Physiology & Behavior, Vol. 48, pp. 701-706. ©Pergamon Press plc, 1990. Printed in the U.S.A.
0031-9384/90 $3.00 + .00
Olfactory But Not Vomeronasal Mediation of Scent Marking by Male Golden Hamsters R O B E R T E. J O H N S T O N A N D U L R I C H G. M U E L L E R
Department o f Psychology, Uris Hall, Cornell University, Ithaca, N Y 14853 R e c e i v e d 12 April 1990
JOHNSTON, R. E. AND U. G. MUELLER. Olfactory but not vomeronasal mediation of scent marking by male golden hamsters. PHYSIOL BEHAV 48(5) 701-706, 1990.--Although olfactory and vomeronasal projections to the olfactory bulb and to the rest of the brain are quite distinct, it is not clear how the functions of these two chemosensory systems differ. In these experiments we attempted to determine the roles of the main olfactory and vomeronasal sensory systems in mediating scent marking by male golden hamsters, a behavior known to be stimulated by odors of other hamsters. Lesions of the main olfactory mucosa by irrigation of the nasal cavity with a solution of zinc sulfate dramatically reduced the frequency of marking. The degree of reduction in marking was correlated with an independent test of ability to detect volatile odorants. In contrast, surgical removal of the vomeronasal organ had no effect on marking frequency. Thus scent marking by male hamsters is primarily mediated by the main olfactory system and does not require the presence of the vomeronasal system. These results are discussed in the context of a theory that stresses the importance of the main olfactory system in pattern recognition and, in particular, in many types of social recognition. Olfaction Vomeronasal organ Scent marking Social recognition Animal communication Olfactory communication Odor discrimination Scent glands Pheromones
AN important advance in the study of nasal chemosensory function in recent years has been the realization that the olfactory and vomeronasal systems have separate neural projections and may have some separate functions (13, 27, 31, 38). The vomeronasai system, for example, seems to be especially important for mediating responses to large, nonvolatile molecules (21,41); it is also necessary for mediating many endocrine responses to chemical communication signals (27,39). The main olfactory system, although it is influential in some cases of endocrine response to social signals (16), has not been shown to be necessary for any of them. In contrast, the mediation of many behavioral responses to chemical signals seems to be influenced by both the main olfactory and vomeronasal systems (27). For example, male hamster copulatory behavior, which was originally thought to be dependent on the vomeronasal organ, has turned out to be influenced strongly by both systems (13, 26, 27). In this and other examples in which both systems are involved, it is not clear whether the two system~ have redundant or distinct functions. Although it is generally believed that the main olfactory system is responsible for the majority of functions of nasal chemoreception, the vomeronasal system has been implicated in a wide variety of responses (39) and the extent to which the olfactory system has unique functions needs to be more clearly determined. Johnston (13) has argued that, based on the differences in neuroanatomy of the two systems, one would expect that the olfactory system should be particularly good at discrimination and/or recognition of objects or other animals. Such abilities probably depend on pattern recognition mechanisms; that is, the recognition of unique odor qualities (or " i m a g e s " ) that derive their unique character from the proportions of chemicals in complex mixtures. In many cases these mixtures may differ in relatively subtle ways, such as the differences in scent compositions between individuals of the same
species. The widespread projections of the main olfactory bulb to the pyriform and entorhinal cortex suggest a more extensive involvement in pattern recognition functions than the much more restricted projections of the accessory (vomeronasal) olfactory bulb to the amygdala and bed nucleus of the stria terminalis. Although there is some evidence to support this hypothesis (5, 17, 30, 34), there is also evidence that the vomeronasal system may be essential for, or contribute to, these kinds of social discriminations (3, 18, 35, 40). At present more information is needed, especially about behaviors that are mediated primarily by the main olfactory system, in order to evaluate the hypothesis. Much of the research to date on the olfactory and/or vomeronasal mediation of behavior has involved global, multicomponent behavioral complexes such as male or female copulatory, aggressive or maternal behavior. In order to discover more discrete functions of the two systems it may be more productive to investigate smaller, more discrete units of behavior. This approach has been successfully applied to the ultrasonic calling behavior of male house mice (40), and in this paper we investigate the main olfactory and vomeronasal contributions to scent marking in male hamsters. Flank marking is a stereotyped behavioral pattern used by golden hamsters to deposit secretions of the flank gland and probably secretions from the head and other regions of the body (4, 9, 1 1, 14). This behavior occurs as part of normal maintenance activities, but is facilitated by contact with other hamsters or their odors. In males flank marking is also supported by testicular androgens (12). The most important scent that stimulates flank marking by a male is the scent of another male's flank gland (10); males mark much less frequently in response to the scent of males lacking the flank gland than to the scent of intact males (7). In addition it is only conspecific male odor that stimulates flank
701
702
JOHNSTON AND MUELLER
marking. Males mark less frequently when exposed to the scent of the closely related Turkish hamster (Mesocricetus brandti) than to conspecific male scent [Mesocricetus auratus (15)]. Thus this behavior is stimulated by odors of a very specific nature: conspecific scent that includes the components contributed by the flank gland. On the basis of the hypothesis discussed above, one would predict that this behavior would be mediated by the main olfactory system but not by the vomeronasal system. EXPERIMENT 1
In this experiment we investigated the effects of lesions of the main olfactory system on flank marking by male hamsters in response to odors of other males.
Method From an initial pool of 23 male hamsters (Mesocricetus auratus), 17 males were selected as subjects on the basis of consistent flank marking during 4 days of habituation trials (see below); 23 additional males served as stimulus animals. All males were between 6 and 7 months of age. The hamsters were laboratory stock derived from the random-bred strain of LakeviewCharles River. Animals were weaned at 25-30 days of age and transferred to individual, solid-bottom cages (30 x 15 × 15 cm). The colony was maintained on a reversed light cycle of 14 h bright (average 35.5 lux) and 10 h dim (0.014 lux) illumination. Food and water were available in home cages at all times. Behavioral testing procedures were similar to those used before (9, 11, 12). Males were tested in the temporarily vacant home cages of stimulus males; each experimental male was matched with a stimulus male and was always tested in his cage. Trials were run during the same time period each day, 1 to 2 hours after the start of the dim phase of the cycle. Trials lasted for 7 min, after which the experimental males were returned to their own home cages. The number of flank marks were recorded with a hand-held counter. Males were observed for 24 consecutive days in a repeated measures design: 4 days of habituation to the procedures, 8 days of baseline observations, 6 days of observation after treatment with saline and 6 days of observation after treatment with zinc sulfate. The zinc sulfate and saline treatments of the nasal mucosa were carried out on the last day of the preceding test series, after trials for that day had been completed. Zinc sulfate treatment was accomplished following the procedure of Winans and Powers (17,37). Males were lightly anesthetized with ether and secured on their backs on an inclined board. Zinc sulfate (0.5 cc of a 5% solution of zinc sulfate in a 0.5% saline solution) was introduced into the nasal cavity via the nasopharyngeal meatus using a curved tube. The animal was held with its head lowered so that excess solution drained out the nares; drainage was assisted by gentle aspiration. This method effectively destroys a large percentage of the nasal epithelium, including the sensory receptors of the main olfactory system, without damaging the receptors of the vomeronasal organ (op cit). Saline treatments were carried out in the same manner. Zinc sulfate is toxic. If ingested, animals can become lethargic or even die; 6 males were removed from the experiment because of adverse effects of zinc sulfate, leaving 11 males in the experimental group. The effects of the zinc sulfate treatment were assessed by histology of the nasal mucosa. The day after the final behavioral test males were deeply anesthetized with sodium pentobarhital and perfused transcardially with saline, followed by 10% formalin. The snout was decalcified, embedded in paraffin, sectioned at 40 microns and stained with hematoxylin and eosin. Every fifth section was examined for healthy olfactory and vomeronasal
mucosa; a quantitative estimate of the percentage of normallooking receptor mucosa was made by recording, for each 0.60.7-ram length of olfactory mucosa, the presence or absence of any normal receptor tissue. The totals (the number of "lengths" and the number of lengths with normal tissue) for each side of the nasal cavity were obtained and then averaged to obtain a single percentage of normal tissue for each male. In questionable cases, the tissue was always scored as " n o r m a l . " For the first four animals we attempted to assess remaining functional tissue by the horseradish peroxidase procedure (28), but despite good results with pilot animals these results were inconsistent and could not be used. Independent assessment of the animals' ability to detect volatile chemicals was carded out by measuring the time it took the male to find a cup containing peanut butter that was buried in the male's home cage under approximately 1 cm of bedding. Peanut butter was placed on the bottom of a disposable plastic cup (2 cm diameter, 2 cm high) that was covered with Kimwipe tissue, which was in turn punctured 10 times, creating small holes. After a male was removed from his cage, the cup was buried and the male was returned to his cage; the latency to discover the peanut butter was measured. The criteria for detection were that the male had to stop and sniff with his nose at the surface of the bedding (or below) for at least 2 sec and uncover at least part of the top surface of the cup by snout movements or digging with its paws. Males did not have the opportunity to contact or ingest the peanut butter; they did not dig through the covering of the cup, although they probably could have. These tests were carried out 1-2 hours after the daily marking trial on days 4 and 6 of the saline and zinc sulfate conditions. Since it is clear that some olfactory function returns within 4-8 days after zinc sulfate treatment (1,33) the data is presented as group means per test day so that changes in marking rate over time can be examined.
Results and Discussion Zinc sulfate treatment of the nasal mucosa resulted in a dramatic drop in flank marking rate whereas the saline control treatment had no significant effect (Fig. 1; Friedman nonparametric ANOVA for treatment effects, chi r 2= 17.636, p
0031-9384/90 $3.00 + .00
Olfactory But Not Vomeronasal Mediation of Scent Marking by Male Golden Hamsters R O B E R T E. J O H N S T O N A N D U L R I C H G. M U E L L E R
Department o f Psychology, Uris Hall, Cornell University, Ithaca, N Y 14853 R e c e i v e d 12 April 1990
JOHNSTON, R. E. AND U. G. MUELLER. Olfactory but not vomeronasal mediation of scent marking by male golden hamsters. PHYSIOL BEHAV 48(5) 701-706, 1990.--Although olfactory and vomeronasal projections to the olfactory bulb and to the rest of the brain are quite distinct, it is not clear how the functions of these two chemosensory systems differ. In these experiments we attempted to determine the roles of the main olfactory and vomeronasal sensory systems in mediating scent marking by male golden hamsters, a behavior known to be stimulated by odors of other hamsters. Lesions of the main olfactory mucosa by irrigation of the nasal cavity with a solution of zinc sulfate dramatically reduced the frequency of marking. The degree of reduction in marking was correlated with an independent test of ability to detect volatile odorants. In contrast, surgical removal of the vomeronasal organ had no effect on marking frequency. Thus scent marking by male hamsters is primarily mediated by the main olfactory system and does not require the presence of the vomeronasal system. These results are discussed in the context of a theory that stresses the importance of the main olfactory system in pattern recognition and, in particular, in many types of social recognition. Olfaction Vomeronasal organ Scent marking Social recognition Animal communication Olfactory communication Odor discrimination Scent glands Pheromones
AN important advance in the study of nasal chemosensory function in recent years has been the realization that the olfactory and vomeronasal systems have separate neural projections and may have some separate functions (13, 27, 31, 38). The vomeronasai system, for example, seems to be especially important for mediating responses to large, nonvolatile molecules (21,41); it is also necessary for mediating many endocrine responses to chemical communication signals (27,39). The main olfactory system, although it is influential in some cases of endocrine response to social signals (16), has not been shown to be necessary for any of them. In contrast, the mediation of many behavioral responses to chemical signals seems to be influenced by both the main olfactory and vomeronasal systems (27). For example, male hamster copulatory behavior, which was originally thought to be dependent on the vomeronasal organ, has turned out to be influenced strongly by both systems (13, 26, 27). In this and other examples in which both systems are involved, it is not clear whether the two system~ have redundant or distinct functions. Although it is generally believed that the main olfactory system is responsible for the majority of functions of nasal chemoreception, the vomeronasal system has been implicated in a wide variety of responses (39) and the extent to which the olfactory system has unique functions needs to be more clearly determined. Johnston (13) has argued that, based on the differences in neuroanatomy of the two systems, one would expect that the olfactory system should be particularly good at discrimination and/or recognition of objects or other animals. Such abilities probably depend on pattern recognition mechanisms; that is, the recognition of unique odor qualities (or " i m a g e s " ) that derive their unique character from the proportions of chemicals in complex mixtures. In many cases these mixtures may differ in relatively subtle ways, such as the differences in scent compositions between individuals of the same
species. The widespread projections of the main olfactory bulb to the pyriform and entorhinal cortex suggest a more extensive involvement in pattern recognition functions than the much more restricted projections of the accessory (vomeronasal) olfactory bulb to the amygdala and bed nucleus of the stria terminalis. Although there is some evidence to support this hypothesis (5, 17, 30, 34), there is also evidence that the vomeronasal system may be essential for, or contribute to, these kinds of social discriminations (3, 18, 35, 40). At present more information is needed, especially about behaviors that are mediated primarily by the main olfactory system, in order to evaluate the hypothesis. Much of the research to date on the olfactory and/or vomeronasal mediation of behavior has involved global, multicomponent behavioral complexes such as male or female copulatory, aggressive or maternal behavior. In order to discover more discrete functions of the two systems it may be more productive to investigate smaller, more discrete units of behavior. This approach has been successfully applied to the ultrasonic calling behavior of male house mice (40), and in this paper we investigate the main olfactory and vomeronasal contributions to scent marking in male hamsters. Flank marking is a stereotyped behavioral pattern used by golden hamsters to deposit secretions of the flank gland and probably secretions from the head and other regions of the body (4, 9, 1 1, 14). This behavior occurs as part of normal maintenance activities, but is facilitated by contact with other hamsters or their odors. In males flank marking is also supported by testicular androgens (12). The most important scent that stimulates flank marking by a male is the scent of another male's flank gland (10); males mark much less frequently in response to the scent of males lacking the flank gland than to the scent of intact males (7). In addition it is only conspecific male odor that stimulates flank
701
702
JOHNSTON AND MUELLER
marking. Males mark less frequently when exposed to the scent of the closely related Turkish hamster (Mesocricetus brandti) than to conspecific male scent [Mesocricetus auratus (15)]. Thus this behavior is stimulated by odors of a very specific nature: conspecific scent that includes the components contributed by the flank gland. On the basis of the hypothesis discussed above, one would predict that this behavior would be mediated by the main olfactory system but not by the vomeronasal system. EXPERIMENT 1
In this experiment we investigated the effects of lesions of the main olfactory system on flank marking by male hamsters in response to odors of other males.
Method From an initial pool of 23 male hamsters (Mesocricetus auratus), 17 males were selected as subjects on the basis of consistent flank marking during 4 days of habituation trials (see below); 23 additional males served as stimulus animals. All males were between 6 and 7 months of age. The hamsters were laboratory stock derived from the random-bred strain of LakeviewCharles River. Animals were weaned at 25-30 days of age and transferred to individual, solid-bottom cages (30 x 15 × 15 cm). The colony was maintained on a reversed light cycle of 14 h bright (average 35.5 lux) and 10 h dim (0.014 lux) illumination. Food and water were available in home cages at all times. Behavioral testing procedures were similar to those used before (9, 11, 12). Males were tested in the temporarily vacant home cages of stimulus males; each experimental male was matched with a stimulus male and was always tested in his cage. Trials were run during the same time period each day, 1 to 2 hours after the start of the dim phase of the cycle. Trials lasted for 7 min, after which the experimental males were returned to their own home cages. The number of flank marks were recorded with a hand-held counter. Males were observed for 24 consecutive days in a repeated measures design: 4 days of habituation to the procedures, 8 days of baseline observations, 6 days of observation after treatment with saline and 6 days of observation after treatment with zinc sulfate. The zinc sulfate and saline treatments of the nasal mucosa were carried out on the last day of the preceding test series, after trials for that day had been completed. Zinc sulfate treatment was accomplished following the procedure of Winans and Powers (17,37). Males were lightly anesthetized with ether and secured on their backs on an inclined board. Zinc sulfate (0.5 cc of a 5% solution of zinc sulfate in a 0.5% saline solution) was introduced into the nasal cavity via the nasopharyngeal meatus using a curved tube. The animal was held with its head lowered so that excess solution drained out the nares; drainage was assisted by gentle aspiration. This method effectively destroys a large percentage of the nasal epithelium, including the sensory receptors of the main olfactory system, without damaging the receptors of the vomeronasal organ (op cit). Saline treatments were carried out in the same manner. Zinc sulfate is toxic. If ingested, animals can become lethargic or even die; 6 males were removed from the experiment because of adverse effects of zinc sulfate, leaving 11 males in the experimental group. The effects of the zinc sulfate treatment were assessed by histology of the nasal mucosa. The day after the final behavioral test males were deeply anesthetized with sodium pentobarhital and perfused transcardially with saline, followed by 10% formalin. The snout was decalcified, embedded in paraffin, sectioned at 40 microns and stained with hematoxylin and eosin. Every fifth section was examined for healthy olfactory and vomeronasal
mucosa; a quantitative estimate of the percentage of normallooking receptor mucosa was made by recording, for each 0.60.7-ram length of olfactory mucosa, the presence or absence of any normal receptor tissue. The totals (the number of "lengths" and the number of lengths with normal tissue) for each side of the nasal cavity were obtained and then averaged to obtain a single percentage of normal tissue for each male. In questionable cases, the tissue was always scored as " n o r m a l . " For the first four animals we attempted to assess remaining functional tissue by the horseradish peroxidase procedure (28), but despite good results with pilot animals these results were inconsistent and could not be used. Independent assessment of the animals' ability to detect volatile chemicals was carded out by measuring the time it took the male to find a cup containing peanut butter that was buried in the male's home cage under approximately 1 cm of bedding. Peanut butter was placed on the bottom of a disposable plastic cup (2 cm diameter, 2 cm high) that was covered with Kimwipe tissue, which was in turn punctured 10 times, creating small holes. After a male was removed from his cage, the cup was buried and the male was returned to his cage; the latency to discover the peanut butter was measured. The criteria for detection were that the male had to stop and sniff with his nose at the surface of the bedding (or below) for at least 2 sec and uncover at least part of the top surface of the cup by snout movements or digging with its paws. Males did not have the opportunity to contact or ingest the peanut butter; they did not dig through the covering of the cup, although they probably could have. These tests were carried out 1-2 hours after the daily marking trial on days 4 and 6 of the saline and zinc sulfate conditions. Since it is clear that some olfactory function returns within 4-8 days after zinc sulfate treatment (1,33) the data is presented as group means per test day so that changes in marking rate over time can be examined.
Results and Discussion Zinc sulfate treatment of the nasal mucosa resulted in a dramatic drop in flank marking rate whereas the saline control treatment had no significant effect (Fig. 1; Friedman nonparametric ANOVA for treatment effects, chi r 2= 17.636, p
