Download The location and type of mutation predict

© 2000 Oxford University Press
Human Molecular Genetics, 2000, Vol. 9, No. 20 3019–3028
The location and type of mutation predict malformation
severity in isolated lissencephaly caused by
abnormalities within the LIS1 gene
Carlos Cardoso1,+, Richard J. Leventer1,2,+, Naomichi Matsumoto4, Julie A. Kuc1,
Melissa B. Ramocki1, Stephanie K. Mewborn1, Laura L. Dudlicek1, Lorraine F. May1,
Patti L. Mills1, Soma Das1, Daniela T. Pilz5, William B. Dobyns1,2,3 and David H. Ledbetter1,3,§
Departments of 1Human Genetics, 2Neurology and 3Pediatrics, University of Chicago, 920 East 58th Street, Chicago,
IL 60637, USA, 4Department of Human Genetics, Nagasaki University School of Medicine, Sakamoto 1-12-4,
Nagasaki 852-8523, Japan and 5Institute for Medical Genetics, University Hospital of Wales, Heath Park, Cardiff
CF14 4WX, UK
Received 9 August 2000; Revised and Accepted 18 October 2000
Lissencephaly is a cortical malformation secondary
to impaired neuronal migration resulting in mental
retardation, epilepsy and motor impairment. It shows
a severity spectrum from agyria with a severely thickened cortex to posterior band heterotopia only. The
LIS1 gene on 17p13.3 encodes a 45 kDa protein
named PAFAH1B1 containing seven WD40 repeats.
This protein is required for optimal neuronal migration by two proposed mechanisms: as a microtubuleassociated protein and as one subunit of the enzyme
platelet-activating factor acetylhydrolase. Approximately 65% of patients with isolated lissencephaly
sequence (ILS) show intragenic mutations or deletions of the LIS1 gene. We analyzed 29 non-deletion
ILS patients carrying a mutation of LIS1 and we
report 15 novel mutations. Patients with missense
mutations had a milder lissencephaly grade
compared with those with mutations leading to a
shortened or truncated protein (P = 0.022). Early truncation/deletion mutations in the putative microtubule-binding domain resulted in a more severe
lissencephaly than later truncation/deletion mutations (P < 0.001). Our results suggest that the lissencephaly severity in ILS caused by LIS1 mutations
may be predicted by the type and location of the
mutation. Using a spectrum of ILS patients, we
confirm the importance of specific WD40 repeats and
a
putative
microtubule-binding
domain
for
PAFAH1B1 function. We suggest that the small
number of missense mutations identified may be due
to underdiagnosis of milder phenotypes and hypothesize that the greater lissencephaly severity seen
in Miller–Dieker syndrome may be secondary to the
+These
§To
loss of another cortical development gene in the
deletion of 17p13.3.
INTRODUCTION
Lissencephaly (LIS) is a diffuse cortical malformation
resulting from incomplete neuronal migration during early
brain development. It encompasses a continuous spectrum of
malformations from complete agyria to variable degrees of
agyria and pachygyria to subcortical band heterotopia (SBH)
only (1). Clinical manifestations range from profound mental
retardation, intractable epilepsy, spasticity and reduced
longevity (2) to milder forms with infrequent seizures and
intellectual disability only (3). Lissencephaly is one component of the contiguous gene deletion disorder known as Miller–
Dieker syndrome (MDS) or may occur as an isolated brain
malformation in patients with isolated lissencephaly sequence
(ILS) (2). The clinical severity generally correlates with the
degree of agyria and cortical thickening (4). Approximately
80% of typical LIS spectrum patients show abnormalities of
either the LIS1 gene or the DCX gene (alternatively known as
XLIS) (5). The majority of patients with LIS show deletions or
mutations of the LIS1 gene whereas the majority of patients
with typical SBH show mutations of the DCX gene (5).
Patients with MDS and ILS have been shown to carry heterozygous mutations or deletions of LIS1 suggesting haplo-insufficiency of the LIS1 gene product as the pathogenic mechanism
(6). The functions of the LIS1 protein in normal brain development are currently being elucidated, yet our understanding of
the functional domains within this protein and the biological
basis for the phenotypic spectrum of LIS remains incomplete.
LIS1, located on chromosome 17p13.3, encodes a protein of
410 amino acids named PAFAH1B1, which contains seven
WD40 repeats and has a molecular weight of ∼45 kDa (7).
PAFAH1B1 is highly conserved showing only a single amino
acid difference between mouse and human (8), three amino acid
differences between human and bovine (9), 70% identity with
authors contributed equally to this work
whom correspondence should be addressed. Tel: +1 773 834 0525; Fax: +1 773 834 0505; Email: [email protected]
3020 Human Molecular Genetics, 2000, Vol. 9, No. 20
Drosophila LIS1 (10) and 29% homology with the yeast
homolog, PAC1 (11). In addition, it shares 42% homology
with the nuclear migration protein NudF in Aspergillus
nidulans (12). PAFAH1B1 forms the non-catalytic subunit of
the G protein-like heterotrimeric cytosolic platelet-activating
factor acetylhydrolase (PAF-AH) brain isoform Ib
(PAFAH1B1) (13). Together with the 29 kDa subunit
(PAFAH1B3) and the 30 kDa subunit (PAFAH1B2),
PAFAH1B1 forms a trimeric complex. This protein complex
regulates the level of platelet-activating factor (PAF) in the
brain through the removal of the acetyl group of PAF at the
sn-2 position to produce biologically inactive lyso-PAF (14).
Optimal concentrations of PAF may be necessary for correct
neuronal migration. The precise role of PAF in neuronal
migration is uncertain, although it has been proposed that PAF
affects migrating neurons by influencing cell morphology or
adhesion properties (15). Studies demonstrate that addition of
PAF or inhibition of PAF-AH decreases migration of cerebellar granule cells in vitro (16,17) and addition of PAF to
cultured hippocampal neurites produces growth cone collapse,
neurite retraction and neurite varicosity formation (18). The
three subunits of PAF-AH are co-expressed in the developing
brain in regions containing migrating neurons. The highest
expression level for each of the three subunits is concomitant
with peak times of neuronal migration (15,19). The targeted
knockout of the LIS1 gene created a mouse model for LIS.
Homozygous null mice die early in embryogenesis whereas
heterozygous mice survive, showing evidence of delayed
neuronal migration (cortical, hippocampal, cerebellar and
olfactory bulb disorganization) confirmed by in vitro and in
vivo cell migration assays (20).
Recent evidence suggests that an important domain for
PAFAH1B1 function may be localized to the second WD40
repeat at the region where interactions with the other two PAFAH subunits occur (21,22). Apart from its role as a subunit of
PAF-AH, PAFAH1B1 may enhance neuronal migration by
acting as a microtubule-associated protein (MAP). Evidence
for this role comes from studies that demonstrate the co-localization and interaction of PAFAH1B1 with tubulin, the major
component of microtubules. This interaction results in the
developmentally regulated phosphorylation of PAFAH1B1
(23). The co-assembly of PAFAH1B1 with microtubules
results in a reduction of microtubule catastrophic events, and
thus an increase in the length of microtubules (24). This would
be predicted to enhance and maintain microtubule elongation,
a process necessary for optimal nuclear (and thus neuronal)
migration. In vitro studies show that the microtubule-associated protein-binding domain (MAP-D) of the PAFAH1B1
protein is near the N-terminus (23), but the precise boundaries
and characteristics of this domain are not well defined.
As part of our ongoing lissencephaly research project, we
have collected a large number of patients with lissencephaly
secondary to abnormalities of the LIS1 gene. The goal of this
study was to investigate patients with ILS and a mutation
within LIS1 to determine the pathophysiological basis of the
spectrum of brain malformations seen in patients with LIS1
mutations. To do this, we performed an analysis of the correlation between mutation type and phenotype (lissencephaly
severity grade) assessed by magnetic resonance imaging
(MRI) or computed tomography (CT). We performed detailed
mutation analysis using direct gene sequencing, Southern blot
and RT–PCR analysis and identified 15 novel mutations, not
previously reported, the majority of which result in protein
truncation or shortening due to intragenic deletion, frameshift
or exon skipping (herein referred to as ‘truncation/deletion
mutations’). By comparing mutation type and location to the
lissencephaly severity grade, we show that missense mutations
are usually less severe than truncation/deletion mutations and
that more 5′ truncation/deletion mutations produce a more
severe phenotype than similar mutations located towards the 3′
region of LIS1. Our results, based on the study of a large
number of patients with a spectrum of LIS phenotypes, demonstrate that the mutation type and location predict malformation
severity and that this correlation is consistent with the current
understanding of the functional domains of the LIS1 gene.
RESULTS
Range and distribution of LIS1 mutations in ILS
Among 65 patients with ILS and no LIS1 deletion by fluorescence in situ hybridization (FISH), we found 29 unrelated
patients (18 boys and 11 girls) carrying a heterozygous LIS1
mutation, 15 of which have not been reported previously.
Mutation analysis was performed by direct sequencing and
confirmed in both the forward and reverse directions. In five of
these patients (LP85-002, LP88-009, LP90-012, LP97-064 and
LP91-022), we detected a LIS1 rearrangement by Southern blot
analysis and used a restriction map from the genomic sequence
of the LIS1 gene (see Materials and Methods) to determine the
extent of the coding region deletions (Fig. 1 and Table 1). For
patients LP85-002 and LP91-022, subsequent RT–PCR
analysis followed by direct sequencing of the cDNA confirmed
the deletion of exon 2 for LP85-002 and the deletion of exons
3–6 for LP91-022 that was previously detected by Southern
analysis (Fig. 1). Lymphoblastoid cell lines were not available
on patients LP88-009, LP90-012 and LP97-064; therefore,
further studies to delineate the deletion boundaries could not
be performed.
For the remaining 24 patients, direct sequencing detected
five missense mutations, six splicing defects confirmed by
RT–PCR analysis, five nonsense mutations and eight
frameshift mutations (Fig. 2, Tables 1 and 2). Only two mutations were recurrent, including a splice donor site change
(T→C) in intron 6, and a C→T substitution at position 817 in
exon 8 resulting in a protein truncation (R273X). Each of these
two mutations was found in two unrelated patients. Missense
mutations were found in exons 3, 6 and 9, with three (H149R,
G162S and S169P) clustered in exon 6 which corresponds to
the coding region for the second WD40 repeat (Fig. 2 and
Table 2). Four of the five missense mutations were demonstrated to be de novo (Table 2). For patient LP98-063 (amino
acid change D317H), parental samples were unavailable for
further analysis but the mutation was not found in 100 normal
chromosomes sequenced (Table 2). Large intragenic deletions
were clustered between exons 2 and 6. Other mutation types
(nonsense mutations, small insertions and deletions and
splicing mutations) were found scattered throughout the gene
except in the third WD40 repeat encoded by exon 7.
Sequencing was performed on the available parents and
showed that the mutations were de novo.
Human Molecular Genetics, 2000, Vol. 9, No. 20 3021
Figure 1. Example of LIS1 mutation detection demonstrated using patient LP91-022 with an intragenic deletion of LIS1 detected by the sequential use of Southern
analysis, RT–PCR and direct cDNA sequencing. (a) Southern blot analysis using EcoR1, showing a band at ∼20 kb (double asterisk) which is absent in the control.
The stronger band at ∼7.5 kb (single asterisk) corresponds to the LIS1 pseudogene localized to chromosome 2 (37) and the other bands correspond to the LIS1
cDNA (see Materials and Methods). (b) RT–PCR, showing the presence of an abnormal transcript of LIS1 of 250 bp (arrow). (c) Direct sequencing chromatogram
of the abnormal transcript in b shows the absence of exons 3–6 as indicated in the overlying schematic representation of LIS1.
Table 1. LIS1 protein truncation mutations and lissencephaly severity grade found in ILS patients
Patients
Sex
Exon
Mutation
Type of mutation
LIS grade
LP85-002
Female
2
∆exon 2
Deletion
2
This report
LP88-009
Female
2+3
Rearrangement Southern blot
Deletion
2
This report
LP90-012
Female
3+4+5
Rearrangement Southern blot
Deletion
2
5
LP97-064
Female
3+4+5
Rearrangement Southern blot
Deletion
2
5
LP91-022
Male
2–7
∆exons 3, 4, 5 and 6
Deletion
3
This report
LP97-068
Female
4
162insA
Frameshift
2
5
LP99-053
Male
4
166delA
Frameshift
4
This report
LP99-058
Male
5
R89X
Nonsense
3
This report
LP97-078
Male
6
R144X
Nonsense
3
5
LP99-014
Male
6+7
569–10 bp T→C
Splicing defect
3
This report
LP99-088
Female
6+7
569–10 bp T→C
Splicing defect
3
This report
LP97-071
Female
8
703delGA
Frameshift
3
5
LP91-019
Female
8
805delA
Frameshift
3
5
LP89-005
Male
8
R273X
Nonsense
3
6
DP-0201
Male
8
R273X
Nonsense
4
This report
LP97-086
Female
8
S289X
Nonsense
3
This report
LP99-039
Male
8
881delCT
Frameshift
3
This report
LP97-021
Male
9
903insT
Frameshift
3
5
LP93-005
Male
9
988del 22bp
Splicing defect
4
6
LP91-012
Male
9 + 10
1002+1 bp G→A
Splicing defect
3
This report
LP97-069a
Female
9 + 10
1002+5 bp G→A
Splicing defect
3
5
LP90-014
Male
10
1018insT
Frameshift
4
5
LP98-100
Male
10
1049insG
Frameshift
3
This report
LP90-020
Male
10 + 11
1159+3 bp ins CA
Splicing defect
3
5
aIn
Pilz et al. (5), LP97-069 was incorrectly numbered as LP97-067.
Reference
3022 Human Molecular Genetics, 2000, Vol. 9, No. 20
Figure 2. Summary of LIS1 mutations in 29 ILS patients. Missense mutations are denoted by green circles above the gene and nonsense and frameshift mutations
by blue arrows, splicing defects by orange boxes and deletions by solid black lines (indicating the deleted region) below the gene. An asterisk denotes recurrent
mutations. Red shading of the gene indicates the putative microtubule associated protein-binding domain and gray shading indicates the seven WD40 repeats.
Table 2. LIS1 missense mutations and lissencephaly severity grade found in ILS patients
Patients
Sex
Exon
Mutation
Nucleotide change
Domain
De novo
LIS grade
Reference
LP97-077
Female
3
F31S
92T→C
MAP-D
Yes
4
This report
LP93-012
Male
6
H149R
446A→G
WD2
Yes
3
5
LP99-127
Male
6
G162S
486C→T
WD2
Yes
4
This report
LP94-051
Male
6
S169P
499T→C
WD2
Yes
6
36
LP98-063
Male
9
D317H
949G→C
WD5
–
4
This report
Two silent polymorphic variants, an A→T change at nucleotide position 687 in exon 8 (T229) and a C→T change at
nucleotide position 1188 in exon 11 (V396) were found in two
unrelated boys and their mothers. We also identified two polymorphisms in the 3′ untranslated region (UTR) of the LIS1
transcript, a C→T substitution at nucleotide position 1250 (24)
and a T→G substitution at nucleotide position 1236 which we
have observed in normal individuals (unpublished data).
Characterization of the putative MAP-D
In vitro and in vivo analyses show that PAFAH1B1 interacts
with tubulin at the N-terminal region (23,25). By in silico
analysis (see Materials and Methods) we attempted to determine whether this region contains a domain with homology to
other motifs allowing protein–protein interaction. We found a
coiled-coil domain in the N-terminal region (Fig. 3a and b)
conserved in all LIS1 proteins and also present in the LIS1
homolog NudF (Fig. 3c). We also identified the presence of a
moderate number of basic amino acids (such as lysine, arginine
and histidine), one specific KKEX motif and three related
KKXX repeats characteristic of the MAP 1B (26) (Fig. 3c).
Genotype–phenotype correlation
MRI scans were available for review on 26 patients and CT
scans were available for the remaining 3 patients. The lissencephaly severity ranged from grade 2 to grade 6, excepting
grade 5 (see Materials and Methods; Fig. 4). Of these, 17% had
diffuse agyria with shallow sulci over the frontal and temporal
poles (grade 2), 55% had posterior agyria and frontal pachygyria (grade 3), 24% had diffuse pachygyria that was more
severe posteriorly (grade 4) and 4% had posterior subcortical
band heterotopia (grade 6). Patients with missense mutations
were grades 3, 4 and 6, and those with other mutations ranged
from grade 2 to grade 4 (Tables 1 and 2).
As shown in Table 3, patients with missense mutations had a
significantly milder mean lissencephaly severity grade
compared with those with truncation/deletion mutations (P =
0.022). Patients with truncation/deletion mutations were
Human Molecular Genetics, 2000, Vol. 9, No. 20 3023
Figure 3. Delineation of the putative microtubule-associated protein domain of PAFAH1B1. (a) Graphical representation of the output from the coils program (see
Materials and Methods) demonstrating that PAFAH1B1 (GenBank accession no. P43034) has a 1.0 probability of forming a coiled-coil structure in the N-terminal
region. (b) The region of PAFAH1B1 corresponding to amino acids 44–78 are listed corresponding to their position within each heptad repeat of the coiled-coil
structure. The boxed regions contain the hydrophobic amino acids which are predicted to lie in positions a and d of the heptad repeat forming the coiled-coil motif
(4-3 hydrophobic repeat). (c) Conservation of the coiled-coil domain in PAFAH1B1 in the species bovine (GenBank accession no. P43033), chicken (AAF18938),
mouse (P43035), Drosophila (AAD38390), Caenorhabditis elegans (AAF82632) and two homologous proteins in Aspergillus nidulans (NudF; AAA91301) and
yeast (PAC1; S67166). For LIS1 from rat and zebrafish, the protein sequences were predicted using the translation of the expressed sequence tag available in GenBank. The N-terminal region of PAFAH1B1 shows homology to MAP-1B, which is lysine- and arginine-rich and the occurrence of one specific KKEX and three
related KKXX repeats (denoted by boxed regions).
Figure 4. Axial T2-weighted MRI of four patients representing the four lissencephaly grades found in patients in this study. The lissencephaly grade is noted in
the upper left-hand corner of each image. Lissencephaly severity decreases with increasing grade (left to right). Grade 2 shows predominant agyria (lack of gyri)
with minimal pachygyria (broadened gyri), grade 3 shows mixed agyria/pachygyria, grade 4 shows pachygyria only and grade 6 shows subcortical band heterotopia
(arrows).
3024 Human Molecular Genetics, 2000, Vol. 9, No. 20
Table 3. Mean lissencephaly severity grade by LIS1 gene mutation type
Lissencephaly grade
Missense mutations
(n = 5)
Truncating mutations P value
(n = 24)
4.20
2.96
0.022
Table 4. Mean lissencephaly severity grade by location of truncating LIS1
gene mutation
Lissencephaly grade
Early truncating
(n = 6)
Late truncating
(n = 18)
P value
2.43
3.18
<0.001
divided into two groups based on an estimate of the boundary
of the putative MAP-D spanning exon 2 to the beginning of
exon 5 (at the end of the coiled-coil domain) (24). Mutations in
this region predicted to compromise the MAP-D were named
‘early truncation/deletion’ mutations, whereas the remaining
were named ‘late truncation/deletion’ mutations. As shown in
Table 4, patients with early truncation/deletion mutations had a
significantly more severe mean lissencephaly grade compared
with those with late truncation/deletion mutations (P < 0.001).
The functional consequences of the LIS1 mutation in patient
LP99-053 are uncertain as he has a relatively mild LIS grade
despite a mutation at the end of the putative MAP-D.
Mutant PAFAH1B1 protein was not detected in patients
with truncation or deletion mutations
In an attempt to detect the presence of mutant PAFAH1B1
protein in patients with mutations leading to a truncated or
deleted protein that could correlate with the different lissencephaly grades, we performed western blot analysis on a subset
of 11 patients. For our experiments, we used a commercial
polyclonal antibody directed against the N-terminal region of
PAFAH1B1 on total cellular protein extracts from lympho-
blastoid cell lines of three normal individuals as negative
controls, three patients carrying missense mutations as positive
controls (LP97-077, LP93-012, LP94-051) and eight patients
with truncation or deletion mutations (LP90-012, LP97-068,
LP97-078, LP91-019, LP89-005, LP97-021, LP93-005 and
LP97-069). A band of ∼45 kDa, attributable to PAFAH1B1,
was detected in the normal control and in all patients. However,
we were unable to detect the presence of truncated or abnormal
proteins (<45 kDa) in any of these patients (Fig. 5).
We also performed simultaneous and sequential hybridization of PAFAH1B1 with a monoclonal antibody directed
against actin (as an internal normalization control of protein
expression) to determine whether the expression level of
PAFAH1B1 was reduced in these patients. Although patients
with truncation/deletion mutations appear to have a reduction
of PAFAH1B1 by this method, the differences in the levels
between patient groups could not be consistently discriminated
within the sensitivity of densitometry, indicating that the polyclonal PAFAH1B1 antibody might not be suitable for accurate
quantitative analysis.
DISCUSSION
Lissencephaly is one disorder in a large group of malformations of cortical development with increased recognition since
the widespread use of MRI scanning. Although individually
rare, collectively these malformations are responsible for a
significant proportion of epilepsy and chronic neurological
disability, especially in childhood. Through molecular genetic
analysis of these patients, it is possible to identify genes that
are essential to normal cortical development. Once these genes
are identified, the task is to determine the precise function of
the protein products in complex cortical development pathways, especially during the stage of neuronal migration. Elucidation of the aberrant developmental mechanisms that generate
these malformations requires the study of affected patients and
mutant models to complement the studies of normal protein
function.
Optimal functioning of the PAFAH1B1 protein appears to be
a critical requirement for normal development of the human
Figure 5. Immunoblot analysis of ILS patients carrying missense mutations and mutations predicted to produce truncated or shortened proteins. All ILS patients
with mutations leading to a shortened or truncated protein (LP90-012, LP97-068, LP97-078, LP91-019, LP89-005, LP97-021, LP93-005 and LP97-069) showed
expression of an ∼45 kDa PAFAH1B1 protein similar to controls (normal individuals) or the patients with missense mutations (LP97-077, LP93-012 and LP94051) and did not show the presence of abnormal or shortened PAFAH1B1 mutant protein.
Human Molecular Genetics, 2000, Vol. 9, No. 20 3025
cortex. The fact that PAFAH1B1 is so highly conserved across
species confirms its evolutionary importance. It was not until
very recently that the spectrum of malformation severity in
lissencephaly was appreciated. There are few human data
available to explain the functional basis of this range of malformation severity. Such data will have important implications
both to elucidate the functional regions of PAFAH1B1 and to
improve diagnosis and genetic counseling. This is the largest
study to date of ILS patients with abnormalities within the LIS1
gene that attempts to explain the spectrum of severity seen in
ILS based on the nature of the mutation and its functional
consequences.
Functional domains of PAFAH1B1
Recent studies suggest the existence of two important functional domains within PAFAH1B1. We divided our patients
into three groups (missense mutations, early truncation/
deletion and late truncation/deletion mutations) according to
the predicted functional consequences of their mutations based
on these domains. The first of these is a poorly defined MAP-D
at the N-terminal region (24). Based on in vitro studies, this
region of PAFAH1B1 may interact with tubulin, leading to
microtubule assembly and elongation, a process critical to
neuronal migration (24). Using in silico analysis we confirmed
the presence of a coiled-coil domain spanning most of the Nterminal region that may act as a domain for protein–protein
interaction (27). This domain is also present in PAFAH1B1 of
other species and in two homologous proteins, NudF and
PAC1 (11,28). Such domains allow proteins such as
PAFAH1B1 to interact with themselves to form homodimers,
or with other proteins (such as tubulin) to form heterodimers
(29). This finding is consistent with experiments performed by
Sapir et al. (24) that suggest the presence of a domain of interaction between PAFAH1B1 and tubulin at the N-terminal
region of PAFAH1B1. Further studies, such as a yeast twohybrid analysis using constructs expressing different portions
of the N-terminal region and site-directed mutagenesis
analysis, will be required to better define the amino acids
within this domain that are essential for the interaction with
tubulin or other proteins. The second proposed functional
domain of PAFAH1B1 is the region containing seven WD40
repeats, with greatest functional significance involving the
second WD40 repeat encoded by exon 6. This particular part of
the domain appears to be critical for the correct interaction of
PAFAH1B1 with the other two subunits of PAF-AH (21,22).
As hypothesized, the three groups of patients that we defined
according to the predicted consequences of their mutations
within these functional domains have a statistically different
phenotype as measured by lissencephaly severity grade.
Patients with missense mutations
We report five missense mutations among the 29 patients of
which three are not previously reported. The low number of
missense mutations strongly suggests that patients with such
mutations may be underdiagnosed. An alternative explanation
is that most missense mutations are lethal, although our finding
of milder phenotypes among patients with missense mutations
argues against this. With the exception of patient LP93-012,
who has a point mutation which alters the critical histidine
amino acid in the second WD40 repeat, all missense mutations
result in a relatively mild phenotype (grades 4 and 6),
including one patient with posterior band heterotopia and
another with occipital pachygyria only. Three of the five
missense mutations affect the second WD40 repeat, a domain
necessary for the interaction of PAFAH1B1 with the other two
subunits of PAF-AH (22). These mutations are predicted to
result in either a partial alteration or a complete loss of PAFAH activity. Although the patient numbers are small, we found
a significant difference in malformation severity between
patients with missense and truncation/deletion mutations (P =
0.022) suggesting that missense mutations generally produce a
less severe phenotype. Analysis of a greater number of patients
with missense mutations is required. Ascertainment of such
patients will require an increased referral of patients with
milder phenotypes (such as partial band heterotopia and localized pachygyria) for LIS1 sequencing.
Patients with mutations leading to a truncated or
shortened protein
Our observation that the majority of LIS1 mutations result in
protein truncation confirms the published results of previous
studies (5,6,30). For patients with truncation/deletion mutations, our data show that mutations resulting in the most severe
phenotype (grade 2) cluster in the N-terminal region of the
LIS1 gene from exon 2 to 5. All such patients have deletions or
mutations that compromise the putative MAP-D or the MAP-D
and the PAF-AH subunit interaction domain, producing a more
severe defect of neuronal migration and thus a more severe LIS
grade. Although the numbers are small, this trend is supported
by results showing that patients with 5′ truncation/deletion
mutations have a more severe LIS grade than do patients with
later truncation/deletion mutations (P < 0.001), which would
be predicted to spare the MAP-D.
Evidence from patients with ILS secondary to large deletions
of 17p13.3 supports the hypothesis that the N-terminal region
of LIS1 contains a domain of functional significance. A review
of our FISH data shows that the majority of patients with a
telomeric deletion including the 5′ end of LIS1 (defined by a
deletion of the D17S379 probe and/or the c37E9 cosmid) have
grade 2 LIS and the majority of patients with a deletion of the
3′ end of LIS1 only (defined by a deletion of the c120A7
cosmid) have grade 3 or 4 LIS (31) (unpublished data). Unfortunately, the boundaries of the large deletions observed in
these patients are not precisely defined. The most severe form
of lissencephaly (grade 1) has been found only in patients with
MDS, which is a contiguous gene deletion syndrome of
17p13.3 including a deletion of LIS1. This may suggest the
presence of a second gene required for neuronal migration
located within the MDS deletion in addition to the deletion of
LIS1. Deletion of these two genes would explain the severe
LIS seen in MDS. Identification of this second gene will
require an improvement of the physical and transcriptional
map of 17p13.3 followed by testing of candidate genes in suitable patients.
Graded function of PAFAH1B1 in patients with intragenic
mutations or internal deletions of the LIS1 gene
The data from our patient group support the conclusions of
previous studies suggesting haploinsufficiency of LIS1 as the
cause of LIS in most patients (2). In addition, our data suggest
3026 Human Molecular Genetics, 2000, Vol. 9, No. 20
that the location and type of mutation of the LIS1 gene may
account for the phenotypic variability now evident in patients
with LIS. This further implies that patients with missense
mutations and mutations predicted to lead to a truncated or
shortened protein should result in the production of a partially
functional protein. According to our genotype–phenotype
correlation data, we expect the mutant PAFAH1B1 proteins
caused by missense mutations to be more stable and functional
than those caused by truncation/deletion mutations and 3′
mutations should result in a more stable or functionally active
protein than 5′ mutations.
We did not observe an altered protein by western blot analysis
of 11 patients. Fogli et al. (30), have previously performed a
similar analysis using cells from four LIS patients, all of whom
had a mutation predicted to lead to protein truncation or shortening. They reported reduced PAFAH1B1 protein levels with
truncation mutations but no abnormal protein product in three of
four patients. For these 14 patients, it is possible that mutant
proteins were not detected due to instability and rapid protein
degradation (32). Alternatively, it is possible that such proteins
are expressed and functionally active only during development
in the time window of neuronal migration. We are aware of other
studies that fail to show partial PAFAH1B1 recombinant protein
products, even when the truncation or deletion is small (A.
Wynshaw-Boris, personal communication). Further studies of
PAFAH1B1 expression are required to detect differences in
PAFAH1B1 levels between patients with different mutation
types. Such studies should include western blots using a specific
monoclonal antibody directed against PAFAH1B1 and
immunocytofluorescence techniques designed to detect the
localization and distribution of PAFAH1B1.
One patient, LP99-053, has a relatively mild LIS grade (grade
4) despite a truncation mutation located within the latter part of
the putative MAP-D. Further study of this patient may help in
defining the precise 3′ boundary of the region of the
PAFAH1B1 domain that is essential for the interaction with
tubulin or other proteins required for cell migration. For our
analysis, we included this sample as an early truncation/
deletion mutation, but further elucidation of the boundaries of
this domain may classify this sample as a late truncation/
deletion mutation. An additional important observation from
our data is the absence of any mutations in exon 7 (coding for
the third WD40 repeat). This raises the question of whether this
domain has great functional importance (with mutations
resulting in lethality through a dominant negative effect) or
little functional importance. The answer to this question will
come only with further functional studies of PAFAH1B1 or
with the identification of patients with mutations within exon 7.
Using data derived from a large number of ILS patients, our
results suggest that the phenotypic spectrum observed in these
patients may be a result of the functional consequences of the
specific location and type of mutation. Our results support the
hypothesis that PAFAH1B1 may contain two important functional domains. However, an understanding of the relative
importance of these domains in the complex process of cortical
development remains incomplete. Further studies are required
to determine the precise role of PAF (and consequently
PAF-AH) in neuronal migration. Evidence from the study of
other human cortical malformations such as band and nodular
heterotopia demonstrates the importance of cytoskeletal
proteins such as doublecortin (33) and filamin 1 (34) in
neuronal migration. Consideration of PAFAH1B1 as a MAP is
strengthened by the presence of an appropriate functional
domain at the N-terminal region and our finding that patients
with the most severe LIS grade had mutations that compromise
this domain. Whether PAFAH1B1 interacts directly with
tubulin or indirectly through other proteins remains unknown
at present. Appropriate functional experiments combined with
the study of patients with cortical malformations, may help in
answering this question as well as in identifying other genes
involved in cortical development pathways.
MATERIALS AND METHODS
Patients
Patients were ascertained through the ongoing Lissencephaly
Research Project from a variety of sources including clinical
referrals from physicians, contact through the Lissencephaly
Network and direct inquiry from parents of affected children.
Patients were included in the study if they had ILS and a mutation of the LIS1 gene. ILS patients with deletions detected by
FISH were excluded from the study, as determination of the
precise boundaries of their deletions could not always be
assessed with certainty. MRI (n = 26) and CT (n = 3) scans
were reviewed and each patient’s malformation was classified
using an established classification scheme for LIS which
involves both the severity (grades 1–6) and the gradient (1).
According to this scale, LIS grade ranges from generalized
agyria (grade 1), to variable degrees of agyria and pachygyria
(grades 2 and 3), to generalized pachygyria (grade 4), to mixed
pachygyria and SBH (grade 5) to SBH alone (grade 6). The
LIS gradient may be either posterior→anterior (‘a’ gradient) as
observed with LIS1 mutations, or anterior→posterior (‘b’
gradient) as observed with DCX mutations. The precise mutation data were not available to the investigators at the time of
neuroimaging review. Blood samples from probands were
obtained with informed consent and all protocols were
approved by the appropriate Institutional Review Board
Human Subjects Committee.
DNA extraction, PCR amplification and direct DNA
sequencing
DNA was extracted from lymphoblast cell lines or peripheral
blood using a Puregene DNA isolation kit (Gentra System,
Minneapolis, MN) according to the manufacturer’s protocol.
Ten LIS1 exons covering the coding region (exons 2–11) were
amplified by PCR. PCR was cycled 35 times at 94°C for 30 s,
58°C for 30 s, 72°C for 1 min in a volume of 50 µl, containing
1× PCR buffer with 1.5 mM MgCl2, 0.2 mM each dNTP, 1 µM
each primer and 2.5 U of TaqGold polymerase (PE Applied
Biosystems, Foster, CA). Primer sequences have been
described previously (5). PCR products were purified using the
QIAquick PCR purification kit (Qiagen, Chatsworth, CA) and
sequenced on both strands with BigDye Terminator chemistry
by a standard protocol (PE Applied Biosystems) as described
previously (5). The resulting sequences from the patients and
normal controls were aligned and compared with the published
sequences deposited in GenBank (accession nos U72333–
U72342).
Human Molecular Genetics, 2000, Vol. 9, No. 20 3027
RT–PCR
Protein extraction and western blot analysis
RNA isolated from the cell lines of patients with a LIS1 splice
site mutation or an intragenic LIS1 deletion and normal control
RNA was reverse transcribed using the Advantage RT for PCR
kit (Clontech, Palo Alto, CA). Transcription was undertaken
with both the random hexamer and oligo(dT)18 primers.
cDNA (5 µl) was used for the PCR reaction. cDNA primers
were designed to cover the entire LIS1 coding region. PCR
conditions were as follows: 50 µl reaction containing 1× PCR
buffer, 2.5 mM MgCl2, 0.2 mM of each dNTP, 0.1 µM of each
primer and 1 U of TaqGold polymerase (PE Applied
Biosystems). The reactions were denatured at 95°C for 10 min,
followed by 35 cycles of 94°C for 45 s, 55 or 58°C for 30 s and
72°C for 1 min. For patients LP90-020, LP97-069 and
LP93-005, PCR products were separated on a 4% polyacrylamide gel and the DNA extracted from the aberrant band.
The DNA was then sequenced using DyeTerminator chemistry
(PE Applied Biosystems). The primers used to amplify the
aberrant RT–PCR products are as follows: patient LP85-002,
LIS1-ex1F2 (5′-CGGTGGATGGGAGTGAAGGA-3′) and
LIS1-ex7R (5′-CCAGTTTGCACTTCCCACAT-3′); patient
91-022, LIS1-ex2F (5′-TGTGGAAGACACTTAGTGGCATA-3′) and LIS1-ex7R; and the products were sequenced
directly by DyeTerminator chemistry (PE Applied Biosystems).
Proteins were extracted from lymphoblastoid cell lines of
normal individuals and patients carrying missense and truncation/deletion mutations using a modified lysis buffer (50 mM
Tris–HCl pH 5.4), 100 mM NaCl, 0.4% NP-40, pepstatine
10 µg/ml, 10 mM β-mercaptoethanol, 50 mM NaF, 1 mM
Na3VO4, 30 mM pyrophosphate, 1 mM phenylmethylsulfonyl
fluoride and protease inhibitors (CompleteMini; Boehringher
Mannheim, Indianapolis, IN) by mixing for 5 min at 4°C. The
extracts were centrifuged for 35 000 r.p.m. 160 000 g for
30 min at 4°C in an ultracentrifuge tube and the supernatant
was frozen at –80°C.
Protein expression was assessed by western blot analysis. In
brief, samples were boiled and resolved by SDS–PAGE on
10% gels, which were subsequently transferred to an ECL
nylon membrane (Amersham, Piscataway, NJ) for 1 h 30 min
at 20 V using the genie electrophoretic blotter (Idea Scientific,
Minneapolis, MN) in a buffer of Tris–glycine/20% methanol.
The membrane was blocked for 1 h in 5 or 10% skim milk
powder in Tris–buffered saline with 0.05% Tween 20 (TBST)
at room temperature. The membranes were then incubated only
with primary polyclonal anti-LIS1 (N-19) or monoclonal antiActin (C-2) antibodies or with both simultaneously (Santa
Cruz Biotechnolgy, Santa Cruz, CA) for 1 h. HRP-conjugated
secondary antibodies were incubated with the blot for 45 min
at room temperature. Enhanced chemiluminescence reagents
were used for detection (ECL plus; Amersham). An optical
scanner (Storm 860; Molecular Dynamics) was used to image
the western blot or alternatively X-ray films of western blots
were scanned and densitometry was performed by using
ImageQuant version 1.2 for Macintosh (Molecular Dynamics).
Southern blot analysis
Aliquots of genomic DNA (8 µg) from the patients and normal
controls were digested with the restriction enzymes EcoRI,
BamHI and HindIII and separated on a 1.0% agarose gel. The
LIS1 cDNA clone 47 (20) was digested with EcoRI and the
insert was used as a probe. Southern transfer, radiolabeling of
the cDNA probe, hybridization to Southern blots and washes
were performed as described previously (35). Computer database analysis using NCBI (http://www.ncbi.nlm.nih.gov/blast/ )
against human throughput genome sequences database
(HTGS) indicated that the complete LIS1 gene is contained
within the RPCI-11 BAC clones 135N5 (AC015799) and
74E22 (NT_000744) (unpublished data). BAC 135N5 contains
the promoter region of the LIS1 gene and exons 1 and 2 only.
BAC 74E22 overlaps BAC 135N5 in intron 2 and contains
LIS1 exons 3–11 and the 3′ UTR. The complete genomic
sequence of these two BACs is known. This sequence was
used to generate an EcoRI restriction map. Southern blot
hybridization was performed on EcoRI digests of control
genomic DNA using each of the LIS1 exons as probes and
compared with the same Southern blots hybridized with the
complete LIS1 cDNA (clone 47) (31). Based on these data, we
assigned each LIS1 exon to a specific band of the EcoRI digest.
Therefore, Southern blot hybridization of EcoRI digests of
patient genomic DNA was used to determine the extent of
coding region deletions. The intronic borders of the deletions
are unknown. An optical scanner (Storm 860; Molecular
Dynamics, Sunnyvale, CA) was used to image the blot and the
software ImageQuant (Molecular Dynamics) was used for
densitometric analysis.
Computer and statistical analysis
The region of the coiled-coil domain was predicted with Stripe
(version 2.0b1) (http://www.york.ac.uk/depts/biol/units/coils/
coilcoil.html ) using length 28 residues. Stripe employs
algorithms described by Lupas et al. (36). These findings were
confirmed using COILS version 2.2 (http://www.ch.embnet.org/
software/coils ).
Due to the discreteness of lissencephaly severity grade, for
which only four distinct values occurred among the 29
patients, categorical data methods were used to describe
severity within each LIS1 genotype and to compare severity
between genotypes. Within genotypes, the proportion of
patients at each severity grade was tabulated. For comparisons,
patients were cross-classified by severity grade and mutation
type and Fisher’s exact test was applied to test the hypothesis
of no association between genotype and phenotype. Fisher’s
exact test is a method of performing inference in two-way
contingency tables using exact distributions, when small
sample sizes proscribe the use of χ2 tests. The Mann–Whitney
U-test, an alternative non-parametric approach, produced the
same qualitative results, both with and without adjustment for
tied ranks. Statistical analysis was performed in Stata 6.0
(Stata, College Station, TX).
ACKNOWLEDGEMENTS
We would like to thank the patients with lissencephaly and
their families as well as their physicians who referred them for
3028 Human Molecular Genetics, 2000, Vol. 9, No. 20
study. We gratefully acknowledge the help of Dr Melinda
Drum and Dr Amy Elkund with the statistical analysis and Dr
Francesca Zazzeroni with technical advice. This work was
supported in part by a grant from the National Institutes of
Health to W.E.B. and D.H.L. (PO1 NS39404) and the Lissencephaly Network. C.C. is supported by an INSERM Fellowship and the David and Janice Katz Fellowship in Human
Genetics. R.J.L. is supported by the Australian College of
Pediatrics Nestle Traveling Scholarship.
REFERENCES
1. Dobyns, W.B., Truwit, C.L., Ross, M.E., Matsumoto, N., Pilz, D.T.,
Ledbetter, D.H., Gleeson, J.G., Walsh, C.A. and Barkovich, A.J. (1999)
Differences in the gyral pattern distinguish chromosome 17-linked and Xlinked lissencephaly. Neurology, 53, 270–277.
2. Dobyns, W.B., Reiner, O., Carrozzo, R. and Ledbetter, D.H. (1993) Lissencephaly: a human brain malformation associated with deletion of the
LIS1 gene located at chromosome 17p13. J. Am. Med. Assoc., 23, 2838–
2842.
3. Barkovich, A.J., Guerrini, R., Battaglia, A., Kalifa, G., N’Guyen, T.,
Parmeggiani, A., Santucci, M., Giovanardi-Rossi, P., Granata, T. and
D’Incerti, L. (1994) Band heterotopia: correlation of outcome with magnetic resonance imaging parameters. Ann. Neurol., 36, 609–617.
4. Barkovich, A.J., Koch, T.K. and Carrol, C.L. (1991) The spectrum of lissencephaly: report of ten patients analyzed by magnetic resonance imaging. Ann. Neurol., 30, 139–146.
5. Pilz, D.T., Matsumoto, N., Minnerath, S.R., Mills, P., Gleeson, J.G.,
Allen, K.M., Walsh, C.A., Barkovich, A.J., Dobyns, W.B., Ledbetter,
D.H. and Ross, M.E. (1998) LIS1 and XLIS (DCX) mutations cause most
classical lissencephaly, but different patterns of malformation. Hum. Mol.
Genet., 7, 2029–2037.
6. Lo Nigro, C., Chong, C.S., Smith, A.C., Dobyns, W.B., Carrozzo, R. and
Ledbetter, D.H. (1997) Point mutations and an intragenic deletion in LIS1,
the lissencephaly causative gene in isolated lissencephaly sequence and
Miller–Dieker syndrome. Hum. Mol. Genet., 6, 157–164.
7. Reiner, O., Carrozzo, R., Shen, Y., Wehnert, M., Faustinella, F., Dobyns,
W.B., Caskey, C.T. and Ledbetter, D.H. (1993) Isolation of a Miller–
Dieker lissencephaly gene containing G protein beta-subunit-like repeats.
Nature, 364, 717–721.
8. Peterfy, M., Gyuris, T., Basu, R. and Takacs, L. (1994) Lissencephaly-1 is
one of the most conserved proteins between mouse and human: a single
amino-acid difference in 410 residues. Gene, 150, 415–416.
9. Hattori, M., Adachi, H., Aoki, J., Tsujimoto, M., Arai, H. and Inoue, K.
(1995) Cloning and expression of a cDNA encoding the beta-subunit (30
kDa subunit) of bovine brain platelet-activating factor acetylhydrolase. J.
Biol. Chem., 270, 31345–31352.
10. Liu, Z., Xie, T. and Steward, R. (1999) LIS1, the Drosophila homolog of
a human lissencephaly disease gene, is required for germline cell division
and oocyte differentiation. Development, 126, 4477–4488.
11. Rohan, P.J., Davis, P., Moskaluk, C.A., Kearns, M., Krutzsch, H.,
Siebenlist, U. and Kelly, K. (1993) PAC-1: a mitogen-induced nuclear
protein tyrosine phosphatase. Science, 259, 1763–1766.
12. Morris, N.R., Efimov, V.P. and Xiang, X. (1998) Nuclear migration,
nucleokinesis and lissencephaly. Trends Cell Biol., 8, 467–470.
13. Hattori, M., Adachi, H., Tsujimoto, M., Arai, H. and Inoue, K. (1994)
Miller–Dieker lissencephaly gene encodes a subunit of brain platelet-activating factor acetylhydrolase. Nature, 370, 216–218. [Erratum (1994)
Nature, 370, 391.]
14. Hattori, M., Aoki, J., Arai, H. and Inoue, K. (1996) PAF and PAF acetylhydrolase in the nervous system. J. Lipid Mediators Cell Signal., 14, 99–
102.
15. Albrecht, U., Abu-Issa, R., Ratz, B., Hattori, M., Aoki, J., Arai, H., Inoue,
K. and Eichele, G. (1996) Platelet-activating factor acetylhydrolase
expression and activity suggest a link between neuronal migration and
platelet-activating factor. Dev. Biol., 180, 579–593.
16. Adachi, T., Aoki, J., Manya, H., Asou, H., Arai, H. and Inoue, K. (1997)
PAF analogues capable of inhibiting PAF acetylhydrolase activity suppress migration of isolated rat cerebellar granule cells. Neurosci. Lett.,
235, 133–136.
17. Bix, G.J. and Clark, G.D. (1998) Platelet-activating factor receptor stimulation disrupts neuronal migration in vitro. J. Neurosci., 18, 307–318.
18. Clark, G.D., McNeil, R.S., Bix, G.J. and Swann, J.W. (1995) Platelet-activating factor produces neuronal growth cone collapse. Neuroreport, 6,
2569–2575.
19. Reiner, O., Albrecht, U., Gordon, M., Chianese, K.A., Wong, C., GalGerber, O., Sapir, T., Siracusa, L.D., Buchberg, A.M., Caskey, C.T. et al.
(1995) Lissencephaly gene (LIS1) expression in the CNS suggests a role
in neuronal migration. J. Neurosci., 15, 3730–3738.
20. Hirotsune, S., Fleck, M.W., Gambello, M.J., Bix, G.J., Chen, A., Clark,
G.D., Ledbetter, D.H., McBain, C.J. and Wynshaw-Boris, A. (1998)
Graded reduction of Pafah1b1 (LIS1) activity results in neuronal migration defects and early embryonic lethality. Nature Genet., 19, 333–339.
21. Sapir, T., Eisenstein, M., Burgess, H.A., Horesh, D., Cahana, A., Aoki, J.,
Hattori, M., Arai, H., Inoue, K. and Reiner, O. (1999) Analysis of lissencephaly-causing LIS1 mutations. Eur. J. Biochem., 266, 1011–1020.
22. Sweeney, K.J., Clark, G.D., Prokscha, A., Dobyns, W.B. and Eichele, G.
(2000) Lissencephaly associated mutations suggest a requirement for the
PAFAH1B heterotrimeric complex in brain development. Mech. Dev., 92,
263–271.
23. Sapir, T., Cahana, A., Seger, R., Nekhai, S. and Reiner, O. (1999) LIS1 is
a microtubule-associated phosphoprotein. Eur. J. Biochem., 265, 181–
188.
24. Koch, A., Tonn, J., Albrecht, S., Sorensen, N., Wiestler, O.D. and Pietsch,
T. (1996) Frequent intragenic polymorphism in the 3′ untranslated region
of the lissencephaly gene 1 (LIS1). Clin. Genet., 50, 527–528.
25. Sapir, T., Elbaum, M. and Reiner, O. (1997) Reduction of microtubule
catastrophe events by LIS1, platelet-activating factor acetylhydrolase subunit. EMBO J., 16, 6977–6984.
26. Lien, L.L., Boyce, F.M., Kleyn, P., Brzustowicz, L.M., Menninger, J.,
Ward, D.C., Gilliam, T.C. and Kunkel, L.M. (1991) Mapping of human
microtubule-associated protein 1B in proximity to the spinal muscular
atrophy locus at 5q13. Proc. Natl Acad. Sci. USA, 88, 7873–7876.
27. Harbury, P.B., Zhang, T., Kim, P.S. and Alber, T. (1993) A switch
between two-, three- and four-stranded coiled coils in GCN4 leucine zipper mutants. Science, 262, 1401–1407.
28. Xiang, X., Osmani, A.H., Osmani, S.A., Xin, M. and Morris, N.R. (1995)
NudF, a nuclear migration gene in Aspergillus nidulans, is similar to the
human LIS1 gene required for neuronal migration. Mol. Biol. Cell, 6, 297–
310.
29. Smeal, T., Angel, P., Meek, J. and Karin, M. (1989) Different requirements for formation of Jun:Jun and Jun:Fos complexes. Genes Dev., 3,
2091–2100.
30. Fogli, A., Guerrini, R., Moro, F., Fernandez-Alvarez, E., Livet, M.O.,
Renieri, A., Cioni, M., Pilz, D.T., Veggiotti, P., Rossi, E. et al. (1999)
Intracellular levels of the LIS1 protein correlate with clinical and neuroradiological findings in patients with classical lissencephaly. Ann. Neurol.,
45, 154–161.
31. Chong, S.S., Pack, S.D., Roschke, A.V., Tanigami, A., Carrozzo, R.,
Smith, A.C., Dobyns, W.B. and Ledbetter, D.H. (1997) A revision of the
lissencephaly and Miller–Dieker syndrome critical regions in chromosome 17p13.3. Hum. Mol. Genet., 6, 147–155.
32. Capecchi, M.R., Capecchi, N.E., Hughes, S.H. and Wahl, G.M. (1974)
Selective degradation of abnormal proteins in mammalian tissue culture
cells. Proc. Natl Acad. Sci. USA, 71, 4732–4736.
33. Gleeson, J.G., Lin, P.T., Flanagan, L.A. and Walsh, C.A. (1999) Doublecortin is a microtubule-associated protein and is expressed widely by
migrating neurons. Neuron, 23, 257–271.
34. Fox, J.W., Lamperti, E.D., Eksioglu, Y.Z., Hong, S.E., Feng, Y., Graham,
D.A., Scheffer, I.E., Dobyns, W.B., Hirsch, B.A., Radtke, R.A. et al.
(1998) Mutations in filamin 1 prevent migration of cerebral cortical neurons in human periventricular heterotopia. Neuron, 21, 1315–1325.
35. Kubota, T., Sutcliffe, J.S., Aradhya, S., Gillessen-Kaesbach, G., Christian,
S.L., Horsthemke, B., Beaudet, A.L. and Ledbetter, D.H. (1996) Validation studies of SNRPN methylation as a diagnostic test for Prader–Willi
syndrome. Am. J. Med. Genet., 66, 77–80.
36. Lupas, A., Van Dyke, M. and Stock, J. (1991) Predicting coiled coils from
protein sequences. Science, 252, 1162–1164.
37. Fogli, A., Giglio, S., Arrigo, G., Lo, N.C., Zollo, M., Viggiano, L., Rocchi, M., Archidiacono, N., Zuffardi, O. and Carrozzo, R. (1999) Identification of two paralogous regions mapping to the short and long arms of
human chromosome 2 comprising LIS1 pseudogenes. Cytogenet. Cell
Genet., 86, 225–232.