Day length and estradiol affect same-sex affiliative behavior in the female meadow vole

Abstract

Non-sexual social bonding between adult mammals is poorly understood, despite its importance in many species. Female meadow voles are territorial and nest alone in long summer day lengths when circulating estradiol concentrations are high, but cohabit in groups in short winter photoperiods when estradiol secretion is low. The influence of day length and estradiol on same-sex huddling behavior was assessed in adult female pairs housed together in long day lengths (LDs) or short day lengths (SDs) from weaning. The behavior of intact, ovariectomized, and estradiol-treated ovariectomized females from each photoperiod was assessed during 3 h partner-preference tests. Intact SD voles, unlike intact LD voles, spent the majority of the test in proximity to their cage-mates. Estradiol treatment of SD voles significantly reduced time spent huddling with the partner. Neither ovariectomy nor estradiol treatment significantly affected the amount of time LD females spent in contact with their partners. Low estradiol availability is therefore a necessary but not sufficient condition for maintenance of high levels of huddling. These results establish that ovarian hormones interact with photoperiod to affect same-sex social behavior.

Fig 1

(A) Mean (± SEM) time focal females spent huddling with the partner (■) or the stranger () during a 3 h test. (B) Mean (± SEM) time focal females spent in the same chamber as the partner or the stranger during a 3 h test. Asterisks denote significant differences within groups (between partners and strangers). Letters denote differences between treatment groups — groups labeled “a” are not different from LD intact voles, whereas groups labeled “b” are significantly different from LD intact animals (P <0.005, Dunnet’s test).

Fig 2

(A) Mean uterine mass (± SEM) for a subset of focal females from each treatment group (uterine mass is a proxy for, and is correlated with exposure to circulating estradiol). Groups labeled “a” are significantly different from those labeled “b” (Tukey-Kramer test). (B) Bivariate analysis of total huddling time by uterine mass. At low uterine masses (< 30 mg) social behavior was varied, whereas voles with high uterine masses exhibited little or no huddling behavior.

Fig 3

Rhythms in body temperature of LD and SD voles (n=6 females/group; mean ± SEM). (A) Body temperature as a function of clock time. The shaded area represents the time of social testing; groups did not differ in T_b during this interval. (B) Body temperature as a function of hours since light onset. In this view T_b rhythms in LD and SD are strikingly similar, with an ultradian period of ~3 hrs.