- Email: [email protected]
Alterations in gastric acid secretion following hepatic portal injections of d -glucose and its anomers
Journal of the Autonomic Nervous System, 5 (1982) 337-344
337
Elsevier BiomedicalPress
Alterations in gastric acid secretion following hepatic portal injections of D-glucose and its anomers Takeo Sakaguchi Department of Physiology, Niigata University School of Medicine, Niigata (Japan)
(ReceivedJuly 19th, 1981) (AcceptedOctober 30th, 1981)
Key words: blood gluc0se--gastric acid secretion--stomach--glucose anomers--
vagus nerve--liver
Abstract
Changes in acid outputs from the stomach were examined after portal injections of D-glucose and its optical anomers in the bilaterally adrenalectomized rats with insulin hypoglycaemia. Significant decrease in gastric acid outputs was noted after portal injections of a-D-glucose, optically equilibrated D-glucose (OEDG) consisting of 36% a-anomer and 64% fl-anomer and fl-D-glucose. The effect of fl-D-glucose was most potent in reducing the acid outputs and the inhibitory response was entirely prevented by prior vagotomy at the hepatic level. The injections of isotonic NaCI solution, however, produced no change in the acid outputs. Results suggest that changes in glucose levels in the portal vein may modulate gastric acid secretion through hepatic vagal afferents and gastric vagal efferents and suggest that activation of hepatic glucosensitive mechanisms may be dependent on the anomeric stereospecificity of D-glucose in the blood.
Introduction
Electrophysiological studies have revealed that D-glucose injected into the portal vein affects tonic activities of the vagus nerve innervating the stomach [17]. This, at least, suggested a neural connection existing between hepatic vagal afferents and gastric vagal efferents in the brain. Considering the observation together with the 0165-1838/82/0000-0000/$02.75 © 1982ElsevierBiomedicalPress
338
finding that electrical stimulation of the proximal cut end of the hepatic vagus nerve changes acidity in the gastric perfusing fluid [18], it has also been speculated that hepatic afferent glucose signals may relate to gastric acid secretion through the central vagal connection. However, there is no report concerning the effect of portal glucose injection on gastric acid secretion. On the other hand, D-glucosedissolved in aqueous solution has been known to exist as an optically equilibrium mixture of its two anomers; 36% a-anomer and 64% fl-anomer [15]. Moreover, almost the same anomeric composition has been shown in blood [7, t0]. The present experiments were thus designed to investigate whether D-gluc0se injected into the portal vein influences gastric acid secretion, especially in consideration of the D-glucose anomers.
Materials and Methods
One hundred and eighteen male rats weighing about 300 g were used. They were fasted for 22 h before each experiment under controlled conditions of room temperature (22°C) and light (12:12 light-dark cycle), although allowed tree access to tap water. The animals anaesthetized with pentobarbital sodium (45 mg/kg, i.p.) were adrenalectomized bilaterally: this diminished scatter in plasma levels of glucose and insulin [19] and prevented inlu'bition of gastric acid secretion by the adrenalin which would have been released in response to hypoglycaemia [1]. The experiments were further performed in the morning to eliminate any diurnal variation in levels of glucose and insulin [9]. The experimental preparation for gastric acid estimation was based on the methods described earlier [3,6]. After a tracheal intubation, a polyethylene tube was introduced into the stomach through the esophagus and was tied in position by a ligature around the esophagus in the neck. The abdominal cavity was opened by a mid-line incision. A cannula was then inserted in the pyloro-duodenal junction through an incision and passed up into the stomach. The cannula was tied into position by purse string sutures around the junction [6]. In some experiments, loose threads were set around the hepatic branch of the vagus nerve or around the left vagus nerve in the neck and both ends of each thread led through a plastic tube; the nerves could be cut easily by pulling the loop of thread. Residual food particles in the stomach were washed out with warm saline. Throughout the experiment body temperature of each animal was kept at 36 - 0.5°C by a heating lamp. Physiological saline in a bottle placed 3 cm above the level of the animal was warmed at 36°C and it was gravitationally perfused through the stomach into a collecting tube. The perfusion rate was 2.6 ml,/min and as the gastric perfusate was gathered into the tube, the volume of each sample was measured and titratable acidity (end point pH 7.0) was determined using N/100 NaOH [3]. With this system acid output was calculated every 3-15 min. D-Glucose and NaC1 dissolved in distilled water kept at 36°C were infused through a catheter placed in the portal vein. a- and/~-anomers of crystallized D-glucose were dissolved in the water immediately before use to eliminate mutarotation of the glucose. Purity of the two anomers was more than 98% a n d the rate o f mutarotation of the two anomers in the water was negligible. Optically eqe/librated
339 D-glucose solution consisting of 36% a-D-glucose and 64% B-D-glucose was obtained by incubating either a- or B-D-glucose solution for 24 h at room temperature (22°C) to induce the glucose to mutarotate completely. The amount of a portal injection was 0.2 ml and each injection lasted 15 s via an infusion pump. Insulin (regular insulin: N o v o Ind., Copenhagen) was injected through a catheter inserted in the right jugular vein by an infusion pump. The concentration of glucose in the blood was measured by the method reported earlier [20]. Blood for glucose estimation (0.1 ml) was drawn through the same catheter in the jugular vein. Data were collected from the first response of each animal to a certain drug. Statistical significance of differences between values were determined by Student's t-test: P < 0.05 was taken as indicative of a significance between means.
Results When gastric acid outputs were measured every 15 min following insulin injections, the cumulative levels of acid outputs (the mean-+ S.E.M.) 90 min after the injections of 0.5, 1.0 and 2.0 U / k g / h insulin were 48.5-+ 1.2 (n = 9), 54.5 -+ 0.7 (n = 9) and 46.3 -+ 1.1 (n = 10)/~-equiv/90 min, respectively (Fig. 1). The acid output produced by 1.0 U / k g / h insulin injections was significantly ( P < 0.02 and P < 0.01) greater than that induced by 0.5 and 2.0 U / k g / h insulin injections. The patterns of acid output by the 3 insulin injections were also different from each other and a relatively constant stimulated acid output was observed during 60-90 min only after
'oo t
ii '°i]
611 6iI 6i I 611 6i I 611
20"
gT
o~
It
15
i
t
i
6 6 iI I
Oil
U
6, i 6i i
7o Time
6'0
IO
(mln)
Fig. 1. Time course of plasma levels of glucose and gastric acid output after administrations of 0.5 (©: n=9), 1.0 (×: n=9) and 2.0 (O: n= 10) U/kg/h insulin. Horizontal bar represents duration of the insulin injections. Values are the mean+ S.E.M. A relativelyconstant enhanced acid output was noted for 60-90 min after 1.0 U/kg/h insulin injections.
340
3 2=. 200 E
100
~Ai*
6AA
Q
~i
0
200 "~
10
e~
0 15
l
i L
,
6Al
5
=~
10
"~ ._> •-~ g
5
0
.~_
io
io
Time (min)
o
0 0
60
7'5 Time (min)
Fig. 2. Denervation effect of the vagus nerve on the gastric acid outputs evoked by 1.0 U / k g / h insulin ( n = l l). Arrow indicates division of the left vagus nerve. Horizontal bar Shows duration of the insulin injection. Values are the m e a n +- S,E.M. * P<0.05, ** P<0.01: significantly different from the vatue just before vagal denervations. Fig. 3. The effect of portal injections of 30 ( ~ : n = l l ) , 60 ( A : n = 1 2 ) and 120 (&: n = 8 ) m g / k g equilibrated D-glucose on the gastric acid outputs caused by 1.0 U / k g / h insulin. Arrow indicates the time of portal injections. Horizontal bar shows duration of the insulin administration, Values are meansm S.E,M. * P <0.05, ** P <0.02, *** P <0.01 : significantly different from the value immediately before injection:
the 1.0 U / k g / h insulin injection. Acid outputs were then calculated every 5 min following the 1.0 U / k g / h insulin administration and it was observed that the enhanced acid outputs by the insulin injection were significantly inhibited by left vagotomy at the cervical level (Fig. 2). From the two preliminary findings, successive administration of 1.0 U / k g / h insulin was employed in the following experiments and portal test injections were always begun about 60 min after the insulin injection. Optically equilibrated D-glucose (OEDG) was portally injected at various doses from 30 to 120 m g / k g (Fig. 3). The levels of acid output 0 and 3 min before portal injections of the glucose were not significantly different from each other. The acid output 3 min after the injections of 30 and 60 mg,/kg OEDG was reduced significantly compared to the level before injection without changes in the blood glucose levels and the inhibitory response had disappeared 6 min after the injections. A sustained suppression of acid output was observed after the 120 m g / k g OEDG injections; however, the glucose levels 3 min after the injection increased significantly to the level 3 rain before injection. In the rats with vagotomy at the hepatic level, no change in the acid output was
341 4 : E
5 "Z E
10 ~ f
ab
~
6, 6,
5
0
*i 0
1
!
s
2
g,
>
.:~
11
!
10
0
60
*i
O
7'5 Time(min)
o
61 64 6, "6
0
0
60
7'5 Time (rain)
Fig. 4. The effect of portal injections of equilibrated D-glucoseon the gastric acid outputs evoked by 1.0 U/kg/h insulin in the hepatic vagotomized rats. Sixty (/x: n = 9) and 120 (A: n = 8) mg/kg equilibrated D-glucose were portally injected. The NaC1 solutions (11: n=7) used were equitonic to 60 mg/kg D-glucose solutions. Arrows show the hepatic vagotomy (a) and the portal injections (b). Horizontal bar represents duration of the insulin injection. Values are the mean±S.E.M. * P<0.02: significantly different from the value immediately before injection. Fig. 5. Changes in the acid outputs from the stomach in response to portal injections of 60 mg/kg a-D-glucose (~: n= 13) and 60 mg/kg fl-D-glucose ( , : n= 11) in the rats treated with 1.0 U/kg/h insulin. Horizontal bar expresses duration of the insulin injection. Arrow indicates the time of injection. Values are means±S.E.M. * P<0.05. ** P<0~01: significantly different from the value immediately before injection.
noted after the 60 m g / k g O E D G injections. Significant depression in the acid o u t p u t was elicited 3 and 6 min after the 120 m g / k g O E D G injections in spite of division of the hepatic vagus nerve. The NaC1 solutions, which are equitonic to 60 m g / k g O E D G solutions, caused no change in the acid output (Fig. 4). The level of gastric acid output 3 m i n after portal injections of 60 m g / k g a-D-glucose was decreased significantly c o m p a r e d to the level before injection, whereas the 60 m g / k g B-D-glucose injections also reduced significantly the acid output levels 3 min after the injection (Fig. 5). The injections of the B-D-glucose seemed to induce a further reduction in the acid output; the change by the /3-D-glucose injections was greater ( P < 0.01) than that by the a-D-glucose injections.
Discussion
Electrophysiological studies have suggested that hepatic glucose signals may affect gastric function through a central vagal pathway existing between the liver and the stomach [17]. In addition, afferent hepatic signals have been shown to influence gastric acid secretion, because electrical stimulation to the hepatic vagus nerve increased gastric acid secretion [18]. In the experiment, electrical stimulation
342 given to the nerve with elastic electrodes [ 16] has been considered to be equivalent to hypoglycaemia in the portal vein, especially since the inverse relationship between afferent hepatic vagus discharge and the glucose level in the portal perfusing fluid was established [12]. However, the effect of portal injection of glucose on gastric acid secretion has not yet been examined. The plasma concentrations of glucose appeared to have fallen to between 70 and 90 mg/dl 45 min after the injections of 0.5, 1.0 and 2.0 U / k g / h insulin, while the gastric acid outputs were acutely increased 45 min after the insulin injections (Fig. 1). This supports the view that the onset of the gastric acid secretion during insulin hypoglycaemia corresponds to the glucose concentration in the blood [2]. Although the insulin effects on the acid outputs were not dose-dependent, a relatively constant enhanced acid output was observed for 60-90 min after 1.0 U / k g / h insulin injections and an inhibitory response in the acid output to glucose was analysed for 60-90 min after the insulin treatment. During a period of 60-90 min after the insulin injection, denervation of the left cervical vagus inhibited the elevated outputs of gastric acid, indicating that the acid response by insulin is mediated by vagal activity (Fig. 2). Portal injections of 30 and 60 mg/kg OEDG solutions transiently lowered the gastric acid outputs (Figs. 3, 4) and the effect of the glucose was abolished by vagotomy at the hepatic level (Fig. 4). Thus. it appears that the high level of glucose kept locally around the portal areas stimulates hepatic mechanisms sensitive to glucose and the inhibitory response results from a change in central nervous activity. The 120 mg/kg OEDG injections also produced a reliable decrease in the acid outputs; however, the response was not blocked by hepatic vagotomy (Figs. 3, 4). Since the glucosensitive mechanism in the brain has been well documented to show control of the vagally mediated secretion of gastric acid, and glucose inhibits acid secretion by reducing vagal activity [2,3,8], it is likely that the glucose injected into the portal vein in the utilized dose may not have been diluted to an ineffective concentration in the circulation by the time it reaches the brain and, in this case, the mechanism in the brain may be activated. However, the possibility should also be entertained that this dose may have affected nonneural (e.g. hormonal) or possibly non-vagal neural fibres involved in gastric secretory functions. The inhibitory response by the 60 mg/kg OEDG solutions was not reproduced by the equitonic NaCI solutions, suggesting that the response is not due to osmotic effects of the glucose (Fig. 4). D-Glucose dissolved in aqueous solution has been known to exist as an optically equilibrium mixture of its two anomers: 36% a-D-glucose and 64~ fl-D-glucose [15]. Furthermore, almost the same anomeric fraction is found in blood [7,10]. Concerning the physiological function of D-glucose anomers, several works have disclosed that fl-D-glucose is more rapidly transported and metabolized by various ceils and tissues than a-D-glucose [4,5,11,14]. In this experiment, fl-O-glUCO~'d~ was most potent in reducing the gastric acid outputs (Fig. 5). Histologically, afferent sensory nerve fibers in the hepatic portal vessels have been identified as free nerve endings [21]. Thus, the observations provide a possibility that the action of D-glucose on the hepatic nerve terminals might be ascribed to the intracellular glucose metabolism.
343
a-Anomeric stereospecificity has been known in insulin release responsive to glucose; a-D-glucose is more effective in stimulating insulin secretion from the pancreatic islets than fl-D-glucose and the action of a-D-glucose has been regarded as an action of glucose independent of intracellular glucose metabolism [13]. These results suggest that hepatic glucosensitive mechanisms may affect gastric acid secretion and suggest that fl-D-glucose in the blood may play an important role in activating such mechanisms.
Acknowledgements The author is greatly indebted to Mr. K. Kunihara for his assistance, to Prof. J. Okuda (Department of Clinical Biochemistry, Faculty of Pharmacy, Meijo University, Nagoya) for providing the D-glucose used in this study, and to Prof. A. Niijima for his constant encouragement of the project. This work was supported by Research Grant 56770096 from the Ministry of Education, Science and Culture of Japan.
References 1 Armin, J. and Grant, R.T., Adrenaline release during insulin hypoglycaemia in the rabbit, J. Physiol. (Lond.), 149 (1959) 228-249. 2 Brooks, F.P., Central neural control of acid secretion. In American Physiological Society (Ed.), Handbook of Physiology, Section 6, Williams and Wilkins, Baltimore, 1967, pp. 805-826. 3 Colin-Jones, D.G. and Himsworth, R.L., The location of the chemoreceptor controlling gastric acid secretion during hypoglycaemia, J. Physiol. (Lond.), 206 (1970) 397-409. 4 Faust, R.G., Monosaccharide penetration into human red blood cells by an altered diffusion mechanism, J. Cell. Comp. Physiol., 56 (1960) 103-121. 5 Fishman, P.H. and Bailey, J.M., Mutarotases X. Anomer specific glucose transport in ascites tumor cells, Amer. J. Physiol., 226 (1974) 1007-1014. 6 Ghosh, M.N. and Schild, H.O., Continuous recording of acid gastric secretion in the rat, Brit. J. Pharmacol., 13 (1958) 54-61. 7 Hill, J.B. and Cowart, D.S., Mutarotase and glucose anomers. Biochem. Med., 1 (1967) 62-79. 8 Hirano, T. and Niijima, A., Effects of 2-deoxy-D-glucose, glucose and insulin on efferent activity in gastric vagus nerve, Experientia, 36 (1980) 1197-1198. 9 Jarrett, R.J., Rhythms in insulin and glucose. In D.T. Krieger (Ed.), Endocrine Rhythms, Raven Press, New York, 1979, pp. 247-258. l0 Miwa, I., Maeda, K. and Okuda, J., Anomeric compositions of D-glucose in tissues and blood of rat, Experientia, 34 (1978) 167-169. I I Miwa, I., Okuda, J., Niki, H. and Niki, A., Uptake of radioactive D-glucose anomers by pancreatic islets, J. Biochem., 78 (1975) 1109-1111. 12 Niijima, A., Afferent impulse discharges from glucoreceptors in the liver of the guinea pig, Ann. N.Y. Acad. Sci., 157 (1969) 690-700. 13 Niki, A. and Niki, H., Hexose anomers, insulin release, and diabetes mellitus, Biomed. Res., 1 (1980) 189-206. 14 Okuda, J., Miwa, I., Sato, M. and Murata, T., Uptake of D-glucose anomers by rat retina, Experientia, 33 (1977) 19-20. 15 Pigman, W. and Isbell, H.S., Mutarotation of sugars in solution, Advanc. Carbohyd. Chem., 23 (1968) 11-57.
344 16 Sakaguchi, T., Warashina, A. and Niijima, A., Production of elastic electrodes for nerve stimulation, Pfltigers Arch., 380 (1979) 283. 17 Sakaguchi, T. and Yamaguchi, K., Changes in efferent activities of the gastric vagus nerve by administration of glucose in the portal vein, Experientia, 35 (1979) 875-876. 18 Sakaguchi, T. and Yamaguchi, K., The effect of electrical stimulation of the hepatic branch of the vagus nerve on the secretion of gastric acid in the rat, Neurosci. Lett., 13 (1979) 25-28. 19 Sakaguchi, T. and Yamaguchi, K., Effects of vagal stimulation, vagotomy and adrenalectomy on release of insulin in the rat, J. Endocrinol., 85 (1980) 131 - 136. 20 Salomon, L.L. and Johnson, J.E., Enzymatic microdetermination of glucose in blood and urine, Analyt. Chem., 31 (1959) 453-456. 21 Tsai, T.L., A histological study of sensory nerves in the liver, Acta Neuroveg., 17 (1958) 354-385.
337
Elsevier BiomedicalPress
Alterations in gastric acid secretion following hepatic portal injections of D-glucose and its anomers Takeo Sakaguchi Department of Physiology, Niigata University School of Medicine, Niigata (Japan)
(ReceivedJuly 19th, 1981) (AcceptedOctober 30th, 1981)
Key words: blood gluc0se--gastric acid secretion--stomach--glucose anomers--
vagus nerve--liver
Abstract
Changes in acid outputs from the stomach were examined after portal injections of D-glucose and its optical anomers in the bilaterally adrenalectomized rats with insulin hypoglycaemia. Significant decrease in gastric acid outputs was noted after portal injections of a-D-glucose, optically equilibrated D-glucose (OEDG) consisting of 36% a-anomer and 64% fl-anomer and fl-D-glucose. The effect of fl-D-glucose was most potent in reducing the acid outputs and the inhibitory response was entirely prevented by prior vagotomy at the hepatic level. The injections of isotonic NaCI solution, however, produced no change in the acid outputs. Results suggest that changes in glucose levels in the portal vein may modulate gastric acid secretion through hepatic vagal afferents and gastric vagal efferents and suggest that activation of hepatic glucosensitive mechanisms may be dependent on the anomeric stereospecificity of D-glucose in the blood.
Introduction
Electrophysiological studies have revealed that D-glucose injected into the portal vein affects tonic activities of the vagus nerve innervating the stomach [17]. This, at least, suggested a neural connection existing between hepatic vagal afferents and gastric vagal efferents in the brain. Considering the observation together with the 0165-1838/82/0000-0000/$02.75 © 1982ElsevierBiomedicalPress
338
finding that electrical stimulation of the proximal cut end of the hepatic vagus nerve changes acidity in the gastric perfusing fluid [18], it has also been speculated that hepatic afferent glucose signals may relate to gastric acid secretion through the central vagal connection. However, there is no report concerning the effect of portal glucose injection on gastric acid secretion. On the other hand, D-glucosedissolved in aqueous solution has been known to exist as an optically equilibrium mixture of its two anomers; 36% a-anomer and 64% fl-anomer [15]. Moreover, almost the same anomeric composition has been shown in blood [7, t0]. The present experiments were thus designed to investigate whether D-gluc0se injected into the portal vein influences gastric acid secretion, especially in consideration of the D-glucose anomers.
Materials and Methods
One hundred and eighteen male rats weighing about 300 g were used. They were fasted for 22 h before each experiment under controlled conditions of room temperature (22°C) and light (12:12 light-dark cycle), although allowed tree access to tap water. The animals anaesthetized with pentobarbital sodium (45 mg/kg, i.p.) were adrenalectomized bilaterally: this diminished scatter in plasma levels of glucose and insulin [19] and prevented inlu'bition of gastric acid secretion by the adrenalin which would have been released in response to hypoglycaemia [1]. The experiments were further performed in the morning to eliminate any diurnal variation in levels of glucose and insulin [9]. The experimental preparation for gastric acid estimation was based on the methods described earlier [3,6]. After a tracheal intubation, a polyethylene tube was introduced into the stomach through the esophagus and was tied in position by a ligature around the esophagus in the neck. The abdominal cavity was opened by a mid-line incision. A cannula was then inserted in the pyloro-duodenal junction through an incision and passed up into the stomach. The cannula was tied into position by purse string sutures around the junction [6]. In some experiments, loose threads were set around the hepatic branch of the vagus nerve or around the left vagus nerve in the neck and both ends of each thread led through a plastic tube; the nerves could be cut easily by pulling the loop of thread. Residual food particles in the stomach were washed out with warm saline. Throughout the experiment body temperature of each animal was kept at 36 - 0.5°C by a heating lamp. Physiological saline in a bottle placed 3 cm above the level of the animal was warmed at 36°C and it was gravitationally perfused through the stomach into a collecting tube. The perfusion rate was 2.6 ml,/min and as the gastric perfusate was gathered into the tube, the volume of each sample was measured and titratable acidity (end point pH 7.0) was determined using N/100 NaOH [3]. With this system acid output was calculated every 3-15 min. D-Glucose and NaC1 dissolved in distilled water kept at 36°C were infused through a catheter placed in the portal vein. a- and/~-anomers of crystallized D-glucose were dissolved in the water immediately before use to eliminate mutarotation of the glucose. Purity of the two anomers was more than 98% a n d the rate o f mutarotation of the two anomers in the water was negligible. Optically eqe/librated
339 D-glucose solution consisting of 36% a-D-glucose and 64% B-D-glucose was obtained by incubating either a- or B-D-glucose solution for 24 h at room temperature (22°C) to induce the glucose to mutarotate completely. The amount of a portal injection was 0.2 ml and each injection lasted 15 s via an infusion pump. Insulin (regular insulin: N o v o Ind., Copenhagen) was injected through a catheter inserted in the right jugular vein by an infusion pump. The concentration of glucose in the blood was measured by the method reported earlier [20]. Blood for glucose estimation (0.1 ml) was drawn through the same catheter in the jugular vein. Data were collected from the first response of each animal to a certain drug. Statistical significance of differences between values were determined by Student's t-test: P < 0.05 was taken as indicative of a significance between means.
Results When gastric acid outputs were measured every 15 min following insulin injections, the cumulative levels of acid outputs (the mean-+ S.E.M.) 90 min after the injections of 0.5, 1.0 and 2.0 U / k g / h insulin were 48.5-+ 1.2 (n = 9), 54.5 -+ 0.7 (n = 9) and 46.3 -+ 1.1 (n = 10)/~-equiv/90 min, respectively (Fig. 1). The acid output produced by 1.0 U / k g / h insulin injections was significantly ( P < 0.02 and P < 0.01) greater than that induced by 0.5 and 2.0 U / k g / h insulin injections. The patterns of acid output by the 3 insulin injections were also different from each other and a relatively constant stimulated acid output was observed during 60-90 min only after
'oo t
ii '°i]
611 6iI 6i I 611 6i I 611
20"
gT
o~
It
15
i
t
i
6 6 iI I
Oil
U
6, i 6i i
7o Time
6'0
IO
(mln)
Fig. 1. Time course of plasma levels of glucose and gastric acid output after administrations of 0.5 (©: n=9), 1.0 (×: n=9) and 2.0 (O: n= 10) U/kg/h insulin. Horizontal bar represents duration of the insulin injections. Values are the mean+ S.E.M. A relativelyconstant enhanced acid output was noted for 60-90 min after 1.0 U/kg/h insulin injections.
340
3 2=. 200 E
100
~Ai*
6AA
Q
~i
0
200 "~
10
e~
0 15
l
i L
,
6Al
5
=~
10
"~ ._> •-~ g
5
0
.~_
io
io
Time (min)
o
0 0
60
7'5 Time (min)
Fig. 2. Denervation effect of the vagus nerve on the gastric acid outputs evoked by 1.0 U / k g / h insulin ( n = l l). Arrow indicates division of the left vagus nerve. Horizontal bar Shows duration of the insulin injection. Values are the m e a n +- S,E.M. * P<0.05, ** P<0.01: significantly different from the vatue just before vagal denervations. Fig. 3. The effect of portal injections of 30 ( ~ : n = l l ) , 60 ( A : n = 1 2 ) and 120 (&: n = 8 ) m g / k g equilibrated D-glucose on the gastric acid outputs caused by 1.0 U / k g / h insulin. Arrow indicates the time of portal injections. Horizontal bar shows duration of the insulin administration, Values are meansm S.E,M. * P <0.05, ** P <0.02, *** P <0.01 : significantly different from the value immediately before injection:
the 1.0 U / k g / h insulin injection. Acid outputs were then calculated every 5 min following the 1.0 U / k g / h insulin administration and it was observed that the enhanced acid outputs by the insulin injection were significantly inhibited by left vagotomy at the cervical level (Fig. 2). From the two preliminary findings, successive administration of 1.0 U / k g / h insulin was employed in the following experiments and portal test injections were always begun about 60 min after the insulin injection. Optically equilibrated D-glucose (OEDG) was portally injected at various doses from 30 to 120 m g / k g (Fig. 3). The levels of acid output 0 and 3 min before portal injections of the glucose were not significantly different from each other. The acid output 3 min after the injections of 30 and 60 mg,/kg OEDG was reduced significantly compared to the level before injection without changes in the blood glucose levels and the inhibitory response had disappeared 6 min after the injections. A sustained suppression of acid output was observed after the 120 m g / k g OEDG injections; however, the glucose levels 3 min after the injection increased significantly to the level 3 rain before injection. In the rats with vagotomy at the hepatic level, no change in the acid output was
341 4 : E
5 "Z E
10 ~ f
ab
~
6, 6,
5
0
*i 0
1
!
s
2
g,
>
.:~
11
!
10
0
60
*i
O
7'5 Time(min)
o
61 64 6, "6
0
0
60
7'5 Time (rain)
Fig. 4. The effect of portal injections of equilibrated D-glucoseon the gastric acid outputs evoked by 1.0 U/kg/h insulin in the hepatic vagotomized rats. Sixty (/x: n = 9) and 120 (A: n = 8) mg/kg equilibrated D-glucose were portally injected. The NaC1 solutions (11: n=7) used were equitonic to 60 mg/kg D-glucose solutions. Arrows show the hepatic vagotomy (a) and the portal injections (b). Horizontal bar represents duration of the insulin injection. Values are the mean±S.E.M. * P<0.02: significantly different from the value immediately before injection. Fig. 5. Changes in the acid outputs from the stomach in response to portal injections of 60 mg/kg a-D-glucose (~: n= 13) and 60 mg/kg fl-D-glucose ( , : n= 11) in the rats treated with 1.0 U/kg/h insulin. Horizontal bar expresses duration of the insulin injection. Arrow indicates the time of injection. Values are means±S.E.M. * P<0.05. ** P<0~01: significantly different from the value immediately before injection.
noted after the 60 m g / k g O E D G injections. Significant depression in the acid o u t p u t was elicited 3 and 6 min after the 120 m g / k g O E D G injections in spite of division of the hepatic vagus nerve. The NaC1 solutions, which are equitonic to 60 m g / k g O E D G solutions, caused no change in the acid output (Fig. 4). The level of gastric acid output 3 m i n after portal injections of 60 m g / k g a-D-glucose was decreased significantly c o m p a r e d to the level before injection, whereas the 60 m g / k g B-D-glucose injections also reduced significantly the acid output levels 3 min after the injection (Fig. 5). The injections of the B-D-glucose seemed to induce a further reduction in the acid output; the change by the /3-D-glucose injections was greater ( P < 0.01) than that by the a-D-glucose injections.
Discussion
Electrophysiological studies have suggested that hepatic glucose signals may affect gastric function through a central vagal pathway existing between the liver and the stomach [17]. In addition, afferent hepatic signals have been shown to influence gastric acid secretion, because electrical stimulation to the hepatic vagus nerve increased gastric acid secretion [18]. In the experiment, electrical stimulation
342 given to the nerve with elastic electrodes [ 16] has been considered to be equivalent to hypoglycaemia in the portal vein, especially since the inverse relationship between afferent hepatic vagus discharge and the glucose level in the portal perfusing fluid was established [12]. However, the effect of portal injection of glucose on gastric acid secretion has not yet been examined. The plasma concentrations of glucose appeared to have fallen to between 70 and 90 mg/dl 45 min after the injections of 0.5, 1.0 and 2.0 U / k g / h insulin, while the gastric acid outputs were acutely increased 45 min after the insulin injections (Fig. 1). This supports the view that the onset of the gastric acid secretion during insulin hypoglycaemia corresponds to the glucose concentration in the blood [2]. Although the insulin effects on the acid outputs were not dose-dependent, a relatively constant enhanced acid output was observed for 60-90 min after 1.0 U / k g / h insulin injections and an inhibitory response in the acid output to glucose was analysed for 60-90 min after the insulin treatment. During a period of 60-90 min after the insulin injection, denervation of the left cervical vagus inhibited the elevated outputs of gastric acid, indicating that the acid response by insulin is mediated by vagal activity (Fig. 2). Portal injections of 30 and 60 mg/kg OEDG solutions transiently lowered the gastric acid outputs (Figs. 3, 4) and the effect of the glucose was abolished by vagotomy at the hepatic level (Fig. 4). Thus. it appears that the high level of glucose kept locally around the portal areas stimulates hepatic mechanisms sensitive to glucose and the inhibitory response results from a change in central nervous activity. The 120 mg/kg OEDG injections also produced a reliable decrease in the acid outputs; however, the response was not blocked by hepatic vagotomy (Figs. 3, 4). Since the glucosensitive mechanism in the brain has been well documented to show control of the vagally mediated secretion of gastric acid, and glucose inhibits acid secretion by reducing vagal activity [2,3,8], it is likely that the glucose injected into the portal vein in the utilized dose may not have been diluted to an ineffective concentration in the circulation by the time it reaches the brain and, in this case, the mechanism in the brain may be activated. However, the possibility should also be entertained that this dose may have affected nonneural (e.g. hormonal) or possibly non-vagal neural fibres involved in gastric secretory functions. The inhibitory response by the 60 mg/kg OEDG solutions was not reproduced by the equitonic NaCI solutions, suggesting that the response is not due to osmotic effects of the glucose (Fig. 4). D-Glucose dissolved in aqueous solution has been known to exist as an optically equilibrium mixture of its two anomers: 36% a-D-glucose and 64~ fl-D-glucose [15]. Furthermore, almost the same anomeric fraction is found in blood [7,10]. Concerning the physiological function of D-glucose anomers, several works have disclosed that fl-D-glucose is more rapidly transported and metabolized by various ceils and tissues than a-D-glucose [4,5,11,14]. In this experiment, fl-O-glUCO~'d~ was most potent in reducing the gastric acid outputs (Fig. 5). Histologically, afferent sensory nerve fibers in the hepatic portal vessels have been identified as free nerve endings [21]. Thus, the observations provide a possibility that the action of D-glucose on the hepatic nerve terminals might be ascribed to the intracellular glucose metabolism.
343
a-Anomeric stereospecificity has been known in insulin release responsive to glucose; a-D-glucose is more effective in stimulating insulin secretion from the pancreatic islets than fl-D-glucose and the action of a-D-glucose has been regarded as an action of glucose independent of intracellular glucose metabolism [13]. These results suggest that hepatic glucosensitive mechanisms may affect gastric acid secretion and suggest that fl-D-glucose in the blood may play an important role in activating such mechanisms.
Acknowledgements The author is greatly indebted to Mr. K. Kunihara for his assistance, to Prof. J. Okuda (Department of Clinical Biochemistry, Faculty of Pharmacy, Meijo University, Nagoya) for providing the D-glucose used in this study, and to Prof. A. Niijima for his constant encouragement of the project. This work was supported by Research Grant 56770096 from the Ministry of Education, Science and Culture of Japan.
References 1 Armin, J. and Grant, R.T., Adrenaline release during insulin hypoglycaemia in the rabbit, J. Physiol. (Lond.), 149 (1959) 228-249. 2 Brooks, F.P., Central neural control of acid secretion. In American Physiological Society (Ed.), Handbook of Physiology, Section 6, Williams and Wilkins, Baltimore, 1967, pp. 805-826. 3 Colin-Jones, D.G. and Himsworth, R.L., The location of the chemoreceptor controlling gastric acid secretion during hypoglycaemia, J. Physiol. (Lond.), 206 (1970) 397-409. 4 Faust, R.G., Monosaccharide penetration into human red blood cells by an altered diffusion mechanism, J. Cell. Comp. Physiol., 56 (1960) 103-121. 5 Fishman, P.H. and Bailey, J.M., Mutarotases X. Anomer specific glucose transport in ascites tumor cells, Amer. J. Physiol., 226 (1974) 1007-1014. 6 Ghosh, M.N. and Schild, H.O., Continuous recording of acid gastric secretion in the rat, Brit. J. Pharmacol., 13 (1958) 54-61. 7 Hill, J.B. and Cowart, D.S., Mutarotase and glucose anomers. Biochem. Med., 1 (1967) 62-79. 8 Hirano, T. and Niijima, A., Effects of 2-deoxy-D-glucose, glucose and insulin on efferent activity in gastric vagus nerve, Experientia, 36 (1980) 1197-1198. 9 Jarrett, R.J., Rhythms in insulin and glucose. In D.T. Krieger (Ed.), Endocrine Rhythms, Raven Press, New York, 1979, pp. 247-258. l0 Miwa, I., Maeda, K. and Okuda, J., Anomeric compositions of D-glucose in tissues and blood of rat, Experientia, 34 (1978) 167-169. I I Miwa, I., Okuda, J., Niki, H. and Niki, A., Uptake of radioactive D-glucose anomers by pancreatic islets, J. Biochem., 78 (1975) 1109-1111. 12 Niijima, A., Afferent impulse discharges from glucoreceptors in the liver of the guinea pig, Ann. N.Y. Acad. Sci., 157 (1969) 690-700. 13 Niki, A. and Niki, H., Hexose anomers, insulin release, and diabetes mellitus, Biomed. Res., 1 (1980) 189-206. 14 Okuda, J., Miwa, I., Sato, M. and Murata, T., Uptake of D-glucose anomers by rat retina, Experientia, 33 (1977) 19-20. 15 Pigman, W. and Isbell, H.S., Mutarotation of sugars in solution, Advanc. Carbohyd. Chem., 23 (1968) 11-57.
344 16 Sakaguchi, T., Warashina, A. and Niijima, A., Production of elastic electrodes for nerve stimulation, Pfltigers Arch., 380 (1979) 283. 17 Sakaguchi, T. and Yamaguchi, K., Changes in efferent activities of the gastric vagus nerve by administration of glucose in the portal vein, Experientia, 35 (1979) 875-876. 18 Sakaguchi, T. and Yamaguchi, K., The effect of electrical stimulation of the hepatic branch of the vagus nerve on the secretion of gastric acid in the rat, Neurosci. Lett., 13 (1979) 25-28. 19 Sakaguchi, T. and Yamaguchi, K., Effects of vagal stimulation, vagotomy and adrenalectomy on release of insulin in the rat, J. Endocrinol., 85 (1980) 131 - 136. 20 Salomon, L.L. and Johnson, J.E., Enzymatic microdetermination of glucose in blood and urine, Analyt. Chem., 31 (1959) 453-456. 21 Tsai, T.L., A histological study of sensory nerves in the liver, Acta Neuroveg., 17 (1958) 354-385.