Digeorge Syndrome

A number sign (#) is used with this entry because DiGeorge syndrome is caused by a 1.5- to 3.0-Mb heterozygous deletion of chromosome 22q11.2. Haploinsufficiency of the TBX1 gene (602054) in particular is responsible for most of the physical malformations. There is evidence that point mutations in the TBX1 gene can also cause the disorder.

Description

DiGeorge syndrome (DGS) comprises hypocalcemia arising from parathyroid hypoplasia, thymic hypoplasia, and outflow tract defects of the heart. Disturbance of cervical neural crest migration into the derivatives of the pharyngeal arches and pouches can account for the phenotype. Most cases result from a deletion of chromosome 22q11.2 (the DiGeorge syndrome chromosome region, or DGCR). Several genes are lost including the putative transcription factor TUPLE1 which is expressed in the appropriate distribution. This deletion may present with a variety of phenotypes: Shprintzen, or velocardiofacial, syndrome (VCFS; 192430); conotruncal anomaly face (or Takao syndrome); and isolated outflow tract defects of the heart including tetralogy of Fallot, truncus arteriosus, and interrupted aortic arch. A collective acronym CATCH22 has been proposed for these differing presentations. A small number of cases of DGS have defects in other chromosomes, notably 10p13 (see 601362). In the mouse, a transgenic Hox A3 (Hox 1.5) knockout produces a phenotype similar to DGS as do the teratogens retinoic acid and alcohol.

Nomenclature

DiGeorge syndrome overlaps clinically with the disorder described by the Japanese as 'conotruncal anomaly face syndrome' (Kinouchi et al., 1976; Takao et al., 1980; Shimizu et al., 1984), where the cardiovascular presentation is the focus of attention. The term conotruncal anomaly face syndrome is cumbersome and has the disadvantage of using embryologic assumptions as a title. It would be appropriate to use Takao syndrome for those cases with a preponderant cardiac presentation in contrast to the low T cell and hypocalcemic presentation in infancy of DiGeorge syndrome and the craniofacial and palatal abnormalities typical of Shprintzen syndrome. These 3 phenotypes may be seen in the same family and most cases of all 3 categories have been shown to have a 22q11 deletion. This led Wilson et al. (1993) to propose the acronym CATCH22 (Cardiac Abnormality/abnormal facies, T cell deficit due to thymic hypoplasia, Cleft palate, Hypocalcemia due to hypoparathyroidism resulting from 22q11 deletion) as a collective acronym for those with the common genetic etiology. Shprintzen (1994) objected to 'lumping' velocardiofacial syndrome with the DiGeorge anomaly, arguing that there is 'no valid evidence to suggest that velocardiofacial syndrome is etiologically heterogeneous...[whereas] the DiGeorge anomaly is known to be so.' Hall (1993) cited data of Driscoll et al. (1993) indicating that velocardiofacial syndrome is etiologically heterogeneous. She stated that '...68% of Shprintzen syndrome patients...have been recognised to have deletions of 22q11.' Shprintzen (1994) refuted her statement, maintaining that it could accurately be stated that deletion was found in 68% of patients sent to the Driscoll laboratory with a diagnosis of velocardiofacial syndrome made by other clinicians. Shprintzen (1994) said that in his sample, 100% had deletion.

Burn (1999), one of the original proposers of the acronym CATCH22, reviewed the discussion of nomenclature. He recognized that the term CATCH22 had a number of negative connotations and that in practice different terms were in use for this phenotype and would continue to be so. Burn (1999) proposed that the term DiGeorge syndrome be reserved for those with neonatal presentation, particularly with thymic hypoplasia and hypocalcemia, and that the designation VCFS be used for children with a presentation dominated by nasal speech due to palatal insufficiency. He also suggested that 'conotruncal anomaly face' be replaced by 'Takao syndrome' and pointed out that the term '22q11 deletion syndrome' was reasonable. Finally, Burn (1999) proposed that 'CATCH phenotype' be used rather than CATCH22 and that the acronym be taken to represent cardiac abnormality, T cell deficit, clefting, and hypocalcemia.

Clinical Features

DiGeorge syndrome is characterized by neonatal hypocalcemia, which may present as tetany or seizures, due to hypoplasia of the parathyroid glands, and susceptibility to infection due to a deficit of T cells. The immune deficit is caused by hypoplasia or aplasia of the thymus gland. A variety of cardiac malformations are seen in particular affecting the outflow tract. These include tetralogy of Fallot, type B interrupted aortic arch, truncus arteriosus, right aortic arch and aberrant right subclavian artery. In infancy, micrognathia may be present. The ears are typically low set and deficient in the vertical diameter with abnormal folding of the pinna. Telecanthus with short palpebral fissures is seen. Both upward and downward slanting eyes have been described. The philtrum is short and the mouth relatively small. In the older child the features overlap Shprintzen syndrome (velocardiofacial syndrome) with a rather bulbous nose and square nasal tip and hypernasal speech associated with submucous or overt palatal clefting. Cases presenting later tend to have a milder spectrum of cardiac defect with ventricular septal defect being common.

Short stature and variable mild to moderate learning difficulties are common. A variety of psychiatric disorders have been described in a small proportion of adult cases of velocardiofacial syndrome. These have included paranoid schizophrenia and major depressive illness. Clinical features seen more rarely include hypothyroidism, cleft lip, and deafness.

Goodship et al. (1995) described monozygotic twin brothers with precisely the same 22q11.2 deletion but somewhat discordant clinical phenotype. Both twins had a small mouth, square nasal tip, short palpebral fissures, and small ears with deficient upper helices. Twin 1 had bilateral hair whorls and twin 2 had a right-sided hair whorl. Toes 4 and 5 were curled under bilaterally in both boys, this being more marked in twin 1. The twins were said to have had a single placenta although the findings of a detailed examination were not recorded. Twin 1 weighed 2,200 g and twin 2 weighed 2,800 g. Twin 1 had tetralogy of Fallot, which was repaired at 1 year of age. Twin 2 had a normal cardiovascular system. Twin 1 started taking steps at 24 months of age, while his brother stood at 13 months and walked steadily at 18 months. These observations indicated to Goodship et al. (1995) that differences in deletion size and modifying genetic loci are not responsible for all the phenotypic differences observed in CATCH22.

Vincent et al. (1999) reported the case of female monozygotic twins with 22q11 deletions. The twins shared facial characteristics of DGS/VCFS and immunologic defect. However, only one, who died on day 5, had a cardiac defect, comprised of an interrupted aortic arch with a ventricular septal defect, a truncus arteriosus, and a large arterial duct. The authors stated that this was the fourth report of a discrepant cardiac status between monozygotic twins harboring 22q11 deletions.

Wilson et al. (1992) looked for deletions in 9 families with 2 or more cases of outflow tract heart defects. In 5 of the families, chromosome 22 deletions were detected in all living affected persons studied and also in the clinically normal father of 3 affected children. The deletion was transmitted from parents to offspring and was associated with an increase in the severity of cardiac defects. No deletions were found in 4 families in which the parents were normal and affected sibs had anatomically identical defects, presumably an autosomal recessive form of congenital heart defect.

Fokstuen et al. (1998) analyzed 110 patients with nonselective syndromic or isolated nonfamilial congenital heart malformations by fluorescence in situ hybridization using the D22S75 DGS region probe. A 22q11.2 microdeletion was detected in 9 of 51 (17.6%) syndromic patients. Five were of maternal origin and 4 of paternal origin. None of the 59 patients with isolated congenital cardiac defect had the 22q11.2 deletion. In a study of 157 consecutively catheterized patients with isolated, nonsyndromic cardiac defects, and 25 patients with cardiac malformation and additional abnormalities (10 of whom had been clinically diagnosed as DiGeorge syndrome or velocardiofacial syndrome), Borgmann et al. (1999) found the 22q11.2 microdeletion only in the latter group.

Jawad et al. (2001) studied 195 patients with chromosome 22q11 deletion syndrome and found that diminished T-cell counts in the peripheral blood are common. The pattern of changes seen with aging in normal control patients was also seen in patients with the chromosome 22q11.2 deletion syndrome, although the decline in T cells was blunted. Autoimmune disease was seen in most age groups, although the types of disorders varied according to age. Infections were also common in older patients, although they were seldom life-threatening. Juvenile rheumatoid arthritis with onset between 1.5 and 6 years of age was seen in 4 of the 195 patients; idiopathic thrombocytopenia purpura with onset at 1 to 8 years of age was seen in 8 of 195 patients; autoimmune hemolytic anemia, psoriasis, vitiligo, inflammatory bowel disease, adult rheumatoid arthritis, and rheumatic fever with chorea were each seen in 1 patient of the 195 patients sampled.

Kawame et al. (2001) reported 5 patients with chromosome 22q11.2 deletion that manifested Graves disease between the ages of 27 months and 16 years, and suggested that Graves disease may be part of the clinical spectrum of this disorder.

Bassett et al. (2005) described the phenotypic features of 78 adults with 22q11 deletion syndrome and identified 43 distinct features present in more than 5% of patients. Common characteristic features included intellectual disabilities (92.3%), hypocalcemia (64%), palatal anomalies (42%), and cardiovascular anomalies (25.8%). Other less commonly appreciated features included obesity (35%), hypothyroidism (20.5%), hearing deficits (28%), cholelithiasis (19%), scoliosis (47%), and dermatologic abnormalities (severe acne, 23%; seborrhea, 35%). Significantly, schizophrenia was present in 22.6% of patients.

Maalouf et al. (2004) reported an African American male diagnosed at age 32 years with dysgenesis of the parathyroid glands due to a chromosome 22 microdeletion. Symptomatic hypocalcemia did not develop until age 14 years, a few weeks after initiation of anticonvulsant therapy for generalized tonic-clonic seizures. Because of the timing for onset of symptomatic hypocalcemia, it was presumed that the patient had anticonvulsant-induced hypocalcemia, and he carried that diagnosis for 18 years. Chromosome 22q11 deletion syndrome was first suspected at age 32 years. The diagnosis was confirmed by fluorescence in situ hybridization analysis. This case underscores the variable clinical presentation of this congenital form of hypoparathyroidism.

Kousseff (1984) described 3 sibs with a syndrome of sacral meningocele, conotruncal cardiac defects, unilateral renal agenesis (in 1 sib), low-set and posteriorly angulated ears, retrognathia, and short neck with low posterior hairline. Kousseff (1984) suggested autosomal recessive inheritance. Toriello et al. (1985) reported a similar, isolated case and designated the disorder Kousseff syndrome. Forrester et al. (2002) restudied the family reported by Kousseff (1984) and identified a 22q11-q13 deletion in the proband, his deceased brother, and his father. The proband had spina bifida, shunted hydrocephalus, cleft palate, short stature, cognitive impairment, and the typical craniofacial features of velocardiofacial syndrome, including low-set and dysplastic ears, broad base of the nose, narrow alae nasi, and retrognathia. His brother had died at 2 weeks of age with myelomeningocele, hydrocephalus, transposition of the great vessels, and unilateral renal agenesis, and his sister had died at 22 days of age with myelomeningocele, truncus arteriosus, hypocalcemia, and autopsy findings of absent thymus and parathyroid glands, consistent with DiGeorge anomaly.

Maclean et al. (2004) reported 2 unrelated patients with Kousseff syndrome, 1 with a 22q11.2 deletion and the other without. The first was a 4-year-old girl with a sacral myelomeningocele, tetralogy of Fallot, microcephaly, hydrocephalus, hypoplasia of the corpus callosum, and moderate developmental delay, who had a normal chromosome 22q11.2 FISH test and did not exhibit the facial phenotype of VCFS. The second patient, a male infant who died at 10 days of age, had a large sacral myelomeningocele, hydrocephalus, Arnold-Chiari malformation, atrial septal defect, conoventricular ventricular septal defect, type B interrupted aortic arch, hypocalcemia, and suspected duodenal stenosis; FISH testing revealed a 22q11.2 microdeletion. Maclean et al. (2004) concluded that Kousseff syndrome is a distinct clinical entity that is genetically heterogeneous.

Kujat et al. (2006) reported that 5 (83%) of 6 patients with a 22q11.2 microdeletion had renal anomalies, including renal dysplasia, hydronephrosis, and unilateral renal agenesis.

Robin et al. (2006) reviewed clinical data including brain imaging on 21 patients with polymicrogyria associated with deletion 22q11.2 and another 11 patients from the literature. The authors found that the cortical malformation consisted of perisylvian polymicrogyria of variable severity and frequent asymmetry with a striking predisposition for the right hemisphere (p = 0.008).

Forbes et al. (2007) reported the ocular features of 90 consecutive patients with confirmed 22q11.2 deletion syndrome. Posterior embryotoxon was found in 49%, tortuous retinal vessels in 34%, eyelid hooding in 20%, strabismus in 18%, ptosis in 4%, amblyopia in 4%, and tilted optic nerves in 1%.

Sundaram et al. (2007) described 2 patients with 22q11.2 deletion who had absent uterus and unilateral renal agenesis. One patient also had mild developmental delay, hypoparathyroidism, and psychiatric symptoms; the other patient also had high-arched palate, bulbous nasal tip, bicuspid aortic valve, short stature, and primary amenorrhea. Sundaram et al. (2007) suggested that mullerian or uterine/vaginal agenesis be included as part of the clinical spectrum of 22q11.2 deletion syndrome. Scheuerle (2008) reported a 14-year-old Latin American girl with 22q11.2 deletion syndrome who was found to have unilateral renal agenesis, uterine didelphys with duplication of the cervix, and imperforate vaginal hymen with hematometrocolpos.

Binenbaum et al. (2008) reported 4 boys and 3 girls with 22q11.2 deletion syndrome, including 5 who had bilateral sclerocornea. Other eye findings included descemetocele in 5 eyes, microphthalmia in 1 eye, severe anterior segment dysgenesis in 1 eye, and bilateral iridocorneal adhesions in 1 patient. Binenbaum et al. (2008) suggested that a genetic locus at chromosome 22q11.2 may be involved in anterior segment embryogenesis, and that sclerocornea should be added to the clinical manifestations of the 22q11.2 deletion syndrome.

Cheung et al. (2014) used a logistic regression model to investigate potential predictors of intellectual disability severity, including neonatal hypocalcemia, neonatal seizures, and complex congenital heart disease in 149 adults with 22q11.2 deletion syndrome, 10 of whom had moderate to severe intellectual disability. The model was highly significant (p less than 0.0001), showing neonatal seizures (p = 0.0018) and neonatal hypocalcemia (p = 0.047) to be significant predictors of a more severe level of intellectual disability. Neonatal seizures were significantly associated with hypocalcemia in the entire sample, regardless of intellectual level.

Biochemical Features

Hypocalcemia secondary to hypoparathyroidism is the key biochemical feature and may be sufficiently severe to be symptomatic. Resolution in early childhood is typical, although the deficient function of the parathyroids may be exposed in adulthood by infusion of disodium edetate (EDTA) (Gidding et al., 1988).

The patient of Gidding et al. (1988) had isolated conotruncal cardiac defect and, despite normal baseline ionized calcium and midmolecule parathyroid hormone levels, she failed to increase the secretion of midmolecular parathyroid hormone appropriately in response to a hypocalcemic challenge. They speculated that this combination of latent-hypoparathyroidism (LHP) and conotruncal cardiac defects should be included in the clinical spectrum of DiGeorge anomaly. Indeed, this woman's fourth child died with DiGeorge anomaly. Seven years after the report by Gidding et al. (1988), Cuneo et al. (1997) restudied the index patient with LHP and evaluated 3 generations of her family for parathyroid dysfunction, cardiac anomalies, and del22q11. Deletions were found in 6 relatives, 3 with conotruncal cardiac defects and 3 with a structurally normal heart. They found significant transgenerational noncardiac phenotypic variability, including learning difficulties, dysmorphic facial appearance, and psychiatric illness. A spectrum of parathyroid gland dysfunction associated with the del22(q11) was seen, ranging from hypocalcemic hypoparathyroidism to normocalcemia with abnormally low basal intact parathyroid hormone levels. In addition, LPH found in the index patient 7 years previously had evolved to frank hypocalcemic hypoparathyroidism.

Other Features

The deficit in thymic function results in a lack of T cells which may be demonstrated by measuring the proportion of CD4 cells (Wilson et al., 1993). Immunohistochemical analysis of the parathyroids reveals a deficit of thyrocalcitonin immunoreactive cells (C cells) (Palacios et al., 1993).

Levy et al. (1997) stated that 10 to 25% of parents of patients with DGS exhibit the 22q11 deletion but are nearly asymptomatic. The authors described 2 female patients carrying a 22q11 microdeletion who presented with idiopathic thrombocytopenic purpura. Both had children with typical manifestations of DGS. The possibility that defective thymic function predisposes patients with DGS to autoimmune diseases was raised.

Evers et al. (2006) reported a 52-year-old man with 22q11.2 deletion. As a child he showed learning disabilities and behavioral problems. As a young adult, he exhibited aggressive outbursts, apathy, echolalia, perseverations, and psychotic features, including delusional thoughts and hallucinations, necessitating long-term care in a psychiatric facility. Since then, he has demonstrated aggressive behavior, periods of withdrawal, and progressive cognitive decline consistent with dementia, particularly since the age of 36 years. An affected autistic sister also had the deletion.

Inheritance

DiGeorge syndrome is usually sporadic and results from de novo deletion within chromosome 22. A long series of reports has recognized the variable features resulting from this deletion in multiple family members with the variable phenotype behaving as an autosomal dominant trait (Steele et al., 1972; Raatikka et al., 1981; Atkin et al., 1982; Rohn et al., 1984; Keppen et al., 1988; Stevens et al., 1990). Stevens et al. (1990) suggested that such familial cases should be regarded as being velocardiofacial syndrome. The variable phenotype was described by Strong (1968) prior to the recognition of DGS. The mother in that family developed a psychotic illness. The first dominant pedigree in which marked clinical variability was associated with dominant transmission of a 22q11 deletion was reported by Wilson et al. (1991); the mother had the typical dysmorphic features. Of the 3 affected offspring, one had coarctation of the aorta, one a ventricular septal defect, and one DGS. Wilson et al. (1991) found 5 of 9 families ascertained on the basis of familial outflow tract defects to have 22q11 deletion. Subtle dysmorphic features typical of those seen in DGS were apparent in several of these affected family members.

Carelle-Calmels et al. (2009) noted that deletion of 22q11.2, resulting in DGS or VCFS, is usually sporadic but has been reported to be inherited in 6 to 28% of patients with these syndromes. They performed cytogenetic studies of the parents of a girl with DGS (or VCFS) who had a deletion of 22q11.2 and found an unexpected rearrangement of both 22q11.2 regions in the unaffected father. He carried a 22q11.2 deletion on one copy of chromosome 22 and a reciprocal 22q11.2 duplication (see 608363) on the other copy of chromosome 22. Genetic compensation, which is consistent with the normal phenotype of the father, was shown through quantitative-expression analyses of genes located within the genetic region associated with the 22q11 deletion syndrome. Carelle-Calmels et al. (2009) noted that this finding has implications for genetic counseling.

Delio et al. (2013) genotyped a total of 389 DNA samples from 22q11 deletion syndrome-affected families. A total of 219 (56%) individuals with 22q11 deletion had maternal origin and 170 (44%) had paternal origin of the de novo deletion, which represents a statistically significant bias for maternal origin (p = 0.0151). Combined with many smaller previous studies, 465 (57%) individuals had maternal origin and 345 (43%) had paternal origin, amounting to a ratio of 1.35 or a 35% increase in maternal compared to paternal origin (p = 0.000028). Among 1,892 probands with the de novo 22q11.2 deletion, the average maternal age at time of conception was 29.5, similar to data for the general population in 11 countries. Of interest, the female recombination rate in the 22q11.2 region was about 1.6 to 1.7 times greater than that for males, suggesting that for this region in the genome enhanced meiotic recombination rates, as well as other 22q11.2-specific features, could be responsible for the observed excess in maternal origin.

Cytogenetics

De la Chapelle et al. (1981) suggested that DiGeorge syndrome may be due to a deletion within chromosome 22 or partial duplication of 20p, based on finding the syndrome in members of a family with a 20;22 translocation. Specifically, they observed DGS in 4 members of 1 family and demonstrated monosomy of 22pter-q11 and 20p duplication. Their interpretation that DGS might result from monosomy for 22q11 was confirmed by Kelley et al. (1982) in 3 patients with translocation of 22q11-qter to other chromosomes.

Greenberg et al. (1984) observed partial monosomy due to an unbalanced 4;22 translocation in a 2-month-old male with type 1 truncus arteriosus and features of DGS. The asymptomatic mother showed partial T-cell deficiency and the same unbalanced translocation with deletion of proximal 22q11.

Augusseau et al. (1986) observed telecanthus, microretrognathia, severe aortic coarctation with hypoplastic left aortic arch, decreased E rosettes, and mild neonatal hypocalcemia. The same translocation was present in the clinically normal mother and maternal aunt. The latter had had her fourth pregnancy aborted because of cardiac and other malformations detected on ultrasound. This translocation has proved important in analysis of the expressed sequences in the deleted segment.

The recognition of the importance of 22q11 deletion grew with improving techniques. Greenberg et al. (1988) found chromosome abnormalities in 5 of 27 cases of DGS, 3 with 22q11 deletion though only one of these was an interstitial deletion.

Wilson et al. (1992) reported high resolution banding (more than 850 bands per haploid set) in 30 of 36 cases of DGS and demonstrated 9 cases of interstitial deletion. All other cases were apparently normal. Use of molecular dosage analysis and fluorescence in situ hybridization with probes isolated from within the deleted area revealed deletion in 21 of the 22 cases with normal karyotypes (Carey et al., 1992) giving pooled results of 33 deleted among the consecutive series of 35 cases. Driscoll et al. (1992) also found deletions at the molecular level in all 14 cases studied.

Whereas 90% of cases of DGS may now be attributed to a 22q11 deletion, other chromosome defects have been identified. In the report of Greenberg et al. (1988), there was 1 case of DGS with del10p13 and one with an 18q21.33 deletion. Fukushima et al. (1992) found a female infant with a deletion of 4q21.3-q25 associated with interrupted aortic arch, VSD, ASD, and PDA; T cell deficit and a small thymus at surgery; absent corpus callosum; and dysmorphic features. The possibility of an unrecognized submicroscopic deletion of 22q11 should be considered in such cases, although it is clear that the disturbance of neural crest migration presumed to underlie DGS may be caused by several distinct defects at the molecular level.

Pinto-Escalante et al. (1998) described a premature male infant with mosaic monosomy of chromosome 22. His facial appearance was similar to that in DiGeorge syndrome; hypertonicity, limitation of extension of major joints, and flexion contracture of all fingers were also present. They found previous reports of monosomy 22 in 6 cases, 3 of which were nonmosaic and 3 mosaic. There was great variability in anomalies in these patients; however, the most common anomalies were in the face and joints.

Gottlieb et al. (1998) determined the location and extent of the deletion on chromosome 10 in 5 DiGeorge syndrome patients by means of a combination of heterozygosity tests and fluorescence in situ hybridization analysis. The results did not support the existence of a single, commonly deleted region on 10p in these 5 patients. Rather, they suggested that deletion of more than 1 region on 10p could be associated with the DGS phenotype. Furthermore, there was no obvious correlation between the phenotypic traits of the patients and the extent of the deletion. The patient with the largest deletion exhibited one of the less severe phenotypes. The authors commented that the lack of a correlation between the size of a deletion and the phenotype is observed also with deletions on chromosome 22 and may be a characteristic of haploinsufficiency disorders.

Mapping

A large series of polymorphic markers and some expressed sequences have now been identified in the critical region (Fibison and Emanuel, 1987; Fibison et al., 1990; Scambler et al., 1990). The deletion lies proximal to the breakpoint critical region (151410). Details of the mapping of DGS to 22q11 are located in the Molecular Genetics and Cytogenetics sections of this entry.

Galili et al. (1997) documented homology of synteny between a 150-kb region on mouse chromosome 16 and the portion of 22q11.2 most commonly deleted in DiGeorge syndrome and VCFS. They identified 7 genes, all of which are transcribed in the early mouse embryo.

In 2 children with a DiGeorge syndrome phenotype from a consanguineous family, in whom deletion analysis at 22q11.2 and 10p14-p13 did not reveal any abnormality, Henwood et al. (2001) carried out microsatellite analysis. The affected children were homozygous at 3 markers within the 22q11.2 region, the markers being those at NLJH1, D22S941, and D22S944. The unaffected sib and the unaffected parents were heterozygous at these markers. A subsequent child who appeared to be unaffected was also found to be homozygous for the markers at these loci. Henwood et al. (2001), however, pointed out that nonpenetrance might be possible.

Molecular Genetics

Several expressed sequences have been identified in the region commonly deleted. Aubry et al. (1993) have identified a zinc finger gene ZNF74, and Halford et al. (1993) reported the expressed sequence T10. The gene TUPLE1 (TUP-like enhancer of split gene-1; 600237) reported by Halford et al. (1993) was an attractive candidate for the central features of the syndrome. This putative transcription factor shows homology to the yeast transcription factor TUP, and to Drosophila enhancer of split. It contains 4 WD40 domains and shows evidence of expression at the critical period of development in the outflow tract of the heart and the neural crest derived aspects of the face and upper thorax. The gene localizes to the critical DiGeorge region but was not disrupted by the translocation breakpoint described by Augusseau et al. (1986).

Augusseau et al. (1986) described a patient (ADU) with 'partial' DGS. She had telecanthus, microretrognathia, severe aortic coarctation with hypoplastic left aortic arch, decreased E rosettes, and mild neonatal hypocalcemia. The apparently balanced translocation involved chromosomes 2 and 22: t(2;22)(q14;q11). The same translocation was present in her mother (VDU). The original paper reported that VDU had no features of DGS. However, Budarf et al. (1995) observed that subsequent publications cited VDU as being mildly affected with hypernasal speech, micrognathia, and inverted T4/T8 ratio, which are all features seen in VCFS and DGS. The DGS phenotype in ADU, the VCFS phenotype in VDU, and a balanced translocation of chromosome 22 in both led Budarf et al. (1995) to clone the translocation, sequence the region containing the breakpoint, and analyze the DNA sequence for transcript identification. A gene disrupted by the rearrangement was identified. Their analysis suggested that there are at least 2 transcripts on opposite strands in the region of the t(2;22) breakpoint. The breakpoint disrupted a predicted ORF of one of these genes, deleting 11 nucleotides at the translocation junction. Additional fluorescence in situ hybridization studies and Southern blot analysis demonstrated that the deletions in chromosome 22 deletion-positive patients with DGS/VCFS include both of the transcripts at the t(2;22) breakpoint. Support that either of these putative genes is of significance in the etiology of DGS might come from determining whether all deleted patients are hemizygous for these loci and whether mutations in these genes are detectable in nondeletion patients with features of DGS. Lacking such evidence, the possibility remains that the translocation separates a locus control region from its target gene or produces a position effect. This has been suggested for the role of translocations seen in association with autosomal sex reversal and campomelic dysplasia (CMPD; 114290), where several disease-causing translocation breakpoints map 50 kb or more 5-prime of the SOX9 gene (608160).

Bartsch et al. (2003) used cytogenetic and analyses to study a series of 295 patients with suspected DiGeorge/velocardiofacial syndrome. They identified 58 subjects with a 22q11 deletion, and none with a 10p deletion. The common deletion was present in 52 subjects, the proximal deletion in 5, and an atypical proximal deletion due to a 1;22 translocation in 1. Bartsch et al. (2003) suggested that intellectual and/or behavioral outcome may be better with the proximal versus the common 22q11 deletion.

Demczuk et al. (1995) pointed to the existence of a strong tendency for 22q11.2 deletions in DGS, VCFS, and isolated conotruncal cardiac disease to be of maternal origin. With their experience of 22 cases in which parental origin could be determined, combined with recent results from the literature, 24 cases were found to be of maternal origin and 8 of paternal origin, yielding a probability of less than 0.01.

Demczuk et al. (1995) reported the isolation and cloning of a gene encoding a potential adhesion receptor protein (600594) in the DGCR. They designated the gene DGCR2 and suggested DGCR1 as a symbol for the TUPLE1 gene.

Pizzuti et al. (1996) described the cloning and tissue expression of a human homolog of the Drosophila 'dishevelled' gene (601225), a gene required for the establishment of fly embryonic segments. The 3-prime untranslated region of the gene was positioned within the DGS critical region and was found to be deleted in DGS patients. The authors stated that the gene may be involved in the pathogenesis of DGS.

Demczuk et al. (1996) described the cloning of a gene, which they referred to as DGCR6 (601279), from the DGS critical region. The putative protein encoded by this gene shows homology with Drosophila melanogaster gonadal protein (gdl) and with the gamma-1 chain of human laminin (150290), which maps to chromosome 1q31.

Edelmann et al. (1999) developed hamster-human somatic hybrid cell lines from VCFS/DGS patients and showed by use of haplotype analysis with a set of 16 ordered genetic markers on 22q11 that the breakpoints occurred within similar low copy repeats, designated LCR22s. Models were presented to explain how the LCR22s can mediate different homologous recombination events, thereby generating a number of rearrangements that are associated with congenital anomaly disorders.

Shaikh et al. (2000) completed sequencing of the 3-Mb typically deleted region (TDR) and identified 4 LCRs within it. Although the LCRs differed in content and organization of shared modules, those modules that were common between them shared 97 to 98% sequence identity with one another. Sequence analysis of rearranged junction fragments from variant deletions in 3 DGS/VCFS patients implicated the LCRs directly in the formation of 22q11.2 deletions. FISH analysis of nonhuman primates suggested that the duplication events which generated the nest of LCRs may have occurred at least 20 to 25 million years ago.

Stalmans et al. (2003) reported that absence of the 164-amino acid isoform of Vegf (Vegf164; see 192240), the only one that binds neuropilin-1 (602069), causes birth defects in mice reminiscent of those found in patients with deletion of 22q11. The close correlation of birth and vascular defects indicated that vascular dysgenesis may pathogenetically contribute to the birth defects. Vegf interacted with Tbx1, as Tbx1 expression was reduced in Vegf164-deficient embryos and knocked-down Vegf levels enhanced the pharyngeal arch artery defects induced by Tbx1 knockdown in zebrafish. Moreover, initial evidence suggested that a Vegf promoter haplotype was associated with an increased risk for cardiovascular birth defects in del22q11 individuals. Stalmans et al. (2003) concluded that genetic data in mouse, fish, and human indicated that VEGF is a modifier of cardiovascular birth defects in the del22q11 syndrome.

Baldini (2002) reviewed the molecular basis of DiGeorge syndrome, with special emphasis on mouse models and the role of TBX1 in development of the pharyngeal arches.

Yagi et al. (2003) screened for mutations in the coding sequence of TBX1 in 13 patients from 10 families who had the 22q11.2 syndrome phenotype but no detectable deletion in 22q11.2. They identified 3 mutations in TBX1 in 2 unrelated patients: 1 mutation was found in a case of sporadic conotruncal anomaly face syndrome/velocardiofacial syndrome and a second in a sporadic case of DiGeorge syndrome (602054.0002). A third mutation was found in 3 patients from a family with conotruncal anomaly face syndrome/velocardiofacial syndrome. The findings of Yagi et al. (2003) indicated that TBX1 mutations are responsible for 5 major phenotypes of the 22q11.2 syndrome, namely, abnormal facies (conotruncal anomaly face), cardiac defects, thymic hypoplasia, velopharyngeal insufficiency of the cleft palate, and parathyroid dysfunction with hypocalcemia; these mutations did not appear to be responsible for typical mental retardation that is commonly seen in patients with the deletion form of 22q11.2 syndrome.

Saitta et al. (2004) traced the grandparental origin of regions flanking de novo 3-Mb deletions in 20 informative 3-generation families with DiGeorge or velocardiofacial syndromes. Haplotype reconstruction of the flanking regions showed an unexpectedly high number of proximal interchromosomal exchanges between homologs, occurring in 19 of 20 families, whereas the normal chromosome 22 in these probands showed interchromosomal exchanges in 2 of 15 informative meioses, a rate consistent with the genetic distance. Immunostaining with MLH1 antibody showed meiotic exchanges localized to the distal region of chromosome 22q in 75% of human spermatocytes tested, also reflecting the genetic map. There was no effect of proband gender or parental age on crossover frequency, and parental origin studies in 65 de novo 3-Mb deletions demonstrated no bias. Unlike Williams syndrome (194050), FISH analysis showed no chromosomal inversions flanked by LCRs in 22 sets of parents of 22q11-deleted patients or in 8 nondeleted patients with a DGS/VCFS phenotype. Saitta et al. (2004) concluded that significant aberrant interchromosomal exchange events during meiosis I in the proximal region of the affected chromosome 22 are the likely etiology for these deletions. Since this type of exchange occurs more often for 22q11 deletions than for deletions of 7q11, 15q11, 17p11, and 17q11, they suggested that there is a difference in the meiotic behavior of chromosome 22.

Fernandez et al. (2005) found that 7 (13%) of 55 index patients with 22q11.2 deletion syndrome diagnosed by FISH analysis had inherited the deletion; 2 of the index patients were related as half sibs and had received the deletion from their shared mother. Using molecular techniques to characterize the size of the deletion, The authors found that 3 of 5 families had the smaller 1.5- to 2-Mb deletion and 2 families had the larger 3-Mb deletion; the size of the deletion in 1 family could not be determined. The findings suggested that small deletions may be more common in familial inheritance than larger deletions. Although the clinical severity did not differ between the 2 groups of patients, Fernandez et al. (2005) postulated that the smaller deletion may be associated with higher fecundity than the larger deletion.

Paylor et al. (2006) identified a heterozygous 23-bp deletion in the TBX1 gene (602054.0004) in a mother and 2 sons with VCFS. The mother also had major depression (608516) and 1 of the sons was diagnosed with Asperger syndrome (see, e.g., 608638 and 209850). Paylor et al. (2006) suggested that the TBX1 gene is a candidate for psychiatric disease in patients with VCFS and DiGeorge syndrome.

Kaminsky et al. (2011) presented the largest copy number variant case-control study to that time, comprising 15,749 International Standards for Cytogenomic Arrays cases and 10,118 published controls, focusing on recurrent deletions and duplications involving 14 copy number variant regions. Compared with controls, 14 deletions and 7 duplications were significantly overrepresented in cases, providing a clinical diagnosis as pathogenic. The 22q11.2 deletion was identified in 93 cases and no controls for a p value of 9.15 x 10(-21) and a frequency in cases of 1 of 169.

Genotype/Phenotype Correlations

Patients with DiGeorge syndrome are hemizygous for the COMT gene (116790). In a study of 21 nonpsychotic DiGeorge syndrome patients aged 7 to 16 years, Shashi et al. (2006) found that those carrying the met allele of the COMT V158M polymorphism (116790.0001), which results in increased dopamine in the prefrontal cortex, performed better on tests of general cognitive ability and on a specific test of prefrontal cognition compared to those with the val allele. Glaser et al. (2006) tested measures of executive function, IQ, and memory in 34 children and young adults with the 22q11.2 deletion syndrome (14 hemizygous for val158 and 30 for met158). No significant differences were detected between met- and val-hemizygous participants on measures of executive function. The groups did not differ on full-scale, performance, and verbal IQ or on verbal and visual memory. Glaser et al. (2006) suggested that either the COMT polymorphism has a small effect on executive function in 22q11.2 deletion syndrome or no effect exists at all.

Lopez-Rivera et al. (2017) conducted a genomewide search for structural variants in 2 cohorts: 2,080 patients with congenital kidney and urinary tract anomalies and 22,094 controls. Exome and targeted resequencing was performed in samples obtained from 586 additional patients with congenital kidney anomalies. Functional studies were also performed in zebrafish and mice. Lopez-Rivera et al. (2017) identified heterozygous deletion of chromosome 22q11.2 in 1% of patients with congenital kidney anomalies and in 0.01% of population controls (OR = 81.5, p = 4.5 x 10(-14)). The main driver of renal disease in DiGeorge syndrome was a 370-kb region containing 9 genes. In zebrafish embryos, an induced loss of function in snap29 (604202), aifm3 (617298), and crkl (602007) resulted in renal defects; the loss of crkl alone was sufficient to induce defects. Five of 586 patients with congenital urinary anomalies had newly identified heterozygous protein-altering variants, including a premature termination codon, in CRKL. The inactivation of Crkl in the mouse model induced developmental defects similar to those observed in patients with congenital urinary anomalies. Lopez-Rivera et al. (2017) concluded that a recurrent 370-kb deletion in the 22q11.2 locus is the driver of kidney defects in DiGeorge syndrome and in sporadic congenital kidney and urinary tract anomalies. Of the 9 genes at this locus, SNAP29, AIFM3, and CRKL appear to be critical to the phenotype, with haploinsufficiency of CRKL emerging as the main genetic driver.

Heterogeneity

The association of the DiGeorge syndrome with at least 2 and possibly more chromosomal locations suggests strongly that several genes are involved in control of migration of neural crest cells and their subsequent fixation and differentiation at different sites. In the mouse, Chisaka and Capecchi (1991) described a knockout of Hox A3(1.5) which produced a recessive phenocopy of DGS. This gene maps to human chromosome 7, an area not yet implicated in the cause of the human syndrome.

One explanation for the wide variation in phenotype would be the need for more than 1 gene defect to produce the severe version. Thus, for example, impaired signal and receptor may be needed to produce the full phenotype. Environmental factors could also play an additive role. Features of DGS have been described in children with clinical evidence of fetal alcohol syndrome. Ammann et al. (1982) found 4 children among a referral population with immunodeficiency who had hypocalcemia with decreased levels of parathormone, and T cell rosette formation of between 9 and 50% (normal over 65%). All 4 had cardiovascular lesions compatible with DGS; VSD with right aortic arch, truncus arteriosus and pulmonary stenosis, aberrant subclavian artery and pulmonary valve stenosis respectively. Two of the children had absent thymus at direct examination. The alcohol may have directly disrupted neural crest migration or have exposed a genetic predisposition. Among a series of pregnancies exposed to the teratogen isotretinoin (vitamin A) reported by Lammer et al. (1985) 21 malformed infants were investigated; 8 had conotruncal defects or aortic arch anomalies, 6 had micrognathia, 3 had cleft palate and 7 had thymic defects. Several of these children would satisfy the diagnostic criteria of DGS. Again, it is likely that this environmental challenge is exposing the same susceptible pathways of development as are impaired by the 22q11 deletion though the possibility of an interaction between the insult and genotype remains open.

Diagnosis