Journal o f Chemical Ecology, Vol. 8, No. 8, 1982
O L F A C T O R Y SENSITIVITY TO G R O U P - S P E C I F I C S U B S T A N C E S IN A T L A N T I C S A L M O N (Salmo salar L.)
B R I T F I S K N E S ~ a n d K J E L L B. D ~ V I N G Institute of Zoophysiology University o f Oslo P.O. Box 1051 Blindern Oslo 3, Norway
(Received October 27, 1981; revised December 14, 1981) Abstract--The olfactory sensitivity of three groups of Atlantic salmon (Salmo salar L.) toward substances emanating from their own groups was studied. Thresholds were determined by electrophysiological recordings of the induced waves from the medial and lateral part of the olfactory bulb surface. The intestine contained more potent olfactory substances on a gram per liter basis than skin mucus, urine, or amino acids. Chemical fractions were obtained from a parallel study on the nature of the naturally occurring substances by Stabell et al. (1982). A retarded fraction from chromatography on a Sephadex G-25 column contained the most potent material. The most potent fractions of the intestinal content evoked responses mainly in the medial part of the olfactory bulb, whereas the lateral part responded to amino acids. The results suggest that all salmon smolts of the waterways contribute to an odor trail in the coastal currents, thus facilitating the odor-dependent migration of the mature salmon. Key Words--Salmo salar, olfactory sensitivity, olfactory bulb, induced waves, pheromones, migrational cues, homing.
INTRODUCTION W h e n a n a d r o m o u s s a l m o n i d fishes r e t u r n to f r e s h w a t e r t o s p a w n a f t e r s t a y i n g in the sea, t h e y will h o m e to t h e i r n a t i v e river. T h e sense o f s m e l l has b e e n s h o w n to be e s s e n t i a l f o r c o r r e c t h o m i n g in s a l m o n ( W i s b y a n d H a s l e r , 1954; G r o v e s et al., 1968; T 0 i l , 1975). T h i s i m p l i e s t h e p r e s e n c e o f specific s u b s t a n c e s in e a c h river. A c c o r d i n g to the " p h e r o m o n e " h y p o t h e s i s p r o p o s e d b y N o r d e n g (1971, ~Present address: NORAD, P.O. Box 2646, Dar-es-Salaam, Tanzania. 1083 009841331 / 82/0800-1083503.00/0 9 1982 Plenum Publishing Corporation
1084
FISKNESAND Dg"VING
1977), the salmon smolts migrating to sea release population specific substances, forming an odor trail that guides the mature salmon to their native river. This hypothesis is supported by the results from behavioral experiments which have shown that adult sea char (Salmo salvelinus L.) is attracted to water where smolts from their own population have been kept (Selset and Doving, 1980). Physiological recordings from single cells in the olfactory bulb of sea char have given evidence of a sensory basis for discrimination between populations of char (Doving et al., 1974). The mechanism that is used by the fish olfactory system to discriminate specific substances of its own population from others is unknown. Two possibilities can be considered: (1) the returning salmon has a higher olfactory sensitivity to substances originating from its own population; and (2) the olfactory sensitivity is the same towards all population-specific substances. In the latter case the specific effect on the behavior of the fish must rely on specific interaction in the central nervous system. The aim of the present study was twofold: first, to investigate the olfactory sensitivity in Atlantic salmon to substances from salmon smolts of different groups; and second, to approach, by electrophysiological means, the most interesting chemical fraction of the samples with respect to populationspecific substances. A short abstract of the results of this study has been published (Fisknes, 1979). METHODS AND MATERIALS
Donor and Experimental Fishes. The fishes ( Salmo salar L.) were raised at the Research Station for Salmonids, Sunndalsr Unit, Western Norway and were offspring of wild fish caught in three Norwegian rivers: Driva, Namsen, and Tafjord. The fishes were thus the first generation in captivity. They were hatched in the spring of 1976. From autumn 1976 they were marked (fin clipped and freeze-branded) and mixed with other fish. They smoltified in spring 1977 and from then on they were held in culture in the sea. One to two weeks before the experiment the test fishes were transferred by air to the aquarium facilities at the University of Oslo and kept in recirculating seawater at a temperature of I0-12~ Handling o f Experimental Fish. The experimental fishes were 2--3-yearold salmon weighing 690-2630 g. They were initially anesthetized with tricain-methane sulfonate (MS-222, Sandoz) at 1:5000 concentration and then immobilized with Alloferin (Roche) given intramuscularly. The dose of the active principle, toxiferin, injected was about 4 mg/kg body weight. The fish were wrapped in a wet sponge and placed in a holder in the experimental set-up. The gills were continuously perfused with seawater at 14-16~ through a tube inserted into the mouth, at a flow about 1.8 liter/min. The
SALMON POPULATIONS AND ODORS
1085
Basin water
L - s e r i n e 10-3M
olfactory nerves
FIG. 1. Recording traces from the medial (e) and lateral (o) part of the left olfactory bulb in salmon. Stimuli were water from a saltwater basin containing salmon and L-serine at 10-3 M. Time bar = 1 sec; vertical bar = lO0/.tV. dorsal part of the skull, together with the underlying fat and connective tissue, was carefully removed and the olfactory bulbs exposed. Recording. The recording sites were at the lateral and medial part of the dorsal surface of the left olfactory bulb, positioned as shown in Figure 1. At each site a differential recording system was used. The electrodes were of glass and filled with Hanks' Ringer for marine fishes (Wolf and Quimby, 1969). The electrodes were inserted into electrode holders containing sintered Ag/AgC1. Tip diameter of the electrodes ranged from 10 to 50 #m. A stainless-steel wire served as a c o m m o n indifferent electrode. It was connected to the fish in the connective tissue caudal to the opening made in the skull. The potentials were amplified with two differential preamplifiers (WPI, D A M - 5 A ) and fed through a bandpass filter, 1-50 Hz. The signals were displayed on a pen recorder (Hewlett-Packard, type 7402 A). Stimuli. The stimulus donors were hatched at the same time as the experimental fishes and some of them were siblings or half-siblings of the experimental fishes. Skin mucus, intestinal content, and urine were collected in May 1977 from two groups of smolts from each of the three rivers. Each group contained two to seven individuals that were siblings. The methods for collection of these samples and the chemical procedure, including fractioning, are described in detail by Stabell et al., (1982). Briefly, the original samples of skin mucus, intestinal content, or urine were centrifuged (5700 rpm, 15 min). The samples of skin mucus and intestinal content were treated with acetic acid to a final concentration of 5%. The precipitate after centrifugation (25,000g, 30 rain) was separated from the supernatant. For olfactory tests the precipitate was stirred in distilled water after washing in 5% H A C and then centrifuged. The supernatants of skin mucus and intestinal content were run in a Sephadex G-25 column and separated into fractions I - V I I I . Fractions I V - V I I I
1086
FISKNESAND D~VING
appeared after the salt volume and will be mentioned as retarded fractions. Aliquots for biological tests were taken out at all stages of the chemical procedure. The dry weight of all samples was determined. The samples were kept frozen at - 2 0 ~ Just prior to the experiment they were thawed and diluted with artificial sea water in concentration steps of 1:10. Controls. As control stimuli L-serine and L-glutamine (Sigma) were made up in a series of dilution steps just prior to use. In addition "basin water," a water sample from the basin in which the fishes were kept, was used as a control stimulus. Stimulation. The olfactory epithelium was continuously flushed with artificial sea water through a polyethylene tube inserted in the anterior nares. The waterflow was ca. 13 ml/min. The stimulus solution was introduced from a pipet via a manually operated three-way stopcock. Procedure. All samples were applied in series of increased concentration. Each concentration was applied twice with a 30-sec interval between each stimulation. At intervals of 30 sec or more, no interaction of the stimuli was evident from observation of the response amplitudes. The time interval between two different dilution series was a minimum of 60 sec. Threshold concentration was determined as the lowest concentration of the stimulus that elicited a response (see Belghaug and D Cving, 1977). The concentrations were expressed as gram per liter.
RESULTS When an olfactory stimulus was introduced into the olfactory pit, a change in the potentials recorded from the surface of the olfactory bulb was seen (Ottoson, 1959a,b). Under present recording conditions, the response was characterized by oscillatory waves with constant frequency. The amplitude of the "induced waves" increased with increasing concentration of the stimulus. The frequency was 8-11 Hz at the temperature conditions used in the present study (cf. Dr and Belghaug, 1977). Figure 1 shows the responses to two of the control stimuli, "basin water" and I.-serine, recorded from the medial and lateral part of the bulb. Thresholds. Samples from both skin mucus and intestinal content were highly stimulatory to the salmon olfactory organ. Responses were elicited both in the medial and the lateral part of the olfactory bulb. The threshold yalues (g/liter) showed, however, that there exists a difference in the stimulatory effectiveness between skin mucus and intestinal content, the intestinal content being 10-1000 times more potent than the skin mucus (Figure 2). This was the case both for the original sample and the supernatant. The urine samples were at best equipotent to the skin mucus. In the second step in the chemical fractionation procedure, samples of the
1087
SALMON P O P U L A T I O N S AND ODORS
25 ~
[]s kin
intestinal content (50)
mucus (57)
20 t-
"5
15
I...0
E
TO
Z
-5
-4
-3
-2
-1
0
No response
Log concentration (g/t)
FIG. 2. Distribution of thresholds to supernatants of intestinal content and skin mucus from salmon smolts. The numbers of observations are given in parenthesis.
skin mucus and intestinal content were divided into supernatant and precipitat e fractions. The stimulating efficiency of these fractions was compared with that of the respective crude material. It thus showed that the stimulating substances were not completely isolated in the supernatant, but also remained in the precipitate. When stimulating with successive fractions from the G-25 Sephadex gel filtration, a change occurred in the response between the medial and lateral parts of the bulb. The first three fractions (I-III), which precede the salt volume of the column, elicited a response with higher amplitude from the lateral part of the olfactory bulb than the medial part. The retarded fractions (IV-VIII) were more likely to elicit the highest response in the medial part of the bulb (Figure 3). This was evident in experiments with fractions of intestinal content, but not for fractions of the skin mucus. Substances in the retarded fractions seem to have qualitative and stimulatory properties different from those in the first, unretarded fractions. Fractions I-III, which elicited responses with highest amplitude in the lateral part of the bulb, showed an increasing stimulatory effect from I to III. The lowest threshold was found for fraction III and 1 • 1014 g/liter in four test fishes. Most effective were the first three of the retarded fractions (IV, V, and VI). The most potent of these was fraction V, with a threshold concentration observed down to 1 • 10-5 g/liter. The amino acids used in the experiment always elicited a response in the lateral part of the bulb, but did so only sometimes in the medial part. The
1088
FISKNES AND D~VING
{} ~-
2F
39 31
T
30
E
15
35
4/*
{1 13 D.
E
o
i I
i II
i
I
III
IV
V
I
I
I
VI
VII
VIII
F r a c t i o n no. FIG. 3. The ratio between the amplitude of the responses at the medial versus the lateral part of the s a l m o n olfactory bulb observed when stimulating with different fractions of the s u p e r n a t a n t of the intestinal content (see text). F o r each fraction the m e a n a n d one s t a n d a r d deviation of the ratio is indicated. The n u m b e r of observations is given above each bar.
threshold were found to 8 X 10 -6 M ( N = 8) for 4 X 10 -3 g/liter and 1 X higher than those of the content.
have a mean of 4 • 10-5 M (N = 8) for L-serine and L-glutamine. These concentrations correspond to 10 -3 g/liter, respectively, i.e. about 10 to 100 times most potent fractions of skin mucus and intestinal
Cross-Comparisons. One of the objectives of the present study was to seek evidence for group-specific sensitivity to natural substances. The experimental fishes were therefore exposed to stimuli from donor fishes of the three groups. Table 1 shows an example of thresholds in three fishes, one from TABLE 1. EXAMPLE OF A SERIES OF CROSS-STIMULATIONS a Donor fish Experimental fish Driva Namsen Tat]ord
Driva
Namsen
Tafjord
III I II
II III II
III III II
aThe entries indicate the threshold values obtained when stimulating with fraction V from the intestinal content of the three smolt groups. I, threshold range 10-5-10-4;11, 10 - 4 10 -3 ; III, 10-3-10 -2 g/liter.
1089
SALMON POPULATIONS AND ODORS
each river, to the most potent fraction (fraction V) of intestinal content from smolts of the same three groups. In this case the Driva fish was most sensitive to the Namsen donor and the Namsen fish to the sample of Driva fish. All these samples were equally potent to the experimental fish from Tafjord. A systematic search through the experimental fishes revealed no specific sensitivity to any particular donor sample. Neither was a particular donor sample specifically potent to the experimental fish.
DISCUSSION
The results of the present experiments open perspectives in various aspects of salmon migration. First, the threshold of the amino acids versus the fractions of intestinal content indicate a potency of natural substances from the fish superior to those of the amino acids. Second, the qualitative differences in the responses of the olfactory bulb point to an important separation of nervous function toward the natural substances that will need further investigation. Third, the results of the cross-comparisons performed in the present study focus on the mechanisms that provide the salmon with their olfactory cues for homing. In our present experiments the threshold towards the amino acids used was at best I • 10-3 g/liter. The samples of crude intestinal content contained substances that evoked responses in concentrations down to 1 • 10-5 g/liter. Every fraction tested contained a number of substances. Since the chemical nature and quantity of the active substances are unknown so far, we may only speculate upon the actual threshold for a single component. It is hardly probable that they are substances with molecular weights lower than the amino acids. Thus the actual threshold on a mole basis will be considerably lower than the mass per liter measurement given. A molecular weight of 500 will give a threshold of 20 nM. It should be noted that the concentrations used by Selset and Dcving (1980) in the behavior studies were 1.5 X 10 -9 g/liter. The methods used in the present experiments do not reflect a sensitivity in accord with that obtained in behavior experiments. The reasons for this discrepancy are at least twofold. The activity, as seen from the electrodes on the bulbar surface, does not reflect the activity changes in one single neuron in the olfactory bulb, which in theory can be responsible for the behavior of the fish. Second, there exists a possibility that there are seasonal variations in sensitivity and that the period at which the present fishes were tested (JuneOctober) did not coincide with that of their peak sensitivity. The most potent fraction of intestinal conten t evoked larger responses in the medial part of the olfactory bulb than in the lateral part. D~ving et al. (1980) have shown that this part of the olfactory bulb in grayling and sea char responds to bile salts and derivatives of bile salts at concentrations below the ones found for amino acids in the lateral part of the bulb. Thus, the intestinal
1090
F1SKNES AND D~VING
sample contains substances that are qualitatively different from the amino acids, and they evoke responses in a part of the olfactory bulb that has been proposed to serve functions concerning social behavior in the fish while the lateral part of the olfactory bulb, which responds to amino acids, is supposed to serve functions concerning feeding (Thommesen, 1976, 1978). The samples that evoke reactions in the medial part of the bulb and contain the most potent substances are equivalent to those chemical fractions of sea-char smolts that attract mature migrating sea chars (Selset, 1980; Selset and Dcving, 1980). Evidence from the present study concurs with the results found in other salmonid species with different techniques mentioned above. Priesner (1968, 1969) studied the response in male antennae of saturniid moths to female abdominal gland substances. In this extensive study it was shown that different species produced specific substances. The receptors were specifically sensitive to substances produced by their own species. Between very closely related species, the specificity was absent. The present study indicates similar mechanisms. Data from Stabell et al. (1982) indicated chemical variability in the samples used. When presented to the fish, however, there was no evidence indicating that the samples from one group of donor smolts contained substances that were specifically potent to the experimental fishes of the same group. The present results show that no single donor sample was a specifically potent odor to the siblings. In other words, all fish seem equally sensitive to all donors. These findings indicate two important points in salmonid migration. First, all the salmon smolts of the waterways will contribute to a gigantic odor trail in the fjords and coastal currents. According to Nordeng's "pheromone" hypothesis, the adult salmon return to their native river following this trail. The specificity of a current containing odors from salmon smolts of different rivers will increase as the counter-current swimming fish approaches the origin of its population. Chemical specificity is indicated by the parallel study performed by Stabell et al. (1982). Their results from thin-layer chromatography of the fractions used in the present experiments demonstrated chemical variation in the retarded fractions and in particular in the most potent fraction found, fraction V. Second, the specificity in behavior which would permit the salmon to return correctly to its spawningsite will depend upon proper connections within the nervous substrate underlying the fish migration. Further progress can be made in elucidating the mechanisms behind the wiring of the nervous substrate in migratory behavior after identification of the substances that are responsible for the correct choice of spawning site. In considering the nervous mechanisms we note the results of the experiments described by Hasler et al. (1978), who have shown that returning coho salmon (Oncorhynchus kisuteh Walbaum) are guided by artificial odors to which they have previously been exposed. Their results indicate that specific memory processes might be involved in salmonid
SALMON POPULATIONS AND ODORS
1091
m i g r a t i o n . W e believe t h a t g e n e t i c f a c t o r s a r e i m p o r t a n t in s a l m o n i d m i g r a t i o n b e c a u s e o f t h e specificity o f n a t u r a l p o p u l a t i o n s (St~hl, 1981) a n d t h e v a r i a t i o n s in c h e m i c a l c o m p o s i t i o n o f t h e o d o r a n t s u b s t a n c e s ( S t a b e l l et al., 1982). A c o m b i n a t i o n o f g e n e t i c f a c t o r s a n d m e m o r y s h o u l d also be c o n s i d e r e d in f u t u r e e x p e r i m e n t s . A d v a n c e s in o u r u n d e r s t a n d i n g o f s a l m o n i d m i g r a t i o n r e q u i r e t h e t o o l s p r o v i d e d by t h e k n o w l e d g e o f t h e n a t u r a l l y occurring odorants.
Acknowledgments--This study was made possible by funds to K.B.D. from the Norwegian Fisheries Research Council grant 203.01. We are grateful to the personnel at the Research Station for Salmonids, Sunndalsera and Averey Units, for help and guidance in supplying and handling the fish and to O.B. Stabell for determining the dry weight of the natural samples and their fractions. REFERENCES BELGttAUG, R., and D!~VING,K.B. 1977. Odour thresholds determined by studies of the induced waves in the olfactory bulb of the char (Salmo alpinus L.) Comp. Biochem. Physiol. 57A:577-579. DelVING, K.B., NORDENG,I-I., and OAKLEY,B. 1974. Single unit discrimination of fish odours released by char (Salmo alpinus L.) populations. Cbmp. Biochem. Physiol. 47A: 1051-1063. DOVIN~, K.B., SELSET, R., and TnOMMESEN, G. 1980. Olfactory sensitivity to bile acids in salmonid fishes. Acta Physiol. Scand. 108:123-131. FIS~NES, B. 1979. The sensitivity to possible population specific odours in Atlantic salmon (Salmo salar L.). Abstract from oral communication. 2nd Symposium on Fish Physiology. Geteborg, June 1979, p. 41. GROVES,A.B., COLLINS,G.B., and TREFETHEN,P.S. 1968. Roles of olfaction and vision in choice of spawning site by homing adult chinook salmon (Oncorhynchus tshawytscha) J. Fish. Res. Board Can. 25:867-876. HASLER, A.D., SCHOLZ, A.T., and HORRALL,R.M. 1978. Olfactory imprinting and homing in Salmon. Am. Sci. 66:347-355. NORDENG, H. 1971. Is the local orientation of anadromus fishes determined by pheromones? Nature 233:411-413. NORDENG,H. 1977. A pheromone hypothesis for homeward migration in anadromous salmonids, Oikos 28:155-159. OTTOSON,D. 1959a. Studies on slow potentials in the rabbit's olfactory bulb and nasal mueosa. Acta Physiol. Scand. 47:136-148. OTTOSON, D. 1959b. Comparison of slow potentials evoked in the frog's nasal mucosa and olfactory bulb by natural stimulation. Acta Physiol. Scand. 47:149-159. PRIESNER, E. 1968. Die interspezifischen Wirkungen der Sexuallockstoffe der Saturniidae (Lepidoptera). Z. Vergl. Physiol. 61:263-297. PRIESNER, E., 1969. A new approach to insect pheromone specificity, pp. 235-240, in C. Pfaffmann (ed.). i l l Int. Symp. Olfaction and Taste (1968). Rockefeller University Press, New York. SELSEX,R. 1980. Chemical methods for fractionation of odorants produced by char smolts and tentative suggestions for pheromone origins. Acta Physiol. Scand. 108:97-103. SELS~X, R., and D~vIN~, K.B. 1980. Behaviour of mature anadromous char (Salmo alpinus L.) towards odorants produced by smolts of their own population. Acta Physiol. Scand. 108:113-122.
1092
FISKNES AND D~VING
STABELL,O.B., SELSET,R., and SLETTEN,K. 1982. A comparative chemical study on population specific odorants from Atlantic salmon. J. Chem. Ecol. 8(1):201-217. STSrtL, G. 1981. Genetic differentiation among natural populations of Atlantic salmon (Salmo salar) in northern Sweden. Ecol. Bull. 34:95-105. THOMMESEN, G. 1976. Spatial differences in specificity of trout (Salmo trutta L.) olfactory bulb. Acta PhysioL Scand. 96:6A-7A. THOr~MES~N, G. 1978. The spatial distribution of odour induced potentials in the olfactory bulb of char and trout ( Salmonidae). Acta Physiol. Scand. 102:205-217. TOFT, R. 1975. Lukt ock synsinnets roll fOr lekvandringsbeteendet hos Ostersj61ax. Swed. Salmon Res. Inst. Rep. L F I Medd. 10:l-40. WIssY, W.J., and HASLER,A.D. 1954. Effect of olfactory occlusion on migrating silver salmon (0. kisutch). J. Fish. Res. Board Can. 11:472-478. WOLF, K., and QUIMBY, M.C. 1969. Fish cell and tissue culture, pp. 253-305, in W.S. Hoar and D.J. Randall (eds.). Fish Physiology, Vol. 3. Academic Press, New York.
O L F A C T O R Y SENSITIVITY TO G R O U P - S P E C I F I C S U B S T A N C E S IN A T L A N T I C S A L M O N (Salmo salar L.)
B R I T F I S K N E S ~ a n d K J E L L B. D ~ V I N G Institute of Zoophysiology University o f Oslo P.O. Box 1051 Blindern Oslo 3, Norway
(Received October 27, 1981; revised December 14, 1981) Abstract--The olfactory sensitivity of three groups of Atlantic salmon (Salmo salar L.) toward substances emanating from their own groups was studied. Thresholds were determined by electrophysiological recordings of the induced waves from the medial and lateral part of the olfactory bulb surface. The intestine contained more potent olfactory substances on a gram per liter basis than skin mucus, urine, or amino acids. Chemical fractions were obtained from a parallel study on the nature of the naturally occurring substances by Stabell et al. (1982). A retarded fraction from chromatography on a Sephadex G-25 column contained the most potent material. The most potent fractions of the intestinal content evoked responses mainly in the medial part of the olfactory bulb, whereas the lateral part responded to amino acids. The results suggest that all salmon smolts of the waterways contribute to an odor trail in the coastal currents, thus facilitating the odor-dependent migration of the mature salmon. Key Words--Salmo salar, olfactory sensitivity, olfactory bulb, induced waves, pheromones, migrational cues, homing.
INTRODUCTION W h e n a n a d r o m o u s s a l m o n i d fishes r e t u r n to f r e s h w a t e r t o s p a w n a f t e r s t a y i n g in the sea, t h e y will h o m e to t h e i r n a t i v e river. T h e sense o f s m e l l has b e e n s h o w n to be e s s e n t i a l f o r c o r r e c t h o m i n g in s a l m o n ( W i s b y a n d H a s l e r , 1954; G r o v e s et al., 1968; T 0 i l , 1975). T h i s i m p l i e s t h e p r e s e n c e o f specific s u b s t a n c e s in e a c h river. A c c o r d i n g to the " p h e r o m o n e " h y p o t h e s i s p r o p o s e d b y N o r d e n g (1971, ~Present address: NORAD, P.O. Box 2646, Dar-es-Salaam, Tanzania. 1083 009841331 / 82/0800-1083503.00/0 9 1982 Plenum Publishing Corporation
1084
FISKNESAND Dg"VING
1977), the salmon smolts migrating to sea release population specific substances, forming an odor trail that guides the mature salmon to their native river. This hypothesis is supported by the results from behavioral experiments which have shown that adult sea char (Salmo salvelinus L.) is attracted to water where smolts from their own population have been kept (Selset and Doving, 1980). Physiological recordings from single cells in the olfactory bulb of sea char have given evidence of a sensory basis for discrimination between populations of char (Doving et al., 1974). The mechanism that is used by the fish olfactory system to discriminate specific substances of its own population from others is unknown. Two possibilities can be considered: (1) the returning salmon has a higher olfactory sensitivity to substances originating from its own population; and (2) the olfactory sensitivity is the same towards all population-specific substances. In the latter case the specific effect on the behavior of the fish must rely on specific interaction in the central nervous system. The aim of the present study was twofold: first, to investigate the olfactory sensitivity in Atlantic salmon to substances from salmon smolts of different groups; and second, to approach, by electrophysiological means, the most interesting chemical fraction of the samples with respect to populationspecific substances. A short abstract of the results of this study has been published (Fisknes, 1979). METHODS AND MATERIALS
Donor and Experimental Fishes. The fishes ( Salmo salar L.) were raised at the Research Station for Salmonids, Sunndalsr Unit, Western Norway and were offspring of wild fish caught in three Norwegian rivers: Driva, Namsen, and Tafjord. The fishes were thus the first generation in captivity. They were hatched in the spring of 1976. From autumn 1976 they were marked (fin clipped and freeze-branded) and mixed with other fish. They smoltified in spring 1977 and from then on they were held in culture in the sea. One to two weeks before the experiment the test fishes were transferred by air to the aquarium facilities at the University of Oslo and kept in recirculating seawater at a temperature of I0-12~ Handling o f Experimental Fish. The experimental fishes were 2--3-yearold salmon weighing 690-2630 g. They were initially anesthetized with tricain-methane sulfonate (MS-222, Sandoz) at 1:5000 concentration and then immobilized with Alloferin (Roche) given intramuscularly. The dose of the active principle, toxiferin, injected was about 4 mg/kg body weight. The fish were wrapped in a wet sponge and placed in a holder in the experimental set-up. The gills were continuously perfused with seawater at 14-16~ through a tube inserted into the mouth, at a flow about 1.8 liter/min. The
SALMON POPULATIONS AND ODORS
1085
Basin water
L - s e r i n e 10-3M
olfactory nerves
FIG. 1. Recording traces from the medial (e) and lateral (o) part of the left olfactory bulb in salmon. Stimuli were water from a saltwater basin containing salmon and L-serine at 10-3 M. Time bar = 1 sec; vertical bar = lO0/.tV. dorsal part of the skull, together with the underlying fat and connective tissue, was carefully removed and the olfactory bulbs exposed. Recording. The recording sites were at the lateral and medial part of the dorsal surface of the left olfactory bulb, positioned as shown in Figure 1. At each site a differential recording system was used. The electrodes were of glass and filled with Hanks' Ringer for marine fishes (Wolf and Quimby, 1969). The electrodes were inserted into electrode holders containing sintered Ag/AgC1. Tip diameter of the electrodes ranged from 10 to 50 #m. A stainless-steel wire served as a c o m m o n indifferent electrode. It was connected to the fish in the connective tissue caudal to the opening made in the skull. The potentials were amplified with two differential preamplifiers (WPI, D A M - 5 A ) and fed through a bandpass filter, 1-50 Hz. The signals were displayed on a pen recorder (Hewlett-Packard, type 7402 A). Stimuli. The stimulus donors were hatched at the same time as the experimental fishes and some of them were siblings or half-siblings of the experimental fishes. Skin mucus, intestinal content, and urine were collected in May 1977 from two groups of smolts from each of the three rivers. Each group contained two to seven individuals that were siblings. The methods for collection of these samples and the chemical procedure, including fractioning, are described in detail by Stabell et al., (1982). Briefly, the original samples of skin mucus, intestinal content, or urine were centrifuged (5700 rpm, 15 min). The samples of skin mucus and intestinal content were treated with acetic acid to a final concentration of 5%. The precipitate after centrifugation (25,000g, 30 rain) was separated from the supernatant. For olfactory tests the precipitate was stirred in distilled water after washing in 5% H A C and then centrifuged. The supernatants of skin mucus and intestinal content were run in a Sephadex G-25 column and separated into fractions I - V I I I . Fractions I V - V I I I
1086
FISKNESAND D~VING
appeared after the salt volume and will be mentioned as retarded fractions. Aliquots for biological tests were taken out at all stages of the chemical procedure. The dry weight of all samples was determined. The samples were kept frozen at - 2 0 ~ Just prior to the experiment they were thawed and diluted with artificial sea water in concentration steps of 1:10. Controls. As control stimuli L-serine and L-glutamine (Sigma) were made up in a series of dilution steps just prior to use. In addition "basin water," a water sample from the basin in which the fishes were kept, was used as a control stimulus. Stimulation. The olfactory epithelium was continuously flushed with artificial sea water through a polyethylene tube inserted in the anterior nares. The waterflow was ca. 13 ml/min. The stimulus solution was introduced from a pipet via a manually operated three-way stopcock. Procedure. All samples were applied in series of increased concentration. Each concentration was applied twice with a 30-sec interval between each stimulation. At intervals of 30 sec or more, no interaction of the stimuli was evident from observation of the response amplitudes. The time interval between two different dilution series was a minimum of 60 sec. Threshold concentration was determined as the lowest concentration of the stimulus that elicited a response (see Belghaug and D Cving, 1977). The concentrations were expressed as gram per liter.
RESULTS When an olfactory stimulus was introduced into the olfactory pit, a change in the potentials recorded from the surface of the olfactory bulb was seen (Ottoson, 1959a,b). Under present recording conditions, the response was characterized by oscillatory waves with constant frequency. The amplitude of the "induced waves" increased with increasing concentration of the stimulus. The frequency was 8-11 Hz at the temperature conditions used in the present study (cf. Dr and Belghaug, 1977). Figure 1 shows the responses to two of the control stimuli, "basin water" and I.-serine, recorded from the medial and lateral part of the bulb. Thresholds. Samples from both skin mucus and intestinal content were highly stimulatory to the salmon olfactory organ. Responses were elicited both in the medial and the lateral part of the olfactory bulb. The threshold yalues (g/liter) showed, however, that there exists a difference in the stimulatory effectiveness between skin mucus and intestinal content, the intestinal content being 10-1000 times more potent than the skin mucus (Figure 2). This was the case both for the original sample and the supernatant. The urine samples were at best equipotent to the skin mucus. In the second step in the chemical fractionation procedure, samples of the
1087
SALMON P O P U L A T I O N S AND ODORS
25 ~
[]s kin
intestinal content (50)
mucus (57)
20 t-
"5
15
I...0
E
TO
Z
-5
-4
-3
-2
-1
0
No response
Log concentration (g/t)
FIG. 2. Distribution of thresholds to supernatants of intestinal content and skin mucus from salmon smolts. The numbers of observations are given in parenthesis.
skin mucus and intestinal content were divided into supernatant and precipitat e fractions. The stimulating efficiency of these fractions was compared with that of the respective crude material. It thus showed that the stimulating substances were not completely isolated in the supernatant, but also remained in the precipitate. When stimulating with successive fractions from the G-25 Sephadex gel filtration, a change occurred in the response between the medial and lateral parts of the bulb. The first three fractions (I-III), which precede the salt volume of the column, elicited a response with higher amplitude from the lateral part of the olfactory bulb than the medial part. The retarded fractions (IV-VIII) were more likely to elicit the highest response in the medial part of the bulb (Figure 3). This was evident in experiments with fractions of intestinal content, but not for fractions of the skin mucus. Substances in the retarded fractions seem to have qualitative and stimulatory properties different from those in the first, unretarded fractions. Fractions I-III, which elicited responses with highest amplitude in the lateral part of the bulb, showed an increasing stimulatory effect from I to III. The lowest threshold was found for fraction III and 1 • 1014 g/liter in four test fishes. Most effective were the first three of the retarded fractions (IV, V, and VI). The most potent of these was fraction V, with a threshold concentration observed down to 1 • 10-5 g/liter. The amino acids used in the experiment always elicited a response in the lateral part of the bulb, but did so only sometimes in the medial part. The
1088
FISKNES AND D~VING
{} ~-
2F
39 31
T
30
E
15
35
4/*
{1 13 D.
E
o
i I
i II
i
I
III
IV
V
I
I
I
VI
VII
VIII
F r a c t i o n no. FIG. 3. The ratio between the amplitude of the responses at the medial versus the lateral part of the s a l m o n olfactory bulb observed when stimulating with different fractions of the s u p e r n a t a n t of the intestinal content (see text). F o r each fraction the m e a n a n d one s t a n d a r d deviation of the ratio is indicated. The n u m b e r of observations is given above each bar.
threshold were found to 8 X 10 -6 M ( N = 8) for 4 X 10 -3 g/liter and 1 X higher than those of the content.
have a mean of 4 • 10-5 M (N = 8) for L-serine and L-glutamine. These concentrations correspond to 10 -3 g/liter, respectively, i.e. about 10 to 100 times most potent fractions of skin mucus and intestinal
Cross-Comparisons. One of the objectives of the present study was to seek evidence for group-specific sensitivity to natural substances. The experimental fishes were therefore exposed to stimuli from donor fishes of the three groups. Table 1 shows an example of thresholds in three fishes, one from TABLE 1. EXAMPLE OF A SERIES OF CROSS-STIMULATIONS a Donor fish Experimental fish Driva Namsen Tat]ord
Driva
Namsen
Tafjord
III I II
II III II
III III II
aThe entries indicate the threshold values obtained when stimulating with fraction V from the intestinal content of the three smolt groups. I, threshold range 10-5-10-4;11, 10 - 4 10 -3 ; III, 10-3-10 -2 g/liter.
1089
SALMON POPULATIONS AND ODORS
each river, to the most potent fraction (fraction V) of intestinal content from smolts of the same three groups. In this case the Driva fish was most sensitive to the Namsen donor and the Namsen fish to the sample of Driva fish. All these samples were equally potent to the experimental fish from Tafjord. A systematic search through the experimental fishes revealed no specific sensitivity to any particular donor sample. Neither was a particular donor sample specifically potent to the experimental fish.
DISCUSSION
The results of the present experiments open perspectives in various aspects of salmon migration. First, the threshold of the amino acids versus the fractions of intestinal content indicate a potency of natural substances from the fish superior to those of the amino acids. Second, the qualitative differences in the responses of the olfactory bulb point to an important separation of nervous function toward the natural substances that will need further investigation. Third, the results of the cross-comparisons performed in the present study focus on the mechanisms that provide the salmon with their olfactory cues for homing. In our present experiments the threshold towards the amino acids used was at best I • 10-3 g/liter. The samples of crude intestinal content contained substances that evoked responses in concentrations down to 1 • 10-5 g/liter. Every fraction tested contained a number of substances. Since the chemical nature and quantity of the active substances are unknown so far, we may only speculate upon the actual threshold for a single component. It is hardly probable that they are substances with molecular weights lower than the amino acids. Thus the actual threshold on a mole basis will be considerably lower than the mass per liter measurement given. A molecular weight of 500 will give a threshold of 20 nM. It should be noted that the concentrations used by Selset and Dcving (1980) in the behavior studies were 1.5 X 10 -9 g/liter. The methods used in the present experiments do not reflect a sensitivity in accord with that obtained in behavior experiments. The reasons for this discrepancy are at least twofold. The activity, as seen from the electrodes on the bulbar surface, does not reflect the activity changes in one single neuron in the olfactory bulb, which in theory can be responsible for the behavior of the fish. Second, there exists a possibility that there are seasonal variations in sensitivity and that the period at which the present fishes were tested (JuneOctober) did not coincide with that of their peak sensitivity. The most potent fraction of intestinal conten t evoked larger responses in the medial part of the olfactory bulb than in the lateral part. D~ving et al. (1980) have shown that this part of the olfactory bulb in grayling and sea char responds to bile salts and derivatives of bile salts at concentrations below the ones found for amino acids in the lateral part of the bulb. Thus, the intestinal
1090
F1SKNES AND D~VING
sample contains substances that are qualitatively different from the amino acids, and they evoke responses in a part of the olfactory bulb that has been proposed to serve functions concerning social behavior in the fish while the lateral part of the olfactory bulb, which responds to amino acids, is supposed to serve functions concerning feeding (Thommesen, 1976, 1978). The samples that evoke reactions in the medial part of the bulb and contain the most potent substances are equivalent to those chemical fractions of sea-char smolts that attract mature migrating sea chars (Selset, 1980; Selset and Dcving, 1980). Evidence from the present study concurs with the results found in other salmonid species with different techniques mentioned above. Priesner (1968, 1969) studied the response in male antennae of saturniid moths to female abdominal gland substances. In this extensive study it was shown that different species produced specific substances. The receptors were specifically sensitive to substances produced by their own species. Between very closely related species, the specificity was absent. The present study indicates similar mechanisms. Data from Stabell et al. (1982) indicated chemical variability in the samples used. When presented to the fish, however, there was no evidence indicating that the samples from one group of donor smolts contained substances that were specifically potent to the experimental fishes of the same group. The present results show that no single donor sample was a specifically potent odor to the siblings. In other words, all fish seem equally sensitive to all donors. These findings indicate two important points in salmonid migration. First, all the salmon smolts of the waterways will contribute to a gigantic odor trail in the fjords and coastal currents. According to Nordeng's "pheromone" hypothesis, the adult salmon return to their native river following this trail. The specificity of a current containing odors from salmon smolts of different rivers will increase as the counter-current swimming fish approaches the origin of its population. Chemical specificity is indicated by the parallel study performed by Stabell et al. (1982). Their results from thin-layer chromatography of the fractions used in the present experiments demonstrated chemical variation in the retarded fractions and in particular in the most potent fraction found, fraction V. Second, the specificity in behavior which would permit the salmon to return correctly to its spawningsite will depend upon proper connections within the nervous substrate underlying the fish migration. Further progress can be made in elucidating the mechanisms behind the wiring of the nervous substrate in migratory behavior after identification of the substances that are responsible for the correct choice of spawning site. In considering the nervous mechanisms we note the results of the experiments described by Hasler et al. (1978), who have shown that returning coho salmon (Oncorhynchus kisuteh Walbaum) are guided by artificial odors to which they have previously been exposed. Their results indicate that specific memory processes might be involved in salmonid
SALMON POPULATIONS AND ODORS
1091
m i g r a t i o n . W e believe t h a t g e n e t i c f a c t o r s a r e i m p o r t a n t in s a l m o n i d m i g r a t i o n b e c a u s e o f t h e specificity o f n a t u r a l p o p u l a t i o n s (St~hl, 1981) a n d t h e v a r i a t i o n s in c h e m i c a l c o m p o s i t i o n o f t h e o d o r a n t s u b s t a n c e s ( S t a b e l l et al., 1982). A c o m b i n a t i o n o f g e n e t i c f a c t o r s a n d m e m o r y s h o u l d also be c o n s i d e r e d in f u t u r e e x p e r i m e n t s . A d v a n c e s in o u r u n d e r s t a n d i n g o f s a l m o n i d m i g r a t i o n r e q u i r e t h e t o o l s p r o v i d e d by t h e k n o w l e d g e o f t h e n a t u r a l l y occurring odorants.
Acknowledgments--This study was made possible by funds to K.B.D. from the Norwegian Fisheries Research Council grant 203.01. We are grateful to the personnel at the Research Station for Salmonids, Sunndalsera and Averey Units, for help and guidance in supplying and handling the fish and to O.B. Stabell for determining the dry weight of the natural samples and their fractions. REFERENCES BELGttAUG, R., and D!~VING,K.B. 1977. Odour thresholds determined by studies of the induced waves in the olfactory bulb of the char (Salmo alpinus L.) Comp. Biochem. Physiol. 57A:577-579. DelVING, K.B., NORDENG,I-I., and OAKLEY,B. 1974. Single unit discrimination of fish odours released by char (Salmo alpinus L.) populations. Cbmp. Biochem. Physiol. 47A: 1051-1063. DOVIN~, K.B., SELSET, R., and TnOMMESEN, G. 1980. Olfactory sensitivity to bile acids in salmonid fishes. Acta Physiol. Scand. 108:123-131. FIS~NES, B. 1979. The sensitivity to possible population specific odours in Atlantic salmon (Salmo salar L.). Abstract from oral communication. 2nd Symposium on Fish Physiology. Geteborg, June 1979, p. 41. GROVES,A.B., COLLINS,G.B., and TREFETHEN,P.S. 1968. Roles of olfaction and vision in choice of spawning site by homing adult chinook salmon (Oncorhynchus tshawytscha) J. Fish. Res. Board Can. 25:867-876. HASLER, A.D., SCHOLZ, A.T., and HORRALL,R.M. 1978. Olfactory imprinting and homing in Salmon. Am. Sci. 66:347-355. NORDENG, H. 1971. Is the local orientation of anadromus fishes determined by pheromones? Nature 233:411-413. NORDENG,H. 1977. A pheromone hypothesis for homeward migration in anadromous salmonids, Oikos 28:155-159. OTTOSON,D. 1959a. Studies on slow potentials in the rabbit's olfactory bulb and nasal mueosa. Acta Physiol. Scand. 47:136-148. OTTOSON, D. 1959b. Comparison of slow potentials evoked in the frog's nasal mucosa and olfactory bulb by natural stimulation. Acta Physiol. Scand. 47:149-159. PRIESNER, E. 1968. Die interspezifischen Wirkungen der Sexuallockstoffe der Saturniidae (Lepidoptera). Z. Vergl. Physiol. 61:263-297. PRIESNER, E., 1969. A new approach to insect pheromone specificity, pp. 235-240, in C. Pfaffmann (ed.). i l l Int. Symp. Olfaction and Taste (1968). Rockefeller University Press, New York. SELSEX,R. 1980. Chemical methods for fractionation of odorants produced by char smolts and tentative suggestions for pheromone origins. Acta Physiol. Scand. 108:97-103. SELS~X, R., and D~vIN~, K.B. 1980. Behaviour of mature anadromous char (Salmo alpinus L.) towards odorants produced by smolts of their own population. Acta Physiol. Scand. 108:113-122.
1092
FISKNES AND D~VING
STABELL,O.B., SELSET,R., and SLETTEN,K. 1982. A comparative chemical study on population specific odorants from Atlantic salmon. J. Chem. Ecol. 8(1):201-217. STSrtL, G. 1981. Genetic differentiation among natural populations of Atlantic salmon (Salmo salar) in northern Sweden. Ecol. Bull. 34:95-105. THOMMESEN, G. 1976. Spatial differences in specificity of trout (Salmo trutta L.) olfactory bulb. Acta PhysioL Scand. 96:6A-7A. THOr~MES~N, G. 1978. The spatial distribution of odour induced potentials in the olfactory bulb of char and trout ( Salmonidae). Acta Physiol. Scand. 102:205-217. TOFT, R. 1975. Lukt ock synsinnets roll fOr lekvandringsbeteendet hos Ostersj61ax. Swed. Salmon Res. Inst. Rep. L F I Medd. 10:l-40. WIssY, W.J., and HASLER,A.D. 1954. Effect of olfactory occlusion on migrating silver salmon (0. kisutch). J. Fish. Res. Board Can. 11:472-478. WOLF, K., and QUIMBY, M.C. 1969. Fish cell and tissue culture, pp. 253-305, in W.S. Hoar and D.J. Randall (eds.). Fish Physiology, Vol. 3. Academic Press, New York.
