Research ArticleNew Research, Cognition and Behavior

Atypical Social Development in Vasopressin-Deficient Brattleboro Rats

Matthew J. Paul, Nicole V. Peters, Mary K. Holder, Anastasia M. Kim, Jack Whylings, Joseph I. Terranova and Geert J. de Vries
eNeuro 24 March 2016, 3 (2) ENEURO.0150-15.2016; DOI: https://doi.org/10.1523/ENEURO.0150-15.2016

Abstract

Over the past 3 decades, a large body of evidence has accumulated demonstrating that the neuropeptide arginine vasopressin (AVP) plays a critical role in regulating social behavior. The overwhelming majority of this evidence comes from adults, leaving a gap in our understanding of the role of AVP during development. Here, we investigated the effect of chronic AVP deficiency on a suite of juvenile social behaviors using Brattleboro rats, which lack AVP due to a mutation in the Avp gene. Social play behavior, huddling, social investigation & allogrooming, and ultrasonic vocalizations (USVs) of male and female rats homozygous for the Brattleboro mutation (Hom) were compared with their wild-type (WT) and heterozygous (Het) littermates during same-sex, same-genotype social interactions. Male and female Hom juveniles exhibited less social play than their Het and WT littermates throughout the rise, peak, and decline of the developmental profile of play. Hom juveniles also emitted fewer prosocial 50 kHz USVs, and spectrotemporal characteristics (call frequency and call duration) of individual call types differed from those of WT and Het juveniles. However, huddling behavior was increased in Hom juveniles, and social investigation and 22 kHz USVs did not differ across genotypes, demonstrating that not all social interactions were affected in the same manner. Collectively, these data suggest that the Avp gene plays a critical role in juvenile social development.

  • play behavior
  • postnatal development
  • social behavior
  • ultrasonic vocalizations
  • vasopressin

Significance Statement

Several neurodevelopmental disorders are characterized by deficits in social behaviors, the underlying neurobiology of which is not understood. Arginine vasopressin (AVP) has emerged as a candidate neuropeptide through which two such groups of disorders, autism spectrum disorders and schizophrenia, might impact social function. Nonetheless, only a few studies have investigated the role of AVP in social development. Here, we find that rats with a mutation in the Avp gene exhibit “atypical” juvenile social behaviors and vocal communication. These findings suggest that AVP plays a critical role in the regulation of the quantity, quality, and type of social behaviors expressed during development.

Introduction

Childhood and adolescence are periods of marked social development, when individuals acquire the necessary skills for independence (for review, see Spear, 2000). The most prominent social behavior of juveniles across many species is social play, where individuals engage in mock fighting behavior (Bekoff and Byers, 1998; Pellis and Pellis, 1998). In rats, social play emerges during the juvenile phase (∼18 d of age), peaks during early adolescence (∼35 d of age), and declines thereafter (Panksepp, 1981; Pellis and Pellis, 1990). This well characterized developmental profile makes play ideal for studying juvenile and adolescent social development. Furthermore, play contributes to social and emotional development (Pellegrini, 1988; Vanderschuren et al., 1997; Hol et al., 1999; van den Berg et al., 1999). During social interactions, such as play, rats emit ultrasonic vocalizations (USVs) as a form of affective communication (for review, see Wöhr and Schwarting, 2013). Calls with frequencies close to 50 kHz are thought to signal positive affect, whereas ∼22 kHz calls are thought to signal distress (for review, see Brudzynski, 2013). Infant rats and mice also emit ∼40 kHz calls when separated from their mother (for review, see Scattoni et al., 2009).

Many neurodevelopmental disorders are characterized by deficits in social behaviors such as play and communication (e.g., autism spectrum disorders, schizophrenia, and attention deficit hyperactivity disorder; Alessandri, 1992; Jones et al., 1994; Jordan, 2003; Scattoni et al., 2009). Uncovering the underlying neurobiology by which neurodevelopmental disorders impact social function is a difficult task, especially given that the neural mechanisms that regulate “normal” social development are not understood. Here, we focus on the role of arginine vasopressin (AVP) in social development. This peptide is often referred to as a “social neuropeptide” because of its actions on a number of social and antisocial behaviors, including pair bonding, parental behaviors, social recognition, flank marking, and aggression (for review, see Caldwell et al., 2008; Albers, 2012; Bosch and Neumann, 2012). The overwhelming majority of this research has been conducted in adults, but emerging evidence indicates that AVP also influences juvenile social behaviors. The most direct evidence comes from intracranial injections of AVP agonists or antagonists, which impact social play (Cheng and Delville, 2009; Veenema et al., 2013), social recognition (Veenema et al., 2012), and USVs (Lukas and Wöhr, 2015) of juvenile rodents. While these findings provide strong evidence that AVP influences the immediate expression of juvenile social behaviors, the direction of the effects often depends on the age and sex of the subjects, the context of the experiment, and the brain area injected (Veenema et al., 2012, 2013; Bredewold et al., 2014). Hence, we do not yet understand the role of AVP in social development.

Brattleboro rats provide a unique model to study the effects of lifelong deficits in AVP on social behaviors. These rats have a single base pair deletion in exon 2 of the Avp gene that disrupts the production of AVP (Schmale and Richter, 1984). The behavior of adult homozygous Brattleboro (Hom) rats has been well studied, and deficits have been found in the major functions assigned to AVP, including social behaviors such as social recognition/discrimination (Engelmann and Landgraf, 1994; Feifel et al., 2009) and social interactions (Lin et al., 2013). Studies on the behavioral development of Brattleboro rats have been confined to early postnatal life (first 2 weeks of life; Zelena et al., 2008; Lin et al., 2013). Infant Hom rats exhibit decreased aggregation (Schank, 2009) and emit fewer maternal separation-induced USVs (Varga et al., 2015), suggesting that the development of social behaviors might be affected by the Brattleboro mutation. Juvenile social development has not been studied in Brattleboro rats.

In the present study, we test the impact of chronic AVP deficiency on juvenile social development by assessing the effects of the Brattleboro mutation on several social behaviors (social play, USVs, huddling, and social investigation & allogrooming). We find that male and female AVP-deficient Hom rats exhibit lower levels of social play at all stages of play development (onset, peak, and decline of play). Juvenile Hom rats also emit fewer 50 kHz USVs with altered spectrotemporal characteristics. Not all social behaviors are affected in the same manner, however, as juvenile Hom rats display more huddling episodes, and social investigation & allogrooming do not differ between genotypes. These data demonstrate that deficits in AVP throughout development impact the quantity and quality of juvenile social interactions and communication.

Materials and Methods

Animals and housing conditions

A colony of Brattleboro rats (with Long–Evans background) was established in our laboratory from rats purchased from the Rat Resource and Research Center (University of Missouri, Columbia, MO). Brattleboro rats were housed in either opaque plastic cages with Carefresh bedding and wood chips (48 × 27 × 20 cm) or ventilated transparent OptiRat plastic cages with Bed-O-Cobs bedding (35.6 × 48.5 × 21.8 cm). For all experiments, the day of birth was considered postnatal day 0 (P0). Room lights were set to a 12 h light/12 h dark cycle (lights off at 5:00 P.M. Eastern time), and ambient temperature was maintained at 23°C. Food and water were available ad libitum. All procedures were performed in accordance with the Guide for the Care and Use of Laboratory Animals and were approved by the Animal Care and Use Committee at Georgia State University and the University of Massachusetts, Amherst.

Experiment 1: emergence of play behavior in Brattleboro rats

Wild-type (WT), heterozygous (Het), and homozygous (Hom) Brattleboro offspring were obtained from Het × Het breeding pairs from our colony. Overall, the distribution of genotypes was 1.3/2.0/0.9 (WT/Het/Hom). Each rat pup was tested for play behavior once at P17, P19, P21, or P23. All rats were tested prior to weaning, which occurred at P24. Rats were removed from their litters 2–3 h before being paired with a similarly treated age-matched, same-genotype, same-sex rat in a clean cage for a social behavior test (Panksepp, 1981; Paul et al., 2014).

Experiment 2: social behaviors and ultrasonic vocalizations of juvenile Brattleboro rats

WT, Het, and Hom Brattleboro offspring were obtained from Het × Het breeding pairs from our colony. Overall, the distribution of genotypes was 0.9/2.0/0.9 (WT/Het/Hom). Rats were weaned at P22, at which point they were housed with an age-matched, same-genotype, same-sex cage mate. At P33 (±2 d) or P43 (±2 d), cage mates were single housed for ∼24 h before being reunited in a social behavior test the following day.

Social behavior tests

All tests were conducted within the first 2.5 h of lights off under red light. Animals were paired with an age-matched, same-sex, same-genotype playmate in a fresh cage for 20 min, and their behavior was videotaped. In Experiment 2, bedding was removed from the test cage to minimize background noise interference with the ultrasonic recordings. Play attacks (i.e., lunges toward the nape of the playmate’s neck), pins (i.e., animal lying in the supine position with playmate on top), and boxing events (i.e., both animals standing on their hindpaws and pushing each other with their forepaws), as described in studies by Meaney and Stewart (1981a) and Vanderschuren et al. (1997), were scored by a researcher who was blind to the treatment conditions using JWatcher software (http://www.jwatcher.ucla.edu/; Experiment 1) or The Observer XT11 (Noldus Information Technology Inc.; Experiment 2). In Experiment 2, the number of social investigation & allogrooming events and huddling episodes were also scored using The Observer XT11.

Ultrasonic vocalization recordings

Vocal emissions were recorded for the duration of the social behavior tests using an UltraSoundGate CM16/CMPA microphone (Avisoft Bioacoustics) placed just above the testing cage. The microphone was connected to a computer via an Avisoft Bioacoustics UltraSoundGate 116Hb. Acoustic data were recorded with a sampling rate of 250 kHz in 16 bit format, and spectrograms were constructed by fast Fourier transformation (FFT; 256 FFT length, 100% frame, FlatTop window, and 50% time window overlap; SASLab Pro, Avisoft Bioacoustics). All USVs made within the first 10 min of the play behavior trial were manually marked by investigators who were blind to the age, sex, and genotype of the rats. In order to be marked, calls had to be at least 10 ms in length, and distinct calls had to be separated by at least 10 ms. Several call parameters were quantified, including fundamental frequency, duration, and number of calls emitted. Call frequency (in hertz) was calculated by averaging the fundamental frequency at call onset, call offset, and peak amplitude of the call (integrated frequency). A subset (20% random sampling) of the calls was selected and manually classified into the 15 call categories described in the study by Wright et al. (2010).

Genotyping of Brattleboro rats

The Brattleboro mutation is a single base pair deletion in exon 2 of the Avp gene that disrupts processing of the AVP prohormone (Schmale and Richter, 1984). Previous genotyping protocols required DNA sequencing after PCR amplification to detect the single base pair deletion. In the present study, we developed a faster and cheaper method, replacing the sequencing step with restriction enzyme digestion followed by gel electrophoresis. Tail tissue was harvested from rat pups between 8 and 12 d of age using ice-cold ethanol as a local anesthetic. For animals in Experiment 1, tails were digested at 55°C overnight in 400 μl of Tail Lysis Buffer containing 4 μl of Proteinase K. DNA was extracted and purified with phenol, chloroform:isoamyl alcohol (24:1), isopropanol, and 70% ethanol. For animals in Experiment 2, the REDExtract-N-Amp Tissue PCR Kit (Sigma-Aldrich) was used for tail digestion and DNA extraction. DNA surrounding the base pair deletion was amplified by PCR using the following primers: forward, GACGAGCTGGGCTGCTTC; reverse, CCTCAGTCCCCCACTTAGCC. Twenty microliters of PCR product was then digested with BcgI restriction endonuclease (New England BioLabs) at 37°C overnight using the following concentrations: 3 μl of 10× NEBuffer 3, 4 μl of 10× S-adenosylmethionine, 2 μl of nuclease-free water, and 1 μl of BcgI. BcgI recognizes and cuts only the mutant Brattleboro PCR product, resulting in two DNA fragments of similar size (92 and 97 bp). Therefore, samples from WT rats exhibit a single 222 bp band after gel electrophoresis, whereas those of Hom rats exhibit a single ∼95 bp band (the two fragments do not separate on a 2% agarose gel). Samples from Het rats exhibit both WT and Hom bands. To validate this procedure, we confirmed the genotyping results from a subset of samples with the traditional sequencing methodology.

Statistical analyses

Data were analyzed by three-way ANOVA with genotype, age, and sex as the independent variables. Because the main effect of age was significant for the overwhelming majority of measures (22 of 27), data were also analyzed separately at each age by two-way ANOVA, and this analysis was depicted in all figures. Where the main effect of genotype was significant, and in a few cases where it approached significance (p = 0.06), post hoc comparisons between each genotype were assessed using Fisher’s PLSD. For a number of comparisons, the data were not normally distributed due to the high number of zero values, and the distribution could not be corrected by using a log(x + 0.01) transformation (Tables 1–3 located at the end of the article). This is to be expected when age groups prior to the onset of the behavior are included in the analysis (in Experiment 1) and when one assesses behaviors not highly expressed during motivated social behavior tests (e.g., distress USVs and huddling in Experiment 2). Although multifactorial ANOVA is robust when the data are not normally distributed, we confirmed significant main effects and post hoc comparisons from non-normally distributed data with the appropriate nonparametric test (Kruskal–Wallis or Mann–Whitney U tests). For all of these comparisons, except for post hoc tests assessing age differences in Experiment 1, the ANOVA and nonparametric tests yielded the same result. For the post hoc tests where the results differed, we reported the nonparametric statistic. Significance was assumed at p < 0.05.

View this table:
Table 1:

Experiment 1 statistical analyses

View this table:
Table 2:

Experiment 2 statistical analyses

View this table:
Table 3:

USV call type statistical analyses

Results

Experiment 1: emergence of play behavior in Brattleboro rats

Consistent with other reports (Panksepp, 1981; Paul et al., 2014), the developmental onset of play occurred at ∼P19–P21 (Fig. 1). Play behavior was virtually absent at P17, increased slightly at P19, and further increased at P21 (main effect of age, p < 0.0001 for total playa1, pinsb1, and pouncesc1; P17 vs P19 and P19 vs P21, p < 0.0001, for total playa4,a5, pinsb4,b5, and play attacksc4,c5; superscripts used here and further in the document indicate rows in Tables 1–3 located at the end of the article). Play behavior of males and females did not differ (main effect of sex, p > 0.38 for total playa3, pinsb3, and play attacksc3).

Overall, Hom rats played less than their WT and Het littermates (Fig. 1A; total play, main effect of genotype, p < 0.04a2; Hom vs WTa6 or Heta7, p < 0.05). This was due to fewer numbers of pins (Fig. 1B; main effect of genotype, p < 0.02b2; Hom vs WTb6 or Hetb7, p < 0.009) and play attacks: while the main effect of genotype fell short of significance for play attacks (p = 0.06 for ANOVA and p = 0.09 for Kruskal–Wallis testc2), post hoc comparisons indicated that Hom weanlings did exhibit fewer play attacks than WT weanlings (p < 0.04c6; not illustrated). Boxing events were rare in all genotypes, with <0.25 events during the 20 min test for all groups; no Hom pairs exhibited a boxing event. Decreased total play and pins of Hom rats were evident at P21 (main effect of genotype, p < 0.003 for both behaviorsa8,b8; Hom vs WT, p < 0.0005 for both behaviorsa9,b9) and P23, although comparisons for total play fell short of significance at P23 (main effect of genotype, p < 0.03 for pinsb12, p = 0.06 for total playa12; Hom vs WT, p < 0.04 for pinsb13, p = 0.09 for total playa13). Notably, Het animals exhibited less total play and fewer pins than WT animals at P21 (Het vs WT, p < 0.04 for both behaviorsa11,b11, Fisher’s PLSD). Genotype did not impact total play or pins at P19 (main effect of genotype, p > 0.32 for total playa15 and pinsb15), when play was low for all genotypes; levels of play at P17 were too low for statistical analyses.

Figure 1.
Figure 1.

Hom Brattleboro weanlings play less than their WT and Het littermates. A, B, Number of total play behaviors (A) and pins (B) of Hom, Het, and WT rats during a 20 min test at P17, P19, P21, or P23. Sample sizes are indicated within each bar. Data from each age were obtained from separate cohorts of animals. Genotypes with differing letters differ significantly from each other (p < 0.05, Fisher’s PLSD); where differences approach significance, the p value is included in parentheses next to the letter representing the appropriate comparison. See Results for ANOVA details.

Experiment 2: social behaviors and ultrasonic vocalizations of juvenile Brattleboro rats

Social behaviors

As for weanling-aged rats, Hom juveniles played less than WT and Het juveniles due to reductions in both pins and play attacks (Figs. 2, 3B; main effect of genotype, p < 0.0001 for total playd2, pinse2, and play attacksf2; Hom vs WTd4,e4,f4 or Hetd5,e5,f5, p < 0.005 for all three behaviors). Analysis of the temporal profile of total play revealed that Hom juveniles played less than WT and Het rats across the entire 20 min test (Fig. 2C,D). Hom juveniles exhibited fewer total social behaviors than WT and Het juveniles (Fig. 3A; main effect of genotype, p < 0.0001g2; Hom vs WTg4 or Hetg5, p < 0.0001). Reductions in play and social behaviors of Hom juveniles were evident at both P34 and P44, although the post hoc comparison between WT and Hom groups for pins at P34 fell short of significance when each age was analyzed separately (Figs. 2, 3). Social investigation & allogrooming did not differ between genotypes (Fig. 3C; main effect of genotype, p > 0.51h2), and huddling episodes were increased in Hom juveniles (Fig. 3D; main effect of genotype, p < 0.0001i2; Hom vs WTi4 or Heti5, p < 0.0001), indicating that not all social behaviors are affected in the same manner by the Brattleboro mutation.

Figure 2.
Figure 2.

Social play is decreased in Hom Brattleboro juveniles. A, B, Number of pins (A) and play attacks (B) of male and female Hom, Het, and WT rats during a 20 min test at P34 or P44. C, D, The temporal profile of play is illustrated in C (for P34) and D (for P44) as the number of total play behaviors binned every 5 min. Sample sizes are indicated within each bar in A and B. Data from each age were obtained from separate cohorts of animals. Genotypes with differing letters differ significantly from each other (p < 0.05, Fisher’s PLSD); where differences approach significance, the p value is included in parentheses next to the letter representing the appropriate comparison. In C and D, significant differences between Hom and WT or Het rats within each bin are indicated by * and #, respectively (p < 0.005, Fisher’s PLSD). See Results for ANOVA details.

Figure 3.
Figure 3.

Social behavior is altered in Hom Brattleboro juveniles. A–D, Number of total social behaviors (A), total play behaviors (B), social investigation & allogrooming behaviors (C), and huddling episodes (D) of Hom, Het, and WT rats during a 20 min test at P34 or P44. Sample sizes are indicated within each bar. Data from each age were obtained from separate cohorts of animals. Genotypes with differing letters differ significantly from each other (p < 0.05, Fisher’s PLSD). See Results for ANOVA details.

The number of all social behaviors decreased between P34 and P44 (Figs. 2, 3; main effect of age, p < 0.01 for total playd1, pinse1, play attacksf1, total social behaviorsg1, social investigation & allogroomingh1, and huddlingi1). There were no significant sex differences in the number of social behaviors (main effect of sex, p > 0.46 for total playd3, pinse3, play attacksf3, total social behaviorsg3, social investigation & allogroomingh3, and huddlingi3), except that male Het juveniles exhibited more pins than female Het juveniles at 44 d of age (Fig. 2B; genotype × sex interaction at P44, p < 0.008e6; Het male vs Het female, p < 0.002e7).

Ultrasonic vocalizations

Similar to play and overall social behaviors, Hom rats emitted fewer USVs than WT and Het rats (Fig. 4; main effect of genotype, p < 0.0001j1; Hom vs WTj2 or Hetj3, p < 0.003) due to a selective reduction in 50 kHz USVs (Fig. 4B,C; 50 kHz USVs: main effect of genotype, p < 0.0001k2; Hom vs WTk3 or Hetk4, p < 0.002; 22 kHz USVs: main effect of genotype, p > 0.25l2). Decreased 50 kHz USVs of Hom rats were evident at both juvenile ages, although the post hoc comparisons between WT and Hom rats fell short of significance when P44 data were analyzed separately (Fig. 4B). Unlike social behaviors, the number of 50 and 22 kHz USVs increased across age (main effect of age, p < 0.04 for 50 kHz and 22 kHz USVs).

Figure 4.
Figure 4.

Hom Brattleboro juveniles emit fewer 50 kHz USVs. A–C, Number of all (A), 50 kHz (B), and 22 kHz (C) USVs of male and female Hom, Het, and WT rats during the first 10 min of a 20 min test at P34 or P44. Sample sizes are indicated within each bar. Data from each age were obtained from separate cohorts of animals. Genotypes with differing letters differ significantly from each other (p < 0.05, Fisher’s PLSD); where differences approach significance, the p value is included in parentheses next to the letter representing the appropriate comparison. See Results for ANOVA details.

The 50 kHz USV category consists of calls with a broad range of frequencies (30–117 kHz in the present study) and spectral-temporal structures (e.g., constant frequency, frequency steps, frequency trills), and it is not known whether these calls are functionally equivalent. To determine which types of 50 kHz USVs were impacted by the Brattleboro mutation, we classified a subset (20%) of the USVs of each animal according to the call types proposed by Wright et al. (2010) and assessed the impact of the Brattleboro mutation on the quantity (number) and quality (duration and integrated frequency, which was defined as the mean of the call onset, peak amplitude, and call end frequency) of the vocalizations most frequently emitted during social behavior testing.

Figure 5 illustrates the percentage of all classified calls, regardless of the genotype of the caller. Most USVs fell within the 50 kHz category, with type 4 calls (flat calls) being the most common (33.3%) followed by type 1 (complex calls, 15.0%), type 10 (trills, 11.1%), type 2 (upward-ramp calls, 10.1%), and type 7 calls (step-up calls, 9.7%). The percentage for each of the remaining call types was <5%, including 22 kHz USVs (type 15), which comprised 2.5% of calls.

Figure 5.
Figure 5.

Ultrasonic vocalization call types emitted during social behavior testing. A, Percentage of USV call types, as defined by Wright et al. (2010), emitted by male and female Hom, Het, and WT juveniles (P34 or P44) during the first 10 min of a 20 min test; data are combined across sex, genotype, and age. B–G, Representative spectrograms of the most common 50 kHz calls [complex (B), upward-ramp (C), flat (D), step-up (E), and trill (F)] as well as the 22 kHz call (G).

Quantity of ultrasonic vocalization call types

In general, juvenile Hom rats emitted fewer calls of each type than their WT and Het littermates, and for most this was due to a decreased call number at P34 (Fig. 6A–E). At P34, Hom rats emitted fewer upward-ramp, flat, and step-up calls (Fig. 6B–D; main effect of genotype, p < 0.03 for each call typen2,o2,p2; Hom vs WT, p < 0.02 for upward-rampn3 and step-upp3 calls; Hom vs Het, p < 0.02 for each call typen4,o4,p4), although the comparison between WT and Hom flat calls fell short of significance (p = 0.06°3). At P44, there were no significant differences between genotypes for these calls (main effect of genotype, p > 0.05 for upward-rampn5, flato5, and step-upp5 calls). Trills were reduced in Hom juveniles at both P34 and P44 (Fig. 6E; main effect of genotype, p < 0.01 for both P34q3 and P44q6; Hom vs WT, p < 0.04 for P44q7; Hom vs Het, p < 0.005 for both P34q5 and P44q8), although the difference between Hom and WT rats was not significant at P34 (p = 0.10q4). When analyzed across both ages, Hom rats emitted fewer complex calls than WT and Het rats (main effect of genotype, p < 0.03m2; Hom vs WT m3or Hetm4, p < 0.04), but these comparisons were not significant when assessed separately for each age (Fig. 6A; main effect of genotype, p > 0.15 for P34m5 and P44m6).

Figure 6.
Figure 6.

The quantity and quality of USV calls is altered in Hom Brattleboro rats. A–O, number (A–E), integrated frequency (F–J), and duration (K–O) of type 1 (complex), type 2 (upward-ramp), type 4 (flat), type 7 (step-up), and type 10 (trill) USV calls of male and female Hom, Het, and WT rats during the first 10 min of a 20 min test at P34 or P44. Data from each age were obtained from separate cohorts of animals. Genotypes with differing letters differ significantly from each other (p < 0.05, Fisher’s PLSD); where differences approach significance, the p value is included in parentheses next to the letter representing the appropriate comparison. See Results for ANOVA details.

Upward-ramp, step-up, and trill calls increased from P34 to P44 (Fig. 6B,D,E; main effect of age, p < 0.04 for each call typen1,p1,q1); complex and flat calls did not differ across age (Fig. 6A,C; main effect of age, p > 0.13 for both callsm1,o1). Only trills differed between the sexes, with males emitting more than females (Fig. 6E; main effect of sex, p < 0.005q2).

Quality of ultrasonic vocalization call types

For most USV types, Hom rats emitted calls with a lower integrated frequency, but duration was only altered for step-up calls and trills. Upward-ramp, flat, and step-up calls of Hom rats had a lower integrated frequency than those of WT and Het rats (main effect of genotype, p < 0.02 for each call types1,t1,u1; Hom vs WTs2,t2,u2 or Hets3,t3,u3, p < 0.05 for each call type). The age at which these effects were significant depended on the call type (Fig. 6G–I). Complex calls of Hom rats also had a lower integrated frequency than those of Het rats (main effect of genotype, p < 0.02r1; Hom vs Het, p < 0.006r3), but did not differ significantly from WT rats (p = 0.10r2); Hom and Het differences in complex calls were not significant when those from P34 and P44 were analyzed separately (Fig. 6F). Trills were the only call type analyzed for which integrated frequency was unaffected by genotype (Fig. 6J; main effect of genotype, p > 0.24v1).

The duration of USVs also differed by genotype, but only for step-up calls and trills, and only at P44 (Fig. 6K–O). Hom rats emitted longer duration step-up calls than WT and Het rats at P44 (Fig. 6N; main effect of genotype, p < 0.003 for P44w1; Hom vs WTw2 or Hetw3, p < 0.02). For trills, both Hom and Het rats emitted shorter calls than WT rats at P44, and Hom and Het rats did not differ from each other (Fig. 6O; main effect of genotype, p < 0.01x1; Homx2 or Hetx3 vs WT, p < 0.02; Het vs Hom, p > 0.26x4).

The quality of USV call types also changed with age. The integrated frequency decreased from P34 to P44 for each call type except trills (Fig. 6F–J; main effect of age, p < 0.05 for complexr4, upward-ramps4, flatt4, and step-upu4 calls; main effect of age, p > 0.18 for trillsv2). This reduction in integrated frequency was evident in all genotypes; no interaction between age and genotype was found (age × genotype, p > 0.05 for complexr5, upward-ramps5, flatt5, and step-upu5 calls). In addition, the duration of complex, upward-ramp, and flat calls also decreased from P34 to P44 (Fig. 6K–M; main effect of age, p < 0.03 for each call typey1,z1,aa1); the duration of step-up calls and trills did not vary with age (Fig. 6N,O; main effect of age, p > 0.10 for step-up callsw4 and trillsx5).

Discussion

The present study suggests that the Avp gene plays an important role in social development. The Brattleboro mutation, which disrupts the production of AVP, impacted both social behaviors and ultrasonic communication of juvenile rats. Hom rats played less and emitted fewer 50 kHz USVs than their WT and Het littermates. In addition, the spectrotemporal characteristics of USVs emitted by Hom rats differed from those of WT and Het rats. Social deficits, however, were behavior and USV specific. Huddling episodes were increased in Hom rats, and social investigation & allogrooming, and 22 kHz USVs did not differ across genotypes. Hence, Hom Brattleboro rats are not simply asocial, rather their social behaviors are “atypical” compared with those of WT and Het rats.

Deficits in the social play of Hom rats were evident throughout the developmental profile of play [onset (P21 and P23), peak (P34), and decline (P44)], suggesting that AVP is important for the overall level of play rather than its developmental timing. This is similar to the persistent developmental deficits reported for body and brain weights of Hom Brattleboro rats (for review, see Boer, 1985), but differs from other measures (e.g., eye opening, ear opening, incisor eruption), which occur earlier in Hom Brattleboro rats (Boer et al., 1980; Zelena et al., 2009). Notably, the greatest deficits in the neural development of Hom Brattleboro rats occur in the cerebellum (Boer et al., 1982), a brain region whose development correlates with the ontogeny of play across several species (Byers and Walker, 1995). It has been proposed that play behavior contributes to cerebellar development (Byers and Walker, 1995). Following this logic, it is possible that the decreased play of Hom rats contributes to developmental deficits in their cerebellar size and morphology. The reverse, however, is also possible.

Adult Het rats exhibit partial reductions in AVP neural mRNA expression and pituitary peptide content (Dorsa and Bottemiller, 1982), and sometimes exhibit behavioral differences from WT rats (Brot et al., 1992). Nonetheless, the social behaviors of Het rats in the present study did not differ statistically from those of WT rats except for a transient reduction in social play during the developmental onset of play (at P21). These data raise the possibility of a gene dosage effect during the developmental onset of play. Perhaps the onset of play requires higher levels of AVP than its maintenance or Het rats have insufficient AVP at this age to stimulate play.

The Brattleboro mutation also affects USVs. Infant Hom Brattleboro pups emit fewer maternal separation-induced 40 kHz USVs (Lin et al., 2013; Varga et al., 2015). Here, we demonstrate that USV deficits of Hom Brattleboro rats persist into the juvenile stage and include prosocial vocalizations. Juvenile Hom rats emitted fewer USVs during the social interaction test due to a selective reduction in 50 kHz calls. Fifty kilohertz USVs reflect a positive affective state and are considered a form of prosocial communication (for review, see Brudzynski, 2013; Wöhr and Schwarting, 2013). Fifty kilohertz calls are emitted during appetitive interactions and in anticipation of reward stimuli, such as mating, play, addictive drugs, and “tickling” (Barfield et al., 1979; Knutson et al., 1998, 1999; Panksepp and Burgdorf, 2000; Burgdorf et al., 2008). Furthermore, 50 kHz calls elicit approach behavior (Wöhr and Schwarting, 2007) and “self administration” for their playback (Burgdorf et al., 2008). Hence, decreased 50 kHz USVs in Hom Brattleboro rats may indicate decreased prosocial motivation for, or reward value of, social interactions in AVP-deficient animals.

In contrast, 22 kHz calls are emitted in response to aversive stimuli (e.g., electric shocks, predators, drug withdrawal, and aggressive interactions; Sales, 1972; Tonoue et al., 1986; Cuomo et al., 1988; Blanchard et al., 1991; Vivian and Miczek, 1991; Barros and Miczek, 1996; Covington and Miczek, 2003), are thought to reflect a negative affective state akin to anxiety or distress (for review, see Brudzynski, 2013; Wöhr and Schwarting, 2013), and are not affected by the Brattleboro mutation (present findings). Hence, AVP deficiency does not affect all forms of vocal communication, with distress-like calls being particularly independent of AVP status. This conclusion should be tempered by the low levels of 22 kHz calls for all genotypes in the present experiment, which is consistent with the findings of previous studies measuring USVs during prosocial playful interactions (Burgdorf et al., 2006). In addition, while the 22 kHz USVs in the present study were within the frequency range of distress-like USVs, their duration was much shorter than that typically reported: ∼24 ms in the current study (Fig. 5G) versus 300–1200 ms in studies investigating USVs in response to aversive stimuli (Tonoue et al., 1986; Brudzynski and Ociepa, 1992). Brudzynski et al. (1993) reported two distinct populations of 22 kHz USVs in response to experimenter handling: short calls of 20–300 ms and long calls of 300 to >2000 ms, with most long calls falling between 500 and 600 ms. The functional significance of these short 22 kHz calls is not known. Therefore, it is possible that the 22 kHz USVs in the present study were not true anxiety or distress-like calls. Future studies are needed to determine whether 22 kHz USVs are altered in Hom Brattleboro rats tested under aversive conditions.

We further analyzed the USVs according to subcategories suggested by Wright et al. (2010) to determine whether the Brattleboro mutation differentially impacted different types of calls (i.e. altered their vocal repertoire). In general, Hom Brattleboro rats emitted fewer of each USV call type analyzed. Reductions were evident in each of the five following most common call types: flat calls, complex calls, trills, upward-ramp calls, and step-up calls (all 50 kHz calls). Deficits were most robust for trills, which were present at both P34 and P44, and least robust for complex calls, which were not significant when each age was analyzed separately. In addition to the quantity of USVs, the Brattleboro mutation impacted spectrotemporal characteristics of USVs. Flat, upward-ramp, and step-up calls of Hom rats had lower integrated frequencies. In addition, step-up calls were longer and trills were shorter in Hom rats. While it is not clear why AVP deficiency impacts the spectrotemporal quality of USVs in a call-specific manner, it is clear that several USV call types of Hom rats sound different than those of WT and Het rats. It is interesting to speculate that the reduced number and integrated frequency of 50 kHz USVs of Hom rats may contribute to their “atypical” social behaviors. Call frequency is an important feature for USV call structure. Frequency is the dominant feature mice use to discriminate between tone categories (Radziwon and Dent, 2014). In addition, rats will approach a speaker playing 50 kHz calls and tones (Wöhr and Schwarting, 2007). Hence, the decreased integrated frequency of flat, upward-ramp, and step-up calls of Hom rats might impact call meaning or appetitive quality. Therefore, the reduced number and frequency of 50 kHz calls might lead to less play by reducing the amount of prosocial stimulation during social interactions. This rationale, however, cannot explain the increased huddling seen in Hom rats. Perhaps the prosocial nature of 50 kHz USVs depends upon the type of social behavior. For example, because 50 kHz USVs stimulate locomotor behavior (Wöhr and Schwarting, 2007), it is possible that they stimulate “active” social behaviors such as play, but inhibit “passive” social behaviors such as huddling. It should be noted that a direct relationship between the number of USVs and play events has not been established, and it is also possible that play triggers USVs. Our data suggest that a simple relationship between play and USVs as a whole is unlikely; in Experiment 2, for example, the number of USVs increased with age, whereas the number of play events (and social interactions) decreased with age.

Despite significant interest in adolescent social development, most studies on the development of USVs have focused on infant maternal separation-induced 40 kHz calls. A few reports have found that adult rodents emit more USVs than adolescents during same-sex or opposite-sex interactions (Cherry, 1987; Willey et al., 2009; Willey and Spear, 2012; Kabitzke et al., 2015). Similarly, we found that the number of both 50 and 22 kHz USVs emitted during same-sex juvenile social interactions increase across a short 10 d interval from P34 to P44, which approximates early/mid-adolescence in rats (Vetter-O’Hagen and Spear, 2012). We further found that the developmental increases in 50 kHz USVs were specific to call type, occurring in upward-ramp, step-up, and trill calls, but not complex or flat calls. In addition, the spectrotemporal characteristics of several 50 kHz call types changed across these ages: integrated frequency decreased for complex, upward-ramp, flat, and step-up calls; and duration decreased for complex, upward-ramp, and flat calls. These findings raise the possibility that spectrotemporal characteristics of some rat USVs convey age-related information of the caller and thereby influence age-dependent social interactions. For example, perhaps the spectrotemporal characteristics of prepubertal calls elicit less aggression from same-sex adults, whereas those of postpubertal calls may better stimulate sex behaviors in the opposite sex.

Males emit more USVs than females, both as infants in response to maternal separation (Bowers et al., 2013), and as juveniles immediately preceding and following play bouts (Himmler et al., 2014). In the present experiment, we found that the sex difference in the number of USVs during juvenile play is restricted to trills, with males again emitting more than females. Although Himmler et al. (2014) did not report whether sex differences were present in all or some USV call types, trills comprised 77% of calls in their analysis. We did not detect any sex differences in the integrated frequency or duration of any call type, including trills, suggesting that juvenile sex differences in USVs are limited to quantity rather than spectrotemporal quality. With the exception of a single comparison in 44-d-old Het rats (Fig. 4A), we did not detect sex differences in other social behaviors, including play. Although males are often reported to engage in more rough-and-tumble play than females (Meaney and Stewart, 1981b; Pellis, 2002; Olesen et al., 2005), this sex difference is dependent upon the testing conditions and behaviors measured (Thor and Holloway, 1984; Argue and McCarthy, 2015). The absence of sex differences in play in the present experiments is not surprising as studies testing same-sex pairs of rats after a period of isolation, as we did in the present experiments, generally do not detect sex differences in juvenile social play (Panksepp and Beatty, 1980; Panksepp, 1981; Veenema et al., 2013; Paul et al., 2014).

Currently, we cannot determine whether the altered social behavior and USVs of Brattleboro rats are due to the absence of central or peripheral actions of AVP. Hom Brattleboro rats develop diabetes insipidus due to the absence of AVP-mediated water reabsorption at the level of the kidney (Valtin and Schroeder, 1964), and symptoms are evident before play onset: increased plasma osmolarity is present at 10-14 d of age, and polydipsia develops between 15 and 16 d of age (Dlouhá et al., 1982; Zelena et al., 2009). Nonetheless, acute intracerebroventricular injections of a vasopressin 1a receptor (V1aR) antagonist decrease maternal separation-induced 40 kHz USVs of infant rat pups (Winslow and Insel, 1993; Bleickardt et al., 2009), as well as 50 kHz USVs and play behavior of male juvenile rats (Veenema et al., 2013; Lukas and Wöhr, 2015). These findings argue that the altered play and USVs of Brattleboro rats is due to a direct disruption of the actions of AVP in the brain rather than a disruption of the peripheral actions of AVP or indirect compensatory changes resulting from the absence of AVP during development. Decreased anxiety-like, depressive-like, and maternal behaviors of Brattleboro rats persist after the restoration of the peripheral actions of AVP, indicating that several behavioral abnormalities of Brattleboro rats are due to the loss of the central actions of AVP (Fodor et al., 2012; Balázsfi et al., 2015).

We do not yet know which AVP system is responsible for the deficits seen in the Brattleboro rats. The limited available data focus on juvenile social play and do not provide a comprehensive understanding of the regulation by AVP of social development. AVP cells in the bed nucleus of the stria terminalis (BNST) appear to play an inhibitory, rather than stimulatory, role in juvenile social play, at least in males. During the developmental emergence in weanling-aged rats (P18–P23), BNST AVP mRNA expression of males correlates negatively with play behavior, whereas in females, BNST AVP mRNA expression is not detectable (Paul et al., 2014). Furthermore, V1aR antagonist injections into the septum, a projection area of BNST AVP cells, increases the play behavior of juvenile males, but not females (Veenema et al., 2013). In the same study, intracerebroventricular V1aR antagonist injections decreased the play behavior of male rats but increased the play behavior of female rats. AVP mRNA expression in the paraventricular nucleus of the hypothalamus (PVN) of male, but not female, rats correlates positively with their play behavior during the developmental emergence of play, raising the possibility that the PVN is the site of the stimulatory actions of AVP, at least in males (Paul et al., 2014). At present, however, it is difficult to incorporate these sex-specific findings with the present results in which “atypical” social behaviors of Hom rats were seen in both sexes (including play deficits). Complicating matters further, play behavior and the effects of pharmacological manipulations of AVP on play can depend upon the context in which the animals are tested (Bredewold et al., 2014). AVP in the PVN regulates the stress axis and autonomic function, both of which could influence social behavior in a context-specific manner through their actions on stress or arousal. Adult Brattleboro rats are less reactive to various stressors (Balázsfi et al., 2015). If true for juveniles, this decreased stress reactivity could have contributed to the present findings where play was tested in a novel home cage after a period of isolation. Although rats play less after restraint stress (Klein et al., 2010), isolation increases play in rats (Panksepp and Beatty, 1980). Hence, it is not clear whether decreased stress reactivity would lead to higher or lower levels of play. The low levels of 22 kHz USVs (and the absence of long 22 kHz USVs) across all genotypes suggest that rats were not distressed or anxious in the novel cage during testing. Genotype differences in novel cage exploration at the beginning of the test are unlikely to account for the decreased play as Hom Brattleboro rats exhibited lower levels of play across the entire 20 min test (Fig. 2C,D). More studies are needed to dissect out which AVP systems contribute to social development and how.

Early life experiences can significantly impact behavioral development (Kundakovic and Champagne, 2015), and juvenile social play is particularly sensitive to both prenatal and early postnatal environments (Veenema and Neumann, 2009; Kirsten et al., 2010; Taylor et al., 2012; Karkow and Lucion, 2013), including natural variations in maternal care (Parent and Meaney, 2008). Given that the Brattleboro mutation impacts maternal behaviors (Fodor et al., 2012), care must be taken when designing and interpreting results using this model. Indeed, Hom Brattleboro dams influence several behavioral and physiological characteristics of their offspring (e.g., body weight, brain weight, startle response, and stress reactivity; Snijdewint et al., 1988; Zelena et al., 2003, 2009; Feifel and Priebe, 2007). To minimize the potential impact of the early-life environment in the present experiment, we tested Hom, Het, and WT littermates born to Het dams, thereby removing potential prenatal and postnatal confounds of maternal genotype. Nonetheless, we cannot rule out possible confounds of differential maternal or sibling treatment toward Hom Brattleboro pups.

While it is now clear that AVP influences juvenile social behaviors, we know very little about how AVP acts to regulate social development. This represents a critical gap in our knowledge as altered AVP function has been implicated in neurodevelopmental disorders such as autism spectrum disorders and schizophrenia [Heinrichs et al., 2009; Rubin et al., 2014 (and references therein)]. Notably, Hom Brattleboro rats exhibit behavioral abnormalities associated with schizophrenia and autism spectrum disorders, including decreased social interactions, social cognitive deficits, and attenuated prepulse inhibition, which is consistent with the hypothesis that AVP signaling is disrupted in these disorders (Engelmann and Landgraf, 1994; Feifel and Priebe, 2007; Feifel et al., 2009; Lin et al., 2013). The present findings add social play and vocal communication to this list. Our findings also provide insight into the role of AVP in social development. By measuring multiple social behaviors in the same experiment, we were able to demonstrate that long-term disruption of AVP production does not simply decrease overall levels of social behavior, but rather alters the type of social behavior the animal expresses, leading to an atypical rather than asocial phenotype. Future studies are needed to uncover the neural mechanisms through which AVP influences both normal and disordered social development.

Acknowledgments

Acknowledgements: We thank Stefanie Krug, Krishna Mehta, Christine Badeau, Quentin Richardson, and Beth Parker for technical assistance. We also thank Ann-Marie Torregrossa for assistance with statistical analyses.

Footnotes

  • 1 The authors declare no competing financial interests.

  • 3 This research was supported by National Institutes of Health Grant R01-MH-047538 to G.J.d.V.

  • Received December 5, 2015.
  • Revision received February 14, 2016.
  • Accepted March 1, 2016.

This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International, which permits unrestricted use, distribution and reproduction in any medium provided that the original work is properly attributed.

References

  1. Albers HE (2012) The regulation of social recognition, social communication and aggression: vasopressin in the social behavior neural network. Horm Behav 61:283–292. doi:10.1016/j.yhbeh.2011.10.007 pmid:22079778
  2. Alessandri SM (1992) Attention, play, and social behavior in ADHD preschoolers. J Abnorm Child Psychol 20:289–302. pmid:1619135
  3. Argue KJ, McCarthy MM (2015) Characterization of juvenile play in rats: importance of sex of self and sex of partner. Biol Sex Differ 6:16. doi:10.1186/s13293-015-0034-x pmid:26361539
  4. Balázsfi D, Pintér O, Klausz B, Kovács KB, Fodor A, Török B, Engelmann M, Zelena D (2015) Restoration of peripheral V2 receptor vasopressin signaling fails to correct behavioral changes in Brattleboro rats. Psychoneuroendocrinology 51:11–23. doi:10.1016/j.psyneuen.2014.09.011 pmid:25278460
  5. Barfield RJ, Auerbach P, Geyer LA, McIntosh TK (1979) Ultrasonic vocalizations in rat sexual behavior. Am Zool 19:469–480. doi:10.1093/icb/19.2.469
  6. Barros HM, Miczek KA (1996) Withdrawal from oral cocaine in rate: ultrasonic vocalizations and tactile startle. Psychopharmacol (Berl) 125:379–384. pmid:8826543
  7. Bekoff M, Byers JA (1998) Animal play: evolutionary, comparative and ecological perspectives. Cambridge, UK: Cambridge UP.
  8. Blanchard RJ, Blanchard DC, Agullana R, Weiss SM (1991) Twenty-two kHz alarm cries to presentation of a predator, by laboratory rats living in visible burrow systems. Physiol Behav 50:967–972. pmid:1805287
  9. Bleickardt CJ, Mullins DE, Macsweeney CP, Werner BJ, Pond AJ, Guzzi MF, Martin FDC, Varty GB, Hodgson RA (2009) Characterization of the V1a antagonist, JNJ-17308616, in rodent models of anxiety-like behavior. Psychopharmacol (Berl) 202:711–718. doi:10.1007/s00213-008-1354-x pmid:18923820
  10. Boer GJ (1985) Vasopressin and brain development: studies using the Brattleboro rat. Peptides 6 [Suppl 1]:49–62. pmid:4047982
  11. Boer GJ, Buijs RM, Swaab DF, De Vries GJ (1980) Vasopressin and developing rat brain. Peptides 1:203–209. doi:10.1016/0196-9781(80)90118-7
  12. Boer GJ, Van Rheenen-Verberg CM, Uylings HB (1982) Impaired brain development of the diabetes insipidus Brattleboro rat. Brain Res 255:557–575. pmid:7074363
  13. Bosch OJ, Neumann ID (2012) Both oxytocin and vasopressin are mediators of maternal care and aggression in rodents: from central release to sites of action. Horm Behav 61:293–303. doi:10.1016/j.yhbeh.2011.11.002 pmid:22100184
  14. Bowers JM, Perez-Pouchoulen M, Edwards NS, McCarthy MM (2013) Foxp2 mediates sex differences in ultrasonic vocalization by rat pups and directs order of maternal retrieval. J Neurosci 33:3276–3283. doi:10.1523/JNEUROSCI.0425-12.2013 pmid:23426656
  15. Bredewold R, Smith CJW, Dumais KM, Veenema AH (2014) Sex-specific modulation of juvenile social play behavior by vasopressin and oxytocin depends on social context. Front Behav Neurosci 8:216. doi:10.3389/fnbeh.2014.00216 pmid:24982623
  16. Brot MD, Bernstein IL, Dorsa DM (1992) Vasopressin deficiency abolishes a sexually dimorphic behavior in Brattleboro rats. Physiol Behav 51:839–843. pmid:1594683
  17. Brudzynski SM (2013) Ethotransmission: communication of emotional states through ultrasonic vocalization in rats. Curr Opin Neurobiol 23:310–317. doi:10.1016/j.conb.2013.01.014 pmid:23375168
  18. Brudzynski SM, Ociepa D (1992) Ultrasonic vocalization of laboratory rats in response to handling and touch. Physiol Behav 52:655–660. pmid:1409936
  19. Brudzynski SM, Bihari F, Ociepa D, Fu XW (1993) Analysis of 22 kHz ultrasonic vocalization in laboratory rats: long and short calls. Physiol Behav 54:215–221. pmid:8372113
  20. Burgdorf J, Panksepp J, Beinfeld MC, Kroes RA, Moskal JR (2006) Regional brain cholecystokinin changes as a function of rough-and-tumble play behavior in adolescent rats. Peptides 27:172–177. doi:10.1016/j.peptides.2005.07.005 pmid:16143427
  21. Burgdorf J, Kroes RA, Moskal JR, Pfaus JG, Brudzynski SM, Panksepp J (2008) Ultrasonic vocalizations of rats (Rattus norvegicus) during mating, play, and aggression: behavioral concomitants, relationship to reward, and self-administration of playback. J Comp Psychol 122:357–367. doi:10.1037/a0012889 pmid:19014259
  22. Byers JA, Walker C (1995) Refining the motor training hypothesis for the evolution of play. Am Nat 146:25-40. doi:10.1086/285785
  23. Caldwell HK, Lee H-J, Macbeth AH, Young WS (2008) Vasopressin: behavioral roles of an “original” neuropeptide. Prog Neurobiol 84:1–24. doi:10.1016/j.pneurobio.2007.10.007 pmid:18053631
  24. Cheng S-Y, Delville Y (2009) Vasopressin facilitates play fighting in juvenile golden hamsters. Physiol Behav 98:242–246. doi:10.1016/j.physbeh.2009.04.019 pmid:19414025
  25. Cherry JA (1987) The effect of photoperiod on development of sexual behavior and fertility in golden hamsters. Physiol Behav 39:521–526. pmid:3575500
  26. Covington HE, Miczek KA (2003) Vocalizations during withdrawal from opiates and cocaine: possible expressions of affective distress. Eur J Pharmacol 467:1–13. pmid:12706449
  27. Cuomo V, Cagiano R, De Salvia MA, Maselli MA, Renna G, Racagni G (1988) Ultrasonic vocalization in response to unavoidable aversive stimuli in rats: effects of benzodiazepines. Life Sci 43:485–491. pmid:3136280
  28. Dlouhá H, Krecek J, Zicha J (1982) Postnatal development and diabetes insipidus in Brattleboro rats. Ann N Y Acad Sci 394:10–20. pmid:6960750
  29. Dorsa DM, Bottemiller L (1982) Age-related changes of vasopressin content of microdissected areas of the rat brain. Brain Res 242:151–156. pmid:7104725
  30. Engelmann M, Landgraf R (1994) Microdialysis administration of vasopressin into the septum improves social recognition in Brattleboro rats. Physiol Behav 55:145–149. pmid:8140159
  31. Feifel D, Mexal S, Melendez G, Liu PYT, Goldenberg JR, Shilling PD (2009) The Brattleboro rat displays a natural deficit in social discrimination that is restored by clozapine and a neurotensin analog. Neuropsychopharmacoly 34:2011–2018. doi:10.1038/npp.2009.15 pmid:19322170
  32. Feifel D, Priebe K (2007) The effects of cross-fostering on inherent sensorimotor gating deficits exhibited by Brattleboro rats. J Gen Psychol 134:173–182. doi:10.3200/GENP.134.2.172-182 pmid:17503693
  33. Fodor A, Klausz B, Pintér O, Daviu N, Rabasa C, Rotllant D, Balazsfi D, Kovacs KB, Nadal R, Zelena D (2012) Maternal neglect with reduced depressive-like behavior and blunted c-fos activation in Brattleboro mothers, the role of central vasopressin. Horm Behav 62:539–551. doi:10.1016/j.yhbeh.2012.09.003 pmid:23006866
  34. Heinrichs M, von Dawans B, Domes G (2009) Oxytocin, vasopressin, and human social behavior. Front Neuroendocrinol 30:548–557. doi:10.1016/j.yfrne.2009.05.005 pmid:19505497
  35. Himmler BT, Kisko TM, Euston DR, Kolb B, Pellis SM (2014) Are 50-kHz calls used as play signals in the playful interactions of rats? I. Evidence from the timing and context of their use. Behav Processes 106:60–66. doi:10.1016/j.beproc.2014.04.014 pmid:24811452
  36. Hol T, Van den Berg CL, Van Ree JM, Spruijt BM (1999) Isolation during the play period in infancy decreases adult social interactions in rats. Behav Brain Res 100:91–97. pmid:10212056
  37. Jones P, Rodgers B, Murray R, Marmot M (1994) Child development risk factors for adult schizophrenia in the British 1946 birth cohort. Lancet 344:1398–1402. pmid:7968076
  38. Jordan R (2003) Social play and autistic spectrum disorders a perspective on theory, implications and educational approaches. Autism 7:347–360. doi:10.1177/1362361303007004002 pmid:14678675
  39. Kabitzke PA, Simpson EH, Kandel ER, Balsam PD (2015) Social behavior in a genetic model of dopamine dysfunction at different neurodevelopmental time points. Genes Brain Behav 14:503–515. doi:10.1111/gbb.12233
  40. Karkow ARM, Lucion AB (2013) Mild environmental intervention in mother-infant interactions reduces social play behavior in rats. Psychol Neurosci 6:39–43. doi:10.3922/j.psns.2013.1.07
  41. Kirsten TB, Taricano M, Maiorka PC, Palermo-Neto J, Bernardi MM (2010) Prenatal lipopolysaccharide reduces social behavior in male offspring. Neuroimmunomodulation 17:240–251. doi:10.1159/000290040 pmid:20203530
  42. Klein ZA, Padow VA, Romeo RD (2010) The effects of stress on play and home cage behaviors in adolescent male rats. Dev Psychobiol 52:62–70. doi:10.1002/dev.20413 pmid:19937741
  43. Knutson B, Burgdorf J, Panksepp J (1998) Anticipation of play elicits high-frequency ultrasonic vocalizations in young rats. J Comp Psychol 112:65–73. pmid:9528115
  44. Knutson B, Burgdorf J, Panksepp J (1999) High-frequency ultrasonic vocalizations index conditioned pharmacological reward in rats. Physiol Behav 66:639–643. pmid:10386908
  45. Kundakovic M, Champagne FA (2015) Early-life experience, epigenetics, and the developing brain. Neuropsychopharmacology 40:141–153. doi:10.1038/npp.2014.140 pmid:24917200
  46. Lin RE, Ambler L, Billingslea EN, Suh J, Batheja S, Tatard-Leitman V, Featherstone RE, Siegel SJ (2013) Electroencephalographic and early communicative abnormalities in Brattleboro rats. Physiol Rep 1:e00100. doi:10.1002/phy2.100 pmid:24303172
  47. Lukas M, Wöhr M (2015) Endogenous vasopressin, innate anxiety, and the emission of pro-social 50-kHz ultrasonic vocalizations during social play behavior in juvenile rats. Psychoneuroendocrinology 56:35–44. doi:10.1016/j.psyneuen.2015.03.005 pmid:25800147
  48. Meaney MJ, Stewart J (1981a) A descriptive study of social development in the rat (Rattus norvegicus). Anim Behav 29:34–45.
  49. Meaney MJ, Stewart J (1981b) Neonatal-androgens influence the social play of prepubescent rats. Horm Behav 15:197–213. pmid:7250910
  50. Olesen KM, Jessen HM, Auger CJ, Auger AP (2005) Dopaminergic activation of estrogen receptors in neonatal brain alters progestin receptor expression and juvenile social play behavior. Endocrinology 146:3705–3712. doi:10.1210/en.2005-0498 pmid:15919740
  51. Panksepp J (1981) The ontogeny of play in rats. Dev Psychobiol 14:327–332. doi:10.1002/dev.420140405 pmid:7250521
  52. Panksepp J, Beatty WW (1980) Social deprivation and play in rats. Behav Neural Biol 30:197–206. pmid:7447871
  53. Panksepp J, Burgdorf J (2000) 50-kHz chirping (laughter?) in response to conditioned and unconditioned tickle-induced reward in rats: effects of social housing and genetic variables. Behav Brain Res 115:25–38. pmid:10996405
  54. Parent CI, Meaney MJ (2008) The influence of natural variations in maternal care on play fighting in the rat. Dev Psychobiol 50:767–776. doi:10.1002/dev.20342 pmid:18846499
  55. Paul MJ, Terranova JI, Probst CK, Murray EK, Ismail NI, De Vries GJ (2014) Sexually dimorphic role for vasopressin in the development of social play. Front Behav Neurosci 8:58. doi:10.3389/fnbeh.2014.00058 pmid:24616675
  56. Pellegrini AD (1988) Elementary-school children’s rough-and-tumble play and social competence. Dev Psychol 24:802–806. doi:10.1037/0012-1649.24.6.802
  57. Pellis SM (2002) Sex differences in play fighting revisited: traditional and nontraditional mechanisms of sexual differentiation in rats. Arch Sex Behav 31:17–26. pmid:11910788
  58. Pellis SM, Pellis VC (1990) Differential rates of attack, defense, and counterattack during the developmental decrease in play fighting by male and female rats. Dev Psychobiol 23:215–231. doi:10.1002/dev.420230303 pmid:2379760
  59. Pellis SM, Pellis VC (1998) Play fighting of rats in comparative perspective: a schema for neurobehavioral analyses. Neurosci Biobehav Rev 23:87–101. pmid:9861614
  60. Radziwon KE, Dent ML (2014) Frequency difference limens and auditory cue trading in CBA/CaJ mice (Mus musculus). Behav Processes 106:74–76. doi:10.1016/j.beproc.2014.04.016 pmid:24811450
  61. Rubin LH, Carter CS, Bishop JR, Pournajafi-Nazarloo H, Drogos LL, Hill SK, Ruocco AC, Keedy SK, Reilly JL, Keshavan MS, Pearlson GD, Tamminga CA, Gershon ES, Sweeney JA (2014) Reduced levels of vasopressin and reduced behavioral modulation of oxytocin in psychotic disorders. Schizophr Bull 40:1374–1384. doi:10.1093/schbul/sbu027 pmid:24619535
  62. Sales GD (1972) Ultrasound and aggressive behaviour in rats and other small mammals. Anim Behav 20:88–100. pmid:4677167
  63. Scattoni ML, Crawley J, Ricceri L (2009) Ultrasonic vocalizations: a tool for behavioural phenotyping of mouse models of neurodevelopmental disorders. Neurosci Biobehav Rev 33:508–515. doi:10.1016/j.neubiorev.2008.08.003 pmid:18771687
  64. Schank JC (2009) Early locomotor and social effects in vasopressin deficient neonatal rats. Behav Brain Res 197:166–177. doi:10.1016/j.bbr.2008.08.019 pmid:18786571
  65. Schmale H, Richter D (1984) Single base deletion in the vasopressin gene is the cause of diabetes insipidus in Brattleboro rats. Nature 308:705–709. pmid:6717565
  66. Snijdewint FG, Boer GJ, Swaab DF (1988) Prenatal development of the Brattleboro rat is influenced by genotype and lysine vasopressin treatment of the mother. Biol Neonate 53:295–304. pmid:3390492
  67. Spear LP (2000) The adolescent brain and age-related behavioral manifestations. Neurosci Biobehav Rev 24:417–463. pmid:10817843
  68. Taylor PV, Veenema AH, Paul MJ, Bredewold R, Isaacs S, de Vries GJ (2012) Sexually dimorphic effects of a prenatal immune challenge on social play and vasopressin expression in juvenile rats. Biol Sex Differ 3:15. doi:10.1186/2042-6410-3-15 pmid:22697211
  69. Thor DH, Holloway WR (1984) Sex and social play in juvenile rats (Rattus norvegicus). J Comp Psychol 98:276–284. doi:10.1037/0735-7036.98.3.276
  70. Tonoue T, Ashida Y, Makino H, Hata H (1986) Inhibition of shock-elicited ultrasonic vocalization by opioid peptides in the rat: a psychotropic effect. Psychoneuroendocrinol 11:177–184. pmid:2428074
  71. Valtin H, Schroeder HA (1964) Familial hypothalamic diabetes insipidus in rats (Brattleboro strain). Am J Physiol 206:425–430. pmid:14120453
  72. Van Den Berg CL, Hol T, Van Ree JM, Spruijt BM, Everts H, Koolhaas JM (1999) Play is indispensable for an adequate development of coping with social challenges in the rat. Dev Psychobiol 34:129–138. pmid:10086231
  73. Vanderschuren LJ, Niesink RJM, Van Ree JM (1997) The neurobiology of social play behavior in rats. Neurosci Biobehav Rev 21:309–326. pmid:9168267
  74. Varga J, Fodor A, Klausz B, Zelena D (2015) Anxiogenic role of vasopressin during the early postnatal period: maternal separation-induced ultrasound vocalization in vasopressin-deficient Brattleboro rats. Amino Acids 47:2409–2418. doi:10.1007/s00726-015-2034-x pmid:26133736
  75. Veenema AH, Neumann ID (2009) Maternal separation enhances offensive play-fighting, basal corticosterone and hypothalamic vasopressin mRNA expression in juvenile male rats. Psychoneuroendocrinology 34:463–467. doi:10.1016/j.psyneuen.2008.10.017 pmid:19056182
  76. Veenema AH, Bredewold R, De Vries GJ (2012) Vasopressin regulates social recognition in juvenile and adult rats of both sexes, but in sex- and age-specific ways. Horm Behav 61:50–56. doi:10.1016/j.yhbeh.2011.10.002 pmid:22033278
  77. Veenema AH, Bredewold R, De Vries GJ (2013) Sex-specific modulation of juvenile social play by vasopressin. Psychoneuroendocrinology 38:2554–2561. doi:10.1016/j.psyneuen.2013.06.002 pmid:23838102
  78. Vetter-O’Hagen CS, Spear LP (2012) Hormonal and physical markers of puberty and their relationship to adolescent-typical novelty-directed behavior. Dev Psychobiol 54:523–535. doi:10.1002/dev.20610 pmid:21953609[Mismatch]
  79. Vivian JA, Miczek KA (1991) Ultrasounds during morphine withdrawal in rats. Psychopharmacol (Berl) 104:187–193. pmid:1876663
  80. Willey AR, Spear LP (2012) Development of anticipatory 50 kHz USV production to a social stimuli in adolescent and adult male Sprague-Dawley rats. Behav Brain Res 226:613–618. doi:10.1016/j.bbr.2011.10.001 pmid:22004980
  81. Willey AR, Varlinskaya EI, Spear LP (2009) Social interactions and 50 kHz ultrasonic vocalizations in adolescent and adult rats. Behav Brain Res 202:122–129. doi:10.1016/j.bbr.2009.03.025 pmid:19447288
  82. Winslow JT, Insel TR (1993) Effects of central vasopressin administration to infant rats. Eur J Pharmacol 233:101–107. pmid:8472738
  83. Wöhr M, Schwarting RKW (2007) Ultrasonic communication in rats: can playback of 50-kHz calls induce approach behavior? PLoS One 2:e1365. doi:10.1371/journal.pone.0001365 pmid:18159248
  84. Wöhr M, Schwarting RKW (2013) Affective communication in rodents: ultrasonic vocalizations as a tool for research on emotion and motivation. Cell Tissue Res 354:81–97. doi:10.1007/s00441-013-1607-9 pmid:23576070
  85. Wright JM, Gourdon JC, Clarke PBS (2010) Identification of multiple call categories within the rich repertoire of adult rat 50-kHz ultrasonic vocalizations: effects of amphetamine and social context. Psychopharmacology (Berl) 211:1–13. doi:10.1007/s00213-010-1859-y
  86. Zelena D, Mergl Z, Makara GB (2003) Maternal genotype influences stress reactivity of vasopressin-deficient Brattleboro rats. J Neuroendocrinol 15:1105–1110. pmid:14636171
  87. Zelena D, Domokos A, Barna I, Mergl Z, Haller J, Makara GB (2008) Control of the hypothalamo-pituitary-adrenal axis in the neonatal period: adrenocorticotropin and corticosterone stress responses dissociate in vasopressin-deficient Brattleboro rats. Endocrinology 149:2576–2583. doi:10.1210/en.2007-1537 pmid:18276753
  88. Zelena D, Mergl Z, Makara GB (2009) Postnatal development in vasopressin deficient Brattleboro rats with special attention to the hypothalamo-pituitary-adrenal axis function: the role of maternal genotype. Int J Dev Neurosci 27:175–183. doi:10.1016/j.ijdevneu.2008.11.003 pmid:19059472

Synthesis

The decision was a result of the Reviewing Editor Jeffrey Blaustein and the peer reviewers coming together and discussing their recommendations until a consensus was reached. A fact-based synthesis statement explaining their decision and outlining what is needed to prepare a revision is listed below. The following reviewers agreed to reveal their identity: Yvon Delville, J. Curtis.

In general, the reviewers believe that the manuscript has merit. However, there are a number of important comments (see below). However after consultation, the reviewers and editor agree with the following important points:

1. The authors must either perform a more complete analysis of USVs as they relate to social play and other events or provide a compelling argument as to why this is not necessary. The reviewers suggest that the numbers of USVs should be adjusted to social play time or events.

2. Considering the shortcomings of using the Brattleboro rat to draw conclusions about the role of VP, the authors should be more cautious in attributing the effects to VP directly. Terms like "suggest" would be more appropriate.

3. A caveat regarding the possibility that there is a role of stress in behavioral phenotype is essential for reasons delineated in the reviews. The authors need to provide an explanation re: why the reader should not be concerned about the confound of testing in a new environment.

Reviews:

Reviewer 1

Major Issues:

1. The authors seem to like using the word "demonstrate" in their Abstract, Introduction and Discussion sections. I strongly recommend using more cautious words for this particular study. The data are consistent with a role of vasopressin in the control of these behaviors, but they do not demonstrate a hypothesis.

2. The reason these data do not demonstrate the role of vasopressin in these behaviors has to do with the type of study carried out by the authors. In essence, this study is similar to gene knockout experiments. Yes, a gene is not expressed, but the behavioral outcome does not necessarily prove a behavioral role for the missing peptide. Vasopressin is only one of several neurotransmitters or neuromodulators mediating social play and ultrasonic vocalizations. For instance, vasopressin is well known to interact with serotonin which is also associated with social behavior. The absence of vasopressin throughout development may result in over-expression of serotonin, which could be the direct cause explaining the behavior of Brattleboro rats. Over time, compensatory systems could also take over for vasopressin. I think the authors are lucky to be working within a developmental time period when serotonin systems are not fully mature. While the authors did acknowledge the possibility of compensatory systems either within or outside of the brain on p. 20, this sections could be developed further or at the very least the authors need to acknowledge that their data do not demonstrate a role for vasopressin.

3. As noted by the authors social play and the role of vasopressin in this behavior is context-dependent. I am glad the authors addressed behavior before and after weaning. However, I am not entirely satisfied with the behavioral procedures. Before weaning, animals were taken out of their litters and placed in a new environment and tested with a strange (unknown) individual. Obviously, if the animals respond to this type of stress differently, their social behavior would also be different. Rats are social animals. Being isolated is stressful to them, particularly early in development. The authors need to address possible differential responses to stress as a possible confounding variable in their experimental design. Perhaps observations of the entire litters would be useful, as long as it is possible to track specific individuals inside the litters. (JDB: Possibility of stress interaction should be discussed.)

4. The procedure used to test animals after weaning is not entirely trouble-free either. The issue of isolation stress is also relevant here. Typically a period of isolation enhances play fighting in rats, and that could be different in Brattleboro rats. Maybe the behavioral differences are not specifically related to social play itself. This other confounding variable needs to be addressed as well. Personally I would have preferred observations between cagemates inside their home cages, or in response to an intruder. (JDB: Possibility of stress interaction should be discussed.)

5. It would be useful to provide some information about the distribution of the genotypes inside the litters. Could variability in the datasets originate from variability in the genotypes? (JDB: This will not require a full analysis, just mention of the distribution and speculation re: variability in genotype.)

6. I am very pleased to note the complex analysis of ultrasonic recordings. It would have been nice to match it with an equally complex analysis of social play observations. Social interactions in juvenile or infant rats are not limited to huddling, attacks, pins, and allogrooming. I would recommend future studies to include defensive components of the behavior as described by Sergio Pellis.

Minor Issues:

1. There is no need to explain the outcome of the study at the end of the Introduction section. This is anticlimatic for the reader. A hypothesis or predictions would be sufficient.

2. The beginning of the Results section (line 186) mentions a consistency with other reports. Where are the references?

Perhaps a publication by Sergio Pellis? These are missing from the reference list.

3. I am not convinced the tables of ANOVAs are really necessary. They are just making me worried that some analyses will come up significant by chance.

4. line 391. from not form.

5. I am not sure I see any relevance for autism in these studies. As noted by the authors, the Brattleboro rats were not asocial. I would recommend the authors be more careful when relating to mental disorders in humans.

Reviewer 2

Minor details:

In figure 3c, the y-axis title is "#Soc. Inv./Allo. Behaviors", which at first glance suggests a ratio of the two measures. It might be clearer to use “and” or “&”.

“Emit” is misspelled in the caption for figure 4.

The Byers and Walker ref is missing page numbers

In the Feifel et al. ref, “brattelboro” should be capitalized unless it was lower case in the original

In general, go through the references to ensure consistency of capitalization, abbreviation of journal names, etc.

I am going to assume that the figures will be published in color. If not, the bars should be rendered more distinguishable. (JDB: Because readers often print in black and white, enhancing the contrast when printed black and white is desirable.)

Genotype is misspelled on lines a8 and b8 of table 1

Since the data were collected from different groups of animals at each time point, it would be more appropriate to describe the groups as e.g., P34 “or” P44, rather than as P34 “and” P44. The latter implies the same animals were tested at both time points

Other comments

The authors cite Himmler's report of sex differences in USVs before or after play bouts, suggesting that the timing of vocalizations may contain information. I don't know if the data collection methods would allow for assessment of the timing of USVs relative to the onset (or termination) of the various social interactions, but doing so may provide further insights. Similarly, it might be useful to see whether specific types of calls consistently accompany particular play behaviors. I will leave it to the authors/ editor to decide whether this is worth pursuing.

It also might be interesting to test whether the AVP-deficient rats differ in their latency to exhibit particular behaviors, especially when the frequency of a behavior decreases. Is it possible that the motivation to initiate playing differs, but that once initiated, play behavior does not differ between genotypes? As above, I will leave it to the authors/editor to decide whether this is worth pursuing. (JDB: This is something for the authors to consider, but not required by Editor.)

Although the authors mention differences Het and WT individuals in a few cases, relatively little attention is paid to this type of result. Is it possible that a “dose/response” effect may come into play. For example, in figure 2, it appears that the Het responses for both Pins and Attacks at day 44 may be “dose-dependent” in males but not in females. In figure 1, a similar pattern is seen, but for both sexes, and with the Het group significantly different from the Hom and WT groups for pins and nearly so for total play at 21 days. Once again, I will leave it to the authors/editor to decide whether this is worth pursuing. (JDB: This is something for the authors to consider, but not required by Editor.)