Introduction

A number of experiments have provided information on behavioral emotional expression and rat communication. In rodents, this communication takes place through the emission of sounds that are inaudible to humans (in bands above 18 kHz), so-called ultrasound vocalizations (USVs). Many years of behavioral analysis determined the frequency of the emitted signals, intervals, and sound architecture and classified ultrasound into two basic bands of USV emission: 22- and 50-kHz USVs (Sales 1972; Fu and Brudzynski 1994; Vivian and Miczek 1998; Wöhr et al. 2005; Litvin et al. 2007; Panksepp et al. 2004; Brudzynski 2013; Wright et al. 2010; Hamed et al. 2009).

It is believed that these two distinct bands can be assigned to separate (respective) emotional states of the animals (Brudzynski 2013); 22-kHz USVs are usually associated with the expression of the negative emotional states, manifesting anxiety, fear, and pain (Brudzyński 2007; Wöhr and Schwarting 2008; Wöhr et al. 2005). After ejaculation (Barfield and Geyer 1975), 22-kHz USV calls were also detected. Aversive 22-kHz USVs (alarm calls) are uniform and stable in frequency (from 18 to 28 kHz). It is believed that these calls are a method of communication to warn other rats of dangers; they also serve as an expression of fear (Brudzynski 2009; Panksepp et al. 2004; Litvin et al. 2007; Wöhr et al. 2005; Yee et al. 2012). The 50-kHz USVs are shorter than the 22-kHz (3–300 ms) USVs and are more complex in acoustic architecture (Wright et al. 2010; Fu and Brudzynski 1994; Hamed et al. 2009; Brudzynski 2013); 50-kHz USVs may reflect an appetitive emotional state, and therefore this method is used in studies of drug addiction, sexual behavior, and social interactions (Knutson et al. 2002; Brudzyński 2007; Wang et al. 2008; Brudzynski 2009; Burgdorf et al. 2011; Brudzyński and Pniak 2002). This type of USV is also considered to be a method of social communication (ethotransmission) (Brudzynski 2013) and was recorded during social interactions (Brudzyński and Pniak 2002; Łopuch and Popik 2011; Manduca et al. 2014; Hamed et al. 2009). The frequency range of 50-kHz USVs can extend from 30 to 130 kHz. The sound frequency distribution and the meaning of various types of 50-kHz USV episodes remain poorly understood. Wright et al. (2010) extracted 14 subtypes of emitted sounds in the 50-kHz USV band (Wright et al. 2010). Several of these subtypes are characterized by the modulation of sound frequency—frequency modulated (FM). Despite this classification, it is difficult to assign the different 50-kHz subtypes to specific animal behaviors.

Individual differences in the intensity of the emission of 50-kHz USVs are useful in research related to addictive and psychoactive substances (Meyer et al. 2011; Ma et al. 2010; Taracha et al. 2012). Appetitive 50-kHz USVs were successfully used in studies with cocaine (Maier et al. 2010; Barker et al. 2010; Ma et al. 2010) and amphetamine (Ahrens et al. 2009; Burgdorf et al. 2001; Wang et al. 2008; Wright et al. 2010; Brudzynski et al. 2011; Taracha et al. 2012) and occurred in response to these substances and also during drug anticipation (Burgdorf et al. 2001; Ma et al. 2010; Hamed et al. 2012).

Morphine (a derivative of Papaver somniferum) is a psychoactive substance with addictive potential; it has strong affinity to the μ-opioid receptor (MOR) and lower affinity to the κ-opioid (KOR) and δ-opioid receptors (Zhang et al. 2012). Morphine also activates the mesolimbic reward system (including the medial prefrontal cortex (mPFC), nucleus accumbens (NAcc), and ventral tegmental area (VTA)) by stimulating opioid receptors in the VTA and NAcc (Bozarth 1987; Trigo et al. 2010), which are related to pleasure, motivation, and a positive emotional state (Johnson and North 1992).

It has been proven that dopaminergic (DA) neurons in the VTA are controlled by GABAergic interneurons (Johnson and North 1992; Adell and Artigas 2004), which coordinate DA release (Riegel and Kalivas 2010). By activating MOR, morphine has been shown to increase dopamine transmission by decreasing the release of GABA onto dopamine neurons in the VTA (Johnson and North 1992). Although morphine meets classical criteria for inducing USVs (it causes place preference (CPP) and induces dopamine release in the mesolimbic system), it inhibits appetitive vocalizations (50-kHz USVs) (Hamed et al. 2012; Wright et al. 2012; Simola et al. 2012; Nagai et al. 2004).

In an attempt to define the effects of opioid-receptor activation on the reward system (measured by 50-kHz USVs), we used a social interaction (pair encounter) test after 21 days of isolation (long social deprivation) (Hamed et al. 2009). Previously, we observed that long-term (21 days) social deprivation significantly intensified the frequency of 50-kHz call emission during social encounters (Hamed et al. 2009). It is well known that the mesolimbic reward system is highly involved in the production of 50-kHz USVs, and the activation of dopaminergic neurons projecting from the VTA to the NAcc is associated with 50-kHz USVs emission (Knutson et al. 1999; Burgdorf et al. 2001, 2007). Taking into account that DA also plays a role in aversion (Bromberg-Martin et al. 2010; Zweifel et al. 2011; Lammel et al. 2011), we have analyzed fast Fourier transform (FFT) spectrograms in whole recorded frequency (10–130 kHz) to exclude the presence of 22-kHz USVs.

The aims of this study were to (i) study the pharmacological impact of KOR and MOR ligands on the emission of 50-kHz USVs triggered by social interaction after long-term isolation and (ii) analyze the concentrations of the main neurotransmitters in reward-related structures (VTA, NAcc, and mPFC) after pharmacological and behavioral modifications.

Materials and methods

Animals

Adult male Sprague–Dawley rats (n = 154; n (pairs) = 77; 180 ± 20 g) were used in the experiment. The animals were purchased from a licensed breeder (the Polish Academy of Science Medical Research Center, Warsaw, Poland). The animals were housed in standard laboratory conditions under 12/12-h light/dark cycles (lights on at 7 a.m.) at a constant temperature (21 ± 2 °C) and 70 % humidity. The rats had free access to food and water. The experiments were performed in accordance with the European Communities Council Directive of 24 November 1986 (86/609 EEC). The Local Committee for Animal Care and Use of Warsaw Medical University approved all experimental procedures using animal subjects.

Experimental protocol

Procedure

Sprague–Dawley rats (n = 154) at 8–10 weeks of age (180 ± 20 g) were housed in groups of eight (and in one cage group of ten; procedures A and B; n = 154) for 2 weeks in acrylic cages (cage size, 54 × 34 × 21 cm). After 2 weeks, the animals were separated and housed individually in non-transparent cages (cage size, 36 × 27 × 19 cm) in the same room. The rats were socially isolated for 3 weeks with food and water provided ad libitum. Then, pairs of rats (randomly selected from the same cage) received drug injections; paired rats were injected with the same substance: morphine in doses of 0.1 mg/kg (Morph 0.1; n = 14), 1.0 mg/kg (Morph 1.0; n = 16), 5.0 mg/kg (Morph 5.0; n = 12), naltrexone in a dose of 5.0 mg/kg (naltr 5; n = 16), or U-50488 in a dose of 10 mg/kg (U-50488; n = 14) in “procedure A” (I injection) and saline, naltrexone in 5.0 mg/kg (intraperitoneal (i.p.); naltr 5) (n = 10), or nor-BNI 10 mg/kg (subcutaneous (s.c.); nor-BNI 10; n = 12) (I injection). Thirty minutes later, the rats were paired and placed on soundproof testing cages (45 × 30 × 45 cm), and their common USVs were recorded for the next 15 min. In procedure B after that time, all of the animals received an injection of morphine (5.0 mg/kg, s.c.; n = 32; II injection; “procedure B”). USVs were recorded for the next 30 min (procedure B). The total USVs emitted by the two rats were recorded and measured.

Apparatus and USV recordings

To see the details of USVs recording and analysis of spectrograms, please refer to “Apparatus and USV recordings” previously described in Hamed et al. (2009, 2012).

Taking into account that dopaminergic system plays a role in the processing of both appetitive and aversive state (Bromberg-Martin et al. 2010; Zweifel et al. 2011; Lammel et al. 2011), we have analyzed FFT spectrograms in whole recorded frequency spectrum (10–130 kHz) to evaluate occurrence not only “50-kHz appetitive” but also “22-kHz aversive” calls. Detailed analysis of the FFT spectrograms showed the absence of 22-kHz (alarm calls) in presented model.

To compare locomotor activity between groups, a Sony YR600 1/3 in. CCD 540 TVL camera (Sony, Japan) was fixed to the room ceiling and a PC equipped with the EthoVision® XT Video Tracking System v.7 (Noldus Information Technology B.V., Wageningen, The Netherlands) were used. The total distance moved (cm) and the mean velocity (cm/s) were measured.

Video analysis of the animal behavior during social interactions was measured manually using BehaView software (custom made, Warsaw). The parameters such as number of pouncing, pouncing duration, number of pinning, pinning duration, number of “nose to nose,” nose-to-nose duration, number of anogenital investigation, duration of anogenital investigation, number of following, duration of following, number of cleaning, cleaning duration, number of cage exploration, and duration of cage exploration were manually measured using BehaView software.

Drugs

Morphine hydrochloride (Polfa, Poland) and naltrexone hydrochloride (MOR and KOR antagonist; Tocris, UK), U-50488 hydrochloride (KOR agonist; Tocris, UK), and nor-binaltorphimine dihydrochloride (nor-BNI; KOR antagonist; Tocris, UK) was dissolved in 0.9 % isotonic saline. Morphine was administered (s.c.) in doses of 0.1, 1.0, and 5.0 mg/kg (1 ml/kg). Naltrexone was administered (i.p.) in doses of 5.0 mg/kg (1 ml/kg). U-50488 hydrochloride was administered (i.p.) in a dose of 10 mg/kg (2 ml/kg). Nor-BNI was administered (s.c.) in a dose of 10 mg/kg (1 ml/kg). Saline was used as a control vehicle for all drugs.

Determination of monoamines in brain structures (VTA, NAcc, and mPFC)

Frozen brains were cut into slices (−20 °C) in a cryostat. The following structures were dissected: prefrontal cortex (3.7 to 3.2 mm anterior to the bregma), nucleus accumbens (1.70 to 1.0 mm posterior to the bregma), and ventral tegmental area (−5.80 to −6.30 mm anterior from bregma) based on the rat brain atlas of Paxinos and Watson (1998). Tissue samples were weighed and homogenized for 30 s in 15 volumes of ice-cold 0.2 M perchloric acid containing dihydroxybenzylamine as an internal standard. The homogenates were then centrifuged at 26,900×g for 8 min at 4 °C. The obtained supernatants were filtered through 0.45-μm pore filters and then stored at −70 °C until analyzed for noradrenaline (NA), dopamine (DA), 3,4-dihydroxyphenylacetic acid (DOPAC), homovanillic acid (HVA), 3-methoxytyramine (3-MT), serotonin (5-HT), and 5-hydroxyindoleacetic acid (5-HIAA) via HPLC as described by Kaneda et al. (1986), with minor modifications (Szyndler et al. 2010). The concentrations of NA, DA, DOPAC, 3-MT, HVA, 5-HT, and 5-HIAA were calculated as nanograms per gram of brain tissue.

HPLC analysis of amino acids

The brain levels of the amino acids were determined using HPLC method with an electrochemical detection as described previously (Szyndler et al. 2008).

In brief, the compounds were eluted isocratically with the mobile phase delivered at 0.7 ml/min using a Shimadzu Class LC-10ADvp pump. An electrochemical detector with a flow-through cell (Intro-Antec Leyden) linked to the Shimadzu Class VP Integrator SCL-10 Avp was used. A high-density glassy carbon-working electrode (Antec) was operated at +0.85 V. A Rheodyne injection valve with 20-μl sample loops was used to manually inject the samples. The preparation of the mobile phase and the derivatizing agents was based on the method of Rowley et al. (1995), with some modifications (Szyndler et al. 2008). The concentrations of GABA and glutamate were calculated in micromoles per gram of tissue.

Statistics

The data were expressed as the means ± standard error of the means (SEM). All behavioral data were analyzed using one-way analysis of variance (ANOVA). These measures were followed using a Newman-Keuls post hoc test. To analyze the neurochemical data, Student’s t test was used to compare the non-isolated and isolated groups. Additional neurochemical data were analyzed using one-way ANOVA followed by the post hoc Newman-Keuls test. The correlation between 50-kHz ultrasonic vocalization and another behavioral data and correlation between 50-kHz ultrasonic vocalization and biochemical effects, obtained from the same animal, was performed using the Pearson’s r correlation test. A probability value of p < 0.05 was considered significant in this study. All of the statistical analyses were performed using StatSoft Statistica 10 for Windows (StatSoft Inc. USA).

Results

Effects of morphine, naltrexone, nor-BNI, and U-50488 on isolation-induced 50-kHz USVs during pair encounters

One-way ANOVA revealed significant differences between the effects of the acute administration of morphine, naltrexone, nor-BNI, and U-50488 on the number of 50-kHz USV episodes (F (7, 51) = 34.35, p < 0.001). A Newman-Keuls post hoc analysis showed a decrease in a number of 50-kHz USV episodes after the acute administration of morphine at a dose of 1.0 mg/kg (p < 0.01), morphine at a dose of 5.0 mg/kg (p < 0.001), naltrexone at a dose of 5.0 mg/kg (p < 0.001), and U-50488 at a dose of 10 mg/kg (p < 0.001) compared with the saline group (isolated). A Newman-Keuls post hoc analysis also showed an increase in the number of 50-kHz USV episodes during pair encounters after acute nor-BNI at a dose of 10 mg/kg (s.c.; p < 0.01) during the rat encounters after 21 days of social isolation (Fig. 1). Detailed analysis of the FFT spectrograms showed the absence of 22-kHz (alarm calls) in the presented model.

Fig. 1
figure 1

The effects of morphine, naltrexone, nor-BNI, and U-50488 on the number of 50-kHz USV episodes during pair encounters after 21 days of isolation. Group legend: the acute administration of morphine in doses of 0.1 mg/kg (Morph 0.1), 1.0 mg/kg (Morph 1.0), and 5.0 mg/kg (Morph 5.0), naltrexone (Naltr 5.0), nor-BNI 10 mg/kg (s.c.; nor-BNI 10) and U-50488 10 mg/kg (i.p.; U-50488). **p < 0.01; ***p < 0.001—differs from the isolated group (saline)

Full size image

Effects of morphine, naltrexone, nor-BNI, and U-50488 on animal behavior during pair encounters

One-way ANOVA revealed significant differences between the effects of the acute administration of saline, morphine, naltrexone, nor-BNI, U-50488, and non-isolated group (saline) on the number of pouncing (F (7, 32) = 15.74, p < 0.001), pouncing duration (s) (F (7, 32) = 20.73, p < 0.001), number of pinning (F (7, 31) = 11.21, p < 0.001)), pinning duration (s) (F (7, 32) = 3.24, p < 0.05), number of nose to nose (F (7, 32) = 10.58, p < 0.001), nose-to-nose duration (s) (F (32, 7) = 5.61, p < 0.001), number of anogenital investigation (F (7, 32) = 13.41, p < 0.001), duration of anogenital investigation (s) (F (7, 32) = 16.15, p < 0.001), following (n) (F (7, 32) = 20.55, p < 0.001), following duration (s) (F (7, 32) = 11.37, p < 0.001), cleaning (n) (F (7, 32) = 9.02, p < 0.001), cleaning duration (s) (F (7, 32) = 4.41, p < 0.01), number of cage exploration (F (7, 32) = 5.54, p < 0.001), and duration of cage exploration (s) (F (7, 32) = 2.90, p < 0.05). One-way ANOVA also revealed significant differences between the effects of the acute administration of morphine, naltrexone, nor-BNI, and U-50488 and non-isolated group (saline) on total distance (cm) (F (7, 50) = 16.50, p < 0.001) and mean velocity (cm/s) (F (7, 50) = 16.50, p < 0.001). The results of post hoc Newman-Keuls analysis are shown in Table 1.

Table 1 The effects of 21 days isolation on rats behavior (video analysis) during encounter; effects of morphine, naltrexone, nor-BNI, and U-50488 acute administration before isolated rat encounters
Full size table

An animal behavior during pair encounters were correlated with a number of 50-kHz ultrasonic vocalization. Positive correlations were present between the number of 50-kHz USVs and number of pouncing (r = (+) 0.6581, p < 0.001), pouncing duration (r = (+) 0.7595, p < 0.001), number of pinning (r = (+) 0.5624, p < 0.001), number of “nose-to-nose” episodes (r = (+) 0.7067, p < 0.001), duration of nose to nose (r = (+) 0.6529, p < 0.001), number of anogenital investigations (r = (+) 0.6415, p < 0.001), duration of anogenital investigation (r = (+) 0.5872, p < 0.001), number of “following” episodes (r = (+) 0.5848, p < 0.001), number of “cage exploration” episodes (r = (+) 0.5829, p < 0.001), total distance moved (r = (+) 0.6142, p < 0.001), and mean velocity (r = (+) 0.6162, p < 0.001) (Table 1).

Neurochemical effects of 21-day social isolation on pair encounters

Student’s t test revealed a significantly higher concentration of noradrenaline in the homogenates of the VTA (t = 2.89; df = 18; p < 0.01) and mPFC (t = 2.4; df = 18; p < 0.05) in the isolated group compared with the non-isolated group, higher dopamine concentrations in the VTA (t = 4.99; df = 18; p < 0.001), NAcc (t = 3.87; df = 18; p < 0.01), and mPFC (t = 2.37; df = 18; p < 0.05), higher concentrations of the dopamine metabolites DOPAC (in VTA; t = 3.56; df = 18; p < 0.01), HVA (in VTA; t = 3.42; df = 18; p < 0.01), and 3-MT (t = 4.26; df = 18; p < 0.001) in the VTA, as well as higher 3-MT concentrations in the NAcc (t = 4.09; df = 18; p < 0.01) in the isolated group compared with the non-isolated group. Student’s t test also revealed higher concentrations of serotonin in the NAcc (t = 6.36; df = 18; p < 0.001) and in the mPFC (t = 2.7; df = 18; p < 0.05) and a higher concentration of the serotonin metabolite 5-HIAA in the NAcc (t = 3.06; df = 18; p < 0.01) in the isolated group compared with the non-isolated group. This test also revealed a higher concentration of GABA in the VTA (t = 4.00; df = 18; p < 0.001) in the isolated group compared with the non-isolated group as well as a higher concentration of glutamate in the VTA (t = 2.18; df = 18; p < 0.05) (Fig. 2).

Fig. 2
figure 2figure 2

The neurotransmitter concentrations after the rat encounter comparing the isolated (21-day isolation) and non-isolated groups. *p < 0.05; **p < 0.01; ***p < 0.001—differs from the non-isolated group

Full size image

Neurochemical effects of the acute administration of morphine at a dose of 5.0 mg/kg, nor-BNI at a dose of 10 mg/kg, and U-50488

One-way ANOVA revealed significant differences between the effects of the acute administration of morphine, nor-BNI, or U-50488 on the dopamine concentrations in the VTA (F (3, 34) = 11.13, p < 0.001) and in the mPFC (F (3, 34) = 3.19, p < 0.05), as well as differences in the dopamine metabolites concentrations: DOPAC in the VTA (F (3, 34) = 12.29, p < 0.001), HVA in the VTA (F (3, 34) = 6.27, p < 0.01), HVA in the NAcc (F (3, 34) = 8.65, p < 0.001), HVA in the mPFC (F (3, 34) = 7.25, p < 0.001), 3-MT in the NAcc (F (3, 34) = 5.05, p < 0.01), and 3-MT in the mPFC (F (3, 34) = 10.11, p < 0.001). One-way ANOVA also revealed significant differences in the serotonin concentrations of the VTA (F (3, 34) = 3.33, p < 0.05), serotonin in the NAcc (F (3, 34) = 4.49, p < 0.01), and differences in the serotonin metabolite 5-HIAA concentrations in the VTA (F (3, 34) = 8.13, p < 0.001), the NAcc (F (3, 34) = 5.44, p < 0.01), and the mPFC (F (3, 34) = 3.76, p < 0.05), as well as differences in GABA concentrations in the VTA (F (3, 34) = 32.13, p < 0.001) and NAcc (F (3, 34) = 6.91, p < 0.001) and differences in the glutamate concentration in the VTA (F (3, 34) = 13.67, p < 0.001). One-way ANOVA also revealed significant differences between the effects of the acute administration of morphine, nor-BNI and U-50488 on the 5-HIAA/3-MT ratio in the VTA (F (3, 34) = 9.74, p < 0.001).

A post hoc Newman-Keuls analysis showed a decreased dopamine concentration in the VTA after morphine at a dose of 5.0 mg/kg (p < 0.001), nor-BNI (p < 0.01), and U-50488 (p < 0.001), as well as in the mPFC in the nor-BNI group (p < 0.05) in this model compared with the saline (isolated) group.

The post hoc Newman-Keuls test also showed a decrease in the dopamine metabolites concentrations (the DOPAC in the VTA in the morphine 5.0 (p < 0.01) group, in the U-50488 group (p < 0.001), and in the nor-BNI group (p < 0.01) compared with the saline (isolated) group). Decreased concentrations of HVA were observed in the VTA of the U-50488 group (p < 0.01) compared with the saline (isolated) group. However, a post hoc Newman-Keuls analysis showed an increased HVA (dopamine metabolite) concentration in the NAcc (p < 0.001) and the mPFC (p < 0.01) in the morphine group compared with the saline group as well as an increased concentration of dopamine metabolite 3-MT in the NAcc (p < 0.01) and mPFC (p < 0.001) in the nor-BNI group. An increased concentration of 3-MT was observed in the NAcc of the morphine group (p < 0.05) compared with the control group. A post hoc Newman-Keuls test showed a decreased serotonin concentration in the NAcc of the morphine group (p < 0.01) and an increased concentration of serotonin in the VTA (p < 0.05) in the morphine group compared with the saline (isolated) group. This test also showed an increased level of 5-HIAA (serotonin metabolite) in the VTA (p < 0.001) in the morphine-treated group compared with the control group and a decreased concentration of 5-HIAA (serotonin metabolite) in the NAcc in the morphine (p < 0.01), U-50488 (p < 0.01), and nor-BNI (p < 0.01) groups compared with the saline (isolated) group, as well as a decreased concentration of 5-HIAA in the mPFC in the nor-BNI group (p < 0.05). A post hoc Newman-Keuls test also demonstrated a decreased concentration of GABA in the VTA in the morphine (p < 0.001), nor-BNI (p < 0.001), and U-50488 (p < 0.001) groups, as well as a decreased GABA concentration in the NAcc in the morphine (p < 0.01), U-50488 (p < 0.01), and nor-BNI groups (p < 0.01) compared with the saline group. A post hoc Newman-Keuls also showed a decreased concentration of glutamate in the VTA of the morphine group (p < 0.01), U-50488 group (p < 0.05), and nor-BNI group (p < 0.001) compared with the saline (isolated) group (Fig. 3). A post hoc Newman-Keuls test also demonstrated an increased 5-HIAA/3-MT ratio in the morphine (p < 0.001) and U-50488 groups (p < 0.05) compared with the isolated (saline) group (Fig. 5).

Fig. 3
figure 3

The effects of morphine (5.0 mg/kg), KOR agonist U-50488 (10 mg/kg), and antagonist nor-BNI (10 mg/kg (s.c)) on the dopamine and dopamine metabolites (DOPAC, HVA, 3-MT), serotonin and serotonin metabolite 5-HIAA, GABA and glutamate concentrations in the VTA, NAcc, and mPFC after the rat encounters (all groups were socially isolated for 21 days). Group legend: the acute administration of morphinein a dose of 5.0 mg/kg (morph), U-50488 in a dose of 10 mg/kg, and nor-BNI (s.c.) in a dose of 10 mg/kg (s.c.). *p < 0.05; **p < 0.01; ***p < 0.001—differs from the isolated (saline) group

Full size image

Effects of naltrexone and nor-BNI (s.c.) on isolation-induced 50-kHz USVs during pair encounters after morphine injection

One-way ANOVA revealed significant differences between the effects of pretreatment of naltrexone and nor-BNI on the number of 50-kHz USVs episodes after morphine at a dose of 5.0 mg/kg injection (F (2, 13) = 10.77, p < 0.01).

A Newman-Keuls post hoc analysis showed an increase in the number of 50-kHz USVs in the 5.0 mg/kg naltrexone pretreated group (naltr + morph) compared with the saline + morph group (saline + morph; p < 0.01), and also an increased number of 50-kHz USVs episodes in the nor-BNI pretreated group (nor-BNI + morph; p < 0.01) compared with the saline + morph group (Fig. 4).

Fig. 4
figure 4

The effects of naltrexone (KOR and MOR antagonist) and nor-BNI (KOR antagonist) on morphine (5.0 mg/kg) suppression on the number of 50-kHz USV episodes during pair encounters after 21 days of isolation—procedure B. Group legend: the acute administration of saline and morphine (saline + morph; control group), acute administration of naltrexone at a dose of 5.0 mg/kg and morphine (naltr + morph) and nor-BNI (s.c.) at doses of 10 mg/kg plus morphine (nor-BNI + morph). **p < 0.01—differs from the “saline + morph” group

Full size image

Discussion

Social interaction after 21 days isolation leads to enhanced emission of 50-kHz USVs (Hamed et al. 2009). Ultrasonic vocalizations analysis showed a decrease in a number of 50-kHz USV episodes during rat encounters after the acute administration of morphine at a dose of 1.0 and 5.0 mg/kg, naltrexone at a dose of 5.0 mg/kg, and U-50488 at a dose of 10 mg/kg. An increased 50-kHz USV emission in this model has been observed after acute nor-BNI at a dose of 10 mg/kg (s.c.). Naltrexone in a dose of 5.0 mg/kg and nor-BNI in a dose of 10 mg/kg (s.c.) reduced the 50-kHz suppressive effect produced by morphine in a dose of 5.0 mg/kg.

An increased 50-kHz ultrasonic vocalization after administration of KOR antagonist closely corresponds to data on the increase of social interaction demonstrated in earlier publications, using nor-BNI at 3.0 mg/kg (Vanderschuren et al. 1995) as well as the decrease of behaviors associated with reward after administration of KOR agonist U-69593. In our study, these effects were observed in the analysis of USVs emission; however, in the case of KOR antagonist, they were not confirmed in the video analysis. The same team also showed an increase of pinning behavior after 0.1 mg/kg of morphine and demonstrated the participation of MOR in strengthening reward-related social behaviors (Trezza et al. 2011a, b). In our study, the increase of social reward-related behaviors after administration of morphine did not occur; however, we noticed a reduction of these behaviors and also increased mean velocity after low doses of morphine. These differences may be a result of different procedures, rat species, test cages used, and rats’ age.

There have been also reports of indicating both an increase in the 50-kHz ultrasonic vocalization and reward-related behaviors in rats during social interaction after administration of morphine (Manduca et al. 2014), as well as indicating a decrease of these behaviors measured with 50-kHz USV method (Wright et al. 2012). The applied apparatus and method of 50-kHz ultrasonic vocalization measurement (detector vs microphone) seem to be a key differentiating factor. We cannot exclude that the presence of experimenter may affect the outcome of behavioral tests; however, substances were administered in another room, as well as during the measurement the experimenter was outside the room. It is worth noting that in the studies, which compare methods of substance administration that is experimenter delivered and self-administrated, biochemical effects did not differ (Roth-Deri et al. 2003). In our study, the emission of 50-kHz ultrasonic vocalizations correlates most strongly with the duration of pouncing behavior and with the number of pinning, as well as duration of anogenital investigation (Table 1), which is considered to be a pro-social (social exploration) behavior (Vanderschuren et al. 1997). The comparative observation of video analysis with episodes marked on the timeline on FFT spectrograms showed that such behaviors as pinning, pouncing, and anogenital investigation trigger the most 50-kHz vocalization in rats after 21-day isolation. The statistical analysis also revealed a strong correlation with “nose-to-nose” behaviors; however, all aggregate behaviors were correlated during the session. In the comparative observation of episodes with behaviors from video analysis, we observed that nose-to-nose behavior was not a reaction triggering the 50-kHz USVs, what would suggest the 50-kHz ultrasonic vocalization’s ability to increase social interaction. It is also worth mentioning that another study which used 50-kHz USV playback, showed that appetitive sounds are very important for increasing rats’ social interactions (Wöhr and Schwarting 2007), and sound emission at 50-kHz as opposed to 22-kHz (alarm calls) triggers increases in dopamine levels in NAcc (Willuhn et al. 2014).

There is a certain analogy in the 50-kHz USVs induction of pinning behavior and pouncing behavior with reactions induced by tickling procedure, which also triggers appetitive ultrasonic vocalizations at 50-kHz band (Panksepp and Burgdorf 2000; Burgdorf and Panksepp 2001; Schwarting et al. 2007; Rygula et al. 2012).

Social interaction after 21 days of isolation leads to enhanced concentrations of dopamine and its metabolites (DOPAC, HVA, and 3-MT) in the VTA, NAcc, and mPFC as well as enhanced concentrations of norepinephrine, GABA, and glutamate. Similarly, increased levels of serotonin and serotonin metabolite (5-HIAA) concentrations were observed in the NAcc of the isolated (saline) group. We did not observe a significant increase in the concentration of GABA and glutamate in the NAcc in long-term isolated rats after social encounters. However, in the mPFC, higher concentrations of norepinephrine, dopamine, and serotonin were observed in the isolated compared with non-isolated rats. This neurochemical changes profile corresponds with increases in the number of emitted appetitive 50-kHz USVs (Figs. 1 and 2).

Literature data indicates that the dopaminergic mesolimbic system is involved in the rewarding effects of addictive drugs and 50-kHz USV seems to be related with this phenomena (Wintink and Brudzyński 2001; Schultz 2010; Thompson et al. 2006; Brudzynski 2009; Taracha et al. 2012; Simola et al. 2012). It should be noted that increases of dopamine levels in the mesolimbic system cannot be assigned only with reward. There is strong evidence reported also in the optogenetics studies which indicate the differentiation of brain dopamine functions in various structures involved in emotional processing such as signaling of aversion and reward (Abercrombie et al. 1989; Salamone 1994; Joseph et al. 2003; Lammel et al. 2008, 2011, 2014; Ungless et al. 2010; Ungless 2004; Bromberg-Martin et al. 2010; Zweifel et al. 2011; Britt and Bonci 2013; Jones et al. 1990; Ikemoto and Bonci 2014; Tanimoto et al. 2004). It is worth noting that other studies have shown that aversive stimuli increase levels of dopamine in mPFC in a greater extent than in Nacc (Mantz et al. 1989; Abercrombie et al. 1989; Bassareo et al. 2002).

In analyzing the ratio of the metabolites of serotonin–5-HIAA and dopamine–3-MT, we noticed that the 5-HIAA/3-MT ratio in the VTA is inversely proportional to the number of emitted ultrasound vocalizations at 50 kHz in the isolated group (correlation analysis r = (−) 0.7197; p < 0.01). Interestingly, in the morphine (5.0 mg/kg) group, the concentrations analysis of the VTA neurotransmitters showed increased levels of serotonin and 5-HIAA (serotonin metabolite) in this model and also an increased concentration of 3-MT and HVA (metabolite of dopamine) in the NAcc. It is well known that morphine increases the levels of dopamine metabolite in the NAcc (Johnson and North 1992) and may also increase the levels of serotonin in the nucleus accumbens and other forebrain regions (Spampinato et al. 1985; Tao and Auerbach 1994; Harris and Aston-Jones 2001; Munn and Borszcz 2002). In the morphine group, the ratio of the serotonin metabolite 5-HIAA to the dopamine metabolite 3-MT in the VTA was higher but 50-kHz was lower compared with saline-isolated group. Our biochemical data also suggest that the 50-kHz USVs reduction both in the morphine and U-50488 groups were associated with a high 5-HIAA/3-MT ratio in the VTA (Fig. 5). Conversely, in the nor-BNI group, the 5-HIAA/3-MT ratio in the VTA was low (due to increased levels of 3-MT), which corresponded with the increased 50-kHz USVs during social pair encounters (Figs. 1 and 5).

Fig. 5
figure 5

Changes in the 5-HIAA/3-MT concentration ratio in the VTA in morphine at a dose of 5.0 mg/kg, U-50488, and nor-BNI groups. *p < 0.05; ***p < 0.001—differs from the isolated (saline) group

Full size image

In our study, the concentration of serotonin in the VTA did not change after the administration of either the KOR agonist U-50488 or antagonist nor-BNI; the bi-directional influence of KOR ligands on the 50-kHz USVs resulted from changes in the concentration of dopamine and dopamine metabolites, as presented in this study and previously described in the literature (Spanagel et al. 1992; Margolis et al. 2003; Di Chiara and Imperato 1988a, b). KOR and MOR agonists exert opposing influences on a number of central systems, including those that mediate rewarding stimuli (Pan 1998). The local administration of selective MOR (DAMGO) and KOR (U-50488) agonists to the VTA evoke opposing behavioral effects on conditioned place preference (CPP): MOR agonists produce dose-dependent preferences for the drug-associated site, and KOR agonists produce place aversion (Bals-Kubik et al. 1993). Margolis et al. (2003) stated that the opposing behavioral effects of the MOR and KOR agonists acting in the VTA could be explained by the specific localization of MORs and KORs onto interconnected subpopulations of neurons that produce opposite effects on dopaminergic outputs into the NAcc (Margolis et al. 2003). It was also demonstrated that DAMGO produced a significant dose-related increase in dopamine release in the NAcc and that CTOP (MOR antagonist) produced a significant decrease in dopamine release; higher doses of the KOR agonist U-69593 (infusion to NAcc) were shown to be less effective in decreasing dopamine release than lower doses (Spanagel et al. 1992). Similarly, opposite behavioral effects of MOR and KOR were observed in another study, which demonstrated that core and shell intra-NAcc infusions of DAMGO (a selective MOR agonist) increased social play, whereas the intra-NAcc infusion of KOR agonist U-69593 decreased social play (Trezza et al. 2011a, b). Some data indicated that the acute systemic administration of KOR agonists decreased DA levels in the NAcc and dorsal striatum (Di Chiara and Imperato 1988a, b; Spanagel et al. 1990). Additionally, it is well known that U-50488 produces anxiety-like behaviors in the elevated plus maze test (Wiley et al. 2009; Valdez and Harshberger 2012).

It is noteworthy that the systemic administration of KOR agonists and their infusion into the VTA and NAcc or mPFC produced conditioned place aversions (Bals-Kubik et al. 1993). Tejada et al. (2013) showed evidence that the systemic administration of U-69593 (KOR agonist) resulted in decreased mPFC DA levels and led to place aversion in a CPA model (Tejada et al. 2013). The above data indicate that the activation of the KOR system inhibits dopamine-related behavior. Therefore, it is not surprising that the administration of U-50488 (KOR agonist) in this study caused a dramatic inhibition of 50-kHz USVs. Endogenous KOR systems appear to strictly regulate DA release in the mesolimbic system. It was demonstrated that KORs co-localize with dopamine transporter (DAT) in NAcc nerve terminals (Svingos et al. 2001; Shippenberg et al. 2007). It appears that KOR activation decreases DA neurotransmission in the mesolimbic reward system through two mechanisms: the inhibition of DA release and the stimulation of the uptake of dopamine in NAcc (Shippenberg et al. 2007). Another study demonstrated that in mice with the KORs deleted on their DA neurons, the administration of U-69593 (a KOR agonist) did not produce a conditioned place aversion. Furthermore, a lack of the inhibition of dopamine release in the NAcc was observed after KOR agonist administration (Chefer et al. 2013; Margolis et al. 2003). It is interesting that also aversive stimuli may increase dopamine levels in the NAcc and in the VTA (Abercrombie et al. 1989; Leknes and Tracey 2008; Brischoux et al. 2009). Some authors suggested that different subpopulations of VTA dopamine neurons are involved in aversive and reward response (Ikemoto 2007; Brischoux et al. 2009; Lammel et al. 2011; Budygin et al. 2012).

The presented data and data from the literature show that the selective KOR antagonist nor-BNI increases dopamine release in reward-related structures (Spanagel et al. 1992). In our study, subcutaneous injection of 10 mg/kg (s.c.) nor-BNI (KOR antagonist) increased 50-kHz USVs and also reduced the 50-kHz suppressive effect produced by morphine. Considering the presented biochemical data, we can presume that the 50-kHz USV suppression due to morphine were reduced by nor-BNI (KOR antagonist pretreatment) via increasing the dopamine and dopamine metabolite levels in VTA, consequently decreasing 5-HIAA/3-MT ratio and resulting in an increased number of 50-kHz USVs.

Literature data indicate that the endogenous opioid system may play a role in the motivational and hedonic properties of food, drug rewards, and social play (Panksepp et al. 1985, 1994; Vanderschuren et al. 1997; Siviy et al. 1997; Ikemoto and Pankepp 1999; Trezza et al. 2010; Wassum et al. 2009; Barbano and Cador 2007; Castro and Berridge 2014; Peciña and Berridge 2013; Pitchers et al. 2014; Laurent et al. 2014). It may be assumed that MOR endogenous opioids in the mesolimbic reward system were released during rat encounters. Thus, it is possible that the administration of naltrexone (with a high affinity to KOR and MOR (Peng et al. 2007)) prior to the rat encounters inhibited 50-kHz USVs in this model by antagonizing MOR, thus preventing the impact of endogenous MOR ligands. In this study, the neurotransmitter concentration analysis was not performed in the naltrexone group, and further research is needed to define the effects of naltrexone on the neurotransmitter balance in reward-related structures.

In turn, we also showed that naltrexone pretreatment (prior to morphine administration; procedure B) presumably reduced the 50-kHz suppression induced by morphine (Fig. 4). Similarly, other authors showed that the rate and duration of low- and high-frequency USVs (interactions with the aggressive component) were decreased by morphine in a naltrexone-reversible manner (Vivian and Miczek 1993). Considering the observation that the level of sound emitted in the 50-kHz band was inversely proportional to the 5-HIAA/3-MT ratio in the VTA, our data confirm the canonical conviction that the activation of the mesolimbic dopaminergic system (particularly in the NAcc) is the predictor of the emission of 50-kHz ultrasonic vocalizations; however, it appears that the key predictor for 50-kHz USVs emissions is the change in the balance between the serotonin and dopamine concentrations in the VTA. It should be emphasized that increased serotonin concentrations in the VTA may reduce 50-kHz USVs. Sadananda et al. (2012) observed that MDMA, the amphetamine derivative, did not lead to increased 50-kHz calling and suggested that a strong postsynaptic serotonergic impact may inhibit 50-kHz USVs (Sadananda et al. 2012). In another study, MDMA increased the extracellular concentrations of serotonin in the VTA after systemic administration (Bankson and Yamamoto 2004).

Altogether, our results and considerable data in the literature concerning opioid systems indicate that KOR activation may be a mediator in stress-related behavior that is associated with drugs of abuse (Valdez and Harshberger 2012; Shippenberg et al. 2007; Wall and Messier 2000, 2002). The effects of the MOR and KOR ligands on 50-kHz USVs indicate that the activation of both receptors play an important role in the regulation of the emotional expression measured via 50-kHz USVs in rats. Our study showed a bi-directional influence of KOR ligands, where KOR nor-BNI antagonist caused the increase of 50-kHz ultrasonic vocalization and KOR agonist U-50488 reduced the appetitive vocalization at 50-kHz band. These indicate a key importance of the KOR receptors in 50-kHz modulation associated with processing positive emotions.

The important finding in this paper is that changes in the balance of the serotonin and dopamine concentrations in the VTA may be a key predictor for 50-kHz USVs emission. We cannot rule out that the changes in the serotonin and dopamine concentration balance were the consequence of the endogenous opioids or other neurotransmitter actions (Wassum et al. 2009; Saunders and Robinson 2012; Narayanan et al. 2004). Studying the complex mechanisms and pharmacological modification of the mesolimbic system (measured via USVs in the 50-kHz band) using receptor ligands, such as KOR agonists and antagonists, may support further research on new drugs for the treatment and prevention of addictions. Bi-directional effect of KOR ligands on processing emotions as well as further studies with the use of KOR antagonist may also result in future clinical implementation in mood disorders or negative symptoms of schizophrenia treatment.