Effects of apomorphine on mating behavior, flank marking and aggression in male hamsters

Pharmacology, Biochemistry and Behavior 101 (2012) 520–527

Contents lists available at SciVerse ScienceDirect

Pharmacology, Biochemistry and Behavior journal homepage: www.elsevier.com/locate/pharmbiochembeh

Effects of apomorphine on mating behavior, flank marking and aggression in male hamsters Molly M. Hyer, Laura M. Rycek, Owen R. Floody ⁎ Department of Psychology and Program in Animal Behavior, Bucknell University, Lewisburg, PA 17837, United States

a r t i c l e

i n f o

Article history: Received 29 September 2011 Received in revised form 19 February 2012 Accepted 26 February 2012 Available online 3 March 2012 Keywords: Aggression Apomorphine Dopamine Flank marking Hamster Male Scent marking Sexual behavior

a b s t r a c t In male rats, the dopamine agonist apomorphine (APO) generally facilitates copulatory behavior. However, disruptive effects of high APO doses have been reported. These have been interpreted in diverse ways, as products of a dopaminergic system that inhibits sexual behavior or as consequences of APO's stimulation of competing responses. To test the generality of these effects, we observed APO's impact on copulatory behavior in male hamsters. Several effects were observed, all attributable to a relatively high dose and involving the disruption of male behavior. More unexpectedly, APO treatment caused males to attack estrous stimulus females in the course of these tests. To clarify these effects, we observed the effects of APO on flank marking, a type of scent marking closely allied to aggression and dominance in hamsters. Treatment reliably decreased the latency of marking. It also increased the rate of marking when appropriate measures were taken to prevent this effect from being obscured by drug-induced cheek pouching. Together, these results confirm and extend APO's well-known ability to increase aggression. Further, they suggest that APO-induced aggression can intrude into other contexts so as to disrupt, or possibly facilitate, other forms of social behavior. © 2012 Elsevier Inc. All rights reserved.

1. Introduction Dopamine (DA) is widely considered to be one of the most influential neurotransmitters in the control of male-typical mating behavior (reviews in Bitran and Hull, 1987; Hull and Dominguez, 2007; Meisel and Sachs, 1994). This view is supported by many forms of evidence, including correlations of central dopaminergic activity with sexual stimulation or performance (Hull et al., 1993, 1995; Tsai et al., 2006), responses to central applications of dopaminergic drugs (Hull et al., 1986), and the effects of lesions specific to dopaminergic neurons (Bazzett et al., 1992). But some of the earliest and most important evidence of this type emerged from descriptions of the behavioral effects of systemic treatment with the nonselective DA receptor mimic apomorphine (APO). The effects of APO on male sexual behavior have been the focus of many studies (Agmo and Fernández, 1989; Arteaga et al., 2002; Butcher et al., 1969; Clark and Smith, 1987; Paglietti et al., 1978; Scaletta and Hull, 1990; Tagliamonte et al., 1974). These describe a preponderance of facilitory effects, justifying the common characterization of DA and dopaminergic systems as net facilitators of male behavior (e.g., Bitran and Hull, 1987; Hull and Dominguez, 2007; Meisel and Sachs, 1994). At the same time, these studies are limited in two respects. First, nearly all have focused on male rats, leaving the role

⁎ Corresponding author. Tel.: + 1 570 577 1432; fax: + 1 570 577 7007. E-mail address: ofl[email protected] (O.R. Floody). 0091-3057/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.pbb.2012.02.019

of DA in other species less clear. Second, studies of the responses to varying doses of APO have sometimes described both facilitory and disruptive effects. In many, but not all, cases these seem to assort by dose, with relatively low doses tending to facilitate and some higher doses tending to disrupt (e.g., Clark and Smith, 1987). These contrasting effects have led to some uncertainty regarding DA's role in male behavior (e.g., Bitran and Hull, 1987; Clark and Smith, 1987; Meisel and Sachs, 1994). To a large extent, this controversy has been resolved by the other forms of evidence described earlier. Nevertheless, the reasons for APO's sometimes contrasting behavioral effects are of interest and remain less than completely clear. These limitations could be related in the sense that uncertainty regarding some of APO's actions could reflect the concentration of past research on rats: The factors explaining these contrasting effects at the two ends of the dose–response curve could be clearer in other species. It was partly to test this possibility that we sought to describe the effects of systemic APO treatment on mating behavior in male hamsters. 2. Experiment 1: Apomorphine and copulatory behavior Male hamsters and rats share a basic copulatory pattern characterized by a single intravaginal thrust per intromission, multiple intromissions prior to ejaculation, and the potential for multiple ejaculations during a sexual interaction (Dewsbury, 1975). At the same time, male-typical sexual behavior in these species differs both behaviorally and neurochemically.

M.M. Hyer et al. / Pharmacology, Biochemistry and Behavior 101 (2012) 520–527

Behavioral differences are evident on many of the measures commonly used to describe male performance, including the latencies to mount, intromit and ejaculate (Dewsbury, 1979; Floody, 2011a; Pfaus et al., 1990; Sachs, 1978). Other differences emerge from the studies that have used factor analysis to describe patterns of interindividual correlation suggestive of basic processes underlying the behavior (Dewsbury, 1979; Floody, 2011a; Pfaus et al., 1990; Sachs, 1978). Though factors focused on the initiation and efficiency of copulation have been described in both hamsters and rats, these factors seem to be defined differently in the two species. Further, a factor emphasizing the rate of performance is a prominent feature in rats but seems absent from the factor structure of hamsters. On the neurochemical level, the role of acetylcholine (ACh) has been studied in both rats and hamsters. In rats, responses to systemic treatment with a muscarinic agonist such as oxotremorine suggest for ACh a quite specific role in the control of intromission frequency and ejaculation latency (e.g., Ahlenius and Larsson, 1985; RetanaMarquez et al., 1993). In contrast, similar treatments seem to produce much broader effects in hamsters, involving changes in most of the common measures of male behavior (Floody, 2011b). Though the data on cholinergic control raise the possibility of other species differences in neurochemical mechanisms for sexual behavior, the existing studies instead emphasize similar, facilitory, responses by male rats and hamsters to DA. Indeed, the one previous study of APO's effects on male behavior in hamsters seems to offer especially clear support for such effects (Arteaga et al., 2002). On the other hand, this study examined the impact of just one, relatively low dose (0.025 mg/kg), whereas most of the uncertainty regarding APO's effects revolves around its dose–response curve, and especially the tendency of low and high doses to sometimes produce opposite effects (e.g., Clark and Smith, 1987). This suggests that further work is required to describe the responses of male hamsters to a wider range of APO doses, such as used in the present study. 2.1. Methods 2.1.1. Animals and drug treatments The subjects were 18 adult male golden hamsters (Mesocricetus auratus, LVG:Lak outbred strain) that averaged 169.9 g in weight (SEM = 5.0) at the time of their first test. These were selected from a larger group of 25 on the basis of their successful completion of 2 screening tests requiring the achievement of ejaculation within 10 min of social contact. The experimental stimuli included 12 adult female hamsters, each of which was bilaterally ovariectomized approximately 3 months before the start of testing. Each animal was housed in a 34 × 18 × 18 or 31 × 21 × 21 cm stainless steel cage in a colony maintained at 20–25 °C and on a reversed 14:10 light:dark cycle. All had free access to food and water except during tests. All methods were approved by Bucknell University's Institutional Animal Care and Use Committee. Behavioral tests were conducted weekly. At 15 min before testing, each male received an intraperitoneal (ip) injection containing 0, 0.05 or 0.5 mg/kg of APO (apomorphine hydrochloride hemihydrate, Sigma-Aldrich, Inc) in a volume of physiological saline equal in ml to his body weight/1000. All APO solutions were prepared shortly before use. Each male was tested twice after exposure to each treatment. To determine treatments on the first 3 tests, the 6 possible orders of treatment were randomly assigned to subjects with the constraint that each be equally represented. For each subject, the second series of 3 tests duplicated the first. Tests were staged and scored without knowledge of the treatment. Each of the stimulus females was ovariectomized under sodium pentobarbital anesthesia (70 mg/kg, ip) supplemented by a subcutaneous (sc) injection of 0.4 mg of butorphanol tartrate (both from Henry Schein, Inc). To ensure sexual responsiveness, each female was primed with two sc injections of gonadal hormone in 0.05 ml of

521

peanut oil, the first at approximately 48 hr before use and containing 10 μg of estradiol benzoate and the second at 4–6 hr before use and containing 500 μg of progesterone (both from Steraloids, Inc). 2.1.2. Behavioral tests Each test began with the introduction of a male into a 40 × 20 × 25 cm glass aquarium. After 1–2 min of adaptation, a female was presented, the timing of the encounter beginning with the first social contact. Tests then normally continued through 2 copulatory series (2 ejaculations plus the first intromission thereafter). However, males sometimes failed to achieve this criterion. Specifically, some tests were terminated when males failed to intromit within 10 min of contact, failed to complete 2 copulatory series within 15 min, or upon the occurrence of a fight. Though encounters with fighting were terminated as quickly as possible, only obvious fights (e.g., as described in Floody and Pfaff, 1977) were scored as such: Males sometimes use relatively gentle bites to inspect or reposition females, but biting of this type was not mistaken for fighting. Fights between hamsters sometimes develop so quickly that it can be impossible to determine the instigator. However, most or all of the fights reported here were initiated by the male: The instigator could be identified in 10 of the 13 fights observed and was the male in each case. The data collected during each completed test included the timing of the first mount and intromission in each copulatory series, the timing of each ejaculation, and the total numbers of mounts and intromissions in each series. From these scores we derived each of the 14 dependent variables that typically would be used to describe male copulatory behavior in encounters of this length (e.g., Arteaga et al., 2002; Bunnell et al., 1977). This set includes 2 measures that are considered to initiate the interaction and so are not tied to a copulatory series, i.e., mount latency (ML, the delay between the initiation of contact and the first mount), and intromission latency (IL, the corresponding delay for the first intromission). The remaining 12 measures include 6 dependent variables, each of which is defined for each of the 2 copulatory series. These include ejaculation latency (the interval separating the first intromission of a series from the ejaculation that concludes that series, identified as EL-1 for the first series and EL-2 for the second), mount frequency (the number of mounts in a series; MF-1, MF-2), intromission frequency (the number of intromissions in a series; IF-1, IF-2), intromission ratio (the proportion of all mounts and intromissions in a series that are intromissions, or IF/(MF + IF) for the relevant series; IR-1, IR-2), interintromission interval (the average interval separating successive intromissions in a series, or EL/IF for the series; III-1, III-2), and postejaculatory interval (the interval separating the ejaculation of a focal series from the first intromission of the next series; PEI-1, PEI-2). Most of these measures were defined in standard ways (e.g., as in Arteaga et al., 2002; Bunnell et al., 1977). Our few departures from some earlier methods are detailed in Floody (2011a), along with evidence for the validity and reliability of the entire system of observation and scoring. Of special relevance to the present study is the recent successful use of this system in the analysis of cholinergic influences on male behavior in hamsters (Floody, 2011b; Floody et al., 2011). Dopaminergic agonists are well known to stimulate stereotyped behaviors. In hamsters, the most likely of these to appear over the present range of APO doses is reported to be gnawing (Schnur and Martinez, 1989). To reduce the chances of misinterpreting a change in sexual behavior due to the occurrence of an incompatible stereotyped act, each test cage was provided with a potential stimulus for gnawing (a wooden dowel, 5.6–6.0 cm long, 1.2 cm diameter) and the measures recorded during each test included the male's total duration of gnawing. 2.1.3. Analysis Our analyses distinguished the likelihood of failing a test from the quality of behavior on successful tests. Treatments were compared for

522

M.M. Hyer et al. / Pharmacology, Biochemistry and Behavior 101 (2012) 520–527

possible effects on the likelihood of failure due to inactivity, and on the incidence of fighting, using the Cochran Q and sign tests (Siegel, 1956). Effects on other measures were assessed by analysis of variance (ANOVA) after having averaged the results across the 2 days of testing at each APO dose. For the 2 measures that relate to an encounter as a whole rather than a specific copulatory series (ML, IL), ANOVA treated APO Dose as a within-subjects factor. For all other measures, the ANOVAs also included copulatory Series as a second withinsubjects factor. In this design, any drug effect that depended on copulatory series was revealed by a reliable (p ≤ .05) Dose × Series interaction; any drug effect that did not depend on series was revealed by a reliable Dose main effect. 2.2. Results 2.2.1. Effects on male copulatory behavior Though males occasionally failed tests due to sexual inactivity, the likelihood of such failures was not related to drug treatment. However, the combination of failures due to inactivity and fighting (next section) prevented us from collecting complete data (at least 1 completed test at each drug dose) on 3 males, forcing their exclusion from analyses of the quality of sexual performance. The analysis of the data from the remaining 15 males revealed reliable main effects of APO treatment on IR (F(2,28) = 4.91, p = .015), III (F(2,28) = 7.67, p = .002) and PEI (F(2,28) = 3.95, p = .031). The further examination of these effects suggested that the 0.5 mg/kg dose of APO increased III relative to both of the other treatments, and that it depressed IR and elevated PEI relative to just the 0.05 mg/kg dose of the same drug (each p b .05, Tukey test (Winer, 1971); see Figs. 1 and 2). In addition to these main effects, ANOVA revealed a reliable Dose × Series interaction affecting EL (F(2,28) = 4.43, p = .021). Further analysis suggested the restriction of the drug effect to the first copulatory series (i.e., to EL-1, for which F(2,28) = 5.53, p = .009). At this time, treatment with the 0.5 mg/kg dose of APO reliably

Fig. 2. Panel A describes the mean (and SEM) ejaculation latencies (ELs, in s) observed after treatment with APO doses of 0, 0.05 or 0.5 mg/kg. Panel B does the same for the postejaculatory interval (PEI, in s). In each case, scores in the first copulatory series are described by open symbols and those in the second series are described using filled symbols. The application of ANOVA to the data on PEI revealed a reliable Dose main effect (F(2,28) = 3.95, p = .031), attributable to consistently longer intervals after treatment with 0.5 mg/kg of APO than after 0.05 mg/kg of this drug (p b .05, Tukey test, 2-tailed). The analysis of EL scores revealed a reliable Dose × Series interaction (F(2,28) = 4.43, p = .021). This reflects a drug effect confined to the first copulatory series (and EL-1) and involving the elevation of latencies by 0.5 mg/kg of APO relative to each of the other treatments (each p b .05, Tukey test, 2-tailed).

increased EL relative to both of the other treatments (p b .05, Tukey test, Fig. 2). Though the focus of this report is on responses to APO treatment, the performance of subjects also varied as a function of copulatory series. The modulation by series of APO's impact on EL has already been described. This interaction reflects levels of EL that were much greater in the first copulatory series regardless of drug treatment (each p b .01, Tukey test, Fig. 2). For each of the other measures that distinguished copulatory series, this variable affected performance independently of drug treatment, resulting in reliable Series main effects (F(1,14) ≥ 6.67, p ≤ .022). In nearly every case, the quality of performance increased over series, signaled by reliable decreases in MF, IF, and III, and by a reliable increase in IR (Table 1). The one exception was PEI, which increased slightly but consistently over series (Table 1). 2.2.2. Effects on other behaviors Average rates of gnawing were very low (≤0.2 s/min of observation) and failed to differ reliably across drug treatments. However, the likelihood of fighting clearly was drug related. Whereas no male failed a test due to fighting after placebo treatment, 11% failed at least one test for this reason after treatment with 0.05 mg/kg of APO Table 1 Mean (and SEM) levels of male behavior across copulatory series.

Fig. 1. Panel A describes the mean (and SEM) intromission ratios (IRs) observed after treatment with APO doses of 0, 0.05 or 0.5 mg/kg. Panel B does the same for interintromission interval (III, in s). In each case, scores in the first copulatory series are described by open symbols and those in the second series are described using filled symbols. The application of ANOVA to these data revealed reliable Dose main effects for each measure (F(2,28) ≥ 4.91, p ≤ .015). In the case of IR, this reflects reliably lower ratios after treatment with 0.5 mg/kg of APO than after 0.05 mg/kg of this drug (p b .05, Tukey test, 2-tailed). For III, the main effect reflects a reliable elevation in scores by the higher dose of APO relative to each of the other treatments (each p b .05, Tukey test, 2-tailed).

Measure

Series 1

Series 2

MF IF IR III EL PEI

2.3 7.5 0.78 9.4 68.5 23.8

0.1 1.6 0.97 8.4 13.6 26.6

(0.3) (0.5) (0.02) (0.4) (4.6) (1.2)

(0.0) (0.1) (0.01) (0.4) (1.8) (1.3)

Notes: For each of these measures, ANOVA revealed a reliable main effect of copulatory series (F(1,14) ≥ 6.67, p ≤ .022). In the case of EL, this was in addition to a reliable Dose × Series interaction that is depicted more fully in Fig. 2.

M.M. Hyer et al. / Pharmacology, Biochemistry and Behavior 101 (2012) 520–527

and 50% did so after treatment with the 0.5 mg/kg APO dose (Q(2) = 6.70, p b .05, Cochran test, 2-tailed). Further analysis confirmed that this effect is due to the higher APO dose, which led to more frequent fights (instances of males attacking estrous females) than either of the other treatments (each p ≤ .04, sign tests, 2-tailed). As suggested by its association with test failure, fighting was negatively correlated with successful mating: In no encounter terminated due to fighting did the male achieve intromission or ejaculation. Further, this is not because fighting appeared immediately, limiting the time available for copulation: The average duration of encounters terminated by fights was comparable to that of successfully completed encounters (227 and 191 s, respectively).

2.3. Discussion These results document two consequences of exposing male hamsters to relatively high doses of APO. First, such treatments can disrupt male copulatory behavior. This effect is consistent with the ability of similar APO doses to disrupt or prevent copulation in male rats (Clark and Smith, 1987; also see Butcher et al., 1969; Paglietti et al., 1978). Second, our tests revealed an unexpected ability of APO to cause males to attack their estrous female partners. Though APO is well known to stimulate aggression (e.g., Kask and Harro, 2000; Matto et al., 1998; Skrebuhhova-Malmros et al., 2000), we are not aware of any previous report of APO-induced aggression intruding into copulatory interactions. However, such an effect may be more likely, or evident, in a species in which male–female fights are relatively common and violent (Blanchard et al., 1988; Johnston, 1975; Payne and Swanson, 1971, 1972b). These effects could be independent and their combination coincidental. However, such an interpretation seems unable to provide a plausible account of the disruption in copulation. Some such effects have been interpreted as indirect products of APO-induced increases in competing stereotyped behaviors such as gnawing or sniffing (Butcher et al., 1969; Paglietti et al., 1978; Tagliamonte et al., 1974). But such an explanation of the present results seems inappropriate. The most likely stereotyped behavior (Schnur and Martinez, 1989) was monitored and found to occur at levels that were low and unrelated to drug treatment. An alternative approach is to argue that postsynaptic DA inhibits sexual behavior, and that facilitory effects of low APO doses reflect decreases in dopaminergic transmission due to the selective activation of inhibitory autoreceptors (Clark and Smith, 1987). This interpretation also seems unlikely. Most of the evidence suggesting selective actions of APO on subtypes of DA receptors seems to have been based on inappropriate behavioral models or drugs later found to affect multiple neurotransmitter systems (Bitran and Hull, 1987). Alternatively, it could be argued that the effects reported here are related, and that a direct effect of APO on one behavior or motivational state indirectly affected the other. Further, both variants of this position seem plausible. It seems clear that excessive or inappropriately directed aggressive behavior could disrupt sexual interaction. But the failure to achieve expected rewards is widely viewed as having the potential to increase aggression (brief reviews in Cunningham and McGinnis, 2007; Miczek et al., 2003). In a recent study of such “frustration-induced aggression,” it was found that disrupting copulation by blocking access to the vagina did not affect aggression in peripubertal male rats (Cunningham and McGinnis, 2007). However, this does not exclude the possibility of increased aggression due to other manipulations or in other subjects. Such effects may be especially likely in animals predisposed to aggression, as seems true of hamsters (Blanchard et al., 1988; Johnston, 1975; Payne and Swanson, 1971, 1972a,b). Such an interpretation of the fighting observed here is consistent with the apparent inability of the males in question to achieve intromission or ejaculation.

523

In order to decide among these alternatives, it would be helpful to know more about APO's effects on aggression in hamsters. In particular, it would be useful to know if APO can increase aggression in male hamsters outside of a sexual context. The remaining studies address this issue by examining the effects of APO on flank marking (Johnston, 1975), a type of scent marking that correlates directly with aggression and dominance in hamsters. 3. Experiment 2: Apomorphine and flank marking I Uncertainty regarding the interpretation of the earlier results suggests an incomplete understanding of dopaminergic systems. But it also may reflect the complexity of standard mating tests. Each of the animals paired in such tests may present a variety of cues with the potential to affect a corresponding variety of motivational or behavioral systems (Floody et al., 1977; Lehman et al., 1983; Petrulis and Johnston, 1997). For example, it seems easy to imagine a primary effect of APO on either sexual or aggressive motivation partly because there is reason to think that either could build upon a natural reaction of male hamsters to stimulus females (Floody and Pfaff, 1977; Lehman et al., 1983; Payne and Swanson, 1971, 1972a,b). Consistent with this problem, one goal of the remaining studies was to focus on situations, stimuli and behaviors that are relatively simple, and that relate closely and simply to specific motivational states. This was not done in the expectation of identifying a perfect stimulus or measure. Instead, it was driven by the desire to improve upon a stimulus and set of measures that may be especially complex. To combine these goals with that of clarifying the relationship between APO treatment and aggression, one appropriate system seems to be that revolving around the flank marking that male hamsters exhibit upon exposure to the odors of other males. As the term implies, flank marking involves the deposition of the product of a sebaceous flank gland as an animal rubs its flank against some part of the environment, usually a wall or corner (Johnston, 1975). Rates of marking fluctuate in a way suggesting a direct, though not perfect, correlation with levels of aggression and dominance. Prior to the establishment of dominance, marking by a male can be increased by exposure to a potential source or target of aggression, such as a nonestrous female or second male (Johnston, 1975). During the establishment of dominance, rates of marking are more variable and may be low in both competitors. However, once dominance has been established, marking rates consistently are higher in dominant males than in subordinates. On the basis of such evidence, flank marking has been widely recognized as a form of agonistic behavior (Ferris et al., 1986; Hayden-Hixson and Ferris, 1991; Johnston, 1975; Melloni et al., 2001) and has been used as an indicator of drug effects on aggression and dominance (Ferris et al., 1986; Melloni et al., 2001). Rates of flank marking are highly responsive to vasopressin (AVP) or its antagonists, especially when delivered in or near the medial preoptic area (MPOA; e.g., Ferris et al., 1984, 1985). Rates also have been reported to depend on other neurotransmitters or neuropeptides, including glutamate, serotonin and galanin (Bamshad et al., 1996; Ferris and Delville, 1994; Ferris et al., 1999). However, little evidence is available on the responsiveness of flank marking to dopaminergic manipulations. Whitman et al. (1992; also see Ferris et al., 1985) found it to be unaffected by dopamine infusions into the MPOA, leaving open the possibility of responses to treatments elsewhere. Melloni et al. (2001) studied the effects of chronic cocaine treatment on marking by adolescent hamsters. But the effects they observed were restricted to females and attributed to changes in vasopressin or noradrenergic, rather than dopaminergic, activity. In view of this gap, along with the possibility that effects on flank marking could clarify those on aggression more generally, we have examined the effects on marking of APO treatments similar to those that disrupted sexual behavior and increased fighting in Experiment 1.

524

M.M. Hyer et al. / Pharmacology, Biochemistry and Behavior 101 (2012) 520–527

3.1. Methods 3.1.1. Animals and treatments Data were collected from 18 adult male hamsters that averaged 174.8 g (SEM = 4.3) at the time of their first tests. All were housed, maintained and treated much as in the preceding study. In particular, each received an ip injection containing 0, 0.25 or 0.5 mg/kg of APO at 15 min before each test. These solutions were prepared shortly before use. Each male was tested twice after exposure to each treatment with the order of exposure counterbalanced as in the first study. Injections were given and tests were staged and scored without knowledge of the drug treatment. Successive tests were separated by intervals of approximately 48 hr. 3.1.2. Behavioral tests Flank marking was defined and scored essentially as described by Johnston (1975). We deviated from those methods only by collapsing Johnston's categories of “flank marks” (defined by movement with physical contact between the flank and substrate) and “intention marks” (defined by comparable movements but with some separation of flank and substrate). However, marks of the latter type were rare and so are unlikely to have had more than a negligible effect on the results. Several other aspects of the methods used to elicit and measure flank marking were based on Johnston's (1975) suggestion that this behavior can be facilitated by familiar odors. First, observations of marking began by randomly pairing subjects and then maintaining these pairings over the remainder of a test series. Second, tests of drug effects on marking were preceded by approximately 9 pretests at intervals of 2–3 days. These sought to increase and stabilize rates of marking. Each was identical to a standard test except for the absence of drug treatment. Third, we did not manipulate the contents of a male's home cage at any point during testing. Therefore, tests exposed males both to substrate-borne odors and to any stimuli associated with the food and bedding of the test cage's normal resident. Each test of marking began with a male's removal from his home cage and transfer into the home cage of his assigned partner. This exchange initiated a 10-min period during which we recorded the latency of flank marking (in s from a male's introduction into the test cage), the total number of marks (converted into marks/min), and the incidence of cheek pouching (the introduction of food or bedding into both cheek pouches; see Hoffman et al., 1968). 3.1.3. Analysis Possible effects on flank marking were assessed by ANOVA, treating APO Dose (0, 0.25, 0.5 mg/kg) as a within-subjects factor. As necessary, logarithmic transformations were used to reduce heterogeneity of variance and reliable effects (p≤ .05) were clarified using the Tukey test (Winer, 1971). Effects on the incidence of cheek pouching were assessed using the Cochran Q and sign tests (Siegel, 1956). 3.2. Results Analyses of the latency to flank mark were restricted to the 15 subjects that marked on at least 1 test at each APO dose. The application of ANOVA to these data revealed a highly reliable Dose effect (F(2,28) = 19.16, p b .001; Fig. 3A). Further analysis confirmed that this effect is due to the significant reduction of latency by each drug dose (each p b .05, Tukey test, 2-tailed). An initial analysis of flank marking rate included all animals and tests. This revealed no reliable drug effect (Fig. 3B). However, in view of data suggesting the disruption of flank marking by APOinduced cheek pouching, rates of marking also were determined for the 8 males with at least 1 test without pouching at each APO dose. This analysis revealed a reliable Dose effect (F(2,14) = 6.17, p = .012), attributable to a rate after the injection of 0.25 mg/kg of

Fig. 3. Panel A describes the mean (and SEM) latencies to flank mark (s) after treatment with APO doses of 0, 0.25 and 0.5 mg/kg. The analysis of these data revealed a Dose main effect (F(2,28) = 19.16, p b .001) due to the differences between the placebo and other doses (each p b .05, Tukey test, 2-tailed). Panel B describes the mean (and SEM) rates of flank marking (marks/min) observed in the same study on all tests (filled squares interconnected by solid lines, N = 18) and tests without cheek pouching (open circles interconnected by dashed lines, N = 8). The analysis of the latter revealed a reliable Dose main effect (F(2,14) = 6.17, p = .012) due to the difference between the 0 and 0.25 mg/kg treatments (p b .05, Tukey test, 2-tailed).

APO that reliably exceeded that after placebo treatment (p b .05, Tukey test, 2-tailed; Fig. 3B). As suggested just above, APO treatment also reliably affected the likelihood of cheek pouching. Specifically, 33.3, 77.8 and 61.1% of males, respectively, cheek pouched on at least 1 test at the APO doses of 0, 0.25 and 0.5 mg/kg (Q(2) = 9.80, p b .05, Cochran test, 2-tailed). Further analysis confirmed that the incidence of cheek pouching was reliably increased by each APO dose (each p b .05, sign test, 2-tailed). 3.3. Discussion These results clearly support the ability of APO to reduce the latency to flank mark and increase the incidence of cheek pouching. They also suggest a drug-related increase in the extent of flank marking. These effects seem related in several respects. The facilitation of cheek pouching reflected a combination of APO treatment and exposure to another male's food and bedding: Under these test conditions, pouching only occurred after males had been transferred, never during the 15-min interval between APO treatment and the transfer that initiated the behavioral test. This delay in pouching may have created a window of time in which some marking was possible, thereby helping to reveal the drug effect on latency to mark. Once pouching had occurred, however, it seems to have prevented further marking, thus possibly obscuring a drug effect on marking rate, at least in the affected tests. Consistent with this suggestion is the fact that a drug effect on marking rate did emerge from an analysis focused on tests without pouching. Other evidence for this incompatibility includes the simple fact that an animal with his pouches full never was observed to mark. Further, in a pilot study in which cheek pouching sometimes did occur prior to the exchange of males, the likelihood of failing to flank mark altogether was consistently greater on tests with pouching than on those without this response (p = .022, sign test, 2-tailed, data not shown). The net effect of these results is to suggest the facilitation by DA of both flank marking and cheek pouching. The first of these extends in

M.M. Hyer et al. / Pharmacology, Biochemistry and Behavior 101 (2012) 520–527

several respects a description of increased flank marking in adolescent female hamsters exposed to repeated cocaine treatments (Melloni et al., 2001). In particular, the present data show that marking also can be increased in adult male hamsters, and by an acute treatment that may be more specific to DA. At the same time, the effects on marking seem inconsistent with work suggesting that flank marking is unaffected by DA (Ferris et al., 1985; Whitman et al., 1992). However, the treatments in this work were applied directly to the MPOA and so do not exclude the possibility of a role for DA elsewhere. In turn, these effects on flank marking support the ability of APO to increase levels of aggression in male hamsters. It could instead be argued that APO acts directly on the motor program for marking without producing any motivational change. However, we are aware of no evidence suggesting such a direct effect of DA on marking (Whitman et al., 1992) whereas the interpretation in terms of aggression is supported by APO's ability to increase other indices of aggression, both here (Experiment 1) and in many earlier studies (e.g., Kask and Harro, 2000; Matto et al., 1998; Skrebuhhova-Malmros et al., 2000). Previous studies of cheek pouching in hamsters suggest that this can be affected by dopaminergic drugs, but that the effects depend on a variety of factors including test conditions. The results of Poignant and Rismondo (1975) suggest that APO doses of 2–4 mg/kg cause males to cease pouching food and to instead collect bedding: It is unclear if this change in substrate also entailed one in the total amount of pouching. Chester et al. (2006) seem to have created a situation in which pouching by females was focused on newly introduced food pellets. They found this food pouching to be facilitated by repeated injections of quinpirole, a D2-like DA receptor agonist. Taken together, these results seem consistent with our finding of increased cheek pouching in male hamsters given access to another male's food and bedding. Our results also are consistent with several reports suggesting the dopaminergic control of food hoarding in rats or gerbils. For example, the extent of hoarding in gerbils seems to correlate directly with levels of dopaminergic activity in the ventral tegmental area (Yang et al., 2011). Conversely, hoarding in rats can be decreased by lesions or drug treatments that should reduce dopaminergic activity (Blundell et al., 1977; Kelley and Stinus, 1985; Stinus et al., 1978). Further, such decrements sometimes can be reversed by neural grafts or other treatments with the potential to increase dopaminergic activity (Herman et al., 1986; Kelley and Stinus, 1985).

525

significant impact on the results. All animals were housed and maintained as in the preceding studies. Subjects also were treated much as before, but focusing on the more extreme drug doses. In particular, each received an ip injection containing 0 or 0.5 mg/kg of APO at 15 min before each test. These solutions were prepared shortly before use. Each male was tested twice after exposure to each treatment with the order of exposure counterbalanced much as in the previous studies. Injections were given and tests were staged and scored without knowledge of the drug treatment. Successive tests were separated by intervals of approximately 48 hr. With two exceptions, each test was identical to those in Experiment 2. First, to prevent any distortion of drug effects on marking by cheek pouching, test cages were emptied as males were transferred. To minimize any disruption in a male's home environment and later behavior, these materials were returned to the appropriate cages upon the conclusion of each test. Second, to guard against the possibility that the preceding study missed a short-lived drug effect on marking, rates of marking here were determined for each in a succession of five 2-min blocks. Possible effects on the latency of flank marking were assessed using paired t-tests. Those on the frequency of marking were assessed by ANOVA, treating APO Dose (0, 0.5 mg/kg) and test Block as withinsubjects factors. Reliable effects (p ≤ .05) were clarified as necessary using the Tukey test (Winer, 1971). 4.2. Results The analysis of flank marking latencies revealed a highly reliable difference (t(11) = 3.28, p = .007, 2-tailed), reflecting consistently lower latencies after APO treatment (Fig. 4A). The application of ANOVA to the rates of flank marking revealed reliable main effects of Dose (F(1,11) = 4.80, p = .051) and Block

4. Experiment 3: Apomorphine and flank marking II Though the previously described effect of APO on the latency to flank mark was clear, there was no reliable effect on the extent of marking when all animals were included in the analysis. We have explained why we think that this is due to the interruption of marking by cheek pouching. However, it could instead reflect a drug effect of short duration, affecting marking rate over just a small fraction of a test. Some support for this possibility is provided by the substantial decrease in extent of gnawing observed by Schnur and Martinez (1989) over the second 10-min interval following the treatment of female hamsters with 0.1 mg/kg of APO. Therefore, to better define the relationship between APO and marking, we conducted another study of this response, in which rates of marking were tracked in greater detail and animals were denied the opportunity to engage in potentially disruptive cheek pouching. 4.1. Methods Data were collected from 12 adult male hamsters that averaged 171.3 g (SEM = 4.4) at the time of their initial tests. These included 8 of the subjects from Experiment 2 and 4 naive males. However, the results of preliminary analyses suggest that this variable had no

Fig. 4. Panel A describes the mean (and SEM) latencies to flank mark (s) observed after treatment with APO doses of 0 and 0.5 mg/kg. The analysis of these data revealed consistently shorter latencies after APO than control treatment (t(11) = 3.28, p = .007, 2-tailed). Panel B describes the mean (and SEM) rates of flank marking (marks/min) observed in successive 2-min blocks after treatment with APO doses of 0 (filled squares interconnected by solid lines) or 0.5 (open circles interconnected by dashed lines) mg/kg. The analysis of these data revealed reliable main effects of Dose (F(1,11) = 4.80, p = .051) and Block (F(4,44) = 32.99, p b .001). The first reflects the higher marking rates observed after APO treatment whereas the second reflects the decline in rate over the first several blocks.

526

M.M. Hyer et al. / Pharmacology, Biochemistry and Behavior 101 (2012) 520–527

(F(4,44) = 32.99, p b .001). The latter clearly reflects a gradual decrease in marking rate, especially over the first 6 min of exposure to another male's home cage (Fig. 4B). The dose effect reflects consistently higher rates of marking after APO than after control (placebo) treatment (Fig. 4B). 4.3. Discussion These results confirm that APO can significantly reduce flank marking latency in male hamsters. They also strengthen the case for an allied drug effect on the rate of marking. Though this rate did decline over the 10-min test period, this seems due to factors (e.g., habituation to the resident's odors or an accumulation of the test male's odors) other than a rapid reduction in drug action, as the pattern of the change in rate was similar after treatment with APO or placebo. To the contrary, the near constancy of the difference in marking rate across treatments suggests that the effect of APO was both significant and stable over the entire test period (Fig. 4). Taken together, the results of this and the preceding study support the facilitation of flank marking by APO, in turn strongly suggesting the facilitation of marking by DA. As mentioned earlier, this supports and extends previous observations of cocaine-induced changes in marking by young female hamsters (Melloni et al., 2001). Similarly, the inconsistency of these results with some past reports (Ferris et al., 1985; Whitman et al., 1992) seems understandable in terms of very different methods of drug administration: The earlier work focused on responses to DA applied directly to the MPOA and so does not exclude effects of systemic treatments with the potential to affect many brain areas. It should not be assumed that all drug effects on flank marking are mediated by changes in aggression. For example, the power and immediacy of AVP's stimulation of marking suggest a motor, not motivational, effect (Ferris et al., 1984). Consistent with this are results showing that the increases in marking provoked by AVP treatment are not accompanied by changes in overt aggression in either subordinate or dominant males (Ferris et al., 1986). In contrast, the effects of APO on flank marking do seem likely to reflect a more general change in aggression. As previously discussed, rates of marking correlate directly, though not perfectly, with levels of aggression and dominance (Johnston, 1975). Such evidence has led to the recognition of flank marking as a form of agonistic behavior (Ferris et al., 1986; Hayden-Hixson and Ferris, 1991; Johnston, 1975; Melloni et al., 2001) and to its use as an indicator of drug effects on aggression and dominance (Ferris et al., 1986; Melloni et al., 2001). Further evidence in support of this interpretation of APO's effect on marking is evident in this drug's ability to stimulate overt aggression, both in previous studies of rats (Kask and Harro, 2000; Matto et al., 1998; Skrebuhhova-Malmros et al., 2000) and in the male hamsters observed in our first study. The ability of APO to cause these changes and to facilitate marking seems best explained by a drug effect on the most likely common denominator of increased aggression. 5. Conclusions The results of these studies document the ability of APO to stimulate overt aggression that may be especially impressive by virtue of its direction at the unlikely target of a sexually-receptive estrous female. In addition, they establish APO's ability to facilitate a scent-marking behavior closely allied to aggression and dominance. Taken together, these results strongly support the stimulation of aggression in male hamsters by APO and DA. This conclusion reinforces and extends a widely held view of one of the many motivational or behavioral systems influenced by DA. But it may have especially interesting implications for interactions of the sort that initiated this series of studies, male–female

interactions that normally would be categorized as sexual, not aggressive, in character. For example, they suggest that drug-induced declines in sexual performance sometimes may reflect the intrusion of elevated aggression, and a resulting competition between forms of social interaction that often are thought of as mutually exclusive. Such an effect seems highly likely to explain the changes in sexual behavior observed in Experiment 1 and could possibly help to account for the disruptive effects of high doses of APO that sometimes have been reported in studies of the dopaminergic control of sexual behavior in male rats (e.g., Clark and Smith, 1987). But is this the only way in which sexual and aggressive motivation can interact? Might there be conditions under which an APO-induced increase in aggression could facilitate, rather than disrupt, sexual behavior? This question may relate to the contrast between the disruption of sexual behavior by 0.5 mg/kg of APO in Experiment 1 and the clear facilitation of this behavior by 0.025 mg/kg of the same drug reported by Arteaga et al. (2002). In several unpublished studies, we have tried to replicate the latter effect. In combination, these have yielded some data suggesting a facilitation of male behavior by 0.025 mg/kg of APO. However, even these positive effects pale in comparison to those described by Arteaga et al. (2002). In the course of our efforts to understand this discrepancy, we have considered several methodological differences across the relevant studies. One of these relates to the ages of the subjects: Arteaga et al. (2002) used males that were much smaller, and probably younger, than the subjects in most of our studies. However, in one unpublished study we explicitly tested the effects of APO treatments of 0, 0.025 and 0.1 mg/kg on both young (approximately 3.5 month old) and very old (approximately 21.5 months) male hamsters. We found several age effects (lower levels of IF, EL and PEI in the younger males, suggesting more rapid and efficient performance) but no interactions of age and drug treatment. The other major methodological difference that we see between our studies and that of Arteaga et al. (2002) relates to the ways in which subjects were housed. Whereas we study animals that have lived in isolation (individual cages) since about the time of weaning, Arteaga et al. (2002) studied group-housed males. This seems potentially relevant in light of previous reports of higher levels of aggression (Onyekwere and Ramírez, 1994) and responsiveness to the aggression-promoting effects of some hormone treatments (Grelk et al., 1974) in isolated males. This raises the possibility that effects of APO on aggression could facilitate or disrupt sexual behavior as a function of baseline levels of aggression. It seems likely that some level of aggression is required for social interaction, especially with a potentially threatening social partner, a category that, for a male hamster, may include conspecific females (Lehman et al., 1983). If this is true, then a drug-induced increase in aggression might facilitate sex behavior if it is not excessive in itself, and is combined with a low baseline level of aggression, as seems likely in the case of a group-housed male hamster. Conversely, it seems possible that APO-induced aggression would be most likely to disrupt sex behavior if the increase is large or superimposed on an already high level of aggression, as in the case of a hamster raised in isolation. In addition to its possible relevance to the effects of APO on hamster behavior, this reasoning could have implications for other studies, including those of rats and other species. APO clearly can increase aggression in rats (e.g., Kask and Harro, 2000; Matto et al., 1998; Skrebuhhova-Malmros et al., 2000) but has never, to our knowledge, been accused of causing males to fail tests of copulation. In some cases, this could reflect levels of aggression that fall short of those required to trigger attacks but that still could be disrupting sexual behavior to a degree. In the majority of cases, however, the absence of any perceived problem may reflect levels of aggression that facilitate, rather than disrupting, the sexual interaction. In such a case, APO could still be facilitating sexual behavior, but possibly not in the generally expected way.

M.M. Hyer et al. / Pharmacology, Biochemistry and Behavior 101 (2012) 520–527

Acknowledgments Thanks to the Bucknell PSYC 290 classes of 2008–2009 for help in the collection of some of these data. References Agmo A, Fernández H. Dopamine and sexual behavior in the male rat: a reevaluation. J Neural Transm 1989;77:21–37. Ahlenius S, Larsson K. Central muscarinic receptors and male rat sexual behavior: facilitation by oxotremorine but not arecoline or pilocarpine in methscopolamine pretreated animals. Psychopharmacology 1985;87:127–9. Arteaga M, Motte-Lara J, Velázquez-Moctezuma J. Effects of yohimbine and apomorphine on the male sexual behaviour pattern of the golden hamster (Mesocricetus auratus). Eur Neuropsychopharmacol 2002;12:39–45. Bamshad M, Cooper TT, Karom M, Albers HE. Glutamate and vasopressin interact to control scent marking in Syrian hamsters (Mesocricetus auratus). Brain Res 1996;731:213–6. Bazzett T, Lumley L, Bitran D, Markowski V, Warner R, Hull E. Male rat copulation following 6-OHDA lesions of the medial preoptic area: resistance to repeated administration and rapid behavioral recovery. Brain Res 1992;580:164–71. Bitran D, Hull EM. Pharmacological analysis of male rat sexual behavior. Neurosci Biobehav Rev 1987;11:365–89. Blanchard RJ, Flannelly KJ, Blanchard DC. Life-span studies of dominance and aggression in established colonies of laboratory rats. Physiol Behav 1988;43:1–7. Blundell JE, Strupp BJ, Latham CJ. Pharmacological manipulation of hoarding: further analysis of amphetamine isomers and pimozide. Physiol Psychol 1977;5:462–8. Bunnell BN, Boland BD, Dewsbury DA. Copulatory behavior of golden hamsters (Mesocricetus auratus). Behavior 1977;61:180–206. Butcher LL, Butcher SG, Larsson K. Effects of apomorphine, (+)-amphetamine, and nialamide on tetrabenazine-induced suppression of sexual behavior in the male rat. Eur J Pharmacol 1969;7:283–8. Chester JA, Mullins AJ, Nguyen CH, Watts VJ, Meisel RL. Repeated quinpirole treatments produce neurochemical sensitization and associated behavioral changes in female hamsters. Psychopharmacology 2006;188:53–62. Clark JT, Smith ER. Effects of apomorphine on sexual behavior in young and middleaged rats. Neurobiol Aging 1987;8:153–7. Cunningham RL, McGinnis MY. Factors influencing aggression toward females by male rats exposed to anabolic steroids during puberty. Horm Behav 2007;51:135–41. Dewsbury DA. Diversity and adaptation in rodent copulatory behavior. Science 1975;190:947–54. Dewsbury DA. Factor analysis of measures of copulatory behavior in three species of muroid rodents. J Comp Physiol Psychol 1979;93:868–78. Ferris CF, Delville Y. Vasopressin and serotonin interactions in the control of agonistic behavior. Psychoneuroendocrinology 1994;19:593–601. Ferris CF, Albers HE, Wesolowski SM, Goldman BD, Leeman SE. Vasopressin injected into the hypothalamus triggers a sterotypic behavior in golden hamsters. Science 1984;224:521–3. Ferris CF, Pollock J, Albers HE, Leeman SE. Inhibition of flank-marking behavior in golden hamsters by microinjection of a vasopressin antagonist into the hypothalamus. Neurosci Lett 1985;55:239–43. Ferris CF, Meenan DM, Axelson JF, Albers HE. A vasopressin antagonist can reverse dominant/subordinate behavior in hamsters. Physiol Behav 1986;38:135–8. Ferris CF, Delville Y, Bonigut S, Miller MA. Galanin antagonizes vasopressin-stimulated flank marking in male golden hamsters. Brain Res 1999;832:1–6. Floody OR. Organization of mating behavior in male hamsters (Mesocricetus auratus). J Comp Psychol 2011a;125:185–93. Floody OR. Cholinergic control of male mating behavior in hamsters: effects of systemic agonist or antagonist treatment. Pharmacol Biochem Behav 2011b;100:289–98. Floody OR, Pfaff DW. Aggressive behavior in female hamsters: the hormonal basis for fluctuations in female aggressiveness correlated with estrous state. J Comp Physiol Psychol 1977;91:443–64. Floody OR, Pfaff DW, Lewis CD. Communication among hamsters by high-frequency acoustic signals: II. Determinants of calling by females and males. J Comp Physiol Psychol 1977;91:807–19. Floody OR, Katin MJ, Harrington LX, Schassburger RL. Cholinergic control of male mating behavior in hamsters: effects of central oxotremorine treatment. Pharmacol Biochem Behav 2011;100:299–310. Grelk DF, Papson BA, Cole JE, Rowe FA. The influence of caging conditions and hormone treatments on fighting in male and female hamsters. Horm Behav 1974;5:355–66. Hayden-Hixson DM, Ferris CF. Steroid-specific regulation of agonistic responding in the anterior hypothalamus of male hamsters. Physiol Behav 1991;50:793–9. Herman JP, Choulli K, Geffard M, Nadaud D, Taghzouti K, Le Moal M. Reinnervation of the nucleus accumbens and frontal cortex of the rat by dopaminergic grafts and effects on hoarding behavior. Brain Res 1986;372:210–6. Hoffman RA, Robinson PF, Magalhaes H. The golden hamster; its biology and use in medical research. Ames, Iowa: Iowa State University Press; 1968. Hull EM, Dominguez JM. Sexual behavior in male rodents. Horm Behav 2007;52:45–55. Hull EM, Bitran D, Pehek EA, Warner RK, Band LC, Holmes GM. Dopaminergic control of male sex behavior in rats: effects of an intracerebrally-infused agonist. Brain Res 1986;370:73–81.

527

Hull EM, Eaton RC, Moses J, Lorrain D. Copulation increases dopamine activity in the medial preoptic area of male rats. Life Sci 1993;52:935–40. Hull EM, Du J, Lorrain DS, Matuszewich L. Extracellular dopamine in the medial preoptic area: implications for sexual motivation and hormonal control of copulation. J Neurosci 1995;15:7465–71. Johnston RE. Scent marking by male golden hamsters (Mesocricetus auratus): I. Effects of odors and social encounters. Z Tierpsychol 1975;37:75–98. Kask A, Harro J. Inhibition of amphetamine- and apomorphine-induced behavioural effects by neuropeptide Y Y1 receptor antagonist BIBO 3304. Neuropharmacology 2000;39:1291–302. Kelley AE, Stinus L. Disappearance of hoarding behavior after 6-hydroxydopamine lesions of the mesolimbic dopamine neurons and its reinstatement with L-dopa. Behav Neurosci 1985;99:531–45. Lehman MN, Powers JB, Winans SS. Stria terminalis lesions alter the temporal pattern of copulatory behavior in the male golden hamster. Behav Brain Res 1983;8:109–28. Matto V, Allikmets L, Skrebuhhova T. Apomorphine-induced aggressiveness and [3H] citalopram binding after antidepressant treatment in rats. Pharmacol Biochem Behav 1998;59:747–52. Meisel RL, Sachs BD. The physiology of male sexual behavior. In: Knobil E, Neill JD, editors. The physiology of reproduction. 2nd edition. New York: Raven Press; 1994. p. 3-105. Vol 2. Melloni RH, Connor DF, Todtenkopf MS, DeLeon KR, Sanyal P, Harrison RJ. Repeated cocaine treatment activates flank marking in adolescent female hamsters. Physiol Behav 2001;73:561–70. Miczek KA, Fish EW, DeBold JF. Neurosteroids, GABAA receptors, and escalated aggressive behavior. Horm Behav 2003;44:242–57. Onyekwere D, Ramírez JM. Influence of timing of post-weaning isolation on play fighting and serious aggression in the male golden hamsters (Mesocricetus auratus). Aggress Behav 1994;20:115–22. Paglietti E, Pellegrini Quarantotti B, Mereu G, Gessa GL. Apomorphine and L-DOPA lower ejaculation threshold in the male rat. Physiol Behav 1978;20:559–62. Payne AP, Swanson HH. Hormonal control of aggressive dominance in the female hamster. Physiol Behav 1971;6:355–7. Payne AP, Swanson HH. The effects of sex hormones on the agonistic behavior of the male golden hamster (Mesocricetus auratus Waterhouse). Physiol Behav 1972a;8: 687–91. Payne AP, Swanson HH. The effects of sex hormones on the aggressive behaviour of the female golden hamster (Mesocricetus auratus Waterhouse). Anim Behav 1972b;20: 782–7. Petrulis A, Johnston RE. Causes of scent marking in female golden hamsters (Mesocricetus auratus): specific signals or classes of information? J Comp Psychol 1997;111:25–36. Pfaus JG, Mendelson SD, Phillips AG. A correlational and factor analysis of anticipatory and consummatory measures of sexual behavior in the male rat. Psychoneuroendocrinology 1990;15:329–40. Poignant J-C, Rismondo N. Influence of the administration of psychopharmacological compounds on the take up time of an alimentary material in the hamster and a study of the associated behavior. Psychopharmacologia 1975;43:47–52. Retana-Marquez S, Salazar ED, Velazquez-Moctezuma J. Muscarinic and nicotinic influences on masculine sexual behavior in rats: effects of oxotremorine, scopolamine, and nicotine. Pharmacol Biochem Behav 1993;44:913–7. Sachs BD. Conceptual and neural mechanisms of masculine copulatory behavior. In: McGill TE, Dewsbury DA, Sachs BD, editors. Sex and behavior: status and prospectus. New York: Plenum Press; 1978. p. 267–95. Scaletta LL, Hull EM. Systemic or intracranial apomorphine increases copulation in long-term castrated male rats. Pharmacol Biochem Behav 1990;37:471–5. Schnur P, Martinez RA. Independent effects of morphine and apomorphine on sterotyped gnawing in the hamster. Pharmacol Biochem Behav 1989;32:589–94. Siegel S. Nonparametric statistics for the behavior sciences. New York: McGraw Hill; 1956. Skrebuhhova-Malmros T, Katrin P, Rudissaar R, Allikmets L, Matto V. The serotonin 5HT2A receptor subtype does not mediate apomorphine-induced aggressive behaviour in male Wistar rats. Pharmacol Biochem Behav 2000;67:339–43. Stinus L, Gaffori O, Simon H, Le Moal M. Disappearance of hoarding and disorganization of eating behavior after ventral mesencephalic tegmentum lesions in rats. J Comp Physiol Psychol 1978;92:289–96. Tagliamonte A, Fratta W, Del Fiacco M, Gessa GL. Possible stimulatory role of brain dopamine in the copulatory behavior of male rats. Pharmacol Biochem Behav 1974;2:257–60. Tsai H-W, Shui H-A, Liu H-S, Tai M-Y, Tsai Y-F. Monoamine levels in the nucleus accumbens correlate with male sexual behavior in middle-aged rats. Pharmacol Biochem Behav 2006;83:265–70. Whitman DC, Hennessey AC, Albers HE. Norepinephrine inhibits vasopressinstimulated flank marking in the Syrian hamster by acting within the medial preoptic-anterior hypothalamus. J Neuroendocrinol 1992;4:541–6. Winer BJ. Statistical principles in experimental design. New York: McGraw Hill; 1971. Yang H-D, Wang Q, Wang Z, Wang D-H. Food hoarding and associated neuronal activation in brain reward circuitry in Mongolian gerbils. Physiol Behav 2011;104: 429–36.