Williams-Beuren Syndrome

A number sign (#) is used with this entry because Williams-Beuren syndrome (WBS) is a contiguous gene deletion syndrome resulting from the hemizygous deletion of 1.5 to 1.8 Mb on chromosome 7q11.23.

For a discussion of the genes deleted in this syndrome and possible genotype/phenotype correlations, see below.

Description

Williams-Beuren syndrome is a multisystem disorder caused by hemizygous deletion of 1.5 to 1.8 Mb on chromosome 7q11.23, which contains approximately 28 genes. Pober (2010) reviewed the clinical features of Williams-Beuren syndrome as well as the genomic and genetic basis and clinical management.

See also the distal chromosome 7q11.23 deletion syndrome (613729), which occurs between the WBS region and the MAGI2 gene (606382).

Clinical Features

Williams et al. (1961) described a syndrome characterized by supravalvular aortic stenosis (SVAS), mental retardation, and distinctive facial features. Beuren et al. (1962) described a similar syndrome with the additional features of dental anomalies and peripheral pulmonary artery stenosis. Two features of the syndrome had been described as distinct entities: supravalvular aortic stenosis (Sissman et al., 1959) and infantile hypercalcemia (Fanconi et al., 1952). Black and Bonham-Carter (1963) pointed out the similarity of the facial features described in these reports to those described by Williams et al. (1961) and Beuren et al. (1962).

Grimm and Wesselhoeft (1980) suggested that full-blown Williams syndrome includes supravalvular aortic stenosis, multiple peripheral pulmonary arterial stenoses, 'elfin face,' mental and statural deficiency, characteristic dental malformation, and infantile hypercalcemia. In a series of cases ascertained through supravalvular aortic stenosis, they found patients with mental retardation without 'elfin facies' and patients with 'elfin facies' who were mentally normal. Beuren (1972) presented compelling evidence that supravalvular aortic stenosis and idiopathic infantile hypercalcemia are part of the same disorder.

Among 19 patients with the Williams syndrome not ascertained through a cardiologic hospital, Jones and Smith (1975) found 6 without supravalvular aortic stenosis, peripheral pulmonary stenosis, or hypoplastic aorta. Oppenheimer (1938) reported a 17-month-old child with pulmonary artery stenosis and calcification of the aorta and pulmonary artery; this may have been an early case.

White et al. (1977) described second cousins with the characteristic facies and mental retardation but no documented hypercalcemia and no cardiovascular abnormality.

Preus (1984), in 2 companion articles, used numerical taxonomy (Preus, 1980) to sharpen the definition of the Williams syndrome and used the diagnostic index so derived in the differential diagnosis of the Williams and Noonan syndromes.

Burn (1986) reviewed the features of Williams syndrome and suggested that the term 'elfin facies' be dropped. He described the characteristic facial features as broad forehead, medial eyebrow flare, periorbital fullness, strabismus, stellate iris pattern, flat nasal bridge, malar flattening, full cheeks and lips, a long smooth philtrum, a pointed chin, and a wide mouth. The face becomes more coarse with age.

Preus (1975) pointed out that the iris pattern, described by her as 'lacey' and by others as 'stellate,' can be a useful diagnostic clue in infants. Holmstrom et al. (1990) had 3 ophthalmologists and 4 geneticists examine eye photographs from 43 children with Williams syndrome and 124 control subjects. A stellate pattern was noted in the irides of 51% of the Williams syndrome patients and in 12% of the control subjects. The pattern was more difficult to detect or was absent in heavily pigmented irides. Hotta et al. (1990) reported on the iris pattern in 3 cases. Winter et al. (1996) assessed the frequency and severity of ophthalmologic features in 152 patients with Williams-Beuren syndrome. Eighty-two (54%) had strabismus, while 149 had esotropia. Blue irides were present in 117 (77%), green irides in 10 (7%), and brown irides in 25 (16%). A typical stellate iris pattern of the anterior stroma was found in 112 (74%). Whitish anomalies were also detected in brown irides. Retinal vascular tortuosity was found in 22% of patients with funduscopy. Two 9-year-old patients and a 46-year-old patient had initial cataract. No ocular manifestations of hypercalcemia were noted.

Pankau et al. (1992) analyzed the statural growth in 165 patients (75 girls and 90 boys). Intrauterine growth retardation was present in 35% of the girls and 22% of the boys. Poor growth was noted during the first 2 years of life. Until age 9 years in girls and 11 years in boys, mean growth followed the 3rd percentile. A pubertal growth spurt with normal growth rate was seen at age 10 years in girls and 13 years in boys, i.e., 1 to 2 years earlier than normal. Menarche also occurred earlier than normal. Mean adult height was 153.9 +/- 6.9 cm in 17 girls and 168.2 +/- 6.9 cm in 27 boys, approximately corresponding to the 3rd percentile in both sexes. The mean deficit of adult height compared to target height was 10.2 cm in girls and 9.1 cm in boys. Skeletal development progressed at an approximately normal rate in both sexes.

Pankau et al. (1993) conducted a retrospective study of 119 patients with Williams syndrome. Results showed limitation of supination at the elbow with radioulnar synostosis in 9 patients. One patient had bilateral radioulnar synostosis. Pankau et al. (1993) suggested that radioulnar synostosis should be considered a common manifestation of the syndrome.

Patients with Williams syndrome are often described as having a harsh, brassy, or hoarse voice (Gosch et al., 1994). Stewart et al. (1993) described a patient with bilateral vocal cord paralysis, developing at the age of 9 years, which required tracheostomy. Takamatsu (1996) studied 18 cases of bilateral vocal cord paralysis in children, including 1 patient with WS. Vaux et al. (2003) described 2 WS patients who had bilateral vocal cord abnormalities (1 of whom required tracheostomy because of bilateral vocal cord paralysis), bringing to 4 the number of children with WS in whom such defects had been documented. They suggested that vocal cord abnormalities may be a far more common feature of WS than previously suspected, and that mild vocal cord dysfunction caused by abnormal vocal cord elastin may be the cause of the hoarse voice in this condition.

Narin et al. (1993) reported an 8-year-old boy with Williams syndrome who had subvalvular aortic stenosis--seemingly the first report of subvalvular location of obstruction in this disorder. Wollack et al. (1996) described a 19-year-old girl with Williams syndrome who developed an ischemic stroke of the internal capsule and putamen but who was not found to have stenotic lesion on angiography. They reviewed 5 other cases of stroke in Williams syndrome.

Cortada et al. (1980) reported the disorder in mother and both twin daughters, presumably dizygotic. One twin had supravalvular and valvular aortic stenosis. The other twin had mild peripheral pulmonary stenosis and mild coarctation of the left pulmonary artery. One twin, who died during cardiac surgery, and the mother had mitral valve prolapse. Intelligence was normal. A stellate pattern of the irides was present in both twins. All 3 had pectus excavatum, hypoplastic nails, and hallux valgus. Murphy et al. (1990) added 2 sets of concordantly affected monozygotic twins to the 2 previously reported sets. To the 5 sets of monozygotic twins with WMS previously reported, Pankau et al. (1993) added a pair concordant for the disorder but showing variable expression. Both had typical facial appearance, developmental delay, mild supravalvular aortic stenosis, hypoplasia of both pulmonary arteries, multiple peripheral pulmonary stenoses, and inguinal hernia. One twin had unilateral renal agenesis. A presumably separate disorder was cleft palate in both twins; the father, grandfather, and great-grandfather all had cleft lip with or without cleft palate.

To the 6 pairs of previously reported monozygotic twins with Williams syndrome, Castorina et al. (1997) added 2 further sets. Monozygosity was confirmed by DNA microsatellite analysis and the clinical diagnosis was confirmed by FISH using a WS-specific probe. Analysis of concordance was assisted by a long follow-up. Most clinical signs were concordant in the twins of each pair, with differences present at younger ages, mainly minor facial anomalies, being attenuated with time. Developmental delay was substantially concordant. Inguinal hernia was present in a single twin in 1 pair. Facial anomalies and other signs attributable to connective tissue abnormalities were also displayed by only 1 twin in both sets, suggesting that the WS genotype has only a predisposing role in the development of these signs.

Biesecker et al. (1987) described a 19-year-old patient with Williams syndrome who had renal cystic dysplasia and gradual deterioration of renal function, with recurrent episodes of dehydration secondary to a concentrating defect. They suggested that this is a more frequent complication than previously realized. In studies of 40 persons with Williams syndrome who were assessed at an average age of about 7 years, Pober et al. (1993) found renal abnormalities in 7: nephrocalcinosis in 2, marked asymmetry in kidney size in 2, small kidneys in 1, solitary kidney in 1, and pelvic kidney in 1. Renal artery stenosis was sought in 9 persons who underwent abdominal angiography during cardiac catheterization. Unilateral or bilateral mild renal artery narrowing was found in 4 persons and normal renal arteries in the remaining 5. Persistent hypertension was found in only 2 individuals and did not correlate with renal artery status.

Knudtzon et al. (1987) described 2 brothers with Williams syndrome who did not have hypercalcemia. One boy died during the first month of life. His brother developed severe microcephaly and cataract and died at the age of 9 years. The skeleton was osteosclerotic at birth and became osteoporotic by the age of 2 years. This brother had persistently elevated 1,25-dihydroxyvitamin D levels during the first 2 years of life, in spite of normocalcemia. At autopsy, microcalcifications were found in the brain and kidneys. Maisuls et al. (1987) described 2 patients with Williams syndrome and severe mitral regurgitation requiring surgical treatment at ages 8 and 11. Another patient had coarctation of the abdominal aorta. Hallidie-Smith and Karas (1988) described the cardiologic findings in 66 patients with the Williams-Beuren syndrome; systemic hypertension was present in 7.8% of the patients, mitral valve prolapse by clinical and echocardiographic criteria in 15%, and bicuspid aortic valve in 11.6%.

Morris et al. (1988) reviewed the natural history of Williams syndrome. After delayed growth in the first 4 years of life, catch-up growth occurred with the ultimate attainment of low-normal adult height. Older children developed progressive joint limitation and hypertonia. Hypertension was frequent in adulthood, being present in 8 of 17 adults. Morris et al. (1988) referred to the Williams Syndrome National Association, which was a source of patients for review. Morris et al. (1990) evaluated 13 adults with Williams syndrome and reviewed the case reports of 16 patients older than 16 years. Hypercalcemia may persist into adulthood. Hypertension was common. Recurrent urinary tract infections led to studies that showed urethral stenosis in some patients and bladder diverticula and vesicoureteral reflux in others. Gastrointestinal problems included chronic constipation and diverticulosis.

In 10 adults with Williams syndrome, Lopez-Rangel et al. (1992) found supravalvular aortic stenosis in 4, mitral valve prolapse in 3, bicuspid aortic valve in 1, valvular aortic stenosis in 1, and pulmonary stenosis with right ventricular hypertrophy in 1. Mental retardation was seen in all patients. Verbal skills were better developed than motor skills. All patients led active lives and most were involved in sports. Some held supervised jobs. Conway et al. (1990) reported 3 children, aged 8 years, 16 years, and 59 months, who died suddenly with myocardial ischemia following cardiac catheterization. In addition to supravalvular aortic stenosis, all showed stenosis of the left coronary artery and its branches and regions of recent and/or remote myocardial infarction. Voit et al. (1991) pointed to clinical and morphologic evidence of myopathy in this syndrome giving rise to hypotonia in infancy, delayed walking, joint contractures, scoliosis, and increased exhaustion on exertion.

Wessel et al. (1994) reported results of follow-up cardiologic examination of 59 patients with Williams syndrome. Supravalvular aortic stenosis was found in 57 patients, 17 of whom underwent surgery because of severe stenosis. Aortic hypoplasia was diagnosed in 24 patients, peripheral pulmonary stenosis in 49, and coarctation of the aorta in 4. If patients with SVAS had a pressure gradient of less than 20 mm Hg in infancy, their gradient remained unchanged for the next 20 years. If patients with SVAS had a pressure gradient of more than 20 mm Hg in infancy, their gradient increased in later life. Four of 6 patients with aortic hypoplasia and surgery for SVAS developed restenosis, whereas patients without aortic hypoplasia remained free of restenosis.

Greenberg (1990) expressed the opinion that no well-documented cases of parent-to-child transmission of classic Williams syndrome have been reported.

In 3 unrelated families, Morris et al. (1993) described Williams syndrome in parent and child: father and son in 1 family and mother and daughter in the other 2. None of the patients had supravalvular aortic stenosis or chromosomal abnormalities. In all 3 families, the parent was diagnosed after identification of the syndrome in the affected child. Sadler et al. (1993) reported Williams syndrome in mother and son. Ounap et al. (1998) reported WBS in mother and son. The diagnosis was confirmed in the son by molecular cytogenetic analysis using FISH; the mother was deceased and was thus not studied by FISH. Two traits uncommon in WBS were unilateral renal hypoplasia in the mother and a hemivertebra at L5 in the son.

Kaplan et al. (1995) pointed out that stenoses in the cerebral arteries can cause strokes with brain damage and chronic hemiparesis in children with Williams syndrome. Increased irritability, loss of consciousness, and seizures were initial signs in 2 patients. One patient, aged 22 years, had episodes of cerebral vascular insufficiency beginning at the age of 3 years at which time moyamoya was diagnosed.

Scothorn and Butler (1997) reported the case of a girl with Williams syndrome who had onset of puberty at 7.5 years of age and menarche at 8.5 years of age. They suggested that because intellectual and emotional development of children with this disorder are delayed, pharmacologic and hormonal intervention to delay puberty may be warranted to allow for intellectual and emotional maturation.

Partsch et al. (2002) reported a mean age of menarche of 11.5 +/- 1.7 years in 86 females with Williams syndrome compared with 12.9 +/- 1.1 years in a contemporary cohort of 759 girls. They estimated the prevalence of precocious puberty in Williams syndrome as 1 in 5 to 6 girls (18.3%).

Broder et al. (1999) confirmed previous findings of hypertension in Williams syndrome. They studied blood pressure using 24-hour ambulatory BP monitoring in 20 WS subjects and found that they had significantly higher ambulatory blood pressures than controls. The diagnosis of WS added approximately 10 mm Hg to mean daytime and nighttime BPs. Hypertension, defined by elevated mean daytime BP, was present in 40% of WS patients versus 14% of controls; among the children studied, this difference was even more dramatic, with 46% of WS children versus 6% of control children classified as hypertensive. Parental reporting of a history of infantile hypercalcemia was strongly associated with the presence of hypertension.

Since the elastin protein is a major component of elastic fibers in the dermis of the skin, Dridi et al. (1999) evaluated elastic fibers in the dermis of 10 Williams syndrome patients, all of whom were shown by FISH to have 7q11.23 deletions. Patients with Williams syndrome showed disorganized pre-elastic and mature elastic fibers when compared with 5 healthy children and 1 patient with isolated supravalvular aortic stenosis. The authors concluded that skin biopsies may provide a simple means to elucidate the extracellular matrix anomalies associated with Williams syndrome.

Kozel et al. (2014) specifically evaluated the dermatologic features of 94 WBS patients ranging from 7 to 50 years of age. Compared to the general population, WBS patients had higher incidences of soft skin (83%), premature graying of the hair (80% of those 20 years or older), wrinkles (92%), and abnormal scarring (33%). Biomechanical studies showed that patients had statistically significant differences in the pressure required to lift the skin, the time required to raise the skin through a prescribed gradient, viscoelasticity, and skin displacement parameters compared to controls, all consistent with easier stretching and decreased stiffness of WBS skin. The differences in skin elasticity observed in WBS patients did not correlate with the presence of vascular defects, suggesting the presence of tissue-specific modifiers that modulate the impact of elastin insufficiency resulting from haploinsufficiency of the elastin gene (ELN; 130160).

Sadler et al. (2001) did a retrospective analysis of the incidence and severity of cardiovascular disease in Williams syndrome in 127 patients. The prevalence of SVAS was 44 (35%) of 127. Statistical analysis revealed that the severity of both SVAS and total cardiovascular disease was significantly greater in male than female patients. Sadler et al. (2001) also observed that the clinical diagnosis of WS was made at a significantly younger age in male patients and that this was partly because of increased incidence and severity of cardiovascular disease. Rose et al. (2001) followed 112 patients with WS since 1975 and studied 25 of them by aortography. Twenty of 25 patients had vascular stenosis, of whom 19 were affected by segmental narrowing either of the thoracic aorta (9) or the abdominal aorta (7) or both (3). Hypoplasia of the abdominal aorta was characterized by the smallest diameters at the renal artery level and an increased diameter of the infrarenal abdominal aorta. Eleven patients had renal artery stenosis associated with narrowing of other aortic segments in 10 cases. Of 17 patients with hypertension, 2 had no vascular lesions; and in the remaining 15 patients, stenosis was present in more than 1 segment. Rose et al. (2001) concluded that hypertension is a common symptom and must be regarded as a manifestation of generalized arteriopathy rather than renal hypoperfusion.

Giannotti et al. (2001) reported a study of celiac disease (212750) in 63 Italian WS patients. The dosage of antigliadin antibodies and antiendomisium antibodies was analyzed, and 6 patients positive for these antibodies underwent small bowel biopsy. Celiac disease was present in 6 (9.5%) WS patients, compared with 1 of 184 (0.54%) Italian children (p less than 0.001). Giannotti et al. (2001) suggested screening for celiac disease in patients with WS.

In a retrospective study of 75 patients with WS, Eronen et al. (2002) found that cardiovascular symptoms were evident in 35 patients (47%) at birth. The most common abnormalities were SVAS (73%) and pulmonary artery stenosis (41%). Arterial hypertension was found in 55% of adults.

In a study of the natural history of Williams syndrome, Cherniske et al. (2004) performed multisystem assessment of 20 affected adults over 30 years of age and documented a high frequency of problems in multiple organ systems. The most consistent and striking findings were: abnormal body habitus; mild to moderate high-frequency sensorineural hearing loss; cardiovascular disease and hypertension; gastrointestinal symptoms, including diverticular disease; diabetes and abnormal glucose tolerance on standard oral glucose tolerance testing; subclinical hypothyroidism; decreased bone mineral density on dual energy x-ray absorptiometry (DEXA) scanning; and a high frequency of psychiatric symptoms, most notably anxiety, often requiring multimodal therapy. Brain MRI scans did not demonstrate consistent pathology. The adults were not living independently and the great majority were not competitively employed. One of the patients reported by Cherniske et al. (2004) had hypercalcemia, indicating that this feature is not restricted to infancy. Most of the adults had premature graying of the hair starting as early as 16 years of age, a finding that had been reported by Morris et al. (1988). This feature, together with an earlier than expected onset of cataracts and high-frequency sensorineural hearing loss, suggested mild accelerated aging, which may additionally complicate the long-term course of older adults with WS.

Marler et al. (2005) studied auditory system function in 27 William syndrome patients aged 6 to 48 years. They found sensorineural hearing loss in 14 of 18 patients aged 21 or younger. The degree of hearing loss was greater in adults than in children, suggesting early-onset, progressive hearing loss.

Gothelf et al. (2006) found that 41 (84%) of 49 patients with Williams syndrome had moderate to severe hyperacusis beginning in infancy. The most frequent sounds of daily life to which the children were sensitive included electric machines, thunder, bursting balloons, and fireworks. The children responded with marked fear and exhibited aversive behaviors. Hyperacusis peaked at age 5.7 years and tended to decrease somewhat thereafter. Quantitative testing of 21 of these patients revealed discomfort at sound intensities on average 20 dB lower than control individuals. Pure-tone audiometry and distortion product otoacoustic emission tests revealed high-frequency cochlear hearing loss. An absence of ipsilateral acoustic reflex responses to maximum stimulation was also observed. On brain auditory evoked response (BAER) testing, patients with Williams syndrome had a significant prolongation in wave I latency. Gothelf et al. (2006) noted that hearing loss in Williams syndrome resembled the configuration of noise-induced hearing loss and suggested that hyperacusis and hearing loss in Williams syndrome resulted from a deficiency in the normally protective acoustic reflex as a result of auditory nerve dysfunction.

Game et al. (2010) emphasized the urinary abnormalities in patients with WBS, including urinary frequency, urgency, nocturia, bladder diverticula, structural renal anomalies, and recurrent urinary tract infections. The authors noted that urodynamic testing has suggested evidence of detrusor overactivity and detrusor-sphincter dyssynergia in patients with WBS.

Neurodevelopmental Features

Ewart et al. (1993) commented that the IQ in patients with Williams syndrome varies from 20 to 106 (mean = 58). Specific cognitive deficits include poor visual-motor integration. As a result, affected individuals have problems visualizing a complete picture but instead see only the parts. Affected individuals also suffer from attention deficit disorder. Language development, by contrast, is relatively spared and some elements of speech may be enhanced, particularly the quantity and quality of vocabulary, auditory memory, and social use of language. Many patients sing or play musical instruments with considerable expertise and they rarely forget a name. Because of their engaging personalities, language skills, and loquaciousness, mental retardation is often underestimated in children with Williams syndrome. Gosch and Pankau (1996) used 2 methods to examine the cognitive abilities of 18 affected children (9 girls and 9 boys) with a mean age of 6.6 years at year one (T1) and approximately 2 years later (T2). The Draw A Person Test showed stable results (mean IQ of 63.5 at T1 and 65 at T2). The Columbia Mental Maturity Scale revealed a significant decrease of IQ (mean IQ of 77 at T1 and 68 at T2). Gosch and Pankau (1996) contended that this change represented a decrease of developmental rate of special abilities such as the application of classifications.

Plissart et al. (1994) studied the psychologic and behavioral characteristics of 11 adult Belgian patients, aged 17 to 66 years. Mental retardation in all patients was moderate or severe. Verbal skills were superior to visuospatial and motor abilities. The most frequent behavioral problems were poor concentration, attention-seeking behavior, and restlessness. The behavioral and emotional disturbances typical for children with Williams syndrome persisted into adulthood. Most patients achieved a good level of autonomy, with the majority living at home with parents and attending a day center.

Lenhoff et al. (1997) described the remarkable musical and verbal abilities of individuals with Williams syndrome, who perform poorly on standard IQ tests. They usually read and write poorly and struggle with simple arithmetic, but display a facility not only for spoken language but also for recognizing faces. As a group, they tend to be empathetic, loquacious, and sociable. Lenhoff et al. (1997) presented pictures, suggesting that children with Williams syndrome were an inspiration for pixie legends, and pointed out that the 'wee, magical people' of assorted folktales were often musicians and storytellers.

Gosch and Pankau (1994) compared behavioral characteristics in 19 children with Williams syndrome, aged 4 to 10 years, to those in a control group matched for age, gender, and nonverbal reasoning abilities. The children with Williams syndrome were more unreserved with and more willing to follow strangers, hypersensitive to sounds, and less socially adjusted than the control children.

Mervis et al. (1999) discussed the subject of visuospatial constructive abilities in persons with normal intelligence and in persons with Williams syndrome or small deletions in the Williams syndrome region. They reviewed behavioral genetic studies of visuospatial constructive ability, which suggested that a substantial portion of the individual differences found among people of normal intelligence has a genetic basis.

The behavioral phenotype in Williams syndrome suggests a dorsal and/or ventral developmental dissociation, with defects in dorsal but not the ventral hemispheric visual stream. A shortened extent of the dorsal central sulcus had been observed in autopsy specimens. Galaburda et al. (2001) compared gross anatomic features between the dorsal and ventral portions of the cerebral hemispheres by examining the dorsal extent of the central sulcus in MRI images from 21 subjects with WMS and age- and sex-matched control subjects. They found that the dorsal central sulcus was less likely to reach the interhemispheric fissure in subjects with WMS than in controls for both right and left hemispheres. No differences between the groups were found in the ventral extent of the central sulcus. They concluded that early neurodevelopmental problems affect the development of the dorsal forebrain and are probably related to the deficits in visuospatial ability and behavioral timing often observed in Williams syndrome.

Schmitt et al. (2001) performed brain MRI on 20 patients with Williams syndrome to determine how cerebral shape differs from that of normal controls. In Williams syndrome, both cerebral hemispheres and the corpus callosum bend to a lesser degree in the sagittal plane, which the authors believed to be due to variation in the parietooccipital region. In addition, the cerebral hemispheres and corpus callosum midline lengths were decreased in Williams syndrome. Schmitt et al. (2001) suggested that the brain findings are consistent with aberrant premature termination of brain development, which proceeds normally in the rostrocaudal direction.

Lenhoff et al. (2001) evaluated 5 patients with Williams syndrome for absolute musical pitch (AP; see 159300), which is the ability to recognize, name, and reproduce the pitch of a musical note without reference. The 5 patients had a mean IQ of 58 but were able to read musical notation. They began to play music at ages 5, 7, 8, 10, and 11 years, respectively. As a group, the 5 patients scored 97.5% on 1,084 absolute pitch trials, indicating that they possessed exceptional abilities in absolute pitch. By comparison, cognitively intact musicians who claim to have AP scored 84.3% on similar tests. Lenhoff et al. (2001) suggested that the prevalence of AP in individuals with Williams syndrome is higher than that in the general Western population (1 in 10,000) and noted that the age window of AP acquisition in Williams syndrome appears to be extended compared to the general population. Hickok et al. (1995) reported that brain imaging of patients with Williams syndrome suggested an exaggerated left-right asymmetry of the planum temporale, which had also been found in musicians with absolute pitch (Schlaug et al., 1995), suggesting a neuroanatomical correlate to the ability.

Patients with Williams syndrome have relatively good abilities in face recognition and discrimination. Using functional MRI to assess facial recognition, Mobbs et al. (2004) found that 11 patients with WS showed increased activation in the right fusiform gyrus and several frontal and temporal regions, including subcortical structures. By contrast, control individuals showed greater activation in the primary and secondary visual cortices. The findings suggested that patients with WS have impairments in the visual cortical regions and use frontal and temporal regions as a compensatory mechanism.

Primate visual cortex is organized into 2 functionally specialized, hierarchically organized processing pathways: a ventral stream for object processing and a dorsal stream for spatial processing. Patients with Williams syndrome show a visuospatial constructive deficit, which is an inability to visualize an object as a set of parts or to construct a replica. Using multimodal neuroimaging techniques, Meyer-Lindenberg et al. (2004) found that 13 high-functioning individuals with WS showed significant hypoactivation in dorsal stream areas during different visual tasks compared to controls. No differences were found in the ventral stream. Structural imaging studies showed that individuals with WS had gray matter volume reduction in the parietooccipital/intraparietal sulcus, immediately adjacent to the region of hypofunction, suggesting a structural-functional connection.

Meyer-Lindenberg et al. (2005) used multimodal neuroimaging to characterize hippocampal structure, function, and metabolic integrity in 12 normal-intelligence patients with Williams syndrome and 12 age-, sex-, and IQ-matched healthy controls. PET and functional MRI studies showed profound reduction in resting blood flow and absent differential response to visual stimuli in the anterior hippocampal formation in patients with Williams syndrome. Spectroscopic measures of N-acetylaspartate, a marker of synaptic activity, were reduced. Hippocampal size was preserved, but subtle alterations in shape were present. Meyer-Lindenberg et al. (2005) suggested that hippocampal dysfunction might contribute to neurocognitive abnormalities in Williams syndrome.

Castelo-Branco et al. (2007) presented evidence of a neural defect in the retina of WBS patients. High-resolution imaging techniques found that WBS patients had decreased retinal thickness, abnormal optic disc concavity, and impaired visual responses compared to controls. Low-level magnocellular performance was independent of deficits in the integration of information at higher levels.

Marenco et al. (2007) performed brain diffusion tensor MRI to assess white matter integrity in 5 high-functioning WBS patients. Patients showed significant differences in white matter tissue organization compared to controls, particularly with respect to alterations in the main orientation of fibers underlying abnormalities in the gray matter. There appeared to be an increase in anterior-posterior longitudinal fibers and a reduction in right-to-left transverse axis fibers in the patients, consistent with the finding of other midline defects, such as dysgenesis of the corpus callosum. Marenco et al. (2007) hypothesized that there is specific alteration in the development of U fibers in the later stages of neuronal migration in patients with WBS and suggested that these abnormal patterns result from deletions of genes within the critical region.

Atypical Williams-Beuren Syndrome

Morimoto et al. (2003) reported a Japanese male with a severe form of WBS associated with craniosynostosis and refractory infantile seizures. At age 5 months, he was diagnosed with peripheral pulmonary stenosis and mild ventricular hypertrophy. The seizures responded to ACTH, which had to be discontinued due to progression of the cardiac hypertrophy. EEG showed a variant of hypsarrhythmia. He also had an elfin face, failure-to-thrive, severe developmental delay, and dental malformation, in addition to congenital heart defects. FISH showed deletion of the elastin gene, and high-resolution chromosome analysis revealed interstitial deletion of 7q11.22-q11.23, consistent with Williams syndrome. At 2 years, his seizures were controlled, but his psychomotor development was severely delayed. Treatment with thyrotropin-releasing hormone (TRH) offered improvement in seizure control. Marshall et al. (2008) noted that seizures are not common in WBS. Using high resolution mapping to reexamine the patient reported by Morimoto et al. (2003), Marshall et al. (2008) found that the deletion was 4.4-Mb in length and extended telomeric to the classic WBS region. The deletion included the YWHAG gene (605356) but did not include the MAGI2 gene (606382); see the distal 7q11.23 deletion syndrome (613729).

Tassabehji et al. (2005) identified an atypical Williams-Beuren syndrome individual with a smaller genetic deletion relative to classic Williams-Beuren syndrome cases but including 2 extratelomeric genes, CYLN2 (603432) and GTF2IRD1 (604318). The patient was a 4.5-year-old girl with surgically corrected pulmonary artery stenosis. Her birth weight and growth appeared normal, and at 4.5 years her height was just above the 50th centile. Facial features were suggestive of but not classic for Williams-Beuren syndrome. Early developmental milestones such as sitting and walking were within normal limits; however, by 18 months she had a vocabulary of only a few single words and by age 4 she continued to show a delay in language acquisition as well as serious deficits in spatial cognition, but to a lesser degree than that seen in Williams-Beuren syndrome patients.

Biochemical Features

In a study of patients with Williams syndrome, Rae et al. (1998) found a correlation between performance on neuropsychologic tests and decreases in the amount of neocerebellar N-acetylaspartate when normalized to choline or creatine. They speculated that this could either reflect a global decrease of this neuronal marker in the entire brain, or perhaps evidence of cerebellar involvement.

Grimm and Wesselhoeft (1980) pointed out that supravalvular aortic stenosis has been described as a rare feature of the Marfan syndrome and occurs as a phenocopy of the genetic disorder induced by rubella embryopathy (Varghese et al., 1969), by experimental vitamin (Jorgensen, 1972). Taylor et al. (1982) investigated the effects of pharmacologic doses of vitamin D2 given for 4 days to normal children and to children with Williams disease and their sibs. The results indicated an exaggerated increase in serum 25-OH-D in response to challenge with vitamin D in patients with the Williams syndrome and in some of their sibs with no clinical features of the syndrome. Despite the increases in serum 25-OH-D, none of the patients became hypercalcemic. Garabedian et al. (1985) found high plasma concentrations of 1,25-(OH)2D in 4 children with hypercalcemia and 'elfin facies.' The levels were higher than in 3 children with 'elfin facies' but without hypercalcemia or dysmorphia. In Williams syndrome, a low calcium diet controlled the hypercalcemia. They suggested that an abnormal synthesis or degradation of 1,25-(OH)2D is present in this syndrome. Others (e.g., Martin et al., 1985) questioned this work. From a study of calcium metabolism in 27 normocalcemic children and adults, aged 2 to 47 years, with WBS, Kruse et al. (1992) concluded that neither deficient calcitonin secretion nor increased renal sensitivity to parathyroid hormone is a feature in normocalcemic patients. Furthermore, they did not find a significant disturbance in vitamin D metabolism.

Cherniske et al. (2004) reported studies of 20 adults with Williams syndrome (age range 30 to 51 years) in which they observed a 25% prevalence of elevation of serum thyroid-stimulating hormone (TSH; see 188540) concentration. Stagi et al. (2005) analyzed thyroid function and morphology in another 20 patients with Williams syndrome (age range 1.7 to 34.9 years). They likewise found that 25% of the patients showed a TSH elevation; they related this finding to the hypoplasia of the thyroid gland which was evident in about 70% of their patients. Selicorni et al. (2006) reported the results of a morphologic and functional study of the thyroid gland in 95 patients with WS, who periodically underwent a complete survey to detect early complications related to the condition. The study confirmed the increased incidence of both elevated TSH serum values (37.9%) and thyroid gland hypoplasia (74.7%). Moreover, they demonstrated that TSH elevation declined with age.

Population Genetics

Grimm and Wesselhoeft (1980) estimated the frequency of Williams syndrome to be 1 in 10,000.

Stromme et al. (2002) estimated that the Williams-Beuren syndrome occurs at a frequency of approximately 1 in 7,500 live births, with approximately two-thirds of the deletion events being intrachromosomal.

Clinical Management

The Committee on Genetics American Academy of Pediatrics (2001) published a set of guidelines to assist in the health care supervision of children with Williams syndrome.

Cytogenetics

Osborne (1999) reviewed Williams-Beuren syndrome, including the phenotype and the genes that have been identified as mapping within the WBS common deletion. They discussed the mechanism of deletion and the correlation between extent of deletion and phenotype.

Perez Jurado et al. (1996) investigated the deletion size and frequency on chromosome 7q11.23, determined the parental origin, and correlated the molecular results with the clinical findings in 65 patients with Williams syndrome. They carried out genotyping of WS patients and available parents for 13 polymorphisms and determined that 94% of patients had a deletion of the ELN (130160) locus. Analysis of polymorphic markers suggested that the commonly deleted region extended from D7S489B through D7S1870. The D7S489B locus was deleted in all informative patients. No variability in the size of the deletion was detected in the WS patients by genotyping of polymorphic markers. The investigators were able to visualize the common deletion in WS, estimated to be 1.5-2.5 Mb. The D7S489B locus constitutes a lower-copy repeat with at least 2 copies which map close to the WS deletion breakpoints. Perez Jurado et al. (1996) proposed that these repeats may provide a mechanism for aberrant recombination or replication events. All 4 patients with normal dosage at the ELN locus had biparental inheritance at all informative loci tested. Perez Jurado et al. (1996) noted that clinical reevaluation of these 4 patients was consistent with a diagnosis of WS based on the presence, during some period of development, of characteristic facial features, mental retardation, and strongly suggestive cognitive and personality profiles. They noted that 3 of the 4 patients were above the 50th centile for height and head circumference. None of them had hypercalcemia or vascular stenoses. Perez Jurado et al. (1996) reported that in 39 families informative for parental origin, all deletions were de novo and 18 were paternally and 21 maternally derived. They noted that comparison of clinical data collected in a standardized quantifiable format revealed more severe growth retardation and microcephaly in the maternal deletion group. Perez Jurado et al. (1996) proposed that an imprinted locus, silent on the paternal chromosome and contributing to statural growth, may be affected by the deletion.

Dutly and Schinzel (1996) carried out molecular genetic studies in 15 families with WBS. They demonstrated deletion of the ELN gene in all of the probands. The 15 families consisting of patients, parents, and paternal or maternal grandparents were genotyped using microsatellites adjacent to the centromeric or telomeric end of ELN. They demonstrated that in 10 out of 15 WBS families (67%) with a de novo deletion within 7q11.23, the segment flanking the deleted region contained recombined haplotypes. These recombination events indicated that deletion was the result of an unequal crossing-over event between the chromosome 7 homologs during gametogenesis. In 5 of the 15 families there was no recombination on either side of the deletion. Dutly and Schinzel (1996) postulated that in these families there may be intrachromosomal recombination. They noted that unequal recombination events are mediated by related gene sequences or repetitive elements and that the elastin gene has relatively large introns characterized by repetitive elements. Frangiskakis et al. (1996) reported that breakpoints in the LIM kinase-1 gene (LIMK1; 601329), which is adjacent to ELN, occur within Alu repeats. Dutly and Schinzel (1996) concluded that a practical consequence of their findings is improved prediction of recurrence risks for sibs of a WBS-affected proband since a recombination event around the deleted segment indicates meiotic recombination which is unlikely to recur.

Pankau et al. (2001) reported 2 families in which girls had inherited Williams-Beuren syndrome from their mothers. In all 4 patients the clinical diagnosis was supported by the molecular cytogenetic detection of a hemizygous deletion at 7q11.23. Considerable variation in the clinical manifestations of the syndrome within and between these families was noted.

Deletion at the ELN Gene Locus

Loss of an ELN (130160) allele produces the cardiovascular pathology of Williams-Beuren syndrome (review by Pober, 2010).

In studies in 4 familial and 5 sporadic cases of Williams syndrome, Ewart et al. (1993) identified hemizygosity at the elastin locus (ELN) resulting from deletion. Loss of heterozygosity for DNA markers was the first clue; fluorescence in situ hybridization and quantitative Southern analysis confirmed this finding. The neurobehavioral features of Williams syndrome described earlier are not easily explained by hemizygosity at the ELN locus. The linkage and physical mapping data of Ewart et al. (1993) suggested that the deletions associated with Williams syndrome extend beyond the ELN (130160) locus, spanning at least 114 kb.

Taking advantage of a large series (27 cases) of sporadic Williams syndrome, Gilbert-Dussardier et al. (1995) explored the potential application of novel microsatellite DNA markers in the rapid detection of hemizygosity in WBS. They found that a highly informative marker at locus D7S1870 could detect failure of parental inheritance in almost 75% of cases in their series.

Nickerson et al. (1995) investigated the frequency of deletions of the ELN gene in patients with Williams syndrome using both fluorescence in situ hybridization (FISH) and PCR amplification of a dinucleotide repeat polymorphism. In 40 of the 44 patients tested (91%), FISH demonstrated deletion of the ELN gene. Using the DNA polymorphism, both maternally (39%) and paternally (61%), derived deletions were found. Thus, FISH analysis proved a rapid and informative test to confirm a clinical diagnosis of Williams syndrome. However, the presence of 2 copies of the ELN locus in a patient does not rule out the diagnosis.

In a series of 235 patients, Lowery et al. (1995) identified molecular cytogenetic deletions by FISH in 96% of patients with classic WBS. Patients included 195 solicited through the Williams Syndrome Association, plus 40 clinical cytogenetics cases referred by primary-care physicians. On the basis of photographs and medical records of most subjects from the Association, 114 of the patients were identified as 'classic' and 39 'uncertain.' Whereas 96% of the classic WMS subjects showed deletion, only 3 of 39 of the uncertain patients showed a deletion. Of the 42 who were not classified phenotypically, because of lack of clinical information, 25 patients (60%) showed a deletion. In 15 of 40 (38%) of clinical cytogenetics cases, they found an ELN deletion and no cytogenetic deletion by banded analysis. Results supported the usefulness of FISH for the detection of elastin deletions as an initial diagnostic assay for WMS.

Mari et al. (1995) used an intragenic RFLP and gene dosage of the elastin gene with a new probe to analyze 60 sporadic cases with the clinical diagnosis of Williams syndrome. Deletion of the ELN gene was shown in 54 cases; clinical reevaluation of the 6 patients without demonstrable deletion did not confirm the diagnosis of WBS. The results supported the genetic homogeneity of WBS and the high accuracy of ELN molecular analysis. By using FISH as a diagnostic tool, Borg et al. (1995) demonstrated hemizygosity at the ELN locus in all 5 cases considered to be classic Williams syndrome and in 3 of 5 atypical cases. Prominence of the thyroid cartilage and thinning of the cheeks (with loss of jowls) occurred with advancing age. A friendly disposition was found in all patients with the microdeletion, but the degree of loquacity decreased as the severity of mental retardation increased. Hyperacusis was also a constant feature. Hypercalcemia was documented in only 2 of the patients with submicroscopic microdeletion but surprisingly was documented in both patients lacking the chromosomal abnormality. By FISH, Brewer et al. (1996) found hemizygosity for an ELN gene probe in all of 16 children