**2. Genetics**

Autistic disorder is a multifactorial genetic disorder not following classical Mendelian inher‐ itance. Impairment in social interaction and verbal communication as well as genetic differ‐ entiation in rigid-repetitive behaviours indicates that different features in autistic disorder

© 2013 Guney and Iseri; licensee InTech. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2013 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

may be caused by different genes associated with distinct brain regions and be related to cognitive impairment and functional abnormalities [13].

Another region related to autism is a deletion region located on chromosome 16p11. This re‐ gion has also been demonstrated to be in relationship with Asperger Syndrome, mental re‐

Genetic and Environmental Factors in Autism

http://dx.doi.org/10.5772/53295

323

It has been showed that, in individuals with autism, there is a significant increase in the fre‐ quency of allelic variations of HOXA1 gene (7p15). HOXA1 and HOXB1, which have a criti‐ cal role for development of fetal caudal medullary structures, are only expressed at the third week following fertilization, a period when neural tube is formed, and they appear to be partly associated with development of superior olivary, facial and abducens nuclei. It has been suggested that HOXA1 has a role in autism tendency and is associated with early phase of brain stem development in autism etiology [16, 35]. On the other hand, there are studies where no significant association with HOXA1 gene variants and autism could be

Engrailed-2 (EN-2) which is the human homologue of drosophila engrailed gene and located on the long arm of Chromosome 7 (7q36) is a homeobox gene having a critical role in mid‐ brain and cerebellar development. Temporal and spatial pattern of engrailed gene expres‐ sion occurs simultaneously with the development of cerebellar precursor cells. Thus, it has been suggested to be important to determine correct cell number in cerebellum [38]. Petit and his colleagues (1995) reported a significant association between *Pvull* polymorphism at the 5' region of EN-2 gene and autism [39]. However, this association could not be con‐

MET oncogene coding pleiotropic MET receptor thyrosine kinase is located on the long arm of Chromosome 7. MET signalization has a role in neocortex and cerebellar growth and ma‐ turation, and immune functions. MET gene and its ligand, hepatocyte growth factor (HGF), have been related to autism. Studies conducted by Campbell and his colleagues ([41,42] showed that C allele in the promoter region of MET gene decreases MET promoter activity

Another gene on Chromosome 7 is CNTNAP2 (contactin-associated-protein-like 2) gene. CNTNAP2 gene has been associated in various studies with autism, language delay, and

FOXP2 (forkhead box P2), a forkhead transcription factor gene, is a member of family fork‐ head known as the key regulators of embryogenesis; it encodes a transcription factor con‐ taining polyglutamine and is associated with development of lingual functions. In a study in Chinese society, FOXP2 gene located in the 7q31 region was linked with autism pathogene‐

Another gene investigated for autism relationship is Wingless-Int (Wnt2) gene located on the long arm of Chromosome 7 (7q31-33). Wnt genes encode glycoproteins rich in cysteine, which regulate various cellular movements during the embryonic development [49]. It has been shown that Wnt has a role in regulation of activity-dependent dendritic branching in hippocampal pyramidal neurons [50]. Wnt2 gene was linked with autism in a study by Was‐

by two fold and decreased MET gene expression is associated with autism tendency.

sis [46]. However, other studies did not replicate these findings [47, 48].

tardation, and developmental abnormalities [33, 34].

demonstrated [36, 37].

epilepsy [43-45].

sink and his colleagues [51].

firmed in a later family study [40].

Genetic studies in the field of autistic disorder have mainly focused on molecular genetic studies, assessment of chromosomal abnormalities, twin studies and family studies. In fami‐ lies having an autistic child the recurrence rate has been reported as 3-8% [14, 15, 16]. The studies on twins and adopted children are important in identifying the actual importance of genetic factors. Concordance among twins enables to measure heritability, and thus to as‐ sess what percentage of the phenotype is affected by genetic factors. Monozygotic (identical) twins share 100% of the genetic material whereas dizygotic (fraternal) twins share 50% of the genetic material. Monozygotic twins' higher rate of concordance compared to dizygotic twins may be used for calculation of heritability. Twin studies generally showed a higher concordance rate for monozygotic twins compared to dizygotic twins. The concordance rate of monozygotic twins is at least 60% when diagnostic criteria for autism (DSM-IV) are used whereas the number is as high as 71% for autism spectrum and 92% for a broader spectrum of verbal/social interaction disorders [11, 12, 16, 17]. On the other hand, the concordance rate of dizygotic twins has been reported as 1-30% [9, 17-20]. Twin studies demonstrated an average autism inheritance of 90% [21]. On the basis of these studies autism is considered to be among the most inherited psychiatric diseases [22, 23].

Although autism has a high inheritance rate, its mode of inheritance remains unclear. Multigene interactions and multiple loci are believed to play role in genetic susceptibility to the dis‐ ease [24]. There are 3 basic approaches in this area: 1) in whole genome scanning method, it is aimed to predict the localization of a disease, about chromosomal localization of which we have no preliminary information, by starting from common genetic determinants in a com‐ munity composed of multiplex families (families with more than one involved member). 2) cy‐ togenetic studies guide molecular studies by showing inherited or de novo chromosomal anomalies in involved persons or families. 3) candidate gene studies examine the relationship of genes known to affect brain development in associated regions or alternatively, a selected precursor gene considered to hypothetically contribute to autism pathogenesis.

It has been demonstrated that structural chromosomal variations comprising also copy number variations play an important role in etiology of autism. De novo copy number varia‐ tions have been identified in 7-10% of sporadic autism cases [25, 26].

In studies employing genome scanning method to reveal genetic etiology of autism, cogent evidence for an association with chromosomes 2, 7, 1, and 17, especially long arm of chro‐ mosomes 2 (2q) and 7 (7q) has been obtained. Other chromosomes less associated with au‐ tism are chromosomes 1, 9, 13, 15, 19, 22, and X chromosome [14, 16, 27]. Although a lot of genomic regions have been explored for etiology, consistent results for a limited number of regions such as 7q11, 7q31, 22q11 have been obtained [16, 28, 29]. Particularly 15q11-q13 re‐ gion on chromosome 15 has been widely related to autism. It has been suggested that dupli‐ cations in this region of chromosome 15 may contribute to autism development. There exist in this area a series of potential candidate genes containing gamma aminobutyric acid A (GABAA) receptor gene complex [30]. These duplications inherited maternally have been re‐ ported to be present in 1-3% of individuals with idiopathic autism [31, 32].

Another region related to autism is a deletion region located on chromosome 16p11. This re‐ gion has also been demonstrated to be in relationship with Asperger Syndrome, mental re‐ tardation, and developmental abnormalities [33, 34].

may be caused by different genes associated with distinct brain regions and be related to

Genetic studies in the field of autistic disorder have mainly focused on molecular genetic studies, assessment of chromosomal abnormalities, twin studies and family studies. In fami‐ lies having an autistic child the recurrence rate has been reported as 3-8% [14, 15, 16]. The studies on twins and adopted children are important in identifying the actual importance of genetic factors. Concordance among twins enables to measure heritability, and thus to as‐ sess what percentage of the phenotype is affected by genetic factors. Monozygotic (identical) twins share 100% of the genetic material whereas dizygotic (fraternal) twins share 50% of the genetic material. Monozygotic twins' higher rate of concordance compared to dizygotic twins may be used for calculation of heritability. Twin studies generally showed a higher concordance rate for monozygotic twins compared to dizygotic twins. The concordance rate of monozygotic twins is at least 60% when diagnostic criteria for autism (DSM-IV) are used whereas the number is as high as 71% for autism spectrum and 92% for a broader spectrum of verbal/social interaction disorders [11, 12, 16, 17]. On the other hand, the concordance rate of dizygotic twins has been reported as 1-30% [9, 17-20]. Twin studies demonstrated an average autism inheritance of 90% [21]. On the basis of these studies autism is considered to

Although autism has a high inheritance rate, its mode of inheritance remains unclear. Multigene interactions and multiple loci are believed to play role in genetic susceptibility to the dis‐ ease [24]. There are 3 basic approaches in this area: 1) in whole genome scanning method, it is aimed to predict the localization of a disease, about chromosomal localization of which we have no preliminary information, by starting from common genetic determinants in a com‐ munity composed of multiplex families (families with more than one involved member). 2) cy‐ togenetic studies guide molecular studies by showing inherited or de novo chromosomal anomalies in involved persons or families. 3) candidate gene studies examine the relationship of genes known to affect brain development in associated regions or alternatively, a selected

It has been demonstrated that structural chromosomal variations comprising also copy number variations play an important role in etiology of autism. De novo copy number varia‐

In studies employing genome scanning method to reveal genetic etiology of autism, cogent evidence for an association with chromosomes 2, 7, 1, and 17, especially long arm of chro‐ mosomes 2 (2q) and 7 (7q) has been obtained. Other chromosomes less associated with au‐ tism are chromosomes 1, 9, 13, 15, 19, 22, and X chromosome [14, 16, 27]. Although a lot of genomic regions have been explored for etiology, consistent results for a limited number of regions such as 7q11, 7q31, 22q11 have been obtained [16, 28, 29]. Particularly 15q11-q13 re‐ gion on chromosome 15 has been widely related to autism. It has been suggested that dupli‐ cations in this region of chromosome 15 may contribute to autism development. There exist in this area a series of potential candidate genes containing gamma aminobutyric acid A (GABAA) receptor gene complex [30]. These duplications inherited maternally have been re‐

precursor gene considered to hypothetically contribute to autism pathogenesis.

tions have been identified in 7-10% of sporadic autism cases [25, 26].

ported to be present in 1-3% of individuals with idiopathic autism [31, 32].

cognitive impairment and functional abnormalities [13].

322 Recent Advances in Autism Spectrum Disorders - Volume I

be among the most inherited psychiatric diseases [22, 23].

It has been showed that, in individuals with autism, there is a significant increase in the fre‐ quency of allelic variations of HOXA1 gene (7p15). HOXA1 and HOXB1, which have a criti‐ cal role for development of fetal caudal medullary structures, are only expressed at the third week following fertilization, a period when neural tube is formed, and they appear to be partly associated with development of superior olivary, facial and abducens nuclei. It has been suggested that HOXA1 has a role in autism tendency and is associated with early phase of brain stem development in autism etiology [16, 35]. On the other hand, there are studies where no significant association with HOXA1 gene variants and autism could be demonstrated [36, 37].

Engrailed-2 (EN-2) which is the human homologue of drosophila engrailed gene and located on the long arm of Chromosome 7 (7q36) is a homeobox gene having a critical role in mid‐ brain and cerebellar development. Temporal and spatial pattern of engrailed gene expres‐ sion occurs simultaneously with the development of cerebellar precursor cells. Thus, it has been suggested to be important to determine correct cell number in cerebellum [38]. Petit and his colleagues (1995) reported a significant association between *Pvull* polymorphism at the 5' region of EN-2 gene and autism [39]. However, this association could not be con‐ firmed in a later family study [40].

MET oncogene coding pleiotropic MET receptor thyrosine kinase is located on the long arm of Chromosome 7. MET signalization has a role in neocortex and cerebellar growth and ma‐ turation, and immune functions. MET gene and its ligand, hepatocyte growth factor (HGF), have been related to autism. Studies conducted by Campbell and his colleagues ([41,42] showed that C allele in the promoter region of MET gene decreases MET promoter activity by two fold and decreased MET gene expression is associated with autism tendency.

Another gene on Chromosome 7 is CNTNAP2 (contactin-associated-protein-like 2) gene. CNTNAP2 gene has been associated in various studies with autism, language delay, and epilepsy [43-45].

FOXP2 (forkhead box P2), a forkhead transcription factor gene, is a member of family fork‐ head known as the key regulators of embryogenesis; it encodes a transcription factor con‐ taining polyglutamine and is associated with development of lingual functions. In a study in Chinese society, FOXP2 gene located in the 7q31 region was linked with autism pathogene‐ sis [46]. However, other studies did not replicate these findings [47, 48].

Another gene investigated for autism relationship is Wingless-Int (Wnt2) gene located on the long arm of Chromosome 7 (7q31-33). Wnt genes encode glycoproteins rich in cysteine, which regulate various cellular movements during the embryonic development [49]. It has been shown that Wnt has a role in regulation of activity-dependent dendritic branching in hippocampal pyramidal neurons [50]. Wnt2 gene was linked with autism in a study by Was‐ sink and his colleagues [51].

Reelin is an important extracellular matrix glycoprotein that has an important role in devel‐ opment of neuronal migration, lamination, and connection during embryonic brain develop‐ ment and is associated with a signal pathway forming the basis of neuro-conduction, memory formation, and synaptic plasticity [52]. It is responsible for lamination in embryonic period whereas it has a role in cell signalization and synaptic plasticity in adulthood period [53]. Decrease in reelin expression has been associated with autism. RELN gene, which is lo‐ cated in 7q22 region and encodes reelin protein which is important in neurodevelopment, involves a polymorphic GGC repeat in 5' region. Long GGC alleles of RELN gene cause blunt gene expression; therefore, they are considered to be linked with autism [52]. There are studies reporting a significant relationship between RELN alleles with larger numbers of CGG repeats and autism [52, 54] while there are also negative studies in terms of such a rela‐ tionship [55]. Besides the genetic complexity in the etiopathogenesis of this disorder, nonreplication of the results of different studies should also be taken into consideration.

complex in development of autism spectrum disorders provide potential evidence for syn‐

Genetic and Environmental Factors in Autism

http://dx.doi.org/10.5772/53295

325

A large-scale study by Wang and his colleagues [65] revealed a significant relationship be‐ tween a single nucleotide polymorphism located in the 5p14.1 region and autism spectrum disorders. The associated single nucleotide polymorphism is located in a region placed be‐ tween cadherin 10 (CDH 10) and cadherin 9 (CDH 9) genes. CDH 9 and CDH 10 encode type II classic cadherins of the cadherin family, which are transmembraneous glycoprotiens responsible for calcium-dependent cell-cell adhesion. This finding shows the role of neuro‐

Neurotrophins have many functions such as neuronal survival, target innervation, and synap‐ togenesis in development of peripheral and central nervous system. Neurotrophins exert their biologic functions by binding to a Trk tyrosine kinase receptor which is a high-affinity recep‐ tor. Neurotrophin family has 4 members. These are nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), neurotrophin-3 and neurotrophin-4. BDNF is the most important member of neurotrophin family. BDNF has many roles in neuronal differentiation such as neu‐ ronal survival, dendritic and axonal growth/branching, synapse formation, and neuronal plas‐ ticity [66, 67]. Various studies have investigated the relationship between BDNF gene and autism. Nishimura and colleagues [68] detected an increase in BDNF expression in autistic in‐ dividuals. Subsequent studies confirmed the potential role of BDNF gene mutations in autism pathogenesis [69]. A recent study in which serum BDNF levels significantly increased in autis‐ tic children found no significant impact of genetic variations of BDNF gene on autism risk; however, a significant relationship between neurotrophic tyrosine kinase receptor type 2 (NTRK2) and autism was reported [70]. A large-scale study on patients diagnosed with autism spectrum disorder and mental retardation without autism diagnoses showed that, when com‐ pared to control group, autism spectrum disorders and mental retardation had a significant in‐ crease in serum neurotrophin 4 and BDNF (both are Trk B ligands) [71, 72]. On the other hand, no changes were observed in NGF (trk A ligand) and neurotrophin 3 (Trk C ligand) levels. In light of these findings, it has been suggested that trkB ligands may be overexpressed or secret‐ ed in central nervous system of autistic or mental retarded children during infantile period. It has also been suggested that the effect of BDNF and neurotrophin- 4 on activity-dependent dendritic growth and branching [66] may be related to early and transient brain development seen in autistic infants [67, 73]. This increase in BDNF expression and/or secretion was suggest‐ ed to be linked with the role of Metil-CpG-binding protein 2 (MeCP2) gene in BDNF transcrip‐ tion [74]. A mutation in MeCP2 gene encoding a protein functioning as a general transcriptional receptor is responsible for Rett Syndrome. It has been shown that MeCP2 selec‐ tively bind to BDNF promoter III and suppresses BDNF gene expression. MeCP2 has an impor‐ tant role in regulation of neuronal activity [75]. It has been suggested that MeCP2 mutations located on Xq28 locus may be a risk factor for autism by affecting BDNF expression and den‐ dritic differentiation in cortex. In a study investigating MeCP2 gene mutation in autistic indi‐

aptic alterations in etiology of the disorder.

nal cellular adhesion molecules in autism pathogenesis [65].

viduals for that purpose, 2 girls exhibited de novo mutations [67, 76].

Another gene linked with autism is the Fragile X mental retardation 1 (FMR1) gene encoding Fragile X mental retardation protein (FMRP). FMR1 is associated with autism secondary to

Neuroligins are cellular adhesion molecules located at the postsynaptic side of the synapse. Neuroligins and neuroxins, neuronal cell surface proteins, form an asymmetrical intercellular connection by adhering to each other. Interaction of neuroligins with beta neuroxins forms functional synapses [56]. Neuroligin family is composed of 5 members, i.e. NLGN1, NLGN2, NLGN3, NLGN4, and NLGN4Y. Although all of the neuroligin family is linked with autism spectrum disorder [57], the most robust evidence comes from NLGN3 (Xq13) and NLGN4 (Xp22.3) genes. Jamain and colleagues [58] found that mutations in NLGN3 and NLGN4, two X-linked neuroligin genes, are associated with autism spectrum disorders [58]. Following this, it has been demonstrated that a 2-base-pair deletion in NLGN4 gene causes premature stop co‐ don in mental-retarded men with or without autism. This finding indicates that NLGN4 gene is not only associated with autism, but also with mental retardation [59]. Since mutations in neu‐ roligin genes impair the functions of synaptic cell adhesion molecules, they are considered to be related with autism and neurodevelopmental defects in mental retardation [60]. Since neu‐ roligins are abundant particularly in excitatory synapses, a defect in synaptogenesis has been suggested to result in derangement in cognitive development and communication [59]. None‐ theless, some other studies revealed negative results [61].

Genetic studies examining the relationship of neuroxins, the connection partners of neuroli‐ gins, with autism revealed that a mutation in neuroxin 1beta gene results in autism susceptibil‐ ity [62]. Structural variants of neuroxin 1alpha gene have also been linked with autism [63].

Another protein adhering to neuroligins is SHANK3. Some forms of autism are considered to stem from a single gene, and particularly from a rare allele having a major effect. Doctor Joseph Buxbaum has reported that one of these genes is SHANK3 gene located on Chromo‐ some 22 (22q13) which is responsible for 1% of autism and some forms of mental retarda‐ tion, microcephaly, and delay in expressive language [34]. SHANK proteins are believed to be the primary regulator of postsynaptic density thanks to their ability to form multimeric complexes with postsynaptic receptors, signal molecules, and cellular skeleton proteins found in dendritic spikes. Postsynaptic density is the measurement of how synapses are linked to each other. A mutation in SHANK3 gene has been reported to be related with au‐ tism spectrum disorders [64]. Role of various mutations in Neuroligin/neuroxin/SHANK3 complex in development of autism spectrum disorders provide potential evidence for syn‐ aptic alterations in etiology of the disorder.

Reelin is an important extracellular matrix glycoprotein that has an important role in devel‐ opment of neuronal migration, lamination, and connection during embryonic brain develop‐ ment and is associated with a signal pathway forming the basis of neuro-conduction, memory formation, and synaptic plasticity [52]. It is responsible for lamination in embryonic period whereas it has a role in cell signalization and synaptic plasticity in adulthood period [53]. Decrease in reelin expression has been associated with autism. RELN gene, which is lo‐ cated in 7q22 region and encodes reelin protein which is important in neurodevelopment, involves a polymorphic GGC repeat in 5' region. Long GGC alleles of RELN gene cause blunt gene expression; therefore, they are considered to be linked with autism [52]. There are studies reporting a significant relationship between RELN alleles with larger numbers of CGG repeats and autism [52, 54] while there are also negative studies in terms of such a rela‐ tionship [55]. Besides the genetic complexity in the etiopathogenesis of this disorder, non-

replication of the results of different studies should also be taken into consideration.

theless, some other studies revealed negative results [61].

324 Recent Advances in Autism Spectrum Disorders - Volume I

Neuroligins are cellular adhesion molecules located at the postsynaptic side of the synapse. Neuroligins and neuroxins, neuronal cell surface proteins, form an asymmetrical intercellular connection by adhering to each other. Interaction of neuroligins with beta neuroxins forms functional synapses [56]. Neuroligin family is composed of 5 members, i.e. NLGN1, NLGN2, NLGN3, NLGN4, and NLGN4Y. Although all of the neuroligin family is linked with autism spectrum disorder [57], the most robust evidence comes from NLGN3 (Xq13) and NLGN4 (Xp22.3) genes. Jamain and colleagues [58] found that mutations in NLGN3 and NLGN4, two X-linked neuroligin genes, are associated with autism spectrum disorders [58]. Following this, it has been demonstrated that a 2-base-pair deletion in NLGN4 gene causes premature stop co‐ don in mental-retarded men with or without autism. This finding indicates that NLGN4 gene is not only associated with autism, but also with mental retardation [59]. Since mutations in neu‐ roligin genes impair the functions of synaptic cell adhesion molecules, they are considered to be related with autism and neurodevelopmental defects in mental retardation [60]. Since neu‐ roligins are abundant particularly in excitatory synapses, a defect in synaptogenesis has been suggested to result in derangement in cognitive development and communication [59]. None‐

Genetic studies examining the relationship of neuroxins, the connection partners of neuroli‐ gins, with autism revealed that a mutation in neuroxin 1beta gene results in autism susceptibil‐ ity [62]. Structural variants of neuroxin 1alpha gene have also been linked with autism [63].

Another protein adhering to neuroligins is SHANK3. Some forms of autism are considered to stem from a single gene, and particularly from a rare allele having a major effect. Doctor Joseph Buxbaum has reported that one of these genes is SHANK3 gene located on Chromo‐ some 22 (22q13) which is responsible for 1% of autism and some forms of mental retarda‐ tion, microcephaly, and delay in expressive language [34]. SHANK proteins are believed to be the primary regulator of postsynaptic density thanks to their ability to form multimeric complexes with postsynaptic receptors, signal molecules, and cellular skeleton proteins found in dendritic spikes. Postsynaptic density is the measurement of how synapses are linked to each other. A mutation in SHANK3 gene has been reported to be related with au‐ tism spectrum disorders [64]. Role of various mutations in Neuroligin/neuroxin/SHANK3 A large-scale study by Wang and his colleagues [65] revealed a significant relationship be‐ tween a single nucleotide polymorphism located in the 5p14.1 region and autism spectrum disorders. The associated single nucleotide polymorphism is located in a region placed be‐ tween cadherin 10 (CDH 10) and cadherin 9 (CDH 9) genes. CDH 9 and CDH 10 encode type II classic cadherins of the cadherin family, which are transmembraneous glycoprotiens responsible for calcium-dependent cell-cell adhesion. This finding shows the role of neuro‐ nal cellular adhesion molecules in autism pathogenesis [65].

Neurotrophins have many functions such as neuronal survival, target innervation, and synap‐ togenesis in development of peripheral and central nervous system. Neurotrophins exert their biologic functions by binding to a Trk tyrosine kinase receptor which is a high-affinity recep‐ tor. Neurotrophin family has 4 members. These are nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), neurotrophin-3 and neurotrophin-4. BDNF is the most important member of neurotrophin family. BDNF has many roles in neuronal differentiation such as neu‐ ronal survival, dendritic and axonal growth/branching, synapse formation, and neuronal plas‐ ticity [66, 67]. Various studies have investigated the relationship between BDNF gene and autism. Nishimura and colleagues [68] detected an increase in BDNF expression in autistic in‐ dividuals. Subsequent studies confirmed the potential role of BDNF gene mutations in autism pathogenesis [69]. A recent study in which serum BDNF levels significantly increased in autis‐ tic children found no significant impact of genetic variations of BDNF gene on autism risk; however, a significant relationship between neurotrophic tyrosine kinase receptor type 2 (NTRK2) and autism was reported [70]. A large-scale study on patients diagnosed with autism spectrum disorder and mental retardation without autism diagnoses showed that, when com‐ pared to control group, autism spectrum disorders and mental retardation had a significant in‐ crease in serum neurotrophin 4 and BDNF (both are Trk B ligands) [71, 72]. On the other hand, no changes were observed in NGF (trk A ligand) and neurotrophin 3 (Trk C ligand) levels. In light of these findings, it has been suggested that trkB ligands may be overexpressed or secret‐ ed in central nervous system of autistic or mental retarded children during infantile period. It has also been suggested that the effect of BDNF and neurotrophin- 4 on activity-dependent dendritic growth and branching [66] may be related to early and transient brain development seen in autistic infants [67, 73]. This increase in BDNF expression and/or secretion was suggest‐ ed to be linked with the role of Metil-CpG-binding protein 2 (MeCP2) gene in BDNF transcrip‐ tion [74]. A mutation in MeCP2 gene encoding a protein functioning as a general transcriptional receptor is responsible for Rett Syndrome. It has been shown that MeCP2 selec‐ tively bind to BDNF promoter III and suppresses BDNF gene expression. MeCP2 has an impor‐ tant role in regulation of neuronal activity [75]. It has been suggested that MeCP2 mutations located on Xq28 locus may be a risk factor for autism by affecting BDNF expression and den‐ dritic differentiation in cortex. In a study investigating MeCP2 gene mutation in autistic indi‐ viduals for that purpose, 2 girls exhibited de novo mutations [67, 76].

Another gene linked with autism is the Fragile X mental retardation 1 (FMR1) gene encoding Fragile X mental retardation protein (FMRP). FMR1 is associated with autism secondary to Fragile X syndrome [28]. However, fragile X mutations may be found in 7-8% idiopathic au‐ tism patients [77]. FMRP is a selective RNA-binding protein; it transports RNAs to dendrites and regulates local translation of synaptic mRNAs as a response to activation of metabotropic glutamate receptors. This protein is considered to have a role in synaptic plasticity and devel‐ opment of synaptic connections between neural cells. Impaired mRNA translation in the ab‐ sence of FMRP leads to an alteration in protein-synthesis-dependent plasticity [28, 78].

children [96]. Other glutamatergic receptor genes associated with autism are metabotropic GluR8 and GRIN2A (glutamate receptor, ionotropic, N-methyl-D-aspartate 2A) [97, 98].

Genetic and Environmental Factors in Autism

http://dx.doi.org/10.5772/53295

327

Gama aminobutyric acid (GABA) is the major inhibitor neurotransmitter in the brain. GABAA receptors are formed by different homologous subunits. Among GABA receptor subunit genes, GABRA4 with 4p12 location has been shown to play a role in etiology of autism and increases au‐ tism risk by interaction with GABRB1 [99]. Other genes associated with autism in some other studies are GABRG3, GABRA5, GABRB3 located on 15q11-q13, and GABRA2 located on 4p [100-102]. Contrary to these findings, there are other studies with negative results in terms of the

Proenkephalin, prodinorphine of opioid metabolism; tyrosine hydroxylase, dopamin beta hydroxylase (DBH), D2, D3, and D5 dopamin receptors, monoaminooxidase A (MAOA) and B genes of monoaminergic system have no major role in etiology of autism shown in studies [104, 105]. However, a recent study revealed a significant relationship between MAOA gene

Mutations detected in autism in conjunction with all other genetic factors explored so far have been reported to explain no more than 20% of cases with autism spectrum disorder. Thus, a gene-dosage model has been proposed according to which the susceptibility for au‐ tism is determined by the sum of effects of threshold genetic and non-genetic factors [107, 108]. For autism etiology, it has been suggested that the detected chromosomal abnormali‐ ties in combination with other undetected loci cause autism. It has been considered that the inconsistencies between the results of the studies aimed to determine the role of genetic fac‐ tors may be the product of genetic heterogeneity, clinical heterogeneity, and sample size and

In addition to effects of a number of genes of small effect, various environmental factors are believed to be responsible for susceptibility to autism. Development of autism seems to be dependent on interaction of susceptibility genes with each other and with the environment [110]. It has been claimed that among environmental factors related to autism are toxins (en‐ vironment-polluting matters, insecticides, thimerosal in vaccines, lead), viruses (prenatal ex‐ posure to influenza, rubella, and cytomegalovirus infections), and premature birth with premature retinopathy [111-115]. Although there has been a debate regarding the relation‐ ship of autism with thimerosal in measles, rubella, and mumps vaccines; further careful evaluation of data could not support the relationship between autism and vaccines [116, 117]. The relationship between exposure to Rh immune globulin, which contained the pres‐ ervative thimerosal until 2001 in the United States, and autism has also been investigated; however, no significant association has been revealed between exposure of antepartum RhIg preserved with thimerosal and an increase in risk of autistic disorder. The latter findings are in accordance with the consensus that exposure to ethymercury in thimerosal is not the

relationship between GABA receptor genes and autism in various ethnic groups [103].

and autism [106].

ethnic differences among different studies [109].

cause of increased prevalence of autism [118].

**3. Environmental risk factors**

Autism risk is higher than general population in neurofibromatosis, tuberosclerosis, or Cow‐ den Syndrome, a rare syndrome which is characterized by multiple tumor-like growths called hamartomas and affects the intellectual abilities. These diseases develop due to domi‐ nant mutations in tumor suppressor genes NF1, TSC1/TSC2, and PTEN. Mutations in these autism-associated genes affect synaptic protein level by impairing cellular translation. Alter‐ ations in protein level results in abnormal synaptic functions [28].

Angelman syndrome and Prader-Willi syndrome mainly develop due to genetic deletions in 15q11-q13 locus or disomy (condition where two copies of a chromosome comes from a sin‐ gle parent) belonging to a single parent [79]. Deficiencies in paternal genes cause Prader-Willi syndrome; Angelman Syndrome which is more commonly associated with autism may be caused by deletion or mutation in maternal ubiquitin protein ligase gene UBE3A or ATP10C [80, 81]. Other rare single gene defects associated with autism are found in Wil‐ liams Syndrome, Sotos Syndrome, hipomelanozis Ito, and Moebius Syndrome [82-85].

Since serotonin reuptake inhibitors have favourable effects on rituals and routines in autistic individuals and serotonin transporter gene has important role in serotonergic neurotrans‐ mission, serotonin transport gene has been investigated as candidate gene in autism. One of the polymorphisms examined in this gene is the one that is formed by long (L) and short (S) alleles owing to different number of insertion/deletion repeats of a 44-base-pair sequence in the transcriptional control region. Cook and his colleagues [86] reported a significant rela‐ tionship between autism and short allele while Klauck and his colleagues [87] revealed a significant relationship between autism and long allele. A subsequent study did not dupli‐ cate these findings [88]. A different polymorphism investigated at the serotonin transport gene is the variable number of tandem repeats (VNTR) polymorphism due to repeat of a 17 base-pair region at 2nd intron of the gene 9,10, and 12 times. This polymorphism could not be related to autism [89]. Evidence has been accumulated on the relationship of many sero‐ tonin genes, notably serotonin receptor (HTR) 1B, HTR2A, HTR3A, and HTR5A, with au‐ tism [90-93].

Glutamate is the main excitatory neurotransmitter associated with cognitive functions such as memory and learning. Autism has been hypothesized as a hypoglutamatergic disorder by virtue of neuroanatomic studies and the similarities glutamate antagonists generate in healthy persons [94]. It has been demonstrated in genome scanning studies that one of the candidate regions for autism is 6q21 region [95]. This region contains glutamate receptor 6 (GluR6) gene. A study by Jamain and his colleagues [96] found a significant relationship be‐ tween GluR6 gene and autism. It has been thought that GluR6 dysfunction may contribute the deterioration of communication and learning process in autism and any dysregulation of GluR expression may be related to an increase in the rate of epileptic disorder in autistic children [96]. Other glutamatergic receptor genes associated with autism are metabotropic GluR8 and GRIN2A (glutamate receptor, ionotropic, N-methyl-D-aspartate 2A) [97, 98].

Gama aminobutyric acid (GABA) is the major inhibitor neurotransmitter in the brain. GABAA receptors are formed by different homologous subunits. Among GABA receptor subunit genes, GABRA4 with 4p12 location has been shown to play a role in etiology of autism and increases au‐ tism risk by interaction with GABRB1 [99]. Other genes associated with autism in some other studies are GABRG3, GABRA5, GABRB3 located on 15q11-q13, and GABRA2 located on 4p [100-102]. Contrary to these findings, there are other studies with negative results in terms of the relationship between GABA receptor genes and autism in various ethnic groups [103].

Proenkephalin, prodinorphine of opioid metabolism; tyrosine hydroxylase, dopamin beta hydroxylase (DBH), D2, D3, and D5 dopamin receptors, monoaminooxidase A (MAOA) and B genes of monoaminergic system have no major role in etiology of autism shown in studies [104, 105]. However, a recent study revealed a significant relationship between MAOA gene and autism [106].

Mutations detected in autism in conjunction with all other genetic factors explored so far have been reported to explain no more than 20% of cases with autism spectrum disorder. Thus, a gene-dosage model has been proposed according to which the susceptibility for au‐ tism is determined by the sum of effects of threshold genetic and non-genetic factors [107, 108]. For autism etiology, it has been suggested that the detected chromosomal abnormali‐ ties in combination with other undetected loci cause autism. It has been considered that the inconsistencies between the results of the studies aimed to determine the role of genetic fac‐ tors may be the product of genetic heterogeneity, clinical heterogeneity, and sample size and ethnic differences among different studies [109].
