Acta Physiol Scand 1992, 146, 349-356
Vagally mediated release of gastrin and cholecystokinin following sensory stimulation K. U V N A S - M O B E R G " , T. LUNDEBERGT, G. B R U Z E L I U S " and P. ALSTER" Department of Pharmacology and Department of Physiology, Karolinska Institutet, Stockholm, Sweden
"
t
UVNAS-MOBERG, K., LUNDEBERF, T., BRUZELIUS, G. & ALSTER, P. 1992. Vagally mediated release of gastrin and cholecystokinin following sensory stimulation. Acta Physiol Scand 146, 349-356. Received 20 February 1992, accepted 22 May 1992. ISSN 0001-6772. Department of Pharmacology and Department of Physiology, Karolinska Institute, Stockholm, Sweden. The aim of the present study was to investigate whether gastrin, cholecystokinin (CCK) and somatostatin secretion can be influenced by sensory stimulation and if so, whether such effects are mediated via the vagal nerves. Male rats anaesthetized with chloral hydrate were exposed to three different stimuli, i.e. to low frequency (2 Hz) electrical stimulation of muscles via needles (electro-acupuncture), to thermal stimulation at 40 "C or to vibration at 100 Hz. The two former stimuli activate mainly small and medium sized myelinated fibres from muscles and skin respectively, whereas vibration activates large myelinated fibres from skin, subcutaneous tissue and muscles. Experiments were also performed on animals that were vagotomized or exposed to prior treatment with atropine (0.5 mg kg-I). Blood was collected at various time intervals and plasma levels of gastrin, CCK and somatostatin were measured with radioimmunoassay (RIA). All three stimuli, i.e. electro-acupuncture, vibration and thermal stimulation caused significant elevations of gastrin (103 f 11-151 f 16 pM, 105*8-140& 12 PM and 105&14162&4 PM) and cholecystokinin (9k0.8-15k2.8 PM, 8+0.5-10*1.5 PM and 8.0 k0.5-10.5 f1.5). Somatostatin was raised in response to electro-acupuncture (10f 1-14? 3 p ~ ) Vagotomy . and atropinization abolished the release of gastrin and CCK in response to all three stimuli. CCK levels were significantly reduced following electro-acupuncture in atropinized rats. In conclusion, gastrin and cholecystokinin release is stimulated by activation of sensory afferent, originating in skin, subcutaneous tissue as well as in muscle. Some of these effects are vagally mediated.
Key words : A-,8 and A-S afferent, CCK, electro-acupuncture, gastrin, non-noxious stimulation, somatosensory afferents, somatostatin, thermal stimulation, vibration.
T h e endocrine system of the gut is activated in response to feeding. Part of this effect is mediated by vagal efferents. Thus, in dogs sham feeding causes a rise of gastrin and CCK (Nilsson 1975, Schafmayer et al. 1988). Electrical vagal stimulation in rats also causes a release of gastrin and CCK and a decrease or increase in somatostatin levels (LindCn et al. 1989, Aliiio Correspondence : Kerstin Uvnas-Moberg, Department of PharmacoIogy, Karolinska Institutet, Box 60400, S-104 01 Stockholm, Sweden.
et al. 1983). Recently, we have been able to show that gastro-intestinal hormones such as gastrin and CCK are released in response to suckling in lactating animals of several species including rats (Linden et al. 1987, 1990). I n rats the maternal CCK release triggered by suckling of the pups is vagally mediated via a central reflex possibly involving oxytocinergic transmission (LindCn et al. 1990). Electrical stimulation of the sciatic nerve induced a release of gastro-intestinal hormones in anaesthetized cats, indicating that activation of somatosensory afferent, not only
349
from the teats b u t also from other parts of the skin, ma!- influence the release of gastro-intestinal hormones (Uvnas-.\Ioberg rt al. 1986). Sin( thin (;-fibre afferents are not likelj- t o be activated by such stimulation, the effect is possibly mediated via thicker fibres, normally actii-atcd by 1-arious kinds of non-noxious stimuli ( U m a s %loberg t t aI. 1986). I t has also been shown that ' different modes ofperipheral stimulation activate spinal cord mechanisms that affect pain and peripheral blood flow (Lundeberg rt ul. 1988a & b, Kjartansson et al. 1988). T h e present stud!- in\-estigates in more detail hon sensory stimulation influences the release of gastrin, CXK and somatostatin. 3lale rats a-ere exposed to three types o f treatment, known to set up acthit!- in different types of somatosensory n e r w fibres, i.e. vibration at 100 Hz which activates mainly thick mvelinated fibres from the skin, subcutaneous tissue and muscles. Mild thermal stimulation a t 40 "C which activates medium sized '4-8 and C-afferents, originating in the skin and electro-acupuncture ( 2 H z ) which activates thin and medium sized myelinated muscular afferents ( C h u n g el 01. 1984a & b, L u r t ul. 1986). I n order t o explore the involvement o f t h e vagal nerves and of cholinergic mechanisms in hormonal effects caused by such stimulations, experiments were also performed on vagotomized and atropinized animals.
\I E T H0 D S 1;uperiments were performed on male Sprague Danley rats (300 g) all~wedfree access to food and water. The animals \\ere anaesthetized with chloral h!-drate (.NO mg kg ', given i.p,), Experiments were routincl!performed between 9 and 11 a.m. -4fter the cxperiment, the animals were killed b!- decapitation and blood (.? ml) was collected. (The study mas approved by the local Ethics Committee.) SurKizal procedures. Vagotomy was carried out under anaesthesia. The major branches of the subdiaphragmatic vagus nerve were identified on or near the oesophagus and the anterior and posterior subdiaphragmatic vagi were dissecred free and cur. Sham operations w r e performed according to the same procedure, but the Yagi were not cut. .Stimu/utionr. 1-ibration stimulation (100 Hz) was carried out for a period of 30 min. An electromechanical vibrator (Somedic AB, Stockholm, Sueden) with a probe area of 9 cm' was applied with the constant pressure (3 kPa) to the thorax and abdomen ( T h , I , , ) ofthe rat. The vibratory movement (peak to
peak 1000 pm) was continuously measured throughout the stimulation period (Lundeberg et ul. 1984). Electro-acupuncture was performed for a period of 30 min. Points chosen were U B l l (close to the shoulder joint) and UR54 (close to the hip joint) bilaterall!. Stimulation was brought about by manual rotations of the needles (0.3 mm) after insertion to depths of0.5-0.7 cm, at the points chosen. Thereafter, the rats receiwd electrical stimulation through all the needles (CBI 1-UBll, UB54UB54) that were connected to an acupuncture pulse stimulator (B.V. Enraf-Nonius Delft, Delft, the Netherlands) producing bipolar square wave pulses of 0.2 ms duration, and 2 Hz frequency. The current intensity was adjusted so that localized muscle contractions were seen (2-5 1.) (Lundeberg et al. 1988~). Thermal stimulation was applied for 30 min to the skin of the thorax and abdomen (Th,-L,) of the rat using a feedback controlled contact thermode. T h e thermode had a contact surface of 9 cm', was fixed to a glass chamber with a constant temperature of40 "C and applied with a constant pressure of 3 kPa (Lundeberg & Ottoson 1982). Experimental design. The stimulation was started 10 min after the induction of anaesthesia and was applied for 30 min. The effect of the various stimulations was tested in two different experimental series. In series I blood samples were collected immediately - 0 min (n = 6) or 60 min ( n = 6) after the stimulation n-as finished. Control samples ( n = 6) nere collected from anaesthetized non-stimulated animals a t time point 0 min. I n the group o f animals in which the effect of temperature (40 "C) was tested, additional controls (n = 6) were performed in which the animals were exposed to the same thermode at body temperature (37 "C) for 30 min. I n the electroacupuncture group and controls, samples were also collected 330 min ( n = 6) after the end of the stimulation. T o test the effect of anaesthesia alone samples were collected 40 and 100 rnin after induction of anaesthesia ~
(ti
= 5).
In series 11. stimuli were applied to control animals as nell as atropinized, vagotomized or sham-operated animals. Atropine (0.5 mg kg-' s.c.) was given immediatel!. after induction of anaesthesia (n = 6). Vagotomy or sham operation was performed 1 week before the experiments. Samples were collected 60 rnin (n = 6) after the end of stimulation in the different experimental groups and controls. Separate controls were performed for each type of treatment. Trerittnent of blood samples. Blood samples were collected in tubes containing heparin (10 IU ml-') and Trasylol (500 I L ml-I). They were centrifuged, the plasma was removed and frozen at -20 "C. RZAs. Gastrin levels were measured in unextracted plasma and by RIA (Nilsson 1975) using antiserum No. 2604, N-hich recognizes gastrin 17 and 34 with
Somiltosensory control of GI hormones
35 1
equimolar potency. The detection limit of the assay unchanged in all groups (Table 1). Kesults w a s 2 pmol I-’ and the intra- and inter-assay co- obtained froni sham operation experiments did ctiicients of variation were 10 and 13% respectively. not diffcr from those obtained in unoperated CCK le\-els in plasma were measured after SEP- animals and \$ere therefore included in the PAK,, extraction using acetonitrile and acetic acid. control group. The RIA method used has been described by Flimcno rt 01. (1983). The antiserum OAL 656 (Otsuka Pharmaceutical Co. Ltd., ‘I’okushima, Japan) raised against the N-terminal amino acid residue of sulphated CCK-8 detects CCK-8, CCK-33 and CCK-39 but not gastrin. The intra- and inter-assay variation was 10 and 12% respectively. Somatostatin in plasma was measured by RIA after f c SEP-PAK,, extraction of acidified samples using ,r *e methanol. The antibodq Rl4lE was used with the U m method described by Efendic et al. (1980). The 17 detection limit of the assay was 2 pmol I-’ and the intra- and inter-assay coefficients of variation were 9 and 11 O 0 respectively. Stutisticnl ecaluativns. ‘I‘hc hormone levels in the figures are presented as mcan valueskSD. Comparisons between groups were evaluated with the Kruskal-Wallis one-way analysis of variance (ANOVA) and subsequent two-group comparisons were made with the Maim Whitney U-test. Twotailed tests were used and P-values < 0.05 were considered significant.
RESULTS Effect of’ anaesthesia Gastrin, CCK and somatostatin levels were not significantly different in samples collected 40 and 100 min after induction of anaesthesia i.e. corresponding to 0 and 60 min after stimulation. I n the five experiments performed, gastrin levels were I 1 5 8 PM and 116 i10 PM, CCK 1 0 i - l PM and 12k4 PM and somatostatin 12f 4 PM and 1 4 i 3 p w
Effect of vibration Gastrin and CCK levels were significantly (P< 0.05) elevated when comparcd to control values 60 min after the stimulation, whereas somatostatin levels were unchanged (Fig. 1).
f
EfSect of atropine mid z!agotowiy
A significant rise (P < 0.05) of gastrin and CXK levels was seen in response to vibration also in this series of experiments. This effect was abolished by vagotomy or following administration of atropine. Somatostatin levels were
Fig. 1. Gastrin (a), CCK (b) and somatostatin (c) levels in control rats as well as 0 and 60 min after exposure to vibration at 100 Hz for 30 min.
K. C'vnas-Moberg et al.
352
fc
.c
L
U
m
rn (3
control
''
0
60
control
0'
330'
**
-I-
l(b\
60
I
'2 E
5, Y
u u
f Y
6
10
u u
4
2 0
c control
0
60
r,c .c
.v rg
fe
c 0
U
I
1s
IC
J
I
10
c 0
tn
I
0
r
0 60 Time (min) nost stimulation
5
conipol
Fig. 2. Gastrin (a), CCK (b) and somatostatin (c) levels in control rats as well as 0 and 60 min after exposure to thermal stimulation at 40 "C.
0 control
0
60'
330'
Time (min) post stimulation
Fig. 3. Gastrin (a), CCK (b) and somatostatin (c) levels in control rats as well as 0, 60 and 330 min after exposure to electro-acupuncture at 2 Hz.
Efect of rkermaj strrnulation
A clearly significant ( P < 0.001) increase of gastrin levels was seen in response to thermal stimulation at 40 O C . CCK levels nere also raised (P < 0.05) but somatostatin levels were not altered (Fig. 2 ) . No effect was induced by the thermode at body temperature (data not shown).
Effect of atropine and vagotomy Gastrin levels were elevated ( P < 0.05) in response to thermal stimulation - an effect which was abolished by atropine or vagotomy (Table 2 ) . Results obtained in sham-operation experiments did not differ from those obtained in
Somatosensory control of GI hormones
353
T a b l e 1. Effect of vibration at 100 Hz on gastrin, CCK and somatostatin levels collected 60 rnin after end of stimulation in control animals as well as in animals treated with atropine, 0.5 mg kg-' and vagotomy
Control 60 min after stimulation Control - atropine 60 min after stimulation Control - vagotomy 60 min after stimulation
n
Gastrin (pmol I-')
n
11 12 5 5 5 6
104k18 132 f25" 126+21 119 30 190f19 186i23
12 8 + 2 12 11 3" 6 10+3 6 9 2 6 7f2 6 7 f2
+
CCK (pmol I-')
+ +
n
Somatostatin (pmol It')
12 12 6 6 6 6
11+2 9 3 llk3 10+1 11+3 10k1
+-
T a b l e 2. Effect of thermal stimulation at 40 "C on gastrin, CCK and somatostatin levels collected 60 min after end of stimulation in control animals as well as in animals treated with atropine, 0.5 mg kg-' and vagotomy
Control 60 min after stimulation Control - atropine 60 min after stimulation Control - vagotomy 60 min after stimulation
n
Gastrin (pmol I-')
n
CCK (pmoll-')
n
Somatostatin (pmol I-')
10 11 4 6 6 6
112k22 132+21" 117.t.29 120f31 179+34 163f31
12 12 6 6 6 6
9+2 1051 llf2 13 f3 11&4 10k3
12 8 + 2 12 10f 3 5 12$3 6 6k2 6 7+3 6 7+3
T a b l e 3. Effect of electro-acupuncture at 2 Hz on gastrin, CCK and somatostatin levels collected 60 rnin after end of stimulation in control animals as weil as in animals treated with atropine, 0.5 mg kg-' and vagotomy
Control 60min after stimulation Control - atropine 60minafterstimulation Control - vagotomy 60 min after stimulation
n
Gastrin (pmol I-')
n
CCK (pmoll-')
12 11 8 6 5 5
115+23 152+_26"" 120f25 121f18 186$.32 179+ 19
12 9 + 2 12 13+2" 6 15+2 6 10L-2" 6 8fl 6 9f2
n
Somatostatin (pmol 1-')
12 9 + 3 12 l l k l 8 11+2 6 10k3 6 9f2 5 7&1
Table 4. Basal levels of gastrin, CCK and somatostatin levels in controls, vagotomized and atropine-treated (0.5 mg kg-') animals Gastrin Basal levels n = 3 3 n = 17 Atropine n = 16 Vagotomy
CCK
112+30 n = 3 6 120+25 n = 18 185+34""n= 18
Somatostatin
9+2 13k2"" 9f2
n=36 n = 19 n = 18
9f2 11$2 9+2
fasted contributed to the high gastrin and CCK levels. T h e animals had to be anaesthetized in order to receive the standardized simulations. Furthermore, we chose to work with freely fed L#i,i.t ?/'ii cupu 11 ctiirt animals in which intragastric p H is high due to CCX levels u r r e significantly elevated 60 (t' < the buffering effect of food in the stomach to be 0.01) and 330 ( P < 0.05) min after stimulation able to demonstrate increases of gastrin levels. u as ended. Somatostatin levels were significantly At low antral pH, gastrin release is inhibited and eleiated at 0 and 330 min ( P < 0.05). Gastrin previous attempts to evoke a vagally induced Ic~-el.sHere significantly increased (P < 0.05) a t release of gastrin in fasted rats had proved 60 min (Fig. 3). T h e results of control experi- unsuccessful (Linden et nl. 1990). Gastrin and cholecystokinin secretion is in ments performed at 0 and 330 min did not differ part under control of the autonomic nervous and \%eretherefore combined. s!-stem. Vagal activation induced electrically or b!- sham feeding raises the plasma levels of these &'t/i~L.tu/'trr,.oplnr a t d rrrgotot?!)' hormones in dogs, cats and rats (Nilsson 1975, .\lift0 rf ul. 1983, Schafmayer er nl. 1988, Linden Gastrin and C C K levels were again influenced et tri. 1989). I n contrast, activation of the hj- acupuncture ( P < 0.01). This effect was sympatho-adrenal system inhibits vagally-inabolished by \-agotomy and by administration of duced hormone responses. T h u s gastrin release atropine. I n the case of CCK, a significant caused by electrical vagal stimulation in cats is decrease \\-as seen after atropinization. SomAtoabolished during concomitant stimulation of the statin h e l s \\ere not influenced (Table 3). splanchnic nerve (Uvnas-Wallensten & Jarhult Results obtained in sham-operation esperiments 1982). Furthermore, in rats, administration of did not difkr from those obtained in unoperated 3,-adrenergic agonists such as clonidine inhibits anin\als and were therefore included in the the \-agall>--induced release of somatostatin into control group. the gastric lumen of anaesthetized rats (UvnasMoberg & Aliiio 1991). Activation of the unm!-elinated C-fibre afferents is related to pain qf'trtroprne atid ragotomy on biisrri Irrtis transmission and activation of the sympathooJ:listuti, C C K &nu' soin(rtmtatrn adrenal system and consequently to an inhibition U'hen data from the three separate experimental of vagally induced effects on the release of GI groups ere analvsed together, atropine (0.j mg hormones. hg '1 caused a significant ( P < 0.01) increase in I n a previous study performed on anaesCCK and vagotom!- produced an increase of thetized cats, the levels of gastrin and some other gastrin le\ els in comparison to untreated animals gastro-intestinal hormones were raised in reSomatostatin l e d s were not influenced sponse to afferent electrical stimulation of the (1';ihle 4). sciatic nerve at low stimulatory intensity (Uvnas\loberg et a(. 1986). At such stimulation, thick and medium sized afferents, but not C-fibres, are likely to be activated. I n order to further explore the eEect of nonT h e main findings of the present study vere that vibration a t 100 H z and thermal stimulation at nosious somatosensory stimulation on the release 40 "C as well as electro-acupuncture ( 2 Hz) of gastro-intestinal hormones, three different significantly increased gastrin and C C K levels, kinds of afferent stimulations known to activate efects ahich were abolished by \-agotom!- or specific populations of sensor!- nerve fibres (i.e. atropine pretreatment. T h e results show that 1-ibration at 100 Hz, thermal stimulation at 40 "C difterent modes of sensor!- stimulation may and deep muscle stimulation a t 2 Hz) were induce a vagally mediated cholinergic release of applied. More specifically vibratory stimulation some gastro-intestinal hormones. acti\-ates all types of low threshold mechanoRasaf rastrin and C C K levels were high under receptors in the stimulated area and the activity thc present experimental conditions. Both the is transmitted in A-/? fibres (Merzenich & anaesthesia and the fact that animals were not Harrington 1969). Thermal stimulation is com-
unoperdted animals and were therefore included in the control group.
cf/t,~~
Somatosensory control of GI hormones plex and the types of fibres that are activated depend on the temperature. Above 43 "C heat receptors are activated and the sensation of heat is conducted in afferent Cfibres. I n the present studies in which a thermode of 40 "C was used, A-6 and C-fibres are likely to have been activated (Kenshalo 1970). Finally, electro-acupuncture (i.e. low frequency electrical stimulation of muscles) activates several types of mechanoreceptors within the muscle. Among these are the so-called ergoreceptors. T h e activity set u p in these receptors is transmitted in A-6 fibres (Kniffki et al. 1981). T h e present data show that stimulation of sensory afferents from skin and muscle caused a release of gastrin and C C K (and sometimes of somatostatin). Whether the sensory stimuli used really are equipotent as to their hormonestimulating potential is difficult to say. I t seemed as if thermal stimulation was particularly efficient in stimulating gastrin release and electro-acupuncture in CCK release. However, the intensities of the stimulations are difficult to compare and also basal levels which are of importance for the outcome of the stimulations varied somewhat between the experimental groups and series. T h e finding that vagotomy abolished the effects on gastrin and C C K release caused by vibration, thermal stimulation and acupuncture indicates that the effects were vagally mediated. I t is also possible that it is difficult to further enhance the high basal values of gastrin seen after vagotomy. Pretreatment with atropine also abolished the responses, indicating that the vagal effect of gastrin and C C K secretion is mediated by cholinergic mechanisms. However, administration of atropine alone also caused a rise of basal CCK levels, indicating the presence of a cholinergic inhibitory tone on C C K release. Indeed, an enhanced secretion of C C K following feeding in rat has been demonstrated to occur following treatment of atropine in rats. T h i s increase was suggested to be a consequence of a reduced secretion of pancreatic juice and hence to lower levels of inhibitors of C C K secretion in the duodenum (Nakano et al.
1990). Electro-acupuncture caused a significant decrease of C C K levels when they were previously raised by treatment with atropine. T h i s finding indicates that electro-acupuncture may not only increase C C K levels but also in case of elevated
355
basal C C K levels lead to an inhibition of CCK release. By which mechanism this effect is mediated is not known. Acupuncture stimulation has previously been shown to reduce sympathetic nerve activity and in addition Sato et al. have shown that nonnoxious stimulation of the skin leads to a reflex inhibition of both sympathetic nerve activity and catecholamine secretion (Sato 1987). T h e present data show that not only is sympathetic transmission inhibited by these kinds of somatosensory stimulation, in addition some parasympathetic vagally mediated effects are induced (Sato & Schmidt 1987). I n lactating rats the maternal levels of gastrin, and particularly of CCK, rise in response to suckling of the pups. This effect is abolished by vagotomy. Since lesions of the afferent nervous pathways within the CNS, leading to oxytocin secretion inhibit suckling related C C K release and since oxytocinergic neurons from the paraventricular nucleus innervate the dorsal vagal motor nucleus, the suckling related release may involve a central oxytocinergic pathway (Linden et al. 1990). Whether central mechanisms are also involved in the gastrin and CCK release caused by vibration, thermal stimulation at 40 "C and electro-acupuncture remains to be established. We thank the Swedish Medical- Research Council (project No. B9214X-05207-15C, Ake Wiberg's Foundation, RMR (49-9 l), Magn. Bergvall's Foundation, Nanna Svartz Foundation, King Gustaf V Anniversary Found, SCAA, Anna-Greta Crafoord's Foundation. REFERENCES A L I ~ ~ oS.F., , GARCIA, D. & UVNAS-MOBERG, K. 1983. On the interaction between intragastric pH and electrical vagal stimulation in causing gastric acid secretion and intraluminal release of gastrin and somatostatin in anesthetized rats. Acta Physiol Scand 117, 491495. CHUNG,J.M., FANG,Z.R., HORI,Y., LEE, K.H. & WILLIS, W.D. 1984a. Prolonged inhibition of primate spinothalamic tract cells by peripheral nerve stimulation. Pain 19, 259-275. CHUNG,J.M., LEE, K.H., HORI,Y., ENDO,K. & WILLIS, W.D. 1984b. Factors influencing peripheral nerve stimulation produced inhibition of primate spinothalamic tract cells. Pain 19, 277-293. F., NYLEN,A. & UVNASEFENDIC,S., ENZMANN, WALLENSTEN, K. 1980. Sulphonylurea (Glibenclamide) enhances somatostatin and inhibits gluca-
356
K. Urnas-Moberg et al.
imentally induced pain. Ear Nose Throat 3 67, gon release induced by arginine. .4cta Ph.ysiol Scand 565-57.5. 108, 231--233. T., NORDEMAR, R. & OTTOSON, D. 1984. H I hl E U O , S., TARCI, S . , KANAYAMA, s., KCROSHIIIA, LCNDEBERG, Pain alleviation b j vibratory stimulation. Pain 20, *r.,SHINOMURA,Y., HAYASHI,c., TATEISHI,K . , 2544. IMAGAWA, K., HASHIMURA, E. & HAMOUKA, T. LCSDEBERG, T., ABRAHAMSSON, P., BONDESSON, L. & 1983. Plasma cholecystokinin responses after inHAKER, E. 1988a. Effect of vibratory stimulation on gestion of liquiol meal and intraduodenal infusion of experimental and clinical pain. Scand 3 Rehab M e d fat, amino acids, or hydrochloric acid in man: 20. 149-159. Analysis with region specific radioimmunoassay. LL-NDEBERG, T., HURTIG,T., LUNDEBERG, S. & Ani 3 Gastroenterol 78 (1 l), 703-707. THOMAS,M. 1988c. Long-term results of acuKEKSHALO,D.R. 1970. Psychophysical studies of puncture. Pain Clin 2, 15-31. temperature sensibility. I n : \V.D. Neff (ed.) Con,\IERZENICH, ht..u. & HARRINGTON, T. 1969. The tributions to Sensory Ph,yslohgjJ, vol. 4, pp. 17-74, sense of flutter-vibration evoked by stimulation of Academic Press, New York. the hairy skin of primates: Comparison of human KJART~SSSON, J., LLSDEBERG, T . , SAXELSOT, U E., sensory capacity with the response of mechanoDALSGAARD, C-J. & HEDEN, P. 1988. Calcitonin receptive afferents innervating the hairy skin of gene-related peptide (CGRP) and transcutaneous monkeys. Exp Brain Res 9, 236-260. electrical nerve stimulation (TENS) increase cuN.AK.4S0, I., TAWIL, T., SPANNAGEL, A.W., LIDDLE, taneous blood f l o ~in - a musculocutaneous flap in R.A. & GREEK,G.M. 1990. i\tropine enhances the rat. .4cta Ph,ysiol Scand 134, 89-94. MENSE,S . & SCHMIDT,R.F. 1981. K ~ I F F KK.D., I, food-stimulated CCK secretion in the rat. Pancreas 5, 621-625. Muscle receptors with fine afferent fibres which G. 1975. Increased plasma gastrin levels in may evoke circulatory reflexes. Circ Res 48 (suppl SII.SSOS, connection with inhibition of gastric acid in I), 125-131. response to sham feeding following bulbar perfusion L I X D ~ S.%,, , CARLQLIST, Lu., H.ASSEN,s. & with acid in dogs. ScandJGastroenterol 10,273-277. ~ I O B E RK G., 1989. Plasma concentrations of choleSATO, T. 1987. Neural mechanisms of somatic sensory cystokinin, CCK-8 and CCK-33,39 in rats, determined by a method based on e n q m e digestion of regulation of catecholamine secretion from the gastrin before HPLC and RI.4 detection of CCK. adrenal gland. Adz. Biophys 23, 39-80. Glii 30 (2), 213-222. SATO, 3.& SCHMIDT,R.F. 1987. T h e modulation of I,INDES, .I,, ERIKSSON, &I., CARLQCIST, 51. & \-isceral functions by somatic afferent activity. 3ap 3 MOBERG, K . 1987. Plasma levels of gastrin, somatoPhjisiol 37, 1-17. statin and cholecystokinin immunoreactib-it! during A, YUSTEDE,A., POMPINO,A. & SCHAFVAYER, KOHLER, H . 1088. Vagal influence of cholepregnancy and lactation in dogs. Gastroenterdog), 92, 578-584. cystokinin and neurotensin release in conscious I,IuDE\, .I., ERIKSSOS,>IH.~NSES, ., s., Lvxisdogs. Scand 3 Gastroenterol 23, 3 15-320. ~ I O B E RK. G 1990. Suckling-induced release of , 1991. Vagal and -MOBERG,K . & A L I ~ OS.F. HCI-induced release of somatostatin and gastrin cholecystokinin into plasma in the lactating rat : into the gastric lumen of rats. In : R. Hdkanson & F. effects of abdominal vagotomy and lesions of central patha-aJ-s concerned with milk-ejection. 3 EnduSundler (eds.) The Stomach as an Endocrine Organ, m n o l 127, 257-263. 51-163. Elsevier Science Publisher. I,[, G., Z m o , G., LIASG, R.Z., ~VASG,I-.,SISG, ii-, ~ I O B E R K., G , POSLONCEC, B. & AHLBERG,I,. & S H I , G. 1986. .%nalysis of afferent fibres . Influence on plasma levels of somatostatin, transmitting needling sensations of acupoint. In : X. gastrin, glucagon, insulin and VIP-like immunoZhang (ed.) Research on ..lrupunr.ture, .tlosibuJtion reactkit! in peripheral venous blood of anaesmil Acupuncture ,4naesthesia, pp. 16tr177. Science thetized cats induced by low intensity afferent Press, Beijing. stimulation of the sciatic nerve. Acta Physiol Scand I,CSDEBERG, T. & OTTOSOU, D. 1982. The effect of 225-230. peripheral stimulation and placebo in chronic -W-ALLENSTEN, K. & JARHLLT, J. 1982. Reflex neuralgic pain states. Acra Ph,ysiol Scand, suppl. activation of the sympatho-adrenal system inhibits S38. the gastrin release caused by electrical vagal LCVDEBERG, T., I,ALItEI.L, G. & THOMAS,31. 1988 b. stimulation in cats. Acta Ph.ysiol Scand 114, Effect of acupuncture on sinus pain and exper297-302.
Vagally mediated release of gastrin and cholecystokinin following sensory stimulation K. U V N A S - M O B E R G " , T. LUNDEBERGT, G. B R U Z E L I U S " and P. ALSTER" Department of Pharmacology and Department of Physiology, Karolinska Institutet, Stockholm, Sweden
"
t
UVNAS-MOBERG, K., LUNDEBERF, T., BRUZELIUS, G. & ALSTER, P. 1992. Vagally mediated release of gastrin and cholecystokinin following sensory stimulation. Acta Physiol Scand 146, 349-356. Received 20 February 1992, accepted 22 May 1992. ISSN 0001-6772. Department of Pharmacology and Department of Physiology, Karolinska Institute, Stockholm, Sweden. The aim of the present study was to investigate whether gastrin, cholecystokinin (CCK) and somatostatin secretion can be influenced by sensory stimulation and if so, whether such effects are mediated via the vagal nerves. Male rats anaesthetized with chloral hydrate were exposed to three different stimuli, i.e. to low frequency (2 Hz) electrical stimulation of muscles via needles (electro-acupuncture), to thermal stimulation at 40 "C or to vibration at 100 Hz. The two former stimuli activate mainly small and medium sized myelinated fibres from muscles and skin respectively, whereas vibration activates large myelinated fibres from skin, subcutaneous tissue and muscles. Experiments were also performed on animals that were vagotomized or exposed to prior treatment with atropine (0.5 mg kg-I). Blood was collected at various time intervals and plasma levels of gastrin, CCK and somatostatin were measured with radioimmunoassay (RIA). All three stimuli, i.e. electro-acupuncture, vibration and thermal stimulation caused significant elevations of gastrin (103 f 11-151 f 16 pM, 105*8-140& 12 PM and 105&14162&4 PM) and cholecystokinin (9k0.8-15k2.8 PM, 8+0.5-10*1.5 PM and 8.0 k0.5-10.5 f1.5). Somatostatin was raised in response to electro-acupuncture (10f 1-14? 3 p ~ ) Vagotomy . and atropinization abolished the release of gastrin and CCK in response to all three stimuli. CCK levels were significantly reduced following electro-acupuncture in atropinized rats. In conclusion, gastrin and cholecystokinin release is stimulated by activation of sensory afferent, originating in skin, subcutaneous tissue as well as in muscle. Some of these effects are vagally mediated.
Key words : A-,8 and A-S afferent, CCK, electro-acupuncture, gastrin, non-noxious stimulation, somatosensory afferents, somatostatin, thermal stimulation, vibration.
T h e endocrine system of the gut is activated in response to feeding. Part of this effect is mediated by vagal efferents. Thus, in dogs sham feeding causes a rise of gastrin and CCK (Nilsson 1975, Schafmayer et al. 1988). Electrical vagal stimulation in rats also causes a release of gastrin and CCK and a decrease or increase in somatostatin levels (LindCn et al. 1989, Aliiio Correspondence : Kerstin Uvnas-Moberg, Department of PharmacoIogy, Karolinska Institutet, Box 60400, S-104 01 Stockholm, Sweden.
et al. 1983). Recently, we have been able to show that gastro-intestinal hormones such as gastrin and CCK are released in response to suckling in lactating animals of several species including rats (Linden et al. 1987, 1990). I n rats the maternal CCK release triggered by suckling of the pups is vagally mediated via a central reflex possibly involving oxytocinergic transmission (LindCn et al. 1990). Electrical stimulation of the sciatic nerve induced a release of gastro-intestinal hormones in anaesthetized cats, indicating that activation of somatosensory afferent, not only
349
from the teats b u t also from other parts of the skin, ma!- influence the release of gastro-intestinal hormones (Uvnas-.\Ioberg rt al. 1986). Sin( thin (;-fibre afferents are not likelj- t o be activated by such stimulation, the effect is possibly mediated via thicker fibres, normally actii-atcd by 1-arious kinds of non-noxious stimuli ( U m a s %loberg t t aI. 1986). I t has also been shown that ' different modes ofperipheral stimulation activate spinal cord mechanisms that affect pain and peripheral blood flow (Lundeberg rt ul. 1988a & b, Kjartansson et al. 1988). T h e present stud!- in\-estigates in more detail hon sensory stimulation influences the release of gastrin, CXK and somatostatin. 3lale rats a-ere exposed to three types o f treatment, known to set up acthit!- in different types of somatosensory n e r w fibres, i.e. vibration at 100 Hz which activates mainly thick mvelinated fibres from the skin, subcutaneous tissue and muscles. Mild thermal stimulation a t 40 "C which activates medium sized '4-8 and C-afferents, originating in the skin and electro-acupuncture ( 2 H z ) which activates thin and medium sized myelinated muscular afferents ( C h u n g el 01. 1984a & b, L u r t ul. 1986). I n order t o explore the involvement o f t h e vagal nerves and of cholinergic mechanisms in hormonal effects caused by such stimulations, experiments were also performed on vagotomized and atropinized animals.
\I E T H0 D S 1;uperiments were performed on male Sprague Danley rats (300 g) all~wedfree access to food and water. The animals \\ere anaesthetized with chloral h!-drate (.NO mg kg ', given i.p,), Experiments were routincl!performed between 9 and 11 a.m. -4fter the cxperiment, the animals were killed b!- decapitation and blood (.? ml) was collected. (The study mas approved by the local Ethics Committee.) SurKizal procedures. Vagotomy was carried out under anaesthesia. The major branches of the subdiaphragmatic vagus nerve were identified on or near the oesophagus and the anterior and posterior subdiaphragmatic vagi were dissecred free and cur. Sham operations w r e performed according to the same procedure, but the Yagi were not cut. .Stimu/utionr. 1-ibration stimulation (100 Hz) was carried out for a period of 30 min. An electromechanical vibrator (Somedic AB, Stockholm, Sueden) with a probe area of 9 cm' was applied with the constant pressure (3 kPa) to the thorax and abdomen ( T h , I , , ) ofthe rat. The vibratory movement (peak to
peak 1000 pm) was continuously measured throughout the stimulation period (Lundeberg et ul. 1984). Electro-acupuncture was performed for a period of 30 min. Points chosen were U B l l (close to the shoulder joint) and UR54 (close to the hip joint) bilaterall!. Stimulation was brought about by manual rotations of the needles (0.3 mm) after insertion to depths of0.5-0.7 cm, at the points chosen. Thereafter, the rats receiwd electrical stimulation through all the needles (CBI 1-UBll, UB54UB54) that were connected to an acupuncture pulse stimulator (B.V. Enraf-Nonius Delft, Delft, the Netherlands) producing bipolar square wave pulses of 0.2 ms duration, and 2 Hz frequency. The current intensity was adjusted so that localized muscle contractions were seen (2-5 1.) (Lundeberg et al. 1988~). Thermal stimulation was applied for 30 min to the skin of the thorax and abdomen (Th,-L,) of the rat using a feedback controlled contact thermode. T h e thermode had a contact surface of 9 cm', was fixed to a glass chamber with a constant temperature of40 "C and applied with a constant pressure of 3 kPa (Lundeberg & Ottoson 1982). Experimental design. The stimulation was started 10 min after the induction of anaesthesia and was applied for 30 min. The effect of the various stimulations was tested in two different experimental series. In series I blood samples were collected immediately - 0 min (n = 6) or 60 min ( n = 6) after the stimulation n-as finished. Control samples ( n = 6) nere collected from anaesthetized non-stimulated animals a t time point 0 min. I n the group o f animals in which the effect of temperature (40 "C) was tested, additional controls (n = 6) were performed in which the animals were exposed to the same thermode at body temperature (37 "C) for 30 min. I n the electroacupuncture group and controls, samples were also collected 330 min ( n = 6) after the end of the stimulation. T o test the effect of anaesthesia alone samples were collected 40 and 100 rnin after induction of anaesthesia ~
(ti
= 5).
In series 11. stimuli were applied to control animals as nell as atropinized, vagotomized or sham-operated animals. Atropine (0.5 mg kg-' s.c.) was given immediatel!. after induction of anaesthesia (n = 6). Vagotomy or sham operation was performed 1 week before the experiments. Samples were collected 60 rnin (n = 6) after the end of stimulation in the different experimental groups and controls. Separate controls were performed for each type of treatment. Trerittnent of blood samples. Blood samples were collected in tubes containing heparin (10 IU ml-') and Trasylol (500 I L ml-I). They were centrifuged, the plasma was removed and frozen at -20 "C. RZAs. Gastrin levels were measured in unextracted plasma and by RIA (Nilsson 1975) using antiserum No. 2604, N-hich recognizes gastrin 17 and 34 with
Somiltosensory control of GI hormones
35 1
equimolar potency. The detection limit of the assay unchanged in all groups (Table 1). Kesults w a s 2 pmol I-’ and the intra- and inter-assay co- obtained froni sham operation experiments did ctiicients of variation were 10 and 13% respectively. not diffcr from those obtained in unoperated CCK le\-els in plasma were measured after SEP- animals and \$ere therefore included in the PAK,, extraction using acetonitrile and acetic acid. control group. The RIA method used has been described by Flimcno rt 01. (1983). The antiserum OAL 656 (Otsuka Pharmaceutical Co. Ltd., ‘I’okushima, Japan) raised against the N-terminal amino acid residue of sulphated CCK-8 detects CCK-8, CCK-33 and CCK-39 but not gastrin. The intra- and inter-assay variation was 10 and 12% respectively. Somatostatin in plasma was measured by RIA after f c SEP-PAK,, extraction of acidified samples using ,r *e methanol. The antibodq Rl4lE was used with the U m method described by Efendic et al. (1980). The 17 detection limit of the assay was 2 pmol I-’ and the intra- and inter-assay coefficients of variation were 9 and 11 O 0 respectively. Stutisticnl ecaluativns. ‘I‘hc hormone levels in the figures are presented as mcan valueskSD. Comparisons between groups were evaluated with the Kruskal-Wallis one-way analysis of variance (ANOVA) and subsequent two-group comparisons were made with the Maim Whitney U-test. Twotailed tests were used and P-values < 0.05 were considered significant.
RESULTS Effect of’ anaesthesia Gastrin, CCK and somatostatin levels were not significantly different in samples collected 40 and 100 min after induction of anaesthesia i.e. corresponding to 0 and 60 min after stimulation. I n the five experiments performed, gastrin levels were I 1 5 8 PM and 116 i10 PM, CCK 1 0 i - l PM and 12k4 PM and somatostatin 12f 4 PM and 1 4 i 3 p w
Effect of vibration Gastrin and CCK levels were significantly (P< 0.05) elevated when comparcd to control values 60 min after the stimulation, whereas somatostatin levels were unchanged (Fig. 1).
f
EfSect of atropine mid z!agotowiy
A significant rise (P < 0.05) of gastrin and CXK levels was seen in response to vibration also in this series of experiments. This effect was abolished by vagotomy or following administration of atropine. Somatostatin levels were
Fig. 1. Gastrin (a), CCK (b) and somatostatin (c) levels in control rats as well as 0 and 60 min after exposure to vibration at 100 Hz for 30 min.
K. C'vnas-Moberg et al.
352
fc
.c
L
U
m
rn (3
control
''
0
60
control
0'
330'
**
-I-
l(b\
60
I
'2 E
5, Y
u u
f Y
6
10
u u
4
2 0
c control
0
60
r,c .c
.v rg
fe
c 0
U
I
1s
IC
J
I
10
c 0
tn
I
0
r
0 60 Time (min) nost stimulation
5
conipol
Fig. 2. Gastrin (a), CCK (b) and somatostatin (c) levels in control rats as well as 0 and 60 min after exposure to thermal stimulation at 40 "C.
0 control
0
60'
330'
Time (min) post stimulation
Fig. 3. Gastrin (a), CCK (b) and somatostatin (c) levels in control rats as well as 0, 60 and 330 min after exposure to electro-acupuncture at 2 Hz.
Efect of rkermaj strrnulation
A clearly significant ( P < 0.001) increase of gastrin levels was seen in response to thermal stimulation at 40 O C . CCK levels nere also raised (P < 0.05) but somatostatin levels were not altered (Fig. 2 ) . No effect was induced by the thermode at body temperature (data not shown).
Effect of atropine and vagotomy Gastrin levels were elevated ( P < 0.05) in response to thermal stimulation - an effect which was abolished by atropine or vagotomy (Table 2 ) . Results obtained in sham-operation experiments did not differ from those obtained in
Somatosensory control of GI hormones
353
T a b l e 1. Effect of vibration at 100 Hz on gastrin, CCK and somatostatin levels collected 60 rnin after end of stimulation in control animals as well as in animals treated with atropine, 0.5 mg kg-' and vagotomy
Control 60 min after stimulation Control - atropine 60 min after stimulation Control - vagotomy 60 min after stimulation
n
Gastrin (pmol I-')
n
11 12 5 5 5 6
104k18 132 f25" 126+21 119 30 190f19 186i23
12 8 + 2 12 11 3" 6 10+3 6 9 2 6 7f2 6 7 f2
+
CCK (pmol I-')
+ +
n
Somatostatin (pmol It')
12 12 6 6 6 6
11+2 9 3 llk3 10+1 11+3 10k1
+-
T a b l e 2. Effect of thermal stimulation at 40 "C on gastrin, CCK and somatostatin levels collected 60 min after end of stimulation in control animals as well as in animals treated with atropine, 0.5 mg kg-' and vagotomy
Control 60 min after stimulation Control - atropine 60 min after stimulation Control - vagotomy 60 min after stimulation
n
Gastrin (pmol I-')
n
CCK (pmoll-')
n
Somatostatin (pmol I-')
10 11 4 6 6 6
112k22 132+21" 117.t.29 120f31 179+34 163f31
12 12 6 6 6 6
9+2 1051 llf2 13 f3 11&4 10k3
12 8 + 2 12 10f 3 5 12$3 6 6k2 6 7+3 6 7+3
T a b l e 3. Effect of electro-acupuncture at 2 Hz on gastrin, CCK and somatostatin levels collected 60 rnin after end of stimulation in control animals as weil as in animals treated with atropine, 0.5 mg kg-' and vagotomy
Control 60min after stimulation Control - atropine 60minafterstimulation Control - vagotomy 60 min after stimulation
n
Gastrin (pmol I-')
n
CCK (pmoll-')
12 11 8 6 5 5
115+23 152+_26"" 120f25 121f18 186$.32 179+ 19
12 9 + 2 12 13+2" 6 15+2 6 10L-2" 6 8fl 6 9f2
n
Somatostatin (pmol 1-')
12 9 + 3 12 l l k l 8 11+2 6 10k3 6 9f2 5 7&1
Table 4. Basal levels of gastrin, CCK and somatostatin levels in controls, vagotomized and atropine-treated (0.5 mg kg-') animals Gastrin Basal levels n = 3 3 n = 17 Atropine n = 16 Vagotomy
CCK
112+30 n = 3 6 120+25 n = 18 185+34""n= 18
Somatostatin
9+2 13k2"" 9f2
n=36 n = 19 n = 18
9f2 11$2 9+2
fasted contributed to the high gastrin and CCK levels. T h e animals had to be anaesthetized in order to receive the standardized simulations. Furthermore, we chose to work with freely fed L#i,i.t ?/'ii cupu 11 ctiirt animals in which intragastric p H is high due to CCX levels u r r e significantly elevated 60 (t' < the buffering effect of food in the stomach to be 0.01) and 330 ( P < 0.05) min after stimulation able to demonstrate increases of gastrin levels. u as ended. Somatostatin levels were significantly At low antral pH, gastrin release is inhibited and eleiated at 0 and 330 min ( P < 0.05). Gastrin previous attempts to evoke a vagally induced Ic~-el.sHere significantly increased (P < 0.05) a t release of gastrin in fasted rats had proved 60 min (Fig. 3). T h e results of control experi- unsuccessful (Linden et nl. 1990). Gastrin and cholecystokinin secretion is in ments performed at 0 and 330 min did not differ part under control of the autonomic nervous and \%eretherefore combined. s!-stem. Vagal activation induced electrically or b!- sham feeding raises the plasma levels of these &'t/i~L.tu/'trr,.oplnr a t d rrrgotot?!)' hormones in dogs, cats and rats (Nilsson 1975, .\lift0 rf ul. 1983, Schafmayer er nl. 1988, Linden Gastrin and C C K levels were again influenced et tri. 1989). I n contrast, activation of the hj- acupuncture ( P < 0.01). This effect was sympatho-adrenal system inhibits vagally-inabolished by \-agotomy and by administration of duced hormone responses. T h u s gastrin release atropine. I n the case of CCK, a significant caused by electrical vagal stimulation in cats is decrease \\-as seen after atropinization. SomAtoabolished during concomitant stimulation of the statin h e l s \\ere not influenced (Table 3). splanchnic nerve (Uvnas-Wallensten & Jarhult Results obtained in sham-operation esperiments 1982). Furthermore, in rats, administration of did not difkr from those obtained in unoperated 3,-adrenergic agonists such as clonidine inhibits anin\als and were therefore included in the the \-agall>--induced release of somatostatin into control group. the gastric lumen of anaesthetized rats (UvnasMoberg & Aliiio 1991). Activation of the unm!-elinated C-fibre afferents is related to pain qf'trtroprne atid ragotomy on biisrri Irrtis transmission and activation of the sympathooJ:listuti, C C K &nu' soin(rtmtatrn adrenal system and consequently to an inhibition U'hen data from the three separate experimental of vagally induced effects on the release of GI groups ere analvsed together, atropine (0.j mg hormones. hg '1 caused a significant ( P < 0.01) increase in I n a previous study performed on anaesCCK and vagotom!- produced an increase of thetized cats, the levels of gastrin and some other gastrin le\ els in comparison to untreated animals gastro-intestinal hormones were raised in reSomatostatin l e d s were not influenced sponse to afferent electrical stimulation of the (1';ihle 4). sciatic nerve at low stimulatory intensity (Uvnas\loberg et a(. 1986). At such stimulation, thick and medium sized afferents, but not C-fibres, are likely to be activated. I n order to further explore the eEect of nonT h e main findings of the present study vere that vibration a t 100 H z and thermal stimulation at nosious somatosensory stimulation on the release 40 "C as well as electro-acupuncture ( 2 Hz) of gastro-intestinal hormones, three different significantly increased gastrin and C C K levels, kinds of afferent stimulations known to activate efects ahich were abolished by \-agotom!- or specific populations of sensor!- nerve fibres (i.e. atropine pretreatment. T h e results show that 1-ibration at 100 Hz, thermal stimulation at 40 "C difterent modes of sensor!- stimulation may and deep muscle stimulation a t 2 Hz) were induce a vagally mediated cholinergic release of applied. More specifically vibratory stimulation some gastro-intestinal hormones. acti\-ates all types of low threshold mechanoRasaf rastrin and C C K levels were high under receptors in the stimulated area and the activity thc present experimental conditions. Both the is transmitted in A-/? fibres (Merzenich & anaesthesia and the fact that animals were not Harrington 1969). Thermal stimulation is com-
unoperdted animals and were therefore included in the control group.
cf/t,~~
Somatosensory control of GI hormones plex and the types of fibres that are activated depend on the temperature. Above 43 "C heat receptors are activated and the sensation of heat is conducted in afferent Cfibres. I n the present studies in which a thermode of 40 "C was used, A-6 and C-fibres are likely to have been activated (Kenshalo 1970). Finally, electro-acupuncture (i.e. low frequency electrical stimulation of muscles) activates several types of mechanoreceptors within the muscle. Among these are the so-called ergoreceptors. T h e activity set u p in these receptors is transmitted in A-6 fibres (Kniffki et al. 1981). T h e present data show that stimulation of sensory afferents from skin and muscle caused a release of gastrin and C C K (and sometimes of somatostatin). Whether the sensory stimuli used really are equipotent as to their hormonestimulating potential is difficult to say. I t seemed as if thermal stimulation was particularly efficient in stimulating gastrin release and electro-acupuncture in CCK release. However, the intensities of the stimulations are difficult to compare and also basal levels which are of importance for the outcome of the stimulations varied somewhat between the experimental groups and series. T h e finding that vagotomy abolished the effects on gastrin and C C K release caused by vibration, thermal stimulation and acupuncture indicates that the effects were vagally mediated. I t is also possible that it is difficult to further enhance the high basal values of gastrin seen after vagotomy. Pretreatment with atropine also abolished the responses, indicating that the vagal effect of gastrin and C C K secretion is mediated by cholinergic mechanisms. However, administration of atropine alone also caused a rise of basal CCK levels, indicating the presence of a cholinergic inhibitory tone on C C K release. Indeed, an enhanced secretion of C C K following feeding in rat has been demonstrated to occur following treatment of atropine in rats. T h i s increase was suggested to be a consequence of a reduced secretion of pancreatic juice and hence to lower levels of inhibitors of C C K secretion in the duodenum (Nakano et al.
1990). Electro-acupuncture caused a significant decrease of C C K levels when they were previously raised by treatment with atropine. T h i s finding indicates that electro-acupuncture may not only increase C C K levels but also in case of elevated
355
basal C C K levels lead to an inhibition of CCK release. By which mechanism this effect is mediated is not known. Acupuncture stimulation has previously been shown to reduce sympathetic nerve activity and in addition Sato et al. have shown that nonnoxious stimulation of the skin leads to a reflex inhibition of both sympathetic nerve activity and catecholamine secretion (Sato 1987). T h e present data show that not only is sympathetic transmission inhibited by these kinds of somatosensory stimulation, in addition some parasympathetic vagally mediated effects are induced (Sato & Schmidt 1987). I n lactating rats the maternal levels of gastrin, and particularly of CCK, rise in response to suckling of the pups. This effect is abolished by vagotomy. Since lesions of the afferent nervous pathways within the CNS, leading to oxytocin secretion inhibit suckling related C C K release and since oxytocinergic neurons from the paraventricular nucleus innervate the dorsal vagal motor nucleus, the suckling related release may involve a central oxytocinergic pathway (Linden et al. 1990). Whether central mechanisms are also involved in the gastrin and CCK release caused by vibration, thermal stimulation at 40 "C and electro-acupuncture remains to be established. We thank the Swedish Medical- Research Council (project No. B9214X-05207-15C, Ake Wiberg's Foundation, RMR (49-9 l), Magn. Bergvall's Foundation, Nanna Svartz Foundation, King Gustaf V Anniversary Found, SCAA, Anna-Greta Crafoord's Foundation. REFERENCES A L I ~ ~ oS.F., , GARCIA, D. & UVNAS-MOBERG, K. 1983. On the interaction between intragastric pH and electrical vagal stimulation in causing gastric acid secretion and intraluminal release of gastrin and somatostatin in anesthetized rats. Acta Physiol Scand 117, 491495. CHUNG,J.M., FANG,Z.R., HORI,Y., LEE, K.H. & WILLIS, W.D. 1984a. Prolonged inhibition of primate spinothalamic tract cells by peripheral nerve stimulation. Pain 19, 259-275. CHUNG,J.M., LEE, K.H., HORI,Y., ENDO,K. & WILLIS, W.D. 1984b. Factors influencing peripheral nerve stimulation produced inhibition of primate spinothalamic tract cells. Pain 19, 277-293. F., NYLEN,A. & UVNASEFENDIC,S., ENZMANN, WALLENSTEN, K. 1980. Sulphonylurea (Glibenclamide) enhances somatostatin and inhibits gluca-
356
K. Urnas-Moberg et al.
imentally induced pain. Ear Nose Throat 3 67, gon release induced by arginine. .4cta Ph.ysiol Scand 565-57.5. 108, 231--233. T., NORDEMAR, R. & OTTOSON, D. 1984. H I hl E U O , S., TARCI, S . , KANAYAMA, s., KCROSHIIIA, LCNDEBERG, Pain alleviation b j vibratory stimulation. Pain 20, *r.,SHINOMURA,Y., HAYASHI,c., TATEISHI,K . , 2544. IMAGAWA, K., HASHIMURA, E. & HAMOUKA, T. LCSDEBERG, T., ABRAHAMSSON, P., BONDESSON, L. & 1983. Plasma cholecystokinin responses after inHAKER, E. 1988a. Effect of vibratory stimulation on gestion of liquiol meal and intraduodenal infusion of experimental and clinical pain. Scand 3 Rehab M e d fat, amino acids, or hydrochloric acid in man: 20. 149-159. Analysis with region specific radioimmunoassay. LL-NDEBERG, T., HURTIG,T., LUNDEBERG, S. & Ani 3 Gastroenterol 78 (1 l), 703-707. THOMAS,M. 1988c. Long-term results of acuKEKSHALO,D.R. 1970. Psychophysical studies of puncture. Pain Clin 2, 15-31. temperature sensibility. I n : \V.D. Neff (ed.) Con,\IERZENICH, ht..u. & HARRINGTON, T. 1969. The tributions to Sensory Ph,yslohgjJ, vol. 4, pp. 17-74, sense of flutter-vibration evoked by stimulation of Academic Press, New York. the hairy skin of primates: Comparison of human KJART~SSSON, J., LLSDEBERG, T . , SAXELSOT, U E., sensory capacity with the response of mechanoDALSGAARD, C-J. & HEDEN, P. 1988. Calcitonin receptive afferents innervating the hairy skin of gene-related peptide (CGRP) and transcutaneous monkeys. Exp Brain Res 9, 236-260. electrical nerve stimulation (TENS) increase cuN.AK.4S0, I., TAWIL, T., SPANNAGEL, A.W., LIDDLE, taneous blood f l o ~in - a musculocutaneous flap in R.A. & GREEK,G.M. 1990. i\tropine enhances the rat. .4cta Ph,ysiol Scand 134, 89-94. MENSE,S . & SCHMIDT,R.F. 1981. K ~ I F F KK.D., I, food-stimulated CCK secretion in the rat. Pancreas 5, 621-625. Muscle receptors with fine afferent fibres which G. 1975. Increased plasma gastrin levels in may evoke circulatory reflexes. Circ Res 48 (suppl SII.SSOS, connection with inhibition of gastric acid in I), 125-131. response to sham feeding following bulbar perfusion L I X D ~ S.%,, , CARLQLIST, Lu., H.ASSEN,s. & with acid in dogs. ScandJGastroenterol 10,273-277. ~ I O B E RK G., 1989. Plasma concentrations of choleSATO, T. 1987. Neural mechanisms of somatic sensory cystokinin, CCK-8 and CCK-33,39 in rats, determined by a method based on e n q m e digestion of regulation of catecholamine secretion from the gastrin before HPLC and RI.4 detection of CCK. adrenal gland. Adz. Biophys 23, 39-80. Glii 30 (2), 213-222. SATO, 3.& SCHMIDT,R.F. 1987. T h e modulation of I,INDES, .I,, ERIKSSON, &I., CARLQCIST, 51. & \-isceral functions by somatic afferent activity. 3ap 3 MOBERG, K . 1987. Plasma levels of gastrin, somatoPhjisiol 37, 1-17. statin and cholecystokinin immunoreactib-it! during A, YUSTEDE,A., POMPINO,A. & SCHAFVAYER, KOHLER, H . 1088. Vagal influence of cholepregnancy and lactation in dogs. Gastroenterdog), 92, 578-584. cystokinin and neurotensin release in conscious I,IuDE\, .I., ERIKSSOS,>IH.~NSES, ., s., Lvxisdogs. Scand 3 Gastroenterol 23, 3 15-320. ~ I O B E RK. G 1990. Suckling-induced release of , 1991. Vagal and -MOBERG,K . & A L I ~ OS.F. HCI-induced release of somatostatin and gastrin cholecystokinin into plasma in the lactating rat : into the gastric lumen of rats. In : R. Hdkanson & F. effects of abdominal vagotomy and lesions of central patha-aJ-s concerned with milk-ejection. 3 EnduSundler (eds.) The Stomach as an Endocrine Organ, m n o l 127, 257-263. 51-163. Elsevier Science Publisher. I,[, G., Z m o , G., LIASG, R.Z., ~VASG,I-.,SISG, ii-, ~ I O B E R K., G , POSLONCEC, B. & AHLBERG,I,. & S H I , G. 1986. .%nalysis of afferent fibres . Influence on plasma levels of somatostatin, transmitting needling sensations of acupoint. In : X. gastrin, glucagon, insulin and VIP-like immunoZhang (ed.) Research on ..lrupunr.ture, .tlosibuJtion reactkit! in peripheral venous blood of anaesmil Acupuncture ,4naesthesia, pp. 16tr177. Science thetized cats induced by low intensity afferent Press, Beijing. stimulation of the sciatic nerve. Acta Physiol Scand I,CSDEBERG, T. & OTTOSOU, D. 1982. The effect of 225-230. peripheral stimulation and placebo in chronic -W-ALLENSTEN, K. & JARHLLT, J. 1982. Reflex neuralgic pain states. Acra Ph,ysiol Scand, suppl. activation of the sympatho-adrenal system inhibits S38. the gastrin release caused by electrical vagal LCVDEBERG, T., I,ALItEI.L, G. & THOMAS,31. 1988 b. stimulation in cats. Acta Ph.ysiol Scand 114, Effect of acupuncture on sinus pain and exper297-302.
