Article Text

PDF

A case of Williams syndrome with a large, visible cytogenetic deletion
  1. YUAN-QING WU,
  2. ELIZABETH NICKERSON,
  3. LISA G SHAFFER
  1. Department of Molecular and Human Genetics
  2. Baylor College of Medicine, One Baylor Plaza
  3. Room 15E, Houston, TX 77030, USA
  4. Department of Pediatrics, Division of Medical Genetics
  5. University of Iowa Hospitals & Clinics
  6. Iowa City, IA, USA
  1. Dr Shaffer
  1. KIM KEPPLER-NOREUIL,
  2. ANN MUILENBURG
  1. Department of Molecular and Human Genetics
  2. Baylor College of Medicine, One Baylor Plaza
  3. Room 15E, Houston, TX 77030, USA
  4. Department of Pediatrics, Division of Medical Genetics
  5. University of Iowa Hospitals & Clinics
  6. Iowa City, IA, USA
  1. Dr Shaffer

Statistics from Altmetric.com

Editor—Williams syndrome (WS) is generally characterised by mental deficiency, gregarious personality, dysmorphic facies, supravalvular aortic stenosis (SVAS), and idiopathic infantile hypercalcaemia. Patients with WS show allelic loss ofSTX1A,1 elastin (ELN),2 3 andLIMK1,4 with most exhibiting a submicroscopic deletion at 7q11.23, detectable by FISH.3 5 The common deletion size is about 1.5 Mb.6 Previous studies have shown that WS patients have consistent deletion sizes and share common proximal and distal breakpoints.7 8 Here we report a patient who has a large, atypical, visible chromosomal deletion of 7q11.2 and features consistent with, and in addition to, those typically seen in Williams syndrome.

The patient was originally referred to the genetics clinic at 5 months of age for evaluation of global developmental delay and dysmorphic features. She was delivered at 37 weeks' gestation by caesarean section weighing 2350g (<5th centile). The initial course included a history of poor feeding in the newborn period. Clinical examination showed macrocephaly, cutaneous haemangioma, and craniofacial features consisting of a large anterior fontanelle, frontal bossing, depressed nasal bridge, cup shaped ears, hypertelorism, and prominent lips (fig1A). Neurological examination showed generalised hypotonia with heel cord and hamstring tightness. CT scan of the head and renal ultrasound were normal. Because of a grade III/VI systolic murmur, echocardiogram was performed, which showed a slightly thickened aortic valve. Cytogenetic analysis showed a 46,XX karyotype.

Figure 1

Front view of the patient at 22 months (A) and 6½ years of age (B). The child has characteristic Williams syndrome facies including frontal bossing, prominent supraorbital ridging, periorbital fullness, stellate pattern to her irides, cup shaped ears in normal position, prominent, full lips, long philtrum, and a broad nose with anteverted nostrils.

Re-evaluation at 4 years of age, showed short stature (90 cm, <3rd centile), continued significant developmental delay, and coarsened facial features with stellate irides. Repeat echocardiogram showed moderately severe supravalvular aortic stenosis. Cardiac catheterisation confirmed these findings without involvement of the leaflets or the rest of the ascending aorta, and no evidence of aortic insufficiency. Ophthalmological examination showed bilateral exotropia and hyperopia requiring corrective lenses and very grey optic discs with surrounding peripapillary retinal pigment epithelial changes. Serum calcium levels were raised once (10.8 mg/dl, normal range 8.5-10.5 mg/dl), but have since been in the normal range. FISH analysis for the elastin locus on chromosome 7q11.23 showed a deletion consistent with Williams syndrome3 and repeat cytogenetic analysis showed a visible deletion in band 7q11.2 (46,XX,del(7)(q11.1q11.23)) (fig 2).

Figure 2

GTG banded partial karyotype of the patient. The normal chromosome 7 is shown on the left and the deleted chromosome 7 is shown on the right (arrow).

Follow up at 6½ years of age showed weight 19.8 kg (40th centile), height 107 cm (5th centile), and head circumference 52.5 cm (70th centile), with other notable findings that included prominent supraorbital ridging, periorbital fullness, stellate pattern to her irides, cup shaped ears in normal position, hyperplastic gums, prominent, full lips, long philtrum, and broad nose with anteverted nostrils (fig 1B). She had fifth finger clinodactyly and brachydactyly. Examination of her skin showed a haemangioma in the midline lumbosacral region, which had reportedly once extended from her occiput to her buttocks. She had a hoarse, raspy voice and frequent drooling. Developmentally, she functioned in the severe mental retardation range. She showed significant delays in communication; expressive language was severely delayed with rare speech (a few words) and limited sign language and receptive language at <1 year of age. She had atypical behaviour including diminished interest in social interaction with others, self-injurious behaviour, intermittent stereotypic behaviour, and sleep disturbance.

The patient had a sister (aged 8 years) and a brother (aged 10 years) who were healthy with normal development. There were no other family members with mental retardation, short stature, or birth defects. The parents were non-consanguineous. Blood samples were obtained from both parents and the proband for additional molecular studies.

In order to identify a FISH probe for the gene encoding the α2/δ subunits of the L type voltage dependent skeletal muscle calcium channel (CACNL2A), a PCR primer set (exnCa-A: 5′-CGGTGAGTGCTAAGACCTGAATG-3′, exnCa-B: 5′-CAGCCCTCATAGATGTCAGTAGG-3′) was designed from exon sequence obtained from the EMBL database under accession number Z28599. Primers were used at a final concentration of 1.5 μmol/l in a 20 μl reaction. The amplification was performed with an annealing temperature of 60°C for 30 cycles. Total human DNA (25 ng) was used as a template. The resulting 311 bp product was electrophoresed on a 1% low melting point agarose gel. The gel fragment was excised and diluted 1:3 with sterile water and was labelled with 32P-dCTP by standard methodology. The probe was hybridised to high density filters arrayed with clones from a human total genomic P1 library. Four positive coordinates were submitted to the Baylor College of Medicine YAC core and streaks, representing 12 P1 clones per coordinate, were received. For each, single colonies were tested by PCR using the exnCa primer pair. Clones from coordinate 101F1 were found to be positive for the primer sequence. The positive clone was grown and DNA was prepared using the Qiagen Plasmid Midi Kit following the amended instructions for P1 clones distributed by Qiagen. A cosmid size FISH probe was made by subcloning the P1 clone using the SuperCos 1 Cosmid Vector Kit available from Stratagene. A clone (CaE-1) specific to the exnCa primer pair was isolated by testing subclones by PCR as described above.

A cosmid containing the D7S849 locus, known to be linked to theCACNL2A locus,9 but not contained in the previous P1 clone, was subcloned from a yeast artificial chromosome (YAC) isolated from the CEPH Mark I YAC library by the Baylor College of Medicine YAC core using the D7S849 primer set. YAC DNA was prepared by a standard caesium chloride protocol and subcloned using the SuperCos I Cosmid Vector Kit. Human clones were identified by hybridisation of colony lifts with 25 ng of total human DNA radiolabelled with 32P. Human clones were then screened by PCR with D7S849. A single clone, positive for D7S849, was identified (CaE-2) and DNA was isolated using a Qiagen Plasmid Midi Kit and was used (20 ng/μl) as a FISH probe against patient metaphase chromosomes.

Additional probes used in the FISH analyses included a cosmid containing the 5′ end of the elastin gene (cELN272),3 7 a cosmid containing the full sequence ofLIMK1,4 7 and a bacterial artificial chromosome clone (BAC) containing theSTX1A gene (BAC 137N23; Research Genetics, Huntsville, AL).

All FISH probes were labelled by nick translation with digoxigenin-dUTP and detected with anti-digoxigenin conjugated to rhodamine. Either a biotin labelled chromosome 7 alpha satellite centromere probe or a digoxigenin labelled chromosome 7q telomere probe (Oncor, Inc, Gaithersburg, MD) was used as a control to identify the chromosomes 7. The centromere probe was detected using avidin conjugated to fluorescein isothiocyanate (FITC). Slides were counterstained with DAPI. FISH analyses were performed as recently described.7

This patient was deleted for all the FISH probes tested, including theCACNL2A gene and the locus D7S849 (fig3).

Figure 3

Fluorescence in situ hybridisation using cosmid CaE-1 to the CACNL2A locus. A 7q telomeric probe is used as a control. Two signals are seen on the normal chromosome 7. Only the control signal is present on the deleted chromosome 7 (arrow), indicating a deletion of the CACNL2A locus.

DNA was extracted from peripheral blood from the patient and each parent using standard methodology. Polymorphic dinucleotide repeat markers were used to detect deletions and determine the parental origin of the deletion as previously described.3 7 A deletion was evident when the proband failed to inherit an allele from one of the parents. The following loci were examined (listed centromeric to telomeric): D7S672, D7S1816, D7S489U, D7S2476,ELN, LIMK1, D7S613, D7S2472, D7S1870, D7S489L, D7S849, D7S675, D7S699, D7S440, and D7S634.

The patient was deleted for markers D7S489U (centromeric) to D7S440 (telomeric) and uninformative for D7S634. The centromeric breakpoint was the same as seen in classical WS patients.7 The patient's distal deletion breakpoint was telomeric to the classical breakpoint (D7S1870), beyond the marker D7S440, with the exact distal breakpoint undetermined.

WS presents as a remarkable collection of features with significant phenotypic variability among patients. Variability in the phenotype could be the result of different sized deletions aroundELN or the variation in gene content or gene activity of the hemizygous alleles on the non-deleted chromosome. Our previous studies have shown the size of the deletions in the majority of WS patients studied to be consistent between the markers D7S489U and D7S1870.7 The current patient represents a rare exception.

The gene encoding the α2/δ subunits of the L type voltage dependent skeletal muscle calcium channel (CACNL2A) was mapped to 7q21-q22.10 In addition, a form of malignant hyperthermia susceptibility (MHS) has been linked toCACNL2A by analysis of a (CA)nrepeat polymorphism, D7S849, mapping to 7q11.23-q21.1.9Given the occurrence of hypercalcaemia and the reports of masseter spasm and sudden death during surgical procedures in WS patients11 and the mapping ofCACNL2A near the WS critical region on chromosome 7, Mammi et al 12investigated the inclusion of this gene in microdeletions in WS. Although CACNL2A was excluded from the common critical region in the WS patients studied,12 this locus was deleted in the current case of a visible deletion of 7q11.2. It is not known if this patient would be susceptible to malignant hyperthermia, but caution should be exercised if anaesthesia is ever necessary.

To date, this is the only case of a visible deletion of 7q11.2 in which the extent of the deletion has been delineated by molecular methods. The present deletion appears to extend distally beyond D7S849 and CACNL2A. This patient has the characteristic craniofacial and cardiovascular abnormalities, including supravalvular aortic stenosis, described in patients with Williams syndrome. By the scoring system developed by Preus,13 this patient's score was +9.75, well within the Williams syndrome range (+12.59 ± 4.18). She has typical complications reported in patients with Williams syndrome, such as esotropia, hyperopia, enamel hypoplasia, microdontia, early hypercalcaemia, chronic otitis, short stature, and feeding problems. She has not developed scoliosis, kyphosis, or contractures. In addition, she has some atypical findings not seen in Williams syndrome including macrocephaly, retinal problems, severe mental retardation, and minimal speech. She also has had a history of seizures (petit mal type) not often described in patients with Williams syndrome. Her additional features are probably the result of hemizygosity for genes outside the classical WS deletion region.

In examining previously published reviews of cases of 7q deletion, there is considerable clinical variation.14-19 In an attempt to determine if specific phenotypic features are associated with proximal or distal deletions of 7q, deletions of 7q were grouped into terminal deletions14-17 and interstitial deletions.20-22 Interstitial deletions of 7q have been divided into three categories based on the region involved: (1) cen→q21/q22 (proximal), (2) q21→q31/32 (intermediate), and (3) q32→q34 (distal).17 Although correlations between phenotypes and deletions are difficult to establish owing to the variable breakpoints, there are many common but non-specific findings reported in patients with proximal deletions or rearrangements of 7q, including low birth weight, mental retardation, microcephaly, growth retardation, early feeding problems in infancy, abnormal EEG/seizures, hypotonia, and abnormal skull shape.23-25 Comparison of the phenotypes between WS and proximal deletion of 7q shows many common features, such as mental retardation, developmental delay, growth retardation, low birth weight, cardiovascular defects, eye abnormalities, facial dysmorphism and clinodactyly. However, the unique cognitive profile seen in WS patients, significant deficits in motor skills and impaired visuospatial recognition,26 27loquaciousness, sociability, weak adaptive skills, dependency, hyperactivity, distractibility, inattention, and limited perseverance,27-31 is not seen in patients with large deletions, including our patient, perhaps because of more severe mental retardation. Additionally, seizures are common among visible deletion patients, but not generally seen as a feature in WS. This finding suggests that the genes responsible for seizures are outside the common WS deletion. The proximal breakpoint of this case is the same as the proximal breakpoint of the critical deletion region of Williams syndrome, indicating, perhaps, a common mechanism for the deletion in this case and the classical deletion.32 33

Acknowledgments

The authors thank the family for their participation, Dr M Keating (University of Utah) for the elastin containing cosmid, and Dr M Tassabehji (St Mary's Hospital, UK) for the LIMK1 containing cosmid. This research was supported in part by an American Heart Association grant in aid (LGS) and NIH grant RO3 HD 35112 (LGS).

References

View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.