Background

Anophthalmia and microphthalmia (A/M) are part of a spectrum of ocular anomalies that includes coloboma. These conditions have a combined incidence of about 2/10,000 births [1–4]. The A/M spectrum may be isolated or syndromic and is etiologically and clinically heterogeneous with substantial phenotypic overlap in many cases. This makes accurate syndrome identification crucial in order to provide families and individuals with appropriate recurrence risk counseling.

One-third of individuals with A/M have associated malformations [4]. Both genetic and environmental causes of A/M have been noted. Mutations causing a phenotype that segregates in an autosomal recessive pattern have been found in PAX6 [5], RAX [6] and CHX10 [7] (MIM 607108, 601881 and 142993, respectively). Mutations in SOX2 (MIM 184429) [8, 9] were recently described in individuals with A/M that segregates in an autosomal dominant pattern. Current estimates suggest that up to 15% of individuals with bilateral A/M have a mutation in SOX2 [9]. Bakrania et al., [10] also found individuals with deletions of SOX2. Mutations in OTX2 (MIM 600037) have been reported in up to 5% of individuals with A/M [9]. Mutations in BCOR (MIM 300485) have been seen in families with X-linked recessive inheritance of a phenotype strikingly similar to Lenz microphthalmia syndrome [11, 12]. Lenz microphthalmia has cardinal manifestations of microphthalmia, growth retardation and limb anomalies [13]. Other malformations thought to be typical of this disorder include microcephaly, cognitive impairment, protuberant ears, missing upper central incisors, syndactyly, and a cylindrical thorax with sloping shoulders. However, it is clear that Lenz microphthalmia is clinically variable and genetically heterogeneous and prior data show that phenotypes compatible with this description map both to Xp and Xq, the former caused by mutations in the BCOR gene (MIM 309800) which causes a phenotype that overlaps extensively with Lenz microphthalmia, and shares the X-linked inheritance pattern in two families [12, 14]. The Xq Lenz locus (MCOPS1 MIM 309800) has not yet been identified [15].

Here we report a patient with syndromic A/M that overlaps with Lenz microphthalmia syndrome (MIM 309800). Molecular evaluation of this patient showed a deletion of 16q, which includes the SALL1 gene, which causes Townes-Brocks syndrome (TBS, MIM 107480). This finding extends the notion of etiologic heterogeneity of syndromic microphthalmia and has implications for the approach to molecular diagnostics for this disorder.

Case Presentation

The patient was born at full term weighing 6 lb 2 oz (~25th centile) and length was 21 in (~75th centile). He was apneic at birth and was intubated and transferred to the NICU. He was weaned from the ventilator at 7 days of age with no complications. Bilateral clinical anophthalmia was diagnosed on clinical examination. A cranial CT scan at one day of age showed mild prominence of the left lateral ventricle. A small cavum septum pellucidum was also noted. A focal area of parenchymal lucency was noted in the left parietal parenchyma. This raised the possibility of an in utero ischemic event. A CT scan of the orbits at 6 days of age showed some orbital contents in the right eye with an abnormal lens and soft tissue overlying the globe, and his clinical diagnosis was modified to left clinical anophthalmia and right microphthalmia. The left globe had a small oval focus of soft tissue, but this tissue could not confidently be determined to be optic. The optic nerves appeared severely hypoplastic. A cranial MRI done at 7 days of age identified absent optic nerves with a small optic chiasm. Renal ultrasound at 2 days of age showed the kidneys measured 3.7 cm bilaterally (normal 4.0-6.0 cm). Echocardiogram identified a small ventriculoseptal defect and patent ductus arteriosus, which both resolved spontaneously. He was also noted to have an anteriorly placed anus with a fistula that was corrected surgically at 2 months of age.

Clinical Genetics evaluation was performed at 5 years 9 months by one of the authors. Additional history at the time of this evaluation showed he failed to meet multiple developmental milestones. His height was 102 cm (<<5th centile; 50th centile for 4 years), weight 29.5 pounds (<<5th centile; ~50th centile for 2 years 6 months), and head circumference was 50 cm (10-25th centile). In addition to the findings described above, he was found to have apparently small, overfolded, and cupped ears, long trunk with narrow shoulders, low set thumbs, and partial cutaneous syndactyly of toes 4-5. Peripheral blood karyotype was performed at the referring institution and was reported as normal, 46,XY. The patient was diagnosed with Lenz microphthalmia syndrome (OMIM 309800 MCOPS1). Complete sequencing of coding exons and flanking introns of BCOR (the single known Lenz syndrome gene) was normal (data not shown).

He was recently re-evaluated at the age of 13 years. He was treated with melatonin to help with his sleep cycle. On examination, his weight was 65 lbs (<<5th centile; ~50th centile for 9 year old), Height 57 in (~10th centile) and head circumference 52.5 cm (~10th centile). In addition to the previously mentioned findings, he had multiple nevi on his chest, back, and limbs. All were 0.3 cm or less. He has made developmental progress but is in special education. Sequencing of SOX2 was normal (data not shown).

The array platform was of the oligonucleotide CGH type with average genome spacing of 37 kb and targeted probes in candidate gene regions with 5-20 kb probe spacing (for details see http://www.genedx.com/site/genomedx). Hybridization was performed on DNA isolated from peripheral blood leukocytes as previously described [16]. This assay identified an approximately 5.6 Mb deletion of chromosome 16. The deletion likely has a minimum boundary size of Chr16:45,018,886 - 50,571,154 (Human Genome Build 36) and includes 26 known genes (Table 1). This deletion was not present in the parents. The deletion of SALL1 was confirmed by a MLPA assay using the SALSA MLPA kit P180 (MRC Holland, data not shown). Assessment of 6 STRP markers on chromosome 16 showed inheritance of alleles consistent with biologic parentage. The proband is apparently homo/hemizygous at marker D16S3396 (Chr16:49,749,809 - 49,750,146) and does not share an allele with the mother suggesting the deletion occurred on the maternal chromosome. We conclude that this variant is de novo in the proband.

Table 1 Genes in the deleted region
Full size table

Conclusions

We report a boy with microdeletion of 16q11.2q12.1 identified by array CGH. Microdeletion of 16q11.2q12.2 has been reported as an emerging microdeletion syndrome [17]. That report includes two patients with microdeletions of 16q that overlap the deletion seen in the patient reported here. All three patients share similar centromeric endpoints (as defined by the last centromeric oligonucleotide detected as a single-copy) and two of the three patients share similar telomeric endpoints. The deletion in Patient 1 in Ballif et al [17] extended an additional 1.4 Mb towards the telomere when compared to the present patient. Other patients with deletions of 16q11q12 have been reported with limited molecular data (Table 2).

Table 2 Summary of patients with proximal 16 deletions.
Full size table

The deleted region in the patient reported here includes 26 genes, including three recognized disease-causing genes (Table 1). This list includes the SALL1 gene, which when mutated causes Townes-Brocks syndrome [18]. Individuals with whole gene deletions have a milder phenotype than those who have dominant negative point mutations [19]. The classic phenotype for TBS includes imperforate anus, dysplastic ears with hearing impairment, and thumb malformations, although many affected patients do not have the typical phenotypic features. Less frequent manifestations of TBS include renal, heart, foot, and genitourinary anomalies, and mental retardation. This patient has a number of phenotypic manifestations that are not typical of TBS including A/M, cognitive impairment, borderline small head size, growth retardation, abnormal body habitus, and relatively normal thumbs. According to Kohlhase, the clinical diagnosis of TBS requires two of:

* Imperforate anus

* Dysplastic ears (overfolded superior helices, microtia)

* Typical thumb malformations (preaxial polydactyly, triphalangeal thumbs, hypoplastic thumbs) without shortening of the radius http://www.ncbi.nlm.nih.gov/sites/GeneTests/?db=GeneTests

The present patient only has one of these manifestations and therefore does not meet these stringent clinical criteria, though as noted above, the phenotype is acknowledged to be variable. One may speculate that an anterior-placed anus with a fistula is reminiscent of TBS, however it does not formally meet the criteria.

In contrast, the classical manifestations of Lenz microphthalmia are microphthalmia, growth retardation, and ear anomalies, all of which this patient manifested [13]. Clearly, this triad is not specific for Lenz microphthalmia and it is reported that > 90% have cognitive impairment [20], as seen in the present patient. A variety of skeletal and limb manifestations have been reported, although frequency data are not available.

Although some of the clinical features of the present patient overlap with the patients previously reported with 16q11.2 deletions and patients with TBS, the patient reported here had a major anomaly, A/M, which is clearly atypical for either of these molecular lesions. Microphthalmia has been reported in a pair of twins with TBS and a point mutation in SALL1 [21] and a patient from the pre-molecular era with a clinical phenotype consistent with TBS [22]. There are several possible hypotheses to explain this observation. First, it is possible that A/M is a very uncommon manifestation of TBS. Insofar as the twins reported by Botzenart et al are concerned, it is remarkable that although both twins had unilateral microphthalmia, neither their TBS-affected siblings nor their TBS-affected father had microphthalmia. It is reported that microphthalmia is two to five times more common in twins than in singletons [23, 24], which limits the utility of this case in assigning causality of the microphthalmia in this pair of twins to the SALL1 mutation. Second it is possible that the deleted region in the patient reported here includes a gene, that when deleted, leads to A/M but with reduced penetrance. The prior report of a single patient with the co-occurrence of TBS and microphthalmia is intriguing [22] in this regard. As this patient has not be characterized molecularly and is now deceased (A.R. Cooper, personal communication), we cannot determine if that child had microphthalmia and TBS because of a 16q deletion or that s/he had a point mutation in SALL1 and microphthalmia is a rare manifestation of TBS. The third possibility is that in the present patient and the patient reported by Fraser and Cooper, [22] the 16q11.2q12.2 deletion is in trans with another variant, likely to be uncommon, that leads to a null for a putative A/M gene in this interval. It will be important to analyze additional cases with deletions of this region to distinguish these two hypotheses.

The deletion region also includes the genes NOD2 associated with Blau syndrome (OMIM 266600) [25] and CYLD associated with Brooke-Spiegler syndrome (MIM 605041) [26]. Missense mutations in NOD2 cause Blau syndrome and an increased susceptibility to Crohn's disease (MIM 266600) [27]. Blau syndrome is inherited in an autosomal dominant pattern and characterized by early-onset granulomatous arthritis, rashes, and camptodactyly. Ocular manifestations include uveitis, glaucoma, retinal detachment, and cataract [28, 29]. No published reports of A/M and Blau syndrome were identified (negative results from PubMed using the two search strings: "anophth* AND Blau" and "microphth* AND Blau").

Heterozygous loss of function mutations in CYLD cause Brooke-Spiegler syndrome, which is characterized by a number of rare skin appendage tumors such as cylindroma, trichoepithelioma, and spiradenoma [30]. Early detection and aggressive treatment can prevent the significant deformity that can be caused by these tumors. These data suggest that the molecular delineation of this deletion in this patient present an opportunity to provide presymptomatic care that will significantly ameliorate the morbidity of this syndrome.

The patient reported here shares some clinical findings with the two previously reported patients [17] with deletion 16q11.2;12.2 namely; ear anomalies, toe abnormalities, hypotonia, and significant developmental delay. Interestingly, Patient 2 from Ballif et al., [17] was reported to have scattered pigmentary nevi, which was also found in the patient reported here.

We conclude from these data that the pleiotropic syndrome in this patient is caused by the deletion, which was detected by array CGH. The phenotype in this patient has some similarities to Townes-Brocks syndrome and to Lenz microphthalmia. The other suggestion from this report is that array CGH analysis should be considered in all patients with syndromic A/M. This is in addition to previous recommendations that such patients undergo chromosome analysis and SOX2 mutation analysis [9, 10].

Consent

The parent of this minor patient has given written consent for the report to be published. The patient was also consented to a clinical research protocol reviewed and approved by the NHGRI IRB.