Genetics of Autism Spectrum Disorders 

Autism spectrum disorders (ASDs) are characterized by impaired socialization, reduced communication, and restricted, repetitive, or stereotyped activities and interests.^[1] As defined by the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (DSM-IV-TR), the spectrum includes diverse phenotypic manifestations, such as classic autism, Asperger syndrome, childhood disintegrative disorder, Rett syndrome, and pervasive developmental disorder not otherwise specified.^[2]

The symptoms of ASDs typically are present before 3 years of age and often are accompanied by abnormalities in cognitive functioning, learning, attention, and sensory processing.^[3] ASDs are associated with numerous comorbidities and disabling symptoms, such as aggression and self-injurious behaviors, for which behavioral and psychopharmacologic interventions are the mainstay of treatment.^[4] There is a well-defined increased prevalence in males, with an affected male–to–affected female ratio of approximately 4:1.^{[5, 6]}

Although ASDs were once considered rare, they are common today, and their prevalence has been dramatically increasing.^[7] On the basis of data from medical records of children who had been diagnosed with or showed signs of ASDs across the United States, the overall US prevalence has been estimated to be approximately 1%, or about 1 in every 110 children.^[8]

According to data from monozygotic twin studies, ASDs have an estimated heritability of more than 90%.^{[9, 10, 11, 12]} Patients with various Mendelian or monogenic diseases—such as Rett syndrome (defective MECP2), fragile X syndrome (defective FMR1), neurofibromatosis 1 (defective NF1), and tuberous sclerosis (defective TSC1 or TSC2)—display features characteristic of the ASDs,^[13] but these causes of ASDs are rare.

The vast majority of family studies suggest that the ASDs do not segregate as a simple Mendelian disorder but, rather, display patterns consistent with a complex trait.^[14] Coupled with this genetic heterogeneity is considerable clinical heterogeneity, as illustrated by substantial differences in the extent and quality of symptoms.^[15]

Genome-wide association studies (GWAS) have implicated the region on chromosome 5p14.1 between CDH9 and CDH10 as the first potential common genetic risk factor in Caucasian populations.^[16] However, resolving the GWAS signal from single nucleotide polymorphisms across large genomic regions to specific causal mutations requires large-scale sequencing; studies using this approach are forthcoming.

Therefore, although ASDs are known to be extremely heritable, their common genetic causes remain largely elusive because of the complex behavioral phenotypes and multigenic etiology of these disorders.^[17]

Although the common genetic mechanisms underlying autism spectrum disorders (ASDs) may be varied and unknown, mounting evidence suggests that defects at the neuronal synapse may underlie the pathophysiology.^{[18, 19]} Initial studies showed that genes involved in serotonin physiology are among the most significant reproducible susceptibility loci in both linkage analyses^{[20, 21, 22, 23, 24]} and biochemical studies.^[25]

Subsequently, models involving more generalized deficits or disturbances in ASDs at the level of central processing emerged through studies addressing weak central coherence,^[26] impaired complex processing,^[27] network abnormalities,^{[28, 29]} disordered information processing,^[30] and defects at the neurochemical level.^[31]

Multiple lines of evidence implicate rare mutations in candidate genes involved in neurotransmission in the pathogenesis of ASDs. Early on, rare mutations of the X-linked genes encoding neuroligins (NLGN),^{[32, 33]} as well as polymorphisms around autosomal genes encoding neurexins (NRXN),^[34] were found to be implicated in the ASDs. Interaction of postsynaptic neuroligins with presynaptic beta-neurexins is involved in the formation of functional synapses, suggesting that defects in synaptogenesis may underlie the etiology.

Since these findings were made, a host of defects in candidate genes that coordinate synaptic transmission have been implicated in ASDs in sporadic families,^[35] and some of these defective candidate genes have been validated with mouse models.^{[36, 37, 38, 39, 40]}

A portion of the sporadic nature of ASD may be attributed to spontaneous mutation, and this may explain why the search for common shared mutations has yielded such inconclusive results.

In one of the first attempts to use next-generation sequencing to study the genetic etiology of ASDs by analyzing the sequenced exomes of 20 parent-child trios, or 60 exomes in total, 4 attractive candidate genes (FOXP1, GRIN2B, SCN1A and LAMC3) involved in neurotransmission were found to harbor functional de novo mutations in sporadic families with ASDs.^[41] That 4 novel mutations could be identified in as few as 20 families suggests that de novo point mutations may be more widespread in ASDs than was previously appreciated.

Although current efforts are moving toward the use of next-generation sequencing to discover novel causative variants, the bulk of the de novo mutations identified to date have been submicroscopic chromosomal deletions and duplications in the form of copy number variations (CNVs) gleaned from genotyping large numbers of individuals.^{[42, 43]}

More sophisticated analyses of these sporadic CNVs in the more biologically relevant context of gene networks demonstrates that CNV defects in ASD patients tend to cluster in genes controlling various biologic pathways important in neurotransmission. For example, CNV defects in cadherins and protocadherins have implicated the neuronal cell adhesion pathway, which plays a critical role in the development of the nervous system by contributing to axonal guidance, synaptic formation and plasticity, and neuronal-glial interactions.

CNVs in other sets of genes have implicated the ubiquitin-proteasome system, which regulates synaptic attributes such as neurotransmitter release, synaptic vesicle recycling in presynaptic terminals, and dynamic changes in dendritic spines and postsynaptic density.^[44] Finally, CNV defects in other sets of molecules, including NRXN1, NLGN3/4X and SHANK3, all of which localize to the synaptic terminal, highlights maturation and function of glutamatergic synapses as important neurodevelopmental pathways that may be disrupted in the ASDs.^[45]

Because of the considerable clinical and genetic heterogeneity of the autism spectrum disorders (ASDs), it is not surprising that there are currently no approved treatments for their core symptoms.

However, in view of the mounting evidence that neurophysiologic processes are disrupted in the ASDs, it also is not surprising that various psychotropic drugs—including stimulants, antidepressants, adrenergic agonists, antipsychotics, antiepileptics, and cholinesterase inhibitors approved for Alzheimer disease (eg, donepezil, galantamine, rivastigmine, and tacrine)^{[13, 46, 47, 48, 49]} —are typically prescribed to mitigate ASD symptoms.

Both first- and second-generation antipsychotics are used to manage aggressive symptoms in autism and have shown efficacy across different studies.^[4] The US Food and Drug Administration (FDA) has approved 2 atypical antipsychotics, risperidone^[50] and aripiprazole,^[51] for the treatment of severe irritability symptoms such as the aggression and self-injurious behavior that are associated with the ASDs.

Other antipsychotics have also been shown to be efficacious in managing disabling ASD symptoms. For example, haloperidol has long been shown to reduce social isolation while improving learning, anger-related behaviors, hyperactivity, and tasks of language acquisition.^{[52, 53]}

Another therapeutic strategy based on translational genomics targets the serotonergic system, which is altered in some patients with ASDs.^{[54, 55, 56]} Rare nonsynonymous mutations in the serotonin transporter (5-HTT) are associated with rigid-compulsive variants of autism,^{[55, 56]} and circulating levels of serotonin are high in some individuals with ASD.^[57]

In addition, the melatonin production pathway begins with serotonin, and individuals with ASD exhibit both abnormal melatonin levels and sleep disturbances,^[58] ; thus, it has been suggested that drugs targeting the serotonergic system may help children with ASDs.^[13] However, clinical trials with selective serotonin reuptake inhibitors (SSRIs) have yielded mixed results.^[59] Fluoxetine has a designated orphan indication for autism.

The finding that as few as 60 exomes could facilitate the identification of 4 functional mutations underlying the autism spectrum disorders (ASDs) suggests that as next-generation sequencing of larger numbers of samples becomes more commonplace, the genomic landscape of rare mutations underlying ASDs will expand considerably.

Just as a better understanding of the molecular genetics of cancer cells has revolutionized the treatment of many cancer treatments, an enhanced ability to identify rare mutations underlying ASDs will facilitate the development of molecular tests with improved diagnostic yields that will be able to aid clinicians in diagnosing the ASDs and their particular genetic subtypes.

For example, approximately 30% of patients with breast cancer overexpress HER2. When the monoclonal antibody trastuzumab binds to the excess HER2 receptors, it prevents the HER2 receptor from signaling the cells in the breast to reproduce uncontrollably; its use in patients with HER2-positive metastatic breast cancer in combination with chemotherapy has been shown to increase survival significantly.^[60]

With better molecular diagnostics that dissect the sporadic genetic mutations underlying the ASDs, a similar personalized approach to treating patients’ specific molecular defects becomes feasible.

For example, selective serotonin reuptake inhibitors (SSRIs) have proven efficacious for rare cases of ASDs where the serotonin system is defective, and work is under way to define analogous treatments for ASDs that are due to malfunctioning NLGN/NRXN pathways, neuronal cell adhesion pathways, glutamatergic receptor pathways, and various other neurophysiologic pathways and networks that have been shown to be defective in sporadic cases.

This type of approach to the ASDs will facilitate the development of a test-and-treat model for drugs that target genetically defined responder populations just as trastuzumab does for patients with HER2-positive breast cancer (see the image below). Such a rational approach to personalized drug design would both restore normal neurophysiology in patients with ASDs by rescuing specific disrupted genetic pathways and avoid exposing these patients to drugs that will precipitate adverse side effects.

Given the clinical and genetic heterogeneity of the ASDs, it is likely that the combination of early tailored behavioral and psychopharmacogenomic intervention will have a positive impact on the prognosis and the outlook for patients with these diseases.