Combining the Sibling Disequilibrium Test and Transmission/Disequilibrium Test for Multiallelic Markers

The American Journal of Human Genetics (Impact Factor: 10.93). 07/1999; 64(6):1785-6. DOI: 10.1086/302421
Source: PubMed


Available from: David Curtis
Letters to the Editor
Am. J. Hum. Genet. 64:1775–1778, 1999
Mutations of the TIGR/MYOC Gene in Primary
Open-Angle Glaucoma in Korea
To the Editor:
Glaucoma affects
13.5 million people in North America.
Although treatable in the early stages, often it is not
diagnosed and treated in time, which results in irre-
versible blindness. Primary open-angle glaucoma
(POAG), the most common form of glaucoma, repre-
150% of glaucoma cases in Western countries
(Raymond 1997). POAG is an eye disorder characterized
by the progressive excavation of optic disks, typical vi-
sual-field defects, and optic-nerve damage. Many fam-
ilies with autosomal dominant POAG have been re-
ported (Brezin et al. 1997). The initial finding of a
linkage between the juvenile-onset form of POAG
(JOAG) and markers at the GLC1A locus on 1q21–31
(Sheffield et al. 1993) was subsequently confirmed
in other families (MIM 137750). Recently, Stone et
al. (1997) showed that mutations of the TIGR/MYOC
(trabecular meshwork–induced glucocorticoid-response
protein/myocilin) gene (MIM 601652), which maps at
the GLC1A locus, were responsible for JOAG, as well
as for middle-age–onset POAG. After publication of that
report, other investigators have described various mu-
tations in TIGR/MYOC in patients with JOAG/POAG
(Adam et al. 1997; Suzuki et al. 1997; Mansergh et al.
1998; Michels-Rautenstrauss et al. 1998). Although a
few mutations were found in different exons (Alward et
al. 1998), most mutations reported to date are clustered
in the third exon. This exon is evolutionarily conserved
and bears sequence homology with the olfactomedin
gene (Yokoe and Anolt 1993). We investigated whether
Korean patients with JOAG/POAG have the mutations
of the TIGR/MYOC gene. In our report we present two
patients whose genomes harbor different TIGR/MYOC
gene mutations.
After obtaining informed consent, we collected pe-
ripheral-blood samples from 45 unrelated patients with
POAG who visited the Department of Ophthalmology
at the Catholic University Medical Center, Korea. The
patients were given diagnoses of POAG on the basis of
findings from ocular examinations. Patients with POAG
were determined to be affected if intraocular pressure
(IOP) was
122 mmHg in both eyes and if the cup/disk
ratio was
10.3, with characteristic visual-field loss and
gonioscopic grade III or IV. Blood samples were also
obtained from 106 patients who had visited the Catholic
University Medical Center because of diseases other than
POAG and who served as controls. Genomic DNA was
extracted from each blood sample by means of DNA-
isolation kits for mammalian blood (Boehringer Mann-
heim). The DNA fragments encoding portions of TIGR/
MYOC protein were amplified by means of PCR and
were analyzed by cold SSCP. The primers used for PCR
are shown in table 1. The nucleotide numbers corre-
spond to those in the work of Nguyen et al. (1998). A
PCR reaction was performed in a 30-ml volume con-
taining 50 ng of genomic DNA, 0.2 mM each of forward
and reverse primers, 0.19 mM of each deoxyribonucle-
otide triphosphate (dATP, dCTP, dGTP, and dTTP), 50
mM KCl, 1.5 mM MgCl
, and 10 mM Tris-HCl, pH
8.3. The PCR product was denatured and separated on
a 20% polyacrylamid Tris-borate EDTA gel (Novex).
The DNA fragments were visualized by staining the gel
with ethidium bromide solution. The PCR products ex-
hibiting aberrant SSCP patterns were subcloned and se-
quenced by means of an ALF express DNA sequencer
(Pharmacia) with fluorescent dye–primer chemistry.
Multiple clones were sequenced to confirm the presence
of both normal and mutant clones.
Of the 45 patients who were screened for mutations,
2 were found to carry variants in the TIGR/MYOC gene.
We were able to recruit one patient for a family study.
This family consists of five members (fig. 1A). The pro-
band (individual 3; fig. 1A) was given a diagnosis of
JOAG at age 15 years. At the time of diagnosis, her IOPs
were 29 mmHg (right eye [OD]) and 30 mmHg (left eye
[OS]), and she exhibited severe visual-field loss and op-
tic-nerve damage. Because the disease progressed ag-
gressively and medical treatment was not effective, sur-
gery was required to control the progress of glaucoma
in both eyes. None of the other family members was
given a diagnosis of POAG, although individuals 1, 2,
and 4 had slightly elevated IOPs compared with indi-
vidual 5 (table 2), as well as with those in the control
Another patient whom we were unable to recruit for
a family study received a diagnosis of POAG at age 59
Page 1
1776 Letters to the Editor
Table 1
Primer Sequences for PCR and Conditions for PCR and SSCP
(bp)Forward Reverse
for PCR
for SSCP
88 to 281 5
56 10 204
1014 to 1202 5
62 14 189
1296 to 1493 5
60 8 198
283 to 281 5
60 384
177 to int1A 5
62 593
int1B to 795 5
56 170
796 to 1316 5
56 521
1296 to 3
60 283
Numbers correspond to those in the work of Nguyen et al. (1998).
Three parts of the MYOC coding region were analyzed by SSCP, and the rest were used for sequencing the MYOC gene of the proband,
individual 3.
Figure 1 Pedigree of a family, analysis of SSCP, and sequences
of regions with mutations in the TIGR/MYOC gene. A, Pedigree of
a family with the C201T mutation. The unblackened symbol denotes
a genotypically and phenotypically unaffected individual. The black-
ened symbol indicates an individual with documented evidence of
POAG. A symbol with a dot indicates an obligatory carrier. The arrow
indicates the proband. B, SSCP analysis of C201T (lanes 1–5) and
C1123T (lanes 6 and 7) mutations, described in the text. Lane M,
HaeIII fragments of f # 174RF. Lanes 1, 2, 4, and 6, Heterozygotes.
Lane 3, Homozygote. Lanes 5 and 7, Normal. C, Sequence comparison
between the TIGR/MYOC gene harboring mutations and the wild
type. The altered nucleotides are shown in boldface and are denoted
by an asterisk (*). Rows TIGR, wild-type sequence. Rows mTIGR,
sequence from patients with POAG.
years and displayed a moderate phenotype. Her IOPs
before medication treatment (betaxolol HCl two times
a day and dorzolamide HCl three times a day) were 26
mmHg (OD) and 24 mmHg (OS). Severe loss (OU) of
visual field was observed, and cup/disk ratios were 0.8
(OD) and 0.6 (OS). Because her sister and one of her
daughters were also given diagnoses of POAG, a genetic
basis for its etiology could be suggested.
Mutations of the TIGR/MYOC gene were detected by
SSCP analysis and were confirmed by sequence analysis.
SSCP followed by sequence analysis of the proband, in-
dividual 3, revealed a CrT transition at nucleotide 201
in exon 1. This alteration resulted in a nonsense mu-
tation at codon 46 (argininerTAG, amber; fig. 1Ca).
The complete TIGR/MYOC coding region of the pro-
band, individual 3, was sequenced to confirm that there
was no alteration other than a C201T transition. This
mutation was presumed to result in a truncated protein
with 45 amino acids. This proband carried only mutated
alleles (fig. 1B), which was indicative of homozygosity
for the TIGR/MYOC gene. Further analysis of her fam-
ily revealed that her father, mother, and sister were het-
erozygous for TIGR/MYOC, apparently without any de-
tectable symptom. The proband’s brother, individual 5,
had two normal copies and did not have any symptoms
of POAG. The homozygote with this mutation is severely
affected, whereas heterozygotes do not display any de-
tectable POAG symptom. Thus, this family shows pos-
sible autosomal recessive inheritance of JOAG, whereas
other families, which have been described elsewhere,
show autosomal dominant inheritance. The heterozy-
gotes may develop POAG later in life, because it is
known that late-onset POAG shows age-dependent pen-
etrance. Another possibility is that, because of consan-
guinity, the proband is homozygous at other loci that
may modify the glaucoma phenotype. Importantly, there
was an individual with the C201T mutation in the con-
trol group (1/106), whose POAG status has not been
documented. This suggests that the C201T mutation
may not be so rare in the Korean population, which
supports the notion of autosomal recessive inheritance
of JOAG. However, at this point, we cannot exclude the
possibility that the proband’s parents are genetically re-
Page 2
Letters to the Editor 1777
Table 2
Clinical Data on Members of a Family with Familial JOAG and Their Genotypes for the MYOC Gene, Determined by SSCP and
1 42 .5 .5 22 24 IV(D40r) NA (OU) None Heterozygous
2 40 .4 .3 20 20 IV(D40r) NA (OU) None Heterozygous
3 15 .4 .5 29 30 IV(D45r) Nasal step (OU) Trabe (OU) Homozygous
4 13 .4 .5 22 22 IV(D40r) NA (OU) None Heterozygous
5 12 .6 .4 17 17 IV(D40r) NA (OU) None Wild type
As in figure 1A.
NA 5 not affected.
As determined by SSCP and sequencing.
lated. Whereas further study is required to determine the
recessive inheritance for JOAG/POAG, this family pro-
vides an opportunity to elucidate a molecular mecha-
nism for TIGR/MYOC in POAG pathogenesis.
The mutation identified in a different family was het-
erozygous and was confirmed to be a CrT transition at
nucleotide 1123 (fig. 1Cb). None of the 106 patients
in the control group was found to contain the same
mutation. The change resulted in the conversion
Thr353 Ile (fig. 1B, lane 6, and Cb). This mutation re-
sided in the olfactomedin-homology region in the third
exon yet was different from all the mutations in the
TIGR/MYOC gene that have been reported to date.
Threonine at 353 residue is a putative phosphorylation
site by protein kinase C, predicted by the PROSITE Pat-
tern DB search program. Thus, the phosphorylation of
TIGR/MYOC may play a role in the regulation of IOP
in trabecular-meshwork cells.
The prevalence of C201T and C1123T mutations in
the TIGR/MYOC gene in Korean patients was estimated
by screening 45 unrelated patients with JOAG/POAG.
The prevalence of each mutation in the TIGR/MYOC
gene was 1 (2.2%) of 45 in patients with JOAG/POAG;
thus the combined prevalence was 4.4%.
With support from another report (Kee and Ahn
1997), the present study indicates that the mutations in
the TIGR/MYOC gene are responsible for JOAG/POAG
in Korean patients. That report also described a phe-
notypic homozygote with JOAG linked to GLC1A, and
it alluded to the autosomal recessive inheritance of
JOAG. The analysis of the function of the different mu-
tant forms of TIGR/MYOC in the regulation of IOP will
enhance our understanding of POAG pathogenesis.
We are most grateful to Dr. T. Nguyen, for sharing valuable
information regarding TIGR/MYOC sequences, and to Min-
jung Kim, for technical help. This work was supported by a
grant from the Catholic University Medical Center Research
Fund for Special Project (1998).
1, 2
1, 2
2, 3
1, 2
1, 2
Department of Ophthalmology, and
Institutes of Medical Science, Catholic University
Medical College, and
Department of Ophthalmology,
St. Mary’s Hospital, Seoul
Electronic-Database Information
The URLs for data in this study are as follows:
Online Mendelian Inheritance in Man (OMIM), http:// (for POAG [MIM 137750]
and TIGR/MYOC [MIM 601652])
Adam MF, Belmouden A, Binisti P, Brezin AP, Valtot F,
Bechetoille A, Dascotte J-C, et al (1997) Recurrent muta-
tions in a single exon encoding the evolutionarily conserved
olfactomedin-homology domain of TIGR/MYOC in familial
open-angle glaucoma. Hum Mol Genet 6:2061–2097
Alward WL, Fingert JH, Coote MA, Johnson AT, Lerner SF,
Junqua D, Durcan FJ, et al (1998) Clinical features asso-
ciated with mutations in the chromosome 1 open-angle glau-
coma gene (GLC1A). N Engl J Med 338:1022–1027
Brezin AP, Bechetoille A, Harmad P, Valtot F, Berkani M, Bel-
mouden A, Adam MF, et al (1997) Genetic heterogeneity of
primary open angle glaucoma and ocular hypertension: link-
age to GLC1A associated with an increased risk of severe
glaucomatous optic neuropathy. J Med Genet 34:546–552
Kee C, Ahn BH (1997) TIGR gene in primary open-angle glau-
coma and steroid-induced glaucoma. Korean J Ophthalmol
Mansergh FC, Kenna PF, Ayuso C, Kiang AS, Humphries P,
Farrar GJ (1998) Novel mutations in the TIGR/MYOC gene
in early and late onset open angle glaucoma. Hum Mutat
Page 3
1778 Letters to the Editor
Michels-Rautenstrauss KG, Mardin CY, Budde WM, Liehr T,
Polansky J, Nguyen T, Timmerman V, et al (1998) Juvenile
open angle glaucoma: fine mapping of the TIGR/MYOC
gene to 1q24.3–q25.2 and mutation analysis. Hum Genet
Nguyen TD, Chen P, Huang WD, Chen H, Johnson D, Po-
lansky JR (1998) Gene structure and properties of TIGR/
MYOC, an olfactomedin-related glycoprotein cloned from
glucocorticoid-induced trabecular meshwork cells. J Biol
Chem 273:6341–6350
Raymond V (1997) Molecular genetics of the glaucomas: map-
ping of the first five “GLC” loci. Am J Hum Genet 60:
Sheffield VC, Stone EM, Alward WLM, Drack AV, Johnson
AT, Atreb LM, Nichols BE (1993) Genetic linkage of familial
open-angle glaucoma to chromosome 1q21–q31. Nat Genet
Stone EM, Fingert JH, Alward WLM, Nguyen TD, Polansky
JR, Sunden SLF, Nishimura D, et al (1997) Identification of
a gene that causes primary open angle glaucoma. Science
Suzuki Y, Shirato S, Taniguchi F, Ohara K, Nishimaki K, Ohta
S (1997) Mutations in the TIGR gene in familial primary
open-angle glaucoma in Japan. Am J Hum Genet 61:
Yokoe H, Anolt RRH (1993) Molecular cloning of olfacto-
medin, an extracellular matrix protein specific to olfactory
neuroepithelium. Proc Natl Acad Sci USA 90:4655–4659
Address for correspondence and reprints: Dr. Choun-Ki Joo, Department of
Ophthalmology, Catholic University Medical College, 505 Banpo-dong, Seocho-
ku, Seoul 137-701, Korea. E-mail:
q 1999 by The American Society of Human Genetics. All rights reserved.
Am. J. Hum. Genet. 64:1778–1781, 1999
Evidence for the Genetic Heterogeneity of
Nephropathic Phenotypes Associated with Denys-
Drash and Frasier Syndromes
To the Editor:
The association of constitutional heterozygous muta-
tions of the Wilms tumor 1 (WT1) gene with the majority
of cases of both Denys-Drash syndrome (DDS [MIM
194080]) (Denys et al. 1967; Drash et al. 1970) and
Frasier syndrome (FS [MIM 136680]) (Frasier et al.
1964) has been well described; however, the char-
acteristic nephropathies connected with these syn-
dromes—that is, diffuse mesangial sclerosis (DMS [MIM
256370]) and focal segmental glomerulosclerosis (FSGS
[MIM 603278])—occur more commonly in early life, as
disorders confined to the kidney. It is unclear how fre-
quent WT1 gene mutations are in this population.
The WT1 gene encodes a transcription factor critical
for the normal development and function of the uro-
genital tract, reflected by the knockout-mouse homo-
logue of WT1, in which homozygous inactivation of
Wt1 causes absent kidneys and malformation of the go-
nads (Kreidberg et al. 1993). Alternative splicing results
in at least 4 different zinc-finger protein isoforms, and
the possibility of RNA editing and use of an alternative
initiation codon increases this number to 16 (Bruening
and Pelletier 1996). A correct ratio of isoforms appears
crucial for normal gene function. During renal devel-
opment, maximum WT1 expression occurs in condens-
ing mesenchyme and during the mesenchymal-epithelial
switch. In the mature nephron, WT1 expression is con-
fined to the podocytes, a highly specialized layer of ep-
ithelial cells in the glomerulus. Here it may have a role
in the maintenance of these cells, thus affecting the in-
tegrity of the glomerular filter (Pritchard-Jones et al.
1990). Accumulating evidence supports a regulatory role
for WT1 in kidney development, although little is known
about either its targets or which genetic cascades are
affected by its abnormal function.
A wide variety of WT1 mutations is seen in DDS—a
triad of intersex, nephropathy due to DMS, and Wilms
tumor—making genotype-phenotype correlation diffi-
cult even with the aid of computer programs (Jeanpierre
et al. 1998a). In contrast, FS is caused by specific intronic
point mutations that disrupt the exon 9 alternative
splice-donor site, reversing the normal WT1 1/2 KTS
isoform ratio (Klamt et al. 1998). It is also associated
with intersex, but there is no predisposition to Wilms,
and the nephropathy typically results from FSGS. These
conditions act as human disease models of the effects of
WT1 gene mutations and provide further strong evi-
dence of WT1’s crucial role in both renal and gonadal
development. How WT1 mutations affect glomerular
development remains a matter of debate; some muta-
tions may be mediated during the progression of ne-
phrogenesis, perhaps through aberrant interactions of
WT1 with genes and proteins important for this process,
thereby causing abnormal glomerular differentiation. It
is equally possible is that the problem could lie in the
terminally differentiated, nondividing podocytes, partic-
ularly if silencing of the normal allele were to occur, as
in some tissues (e.g., placenta and brain [Jinno et al.
1994]), and if the mutant WT1 protein were incapable
of maintaining normal podocyte function. Further study
of animal models with targeted gene mutations may shed
light on this increasingly complex picture.
Jeanpierre et al. (1998b) demonstrated that 4 of their
10 patients with DMS but no other features of DDS had
mutations in exons 8 and 9 of WT1; 3 of these 4 patients
were female. One mutation, 1147TrC (F383L) in exon
9, was novel. Two other mutations, 1186GrA (D396N)
in exon 9 and 1129CrT (H377Y) in exon 8, had been
found in previously reported cases of DDS. The fourth
Page 4
Letters to the Editor 1779
mutation, (14CrT) in intron 9, had been shown to
cause FS (Barbaux et al. 1997; Kikuchi et al. 1998;
Klamt et al. 1998). Schumacher et al. (1998) studied a
broader spectrum of patients with early-onset nephrotic
syndrome, to identify possible WT1 gene mutations; two
of four patients who had DMS but lacked other features
of either DDS or FS had mutations; one of them had a
newly discovered mutation, 1135GrT (G379C) in exon
8, and the other had a mutation, 1180CrT (R394W)
in exon 9, that previously had been reported described
in a case of DDS; both patients were female and were
reported to have other unspecified features consistent
with this syndrome.
We tested the hypothesis that WT1 gene mutations
occur in cases of DMS and of congenital/early-onset
FSGS occurring in the absence of other features of DDS
or FS. Our intent was to identify how common muta-
tions of the WT1 gene were in this population and to
begin establishing the boundaries of the DDS/FS spec-
trum of disease.
A series of 30 patients, 22 with DMS and 8 with FSGS,
were screened for mutations of WT1 (table 1). The di-
agnosis of DMS or FSGS was established on the basis
of either renal biopsy or postmortem renal histological
findings. The majority of surviving patients with DMS
were followed-up with yearly renal ultrasounds, to ex-
clude Wilms tumor. Twenty-seven of the 30 patients had
documented karyotype analysis. Congenital malforma-
tions were investigated by clinical examination and the
appropriate investigations. None of the patients with
psychomotor abnormalities were given a diagnosis of
Galloway-Mowat syndrome. Abnormalities of the in-
ternal gonads were examined by pelvic ultrasound, and
in some cases, by laparotomy. Patient 14 had an equiv-
ocal report with regard to karyotype—and, therefore,
possible intersex status—but no other features of DDS.
Two patients with DMS (patients 21 and 22) had DDS
with known 1180CrT (R394W) mutations in exon 9
and were used as positive controls. Genomic DNA was
obtained from blood of 23 patients and from paraffin-
section material in 7 patients, by standard phenol-chlo-
roform extraction methods. The two most common DDS
mutations located in exon 9 (i.e., 1180CrT [R394W],
60%; and 1186GrA [D396N], 15%) abolish an RsrII
site in zinc finger 3 (Little et al. 1993). Heterozygous
loss of the site was observed in patients 21 and 22, as
expected, whereas patients 1–20 showed normal restric-
tion-enzyme digests. We used SSCP to analyze exons
1–10 after PCR amplification, using oligonucleotides
that had also been published by Baird et al. (1992a) and
radiolabeling them with [
P]. Products were electro-
phoresed, at 47C, on both 5% and 10% nondenaturing
polyacrylamide gels. In addition, exons 5, 8, and 9 were
directly sequenced on an ABI 377 automated sequencer.
PCR products were purified either on spin columns (Pro-
mega Biotec) or by gel extraction (Qiagen Gel Extraction
Kit). Samples (2–5 ml) of the purified product were se-
quenced by use of a Thermo-Sequenase dye terminator
cycle sequencing kit and 5 pmol of the original primer.
No WT1 mutations were detected in either the 20
patients with isolated DMS or the 7 patients with iso-
lated FSGS, although patient 14 demonstrated a newly
discovered polymorphism in codon 178 CrT in exon 4,
which conserves a threonine, and patient 16 a demon-
strated previously reported (Groves et al. 1992) poly-
morphism, an ArG transition in codon 313 (arginine)
in exon 7, which destroys an AflIII restriction-enzyme
recognition site. Both DDS patients showed the expected
(as reported by Baird et al. 1992b) 1180CrT mutations
in exon 9; in addition, a 15GrA mutation in intron 9
was detected during this study in patient 30, who had
classic features of FS (previously reported by Klamt et
al. 1998).
In conclusion, when a larger, more generalized pop-
ulation of cases with DMS is examined, mutations of
WT1 are both less frequent than initial data have sug-
gested and absent in isolated FSGS. This confirms the
genetic heterogeneity of DMS and FSGS, despite the uni-
form renal histological findings seen throughout this
group of conditions. Supportive evidence comes from
linkage studies of familial FSGS, which show that the
candidate gene for this condition lies on chromosome
1q25-q31 (Fuchshuber et al. 1995). Isolated DMS and
FSGS may therefore also result from abnormalities of
other glomerular genes, perhaps downstream of WT1
and mimicking the effects of WT1 mutations seen in
DDS and FS.
We propose that, if WT1 gene mutations are present
in isolated renal DMS or FSGS, this lends strong support
to a diagnosis of DDS or FS, and the spectrum of DDS
should be broadened to include occasional cases in
which the characteristic nephropathy does occur alone.
In this situation, WT1 mutations appear more common
in phenotypic females, which affirms the less critical role
of WT1 in female gonadal development, as has been
suggested by experimental data (Nachtigal et al. 1998).
This finding also correlates with cases of FS in which
individuals with characteristic WT1 mutations in intron
9 and with a 46,XX karyotype develop nephropathy but
have no obvious gonadal abnormality (Klamt et al.
1998). Karyotype analysis remains the most important
first-line investigation in phenotypic females with iso-
lated renal DMS or FSGS; however, WT1 mutation anal-
ysis should also be considered in isolated DMS and
FSGS, since the presence of characteristic WT1 gene mu-
tations may be important in the determination of Wilms
tumor risk, especially in 46,XX individuals who may
not demonstrate any other clues for an underlying di-
agnosis of DDS.
Page 5
Table 1
Results and Clinical Status
Case Karyotype
Genital Status
(Internal or
Other Renal
or Extrarenal
History? WT1 Gene Mutation? Age at Last Follow-up
Onset of
Dialysis? Nephrectomy? Transplant?
1 46,XX Day 1 Yes No Yes No Female Psychomotor Yes No; exons 1–10 10 years
2 46,XX Day 1 Yes Yes, right and left Yes No Female No No No; exons 1–10 4 years
3 46,XX Day 1 Yes Yes, right and left No No Female No Yes No; exons 1–10 8 years
4 46,XY Day 1 Yes No No No Male No No No; exons 1–10 (Died at age 3 years)
5 46,XY 11 year Yes Yes, right and left Yes No Male No No No; exons 1–10 13 years
6 46,XY
11 year Yes Yes, right and left Yes No Male No No No; exons 1–10 11 years
7 46,XY 11 year No No No No Male No No No; exons 1–10 3 years
8 46,XX Day 1 No No No No Female Psychomotor,
No No; exons 1–10 (Died at age 7 mo)
9 46,XY 6 mo Yes No Yes No Male Pulmonary stenosis No No; exons 1–10 14 years
10 46,XX 11 year No No No No Female Psychomotor No No; exons 1–10 (Died at age 3 years)
11 46,XY 18 mo Yes No No No Male No No No; exons 1–10 5 years
12 46,XX Day 1 Yes No No No Female Psychomotor ? No; exons 2–10 (Died at age 2 mo)
13 Unknown Day1 Yes No No No Female No Yes No; exons 2–10 (Died at age 7 mo)
14 46,XY Day1 No No No No Female No No Exon 4 polymorphism (Died at age 3 mo)
15 46,XX Day1 No No No No Female Psychomotor No No; exons 2–10 (Died at age 10 mo)
16 46,XY Day 1 No No No No Male Cryptorchidism? No AflIII polymorphism (Died at age 2 mo)
17 Unknown Day 1 No No No No Unknown No No No; exons 2–10 (Died at age 10 wk)
18 Unknown Day 1 No No No No Unknown No No No; exons 2–10 (Died at age 8 mo)
19 46,XX 1 year Yes No No No Female Left kidney
No No; exons 5, 8, 9 2 years
20 46,XY 9 mo No No No No Male no No No; exons 5, 8, 9 14 mo
21 46,XY
12 years Yes Yes, right and left Yes Wilms Female No No Exon 9: 1180CrT
13 years
22 46,XY 1 year Yes Yes, right and left Yes Wilms Ambiguous No No Exon 9: 1180CrT
11 years
23 46,XX 4 mo No No No Female No No Exons 1–10 7 years
24 46,XX 11 year ? ? ? No Female No Yes Exons 1–10 Lost to follow-up
25 46,XY
11 year ? ? ? No Male No Yes Exons 1–10 Lost to follow-up
26 46,XX Day 1 No No No No Female No Yes Exons 1–10 3 years
27 46,XY 18 mo Yes No Yes No Male No Yes Exons 1–10 13 years
28 46,XY
11 year Yes ? Yes No Male No No Exons 1–10 7 years
29 46,XX 1 year Yes Yes, right and left No No Female Spondyloepiphyseal
No Exons 1–10 (Died at age 6 years)
30 46,XY 5 years 8 years No Yes Gonadoblastoma Female No ? Intron 9 15GrA 16 years
Page 6
Letters to the Editor 1781
A.B.K. is funded by a National Kidney Research Fund Twis-
tington Higgins Fellowship. We would like to thank the renal
units throughout the United Kingdom, which kindly provided
patients and clinical information for this study.
A. B. K
R. G
T. M. B
P. S
Molecular Medicine,
and Oncology Units, Institute of Child Health,
Electronic-Database Information
Online Mendelian Inheritance in Man (OMIM): http:// (for FS [MIM 136680], DDS
[MIM 194080], DMS [MIM 256370], and FSGS [MIM
Baird PN, Groves N, Haber DA, Housman DE, Cowell JK
(1992a) Identification of mutations in the WT1 gene in tum-
ours from patients with the WAGR syndrome. Oncogene 7:
Baird PN, Santos A, Groves N, Jadresic L, Cowell JK (1992b)
Constitutional mutations in the WT1 gene in patients with
Denys-Drash syndrome. Hum Mol Genet 1:301–305
Barbaux S, Niaudet P, Gubler MC, Grunfeld JP, Jaubert F,
Kuttenn F, Fekete CN, et al (1997) Donor splice site mu-
tations are responsible for Frasier syndrome. Nat Genet 17:
Bruening W, Pelletier J (1996) A non-AUG translational ini-
tiation event generates novel WT1 isoforms. J Biol Chem
Denys P, Malvaux P, van den Berghe H, Tanghe W, Proemans
W (1967) Association d’un syndrome anatomo-patholo-
gique de pseudo-hermaphrodisme masculin, d’un tumeur de
Wilms’ d’un nephropathie parenchymarteuse et d’un mo-
saicisme XX/XY. Arch Pediatr 24:729–739
Drash A, Sherman F, Harmann W, Blizzard R (1970) A syn-
drome of pseudohermaphroditism, Wilm’s tumour, hyper-
tension and degenerative renal disease. J Pediatr 76:585–593
Frasier SD, Bashore RA, Mosier HD (1964) Gonadoblastoma
associated with pure gonadal dysgenesis in monozygotic
twins. J Pediatr 64:740–745
Fuchshuber A, Jean G, Gribouval O, Gubler MC, Broyer M,
Beckmann JS, Niaudet P, et al (1995) Mapping a gene
(SRN1) to chromosome 1q25-q31 in idiopathic nephrotic
syndrome confirms a distinct entity of autosomal recessive
nephrosis. Hum Mol Genet 4:2155–2158
Groves N, Baird PN, Hogg A, Cowell JK (1992) A single base
pair polymorphism in the WT1 gene detected by single-
stranded conformational polymorphism analysis. Hum Ge-
net 90:440–442
Jeanpierre C, Beroud C, Niaudet P, Junien C (1998a) Software
and database for the analysis of mutations in the human
WT1 gene. Nucleic Acids Res 26:271–274
Jeanpierre C, Denamur E, Henry I, Cabanis M-O, Luce S,
Ce´cille A, Elion J, et al (1998b) Identification of constitu-
tional WT1 mutations, in patients with isolated diffuse mes-
angial sclerosis, and analysis of genotype/phenotype corre-
lations by use of a computerized mutation database. Am J
Hum Genet 62:824–833
Jinno Y, Yun K, Nishiwaki K, Kubota T, Ogawa O, Reeve AE,
Niikawa N (1994) Mosaic and polymorphic imprinting of
the WT1 gene in humans. Nat Genet 6:305–309
Kikuchi H, Takata A, Akasaka Y, Fukuzawa R, Yoneyama H,
Kurosawa Y, Honda M, et al (1998) Do intronic mutations
affecting splicing of WT1 exon 9 cause Frasier syndrome?
J Med Genet 35:45–48
Klamt B, Koziell AB, Poulat F, Wieacker P, Scambler P, Berta
P, Gessler M (1998) Frasier syndrome is caused by defective
alternative splicing leading to an altered ratio of WT1 1/2
KTS splice isoforms. Hum Mol Genet 7:709-714
Kreidberg JA, Sariola H, Loring JM, Maeda M, Pelletier J,
Houseman DE, Jaenisch R (1993) WT1 is required for early
kidney development. Cell 74:679-691
Little MH, Williamson KA, Mannens M, Kelsey A, Gosden
C, Hastie ND, van Heyningen V (1993) Evidence that WT1
mutations in Denys-Drash syndrome patients may act in a
dominant-negative fashion. Hum Mol Genet 2:259-264
Nachtigal MW, Hirokawa Y, Enyeart-VanHouten DL, Flan-
agan JN, Hammer GD, Ingraham HA (1998) Wilms’ tumor
1 and Dax-1 modulate the orphan nuclear receptor SF-1 in
sex-specific gene expression. Cell 93:445-454
Pritchard-Jones K, Fleming S, Davidson D, Bickmore W, Por-
teous D, Gosden C, Bard J, et al (1990) The candidate
Wilms’ tumour gene is involved in genitourinary develop-
ment. Nature 346:194-197
Schumacher V, Scharer K, Wuhl E, Altrogge H, Bonzel KE,
Guschmann M, Neuhaus TJ, et al (1998) Spectrum of early
onset nephrotic syndrome associated with WT1 missense
mutations. Kidney Int 53:1594-1600
Address for correspondence and reprints: Dr. Ania Koziell, Molecular Medi-
cine Unit, Institute of Child Health, 30 Guilford Street, London WC1N 1EH.
*Present affiliation: Department of Oncology, The Birmingham Children’s
Hospital NHS Trust, Birmingham, England.
q 1999 by The American Society of Human Genetics. All rights reserved.
Am. J. Hum. Genet. 64:1781–1785, 1999
Rett Syndrome in a Boy with a 47,XXY Karyotype
To the Editor:
Rett syndrome (RS [MIM 312750]) is a progressive en-
cephalopathy characterized by severe mental retarda-
tion, autism, apraxia, seizures, stereotypical hand move-
ments, and deceleration of head growth. Its prevalence
is estimated at 1:10,000–15,000 female births (Hagberg
1995). The majority of cases are sporadic, but rare re-
ports of familial recurrence have been made. In addition,
all but 1 of the 10 MZ twins reported in the literature
are concordant, whereas all 11 DZ twins reported are
Page 7
1782 Letters to the Editor
discordant for the disorder (Migeon et al. 1995). Lab-
oratory investigations have not revealed any metabolic
abnormalities in affected individuals.
Chromosomal abnormalities and/or association with
another syndrome have already been reported in patients
with RS: a translocation t(X;22)(p11.22;p11) by Journel
et al. (1990), a translocation t(X;3)(p21.3;p25.2) by
Zoghbi et al. (1990) and Ellison et al. (1993), a deletion
del(3)(3p25.1-p25.2) by Wahlstro¨ m et al. (1996), and a
deletion del(13)(13q12.1-q21.2) by Herder et al. (1996).
RS was described in association with fragile X by Alem-
bick et al. (1995) and with Down syndrome by Eas-
thaugh et al. (1996). No concordance for the chromo-
somal abnormalities has been found, however, since
different chromosomes and/or breakpoints were in-
volved in each case. Vorsanova et al. (1996) reported
a boy with RS and karyotype 46,XY/47,XXY (the
47,XXY cell line was observed in 6%–12% of the stud-
ied lymphocytes).
Here, we describe a patient with RS and a 47,XXY
karyotype. The propositus, a male patient born in Jan-
uary 1995, was referred for genetic studies at age 28
mo. His parents are healthy, were aged 30 years (father)
and 29 years (mother) at the time of the birth, and are
not consanguineous. The child was born at term, after
an uneventful pregnancy. His birth weight was 3.330 g
(25th–50th percentile), his Apgar indices were 6 (1st
minute) and 7 (5th minute), and his birth occipitofrontal
head circumference was 32 cm (2.5 percentile). The per-
inatal period was uneventful. The propositus is the
fourth child, and his older sibs—two boys aged 16 and
9 years and one girl aged 13 years—are normal. There
is no history of neuropsychiatric diseases in the families
of the mother or the father. The propositus showed nor-
mal development until age 8 mo. At that time, he sat
without support, played normally, and was able to grasp
objects and to put food into his mouth. He had also
started to say some words comprehensibly.
The family noticed that, at age 11 mo, he had lost
purposeful hand movements and language skills. He also
began to show regression in social contact. At age 1 year,
he began to show stereotypical hand movements, brux-
ism, and constipation. At age 28 mo, he presented severe
global retardation and slightly diffuse hypotonia. He
was socially isolated and made few spontaneous move-
ments (other than the stereotypical hand movements).
He did not grasp or otherwise show interest in any object
or toy. He could vocalize but did not form any words.
He reacted to luminous and sonorous stimuli. When
standing up with support, he presented axial ataxia.
Bruxism and short episodes of apnea were observed dur-
ing consultation. No focal neurological signs or altera-
tion in cranial nerves were observed. His occipitofrontal
head circumference was 45 cm (2.5 percentile), his
weight was 12.220 g (35th percentile), and his height
was 87 cm (25th percentile).
When the patient was last seen, at age 37 mo, the loss
of purposeful hand movements, the manual apraxia, and
the slight global hypotonia were persistent. The stereo-
typy of his hand movements was midline, was constant
in vigil, and showed a slightly athetoid component.
When walking with support, he presented ataxia/
apraxia. He reacted to luminous and sonorous stimuli.
The episodes of apnea were more frequent and more
sustained. His occipitofrontal head circumference was
46 cm (2.5 percentile), his weight was 15.200 g (35th
percentile), and his height was 94 cm (25th percentile).
Results of electroretinogram, magnetic resonance im-
aging of the brain, and electroencephalogram were nor-
mal. The results for rubella, syphilis, HIV I and HIV II,
cytomegalovirus, herpes, cerebrospinal fluid, and serum
amino acid testing were all normal. Toxoplasmosis test-
ing showed that the patient’s IgG level was slightly in-
creased. However, acquired neurological disorders re-
sulting from congenital toxoplasmosis infection were
ruled out, since the boy was normal from birth until age
8 mo.
Chromosomal analysis, including GTG banding, was
performed on peripheral blood leukocytes as described
by Seabright (1971). Karyotype analyses from all 300
banded metaphase preparations showed 47 chromo-
somes with an extra X chromosome (47,XXY).
To establish the origin of the nondisjunction, we
analyzed DNA from the mother and the propositus
with eight microsatellite markers from the dystrophin
DYSMS; STR 44; STR 45;
STR 49; STR 50; and 3
-19n8. DNA from the father
was not available. DNA analysis showed that the pro-
positus had an allele that was not present in his mother,
indicating, therefore, that the additional sex chromo-
some was paternal in origin—that is, it resulted from
nondisjunction at the paternal first meiotic division.
For X-inactivation analyses, DNA was extracted from
peripheral blood from the mother and the propositus,
and 1 mg of digested (with AluI and CfoI) and nondig-
ested DNA samples were used as templates for ampli-
fication of the androgen receptor (AR) highly polymor-
phic (CAG)
repeat, as reported (Allen et al. 1992;
Edwards et al. 1992). All samples were run in duplicate
in a 5% polyacrylamide gel (19:1 acrylamide:bis-acry-
lamide). A densitometer (Shimadzu CS-9000) was used
to determine the ratio of X inactivation in each sample,
and the mean of two readings was considered for each
case. Since one allele may amplify more than the other,
a correction factor was applied to compensate for un-
equal amplification of alleles. We did this for the mother
and for the son, calculating, first, the ratio between the
two alleles of the undigested DNA and correcting the
final values for preferential PCR amplification (Pegoraro
Page 8
Letters to the Editor 1783
et al. 1994). We calculated the degree of X inactivation
on the digested DNA by normalizing the sum of allele
A plus allele B to 100%, as reported in Sumita et al.
(1998). The analysis of the X-chromosome–inactivation
pattern in blood DNA showed X-inactivation ratios of
73:27 in the mother and 41X
in the affected son.
To rule out a possible diagnosis of Angelman syn-
drome (AS), the methylation status of the locus SNRPN
mapped within the PWS/AS region was assessed by
Southern blotting. The probe used was a 0.6-kb EcoRI-
Notl fragment that contains exon 1 of SNRPN (Glenn
et al. 1996). Methylation assay for AS was analyzed at
the SNRPN CpG island and a normal result was ob-
tained, with the presence of the 0.9-kb band from the
unmethylated paternal allele and a 4.2-kb band from the
methylated maternal allele. This method confirms the
diagnosis in 80% of cases, since in the remaining 20%
AS may be due to UBE3A mutations or other unknown
mechanisms (Kishino et al. 1997; Matsuura et al. 1997)
The parental origin of additional sex chromosomes
was studied by Lorda-Sanchez et al. (1992) in 47 patients
with a 47,XXY chromosome constitution. In 23 (49%)
cases, the error occurred during the first paternal meiotic
division, as observed in the present case. No significant
clinical differences were found among patients of distinct
parental origin.
To date, RS has been convincingly described only in
females. Some cases described as RS syndrome in males
have been reported (Coleman 1990; Eeg-Olofsson et al.
1990; Philippart 1990; Topc¸u et al. 1991; Christen and
Hanefeld 1995; Vorsanova et al. 1996). The clinical
signs and symptoms, however, were but suggestive, atyp-
ical, and/or partial. In the present report, the clinical and
laboratory findings do not overlap with any described
for Klinefelter syndrome. AS was excluded with 80%
certainty, and extensive testing did not disclose any other
alternative etiology, such as infantile neuronal ceroid-
lipofuscinosis. The clinical findings met the criteria of
inclusion and exclusion for the diagnosis of RS (Tre-
vathan and Naidu 1988).
Several authors (Zoghbi et al. 1990; Webb et al. 1993;
Camus et al. 1996; Webb and Watkiss 1996; Krepischi
et al. 1998) reported that, as a group, RS patients tended
to present a higher frequency of moderate skewing
(20%–35% or 65%–80%) of X inactivation in lym-
phocytes, when compared with their mothers and nor-
mal controls, and that this skewing, when present,
favors, in most cases, preferential inactivation of the pa-
ternally inherited X chromosome. On the other hand, it
has been suggested that extreme skewed X inactivation
could prevent manifestation of the RS phenotype in mu-
tant-gene female carriers, which would be consistent
with RS being a male-lethal trait (Schanen and Franke
1998; Xiang et al. 1998). In the present report, analysis
of X inactivation in the proband and his mother did not
show extreme skewed X inactivation, suggesting that the
proband might be the result of a new paternal or ma-
ternal germ line mutation event. However, as shown pre-
viously, it is not known whether the X-inactivation pat-
tern found in DNA from blood is representative of other
tissues and, furthermore, a skewed pattern of X-inacti-
vation in blood is not rare in normal females (Naumova
et al. 1996; Sumita et al. 1998). Therefore, although the
occurrence of moderate skewing is more frequent in RS
patients and extreme skewed X inactivation has been
observed in obligate RS carriers (Sirianni et al. 1998), a
correlation between X-inactivation skewing and the RS
phenotype must be interpreted with caution.
An explanation for the exclusive occurrence of RS in
females, without evidence of male lethality, was pro-
posed by Thomas (1966) on the basis of the fact that
de novo X-linked mutations occurring exclusively in
male germ cells could only be passed on to, and result
in, an affected daughter. Under such a hypothesis, the
absence of affected males is explained by the fact that
sons do not inherit their X chromosomes from their
fathers. Since our patient inherited one of his two X
chromosomes from his father, his RS phenotype would
be consistent with Thomas’s hypothesis if the mutated
gene was on the paternal X chromosome. On the other
hand, RS-affected half sisters with the same mothers
have been described (Archidiacono et al. 1991; Sirianni
et al. 1998). However, under Thomas’s hypothesis, it
would be expected, in rare instances, to find families
with half sisters with the same father, because of ger-
minal mosaicism. This has already been demonstrated
for other disorders such as achondroplasia (Philip et al.
1988) and Duchenne muscular dystrophy (Darras and
Francke 1987) but apparently has not been reported for
In a recent report, Sirianni et al. (1998) postulated
that the relatively high frequency for RS would be ex-
plained by a high mutation rate in either male or female
germ lines. In the present case, it was not possible to
determine whether the mutation was inherited through
paternal or maternal gametes.
With respect to the etiology of RS, several investiga-
tors have suggested the possibility of an alteration in the
timing of replication of a gene (or genes) on the late X
chromosome in RS patients (Riccardi 1986; Martinho
et al. 1990; Kormann-Bortolotto 1992; Webb and Wat-
kiss 1996). If this alteration represents the 0misbehavior0
of a gene (or genes) that should be inactive on the in-
activated X chromosome but, when mutated, does not
respond to XIST (the product of the X-inactivation
center gene), the consequence would be transient func-
tional disomy at one or more loci. Partial functional
disomy as a cause for RS (Webb et al. 1993) and other
abnormal phenotypes, such as hypomelanosis of Ito or
mental retardation, has already been suggested (Journel
Page 9
1784 Letters to the Editor
1990; Schmidt and Du Sart 1992; Correa-Cerro et al.
1997; Wolff et al. 1998). If such a mechanism occurred
in RS patients, this condition could be the result of func-
tional disomy.
The present report, confirming an RS phenotype in a
47,XXY male, is consistent with the hypothesis that two
X chromosomes are required for the manifestation of
Rett syndrome.
The collaboration of Drs. Mariz Vainzof, Maria Rita Passos-
Bueno, and Lygia V. Pereira and of Constancia Urbani is grate-
fully acknowledged. This research was supported with grants
from the Fundac¸a˜o de Amparo a` Pesquisa do Estado de Sa˜o
Paulo, Programa de Apoio a` Nu´ cleos de Exceleˆncia, and Con-
selho Nacional de Desenvolvimento Cientı´fico e Tecnolo´ gico.
P. K
Universidade Mackenzie, and
Departamento de
Biologia, Instituto de Biocieˆncias,
Universidade de Sa˜o Paulo, Sa˜ o Paulo
Electronic-Database Information
The URL for data in this article is as follows:
Online Mendelian Inheritance in Man (OMIM), http:// (for Rett syndrome [MIM
Alembick Y, Dott B, Stoll C (1995) Rett-like syndrome in frag-
ile X syndrome. Genet Couns 6:207–210
Allen RC, Zoghbi HY, Moseley AB, Rosenblatt HM, Belmont
JW (1992) Methylation 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 51:1229–1239
Archidiacono N, Lerone M, Rocchi M, Anvret M, Ozcelik T,
Francke U, Romeo G (1991) Rett syndrome: exclusion map-
ping following the hypothesis of germinal mosaicism for new
X-linked mutations. Hum Genet 86:604–606
Camus P, Abbadi N, Perrier M-C, Che´ry M, Gilgenkrantz S
(1996) X chromosome inactivation in 30 girls with Rett
syndrome: analysis using the probe. Hum Genet 97:247–250
Christen HJ, Hanefeld F (1995) Male Rett variant. Neuro-
pediatrics 26:81–82
Coleman M (1990) Is classical Rett syndrome ever present in
males? Brain Dev 12:31–32
Correa-Cerro LS, Rivera H, Vasquez AI (1997) Functional Xp
disomy and de novo t(X;13)(q10;q10) in a girl with hypo-
melanosis of Ito. J Med Genet 34:161–163
Darras BT, Francke U (1987) A partial deletion of the muscular
dystrophy gene transmitted twice by an unaffected male.
Nature 329:556–558
Easthaugh P, Smith L, Leonard H (1996) Trisomy 21 associ-
ated with Rett syndrome phenotype. Paper presented at the
World Congress on Rett Syndrome, Gothenburg, Sweden,
30 August–1 September
Edwards AL, Hammond HA
, Jin L, Cakey CT, Chakraborty
R (1992) Genetic variation at five trimeric and tetrameric
tandem repeat loci in four human population groups. Gen-
omics 12:241–253
Eeg-Olofsson O, Al-Zuhair AGH, Teebi AS, Zaki M, Daoud
AS (1990) A boy with Rett syndrome? Brain Dev 12:529–
Ellison KA, Roth EJ, McCabe ERB, Chinault AC, Zoghbi HY
(1993) Isolation of a yeast artificial chromosome contig
spanning the chromosomal translocation breakpoint in a
patient with Rett syndrome. Am J Med Genet 47:1124–1134
Glenn CC, Saitoh S, Jong MTC, Filbrandt MM, Surti U, Dris-
coll DJ, Nicholls RD (1996) Gene structure, DNA methyl-
ation, and imprinted expression of the human SNRPN gene.
Am J Hum Genet 58:335–346
Hagberg B (1995) Rett syndrome: clinical peculiarities and
biological mysteries. Acta Paediatr 84:971–976
Herder GA, Skjeldal O, Hagberg B, Tranebjærg L (1996) Con-
genital Rett syndrome phenotype–interstitial deletion chro-
mosome 13 and retinoblastoma. Paper presented at the
World Congress on Rett syndrome, Gothenburg, Sweden,
30 August–1 September
Journel H, Melki J, Turleau C, Munnich A, Grouchy J (1990)
Rett phenotype with X/autosome translocation: possible
mapping to the short arm of chromosome X. Am J Med
Genet 35:142–147
Kishino T, Lalande M, Wagstaff J (1997): UBE3A/E6AP mu-
tations cause Angelman syndrome. Nat Genet 15:70–73
Kormann-Bortolotto MH, Woods CG, Green SH, Webb T
(1992) X-inactivation in girls with Rett syndrome. Clin Ge-
net 42:296–301
Krepischi ACV, Kok F, Otto PG (1998) X chromosome in-
activation patterns in patients with Rett syndrome. Hum
Genet 102:319–321
Lorda-Sanchez I, Binkert F, Maechler M, Robinson WP, Schin-
zel A (1992) Reduced recombination and paternal age effect
in Klinefelter syndrome. Hum Genet 89:524–530
Martinho PS, Otto PG, Kok F, Diament A, Marques-Dias MJ,
Gonzalez CH (1990) In search of a genetic basis for the Rett
syndrome. Hum Genet 86:131–134
Matsuura T, Sutcliffe JS, Fang P, Galjaard R-J, Jiang Y, Benton
CS, Rommens JM, et al. (1997): De novo truncating mu-
tations in E6-AP ubiquitin-protein ligase gene (UBE3A) in
Angelman syndrome. Nat Genet 15:74–73
Migeon BR, Dunn MA, Thomas G, Schmeckpeper BJ, Naidu
S (1995) Studies of X inactivation and isodisomy in twins
provide further evidence that the X chromosome is not in-
volved in Rett syndrome. Am J Hum Genet 56:647–653
Naumova AK, Plenge RM, Bird LM, Leppert M, Morgan K,
Willard HF, Sapienza C (1996) Heritability of X chromo-
some-inactivation phenotype in a large family. Am J Hum
Genet 58:1111–1119
Pegoraro E, Schimke RN, Arahata K, Hayashi Y, Stern H,
Marks H, Glasberg MR, et al (1994) Detection of new pa-
Page 10
Letters to the Editor 1785
ternal dystrophin gene mutations in isolated cases of dys-
trophinopathy in females. Am J Hum Genet 54:989–1003
Philip N, Auger M, Mattei JF, Giraud F (1988) Achondroplasia
in sibs of normal parents. J Med Genet 25:857–859
Philippart M (1990) The Rett syndrome in males. Brain Dev
Riccardi VM (1986) The Rett syndrome: genetics and the fu-
ture. Am J Med Genet Suppl 24:389–402
Schanen C, Francke U (1998) A severely affected male born
into a Rett syndrome kindred supports X-linked inheritance
and allows extension of the exclusion map. Am J Hum Genet
Schmidt M, Du Sart D (1992) Functional disomies of the X
chromosome influence the cell selection and hence the X
inactivation pattern in females with balanced X-autosome
translocations: a review of 122 cases. Am J Med Genet 42:
Seabright M (1971) A rapid banding technique for human
chromosomes. Lancet 2:971–972
Sirianni N, Naidu S, Pereira JL, Pillotto RF, Hoffman EP
(1998) Rett syndrome: confirmation of X-linked dominant
inheritance, and localization of the gene to Xq28. Am J Hum
Genet 63:1552–1558
Sumita DR, Vainzof M, Campiotto S, Cerqueira AM, Ca´novas
M, Otto PA, Passos-Bueno MR, et al (1998) Absence of
correlation between skewed X inactivation in blood and
serum creatine-kinase (CK) levels in Duchenne/Becker fe-
male carriers. Am J Med Genet 80:356–361
Thomas GH (1996) High male:female ratio of germ-line mu-
tations: an alternative explanation for postulated gestational
lethality in males in X-linked dominant disorders. Am J
Hum Genet 58:1364–1368
Topc¸u M, Topaglu H, Renda Y, Berket M, Turani G (1991)
The Rett syndrome in males. Brain Dev 13:62
Trevathan E, Naidu S (1988) The clinical recognition and
differential diagnosis of Rett syndrome. J Child Neurol
Vorsanova SG, Demidova IA, Ulas V Yu, Soloviev IV, Ka-
zantzeva LZ, Yurov Yu B (1996) Cytogenetic and molecular-
cytogenetic investigation of Rett syndrome: analysis of 31
cases. Neuroreport 8:187–189
Wahlstro¨ m J, Uller A, Tonnby B, Darnfors C, Martinsson T,
Vujuic M (1996) Congenital Rett Syndrome phenotype-
deletion short arm chromosome 3. Paper presented at the
World Congress on Rett Syndrome, Gothenburg, Sweden,
30 August–1 September
Webb T, Watkiss E (1996) A comparative study of X inacti-
vation in Rett syndrome probands and control subjects. Clin
Genet 49:189–195
Webb T, Watkiss E, Woods CG (1993) Neither uniparental
disomy nor skewed X-inactivation explains Rett syndrome.
Clin Genet 44:236–240
Wolff DJ, Schwartz S, Montgomery T, Zackowski JL (1998)
Random X inactivation in a girl with a balanced t(X;9) and
an abnormal phenotype. Am J Med Genet 77:401–404
Xiang F, Zhang Z, Clarke A, Pereira JL, Naidu S, Budden S,
Delozier-Blanchet CD, et al (1998) Chromosome mapping
of Rett syndrome: a likely candidate region on the telomere
of Xq. J Med Genet 35:297–300
Zoghbi HY, Percy AK, Schultz RJ, Fill C (1990) Patterns of
X chromosome inactivation in Rett syndrome. Brain Dev
Address for correspondence and reprints: Dr. Mayana Zatz, Centro de Estudos
do Genoma Humano, Departamento de Biologia, Instituto de Biocieˆncias,
Universidade de Sa˜o Paulo, CEP:05508-900, Sa˜o Paulo, SP, Brazil. E-mail:
q 1999 by The American Society of Human Genetics. All rights reserved.
Am. J. Hum. Genet. 64:1785–1786, 1999
Combining the Sibling Disequilibrium Test and
Transmission/Disequilibrium Test for Multiallelic
To the Editor:
Horvath and Laird (1998) describe the SDT (sibling dis-
equilibrium test) which, like the sibling-association test
(Curtis 1997), is a test for association in addition to
linkage even when applied to sibships larger than sib
pairs. These tests thus differ from the sibling transmis-
sion disequilibrium test (S-TDT [Spielman and Ewens
1998]), which is a test for linkage but not for association
(unless attention is restricted to sib pairs). The possible
advantage that the SDT has over Curtis’s test is that it
uses all affected sibs in the sibship, although it does not
allow for special provision to be made to detect a re-
cessive effect by testing whether there is an excess of
affected sibs homozygous for one particular allele. Hor-
vath and Laird demonstrate how the SDT can be applied
to a multiallelic marker and how, in the case of a biallelic
marker, the SDT and TDT can be combined, but they
do not show how the tests can be combined for a mul-
tiallelic maker. Curtis described by using logistic regres-
sion how his test could be combined with multiallelic
TDT data as implemented in the extended TDT (ETDT
[Sham and Curtis 1995]), and here we show, using their
multivariate sign test, that it is straightforward to apply
Horvath and Laird’s own approach to combine the mul-
tiallelic SDT with multiallelic TDT data.
Horvath and Laird use the 0component0 sign test
(Bickel 1965; Randles 1989) as follows. For N sibships
and a marker with m alleles, let be 1, 0, or -1, according
to whether, in the ith sibship, the frequency of allele j
in affected sibs is higher than, equal to, or lower than
that in unaffected sibs. Then define
1 2 m21
S 5 (S ,S , ) ,S )
where and a matrix W having elements
j N j
S 5 S s
i51 i
. The multiallelic SDT statistic is then
N j k
W 5 S s s
jk i51 i i
, which is asymptotically under the null
0 21 2
T 5 S W S x
hypothesis of no association or no linkage. In order to
extend this approach to include TDT data, we note that
we can apply exactly the same formula to a sample of
Page 11
1786 Letters to the Editor
N/2 trios (containing N parents) by using s
to denote,
instead, the transmission for the ith parent, being 1 or
21 if the parent has one copy of allele j and, respectively,
does or does not transmit it to the affected subject and
being 0 if the parent is uninformative for allele j (i.e.,
has 0 or 2 copies). Then the same statistic, T 5 S9W
provides a non-parametric multiallelic TDT statistic.
(This test is mathematically identical to the Stuart [1955]
test presented by Sham [1997], and is in fact the score
test of the Bradley-Terry model [Bradley and Terry
1952].) Of course it is obvious that we can sum both
forms of over a mixed sample of sibships and trios in
a combined multiallelic SDT and TDT analysis. For-
mally, if we write S
and W
for the totals derived
from the sibship data and S
and W
for those from
the trios then ,
S 5 S 1 S W 5 W 1
and is the combined
0 21
W T 5 S W S
In order to use TDT data from families with more
than one affected child, we can follow Martin et al.
(1997) and can define for the ith parent as being 1 if
the parent is heterozygous for allele j and transmits this
allele to more than half the affected children, 21 if the
allele is transmitted to fewer than half the affected chil-
dren, and 0 if the allele is transmitted to exactly half the
affected children or if the parent is uninformative for
this allele. (When there is only one affected child, this
scoring scheme is equivalent to that given above.) TDT
data can be used if only one parent is genotyped, pro-
vided that affected children homozygous for the marker
are disregarded (Curtis and Sham 1995). When all these
procedures are combined, the summations of the ap-
propriate can be performed over families consisting of
discordant sibships and consisting of one or two parents
having one or more affected children. The overall sta-
tistic provides a test, for association
0 21
T 5 S W S
with linkage, that makes appropriate use of all the avail-
able information from these different family types and
that is asymptotically .
We propose that further efforts could proceed in three
directions. First, the work of Horvath and Laird that
considers the relative power of SDT and TDT could be
extended in order to determine which is preferable to
apply to a family suitable for either. This would depend
on the transmission model of the disease and on the
numbers of parents, affected siblings, and unaffected sib-
lings who were genotyped. Second, a comparison of the
performance of the above test versus those of tests util-
izing logistic regression would be of interest. Third, the
appropriateness of the asymptotic distribution could be
investigated, since, for markers having large numbers of
alleles, it might be that a Monte Carlo approach to as-
sessment of significance could be desirable.
This work was supported by Wellcome Trust Project Grant
B. M
C. S
Academic Department of Psychological Medicine, St.
Bartholomew’s and Royal London School of Medicine
and Dentistry,
Department of Psychology, University
of Missouri and
Department of Psychological
Medicine, Institute of Psychiatry, London
Bickel PJ (1965) On some asymptotic competitors to Hotell-
ing’s T
. Ann Math Stat 36:160–173
Bradley RA, Terry ME (1952) Rank analysis of incomplete
block designs. I. The method of paired comparisons. Biom-
etrika 39:324–345
Curtis D (1997) Use of siblings as controls in case-control
association studies. Ann Hum Genet 61:319–333
Curtis D, Sham PC (1995) A note on the application of the
transmission disequilibrium test when a parent is missing.
Am J Hum Genet 1995 56:811–812
Horvath S, Laird NM (1998) A discordant-sibship test for
disequilibrium and linkage: no need for parental data. Am
J Hum Genet 63:1886–1897
Martin ER, Kaplan NL, Weir BS (1997) Tests for linkage and
association in nuclear families. Am J Hum Genet 61:
Randles RH (1989) A distribution-free multivariate sign test
based on interdirections. J Am Stat Assoc 84:1045–1050
Sham P (1997) Transmission/disequilibrium tests for multial-
lelic loci. Am J Hum Genet 61:774–778
Sham PC, Curtis D (1995) An extended transmission/disequi-
librium test (TDT) for multi-allele marker loci. Ann Hum
Genet 59:323–336
Spielman RS, Ewens WJ (1998) A sibship test for linkage in
the presence of association: the sib transmission/disequilib-
rium test. Am J Hum Genet 62:450–458
Stuart A (1955) A test of homogeneity of the marginal distri-
bution in a two-way classification. Biometrika 42:412–416
Address for correspondence and reprints: Dr. David Curtis, Department of
Adult Psychiatry, 3d Floor, Outpatient Building, Royal London Hospital, Whi-
techapel, London E1 1BB, United Kingdom. E-mail:
q 1999 by The American Society of Human Genetics. All rights reserved.
Page 12
  • Source
    • "In the very other extreme of only one marker, there will be only two different alleles and very small data sets may be enough for accurate estimators of population models, models which will also replicate in a different data set. As abovementioned, one example of a multimarker TDT is mTDT [3,4], a straightforward extension of TDT to be used with haplotypes defined as: "
    [Show abstract] [Hide abstract] ABSTRACT: Multimarker Transmission/Disequilibrium Tests (TDTs) are very robust association tests to population admixture and structure which may be used to identify susceptibility loci in genome-wide association studies. Multimarker TDTs using several markers may increase power by capturing high-degree associations. However, there is also a risk of spurious associations and power reduction due to the increase in degrees of freedom. In this study we show that associations found by tests built on simple null hypotheses are highly reproducible in a second independent data set regardless the number of markers. As a test exhibiting this feature to its maximum, we introduce the multimarker -Groups TDT (), a test which under the hypothesis of no linkage, asymptotically follows a distribution with degree of freedom regardless the number of markers. The statistic requires the division of parental haplotypes into two groups: disease susceptibility and disease protective haplotype groups. We assessed the test behavior by performing an extensive simulation study as well as a real-data study using several data sets of two complex diseases. We show that test is highly efficient and it achieves the highest power among all the tests used, even when the null hypothesis is tested in a second independent data set. Therefore, turns out to be a very promising multimarker TDT to perform genome-wide searches for disease susceptibility loci that may be used as a preprocessing step in the construction of more accurate genetic models to predict individual susceptibility to complex diseases.
    Full-text · Article · Feb 2012 · PLoS ONE
  • Source
    • "To the best of our knowledge, the consequences of using multimarker mTDT for a small number of data, which is a very common problem, have not been studied. The most widely used solution is to disregard haplotypes with a total count of less than 10 (Sham and Curtis 1995). In the present study we considered two different approaches to the problem of small numbers of data for mTDT instead of disregarding low-count haplotypes. "
    [Show abstract] [Hide abstract] ABSTRACT: Multimarker transmission/disequilibrium tests (TDTs) are powerful association and linkage tests used to perform genome-wide filtering in the search for disease susceptibility loci. In contrast to case/control studies, they have a low rate of false positives for population stratification and admixture. However, the length of a region found in association with a disease is usually very large because of linkage disequilibrium (LD). Here, we define a multimarker proportional TDT (mTDT P ) designed to improve locus specificity in complex diseases that has good power compared to the most powerful multimarker TDTs. The test is a simple generalization of a multimarker TDT in which haplotype frequencies are used to weight the effect that each haplotype has on the whole measure. Two concepts underlie the features of the metric: the ‘common disease, common variant’ hypothesis and the decrease in LD with chromosomal distance. Because of this decrease, the frequency of haplotypes in strong LD with common disease variants decreases with increasing distance from the disease susceptibility locus. Thus, our haplotype proportional test has higher locus specificity than common multimarker TDTs that assume a uniform distribution of haplotype probabilities. Because of the common variant hypothesis, risk haplotypes at a given locus are relatively frequent and a metric that weights partial results for each haplotype by its frequency will be as powerful as the most powerful multimarker TDTs. Simulations and real data sets demonstrate that the test has good power compared with the best tests but has remarkably higher locus specificity, so that the association rate decreases at a higher rate with distance from a disease susceptibility or disease protective locus.
    Full-text · Article · Sep 2010 · Human Genetics
  • Source
    • "Thus, our analyses focused on these haplotypes, rather than the individual SNPs, although TNF À308 was also assessed to enable direct comparison to the results of Boin et al. [2001]. Two types of analyses were conducted within the four sampled populations: (1) a case-control comparison of the TNF haplotypes using CLUMP [Sham and Curtis, 1995a] and (2) the SDT test [Curtis et al., 1999], both of which output appropriately calculated P-values. We chose to use the CLUMP ''T4'' statistic which compares the combined alleles over-represented in the case group with the remainder, allowing a form of w 2 -test with one-degree of freedom in which the inherently two-tailed significance is assessed using Monte–Carlo simulations (CLUMP manual). "
    [Show abstract] [Hide abstract] ABSTRACT: A single nucleotide polymorphism (TNF-308A) within the promoter region of the gene encoding tumor necrosis factor (TNF), has been significantly associated with schizophrenia in a study of Italian patients and control subjects Boin et al. [2001: Mol Psychiatry 6:79-82]. We have applied case-control analyses to examine TNF promoter haplotypes (containing TNF -308 and two additional promoter variants: TNF-376 and TNF-238) in four schizophrenia cohorts drawn from Australian, Indian Fijian, Indigenous Fijian, and Brahmin populations. In addition, we have applied the sibling transmission disequilibrium (STD) test to promoter haplotypes within 81 trios drawn from Australian Caucasian pedigrees with multiple schizophrenia cases, and 86 trios drawn from the Brahmin population of Tamil Nadu province in Southern India. Within each of these cohorts, we found no evidence of recombination between these tightly linked promoter variants, supporting previous studies which demonstrated that only a subset of the eight possible haplotypes exist. Of the four observed haplotypes, we and others have observed only one carries the TNF-308A variant allele. We report no significant differences in TNF promoter haplotype frequencies between the patient and control groups within each population, although the Indian Fijian cohort showed a trend towards reduced TNF-308A alleles amongst schizophrenia cases (P = 0.07). We found no evidence of bias in TNF promoter haplotype transmission to schizophrenia probands. Very similar results were obtained when only the TNF-308 polymorphism was considered. Taken together, these data provide no support for the involvement of TNF promoter variants TNF-308, TNF-376, and TNF-238 in schizophrenia susceptibility within four ethnically distinct cohorts.
    Full-text · Article · Aug 2003 · American Journal of Medical Genetics Part B Neuropsychiatric Genetics
Show more