New Developments in Understanding Autism:

1
New Developments in Understanding Autism

• W. Ted Brown, MD, PhD
• Director, New York State
• Institute for Basic Research
• In Developmental Disabilities

2
NEW YORK STATE AUTISM CONSORTIUM

• A public-private partnership
• 5 initial goals
• Professional training
• Early identification and intervention
• Epidemiology and registry
• Treatment research (lab school)
• Brain and tissue banking

3
Epidemiology of Autism

• Rates of autism are increasing.
• The reasons are unclear.

• Better diagnosis?
• Better recognition?
• Better programs?
• True increase?

4
How Common is Autism?

• The general observation is of a dramatic
  increase.
• The standard figure up to about 1980
• was 1/2000.
• Since then, a consistent increase has been
  observed.
• CDC (Feb 07) ? 6.6/1000 or about 1/150 with
  10.6/1000 in NJ.

5
How Common is Autism?
6
Autistic Children in CA System
7
California Data and Changes in DSM Criteria
8
Autistic Age Distribution 1987 vs 1998
9
Mental Ability and Autism
10
What is Autism?

• A poorly understood Developmental Disability
• Hallmark Severe impairment in social relations
• DSM IV - 5 different Pervasive Dev Disorders
• 1. Autism - 6 or more of 12 features
  involving
• social interactions,
• communication,
• repetitive stereotypical behaviors
• 2. Onset prior to three yrs of age
• 3. Not Rett syndrome or CDD
• (childhood disintegrative disorder)

11
Clinical and Research Identification of Autism

• ADI-R Autism Diagnostic Interview- Revised
• ADOS-G Autism Diagnostic Observation
  Schedule-Generic
• PDD-BI Pervasive Developmental Disability
  Behavior Inventory

12
Other PDDs

• Aspergers Disorder
• - Marked impairment in social interactions
• eye contact, peer relations, social sharing.
• - No delay in language, self help skills, or
• general cognitive development.
• - Apparently very common, but still is little
  known.

13
Other PDDs

• Rett Syndrome
• - Females
• - Onset of regression gt5 mo
• - Hand-wringing
• - MECP2 gene mutation

14
Other PDDs

• PDD-NOS (Pervasive Developmental Disorder -
  Not Otherwise Specified)
• Reserved for cases not meeting the full criteria
  for autism because of later age of onset,
  atypical or subthreshold symptoms.
• Concept of a broader spectrum
• Autistic Spectrum Disorders

15
Autism Spectrum Disorders
AUTISM
RETTS
CDD
ASPERGERS SYNDROME
PDD-NOS
16
Time to give up on a single explanation for
autism. Happe et al. Nature Neurocience, Oct 2006

• 3,400 eight year old twin pairs were studied.
• Modest to low correlations between autistic-like
  traits in the three core areas (social, language,
  sameness).
• Autism like traits are in a smooth continuum
  between individuals with ASD and the general
  population-no humps separating disease from not
  disease.
• Social/communication impairments correlated in
  the range of 0.2-0.4
• Considerable numbers showed isolated difficulty
  in one area - 59 with social difficulty had only
  social difficulty.
• Around 10 of all children showed only social
  impairment.
• Conclusion largely nonoverlapping genes act on
  each of these three traits.

17
Syndromal vs Pure Autism

• Pure Autism (non syndromal) -Limited to moderate
  mental retardation to normal intellectual
  functioning and no associated signs or symptoms
  (except seizures).
• Syndromal autism- one or more morphologic signs-
  estimated at 20
• Cohen et al JADD
  2005

18
Essential vs Complex AutismMiles et al AJMG 2005

• Complex autism (260 total)
• 5 microcephalic
• 16 dysmorphic - total 21
• Lower IQ plt.006
• More seizures plt.0008
• Less sib reoccurence 0 vs 4
• Less relatives with autism 9 vs 20
• Lower male to female ratio 3.21/6.51

19
Genetics of Autism

• Identical twins with autism - 70-90
• Sibling recurrence risk - 5
• ?A high heritability relative to diabetes,
  asthma, schizophrenia, hypertension, etc
• A polygenetic (8-15 genes), multi-factorial
  disorder with MF ratio of 41

20
Autism Genetic Etiologies?

• The high concordance in MZ twins indicates a high
  degree of genetic control.
• The rapid fall off of concordance in DZ twins
  suggested a multilocus, epistatic model.
• Caveats Epigenetics Mitochondria

21
Some Genetic Disorders associated with Autism

• Fragile X 25
• Tuberous Sclerosis 25
• Down syndrome 10
• Angelman syndrome 40
• Prader-Willi syndrome 25
• San Filippo syndrome 90
• Smith-Magenis (del 17p11.2) 90
• VCF/ DiGeorge (del 22q11) 25

22
Linkage Studies
Folstein and Rosen-Sheidley, Nature Rev Genet,
2001
23
Possible Autism Susceptibility Genes

• GENE
• Immune (HLA)
• Glutamate receptor (GluR6)
• Homeobox (HOXA1)
• Reelin (RELN)
• Speech1 (FOXP2) MET
• WNT2 EN2
• Serotonin Receptor (5-HTR-7)
• GABA Receptors (GABRB3)
• Ubiquitin-P Ligase (UBE3A/E6-AP)
• Serotonin Transporter (5-HTR7)
• Neuroligins (NLGN3 4)

• LOCATION
• 6p
• 6q
• 7p
• 7q22
• 7q31
• 7q32
• 10q
• 15q
• 15q
• 17q
• X

24
Genetics of Autism

• Preliminary genome wide screenings indicate loci
  on at least 21 chromosomes.
• Chrs 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12,
• 13, 15, 16, 17, 18, 19, 20, 22, X
• Cytogenetic Abnormalities have been reported in
  5-7 of cases
• IMGSAC et al. Hum. Mol Genet. 7(3)571-578, 1998
• Philippe et al. Hum Mol. Genet. 8(5)805-812,
  1999
• Liu et al. Am. J. Hum. Genet. 68327-340, 2001
• Yonan et al. Am J. Hum. Genet. 73(4)2003

25
Array-based comparative genomic hybridisation
identifies high frequency of cryptic chromosomal
rearrangements in patients with syndromic autism
spectrum disorders. Jacquemont et al. J Med
Genet, 43 843-9, 2006
Abnormalities found in 8/29 (27.5) patients 6
deletions and 2 duplications. Altered segments
ranged in size from 1.4 to 16 Mb.
26
(No Transcript)
27
Strong Association of De Novo Copy Number
Mutations with Autism Sebat, et al. Science 316,
445-9, (2007)

• Sampled 264 families, 118 simplex, 47
  multiplex, and 99 control families. Majority came
  from AGRE. Genome scans were performed by a form
  of comparative genomic hybridization (CGH)
  providing a mean resolution of 35 kb.
• CNVs were identified in 12 of 118 (10) of
  simplex, 2 of 77 (3) multiplex, and in 2 of 196
  (1) of controls.
• CNVs were in the range of 100kb-12mb.
• Adding the known rate of cytogenetically
  visible abnormalities, the total frequency of de
  novo variation detectable in sporadic cases is
  15 at this resolution.
• Affected genomic regions were highly
  heterogeneous and included mutations of single
  genes.
• These findings establish de novo germline
  mutation as a more significant risk factor for
  ASD than previously recognized.

28
Structural Variation of Chromosomes in Autism
Spectrum Disorder Marshall et al. AJHG 82,
477-88, (2008)

• Sampled 427 families. Genome scans were
  performed by CGH arrays.
• 277 unbalance CNVs were found in 44 of ASD
  families, not present in controls,.
• de novo CNVs were identified in 7 of
  simplex, 2 of multiplex
• CNVs were in the range of 100kb-20mb.
• Affected genomic regions were highly
  heterogeneous and included mutations of single
  genes (NLGN4, SHANK3, NRXN1,, etc) and regions
  previously associated with syndromes (22q11.2,
  16p11.2, etc).

29
CNVs may also be involved with Schizophrenia
Although complex disorders such as schizophrenia
have a heritable component, identifying the
genetic components associated has been very
difficult. Walsh et al (Sci April 08) found that
multiple, individually rare, structural mutations
(genomic microdeletions and microduplications)
occurred more frequently in 150 individuals with
schizophrenia than in controls. The enrichment
was more than threefold among schizophrenia cases
generally and more than fourfold among
schizophrenia cases with onset by age 18. The
genes disrupted by the genomic breakpoints of
mutations in the schizophrenia patients were not
random, but were disproportionately members of
pathways controlling neuronal signaling and brain
development.
30
Autistic children linked to same sperm donor
Gwenyth Jackaway is a professor at New York's
Fordham University. She's single but had always
wanted to have a child. So she contacted
California Cryobank, one of the largest sperm
donor banks in the country. Cryobank doesn't
reveal the identities of donors but allows people
to choose based on the traits they'd like their
child to have. Jackaway decided on "Donor X".
What she couldn't know then is that her son would
have autism. So she started to wonder whether
Donor X might carry a gene that could have
contributed. In six families Jackaway contacted
that had used Donor X, three of the children are
autistic, and one is showing signs of autism.
The cause or causes of autism are not known and
are hotly debated. Most experts believe that
genetics are a component, making a child
predisposed to autism or responsive to an
environmental trigger. Researchers have found
some genetic areas associated with autism, but it
could take years before the gene or genes that
cause autism or contribute to it will be
determined. Until then, Geri Dawson, chief
science officer for Autism Speaks, says there's
no way to screen for those genes and prevent them
from being passed to a child.
CNN April 04
31
PATERNAL AGE AND AUTISM
Offspring of men 40 years or older were 5.75
times (95 confidence interval, 2.65-12.46
Plt.001) more likely to have ASD compared with
offspring of men younger than 30 years, after
controlling for year of birth, socioeconomic
status, and maternal age. Advancing maternal age
showed no association with ASD after adjusting
for paternal age. Advancing Paternal Age and
Autism. Reichenberg et al. Arch Gen Psych
631026,2006.
Paternal age US 'controls' a AGRE autistic first child
  n2 492 308 n312
2029b 55.1 41.6c
3039 41.9 54.7
4049 3.0 3.7
c Plt0.005
Paternal age and autism are associated in a
family-based sample. Cantor et al. Mol Psych
12419-21,2007.
32
Advancing paternal age has been consistently
associated with increased risk of autism and
ASDs.

• Advanced paternal age has been associated with
  several congenital disorders, including Apert
  syndrome, craniosynostosis, neurofibromatosis,
  progeria, situs inversus, syndactyly, cleft lip
  and/or palate, hydrocephalus, and neural tube
  defects . In addition, advanced paternal age has
  been associated with schizophrenia and decreased
  intellectual capacities in the offspring.
• The most widely proposed mechanism underlying
  these congenital anomalies is known as the copy
  error hypothesis, first proposed by Penrose.
  After puberty, spermatocytes divide every 16
  days, and by the age of 35 years, approximately
  540 cell divisions have occurred. As a result, de
  novo genetic mutations that result from
  replication errors and defective DNA repair
  mechanisms are believed to propagate in
  successive clones of spermatocytes. These
  mutations accumulate with advancing paternal age
  and thus help explain how this disorder, which
  has a large genetic component, can be maintained
  in the population despite reduced reproduction in
  affected individuals.
• Prenatal and perinatal risk factors for autism a
  review and integration of findings. Kolevzon A,
  Gross R, Reichenberg A. Arch Pediatr Adolesc Med.
  161326-33. 2007

33
25 of very premature babies show autism signs
A study of toddlers found that about one-quarter
(26 of 91) of babies born very prematurely had
signs of autism on an early screening test. The
research is preliminary since formal autism
testing wasn't done. But the results are
provocative, suggesting that tiny preemies may
face greater risks of developing autism than
previously thought. That suggests autism may be
an under-appreciated consequence of medical
advances enabling the tiniest of premature babies
to survive, said lead author Dr Limperopoulos, a
researcher at McGill University in Montreal and
Children's Hospital in Boston.
Pediatrics April 08
34
Fever Improves Autism Symptoms Children with
autism appear to improve when they have a fever,
according to intriguing new research that could
lead to a better understanding of the
disorder.Fever was associated with less
hyperactivity, improved communication, and less
irritability in the study involving children with
autism and related disorders.Anecdotal reports
of improvements in autism symptoms related to
fever have circulated for years, but the research
represents the first scientific investigation
into the observed association.While kids with
autism might be expected to be calmer and less
hyperactive when they have fevers, the
improvement in communication and socialization
seen in the study suggests that fever directly
affects brain function says pediatric neurologist
Andrew Zimmerman, MD, of Baltimore's Kennedy
Krieger Institute
Curran et al. Pediatrics Dec 07
35
What Is Autism Caused by?
Dr Tom Insel, NIMH Director and chair of the NIH Interagency Autism Coordinating Committee, suggested In Autism, the phenotype is diverse Language development is diverse. Social cognition is variable. Regressive form seen in 30. CNV differences multiplex vs. simplex. Male only vs. female in family. Male ratio of 41 higher in high functioning.
? piRNA
36
piRNAs
The recent discovery of a new class of
mammalian small regulatory RNAs termed
PIWI-interacting RNA (piRNA) has extended the
diverse family of small regulatory RNAs.
piRNAs are a novel class of small RNAs isolated
from mammalian germline cells. piRNAs interact
with the Piwi subfamily of proteins and form a
ribonucleoprotein complex called Piwi-interacting
RNA complex (piRC). piRNAs consist of more than
50,000 (perhaps gt200,000 _at_ Betel PLoS 07)
different species, in contrast to several hundred
species of miRNAs. http//pirnabank.ibab.ac.
in/index.shtml
37
Small non-coding RNAs may play a key role in
Autism
http//pirnabank.ibab.ac.in/index.shtml piRNABank
is a web analysis system and resource
Sai Lakshmi S, Agrawal S.Nucleic Acids Research
2007 Sep 18
38
Enrichment and variability of PIWI-interacting
RNAs (piRNAs) in segmental duplications and copy
number variants (CNVs) suggest a functional role
in the integrity of the genome. Armengol,
Caceres, Brunet, X. Estivill (153 ASHG)
   PIWI-interacting RNAs (piRNAs) are a novel
class of small (30 nt) noncoding RNAs identified
in mammalian germline cells and constitute the
most abundant known class of genes with over
32,000 elements in humans. We have examined
piRNAs' organization in the human genome and have
found that currently known human piRNAs map to
70,736 sites and are structured in about 400
clusters, containing each at least 10 piRNAs. A
large proportion of the piRNA loci (about 65)
are located within repeat sequences, mainly LTRs
and LINE sequences, and over 50 contain repeated
units of a single piRNA, 71 being composed of
tandem copies of a unique piRNAs. Surprisingly,
over 25 of total piRNAs are located in regions
that contain segmental duplications (SDs) and
about 37 are within copy number variant (CNVs)
regions. In addition, 233 (58) and 220 (55)
piRNA clusters are within SDs and CNVs,
respectively. Similarly, a significant subset of
SDs (43), especially those with the highest
level of nucleotide identity, contains piRNAs.
Finally, we have confirmed experimentally that
the genomic sequences in which piRNAs are
embedded in vary in copy number in humans. Since
SDs and CNVs account for 5 and 12 of the human
genome sequence, respectively, the significant
enrichment of piRNAs is suggestive of a
functional role of these elements. This
association provides the first link between SDs
and CNVs with elements that could have a putative
functional role in the integrity of the genome.
39
The Fragile X Syndrome

• The leading cause of inherited mental
  retardation. It is associated with autism.
• This common mental retardation syndrome is due
  to the silencing of the FMR1 gene on the X
  chromosome.
• The FMR1 protein (FMRP) is an RNA binding
  protein and modifies the expression of a set of
  other genes.

40
Association of Fragile X and Autism

• 17 screening studies of 1,243 autistic subjects
  for FX had a range of 0 -16, and a mean of 4.
• 14 studies of 479 FX males had a range of 7- 60,
  and a mean of 24 with autism.
• A recent study by Don Bailey of 57 boys with FX
  found 25 with autism.

41
Association of Fragile X and Autism

• Behavioral characteristics found in many FX males
  are similar to that in autism (perseveration,
  stereotypies, regulating affect, attention,
  hyperactivity).
• Anxiety-based symptoms more common in FX (gaze
  aversion).
• Social interests, perception of others emotions,
  attachment to parents are higher in FX.

42
Chromosome 15

• A maternally inherited chromosomal duplication
  of the region 15q11-q13 has been found in 1-3 of
  several large autism samples.
• This region is paternally deleted in Angelman
  syndrome, and maternally deleted in Prader-Willi
  syndrome.
• Contains several potentially relevant genes
  Gaba receptor, Ubiquitin related gene.

43
Chromosome 15

• Angelman Syndrome Develpmental delay,
  retardation speech impairment, gait ataxia and
  happy behavior. Often microcephaly and seizures.
  1/12000
• Silence or disruption of UBE3A maternally derived
  gene
• Autism in Angelman 40

44
Chromosome 15

• Prader-Willi is the lack of paternal contribution
  within the same region 15q11-13
• Hypotonia, dev delay, and MR.
• Prevalence 1/29,000
• Autism in PW 25

45
Isodicentric 15 - idic(15)
46
IDIC(15)

• Phenotype includes MR, seizures, language
  disorders, autism, hypotonia
• One series- 8/9 exhibited autism
• Brain size tends towards small 20 microcephalic
• Autism shown to have larger brain size on
  neuropathologic exam, neuroimaging and is present
  in first degree relatives.

47
Brain Abnormalities in Autism

• Rare postmortem studies in autism (lt30).
  Typically larger brain size 10-20.
• Brain size enlarged by 10-20 using MRI during
  2-4 yrs, then falls off after 8-10 yrs.
• Smaller nerve cell size and more dense packing
  seen in internal regions limbic system and brain
  stem nuclei.
• Minicolumns are smaller and less compact.
• More white matter in frontal lobes, the
  cerebellum, various other areas, and associated
  with signs of inflamation.

48
Brain Abnormalities in Autism

• Dr. George Wegiel at IBR has studied 14 autism
  cases and age matched controls.
• Finds smaller and decreased numbers of neurons
  in many areas.
• Finds elevated levels of intracellular
  Beta-Amyloid in 40 of autistic specimens,
  unrelated to age.

49
Material   14 brains of autistic subjects and 14
age matched control brains Age range 4 to 64
years. Autism Tissue Program (Princeton)
coordinated selection of material for
morphometric study of cortical and subcortical
subdivisions in several European and US research
centers engaged in Autism Brain Atlas
Project. Inclusion/exclusion criteria. Selection
of brains was based on clinical,
neuropathological and technical inclusion and
exclusion criteria. This study of historically
the largest collection of brains of autistic
individuals and age matched control subjects is
the product of coordinated efforts of several
institutions and dozens of unnamed contributors.
50
Structures examined Striatum
Caudate Putamen
N. accumbens
Amygdala Lateral, basal, accessory
and central nucleus Cerebellum
Purkinje cells
Dentate nucleus Brain stem Nucleus of
facial nerve and
Nucleus olivaris inferior Entorhinal cortex
Layers II (stellate neurons), III, V, and
VI. Structures to be examined Examination of 11
other structures (in progress) will complete
comprehensive mapping of developmental and
age-associated changes in brains of
autistic subjects.
51
IDIOPATHIC AUTISM GRADING
OF DEVELOPMENTAL DEFICITS IN THE SIZE OF
NEURONAL BODY IN 4-7 YEAR OLD AUTISTIC
CHILDREN Nucleus accumbens -46 Amygdala
(lateral basal nucleus) -29 Putamen
-27 Dentate nucleus -25 Caudate -20 Nucleus
of facial nerve 5 (not significant)
52
IDIOPATHIC AUTISM
PERCENTAGE OF SMALL NEURONAL NUCLEI IN
AUTISTIC AND CONTROL PEOPLE Structure
Neuronal Autism Control
Ratio
nucleus
autism/
volume µm3
control Amygdala. Lateral
n lt300 45 15
2.5x Amygdala. Lat.Bas.n. lt300
46 15 3.0x N.
accumbens lt200
82 12 6.8x Putamen
lt200 66 31
2.1x Caudate
lt200 72 27
2.7x Dentate n lt300
55 18 3.0x N.
olivaris lt400
47 16 2.9x
Average
59 20 3.0x
N. Facial n.
lt1,500 50 58
0.9x
53
Triple Hit Hypothesis

• UNDERLYING VULNERABILITY
• EXOGENOUS STRESSOR
• CRITICAL PERIOD OF BRAIN DEVELOPMENT

54
Some fundamental questions

• How many autisms are there?
• Is there is a real autism and one or more
  pretenders?
• Should autism be studied as an entity or as part
  of a continuum?
• Where do we draw the lines? Who are we excluding?
• What do words like co-morbid and dual diagnosis
  really mean?