REVIEW ARTICLE published: 10 May 2013 SYSTEMS NEUROSCIENCE doi: 10.3389/fnsys.2013.00015 Is autism a disease of the cerebellum? An integration of clinical and pre-clinical research

Tiffany D. Rogers1, Eric McKimm1, Price E. Dickson1, Dan Goldowitz 2, Charles D. Blaha1* and Guy Mittleman1

1 Department of Psychology, The University of Memphis, Memphis, TN, USA 2 Centre for Molecular Medicine and Therapeutics, Department of Medical Genetics, University of British Columbia, Vancouver, BC, Canada

Edited by: Autism spectrum disorders are a group of neurodevelopmental disorders characterized Reza Shadmehr, Johns Hopkins by deficits in social skills and communication, stereotyped and repetitive behavior, and a School of Medicine, USA range of deficits in cognitive function. While the etiology of autism is unknown, current Reviewed by: research indicates that abnormalities of the cerebellum, now believed to be involved in Stewart H. Mostofsky, Kennedy Krieger Institute, USA cognitive function and the prefrontal cortex (PFC), are associated with autism. The current Mollie K. Marko, Johns Hopkins paper proposes that impaired cerebello-cortical circuitry could, at least in part, underlie School of Medicine, USA autistic symptoms. The use of animal models that allow for manipulation of genetic and *Correspondence: environmental influences are an effective means of elucidating both distal and proximal Charles D. Blaha, Department of etiological factors in autism and their potential impact on cerebello-cortical circuitry. Some Psychology, The University of Memphis, 400 Innovation existing rodent models of autism, as well as some models not previously applied to the Dr., Memphis, TN 38152-6400, USA. study of the disorder, display cerebellar and behavioral abnormalities that parallel those e-mail: [email protected] commonly seen in autistic patients. The novel findings produced from research utilizing rodent models could provide a better understanding of the neurochemical and behavioral impact of changes in cerebello-cortical circuitry in autism.

Keywords: autism, cerebellum

INTRODUCTION approximately 60–90% (Bailey et al., 1995; Hallmayer et al., AUTISM SPECTRUM DISORDERS (ASD) 2011). Studies exploring the contribution of genetics have Autism spectrum disorders are a group of neurodevelopmen- revealed that multiple genes are likely contributors to this disorder tal disorders that have traditionally included autism, Asperger’s (Muhle et al., 2004). Cytogenetic studies have found chromo- syndrome, Rett syndrome, childhood disintegrative disorder, somal abnormalities in regions such as 15q and 7q in ASD and pervasive developmental disorder not otherwise specified individuals (Muhle et al., 2004; Campbell et al., 2006; Schanen, (American Psychiatric Association, 2000). With the impending 2006; Dimitropoulos and Schultz, 2007). Variations within these arrival of the DSM-V, Asperger’s syndrome, pervasive devel- chromosomal loci are associated with an increased diagnosis of opmental disorder and childhood disintegrative disorder have autism (Campbell et al., 2006). Whole-genome scans implicate been consolidated into the overarching category of ASD, while chromosomes 1, 2, 4, 7, 10, 13, 15, 16, 17, 19, 22, and X (Muhle Rett syndrome has been removed from the diagnostic manual et al., 2004). Cytogenetic studies and whole-genome searches have (Lauritsen, 2013). Disorders within the ASD spectrum are charac- yielded several candidate genes to be further researched (Muhle terized by deficits in social skills and communication, stereotyped et al., 2004). However, all the currently known genetic variations and repetitive behavior, and a range of cognitive function deficits contributing to autism only account for approximately 5–15% of that are diagnosed on average by 4.8 years of life, although cases (Devlin and Scherer, 2012). It is therefore likely that genetic symptoms suggestive of autism have been observed as early as factors alone are not sufficient to explain the etiology of autism. 6monthsofage(Rogers and DiLalla, 1990; American Psychiatric Prenatal or perinatal environmental insults are also believed Association, 2000; Centers for Disease Control and Prevention, to contribute to autism. The developing brain is particularly sus- 2007; Ozonoff et al., 2010; Elsabbagh et al., 2012; Wolff et al., ceptible to environmental injury early in development (Rice and 2012). Deficits in cognitive function most commonly observed Barone, 2000). Prenatal or neonatal exposure to certain chemi- in ASD patients include impairments in memory and attention cals, such as ethyl alcohol, produces neuropsychological deficits as well as impairments in executive function including plan- similar to autism, and several environmental causes have recently ning, cognitive flexibility, rule acquisition, and abstract thinking been shown specifically to result in both autistic symptoms and an (Ozonoff et al., 2007). increased likelihood of a diagnosis of autism (Landrigan, 2010). These chemicals include thalidomide, misoprostol, and valproic THE ETIOLOGY OF AUTISM acid (VPA) (Moore et al., 2000; Rodier, 2002; Bandim et al., 2003). The etiology of autism is multifaceted. Genetic factors are known However, similar to genetic explanations of autism, exposure to to contribute to this disorder. According to studies investigat- known environmental agents accounts for a relatively small per- ing concordance rates in twins, the heritability of autism is centage of cases (Landrigan, 2010). Therefore, it is possible that

Frontiers in Systems Neuroscience www.frontiersin.org May 2013 | Volume 7 | Article 15 | 1 Rogers et al. Autism and the cerebellum

autism results from a specific and as yet unknown combination of of the superior cerebellar peduncles, suggesting abnormal con- genetic predisposition and environmental insults with variation nectivity between the cerebellum and the rostral-going cerebellar in these causes determining the severity of the phenotype. projections (Sivaswamy et al., 2010). Despite the current gap in understanding the etiology of autism, abnormalities in neuroanatomy, behavior, and the under- ABNORMALITIES OF CEREBELLAR FUNCTION IN AUTISM AND ASD lying genetics of patients diagnosed with autism may provide ARE ASSOCIATED WITH DEFICITS IN COGNITIVE AND MOTOR some insight into the causes of autistic symptomology. The aim BEHAVIOR, AND SOCIAL REWARD of the current review is to suggest that accumulating clinical and Imaging studies indicate that in addition to neuropathological preclinical research indicates that, regardless of etiology, develop- changes, the cerebellum is functionally abnormal in ASD patients. mental pathology of the cerebellum likely plays a very important Functional MRI (fMRI) work has revealed that during a visually role in autism and ASD. This hypothesis may provide a unify- guided saccade task, ASD patients have increased activation of ing framework for understanding the diversity of autism research cerebello-thalamic circuitry compared to controls (Takarae et al., findings as well as providing direction in the search for etiological 2007). It has been reported that activation of the ipsilateral ante- factorsinthisdisease. rior cerebellar hemisphere is increased in magnitude and larger in spatial extent during simple motor tasks in ASD patients as com- EVIDENCE THAT CEREBELLUM IS INVOLVED IN AUTISM pared to control participants (Allen et al., 2004). In adults with AND ASD Asperger’s syndrome, diffusion tensor magnetic resonance imag- Although traditionally implicated in motor function, accumu- ing showed that the level of severity of social impairment was lating evidence indicates that the cerebellum is also involved significantly negatively correlated with the degree of anisotropy in cognitive function (Schmahmann and Caplan, 2006). The of the superior cerebellar peduncle, the primary output path cerebellum is structurally and functionally abnormal in patients of the cerebellum. This suggests that as cerebellar white matter diagnosed with autism or within the ASD spectrum (Fatemi decreases, symptom severity increases (Catani et al., 2008). et al., 2012). Syndromes that share cognitive symptomology with ASD individuals and individuals with cerebellar abnormal- autism also frequently share genetic mutations associated with ities occurring during development or following acute injury abnormal cerebellar development. have similar cognitive impairments, suggesting that the cere- bellum is directly involved in autistic symptoms. Clinical stud- STRUCTURAL CEREBELLAR ABNORMALITIES ARE COMMON IN ies have shown that progressive cerebellar atrophy as well as AUTISM AND ASD acute cerebellar lesions caused by tumors or infarction result in Cerebellar neuropathology commonly occurs in ASD individuals. impairments to attention, planning, set-shifting, reversal learn- Cerebellar hypoplasia and reduced cerebellar Purkinje cell num- ing, verbal fluency, abstract reasoning, memory, and associa- bers are the most consistent neuropathologies linked to autism tive learning (Bracke-Tolkmitt et al., 1989; Botez-Marquard and (Courchesne et al., 1988, 1994; Bauman, 1991; Courchesne, 1997; Botez, 1993; Courchesne et al., 1994; Schmahmann and Sherman, Palmen et al., 2004; DiCicco-Bloom et al., 2006). MRI studies 1998; Thoma et al., 2008; Baillieux et al., 2010). In individuals report that autistic children have smaller cerebellar vermal vol- with developmental reductions in cerebellar volume, cognitive ume as compared to typically developing children (Webb et al., functions such as associative learning, verbal ability, planning, 2009). Postmortem studies report that in addition to reduced and working memory are impaired, and the degree of volume Purkinje cell numbers, microanatomic abnormalities of the cere- reduction is correlated with the degree of cognitive impairment bellum in this population include excess Bergmann glia, reduc- (Steinlin, 2008; Bolduc et al., 2011, 2012). tions in the size and number of cells in the cerebellar nuclei, and In addition to cognitive deficits, individuals with autism an active neuroinflammatory process within cerebellar white mat- exhibit motor deficits [reviewed in Gowen and Hamilton (2013)] ter (Bailey et al., 1998; Bauman and Kemper, 2005; Vargas et al., which begin in infancy (Provost et al., 2007; Brian et al., 2008) 2005). and continue through adolescence and into adulthood (Fournier Multiple disorders within the autism spectrum are also asso- et al., 2010; Van Waelvelde et al., 2010). Motor deficits appear to ciated with cerebellar abnormalities. For example, Asperger’s be relatively prevalent in this patient group, having been observed syndrome has been associated with lower total cerebellar volume in 21–100% of individuals with autism, depending on the sample and lower gray matter volume in the right cerebellum (McKelvey (Ghaziuddin et al., 1994; Manjiviona and Prior, 1995; Miyahara et al., 1995; Hallahan et al., 2009; Yu et al., 2011). Diffusion ten- et al., 1997; Green et al., 2002; Pan et al., 2009). Individuals with sor magnetic resonance imaging has demonstrated that adults autism exhibit both gross and fine motor deficits including slow with Asperger’s syndrome have reduced fractional anisotropy in and repetitive hand and foot movements (Dowell et al., 2009), the cortico-cortical parallel fibers and Purkinje cell fibers of the slow and inaccurate manual dexterity (Green et al., 2002), unsta- cerebellum and superior cerebellar peduncle of the right hemi- ble balance (Freitag et al., 2007), and impaired gait (Jansiewicz sphere, which indicates abnormal microstructure of cerebellar et al., 2006). Interestingly, motor deficits and cognitive deficits white matter. According to postmortem studies, patients with appear to be related in autism. Specifically, better motor skills pre- Rett syndrome display reduced volume of the cerebellum, cere- dict better daily living skills (Jasmin et al., 2009)anddeficitsin bellar atrophy, and reduced Purkinje cell numbers (Oldfors et al., motor control predict increased severity of autistic symptoms in 1990; Murakami et al., 1992). Diffusion tensor imaging of cere- adulthood (Sutera et al., 2007). The relationship between autism bellar pathways in children with ASD found increased diffusivity and motor deficits is unsurprising considering that (1) cerebellar

Frontiers in Systems Neuroscience www.frontiersin.org May2013|Volume7|Article15| 2 Rogers et al. Autism and the cerebellum

Purkinje cell loss is consistently reported in autism (Fatemi et al., to controls, children with autism exhibited decreased pupillary 2012) and (2) the cerebellum is able to affect regions which con- diameter when looking at happy faces. These authors interpreted trol cognitive and motor functions via reciprocal connections to these differential findings as indicative of reduced sensitivity to prefrontal cortex (PFC), posterior parietal cortex, and cortical reward value of social stimuli in children with autism. Gaze ori- motor regions (Strick et al., 2009). entation studies in which children with autism fail to orient The mechanisms through which cerebellar deficits may affect to naturally occurring social stimuli also provide evidence of motor function in autism are currently unknown. Considering impaired social reward processing in autism (Dawson et al., 1998; that the cerebellum is an important region in the development Klin et al., 2009). Neuroimaging studies indicate that differences of internal models of action (i.e., models which predict the sen- in reward circuitry likely underlie the observed differential perfor- sory consequences of motor commands and adapt to errors), one mance of controls and children with autism on behavioral tasks possibility is that motor deficits in autism result from impair- (Dichter et al., 2012a,b). In support of the hypothesis that the ments in this ability. Haswell et al. (2009) observed children with reward impairment in autism is specific to social stimuli, Cascio autism and age-matched controls as they learned to maneuver a et al. (2012) found similar responses of neural reward regions to robotic arm. These authors reported that children with autism food cues in children with autism and typically developing con- built a stronger than normal association between self-generated trols. Collectively, these studies suggest that impaired social skills motor commands and proprioceptive feedback, suggesting that in autism may result from reduced feelings of reward in response individuals with autism process and integrate motor information to social stimuli. differently than controls. Interestingly, these authors also found a Involvement of the cerebellum in deficits in social reward strong positive relationship between the reliance on propriocep- has not been investigated in humans. It is well known, how- tion and impairments in social function, suggesting the possibility ever, that disruption of afferents to the cerebellum results in of a relationship between cerebellar pathology and motor and deficits in the acquisition of classical eyeblink conditioning and cognitive function in autism. causes extinction of eyeblink conditioning in well trained ani- However, the relationship between autism and impairments in mals (McCormick et al., 1985; McCormick, 2005). Most recently the formation or use of internal models of action has not con- it has been reported that temporary, bilateral inactivation of the sistently been found. Gidley Larson et al. (2008) tested children cerebellar dentate nuclei reduced operant responding for a food with autism on a prism adaptation task and a reaching task to reward (Peterson et al., 2012). These studies are therefore con- assess their ability, relative to controls, to form internal mod- sistent with the idea that the cerebellum potentially may play an els of action. In the prism adaptation task, children were asked important role in human reward processing. to throw balls to visual targets with and without prism goggles which shifted the map of the visual environment. In the reaching MULTIPLE GENES NECESSARY FOR NORMAL CEREBELLAR task, children were required to move the handle of a robotic arm DEVELOPMENT ARE ASSOCIATED WITH THE SYMPTOMOLOGY OF which imposed forces on the hand or displaced the cursor associ- AUTISM ated with the handle position. In both tasks, children with autism Although single genes have diverse influences on brain devel- performed indistinguishably from controls, suggesting that motor opment, mutations of several genes that contribute to normal impairments seen in autism do not result from impaired abil- cerebellar development are consistently associated with increased ity to form or use internal models of action. These results are susceptibility to autism. These genes include the retinoic acid- similar to those of Mostofsky et al. (2004) who found no perfor- related orphan receptor alpha (RORα), Engrailed 2 (EN2), mance differences between children with high functioning autism Ca2+-dependent activator protein for secretion 2 (CAPS2), andcontrolsonacatchingtaskknowntodependonthecerebel- and mesenchymal-epithelial transition (MET) receptor tyrosine lum. Thus, although it appears that motor deficits in autism may kinase (Campbell et al., 2006; Wang et al., 2008; Yang et al., be related to an underlying deficit in the ability to form internal 2008; Sadakata and Furuichi, 2009; Nguyen et al., 2010; Sen et al., models of action, the factors which modulate this relationship are 2010). RORα is a gene required for differentiation of cerebellar not yet clear. Purkinje cells (Boukhtouche et al., 2010). The EN2 gene has been It should also be noted that individuals diagnosed with ASD associated with normal cerebellar development, and mutations also exhibit impairments in reward processing [reviewed in or deletions of EN2 result in reduced volume in the cerebellum Dichter and Adolphs (2012)]. Learning studies indicate that this and foliation abnormalities (Joyner, 1996; Kuemerle et al., 2007). deficit appears to be specific to social as opposed to non-social CAPS2 contributes to normal cerebellar development by enhanc- rewards. For example, Lin et al. (2012) compared performance ing release of brain-derived neurotrophic factor (BDNF) and of individuals with and without autism on an instrumental neurotrophin-3 (NT-3) (Sato et al., 2008; Sadakata and Furuichi, reward learning task in which the reward was social (pictures 2009). The MET receptor gene has been found to contribute to of positive and negative faces) or non-social (winning or losing pre- and post-natal cerebellar proliferation as well as migration of money). Individuals with autism exhibited a specific behavioral cerebrocortical cells (Powell et al., 2001, 2003; Ieraci et al., 2002). insensitivity to social rewards. In addition to learning studies, Genetic mutations associated with increased risk for syn- autonomic response studies also indicate that individuals with dromes associated with autistic symptomology are also associated autism are impaired in social reward processing. For example, with abnormal cerebellar development. Patients with fragile X Sepeta et al. (2012) examined the pupillary response in chil- syndrome, which results from a mutation of the fragile X mental dren with and without autism and found that, in comparison retardation 1 (FMR1) gene, exhibit cognitive symptoms similar

Frontiers in Systems Neuroscience www.frontiersin.org May 2013 | Volume 7 | Article 15 | 3 Rogers et al. Autism and the cerebellum

to autism such as hypersensitivity to auditory, tactile, gustatory, These changes may then result in a loss or dysregulation of cere- and olfactory stimuli, deficits in attention, mental disfunction, bellar output leading to a cascade of events which either directly and perseveration (Jin and Warren, 2000; Tsiouris and Brown, or indirectly results in the behavioral and cognitive symptoms 2004). Cerebellar abnormalities such as ectopic Purkinje cells, associated with the diagnosis of ASD. Thus, the autism discon- focal cerebellar Purkinje cell loss, and Bergmann gliosis are asso- nection hypothesis presented here posits that a disconnection in ciated with fragile X syndrome (Sabaratnam, 2000; Greco et al., the autistic brain occurs as a result of developmental cerebellar 2011). Mutations of the methyl CpG-binding protein 2 (MECP2) neuropathology and particularly the loss of Purkinje cells in the gene are known to produce Rett syndrome, a disorder within the cerebellar cortex as these cells constitute the outflow of cerebellar autism spectrum characterized by language impairments, motor cortex (Palmen et al., 2004; Amaral et al., 2008; Whitney et al., impairments, and stereotypical behavior (Amir et al., 1999). 2008). Patients with Rett syndrome frequently have cerebellar atrophy The disconnection hypothesis asserts that the etiology of a that increases with age (Murakami et al., 1992). given disorder can be accounted for by a loss of connectivity Multiple environmental agents, including VPA, an antiepilep- between two or more brain areas (Catani and Ffytche, 2008). tic drug, and chlorpyrifos, an organophosphate pesticide, have While this theory was originally used to explain literal severances been associated with an increased risk of autism (Moore et al., of circuitry, this hypothesis has more recently been expanded to 2000; de Cock et al., 2012). Basic research also indicates that include developmental neuropathology or disturbance or dys- these agents interrupt normal cerebellar development. Yochu m regulation of neural circuitry resulting in impaired functional et al. (2008) report that rats prenatally exposed to VPA exhibit connectivity (Geschwind, 1965; Geshwind and Levitt, 2007). an increased number of apoptotic cells in the cerebellum. Ingram A disconnection occurring due to a change in cerebellar output et al. (2000) also showed that prenatal exposure to VPA in rats in ASD patients could result not only in impaired communica- results in a decrease of Purkinje cells. Krishnan et al. (2012) tion between the cerebellum and its efferent targets, but also in showed that in both young and adult mice, dermal exposure compensations or adaptations of affiliated neuronal circuitry. to chlorpyrifos resulted in increased glial fibrillary acidic pro- It is possible that a disconnection of cerebello-cortical pro- tein expression of the cerebellum, suggesting neurotoxic effects jections could account for many of the cognitive symptoms in the cerebellums of these mice. Cole et al. (2011) reported that of autism. Middleton and Strick (2000) used retrograde trans- repeated postnatal exposure to chlorpyrifos resulted in changes in portation of herpes simplex virus to trace multiple closed-loop, cerebellar gene expression. Abou-Donia et al. (2006) also reported basal ganglia- and cerebello-cortical circuits. Notably, this study that rats prenatally exposed to chlorpyrifos had fewer Purkinje reported that the cerebellum projects, via the thalamus, to areas cells at postnatal day 90. 46 and 9 of the PFC which is known to be involved in cogni- It is also worth noting that an association between maternal tive functions such as memory, planning, and decision making fever during pregnancy and the risk of ASD has recently been (Middleton and Strick, 2000; Schmahmann, 2001). Research also reported. Zerbo et al. (2013) compared the incidence of mater- provides a second cerebellum-PFC circuit that could influence nal fever during pregnancy in children with ASD, developmental cognitive functions commonly attributed to the frontal cor- delay (DD), or typically developing controls. It was found that tex. The cerebellum projects indirectly to the ventral tegmental neither ASD or DD were associated with maternal influenza, but area (VTA) which contains dopamine cell bodies comprising both conditions were associated with maternal fever during preg- the mesocortical dopaminergic system that terminates in the nancy. Further, the risk of ASD was reduced in mothers who PFC (Fallon and Moore, 1978). Abundant evidence links PFC also reported taking fever reducing medications, in comparison dopaminergic function to a number of important cognitive abili- to mothers who did not. Although mechanisms underlying the ties including working memory, attentional selection, and cogni- association between maternal fever and ASD have not been inves- tive flexibility (Robbins and Roberts, 2007). Therefore, a loss of tigated, it should be remembered that the cerebellum is one brain cerebellar output could result in significant downstream effects structure that has a very prolonged developmental time course on either of these cerebello-PFC circuits that, in turn, could result that extends in humans from early embryogenesis until the first in deficits in higher cognitive functions that commonly occur postnatal years (ten Donkelaar et al., 2003). It has been suggested in ASD. that this protracted development makes the cerebellum uniquely The proposed neuronal circuitry, based on these findings, vulnerable to a broad spectrum of developmental disorders (ten by which the cerebellum modulates PFC dopamine release is Donkelaar et al., 2003). The cerebellum may also be differentially illustrated in Figure 1. Both of these neuronal circuits originate vulnerable to fever. Preclinical research indicates that hyperther- in cerebellar cortex Purkinje neurons that project to the deep mic conditions induce apoptosis, disrupt neuronal maturation cerebellar nuclei including cerebellar dentate nucleus (DN). The and result in an enhanced activation of heat shock proteins in cerebello-ventral tegmental-cortical circuit involves indirect acti- the developing cerebellum (Khan and Brown, 2002; Maroni et al., vation of mesocortical dopaminergic neurons via contralateral 2003; Aydin et al., 2011; Dean et al., 2012). projections of the DN to reticulo-tegmental nuclei (RTN) that, in turn, project to pedunculopontine nuclei (PPT) and then AUTISM DISCONNECTION HYPOTHESIS project to, and stimulate directly, VTA dopaminergic cell bod- Together, the recent findings implicating the cerebellum in autism ies projecting to the medial PFC (mPFC) (Snider et al., 1976; suggest that by genetic and/or environmental causes, neuropatho- Perciavalle et al., 1989; Schwarz and Schmitz, 1997; Garcia-Rill logical changes in the cerebellum occur in individuals with ASD. et al., 2001; Forster and Blaha, 2003; Mittleman et al., 2008;

Frontiers in Systems Neuroscience www.frontiersin.org May2013|Volume7|Article15| 4 Rogers et al. Autism and the cerebellum

the frontal cortex was directly and positively correlated with the degree of hypoplasia in the cerebellum (Carper and Courchesne, 2000). In addition, imaging studies indicate abnormal PFC func- tion in patients with autism. Positron emission tomography (PET) studies demonstrate abnormal mPFC activation during theory of mind tasks and facial expression processing in ASD patients (Castelli et al., 2002; Wang et al., 2007; Wong et al., 2008; Gilbert et al., 2009). Recent fMRI studies also show lower activ- ity levels within the mPFC in individuals with autism during a spatial working memory task (Luna et al., 2002). Furthermore, multi-voxel pattern analyses following fMRI during attentional tasks have shown that the spatial distribution of activation of the mPFC differs between those with and without autism suggesting abnormal functional organization of the mPFC in ASD patients (Gilbert et al., 2009). The involvement of either the cerebello-ventral tegmental- cortical pathway or the cerebello-thalamocortical pathway in FIGURE 1 | Neural circuitry involved in cerebellar modulation of medial autism would implicate dopaminergic dysfunction, as the cerebel- prefrontal cortex dopamine proposed to be affected by a lum modulates PFC dopamine release by both of these neuronal developmental disconnection in autism. With the exception of inhibitory cerebellar to dentate nucleus projections, red arrows indicate glutamatergic circuits. Although to-date this possibility has been sparsely inves- pathways. The green arrow indicates the mesocortical dopaminergic tigated in human patients, there is a small amount of concordant pathway. Dotted black arrow indicates feedback loop. See text for additional evidence. For example, systemic administration of dopamine details and references. antagonists such as haloperidol or risperidone has been shown to reduce aggressive behaviors, social impairments, and stereotyp- ies in ASD patients (Cohen et al., 1980; Campbell et al., 1982; Rogers et al., 2011). The cerebello-thalamocortical circuit involves Canitano, 2006). In contrast, administration of bromocriptine, activation of the contralateral projections of the DN to thalamic a dopamine agonist, has been shown to have therapeutic effects mediodorsal/ventrolateral nuclei (ThN md/vl) that send afferents such as a reduction in hyperactivity and attention deficits for to the mPFC to modulate mesocortical dopaminergic terminal ASD patients (Dollfus et al., 1992). Recent biochemical analysis release in the mPFC via appositional excitatory glutamatergic of cerebrospinal fluid (CSF) levels of dopamine metabolites such synapses (Middleton and Strick, 2001; Pinto et al., 2003; Del Arco as homovanillic acid (HVA) have produced conflicting results, and Mora, 2005; Mittleman et al., 2008). Considered together, this with some studies reporting elevated levels in ASD patients and anatomical connectivity suggests that changes in either or both others reporting no differences between patients and controls of these circuits could result in aberrant dopaminergic activity in (Lam et al., 2006). These disparate systemic drug studies and the mPFC. biochemical findings may not be surprising given that centrally If developmental cerebellar damage affects PFC processing, it administered dopamine drugs act at diverse sites throughout should be expected that abnormalities in the target structure (i.e., the brain and CSF levels of neurotransmitter metabolites reflect PFC), as well as at the level of the neural junctions within these whole brain neurotransmission activity. A single PET study has circuits (i.e., thalamus and VTA), should occur in ASD individ- demonstrated that children with autism have reduced dopamin- uals. As previously discussed, several cerebellar abnormalities are ergic activity in the mPFC (Ernst et al., 1997). While current affiliated with autism, but recent research also suggests that the evidence of dopamine dysfunction in ASD patients is inconsis- PFC and thalamus are abnormal in these individuals (Tsatsanis tent, relatively few studies have investigated dopamine localized to et al., 2003; Carper and Courchesne, 2005). While abnormalities specific brain areas. Therefore, it remains possible that dopamin- of the VTA in ASD individuals have been infrequently investi- ergic activity within certain brain areas, such as the mPFC, is gated, it has been suggested that abnormal dopamine function in abnormal in ASD individuals. the mPFC occurs in these individuals, thus implicating a role of this pathway in cognitive symptoms of ASD (Ernst et al., 1997). ABNORMALITIES OF THE THALAMUS ARE ASSOCIATED WITH AUTISM If the cerebello-thalamocortical circuit is involved in autism, ABNORMALITIES OF THE PFC ARE ASSOCIATED WITH AUTISM abnormalities of the thalamus should also be expected in ASD It might be expected that as a function of loss or dysregulation patients. Recent MRI studies indirectly corroborate this notion of cerebellar input to the PFC, that the PFC becomes structurally by concluding that while total brain volume is positively cor- and/or functionally abnormal in ASD patients. Studies investigat- related with the volume of the thalamus in control brains, a ing structural abnormalities of the PFC indicate that the frontal lack of correlation exists between total brain volume and vol- cortex is larger in individuals with autism, and the increased vol- ume of the thalamus in autistic brains (Tsatsanis et al., 2003; ume of frontal cortex has been shown to be positively correlated Hardan et al., 2006; Tamura et al., 2010).Thesizeofthetha- with autistic symptoms (Carper and Courchesne, 2005; Kumar lamus in ASD individuals is reduced when compared to con- et al., 2010). Most significantly, the degree of enlargement of trols, and the size of the left thalamus in children with ASD is

Frontiers in Systems Neuroscience www.frontiersin.org May 2013 | Volume 7 | Article 15 | 5 Rogers et al. Autism and the cerebellum

inversely correlated with stereotypical and repetitive behaviors those commonly seen in ASD patients. Genetic, viral infection, (Tamura et al., 2010; Estes et al., 2011). Mizuno et al. (2006) toxic exposure, and developmental neuropathology models exist also report that the temporal correlation between neurophysio- that display the signature psychomotor impairments and cerebel- logical events in the thalamus and the cortex in ASD individuals lar pathology of autism, and are therefore useful for investigating exceeds that of controls, as measured by functional connectivity the behavioral impact of changes in cerebello-cortical circuitry. MRI (fcMRI). Additionally, during a visually guided saccade task, The novel findings produced from research utilizing rodent mod- ASD individuals have been shown to have greater activation of els could therefore confirm and extend our understanding of the the medial thalamus as compared to controls (Takarae et al., neurochemical and behavioral impact of changes in cerebellar- 2007). The thalamus has also been implicated in cognition in cortical circuitry. non-autistic individuals as lesions to the thalamus, particularly the mediodorsal nuclei, result in autism-like cognitive deficits GENETIC MODELS such as poor working memory, perseveration, failure to inhibit As previously discussed, autism is known to be a product of mul- inappropriate behaviors, and language deficits during recall and tiple genetic mutations and is highly heritable. Genetic mutations recognition tasks (Schmahmann and Pandya, 2008; Gong et al., associated with autism, ASD, and similar cognitive disorders have 2011). been replicated in rodents. As shown in Table 1,someofthese rodent models display both autism-like behavior and cerebel- RODENT MODELS WITH CEREBELLAR ABNORMALITIES lar pathology including the FMR1 knockout (KO) mouse, the The above-mentioned research clearly implicates the cerebellum engrailed homeobox 2 (EN2) KO mouse, and the staggerer mouse in autism (Courchesne et al., 1988). Similarly, the PFC and tha- (Goldowitz and Koch, 1986; Petit et al., 1995; Lalonde et al., lamus have also been implicated in this disorder, which could 1996; Kuemerle et al., 1997; Doulazmi et al., 2001; Rogers et al., possibly be explained by downstream effects of cerebellar pathol- 2001; Ellegood et al., 2010). Similarly, the Shank3 gene has been ogy on cerebello-cortical circuitry. Abnormalities in structure and associated with behavioral and cerebellar abnormalities, making function of all these areas have been associated with the appear- Shank3 mutant mice potential candidates for investigations of ance or severity of autism symptoms (Carper and Courchesne, impaired cerebello-cortical circuitry (Peça et al., 2011; Wang et al., 2005; Tamura et al., 2010). The use of animal models, which allow 2011). Also included in this category is the Lurcher (Lc/+)mouse, for manipulation of genetic and environmental influences, is an a mutant with a gain of function in the gene that encodes the effective means of further elucidating both distal and proximal Grid2 locus (glutamate receptor, ionotropic, delta2) (Caddy and etiological factors and their potential impact on cerebello-cortical Biscoe, 1979). Although the Grid2 locus has not been associated circuitry. Some existing rodent models of autism, as well as some with ASD, there is substantial overlap between the morphological models not previously applied to the study of this disorder, dis- and behavioral phenotypes between these mice and patients with play cerebellar, behavioral, and motor abnormalities that parallel these disorders.

Table 1 | Summary of genetic rodent models with cerebellar and behavioral abnormalities that parallel those observed in autism.

Rodent model Behavioral abnormalities Cerebellar abnormalities Prefrontal cortex abnormalities

FMR1 KO mice Hyperactivity Elongated spines on Purkinje cells Hyperconnectivity of layer 5 Reduced spatial learning Decreased volume of deep cerebellar pyramidal cells Memory deficits nuclei Synapses between layer 5 Reduced fear conditioning pyramidal cells do not recover from LTD as quickly as controls

EN2 KO mice Decreased play Cerebellar hypoplasia Decreased social behaviors Decreased Purkinje cell number Reduced aggressive behavior Foliation defects

Staggerer mutant mice Impaired spatial and reversal learning Cerebellar hypoplasia Memory deficits Decreased Purkinje cell number Perseverative behavior Ectopic and atrophic Purkinje cells Abnormal responses to novel Decreased number of granule cells environments Reduced volume in deep cerebellar nuclei

Shank3 mutant mice Social abnormalities Repetitive behavior Learning and memory deficits

Lurcher mutant mice Impaired behavioral flexibility Decreased Purkinje cell number Decreased mPFC dopamine release Repetitive behavior following DN stimulation

Frontiers in Systems Neuroscience www.frontiersin.org May2013|Volume7|Article15| 6 Rogers et al. Autism and the cerebellum

FMR1 KNOCKOUT MICE to display behavioral deficits including impaired spatial learning The FMR1 gene codes for the production of the fragile X and reversal learning, memory deficits, perseverative behavior, mental retardation protein known to be involved in cognitive and abnormal responses to novel environments (Goldowitz and development, and mutations of this gene result in fragile X Koch, 1986; Misslin et al., 1986; Lalonde et al., 1996; Lalonde and syndrome (Verkerk et al., 1991; Goodrich-Hunsaker et al., 2011). Strazielle, 2008). As discussed previously, patients with fragile X syndrome exhibit autism-like cognitive symptoms and have an increased proba- Shank3 MUTANT MICE bility of an autism diagnosis (Rogers et al., 2001; Tsiouris and In humans, the Shank3 gene is found at chromosome 22q13 and Brown, 2004). Because of this, FMR1 KO mice have been used to codes for a structural protein necessary for the maintenance and model autism. These mice display several behavioral deficits sim- plasticity of postsynaptic densities located in excitatory synapses ilar to those seen in autism such as hyperactivity, perseverative (Boeckers et al., 2002; Durand et al., 2007). An abnormal Shank3 behavior, reduced spatial learning abilities, memory deficits, and mutation accounts for approximately 2% of autism cases and reduced fear conditioning (Bernardet and Crusio, 2006; Ey et al., has been associated with the occurrence of cognitive deficits and 2011; Olmos-Serrano et al., 2011). FMR1 KO mice have cerebel- disorders with similar cognitive phenotypes such as chromo- lar abnormalities such as elongated spines on cerebellar Purkinje some 22q13 deletion syndrome (Bonaglia et al., 2006; Betancur cells and decreased volume of cerebellar nuclei (Koekkoek et al., et al., 2009; Bozdagi et al., 2010; Chen et al., 2011; Peça et al., 2005; Ellegood et al., 2010). Also, eye-blink conditioning, known 2011). ASD patients with Shank3 deletions are also noted to to be dependent on synaptic plasticity of cerebellar neurons (Kim have severe core symptoms and mental disabilities (Chen et al., and Thompson, 1997), is attenuated in FMR1 KO mice suggest- 2011). Multiple models exist with Shank3 mutations, and sev- ing that these cerebellar abnormalities have functional relevance eral display behaviors analogous to the core symptoms of autism (Koekkoek et al., 2005). Abnormalities also exist in the PFC of (Bozdagi et al., 2010; Bangash et al., 2011; Peça et al., 2011; FMR1 KO mice. Layer 5 pyramidal neurons display hyperconnec- Wang et al., 2011). For example, the isoform-specific Shank3 tivity in the mPFC, and synapses between these neurons are not (e4-9) homozygous mutant exhibits abnormal social behaviors, able to recover from long term depression as rapidly as the same repetitive behaviors, and learning and memory deficits (Wang neurons in control mice (Testa-Silva et al., 2012). et al., 2011). Although research has not yet directly addressed the effect of Shank3 mutations on cerebellar anatomy and func- EN2 KNOCKOUT MICE tion, recent research suggests that mutations of Shank3 may be The EN2 gene has been associated with autism (Petit et al., 1995; related to cerebellar abnormalities in both humans and rodents. Sen et al., 2010). EN2 is a transcription factor that regulates Shank3 has been shown to be tissue-specific with high expres- gene expression necessary for normal cerebellar development. sion in the cerebellar granule cells (Beri et al., 2007; Peça et al., Thus, EN2 KO mice also have cerebellar abnormalities that resem- 2011). Shank3 has also been suggested to play an important role in ble those observed in autism such as hypoplasia, a reduction the recruitment of axon terminals to cerebellar granule cell den- in the number of Purkinje cells by approximately 40%, and drites (Roussignol et al., 2005). In addition, mutations of Shank3 foliation defects (Millen et al., 1994; Kuemerle et al., 1997). are associated with abnormalities or reduced synaptic expres- EN2 has also been identified as important for the development sion of glutamate receptors (Bangash et al., 2011; Verpelli et al., and survival of midbrain dopaminergic neurons such as those 2011). located in the VTA (Wallén and Perlmann, 2003; Sonnier et al., 2007). EN2 KO mice display social impairments which parallel LURCHER MUTANT MICE symptoms of autism including decreased play, social behavior, Lurcher mutant mice have an autosomal dominant mutation and aggressive behavior as compared to controls (Cheh et al., that results in nearly a complete loss of cerebellar Purkinje cells 2006). between the 2nd and 4th weeks of life (Caddy and Biscoe, 1979; Zuo et al., 1997). Since these mice are ataxic and display charac- STAGGERER MUTANT MICE teristic lurching movements, chimeric mice, which have a vari- The staggerer mouse has a mutation of the retinoic acid receptor- able loss of Purkinje cells dependent upon the incorporation of related orphan receptor alpha (RORα) gene which has been the wildtype lineage, have been used to examine the behavioral associated with autism (Hamilton et al., 1996; Nguyen et al., impact of Purkinje cell loss (Goldowitz et al., 1992). Lurcher mice 2010). This gene is responsible for Purkinje cell survival and dif- and chimeras display impaired executive function as measured by ferentiation (Boukhtouche et al., 2006). The staggerer mouse has a serial reversal learning task (executive function), and learning cerebellar hypoplasia resulting from a perinatal loss of approx- errors were negatively correlated with the number of Purkinje imately 80% of Purkinje cells with the remaining Purkinje cells cells in subsequent cell counts (Dickson et al., 2010). Belzung displaying defects such as reduced size and ectopic location et al. (2001) found that Lurcher mutant mice showed memory (Herrup and Mullen, 1979; Herrup et al., 1996; Doulazmi et al., deficits in a radial arm maze. Martin et al. (2010) found a negative 2001). Other cerebellar abnormalities such as a near 100% reduc- correlation between repetitive behaviors and cerebellar Purkinje tion in the number of granule cells and reduced volume of the cell number in chimeric mice. Thus, these mice exhibit executive cerebellar nuclei are also noted in these animals (Landis and dysfunction and repetitive behaviors directly related to cerebellar Sidman, 1978; Roffler-Tarlov and Herrup, 1981). These mice are Purkinje cell numbers and approximate the symptoms commonly ataxic, which limits behavioral research, but have been found observed in autism.

Frontiers in Systems Neuroscience www.frontiersin.org May 2013 | Volume 7 | Article 15 | 7 Rogers et al. Autism and the cerebellum

Recent research in Lurcher mice has also explored the neuropathology in the form of Purkinje cell loss results in autistic- cerebello-thalamocortical and cerebello-ventral tegmental- like symptomatology as well as a dysregulation of dopamine cortical circuits. Mittleman et al. (2008) used fixed potential release in the mPFC. amperometry (FPA) to detect dopamine release in the mPFC fol- It should be noted that FMR1 mutant mice that have an lowing electrical stimulation of the cerebellar Purkinje cell layer Fmr1tm1Cgr targeted mutation and Lurcher mutants have recently or DN of the cerebellum in mice. In Lurcher mice with significant been compared by Rogers et al. in order to determine if both developmental cerebellar Purkinje cell loss, mPFC dopamine mutant strains show similar changes in dopamine release in release following stimulation at either site was markedly attenu- the mPFC following electrical stimulation of the cerebellar DN ated in comparison to littermate mice with normal Purkinje cell (Rogers et al., 2013). Similar to Mittleman et al. (2008) FPA was numbers. used to monitor mPFC dopamine release evoked by DN electrical Rogers et al. (2011) further examined these two neuronal cir- stimulation in both mutant strains and their littermate controls. cuits by using FPA to detect DN stimulation-evoked dopamine As shown in Figure 2, in comparison to their respective littermate release in the mPFC of mice before and after infusions of lido- controls, mutants of both strains showed markedly attenuated caine, a local anesthetic, or kynurenate, a broad spectrum gluta- dopamine release in the mPFC following brief electrical stim- mate ionotropic receptor antagonist, into the VTA, mediodorsal ulation of the DN, a main output nucleus of the cerebellum. thalamus, or ventrolateral thalamus. Both lidocaine and kynure- The attenuation in PFC dopamine release was accompanied by nate infusions into the VTA resulted in a comparable decrease a functional reorganization of VTA and thalamic projections. in dentate stimulation-evoked mPFC dopamine release (approx- Rogers et al. (2013) observed that in Lurcher and FMR1 wildtype imately 50%). In a similar fashion, lidocaine or kynurenate infu- mice, cerebellar modulation of mPFC dopamine was mediated sions into the thalamus resulted in approximately 50% decrease equally by the VTA and thalamic pathways. However, in both the in dentate stimulation-evoked dopamine release in the mPFC Lurcher mutant and FMR1 mutants, there was a shift in modu- suggesting that these pathways equally contribute to mPFC latory control away from the VTA toward the thalamic pathway, dopamine modulation and that they are primarily, if not entirely, with a specific increase in modulatory strength of the dopamine glutamatergic. Although the Lurcher and chimera models have signal through the ventrolateral nucleus of the thalamus. These no known genetic overlap with autism, these findings are impor- results support the notion that compromised functionality of tant because they demonstrate that developmental cerebellar the projections from the DN to mPFC is a consistent result

FIGURE 2 | Individual examples of medial prefrontal cortex dopamine release responses evoked by electrical stimulation (black bar, 100 pulses at 50 Hz) of the dentate nucleus.

Frontiers in Systems Neuroscience www.frontiersin.org May2013|Volume7|Article15| 8 Rogers et al. Autism and the cerebellum

of developmental damage to the cerebellum and provides an Solbrig et al. (1996) employed high performance liquid chro- explanation for the observed attenuation of mPFC dopamine matography (HPLC) to investigate the PFC of infected rats and release. found high levels of viral nucleic acid and elevated levels of the dopamine metabolite 3,4-dihydroxy-phenylacetic acid (DOPAC) VIRAL INFECTION MODELS in the PFC suggesting a role for PFC dopamine hyperactivity in Children with autism display a variety of inflammatory-like the behavioral changes observed in this model. abnormalities in the brain and periphery suggesting autoimmune system activation (Vargas et al., 2005; Careaga et al., 2010; Rose MATERNAL INFLUENZA INFECTION MICE et al., 2012). Viral infections and maternal immune activation In a sample of mothers hospitalized during pregnancy, 50% of have therefore been suggested to play a role in the etiology of those whose offspring was diagnosed with autism were hospi- autism (Libbey et al., 2005; Patterson, 2009, 2011). Atladóttir talized due to influenza. Also, according to this study, the fre- et al. (2010) determined that maternal viral infection occurring quency of autism diagnosis among children of mothers exposed in the first trimester is associated with the diagnosis of autism. to influenza during the first trimester was approximately 6 Viral infections of developing nervous systems can alter normal times higher than in the general population (Atladóttir et al., development making the risk of neuropathology following pre- 2010). Similarly, rodents prenatally or neonatally exposed to natal or neonatal exposure particularly high for areas including influenza virus display a range of behavioral deficits including the cerebellum and neocortex that continue to develop after birth social impairments and cognitive deficits, such as impaired work- (Johnson, 1972; Sells et al., 1975). As shown in Table 2, cerebel- ing memory and impaired emotional behavior (Shi et al., 2003; lar and frontal cortex abnormalities following viral infection have Beraki et al., 2005; Asp et al., 2009). Atrophy of the cerebellum been demonstrated in two rodent models, the neonatal Borna dis- and gene expression changes in the PFC and cerebellum have ease virus (BDV) infection rat model and the maternal influenza also been observed in mice prenatally exposed to influenza virus infection mouse model. These two animal models also display (Fatemi et al., 2008, 2009). Most significantly, prenatal exposure autism-like behavioral deficits making them useful for evaluating to influenza in mice results in approximately 33% decrease in behavioral changes following developmental cerebellar pathology. Purkinje cell numbers, heterotopic Purkinje cells, and delayed migration of granule cells in the cerebellum (Shi et al., 2009). NEONATAL BORNA DISEASE VIRUS INFECTION RATS Also, Fatemi et al. (2004) observed an increase in GAD 67-kDa BDV is a non-segmented, negative, single-strand virus that repli- protein, a rate limiting enzyme that catalyzes synthesis of GABA cates in neuronal nuclei of many species and produces a range from L-glutamate, in mice exposed to influenza and suggested of neurological symptoms such as motor, sensory, and emo- that this was evidence of aberrant glutamatergic activity. tional impairments (Richt et al., 1992; Eisenman et al., 1999; Pletnikov et al., 1999). Although BDV is not considered a poten- TOXIC EXPOSURE MODEL tial cause of autism, BDV rodent models have neuroanatomical Environmental factors such as exposure to toxic chemicals dur- and behavioral abnormalities representative of the disorder. Rats ing pregnancy are also thought to contribute to autism. Thus, neonatally exposed to BDV display several behavioral impair- some rodent models utilize prenatal or neonatal exposure to ter- ments including hyperactivity, impaired spatial learning, memory atogens known to be associated with autism. As shown in Table 3, deficits, decreased open-field exploration, stereotypic behavior, one such model is the prenatal valproic acid exposure rat model and deficits in social behavior (Hornig et al., 1999; Pletnikov et al., which displays signature behavioral and cerebellar abnormalities 1999; Rubin et al., 1999). Neonatal BDV infection in rats pro- that parallel those observed in autism (Wagner et al., 2006). duces cerebellar hypoplasia and has a relatively high specificity for Purkinje cells resulting in up to 75% loss of these cells (Bautista PRENATAL VALPROIC ACID EXPOSURE RATS et al., 1995; Eisenman et al., 1999; Hornig et al., 1999; Zocher VPA is an antiepileptic drug found to be positively correlated with et al., 2000; Pletnikov et al., 2003; Williams and Lipkin, 2006). theoccurrenceofautismintheoffspringofmothersexposed

Table 2 | Summary of viral infection rodent models with cerebellar and behavioral abnormalities that parallel those observed in autism.

Rodent model Behavioral abnormalities Cerebellar abnormalities Prefrontal cortex abnormalities

Neonatal Borna disease virus Hyperactivity Cerebellar hypoplasia High levels of viral nucleic acid in PFC infection rats Stereotypic behavior Decreased Purkinje cell number Elevated levels of DOPAC Deficits in social behavior

Maternal influenza infection Reduced exploratory behavior Atrophy of cerebellum Gene expression changes in PFC mice Social impairments Gene expression changes in Working memory deficit cerebellum Impaired emotional behavior Decreased Purkinje cell number Heterotopic Purkinje cells Delayed migration of granule cells

Frontiers in Systems Neuroscience www.frontiersin.org May 2013 | Volume 7 | Article 15 | 9 Rogers et al. Autism and the cerebellum

Table 3 | Summary of toxic exposure rodent model with cerebellar and behavioral abnormalities that parallel those observed in autism.

Rodent model Behavioral abnormalities Cerebellar abnormalities Prefrontal cortex abnormalities

Prenatal valproic acid exposure Deficits in olfactory discrimination Apoptotic cells in cerebellum Aberrant dopamine activity in PFC rats Decreased prepulse inhibition Decreased Purkinje cell number Hyperactivity Increased stereotypy Decreased social play Decreased exploratory behavior to the drug during pregnancy (Arndt et al., 2005). In humans, this question is “No,” simply because the heterogeneity of symp- the behavior resulting from prenatal exposure includes delayed toms as well as the range of symptom severity from savant skills speech and motor development, hyperactivity, poor social and to profound mental and physical handicaps makes it unlikely that communication skills, and deficits in cognitive functions such as there is a unitary cause for these disorders. attention (Moore et al., 2000). Likewise, mice and rats prenatally A third question that also should be addressed is, can glu- exposed to VPA exhibit abnormal behaviors including deficits in tamatergic projections to the PFC be affected independently olfactory discrimination, decreased prepulse inhibition, hyperac- of developmental cerebellar damage? Pre-clinical research has tivity, increased stereotypy, decreased social play, and decreased demonstrated that two glutamatergic pathways from the cerebel- exploratory behavior, as compared to controls (Schneider and lum to the PFC can modulate dopamine release in the mPFC. This Przewlocki, 2005; Roullet et al., 2010). Rats prenatally exposed is significant because impairments in attention, memory, and a to VPA also exhibit cerebellar abnormalities. Yochum et al. (2008) marked tendency to perseverate in the face of changing contin- found that prenatal exposure to VPA in rats induced an increase in gencies (i.e., executive function) are well documented in patients apoptotic cells in the cerebellum, while Ingram et al. (2000) found with ASD, and are well known to be dependent on on-going that rats prenatally exposed to a single VPA intra-peritoneal injec- dopaminergic activity in the mPFC (Seamans and Yang, 2004). tion had approximately 30% fewer Purkinje cells and significantly Research in lurcher and FMR1 mutant mice has shown that the differed from control animals. Consistent with the disconnection ability of the cerebellum to modulate mPFC dopamine release is hypothesis, VPA seems to have an effect on dopamine tissue lev- significantly affected by developmental cerebellar damage involv- els in adult rodents following acute or chronic exposure (Löscher, ing Purkinje cells, and in Lurchers, that cerebellar Purkinje cell 1999). Adult rats that received chronic i.p. injections of VPA had number is significantly correlated with autism-like deficits in increased dopamine tissue levels in multiple brain areas, includ- executive function, working memory and repetitive behavior. ingfrontalcortex,asmeasuredbyHPLC(Baf et al., 1994). Also, Thus it is possible that cerebellar neuropathology which results an in vivo microdialysis study found that an acute subcutaneous in changes in cerebellar output and subsequent effects on down- injection of VPA produced increased extracellular dopamine lev- stream circuitry might underlie some of the cognitive symptoms els in the mPFC of rats (Ichikawa and Meltzer, 1999). of autism. In terms of this question we currently believe that, depending SUMMARY AND CONCLUSIONS on the severity of developmental cerebellar neuropathology, it is In attempting to answer the question, “Is autism a disease of likely that glutamatergic projections to the PFC will be effected the cerebellum,” it is apparent from clinical research that devel- and this will result in autistic symptomatology. This may seem opmental damage to the cerebellum is a common occurrence surprising given that it has been effectively argued that in a mouse in ASD. Pre-clinical research confirms clinical observations that model of Rett syndrome, mutation of the MECP2 gene results there are multiple routes to developmental cerebellar damage in the appearance of autism-like symptoms including increased including genetic mutations, viral exposure, or exposure to envi- repetitive behaviors which can be linked to dysfunctions in GABA ronmental toxins. Rodent models of autism both confirm often signaling (Chao et al., 2010). It should be noted, however, that reported clinical findings and also extend these findings by pro- paralleling the time course in the development of symptoms, viding candidate pathways deserving of further research, although mutations involving the MECP2 locus also cause progressive both types of research are potentially limited by questions about degenerative changes in the cerebellum (Murakami et al., 1992; the selectivity of cerebellar abnormalities as opposed to the rest Reardon et al., 2010), as well as altering glutamate homeostasis of the brain. It is however intriguing to note that developmental (Zhong et al., 2012). damage to the cerebellum is frequently associated with a diagno- In support of the notion that cerebellar pathology and glu- sis of ASD in humans, and in rodents cerebellar neuropathology tamatergic dysregulation co-occur, in a mouse model of frag- is associated with many autism-like cognitive deficits. Thus, a par- ile X, neuroligin-3 KO mice were found to exhibit deficits in simonious answer to this initial question is, “Yes,” in many cases social behaviors analogous to those seen in autism (Baudouin it is likely that autism is associated with developmental cerebellar et al., 2012). These mice showed cerebellar pathology involving damage. increases in the metabotropic glutamate receptor mGLUR1α Is developmental cerebellar damage the only route to the which interfered with the structure and function of circuitry in symptoms of autism? We believe that the most likely answer to the cerebellum. Reactivation of the production of neuroligin-3

Frontiers in Systems Neuroscience www.frontiersin.org May2013|Volume7|Article15| 10 Rogers et al. Autism and the cerebellum

ameliorated behavioral deficits in these mice and reversed the While it seems clear that cerebellar modulation of mPFC cerebellar structural deficits that were previously observed. dopamine release is dependent on glutamatergic pathways, there Our goal in the present manuscript was to provide a galvaniz- remains a significant gap in our knowledge of the role of ing theory of what may be a major neurophysiological mechanism dopaminergic and glutamatergic systems in autism. Several that could account for the cognitive, motor and reward symp- molecular biological techniques are available for investigating toms of autism, as well as the apparent variability in symptom this in rodents, including the use of viral vectors to selectively severity. Thus, acceptance of the idea that developmental dam- transduce, or introduce new genetic material into glutamatergic age to the cerebellum is a key player in the etiology of autism neurons. This might provide a way to understand the behavioral and ASDs may have explanatory power in realms other than consequences and ameliorate behavioral impairments associated cognitive symptomatology. As noted previously, the diagnosis with altered glutamatergic transmission within this circuitry. of autism is frequently associated with co-occurring deficits in Future research should also aim to determine when during the motor function (Gowen and Hamilton, 2013). Given the rela- developmental time course cerebellar lesions occur as well as tionship between cerebellar Purkinje cell loss and the occurrence the anatomical specificity of the cerebellar deficits observed in of ataxia as well as the cerebellar connections to premotor and the reviewed rodent models. Understanding the developmental motor cortical regions (Strick et al., 2009; Fatemi et al., 2012)it timing and the anatomical specificity of cerebellar pathology is seems reasonable to suggest that variability in the degree of cere- important for understanding the causal mechanisms underlying bellar damage is likely related to the severity of motor deficits in this consistent feature of autism. Additionally, given the apparent ASD individuals. glutamatergic dysregulation in the cerebello-ventral tegmental- With regard to underlying mechanisms of the social reward cortical pathway that occurs along with developmental cerebellar impairment in autism, preclinical research links dysregulation of damagewebelievethescopeofresearchshouldbeexpandedto meso-accumbens dopamine to deficits in social engagement and include investigations of other brain regions that receive effer- social bond formation (Dichter et al., 2012c). As there is now ents from nuclei in this circuit, and especially the PPT. It is well clear evidence that developmental cerebellar damage is consis- known that projections from the PPT to the substantia nigra tently associated with dysregulation of mesocortical dopamine modulate dopamine transmission in the dorsal striatum (Forster projections arising from the VTA in Lurcher and FMR1 mice and Blaha, 2003). Thus, it would be expected that developmental (Rogers et al., 2013), it is highly likely that the meso-accumbens cerebellar damage is also associated with dysregulation in striatal dopaminergic projection, also arising from the VTA, may also dopamine transmission. Ultimately, a better understanding of the show deficits in neurotransmission. As the meso-accumbens sys- direct effects of cerebellar pathology in autism could lead to better tem has been heavily implicated in reward and reinforcement diagnostic measures, earlier diagnosis, and novel pharmacological (e.g., Salamone and Correa, 2012) it is reasonable to suggest targets for the treatment of autism and ASD. that a putative dysfunction in meso-accumbens dopamine, that is driven by developmental cerebellar neuropathology, could ACKNOWLEDGMENTS account for the deficits in reward responsivity displayed by ASD Contract grant sponsor: NINDS; Contract grant number: R01 patients. NS063009.

REFERENCES Amir,R.E.,VandenVeyver,I.B., disorders. J. Autism Dev. Disord. 40, A clinicopathological study of Abou-Donia, M. B., Khan, W. A., Wan, M., Tran, C. Q., Francke, U., 1423–1430. autism. Brain 121, 889–905. Dechkovskaia, A. M., Goldstein, L. and Zoghbi, H. Y. (1999). Rett syn- Aydin, M., Kislal, F. M., Ayar, A., Baillieux, H., De Smet, H. J., Dobbeleir, B., Bullman, S. L., and Abdel- drome is caused by mutations in Demirol, M., Kabakus, N., Canatan, A.,Paquier,P.F.,DeDeyn,P.P., Rahman, A. (2006). In utero expo- X-linked MECP2, encoding methyl- H., et al. (2011). The effects and Mariën, P. (2010). Cognitive sure to nicotine and chlorpyrifos CpG-binding protein 2. Nat. Genet. of lipopolysaccharide-induced and affective disturbances following alone, and in combination produces 23, 185–188. endogenous hyperthermia and focal cerebellar damage in adults: persistent sensorimotor deficits and Arndt, T., Guessregen, B., Hohl, A., different antipyretic treatment a neuropsychological and SPECT Purkinje neuron loss in the cerebel- and Reis, J. (2005). Determination modalities on rat brain. Bratisl. Lek. study. Cortex 46, 869–879. lum of adult offspring rats. Arch. of serum amantadine by liquid Listy 112, 227–234. Bandim, J. M., Ventura, L. O., Miller, Toxicol. 80, 620–631. chromatography-tandem mass Baf,M.H.,Subhash,M.N., M.T.,Almeida,H.C.,andCosta, Allen, G., Müller, R. A., and spectrometry. Clin. Chim. Acta 359, Lakshmana, K. M., and Rao, A. E. S. (2003). Autism and Courchesne, E. (2004). Cerebellar 125–131. B. S. (1994). Sodium valproate Möbius sequence: an exploratory function in autism: functional mag- Asp, L., Beraki, S., Kristensson, K., induced alterations in monoamine study of children in northeastern netic resonance image activation Ogren, S. O., and Karlsson, H. levels in different regions of the Brazil. Arq. Neuropsiquiatr. 61, during a simple motor task. Biol. (2009). Neonatal infection with rat brain. Neurochem. Int. 24, 181–185. Psychiatry 56, 269–278. neurotropic influenza A virus affects 67–72. Bangash, M. A., Park, J. M., Melnikova, Amaral, D. G., Schumann, C. M., working memory and expression of Bailey, A., Le Couteur, A., Gottesman, T., Wang, D., Jeon, S. K., Lee, D., and Nordahl, C. W. (2008). type III Nrg1 in adult mice. Brain I., Bolton, P., Simonoff, E., Yuzda, et al. (2011). Enhanced polyubiq- Neuroanatomy of autism. Trends Behav. Immun. 23, 733–741. E., et al. (1995). Autism as a strongly uitination of Shank3 and NMDA Neurosci. 31, 137–145. Atladóttir, H. O., Thorsen, P., genetic disorder: evidence from a receptor in a mouse model of American Psychiatric Association. Østergaard, L., Schendel, D. British twin study. Psychol. Med. 25, autism. Cell 145, 758–772. (2000). Diagnostic and Statistical E., Lemcke, S., Abdallah, M., 63–77. Baudouin, S. J., Gaudias, J., Gerharz, S., Manual of Mental Disorders-IV-TR. et al. (2010). Maternal infection Bailey, A., Luthert, P., Dean, Hatstatt, L., Zhou, K., Punnakkal, 4th Edn. Washington, DC: requiring hospitalization during A., Harding, B., Janota, I., P., et al. (2012). Shared synap- American Psychiatric Association. pregnancy and autism spectrum Montgomery, M., et al. (1998). tic pathophysiology in syndromic

Frontiers in Systems Neuroscience www.frontiersin.org May2013|Volume7|Article15| 11 Rogers et al. Autism and the cerebellum

and nonsyndromic rodent models Bonaglia, M. C., Giorda, R., Mani, Careaga, M., Van de Water, J., and design. J. Am. Acad. Child Psychiatry of autism. Science 338, 128–132. E., Aceti, G., Anderlid, B. M., Ashwood, P. (2010). Immune 19, 665–677. Bauman, M. L. (1991). Microscopic Baroncini, A., et al. (2006). dysfunction in autism: a pathway Cole, T. B., Beyer, R. P., Bammler, neuroanatomic abnormalities in Identification of a recurrent to treatment. Neurotherapeutics 7, T.K.,Park,S.S.,Farin,F.M., autism. Pediatrics 87, 791–796. breakpoint within the Shank3 gene 283–292. Costa, L. G., et al. (2011). Repeated Bauman,M.L.,andKemper,T.L. in the 22p13.3 deletion syndrome. Carper, R. A., and Courchesne, E. developmental exposure of mice to (2005). Neuroanatomic observa- J. Med. Genet. 43, 822–828. (2000). Inverse correlation between chlorpyrifos oxon is associated with tions of the brain in autism: a Botez-Marquard, T., and Botez, M. I. frontal lobe and cerebellum sizes paraoxonase 1 (PON1)-modulated review and future directions. Int. J. (1993). Cognitive behavior in here- in children with autism. Brain 123, effects on cerebellar gene expres- Dev. Neurosci. 23, 183–187. dodegenerative ataxias. Eur. Neurol. 836–844. sion. Toxicol. Sci. 123, 155–169. Bautista, J. R., Rubin, S. A., Moran, T. 33, 351–357. Carper, R. A., and Courchesne, E. Courchesne, E. (1997). Brainstem, cere- H.,Schwartz,G.J.,andCarbone, Boukhtouche, F., Brugg, B., Wehrlé, (2005). Localized enlargement of bellar and limbic neuroanatomical K. M. (1995). Developmental injury R., Bois-Joyeux, B., Danan, J. L., the frontal lobe in autism. Biol. abnormalities in autism. Curr. Opin. to the cerebellum following perinata Dusart, I., et al. (2010). Induction Psychiatry 57, 126–133. Neurobiol. 7, 269–278. Borna disease virus infection. Brain of early Purkinje cell dendritic dif- Cascio, C. J., Foss-Feig, J. H., Heacock, Courchesne, E., Townsend, J., Res. Dev. Brain Res. 90, 45–53. ferentiation by thyroid hormone J. L., Newsom, C. R., Cowan, R. L., Akshoomoff, N. A., Saitoh, O., Belzung, C., Chapillon, P., and Lalonde, requires RORα. Neural Dev. 5:18. Benningfield, M. M., et al. (2012). Yeung-Courchesne, R., Lincoln, R. (2001). The effects of the lurcher doi: 10.1186/1749-8104-5-18 Response of neural reward regions A. J., et al. (1994). Impairment in mutation on object localization, Boukhtouche, F., Janmaat, S., Vodjdani, to food cues in autism spectrum dis- shifting attention in autistic and T-maze discrimination, and radial G., Gautheron, V., Mallet, J., Dusart, orders. J. Neurodev. Disord. 4:9. doi: cerebellar patients. Behav. Neurosci. arm maze tasks. Behav. Genet. 31, I., et al. (2006). Retinoid-related 10.1186/1866-1955-4-9 108, 848–865. 151–155. orphan receptor alpha controls the Castelli, F., Frith, C., Happe, F., Courchesne, E., Yeung-Courchesne, R., Beraki, S., Aronsson, F., Karlsson, H., early steps of purkinje cell den- and Frith, U. (2002). Autism, Press, G. A., Hesselink, J. R., and Ogren, S. O., and Kristensson, K. dritic differentiation. J. Neurosci. 26, Asperberger’s syndrome and brain Jernigan, T. L. (1988). Hypoplasia (2005). Influenza A virus infec- 1531–1538. mechanisms for the attribution of of cerebellar vermal lobules VI and tion causes alterations in expression Bozdagi, O., Sakurai, T., Papapetrou, mental states to animated shapes. VII in autism. N. Engl. J. Med. 318, of synaptic regulatory genes com- D., Wang, X., Dickstein, D. L., Brain 125, 1839–1849. 1349–1354. bined with changes in cognitive and Takahashi, N., et al. (2010). Catani, M., and Ffytche, D. H. (2008). Dawson, G., Meltzoff, A. N., Osterling, emotional behaviors in mice. Mol. Haploinsufficiency of the autism- The rises and falls of disconnection J., Rinaldi, J., and Brown, E. (1998). Psychiatry 10, 299–308. associated Shank3 gene leads to syndromes. Brain 128, 2224–2239. Children with autism fail to ori- Beri, S., Tonna, N., Menozzi, G., deficits in synaptic function, social Catani, M., Jones, D. K., Daly, E., ent to naturally occurring social Bonaglia, M. C., Sala, C., and interaction, and social commu- Embiricos, N., Deeley, Q., Pugliese, stimuli. J. Autism Dev. Disord. 28, Giorda, R. (2007). DNA methyla- nication. Mol. Autism 1:15. doi: L., et al. (2008). Altered cere- 479–485. tion regulates tissue-specific expres- 10.1186/2040-2392-1-15 bellar feedback projections in Dean, S. L., Wright, C. L., Hoffman, sion of Shank3. J. Neurochem. 101, Bracke-Tolkmitt, R., Linden, A., Asperger syndrome. Neuroimage 41, J. F., Wang, M., Alger, B. E., 1380–1391. Canavan, A. G. M., Rockstroh, B., 1148–1191. and McCarthy, M. M. (2012). Bernardet, M., and Crusio, W. E. Scholz, E., Wessel, K., et al. (1989). Centers for Disease Control and Prostaglandin E2 stimulates estra- (2006). Fmr1 KO mice as a The cerebellum contributes to Prevention. (2007). Prevalence of diol synthesis in the cerebellum possible model of autistic fea- mental skills. Behav. Neurosci. 103, autism spectrum disorders—autism postnatally with associated effects tures. ScientificWorldJournal 6, 442–446. and developmental disabilities on Purkinje neuron dendritic arbor 1164–1176. Brian, J., Bryson, S. E., Garon, N., monitoring network, 14 sites, and electrophysiological properties. Betancur, C., Sakurai, T., and Roberts, W., Smith, I. M., Szatmari, United States, 2002. MMWR Endocrinology 153, 5415–5427. Buxbaum, J. D. (2009). The emerg- P., et al. (2008). Clinical assessment Surveill. Summ. 56, 12–28. deCock,M.,Maas,Y.G.,andvan ing role of synaptic cell-adhesion of autism in high-risk 18-month- Chao, H. T., Chen, H., Samaco, R. de Bor, M. (2012). Does perina- pathways in the pathogenesis of olds. Autism 12, 433–456. C.,Xue,M.,Chahrour,M.,Yoo, tal exposure to endocrine disruptors autism spectrum disorders. Trends Caddy, K. W., and Biscoe, T. J. (1979). J., et al. (2010). Dysfunction induce autism spectrum and atten- Neurosci. 32, 402–412. Structural and quantitative studies in GABA signalling mediates tion deficit hyperactivity disorders? Boeckers, T. M., Bockmann, J., Kreutz, on the normal C3H and Lurcher autism-like stereotypies and Rett Review. Acta Paediatr. 101, 811–818. M. R., and Gundelfinger, E. D. mutant mouse. Philos. Trans. syndrome phenotypes. Nature 468, Del Arco, A., and Mora, F. (2005). (2002). ProSAP/Shank proteins – a R.Soc.Lond.BBiol.Sci.287, 263–269. Glutamate-dopamine in vivo inter- family of higher order organizing 167–201. Cheh, M. A., Millonig, J. H., Roselli, L. action in the prefrontal cortex mod- molecules of the postsynaptic den- Campbell, D. B., Sutcliffe, J. S., Ebert, M., Ming, X., Jacobsen, E., Kamdar, ulates the release of dopamine and sity with an emerging role in human P. J., Militerni, R., Bravaccio, C., S., et al. (2006). En2 knockout acetylcholine in the nucleus accum- neurological disease. J. Neurochem. Trillo, S., et al. (2006). A genetic mice display neurobehavioral and bens of the awake rat. J. Neural 81, 903–910. variant that disrupts MET tran- neurochemical alterations relevant Transm. 112, 97–109. Bolduc, M., Du Plessis, A. J., Sullivan, scription is associated with autism. to autism spectrum disorder. Brain Devlin, B., and Scherer, S. W. (2012). N., Khwaja, O. S., Zhang, X., Barnes, Proc. Natl. Acad. Sci. U.S.A. 103, Res. 1116, 166–176. Genetic architecture in autism spec- K., et al. (2011). Spectrum of 166834–166839. Chen, B. Y., Zou, X. B., Zhang, J., Deng, trum disorders. Curr. Opin. Genet. neurodevelopmental disabilities in Campbell, M., Anderson, L. T., Small, H. Z., Li, J. Y., Li, L. Y., et al. (2011). Dev. 22, 229–237. children with cerebellar malforma- A. M., Perry, R., Green, W. H., and Copy-number variations of Shank3 Dichter, G., and Adolphs, R. (2012). tions. Dev. Med. Child Neurol. 53, Caplan, R. (1982). The effects of and related clinical phenotypes in Reward processing in autism: a the- 409–416. haloperidol on learning and behav- children with autism. Zhonghua Er matic series. J. Neurodev. Disord. Bolduc, M. E., Du Plessis, A. J., ior in autistic children. J. Autism Ke Za Zhi 49, 607–611. 4:20. doi: 10.1186/1866-1955-4-20 Sullivan, N., Guizard, N., Zhang, Dev. Disord. 12, 167–175. Cohen, I. L., Campbell, M., Posner, Dichter, G. S., Felder, J. N., Green, S. X., Robertson, R. L., et al. (2012). Canitano, R. (2006). Self injuri- D.,Small,A.M.,Triebel,D.,and R.,Rittenberg,A.M.,Sasson,N.J., Regional cerebellar volumes predict ous behavior in autism: clinical Anderson, L. T. (1980). Behavioral and Bodfish, J. W. (2012a). Reward functional outcome in children aspects and treatment with risperi- effects of haloperidol in young circuitry function in autism spec- with cerebellar malformations. done. J. Neural Transm. 113, autistic children. An objective anal- trum disorders. Soc. Cogn. Affect. Cerebellum 11, 531–542. 425–431. ysis using a within-subjects reversal Neurosci. 7, 160–172.

Frontiers in Systems Neuroscience www.frontiersin.org May2013|Volume7|Article15| 12 Rogers et al. Autism and the cerebellum

Dichter, G. S., Richey, J. A., Rittenberg, Ellegood, J., Pacey, L. K., Hampson, efflux by activation of acetylcholine in patients with single subcorti- A. M., Sabatino, A., and Bodfish, J. D. R., Lerch, J. P., and Henkelman, and glutamate receptors in the cal lesion stroke of four different W. (2012b). Reward circuitry func- R. M. (2010). Anatomical pheno- midbrain and pons of the rat. Eur. areas. Zhonghua Yi Xue Za Zhi 91, tion in autism during face anticipa- typing in a mouse model of frag- J. Neurosci. 17, 751–762. 1904–1908. tion and outcomes. J. Autism Dev. ile X syndrome with magnetic res- Fournier, K. A., Hass, C. J., Naik, S. Goodrich-Hunsaker, N. J., Wong, L. M., Disord. 42, 147–160. onance imaging. Neuroimage 53, K., Lodha, N., and Cauraugh, J. McLennan, Y., Tassone, F., Harvey, Dichter,G.S.,Damiano,C.A., 1023–1029. H. (2010). Motor coordination in D., Rivera, S. M., et al. (2011). Adult and Allen, J. A. (2012c). Reward Elsabbagh, M., Mercure, E., Hudry, K., autism spectrum disorders: a syn- female fragile X permutation car- circuitry dysfunction in psychi- Chandler, S., Pasco, G., Charman, thesis and meta-analysis. J. Autism riers exhibit age- and CGG repeat atric and neurodevelopmental T., et al. (2012). Infant neu- Dev. Disord. 40, 1227–1240. length-related impairments on an disorders and genetic syndromes: ral sensitivity to dynamic eye Freitag, C. M., Kleser, C., Schneider, attentionally based enumeration animal models and clinical findings. gaze is associated with later M., and von Gontard, A. (2007). task. Front. Hum. Neurosci. 5:63. J. Neurodev. Disord. 4, 1–43. emerging autism. Curr. Biol. 22, Quantitative assessment of neu- doi: 10.3389/fnhum.2011.00063 DiCicco-Bloom, E., Lord, C., 338–342. romotor function in adolescents Gowen, E., and Hamilton, A. (2013). Zwaigenbaum, L., Courchesne, Ernst, M., Zametkin, A. J., Matochik, with high functioning autism and Motor abilities in autism: a review E.,Dager,S.R.,Schmitz,C.,etal. J. A., Pascualvaca, D., and Cohen, Asperger Syndrome. J. Autism Dev. using a computational context. (2006). The developmental neu- R. M. (1997). Low medial prefrontal Disord. 37, 948–959. J. Autism Dev. Disord. 43, 323–344. robiology of autism spectrum dopaminergic activity in autistic Garcia-Rill, E., Skinner, R. D., Greco, C. M., Navarro, C. S., Hunsaker, disorder. J. Neurosci. 26, 6897–6906. children. Lancet 350, 638. Miyazato,H.,andHomma,Y. M. R., Maezawa, I., Shuler, J. Dickson, P. E., Rogers, T. D., Del Mar, Estes, A., Shaw, D. W., Sparks, B. F., (2001). Pedunculopontine stimula- F., Tassone, F., et al. (2011). N., Martin, L. A., Heck, D., Blaha, Friedman, S., Giedd, J. N., Dawson, tion induces prolonged activation Neuropathologic features in the C. D., et al. (2010). Behavioral flex- G., et al. (2011). Basal ganglia of pontine reticular neurons. hippocampus and cerebellum of ibility in a mouse model of devel- morphometry and repetitive behav- Neuroscience 104, 455–465. three older men with fragile X opmental cerebellar Purkinje cell ior in young children with autism Geschwind, N. (1965). Disconnexion syndrome. Mol. Autism 2:2. doi: loss. Neurobiol. Learn. Mem. 94, spectrum disorder. Autism Res. 4, syndromes in animals and man. 10.1186/2040-2392-2-2 220–228. 212–220. Brain 88, 237–294. Green, D., Baird, G., Barnett, A. Dimitropoulos, A., and Schultz, R. Ey, E., Leblond, C. S., and Bourgeron, T. Geshwind, D. H., and Levitt, P. L., Henderson, L., Huber, J., and T. (2007). Autistic-like symptoma- (2011). Behavioral profiles of mouse (2007). Autism spectrum disorders: Henderson, S. E. (2002). The sever- tology in Prader-Willi syndrome: models for autism spectrum disor- developmental disconnection syn- ity and nature of motor impairment a review of recent findings. Curr. ders. Autism Res. 4, 5–16. dromes. Curr. Opin. Neurobiol. 17, in Asperger’s syndrome: a compari- Psychiatry Rep. 9, 159–164. Fallon, J. H., and Moore, R. Y. (1978). 103–111. son with specific developmental dis- Dollfus, S., Petit, M., Menard, J. F., Catecholamine innervation of the Ghaziuddin, M., Butler, E., Tsai, L., order of motor function. J. Child and Lesieur, P. (1992). Amisulpride basal forebrain. IV. Topography of and Ghaziuddin, N. (1994). Is clum- Psychol. Psychiatry 43, 655–668. versus bromocriptine in infantile the dopamine projection to the siness a marker for Asperger syn- Hallahan, B., Daly, E. M., McAlonan, autism: a controlled crossover com- basal forebrain and neostriatum. drome? J. Intellect. Disabil. Res. G., Loth, E., Toal, F., O’Brien, F., parative study of two drugs with J. Comp. Neurol. 180, 545–580. 38(Pt 5), 519–527. et al. (2009). Brain morphometry opposite effects on dopaminergic Fatemi, S. H., Aldinger, K. A., Ashwood, Gidley Larson, J. C., Bastian, A. J., volume in autistic spectrum disor- function. J. Autism Dev. Disord. 22, P.,Bauman,M.L.,Blaha,C.D., Donchin, O., Shadmehr, R., and der: a magnetic resonance imaging 47–60. Blatt, G. J., et al. (2012). Consensus Mostofsky, S. H. (2008). Acquisition study of adults. Psychol. Med. 39, Doulazmi,M.,Frédéric,F.,Capone, paper: pathological role of the cere- of internal models of motor tasks 337–346. F., Becher-André, M., Delhaye- bellum in autism. Cerebellum 11, in children with autism. Brain 131, Hallmayer, J., Cleveland, S., Torres, A., Bouchaud, N., and Mariani, J. 777–807. 2894–2903. Phillips, J., Cohen, B., Torigoe T., (2001). A comparative study of Fatemi, S. H., Araghi-Niknam, N., Gilbert, S. J., Meuwese, J. D. I., et al. (2011). Genetic heritability Purkinje cells in two RORalpha Laurence, J. A., Stary, J. M., Towgood, K. J., Frith, C. D., and and shared environmental factors gene mutant mice: stagger and Sidwell, R. W., and Lee, S. (2004). Burgess, P. W. (2009). Abnormal among twin pairs with autism. Arch. RORalpha(-/-). Brain Res. Dev. Glial fibrillary acidic protein functional specialization within Gen. Psychiatry 68, 1095–1102. Brain Res. 127, 165–174. and glutamic acid decarboxy- medial prefrontal cortex in high- Hamilton, B. A., Frankel, W. N., Dowell, L. R., Mahone, E. M., lase 65 and 67 kDa proteins are functioning autism: a multi-voxel Kerrebrock, A. W., Hawkins, T. L., and Mostofsky, S. H. (2009). increased in brains of neonatal similarity analysis. Brain 132, Fitzhugh, W., Kusumi, K., et al. Associations of postural knowl- BALB/c mice following viral infec- 869–878. (1996). Disruption of the nuclear edge and basic motor skill with tion in utero. Schizophr. Res. 69, Goldowitz, D., and Koch, J. (1986). hormone receptor RORalpha dyspraxia in autism: implication 121–123. Performance of normal and neu- in staggerer mice. Nature 379, for abnormalities in distributed Fatemi, S. H., Folsom, T. D., Reutiman, rological mutant mice on radial 736–739. connectivity and motor learning. T. J., Abu-Odeh, D., Mori, S., arm maze and active avoidance Hardan,A.Y.,Girgus,R.R.,Adams, Neuropsychology 23, 563–570. Huang, H., et al. (2009). Abnormal tasks. Behav. Neural. Biol. 46, J.,Gilbert,A.R.,Keshaven,M. Durand, C. M., Betancur, C., Boeckers, expression of myelination genes and 216–226. S., and Minshew, N. J. (2006). T.M.,Bockmann,J.,Chaste, alterations in white matter frac- Goldowitz, D., Moran, H., and Wetts, Abnormal brain size effect on the P., Fauchereau, F., et al. (2007). tional anisotropy following pre- R. (1992). “Mouse chimeras in thalamus in autism. Psychiatry Res. Mutations in the gene encoding the natal viral influenza infection at the study of genetic and struc- 147, 145–151. synaptic scaffolding protein Shank3 E16 in mice. Schizophr. Res. 112, tural determinants of behavior,” in Haswell, C. C., Izawa, J., Dowell, L. are associated autism spectrum 46–53. Techniques for the Genetic Analysis R.,Mostofsky,S.H.,andShadmehr, disorders. Nat. Genet. 39, 25–27. Fatemi,S.H.,Reutiman,T.J.,Folsom, of Brain and Behavior: Focus on R. (2009). Representation of inter- Eisenman, L. M., Brothers, R., Tran, T. D., and Sidwell, R. W. (2008). The the Mouse,edsD.Goldowitz, nal models of action in the autistic M.H.,Kean,R.B.,Dickson,G. role of cerebellar genes in pathol- D. Wahlsten, and R. E. Wimer brain. Nat. Neurosci. 12, 970–972. M., Dietschold, B., et al. (1999). ogy of autism and schizophrenia. (Amsterdam: Elsevier), 271–290. Herrup, K., and Mullen, R. J. (1979). Neonatal Borna disease virus infec- Cerebellum 7, 279–294. Gong,W.P.,Ding,M.P.,Guo,Q. Staggerer chimeras: intrinsic nature tion in the rat causes a loss of Forster, G. L., and Blaha, C. D. (2003). H.,Qiu,L.Q.,Huang,S.Y.,and of Purkinje cell defects and implica- Purkinje cells in the cerebellum. Pedunculopontine tegmental stim- Zhou, X. X. (2011). A compar- tions for normal cerebellar develop- J. Neurovirol. 5, 181–189. ulation evokes striatal dopamine ison study on cognitive function ment. Brain Res. 178, 443–457.

Frontiers in Systems Neuroscience www.frontiersin.org May 2013 | Volume 7 | Article 15 | 13 Rogers et al. Autism and the cerebellum

Herrup, K., Shojaeian-Zanjani, H., Klin,A.,Lin,D.J.,Gorrindo,P., Libbey, J. E., Sweeten, T. L., McMahon, Misslin, R., Cigrang, M., and Panzini, L., Sunter, K., and Mariani, Ramsay, G., and Jones, W. (2009). W. M., and Fujinami, R. S. (2005). Guastavino, J. (1986). Responses to J. (1996). The numerical matching Two-year-olds with autism orient Autistic disorder and viral infec- novelty in staggerer mutant mice. of source and target populations to non-social contingencies rather tions. J. Neurovirol. 11, 1–10. Behav. Process 12, 51–56. in the CNS: the inferior olive to than biological motion. Nature 459, Lin, A., Rangel, A., and Adolphs, R. Mittleman, G., Goldowitz, D., Heck, Purkinje cell projection. Brain Res. 257–261. (2012). Impaired learning of social D. H., and Blaha, C. D. (2008). Dev. Brain Res. 96, 28–35. Koekkoek, S. K., Yamaguchi, K., compared to monetary rewards in Cerebellar modulation of frontal Hornig, M., Weissenböck, H., Milojkovic,B.A.,Dortland, autism. Front. Neurosci. 6:143. doi: cortex dopamine efflux in mice: rel- Horscroft, N., and Lipkin, W. B.R.,Ruigrok,T.J.,Maex,R., 10.3389/fnins.2012.00143 evance to autism and schizophrenia. I. (1999). An infection-based model et al. (2005). Deletion of Fmr1 in Löscher, W. (1999). Valproate: a reap- Synapse 62, 544–550. of neurodevelopmental damage. Purkinje cells enhances parallel fiber praisal of its pharmacodynamic Miyahara, M., Tsujii, M., Hori, M., Proc. Natl. Acad. Sci. U.S.A. 96, LTD, enlarges spines, and attenuates properties and mechanisms of Nakanishi, K., Kageyama, H., and 12102–12107. cerebellar eyelid conditioning in action. Prog. Neurobiol. 58, 31–59. Sugiyama, T. (1997). Brief report: Ichikawa, J., and Meltzer, H. Y. (1999). fragile X syndrome. Neuron 47, Luna, B., Minshew, N. J., Garver, K. E., motor incoordination in children Valproate and carbamazepine 339–352. Lazar, N. A., Thulborn, K. R., Eddy, with Asperger syndrome and increase prefrontal dopamine Krishnan, K., Mitra, N. K., Yee, L. S., W. F., et al. (2002). Neocortical learning disabilities. J. Autism Dev. release by 5-HT1A receptor acti- and Yang, H. M. (2012). A com- system abnormalities in autism: an Disord. 27, 595–603. vation. Eur. J. Pharmacol. 380, parison of neurotoxicity in cerebel- fMRI study of spatial working mem- Mizuno, A., Villalobos, M. E., Davies, R1–R3. lum produced by dermal applica- ory. Neurology 59, 834–840. M. M., Dahl, B. C., and Müller, R. Ieraci, A., Forni, P. E., and Ponzetto, tion of chlorpyrifos in young and Manjiviona, J., and Prior, M. (1995). (2006). Partially enhanced thalam- C. (2002). Viable hypomorphic sig- adult mice. J. Neural Transm. 119, Comparison of Asperger syndrome ocortical functional connectivity in naling mutant of the Met receptor 345–352. and high-functioning autistic chil- autism. Brain Res. 1104, 160–174. reveals a role for hepatocyte growth Kuemerle, B., Gulden, F., Cherosky, N., dren on a test of motor impair- Moore, S. J., Turnpenny, P., Quinn, A., factor in postnatal cerebellar devel- Williams, E., and Herrup, K. (2007). ment. J. Autism Dev. Disord. 25, Glover, S., Lloyd, D. J., Montgomery, opment. Proc. Natl. Acad. Sci. U.S.A. The mouse engrailed genes: a win- 23–39. T., et al. (2000). A clinical study of 99, 15200–15205. dow into autism. Behav. Brain Res. Maroni, P., Bendinelli, P., Tiberio, 57 children with fetal anticonvul- Ingram, J. L., Peckham, S. M., Tisdale, 176, 121–132. L., Rovetta, F., Piccoletti, R., and sant syndromes. J. Med. Genet. 37, B., and Rodier, P. M. (2000). Kuemerle, B., Zanjani, H., Joyner, A., Schiaffonati, L. (2003). In vivo heat- 489–497. Prenatal exposure of rats to valproic and Herrup, K. (1997). Pattern shock response in the brain: sig- Mostofsky, S. H., Bunoski, R., Morton, acid reproduces the cerebellar deformities and cell loss in nalling pathway and transcription S. M., Goldberg, M. C., and Bastian, anomalies associated with autism. Engrailed-2 mutant mice sug- factor activation. Brain Res. Mol. A. J. (2004). Children with autism Neurotoxicol. Teratol. 22, 319–324. gest two separate patterning events Brain Res. 119, 90–99. adapt normally during a catch- Jansiewicz,E.M.,Goldberg,M.C., during cerebellar development. Martin,L.A.,Goldowitz,D.,and ing task requiring the cerebellum. Newschaffer, C. J., Denckla, M. J. Neurosci. 17, 7881–7889. Mittleman, G. (2010). Repetitive Neurocase 10, 60–64. B., Landa, R., and Mostofsky, S. Kumar, A., Sundaram, S. K., behavior and increased activity in Muhle, R., Trentacoste, S. V., and H. (2006). Motor signs distin- Sivaswamy, L., Behen, M. E., mice with Purkinje cell loss: a model Rapin, I. (2004). The genetics of guish children with high function- Makki, M. I., Ager, J., et al. (2010). for understanding the role of cere- autism. Pediatrics 113, 472–486. ing autism and Asperger’s syndrome Alterations in frontal lobe tracts and bellar pathology in autism. Eur. J. Murakami, J. W., Courchesne, E., Haas, from controls. J. Autism Dev. Disord. corpus callosum in young children Neurosci. 31, 544–555. R. H., Press, G. A., and Yeung- 36, 613–621. with autism spectrum disorder. McCormick, D. A. (2005). Neuronal Courchesne, R. (1992). Cerebellar Jasmin, E., Couture, M., McKinley, P., Cereb. Cortex 20, 2103–2113. networks: flip-flops in the brain. and cerebral abnormalities in Rett Reid, G., Fombonne, E., and Gisel, Lalonde, R., Filali, M., Bensoula, Curr. Biol. 15, R294–R296. syndrome: a quantitative MR anal- E. (2009). Sensori-motor and daily A. N., and Lestienne, F. (1996). McCormick, D. A., Steinmetz, J. E., and ysis. AJR Am. J. Roentgenol. 159, living skills of preschool children Sensorimotor learning in three Thompson, R. F. (1985). Lesions of 177–183. with autism spectrum disorders. cerebellar mutant mice. Neurobiol. the inferior olivary complex cause Nguyen, A., Rauch, T. A., Pfeifer, J. Autism Dev. Disord. 39, 231–241. Learn. Mem. 65, 113–120. extinction of the classically condi- G. P., and Hu, V. W. (2010). Jin, P., and Warren, S. T. (2000). Lalonde, R., and Strazielle, C. (2008). tioned eyeblink response. Brain Res. Global methylation profiling of Understanding the molecular basis Discrimination learning in Rora(sg) 359, 120–130. symphoblastoid cell lines reveals of fragile X syndrome. Hum. Mol. and Grid2(ho) mutant mice. McKelvey, J. R., Lambert, R., Mottron, epigenetic contributions to autism Genet. 9, 901–908. Neurobiol. Learn. Mem. 90, L., and Shevell, M. I. (1995). spectrum disorders and a novel Johnson, R. T. (1972). Effects of viral 472–474. Right-hemisphere dysfunction autism candidate gene, RORA, infection on the developing ner- Lam, K. S., Aman, M. G., and Arnold, in Asperger’s syndrome. J. Child whose protein product is reduced vous system. N.Engl.J.Med.287, L. E. (2006). Neurochemical corre- Neurol. 10, 310–314. in autistic brain. FASEB J. 24, 599–604. lates of autistic disorder: a review of Middleton, F. A., and Strick, P. L. 3036–3051. Joyner, A. L. (1996). Engrailed, Wnt the literature. Res. Dev. Disabil. 27, (2000). Basal ganglia and cerebel- Oldfors, A., Sourander, P., Armstrong, and Pax genes regulate midbrain- 254–289. lar loops: motor and cognitive cir- D. L., Percy, A. K., Witt-Engerström, hindbrain development. Trends Landis,D.M.,andSidman,R.L. cuits. Brain Res. Brain Res. Rev. 31, I., and Hagberg, B. A. (1990). Genet. 12, 15–20. (1978). Electron microscopic anal- 236–250. Rett syndrome: cerebellar pathol- Khan, V. R., and Brown, I. R. (2002). ysis of postnatal histogenesis in Middleton, F. A., and Strick, P. L. ogy. Pediatr. Neurol. 6, 310–314. The effect of hyperthermia on the cerebellar cortex of staggerer (2001). Cerebellar projections to the Olmos-Serrano,J.L.,Corbin,J.G.,and the induction of cell death in mutant mice. J. Comp. Neurol. 179, prefrontal cortex of the primate. Burns, M. P. (2011). The GABA(A) brain, testis, and thymus of the 831–863. J. Neurosci. 21, 700–712. receptor antagonist THIP amelio- adult and developing rat. Cell Stress Landrigan, P. J. (2010). What causes Millen, K. J., Wurst, W., Herrup, K., rates specific behavioral deficits in Chaperones 7, 73–90. autism? Exploring the environmen- and Joyner, A. L. (1994). Abnormal the mouse model of fragile X syn- Kim, J. J., and Thompson, R. F. (1997). tal contribution. Curr. Opin. Pediatr. embryonic cerebellar develop- drome. Dev. Neurosci. 33, 395–403. Cerebellar circuits and synaptic 22, 219–225. ment and patterning of postnatal Ozonoff,S.,Iosif,A.M.,Baquio,F., mechanisms involved in classical Lauritsen, M. B. (2013). Autism spec- foliation in two mouse Engrailed- Cook, I. C., Hill, M. M., Hutman, eyeblink conditioning. Trends trum disorders. Eur. Child Adolesc. 2mutants.Development 120, T., et al. (2010). A prospective study Neurosci. 20, 177–181. Psychiatry Suppl. 1, S37–S42. 695–706. of the emergence of early behavioral

Frontiers in Systems Neuroscience www.frontiersin.org May2013|Volume7|Article15| 14 Rogers et al. Autism and the cerebellum

signs of autism. J. Am. Acad. Child Pletnikov,M.V.,Rubin,S.A., in young children with pervasive Salamone, J. D., and Correa, M. (2012). Adolesc. Psychiatry 49, 256–266. Vasudevan,K.,Moran,T.H., developmental disorders. J. Am. The mysterious motivational func- Ozonoff, S., South, M., and Provencal, and Carbone, K. M. (1999). Acad. Child Adolesc. Psychiatry 29, tions of mesolimbic dopamine. S. (2007). “Executive functions in Developmental brain injury associ- 863–872. Neuron 76, 470–485. autism: theory and practice,” in New ated with abnormal play behavior Rogers, S. J., Wehner, D. E., and Sato, A., Sekine, Y., Saruta, C., Nishibe, Developments in Autism: The Future in neonatally Borna disease virus- Hagerman, R. (2001). The behav- H., Morita, N., Sato, Y., et al. (2008). is Today, eds J. M. Pérez, P. M. infected Lewis rats: a model of ioral phenotype in fragile X: symp- Cerebellar development transcrip- González,M.C.Comí,andC.Nieto autism. Behav. Brain Res. 100, toms of autism in very young tome database (CDT-DB): profiling (Philadelphia, PA: Asociación de 43–50. children with fragile X syndrome, of spatio-temporal gene expression Padres de Personas con Autismo), Powell, E. M., Campbell, D. B., idiopathic autism, and other devel- during the postnatal development 185–213. Stanwood, G. D., Davis, C., opmental disorders. J. Dev. Behav. of mouse cerebellum. Neural Netw. Palmen, S. J., van Engeland, H., Hof, Noebels, J. L., and Levitt, P. (2003). Pediatr. 22, 409–417. 21, 1056–1059. P. R., and Schmitz, C. (2004). Genetic disruption of cortical Rogers,T.D.,Dickson,P.E.,Heck,D. Schanen, N. C. (2006). Epigenetics of Neuropathological findings in interneuron development causes H., Goldowitz, D., Mittleman, G., autism spectrum disorders. Hum. autism. Brain 127, 2572–2583. region- and GABA cell type-specific and Blaha, C. D. (2011). Connecting Mol. Genet. 15, R138–R150. Pan, C. Y., Tsai, C. L., and Chu, deficits, epilepsy, and behav- the dots of the cerebro-cerebellar Schmahmann, J. D. (2001). The cere- C. H. (2009). Fundamental move- ioral dysfunction. J. Neurosci. 23, role in cognitive function: neu- brocerebellar system: anatomic sub- ment skills in children diagnosed 622–631. ronal pathways for cerebellar mod- strates of the cerebellar contribution with autism spectrum disorders and Powell, E. M., Mars, W. M., and ulation of dopamine release in to cognition and emotion. Int. Rev. attention deficit hyperactivity dis- Levitt, P. (2001). Hepatocyte growth the prefrontal cortex. Synapse 65, Psychiatry 13, 247–260. order. J. Autism Dev. Disord. 39, factor/scatter factor is a motogen 1204–1212. Schmahmann, J. D., and Caplan, D. 1694–1705. for interneurons migrating from Rogers, T. D., Dickson, P. E., McKimm, (2006). Cognition, emotion and the Patterson, P. H. (2009). Immune the ventral to dorsal telencephalon. E.,Heck,D.,Goldowitz,D.,Blaha, cerebellum. Brain 129, 290–292. involvement in schizophrenia and Neuron 30, 79–89. C. D., et al. (2013). Reorganization Schmahmann, J. D., and Pandya, D. N. autism: etiology, pathology and Provost, B., Lopez, B. R., and Heimerl, of circuits underlying cerebellar (2008). Disconnection syndromes animal models. Behav. Brain Res. S. (2007). A comparison of motor modulation of prefrontal corti- of basal ganglia, thalamus, and cere- 204, 313–321. delays in young children: autism cal dopamine in mouse models brocerebellar systems. Cortex 44, Patterson, P. H. (2011). Maternal spectrum disorder, developmental of Autism Spectrum Disorder. 1037–1066. infection and immune involvement delay, and developmental con- Cerebellum. doi: 10.1007/s12311- Schmahmann, J. D., and Sherman, J. in autism. Trends Mol. Med. 17, cerns. J. Autism Dev. Disord. 37, 013-0462-2. [Epub ahead of C. (1998). The cerebellar cogni- 389–394. 321–328. print]. tive affective syndrome. Brain 121, Peça, J., Feliciano, C., Ting, J. T., Wang, Reardon, W., Donoghue, V., Murphy, Rose, S., Melnyk, S., Pavliv, O., Bai, 561–579. W.,Wells,M.F.,Venkatraman,T. A. M., King, M. D., Mayne, P. D., S., Nick, T. G., Frye, R. E., et al. Schneider, T., and Przewlocki, R. N., et al. (2011). Shank3 mutant Horn, N., et al. (2010). Progressive (2012). Evidence of oxidative dam- (2005). Behavioral alterations in mice display autistic-like behav- cerebellar degenerative changes in age and inflammation associated rats prenatally exposed to valproic iors and striatal dysfunction. Nature the severe mental retardation syn- with low glutathione redox status in acid: animal model of autism. 472, 437–442. drome caused by duplication of the autistic brain. Transl. Psychiatry Neuropsychopharmacology 30, Perciavalle, V., Berretta, S., and MECP2 and adjacent loci on Xq28. 2:e134. doi: 10.1038/tp.2012.61 80–89. Raffaele, R. (1989). Projections Eur. J. Pediatr. 169, 941–949. Roullet, F. I., Wollaston, L., Schwarz, C., and Schmitz, Y. (1997). from the intracerebellar nuclei to Rice, D., and Barone, S. (2000). Critical Decatanzaro, D., and Foster, J. Projection from the cerebellar lat- the ventral midbrain tegmentum in periods of vulnerability for the A. (2010). Behavioral and molec- eral nucleus to precerebellar nuclei the rat. Neuroscience 29, 109–119. developing nervous system: evi- ular changes in the mouse in in the mossy fiber pathway is gluta- Peterson, T. C., Villatoro, L., Arneson, dence from humans and animal response to prenatal exposure to the matergic: a study combining antero- T., Ahuja, B., Voss, S., and Swain, models. Environ. Health Perspect. anti-epileptic drug valproic acid. grade tracing with immunogold R. A. (2012). Behavior modifica- 108, 511–533. Neuroscience 170, 514–522. labeling in the rat. J. Comp. Neurol. tion after inactivation of cerebellar Richt, J. A., VandeWoude, S., Zink, Roussignol, G., Ango, F., Romorini, S., 381, 320–334. dentate nuclei. Behav. Neurosci. 126, M. C., Clements, J. E., Herzog, S., Tu,J.C.,Sala,C.,Worley,P.F.,etal. Seamans, J. K., and Yang, C. R. 551–562. Stitz, L., et al. (1992). Infection with (2005). Shank expression is suffi- (2004). The principal features and Petit, E., Hérault, J., Martineau, Borna disearse virus: molecular and cient to induce functional dendritic mechanisms of dopamine modula- J., Perrot, A., Barthélémy, C., immunobiological characterization spine synapses on aspiny neurons. tion in the prefrontal cortex. Prog. Hameury, L., et al. (1995). of the agent. Clin. Infect. Dis. 14, J. Neurosci. 25, 3560–3570. Neurobiol. 74, 1–58. Association study with two markers 1240–1250. Rubin,S.A.,Sylves,P.,Vogel,M., Sells, C. J., Carpenter, R. L., and Ray, of a human homeogene in infantile Robbins,T.W.,andRoberts,A.C. Pletnikov, M., Moran, T. H., C. G. (1975). Sequelae of central- autism. J. Med. Genet. 32, 269–274. (2007). Differential regulation Schwartz, G. J., et al. (1999). Borna nervous-system enterovirus infec- Pinto, A., Jankowski, M., and Sesack, of fronto-executive function by disease virus-induced hippocampal tions. N. Engl. J. Med. 293, 1–4. S. R. (2003). Projections from the the monoamines and acetyl- dentate gyrus damage is associated Sen, B., Singh, A. S., Sinha, S., paraventricular nucleus of the tha- choline. Cereb. Cortex 17(Suppl. 1), with spatial learning and memory Chatterjee, A., Ahmed, S., Ghosh, lamus to the rat prefrontal cortex i151–i160. deficits. Brain Res. Bull. 48, 23–30. S., et al. (2010). Family-based stud- and nucleus accumbens shell: ultra- Rodier, P. M. (2002). Converging Sabaratnam, M. (2000). Pathological ies indicate association of Engrailed structural characteristics and spa- evidence for brain stem injury and neuropathological findings in 2 gene with autism in an Indian tial relationships with dopamine in autism. Dev. Psychopathol. 14, two males with fragile-X syndrome. population. Genes Brain Behav. 9, afferents. J. Comp. Neurol. 459, 537–557. J. Intellect. Disabil. Res. 44, 81–85. 248–255. 142–155. Roffler-Tarlov, S., and Herrup, K. Sadakata, T., and Furuichi, T. (2009). Sepeta, L., Tsuchiya, N., Davies, M. Pletnikov,M.V.,Rubin,S.A.,Moran, (1981). Quantitative examination Developmentally regulated Ca2+- S., Sigman, M., Bookheimer, S. Y., T. H., and Carbone, K. M. (2003). of the deep cerebellar nuclei in the dependent activator protein for and Dapretto, M. (2012). Abnormal Exploring the cerebellum with a staggerer mutant mouse. Brain Res. secretin 2 (CAPS2) is involved social reward processing in autism new tool: neonatal Borna disease 215, 49–59. in BDNF secretin and is associ- as indexed by pupillary responses virus (BDV) infection of the rat’s Rogers,S.J.,andDiLalla,D.L. ated with autism susceptibility. to happy faces. J. Neurodev. Disord. brain. Cerebellum 2, 62–70. (1990). Age of symptom onset Cerebellum 8, 312–322. 4:17. doi: 10.1186/1866-1955-4-17

Frontiers in Systems Neuroscience www.frontiersin.org May 2013 | Volume 7 | Article 15 | 15 Rogers et al. Autism and the cerebellum

Shi, L., Fatemi, S. H., Sidwell, R. W., and Testa-Silva, G., Loebel, A., Giugliano, Wang,A.T.,Lee,S.S.,Sigman,M., behavioral deficits in neonatal mice. Patterson, P. H. (2003). Maternal M., de Kock, C. P., Mansvelder, H. and Dapretto, M. (2007). Reading Brain Res. 1203, 126–132. influenza infection causes marked D., and Meredith, R. M. (2012). affect in the face and voice: neu- Yu,K.K.,Cheung,C.,Chua,S. behavioral and pharmacological Hyperconnectivity and slow ral correlates of interpreting com- E., and McAlonan, G. M. (2011). changes in the offspring. J. Neurosci. synapses during early development municative intent in children and Can Asperger syndrome be distin- 23, 297–302. of medial prefrontal cortex in a adolescents with autism spectrum guished from autism? An anatomic Shi,L.,Smith,S.E.,Malkova,N., mouse model for mental retarda- disorders. Arch. Gen. Psychiatry 64, likelihood meta-analysis of MRI Tse, D., Su, Y., and Patterson, P. tion and autism. Cereb. Cortex 22, 698–708. studies. J. Psychiatry Neurosci. 36, H. (2009). Activation of the mater- 1333–1342. Wang, L., Jia, M., Yue, W., Tang, F., 412–421. nal immune system alters cerebellar Thoma, P., Bellebaum, C., Koch, B., Qu, M., Ruan, Y., et al. (2008). Zerbo,O.,Iosif,A.M.,Walker,C., development in the offspring. Brain Schwarz, M., and Daum, I. (2008). Association of the ENGRAILED 2 Ozonoff, S., Hansen, R. L., and Behav. Immun. 23, 116–123. The cerebellum is involved in (EN2) gene with autism in Chinese Hertz-Picciotto, I. (2013). Is mater- Sivaswamy, L., Kumar, A., Rajan, D., reward-based reversal learning. Han population. Am. J. Med. Genet. nal influenza or fever during Behen, M., Muzik, O., Chugani, Cerebellum 7, 433–443. B Neuropsychiatr. Genet. 147B, pregnancy associated with autism D., et al. (2010). A diffusion ten- Tsatsanis, K. D., Rourke, B. P., Klin, A., 434–438. or developmental delays? Results sor imaging study of the cerebellar Colkmar, F. R., Cicchetti, D., and Wang,X.,McCoy,P.A.,Rodriguiz, from the CHARGE (CHildhood pathways in children with autism Schultz, R. T. (2003). Reduced tha- R.M.,Pan,Y.,Je,H.S.,Roberts, Autism Risks from Genetics and spectrum disorder. J. Child Neurol. lamic volume in high-functioning A. C., et al. (2011). Synaptic dys- Environment) Study. J. Autism Dev. 25, 1223–1231. individuals with autism. Biol. function and abnormal behaviors Disord. 43, 25–33. Snider, R. S., Maiti, A., and Snider, Psychiatry 53, 121–129. in mice lacking major isoforms Zhong, X., Li, H., and Chang, Q. S. R. (1976). Cerebellar pathways Tsiouris, J. A., and Brown, W. T. (2004). of Shank3. Hum. Mol. Genet. 20, (2012). MeCP2 phosphoryla- to ventral midbrain and nigra. Exp. Neuropsychiatric symptoms of frag- 3093–3108. tion is required for modulating Neurol. 53, 714–728. ile X syndrome: pathophysiology Webb, S. J., Sparks, B. F., Friedman, S. synaptic scaling through mGluR5. Solbrig, M. V., Koob, G. F., Fallon, and pharmacotherapy. CNS Drugs D.,Shaw,D.W.,Giedd,J.,Dawson, J. Neurosci. 32, 12841–12847. J.H.,Reid,S.,andLipkin,W.I. 18, 687–703. G., et al. (2009). Cerebellar ver- Zocher, M., Czub, S., Schulte-Mönting, (1996). Prefrontal cortex dysfunc- Van Waelvelde, H., Oostra, A., Dewitte, mal volumes and behavioral corre- J.,deLaTorre,J.C.,andSauder, tion in Borna disease virus (BDV)— G., Van Den Broeck, C., and lates in children with autism spec- C. (2000). Alterations in neu- infected rats. Biol. Psychiatry 40, Jongmans, M. J. (2010). Stability of trum disorder. Psychiatry Res. 172, rotrophin and neurotrophin 629–636. motor problems in young children 61–67. receptor gene expression patterns Sonnier, L., Le Pen, G., Hartmann, with or at risk of autism spec- Whitney, E. R., Kemper, T. L., Bauman, in the rat central nervous system A., Bizot, J. C., Trovero, F., Krebs, trum disorders, ADHD, and or M. L., Rosene, D. L., and Blatt, following perinatal Borna disease M. O., et al. (2007). Progressive developmental coordination dis- G. J. (2008). Cerebellar Purkinje virus infection. J. Neurovirol. 6, loss of dopaminergic neurons order. Dev. Med. Child Neurol. 52, cells are reduced in a subpop- 462–477. in the ventral midbrain of adult e174–e178. ulation of autistic brains: a Zuo, J., De Jager, P. L., Takahashi, K. A., mice heterozygote for Engrailed1. Vargas, D. L., Nascimbene, C., stereologic experiment using Jiang,W.,Linden,D.J.,andHeintz, J. Neurosci. 27, 1063–1071. Krishnan, C., Zimmerman, A. W., calbindin-D28k. Cerebellum 7, N. (1997). Neurodegeneration in Steinlin, M. (2008). Cerebellar disor- and Pardo, C. A. (2005). Neuroglial 406–416. Lurchermicecausedbymutation ders in childhood: cognitive prob- activation and neuroinflammation Williams, B. L., and Lipkin, W. I. in delta2 glutamate receptor gene. lems. Cerebellum 7, 607–610. in the brain of patients with autism. (2006). Edoplasmic reticulum Nature 388, 769–773. Strick,P.L.,Dum,R.P.,andFiez,J.A. Ann. Neurol. 57, 67–81. stress and neurodegenration in (2009). Cerebellum and nonmotor Verkerk, A. J., Pieretti, M., Sutcliffe, J. rats neonatally infected with Conflict of Interest Statement: The function. Annu. Rev. Neurosci. 32, S., Fu, Y. H., Kuhl, D. P., Pizzuti, A., borna disease virus. J. Virol. 80, authors declare that the research 413–434. et al. (1991). Identification of a gene 8613–8626. was conducted in the absence of any Sutera, S., Pandey, J., Esser, E. L., (FMR-1) containing a CGG repeat Wolff,J.J.,Gu,H.,Gerig,G.,Elison, commercial or financial relationships Rosenthal, M. A., Wilson, L. B., coincident with a breakpoint clus- J. T., Styner, M., Gouttard, S., et al. that could be construed as a potential Barton, M., et al. (2007). Predictors ter region exhibiting length varia- (2012). Differences in white mat- conflict of interest. of optimal outcome in toddlers tion in fragile X syndrome. Cell 65, ter fiber tract development present diagnosed with autism spectrum 905–914. from 6 to 24 months in infants Received: 23 October 2012; accepted: 23 disorders. J. Autism Dev. Disord. 37, Verpelli, C., Dvoretskova, E., with autism. Am. J. Psychiatry 169, April 2013; published online: 10 May 98–107. Vicidomini, C., Rossi, F., 589–600. 2013. Takarae, Y., Minshew, N. J., Luna, Chiappalone, M., Schoen, M., Wong,T.K.,Fung,P.C.,Chua,S. Citation: Rogers TD, McKimm E, B., and Sweeney, J. A. (2007). et al. (2011). Importance of E., and McAlonan, G. M. (2008). Dickson PE, Goldowitz D, Blaha CD and Atypical involvement of frontostri- Shank3 protein in regulating Abnormal spatiotemporal process- Mittleman G (2013) Is autism a disease atal systems during sensorimotor metabotropic glutamate receptor 5 ing of emotional facial expressions of the cerebellum? An integration of control in autism. Psychiatry Res. (MGluR5) expression and signaling in childhood autism: dipole clinical and pre-clinical research. Front. 156, 117–127. at synapses. J. Biol. Chem. 286, source analysis of event-related Syst. Neurosci. 7:15. doi: 10.3389/fnsys. Tamura, R., Kitamura, H., Endo, T., 34839–34850. potentials. Eur. J. Neurosci. 28, 2013.00015 Hasegawa, N., and Someya, T. Wagner, G. C., Reuhl, K. R., Cheh, 407–416. Copyright © 2013 Rogers, McKimm, (2010). Reduced thalamic volume M.,McRae,P.,andHalladay,A. Yang,P.,Lung,F.W.,Jong,Y.J., Dickson, Goldowitz, Blaha and observed across different subgroups K. (2006). A new neurobehavioral Hsieh,H.Y.,Liang,C.L.,and Mittleman. This is an open-access of autism spectrum disorders. model of autism in mice: pre- and Juo, S. H. (2008). Association of article distributed under the terms of the Psychiatry Res. 184, 186–188. postnatal exposure to sodium val- the homeobox transcription factor Creative Commons Attribution License, ten Donkelaar, H. J., Lammens, M., proate. J. Autism Dev. Disord. 36, gene ENGRAILED 2 with autis- which permits use, distribution and Wesseling, P., Thijssen, H. O., and 779–793. tic disorder in Chinese children. reproduction in other forums, provided Renier, W. O. (2003). Development Wallén, A., and Perlmann, T. Neuropsychobiology 57, 3–8. the original authors and source are and developmental disorders of the (2003). Transcriptional control Yochum,C.L.,Dowling,P.,Reuhl,K. credited and subject to any copyright human cerebellum. J. Neurol. 250, of dopamine neuron development. R., Wagner, G. C., and Ming, X. notices concerning any third-party 1025–1036. Ann. N.Y. Acad. Sci. 991, 48–60. (2008). VPA-induced apoptosis and graphics etc.

Frontiers in Systems Neuroscience www.frontiersin.org May2013|Volume7|Article15| 16