A postpartum separation induces c‐Fos expression in the supramammillary nucleus of lactating rats

Abstract

Aim

Elucidation of the neural mechanism of maternal behaviors is a medically and biologically important research task. The rat is the laboratory animal most extensively analyzed for maternal behaviors. However, the neural mechanism that maintains the motivation of postpartum rats for maternal behaviors has not yet been elucidated. In this study, we aimed to identify brain regions involved in the maintenance of motivation for maternal behaviors by detecting brain regions that exhibit changes in nerve activity when the mother rat is separated from her pups.

Methods

Lactating mother rats were separated from their pups on postpartum day 3 and kept away from the pups for a certain period of time, and brain regions that exhibited changes in nerve activity when the rats were separated from their pups and those that exhibited changes in nerve activity when the pups are returned were detected by immunohistochemistry using anti‐c‐Fos antibody, a marker for increased nerve activity.

Results

Rats that were separated from their pups and with the pups returned later showed increases in the number of c‐Fos immunoreactive (c‐Fos‐IR) cells in the medial preoptic area (MPA), the bed nucleus of the stria terminalis (BST), the caudal portion of posterior hypothalamic area (PH) and the supramamillary nucleus (SUM). In mother rats permanently separated from their pups, only the PH and SUM exhibited an increase in the number of c‐Fos‐IR cells.

Conclusion

In rats, the SUM is involved in aversive memory and changes in the postpartum anxiety level. The observed increase in the number of c‐Fos‐IR cells in the SUM of mother rats separated from their pups suggests that the nerve activity change in the SUM, which is involved in aversive memory and anxiety, is involved in the maintenance of maternal behaviors.

Introduction

Elucidation of the neural mechanism involved in the onset and maintenance of maternal behaviors is an important research task in the human society, where neglect and child abuse are serious issues [1], as well as in the field of animal sciences, where an improvement in breeding technology for rare animals is demanded [2, 3]. The rat is the laboratory animal most extensively analyzed for the neural mechanisms of maternal behaviors [4, 5]. A previous study has revealed that the medial preoptic area (MPA) and bed nucleus of the stria terminalis (BST) of postpartum rats play the most important role in perceiving the presence of their pups [6]. A review article by Numan [4] states that information from pups input into the sex hormone‐sensitized MPA and BST generates an appetitive response related to maternal behaviors. This information is further transmitted to the periaqueductal gray (PGA) and ventral tegmental area (VTA) located in the posterior part of the diencephalon, resulting in the onset of actual maternal behaviors. Although separating a lactating mother rat from her pups causes certain changes to the brain of the mother rat, it is unclear what kind of change leads to the maintenance or loss of motivation for maternal behaviors. Elucidation of the neural mechanisms involved in the maintenance of maternal behaviors of mother rats that have been separated from their pups will provide useful information for understanding the mechanisms of mothers’ neglecting their children and aggressive behaviors to their newborn babies (human) or pups (other animals). The c‐Fos is a good marker for neuronal activation in the brain [7, 8], and immunohistochemistry with anti‐c‐Fos antibody is used for identifying brain regions involved in the reproductive behavior of various animals [9].

In this study, we aimed to identify brain regions involved in the maintenance of maternal behaviors of postpartum mother rats by detecting brain regions that exhibit neuronal activity change when the mother rat is separated from her pups, using immunohistochemistry for c‐Fos.

Materials and methods

Behavioral testing and tissue preparation

Virginal female Wistar–Imamichi rats (230–250 g) obtained from the Imamichi Institute for Animal Reprodution (Ibaraki, Japan) were used. On arrival in our laboratory, the rats were housed in standard laboratory cages with wood‐chip bedding on the floor (2 animals/cage) in an air‐conditioned room (24°C) with a controlled light–dark cycle (12 h light/12 h dark). Food and water were provided ad libitum. The rats were mated by placing one stud male into a female cage, and approximately 18 days later, pregnant females were housed individually in cages. Animal protocols used were approved by the Nippon Veterinary and Life Science University Animal Care and Use Committee.

On postpartum day 1 (PD1), all litters were culled to eight pups. Then, on the day of the experiment at midnight on PD3, the postpartum animals were divided into three groups as follows (Fig. 1): (1) Lactating females housed with their pups (Control), (2) Postpartum lactating females with their pups withdrawn at 0:00 on PD3 (Remove), and (3) Postpartum lactating females with their pups withdrawn at 0:00 on PD3 and then returned between 8:00 and 10:00 on PD3 (Return). We repeated these experiments seven times for the three groups. Pups that separated from their original mother were cared for by another mother that delivered within 2 days. At 11:30 on PD3, each postpartum rat was deeply anesthetized by injection of sodium pentobarbital (40 mg/kg, i.p.) and then perfused transcardially with saline followed by a fixative solution (4% paraformaldehyde in 0.1 M phosphate buffer (PB), pH 7.4). The brain was removed, post‐fixed overnight in the same fixative solution at 4°C, and incubated in 30% sucrose in 0.1 M PB at 4°C until it sank.

Figure 1.

a Experimental schedule. Each arrow head indicates the following; a remove all pups, b return all pups, c remove all pups, d perfused transcardially with fixative solution. b The schematic illustrates the brain areas for c‐Fos‐IR cell counts (gray part). MPA medial preoptic area, BST bed nucleus of the stria terminalis, PH caudal portion of posterior hypothalamic area, SUM supramamillary nucleus, LV lateral ventricle, V3 third ventricle, MM medial mammillary nucleus, PMD premammillary nucleus dorsal part, cc corpus callosum

Immunocytochemistry for c‐Fos

Serial frontal sections of brain tissue (40 μm thick) were cut on a freezing microtome (Yamato, Japan) from the area just in front of the MPA to the area just behind of the arcuate nucleus. Every second section was used for immunocytochemistry to detect c‐Fos protein according to our previously reported method [10]. In brief, sections were rinsed with 0.1 M phosphate‐buffered saline (PBS, pH 7.4), and then treated with 3% hydrogen peroxide in absolute methanol for 15 min. The sections were thereafter rinsed with 0.3% Triton X‐100 in 50 mM PBS (PBST, pH 7.4) for 45 min with three changes, and then nonspecific binding sites were blocked by incubation with 1% bovine serum albumin (BSA) in PBST (BSA‐PBST) for 1 h at room temperature. The sections were subsequently incubated with an anti‐human c‐Fos antibody (rabbit polyclonal IgG, Ab‐5, Lot no. D07130, Oncogene Research Products, Cambridge, MA, UK; diluted 1:40,000 with BSA‐PBST) for approximately 50 h at 4°C. Next, the sections were incubated with biotinylated anti‐rabbit IgG (7.5 μg/ml; Vector Labs, Burlingame, CA, USA) in BSA‐PBST for 1 h at room temperature, followed by avidin–biotin complex solution (1:100; Amersham Life Science, Amersham, Buckinghamshire, UK) in BSA‐PBST for 1 h at room temperature. Each step was followed by three 15‐min washes with PBST. After the last wash, the sections were immersed in 0.175 M sodium acetate buffer (pH 7.4) for 30 min with two changes, and then incubated with the chromogen solution (0.25 mg/ml nickel chloride, 0.2 mg/ml 3,3′‐diaminobenzidine and 0.0025% hydrogen peroxide) in the same buffer for 8 min. The reaction was stopped by transferring the sections to sodium acetate buffer, after which they were washed with 10 mM PBS for 20 min with three changes. Finally, the sections were mounted on gelatin‐coated glass slides. The alternate sections were stained with cresyl violet for identification of brain structures.

Quantification of c‐Fos‐immunoreactive cells

To quantify c‐Fos‐immunoreactive (c‐Fos‐IR) cells, sections were carefully matched across animals under a microscope according to the appearance of brain structures in the sections stained with cresyl violet and by immunocytochemistry. For each rat (n = 7 per group), three sections for each brain region (Fig. 1a) corresponding to sections in the atlas of the rat brain [11] were selected as follows: the medial preoptic area (MPA: Figs. 21 and 22); all regions of the bed nucleus of the stria terminalis (BST: Figs. 22 and 23), the caudal portion of posterior hypothalamic area (PH: from Figs. 34 to 36) and supramamillary nucleus (SUM: Figs 35 and 36). The number of c‐Fos‐IR cells within hemisections of the MPA, BST, PH and SUM were counted using a computer‐assisted image analysis system (KS‐300, Zeiss, Germany). Each pixel within the area to be measured was scaled to 0–255 (0 = black, 255 = white). An image brighter than 201 was discarded as background level, while darker signals less than 60 were transformed to scale 0 as positive signals. The scaled image between 50 and 200 was then expanded to 0–250. Signals below the threshold level of 170 were considered as positive. The numbers of c‐Fos‐IR cells were calculated as the value of the total area of each nucleus. Data for the numbers of c‐Fos‐IR cells between three groups were analyzed using one‐way ANOVA, followed by the post‐hoc Newman–Keuls multiple comparison test. P < 0.05 was considered significant (GraphPad Prism version 4; GraphPad Software, Inc., USA).

Results

MPA and BST

The results for the MPA and BST as representative areas exhibiting different c‐Fos‐IR responses are shown in Fig. 2b. After returning the pups, the number of c‐Fos‐IR cells in the MPA and BST significantly increased compared with other experimental groups (Figs. 2b, 3 MPN, BST).

Figure 2.

Photomicrographs showing c‐Fos‐IR cells in the MPA, BST (a, b), PH, and SUM (c–f). a Histological positions of the MPN and BST (only the ventral part of the BST are shown). Cresyl violet staining section. b c‐Fos‐IR cells in the “Return” female. c Histological positions of the SUM and PH (only the ventral part of the PH are shown). Cresyl violet staining. d c‐Fos‐IR cells in the “Control” female. e c‐Fos‐IR cells in the “Remove” female. f c‐Fos‐IR cells in the “Return” female. Scale bar in a and c = 200 μm

PH and SUM

After returning the pups, the number of c‐Fos‐IR cells in the PH and SUM significantly increased compared with other experimental groups (Fig. 3 PH, SUM). Moreover, in the “Remove” group c‐Fos‐IR expression was also elicited in the PH and SUM compared with the control animals (Fig. 3 PH, SUM). However, the rate of increase in c‐Fos‐IR cells in the “Remove” group was significantly smaller than that in the “Return” group.

Figure 3.

Quantitative analyses of the number of c‐Fos‐IR cells in the medial preoptic area (MPA), the bed nucleus of the stria terminalis (BST), the posterior hypothalamic area (PH) and the supramamilary nucleus (SUM). Histograms represent mean (±SEM) values. Bars not labeled with the same alphabetical letters are significantly different at P < 0.05

Other maternal behavior‐related brain areas

Significant changes in the number of c‐Fos‐IR cells were not detected in the amygdaloid complex between animals (data not shown). Moreover, c‐Fos‐IR cells were not detected in the ventromedial hypothalamic nucleus in any experimental animals including the controls (data not shown).

Discussion

Separating postpartum lactating mother rats from their pups and returning the pups after 8 h resulted in a marked increase in the number of c‐Fos‐IR cells in the MPA and BST of the mother rats (Fig. 3). The MPA and BST have been shown to exhibit increased expression of c‐Fos in association with maternal behaviors of mother rats. The neuronal activities in these nuclei appear to be stimulated by new information input from pups [4, 5]. Thus, the reason for the increase in the number of c‐Fos‐IR cells in the MPN and BST of “Return” mother rats was probably because “new” stimuli from returned pups were inputted into the nuclei. In contrast, mother rats housed continuously with their pups (Control) and those permanently separated from their pups (Remove) exhibited no increase in the number of c‐Fos‐IR cells in the MPA and BST (Fig. 3). In the “Remove” mother rats, which were permanently separated from their pups, the lack of information input from pups should explain the absence of an increase in the number of c‐Fos‐IR cells. Meanwhile, the reason for the absence of an increase in the number of c‐Fos‐IR cells in “Control” rats, which were continuously housed with their pups, appears to be because these rats were exposed to their pups all the time. Given that the neuronal activities in the MPA and BST are stimulated by new information input from pups and that c‐Fos is a marker of newly excited neurons, it is reasonable to think that “Control” mother rats showed no increase in the number of c‐Fos‐IR cells in the MPA and BST because these rats did not receive “new” information from pups.

The novel finding in the present study is c‐Fos expression in the SUM and PH. The SUM is a nucleus that receives efferent projections from neurons in the BST [12] and thus is believed to be part of the neural circuit for maternal behaviors that receives information from the BST. The involvement of the SUM in aversive memory has also been reported [13]. Smith and Lonstein [14] have recently analyzed the relationship between fear level and c‐Fos expression in postpartum rats using an elevated plus‐maze and found that the SUM, as well as the BST, exhibited the most prominent change in c‐Fos expression. These findings suggest that the SUM is a brain region in which the neural circuit related to changes in the fear threshold of postpartum mother rats intersects with the neural circuit that induces maternal behaviors in the presence of pups.

Taken together, the marked increase in the number of c‐Fos‐IR cells in the SUM of the “Return” mother rats appears to be associated with an increased neuronal activity in the BST induced by the presence of pups. On the other hand, the relatively modest increase in the number of c‐Fos‐IR cells in the SUM of the “Remove” mother rats may be due to sustained fear or anxiety of the mother rats caused by separation from their pups. While the neuronal activities in the MPN and BST, well‐known centers of maternal behaviors, are stimulated by the presence of pups, the activation of the SUM may maintain the motivation of mother rats for maternal behaviors by having the brain memorize fear and anxiety of losing pups. If “presence” information is re‐transmitted to the SUM through the BST, a comparable level of neuronal activity to that induced in the MPA and BST is also induced in the SUM, which might have led to a prominent increase in the number of c‐Fos‐IR cells. If this hypothesis is true, the SUM should be considered as a brain region that plays an important role in the maintenance of maternal behaviors of postpartum lactating mother rats.

With regard to the PH, which exhibited a similar pattern of increase in c‐Fos‐IR cells to that for the SUM, we were unable to find evidence suggesting its involvement in maternal behaviors based on its physiological characteristics or neural circuits. It is thus currently unclear how the increased c‐Fos expression in the PH might influence maternal behaviors. The PH is adjacent to the ventral portion of the periaqueductal gray (PAG), whose involvement in the avoidance and defensive behaviors of mother rats have been demonstrated [15, 16]. In addition, a study suggests that the ventral tegmental area (VTA), which is located posterior to the PAG on the ventral side of the brain, has been shown to be involved in the maternal responsiveness of mother rats [5]. We did not prepare specimens of the PAG and VTA for analysis in this study and therefore were unable to determine whether the change in c‐Fos expression in the PH is correlated with that in the PGA and/or VTA. It is still worthwhile to explore the possibility that these adjacent areas are somehow related to each other, regulate the onset of the avoidance and defensive behaviors and are involved in the maintenance of motivation for maternal behaviors of postpartum lactating mother rats.

It is reported that the suckling stimulus by pups is also an important factor to have an influence on motivation and stress of the maternal behavior [18, 19, 20]. However, the suckling stimulus by pups induces activation of the brain area such as the nucleus of the solitary tract, ventrolateral medulla, some hypothalamic neurons which including the tuberoinfundibular dopaminergic neurons, the arcuate nucleus, and periventricular hypothalamic nucleus of the postpartum mother rats [17, 18, 19, 20]. Therefore, it seems that the increasing c‐Fos‐IR cells in SUM and PH is not a response for the direct contact with pups such as suckling.

Further studies on the involvement of the SUM and PH, as well as the PGA and VTG, are currently in progress.