- Email: [email protected]
Pheromones and Novel Male-Induced Pregnancy Disruptions in Mice
Physiology & Behavior, Vol. 66, No. 1, pp. 153–157, 1999 © 1999 Elsevier Science Inc. Printed in the USA. All rights reserved 0031-9384/99/$–see front matter
PII S0031-9384(98)00277-7
Pheromones and Novel Male-Induced Pregnancy Disruptions in Mice: Exposure to Conspecifics Is Necessary for Urine Alone to Induce an Effect DENYS DE CATANZARO,*1 CAMERON MUIR,* CHRISTINE SULLIVAN* AND ALAIN BOISSY† *Department of Psychology, McMaster University, Hamilton, Ontario, L8S 4K1 Canada, and †L.A.H.M., I.N.R.A., Clermont-Ferrand, Theix, 63122, St.-Genès, France Received 3 July 1998; Accepted 28 September 1998 DECATANZARO, D., C. MUIR, C. SULLIVAN AND A. BOISSY. Pheromones and novel male-induced pregnancy disruptions in mice: Exposure to conspecifics is necessary for urine alone to induce an effect. PHYSIOL BEHAV 66(1) 153– 157, 1999.—Previous research indicates a role of pheromones in novel male-induced early pregnancy disruptions. Although some reports suggest that urine alone is sufficient to produce this effect, others raise procedural concerns and fail to replicate such effects. On Days 1 to 5 after insemination, female CF-1 mice had their nasal regions repeatedly painted with water, urine from males housed in isolation, or urine from males housed in proximity to females. Almost all (87.5%) of the control females delivered litters. There was a small nonsignificant reduction in proportion parturient (78.5%) among females exposed to urine of males housed without social contact. The proportion of females parturient (57.1%) after treatment with urine from males housed in proximity to females was significantly different from controls. The magnitude of the effect of socially stimulated male urine is substantially less than that recorded when males are housed directly above inseminated females separated by a wire-mesh grid. These data suggest that production of pregnancy-disrupting male pheromones is stimulated by contact with conspecifics. © 1999 Elsevier Science Inc. Pregnancy disruption
Implantation
Pheromones
Urine
Mice
disruption. However, a disruption could be obtained by housing females in glass jars with restricted ventilation on cloth bedding that was highly retentive of odors, and renewing the soiled container twice daily over 3 days. In subsequent studies (35), pregnancy disruption required soiling of the female’s bedding by as many as five novel males, with renewal of the soiled bedding twice daily for 3 days. It is conceivable that such manipulations stress females in manners unrelated to pheromones, via air quality, bacteria, and human handling. This could confound interpretation of these results, given that chronic physical restraint (10) and daily human handling (38) can disrupt implantation on their own. There are reports that the Bruce Effect has been prevented when females were rendered anosmic by olfactory bulb removal (7) or olfactory tract transection (36) or lesioning of the vomeronasal organs (1,29). On the other hand, one
EXPOSURE to novel males can disrupt early pregnancy in various rodent species (4–6,14–16,24,28,40), a phenomenon known as the “Bruce Effect.” Similar disruptions of early pregnancy can be induced by diverse stressors in many mammals (9). It is clear that most of these stimuli terminate pregnancy by disrupting intrauterine implantation of fertilized ova. Much investigation has focused on stimulus elements producing the Bruce Effect, particularly the female’s sensitivity to pheromonal cues from the strange male. Several studies suggest that chemical products or pheromones produced by strange males, particularly in their urine, mediate pregnancy termination. Nevertheless, production of a pregnancy disruption via pheromones from novel males may require more stringent conditions than does direct exposure to novel males. Bruce (5) originally found that exposing females to cages recently soiled by novel males was not sufficient for a pregnancy
1To
Bruce Effect
whom requests for reprints should be addressed. E-mail: [email protected]
153
154 study found that the original inseminating male can disrupt pregnancy when the female’s vomeronasal system is damaged (26). Furthermore, some reports show strong pregnancy disruptions when females were merely exposed to novel male urine or extracts of such urine. Urine alone was reported to be sufficient, either when dripped into females’ bedding or when painted directly on the female’s nasal region (17–19). It has also been reported that proteins salted out of male urine (31,32) were sufficient to disrupt pregnancy. However, many of these studies (17–19,31,32,36) involved indirect assessments of pregnancy outcome. Disruption was inferred from vaginal smear measures, taken daily during exposure to novel males, with cell cornification indicating a return to estrus. Human manipulation of females in these protocols is of concern, as is need for validation of vaginal cell measures through female litter production. When novel males are housed directly with females, there are typically intense behavioral interactions between the male and female, adding nonpheromonal qualities to the effect. Novel males housed directly in the presence of recently inseminated females are agitated and prone to approach and mount females (14,16). In many instances, the female is reinseminated by the novel male, as evidenced by mating behavior, gestation length, and phenotypes of the litter resembling the novel male. Some studies suggest that sexual interactions with the novel male, especially intromissions, could contribute to such pregnancy disruptions (14,15,41), and vaginal stimulation is known to have neuroendocrine influences inducing estrus (27). Using procedures that minimize human handling and allow females to bear their litters, we have previously found a reliable disruption to early pregnancy with behavioral interactions between male and female minimized, simply by housing males above females separated by a wire-mesh grid (15,16). The effect is increased with more than one male above each female. Such an effect is absent, however, when the female is housed above males, even when as many as three males are housed below the female. These data implicate nonvolatile excretions of the novel males. Nevertheless, we have so far been unable to produce any strong disruption by urine alone. A previous report (15) indicated little effect of the application of urine to the nasal area of inseminated females during early pregnancy. However, the urine was taken from males without contact with females, with urination induced by human handling. It is conceivable that such stress-induced urination, in the absence of female exposure, is not representative of urine quality that would occur if males were in the presence of females. Exposure to females is known to alter the hormonal state of males; in particular, males exposed to females tend to show increased levels of testosterone (30). Accordingly, we reasoned that there may be context-specific release in male urine of pregnancy-disrupting chemical agents, provoked by the odor of conspecifics. The present study was designed to compare the influences upon pregnancy of urine from males housed without social contact to urine from males housed in proximity to females. The males were housed according to procedures that mimic those previously established to produce a strong and reliable pregnancy disruption (16), above a second compartment either containing or not containing a stimulus adult female. Two males were housed in the upper compartment, separated from each other by a double opaque plastic shield. Males were of a distinct strain (HS) from the females (CF-1), given data indicating a more reliable effect than when males are of the same strain as females (34,39). Urine taken from such males was painted on the nasal region of pre-
DE CATANZARO ET AL. viously inseminated female subjects, which were individually housed without any direct social contact. METHODS
Subjects CF-1 strain mice were obtained from Charles River Breeding Farms, La Prairie, Quebec, or bred in our laboratory from such stock. HS (heterogeneous strain) mice were bred in this laboratory from stock originally obtained from the Department of Zoology, University of Toronto, subsequently crossbred with C-57, DBA, Swiss, and CF-1 mice to maintain genetic heterogeneity. Housing involved standard polypropylene cages measuring 28 3 16 3 11 (height) cm with straight-wire tops allowing continuous access to food and water, except where otherwise specified. The colony room was maintained under a reversed 14:10 light:dark cycle at 218C. Insemination When CF-1 female subjects were between 75 and 100 days of age, they were each housed alone at the commencement of the dark phase of the light cycle with one CF-1 male in a standard cage. These males were sexually experienced and had been deprived of access to a female for at least 7 days. The hindquarters of each female were inspected on three occasions daily during the dark phase of their cycle for the presence of a sperm plug. At the end of the dark phase, all females with plugs were identified as subjects, and the day of detection was designated as day 0 of pregnancy. Each female remained housed with the inseminating male until the morning after detection of a sperm plug, about the start of the dark phase of the lighting cycle. The female was then removed from the male, housed individually in a clean cage with fresh bedding, and assigned to one of the experimental conditions. Assignment to conditions was counterbalanced across age and date of insemination. Novel Male Housing A double-decker apparatus (15,16) was used for housing HS male mice for urine collection. This apparatus was constructed from clear Plexiglas, measuring 30 3 21 3 27 cm, divided into two compartments measuring approximately 30 3 21 3 13 cm. These two compartments were separated by a wire-mesh grid with square openings measuring 0.5 cm2. The lower compartment was filled with approximately 0.5 L of clean bedding, had independent food and water supply, and was prepared for housing a female mouse that could not be directly contacted by the male above. The upper compartment was covered with a standard straight-wire mouse cage lid providing continuous access to food and water. The upper compartment was divided into two portions by a Plexiglas barrier at the lowest point of the cage lid. Two males were housed on each side of this barrier in each cage. First Replication In an initial experiment, inseminated females were assigned to two groups: Those receiving water stimulation (control condition) and those receiving urine taken from males housed in proximity to other females (experimental condition). For the experimental condition, HS males were housed in pairs above a normally cycling noninseminated adult female CF-1 mouse in the double-decker apparatus. Urine was
URINE QUALITY AND THE BRUCE EFFECT collected from these males by a human holding each male above a Petri dish, which usually elicited spontaneous urination. Care was taken to ensure that feces did not enter the sample. Inseminated female subjects in the experimental condition each received topical application of freshly collected male urine on the nasal area at 2.5, 6.5, 9.5, and 13.5 h after the start of the dark phase of the lighting cycle on each of days 1 through 5 after insemination. Inseminated female subjects in the control condition each received a similar topical application of tap water at the same points in the light cycle and after insemination. Topical application was accomplished by lowering a standard No. 6 artist’s paintbrush, saturated with the given substance, into the cage and gently swabbing the mouse’s snout area. Repeated measurement of the quantity of fluid delivered by this method, by weighing the brush before and after application, indicated a mean weight of fluid delivered in a single application of 82.8 mg, with a range of 55 to 104 mg. Direct handling of the mice was avoided. Different brushes were used for each condition. In the experimental condition, urine was applied within 2 min of collection from the male. Second Replication In a subsequent experiment, inseminated females were assigned to three groups: Those receiving water stimulation (control condition), those receiving urine from males housed without proximity to females (unstimulated condition), and those receiving urine taken from males housed in proximity to other females (female stimulated condition). For the unstimulated condition, HS males were housed in pairs as described above in the upper portion of the double-decker apparatus, with no animal housed in the lower compartment. For the female stimulated condition, a normally cycling noninseminated adult female CF-1 mouse was housed in the lower compartment in the double-decker apparatus described above, with two HS males above. Urine was collected from these males more efficiently than in the first replication, simply by removing the upper portion of the double-decker compartment, which contained the males, and placing it over a stainless-steel surface, such that urine from the males was deposited through the wire-mesh grid on the stainless-steel surface. Care was taken to ensure that feces did not enter into the sample. Urine or water application to inseminated females in the three conditions was accomplished as described above for the first replication. Dependent Measures After the experimental manipulations ended, subjects were not disturbed until 18 days following sperm plug detection, at which point they were observed twice daily for potential parturition until day 23 after sperm plug detection. The maximal length of normal gestation in this strain of mice in this laboratory is 22 days. Pregnancy outcome was measured through counts of the number of pups born, as described elsewhere (14–16). RESULTS
In animals run initially, where a water control group was compared to a group exposed to urine of males housed in proximity to females, 11 of 12 water-exposed controls delivered litters, while 7 of 13 urine-exposed females delivered litters. A chi-square test of association, relating conditions to the presence or absence of a litter, showed significance, x2(1) 5 4.43,
155 p , 0.05. In the subsequent replication with all three conditions, 24 of 28 water-exposed controls delivered litters, while 22 of 28 females treated with urine from males not exposed to females delivered litters, while 17 of 29 females treated with urine from males exposed to females delivered litters. A chisquare test of association comparing all three conditions marginally escaped the conventional level of significance, x(2) 5 5.88, p , 0.10, but one comparing just the water control and the female-stimulated urine groups did reach significance, x2(1) 5 5.18, p , 0.025. Thus, with combined data, the percent parturient was 87.5% of control females, 78.5% of females treated with urine of males alone, and 57.1% of females treated with urine from males housed in proximity to females. Considering all data, chi-square tests showed significance comparing all three groups, x2(2) 5 10.2, p , 0.01, and comparing the water control and stimulated-urine groups, x2(1) 5 9.36, p , 0.005, but not for the other pairs of comparisons. The mean (6SE) number of pups born for all females, including zero for each nonparturient female, was 10.90 (60.81) for the water control females, 9.25 (61.04) for females exposed to urine of nonstimulated males, and 6.31 (60.95) for females exposed to urine of female-stimulated males. Analysis of variance conducted on number of pups born was significant, F(2, 107) 5 6.99, p 5 0.0018. Multiple comparisons (Newman–Keuls) indicated that the water control group differed from the female-stimulated urine group (,0.01), and that the nonstimulated urine group differed from the female-stimulated urine group (,0.05), but did not show a difference between the water control group and the nonstimulated urine group. The differences among conditions in number of pups born were due primarily to differences in proportions of females bearing litters, as litter sizes were comparable when nonparturient females were excluded. DISCUSSION
An effect of novel male urine upon early pregnancy depends upon the social condition of males. Such an effect is demonstrated when urine is taken from males housed near females, but negligible when the males are isolated from females. The urine that effectively diminished the proportion of females delivering litters was taken from males housed above females, and elicited from the male by brief handling of the male and/or its caging. The urine from males without contact with females, elicited by similar brief handling, did not significantly diminish the proportion of females parturient. These data may help to reconcile previous failure to observe disruption of pregnancy using novel-male urine (15) and other reports of strong effects of such urine [e.g., (17)]. Nevertheless, the proximity of females to males from which urine was collected was not specified in that previous positive report (17) or reports of the use of male urine extracts (31,32). This is the first clear observation that urine alone can disrupt early pregnancy from this laboratory, using methods allowing females to bear their litters and that minimize human handling. Nevetheless, the effect is substantially smaller than that found when males are housed directly above inseminated females, separated only by a wire-mesh grid (16). Strain differences among laboratories could also be relevant, especially given that strains differ in susceptability to and capacity to induce the Bruce effect (8,33). Control of the amount of fecal matter in the urine may also be critical. The data indicate that social stimulation alters the chemical nature of urine such that it has a greater capacity to disrupt
156
DE CATANZARO ET AL.
pregnancy. Arguably, the present data do not exclude the possibility that exposure to conspecifics in general, rather than adult females specifically, alters the quality of urine such that it can disrupt pregnancy. However, in all conditions, each novel male was in olfactory proximity to a second male across the barrier in the upper portion of the exposure apparatus. Also, a previous report from this laboratory (15) indicated that urine of males that were group housed in their home cages failed to have a significant influence upon early pregnancy when painted on the noses of inseminated females. Evidence indicates that males’ behavior and endocrine states are altered by proximity to adult female mice. Males become highly agitated when presented with females through a wiremesh grid (14,16), and exposure to females with or without sexual access elevates testosterone levels (2,30), which conceivably plays a role in altering urine quality. On the other hand, exposure to conspecific males tends to lower gonadotropin levels in male mice, particularly among those subordinated in aggressive encounters (3). There can be little doubt that female house mice are generally responsive to chemical cues from male conspecifics. To some degree, inseminated female mice can discriminate among odors of different males (20), and estrous female house mice discriminate dominant from subordinate males (21,25). It has been shown that rates of urine excretion by mice are affected by age, sex, social status, and reproductive condition (22); in particular, dominant males produce more urine than do subordinate males, and castrated males produce less urine than do intact males. Castration eliminates the capacity of males to disrupt
pregnancy, both with direct exposure of inseminated females to males, where the males fail to mount females (14), and where such behavior is removed by indirect exposure through wire mesh (14,15,37) or mere exposure to male urine (17). Testosterone (15,17) or 17b-estradiol (13) administration restores the capacity of castrated males to disrupt pregnancy. Females can disrupt pregnancy when given testosterone (15,17,37). However, removal of androgen-dependent male accessory glands, such as the preputial glands, does not affect males’ capacity to disrupt pregnancy (16). Increasingly, evidence has implicated elevations of androgen and estrogen levels in the female to disruptions of intrauterine implantation of fertilized ova (9). Very low exogenous dosages of these steroids given directly to females completely disrupt implantation (11,23). Antibodies to 17b-estradiol can preserve pregnancy in the face of chronic restraint stress (10) and strange-male exposure (12). One possibility is that male urinary and fecal excretions transmit steroids to the inseminated female. On the other hand, some research implicates discriminations among individual male odors in the Bruce effect (1,7,20,26,29,36), which needs to be distinguished in future research from more nonspecific male factors such as urinary and fecal excretions. ACKNOWLEDGEMENTS
This research was supported by a grant from the Natural Sciences and Engineering Research Council of Canada to D. deCatanzaro, and from the North Atlantic Treaty Organization awarded to D. deCatanzaro and A. Boissy. The authors would like to thank Luc Lapointe and Miriam Hansen for their assistance in conducting this research.
REFERENCES 1. Bellringer, J. F.; Pratt, H. P. M.; Keverne, E. B.: Involvement of the vomeronasal organ and prolactin in pheromonal induction of delayed implantation in mice. J. Reprod. Fertil. 50:223–228; 1980. 2. Bliss, E. L.; Frischat, A.; Samuels, L.: Brain and testicular function. Life Sci. 11:231–238; 1972. 3. Bronson, F. H.; Stetson, M. H.; Stiff, M. E.: Serum FSH and LH in male mice following aggressive and nonaggressive interaction. Physiol. Behav. 10:369–372; 1973. 4. Bruce, H. M.: A block to pregnancy in mice caused by the proximity of strange males. J. Reprod. Fertil. 1:96–103; 1960. 5. Bruce, H. M.: Further observations of the pregnancy block in mice caused by the proximity of strange males. J. Reprod. Fertil. 1:311–312; 1960. 6. Bruce, H. M.: Olfactory block to pregnancy among grouped mice. J. Reprod. Fertil. 6:451–460; 1963. 7. Bruce, H. M.; Parrott, D. M. V.: Role of olfactory sense in pregnancy block by strange males. Science 131:1526; 1960. 8. Chapman, V. M.; Whitten, W. K.: The occurrence and inheritance of pregnancy block in inbred mice. Genetics 61:59; 1969. 9. deCatanzaro, D.; MacNiven, E.: Psychogenic pregnancy disruptions in mammals. Neurosci. Biobehav. Rev. 16:43–53; 1992. 10. deCatanzaro, D.; MacNiven, E.; Goodison, T.; Richardson, D.: Estrogen antibodies reduce vulnerability to stress-induced failure of intrauterine implantation in inseminated mice. Physiol. Behav. 55:35–38; 1994. 11. deCatanzaro, D.; MacNiven, E.; Ricciuti, F.: Comparison of the adverse effects of adrenal and ovarian steroids on early pregnancy in mice. Psychoneuroendocrinology 16:525–536; 1991. 12. deCatanzaro, D.; Muir, C.; O’Brien, J.; Williams, S.: Strangemale-induced pregnancy disruption in mice: Reduction of vulnerability by 17b-estradiol antibodies. Physiol. Behav. 58:401–404; 1995. 13. deCatanzaro, D.; Smith, M.; Muir, C.: Strange-male-induced pregnancy disruption in mice: Potentiation by administration of
14. 15.
16.
17. 18. 19. 20. 21. 22. 23. 24.
17b-estradiol to castrated males. Physiol. Behav. 58:405–408; 1995. deCatanzaro, D.; Storey, A.: Partial mediation of strange-maleinduced pregnancy blocks by sexual activity in mice (Mus musculus). J. Comp. Psychol. 103:381–388; 1989. deCatanzaro, D.; Wyngaarden, P.; Griffiths, J.; Ham, M.; Hancox, J.; Brain, D.: Interactions of contact, odor cues, and androgens in strange-male-induced early pregnancy disruptions in mice (Mus musculus). J. Comp. Psychol. 109:115–122; 1995. deCatanzaro, D.; Zacharias, R.; Muir, C.: Disruption of early pregnancy by direct and indirect exposure to novel males in mice: Comparison of influences of preputialectomized and intact males. J. Reprod. Fertil. 106:269–274; 1996. Dominic, C. J.: The origin of pheromones causing pregnancy block in mice. J. Reprod. Fertil. 10:469–472; 1965. Dominic, C. J.: Observations on the reproductive pheromones of mice. I. Source. J. Reprod. Fertil. 11:407–414; 1966. Dominic, C. J.: Observations on the reproductive pheromones of mice. II. Neuroendocrine mechanisms involved in the olfactory block to pregnancy. J. Reprod. Fertil. 11:415–421; 1966. Drickamer, L. C.: Pregnancy block in wild stock house mice, Mus domesticus: Olfactory preferences of females during gestation. Anim. Behav. 37:690–692; 1989. Drickamer, L. C.: Oestrus female house mice discriminate dominant from subordinate males and sons of dominant from sons of subordinate males by odour cues. Anim. Behav. 43:868–870; 1992. Drickamer, L. C.: Rates of urine excretion by house mouse (Mus domesticus): Differences by age, sex, social status, and reproductive condition. J. Chem. Ecol. 21:1481–1493; 1995. Harper, M. J. K.: Estrogenic effects of dehydroepiandrosterone and its sulfate in rats. Endocrinology 84:229–235; 1969. Heske, E. J.; Nelson, R. J.: Pregnancy interruption in Microtus ochrogaster: Laboratory artifact or field phenomenon. Biol. Reprod. 31:97–103; 1984.
URINE QUALITY AND THE BRUCE EFFECT 25. Hurst, J. L.: Urine marking in populations of wild house mice Mus domesticus Rutty. III. Communication between the sexes. Anim. Behav. 40:233–243; 1990. 26. Keverne, E. B.; de la Riva, C.: Pheromones in mice: Reciprocal interactions between the nose and brain. Nature 296:148–150; 1982. 27. Komisaruk, B. R.; Steinman, J. L.: Genital stimulation as a trigger for neuroendocrine and behavioral control of reproduction. Ann. NY Acad. Sci. 474:64–74; 1986. 28. Labov, J.: Pregnancy blocking in rodents: Adaptive advantages for females. Am. Natur. 118:297–303; 1981. 29. Lloyd–Thomas, A.; Keverne, E. B.: Role of the brain and accessory olfactory system in the block to pregnancy in mice. Neuroscience 7:907–913; 1982. 30. Macrides, F.; Bartke, A.; Dalterio, S.: Strange females increase plasma testosterone levels in male mice. Science 189:189–191; 1975. 31. Marchlewska–Koj, A.: Pregnancy block elicited by urinary proteins of male mice. Biol. Reprod. 18:729–732; 1977. 32. Marchlewska–Koj, A.: Pregnancy block elicited by male urinary peptides in mice. J. Reprod. Fertil. 61:221–224; 1981. 33. Marsden, H. M.; Bronson, F. H.: Strange male block to pregnancy: Its absence in inbred mouse strains. Nature 207:878; 1965.
157 34. Parkes, A. S.; Bruce, H. M.: Olfactory stimuli in mammalian reproduction. Science 134:1049–1054; 1961. 35. Parkes, A. S.; Bruce, H. M.: Pregnancy block in female mice placed in boxes soiled by males. J. Reprod. Fertil. 4:303–308; 1962. 36. Rajendren, G.; Dominic, C. J.: Effect of bilateral transection of the lateral olfactory tract on the male-induced implantation failure. Physiol. Behav. 36:587–590; 1986. 37. Rajendren, G.; Dominic, C. J.: Effect of cyproterone acetate on the pregnancy-blocking ability of male mice and the possible chemical nature of the pheromone. J. Reprod. Fertil. 84:387–392; 1988. 38. Runner, M. L.: Embryocidal effect of handling pregnant mice and its prevention with progesterone. Anat. Rec. 133:330–331; 1959. 39. Spironello, E.; deCatanzaro, D.: Sexual satiety diminishes the capacity of novel males to disrupt early pregnancy in inseminated female mice (Mus musculus). J. Comp. Psychol. (in press). 40. Storey, A. E.: Influence of sires on male-induced pregnancy disruptions in meadow voles (Microtus pennsylvanicus) differs with stage of pregnancy. J. Comp. Psychol. 100:15–20; 1986. 41. Storey, A. E.; Snow, D. T.: Postimplantation pregnancy disruptions in meadow voles: Relationship to variation in male sexual and aggressive behavior. Physiol. Behav. 47:19–25; 1990.
PII S0031-9384(98)00277-7
Pheromones and Novel Male-Induced Pregnancy Disruptions in Mice: Exposure to Conspecifics Is Necessary for Urine Alone to Induce an Effect DENYS DE CATANZARO,*1 CAMERON MUIR,* CHRISTINE SULLIVAN* AND ALAIN BOISSY† *Department of Psychology, McMaster University, Hamilton, Ontario, L8S 4K1 Canada, and †L.A.H.M., I.N.R.A., Clermont-Ferrand, Theix, 63122, St.-Genès, France Received 3 July 1998; Accepted 28 September 1998 DECATANZARO, D., C. MUIR, C. SULLIVAN AND A. BOISSY. Pheromones and novel male-induced pregnancy disruptions in mice: Exposure to conspecifics is necessary for urine alone to induce an effect. PHYSIOL BEHAV 66(1) 153– 157, 1999.—Previous research indicates a role of pheromones in novel male-induced early pregnancy disruptions. Although some reports suggest that urine alone is sufficient to produce this effect, others raise procedural concerns and fail to replicate such effects. On Days 1 to 5 after insemination, female CF-1 mice had their nasal regions repeatedly painted with water, urine from males housed in isolation, or urine from males housed in proximity to females. Almost all (87.5%) of the control females delivered litters. There was a small nonsignificant reduction in proportion parturient (78.5%) among females exposed to urine of males housed without social contact. The proportion of females parturient (57.1%) after treatment with urine from males housed in proximity to females was significantly different from controls. The magnitude of the effect of socially stimulated male urine is substantially less than that recorded when males are housed directly above inseminated females separated by a wire-mesh grid. These data suggest that production of pregnancy-disrupting male pheromones is stimulated by contact with conspecifics. © 1999 Elsevier Science Inc. Pregnancy disruption
Implantation
Pheromones
Urine
Mice
disruption. However, a disruption could be obtained by housing females in glass jars with restricted ventilation on cloth bedding that was highly retentive of odors, and renewing the soiled container twice daily over 3 days. In subsequent studies (35), pregnancy disruption required soiling of the female’s bedding by as many as five novel males, with renewal of the soiled bedding twice daily for 3 days. It is conceivable that such manipulations stress females in manners unrelated to pheromones, via air quality, bacteria, and human handling. This could confound interpretation of these results, given that chronic physical restraint (10) and daily human handling (38) can disrupt implantation on their own. There are reports that the Bruce Effect has been prevented when females were rendered anosmic by olfactory bulb removal (7) or olfactory tract transection (36) or lesioning of the vomeronasal organs (1,29). On the other hand, one
EXPOSURE to novel males can disrupt early pregnancy in various rodent species (4–6,14–16,24,28,40), a phenomenon known as the “Bruce Effect.” Similar disruptions of early pregnancy can be induced by diverse stressors in many mammals (9). It is clear that most of these stimuli terminate pregnancy by disrupting intrauterine implantation of fertilized ova. Much investigation has focused on stimulus elements producing the Bruce Effect, particularly the female’s sensitivity to pheromonal cues from the strange male. Several studies suggest that chemical products or pheromones produced by strange males, particularly in their urine, mediate pregnancy termination. Nevertheless, production of a pregnancy disruption via pheromones from novel males may require more stringent conditions than does direct exposure to novel males. Bruce (5) originally found that exposing females to cages recently soiled by novel males was not sufficient for a pregnancy
1To
Bruce Effect
whom requests for reprints should be addressed. E-mail: [email protected]
153
154 study found that the original inseminating male can disrupt pregnancy when the female’s vomeronasal system is damaged (26). Furthermore, some reports show strong pregnancy disruptions when females were merely exposed to novel male urine or extracts of such urine. Urine alone was reported to be sufficient, either when dripped into females’ bedding or when painted directly on the female’s nasal region (17–19). It has also been reported that proteins salted out of male urine (31,32) were sufficient to disrupt pregnancy. However, many of these studies (17–19,31,32,36) involved indirect assessments of pregnancy outcome. Disruption was inferred from vaginal smear measures, taken daily during exposure to novel males, with cell cornification indicating a return to estrus. Human manipulation of females in these protocols is of concern, as is need for validation of vaginal cell measures through female litter production. When novel males are housed directly with females, there are typically intense behavioral interactions between the male and female, adding nonpheromonal qualities to the effect. Novel males housed directly in the presence of recently inseminated females are agitated and prone to approach and mount females (14,16). In many instances, the female is reinseminated by the novel male, as evidenced by mating behavior, gestation length, and phenotypes of the litter resembling the novel male. Some studies suggest that sexual interactions with the novel male, especially intromissions, could contribute to such pregnancy disruptions (14,15,41), and vaginal stimulation is known to have neuroendocrine influences inducing estrus (27). Using procedures that minimize human handling and allow females to bear their litters, we have previously found a reliable disruption to early pregnancy with behavioral interactions between male and female minimized, simply by housing males above females separated by a wire-mesh grid (15,16). The effect is increased with more than one male above each female. Such an effect is absent, however, when the female is housed above males, even when as many as three males are housed below the female. These data implicate nonvolatile excretions of the novel males. Nevertheless, we have so far been unable to produce any strong disruption by urine alone. A previous report (15) indicated little effect of the application of urine to the nasal area of inseminated females during early pregnancy. However, the urine was taken from males without contact with females, with urination induced by human handling. It is conceivable that such stress-induced urination, in the absence of female exposure, is not representative of urine quality that would occur if males were in the presence of females. Exposure to females is known to alter the hormonal state of males; in particular, males exposed to females tend to show increased levels of testosterone (30). Accordingly, we reasoned that there may be context-specific release in male urine of pregnancy-disrupting chemical agents, provoked by the odor of conspecifics. The present study was designed to compare the influences upon pregnancy of urine from males housed without social contact to urine from males housed in proximity to females. The males were housed according to procedures that mimic those previously established to produce a strong and reliable pregnancy disruption (16), above a second compartment either containing or not containing a stimulus adult female. Two males were housed in the upper compartment, separated from each other by a double opaque plastic shield. Males were of a distinct strain (HS) from the females (CF-1), given data indicating a more reliable effect than when males are of the same strain as females (34,39). Urine taken from such males was painted on the nasal region of pre-
DE CATANZARO ET AL. viously inseminated female subjects, which were individually housed without any direct social contact. METHODS
Subjects CF-1 strain mice were obtained from Charles River Breeding Farms, La Prairie, Quebec, or bred in our laboratory from such stock. HS (heterogeneous strain) mice were bred in this laboratory from stock originally obtained from the Department of Zoology, University of Toronto, subsequently crossbred with C-57, DBA, Swiss, and CF-1 mice to maintain genetic heterogeneity. Housing involved standard polypropylene cages measuring 28 3 16 3 11 (height) cm with straight-wire tops allowing continuous access to food and water, except where otherwise specified. The colony room was maintained under a reversed 14:10 light:dark cycle at 218C. Insemination When CF-1 female subjects were between 75 and 100 days of age, they were each housed alone at the commencement of the dark phase of the light cycle with one CF-1 male in a standard cage. These males were sexually experienced and had been deprived of access to a female for at least 7 days. The hindquarters of each female were inspected on three occasions daily during the dark phase of their cycle for the presence of a sperm plug. At the end of the dark phase, all females with plugs were identified as subjects, and the day of detection was designated as day 0 of pregnancy. Each female remained housed with the inseminating male until the morning after detection of a sperm plug, about the start of the dark phase of the lighting cycle. The female was then removed from the male, housed individually in a clean cage with fresh bedding, and assigned to one of the experimental conditions. Assignment to conditions was counterbalanced across age and date of insemination. Novel Male Housing A double-decker apparatus (15,16) was used for housing HS male mice for urine collection. This apparatus was constructed from clear Plexiglas, measuring 30 3 21 3 27 cm, divided into two compartments measuring approximately 30 3 21 3 13 cm. These two compartments were separated by a wire-mesh grid with square openings measuring 0.5 cm2. The lower compartment was filled with approximately 0.5 L of clean bedding, had independent food and water supply, and was prepared for housing a female mouse that could not be directly contacted by the male above. The upper compartment was covered with a standard straight-wire mouse cage lid providing continuous access to food and water. The upper compartment was divided into two portions by a Plexiglas barrier at the lowest point of the cage lid. Two males were housed on each side of this barrier in each cage. First Replication In an initial experiment, inseminated females were assigned to two groups: Those receiving water stimulation (control condition) and those receiving urine taken from males housed in proximity to other females (experimental condition). For the experimental condition, HS males were housed in pairs above a normally cycling noninseminated adult female CF-1 mouse in the double-decker apparatus. Urine was
URINE QUALITY AND THE BRUCE EFFECT collected from these males by a human holding each male above a Petri dish, which usually elicited spontaneous urination. Care was taken to ensure that feces did not enter the sample. Inseminated female subjects in the experimental condition each received topical application of freshly collected male urine on the nasal area at 2.5, 6.5, 9.5, and 13.5 h after the start of the dark phase of the lighting cycle on each of days 1 through 5 after insemination. Inseminated female subjects in the control condition each received a similar topical application of tap water at the same points in the light cycle and after insemination. Topical application was accomplished by lowering a standard No. 6 artist’s paintbrush, saturated with the given substance, into the cage and gently swabbing the mouse’s snout area. Repeated measurement of the quantity of fluid delivered by this method, by weighing the brush before and after application, indicated a mean weight of fluid delivered in a single application of 82.8 mg, with a range of 55 to 104 mg. Direct handling of the mice was avoided. Different brushes were used for each condition. In the experimental condition, urine was applied within 2 min of collection from the male. Second Replication In a subsequent experiment, inseminated females were assigned to three groups: Those receiving water stimulation (control condition), those receiving urine from males housed without proximity to females (unstimulated condition), and those receiving urine taken from males housed in proximity to other females (female stimulated condition). For the unstimulated condition, HS males were housed in pairs as described above in the upper portion of the double-decker apparatus, with no animal housed in the lower compartment. For the female stimulated condition, a normally cycling noninseminated adult female CF-1 mouse was housed in the lower compartment in the double-decker apparatus described above, with two HS males above. Urine was collected from these males more efficiently than in the first replication, simply by removing the upper portion of the double-decker compartment, which contained the males, and placing it over a stainless-steel surface, such that urine from the males was deposited through the wire-mesh grid on the stainless-steel surface. Care was taken to ensure that feces did not enter into the sample. Urine or water application to inseminated females in the three conditions was accomplished as described above for the first replication. Dependent Measures After the experimental manipulations ended, subjects were not disturbed until 18 days following sperm plug detection, at which point they were observed twice daily for potential parturition until day 23 after sperm plug detection. The maximal length of normal gestation in this strain of mice in this laboratory is 22 days. Pregnancy outcome was measured through counts of the number of pups born, as described elsewhere (14–16). RESULTS
In animals run initially, where a water control group was compared to a group exposed to urine of males housed in proximity to females, 11 of 12 water-exposed controls delivered litters, while 7 of 13 urine-exposed females delivered litters. A chi-square test of association, relating conditions to the presence or absence of a litter, showed significance, x2(1) 5 4.43,
155 p , 0.05. In the subsequent replication with all three conditions, 24 of 28 water-exposed controls delivered litters, while 22 of 28 females treated with urine from males not exposed to females delivered litters, while 17 of 29 females treated with urine from males exposed to females delivered litters. A chisquare test of association comparing all three conditions marginally escaped the conventional level of significance, x(2) 5 5.88, p , 0.10, but one comparing just the water control and the female-stimulated urine groups did reach significance, x2(1) 5 5.18, p , 0.025. Thus, with combined data, the percent parturient was 87.5% of control females, 78.5% of females treated with urine of males alone, and 57.1% of females treated with urine from males housed in proximity to females. Considering all data, chi-square tests showed significance comparing all three groups, x2(2) 5 10.2, p , 0.01, and comparing the water control and stimulated-urine groups, x2(1) 5 9.36, p , 0.005, but not for the other pairs of comparisons. The mean (6SE) number of pups born for all females, including zero for each nonparturient female, was 10.90 (60.81) for the water control females, 9.25 (61.04) for females exposed to urine of nonstimulated males, and 6.31 (60.95) for females exposed to urine of female-stimulated males. Analysis of variance conducted on number of pups born was significant, F(2, 107) 5 6.99, p 5 0.0018. Multiple comparisons (Newman–Keuls) indicated that the water control group differed from the female-stimulated urine group (,0.01), and that the nonstimulated urine group differed from the female-stimulated urine group (,0.05), but did not show a difference between the water control group and the nonstimulated urine group. The differences among conditions in number of pups born were due primarily to differences in proportions of females bearing litters, as litter sizes were comparable when nonparturient females were excluded. DISCUSSION
An effect of novel male urine upon early pregnancy depends upon the social condition of males. Such an effect is demonstrated when urine is taken from males housed near females, but negligible when the males are isolated from females. The urine that effectively diminished the proportion of females delivering litters was taken from males housed above females, and elicited from the male by brief handling of the male and/or its caging. The urine from males without contact with females, elicited by similar brief handling, did not significantly diminish the proportion of females parturient. These data may help to reconcile previous failure to observe disruption of pregnancy using novel-male urine (15) and other reports of strong effects of such urine [e.g., (17)]. Nevertheless, the proximity of females to males from which urine was collected was not specified in that previous positive report (17) or reports of the use of male urine extracts (31,32). This is the first clear observation that urine alone can disrupt early pregnancy from this laboratory, using methods allowing females to bear their litters and that minimize human handling. Nevetheless, the effect is substantially smaller than that found when males are housed directly above inseminated females, separated only by a wire-mesh grid (16). Strain differences among laboratories could also be relevant, especially given that strains differ in susceptability to and capacity to induce the Bruce effect (8,33). Control of the amount of fecal matter in the urine may also be critical. The data indicate that social stimulation alters the chemical nature of urine such that it has a greater capacity to disrupt
156
DE CATANZARO ET AL.
pregnancy. Arguably, the present data do not exclude the possibility that exposure to conspecifics in general, rather than adult females specifically, alters the quality of urine such that it can disrupt pregnancy. However, in all conditions, each novel male was in olfactory proximity to a second male across the barrier in the upper portion of the exposure apparatus. Also, a previous report from this laboratory (15) indicated that urine of males that were group housed in their home cages failed to have a significant influence upon early pregnancy when painted on the noses of inseminated females. Evidence indicates that males’ behavior and endocrine states are altered by proximity to adult female mice. Males become highly agitated when presented with females through a wiremesh grid (14,16), and exposure to females with or without sexual access elevates testosterone levels (2,30), which conceivably plays a role in altering urine quality. On the other hand, exposure to conspecific males tends to lower gonadotropin levels in male mice, particularly among those subordinated in aggressive encounters (3). There can be little doubt that female house mice are generally responsive to chemical cues from male conspecifics. To some degree, inseminated female mice can discriminate among odors of different males (20), and estrous female house mice discriminate dominant from subordinate males (21,25). It has been shown that rates of urine excretion by mice are affected by age, sex, social status, and reproductive condition (22); in particular, dominant males produce more urine than do subordinate males, and castrated males produce less urine than do intact males. Castration eliminates the capacity of males to disrupt
pregnancy, both with direct exposure of inseminated females to males, where the males fail to mount females (14), and where such behavior is removed by indirect exposure through wire mesh (14,15,37) or mere exposure to male urine (17). Testosterone (15,17) or 17b-estradiol (13) administration restores the capacity of castrated males to disrupt pregnancy. Females can disrupt pregnancy when given testosterone (15,17,37). However, removal of androgen-dependent male accessory glands, such as the preputial glands, does not affect males’ capacity to disrupt pregnancy (16). Increasingly, evidence has implicated elevations of androgen and estrogen levels in the female to disruptions of intrauterine implantation of fertilized ova (9). Very low exogenous dosages of these steroids given directly to females completely disrupt implantation (11,23). Antibodies to 17b-estradiol can preserve pregnancy in the face of chronic restraint stress (10) and strange-male exposure (12). One possibility is that male urinary and fecal excretions transmit steroids to the inseminated female. On the other hand, some research implicates discriminations among individual male odors in the Bruce effect (1,7,20,26,29,36), which needs to be distinguished in future research from more nonspecific male factors such as urinary and fecal excretions. ACKNOWLEDGEMENTS
This research was supported by a grant from the Natural Sciences and Engineering Research Council of Canada to D. deCatanzaro, and from the North Atlantic Treaty Organization awarded to D. deCatanzaro and A. Boissy. The authors would like to thank Luc Lapointe and Miriam Hansen for their assistance in conducting this research.
REFERENCES 1. Bellringer, J. F.; Pratt, H. P. M.; Keverne, E. B.: Involvement of the vomeronasal organ and prolactin in pheromonal induction of delayed implantation in mice. J. Reprod. Fertil. 50:223–228; 1980. 2. Bliss, E. L.; Frischat, A.; Samuels, L.: Brain and testicular function. Life Sci. 11:231–238; 1972. 3. Bronson, F. H.; Stetson, M. H.; Stiff, M. E.: Serum FSH and LH in male mice following aggressive and nonaggressive interaction. Physiol. Behav. 10:369–372; 1973. 4. Bruce, H. M.: A block to pregnancy in mice caused by the proximity of strange males. J. Reprod. Fertil. 1:96–103; 1960. 5. Bruce, H. M.: Further observations of the pregnancy block in mice caused by the proximity of strange males. J. Reprod. Fertil. 1:311–312; 1960. 6. Bruce, H. M.: Olfactory block to pregnancy among grouped mice. J. Reprod. Fertil. 6:451–460; 1963. 7. Bruce, H. M.; Parrott, D. M. V.: Role of olfactory sense in pregnancy block by strange males. Science 131:1526; 1960. 8. Chapman, V. M.; Whitten, W. K.: The occurrence and inheritance of pregnancy block in inbred mice. Genetics 61:59; 1969. 9. deCatanzaro, D.; MacNiven, E.: Psychogenic pregnancy disruptions in mammals. Neurosci. Biobehav. Rev. 16:43–53; 1992. 10. deCatanzaro, D.; MacNiven, E.; Goodison, T.; Richardson, D.: Estrogen antibodies reduce vulnerability to stress-induced failure of intrauterine implantation in inseminated mice. Physiol. Behav. 55:35–38; 1994. 11. deCatanzaro, D.; MacNiven, E.; Ricciuti, F.: Comparison of the adverse effects of adrenal and ovarian steroids on early pregnancy in mice. Psychoneuroendocrinology 16:525–536; 1991. 12. deCatanzaro, D.; Muir, C.; O’Brien, J.; Williams, S.: Strangemale-induced pregnancy disruption in mice: Reduction of vulnerability by 17b-estradiol antibodies. Physiol. Behav. 58:401–404; 1995. 13. deCatanzaro, D.; Smith, M.; Muir, C.: Strange-male-induced pregnancy disruption in mice: Potentiation by administration of
14. 15.
16.
17. 18. 19. 20. 21. 22. 23. 24.
17b-estradiol to castrated males. Physiol. Behav. 58:405–408; 1995. deCatanzaro, D.; Storey, A.: Partial mediation of strange-maleinduced pregnancy blocks by sexual activity in mice (Mus musculus). J. Comp. Psychol. 103:381–388; 1989. deCatanzaro, D.; Wyngaarden, P.; Griffiths, J.; Ham, M.; Hancox, J.; Brain, D.: Interactions of contact, odor cues, and androgens in strange-male-induced early pregnancy disruptions in mice (Mus musculus). J. Comp. Psychol. 109:115–122; 1995. deCatanzaro, D.; Zacharias, R.; Muir, C.: Disruption of early pregnancy by direct and indirect exposure to novel males in mice: Comparison of influences of preputialectomized and intact males. J. Reprod. Fertil. 106:269–274; 1996. Dominic, C. J.: The origin of pheromones causing pregnancy block in mice. J. Reprod. Fertil. 10:469–472; 1965. Dominic, C. J.: Observations on the reproductive pheromones of mice. I. Source. J. Reprod. Fertil. 11:407–414; 1966. Dominic, C. J.: Observations on the reproductive pheromones of mice. II. Neuroendocrine mechanisms involved in the olfactory block to pregnancy. J. Reprod. Fertil. 11:415–421; 1966. Drickamer, L. C.: Pregnancy block in wild stock house mice, Mus domesticus: Olfactory preferences of females during gestation. Anim. Behav. 37:690–692; 1989. Drickamer, L. C.: Oestrus female house mice discriminate dominant from subordinate males and sons of dominant from sons of subordinate males by odour cues. Anim. Behav. 43:868–870; 1992. Drickamer, L. C.: Rates of urine excretion by house mouse (Mus domesticus): Differences by age, sex, social status, and reproductive condition. J. Chem. Ecol. 21:1481–1493; 1995. Harper, M. J. K.: Estrogenic effects of dehydroepiandrosterone and its sulfate in rats. Endocrinology 84:229–235; 1969. Heske, E. J.; Nelson, R. J.: Pregnancy interruption in Microtus ochrogaster: Laboratory artifact or field phenomenon. Biol. Reprod. 31:97–103; 1984.
URINE QUALITY AND THE BRUCE EFFECT 25. Hurst, J. L.: Urine marking in populations of wild house mice Mus domesticus Rutty. III. Communication between the sexes. Anim. Behav. 40:233–243; 1990. 26. Keverne, E. B.; de la Riva, C.: Pheromones in mice: Reciprocal interactions between the nose and brain. Nature 296:148–150; 1982. 27. Komisaruk, B. R.; Steinman, J. L.: Genital stimulation as a trigger for neuroendocrine and behavioral control of reproduction. Ann. NY Acad. Sci. 474:64–74; 1986. 28. Labov, J.: Pregnancy blocking in rodents: Adaptive advantages for females. Am. Natur. 118:297–303; 1981. 29. Lloyd–Thomas, A.; Keverne, E. B.: Role of the brain and accessory olfactory system in the block to pregnancy in mice. Neuroscience 7:907–913; 1982. 30. Macrides, F.; Bartke, A.; Dalterio, S.: Strange females increase plasma testosterone levels in male mice. Science 189:189–191; 1975. 31. Marchlewska–Koj, A.: Pregnancy block elicited by urinary proteins of male mice. Biol. Reprod. 18:729–732; 1977. 32. Marchlewska–Koj, A.: Pregnancy block elicited by male urinary peptides in mice. J. Reprod. Fertil. 61:221–224; 1981. 33. Marsden, H. M.; Bronson, F. H.: Strange male block to pregnancy: Its absence in inbred mouse strains. Nature 207:878; 1965.
157 34. Parkes, A. S.; Bruce, H. M.: Olfactory stimuli in mammalian reproduction. Science 134:1049–1054; 1961. 35. Parkes, A. S.; Bruce, H. M.: Pregnancy block in female mice placed in boxes soiled by males. J. Reprod. Fertil. 4:303–308; 1962. 36. Rajendren, G.; Dominic, C. J.: Effect of bilateral transection of the lateral olfactory tract on the male-induced implantation failure. Physiol. Behav. 36:587–590; 1986. 37. Rajendren, G.; Dominic, C. J.: Effect of cyproterone acetate on the pregnancy-blocking ability of male mice and the possible chemical nature of the pheromone. J. Reprod. Fertil. 84:387–392; 1988. 38. Runner, M. L.: Embryocidal effect of handling pregnant mice and its prevention with progesterone. Anat. Rec. 133:330–331; 1959. 39. Spironello, E.; deCatanzaro, D.: Sexual satiety diminishes the capacity of novel males to disrupt early pregnancy in inseminated female mice (Mus musculus). J. Comp. Psychol. (in press). 40. Storey, A. E.: Influence of sires on male-induced pregnancy disruptions in meadow voles (Microtus pennsylvanicus) differs with stage of pregnancy. J. Comp. Psychol. 100:15–20; 1986. 41. Storey, A. E.; Snow, D. T.: Postimplantation pregnancy disruptions in meadow voles: Relationship to variation in male sexual and aggressive behavior. Physiol. Behav. 47:19–25; 1990.