 5-HT and TRH are co-localized in medullary raphe nuclei projecting to the DVC [8] and there is evidence that electrical activation of raphe obscurus increases 5-HT release in the DVC [2] where high levels of 5-HTw receptors are expressed [18]. (4) Stress-induced gastric lesions were shown to increase 5-HT release in pons/medulla [7]. Taken together, these data provide evidence that the corelease of TRH and 5-HT in the DVC induced by cold can both contribute to gastric lesion formation through potentiating interactions. 5-HT can amplify the central vagally mediated gastric erosions induced by TRH analog whereas the 5-HT2/~c antagonist, ketanserin, decreased lesion formation induced by cold restraint in indomethacin-treated rats. These data suggest that endogenous 5-HT plays a role in cold stress-induced gastric lesions, probably through potentiating interactions with TRH in the DVC leading to enhanced vagal stimulation of gastric secretory and motor function and related lesion formation. In addition these studies indicate that gastric lesion formation induced by central vagal stimulation represents the net result of aggressive as well as defensive factors such as prostaglandins. Supported by the NIHDDK, Grants DK-30110 and DK 41301 (Blood Flow and Experimental Ulcer Core) and the NIMH Grant MH-00663. The authors thank J.G. Loyd-Jones (Reckitt & Collman, Kingston-uponHill, UK) for the generous donation of RX 77368. 1 Bonaz, B. and Tach6, Y.. C-Fos expression in specific brain nuclei during cold restraint stress-induced increase in colonic transit and gastric lesions in rats, Gastroenterology, 104 (1993) A43.
202
2 Brodin, E., Linderoth, B., Giony, M., Yamamoto, Y., Gazelius, B.. Millhorn, D.E., Hokfelt, T. and Ungerstedt, U., In vivo release of serotonin in cat dorsal vagal complex and cervical ventral horn by electrical stimulation of the medullary raphe nuclei. Brain Res.. 535 (19901 227 236. 3 Feldberg, W.S., Fifty years on: looking back on some developments in neurohumoral physiology, Liverpool University Press, Liverpool, 1982, 31 pp. 4 Garrick, T., Buack, S. and Bass, R, Gastric motilit? is a major factor in cold restraint-induced lesion formation in rats, Am. J. Physiol., 250 (1986) G191-G199. 5 Glennon, R.A. and Dukat, M., Serotonin receptors and their ligands: a lack of selective agents, Pharmacol. Biochem. Behav., 40 (19911 1009-1017. 6 Goto, Y. and Tach6, Y., Gastric erosions induced by intracisternal thyrotropin-releasing hormone (TRH) in rats, Peptides, 6 (1985) 153 156. 7 Hellhammer, D.H., Hingtgen, J.N., Wade, S.E., Shea, P.A. and Aprison, M.H., Serotonergic changes in specific areas of rat brain associated with activity-stress gastric lesions, Psychosom. Med., 45 (1983) t15 122. 8 Kachidian, E, Poulat, R, Marlier, L. and Privat, A., Immunohistochemical evidence for the coexistence of substance R thyrotropinreleasing hormone, GABA, methionin-enkephalin, and leucinenkephalin in the serotonergic neurons of the caudal raphe nuclei: a dual labeling in the rat, J. Neurosci. Res., 30 (199l) 521 530. 9 Lin, M.T., Wang, ES., Chuang, J., Fan, L.J. and Won, S.J., Cold stress or a pyrogenic substance elevates thyrotropin-releasing hormone levels in the rat hypothalamus and induces thermogenic reactions, Neuroendocrinology, 50 (1989) 177 181. 10 Manaker, S. and Rizio, G., Autoradiographic localization of thyrotropin-releasing hormone and subtance P receptors in the rat dorsal vagal complex, J. Comp. Neurol., 290 (1989) 5t6-526. 11 McCann, M.J., Hermann, G.E. and Rogers, R.C., Dorsal medullary serotonin and gastric motility: enhancement of effects by thyrotropin-releasing hormone, J. Aut. Nerv. Syst., 25 (1988) 3540. t2 McTigue, D.M., Rogers, R.C. and Stephens, R.L., Thyrotropin releasing hormone analogue and serotonin interact within the dorsal vagal complex to augment gastric acid secretion, Neurosci. Letl.. 144 (1992) 61 64.
13 Niida, H., Takeuchi, K. and Okabe, S., Role of thyrotropin-releasing hormone in acid secretory response induced by lowering of body temperature in the rat, Eur. J. Pharmacol., 198 !1991) 137 t42. 14 Robert, A., Leung, F.W. and Guth, EH., Morphological and lun~tional gastric cytoprotection by prostaglandin m rats receiving absolute ethanol orally, Gut, 33 t 1992)444 451 15 Robert, A.. Nezamis, J.E.. Lancaster, C. a~ld Hanchar, A.J. (ktoprotection by prostaglandins m rats. Prevmmon of gastric necrosis produced by alcohol, HCL, NaOH, hypertonic NaC1, and thermal injury, Gastroenterology, 77 (1979) 433 443. 16 Saperas, E., Kauffman, G. and/'ache, Y., Role of central prostaglandin E: in the regulation of gastric acid secretion in the rat, Eur. J. Pharmacol., 209 (1991) 1 7 17 Tache, Y., Yang, H. and Yoneda. M., Vagal regulation of gastric function involves thyrotropin-releasing hormone in the medulary raphe nuclei and dorsal vagal complex, Digestion, 54 (1993) 65 72. 18 Thor, K.B., Blitz-Siebert, A. and Helke, C.J., Autoradiographic localization of 5HT~ binding sites in autonomic areas of the rat dorsomedial medulla oblongata, Synaps, 10 (1992) 217- 227. 19 Travagli, R.A. and Gillis, R.A.. Effects of serotonin alone, and its interaction with thyrotropin-releasing hormone on neurons in rat dorsal motor nucleus of the wtgus, J. Pharmacol. Exp. Ther., in press. 20 Yang, H., Wu, V. and Tachd, Yi, Thyrotropin-releasing hormone (TRH) mRNA expression is increased in medullary raphe nuclei during cold-induced gastric ulcer formation. Gastroenterology, 102 ( 1991 ) A767. 21 Yoneda, M. and Tach6, Y., Serotonin (5-HT) in the dorsal vagat complex (DVC) potentiates gastric acid stimulation induced by TRH analog through 5-HT2/Ic receptor in rats, Gastroenterology, 102 (1992) A768. 22 Yoneda, M. and Tach6, Y., Central thyrotropin-releasing factor analog prevents ethanol-induced gastric damage through prostaglandins in rats, Gastroenterology, 102 (1992) 1568-1574. 23 Yoneda, M. and Tach6, Y., Vagal regulation of gastric prostaglandin E, release by central TRH in rats, Am. J. Physiol., 264 11993) G231-G236.