Our patient carries the smallest pure distal duplication of chromosome 5q (i.e. terminal part of 5q35.2 band and the entire 5q35.3 band). The phenotype of our patient showed significant overlap with Hunter-McAlpine syndrome, with such common features as short stature, heart defect, cryptorchidism, and craniofacial dysmorphism including prominent widened nasal bridge, almond-shaped eyes, thin vermillion of upper lip, and low-set dysplastic ears. In addition, our patient manifested less frequent symptoms (i.e. hypothyroidism) or even unique findings such as bilateral radial agenesis with absent thumbs, bilateral ulnar hypoplasia, abnormal biliary vesicle, and unilateral choroidal and retinal coloboma. Interestingly, absent thumbs were reported so far only in a single patient with an atypical copy number variation (CNV) on 5q, which was the interstitial triplication of the distal 5q segment encompassing MSX2 (6.56 Mb, coordinates according to HG18: 173897858–180456069). Since absent thumbs were never noted in cases with a distal 5q duplication, the authors hypothesized that the relatively severe limb malformation was due to the increased dosage of 5q copies
. Although caused solely by a pure duplication, the skeletal phenotype in our patient was more severe and involved bilateral radial aplasia with completely absent Ist digital rays (thumbs and Ist metacarpals), and ulnar hypoplasia with bowing. The defect seen in our patient as well as in the former case can be both categorized to radial ray deficiency spectrum. We therefore suggest that limb malformation presented by our index was a more severe manifestation of a common defect, namely radial ray deficiency. Of note, the chromosomal microduplication identified in our proband (terminal 5.4-5.6 Mb) did not encompass the MSX2 gene, which is located 1.1 Mb centromeric relative to its beginning. The MSX family, comprises MSX1 and MSX2 homeobox containing genes, which are important developmental regulators involved in the processes of limb, craniofacial, and ectoderm formation in vertebrates
. For example, MSX1 is essential for tooth and facial bone development and its mutations lead to Witkop syndrome also known as nail dysplasia with hypodontia
. Moreover, duplications of MSX2 cause Boston type craniosynostosis
[6, 12], whereas intragenic alterations or gene deletions result in parietal foramina, a disorder of deficient ossification of the skull
[13, 14]. It has been therefore postulated that duplications of MSX2 are responsible for craniosynostosis and brachydactyly in Hunter-McAlpine patients. Thus far, 5q distal trisomy has never been associated with the absence of digits. However, based on the observation of a 5q tetrasomy carrying patient, it has been hypothesized that multiple copies of 5q (including MSX2) result in a more severe skeletal anomaly such as absent thumbs. Clinical manifestation of our proband and the underlying genetic defect show that in the case of 5q gain, an extra copy (copies) of MSX2 is not necessary to give rise to a severe limb phenotype involving not only absent thumbs, but also bilateral radial aplasia and hypoplastic ulnae. Importantly, the size of the duplication in our case was smaller than that reported for other trisomy 5q patients and mapped 1.1 Mb telomeric to the MSX2. A plausible candidate gene causative for the limb malformation in our proband could be FGFR4. This gene is duplicated in our patient and maps around 1.3 Mb from the beginning of CNV. FGFR4 encodes for a protein, which is a type 4 receptor for fibroblast growth factors (FGFs). Members of FGF protein family are involved in FGF signalling pathway and play an important role during limb development. FGF4 along with FGF8 is secreted by apical ectodermal ridge (AER) which maintains the FGF10 signal and induces proliferation in the mesoderm
[15, 16]. For example, loss of both Fgf4 and Fgf8 in mice is thought to result in a reduction of the proliferation rate in distal mesenchyme, followed by downregulation of Fgf10 and premature degeneration of AER. Hence, in the absence of both Fgf4 and Fgf8, increased mesenchymal cell death results in a reduction in limb bud size
. So far there has been no report on radial agenesis and absent thumbs in other patients carrying 5q duplication encompassing FGFR4, suggesting that an extra copy of this gene is not sufficient to give rise to the limb phenotype. Noteworthy, FGFR4 maps about 1.3 Mb from the beginning of the duplication detected in our proband, which is relatively the closest position to the chromosomal breakage site for all known 5q duplications. Since both the limb malformation as well as the underlying genetic defect are unique in our patient we propose that a position effect resulting in altered long-range regulation of the FGFR4 (or possibly MSX2) may be the underlying patomechanism for limb malformation in both our and 5q tetrasomy patient. Alternatively, an extra copy and/or dysregulation of another gene may be responsible for radial agenesis.
The high frequency of congenital heart abnormalities in patients with a 5q trisomy was attributed to the altered dosage of one or two cardiac developmental genes, NKX2-5 and CSX1, both mapping to chromosome 5q34
[18, 19]. Interestingly, patient with the smallest pure distal duplication of 5q (encompassing terminal 6.4 of 5q35.2-q35.3) described to date in the literature, did not show any cardiac abnormality. This pointed to an observation that the direct duplication of 5q35.2-q35.3 may not lead to a cardiac phenotype
. In contrast, our patient carried even smaller distal duplication of 5q, however presented with a complex congenital heart defect including dextrocardia, dextroversion, and PFO. This may suggest that not only NKX2-5 or CSX1, but also other genes or regulatory elements located in distal 5q play an important role in the process of embryonic heart formation.
A duplication of the distal arm of chromosome 5q is known to be associated with short stature and microcephaly, and an increased dosage of NSD1 gene was proposed to be responsible for a combination of these two features
. Deletions and point mutations of the NSD1 gene cause Sotos syndrome with cerebral gigantism, overgrowth and macrocephaly
. It is theoretically possible, as suggested by Chen et al.
, that dosage changes (decrease or increase) of NSD1 lead to opposite phenotypes. Nonetheless, our patient had a duplication encompassing NSD1 but although short statured, he did not presented with a microcephaly. This may be explained by the incomplete penetrance of the candidate microcephaly gene.