The Neurobiology of Autism:
New Pieces of the Puzzle
Maria T. Acosta, MD, and Phillip L. Pearl, MD
Address ligence is average, cognitive profiles are frequently mark-
Department of Neurology, Children's National Medical Center,
edly uneven. Significant deficits (eg, in social cognition)
111 Michigan Avenue NW, Washington, DC 20010-2970, USA.
may coexist with islets of strength or even superiority (eg,
E-mail: [e-mail]
visuospatial skills). In addition to the triad of core deficits,
Current Neurology and Neuroscience Reports 2003, 3:149 156
multiple other domains, including motor, sensory, and
Current Science Inc. ISSN 1528 4042
perceptual (eg, auditory, tactile), may be affected. The
Copyright © 2003 by Current Science Inc.
developmental course varies, with regression in language
and social relatedness noted in perhaps one third of cases.
The neurobiologic basis of autism is reviewed, with discus-
Epilepsy affects approximately one third of autistic chil-
sion of evidence from genetic, magnetic resonance imaging,
dren as well, and subclinical epilepsy has been hypothe-
neuropathology, and functional neuroimaging studies.
sized to play a role in clinical regression [1,2].
Although autism is a behaviorally valid syndrome, it is
Prevalence is considered approximately four to five per
remarkably heterogeneous and involves multiple develop-
1000 people, but numbers differ according to criteria used
mental domains as well as a wide range of cognitive, lan-
for diagnosis. Earlier identification and better availability
guage, and socioemotional functioning. Although multiple
of services can increase the awareness of this entity [3].
etiologies are implicated, recent advances have identified
common themes in pathophysiology. Genetic factors play a
primary role, based on evidence from family studies, identi-
Nosology: Autism as a Complex
fication of putative genes using genome-wide linkage analy-
Neurobiologic Disorder
ses, and comorbidities with known genetic mutations. The
Whether autism should be considered a single syndrome
RELN gene, which codes for an extracellular protein guiding
with highly variable severity (ie, the autistic spectrum) or
neuronal migration, has been implicated in autism. Numer-
an aggregate of specific disorders that share common fea-
ous neuropathologic changes have been described, includ-
tures is unresolved. It is clear today that autism is a syn-
ing macroencephaly, acceleration and then deceleration in
drome with multiple etiologies. Recently, enormous
brain growth, increased neuronal packing and decreased
advances in genetics, neuroimaging, and neurotransmitter
cell size in the limbic system, and decreased Purkinje cell
studies have advanced our knowledge of the pathophysiol-
number in the cerebellum. Abnormalities in organization of
ogy of autism, although much uncertainty remains.
the cortical minicolumn, representing the fundamental sub-
There is an effort to develop a validated typology based
unit of vertical cortical organization, may underlie the
on behavioral criteria, based on the hypothesis that dys-
pathology of autism and result in altered thalamocortical
function of a particular brain region or neural network
connections, cortical disinhibition, and dysfunction of the
may produce reasonably predictable behavioral deficits. In
arousal-modulating system of the brain. The role of
autism, this dysfunction would potentially be a result of
acquired factors is speculative, with insufficient evidence to
genetic and environmental factors, expressed morphologi-
link the measles-mumps-rubella (MMR) vaccine with autism
cally as alterations in early brain development.
or to change immunization practices.
Advances in Genetics:
Introduction Genotype versus Phenotype
Autism is a behaviorally defined syndrome characterized There is a body of evidence implicating genetic factors as
by atypical social interaction, disordered verbal and non- playing a primary role in the etiology of autism [4]. Multi-
verbal communication, restricted areas of interest, and lim- ple twin and family studies demonstrate significant differ-
ited imaginative play. Although autism has been well ences in monozygotic and dizygotic twin concordance
validated as a syndrome, it is heterogeneous in its presenta- rates, suggesting an underlying genetic predisposition to
tion. Severity varies widely; intelligence ranges from autism exceeding 90% [5]. In addition, the rate of autism
severely deficient to superior, and language from absent to among a proband's siblings is 3% to 6%, or 50 to 100
grammatically complex but dysprosodic. Even when intel- times greater than the general population [6]. The signifi-
150 Pediatric Neurology
cant sibling risk versus the rapid fall-off in risk for distant with alterations in autism [24]. However, data from differ-
relatives is consistent with the involvement of multiple ent groups have shown that changes in RELN expression
susceptibility genes. The existing genetic models postulate exhibit a broad phenotypic spectrum, including several
different numbers of interacting genes, ranging from less neuropsychiatric disorders [25,26], neuromuscular disor-
than 10, with three loci providing the best fit [7], to more ders, and lissencephaly [27].
than 15 genes, each with a minor effect [8]. Approximately 10% to 15% of autistic individuals
Because autism is a complex disorder, it is clear that the demonstrate comorbility with known genetic conditions,
design of genetic studies plays an important role. One including tuberous sclerosis, neurofibromatosis, fragile X
source of discrepancy may be the heterogeneity in autism, syndrome, and chromosome abnormalities [28]. Cytoge-
both genetic and phenotypic. Among these considerations, netic screening for chromosomal abnormalities has dem-
two types of genetic studies have been designed. One onstrated chromosome 15 and X chromosome break-
focuses on individuals with well-defined subtypes and points as the most frequently associated with autism [29].
groups of patients that are homogeneous and, therefore, Researchers have assumed that gene alterations arising
reduce the "noise" of pooling subjects. This should facili- from such mutations, or the presence of duplicated or
tate detection of the genetic "signal" [9]. missing genes, are directly responsible for the behavior
The other type of study design looks for a wide range phenotype [9]. Special interest has developed in chromo-
of clinical and subclinical forms of the disorder in rela- some 15, as cytogenetic abnormalities in the Prader-
tives of affected patients. For example, an increased fre- Willi/Angelman syndrome critical region (15q11-13)
quency of speech disorders has been found [7]. Also, an have been described in several individuals with autism.
increased prevalence of a series of personality traits (eg, Markers across this region have been screened for evi-
impulsive, aloof, shy, and eccentric) [10] and specific pat- dence of linkage association. Three of the gamma-ami-
terns of social behavior have been identified [11]. As the nobutyric acid (GABA)-A receptor subunits are localized
measurement and detection of these differences become in this critical region. Several studies found an association
more refined, the information may be useful in genetic with polymorphisms of a GABA-A receptor subtype,
analyses. Monozygotic twin studies have shown a concor- GABRB3, and autism [30] or other genes nearby [31,32].
dance rate as high as 60% for the diagnosis of autism. Other studies have not replicated these findings [33].
Milder forms of social and communicative abnormalities UBE3A, an Angelman syndrome gene, has been associ-
were found in up to 95% of monozygotic twins [5]. The ated with autism [34,35]. Although it is premature to
rate of these abnormalities among relatives of probands is consider these findings determinants in the etiology of
remarkably increased compared with the general popula- autism, they open additional opportunities for genetic
tion, and falls rapidly with a decrease in the degree of and neurotransmitter studies.
familial relationship [12]. The higher incidence of autism in boys compared with
girls and the relatively high frequency of autistic character-
Cytogenetic and molecular studies istics in patients with Fragile X syndrome have led to signif-
Genome-wide linkage analyses of families with autism icant interest in the X chromosome. Family studies have
have yielded positive signals for chromosomes 1, 2q, 7q, 9, demonstrated the presence of an association between
13, 15q, 16p, 17q, 19, 22, and X [13 16]. The best candi- known genes of major significance on the X chromosome
dates for regions containing an autism locus are chromo- and autism [16,36]. Mutations in the X-linked gene
somes 2q and 7q [17,18]. An independent linkage study MECP2 have been identified in up to 80% of typical Rett
identified a region in chromosome 7q31 as the locus syndrome patients, with most of the mutations originating
responsible for the speech-language disorder 1 (SPCH1) de novo [37]. However mutations in MECP2 have been
[19]. The possible relationship between SPCH1 and autism observed in a variety of neurobehavioral problems, includ-
is supported by increased rates of language impairment in ing autistic syndrome, learning disabilities, and neonatal
relatives of autistic individuals [20,21]. Recently, the identi- encephalopathies [38,39" ]. Table 1 shows a list of some of
fication of the gene FOXP2, which is responsible for the genes that have been associated with autism [39" ]
SPCH1 and encodes a putative transcription factor, pro-
vides light on the neurodevelopmental process that culmi-
nates in speech and language [22]. Additional associations Neuroimaging Studies:
in chromosome 7 have been found with another candidate Magnetic Resonance Imaging
gene for autism, RELN, which encodes for reelin, an extra- Several neuroimaging studies have reported abnormalities
cellular protein guiding neuronal migration during brain in specific brain regions, including cerebellum [40 44],
development. Recent studies have found a significant asso- mesial temporal structures [45" ,46], brain stem [47], basal
ciation between autistic disorder and a GGC repeat located ganglia, and corpus callosum [48].
at 5 on the RELN gene [23]. Reelin plays a pivotal role in Increase in the overall size and weight of the brain in
the development of the cerebral cortex, cerebellum, hip- individuals with autism compared with age-matched con-
pocampus, and brain stem, which are structures associated trol subjects has been reported [49,50]. Brain size in chil-
The Neurobiology of Autism: New Pieces of the Puzzle " Acosta and Pearl 151
grey and white matter volumes increased by 19% and 46%,
Table 1. Genes frequently associated with autism
respectively, from 2 to 4 years of age compared with 9 to 12
Gene Location
years, but increased by only 1% and 14%, respectively, in
autistic children during this same age span [55" " ,56" " ].
MECP2 (methyl-CpG binding protein - 2) Xq28
Results from multiple studies, therefore, support the
5-HTT (serotonin transporter) 17q11.1-q12
5-HTR 7 (serotonin receptor 7) 10q21-q24 conclusions that autism is associated with acceleration in
GABRB3 (gamma-aminobutyric acid 15q11-q13
brain growth during early childhood, and that this increase
receptor subunit beta 3)
in brain volume does not persist throughout the lifespan
UBE3A/E6-AP (ubiquitin-protein ligase) 15q11-q13
[54" " ]. Tissue enlargement, however, does not appear to be
HRAS1 (c-Harvey-ras-1) 11p15.5
a global phenomenon in the developing autistic brain.
HOXA-1 (Homebox A-1) 7p15.3
Variations have been observed in areas including the cere-
RAY1 or FAM4A1 (supressor of 7q31.3
bellum, limbic system, and corpus callosum [43,56" " ,57].
tumorigenicity 7)
WNT2 7q31-33 In a study of boys with autism aged 2 to 3 years, there was
RELN (reelin) 7q22
less grey matter in the cerebellum, a smaller ratio of grey to
ALDH51A1 (succinic semialdehyde 6p22
white matter, and smaller vermian lobules VI-VII than in
dehydrogenase)
normal control subjects [52" ]. There are 16 MRI studies
HLA genes 6p
reporting significantly reduced size of the cerebellar hemi-
GluR6 (glutamate receptor 6) 6q21
spheres or vermis in patients with autism, making the cere-
ASL (adenylosuccinate lyase) 22q13.3-q13.2
bellum the most widely replicated site of MRI abnormality
Mitochondrial transfer RNA (Lys) Mitochondrial
DNA
in the autism literature [43]. These neuroimaging findings
HTR2A 13q
correlate with post-mortem examinations in which 95% of
Bcl-2 18q 21.3
autistic cases have loss of cerebellar Purkinje neurons.
FOXP2 (speech-language disorder 1) 7q31
A developmental defect has been hypothesized in the
limbic system, given its mediation of memory, social, and
dren with autism appears to be normal at birth [43,51]. affective functions. A developmental MRI study of
This conclusion is based on measures of head circumfer- patients ranging from ages 29 months to 42 years showed
ence with an index that has been shown to be highly pre- a smaller cross-sectional area of the dentate gyrus [41].
dictive of computed tomography (CT)-based or magnetic Posterior regions of the corpus callosum were also
resonance imaging (MRI)-based brain volume [52" ,53]. reduced in size in autism in a study of patients ranging
By 2 to 4 years of age, however, 90% of autistic patients from 3 to 42 years of age [58].
have brain volumes that are larger than average. More- In conclusion, recent MRI observations suggest abnor-
over, 37% of 2- to 4-year-old autistic toddlers meet crite- mal regulation of brain growth in autism, characterized by
ria for developmental macroencephaly [51]. early overgrowth followed by abnormally slow growth in
Several volumetric studies using MRI techniques have some regions, but premature arrest of growth in others.
explored this issue. Total brain volume by MRI volumetric This evidence raises a growth dysregulation hypothesis of
techniques was recently measured in a group of 67 autistic autism in which there is pathologic dysregulation in the
individuals from 8 to 46 years of age and 83 normal con- timing and amount of growth as well as cessation of
trol patients matched by sex, age, IQ, ethnicity, and socio- growth. Pathologic growth regulation during this critical
economic status [54" " ]. All autistic individuals were non- period of life could lead to widespread and pervasive con-
mentally retarded. An increase in total brain volume in sequences for the functional differentiation of systems
subjects with autism between 8 and 12 years of age was mediating many neurobehavioral domains. Evidence con-
found. No increase in brain volume was found for autistic sistent with this prediction of aberrant functional organi-
individuals 12 years or older. Similar findings in autistic zation in autism comes from recent functional MRI (fMRI)
children aged 2 to 4 years has been reported [52" ]. In this studies of mature patients performing tasks involving
study, 90% of autistic boys aged 2 to 4 years had more cere- movement [59,60], face perception [59,60], processing
bral (18%) and cerebellar (39%) white matter and more emotions [61], and visual attention [58].
cerebral cortical grey matter (12%) than normal patients, This abnormal pattern of brain growth in autism
whereas older autistic children and adolescents did not may also include a lack of normal acceleration in
have such enlargement of grey and white matter. Addition- growth that occurs in typically developing adolescents.
ally, increased cerebral size in young autistic children This period of maturation is associated with the emer-
showed an anterior to posterior gradient, with frontal gence of a second phase of higher-order abilities, par-
lobes being the most enlarged and occipital lobes showing ticularly frontal lobe functions [62]. This means that
the least effect. After a period of accelerated growth, cross- even though the brain size is "normalized" in adoles-
sectional MRI data show slowed growth velocity through- cents with autism, this normalization may be the result
out cerebral and cerebellar regions, in contrast with nor- of both early acceleration and later deceleration in
mal subjects. For example, in normal children, frontal lobe brain growth.
152 Pediatric Neurology
Elevated brain neurotrophins and neuropeptides anatomic and physiologic unit of the cerebral cortex, is the
(brain-derived neurotrophic factor, neurotrophin 4/5, smallest level of cortical vertical organization.
vasoinhibitory peptide, and calcitonin gene-related pep- Early in development, post-mitotic neurons leave the
tide) have been found in neonatal blood spots of individu- ventral wall and travel in a path ending with their radial
als who later developed autism and mental retardation arrangement within the cortical plate [69]. In the adult
[63" " ]. These and other growth factors play roles in neu- brain, minicolumns appears as thin radial structures, rang-
ronal proliferation, migration, differentiation, growth, and ing from 30- to 60-µm wide depending on the cortical
circuit organization. These elevations or others could be area. Each minicolumn contains a repetitive array of affer-
the molecular basis for the early and accelerated brain ent inputs, intrinsic microcircuitry, and efferent outputs
growth in autism. Neurotrophins typically stimulate the infusing the structure with a putative role as a physiologic
growth of both neuronal and glial elements and endothe- unit [70" " ].
lium. The increased levels of neurotrophins suggest that In autism, more numerous, smaller [67" " ,68], and
there is a premature over-expression of genes that leads to less compact minicolumns have been described
the production of neurotropins and neuropeptides. Future [52" ,55" " ,65]. This may imply defects in the prolifera-
studies should examine possible relationships between tion of neuronal precursor cells or environmental-
nerve growth factors and brain growth patterns. induced changes affecting minicolumnar architecture.
Altered minicolumnar organization has also been
described in Down syndrome, where minicolumns are of
Neuropathologic Studies: Morphologic normal width, but the radial structures attain adult mini-
Developmental Alterations columnar size earlier than normal. This is consistent with
Neuroanatomic abnormalities have been observed in the the accelerated aging observed in this condition.
limbic system and cerebellum. Three major neuropatho- The larger brain size in autistic children, coupled with
logic findings have been described. They are as follows: 1) the observation that cortical minicolumns are smaller and
curtailed development of neurons in the forebrain limbic more numerous, results in a novel cytoarchitectural
system (anterior cingulate gyrus, hippocampus, subicu- arrangement of a relative and absolute increase in columns
lum, entorhinal cortex, and mammillary body); 2) congen- per brain surface area, or processing units [67" " ,71]. This
ital decrease in the number of Purkinje cells in the configuration may originate during the genesis of neurons
cerebellum; and 3) age-related differences in cell size and and the minicolumn. As cortical surface area has increased
neuronal number in the cerebellar nuclei and the inferior evolutionarily with essentially constant size of the minicol-
olivary nucleus of the brainstem, suggesting an evolving umn, the result is an increase in the number and complex-
process and disturbance in the synaptic relationships of ity of processing units. During this slow process
these nuclei [64,65]. throughout evolution, selection pressures would have
Although gross neuropathologic changes are not typi- occurred to benefit the organism, or at least not prove mal-
cal in autism, abnormalities observed at the cellular level adaptive [70" " ]. In autism, a significant increase in pro-
may underlie the basis of clinical manifestations. Increased cessing units may occur as an acute event, not subject to
neuronal cell density and decreased nerve cell size have normal selection pressures. If thalamic terminations
been found bilaterally in the amygdala, entorhinal cortex, remain the same in autism and minicolumns are smaller,
mammillary body, anterior cingulate gyrus, medial septal then more minicolumns will be innervated per thalamic
nucleus, and hippocampal complex [66]. These changes in afferent terminal than in the normal brain. The failure to
the limbic system may be related to the emotional and assimilate extra processing units may result in cortical
mood manifestations of the clinical syndrome. Although "noise" that then overtaxes the system [67" " ].
dysmorphic neurons are not usually found, a simplified Among several conceptual classifications, autism has been
dendritic pattern in hippocampal regions CA4 and CA1 considered a disorder of the arousal-modulating system of the
has been described [46]. brain. Accordingly, autistic individuals experience a chronic
state of over-arousal and subsequently exhibit abnormal
The minicolumn hypothesis behaviors. The arousal theory is consistent with a reduction in
Microscopic examination of brain tissue in autistic patients inhibitory interneuronal activity. The cortex contains inhibi-
provides evidence for altered central nervous system (CNS) tory cells that define minicolumnar organization. Lateral
development. The processes of neuronal genesis and inhibition caused by GABAergic neurons helps to ensure indi-
migration are accomplished mainly before birth in mam- vidual minicolumn discreteness and, during development,
mals. Both events must occur at the proper tempo and in a compels adjacent minicolumns into establishing connections
coordinated fashion for the critical events of neuronal dif- with functionally dissimilar sets of thalamic neurons. A lack
ferentiation and synaptogenesis to occur properly. Recent of this inhibition would grossly alter the connective patterns
observations regarding minicolumnar organization pro- between thalamic input and the cerebral cortex. The result
vide important evidence about developmental alterations would affect the ability to discriminate between competing
in autism [67" " ,68]. The minicolumn, the fundamental types of sensory information [67" " ].
The Neurobiology of Autism: New Pieces of the Puzzle " Acosta and Pearl 153
Other developmental alterations gic afferents to the cortex influence neural stem cell pro-
Other studies have reported dysregulation of the proteins liferation [84,85] and may contribute to an increased
reelin and Bcl-2, with reduction of these two proteins in number of cortical minicolumns. Further studies on the
autistic cerebellar tissue compared with human control tis- role of serotonin in prenatal cortical development are
sue [24]. Reelin, a signaling protein that guides neuronal needed, as abnormal neurogenesis or migration may
migration in the developing fetus, as well as a cellular sig- underlie the macrocephaly associated with autism.
naling system subserving cognition in adult brain [26], is Whole blood serotonin levels are increased in patients,
the product of the RELN gene, a candidate gene for autism and serotonin transporter inhibitors reduce rituals and
[23,72" ]. The Bcl-2 protein governs programmed cell aggression in autism. Comparative studies of monoamine
death, or apoptosis, in the developing and maturing levels in autistic children versus control patients suggest
human brain. Reductions in these neuroregulatory pro- abnormal maturation of the serotoninergic system during
teins may explain the neuronal migration and cell density child development [86].
abnormalities observed in autism, and are intriguing in Chugani et al. [87] demonstrated developmental
light of the aforementioned putative peptide biomarkers changes in serotonin synthesis capacity in children using
found in newborns [63" " ]. alpha[C-11]methyl-L-tryptophan and positron emission
Hippocampal studies in autistic individuals have tomography. For nonautistic children, serotonin synthe-
shown that cells in this region are often smaller, packed sis capacity was greater than 200% of adult values until
more densely, and have less dendritic branching than is the age of 5 years, and then declined toward adult values.
expected [73]. Proteomic studies have demonstrated In autistic children, serotonin synthesis capacity
changes in protein in frontal cortex. Reduction or absence increased gradually between the ages of 2 and 11 years to
of Ä…B-crystallin in autistic brains has been reported [74]. values 1.5 times of adult normal values. These findings
Crystallin belongs to a group of small heat shock proteins suggest that humans undergo a period of high brain sero-
and functions as a molecular chaperone. tonin-synthesis capacity during childhood, and that this
developmental process is disrupted in autistic children,
with some brain regions more severely impacted [77,87].
Functional Neuroimaging Changes in serotonin synthesis and receptor density with
and Neurotransmitter Studies: age suggest that serotonin plays an important role in
Serotoninergic Dysfunction brain development [88].
Serotonin, like other monoamine transmitters, has been
shown to play a role in regulating brain development Other neurotransmitters
[75" " ]. It influences the processes of neurogenesis, neu- Other neurotransmitter alterations associated with autism
ronal differentiation, neuropil formation, axon myelina- include glutamate and GABA. The glutamate receptor 6
tion, and synaptogenesis. Removal of serotonin during (GluR6) gene is localized in 6q21 and has been implicated
very early fetal development in rats causes a permanent in autism [89]. It is highly expressed in brain regions
reduction in the number of neurons in adult brain hippoc- involved in learning and memory, and in motor and moti-
ampus and cortex [75" " ]. At a later time in development, vational aspects of behavior [17]. GABAergic interneurons
serotonin plays a role in dendritic development, including in mouse cortex originate from the ganglionic eminence of
overall dendritic length, spine formation, and branching in the ventral telencephalon. Disturbances of interneuron
both hippocampus and cortex. Additionally, serotonin lev- migration and integration have been implicated in a vari-
els can be affected by multiple and nonspecific post-natal ety of developmental disorders, including cortical dyspla-
factors, including hypoxia, viral infections, malnutrition sia with epilepsy [90], schizophrenia [91], Tourette's
[76], social enrichment, and stress. Drugs, including syndrome [92], and autism [67" " ,68].
cocaine, nicotine, and alcohol, can alter serotonin levels.
Increased evidence points to an imbalance in seroto-
nin levels as important for the etiology of autism Acquired Factors/Immunizations
[75" " ,77 79]. Although several reports conclude that A possible etiologic connection between viral infections
dysgenesis of several brain systems occurs in autism and autism has received recent attention. Despite a pro-
[52" ,55" " ,65,66,67" " ,68], only recently has a link been posed link between autism and several pathogens, includ-
made between the role of serotonin in brain development ing rubella, cytomegalovirus, and herpes simplex [93,94], a
and autism [75" " ,77,80]. When assessing the effects of direct search for viral genome or protein in blood and cere-
serotonin on cortical growth, both an abundance and a brospinal fluid of patients with autism has not produced
paucity may be deleterious. Depletion of serotonin consistent results [95]. Congenital rubella has been associ-
results in a significant delay in maturation of the soma- ated with autism, although less than 1% of patients with
tosensory cortex [81,82]. In contrast, excessive serotonin autism have congenital rubella [96].
during early development results in hyperinnervation and A speculative association between the measles-mumps-
expansion of cortical architecture [83]. Early serotoniner- rubella (MMR) vaccine and autism has been described as a
154 Pediatric Neurology
new, chronic inflammatory bowel syndrome with autistic- References and Recommended Reading
Papers of particular interest, published recently, have been
like developmental regression [97]. Parallel epidemiologic
highlighted as:
studies of central nervous system complications from
" Of importance
native and attenuated measles, mumps, and rubella viruses
" " Of major importance
show that these manifestations do not mimic autism [39].
1. Shinnar S, Rapin I, Arnold S, et al.: Language regression in
Given that MMR vaccination is practically ubiquitous in
childhood. Pediatr Neurol 2001, 24:183 189.
developed countries, and that the time of its administra-
2. Tuchman RF, Rapin I: Regression in pervasive developmental
tion overlaps with the onset of autism, a causal link
disorders: seizures and epileptiform electroencephalogram
between these events cannot be confirmed or ruled out by correlates. Pediatrics 1997, 99:560 566.
3. Yeargin-Allsopp M: Past and future perspectives in autism epi-
anything other than statistical evidence. Since the publica-
demiology. Mol Psychiatry 2002, 7(suppl 2):S9 S11.
tion of the presumptive association between autism and
4. Lamb JA, Moore J, Bailey A, Monaco AP: Autism: recent molec-
MMR, several large, epidemiologic studies have been devel-
ular genetic advances. Hum Mol Genet 2000, 9:861 868.
oped. A large, 14-year retrospective study in Finland did 5. Bailey A, Le Couteur A, Gottesman I, et al.: Autism as a strongly
genetic disorder: evidence from a British twin study. Psychol
not find an association between MMR and cases of autism
Med 1995, 25:63 77.
[98]. Other investigators have likewise been unable to
6. Rutter M: The Emanuel Miller Memorial Lecture 1998.
demonstrate an association between autism and MMR
Autism: two-way interplay between research and clinical
work. J Child Psychol Psychiatry 1999, 40:169 188.
[99 101]. Recently, the American Academy of Pediatrics
7. Pickles A, Starr E, Kazak S, et al.: Variable expression of the
published the results of a Consensus Development Confer-
autism broader phenotype: findings from extended
ence regarding MMR and autism, concluding that an asso-
pedigrees. J Child Psychol Psychiatry 2000, 41:491 502.
ciation is based on single cases, personal observations, and 8. Risch N, Spiker D, Lotspeich L, et al.: A genomic screen of
autism: evidence for a multilocus etiology. Am J Hum Genet
isolated and circumstantial information [102" " ]. Separate
1999, 65:493 507.
administration of measles, mumps, and rubella vaccines to
9. Spence MA: The genetics of autism. Curr Opin Pediatr
children provides no benefit over the administration of the
2001, 13:561 565.
10. Murphy M, Bolton PF, Pickles A, et al.: Personality traits
combination MMR vaccine, and may result in delayed or
of the relatives of autistic probands. Psychol Med
missed immunizations [102" " ].
2000, 30:1411 1424.
11. Constantino JN, Todd RD: Genetic structure of reciprocal
social behavior. Am J Psychiatry 2000, 157:2043 2045.
12. Bailey A, Palferman S, Heavey L, Le Couteur A: Autism: the phe-
Conclusions
notype in relatives. J Autism Dev Disord 1998, 28:369 392.
The past decade has provided enormous advances in the
13. Smalley SL, Kustanovich V, Minassian SL, et al.: Genetic linkage
scientific evidence surrounding the neurobiologic basis of
of attention-deficit/hyperactivity disorder on chromosome
autism. Based on this evidence, autism may be hypothe- 16p13, in a region implicated in autism. Am J Hum Genet
2002, 71:959 963.
sized to result from genetically determined prenatal alter-
14. A genomewide screen for autism: strong evidence for linkage
ations in brain development. Multiple genes,
to chromosomes 2q, 7q, and 16p. Am J Hum Genet
neurotransmitter systems, and brain regions are impli- 2001, 69:570 581.
15. Gutknecht L: Full-genome scans with autistic disorder: a
cated. The timing of prenatal factors, both genetically and
review. Behav Genet 2001, 31:113 123.
environmentally based, appears to have a fundamental
16. Shao Y, Wolpert CM, Raiford KL, et al.: Genomic screen and
role. Alterations in the minicolumnar organization of the
follow-up analysis for autistic disorder. Am J Med Genet
2002, 114:99 105.
cerebral cortex appears to be a significant factor in the
17. Licinio J, Alvarado I: Progress in the genetics of autism. Mol
pathogenesis of autism, and this process appears to be reg-
Psychiatry 2002, 7:229.
ulated by neurotransmitters such as serotonin.
18. Yu CE, Dawson G, Munson J, et al.: Presence of large deletions
Central nervous system development is not finished at
in kindreds with autism. Am J Hum Genet 2002, 71:100 115.
19. Fisher SE, Vargha-Khadem F, Watkins KE, et al.: Localization of
birth, and the post-natal onset of developmental arrest or
a gene implicated in a severe speech and language disorder.
regression invokes the presence of external or newly
Nat Genet 1998, 18:168 170.
acquired cerebral insults. Yet, there is strong evidence from
20. Folstein SE, Mankoski RE: Chromosome 7q: where autism
genetic, neuroimaging, and neurotransmitter studies that meets language disorder?. Am J Hum Genet 2000, 67:278 281.
21. Warburton P, Baird G, Chen W, et al.: Support for linkage of
individuals with autism have prenatal onset or pre-pro-
autism and specific language impairment to 7q3 from two
grammed alterations in brain function and development.
chromosome rearrangements involving band 7q31. Am J Med
Additional studies are needed to understand the early fac-
Genet 2000, 96:228 234.
tors regulating nerve cell formation, migration, and corti- 22. Lai CS, Fisher SE, Hurst JA, et al.: A forkhead-domain gene is
mutated in a severe speech and language disorder. Nature
cal column formation, as well as acquired or
2001, 413:519 523.
environmental factors that may interact with the earliest
23. Persico AM, D'Agruma L, Maiorano N, et al.: Reelin gene alleles
events in brain development. and haplotypes as a factor predisposing to autistic disorder.
Mol Psychiatry 2001, 6:150 159.
24. Fatemi SH, Stary JM, Halt AR, Realmuto GR: Dysregulation of
reelin and Bcl-2 proteins in autistic cerebellum. J Autism Dev
Disord 2001, 31:529 535.
The Neurobiology of Autism: New Pieces of the Puzzle " Acosta and Pearl 155
25. Petek E, Windpassinger C, Vincent JB, et al.: Disruption of a 47. Hashimoto T, Tayama M, Murakawa K, et al.: Development of
novel gene (IMMP2L) by a breakpoint in 7q31 associated the brainstem and cerebellum in autistic patients. J Autism
with Tourette syndrome. Am J Hum Genet 2001, 68:848 858. Dev Disord 1995, 25:1 18.
26. Fatemi SH, Kroll JL, Stary JM: Altered levels of Reelin and its 48. Hardan AY, Minshew NJ, Keshavan MS: Corpus callosum size
isoforms in schizophrenia and mood disorders. Neuroreport in autism. Neurology 2000, 55:1033 1036.
2001, 12:3209 3215.
49. Fidler DJ, Bailey JN, Smalley SL: Macrocephaly in autism and
27. Hong SE, Shugart YY, Huang DT, et al.: Autosomal recessive other pervasive developmental disorders. Dev Med Child Neu-
lissencephaly with cerebellar hypoplasia is associated with rol 2000, 42:737 740.
human RELN mutations. Nat Genet 2000, 26:93 96.
50. Fombonne E, Roge B, Claverie J, et al.: Microcephaly and mac-
28. Fombonne E, Chakrabarti S: No evidence for a new variant of rocephaly in autism. J Autism Dev Disord 1999, 29:113 119.
measles-mumps-rubella-induced autism. Pediatrics
51. Lainhart JE, Piven J, Wzorek M, et al.: Macrocephaly in chil-
2001, 108:E58.
dren and adults with autism. J Am Acad Child Adolesc Psychiatry
29. Gillberg C: Chromosomal disorders and autism. J Autism Dev 1997, 36:282 290.
Disord 1998, 28:415 425.
52." Courchesne E, Karns CM, Davis HR, et al.: Unusual brain
30. Buxbaum JD, Silverman JM, Smith CJ, et al.: Association growth patterns in early life in patients with autistic
between a GABRB3 polymorphism and autism. Mol Psychiatry disorder: an MRI study. Neurology 2001, 57:245 254.
2002, 7:311 0316. This study shows evidence of abnormal regulation of brain growth,
with early overgrowth and later abnormal slow growth.
31. Menold MM, Shao Y, Wolpert CM, et al.: Association analysis
of chromosome 15 gaba receptor subunit genes in autistic 53. Lemons JA, Schreiner RL, Gresham EL: Relationship of brain
disorder. J Neurogenet 2001, 15:245 259. weight to head circumference in early infancy. Hum Biol
1981, 53:351 354.
32. Martin ER, Menold MM, Wolpert CM, et al.: Analysis of linkage
disequilibrium in gamma-aminobutyric acid receptor 54." " Aylward EH, Minshew NJ, Field K, et al.: Effects of age on brain
subunit genes in autistic disorder. Am J Med Genet volume and head circumference in autism. Neurology
2000, 96:43 48. 2002, 59:175 183.
33. Maestrini E, Lai C, Marlow A, et al.: Serotonin transporter (5- Supportive evidence for an abnormal pattern of brain development
early in life and during adolescence in autism.
HTT) and gamma-aminobutyric acid receptor subunit beta3
(GABRB3) gene polymorphisms are not associated with 55." " Carper RA, Courchesne E: Inverse correlation between frontal
autism in the IMGSA families. The International Molecular lobe and cerebellum sizes in children with autism. Brain
Genetic Study of Autism Consortium. Am J Med Genet 2000, 123(Pt 4):836 844.
1999, 88:492 496. This study shows evidence of concurrent abnormalities in both the
frontal lobe and the cerebellum.
34. Nurmi EL, Bradford Y, Chen Y, et al.: Linkage disequilibrium at
the Angelman syndrome gene UBE3A in autism families. 56." " Carper R, Moses P, Tigue Z, Courchesne E: Cerebral lobes in
Genomics 2001, 77:105 113. autism: early hyperplasia and abnormal age effects. Neuroim-
age 2002, 16:1038.
35. Slopien A, Rajewski A: Genetic studies in autistic disorders.
Evaluation of volumetric variations in different brain regions. Mecha-
Psychiatry Pol 2000, 34:435 446.
nisms that might account for early hyperplasia are discussed.
36. Schutz CK, Polley D, Robinson PD, et al.: Autism and the X
57. Saitoh O, Karns CM, Courchesne E: Development of the hip-
chromosome: no linkage to microsatellite loci detected using
pocampal formation from 2 to 42 years: MRI evidence of
the affected sibling pair method. Am J Med Genet
smaller area dentata in autism. Brain 2001, 124:1317 1324.
2002, 109:36 41.
58. Egaas B, Courchesne E, Saitoh O: Reduced size of corpus
37. Amir RE, Zoghbi HY: Rett syndrome: methyl-CpG-binding
callosum in autism. Arch Neurol 1995, 52:794 801.
protein 2 mutations and phenotype- genotype correlations.
Am J Med Genet 2000, 97:147 152. 59. Muller RA, Pierce K, Ambrose JB, et al.: Atypical patterns of
38. Hammer S, Dorrani N, Dragich J, et al.: The phenotypic conse- cerebral motor activation in autism: a functional magnetic
resonance study. Biol Psychiatry 2001, 49:665 676.
quences of MECP2 mutations extend beyond Rett syndrome.
Ment Retard Dev Disabil Res Rev 2002, 8:94 98. 60. Pierce K, Muller RA, Ambrose J, et al.: Face processing occurs
outside the fusiform 'face area' in autism: evidence from
39." Korvatska E, Van de WJ, Anders TF, Gershwin ME: Genetic and
functional MRI. Brain 2001, 124:2059 2073.
immunologic considerations in autism. Neurobiol Dis
2002, 9:107 125. 61. Critchley HD, Daly EM, Bullmore ET, et al.: The functional
Overview of genetic and inmunologic aspects of autism. neuroanatomy of social behaviour: changes in cerebral
blood flow when people with autistic disorder process facial
40. Kemper TL, Bauman ML: The contribution of neuropathologic
expressions. Brain 2000, 123(Pt 11):2203 2212.
studies to the understanding of autism. Neurol Clin
1993, 11:175 187. 62. Casey BJ, Giedd JN, Thomas KM: Structural and functional
brain development and its relation to cognitive develop-
41. Saitoh O, Karns CM, Courchesne E: Development of the hip-
ment. Biol Psychol 2000, 54:241 257.
pocampal formation from 2 to 42 years: MRI evidence of
smaller area dentata in autism. Brain 2001, 124:1317 1324. 63." " Nelson KB, Grether JK, Croen LA, et al.: Neuropeptides and
neurotrophins in neonatal blood of children with autism or
42. Courchesne E, Karns CM, Davis HR, et al.: Unusual brain
mental retardation. Ann Neurol 2001, 49:597 606.
growth patterns in early life in patients with autistic disor-
Evaluation of expression of neuropeptides and neurotrophins
der: an MRI study. Neurology 2001, 57:245 254.
in peripheral blood in children with autism drawn in the first days
43. Courchesne E: Abnormal early brain development in autism.
of life.
Mol Psychiatry 2002, 7(suppl 2):S21 S23.
64. Kemper TL, Bauman M: Neuropathology of infantile autism.
44. Hardan AY, Minshew NJ, Harenski K, Keshavan MS: Posterior
J Neuropathol Exp Neurol 1998, 57:645 652.
fossa magnetic resonance imaging in autism. J Am Acad Child
65. Bauman M, Kemper TL: Histoanatomic observations of the
Adolesc Psychiatry 2001, 40:666 672.
brain in early infantile autism. Neurology 1985, 35:866 874.
45." Sparks BF, Friedman SD, Shaw DW, et al.: Brain structural
66. Kemper TL, Bauman ML: Neuropathology of infantile autism.
abnormalities in young children with autism spectrum
Mol Psychiatry 2002, 7(suppl 2):S12 S13.
disorder. Neurology 2002, 59:184 192.
Supportive evidence of volumetric changes in brain development in 67." " Casanova MF, Buxhoeveden DP, Switala AE, Roy E: Minicolum-
children with autism. nar pathology in autism. Neurology 2002, 58:428 432.
Description of differences between brain of autistic patients and con-
46. Raymond GV, Bauman ML, Kemper TL: Hippocampus in
trol patients in the number and structure of minicolumns
autism: a Golgi analysis. Acta Neuropathol (Berl)
1996, 91:117 119.
156 Pediatric Neurology
68. Casanova MF, Buxhoeveden DP, Switala AE, Roy E: Asperger's 85. Lauder JM, Wallace JA, Krebs H: Roles for serotonin in neuro-
syndrome and cortical neuropathology. J Child Neurol embryogenesis. Adv Exp Med Biol 1981, 133:477 506.
2002, 17:142 145.
86. Martineau J, Barthelemy C, Jouve J, et al.: Monoamines (sero-
69. Kornack DR, Rakic P: Radial and horizontal deployment of tonin and catecholamines) and their derivatives in infantile
clonally related cells in the primate neocortex: relationship autism: age-related changes and drug effects. Dev Med Child
to distinct mitotic lineages. Neuron 1995, 15:311 321. Neurol 1992, 34:593 603.
70." " Buxhoeveden DP, Casanova MF: The minicolumn hypothesis 87. Chugani DC, Muzik O, Rothermel R, et al.: Altered serotonin
in neuroscience. Brain 2002, 125:935 951. synthesis in the dentatothalamocortical pathway in autistic
Review of the concept of minicolumn as an anatomic and boys. Ann Neurol 1997, 42:666 669.
functional unit.
88. Chugani DC: Role of altered brain serotonin mechanisms in
71. Casanova MF, Buxhoeveden DP, Cohen M, et al.: Minicolumnar autism. Mol Psychiatry 2002, 7(suppl 2):S16 S17.
pathology in dyslexia. Ann Neurol 2002, 52:108 110.
89. Jamain S, Quach H, Quintana-Murci L, et al.: Y chromosome
72." Pickett J: Current investigations in autism brain tissue haplogroups in autistic subjects. Mol Psychiatry
research. J Autism Dev Disord 2001, 31:521 527. 2002, 7:217 219.
Update on current autism brain research efforts.
90. Roper SN, Eisenschenk S, King MA: Reduced density of parval-
73. Blatt GJ, Fitzgerald CM, Guptill JT, et al.: Density and distribu- bumin- and calbindin D28-immunoreactive neurons in
tion of hippocampal neurotransmitter receptors in autism: experimental cortical dysplasia. Epilepsy Res 1999, 37:63 71.
an autoradiographic study. J Autism Dev Disord
91. Benes FM, McSparren J, Bird ED, et al.: Deficits in small inter-
2001, 31:537 543.
neurons in prefrontal and cingulate cortices of schizo-
74. Junaid MA, Pullarkat RK: Proteomic approach for the elucida- phrenic and schizoaffective patients. Arch Gen Psychiatry
tion of biological defects in autism. J Autism Dev Disord 1991, 48:996 1001.
2001, 31:557 560.
92. Leckman JF, Riddle MA: Tourette's syndrome: when
75." " Whitaker-Azmitia PM: Serotonin and brain development: role habit-forming systems form habits of their own? Neuron
in human developmental diseases. Brain Res Bull 2000, 28:349 354.
2001, 56:479 485.
93. Stubbs EG, Ash E, Williams CP: Autism and congenital
Overview of the role of serotonin in prenatal and post-natal
cytomegalovirus. J Autism Dev Disord 1984, 14:183 189.
brain development.
94. Gillberg IC: Autistic syndrome with onset at age 31 years:
76. Barreto Medeiros JM, Cabral Filho JE, De Souza SL, et al.: Early
herpes encephalitis as a possible model for childhood
malnourished rats are not affected by anorexia induced by a
autism. Dev Med Child Neurol 1991, 33:920 924.
selective serotonin reuptake inhibitor in adult life. Nutr
95. Van Gent T, Heijnen CJ, Treffers PD: Autism and the immune
Neurosci 2002, 5:211 214.
system. J Child Psychol Psychiatry 1997, 38:337 349.
77. Chugani DC, Muzik O, Behen M, et al.: Developmental
96. Fombonne E: The epidemiology of autism: a review. Psychol
changes in brain serotonin synthesis capacity in autistic and
Med 1999, 29:769 786.
nonautistic children. Ann Neurol 1999, 45:287 295.
97. O'Leary JJ, Uhlmann V, Wakefield AJ: Measles virus and
78. Tordjman S, Gutknecht L, Carlier M, et al.: Role of the seroto-
autism. Lancet 2000, 356:772.
nin transporter gene in the behavioral expression of autism.
98. Peltola H, Patja A, Leinikki P, et al.: No evidence for measles,
Mol Psychiatry 2001, 6:434 439.
mumps, and rubella vaccine-associated inflammatory bowel
79. Carlsson ML: Hypothesis: is infantile autism a hypo-
disease or autism in a 14-year prospective study. Lancet
glutamatergic disorder? Relevance of glutamate-serotonin
1998, 351:1327 1328.
interactions for pharmacotherapy. J Neural Transm
99. Taylor B, Miller E, Farrington CP, et al.: Autism and measles,
1998, 105:525 535.
mumps, and rubella vaccine: no epidemiological evidence
80. Donovan SL, Mamounas LA, Andrews AM, et al.: GAP-43 is
for a causal association. Lancet 1999, 353:2026 2029.
critical for normal development of the serotonergic innerva-
100. Dales L, Hammer SJ, Smith NJ: Time trends in autism and in
tion in forebrain. J Neurosci 2002, 22:3543 3552.
MMR immunization coverage in California. JAMA
81. Blue ME, Erzurumlu RS, Jhaveri S: A comparison of pattern
2001, 285:1183 1185.
formation by thalamocortical and serotonergic afferents in
101. Kaye JA, Mar Melero-Montes M, Jick H: Mumps, measles, and
the rat barrel field cortex. Cereb Cortex 1991, 1:380 389.
rubella vaccine and the incidence of autism recorded by
82. Bennett-Clarke CA, Leslie MJ, Lane RD, Rhoades RW: Effect of
general practitioners: a time trend analysis. BMJ
serotonin depletion on vibrissa-related patterns of thalamic
2001, 322:460 463.
afferents in the rat's somatosensory cortex. J Neurosci
102." " Halsey NA, Hyman SL: Measles-mumps-rubella vaccine and
1994, 14:7594 7607.
autistic spectrum disorder: report from the New Challenges
83. Cases O, Vitalis T, Seif I, et al.: Lack of barrels in the soma-
in Childhood Immunizations Conference convened in Oak
tosensory cortex of monoamine oxidase A- deficient mice:
Brook, Illinois, June 12-13, 2000. Pediatrics 2001, 107:E84.
role of a serotonin excess during the critical period. Neuron
Conclusions from American Academy of Pediatrics Consensus Devel-
1996, 16:297 307.
opment Conference entitled, "New Challenges in Childhood Inmuni-
84. Lauder JM, Sze PY, Krebs H: Maternal influences on
zations." Available evidence does not support the hypothesis that the
tryptophan hydroxylase activity in embryonic rat brain.
measles-mumps-rubella vaccine causes autism.
Dev Neurosci 1981, 4:291 295.