Juvenile Moyamoya and Craniosynostosis in a Child with Deletion 1p32p31: Expanding the Clinical Spectrum of 1p32p31 Deletion Syndrome and a Review of the Literature

Moyamoya angiopathy (MA) is a rare cerebrovascular disorder characterised by the progressive occlusion of the internal carotid artery. Its aetiology is uncertain, but a genetic background seems likely, given the high MA familial rate. To investigate the aetiology of craniosynostosis and juvenile moyamoya in a 14-year-old male patient, we performed an array-comparative genomic hybridisation revealing a de novo interstitial deletion of 8.5 Mb in chromosome region 1p32p31. The deletion involved 34 protein coding genes, including NF1A, whose haploinsufficiency is indicated as being mainly responsible for the 1p32-p31 chromosome deletion syndrome phenotype (OMIM 613735). Our patient also has a deleted FOXD3 of the FOX gene family of transcription factors, which plays an important role in neural crest cell growth and differentiation. As the murine FOXD3−/− model shows craniofacial anomalies and abnormal common carotid artery morphology, it can be hypothesised that FOXD3 is involved in the pathogenesis of the craniofacial and vascular defects observed in our patient. In support of our assumption, we found in the literature another patient with a syndromic form of MA who had a deletion involving another FOX gene (FOXC1). In addition to describing the clinical history of our patient, we have reviewed all of the available literature concerning other patients with a 1p32p31 deletion, including cases from the Decipher database, and we have also reviewed the genetic disorders associated with MA, which is a useful guide for the diagnosis of syndromic form of MA.


Introduction
Moyamoya angiopathy (MA) is a cerebrovascular occlusive disorder characterised by bilateral progressive stenoses of the terminal portions of the internal carotid and the proximal anterior and Table 1.

Type 1 Neurofibromatosis [10] NF1
Noonan syndrome [11] PTPN11, SOS1, RAF1 and, more rarely: KRAS; NRAS; BRAF, and; MAP2K1 Costello syndrome [12] HRAS Alagille syndrome [13] JAG1, NOTCH2 Marfan syndrome [14] FBN1 Sickle cell disease [15] HBB Moyamoya disease-6 with achalasia [16] GUCY1A3 SAMHD1-related disorders [17] SAMHD1 MOPD2/Majewski syndrome [18] PCNT Seckel syndrome (microcephalic primordial dwarfism) [19] ATR, RBBP8, CENPJ, CEP152, CEP63, NIN The aetiology of moyamoya is still unknown, but familial aggregation and association studies suggest a genetic background, although the studies carried out so far have not revealed any significant locus associated with moyamoya other than RNF213, which seems to confer susceptibility to the disease in Asian countries [6,20,21]. Rare missense mutations in this gene have been significantly associated to MA European patients, particularly in childhood-onset and familial cases [2]. However, recently familial cases of MA and stereotyped facial dysmorphisms and early-onset achalasia, carrying respectively mutations in BRCC3 deubiquitinase and in GUCY1A3, the gene encoding the major nitric oxide receptor in vascular smooth muscle cells (vSMCs), have been reported [16,22]. We here describe the case of a 14-year-old patient with an intellectual disability (ID), craniosynostosis and juvenile moyamoya, carrying an 8.5 Mb de novo deletion in the chromosomal region 1p32.2p31.3, a newly recognized genomic disorder to be added to the growing number of genetic and genomic syndromes associated with MA.

Patient Presentation
Informed consent to the study and the publication of the results was given by the proband's parents.
The 14-year-old boy was born to healthy non-consanguineous parents. At birth, he showed complex craniofacial abnormalities ( Figure 1D), and 3D cerebral computed tomography (CT) revealed a hyperostotic metopic suture, bilateral aplasia of the frontal bones, a hypoplastic supraorbital ridge ( Figure 1A,B); additionally, magnetic resonance imaging (MRI) revealed a cyst in the septum pellucidum associated with mild ventricular dilatation. At the age of six months, he underwent neurosurgery and plastic reconstruction, including bi-frontal craniotomy, orbital osteotomies, and the remodelling of the frontal skull vault and the supraorbital margins ( Figure 1C,E). The post-operative course was uneventful. His motor development was normal (walking at the age of 12 months) but a speech delay required speech therapy. The results of a neuropsychiatric evaluation when he was four years old indicated mild intellectual disability (IQ 70). Follow-up MRI when he was eight years old confirmed the presence of the cyst in the septum pellucidum and revealed the presence of multiple punctate areas of altered signal, which were more pronounced in the right hemisphere ( Figure 2D), and an atrophic aspect of the corpus callosum. At the age of 10, he experienced an episode of right-sided weakness and a subsequent brain MRI showed acute ischemic stroke of the basal ganglia bilaterally, mainly on the left side (Figure 2A,B).
One year later, during a febrile episode, he experienced a tonic-clonic seizure involving the right upper limb. He repeated brain MRI associated with cerebral angiography, which failed to visualise the right internal carotid artery but showed a severe stenosis of the left internal carotid artery in association with dilated collateral middle cerebral artery (MCA) vessels having the typical "puff of smoke" aspect ( Figure 2C). A CT scan with acetazolamide showed a marked reduction in the time of transit of the contrast medium in bi-hemispheric cortical space that suggested reduced perfusion in both the Sylvian and the anterior circulation territories of the carotids bilaterally, with no signs of reserve. On the basis of the established criteria [23], the neuroradiological findings were diagnostic of MA.
At the age of 11, the patient underwent right encephalomiosynangiosis surgery, which was followed by an improvement in his school performance and the absence of any further episodes of stroke or stroke-like activity. No other surgical or pharmacological interventions have been carried out since.

Discussion
We here describe the first moyamoya patient carrying a deletion at 1p32p31. The rarely reported deletions at 1p32p31 are mainly associated with brain malformations (absence or hypoplasia of the corpus callosum, ventriculomegaly, and macrocephaly), urinary tract defects (vescicouretral reflux, urinary incontinence), facial dysmorphisms, and developmental delay. This condition has recently been annotated as "1p32p31 chromosome deletion syndrome" (OMIM 613735), and it has been suggested that the phenotype is the result of haploinsufficiency of the NFIA gene [26]. Only eight patients have been reported in the literature [27], but we found 11 further patients with 1p32p31 deletions involving the NFIA gene recorded in the Decipher v9.1 database (Wellcome Trust Sanger Institute: http://decipher.sanger.ac.uk) [28]. Of the 20 patients with 1p32p31 deletions, 10 (50%) show , showing normal red (specific probe for the subtelomeric 1q region) and green (specific probe for the 1p32 region) signals in the parents (target chromosomes encircled by yellow ovals), indicating the absence of the deletion of 1p32 (B,C), that instead is present in the patient (D), where the green signal is absent on the chromosome 1p32, indicated by the white arrow, the normal chromosome is also encircled by a yellow oval.

Discussion
We here describe the first moyamoya patient carrying a deletion at 1p32p31. The rarely reported deletions at 1p32p31 are mainly associated with brain malformations (absence or hypoplasia of the corpus callosum, ventriculomegaly, and macrocephaly), urinary tract defects (vescicouretral reflux, urinary incontinence), facial dysmorphisms, and developmental delay. This condition has recently been annotated as "1p32p31 chromosome deletion syndrome" (OMIM 613735), and it has been suggested that the phenotype is the result of haploinsufficiency of the NFIA gene [26]. Only eight patients have been reported in the literature [27], but we found 11 further patients with 1p32p31 deletions involving the NFIA gene recorded in the Decipher v9.1 database (Wellcome Trust Sanger Institute: http://decipher.sanger.ac.uk) [28]. Of the 20 patients with 1p32p31 deletions, 10 (50%) show the involvement of both the NFIA and FOXD3 genes; only one patient (Decipher database ID: 252422), for whom no clinical information is available, has the deletion of FOXD3 but not NFIA. The variability in the extent of these deletions is probably responsible for the phenotypic variability of the patients but consideration of all of the reported cases, including our patient, makes it possible to define a pattern of recurrent clinical signs and symptoms (Table 2). As MA may be due to the haploinsufficiency of a gene in the 1p32p31 region, and this could be a starting point for future investigations of its genetics aspects, we carefully evaluated all of the genes deleted in our patient. In particular the FOXD3 gene (Forkhead box D3-OMIM 611539) aroused our interest because of its function and its analogy to another FOX gene (FOXC1) that was found to be deleted in another seven-year-old patient with juvenile moyamoya, ID and craniofacial dysmorphisms [5].
The FOX gene family consists of a large number of genes encoding for transcription factors that play a critical role in the differentiation and development of neural crest cells (NCCs). FOXD3 is expressed in the epiblast during early embryogenesis and later in NCCs. Many studies of various animal models (early chicken embryos, mouse embryos, zebrafish) have highlighted the critical role of FOXD3 neural crest (NC) development as it participates in segregating the NC lineage from the neural epithelium. In mouse embryos, FOXD3 is expressed in pre-migratory and migratory NCCs, and is required for the maintenance of multipotent NC progenitors by self-renewing and repressing differentiation [29,30]. Notably, cephalic NCCs migrate to various regions in the head and neck where they contribute to the development of structures as diverse as the anterior skull base, the walls of the craniofacial arteries, the forebrain, and the face [31], which is in line with the pattern of craniofacial and vascular malformations shown by FOXD3 knock-out mice.
The FOXD3 tm2Lby /FOXD3 tm3Lby murine model (Mouse Genome Informatics: http://www. informatics.jax.org) [32] presents a morphologically abnormal neurocranium due to the reduced growth rate of cranial bones. It is widely known that craniosynostosis is characterised by a primary abnormality of skull growth, with the premature fusion of the cranial sutures that appear as a result of the difference in the rate of growth between the skull and developing brain. Furthermore, this animal model also shows a morphologically abnormal common carotid artery from which the abnormal internal carotid arteries of moyamoya arise. These findings suggest the possible role of FOXD3 haploinsufficiency in the pathogenesis of the craniosynostosis and MA of our patient but, as this is not supported by other case reports of patients with FOXD3 deletions, we can hypothesise the reduced penetrance of a single gene defect (clinical variability is very frequent when a transcription factor is involved in the pathogenesis of human diseases) [33], or a polygenic or multifactorial origin of the vascular defects. Intriguingly, NC disease could explain the association between craniosynostosis and MA in our patient and others as craniosynostosis is one of the most frequent comorbidities found in patients with syndromic moyamoya [34], and could also be helpful in clarifying the presence of the segmental vascular alterations.

Conclusions
Patients with 1p32p31 deletion do not show patognomonic features, but the general clinical presentation (developmental delay, facial dysmorphism, brain and/or genito-urinary malformations) is suggestive of a genomic disorder, prompting to the diagnosis by a-CGH or single nucleotide polymorphism (SNP)-array analyses. Our case expands the clinical spectrum of the diseases associated with 1p32-p31 deletions, further underlining the importance of genomic analyses of patients with MMS, and contextually suggests adding mutations of FOX gene family to the heterogeneous genetic causes of MA.