Autism Endophenotypes are Not Unified by Gene Variants or Chromosome Number Variants
Lynn Waterhouse
The College of New Jersey
preprint
In Deborah A. Fein, Neuropsychology of Autism, Oxford University Press.
Introduction
Researchers have claimed that it is crucial to study homogeneous autism samples in order to determine the genetic basis of autism (Folstein, 2006; Goos, 2008; Vernes et al., 2008). Forming homogeneous autism samples, though, is difficult because the diagnosis of autism includes immense phenotypic heterogeneity. Goos (2008) asserted that using diagnosis as a phenotype in genetic studies “is a very serious mistake, as heterogeneity is rampant within diagnostic categories, and individuals with the same diagnosis may vary significantly in phenotype and etiology, even in the presence of high heritability” (p. 270). Consequently, autism endophenotypes—brain deficits or behavioral impairments thought to be linked to a single gene mutation or chromosome alteration — may offer a better opportunity to map brain or behavior autism traits onto genetic variants (Duvall et al., 2007; Happé & Ronald, 2008; Lam, Bodfish, & Piven, 2008; Liu, Paterson & Szatmari, 2008; Reiss, 2009; Spence et al., 2006).
However, the majority of gene mutations and chromosome copy number variants (CNV) associated with autism are also associated with non-autism phenotypes (Abrahams & Geschwind, 2008; Happé & Ronald, 2008; Joober & Bokva, 2009; Kramer & Bokhov, 2009; Lintas & Persico, 2009; Marshall et al., 2008; Morrow et al., 2008; Schanen, 2006; Veenstra-VanderWeele, Christian & Cook, 2004). This link between autism and non-autism phenotypes is both the result of pleiotrophic effects of single gene mutations on phenotypic traits and the result of dosage effects in chromosome duplications and deletions. Consequently, autism endophenotypes also share gene variants and chromosome copy number variants with other syndromes and conditions (Christian et al., 2008; Lintas & Persico, 2009; Marshall et al., 2008; Schanen, 2006). Therefore, as the autism phenotype and individual autism endophenotypes share nearly all their many gene, chromosomal or epigenetic sources with other disorders, it may be more useful to define multiple subsyndromes by unitary gene system causes rather than by the general behavioral syndrome of autism. One reason is that future genetic treatments will require identifying gene system causes. The problem of diagnosis and gene causal factors is discussed in section three below.
This chapter has three sections. The first section summarizes current revisions to molecular genetics relevant to the increasing complexity facing genetics research in autism. This section argues that new findings for genetic and chromosomal variation and molecular epigenetics will help uncover many more gene system sources for autism. The second section reviews gene system sources associated with autism endophenotypes. This section reports that the three diagnostic endophenotypes—social impairment, language impairment, and restricted and repetitive behaviors—and two non-diagnostic endophenotypes—macrocephaly and mental retardation—are each associated with more than one gene system source. As these five endophenotypes are not differentiated by gene system source, they do not serve to form meaningful causal groups. The third section considers the efforts to maintain the syndrome diagnosis of autism despite the wide range of recent genetic and chromosomal findings. This section argues that because single gene mutations and chromosome copy number variants associated with autism have pleiotrophic expressions, wherein autism is just one expression of many varied expressions, the coherence of the autism syndrome is increasingly frayed as more gene system sources for autism are discovered. The chapter concludes that autism is a portmanteau syndrome whose contents must be taken out one by one in order to provide a coherent basis for genetic screening, or possible future treatment research.
Section One: Recent Revisions to Molecular Genetics
Until recently, molecular genetics research was based on a model in which only 1.2% of the human genome contained all ~20,000 genes wherein each gene generates one protein, and each protein has one function. Gene-based diseases and disorders were thought to be caused either by rare single gene mutations or by the accumulation of multiple common chromosomal copy number variants (CNVs). A CNV is a segment of DNA of 1000 base pairs or more containing either multiple duplications or multiple deletions of DNA base pairs. Although these DNA base pair duplications or deletions usually occur at a single locus, surprisingly the duplications or deletions may occur at different loci on the chromosome. CNV regions may include hundreds of genes, disease-linked loci and functional elements (Joober & Bokstra, 2009). Although researchers have identified more than 300 single gene mutations linked to specific disorders, and more than 1400 CNVs have been discovered, these single gene mutations and CNVs have only been able to account for less than 5% of genetic heritability of common diseases such as diabetes and heart disease. A massive amount of “genetic dark matter” is missing (Maher, 2008). Although twin and family studies suggest a 90% heritability for autism, autism-linked gene mutations and known CNVs identified to date explain perhaps 7% to 10% of that heritability (Abrahams & Geschwind, 2008; and see review of genetics in autism by Abrahams in this volume).
The missing “genetic dark matter” that might explain the heritability of autism and other disorders such as diabetes and schizophrenia may be found in novel mutations in genes and CNVs, and in the myriad of processes in the regulation of gene expression. Research has now shown that alternate splicing allows a single gene to generate multiple transcripts, each of which may produce a different protein (Seringhaus & Gerstein, 2008). Moreover, although gene exon regions were previously thought to be the only coding elements of the gene, it is now known that exons can be excluded from the final product, and non-gene sequences can be spliced into the transcript to produce a different protein. Still more surprising, Kapranov, Willingham and Gingeras (2007) reported that more than 90% of the human genome could be transcribed. Transcription outside the known genes includes “pseudogenes, regions of the genome long considered fossils of past genes” (Seringhaus & Gerstein, 2008, p. 468). Adding further complexity to the system, many different genome variants overlap and “multifunctional usage of the same genomic space is common” (Kapranov et al., 2007, p. 414).
Greater variation in the genome base pair sequences has also recently been discovered. Khaja et al. (2006) reported finding 1,447 copy number variant (CNV) chromosome regions including ~30 million base pairs that span the entire 3 billion base pairs of our genome, and cover more than 2,900 genes including many of the 285 single genes known to be associated with diseases. These 1,447 copy number variants are likely to co-function with the 2,900 genes they overlap, and perhaps co-function with non-gene regions of the genome as well. Christian et al. (2008) discovered 51 CNVs in 12% of 397 individuals with autism, and Marshall et al. (2008) reported 277 CNVs in 44% of ASD families. Joober and Bokstra (2009) noted that although humans are resilient to many structural variants that affect hundreds of genes, CNVs are likely to explain a significant subpopulation of autism because autism appears early in development and usually has a severe clinical expression.
In addition to the expanding DNA sequence variation, greater epigenetic variation has also been uncovered, and the epigenome code is now seen as a significant force in the production of phenotypic traits. Epigenetics includes both the transitory and the heritable changes in gene function that occur without a change in the DNA sequence. Epigenetic action involves DNA methylation, histone acetylation, and RNA interference. DNA methylation marks the genomic chromatin to regulate gene expression. Even a histone code has been identified: histones control access to regulatory proteins and act in phosphorylation for DNA replication, thus contributing to regulation of gene expression (Happel & Doenecke, 2008).
Kaminsky et al. (2009) found significant variation in MZ co-twin DNA methylation in the absence of DNA sequence variation. They concluded that an individual’s epigenome could account for considerable phenotypic variation. Reid, Gallais, and Metivier (2009) asserted that gene expression depends on which gene transcript is produced and what epigenetic modifications of chromatin have been made. Epigenetic mechanisms include imprinting— in which one parent’s allele controls gene expression, X-inactivation of one of the two copies of the X chromosome, gene silencing wherein histone modification switches off a gene, and many other mechanisms as well. Kaduake and Blobel (2009) proposed that even the physical structure of chromatin is likely to have phenotypic effects. They reported that chromatin loops form to block gene expression through blocking gene “enhancers and promoters from interacting productively perhaps by separating them physically” (p. 22).
In sum, the new data on the epigenome, the histone code, the CNV findings, and genome sequence functional complexity together create a revised model that seriously “challenges the notion that a DNA sequence in a single region is sufficient to describe a gene” (Seringhaus & Gerstein, 2008, p. 469).
This complexity can be seen in the finding that mutations in MECP2, a chromatin-associated protein that binds methylated CpGs to activate or repress transcription, is the cause of Rett syndrome, an autism spectrum disorder (Swanberg, Nagarajan, Peddada, Yasui, & LaSalle, 2009). The researchers found an interactive relationship between MECP2 and theactivity-dependent early growth response gene 2 (EGR2), requiredfor both brain development and mature neuron function. Most importantly, they reported that Rett syndrome andautism postmortem cortex samples showed a significant reductionin EGR2 protein, suggesting that EGR2 affects neuron development in both syndromes, even though MECP2 function in autism without Rett syndrome does not produce typical Rett syndrome phenotypic features.
Schanen (2006) claimed that autism linkage peaks met imprinting zones at 15q11–13, 7q21–31.31, 7q32.3–36.3, 4q21–31, 11p11.2–13 and 13q12.3, and epigenetic imprinting theories of autism have been proposed. Crespi and Badcock (2008) theorized that aberrant DNA methylation in imprinting can yield both psychosis and autism, leaving social cognition underdeveloped in autism, but dysfunctionally overdeveloped in psychosis. Jones, Skinner, Friez, Schwartz, and Stevenson (2007) theorized that autism is caused by hypomethylation or hypermethylation of CpG sites within gene promoters on the X chromosome that leads to overexpression or partial silencing of one or more brain-expressed genes. Mehler and Purpura (2008) theorized that a dynamic epigenetic remodeling of the locus coeruleus and of the noradrenergic network of pre- and postsynaptic receptors might be the basis for autism.
Section Two: Gene System Sources of Autism Endophenotypes
Goos (2008) argued, “The use of endophenotypes in the study of complex psychiatric disorders is increasing, and has been shown to facilitate the identification of genetic risk factors” (p. 270). The majority of the few genetic studies that have explored autism endophenotypes, have identified endophenotypes as the triad of diagnostic features of autism—impaired social function, impaired communication, and restricted, repetitive, and stereotyped patterns of behavior, interests, and activities.
Gene System Findings for Endophenotypes of Diagnostic Criteria for Autism
Happé and Ronald (2008) reported that each member of the diagnostic triad (impaired social interaction, communication deficits, and rigid or repetitive behaviors) is separately heritable. Vernes et al. (2008) also proposed that the diagnostic triad should be the basis for endophenotypes of autistic-spectrum disorders. Happé, Ronald and Plomin (2006) noted the lack of replication of linkage studies in autism, and they claimed that the triad of autism diagnostic features was only weakly correlated. Happé et al. (2006) and Mandy and Skuse (2008) proposed that studies should be conducted in autism samples that expressed primarily social impairment, or primarily communication impairment, or primarily restricted, repetitive, and stereotyped patterns of behavior, interests, and activities.
Table 1 outlines a selected set of recent findings for diagnostic and non-diagnostic autism endophenotypes. These findings for genetic studies of the three diagnostic autism endophenotypes are mixed.
Social impairment. Liu and colleagues (2008) found no genome-wide linkage for reciprocal social interaction in 976 multiplex families from the Autism Genome Project consortium. Duvall and colleagues (2007) also reported that a diagnostic endophenotype measure of social responsiveness was not significantly linked to any chromosomal regions they explored, although they did detect two possible associated regions on chromosome 11 and 17.
Conversely, Yrigollen and colleagues (2008) reported evidence for polymorphisms of PRL, PRLR, and OXTRgenes in 177 individuals diagnosed with ASD. Oxytocin (OT) is a peptide involved in affiliative behavior. Autistic children have been shown to have abnormal levels of plasma OT (Green et al., 2001; Modahl et al., 1998). Prolactin (PRL) is a pituitary hormonal peptide also found to be important for affiliative behaviors. The researchers found significant correlations between the presence of PRL receptor gene (PRLR) polymorphisms and an endophenotype of social skill impairment,
Communication impairment. Vernes et al. (2008) found an association between the language endophenotype of nonsense word repetition and polymorphisms in the exon 13–15 region of CNTNAP2 in children with specific language impairment. Alarcón et al. (2008) reported evidence of a significant link between a language endophenotype (age of first spoken word) and polymorphisms in the exon 13–15 region of CNTNAP2 in autism. Vernes et al. argued that CNTNAP2-caused deficits appear in a pure form in specific language impairment, but CNTNAP2-caused deficits are only one contributor to the gene system basis of autism. A genome wide association study (GWAS) of individuals with autism by Arking and colleagues (2008) detected a significant linkage signal in the 7q35 region involving the CNTNAP2 gene. Bakkaloglu and colleagues (2008) reported evidence for an association of autism and rare CNTNAP2 variants.
Yrigollen et al. (2008) reported a pattern of significant correlations between polymorphisms of the OXTR and OXTgenes and diagnostic endophenotypes for impaired communication skills in a sample of 177 autistic patients. However, Spence and colleagues (2006) reported finding no linkage in a sample of 133 multiplex autism families for any chromosomal loci and either of two language endophenotypes, the delayed production of words, and the delayed production of phrases. Their study, however, did identify loci on chromosomes 1, 2, 4, 6, 7, 8, 9, 10, 12, 15, and 19 that showed somewhat higher linkage signals in the two language endophenotype subgroups.
Restricted or repetitive activities and interests. Lam and colleagues (2008) found familial aggregation associated with subtypes of the diagnostic endophenotype of restricted or repetitive activities and interests. Lam et al. (2008) identified three subgroups within the endophenotype: repetitive motor behaviors, insistence on sameness, and circumscribed interests. Using sib-pair correlations the researchers found that insistence on sameness and circumscribed interests were family-linked features. Yrigollen et al. (2008) found significant correlations between the presence of PRL receptor gene (PRLR) polymorphisms and an endophenotype of stereotyped behaviors. The researchers also found a pattern of significant correlations between polymorphisms of the OXTR and OXTgenes and diagnostic endophenotypes for stereotyped behaviors.
However, Liu and colleagues (2008) found no genome-wide linkage for restricted, repetitive, and stereotyped patterns of behavior in 976 multiplex families from the Autism Genome Project consortium.
Summary. Taken together, the evidence for gene system sources for endophenotypes of autism based on diagnostic criteria suggests that not much explanatory ground has been gained by limiting studies to endophenotypes of autism based on diagnostic criteria. For example, although Yrigollen et al. (2008) found significant correlations between autism diagnostic endophenotypes and oxytocin and prolactin gene polymorphisms, the correlations were indiscriminate: i.e., polymorphisms of the OXTR, OXT and PRLR genes were significantly correlated with both the diagnostic endophenotype for impaired communication skills and the diagnostic phenotype for stereotyped behaviors.
Gene System Findings for Endophenotypes of Non-Diagnostic Features of Autism
Macrocephaly and mental retardation (MR) and are two non-diagnostic features commonly found in significant subgroups of individuals diagnosed with autism. The prevalence of atypical increased head size (macrocephaly) or atypical increased brain size (megalencephaly) in autism has been identified variously at 0%, 14% , and 75% depending on sample size and measurement employed (Redcay and Courchesne, 2005), and the prevalence of mental retardation in autism is about 60% (Amaral, Schumann, & Nordahl, 2008). While both macrocephaly and MR may be meaningful endophenotypes of autism, neither of them can help to limit associated gene and chromosome variants because it already known that in samples of non-autistic patients both MR and macrocephaly are associated with multiple single gene mutations and CNVs.
Macrocephaly. Williams, Dagli and Battaglia (2008) identified 164 conditions associated with macrocephaly. The condition most commonly found in autism with macrocephaly is a PTEN gene mutation. The PTEN gene generates the phosphatase and tensin homolog protein in which the phosphatase is involved in preventing cells from growing and dividing too rapidly. As can be seen on Table 1, Orrico et al. (2008) reported that in 40 patients with neurodevelopmental disorders and macrocephaly, three novel de novo missense PTEN gene mutations were found (p.H118P, p.Y176C, p.N276S) in two severely mentally retarded patients with autism and in a subject with neurodevelopmental disorders without autistic features. Butler et al. (2005) studied 18 children with autism spectrum disorder and macrocephaly. Of these 13 boys and girls, each with a head circumference range from 2.5 to 8.0 standard deviations above the mean, the three boys with the largest head circumference were found to carry previously undescribed germline PTEN mutations: H93R (exon 4), D252G (exon 7), and F241S (exon 7). Buxsbaum et al. (2007a) found only one boy with PTEN mutations in a sample of 88 individuals with autism who also had macrocephaly.
However, Buxbaum et al. (2007b) screened for the NSD1 gene in a sample of 88 individuals diagnosed with autism who also had macrocephaly: the researchers found no mutations or deletions in the NSD1gene and concluded that the related Sotos syndrome is a rare cause of autism macrocephaly. Moreover, Sacco et al. (2007) studied 241 patients with autism and found no association between macrocephaly and blood levels of serotonin (5-HT).
Other genes and chromosome copy number variants have been associated with macrocephaly in autism. Tsuchiya et al. (2008) reported that serum levels of platelet-endothelial adhesion molecule PECAM-1 and vascular cell adhesion molecule VCAM-1 in subjects with high-functioning autism were lower than those of age-matched controls, and that serum levels of PECAM-1 were negatively correlated with head circumference at birth in autism. Both VCAM-1 and PECAM-1 contribute to the regulation of endothelial cells.
Mental Retardation. The 60% of individuals with autism and mental retardation have not often been studied as a group because the presence of mental retardation has been incorrectly argued to be a separate co-morbid syndrome affecting individuals with autism (Waterhouse, 2008). However, the 66 single gene mutations Wall and colleagues (2009) identified as present in autism families are associated with mild to moderate retardation. Moreover, the chromosome region copy number variants associated with autism (1p, 1q, 2q37, 3q, 4q21–31, 5p, 6q, 7q21–31.31, 7q32.3–36.3, 8q, 11p11.2–13, 13q12.3, 15q11–13, 15q24, 16p11.2, 17q11, 19p, 22q11.2, and Xq (6, 21, 30, 62, 82, 102, 110, 120)) (Veenstra-VanderWeele et al., 2004) are associated with phenotypic expression of mild to moderate retardation. Joober and Bokva (2009) noted, “low IQ is a quasi-constant manifestation of large chromosomal anomalies” (p. 58).
Kramer and van Bokhoven (2009) stated that 2-3% of the population has mild to moderate mental retardation, and over 300 MR genes have been identified, and there are already more than a thousand Mendelian disorders known for which MR is a characterizing feature. The current prevalence of autism is 6-8 in 1,000, of which, 4-6 are likely to have mental retardation. Thus, as only 35-50% of mental retardation has a genetic basis (Kramer & van Bokhoven, 2009), but at least 90% of autism has a genetic basis, therefore 0.70% to 1.5% of a population would be expected to have gene-based MR, and 0.36% to 0.54% of a population would be expected to have gene-based autism with MR. Gene-based autism and MR overlap in population ascertainment because many of the identified MR genes are also associated with autism.
In fact, the pleiotrophic effects of a given individual gene mutation or the dosage effects of chromosome copy number variant do not just include autism features co-occurring with mental retardation: the pleiotrophic variation and dosage range of phenotypic traits can be extremely large. For example, Miller et al. (2008) reported that 10 of 1,441 individuals with autism spectrum disorders (ASD) had segmental alterations at breakpoints four and five (BP4-BP5) of chromosome 15q13.2q13.3. The BP4-BP5 alterations included deletions and duplications spanning CHRNA7, a candidate gene for seizures. However, none of the 10 individuals with the BP4-BP5 alterations had epilepsy or seizures, although one individual had an abnormal EEG. These ten individuals did have significant expressive language deficits, and subsets of the 10 individuals were diagnosed with autism, ADHD, anxiety disorder, and mood disorder. Cognitive impairment varied from moderate mental retardation to normal IQ with learning disability. This pleiotrophic range illustrated by Miller and colleagues findings (2008) is more likely to account for the appearance of mental retardation in autism, than is the possibility that mental retardation is a gene-based co-morbid disorder separate from autism.
Table 1 outlines the association of mental retardation and autism. Cai and colleagues (2008) found that about 1% of their sample diagnosed with autism had chromosome and gene duplications and deletions. Duplications included 15q11-q13, 22q11, TM4SF2 gene, ASMT Xp22.32, and deletions occurred in ASPA in 17p13, PAX6 in 11p13, EXT1 in 8q24, and ARHGEF6 in Xq26. These duplications and deletions also produced mental retardation.
Loat et al. (2008) found MECP2 in association with autism and developmental delay. Weiss et al (2008) found mental retardation and autism in association with deletions and duplications in 16p11.2 . Brunetti-Pierri et al. (2008) and Mefford et al. (2008) found mental retardation and autism in association with deletions or duplications in 1q21.1.
Mental retardation and macrocephaly. A problem for defining mental retardation and macrocephaly as separate endophenotypes of autism is that they can occur together as related features. At least one chromosomal locus linked to autism, 1q21.1, yields both abnormal head size and mental retardation. Brunetti-Pierri and colleagues (2008) studied a large sample of individuals with microdeletion and microduplication at 1q21.1. The researchers found dosage effects such that head circumference was significantly smaller in 21 of 29 individuals with a microdeletion at 1q21.1, whereas head circumference was significantly larger in 10 of 24 individuals with microduplication at 1q21.1. Mefford and colleagues (2008) screened 5218 individuals with idiopathic mental retardation, autism, or congenital anomalies and found 25 unrelated individuals with overlapping deletions of 1q21.1 and 8 individuals with the reciprocal duplication of 1q21.1. Mefford et al. (2008) also reported significant dosage effects for 1q21.1: deletions were associated with microcephaly, and duplications were associated with macrocephaly. Both deletions and duplications, however, were associated with mild-to-moderate mental retardation. Seven of the eight individuals who were carriers of the 1q21.1 duplication had mental retardation, and four of the seven had macrocephaly. The researchers also found that four of the seven had autistic behaviors or autism. In these two studies macrocephaly, autism, and mental retardation are co-occurring phenotypic features of the 1q21.1 duplication.
As can be seen in the data from Table 1, and from Miller et al. (2008), Joober and Bokva (2009), and Wall et al. (2009), and the reviews of Abrahams and Geschwind (2008) and Veenstra-VanderWeele et al. (2004), as well as the chapter by Abrahams in the current volume, mental retardation cannot be an endophenotype of autism. Ramocki and Zoghbi (2008) concluded that chromosome deletion and duplication, the “functional loss or gain of proteins or RNAs involved in diverse processes leads to mental retardation, autism and other neuropsychiatric symptoms”(p. 217). They also argued that there are likely to be thousands of genes whose alteration results in mental retardation or autism or both.
Mental retardation and autism diagnostic features occur together because they are pleiotropic effects of a single gene mutation, or because they are phenotypic expressions of the same chromosome copy number variant. As there are so very many CNVs and single gene mutations that give rise to autism co-occurring with mental retardation, creating an MR endophenotype of autism gains no inferential ground.
Summary. As with the diagnostic endophenotypes, explorations of mental retardation and macrocephaly as endophenotypes of autism have so far done little to increase the homogeneity of gene sources. Instead, new findings have suggested the inherent complexity of single gene mutation pleiotrophy and chromosome number duplication and deletion dosage effects in creating a wide range of aspects of phenotypic expression, and have highlighted the co-occurrences of autism and mental retardation, as well as autism and mental retardation and macrocephaly.
Section Three: Gene Sources Undermine the Autism Diagnostic Phenotype
Maher (2008) noted that, “Medicine tries hard to lump together a complex collection of symptoms and call it a disease. But if thousands of rare genetic variants contribute to a single disease, and the genetic underpinnings can vary radically for different people, how common is it? Are these, in fact, different diseases? “(p. 7). At present, despite the wide variation in gene-based sources, and despite the syndrome overlaps between autism and other syndromes, autism continues to be diagnosed as a unitary behavioral syndrome, and research is conducted on samples defined by diagnosis. The problems with maintaining the syndrome of autism are revealed in the effort to isolate syndromic autism from idiopathic autism, in the effort to construct unified gene-based theories of autism, in the lack of coverage of gene and CNV findings to date, and in the emerging case-carrier examinations of individual chromosome copy number variants.
The Effort to Maintain Syndromic Autism versus Idiopathic Autism: Fragile X Syndrome, Rett Syndrome, Cortical Dysplasia-Focal Epilepsy Syndrome and Autism
The problem of “what makes autism a syndrome?” can be seen in the relationship of autism to Fragile X syndrome (FRX), Rett syndrome (RTT), and to Cortical dysplasia-focal epilepsy syndrome (CDFES). A diagnosis of autism in Fragile X syndrome, Rett syndrome, or Cortical dysplasia-focal epilepsy syndromehas been defined as “syndromic autism” and has been argued to include only 10% of cases of autism (Lintas & Persico, 2009). The remainder of autism is identified as “idiopathic autism”. Although Fragile X and Rett syndrome are both now excluded from the diagnosis of autism, shared gene system sources do link autism and Rett syndrome, autism and Fragile X syndrome, and autism and Cortical dysplasia-focal epilepsy syndrome.
Rett syndrome and autism. Monteggia and Kavalali (2009) noted that Rett syndrome is defined by mutations in the MECP2 gene in the 15q11-13 region. They concluded that epigenetic regulation of gene expression for those genes within the 15q11-13 region for all these disorders is critical for development of the neural circuits involved in social behaviors, language and cognition. Thus, Rett syndrome with autism diagnostic features, Rett syndrome without autism diagnostic features, and autism without Rett syndrome may all arise from mutations in the MECP2 gene within the CNV in 15q11-13.
Schanen (2006) claimed that duplications of the MECP2 gene in chromosome 15q11–13 occurred in up to 5% of individuals with ASD, and that parent of origin effect on chromosome 15q duplications indicated that imprinted genes in this region contributed to ASD. Hogart, Nagarajan, Patzel, Yasui, and LaSalle (2007) posited that epigenetic dysregulation of the 15q11-13 GABAA receptor cluster results in aberrant expression levels of GABRB3 in multiple neurodevelopmental disorders, including Rett syndrome and autism. The researchers noted that epigenetic methylation of an intronic sequence of GABRB3 serves as a binding site for MECP2, and that MECP2 is a positive regulator of GABRB3 expression. Hogart et al. (2007) found that autism samples with loss of biallelic expression of any one of the 15q11-13 GABAA receptor subunit genes had significantly reduced GABRB3 protein levels. Swanberg and colleagues(2009) reported that EGR2 and MECP2 co-regulate one another in both Rett syndrome and autism.
Should there be an EGR2-MECP2-GABRB3 gene mutation network Rett syndrome? This would imply an EGR2-MECP2-GABRB3 gene mutation network autism-without Rett syndrome features, and an EGR2-MECP2-GABRB3 gene network mutation Rett syndrome without autism features.
How should non-Rett syndrome autism and Rett syndrome with an associated diagnosis of autism be differentiated where both are linked to the presence of a MECP2 gene mutation and an associated dysfunction in a gene network?
Fragile X syndrome and autism. Zingerevich et al. (2009) reported that 60% of a sample of 48 children with Fragile X syndrome met the diagnostic criteria for autism or ASD. Similarly, Bearden et al. (2008) reported that 30% of males with Fragile X syndrome could be diagnosed with autism, and that a much greater percentage had some features of autism. Fragile X syndrome, the most common inherited form of mental disability in males, results from an expanded repeat mutation of the fragile X mental retardation 1 (FMR1) gene on the X chromosome. Affected females have a milder form of the disorder. The associated protein, FMRP, contributes to the organization of neuron structure in development, and reduced FMRP is associated with abnormalities of the cerebellar vermis and enlargement of the caudate, both of which have been found in autism without Fragile X syndrome phenotypic features (Redcay & Courchesne, 2005; Stanfield et al. 2008).
As noted above, the current convention has been to define an autism diagnosis in Fragile X syndrome as Fragile X syndrome, and define the presence of the FMR1 gene mutation in autism without Fragile X syndrome physical features as “syndromic autism”. However, the behavioral diagnosis of autism should provide a label for all individuals whose behaviors meet the criteria for autism.
Cortical dysplasia-focal epilepsy syndrome and autism. Cortical dysplasia-focal epilepsy syndrome is associated with a mutation of the Contactin Associated Protein- like 2 (CNTNAP2) gene, a member of the Neurexin family. On the basis of research reports by Alarcón et al. (2008), Arking et al., (2008) and Bakkaloglu et al. (2008), Stephan (2008) argued that it was “reasonable at this point to define CNTNAP2 mutation-positive autistic cases as having ‘Type 1 autism’ “ (p. 7). However, CNTNAP2 mutation-positive individuals already have their own defined syndrome, Cortical dysplasia-focal epilepsy syndrome. More importantly, Bakkaloglu et al. (2008) reported that there was no significant increased CNTNAP2 gene mutation in the individuals with autism compared to controls. The researchers identified only 27 of 635 individuals with autism as having variants of the CNTNAP2 gene, and identified 35 of 942 individuals in a control group as having variants of the CNTNAP2 gene. Although Alarcón et al. (2008) admitted that “large CNVs at the CNTNAP2 locus are not a common cause of autism or a major contributor to the language disorders” (p. 156), the researchers did claim that both rare and common variants of the CNTNAP2 gene contributed to a behavioral endophenotype of autism. However, the behavioral endophenotype is broadly conceived to include mental retardation, repetitive motor behaviors, seizures and language delay (Alarcón et al., 2008). Importantly, as was noted in the discussion of the work by Yrigollen et al. (2008) above, an association between the presence of the CNTNAP2 gene mutation and many behaviors—here, mental retardation, repetitive motor behaviors, seizures and language delay—in individuals is not circumscribed by finding a significant correlation between the presence of the CNTNAP2 gene mutation and language delay.
- The mutation of the CNTNAP2 gene is associated with Cortical dysplasia-focal epilepsy syndrome, the FMR1 gene variant is associated with Fragile X syndrome, and the MECP2 gene variant associated with Rett syndrome. All three gene variants are linked to delayed language and cognition problems. Delayed language and cognition problems are symptoms of many neurodevelopmental disorders including autism. If the pleiotrophic phenotypic expression patterns for many different single gene mutations are similar, what should be the defining point for a syndrome? Should there be a CNTNAP2 gene associated autism, and a CNTNAP2 gene associated learning disability, and a CNTNAP2 gene associated language delay as well as the originally defined CNTNAP2 gene syndrome–Cortical dysplasia-focal epilepsy syndrome?
The Effort to Construct Unified Gene-Based Theories of Autism
Researchers are aware of the many single gene mutations and chromosomal duplications and deletions (See Abrahams & Geschwind, 2008; Abrahams, this volume; Glessner et al., 2009; Gupta & State, 2007; Lintas & Persico, 2009; Ma et al., 2009; Marshall et al., 2008; Skuse, 2007; van der Zwaag et al., 2009; Veenstra-VanderWeele et al., 2004; Wang et al., 2009) associated with autism. Efforts to unify the gene bases for autism have included the creation of models of gene and CNV relationships and the selection of a limited set of genes or CNVs as crucial to autism.
Models of gene mutation and CNV relationships in autism. Ramocki and Zoghbi (2008) argued that the wide range of single gene mutations and CNVs could be bound into a unified pattern by considering homeostasis. They claimed that failure of homeostasis is the mechanism common to all such gene mutations and they argued that any loss or gain of a protein that influences synaptic function might be the source of neurological or psychiatric phenotypes because changes in synaptic function will eventually exhaust the ability of neural circuits to establish homeostasis. Morrow et al. (2008) theorized that impairment of neural activity–dependent regulation of synapse development might be common to a group of mutations associated with autism and that loss of proper regulation of gene dosage may be a core genetic deficit in autism.
Zhao et al. (2007) proposed a template pattern of inheritance that needs to be filled in by a variety of spontaneous mutations, as well as specific modifier genes. The researchers argued that most forms of autism are the result of de novo mutations that appear in the parental germ line. These mutations affect males more than females. Resistant females carry the de novo mutation and pass the mutation to their children who are more likely to display the autism symptoms if they are male. The researchers claimed that the de novo and inherited mutation expressions are likely to account for the majority of cases of autism.
Identifying a limited set of genes or CNVs as crucial to autism. Ma et al. (2009), Wang et al. (2009), and Glessner et al. (2009) reported concurrent GWAS findings for common variants in the region of chromosome 5p14.1 in a combined cross-study sample of over 10,000 individuals. Wang et al. (2009) noted that SNPs with higher association P values 9dentified a linkage disequilibrium block within the intergenic region between CDH10 (cadherin 10) and CDH9 (cadherin 9) genes. Wang and colleagues (2009) further reported that a group of 25 related cadherin genes showed more significant association with ASD than all other genes, and also noted that combining the 25 cadherin genes with eight neurexin family genes (NRXN1 to NRXN3, CNTNAP1 to CNTNAP5) revealed a still more significant association with ASD. Wang et al. (2009) claimed that their data, combined with evidence for brain underconnectivity in ASD “convergently indicate that ASDs may result from structural and functional disconnection of brain regions that are involved in higher-order associations , suggesting that ASDs may represent a neuronal disconnection syndrome” (p. 5).
However, Wang and colleagues (2009) noted that several other loci contained SNPs with suggestive association signals, including 13q33.3, 14q21.1, LRFN5, Xp22.32, NLGN4X, and a SNP on the Y chromosome located within an ubiquitin gene. Additionally, Glessner and colleagues (2009) also reported evidence that CNVs in their ASD sample were associated with the genes outside the cadherin/neurexin gene families. They found that genes from the ubiquitin pathway (UBE3A, PARK2, RFWD2 and FBXO40) were another possible ASD susceptibility source.
Van der Swaag et al. (2009) explored CNVs in 105 ASD patients and 267 healthy individuals and found evidence for an association between ASD and genes RAI1, BRD1, and LARGE. They reported that a group of seven genes functioning in glycobiology that included the LARGE gene was associated with seven CNVs specifically identified in autism patients, where three of the seven CNVs were de novo in the patients. They argued that gains and losses of genes associated with glycobiology are important contributors to the development of ASD.
Wall et al. (2008) proposed that sets of genes that are under differential gene regulation in autism include multiple genes that influence transmission of nerve impulses, nervous system development, synaptic transmission, cell–cell signaling, brain development, generation of neurons, regulation of cell proliferation, cell migration, and homeostasis, cell morphogenesis, ion transport, and cell differentiation. Through an analytic network strategy they discovered 9 new candidate genes (SLC16A2, SLC6A8, OPHN1, FXN,AR, L1CAM, FLNA, MYO5A, PAFAH1B1) that were differentially expressed in “autism sibling disorders”, which they defined as Rett, Fragile X, Asperger Syndrome, mental retardation, Angelman syndrome, tuberous sclerosis, hypotonia, ataxia, hypoxia, seizure disorders, spasticity, and microcephaly.
Lintas and Persico (2009) argued that the six most important single gene mutation causes of autism would be HOXA1, NLGN3, NLGN4, NRXN1, PTEN, and SHANK3. Abrahams and Geschwind (2008) identified nine genes associated with autism (UBE3A; SHANK3; CNTNAP2; FMR1; DHCR7; MECP2; CNA1C; TSC1; and, TSC2), but they claimed that the most promising candidate gene for autism was RELN. In order to boost the possibility of finding inherited factors, Morrow et al. (2008) recruited a sample of individuals with autism-spectrum disorders whose parents shared ancestors. The researchers reported deletions in PCDH10 (protocadherin 10), NHE9, and ina researcher-defined potential new gene region DIA1 (deleted in autism1, or c3orf58).
Weiss and colleagues (2008) and Marshall and colleagues (2008) both reported that a chromosome number variant alteration at 16p11.2 was present in 1% of their autism spectrum samples. However, Weiss and colleagues (2008) also reported that that 1 of 648 patients with schizophrenia, 1 of 420 patients with bipolar disorder, 1 of 203 patients with ADHD, 1 of 748 patients with dyslexia and 1 of 3000 patients with panic disorder, anxiety, depression, or addiction had the same CNV at 16p11.2. As noted earlier, Marshall and colleagues (2008) reported finding 277 CNVs in 44% of ASD families, Christian et al. (2008), reported finding 51 CNVs in 12% of their sample of individuals with autism, and Sebat et al. (2007) reported that de novo CNVs occurred in 10% of their sample of individuals with autism.
Summary. Many gene variants and CNVs have been interpreted as “the set of genetic
sources” for autism. Evidence does suggest that Rett and Fragile X syndromes arise from gene variants that generate the symptoms of autism, that CNVs at 16p11.2 are likely to represent 1% of ASD (Weiss et al., 2008; Marshall et al., 1008), and that cadherin and neurexin genes variants may be linked to some, as yet unknown, percentage of ASD patients. These sources, though, cannot define autism as its own multi-gene mutation multi-CNV syndrome for two reasons. First, a majority of these single gene mutations and chromosome alterations are not unique to ASD and thus serve to penetrate other phenotypic syndromes. Second, as gene variants and CNVs account for less than 10% of ASD (Lintas & Persico, 2009), at present it is most likely that more complex epigenetic and multi-variant potential sources of ASD have yet to be discovered.
The Lack of Coverage of Gene and CNV Findings to Date
Despite the large number of individual gene and CNV variants discovered to date, nonetheless, as noted above, the coverage for autism is low. Lintas and Persico (2009) claimed that only 10% of autism could be linked to known genetic syndromes such as Fragile X, Rett Syndrome, neurofibromatosis, tuberous sclerosis, and Angelman Syndrome, or CNVs such as duplication of the maternal 15q11-13 region, deletions of chromosome 2q37, 7q31, 22q11, and microdeletions of chromosome 22q11.2. Sebat and colleagues (2007) reported that 10% of individuals diagnosed with autism were carriers of novel CNVs that were quite varied and included mutations of single genes. Abrahams and Geschwind (2008) argued that 1-2 % of ASD was linked to 15q11–15q13 duplications, 2-4% of ASD was linked to FXS and Angelman syndrome, and 1-3% of ASD was linked to all other chromosomal abnormalities, but Schanen (2006) argued that up to 5% of ASD was linked to 15q11–15q13 duplications alone. Weiss and colleagues (2008) and Marshall and colleagues (2008) both reported that CNV alteration at 16p11.2 was present in 1% of their autism spectrum sample.
Summary. The current coverage of gene system findings is generally thought to cover about 10% of gene-based autism. The missing “genetic dark matter” may include more single gene mutations and more CNVs. It is also possible that the complex processes in regulating gene expression will be a large component of the missing genetic matter.
Case-Carrier Examinations of Individual Chromosome Copy Number Variants
Abrahams and Geschwind (2008) argued that the autism phenotype can be examined for single gene mutations and CNVs based on a search for mutation specific phenotypic signatures. However, this method, as has already been demonstrated by work conducted to date, has yielded, at best, 10% coverage. Moreover, case carrier phenotype studies have begun to reveal the difficulty of attempting to find phenotypic signatures that are isolable to a particular single mutation.
Case carrier phenotype studies. Case carrier phenotype studies are best illustrated by the explorations of Mefford et al. (2008) and Brunetti-Pierri et al. (2008) of the phenotypes found for carriers of 1q21.1 duplications and deletions. As was outlined in Section Two above, these two groups of researchers screened large samples to find small groups of carriers of 1q21.1 duplications and deletions. Exploring these carriers in a case-by-case fashion revealed the massive pleiotrophic effects and dosage effects of the 1q21.1 chromosomal alterations. Both groups found differential microdeletion and microduplication dosage effects influencing head size. Both groups found a wide range of outcomes for the carriers. Duplications of 1q21.1 were associated with macrocephaly, autism, depression, anxiety, speech delay, learning disability, seizure disorder, macrocephaly, dysphagia, Chiari malformation, hydrocephalus, toe-walking, cryptorchidism, right-sided hyperpigmentation, Raynaud’s phenomenon, hypospadias, hypotonia, nerve paresis, fifth-finger clinodactyly, scoliosis, and advanced bone age. Deletions of 1q21.1 were associated with microcephaly, autism, schizophrenia, hallucinations, depression, anxiety, antisocial behavior, ADHD, learning problems, speech delay, seizure disorder, Chiari malformation, hydrocephalus, agenesis of the corpus callosum, trigonocephaly, scoliosis, 11 pairs of ribs, short stature, various eye disorders including cataracts, postaxial polydactyly, two- to three-toe syndactyly, precocious puberty, cryptorchidism, multiple sclerosis, hemangioma, and isolated heart defects.
Summary. Although these widely varied phenotypic findings are likely to reflect mutation across multiple genes within the 1q21.1 region, the phenotypic variation is extremely wide. This wide phenotypic variation is a clear example of the significant limitation to the goal of determining mutation specific phenotypic signatures.
Conclusion
The syndrome definition problem considered throughout this chapter is not a trivial one. The study of brain deficits and behavioral impairment in autism research has not produced any reasonable standard causal theory of autism (Happé & Ronald, 2008; Reiss, 2009; Waterhouse, 2008). The myriad of competing neural and behavioral theories of autism, while supported by evidence are, nonetheless, unsynthesized, and remain competing visions.
The increased prevalence of autism and the widening range of social action groups pushing for a “cure” for autism create a strong social force against admitting that autism is not one disorder. However, heightened public concern and attention increases the need to generate productive and predictive understanding of autism. Although all researchers understand that gene-based autism is a behavioral aggregation of phenotypic traits based on pleiotrophic and dosage and epigenetic effects, nonetheless research continues to focus on the genetics of “autism”. Real progress leading to treatment will not happen if genetic researchers continue to treat autism as a unitary syndrome (Stephan, 2008; Wall et al., 2008; Zhao et al., 2007). Researchers should concede a paradigm shift that posits that gene system variation in autism cannot be encompassed by an overarching model.
Veenstra-VanderWeele et al. (2004) argued that the integration of genetic findings for autism “may not be feasible” (p. 396) or may not happen because “some of those investigating autism… are sometimes too rigid and compulsive to make the creative leaps necessary to solve the puzzle of autism most efficiently and rapidly”(p. 396). But puzzle may be the wrong metaphor. A better metaphor may be portmanteau, or carryall. Given that only 10% of autism is tied to gene system variants, and these variants are associated with other phenotypic syndromes, it appears that autism is a portmanteau syndrome, a carryall phenotype carrying so many gene and non-gene sources (as well as so many neural deficits) that no solution is possible that could provide a unified basis for genetic screening, or possible future genetic treatment research.
Volkmar, State and Klin (2009) proposed, “The use of newly developed alternative, dimensional assessments may help disentangle much of the current confusion about ASDs broadly defined and their relationship to more strictly diagnosed autism. The ability to provide better sample specification, e.g., through additional ratings of levels of communicative or cognitive ability would greatly add to the diagnostic system” (p. 112). The genetic research conducted to date suggests, however, that “more strictly diagnosed autism”, will not prove helpful to future genetic research or treatment.
Worse still, Joober and Boksa (2009) cautioned that genome-wide association studies and CNV studies have only been able to examine a tiny portion of the human genome, and study of the entire human genome is needed to completely understand the role of genes in mental illnesses. Kapranov et al (2007) concluded that the possibility of thousands of additional coding sites and thousands of additional regulators of gene expression “significantly increase the diversity of both transcripts and proteins” (p. 417). Moreover, Janssens and van Duijn (2008) argued that simply finding all gene mutations and chromosomal alterations would not be enough. The study of the epigenetic code and gene networks is just beginning and it may require a complete understanding of the dynamic mechanisms in gene expression in order to understand the complete causal mechanisms of autism. Jannsens and Duijn (2008) proposed that disorders like autism might never be completely understood in individuals, as the unraveling of their complete unique causal pathways may prove impossible.
Gene-based autism is a portmanteau phenotype containing pleiotrophic expressions of many different single gene variants, of the dosage effects of many chromosome copy number duplications and deletions, and of presently unknown epigenetic effects. Genetic research based on the autism diagnostic phenotype and autism endophenotypes, although productive, continue to reveal increasing gene source heterogeneity. Two studies that explored the range of individual phenotypes for carriers of 1q21.1 duplications and deletions (Brunetti-Pierri et al., 2008; Mefford et al., 2008) offered a detailed description of dosage variation and apparent pleiotropic variation within a dosage level expressed in the phenotypes of these carriers. If this case-carrier analytic approach is applied to all the gene mutations and chromosome copy number variants associated with autism, it might be possible to ultimately generate a description of the majority of gene system variants that result in the autism phenotype. This description, however, will most certainly fail to find any overarching “genome variant story” for autism.
Similar to the case-carrier studies outlined above, Reiss (2009) argued that research in developmental disorders such as Fragile X, Rett syndrome and autism should consider using gene and chromosomal variants as a means to sort associated behaviors. Reiss also argued that the use of endophenotypes would be valuable. But, as reviewed in this chapter, to date no unifying endophenotypes have emerged. It may be that micro-endophenotypes based on narrow neural deficits will be discovered in future research. However, the evidence at present suggests that research employing diagnostic or non-diagnostic endophenotypes of autism will continue to uncover increasing gene system heterogeneity within endophenotypes. The portmanteau of autism must be carefully and completely unpacked to allow for the development of genetic testing for those individuals with gene-based autism, and for future possible case-specific treatment interventions.
References
Abrahams, B. S., & Geschwind, D. H. (2008). Advances in autism genetics: on the
threshold of a new neurobiology. Nature Review Genetics, 9, 341-355.
Alarcón, M., Abrahams, B. S., Stone, J. L., Duvall, J. A., Perederiy, J. V., Bomar, S. M.,
et al. (2008). Linkage, association, and gene-expression analyses identify
CNTNAP2 as an autism-susceptibility gene. American Journal of Human
Genetics, 82, 150-159.
Amaral, D. G., Schumann, C. M., & Nordahl, C. W. (2008). Neuroanatomy of autism.
Trends in Neuroscience, 31, 137-145.
Arking, D. E., Cutler, D. J., Brune, C. W., Teslovich, T. M., West, K., Ikeda, M., et al.
(2008). A common genetic variant in the neurexin superfamily member
CNTNAP2 increases familial risk of autism. American Journal of Human
Genetics, 82, 160–164.
Bakkaloglu, B., O’Roak, B. J., Louvi, A., Gupta, A. R., Abelson, J. F., Morgan, T. M., et
al. (2008). Molecular cytogenetic analysis and resequencing of contactin
associated proteinlike 2 in autism spectrum disorders. American Journal of Human Genetics, 82, 165-173.
Bearden, C. E., Glahn, D. C., Lee, A. D., Chiang, M-C., van Erp, T. G. M., Cannon, T.
D., et al. (2008). Neural phenotypes of common and rare genetic variants. Biological Psychology 79, 43–57.
Brunetti-Pierri, N., Berg, J. S., Scaglia, F., Belmont, J., Bacino, C. A., Sahoo, T., et al.
(2008). Recurrent reciprocal 1q21.1 deletions and duplications associated with
microcephaly or macrocephaly and developmental and behavioral abnormalities.
Nature Genetics, 40, 1466-1471.
Butler, M. G., Dasouki, M. J., Zhou, X. P., Talebizadeh, Z., Brown, T., Takahashi, T. N.,
et al. (2005)Subset of individuals with autism spectrum disorders and extreme
macrocephaly associated with germline PTEN tumour suppressor gene mutations.
Journal of Medical Genetics, 42, 318–321.
Buxbaum, J. D., Cai, G., Chaste, P., Nygren, G., Goldsmith, J., Reichert, J., et al.
(2007a). Mutation screening of the PTEN gene in patients with autism spectrum
disorders and macrocephaly. American Journal of Medical Genetics Part B
Neuropsychiatric Genetics, 144B, 484–491.
Buxbaum, J. D., Cai, G., Nygren, G., Chaste, P., Delormes, R., Goldsmith, J., et al.
(2007b). Mutation analysis of the NSD1 gene in patients with autism spectrum
disorders and macrocephaly. BMC Medical Genetics, 8, 68.
Cai, G., Edelmann, L., Goldsmith, J. E., Cohen, N., Nakamine, A., Reichert, J. G., et al.
(2008). Multiplex ligation-dependent probe amplification for genetic screening in
autism spectrum disorders: Efficient identification of known microduplications
and identification of a novel microduplication in ASMT.BMC Medical
Genomics, 1, 50.
Christian, S. L., Brune, C. W., Sudi, J., Kumar, R. A., Liu, S., Karamohamed, S., et al.
(2008). Novel Submicroscopic chromosomal abnormalities detected in autism
spectrum disorder. Biological Psychiatry, 63, 1111–1117.
Crespi, B., & Badcock, C. (2008). Psychosis and autism as diametrical disorders of the
social brain. Behavioral and Brain Sciences, 31, 241–320.
Duvall, J. A., Lu, A., Cantor, R. M., Constantino, J. N., & Geschwind, D. H. (2007). A
quantitative trait locus analysis of social responsiveness in multiplex autism
families. American Journal of Psychiatry, 164, 656–662.
Folstein, S. (2006). The clinical spectrum of autism. (2006). Clinical Neuroscience
Research, 113-117.
Freitag, C. M. (2007). The genetics of autistic disorders and its clinical relevance: A
review of the literature. Molecular Psychiatry, 12, 2-22.
Glessner, J. T., Wang, K., Cai, G., Korvatska, O., Kim, C.E., Wood, S., et al. (2009). Autism genome-wide copy number variation reveals ubiquitin and neuronal genes. Nature, 459, 569-573.
Goos, L. M. (2008). Imprinting and psychiatric genetics: Beware the diagnostic
phenotype. Behavioral and Brain Sciences, 31, 271-272.
Green, L., Fein, D., Modahl. C., Feinstein C., Waterhouse, L., Morris, M., et al. (2001)
Oxytocin and autistic disorder: Alterations in peptide forms. Biological
Psychiatry, 50, 609–613.
Happé, F., & Ronald, A. (2008). ‘Fractionable Autism Triad’: A Review of Evidence
from Behavioural, Genetic, Cognitive and Neural Research. Neuropsychology
Review, 18, 287-304.
Happé, F., Ronald, A., & Plomin, R. (2006). Time to give up on a single explanation for
autism. Nature Neuroscience, 9(10), 1218–1220.
Happel, N. & Doenecke, D. (2009). Histone H1 and its isoforms: contribution to
chromatin structure and function. Gene, 431(1-2), 1-12
Hogart, A., Nagarajan, R. P., Patzel, K. A., Yasui, D. H., & Lasalle, J. M. (2007).15q11–
13 GABAA receptor genes are normally biallelically expressed in brain yet are
subject to epigenetic dysregulation in autism-spectrum disorders. Human
Molecular Genetics, 16, 691–703.
Janssens, A. C., & van Duijn, C. M. (2008). Genome-based prediction of common
diseases: advances and prospects. Human Molecular Genetics, 17(R2), R166-173.
Jones, J. R., Skinner, C., Friez, M. J., Schwartz, C. E., & Stevenson, R. E. (2007).
Hypothesis: dysregulation of methylation of brain-expressed genes on the X
chromosome and autism spectrum disorders. American Journal of Medical
Genetics Part A, 146A (17), 2213-2220.
Joober, R., & Bokva, P. (2009). A new wave in the genetics of psychiatric disorders:
the copy number variant tsunami. Journal of Psychiatry and Neuroscience, 34,
55-59.
Kaduake, S. & Blobe., G. A. (2009). Chromatin loops in gene regulation. Biochimica et
Biophysica Acta, 1789, 17–25.
Kaminsky, Z. A., Tang, T., Wang, S-C., Ptak, C., Oh, G. H., Wong, A. H., et al. (2009).
DNA methylation profiles in monozygotic and dizygotic twins. Nature Genetics,
41, 240-245.
Kapranov, P., Willingham, A.T., & Gingeras, T. R. 2007. Genome-wide transcription
and the implications for genomic organization. Nature Review Genetics, 8,413– 423.
Khaja, R., Zhang, J., MacDonald, J. R., He, Y., Joseph-George, A. M., Wei, J., et al.
(2006). Genome assembly comparison to identify structural variants in the human
- Nature Genetics, 38, 1413-1418.
Kramer, J. M., & van Bokhoven, H. (2009). Genetic and epigenetic defects in mental
retardation. The International Journal of Biochemistry and Cell Biology, 41, 96–
Lam, K. S. L., Bodfish, J. W., & Piven, J. (2008). Evidence for three subtypes of
repetitive behavior in autism that differ in familiality and association with other
symptoms. Journal of Child Psychology and Psychiatry, 49, 1193–1200.
Lintas, C., & Persico, A. M. (2009). Autistic phenotypes and genetic testing: state-of-the
art for the clinical geneticist. Journal of Medical Genetics, 46, 1-8.
Liu, X. Q., Paterson, A. D., Szatmari, P., & The Autism Genome Project Consortium.
(2008). Genome-wide linkage analyses of quantitative and categorical autism subphenotypes. Biological Psychiatry, 64, 561-570.
Loat, C. S., Curran, S., Lewis, C. M. , Duvall, J., Geschwind, D., Bolton, P., et al. (2008).
Methyl-CpG-binding protein 2 polymorphisms and vulnerability to autism.
Genes, brain, and behavior, 7, 754-760.
Ma, D., Salyakina, D., Jaworski, J. M., Konidari, I., Whitehead, P. L., Andersen, A. N., et
al. (2009). A genome-wide study of autism reveals a common novel risk locus at 5p14.1. Annals of Human Genetics, 73, pt. 3, 263-273.
Maher, B. (2008). The case of the missing heritability. Nature, 456, 6, 18-21.
Mandy, W. P., & Skuse, D. H. (2008). Research review: What is the association between
the social-communication element of autism and repetitive interests, behaviours,
and activities? Journal of Child Psychology and Psychiatry, 49, 795-808.
Marshall, C. R., Noor. A.,Vincent, J. B., Lionel, A. C., Feuk, L., Skaug, J., et al. (2008).
Structural variation of chromosomes in autism spectrum disorder. American
Journal of Human Genetics, 82, 477–488.
Mefford, H., Sharp, A., Baker, C., Itsara, A., Jiang, Z., Buysse, K., et al. (2008).
Recurrent rearrangements of chromosome 1q21.1 and variable pediatric
phenotypes. The New England Journal of Medicine, 359, 16, 1685-1699.
Mehler, M. F., & Purpura, D. P. (2008). Autism, fever, epigenetics and the locus
coeruleus. Brain Research Reviews, in press.
Miller, D. T., Shen, Y., Weiss, L. A., Korn J, Anselm, I., Bridgemohan, C., et al. (2008).
Microdeletion/duplication at 15q13.2q13.3 among individuals with features of
autism and other neuropsychiatric disorders. Journal of Medical Genetics, in
Modahl, C., Green, L., Fein, D., Morris, M., Waterhouse, L., Feinstein, C., et al. (1998).
Plasma oxytocin levels in autistic children. Biological Psychiatry 43, 270–277.
Monteggia, L. M., & Kavalali, E. T. (2009). Rett syndrome and the impact of MeCP2
associated transcriptional mechanisms on neurotransmission. Biological
Psychiatry, 65, 204-210.
Morrow, E. M. , Yoo, S. Y., Flavell, S. W., Kim, T. K., Lin, Y., Hill, R. S., et al. (2008).
Identifying autism loci and genes by tracing recent shared ancestry. Science,
321(5886), 218-223.
Ramocki, M. B., & Zoghbi, H. Y. (2008) Failure of neuronal homeostasis results in common neuropsychiatric phenotypes. Nature, 455(7215), 912-918.
Redcay, E., & Courchesne, E. (2005). When is the brain enlarged in autism? A meta-
analysis of all brain size reports. Biological Psychiatry, 58, 1-9.
Reid, G., Gallais, R., & Metiviér, T. (2009). Marking time: The dynamic role of
chromatin and covalent modification in transcription. International Journal of
Biochemistry and Cell Biology, 41, 155-163.
Reiss, A. (2009). Childhood developmental disorders: an academic and clinical
convergence point for psychiatry, neurology, psychology and pediatrics. Journal
of Child Psychology and Psychiatry, 50, 87–98.
Schanen, N. C. (2006). Epigenetics of autism spectrum disorders. Human Molecular
Genetics, 15, R138–R150.
Sebat, J., Lakshmi, B., Malhotra, D., Troge, J., Lese-Martin, C., Walsh, T., et al.,(2007).
Strong association of de novo copy number mutations with autism. Science, 316
(5823), 445–449.
Seringhaus, M. & Gerstein, M. (2008). Genomics confounds gene classification.
American Scientist, 96, 466-473.
Spence, S. J., Cantor, R. M., Chung, L., Kim, S., Geschwind, D. H., & Alarcon, M.
(2006). Stratification based on language-related endophenotypes in autism:
Attempt to replicate reported linkage. American Journal of Medical Genetics Part
| B Neuropsychiatric Genetics, 141B, 591-598. |
Stephan, D. A. (2008). Unraveling autism. The American Journal of Human Genetics, 82,
7–9.
Stanfield, A. C., McIntosh, A. M., Spencer, M. D., Philip, R., Gaur, S., & Lawrie, S. M.
(2008). Towards a neuroanatomy of autism: A systematic review and meta-
analysis of structural magnetic resonance imaging studies. European Psychiatry
23, 289-299.
Swanberg, S. E., Nagarajan, R. P., Peddada, S., Yasui, D. H., & LaSalle, J. M. (2009).
Reciprocal co-regulation of EGR2 and MECP2 is disrupted in Rett syndrome and autism. Human Molecular Genetics, 18(3), 524-534.
Tsuchiya, K. J., Hashimoto, K., Iwata, Y., Tsujii, M., Sekine, Y., Sugihara, G., et al.
(2008). Decreased serum levels of platelet-endothelial adhesion molecule
(PECAM-1) in subjects with high-functioning autism: a negative correlation with
head circumference at birth. Biological Psychiatry, 62, 1056-1058.
van der Zwaag, B., Franke, L., Poot, M., Hochstenbach, R., Spierenburg, H. A., Vorstman, J. A., et al. (2009). Gene-network analysis identifies susceptibility genes related to glycobiology in autism. PLoS ONE, 4(5), e5324.
Veenstra-VanderWeele, J., Christian, S. L., & Cook, Jr., E. H. (2004). Autism as a
paradigmatic complex genetic disorder. Annual Review of Genomics and Human
Genetics, 5, 379-405.
Vernes, S. C., Newbury, D. F., Abrahams, B., Winchester, L., Nicod, J., Groszer, M., et
al. (2008). A functional genetic link between distinct developmental language
disorders. The New England Journal of Medicine, 359, 22, 2337-2345.
Volkmar, F. R., State, M., & Klin, A. (2009). Autism and autism spectrum disorders:
diagnostic issues for the coming decade. Journal of Child Psychology and
Psychiatry, 50, 108–115.
Wall, D. P., Esteban, T. F., DeLuca, M., Huyck, T. Monaghan, N., Velez de Mendizabal,
N., et al. (2008). Comparative analysis of neurological disorders focuses genome-
wide search for autism genes. Genomics, 93,120-129.
Wang, K., Zhang, H., Ma, D., Bucan, M., Glessner, J. T., Abrahams, B. S., et al.
(2009).Common genetic variants on 5p14.1 associate with autism spectrum
disorders. Nature, 459, 528-533.
Waterhouse, L. (2008). Autism overflows: increasing prevalence and proliferating
theories. Neuropsychology Review, 18, 273-286.
Weiss, L. A., Shen, Y., Korn, J. M., Arking, D. E., Miller, D. T., Fossdal, R., et al.
(2008). Association between microdeletion and microduplication at 16p11.2 and
autism. The New England Journal of Medicine, 358, 667–675.
Williams, C. A., Dagli, A., & Battaglia, A. (2008). Genetic disorders associated with
macrocephaly. American Journal of Medical Genetics Part, 146A, 2023–2037.
Yrigollen, C. M., Han, S. S., Kochetkova, A., Babitz, T., Volkmar, F. R., Leckman, J. F.,
et al. (2008). Genes controlling affiliative behavior as candidate genes for autism.
Biological Psychiatry, 63, 911-916.
Zhao, X., Leotta, A., Kustanovich, V., Lajonchere, C., Geschwind, D. H., Law, K., et al.
(2007). A unified genetic theory for sporadic and inherited autism. Proceedings of
the National Academy of Sciences U S A, 104, 12831-12836.
Zingerevich, C., Greiss-Hess, L., Lemons-Chitwood, K., Harris, S. W., Hessl, D., Cook,
K., et al. (2009). Motor abilities of children diagnosed with fragile X syndrome
with and without autism. Journal of Intellectual Disability Research, 53, 11-18.