- Email: [email protected]
Periaqueductal gray μ and κ opioid receptors determine behavioral selection from maternal to predatory behavior in lactating rats
Behavioural Brain Research 274 (2014) 62–72
Contents lists available at ScienceDirect
Behavioural Brain Research journal homepage: www.elsevier.com/locate/bbr
Research report
Periaqueductal gray and opioid receptors determine behavioral selection from maternal to predatory behavior in lactating rats Marianne Orlandini Klein a,b , Aline de Mello Cruz a , Franciele Corrêa Machado b,d , Gisele Picolo d , Newton Sabino Canteras c , Luciano Freitas Felicio a,∗ a Department of Pathology, School of Veterinary Medicine, University of São Paulo, Avenida Professor Doutor Orlando Marques de Paiva, 87, São Paulo, SP CEP 05508-270, Brazil b Departments of Pharmacology, Institute of Biomedical Sciences, University of São Paulo, Avenida Prof. Lineu Prestes, 1524, Biomédicas I, São Pauo, SP CEP 05508-900, Brazil c Department of Anatomy, Institute of Biomedical Sciences, University of São Paulo, 2415, São Paulo, SP CEP 05508-900, Brazil d Special Laboratory of Pain and Signaling, Butantan Institute, Avenida Vital Brazil, 1500, São Paulo CEP 05503-900, Brazil
h i g h l i g h t s • • • •
The rlPAG and opioid receptors have a role in behavioral selection. Morphine in the rlPAG inhibits maternal behavior without interfering with hunting. Receptor blockade increases hunting and increases receptor activation. Multiple opioid receptor interactions play a role in behavioral selection.
a r t i c l e
i n f o
Article history: Received 17 April 2014 Received in revised form 21 July 2014 Accepted 4 August 2014 Available online 10 August 2014 Keywords: Behavioral selection Maternal behavior Predatory hunting Periaqueductal gray
a b s t r a c t Every mother must optimize her time between caring for her young and her subsistence. The rostro lateral portion of the periaqueductal grey (rlPAG) is a critical site that modulates the switch between maternal and predatory behavior. Opioids play multiple roles in both maternal behavior and this switching process. The present study used a pharmacological approach to evaluate the functional role of rlPAG and opioid receptors in behavioral selection. Rat dams were implanted with a guide cannula in the rlPAG and divided into three experiments in which we tested the role of opioid agonists (Experiment 1), the influence of and opioid receptor blockade in the presence of morphine (Experiment 2), and the influence of and opioid receptor blockade (Experiment 3). After behavioral test, in Experiment 4, we evaluated rlPAG and receptor activation in all Experiments 1–3. The results showed that massive opioidergic activation induced by morphine in the rlPAG inhibited maternal behavior without interfering with predatory hunting. No behavioral changes and no receptor activation were promoted by the specific agonist alone. However, receptor blockade increased hunting behavior and increased the level of receptor activation in the rlPAG. Thus, endogenous opioidergic tone might be modulated by a functional interaction between opioid receptor subtypes. Such a compensatory receptor interaction appears to be relevant for behavioral selection among motivated behaviors. These findings indicate a role for multiple opioid receptor interactions in the modulation of behavioral selection between maternal and predatory behaviors in the PAG. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Lactating mammalian mothers must optimize the constant exchange between reproduction and subsistence [1,2]. Indeed, a
∗ Corresponding author. Tel.: +55 11 3091 7934; fax: +55 11 3091 7829. E-mail address: [email protected] (L.F. Felicio). http://dx.doi.org/10.1016/j.bbr.2014.08.008 0166-4328/© 2014 Elsevier B.V. All rights reserved.
mother has to use her time more efficiently than a virgin female to express maternal and predatory behaviors, which has adaptive value [3–5]. In mammals, maternal behavior is fundamental for perpetuation of the species, in which the neonate’s survival depends on the mother and her ability to provide food, heat, shelter, and protection [6]. Predatory behavior is also essential for individual survival, and it is characteristic of the specie. During the post-partum period,
M.O. Klein et al. / Behavioural Brain Research 274 (2014) 62–72
the time spent foraging and hunting depends on environmental circumstances [2]. As a stimulus to elicit predatory behavior, we used cockroaches, which appear to be suitable prey and have an inherent hedonic value for the rat [7–11]. The periaqueductal gray (PAG) is an important region of the central nervous system that is involved in multiple behavioral and physiological processes, including nociception, fear, anxiety, cardiovascular control, sexual behavior, and vocalization [12]. Beyond these functions, previous studies have demonstrated that the PAG, mainly its rostrolateral portion (rlPAG), also plays an important role in maternal and predatory behavioral selection, and opioids are involved in this switch between behavioral modes [2,3,11,13]. Opioids are well established to be involved in the regulation of maternal behavior [14–16]. Morphine is a nonselective opioid agonist that acts preferentially at opioid receptors [17]. A single injection of morphine directly into the rlPAG impairs maternal behavior [18]. Nevertheless, the specific effects of a morphine injection in the rlPAG on predatory hunting vs. maternal behavior have not been investigated previously. Different types of opioid receptors have been described, and the and subtypes are the most widely expressed and functionally relevant [19]. Nevertheless, unknown is the functional relevance of these receptors in the rlPAG for behavioral selection during the post-partum period. An intracerebroventricular injection of a specific agonist in lactating females dose-dependently disrupted maternal behavior [20]. Acute peripheral injections of a specific opioid agonist increased the latency to retrieve pups [21]. In the present study, lactating females were tested for behavioral selection using various pharmacological challenges that involved both stimulation and blockade of rlPAG and opioid receptors. The state of receptor activation was assessed by an immunological assay. Our results showed that the central functional interaction between both receptors might be important for maintaining the normal expression of predatory behavior, and morphine injected directly into the rlPAG was shown to act on both receptors to inhibit maternal behavior. 2. Materials and methods 2.1. Animals Female nulliparous Wistar rats were ∼90 days of age at the beginning of the experiments. They were mated with sexually experienced males, and the day that sperm was observed in the vaginal lavage was considered day 1 of pregnancy. Pregnant females were individually housed in polypropylene cages (30 cm × 40 cm × 18 cm) with pine flakes. They had free access to water and food during all of the experiments. The light/dark cycle (12 h/12 h) and temperature (23 ± 2 ◦ C) were controlled. On day 15–17 of pregnancy, females underwent stereotaxic surgery to implant a unilateral guide cannula in the rlPAG. After giving birth (day 0 of lactation), the dams remained with their litter, which was culled to eight pups (four males and four females) on day 2 of lactation until the day of the behavioral test. All of the procedures were in accordance with the Ethical Principles of Animal Experimentation adopted by the Sociedade Brasileira de Ciências em Animais de Laboratório (SBCAL) and approved by the Comissão de Ética no Uso de Animais (CEUA) of the Biomedical Sciences Institute and School of Veterinary Medicine at the University of São Paulo. 2.2. Stereotaxic surgery On day 15–17 of pregnancy, the females were intraperitoneally anesthetized with a mixture of ketamine (60 mg/kg Vetanarcol, Santana de Parnaíba, SP, Brazil) and xylazine (5 mg/kg Kensol,
63
Santana de Parnaíba, SP, Brazil) and received 3 mg/kg dipyrone (Ibasa, Porto Alegre, RS, Brazil) for postsurgical analgesia. The females were then placed in a stereotaxic apparatus to implant a guide cannula (Plastic One, Roanoke, VA, USA) in the rlPAG. The stereotaxic coordinates were −6.0 mm from bregma, −0.6 mm lateral to the midline, and 4.2 mm ventral to the surface of the brain [22]. On the test day, the volume of drug or saline (0.6 l) was injected over 30 s, and the internal cannula was left in place for an additional 30 s to allow for diffusion. The internal cannula was 1 mm longer than the guide cannula, allowing precise injection into the rlPAG. Only animals with correct cannula placements were used in the study. 2.3. Behavioral analyses Behavioral test was not performed in the home cage of animals. The test cage was constructed of transparent Plexiglas to allow complete visualization of the pups and cockroaches and prevent the insects from escaping. To permit complete observation of both the insects and mammals’ behaviors, no pine flakes were used in these cages. One day before the tests, the dams were habituated to the test cage. Females without their pups were habituated to exploring the experimental cage for 30 min. The bottom of the experimental cage was divided by two perpendicular lines. Each time the animal crossed a line, it was counted as an indication of locomotion. Behavioral tests were performed on day 5 or 6 of lactation. The animals were tested for behavioral selection (i.e., caring for pups or predatory hunting) and were not food deprived before the tests. On the test day, 60 min before the test, lactating females were placed in the test cage without their pups. Thirty minutes later, the dams received a drug/saline injection, and the behavioral test began 30 min after the injection. Eight pups and five mature cockroaches (Leurolestes circunvagans) were dispersed introduced to the cage at the beginning of the behavioral test. All of the trials were recorded for subsequent analysis. During the 30 min trial, the following maternal behavior parameters were analyzed: latency to retrieve the 1st, 5th, and 8th pups, the percentage of dams that retrieved these pups, the number of contacts with the pups (i.e., when lactating females only touched them without retrieving them), the number of times the pups were licked, the percentage of dams that nursed their pups, and the expression of full maternal behavior (FMB), in which the dam, after retrieving and grouping at least five pups, arched over them with her legs splayed, with the pups attached to the nipples, and the dam remained in this position for at least three consecutive minutes. For the predatory hunting analysis, we recorded the latency to capture each cockroach, percentage of dams that captured the 1st, 2nd, 3rd, 4th, and 5th insects, and total number of insects hunted. General parameters, such as the number of lines crossed, time spent exploring the cage, and self-grooming time, were also analyzed. In all of the experiments, the experimental cage was cleaned with a 5% alcohol solution before each behavioral test to eliminate possible bias caused by odors left by previous animals. Experimental and control observations were intermixed to minimize possible circadian influences on the dams’ behavior. 2.4. Experiment 1 The goal of this experiment was to determine whether injections of morphine and the and opioid agonists in the rlPAG interfere with maternal behavioral selection and predatory hunting. Behavioral analyses were performed in four different groups: morphine (MOR; n = 13; 7.91 nmol morphine sulfate; Dimorf, São Paulo, SP, Brazil), opioid receptor agonist (; n = 12; 10 nmol DAMGO; Sigma-Aldrich, St Louis, MO, USA), opioid receptor agonist (; n = 12; 10 nmol U69593; Tocris Bioscience, Bristol, UK), and
64
M.O. Klein et al. / Behavioural Brain Research 274 (2014) 62–72
Fig. 1. Injection loci. (A) Photomicrograph of the periaqueductal gray (PAG) with Nissl staining. (B) Diagram that shows the main injection loci in each group. Because of overlap, the number of illustrated points may be less than the actual number of injections. The injections were made unilaterally. Each figure represents the values of PAG regions posterior to bregma (−5.8 mm, −6.04 mm, −6.3 mm, −6.72 mm, and −6.80 mm). Only animals that had correct injection sites were considered in this study. Aq, aqueduct; EW, Edinger–Westphal nucleus. Adapted from reference [22].
control (CS; n = 10; 0.9% physiological saline) [20]. The injections were performed as described above. The dose of morphine sulfate was not equimolar to the other drug doses because of technical issues. Morphine is already sold diluted; thus, an equimolar morphine dose would imply a drug volume that is too large to be injected in the animal brain [18]. 2.5. Experiment 2 In this experiment, we tested the importance of and opioid receptors in morphine’s action. Four groups were formed: morphine (MOR; n = 13; 7.91 nmol morphine sulfate; Dimorf, São Paulo, SP, Brazil), opioid receptor antagonist plus morphine (NM; n = 11; 10 nmol of the opioid receptor antagonist CTAP [Tocris Bioscience, Bristol, UK] plus 7.91 nmol morphine sulfate), opioid receptor antagonist plus morphine (NM; n = 12; 10 nmol of the opioid antagonist nor-binaltorphimine dihydrochloride (Tocris Bioscience, Bristol, UK) plus 7.91 nmol morphine sulfate), and control (CS; n = 10; which received 0.9% physiological saline) [20]. The injections and behavioral analyses were performed as described above. For the groups that was necessary the injection of two drugs, first it was administered the antagonist drug and 30 s after, the morphine. 2.6. Experiment 3 Considering the hypothesis that opioid receptor subtypes play differential roles in behavioral selection, this experiment was performed to investigate the role of endogenous and opioid receptor subtypes in behavioral selection using their respective antagonists. Three different groups were tested: opioid receptor
antagonist (N; n = 12; 10 nmol of the opioid receptor antagonist CTAP), opioid antagonist (N; n = 11; 10 nmol of the opioid receptor antagonist nor-binaltorphimine dihydrochloride), and control (CS; n = 10; which received 0.9% physiological saline). All of the behavioral analyses and drug injections were performed as described above.
2.7. Experiment 4 This experiment was conducted to evaluate the activation of and opioid receptors under the conditions of the previous experiments, i.e., each group for each experiment (1–3) as previously reported was submitted to an immunofluorescence assay after the behavioral test to evaluate the activation of and receptors under drugs action in rlPAG. Thus, conformation state-sensitive anti- and anti- antibodies were used. When the receptors are activated, they undergo a conformational change that is detected by these antibodies, thus recognizing the activated state of the receptor. It is a well established and validated technique [23,24]. To perform this experiment, 60 min after the end of the behavioral test, the dams were anesthetized with 70 mg/kg ketamine (Vetanarcol, Santana de Parnaíba, SP, Brazil) and 10 mg/kg xylazine (Kensol, Santana de Parnaíba, SP, Brazil) and transcardially perfused with a solution of 4.0% formaldehyde and sodium tetraborate, pH 7.4. The brains were removed and left overnight in a solution of 20% sucrose in 0.1 M phosphate buffer at 4 ◦ C. Frozen brain sections (30 m) of the rlPAG were cut with a sliding microtome in the frontal plane. The sections were mounted on gelatin-coated slides and blocked with 1% bovine serum albumin + 5% sucrose and processed for immunostaining. Tissue samples were incubated overnight with conformation state-sensitive anti- and anti- primary antibodies
M.O. Klein et al. / Behavioural Brain Research 274 (2014) 62–72
65
Fig. 2. Predatory hunting parameters in Experiment 1. CS, control group (n = 10); MOR, morphine group (n = 13); , receptor agonist group (DAMGO; n = 12); , receptor agonist group (U69593; n = 12). (A) Number of insects captured during the behavioral test. The data are expressed as mean ± SEM (one-way ANOVA). (B) Latency to capture the 1st, 2nd, 3rd, 4th, and 5th cockroaches. The data are expressed as mean ± SEM (Kruskal–Wallis test). (C) Percentage of dams that captured the 1st, 2nd, 3rd, 4th, and 5th cockroaches. p > 0.05 (Fisher’s exact test).
(Proteimax Biotechnology, Cotia, SP, Brazil) labeled with fluorescent Alexa Fluor dye (682 nm for opioid receptors and 800 nm for opioid receptors) diluted 1:4000. The Odyssey system (LiCOR) was used to analyze and quantify the intensity of fluorescence. Differences in receptor activation were evaluated by considering, beyond the tissue area, the total number of each receptor expressed in the same brain tissue using opioid receptor (N-terminus) and opioid receptor-1 (N-19) rabbit polyclonal immunoglobulin G diluted 1:100 (Santa Cruz Biotechnology, Santa Cruz, CA, USA). 2.8. Statistical analysis Because of the number of mothers that did not retrieve their pups, we expressed the latency to retrieve each pup as the following latency scores: 1 (latency = 0–300 s), 2 (latency = 301–600 s), 3 (latency = 601–900 s), 4 (latency: 901–1200 s), 5 (latency = 1201–1500 s), 6 (latency: 1501–1800 s), and 7 (when the dams do not retrieve the pups during the 1800 s of the behavioral test). We were thus able to include all of the dams in the statistical analysis. One-way analysis of variance (ANOVA) was used for predatory parameters (i.e., the number of cockroaches captured and latency to capture each cockroach), general parameters (i.e., the time exploring the cage, number of lines crossed in the cage, and self-grooming time), and maternal behavior of licking the pups. The Kruskal–Wallis test was used
for pup retrieval latency scores. Fisher’s exact test was used to analyze the percentage parameters, such as the dams that captured cockroaches, dams that retrieved pups, dams that expressed nursing behavior, and dams that expressed FMB. In Experiment 4, a one-way ANOVA was used to do a comparison of the groups of each experiment (1–3; a group of Experiment 1 was not compared to a group of Experiment 3, for example; and the activation of and receptors were not compared to each other). Data are expressed as a percentage of control activation, i.e. the closer the activation is of 100% more similar to control’s activation it is. Statistical significance was set at p < 0.05 for all the data. The statistical analyses were performed using GraphPad Prism 5.0. 3. Results For all of the experiments, only the animals that had the guide cannula implanted correctly into the rlPAG were considered (Fig. 1). Dams did not exhibit any difference in the number of lines crossed, time spent exploring the cage, and self-grooming time in any of the behavioral experiments, indicating that the drugs did not impair general parameters (p > 0.05, data not shown). Throughout the observation period, the dams in the MOR group and other groups that received morphine or specific opioid drug injections in Experiments 2 and 3 did not show any stereotypical behavior, motor rigidity, or signs of sedation that morphine may cause at
66
M.O. Klein et al. / Behavioural Brain Research 274 (2014) 62–72
Fig. 3. Maternal parameters in Experiment 1. CS, control group (n = 10); MOR, morphine group (n = 13); , receptor agonist group (DAMGO; n = 12); , receptor agonist group (U69593; n = 12). (A) Latency to retrieve the 1st, 5th, and 8th pups after being transformed into rank-order data (1: 0–300 s; 2: 301–600 s; 3: 601–900 s; 4: 901–1200 s; 5: 1201–1500 s; 6: 1501–1800 s; 7: no retrieval during 30 min test). The data are expressed as mean ± SEM (Kruskal–Wallis test followed by Dunn’s test). (B) Percentage of dams that retrieved each of their pups. *p < 0.05, compared with control (CS) and groups; # p < 0.05, compared with CS group (Fisher’s exact test).
higher doses [25]. Animals that expressed hunting behavior sniffed around the cage, some more vigorously than others, depending on the group, foraged the prey normally, and attempted to seize the prey with apparently no difficulty. These results indicate that the animals exhibited no motor alterations caused by the drugs. 3.1. Experiment 1: Role of morphine, and opioid agonists in rlPAG in behavioral selection In Experiment 1, the effects opioids’ action in the rlPAG by morphine and the agonists of and receptors varied. Both morphine and the and opioid receptor-specific agonists did not interfere with the predatory parameters (Fig. 2). No differences were found among groups in the number of cockroaches captured, latency to capture each cockroach, and percentage of dams that captured each of the five cockroaches. Maternal parameters were impaired in the MOR group compared with the CS group (Fig. 3). The latencies scores to retrieve the 1st (p = 0.007), 5th (p = 0.005), and 8th (p = 0.050) pups were longer in the dams that received morphine compared with controls. The percentage of animals that retrieved each pup was smaller in the MOR group compared with the CS group (p < 0.05 for all comparisons). The time spent licking the pups was less in the MOR group compared with the CS and groups (F3,46 = 4.526, p = 0.0076), but the percentages of animals that displayed FMB and nursing and number of pup contacts were not different among groups (Table 1). Stimulation of a single or opioid receptor did not significantly interfere with the predatory or maternal parameters. No significant differences were found in the predatory parameters during the entire time of observation. Thus, considering the total time of the test, the control group hunted as well as the MOR group. To test whether the groups had different hunting strategies, Table 1 Maternal parameters analyzed for 1800 s in Experiment 1. Parameter
FMB (%) Nursing (%) Pup contact Licking (s)
we evaluated primary motivation at the beginning of the test (i.e., whether the dams preferred to hunt before or after caring for their pups). Therefore, we decided to compare the number of insects captured in the initial 15 min (0–900 s) with the total time of the test (1800 s). We did the same with the number of pups retrieved (Table 2). Indeed, the control group preferred to retrieve most of their pups prior to capturing the cockroaches, whereas the MOR group captured almost all of the cockroaches at the beginning of the test. The group, although not statistically significant, tended to behave similarly to the MOR group, whereas the group was more similar to the control group. Thus, morphine appeared to act at both receptors at the same time to disrupt maternal behavior. To evaluate the importance of each of these opioid receptor subtypes in morphine’s action, we performed Experiment 2. 3.2. Experiment 2: Blockade of and opioid receptors face of morphine action in rlPAG In this experiment, opioid receptor blockade under the action of morphine in the rlPAG impaired hunting behavior in these dams (Fig. 4). The NM group captured fewer cockroaches than the CS and MOR groups (F3,45 = 4.147, p = 0.0116). Furthermore, the latencies to capture the 1st to 4th cockroaches were longer in the NM group compared with the CS and MOR groups (p < 0.05 for all comparisons). The percentage of lactating females in the NM group that captured each of the five insects was lower (p < 0.05 for all comparisons). With regard to maternal parameters, all of the groups that received a morphine injection (i.e., MOR, NM, and NM groups) exhibited some disruptions of aspects of this behavior (Fig. 5). The latency scores to retrieve the 1st pup in all of these groups were longer compared with the CS group (p = 0.001). The MOR and NM group also longer latency scores to retrieve the 5th Table 2 Insects captured and pups retrieved in the initial 900 s and during the entire behavioral test (1800 s) in Experiment 1.
Group CS
MOR
20.0 60.0 7.0 ± 2.6 130.1 ± 24.9
0.0 38.5 4.8 ± 1.0 35.8 ± 14.0*
16.7 50.0 5.7 ± 1.4 131.9 ± 29.9
16.7 75.0 6.8 ± 1.1 116.6 ± 18.1
FMB, full maternal behavior; CS, control group (n = 10); MOR, morphine group (n = 13); , receptor agonist group (DAMGO; n = 12); , receptor agonist group (U69593; n = 12). FMB and nursing behavior were analyzed using Fisher’s exact test. Pup contacts were analyzed using the Kruskal–Wallis test. Licking behavior was analyzed using one-way ANOVA followed by Tukey’s test. The data are expressed as mean ± SEM. * p < 0.05, compared with CS and groups.
Parameter
Insects captured until 900 s Insects captured during 1800 s Pups retrieved until 900 s Pups retrieved during 1800 s
Group CS
MOR
1.5 (0–5) 4 (0–5) 7 (0–8) 8 (7–8)
4 (0–5) 5 (0–5) 0 (0–8)* 0 (0–8)**
3.5 (0–5) 5 (0–5) 5 (0–8) 5.5 (0–8)
0 (0–5) 3 (0–5) 5.5 (0–8) 8 (0–8)
CS, control group (n = 10); MOR, morphine group (n = 13); , receptor agonist group (DAMGO; n = 12); , receptor agonist group (U69593; n = 12). The data are expressed as median (range). * p < 0.05, compared with CS and groups. ** p < 0.05, compared with CS group (Kruskal–Wallis test followed by Dunn’s test).
M.O. Klein et al. / Behavioural Brain Research 274 (2014) 62–72
67
Fig. 4. Predatory hunting parameters in Experiment 2. CS, control group (n = 10); MOR, morphine sulfate group (n = 13); NM, receptor antagonist (CTAP) + morphine group (n = 11); NM, receptor antagonist (nor-binaltorphimine) + morphine group (n = 12). (A) Number of insects captured during the behavioral test. The data are expressed as mean ± SEM (one-way ANOVA followed by Tukey’s test). (B) Latency to capture the 1st, 2nd, 3rd, 4th, and 5th cockroaches. The data are expressed as mean ± SEM (Kruskal–Wallis test). (C) Percentage of dams that captured the 1st, 2nd, 3rd, 4th, and 5th cockroaches. *p < 0.05, compared with CS and MOR groups; # p < 0.05, compared with CS group; **p < 0.05, compared with MOR group (Fisher’s exact test).
Fig. 5. Maternal parameters in Experiment 2. CS, control group (n = 10); MOR, morphine sulfate group (n = 13); NM, receptor antagonist (CTAP) + morphine group (n = 11); NM, receptor antagonist (U69593) + morphine group (n = 12). (A) Latency to retrieve the 1st, 5th, and 8th pups after being transformed into rank-order data (1: 0–300 s; 2: 301–600 s; 3: 601–900 s; 4: 901–1200 s; 5: 1201–1500 s; 6: 1501–1800 s; 7: no retrieval during 30 min test). The data are expressed as mean ± SEM (Kruskal–Wallis test followed by Dunn’s test). (B) Percentage of dams that retrieved each of their 1st, 5th, and 8th pups. *p < 0.05, compared with control (CS) group (Fisher’s exact test).
68
M.O. Klein et al. / Behavioural Brain Research 274 (2014) 62–72
altered by the blockade of opioid receptors, we further investigated the endogenous tone of and receptors in Experiment 3.
Table 3 Maternal parameters analyzed during 1800 s in Experiment 2. Parameter
FMB (%) Nursing (%) Pup contact Licking (s)
Group CS
MOR
NM
NM
10.0 80.0 5.4 ± 1.2 120.8 ± 26.0
0.0 38.5 4.3 ± 0.9 44.1 ± 15.8
0.0 9.0* 2.09 ± 0.6 9.8 ± 5.07*
0.0 8.3* 5.1 ± 1.5 36.9 ± 15.1*
FMB, full maternal behavior; CS, control group (n = 10); MOR, morphine sulfate group (n = 13); NM, receptor antagonist (CTAP) + morphine group (n = 11); NM, receptor antagonist (U69593) + morphine group (n = 12). FMB and nursing behavior were analyzed using Fisher’s exact test. Pup contacts and licking behavior were analyzed using the Kruskal-Wallis test followed by Dunn’s test. The data are expressed as mean ± SEM. * p < 0.05, compared with CS group.
(p = 0.0041) and 8th (p = 0.0003) pups compared with controls. Less than 40% of the dams in the MOR, NM, and NM groups retrieved the pups, whereas at least 80% of the dams in the CS group retrieved them (p < 0.05 for all comparisons). The number of dams that displayed nursing and licking behaviors was less in the NM and NM groups compared with the CS group (Table 3; p < 0.01 for all comparisons). Because we found that predatory parameters were
3.3. Experiment 3: Blockade of and opioid receptors in rlPAG behavioral selection In this experiment, dams in the N group that received the opioid receptor antagonist into the rlPAG captured more cockroaches (F2,32 = 4.754, p = 0.0161) and had shorter latencies to capture each cockroach (p < 0.05 for all comparisons). Additionally, 100% of the N dams captured at least the 1st cockroach, whereas only 70% and 41% of the animals displayed such behavior in the control and N groups, respectively (p < 0.05; Fig. 6). With regard to the maternal parameters, the latency scores to retrieve the 1st, 5th, and 8th pups were longer in the N group compared with the CS group (Fig. 7A; p < 0.04 for all comparisons). The percentage of dams that retrieved each pup was lower in the N group (Fig. 7B; p < 0.04). Dams in the N group displayed less nursing behavior (p < 0.04) and spent less time licking their pups (F2,32 = 4.390, p = 0.0213; Table 4). Animals in Experiment 3 had different hunting strategies (Table 5). In the first 900 s of the test, N dams caught all of the
Fig. 6. Predatory hunting parameters in Experiment 3. CS, control group (n = 10); N, receptor antagonist group (CTAP; n = 12), N, receptor antagonist group (norbinaltorphimine; n = 11). (A) Number of insects captured during the behavioral test. The data are expressed as mean ± SEM (one-way ANOVA followed by Tukey’s test). (B) Latency to capture each cockroach. The data are expressed as mean ± SEM (Kruskal–Wallis test followed by Dunn’s test). (C) Percentage of dams that captured the 1st, 2nd, 3rd, 4th, and 5th cockroaches. *p < 0.05, compared with N group (Fisher’s exact test).
M.O. Klein et al. / Behavioural Brain Research 274 (2014) 62–72
69
Fig. 7. Maternal parameters in Experiment 3. CS, control group (n = 10); N, receptor antagonist group (CTAP; n = 12); N, receptor antagonist group (nor-binaltorphimine; n = 11). (A) Latency to retrieve the 1st, 5th, and 8th pups after being transformed into rank-order data (1: 0–300 s; 2: 301–600 s; 3: 601–900 s; 4: 901–1200 s; 5: 1201–1500 s; 6: 1501–1800 s; 7: no retrieval during 30 min test). The data are expressed as mean ± SEM (Kruskal–Wallis test followed by Dunn’s test). (B) Percentage of dams that retrieved each of their pups. *p < 0.05, compared with control (CS) group; **p < 0.05, compared with CS and N groups (Fisher’s exact test).
insects and retrieved most of the pups, whereas dams in the CS group first retrieved the pups and then captured the cockroaches. Most of the animals in the N group did not capture roaches (nine of 12 dams) or take care of pups (six of 12 dams). 3.4. Experiment 4: Opioids receptor activation in the rlPAG To evaluate the activation of and opioid receptors, we performed Experiment 4. The intensity of fluorescence (red for receptors and green for receptors) indicates the level of activation of these receptors caused by drug stimulation in each of the groups in Experiments 1–3 (Fig. 8A). The control group exhibited a basal tone of activation of both receptors, whereas morphine sulfate significantly activated both and receptors in the MOR group Table 4 Maternal parameters analyzed during 1800 s in Experiment 3.
FMB (%) Nursing (%) Pup contact Licking (s)
4. Discussion
Group
Parameter CS
N
N
20.0 70.0 5.9 ± 1.0 104.5 ± 21.4
0.0 16.7* 4.92 ± 1.1 34.5 ± 15.4#
27.3 63.6 4.8 ± 0.8 70.3 ± 13.2
FMB, full maternal behavior; CS, control group (n = 10); N, receptor antagonist group (CTAP; n = 12); N, receptor antagonist group (nor-binaltorphimine; n = 11). FMB and nursing behavior were analyzed using Fisher’s exact test. Pup contacts and licking behavior were analyzed using one-way ANOVA followed by Tukey’s test. The data are expressed as mean ± SEM. * p < 0.05, compared with CS and N groups. # p < 0.05, compared with control group.
Table 5 Insects captured and pups retrieved in the initial 900 s and during the entire behavioral test (1800 s) in Experiment 3. Parameter
Insects captured until 900 s Insects captured during 1800 s Pups retrieved until 900 s Pups retrieved during 1800 s
(: F3,17 = 12.18, p = 0.0003, Fig. 8B; : F3,17 = 3.68, p = 0.04, Fig. 8E). Although they received a specific agonist, the and groups did not exhibit significant levels of opioid receptor activation. In the groups in Experiment 2 (Fig. 8C and F), the injection of a single opioid receptor antagonist in the presence of morphine (i.e., in the NM and NM groups) reversed the overactivation caused by morphine alone ( receptor: F3,17 = 8.487, p = 0.0018; receptor: F3,16 = 4.457, p = 0.023). Fig. 8D and G show the groups in Experiment 3. Activation of the opioid receptor in the N group was not significantly different from the CS group, but receptor activation in the N group was higher than in the CS and N groups (140.4% ± 21.8% of control [mean ± SEM], F2,10 = 4.601, p = 0.0468). With regard to opioid receptor activation in the groups in Experiment 3 (Fig. 8G), no significant differences were found among groups (F2,10 = 0.5105, p > 0.05).
Group CS
N
N
1 (0–5) 3.5 (0–5) 8 (0–8) 8 (7–8)
0 (0–5) 0 (0–5) 0.5 (0–8)# 0.5 (0–8)#
5 (0–5)* 5 (1–5)* 6 (0–8) 7 (0–8)
CS, control group (n = 10); N, receptor antagonist group (CTAP; n = 12); N, receptor antagonist group (nor-binaltorphimine; n = 11). The data are expressed as median (range). * p < 0.03, compared with N group. # p < 0.01, compared with CS group (Kruskal–Wallis test followed by Dunn’s test).
The PAG is a putative center that switches adaptive behavioral responses [26]. The rlPAG plays an important role in switching from maternal to predatory behavior in lactating rats [11]. Morphine injected directly into this area impairs pup retrieval, nest building, and FMB in mothers [18]. In the present study, we tested the role of morphine injected into the rlPAG in the context of selecting maternal vs. predatory behaviors. We also analyzed the functional relevance of and opioid receptors in this paradigm. We injected and opioid receptor agonists in the rlPAG in dams in Experiment 1. Unlike what happens when injected peripherally [13], morphine administered centrally in the rlPAG disrupted maternal behavior but did not increase predatory hunting in the dams. Thus, peripheral injections of morphine may cause the activation of opioid receptors in other areas together with the PAG. This appears to be necessary to generate morphine-induced increases in predatory hunting. Other areas, such as the ventrolateral caudate putamen, amygdala, hypothalamus, and superior colliculus, play a role in predatory hunting, the motivational drive to forage and hunt, and opioid receptor expression [8,9,19,27–29]. Lactating females appear to be more efficient in hunting and catching prey compared with pregnant and virgin females [4]. Lactating females in the CS group had a greater level of expression of hunting (i.e., beyond caring for pups, females in the CS group also foraged and captured prey). Interestingly, CS females retrieved their pups first and hunted for insects afterward. This did not occur in the other groups, mainly in females who received a morphine injection. These animals displayed only hunting behavior.
70
M.O. Klein et al. / Behavioural Brain Research 274 (2014) 62–72
Fig. 8. Opioid receptor activation. (Left) (A) Fluorescent photomicrograph of the periaqueductal gray in Experiment 4. The intensity of fluorescence indicates the activation of each opioid receptor subtype ( and ) in the tissue. Scale bar = 150 m. (Right) Activation of and opioid receptors in the dams after the behavioral test. For presentation purposes, the data are shown as a percentage of activation relative to the control (CS) group (n = 5). Just the groups within each Experiment (1–3) were compared to each other for each receptor. MOR (morphine group; n = 4), ( opioid receptor agonist [DAMGO] group; n = 4), ( opioid receptor agonist [U69593] group; n = 5), NM ( opioid receptor antagonist [CTAP] with consecutive morphine injection; NM (n = 5): opioid receptor antagonist [nor-binaltorphimine] plus morphine; n = 4), N ( opioid receptor antagonist [CTAP]; n = 4), N ( opioid receptor antagonist [nor-binaltorphimine] group; n = 4). The red line indicates the CS group. (B) Activation of opioid receptors in Experiment 1. ***p < 0.001, compared with CS, , and groups (one-way ANOVA followed by Tukey’s test). (C) Activation of opioid receptors in Experiment 2. **p < 0.01, compared with CS and NM groups (one-way ANOVA followed by Tukey’s test). (D) Activation of opioid receptors in Experiment 3. *p < 0.05, compared with CS and N groups (one-way ANOVA followed by unpaired t-test). (E) Activation of opioid receptors in Experiment 1. # p < 0.05, compared with CS group (one-way ANOVA followed by Tukey’s test). (F) Activation of opioid receptors in Experiment 2. # p < 0.05, compared with CS and NM groups (one-way ANOVA followed by Tukey’s test). (G) Activation of opioid receptors in Experiment 3 (one-way ANOVA).
Intracerebroventricular injection of the opioid agonist is as effective as morphine or -endorphin in inhibiting maternal behavior [15,20]. In the present study, the same opioid agonist injected into the rlPAG did not interfere either with maternal behavior or predatory hunting. Thus, the activation of opioid receptors in other areas beyond the rlPAG might be necessary to inhibit maternal behavior, or the dose used in this study was not enough even being a high dose [20]. Experiment 4 showed that intrarlPAG injection of the opioid agonist did not significantly activate this receptor. This finding may be related to insufficient receptor activation to promote behavioral changes in this paradigm. Similarly, peripheral injections of the agonist increased the latency to retrieve the pups [21,30]. However, intracerebroventricular injection of the agonist did not have any effect on maternal behavior [20]. In the present study, the opioid receptor agonist injected directly into the rlPAG also did not interfere with maternal behavior or behavioral selection. Similarly, a single injection of the agonist did not activate a significant number of receptors in Experiment 4. Thus, at the tested doses, these opioid agonists might be ineffective in activating their respective receptors in the rlPAG. Beyond that, the peripheral injection of the specific and agonists may act in other areas of the brain, which together can impair the
maternal behavior. More studies needs to be performed to clarify this issue. In Experiment 1, a morphine injection into the rlPAG alone inhibited maternal behavior. Such a behavioral effect may be attributable to morphine’s action on more than one type of opioid receptor simultaneously. Furthermore, it is important to state that in this study, the role of ␦ opioid receptor was not investigated, and morphine, even in a minor scale, acts in this type of receptor. Therefore, ␦ receptor in rlPAG could also plays a role in modulating morphine inhibition of maternal behavior and behavioral selection in all experimental paradigms presented [31]. To indirectly address the possible specific functional meaning of the morphine-induced activation of and opioid receptor, we specifically blocked and receptors in Experiment 2. All of the morphine-treated groups exhibited impairments in maternal behavior, even with and receptors blocked. This supports the hypothesis that morphine may act on more than one type of opioid receptor to inhibit maternal behavior. The presence of the receptor antagonist in the rlPAG concurrently with morphine inhibited predatory hunting, whereas the presence of the receptor antagonist with morphine or morphine alone did not alter this behavior. Therefore, opioid receptors in the rlPAG appear to have a central function in the
M.O. Klein et al. / Behavioural Brain Research 274 (2014) 62–72
expression of predatory hunting. This finding led us to investigate the role of endogenous and opioid receptor tone in behavioral selection in Experiment 3. The injection of the receptor antagonist did not significantly interfere with predatory behavior but impaired the care of pups. This suggests that the basal tone of receptors is important for the regular expression of maternal behavior. Other studies have shown that diminishing or increasing receptor activation inhibits maternal behavior [13,32,33]. The blockade of receptors facilitated predatory behavior without interfering with parameters of maternal behavior. Receptor activation may restrain the expression of predatory hunting, and the blockade of receptors may facilitate this behavior. Experiment 4 failed to show significant effects of and receptor antagonists on the endogenous activation of their specific receptors. Since the specificity of these antagonists is well established [34–36], these findings may be due to the fact that the basal activation was so low in the present experimental situation that made it difficult to decrease to even lower levels. However, receptor blockade increased the activation of opioid receptors in the rlPAG in Experiment 4. This finding suggests the existence of an important functional balance between these two receptor subtypes that may promote predatory hunting, i.e., once the action of one type of receptor is diminished, the activity of the other one would increase to compensate the lack of action from the first. Thus, the activity of opioid receptors appears to be affected by inhibitory endogenous receptor tone. Some important general aspects of the behavioral tests in Experiments 1–3 should be considered. Previous studies have shown that central injections of morphine impair pup retrieval, nest building, and the expression of FMB [18]. In the present study, nest building was not evaluated because of the absence of pine flakes in the experimental cages. Behavior was tested without pine flakes to better analyze predatory behavior (i.e., the visualization of cockroaches would be hampered by pine flakes in the cages). Limited access to nest material is a well-established model of neglected maternal behavior [37,38]. This may explain why the percentage of dams that expressed FMB and nursing in all of the groups, including the CS group, was lower in the present study compared with other studies that utilized free access to nest material [13]. Because this test was the first time that the dams had contact with roaches, the novelty of the insects may have indirectly influenced the expression of FMB. Changes in the time taken to begin catching the cockroaches and latency scores to retrieve the 1st pup might reflect the motivational drive of the dams and may be useful to evaluate this drive [11]. The rlPAG is an integrative region that plays an important role in motivational and rewarded behaviors (e.g., hunting, foraging, and reward seeking) [29,39]. This feature may have influenced the present results. The central modulation exerted by the drugs may have influenced the motivation to elicit behavioral responses. Interestingly, beyond interfering with the retrieve of pups, drugs also impaired the nursing and licking behavior. It shows that in addition to interfere with the motivation to care pups, they also disrupted the maintenance of this behavior, mainly regarding to opioid receptor. The drug-induced behavioral changes reported herein might be attributable to changes in the activational state of PAG opioid receptors. The level of activation of and opioid receptors in the rlPAG was measured in all of the experimental paradigms. As expected, morphine increased the activation of both and receptors. However, the blockade of one of these receptors in the presence of morphine decreased the activation caused by morphine alone, restoring it to control levels. Interestingly, treatment with and receptor agonists did not result in differences in and receptor activation compared with controls. Indeed, the and groups showed no significant changes in behavioral parameters.
71
This may be explained by an inability of the agonists to activate a significant number of receptors that would be sufficient to influence behavior in this paradigm. Also, as previously said, the doses used might be not enough to show any behavioral or molecular alteration. In Experiment 3, the N group, which received a receptor antagonist, exhibited an increase in the activation of opioid receptors. This group also presented an improvement in predatory hunting, whereas receptor blockade in the N group impaired the expression of predatory behavior. This may be attributable to a functional balance between opioid receptors, in which the suppression of receptors, directly or indirectly, may facilitate receptor activation, always keeping a basal tone of receptors activation. This suggests the existence of a dynamic interaction among opioid receptors that may involve other central systems. Such a system may compensate for the lack of action of one receptor by increasing the activation of another. This mechanism might play a role in the behavioral changes observed in the present study. 5. Conclusion Our data suggest that endogenous opioid receptor tone in the rlPAG is essential for the expression of hunting behavior and can be modulated by the degree of receptor activation. Thus, receptor-mediated opioidergic transmission appears to be critical for behavioral selection during lactation. The pharmacological blockade of opioid receptors leads to a behaviorally meaningful increase in endogenous receptor activation, which can be directly or indirectly mediated. A role for compensatory opioidergic multireceptor interactions that determine behavioral responses now appears likely. Conflict of interest The authors have no conflict of interest to report. Funding source This work was supported by grant from Fundac¸ão de Amparo à Pesquisa do Estado de São Paulo (FAPESP) processes 2010/06774-0 and 2013/01610-7. Funding source played no role in experimental design or decision to submit the paper for publication. Acknowledgment The authors thank to Miriam Aline Geigner for her technical assistance in the immunofluorescence assay. References [1] Blaffer-Hrdy S. Mother nature: maternal instincts and how they shape the human species. New York, NY: Ballentine; 1999. [2] Felicio LF, Canteras NS. Maternal choices: neural mediation: caring for young or hunting. In: Bridges RS, editor. Neurobiology of the parental brain. New York: Elsevier; 2008. p. 75–82. [3] Cruz Ade M, Maiorka PC, Canteras NS, Sukikara MH, Felicio LF. Morphine treatment during pregnancy modulates behavioral selection in lactating rats. Physiol Behav 2010;101:40–4. [4] Kinsley CH, Amory-Meyer E. Why the maternal brain? J Neuroendocrinol 2011;23:974–83. [5] Teodorov E, Felicio LF, Bernardi MM. Maternal behavior. In: Andersen ML, Tufik S, editors. Animal models as tools in ethical biomedical research. São Paulo: UNIFESP; 2010. p. 149–65. [6] Numan M, Fleming AS, Levy F. Maternal behavior. In: Neil JD, editor. Knobil and Neill’s physiology of reproduction. Amsterdam: Elsevier; 2006. p. 1921–93. [7] Comoli E, Ribeiro-Barbosa ER, Canteras NS. Predatory hunting and exposure to a live predator induce opposite patterns of Fos immunoreactivity in the PAG. Behav Brain Res 2003;138:17–28. [8] Comoli E, Ribeiro-Barbosa ER, Negrao N, Goto M, Canteras NS. Functional mapping of the prosencephalic systems involved in organizing predatory behavior in rats. Neuroscience 2005;130:1055–67.
72
M.O. Klein et al. / Behavioural Brain Research 274 (2014) 62–72
[9] Furigo IC, de Oliveira WF, de Oliveira AR, Comoli E, Baldo MV, Mota-Ortiz SR, et al. The role of the superior colliculus in predatory hunting. Neuroscience 2010;165:1–15. [10] Miranda-Paiva CM, Ribeiro-Barbosa ER, Canteras NS, Felicio LF. A role for the periaqueductal grey in opioidergic inhibition of maternal behaviour. Eur J Neurosci 2003;18:667–74. [11] Sukikara MH, Mota-Ortiz SR, Baldo MV, Felicio LF, Canteras NS. A role for the periaqueductal gray in switching adaptive behavioral responses. J Neurosci 2006;26:2583–9. [12] Behbehani MM. Functional characteristics of the midbrain periaqueductal gray. Prog Neurobiol 1995;46:575–605. [13] Sukikara MH, Platero MD, Canteras NS, Felicio LF. Opiate regulation of behavioral selection during lactation. Pharmacol Biochem Behav 2007;87:315–20. [14] Bridges RS, Grimm CT. Reversal of morphine disruption of maternal behavior by concurrent treatment with the opiate antagonist naloxone. Science 1982;218:166–8. [15] Felicio LF, Mann PE, Bridges RS. Intracerebroventricular cholecystokinin infusions block beta-endorphin-induced disruption of maternal behavior. Pharmacol Biochem Behav 1991;39:201–4. [16] Grimm CT, Bridges RS. Opiate regulation of maternal behavior in the rat. Pharmacol Biochem Behav 1983;19:609–16. [17] Martin WR. Pharmacology of opioids. Pharmacol Rev 1984;35:283–323. [18] Moura LM, Canteras NS, Sukikara MH, Felicio LF. Morphine infusions into the rostrolateral periaqueductal gray affect maternal behaviors. Braz J Med Biol Res 2010;43:899–905. [19] Mansour A, Fox CA, Burke S, Meng F, Thompson RC, Akil H, et al. Mu, delta, and kappa opioid receptor mRNA expression in the rat CNS: an in situ hybridization study. J Compar Neurol 1994;350:412–38. [20] Mann PE, Kinsley CH, Bridges RS. Opioid receptor subtype involvement in maternal behavior in lactating rats. Neuroendocrinology 1991;53:487–92. [21] Teodorov E, Tomita AT, Banon GP, Gil IG, Bernardi MM, Felicio LF. Behavioral effects of acute stimulation of -opioid receptors during lactation. Pharmacol Biochem Behav 2008;90:534–9. [22] Paxinos G, Watson C. The rat brain in stereotaxic coordinates. 4th edition San Diego, CA: Academic Press; 1998. [23] Gupta A, Decaillot FM, Gomes I, Tkalych O, Heimann AS, Ferro ES, et al. Conformation state-sensitive antibodies to G-protein-coupled receptors. J Biol Chem 2007;282:5116–24. [24] Machado FC, Zambelli VO, Fernandes ACO, Heimann AS, Cury Y, Picolo G. Peripheral interactions between cannabinoid and opioid systems contribute to the antinociceptive effectof crotalphine. Br J Pharmacol 2014;171(4):961–72, http://dx.doi.org/10.1111/bph.12488. [25] Nasello AG, Depiante R, Tannhauser M, Izquierdo I. Effect of morphine on the RNA and ATP concentration of brain structures of the rat. Pharmacology 1973;10:56–9.
[26] Sukikara MH, Mota-Ortiz SR, Baldo MV, Felicio LF, Canteras NS. The periaqueductal gray and its potential role in maternal behavior inhibition in response to predatory threats. Behav Brain Res 2010;209:226–33. [27] dos Santos LM, Ferro MM, Mota-Ortiz SR, Baldo MV, da Cunha C, Canteras NS. Effects of ventrolateral striatal inactivation on predatory hunting. Physiol Behav 2007;90:669–73. [28] dos Santos LM, Boschen SL, Bortolanza M, de Oliveira WF, Furigo IC, Mota-Ortiz SR, et al. The role of the ventrolateral caudoputamen in predatory hunting. Physiol Behav 2012;105:893–8. [29] Mota-Ortiz SR, Sukikara MH, Felicio LF, Canteras NS. Afferent connections to the rostrolateral part of the periaqueductal gray: a critical region influencing the motivation drive to hunt and forage. Neural Plast 2009;2009:612698. [30] Teodorov E, Ferrari MF, Fior-Chadi DR, Camarini R, Felicio LF. Behavioral meaningful opioidergic stimulation activates kappa receptor gene expression. Braz J Med Biol Res 2012;45:982–7. [31] Gendron L, Mittal N, Beaudry H, Walwyn W. Recent advances on the delta opioid receptor: from trafficking to function. Br J Pharmacol 2014;26, http://dx.doi.org/10.1111/bph.12706. [32] Byrnes EM, Rigero BA, Bridges RS. Opioid receptor antagonism during early lactation results in the increased duration of nursing bouts. Physiol Behav 2000;70:211–6. [33] Mann PE, Pasternak GW, Bridges RS. Mu1 opioid receptor involviment in maternal behavior. Physiol Behav 1990;47:133–8. [34] Kramer TH, Shook JE, Kazmierski W, Ayres EA, Wire WS, Hruby VJ, et al. Novel peptidic mu opioid antagonists: pharmacologic characterization in vitro and in vivo. J Pharmacol Exp Ther 1989;249(2):544–51. [35] Abbruscato TJ, Thomas SA, Hruby VJ, Davis TP. Blood–brain barrier permeability and bioavailability of a highly potent and mu-selective opioid receptor antagonist, CTAP: comparison with morphine. J Pharmacol Exp Ther 1997;280(1):402–9. [36] Portoghese PS, Lin CE, Farouz-Grant F, Takemori AE. Structure–activity relationship of N17 -substituted norbinaltorphimine congeners. Role of the N17 basic group in the interaction with a putative address subsite on the kappa opioid receptor. J Med Chem 1994;37(10):1495–500. [37] Dalle Molle R, Portella AK, Goldani MZ, Kapczinski FP, Leistner-Segal S, Salum GA, et al. Associations between parenting behavior and anxiety in a rodent model and a clinical sample: relationship to peripheral BDNF levels. Trans Psychiatry 2012;2:e195. [38] Ivy AS, Brunson KL, Sandman C, Baram TZ. Dysfunctional nurturing behavior in rat dams with limited access to nesting material: a clinically relevant model for early-life stress. Neuroscience 2008;154:1132–42. [39] Mota-Ortiz SR, Sukikara MH, Bittencourt JC, Baldo MV, Elias CF, Felicio LF, et al. The periaqueductal gray as a critical site to mediate reward seeking during predatory hunting. Behav Brain Res 2012;226:32–40.
Contents lists available at ScienceDirect
Behavioural Brain Research journal homepage: www.elsevier.com/locate/bbr
Research report
Periaqueductal gray and opioid receptors determine behavioral selection from maternal to predatory behavior in lactating rats Marianne Orlandini Klein a,b , Aline de Mello Cruz a , Franciele Corrêa Machado b,d , Gisele Picolo d , Newton Sabino Canteras c , Luciano Freitas Felicio a,∗ a Department of Pathology, School of Veterinary Medicine, University of São Paulo, Avenida Professor Doutor Orlando Marques de Paiva, 87, São Paulo, SP CEP 05508-270, Brazil b Departments of Pharmacology, Institute of Biomedical Sciences, University of São Paulo, Avenida Prof. Lineu Prestes, 1524, Biomédicas I, São Pauo, SP CEP 05508-900, Brazil c Department of Anatomy, Institute of Biomedical Sciences, University of São Paulo, 2415, São Paulo, SP CEP 05508-900, Brazil d Special Laboratory of Pain and Signaling, Butantan Institute, Avenida Vital Brazil, 1500, São Paulo CEP 05503-900, Brazil
h i g h l i g h t s • • • •
The rlPAG and opioid receptors have a role in behavioral selection. Morphine in the rlPAG inhibits maternal behavior without interfering with hunting. Receptor blockade increases hunting and increases receptor activation. Multiple opioid receptor interactions play a role in behavioral selection.
a r t i c l e
i n f o
Article history: Received 17 April 2014 Received in revised form 21 July 2014 Accepted 4 August 2014 Available online 10 August 2014 Keywords: Behavioral selection Maternal behavior Predatory hunting Periaqueductal gray
a b s t r a c t Every mother must optimize her time between caring for her young and her subsistence. The rostro lateral portion of the periaqueductal grey (rlPAG) is a critical site that modulates the switch between maternal and predatory behavior. Opioids play multiple roles in both maternal behavior and this switching process. The present study used a pharmacological approach to evaluate the functional role of rlPAG and opioid receptors in behavioral selection. Rat dams were implanted with a guide cannula in the rlPAG and divided into three experiments in which we tested the role of opioid agonists (Experiment 1), the influence of and opioid receptor blockade in the presence of morphine (Experiment 2), and the influence of and opioid receptor blockade (Experiment 3). After behavioral test, in Experiment 4, we evaluated rlPAG and receptor activation in all Experiments 1–3. The results showed that massive opioidergic activation induced by morphine in the rlPAG inhibited maternal behavior without interfering with predatory hunting. No behavioral changes and no receptor activation were promoted by the specific agonist alone. However, receptor blockade increased hunting behavior and increased the level of receptor activation in the rlPAG. Thus, endogenous opioidergic tone might be modulated by a functional interaction between opioid receptor subtypes. Such a compensatory receptor interaction appears to be relevant for behavioral selection among motivated behaviors. These findings indicate a role for multiple opioid receptor interactions in the modulation of behavioral selection between maternal and predatory behaviors in the PAG. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Lactating mammalian mothers must optimize the constant exchange between reproduction and subsistence [1,2]. Indeed, a
∗ Corresponding author. Tel.: +55 11 3091 7934; fax: +55 11 3091 7829. E-mail address: [email protected] (L.F. Felicio). http://dx.doi.org/10.1016/j.bbr.2014.08.008 0166-4328/© 2014 Elsevier B.V. All rights reserved.
mother has to use her time more efficiently than a virgin female to express maternal and predatory behaviors, which has adaptive value [3–5]. In mammals, maternal behavior is fundamental for perpetuation of the species, in which the neonate’s survival depends on the mother and her ability to provide food, heat, shelter, and protection [6]. Predatory behavior is also essential for individual survival, and it is characteristic of the specie. During the post-partum period,
M.O. Klein et al. / Behavioural Brain Research 274 (2014) 62–72
the time spent foraging and hunting depends on environmental circumstances [2]. As a stimulus to elicit predatory behavior, we used cockroaches, which appear to be suitable prey and have an inherent hedonic value for the rat [7–11]. The periaqueductal gray (PAG) is an important region of the central nervous system that is involved in multiple behavioral and physiological processes, including nociception, fear, anxiety, cardiovascular control, sexual behavior, and vocalization [12]. Beyond these functions, previous studies have demonstrated that the PAG, mainly its rostrolateral portion (rlPAG), also plays an important role in maternal and predatory behavioral selection, and opioids are involved in this switch between behavioral modes [2,3,11,13]. Opioids are well established to be involved in the regulation of maternal behavior [14–16]. Morphine is a nonselective opioid agonist that acts preferentially at opioid receptors [17]. A single injection of morphine directly into the rlPAG impairs maternal behavior [18]. Nevertheless, the specific effects of a morphine injection in the rlPAG on predatory hunting vs. maternal behavior have not been investigated previously. Different types of opioid receptors have been described, and the and subtypes are the most widely expressed and functionally relevant [19]. Nevertheless, unknown is the functional relevance of these receptors in the rlPAG for behavioral selection during the post-partum period. An intracerebroventricular injection of a specific agonist in lactating females dose-dependently disrupted maternal behavior [20]. Acute peripheral injections of a specific opioid agonist increased the latency to retrieve pups [21]. In the present study, lactating females were tested for behavioral selection using various pharmacological challenges that involved both stimulation and blockade of rlPAG and opioid receptors. The state of receptor activation was assessed by an immunological assay. Our results showed that the central functional interaction between both receptors might be important for maintaining the normal expression of predatory behavior, and morphine injected directly into the rlPAG was shown to act on both receptors to inhibit maternal behavior. 2. Materials and methods 2.1. Animals Female nulliparous Wistar rats were ∼90 days of age at the beginning of the experiments. They were mated with sexually experienced males, and the day that sperm was observed in the vaginal lavage was considered day 1 of pregnancy. Pregnant females were individually housed in polypropylene cages (30 cm × 40 cm × 18 cm) with pine flakes. They had free access to water and food during all of the experiments. The light/dark cycle (12 h/12 h) and temperature (23 ± 2 ◦ C) were controlled. On day 15–17 of pregnancy, females underwent stereotaxic surgery to implant a unilateral guide cannula in the rlPAG. After giving birth (day 0 of lactation), the dams remained with their litter, which was culled to eight pups (four males and four females) on day 2 of lactation until the day of the behavioral test. All of the procedures were in accordance with the Ethical Principles of Animal Experimentation adopted by the Sociedade Brasileira de Ciências em Animais de Laboratório (SBCAL) and approved by the Comissão de Ética no Uso de Animais (CEUA) of the Biomedical Sciences Institute and School of Veterinary Medicine at the University of São Paulo. 2.2. Stereotaxic surgery On day 15–17 of pregnancy, the females were intraperitoneally anesthetized with a mixture of ketamine (60 mg/kg Vetanarcol, Santana de Parnaíba, SP, Brazil) and xylazine (5 mg/kg Kensol,
63
Santana de Parnaíba, SP, Brazil) and received 3 mg/kg dipyrone (Ibasa, Porto Alegre, RS, Brazil) for postsurgical analgesia. The females were then placed in a stereotaxic apparatus to implant a guide cannula (Plastic One, Roanoke, VA, USA) in the rlPAG. The stereotaxic coordinates were −6.0 mm from bregma, −0.6 mm lateral to the midline, and 4.2 mm ventral to the surface of the brain [22]. On the test day, the volume of drug or saline (0.6 l) was injected over 30 s, and the internal cannula was left in place for an additional 30 s to allow for diffusion. The internal cannula was 1 mm longer than the guide cannula, allowing precise injection into the rlPAG. Only animals with correct cannula placements were used in the study. 2.3. Behavioral analyses Behavioral test was not performed in the home cage of animals. The test cage was constructed of transparent Plexiglas to allow complete visualization of the pups and cockroaches and prevent the insects from escaping. To permit complete observation of both the insects and mammals’ behaviors, no pine flakes were used in these cages. One day before the tests, the dams were habituated to the test cage. Females without their pups were habituated to exploring the experimental cage for 30 min. The bottom of the experimental cage was divided by two perpendicular lines. Each time the animal crossed a line, it was counted as an indication of locomotion. Behavioral tests were performed on day 5 or 6 of lactation. The animals were tested for behavioral selection (i.e., caring for pups or predatory hunting) and were not food deprived before the tests. On the test day, 60 min before the test, lactating females were placed in the test cage without their pups. Thirty minutes later, the dams received a drug/saline injection, and the behavioral test began 30 min after the injection. Eight pups and five mature cockroaches (Leurolestes circunvagans) were dispersed introduced to the cage at the beginning of the behavioral test. All of the trials were recorded for subsequent analysis. During the 30 min trial, the following maternal behavior parameters were analyzed: latency to retrieve the 1st, 5th, and 8th pups, the percentage of dams that retrieved these pups, the number of contacts with the pups (i.e., when lactating females only touched them without retrieving them), the number of times the pups were licked, the percentage of dams that nursed their pups, and the expression of full maternal behavior (FMB), in which the dam, after retrieving and grouping at least five pups, arched over them with her legs splayed, with the pups attached to the nipples, and the dam remained in this position for at least three consecutive minutes. For the predatory hunting analysis, we recorded the latency to capture each cockroach, percentage of dams that captured the 1st, 2nd, 3rd, 4th, and 5th insects, and total number of insects hunted. General parameters, such as the number of lines crossed, time spent exploring the cage, and self-grooming time, were also analyzed. In all of the experiments, the experimental cage was cleaned with a 5% alcohol solution before each behavioral test to eliminate possible bias caused by odors left by previous animals. Experimental and control observations were intermixed to minimize possible circadian influences on the dams’ behavior. 2.4. Experiment 1 The goal of this experiment was to determine whether injections of morphine and the and opioid agonists in the rlPAG interfere with maternal behavioral selection and predatory hunting. Behavioral analyses were performed in four different groups: morphine (MOR; n = 13; 7.91 nmol morphine sulfate; Dimorf, São Paulo, SP, Brazil), opioid receptor agonist (; n = 12; 10 nmol DAMGO; Sigma-Aldrich, St Louis, MO, USA), opioid receptor agonist (; n = 12; 10 nmol U69593; Tocris Bioscience, Bristol, UK), and
64
M.O. Klein et al. / Behavioural Brain Research 274 (2014) 62–72
Fig. 1. Injection loci. (A) Photomicrograph of the periaqueductal gray (PAG) with Nissl staining. (B) Diagram that shows the main injection loci in each group. Because of overlap, the number of illustrated points may be less than the actual number of injections. The injections were made unilaterally. Each figure represents the values of PAG regions posterior to bregma (−5.8 mm, −6.04 mm, −6.3 mm, −6.72 mm, and −6.80 mm). Only animals that had correct injection sites were considered in this study. Aq, aqueduct; EW, Edinger–Westphal nucleus. Adapted from reference [22].
control (CS; n = 10; 0.9% physiological saline) [20]. The injections were performed as described above. The dose of morphine sulfate was not equimolar to the other drug doses because of technical issues. Morphine is already sold diluted; thus, an equimolar morphine dose would imply a drug volume that is too large to be injected in the animal brain [18]. 2.5. Experiment 2 In this experiment, we tested the importance of and opioid receptors in morphine’s action. Four groups were formed: morphine (MOR; n = 13; 7.91 nmol morphine sulfate; Dimorf, São Paulo, SP, Brazil), opioid receptor antagonist plus morphine (NM; n = 11; 10 nmol of the opioid receptor antagonist CTAP [Tocris Bioscience, Bristol, UK] plus 7.91 nmol morphine sulfate), opioid receptor antagonist plus morphine (NM; n = 12; 10 nmol of the opioid antagonist nor-binaltorphimine dihydrochloride (Tocris Bioscience, Bristol, UK) plus 7.91 nmol morphine sulfate), and control (CS; n = 10; which received 0.9% physiological saline) [20]. The injections and behavioral analyses were performed as described above. For the groups that was necessary the injection of two drugs, first it was administered the antagonist drug and 30 s after, the morphine. 2.6. Experiment 3 Considering the hypothesis that opioid receptor subtypes play differential roles in behavioral selection, this experiment was performed to investigate the role of endogenous and opioid receptor subtypes in behavioral selection using their respective antagonists. Three different groups were tested: opioid receptor
antagonist (N; n = 12; 10 nmol of the opioid receptor antagonist CTAP), opioid antagonist (N; n = 11; 10 nmol of the opioid receptor antagonist nor-binaltorphimine dihydrochloride), and control (CS; n = 10; which received 0.9% physiological saline). All of the behavioral analyses and drug injections were performed as described above.
2.7. Experiment 4 This experiment was conducted to evaluate the activation of and opioid receptors under the conditions of the previous experiments, i.e., each group for each experiment (1–3) as previously reported was submitted to an immunofluorescence assay after the behavioral test to evaluate the activation of and receptors under drugs action in rlPAG. Thus, conformation state-sensitive anti- and anti- antibodies were used. When the receptors are activated, they undergo a conformational change that is detected by these antibodies, thus recognizing the activated state of the receptor. It is a well established and validated technique [23,24]. To perform this experiment, 60 min after the end of the behavioral test, the dams were anesthetized with 70 mg/kg ketamine (Vetanarcol, Santana de Parnaíba, SP, Brazil) and 10 mg/kg xylazine (Kensol, Santana de Parnaíba, SP, Brazil) and transcardially perfused with a solution of 4.0% formaldehyde and sodium tetraborate, pH 7.4. The brains were removed and left overnight in a solution of 20% sucrose in 0.1 M phosphate buffer at 4 ◦ C. Frozen brain sections (30 m) of the rlPAG were cut with a sliding microtome in the frontal plane. The sections were mounted on gelatin-coated slides and blocked with 1% bovine serum albumin + 5% sucrose and processed for immunostaining. Tissue samples were incubated overnight with conformation state-sensitive anti- and anti- primary antibodies
M.O. Klein et al. / Behavioural Brain Research 274 (2014) 62–72
65
Fig. 2. Predatory hunting parameters in Experiment 1. CS, control group (n = 10); MOR, morphine group (n = 13); , receptor agonist group (DAMGO; n = 12); , receptor agonist group (U69593; n = 12). (A) Number of insects captured during the behavioral test. The data are expressed as mean ± SEM (one-way ANOVA). (B) Latency to capture the 1st, 2nd, 3rd, 4th, and 5th cockroaches. The data are expressed as mean ± SEM (Kruskal–Wallis test). (C) Percentage of dams that captured the 1st, 2nd, 3rd, 4th, and 5th cockroaches. p > 0.05 (Fisher’s exact test).
(Proteimax Biotechnology, Cotia, SP, Brazil) labeled with fluorescent Alexa Fluor dye (682 nm for opioid receptors and 800 nm for opioid receptors) diluted 1:4000. The Odyssey system (LiCOR) was used to analyze and quantify the intensity of fluorescence. Differences in receptor activation were evaluated by considering, beyond the tissue area, the total number of each receptor expressed in the same brain tissue using opioid receptor (N-terminus) and opioid receptor-1 (N-19) rabbit polyclonal immunoglobulin G diluted 1:100 (Santa Cruz Biotechnology, Santa Cruz, CA, USA). 2.8. Statistical analysis Because of the number of mothers that did not retrieve their pups, we expressed the latency to retrieve each pup as the following latency scores: 1 (latency = 0–300 s), 2 (latency = 301–600 s), 3 (latency = 601–900 s), 4 (latency: 901–1200 s), 5 (latency = 1201–1500 s), 6 (latency: 1501–1800 s), and 7 (when the dams do not retrieve the pups during the 1800 s of the behavioral test). We were thus able to include all of the dams in the statistical analysis. One-way analysis of variance (ANOVA) was used for predatory parameters (i.e., the number of cockroaches captured and latency to capture each cockroach), general parameters (i.e., the time exploring the cage, number of lines crossed in the cage, and self-grooming time), and maternal behavior of licking the pups. The Kruskal–Wallis test was used
for pup retrieval latency scores. Fisher’s exact test was used to analyze the percentage parameters, such as the dams that captured cockroaches, dams that retrieved pups, dams that expressed nursing behavior, and dams that expressed FMB. In Experiment 4, a one-way ANOVA was used to do a comparison of the groups of each experiment (1–3; a group of Experiment 1 was not compared to a group of Experiment 3, for example; and the activation of and receptors were not compared to each other). Data are expressed as a percentage of control activation, i.e. the closer the activation is of 100% more similar to control’s activation it is. Statistical significance was set at p < 0.05 for all the data. The statistical analyses were performed using GraphPad Prism 5.0. 3. Results For all of the experiments, only the animals that had the guide cannula implanted correctly into the rlPAG were considered (Fig. 1). Dams did not exhibit any difference in the number of lines crossed, time spent exploring the cage, and self-grooming time in any of the behavioral experiments, indicating that the drugs did not impair general parameters (p > 0.05, data not shown). Throughout the observation period, the dams in the MOR group and other groups that received morphine or specific opioid drug injections in Experiments 2 and 3 did not show any stereotypical behavior, motor rigidity, or signs of sedation that morphine may cause at
66
M.O. Klein et al. / Behavioural Brain Research 274 (2014) 62–72
Fig. 3. Maternal parameters in Experiment 1. CS, control group (n = 10); MOR, morphine group (n = 13); , receptor agonist group (DAMGO; n = 12); , receptor agonist group (U69593; n = 12). (A) Latency to retrieve the 1st, 5th, and 8th pups after being transformed into rank-order data (1: 0–300 s; 2: 301–600 s; 3: 601–900 s; 4: 901–1200 s; 5: 1201–1500 s; 6: 1501–1800 s; 7: no retrieval during 30 min test). The data are expressed as mean ± SEM (Kruskal–Wallis test followed by Dunn’s test). (B) Percentage of dams that retrieved each of their pups. *p < 0.05, compared with control (CS) and groups; # p < 0.05, compared with CS group (Fisher’s exact test).
higher doses [25]. Animals that expressed hunting behavior sniffed around the cage, some more vigorously than others, depending on the group, foraged the prey normally, and attempted to seize the prey with apparently no difficulty. These results indicate that the animals exhibited no motor alterations caused by the drugs. 3.1. Experiment 1: Role of morphine, and opioid agonists in rlPAG in behavioral selection In Experiment 1, the effects opioids’ action in the rlPAG by morphine and the agonists of and receptors varied. Both morphine and the and opioid receptor-specific agonists did not interfere with the predatory parameters (Fig. 2). No differences were found among groups in the number of cockroaches captured, latency to capture each cockroach, and percentage of dams that captured each of the five cockroaches. Maternal parameters were impaired in the MOR group compared with the CS group (Fig. 3). The latencies scores to retrieve the 1st (p = 0.007), 5th (p = 0.005), and 8th (p = 0.050) pups were longer in the dams that received morphine compared with controls. The percentage of animals that retrieved each pup was smaller in the MOR group compared with the CS group (p < 0.05 for all comparisons). The time spent licking the pups was less in the MOR group compared with the CS and groups (F3,46 = 4.526, p = 0.0076), but the percentages of animals that displayed FMB and nursing and number of pup contacts were not different among groups (Table 1). Stimulation of a single or opioid receptor did not significantly interfere with the predatory or maternal parameters. No significant differences were found in the predatory parameters during the entire time of observation. Thus, considering the total time of the test, the control group hunted as well as the MOR group. To test whether the groups had different hunting strategies, Table 1 Maternal parameters analyzed for 1800 s in Experiment 1. Parameter
FMB (%) Nursing (%) Pup contact Licking (s)
we evaluated primary motivation at the beginning of the test (i.e., whether the dams preferred to hunt before or after caring for their pups). Therefore, we decided to compare the number of insects captured in the initial 15 min (0–900 s) with the total time of the test (1800 s). We did the same with the number of pups retrieved (Table 2). Indeed, the control group preferred to retrieve most of their pups prior to capturing the cockroaches, whereas the MOR group captured almost all of the cockroaches at the beginning of the test. The group, although not statistically significant, tended to behave similarly to the MOR group, whereas the group was more similar to the control group. Thus, morphine appeared to act at both receptors at the same time to disrupt maternal behavior. To evaluate the importance of each of these opioid receptor subtypes in morphine’s action, we performed Experiment 2. 3.2. Experiment 2: Blockade of and opioid receptors face of morphine action in rlPAG In this experiment, opioid receptor blockade under the action of morphine in the rlPAG impaired hunting behavior in these dams (Fig. 4). The NM group captured fewer cockroaches than the CS and MOR groups (F3,45 = 4.147, p = 0.0116). Furthermore, the latencies to capture the 1st to 4th cockroaches were longer in the NM group compared with the CS and MOR groups (p < 0.05 for all comparisons). The percentage of lactating females in the NM group that captured each of the five insects was lower (p < 0.05 for all comparisons). With regard to maternal parameters, all of the groups that received a morphine injection (i.e., MOR, NM, and NM groups) exhibited some disruptions of aspects of this behavior (Fig. 5). The latency scores to retrieve the 1st pup in all of these groups were longer compared with the CS group (p = 0.001). The MOR and NM group also longer latency scores to retrieve the 5th Table 2 Insects captured and pups retrieved in the initial 900 s and during the entire behavioral test (1800 s) in Experiment 1.
Group CS
MOR
20.0 60.0 7.0 ± 2.6 130.1 ± 24.9
0.0 38.5 4.8 ± 1.0 35.8 ± 14.0*
16.7 50.0 5.7 ± 1.4 131.9 ± 29.9
16.7 75.0 6.8 ± 1.1 116.6 ± 18.1
FMB, full maternal behavior; CS, control group (n = 10); MOR, morphine group (n = 13); , receptor agonist group (DAMGO; n = 12); , receptor agonist group (U69593; n = 12). FMB and nursing behavior were analyzed using Fisher’s exact test. Pup contacts were analyzed using the Kruskal–Wallis test. Licking behavior was analyzed using one-way ANOVA followed by Tukey’s test. The data are expressed as mean ± SEM. * p < 0.05, compared with CS and groups.
Parameter
Insects captured until 900 s Insects captured during 1800 s Pups retrieved until 900 s Pups retrieved during 1800 s
Group CS
MOR
1.5 (0–5) 4 (0–5) 7 (0–8) 8 (7–8)
4 (0–5) 5 (0–5) 0 (0–8)* 0 (0–8)**
3.5 (0–5) 5 (0–5) 5 (0–8) 5.5 (0–8)
0 (0–5) 3 (0–5) 5.5 (0–8) 8 (0–8)
CS, control group (n = 10); MOR, morphine group (n = 13); , receptor agonist group (DAMGO; n = 12); , receptor agonist group (U69593; n = 12). The data are expressed as median (range). * p < 0.05, compared with CS and groups. ** p < 0.05, compared with CS group (Kruskal–Wallis test followed by Dunn’s test).
M.O. Klein et al. / Behavioural Brain Research 274 (2014) 62–72
67
Fig. 4. Predatory hunting parameters in Experiment 2. CS, control group (n = 10); MOR, morphine sulfate group (n = 13); NM, receptor antagonist (CTAP) + morphine group (n = 11); NM, receptor antagonist (nor-binaltorphimine) + morphine group (n = 12). (A) Number of insects captured during the behavioral test. The data are expressed as mean ± SEM (one-way ANOVA followed by Tukey’s test). (B) Latency to capture the 1st, 2nd, 3rd, 4th, and 5th cockroaches. The data are expressed as mean ± SEM (Kruskal–Wallis test). (C) Percentage of dams that captured the 1st, 2nd, 3rd, 4th, and 5th cockroaches. *p < 0.05, compared with CS and MOR groups; # p < 0.05, compared with CS group; **p < 0.05, compared with MOR group (Fisher’s exact test).
Fig. 5. Maternal parameters in Experiment 2. CS, control group (n = 10); MOR, morphine sulfate group (n = 13); NM, receptor antagonist (CTAP) + morphine group (n = 11); NM, receptor antagonist (U69593) + morphine group (n = 12). (A) Latency to retrieve the 1st, 5th, and 8th pups after being transformed into rank-order data (1: 0–300 s; 2: 301–600 s; 3: 601–900 s; 4: 901–1200 s; 5: 1201–1500 s; 6: 1501–1800 s; 7: no retrieval during 30 min test). The data are expressed as mean ± SEM (Kruskal–Wallis test followed by Dunn’s test). (B) Percentage of dams that retrieved each of their 1st, 5th, and 8th pups. *p < 0.05, compared with control (CS) group (Fisher’s exact test).
68
M.O. Klein et al. / Behavioural Brain Research 274 (2014) 62–72
altered by the blockade of opioid receptors, we further investigated the endogenous tone of and receptors in Experiment 3.
Table 3 Maternal parameters analyzed during 1800 s in Experiment 2. Parameter
FMB (%) Nursing (%) Pup contact Licking (s)
Group CS
MOR
NM
NM
10.0 80.0 5.4 ± 1.2 120.8 ± 26.0
0.0 38.5 4.3 ± 0.9 44.1 ± 15.8
0.0 9.0* 2.09 ± 0.6 9.8 ± 5.07*
0.0 8.3* 5.1 ± 1.5 36.9 ± 15.1*
FMB, full maternal behavior; CS, control group (n = 10); MOR, morphine sulfate group (n = 13); NM, receptor antagonist (CTAP) + morphine group (n = 11); NM, receptor antagonist (U69593) + morphine group (n = 12). FMB and nursing behavior were analyzed using Fisher’s exact test. Pup contacts and licking behavior were analyzed using the Kruskal-Wallis test followed by Dunn’s test. The data are expressed as mean ± SEM. * p < 0.05, compared with CS group.
(p = 0.0041) and 8th (p = 0.0003) pups compared with controls. Less than 40% of the dams in the MOR, NM, and NM groups retrieved the pups, whereas at least 80% of the dams in the CS group retrieved them (p < 0.05 for all comparisons). The number of dams that displayed nursing and licking behaviors was less in the NM and NM groups compared with the CS group (Table 3; p < 0.01 for all comparisons). Because we found that predatory parameters were
3.3. Experiment 3: Blockade of and opioid receptors in rlPAG behavioral selection In this experiment, dams in the N group that received the opioid receptor antagonist into the rlPAG captured more cockroaches (F2,32 = 4.754, p = 0.0161) and had shorter latencies to capture each cockroach (p < 0.05 for all comparisons). Additionally, 100% of the N dams captured at least the 1st cockroach, whereas only 70% and 41% of the animals displayed such behavior in the control and N groups, respectively (p < 0.05; Fig. 6). With regard to the maternal parameters, the latency scores to retrieve the 1st, 5th, and 8th pups were longer in the N group compared with the CS group (Fig. 7A; p < 0.04 for all comparisons). The percentage of dams that retrieved each pup was lower in the N group (Fig. 7B; p < 0.04). Dams in the N group displayed less nursing behavior (p < 0.04) and spent less time licking their pups (F2,32 = 4.390, p = 0.0213; Table 4). Animals in Experiment 3 had different hunting strategies (Table 5). In the first 900 s of the test, N dams caught all of the
Fig. 6. Predatory hunting parameters in Experiment 3. CS, control group (n = 10); N, receptor antagonist group (CTAP; n = 12), N, receptor antagonist group (norbinaltorphimine; n = 11). (A) Number of insects captured during the behavioral test. The data are expressed as mean ± SEM (one-way ANOVA followed by Tukey’s test). (B) Latency to capture each cockroach. The data are expressed as mean ± SEM (Kruskal–Wallis test followed by Dunn’s test). (C) Percentage of dams that captured the 1st, 2nd, 3rd, 4th, and 5th cockroaches. *p < 0.05, compared with N group (Fisher’s exact test).
M.O. Klein et al. / Behavioural Brain Research 274 (2014) 62–72
69
Fig. 7. Maternal parameters in Experiment 3. CS, control group (n = 10); N, receptor antagonist group (CTAP; n = 12); N, receptor antagonist group (nor-binaltorphimine; n = 11). (A) Latency to retrieve the 1st, 5th, and 8th pups after being transformed into rank-order data (1: 0–300 s; 2: 301–600 s; 3: 601–900 s; 4: 901–1200 s; 5: 1201–1500 s; 6: 1501–1800 s; 7: no retrieval during 30 min test). The data are expressed as mean ± SEM (Kruskal–Wallis test followed by Dunn’s test). (B) Percentage of dams that retrieved each of their pups. *p < 0.05, compared with control (CS) group; **p < 0.05, compared with CS and N groups (Fisher’s exact test).
insects and retrieved most of the pups, whereas dams in the CS group first retrieved the pups and then captured the cockroaches. Most of the animals in the N group did not capture roaches (nine of 12 dams) or take care of pups (six of 12 dams). 3.4. Experiment 4: Opioids receptor activation in the rlPAG To evaluate the activation of and opioid receptors, we performed Experiment 4. The intensity of fluorescence (red for receptors and green for receptors) indicates the level of activation of these receptors caused by drug stimulation in each of the groups in Experiments 1–3 (Fig. 8A). The control group exhibited a basal tone of activation of both receptors, whereas morphine sulfate significantly activated both and receptors in the MOR group Table 4 Maternal parameters analyzed during 1800 s in Experiment 3.
FMB (%) Nursing (%) Pup contact Licking (s)
4. Discussion
Group
Parameter CS
N
N
20.0 70.0 5.9 ± 1.0 104.5 ± 21.4
0.0 16.7* 4.92 ± 1.1 34.5 ± 15.4#
27.3 63.6 4.8 ± 0.8 70.3 ± 13.2
FMB, full maternal behavior; CS, control group (n = 10); N, receptor antagonist group (CTAP; n = 12); N, receptor antagonist group (nor-binaltorphimine; n = 11). FMB and nursing behavior were analyzed using Fisher’s exact test. Pup contacts and licking behavior were analyzed using one-way ANOVA followed by Tukey’s test. The data are expressed as mean ± SEM. * p < 0.05, compared with CS and N groups. # p < 0.05, compared with control group.
Table 5 Insects captured and pups retrieved in the initial 900 s and during the entire behavioral test (1800 s) in Experiment 3. Parameter
Insects captured until 900 s Insects captured during 1800 s Pups retrieved until 900 s Pups retrieved during 1800 s
(: F3,17 = 12.18, p = 0.0003, Fig. 8B; : F3,17 = 3.68, p = 0.04, Fig. 8E). Although they received a specific agonist, the and groups did not exhibit significant levels of opioid receptor activation. In the groups in Experiment 2 (Fig. 8C and F), the injection of a single opioid receptor antagonist in the presence of morphine (i.e., in the NM and NM groups) reversed the overactivation caused by morphine alone ( receptor: F3,17 = 8.487, p = 0.0018; receptor: F3,16 = 4.457, p = 0.023). Fig. 8D and G show the groups in Experiment 3. Activation of the opioid receptor in the N group was not significantly different from the CS group, but receptor activation in the N group was higher than in the CS and N groups (140.4% ± 21.8% of control [mean ± SEM], F2,10 = 4.601, p = 0.0468). With regard to opioid receptor activation in the groups in Experiment 3 (Fig. 8G), no significant differences were found among groups (F2,10 = 0.5105, p > 0.05).
Group CS
N
N
1 (0–5) 3.5 (0–5) 8 (0–8) 8 (7–8)
0 (0–5) 0 (0–5) 0.5 (0–8)# 0.5 (0–8)#
5 (0–5)* 5 (1–5)* 6 (0–8) 7 (0–8)
CS, control group (n = 10); N, receptor antagonist group (CTAP; n = 12); N, receptor antagonist group (nor-binaltorphimine; n = 11). The data are expressed as median (range). * p < 0.03, compared with N group. # p < 0.01, compared with CS group (Kruskal–Wallis test followed by Dunn’s test).
The PAG is a putative center that switches adaptive behavioral responses [26]. The rlPAG plays an important role in switching from maternal to predatory behavior in lactating rats [11]. Morphine injected directly into this area impairs pup retrieval, nest building, and FMB in mothers [18]. In the present study, we tested the role of morphine injected into the rlPAG in the context of selecting maternal vs. predatory behaviors. We also analyzed the functional relevance of and opioid receptors in this paradigm. We injected and opioid receptor agonists in the rlPAG in dams in Experiment 1. Unlike what happens when injected peripherally [13], morphine administered centrally in the rlPAG disrupted maternal behavior but did not increase predatory hunting in the dams. Thus, peripheral injections of morphine may cause the activation of opioid receptors in other areas together with the PAG. This appears to be necessary to generate morphine-induced increases in predatory hunting. Other areas, such as the ventrolateral caudate putamen, amygdala, hypothalamus, and superior colliculus, play a role in predatory hunting, the motivational drive to forage and hunt, and opioid receptor expression [8,9,19,27–29]. Lactating females appear to be more efficient in hunting and catching prey compared with pregnant and virgin females [4]. Lactating females in the CS group had a greater level of expression of hunting (i.e., beyond caring for pups, females in the CS group also foraged and captured prey). Interestingly, CS females retrieved their pups first and hunted for insects afterward. This did not occur in the other groups, mainly in females who received a morphine injection. These animals displayed only hunting behavior.
70
M.O. Klein et al. / Behavioural Brain Research 274 (2014) 62–72
Fig. 8. Opioid receptor activation. (Left) (A) Fluorescent photomicrograph of the periaqueductal gray in Experiment 4. The intensity of fluorescence indicates the activation of each opioid receptor subtype ( and ) in the tissue. Scale bar = 150 m. (Right) Activation of and opioid receptors in the dams after the behavioral test. For presentation purposes, the data are shown as a percentage of activation relative to the control (CS) group (n = 5). Just the groups within each Experiment (1–3) were compared to each other for each receptor. MOR (morphine group; n = 4), ( opioid receptor agonist [DAMGO] group; n = 4), ( opioid receptor agonist [U69593] group; n = 5), NM ( opioid receptor antagonist [CTAP] with consecutive morphine injection; NM (n = 5): opioid receptor antagonist [nor-binaltorphimine] plus morphine; n = 4), N ( opioid receptor antagonist [CTAP]; n = 4), N ( opioid receptor antagonist [nor-binaltorphimine] group; n = 4). The red line indicates the CS group. (B) Activation of opioid receptors in Experiment 1. ***p < 0.001, compared with CS, , and groups (one-way ANOVA followed by Tukey’s test). (C) Activation of opioid receptors in Experiment 2. **p < 0.01, compared with CS and NM groups (one-way ANOVA followed by Tukey’s test). (D) Activation of opioid receptors in Experiment 3. *p < 0.05, compared with CS and N groups (one-way ANOVA followed by unpaired t-test). (E) Activation of opioid receptors in Experiment 1. # p < 0.05, compared with CS group (one-way ANOVA followed by Tukey’s test). (F) Activation of opioid receptors in Experiment 2. # p < 0.05, compared with CS and NM groups (one-way ANOVA followed by Tukey’s test). (G) Activation of opioid receptors in Experiment 3 (one-way ANOVA).
Intracerebroventricular injection of the opioid agonist is as effective as morphine or -endorphin in inhibiting maternal behavior [15,20]. In the present study, the same opioid agonist injected into the rlPAG did not interfere either with maternal behavior or predatory hunting. Thus, the activation of opioid receptors in other areas beyond the rlPAG might be necessary to inhibit maternal behavior, or the dose used in this study was not enough even being a high dose [20]. Experiment 4 showed that intrarlPAG injection of the opioid agonist did not significantly activate this receptor. This finding may be related to insufficient receptor activation to promote behavioral changes in this paradigm. Similarly, peripheral injections of the agonist increased the latency to retrieve the pups [21,30]. However, intracerebroventricular injection of the agonist did not have any effect on maternal behavior [20]. In the present study, the opioid receptor agonist injected directly into the rlPAG also did not interfere with maternal behavior or behavioral selection. Similarly, a single injection of the agonist did not activate a significant number of receptors in Experiment 4. Thus, at the tested doses, these opioid agonists might be ineffective in activating their respective receptors in the rlPAG. Beyond that, the peripheral injection of the specific and agonists may act in other areas of the brain, which together can impair the
maternal behavior. More studies needs to be performed to clarify this issue. In Experiment 1, a morphine injection into the rlPAG alone inhibited maternal behavior. Such a behavioral effect may be attributable to morphine’s action on more than one type of opioid receptor simultaneously. Furthermore, it is important to state that in this study, the role of ␦ opioid receptor was not investigated, and morphine, even in a minor scale, acts in this type of receptor. Therefore, ␦ receptor in rlPAG could also plays a role in modulating morphine inhibition of maternal behavior and behavioral selection in all experimental paradigms presented [31]. To indirectly address the possible specific functional meaning of the morphine-induced activation of and opioid receptor, we specifically blocked and receptors in Experiment 2. All of the morphine-treated groups exhibited impairments in maternal behavior, even with and receptors blocked. This supports the hypothesis that morphine may act on more than one type of opioid receptor to inhibit maternal behavior. The presence of the receptor antagonist in the rlPAG concurrently with morphine inhibited predatory hunting, whereas the presence of the receptor antagonist with morphine or morphine alone did not alter this behavior. Therefore, opioid receptors in the rlPAG appear to have a central function in the
M.O. Klein et al. / Behavioural Brain Research 274 (2014) 62–72
expression of predatory hunting. This finding led us to investigate the role of endogenous and opioid receptor tone in behavioral selection in Experiment 3. The injection of the receptor antagonist did not significantly interfere with predatory behavior but impaired the care of pups. This suggests that the basal tone of receptors is important for the regular expression of maternal behavior. Other studies have shown that diminishing or increasing receptor activation inhibits maternal behavior [13,32,33]. The blockade of receptors facilitated predatory behavior without interfering with parameters of maternal behavior. Receptor activation may restrain the expression of predatory hunting, and the blockade of receptors may facilitate this behavior. Experiment 4 failed to show significant effects of and receptor antagonists on the endogenous activation of their specific receptors. Since the specificity of these antagonists is well established [34–36], these findings may be due to the fact that the basal activation was so low in the present experimental situation that made it difficult to decrease to even lower levels. However, receptor blockade increased the activation of opioid receptors in the rlPAG in Experiment 4. This finding suggests the existence of an important functional balance between these two receptor subtypes that may promote predatory hunting, i.e., once the action of one type of receptor is diminished, the activity of the other one would increase to compensate the lack of action from the first. Thus, the activity of opioid receptors appears to be affected by inhibitory endogenous receptor tone. Some important general aspects of the behavioral tests in Experiments 1–3 should be considered. Previous studies have shown that central injections of morphine impair pup retrieval, nest building, and the expression of FMB [18]. In the present study, nest building was not evaluated because of the absence of pine flakes in the experimental cages. Behavior was tested without pine flakes to better analyze predatory behavior (i.e., the visualization of cockroaches would be hampered by pine flakes in the cages). Limited access to nest material is a well-established model of neglected maternal behavior [37,38]. This may explain why the percentage of dams that expressed FMB and nursing in all of the groups, including the CS group, was lower in the present study compared with other studies that utilized free access to nest material [13]. Because this test was the first time that the dams had contact with roaches, the novelty of the insects may have indirectly influenced the expression of FMB. Changes in the time taken to begin catching the cockroaches and latency scores to retrieve the 1st pup might reflect the motivational drive of the dams and may be useful to evaluate this drive [11]. The rlPAG is an integrative region that plays an important role in motivational and rewarded behaviors (e.g., hunting, foraging, and reward seeking) [29,39]. This feature may have influenced the present results. The central modulation exerted by the drugs may have influenced the motivation to elicit behavioral responses. Interestingly, beyond interfering with the retrieve of pups, drugs also impaired the nursing and licking behavior. It shows that in addition to interfere with the motivation to care pups, they also disrupted the maintenance of this behavior, mainly regarding to opioid receptor. The drug-induced behavioral changes reported herein might be attributable to changes in the activational state of PAG opioid receptors. The level of activation of and opioid receptors in the rlPAG was measured in all of the experimental paradigms. As expected, morphine increased the activation of both and receptors. However, the blockade of one of these receptors in the presence of morphine decreased the activation caused by morphine alone, restoring it to control levels. Interestingly, treatment with and receptor agonists did not result in differences in and receptor activation compared with controls. Indeed, the and groups showed no significant changes in behavioral parameters.
71
This may be explained by an inability of the agonists to activate a significant number of receptors that would be sufficient to influence behavior in this paradigm. Also, as previously said, the doses used might be not enough to show any behavioral or molecular alteration. In Experiment 3, the N group, which received a receptor antagonist, exhibited an increase in the activation of opioid receptors. This group also presented an improvement in predatory hunting, whereas receptor blockade in the N group impaired the expression of predatory behavior. This may be attributable to a functional balance between opioid receptors, in which the suppression of receptors, directly or indirectly, may facilitate receptor activation, always keeping a basal tone of receptors activation. This suggests the existence of a dynamic interaction among opioid receptors that may involve other central systems. Such a system may compensate for the lack of action of one receptor by increasing the activation of another. This mechanism might play a role in the behavioral changes observed in the present study. 5. Conclusion Our data suggest that endogenous opioid receptor tone in the rlPAG is essential for the expression of hunting behavior and can be modulated by the degree of receptor activation. Thus, receptor-mediated opioidergic transmission appears to be critical for behavioral selection during lactation. The pharmacological blockade of opioid receptors leads to a behaviorally meaningful increase in endogenous receptor activation, which can be directly or indirectly mediated. A role for compensatory opioidergic multireceptor interactions that determine behavioral responses now appears likely. Conflict of interest The authors have no conflict of interest to report. Funding source This work was supported by grant from Fundac¸ão de Amparo à Pesquisa do Estado de São Paulo (FAPESP) processes 2010/06774-0 and 2013/01610-7. Funding source played no role in experimental design or decision to submit the paper for publication. Acknowledgment The authors thank to Miriam Aline Geigner for her technical assistance in the immunofluorescence assay. References [1] Blaffer-Hrdy S. Mother nature: maternal instincts and how they shape the human species. New York, NY: Ballentine; 1999. [2] Felicio LF, Canteras NS. Maternal choices: neural mediation: caring for young or hunting. In: Bridges RS, editor. Neurobiology of the parental brain. New York: Elsevier; 2008. p. 75–82. [3] Cruz Ade M, Maiorka PC, Canteras NS, Sukikara MH, Felicio LF. Morphine treatment during pregnancy modulates behavioral selection in lactating rats. Physiol Behav 2010;101:40–4. [4] Kinsley CH, Amory-Meyer E. Why the maternal brain? J Neuroendocrinol 2011;23:974–83. [5] Teodorov E, Felicio LF, Bernardi MM. Maternal behavior. In: Andersen ML, Tufik S, editors. Animal models as tools in ethical biomedical research. São Paulo: UNIFESP; 2010. p. 149–65. [6] Numan M, Fleming AS, Levy F. Maternal behavior. In: Neil JD, editor. Knobil and Neill’s physiology of reproduction. Amsterdam: Elsevier; 2006. p. 1921–93. [7] Comoli E, Ribeiro-Barbosa ER, Canteras NS. Predatory hunting and exposure to a live predator induce opposite patterns of Fos immunoreactivity in the PAG. Behav Brain Res 2003;138:17–28. [8] Comoli E, Ribeiro-Barbosa ER, Negrao N, Goto M, Canteras NS. Functional mapping of the prosencephalic systems involved in organizing predatory behavior in rats. Neuroscience 2005;130:1055–67.
72
M.O. Klein et al. / Behavioural Brain Research 274 (2014) 62–72
[9] Furigo IC, de Oliveira WF, de Oliveira AR, Comoli E, Baldo MV, Mota-Ortiz SR, et al. The role of the superior colliculus in predatory hunting. Neuroscience 2010;165:1–15. [10] Miranda-Paiva CM, Ribeiro-Barbosa ER, Canteras NS, Felicio LF. A role for the periaqueductal grey in opioidergic inhibition of maternal behaviour. Eur J Neurosci 2003;18:667–74. [11] Sukikara MH, Mota-Ortiz SR, Baldo MV, Felicio LF, Canteras NS. A role for the periaqueductal gray in switching adaptive behavioral responses. J Neurosci 2006;26:2583–9. [12] Behbehani MM. Functional characteristics of the midbrain periaqueductal gray. Prog Neurobiol 1995;46:575–605. [13] Sukikara MH, Platero MD, Canteras NS, Felicio LF. Opiate regulation of behavioral selection during lactation. Pharmacol Biochem Behav 2007;87:315–20. [14] Bridges RS, Grimm CT. Reversal of morphine disruption of maternal behavior by concurrent treatment with the opiate antagonist naloxone. Science 1982;218:166–8. [15] Felicio LF, Mann PE, Bridges RS. Intracerebroventricular cholecystokinin infusions block beta-endorphin-induced disruption of maternal behavior. Pharmacol Biochem Behav 1991;39:201–4. [16] Grimm CT, Bridges RS. Opiate regulation of maternal behavior in the rat. Pharmacol Biochem Behav 1983;19:609–16. [17] Martin WR. Pharmacology of opioids. Pharmacol Rev 1984;35:283–323. [18] Moura LM, Canteras NS, Sukikara MH, Felicio LF. Morphine infusions into the rostrolateral periaqueductal gray affect maternal behaviors. Braz J Med Biol Res 2010;43:899–905. [19] Mansour A, Fox CA, Burke S, Meng F, Thompson RC, Akil H, et al. Mu, delta, and kappa opioid receptor mRNA expression in the rat CNS: an in situ hybridization study. J Compar Neurol 1994;350:412–38. [20] Mann PE, Kinsley CH, Bridges RS. Opioid receptor subtype involvement in maternal behavior in lactating rats. Neuroendocrinology 1991;53:487–92. [21] Teodorov E, Tomita AT, Banon GP, Gil IG, Bernardi MM, Felicio LF. Behavioral effects of acute stimulation of -opioid receptors during lactation. Pharmacol Biochem Behav 2008;90:534–9. [22] Paxinos G, Watson C. The rat brain in stereotaxic coordinates. 4th edition San Diego, CA: Academic Press; 1998. [23] Gupta A, Decaillot FM, Gomes I, Tkalych O, Heimann AS, Ferro ES, et al. Conformation state-sensitive antibodies to G-protein-coupled receptors. J Biol Chem 2007;282:5116–24. [24] Machado FC, Zambelli VO, Fernandes ACO, Heimann AS, Cury Y, Picolo G. Peripheral interactions between cannabinoid and opioid systems contribute to the antinociceptive effectof crotalphine. Br J Pharmacol 2014;171(4):961–72, http://dx.doi.org/10.1111/bph.12488. [25] Nasello AG, Depiante R, Tannhauser M, Izquierdo I. Effect of morphine on the RNA and ATP concentration of brain structures of the rat. Pharmacology 1973;10:56–9.
[26] Sukikara MH, Mota-Ortiz SR, Baldo MV, Felicio LF, Canteras NS. The periaqueductal gray and its potential role in maternal behavior inhibition in response to predatory threats. Behav Brain Res 2010;209:226–33. [27] dos Santos LM, Ferro MM, Mota-Ortiz SR, Baldo MV, da Cunha C, Canteras NS. Effects of ventrolateral striatal inactivation on predatory hunting. Physiol Behav 2007;90:669–73. [28] dos Santos LM, Boschen SL, Bortolanza M, de Oliveira WF, Furigo IC, Mota-Ortiz SR, et al. The role of the ventrolateral caudoputamen in predatory hunting. Physiol Behav 2012;105:893–8. [29] Mota-Ortiz SR, Sukikara MH, Felicio LF, Canteras NS. Afferent connections to the rostrolateral part of the periaqueductal gray: a critical region influencing the motivation drive to hunt and forage. Neural Plast 2009;2009:612698. [30] Teodorov E, Ferrari MF, Fior-Chadi DR, Camarini R, Felicio LF. Behavioral meaningful opioidergic stimulation activates kappa receptor gene expression. Braz J Med Biol Res 2012;45:982–7. [31] Gendron L, Mittal N, Beaudry H, Walwyn W. Recent advances on the delta opioid receptor: from trafficking to function. Br J Pharmacol 2014;26, http://dx.doi.org/10.1111/bph.12706. [32] Byrnes EM, Rigero BA, Bridges RS. Opioid receptor antagonism during early lactation results in the increased duration of nursing bouts. Physiol Behav 2000;70:211–6. [33] Mann PE, Pasternak GW, Bridges RS. Mu1 opioid receptor involviment in maternal behavior. Physiol Behav 1990;47:133–8. [34] Kramer TH, Shook JE, Kazmierski W, Ayres EA, Wire WS, Hruby VJ, et al. Novel peptidic mu opioid antagonists: pharmacologic characterization in vitro and in vivo. J Pharmacol Exp Ther 1989;249(2):544–51. [35] Abbruscato TJ, Thomas SA, Hruby VJ, Davis TP. Blood–brain barrier permeability and bioavailability of a highly potent and mu-selective opioid receptor antagonist, CTAP: comparison with morphine. J Pharmacol Exp Ther 1997;280(1):402–9. [36] Portoghese PS, Lin CE, Farouz-Grant F, Takemori AE. Structure–activity relationship of N17 -substituted norbinaltorphimine congeners. Role of the N17 basic group in the interaction with a putative address subsite on the kappa opioid receptor. J Med Chem 1994;37(10):1495–500. [37] Dalle Molle R, Portella AK, Goldani MZ, Kapczinski FP, Leistner-Segal S, Salum GA, et al. Associations between parenting behavior and anxiety in a rodent model and a clinical sample: relationship to peripheral BDNF levels. Trans Psychiatry 2012;2:e195. [38] Ivy AS, Brunson KL, Sandman C, Baram TZ. Dysfunctional nurturing behavior in rat dams with limited access to nesting material: a clinically relevant model for early-life stress. Neuroscience 2008;154:1132–42. [39] Mota-Ortiz SR, Sukikara MH, Bittencourt JC, Baldo MV, Elias CF, Felicio LF, et al. The periaqueductal gray as a critical site to mediate reward seeking during predatory hunting. Behav Brain Res 2012;226:32–40.