Experience-based vocalization of male mice to female chemosignals

Physiology & Behavior, Vol. 31, pp. 26%272. PergamonPress Ltd., 1983. Printedin the U.S.A.

Experience-Based Vocalization of Male Mice to Female Chemosignals JOHN C. MAGGIO, JEANNE H. M A G G I O A N D G L A Y D E W H I T N E Y l D e p a r t m e n t o f Psychology, Florida State University, Tallahassee, F L 32306 R e c e i v e d 3 F e b r u a r y 1983 MAGGIO, J. C., J. H. MAGGIO AND G. WHITNEY. Experience-based vocalization of male mice to female chemosigrials. PHYSIOL BEHAV 31(3) 269--272, 1983.--Previous experimentation led to a suggestion that pituitary-mediated metabolites present in female urine are important determinants of stimulus potency in eliciting 70 kHz ultrasonic vocalizations from adult male mice (Mus musctdus). In the present experiment this hypothesis was reexamined, taking into account the prior experience of the male test subjects. Male mice were given social experience with either hypophysectomized females or intact, normal females and were then monitored for ultrasonic vocalizations to chemosensory stimuli from both. Results indicated that males (1) will emit ultrasonic vocalizations to urine from hypophysectomized females and (2) tend to preferentially emit ultrasound to stimuli from the type of female previously encountered during the social experience regimen. It is proposed that the previously reported lack of ultrasound emission by males to chemosignals from hypophysectomized females can be accounted for via straightforward associative learning mechanisms. Conditioning Ultrasound

Hypophysectomy Urine

Learning

Mouse

A D U L T male mice which have never encountered a female mouse after weaning rarely emit ultrasonic vocalizations to urinary chemosignals from females, yet emit many "courtship" ultrasounds upon first exposure to an intact adult female mouse. However, following experience with a female, adult male mice readily emit ultrasonic vocalizations to both female urine and female-soiled cage shavings [7, 9, ll]. Apparently adult males learn the signal value of female urine as a direct consequence of experience with females [3]. Consistent with this learning interpretation are findings that: (1) repeated exposures to female urine, in the absence of a female, are not sufficient to induce ultrasonic vocalizations to female urine (therefore the response acquisition cannot be attributed to sensitization) [5]; (2) the initially vigorous response to female urine following exposure to adult females "extinguishes" over repeated trials with urine alone [3]; and (3) a normally neutral synthetic chemosubstance (Wild Musk Spray, Coty) can acquire ultrasound-eliciting potency if paired with an adult female stimulus animal [6]. Thus, when viewed in terms of classical conditioning [4], rather than displaying the functional properties of unconditioned stimuli (US), urinary cues from females seem to act as conditioned stimuli (CS) with regard to their potency to elicit ultrasonic vocalizations from males. Following an exposure to adult conspecifics, adult male mice sometimes also differentiate among urines from various conspecific classes. Males emit more ultrasonic vocalizations in the presence of urine from adult females than they do in the presence of urine from either adult males or from immatures of either sex [10]. Although no specific female

Pheromone

Pituitary

Reproduction

urinary chemosubstance(s) responsible for ultrasound elicitation has been discovered, the pituitary gland has recently been implicated. Urine and soiled cage shavings collected from hypophysectomized females was found to be ineffective in eliciting ultrasounds from experienced males [8]. Further, the decreased ultrasounding to these stimuli was not simply attributable to post-hypophysectomy diabetes insipidus-related dilution of ultrasound-eliciting factors. One interpretation that has been offered of these findings is that the presence of some pituitary-dependent urinary chemosubstance(s) is necessary for female urine to acquire ultrasound eliciting capability [8,10]. However, an alternative, associative explanation for the lack of ultrasounding by socially experienced males to chemosignals from hypophysectomized females is also possible. In the prior experiments normal, intact females were employed to provide the initial social experience to the male test subjects. Thus, the subsequent ultrasonic vocalizations emitted to normal, intact females' urine but not to hypophysectomized females' urine could represent a straightforward conditioned discrimination. Perhaps males ultrasounded to the intact females' urine (CS) because it had been paired with a female (US), but failed to ultrasound to hypophysectomized females' urine because it had not been paired with a female. Since mice are capable of discriminating between conspecifics of the same age, sex, and strain via chemical cues [1] conditioned discriminative ultrasonic responding to even subtle urinary differences among females is not unlikely. In order to differentiate between these interpretations, we gave male mice experience with either intact or hypophysectomized females and then tested them for ultrasonic respond-

~Requests for reprints should be addressed to G. Whitney.

Copyright © 1983 Pergamon Press Ltd.--0031-9384/83/090269-04503.00

270

MAGGIO, MAGGIO A N D W H I T N E Y

ing to chemical cues from both. If males emit ultrasonic vocalizations to chemical cues from the female type previously encountered, then the results should be quite different than those predicted under the supposition that some pituitary-dependent metabolite is necessary for female urine to acquire ultrasound eliciting potency. METHOD

Animals Twenty DBA/2J inbred male mice (Mus musculus) from Jackson Laboratory (Bar Harbor, ME), that arrived in our laboratory at 50_+3 days of age, served as subjects. They were 78_+3 days of age on the first day of the social experience regime. Social experience/stimulus donor animals consisted of 3 adult C57BL/6J male mice, 3 intact C57BL/6J adult female mice (all from Jackson Laboratory, Bar Harbor, ME), and 3 hypophysectomized C57BL/6 adult female mice (from Charles River, Willmington, MA). The latter females were hypophysectomized by the supplier at 55 days of age and arrived in our laboratory at 60 days of age. Social experience/stimulus donor animals were 118-151-+3 days of age at the beginning of the experiment.

Apparatus Ultrasounds were monitored with a QMC Bat Detector (Model S 100, QMC Instruments, London) maintained in the tuned mode, broadband trim fully off, and set at a frequency of 70 kHz. Volume was kept fully on during all monitoring. Ultrasonic signals were received via a microphone positioned 13 cm above the wire top of a 29× 18× 13 cm transparent plastic test chamber containing a 1-2 cm layer of clean wood shavings. Stimulus urine was collected via plastic metabolic cages (Maryland Plastics, E 110-Metabolism Unit).

Procedure On the day of arrival all animals were individually housed in transparent 29 x 18 × 13 cm plastic cages with wood shavings for bedding and a wire mesh top supporting a water bottle and food. Food and water were provided ad lib. All animals were maintained on a 12:12 L/D cycle, with all socialization and testing occuring in the light portion. Social experience. F o r all subjects social experience consisted of eight consecutive days of a 3-min/day exposure to a male and to a female conspecific. On each day each subject was removed from the main colony and carried in his home cage to a separate room. Either a male or a female stimulus animal was then introduced into the subject's home cage for a 3-min period. At the end of that time the first stimulus animal was removed and a second stimulus animal of opposite sex was immediately placed in the subject's cage for three more minutes. Upon termination of the second 3-min period the stimulus animal was removed and the subject was returned to the main colony and left undisturbed until the next day. For half the subjects (Group I) the female stimulus animals encountered during social experience were always intact, normal females. The remaining subjects (Group H) received social experience with hypophysectomized females. The same intact males were employed in the social experience of both groups. A Latin Square design was employed to systematically counterbalance, across groups and subjects within days and reversed on each day,

the sex of the stimulus animal first presented during social experience. The social experience regimes were also counterbalanced so that throughout the 8-day period all subjects were exposed to different male and female stimulus animals across three consecutive days before reencountering an individual stimulus animal. On Day 8 all subjects were briefly placed under the bat detector microphone to determine if they were producing ultrasounds differentially to males versus females. All subjects in both groups emitted ultrasound to the stimulus female, while none emitted ultrasound to stimulus males. Testing. Each subject was monitored for ultrasonic vocalizations in the presence of (Phase 1) urines or distilled water presented on cotton swabs, and (Phase 2) soiled cage shavings or clean, control cage shavings. Phase 1. On the day following completion of social experience (86___3 days of age) all subjects were given their first test in response to either (1) intact female urine, (2) hypophysectomized female urine, or (3) distilled water. A within-subjects experimental design was employed with the order of stimulus presentation counterbalanced across subjects according to a Latin Square design. Trials were separated by 24 hr so that over 3 consecutive days each subject was tested once with each of the 3 stimulus types. Twelve hours prior to each day's testing, the stimulus donor animals were placed in metabolic cages according to stimulus type (n=3/cage). Intact female urine and hypophysectomized female urine was then collected for a 12 hr period. Urine was removed from the metabolic cage collecting vials with a syringe immediately preceding that d a y ' s testing. All stimulus preparations occurred outside the testing room, wtih each stimulus prepared within seconds of delivery into the test chamber. F o r each subject tested, 0.1 ml of the appropriate urine was injected onto a cotton swab. Control stimuli were also prepared just prior to use by injecting 0.1 ml of distilled water onto swabs. The wood part of the swab touched by the experimenter was broken off and the stimulus-containing cotton tip was briefly stored in a clean test tube. Each subject was individually placed in a clean test chamber, transported to the testing room, and placed under the microphone. If any ultrasound occurred during a l-min habituation period, the stimulus was not presented until two full minutes of nonultrasounding had elapsed. If no ultrasounds occurred for a I-min habituation period, the appropriate stimulus was introduced into the center of the test chamber. Ultrasounding rarely occurred during habituation. After each subject had met the habituation criteria, the swab stimulus was " p o u r e d " from the test tube into the center of the test chamber, and the test was begun. UItrasounding was then monitored for a 3-min period and quantified by counting the number of 5-sec intervals containing at least one ultrasonic vocalization. Thus, no ultrasounding resulted in a score of 0 while ultrasounding in every 5-sec interval resulted in a maximum score of 36. The person monitoring ultrasounds was always kept uninformed as to which stimulus type was presented. After each trial the subject was housed in a clean cage and returned to the colony until the next test. On completion of Phase 1, subjects were placed in fresh cages, returned to the main colony, and left undisturbed (aside from general food and water maintenance) until Phase 2 testing. Phase 2. Nineteen days after their last test in Phase 1 all 20 subjects began testing in Phase 2. The experimental design, habituation procedures, and ultrasound quantification

U L T R A S O N I C VOCALIZATIONS

16

0 +1 .J

12

~

{E m ~

0 7

Z

Z

<0

16

SWAB STIMULUS [ ] INTACT ~ URINE [ ] HYPOPHYSECTOMIZED URINE I~1 H20

(/:1

0

271

I

CAGE SHAVING STIMULUS [ ] I N T A C T ~ SOILED

0 -J

+t

COZ~ 12 O D "' O u ) ¢/~ u.

I-..-

in

_

7

7

[ ] HYPOPHYSECTOMIZED 9. SOILED [ ] FRESH

<0 w 0

-r

H

SOCIAL EXPERIENCE GROUP

T

H

SOCIAL EXPERIENCE GROUP

FIG. 1. Amount of ultrasound (mean_+SE) emitted by males socially experienced with intact (Group I) or hypophysectomized (Group H) females in response to urine from intact females ~ , urine from hypophysectomized females [], and distilled water [2.

FIG. 2. Amount of ultrasound (mean_+SE) emitted by males socially experienced with intact (Group I) or hypophysectomized (Group H) females in response to intact female-soiled~l, hypophysectomized female-soiled ~, and fresh [] cage shavings.

were all identical to that employed in Phase 1. However, instead of cotton swab stimuli, l0 ml of either (l) intact female-soiled, (2) hypophysectomized female-soiled, or (3) fresh cage shavings were poured through the wire mesh top into the center of the test chamber. Fourteen days prior to the initial Phase 2 test session, each of the female stimulus donors was placed in a cage with clean bedding. Stimulus donors were then left undisturbed until the first day of Phase 2 testing when, just prior to the first session, the soiled cage shavings were removed from the donors' cages. Cage shavings soiled by the 3 intact female donors were pooled together, well mixed, and stored in plastic resealable bags. Soiled shavings from the 3 hypophysectomized female donors were treated similarly. A comparable quantity of fresh, unsoiled shavings was also stored. Statistical analyses. For each of the two social experience groups (I and H), within-group comparisons of the amount of ultrasound emitted to the three stimulus types in each phase were performed via one-tailed matched pairs t-tests [2].

Also shown in Fig. l, subjects having prior social experience with intact females (Group I) emitted more ultrasound to both urine from hypophysectomized females, t(9)=2.41, p<0.025, and to urine of intact females, t(9)=2.62, p<0.025, than to distilled water. Group I subjects also emitted nominally more ultrasound to urine of intact females than to urine of hypophysectomized females (mean intact=6.8, mean hypox--4.0), but again, this difference was not statistically significant, t(9)= 1.34, p>0.05. Subjects having prior social experience with hypophysectomized females (Group H) emitted significantly more ultrasound to cage shavings soiled by hypophysectomized females than they did to fresh cage shavings, t(9)=2.46, p<0.025 (Fig. 2). Although subjects of group H emitted over twice as many ultrasounds in the presence of cage shavings soiled by hypophysectomized females than they emitted in the presence of cage shavings soiled by intact females (mean h y p o x = l l . 4 , mean intact=4.6), this difference was not statistically significant, t(9)= 1.56, p>0.05. For group H subjects, soiled cage shavings from intact females marginally differed from fresh unsoiled cage shavings in ultrasound elicitability, t(9)=2.05, p<0.05. Also illustrated in Fig. 2, subjects having prior social experience with intact females (Group I) emitted significantly more ultrasound to cage shavings soiled by intact females than they did to either fresh cage shavings, t(9)=3.18, p<0.01, or to cage shavings soiled by hypophysectomized females, t(9)=3.11, p<0.01. The amount of ultrasound emitted by group I subjects to cage shavings soiled by hypophysectomized females did not differ from the amount emitted to fresh cage shavings, t(9)= 1.50, p>0.05.

RESULTS

As shown in Fig. l, subjects having prior social experience with hypophysectomized females (Group H) emitted significantly more ultrasonic vocalizations to a swab containing urine from hypophysectomized females than they did to a control swab containing distilled water, t(9)=3.76, p<0.005. Further, for males of group H there was no significant difference between the amount of ultrasound to water control swabs versus swabs containing urine from intact females, t(9)= 1.81, p>0.05. Thus, urinary stimuli from hypophysectomized females, but not from intact females, were more effective than the water control in eliciting ultrasounds from males whose prior social experience was with hypophysectomized females. Although group H subjects emitted more ultrasounds to urine from hypophysectomized females than to urine of intact females (mean hypox=12.5, mean intact=8.0), the difference was not statistically significant, t(9)= 1.3 l, p>0.05.

DISCUSSION

The results of the above experiment a r e consistent with the suggestion that the male mouse ultrasonic vocalization response to female urinary substances is not under strict control of pituitary-mediated metabolites of females. The results of Phase 1 demonstrated that either intact or

272

MAGGIO, MAGGIO AND WHITNEY

hypophysectomized females' urine was a more potent stimulus in eliciting ultrasound from appropriately socially experienced males than was distilled water. Although a social-experience based discrimination by males of female urine types was not observed within Phase 1, the finding that males emitted more ultrasound to hypophysectomized female urine than to a neutral stimulus (i.e., water) indicates that hypophysectomized females' urine can acquire ultrasound eliciting potency. Results from Phase 2 support this interpretation. In Phase 2, males which had received social experience with intact females emitted more ultrasound to soiled cage shavings from intact females than to cage shavings soiled by hypophysectomized females. This finding replicates previous reports [8,10] demonstrating that hypophysectomized female-soiled cage shavings are weak in eliciting ultrasound from males that are socially experienced with intact females. Further, Phase 2 also showed that hypophysectomized females' soiled cage shavings were no better in eliciting ultrasounds from group I males than was the "neutral" fresh cage shaving stimulus. Thus, group I males specifically emitted ultrasounds to the type of female-soiled cage shavings congruent with their prior experience. Results from Phase 2 also demonstrate that if the social experience regimen is one in which males encounter hypophysectomized females, then males emit significantly more ultrasounds to hypophysectomized females' soiled cage shavings than they do to neutral cage shavings. Cage shavings soiled by intact females also elicited some ultrasound from group H males; the amount was nominally intermediate between that elicited by hypophysectomized females' soiled cage shavings and the neutral fresh cage shavings.

Although social experience based discriminatory ultrasounding to the different female chemosubstance types within groups was observed only in Phase 2 for males experienced with intact females, the mean amount of ultrasound emitted by males of any of the social experience groups was always nominally higher for the specific type of stimulus which served as a CS during the social experience conditioning trials. Thus, considering two social experience groups, two phases and three stimulus types, the random probability of males emitting the highest mean amount of ultrasound always to the specific stimulus type encountered during social experience is extremely low (p=(I/3p=0.012). Although extrapolation to considerations of the relative importance in ecological contexts of pituitary-mediated metabolites may be limited by the "artificiality" of hypophysectomized female stimuli, two conclusions can be suggested from these data. First, the absence of pituitary metabolites from female urine reduces the stimulus value (i.e., ultrasound-eliciting potency) of female urine for males only if the males were previously conditioned with intact females. Second, an intact pituitary is not necessary in order for a female's chemosubstances to serve as a conditioned stimulus for male ultrasonic emissions. Rather, it appears that the potency of various female chemosubstances in eliciting ultrasounds from a male is contingent on the prior experience of the male. ACKNOWLEDGEMENTS This research was supported, in part, by a Florida State University Psychobiology Fellowship awarded to J,C. M. and by NINCDS grant No. NS 15560.

REFERENCES

1. Bowers, J. M. and B. L. Alexander. Mice: Individual recognition by olfactory cues. Science 158: 1208-1210, 1967. 2. Brown, B. W. and M. Hollander. Statistics: A Biomedical Introduction. New York: John Wiley, 1977. 3. Dizinno, G., G. Whitney and J. Nyby. Ultrasonic vocalizations by male mice (Mus musculus) to female sex pheromone: Experiential determinants. Behav Biol 22:104-113, 1978. 4. Mackintosh, N. J. The Psychology o f Animal Learning. New York: Academic Press, 1974, pp. 730. 5. Nyby, J. and G. Whitney. Experience affects behavioral responses to sex odors. In: Chemical Signals: Vertebrates and Aquatic Invertebrates, edited by D. Muller-Schwartz and R. M. Silverstein. New York: Plenum Press, 1980, pp. 173-192. 6. Nyby, J., G. Whitney, S. Schmitz and G. Dizinno. Postpubertal experience establishes signal value of mammalian sex odor. Behav Biol 22: 545-552, 1978.

7. Nyby, J., C. J. Wysocki, G. Whitney and G. Dizinno. Pheromonal regulation of male mouse ultrasonic courtship (Mus musculus). Anita Behav 25: 333-341, 1977. 8. Nyby, J., C. J. Wysocki, G. Whitney, G. Dizinno and J. Schneider. Elicitation of male mouse (Mus musculus) ultrasonic vocalizations: 1. Urinary cues. J Comp PhyMol Psychol 93: 957-975, 1979. 9. Whitney, G., M. Alpern, G. Dizinno and G. Horowitz. Female odors evoke ultrasounds from male mice. Anim Learn Behav 2: 13-18, 1974. 10. Whitney, G. and J. Nyby. Cues that elicit ultrasounds from adult male mice. Am Zool 19: 457-463, 1979. 11. Whitney, G. and J. Nyby. Sound communication among adults. In: Auditory Psychobiology of the Mouse, edited by J. F. Willot. Springfield, IL. Charles C. Thomas, 1983, pp. 98-129.