The acoustic properties, syllable structure, and syllable sequences of ultrasonic vocalizations (USVs) during neonatal opioid withdrawal in FVB/N mouse substrains

Abstract

Concomitant with the opioid epidemic, there has been a rise in pregnant women diagnosed with opioid use disorder and cases of infants born with neonatal opioid withdrawal syndrome (NOWS). NOWS refers to signs and symptoms following cessation of prenatal opioid exposure that comprise neurological, gastrointestinal, and autonomic system dysfunction. A critical indicator of NOWS severity is excessive, high-pitched crying. However, NOWS evaluation is, in large part, subjective, and additional cry features may not be easily recognized during clinical assessment. Thus, there is a need for more objective measures to determine NOWS severity. We used a third trimester-approximate opioid exposure paradigm to model NOWS traits in genetically similar inbred substrains of FVB/N mice (NJ, NCrl, NHsd, and NTac). Pups were injected twice daily from postnatal day 1 (P1) to P14 with morphine (10 mg/kg, s.c.) or saline (20 ml/g, s.c.). Because there were only very minor substrain differences in spontaneous withdrawal-induced ultrasonic vocalization (USV) profiles, we collapsed across substrains to evaluate the effects of morphine withdrawal on additional USV properties. We identified syllable sequences unique to morphine-withdrawn and saline-control FVB/N pups on P7 and P14. We also observed an effect of spontaneous morphine withdrawal on the acoustic properties of USVs and specific syllables on P7 and P14. Multiple withdrawal traits correlated with some acoustic properties of USVs and syllable type emission in morphine-withdrawn FVB/N pups on P7 and P14. These data provide an in-depth investigation of mouse USV syllable profiles and acoustic features during spontaneous neonatal opioid withdrawal in mice.

Keywords: emotional-affective withdrawal; morphine; mouse substrains; neonatal opioid withdrawal syndrome; spectrotemporal profile; ultrasonic vocalizations.

Figure 1.. Morphine exposure from P1 – P14 is sufficient to induce opioid withdrawal traits in FVB/N pups.

Data are plotted as the mean ± SEM. Closed circles = females. Open circles = males. Data were analyzed using linear mixed models with Saline Treatment, Female Sex, and NCrl Substrain as the reference variables and Pup as a random effect (for repeated measures). A. Experimental Timeline. B. Body Weight: The effect of Morphine Treatment on body weight was dependent on Postnatal Day (β = −0.29, SE = 0.012, t(166) = −24.14, ****p < 0.0001), where morphine-withdrawn pups weighed significantly less than saline-control pups from P2 – P14 (P2: **p = 0.0027; P3 – P14: ****p < 0.0001). C. Temperature: There were main effects of Morphine Treatment (β = −1.98, SE = 0.30, t(475) = −6.65, ****p < 0.0001) and Postnatal Day (β = 0.65, SE = 0.041, t(2182) = 15.97, ****p < 0.0001), but no interaction (β = 0.00042, SE = 0.026, t(2181) = 0.016, p = 0.99). SAL, n = 84; MOR, n = 84. D. P7 Hot Plate Latency: There were no interactions (all p ≥ 0.11). Morphine pups displayed decreased hot plate latency compared to saline-control pups (β = −13.03, SE = 0.95, t(177) = −13.70, ****p < 0.0001). SAL, n = 90 (52F, 38M); MOR, n = 89 (43F, 46M). E. P7 Hot Plate Velocity: There were no interactions (all p ≥ 0.91). There was no effect of Morphine Treatment on hot plate velocity (β = −0.00072, SE = 0.0026, t(158) = −0.28, p = 0.78). SAL, n = 83 (48F, 35M) ; MOR, n = 79 (39F, 40M) . F. P7 Tail Withdrawal Latency: There were no interactions (all p ≥ 0.15). Morphine-withdrawn pups displayed reduced tail withdrawal latencies compared to saline-control pups (β = −0.76, SE = 0.34, t(173) = −2.25, *p = 0.026). SAL, n = 77 (44F, 33M); MOR, n = 86 (41F, 45M). G. P14 Hot Plate Latency: There were no interactions (all p ≥ 0.29). Morphine-withdrawn pups displayed reduced hot plate latencies compared to saline-control pups (β = −10.82, SE = 2.14, t(157) = −5.053, ****p < 0.0001). SAL, n = 83 (48F, 35M; MOR, n = 75 (36F, 39M). H. P14 Hot Plate Velocity: There were no interactions (all p ≥ 0.63). Morphine-withdrawn pups displayed increased hot plate velocity compared to saline-control pups (β = 0.0033, SE = 0.0017, t(158) = 1.99, *p = 0.049). SAL, n = 83 (47F, 36M); MOR, n = 77 (37F, 40M). I. P14 Tail Withdrawal Latency: There were no interactions (all p ≥ 0.18). There was no effect of Morphine Treatment on tail withdrawal latency (β = −0.32, SE = 0.21, t(164) = −1.47, p = 0.14). SAL, n = 82 (48F, 34M); MOR, n = 75 (35F, 40M).

Figure 2.. USV locomotor activity during spontaneous morphine withdrawal on P7 and P14 in FVB/N pups.

Data are plotted as the mean ± SEM. Closed circles = females. Open circles = males. Data were analyzed using linear mixed models with Saline Treatment, Female Sex, and NCrl Substrain as the reference variables and Pup as a random effect (for repeated measures). A. P7 USV Distance: The effect of Morphine Treatment on USV distance was dependent on Time (β = −0.020, SE = 0.0026, t(1488) = −7.67, ****p < 0.0001), where morphine-withdrawn pups traveled a greater distance than saline-control pups from 1 – 5 min of the recording session (all **p ≤ 0.0062). B. P7 Total USV Distance: There were no interactions (all p ≥ 0.57). Morphine Treatment was associated with a greater distance traveled during USV recordings associated with spontaneous opioid withdrawal (β = 0.68, SE = 0.14, t(165) = 4.98, ****p < 0.0001). C. P7 Average USV Velocity: There were no interactions (all p ≥ 0.54). There was no effect of Morphine Treatment on average velocity (β = −0.00097, SE = 0.0025, t(167) = −0.39, p = 0.70). SAL, n = 78 (44F, 34M); MOR, n = 82 (43F, 39M). D. P14 USV Distance: There was no Morphine Treatment x Time interaction on USV distance (β = 0.005, SE = 0.014, t(1981) = 0.37, p = 0.71). E. P14 Total USV Distance: There were no interactions (all p ≥ 0.078). Overall, Morphine Treatment was associated with a greater distance traveled during USV recordings associated with spontaneous opioid withdrawal (β = 4.48, SE = 1.69, t(134) = 2.65, **p = 0.0091). F. P14 Average USV Velocity: There were no interactions (all p ≥ 0.63). Morphine Treatment was associated with increased average velocity (β = 0.0083, SE = 0.0025, t(133) = 3.39, ***p = 0.00092). SAL, n = 67 (37F, 30M) MOR, n = 62 (29F, 33M)

Figure 3.. USV emissions during spontaneous morphine withdrawal on P7 in FVB/N pups.

Data are plotted as the mean ± SEM. Closed circles = females. Open circles = males. Data were analyzed using linear mixed models with Saline Treatment, Female Sex, and NCrl Substrain as the reference variables and Pup as a random effect (for repeated measures). The following data are collapsed across Substrain and Sex for simplicity. A. P7 USVs across time: The effect of Morphine Treatment was dependent on Time (β = −5.38, SE = 0.62, t(1573) = −8.63, ****p < 0.001, where Morphine Treatment was associated with an increase in USVs during the first 4 min of the recording session (all **p < 0.0055), and a decrease in USVs during the 6 – 10 min time intervals (all *p < 0.040). B. P7 Total USVs: Overall, there was no effect of Morphine Treatment on the total number of USVs (β = −0.99, SE = 40.64, t(181) = −0.024, p = 0.98). C. P7 Syllable Profile: Morphine Treatment was associated with a decrease in Complex 3 (β = −0.12, SE = 0.025, t(181) = −5.04, ****p < 0.001) and Upward (β = −0.014, SE = 0.006, t(181) = −2.28, *p = 0.024), and an increase in Downward (β = 0.036, SE = 0.0089, t(181) = 4.04, ****p < 0.0001) and Short (β = 0.015, SE = 0.0057, t(181) = 2.71, **p = 0.0074). D. P14 USVs across time: There was a Morphine Treatment x Sex x Time interaction (β = −2.32, SE = 0.53, t(2170) = −4.37, ****p < 0.0001). Both morphine-withdrawn females and males vocalized more over time compared to saline-control pups; however, morphine-withdrawn females vocalized more than morphine-withdrawn males during the 3, 4, and 6 – 15 min intervals (all *p < 0.015). E. P14 Total USVs: There was a Morphine Treatment x Sex interaction (β = −452.5, SE = 153.5, t(151) = −2.95, **p = 0.0037). Both morphine-withdrawn females (β = 883.63, SE = 93.49, t(80) 9.45, ****p < 0.0001) and morphine-withdrawn males (β = 431.1, SE = 124.2, t(71) = 3.47, ***p = 0.00089) vocalized more than saline-control pups. Furthermore, morphine-withdrawn females emitted more USVs than morphine-withdrawn males (β = 471.6, SE = 109, t(151) = 4.32, ****p < 0.0001). F. P14 Syllable Profile: Morphine Treatment was associated with an increase in the proportion of Complex 3 syllables emitted during spontaneous opioid withdrawal (β = 0.14, SE = 0.20, t(155) = 6.89, ****p < 0.0001), and a decrease in Downward (β = −0.015, SE = 0.0077, t(155) = −2.02, *p = 0.045), Flat (β = −0.034, SE = 0.0071, t(155) = −4.73, ****p < 0.0001), Short (β = −0.058, SE = 0.013, t(155) = −4.61, ****p < 0.0001), and Upward (β = −0.056, SE = 0.016, t(155) = −3.48, ***p = 0.00065) syllables. G. Zipf Slope: The Zipf slope of morphine-withdrawn pups decreased from P7 to P14, demonstrating that the syllable repertoire became less random and more repetitious. The Zipf slope of saline-treated pups increased from P7 to P14, depicting increased syllable usage diversity. P7: SAL, n = 88 (50F, 38M); MOR, n = 87 (45F, 42M). P14: SAL, n = 79 (44F, 25M); MOR, n = 76 (38F, 38M).

Figure 4.. Unique USV syllable sequences in FVB/N pups on P7 and P14.

Data are plotted as the mean ± SEM. Closed circles = females. Open circles = males. Data were collapsed across Substrain to increase statistical power and were analyzed using linear mixed models with Saline Treatment and Female Sex as the reference variables. A. P7 Total USV Sequences: There was no effect of Morphine Treatment on the total number of sequences emitted (β = −5.15, SE = 3.96, t(136) = −1.30, p = 0.20). SAL = 74 (40F, 34M); MOR = 65 (37F, 28M). B. P14 Total USV Sequences: There was a Morphine Treatment x Sex Interaction (β = 43.22, SE = 15.60, t(126) = 2.77, **p = 0.00065) where morphine-withdrawn females emitted more unique sequences than males (β = 45.20, SE = 11.31, t(126) = 3.89, ***p = 0.0001). The following data is presented as the percentage of sequences of the top 10 most common sequences observed in morphine-withdrawn and saline-treated pups on C. P7 % Sequences. and D. P14 % Sequences. Complex 2 (C2), Complex 3 (C3), Chevron (Ch), Down (Dn), Upward (Up). SAL, n = 69 (38F, 31M); MOR, n = 61 (29F, 32M).

Figure 5.. Acoustic features of USVs emitted during spontaneous morphine withdrawal in FVB/N pups on P7 and P14.

Data are plotted as the mean ± SEM. For A,B, D, and E: closed circles = females; open circles = males. For C and F, closed circles include both sexes to avoid clutter. Data were analyzed using linear mixed models with Saline Treatment, Female Sex, and NCrl Substrain as the reference variables. The following data are collapsed across Substrain for simplicity. A. P7 Length: There were no interactions (all p ≥ 0.63). Morphine Treatment was associated with shorter USV duration (β = −0.0063, SE = 0.0021, t(144) = −3.054, **p = 0.0027). B. P7 Power: Morphine Treatment was associated with reduced USV power (β = −2.69, SE = 0.76, t(144) = −3.54, ***p = 0.00055). C. P7 Frequencies: Morphine Treatment was associated with higher lowest frequency (β = 3.056, SE = 0.92, t(144) = 3.31, **p = 0.0012), reduced change in frequency (β = −2.94, SE = 0.88, t(144) = −3.34, **p = 0.0011), increased principal frequency (β = 3.36, SE = 1.08, t(144) = 3.12, **p = 0.0022), and increased peak frequency (β = 3.71, SE = 1.03, t(144) = 3.61, ***p = 0.00042). There was no effect of Morphine Treatment on high frequency (β = 0.12, SE = 1.13, t(44) = 0.11, p = 0.92). SAL, n = 73 (41F, 32M); MOR, n = 66 (37F, 29M). D. P14 Length: There were no interactions (all p ≥ 0.091). Morphine Treatment was associated with increased USV duration (β = 0.010, SE = 0.0017, t(145) = 6.18, ****p < 0.0001). E. P14 Power: There were no interactions (all p ≥ 0.36). Morphine Treatment was associated with increased power (β = 2.32, SE = 0.97, t(145) = 2.39, *p = 0.018). F. P14 Frequencies: Morphine Treatment was associated with reduced low frequency (β = −3.38, SE = 0.79, t(145) = −4.27, ****p < 0.0001), a greater change in frequency (β = 5.41, SE = 0.93, t(145) = 4.04, ****p < 0.0001), reduced principal frequency (β = −.89, SE = 0.73, t(145) = −2.59, *p = 0.011), and reduced peak frequency (β = −1.72, SE = 0.77, t(145) = −2.23, *p = 0.027). There was no effect of Morphine Treatment on high frequency (β = 1.43, SE = 0.93, t(145) = 1.54, p = 0.13). SAL, n = 75 (42F, 33M); MOR, n = 67 (33F, 34M).

Figure 6.. Correlation of withdrawal traits in morphine-withdrawn FVB/N pups on P7 and P14.

Colors indicate Pearson’s correlation coefficient: blue = positive, red = negative. Darker colors reflect stronger correlations. *p < 0.01, **p < 0.001, ***p < 0.0001. P7, n = 54 (28F, 26M); P14, n = 44 (21F, 23M).