Autism is a severe neurodevelopmental disorder with childhood onset and a rising prevalence rate of 16–20 per 10 000 which is characterized by severe problems in communication, social skills and repetitive behavior. The causes for autism may be genetic1 and/or environmental.2 The preponderance of evidence, however, supports genetic causes for evolution of this disabling disorder,3 with high linkage between autism and several markers on chromosome 7.4 Many areas of the brain exhibit abnormalities in autism, including cerebellum, hippocampus, parietal and frontal cortices.5,6,7 Courchesne and coworkers5 studied the MRIs of 60 autistic and 52 normal boys and showed brain abnormalities such as hyperplasia of cerebellar white matter, neocortical gray matter, and cerebral white matter at the earliest ages, with slowed growth thereafter. These investigators also described evidence of slowed growth in the cerebellar hemispheres and a reduction in the size of vermal gray matter.5 Saitoh et al6 studied the MRIs of 59 autistic and 51 healthy normal controls and showed hypoplasia of the area dentata in the autistic subjects during the earliest years of the disorder (age 29 months–4 years), consistent with previous neuroanatomic work showing increased pyramidal cell packing density and reduced neuronal cell size7 with small dendritic neuronal cell branching.8 Recent biochemical evidence shows significant alterations in levels of several important neuroregulatory proteins affecting growth of central nervous system, in brains of age, sex and postmortem interval-matched (PMI) autistic subjects vs controls; these changes include deficits in Bcl2,9,10 glutamic-acid decarboxylase proteins of 65 and 67 kDa,11 and Reelin10, and increases in p5312 and glial fibrillary acidic proteins.13

Reelin is an extracellular matrix protein with serine protease activity,14 which is responsible for correct lamination of the brain during the embryonic period and cell signaling and synaptic plasticity in the adult life.15 Recent emerging evidence points to pathologic involvement of Reelin in several neuropsychiatric disorders and in several animal models of brain development.16 Costa and coworkers were the first investigators to report deficits in brain levels of Reelin mRNA and protein, in subjects with schizophrenia and psychotic bipolar disorder.17 These reports were later confirmed18,19 and extended to involve patients with nonpsychotic bipolar disorder,17 major depression,19 lissencephaly20 and autism.10 These reports clearly indicated that Reelin protein deficiency may be a common phenomenon subserving cognitive deficits (including psychotic thinking) in multiple neuropsychiatric disorders. A recent report21 also implicates Reelin signaling system in the neuropathology of Alzheimer's disease. These authors showed the presence of Reelin immunostaining in association with amyloid deposits in brains of double-transgenic mice that expressed both human mutant β-amyloid precursor protein and human mutant presenelin-1.21

Several recent genetic studies describe the association of polymorphisms in the RELN gene with three neurodevelopmental disorders: lissencephaly,20 schizophrenia22 and autism.23,24 The study by Grayson and coworkers22 described a polymorphic GGC repeat in the 5′ UTR of the human RELN gene, upstream of the ATG initiation codon.

Polymorphisms in this repeat were also described by Persico et al.23 In that study, six distinct alleles (containing 8–14 copies of the repeats) were found in the general population, and four additional alleles (containing 4,7,15 or 23 repeats) were found in the autistic subjects and their first-degree relatives.23 Alleles containing 8–10 repeats were most common in the general population, although the genotypic and allelic distributions were different for different ethnic groups. Individuals with autism, however, tended to have more repeats. For Caucasians of Italian descent, 17.9% of autistic individuals vs 9.1% unaffected had at least one allele containing 11 or more repeats and the frequencies of ‘long’ alleles were twice as high in affected individuals compared to normal individuals.

Preferential transmission of ‘long’ alleles (11 or more repeats) was also found in a study of 89 Caucasian Americans of European descent and their first-degree relatives.23 In addition to the polymorphic GGC repeat, Persico et al,23 also described two additional polymorphisms eg (i) a A/G transversion predicted to affect RNA splicing immediately 5′ to exon 6; (ii) a conservative T/C transversion in the codon for histidine 2682 in exon 50. Haplotype analysis using these three genetic markers revealed statistically significant increases in the frequency of occurrence in autistic patients vs controls for several specific combinations of alleles.23 Now, in a recent report (see the current issue) Zhang and coworkers,24 test the hypothesis that an instability in the number of repeats in the RELN gene, or an altered distribution in these alleles, may be associated with increased susceptibility towards development of autism. These investigators examined the distribution of GGC repeat numbers in affected members from 126 multiplex and 68 simplex families and found no evidence for expansion, or instability of transmission of this repeat in the autistic subjects. The authors discovered 14 alleles of variable number of GGC repeats (3–16 repeats) in the affected and control subjects;24 there were no differences in the case-control studies for the most common alleles ie, 8 or 10 alleles or genotypes.

Additionally, Zhang et al,24 did not observe an increased frequency of large alleles (more than or equal to 11 repeats) in the autistic cases vs controls. Later, using a family-based association test, these authors showed that the larger alleles (more than or equal to 11 repeats) were transmitted higher than expected in the affected children (S = 43, E(S) = 34.5, P = 0.035) and in particular with the 13 repeat allele (S = 22, E(S) = 16, P = 0.034).24 This increased risk of developing autism supports Persico et al’s23 findings, without evidence of increased allele sharing previously described in the siblings of the autistic children.23 While Zhang et al’s24 case-control studies did not replicate Persico et al’s23 data, their family-based data support Persico et al’s23 study, adding to the weight of genetic evidence implicating RELN in the etiology of autism.23,24 Despite these positive findings, a recent study,29 could not show any association between a polymorphic GGC repeat in the 5′ untranslated region of the RELN gene and autism in a population of mixed European descent emphasizing the need for more future rigorous genetic studies to establish a definitive link between the RELN gene and autism.

Recently, Fatemi et al25 measured blood levels of unprocessed Reelin (410 kDa) and its proteolytic cleavage products (Reelin 330 and 180 kDa) as well as albumin and ceruloplasmin in 28 autistic monozygotic twins (AGRE consortium), their parents (13 fathers, 13 mothers), six normal siblings, and eight normal controls using SDS-PAGE and Western blotting. Results obtained indicated significant reductions in 410 kDa Reelin species in autistic twins (−70%, P < 0.01), their fathers (−62%, P < 0.01), their mothers (−72%, P < 0.01) , and their phenotypically normal siblings (−70%, P < 0.01) vs controls.25

Additionally, autistic cerebellar Reelin 410 kDa protein deficits10 have now been extended to parietal cortices of matched autistic subjects (Fatemi, Stary, Halt, unpublished observations) vs controls, showing further evidence for Reelin deficiency in brains and sera of autistic subjects. These biochemical data10,25 confirm the recent genetic data,23,24 implicating Reelin as a vulnerability factor in the etiopathology of autistic disorder.

Finally, supportive experimental animal data,16,26,27,28 also extend the important role of Reelin glycoprotein, in normal brain development and synaptic plasticity, as exemplified in various rodent models which exhibit Reelin deficits in association with development of abnormal brain structure, leading to the genesis of abnormal behavior. Future studies should focus on and expand the scope of postmortem, genetic and animal model studies to better define the etiology and potential treatments for autism.