- Email: [email protected]
Role of maternal behavior on aggression, fear and anxiety
Physiology & Behavior 77 (2002) 197 – 204
Role of maternal behavior on aggression, fear and anxiety A. Ferreiraa, M. Pereiraa, D. Agratia, N. Uriartea, A. Ferna´ndez-Guastib,* a
b
Facultad de Ciencias, Universidad de la Repu´blica, Montevideo, Uruguay Departamento de Farmacobiologı´a, CINVESTAV/IPN, Calzada De los Tenorios 235, Col. Granjas Coapa, Me´xico 14330 D.F., Mexico Received 14 February 2001; received in revised form 10 April 2002; accepted 28 May 2002
Abstract Concomitant to the expression of maternal behavior, the lactating female develops anxiolysis in the elevated plus maze test, aggression towards intruders and reduced fear in response to a sudden auditory stimulus. This study aims to determine if these behavioral changes are associated with maternal behavior independently of the endocrine status that characterizes gestation, parturition and lactation. To assess this purpose, the behavior of lactating females was compared to that exhibited by maternal and nonmaternal ovariectomized rats untreated with steroid hormones. In contrast with lactating dams, sensitized animals (rats that displayed maternal behavior after a continuous contact with young pups) did not display reduced anxiety in the plus maze test. However, the sensitized females showed behaviors characteristic of lactating rats, such as some components of maternal aggression and reduced fear, though much less intensely than dams. These results suggest that aggression and reduced fear, but not anxiolysis, partially depend on the development of maternal behavior. D 2002 Elsevier Science Inc. All rights reserved. Keywords: Maternal aggression; Fear; Anxiety; Sensitization; Maternal behavior
1. Introduction The behavioral repertoire of some mammals during motherhood differs from that exhibited by females in other periods of their reproductive cycle. Concomitant with the display of maternal behavior, lactating rats show lower levels of anxiety in conflict tests [14], open-field paradigms [16] and plus maze test [2,23,27], exhibit less fear towards a sudden auditory stimulus [21] and develop aggression when confronted with intruders [9,36,49]. This behavioral pattern relies, at least partly, on the hormonal changes that characterize late gestation, parturition and lactation [2,15,20]. As maternal care parallels the course of these behaviors, it might be suggested that maternal behavior is a condition needed for the expression of anxiolysis, aggression and fear reduction. Most components of maternal behavior can be induced in virgin or even ovariectomized animals by a continuous exposure to foster pups. This process, called ‘‘sensitization’’ [8,17,42], is shortened by steroid treatment simulating the endocrine profile of late gestation and parturition [5,29]. Most studies show that sensitized females treated with sex * Corresponding author. Tel.: +52-54-83-28-56; fax: +52-54-83-28-63. E-mail address: [email protected] (A. Ferna´ndez-Guasti).
steroids, develop aggression [29,30] and show less fear [20], behavioral features characteristic of lactating animals. These results suggest that both endocrine factors and maternal care contribute to the expression of aggression and fear reduction. The purpose of the present study was to analyze whether the reduced anxiety and fear and the increased aggression observed in lactating rats are related to the expression of maternal care. The sensitization of ovariectomized females—without hormonal treatment—is a useful model to examine the role of maternal behavior on the reduced anxiety and fear and on the expression of maternal aggression, independently of the endocrine changes that characterize gestation, parturition and lactation. With this aim, anxiety (measured as the time spent and number of entries into the open arms of an elevated plus maze) [37], maternal aggression [36] and fear (measured as the duration of immobility in response to a sudden auditory stimulus) [1] were tested in ovariectomized, sensitized and lactating rats. Sensitized females were ovariectomized to eliminate the possibility that ovarian steroid fluctuations, produced after prolonged contact with pups [10], affect the behaviors here recorded. This is particularly important, given the influence of progesterone and its metabolites on anxiety-related behaviors [2,3,38].
0031-9384/02/$ – see front matter D 2002 Elsevier Science Inc. All rights reserved. PII: S 0 0 3 1 - 9 3 8 4 ( 0 2 ) 0 0 8 4 5 - 4
198
A. Ferreira et al. / Physiology & Behavior 77 (2002) 197–204
As the presence of the litter is relevant for the reduced freezing and can induce or potentiate anxiolysis in the openfield and conflict tests [14,21], it is hypothesized that the litter presence may induce anxiolysis in the sensitized females. On these bases, the tests were performed in the attendance of the litter. The site where the pups should be located in the plus maze test remained as a puzzling question. In the present study, the pups were always located in the closed arms, since it was observed that when the newborn were placed in the open arms, the lactating females spend long intervals in these arms, without necessarily revealing a decreased anxiety but rather maternal motivation (unpublished results). Prolonged isolation increases aggression in males [52,53]. Although the groups of lactating and sensitized rats were not completely isolated, but obviously housed with pups, a putative behavioral change in aggression might be caused by the fact that maternal females were individually housed while the ovariectomized animals inhabited in groups. On these bases, the behavior of ovariectomized females housed in groups was compared to that of ovariectomized animals housed individually for a similar period to that required for the sensitization procedure. Finally, to control for putative differences between lactating and sensitized females in the display of aggression, fear and anxiety due to various factors involved in the development of sensitization, an experiment was performed considering the number and pups manipulation as well as the cage measurements.
2. Materials and methods 2.1. Animals Three-month-old nulliparous Wistar females, males (weighing 250 –350 g) and 2- to 10-day-old pups were used in this study. All animals were housed in a room under inverted 12-h light–dark cycle (lights off at 15:00 h) with ad libitum access to water and rat chow throughout the experiments. Females were divided into three main groups: ovariectomized (n = 8), sensitized (n = 8) and lactating (n = 8). Ovariectomy was performed through a ventral incision under 40 mg/kg ip of pentobarbital anesthesia. An extra group of donor lactating females with their pups were maintained to provide a supply of foster pups for the sensitization procedure and for the elevated plus maze, aggression and freezing tests. In general, animals were housed as follows: Nonsensitized ovariectomized females were kept in groups of four in a cage measuring 45 28 26 cm. Sensitized ovariectomized females were individually placed in small cages (27 19 11 cm) to reduce the sensitization latency [39,51]. Lactating females were maintained with their pups (adjusted to six on the delivery day) in cages measuring 45 28 26 cm. For controlling the role of isolation on
aggression, a group of ovariectomized rats housed individually in small cages (27 19 11 cm) for 17 days was also included (n = 8). Additionally, to assess the role of cage measurements, number of pups and manipulation, a group of dams housed in small cages together with three foster pups daily changed (treated in the same way as the sensitized females) (n = 8) were compared to a group of lactating females placed in big cages together with six pups (n = 8). 2.2. Induction of maternal behavior (sensitization procedure) Three days after ovariectomy, the females were housed in individual cages. Three freshly fed pups (2 –10 days of age) were scattered in the corner of the home cage opposite to the female’ sleeping area. They were removed daily 1 h after the onset of darkness and replaced 5 min later with three new pups. Immediately after placing the pups, maternal behavior was tested [17,41]. The number of the following behaviors was recorded during a 5-min test: retrieving (the female picks up a pup and carries it to another location of the cage); crouching posture (the mother hovers over the pups remaining immobile except for occasional postural adjustment; every 10 s, a new score is recorded if the female remains in this position); licking (the female licks the pups) and nest building (the female pushes or picks up nest material in her mouth). Based on previous studies [6,17,34], retrieval of the three pups within the test period (5 min) during two consecutive days was used as a marker of maternal care. Subjects were assigned a sensitization latency corresponding to the second day of the two consecutive retrievals. Those animals that cannibalized pups and those that did not display maternal behavior after 26 days of pup exposure were excluded from the experiment. Because only 25% of the Wistar rats became maternal in a period of 10 –15 days [22], we selected a long period for the sensitization procedure. 2.3. Experimental procedure Each animal was tested in the experimental paradigms in the following sequence: Day 1, elevated plus maze test; Day 2, aggression test; and Day 3, freezing test. The lactating rats started the testing sequence on the 7th day postpartum, the sensitized group 1 day after being considered maternal and the ovariectomized group 15 days after surgery. The extra control group of ovariectomized females was tested for aggression after an isolation period of 17 days. In all the tests, the behavior of a single rat was recorded at a time by an observer placed 1 m from the testing model under red dim light. 2.4. Behavioral tests 2.4.1. Elevated plus maze test This test was carried out 2 h after the onset of darkness (at 17:00 h). This model consists of an elevated (40 cm
A. Ferreira et al. / Physiology & Behavior 77 (2002) 197–204
above the floor) plus-shaped maze with two opposite enclosed and two open arms, measuring 50 cm long 10 cm wide, placed in a red dimly lit room. Each animal was placed in the center of the maze facing a closed arm. Three foster pups were placed at the end of each closed arm. An entry into an arm was defined as the animal placing all four paws into it. The cumulative time spent in the open arms and the number of open- and closed-arm entries were recorded over a 5-min session. Data were expressed as the percentage of open arms entries (100 open/total), the percentage of time spent in the open arms (100 open/ total), the number of entries to the closed arms and the total number of entries (open + enclosed arms). The number of retrievals, groomings, defecations and rearings was also recorded. 2.4.2. Aggressive behavior The aggression test towards a male (300 – 350 g) was carried out in the female’s home cage 2 h after the onset of darkness (at 17:00 h). Aggression was assessed using a previously described procedure [9,10]. A male rat was placed in the female’s cage, and the frequency of the following behaviors recorded during a 5-min test: attack (or ‘‘frontal attack’’: female lunges toward the male, sometimes followed by rolling and scuffing for a brief period); bite (female mouth in contact with the intruder body and the male vocalizes); lateral posture (or ‘‘lateral attack’’: female approaches male side way presenting the lateral part of her body to him); and kick (with rear leg). The number of times the intruder male adopted an upright subordinate posture (sitting on haunches, curved back, forepaws at the female’s face) during the test was also recorded. The latencies to the first aggressive response displayed by the female (attack, bite or lateral posture) and to the first subordinate posture by the male were also recorded. Nonresponder animals were assigned a latency of 300 s. The stimulus males were never used more than once in the tests. Females were considered aggressive during an intruder test if they attacked more than once, or attacked once eliciting a submissive posture [29]. 2.4.3. Freezing behavior This test was carried out 2 h after the onset of darkness (at 17:00 h). This paradigm includes a circular Plexiglas arena (39-cm diameter), whose floor is covered by a filter paper divided by two lines to form four 90 sectors. The floor cover was changed after testing each rat. The circular arena is enclosed in a soundproof chamber 70 41 40 cm, which is illuminated by a 15-W white light bulb suspended together with a door bell (95 dB) on the ceiling of the test chamber. Each rat was placed together with three foster pups and its behavior recorded through a window on the front wall. The animal was allowed a 5-min adaptation period in the novel observation cage. During this time, the number of retrievals, groomings, rearings and defecations was recorded. After this period, the doorbell was activated for 6 s. During the sound, the rat typically rushed around in the
199
cage and eventually immobilized either concurrently with or slightly before the end of the signal. In both cases, the freezing duration comprises the time (in seconds) of immobility shown by the rat from the cessation of the signal to the first distinct body movement excluding eye blinks and respiratory or vibrissae movements [21]. The local ethical committee approved all the experimental procedures performed in this study. 2.5. Statistical analysis Most data are presented as median (IQR) (interquartile ranges) [43]. As most variables did not follow a normal distribution (Kolmogorov –Smirnov test) nor were variances homogeneous (Cochran C test), data were analyzed with nonparametric tests. The number of responses among the groups in the tests of anxiety, aggression and fear was analyzed by the Kruskal – Wallis one-way analysis of variance followed by Mann –Whitney U test. The nominal scale data were analyzed using Fisher exact probability test and c2 test for independent samples. The latencies for sensitized animals to show maternal behavior are expressed as means ± S.E.
3. Results Eight of nine animals (89%) submitted to the sensitization procedure showed pups’ retrieval, licking, nursing behavior and nest building. The latencies (mean days ± S.E.) required for the sensitized females to exhibit the following behaviors were: pups’ retrieval (after two consecutive days), 17.0 ± 2.0; nursing behavior, 6.8 ± 2.2; nest building, 10.5 ± 2.5; and licking, 2.3 ± 1.5. 3.1. Elevated plus maze The lactating females spent a higher percentage of time in the open arms (Fig. 1, Panel A, H = 19.6; P < .01) and showed higher percentage of entries into open arms (Panel B, H = 11.7; P < .01) than the sensitized and the ovariectomized rats. Lactating, as compared to both sensitized and control females, exhibited a statistically significant increase in the total number of entries and in the entries to the closed arms. Sensitized animals also showed an increase in these parameters when compared to ovariectomized rats (see Table 1). Both lactating and sensitized rats retrieved the pups, either in the same closed arm or from one closed arm to the other, but never took them into the open arms. No significant differences were found in the number of animals that showed retrieval (lactating: 8/8 vs. sensitized: 5/8; P = ns); however, the number of retrievals was significantly higher in lactating compared to sensitized females (median (IQR): lactating group, 5.0 (1.1); sensitized group, 1.0 (0.7); U = 2.5, P < .005). No differences were found in the number
200
A. Ferreira et al. / Physiology & Behavior 77 (2002) 197–204
Fig. 1. Percentage of time spent in the open arms (Panel A) and percentage of open arm entries (Panel B) in the plus maze test of ovariectomized, n = 8 (OVX), sensitized, n = 8 (SENS) and lactating, n = 8 (LACT) rats. Data are expressed as median (IQR) and were compared by the Kruskal – Wallis analysis of variance followed by Mann – Whitney U test. aP < .05 versus ovariectomized controls and sensitized.
of groomings, defecations and rearings among the three groups (data not shown). 3.2. Aggressive behavior Fig. 2A shows that the median (IQR) of the sum of aggressive behaviors (attacks, bites and lateral postures) was significantly different among the groups (H = 29.1; P < .001). Thus, lactating females showed more aggressive responses than the sensitized and the nonmaternal ovariectomized rats. The sensitized animals displayed more aggression than the ovariectomized control animals. Fig. 2B shows that male upright subordinate postures consequently follows the aggression levels of the females (H = 29.3; P < .001). In
Table 1 Number of entries into the closed-arm and total entries in the plus maze test of ovariectomized, sensitized and lactating rats
Ovariectomized Sensitized Lactating H P
Closed-arm entries
Total entries
1.0 (0.5)a 5.5 (0.75)b 11.0 (1.3)c 21.5 < .001
1.0 (0.5)a 6.0 (1.4)b 13.0 (2.0)c 21.5 < .001
Data are expressed as median (IQR) and were compared by the Kruskal – Wallis analysis of variance followed by Mann – Whitney U test. Different letters indicate statistical differences ( P < .01) among groups.
Fig. 2. (A) Number of aggressive responses (sum of attacks, lateral postures and bites) displayed by females towards male intruders. (B) Number of submission postures adopted by the male intruders. OVX: ovariectomized housed in groups (n = 8); OVX isolated: ovariectomized isolated during 17 days (n = 8); SENS: sensitized (n = 8); and LACT: lactating females (n = 8). Data are expressed as median (IQR). OVX, OVX isolated, SENS and LAC were compared by the Kruskal – Wallis analysis of variance followed by Mann – Whitney U test. aP < .01 versus ovariectomized controls; bP < .01 versus lactating rats.
contrast, none of the ovariectomized nonmaternal rats (housed in groups or kept isolated for 17 days) displayed aggressive responses during the test (see Fig. 2). Table 2 shows that the lactating females exhibited all components of maternal aggression—attacks (or frontal attacks), lateral postures (or lateral attacks), bites and kicks—while the sensitized females mainly showed attacks towards intruders. The number of attacks, bites and kicks was considerably higher in lactating than in sensitized females, although most sensitized females attacked the intruder at least once eliciting a subordinate posture in the male (see Table 2). While all lactating (100%) and most sensitized females (87.5%) displayed aggression, none of the ovariectomized animals were aggressive. It is worth clarifying that few sensitized females showed high levels of aggression analogous to those exhibited by the lactating females (3/8 sensitized vs. 8/8 lactating; P < .05). High levels of aggression consisted of at least four attacks, low latencies to the first attack and other aggressive responses such as bites, lateral postures and boxings. Table 2 also shows that the latencies to the first aggressive response performed by the female and to the first subordinate posture exhibited by the male were significantly lower in lactating than in sensitized animals. None of the ovariectomized rats
A. Ferreira et al. / Physiology & Behavior 77 (2002) 197–204
201
Table 2 Number and latency to some aggressive responses during the 5-min test of ovariectomized, sensitized and lactating rats, as well as the latency of subordinated posture of the male intruders Attacks Ovariectomized housed in groups Ovariectomized isolated Sensitized Lactating H P
Lateral postures
Bites
Kicks
0.0 (0.0)a
0.0 (0.0)a
0.0 (0.0)a
0.0 (0.0)a
0.0 (0.0)a 2.0 (2.0)b 8.5 (2.5)c 27.9 < .001
0.0 (0.0)a 0.0 (0.0)a 2.0 (1.3)b 29.6 < .001
0.0 (0.0)a 0.0 (1.0)a 3.5 (1.6)b 25.3 < .001
0.0 (0.0)a 0.0 (0.0)a 4.5 (1.9)b 21.6 < .001
Aggression latency (s)
Subordinate latency (s)
Aggressive females
300.0 (0.0)a
300.0 (0.0)a
0/8a
300.0 (0.0)a 135.5 (134.1)b 40.5 (8.6)b 25.8 < .001
300.0 (0.0)a 135.0 (121.8)b 40.5 (8.6)b 24.1 < .001
0/8a 7/8b 8/8b (c2 = 14.7) < .01
The number of responses (median (IQR)) was compared by the Kruskal – Wallis analysis of variance followed by Mann – Whitney U test, different letters indicate statistical differences ( P < .01) among groups. The number of aggressive females was analyzed using c2 test for independent samples.
(isolated or housed in groups) displayed aggressive responses. 3.3. Freezing behavior A significant difference in the duration of the immobility response among the three groups was found (H = 11.3; P < .01). Lactating and sensitized females showed a statistically significant shorter freezing period than the ovariectomized control group (Fig. 3). Both lactating and sensitized rats retrieved the pups during the 5-min adaptation period, although the lactating group showed an increased number of retrievals when compared to that exhibited by the sensitized one (median (IQR): lactating, 8.0 (1.7); sensitized, 1.0 (2.0); U = 4; P < .05). No significant differences were found in the number of groomings, defecations and rearings among the three groups. The comparison between lactating animals housed in small cages with three foster pups changed daily, and dams placed in big cages together with six unchanged pups did not show significant differences. Thus, in the anxiety test, neither the percentage of time in the open arms (U = 10.0; P = ns) nor the percentage of entries into open arms (U = 13.5; P = ns) differed between the groups. Similarly,
Fig. 3. Duration of freezing reaction in response to a sudden auditory stimulus (median (IQR)) in ovariectomized (OVX, n = 8), sensitized, (SENS, n = 8) and lactating (LACT, n = 8) rats. Data were compared by the Kruskal – Wallis analysis of variance followed by Mann – Whitney U test. a P < .05 versus ovariectomized controls.
no differences were found in the aggression (sum of aggressive behaviors: U = 11.5; P = ns; male upright subordinate postures: U = 12.0; P = ns) and fear (duration of immobility: U = 17.5; P = ns) tests.
4. Discussion The main findings of the present study are: 1. Lactating females showed a decreased anxiety in the plus maze test in contrast to the levels exhibited both by the sensitized and ovariectomized rats. 2. Sensitized females displayed attacks towards intruders, although the aggression levels were significantly lower than those of lactating dams. 3. Both lactating and sensitized females showed lower levels of fear compared to those exhibited by nonmaternal ovariectomized rats. It has been previously demonstrated that lactating dams show anxiolysis in conflict paradigms [14], open-field [16] and plus maze [2,27] tests. In accordance with these observations, we found that the anxiety levels observed in the elevated plus maze test were low in lactating rats. In contrast, both the sensitized and the ovariectomized animals showed high levels of anxiety. The selective reduction in anxiety in lactating but not in sensitized rats suggest that maternal behavior per se is not the essential cue to produce anxiolysis in the elevated plus maze, but other factors, possibly of endocrine origin. Supporting this idea, it has been demonstrated that progesterone and its metabolites exert an anxiolytic effect in females in different animal models [3,4,11]. In future research, it would be important to include a lactating ovariectomized group in order to explore the possible role of this hormone in the anxiolysis observed in lactating rats. In some paradigms, such as the conflict and open-field tests, the presence of the litter induces or enhances anxiolysis in lactating rats [14]. In the present study, the presence of the litter in the closed arms did not induce anxiolysis in sensitized females. The total number of entries was highest in lactating animals followed by sensitized rats and then
202
A. Ferreira et al. / Physiology & Behavior 77 (2002) 197–204
ovariectomized controls. It has been proposed that this parameter reflect exploration associated with anxiolysis [40]. However, the fact that sensitized females showed an intermediate number of total arms entries compared to lactating and ovariectomized animals could be due to the fact that they retrieved the pups from one closed arm to the other, but less frequently than did the lactating rats. This behavior seems to reflect maternal responsiveness rather than exploration. Such observation is in accordance with previous experiments [47] that showed that lactating females enter the T-maze to retrieve pups more frequently than virgin sensitized rats. In general, it is considered that maternal aggression is triggered by changes in the endocrine pattern characteristic of late gestation, parturition and lactation [34]. Although being controversial, most studies show changes in ovarian hormones [29,31 –33], prolactin [50] and oxytocin [7,19,20,35] associated to the expression of maternal aggression. The role of these hormones on maternal aggression is further supported by the results showing that sensitized females displayed this behavior either after a prolonged treatment with steroids [29,30] or after endocrine changes such as pseudopregnancy produced by long periods of exposure to pups [9]. Present data reveal that without any gonadal hormonal influence, sensitized ovariectomized females attacked intruders. This result is in agreement with data from Mayer and Rosenblatt [34] who found aggression in sensitized Sprague –Dawley females untreated with steroid hormones. However, numerous reports maintain that nonpregnant females do not display maternal aggression after sensitization [10,18,20,26,33]. The reasons underlying the disparity between these results and present findings could rely, among others, on differences in the duration of the aggression test, the period of pup stimulation and the strain used. Although the sensitized females in our study displayed aggression without steroid hormonal treatment, we cannot completely rule out hormonal factors in the control of aggression, since the intensity and quality of the offensive behavior in sensitized females were much lower than those exhibited by lactating dams (present data; [34]). Thus, the maternal aggression observed in lactating animals included attacks, bites, lateral postures and kicks, while in sensitized females consisted of attacks, vocalizations and other warning behaviors were seen. From these results, it can be proposed that for the full expression of maternal aggression, steroid hormonal factors are required. Sensory stimuli from the pups are necessary for the initiation [24,45] and maintenance [9,13,30,46] of maternal aggression. One of the most important cues received by the mother for the display of this behavior is the litter ventral trunk stimulation [46], which can result in prolactin release [48]. Unfortunately, we did not distinguish between different nursing behaviors [44], and, therefore, we could not correlate the amount of ventral stimulation received by the sensitized females and the aggression levels displayed.
In the expression of maternal aggression, other factors may contribute, such as changes in the perception of olfactory cues from the pups and a contrast of these odors with those of the intruder [12,13,25,28]. The development of maternal behavior, as previously proposed [17], depends on the fear reduction to the natural aversion to pups’ odors. Pups stimulation, independent of endocrine factors, could induce changes in brain areas such as the ventro-lateral caudal periaqueductual gray (vlcPAG), which regulate some of the behavioral characteristics of the lactating rat. Thus, it has been shown that suckling and pup ventral stimulation affects the vlcPAG and its efferents. Furthermore, its lesioning increases maternal aggression and reduces anxiety in lactating rats [27]. Present results did not find aggression in the isolated ovariectomized control group, indicating that the aggression observed in sensitized rats cannot be explained on the basis of isolation but may be considered as specific maternal aggression. These results are in accordance with former reports [9]. Interestingly, our results showed that lactating females and sensitized rats without hormonal treatment exhibited a similar reduced freezing behavior compared to control ovariectomized animals. Previous results reported a reduced freezing in lactating and sensitized females treated with steroid hormones [20,21]. In our study, the sensitized ovariectomized females were not treated with steroid hormones, ruling out the role of these hormones in the reduced freezing. Previous studies showed that lesions to the ascending milk ejection pathway [20] or the olfactory bulb [12] failed to affect the short freezing reaction of mother rats, suggesting that neither suckling nor olfactory cues coming from the pups [12] are essential for fear reduction. The possible role of other factors is unknown, however, it may be suggested that fear reduction depends on the display of maternal behavior. It is also proposed that the reduced fear, as maternal behavior, is controlled by a multisensorial stimulation from the newborn pups [15]. The number of pups, the housing conditions and the daily manipulation of the animals do not seem to affect the behavior observed since no differences were found in anxiety, aggression and fear between the two groups of lactating females with different housing conditions. To conclude, this paper reports that maternal behavior induced without hormonal treatment was not sufficient to reduce anxiety in the plus maze test. Conversely, aggression and fear reduction parallels the expression of maternal behavior in the sensitized animals, although the aggression levels were not comparable in quality and intensity to those exhibited by lactating rats.
Acknowledgements Authors wish to thank Marcelo Ferna´ndez for animal care, Marı´a Jose´ Zuluaga for technical assistance and
A. Ferreira et al. / Physiology & Behavior 77 (2002) 197–204
Mariana Meerhoff for English checking. This paper was supported by CSIC, PEDECIBA (A.F., M.P. and N.U) and FNI (A.F).
[21] [22]
References [23] [1] Archer J. Test for emotionality in rats and mice: a review. Anim Behav 1973;21:205 – 35. [2] Bitran D, Hilvers RJ, Kellogg CK. Ovarian endocrine status modulates the anxiolytic potency of diazepam and the efficacy of g-aminobutyric acid – benzodiazepine receptor-mediated chloride ion transport. Behav Neurosci 1991;105:653 – 62. [3] Bitran D, Hilvers RJ, Kellogg CK. Anxiolytic effects of 3a-hydroxy5a[b]-pregnan-20-one: endogenous metabolites of progesterone that are active at the GABAA receptor. Brain Res 1991;561:157 – 61. [4] Bitran D, Purdy RH, Kellogg CK. Anxiolytic effect of progesterone is associated with increases in cortical allopregnanolone and GABAA receptor function. Pharmacol Biochem Behav 1993;45:423 – 8. [5] Bridges RS. A quantitative analysis of the roles of dosage, sequence and duration of estradiol and progesterone exposure in the regulation of maternal behavior in the rat. Endocrinology 1984;114:930 – 40. [6] Bridges RS, Zarrow MX, Gandelman R, Denenberg VH. Differences in maternal responsiveness between lactating and sensitized rats. Dev Psychobiol 1972;5:123 – 7. [7] Consiglio AR, Lucion AB. Lesions of hypothalamic paraventricular nucleus and maternal aggressive behavior in female rats. Physiol Behav 1996;59:591 – 6. [8] Cosnier J. Quelques problemes pose´s par le ‘‘comportement maternel provoque´’’ chez la ratte. C R Seances Soc Biol 1963;157:1611 – 3. [9] Erskine MS, Barfield RJ, Goldman BD. Intraspecific fighting during late pregnancy and lactation in rats and effects of litter removal. Behav Biol 1978;23:206 – 13. [10] Erskine MS, Barfield RJ, Goldman BD. Postpartum aggression in rats: II. Dependence on maternal sensitivity to young and effects of experience with pregnancy and parturition. J Comp Physiol Psychol 1980; 94:495 – 505. [11] Ferna´ndez-Guasti A, Picazo O. Changes in burying behavior during the estrous cycle: effect of estrogen and progesterone. Psychoneuroendocrinology 1992;17:681 – 9. [12] Ferreira A, Dahlo¨f LS, Hansen S. Olfactory mechanisms in the control of maternal aggression, appetite, and fearfulness: effects of lesions to olfactory receptors, mediodorsal thalamic nucleus, and insular prefrontal cortex. Behav Neurosci 1987;101:709 – 17. [13] Ferreira A, Hansen S. Sensory control of maternal aggression in Rattus norvegicus. J Comp Psychol 1986;100:173 – 7. [14] Ferreira A, Hansen S, Nielsen M, Archer T, Minor BC. Behavior of mother rats in conflict tests sensitive to anti-anxiety agents. Behav Neurosci 1989;103:193 – 201. [15] Fleming AS, Cheung U, Myhal N, Kessler Z. Effects of maternal hormones on ‘timidity’ and attraction to pup-related odors in females rats. Physiol Behav 1989;46:449 – 53. [16] Fleming AS, Luebke C. Timidity prevents the virgin female rat from being a good mother: emotionality differences between nulliparous and parturient females. Physiol Behav 1981;27:863 – 70. [17] Fleming AS, Rosenblatt JS. Maternal behavior in the virgin and lactating rat. J Comp Physiol Psychol 1974;86:957 – 72. [18] Gandelman R, Simon NG. Postpartum fighting in the rat: nipple development and the presence of young. Behav Neural Biol 1980;28: 350 – 60. [19] Giovenardi M, Padoin MJ, Cadore LP, Lucion AB. Hypothalamic paraventricular nucleus modulates maternal aggression in rats: effects of ibotenic acid lesion and oxitocin antisense. Physiol Behav 1998;63: 351 – 9. [20] Hansen S, Ferreira A. Food intake, aggression and fear behavior in
[24]
[25]
[26] [27]
[28] [29]
[30]
[31]
[32]
[33] [34]
[35]
[36] [37]
[38]
[39]
[40]
[41] [42] [43]
203
mother rats: control by neural systems concerned with milk ejection and maternal behavior. Behav Neurosci 1986;100:67 – 70. Ha˚rd E, Hansen S. Reduced fearfulness in the lactating rat. Physiol Behav 1985;35:641 – 3. Jakubowski M, Terkel J. Incidence of pup killing and parental behavior in virgin female and male rats (Rattus norvergicus): differences between Wistar and Sprague – Dawley stocks. J Comp Psychol 1985;99: 93 – 7. Kellog CK, Barrett KA. Reduced progesterone metabolites are not critical for plus-maze test performance of lactating females rats. Pharmacol Biochem Behav 1999;63:441 – 8. Kolunie JM, Stern JM. Maternal aggression: disruption by perioral anesthesia in lactating Long – Evans rats (Rattus norvegicus). J Comp Psychol 1990;104:352 – 60. Kolunie JM, Stern JM. Maternal aggression in rats: effects of olfactory bulbectomy, ZnSo4-induced anosmia, and vomeronasal organ removal. Horm Behav 1995;29:418 – 92. Krehbiel DA, LeRoy LM. The quality of hormonally stimulated maternal behavior in ovariectomized rats. Horm Behav 1979;12:243 – 52. Lonstein JS, Simmons DA, Stern JM. Functions of the caudal periaqueductal gray in lactating rats: kyphosis, lordosis, maternal aggression, and fearfulness. Behav Neurosci 1998;112:1502 – 18. Lynds PG. Olfactory control of aggression in lactating female house mice. Physiol Behav 1976;17:157 – 9. Mayer AD, Ahdieh HB, Rosenblatt JS. Effects of prolonged estrogen – progesterone treatment and hypophysectomy on the stimulation of short-latency maternal behavior and aggression in female rats. Horm Behav 1990;24:152 – 73. Mayer AD, Monroy MA, Rosenblatt JS. Prolonged estrogen – progesterone treatment of nonpregnant ovariectomized rats: factors stimulating home-cage and maternal aggression and short-latency maternal behavior. Horm Behav 1990;24:3342 – 64. Mayer AD, Reisbick S, Siegel HI, Rosenblatt JS. Maternal aggression in rats: changes over pregnancy and lactation in a Sprague – Dawley strain. Aggressive Behav 1987;13:29 – 43. Mayer AD, Rosenblatt JS. Prepartum changes in maternal responsiveness and nest defense in Rattus novegicus. J Comp Psychol 1984;98: 177 – 88. Mayer AD, Rosenblatt JS. Hormonal factors influence the onset of maternal aggression in laboratory rats. Horm Behav 1987;21:253 – 67. Mayer AD, Rosenblatt JS. Persistent effects on maternal aggression of pregnancy but not of estrogen/progesterone treatment of non pregnant ovariectomized rats revealed when initiation of maternal behavior is delayed. Horm Behav 1993;27:1132 – 55. Olazabal DE, Ferreira A. Maternal behavior in rats with kainic acidinduced lesions of the hypothalamic paraventricular nucleus. Physiol Behav 1997;61:779 – 84. Ostermeyer MC. Maternal aggression. In: Elwood RW, editor. Parental behaviour of rodents. New York: Wiley, 1983. p. 151 – 79. Pellow S, Chopin P, File SE, Briley M. Validation of open: closed arm entries in an elevated plus-maze as a measure of anxiety in the rat. J Neurosci Methods 1985;14:149 – 67. Picazo O, Fe´rnandez-Guasti A. Antianxiety effects of progesterone and some of its reduced metabolites: an evaluation using the burying behavior test. Brain Res 1995;680:135 – 41. Quadagno DM, Debold JF, Gorzalka BB, Whalen RE. Maternal behavior in the rat: aspects of concaveation and neonatal androgen treatment. Physiol Behav 1974;12:1071 – 4. Rodgers RJ, Cole JC. The elevated plus-maze: pharmacology, methodology and ethology. In: Cooper SJ, Hendrie CA, editors. Ethology and psychopharmacology. Chichester: Wiley, 1994. p. 9 – 44. Rosenblatt JS. Nonhormonal basis of maternal behavior in the rat. Science 1967;156:1512 – 4. Rosenblatt JS. Prepartum and postpartum regulation of maternal behavior in the rat. Ciba Found Symp (new series) 1975;33:17 – 32. Siegel S. Nonparametric statistics for the behavioral sciences. New York: McGraw-Hill, 1956.
204
A. Ferreira et al. / Physiology & Behavior 77 (2002) 197–204
[44] Stern J, Johnson SK. Ventral somatosensory determinants of nursing behavior in Norway rats: I. Effects of variations in the quality and quantity of pups stimuli. Physiol Behav 1990;47:993 – 1011. [45] Stern JM, Kolunie JM. Trigeminal lesions and maternal behavior in Norway rats: I. Effects of cutaneous rostral snout denervation on maintenance of nurturance and maternal aggression. Behav Neurosci 1991;105:984 – 97. [46] Stern JM, Kolunie JM. Maternal aggression of rats is impaired by cutaneous anesthesia of ventral trunk, but not by nipple removal. Physiol Behav 1993;54:861 – 8. [47] Stern JM, Mackinnon DA. Postpartum, hormonal, and nonhormonal induction of maternal behavior in rats: effects on T-maze retrieval of pups. Horm Behav 1970;7:305 – 16. [48] Stern J, Siegel S. Prolactin release in lactating, primiparous and multiparous thelectomized and maternal virgin rats exposed to pups stimulis. Biol Reprod 1978;19:177 – 82.
[49] Svare BB. Maternal aggression in mammals. In: Gubernick DJ, Klopfer PH, editors. Parental care in mammals. New York: Plenum, 1981. p. 179 – 210. [50] Svare BB, Mann MA. Hormonal influence on maternal aggression. In: Svare BB, editor. Hormones and aggressive behavior. New York: Plenum, 1983. p. 91 – 104. [51] Terkel J, Rosenblatt JS. Aspects of nonhormonal maternal behavior in the rat. Horm Behav 1971;2:161 – 71. [52] Valzelli L. The ‘‘isolation syndrome’’ in mice. Psychopharmacology 1973;31:305 – 20. [53] Wongwidecha N, Mardsen CA. Social isolation increases aggressive behavior and alters the effect of diazepam in the rat social interaction test. Behav Brain Res 1996;75:27 – 32.
Role of maternal behavior on aggression, fear and anxiety A. Ferreiraa, M. Pereiraa, D. Agratia, N. Uriartea, A. Ferna´ndez-Guastib,* a
b
Facultad de Ciencias, Universidad de la Repu´blica, Montevideo, Uruguay Departamento de Farmacobiologı´a, CINVESTAV/IPN, Calzada De los Tenorios 235, Col. Granjas Coapa, Me´xico 14330 D.F., Mexico Received 14 February 2001; received in revised form 10 April 2002; accepted 28 May 2002
Abstract Concomitant to the expression of maternal behavior, the lactating female develops anxiolysis in the elevated plus maze test, aggression towards intruders and reduced fear in response to a sudden auditory stimulus. This study aims to determine if these behavioral changes are associated with maternal behavior independently of the endocrine status that characterizes gestation, parturition and lactation. To assess this purpose, the behavior of lactating females was compared to that exhibited by maternal and nonmaternal ovariectomized rats untreated with steroid hormones. In contrast with lactating dams, sensitized animals (rats that displayed maternal behavior after a continuous contact with young pups) did not display reduced anxiety in the plus maze test. However, the sensitized females showed behaviors characteristic of lactating rats, such as some components of maternal aggression and reduced fear, though much less intensely than dams. These results suggest that aggression and reduced fear, but not anxiolysis, partially depend on the development of maternal behavior. D 2002 Elsevier Science Inc. All rights reserved. Keywords: Maternal aggression; Fear; Anxiety; Sensitization; Maternal behavior
1. Introduction The behavioral repertoire of some mammals during motherhood differs from that exhibited by females in other periods of their reproductive cycle. Concomitant with the display of maternal behavior, lactating rats show lower levels of anxiety in conflict tests [14], open-field paradigms [16] and plus maze test [2,23,27], exhibit less fear towards a sudden auditory stimulus [21] and develop aggression when confronted with intruders [9,36,49]. This behavioral pattern relies, at least partly, on the hormonal changes that characterize late gestation, parturition and lactation [2,15,20]. As maternal care parallels the course of these behaviors, it might be suggested that maternal behavior is a condition needed for the expression of anxiolysis, aggression and fear reduction. Most components of maternal behavior can be induced in virgin or even ovariectomized animals by a continuous exposure to foster pups. This process, called ‘‘sensitization’’ [8,17,42], is shortened by steroid treatment simulating the endocrine profile of late gestation and parturition [5,29]. Most studies show that sensitized females treated with sex * Corresponding author. Tel.: +52-54-83-28-56; fax: +52-54-83-28-63. E-mail address: [email protected] (A. Ferna´ndez-Guasti).
steroids, develop aggression [29,30] and show less fear [20], behavioral features characteristic of lactating animals. These results suggest that both endocrine factors and maternal care contribute to the expression of aggression and fear reduction. The purpose of the present study was to analyze whether the reduced anxiety and fear and the increased aggression observed in lactating rats are related to the expression of maternal care. The sensitization of ovariectomized females—without hormonal treatment—is a useful model to examine the role of maternal behavior on the reduced anxiety and fear and on the expression of maternal aggression, independently of the endocrine changes that characterize gestation, parturition and lactation. With this aim, anxiety (measured as the time spent and number of entries into the open arms of an elevated plus maze) [37], maternal aggression [36] and fear (measured as the duration of immobility in response to a sudden auditory stimulus) [1] were tested in ovariectomized, sensitized and lactating rats. Sensitized females were ovariectomized to eliminate the possibility that ovarian steroid fluctuations, produced after prolonged contact with pups [10], affect the behaviors here recorded. This is particularly important, given the influence of progesterone and its metabolites on anxiety-related behaviors [2,3,38].
0031-9384/02/$ – see front matter D 2002 Elsevier Science Inc. All rights reserved. PII: S 0 0 3 1 - 9 3 8 4 ( 0 2 ) 0 0 8 4 5 - 4
198
A. Ferreira et al. / Physiology & Behavior 77 (2002) 197–204
As the presence of the litter is relevant for the reduced freezing and can induce or potentiate anxiolysis in the openfield and conflict tests [14,21], it is hypothesized that the litter presence may induce anxiolysis in the sensitized females. On these bases, the tests were performed in the attendance of the litter. The site where the pups should be located in the plus maze test remained as a puzzling question. In the present study, the pups were always located in the closed arms, since it was observed that when the newborn were placed in the open arms, the lactating females spend long intervals in these arms, without necessarily revealing a decreased anxiety but rather maternal motivation (unpublished results). Prolonged isolation increases aggression in males [52,53]. Although the groups of lactating and sensitized rats were not completely isolated, but obviously housed with pups, a putative behavioral change in aggression might be caused by the fact that maternal females were individually housed while the ovariectomized animals inhabited in groups. On these bases, the behavior of ovariectomized females housed in groups was compared to that of ovariectomized animals housed individually for a similar period to that required for the sensitization procedure. Finally, to control for putative differences between lactating and sensitized females in the display of aggression, fear and anxiety due to various factors involved in the development of sensitization, an experiment was performed considering the number and pups manipulation as well as the cage measurements.
2. Materials and methods 2.1. Animals Three-month-old nulliparous Wistar females, males (weighing 250 –350 g) and 2- to 10-day-old pups were used in this study. All animals were housed in a room under inverted 12-h light–dark cycle (lights off at 15:00 h) with ad libitum access to water and rat chow throughout the experiments. Females were divided into three main groups: ovariectomized (n = 8), sensitized (n = 8) and lactating (n = 8). Ovariectomy was performed through a ventral incision under 40 mg/kg ip of pentobarbital anesthesia. An extra group of donor lactating females with their pups were maintained to provide a supply of foster pups for the sensitization procedure and for the elevated plus maze, aggression and freezing tests. In general, animals were housed as follows: Nonsensitized ovariectomized females were kept in groups of four in a cage measuring 45 28 26 cm. Sensitized ovariectomized females were individually placed in small cages (27 19 11 cm) to reduce the sensitization latency [39,51]. Lactating females were maintained with their pups (adjusted to six on the delivery day) in cages measuring 45 28 26 cm. For controlling the role of isolation on
aggression, a group of ovariectomized rats housed individually in small cages (27 19 11 cm) for 17 days was also included (n = 8). Additionally, to assess the role of cage measurements, number of pups and manipulation, a group of dams housed in small cages together with three foster pups daily changed (treated in the same way as the sensitized females) (n = 8) were compared to a group of lactating females placed in big cages together with six pups (n = 8). 2.2. Induction of maternal behavior (sensitization procedure) Three days after ovariectomy, the females were housed in individual cages. Three freshly fed pups (2 –10 days of age) were scattered in the corner of the home cage opposite to the female’ sleeping area. They were removed daily 1 h after the onset of darkness and replaced 5 min later with three new pups. Immediately after placing the pups, maternal behavior was tested [17,41]. The number of the following behaviors was recorded during a 5-min test: retrieving (the female picks up a pup and carries it to another location of the cage); crouching posture (the mother hovers over the pups remaining immobile except for occasional postural adjustment; every 10 s, a new score is recorded if the female remains in this position); licking (the female licks the pups) and nest building (the female pushes or picks up nest material in her mouth). Based on previous studies [6,17,34], retrieval of the three pups within the test period (5 min) during two consecutive days was used as a marker of maternal care. Subjects were assigned a sensitization latency corresponding to the second day of the two consecutive retrievals. Those animals that cannibalized pups and those that did not display maternal behavior after 26 days of pup exposure were excluded from the experiment. Because only 25% of the Wistar rats became maternal in a period of 10 –15 days [22], we selected a long period for the sensitization procedure. 2.3. Experimental procedure Each animal was tested in the experimental paradigms in the following sequence: Day 1, elevated plus maze test; Day 2, aggression test; and Day 3, freezing test. The lactating rats started the testing sequence on the 7th day postpartum, the sensitized group 1 day after being considered maternal and the ovariectomized group 15 days after surgery. The extra control group of ovariectomized females was tested for aggression after an isolation period of 17 days. In all the tests, the behavior of a single rat was recorded at a time by an observer placed 1 m from the testing model under red dim light. 2.4. Behavioral tests 2.4.1. Elevated plus maze test This test was carried out 2 h after the onset of darkness (at 17:00 h). This model consists of an elevated (40 cm
A. Ferreira et al. / Physiology & Behavior 77 (2002) 197–204
above the floor) plus-shaped maze with two opposite enclosed and two open arms, measuring 50 cm long 10 cm wide, placed in a red dimly lit room. Each animal was placed in the center of the maze facing a closed arm. Three foster pups were placed at the end of each closed arm. An entry into an arm was defined as the animal placing all four paws into it. The cumulative time spent in the open arms and the number of open- and closed-arm entries were recorded over a 5-min session. Data were expressed as the percentage of open arms entries (100 open/total), the percentage of time spent in the open arms (100 open/ total), the number of entries to the closed arms and the total number of entries (open + enclosed arms). The number of retrievals, groomings, defecations and rearings was also recorded. 2.4.2. Aggressive behavior The aggression test towards a male (300 – 350 g) was carried out in the female’s home cage 2 h after the onset of darkness (at 17:00 h). Aggression was assessed using a previously described procedure [9,10]. A male rat was placed in the female’s cage, and the frequency of the following behaviors recorded during a 5-min test: attack (or ‘‘frontal attack’’: female lunges toward the male, sometimes followed by rolling and scuffing for a brief period); bite (female mouth in contact with the intruder body and the male vocalizes); lateral posture (or ‘‘lateral attack’’: female approaches male side way presenting the lateral part of her body to him); and kick (with rear leg). The number of times the intruder male adopted an upright subordinate posture (sitting on haunches, curved back, forepaws at the female’s face) during the test was also recorded. The latencies to the first aggressive response displayed by the female (attack, bite or lateral posture) and to the first subordinate posture by the male were also recorded. Nonresponder animals were assigned a latency of 300 s. The stimulus males were never used more than once in the tests. Females were considered aggressive during an intruder test if they attacked more than once, or attacked once eliciting a submissive posture [29]. 2.4.3. Freezing behavior This test was carried out 2 h after the onset of darkness (at 17:00 h). This paradigm includes a circular Plexiglas arena (39-cm diameter), whose floor is covered by a filter paper divided by two lines to form four 90 sectors. The floor cover was changed after testing each rat. The circular arena is enclosed in a soundproof chamber 70 41 40 cm, which is illuminated by a 15-W white light bulb suspended together with a door bell (95 dB) on the ceiling of the test chamber. Each rat was placed together with three foster pups and its behavior recorded through a window on the front wall. The animal was allowed a 5-min adaptation period in the novel observation cage. During this time, the number of retrievals, groomings, rearings and defecations was recorded. After this period, the doorbell was activated for 6 s. During the sound, the rat typically rushed around in the
199
cage and eventually immobilized either concurrently with or slightly before the end of the signal. In both cases, the freezing duration comprises the time (in seconds) of immobility shown by the rat from the cessation of the signal to the first distinct body movement excluding eye blinks and respiratory or vibrissae movements [21]. The local ethical committee approved all the experimental procedures performed in this study. 2.5. Statistical analysis Most data are presented as median (IQR) (interquartile ranges) [43]. As most variables did not follow a normal distribution (Kolmogorov –Smirnov test) nor were variances homogeneous (Cochran C test), data were analyzed with nonparametric tests. The number of responses among the groups in the tests of anxiety, aggression and fear was analyzed by the Kruskal – Wallis one-way analysis of variance followed by Mann –Whitney U test. The nominal scale data were analyzed using Fisher exact probability test and c2 test for independent samples. The latencies for sensitized animals to show maternal behavior are expressed as means ± S.E.
3. Results Eight of nine animals (89%) submitted to the sensitization procedure showed pups’ retrieval, licking, nursing behavior and nest building. The latencies (mean days ± S.E.) required for the sensitized females to exhibit the following behaviors were: pups’ retrieval (after two consecutive days), 17.0 ± 2.0; nursing behavior, 6.8 ± 2.2; nest building, 10.5 ± 2.5; and licking, 2.3 ± 1.5. 3.1. Elevated plus maze The lactating females spent a higher percentage of time in the open arms (Fig. 1, Panel A, H = 19.6; P < .01) and showed higher percentage of entries into open arms (Panel B, H = 11.7; P < .01) than the sensitized and the ovariectomized rats. Lactating, as compared to both sensitized and control females, exhibited a statistically significant increase in the total number of entries and in the entries to the closed arms. Sensitized animals also showed an increase in these parameters when compared to ovariectomized rats (see Table 1). Both lactating and sensitized rats retrieved the pups, either in the same closed arm or from one closed arm to the other, but never took them into the open arms. No significant differences were found in the number of animals that showed retrieval (lactating: 8/8 vs. sensitized: 5/8; P = ns); however, the number of retrievals was significantly higher in lactating compared to sensitized females (median (IQR): lactating group, 5.0 (1.1); sensitized group, 1.0 (0.7); U = 2.5, P < .005). No differences were found in the number
200
A. Ferreira et al. / Physiology & Behavior 77 (2002) 197–204
Fig. 1. Percentage of time spent in the open arms (Panel A) and percentage of open arm entries (Panel B) in the plus maze test of ovariectomized, n = 8 (OVX), sensitized, n = 8 (SENS) and lactating, n = 8 (LACT) rats. Data are expressed as median (IQR) and were compared by the Kruskal – Wallis analysis of variance followed by Mann – Whitney U test. aP < .05 versus ovariectomized controls and sensitized.
of groomings, defecations and rearings among the three groups (data not shown). 3.2. Aggressive behavior Fig. 2A shows that the median (IQR) of the sum of aggressive behaviors (attacks, bites and lateral postures) was significantly different among the groups (H = 29.1; P < .001). Thus, lactating females showed more aggressive responses than the sensitized and the nonmaternal ovariectomized rats. The sensitized animals displayed more aggression than the ovariectomized control animals. Fig. 2B shows that male upright subordinate postures consequently follows the aggression levels of the females (H = 29.3; P < .001). In
Table 1 Number of entries into the closed-arm and total entries in the plus maze test of ovariectomized, sensitized and lactating rats
Ovariectomized Sensitized Lactating H P
Closed-arm entries
Total entries
1.0 (0.5)a 5.5 (0.75)b 11.0 (1.3)c 21.5 < .001
1.0 (0.5)a 6.0 (1.4)b 13.0 (2.0)c 21.5 < .001
Data are expressed as median (IQR) and were compared by the Kruskal – Wallis analysis of variance followed by Mann – Whitney U test. Different letters indicate statistical differences ( P < .01) among groups.
Fig. 2. (A) Number of aggressive responses (sum of attacks, lateral postures and bites) displayed by females towards male intruders. (B) Number of submission postures adopted by the male intruders. OVX: ovariectomized housed in groups (n = 8); OVX isolated: ovariectomized isolated during 17 days (n = 8); SENS: sensitized (n = 8); and LACT: lactating females (n = 8). Data are expressed as median (IQR). OVX, OVX isolated, SENS and LAC were compared by the Kruskal – Wallis analysis of variance followed by Mann – Whitney U test. aP < .01 versus ovariectomized controls; bP < .01 versus lactating rats.
contrast, none of the ovariectomized nonmaternal rats (housed in groups or kept isolated for 17 days) displayed aggressive responses during the test (see Fig. 2). Table 2 shows that the lactating females exhibited all components of maternal aggression—attacks (or frontal attacks), lateral postures (or lateral attacks), bites and kicks—while the sensitized females mainly showed attacks towards intruders. The number of attacks, bites and kicks was considerably higher in lactating than in sensitized females, although most sensitized females attacked the intruder at least once eliciting a subordinate posture in the male (see Table 2). While all lactating (100%) and most sensitized females (87.5%) displayed aggression, none of the ovariectomized animals were aggressive. It is worth clarifying that few sensitized females showed high levels of aggression analogous to those exhibited by the lactating females (3/8 sensitized vs. 8/8 lactating; P < .05). High levels of aggression consisted of at least four attacks, low latencies to the first attack and other aggressive responses such as bites, lateral postures and boxings. Table 2 also shows that the latencies to the first aggressive response performed by the female and to the first subordinate posture exhibited by the male were significantly lower in lactating than in sensitized animals. None of the ovariectomized rats
A. Ferreira et al. / Physiology & Behavior 77 (2002) 197–204
201
Table 2 Number and latency to some aggressive responses during the 5-min test of ovariectomized, sensitized and lactating rats, as well as the latency of subordinated posture of the male intruders Attacks Ovariectomized housed in groups Ovariectomized isolated Sensitized Lactating H P
Lateral postures
Bites
Kicks
0.0 (0.0)a
0.0 (0.0)a
0.0 (0.0)a
0.0 (0.0)a
0.0 (0.0)a 2.0 (2.0)b 8.5 (2.5)c 27.9 < .001
0.0 (0.0)a 0.0 (0.0)a 2.0 (1.3)b 29.6 < .001
0.0 (0.0)a 0.0 (1.0)a 3.5 (1.6)b 25.3 < .001
0.0 (0.0)a 0.0 (0.0)a 4.5 (1.9)b 21.6 < .001
Aggression latency (s)
Subordinate latency (s)
Aggressive females
300.0 (0.0)a
300.0 (0.0)a
0/8a
300.0 (0.0)a 135.5 (134.1)b 40.5 (8.6)b 25.8 < .001
300.0 (0.0)a 135.0 (121.8)b 40.5 (8.6)b 24.1 < .001
0/8a 7/8b 8/8b (c2 = 14.7) < .01
The number of responses (median (IQR)) was compared by the Kruskal – Wallis analysis of variance followed by Mann – Whitney U test, different letters indicate statistical differences ( P < .01) among groups. The number of aggressive females was analyzed using c2 test for independent samples.
(isolated or housed in groups) displayed aggressive responses. 3.3. Freezing behavior A significant difference in the duration of the immobility response among the three groups was found (H = 11.3; P < .01). Lactating and sensitized females showed a statistically significant shorter freezing period than the ovariectomized control group (Fig. 3). Both lactating and sensitized rats retrieved the pups during the 5-min adaptation period, although the lactating group showed an increased number of retrievals when compared to that exhibited by the sensitized one (median (IQR): lactating, 8.0 (1.7); sensitized, 1.0 (2.0); U = 4; P < .05). No significant differences were found in the number of groomings, defecations and rearings among the three groups. The comparison between lactating animals housed in small cages with three foster pups changed daily, and dams placed in big cages together with six unchanged pups did not show significant differences. Thus, in the anxiety test, neither the percentage of time in the open arms (U = 10.0; P = ns) nor the percentage of entries into open arms (U = 13.5; P = ns) differed between the groups. Similarly,
Fig. 3. Duration of freezing reaction in response to a sudden auditory stimulus (median (IQR)) in ovariectomized (OVX, n = 8), sensitized, (SENS, n = 8) and lactating (LACT, n = 8) rats. Data were compared by the Kruskal – Wallis analysis of variance followed by Mann – Whitney U test. a P < .05 versus ovariectomized controls.
no differences were found in the aggression (sum of aggressive behaviors: U = 11.5; P = ns; male upright subordinate postures: U = 12.0; P = ns) and fear (duration of immobility: U = 17.5; P = ns) tests.
4. Discussion The main findings of the present study are: 1. Lactating females showed a decreased anxiety in the plus maze test in contrast to the levels exhibited both by the sensitized and ovariectomized rats. 2. Sensitized females displayed attacks towards intruders, although the aggression levels were significantly lower than those of lactating dams. 3. Both lactating and sensitized females showed lower levels of fear compared to those exhibited by nonmaternal ovariectomized rats. It has been previously demonstrated that lactating dams show anxiolysis in conflict paradigms [14], open-field [16] and plus maze [2,27] tests. In accordance with these observations, we found that the anxiety levels observed in the elevated plus maze test were low in lactating rats. In contrast, both the sensitized and the ovariectomized animals showed high levels of anxiety. The selective reduction in anxiety in lactating but not in sensitized rats suggest that maternal behavior per se is not the essential cue to produce anxiolysis in the elevated plus maze, but other factors, possibly of endocrine origin. Supporting this idea, it has been demonstrated that progesterone and its metabolites exert an anxiolytic effect in females in different animal models [3,4,11]. In future research, it would be important to include a lactating ovariectomized group in order to explore the possible role of this hormone in the anxiolysis observed in lactating rats. In some paradigms, such as the conflict and open-field tests, the presence of the litter induces or enhances anxiolysis in lactating rats [14]. In the present study, the presence of the litter in the closed arms did not induce anxiolysis in sensitized females. The total number of entries was highest in lactating animals followed by sensitized rats and then
202
A. Ferreira et al. / Physiology & Behavior 77 (2002) 197–204
ovariectomized controls. It has been proposed that this parameter reflect exploration associated with anxiolysis [40]. However, the fact that sensitized females showed an intermediate number of total arms entries compared to lactating and ovariectomized animals could be due to the fact that they retrieved the pups from one closed arm to the other, but less frequently than did the lactating rats. This behavior seems to reflect maternal responsiveness rather than exploration. Such observation is in accordance with previous experiments [47] that showed that lactating females enter the T-maze to retrieve pups more frequently than virgin sensitized rats. In general, it is considered that maternal aggression is triggered by changes in the endocrine pattern characteristic of late gestation, parturition and lactation [34]. Although being controversial, most studies show changes in ovarian hormones [29,31 –33], prolactin [50] and oxytocin [7,19,20,35] associated to the expression of maternal aggression. The role of these hormones on maternal aggression is further supported by the results showing that sensitized females displayed this behavior either after a prolonged treatment with steroids [29,30] or after endocrine changes such as pseudopregnancy produced by long periods of exposure to pups [9]. Present data reveal that without any gonadal hormonal influence, sensitized ovariectomized females attacked intruders. This result is in agreement with data from Mayer and Rosenblatt [34] who found aggression in sensitized Sprague –Dawley females untreated with steroid hormones. However, numerous reports maintain that nonpregnant females do not display maternal aggression after sensitization [10,18,20,26,33]. The reasons underlying the disparity between these results and present findings could rely, among others, on differences in the duration of the aggression test, the period of pup stimulation and the strain used. Although the sensitized females in our study displayed aggression without steroid hormonal treatment, we cannot completely rule out hormonal factors in the control of aggression, since the intensity and quality of the offensive behavior in sensitized females were much lower than those exhibited by lactating dams (present data; [34]). Thus, the maternal aggression observed in lactating animals included attacks, bites, lateral postures and kicks, while in sensitized females consisted of attacks, vocalizations and other warning behaviors were seen. From these results, it can be proposed that for the full expression of maternal aggression, steroid hormonal factors are required. Sensory stimuli from the pups are necessary for the initiation [24,45] and maintenance [9,13,30,46] of maternal aggression. One of the most important cues received by the mother for the display of this behavior is the litter ventral trunk stimulation [46], which can result in prolactin release [48]. Unfortunately, we did not distinguish between different nursing behaviors [44], and, therefore, we could not correlate the amount of ventral stimulation received by the sensitized females and the aggression levels displayed.
In the expression of maternal aggression, other factors may contribute, such as changes in the perception of olfactory cues from the pups and a contrast of these odors with those of the intruder [12,13,25,28]. The development of maternal behavior, as previously proposed [17], depends on the fear reduction to the natural aversion to pups’ odors. Pups stimulation, independent of endocrine factors, could induce changes in brain areas such as the ventro-lateral caudal periaqueductual gray (vlcPAG), which regulate some of the behavioral characteristics of the lactating rat. Thus, it has been shown that suckling and pup ventral stimulation affects the vlcPAG and its efferents. Furthermore, its lesioning increases maternal aggression and reduces anxiety in lactating rats [27]. Present results did not find aggression in the isolated ovariectomized control group, indicating that the aggression observed in sensitized rats cannot be explained on the basis of isolation but may be considered as specific maternal aggression. These results are in accordance with former reports [9]. Interestingly, our results showed that lactating females and sensitized rats without hormonal treatment exhibited a similar reduced freezing behavior compared to control ovariectomized animals. Previous results reported a reduced freezing in lactating and sensitized females treated with steroid hormones [20,21]. In our study, the sensitized ovariectomized females were not treated with steroid hormones, ruling out the role of these hormones in the reduced freezing. Previous studies showed that lesions to the ascending milk ejection pathway [20] or the olfactory bulb [12] failed to affect the short freezing reaction of mother rats, suggesting that neither suckling nor olfactory cues coming from the pups [12] are essential for fear reduction. The possible role of other factors is unknown, however, it may be suggested that fear reduction depends on the display of maternal behavior. It is also proposed that the reduced fear, as maternal behavior, is controlled by a multisensorial stimulation from the newborn pups [15]. The number of pups, the housing conditions and the daily manipulation of the animals do not seem to affect the behavior observed since no differences were found in anxiety, aggression and fear between the two groups of lactating females with different housing conditions. To conclude, this paper reports that maternal behavior induced without hormonal treatment was not sufficient to reduce anxiety in the plus maze test. Conversely, aggression and fear reduction parallels the expression of maternal behavior in the sensitized animals, although the aggression levels were not comparable in quality and intensity to those exhibited by lactating rats.
Acknowledgements Authors wish to thank Marcelo Ferna´ndez for animal care, Marı´a Jose´ Zuluaga for technical assistance and
A. Ferreira et al. / Physiology & Behavior 77 (2002) 197–204
Mariana Meerhoff for English checking. This paper was supported by CSIC, PEDECIBA (A.F., M.P. and N.U) and FNI (A.F).
[21] [22]
References [23] [1] Archer J. Test for emotionality in rats and mice: a review. Anim Behav 1973;21:205 – 35. [2] Bitran D, Hilvers RJ, Kellogg CK. Ovarian endocrine status modulates the anxiolytic potency of diazepam and the efficacy of g-aminobutyric acid – benzodiazepine receptor-mediated chloride ion transport. Behav Neurosci 1991;105:653 – 62. [3] Bitran D, Hilvers RJ, Kellogg CK. Anxiolytic effects of 3a-hydroxy5a[b]-pregnan-20-one: endogenous metabolites of progesterone that are active at the GABAA receptor. Brain Res 1991;561:157 – 61. [4] Bitran D, Purdy RH, Kellogg CK. Anxiolytic effect of progesterone is associated with increases in cortical allopregnanolone and GABAA receptor function. Pharmacol Biochem Behav 1993;45:423 – 8. [5] Bridges RS. A quantitative analysis of the roles of dosage, sequence and duration of estradiol and progesterone exposure in the regulation of maternal behavior in the rat. Endocrinology 1984;114:930 – 40. [6] Bridges RS, Zarrow MX, Gandelman R, Denenberg VH. Differences in maternal responsiveness between lactating and sensitized rats. Dev Psychobiol 1972;5:123 – 7. [7] Consiglio AR, Lucion AB. Lesions of hypothalamic paraventricular nucleus and maternal aggressive behavior in female rats. Physiol Behav 1996;59:591 – 6. [8] Cosnier J. Quelques problemes pose´s par le ‘‘comportement maternel provoque´’’ chez la ratte. C R Seances Soc Biol 1963;157:1611 – 3. [9] Erskine MS, Barfield RJ, Goldman BD. Intraspecific fighting during late pregnancy and lactation in rats and effects of litter removal. Behav Biol 1978;23:206 – 13. [10] Erskine MS, Barfield RJ, Goldman BD. Postpartum aggression in rats: II. Dependence on maternal sensitivity to young and effects of experience with pregnancy and parturition. J Comp Physiol Psychol 1980; 94:495 – 505. [11] Ferna´ndez-Guasti A, Picazo O. Changes in burying behavior during the estrous cycle: effect of estrogen and progesterone. Psychoneuroendocrinology 1992;17:681 – 9. [12] Ferreira A, Dahlo¨f LS, Hansen S. Olfactory mechanisms in the control of maternal aggression, appetite, and fearfulness: effects of lesions to olfactory receptors, mediodorsal thalamic nucleus, and insular prefrontal cortex. Behav Neurosci 1987;101:709 – 17. [13] Ferreira A, Hansen S. Sensory control of maternal aggression in Rattus norvegicus. J Comp Psychol 1986;100:173 – 7. [14] Ferreira A, Hansen S, Nielsen M, Archer T, Minor BC. Behavior of mother rats in conflict tests sensitive to anti-anxiety agents. Behav Neurosci 1989;103:193 – 201. [15] Fleming AS, Cheung U, Myhal N, Kessler Z. Effects of maternal hormones on ‘timidity’ and attraction to pup-related odors in females rats. Physiol Behav 1989;46:449 – 53. [16] Fleming AS, Luebke C. Timidity prevents the virgin female rat from being a good mother: emotionality differences between nulliparous and parturient females. Physiol Behav 1981;27:863 – 70. [17] Fleming AS, Rosenblatt JS. Maternal behavior in the virgin and lactating rat. J Comp Physiol Psychol 1974;86:957 – 72. [18] Gandelman R, Simon NG. Postpartum fighting in the rat: nipple development and the presence of young. Behav Neural Biol 1980;28: 350 – 60. [19] Giovenardi M, Padoin MJ, Cadore LP, Lucion AB. Hypothalamic paraventricular nucleus modulates maternal aggression in rats: effects of ibotenic acid lesion and oxitocin antisense. Physiol Behav 1998;63: 351 – 9. [20] Hansen S, Ferreira A. Food intake, aggression and fear behavior in
[24]
[25]
[26] [27]
[28] [29]
[30]
[31]
[32]
[33] [34]
[35]
[36] [37]
[38]
[39]
[40]
[41] [42] [43]
203
mother rats: control by neural systems concerned with milk ejection and maternal behavior. Behav Neurosci 1986;100:67 – 70. Ha˚rd E, Hansen S. Reduced fearfulness in the lactating rat. Physiol Behav 1985;35:641 – 3. Jakubowski M, Terkel J. Incidence of pup killing and parental behavior in virgin female and male rats (Rattus norvergicus): differences between Wistar and Sprague – Dawley stocks. J Comp Psychol 1985;99: 93 – 7. Kellog CK, Barrett KA. Reduced progesterone metabolites are not critical for plus-maze test performance of lactating females rats. Pharmacol Biochem Behav 1999;63:441 – 8. Kolunie JM, Stern JM. Maternal aggression: disruption by perioral anesthesia in lactating Long – Evans rats (Rattus norvegicus). J Comp Psychol 1990;104:352 – 60. Kolunie JM, Stern JM. Maternal aggression in rats: effects of olfactory bulbectomy, ZnSo4-induced anosmia, and vomeronasal organ removal. Horm Behav 1995;29:418 – 92. Krehbiel DA, LeRoy LM. The quality of hormonally stimulated maternal behavior in ovariectomized rats. Horm Behav 1979;12:243 – 52. Lonstein JS, Simmons DA, Stern JM. Functions of the caudal periaqueductal gray in lactating rats: kyphosis, lordosis, maternal aggression, and fearfulness. Behav Neurosci 1998;112:1502 – 18. Lynds PG. Olfactory control of aggression in lactating female house mice. Physiol Behav 1976;17:157 – 9. Mayer AD, Ahdieh HB, Rosenblatt JS. Effects of prolonged estrogen – progesterone treatment and hypophysectomy on the stimulation of short-latency maternal behavior and aggression in female rats. Horm Behav 1990;24:152 – 73. Mayer AD, Monroy MA, Rosenblatt JS. Prolonged estrogen – progesterone treatment of nonpregnant ovariectomized rats: factors stimulating home-cage and maternal aggression and short-latency maternal behavior. Horm Behav 1990;24:3342 – 64. Mayer AD, Reisbick S, Siegel HI, Rosenblatt JS. Maternal aggression in rats: changes over pregnancy and lactation in a Sprague – Dawley strain. Aggressive Behav 1987;13:29 – 43. Mayer AD, Rosenblatt JS. Prepartum changes in maternal responsiveness and nest defense in Rattus novegicus. J Comp Psychol 1984;98: 177 – 88. Mayer AD, Rosenblatt JS. Hormonal factors influence the onset of maternal aggression in laboratory rats. Horm Behav 1987;21:253 – 67. Mayer AD, Rosenblatt JS. Persistent effects on maternal aggression of pregnancy but not of estrogen/progesterone treatment of non pregnant ovariectomized rats revealed when initiation of maternal behavior is delayed. Horm Behav 1993;27:1132 – 55. Olazabal DE, Ferreira A. Maternal behavior in rats with kainic acidinduced lesions of the hypothalamic paraventricular nucleus. Physiol Behav 1997;61:779 – 84. Ostermeyer MC. Maternal aggression. In: Elwood RW, editor. Parental behaviour of rodents. New York: Wiley, 1983. p. 151 – 79. Pellow S, Chopin P, File SE, Briley M. Validation of open: closed arm entries in an elevated plus-maze as a measure of anxiety in the rat. J Neurosci Methods 1985;14:149 – 67. Picazo O, Fe´rnandez-Guasti A. Antianxiety effects of progesterone and some of its reduced metabolites: an evaluation using the burying behavior test. Brain Res 1995;680:135 – 41. Quadagno DM, Debold JF, Gorzalka BB, Whalen RE. Maternal behavior in the rat: aspects of concaveation and neonatal androgen treatment. Physiol Behav 1974;12:1071 – 4. Rodgers RJ, Cole JC. The elevated plus-maze: pharmacology, methodology and ethology. In: Cooper SJ, Hendrie CA, editors. Ethology and psychopharmacology. Chichester: Wiley, 1994. p. 9 – 44. Rosenblatt JS. Nonhormonal basis of maternal behavior in the rat. Science 1967;156:1512 – 4. Rosenblatt JS. Prepartum and postpartum regulation of maternal behavior in the rat. Ciba Found Symp (new series) 1975;33:17 – 32. Siegel S. Nonparametric statistics for the behavioral sciences. New York: McGraw-Hill, 1956.
204
A. Ferreira et al. / Physiology & Behavior 77 (2002) 197–204
[44] Stern J, Johnson SK. Ventral somatosensory determinants of nursing behavior in Norway rats: I. Effects of variations in the quality and quantity of pups stimuli. Physiol Behav 1990;47:993 – 1011. [45] Stern JM, Kolunie JM. Trigeminal lesions and maternal behavior in Norway rats: I. Effects of cutaneous rostral snout denervation on maintenance of nurturance and maternal aggression. Behav Neurosci 1991;105:984 – 97. [46] Stern JM, Kolunie JM. Maternal aggression of rats is impaired by cutaneous anesthesia of ventral trunk, but not by nipple removal. Physiol Behav 1993;54:861 – 8. [47] Stern JM, Mackinnon DA. Postpartum, hormonal, and nonhormonal induction of maternal behavior in rats: effects on T-maze retrieval of pups. Horm Behav 1970;7:305 – 16. [48] Stern J, Siegel S. Prolactin release in lactating, primiparous and multiparous thelectomized and maternal virgin rats exposed to pups stimulis. Biol Reprod 1978;19:177 – 82.
[49] Svare BB. Maternal aggression in mammals. In: Gubernick DJ, Klopfer PH, editors. Parental care in mammals. New York: Plenum, 1981. p. 179 – 210. [50] Svare BB, Mann MA. Hormonal influence on maternal aggression. In: Svare BB, editor. Hormones and aggressive behavior. New York: Plenum, 1983. p. 91 – 104. [51] Terkel J, Rosenblatt JS. Aspects of nonhormonal maternal behavior in the rat. Horm Behav 1971;2:161 – 71. [52] Valzelli L. The ‘‘isolation syndrome’’ in mice. Psychopharmacology 1973;31:305 – 20. [53] Wongwidecha N, Mardsen CA. Social isolation increases aggressive behavior and alters the effect of diazepam in the rat social interaction test. Behav Brain Res 1996;75:27 – 32.