| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Departments of Psychology (S.R.W.) and Pediatrics (J.L.T.), University at Buffalo, State University of New York, Buffalo, New York 14260-1660; and Section on Neural Gene Expression (H.K.C., W.S.Y.), National Institute of Mental Health, National Institutes of Health, Department of Health and Human Services, Bethesda, Maryland 20892-4483
Address all correspondence and requests for reprints to: Dr. W. Scott Young, Building 49, Room 5A51, 9000 Rockville Pike, Bethesda, Maryland 20892-4483. E-mail: wsy{at}mail.nih.gov.
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Many of the brain regions that receive accessory olfactory information, either directly or indirectly, contain oxytocin (Oxt) and vasopressin (Avp) cell bodies, fibers, or receptors (for recent reviews, see Refs. 14, 15, 16). The distribution of the Avp receptor (Avpr) 1b has been less well characterized than the distribution of the Avpr1a or Oxt receptor (Oxtr). It is clear that the neuropeptides and/or their receptors are well represented in many regions receiving olfactory cues. This places the Oxt and Avp system in a position to modulate the brains response to olfactory cues.
A key question then is, why do chemosensory cues from the mate fail to trigger this neuroendocrine reflex? Electrical stimulation of the accessory olfactory bulb coincident with the twice-daily prolactin surges reported in the pregnant mouse induces pregnancy block in the absence of a male (17). Brennan et al. (18) have hypothesized that neural activation by the pregnancy blocking signal from the mate, but not other males, is disrupted at the level of the accessory olfactory bulb. This necessitates a mechanism whereby mating signals the formation of an olfactory memory of the animal with whom the female mated. Oxt could signal that mating has occurred and/or directly induce the formation of the memory by acting in the olfactory bulb. In the ewe, vaginocervical stimulation induces Oxt release and the induction of maternal behavior (19). In this context, Oxt appears to signal that parturition has occurred. Infusion of Oxt into the olfactory bulb of a nonpregnant ewe can induce the formation of a maternal bond (20). Moreover, male and female mice lacking Oxt appear unable to recognize individuals (21, 22). Oxt knockout (KO) mice fail to reduce their investigation of an animal to which they have already been exposed. Delivery of Oxt to the medial amygdala prevents this deficit (21). To test the hypothesis that the Oxt in female mice is essential for the formation of a memory of their mate, we assessed the Bruce effect in mice lacking Oxt.
Once the memory of the mate is formed, the social stimuli currently present in the environment must be detected and compared with that of the memory. If the current stimuli and the memory are the same (or similar enough), then the neuroendocrine mechanism underlying pregnancy block is either inhibited or simply not triggered. If the current stimuli and the memory differ, then the neuroendocrine mechanism underlying pregnancy block is triggered or the inhibition of the reflex lifted. This requires the proper processing of social cues currently in the environment.
Social recognition is an olfactory based memory that has been well described in the male. It is often assessed using a social discrimination paradigm in which the subjects behavior varies toward a familiar mouse than an unfamiliar mouse. Many studies have shown that Oxt and Avp modulate this phenomenon, although the direction of the effect varies with factors such as timing of manipulation, route of administration, and behavioral paradigm used. Previous results demonstrate that male mice in which the Avpr1b gene is inactivated do not respond normally to social cues and exhibit deficits in olfactory social recognition using a discrimination paradigm (23, 25). There are conflicting reports concerning the role of the Avpr1a gene in the regulation of olfactory social recognition. One report suggests a profound disruption in social recognition in male Avpr1a KO mice (26), whereas another report failed to find such a deficit (38). However, pharmacological studies suggest that Avpr1a agonists facilitate social recognition, whereas antagonists disrupt it (28, 29, 30, 31, 32). Although olfactory social recognition and pregnancy block appear to be mediated by divergent neural circuitry, both rely on the detection and processing of olfactory social cues. Therefore, we hypothesized that pregnancy block would be disrupted in mice lacking either Avpr.
| Materials and Methods |
|---|
|
|
|---|
Housing.
Unless otherwise stated, all animals were initially group housed in same-sex cages until 60–100 d of age. Although this reflects a broad range of ages, there were no significant genotypic differences in the ages of the subjects. Because control and KO animals were littermates for each experiment, the subjects in each group were, necessarily, of approximately the same distribution of ages. The subjects were then singly housed with lights on from 0600–1800 h and with food and water available ad libitum for at least 2 wk before testing. High-density racks with clean airflow to each individual cage were used.
Stimulus animals.
The stimulus animals were adult (60–180 d of age) male DBA (Jackson Laboratories, Bar Harbor, ME) or BalbC (Jackson Laboratories) mice. The stimulus animals were singly housed in the same room as the subjects. Each male was identified by thin lines drawn on the tail using a black Sharpie permanent marker (Sanford Corp., Oak Brook, IL). The lines were redrawn when deemed necessary, generally every few days. All stimulus males were sexually experienced and were used with many different females over the course of the experiment. Males were screened for their sexual performance before being used as stimulus animals in the experiments and their sexual performances confirmed throughout the duration of the study. There were no differences in the sexual performances of the stimulus males in terms of length of time between pairing and the observation of a sperm plug or in the percentages of females in which sperm plugs were observed (data not shown).
Mating.
A male DBA, arbitrarily selected by an investigator unaware of the genotype of the subject, was placed in the home cage of a subject between 1600 and 1630 h. Every subsequent morning between 0800 and 0900 h, the female was inspected for the presence of a sperm plug. The day the plug was observed was defined as d 0. If a plug was not observed within 10-d pairing, the female was excluded from the study. The number of females excluded from the study for this reason did not significantly differ among the lines of mice or among the genotypes within each line (data not shown). Once a sperm plug was found, the subject was randomly assigned to one of the treatment groups described below. Subjects were added to each group until the target number of subjects was reached.
Treatment groups
Each subject was assigned to one of the following groups:
1. Alone (final n = 10 for each genotype of each line). Once a sperm plug was observed, the DBA male was removed from the cage and the female housed alone until euthanized 10–14 d later.
2. Paired continuously (PC) (final n = 5 for each genotype of each line). After the sperm plug was observed, the DBA male was allowed to remain housed with the female until euthanized 10–14 d later.
3. Familiar (final n = 10 for each genotype of each line). Immediately after the sperm plug was observed, the DBA male was removed and returned to his home cage. The morning of d 1, the same stimulus DBA male was returned to the females cage. On d 2 the DBA male was removed from the females cage. The female was housed alone until euthanized 10–12 d later.
4. Unfamiliar (final n = 10 for each genotype of each line). Immediately after the sperm plug was observed, the DBA male was removed and returned to his home cage. The morning of d 1, a BalbC male was placed in the females cage. On d 2 the BalbC male was removed from the females cage. The female was housed alone until euthanized 10–12 d later.
Ten subjects per group in our experiments generally give enough statistical power to reach significance in a measure that may have high variability within groups. We elected to use only five in the PC group because we have been breeding these lines for weeks and have never encountered any issues with pregnancy.
Assessment of pregnancy.
Ten to 14 d after observation of a sperm plug, the number of fetuses in each uterine horn was counted after isoflurane euthanasia.
Statistics
Within each KO line and genotype, the proportion of subjects that were pregnant was compared using a two-tailed Fishers exact test of probability using the data from the alone group as the expected values. The number of fetuses (of those subjects that were pregnant) and the number of days between pairing and the detection of a sperm plug were compared using a two-way ANOVA, followed by an all-pairwise multiple comparison using the Holm-Sidak method. No comparisons were made among the different genetic lines because they were on different genetic backgrounds.
| Results |
|---|
|
|
|---|
|
|
Of those subjects that were pregnant, there was no effect of genotype (F2,77 = 0.770; P = 0.467) or treatment (F3,77 = 0.260; P = 0.854) on the number of fetuses (Table 1
). There was no interaction between the two (F6,77 = 0.678; P = 0.667). There was no effect of genotype (F2,93 = 0.426; P = 0.654) or treatment (F3,93 = 0.541; P = 0.655) on the number of days between pairing with the stimulus male and the observation of a sperm plug (Table 1
), suggesting that the basic reproductive physiology is intact in all genotypes. There was no interaction between genotype and treatment (F6,93 = 0.236; P = 0.964).
Avpr1a KO mice reexposed to novel, but not familiar males, are not pregnant after 10-d mice
Exposure to an unfamiliar male significantly reduced the percentage of pregnant females in subjects of all genotypes (Fig. 1
, lower panel). Very few females of all three genotypes that were left alone, or left PC or reexposed to the familiar male after mating had no fetuses.
Of those subjects that were pregnant, there was no effect of genotype (F2,72 = 0.381; P = 0.685) or treatment (F3,72 = 1.288; P = 0.285) on the number of fetuses (Table 1
). There was no interaction between the two (F6,72 = 0.987; P = 0.441). There was no effect of genotype (F2,93 = 0.760; P = 0.471) on the number of days between pairing with the stimulus male and the observation of a sperm plug (Table 1
), suggesting that the basic reproductive physiology is intact in all genotypes. There was a significant effect of treatment (F3,93 = 2.783; P = 0.045) on this measure. An all-pairwise post hoc comparison using the Holm-Sidak method did not reveal any groups that were significantly different (P > 0.05 for all comparisons). There was no interaction between genotype and treatment (F6,93 = 2.140; P = 0.412).
| Discussion |
|---|
|
|
|---|
The fact that continuously paired Oxt –/– females remain pregnant is somewhat perplexing. One might reason that if Oxt –/– mice fail to form an olfactory memory to their mate, they would fail to recognize any male, even one paired with them continuously, as their mate. Thus, pregnancy block would occur even when the animals were never separated. Unlike females who were separated from their mate for 24 h before being reintroduced to them, the continuously paired females maintained their pregnancies. We speculate that females PC with their mates do not have an opportunity to "forget" their mate. Sensory systems are optimized to detect changes in the environment. Perhaps in the absence of a new odor in the environment, the females have nothing to use as a comparison with the memory (or the lack of the memory). This is a question that deserves further investigation.
Mice lacking the Avpr1b gene, by contrast, remained pregnant even when exposed to an unfamiliar male. The attack component of territorial aggressive behavior and maternal behavior is reduced in Avpr1b –/– mice. Olfactory recognition measured by a social discrimination task is also disrupted. Avpr1b KOs do not investigate soiled bedding as long as their wild-type littermates do. Olfactometry measurements indicate that there is no apparent main olfactory dysfunction in Avpr1b –/– mice (23, 25). These previous results from our laboratory have led us to hypothesize that the function of the Avpr1b is to couple chemosensory cues mediated by the olfactory system with the proper behavioral or physiological response (23, 25, 37). In the absence of a functional Avpr1b, this coupling is broken, and the proper response is not generated. The present data further refine this hypothesis to include coupling of accessory olfactory system cues that are necessary to generate pregnancy block in response to odor cues from an unfamiliar male (8). Although it is possible that the Avpr1b is a critical facet of the neuroendocrine mechanism underlying pregnancy block, reproductive physiology appears normal in Avpr1b –/– mice, and we are not aware of large effects of Avp on reproductive physiology.
Despite several reports of deficits in social discrimination memory tasks (26) and olfactory discrimination in Avpr1a –/– mice (38), no differences between Avpr1a +/+, Avpr1a +/–, and Avpr1a –/– mice were found. This is consistent with our data indicating that Avpr1a –/– mice have no deficits in olfactory social recognition (38). Some of the pharmacological literature on the Avpr1a and social recognition (28, 29, 30, 31) suggests a strong link between Avp and social recognition in males. However, in females the role of Avp in social recognition is more tenuous (39).
Halem et al. (40) showed that the pattern of fos-like immunoreactivity induced when a female is exposed to chemosensory cues from her mate is attenuated in the medial amygdala, medial preoptic area, and the bed nucleus of the stria terminalis, but not the accessory olfactory bulb, compared with females exposed to bedding from unfamiliar males. Because each of these regions contains Oxt or Oxtr and Avpr1a in the mouse, the differential distribution of the receptors does not provide immediate insight as to the specific role of each in pregnancy block. Avpr1b has been detected using in situ hybridization in the amygdala, but RT-PCR suggests that the distribution is broader than is detected by in situ hybridization.
To the best of our knowledge, there are no reports of Oxt and Avp or their receptors in the tuberoinfundibular dopaminergic (TIDA) neurons. Therefore, any effects of our gene manipulations must affect neural circuitry feeding into the TIDA neurons to alter their activity. It is interesting that Oxt affects the activity of the TIDA neurons as well as the release of prolactin (41, 42). Central Oxt appears to inhibit prolactin release, possibly by activating the TIDA neurons, whereas peripheral Oxt acts on pituitary lactotrophs to stimulate prolactin release (24). We would expect this to disrupt the establishment of pregnancy in all Oxt –/– subjects. Because only Oxt –/– animals in the familiar and unfamiliar groups are not pregnant, Oxt acting on the TIDA neurons is unlikely to underlie this phenotype.
Based on these and others data, we hypothesize that Oxt is critical for the formation of an olfactory memory to the mate. We also hypothesize that the Avpr1b is critical for the ability of accessory olfactory cues from a stranger to trigger pregnancy block. These hypotheses generate testable predictions that will broaden our knowledge of the role of the neuropeptides in the regulation of social behavior when tested.
| Acknowledgments |
|---|
| Footnotes |
|---|
Current address for H.K.C.: Kent State University, Kent, Ohio 44242.
Disclosure Statement: The authors have nothing to declare.
First Published Online October 18, 2007
Abbreviations: Avp, Vasopressin; Avpr, vasopressin receptor; KO, knockout; Oxt, oxytocin; Oxtr, oxytocin receptor; PC, paired continuously; TIDA, tuberoinfundibular dopaminergic.
Received August 1, 2007.
Accepted for publication October 9, 2007.
| References |
|---|
|
|
|---|
-aminobutyric acid, glutamate and noradrenaline release. Brain Res 669:197–206[CrossRef][Medline]
and -β knockout mice. Proc Natl Acad Sci USA 100:6192–6197
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |