Chromosome 2 interstitial deletion (del(2)(q14.1q21)) associated with connective tissue laxity and an attention deficit disorder
- K L Bakera,
- M I Reesb,c,
- P W Thompsona,
- R T Howelld,
- T R Colee,
- H E Hughesa,
- M Upadhyayaa,
- D Ravinea
- aMedical Genetics Service for Wales, University Hospital of Wales, Heath Park, Cardiff CF4 4XW, UK, bDepartment of Psychological Medicine and Medical Genetics, University of Wales College of Medicine, Cardiff, UK, cDepartment of Molecular Medicine, University of Auckland, Private Bag 92019, Auckland, New Zealand, dRegional Cytogenetics Centre, Southmead Hospital, Bristol, UK, eWest Midlands Regional Clinical Genetics Service, Birmingham, UK
- Dr Ravine
Editor—Reports of interstitial deletions involving the long arm of chromosome 2 are uncommon.1-10 Among these, there are only four which involve the region q14q21. We report a further case with a paternally derived de novo interstitial deletion of chromosome 2q14.1q21.
The proband was a male born by spontaneous vaginal delivery at term following an uneventful pregnancy. The parents are healthy, unrelated, and white. Birth weight was 4140 g (97th centile). Early childhood was complicated by hypotonia and recurrent sleep apnoea which resolved following adenoidectomy at 2 years of age. Otherwise, his medical history showed the normal range of intercurrent childhood viral illnesses. While childhood linear growth was rapid, during the second year there was considerable concern about poor weight gain. At 6 years of age he was noted to have a high, bossed forehead with a large head circumference (90-97th centile). A thoracolumbar kyphoscoliosis and a mild sternal depression was noted. He attended normal school although moderate learning difficulties were experienced. An attention deficit defect was identified and managed with the aid of methylphenidate hydrochloride. At 15 years of age, he was tall and thin (height 176 cm, 80th centile; weight 43.3 kg, 5th centile) with an associated moderate thoracolumbar kyphoscoliosis and pectus carinatum deformity. While the upper segment:lower segment ratio was 0.804, it was apparent that spinal height was somewhat reduced by the curvature of the scoliosis. The span measurement was 175 cm and head circumference was 56.0 cm (60th centile). He was a generally thinly muscled adolescent with little subcutaneous tissue. Some mild proximal upper limb weakness was detected and winging of the scapulae was evident, particularly on the right side. Examination of the musculature around the scapulae showed that the trapezius muscles were absent or possibly extremely hypoplastic. Ophthalmic examination showed normal fundi with no evidence of corneal or anterior chamber abnormalities, lens opacities, or dislocation. Other than findings of brisk lower limb reflexes, no other neuromuscular signs were present. No striae were evident. The forehead was high and the mandible was prominent owing to obtuse angulation. The ears were low set and dysplastic with some overfolding of the pinnae. The palate was high arched. Puberty had started, with stage 4 pubic hair development and a testicular volume of 25 ml. Cardiac echocardiography showed a mild degree of aortic root dilatation (aortic diameter 2.9 cm, calculated body surface area of 1.5 m2) but otherwise normal cardiac anatomy. Cytogenetic studies showed a small proximal interstitial deletion on the long arm of chromosome 2 (46,XY,del(2)(q14.1-21)). Before this result, no syndromic diagnosis was immediately apparent, although the occurrence of a high birth weight, a markedly prominent forehead in early childhood (fig 1A, B), and later development of mandibular prominence, hypotonia, and disproportionately long limbs had raised the question of Sotos syndrome.11 No bone age assessments were performed earlier in childhood and it was concluded that there were insufficient features present to confirm this diagnosis. Later photographs taken in mid childhood (fig 1C-F) were not supportive of the diagnosis.
In view of the phenotype observed in this child and the rarity of interstitial deletions within this region of chromosome 2, it was decided to delineate the breakpoints of the deletion further by both fluorescence in situ hybridisation and microsatellite analysis. FISH analysis with three YACS identified a region of deletion defined by YAC694-d-4. The deleted YAC contains marker D2S110 which provided an anchor point for the microsatellite work. Microsatellite markers (Genethon map) were selected and loss of heterozygosity analysis further defined the deletion within a genetic distance of approximately 10-12 cM and involves markers ranging from 2q14.1 to 2q21.1 (fig 2). The loss of alleles was paternal for all markers and the patient displayed only the maternal alleles for the deleted region.
Common clinical features among the few reports of proximal interstitial deletions of chromosome 2 involving the region q14.1q21 include developmental delay, microcephaly, defects of the corpus callosum, prominence of the forehead, low set and malformed ears, cardiac anomalies, and a tendency to recurrent, severe infections (table 1). Our case has some of the characteristics reported in these earlier cases, including a prominent forehead and low set and malformed ears. Like the case reported by Frydman etal,6 the birth weight of our case was unusually high. The older age of our patient, compared with those in earlier reports, provided an opportunity to document a more extensive medical history than has been recorded previously.
Weight gain was disproportionately poor from early childhood onwards, despite linear growth remaining above the 50th centile. In contrast to the earlier reports of marked microcephaly,1 6 head growth in the case reported here was proportionate to linear growth. Similarly, the occurrence of moderate learning difficulties is in contrast to the severe developmental delay reported in the case of Frydman et al.6
The presence of kyphoscoliosis with pectus carinatum deformity and mild aortic root dilatation suggests an abnormality involving connective tissue. Although these features occur commonly in Marfan syndrome, there were insufficient other physical features present to support this diagnosis. Nevertheless, there are some similarities, which raise the interesting possibility of another potential FBN gene within the chromosome 2 deleted region. Similarly, the associated occurrence of a moderate learning disability, together with an attention deficit defect, suggests that genes involved with higher level cerebral function are located within the deleted region. Occurrence of callosal defects among the other cases with interstitial deletions of this chromosomal region points to an abnormality of neuronal migration. Unfortunately, it was not possible to arrange neuroimaging in our own case. Nevertheless, the more moderate neurodeficit in our case contrasted with the earlier case reports of proximal 2q deletions and indicates that a variable CNS phenotype is associated with interstitial deletions in this region.
Haploinsufficiency involving one or more genes is a likely explanation for the observed phenotype. There are a number of genes within or close to the deleted region that are determinants for growth and development. Genes that could contribute to the phenotypes observed among those with interstitial deletions in this region includeGLI2 and the interleukin 1 genes and their receptors (IL1A, IL1B, IL1RN, and IL1R2). Gli2, a zinc finger transcription factor whose human homologueGLI2 is positioned at 2q14, has been shown to have overlapping functions with Gli3 in Shh signalling.12 It is of interest that the most proximal 2q interstitial deletion (q12-q14) has lower limb postaxial hexadactyly4 in keeping with the polydactyly observed withGLI3 mutations in Greig cephalopolysyndactyly syndrome, Pallister-Hall syndrome, and postaxial polydactyly type A syndrome.13 The occurrence of severe childhood infections, and unexplained febrile episodes as well as leukaemoid reactions among many of these cases with interstitial deletions in this proximal region of 2q is of interest. In this context, it is notable that our case did not have such a history. Candidate genes for these features include the interleukin-1 (IL-1) gene cluster and the IL-1 receptor gene cluster which map in the region of chromosome 2q12 to 2q13.14 15
Other genetic explanations for the observed phenotype include the unmasking of autosomal recessive disease owing to hemizygosity, or a parent of origin, or imprinting, effect contributing in part or in whole to the observed differences. While microsatellite analysis showed that the deletion in this case was of paternal origin, the parental origins of the previously reported deletions were not determined. Interestingly, the father of the case with an interstitial deletion of the region q13q21, reported by Davis etal,8 had a balanced translocation involving chromosome 2 as well as a pericentric inversion of chromosome 9 (46,XY,inv(9),t(2,7)(q32.2;p11). Other than this unusual report of co-occurrence of two chromosomal rearrangements involving the long arm of chromosome 2 in parent and child, there is insufficient evidence arising from earlier reports to determine whether a parent of origin effect occurs in this chromosomal region. Should such an effect occur, the paternal origin of the deletion reported here makes it possible to predict that maternal imprinting is involved. While reports of maternal disomy of chromosome 2 associated with phenotypic abnormalities is supportive of this concept,16 a report of maternal isodisomy 2 owing to the de novo inheritance of two isochromosomes for chromosome 2 in a normal healthy female, karyotyped because of recurrent spontaneous abortions,17 is strong evidence against.
To conclude, this case provides some additional insights into the effects of haploinsufficiency arising from a deletion of paternal origin in the proximal region of the long arm of chromosome 2.
The contribution of Dr T Hutchison, who was central to the patient's coordinated paediatric management, is gratefully acknowledged. We thank Dr N Rahman for helpful criticism and suggestions.