Fish Physiology and Biochemistry vol. 4 no. 3 pp 143-155 (1987) Kugler Publications, Amsterdam/Berkeley
Control of gallbladder motility in the rainbow trout, Salmo gairdneri G6ran Aldman and Susanne Holmgren Comparative Neuroscience Group, Department of Zoophysiology, University of GOteborg, Box 250 59, S-400 31 G6teborg, Sweden Keywords: gallbladder, rainbow trout, neuropeptides, hormones, motility
Abstract
Immunohistochemistry revealed nerves containing VlP-like and 5-HT-like material in both gallbladder wall and bile duct of the rainbow trout, while endocrine cells containing gastrin/CCK-like and substance P-like material were present in the mucosa of the bile duct and the duodenum. Fluorescence histochemistry showed adrenergic nerves close to the muscle layer of the gallbladder. Sulphated CCK8, caerulein and non-sulphated CCK8 (in this order of potency), 5-HT and acetylcholine were excitatory on isolated strip preparations, while VIP and adrenergic agonists were inhibitory. The adrenergic drugs were probably acting via a beta-adrenergic receptor, while the effects of 5-HT and cholinergic drugs were antagonized by methysergide and atropine respectively. Electrical stimulation of the gallbladder nerves in situ failed to show any effect or under certain conditions induced a rebound effect. It is concluded that the motility control of the rainbow trout gallbladder may involve an inhibitory innervation by adrenergic and possibly VIP-releasing nerves, while 5-HT, acetylcholine and a CCK-like substance may be involved in the excitatory control.
Introduction
In the rainbow trout, Sahno gairdneri, the bile secreted from the liver is stored in a gallbladder. The bile duct leads bile both from the liver to the gallbladder and from the gallbladder to the proximal intestine (duodenum). A well developed smooth muscle layer in the gallbladder wall (Yasutake and Wales 1983) ensures emptying of bile into the intestine. Little is known about the nervous or hormonal control of the gallbladder motility in fish. In the cyclostome Myxine glutinosa vagal stimulation causes contraction of the gallbladder (F~inge 1958). Cholinergic receptors of the muscarinic type are
present in the gallb'ladder muscle, but cholinergi~ innervation is sparse judging from the measured activity of choline acetyltransferase (Holmgren and F/:inge 1981). The gallbladder of the coho salmon, Oncorhynchus kisutch, is contracted by porcine cholecystokinin (CCK) in vitro (Vigna and Gorbman 1977) while that of the spiny dogfish, Squalus acanthias, shows only weak responses if any to CCK (Vigna 1983). In the cyclostome Eptatretus stouti both CCK and the closely related peptide caerulein fail to produce measurable effects (Vigna 1979). In mammals (guinea pig, cat), tonus and motility of the gallbladder is regulated by the autonomic nervous system, which forms several plexuses in the
144 gallbladder wall. Both catecholamine containing and acetylcholinesterase containing fibres are present (Cai and Gabella 1983) Adrenergic nerves may cause contraction or inhibition via alpha and beta adrenoceptors respectively (Persson 1972; Doggrell and Scott 1980), while CCK8 and cholinergic nerves stimulate contraction (Hedner 1969; Berry and Flowers 1971; Behar and Bianchi 1980; Nakata and Kurhashi 1981). VIP-like immunoreactivity and substance P-like immunoreactivity are reported in the mammalian gallbladder (Cai and Gabella 1983), and VIP relaxes the gallbla.dder and induces fluid secretion into the lumen in the cat (Jansson et al. 1978). This study was carried out as an attempt to elucidate the nervous and hormonal control of the gallbladder motility in a teleost, the rainbow trout, Sahno gairdneri.
Materials and methods
Rainbow trout, Sahno gairdneri, ca 600 g and 100- 150g were bought from a local hatchery. They were kept in aerated, recirculating fresh water at 10~ and fed once a week until used. They were killed by concussion and opened ventrally and laterally for in situ studies of the gallbladder innervation or for removal of the gallbladder.
Light microscopy Tissues were fixed in phosphate buffered 4~ formaldehyde solution, dehydrated and embedded in paraffin. Seven ~m thick sections were stained with hematoxylin and eosin.
Falck-Hillarp fluorescence microscopy Preparations were treated according to the FalckHillarp technique (Falck and Owman 1965) for visualization of adrenergic nerves. Pieces of tissue were snap frozen in liquid propane cooled in liquid nitrogen, freeze-dryed for 3 - 5 days, treated with paraformaldehyde vapour at 80~ for one hour
and vacuum-embedded in paraffin. Sections (7 ~m thick) were mounted in Entellan (Merck) and viewed immediately in a Leitz Dialux fluorescence microscope with incident light illumination and a barrier filter at 460 nm.
Immunohistochemistry Whole-mounts: Preparations were made according to the technique described by Costa et al. (1980). Pieces of gallbladder were stretched out and pinned on flat sheets of dental wax. They were fixed in a solution of formaldehyde (2%) and picric acid (15~ in a phosphate buffer (0.1M, pH 7.2) for 18 h at 4~ After rinsing in 80~ alcohol the preparations were dehydrated, xylene-treated for 30 min and then rehydrated. Thin sheets were peeled off and incubated with primary antisera (see Table 1) in a moist chamber at room temperature for 20 h. After incubation the preparations were rinsed several times in phosphate-buffered saline (PBS, 0.1M, pH 7.2) and then incubated with a second antibody (swine-anti-rabbit IgG diluted I:10 (DAKO)), which was labelled with FITC (fluorescein isothiocyanate) (Coons 1956). After rinsing in PBS the preparations were mounted on slides in carbonate-buffered glycerol (1:1, pH 8.5). The slides were viewed in a Leitz Dialux fluorescence microscope with a barrier filter at 510 nm. Film used was Ilford FP4 or Kodak TriX. Controls were carried out for those antisera giving a positive immunoreaction. The primary antiserum was preincubated with its hapten, i.e. the antiserum raised against VIP was preincubated with porcine VIP (10 nmol/ml), the antiserum raised against 5-HT was preincubated with 5-HT (10 nmol/ml), the antiserum raised against gastrin (G4) was preincubated with gastrin (G17, 10 nmol/ml) and the antiserum raised against substance P was preincubated with substance P (10 nmol/ml) at 4~ for three days. All preabsorbed antisera failed to stain any nerve fibres or cells on subsequent use. Sections: Preparations were made according to the technique described by Bishop et al. (1978). Pieces of gallbladder and bile duct were fixed in parabenzoquinone (4 mg/ml) in sodium cacodylate buffer
145 l-ahle I. Antisera used in immunohistochemical experiments
Antisera raised against
(._'ode
Hapten
Bombesin Bombesin
L89 k90
Gastrin
L112
Synthetic bombesin G4
Mct-enkephalin
L 146
Neurotensin
Dilution
Specificity
1:100 1:50 C-terminus
Mel-enkephalin
1:50
Free C-terminus
NTab
Neurotensin
1:50
Subslance P
CRBAB30
Synthetic SP 11
1:100
C-terminal specific in RIA C-terminus
VlP
L85
1:400
C-terminus
5-HI
71954
Synthetic 18-28 VIP 5-HT serum albumin
Strip preparation
experiments
Gallbladders were dissected out and opened longitudinally from one end to the other. The bile duct was cut off and the bladder was cut in half by a second cut at 90~ angle to the initial longitudinal section, producing two equal preparations, in which it was possible to measure the tension of circularly arranged muscle fibres. The preparations were mounted with an initial tension of 1 milliNewton (mN) in organ baths containing rainbow trout Ringer's solution (NaCI 7.41, KCI 0.36, C a C k 0.17, NaCO 3 0.31, N a z H P O 4 • 2 H 2 0 2.0, Na HzPO 4 0.40, MgSO 4 0.15 and glucose 1.0 g/l; Holmgren 1983 modified from Lockwood 1961) at 10~ bubbled with a gas mixture of O~ (97%) and CO~ (3%) and left for up to 18 h before the experiment.
1:50
Source Dockray Dockray
I: 100
(0.1M, pH 7.3) at 4~ for 4 h. After rinsing in sodium cacodylate buffer containing 20% sucrose for 24 h at 4~ the tissue pieces were snap-frozen in liquid nitrogen and cut in sections of 10 um on a cryostat. The sections were then placed on glass slides covered by a film of chrome-alum gelatine. Incubation with antisera was made as for whole mounts.
Reference
Dockray el al. 1981 Bulochet al. 1983 Dockray, pers. comm.
Dimaline et al. 1980
Dockray Dockray Dockray Cambridge Res. Biochemicals Dockray Immunonuclear
The responses to cumulative concentrations of agonists were studied. Tensions of the preparations were recorded via Grass FT03 isometric transducers on a Grass Polygraph rood 7. Drugs used were acetylcholine chloride (ACh), L-adrenaline bitartrate (A), atropine sulphate (Atr), caerulein (Cae) (Peninsula), carbachol (CCh), cholecystokinin octapeptide (CCK8), sulphated cholecystokinin octapeptide (CCK8S) (Research plus), gastrin 17 (GI7), sulphated gastrin (G17S) (Peninsula), 5-hydroxytryptamine creatinine sulphate (5-HT), Dkisoproterenol HCI (Iso), methysergide (1-methyllysergic acid butanolamine bimaleate) (Met) (Sandoz), pentagastrin (PG) (Peninsula), phentolaminr hydrocloride (Phent), DL-propranlol HCI (Propr), tetrodotoxin (TTX), vasoactive intestinal polypeptide (VIP) (a kind gift from Prof. V. Mutt, Stockholm). Drugs were purchased from Sigma if not otherwise stated. Dose ranges tested were: CCK8S and caerulein 10 -14 - 10-TM; CCK8, GI7 and GI7S 10 -I1 10-VM; PG 10 -l~ - 10-6M; A 10 -9 - 10-TM; NA 10 -9 - 10-SM; Iso 10 -12 - 10-6M; ACh and CCh 10 -8 - 10-4M.
146
Calculations
g.c i ~ Response was calculated as the work performed (surface area on tracing) p~r minute, when a maximal effect of an addition of agonist was achieved. This was compaired to the total maximal effect obtained (100%), and concentrations-response curves were plotted. Mean individual sensitivity (Ari6ns and Simonis 1961) was calculated, and pD 2 (-log ECs0) -values (van Rossum 1964) were determined.
Q.c.-m/~)
liver
Resulls
Extrinsic innervation of the gallbladder Dissection, low power magnification microscopy and osmium staining, showed that small branches of the splancnic nerves running along small mesentric vessels reached the surface of the gallbladder (Fig. 1). No direct connection from the vagal branches could be observed, but it must be emphasized that this region of the gut is difficult to dissect due to the numerous small vessels with accomparting nerves running in different directions in the mesenterium, and the amounts of fat stored in this area. Stimulation in situ of the gastrointestinal vagal branc[aes and of the splancnic nerves running along the coeliac and mesenteric arteries with 4 - 20 Hz, 0.5 - 2 m s a n d 5 - 2 0 V f o r 5 - 20 sec (n = 5) did not produce any macroscopically detectable contractions of the gallbladder, while occasional spontaneous contractions were readily visible. However, if the gallbladder was detached from the liver and the mesenterium shared with the proximal part of the coeliac artery, rebound contractions of the gallbladder were clearly visible 2 - 5 sec after stimulation (10 Hz, 1 ms, 10 V, 5 - 2 0 sec) had ended (n = 5). Stimulation in vitro of whole intact gallbladders, whole longitudinally opened gallbladders or strip preparations produced weak and inconsistent results.
Fig. 1. Highly schematic drawing of the extrinsic innervation of the gallbladder of the rainbow trout, Salmo gairdneri. Branches of the splanchnic nerve (n.sp[.) reach the gallbladder along small vessel branches from the mesenteric artery. Nerves were also observed along the small vessels passing between the liver and the gallbladder. It is possible that nerves also enter the gallbladder from the region of connective tissue, vessels, nerves and fat indicated by the hatched area at the neck of the gallbladder. Legend: a.c.-m. - arteria coeliaco-mesenterica, g.c. - ganglion coeliacure, n.spl. - nervus splanchnicus.
Histochemistry The gallbladder wall consists of an inner mucosa (lamina epithelialis) attached to a basal m e m b r a n e (lamina propria), outside this the submucosa (strata c o m p a c t u m and granulosum), a layer of smooth muscle-cells (muscularis) and the serosa (Yatsutake and Wales 1983; Fig. 2). In both the gallbladder and bile duct walls, the Falck-Hillarp technique reveals a large number of varicose nerve fibres in the outer part of the smooth muscle layer showing the blue/ green fluorescence specific for catecholamines (adrenaline/noradrenaline). In the submucosa, especially close to the basal m e m b r a n e , nerve fibres and occasional cells were observed showing the yellow fluorescence characteristic of 5-hydroxytryptamine (5-HT; Fig. 3).
148
hnmunohistochemistry The presence of a 5-HT-lil~e substance in nerves of the gallbladder was further confirmed with immunohistochemistry. The 5-HT-like immunoreactivity wasweak, but a loose plexus of nerve fibres could be observed in the gallbladder wall (Fig. 4). Within this plexus cell bodies of a bipolar appearance were noticed (Fig. 5), while occasional star shaped cells found outside the plane of focus of the nerve plexus (Fig. 4) did not seem to be in direct contact with the plexus. Varicose nerve fibres showing VIP-like immunoreactivity formed .a well developed dense plexus throughout the muscle layer with an increasing density towards the inner parts in both the gallbladder (Fig. 6) and the bile duct (Fig. 8) wall. The fibres were seen concentrated in ganglia in the nexuses of the plexus (Fig. 7). Populations of endocrine cells showing substance P-like immunoreactivity (Figs. 9 and 10) and gastrin/CCK-like immunoreactivity (Figs. 11 and 12) were present in the mucosa of the bile duct and in the mucosa o f the duodenum in the area of the bile duct entrance. Antibodies raised against bombesin, metenkephalin and neurotensin failed to reveal specifically immunoreactive material.
Strip preparations About 40% of the preparations showed spontaneous rhythmic activity within 18 h. This activity was not blocked by tetrodotoxin (TTX, 10-6M, n = 5; Fig. 13), atropine (10-6M, n = 14) or the alpha adrenergic antogonist phentolamine (10-6M, n = 5), but was often enhanced by the beta adrenergic antagonist propranolol (10-6M, n = 14, Fig. 13). This indicates the presence of a beta adrenergic tonus, but treatment of inactive preparations with tetrodotoxin (10-6M, n = 6) to antagonize a nerve mediated inhibition failed to induce rhythmic activity of the preparations. Agonist drugs could affect the preparations in three different ways: Change of tonus, or change in frequency or force of contraction. Usually the ef-
fect was a combination of the three. Cumulative concentration-response tests were performed, Excitatory, concentration dependent effects on the preparations were obtained with cholinergic drugs, 5-HT and CCK/gastrin related peptides, while VIP and adrenergic drugs were inhibitory. Adrenaline (10-6M, n = 5 ) abolished spontaneous activity (Fig. 13) and activity induced by 5-HT (10-6M, n = 5 ) . Pretreatment with propranolol (10-6M) inhibited this effect of adrenaline ( n = 6 ; Fig. 13). Adrenaline (PD2=7.70-+0.15, n =4) and isoprenaline (PD2=7.45 _+ 10.35, n = 4) were significantly more potent (p > 0.01) than noradrenaline (pD 2 = 6.24_+ 0.28, n = 4) to inhibit the spontaneous activity of the preparations (Fig. 14). VIP had an inhibitory effect on spontaneously active preparations, even in low concentrations (10-1~ _ 10-7M, n = 4 ; Fig. 13). Cholinergic drugs were excitatory on the gallbladder strips. Carbachol (pD 2 = 6.01, n = 5; Figs. 15, 16) was more potent than acetylcholine, indicating the presence of a cholinesterase activity in the preparations. Contractions caused by carbachol (10-6M) were reduced or abolished by atropine (10-6M, n = 5 , Fig. 15), indicating an effect via muscarinic receptors in the preparation. 5-HT ( P D 2 = 5 . 8 5 + 0 . 1 9 , n = 5 ) produced an excitatory effect in the same concentration range as carbachol (Figs. 16, 17). The effect of 5-HT (10-6M) was inhibited by methysergide (10-6M, n = 5 ; Fig. 17), indicating a specific receptor-mediated effect of 5-HT. Of the CCK/gastrin related peptides tested, sulphated CCK 8 (pD2= 10.69_+0.28, n = 7 ) was the most potent, followed by caerulein (PD2=8.83_+ 0.23, n = 6), while sulphated or non-sulphated gastrin 17 and nonsulphated CCK 8 produced weaker responses in a much higher concentration range (Fig. 19).
Discussion
The existence of an autonomic innervation of the gallbladder of the rainbow trout is indicated by the
150 0
TETRODOTOXIN
5min % RESPONSE
mN
I
0
"
~
20
10-"-,
'
-6
b
2
ADRENALINE
A ~~ISO
0
-85 -8 C
PROPR
2
A
-75 -7 10"3M
50
~l, qll
0 A-6
d
i
Vlp
80
-II
-[0
-7 LOG MOLAR CONC
log molor conc
-6
Fig. 13. Circular smooth muscle preparations of the gallbladder o f Salmt~ gairdneri, a. Tetrodotoxin has no effect on spontane-
Fig. 14. Concentration response curves showing thc mean in-
ous r h y t h m i c activity, b. Adrenaline is inhibitor,,' on the spontaneous activity, c. The 13-adrenergic antagonist propranolol enh a n c e s the spontaneous activity and blocks the inhibitory effect o f adrenergic drugs, d. V I P is inhibitor:,, on the spontaneous activity.
range 20-80~ o f total abolishment of activity, calculated as w o r k performed per unit time • S.E. A - a d r e n a l i n e , n 4; ISO - isoprenaline, n = 4 ; N A - n o r a d r e n a l i n e , n = 4 ; **, p > 0.01, students t-test.
d i v i d u a l sensitivity of the preparations to adrenergic drugs in the
CARBACHOL
anatomical and histochemical results in this study. The Falck-Hillarp technique shows the presence of an ad,renergic innervation, and also indicates the presence of a serotonergic innervation, although not of the same extent. The immunohistochemical results further confirm the presence of a low number of nerves containing a serotonin-like substance, and in addition indicate the presence o f nerves containing a VIP-like peptide. The antisera used in the immunohistochemical experiments were raised against mammalian or amphibian peptides, which in most cases differ somewhat in the amino acid sequence from that o f their counterparts occurring in teleost fish. A positive reaction between the mammalian antibody and the teleost peptide, is still often achieved, but this positive reaction may only indicate the presence o f a peptide similar and not necessarily identical to the original hapten. An antibody raised to a mammalian peptide may, on the other hand, fail to react
i ]
mN o
411
~ ~'"'~" ,,~L,~,.~"~'2;.,.2",',~
,.,. , ~'.:::1::~--" ' "
log molor ATR 10"6M 9 I J
75 ~
mN
I ',~ll.
qh~N, 0
hlllll
'
~'
.
.
.
.
.
conc
._JL_~L__ .
log molor conc Fig. 15. The effect of cumulative concentrations o f c a r b a c h o l ( C C h ) on gallbladder smooth muscle (upper tracing). T h e contractile effect produced by C C h is abolished in the presence o f
atropine (ATR; lower tracing).
with the teleost variant of the same peptide, and a negative result in our study must therefore not be taken as evidence for the absence of a certain peptide. The presence o f muscarinic cholinergic receptors mediating an excitation o f the gallbladder smooth
151 % RESPONSE
% RESPONSE
80.
80
CClT 50
CCK8 50
721-~
20
5
l/
00l
_--
20
-&
-,~
-4
-,~
LOG MOLAR CONC
,I.
-,i
-,'o
-9
-8
LOG MOLAR CONe
Fig. 16. Concentration response cur',es showing the mean individual sensitivity ( • S.E.) of gallbladder smoolh muscle preparation,, to carbachol (CCh, n = 5) and 5-hydroxytryptamine (5-HT, n = 5) in the range 2 0 - 8 0 % of maximal effecl, calculated as work performed per unit time.
Fi~. 18. Concentration response curves shov,ing the mean individual sensitivity (_+ S.E.) of the gallbladder smooth muscle preparations to sulphated cholec.vstokinin 8 (CCK8S) and caerulein (CAE) in the range 20-80~ of maximal effect, calculated as work performed per unit time. CCK8S is significantly 1"** = P < 0.00l) more potent than caerulein, students t-test.
5 ,'11in
5-HT mN I 0
4
-?
-&
-g
METH 10"6M
I0
9
o~
m'~, -6
;~_d~.~,~._~,G,jt
-
-6 log molar conc
Fi,e. 17. The effect of cumulative concentrations of 5-hydroxytryptamine (5-HT) on the gallbladder smooth muscle (upper tracingL The stimulator v effect produced by 5-HT is reduced in the presence of the serotonergic antagonist methysergide (METH; lower tracing).
muscle is shown in this study, but similar to the situation in the cyclostome Myxine glutinosa (Holmgren and F~inge 1981) it is doubtful whether the gallbladder has an extrinsic cholinergic innerva-
tion. The presence of a functioning adrenergic innervation, excerting an inhibitory action via /3adrenoceptors, seems more unequivocal. It is also likely that this adrenergic innervation normally exerts an inhibitory tonus on the preparations, since treatment with the f3-adrenergic antagonist propranolol usually caused the onset of or increase in rhythmic activity. A profound effect of 5-HT on'smooth muscle from the fish gut has earlier been shown in Salmo gairdneri as well as in other teleost and elasmobranch species Young 1980, 1983; Holmgren 1983; Holmgren and Nilsson 1983; Jensen and Holmgren 1985). This effect was in most cases demonstrated as a clearly excitatory action, in accordance with the results in the present study. Serotonergic nerves have also earlier been demonstrated by histochemical methods in the gut of both cyclostomes, elasmobranchs and teleosts (Baumgarten 1967; Watson 1979; Sakharov and Salimova 1980; Anderson 1983; Holmgren and Nilsson 1983;
152
Cae 2 mN
5mm
0
-,b
-9
-8
5 log molar conc
CCK8S
2
CCK8
2
mN 0
0
-,2
-tl
-,o
-;
-8
" -
-7 lOg molar cone
2
GI7S
2
GI7
mN 0
-G
-8
5
-8
-7 log molar conc
Fig. 19. The effects of cumulative concentrations of different CCK/gastrin-related peptides on gallbladder smooth muscle. The responses to caerulein and sulphated CCK8 reached a maximum within the concentration range tested, while the effects of non-sulphated CCKg and sulphated and non-sulphated gastrin 17 occurred in a much higher concentration range and never reached a maximum.
Jensen and Holmgren 1985; Holmgren et al. 1985). In the present study of S a l m o gairdneri 5-HT (or a closely related substance) is present in nerves of the gallbladder wall according to both immunohistochemical and fluorescence histochemical results. However, the innervation seems rather sparse and is situated closer to the mucosa than to the muscle layer, and it is possible that at least part of the excitatory effect that can be induced by exogenous 5-HT may in physiological conditions be caused by circulating 5-HT released from endocrine cells or mast cells in the gallbladder or elsewhere. The starshaped 5-HT-containing cells in the gallbladder wall, probably without contact with the 5-HT containing nerves, may be involved in such an endocrine/paracrine function. Previous studies in Sahno gairdneri have shown the presence of such cells in the gastrointestinal canal (Holmgren et al. 1985). Gastrin, cholecystokinin (CCK) and caerulein
share the C-terminal pentapeptide sequence. The antibodies used to localize these substance were raised towards the c o m m o n C-terminal tetrapeptide, hence it is not possible to use this antibody to differentiate between these substances. No gastrin/CCK-like immunoreactive material was, however, encountered in nerves of the gallbladder, and g a s t r i n / C C K affecting the gallbladder motility may originate from the endocrine cells of the gut and bile duct mucosa reported in the present study and previously (Holmgren et al. 1982). The identity of gastrin/CCK-like peptides and their point of evolution has been much discussed (Walsh 1981; Crim and Vigna 1983). An evolution at the amphibian-reptile level of gastrins from a CCK-like peptide, may be a peptide closely resembling caerulein, has been suggested (Larsson and Rehfeld 1977). Other authors have instead even questioned a closer relationship between caerulein and gastrin/CCKs (Walsh 1981), although
153 the shared C-terminal pentapeptide is considered to constitute the biologically active center of these peptides. A recent radioimmunological study now indicates the presence of several gastrin/CCK/ caerulein-like peptides in gut extracts from the rainbow trout; regional differences exist and it is suggested that the occurrence of gastrin/CCK-like variants is even more extensive than in mammals (Vigna 1985). Little conclusive evidence of the true identity of the gastrin/CCK-like peptides present in the fish gut is thus available, and consequently little is known about the exact function of these peptides. Although, in mammals, the effects of gastrins and CCKs are overlapping, it is considered that CCK is the predominant regulator of the gallbladder motility, while gastrins have their major effects on e.g. gastric secretions. In the present study, CCK is also more potent than gastrin in initiating rhythmic activity in the gallbladder smooth muscles of the rainbow trout. The presence of a sulphate group (in position 7 from the C-terminus in CCK and in position 6 in gastrin) has been found important for the biological activity of CCK and gastrin in mammals. A comparison of the threshold concentrations needed to produce an effect in our experiments indicates that this is also the case in the fish gallbladder; sulphation of CCK8 decreases the amount needed to produce an effect at least a thousand fold. In the coho salmon, Vigna and Gorbman (1977) concluded thai, since sulphated gastrin 17 and sulphated CCK8 were equipotent in contracting the gallbladder, sulphation was important, while the position of the sulphate group (6 or 7 from the Cterminus) was of no significance. However, in the rainbow trout, sulphated CCK8 and caerulein were more potent than sulphated gastrin 17. This indicates that the position of the sulphated tyrosine is important in our preparation. The discrepancy with the results in the coho salmon may imply a species difference. The existence o f a VIP-like peptide in the fish gut is firmly established (Langer el al. 1979; Holmgren et al. 1982; Holmgren and Nilsson 1983; Dimaline el al. 1986). Porcine VIP has an inhibitory effect on the swimbladder of the cod, G a d u s m o r h u a
(Lundin and Holmgren 1984), while the effects on the stomach of the rainbow trout are inconsistent (Holmgren 1983). In the present study, a dense meshwork of varicose nerve fibres was found in both gallbladder and bile duct, and exogenous VIP was a potent relaxatory agent on strip preparations of the gallbladder, indicating the presence of a functioning inhibitory innervation by VIP-releasing fibres in the rainbow trout gallbladder. Considering that the sphincter of Oddi (SO) is relaxed by CCK in mammals (Lin 1975) a highly speculative model for the control of the gallbladder emptying in rainbow trout would include the following elements: CCK secreted from endocrine cells in the duodenal mucosa on entrance of food into the duodenum, affects both the gallbladder and the SO. The gallbladder is contracted and the SO is relaxed, resulting in emptying of bile into the duodenum. An inhibitory tonus of adrenergic a n d / o r V1Pnerves reduces spontaneous contractile activity between food digestions. The sparse serotonergic innervation close to the gallbladder mucosa could be involved in the bile processing function, while circulating serotonin may affect the motility of the gallbladder.
Acknowledgements The authors wish to thank Mrs. I. Holmqvist and Mrs. B. Vallander for skilled help with the illustrations, and Professor S. Nilsson for critical reading of the manuscript. We also thank Professor 'r Mutt, Stockholm, for kind gifts of.porcine V1P and Professor G. Dockray, Liverpool, for kind gifts of antisera. This study was supported by grants from the Swedish Natural Science Research Council (NFR), the Magn. Bergvall Foundation, and the Jeanson Foundation. Part of this study was presented at the Bayliss and Starling Society 1986.
References cited Andersson, C. 1983. Evidence for 5-HT-containing intrinsic neurons in the teleost intestine. Cell Tiss. Res. 230: 377-386.
154
Arien,,, t_'..I, and Simonis, A.M. 1961. Analysis of tile actions of drugs and drug combinations, ht Quantative Methods in Pharmacology. pp. 286-31 I, Edited by H. de `longe. NorthHolland, A m s t e r d a m . Baumgarten, H.G. 1967. V o r k o m m e n lind Verteilung adrenerger Ncrvenfasern im Darm der Scheilc. Z.Zellforsch 76: 248-259. Behar. J and Biancani, P. 1980. Effect of cholecystokinin and the octapeptide of cholecystokinin on the feline sphincter of Oddi and gallbladder. `l. Clin. In',est. 66: 1231-1239. Berry, H. and FIo',,,crs, R.`l. 1971. The assay of endogenous cholec~,,;tokinin and factors influencing its release in tile dog ztnd cat. (}astrocntcrology 60: 409-42(I. Bishop, A.(_., Polak, J.M., Bloom, S.R. and Pearse, A.G.E. 1978. A new universal technique.for the inlmunocytochentical localization of peptidcrgic innervation. `l. Endocrinol. 77: 2~ -26P. ('ai, w - o and Gabetla, G. 1983. lnnervatic, n of the gallbladder and the biliar.,, pathv, ays m the guinea pig. `l. Anat. 136: 9 7 - 109. Coons, A.C. 1956. Histochemistry with labelled antibody. Int. Re,,. Cytol. 5: 1-23. Costa, M., Buffa, R., Fur'ncss, .l.B. and Solcia, E. 1980. Immunohistochemical localization of polypeptides in peripheral aulononlic nerves using v,holemotmt preparations. Histochemistry 65: 157- 169. ( r i m , J.\V. and Vigna S.R. 1983. Brain, gut and skin peptide hormones in lower vertebrates. A m . Zooi. 23:621 638. Dimaline, R., Thorndykc, M.C. and Young, .l. 1986. Isolation and partial sequence of elasmobranch VIP. Regul. Pept. 14: I-I0. Doggrell, S.A. and Scott, G.W. 1980. Tile occurrence of posts.vnaptic alpha and beta adrcnoceptors in the guinea pig gallbladder. Br. J. Pharmacol. 71: 185-189. Falck. B and O w m a n , Ch. 1965. A detailed methodological de,,cri'plion of the fluorcscense method for the cellular demonstration of biogenic m o n o a m i n e s . Acta Univ. Lundensis 7: 1-23. Fangc, R. and `lohnels, A.G. 1958. An a u t o n o m i c nerve plexus control of the gallbladder in M3:vine. Acta Zool. 39: 1-8. Holmgren, S. 1983. Tile effects of putative non-adrencrgic noncolinergic autonomic transmitters on isolated strips from tile stomach of tile rainbows, trout, Salmo gairdneri. C o m p . Biochem. F'hysiol. 74(': 229-238. Hohngren, S. and Fange, R. 1981. Effects of cholinergic drugs on the intestinc and gallbladder of the hagfish, M v.~ine ghttitto~a [.. wilh a report on the inconsistent effects of catecholamines. Mar. Biol. Left. 2: 265-277. Holmgrcn, S., Vaillant, C. and I)imalinc, R. 1982. VIP-, subqance P-, gastrin~(_'CK-, bombcsin-, sonlatostatin- and gh, cagon-like immunoreacti,,ities in tile gut of the rainbov, Irout, Salmogairdneri. CelITiss. Res. 223:141 153. Hoh'ngrcn. S. and Nilsson, S. 1983. Bombesin-, gastrin/CCK5-hydroxytryptamine-, nct, rolensin-, somatostatin-, and VIPlike inmmnoreactivit3 and catccholanlinc fluorescence in the gut of the elasmobranch, .Squtthts acunthius. Cell Tiss. Res. 234:595 618.
Holmgren, S., Grove, D.J. and Nils,on, S. 1985. Substance P acts by releasing 5-hydroxytryptamine from enteric neurons in the stomach of the rainbov, trout, Salmo gairdneri. Neurosciencc 14: 683-693. Jansson, R., Steen, G. and Svanvik, J. 1978. Effects of intravenous vasoactive intestinal peptide (VIP) on gallbladder function in the cat. Gastroenterology 75: 4 7 - 5 0 . Jcnsen, J. and Hotmgren, S. 1985. Neurotransmitters in the intestine of the atlantic cod, Gadus morhua. C o m p . Biocbem. Physiol. 82(': 8 1 - 8 9 . kanger, M., van Noorden, S., Polak, J.M. and Pearse, A.G.E. 1979. Peptide hormone-like immunoreactivity in the gastrointestinal tract and endocrine pancreas of eleven teleost species. Cell Tiss. Res. 199: 493-508. Larsson, k.l. and Rehfeld, J.F. 1977. Evidence for a c o m m o n cvolutionar,, origin of gastrin and cholccystokinin. Nature, f o n d . 269: 335-338. kin, T-M. 1975. Progress in gastroenterology. Actions o f gastrointestinal h o r m o n e s and related peptides on the motor function of the biliary tract. Gastroenterology 69: 10061022. Lundin, K. and Holmgren, S. 1984. Vasoactive intestinal polypeptide-likc immunoreactivity and effects of VIP in the sv,'imbladder of the cod, Gadus morhua. C o m p . Physiol. 154B: 627-633. Nakata, K. and Kurahashi, K. 1981. Effects of C-terminal octapeptide of cbolecyslokinin and proslaglandins on adrenergic functions in the guincapig gallbladder and sphincter of O d d i . . l a p . `l. Pharmacol. 31: 7 7 - 8 3 . Persson, C . G . A . 1972. Adrenoceptors in the gallbladder. Acta P h a r m . Tox. 31: 177-185. Sakharov, D.A. and Salimova, N.B. 1980. Serotonin neurons in the peripheral nervous system of the larval lamprey, Lampetra planeri. A histochemical, microspectrofluorimetric and ultrastructural study. Zool. J. Physiol. 84: 231-239. ,,an Rossum, .l.M. 1963. Cumulative dose-response curves. 11. Technique for the making o f dose-response curves in isolated organs and the e',aluation of drug parameters. Arch. Int. P h a r m a c o d y n . Ther. 143: 299-330. Vigna, S.R. and G o r b m a n , A. 1977. Effects o f cholecystokinin, gastrin and related peptides on coho salmon gallbladder contraction in vitro. A m . J. Physiol. 232: E485-E491. Vigna, S.R. 1979. Distinction betv,,een cholecystokinin-like and gastrin-like biological activities extracted from gastrointestinal tissues of some lower ',ertebrates. Gen. C o m p . Endocrinol. 39: 512-520. Vigna, S.R. 1983. Evolution of endocrine regulation of gastrointestinal function in lower vertebrates. A m . Zool. 23: 7 2 9 738. Vigna, S.R., Fisher, B.L., Morgan, J i . M . and Rosenquist, G.I.. 1985. Distribution and molecular heterogeneity of cholccystokinin-like immunoreactive peptides in the brain and gut of tile rainbow trout, Sahno gairdneri. C o m p . Biochem. Physiol. 82C: 143-146. Walsh, J.H. 1981. Gastrin. In Gut Hormones. pp. 163-170. Edited by S.R. Bloom and J.M. Polak. Churchill Livingstone, London.
155 Watson, A.H.D. 1979. Fluorescent histochcmistry of the teleost gut: E~idence for the presence of scrotoncrgic neurons. Cell Tiss. Res. 197:155 164. Yasutakc, W.T. and Wales, J.H. 1983. Microscopic anatomy of salmonides: An alias. Resourcc Publications, U.S. Departmcnt of the Interior, Fish and Wildlife Service. No 150.
Young, J.Z. 1980. Nervous control of stomach movement in dog-fish and rays..I. ,Mar. Biol. Ass. U.K. 60: 1-17. Young, J.Z. 1983. Control of movements of the stomach and spiral intestine of R~Ua and Scvliorllimt.~. J. Mar. Biol. Ass. U.K. 63: 557-574.
Control of gallbladder motility in the rainbow trout, Salmo gairdneri G6ran Aldman and Susanne Holmgren Comparative Neuroscience Group, Department of Zoophysiology, University of GOteborg, Box 250 59, S-400 31 G6teborg, Sweden Keywords: gallbladder, rainbow trout, neuropeptides, hormones, motility
Abstract
Immunohistochemistry revealed nerves containing VlP-like and 5-HT-like material in both gallbladder wall and bile duct of the rainbow trout, while endocrine cells containing gastrin/CCK-like and substance P-like material were present in the mucosa of the bile duct and the duodenum. Fluorescence histochemistry showed adrenergic nerves close to the muscle layer of the gallbladder. Sulphated CCK8, caerulein and non-sulphated CCK8 (in this order of potency), 5-HT and acetylcholine were excitatory on isolated strip preparations, while VIP and adrenergic agonists were inhibitory. The adrenergic drugs were probably acting via a beta-adrenergic receptor, while the effects of 5-HT and cholinergic drugs were antagonized by methysergide and atropine respectively. Electrical stimulation of the gallbladder nerves in situ failed to show any effect or under certain conditions induced a rebound effect. It is concluded that the motility control of the rainbow trout gallbladder may involve an inhibitory innervation by adrenergic and possibly VIP-releasing nerves, while 5-HT, acetylcholine and a CCK-like substance may be involved in the excitatory control.
Introduction
In the rainbow trout, Sahno gairdneri, the bile secreted from the liver is stored in a gallbladder. The bile duct leads bile both from the liver to the gallbladder and from the gallbladder to the proximal intestine (duodenum). A well developed smooth muscle layer in the gallbladder wall (Yasutake and Wales 1983) ensures emptying of bile into the intestine. Little is known about the nervous or hormonal control of the gallbladder motility in fish. In the cyclostome Myxine glutinosa vagal stimulation causes contraction of the gallbladder (F~inge 1958). Cholinergic receptors of the muscarinic type are
present in the gallb'ladder muscle, but cholinergi~ innervation is sparse judging from the measured activity of choline acetyltransferase (Holmgren and F/:inge 1981). The gallbladder of the coho salmon, Oncorhynchus kisutch, is contracted by porcine cholecystokinin (CCK) in vitro (Vigna and Gorbman 1977) while that of the spiny dogfish, Squalus acanthias, shows only weak responses if any to CCK (Vigna 1983). In the cyclostome Eptatretus stouti both CCK and the closely related peptide caerulein fail to produce measurable effects (Vigna 1979). In mammals (guinea pig, cat), tonus and motility of the gallbladder is regulated by the autonomic nervous system, which forms several plexuses in the
144 gallbladder wall. Both catecholamine containing and acetylcholinesterase containing fibres are present (Cai and Gabella 1983) Adrenergic nerves may cause contraction or inhibition via alpha and beta adrenoceptors respectively (Persson 1972; Doggrell and Scott 1980), while CCK8 and cholinergic nerves stimulate contraction (Hedner 1969; Berry and Flowers 1971; Behar and Bianchi 1980; Nakata and Kurhashi 1981). VIP-like immunoreactivity and substance P-like immunoreactivity are reported in the mammalian gallbladder (Cai and Gabella 1983), and VIP relaxes the gallbla.dder and induces fluid secretion into the lumen in the cat (Jansson et al. 1978). This study was carried out as an attempt to elucidate the nervous and hormonal control of the gallbladder motility in a teleost, the rainbow trout, Sahno gairdneri.
Materials and methods
Rainbow trout, Sahno gairdneri, ca 600 g and 100- 150g were bought from a local hatchery. They were kept in aerated, recirculating fresh water at 10~ and fed once a week until used. They were killed by concussion and opened ventrally and laterally for in situ studies of the gallbladder innervation or for removal of the gallbladder.
Light microscopy Tissues were fixed in phosphate buffered 4~ formaldehyde solution, dehydrated and embedded in paraffin. Seven ~m thick sections were stained with hematoxylin and eosin.
Falck-Hillarp fluorescence microscopy Preparations were treated according to the FalckHillarp technique (Falck and Owman 1965) for visualization of adrenergic nerves. Pieces of tissue were snap frozen in liquid propane cooled in liquid nitrogen, freeze-dryed for 3 - 5 days, treated with paraformaldehyde vapour at 80~ for one hour
and vacuum-embedded in paraffin. Sections (7 ~m thick) were mounted in Entellan (Merck) and viewed immediately in a Leitz Dialux fluorescence microscope with incident light illumination and a barrier filter at 460 nm.
Immunohistochemistry Whole-mounts: Preparations were made according to the technique described by Costa et al. (1980). Pieces of gallbladder were stretched out and pinned on flat sheets of dental wax. They were fixed in a solution of formaldehyde (2%) and picric acid (15~ in a phosphate buffer (0.1M, pH 7.2) for 18 h at 4~ After rinsing in 80~ alcohol the preparations were dehydrated, xylene-treated for 30 min and then rehydrated. Thin sheets were peeled off and incubated with primary antisera (see Table 1) in a moist chamber at room temperature for 20 h. After incubation the preparations were rinsed several times in phosphate-buffered saline (PBS, 0.1M, pH 7.2) and then incubated with a second antibody (swine-anti-rabbit IgG diluted I:10 (DAKO)), which was labelled with FITC (fluorescein isothiocyanate) (Coons 1956). After rinsing in PBS the preparations were mounted on slides in carbonate-buffered glycerol (1:1, pH 8.5). The slides were viewed in a Leitz Dialux fluorescence microscope with a barrier filter at 510 nm. Film used was Ilford FP4 or Kodak TriX. Controls were carried out for those antisera giving a positive immunoreaction. The primary antiserum was preincubated with its hapten, i.e. the antiserum raised against VIP was preincubated with porcine VIP (10 nmol/ml), the antiserum raised against 5-HT was preincubated with 5-HT (10 nmol/ml), the antiserum raised against gastrin (G4) was preincubated with gastrin (G17, 10 nmol/ml) and the antiserum raised against substance P was preincubated with substance P (10 nmol/ml) at 4~ for three days. All preabsorbed antisera failed to stain any nerve fibres or cells on subsequent use. Sections: Preparations were made according to the technique described by Bishop et al. (1978). Pieces of gallbladder and bile duct were fixed in parabenzoquinone (4 mg/ml) in sodium cacodylate buffer
145 l-ahle I. Antisera used in immunohistochemical experiments
Antisera raised against
(._'ode
Hapten
Bombesin Bombesin
L89 k90
Gastrin
L112
Synthetic bombesin G4
Mct-enkephalin
L 146
Neurotensin
Dilution
Specificity
1:100 1:50 C-terminus
Mel-enkephalin
1:50
Free C-terminus
NTab
Neurotensin
1:50
Subslance P
CRBAB30
Synthetic SP 11
1:100
C-terminal specific in RIA C-terminus
VlP
L85
1:400
C-terminus
5-HI
71954
Synthetic 18-28 VIP 5-HT serum albumin
Strip preparation
experiments
Gallbladders were dissected out and opened longitudinally from one end to the other. The bile duct was cut off and the bladder was cut in half by a second cut at 90~ angle to the initial longitudinal section, producing two equal preparations, in which it was possible to measure the tension of circularly arranged muscle fibres. The preparations were mounted with an initial tension of 1 milliNewton (mN) in organ baths containing rainbow trout Ringer's solution (NaCI 7.41, KCI 0.36, C a C k 0.17, NaCO 3 0.31, N a z H P O 4 • 2 H 2 0 2.0, Na HzPO 4 0.40, MgSO 4 0.15 and glucose 1.0 g/l; Holmgren 1983 modified from Lockwood 1961) at 10~ bubbled with a gas mixture of O~ (97%) and CO~ (3%) and left for up to 18 h before the experiment.
1:50
Source Dockray Dockray
I: 100
(0.1M, pH 7.3) at 4~ for 4 h. After rinsing in sodium cacodylate buffer containing 20% sucrose for 24 h at 4~ the tissue pieces were snap-frozen in liquid nitrogen and cut in sections of 10 um on a cryostat. The sections were then placed on glass slides covered by a film of chrome-alum gelatine. Incubation with antisera was made as for whole mounts.
Reference
Dockray el al. 1981 Bulochet al. 1983 Dockray, pers. comm.
Dimaline et al. 1980
Dockray Dockray Dockray Cambridge Res. Biochemicals Dockray Immunonuclear
The responses to cumulative concentrations of agonists were studied. Tensions of the preparations were recorded via Grass FT03 isometric transducers on a Grass Polygraph rood 7. Drugs used were acetylcholine chloride (ACh), L-adrenaline bitartrate (A), atropine sulphate (Atr), caerulein (Cae) (Peninsula), carbachol (CCh), cholecystokinin octapeptide (CCK8), sulphated cholecystokinin octapeptide (CCK8S) (Research plus), gastrin 17 (GI7), sulphated gastrin (G17S) (Peninsula), 5-hydroxytryptamine creatinine sulphate (5-HT), Dkisoproterenol HCI (Iso), methysergide (1-methyllysergic acid butanolamine bimaleate) (Met) (Sandoz), pentagastrin (PG) (Peninsula), phentolaminr hydrocloride (Phent), DL-propranlol HCI (Propr), tetrodotoxin (TTX), vasoactive intestinal polypeptide (VIP) (a kind gift from Prof. V. Mutt, Stockholm). Drugs were purchased from Sigma if not otherwise stated. Dose ranges tested were: CCK8S and caerulein 10 -14 - 10-TM; CCK8, GI7 and GI7S 10 -I1 10-VM; PG 10 -l~ - 10-6M; A 10 -9 - 10-TM; NA 10 -9 - 10-SM; Iso 10 -12 - 10-6M; ACh and CCh 10 -8 - 10-4M.
146
Calculations
g.c i ~ Response was calculated as the work performed (surface area on tracing) p~r minute, when a maximal effect of an addition of agonist was achieved. This was compaired to the total maximal effect obtained (100%), and concentrations-response curves were plotted. Mean individual sensitivity (Ari6ns and Simonis 1961) was calculated, and pD 2 (-log ECs0) -values (van Rossum 1964) were determined.
Q.c.-m/~)
liver
Resulls
Extrinsic innervation of the gallbladder Dissection, low power magnification microscopy and osmium staining, showed that small branches of the splancnic nerves running along small mesentric vessels reached the surface of the gallbladder (Fig. 1). No direct connection from the vagal branches could be observed, but it must be emphasized that this region of the gut is difficult to dissect due to the numerous small vessels with accomparting nerves running in different directions in the mesenterium, and the amounts of fat stored in this area. Stimulation in situ of the gastrointestinal vagal branc[aes and of the splancnic nerves running along the coeliac and mesenteric arteries with 4 - 20 Hz, 0.5 - 2 m s a n d 5 - 2 0 V f o r 5 - 20 sec (n = 5) did not produce any macroscopically detectable contractions of the gallbladder, while occasional spontaneous contractions were readily visible. However, if the gallbladder was detached from the liver and the mesenterium shared with the proximal part of the coeliac artery, rebound contractions of the gallbladder were clearly visible 2 - 5 sec after stimulation (10 Hz, 1 ms, 10 V, 5 - 2 0 sec) had ended (n = 5). Stimulation in vitro of whole intact gallbladders, whole longitudinally opened gallbladders or strip preparations produced weak and inconsistent results.
Fig. 1. Highly schematic drawing of the extrinsic innervation of the gallbladder of the rainbow trout, Salmo gairdneri. Branches of the splanchnic nerve (n.sp[.) reach the gallbladder along small vessel branches from the mesenteric artery. Nerves were also observed along the small vessels passing between the liver and the gallbladder. It is possible that nerves also enter the gallbladder from the region of connective tissue, vessels, nerves and fat indicated by the hatched area at the neck of the gallbladder. Legend: a.c.-m. - arteria coeliaco-mesenterica, g.c. - ganglion coeliacure, n.spl. - nervus splanchnicus.
Histochemistry The gallbladder wall consists of an inner mucosa (lamina epithelialis) attached to a basal m e m b r a n e (lamina propria), outside this the submucosa (strata c o m p a c t u m and granulosum), a layer of smooth muscle-cells (muscularis) and the serosa (Yatsutake and Wales 1983; Fig. 2). In both the gallbladder and bile duct walls, the Falck-Hillarp technique reveals a large number of varicose nerve fibres in the outer part of the smooth muscle layer showing the blue/ green fluorescence specific for catecholamines (adrenaline/noradrenaline). In the submucosa, especially close to the basal m e m b r a n e , nerve fibres and occasional cells were observed showing the yellow fluorescence characteristic of 5-hydroxytryptamine (5-HT; Fig. 3).
148
hnmunohistochemistry The presence of a 5-HT-lil~e substance in nerves of the gallbladder was further confirmed with immunohistochemistry. The 5-HT-like immunoreactivity wasweak, but a loose plexus of nerve fibres could be observed in the gallbladder wall (Fig. 4). Within this plexus cell bodies of a bipolar appearance were noticed (Fig. 5), while occasional star shaped cells found outside the plane of focus of the nerve plexus (Fig. 4) did not seem to be in direct contact with the plexus. Varicose nerve fibres showing VIP-like immunoreactivity formed .a well developed dense plexus throughout the muscle layer with an increasing density towards the inner parts in both the gallbladder (Fig. 6) and the bile duct (Fig. 8) wall. The fibres were seen concentrated in ganglia in the nexuses of the plexus (Fig. 7). Populations of endocrine cells showing substance P-like immunoreactivity (Figs. 9 and 10) and gastrin/CCK-like immunoreactivity (Figs. 11 and 12) were present in the mucosa of the bile duct and in the mucosa o f the duodenum in the area of the bile duct entrance. Antibodies raised against bombesin, metenkephalin and neurotensin failed to reveal specifically immunoreactive material.
Strip preparations About 40% of the preparations showed spontaneous rhythmic activity within 18 h. This activity was not blocked by tetrodotoxin (TTX, 10-6M, n = 5; Fig. 13), atropine (10-6M, n = 14) or the alpha adrenergic antogonist phentolamine (10-6M, n = 5), but was often enhanced by the beta adrenergic antagonist propranolol (10-6M, n = 14, Fig. 13). This indicates the presence of a beta adrenergic tonus, but treatment of inactive preparations with tetrodotoxin (10-6M, n = 6) to antagonize a nerve mediated inhibition failed to induce rhythmic activity of the preparations. Agonist drugs could affect the preparations in three different ways: Change of tonus, or change in frequency or force of contraction. Usually the ef-
fect was a combination of the three. Cumulative concentration-response tests were performed, Excitatory, concentration dependent effects on the preparations were obtained with cholinergic drugs, 5-HT and CCK/gastrin related peptides, while VIP and adrenergic drugs were inhibitory. Adrenaline (10-6M, n = 5 ) abolished spontaneous activity (Fig. 13) and activity induced by 5-HT (10-6M, n = 5 ) . Pretreatment with propranolol (10-6M) inhibited this effect of adrenaline ( n = 6 ; Fig. 13). Adrenaline (PD2=7.70-+0.15, n =4) and isoprenaline (PD2=7.45 _+ 10.35, n = 4) were significantly more potent (p > 0.01) than noradrenaline (pD 2 = 6.24_+ 0.28, n = 4) to inhibit the spontaneous activity of the preparations (Fig. 14). VIP had an inhibitory effect on spontaneously active preparations, even in low concentrations (10-1~ _ 10-7M, n = 4 ; Fig. 13). Cholinergic drugs were excitatory on the gallbladder strips. Carbachol (pD 2 = 6.01, n = 5; Figs. 15, 16) was more potent than acetylcholine, indicating the presence of a cholinesterase activity in the preparations. Contractions caused by carbachol (10-6M) were reduced or abolished by atropine (10-6M, n = 5 , Fig. 15), indicating an effect via muscarinic receptors in the preparation. 5-HT ( P D 2 = 5 . 8 5 + 0 . 1 9 , n = 5 ) produced an excitatory effect in the same concentration range as carbachol (Figs. 16, 17). The effect of 5-HT (10-6M) was inhibited by methysergide (10-6M, n = 5 ; Fig. 17), indicating a specific receptor-mediated effect of 5-HT. Of the CCK/gastrin related peptides tested, sulphated CCK 8 (pD2= 10.69_+0.28, n = 7 ) was the most potent, followed by caerulein (PD2=8.83_+ 0.23, n = 6), while sulphated or non-sulphated gastrin 17 and nonsulphated CCK 8 produced weaker responses in a much higher concentration range (Fig. 19).
Discussion
The existence of an autonomic innervation of the gallbladder of the rainbow trout is indicated by the
150 0
TETRODOTOXIN
5min % RESPONSE
mN
I
0
"
~
20
10-"-,
'
-6
b
2
ADRENALINE
A ~~ISO
0
-85 -8 C
PROPR
2
A
-75 -7 10"3M
50
~l, qll
0 A-6
d
i
Vlp
80
-II
-[0
-7 LOG MOLAR CONC
log molor conc
-6
Fig. 13. Circular smooth muscle preparations of the gallbladder o f Salmt~ gairdneri, a. Tetrodotoxin has no effect on spontane-
Fig. 14. Concentration response curves showing thc mean in-
ous r h y t h m i c activity, b. Adrenaline is inhibitor,,' on the spontaneous activity, c. The 13-adrenergic antagonist propranolol enh a n c e s the spontaneous activity and blocks the inhibitory effect o f adrenergic drugs, d. V I P is inhibitor:,, on the spontaneous activity.
range 20-80~ o f total abolishment of activity, calculated as w o r k performed per unit time • S.E. A - a d r e n a l i n e , n 4; ISO - isoprenaline, n = 4 ; N A - n o r a d r e n a l i n e , n = 4 ; **, p > 0.01, students t-test.
d i v i d u a l sensitivity of the preparations to adrenergic drugs in the
CARBACHOL
anatomical and histochemical results in this study. The Falck-Hillarp technique shows the presence of an ad,renergic innervation, and also indicates the presence of a serotonergic innervation, although not of the same extent. The immunohistochemical results further confirm the presence of a low number of nerves containing a serotonin-like substance, and in addition indicate the presence o f nerves containing a VIP-like peptide. The antisera used in the immunohistochemical experiments were raised against mammalian or amphibian peptides, which in most cases differ somewhat in the amino acid sequence from that o f their counterparts occurring in teleost fish. A positive reaction between the mammalian antibody and the teleost peptide, is still often achieved, but this positive reaction may only indicate the presence o f a peptide similar and not necessarily identical to the original hapten. An antibody raised to a mammalian peptide may, on the other hand, fail to react
i ]
mN o
411
~ ~'"'~" ,,~L,~,.~"~'2;.,.2",',~
,.,. , ~'.:::1::~--" ' "
log molor ATR 10"6M 9 I J
75 ~
mN
I ',~ll.
qh~N, 0
hlllll
'
~'
.
.
.
.
.
conc
._JL_~L__ .
log molor conc Fig. 15. The effect of cumulative concentrations o f c a r b a c h o l ( C C h ) on gallbladder smooth muscle (upper tracing). T h e contractile effect produced by C C h is abolished in the presence o f
atropine (ATR; lower tracing).
with the teleost variant of the same peptide, and a negative result in our study must therefore not be taken as evidence for the absence of a certain peptide. The presence o f muscarinic cholinergic receptors mediating an excitation o f the gallbladder smooth
151 % RESPONSE
% RESPONSE
80.
80
CClT 50
CCK8 50
721-~
20
5
l/
00l
_--
20
-&
-,~
-4
-,~
LOG MOLAR CONC
,I.
-,i
-,'o
-9
-8
LOG MOLAR CONe
Fig. 16. Concentration response cur',es showing the mean individual sensitivity ( • S.E.) of gallbladder smoolh muscle preparation,, to carbachol (CCh, n = 5) and 5-hydroxytryptamine (5-HT, n = 5) in the range 2 0 - 8 0 % of maximal effecl, calculated as work performed per unit time.
Fi~. 18. Concentration response curves shov,ing the mean individual sensitivity (_+ S.E.) of the gallbladder smooth muscle preparations to sulphated cholec.vstokinin 8 (CCK8S) and caerulein (CAE) in the range 20-80~ of maximal effect, calculated as work performed per unit time. CCK8S is significantly 1"** = P < 0.00l) more potent than caerulein, students t-test.
5 ,'11in
5-HT mN I 0
4
-?
-&
-g
METH 10"6M
I0
9
o~
m'~, -6
;~_d~.~,~._~,G,jt
-
-6 log molar conc
Fi,e. 17. The effect of cumulative concentrations of 5-hydroxytryptamine (5-HT) on the gallbladder smooth muscle (upper tracingL The stimulator v effect produced by 5-HT is reduced in the presence of the serotonergic antagonist methysergide (METH; lower tracing).
muscle is shown in this study, but similar to the situation in the cyclostome Myxine glutinosa (Holmgren and F~inge 1981) it is doubtful whether the gallbladder has an extrinsic cholinergic innerva-
tion. The presence of a functioning adrenergic innervation, excerting an inhibitory action via /3adrenoceptors, seems more unequivocal. It is also likely that this adrenergic innervation normally exerts an inhibitory tonus on the preparations, since treatment with the f3-adrenergic antagonist propranolol usually caused the onset of or increase in rhythmic activity. A profound effect of 5-HT on'smooth muscle from the fish gut has earlier been shown in Salmo gairdneri as well as in other teleost and elasmobranch species Young 1980, 1983; Holmgren 1983; Holmgren and Nilsson 1983; Jensen and Holmgren 1985). This effect was in most cases demonstrated as a clearly excitatory action, in accordance with the results in the present study. Serotonergic nerves have also earlier been demonstrated by histochemical methods in the gut of both cyclostomes, elasmobranchs and teleosts (Baumgarten 1967; Watson 1979; Sakharov and Salimova 1980; Anderson 1983; Holmgren and Nilsson 1983;
152
Cae 2 mN
5mm
0
-,b
-9
-8
5 log molar conc
CCK8S
2
CCK8
2
mN 0
0
-,2
-tl
-,o
-;
-8
" -
-7 lOg molar cone
2
GI7S
2
GI7
mN 0
-G
-8
5
-8
-7 log molar conc
Fig. 19. The effects of cumulative concentrations of different CCK/gastrin-related peptides on gallbladder smooth muscle. The responses to caerulein and sulphated CCK8 reached a maximum within the concentration range tested, while the effects of non-sulphated CCKg and sulphated and non-sulphated gastrin 17 occurred in a much higher concentration range and never reached a maximum.
Jensen and Holmgren 1985; Holmgren et al. 1985). In the present study of S a l m o gairdneri 5-HT (or a closely related substance) is present in nerves of the gallbladder wall according to both immunohistochemical and fluorescence histochemical results. However, the innervation seems rather sparse and is situated closer to the mucosa than to the muscle layer, and it is possible that at least part of the excitatory effect that can be induced by exogenous 5-HT may in physiological conditions be caused by circulating 5-HT released from endocrine cells or mast cells in the gallbladder or elsewhere. The starshaped 5-HT-containing cells in the gallbladder wall, probably without contact with the 5-HT containing nerves, may be involved in such an endocrine/paracrine function. Previous studies in Sahno gairdneri have shown the presence of such cells in the gastrointestinal canal (Holmgren et al. 1985). Gastrin, cholecystokinin (CCK) and caerulein
share the C-terminal pentapeptide sequence. The antibodies used to localize these substance were raised towards the c o m m o n C-terminal tetrapeptide, hence it is not possible to use this antibody to differentiate between these substances. No gastrin/CCK-like immunoreactive material was, however, encountered in nerves of the gallbladder, and g a s t r i n / C C K affecting the gallbladder motility may originate from the endocrine cells of the gut and bile duct mucosa reported in the present study and previously (Holmgren et al. 1982). The identity of gastrin/CCK-like peptides and their point of evolution has been much discussed (Walsh 1981; Crim and Vigna 1983). An evolution at the amphibian-reptile level of gastrins from a CCK-like peptide, may be a peptide closely resembling caerulein, has been suggested (Larsson and Rehfeld 1977). Other authors have instead even questioned a closer relationship between caerulein and gastrin/CCKs (Walsh 1981), although
153 the shared C-terminal pentapeptide is considered to constitute the biologically active center of these peptides. A recent radioimmunological study now indicates the presence of several gastrin/CCK/ caerulein-like peptides in gut extracts from the rainbow trout; regional differences exist and it is suggested that the occurrence of gastrin/CCK-like variants is even more extensive than in mammals (Vigna 1985). Little conclusive evidence of the true identity of the gastrin/CCK-like peptides present in the fish gut is thus available, and consequently little is known about the exact function of these peptides. Although, in mammals, the effects of gastrins and CCKs are overlapping, it is considered that CCK is the predominant regulator of the gallbladder motility, while gastrins have their major effects on e.g. gastric secretions. In the present study, CCK is also more potent than gastrin in initiating rhythmic activity in the gallbladder smooth muscles of the rainbow trout. The presence of a sulphate group (in position 7 from the C-terminus in CCK and in position 6 in gastrin) has been found important for the biological activity of CCK and gastrin in mammals. A comparison of the threshold concentrations needed to produce an effect in our experiments indicates that this is also the case in the fish gallbladder; sulphation of CCK8 decreases the amount needed to produce an effect at least a thousand fold. In the coho salmon, Vigna and Gorbman (1977) concluded thai, since sulphated gastrin 17 and sulphated CCK8 were equipotent in contracting the gallbladder, sulphation was important, while the position of the sulphate group (6 or 7 from the Cterminus) was of no significance. However, in the rainbow trout, sulphated CCK8 and caerulein were more potent than sulphated gastrin 17. This indicates that the position of the sulphated tyrosine is important in our preparation. The discrepancy with the results in the coho salmon may imply a species difference. The existence o f a VIP-like peptide in the fish gut is firmly established (Langer el al. 1979; Holmgren et al. 1982; Holmgren and Nilsson 1983; Dimaline el al. 1986). Porcine VIP has an inhibitory effect on the swimbladder of the cod, G a d u s m o r h u a
(Lundin and Holmgren 1984), while the effects on the stomach of the rainbow trout are inconsistent (Holmgren 1983). In the present study, a dense meshwork of varicose nerve fibres was found in both gallbladder and bile duct, and exogenous VIP was a potent relaxatory agent on strip preparations of the gallbladder, indicating the presence of a functioning inhibitory innervation by VIP-releasing fibres in the rainbow trout gallbladder. Considering that the sphincter of Oddi (SO) is relaxed by CCK in mammals (Lin 1975) a highly speculative model for the control of the gallbladder emptying in rainbow trout would include the following elements: CCK secreted from endocrine cells in the duodenal mucosa on entrance of food into the duodenum, affects both the gallbladder and the SO. The gallbladder is contracted and the SO is relaxed, resulting in emptying of bile into the duodenum. An inhibitory tonus of adrenergic a n d / o r V1Pnerves reduces spontaneous contractile activity between food digestions. The sparse serotonergic innervation close to the gallbladder mucosa could be involved in the bile processing function, while circulating serotonin may affect the motility of the gallbladder.
Acknowledgements The authors wish to thank Mrs. I. Holmqvist and Mrs. B. Vallander for skilled help with the illustrations, and Professor S. Nilsson for critical reading of the manuscript. We also thank Professor 'r Mutt, Stockholm, for kind gifts of.porcine V1P and Professor G. Dockray, Liverpool, for kind gifts of antisera. This study was supported by grants from the Swedish Natural Science Research Council (NFR), the Magn. Bergvall Foundation, and the Jeanson Foundation. Part of this study was presented at the Bayliss and Starling Society 1986.
References cited Andersson, C. 1983. Evidence for 5-HT-containing intrinsic neurons in the teleost intestine. Cell Tiss. Res. 230: 377-386.
154
Arien,,, t_'..I, and Simonis, A.M. 1961. Analysis of tile actions of drugs and drug combinations, ht Quantative Methods in Pharmacology. pp. 286-31 I, Edited by H. de `longe. NorthHolland, A m s t e r d a m . Baumgarten, H.G. 1967. V o r k o m m e n lind Verteilung adrenerger Ncrvenfasern im Darm der Scheilc. Z.Zellforsch 76: 248-259. Behar. J and Biancani, P. 1980. Effect of cholecystokinin and the octapeptide of cholecystokinin on the feline sphincter of Oddi and gallbladder. `l. Clin. In',est. 66: 1231-1239. Berry, H. and FIo',,,crs, R.`l. 1971. The assay of endogenous cholec~,,;tokinin and factors influencing its release in tile dog ztnd cat. (}astrocntcrology 60: 409-42(I. Bishop, A.(_., Polak, J.M., Bloom, S.R. and Pearse, A.G.E. 1978. A new universal technique.for the inlmunocytochentical localization of peptidcrgic innervation. `l. Endocrinol. 77: 2~ -26P. ('ai, w - o and Gabetla, G. 1983. lnnervatic, n of the gallbladder and the biliar.,, pathv, ays m the guinea pig. `l. Anat. 136: 9 7 - 109. Coons, A.C. 1956. Histochemistry with labelled antibody. Int. Re,,. Cytol. 5: 1-23. Costa, M., Buffa, R., Fur'ncss, .l.B. and Solcia, E. 1980. Immunohistochemical localization of polypeptides in peripheral aulononlic nerves using v,holemotmt preparations. Histochemistry 65: 157- 169. ( r i m , J.\V. and Vigna S.R. 1983. Brain, gut and skin peptide hormones in lower vertebrates. A m . Zooi. 23:621 638. Dimaline, R., Thorndykc, M.C. and Young, .l. 1986. Isolation and partial sequence of elasmobranch VIP. Regul. Pept. 14: I-I0. Doggrell, S.A. and Scott, G.W. 1980. Tile occurrence of posts.vnaptic alpha and beta adrcnoceptors in the guinea pig gallbladder. Br. J. Pharmacol. 71: 185-189. Falck. B and O w m a n , Ch. 1965. A detailed methodological de,,cri'plion of the fluorcscense method for the cellular demonstration of biogenic m o n o a m i n e s . Acta Univ. Lundensis 7: 1-23. Fangc, R. and `lohnels, A.G. 1958. An a u t o n o m i c nerve plexus control of the gallbladder in M3:vine. Acta Zool. 39: 1-8. Holmgren, S. 1983. Tile effects of putative non-adrencrgic noncolinergic autonomic transmitters on isolated strips from tile stomach of tile rainbows, trout, Salmo gairdneri. C o m p . Biochem. F'hysiol. 74(': 229-238. Hohngren, S. and Fange, R. 1981. Effects of cholinergic drugs on the intestinc and gallbladder of the hagfish, M v.~ine ghttitto~a [.. wilh a report on the inconsistent effects of catecholamines. Mar. Biol. Left. 2: 265-277. Holmgrcn, S., Vaillant, C. and I)imalinc, R. 1982. VIP-, subqance P-, gastrin~(_'CK-, bombcsin-, sonlatostatin- and gh, cagon-like immunoreacti,,ities in tile gut of the rainbov, Irout, Salmogairdneri. CelITiss. Res. 223:141 153. Hoh'ngrcn. S. and Nilsson, S. 1983. Bombesin-, gastrin/CCK5-hydroxytryptamine-, nct, rolensin-, somatostatin-, and VIPlike inmmnoreactivit3 and catccholanlinc fluorescence in the gut of the elasmobranch, .Squtthts acunthius. Cell Tiss. Res. 234:595 618.
Holmgren, S., Grove, D.J. and Nils,on, S. 1985. Substance P acts by releasing 5-hydroxytryptamine from enteric neurons in the stomach of the rainbov, trout, Salmo gairdneri. Neurosciencc 14: 683-693. Jansson, R., Steen, G. and Svanvik, J. 1978. Effects of intravenous vasoactive intestinal peptide (VIP) on gallbladder function in the cat. Gastroenterology 75: 4 7 - 5 0 . Jcnsen, J. and Hotmgren, S. 1985. Neurotransmitters in the intestine of the atlantic cod, Gadus morhua. C o m p . Biocbem. Physiol. 82(': 8 1 - 8 9 . kanger, M., van Noorden, S., Polak, J.M. and Pearse, A.G.E. 1979. Peptide hormone-like immunoreactivity in the gastrointestinal tract and endocrine pancreas of eleven teleost species. Cell Tiss. Res. 199: 493-508. Larsson, k.l. and Rehfeld, J.F. 1977. Evidence for a c o m m o n cvolutionar,, origin of gastrin and cholccystokinin. Nature, f o n d . 269: 335-338. kin, T-M. 1975. Progress in gastroenterology. Actions o f gastrointestinal h o r m o n e s and related peptides on the motor function of the biliary tract. Gastroenterology 69: 10061022. Lundin, K. and Holmgren, S. 1984. Vasoactive intestinal polypeptide-likc immunoreactivity and effects of VIP in the sv,'imbladder of the cod, Gadus morhua. C o m p . Physiol. 154B: 627-633. Nakata, K. and Kurahashi, K. 1981. Effects of C-terminal octapeptide of cbolecyslokinin and proslaglandins on adrenergic functions in the guincapig gallbladder and sphincter of O d d i . . l a p . `l. Pharmacol. 31: 7 7 - 8 3 . Persson, C . G . A . 1972. Adrenoceptors in the gallbladder. Acta P h a r m . Tox. 31: 177-185. Sakharov, D.A. and Salimova, N.B. 1980. Serotonin neurons in the peripheral nervous system of the larval lamprey, Lampetra planeri. A histochemical, microspectrofluorimetric and ultrastructural study. Zool. J. Physiol. 84: 231-239. ,,an Rossum, .l.M. 1963. Cumulative dose-response curves. 11. Technique for the making o f dose-response curves in isolated organs and the e',aluation of drug parameters. Arch. Int. P h a r m a c o d y n . Ther. 143: 299-330. Vigna, S.R. and G o r b m a n , A. 1977. Effects o f cholecystokinin, gastrin and related peptides on coho salmon gallbladder contraction in vitro. A m . J. Physiol. 232: E485-E491. Vigna, S.R. 1979. Distinction betv,,een cholecystokinin-like and gastrin-like biological activities extracted from gastrointestinal tissues of some lower ',ertebrates. Gen. C o m p . Endocrinol. 39: 512-520. Vigna, S.R. 1983. Evolution of endocrine regulation of gastrointestinal function in lower vertebrates. A m . Zool. 23: 7 2 9 738. Vigna, S.R., Fisher, B.L., Morgan, J i . M . and Rosenquist, G.I.. 1985. Distribution and molecular heterogeneity of cholccystokinin-like immunoreactive peptides in the brain and gut of tile rainbow trout, Sahno gairdneri. C o m p . Biochem. Physiol. 82C: 143-146. Walsh, J.H. 1981. Gastrin. In Gut Hormones. pp. 163-170. Edited by S.R. Bloom and J.M. Polak. Churchill Livingstone, London.
155 Watson, A.H.D. 1979. Fluorescent histochcmistry of the teleost gut: E~idence for the presence of scrotoncrgic neurons. Cell Tiss. Res. 197:155 164. Yasutakc, W.T. and Wales, J.H. 1983. Microscopic anatomy of salmonides: An alias. Resourcc Publications, U.S. Departmcnt of the Interior, Fish and Wildlife Service. No 150.
Young, J.Z. 1980. Nervous control of stomach movement in dog-fish and rays..I. ,Mar. Biol. Ass. U.K. 60: 1-17. Young, J.Z. 1983. Control of movements of the stomach and spiral intestine of R~Ua and Scvliorllimt.~. J. Mar. Biol. Ass. U.K. 63: 557-574.
