Long-term ovariectomy alters social and anxious behaviors in semi-free ranging Japanese macaques

Abstract

Symptoms of anxiety and depression often occur in young women after complete hysterectomy and in older women during menopause. There are many variables that are hard to control in human population studies, but that are absent to a large extent in stable nonhuman primate troops. However, macaques exhibit depressive and anxious behaviors in response to similar situations as humans such as isolation, stress, instability or aggression. Therefore, we hypothesized that examination of behavior in ovariectomized individuals in a stable macaque troop organized along matriarchal lineages and in which individuals have social support from extended family, would reveal effects that were due to the withdrawal of ovarian steroids without many of the confounds of human society. We also tested the hypothesis that ovariectomy would elicit and increase anxious behavior in a stressful situation such as brief exposure to single caging. Japanese macaques (Macaca fuscata) were ovariectomized (Ovx) or tubal-ligated (intact controls) at 3 years of age and allowed to mature for 3 years in a stable troop of approximately 300 individuals. Behaviors were recorded in the outdoor corral in the third year followed by individual temperament tests in single cages. There was no obvious difference in anxiety-related behaviors such as scratching between Ovx and tubal-ligated animals in the corral. Nonetheless, compared to tubal-ligated animals, Ovx animals exhibited a significant decrease in (1) positive social behavior, (2) initiating dominance behavior, (3) time receiving grooming, (4) locomoting, (5) mounting behavior, and in (6) consort behavior. However, Ovx females exhibited a significant increase in (1) consummatory behavior and (2) object play compared to tubal-ligated controls. In the individual temperament tests, Ovx individuals exhibited an increase in anxiety-related behaviors. There was no difference in adrenal weight/body weight suggesting that neither group was under chronic stress. These data indicate that ovarian hormones enable females to successfully navigate their social situation and may reduce anxiety in novel situations.

Figure 1

Percent of time OVX and tubal ligated (intact) Japanese macaques spent in positive social contact (touch, groom, ventral contact), frequency of dominance behaviors (aggression, chase, displace, threat) and percent of time monkeys spent being groomed by others (receive groom) during behavioral observations taken during the mating and birthing seasons. All animals spent more time engaged in positive social behaviors in the mating season compared to the birthing season (F (1,8)= 21.208, p=0.002). Regardless of treatment, OVX monkeys spent less time in these behaviors compared to intact animals (F (1,8)= 8.129, p=0.02). Intact monkeys initiated dominance-related behaviors significantly more than Ovx monkeys in the mating season (Mann Whitney U= 2.0, P=0.03), but not in the birthing season (Mann Whitney U= 5.0, p=0.11). The intact monkeys also received more grooming from other monkeys than Ovx monkeys (F (1,8)=16.05, p=0.004).

Figure 2

The frequency of mounting behavior and percent of time spent in consort behavior (e.g., grooming, touching, ventral contact, and close social proximity with males) for OVX and tubal-ligated (intact) Japanese macaques during behavioral observations taking during the mating season. The intact monkeys were mounted by males more than the Ovx (Mann Whitney U= 1.00, p=0.01) and tended to spend more time in consortship behavior with males than Ovx monkeys (t= −2.23, df=8, p=0.056).

Figure 3

Figure 3: Percent of time OVX and tubal ligated (intact) Japanese macaques spent in consummatory behavior, interacting with objects (object play), locomotion and the frequency of scratches during behavioral observations taken during the birthing and mating seasons. Ovx monkeys consumed more food and water than the intact monkeys (F (1,8)=12.94, p=0.007). They also spent more time in object play than intact monkeys during the birthing (Mann Whitney U= 22.5, p= 0.04) but not mating (U= 18.5, p=0.21) seasons. Intact monkeys spent more time walking and running than the Ovx monkeys (F (1,8)= 4.976, p=0.056). There were no difference between OVX and intact animals with respect to frequency of scratch (F (1,8)=0.02, p=0.88).

Figure 4

Illustration of selected behaviors observed during the Human Intruder test conducted in year 3. [A] Ovx females tended to freeze longer than tubal ligated animals when exposed to a human profile (t-test = 1.52, df=8, p = 0.17). [B] During the following control period 2 (C2), the OVX monkeys were more likely to continue to freeze than were the tubal-ligated females (Pearson’s Chi square = 4.29, df = 1, p = 0.04). [C] Ovx females tended to freeze longer than tubal-ligated animals when exposed to a human making direct eye contact (Stare; Mann Whitney U= 19.0, p=0.18) and [D] After the human intruder left (Control 3), the Ovx monkeys tended to sleep more than the tubal-ligated monkeys, but the difference was not statistically significant (Mann Whitney U=18.0, p=0.19)

Figure 5

Illustration of selected behaviors observed during the Novel Object tests conducted in year 3. Each animal was exposed to a novel food (kiwi), a novel toy and a rubber snake. There was no difference between the groups in the latency to touch or amount of time spent exploring the kiwi (data not shown). [A] None of the Ovx animals and only two of the intact touched the toy within the 5 min (300 sec) episode. When we categorized the responses there was no statistically significant difference between the groups (Chi square= 2.50, df= 1, p=0.11). [B] There was no difference between the groups in the time spent exploring the toy (Mann Whitney U= 10.5, p=0.64). [C] There was no difference between the groups in the percent of time spent with eyes closed in the presence of the toy (Mann Whitney U= 17.0, p=0.33). [D] Intact animals were more likely than Ovx animals to approach and touch the snake during the 5 min period (Chi square = 3.60, df= 1, p= 0.058). [E] The Ovx animals spent significantly less time exploring the snake than the intact monkeys (Mann Whitney U= 3.5, p=0.05). [F] There were significant differences in whether or not the monkeys fell asleep in the presence of the snake (Chi square= 6.67, df= 1, p= 0.01). Four of the OVX and none of the intact monkeys fell asleep during this episode.

Figure 6

Adrenal to body weight ratio in ovariectomized and tubal ligated Japanese macaques. Three years of ovariectomy had no effect on the adrenal to body weight ratio (two tailed t-test; t= 0.939, df=7, F= 2.38, p=0.37).