European Journal of Medical Genetics 53 (2010) 93e99

Contents lists available at ScienceDirect

European Journal of Medical Genetics

journal homepage: http://www.elsevier.com/locate/ejmg

Original article

Interstitial microduplication of Xp22.31: Causative of intellectual disability

or benign copy number variant?
Feng Li a, b, Yiping Shen c, Udo Köhler d, Freddie H. Sharkey e, Deepa Menon f, Laurence Coulleaux g,
Valérie Malan g, Marlène Rio g, Dominic J. McMullan h, H. Cox i, Kerry A. Fagan j, Lorraine Gaunt k,
Kay Metcalfe k, Uwe Heinrich l, Gordon Hislop m, Una Maye n, Maxine Sutcliffe o, Bai-Lin Wu c,
Brian D. Thiel p, Surabhi Mulchandani p, Laura K. Conlin p, Nancy B. Spinner p,
Kathleen M. Murphy q, Denise A.S. Batista a, q, r, *
a
McKusick-Nathans Institute of Genetic Medicine, Johns Hopkins University, Baltimore, MD, USA
b
Kaiser Permanente San Jose Medical Center, San Jose, CA, USA
c
Department of Laboratory Medicine, Children's Hospital Boston, Harvard Medical School, Boston, MA, USA
d
Center of Medical Genetics, MGZ Munich, Germany
e
Microarray Facility, SE Scotland Cytogenetics Department, Western General Hospital, Edinburgh, UK
f
Department of Neurology, Kennedy Krieger Institute, Baltimore, MD, USA
g
Department of Genetics and INSERM U781, Necker Hospital, Université Paris Descartes, Paris, France
h
Molecular Cytogenetics and Microarray Development, West Midlands Regional Genetics Laboratories, Birmingham Womens Hospital NHS Trust, Birmingham, UK
i
West Midlands Clinical Genetics Service, Birmingham Womens Hospital NHS Trust, Birmingham, UK
j
Cytogenetics Hunter Area Pathology Service, Newcastle, NSW, Australia
k
Genetic Medicine, Manchester Academic Health Science Centre, Central Manchester University Hospitals NHS Foundation Trust, Manchester, UK
l
Center for Human Genetics and Laboratory Medicine Dr. Klein and Dr. Rost, Martinsried, Germany
m
Human Genetics, Ninewells Hospital, Dundee, UK
n
Cheshire and Merseyside Regional Genetics Laboratory, Liverpool Women's Hospital, Liverpool, UK
o
Departments of Pathology and Pediatrics, All Children's Hospital/University of South Florida, St Petersburg, FL, USA
p
Clinical Cytogenomics, The Children's Hospital of Philadelphia, Philadelphia, PA, USA
q
Department of Pathology, Johns Hopkins University, Baltimore, MD, USA
r
Cytogenetics Laboratory, Kennedy Krieger Institute, Baltimore, MD, USA

a r t i c l e i n f o a b s t r a c t

Article history: The use of comparative genomic hybridization (CGH) and single nucleotide polymorphism (SNP) arrays
Received 24 September 2009 has dramatically altered the approach to identification of genetic alterations that can explain intellectual
Accepted 23 January 2010 disability and /or congenital anomalies. However, the discovery of numerous copy number changes with
Available online 2 February 2010
benign or unknown clinical significance has made interpretation problematic. Submicroscopic duplica-
tion of Xp22.31 has been reported as either a possible cause of intellectual disability and/or develop-
Keywords:
mental delay or a benign variant. Here we report 29 individuals with the microduplication found as part
Xp22.31
of microarray analysis of 7793 samples submitted to an international group of 13 clinical laboratories.
STS
Microarray The referral reasons varied and included developmental delay, intellectual disability, autism, dysmorphic
Intellectual disability features and/or multiple congenital anomalies. The size of the Xp22.31 duplication varied between
Autism 149 kb and 1.74 Mb and included the steroid sulfatase (STS) gene with the male to female ratio of 0.7.
Behavior problems Duplication within this segment is seen at a frequency of 0.15% in a healthy control population, whereas
Microdeletion a frequency of 0.37% was observed in our cohort of individuals with abnormal phenotypes. We present
CNV a detailed comparison of the breakpoints, inheritance, X-inactivation and clinical phenotype in our
SNP cohort and a review of the literature for a total of 41 patients. To date, this report is the largest
compilation of clinical and array data regarding the microduplication of Xp22.31 and will serve to
broaden the knowledge of regions involving copy number variation (CNV).
Ó 2010 Elsevier Masson SAS. All rights reserved.

* Corresponding author at: Johns Hopkins University, 600 N. Wolfe Street, Park Bldg. SB 202, Cytogenetics Laboratory, Baltimore, MD 21287, USA. Tel.: þ1 410 955 8363;
fax: þ1 410 614 7440.
E-mail address: dbatist1@jhmi.edu (D.A.S. Batista).

1769-7212/$ e see front matter Ó 2010 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmg.2010.01.004
94 F. Li et al. / European Journal of Medical Genetics 53 (2010) 93e99
1. Introduction
Intellectual disability (ID) of variable severity has a population
prevalence of 1e3% with a male to female ratio of 1.4 for moderate
to severe and 1.9 for mild ID [23,31,32]. ID can be classified as
syndromic when it includes other phenotypic abnormalities or
non-syndromic when ID is the only feature. However, in some
cases, both types of ID can be caused by the same genetic abnor-
mality. The high genetic heterogeneity, especially for non-syn-
dromic ID, makes finding the cause(s) a difficult task. The
disproportionate ratio of affected males suggests a significant
contribution of X-linked genes in regulating neurocognitive
development [41]. However, excluding FMR1 of fragile X syndrome,
it was estimated that X-linked mental retardation (XLMR) genes
account for only 10e16% of moderate to severe ID in males and
therefore, cannot fully explain the excess of affected males [32].
Mutations in 76 genes [25] as well as deletions and duplications on
the X chromosome have been described in association with XLMR.
X-Linked ichthyosis (XLI) is due to deficiency of the steroid
sulfatase (STS) gene transcript located at Xp22.31. Deletion of this
gene and adjacent regions is found in 90% of the affected individ-
uals and it occurs due to recombination between low copy repeats
(LCRs) located near the VCX genes that flank STS [13]. Some indi-
viduals with XLI also present with ID [13,18] and in addition, recent
data showed an association between STS deficiency and suscepti-
bility to attention deficit hyperactivity disorder (ADHD), autism and
social communication deficits [19]. The common Xp22.31 deletion
is approximately 1.5 Mb in size and is located from 6.3 to 7.9 Mb on
chromosome X [13]. Deletion of VCX3A has been reported in asso-
ciation with abnormal neurocognitive phenotype [13,15,18].
However, this association was not found by others [11,26,30].
Several authors [4,18,21,27,33,34,38e40,42,43] and investigators
from many clinical laboratories (personal communication) have
observed a submicroscopic duplication of Xp22.31 that involves the STS
gene, the counterpart of the deletion causing XLI. These patients all
presented an abnormal phenotype which could be directly related to
the genomic imbalance or simply reflect bias of ascertainment. Indeed,
some authors interpreted this duplication as a normal variant [4,38],
whereas others classified it as pathogenic [17,33,39,42,43], and still
others were unclear about its clinical significance [27,34]. The fact that
the duplication tended to be inherited from a normal parent in most
cases seems to favor the interpretation of a benign variant. However,
the phenotype of a genomic disorder can be variable due to a number of
genetic mechanisms such as incomplete penetrance, variable expres-
sivity and skewed X-inactivation. A benign variant might also behave
differently in different populations or different genomic backgrounds
and could cause a pathological phenotype under different conditions.
In an attempt to elucidate the significance of the microduplication
Xp22.31, we present data from 13 clinical centers from five countries.
These centers utilized whole genome BAC, oligonucleotide, or SNP
arrays as diagnostic tools for individuals with an abnormal pheno-
type. Here we describe the molecular boundaries, inheritance,
X-inactivation pattern and presenting phenotypes of 29 cases/fami-
lies with the microduplication detected amongst 7793 referred
individuals. We also reviewed literature data and compared it with
our own to compile the largest and most comprehensive evaluation
of the significance of dup Xp22.31 to date.
2. Materials and methods
2.1. Patient samples
A total of 7793 individuals were referred for microarray analysis
with a variety of indications including developmental delay (DD),
ID, autism, dysmorphic features and multiple congenital anomalies.
These individuals were examined in 13 clinical centers from five
countries and microarray analysis was performed in each center's
respective cytogenetic/molecular laboratory. The laboratories and
countries included in this report were: one in Australia, one in
France, two in Germany, five in the United Kingdom, and four in the
United States. Clinical evaluation was performed by clinicians from
the referring institutions.
2.2. Chromosome, FISH and microarray analysis
Chromosome analysis was performed in peripheral blood
lymphocyte cultures according to established protocols. FISH
testing utilized BAC probes mapped within the Xp22.31 region
(BlueGnome, Cambridge, UK) or the STS probe (Abbott Molecular/
Vysis, Des Plaines, IL, USA). Both G-banded karyotype and array
analyses were performed for most of the patients. Fluorescence in
situ hybridization (FISH) was utilized to confirm the duplication in
the probands and to evaluate parental inheritance.
Microarray was performed using one of four different platforms:
BAC based high resolution array composed of 3600 targets (Cyto-
Chip v1.1) or 4200 targets (CytoChip v2.0 or 2.0.1) (BlueGnome,
UK); oligonucleotide array comprised of 244,000 probes (Agilent,
USA); SNP array with 250,000 targets (Affymetrix, USA); and SNP
array with 610,000 targets (Illumina, USA). Genomic DNA labeling
and hybridization were performed following each of the manu-
facturers’ instructions. Data were analyzed with different software
depending on the platform utilized: BlueFuse for the CytoChip
arrays, CGH analytics for the Agilent oligonucleotide, Affymetrix
Genotyping Console v2.1-for the Affymetrix SNP array and Bead-
Studio for the Illumina SNP array. Each laboratory utilized its own
standards for interpretation of CNVs, but in general, an observed
change was interpreted as non-pathogenic if it had been previously
well documented in databases cataloging benign CNVs, such as the
Database of Genomic Variants, and/or laboratory internal
databases.
2.3. X-Inactivation analysis
An assay to determine X chromosome inactivation was per-
formed in individuals with the microduplication when DNA
samples were available and included: five female patients, eight
carrier mothers and one family including sister, mother and
grandmother. All DNA samples tested were derived from peripheral
blood cells. Samples were examined by testing the androgen-receptor
gene for assessment of the methylation status (modification of the
HUMARA assay) [2]. 250 ng genomic DNA was double-digested
overnight with 20 units of HhaI and 20 units of HpaII (New England
BioLabs, USA). Mock digestion with no enzyme was set up for each
sample. Restriction enzymes were subsequently inactivated by
heating the samples at 65  C for 20 min. 50 ng digested or mock-
digested DNA was then amplified with PCR primers as previously
described [2]. PCR products were separated and detected by
capillary gel electrophoresis using the 3100 Genetic Analyzer (ABI
PRISM, USA) and analyzed using Genescan 3.7. The X-inactivation
ratio in heterozygotes was calculated as previously described [37].
A sample was considered to have skewed X-inactivation if the same
X chromosome was inactivated in at least 80% of the cells. Samples
with a single homozygous peak were considered not informative.
3. Results
Among 7793 patients referred for microarray testing, 3202 were
analyzed by BAC array, 2600 by oligonucleotide array, and 1991 by
 F. Li et al. / European Journal of Medical Genetics 53 (2010) 93e99
SNP array. A submicroscopic duplication in the Xp22.31 region
involving the STS gene was identified in 29 probands (15 by BAC
array, seven by oligonucleotide and seven by SNP array). FISH
showed that the duplication Xp22.31was in tandem and not
inserted into another chromosome. The patients’ age ranged from
newborn to 18 years old at the time when the tests were per-
formed. The ratio of male to female was 0.7 (12 males/17 females).
In addition, we also detected three individuals with micro-
duplications within Xp22.31 but no STS involvement. These dupli-
cations were smaller in size (457, 549 and 737 kb), included VCX
and PNPLA4 genes only, were located more proximal to the
centromere and were excluded from our analysis. Literature review
yielded 12 individuals with duplication of Xp22.31 described in
sufficient detail (six males, four females, two individuals with
gender not reported) [17,27,33,38e40,42,43]. Although the molec-
ular boundaries were not described in detail in each report, most
cases had breakpoints similar to those we identified in our study
(Fig. 1). We summarized our data with the literature for a total of 41
probands. The duplicated region in the combined sample ranged
from 149 kb to 1.9 Mb and involved the STS gene in all probands.
The smallest duplication (149 kb) was located from 7.18 Mb to
7.32 Mb involving only the STS gene. The largest duplicated region
contained six OMIM genes, including VCX3A, HDHD1A, STS, VCX,
PNPLA4, and VCX2. In most probands (33/40) the duplication
involved HDHD1A, STS, VCX and PNPLA4.
The duplicated X chromosome was inherited in 26/27 probands
for whom both parents were available, including 20 probands from
our study and seven from the literature (14 females, 11 males and
two gender not specified) [17,27,38e40,43]. The males obviously
had maternal inheritance; for the females eight were maternal and
six paternal. In one of the males from our sample, the duplication
was also present in the phenotypically normal maternal grandfa-
ther (Family 1, Fig. 2). The last two individuals (gender not speci-
fied) one had a de novo duplication and in the other it was maternal
[18]. Two of the mothers carrying the duplication (2/20) had
abnormal phenotypes, one had autism (Family 2) and the other had
DD, with psychiatric problems and seizures (Family 3). The six
transmitting fathers were not known to have any major abnormal
phenotypic features but detailed clinical information, specifically
acquisition of milestones, was not available.
X-Inactivation studies were performed in 16 females (15 from
our group, one from the literature) [43] who carry the duplication:
five female patients, nine carrier mothers, including the two
X pter 6 Mb
VCX3A HDHD1A
Fig. 1. Comparison of breakpoints of 29 patients in our study and 11 published cases with available data. On the top, the black bar represents the genomic location in Xp22.31 from
distal (X pter) to proximal (X cen) of X chromosome. The black boxes represent OMIM genes with the gene names indicated underneath. Under the genomic location bar, the
published cases and our patients are shown with gray boxes representing the minimum size of the duplication in each patient according to the molecular boundaries detected by
array. Each patient's gender, when known, is indicated in parentheses. NA, not available.
95
mothers with an abnormal phenotype and in one family including
sister, and maternal grandmother (Table 1, Fig. 2). Using the
skewing criteria defined above as 80:20 (no ratio was provided for
the patient described in the literature [43], only that it was
skewed), three female patients (3/5) had random inactivation and
in two (2/5) inactivation was skewed: in one individual the dupli-
cated X was preferentially inactive (dup Xi) and in one it was
preferentially active (dup Xa). Of the nine mothers tested (Fig. 2),
one was not informative, five had random and three had skewed X-
inactivation. Of the mothers with an abnormal phenotype, one had
random and one had dup Xa. Two phenotypically normal mothers
had dup Xi. The sister and maternal grandmother from the three-
generation family studied, both had skewed X-inactivation with
dupXi. A seemingly unaffected mother was found to have four
copies of the STS gene, resulting from two independent tandem
duplications on both of her X chromosomes (as confirmed by FISH,
data not shown). In this individual regardless of skewing of the X-
inactivation (assay not performed) there are two active copies of
the STS gene.
There were three families with more than one individual with
abnormal phenotype in concordance with presence of dup Xp22.31.
In Family 2 the mother of the affected boy had autistic features. In
Family 3 there were two brothers carrying the duplication affected
with ID, dysmorphism and global DD; one of the brothers was more
severely affected and also had Klinefelter syndrome (47,XXY), but
only one of his X chromosomes had the Xp22.31 duplication. The
mother in Family 3 had DD, with psychiatric problems and seizures.
In Family 5, three brothers had the same duplication and each one
had autistic features varying from moderate to severe. Discordance
of an abnormal phenotype and dup Xp22.31 was noted in two
families from the literature [27,38]. The probands in both discor-
dant families described in the literature were females with simi-
larly affected brothers who did not have the dup Xp22.31. In these
cases, the duplication does not seem to be causative of the
phenotype in the females. Of note, in one of these families the
daughter and her carrier mother both had duplications on Xp22.31
involving STS but with different breakpoints resulting in only
partial overlap of the duplicated genomic region [27].
Clinical presentation had a broad spectrum (Table 2) with most
dysmorphic features present only in a few individuals. ID, varying
from severe to learning disabilities, was the most common finding
with a prevalence of 69% (24/35). Behavioral problems were also
common and present in 37% of probands (13/35), mostly males and
7 Mb 8 Mb X cen
STS VCX PNPLA4 VCX2
Mencarelli et al. 2008 (female)
Horn et al., 2007 (2 NA)
Shur et al. 2008 (male)
Wagenstaller et al. 2007 (male)
Scharer et al.,2008 (2 males;2 females)
Tyson et al., 2005 (male)
Shaw-Smith et al. 2004 (female)
cases 1-13 (5 males; 8 females)
case 14 (female)
case 15 (male)
cases 16-19 (2 males; 2 females)
cases 20-24 (3 males; 2 females)
case 25 (male)
cases 26-27 (2 females)
case 28 (female)
case 29 (female)
96 F. Li et al. / European Journal of Medical Genetics 53 (2010) 93e99

FAMILY 1 FAMILY 2

dup Xa

random

FAMILY 4

FAMILY 3

dup Xi

random

dup Xi

47,XXY

dup Xi
CVS

FAMILY 5

random

Fig. 2. Pedigrees from families cited in the text showing segregation of the dup Xp22.31, phenotype and X-inactivation pattern. Dup Xi, duplicated X chromosome preferentially
inactive; dup Xa, duplicated X preferentially active; random, random X chromosome inactivation; CVS, chorionic villi sampling. Additional chromosome abnormality is shown next
to pedigree symbols. Blank and shaded symbols, normal and abnormal phenotypes respectively; black dot, carrier of dup Xp22.31; diagonal line, normal phenotype and not carrier
of dup Xp22.31; small dots, phenotype not known.

included autism (9/13), auto aggression (1/13), maladaptive These frequencies are probably underestimated due to the paucity
behavior (1/13), ADHD (1/13) and behavior problems non-specified of clinical information available. Three of our female patients had
(1/13). Other findings included: hypotonia (7/35; 20%), seizures additional chromosome abnormalities and their phenotypic
(4/35; 11%), and failure to thrive/feeding difficulties (4/35; 11%). features were not included in the clinical analysis. These additional
abnormalities were a terminal deletion of the distal short arm of
Table 1 chromosome 4 at band p16; interstitial deletion of the distal long
X-Inactivation studies in 16 females with the duplication Xp22.31, including 15 from arm of chromosome 7 at band q35; and a de novo interstitial
our report and one from Wagenstaller et al. [23]: seven females with an abnormal deletion of the long arm of chromosome 17 at band q21.31. These
phenotype (five probands and two mothers) and nine females with a normal
structural unbalanced rearrangements are in areas known to cause
phenotype (seven mothers, one sister and one maternal grandmother). In our
patients, dup X active is defined as preferential activation of this homolog in 80% or an abnormal phenotype.
more of the cells assayed; dup X inactive is defined as preferential inactivation in
80% or more of the cells. 4. Discussion
Dup X Random Dup X Not Total
active inactive informative With the development of microarray CGH technology and its
Abnormal 2 4 1 0 7 broad application in research and clinical diagnosis, numerous
phenotype novel pathogenic copy number variants (CNV) have been discov-
Normal phenotype 0 4 4 1 9 ered. Seemingly benign CNVs have been discovered as well.
Total 2 8 5 1 16
Among the benign CNVs listed in the Database of Genomic
 F. Li et al. / European Journal of Medical Genetics 53 (2010) 93e99 97

Table 2
Phenotypic features in 35 probands with duplication Xp22.31. Clinical information was available for 23/26 probands from our sample and from 12 previously described in the
literature [14,16,17,19-23]. Only features present in at least two probands were included. M, male; F, female; NA, gender not specified.

23 patients from our 12 patients from literature Total 35 patients (15 M/14 F/6 NA) (%)
sample (11 M/12 F) (4 M/2 F/6 NA)

M F M F NA
Short stature 0 1 1 0 0 2 (6)
Microcephaly 1 1 1 0 1 4 (11)
Frontal bossing 1 1 1 0 0 3 (9)
Low set/posteriorly rotated 0 0 1 1 0 2 (6)
ears
Epicanthal folds 2 0 0 0 1 3 (9)
Ptosis 1 0 1 1 0 3 (9)
Short nose 1 0 1 0 0 2 (6)
Anteverted nares 2 0 1 0 0 3 (9)
Long philtrum 1 0 0 0 1 2 (6)
Small teeth 1 0 1 0 0 2 (6)
Cardiac problems 1 1 0 0 0 2 (6)
GE reflux/esophageal atresia 2 0 1 0 0 3 (9)
Failure to thrive/feeding 2 2 0 0 0 4 (11)
difficulties
Intellectual disability 6 7 6 1 4 24 (69)
Hypotonia 2 0 2 0 3 7 (20)
Seizures 1 2 0 1 0 4 (11)
Behavior abnormalities 6 0 2 1 4 13 (37)

Variants (http://projects.tcag.ca/variation), w50% are smaller than
100 kb, w95% are smaller than 500 kb, and only about 5% repre-
sent large scale variation [4]. Almost half of these large scale CNVs
were found at pericentrometric or subtelomeric regions while the
other half were interstitial [4]. Some of these large scale CNVs are
difficult to categorize as pathogenic or benign despite careful
analysis. For example, deletions or duplications at 16p13.11 and
16p11.2 were recently described in both phenotypically normal
and abnormal individuals, sometimes within the same family,
including patients with epilepsy and autism [4,16,44]. Similarly
a CNV at 1q21.1 was observed in several individuals with throm-
bocytopenia with absent radius (TAR) syndrome and the same CNV
was present in 32% of unaffected family members [20]. This led to
the hypotheses of a susceptibility locus in the 1q21.1 region that
was necessary but not sufficient to cause the TAR phenotype.
Another example is the recently described microdeletion at
15q13.3 which presents with a wide phenotypic spectrum varying
from normal to ID, autism and neuropsychiatric disorder [5,6,29].
All these CNVs were present at a lower frequency in controls,
compared to affected individuals, which led the authors to propose
a causative role, perhaps a predisposition, for such changes.
Therefore, careful studies of patients vs. normal control population
are needed to elucidate the role of particular CNVs in genetic
disorders.
Determining the significance of a CNV is a challenge to the
geneticist [22], especially when it is not well characterized. A CNV
in an interstitial region like Xp22.31 of a large size and involving
multiple genes is difficult to interpret. In a large database of 2026
disease-free individuals [36] (http://cnv.chop.edu), we found three
cases with a duplication spanning the STS gene, two of these were
about 1.6 Mb in size and the third case involved only the STS gene,
for a prevalence of 0.15% (3/2026). In our sample of affected
probands the prevalence was 0.37% (29/7793). This 2.5-fold
increase in prevalence compared to a healthy population does not
achieve statistical significance (Fisher's exact test p ¼ 0.1295), but it
might suggest a phenotypic burden or risk factor. However, the
control and our sample data were not analyzed by the same
method. Our samples were analyzed on an individual basis whereas
control data is usually generated only by calling algorithms and
might miss some duplications and deletions. It is perfectly possible
that the control data reflects actual prevalence but further
comparison might be needed. Moreover, since the duplication is on
the X chromosome, it would be ideal to compare its frequency in
relation to gender but the published control population we utilized
does not provide this data.
In a report summarizing 8789 clinical cases referred for array
CGH, 22 individuals (0.25%) were identified with duplication within
Xp22.3 including the STS gene [34]. An earlier report from the same
group including 1500 cases with a variety of clinical indications
showed five individuals with this duplication (0.3%) [35]. All five
cases had developmental delay, two had dysmorphic features, one
had cleft lip, and another had seizures and autism. Other than
indicating that the STS gene is within the duplicated region, these
reports did not provide the size, precise location or parental
inheritance and therefore, were not included in our analysis. In
these samples from Shaffer and colleagues [34,35], the prevalence
of dup Xp22.31 is also somewhat increased over the general
population.
LCRs located at the distal ends of VCX3A and VCX2 seem to
predispose this region to non-allelic homologous recombination
[13] leading to the deletions seen in XLI and to the counterpart
duplications. STS deficiency has been associated with ID [13,18],
ADHD, autism and social communication deficits [19]. Association
of SNP markers within the STS gene have also been described in
individuals with isolated ADHD [7]. The enzyme steroid sulfatase
converts sulfated dehydroepiandrosterone (DHEA-S) to DHEA. Both
are neurosteroids with effects on several neurophysiological and
behavioral processes and there is converging evidence of a role for
STS in attentional functioning [12]. Deletion of VCX3A, a member of
VCX/Y gene family, has also been implicated as causing ID in some
individuals [15]. However, a causative role for VCX3A deletion was
not found by others. In a report of 80 individuals with XLI and
normal intelligence, the authors did not find a correlation between
deletion of VCX3A and intellectual ability [11]. The other four genes
in the duplicated area, HDHD1A, VCX, PNPLA4, and VCX2, are not
well studied yet. It is not difficult to infer that haploinsufficiency for
genes on the X chromosome is pathogenic and could affect males
more than females, but whether the duplication of X-linked genes
might have similar critical roles is not clear. A recent study using an
X chromosome-specific tiling array in individuals with ID did
provide evidence suggesting that increased gene dosage of X-linked
genes may have an impact in the deregulation of normal cognitive
98 F. Li et al. / European Journal of Medical Genetics 53 (2010) 93e99
development [14]. STS could be a candidate gene contributing to
the abnormal phenotype in dup Xp22.31. Three patients from our
sample seem to have a breakpoint within the STS gene (Fig. 1): cases
25 (male patient) and 29 (female) were analyzed by BAC array and
case 28 (female) was analyzed by oligonucleotide array. BAC array
does not yield precise definition of breakpoints, however for these
three cases the breaks were within or at least close enough to STS
that might lead to loss of function rather than a duplication effect.
Break within STS was not seen in the 2026 individuals from the
control sample described by Shaikh et al. [36].
X-Inactivation may play an important role in determining
phenotype. All eight phenotypically normal informative females
tested had either random or preferential inactivation of the dupli-
cated X (dupXi). In contrast, amongst the females with an abnormal
phenotype, the duplicated X was preferentially active in 2/7
patients (dup Xa) and in one it was preferentially inactive (dupXi).
In a large sample of unaffected females, 8% had skewed X-inacti-
vation with a ratio of 80:20 (14.2% when newborns were excluded)
[3]. Although our numbers are small, when we consider the
affected females in our sample, 29% (2/7) showed skewed X-inac-
tivation with dup Xa. Overall, the frequency of skewed X-inactiva-
tion in our sample (7/15) is increased over the general population
(47% versus 14.2%). Although more data is needed, our numbers
might suggest a positive correlation between dup Xa with
abnormal phenotype and random/dup Xi with normal phenotype.
The exception would be the carrier mother with the duplication on
both of her X chromosomes and therefore, regardless of inactiva-
tion pattern (analysis not performed), would have a duplicated X
active (dup Xa). Our X-inactivation assay was performed using DNA
isolated from peripheral blood cells and we cannot be certain that it
reflects other tissues. Although the literature suggests that there is
significant correlation of the X-inactivation ratios between two
tissues analyzed (peripheral blood and urinary epithelia), some
individuals can have wide variation, and therefore, extrapolation
cannot always be accurate [8]. In addition, it has been shown that
about 21% of genes on the X chromosome normally escape X-
inactivation [8,9] and a large cluster of these genes resides on the
short arm. Specifically STS, HDHD1A and PNPLA4 genes were shown
to escape X-inactivation although with less than 100% efficiency
[10] with STS activity varying from 1.3 to 1.7 in normal females [28].
This partial or incomplete expression from the inactive X compli-
cates the genotypeephenotype correlation for the female probands
and carrier mothers and might explain some of the variability we
found.
We observed three families with recurrence of ID in concor-
dance with the duplication Xp22.31 including a family with three
affected brothers with diagnosis of autism (Fig. 2). Although it is
possible that these three brothers share the same X chromosome by
chance alone, this likelihood is 12.5%.
The most frequent clinical features observed were ID, behavior
abnormalities, hypotonia, failure to thrive/feeding difficulties and
seizures. Of particular interest is the high incidence of behavioral
problems (34%) with autistic features as the major contributor.
Based on the proposed role of STS in neurophysiological and
behavioral processes [12] it is possible that increased STS activity, as
already noted for its deficiency, might have an impact in behavior.
As a whole, the clinical features we observed do not reveal
a recognizable syndromic phenotype. Many genomic disorders
have already been described with a high level of phenotypic
heterogeneity. These disorders have been identified largely after
the adoption of microarray analysis and were first defined by the
genomic rearrangement with subsequent phenotypic description
an approach known as genotype first. To explain how a particular
CNV, such as the dup Xp22.31, can occur in both affected and
healthy populations albeit with different prevalence, it was
proposed that some individuals can overcome an initial impair-
ment. This was observed for example in the microdeletion 3q29
and microduplication 7q11.23 syndromes, in which some individ-
uals can become normal functioning adults after initially present-
ing mild developmental delay in infancy [1,24]. Another factor is
the observation that in several genomic abnormalities, such as
3q29, 7q11.23 and 22q11.2, duplications seem to have lower
penetrance than the counterpart deletions.
As a group our observations suggest a possible causal relation-
ship of the dup Xp22.31 and abnormal phenotype, possibly an
additive effect. However, we cannot completely exclude the
possibility that this duplication is a rare population variant. In
conclusion, a clear clinical phenotype for the dup Xp22.31 does not
emerge from this detailed analysis of our large data sample.
However, we caution that there is still enough evidence to consider
this duplication as a risk factor or modifier for ID and behavior
problems. Further data with other large cohorts will contribute to
understanding the role of this duplication.
5. Web resources
http://cnv.chop.edu/
http://projects.tcag.ca/variation
Acknowledgments
The authors wish to acknowledge Dr Constance Griffin for the
critical review of this manuscript, Dr Gail Stetten for providing
control DNA to most of the X-inactivation assays. We thank the staff
of the Cytogenetics Laboratory at the Kennedy Krieger Institute for
their technical expertise: Janet Biscoe, Maria Blazina, Susan Morsey
and Elizabeth Wohler. We also thank the patients and their families
for their cooperation.
References
[1] N. van der Aa, L. Rooms, G. Vandeweyer, J. van den Ende, E. Reyniers,
M. Fichera, C. Romano, B. Delle Chiaie, G. Mortier, B. Menten, A. Destrée,
I. Maystadt, K. Männik, A. Kurg, T. Reimand, D. McMullan, C. Oley, L. Brueton,
E.M.H.F. Bongers, B.W.M. van Bon, R. Pfund, S. Jacquemont, A. Ferrarini,
D. Martinet, C. Schrander-Stumpel, A.P.A. Stegmann, S.G.M. Frints, B.B.A.
de Vries, B. Ceulemans, R.F. Kooy, Fourteen new cases contribute to the
characterization of the 7q11.23 microduplication syndrome. Eur. J. Med. Genet.
52 (2009) 94e100.
[2] R.C. Allen, H.Y. Zoghbi, A.B. Moseley, H.M. Rosenblatt, J.W. Belmont, Methyl-
ation of HpaII and HhaI sites near the polymorphic CAG repeat in the human
androgen-receptor gene correlates with X chromosome inactivation. Am. J.
Hum. Genet. 51 (1992) 1229e1239.
[3] J.M. Amos-Landgraf, A. Cottle, R.M. Plenge, M. Friez, C.E. Schwartz,
J. Longshore, H.F. Willard, X chromosome-inactivation patterns of 1,005
phenotypically unaffected females. Am. J. Hum. Genet. 79 (2006) 493e499.
[4] E.L. Baldwin, J.Y. Lee, D.M. Blake, B.P. Bunke, C.R. Alexander, A.L. Kogan,
D.H. Ledbetter, C.L. Martin, Enhanced detection of clinically relevant genomic
imbalances using a targeted plus whole genome oligonucleotide microarray.
Genet. Med. 10 (2008) 415e429.
[5] S. Ben-Shachar, B. Lanpher, J.R. German, M. Qasaymeh, L. Potocki, S.C. Sreenath
Nagamani, L.M. Franco, A. Malphrus, G.W. Bottenfield, J.E. Spence, S. Amato,
J.A. Rousseau, B. Moghaddam, C. Skinner, S.A. Skinner, S. Bernes, N. Armstrong,
M. Shinawi, P. Stankiewicz, A. Patel, S.-W. Cheung, J.R. Lupski, A.L. Beaudet,
T. Sahoo, Microdeletion 15q13.3: a locus with incomplete penetrance for
autism, mental retardation, and psychiatric disorders. J. Med. Genet. 46 (2009)
382e388.
[6] B.W. van Bon, H.C. Mefford, B. Menten, D.A. Koolen, A.J. Sharp, W.M. Nillesen,
J.W. Innis, T.J. de Ravel, C.L. Mercer, M. Fichera, H. Stewart, L.E. Connell,
K. Ounap, K. Lachlan, B. Castle, N. van der Aa, C. van Ravenswaaij,
M.A. Nobrega, C. Serra-Juhé, I. Simonic, N. de Leeuw, R. Pfundt, E.M. Bongers,
C. Baker, P. Finnemore, S. Huang, V.K. Maloney, J.A. Crolla, M. van Kalmthout,
M. Elia, G. Vandeweyer, J.P. Fryns, S. Janssens, N. Foulds, S. Reitano, K. Smith,
S. Parkel, B. Loeys, C.G. Woods, A. Oostra, F. Speleman, A.C. Pereira, A. Kurg,
L. Willatt, S.J. Knight, J.R. Vermeesch, C. Romano, J.C. Barber, G. Mortier,
L.A. Perez-Jurado, F. Kooy, H.G. Brunner, E.E. Eichler, T. Kleefstra, B.B. de Vries,
Further delineation of the 15q13 microdeletion and duplication syndromes:
a clinical spectrum varying from non-pathogenic to a severe outcome. J. Med.
Genet. 46 (2009) 511e523.
 F. Li et al. / European Journal of Medical Genetics 53 (2010) 93e99
[7] K.J. Brookes, Z. Hawi, A. Kirley, E. Barry, M. Gill, L. Kent, Association of the
steroid sulfatase (STS) gene with attention deficit hyperactivity disorder. Am. J.
Med. Genet. B Neuropsychiatr. Genet. 147B (2008) 1531e1535.
[8] C.J. Brown, J.M. Greally, A stain upon the silence: genes escaping X inactiva-
tion. Trends Genet. 19 (2003) 432e438.
[9] L. Carrel, A.A. Cottle, K.C. Goglin, H.F. Willard, A first-generation X-inactivation
profile of the human X chromosome. Proc. Natl. Acad. Sci. USA 96 (1999)
14440e14444.
[10] L. Carrel, H. Willard, X-inactivation profile reveals extensive variability in X-
linked gene expression in females. Nature 434 (2005) 400e404.
[11] S.A. Cuevas-Covarrubias, L.M. Gonzalez-Huerta, Analysis of the VCX3A, VCX2
and VCX3B genes shows that VCX3A gene deletion is not sufficient to result in
mental retardation in X-linked ichthyosis. Br. J. Dermatol. 158 (2008) 483e486.
[12] W. Davies, T. Humby, W. Kong, T. Otter, P.S. Burgoyne, L.S. Wilkinson,
Converging pharmacological and genetic evidence indicates a role for steroid
sulfatase in attention. Biol. Psychiatry 66 (2009) 360e367.
[13] H. van Esch, K. Hollanders, L. Badisco, C. Melotte, P. van Hummelen,
J.R. Vermeesch, K. Devriendt, J.P. Fryns, P. Marynen, G. Froyen, Deletion of VCX-A
due to NAHR plays a major role in the occurrence of mental retardation in
patients with X-linked ichthyosis. Hum. Mol. Genet. 14 (2005) 1795e1803.
[14] G. Froyen, H. van Esch, M. Bauters, K. Hollanders, S.G. Frints, J.R. Vermeesch,
K. Devriendt, J.P. Fryns, P. Marynen, Detection of genomic copy number
changes in patients with idiopathic mental retardation by high-resolution X-
array-CGH: important role for increased gene dosage of XLMR genes. Hum.
Mutat. 28 (2007) 1034e1042.
[15] M. Fukami, S. Kirsch, S. Schiller, A. Richter, V. Benes, B. Franco, K. Muroya,
E. Rao, S. Merker, B. Niesler, A. Ballabio, W. Ansorge, T. Ogata, G.A. Rappold, A
member of a gene family on Xp22.3, VCX-A, is deleted in patients with X-
linked nonspecific mental retardation. Am. J. Hum. Genet. 67 (2000) 563e573.
[16] F.D. Hannes, A.J. Sharp, H.C. Mefford, T. de Ravel, C.A. Ruivenkamp,
M.H. Breuning, J.P. Fryns, K. Devriendt, G. van Buggenhout, A. Vogels,
H. Stewart, R.C. Hennekam, G.M. Cooper, R. Regan, S.J. Knight, E.E. Eichler,
J.R. Vermeesch, Recurrent reciprocal deletions and duplications of 16p13.11:
the deletion is a risk factor for MR/MCA while the duplication may be a rare
benign variant. J. Med. Genet. 46 (2009) 223e232.
[17] D. Horn, S. Spranger, G. Kruger, J. Wagenstaller, B. Weschke, H.-H. Ropers,
S. Mundlos, R. Ulmann, T.M. Strom, E. Klopocki, Microdeletions and micro-
duplications affecting the STS gene at Xp22.31 are associated with a distinct
phenotypic spectrum. Med. Genet. 19 (2007) P62.
[18] N. Hosomi, N. Oiso, K. Fukai, K. Hanada, H. Fujita, M. Ishii, Deletion of distal
promoter of VCXA in a patient with X-linked ichthyosis associated with
borderline mental retardation. J. Dermatol. Sci. 45 (2007) 31e36.
[19] L. Kent, J. Emerton, V. Bhadravathi, E. Weisblatt, G. Pasco, L.R. Willatt,
R. McMahon, J.R. Yates, X-linked ichthyosis (steroid sulfatase deficiency) is
associated with increased risk of attention deficit hyperactivity disorder,
autism and social communication deficits. J. Med. Genet. 45 (2008) 519e524.
[20] E. Klopocki, H. Schulze, G. Strauss, C.E. Ott, J. Hall, F. Trotier, S. Fleischhauer,
L. Greenhalgh, R.A. Newbury-Ecob, L.M. Neumann, R. Habenicht, R. König,
E. Seemanova, A. Megarbane, H.H. Ropers, R. Ullmann, D. Horn, S. Mundlos,
Complex inheritance pattern resembling autosomal recessive inheritance
involving a microdeletion in thrombocytopenia-absent radius syndrome. Am.
J. Hum. Genet. 80 (2007) 232e240.
[21] A. Kurg, K. Mannik, O. Zilina, H. Puusepp, S. Parkel, P. Palta, A. Veidenberg,
K. Unap, Detection of genomic copy number changes in Estonian patients with
mental retardation by SNP genotyping microarrays. Am. S. Hum. Genet. 58
(2008) P232.
[22] C. Lee, A.J. Iafrate, A.R. Brothman, Copy number variations and clinical cyto-
genetic diagnosis of constitutional disorders. Nat. Genet. 39 (2007) S48eS54.
[23] H. Leonard, X. Wen, The epidemiology of mental retardation: challenges and
opportunities in the new millennium, Ment. Retard. Dev. Disabil. Res. Rev. 8
(2002) 117e134.
[24] F. Li, E.C. Lisi, E.S. Wohler, A. Hamosh, D.A. Batista, 3q29 interstitial micro-
deletion syndrome: an inherited case associated with cardiac defect and
normal cognition. Eur. J. Med. Genet. 52 (2009) 349e352.
[25] M.Z. Lisik, A.L. Sieron, X-linked mental retardation. Med. Sci. Monit. 14 (2008)
RA221eRA229.
[26] M. Macarov, M. Zeigler, J.P. Newman, D. Strich, V. Sury, A. Tennenbaum,
V. Meiner, Deletions of VCX-A and NLGN4: a variable phenotype including
normal intellect. J. Intellect. Disabil. Res. 51 (2007) 329e333.
[27] M.A. Mencarelli, E. Katzaki, F.T. Papa, K. Sampieri, R. Caselli, V. Uliana,
M. Pollazzon, R. Canitano, R. Mostardini, S. Grosso, I. Longo, F. Ariani, I. Meloni,
J. Hayek, P. Balestri, F. Mari, A. Renieri, Private inherited microdeletion/
microduplications: implications in clinical practice. Eur. J. Med. Genet. 51
(2008) 409e416.
99
[28] B.R. Migeon, L.J. Shapiro, R.A. Norum, T. Mohandas, J. Axelman, R.L. Dabora,
Differential expression of steroid sulphatase locus on active and inactive
human X chromosome. Nature 299 (1982) 838e840.
[29] D.T. Miller, Y. Shen, L.A. Weiss, J. Korn, I. Anselm, C. Bridgemohan, G.F. Cox,
H. Dickinson, J. Gentile, D.J. Harris, V. Hegde, R. Hundley, O. Khwaja, S. Kothare,
C. Luedke, R. Nasir, A. Poduri, K. Prasad, P. Raffalli, A. Reinhard, S.E. Smith,
M.M. Sobeih, J.S. Soul, J. Stoler, M. Takeoka, W.H. Tan, J. Thakuria, R. Wolff,
R. Yusupov, J.F. Gusella, M.J. Daly, B.L. Wu, Microdeletion/duplication at
15q13.2q13.3 among individuals with features of autism and other neuro-
psychiatric disorders. J. Med. Genet. 46 (2009) 242e248.
[30] F. Mochel, C. Missirian, R. Reynaud, A. Moncla, Normal intelligence and social
interactions in a male patient despite the deletion of NLGN4X and the VCX
genes. Eur. J. Med. Genet. 51 (2008) 68e73.
[31] N. Roeleveld, G.A. Zielhuis, F. Gabreels, The prevalence of mental retardation:
a critical review of recent literature, Dev. Med. Child. Neurol. 39 (1997)
125e132.
[32] H.H. Ropers, B.C. Hamel, X-linked mental retardation. Nat. Rev. 6 (2005)
46e57.
[33] G. Scharer, D. Manchester, G. Bellus, L. Pickler, M. Saenz, M. Raymond,
K. McKelvie, M. Stillberger, J. March, B. Lund, M. Springer, K. Swisshelm, Copy
number variations (CNVs) in well-defined region of Xp22.31 are associated
with syndromic phenotypes and intellectual disability. Further delineation of
a new microdeletion/duplication syndrome. Am. S. Hum. Genet. 58 (2008)
P136.
[34] L.G. Shaffer, B.A. Bejjani, B. Torchia, S. Kirkpatrick, J. Coppinger, B.C. Ballif, The
identification of microdeletion syndromes and other chromosome abnor-
malities: cytogenetic methods of the past, new technologies for the future,
Am. J. Med. Genet. C Semin. Med. Genet. 145C (2007) 335e345.
[35] L.G. Shaffer, C.D. Kashork, R. Saleki, E. Rorem, K. Sundin, B.C. Ballif,
B.A. Bejjani, Targeted genomic microarray analysis for identification of
chromosome abnormalities in 1500 consecutive clinical cases. J. Pediatr. 149
(2006) 98e102.
[36] T.H. Shaikh, X. Gai, J.C. Perin, J.T. Glessner, H. Xie, K. Murphy, R. O'Hara,
T. Casalunovo, L.K. Conlin, M. D'arcy, E.C. Frackelton, E.A. Geiger, C. Haldeman-
Englert, M. Imielinski, C.E. Kim, L. Medne, K. Annaiah, J.P. Bradfield,
E. Dabaghyan, A. Eckert, C.C. Onyiah, S. Ostapenko, F.G. Otieno, E. Santa,
J.L. Shaner, R. Skraban, R.M. Smith, J. Elia, E. Goldmuntz, N.B. Spinner,
E.H. Zackai, R.M. Chiavacci, R. Grundmeier, E.F. Rappaport, S.F. Grant,
P.S. White, H. Hakonarson, High-resolution mapping and analysis of copy
number variations in the human genome: a data resource for clinical and
research applications. Genome Res. 19 (2009) 1682e1690.
[37] A. Sharp, D. Robinson, P. Jacobs, Age- and tissue-specific variation of X chro-
mosome inactivation ratios in normal women. Hum. Genet. 107 (2000)
343e349.
[38] C. Shaw-Smith, R. Redon, L. Rickman, M. Rio, L. Willatt, H. Fiegler, H. Firth,
D. Sanlaville, R. Winter, L. Colleaux, M. Bobrow, N.P. Carter, Microarray based
comparative genomic hybridisation (array-CGH) detects submicroscopic
chromosomal deletions and duplications in patients with learning disability/
mental retardation and dysmorphic features. J. Med. Genet. 41 (2004)
241e248.
[39] M.I. Shevell, B.A. Bejjani, M. Srour, E.A. Rorem, N. Hall, L.G. Shaffer, Array
comparative genomic hybridization in global developmental delay. Am. J.
Med. Genet. B Neuropsychiatr. Genet. 147B (2008) 1101e1108.
[40] N. Shur, M. Vecchiotti, K. Reddy, S. Gunn, D. Abuelo, A rare familial duplication
of Xp22.31 that adds to characterization of phenotype-genotype correlations
of the short arm of the X chromosome. Am. S. Hum. Genet. 58 (2008) P111.
[41] D.H. Skuse, X-linked genes and mental functioning. Hum. Mol. Genet. 14 (1)
(2005) R27eR32.
[42] C. Tyson, C. Harvard, R. Locker, J.M. Friedman, S. Langlois, M.E. Lewis, M. van
Allen, M. Somerville, L. Arbour, L. Clarke, B. McGilivray, S.L. Yong, J. Siegel-
Bartel, E. Rajcan-Separovic, Submicroscopic deletions and duplications in
individuals with intellectual disability detected by array-CGH. Am. J. Med.
Genet. 139 (2005) 173e185.
[43] J. Wagenstaller, S. Spranger, B. Lorenz-Depiereux, B. Kazmierczak, M. Nathrath,
D. Wahl, B. Heye, D. Glaser, V. Liebscher, T. Meitinger, T.M. Strom, Copy-
number variations measured by single-nucleotide-polymorphism oligonu-
cleotide arrays in patients with mental retardation. Am. J. Hum. Genet. 81
(2007) 768e779.
[44] L.A. Weiss, Y. Shen, J.M. Korn, D.E. Arking, D.T. Miller, R. Fossdal,
E. Saemundsen, H. Stefansson, M.A. Ferreira, T. Green, O.S. Platt, D.M. Ruderfer,
C.A. Walsh, D. Altshuler, A. Chakravarti, R.E. Tanzi, K. Stefansson,
S.L. Santangelo, J.F. Gusella, P. Sklar, B.L. Wu, M.J. Daly, Autism Consortium,
Association between microdeletion and microduplication at 16p11.2 and
autism. N. Engl. J. Med. 358 (2008) 667e675.