Camp. Biochem. Physiol. Vol. IOU, No. 3, pp. 453-457, 1992
030&9629/92 $5.00 + 0.00
Printedin GreatBritain
0 1992PergamonPm.9Ltd
CORRELATIONS BETWEEN SERUM CORTICOSTERONE CONCENTRATION AND REPRODUCTIVE CONDITIONS IN THE WHITE-FOOTED MOUSE (PEROMYSCUS LEUCOPUS NOVEBORACENSIS) STERLINGN.
RANSONE
JR and ERIC L. BRADLEY*
Laboratory of Endocrinology and Population Ecology, Biology Department, College of William and Mary, Williamsburg, VA 23187, U.S.A. Tel.: (804) 221-2220; Fax: (804) 221-6483 (Received
29 November
1991)
Abstract-1. Adrenal size and activity in reproductively inhibited young born into laboratory populations of the white-footed mouse were examined. Measurements were made of body weight, reproductive organ weight and adrenal weight. Serum corticosterone was measured by radioimmunoassay. 2. Data from reproductively inhibited animals were compared with corresponding values from reproductively capable animals of the same sex. The mean paired testis and seminal vesicle, or paired ovary and uterus, weights, were significantly reduced in reproductively inhibited animals of both sexes. The paired adrenal weights were not different in any comparison. 3. Reproductively inhibited males selected from laboratory populations had a mean corticosterone concentration that was significantly higher than the corresponding value for reproductively capable males. Females showed no significant difference in mean serum corticosterone concentrations. 4. The results are discussed in relation to earlier studies in two species of Peromyscus and the apparent paradox of elevated serum corticosterone in the absence of adrenal hypertrophy.
The
prairie
deetmouse,
maniculatus,
rarely 1940),
natural population (Blair, undergoes much population density over a time-span than species of mammals (Terman, Likewise, the mouse, Peromyscus shows a range of size fluctuation, this range typically covered a shorter span (Terman, Apparently, these Peromyscus species an intrinsic to regunatural population at well the physical capacity of environment. Laboratory of the species also numerical growth to either failure of to survive most commonly, a cessation reproduction within population when of the productive population fail to Most dramatically, majority (usually than 90%) the young into the lation do mature reproductively maintain smaller organs compared reproductively animals (Terman, 1966, 1969; and Terman, Haigh, 1983; and Terman, One explanation reproductive inhibition some rodent populations has that an in relative or “crowding” an hypophysial-adrenal response which hypersecretion of hormone (ACTH) adrenal steroids ultimately disrupts (Christian, 1950, 1956, *To
all correspondence
he addressed.
Christian et 1965). Indeed, dependent, adrenal-based alterations have well documented Mus, Rattus Microtus (Christian, 1956; Brain, Andrews and 1979; To Tamarin, 1977). the prairie significantly elevated serum corticosterone has been in both of reproductively animals from relative to serum levels in control However, there no concomitant in the of the in these animals (Sung al., 1977; and Terman, Also, an study with species did demonstrate a difference in serum ACTH of reproductively animals from populations compared reproductively capable (Coppes and 1984). This extends to Peromyscus species, evaluation of adrenal size serum corticosterone related to condition within populations. MATERULS
METBOBS
Animal
The animals this study myxus kucopus
laboratory tained in cage,
white-footed mice outbred Colony and animals were two-chambered, wire-topped, plastic derived from
12.8 x x 14Scm. (Prolab Rat, Hamster 3000, Inc., Syracuse, and tap were continuously Cage bedding of wood which were every 2 Room temperature regulated between f 3°C 5 to ah exchanges nour. The cycle consisted 14 hr
453
454
STERLING N. RANSONE JR and ERIC L. BRADW
light (15 A-c. at the floor) turned on between 7 a.m. and 9p.m. EST. Colony young were weaned at 21 days of age and placed into cages with same-sex sibs. At 60 f 3 days of age, 50 females were paired with 50 non-sib. males. The first 30 pairs to produce young were used to found 10 laboratory populations and the remainder of the pairs were used to produce control animals. Experimental animals
Control animals were produced from 12 colony pairs. Young were weaned at 21 days of age and placed individually into a cage compartment so that each double-sided cage contained one male and one female that was within 1 day of age of each other. Neither animal could see or touch the other. At 35 It 1 days of age each pair was switched to the opposite side of the compartment onto the soiled bedding of the other in order to maintain olfactory contact with the corresponding mate. This regimen was repeated twice more at 2-week intervals. At 70 f 1 days of age the animals were sacrificed for tissue collection. Each of 10 laboratory populations was founded with three pairs of reproductively proven animals ranging in age from 83 to 161 days old. Only young produced in the population context were retained. Each population was maintained within an enclosure consisting of a 1.5 m diameter stainless steel base with aluminum siding walls at a height of at least 68 cm. Six one-quart plastic containers were provided as nest-boxes in each enclosure and bedding consisted of wood chips. Food, water and environmental conditions were the same as those described for controls. Each population was inspected at intervals of 2 weeks. Males were noted as either having scrotal or non-scrotal testes and females were checked for perforate or nonperforate vaginae. All population founding animals were toe-marked and young were marked during the subsequent inspections. Selection of reproductively inhibited animals from populations
When a litter reached 68 days of age, an additional population inspection was performed to identify young that had never been noted as having scrotal testes or a perforate vagina. During the check, a randomly selected male and female from the litter were marked with a non-toxic UVfluorescing dye (Ultraviolet Products Inc., San Gabriel, CA) to facilitate later identification. All animals were returned to the enclosure and remained undisturbed for the next 2 days. Tissue sampling was accomplished between 6:45 and 7: 15 p.m. EST (2 hr f 15 min prior to darkness) by locating the dye-marked animals with a long-range UV light and rapidly removing one to a jar containing diethyl ether vapor. Only one animal per day was removed from a given population and its marked littermate was removed 48hr later. Male population animals were sampled from six different populations and females were sampled from seven populations. Possible inter-population differences were tested using a Kruskal-Wallis one.-way analysis of variance. No significant differences were noted across populations in any of the variables that were measured for either sex and so the data from all populations were combined by sex. Collection of tissues
A ventral abdominal incision was made under diethyl ether anesthesia and blood was collected from the left renal artery in less than 2 min from the time of first contact with the animal. Blood was placed into a 1.5 ml plastic centrifuge tube to clot and then centrifuged at 9000 g for 2 min. The serum was drawn off and frozen at below -70°C until analysed. The body was weighed to the nearest 0.1 g. Both adret& and the reproductive organs (testes and seminal vesicles or ovary and uterus) were removed and placed directly in a
10% buffered formaldehyde solution. Organs were dissected free of fat and weighed to the nearest 0.1 mg after at least 72 hr in the fixative. Radioimmunoassay for corticosterone
Corticosterone antisera B3-163 (Endocrine Sciences, Tanana, CA) was diluted 1: 85 according to the specifications of the manufacturer. Standards for each assay were run in triplicate with 0.063,0.125,0.250,0.500, 1.000, 1.250 and 2.5OOng/tube of authentic corticosterone (SchwartzMann, Grangeburg, NY). A standard pool of P. feucopus female sera was assayed in duplicate at 1,0.5,0.25 and 0.125 times natural concentration. The assay protocol was conducted as previously described (Bradley and Terman, 1981a). Samples and standards were counted to a 1% error (Beckman LS-3 133T), calculated as per cent of total counts added, and then logit transformed. Standard hormone concentration values were transformed to the log of the concentration. The slopes of the four standard curves and the four serial dilutions of the standard sera pools were compared using the BMDLR multiplaregression program (Biomedical Computer Programs, University of California, Berkeley, CA). No significant diiTerences in the slope of any regression were found. The mean limit of detection calcu-
lated from the variation of the buffer control tubes was 0.46 f 0.01 pgjtube. Hormoneconcentrationsarc expressed in ng/ml of sera. Statistics
A one-way ANOVA indicated heterogeneous variance and so the nonparametric Mann-Whitney U-test was used for comparisons. All correlation data was analysed using Spearman’s nonparametric ranked correlation test. All data are reported as the mean value f the standard error of the mean. In all cases P < 0.05 was considered statistically significant.
Comparironsbetween sexes Control males were significantly (P < 0.001) heavier than control females, but no such differences were observed between the sexes among the reproductively inhibited animals selected from populations. Also, there were no significant differeuces in mean absolute adrenal weights between the sexes in control or population animals (Tables 1 and 2). Weight comparbons between control and inhibited animaIs The mean body weight of reproductively inhibited males selected from populations was signi&ntly (P < 0.001) lighter than the corresponding control male value (Table 1). No corresponding differences in body weight were observed between the mean body weights for population vs control females (Table 2). The mean absolute adrenal weights were not significantly diRerent between controls and reproductively inhibited population males or females (cf. Tables 1 and 2). Likewise, the calculated relative adrenal weights for these comparisons were not different. The mean paired testes weights and seminal vesicle weights were both sign&a&y (P < 0.001) larger in controls compared with reproductively inhibited population males (Table 1). Similarly, among females, both the mean paired ovary and uterus weights were signScantly (P < 0.001) larger in controls compared with the corresponding values ftom population females (Table 2).
Corticosterone and reproduction in P. leucopus Table
455
I. Mean body weight, testis weight, seminal vesicle weight, adrenal weight and serum corticosterone concentration
Treatments Control males (N = 19)
Body weight (g) 20.09 * 0.43
Population males
16.8 + 0.71*+*
in control and population males Seminal vesicle weight (mg) 112.6 k 7.56
Testis weight (mg) 273.3 + 14.13 84.3 + 20.77***
11.0 f 4.16***
Values are mean 1?ISEM. *P < 0.05, ***P< 0.001, “‘not significant, with respect to corresponding
Serum corticosterone concentrations
There were no statistically significantly differences between the mean serum corticosterone concentrations in male controls compared with female controls or between reproductively inhibited male and female animals taken from populations. However, the mean serum corticosterone concentration of reproductively inhibited males taken from populations was significantly (P < 0.05) elevated compared with control males (Table 1). DISCUSSION
Of all of the young born into the ten laboratory populations and reaching 70 days of age, 29 of 33 males had undescended testes and 36 of 37 females were always vaginally imperforate. The drastic nature of the reproductive inhibition in males selected from populations was further indicated by the significantly smaller size of both the testes and seminal vesicles compared with control males (Table 1). Likewise, females selected from populations showed significantly smaller ovarian and uterine weights compared with their reproductively proven controls (Table 2). The degree of reproductive inhibition reported here is entirely similar to reports on other laboratory populations of P. leucopus (Haigh, 1987; Creigh and Terman, 1988) and also laboratory populations of P. maniculatus (Terman, 1969, 1973; Bradley and Terman, 1981a,b,c). The mean body weight of reproductively inhibited males selected from populations was significantly (P < 0.001) less than the mean weight for control males. However, no such significant reduction in mean weight was observed between females taken from populations and their respective controls. These results are similar to those reported by Bradley and Terman (1981a) for P. maniculatus, and indicates that, in the female of both species, reproductive inhibition may not be directly associated with smaller body weight.
Adrenal weight (mg) 10.6 f 0.47 9.0 k 0.86”’
Serum corticosterone (ng/ml) 145.3 + 14.41 201.9 + 16.05.
control value.
Neither absolute nor relative mean adrenal weight was significantly different between reproductively inhibited population and control males (Table 1) or females (Table 2). Among males there were no statistically significant correlations between the adrenal weight and reproductive organ weight. However, in control females ovary weight was positively correlated (r = 0.7806; P < 0.001) with adrenal weight. This relationship is very likely a reflection of the interrelationship between the ovary and adrenal during the normal ovarian cycle. These findings in P. Ieucopus are very similar to earlier studies with P. maniculatus where no population density-related adrenal weight increase was found for either growing or asymptotic laboratory populations (Bronson and Eleftheriou, 1963; Terman, 1966; Sung et al., 1977; Bradley and Terman, 1981a) or in crowded natural populations (McKeever, 1964). This lack of an adrenal hypertrophy response in Peromyscus is distinctly different from reports in such genera as Mus, Rattus and Microtus where there is a marked density-dependent adrenal hypertrophy in response to population density increase (Bronson and Eleftherious, 1963; Christian, 1955, 1956; Purushotham et al., 1964; Christian and Davis, 1966; Andrews and Belknap, 1979). Corticosterone comparisons
Control female white-footed mice in this study only tended (P < 0.09) to have higher serum corticosterone concentrations than control males. Bradley and Terman (1981a) reported a significantly higher serum corticosterone concentration in control female P. maniculatus compared with control males. This difference between the two reports might be due to some species variation in adrenal function, but it may also be due to females in different stages of their ovarian cycle (Nequin and Schwartz, 1971; Kitay et al., 1971; Barfield and Lisk, 1974). The mean serum corticosterone concentration in control male P. feucopus was over twice as high as the
Table 2. Mean body weight, ovary weight, uterus weight, adrenal weight and serum corticosterone concentration in control and population females
Treatments Control females (N=l8) Population females (N = 12)
Body weight (8) 17.7 * 0.61 16.6 * 0.68”’
Ovary weight (mg) 11.6kO.95 4.5 f 0.66***
Values are mean f SEM. l**P< 0.001, “hot significant, with respect to corresponding
Uterus weight (me) 49.5 k 5.36 11.8 f 2.58***
control value.
Adrenal weight (mg) 11.4f0.68 9.8 f 0.83”’
Serum wrticosterone (nglml) 189.1 + 19.25 168.5 f 16.35”’
456
STERLING N. RANSONE
mean level (68.8 ng/ml) reported for P. municulu&s by Bradley and Terman (1981a). This difference between the species is believed to be actual because a two-fold difference was also detected in the corticosterone values measured in standard sera pools from both species run with each assay in the present study. This apparent species difference is also interesting because the slightly heavier mature P. leucopus has a relative adrenal weight that is more than three times greater than P. manicula~us. Among control male P. i&opus there is a positive correlation of serum corticosterone concentration with adrenal weight (I = 0.4967; P = 0.042), although among the control females there was only a trend toward this same correlation (r = 0.3841; P = 0.116). Apparently, in reproductively proven P. leucopus of this age and weight range, there is a previously unidentified reIationship between increased adrenal weight and increased serum corticosterone concentration. In earlier studies on reproductively inhibited P. maniculatus selected from laboratory populations and compared with controls, Bradley and Terman (198 la) reported a tendency towards smaller adrenal glands concomitantly with significantly elevated serum corticosterone concentration. In this study, reproductively inhibited P. leucopus males had a mean serum corticosterone concentration that was also significantly (P < 0.04) elevated when compared with control males; however, the mean adrenal weights were not significantly decreased as in P. maniculatus. Also, in female P. Ieucopus there was no significant difference in the serum corticosterone concentration of population female vs control (Table 2), which is distinctly different from the case in female P. maniculatus. The serum concentration of corticosterone in reproductively inhibited P. Zeucopus males in this study, while higher than control males, was much lower than the mean level (348 ng/ml) measured previously in P. maniculatus inhibited males (Bradley and Terman, 1981a). It is also of interest that in earlier reports (Sung et al., 1977; Bradley and Terman, 1981a) the mean serum corticosterone concentration (178.3 ng/ml) for control female deermice was very similar to the value reported here for control female white-footed mice. The observation that there was a corresponding significant increase in the serum corticosterone concentration in white-footed female mice selected from populations is comparable with the si~nifi~ntly elevated mean value (326.3 ng/ml) reported for reproductively inhibited female P. maniculatus in the earlier study (Bradley and Terman, 1981a). The higher serum corticosterone concentrations found in mafes selected from populations may indicate a greater sensitivity to population conditions than shown by females. Males clearly experienced the highest rate of death due to wounding compared with females that were seldom observed to fight. However, no systematic behavioral observations were made that would confirm this notion. Taken together the data from studies on Peromyscus do not support the notion that a stressinduced adrenal hy~rtrophy is central to the reproductive inhibition that is so clearly produced in
JR and ERICL. BRADLEY the vast majority of young born into laboratory populations. However, it is also apparent that an elevated serum corticosterone concentration is associated with the reproductively inhibited condition. Whether the elevated corticosterone level is caused by a reduction in metabolic clearance of the steroid or an increase in steroid sequestration due to an increase in protective plasma protein binding (Bradley and Terman, 1981a) is not yet clear for this species. In order to address the apparent paradox of increased corticosterone concentration without an increase in adrenal weight, work is now in progress to evaluate possible adrenal zone area differences in the otherwise similar weight adrenals of reproductively inhibited animals. Also, the metabolic clearance of corticosterone and the general metabolic state of reproductively inhibited animals is under investigation. REFERENCES Andrews R. V. and Belknap R. W. (1979) Deer mouse and lemming adrenal and pathological responses to increases in animal numbers. Comp. Biochem. Physioi.GA, 15-18. Barfteld M. A. and Lisk R. D. (1974) Relative cont~bution of ovarian and adrenal progesterone to the timing of heat in the 4-day cyclic rat. Endocrinology94, 571-575. Blair W. F. (1940) A study of prairie deer-mouse populations in southern Michigan. Am. Mid!. Nat. 24, 272-304. Bradley E. L. and Terrnan C. R. (198la) A comparison of the adrenal histology, reproductive condition and serum corticosterone concentrations of prairie deermice (Peromyscus manictdatusbairdiiifin captivity. J. Mammol. 62, 353-361. Bradley E. L. and Terman C. R. (1981b) Serum testosterone concentrations in male prairie deermice (Peromyscus maniculatusbairdii). J. Mammol. 62, 811-814. Bradley E. L. and Terman C. R. (1981~) Studies on the nature of ~r~uctive inhibition in animals from laboratory populations of prairie deermice (Peromyscus maniculurusbairdii): serum LH and FSH concentrations. Comp. Biochem. Physiol. MA, 563-570. Brain P. F. (1971) The physiology of population limitation in rodents. Commrm. B&v. Biol. 6, 115-123. Bronson F. H. and Eleftherious B. E. (1963) Adrenal response to crowding in Peromyscus and C57BL/lOJ mice. Physiol. Zool. 36, 161-166. Christian J. J. (1950) The adreno-pituitary system and ~p~ation cycles in mammals. J. Moon. 31,247-259. Christian J. J. (1955) Effects of population size on the adrenal glands and-reproductive-organs of male mice in oonulations of fixed size. Am. J. Phvsiol 182, 292-300. Cl&titian J. J. (1956) Adrenal and reproductive &ponses to population size in mice from freely growing populations. EeoIogy 37, 258-273, Christian J. J. (1971) Population density and reproductive efficiency. Biol Reprod. 4, 248-294. Christian J. J. (197.5) Hormonal control of population growth. In Hormonal Correlates of Behaviour@&ted by Ele~herious B. E. and Sorott R. L.). ,, Vol. 1.. DD. . . 205-274. Plenum, New York. n Christian J. J. and Davis D. E. (1966) Adrenal glands in female voles (Microrus pennsyluanicus)as related to reproduction and population size. J. Mammal. 47, l-l 7. Christian J. J., Lloyd J. A. and Davis D. E. (1965) The role of endocrines in the self-regulations of mammalian populations. Rec. Prog. Hormoie Res. 21, 501-571. Cannes J. C. and Bradlev E. L. (1984) Serum ACTH and adrenal histology in reprodu&ively inhibited male prairie deermice (Peromysc~ ~nj~la~ bairdii). Camp. Biochem Physiol. 78A, 297-306.
Corticosterone
and reproduction in P. Ieucopus
Creigh S. L. and Terman C. R. (1988) Reproductive recovery of inhibited male prairie deermice (Peromyscus maniculatus bairdii) from laboratory populations by contact with females or their urine. J. Mammol. 69, 603607. Haigh G. R. (1983) Reproductive inhibition and recovery in young female (Peromyscus leucopus). J. Mammal. 64,706. Haigh G. R. (1987) Reproductive inhibition of female Peromyscus leucopus: female competition and behavioral regulation. Am. 2001. 27, 8677878. Kitay J. I., Coyne M. D., Swyzert N. H. and Gaines K. A. (1971) Effects of gonadal hormones and ACTH on the nature and rates of secretion of adrenocortical steroids by the rat. Endocrinology 89, 565-570. McKeever S. (1964) Variation in the weight of the adrenal pituitary and thyroid gland of the white-footed mouse, Peromyscus maniculatus. Am. J. Anat. 114, l-15. Nequin L. G. and Schwartz N. B. (1971) Adrenal participation in the timing of mating and LH release in the cyclic rat. Endocrinology 88, 325-331. Purushotham K. R., Mohano Rao A. M. K. and Rajabai B. S. (1978) Adrenal response to crowding in the spiny
457
field mouse, Musplatythrix (Bennett). Gen. camp. Endocr. 36, 439-441. Sung K-L. P., Bradley E. L. and Terman C. R. (1977) Serum corticosterone concentrations in reproductively mature and inhibited deennice (Peromyscus maniculatus bairdii). J. Reprod. Fert. 49, 201-206.
Terman C. R. (1965) A study of population growth and control exhibited in the laboratory by prairie deermice. Ecology 46, 890-895.
TermanC. R. (1966) Population fluctuations of Peromyscus maniculatus and other small mammals as revealed by the North American census of small mammals. Am. Midl. Nat. 76, 419426. Tennan C. R. (1969) Weights of selected organs of deermice (Peromyscus maniculatus bairdii) from asymptotic populations. J. Mammol. SO, 311-320. Terman C. R. (1973) Recovery of reproductive function by prairie deermice from asymptotic populations. Anim. Behav. 21, 44348.
To L. P. and Tamarin R. H. (1977) The relation of population density and adrenal gland weight in cycling and noncycling voles (Microtus). Ecology 58, 928-934.
030&9629/92 $5.00 + 0.00
Printedin GreatBritain
0 1992PergamonPm.9Ltd
CORRELATIONS BETWEEN SERUM CORTICOSTERONE CONCENTRATION AND REPRODUCTIVE CONDITIONS IN THE WHITE-FOOTED MOUSE (PEROMYSCUS LEUCOPUS NOVEBORACENSIS) STERLINGN.
RANSONE
JR and ERIC L. BRADLEY*
Laboratory of Endocrinology and Population Ecology, Biology Department, College of William and Mary, Williamsburg, VA 23187, U.S.A. Tel.: (804) 221-2220; Fax: (804) 221-6483 (Received
29 November
1991)
Abstract-1. Adrenal size and activity in reproductively inhibited young born into laboratory populations of the white-footed mouse were examined. Measurements were made of body weight, reproductive organ weight and adrenal weight. Serum corticosterone was measured by radioimmunoassay. 2. Data from reproductively inhibited animals were compared with corresponding values from reproductively capable animals of the same sex. The mean paired testis and seminal vesicle, or paired ovary and uterus, weights, were significantly reduced in reproductively inhibited animals of both sexes. The paired adrenal weights were not different in any comparison. 3. Reproductively inhibited males selected from laboratory populations had a mean corticosterone concentration that was significantly higher than the corresponding value for reproductively capable males. Females showed no significant difference in mean serum corticosterone concentrations. 4. The results are discussed in relation to earlier studies in two species of Peromyscus and the apparent paradox of elevated serum corticosterone in the absence of adrenal hypertrophy.
The
prairie
deetmouse,
maniculatus,
rarely 1940),
natural population (Blair, undergoes much population density over a time-span than species of mammals (Terman, Likewise, the mouse, Peromyscus shows a range of size fluctuation, this range typically covered a shorter span (Terman, Apparently, these Peromyscus species an intrinsic to regunatural population at well the physical capacity of environment. Laboratory of the species also numerical growth to either failure of to survive most commonly, a cessation reproduction within population when of the productive population fail to Most dramatically, majority (usually than 90%) the young into the lation do mature reproductively maintain smaller organs compared reproductively animals (Terman, 1966, 1969; and Terman, Haigh, 1983; and Terman, One explanation reproductive inhibition some rodent populations has that an in relative or “crowding” an hypophysial-adrenal response which hypersecretion of hormone (ACTH) adrenal steroids ultimately disrupts (Christian, 1950, 1956, *To
all correspondence
he addressed.
Christian et 1965). Indeed, dependent, adrenal-based alterations have well documented Mus, Rattus Microtus (Christian, 1956; Brain, Andrews and 1979; To Tamarin, 1977). the prairie significantly elevated serum corticosterone has been in both of reproductively animals from relative to serum levels in control However, there no concomitant in the of the in these animals (Sung al., 1977; and Terman, Also, an study with species did demonstrate a difference in serum ACTH of reproductively animals from populations compared reproductively capable (Coppes and 1984). This extends to Peromyscus species, evaluation of adrenal size serum corticosterone related to condition within populations. MATERULS
METBOBS
Animal
The animals this study myxus kucopus
laboratory tained in cage,
white-footed mice outbred Colony and animals were two-chambered, wire-topped, plastic derived from
12.8 x x 14Scm. (Prolab Rat, Hamster 3000, Inc., Syracuse, and tap were continuously Cage bedding of wood which were every 2 Room temperature regulated between f 3°C 5 to ah exchanges nour. The cycle consisted 14 hr
453
454
STERLING N. RANSONE JR and ERIC L. BRADW
light (15 A-c. at the floor) turned on between 7 a.m. and 9p.m. EST. Colony young were weaned at 21 days of age and placed into cages with same-sex sibs. At 60 f 3 days of age, 50 females were paired with 50 non-sib. males. The first 30 pairs to produce young were used to found 10 laboratory populations and the remainder of the pairs were used to produce control animals. Experimental animals
Control animals were produced from 12 colony pairs. Young were weaned at 21 days of age and placed individually into a cage compartment so that each double-sided cage contained one male and one female that was within 1 day of age of each other. Neither animal could see or touch the other. At 35 It 1 days of age each pair was switched to the opposite side of the compartment onto the soiled bedding of the other in order to maintain olfactory contact with the corresponding mate. This regimen was repeated twice more at 2-week intervals. At 70 f 1 days of age the animals were sacrificed for tissue collection. Each of 10 laboratory populations was founded with three pairs of reproductively proven animals ranging in age from 83 to 161 days old. Only young produced in the population context were retained. Each population was maintained within an enclosure consisting of a 1.5 m diameter stainless steel base with aluminum siding walls at a height of at least 68 cm. Six one-quart plastic containers were provided as nest-boxes in each enclosure and bedding consisted of wood chips. Food, water and environmental conditions were the same as those described for controls. Each population was inspected at intervals of 2 weeks. Males were noted as either having scrotal or non-scrotal testes and females were checked for perforate or nonperforate vaginae. All population founding animals were toe-marked and young were marked during the subsequent inspections. Selection of reproductively inhibited animals from populations
When a litter reached 68 days of age, an additional population inspection was performed to identify young that had never been noted as having scrotal testes or a perforate vagina. During the check, a randomly selected male and female from the litter were marked with a non-toxic UVfluorescing dye (Ultraviolet Products Inc., San Gabriel, CA) to facilitate later identification. All animals were returned to the enclosure and remained undisturbed for the next 2 days. Tissue sampling was accomplished between 6:45 and 7: 15 p.m. EST (2 hr f 15 min prior to darkness) by locating the dye-marked animals with a long-range UV light and rapidly removing one to a jar containing diethyl ether vapor. Only one animal per day was removed from a given population and its marked littermate was removed 48hr later. Male population animals were sampled from six different populations and females were sampled from seven populations. Possible inter-population differences were tested using a Kruskal-Wallis one.-way analysis of variance. No significant differences were noted across populations in any of the variables that were measured for either sex and so the data from all populations were combined by sex. Collection of tissues
A ventral abdominal incision was made under diethyl ether anesthesia and blood was collected from the left renal artery in less than 2 min from the time of first contact with the animal. Blood was placed into a 1.5 ml plastic centrifuge tube to clot and then centrifuged at 9000 g for 2 min. The serum was drawn off and frozen at below -70°C until analysed. The body was weighed to the nearest 0.1 g. Both adret& and the reproductive organs (testes and seminal vesicles or ovary and uterus) were removed and placed directly in a
10% buffered formaldehyde solution. Organs were dissected free of fat and weighed to the nearest 0.1 mg after at least 72 hr in the fixative. Radioimmunoassay for corticosterone
Corticosterone antisera B3-163 (Endocrine Sciences, Tanana, CA) was diluted 1: 85 according to the specifications of the manufacturer. Standards for each assay were run in triplicate with 0.063,0.125,0.250,0.500, 1.000, 1.250 and 2.5OOng/tube of authentic corticosterone (SchwartzMann, Grangeburg, NY). A standard pool of P. feucopus female sera was assayed in duplicate at 1,0.5,0.25 and 0.125 times natural concentration. The assay protocol was conducted as previously described (Bradley and Terman, 1981a). Samples and standards were counted to a 1% error (Beckman LS-3 133T), calculated as per cent of total counts added, and then logit transformed. Standard hormone concentration values were transformed to the log of the concentration. The slopes of the four standard curves and the four serial dilutions of the standard sera pools were compared using the BMDLR multiplaregression program (Biomedical Computer Programs, University of California, Berkeley, CA). No significant diiTerences in the slope of any regression were found. The mean limit of detection calcu-
lated from the variation of the buffer control tubes was 0.46 f 0.01 pgjtube. Hormoneconcentrationsarc expressed in ng/ml of sera. Statistics
A one-way ANOVA indicated heterogeneous variance and so the nonparametric Mann-Whitney U-test was used for comparisons. All correlation data was analysed using Spearman’s nonparametric ranked correlation test. All data are reported as the mean value f the standard error of the mean. In all cases P < 0.05 was considered statistically significant.
Comparironsbetween sexes Control males were significantly (P < 0.001) heavier than control females, but no such differences were observed between the sexes among the reproductively inhibited animals selected from populations. Also, there were no significant differeuces in mean absolute adrenal weights between the sexes in control or population animals (Tables 1 and 2). Weight comparbons between control and inhibited animaIs The mean body weight of reproductively inhibited males selected from populations was signi&ntly (P < 0.001) lighter than the corresponding control male value (Table 1). No corresponding differences in body weight were observed between the mean body weights for population vs control females (Table 2). The mean absolute adrenal weights were not significantly diRerent between controls and reproductively inhibited population males or females (cf. Tables 1 and 2). Likewise, the calculated relative adrenal weights for these comparisons were not different. The mean paired testes weights and seminal vesicle weights were both sign&a&y (P < 0.001) larger in controls compared with reproductively inhibited population males (Table 1). Similarly, among females, both the mean paired ovary and uterus weights were signScantly (P < 0.001) larger in controls compared with the corresponding values ftom population females (Table 2).
Corticosterone and reproduction in P. leucopus Table
455
I. Mean body weight, testis weight, seminal vesicle weight, adrenal weight and serum corticosterone concentration
Treatments Control males (N = 19)
Body weight (g) 20.09 * 0.43
Population males
16.8 + 0.71*+*
in control and population males Seminal vesicle weight (mg) 112.6 k 7.56
Testis weight (mg) 273.3 + 14.13 84.3 + 20.77***
11.0 f 4.16***
Values are mean 1?ISEM. *P < 0.05, ***P< 0.001, “‘not significant, with respect to corresponding
Serum corticosterone concentrations
There were no statistically significantly differences between the mean serum corticosterone concentrations in male controls compared with female controls or between reproductively inhibited male and female animals taken from populations. However, the mean serum corticosterone concentration of reproductively inhibited males taken from populations was significantly (P < 0.05) elevated compared with control males (Table 1). DISCUSSION
Of all of the young born into the ten laboratory populations and reaching 70 days of age, 29 of 33 males had undescended testes and 36 of 37 females were always vaginally imperforate. The drastic nature of the reproductive inhibition in males selected from populations was further indicated by the significantly smaller size of both the testes and seminal vesicles compared with control males (Table 1). Likewise, females selected from populations showed significantly smaller ovarian and uterine weights compared with their reproductively proven controls (Table 2). The degree of reproductive inhibition reported here is entirely similar to reports on other laboratory populations of P. leucopus (Haigh, 1987; Creigh and Terman, 1988) and also laboratory populations of P. maniculatus (Terman, 1969, 1973; Bradley and Terman, 1981a,b,c). The mean body weight of reproductively inhibited males selected from populations was significantly (P < 0.001) less than the mean weight for control males. However, no such significant reduction in mean weight was observed between females taken from populations and their respective controls. These results are similar to those reported by Bradley and Terman (1981a) for P. maniculatus, and indicates that, in the female of both species, reproductive inhibition may not be directly associated with smaller body weight.
Adrenal weight (mg) 10.6 f 0.47 9.0 k 0.86”’
Serum corticosterone (ng/ml) 145.3 + 14.41 201.9 + 16.05.
control value.
Neither absolute nor relative mean adrenal weight was significantly different between reproductively inhibited population and control males (Table 1) or females (Table 2). Among males there were no statistically significant correlations between the adrenal weight and reproductive organ weight. However, in control females ovary weight was positively correlated (r = 0.7806; P < 0.001) with adrenal weight. This relationship is very likely a reflection of the interrelationship between the ovary and adrenal during the normal ovarian cycle. These findings in P. Ieucopus are very similar to earlier studies with P. maniculatus where no population density-related adrenal weight increase was found for either growing or asymptotic laboratory populations (Bronson and Eleftheriou, 1963; Terman, 1966; Sung et al., 1977; Bradley and Terman, 1981a) or in crowded natural populations (McKeever, 1964). This lack of an adrenal hypertrophy response in Peromyscus is distinctly different from reports in such genera as Mus, Rattus and Microtus where there is a marked density-dependent adrenal hypertrophy in response to population density increase (Bronson and Eleftherious, 1963; Christian, 1955, 1956; Purushotham et al., 1964; Christian and Davis, 1966; Andrews and Belknap, 1979). Corticosterone comparisons
Control female white-footed mice in this study only tended (P < 0.09) to have higher serum corticosterone concentrations than control males. Bradley and Terman (1981a) reported a significantly higher serum corticosterone concentration in control female P. maniculatus compared with control males. This difference between the two reports might be due to some species variation in adrenal function, but it may also be due to females in different stages of their ovarian cycle (Nequin and Schwartz, 1971; Kitay et al., 1971; Barfield and Lisk, 1974). The mean serum corticosterone concentration in control male P. feucopus was over twice as high as the
Table 2. Mean body weight, ovary weight, uterus weight, adrenal weight and serum corticosterone concentration in control and population females
Treatments Control females (N=l8) Population females (N = 12)
Body weight (8) 17.7 * 0.61 16.6 * 0.68”’
Ovary weight (mg) 11.6kO.95 4.5 f 0.66***
Values are mean f SEM. l**P< 0.001, “hot significant, with respect to corresponding
Uterus weight (me) 49.5 k 5.36 11.8 f 2.58***
control value.
Adrenal weight (mg) 11.4f0.68 9.8 f 0.83”’
Serum wrticosterone (nglml) 189.1 + 19.25 168.5 f 16.35”’
456
STERLING N. RANSONE
mean level (68.8 ng/ml) reported for P. municulu&s by Bradley and Terman (1981a). This difference between the species is believed to be actual because a two-fold difference was also detected in the corticosterone values measured in standard sera pools from both species run with each assay in the present study. This apparent species difference is also interesting because the slightly heavier mature P. leucopus has a relative adrenal weight that is more than three times greater than P. manicula~us. Among control male P. i&opus there is a positive correlation of serum corticosterone concentration with adrenal weight (I = 0.4967; P = 0.042), although among the control females there was only a trend toward this same correlation (r = 0.3841; P = 0.116). Apparently, in reproductively proven P. leucopus of this age and weight range, there is a previously unidentified reIationship between increased adrenal weight and increased serum corticosterone concentration. In earlier studies on reproductively inhibited P. maniculatus selected from laboratory populations and compared with controls, Bradley and Terman (198 la) reported a tendency towards smaller adrenal glands concomitantly with significantly elevated serum corticosterone concentration. In this study, reproductively inhibited P. leucopus males had a mean serum corticosterone concentration that was also significantly (P < 0.04) elevated when compared with control males; however, the mean adrenal weights were not significantly decreased as in P. maniculatus. Also, in female P. Ieucopus there was no significant difference in the serum corticosterone concentration of population female vs control (Table 2), which is distinctly different from the case in female P. maniculatus. The serum concentration of corticosterone in reproductively inhibited P. Zeucopus males in this study, while higher than control males, was much lower than the mean level (348 ng/ml) measured previously in P. maniculatus inhibited males (Bradley and Terman, 1981a). It is also of interest that in earlier reports (Sung et al., 1977; Bradley and Terman, 1981a) the mean serum corticosterone concentration (178.3 ng/ml) for control female deermice was very similar to the value reported here for control female white-footed mice. The observation that there was a corresponding significant increase in the serum corticosterone concentration in white-footed female mice selected from populations is comparable with the si~nifi~ntly elevated mean value (326.3 ng/ml) reported for reproductively inhibited female P. maniculatus in the earlier study (Bradley and Terman, 1981a). The higher serum corticosterone concentrations found in mafes selected from populations may indicate a greater sensitivity to population conditions than shown by females. Males clearly experienced the highest rate of death due to wounding compared with females that were seldom observed to fight. However, no systematic behavioral observations were made that would confirm this notion. Taken together the data from studies on Peromyscus do not support the notion that a stressinduced adrenal hy~rtrophy is central to the reproductive inhibition that is so clearly produced in
JR and ERICL. BRADLEY the vast majority of young born into laboratory populations. However, it is also apparent that an elevated serum corticosterone concentration is associated with the reproductively inhibited condition. Whether the elevated corticosterone level is caused by a reduction in metabolic clearance of the steroid or an increase in steroid sequestration due to an increase in protective plasma protein binding (Bradley and Terman, 1981a) is not yet clear for this species. In order to address the apparent paradox of increased corticosterone concentration without an increase in adrenal weight, work is now in progress to evaluate possible adrenal zone area differences in the otherwise similar weight adrenals of reproductively inhibited animals. Also, the metabolic clearance of corticosterone and the general metabolic state of reproductively inhibited animals is under investigation. REFERENCES Andrews R. V. and Belknap R. W. (1979) Deer mouse and lemming adrenal and pathological responses to increases in animal numbers. Comp. Biochem. Physioi.GA, 15-18. Barfteld M. A. and Lisk R. D. (1974) Relative cont~bution of ovarian and adrenal progesterone to the timing of heat in the 4-day cyclic rat. Endocrinology94, 571-575. Blair W. F. (1940) A study of prairie deer-mouse populations in southern Michigan. Am. Mid!. Nat. 24, 272-304. Bradley E. L. and Terrnan C. R. (198la) A comparison of the adrenal histology, reproductive condition and serum corticosterone concentrations of prairie deermice (Peromyscus manictdatusbairdiiifin captivity. J. Mammol. 62, 353-361. Bradley E. L. and Terman C. R. (1981b) Serum testosterone concentrations in male prairie deermice (Peromyscus maniculatusbairdii). J. Mammol. 62, 811-814. Bradley E. L. and Terman C. R. (1981~) Studies on the nature of ~r~uctive inhibition in animals from laboratory populations of prairie deermice (Peromyscus maniculurusbairdii): serum LH and FSH concentrations. Comp. Biochem. Physiol. MA, 563-570. Brain P. F. (1971) The physiology of population limitation in rodents. Commrm. B&v. Biol. 6, 115-123. Bronson F. H. and Eleftherious B. E. (1963) Adrenal response to crowding in Peromyscus and C57BL/lOJ mice. Physiol. Zool. 36, 161-166. Christian J. J. (1950) The adreno-pituitary system and ~p~ation cycles in mammals. J. Moon. 31,247-259. Christian J. J. (1955) Effects of population size on the adrenal glands and-reproductive-organs of male mice in oonulations of fixed size. Am. J. Phvsiol 182, 292-300. Cl&titian J. J. (1956) Adrenal and reproductive &ponses to population size in mice from freely growing populations. EeoIogy 37, 258-273, Christian J. J. (1971) Population density and reproductive efficiency. Biol Reprod. 4, 248-294. Christian J. J. (197.5) Hormonal control of population growth. In Hormonal Correlates of Behaviour@&ted by Ele~herious B. E. and Sorott R. L.). ,, Vol. 1.. DD. . . 205-274. Plenum, New York. n Christian J. J. and Davis D. E. (1966) Adrenal glands in female voles (Microrus pennsyluanicus)as related to reproduction and population size. J. Mammal. 47, l-l 7. Christian J. J., Lloyd J. A. and Davis D. E. (1965) The role of endocrines in the self-regulations of mammalian populations. Rec. Prog. Hormoie Res. 21, 501-571. Cannes J. C. and Bradlev E. L. (1984) Serum ACTH and adrenal histology in reprodu&ively inhibited male prairie deermice (Peromysc~ ~nj~la~ bairdii). Camp. Biochem Physiol. 78A, 297-306.
Corticosterone
and reproduction in P. Ieucopus
Creigh S. L. and Terman C. R. (1988) Reproductive recovery of inhibited male prairie deermice (Peromyscus maniculatus bairdii) from laboratory populations by contact with females or their urine. J. Mammol. 69, 603607. Haigh G. R. (1983) Reproductive inhibition and recovery in young female (Peromyscus leucopus). J. Mammal. 64,706. Haigh G. R. (1987) Reproductive inhibition of female Peromyscus leucopus: female competition and behavioral regulation. Am. 2001. 27, 8677878. Kitay J. I., Coyne M. D., Swyzert N. H. and Gaines K. A. (1971) Effects of gonadal hormones and ACTH on the nature and rates of secretion of adrenocortical steroids by the rat. Endocrinology 89, 565-570. McKeever S. (1964) Variation in the weight of the adrenal pituitary and thyroid gland of the white-footed mouse, Peromyscus maniculatus. Am. J. Anat. 114, l-15. Nequin L. G. and Schwartz N. B. (1971) Adrenal participation in the timing of mating and LH release in the cyclic rat. Endocrinology 88, 325-331. Purushotham K. R., Mohano Rao A. M. K. and Rajabai B. S. (1978) Adrenal response to crowding in the spiny
457
field mouse, Musplatythrix (Bennett). Gen. camp. Endocr. 36, 439-441. Sung K-L. P., Bradley E. L. and Terman C. R. (1977) Serum corticosterone concentrations in reproductively mature and inhibited deennice (Peromyscus maniculatus bairdii). J. Reprod. Fert. 49, 201-206.
Terman C. R. (1965) A study of population growth and control exhibited in the laboratory by prairie deermice. Ecology 46, 890-895.
TermanC. R. (1966) Population fluctuations of Peromyscus maniculatus and other small mammals as revealed by the North American census of small mammals. Am. Midl. Nat. 76, 419426. Tennan C. R. (1969) Weights of selected organs of deermice (Peromyscus maniculatus bairdii) from asymptotic populations. J. Mammol. SO, 311-320. Terman C. R. (1973) Recovery of reproductive function by prairie deermice from asymptotic populations. Anim. Behav. 21, 44348.
To L. P. and Tamarin R. H. (1977) The relation of population density and adrenal gland weight in cycling and noncycling voles (Microtus). Ecology 58, 928-934.
