John C. K. Barber

The 8p23.1 duplication syndrome and copy number variation of the 8p23.1 defensin gene cluster are cytogenetically indistinguishable but distinct at the molecular level. To our knowledge, the 8p23.1 duplication syndrome has been described at prenatal diagnosis only once and we report our experience with four further apparent duplications ascertained at prenatal diagnosis.

The 8p23.1 deletion syndrome is established but not an equivalent duplication syndrome. Here, we report five patients; a de novo prenatal case and two families in which 8p23.1 duplications have been directly transmitted from mothers to children. Dual-colour fluorescent in situ hybridisation, multiplex ligation-dependent probe amplification analysis and customised oligonucleotide array comparative genomic hybridisation (oaCGH) indicated an ~3.75 Mb duplication of most of band 8p23.1 between the olfactory receptor/defensin repeats (ORDRs) in all cases. However, oaCGH revealed an additional duplication of 500 kb adjacent to the proximal ORDR in Family 1 and an additional deletion of 3.14 Mb within the Nablus Mask-Like Facial Syndrome region of 8q22.1 in Family 2. Copy number variation at introns 4–5 of the GATA4 gene was also identified. This 8p23.1 duplication syndrome is associated with a characteristic facial phenotype including a prominent forehead and arched eyebrows. Adrenal insufficiency, Tetralogy of Fallot, partial 2/3 syndactyly of the toes and cleft palate in some individuals may be explained by ascertainment bias, incomplete penetrance and/or the presence of the microdeletion in Family 2. The duplication is compatible with normal early childhood development but, although our adult cases live independent lives with varying degrees of support, learning difficulties have been experienced by some family members. We conclude that the 8p23.1 duplication syndrome is a genomic condition with an emerging but variable phenotype that may be under-diagnosed. Our results demonstrate that direct transmission does not distinguish genuine duplications from euchromatic variants and illustrate the power of array CGH to reveal unexpected additional imbalances in affected patients.

Extra euchromatic material was found within the major heterochromatic block of chromosome 16 (16qh) in one de novo case and seven members of two families. In contrast to the euchromatic variants of chromosome 9 (9qh), which are derived from pericentromeric euchromatin, molecular cytogenetics confirmed that these duplications were of 16q11.2→q12.2 in the de novo case, of 16q11.2→q13 in three members of family 1 and 16q11.2→q12.1 in four members of family 2. The duplication had arisen as a post-zygotic mitotic event in the mother of family 1 and been transmitted paternally in family 2. An insertional mechanism of origin is proposed for the duplications in case 1 and family 1. Expression at the 16q13 matrix metalloproteinase-2 (MMP2)locus in families 1 and 2 was proportional to genomic copy number and not therefore consistent with position effect silencing due to the flanking blocks of heterochromatin. We conclude that proximal 16q duplications within 16qh are not novel euchromatic variants but associated with a variable phenotype including developmental delay, speech delay, learning difficulties and behavioural problems. The behavioural problems in families ascertained through affected children are much less severe than those encountered in previous patients ascertained as adults.

A novel transmitted 2–3 Mb deletion of 2q14.1–q14.2 was found in an affected boy from a consanguineous family with a possible diagnosis of PEHO syndrome (OMIM 260565). BAC FISH showed that the deletion included a minimum of 20 genes including the homeobox engrailed 1 gene (EN1). However, the same deletion was also found in his phenotypically normal father and brother (family 1). The phenotype of the proband may, therefore, have been coincidental to the deletion, a result of a recessive condition within or outside the deleted segment or possibly due to variable dosage compensation of EN1 by the paralogous EN2 gene at 7q36. BAC FISH also showed that this deletion overlapped with a previously reported transmitted deletion of 2q13–q14.1 that had no phenotypic consequences (family 2). The deleted regions contained a total of 32 genes and comprise the final 5.25 Mb of the ancestral chromosome 2B from which chromosome 2 was formed in man. These families provide further evidence that heterozygous deletions of regions of low gene density are compatible with a normal phenotype.

It has been proposed that duplications of 8p23.1 are either euchromatic variants of the 8p23.1 defensin domain with no phenotypic consequences or true duplications associated with developmental delay and heart defects. Here, we provide evidence for both alternatives in two new families. A duplication of most of band 8p23.1 (circa 5 Mb) was found in a girl of 8 years with pulmonary stenosis and mild language delay. BAC fluorescence in situ hybridisation (FISH) and multiplex amplifiable probe hybridisation (MAPH) showed that the two copies of the duplicated segment were sited, in an alternating fashion, between three copies of a circa 300–450 kb segment from 8p23.1 distal to REPD. Copy number of the variable 8p23.1 defensin domain was consistent with duplication but within the normal range. Duplication of the GATA-binding protein 4 gene (GATA4) in this patient and others with and without heart defects, suggests it is a dosage-sensitive gene with variable penetrance. A cytogenetically similar duplication of 8p23.1 was found at prenatal diagnosis in a fetus, father and grandmother. There was no duplication using BAC FISH but MAPH showed 11 copies of the 360 kb variable defensin domain which is within the expanded range found in previous euchromatic variant carriers. Semiquantitative FISH (SQ-FISH) was consistent with a simultaneous expansion of the adjacent olfactory receptor repeats. These results distinguish duplications of 8p23.1 with clinically significant consequences from benign copy number variants, which have not yet been associated with qualitative or quantitative traits.

In total, 200 families were reviewed with directly transmitted, cytogenetically visible unbalanced chromosome abnormalities (UBCAs) or euchromatic variants (EVs). Both the 130 UBCA and 70 EV families were divided into three groups depending on the presence or absence of an abnormal phenotype in parents and offspring. No detectable phenotypic effect was evident in 23/130 (18%) UBCA families ascertained mostly through prenatal diagnosis (group 1). In 30/130 (23%) families, the affected proband had the same UBCA as other phenotypically normal family members (group 2). In the remaining 77/130 (59%) families, UBCAs had consistently mild consequences (group 3). In the 70 families with established EVs of 8p23.1, 9p12, 9q12, 15q11.2, and 16p11.2, no phenotypic effect was apparent in 38/70 (54%). The same EV was found in affected probands and phenotypically normal family members in 30/70 families (43%) (group 2), and an EV co-segregated with mild phenotypic anomalies in only 2/70 (3%) families (group 3). Recent evidence indicates that EVs involve copy number variation of common paralogous gene and pseudogene sequences that are polymorphic in the normal population and only become visible at the cytogenetic level when copy number is high. The average size of the deletions and duplications in all three groups of UBCAs was close to 10 Mb, and these UBCAs and EVs form the “Chromosome Anomaly Collection” at http://www.ngrl.org.uk/Wessex/collection. The continuum of severity associated with UBCAs and the variability of the genome at the sub-cytogenetic level make further close collaboration between medical and laboratory staff essential to distinguish clinically silent variation from pathogenic rearrangement.