Novel USV, Fleeting, is characterized by two consecutive acoustic elements with a narrow interval. To analyze the development of the mouse vocal repertoire, maternal separation-induced USVs of C57BL/6 mouse pups were recorded from postnatal day (P) 6 to P18. While the majority of USVs produced corresponded to the waveforms previously described (Supplementary Fig. 1, Supplementary Table 1), a novel pattern of vocalization was identified, which we named Fleeting (Fig. 1a). This novel vocalization was characterized by two acoustic elements, which were separated by a narrow temporal interval of 0.01525 ± 0.0002438 seconds (mean ± SEM) (Fig. 1b). This was unlike any other USV previously described, which consist of continuous sweeps and/or immediate changes in frequency with no temporal interval.
To confirm our finding, we manually measured temporal interval of each identified novel Fleeting USV, and compared it against the intercall intervals (the silence mediating any two-consecutive independent USVs). The temporal interval between the two acoustic elements of the Fleeting was significantly smaller than the time mediating any two-consecutive independent USVs (median temporal interval = 0.01380; median intercall interval = 1.197; Mann-Whitney, p<0.0001) (Fig. 2a). Our finding indicated that the elements of Fleeting are tightly linked, as also suggested by the low variability of their interval. Moreover, the principal frequency of the Fleeting was significantly higher than the principal frequency of the overall mouse vocal repertoire (median principal frequency of Fleeting = 82.02; median principal frequency for overall USVs = 78.94; Mann-Whitney, p=0.0004) (Fig. 2b).
Fleeting USVs shows a negative association with Complex USVs over time. To determine the production of Fleeting over time, its average production across neurodevelopmental time points analyzed was calculated. As the Fleeting can be classified into the Single category of USVs (because it presents no sudden frequency jumps or harmonic elements), relative production of this vocalization was calculated within the total of Single calls produced by each animal. Importantly, the vocalization composed 5.9 ± 0.76% of total Single USVs on P6, remaining stable until P10 (Fig. 3). From this point onwards, the amount of Fleeting emitted steadily decreased, being extinguished by P16, as no Fleeting was identified at P16 or P18 (Fig. 3; Table 1). Then, this progression pattern of Fleeting was compared with the other Single vocalizations – Complex, Upward, Downward, Chevron and Flat15 - identified in the sonograms of the same animals (Supplementary Fig. 2). We found only a complementary progression pattern in Complex vocalizations, which increased over time, suggesting a negative association between Fleeting and Complex vocalizations (Fig. 3).
Table 1
Percentage of novel vocalization Fleeting produced within Single USVs, across the timepoints analyzed. Data are represented by mean ± SEM. Symbols indicate statistically significant differences, *p < 0.05 when compared to Female WT; $$p<0.01 when compared to Male Tsc2+/−, using Mann-Whitney test. Male WT (n=24), Male Tsc2+/− (n=22), Female WT (n=27), Female Tsc2+/− (n=33).
|
Mean ± SEM
|
|
P6
|
P8
|
P10
|
P12
|
P14
|
P16
|
P18
|
Male WT
|
5.650 ± 1.165
|
5.439 ± 0.7658
|
5.800 ± 0.7274
|
4.395 ± 1.145
|
1.969 ± 0.7772
|
0.000 ± 0.000
|
0.000 ± 0.000
|
Female WT
|
6.175 ± 1.029
|
7.583 ± 0.8185
|
6.738 ± 0.7333
|
4.677 ± 0.8725
|
1.992 ± 0.6421
|
0.000 ± 0.000
|
0.000 ± 0.000
|
Male Tsc2+/−
|
5.624 ± 1.087
|
6.491 ± 1.327
|
7.329 ± 1.082
|
3.776 ± 0.8919
|
0.2881 ± 0.1951
|
0.000 ± 0.000
|
0.000 ± 0.000
|
Female Tsc2+/−
|
3.758 ± 0.6146 *
|
5.553 ± 0.6826
|
6.186 ± 0.8251
|
3.043 ± 0.5304
|
2.429 ± 0.5324 $$
|
0.3846 ± 0.3846
|
0.1674 ± 0.1674
|
To further assess that this novel vocalization may be transformed into the Complex one over time, time interval mediating the acoustic elements of Fleeting was plotted over time. Interestingly, we observed a significant time-effect on the time interval of Fleeting (F(4, 976)=59.30; p<0.0001; Fig. 4). Post-hoc comparisons using Tukey’s multiple comparisons test revealed time interval is significantly reduced on P8 (m=0.01603; SEM=0.0004116) compared with P6 (m=0.02004; SEM=0.0004727), and on P10 (m=0.01258; SEM=0.0003699), P12 (m=0.01035; SEM=0.0005644) and P14 (m=0.009695; SEM=0.001067) in relation to P6 and P8 (Fig. 4). Taken together, these data indicate that Fleeting evolves over time, bringing its two acoustic elements closer together, until it is extinguished by adolescence, or as suggested by its temporal pattern of evolution, converted into Complex USVs.
Novel USV is also present in an animal model of ASD. Since the Fleeting vocalization seemed to be specific to early postnatal neurodevelopment, we analyzed the spectrograms of Tsc2+/− mice, a mouse model for ASD, from P6 until P18. The Fleeting was also identified in the spectrograms of these animals, with a temporal interval mediating its two acoustic elements of 0.01525 ± 0.0002276 seconds, which was not significantly different to that of WT animals (median WT time interval = 0.01380; median Tsc2+/− time interval = 0.01390; Mann-Whitney, p=0.8709) (Fig. 5a). Again, temporal interval was significantly smaller than intercall interval (median temporal interval = 0.01390; median intercall interval = 1.950; Mann-Whitney, p<0.0001) (Fig. 5b), which confirmed that the same Fleeting dual element identified in WT animals was also present in the vocal repertoire of Tsc2+/−. Moreover, the principal frequency of Fleeting was also increased when compared with the principal frequency of the overall Tsc2+/− vocal repertoire (median principal frequency of Fleeting = 83.42; median principal frequency overall USVs = 80.92; Mann-Whitney, p=0.0001). However, no significant differences were found regarding principal frequency of Fleeting between WT and Tsc2+/− (Mann-Whitney, p=0.1904) (Supplementary Fig. 3). Analysis of the relative production of Fleeting over time in Tsc2+/− animals showed a pattern of extinction along time, similar to WT (Table 1). Importantly, however, we identified Fleeting being produced at P16 (1.050 ± 0.826% of Single USVs) and P18 (0.1234 ± 0.1234% of Single USVs), unlike WT animals (Fig. 5c). Importantly, we found that relative production of Fleeting and of Complex vocalization over time also suggested a negative association between them (Fig. 5c). The Fleeting in Tsc2+/− animals also showed a time-dependent enclosing of the time interval (F(6, 1144)=41.64; p<0.0001), with significant decrease of the time interval on P8 (0.01583 ± 0.0003377), P10 (0.01264 ± 0.0003884), P12 (0.01076 ± 0.0005391), P14 (0.009292 ± 0.0009332), P16 (0.009167 ± 0.002072) and P18 (0.006550 ± 0.003232) compared with P6 (0.01959 ± 0.0004722); on P10, P12 and P14 compared with P8; and on P10 compared with P14 (Tukey’s multiple comparisons test, p<0.05) (Fig. 5d). Overall, these results seem to indicate that Tsc2+/− animals are not able to transform Fleeting into Complex USV as fast as the WT animals, as Fleeting is still being produced at a later age in the transgenic animals.
Sexual dimorphism is found in relative production of the novel USV. Since male bias is a characteristic of ASD20, we separated groups between males and females and new analyses were conducted minding both genotype and sex. No significant differences were found between WT or Tsc2+/− males and females regarding time interval and principal frequency of Fleeting (Supplementary Fig. 4), showing that Fleeting can be consistently observed even in the Tsc2+/− model. Regarding relative production of Fleeting (Table 1), Tsc2+/− females produced significantly lower proportion of Fleeting in comparison with WT females on P6 (median female WT = 5.138; median female Tsc2+/− = 3.390; Mann-Whitney, p=0.0357), with no differences between males (Fig. 6a). Within the transgenic group, females produced significantly higher proportion of Fleeting than males on P14 (median male Tsc2+/− = 0; median femaleTsc2+/− = 1.899; Mann-Whitney, p=0.0035) (Fig. 6b). Furthermore, female Tsc2+/− were the sole producers of the Fleeting on P16 and P18 (Supplementary Fig. 5). Overall, these results suggest distinctive patterns of vocal repertoire development, in which both genotype and sex seems to play a role.