Motor Impairments Correlate with Social Deficits and Restricted Neuronal Loss in an Environmental Model of Autism

Motor Impairments Correlate with Social Deficits and Restricted Neuronal Loss in an Environmental... Downloaded from https://academic.oup.com/ijnp/article-abstract/21/9/871/4995459 by Ed 'DeepDyve' Gillespie user on 04 September 2018 International Journal of Neuropsychopharmacology (2018) 21(9): 871–882 doi:10.1093/ijnp/pyy043 Advance Access Publication: May 12, 2018 Regular Research Article regular research article Motor Impairments Correlate with Social Deficits and Restricted Neuronal Loss in an Environmental Model of Autism Tareq Al Sagheer, Obelia Haida, Anais Balbous, Maureen Francheteau, Emmanuel Matas, Pierre-Olivier Fernagut, Mohamed Jaber INSERM U-1084, Experimental and Clinical Neurosciences Laboratory, University of Poitiers, Poitiers, France (Dr Sagheer, Ms Haida, Dr Balbous, Ms Francheteau, Dr Matas, Dr Fernagut, and Dr Jaber); CHU Poitiers, Poitiers, France (Drs Balbous and Jaber); Université de Bordeaux, Institut des Maladies Neurodégénératives, UMR 5293, Bordeaux, France (Dr Fernagut); CNRS, Institut des Maladies Neurodégénératives, UMR 5293, Bordeaux, France (Dr Fernagut). T.A.S. and O.H. contributed equally to this work. Correspondence: Mohamed Jaber, PhD, Laboratoire de Neurosciences Expérimentales et Cliniques-LNEC, INSERM U-1084, Université de Poitiers, Bâtiment B36, 1 rue Georges Bonnet, BP 633, TSA 51106, 86073 POITIERS cedex9 – France (mohamed.jaber@univ-poitiers.fr). ABSTRACT Background: Motor impairments are amongst the earliest and most consistent signs of autism spectrum disorders but are not used as diagnostic criteria. In addition, the relationship between motor and cognitive impairments and their respective neural substrates remain unknown. Methods: Here, we aimed at determining whether a well-acknowledged animal model of autism spectrum disorders, the valproic acid model, displays motor impairments and whether they may correlate with social deficits and neuronal loss within motor brain areas. For this, pregnant female mice (C57BL/6J) received valproic acid (450  mg/kg) at embryonic day 12.5 and offspring underwent a battery of behavioral analyses before being killed for histological correlates in motor cortex, nigrostriatal pathway, and cerebellum. Results: We show that while valproic acid male mice show both social and motor impairments, female mice only show motor impairments. Prenatal valproic acid exposure induces specific cell loss within the motor cortex and cerebellum and that is of higher magnitude in males than in females. Finally, we demonstrate that motor dysfunction correlates with reduced social behavior and that motor and social deficits both correlate with a loss of Purkinje cells within the Crus I cerebellar area. Conclusions: Our results suggest that motor dysfunction could contribute to social and communication deficits in autism spectrum disorders and that motor and social deficits may share common neuronal substrates in the cerebellum. A systematic assessment of motor function in autism spectrum disorders may potentially help the quantitative diagnosis of autism spectrum disorders and strategies aimed at improving motor behavior may provide a global therapeutic benefit. Keywords: valproic acid, cerebellum, motor cortex, gait, Purkinje cells Received: February 9, 2018; Revised: April 4, 2018; Accepted: May 9, 2018 © The Author(s) 2018. Published by Oxford University Press on behalf of CINP. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http:// creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, 871 provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com Downloaded from https://academic.oup.com/ijnp/article-abstract/21/9/871/4995459 by Ed 'DeepDyve' Gillespie user on 04 September 2018 872 | International Journal of Neuropsychopharmacology, 2018 Significance Statement The paper’s main message is that motor impairments in ASD may be indicative of this pathology and may even underlie some of its core cognitive and behavioral features. For this, we conducted a detailed analysis of motor and social behavior in both sexes of a VPA mouse model and correlated these findings to cell loss in specific brain areas such as the cerebellum and motor cortex. Our work brings evidence that motor behavior can be used as a quantitative approach for the diagnosis of ASD and is indicative of social deficits. This opens new avenues in the diagnosis and treatment of ASD targeting the affected brain areas. and postmortem outcomes to identify their potential links and Introduction offer a new perspective on the disease by demonstrating that Autism is one of the most common and disabling neurological motor disorders are linked with social impairments and may disorders of childhood with no commonly implemented diag- share common neural substrates. nostic test or cure available. It affects 3 times more males than females with a prevalence that has been constantly increasing these last decades to reach a proportion of 1 in 68 individuals Materials and Methods (CDC, 2014; Loomes et al., 2017). In the current DSM-5 classifica- tion, 2 global symptoms are recognized as hallmarks of autism Animals spectrum disorder (ASD): disorders of social communication and restricted and repetitive behaviors (APA, 2013). However, Animal housing and experimental procedures were per - in their initial reports of autistic patients, Kanner and Asperger formed in accordance with the European Union directive described early motor deficits, including clumsy gait in what (2010/63/EU) and validated by the regional ethical committee is now recognized as ASD (Kanner, 1943; Asperger, 1944). As (approval no.  2015020415093780). C57BL/6J Mice (Charles River such, motor impairments in ASD have been receiving increased Laboratories) were housed in ventilated cages with access to attention recently, and study of their contribution to the patho- food and water ad libitum. Room temperature was maintained physiology of ASD is currently a growing field of research. For at 23°C on a 12-h-light/-dark cycle. instance, there are a number of qualitative and quantitative reports in ASD describing impairments in visio-motor and man- Experimental Design and Statistical Analysis ual dexterity tasks, limb coordination during tasks requiring balance, agility and speed as well as in gait and ataxia (Fatemi Twenty females and 10 males were used for mating. Four females et al., 2012). Furthermore, motor impairments may be amongst were placed with a single male and left overnight. Pregnant mice the earliest signs of some forms of ASD (Ozonoff et  al., 2008). received a single i.p. injection of either VPA (450 mg/kg) or NaCl Accordingly, the assessment of motor disorders might help the 0.9% at gestational day 12,5 (E12.5) (Nicolini and Fahnestock, 2018) early and quantitative diagnosis of the pathology and the identi- when the neuronal tube is closing in rodents, followed by neuro- fication of dysfunctional brain regions and circuits in ASD. Such genesis (Morriss-Kay et al., 1993). Following mating, pregnant mice approach may also pave the way for novel therapeutic interven- were left undisturbed until they gave birth. At weaning (P21), pups tions that might correct motor impairments and potentially the from different litters were allocated to the 4 experimental groups cognitive and social deficits characteristic of ASD. depending on prenatal treatment: VPA males (n = 34), saline With this in perspective, the cerebellum has attracted a males (n= 27), VPA females (n= 25), and saline females (n= 30). The renewed interest in ASD as a potential brain area at the cross- experiment timeline is presented in Figure 1. The battery of tests roads of cognitive and motor symptoms characteristic of the was performed in the least stressing and challenging order to disease. The cerebellum is not only critical for the coordination avoid potential training and learning effects. Animals were tested and control of movements but also implicated in higher func- during their light cycle, and the experimenter was blind to the tions such as cognition, speech, and emotion. Indeed, the cere- treatment until all study and analyses were completed. bellum controls balance and timing and also has an important Data are expressed as mean± SEM and analyzed using contribution in facilitating language and executive functions, all GraphPad Prism-7 software. Data that followed a normal dis- behaviors associated with ASD (Fatemi et al., 2012). tribution (all data except grooming and raring behavior) were Environmental factors acting in utero are suspected to analyzed using 1- or 2-way ANOVAs whenever appropriate. strongly contribute to the etiology of ASD (Arndt et  al., 2005). Upon significant main effects, Tukey’s or Fisher’s LSD multiple Among these, valproic acid (VPA) exposure during pregnancy comparisons were performed for behavioral or histological to manage epilepsy and mood disorders has been consistently measures, respectively. When data did not follow a normal dis- shown to be a major factor associated with developmental tribution, we conducted nonparametric tests (Kruskal-Wallis) defects leading to ASD (Christianson et al., 1994). Hence, rodent followed by Dunn’s multiple comparison tests. Pearson correla- models prenatally exposed to VPA have been increasingly used tions between behavioral and neuroanatomical readouts were to model ASD with strong construct, face, and, more recently, performed using SPSS Statistics version 21 (IBM). For all analy- predictive validity at the structural, behavioral, and pharmaco- ses, P <.05 was considered significant. logical levels (Nicolini and Fahnestock, 2018). However, most if not all of these studies have solely focused on the social and Assessment of Developmental Milestones (P9–P16) cognitive aspects of ASD and have neglected motor aspects of During early postnatal life (1–3 weeks), pups were kept in their the disease. Here, we used the VPA model of ASD to extensively deter - home cage and righting reflex and eye opening were assessed. To assess righting reflex, mice were placed in the supine posi- mine the motor disturbances associated with this pathology, both in males and females, and to identify associated brain his- tion and the time taken to right was monitored 3 times with a 5-minute inter-trial interval. Eye opening was assessed daily topathology. We then explored correlations between behavioral Downloaded from https://academic.oup.com/ijnp/article-abstract/21/9/871/4995459 by Ed 'DeepDyve' Gillespie user on 04 September 2018 Al Sagheer et al. | 873 Figure 1. Behavioral experimental plan. For mating, 3 females were transferred to the male cage and the presence of vaginal plug designated gestational day 0 (GD 0). Pregnant dams where treated with either the antiepileptic agent valproic acid (VPA) (450 mg/kg) or NaCl 0.9% on embryonic day 12.5 (E12.5). Sex and age matched pups where separated from the dams on postnatal day 21 (P21) and raised by groups of 4 in a randomized fashion to avoid littermate effects. Comprehensive behavioral screening was performed between P30 and 45. Mice were then killed, brains harvested for histological analysis. at postnatal day (P12) to P16 and scored as either 0 = both e yes 1.5-m-long glass corridor with a dim green light beamed into closed, 1= one eye open, or 2 = both eyes open. the glass walkway. The light is reflected downward and a high- speed camera captures footprints’ spatial and kinetic param- eters. Each mouse was assessed individually for 3 consecutive SHIRPA Primary Screen (P30) runs. The following parameters were analyzed: (1) stride length: We implemented the primary SHIRPA screen that serves to iden- distance between 2 consecutive placements of the same paw, (2) tify global disturbances in gait, posture, and muscle tone as well limb base of support: distance between 2 pair prints at contact as motor control and coordination abnormalities. Behavior was during each step cycle, and (3) pair gap: gap between the place- analyzed in a transparent Plexiglas arena (55 × 33 × 22  cm) with ment of the 2 trailing feet, which measures spatial coordination 11- × 11-cm square grid on the bottom and a 3-mm metal wire between the 2 pairs. crossing diagonally on top. The transfer reaction (time to initiate movement after being placed in the arena and distance covered Assessment of Sociability in the Three Chambers over 30 seconds) was first evaluated, followed by the wire man- Test (P35–P45) euver test to assess motor coordination and muscle function. Negative geotaxis to assess postural stability and coordination Social interaction was assessed using the 3-chambers test (Moy et  al., 2004). The apparatus consists of a Plexiglas box in space was evaluated as the time taken to turn and climb a 45° inclined grid (Rogers et al., 1997). (60 × 45 × 22  cm) partitioned into 3 chambers with retract- able doorways. The first phase (PHASE-I) comprises 2 identical nonsocial stimuli (inverted wire cups) placed in the opposite Spontaneous Activity in the Cylinder (P30) chambers. The second phase (PHASE-II) comprises a nonsocial Spontaneous activity in the cylinder was performed as previ- stimulus and a social stimulus (a naïve mouse with no previ- ously described (Fleming et al., 2013). Mice were put in a trans- ous contact with the tested animal). Each phase was of 10 min- parent Plexiglas cylinder (diameter: 12  cm), and their activity utes, during which time spent in each chamber was recorded. was videotaped for 3 minutes. Number of rearings and time Subsequently, a sociability index (SI) was calculated as fol- spent grooming were quantified. lows: (time exploring social chamber – time exploring nonso- cial chamber) / (time exploring social chamber + time exploring nonsocial chamber). Assessment of Motor Coordination on the Challenging Beam (P33) Tissue Processing and Immunohistochemistry The challenging beam was performed as described previously After behavioral assays (P45), males and females from each group (Fleming et al., 20042013 , ). The beam consists of four Plexiglas sec- were randomly selected for histopathological analysis. Mice tions (25 cm length) starting with a width of 3.5 cm and gradually were deeply anesthetized with ketamine/xylazine (120/20  mg/ narrowed to 0.5- by 1-cm decrements. Animals were first trained kg) and transcardially perfused with 0.9% saline at 37°C followed for 2 days to traverse the beam starting at the widest section and by 4% paraformaldehyde at 4°C. Brains were post-fixed in 4% ending at the narrowest section that led into the home cage. On the paraformaldehyde at 4°C for 24 hours before cryoprotection in test day, a mesh grid (1-cm squares) was placed over the beam sur - 30% sucrose at 4°C for 48 hours. Serial 50  µm (cerebellum) and face. Animals were videotaped while traversing the grid-surfaced 40 µm (striatum, cortex, and substantia nigra) free-floating sec- beam for 5 trials. Time to traverse, errors, number of steps, and tions were collected and stored at -20°C until use. errors per step made by each animal were measured and averaged. Purkinje cells (PC) within the cerebellum, dopaminergic neurons within the substantia nigra, and neurons in the stri- Spatial, Temporal, and Kinetic Gait Parameters (P34) atum and the motor cortex were quantified. Every fourth cere- Gait was analyzed during spontaneous walk using an automated bellar section was mounted on gelatin-coated slides, PC were gait analysis system (Viewpoint). The apparatus is made of a identified by cresyl violet staining as previously described Downloaded from https://academic.oup.com/ijnp/article-abstract/21/9/871/4995459 by Ed 'DeepDyve' Gillespie user on 04 September 2018 874 | International Journal of Neuropsychopharmacology, 2018 (Woodruff-Pak, 2006), and their phenotype was further con- firmed with calbindin immunohistochemistry (1:2500; Cb-38a, Swant). However, calbindin was not used for PC quantifi- cation, as VPA treatment can lead to reduced calbindin pro- tein expression (Main and Kulesza, 2017). Every 6th section throughout the entire striatum, substantia nigra, and motor cortex was selected and processed for either neuronal nuclei antigen (NeuN) to quantify neurons in the striatum and the motor cortex or for tyrosine hydroxylase (TH) immunoreactiv- ity to quantify dopaminergic neurons in the substantia nigra pars compacta (SNc). Sections were incubated for 90 minutes in a blocking solution (3% bovine serum albumin, 0.3% Triton X-100 in PBS 1 M, pH 7.4). Rabbit anti-NeuN (1:500; Ab177487, abcam) or mouse anti-TH (1:5000; 22941, Immunostar) was applied overnight at 4°C. Biotinylated anti-rabbit or anti- mouse IgG was used as secondary antibody (1:250; BA-1000 and BA-9200, Vector laboratories) for 1 hour at room tempera- ture. Signal was amplified with an ABC Elite kit and revealed with diaminobenzidine (Vector Laboratories). Sections were mounted on gelatin-coated slides and processed for cresyl violet counterstaining. Stereology Stereological estimates were performed using the optical frac- tionator method and systematic random sampling to obtain the total number of cerebellar PC, motor cortex neurons, stri- atal neurons, and dopaminergic nigral neurons. Each region of interest was outlined based on the mouse brain atlas (Franklin and Paxinos, 2008) at 2.5× objective and neurons were counted at 40× objective using Mercator Software (Explora Nova). Upper and lower guard zones of 1 µm were set at the top and bottom of the section to exclude lost profiles and each neuron or visible nucleus was counted (See Supplementary Table 1). Results VPA Mice Showed Delayed Postnatal Development Delays in eye opening and righting reflex were assessed at P9–P16 as indicative of general postnatal development. For eye opening (Figure 2A), 2-way ANOVA revealed a significant effect of treatment [F(3, 560)= 22.60, P < .0001], age [F(4, 560) = 830.9, P < .0001], and treatment a × ge interaction [F(12, 560)= 6.623, P < .0001]. VPA males showed a significant delay at P13 (P < .001), P14 (P < .001), and P15 (P < .01). Delay appeared earlier and per - sisted longer in VPA males compared with VPA females, which showed a delay only at P14 (P < .001). Regarding the right- ing reflex, 2-way ANOVA showed a significant effect of treat- ment [F(3, 336) = 31.05, P < .0001], age [F(2, 336) = 323.3, P < .0001], and treatment a × ge interaction [F(6, 336)= 18.64, P < .0001] Figure 2. Valproic acid (VPA) prenatal treatment affects several developmental and (Figure 2B). VPA males showed a significant latency in righting behavioral paradigms in young pups and that are more pronounced in males than at P9 and P11 compared with saline (P < .0001), and VPA females females. (A) VPA mice displayed significant delayed eye opening and (B) latency to displayed an increase in the latency only at P9 (P < .0001). Thus, righting in comparison with saline. Eye opening was assessed daily at postnatal day (P12) to P16 and scored as 0= both eyes closed, 1= one eye open, or 2 = both VPA prenatal exposure induces significant developmental eyes open. (C) On P30, only VPA males showed a significant increase in immobility delays in males compared with saline and that are more pro- (freezing) when transferred to the SHIRPA arena as well as (D) a significant decrease nounced than in VPA females. in their locomotor activity. (E) VPA mice show a significant increase in the time climbing the wire and (F) time to climb the grid during the negative geotaxis test in comparison with saline. VPA males (n = 34/34), saline males (n= 27/27), VPA females VPA Mice Showed Altered Behavior in the SHIRPA (n = 25/25), and saline females (n= 30/30). Data are expressed as mean ± SEM; 2-way Primary Screen ANOVA followed by Tukey’s posthoc were performed (*P < .05, **P < .01, ***P < .001). VPA mice showed altered behavior in different SHIRPA param- eters tested at P30 (Figure 2C–F). Time spent immobile following P < .0001], sex [F(1, 112) = 136.7, P < .0001], and treatment × sex transfer to the SHIRPA arena was assessed, and 2-way ANOVA interaction [F(1, 112) = 155.6, P < .0001] (Figure 2C). Only VPA males revealed a significant effect of treatment [F(1, 112)= 214.2, showed a 3-fold increase in immobility compared to saline Downloaded from https://academic.oup.com/ijnp/article-abstract/21/9/871/4995459 by Ed 'DeepDyve' Gillespie user on 04 September 2018 Al Sagheer et al. | 875 (P < .0001). Directly after transfer to the SHIRPA arena, sponta- VPA Mice Showed Core ASD Symptoms neous locomotor activity was assessed and an effect of treat- The 3-chambers test was used to assess social behavior in 5- ment [F(1, 112) = 129.4, P < .0001], sex [F(1, 112) = 14.64, P < .001], to 6-week-old male and female mice prenatally exposed to and treatment × sex interaction [F(1, 112) = 90.78, P < .0001] were VPA or saline. None of the treatment groups, regardless the found (Figure  2D). VPA males showed a significant decrease in sex, showed a spontaneous preference for any of the cham- locomotor activity vs saline males (-50%, P < .0001) while no dif- bers during the 10-minute habituation (PHASE I) (Figure 3A). As ference was observed in females. expected, there was a significant effect of treatment on PHASE For the wire maneuver test, 2-way ANOVA revealed a signifi- II when a mouse was introduced to one of the chambers [F(1, cant effect of treatment [F(1, 112) = 21.40, P < .0001], and posthoc 128) = 243.8, P < .0001]. Indeed, while saline males spent more analysis showed that both VPA males and females were dif- time in the social chamber than in the empty chamber (69% vs ferent from saline as they spent respectively +15% (P < .05) and 31%, P < .0001), VPA males spent similar time in both chambers +25% (P < .001) more time to climb (Figure 2E). Prenatal exposure (Figure  3B). This effect was not observed in females that spent to VPA also affected negative geotaxis with a significant effect more time in the social chamber than in the empty chamber, of treatment [F(1, 112) = 16.62, P < .0001] on time needed to turn regardless of the treatment. The effect of VPA on sociability of on the grid that was only observed in VPA females (P < .001). male mice was even more striking when results were expressed In addition, treatment [F(1, 112) = 117.0, P < .0001] and sex [F(1, as SI, with a significant effect of treatment [2-way ANOVA F(1, 112) = 6.557, P = .0118] had an effect on the time needed to climb 64) = 6.285, P < .05], sex [F(1, 64) = 5.453, P < .05] and treatment × the grid (Figure 2F). Both VPA males and females spent signifi- sex interaction [F(1, 64) = 25.75, P < .0001] (Figure 3C). Posthoc ana- cantly more time than controls to climb the grid (+78% and lysis showed that VPA males have a 7-fold decrease in sociability +59% respectively, P < .0001). This increase in time to climb was (P < .0001) compared with saline males, while females had simi- more pronounced in VPA males compared with VPA females lar SI regardless of the treatment. (P < .05). Figure 3. Valproic acid (VPA) prenatal treatment induced core symptoms of autism spectrum disorder (ASD) mainly in males. (A–C) Social behavior analysis. VPA males (n = 27/34), saline males (n= 10/27), VPA females (n= 16/25), and saline females (n= 15/30). Data are expressed as mean± SEM; 2-way ANOVA followed by Tukey’s posthoc were performed (*P < .05, **P < .01, ***P < .001). (A) Social interaction was assessed using the 3-chambers test. None of the treatment groups showed preference to any of the chambers during the 10-minute habituation (PHASE I). (B) VPA prenatally treated male mice spent less time in the social chamber and more time in the nonsocial chamber relative to controls, (C) they also show a significant decrease in sociability index. (D–E) Grooming and rearing behavior. VPA males (n = 34/34), saline males (n = 27/27), VPA females (n= 25/25), and saline females (n= 30/30). (D) VPA mice presented repetitive behavior and (E) a significant decrease in rearing exploratory behav- ior. Data did not follow a normal distribution and was analyzed using a non-parametric test (Kruskal-Wallis) followed by Dunn’s multiple comparison test. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/9/871/4995459 by Ed 'DeepDyve' Gillespie user on 04 September 2018 876 | International Journal of Neuropsychopharmacology, 2018 Repetitive and restricted behaviors are among the core P < .0001], group [F(3, 448)= 60.61, P < .0001] and beam section × symptoms and the earliest signs of ASD (APA, 2013). As the group interaction [F(9, 448) = 3.135, P < .0011] (Figure  4C). Indeed, grooming and rearing data do not follow a normal distribution, male and female VPA mice made more errors on the second, Kruskal-Wallis nonparametric test was conducted. Time spent third, and fourth beam sections compared with their saline grooming in the cylinder test was significantly affected by treat- counterparts (P < .0001 each). ment [P < .0001] (Figure 3D) in both VPA males (+69%, P < .001) and Gait analysis indicated that VPA treatment did not affect paw females (+41%, P < .05) compared with their saline counterparts area, speed, or regularity of the run in males or females. However, (Dunn’s multiple comparison test). Additionally, Kruskal-Wallis there was a significant effect of treatment [F (3, 220)= 49.86, test revealed a significant change (P < .0001) in rearing behavior P < .0001] on pair gap (Figure  4D). Indeed, when compared with (Figure 3E) as VPA males showed a significant reduction in rear - saline, VPA males and females showed significant increase in ing behavior compared with saline (-41%, P < .001) and with VPA right (+2.3-fold, P < .0001) and left pair gap (+1.62-fold, P < .01). females (-50%, P < .001) (Dunn’s multiple comparison test). In addition, there was a significant effect of treatment [F(3, 220) = 33.74, P < .0001] on limb base of support (Figure 4E). Posthoc analysis showed that both VPA males and females have a signifi- Both VPA Males and Females Show Significant cant increase in both right and left limbs base of support when Deficits in Motor Coordination and Gait compared with saline counterparts (+29 and +40%, respectively, On the challenging beam test, the time needed to traverse P < .0001), indicating that a significant ataxic gait is present in the beam was not affected by the treatment [F(1, 112) = 0.0045, both males and females following prenatal exposure to VPA. P = .9462] (Figure 4A). However, the treatment affected errors per Similarly, there was an effect of VPA [F(1, 109) = 46.39, P < .0001] step [F(1, 112) = 67.04, P < .0001] as both VPA males and females on stride length (Figure  4F), with a decreased stride length in showed a significant increase in errors per step compared with both VPA males and females (-6.5%, P < .0001) compared with saline (+125% and +110%, respectively, P < .0001; Figure 4B). There their respective saline controls. Altogether, these results indi- was also a significant effect of beam section [F(3, 448)= 101.0, cate major deficits in gait in both VPA males and females. Figure 4. Valproic acid (VPA) prenatal treatment induced motor coordination deficits and abnormal gait in both males and females. (A) VPA treatment did not affect the time needed to traverse the beam for any group of mice. (B) VPA males and females showed similar and significant increase in the number of errors per steps in com- parison to saline. (C) VPA mice showed significant deficits on the second, third, and fourth sections of the beam. (D) Gait analysis indicated that VPA males and females have a significant increase in both right and left pair gap, (E) right and left limb base of support as well as (F) a significant decrease in stride length in comparison with saline controls. For challenging beam: VPA males (n = 34/34), saline males (n= 27/27), VPA females (n= 25/25), and saline females (n= 30/30). For the gait analysis: For the challenging beam, VPA males (n = 32/34), saline males (n= 27/27), VPA females (n= 25/25), and saline females (n= 30/30). Data are expressed as mean of 5 trials ± SEM, 2-way ANOVA followed by Tukey’s posthoc were performed (*P < .05, **P < .01, ***P < .001). Downloaded from https://academic.oup.com/ijnp/article-abstract/21/9/871/4995459 by Ed 'DeepDyve' Gillespie user on 04 September 2018 Al Sagheer et al. | 877 saline (Figure  5C). There was also a significant effect of treat- Cell Loss in Motor Brain Areas of VPA Mice ment in Crus II [F (1, 34)= 8.502, P < .01] where VPA females, but In the cerebellum, we focused our analyses on lobules VI and not males, showed a significant reduction of the number of PC VII that were found to be the most affected in ASD (Courchesne (-20%, P < .01) (Figure  5D). These findings indicate that prenatal et  al., 1994; Skefos et  al., 2014) (Figure  5). In lobule VI (SIM, VPA treatment caused sex-dependent and specific regional loss 6cb), there was no effect of treatment on PC cell number of PC in the hemisphere part of the lobule VII. (Figure 5A–B). However, there was an effect of treatment in lob- To determine whether deficits in motor behavior may be ule VII depending on the sub-lobules (Crus I, Crus II, PM, and related to a loss of cortical neurons, we quantified NeuN immu- 7cb, Figure  5J–K) and sex of the VPA-treated mice. Two-way noreactive neurons in the primary and secondary motor cortex. ANOVA revealed a significant effect of treatment [F(1, 34) = 16.28, Two-way ANOVA revealed a significant sex × treatment inter - P < .001] and sex [F(1, 34) = 25.3, P < .0001] in Crus I. Posthoc ana- action [F(1, 37) = 17.61, P < .001], and posthoc analysis showed lysis revealed a significant PC loss in VPA males in Crus I (-28%, P < .001) and a trend in VPA females (P = .0594) compared with a significant cortical cell loss in VPA males (-17%, P < .001; Figure 5. Loss of neurons in mice prenatally exposed to VPA is sex and region specific. (A–F) Stereological Purkinje cell (PC) count after Nissl staining on coronal section of cerebellum. (J–K) Photos of the different sub-lobules of the lobule VII, scale bars = 1 mm. (L) Illustration of the monolayer organization of PC in the cerebellar cortex after DAB-calbindin immunolabeling, scale bars = 200  µm. No difference in PC number was found within the hemispheric part of the lobule VI, SIM (A), or in the vermal part of the lobule VI, 6cb (B). Significant loss of PC was found in the sub-lobule Crus I of the lobule VII in VPA males (C), whereas VPA females showed PC cell loss in the sub-lobule Crus II (D) of the lobule VII. (E) No effect of treatment on the number of PC in the hemispheric sub-lobule PM, or (F) in the vermal sub-lobule 7cb of the lob- ule VII was found. (G) A decrease in the number of NeuN-stained neurons in the M1/M2 motor cortex was found in VPA males (outlined area on N, scale bars = 400  µm). (H) There was no effect of treatment on number of NeuN-stained neurons in the striatum. (I) No change was found in the number of neurons expressing tyrosine hydroxylase (TH) in the substantia nigra pars compacta (SNc) after DAB-TH immunolabeling on ventral mesencephalic coronal sections (M scale bars = 400  µm). n(A) = (saline males/VPA males/saline females/VPA females) = 9/8/10/9; n(B) = 11/8/10/10; n(C) = 11/8/9/10; n(D) = 11/8/9/10; n(E) = 10/7/9/8; n(F) = 9/6/8/9; n(G) = 12/8/11/10; n(H) = 12/8/10/9; n(I) = 12/8/10/10. Statistical test: 2-way ANOVA with posthoc Fisher’s LSD test. *P < .05, **P < .01, ***P < .001 Downloaded from https://academic.oup.com/ijnp/article-abstract/21/9/871/4995459 by Ed 'DeepDyve' Gillespie user on 04 September 2018 878 | International Journal of Neuropsychopharmacology, 2018 Figure 5G). This loss was not observed in females (P = .0916), indi- explore motor disorders may permit early and reproducible cating that VPA treatment induces sex-dependent alterations in diagnosis of this disease. For instance, 5-year-old ASD children number of neurons in the motor cortex. show lower gross and fine motor scores, greater praxis errors, We then explored the effect of VPA treatment on the nigro- lower movement rates, and greater movement variability (Kaur striatal pathway, known to be involved in motor control and pro- et al., 2018). Focusing on motor behavior may also allow a bet- duction of movement. Prenatal VPA exposure did not affect the ter identification of the underlying brain circuitry involved and number of striatal neurons (Figure  5H) or the number of dopa- that may be common to behavioral hallmarks of ASD (Stoodley minergic neurons in the SNc (Figure  5I), indicating that motor et al., 2017). deficits cannot be attributed to striatal or nigral neuronal loss. Here, we show that prenatal exposure to VPA (1) provoked early developmental deficits and delayed central nervous sys- tem maturation; (2) produced ASD social core features in males Relationships between Behavioral and but not in females; (3) resulted in deficits in motor coordination Histopathological Findings and gait in males and, to a lower extent, in females; (4) induced Correlation analyses were performed to determine potential cerebellar PC loss in specific lobules and in relation to sex; and links between behavioral and histological readouts that may be (5) decreased the number of motor cortex neurons only in males. affected by VPA exposure (Table  1). Interestingly, more correla- In addition, we show that motor behavior is directly correlated tions were observed in males (46/136) than in females (17/136) with social deficits and is indicative of the extent of cell loss (Fisher exact test: P < .0001). In males, developmental delays within the cerebellum and the motor cortex. observed at P9-16 correlated with motor impairments at a more The VPA rodent model of ASD is widely used due to its proven advanced age (P30–34). For example, the delay of open eyes cor - construct, face, and, more recently, predictive validity (Nicolini related with the increase of the time spent to climb the grid and and Fahnestock, 2018). This model stems directly from clinical with the increase of the left pair gap. The longer righting reflex observations showing that prenatal VPA exposure increases up correlated with the increase of the right pair gap and both delay to 10-fold the risk to develop ASD (Moore et al., 2000 Rasalam ; of open eyes and righting reflex correlated with an increased et  al., 2005). Injection of VPA to pregnant rodent females sys- immobility time in the transfer reaction and a decreased loco- tematically induces ASD-like syndromes in the offspring (Rodier motor activity. Moreover, the righting reflex also correlated with et al., 1996) such as impaired social interaction (Schneider and the SI. Interestingly, the SI correlated with several motor impair - Przewłocki, 2005; JW Kim et  al., 2014; KC Kim 2014), repetitivAe Q4AQ3 ments such as the number of errors per step, time to climb the behavior (Gandal et  al., 2010; JW Kim et  al., 2014), and delayed grid, right pair gap, immobility time, and locomotor activity. motor development (Schneider and Przewłocki, 2005; Main Altogether, these data indicate that in VPA-exposed males, the and Kulesza, 2017). Few studies explored motor performances severity of motor impairments correlates with decreased soci- in this model and reported impaired swimming performance ability. In addition, neuronal loss in males appears to be linked (Schneider and Przewłocki, 2005) and righting reactions (Wagner to a number of behavioral parameters. Indeed, the number of et al., 2006). PC in Crus I correlated with motor impairments (errors per step, Here, and in agreement with previous results, VPA prenatally stride length, distance between the right pair print) and social exposed males and females displayed a delay in eye opening deficits. The number of neurons in the motor cortex correlated (Roullet et al., 2010) and an increased latency to righting (Iijima with the time of righting reflex and with the distance of right et  al., 2016), indicating that postnatal development and motor and left pair gap. These two histological parameters also both reflexes are affected. VPA males displayed reduced rearing in correlated with the delay of eye opening, time of immobility and the cylinder test, increased immobility, and reduced locomotor to climb the grid, and finally with the SI. activity in the SHIRPA arena. This is in line with previous find- Collectively, this analysis revealed that, in males: (1) develop- ings obtained in older animals showing reduced locomotor mental delays correlate with motor impairments later assessed, activity following prenatal VPA treatment (Kataoka et al., 2013; with social interaction deficits and with neuronal loss in Crus Yamaguchi et  al., 2017) and with clinical studies showing that I and motor cortex; (2) motor disturbances correlate with social children with ASD spend less time actively exploring their en -vir interactions deficits and neuronal loss; (3) neuronal loss also onment (Pierce and Courchesne, 2001). correlate with the decrease of the SI. Finally, (4) developmental Motor stereotypies, repetitive and incontrollable motor delays, motor impairments, sociability and neuronal loss are all behaviors, are among the core symptoms of ASD (APA, 2013). In inter-related in males. rodents, grooming behavior is often suggested to reflect repeti- In females, the time to turn and climb the grid and time to tive/stereotypic behavior (Kalueff et al., 2016). In our study, VPA climb the wire are all correlated. Performances on the grid are mice show increased grooming, confirming previous reports also correlated with the distance between the right pair print, with the same animal model (Gandal et al., 2010 Castr ; o et al., which also correlates with the right pair gap. PC loss in Crus II 2017). Motor coordination was further explored using the chal- correlated with the time to climb the wire and with the stride lenging beam test, usually performed in animal models of length. Interestingly, the number of PC in Crus I also correlated Parkinson’s disease (Fleming et al., 20042013 , ). Both VPA males with stride length and developmental delay observed with the and females showed a significant increase in the number of righting reflex test. However, there was no correlation between errors made during the beam crossing, indicating major motor PC counts in Crus I and II and the number of errors per step. coordination deficits. This is line with recent findings in chron- ically treated rats with VPA showing altered beam walking and motor coordination on a rotarod (Main and Kulesza, 2017). Discussion Gait disturbances and motor development delays were Increasing evidence indicates that motor impairments are pre- reported in ASD patients (Yirmiya and Charman, 2010). We sent in children with ASD (Ming et al., 2007). However, their con- found altered stride length, pair base of support, and pair gap tribution to the overall clinical expression of the disease remains in our VPA mice models, in line with recent findings with mice undetermined. Standardized and quantifiable procedures to bearing a TSC1 genetic knock-out (Tsai et al., 2012) or 15q11-13 Downloaded from https://academic.oup.com/ijnp/article-abstract/21/9/871/4995459 by Ed 'DeepDyve' Gillespie user on 04 September 2018 Al Sagheer et al. | 879 Table 1. Interrelations of the Screened Shifts Behavior and the Anatomical Changes MOTOR SOCIAL DVPMT SHIRPA WIRE GRID BEAM GAIT LAB 3CT HISTO OE RR TR LA WM NG1 NG2 EPS SL LPG RPG LLBS RLBS SI Crus I Crus II M1/M2 DVPMT OE coeff. .414 .596** -.530* .066 .148 .468* .370 -.167 .456* .289 .179 .350 -.474 -.535* .269 -.649** MALES P = .070 .006 .016 .781 .534 .037 .108 .482 .044 .216 .450 .131 .197 .018 .266 .002 RR .143 .643** -.586** -.212 .042 .423 .340 -.030 .270 .624** .152 .406 -.840** -.254 -.144 -.569** .558 .002 .007 .368 .859 .063 .143 .901 .249 .003 .522 .076 .005 .294 .556 .009 MOTOR TR .321 .357 -.776** .217 .209 .685** .713** -.285 .480* .401 .566** .386 -.851** -.602** -.247 -.634** .181 .134 .000 .358 .377 .001 .000 .223 .032 .080 .009 .093 .004 .006 .308 .003 LA -.068 .050 .040 -.312 -.125 -.795** -.521* .050 -.560* -.555* -.394 -.487* .690* .444 .189 .468* .781 .837 .871 .180 .600 .000 .018 .834 .010 .011 .086 .029 .040 .057 .438 .038 WM .257 .270 .020 -.482* -.384 .207 -.145 .219 .233 .173 .392 .001 -.397 -.020 .044 -.021 .289 .263 .937 .036 .095 .381 .543 .353 .322 .465 .088 .997 .290 .937 .858 .930 NG1 -.250 .379 -.167 -.421 .533* .288 .617** -.315 -.356 -.392 .106 -.077 -.724* -.241 .102 .025 .302 .109 .495 .073 .019 .218 .004 .176 .123 .087 .657 .746 .027 .320 .678 .916 NG2 .021 .179 -.148 -.362 .738** .777** .635** -.337 .411 .342 .469* .489* -.705* -.635** -.120 -.542* .932 .464 .546 .127 .000 .000 .003 .147 .072 .141 .037 .029 .034 .004 .623 .013 EPS -.307 .336 .344 .219 -.223 -.006 -.022 -.566** .158 .046 .388 .150 -.902** -.704** -.218 -.339 .202 .160 .149 .367 .359 .980 .930 .009 .506 .846 .091 .529 .001 .001 .370 .144 SL -.077 -.327 -.124 -.200 -.184 .052 -.264 -.127 -.060 .347 -.296 -.270 .368 .847** .044 .264 .754 .172 .613 .412 .450 .832 .275 .605 .803 .134 .206 .250 .330 .000 .858 .261 LPG -.125 .042 .510* .041 -.044 -.022 -.237 .161 .186 .480* .178 .523* -.590 -.340 -.123 -.534* .611 .863 .026 .866 .857 .930 .328 .511 .445 .032 .453 .018 .094 .154 .616 .015 RPG -.012 .300 .296 -.090 .322 .329 .412 .443 -.187 .274 .076 .279 -.697* -.002 -.246 -.470* .962 .212 .219 .713 .179 .169 .080 .058 .444 .255 .752 .234 .037 .993 .311 .037 LLBS -.184 .362 -.007 -.064 .430 .167 .197 .191 -.072 -.025 .349 .017 -.217 -.427 .022 -.278 .450 .128 .977 .796 .066 .495 .418 .434 .769 .918 .144 .943 .575 .068 .927 .235 RLBS -.028 .361 .063 .102 .310 .531* .591** .177 -.286 -.109 .482* -.012 -.515 -.489* -.305 -.420 .908 .128 .798 .679 .196 .019 .008 .468 .235 .657 .036 .960 .156 .034 .205 .066 SOCIAL SI .341 .868** .293 -.585 .759* .627 .920** .178 -.073 .171 .465 .487 .011 .681* -.219 .708* .409 .005 .482 .128 .029 .096 .001 .673 .864 .686 .245 .221 .980 .044 .570 .033 HISTO Crus I -.249 -.557* -.138 -.120 -.110 -.028 -.179 -.295 .510* .268 -.064 -.346 .062 -.849** .021 .443 .319 .016 .585 .635 .664 .912 .478 .234 .031 .281 .802 .160 .806 .008 .935 .057 Crus II -.191 -.317 -.325 .228 -.484* -.050 -.425 -.380 .554* .089 -.374 -.411 -.133 -.397 .455 .067 .448 .200 .188 .363 .042 .843 .079 .120 .017 .724 .126 .090 .600 .331 .058 .785 M1/M2 -.162 .444 .202 .076 -.137 -.149 -.264 .585** -.187 .232 .168 .185 -.126 .558 -.467 -.235 .508 .057 .408 .758 .577 .542 .276 .009 .444 .340 .493 .449 .607 .151 .051 .348 Females Abbreviations: Crus I, Crus II, number of PC in each sub-lobule of the cerebellum; EPS err , ors per step on the challenging beam; LA, locomotor activity: number of crossed squares; LLBS left limbs base of support; LPG , left pair g , ap; M1/M2, number of neurons in the motor cortex; NG1/2, negative geotaxis, 1: time to turn, 2: time to climb; OE, open eyes: age when both eyes are open; RR, mean of time of the righting reflex on P9 and P11; RLBS, right limbs base of support; RPG ,right pair gap; SI, social index: calculated from the 3-chambers test; SL, stride length on the gait lab; TR, transfer reaction: time of immobility; WM, wire maneuver: time to climb.  Correlation analyses using Pearson’s method. Cases filled with dark grey or light grey, respectively, for males or females, and indicated in bold, represent the significant correlations between 2 variates. Note the strong correla- tions among developmental disorders, motor impairments, social interaction deficits, and neuronal loss in males, whereas in females, no relevant correlations were observed. *P < .05, **P < .01; coeff. gives the correlation coefficient r. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/9/871/4995459 by Ed 'DeepDyve' Gillespie user on 04 September 2018 880 | International Journal of Neuropsychopharmacology, 2018 duplication (Piochon et  al., 2014)  they also in line with recent deficits and to cell loss. Moreover, the more profound cell loss, findings reported in rats exposed to VPA (Main and Kulesza, the more deficits are observed both in social interactions and 2017). Interestingly, analyses of family videos have shown that, motor behavior. within a year, movements and postures of children that will later Thus, we point out here that motor impairment is a relevant be diagnosed with ASD were asymmetric (Esposito et al., 2009). and easily implementable biomarker of ASD severity and sug- Karmel et al. (2010) showed that a decrease in arm tone is pre- gest that a biomechanical assessment of the children’s motor sent at the age of 1 month in infants who will later develop ASD abilities may be a precious adjunct tool for the diagnosis of the (Karmel et  al., 2010). Furthermore, the presence or absence of pathology. Our findings also suggest that focusing on motor fine motor delay at 14 months could predict the developmental behavior and its histopathological correlates may pave the way trajectory of children who develop ASD (Landa et al., 2013), and towards the development of therapeutic strategies aimed at spe- 70% of high-risk ASD babies who showed early motor retard- cific brain areas such as the cerebellum and the motor cortex. ation were found to later develop a communication deficit (Bhat et al., 2012). Motor deficits may even aggravate social deficits via an impaired ability to properly interact with the environment Supplementary Material and with peers. For instance, Crus I, which is affected in ASD Supplementary data are available at International Journal of and in the VPA animal model as described below, is involved in Neuropsychopharmacology online. visuospatial, motor, and cognitive processing and is thus critical to interpreting the gestures of others but also to guide skilled behavior and imitation, all essential for normal social inter - action and all affected in ASD (Marko et  al., 2015 Ne ; bel et  al., Funding 2016). This work was supported by grants from the Fondation pour la Beyond motor behavior, the present study points to a major Recherche Médicale. T.A.S. was awarded a scholarship from the sex-difference in the effect of VPA exposure on social behavior. Association of Specialization and Scientific Guidance (Lebanon). We confirm previous findings in male mice prenatally exposed O.H.  was supported by a fellowship from the Institut National to VPA showing deficits in social behavior (Roullet et  al., 2010; de la Santé et de la Recherche Médicale (INSERM) and the region JW Kim et al., 2014) and we further report that, in contrast, VPA- Poitou-Charentes. The University of Poitiers and INSERM pro- exposed females developed normal social interaction skills. This vided infrastructural support. The funders had no role in study suggests that clinical and preclinical exploration based exclu- design, data collection and analysis, decision to publish, or prep- sively on social and cognitive readouts may miss some of the aration of the manuscript. neurodevelopmental consequences of VPA exposure in females. Postmortem and brain imaging studies have consistently identified the cerebellum as one of the most abnormal brain region associated with ASD, and postmortem brains analyses Acknowledgments showed a specific loss of cerebellar PC (Bailey et al., 1998W ; egiel We thank the staff of the PREBIOS animal facility (University et al., 2014). Here we show a loss of PC within the hemispheric of Poitiers-France), Marcello Solinas, Marianne Benoit-Marand, part of lobule VII, in Crus I in males and in Crus II in females. The Afsaneh Gaillard, and Laurie Galvan for kind advice and Denis extent and regionally restricted loss is comparable to those found Couratin for technical support. in postmortem ASD brains (Skefos et al., 2014 W ; egiel et al., 2014). Interestingly, PC loss in Crus I  was found to be correlated with decreased social interaction in VPA males in accordance with the Statement of Interest known cognitive role of this sub-lobule in communication and speech in human, both altered in ASD (Stoodley et al., 2012). None. The cerebellum is reciprocally connected to areas involved in motor control such as the striatum, the motor, and premotor References cortex (Kelly and Strick, 2003Bostan et  ; al., 2010). Our results show no change in the number of striatal or dopamine nigral APA (2013) Diagnostic and statistical manual of mental disor - neurons of VPA mice, in accordance with postmortem analysis ders, 5th ed. Arlington, VA: American Psychiatric Publishing. in the striatum of ASD patients (Wegiel et al., 2014). In contrast, Arndt TL, Stodgell CJ, Rodier PM (2005) The teratology of autism. the number of cortical neurons was decreased in VPA males. Int J Dev Neurosci 23:189–199. This resonates with a previous study showing altered cellular Asperger H (1944) Die autistisehen psychopathen im kinde- morphology in the motor cortex neurons of a VPA rat model of salter. Arch Psychiatr Nurs 117:76–136. autism (Snow et al., 2008). All together our study indicates that Bailey A, Luthert P, Dean A, Harding B, Janota I, Montgomery M, the teratogenic effects of VPA on ASD symptoms extend beyond Rutter M, Lantos P (1998) A clinicopathological study of aut- expression of social skills deficits, mainly observed in males, to ism. Brain 121:889–905. reach motor deficits and specific cell loss that are observed in Bhat AN, Galloway JC, Landa RJ (2012) Relation between early both males and females. As such, some forms of ASD may be motor delay and later communication delay in infants at risk equally expressed in both sexes but underdiagnosed in women for autism. Infant Behav Dev 35:838–846. who develop better social coping strategies than male ASD Bostan AC, Dum RP, Strick PL (2010) The basal ganglia com- patients (Tierney et al., 2016). municate with the cerebellum. Proc Natl Acad Sci U S A This is the first study showing strong interrelationships 107:8452–8456. among development, sociability, motor performances, and neu- Castro K, Baronio D, Perry IS, Riesgo RDS, Gottfried C (2017) ronal loss in any mouse model of ASD. Our study indicates that The effect of ketogenic diet in an animal model of autism motor impairments are directly correlated to social interaction Downloaded from https://academic.oup.com/ijnp/article-abstract/21/9/871/4995459 by Ed 'DeepDyve' Gillespie user on 04 September 2018 Al Sagheer et al. | 881 induced by prenatal exposure to valproic acid. Nutr Neurosci neuronal differentiation regulates autism-like behavior in 20:343–350. prenatally valproic acid-exposed rat offspring. Mol Neurobiol CDC (2014) Health, United States, 2014: with special feature on 49:512–528. adults aged 55–64. Hyattsville, MD: National Center for Health Landa RJ, Gross AL, Stuart EA, Faherty A (2013) Developmental Statistics. trajectories in children with and without autism spectrum Christianson AL, Chesler N, Kromberg JG (1994) Fetal valproate disorders: the first 3 years. Child Dev 84:429–442. syndrome: clinical and neuro-developmental features in two Loomes R, Hull L, Mandy WPL (2017) What is the male-to-female sibling pairs. Dev Med Child Neurol 36:361–369. ratio in autism spectrum disorder? A  systematic review Courchesne E, Saitoh O, Yeung-Courchesne R, Press GA, Lincoln and meta-analysis. J Am Acad Child Adolesc Psychiatry AJ, Haas RH, Schreibman L (1994) Abnormality of cerebellar 56:466–474. vermian lobules VI and VII in patients with infantile autism: Main SL, Kulesza RJ (2017) Repeated prenatal exposure to val- identification of hypoplastic and hyperplastic subgroups proic acid results in cerebellar hypoplasia and ataxia. with MR imaging. AJR Am J Roentgenol 162:123–130. Neuroscience 340:34–47. Esposito G, Venuti P, Maestro S, Muratori F (2009) An exploration Marko MK, Crocetti D, Hulst T, Donchin O, Shadmehr R, Mostofsky of symmetry in early autism spectrum disorders: analysis of SH (2015) Behavioural and neural basis of anomalous motor lying. Brain Dev 31:131–138. learning in children with autism. Brain 138:784–797. Fatemi SH, et al (2012) Consensus paper: pathological role of the Ming X, Brimacombe M, Wagner GC (2007) Prevalence of motor cerebellum in autism. Cerebellum 11:777–807. impairment in autism spectrum disorders. Brain Dev Fleming SM, Salcedo J, Fernagut PO, Rockenstein E, Masliah E, 29:565–570. Levine MS, Chesselet MF (2004) Early and progressive sensori- Moore SJ, Turnpenny P, Quinn A, Glover S, Lloyd DJ, Montgomery motor anomalies in mice overexpressing wild-type human T, Dean JC (2000) A clinical study of 57 children with fetal alpha-synuclein. J Neurosci 24:9434–9440. anticonvulsant syndromes. J Med Genet 37:489–497. Fleming SM, Ekhator OR, Ghisays V (2013) Assessment of sen- Morriss-Kay G, Ruberte E, Fukiishi Y (1993) Mammalian neural sorimotor function in mouse models of Parkinson’s disease. J crest and neural crest derivatives. Ann Anat 175:501–507. Vis Exp doi: 10.3791/50303. Moy SS, Nadler JJ, Perez A, Barbaro RP, Johns JM, Magnuson TR, Franklin KBJ, Paxinos G (2008) The mouse brain in stereotaxic Piven J, Crawley JN (2004) Sociability and preference for social coordinates. Cambridge, MA: Academic Press. novelty in five inbred strains: an approach to assess autistic- Gandal MJ, Edgar JC, Ehrlichman RS, Mehta M, Roberts TP, like behavior in mice. Genes Brain Behav 3:287–302. Siegel SJ (2010) Validating γ oscillations and delayed audi- Nebel MB, Eloyan A, Nettles CA, Sweeney KL, Ament K, Ward RE, tory responses as translational biomarkers of autism. Biol Choe AS, Barber AD, Pekar JJ, Mostofsky SH (2016) Intrinsic Psychiatry 68:1100–1106. visual-motor synchrony correlates with social deficits in aut- Iijima Y, Behr K, Iijima T, Biemans B, Bischofberger J, Scheiffele P ism. Biol Psychiatry 79:633–641. (2016) Distinct defects in synaptic differentiation of neocor - Nicolini C, Fahnestock M (2018) The valproic acid-induced rodent tical neurons in response to prenatal valproate exposure. Sci model of autism. Exp Neurol 299:217–227. Rep 6:27400. Ozonoff S, Young GS, Goldring S, Greiss-Hess L, Herrera AM, Kalueff AV, Stewart AM, Song C, Berridge KC, Graybiel AM, Fentress Steele J, Macari S, Hepburn S, Rogers SJ (2008) Gross motor JC (2016) Neurobiology of rodent self-grooming and its value development, movement abnormalities, and early identifica- for translational neuroscience. Nat Rev Neurosci 17:45–59. tion of autism. J Autism Dev Disord 38:644–656. Kanner L (1943) Autistic disturbances of affective contact. Nerv Pierce K, Courchesne E (2001) Evidence for a cerebellar role in Child 2:217–250. reduced exploration and stereotyped behavior in autism. Biol Karmel BZ, Gardner JM, Meade LS, Cohen IL, London E, Flory MJ, Psychiatry 49:655–664. Lennon EM, Miroshnichenko I, Rabinowitz S, Parab S, Barone A, Piochon C, Kloth AD, Grasselli G, Titley HK, Nakayama H, Harin A (2010) Early medical and behavioral characteristics of Hashimoto K, Wan V, Simmons DH, Eissa T, Nakatani J, NICU infants later classified with ASD. Pediatrics 126:457–467. Cherskov A, Miyazaki T, Watanabe M, Takumi T, Kano M, Wang Kataoka S, Takuma K, Hara Y, Maeda Y, Ago Y, Matsuda T (2013) SS, Hansel C (2014) Cerebellar plasticity and motor learning Autism-like behaviours with transient histone hyperacety- deficits in a copy-number variation mouse model of autism. lation in mice treated prenatally with valproic acid. Int J Nat Commun 5:5586. Neuropsychopharmacol 16:91–103. Rasalam AD, Hailey H, Williams JH, Moore SJ, Turnpenny PD, Kaur M, M Srinivasan S, N Bhat A (2018) Comparing motor per - Lloyd DJ, Dean JC (2005) Characteristics of fetal anticonvul- formance, praxis, coordination, and interpersonal synchrony sant syndrome associated autistic disorder. Dev Med Child between children with and without autism spectrum dis- Neurol 47:551–555. order (ASD). Res Dev Disabil 72:79–95. Rodier PM, Ingram JL, Tisdale B, Nelson S, Romano J (1996) Kelly RM, Strick PL (2003) Cerebellar loops with motor cortex Embryological origin for autism: developmental anoma- and prefrontal cortex of a nonhuman primate. J Neurosci lies of the cranial nerve motor nuclei. J Comp Neurol 370: 23:8432–8444. 247–261. Kim JW, Seung H, Kwon KJ, Ko MJ, Lee EJ, Oh HA, Choi CS, Kim KC, Rogers DC, Fisher EM, Brown SD, Peters J, Hunter AJ, Martin JE Gonzales EL, You JS, Choi DH, Lee J, Han SH, Yang SM, Cheong (1997) Behavioral and functional analysis of mouse pheno- JH, Shin CY, Bahn GH (2014) Subchronic treatment of done- type: SHIRPA, a proposed protocol for comprehensive pheno- pezil rescues impaired social, hyperactive, and stereotypic type assessment. Mamm Genome 8:711–713. behavior in valproic acid-induced animal model of autism. Roullet FI, Wollaston L, Decatanzaro D, Foster JA (2010) Plos One 9:e104927. Behavioral and molecular changes in the mouse in response Kim KC, Lee DK, Go HS, Kim P, Choi CS, Kim JW, Jeon SJ, Song to prenatal exposure to the anti-epileptic drug valproic acid. MR, Shin CY (2014) Pax6-dependent cortical glutamatergic Neuroscience 170:514–522. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/9/871/4995459 by Ed 'DeepDyve' Gillespie user on 04 September 2018 882 | International Journal of Neuropsychopharmacology, 2018 Schneider T, Przewłocki R (2005) Behavioral alterations in rats Autistic-like behaviour and cerebellar dysfunction in pur - prenatally exposed to valproic acid: animal model of autism. kinje cell tsc1 mutant mice. Nature 488:647–651. Neuropsychopharmacology 30:80–89. Wagner GC, Reuhl KR, Cheh M, McRae P, Halladay AK (2006) A new Skefos J, Cummings C, Enzer K, Holiday J, Weed K, Levy E, Yuce T, neurobehavioral model of autism in mice: pre- and postnatal Kemper T, Bauman M (2014) Regional alterations in purkinje exposure to sodium valproate. J Autism Dev Disord 36:779–793. cell density in patients with autism. Plos One 9:e81255. Wegiel J, Flory M, Kuchna I, Nowicki K, Ma SY, Imaki H, Wegiel Snow WM, Hartle K, Ivanco TL (2008) Altered morphology of J, Cohen IL, London E, Wisniewski T, Brown WT (2014) motor cortex neurons in the VPA rat model of autism. Dev Stereological study of the neuronal number and volume of 38 Psychobiol 50:633–639. brain subdivisions of subjects diagnosed with autism reveals Stoodley CJ, Valera EM, Schmahmann JD (2012) Functional top- significant alterations restricted to the striatum, amygdala ography of the cerebellum for motor and cognitive tasks: an and cerebellum. Acta Neuropathol Commun 2:141. fmri study. Neuroimage 59:1560–1570. Woodruff-Pak DS (2006) Stereological estimation of purkinje Stoodley CJ, D’Mello AM, Ellegood J, Jakkamsetti V, Liu P, Nebel MB, neuron number in C57BL/6 mice and its relation to associa- Gibson JM, Kelly E, Meng F, Cano CA, Pascual JM, Mostofsky tive learning. Neuroscience 141:233–243. SH, Lerch JP, Tsai PT (2017) Altered cerebellar connectivity Yamaguchi H, Hara Y, Ago Y, Takano E, Hasebe S, Nakazawa T, in autism and cerebellar-mediated rescue of autism-related Hashimoto H, Matsuda T, Takuma K (2017) Environmental behaviors in mice. Nat Neurosci 20:1744–1751. enrichment attenuates behavioral abnormalities in valproic Tierney S, Burns J, Kilbey E (2016) Looking behind the mask: acid-exposed autism model mice. Behav Brain Res 333:67–73. social coping strategies of girls on the autistic spectrum. Res Yirmiya N, Charman T (2010) The prodrome of autism: early Autism Spectr Disord 23:73–83. behavioral and biological signs, regression, peri- and post- Tsai PT, Hull C, Chu Y, Greene-Colozzi E, Sadowski AR, Leech natal development and genetics. J Child Psychol Psychiatry JM, Steinberg J, Crawley JN, Regehr WG, Sahin M (2012) 51:432–458. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png International Journal of Neuropsychopharmacology Oxford University Press

Motor Impairments Correlate with Social Deficits and Restricted Neuronal Loss in an Environmental Model of Autism

Free
12 pages

Loading next page...
 
/lp/ou_press/motor-impairments-correlate-with-social-deficits-and-restricted-HO9FdAVNu7
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of CINP.
ISSN
1461-1457
eISSN
1469-5111
D.O.I.
10.1093/ijnp/pyy043
Publisher site
See Article on Publisher Site

Abstract

Downloaded from https://academic.oup.com/ijnp/article-abstract/21/9/871/4995459 by Ed 'DeepDyve' Gillespie user on 04 September 2018 International Journal of Neuropsychopharmacology (2018) 21(9): 871–882 doi:10.1093/ijnp/pyy043 Advance Access Publication: May 12, 2018 Regular Research Article regular research article Motor Impairments Correlate with Social Deficits and Restricted Neuronal Loss in an Environmental Model of Autism Tareq Al Sagheer, Obelia Haida, Anais Balbous, Maureen Francheteau, Emmanuel Matas, Pierre-Olivier Fernagut, Mohamed Jaber INSERM U-1084, Experimental and Clinical Neurosciences Laboratory, University of Poitiers, Poitiers, France (Dr Sagheer, Ms Haida, Dr Balbous, Ms Francheteau, Dr Matas, Dr Fernagut, and Dr Jaber); CHU Poitiers, Poitiers, France (Drs Balbous and Jaber); Université de Bordeaux, Institut des Maladies Neurodégénératives, UMR 5293, Bordeaux, France (Dr Fernagut); CNRS, Institut des Maladies Neurodégénératives, UMR 5293, Bordeaux, France (Dr Fernagut). T.A.S. and O.H. contributed equally to this work. Correspondence: Mohamed Jaber, PhD, Laboratoire de Neurosciences Expérimentales et Cliniques-LNEC, INSERM U-1084, Université de Poitiers, Bâtiment B36, 1 rue Georges Bonnet, BP 633, TSA 51106, 86073 POITIERS cedex9 – France (mohamed.jaber@univ-poitiers.fr). ABSTRACT Background: Motor impairments are amongst the earliest and most consistent signs of autism spectrum disorders but are not used as diagnostic criteria. In addition, the relationship between motor and cognitive impairments and their respective neural substrates remain unknown. Methods: Here, we aimed at determining whether a well-acknowledged animal model of autism spectrum disorders, the valproic acid model, displays motor impairments and whether they may correlate with social deficits and neuronal loss within motor brain areas. For this, pregnant female mice (C57BL/6J) received valproic acid (450  mg/kg) at embryonic day 12.5 and offspring underwent a battery of behavioral analyses before being killed for histological correlates in motor cortex, nigrostriatal pathway, and cerebellum. Results: We show that while valproic acid male mice show both social and motor impairments, female mice only show motor impairments. Prenatal valproic acid exposure induces specific cell loss within the motor cortex and cerebellum and that is of higher magnitude in males than in females. Finally, we demonstrate that motor dysfunction correlates with reduced social behavior and that motor and social deficits both correlate with a loss of Purkinje cells within the Crus I cerebellar area. Conclusions: Our results suggest that motor dysfunction could contribute to social and communication deficits in autism spectrum disorders and that motor and social deficits may share common neuronal substrates in the cerebellum. A systematic assessment of motor function in autism spectrum disorders may potentially help the quantitative diagnosis of autism spectrum disorders and strategies aimed at improving motor behavior may provide a global therapeutic benefit. Keywords: valproic acid, cerebellum, motor cortex, gait, Purkinje cells Received: February 9, 2018; Revised: April 4, 2018; Accepted: May 9, 2018 © The Author(s) 2018. Published by Oxford University Press on behalf of CINP. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http:// creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, 871 provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com Downloaded from https://academic.oup.com/ijnp/article-abstract/21/9/871/4995459 by Ed 'DeepDyve' Gillespie user on 04 September 2018 872 | International Journal of Neuropsychopharmacology, 2018 Significance Statement The paper’s main message is that motor impairments in ASD may be indicative of this pathology and may even underlie some of its core cognitive and behavioral features. For this, we conducted a detailed analysis of motor and social behavior in both sexes of a VPA mouse model and correlated these findings to cell loss in specific brain areas such as the cerebellum and motor cortex. Our work brings evidence that motor behavior can be used as a quantitative approach for the diagnosis of ASD and is indicative of social deficits. This opens new avenues in the diagnosis and treatment of ASD targeting the affected brain areas. and postmortem outcomes to identify their potential links and Introduction offer a new perspective on the disease by demonstrating that Autism is one of the most common and disabling neurological motor disorders are linked with social impairments and may disorders of childhood with no commonly implemented diag- share common neural substrates. nostic test or cure available. It affects 3 times more males than females with a prevalence that has been constantly increasing these last decades to reach a proportion of 1 in 68 individuals Materials and Methods (CDC, 2014; Loomes et al., 2017). In the current DSM-5 classifica- tion, 2 global symptoms are recognized as hallmarks of autism Animals spectrum disorder (ASD): disorders of social communication and restricted and repetitive behaviors (APA, 2013). However, Animal housing and experimental procedures were per - in their initial reports of autistic patients, Kanner and Asperger formed in accordance with the European Union directive described early motor deficits, including clumsy gait in what (2010/63/EU) and validated by the regional ethical committee is now recognized as ASD (Kanner, 1943; Asperger, 1944). As (approval no.  2015020415093780). C57BL/6J Mice (Charles River such, motor impairments in ASD have been receiving increased Laboratories) were housed in ventilated cages with access to attention recently, and study of their contribution to the patho- food and water ad libitum. Room temperature was maintained physiology of ASD is currently a growing field of research. For at 23°C on a 12-h-light/-dark cycle. instance, there are a number of qualitative and quantitative reports in ASD describing impairments in visio-motor and man- Experimental Design and Statistical Analysis ual dexterity tasks, limb coordination during tasks requiring balance, agility and speed as well as in gait and ataxia (Fatemi Twenty females and 10 males were used for mating. Four females et al., 2012). Furthermore, motor impairments may be amongst were placed with a single male and left overnight. Pregnant mice the earliest signs of some forms of ASD (Ozonoff et  al., 2008). received a single i.p. injection of either VPA (450 mg/kg) or NaCl Accordingly, the assessment of motor disorders might help the 0.9% at gestational day 12,5 (E12.5) (Nicolini and Fahnestock, 2018) early and quantitative diagnosis of the pathology and the identi- when the neuronal tube is closing in rodents, followed by neuro- fication of dysfunctional brain regions and circuits in ASD. Such genesis (Morriss-Kay et al., 1993). Following mating, pregnant mice approach may also pave the way for novel therapeutic interven- were left undisturbed until they gave birth. At weaning (P21), pups tions that might correct motor impairments and potentially the from different litters were allocated to the 4 experimental groups cognitive and social deficits characteristic of ASD. depending on prenatal treatment: VPA males (n = 34), saline With this in perspective, the cerebellum has attracted a males (n= 27), VPA females (n= 25), and saline females (n= 30). The renewed interest in ASD as a potential brain area at the cross- experiment timeline is presented in Figure 1. The battery of tests roads of cognitive and motor symptoms characteristic of the was performed in the least stressing and challenging order to disease. The cerebellum is not only critical for the coordination avoid potential training and learning effects. Animals were tested and control of movements but also implicated in higher func- during their light cycle, and the experimenter was blind to the tions such as cognition, speech, and emotion. Indeed, the cere- treatment until all study and analyses were completed. bellum controls balance and timing and also has an important Data are expressed as mean± SEM and analyzed using contribution in facilitating language and executive functions, all GraphPad Prism-7 software. Data that followed a normal dis- behaviors associated with ASD (Fatemi et al., 2012). tribution (all data except grooming and raring behavior) were Environmental factors acting in utero are suspected to analyzed using 1- or 2-way ANOVAs whenever appropriate. strongly contribute to the etiology of ASD (Arndt et  al., 2005). Upon significant main effects, Tukey’s or Fisher’s LSD multiple Among these, valproic acid (VPA) exposure during pregnancy comparisons were performed for behavioral or histological to manage epilepsy and mood disorders has been consistently measures, respectively. When data did not follow a normal dis- shown to be a major factor associated with developmental tribution, we conducted nonparametric tests (Kruskal-Wallis) defects leading to ASD (Christianson et al., 1994). Hence, rodent followed by Dunn’s multiple comparison tests. Pearson correla- models prenatally exposed to VPA have been increasingly used tions between behavioral and neuroanatomical readouts were to model ASD with strong construct, face, and, more recently, performed using SPSS Statistics version 21 (IBM). For all analy- predictive validity at the structural, behavioral, and pharmaco- ses, P <.05 was considered significant. logical levels (Nicolini and Fahnestock, 2018). However, most if not all of these studies have solely focused on the social and Assessment of Developmental Milestones (P9–P16) cognitive aspects of ASD and have neglected motor aspects of During early postnatal life (1–3 weeks), pups were kept in their the disease. Here, we used the VPA model of ASD to extensively deter - home cage and righting reflex and eye opening were assessed. To assess righting reflex, mice were placed in the supine posi- mine the motor disturbances associated with this pathology, both in males and females, and to identify associated brain his- tion and the time taken to right was monitored 3 times with a 5-minute inter-trial interval. Eye opening was assessed daily topathology. We then explored correlations between behavioral Downloaded from https://academic.oup.com/ijnp/article-abstract/21/9/871/4995459 by Ed 'DeepDyve' Gillespie user on 04 September 2018 Al Sagheer et al. | 873 Figure 1. Behavioral experimental plan. For mating, 3 females were transferred to the male cage and the presence of vaginal plug designated gestational day 0 (GD 0). Pregnant dams where treated with either the antiepileptic agent valproic acid (VPA) (450 mg/kg) or NaCl 0.9% on embryonic day 12.5 (E12.5). Sex and age matched pups where separated from the dams on postnatal day 21 (P21) and raised by groups of 4 in a randomized fashion to avoid littermate effects. Comprehensive behavioral screening was performed between P30 and 45. Mice were then killed, brains harvested for histological analysis. at postnatal day (P12) to P16 and scored as either 0 = both e yes 1.5-m-long glass corridor with a dim green light beamed into closed, 1= one eye open, or 2 = both eyes open. the glass walkway. The light is reflected downward and a high- speed camera captures footprints’ spatial and kinetic param- eters. Each mouse was assessed individually for 3 consecutive SHIRPA Primary Screen (P30) runs. The following parameters were analyzed: (1) stride length: We implemented the primary SHIRPA screen that serves to iden- distance between 2 consecutive placements of the same paw, (2) tify global disturbances in gait, posture, and muscle tone as well limb base of support: distance between 2 pair prints at contact as motor control and coordination abnormalities. Behavior was during each step cycle, and (3) pair gap: gap between the place- analyzed in a transparent Plexiglas arena (55 × 33 × 22  cm) with ment of the 2 trailing feet, which measures spatial coordination 11- × 11-cm square grid on the bottom and a 3-mm metal wire between the 2 pairs. crossing diagonally on top. The transfer reaction (time to initiate movement after being placed in the arena and distance covered Assessment of Sociability in the Three Chambers over 30 seconds) was first evaluated, followed by the wire man- Test (P35–P45) euver test to assess motor coordination and muscle function. Negative geotaxis to assess postural stability and coordination Social interaction was assessed using the 3-chambers test (Moy et  al., 2004). The apparatus consists of a Plexiglas box in space was evaluated as the time taken to turn and climb a 45° inclined grid (Rogers et al., 1997). (60 × 45 × 22  cm) partitioned into 3 chambers with retract- able doorways. The first phase (PHASE-I) comprises 2 identical nonsocial stimuli (inverted wire cups) placed in the opposite Spontaneous Activity in the Cylinder (P30) chambers. The second phase (PHASE-II) comprises a nonsocial Spontaneous activity in the cylinder was performed as previ- stimulus and a social stimulus (a naïve mouse with no previ- ously described (Fleming et al., 2013). Mice were put in a trans- ous contact with the tested animal). Each phase was of 10 min- parent Plexiglas cylinder (diameter: 12  cm), and their activity utes, during which time spent in each chamber was recorded. was videotaped for 3 minutes. Number of rearings and time Subsequently, a sociability index (SI) was calculated as fol- spent grooming were quantified. lows: (time exploring social chamber – time exploring nonso- cial chamber) / (time exploring social chamber + time exploring nonsocial chamber). Assessment of Motor Coordination on the Challenging Beam (P33) Tissue Processing and Immunohistochemistry The challenging beam was performed as described previously After behavioral assays (P45), males and females from each group (Fleming et al., 20042013 , ). The beam consists of four Plexiglas sec- were randomly selected for histopathological analysis. Mice tions (25 cm length) starting with a width of 3.5 cm and gradually were deeply anesthetized with ketamine/xylazine (120/20  mg/ narrowed to 0.5- by 1-cm decrements. Animals were first trained kg) and transcardially perfused with 0.9% saline at 37°C followed for 2 days to traverse the beam starting at the widest section and by 4% paraformaldehyde at 4°C. Brains were post-fixed in 4% ending at the narrowest section that led into the home cage. On the paraformaldehyde at 4°C for 24 hours before cryoprotection in test day, a mesh grid (1-cm squares) was placed over the beam sur - 30% sucrose at 4°C for 48 hours. Serial 50  µm (cerebellum) and face. Animals were videotaped while traversing the grid-surfaced 40 µm (striatum, cortex, and substantia nigra) free-floating sec- beam for 5 trials. Time to traverse, errors, number of steps, and tions were collected and stored at -20°C until use. errors per step made by each animal were measured and averaged. Purkinje cells (PC) within the cerebellum, dopaminergic neurons within the substantia nigra, and neurons in the stri- Spatial, Temporal, and Kinetic Gait Parameters (P34) atum and the motor cortex were quantified. Every fourth cere- Gait was analyzed during spontaneous walk using an automated bellar section was mounted on gelatin-coated slides, PC were gait analysis system (Viewpoint). The apparatus is made of a identified by cresyl violet staining as previously described Downloaded from https://academic.oup.com/ijnp/article-abstract/21/9/871/4995459 by Ed 'DeepDyve' Gillespie user on 04 September 2018 874 | International Journal of Neuropsychopharmacology, 2018 (Woodruff-Pak, 2006), and their phenotype was further con- firmed with calbindin immunohistochemistry (1:2500; Cb-38a, Swant). However, calbindin was not used for PC quantifi- cation, as VPA treatment can lead to reduced calbindin pro- tein expression (Main and Kulesza, 2017). Every 6th section throughout the entire striatum, substantia nigra, and motor cortex was selected and processed for either neuronal nuclei antigen (NeuN) to quantify neurons in the striatum and the motor cortex or for tyrosine hydroxylase (TH) immunoreactiv- ity to quantify dopaminergic neurons in the substantia nigra pars compacta (SNc). Sections were incubated for 90 minutes in a blocking solution (3% bovine serum albumin, 0.3% Triton X-100 in PBS 1 M, pH 7.4). Rabbit anti-NeuN (1:500; Ab177487, abcam) or mouse anti-TH (1:5000; 22941, Immunostar) was applied overnight at 4°C. Biotinylated anti-rabbit or anti- mouse IgG was used as secondary antibody (1:250; BA-1000 and BA-9200, Vector laboratories) for 1 hour at room tempera- ture. Signal was amplified with an ABC Elite kit and revealed with diaminobenzidine (Vector Laboratories). Sections were mounted on gelatin-coated slides and processed for cresyl violet counterstaining. Stereology Stereological estimates were performed using the optical frac- tionator method and systematic random sampling to obtain the total number of cerebellar PC, motor cortex neurons, stri- atal neurons, and dopaminergic nigral neurons. Each region of interest was outlined based on the mouse brain atlas (Franklin and Paxinos, 2008) at 2.5× objective and neurons were counted at 40× objective using Mercator Software (Explora Nova). Upper and lower guard zones of 1 µm were set at the top and bottom of the section to exclude lost profiles and each neuron or visible nucleus was counted (See Supplementary Table 1). Results VPA Mice Showed Delayed Postnatal Development Delays in eye opening and righting reflex were assessed at P9–P16 as indicative of general postnatal development. For eye opening (Figure 2A), 2-way ANOVA revealed a significant effect of treatment [F(3, 560)= 22.60, P < .0001], age [F(4, 560) = 830.9, P < .0001], and treatment a × ge interaction [F(12, 560)= 6.623, P < .0001]. VPA males showed a significant delay at P13 (P < .001), P14 (P < .001), and P15 (P < .01). Delay appeared earlier and per - sisted longer in VPA males compared with VPA females, which showed a delay only at P14 (P < .001). Regarding the right- ing reflex, 2-way ANOVA showed a significant effect of treat- ment [F(3, 336) = 31.05, P < .0001], age [F(2, 336) = 323.3, P < .0001], and treatment a × ge interaction [F(6, 336)= 18.64, P < .0001] Figure 2. Valproic acid (VPA) prenatal treatment affects several developmental and (Figure 2B). VPA males showed a significant latency in righting behavioral paradigms in young pups and that are more pronounced in males than at P9 and P11 compared with saline (P < .0001), and VPA females females. (A) VPA mice displayed significant delayed eye opening and (B) latency to displayed an increase in the latency only at P9 (P < .0001). Thus, righting in comparison with saline. Eye opening was assessed daily at postnatal day (P12) to P16 and scored as 0= both eyes closed, 1= one eye open, or 2 = both VPA prenatal exposure induces significant developmental eyes open. (C) On P30, only VPA males showed a significant increase in immobility delays in males compared with saline and that are more pro- (freezing) when transferred to the SHIRPA arena as well as (D) a significant decrease nounced than in VPA females. in their locomotor activity. (E) VPA mice show a significant increase in the time climbing the wire and (F) time to climb the grid during the negative geotaxis test in comparison with saline. VPA males (n = 34/34), saline males (n= 27/27), VPA females VPA Mice Showed Altered Behavior in the SHIRPA (n = 25/25), and saline females (n= 30/30). Data are expressed as mean ± SEM; 2-way Primary Screen ANOVA followed by Tukey’s posthoc were performed (*P < .05, **P < .01, ***P < .001). VPA mice showed altered behavior in different SHIRPA param- eters tested at P30 (Figure 2C–F). Time spent immobile following P < .0001], sex [F(1, 112) = 136.7, P < .0001], and treatment × sex transfer to the SHIRPA arena was assessed, and 2-way ANOVA interaction [F(1, 112) = 155.6, P < .0001] (Figure 2C). Only VPA males revealed a significant effect of treatment [F(1, 112)= 214.2, showed a 3-fold increase in immobility compared to saline Downloaded from https://academic.oup.com/ijnp/article-abstract/21/9/871/4995459 by Ed 'DeepDyve' Gillespie user on 04 September 2018 Al Sagheer et al. | 875 (P < .0001). Directly after transfer to the SHIRPA arena, sponta- VPA Mice Showed Core ASD Symptoms neous locomotor activity was assessed and an effect of treat- The 3-chambers test was used to assess social behavior in 5- ment [F(1, 112) = 129.4, P < .0001], sex [F(1, 112) = 14.64, P < .001], to 6-week-old male and female mice prenatally exposed to and treatment × sex interaction [F(1, 112) = 90.78, P < .0001] were VPA or saline. None of the treatment groups, regardless the found (Figure  2D). VPA males showed a significant decrease in sex, showed a spontaneous preference for any of the cham- locomotor activity vs saline males (-50%, P < .0001) while no dif- bers during the 10-minute habituation (PHASE I) (Figure 3A). As ference was observed in females. expected, there was a significant effect of treatment on PHASE For the wire maneuver test, 2-way ANOVA revealed a signifi- II when a mouse was introduced to one of the chambers [F(1, cant effect of treatment [F(1, 112) = 21.40, P < .0001], and posthoc 128) = 243.8, P < .0001]. Indeed, while saline males spent more analysis showed that both VPA males and females were dif- time in the social chamber than in the empty chamber (69% vs ferent from saline as they spent respectively +15% (P < .05) and 31%, P < .0001), VPA males spent similar time in both chambers +25% (P < .001) more time to climb (Figure 2E). Prenatal exposure (Figure  3B). This effect was not observed in females that spent to VPA also affected negative geotaxis with a significant effect more time in the social chamber than in the empty chamber, of treatment [F(1, 112) = 16.62, P < .0001] on time needed to turn regardless of the treatment. The effect of VPA on sociability of on the grid that was only observed in VPA females (P < .001). male mice was even more striking when results were expressed In addition, treatment [F(1, 112) = 117.0, P < .0001] and sex [F(1, as SI, with a significant effect of treatment [2-way ANOVA F(1, 112) = 6.557, P = .0118] had an effect on the time needed to climb 64) = 6.285, P < .05], sex [F(1, 64) = 5.453, P < .05] and treatment × the grid (Figure 2F). Both VPA males and females spent signifi- sex interaction [F(1, 64) = 25.75, P < .0001] (Figure 3C). Posthoc ana- cantly more time than controls to climb the grid (+78% and lysis showed that VPA males have a 7-fold decrease in sociability +59% respectively, P < .0001). This increase in time to climb was (P < .0001) compared with saline males, while females had simi- more pronounced in VPA males compared with VPA females lar SI regardless of the treatment. (P < .05). Figure 3. Valproic acid (VPA) prenatal treatment induced core symptoms of autism spectrum disorder (ASD) mainly in males. (A–C) Social behavior analysis. VPA males (n = 27/34), saline males (n= 10/27), VPA females (n= 16/25), and saline females (n= 15/30). Data are expressed as mean± SEM; 2-way ANOVA followed by Tukey’s posthoc were performed (*P < .05, **P < .01, ***P < .001). (A) Social interaction was assessed using the 3-chambers test. None of the treatment groups showed preference to any of the chambers during the 10-minute habituation (PHASE I). (B) VPA prenatally treated male mice spent less time in the social chamber and more time in the nonsocial chamber relative to controls, (C) they also show a significant decrease in sociability index. (D–E) Grooming and rearing behavior. VPA males (n = 34/34), saline males (n = 27/27), VPA females (n= 25/25), and saline females (n= 30/30). (D) VPA mice presented repetitive behavior and (E) a significant decrease in rearing exploratory behav- ior. Data did not follow a normal distribution and was analyzed using a non-parametric test (Kruskal-Wallis) followed by Dunn’s multiple comparison test. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/9/871/4995459 by Ed 'DeepDyve' Gillespie user on 04 September 2018 876 | International Journal of Neuropsychopharmacology, 2018 Repetitive and restricted behaviors are among the core P < .0001], group [F(3, 448)= 60.61, P < .0001] and beam section × symptoms and the earliest signs of ASD (APA, 2013). As the group interaction [F(9, 448) = 3.135, P < .0011] (Figure  4C). Indeed, grooming and rearing data do not follow a normal distribution, male and female VPA mice made more errors on the second, Kruskal-Wallis nonparametric test was conducted. Time spent third, and fourth beam sections compared with their saline grooming in the cylinder test was significantly affected by treat- counterparts (P < .0001 each). ment [P < .0001] (Figure 3D) in both VPA males (+69%, P < .001) and Gait analysis indicated that VPA treatment did not affect paw females (+41%, P < .05) compared with their saline counterparts area, speed, or regularity of the run in males or females. However, (Dunn’s multiple comparison test). Additionally, Kruskal-Wallis there was a significant effect of treatment [F (3, 220)= 49.86, test revealed a significant change (P < .0001) in rearing behavior P < .0001] on pair gap (Figure  4D). Indeed, when compared with (Figure 3E) as VPA males showed a significant reduction in rear - saline, VPA males and females showed significant increase in ing behavior compared with saline (-41%, P < .001) and with VPA right (+2.3-fold, P < .0001) and left pair gap (+1.62-fold, P < .01). females (-50%, P < .001) (Dunn’s multiple comparison test). In addition, there was a significant effect of treatment [F(3, 220) = 33.74, P < .0001] on limb base of support (Figure 4E). Posthoc analysis showed that both VPA males and females have a signifi- Both VPA Males and Females Show Significant cant increase in both right and left limbs base of support when Deficits in Motor Coordination and Gait compared with saline counterparts (+29 and +40%, respectively, On the challenging beam test, the time needed to traverse P < .0001), indicating that a significant ataxic gait is present in the beam was not affected by the treatment [F(1, 112) = 0.0045, both males and females following prenatal exposure to VPA. P = .9462] (Figure 4A). However, the treatment affected errors per Similarly, there was an effect of VPA [F(1, 109) = 46.39, P < .0001] step [F(1, 112) = 67.04, P < .0001] as both VPA males and females on stride length (Figure  4F), with a decreased stride length in showed a significant increase in errors per step compared with both VPA males and females (-6.5%, P < .0001) compared with saline (+125% and +110%, respectively, P < .0001; Figure 4B). There their respective saline controls. Altogether, these results indi- was also a significant effect of beam section [F(3, 448)= 101.0, cate major deficits in gait in both VPA males and females. Figure 4. Valproic acid (VPA) prenatal treatment induced motor coordination deficits and abnormal gait in both males and females. (A) VPA treatment did not affect the time needed to traverse the beam for any group of mice. (B) VPA males and females showed similar and significant increase in the number of errors per steps in com- parison to saline. (C) VPA mice showed significant deficits on the second, third, and fourth sections of the beam. (D) Gait analysis indicated that VPA males and females have a significant increase in both right and left pair gap, (E) right and left limb base of support as well as (F) a significant decrease in stride length in comparison with saline controls. For challenging beam: VPA males (n = 34/34), saline males (n= 27/27), VPA females (n= 25/25), and saline females (n= 30/30). For the gait analysis: For the challenging beam, VPA males (n = 32/34), saline males (n= 27/27), VPA females (n= 25/25), and saline females (n= 30/30). Data are expressed as mean of 5 trials ± SEM, 2-way ANOVA followed by Tukey’s posthoc were performed (*P < .05, **P < .01, ***P < .001). Downloaded from https://academic.oup.com/ijnp/article-abstract/21/9/871/4995459 by Ed 'DeepDyve' Gillespie user on 04 September 2018 Al Sagheer et al. | 877 saline (Figure  5C). There was also a significant effect of treat- Cell Loss in Motor Brain Areas of VPA Mice ment in Crus II [F (1, 34)= 8.502, P < .01] where VPA females, but In the cerebellum, we focused our analyses on lobules VI and not males, showed a significant reduction of the number of PC VII that were found to be the most affected in ASD (Courchesne (-20%, P < .01) (Figure  5D). These findings indicate that prenatal et  al., 1994; Skefos et  al., 2014) (Figure  5). In lobule VI (SIM, VPA treatment caused sex-dependent and specific regional loss 6cb), there was no effect of treatment on PC cell number of PC in the hemisphere part of the lobule VII. (Figure 5A–B). However, there was an effect of treatment in lob- To determine whether deficits in motor behavior may be ule VII depending on the sub-lobules (Crus I, Crus II, PM, and related to a loss of cortical neurons, we quantified NeuN immu- 7cb, Figure  5J–K) and sex of the VPA-treated mice. Two-way noreactive neurons in the primary and secondary motor cortex. ANOVA revealed a significant effect of treatment [F(1, 34) = 16.28, Two-way ANOVA revealed a significant sex × treatment inter - P < .001] and sex [F(1, 34) = 25.3, P < .0001] in Crus I. Posthoc ana- action [F(1, 37) = 17.61, P < .001], and posthoc analysis showed lysis revealed a significant PC loss in VPA males in Crus I (-28%, P < .001) and a trend in VPA females (P = .0594) compared with a significant cortical cell loss in VPA males (-17%, P < .001; Figure 5. Loss of neurons in mice prenatally exposed to VPA is sex and region specific. (A–F) Stereological Purkinje cell (PC) count after Nissl staining on coronal section of cerebellum. (J–K) Photos of the different sub-lobules of the lobule VII, scale bars = 1 mm. (L) Illustration of the monolayer organization of PC in the cerebellar cortex after DAB-calbindin immunolabeling, scale bars = 200  µm. No difference in PC number was found within the hemispheric part of the lobule VI, SIM (A), or in the vermal part of the lobule VI, 6cb (B). Significant loss of PC was found in the sub-lobule Crus I of the lobule VII in VPA males (C), whereas VPA females showed PC cell loss in the sub-lobule Crus II (D) of the lobule VII. (E) No effect of treatment on the number of PC in the hemispheric sub-lobule PM, or (F) in the vermal sub-lobule 7cb of the lob- ule VII was found. (G) A decrease in the number of NeuN-stained neurons in the M1/M2 motor cortex was found in VPA males (outlined area on N, scale bars = 400  µm). (H) There was no effect of treatment on number of NeuN-stained neurons in the striatum. (I) No change was found in the number of neurons expressing tyrosine hydroxylase (TH) in the substantia nigra pars compacta (SNc) after DAB-TH immunolabeling on ventral mesencephalic coronal sections (M scale bars = 400  µm). n(A) = (saline males/VPA males/saline females/VPA females) = 9/8/10/9; n(B) = 11/8/10/10; n(C) = 11/8/9/10; n(D) = 11/8/9/10; n(E) = 10/7/9/8; n(F) = 9/6/8/9; n(G) = 12/8/11/10; n(H) = 12/8/10/9; n(I) = 12/8/10/10. Statistical test: 2-way ANOVA with posthoc Fisher’s LSD test. *P < .05, **P < .01, ***P < .001 Downloaded from https://academic.oup.com/ijnp/article-abstract/21/9/871/4995459 by Ed 'DeepDyve' Gillespie user on 04 September 2018 878 | International Journal of Neuropsychopharmacology, 2018 Figure 5G). This loss was not observed in females (P = .0916), indi- explore motor disorders may permit early and reproducible cating that VPA treatment induces sex-dependent alterations in diagnosis of this disease. For instance, 5-year-old ASD children number of neurons in the motor cortex. show lower gross and fine motor scores, greater praxis errors, We then explored the effect of VPA treatment on the nigro- lower movement rates, and greater movement variability (Kaur striatal pathway, known to be involved in motor control and pro- et al., 2018). Focusing on motor behavior may also allow a bet- duction of movement. Prenatal VPA exposure did not affect the ter identification of the underlying brain circuitry involved and number of striatal neurons (Figure  5H) or the number of dopa- that may be common to behavioral hallmarks of ASD (Stoodley minergic neurons in the SNc (Figure  5I), indicating that motor et al., 2017). deficits cannot be attributed to striatal or nigral neuronal loss. Here, we show that prenatal exposure to VPA (1) provoked early developmental deficits and delayed central nervous sys- tem maturation; (2) produced ASD social core features in males Relationships between Behavioral and but not in females; (3) resulted in deficits in motor coordination Histopathological Findings and gait in males and, to a lower extent, in females; (4) induced Correlation analyses were performed to determine potential cerebellar PC loss in specific lobules and in relation to sex; and links between behavioral and histological readouts that may be (5) decreased the number of motor cortex neurons only in males. affected by VPA exposure (Table  1). Interestingly, more correla- In addition, we show that motor behavior is directly correlated tions were observed in males (46/136) than in females (17/136) with social deficits and is indicative of the extent of cell loss (Fisher exact test: P < .0001). In males, developmental delays within the cerebellum and the motor cortex. observed at P9-16 correlated with motor impairments at a more The VPA rodent model of ASD is widely used due to its proven advanced age (P30–34). For example, the delay of open eyes cor - construct, face, and, more recently, predictive validity (Nicolini related with the increase of the time spent to climb the grid and and Fahnestock, 2018). This model stems directly from clinical with the increase of the left pair gap. The longer righting reflex observations showing that prenatal VPA exposure increases up correlated with the increase of the right pair gap and both delay to 10-fold the risk to develop ASD (Moore et al., 2000 Rasalam ; of open eyes and righting reflex correlated with an increased et  al., 2005). Injection of VPA to pregnant rodent females sys- immobility time in the transfer reaction and a decreased loco- tematically induces ASD-like syndromes in the offspring (Rodier motor activity. Moreover, the righting reflex also correlated with et al., 1996) such as impaired social interaction (Schneider and the SI. Interestingly, the SI correlated with several motor impair - Przewłocki, 2005; JW Kim et  al., 2014; KC Kim 2014), repetitivAe Q4AQ3 ments such as the number of errors per step, time to climb the behavior (Gandal et  al., 2010; JW Kim et  al., 2014), and delayed grid, right pair gap, immobility time, and locomotor activity. motor development (Schneider and Przewłocki, 2005; Main Altogether, these data indicate that in VPA-exposed males, the and Kulesza, 2017). Few studies explored motor performances severity of motor impairments correlates with decreased soci- in this model and reported impaired swimming performance ability. In addition, neuronal loss in males appears to be linked (Schneider and Przewłocki, 2005) and righting reactions (Wagner to a number of behavioral parameters. Indeed, the number of et al., 2006). PC in Crus I correlated with motor impairments (errors per step, Here, and in agreement with previous results, VPA prenatally stride length, distance between the right pair print) and social exposed males and females displayed a delay in eye opening deficits. The number of neurons in the motor cortex correlated (Roullet et al., 2010) and an increased latency to righting (Iijima with the time of righting reflex and with the distance of right et  al., 2016), indicating that postnatal development and motor and left pair gap. These two histological parameters also both reflexes are affected. VPA males displayed reduced rearing in correlated with the delay of eye opening, time of immobility and the cylinder test, increased immobility, and reduced locomotor to climb the grid, and finally with the SI. activity in the SHIRPA arena. This is in line with previous find- Collectively, this analysis revealed that, in males: (1) develop- ings obtained in older animals showing reduced locomotor mental delays correlate with motor impairments later assessed, activity following prenatal VPA treatment (Kataoka et al., 2013; with social interaction deficits and with neuronal loss in Crus Yamaguchi et  al., 2017) and with clinical studies showing that I and motor cortex; (2) motor disturbances correlate with social children with ASD spend less time actively exploring their en -vir interactions deficits and neuronal loss; (3) neuronal loss also onment (Pierce and Courchesne, 2001). correlate with the decrease of the SI. Finally, (4) developmental Motor stereotypies, repetitive and incontrollable motor delays, motor impairments, sociability and neuronal loss are all behaviors, are among the core symptoms of ASD (APA, 2013). In inter-related in males. rodents, grooming behavior is often suggested to reflect repeti- In females, the time to turn and climb the grid and time to tive/stereotypic behavior (Kalueff et al., 2016). In our study, VPA climb the wire are all correlated. Performances on the grid are mice show increased grooming, confirming previous reports also correlated with the distance between the right pair print, with the same animal model (Gandal et al., 2010 Castr ; o et al., which also correlates with the right pair gap. PC loss in Crus II 2017). Motor coordination was further explored using the chal- correlated with the time to climb the wire and with the stride lenging beam test, usually performed in animal models of length. Interestingly, the number of PC in Crus I also correlated Parkinson’s disease (Fleming et al., 20042013 , ). Both VPA males with stride length and developmental delay observed with the and females showed a significant increase in the number of righting reflex test. However, there was no correlation between errors made during the beam crossing, indicating major motor PC counts in Crus I and II and the number of errors per step. coordination deficits. This is line with recent findings in chron- ically treated rats with VPA showing altered beam walking and motor coordination on a rotarod (Main and Kulesza, 2017). Discussion Gait disturbances and motor development delays were Increasing evidence indicates that motor impairments are pre- reported in ASD patients (Yirmiya and Charman, 2010). We sent in children with ASD (Ming et al., 2007). However, their con- found altered stride length, pair base of support, and pair gap tribution to the overall clinical expression of the disease remains in our VPA mice models, in line with recent findings with mice undetermined. Standardized and quantifiable procedures to bearing a TSC1 genetic knock-out (Tsai et al., 2012) or 15q11-13 Downloaded from https://academic.oup.com/ijnp/article-abstract/21/9/871/4995459 by Ed 'DeepDyve' Gillespie user on 04 September 2018 Al Sagheer et al. | 879 Table 1. Interrelations of the Screened Shifts Behavior and the Anatomical Changes MOTOR SOCIAL DVPMT SHIRPA WIRE GRID BEAM GAIT LAB 3CT HISTO OE RR TR LA WM NG1 NG2 EPS SL LPG RPG LLBS RLBS SI Crus I Crus II M1/M2 DVPMT OE coeff. .414 .596** -.530* .066 .148 .468* .370 -.167 .456* .289 .179 .350 -.474 -.535* .269 -.649** MALES P = .070 .006 .016 .781 .534 .037 .108 .482 .044 .216 .450 .131 .197 .018 .266 .002 RR .143 .643** -.586** -.212 .042 .423 .340 -.030 .270 .624** .152 .406 -.840** -.254 -.144 -.569** .558 .002 .007 .368 .859 .063 .143 .901 .249 .003 .522 .076 .005 .294 .556 .009 MOTOR TR .321 .357 -.776** .217 .209 .685** .713** -.285 .480* .401 .566** .386 -.851** -.602** -.247 -.634** .181 .134 .000 .358 .377 .001 .000 .223 .032 .080 .009 .093 .004 .006 .308 .003 LA -.068 .050 .040 -.312 -.125 -.795** -.521* .050 -.560* -.555* -.394 -.487* .690* .444 .189 .468* .781 .837 .871 .180 .600 .000 .018 .834 .010 .011 .086 .029 .040 .057 .438 .038 WM .257 .270 .020 -.482* -.384 .207 -.145 .219 .233 .173 .392 .001 -.397 -.020 .044 -.021 .289 .263 .937 .036 .095 .381 .543 .353 .322 .465 .088 .997 .290 .937 .858 .930 NG1 -.250 .379 -.167 -.421 .533* .288 .617** -.315 -.356 -.392 .106 -.077 -.724* -.241 .102 .025 .302 .109 .495 .073 .019 .218 .004 .176 .123 .087 .657 .746 .027 .320 .678 .916 NG2 .021 .179 -.148 -.362 .738** .777** .635** -.337 .411 .342 .469* .489* -.705* -.635** -.120 -.542* .932 .464 .546 .127 .000 .000 .003 .147 .072 .141 .037 .029 .034 .004 .623 .013 EPS -.307 .336 .344 .219 -.223 -.006 -.022 -.566** .158 .046 .388 .150 -.902** -.704** -.218 -.339 .202 .160 .149 .367 .359 .980 .930 .009 .506 .846 .091 .529 .001 .001 .370 .144 SL -.077 -.327 -.124 -.200 -.184 .052 -.264 -.127 -.060 .347 -.296 -.270 .368 .847** .044 .264 .754 .172 .613 .412 .450 .832 .275 .605 .803 .134 .206 .250 .330 .000 .858 .261 LPG -.125 .042 .510* .041 -.044 -.022 -.237 .161 .186 .480* .178 .523* -.590 -.340 -.123 -.534* .611 .863 .026 .866 .857 .930 .328 .511 .445 .032 .453 .018 .094 .154 .616 .015 RPG -.012 .300 .296 -.090 .322 .329 .412 .443 -.187 .274 .076 .279 -.697* -.002 -.246 -.470* .962 .212 .219 .713 .179 .169 .080 .058 .444 .255 .752 .234 .037 .993 .311 .037 LLBS -.184 .362 -.007 -.064 .430 .167 .197 .191 -.072 -.025 .349 .017 -.217 -.427 .022 -.278 .450 .128 .977 .796 .066 .495 .418 .434 .769 .918 .144 .943 .575 .068 .927 .235 RLBS -.028 .361 .063 .102 .310 .531* .591** .177 -.286 -.109 .482* -.012 -.515 -.489* -.305 -.420 .908 .128 .798 .679 .196 .019 .008 .468 .235 .657 .036 .960 .156 .034 .205 .066 SOCIAL SI .341 .868** .293 -.585 .759* .627 .920** .178 -.073 .171 .465 .487 .011 .681* -.219 .708* .409 .005 .482 .128 .029 .096 .001 .673 .864 .686 .245 .221 .980 .044 .570 .033 HISTO Crus I -.249 -.557* -.138 -.120 -.110 -.028 -.179 -.295 .510* .268 -.064 -.346 .062 -.849** .021 .443 .319 .016 .585 .635 .664 .912 .478 .234 .031 .281 .802 .160 .806 .008 .935 .057 Crus II -.191 -.317 -.325 .228 -.484* -.050 -.425 -.380 .554* .089 -.374 -.411 -.133 -.397 .455 .067 .448 .200 .188 .363 .042 .843 .079 .120 .017 .724 .126 .090 .600 .331 .058 .785 M1/M2 -.162 .444 .202 .076 -.137 -.149 -.264 .585** -.187 .232 .168 .185 -.126 .558 -.467 -.235 .508 .057 .408 .758 .577 .542 .276 .009 .444 .340 .493 .449 .607 .151 .051 .348 Females Abbreviations: Crus I, Crus II, number of PC in each sub-lobule of the cerebellum; EPS err , ors per step on the challenging beam; LA, locomotor activity: number of crossed squares; LLBS left limbs base of support; LPG , left pair g , ap; M1/M2, number of neurons in the motor cortex; NG1/2, negative geotaxis, 1: time to turn, 2: time to climb; OE, open eyes: age when both eyes are open; RR, mean of time of the righting reflex on P9 and P11; RLBS, right limbs base of support; RPG ,right pair gap; SI, social index: calculated from the 3-chambers test; SL, stride length on the gait lab; TR, transfer reaction: time of immobility; WM, wire maneuver: time to climb.  Correlation analyses using Pearson’s method. Cases filled with dark grey or light grey, respectively, for males or females, and indicated in bold, represent the significant correlations between 2 variates. Note the strong correla- tions among developmental disorders, motor impairments, social interaction deficits, and neuronal loss in males, whereas in females, no relevant correlations were observed. *P < .05, **P < .01; coeff. gives the correlation coefficient r. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/9/871/4995459 by Ed 'DeepDyve' Gillespie user on 04 September 2018 880 | International Journal of Neuropsychopharmacology, 2018 duplication (Piochon et  al., 2014)  they also in line with recent deficits and to cell loss. Moreover, the more profound cell loss, findings reported in rats exposed to VPA (Main and Kulesza, the more deficits are observed both in social interactions and 2017). Interestingly, analyses of family videos have shown that, motor behavior. within a year, movements and postures of children that will later Thus, we point out here that motor impairment is a relevant be diagnosed with ASD were asymmetric (Esposito et al., 2009). and easily implementable biomarker of ASD severity and sug- Karmel et al. (2010) showed that a decrease in arm tone is pre- gest that a biomechanical assessment of the children’s motor sent at the age of 1 month in infants who will later develop ASD abilities may be a precious adjunct tool for the diagnosis of the (Karmel et  al., 2010). Furthermore, the presence or absence of pathology. Our findings also suggest that focusing on motor fine motor delay at 14 months could predict the developmental behavior and its histopathological correlates may pave the way trajectory of children who develop ASD (Landa et al., 2013), and towards the development of therapeutic strategies aimed at spe- 70% of high-risk ASD babies who showed early motor retard- cific brain areas such as the cerebellum and the motor cortex. ation were found to later develop a communication deficit (Bhat et al., 2012). Motor deficits may even aggravate social deficits via an impaired ability to properly interact with the environment Supplementary Material and with peers. For instance, Crus I, which is affected in ASD Supplementary data are available at International Journal of and in the VPA animal model as described below, is involved in Neuropsychopharmacology online. visuospatial, motor, and cognitive processing and is thus critical to interpreting the gestures of others but also to guide skilled behavior and imitation, all essential for normal social inter - action and all affected in ASD (Marko et  al., 2015 Ne ; bel et  al., Funding 2016). This work was supported by grants from the Fondation pour la Beyond motor behavior, the present study points to a major Recherche Médicale. T.A.S. was awarded a scholarship from the sex-difference in the effect of VPA exposure on social behavior. Association of Specialization and Scientific Guidance (Lebanon). We confirm previous findings in male mice prenatally exposed O.H.  was supported by a fellowship from the Institut National to VPA showing deficits in social behavior (Roullet et  al., 2010; de la Santé et de la Recherche Médicale (INSERM) and the region JW Kim et al., 2014) and we further report that, in contrast, VPA- Poitou-Charentes. The University of Poitiers and INSERM pro- exposed females developed normal social interaction skills. This vided infrastructural support. The funders had no role in study suggests that clinical and preclinical exploration based exclu- design, data collection and analysis, decision to publish, or prep- sively on social and cognitive readouts may miss some of the aration of the manuscript. neurodevelopmental consequences of VPA exposure in females. Postmortem and brain imaging studies have consistently identified the cerebellum as one of the most abnormal brain region associated with ASD, and postmortem brains analyses Acknowledgments showed a specific loss of cerebellar PC (Bailey et al., 1998W ; egiel We thank the staff of the PREBIOS animal facility (University et al., 2014). Here we show a loss of PC within the hemispheric of Poitiers-France), Marcello Solinas, Marianne Benoit-Marand, part of lobule VII, in Crus I in males and in Crus II in females. The Afsaneh Gaillard, and Laurie Galvan for kind advice and Denis extent and regionally restricted loss is comparable to those found Couratin for technical support. in postmortem ASD brains (Skefos et al., 2014 W ; egiel et al., 2014). Interestingly, PC loss in Crus I  was found to be correlated with decreased social interaction in VPA males in accordance with the Statement of Interest known cognitive role of this sub-lobule in communication and speech in human, both altered in ASD (Stoodley et al., 2012). None. The cerebellum is reciprocally connected to areas involved in motor control such as the striatum, the motor, and premotor References cortex (Kelly and Strick, 2003Bostan et  ; al., 2010). Our results show no change in the number of striatal or dopamine nigral APA (2013) Diagnostic and statistical manual of mental disor - neurons of VPA mice, in accordance with postmortem analysis ders, 5th ed. Arlington, VA: American Psychiatric Publishing. in the striatum of ASD patients (Wegiel et al., 2014). In contrast, Arndt TL, Stodgell CJ, Rodier PM (2005) The teratology of autism. the number of cortical neurons was decreased in VPA males. Int J Dev Neurosci 23:189–199. This resonates with a previous study showing altered cellular Asperger H (1944) Die autistisehen psychopathen im kinde- morphology in the motor cortex neurons of a VPA rat model of salter. Arch Psychiatr Nurs 117:76–136. autism (Snow et al., 2008). All together our study indicates that Bailey A, Luthert P, Dean A, Harding B, Janota I, Montgomery M, the teratogenic effects of VPA on ASD symptoms extend beyond Rutter M, Lantos P (1998) A clinicopathological study of aut- expression of social skills deficits, mainly observed in males, to ism. Brain 121:889–905. reach motor deficits and specific cell loss that are observed in Bhat AN, Galloway JC, Landa RJ (2012) Relation between early both males and females. As such, some forms of ASD may be motor delay and later communication delay in infants at risk equally expressed in both sexes but underdiagnosed in women for autism. Infant Behav Dev 35:838–846. who develop better social coping strategies than male ASD Bostan AC, Dum RP, Strick PL (2010) The basal ganglia com- patients (Tierney et al., 2016). municate with the cerebellum. Proc Natl Acad Sci U S A This is the first study showing strong interrelationships 107:8452–8456. among development, sociability, motor performances, and neu- Castro K, Baronio D, Perry IS, Riesgo RDS, Gottfried C (2017) ronal loss in any mouse model of ASD. Our study indicates that The effect of ketogenic diet in an animal model of autism motor impairments are directly correlated to social interaction Downloaded from https://academic.oup.com/ijnp/article-abstract/21/9/871/4995459 by Ed 'DeepDyve' Gillespie user on 04 September 2018 Al Sagheer et al. | 881 induced by prenatal exposure to valproic acid. Nutr Neurosci neuronal differentiation regulates autism-like behavior in 20:343–350. prenatally valproic acid-exposed rat offspring. Mol Neurobiol CDC (2014) Health, United States, 2014: with special feature on 49:512–528. adults aged 55–64. Hyattsville, MD: National Center for Health Landa RJ, Gross AL, Stuart EA, Faherty A (2013) Developmental Statistics. trajectories in children with and without autism spectrum Christianson AL, Chesler N, Kromberg JG (1994) Fetal valproate disorders: the first 3 years. Child Dev 84:429–442. syndrome: clinical and neuro-developmental features in two Loomes R, Hull L, Mandy WPL (2017) What is the male-to-female sibling pairs. Dev Med Child Neurol 36:361–369. ratio in autism spectrum disorder? A  systematic review Courchesne E, Saitoh O, Yeung-Courchesne R, Press GA, Lincoln and meta-analysis. J Am Acad Child Adolesc Psychiatry AJ, Haas RH, Schreibman L (1994) Abnormality of cerebellar 56:466–474. vermian lobules VI and VII in patients with infantile autism: Main SL, Kulesza RJ (2017) Repeated prenatal exposure to val- identification of hypoplastic and hyperplastic subgroups proic acid results in cerebellar hypoplasia and ataxia. with MR imaging. AJR Am J Roentgenol 162:123–130. Neuroscience 340:34–47. Esposito G, Venuti P, Maestro S, Muratori F (2009) An exploration Marko MK, Crocetti D, Hulst T, Donchin O, Shadmehr R, Mostofsky of symmetry in early autism spectrum disorders: analysis of SH (2015) Behavioural and neural basis of anomalous motor lying. Brain Dev 31:131–138. learning in children with autism. Brain 138:784–797. Fatemi SH, et al (2012) Consensus paper: pathological role of the Ming X, Brimacombe M, Wagner GC (2007) Prevalence of motor cerebellum in autism. Cerebellum 11:777–807. impairment in autism spectrum disorders. Brain Dev Fleming SM, Salcedo J, Fernagut PO, Rockenstein E, Masliah E, 29:565–570. Levine MS, Chesselet MF (2004) Early and progressive sensori- Moore SJ, Turnpenny P, Quinn A, Glover S, Lloyd DJ, Montgomery motor anomalies in mice overexpressing wild-type human T, Dean JC (2000) A clinical study of 57 children with fetal alpha-synuclein. J Neurosci 24:9434–9440. anticonvulsant syndromes. J Med Genet 37:489–497. Fleming SM, Ekhator OR, Ghisays V (2013) Assessment of sen- Morriss-Kay G, Ruberte E, Fukiishi Y (1993) Mammalian neural sorimotor function in mouse models of Parkinson’s disease. J crest and neural crest derivatives. Ann Anat 175:501–507. Vis Exp doi: 10.3791/50303. Moy SS, Nadler JJ, Perez A, Barbaro RP, Johns JM, Magnuson TR, Franklin KBJ, Paxinos G (2008) The mouse brain in stereotaxic Piven J, Crawley JN (2004) Sociability and preference for social coordinates. Cambridge, MA: Academic Press. novelty in five inbred strains: an approach to assess autistic- Gandal MJ, Edgar JC, Ehrlichman RS, Mehta M, Roberts TP, like behavior in mice. Genes Brain Behav 3:287–302. Siegel SJ (2010) Validating γ oscillations and delayed audi- Nebel MB, Eloyan A, Nettles CA, Sweeney KL, Ament K, Ward RE, tory responses as translational biomarkers of autism. Biol Choe AS, Barber AD, Pekar JJ, Mostofsky SH (2016) Intrinsic Psychiatry 68:1100–1106. visual-motor synchrony correlates with social deficits in aut- Iijima Y, Behr K, Iijima T, Biemans B, Bischofberger J, Scheiffele P ism. Biol Psychiatry 79:633–641. (2016) Distinct defects in synaptic differentiation of neocor - Nicolini C, Fahnestock M (2018) The valproic acid-induced rodent tical neurons in response to prenatal valproate exposure. Sci model of autism. Exp Neurol 299:217–227. Rep 6:27400. Ozonoff S, Young GS, Goldring S, Greiss-Hess L, Herrera AM, Kalueff AV, Stewart AM, Song C, Berridge KC, Graybiel AM, Fentress Steele J, Macari S, Hepburn S, Rogers SJ (2008) Gross motor JC (2016) Neurobiology of rodent self-grooming and its value development, movement abnormalities, and early identifica- for translational neuroscience. Nat Rev Neurosci 17:45–59. tion of autism. J Autism Dev Disord 38:644–656. Kanner L (1943) Autistic disturbances of affective contact. Nerv Pierce K, Courchesne E (2001) Evidence for a cerebellar role in Child 2:217–250. reduced exploration and stereotyped behavior in autism. Biol Karmel BZ, Gardner JM, Meade LS, Cohen IL, London E, Flory MJ, Psychiatry 49:655–664. Lennon EM, Miroshnichenko I, Rabinowitz S, Parab S, Barone A, Piochon C, Kloth AD, Grasselli G, Titley HK, Nakayama H, Harin A (2010) Early medical and behavioral characteristics of Hashimoto K, Wan V, Simmons DH, Eissa T, Nakatani J, NICU infants later classified with ASD. Pediatrics 126:457–467. Cherskov A, Miyazaki T, Watanabe M, Takumi T, Kano M, Wang Kataoka S, Takuma K, Hara Y, Maeda Y, Ago Y, Matsuda T (2013) SS, Hansel C (2014) Cerebellar plasticity and motor learning Autism-like behaviours with transient histone hyperacety- deficits in a copy-number variation mouse model of autism. lation in mice treated prenatally with valproic acid. Int J Nat Commun 5:5586. Neuropsychopharmacol 16:91–103. Rasalam AD, Hailey H, Williams JH, Moore SJ, Turnpenny PD, Kaur M, M Srinivasan S, N Bhat A (2018) Comparing motor per - Lloyd DJ, Dean JC (2005) Characteristics of fetal anticonvul- formance, praxis, coordination, and interpersonal synchrony sant syndrome associated autistic disorder. Dev Med Child between children with and without autism spectrum dis- Neurol 47:551–555. order (ASD). Res Dev Disabil 72:79–95. Rodier PM, Ingram JL, Tisdale B, Nelson S, Romano J (1996) Kelly RM, Strick PL (2003) Cerebellar loops with motor cortex Embryological origin for autism: developmental anoma- and prefrontal cortex of a nonhuman primate. J Neurosci lies of the cranial nerve motor nuclei. J Comp Neurol 370: 23:8432–8444. 247–261. Kim JW, Seung H, Kwon KJ, Ko MJ, Lee EJ, Oh HA, Choi CS, Kim KC, Rogers DC, Fisher EM, Brown SD, Peters J, Hunter AJ, Martin JE Gonzales EL, You JS, Choi DH, Lee J, Han SH, Yang SM, Cheong (1997) Behavioral and functional analysis of mouse pheno- JH, Shin CY, Bahn GH (2014) Subchronic treatment of done- type: SHIRPA, a proposed protocol for comprehensive pheno- pezil rescues impaired social, hyperactive, and stereotypic type assessment. Mamm Genome 8:711–713. behavior in valproic acid-induced animal model of autism. Roullet FI, Wollaston L, Decatanzaro D, Foster JA (2010) Plos One 9:e104927. Behavioral and molecular changes in the mouse in response Kim KC, Lee DK, Go HS, Kim P, Choi CS, Kim JW, Jeon SJ, Song to prenatal exposure to the anti-epileptic drug valproic acid. MR, Shin CY (2014) Pax6-dependent cortical glutamatergic Neuroscience 170:514–522. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/9/871/4995459 by Ed 'DeepDyve' Gillespie user on 04 September 2018 882 | International Journal of Neuropsychopharmacology, 2018 Schneider T, Przewłocki R (2005) Behavioral alterations in rats Autistic-like behaviour and cerebellar dysfunction in pur - prenatally exposed to valproic acid: animal model of autism. kinje cell tsc1 mutant mice. Nature 488:647–651. Neuropsychopharmacology 30:80–89. Wagner GC, Reuhl KR, Cheh M, McRae P, Halladay AK (2006) A new Skefos J, Cummings C, Enzer K, Holiday J, Weed K, Levy E, Yuce T, neurobehavioral model of autism in mice: pre- and postnatal Kemper T, Bauman M (2014) Regional alterations in purkinje exposure to sodium valproate. J Autism Dev Disord 36:779–793. cell density in patients with autism. Plos One 9:e81255. Wegiel J, Flory M, Kuchna I, Nowicki K, Ma SY, Imaki H, Wegiel Snow WM, Hartle K, Ivanco TL (2008) Altered morphology of J, Cohen IL, London E, Wisniewski T, Brown WT (2014) motor cortex neurons in the VPA rat model of autism. Dev Stereological study of the neuronal number and volume of 38 Psychobiol 50:633–639. brain subdivisions of subjects diagnosed with autism reveals Stoodley CJ, Valera EM, Schmahmann JD (2012) Functional top- significant alterations restricted to the striatum, amygdala ography of the cerebellum for motor and cognitive tasks: an and cerebellum. Acta Neuropathol Commun 2:141. fmri study. Neuroimage 59:1560–1570. Woodruff-Pak DS (2006) Stereological estimation of purkinje Stoodley CJ, D’Mello AM, Ellegood J, Jakkamsetti V, Liu P, Nebel MB, neuron number in C57BL/6 mice and its relation to associa- Gibson JM, Kelly E, Meng F, Cano CA, Pascual JM, Mostofsky tive learning. Neuroscience 141:233–243. SH, Lerch JP, Tsai PT (2017) Altered cerebellar connectivity Yamaguchi H, Hara Y, Ago Y, Takano E, Hasebe S, Nakazawa T, in autism and cerebellar-mediated rescue of autism-related Hashimoto H, Matsuda T, Takuma K (2017) Environmental behaviors in mice. Nat Neurosci 20:1744–1751. enrichment attenuates behavioral abnormalities in valproic Tierney S, Burns J, Kilbey E (2016) Looking behind the mask: acid-exposed autism model mice. Behav Brain Res 333:67–73. social coping strategies of girls on the autistic spectrum. Res Yirmiya N, Charman T (2010) The prodrome of autism: early Autism Spectr Disord 23:73–83. behavioral and biological signs, regression, peri- and post- Tsai PT, Hull C, Chu Y, Greene-Colozzi E, Sadowski AR, Leech natal development and genetics. J Child Psychol Psychiatry JM, Steinberg J, Crawley JN, Regehr WG, Sahin M (2012) 51:432–458.

Journal

International Journal of NeuropsychopharmacologyOxford University Press

Published: Sep 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off