Inhibition of 50-kHz ultrasonic vocalizations by dopamine receptor subtype-selective agonists and antagonists in adult rats

Adult rats emit ultrasonic calls at around 22 and 50 kHz, which are often elicited by aversive and rewarding stimuli, respectively. Dopamine (DA) plays a role in aspects of both reward and aversion. The purpose of this study is to investigate the effects of DA receptor subtype-selective agonists on 22- and 50-kHz call rates. Ultrasonic calls were recorded in adult male rats that were initially screened with amphetamine to eliminate low 50-kHz callers. The remaining subjects were tested after acute intraperitoneal or subcutaneous injection of the following DA receptor-selective agonists and antagonists: A68930 (D1-like agonist), quinpirole (D2-like agonist), PD 128907 (D3 agonist), PD 168077 (D4 agonist), SCH 39166 (D1-like antagonist), L-741,626 (D2 antagonist), NGB 2904 (D3 antagonist), and L-745,870 (D4 antagonist). The indirect DA/noradrenaline agonist amphetamine served as a positive control. As expected, amphetamine strongly increased 50-kHz call rates. In contrast, D1-, D2-, and D3-selective DA receptor agonists, when given alone, inhibited calling; combinations of D1- and D2-like agonists also decreased call rate. Given alone, the D1-like and D3 antagonists significantly decreased call rate, with a similar trend for the D2 antagonist. Agonist–antagonist combinations also decreased calling. The D4 agonist and antagonist did not significantly affect 50-kHz call rates. Twenty-two-kilohertz calls occurred infrequently under all drug conditions. Following systemic drug administration, tonic pharmacological activation of D1-like or D2-like DA receptors, either alone or in combination, does not appear sufficient to induce 50-kHz calls. Dopaminergic transmission through D1, D2, and D3 receptors appears necessary for spontaneous calling.


Introduction
Adult rat ultrasonic vocalizations (USVs) are commonly divided into two main categories: calls in the the 20-30-kHz range, termed 22-kHz USVs, and calls in the 35-90-kHz range, termed 50-kHz USVs (Portfors 2007). These two categories have been proposed to indicate negative and positive affective states, respectively (Knutson et al. 2002;Wöhr and Schwarting 2012). For example, 22-kHz calls are emitted during confrontation with an aggressive conspecific or feline predator and in response to painful stimuli (Sales 1972b;Cuomo et al. 1988;van der Poel et al. 1989;Blanchard et al. 1991). In contrast, 50-kHz calls have been reported during rough-and-tumble play, copulation, and in anticipation of food delivery (Sales 1972a;Knutson et al. 1998;Burgdorf et al. 2000). However, the association between USV categories and affective valence appears more complex; notably, male rats emit 22-kHz calls after ejaculation (Barfield and Geyer 1972) and emit 50-kHz calls as well as 22-kHz calls during intermale aggression (Sales 1972b; Thomas et al. 1983).
Evidence from several studies suggests that dopaminergic (DAergic) neurotransmission plays a key role in the emission of 50-kHz calls in adult rats. For example, psychostimulant drugs (amphetamine, methylphenidate, and cocaine) that increase dopamine (DA) release and/or block reuptake at the somatodendritic and terminal level (Kalivas et al. 1989;Sulzer 2011) increase the emission of 50-kHz calls (Burgdorf et al. 2001;Wintink and Brudzynski 2001;Thompson et al. 2006;Ahrens et al. 2009;Wright et al. 2010;Meyer et al. 2011;Browning et al. 2011;Brudzynski et al. 2012;Simola et al. 2012;Wright et al. 2012a, b, c). However, all the above indirect DAergic agonists exert additional, non-DAergic effects, and it is therefore important to note that several DAergic antagonists have been found to markedly inhibit amphetamine-induced 50-kHz calling (Wright et al. 2012b).
Attempts to selectively activate DAergic receptors have, in contrast, produced conflicting findings. For example, the 50-kHz call rate was increased by intra-accumbens microinjection of the D2/D3 agonist (quinpirole; Brudzynski et al. 2012) and by a D2-like receptor antagonist (haloperidol; Thompson et al. 2006). To date, only two studies have investigated the effects of systemically administered D1and D2-like drugs on spontaneously emitted 50-kHz USVs. In the first of these, neither the agonists nor antagonists affected call rate, but baseline call rates were low (Williams and Undieh 2010). In the second, both D1-like and D2-like antagonists inhibited calling (Wright et al. 2012b). However, in the two latter studies, the test drugs were only selective for D1-like vs. D2-like receptor families rather than individual DA receptor subtypes (Andersen and Jansen 1990;Gehlert et al. 1992;Ruskin et al. 1998;Boulougouris et al. 2009).
The main aim of the present study was therefore to investigate the acute effects of DA receptor subtype-selective drugs on 50-kHz calls. To this end, we recorded USVs following acute systemic administration of DA receptor subtypeselective agonists, antagonists, and several agonist-antagonist combinations. Given that DA also plays a role in aversion (Bromberg-Martin et al. 2010, Lammel et al. 2012, we simultaneously recorded 22-kHz vocalizations. Finally, since functional synergy is sometimes observed between D1-like and D2like DA receptors (Clark and White 1987;LaHoste et al. 2000), we also tested combinations of a D1-and a D2-like agonist.

Subjects
Twenty experimentally naïve male Long-Evans rats (Charles River Laboratories, St. Constant, Quebec, Canada) were used in each experiment (total of 140 rats). The rats initially weighed 268-356 g at the beginning of the experiment. Subjects were housed two per cage in a temperatureand humidity-controlled colony room (20-22°C, 50-60 %). Home cage bedding consisted of laboratory grade Sani-Chips (Harlan Laboratories, Indianapolis, IN). Rats were maintained on a reverse 12:12-h light/dark cycle, with lights off at 0700 hours. Behavioral testing took place during the dark phase of the subjects' cycle, between 0800 and 1300 hours. Food and water were available ad libitum, except during testing. Subjects were each handled once daily for 3 min, for 2 days before the first experimental day. Exceptionally, in experiment 1, subjects were handled for 5 days before the start of the testing. All procedures were approved by the McGill Animal Care Committee in accordance with the guidelines of the Canadian Council on Animal Care.

Experimental protocol
Initial amphetamine screen Each experiment began with an initial amphetamine screen (Wright et al. 2012c). This served two purposes: (1) to exclude the significant minority of rats that emit few 50-kHz vocalizations in response to systemic amphetamine (Wright et al. 2010) and (2) to increase the acute response to this drug, which was also used as a positive control in later testing. Briefly, rats (n020) were given an acute injection of 1 mg/kg of amphetamine immediately before placement in the testing chambers (once daily 20-min session, for 3 days, spaced 2 days apart). Ultrasonic vocalizations occurring in the 12th, 14th, and 16th minute of day 3 were counted. The eight rats with the lowest call numbers were not tested further, leaving a group size of 12.
Drug tests A separate group of rats was used for each experiment (n012) except experiments 4 and 8, which were completed with the same group of rats (n012). Each rat received five to 16 test sessions, depending on the experiment, spaced 2 days apart. Fully parametric within-subject designs were employed (i.e., each rat was tested under each drug condition). Amphetamine (1 mg/kg intraperitoneal (IP)) served as a positive control throughout. The order of drug treatment within an experiment was as nearly counterbalanced as subject numbers allowed. By visual inspection, physical appearance and any unusual behaviours were noted before all injections, between injections (where applicable) and after each injection and test session.

Drugs
All test drugs, doses, injection timings and routes of administration are shown in Table 1. All doses were chosen based on behavioural effectiveness in other assays (Hoffman and  In all experiments, each rat was also tested with amphetamine (1 mg/kg IP, positive control) within the counterbalanced design a Given in a volume of 2 mL/kg (otherwise, given in a volume of 1 mL/kg) Psychopharmacology (2013) 226:589-600 Beninger 1988; Al-Naser and Cooper 1994;Bartoszyk 1998;Hsieh et al. 2004;Millan et al. 2004a;Fenu et al. 2005;Melis et al. 2006;Xi and Gardner 2007). The following drugs were used: the D1-like agonist A68930 hydrochloride, D1-like antagonist SCH 39166 hydrobromide, D2/D3 agonist (−)-quinpirole hydrochloride, D2 antagonist L-741,626, D3 agonist (+)-PD 128907 hydrochloride, D3 antagonist NGB 2904, D4 agonist PD 168077 maleate, and the D4 antagonist L-745,870 trihydrochloride. Drugs were purchased from Tocris Bioscience (Minneapolis, MN), except for D-amphetamine (Sigma-Aldrich, Poole, UK), A68930 and quinpirole (Sigma Aldrich, Oakville, ON). All drugs were dissolved in 0.9 % sterile saline, with the following exceptions: (1) L-741,626 was dissolved in 22 % DMSO/78 % deionized water v/v and (2) NGB 2904 was dissolved in a 5 % w/v solution of 2-hydroxypropyl-β-cyclodextrin in deionized water. The timing of each control (vehicle) injection matched that of the respective drug. Drug solutions were pH-matched to the corresponding vehicle solution (pH 5.6-7.0). All doses are expressed as salt. Drugs were administered in a volume of 1 ml/kg except for: (1) A68930 in experiments 1 and 5, (2) NGB 2904, and (3) L-741,626, which were all administered in a volume of 2 mL/kg, as were their corresponding vehicles. All drugs were administered by the IP or subcutaneous (SC) route (see Table 1).

Acoustic data acquisition and analysis of ultrasonic vocalizations
The apparatus, testing procedure, and acoustic analysis were as previously described (Wright et al. 2012c). Testing was carried out in clear Plexiglas™ experimental boxes (ENV-007CT, Med Associates, St. Albans, VT), each enclosed in a separate melamine compartment that was lined with soundattenuating acoustic foam (Primacoustic, Port Coquitlam, BC). Condenser ultrasound microphones (CM16/CMPA, Avisoft Bioacoustics, Berlin, Germany) were placed above a small (5-cm diameter) hole, located at the top center of each experimental box. The microphones were located 15-30 cm from the rat. Microphone signals were delivered to an Ultra-SoundGate 416H data acquisition device (Avisoft Bioacoustics) with a sampling rate of 250 kHz and 16-bit resolution. Avisoft SASLab Pro software (version 5.1.14, Avisoft Bioacoustics, Berlin, Germany) was used for acoustical analysis. Spectrograms were created with a fast Fourier transform length of 512 points and an overlap of 75 % (FlatTop window, 100 % frame size) yielding a frequency resolution of 490 Hz and a time resolution of 0.5 ms. Calls were selected manually from spectrograms by an individual masked to treatment conditions.

Data analysis and statistics
Data were analyzed using commercial software (Systat v11, SPSS, Chicago, IL; GraphPad Software, La Jolla, CA). Calls between 20 and 30 kHz were rarely observed and were not analyzed statistically. Call rate was defined as the total number of 50-kHz calls per minute. Analyzed time bins (see Table 1) were evenly spaced across the session, and the session duration was chosen based on the behavioural time course of each drug. Use of parametric vs. nonparametric tests depended on the distribution of the data. For example, nonparametric tests were used where the variances were heterogeneous. Multiple comparison tests were performed using Wilcoxon signed-rank tests. Single comparisons were done using paired t tests for vehicle conditions in all experiments except experiment 7. Differences between multiple vehicles were assessed by Friedman's nonparametric analysis of variance. For all tests, a two-tailed p value less than 5 % was considered significant.

Initial amphetamine screen, and subsequent saline and amphetamine tests
Since 22-kHz calls were seldom observed, they are reported only under the section "Other observations." During the initial amphetamine screen, the median 50-kHz call rate was 32 calls per minute with an interquartile range (IQR) of 5.5-61 (i.e., pooling all 140 rats from all experiments); with the low callers removed from each experiment (see "Methods"), the median 50-kHz call rate was 54 calls per minute, IQR 40-77 (i.e., pooling the 84 remaining rats). During drug testing blocks, the call rate was much lower under control conditions (i.e., after saline injection, median03 calls per minute, IQR 1-13, n084 rats) than after amphetamine administration (median061 calls per minute, IQR 28-85). This call-promoting effect of amphetamine was significant in all eight experiments (Wilcoxon's signed-rank test, Z02.824 to 3.059, p<0.01 for each).
Experiments 1-4: receptor subtype-selective dopamine agonist dose-response relationships For each experiment, the two control (vehicle) tests did not significantly differ with respect to 50-kHz call rates, and these data were averaged for each rat.
Experiment 6: D3 and D2/D3 agonists in combination with a D3 antagonist The observed effects of quinpirole (D2/D3 agonist) on call rate in experiment 2 resembled the call-suppressive effect of the selective D3 agonist PD 128907 from experiment 3. To test whether quinpirole's effects were due to its actions at the D3 receptor, we administered this drug in combination with a selective D3 receptor antagonist, NGB 2904 (Fig. 3). The call rates in the two control conditions, saline and β-cyclodextrin, did not differ significantly (paired t test, NS) and were averaged for each rat. Quinpirole and PD 128907, given alone, both significantly decreased calling (Wilcoxon: Z02.118 and 2.001, p<0.05). The D3 antagonist NGB 2904 itself did not significantly affect the call rate (Wilcoxon: Z 00.549, NS) and did not appear to reduce the agonist-induced call suppression (Fig. 3). In the presence of the D3 antagonist, both quinpirole and PD 128907 exerted a residual depressant effect on the call rate (antagonist alone vs. antagonist/ agonist combination, Wilcoxon: Z 02.511 and 1.961, p<0.05).
Experiment 7: D1-like, D2/D3, and D3 selective agonists in combination with selective antagonists In experiment 6, the D3 antagonist NGB 2904 failed to counter the call-suppressant effect of the D2/D3 agonist quinpirole and the D3 agonist PD 128907. Therefore, we next tested these agonists in combination with a higher dose of NGB 2904 (i.e., 2 mg/kg instead of 1 mg/kg). The same two agonists were also tested together with the D2-selective antagonist L-741,626. Within the same drug testing block, the D1-like agonist (A68930) was tested in combination with a D1-like antagonist (SCH 39166).

Other observations
Novel 22-kHz calls intermingled with 50-kHz calls We observed frequency modulated long 22-kHz calls that are different from the typically reported long 22-kHz calls. More specifically, these calls comprised a long (400-1,520 ms) low-frequency (24-29 kHz) component, preceded and/or followed by a highfrequency (41-61 kHz) component. These calls, which were intermingled with 50-kHz calls, occurred infrequently (i.e., a total of eight calls, found in two out of 12 rats in experiment 6) and only under amphetamine. In contrast, constant frequency 22-kHz calls were not observed in rats receiving amphetamine and seldom occurred under other drug conditions (14 calls in two rats).
Audible calls In experiment 1, the two highest doses of A68930 (1 and 4 mg/kg), caused audible calls approximately 1 h after the end of the session (2 h postinjection), in three of the 12 rats tested. These calls were emitted in their home cage in the presence of their cage mate and stopped immediately upon social separation.

Discussion
The present study provides the first report that spontaneous 50-kHz call rates can be reduced by systemic administration of DAergic agonists. Call inhibition occurred not only with D1-like, D2, and D3 receptor-selective agonists, but also with DAergic antagonists and agonist/antagonist combinations. These call rate-suppressive effects contrasted strongly with the well-established rate-enhancing effects of the indirect DA/noradrenaline (NA) agonist amphetamine that occurred reliably in the same animals.
Call-suppressive drug effects vs. motor inhibition Several classes of DAergic drugs affect motor function (for a review see Jackson and Westlind-Danielsson 1994). In the present study, the majority of DAergic agents decreased 50-kHz call rates, but only SCH 39166 (D1-like antagonist) produced visible signs of motor impairment or lethargy. For the remaining drugs, there was no consistent relationship with motor output. First, at the doses used there was no visible sign of catalepsy, which is consistent with literature reports (Millan et al. 1998(Millan et al. , 2000Banasikowski and Beninger 2012). Second, quinpirole inhibited calling not only at low, locomotor depressant doses (0.033 and 0.1 mg/kg; Schaub et al. 1997;Schindler and Carmona 2002) but also at higher doses reported to increase locomotor activity (LA) (1-10 mg/kg; Horvitz et al. 2001). Third, the D3 agonist PD 128907 inhibited calling even at low doses that would not be expected to affect LA (Gyertyan and Saghy 2004;Millan et al. 2004b). Fourth, NGB 2904 (D3 antagonist) decreased the call rate at doses that have been shown to increase spontaneous LA (Pritchard et al. 2007). Fifth, A68930 (D1-like agonist) and L-741,626 (D2 antagonist) are reported not to affect LA (Deveney and Waddington 1997;Clifford and Waddington 2000;Isacson et al. 2004;Nergardh et al. 2005;Koffarnus et al. 2011;Chang et al. 2011) at doses which inhibited calling. In conclusion, we cannot exclude the possibility that some drugs at certain doses (notably high-dose SCH 39166 and low-dose quinpirole) reduced 50-kHz call rates by inhibiting motor function. However, the present findings also provide examples where USV emission and locomotor activity can be dissociated, as previously reported with other drugs (Burgdorf et al. 2001;Natusch and Schwarting 2010;Wright et al. 2012a).
Combinations of D1-like and D2/D3-selective agonists decreased 50-kHz vocalizations Concurrent activation of postsynaptic D1-like and D2-like receptors appears to be required for the expression of several DAergic agonist-induced behaviours (Clark and White 1987;Dall'olio et al. 1988;Wachtel et al. 1989;Garrett and Holtzman 1994;Capper-Loup et al. 2002;Hasbi et al. 2011;Ikemoto et al. 1997). To address whether concurrent activation of D1 and D2 receptors is sufficient to elicit USVs, we administered A68930 and quinpirole in combination. Our lower dose of quinpirole (0.033 mg/kg) would selectively target inhibitory DA autoreceptors (Widzowski and Cory-Slechta 1993), whereas the higher dose (0.33 mg/ kg) would be expected to act predominantly at postsynaptic D2 receptors (Cory-Slechta et al. 1996). In the present study, both dose combinations inhibited calling. This result contrasts with a clear stimulant effect reported after systemic administration of the D1/D2-like DAergic agonist apomorphine (Williams and Undieh 2010). However, apomorphine may also have exerted non-DAergic actions at the high dose administered (2 mg/kg SC), for example at adrenergic and 5-HT receptors Newman-Tancredi et al. 2002).
The observed effects of systemically administered DAergic agonists suggest that DA receptor activation in multiple brain regions is insufficient to induce calling. Consistent with this conclusion, we recently observed that the callenhancing effect of systemically administered amphetamine is critically dependent on both dopaminergic and adrenergic receptor mechanisms (Wright et al. 2012b, c).
DA receptor antagonists decreased 50-kHz call rate without affecting agonist-induced inhibition Dopaminergic antagonists were initially reported to have no effect on spontaneous rates of 50-kHz vocalization after systemic administration (Wintink and Brudzynski 2001;Williams and Undieh 2010), but in both these studies the low basal rates of calling could potentially have masked any inhibitory effects. More recently, we observed a suppression of 50-kHz calling following systemic administration of the D1-and D2-like antagonists SCH 39166 and raclopride (Wright et al. 2012b). Extending the latter observations, the D1-like and D3-selective antagonists tested in the present study both decreased the 50-kHz call rate, with a similar trend for the D2-selective antagonist. Taken together, the inhibitory effects of systemically administered DAergic antagonists suggest that DA receptors are necessary for USV emission.
In the present study, the D2-and D3-selective antagonists did not significantly inhibit the effects of their respective agonists. It is likely that the antagonist doses were sufficiently high; first, these drugs appeared to inhibit calling when given alone, and second, comparable doses were effective in other behavioural assays (Fenu et al. 2005;Melis et al. 2006;Collins et al. 2007;Xi and Gardner 2007). In the latter studies, off target actions appear improbable since these drugs are reported to be highly receptor-selective (McQuade et al. 1991;Kebabian et al. 1992;Levant et al. 1993;Pugsley et al. 1995;Bowery et al. 1996;Glase et al. 1997;Patel et al. 1997;Yuan et al. 1998).

Comparisons with amphetamine and cocaine
The inhibition of 50-kHz calling by D1-like and D2-like agonists is particularly striking when set against the robust call stimulation associated with systemic administration of the indirect agonists amphetamine (Wintink and Brudzynski 2001;Thompson et al. 2006;Wright et al. 2010;Simola et al. 2012) and cocaine (Mu et al. 2009;Williams and Undieh 2010;Meyer et al. 2011;Wright et al. 2012c). Psychostimulant drugs, via presynaptic actions, enhance NA as well as DAergic transmission (Kuczenski et al. 1995(Kuczenski et al. , 1997Berridge and Stalnaker 2002), and NAergic mechanisms are clearly critical to amphetamine-induced 50-kHz calling (Wright et al. 2012c). However, a NAergic contribution does not readily explain why DA receptor agonists and antagonists both decreased call rate. Another neuropharmacological difference between amphetamine/cocaine and direct DAergic agonists is that, according to recent in vivo voltammetric evidence, amphetamine and cocaine both enhance phasic DAergic signaling to an important degree (Cheer et al. 2007;Aragona et al. 2008;Ramsson et al. 2011a, b). Transient DA release events are known to occur spontaneously (Wightman and Robinson 2002;Schultz 2007), and their postsynaptic impact would likely be masked after administration of DA receptor agonists, antagonists, and their combination. Therefore, based on the present USV findings, we propose the hypothesis that 50-kHz vocalizations (or certain call subtypes) are driven by DA transients.
Behavioral significance of decreased 50-kHz call rate The relationship of 50-kHz calls with conventional measures of drug reward has been little explored. Specifically, the psychostimulants amphetamine and cocaine, after IP or SC administration, reliably produce CPP (0.5-2 mg/kg and 4-20 mg/kg, respectively; Tzschentke 1998) and acutely promote 50-kHz USVs (AMPH 0.5-2 mg/kg, cocaine 10-20 mg/kg; see above for references), whereas morphine can produce a CPP without a concomitant increase in unconditioned USV emissions (Wright et al. 2012a). The present study provides further evidence that unconditioned drug effects on 50-kHz call rate do not necessarily match the conditioned drug effects that are revealed in the CPP/CPA procedure. Our test drugs that decreased 50-kHz calling either (1) produced CPP or no effect (quinpirole; Hoffman and Beninger 1988;Graham et al. 2007), (2) produced CPA or no effect (SCH 39166; Acquas and Di Chiara 1994;Spina et al. 2006), or (3) produced CPP or CPA, even at the same dose (PD 128907; Khroyan et al. 1997;Gyertyan and Gal 2003). Lastly, no published CPP/CPA data appear available for A68930, PD 168077, L-741,626, L-745,870, or NGB 2904.
Several groups have proposed that 50-kHz calls may represent a behavioural expression of positive affect (Cuomo and Cagiano 1987;Knutson et al. 2002;Panksepp and Burgdorf 2003;Brudzynski 2007;Mallo et al. 2009;Barker et al. 2010;Browning et al. 2011;Burgdorf et al. 2011;Hamdani and White 2011). In the present study, most DAergic antagonists and agonists inhibited calling, and on this basis we speculate that a decrease in 50-kHz call rate may not necessarily reflect a negative shift in affect, but rather a response to an unfamiliar stimulus or context.

Limitations
Route of administration The present study demonstrated that systemically administered DAergic agonists and antagonists, given alone, decreased calling. Inhibition of calling by DAergic agonists indicates that DA receptor activation in multiple brain regions is insufficient to induce calling, whereas the inhibitory effects produced by DAergic antagonists suggest that DA receptors are necessary for USV emission. These conclusions run counter to findings from two studies based on intra-accumbens infusions of DAergic agents. In one study, the D2/D3 agonist quinpirole increased 50-kHz calling, while neither D2-nor D3-selective antagonists produced a significant effect (Brudzynski et al. 2012).
In the other study, the D1-and D2-like antagonists SKF 32957 and raclopride did not alter 50-kHz call rate, whereas the D2-like haloperidol increased it (Thompson et al. 2006). Taken together, these findings highlight the importance of route of administration and raise the possibility of both inhibitory and excitatory DAergic influences on 50-kHz call emissions.
Adverse drug effects High doses of A68930 (1.2 and 3.7 mg/kg) have been reported to trigger motor seizures in adult rats (DeNinno et al. 1991). However, no such effect was noted in several other studies (Salmi 1998;Salmi and Ahlenius 2000;Isacson et al. 2004;Nergardh et al. 2005), including at the two highest doses tested here (1 and 4 mg/ kg; D' Aquila et al. 1994;Deveney and Waddington 1997). Although we did not observe seizures in our rats, we cannot exclude the possibility that our rats suffered convulsions while in the testing chamber since they were not video recorded.
Pharmacology Each DA receptor subtype was probed with a single agonist and antagonist. However, these agents were chosen from the most target-selective available. To our knowledge, no agonists or antagonists currently discriminate between D1 and D5 receptors; for example, the D1-like agonist A68930 and antagonist SCH 39166 have near-equal in vitro affinities for D1 and D5 (Tice et al. 1994;Nergardh et al. 2005). In addition, at the start of each experiment, rats underwent an initial screen comprising three spaced injections of amphetamine; we cannot exclude the possibility that this amphetamine exposure affected subsequent calling to other DAergic agents.
Call-subtype analysis Call subtype analysis was not feasible in view of the low overall call rates (often less than four calls per minute) following receptor subtype-selective agonist and antagonist administration.

Conclusion
Overall, the literature is mixed with regard to the role of dopamine in ultrasonic vocalizations and whether these vocalizations may represent affective state. Here we have shown that following systemic drug administration, tonic pharmacological activation of dopamine receptors is not sufficient to increase 50-kHz vocalization call rates, whereas D1, D2, and D3 receptors may all be necessary for spontaneous calling. The observed drug effects require further investigation with respect to their neurochemical underpinnings (e.g., DA transients) and behavioural significance (e.g., interpretations not based on affect). Elucidation of the neurochemical events underpinning USV emission may provide a clearer understanding of the affective information that these vocalizations putatively convey.