article reviews

DEVELOPMENTAL MEDICINE & CHILD NEUROLOGY REVIEW

The genetic landscape of autism spectrum disorders

RASIM O ROSTI1 | ABDELRAHIM A SADEK2 | KEITH K VAUX1 | JOSEPH G GLEESON1

1 Department of Neurosciences and Pediatrics, Howard Hughes Medical Institute, University of California, San Diego, CA, USA. 2 Pediatric Neurology Unit, Department of Pediatrics, Faculty of Medicine, Sohag University, Sohag, Egypt.

Correspondence to Joseph G Gleeson at University of California San Diego, 9500 Gilman Drive M/C 0665, La Jolla, CA, USA. E-mail: [email protected]

PUBLICATION DATA

Accepted for publication 22nd July 2013.

Published online 1st October 2013.

ABBREVIATIONS

ASD Autism spectrum disorders

CGH Comparative genomic

hybridization

CNV Copy number variants

Autism spectrum disorders (ASDs) are a group of heterogeneous neurodevelopmental disor-

ders that show impaired communication and socialization, restricted interests, and stereotypi-

cal behavioral patterns. Recent advances in molecular medicine and high throughput

screenings, such as array comparative genomic hybridization (CGH) and exome and whole

genome sequencing, have revealed both novel insights and new questions about the nature

of this spectrum of disorders. What has emerged is a better understanding about the genetic

architecture of various genetic subtypes of ASD and correlations of genetic mutations with

specific autism subtypes. Based on this new information, we outline a strategy for advancing

diagnosis, prognosis, and counseling for patients and families.

Autism spectrum disorders (ASDs) are a group of com- plex neurodevelopmental disabilities that affect social inter- action and communication skills. The prevalence of ASDs appears to be constantly and gradually increasing, but it is not clear if this is because of clarification of diagnostic cri- teria or an actual increase in the number of cases. Most recent estimates find that the median of prevalence esti- mates of ASD is 62 out of 10 000.1 Those with molecu- larly defined causes make up roughly 20% of the cases, but the heritability has been estimated to be 90%, suggesting as yet undiscovered causes. However, there are new reports suggesting that the previous estimate of heritability was too high and may need to be adjusted downwards.2

The diagnostic criteria have evolved with increasing clin- ical and molecular understanding of this umbrella term. This aspect makes the diagnosis more challenging as the clinical spectrum is highly variable and the etiological sub- grouping tends to change with the ever-growing molecular data now fed by high throughput techniques such as array comparative genomic hybridization (CGH), whole exome and whole genome sequencing. The judgment of the physician, critical in achieving accurate prognosis and genetic counseling, requires a systematic approach. By incorporating these techniques along with careful clinical and neuropsychological assessment, a more accurate diagnosis of an ASD disorder can be achieved.

Distinguishing between essential autism and complex (syndromic) autism might be considered the starting point of this systematic approach.3 Of all individuals meeting cri- teria for autism, essential autism makes up approximately 75% of the cases. Although essentially a diagnosis of exclu- sion, the main characteristics are the lack of dysmorphic features, higher male to female ratio (6:1), higher sibling recurrence risk (up to 35%), and positive family history

(up to 20%). Syndromic autism, on the other hand, is characterized by accompanying recognizable patterns of dysmorphology, a reduced male to female ratio (3.5:1), lower recurrence risk (4%–6%), and family history to a lesser extent (up to 9%).1

Clinical recognition of well-known phenotypes leading to a targeted molecular testing approach can strengthen the hand of the clinician in answering additional questions about the recurrence risk and prognosis according to the molecular basis identified by targeted testing. However, for most forms of essential autism, there is no familiar pheno- type that points to one particular genetic cause or another. In the absence of a specific genetic test option after clinical evaluation, array CGH testing is recommended as it has been reported to provide the highest diagnostic yield in cases with ASD – ranging from 10%4 to 18%.5 Once such a genetic alteration is identified in an affected individual, more specific information about diagnosis, prognosis, recurrence risks, and possible targeted therapy options can be conveyed to the family.

ASD phenotypes caused by genetic alterations can be divided into three subgroups: cytogenetic alterations that can be detected by conventional karyotyping (not the primary focus of this paper). These comprise classic chro- mosomal syndromes such as Turner and Down syndrome, and large segmental cytogenetic alterations that have been associated with autism.6 Other subgroups are single gene disorders (detected in <5%) and copy number variants (detected in 10%–35% of the cases1).

NON-SYNDROMIC AND POSSIBLE TREATABLE FORMS OF ASD: SINGLE GENE DISORDERS Single gene disorders are mainly composed of metabolic conditions in which accompanying ASD features are

12 DOI: 10.1111/dmcn.12278 © 2013 Mac Keith Press

reported. Among these, phenylketonuria is common enough to be included in newborn screening programs in most countries. Comorbidity of ASD and phenylketonuria was consistently described in the literature in up to 6% of patients.7 The fact that none of the 62 patients with phen- ylketonuria who were treated with the phenylketonuria diet early in life met ASD diagnostic criteria, whereas 5.7% (2 out of 35) of individuals with late diagnosis met diagnostic criteria, underlines the importance of early diagnosis in successful treatment in such metabolic conditions.8 Condi- tions such as mitochondrial disorders, adenylsuccinate lyase and creatine deficiencies may also phenocopy ASD. Accompanying autistic features in these disorders range from 0.4% to 80%.1,9,10 Although mitochondrial disorders can present with autistic features, atypical findings of hypo- tonia, fatigue with activity, failure to thrive, intermittent episodes of regression, especially those after fever and ele- vated plasma lactate concentrations, make diagnosis of the condition more straightforward. Non-specific features such as epilepsy and intellectual disability that accompany autism in the setting of mitochondrial disorders on the other hand, can make the diagnosis more challenging.

A recent addition to this group of disorders was made by Novarino et al.11 who reported mutations in the BCKDK gene in the affected individuals from three consan- guineous families who had epilepsy, autistic features, and intellectual disability. The encoded protein is responsible for phosphorylation-mediated inactivation of the E1a sub- unit of branched-chain ketoacid dehydrogenase enzyme. Patients with BCKDK mutations displayed reductions in BCKDK messenger RNA and protein, E1a phosphoryla- tion, and plasma branched-chain amino acids. Bckdk knock- out mice showed abnormal brain amino acid profiles and neurobehavioral deficits that respond to dietary supple- mentation. By supplementing the diet of humans with branched-chain amino acids, the authors were able to nor- malize plasma branched-chain amino acids levels, but the degree to which neurocognitive changes are treatable or reversible remains to be determined.

ASD ASSOCIATED WITH RECOGNIZABLE PATTERNS OF MALFORMATIONS CAUSED BY SINGLE GENE DISORDERS Clues to recognizable or syndromic forms of autism may be apparent in the clinical evaluation (Table I). Many patients with ASD show unique additional findings (Table II), which will enable the clinician to evoke more targeted testing strategies. A proposed diagnostic work-up for patients with ASD is shown in Figure 1.

Fragile X When ‘ASD’ and ‘syndrome’ are pronounced in the same sentence, fragile X pops into every clinician’s mind. This syndrome is so enmeshed with ASD that general practice is to test almost every patient with ASD for FMR1 muta- tions, which yield positive results only in 1% to 3%.12

Whether it is wise to continue the practice of testing all

patients with ASD with such a low diagnostic yield remains a subject of debate, especially as no treatments for fragile X exist, and as other testing strategies result in a much higher yield. The behavioral overlap between fragile X and ASD is so common that 30% to 72% of patients with frag- ile X were reported to exhibit ASD symptoms in various reports.13,14 In our own practice, we reserve testing for those displaying one or more of the fragile X traits, but recognize that some patients will be missed as CGH and whole exome/whole genome sequencing cannot identify the common repeat expansion mutation.

MECP2-related Another syndrome that overlaps with autism phenotype is Rett syndrome. MECP2 mutations, responsible for approx- imately 96% of classic Rett syndrome, have been reported in approximately 1% of patients diagnosed with essential ASD.15 The most important diagnostic handle in differen- tiating Rett syndrome from autism must be the acquired microcephaly seen in Rett syndrome patients. Nevertheless, early periods can be confusing for definitive diagnosis as the study of Young et al.16 shows, in which analyses of multiple databases determined 17.6% of patients eventually proven to have MECP2 positive Rett syndrome had an initial diagnosis of autism. As most MECP2 mutations will be identified on whole exome/whole genome sequencing, future strategies might relegate the need for specific testing.

Tuberous sclerosis Many children with tuberous sclerosis complex exhibit autistic behaviors. This rate decreases from 66% at age 1.5 years to 50% at age 2.5 years, probably because of other tuberous sclerosis complex features becoming more prominent later in life.17 It has been reported that 1.1% to 1.3% of patients with an initial diagnosis of ASD have been found to test positive for either TSC1 or TSC2 mutations.18,19 Autosomal dominant inheritance pattern and characteristic skin lesions such as Shagreen patches,

Table I: Clues to certain clinical scenarios and unique findings

Presence of epilepsy, dysmorphism, and intellectual disability might suggest a syndrome

Schizophrenia: 1q21.1 del and dup syndromes, 3q29 del syndromes, 17q12 del syndromes, NRXN1

Gilles De La Tourette features: CNTNAP2 Obesity: 16p11.2 del syndrome MODY: 17q12 del syndrome Macrocephaly: 1q21.1 dup syndrome, PTEN Microcephaly: 16p11.2 dup syndrome, 17q21.31 dup syndrome Limb/hand abnormalities: 1q21.1 del syndrome, 15q13.3 del syndrome, 17q21.31 dup syndrome

MODY, maturity onset diabetes of the young.

What this paper adds • Genetic origins of ASDs and a diagnostic strategy for clinicians at a glance. • Summary of the most common ASD-associated phenotypes that display

unique additional findings.

Review 13

adenoma sebaceum, and hypopigmented macules that can flourish in late childhood or early adolescence should be indicators for the clinician to differentiate these disorders.

PTEN-related PTEN gene mutations, known to cause a group of disorders collectively called PTEN hamartoma tumor syndromes, were found in a subset of patients with the autistic phenotype and extreme macrocephaly of average +5 SD.20 Frequency of PTEN mutations in cohorts of children with autism and macrocephaly are repeatedly between 1% and 17%.20,21 As PTEN gene mutations are strongly associated with tumor syndromes (Cowden, Bannayan–Riley–Ruvalcaba, and pro-

teus-like syndromes), it is important that patients and their families are referred to a surveillance program after molecular diagnosis.

Phelan–McDermid syndrome Another syndrome associated with ASD, Phelan–McDer- mid syndrome, is caused by contiguous gene deletion of the 22q13.3 region. SHANK3, a gene within this region, is reported to be mutated as the cause in about 1% to 5% of patients with ASD, epilepsy, overgrowth, hypotonia, and absent language.22 Patients presenting with these symptoms as well as hearing impairment, tendency to overheat, lack of perspiration, increased tolerance to pain,

Table II: Autism spectrum disorder genotypes with clinically unique relevant associated findings

ASD-related genotype Epilepsy Dysmorphism Intellectual disability Other

1q21.1 deletion syndrome

� Mild facial Mild to moderate Schizophrenia, CHA, cataracts, limb/hand abnormalities

1q21.1 duplication syndrome

� Mild facial Mild to moderate Macrocephaly, schizophrenia

2q37 deletion syndrome

� Mild facial Moderate to severe Pain insensitivity, brachydactyly

3q29 deletion syndrome

� Mild facial Mild to moderate Severe schizophrenia

3q29 duplication syndrome

� Mild facial Mild to moderate Excessive hand creases, pes planus

13q14 deletion syndrome

� Facial Mild to severe Retinoblastoma

15q13.3 deletion syndrome

Idiopathic generalized Mild facial Mild to severe Dyspraxia and dysarticulation, skeletal/hand abnormalities

15q11–q13 duplication syndrome

+ – Moderate to severe Ataxia, hypotonia, sudden unexpected death

16p11.2 deletion syndrome

+ Variable, systemic Moderate to severe Obesity

16p11.2 duplication syndrome

� Mild facial Mild to moderate Microcephaly, attention- deficit–hyperactivity disorder

16p12.1 deletion syndrome

+ Facial Mild to moderate CHA and skeletal abnormalities

17q12 deletion syndrome

Complex partial Variable, systemic Mild to moderate Schizophrenia, MODY

17q21.31 deletion syndrome

+ Variable, systemic Mild to severe Severe hypotonia, friendly demeanor

17q21.31 duplication syndrome

� Mild facial Mild to moderate Microcephaly, hirsuitism, limb/hand abnormalities, atopic dermatitis

BCKDK + – Mild to moderate Low plasma BCAA CACNA1C � Webbing of fingers

and toes, no hair at birth Mild to moderate Long QT syndrome

CNTNAP2 Focal – Mild to moderate GTS features FOXP2 � – Mild to moderate Verbal dyspraxia MECP2 dup (males) � – Severe Recurrent respiratory infections NLGN4 (males) � – Moderate to none Depression, anxiety and intellectual

disability in females NRXN1 + Facial Mild to moderate Schizophrenia PTEN � Facial Mild to moderate Macrocephaly SHANK1 (males) � – None Reduced penetrance in females SHANK2 � – Mild to moderate SHANK3 � Mild facial Moderate to severe Essentially non–verbal SLC6A8 (males) + Facial Severe Megacolon, ileus TMLHE (males) �/+ – None

ASD, autism spectrum disorder; BCAA, branched chain amino acids; BCKDK, branched chain alpha keto acid dehydrogenase kinase; CAC- NA1C, calcium channel voltage dependent Alpha 1 C subunit; CHA, congenital heart abnormalities; CNTNAP2, contactin associated protein like-2; FOXP2, forkhead box P2; GTS, Gilles de La Tourette syndrome; MECP2, methyl-CpG binding protein 2; MODY, maturity onset diabe- tes of the young; NLGN4, neuroligin 4; NRXN1, neurexin 1; PTEN, phosphatase and tensin homolog; SHANK1, SH3 and multiple ankyrin repeat domains 1; SHANK2, SH3 and multiple ankyrin repeat domains 2; SHANK3, SH3 and multiple ankyrin repeat domains 3; SLC6A8, solute carrier family 6, member 8; TMLHE, epsilon-trimethyllysine hydroxylase.

14 Developmental Medicine & Child Neurology 2014, 56: 12–18

inappropriate chewing behavior, and associated craniofacial dysmorphisms should especially be considered for this contiguous deletion syndrome.

Duchenne/Becker muscular dystrophy Recent studies have also shown that the incidence of ASD in Duchenne/Becker muscular dystrophy cohorts is higher than observed in the general population and these rates vary from 3.1% to as high as 19%.23 These findings indicate that in patients presenting with ASD, clinicians should be alert for signs that can be observed in muscular dystrophy such as mild intellectual disability, hypotonia, positive Gower’s sign, and calf muscle pseudohypertrophy.

Lastly, clinicians should be aware that mild forms of Mendelian diseases that do not have the expected usual presentation may be diagnosed as and included in ASD populations referred to clinic offices. Confirming this point is a recent publication by Yu et al. in which a surprisingly high frequency of biallelic hypomorphic (i.e. mild or par- tially inactivating of both chromosomal copies) mutations is found in an ASD cohort, which cause conditions that are dramatically different from the null mutations in the same gene. Hypomorphic mutations in genes AMT, PEX7, and SYNE1 normally causing non-ketotic hyperglycinemia, rhizomelic chondrodysplasia punctata and cerebellar ataxia

respectively, were found in patients who have some but not all of the features of the classic syndrome. In all, about 7% of their large ASD cohort had mutations on such genes. The authors argue the interpretation that biallelic muta- tions in some settings can cause a spectrum of clinical phe- notypes, which at one extreme cause a Mendelian disorder, but at the other extreme represent risk alleles or causes for ASD.24

COPY NUMBER VARIANTS LANDSCAPE OF ASD Recent advances in analysis techniques, such as detection of copy number variants (CNVs) and its application to var- ious cohorts, have led to some surprising findings. Certain CNVs disrupt developmental homeostasis of neuronal development resulting in a range of disorders as part of a neurodevelopmental continuum.25 Most recent data sup- port an oligogenic model in which severity of the neurode- velopmental disease increases with the increasing CNV burden (i.e. as the number of genes altered increases).26

ASD is in the middle of this neurodevelopmental disease pyramid along with epilepsy and schizophrenia, whereas bipolar disease marks the less severe end and intellectual disability and developmental delay constitute the more severe end. In such cases, there is often one major CNV that is found to accompany smaller CNVs resulting in dif- ferent combinations of CNVs and different phenotypes

Essential ASD Screen with A-CGH

Consider karyotype

Complex ASD

Specific genetic testing due to accompanying symptoms

(+) Epilepsy:

15q13.3 del syndrome

15q11-q13 dup syndrome

16p11.2 del syndrome

16p12.1 del syndrome

17q12 del syndrome

17q21.31 del syndrome

BCKDK

CNTNAP2

NRXN1

SLC6A8 (males)

TMLHE (males)

(+) Facial and systemic dysmorphism:

Check for del/dup syndromes

(–) Intellectual disability:

NLGN4 (males)

SHANK1 (males)

TMLHE (males)

Complete history, physical, neurological, and

neuropsychological evaluations

Figure 1: Proposed genetic diagnostic strategy work-up for a child with autism spectrum disorders (ASD). A-CGH, array comparative genomic hybrid- ization; del, deletion; dup, duplication.

Review 15

along the neurodevelopmental spectrum. Whether there can be many CNVs piling up to contribute to a specific phenotype is still controversial. This has been called the ‘one to many and many to one’ phenomenon.27

16p11.2 16p11.2 has been highlighted in a number of recent publications. The deletion is associated with more severe clinical phenotypes with dysmorphic features, intellectual disability, autism, and a strong correlation with accompanying obesity.28 The reciprocal duplication has a somewhat milder phenotype, and patients are usually underweight in contrary to the obesity seen in 16p11.2 deletion.29,30

16p12.1 Deletion of a nearby locus, 16p12.1, was found to be asso- ciated with autism phenotype in 42 probands who also had developmental delay, craniofacial dysmorphology, congeni- tal heart defects, and skeletal abnormalities in varying degrees.31

1q21.1 1q21.1 deletions and duplications exhibit some of the most varied phenotypic spectrum in the CNV landscape of ASD. Autism is more prevalent in patients with higher rates of duplication, whereas those with deletions may be more recognizable,29,32 several authors have reported find- ings of microcephaly, intellectual disability, cardiac abnor- malities, cataracts, schizophrenia, and extremity abnormalities associated with this deletion.33,34

15q13.3 15q13.3 deletion is also observed across multiple pheno- types with strong associations to idiopathic generalized epilepsy, dyspraxia, and disarticulation.29,32,35 Frequency of idiopathic generalized epilepsy reaches approximately 1% in the deletion of this locus, which makes it a diagnostic possibility to consider in patients with ASD together with idiopathic generalized epilepsy and an oro-buccal speech disorder.

3q29 Although rare, 3q29 microdeletion is strongly associated with severe childhood-onset schizophrenia.36 These patients also exhibit intellectual disabilities and mild dysmorphism.37

17q12 17q12 deletions are also intriguing from a clinical perspective as patients can exhibit features such as schizo- phrenia, complex partial epilepsy, and maturity onset diabetes of the young type 5 (MODY5). Thus, these patients frequently stand out clinically from other ASD presentations.38

CONVERGENT LOCALIZATION AND MOLECULAR PATHWAYS OF SUSPECTED ASD GENES As the genetic variability of ASDs mirrors the phenotypic variability, there have emerged a few recurring themes for genes mutated in ASD that are expressed in specific cellu- lar populations and where their associated proteins play an important functions in the nervous system. These potential points of molecular convergence are especially important as they may serve as targets for potential therapeutic efforts in the future.

Postsynaptic translational regulation The postsynaptic density is a protein rich specialization at the postsynaptic membrane critical for effective neural transmission and synaptic maturation.39 This complex was one of the first candidates to emerge in the etiology of ASD. The first clue was the localization of the protein mutated in fragile X, FMRP, which regulates activity- dependent protein-synthesis at the postsynaptic density.40,41

High throughput techniques have revealed that the encoded proteins of many suspected ASD genes are associ- ated with FMRP and are located in the postsynaptic den- sity, making this region a hotspot for ASD-causing mutations. Synaptic scaffolding proteins SHANK2 and SHANK3, key players in ASD pathophysiology, are located in the postsynaptic density as well.42,43 Others include autism candidate genes such as CYFIP1 – located in the 15q11–13 duplication region, as well as MET, PTEN, TSC1, TSC2, and NF1.44–48

Ubiquitination pathways that modulate protein metabo- lism at the synapse are also implicated by mutations in genes in ASD. A well known example is the gene for Angelman syndrome, UBE3A, which has been found to play a role in this pathway as well as genes such as PARK2, RFWD2, FBXO40, and USP7.49,50 Clearly the assembly, maintenance as well as the remodeling of the synapse as a result of neuronal activity, are major determinants of nervous system function and play into ASD pathology.

Neuronal cell adhesion Another molecular convergence in ASD pathophysiology is the finding of mutations in neurexin and neuroligin fami- lies. Neurexins, which have a presynaptic localization, bind to postsynaptically localized neuroligins and together they modulate inhibitory and excitatory synaptic function.51

Expression of neuroligin genes 1 and 2 even in non-neural cells is sufficient to drive neurons to develop synaptic con- tacts on these cells, indicating a potent role in neural development.52 Genes of these superfamilies that were identified to play a role in ASD are CNTNAP2, CNTN4, CNTN6, NLGN1, NLGN3, and NRXN1.39

Neuronal activity modulation Recent studies have implicated the protein products of many suspected ASD genes displaying activity dependent

16 Developmental Medicine & Child Neurology 2014, 56: 12–18

expression and known to modulate neuronal activity. Some of these genes, such as SCN2A, SCN1A, and GRIN2B, encode for ion channels and help mediate synaptic plastic- ity,53,54 whereas others, UBE3A, PCDH10, DIA1, and NHE9, are regulated by transcription factors that are themselves regulated by neuronal activity.49,50,55,56

Excitatory and inhibitory function imbalance Yet another commonly encountered defect in ASD models comes from the results of functional studies in mice. Knockout mouse models of genes such as Nrxn1, Cntnap2, Fmr1, and Shank3 have all resulted in imbalances of excita- tion and inhibition in different regions of the brain. All these knockout models had overlapping behavioral end- ophenotypes relevant to ASD, such as repetitive self- grooming, impairments in social interactions, or reduced ultrasonic vocalizations. It is interesting to note that some antipsychotic drugs such as risperidone and gaboxadol were shown to rescue behavioral abnormalities including hyper- activity and perseveration in these mouse models, validat- ing them as useful surrogates of human disease.57–61

FUTURE GOALS High throughput techniques such as array CGH and whole exome and genome sequencing are yielding new regions and genes of interest in ASD phenotypes every day. As these techniques become more accessible to researchers and clinicians alike around the world, these approaches in future ASD cohorts will only grow. As clinical sequencing enters the mainstream, it will take a concerted effort to build knowledge from an ever-expanding clinical/research infrastructure of data.

Multicenter efforts are under way to tackle the problem of impaired pathways therapeutically. One such trial was on the selective GABA(B) receptor agonist STX209 (arba- clofen, R-baclofen), which corrected the increased basal protein synthesis in the hippocampus, and the increased spine density and synaptic abnormalities in Fmr1 knockout mice. This approach is paving the way for other therapy trials in ASD.62 Until such results are available, it is essen- tial to correlate these data with a meticulous examination of the ASD phenotypes in order to provide accurate diag- nosis, prognosis, and counseling to our patients.

REFERENCES

1. Elsabbagh M, Divan G, Koh YJ, et al. Global preva-

lence of autism and other pervasive developmental disor-

ders. Autism Res 2012; 5: 160–79.

2. Hallmayer J, Cleveland S, Torres A, et al. Genetic heri-

tability and shared environmental factors among twin

pairs with autism. Arch Gen Psychiatry 2011; 68:

1095–102.

3. Miles JH. Autism spectrum disorders – a genetics

review. Genet Med 2011; 13: 278–94.

4. Sebat J, Lakshmi B, Malhotra D, et al. Strong associa-

tion of de novo copy number mutations with autism.

Science 2007; 316: 445–9.

5. Shen Y, Dies KA, Holm IA, et al. Clinical genetic test-

ing for patients with autism spectrum disorders. Pediat-

rics 2010; 125: e727–35.

6. Kent L, Evans J, Paul M, Sharp M. Comorbidity of

autistic spectrum disorders in children with Down syn-

drome. Dev Med Child Neurol 1999; 41: 153–8.

7. Gurrieri F. Working up autism: the practical role of

medical genetics. Am J Med Genet C Semin Med Genet

2012; 160C: 104–10.

8. Baieli S, Pavone L, Meli C, Fiumara A, Coleman M.

Autism and phenylketonuria. J Autism Dev Disord 2003;

33: 201–4.

9. Oliveira G, Diogo L, Grazina M, et al. Mitochondrial

dysfunction in autism spectrum disorders: a population-

based study. Dev Med Child Neurol 2005; 47: 185–9.

10. Stathis SL, Cowley DM, Broe D. Autism and adenylo-

succinase deficiency. J Am Acad Child Adolesc Psychiatry

2000; 39: 274–5.

11. Novarino G, El-Fishawy P, Kayserili H, et al. Mutations

in BCKD-kinase lead to a potentially treatable form of

autism with epilepsy. Science 2012; 338: 394–7.

12. Reddy KS. Cytogenetic abnormalities and fragile-X syn-

drome in Autism Spectrum Disorder. BMC Med Genet

2005; 6: 3.

13. Farzin F, Perry H, Hessl D, et al. Autism spectrum dis-

orders and attention-deficit/hyperactivity disorder in

boys with the fragile X premutation. J Dev Behav Pediatr

2006; 27: S137–44.

14. Harris SW, Hessl D, Goodlin-Jones B, et al. Autism

profiles of males with fragile X syndrome. Am J Ment

Retard 2008; 113: 427–38.

15. Moretti P, Zoghbi HY. MeCP2 dysfunction in Rett syn-

drome and related disorders. Curr Opin Genet Dev 2006;

16: 276–81.

16. Young DJ, Bebbington A, Anderson A, et al. The diag-

nosis of autism in a female: could it be Rett syndrome?

Eur J Pediatr 2008; 167: 661–9.

17. Jeste SS, Sahin M, Bolton P, Ploubidis GB, Humphrey A.

Characterization of autism in young children with tuber-

ous sclerosis complex. J Child Neurol 2008; 23: 520–5.

18. Baker P, Piven J, Sato Y. Autism and tuberous sclerosis

complex: prevalence and clinical features. J Autism Dev

Disord 1998; 28: 279–85.

19. de Vries PJ, Hunt A, Bolton PF. The psychopathologies

of children and adolescents with tuberous sclerosis com-

plex (TSC): a postal survey of UK families. Eur Child

Adolesc Psychiatry 2007; 16: 16–24.

20. Buxbaum JD, Cai G, Chaste P, et al. Mutation screen-

ing of the PTEN gene in patients with autism spectrum

disorders and macrocephaly. Am J Med Genet B Neuro-

psychiatr Genet 2007; 144B: 484–91.

21. Butler MG, Dasouki MJ, Zhou XP, et al. Subset of indi-

viduals with autism spectrum disorders and extreme

macrocephaly associated with germline PTEN tumour

suppressor gene mutations. J Med Genet 2005; 42:

318–21.

22. Herbert MR. SHANK3, the synapse, and autism. N Engl

J Med 2011; 365: 173–5.

23. Pagnamenta AT, Holt R, Yusuf M, et al. A family with

autism and rare copy number variants disrupting the

Duchenne/Becker muscular dystrophy gene DMD and

TRPM3. J Neurodev Disord 2011; 3: 124–31.

24. Yu TW, Chahrour MH, Coulter ME, et al. Using

whole-exome sequencing to identify inherited causes of

autism. Neuron 2013; 77: 259–73.

25. Stefansson H, Rujescu D, Cichon S, et al. Large recur-

rent microdeletions associated with schizophrenia.

Nature 2008; 455: 232–6.

26. Coe BP, Girirajan S, Eichler EE. The genetic variability

and commonality of neurodevelopmental disease. Am J

Med Genet C Semin Med Genet 2012; 160C: 118–29.

27. Girirajan S, Rosenfeld JA, Coe BP, et al. Phenotypic

heterogeneity of genomic disorders and rare copy-num-

ber variants. N Engl J Med 2012; 367: 1321–31.

28. Walters RG, Jacquemont S, Valsesia A, et al. A new

highly penetrant form of obesity due to deletions on

chromosome 16p11.2. Nature 2010; 463: 671–5.

29. Marshall CR, Noor A, Vincent JB, et al. Structural vari-

ation of chromosomes in autism spectrum disorder. Am

J Hum Genet 2008; 82: 477–88.

30. Jacquemont S, Reymond A, Zufferey F, et al. Mirror

extreme BMI phenotypes associated with gene dosage at

the chromosome 16p11.2 locus. Nature 2011; 478:

97–102.

31. Girirajan S, Rosenfeld JA, Cooper GM, et al. A recur-

rent 16p12.1 microdeletion supports a two-hit model for

severe developmental delay. Nat Genet 2010; 42: 203–9.

Review 17

32. Cooper GM, Coe BP, Girirajan S, et al. A copy number

variation morbidity map of developmental delay. Nat

Genet 2011; 43: 838–46.

33. Mefford HC, Sharp AJ, Baker C, et al. Recurrent rear-

rangements of chromosome 1q21.1 and variable pediat-

ric phenotypes. N Engl J Med 2008; 359: 1685–99.

34. Brunetti-Pierri N, Berg JS, Scaglia F, et al. Recurrent

reciprocal 1q21.1 deletions and duplications associated

with microcephaly or macrocephaly and developmental

and behavioral abnormalities. Nat Genet 2008; 40:

1466–71.

35. Dibbens LM, Mullen S, Helbig I, et al. Familial and

sporadic 15q13.3 microdeletions in idiopathic general-

ized epilepsy: precedent for disorders with complex

inheritance. Hum Mol Genet 2009; 18: 3626–31.

36. Mulle JG, Dodd AF, McGrath JA, et al. Microdeletions

of 3q29 confer high risk for schizophrenia. Am J Hum

Genet 2010; 87: 229–36.

37. Willatt L, Cox J, Barber J, et al. 3q29 microdeletion

syndrome: clinical and molecular characterization of a

new syndrome. Am J Hum Genet 2005; 77: 154–60.

38. Nagamani SC, Erez A, Shen J, et al. Clinical spectrum

associated with recurrent genomic rearrangements in

chromosome 17q12. Eur J Hum Genet 2010; 18: 278–84.

39. Berg JM, Geschwind DH. Autism genetics: searching for

specificity and convergence. Genome Biol 2012; 13: 247.

40. De Rubeis S, Bagni C. Regulation of molecular path-

ways in the Fragile X Syndrome: insights into Autism

Spectrum Disorders. J Neurodev Disord 2011; 3: 257–69.

41. Bassell GJ, Warren ST. Fragile X syndrome: loss of

local mRNA regulation alters synaptic development and

function. Neuron 2008; 60: 201–14.

42. Ozonoff S, Young GS, Carter A, et al. Recurrence risk

for autism spectrum disorders: a Baby Siblings Research

Consortium study. Pediatrics 2011; 128: e488–95.

43. Berkel S, Marshall CR, Weiss B, et al. Mutations in the

SHANK2 synaptic scaffolding gene in autism spectrum

disorder and mental retardation. Nat Genet 2010; 42:

489–91.

44. Napoli I, Mercaldo V, Boyl PP, et al. The fragile X syn-

drome protein represses activity-dependent translation

through CYFIP1, a new 4E-BP. Cell 2008; 134: 1042–54.

45. Campbell DB, Sutcliffe JS, Ebert PJ, et al. A genetic vari-

ant that disrupts MET transcription is associated with

autism. Proc Natl Acad Sci USA 2006; 103: 16834–9.

46. Kelleher RJ 3rd, Bear MF. The autistic neuron: troubled

translation? Cell 2008; 135: 401–6.

47. Waung MW, Huber KM. Protein translation in synaptic

plasticity: mGluR-LTD, Fragile X. Curr Opin Neurobiol

2009; 19: 319–26.

48. Zoghbi HY, Bear MF. Synaptic dysfunction in neurode-

velopmental disorders associated with autism and intel-

lectual disabilities. Cold Spring Harb Perspect Biol 2012; 4.

pii: a009886.

49. Glessner JT, Wang K, Cai G, et al. Autism genome-

wide copy number variation reveals ubiquitin and neuro-

nal genes. Nature 2009; 459: 569–73.

50. Sanders SJ, Ercan-Sencicek AG, Hus V, et al. Multiple

recurrent de novo CNVs, including duplications of the

7q11.23 Williams syndrome region, are strongly associ-

ated with autism. Neuron 2011; 70: 863–85.

51. Hines RM, Wu L, Hines DJ, et al. Synaptic imbalance,

stereotypies, and impaired social interactions in mice

with altered neuroligin 2 expression. J Neurosci 2008; 28:

6055–67.

52. Scheiffele P, Fan J, Choih J, Fetter R, Serafini T. Neu-

roligin expressed in nonneuronal cells triggers presynap-

tic development in contacting axons. Cell 2000; 101:

657–69.

53. O’Roak BJ, Vives L, Girirajan S, et al. Sporadic autism

exomes reveal a highly interconnected protein network

of de novo mutations. Nature 2012; 485: 246–50.

54. Barnby G, Abbott A, Sykes N, et al. Candidate-gene

screening and association analysis at the autism-suscepti-

bility locus on chromosome 16p: evidence of association

at GRIN2A and ABAT. Am J Hum Genet 2005; 76:

950–66.

55. Morrow EM, Yoo SY, Flavell SW, et al. Identifying

autism loci and genes by tracing recent shared ancestry.

Science 2008; 321: 218–23.

56. Flavell SW, Kim TK, Gray JM, et al. Genome-wide

analysis of MEF2 transcriptional program reveals synap-

tic target genes and neuronal activity-dependent poly-

adenylation site selection. Neuron 2008; 60: 1022–38.

57. Penagarikano O, Abrahams BS, Herman EI, et al.

Absence of CNTNAP2 leads to epilepsy, neuronal

migration abnormalities, and core autism-related deficits.

Cell 2011; 147: 235–46.

58. Etherton MR, Blaiss CA, Powell CM, Sudhof TC.

Mouse neurexin-1alpha deletion causes correlated elec-

trophysiological and behavioral changes consistent with

cognitive impairments. Proc Natl Acad Sci USA 2009;

106: 17998–8003.

59. Peca J, Feliciano C, Ting JT, et al. Shank3 mutant mice

display autistic-like behaviours and striatal dysfunction.

Nature 2011; 472: 437–42.

60. Olmos-Serrano JL, Paluszkiewicz SM, Martin BS, Kauf-

mann WE, Corbin JG, Huntsman MM. Defective GAB-

Aergic neurotransmission and pharmacological rescue of

neuronal hyperexcitability in the amygdala in a mouse

model of fragile X syndrome. J Neurosci 2010; 30:

9929–38.

61. Yang M, Bozdagi O, Scattoni ML, et al. Reduced excit-

atory neurotransmission and mild autism-relevant phe-

notypes in adolescent Shank3 null mutant mice. J

Neurosci 2012; 32: 6525–41.

62. Henderson C, Wijetunge L, Kinoshita MN, et al.

Reversal of disease-related pathologies in the fragile X

mouse model by selective activation of GABA(B) recep-

tors with arbaclofen. Sci Transl Med 2012; 4: 152ra28.

18 Developmental Medicine & Child Neurology 2014, 56: 12–18