Sex-specific modulation of early life vocalization and cognition by Fmr1 gene dosage in a mouse model of Fragile X Syndrome

Abstract

Background: Pup-dam ultrasonic vocalizations (USVs) are essential to cognitive and socio-emotional development. In autism and Fragile X Syndrome (FXS), disruptions in pup-dam USV communication hint at a possible connection between abnormal early developmental USV communication and the later emergence of communication and social deficits.

Methods: Here, we gathered USVs from PND 10 FXS pups during a short period of separation from their mothers, encompassing animals of all possible genotypes and both sexes (i.e., Fmr1-/y vs. Fmr1+/y males and Fmr1+/+, +/-, and -/- females). This allowed comparing the influence of sex and gene dosage on pups' communication capabilities. Leveraging DeepSqueak and analyzing vocal patterns, intricate vocal behaviors such as call structure, duration, frequency modulation, and temporal patterns were examined. Furthermore, homing behavior was assessed as a sensitive indicator of early cognitive development and social discrimination. This behavior relies on the use of olfactory and thermal cues to navigate and search for the maternal or nest odor in the surrounding space.

Results: The results show that FMRP-deficient pups of both sexes display an increased inclination to vocalize when separated from their mothers, and this behavior is accompanied by significant sex-specific changes in the main features of their USVs as well as in body weight. Analysis of the vocal repertoire and syntactic usage revealed that Fmr1 gene silencing primarily alters the USVs' qualitative composition in males. Moreover, sex-specific effects of Fmr1 silencing on locomotor activity and homing behavior were observed. FMRP deficiency in females increased activity, reduced nest-reaching time, and extended nest time. In males, it prolonged nest-reaching time and reduced nest time without affecting locomotion.

Conclusions: These findings highlight the interplay between Fmr1 gene dosage and sex in influencing communicative and cognitive skills during infancy.

Keywords: Autism spectrum disorder; Deep learning; Fragile X Syndrome; Homing behavior; Sex; Ultrasonic vocalizations.

Fig. 1

Schematic illustration of the experimental design. In order to generate litters encompassing all potential genotypes, breeding involved crossing either female Fmr1+/- with male Fmr1 +/y or female Fmr1+/- with male Fmr1 -/y. Mice of all genotypes were divided into two cohorts: one underwent USV recording at PND 10, while the other underwent a homing behavior test at PND 13

Fig. 2

FMRP and sex influence body weight. On PND 10 and PND 13, the absence of FMRP resulted in decreased body weight for male mice (-/y) and increased weight for females (-/-) when compared to their respective control groups (+/y and +/+). Among the control groups, male mice (+/y) are typically heavier than female mice (+/+) at PND 10 but not PND 13. Single dots represent individual mice. The box plots present the data ranging from minimum to maximum values, with median and interquartile range (25–75 percentile) shown. The Mann-Whitney U test was applied for statistical analysis. p-values less than 0.05, indicating statistical significance, are marked on the graphs, while complete statistics can be found in Suppl. Table 1. Sample size 10 PND: +/y males N = 22, -/y males N = 21, +/+ females N = 12, +/- females N = 26 and -/- females N = 6. Sample size 13 PND: +/y males N = 10, -/y males N = 14, +/+ females N = 6, +/- females N = 12 and -/- females N = 7

Fig. 3

Sex-specific differences in vocalizations and vocalization latency in FXS Pups. (A) FXS pups of both sexes display a higher number of vocalizations compared to their control counterparts. (B) Only females show a shorter vocalization latency in the absence of FMRP. (A, B) Data are presented as min. to max. box plots with median and 25–75 percentile. Single dots represent individual mice. Mann-Whitney U tests were conducted, and p-values < 0.05 are indicated in the graphs. Full statistical details can be found in Suppl. Table 2. (C, D) Pie graphs illustrate the percentages of vocalizing (V) and non-vocalizing (NV) male (C) and female (D) pups. The percentages were calculated by dividing the number of vocalizers or non-vocalizers by the total number of animals tested in each group. Sample sizes: (A, C, D) +/y males N = 22, -/y males N = 21, +/+ females N = 12, +/- females N = 26, -/- females N = 6. (B) +/y males N = 17, -/y males N = 18, +/+ females N = 10, +/- females N = 20 and -/- females N = 6

Fig. 4

Sex-dependent alteration of core features in USVs of FMRP-deficient mice. (A) FMRP deficiency specifically leads to longer mean length of vocalizations in males. (B, C) Frequency distribution (%) of USV length shows opposite impacts in male (B) and female (C) pups. (D-F) The principal frequency of vocalizations remains similar across the groups. (G) In the absence of FMRP, only females exhibit a statistically more negative mean power in their USVs. (H, I) Frequency distribution analysis reveals that both sexes show a greater utilization of USVs with more negative power in FMRP-deficient pups. (J-L) The mean change in frequency of USVs does not appear to be affected by the FXS genotype in either sex (J), but frequency distribution analysis indicates wider delta use in the absence of FMRP in males (K) but not in females (L). (A, D, G, J) Single dots represent individual mice. Data are presented as min. to max. box plots with median and 25–75 percentile. Mann-Whitney U tests were performed, and p-values < 0.05 are indicated in the graphs. Full statistical details can be found in Suppl. Table 3. (B, C, E, F, H, I, K, L) Data are represented as a Gaussian curve fit (± CI) of the frequency distribution (%). Sample sizes: (A–L) +/y males N = 9, -/y males N = 14, +/+ females N = 7, +/- females N = 13 and -/- females N = 6

Fig. 5

Vocal repertoire of FXS pups. (A) Representative USVs calls classified into ten distinct categories based on a supervised-call classification neural network. (B, C) In the absence of FMRP, male mice exhibit a limited use of short calls in their vocal repertoire. (D, E) The vocal repertoire of female pups remains unaffected by the absence of FMRP. (B, D) Data are represented as a percentage utilization of each category of USVs for each group. (C, E) Data are shown as a bar graph (mean ± SEM) indicating the percentage utilization of each type of USV category for each group. Significance: * p-values < 0.05, full statistical details can be found in Suppl. Table 6. Sample sizes: (B-E) +/y males N = 9, -/y males N = 14, +/+ females N = 7, +/- females N = 13 and -/- females N = 6

Fig. 6

Syntactic transition probability in male FXS pups. (A – D) The probability of transitions ‘from’ and ‘to’ a specific USV class varies among different vocalization classes in +/y (A, B) and -/y (C, D) males. (E, F) Transition probabilities differ ‘from’ a specific USV class (E), but not ‘to’ a specific class (F). (G) Qualitative illustration of transition probability profiles for males of various genotypes. (A-F) Data are shown as a bar graph (mean ± SEM). The p-values < 0.05 are indicated with an asterisk (*) and the full statistics can be found in Suppl. Tables 9 and 12. The hashtag (#) refers to statistics presented in Suppl. Tables 7 and 10. Statistical analysis was done using Mann-Whitney U tests. (G) Arrow diagrams. Arrows indicate transition directions, with brighter colors signifying higher transition probabilities. C = Complex, DR = Downward ramp, IU = Inverted-U, UR = Upward ramp, CT = Complex trill, S = Short, SD = Step Down, F = Flat, SU = Step up, and T = Trill. Sample sizes: (A-G) +/y males N = 9 and -/y males N = 14

Fig. 7

Syntactic transition probability in female FXS pups. (A–F) The probability of transitions ‘from’ and ‘to’ a specific USV class varies among different vocalization classes in +/+ (A, B), +/- (C, D) and -/- females (E, F). (G, H) These profiles are similar among genotypes in the probability of transition “from” (G) and “to” (H) a specific class of USVs. (I) Qualitative illustration of transition probability profiles for females of various genotypes. (A–H) Data are shown as a bar graph (mean ± SEM). (A-F) The hashtag (#) refers to statistics presented in Suppl. Tables 8 and 11. (G, H) Full statistics can be found in Suppl. Tables 9 and 12. (A-H) Statistical analysis was done using Mann-Whitney U tests. (I) Arrow diagrams. Arrows indicate transition directions, with brighter colors signifying higher transition probabilities. C = Complex, DR = Downward ramp, IU = Inverted-U, UR = Upward ramp, CT = Complex trill, S = Short, SD = Step Down, F = Flat, SU = Step up, and T = Trill. Sample size: (A-I) +/+ females N = 7, +/- females N = 13 and -/- females N = 6

Fig. 8

Sex-specific modification of homing behavior in FXS mice. (A-C) FMRP deficiency has no effect on locomotor activity in male mice while in females it leads to an increase in covered distance (A), time moving (B) and average velocity (C). Among the control groups, males (+/y) move more (A) and with greater average velocity (C) than females (+/+). (D-F) homing behavior in FXS mice displayed distinct sex-related effects. Fmr1 silencing in male mice resulted in a delay in entering the nest (D) and less time spent inside (E). On the contrary, FMRP-deficient females (+/- and -/-) reached the nest more quickly (D) and spent more time inside (E) compared to their controls (+/+). In normal FMRP conditions, male mice (+/y) showed a shorter entry delay (D) and more time spent in the nest (E) than females (+/+). The number of entries into the nest did not differ among the experimental groups (F). (A–F) Data are shown as min. to max. box plots with median and 25–75 percentile. Single dots represent individual mice. Statistical analysis was done using Mann-Whitney U tests. p-values < 0.05 are indicated in the graphs. Full statistics can be found in Suppl. Table 13. Sample sizes: +/y males N = 10, -/y males N = 14, +/+ females N = 6, +/- females N = 12 and -/- females N = 7