Significant outcomes
-
Tickling induced 50 kHz USV calls in FSL and SD young rats.
-
Tickling induced a higher 50 kHz USV response in FSL than SD or FRL.
-
USV pattern can be used to infer emotional states in rodents.
-
FSL does not seem to be a suitable model for studying anhedonia.
-
FSL may serve as a model of non-adaptive reward sensitivity.
Limitations
-
Animals were single-housed to increase playing behaviour during tickling sessions.
Introduction
Depression is a severe and life-threatening disease with extensive personal and societal costs (Whiteford et al., Whiteford et al., Reference Whiteford, Degenhardt, Rehm, Baxter, Ferrari, Erskine, Charlson, Norman, Flaxman, Johns, Burstein, Murray and Vos2013). Despite the recent advancements (Krystal et al., Reference Krystal, Kavalali and Monteggia2024), current pharmacological treatments are unsatisfactory as most present a delayed onset, limited efficacy and poor long-term symptom control (Dwyer & Duman, Reference Dwyer and Duman2013, Gigliucci et al., Reference Gigliucci, O’Dowd, Casey, Egan, Gibney and Harkin2013). Thus, there is an urgent need to refine the preclinical methods used to discover putative antidepressants (Cryan & Lucki, Reference Cryan and Lucki2002, Neumann et al., Reference Neumann, Wegener, Homberg, Cohen, Slattery, Zohar, Olovoer and Mathé2011, Gururajan et al., Reference Gururajan, Reif, Cryan and Slattery2019). The forced swim test is a predictive tool to assess antidepressant-like behaviour (Porsolt et al., Reference Porsolt, Bertin and Jalfre1977a; Porsolt et al., Reference Porsolt, Le Pichon and Jalfre1977b; Porsolt et al., Reference Porsolt, Anton, Blavet and Jalfre1978, Cryan & Lucki, Reference Cryan and Lucki2002, Cryan & Slattery, Reference Cryan and Slattery2007, Slattery et al., Slattery & Cryan, Reference Slattery and Cryan2012). Despite being highly popular due to its feasibility and reproducibility, its use, interpretation and conceptualisation are criticised.
Rats emit ultrasonic vocalisations (USVs), inaudible to the human ear. These reflect the affective state of the rats (Knutson et al., Reference Knutson, Burgdorf and Panksepp1999). For instance, when rats have their tails pinched or are exposed to other aversive stimuli, they elicit 22 kHz vocalisations associated with negative affect (Panksepp, Reference Panksepp1999, Knutson et al., Reference Knutson, Burgdorf and Panksepp2002). Similarly, maternal separation induces calls of 40 or 60 kHz (Boulanger-Bertolus et al., Reference Boulanger-Bertolus, Rincón-Cortés, Sullivan and Mouly2017; Kaidbey et al., Reference Kaidbey, Ranger, Myers, Anwar, Ludwig, Schulz, Barone, Kolacz and Welch2019). Alternatively, administration of amphetamine either systemically or directly in the nucleus accumbens, a vital structure of the reward system, was linked to 50 kHz vocalisations (Burgdorf et al., Reference Burgdorf, Knutson, Panksepp and Ikemoto2001; Burgdorf & Panksepp, Reference Burgdorf and Panksepp2001). These vocalisations are also spontaneously elicited when juvenile rats play with each other, suggesting its association with positive affect (Panksepp, Reference Panksepp1999, Knutson et al., Reference Knutson, Burgdorf and Panksepp2002). The experimenter can imitate the juvenile rats’ rough-and-tumble play through a tickling-like stimulation, increasing 50 kHz vocalisations. The number of calls elicited can be used to infer depressive and anxiety-like behaviour (Burgdorf et al., Reference Burgdorf, Knutson, Panksepp and Ikemoto2001; Mallo et al., Reference Mallo, Matrov, Herm, Koiv, Eller, Rinken and Harro2007).
The Flinders Sensitive Line (FSL) is an inbred line widely validated as a model of depression (Overstreet, Reference Overstreet1992, Overstreet et al., Reference Overstreet and Wegener2013). Studies on face validity demonstrate: (I) reduced sucrose intake following stress and increased immobility in the forced swim test, suggesting a depressive-like phenotype (Wegener G., Reference Wegener, Mathé and Neumann2012); (II) elevated REM sleep and impairments in the novel object recognition task, modelling disturbed sleep and cognitive deficits related to depression (Wegener et al., Reference Wegener, Mathé and Neumann2012). Studies on construct validity in this model show serotonergic, cholinergic, neurotrophic and morphological abnormalities, which have been implicated in depression in humans (Overstreet & Wegener, Reference Overstreet and Wegener2013; Strenn et al., Reference Strenn, Suchankova, Nilsson, Fischer, Wegener, Mathé and Ekman2015; Ardalan et al., Reference Ardalan, Rafati, Nyengaard and Wegener2017; Kirkedal et al., Reference Kirkedal, Elfving, Müller, Moreira, Bindila, Lutz, Wegener and Liebenberg2019; Tillmann & Wegener, Reference Tillmann and Wegener2019, Treccani et al., Reference Treccani, Ardalan, Chen, Musazzi, Popoli, Wegener, Nyengaard and Müller2019; Abildgaard et al., Reference Abildgaard, Kern, Pedersen, Hansen, Lund and Wegener2021; Arjmand et al., Reference Arjmand, Pedersen, Silva, Landau, Joca and Wegener2023; Domingos et al., Reference Domingos, Müller, Da Silva, Filiou, Nielsen, Guimarães, Wegener and Joca2024). Emphasising its predictive validity, the FSL rat shows decreased immobility in the forced swim test when treated chronically and acutely with drugs that have antidepressant effects in humans (du Jardin et al., Reference Du Jardin, Liebenberg, Müller, Elfving, Sanchez and Wegener2016; Marchetti et al., Reference Marchetti, Lauria, Caberlotto, Musazzi, Popoli, Mathé, Domenici and Carboni2020). Therefore, the present study examined the emission of USVs during tickling stimulation in the FSL rat compared to the Flinders Resistant Line (FRL) and outbred Sprague Dawley (SD) rats – both commonly used controls for the FSL rat.
The aims of the study were to i) characterise and ii) investigate the applicability of USVs as a screening method for depressive-like behaviour in a rodent model of depression. It was hypothesised that the depressive-like phenotype of the FSL rat would be reflected in the ultrasonic response to tickling stimulation.
Methods
Animals
Juvenile 22-day-old male FSL (n = 27), FRL (n = 24) and SD (n = 23) rats were obtained from the in-house breeding colony at the Translational Neuropsychiatry Unit, Aarhus University (Denmark). Animals were housed at 22 ± 2oC and kept at a 12-hour light/dark cycle (lights on at 07:00) with access to food and water ad libitum. The rats had access to a hide, nesting material and a wooden stick throughout the experiment. All animal procedures were approved by the Danish National Committee for Ethics in Animal Experimentation (permission ID: 2012-15-2934-00254). Baseline characteristics of the animals are given in Table 1.
Table 1. Group characteristics. Group distributions of age and weight at the beginning of the experiment. Values are expressed as means ± SD. FSL = Flinders Sensitive Line rat, FRL = Flinders Resistant Line rat, SD = Sprague Dawley
Experimental design
On postnatal day 21, the rats were weaned from their mothers. All rats were single-housed during the entire experimental period to increase play behaviour during tickling sessions (Panksepp, Reference Panksepp1981; Panksepp et al., Reference Panksepp, Siviy and Normansell1984). The rats were randomly assigned to the experimental groups (tickling/light-touch). All experimental procedures were carried out in the rat’s light cycle.
Tickling
The rats were tickled daily for 6 weeks starting from postnatal day 21. A tickling session consisted of 15 s acclimatisation to the tickling cage followed by 15 s hand-play with the experimenter, touching the nape of the neck and abdomen, then 15 s of no stimulation followed by 15 s tickling. During the last three seconds of each bout of tickling, attempts were made to pin down the rat and vigorously tickle the abdomen. This method has been described elsewhere (Knutson et al., Reference Knutson, Burgdorf and Panksepp1998; Mallo et al., Reference Mallo, Matrov, Herm, Koiv, Eller, Rinken and Harro2007). A dorsal light-touch group was used as a control group for the tickled rats. As previously described, light touch has been used as a control for tickling, as it is a discernible stimulation but presumably with less reward value (Burgdorf et al., Reference Burgdorf, Knutson, Panksepp and Ikemoto2001; Burgdorf & Panksepp, Reference Burgdorf and Panksepp2001; Yamamuro et al., Reference Yamamuro, Hori, Nakagawa, Hayashi, Sakamoto, Ohnishi, Takeuchi, Mihara, Shiga, Murakami and Urayama2013). Acoustic foam was used to isolate the Plexiglas container where the tickling took place (Brudzynski & Pniak, Reference Brudzynski and Pniak2002). The same experimenter conducted all the tickling sessions and was blinded to the strain of the rats.
Acoustic data acquisition, analysis and classification
The tickling sessions were recorded using Avisoft-RECORDER USGH (v.4.2 Avisoft Bioacoustics, Berlin, Germany). A condenser microphone CM16/CMPA (from AvisoftBioacoustics) was secured 20 cm from the cage floor, and its signal was fed to the Avisoft UltraSoundGate 416H with a sampling rate of 750 kHz. Acoustical analysis was performed with the Avisoft SASlab Pro (v.5.2 Avisoft bioacoustics, Berlin, Germany), and spectograms were generated with a fast Fourier transformation (FFT)-length of 256 points and an overlap of 75% (FlatTop window, 100% frame size). A semi-automatic recording of call parameters was used for the quantitative analysis of the 2-minute tickling or light-touch sessions. In accordance with Reno et al. (Reference Reno, Marker, Cormack, Schallert and Duvauchelle2013), calls with a frequency between 30–90 kHz were considered 50 kHz calls and between 20 and 30 kHz were considered 22 kHz calls (Reno et al., Reference Reno, Marker, Cormack, Schallert and Duvauchelle2013). Furthermore, calls were regarded as individual vocalisations when separated by at least 20 ms. The minimum call length was set at 5 ms and sounds shorter than this were considered noise. For the 50 kHz calls, the differences in morphology and call subtypes were based on the classification by Wright et al. (Wright et al., Reference Wright and Clarke2010). This classification consists of 14 call categories differently modulate by diverse stimuli, environmental conditions and inter-individual differences, ultimately allowing a more refined analysis of the affective state (Wright et al., Reference Wright and Clarke2010). The qualitative evaluation of the calls was carried out on the first 15 s stimulations of experimental day 23, as this was the day with the largest quantity of calls. The categorisation was made manually based on the predefined frequency patterns described by Wright et al.
Data analysis and statistics
All statistics were conducted using SPSS version 22 and GraphPad Prism 10 for Windows (GraphPad Software, San Diego, CA). The statistical analysis of differences between periods of tickling versus periods without tickling and among ticking sessions was done with repeated-measures one-way analysis of variance (ANOVA) with planned post hoc pairwise comparisons using Geisser-Greenhouse correction for sphericity and Bonferroni for multiple comparisons. The statistical analysis of differences between the tickling and the light-touch groups or strain differences for call categories was done using a two-way ANOVA with post hoc Bonferroni for multiple comparisons. Spearman’s correlation coefficient was used to determine the inter-individual stability of calls between days. The qualitative differences in calls between strains were analysed by a one-way ANOVA, followed by a Tukey post hoc test. A two-tailed p-value of less than 5% was considered significant for all tests.
Results
Ultrasonic vocalisations
More than 71.000 calls were counted, and more than 2.100 were classified according to Wright et al., categories (Wright et al., Reference Wright and Clarke2010). As 22 kHz calls accounted for less than 2.9 % of calls on day 1 and 0.6 % on day 6, these were not further analysed. Also, 40 of the 74 rats produced calls which could not be classified as 50 kHz as their frequency was below 30 kHz. They did not have the appearance of 22 kHz either, as they were shorter (<300 ms) or presented frequency modulations. This type of call may be reminiscent of infants’ vocalisations. They were primarily present in the first days of recording, after which they decreased. On experimental day 13, only three rats produced these calls, and they were not further analysed. Finally, in the qualitative evaluation of the call profiles, 26% of the calls could not be classified and went into a miscellaneous group.
Tickling increases the production of 50 kHz calls
The number of calls was significantly different among strains and stimulation type on both days, 1 [Strain: F (2, 66) = 7.358, p = 0.0013; Stimulation: F (1, 66) = 47.75; p < 0.0001; Interaction: F (2, 66) = 5.334, p = 0.0071] and 23 [Strain: F (2, 66) = 11.07, p < 0.0001; Stimulation: F (1, 66) = 132.5, p < 0.0001; Interaction: F (2, 66) = 8.123, p = 0.0007]. The number of calls was higher in tickled animals. This difference was not statistically significant for the FRL rats on day 1 (Fig. 1). Moreover, while no differences among strains were noticed in the light-touch group, tickled FSL emitted more calls than FRL and SD on both days.
Figure 1. Differences in the no. of calls emitted by animals submitted to light-touch or tickle. (A) Number of calls on day 1 by strain. (B) Number of calls on day 23 by strain. Data presented as mean ± SEM. *p < 0.05 compared to light-touched of the same strain. #p < 0.05 compared to SD and FRL.
Furthermore, in the tickled group (Fig. 2), there was a significant difference in the number of calls emitted between periods of stimulation and no-stimulation [Strain: F (2, 45) = 0.8199, p = 0.4427; Time Interval: F (2.707, 121.8) = 17.03, p < 0.0001; Interaction: F (14, 315) = 0.6060, p = 0.8599].
Figure 2. Distribution of 50 kHz calls between strains during tickling sessions on days 1 by interval of 15s. No stimulation intervals: 1, 3, 5, 7. Stimulation intervals: 2, 4, 6, 8. Data presented as mean ± SEM. *p < 0.05 compared to the previous interval.
The FSL rat produces more calls than their controls during tickling
Quantitative analysis of calls during tickling revealed a significant effect of strain, day and interaction on the number of calls produced [Strain: F (1.991, 179.2) = 127.9, p < 0.0001; Day: F (5, 90) = 13.17, p < 0.0001; Interaction: F (10, 180) = 3.014, p = 0.0015]. Post hoc analysis showed that FSL emitted more calls than FRL rats. Moreover, apart from day 1, FSL also produced more calls than SD, and on days 13 and 23, FRL emitted more calls than SD (Fig. 3).
Figure 3. Distribution of calls across experimental days divided by strain. Data presented as mean ± SEM. *p < 0.05 FSL compared to FRL and SD, #p < 0.05 FRL compared to SD.
Furthermore, a correlation study was performed to evaluate whether the quantity of calls was temporally stable. FSL rats did not show a correlation between days; however, SD (from day 6) and FRL (from day 13) did (see Table 2).
Table 2. Stability in call profiles across days in the three strains. Correlation of calls between days across strains. FSL = Flinders Sensitive Line rat, FRL = Flinders Resistant Line rat, SD = Sprague Dawley
Qualitative call profiles among strains
Lastly, 57 and 2074 calls were classified in the light-touch and tickled groups, respectively. The qualitative analysis is graphically presented in Fig. 4. The percentage of calls of each type can be found in the S1 table.
Figure 4. Qualitative analysis of calls in animals tickled or light-touch. (A) Distribution of calls in animals submitted to light-touch. (B) Distribution of calls in tickled animals. The three strains were merged in the two groups (tickled and light-touch).
The qualitative analysis of the calls for each strain in the tickled group is graphically presented in Fig. 5A-C. The percentage of each call type can be found in Table S2. No differences in the relative number of calls were found among strains for trill [Figure 5D: F (2, 45) = 0.3944, p = 0.6764], flat [Figure 5E: F (2, 45) = 1.308, p = 0.2804], multistep [Figure 5G: F (2, 45) = 0.2524, p = 0.7780] and upward ramp calls [Figure 5H: F (2, 45) = 0.9351, p = 0.4000]. However, significant strain differences were observed in flat-trill [Figure 5F: F (2, 45) = 6.162, p = 0.0043] and complex calls [Figure 5I: F (2, 45) = 5.164, p = 0.0096].
Figure 5. Qualitative analysis of calls in tickled animals by strain. (A) Calls distribution across categories in SD. (B) Calls distribution across categories in FRL. (C) Calls distribution across categories in FSL. (D) Relative quantity of trill calls in SD, FRL and FSL. (E) Relative quantity of flat calls in SD, FRL and FSL. (F) Relative quantity of trill-flat calls in SD, FRL and FSL. (G) Relative quantity of multistep calls in SD, FRL and FSL. (H) Relative quantity of upward ramp calls in SD, FRL and FSL. (I) Relative quantity of complex calls in SD, FRL and FSL. D-I Data presented as mean ± SEM. *p < 0.05 as indicated.
Discussion
This is the first time ultrasonic calls emitted by FSL during tickling have been investigated. Our results showed that I) tickling is a suitable procedure to induce 50 kHz vocalisation in rats and II) the ultrasonic response of tickled FSL rats significantly differs from FRL and SD rats.
This study demonstrated a significant difference in the number of calls produced by the tickled rats compared to the light-touch group, which is in line with previous reports (Panksepp, Reference Panksepp1999; Panksepp & Burgdorf, Reference Panksepp and Burgdorf2000; Burgdorf et al., Reference Burgdorf, Knutson, Panksepp and Ikemoto2001; Hori et al., Reference Hori, Shimoju, Tokunaga, Ohkubo, Miyabe, Ohnishi, Murakami and Kurosawa2013). Furthermore, when evaluating the subtypes of calls emitted, the tickled groups produced trill and flat-trill combinations, whereas the light-touch groups did not produce any flat-trill combinations and only very few trill calls. These call types were previously related to tickling (Schwarting et al., Reference Schwarting, Jegan and Wöhr2007) and other rewarding stimuli like psychostimulants (Ahrens et al., Reference Ahrens, Ma, Maier, Duvauchelle and Schallert2009; Simola et al., Reference Simola, Ma and Schallert2010, Wright et al., Reference Wright and Clarke2010). Furthermore, previous reports have shown that pair-tested rats presented a higher proportion of trill calls compared to single-tested rats after saline or amphetamine treatment (Wright et al., Reference Wright and Clarke2010). Despite the rats in our study being single-housed, we observed increased calls after tickling.
Furthermore, tickled FSL rats produced significantly more calls than FRL and SD controls. This contrasts with the pattern observed in other animal models of depression, which showed fewer USV calls (Mallo et al., Reference Mallo, Matrov, Koiv and Harro2009, Rao & Sadananda, Reference Rao and Sadananda2015, Burke et al., Reference Burke, Modlinska, Mauro, Aleksandrova, Pellis, Phillips and Euston2021). The increase in 50 kHz calls observed in juvenile FSL suggests higher sensitivity to rewarding stimulus, which contradicts the core anhedonic state found in depression. However, animal models of depression often resemble some but not all depression features (Gururajan et al., Reference Gururajan, Reif, Cryan and Slattery2019). For instance, whether FSL can emulate anhedonia is a matter of controversy. Previous reports showed mixed findings in adult or prepubertal FSL (Pucilowski et al., Reference Pucilowski, Rezvani and Janowsky1993; Matthews et al., Reference Matthews, Baldo, Markou, Lown, Overstreet and Koob1996; Malkesman et al., Reference Malkesman, Braw, Zagoory-Sharon, Golan, Lavi-Avnon, Schroeder, Overstreet, Yadid and Weller2005; Sanchez et al., Reference Sanchez, El Khoury, Hassan, Wegener and Mathé2018). However, when exposed to specific environmental stressors, anhedonia was observed in FSL, perhaps revealing a heightened sensitivity to external stimuli (Pucilowski et al., Reference Pucilowski, Rezvani and Janowsky1993). Taken together, FSL seems more sensitive to reward and aversive interventions. In agreement with the enhanced sensitivity of FSL, we found an increase in flat-trill combinations: previous data showed that flat-trill combinations induced by reward stimulus (playing, amphetamines) were only increased in pair-tested animals and not in single animals. At the same time, FSL enhanced this call type compared to the other strains (Wright et al., Reference Wright and Clarke2010). Curiously, in humans, high environmental sensitivity during childhood may constitute a risk factor for developing depression as an adult (Lionetti et al., Reference Lionetti, Klein, Pastore, Aron, Aron and Pluess2022). Furthermore, the vulnerability aspects of sensitivity to reward are associated with high expression of traits such as FEAR, SADNESS and ANGER (Pulver et al., Reference Pulver, Kiive and Harro2020) of the Affective Neuroscience Personality Model (Davis & Panksepp, Reference Davis and Panksepp2011). These emotional traits serve as depression vulnerability mediators in adverse environments. Consistently, it has been demonstrated in rats that higher reward sensitivity corresponds to higher sensitivity to chronic mild stress (Koiv et al., Reference Koiv, Vares, Kroon, Metelitsa, Tiitsaar, Laugus, Jaako and Harro2019). Single housing may additionally have been the stressor to precipitate 50 kHz USV response from the more reward-sensitive FSL.
Lastly, we performed a correlational analysis to determine whether the quantity of calls produced by a rat on one day was associated with the number of calls produced on later experimental days. Significant temporal stability was found for the SD rat from day 6 and the FRL rat from day 13. However, the number of calls emitted by FSL was not associated with the number of calls emitted on later days. A previous study on the temporal stability of the USVs produced during tickling stimulation in Wistar rats showed the calls from the beginning of the second week of stimulation to be associated with calls produced on later experimental days (Mallo et al., Reference Mallo, Matrov, Herm, Koiv, Eller, Rinken and Harro2007). The apparent instability in the number of calls produced between days in the FSL rat may thus be a characteristic of this specific phenotype and may relate to their reward sensitivity as a representation of depression vulnerability. One could speculate that this reflects a heightened sensitivity to the tickling stimulation. Hence, minor variations in the stimulation between days may lead to more significant differences in the number of calls produced in the FSL rats compared to the SD and FRL rats.
Thus, instead of a depressive-like phenotype, the FSL rat may present a hypersensitive phenotype, which displays a heightened sensitivity to its environment regardless of whether the stimuli are appetitive or aversive. Thus, our findings have important implications for future studies of the FSL rat within the field of depression. If the depressive-like phenotype is to be modelled, it may be advantageous to expose the FSL rat to an environmental stressor. Moreover, emotional hyperreactivity was related to bipolar disorder in patients (M’bailara et al., Reference M’bailara, Demotes-Mainard, Swendsen, Mathieu, Leboyer and Henry2009, Henry et al., Henry et al., Reference Henry, Phillips, Leibenluft, M’bailara, Houenou and Leboyer2012). Considering the neurochemical profile and the behavioural phenotype of FSL, some authors have already suggested its potential as a preclinical tool for bipolar depression (Mncube et al., Reference Mncube, Möller and Harvey2021). The hypersensitive trait may endorse this hypothesis. Future research should address the performance of FSL in bipolar-like tests such as amphetamine-induced hyperlocomotion, social interaction or their response to mood stabilisers. Moreover, combining environmental stress and tickling would be interesting to explore the USV profile of stressed FSL compared to controls.
Conclusions
The current study aimed to investigate whether USVs may be utilised as an alternative non-stressful tool to assess depressive-like behaviour in rats. The main finding of this study – a profound difference in vocalisations between an animal model of depression and controls during tickling sessions – suggests that this may be a fruitful endeavour. However, more studies on USVs in other rat models of depression are needed as the FSL rat may not be the preferred rat model to use, as it apparently shows heightened reward sensitivity and hence might not model anhedonia, at least at an early age. Instead, USVs in the FSL rat may provide a readout in a model of general vulnerability.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/neu.2024.61.
Acknowledgements
Author contribution
Linda Marie Kai conducted all animal experiments, collected and analysed the data and made the initial draft for the article. Lia Parada Iglesias participated in the analysis and interpretation of data and the writing of the manuscript. Gregers Wegener, Kadri Kõiv and Jaanus Harro contributed to the study’s conception and design, the interpretation of the data and critically revised the intellectual content of the manuscript.
Financial support
This work was supported by Region Midtjyllands Psykiatrisk Forskningsfond. Lia Parada Iglesias is supported by Horizon Europe (MSCA, PoMAM – grant 101110693). Jaanus Harro is supported by the Estonian Research Council (PRG1213).
Competing interests
Gregers Wegener was the Editor-in-Chief of Acta Neuropsychiatrica at the time of submission but was not involved and actively withdrew during this manuscript’s review and decision process.
Ethical statement
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional guides on the care and use of laboratory animals.
Open access
