Characterization of Vocalizations Emitted in Isolation by California Mouse (Peromyscus californicus) Pups throughout the Postnatal Period

Abstract

Rodent species, such as monogamous and biparental California mice, produce vocalizations as a means of communication. A temporal examination of vocalizations produced by California mice pups in isolation was performed. Pup recordings were performed for 3 minutes at ~10.00 and 14.00 hrs on early postnatal days (PND) 2-4, 7, 21, and 28. Once initial recordings were finished, pups were returned to the home cage with parents and any siblings for five minutes to determine if active biparental responses resulted in an enhanced vocalization response when pups were isolated and retested. We also sought to determine whether potential reduction in vocalizations by older pups might be due to procedure-habituation procedure associated with less anxiety and/or possibly decreased need for parental care. Vocalizations were measured in weanling (30 days of age) “naïve” pups not previously isolated. Results show older pups took significantly longer to vocalize, indicated by increased latency before producing their initial syllable compared to earlier ages. With increasing age, pups demonstrated decreased syllable duration, reduced number and duration of phrases, and decreased number of syllables per phrase. No differences in pup vocalizations were observed before and after being placed back with parents, suggestive biparernal potentiation may not exist in California mice pups. Comparison of the naïve to habituated weanling pups indicates the former group had more total calls but no other differences in vocalization parameters were detected between these two groups. Collectively, the findings suggest that as California mice pups mature and approach weaning they generally vocalize less in isolation.

Introduction

Ultrasonic vocalizations (USVs, sounds above 20 kHz) are considered one means of communication in various rodent species, including California mice (Peromyscus californicus). Previous studies have investigated communication patterns in pups (Vieira & Brown, 2002; Wright & Brown, 2004), and adults (Kalcounis-Rueppell et al., 2010). In the wild, different adult Peromyscus species may produce four motifs (types) of USVs that include 1) syllable vocalization (SV) that consist of phrases with one to three syllables with at least 50 milliseconds between syllables, 2) bark syllables that are upside down chevrons or U-shaped calls that are shorter than SV and have harmonics, 3) simple sweeps that extend from higher to lower frequencies, and 4) complex sweep that are short syllables that extend from higher to lower then again to higher frequencies in the same syllable (Briggs & Kalcounis-Rueppell, 2011; Kalcounis-Rueppell, Metheny, & Vonhof, 2006; Pultorak, Fuxjager, Kalcounis-Rueppell, & Marler, 2015). Most rodent studies, including those with California mice, have examined USV production during the neonatal period, although it is clear that juvenile and adult animals can also produce USVs (Briggs & Kalcounis-Rueppell, 2011; Campbell, Pasch, Warren, & Phelps, 2014; Portfors, 2007; Pultorak et al., 2015; Vieira & Brown, 2002; Wright & Brown, 2004). In adults, these vocalizations can be context- and state- dependent and used to communicate identity, emotion, social status, habitat conditions- such as the presence of a predator in the surrounding area or food source, and as a means to stimulate a response by others in the group (Portfors, 2007).

One hypothesis that has been put forward for the function of pup USVs is to induce parental care (Geyer, 1981; Portfors, 2007; Wright & Brown, 2004). This may be the case in pups who cannot regulate their own body temperature until later in postnatal life, such as neotropical singing mice (Scotinomys spp.) (Campbell et al., 2014) and California mice (Gubernick & Alberts, 1987; Gubernick & Teferi, 2000; Rosenfeld, 2013). California mice are monogamous and biparental (Chandler, 1979; Eisenberg, 1962, 1963; Rosenfeld, 2013). It has been postulated that paternal care evolved in this species to allow females increased time outside of the nest to forage for food (Dudley, 1974a, 1974b). During this foraging, the male partner huddles over the pups to prevent a major drop in body temperature- pups are exothermic until about PND 15 (Gubernick & Alberts, 1987; Rosenfeld, 2013). If the primary purpose of pup vocalizations is to elicit maternal or paternal care and warmth, the prediction would be that this communication would decrease once the pups can regulate their own body temperature.

Besides warmth, pup USVs may also serve to strengthen the bond between pups and parent(s), stimulate lactation, and promote licking and grooming of the pups (Brouette-Lahlou, Vernet-Maury, & Vigouroux, 1992; Stern, Thomas, Rabii, & Barfield, 1984). Thus, interruption of the parental-pup bond may result in a social stress and trigger pup vocalizations. Parental licking and grooming might also be associated with pup vocalizations. Other stimulatory factors that induce pup USVs syllables include isolation, cold exposure, handling, presence of home bedding material that presumably includes olfactory cues from the parents (Kapusta, Marchlewska-Koj, & Sales, 1995). Pup vocalizations may result in increased parental behaviors, such as retrieval, by the mother or father (Ehret, 2005; Ehret & Haack, 1981; Ehret & Haack, 1984). Therefore, pup USVs may be considered a form of communication. However, one study suggests that rat pups in the home cage produce negligible amount of USVs, but these types of vocalizations increase when they are placed in isolation (Hofer & Shair, 1978). Comparably, home-cage recordings of albino rat pups for the first 21 days post-partum shows that they produce 9,000 sonic calls but only 134 ultrasonic calls for the first 18 days (McLinn, 1977). The sonic pup calls peeked when the dam shifted herself on the nest but declined when she was nursing or out of contact with the pups.

Other reports with various rodent models, including California mice, vole, and laboratory mice and rat species, indicate that pup USVs increase during the first few days after birth, peak in the first few weeks after birth, and then decline until they cease vocalizing at the time of weaning (Broom, 1977; Campbell et al., 2014; De Ghett, 1974; Naito & Tonoue, 1987; Pontet, 1989; Vieira & Brown, 2002). However, these earlier studies with California mice were performed prior to recent advancements in detecting the full range of rodent USVs. Thus, it was only possible to detect selected ultrasonic frequencies, such as those spanning from 37 to 52 kHz (Vieira & Brown, 2002; Wright & Brown, 2004).

To provide further details on the ontogeny of California mice pup audible and USVs, the current experiments employed a microphone and software program that detects and measures sounds up to 200 kHz. Thus, the main goal of the study was to characterize the ontogeny of vocalizations produced by isolated California mice pups. The hypothesis being that isolated California mice pups would vocalize less as they matured, which could be due to habituation and reduced anxiety from being separated from the parents or represent the fact that with age, the pups required less parental care. The current studies though were not directed at disentangling the motivation why pup vocalizations may decrease with age, only to test whether they do so. Additionally, we sought to determine whether pups would vocalize more after being returned to their parents and then isolated again, otherwise considered a potentiated response (Hofer, Masmela, Brunelli, & Shair, 1999; Myers et al., 2004; Shair, Brunelli, Masmela, Boone, & Hofer, 2003). In essence, maternal (in the case of mice or rats) or both maternal and paternal (for California mice and prairie voles- Microtus ochrogaster) may occur when there is a brief interaction with one or both parents, respectively. It has been previously shown in rats pups that passive (anesthetized dam) results in an enhanced vocal response to isolation in the second to third week of life (PND 10-15) (Myers et al., 2004; Shair et al., 2003). Even though prairie voles are monogamous and biparental, it has recently been shown that only passive (anesthetized) and active maternal potentiation but not paternal potentiation exists in this species (Robison, Myers, Hofer, Shair, & Welch, 2016). In the current set of studies, California mice pups were returned to their awake dam and sire, and thus, the studies were only aimed at testing whether maternal and paternal potentiation exists in this species with no attempt to distinguish one from the other.

Finally, by testing weanling mice that have not previously been isolated from their parents, the current studies tested whether older pups in isolation vocalize less because they become habituated and less anxious to the isolation conditions or whether older pups vocalize less in isolation because they require less parental care. If it is the former case, the prediction would be that older pups placed for the first time in isolation will vocalize to the same extent as younger pups. However, in the latter case, weanling pups placed for the first time in isolation should demonstrate similar vocalizations as older pups that were repeatedly tested throughout the post-natal period.

Materials and Methods

Animal Husbandry

Outbred adult (60–90 days of age) California mouse females and males, who were pathogen-free, were originally purchased from the Peromyscus Genetic Stock Center (PGSC) at the University of South Carolina (Columbia, SC). These animals were placed in quarantine at the University of Missouri for a minimum of 8 weeks to verify their disease-free status. All experiments were approved by University of Missouri Animal Care and Use Committee (Protocol #7753) and performed in accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. Adult breeder females were maintained on a low phytoestrogen AIN 93G diet supplemented with 7% by weight corn oil to minimize potential phytoestrogenic contamination. Females were paired with adult males also fed this same diet. As Peromyscus species do not form a visible plug, the females were weighed weekly to provide some idea of when they might give birth. The day of birth was considered post-natal day 0. Animals were ad libitum provided water and food, and the lights were automatically turned on at 7.00 hrs and turned off at 19.00 hrs.

Ultrasonic Vocalization Measurements and Potentiation Experiments

In preliminary studies, we attempted to measure pup vocalizations when they were in the home cage with their parents. However, as California mice pups latch onto to the dam, it was difficult to discern individual pup vocalizations and vocalizations emitted by pups vs. parents. For these reasons, each individual pup was tested in isolation and away from their home-cage. Pup USVs and potentiation experiments were begun on PND 2. In litters with more than one pup, each pup was demarcated with a distinguishing tattoo mark on either the front or back paw (Fine Science Tools, Foster City, CA). California mice can only accommodate up to four pups but the average litter size is 2.2, and the pups are weaned at 30 days of age (Johnson et al., 2015; Rosenfeld, 2013). All pups in the litter were recorded throughout the entire postnatal period, provided they survived until this time-point. In two litters, one of the two pups died prior to weaning, and thus, only partial data were obtained for these individuals. A total of 11 pups representing nine litters were analyzed. Pup recordings were performed at ~10.00 hrs (3 hrs after lights turned on in the room) and 14.00 hrs (7 hrs after lights turned on in the room). Prior to being placed in the recording box, the experimenter noted whether they were or were not latched on to one of the dam's four teats (Table S1). The pups were placed individually in a polypropylene box (22 inches by 27 inches) that was surrounded on three sides with a 20 inch by 20 inch black polypropylene box lined with 2 inches of convoluted acoustic foam panel (SoundproofCow, Chambersburg, PA) (Figure S1). The pups were placed on a laminated sheet, which could be easily disinfected, and the sheet demarcated pup and litter information, date, and time of day when testing was done. No substrate material, including shavings or bedding material, was present in the recording box. The box was equipped with an Avisoft Bioacoustics CM16/CMPA40-5V microphone (Glienicke, Germany) plugged into an National Instruments USB 6351 data collection board hooked up to a PC. USV data were collected using a custom LabView data collection code written by the laboratory of Dr. Katrin Schenk at Randolph College. A Logitech Carl Zeiss Tessar HD 1080P (Newark, CA) camera was mounted onto a Joby- Gorilla Pod Original Tripod (Daymen US Inc., Petaluma, CA) in the recoding box to monitor isolated pups. Pups were audio-recorded for three minutes, which minimized the time away from their parents but permitted us to collect relevant data for the time period assessed. The pup was then placed back in the home cage with both parents for five minutes.

Afterwards, the pup was isolated again and returned to the audio-recording box and its vocalizations recorded for another three minutes. As per the requirements of our Animal Care and Use Committee, the audio-recording box was sprayed down only between trials with 70% ethanol for disinfection and to remove any potential confounding olfactory cues for subsequent pups tested. It is possible though that the pups were able to smell the residual ethanol, but this would be the case for all pups and time-points tested. In litters with more than one pup, the remaining pups in the litter were then tested in the same manner. Pup vocalizations were measured on PND 2, 3, 4, 7, 14, 21, and 28. Recordings on PND 2-4 were grouped into `early' time points. The other days were considered individually. These time-points were chosen to span the early, mid, and late postnatal periods and key milestones in California mice neonatal development. Based on our observations, the fur begins to emerge and darken between two to four days of age. Our assessments and published results indicate eyes open around 14–15 days of age in this species (Harris, de Jong, Yang, & Saltzman, 2013; Vieira & Brown, 2003). With both mothers and fathers in the cage, as was the case in the current set of experiments, California mice pups begin walking and rearing around 22 and 22.75 days of age (Vieira & Brown, 2003). Pups raised by both parents begin self-grooming around 25 days of age. California mice pups are weaned at 30 days of age and eat solid food at this time (Rosenfeld, 2013).

Vocalizations were detected and measured by Dr. Schenk's program, which allows USVs up to 200 KHz to be detected and analyzed. For each recording, all syllables were segmented for analysis. Number of syllables, syllable duration, syllable's median frequency, average syllable power, power percent below 20 kHz, and power percent above 20 kHz were determined. Syllable power refers to the unscaled decibels for all signals passing a threshold of three SD above the noise level, averaged over bins 390 Hz wide, and then averaged over the entire syllable. In essence, it measures the relative loudness of the syllables. Power percent above and below 20 kHz were measured to determine whether the power was focused in the audible spectrum (below 20 kHz) or the ultrasonic spectrum (above 20 kHz). The noise level was determined by averaging the power in the spectrogram for a region of the recording that had no calls or extraneous noise (this region was detected by eye). A spectrogram bins acoustic power into time and frequency bins. A syllable will have a number of time bins that depends on its length and a number of frequency bins which depends on the number of Fourier components used to decompose the sound. To determine the percentage of power contained in frequencies above and below 20 kHz for a call, we first detected which frequency bins contained signal with power that was higher than noise threshold for each time-bin within the call. This power was then added up separately for frequency bins above 20 kHz and frequency bins below 20 kHz. These were then summed over all time-bins contained in the syllable. To calculate the percentage of power above and below 20 kHz, we divided these sums by the total power in the call that passed the noise threshold.

Additionally, the number of phrases, which refers to a group of syllables that are separated from another group of syllables by an interval in time that is statistically longer the mean time between individual syllables, duration of phrases, and number of syllables in a phrase were measured. Figure S2 demonstrates various California mice pup motifs in both the audible and ultrasonic ranges that are similar to those observed in mice pups, as shown in Figure 2 of Grimsley et al. (Grimsley, Monaghan, & Wenstrup, 2011). At least three out of the four motifs identified in the current set of studies with California mice pups have been previously identified in adults of this species (Briggs & Kalcounis-Rueppell, 2011; Kalcounis-Rueppell et al., 2006; Pultorak et al., 2015).

Figure 2.

Latency to first syllable and syllable duration. A) Latency to first syllable. B) Mean Duration of Syllables. Letter superscripts indicate that the bars are significantly different. Actual p value comparisons are provided in table alongside each graph.

Measurement of Vocalizations Produced in Isolation by Naïve Weanling Pups

At weaning (30 days of age), a separate cohort group of California mice pups (n = 10 pups from four litters) were isolated in the morning and afternoon and vocalizations measured as detailed above. In contrast to the pups tested above, these naïve pups remained with their parents throughout the post-natal period and were only isolated and vocalizations measured at weaning.

Statistics

Data were analyzed by using SAS version 9.2 Software (SAS Institute, Cary, NC). Partial eta squared values (η²_p) were calculated as measure of effect size. For the pup recordings, an ANOVA with a randomized complete block design in which the litter served as the block was used. The variables were arranged as a 3 by 2 by 2 factorial (5 grouped/individual days, two times- AM and PM, and two trials- pre and post being placed back with parents). Mean differences were determined using Fisher's Protected Least Significant Difference (LSD). In this case, the LSD is only calculated if the overall F test is significant and is considered an acceptable method under these conditions (Chew & United, 1977). For the experiments where pup vocalizations were only measured at weaning, the data was analyzed as detailed above, but there were only two time points- AM and PM on the day of weaning and two trials- pre and post-potentiation (i.e., being returned to the home cage with both parents). These weanling results were compared to the last test day measured for the pups who were repeatedly isolated and vocalizations measured to determine if there were differences in vocalizations between older pups that were repeatedly isolated vs. those isolated for the first time at weaning. All data are presented as the $\stackrel{‒}{x}\pm \text{SEM}$.

Results

Vocalizations by Pups Repeatedly Isolated

General features of California mice pup vocalizations

For latency to first syllable there was a significant effect of trial day (age of the pups) (F(4,87) = 4.90, η²_p=0.10, p= 0.001). It took longer on days 21 and 28 than on days 2-4 (p=0.003 to 0.004), day 7 (p=0.002), and day 14 (p=0.01 to 0.02) for the pups to initiate their first syllable (Figure 2A). Trial day also effected syllable duration (F(4,78) = 39.62, η²_p=0.73, p = 0.0001, Figure 2B). Syllable duration gradually decreased over all trial days with PND 2-4 longer than PND 14 (p=0.0002), PND 21 (p=0.0001), and PND 28 (p=0.0001). Syllable duration was longer on PND 7 than PND 14 (p=0.009), PND 21 (p=0.0001), and PND 28 (p=0.0001). For this parameter, PND 14 was longer than PND 21 (p=0.0001) and PND 28 (p=0.0001). Syllable duration on PND 21 was also significantly greater than on PND 28 (p=0.04). Thus, these results show that mean duration of syllables steadily declined from PND 2-4 to PND 28.

Median syllable frequency was also affected by trial day (F(4,78) = 21.29, η²_p = 0.39, p = 0.0001, Figure 3A). The median syllable frequency was significantly lower on PND 2-4 than PND 7 (p=0.0001) and PND 14 (p=0.002), but higher on PND 2-4 than PND 21 (p=0.01) and PND 28 (p=0.05). The mean syllable frequency was greater on PND 7 than PND 14 (p=0.03), 21 (p=0.0001) and 28 (p=0.0001). PND 14 was also significantly higher than day 21 (p=0.0001) and 28 (p=0.0001). There was also a significant effect for trial day to impact the total number of syllables (F(1,104) = 11.85, η²_p = 0.18, p= 0.0001). The total number of syllables was significantly higher on PND 2-4 compared to PND 21 (p=0.0002) and 28 (p=0.0001) (Figure 3B and Table S2). The total number of syllables was greater on PND 7 than PND 21 (p=0.0001) and PND 28 (p=0.0001). For this category, PND 14 was greater than PND 21 (p=0.0001) and 28 (p=0.0001).

Figure 3.

Mean syllable frequency and total number of syllables. A) Median syllable frequency. B) Total number of syllables. Letter superscripts indicate that the bars are significantly different. Actual p value comparisons are provided in table alongside each graph.

Syllable power results

The average syllable power was significantly affected by day (F(4,78) = 11.33, η²_p = 0.18, p = 0.0001). This syllable power was greater on PND 14 compared to PND 7 (p=0.009). On PND 21, the average syllable power was increased compared to PND 2-4 (p=0.0001), PND 7 (p=0.0001), PND 14 (p=0.0004), and PND 28 (p=0.0004) (Figure 4A). The average syllable power was also increased on PND 28 compared to PND 7 (p=0.05). In essence, the average syllable power intensified from PND 2-4 to PND 21 and then slightly declined. However, this parameter was still elevated on PND 28 relative to earlier post-natal days.

Figure 4.

Syllable power results. A) Average syllable power. B) Percentage of syllable power above 20 kHz. Letter superscripts indicate that the bars are significantly different. Actual p value comparisons are provided in table alongside each graph.

The percentage of syllable power above 20 kHz for each syllable was significantly impacted by day (F(4,78) = 9.35, η²_p = 0.31, p = 0.0001). The percentage of syllable power above 20 kHz (Figure 4B) was higher on day 7 than on all other days (p value range=0.01–0.0001). Call power above 20 kHz was also greater on PND 14 than PND 2 (p=0.04) and PND 21 (p=0.02). Collectively, the average call power above 20 kHz peaked on PND 7 and then declined on the later days examined.

Phrase results

Trial day significantly affected the average number of phrases (F(4,104) = 10.48, η²_p = 0.22, p = 0.0001). There were fewer phrases produced on PND 21 and 28 than the other days examined (p=0.001–0.0001, Figure 5A). The mean number of syllables in each phrase was affected by day (F(4,104) = 17.32, η²_p = 0.26, p < 0.0001). Pups produced greater average number of syllables per phrase on days 2-4 compared to PND 14 (p=0.01), 21 (p=0.0001), and day 28 (p=0.0001) (Figure 5B). On PND 7 and 14, pups also produced fewer syllables per phrase than on day 21 (p<0.002) and day 28 (p≤0.003).

Figure 5.

Phrase results. A) Average number of phrases. B) Average number of syllables in each phrase. C) Average phrase duration. Letter superscripts indicate that the bars are significantly different. Actual p value comparisons are provided in table alongside each graph.

The average duration of each phrase was also significantly altered by day (F(4,104) = 17.95, η²_p = 0.21, p = 0.0001). On PND 2-4, the average phrase duration was increased compared to all other trial days, PND 7 (p=0.02), 14 (p=0.0001), 21 (p=0.0001) and 28 (p=0.0001) (Figure 5C). Phrase duration was also higher on PND 7 than PND 21 (p=0.0004) and 28 (p=0.0008).

Maternal and Paternal Potentiation

There were no differences for any of the vocalization parameters above after the pups were placed back with their parents and then isolated again for the re-testing period compared to the pre-potentiation testing (Table S3). Thus, no evidence for maternal and paternal potentiation was detected in California mice pups.

Comparison of Vocalizations Produced by Habituated vs. Naïve Weanling Pups

There was a difference for total number of syllables between habituated and naïve weanling pups (F(1,7) =17.41, η²_p =0.28, p=0.004). Naïve weanling pups called more throughout the three minute trial sessions than habituated pups (p = 0.04–0.005, Figure 6, Table S4). No other differences in the other vocalization parameters assessed were detected between these two groups.

Figure 6.

Comparison of total number of syllables in naïve vs. habituated weanling California mice pups. Naïve weanling pups produced greater number of syllables than habituated weanling pups (p=0.004).

Discussion

In our current studies, we sought to characterize pup vocalizations while in isolation from the early to late postnatal period (or prior to weaning). The current findings show that pup audible and ultrasonic vocalizations decreased as individuals approached weaning. For instance, at 28 days of age, it took pups approximately 3.5X longer to produce their initial syllable during the trial period than on days 2-4. The mean duration of syllables decreased from PND 2-4 to PND 28. However, the mean syllable frequency peaked on PND 7 and then slowly declined, which suggests that neonatal pups cannot emit higher frequency sounds until about the first week after birth. The total number of syllables, which includes a mixture of audible extending to ultrasonic range and USV vocalizations alone peaked at PND 7 but then decreased by almost five-fold at PND 28.

Average syllable power or loudness of the call peaked at PND 21 (Figure 4). However, the percentage of syllable power above 20 kHz is greatest at PND 7. These collective findings suggest that neonatal pups can vocalize, but the ability to fine-tune the vocalizations in terms of frequency and power is acquired with maturity. These maturational changes may be due to alterations in neuromuscular control to the larynx/vocal tract, greater ability to control vocalizations that are emitted, and/or social learning. Analysis of vocalizations produced in isolation by CBA/CaJ mice over the postnatal period reveals similar findings in that by PND 13, vocal bouts become more complex with greater sequential structure than those produced by younger pups (Grimsley et al., 2011).

In contrast to the above results, the average number of phrases declined from PND 2-4 to PND 28. The total number of syllables within a phrase and average phrase duration also decreased from PND 2-4 to 28 days of age. Although 2-4 day old pups produced an average of 4.3 syllables in each phrase, 28 day old pups only produced an average of 1.4 syllables. The phrase duration decreased by approximately five-fold at PND 28 compared to PND 2-4. These findings are similar to an earlier study with California mice pups (Vieira & Brown, 2002) that showed decreased vocalizations as the pups aged. Although we also sought to confirm potential sex differences in California mice pup vocalizations (Vieira & Brown, 2002; Wright & Brown, 2004), we were unable to do so with the current number of pups tested. The current assessments include all of parameters that can be currently analyzed with Dr. Schenk's program. In future studies, we hope to be able to quantify the ontogenetic changes in different type of USVs, changes in pitch, and amplitude of the calls.

Pups may vocalize more, as evidenced by shorter latency to first syllable and greater number of syllables, average number of phrases, average number of syllables in each phrase, and average syllable duration when they are younger potentially to initiate parental care or to signal distress. However, the current study where pups had to be tested in isolation did not proximately examine whether pup vocalizations directly led to some form of parental care. Our findings did not detect any circadian differences in pup vocalizations, even though California mice are nocturnal (King, 1968), and the afternoon tests were done at the height of the inactive period.

Our initial prediction was that pups would vocalize more after being returned and then separated again from the parents (i.e., a potentiated response) (Hofer et al., 1999; Myers et al., 2004; Shair et al., 2003). However, no differences were observed before and after being placed back with the parents. These current results are in contrast with previous studies that demonstrate in rats pups passive maternal potentiation, where the dam is anesthetized, results in an enhanced vocal response to isolation in the second to third week of life (PND 10-15) (Myers et al., 2004; Shair et al., 2003). Both passive and active maternal potentiation but not paternal potentiation is evident in prairie voles (Robison et al., 2016). It is not clear why we did not observe a potentiated response after California mice pups at various ages were placed back in the home cage with both parents and then reisolated. The findings could suggest species differences exist in this behavioral pattern. Past studies indicate that contrasting species require different conditions for potentiation to occur. In domestic piglets (Sus scrofa domesticus), the reunion must occur in the home nest to elicit a potentiation response (Colonnello, Iacobucci, & Newberry, 2010). Presence of littermates may also be necessary. In C57Bl/6 mice, a reunion with the dam in the home cage failed to induce a potentiated USV in subsequent isolation. However, such a response was observed when the reunion occurred with the dam and littermates (Shair, 2014). A potentiated response might also have been observed if the pups were tested individually with one passive or active parent or been left for a longer time with one or both parents in the home-cage prior to being re-tested. It is also not clear if California pup vocalizations differ with age when they remain in the home cage. As detailed above, two-week old rats emit scant USVs when they remain in the home-cage but become highly vocal when placed in isolation (Hofer & Shair, 1978). Although rat pups in the home-cage with the dam may produce many sonic calls, they elicit few ultrasonic calls (McLinn, 1977).

Pups still produce syllables even at these later ages, which is consistent with prior results (Wright & Brown, 2004). Average syllable power, i.e., loudness, that includes sounds below and above 20 kHz increased with age, peaking at 21 days of age. The ontogeny of pup vocalizations might be finely turned by intrinsic factors. Gonadal hormones, especially testosterone, activate male USVs in polygynous species (Dizinno & Whitney, 1977; Nyby, Dizinno, & Whitney, 1977; Pasch, George, Hamlin, Guillette, & Phelps, 2011; Pomerantz, Fox, & Clemens, 1983; Warburton, Sales, & Milligan, 1989). However, testosterone suppresses USV production to novel females in pair-bonded male California mice (Pultorak et al., 2015). If pup vocalizations are also dependent upon gonadal hormones, this behavior could be vulnerable to developmental or multigenerational exposure to endocrine disrupting chemicals (EDCs). Reduced production of overall pup vocalizations may account for decreased maternal and paternal care provided by California mice perinatally exposed to EDCs, such as bisphenol A (BPA) and ethinyl estradiol (EE) (Johnson et al., 2015). This possibility is currently being explored.

In rodent pups, audible and ultrasonic vocalizations are thought to stimulate parental care by one or both parents (Geyer, 1981; Portfors, 2007; Wright & Brown, 2004). Home bedding and cold exposure are likely primary inducers for emittance of USVs by pups (Kapusta et al., 1995; Portfors, 2007). It could be that in response to isolation, younger pups vocalize more to engender parental care in the form of warmth, protection, and nutritional support. Conversely, older pups may vocalize less because with age, they are better able to protect themselves, obtain, nutritional support, and thermo-regulate. Prior to PND 15, California mice pups cannot regulate their own body temperature and instead rely on both of their parents to provide warmth (Gubernick & Alberts, 1987; Gubernick & Teferi, 2000; Rosenfeld, 2013). In rat dams placed in a Y-maze, USVs produced by three to 11 day old hypothermic pups induce maternal search behavior and act as directional cues that are absent in hypothermic but silent (anesthetized) pups of the same age range (Brunelli, Shair, & Hofer, 1994). Social contact may also be an important cue for isolation-induced vocalizations by younger (three to 14 day old) rat pups with the number of isolation-induced vocalizations reduced when they are placed in the testing chamber with a single anesthetized littermate, even if this littermate is at the same temperature as the chamber (22 °C) (Carden & Hofer, 1990, 1992; Hofer & Shair, 1978). When isolated from the dam and littermates and exposed to differing conditions (odor from the nest, social isolation, low temperature isolation, tactile stimulation, or odor from a conspecific adult male), infant (eight day old) CD-1 mice will vary the sound characteristics of call production based on the specific condition to which they are exposed (Branchi, Santucci, Vitale, & Alleva, 1998). Analysis of USVs and ethnogram of seven day old CD-1 Swiss mice pup behaviors in close temporal proximity to when the USVs are elicited suggests that the pups raise their head immediately prior to the USV, which may increase the likelihood of interacting with their mother (Branchi, Santucci, Puopolo, & Alleva, 2004). Rat pup USVs may be genetically inherited as shown with the recent creation of lines bred for high or low call rates of isolation-induced USVs and corresponding changes in acoustic parameters of the calls (Spence, Aslam, Hofer, Brunelli, & Shair, 2016).

Older pups may also vocalize less when isolated because they are habituated to the isolation procedure and/or are generally less anxious. Comparison of the number of vocalizations produced by naïve weanling pups to habituated older pups reveals that the naïve individuals produced a greater number of syllables than habituated pups. However, no other vocalization parameters, in particular latency to first syllable, mean duration of syllables, average number of phrases, or average syllable duration, differed between these two comparable age groups. The findings suggest a combination of reasons might exist as to why older pups vocalize less when tested in isolation. 1) The pups become habituated to the isolation testing procedure, which may reduce their total number of calls, and 2) they may require less parental care at this time, as evidenced by the fact that the latency to the first call and other vocalization indices were the same in the two groups of weanling animals. Besides being better able to thermo-regulate, older pups possess better psychomotor capabilities and can better sense their surroundings as their eyes open and they acquire enhanced visual and hearing abilities. To more thoroughly test the influence of distress and habituation on pup vocalization parameters, additional studies should examine pups who undergo parental deprivations and/or home cage isolations at the same times when experimental pups are isolated in a novel cage.

Although the current work cannot establish definitive causation in terms of whether pups vocalizations trigger parental care, other studies provide additional evidence in support of this notion. Past studies have shown that in response to pup USVs, rodent mothers immediately investigate the emissions and engaged in pup retrieval (Ehret & Haack, 1981, 1982; Ehret & Haack, 1984). In the current project, we did not assess this behavior as the pups were placed in the recording boxes without their parents. Playback of pup vocalizations alone results in dams approaching the speakers emitting the sounds (Ehret, 2005; Ehret & Haack, 1981). Hormonal and maternal state may also increase motivation to engage in parental care in response to pup vocalizations, as lactating mothers are more likely to respond to pup vocalizations than non-lactating mothers (Ehret & Haack, 1984). In ICR mice, which are not naturally biparental, paternal retrieval of pups can be stimulated when their maternal mates produce USV calls (38 kHz) (Liu et al., 2013). However, similar responses are not observed in C57Bl6J male mice (Liang et al., 2014). The frequency and duration of pup sounds seem to be important features in triggering pup retrieval by mouse mothers (Ehret, 1992; Ehret & Haack, 1981). Not all studies though agree as to whether pup USVs are needed to trigger maternal behavior. Genetically deaf (dn/dn) mouse mothers with normal hearing pups show no change in maternal care, as measured by time spent in nursing posture, in nest/nest building, or grooming pups, relative to normal hearing mothers (D'Amato & Populin, 1987). However, normal hearing pups raised by deaf mothers demonstrate reduced number of USVs compared to normal hearing pups raised by non-deaf mothers. Thus, the hearing status of the mother appears to be important in development of normal pup vocalization responses. Future studies should be designed to determine if playback of pup vocalizations alone is a potent stimulus to induce bipaternal responses.

In conclusion, the current work indicates that as California mice pups mature and approach weaning, they generally vocalize less in isolation, which could be because the pups become more habituated and less anxious about the testing procedure, along with the fact that they require less parental care with age, although additional studies are needed to confirm these possibilities. With maturation, pups demonstrated increased latency to the first syllable, decreased syllable duration, reduced number and duration of phrases, and decreased number of syllables per phrase when tested in isolation. In contrast to other rodent studies, no evidence for a biparental potentiation response was observed in California mice pups with the current study design. It remains to be determined whether biparental care is affected by pup vocalizations. The current findings indicate that several properties of pup vocalizations differ throughout the neonatal period with median syllable frequency, total number of syllables, and usage of ultrasonic calls peaking at PND 7, whereas average syllable power reached maximum at PND 21. It remains to be determined what intrinsic and potentially extrinsic factors govern these maturational changes in pup vocalizations. These studies provide evidence that older California mice pups vocalize less in isolation.

Figure 1.

Examples of California mice vocalizations with age of the animal at the time of the assessment in parentheses. A) Individual bark syllables and a phrase of syllables- burst (PND 7). These syllables can have components in the audible range (below 20 kHz) but can also have harmonics up to and above 50 kHz. B) USVs with individual simple sweep syllables (PND 7). These syllables do not have any components in the audible range. The inset includes a higher magnification of some of the individual sweep syllables.