HORMONES
AND
BEHAVIOR
26,
512-521 (1992)
Time Course of Androgenic Modulation of Odor Preferences and Odor Cues in Male Meadow Voles, Microtus pennsylvanicus MICHAEL Department
of Psychology,
University
H. FERKIN* of California,
Berkeley,
California
94720
During the breeding season, male meadow voles prefer female over male odors and females prefer male over female odors. Testosterone control of males’ odor preferences and production of odors attractive to females differ. A male meadow vole’s preference for female versus male odor was still evident 1 week after castration, but not 1 week later. This preference was reinstated in testosteronetreated male voles 2 weeks after the onset of hormone replacement. The attractiveness of male odors to females did not disappear until 3 weeks after castration. The attractiveness of male odors was reinstated 1 week after castrated males were treated with testosterone. The time course for the androgenic modulation of production of odors attractive to females may facilitate breeding. For example, at the end of the breeding season males may emit an odor that is still attractive to females. Similarly, at the beginning of the breeding season males may emit an odor that is attractive to females. 0 1992 Academic press, Inc.
During the breeding season adult meadow voles, Microtus pennsylvanover same-sex conspecifics (Ferkin and Seamon, 1987). In males preferences for the odors of females and emission of odors that are attractive to females depend on testosterone concentrations characteristic of the breeding season (Ferkin and Gorman, 1992; Fe&in, Gorman, and Zucker, 1992). Yet, the time course of the effect of castration and of replacement therapy on these processes is not known. It remains possible, for example, that disappearance of male preferences for female odors and loss of attractiveness of male odors to females occur at markedly different rates after castration. In this study I determined the time course of effects in odor preferences of male voles and attractiveness of male odors to females after castration and testosterone replacement. This information is important for two reasons: (1) The odors of conspecifics are predictive of types of social interactions and result in preferences for odors of conspecifics with whom icus, are attracted to opposite-sex
’ Present address: Department of Psychology, Cornell University, Ithaca, NY 14853. 512 001&506X/92 $4.00 Copyright All rights
0 1992 by Academic Press. Inc. of reproduction in any form reserved.
TESTOSTERONE
AND ODOR COMMUNICATION
523
voles have had amicable, affiliative, and sexual encounters (Ferkin and Seamon, 1987). (2) Male odors are sufficient to induce sexual maturation (Clulow and Mallory, 1970; Baddaloo and Clulow, 1981) and may induce behavioral estrus and facilitate breeding and ovulation in female meadow voles. METHODS Animals
Meadow voles maintained from birth in a long photoperiod that provided 14 hr light/day (lights on at 07.00 hr) were weaned at 18 days of age, housed with littermates until 35 days of age, and singly thereafter in opaque plastic cages (13 x 16 x 31 cm). Wood shavings, cotton nesting material, food (Purina mouse chow No. 5015, Ralston Purina Co., St. Louis, MO) and water were provided ad Zibitum and room temperature was maintained at 22 + 2°C. Cages were cleaned twice weekly. Cotton nesting material was replaced 24 hr before odor testing. Apparatus
Odor preferences were tested in a covered Plexiglas Y-maze (149 x 9 x 13 cmj which consisted of a 15-cm acclimation area at the stem of the Y, a common arm 73 cm in length, and two choice arms each 76 cm long. This apparatus has been detailed elsewhere (Ferkin and Seamon, 1987). Briefly, odor stimuli were placed inside wire baskets in 15-cm chambers at the distal end of each choice arm. These baskets allowed air passage but prevented direct contact of the vole with the cotton. A fan located behind the acclimation area pulled air through the Y-maze toward the vole. Air movement measured with a hot-wire anemometer was 0.60 m/set through the common arm and 0.30 m/set through the choice arms. Preference Test
Voles were placed in the acclimation area behind a moveable screen door for 1 min before the door was raised and they were exposed to two air currents from the separate arms of the maze. The odor stimuli in the maze arms were provided by cotton nesting material taken from a cage that housed a single animal for the preceding 24 hr. Scented cotton was always taken from individuals of similar age as the focal vole. Odor donors and odor receivers were unrelated to and unfamiliar with each other. Each of the two wire baskets contained cotton impregnated with one of the stimulus odors placed in the right or left position at random. Cotton pieces containing stimulus odors were of similar size and weighed 10 g. The odor assignment of the stimulus odors was unknown to the observer. The Y-maze was cleaned with 70% ethanol between trials. Voles were free to move throughout the Y-maze. During each 5-min
514
MICHAEL
H. FERKIN
trial the initial odor choice of the vole and the time spent in the choice arms were recorded to the nearest second. Wilcoxon matched-pairs signedranks tests were used to assess the significance of differences between the paired sets of observations (two-tailed, a = 0.05). Voles in a given treatment condition were considered to display a preference for an odor if the animals spent significantly more time in the maze arm containing that odor. Surgical Procedure
Males were anesthetized with a ketamine “cocktail” (100 mg ketamine hydrochloride/ml, 20 mg xylazine/ml, and 10 mg acetylpromazine/ml); 0.04 ml of cocktail was injected ip per 10 g of body weight. Gonadectomies were performed via a midline incision; blood vessels associated with the testes were ligated, the testes excised, and the abdominal wall and skin closed separately with sterile sutures and treated with 0.2% nitrofurazone (Furacin). Sham surgery was identical to the gonadectomy procedure, except that the testes were exposed but not excised. Discomfort attendant to surgery was treated by administration of 1 .O ml codeine-acetaminophen solution in the drinking water for the first l-4 days postoperatively. Hormone
Replacement
Hormone capsules were constructed from Silastic tubing (Dow Corning, Midland, MI; od 0.077 in., id 0.058 in.) cut into 20-mm lengths and filled with crystalline testosterone (T; Sigma Chemical Co., St. Louis, MO). The ends were plugged with 7.5 mm-long dowels and sealed with silicon rubber cement, leaving 5-mm lengths of testosterone-filled tubing. Blank (BL) capsules were similar to T capsules but were empty. BL and T capsules were incubated separately in saline for 24 hr before insertion into the voles. Capsules were placed subcutaneously in the interscapular area via a small incision that was closed with a wound clip and treated with Furacin. Voles were anesthetized with methoxyflurane vapors (Metofane, Pitman Moore, Washington, NJ) prior to implantation of the capsule. Experimental Latency
Procedures to the loss of males’ preference
for female
over male odor.
Thirty-two sexually inexperienced male voles, between 90-120 days of age, were gonadectomized (n = 16) or underwent a sham procedure (n = 16). One week after surgery, and thereafter at weekly intervals, each male was tested in the Y-maze for preferences between female and male odors. In these tests, intact male and parous female voles, between 90-120 days of age, served as odor donors. Female odor donors were not pregnant or lactating. In each test odor donors and receivers were unrelated and unfamiliar to each other. There were different female and male odor
TESTOSTERONE
AND
ODOR
515
COMMUNICATION
donors for each male in this and subsequent tests. The odor tests were terminated when gonadectomized males (treatment group) no longer displayed a preference for female versus male odor. Persistence
of emission
of castrated
males’ odors attractive
to females.
One week after males were castrated, and thereafter at weekly intervals, adult female voles (n = 16), between 90-120 days of age, were tested in the Y-maze for preferences between gonadectomized male and shamtreated male odors. There were different male donors for each female in this and subsequent tests. The odor preference tests were terminated when female voles (treatment group) no longer displayed a preference for shamtreated versus gonadectomized male odor. Effect of hormone replacement on odor preferences. Four weeks after surgery, gonadectomized and sham-treated males used in the odor preference and odor cue emission experiments received testosterone (gonadectomized male + T, n = 13) or blank (sham-treated male + BL, n = 13) capsules, respectively. One week after surgery, and thereafter at weekly intervals, each male was tested in the Y-maze for preferences between female and male odors. In these tests, intact male and female voles, between 90-120 days of age, served as odor donors. In each test odor donors and receivers were unrelated and unfamiliar. The odor tests were terminated when gonadectomized males + T (treatment group) displayed a preference for female over male odor. Effect of hormone replacement on male odor emissions. One week after males were operated on, and thereafter at weekly intervals, adult female voles (n = 16), not used in the odor cue emission experiment and between 90-120 days of age, were tested in the Y-maze for preferences between ganadectomized male + T (n = 13) and sham-treated male + BL (n = 13) odors. The female odor receivers were not pregnant or lactating. The odor preference tests were terminated when female voles (treatment group) no longer displayed a preference for sham-treated male f BL over gonadectomized male + T odor. RESULTS Gonadectomized and sham-treated display a left- or right-side preference experiments). Initial choice of an odor ferred odor across ail tests. All voles
males and intact females did not (x2 tests, df = 1, P > 0.10, in all coincided with the eventually prealways investigated both odors. Latency to the loss of males’ preference for female over male odor. One week after surgery, gonadectomized and sham-treated males preferred female to male odor (Fig. l), but no such preference was detectable 2 weeks after gonadectomy (Fig. 1). Sham-gonadectomized males continued to prefer female over male odor on each of three weekly tests (Fig. 1). Persistence
of emission
The odors emitted
of castrated
by gonadectomized
males’ odors
attractive
and sham-treated
to females.
males 1 and 2
516
MICHAEL 300G bi 8 8 E
H. FERKIN GX MALES
A
Cl -
FEMALE MALE
c7 m
FEMALE MALE
250 200 -
*
150-
$ 100 fj 5o1 owk 1
‘I
wk 2 SHAM
wk 1
wk 2
wk 3
MALES
wk 3
FIG. 1. Mean time (set) + SE that gonadectomixed (GX) male voles (top) or shamgonadectomixed males (sham males, bottom) spent with female odors vs male odors. Tests were carried out at weekly intervals after surgery on male voles. N = 13 male voles for each paired comparison. Asterisk (*) indicates ‘significant difference between odor pairs, Wilcoxon matched-pairs signed-ranks test at P < 0.05 in this and subsequent figures.
weeks after surgery were equally attractive to adult females (Fig. 2). Three weeks after surgery, odors emitted by sham-treated males were more attractive to adult females than those of gonadectomized males (Fig. 2). Effect of hormone replacement on odor preferences. One week after receiving the hormone implant, gonadectomized + T males did not demonstrate a preference for female over male odors (Fig. 3), whereas sham + BL males preferred female to male odor (Fig. 3). Two weeks after hormone implantation, gonadectomized + T and sham + BL males preferred female to male odor (Fig. 3). Effect of hormone replacement on male odor emissions. One week after hormone implantation, the odors emitted by gonadectomized + T males were more attractive than those of sham + BL males to adult females
TESTOSTERONE
FEMALES
300 G&i? lY
51?
AND ODOR COMMUNICATION
z
SGlGt4ALE
250 200
i F 5 LLi r F $
150 100 5o 0 wk 1
wk 2
wk3
2. Mean time that adult intact females spent with gonadectomized and sham-gonadectomiz.ed male odors. Tests were carried out at weekly intervals after surgery on male voles. N = 16 female voles for each comparison. FIG.
(Fig. 4). In a retest 1 week later, females remained more attracted gonadectomized + T th;m sham + BL male odor (Fig. 4).
to
DISCUSSION
The time course of androgenic modulation of male odor preferences and of emission of female-attractant odors was similar in meadow voles to patterns seen in other species of rodents. In male rats and golden hamsters, as in meadow voles, preferences for female over male odors disappeared 2 weeks after castration. A corresponding interval of 2-3 weeks also elapsed before preferences for female odors was reinstated in males treated with androgens (Hart, 1974; Gregory, Engel, and Pfaff, 1975; Merkx, 1984; Powers, Bergondy, and Matochik, 1985). In addition, the attractiveness of male odors to female voles was lost 3 weeks after castration and reestablished 1 week after testosterone replacement. These tatencies are similar to those demonstrated for female puberty-accelerating odors produced by male house mice. The latter disappear 2 weeks after castration and are reinstated 5-9 days after replacement with testosterone propionate (Lombardi, Vandenbergh, and Whitsett, 1976). The greater attractiveness of odors of testosterone-treated compared with intact males did not correspond to results of a previous study, in which these groups did not differ (Ferkin et al., 1992). The differences between the two studies in the attractiveness of male odors may be due to the 3-week androgen-free interval after castration in this study compared to 6 weeks in the study by Ferkin et al. (1992). In the present study castrated males received hormone capsules containing 5 mm active length of testosterone. Less hormone was needed to reinstate attractiveness of male odors to females. In Ferkin et al. (1992) castrated males received
518
MICHAEL
H. FJZRKIN GX + T MALES
wk 1
Cl -
FEMALE MALE
G
EEM;LE
wk 2 SHAM
+ EL MALES
wk 1 wk 2 3. Mean time that sham-gonadcctomized males implanted with a blank capsule (top, sham + BL males) and gonadectomized males treated with testosterone (bottom, GX + T) spent with male odors vs female odors. Tests were carried out at weekly intervals after males received a hormone capsule. N = 13 male voles for each comparison. FIG.
implants containing 10 mm of testosterone. More hormone was needed to restore the attractiveness of male odors to females. It is well established for other male rodents that investigation of female odors, scent marking, and sexual behavior are more readily stimulated with a lower dosage of testosterone if the androgen-free interval is short and hormone replacement is immediate (Damassa, Smith, Tennent, and Davidson, 1977; Yahr, 1983; Powers et al., 1985). The slower rates of decline of the attractiveness of male odors after castration, and the more rapid reinstatement after testosterone replacement, may be functionally significant. At the end of the breeding season, when testosterone concentrations are decreasing (Adams, Tamarin, and Callard, 1980) males may still be able to emit odors attractive to females and may thereby extend the breeding season. Most adult male meadow
TESTOSTERONE
FEMALES
300 G is -
250
F
150
519
AND ODOR COMMUNICATION z
gzA+MT
MALE
I
5 z F
100
$
5o 0 wk 1
wk 2
FIG. 4. Mean time that females spent with GX f T male and sham + BL male odors. Tests were carried out at weekly intervals after male voles received a testosterone-filled capsule. N = 16 female voles for each comparison.
voles do not readily join over-wintering groups and fail to survive to reproduce the next spring (McShea, 1990); breeding is mostly carried out by males that have overwintered as juveniles (Mihok, 1984). At the beginning of the breeding season, increasing endogenous testosterone concentrations allow males to reach sexual maturity (Adams et al., 1980). The rapid response to testosterone also permits males to produce odors that are attractive to females and thereby induce sexual maturation, behavioral estrus, ovulation, and facilitate breeding (Clulow and Mallory, 1970; Baddaloo and Clulow, 1981), as well as blocking pregnancy (Mallory and Clulow, 1977). Differences between the time course of androgenic modulation of male odor preferences and emissions of odors attractive to females may reflect differences in substrate or activational thresholds. Castration may differentially affect male odor preferences and odor production by altering the number and nature of activity in androgen receptors and aromatase activity. In rats, golden hamsters, and gerbils castration differentially affects several components of scent marking and sexual behavior that depend on androgens (Hart, 1974; Damassa et al., 1977; Roselli and Resko, 1984; Sanborn, Wagle, and Steinberger, 1984; Powers et al., 1985; Yahr and Stephens, 1987). Differences in the time course of the effects of castration and replacement therapy on these two odor-related phenomena also may reflect different sites of neural control and sensitivities to testosterone. The preoptic area and sexually dimorphic areas in the hypothalamus control scent marking in rats (Ferris, Gold, DeVries, and Potegal, 1990) and gerbils (Yahr and Stephens, 1987), whereas the attractiveness of odors to opposite-sex conspecifics may be mediated at sites in the periphery, possibly at the specific sources of these odors.
520
MICHAEL
H. FERKIN
ACKNOWLEDGMENTS I thank E. Adkins-Regan, M. Gorman, R. Johnston, and I. Zucker for comments on earlier drafts of the manuscript, G. Brooks, E. Moore, and D. Starks for excellent animal care, M. Gorman, C. Kudowicz, K. Pelz, and C. Tuthill for technical assistance. This research was supported by NIH Postdoctoral Fellowship 532NS08792 to M.H.F. and NIH Grant HD 02982.
REFERENCES Adams, M. R., Tamarin, R. H., and Callard, I. P. (1980). Seasonal changes in plasma androgen levels and the gonads of the beach vole, Micro&s breweri. Gen. Camp. Endocrinol. 41, 31-40. Baddaloo, E. G. Y., and Clulow, F. V. (1981). Effects of the male on growth, sexual maturation, and ovulation of young female meadow voles, Microtus pennsylvanicus. Can. J. 2001. 59, 415-521. Clulow, F. V., and Mallory, F. F. (1970). Oestrus and induced ovulation in the meadow vole, Microtus pennsylvanicus. J. Reprod. Fertil. 23, 341-343. Damassa, D. A., Smith, E. R., Tennent, B., and Davidson, J. M. (1977). The relationship between circulating testosterone levels and male sexual behavior in rats. Horm. Behav. 8, 275-286. Ferkin, M. H., and Seamon, J. 0. (1987). Odor preferences and social behavior in meadow voles, Microtus pennsylvanicus: Seasonal differences. Can. J. Zool. 65, 2931-2937. Fe&in, M. H., and Gorman, M. R. (1992). Photoperiod and gonadal hormones influence odor preferences of the male meadow vole, Microtus pennsylvanicus. Physiol. Behav. 51, 1087-1091. Fe&in, M. H., Gorman, M. R., and Zucker, I. (1992). Gonadal hormones influence odors emitted by male meadow voles (Microtus pennsylvanicus). J. Reprod. Fertil. 95, 729736. Ferris, C. F., Gold, L., DeVries, G. J., and Potegal, M. (1990). Evidence for a functional and anatomical relationship between the lateral septum and the hypothalamus in the control of flank marking behavior in golden hamsters. J. Comp. Neurol. 293,476-485. Gregory, E., Engel, K., and Pfaff, D. (1975). Male hamster preference for odors of female hamster vaginal discharges: Studies of experiential and hormonal determinants. J. Cornp.
Physiol.
Psychol.
89, 442-446.
Hart, B. L. (1974). Gonadal androgens and sociosexual behavior of male mammals: A comparative analysis. Psychol. Bull. 81, 383-400. Lombardi, J. R., Vandenbergh, J. G., and Whitsett, J. M. (1976). Androgen control of the sexual maturation pheromones in house mouse urine. Biol. Reprod. 15, 179-186. Mallory, F. F., and Clulow, F. V. (1977). Evidence of pregnancy failure in the wild meadow vole, Micro&s pennsylvanicus. Can. J. Zool. 55, l-17. McShea, W. J. (1990). Social tolerance and proximate mechanisms of dispersal among winter groups of meadow voles, Microtus pennsylvanicus. Anim. Behav. 39, 346-351. Me&x, J. (1984). Effect of castration and subsequent substitution with testosterone, dihydrotestosterone and oestradiol on sexual preference behaviour in the male rat. Behav. Brain
Res. 11, 59-65.
Mihok, S. (1984). Life history profiles of boreal meadow voles (Microtzu pennsylvanicus). In J. F. Merritt (Ed.), Winter Ecology of Small Mammals, Special Publication 10, pp. 91-102. Carnegie Mus. Nat. Hist., Pittsburgh, PA. Powers, J. B., Bergondy, M. L., and Matochik, J. A. (1985). Male hamster sociosexual behaviors: Effects of testosterone and its metabolites. Physiol. Behav. 35, 607-616. Roselli, C. E., and Resko, J. A. (1984). Androgens regulate brain aromatase activity in adult male rats through a receptor mechanism. Endocrinology 114, 2183-2189.
TESTOSTERONE
AND ODOR COMMUNICATION
521
Sanborn, B. M., Wagle, J. R., and Steinberger, A. (1984). Control of androgen cytosol receptor concentrations in sertoli cells: Effect of androgens. Endocrinofogy 114,2388-
2393. Yahr, P. (1983). Hormonal influences on territorial marking behavior. In B. B. Svare (Ed.), Hormones and Aggressive Behavior, pp- 14.5-175. Plenum, New York. Yahr, P., and Stephens, D. R. (1987). Hormonal control of sexual and scent marking behaviors of male gerbils in relation to the sexually dimorphic area of the hypothalamus. Horm.
Behav.
21, 331-346.
AND
BEHAVIOR
26,
512-521 (1992)
Time Course of Androgenic Modulation of Odor Preferences and Odor Cues in Male Meadow Voles, Microtus pennsylvanicus MICHAEL Department
of Psychology,
University
H. FERKIN* of California,
Berkeley,
California
94720
During the breeding season, male meadow voles prefer female over male odors and females prefer male over female odors. Testosterone control of males’ odor preferences and production of odors attractive to females differ. A male meadow vole’s preference for female versus male odor was still evident 1 week after castration, but not 1 week later. This preference was reinstated in testosteronetreated male voles 2 weeks after the onset of hormone replacement. The attractiveness of male odors to females did not disappear until 3 weeks after castration. The attractiveness of male odors was reinstated 1 week after castrated males were treated with testosterone. The time course for the androgenic modulation of production of odors attractive to females may facilitate breeding. For example, at the end of the breeding season males may emit an odor that is still attractive to females. Similarly, at the beginning of the breeding season males may emit an odor that is attractive to females. 0 1992 Academic press, Inc.
During the breeding season adult meadow voles, Microtus pennsylvanover same-sex conspecifics (Ferkin and Seamon, 1987). In males preferences for the odors of females and emission of odors that are attractive to females depend on testosterone concentrations characteristic of the breeding season (Ferkin and Gorman, 1992; Fe&in, Gorman, and Zucker, 1992). Yet, the time course of the effect of castration and of replacement therapy on these processes is not known. It remains possible, for example, that disappearance of male preferences for female odors and loss of attractiveness of male odors to females occur at markedly different rates after castration. In this study I determined the time course of effects in odor preferences of male voles and attractiveness of male odors to females after castration and testosterone replacement. This information is important for two reasons: (1) The odors of conspecifics are predictive of types of social interactions and result in preferences for odors of conspecifics with whom icus, are attracted to opposite-sex
’ Present address: Department of Psychology, Cornell University, Ithaca, NY 14853. 512 001&506X/92 $4.00 Copyright All rights
0 1992 by Academic Press. Inc. of reproduction in any form reserved.
TESTOSTERONE
AND ODOR COMMUNICATION
523
voles have had amicable, affiliative, and sexual encounters (Ferkin and Seamon, 1987). (2) Male odors are sufficient to induce sexual maturation (Clulow and Mallory, 1970; Baddaloo and Clulow, 1981) and may induce behavioral estrus and facilitate breeding and ovulation in female meadow voles. METHODS Animals
Meadow voles maintained from birth in a long photoperiod that provided 14 hr light/day (lights on at 07.00 hr) were weaned at 18 days of age, housed with littermates until 35 days of age, and singly thereafter in opaque plastic cages (13 x 16 x 31 cm). Wood shavings, cotton nesting material, food (Purina mouse chow No. 5015, Ralston Purina Co., St. Louis, MO) and water were provided ad Zibitum and room temperature was maintained at 22 + 2°C. Cages were cleaned twice weekly. Cotton nesting material was replaced 24 hr before odor testing. Apparatus
Odor preferences were tested in a covered Plexiglas Y-maze (149 x 9 x 13 cmj which consisted of a 15-cm acclimation area at the stem of the Y, a common arm 73 cm in length, and two choice arms each 76 cm long. This apparatus has been detailed elsewhere (Ferkin and Seamon, 1987). Briefly, odor stimuli were placed inside wire baskets in 15-cm chambers at the distal end of each choice arm. These baskets allowed air passage but prevented direct contact of the vole with the cotton. A fan located behind the acclimation area pulled air through the Y-maze toward the vole. Air movement measured with a hot-wire anemometer was 0.60 m/set through the common arm and 0.30 m/set through the choice arms. Preference Test
Voles were placed in the acclimation area behind a moveable screen door for 1 min before the door was raised and they were exposed to two air currents from the separate arms of the maze. The odor stimuli in the maze arms were provided by cotton nesting material taken from a cage that housed a single animal for the preceding 24 hr. Scented cotton was always taken from individuals of similar age as the focal vole. Odor donors and odor receivers were unrelated to and unfamiliar with each other. Each of the two wire baskets contained cotton impregnated with one of the stimulus odors placed in the right or left position at random. Cotton pieces containing stimulus odors were of similar size and weighed 10 g. The odor assignment of the stimulus odors was unknown to the observer. The Y-maze was cleaned with 70% ethanol between trials. Voles were free to move throughout the Y-maze. During each 5-min
514
MICHAEL
H. FERKIN
trial the initial odor choice of the vole and the time spent in the choice arms were recorded to the nearest second. Wilcoxon matched-pairs signedranks tests were used to assess the significance of differences between the paired sets of observations (two-tailed, a = 0.05). Voles in a given treatment condition were considered to display a preference for an odor if the animals spent significantly more time in the maze arm containing that odor. Surgical Procedure
Males were anesthetized with a ketamine “cocktail” (100 mg ketamine hydrochloride/ml, 20 mg xylazine/ml, and 10 mg acetylpromazine/ml); 0.04 ml of cocktail was injected ip per 10 g of body weight. Gonadectomies were performed via a midline incision; blood vessels associated with the testes were ligated, the testes excised, and the abdominal wall and skin closed separately with sterile sutures and treated with 0.2% nitrofurazone (Furacin). Sham surgery was identical to the gonadectomy procedure, except that the testes were exposed but not excised. Discomfort attendant to surgery was treated by administration of 1 .O ml codeine-acetaminophen solution in the drinking water for the first l-4 days postoperatively. Hormone
Replacement
Hormone capsules were constructed from Silastic tubing (Dow Corning, Midland, MI; od 0.077 in., id 0.058 in.) cut into 20-mm lengths and filled with crystalline testosterone (T; Sigma Chemical Co., St. Louis, MO). The ends were plugged with 7.5 mm-long dowels and sealed with silicon rubber cement, leaving 5-mm lengths of testosterone-filled tubing. Blank (BL) capsules were similar to T capsules but were empty. BL and T capsules were incubated separately in saline for 24 hr before insertion into the voles. Capsules were placed subcutaneously in the interscapular area via a small incision that was closed with a wound clip and treated with Furacin. Voles were anesthetized with methoxyflurane vapors (Metofane, Pitman Moore, Washington, NJ) prior to implantation of the capsule. Experimental Latency
Procedures to the loss of males’ preference
for female
over male odor.
Thirty-two sexually inexperienced male voles, between 90-120 days of age, were gonadectomized (n = 16) or underwent a sham procedure (n = 16). One week after surgery, and thereafter at weekly intervals, each male was tested in the Y-maze for preferences between female and male odors. In these tests, intact male and parous female voles, between 90-120 days of age, served as odor donors. Female odor donors were not pregnant or lactating. In each test odor donors and receivers were unrelated and unfamiliar to each other. There were different female and male odor
TESTOSTERONE
AND
ODOR
515
COMMUNICATION
donors for each male in this and subsequent tests. The odor tests were terminated when gonadectomized males (treatment group) no longer displayed a preference for female versus male odor. Persistence
of emission
of castrated
males’ odors attractive
to females.
One week after males were castrated, and thereafter at weekly intervals, adult female voles (n = 16), between 90-120 days of age, were tested in the Y-maze for preferences between gonadectomized male and shamtreated male odors. There were different male donors for each female in this and subsequent tests. The odor preference tests were terminated when female voles (treatment group) no longer displayed a preference for shamtreated versus gonadectomized male odor. Effect of hormone replacement on odor preferences. Four weeks after surgery, gonadectomized and sham-treated males used in the odor preference and odor cue emission experiments received testosterone (gonadectomized male + T, n = 13) or blank (sham-treated male + BL, n = 13) capsules, respectively. One week after surgery, and thereafter at weekly intervals, each male was tested in the Y-maze for preferences between female and male odors. In these tests, intact male and female voles, between 90-120 days of age, served as odor donors. In each test odor donors and receivers were unrelated and unfamiliar. The odor tests were terminated when gonadectomized males + T (treatment group) displayed a preference for female over male odor. Effect of hormone replacement on male odor emissions. One week after males were operated on, and thereafter at weekly intervals, adult female voles (n = 16), not used in the odor cue emission experiment and between 90-120 days of age, were tested in the Y-maze for preferences between ganadectomized male + T (n = 13) and sham-treated male + BL (n = 13) odors. The female odor receivers were not pregnant or lactating. The odor preference tests were terminated when female voles (treatment group) no longer displayed a preference for sham-treated male f BL over gonadectomized male + T odor. RESULTS Gonadectomized and sham-treated display a left- or right-side preference experiments). Initial choice of an odor ferred odor across ail tests. All voles
males and intact females did not (x2 tests, df = 1, P > 0.10, in all coincided with the eventually prealways investigated both odors. Latency to the loss of males’ preference for female over male odor. One week after surgery, gonadectomized and sham-treated males preferred female to male odor (Fig. l), but no such preference was detectable 2 weeks after gonadectomy (Fig. 1). Sham-gonadectomized males continued to prefer female over male odor on each of three weekly tests (Fig. 1). Persistence
of emission
The odors emitted
of castrated
by gonadectomized
males’ odors
attractive
and sham-treated
to females.
males 1 and 2
516
MICHAEL 300G bi 8 8 E
H. FERKIN GX MALES
A
Cl -
FEMALE MALE
c7 m
FEMALE MALE
250 200 -
*
150-
$ 100 fj 5o1 owk 1
‘I
wk 2 SHAM
wk 1
wk 2
wk 3
MALES
wk 3
FIG. 1. Mean time (set) + SE that gonadectomixed (GX) male voles (top) or shamgonadectomixed males (sham males, bottom) spent with female odors vs male odors. Tests were carried out at weekly intervals after surgery on male voles. N = 13 male voles for each paired comparison. Asterisk (*) indicates ‘significant difference between odor pairs, Wilcoxon matched-pairs signed-ranks test at P < 0.05 in this and subsequent figures.
weeks after surgery were equally attractive to adult females (Fig. 2). Three weeks after surgery, odors emitted by sham-treated males were more attractive to adult females than those of gonadectomized males (Fig. 2). Effect of hormone replacement on odor preferences. One week after receiving the hormone implant, gonadectomized + T males did not demonstrate a preference for female over male odors (Fig. 3), whereas sham + BL males preferred female to male odor (Fig. 3). Two weeks after hormone implantation, gonadectomized + T and sham + BL males preferred female to male odor (Fig. 3). Effect of hormone replacement on male odor emissions. One week after hormone implantation, the odors emitted by gonadectomized + T males were more attractive than those of sham + BL males to adult females
TESTOSTERONE
FEMALES
300 G&i? lY
51?
AND ODOR COMMUNICATION
z
SGlGt4ALE
250 200
i F 5 LLi r F $
150 100 5o 0 wk 1
wk 2
wk3
2. Mean time that adult intact females spent with gonadectomized and sham-gonadectomiz.ed male odors. Tests were carried out at weekly intervals after surgery on male voles. N = 16 female voles for each comparison. FIG.
(Fig. 4). In a retest 1 week later, females remained more attracted gonadectomized + T th;m sham + BL male odor (Fig. 4).
to
DISCUSSION
The time course of androgenic modulation of male odor preferences and of emission of female-attractant odors was similar in meadow voles to patterns seen in other species of rodents. In male rats and golden hamsters, as in meadow voles, preferences for female over male odors disappeared 2 weeks after castration. A corresponding interval of 2-3 weeks also elapsed before preferences for female odors was reinstated in males treated with androgens (Hart, 1974; Gregory, Engel, and Pfaff, 1975; Merkx, 1984; Powers, Bergondy, and Matochik, 1985). In addition, the attractiveness of male odors to female voles was lost 3 weeks after castration and reestablished 1 week after testosterone replacement. These tatencies are similar to those demonstrated for female puberty-accelerating odors produced by male house mice. The latter disappear 2 weeks after castration and are reinstated 5-9 days after replacement with testosterone propionate (Lombardi, Vandenbergh, and Whitsett, 1976). The greater attractiveness of odors of testosterone-treated compared with intact males did not correspond to results of a previous study, in which these groups did not differ (Ferkin et al., 1992). The differences between the two studies in the attractiveness of male odors may be due to the 3-week androgen-free interval after castration in this study compared to 6 weeks in the study by Ferkin et al. (1992). In the present study castrated males received hormone capsules containing 5 mm active length of testosterone. Less hormone was needed to reinstate attractiveness of male odors to females. In Ferkin et al. (1992) castrated males received
518
MICHAEL
H. FJZRKIN GX + T MALES
wk 1
Cl -
FEMALE MALE
G
EEM;LE
wk 2 SHAM
+ EL MALES
wk 1 wk 2 3. Mean time that sham-gonadcctomized males implanted with a blank capsule (top, sham + BL males) and gonadectomized males treated with testosterone (bottom, GX + T) spent with male odors vs female odors. Tests were carried out at weekly intervals after males received a hormone capsule. N = 13 male voles for each comparison. FIG.
implants containing 10 mm of testosterone. More hormone was needed to restore the attractiveness of male odors to females. It is well established for other male rodents that investigation of female odors, scent marking, and sexual behavior are more readily stimulated with a lower dosage of testosterone if the androgen-free interval is short and hormone replacement is immediate (Damassa, Smith, Tennent, and Davidson, 1977; Yahr, 1983; Powers et al., 1985). The slower rates of decline of the attractiveness of male odors after castration, and the more rapid reinstatement after testosterone replacement, may be functionally significant. At the end of the breeding season, when testosterone concentrations are decreasing (Adams, Tamarin, and Callard, 1980) males may still be able to emit odors attractive to females and may thereby extend the breeding season. Most adult male meadow
TESTOSTERONE
FEMALES
300 G is -
250
F
150
519
AND ODOR COMMUNICATION z
gzA+MT
MALE
I
5 z F
100
$
5o 0 wk 1
wk 2
FIG. 4. Mean time that females spent with GX f T male and sham + BL male odors. Tests were carried out at weekly intervals after male voles received a testosterone-filled capsule. N = 16 female voles for each comparison.
voles do not readily join over-wintering groups and fail to survive to reproduce the next spring (McShea, 1990); breeding is mostly carried out by males that have overwintered as juveniles (Mihok, 1984). At the beginning of the breeding season, increasing endogenous testosterone concentrations allow males to reach sexual maturity (Adams et al., 1980). The rapid response to testosterone also permits males to produce odors that are attractive to females and thereby induce sexual maturation, behavioral estrus, ovulation, and facilitate breeding (Clulow and Mallory, 1970; Baddaloo and Clulow, 1981), as well as blocking pregnancy (Mallory and Clulow, 1977). Differences between the time course of androgenic modulation of male odor preferences and emissions of odors attractive to females may reflect differences in substrate or activational thresholds. Castration may differentially affect male odor preferences and odor production by altering the number and nature of activity in androgen receptors and aromatase activity. In rats, golden hamsters, and gerbils castration differentially affects several components of scent marking and sexual behavior that depend on androgens (Hart, 1974; Damassa et al., 1977; Roselli and Resko, 1984; Sanborn, Wagle, and Steinberger, 1984; Powers et al., 1985; Yahr and Stephens, 1987). Differences in the time course of the effects of castration and replacement therapy on these two odor-related phenomena also may reflect different sites of neural control and sensitivities to testosterone. The preoptic area and sexually dimorphic areas in the hypothalamus control scent marking in rats (Ferris, Gold, DeVries, and Potegal, 1990) and gerbils (Yahr and Stephens, 1987), whereas the attractiveness of odors to opposite-sex conspecifics may be mediated at sites in the periphery, possibly at the specific sources of these odors.
520
MICHAEL
H. FERKIN
ACKNOWLEDGMENTS I thank E. Adkins-Regan, M. Gorman, R. Johnston, and I. Zucker for comments on earlier drafts of the manuscript, G. Brooks, E. Moore, and D. Starks for excellent animal care, M. Gorman, C. Kudowicz, K. Pelz, and C. Tuthill for technical assistance. This research was supported by NIH Postdoctoral Fellowship 532NS08792 to M.H.F. and NIH Grant HD 02982.
REFERENCES Adams, M. R., Tamarin, R. H., and Callard, I. P. (1980). Seasonal changes in plasma androgen levels and the gonads of the beach vole, Micro&s breweri. Gen. Camp. Endocrinol. 41, 31-40. Baddaloo, E. G. Y., and Clulow, F. V. (1981). Effects of the male on growth, sexual maturation, and ovulation of young female meadow voles, Microtus pennsylvanicus. Can. J. 2001. 59, 415-521. Clulow, F. V., and Mallory, F. F. (1970). Oestrus and induced ovulation in the meadow vole, Microtus pennsylvanicus. J. Reprod. Fertil. 23, 341-343. Damassa, D. A., Smith, E. R., Tennent, B., and Davidson, J. M. (1977). The relationship between circulating testosterone levels and male sexual behavior in rats. Horm. Behav. 8, 275-286. Ferkin, M. H., and Seamon, J. 0. (1987). Odor preferences and social behavior in meadow voles, Microtus pennsylvanicus: Seasonal differences. Can. J. Zool. 65, 2931-2937. Fe&in, M. H., and Gorman, M. R. (1992). Photoperiod and gonadal hormones influence odor preferences of the male meadow vole, Microtus pennsylvanicus. Physiol. Behav. 51, 1087-1091. Fe&in, M. H., Gorman, M. R., and Zucker, I. (1992). Gonadal hormones influence odors emitted by male meadow voles (Microtus pennsylvanicus). J. Reprod. Fertil. 95, 729736. Ferris, C. F., Gold, L., DeVries, G. J., and Potegal, M. (1990). Evidence for a functional and anatomical relationship between the lateral septum and the hypothalamus in the control of flank marking behavior in golden hamsters. J. Comp. Neurol. 293,476-485. Gregory, E., Engel, K., and Pfaff, D. (1975). Male hamster preference for odors of female hamster vaginal discharges: Studies of experiential and hormonal determinants. J. Cornp.
Physiol.
Psychol.
89, 442-446.
Hart, B. L. (1974). Gonadal androgens and sociosexual behavior of male mammals: A comparative analysis. Psychol. Bull. 81, 383-400. Lombardi, J. R., Vandenbergh, J. G., and Whitsett, J. M. (1976). Androgen control of the sexual maturation pheromones in house mouse urine. Biol. Reprod. 15, 179-186. Mallory, F. F., and Clulow, F. V. (1977). Evidence of pregnancy failure in the wild meadow vole, Micro&s pennsylvanicus. Can. J. Zool. 55, l-17. McShea, W. J. (1990). Social tolerance and proximate mechanisms of dispersal among winter groups of meadow voles, Microtus pennsylvanicus. Anim. Behav. 39, 346-351. Me&x, J. (1984). Effect of castration and subsequent substitution with testosterone, dihydrotestosterone and oestradiol on sexual preference behaviour in the male rat. Behav. Brain
Res. 11, 59-65.
Mihok, S. (1984). Life history profiles of boreal meadow voles (Microtzu pennsylvanicus). In J. F. Merritt (Ed.), Winter Ecology of Small Mammals, Special Publication 10, pp. 91-102. Carnegie Mus. Nat. Hist., Pittsburgh, PA. Powers, J. B., Bergondy, M. L., and Matochik, J. A. (1985). Male hamster sociosexual behaviors: Effects of testosterone and its metabolites. Physiol. Behav. 35, 607-616. Roselli, C. E., and Resko, J. A. (1984). Androgens regulate brain aromatase activity in adult male rats through a receptor mechanism. Endocrinology 114, 2183-2189.
TESTOSTERONE
AND ODOR COMMUNICATION
521
Sanborn, B. M., Wagle, J. R., and Steinberger, A. (1984). Control of androgen cytosol receptor concentrations in sertoli cells: Effect of androgens. Endocrinofogy 114,2388-
2393. Yahr, P. (1983). Hormonal influences on territorial marking behavior. In B. B. Svare (Ed.), Hormones and Aggressive Behavior, pp- 14.5-175. Plenum, New York. Yahr, P., and Stephens, D. R. (1987). Hormonal control of sexual and scent marking behaviors of male gerbils in relation to the sexually dimorphic area of the hypothalamus. Horm.
Behav.
21, 331-346.
