Cornelia De Lange Syndrome 1

A number sign (#) is used with this entry because Cornelia de Lange syndrome-1 (CDLS1) is caused by heterozygous mutation in the NIPBL gene (608667), which encodes a component of the cohesin complex, on chromosome 5p13.

Description

The Cornelia de Lange syndrome (CDLS) is a multisystem malformation syndrome recognized primarily on the basis of characteristic facial dysmorphism, including low anterior hairline, arched eyebrows, synophrys, anteverted nares, maxillary prognathism, long philtrum, thin lips, and 'carp' mouth, in association with prenatal and postnatal growth retardation, mental retardation and, in many cases, upper limb anomalies. However, there is wide clinical variability in this disorder, with milder phenotypes that may be difficult to ascertain on the basis of physical features (summary by Rohatgi et al., 2010).

Boyle et al. (2015) provided a detailed review of CDLS, including clinical features, diagnosis, and genetic counseling.

Genetic Heterogeneity of Cornelia de Lange Syndrome

About 50 to 60% of the cases of CDLS are due to mutation in the NIPBL gene (Musio et al., 2006; Rohatgi et al., 2010).

One X-linked form of CDLS (CDLS2; 300590) is caused by mutation in the SMC1A gene (300040), which accounts for about 5% of cases. Two milder variants of Cornelia de Lange syndrome have been identified: CDLS3 (610759), caused by mutation in the SMC3 gene (606062), and CDLS4 (614701), caused by mutation in the RAD21 gene (606462). All 4 genes, NIPBL, SMC1A, SMC3, and RAD21, encode components of the cohesin complex. Another X-linked form, CDLS5 (300882), is caused by mutation in the HDAC8 gene (300269), the vertebrate histone deacetylase of SMC3.

Clinical Features

Brachmann (1916) reported on a fetus with a very severe form of what is now known as the Cornelia de Lange syndrome. In Amsterdam, Cornelia de Lange (1933) described 2 infant girls with mental deficiency and other features with a less severe form of the same syndrome.

The facies is curious, with eyebrows growing across the base of the nose (synophrys); hair growing well down onto the forehead and low on the neck; unusually long eyelashes; depressed bridge of nose which has uptilted tip and forward-directed nostrils; small, widely spaced teeth; small head; and low-set ears. In a review of 31 cases previously diagnosed as having de Lange syndrome, Ireland et al. (1993) concluded that the facial findings of greatest diagnostic value were the combination of the characteristic eyebrows (neat, well-defined, and arched), long philtrum, thin lips, and crescent-shaped mouth. This combination of anomalies was absent in postpubertal males but not in postpubertal females. Facial abnormalities most likely to lead to an incorrect diagnosis were hypertrichosis, synophrys, and bushy eyebrows.

The ophthalmologic findings in this syndrome have been reported by Levin et al. (1990). De Lange syndrome is associated with ptosis, nystagmus and high myopia, poor macula reflex, hypertropia and nasolacrimal duct fistula. A patient with Peters anomaly was observed by Ponder et al. (1988).

Schlesinger et al. (1963) described radiologic anomalies associated with BDLS: 'The hands are characteristic, with flat spade-like appearance and short tapering fingers, the fifth especially so and curved inwards. A single deep transverse crease was seen over the palms.' The thumbs appear to arise from an abnormally far proximal position. The thenar eminence is inconspicuous so that the thumb suggests a lobster claw. The metacarpophalangeal profile, as described by Halal and Preus (1979) and Filippi (1989) is characteristic: of the metacarpals, the first is shorter than the others, and the second and fifth are shorter than the third and fourth. The middle phalanx of the index fingers is always hypoplastic. Large joints show limitation of motion. At times absence deformity, usually of one arm only, is so severe that only a single finger remains on a short arm. A case was reported by Ullrich (1951). The feet are very short but not malformed. Froster and Gortner (1993) described a typically affected infant with severe involvement of the upper limbs who also had thrombocytopenia, which may have been related to cavernous hemangiomas on the elbow and buttock as in the Kasabach-Merritt syndrome (141000). Fryns and Vinken (1994) described thrombocytopenia in 2 out of 50 patients with BDLS seen over a period of 25 years. Pfeiffer and Correll (1993) reported a male infant with BDLS and ulnar hemimelia and monodactyly but also absence of both tibiae, the right distal femur being bifurcated. Meinecke (1990) reported on a patient closely resembling the case described by Brachmann (1916), with severe ulnar defects and monodactylous hands.

Braddock et al. (1993) presented a review of the radiologic features of de Lange syndrome. The classic radiographic manifestations include microcephaly, limb and digital anomalies, delayed skeletal maturation, abnormal thoracic configuration, and flat acetabular angles in infancy. Unusual radiologic manifestations were related primarily to the limb anomalies, and these were often asymmetric.

Gastroesophageal dysfunction was emphasized by Lachman et al. (1981), Cates et al. (1989), and Rosenbach et al. (1992). Gastroesophageal reflux (GER; 109350) with reflex esophagitis, aspiration pneumonia, and esophageal stenosis had been described. Sommer (1993) examined 17 BDLS patients, ranging in age from 9 months to 19 years, and found that 13 had evidence of Sandifer complex, i.e., gastroesophageal reflux causing paroxysmal dystonic posture including torticollis and opisthotonos. Several children with typical BDLS and congenital diaphragmatic hernia were reported by Fryns (1987), Cunniff et al. (1993), and Jelsema et al. (1993).

In a series of 43 patients with CDLS, Luzzani et al. (2003) evaluated the incidence of GER and the correlation between its presence and severity and the clinical phenotype. Pathologic GER was evident in 28 (65%) of the 43 patients. The incidence was not significantly different in patients with classic (93.3%) versus mild (82.3%) phenotype, whereas a strong correlation was present between the degree of esophageal damage and the clinical phenotype. Hyperactivity was the most frequent sign associated with GER, present in 23 (85%) of the 28 patients.

A spectrum of endocrinopathies may be seen in patients with BDLS (Schwartz et al., 1990). These patients may be at risk for dysfunction of gonadotropin and prolactin secretion and of osmoregulatory mechanisms. A patient with panhypopituitarism of neonatal onset was reported by Tonini and Marinoni (1990).

In a psychosocial assessment of 36 patients, Beck (1987) found that the patients were particularly retarded in verbal communication but functioned relatively well in everyday self-help skills. Self-injurious behavior, frequently observed, can be treated (Menolascino et al., 1982 and Dossetor et al., 1991). Patients with normal intelligence have also been described (Gadoth et al., 1982 and Saal et al., 1993).

Berney et al. (1999) used a postal questionnaire to study 49 individuals with Cornelia de Lange syndrome (both the classic and mild forms) to ascertain behavioral phenotype. Ages ranged from early childhood to adulthood (mean age of 10.2 years) and the degree of mental retardation from borderline (10%), through mild (8%), moderate (18%), and severe (20%) to profound (43%). A wide variety of symptoms occurred frequently, notably hyperactivity (40%), self injury (44%), daily aggression (49%), and sleep disturbance (55%). These correlated closely with the presence of an autistic-like syndrome and with the degree of mental retardation.

BDLS has a variable phenotypic expression, which also evolves with age of the patient. In a clinical review of 310 cases of BDLS, Jackson et al. (1993) demonstrated a higher proportion of mildly affected cases. Only 27% of the cases had the upper limb deficiencies commonly associated with the syndrome. Opitz (1993) suggested that the severe limb defects associated with de Lange syndrome actually occur in a minority of cases. Several patients with a mild phenotype were reported at the Twelfth Annual David W. Smith Workshop on Malformations and Morphogenesis (Bay et al., 1993; Clericuzio, 1993; Leroy et al., 1993; Moeschler and Graham, 1993; Saul et al., 1993; Selicorni et al., 1993). The phenotype can be mild enough to be questionable (Baraitser and Papavasiliou, 1993; Greenberg and Robinson, 1989; Halal and Silver, 1992; Pashayan et al., 1970).

Based on the clinical variability in de Lange syndrome, Van Allen et al. (1993) proposed a classification system. Type I, or classic, BDLS patients have the characteristic facial and skeletal changes of the diagnostic criteria established by Preus and Rex (1983). They have prenatal growth deficiency, moderate to profound psychomotor retardation, and major malformations which result in severe disability or death. Type II, or mild, BDLS patients have similar facial and minor skeletal abnormalities to those seen in type I; however, these changes may develop with time or may be partially expressed. They have mild-to-borderline psychomotor retardation, less severe pre- and postnatal growth deficiency, and the absence of (or less severe) major malformations. Type III, or phenocopy, BDLS includes patients who have phenotypic manifestations of BDLS that are causally related to chromosomal aneuploidies or teratogenic exposures. Preus and Rex (1983) proposed 30 characteristics that best distinguish the de Lange syndrome from other suggestive cases.

Allanson et al. (1997) evaluated 43 subjects with de Lange syndrome, 30 with classic features and 13 with the mild phenotype. They compared gestalt, facial change with time, and detailed craniofacial measurements.

Allanson et al. (1997) concluded that, in the mild phenotype, the characteristic facial appearance may not appear until 2 to 3 years of age, while it is always present at birth in the classic phenotype. They also noted that the characteristic facial appearance decreased with time in the mild phenotype. Craniofacial pattern profiles showed that both groups had microbrachycephaly, but that the dimensions of the mild group were somewhat closer to normal. The correlation coefficient between the mild and classic phenotypes was 0.83 between ages 4 and 9 years and 0.71 in adults. Allanson et al. (1997) concluded that objective assessments supported the clinical impression of 2 distinct phenotypes, and that alternative discriminators, such as birth weight greater than 2,500 grams and absence of major limb anomalies, should be used to distinguish the mild from the severe phenotype early in life because of the similarity of facial features. Allanson et al. (1997) speculated that the 2 distinct phenotypes might be due to allele specificity or to modifying genes. Least likely, in their opinion, was the possibility that mild de Lange syndrome might be a phenocopy of the classic phenotype.

On the basis of 8 cases and a review of the literature, Steinbach et al. (1981) delineated the dup(3q) syndrome, which at least superficially simulates the de Lange syndrome. Features are statomotoric retardation, shortened life span, and a multiple congenital anomalies (MCA) syndrome comprising hypertrichosis, hypertelorism, anteverted nostrils, long philtrum, maxillary prognathism, carp mouth, highly arched or cleft palate, micrognathia, malformed pinnas, short and webbed neck, clinodactyly, simian crease, clubfoot, and congenital heart disease. ('Statomotoric' is a direct translation of the German 'statomotorisch,' which has the same meaning as 'psychomotor' (Opitz, 1991).) Rosenfeld et al. (1981) described a patient who did not show the hirsutism and synophrys present in other cases of dup(3q).

Breslau et al. (1981) provided a clinical comparison of the de Lange and dup(3q) syndromes. Convulsions, eye and palate anomalies, clubfoot, and renal and cardiac anomalies are more common in the dup(3q) syndrome; small hands and feet, limb reduction anomalies, proximally placed thumbs, hirsutism, synophrys, low hairline, cutis marmorata, low birth weight, and growth retardation are more common in the de Lange syndrome. Wilson et al. (1985) provided further delineation of the dup(3q) syndrome. They had data on 40 reported cases. Family studies of new cases are important because only 10 of the 40 represented de novo duplications. The characteristic face (hirsutism, synophrys, broad nasal root, anteverted nares, downturned corners of the mouth, micrognathia, and malformed ears) is recognizable even in the 30-week fetus. In an earlier study, Wilson et al. (1978) concluded that intrauterine growth retardation, prominent philtrum, proximally placed thumbs, oligodactyly/phocomelia, and syndactyly of toes 2 and 3 are more frequent in de Lange syndrome, whereas craniosynostosis, cleft palate, and urinary tract anomalies are more typical of dup(3q).

Selicorni et al. (2005) evaluated 61 patients with CDLS and detected structural anomalies of the kidney and urinary tract either by ultrasound or voiding cystourethrography in 25 (41%), including absent or poor corticomedullary differentiation in 8 patients, pelvic dilation in 6, vesicoureteral reflux in 5, small kidney in 3, isolated renal cyst in 3, and renal ectopia in 2. Renal function was reduced in 9 patients with renal tract abnormalities, 3 of whom had overt proteinuria. The clinical phenotype was more frequently of the classic type in patients with renal tract anomalies than in those without (p less than 0.05).

In a survey of 50 children with CDLS, Marchisio et al. (2008) found that 40 (80%) had hearing loss. Forty-seven (94%) patients had otitis media with effusion that was unrelated to presence of gastroesophageal reflux or respiratory infections. Thirty (60%) children had conductive hearing loss alone due to middle ear effusion, and 10 (20%) children with sensorineural hearing loss also had conductive hearing loss associated with middle ear effusion. Ten (20%) children had normal hearing. Greater hearing loss was associated with more severe developmental impairment.

Using published morphologic definitions of the ear (see, e.g., Allanson et al., 2009), Hunter et al. (2009) analyzed 119 ear photographs from CDLS patients compared to those from 57 controls. The ears of the CDLS patients were significantly different from those of the controls over a number of descriptors, the most significant of which included more frequent apparent posterior rotation, shorter and more serpiginous antihelical stem, sharper antihelical-to-inferior crus angle, shorter crus helix, more V-shaped incisura, and smaller lobe.

Selicorni et al. (2009) performed echocardiographic evaluation of 87 consecutive Italian patients referred with a diagnosis of Brachmann-de Lange syndrome and identified a cardiac anomaly in 29 (33.3%) of the patients, including 28 with a structural anomaly and 1 with isolated nonobstructive CMH (see 192600). Of the 28 patients with a structural anomaly, 12 (42.9%) had an isolated defect, including 10 (36%) with pulmonary stenosis and 8 (28.6%) with an isolated left-to-right shunt. The single most common lesion was valvular pulmonary stenosis, which was present in 11 (39%) of 28 patients. Isolated late-onset mild to moderate mitral or tricuspid valve regurgitation was detected at follow-up examination in 4 patients (14.3%) older than 10 years, who had a previously normal examination and electrocardiogram. Selicorni et al. (2009) noted that in contrast to previous studies, only 2 patients required surgical intervention, 1 for closure of a large ventricular septal defect (VSD) with associated atrial septal defect, and the other for VSD closure and relief of pulmonary valve stenosis.

Other Features

Immunohistochemical examination of two placentae from BDLS patients revealed the absence of pregnancy-associated plasma protein A (PAPPA; 176385) from the syncytiotrophoblast (Westergaard et al., 1983).

Melegh et al. (1996) described a newborn boy with clinical features of de Lange syndrome who manifested dyspnea, hypertonia, and hyperthermia. Muscle biopsy showed severe distortion of the mitochondrial architecture. Multiple deletions of mtDNA were found on Southern blot analysis. The authors suggested that clinical findings of de Lange syndrome in combination with multiple mtDNA deletions and hyperthermia may represent a distinct syndrome.

Inheritance

Most cases are sporadic. In rare instances (e.g., Borghi et al., 1954), multiple presumably affected sibs have had normal parents. Although Ptacek et al. (1963) suggested dominant inheritance, Opitz (1971, 1985) later thought recessive inheritance likely. Pashayan et al. (1969) concluded that the recessive hypothesis can be rejected. The empiric recurrence risk in a sib of an affected child was estimated to be between 2 and 5%. Familial occurrence and parental consanguinity were noted by Pearce et al. (1967). Opitz (1971) found normal parental age (average paternal and maternal age 30.6 and 28.9 years, respectively). Beratis et al. (1971) described 3 affected sibs with normal karyotypes and normal, nonconsanguineous parents. Discordance in dizygotic (Stevenson and Scott, 1976) and monozygotic (Carakushansky and Berthier, 1976) twins has been reported. Opitz (1985) published photographs of concordant monozygotic twins with de Lange syndrome. Carakushansky et al. (1996) gave a follow-up, with photographs, of the discordant twins at the age of 20. DNA fingerprinting with 3 multilocus probes allowed them to establish monozygosity with a high degree of certainty.

Robinson and Jones (1983) supported the conclusion that the de Lange syndrome is autosomal dominant and that the sporadic occurrence in most cases reflects the genetic lethality of the disorder. Their cases were a severely affected 5-month-old boy and his mildly affected 24-year-old mother. She had mildly delayed development, with difficulties in school, and showed synophrys, long philtrum, thin upper lip, fifth finger clinodactyly, and very short right fourth metacarpal.

Kumar et al. (1985) found de Lange syndrome in several members of a family in a pattern consistent with autosomal dominant inheritance. Winter (1986) suggested that the diagnosis was in fact the Ruvalcaba syndrome (180870) because of the combination of eyebrow and hand anomalies. Robinson et al. (1985) reported a mildly affected mother and her 2 severely affected sons, possibly indicating mosaicism in the mother. Mosher et al. (1985) reported the case of a 24-year-old woman with de Lange syndrome who delivered a normal child. Leavitt et al. (1985) reported seemingly typical features in mother and daughter. Fryns (1986) suggested that the disorder in the families reported by Leavitt et al. (1985), Mosher et al. (1985), and Robinson et al. (1985) was Coffin-Siris syndrome (135900), not de Lange syndrome. Bankier et al. (1986) brought to 5 the number of families in which BDLS had been inherited as an autosomal dominant. Feingold and Lin (1993) reported affected mother and daughter.

Fryns et al. (1987) reported 2 infant brothers with a severe form of the syndrome. They died at the age of 3 months and 3 weeks, respectively. The parents were normal, and prometaphase chromosome studies failed to show any abnormality. This would be consistent with autosomal dominant inheritance and gonadal mosaicism. Naguib et al. (1987) described an Arab family with phenotypically normal first-cousin parents and 2 offspring showing variable features of this disorder. The proband had apparently normal chromosomes and had died at the age of 3 months. His sister was less severely affected and lived for 6 years. The authors suggested recessive inheritance. Opitz (1985) suggested that high prenatal lethality of homozygotes explains a segregation ratio that is much lower than one would expect under the recessive hypothesis. In the mother of a child with typical features, de Die-Smulders et al. (1992) observed mild manifestations. They pointed to several reports of similar situations and concluded that 'in all convincingly autosomal dominant cases' the mother was the transmitting parent, suggesting genomic imprinting. They suspected that de novo mutation causes a severe form of the syndrome and that recurrence within sibships with unaffected parents may be explained by germline mosaicism. Chodirker and Chudley (1994) reported apparent male-to-male transmission of mild BDLS. The proband's father was mentally retarded, showed synophrys and some other facial manifestations of BDLS, and was the shortest of his 13 sibs.

Krajewska-Walasek et al. (1995) reported a brother and sister with variable manifestations of a less severe type of Brachmann-de Lange syndrome. There was no significant prenatal growth retardation and no reduction deformities of the forearms. They noted that, with one exception, previously reported sibs with normal parents presented with the severe type of this disorder, the so-called 'classic' or 'full' form, with major upper limb anomalies, severe growth and mental retardation, and, frequently, early death.

Russell et al. (2001) reported a familial case of Cornelia de Lange syndrome transmitted from father to daughter and reviewed the literature on familial cases. They concluded that autosomal dominant inheritance is the most likely mode of transmission, with most cases arising from spontaneous mutations.

Caksen et al. (2001) analyzed 7 infants with this disorder, including 2 who were monozygotic twin sisters. All had normal parents with no consanguinity.

McConnell et al. (2003) reported a family with a classically affected neonate with de Lange syndrome, an affected mother, and a probably affected maternal grandmother, thus suggesting autosomal dominant inheritance.

Diagnosis

Diagnosis is dependent on the recognition of the distinctive facial features (Ireland and Burn, 1993). The diagnosis is seldom in doubt when there is a major longitudinal deficiency defect of the upper limb, severe prenatal and postnatal growth retardation, and severe mental retardation. Uncertainty arises when the patient has the characteristic facial findings but lacks one or more of the other manifestations.

Selicorni et al. (2007) devised a clinical scoring system that assessed auxologic, malformation, and neurodevelopmental parameters to measure the clinical severity of Cornelia de Lange syndrome. A study of 62 Italian patients with a clinical diagnosis of the disorder showed wide phenotypic variability, ranging from mild to severe.

Based on a survey of 65 dysmorphologists who were provided with facial photographs of 32 CDLS patients of varying severity or with features suggestive of the disorder but with another diagnosis, Rohatgi et al. (2010) found that 90% of classic CDLS cases were correctly diagnosed, but only 54% of mild or variant cases were accurately diagnosed. The disorder was most accurately diagnosed in childhood and became more difficult to diagnose with increasing age of the patient. Features used to make the correct diagnosis included penciled and arched eyebrows, high set/short anteverted nose, a long flat philtrum, thin upper lip, downturned corners of the mouth, and micrognathia. Features that proved to be misleading included full or flat brows, a prominent nasal bridge or bulbous tip, and/or a normal or prominent chin. There were some differences between severely and mildly affected patients that could be used to distinguish genotypes: those with mild NIPBL mutations had more typical features, whereas those with SMC1A mutations had mild synophrys, long eyelashes, slightly short, high-set nose with mild anteversion, box-like nose, thin upper lip, and downturned corners of the mouth.

Prenatal Diagnosis

Because there are no genetic or biochemical tests at present, the antenatal detection depends upon identification of some aspects of the phenotype in the fetus using ultrasound imaging, namely growth retardation, limb defects, hirsutism, and diaphragmatic hernia (Kliewer et al., 1993). Manouvrier et al. (1996) reported ultrasonographic prenatal diagnosis of BDLS by the association of intrauterine growth retardation, hypoplastic forearms, underdevelopment of hands, typical facial defects, and diaphragmatic hernia.

Urban and Hartung (2001) reported observations on a 22-week-old female fetus with BDLS. The facial appearance was already characteristic and the associated upper limb malformations (bilateral monodactyly and ulnar agenesis) supported the diagnosis. The prenatal ultrasound images demonstrated a protruding and overhanging upper lip and severe retrognathia.

Clinical Management

Schrier et al. (2011) retrospectively reviewed 426 probands with a confirmed clinical diagnosis of CDLS who died in a 41-year period between 1966 and 2007. Among 295 probands with a known cause of death, respiratory causes, including aspiration/reflux and pneumonias, accounted for 31% of deaths; gastrointestinal disease, including obstruction/volvulus, accounted for 19%; congenital anomalies, including diaphragmatic hernia and congenital heart defects, accounted for 15%. Neurologic causes and accidents each accounted for 8% of deaths, sepsis for 4%, acquired cardiac disease for 3%, cancer for 2%, and renal disease for 1.7%, with other causes of death accounting for 9%. Based on these findings, Schrier et al. (2011) provided recommendations for age-specific monitoring and care.

Population Genetics

Beck (1976) estimated the frequency to be 0.6 per 100,000 in Denmark. The oldest patient found in a nationwide survey was 49 years old. Beck's series contained a half brother and sister (same mother), one instance of parental consanguinity out of 24, and one patient with a low normal IQ. Normal IQ or only mild mental retardation in this disorder was discussed.

Mapping

Krantz et al. (2001) performed linkage analysis in 10 multicase families using markers from the minimal dup(3q) critical region on 3q26.31-q27.3 that encompassed the breakpoint seen in the translocation patient reported by Ireland et al. (1991). Nineteen markers spanning a region of approximately 40 Mb (37 cM) were used. Multipoint linkage analysis demonstrated negative total lod scores across the chromosome 3q26-q27 region. In 4 families, lod scores were less than -2 in the 2-cM region encompassing the translocation, thus excluding linkage in these families. In the remaining 6 families, lod scores could not exclude linkage to this region. The authors concluded that in some multicase families, the disease locus does not map to the CDL1 region at 3q26.3.

Tonkin et al. (2004) analyzed several de novo balanced translocations associated with CDLS and in 1 instance mapped the breakpoints to 5p13.1 and 13q12.1. Because of a report of CDLS in association with a 5p14.2-p13.1 deletion, they focused on the 5p breakpoint and found that it is located in a novel gene they named NIPBL for Nipped-B-like (608667), mutations in which were found to cause CDLS. They also analyzed the translocations t(3;17)(q26.3;q23.1) (Ireland et al., 1991) and t(14;21)(q32;q11) (Wilson et al., 1983). The 3q breakpoint disrupts a large gene undergoing unusual alternative splicing, but they found no mutation specific to any individual with CDLS. Molecular analyses of regions spanning the 17q23, 14q32, and 21q11 breakpoint regions also did not identify a gene likely to underlie CDLS.

Krantz et al. (2004) carried out genomewide linkage exclusion analysis in 12 families with CDLS and identified 4 candidate regions, of which 5p13.1 gave the highest multipoint lod score of 2.7. This information, together with the previous identification of a child with CDLS with a de novo t(5;13)(p13.1;q12.1) translocation and another with classic CDLS and a de novo chromosome 5p14.2-p13.1 deletion (Hulinsky et al., 2003), allowed delineation of a 1.1-Mb critical region on chromosome 5 for the gene mutated in CDLS.

Cytogenetics

The large number of de Lange cases found to have one or another type of chromosomal aberration may be fortuitous, may indicate a predisposition to chromosomal change induced in some way by a point mutation (as in Bloom syndrome and in Fanconi panmyelopathy), or may indeed have a cause-and-effect relationship. According to Craig and Luzzatti (1965), 11 out of 38 patients in whom the chromosomes have been studied showed abnormalities. They felt this was more than chance association. Falek et al. (1966) described 3 affected sibs and their affected first cousins. Patients showed 46 chromosomes with loss of one small acrocentric of the G group and an additional metacentric chromosome resembling, but somewhat smaller than, chromosome 16. Six phenotypically normal relatives, including 1 parent of each of the 2 affected sibships, had the same anomalous chromosome as the affected children but in addition an apparent deletion of one chromosome 3. The authors suggested that the de Lange syndrome is the result of excessive chromosome 3 material. The anomalous chromosome was interpreted as combining one G chromosome with a fragment from one chromosome 3.

McArthur and Edwards (1967) found normal chromosomes in all 20 of their cases. However, they expressed the opinion that the condition is most likely related to a chromosomal deficiency which is not usually detectable. This would explain both the usual sporadic nature and the occasional familial occurrence. Broholm et al. (1968) described a patient with de Lange syndrome and a B-D translocation inherited from the normal mother. The patient was thought to be partially trisomic for a group D chromosome.

Features suggesting the de Lange syndrome are observed with partial trisomy of the distal portion of chromosome 3, specifically the area qter-3q21 (Allderdice et al., 1975). The reported familial cases of de Lange syndrome (e.g., Falek et al., 1966) may be on the basis of this chromosomal anomaly segregating from a balanced rearrangement. A small duplication of the long arm of chromosome 3 is accompanied by features suggestive of the de Lange syndrome; occurrence as an unbalanced segregation in certain families may account for some of the cases of 'familial Cornelia de Lange syndrome' (Francke, 1978). See the earlier discussion of the work of Steinbach et al. (1981), Breslau et al. (1981), and Wilson et al. (1985), comparing the Cornelia de Lange syndrome and the dup(3q) syndrome.

Beck and Mikkelsen (1981) studied 45 de Lange syndrome cases clinically and karyologically, with prometaphase studies in 31. All karyotypes were normal. In 1 other patient, a girl, a 45,X karyotype was found and in a boy, a (13q14q) translocation was found which was also present in the phenotypically normal mother and grandmother. The duplication 3q syndrome was found in none. The authors cited a recurrence risk of 2 to 5% for the de Lange syndrome. A recurrence risk of this order might be observed with a genetic lethal, autosomal dominant disorder with parental gonadal mosaicism. Another case of BDLS associated with a reciprocal translocation 14q;21q was published by Wilson et al. (1983).

Lakshminarayana and Nallasivam (1990) found ring chromosome 3 in an infant with presumed Cornelia de Lange syndrome. Breslau et al. (1981) analyzed the prometaphase chromosomes of 5 patients (1 pair of sibs) with the de Lange syndrome and found no chromosome abnormality in any of them. They suggested that the de Lange and dup(3q) syndromes can be distinguished on clinical and chromosomal grounds. They recommended chromosome studies in any patient with de Lange or de Lange-like manifestations. The possibility remains that the mutation responsible for the de Lange syndrome is located in the same region of 3q that is abnormal in the dup(3q) syndrome.

Ireland et al. (1991) reported a typical case with unusually severe limb reduction defects. Chromosome analysis showed a de novo translocation with breakpoints at 3q26.3 and 17q23.1. After reviewing cases showing phenotypic overlap between de Lange syndrome and partial trisomy 3q and cases of deletions of 3q, they proposed that the gene for Cornelia de Lange syndrome may be located at 3q26.3. Lopez-Rangel et al. (1993) reported the case of a 13-year-old girl with a duplication in the 3q25.1-q26.1 region who had neither BDLS nor the dup(3q) phenotype.

DeScipio et al. (2005) reported 2 half-sibs with clinical features suggestive of de Lange syndrome and an unbalanced chromosomal rearrangement, der(3)t(3;12)(p25.3;p13.3), inherited from a balanced translocation in their unaffected mother, t(3;12)(p25.3;p13.3). The sibs had many features consistent with de Lange syndrome, including microcephaly, growth retardation, mental retardation, hirsutism, synophrys, anteverted nares, single palmar creases, and syndactyly of toes 2 and 3, but also had significant clinical overlap with del(3)(p25) syndrome (see 607416 and 607280). DeScipio et al. (2005) reviewed all reported cases of de Lange syndrome with chromosomal rearrangements.

Molecular Genetics

Tonkin et al. (2004) screened multiple individuals with CDLS for mutations in the NIPBL gene (608667) and identified 9 plausible point mutations, at least 5 of which arose de novo (see, e.g., 608667.0002, 608667.0004, and 608667.0006). They found mutations in individuals with severe and mild CDLS, suggesting that phenotype variation can be explained, at least in part, by allelic heterogeneity. The spectrum and distribution of mutations that implied pathogenesis arises from loss or altered function of a single NIPBL allele.

Krantz et al. (2004) identified mutations in the NIPBL gene in 4 sporadic and 2 familial cases (see, e.g., 608667.0001; 608667.0003, and 608667.0005). They noted that Drosophila Nipped-B facilitates enhancer-promoter communication and regulates Notch signaling and other developmental pathways.

Pehlivan et al. (2012) reported that among 162 patients with CDLS for whom mutations in the known CDLS genes were negative by sequencing, they identified deletions containing NIPBL exons in 7 subjects (approximately 5%). Breakpoint sequences in 5 of the 7 subjects implicated microhomology-mediated replicative mechanisms. Most deletions are predicted to result in haploinsufficiency due to heterozygous loss-of-function mutations, which may result in a more severe CDLS phenotype. Pehlivan et al. (2012) concluded that their findings suggested a potential clinical utility to testing for copy number variations involving NIPBL when clinically diagnosed CDLS cases are mutation-negative by DNA sequencing studies.

Somatic Mosaicism

Huisman et al. (2013) detected pathogenic mutations in the NIPBL gene in buccal cells from 10 of 13 patients with CDLS in whom no mutations were detectable earlier in lymphocytes. Resequencing of the gene in lymphocytes from these 10 patients again failed to detect the NIPBL mutation, indicating somatic mosaicism. Statistical analysis did not show a phenotypic difference between these patients and patients with germline NIPBL mutations. The patients were part of an earlier study of 44 patients with CDLS (Bhuiyan et al., 2006) and thus accounted for 23% of the study group. Huisman et al. (2013) commented on the unusually high frequency of somatic mosaicism found in their study, and suggested that it was due to selection against lymphocytes carrying the mutation ('reversion'). The findings indicated that molecular study of buccal swabs in patients with a CDLS phenotype could facilitate molecular diagnosis.

Exclusion Studies

Smith et al. (1999) excluded the SOX2 (184429) gene as a candidate for Cornelia de Lange syndrome.

In the course of studying the molecular basis of CDLS, Tonkin et al. (2004) focused on the distal 3q region because of the occurrence, in a patient with classic CDLS, of a de novo balanced translocation with a breakpoint at 3q26.3 (Ireland et al., 1991) and because of reports of phenotypic overlap between cases of mild CDLS and individuals trisomic for the 3q26-q27 region. They found that the 3q26.3 breakpoint in the t(3;17)(q26.3;q23.1) translocation severed a previously uncharacterized gene, designated NAALADL2 (608806). Mutation screening of the gene in a panel of CDLS patient DNA samples failed to identify patient-specific mutations.

Genotype/Phenotype Correlations

Yan et al. (2006) identified 13 different NIPBL mutations, including 11 novel mutations, in 13 (46%) of 28 Polish patients with a clinical diagnosis of CDLS. Eleven of the mutations resulted in a premature termination of the protein. Mutation-positive patients were more severely affected than mutation-negative patients with respect to prenatal growth, facial dysmorphism, and speech impairment.

Bhuiyan et al. (2006) stated that to the time of their report, 161 patients were studied molecularly, of whom 63 (39%) were found to have a mutation. Reporting from the Netherlands, the country where CDLS was first described, Bhuiyan et al. (2006) described genotype-phenotype correlations in 39 patients. They found mutations of NIPBL in 56% of the patients. Bhuiyan et al. (2006) found that truncating mutations generally caused a more severe phenotype, but that this correlation was not absolute. By using 3-dimensional facial imaging, they demonstrated the potential for classifying facial features. Behavioral problems were highly correlated with a level of adaptive functioning, and also included autism. No correlation of behavior with the type of mutation was found.

Selicorni et al. (2007) identified 25 different NIPBL mutations in 26 (44%) of 62 unrelated Italian patients with a clinical diagnosis of CDLS. Compared to the 36 patients without NIPBL mutations, patients with NIPBL mutations had more pronounced growth retardation, more limb reduction, and more delayed speech development. There was a correlation between severe phenotype and truncating mutation, mild disease and missense mutation, and moderate disease and splice site mutation.

Among 30 unrelated patients with CDLS, Pie et al. (2010) found that 11 (37%) patients had mutations in the NIPBL gene and 3 (10%) had mutations in the SMC1A gene, with an overall molecular diagnostic yield of 47%. Nine novel NIPBL mutations were reported. None of the patients had mutations in the SMC3 gene. Most of the patients were of Spanish origin. Although those with NIPBL mutations had a more severe phenotype than those with SMC1A mutations, the incidence of palate defects was higher in those with SMC1A mutations.

History

De Lange (1933) (pronounced LANG-eh) described the disorder that carries her name. She was professor of pediatrics in Amsterdam and an immediate predecessor of Van Creveld (1969) in that chair. De Knecht-van Eekelen and Hennekam (1994) provided biographical information on Cornelia de Lange and a bibliography of her publications.

Oostra et al. (1994) reported that a specimen of de Lange syndrome resides in the anatomical collection of the University of Amsterdam. Vrolik (1849) had described this case as an example of 'extreme oligodactyly.'

Opitz (1985) gave a delightful account of his first brush with de Lange syndrome and his long association thereafter. Serendipity was responsible for his insistence on expanding the eponym to Brachmann-de Lange. 'In the fall of 1963...the former head of the...Libraries, came to ask my advice on what to do with a series of volumes of the Jahrbuch fur Kinderheilkunde, which had been damaged...by a burst water pipe. In particular, she was upset by volume 84, dated 1916, the pages of which were completely glued together except for one place, the article beginning on p. 225. I was startled to find out that here was an article on the Cornelia de Lange syndrome written 17 years before de Lange's first paper of 1933. The author, Dr. W. Brachmann, whose subsequent fate is unknown to me, was then a young physician in training, who apologized that his study of this remarkable case was interrupted by sudden orders to report for active duty (in the German Army).'