Maternal behaviours disrupted by Gprasp2 deletion modulate neurodevelopmental trajectory in progeny

Abstract

Autism spectrum disorders (ASDs) are known to present sex-specific differences. At the same time, understanding how maternal behaviours are affected by pathogenic mutations is crucial to translate research efforts since rearing may recursively modulate neurodevelopment phenotype of the progeny. In this work, we focused on the effects of Gprasp2 deletion in females and its impact in progeny care and development. Female mice, wild-type (WT), Gprasp2^+/- (HET) or Gprasp2^-/- (KO) mutants and their progeny were used and behavioural paradigms targeting anxiety, memory, maternal care, and other social behaviours were performed. Analysis of communication was carried out through daily recordings of ultrasonic vocalizations in isolated pups and cross-fostering experiments were performed to understand the effect of maternal genotype in pup development. We found that Gprasp2^-/- females presented striking impairments in social and working memory. Females also showed disruptions in maternal care, as well as physiological and molecular alterations in the reproductive system and hypothalamus, such as the structure of the mammary gland and the expression levels of oxytocin receptor (OxtR) in nulliparous versus primiparous females. We observed alterations in pup communication, particularly a reduced number of calls in Gprasp2 KO pups, which resulted from an interaction effect of the dam and pup genotype. Cross-fostering mutant pups with wild-type dams rescued some of the early defects shown in vocalizations, however, this effect was not bidirectional, as rearing WT pups with Gprasp2^-/- dams was not sufficient to induce significant phenotypical alterations. Our results suggest Gprasp2 mutations perturb social and working memory in a sex-independent manner, but impact female-specific behaviours towards progeny care, female physiology, and gene expression. These changes in mutant dams contribute to a disruption in early stages of progeny development. More generally, our results highlight the need to better understand GxE interactions in the context of ASDs, when female behaviour may present a contributing factor in postnatal neurodevelopmental trajectory.

Keywords: Autism; Cross fostering; GPRASP2; Maternal care; Oxytocin receptor; Sex differences; Ultrasonic vocalisations.

Figure 1

Gpraps2 mutant female mice display increased body weight and deficits in memory and social behaviours. (a) Western blotting of 2-month-old whole brain samples shows decreased expression of GPRASP2 in heterozygous females and protein deletion in Gprasp2^-/- when compared with wild-type littermates, WT n = 7, HET n = 7, KO n = 3, two-tailed Mann–Whitney test. (b) GPRASP2 is highly enriched in the hypothalamus of 1 month-old Gprasp2^+/+ and Gprasp2^+/- females, n = 7, two-tailed Mann–Whitney test. (c) Representative images from 1 year old WT, HET, and KO female mice (scale bar represents 1 cm). (d) Gprasp2^-/- females present an increase in body weight starting approximately at 4-months of age becoming more striking throughout adulthood; WT n = 18, HET n = 23, KO n = 13, two-way repeated measures ANOVA. (e–i) Gprasp2 deficient females present memory and social impairments. (e) Gprasp2 deficient females show no significant alterations in the open field test; WT n = 14, HET n = 13, KO n = 9, two-tailed Mann–Whitney test. (f) In the t-maze for spontaneous alternation, Gprasp2^-/- and Gprasp2^+/- females display impaired alternation when compared with control mice; WT n = 17, HET n = 14, KO n = 12; One sample t test against hypothetical value of chance alternation at 50%. (g) Gprasp2^-/- and Gprasp2^+/- females show no impairments in the sociability component of the three-chamber social test; WT n = 17, HET n = 14, KO n = 12; One sample t test against hypothetical value of chance preference at 0.5. (h) In the social novelty component of the three-chamber test, Gprasp2^-/- females display reduced preference index for a novel social partner when comparing to WT controls; WT n = 17, HET n = 14, KO n = 12; One sample t test against hypothetical value of chance preference at 0.5. (i) Gprasp2^+/- and Gprasp2^-/- females show a slight and significant, respectively, increase in the percentage of marbles unburied in the marble burying test; WT n = 17, HET n = 14, KO n = 12, two-tailed Mann–Whitney test. All data are presented as means ± s.e.m. Statistical significance: *p < 0.05 and **p < 0.01, ***p < 0.001, ****p < 0.0001 (CER-cerebellum; CTX-cortex; STR-striatum; HIPP-hippocampus; HYP-hypothalamus).

Figure 2

Decreased litter survival and deficits in reproductive system in Gprasp2 mutant dams. (a) Schematic diagram of the matings established and possible genotypes of each litter depending on the female progenitor. (b) Females show no significant alterations in the nest building activity; WT n = 17, HET n = 14, KO n = 12, two-tailed Mann–Whitney test. (c) At PND0, no significant differences are seen in terms of the size of the litters from the Gprasp2 deficient females when compared to WT controls, WT n = 12, HET n = 9, KO n = 8, two-tailed Mann–Whitney test. (d) Pups born from Gprasp2^-/- females have a lower chance of survival when compared to pups born from WT controls, WT n = 12, HET n = 9, KO n = 8, two-tailed Mann–Whitney test. (e) Majority of mortality of pups from Gprasp2^-/- females occurs during the two first postnatal days, WT n = 12, HET n = 9, KO n = 8. (f) Weights in the pups nurtured by dams of different genotypes throughout early development, show a small reduction at P5 in animals from KO dams, that is recovered by P15; (+ /y) ^{Gprasp2 +/+dams} n = 38; (+ / +) ^{Gprasp2 +/+dams} n = 42; (+ /y) ^{Gprasp2 +/-dams} n = 21; (-/y)^{Gprasp2 +/-dams} n = 16; (+ / +) ^{Gprasp2 +/-dams} n = 28; (+ /-) ^{Gprasp2 +/-dams} n = 21; (-/y)^{Gprasp2 -/-dams} n = 47; (+ /-) ^{Gprasp2 -/-dams} n = 40, two-tailed Mann–Whitney test. (g) Mammary gland duct complexity is significantly lower in glands from Gprasp2^-/- females (representative images scale bar 5 mm) WT n = 5, HET n = 7, KO n = 9, two-tailed Mann–Whitney test. (h) Oxytocin receptors in the hypothalamus show decreased mRNA expression in both KO and HET nulliparous (NP) and primiparous (PP) females when compared to their WT controls, n = 5/group, two-tailed Mann–Whitney test. (i) No significant alterations are observed between genotypes in terms of mRNA levels of oxytocin in the hypothalamus of these females, being a tendency for increase present in all genotypes when comparing PP to NP females, n = 5/group, two-tailed Mann–Whitney test. (j) Gprasp2 seems to be slightly increased in PP females when compared to NP females, n = 5/group, two-tailed Mann–Whitney test. All data are presented as means ± s.e.m. Statistical significance: *p < 0.05 and **p < 0.01.

Figure 3

Gprasp2 knockout impairs maternal behaviour. (a-d) Gprasp2^-/- females show a significant increase in latency in retrieving the first (a) and third pup (c), being this increased latency still seen although not as strikingly for the second pup (b). Overall, Gprasp2^-/- females show an increased latency to retrieve pups when compared to WT controls (d) data normalised to Gprasp2^+/+ females, trial 1 relative to each pup), WT n = 12, HET n = 11, KO n = 14, two-way repeated measures ANOVA. (e) Schematic diagram of some of the behaviours analysed in the maternal aggression test. (f–h) Gprasp2^-/- females show a tendency for increased aggression with a slight decrease in latency to attack the male intruder (f) with no difference in terms of total number of attacks (g) and slight increase in number of anogenital bites (h), two-tailed Mann–Whitney test WT n = 13, HET n = 13, KO n = 12. All data are presented as means ± s.e.m. Statistical significance: *p < 0.05 and **p < 0.01.

Figure 4

Reduced number and complexity of calls in male pups reared by knockout dams. (a) At PND2, pups from Gprasp2^-/- and Gprasp2^+/- dams show a significant decrease in overall vocalisations, WT n = 4 dams, HET n = 6 dams, KO n = 4 dams, two-tailed Mann–Whitney test. (b) The decrease is present in both males (-/y) and females (+ /-) at this stage when compared with calls from wild-type pups from Gprasp2^+/+ and Gprasp2^+/- females, two-tailed Mann–Whitney test. (c) Examples of the different categories within which the calls were placed and their two broad groups (simple and complex). (d) When looking into the characteristics of these vocalisations, a decrease in terms of the complexity of the vocalisations is observed, at PND2, in pups from Gprasp2^-/- and Gprasp2^+/- dams, especially in the Gprasp2^-/y males. (e) Unravelling the characteristics of these calls even further, a higher predominance of downward calls can be observed at PND2 in the males from Gprasp2^-/- dams when compared to WT controls. (+ /y)^{WT dams} n = 15; (+ / +)^{WT dams} n = 12; (+ /y)^{Gprasp2 +/-dams} n = 13; (-/y)^{Gprasp2 +/-dams} n = 5; (+ / +)^{Gprasp2 +/-dams} n = 14; (+ /-)^{Gprasp2 +/-dams} n = 9; (-/y)^{Gprasp2 -/-dams} n = 12; (+ /-)^{Gprasp2 -/-dams} n = 7, two-tailed Mann–Whitney test. All data are presented as means ± s.e.m. Statistical significance: *p < 0.05 and ****p < 0.0001.

Figure 5

Cross-fostering by wildtype dams rescues neurodevelopmental trajectory in mutant male mice. (a) Schematic representation of the cross-fostering experiment. (b) Gprasp2^-/y pups fostered by Gprasp2^+/+ dams show an increase in body weight at PND15 when compared to wild-type males fostered by Gprasp2^-/- dams, while no differences are observed at PND5, two-tailed Mann–Whitney test. (c) Cross-fostering led to a normalisation of the number of calls at PND2, the pups showing a similar number of calls with both fosters, two-tailed Mann–Whitney test, two-tailed Mann–Whitney test. (d) The cross-fostering effect in the normalisation of the number of calls is not sex-specific with both males and females from both genotypes showing alterations, two-tailed Mann–Whitney test. (e) In terms of call complexity, at PND2, the significant differences previously observed with the normal crossing are also lost, the pups showing similar percentages when cared for by the fosters, two-tailed Mann–Whitney test. (f) As for call characterization, no significant alterations are observed between pups independently of sex and foster genotype (+ /y) n = 8; (+ / +) n = 12; (-/y) n = 8; (+ /-) n = 12, two-tailed Mann–Whitney test. All data are presented as means ± s.e.m. Statistical significance: *p < 0.05.