About 1.7 % to 4.8 % of individuals with ASD have chromosome abnormalities, including unbalanced translocations, inversions, rings, and interstitial deletions and duplications (Table 5). The chromosome abnormalities that have been reported on more than one occasion are duplication of 15q, deletions of 18q, Xp, 2q and the sex chromosome aneuploidies 47,XYY and 45,X 16.

A recent FISH subtelomere study found one out of ten unselected patients with ASD had a subtelomeric 2qter deletion 17. In our experience 7/7 ASD patients were negative for subtelomeric rearrangements.

Children with Down syndrome have autism more commonly than expected. The incidence was at least 7 % in one study 18. This finding suggests that chromosome abnormalities may lower the threshold for the expression of autism.

In our study (4/14 = 28 %) and in other surveys, the common recurring chromosomal abnormality was duplication of the proximal 15q region (Table 5). About 1% of individuals with ASD have a chromosomal duplication in the Prader-Willi/Angelman region of proximal 15q [1920, present study 4/421[0.95%]]. The duplicated region q11.2q13 is on the maternal chromosome 15 in autistic patients 212223. Most commonly, this is a supernumerary isodicentric 15q chromosome detectable by routine cytogenetic studies or, less commonly, an interstitial duplication of the region detected by FISH analysis for the SNRPN gene region. These two chromosome abnormalities have only subtle effects on the physical phenotype. Supernumerary isodicentric 15q chromosomes are de novo occurrences. Duplication of proximal 15q may result from segregation of a parental chromosome translocation or an interstitial 15q duplication. An abnormal gene dosage within 15q11.2-q13 might cause susceptibility to autism.

The 15q11-q13 region is shown schematically in Figures 3 &4. Chromosome 15q11.2-q13 has gained support as an autism candidate region on the basis of the association of maternally derived chromosomal duplications of this region with an autistic phenotype 2024252627 and genetic evidence for linkage and allelic association in the same interval in chromosomally normal autism families 28293031323334. The maternal specificity of chromosome 15 duplications in autism suggests a genomic imprinting effect. There are multiple imprinted genes in 15q11.2-q13, and two neurodevelopmental disorders exhibiting opposite patterns of genomic imprinting have been mapped to this region 353637. Interstitial deletion of 15q11.2-q13 specific for the paternal chromosome is the most frequent cause of Prader-Willi syndrome (PWS; MIM 176270), whereas maternal-specific deletion of the same common interval results in Angelman syndrome (AS; MIM 105830). The converse of Angelman Syndrome is observed in autism, that is a maternal duplication. The four causes of Angelman syndrome are 1) maternal deletion of 15q11.2q13, 2) paternal UPD15 3) mutations in UBE3A 4) mutations leading to imprinting errors of this region. A population-based study showed a high rate of ASD in AS 38. But, a mutation was not identified in the UBE3A putative promoter or coding region in 10 idiopathic ASD patients 39. Lack of expression of the maternally expressed UBE3A gene in the brain is thought to be the cause of AS. Since patients with deletions compared to other types of AS mutations have a more severe phenotype, suggests the involvement of additional gene losses, such as GABAA receptor gene cluster 40. Transcripts increased in patients with duplications 15q11.2q13 are maternal UBE3A 41, maternal ATP10C 4042 and other transcripts including antisense transcripts that could regulate gene expression 43 and may contribute to the duplication phenotype. Therefore, over expression of genes in 15q11-q13 probably confers ASD risk. Region proximal to D15S11 is considered to have no phenotypic effect. However, one of our patients had a marker 15 negative for D15S11, therefore, some duplications of 15, proximal to D15S11 and the autism candidate region may also influence susceptibility to autistic traits.

Initial studies to characterize the phenotype of 15q11.2q13 duplication patients have found variation among affected people including mental retardation, motor coordination problems, seizure disorder, and impairments in attention, communication, and social function (some but not all with ASD or attention deficit hyperactivity disorder (ADHD)) 4445. It appears there may be a parent-of-origin effect on the linkage and association signals in this region of UBE3A and ATP10C 463233. Further studies across data sets, and rigorous evaluation of potential functional effects of associated alleles, and a thorough assessment of haplotype transmission within ATP10C and neighboring genes would be conclusive. The majority of linkage and association data point to the GABRB3 gene, which is one of a cluster of γ-aminobutyric acid (GABA) receptor subunits that map to the distal, apparently nonimprinted segment of the duplicated region (Fig. 3). Although a number of groups have detected genetic effects at GABRB3 in independent autism populations 293031, other studies have failed to replicate these observations 474849.

Yardin et al 2002 50 recommended a systematic screening by FISH, of chromosome region 15q11.2q13 in cases with autistic-like syndrome

A deletion of 2q37.3 was identified by FISH in a patient with autism and macrocephaly in our study. Wolff et al 2002 51 also reported an autistic patient with a 2q37.3 deletion detected using subtelomere probes. Whole genome screens have suggested several chromosomal regions that are potentially associated with a susceptibility gene for autism. 525354555657. Three studies revealed positive linkage to 2q 525357 and a third study demonstrated linkage to distal 2q in a subset of patients with autism and delayed onset of phrase speech 53. Genomic scans are limited by the number of loci that are assessed; therefore, not all areas may be equally represented. It is important to note that telomeric regions may have increased meiotic recombination and may be under-represented in these types of analyses. Thus, the FISH approach is an important correlative study in the search for susceptibility genes. Macrocephaly in ASD: most children with autism are born with normal head circumference and about 20% meet the criteria for macrocephaly 58. The increased rate of growth in head circumference appears to be most dramatic in the first year of life and corresponds to increased growth of the cerebral cortex as measured by MRI 59.

A de novo deletion of chromosome 3, del(3)(p25), was found in one case with ASD and development delay in our patient pool. A deletion of 3q region was found by Konstantareas & Homatidis 1999 60. Genome wide scan found a major susceptibility locus at 3q25-27 and there was also allelic association in families with autism spectrum disorder originating from a subisolate of Finland 6162. Animal models and linkage data from genome screens implicate the oxytocin receptor at 3p25-p26 6162.

To date, there have been no reports of 12q deletions in patients with ASD, to the best of our knowledge. For the first time we report an interstitial deletion 46,XY,del(12)(q21.2q23.3) in a patient with ASD, development delay, mental retardation, multiple congenital abnormality, and family history of Down syndrome.

46,XY,del(13)(q13.2q14.1)de novo was found in one of our patients' with ASD. Hyperserotonemia in autism is one of the longest-standing biochemical findings. The serotonin 2A receptor gene (HTR2A) on chromosome 13q14q21 is a primary candidate gene in autism. Converging data from recent genome screens also implicates the genomic region containing HTR2A 63646566. Correlation of HTR2A disruption or deletion in our case with a 13q13.2q14.1 deletion and other 13q deletion in ASD patients would complement genome screen data.

The recent physical mapping of the serotonin 5-HT(7) receptor gene (HTR7) to 10q23 67 raises the question if the inversion inv(10)(p11.2q21.2) present in our patient, which is considered a normal familial variant could have long range influence on HTR7 and susceptibility to autism in some cases.

So far there has been no observed association or link between chromosome 14 and ASD. Also, the mosaic inversion inv(14)(q11.2q33) [3/20] found in one of our patients is considered a cultural artifact when seen in 1or 2 cells.

In this study, a patient with inv(17)(q23q25)de novo, had ASD, hypotonia and developmental delay. Chromosome 17q shows association with autism by genome wide scans and in linkage studies 576668. There is also interest in 17q since serotonin transporter gene (SERT) has been mapped to17q11-q12 69.

We report a patient with 18p deletion due to an unbalanced translocation between 14 and 18. Majority of the reported cases with autism involve deletion of 18q 707172. A deletion of the 18p-arm (at band 11.3) in about 50% cells and 50% of the cells with a duplication of the long arm in peripheral blood was described in a mildly obese girl with DSM-III-R autistic disorder and moderate mental retardation 73. Another preschool girl with selective autism and a deletion of Chromosome 18p11.l has also been described 74. She had communication problems consistent with a diagnosis of autism. However, in the area of reciprocal social interaction she was a little less deviant than most children and had no major behavior problems typical of autistic disorder. Linkage and association studies have suggested at least 2 candidate loci, one on the short and the other on the long arms of chromosome 18 75.

One of the breakpoints in the balanced translocation, 46,XX,t(1;14)(q25;q31.2) was 1q25, in the present study. A recent report, links D1S1675 that maps to chromosome 1q24 with autism 76 using obsessive-compulsive behaviors as a restricting criterion for the analysis. The proximity of our breakpoint and D1S1675 may be coincidental or causal.

Although no cases with X- rearrangements were identified, in this study, the literature on abnormal X chromosome and autism is discussed as it has been cited in multiple cases. An autistic (ICD-10) woman had a translocation, t(X;8)(p22.13;q22.1) 77, a boy with "autistic disorder" had duplication of Xp22 78, and a de novo Xp22.3 deletion was observed in 3 autistic females 79. Further, mutations in cell adhesion genes NLGN4 on Xp22.3 and NLGN3 on Xq13 are reported in patients with autism 80. Therefore, subtle Xp rearrangements have to be considered in the cytogenetic assessment of ASD patients

The typical clinical picture in FRAXA includes mental retardation, macro-orchidism, large ears, and prominent jaw. Within neurons, the FMR protein (FMRP) interacts with mRNA and ribosomes, suggesting a role in regulating protein synthesis 81. FMRP is heavily synthesized in dendritic spines in response to synaptic activity, and abnormal dendritic spine size and shape have been noted in FRAXA patients and fmr1 knockout mice 82. These abnormalities may correspond to an abnormal postsynaptic response that weakens synaptic connections 83.

A multicenter study in Sweden 84 found fragile X in 13 of 83 boys (16%) with infantile autism but in none of 19 girls with infantile autism. Klauck et al. (1997) 85 concluded from molecular genetic studies of 141 patients from 105 simplex and 18 multiplex families that an association of autism with fragile X is nonexistent and that the Xq27.3 region is not a candidate for autism. Stoll (2001) 86 presented 11 children under the age of 8 years and the difficulties in diagnosis of fragile X syndrome at this age. The author concluded on the importance of fragile X DNA test for all children with mental retardation, autism, or significant developmental delay without a clear etiology. Whereas only a few percent of children with autism have fragile X syndrome, at least half of children with fragile X syndrome have autistic behaviors, including avoidance of eye contact, language delays, repetitive behaviors, sleep disturbances, tantrums, self-injurious behaviors, hyperactivity, impulsiveness, inattention, and sound sensitivities. The frequency of the fragile X syndrome among individuals with autism was ascertained up to 1993 using cytogenetic method. The incidence ranged from 12.7%–1.6%. However, these studies had different criteria for classifying positive cases. The differences were: the number of metaphases analyzed ranged from 20 to 100 and the cut-off range from 1% to 4% metaphases with a fragile X 848788899091929394. Using molecular analysis the incidence of fragile X syndrome was 5% (1/20) 95, 3.3%(1/30) 96, 12% (3/25) 97 and the present study 2.2 % (7/316). The difference in incidence between the present (2.2%) and previous studies (3.3% – 12%) may be due to small sample size in the other studies or clinical criteria for selection of patients or both.