36th Annual David W. Smith Workshop on Malformations and Morphogenesis: Abstracts of the 2015 annual meeting

The 36th Annual David W Smith Workshop on Malformations and Morphogenesis occurred on August 14–19, 2015 at the Harbourtowne Conference Center in St Michaels Maryland. The annual conference, which honors the legacy of David Smith, widely recognized as the father of dysmorphology [Levenson, 2010], brought together over 120 clinicians and researchers interested in how human malformations arise and the underlying mechanisms of morphogenesis.

As is the annual tradition, the Workshop highlighted five themes besides mechanisms of morphogenesis: Rasopathies, Eye Malformations, Therapeutics, Prenatal Diagnosis, and Disorders of Sex Development.

The meeting was organized by Drs. Karen Gripp, Margaret Adams and Louanne Hudgins, coauthors of this Conference Report.

The Workshop was cosponsored by Nemours, Greenwood Genetic Center, the Centers for Disease Control and Prevention, the National Institute of Child Health and Development, the March of Dimes, and John Wiley and Sons, Publishers.

Attached herein are the abstracts presented at the annual Workshop.

RASOPATHIES I

The Many Faces of De Novo Mutations in the RAS Pathway

Anne Goriely

Weatherall Institute of Molecular Medicine, University of Oxford, England

Although mutations, the fundamental process by which genomic variation is acquired, are necessary to the evolution and success of a species, they are also at the origin of all genetic disease. Hence defining the mechanisms that control their occurrence is crucial to our basic understanding of genome biology, disease mechanisms and, ultimately, evolution. It is now possible to directly assess the germline mutation rate using whole-genome sequencing of family trios. These studies concur that 30–100 new point mutations are acquired at each generation, corresponding to a spontaneous mutation rate of 1.2 × 10⁻⁸ per nucleotide. However, this figure only represents a genome-wide average and it is well-recognized that the mutation rate fluctuates across the genome and between individuals.

While de novo mutations in the germline can cause inherited diseases, selection and clonal expansion of pathogenic mutations occurring in somatic tissues is associated with phenotypes such as tumour and tissue overgrowth; as these ‘acquired’ mutations are typically not inherited, they are traditionally seen as a distinct genetic process. However, the situation is different when such somatic mutations take place in the special context of the testis. In this case, clonal expansion of pathogenic mutations may lead not only to the formation of testicular tumours but also to elevated levels of specific mutations in sperm, and consequently an increased risk of transmission of pathogenic alleles to the next generation.

This process, which we termed selfish spermatogonial selection, has been described to explain the paternal age-effect and the high birth prevalence observed for some rare spontaneous disorders, such as Apert syndrome (FGFR2), achondroplasia (FGFR3) or Costello syndrome (HRAS). It relies on principles similar to oncogenesis to explain why these mutations occur spontaneously at levels up to 1000-fold higher than the background rate. The evidence – gathered originally through direct quantification of these ultra-rare pathogenic mutations in human sperm and testes – suggests that selfish mutations, although occurring rarely, confer a selective advantage to mutant spermatogonial stem cells, leading to their clonal expansion and progressive enrichment in sperm over time.

Our understanding of this process so far suggests that molecularly selfish selection relies on the activation of the growth factor receptor-RAS signalling pathway, which is a key regulator of stem cell homeostasis in the testis. As RAS is required in many different cellular contexts, we will discuss to which extend dysregulation of this pathway is likely to be relevant to the pathology of common disorders, including cancer predisposition and neurodevelopmental disorders, such as schizophrenia and autism - for which paternal age-effects have been described epidemiologically.

HRAS p.G12V Mutations in Costello Syndrome and Cancer

D.A. Stevenson¹, J. Swensen², A.M. Hartung³, I.E. Uriz³, M. Lapin³, J.C. Carey⁴, P. Yu⁵, C. Vaughn⁵, S.F. Dobrowolski⁶, M.R. Larsen³, A. Calhoun⁷, B.S. Andresen³

¹Division of Medical Genetics, Stanford University, Stanford, California

²Caris Life Sciences, Phoenix, Arizona

³Department of Biochemistry and Molecular Biology, University of Southern Denmark, Odense M, Denmark

⁴Division of Medical Genetics, University of Utah

⁵ARUP Laboratories, SLC, Utah

⁶Department of Pathology, Children's Hospital of Pittsburgh, Pittsburgh, Pennsylvania

⁷Division of Pediatric Genetics and Metabolism, University of Minnesota, Minneapolis, Minnesota

Costello syndrome (CS) is a disorder caused by activating germline mutations in HRAS. Most individuals with Costello syndrome harbor a c.34G>A (p.G12S) mutation. Somatic HRAS mutations are associated with cancer; p.G12S mutations have modest transforming activity, while p.G12V mutations (a frequent somatic amino acid change of HRAS in cancers) have the highest transforming activity of the codon 12 substitutions. Consistent with this, the few published reports of individuals with germline p.G12V mutations describe severe phenotypes associated with early lethality.

We evaluated a 12-year-old boy with an attenuated Costello phenotype. Sequencing of HRAS detected a novel germline dinucleotide substitution (c.35_36delinsTG), predicted to result in p.G12V. No evidence of mosaicism was observed in additional tissues. In silico analysis showed that HRAS exon 2 had a weak 3' splice site and suggested that c.35_36delinsTG abolished an exonic splicing enhancer (ESE) motif and created an exonic splicing silencer (ESS) motif. HRAS cDNA studies showed that c.35_36delinsTG resulted in increased skipping of exon 2, and indicated that reduced levels of p.G12V protein could explain the patient's mild phenotype.

The other reported germline nucleotide substitutions predicted to result in p.G12V have been c.35G>T, c.35_36delinsTT, and c.35_36delinsTA. We transfected cell lines with minigenes containing each of the p.G12V mutations, as well as the common p.G12S mutation. Only c.35_36delinsTG caused pronounced exon 2 skipping. Deletion of nucleotides c.32–37 also caused exon 2 skipping, indicating the presence of an ESE in this region. Testing in different splicing reporter minigenes showed that ESE strength was increased by c.35G>T (p.G12V), while c.35_36delinsTG inactivated splicing. RNA affinity purification followed by MS/MS showed that c.35_36delinsTG increased binding of hnRNPF/H splicing inhibitory proteins and reduced binding of the SRSF2 splicing stimulatory protein, consistent with the predictions of the in silico analysis. In contrast, c.35G>T increased binding of several splicing stimulatory SR proteins consistent with its observed effect on splicing. Replacement of wild-type sequence with hnRNPF/H- binding ESS motifs in the HRAS minigene confirmed that binding of hnRNPF/H resulted in exon 2 skipping. In addition, knock-down of SRSF2 caused skipping of exon 2, demonstrating the importance of the ESE element. The presence of a critical ESE at c.35 was confirmed by transfection of splice-switching 2-O-Methyl- Phosphorothioate antisense oligonucleotides (SSOs), which caused skipping of exon 2 in wild-type and mutant minigenes as well as from the endogenous HRAS gene in T24 and HepG2 cancer cells. Significantly, SSO-mediated-skipping of exon 2 caused reduced proliferation of cancer cells and significant cell death.

Costello syndrome and somatic cancer phenotypes are a result of the transforming potential of the mutant HRAS protein, but can also be impacted by the efficiency of exon 2 inclusion. This has potential implications for our understanding of the correlation between genotype and phenotype in diseases caused by HRAS mutations, and for development of new therapeutic approaches.

CNS Tumors in Noonan Syndrome: An Exome Analysis Approach to Studying Pathogenesis

C. Clericuzio MD¹, G.G. Pickett PhD¹, K. SantaCruz MD¹, J.S. Edwards PhD¹, G. McWilliams DO², and S.A. Ness PhD¹

¹University of New Mexico Health Sciences Center, Albuquerque, New Mexico

²University of Nevada School of Medicine, Reno, Nevada

CNS tumors have been reported only 18 times in Noonan syndrome (NS), over half in brief abstracts. In 14 cases, constitutional PTPN11 mutations were noted and in none of the cases was mutation analysis of the tumor described. Herein we report an 8 yo boy with typical NS and a common PTPN11 mutation diagnosed with multifocal dysembryoplastic neuroepithelial tumors (DNETs). To investigate the pathogenesis of the tumor, we have undertaken exome analysis of the DNET and the child's constitutional leukocytes to look for acquired genetic differences, with emphasis on RAS pathway genes. In addition, we review the literature on CNS tumors in NS, including those occurring in association with growth hormone (GH) treatment.

NS is the most common RASopathy, and the review by Kratz et al., 2011 confirmed the following NS-associated cancers: neuroblastoma, acute leukemia, glioma and rhabdomyosarcoma. These associations are biologically plausible since RAS pathway mutations are seen at the somatic level in all four of these cancers when they occur sporadically. To date the only published genetic analysis of a NS- associated cancer is by Karow et al., 2007 who reported clonal duplication of a germline PTPN11 mutation in acute leukemia blasts from a NS patient.

Study Methods: DNA samples from the tumor and circulating blood lymphocytes (germline) were sequenced on the Ion Proton sequencer using both the Ampliseq Exome kit (all 25,000 genes) and the Comprehensive Cancer Panel kit (409 cancer- relevant genes). We report here results of the first phase of this project, which compares the tumor and germline targeted DNA sequences of the 409 cancer-relevant genes.

Results: The PTPN11 constitutional mutation was found in both tumor and germline. Eleven altered genes were found unique to the tumor, but none are in the Ras/MAPK pathway. The panel includes BRAF, HRAS, KRAS, NRAS, NF1 and RAF1.

Discussion: We plan to pursue more comprehensive analysis of the exome data, recognizing that this first result is limited to the 409 gene panel. Regarding the occurrence of CNS tumors in NS, the addition of our patient brings to 19 the number of recognized affected individuals. Seven patients have DNETs: typically slow growing, solitary, non-invasive temporal lobe cortical lesions that manifest as seizures and constitute only a small percentage of CNS tumors. The relative rarity of DNETs coupled with the fact that our patient and 2 others had multifocal DNETs raises the possibility of a genetic predisposition to this tumor development in individuals with NS. Lastly, 5 individuals with NS and CNS tumors, including our patient, had received GH. One had GH stopped after 6 months due to tumor regrowth while another had no tumor regrowth after a year. Currently there is no clear evidence of enhanced tumor growth or development associated with GH use in general, but it is theoretically worrisome and our pediatric oncologist has advised against restarting GH given the progression of our patient's tumors over several years before the they were fully recognized. One author has recommended that children with NS have a brain MRI prior to receiving GH.

Constitutional NRAS G12R Mutation With Noonan Features, Extensive Nevi, and Rhabdomyosarcoma

Jacob S. Hogue, Mark J. Stephan

Department of Pediatrics, Madigan Army Medical Center, Tacoma, Washington

Constitutional mutations in NRAS are a rare cause of Noonan syndrome while somatic mutations are present in up to 30% of human cancers and also cause congenital melanocytic and epidermal nevi. We present the case of a woman with unique features to include embryonal rhabdomyosarcoma and diffuse nevi found to have a constitutional NRAS G12R mutation.

The patient was born large for gestational age with nuchal fullness and pleural effusions. She was hypotonic with poor suck and noted to have a tall forehead, midface retrusion, low posterior hairline, and telecanthus. She had mild pulmonic stenosis. She had poor feeding and growth. Global delays were appreciated in the first year. By that time she had hundreds of small nevi. She was found to have bilateral moderate sensorineural hearing loss at age 3. At 6 years of life she was found to have an embryonal rhabdomyosarcoma of the thumb which was treated successfully with surgery, radiation, and chemotherapy. She remains nonverbal. She walks well and can feed herself. She requires assistance with bathing, toileting, and dressing.

On exam at 19 years her height is 141.3 cm (−3.34 SD) and head circumference is 56.9 cm (+2.35 SD). Her forehead is tall and hair in the temporal regions is mildly sparse. Her neck is short and broad and posterior hairline is low. Her ears are mildly low set. Her upper eyelids are thick. Her nasal tip is broad and columella has a low attachment. Her philtrum is short. She has hundreds of brown nevi, mostly <5 mm in diameter, distributed throughout her body.

Chromosome analysis and microarray were normal. Noonan spectrum disorders panel testing (BRAF, HRAS, KRAS, MAP2K1, MAP2K2, PTPN11, RAF1, SHOC2, SOS1, CBL, NRAS) in blood revealed a heterozygous c.34G>C (p.G12R) mutation in the NRAS gene. She has no hematological abnormalities so the mutation was presumed to be constitutional. Parental testing confirmed that the mutation was de novo.

Constitutional mutations in the NRAS gene have been reported in a total of nine patients in the medical literature. The first reported patient had autoimmune lymphoproliferative syndrome, childhood leukemia, and adult lymphoma. Another had juvenile myelomonocytic leukemia along with a high forehead, epicanthal folds, long eyebrows, low nasal bridge, and low set ears. The remainder had typical features of Noonan syndrome.

Disease associated mutations in NRAS are activating missense mutations resulting in increased RAF/MEK/ERK signaling. Mutations at codons 12, 13, and 61 have the strongest effect and presumably for this reason are the ones seen recurrently as somatic mutations in cancer. Rhabdomyosarcoma has been reported in a few patients with Noonan syndrome, with SOS1 mutations identified in three. Rhabdomyosarcoma is also the most common cancer in Costello syndrome due to missense mutations at codon 12 of the HRAS gene. Somatic mutations in multiple RAS pathway genes to include NRAS have also been reported in sporadic embryonal rhabdomyosarcoma.

Constitutional mutations in NRAS at codon 12 have not previously been reported. Our patient represents a unique association of extensive melanocytic nevi, embryonic rhabdomyosarcoma, and severe intellectual disability along with more typical Noonan features in conjunction with mutation at this specific codon.

RASOPATHIES II

Effects of Mosaic RAS-Pathway Changes in Schimmelpenning Syndrome

Joseph Shieh, MD, PhD

Medical Genetics, Pediatrics, Institute for Human Genetics, University of California San Francisco Benioff Children's Hospital

Overactivity of the RAS pathway during development affects organ morphogenesis, but clinical effects may depend on precise mutation, timing during development, and the affected organs. Continued effects of RAS alteration may also contribute to risk for malignancy postnatally. We assessed patients seen at the UCSF RAS Pathway clinic from 2 years and identified two patients that demonstrate nevus sebaceous with additional organ growth abnormality suggesting Schimmelpenning-Feuerstein-Mims syndrome (MIM163200). We present here the clinical manifestations, course, test results, and potential difficulties in management of these patients.

Patient 1: We describe a KRAS alteration detected in skin tissue but not blood leading to Schimmelpenning syndrome. The proband is a 14 month old male who demonstrated prenatal multifocal tachycardia and pleural effusions at 32 weeks. After birth, extensive skin manifestations were noted with nevus sebaceous extending from the head to the lower extremity unilaterally. Clinical complications included hemihypertrophy, multifocal atrial tachycardia, focal seizures, and asymmetric bone abnormality. Interestingly, multifocal atrial tachycardia has also been reported in Costello syndrome (Gripp et al., 2007), and comparison of gene alterations will be discussed. Initial blood testing for RAS gene mutations by high-throughput sequencing in patient 1 was not revealing, however sequencing of the patient's affected scalp skin revealed a mosaic missense alteration in KRAS. Other patients with Schimmelpenning reveal KRAS and HRAS alterations (Groesser et al., 2012), however manifestations differ. Only 7% of overall individuals tested with nevus sebaceous were reported to have significant neurologic manifestations. These observations suggest the degree or distribution of mosaicism may play a role in clinical manifestations. Patient 2: The proband is a 4-year-old female who had an unremarkable prenatal history at birth had a nevus sebaceous of the scalp. An abdominal mass was identified at 2 years of age concerning for rhabdomyosarcoma, but pathology demonstrated a large venous malformation. The spinal cord demonstrated a small syrinx on MRI.

Ongoing management for such patients is not standardized. Patient 1 underwent quadrantectomy of the brain with resolution of seizures. Patient 2 underwent excision of the abdominal mass. The described risks of local tumors in Schimmelpenning include secondary skin tumors, however tumor risk in other organs and surveillance recommendations are unclear. Patient 1 underwent biopsy for an enhancing bone lesion of the humerus, which revealed nonspecific spindle cells that were FGF23 negative. Patient 2 is undergoing further abdominal imaging. Tumor surveillance plans will be discussed given the potential for selection and secondary changes that could drive growth. We conclude that Schimmelpenning manifests with growth abnormalities that can include multiple organs and that mosaic oncogene alterations likely underlie the spectrum of organ involvement. We aim to identify key mechanisms and management steps in presenting these patient and RAS family mutation comparisons.

Fetal Jugular Lymph Sacs— What Is the Significance?

Marjan M. Nezarati^1,2,3, H. Melanie Bedford¹, Karen Chong^2,3, Ants Toi⁴, David Chitayat^2,3

¹Genetics Program, North York General Hospital, Toronto, ON, Canada

²Department of Obstetrics and Gynecology, Toronto, ON, Canada

³The Prenatal Diagnosis and Medical Genetics Program, Toronto, ON, Canada

⁴Department of Medical Imaging Mount Sinai Hospital, Toronto, ON, Canada

Fetal jugular lymph sacs are accumulations of lymphatic fluid in the anterolateral region of the fetal neck. They may be isolated or occur in association with other structural anomalies and in some cases are presenting with an increased nuchal translucency or cystic hygroma. Their significance and the guidelines for the prenatal investigation when identified in the course of routine ultrasonography in pregnancy are still controversial. Our protocol for the investigation of these cases includes microarray analysis and DNA analysis for the Noonan syndrome panel. We report our experience with ten cases of prenatally diagnosed isolated fetal jugular neck cysts identified in the last 5 years (Table I).

In 10 cases identified by us in the last 5 years, 50% had a pathogenic mutation in a Noonan syndrome Panel, one had trisomy 21 and none had abnormality on microarray analysis (in the case with T21 we stopped the analysis following the QF-PCR results). However, all cases with abnormal results had NT > 3.5 mm or cystic hygroma. This study suggests that the identification of fetal jugular lymph sacs with an increased NT/cystic hygroma is associated with an increased incidence of Noonan syndrome and its related disorders and should instigate mutation analysis of the Noonan syndrome panel.

Hypomelanosis of Ito, Overgrowth, Macrocephaly, Intellectual Disability, and Distinct Dysmorphic Features: Role of Activating Mutations in the mTOR Pathway

Steven Klein¹, Hane Lee², Negar Ghahramani¹, Shahnaz Ghahremani³, Spencer Tong², Harry Vinters², Samuel P. Strom², Joshua L. Deignan², Wayne W. Grody^1,2, Stanley F. Nelson^1,2,5, UCLA Clinical Genomics Center, Eric Vilain^1,4,5, and Julian A. Martinez- Agosto^1,5,6

¹Department of Human Genetics, David Geffen School of Medicine at UCLA, Los Angeles, California

²Department of Pathology and Laboratory Medicine, David Geffen School of Medicine at UCLA, Los Angeles, California

³Department of Radiology, David Geffen School of Medicine at UCLA, Los Angeles, California

⁴Department of Urology, David Geffen School of Medicine at UCLA, Los Angeles, California

⁵Division of Medical Genetics, Department of Pediatrics, David Geffen School of Medicine at UCLA, Los Angeles, California

⁶Jonsson Cancer Center, David Geffen School of Medicine at UCLA, Los Angeles, California

The regulation of final organ and body size is a fundamental process in morphogenesis. Human overgrowth is characterized by enhanced growth parameters that exceed those for age, often accompanied by dysmorphic facial features, epidermal manifestations and risk of neoplasms. Here, we report cases of overgrowth that presented with macrosomia, macrocephaly, and intellectual disability with distinctive clinical features including hypomelanosis of Ito, hemimegalencephaly, seizures, and dysmorphic facial features. Whole exome sequencing (WES) identified de novo mutations in the FAT domain of mTOR. One case harbored a germline mutation, c.5395G>A (p.Glu1799Lys). A recurrent de novo mosaic heterozygous c.4448G>A (p.Cys1483Tyr) variant was present in several cases, with increased abundance of the mutant allele in hypopigmented skin versus blood. These two mutations activate mTOR kinase activity, each with distinct preferences for downstream substrates, which may account for the occurrence of hemimegalencephaly versus macrocephaly with somatic overgrowth. These cases demonstrate that mTOR activation is sufficient and required for the phenotypic characteristics exhibited by overgrowth syndromes that result from mutations in upstream components of the PI3K/AKT/mTOR pathway.

Furthermore, in the 1980s Happle, Graham and Flannery recognized Hypomelanosis of Ito as a cutaneous manifestation of a chimeric developmental abnormality due to mosaic genetic alterations. Happle assigned a Hypomelanosis of Ito locus to 9q33-qter, a region that includes the TSC1 gene. The presence of hypopigmented macules and Hypomelanosis of Ito in tuberous sclerosis complex supports this association, suggesting that hyperactivation of mTOR signaling may be involved. Coupled with the finding that inhibition of mTOR signaling promotes melanogenesis, we propose that Hypomelanosis of Ito results from a block of melanogenesis in mosaic melanocytes due to excessive mTOR signaling. Recent studies demonstrating the efficacy of rapamycin, an mTOR inhibitor, in improving pigmentation of hypomelanotic lesions supports this mechanism.

Germline mTOR Mutation in Two Brothers With Megalencephaly, Iris Coloboma and Autism

E. McPherson, MD, K. Rasmussen, MS, CGC

Marshfield Clinic Research Foundation and Marshfield Clinic Medical Genetics

We report a germline heterozygous mTOR mutation, c.5395G>A, identified during exome sequencing of two brothers with megalencephaly (OFC +5SD), iris coloboma, and an autistic spectrum disorder. Because it is present in both brothers, but not in either of the clinically normal parents, the mutation is presumed to be due to gonadal mosaicism. While the specific missense mutation, c.5395G>A, has previously been reported only in human cancer cells, mTOR is a plausible candidate to explain megalencephaly and autism. Because germline loss of function mutations in PTEN, which is an upstream inhibitor of the PIK3-AKT-mTOR pathway, are recognized as a cause of human disease including not only PTEN hamartoma syndrome, but also autism with megalencephaly, activating mutations in this pathway would be expected in children with autism and megalencephaly. Mice with constitutive mTOR hyperactivation have defective neuronal autophagy, synaptic pruning defects and autistic-like behaviors. The lack of germline changes in these genes in cohorts with childhood autism suggests they may result in different and presumably more severe phenotypes. Although somatic mutations of genes in this pathway are frequent in affected brain tissue, but not blood, of children with hemimegalencephaly (HME), only a single case with a mosaic mTOR mutation has been reported. Germline mTOR mutations have recently been reported in two unrelated children with intractable seizures and developmental delay, one of whom also had megalencephaly. Young age and severity of developmental delay precluded autism evaluation in these children.

While the relationship of the colobomas in our patients to the mTOR mutation remains speculative, it is clear that the PIK3-AKT-mTOR pathway affects eye development. Most notably, Rieger anomaly is a cardinal feature of SHORT syndrome, which is due to heterozygous PIK3R1 mutations. Cataracts are also a feature of the PTEN hamartoma syndrome.

The only other finding of possible significance for the intellectual disability/autism in our patients was a variant in OPHN1 hemizygous in both brothers and heterozygous in the mother. Typically OPHN1 causes x-linked intellectual disability/autism, but not megalencephaly or coloboma, and all reported patients had posterior fossa abnormalities which were not present in either brother, making it a less likely candidate.

Given the frequency of somatic mTOR mutations in cancer as well as the risks of tumors associated with the PTEN hamartoma syndrome, our patients are at potentially increased risk for malignancy. At 7 and 23 year old, they have many years of cancer risk ahead of them, and since the previously reported patients with germline mTOR mutations were even younger, there is no empiric risk data available. After discussion with the parents, we have recommended that the brothers undergo screening and follow-up according to the guidelines for PTEN hamartoma syndrome.

EYE MALFORMATIONS

The Taskforce for Neonatal Genomics: An Interdisciplinary Approach for Understanding the Architecture of Genetic Disease in Neonates and Young Children

Nicholas Katsanis, PhD

Duke University Durham, North Carolina

Genome-wide sequencing is emerging as a powerful tool as a first-pass diagnostic test. This has heightened the need for interpretive assays to determine the pathogenic potential of genetic variation. To address these challenges, and to capitalize on the opportunity to shorten the time to secure molecular diagnoses, we have created the Task Force for Neonatal Genomics at Duke University. The Task Force targets a uniquely vulnerable patient cohort: infants and neonates with structural anomalies that can a) be modeled in surrogate systems with high sensitivity and specificity; and b) is of value for clinical management, prognosis and potential therapeutic intervention. Our efforts harness the full spectrum of clinical, genetics and cellular biology expertise, including the use of transient model organisms (primarily zebrafish). I will discuss: 1) the interdisciplinary nature of our efforts; 2) our methodology for recruitment, data generation and analysis, and communication strategies between researchers and clinicians; 3) our analysis progress to date; and 4) our evolving approach to returning primary and secondary molecular findings to clinicians and family members. In phenotype-appropriate patients, we couple whole exome sequencing of trios, a multi-tiered bioinformatic prioritization strategy, and functional modeling of novel variants in physiologically relevant vertebrate and cell-based models to inform allele pathogenicity. This initiative provides an unprecedented model for communication across an interdisciplinary research/clinical team with the ultimate goal of responsible and timely integration of new genetic technologies into clinical care.

Oculofaciocardiodental Syndrome: Three Patients With Novel BCOR Mutations and Review of the Ocular Manifestations

Dorothy K. Grange, MD¹, Lawrence Tychsen, MD²

¹Division of Genetics and Genomic Medicine, Department of Pediatrics

²Department of Ophthalmology, Washington University School of Medicine

Oculofaciocardiodental (OFCD) syndrome is a rare X-linked dominant disorder characterized by congenital ocular abnormalities, cardiac defects, dental anomalies and characteristic facial features. OFCD is caused by truncating and frameshift mutations in BCOR and is allelic to Lenz microphthalmia syndrome. BCOR encodes the BCL-6 interacting corepressor protein, which plays an important role in both embryonic development and tumorigenesis.

Patient 1 is a 14-year-old girl with tetralogy of Fallot, SVT, AV and SA node dysfunction, dental anomalies, 2–3 toe syndactyly and camptodactyly. Ocular abnormalities include bilateral congenital cataracts, microphthalmia, microcornea, severe glaucoma and retinal detachment. She now has no vision in the right eye and has 20/250 vision in the left eye. She has delayed loss of deciduous teeth, abnormally shaped teeth and radioculomegaly. Her ears are small and cup-shaped with narrow ear canals and left hearing loss. She has left radioulnar synostosis. BCOR analysis showed a deletion in exon 6, c.2389_2390delGT (p.Val797CysfsX19) which causes a frameshift and is predicted to cause loss of normal protein function. Mother was shown to have the same mutation in BCOR and has a history of ASD and a unilateral cataract.

Patient 2 is a 9-year-old girl with ASD, VSD and coarctation of the aorta, left radioulnar synostosis, abnormal dentition, 2–3 toe syndactyly and camptodactyly of multiple toes. Ocular abnormalities include bilateral congenital cataracts, optic nerve hypoplasia, microcornea, retinal dysplasia, corneal clouding with iris adhesions and glaucoma. Her lenses were removed with placement of artificial lenses. She has little vision in the left eye and reduced vision in the right eye, and is being considered for a corneal transplant. She has abnormally-shaped teeth, dental decay, delayed loss of deciduous teeth and radiculomegaly. She has a broad nasal tip with a midline indentation and small, dysplastic ears. She was found to have a heterozygous c.4017dupC (p.Arg1340GlnfsX35) mutation in BCOR.

Patient 3 is a 5-year-old girl with complex congenital heart defects, including absence of the ventricular septum, D-malposition of the great arteries, severe pulmonic valve stenosis, ASD, and duplication of the superior vena cava. She has bilateral congenital cataracts, microcornea, nystagmus and left exotropia. Her lenses were removed with placement of artificial lenses. She has 2–3 toe syndactyly and hammer toes. She is missing an incisor and her teeth are abnormally shaped. She has a broad nasal tip with a midline indentation. She was found to have a c.2043_2047delTCACC (p.His682SerfsX56) mutation in BCOR which causes a frameshift with predicted premature protein truncation. Mother does not have this mutation.

OFCD syndrome is associated with complex ocular anomalies, as illustrated by these patients. Review of the literature indicates that congenital cataracts are present in 96% of individuals (69/72 reported cases). Microphthalmia and microcornea are also common (74%). Other eye abnormalities include ptosis (23%), lens dislocation (4%), iris coloboma (4%), nystagmus, strabismus, nasolacrimal duct stenosis or agenesis of puncta, optic nerve and retinal dysplasia, iris synechiae, retinal detachment and phthisis bulbi. Secondary glaucoma may cause progressive vision loss.

Mutations in a Novel CENP Gene May Cause a Recessive Fetal Malformation Phenotype—Cell Division Genes in Early Fetal Development

Isabel Filges¹, Elisabeth Bruder², Max Schubach⁴, Saskia Biskup⁵, Olav Lapaire³, Irene Hoesli³, Sven Schulzke⁶, Sevgi Tercanli⁷, Peter Miny¹

¹Medical Genetics, University Hospital Basel, Basel, Switzerland

²Pathology, University Hospital Basel, Basel, Switzerland

³Obstetrics and Gynecology, University Hospital Basel, Basel, Switzerland

⁴Medical Genetics, Charité Berlin, Germany

⁵CeGaT, Tübingen, Germany

⁶University of Basel Children's Hospital, Basel, Switzerland

⁷Centre for Fetal Medicine and Ultrasound, Basel, Switzerland

The clinical presentation of fetal malformation syndromes, particularly if lethal in utero, and their genetic causes often remain undefined for various reasons including our inability to make a specific diagnosis linking the phenotype to a recognized syndrome or condition. Lethal phenotypes, however, indicate a severe malfunction of early developmental processes, implying a major disruption of a crucial gene. Recessive inheritance may play an important role since a large proportion of animal models with spontaneous recessive mutations and mouse knockouts present with embryonic lethality.

We use whole exome sequencing (WES) for gene discovery in families with recurrent lethal fetal phenotypes suggesting autosomal recessive inheritance. We describe here the first human phenotype of a fetal MCA syndrome for which family-based WES identified truncating autosomal recessive mutations in a novel gene that segregated with this condition.

Common autopsy findings in both fetuses include general physical appearance, facial dysmorphisms, duodenal and multiple jejunal atresias, hyperrotation of the midgut and histopathological signs of intestinal and cardiac myopathy. Additional multiple organ anomalies including bilateral Peter's anomaly and cataract were present in the female and left preaxial polydactyly in the male fetus. We identified compound heterozygous truncating mutations in only one gene which encodes one of the centromere proteins (CENP) required for kinetochore function and chromosome segregation in mitosis. Its regulatory role in cell proliferation, vesicular transport and cell shape through its association with the microtubule network has been demonstrated. It is ubiquituously expressed. A cardiac specific mouse model was reported to present with cardiomyopathy. No human phenotype has been associated to mutations in this gene today.

We describe the fetal phenotype in detail which may overlap with Stromme syndrome, the gene's function and its interacting molecular network. This network also relates to KIF14, playing a role in cytokinesis, for which we recently described the lethal phenotype, and to SHH and ciliary pathways, implied in lethal fetal disorders as well. We will discuss the notion that genes involved in cell division functions may be considered important candidates when elucidating early human maldevelopment.

Broad Spectrum of Phenotypes Associated With Defects of X-Linked N-Terminal Acetyltransferase Deficiency in Males and Females

Bernt Popp¹, Svein I Støve^2,3, Sabine Endele¹, Line M Myklebust², Juliane Hoyer¹, Heinrich Sticht⁴, Silvia Azzarello-Burri⁵, Anita Rauch⁵, Thomas Arnesen^2,3, André Reis¹

¹Institute of Human Genetics, Universität Erlangen-Nürnberg, Erlangen, Germany

²Department of Molecular Biology, University of Bergen, Bergen, Norway

³Department of Surgery, Haukeland University Hospital, Bergen, Norway

⁴Institute of Biochemistry, Universität Erlangen-Nürnberg, Erlangen, Germany

⁵Institute of Medical Genetics, University of Zurich, Schlieren, Switzerland

Human Naa10 deficiency in males has been associated with the lethal X-linked disorder Ogden syndrome (MIM 300855), characterised by an aged appearance, craniofacial anomalies, hypotonia, global developmental delays, cryptorchidism, and cardiac arrhythmias (Rope et al. AJHG 2011). A single hypomorphic missense variant p.Ser37Pro in eight affected males from two different families resulted in death during infancy while female carriers were unaffected. Surprisingly, recent findings of NAA10 mutations also in male and female patients with intellectual disability and global delay as well as in Lenz microophthalmia syndrome (MIM 309800) indicate a much broader phenotypic spectrum.

By trio whole-exome sequencing we identified de novo missense variants in two unrelated individuals, p.Arg116Trp in a boy and p.Val107Phe in a girl, with severe global developmental delay but without any major dysmorphism. The most noticeable overlap was severe global developmental delay with postnatal growth failure, behavioural anomalies, truncal hypotonia with hypertonia of the extremities and some minor facial anomalies. Both de novo variants were predicted to be deleterious. Interestingly, Esmailpour et al (JMG 2014) reported the identification of a splice-donor variant (c.471+2T>A) in the NAA10 gene in a family with Lenz microphthalmia syndrome. The four affected males presented with congenital bilateral anophthalmia, postnatal growth failure, skeletal anomalies, hypotonia and moderate to severe mental retardation with a high degree of intra- familial variation. Three heterozygous carrier females had mild symptoms with abnormally shaped ears, syndactyly of the toes, short terminal phalanges and short stature. This variant resulted in no detectable normal NAA10 gene products but in different aberrant transcripts and low expression of a truncated protein.

Naa10, the N-alpha-acetyltransferase 10, is the catalytic subunit of the N-alpha- acetyltransferase protein complex NatA. The family of NAT enzyme complexes (NatA-F) is involved in multiple cellular processes and acetyl-coenzyme-A dependent N-alpha- acetylation is the most common protein modification in humans. NatA is highly expressed in the developing brain of mice embryos and shows a lower yet uniform and ubiquitous expression in adult mice. Complete knockout is lethal in several animal models. In an attempt to explain phenotypic variability, we used in vitro N-terminal acetylation assays which suggested that the severity of the phenotype correlates with the remaining catalytic activity. The variant in the Ogden syndrome patients exhibited a lower activity than the one seen in the boy with intellectual disability, while the variant in the girl was the most severe exhibiting only residual activity in the acetylation assays used. We propose that N-terminal acetyltransferase deficiency is clinically heterogeneous, can manifest both in boys and girls and the overall catalytic activity determines the phenotypic severity.

Autosomal Recessive Mutations in the VWA Domain of MED25, a Member of the Mediator Complex, Are Associated With a Recognizable Syndrome With Ocular and Cognitive Involvement

A.M. Innes, MD, S.J. Mosca, S.J. Haggstrom, F.P. Bernier, J.S. Parboosingh, R.E. Lamont

Department of Medical Genetics and ACHRI, Cumming School of Medicine, University of Calgary, Calgary, Canada

Accelerated by whole exome sequencing (WES), the genetic basis for over half of the rare Mendelian disorders is now known, including most ‘classic’ malformation syndromes. The majority of unsolved disorders affect few patients, and determining pathogenicity requires replication in other families and/or functional validation. Active or unanswered questions include how to prove pathogenicity of genes causing the remaining Mendelian disorders with few affected individuals, and how many ‘unique’ genes will explain the remaining disorders.

We follow a 10-year-old girl, born to consanguineous parents, with early onset cataracts, epilepsy, intellectual disability (ID), dysmorphic features (nevus flammeus, epicanthal folds, and sparse hair), constipation, club feet and atrial septal defect. She was enrolled in a research study incorporating snp homozygosity mapping and subsequently WES to determine the cause of her disorder. We identified a rare homozygous missense change (p.Asp126Val) affecting a highly conserved amino acid in the von Willebrand factor type A (VWA) domain of MED25. At the time of the discovery MED25 was only linked to a form of Charcot-Marie-Tooth (CMT2B2) in one family that has not been replicated to date; however, other members of the Mediator complex were known to be associated with syndromic and non-syndromic ID. The parents now have a 2nd baby with similar features who is also homozygous for the MED25 variant, while two unaffected sisters are not homozygous.

This month, two publications implicated MED25 in recessive ID, both in single extended founder pedigrees. Figuerierdo et al (JMG 2015) reported seven Brazilian individuals with ID and dysmorphic features who had a homozygous p.Arg140Trp mutation. Subsequently, Basel- Vanagaite et al (Hum Genet 2015) published seven Israeli-Arab patients with a phenotype highly reminiscent of our family (early onset cataracts, ID, epilepsy, nevus flammeus, sparse hair, constipation and congenital heart disease) who had a p.Tyr39Cys mutation. Independent identification of these three separate pedigrees, two with identical phenotypes and all with mutations in the VWA domain (responsible for MED25 recruitment to the Mediator complex), supports the role of MED25 in syndromic ID. As a result MED25 can be added to MED12, MED17 and MED23 as members of the Mediator complex-a transcriptional co-activator- associated with ID. The phenotypes of patients with MED25 mutations defining a novel syndrome, particularly our family and the very similar individuals reported by Basel-Vanagaite will be reviewed in detail.

NEUROLOGY I

A New X-Linked Syndrome With Intellectual Disability, White Matter Anomalies, and a Distinctive Phenotype

Alan Rope¹, A. Micheil Innes², P.Y. Billie Au², Chris S. Smith², Paul Mark³, Gholson J. Lyon⁴

¹Department of Medical Genetics, Northwest Kaiser Permanente, Portland, Oregon

²Department of Medical Genetics, Alberta Children's Hospital Research Institute, University of Calgary, Alberta

³Department of Medical Genetics, Spectrum Health, Grand Rapids, Michigan

⁴Stanley Institute for Cognitive Genomics, Cold Spring Harbor Laboratory, Stony Brook, New York

A proband and his male sibling were ascertained more than 10 years ago when they presented with global developmental delays, severe intellectual disabilities, autistic spectrum behaviors, attention deficit with hyperactivity, and distinctive facial features including: a broad, upturned nose, sagging cheeks, downward sloping palpebral fissures, prominent periorbital ridges, a high-arched palate, and prominent ears. Other features included microcephaly, cerebral atrophy, agenesis of the corpus callosum, growth retardation, hypotonia and a very peculiar surface anatomy of the sacrococcygeal region. Until recently, the only objective evidence for the proposed pattern of inheritance had been the 100% skewing of X-inactivation in their mother. With the advent of whole genome sequencing and implementation of a variety of pipeline analyses, we concluded that this presentation was attributable to a mutation of TAF1, c.4010T>C, p.Ile1337Thr.

TAFs may participate in basal transcription, serve as coactivators, function in promoter recognition or modify general transcription factors (GTFs) to facilitate complex assembly and transcription initiation. TAF1 encodes the largest subunit of TFIID. This subunit binds to core promoter sequences encompassing the transcription start site. It also binds to activators and other transcriptional regulators, and these interactions affect the rate of transcription initiation from binding to these regions.

Mutations of TAF1 have been previously linked to a form of dystonia and was identified as a gene of interest from a large European collaborative study utilizing X-exome sequencing of 405 families affected by idiopathic intellectual disability.

For the purpose of understanding the relevance of this gene and whether or not the phenotype was reproducible in other genetic backgrounds, we actively sought out collaborations with geneticists who had also ascertained cases. Through this effort, we have identified at least three other families affected by mutations in TAF1, and another affected by a duplication of Xq13.1 that encompasses the region occupied by TAF1.

The overall reproducibility of the specific phenotype is somewhat inconclusive. The proband and his brother have a remarkably similar facial appearance as the individuals affected by the duplication of Xq13.1 and several affected members in an extended family from Europe. But the case identified by our collaborators from University of Calgary did not demonstrate the same recognizable facial gestalt. That being said, the very distinctive surface anatomy of the sacrococcygeal region was seen in all but the dup Xq13.1 case. We have entertained the hypothesis that this caudal variation is related to anomalous neural tube closure, but look forward to results from ongoing zebrafish studies involving expression this genetic variant.

Lissencephaly: Expanded Imaging- and Molecular Pathway-Based Classification With a New Category With Autosomal Recessive Inheritance

N. Di Donato^1,2, S. Chiari³, C. Adams¹, S. Collins¹, R. Guerrini³, W.B. Dobyns^1,4

¹Center for Integrative Brain research, Seattle Children's Research Institute, Seattle, Washington

²Institut fuer Klinische Genetik, TU Dresden, Dresden Germany

³Pediatric Neurology Unit, Children's Hospital A. Meyer, University of Firenze, Firenze, Italy

⁴Department of Pediatrics, University of Washington, Seattle, Washington

Introduction: Lissencephaly (“smooth brain,” LIS) is the classic and most severe neuronal migration disorder, and is characterized by absent or abnormally wide gyri in combination with a thick cortex. Further observations of rare subtypes of LIS and rapidly increasing knowledge about underlying genetic causes requires revision and modification of the existing clinical classification.

Materials and Methods: We reviewed brain scans, clinical and molecular findings of 129 LIS patients from 119 families collected in the Dobyns Lab since 2009, with brain scans of another 100 from the Guerrini Lab still under review. We grouped the patients based on severity (LIS grade 1–6), gradient (anterior or posterior predominant, or diffuse), and presence of associated non- cortical brain malformations.

Results: Posterior predominant LIS grade 3 (27%) was the most common type, with most patients having deletions or intragenic mutations of the LIS1 (PAFAH1B1) gene. This was followed by diffuse LIS (17%) most often caused by mutations of LIS1 or DCX (which also causes diffuse subcortical band heterotopia). All other LIS types were observed at frequencies of 2%-9% with anterior predominant LIS forms significantly less common than posterior predominant forms. The tubulinopathies are clinically recognizable and account for 10% of the patients (grade 4 LIS with subtle mixed atypical microgyria). This is the only group (other than severe diffuse LIS) lacking a consistent anterior or posterior gradient. Besides classic LIS with markedly thick (12–20 mm) cortex, we recognized a new group of “thin” LIS (5–7 mm cortex) characterized by anterior predominance, reduced number of gyri, shallow sulcal depth, and usually an undulating appearance of the gyral pattern. When associated with severe cerebellar and hippocampal hypoplasia, this variant is caused by mutations in RELN or (when less severe) VLDLR. We identified 13 probands from 8 families with this new form of LIS. None had mutations in one of the known LIS genes. Based on recurrence in sibs observed in 5 of 8 families and a high percentage of homozygosity on SNP arrays in others, we propose that all or most of the disorders in this new group have autosomal recessive inheritance.

We will present updated diagnostic criteria and current diagnostic yield for each of the major LIS groups.

The Contribution of De Novo Coding Mutations to Cerebellar Malformations

K.A. Aldinger¹, A.E. Timms¹, M. Mehaffey², Brandi Bratrude¹, K.J. Millen¹, D. Doherty²,W.B. Dobyns^1,2

¹Seattle Children's Research Institute, Seattle, Washington

²Division of Medical Genetics, Department of Pediatrics, University of Washington, Seattle, Washington

Cerebellar malformations, especially Dandy-Walker malformation (DWM) and cerebellar hypoplasia, are common birth defects that cause significant intellectual, motor and behavioral disabilities. These developmental brain disorders are characterized by marked phenotypic and genetic heterogeneity. However, a fundamental question is whether specific cerebellar malformations represent causally heterogeneous disorders in which multiple genetic and/or environmental risk factors disrupt common molecular pathways. Our prior studies showed a smaller contribution of de novo copy number variants to cerebellar malformations (<2%) compared to agenesis of the corpus callosum, intellectual disability, or autism. Based on these data, we hypothesize that cerebellar malformations are caused by (1) a higher proportion of intragenic mutations, (2) a higher proportion of non-genetic causes, or (3) both. We tested the first hypothesis using whole exome sequencing, which has become an efficient strategy for identifying rare, disease- causing mutations. Recent studies using whole exome sequencing have further established that de novo mutations are observed in a significant fraction of individuals with common developmental disorders including severe early-onset epilepsy, intellectual disability, autism spectrum disorder, and schizophrenia.

We performed whole exome sequencing in 74 child-parent trios and five singletons with cerebellar malformations including DWM (n = 42), cerebellar vermis hypoplasia without additional features of DWM (n=19), and diffuse cerebellar hypoplasia (n = 5). A small subset (n = 9) had additional brain or other congenital malformations. We identified 73 de novo coding changes in 62 genes predicted to be pathogenic among 58 affected individuals (73%). Compelling mutations were identified in several known disease genes (SYNGAP1, CHD7, VLDLR, STXBP1, TUBA1A, DDX3X, CASK, TMLHE), as well as in numerous genes not previously implicated in neurodevelopmental disorders. Several of these genes are expressed in the cerebellum during early fetal development and are involved in transcriptional regulation and vascular development.

Our data represent the first report on the rate of deleterious de novo mutations in cerebellar malformations. To our knowledge, this is the largest whole exome sequencing study of cerebellar malformations. Our data additionally highlight the critical and divergent roles of many genes associated with epilepsy, intellectual disability, and developmental brain disorders, including cerebellar disorders.

Unilateral Cerebellar Absence/Hypoplasia: Further Evidence Supporting an Acquired Origin

Alasdair G.W. Hunter, MD^1,2, Ashley Smith, GC¹, Naeem Khan, MD³, Michiel Van den Hof, MD⁴

¹Division of Genetics, IWK Health Centre, Halifax

²Department of Genetics CHEO, Ottawa, Canada

³Department of Radiology, IWK Health Centre, Halifax

⁴Department of Obstetrics and Gynecology, IWK Health Centre, Halifax

Unilateral cerebellar absence/hypoplasia (UAH) is an uncommon, but well described, condition that should be distinguished from crossed cerebellar diaschisis, which is secondary to contralateral cortical atrophy. Involvement of the hemisphere ranges from mild hypoplasia to complete absence, and the ispsilateral vermis can vary from normal to absent. Most reported diagnoses have been in individuals presenting postnatally with neurological symptoms such as headache and hyperekplexia, or showing developmental delay, the latter often being associated with complex medical histories and/or associated physical findings. Some authors have concluded that the cerebellar anomaly accounted for the patients' clinical findings; others considered the cerebellar findings unrelated to the clinical presentation.

We report on the pregnancy of a diabetic mother, whose BMI was 38.8 and whose U/S at 155/7 wks was normal. IUGR was noted at 21 6/7 wks and concern was raised about the cerebellar shape. US was repeated frequently, but complicated by reduced amniotic fluid and maternal size. Antenatal concern regarding the cerebellum continued until birth at 335/7 wks. Marked right UAH was confirmed. The findings will be presented as well as the child's overall favourable progress to 6 1/2 years. This case supports the often prenatal, likely vascular, origin of UAC and confirms the view that such early prenatal onset cases can result in minimal impact on development and physical function.

There is support for UAH being an acquired lesion. To date there has been no familial recurrence. Ramaekers et al.¹ reported that six of seven patients with severe isolated UAH had severe pre/perinatal hypoxia; three of five milder cases had a history of cerebellitis or cerebellar strokes, and an adult who was reported with progressive cerebellar signs and cranial nerve palsies had a unilateral porencephalic cavity. The compelling case from Robins et al.,² was of a woman who had a 24wk ultrasound (US) for non-specific abdominal pain and whose fetus was noted to have a 2cm right cerebellar hemispheric mass that expanded to a 5cm cystic lesion by 29wks. UAH was noted on perinatal MRI but development was normal at 18 months.

¹Ramaekers et al. Brain 120:1739, 1997; ²Robins et al. Prenat Diag 18:173, 1998.

Phenotypic and Molecular Insights Into CASK-Related Disorders in Males

U. Moog¹, T. Bierhals², K. Brand², J. Bautsch², S. Biskup³, T. Brune⁴, J. Denecke⁵, C.E. de Die-Smulders⁶, C. Evers¹, M. Hempel², M. Henneke⁷, H. Yntema⁸, B. Menten⁹, J. Pietz¹⁰, R. Pfundt⁸, J. Schmidtke¹¹, D. Steinemann¹², C.T. Stumpel⁶, L. Van Maldergem¹³, K. Kutsche²

¹Institute of Human Genetics, Heidelberg University, Heidelberg, Germany

²Institute of Human Genetics, Germany

³CeGaT GmbH, Tübingen, Germany

⁴Universitäts- kinderklinik, Universitätsklinikum Magdeburg, Magdeburg, Germany

⁵Klinik und Poliklinik für Kinder- und Jugendmedizin, University Medical Center Hamburg- Eppendorf, Hamburg, Germany

⁶Department of Clinical Genetics, Maastricht UMC, Maastricht, The Netherlands

⁷Klinik für Kinder- und Jugend-medizin, Universitätsmedizin Göttingen, Göttingen, Germany

⁸Department of Human Genetics, Radboud UMC, Nijmegen, The Netherlands

⁹Center for Medical Genetics, Ghent University, Ghent, Belgium

¹⁰Section of Neuropediatrics, Center for Child and Adolescent Medicine, Heidelberg, Germany

¹¹Institut für Humangenetik, Hannover, Germany

¹²Institut für Zell- und Molekular-pathologie, Medizinische Hochschule Hannover, Hannover, Germany

¹³Centre de Génétique Humaine, Université de Franche-Comté, Besançon, France

Heterozygous loss-of-function mutations in the X-linked CASK gene cause progressive microcephaly with pontine and cerebellar hypoplasia (MICPCH) and severe intellectual disability (ID) in females. Different CASK mutations have also been reported in males with phenotypes ranging from nonsyndromic ID to Ohtahara syndrome with cerebellar hypoplasia. However, a systematic evaluation of the phenotypic spectrum in males in relation to their genotype has not been performed to date.

We identified a CASK alteration in eight novel unrelated male patients by targeted sequencing, copy number analysis (MLPA and/or FISH), array CGH and whole exome sequencing, and performed transcript (RT-PCR followed by sequencing) and protein analysis (immunoblotting). The clinical data of these eight patients and all CASK-mutation positive males reported previously were reviewed and correlated with available molecular data.

CASK alterations in the eight patients include one nonsense mutation, one 5-bp deletion, one mutation of the start codon, and five partial gene deletions or duplications; seven were de novo, including three somatic mosaics, and one was familial. In three subjects, specific mRNA junction fragments indicated in tandem duplication of exons that disrupted the integrity of CASK. The 5-bp deletion resulted in multiple aberrant CASK mRNAs. In fibroblasts from patients with a CASK loss-of-function germline mutation, no CASK protein could be detected, while individuals who were mosaic for a severe CASK mutation or carried a hypomorphic mutation still showed detectable amount of wild-type protein.

Based on the findings in the novel patients and all CASK-mutation positive males reported previously, CASK alterations in males are associated with 3 distinguishable although overlapping phenotypes: (i) MICPCH with severe epileptic encephalopathy caused by hemizygous loss-of-function mutations, (ii) MICPCH associated with inactivating alterations in the mosaic state or a partly penetrant mutation, and (iii) syndromic/nonsyndromic mild to severe ID with or without nystagmus caused by missense and splice mutations that leave the CASK protein intact but likely alter its function or reduce the amount of normal protein. Our findings facilitate focused testing of the CASK gene and interpreting sequence variants identified by next-generation sequencing in cases with a phenotype resembling either of the three groups.

POSTER PRESENTATIONS

Variable Expressivity of the Scalp-Ear-Nipple Syndrome or a Novel Variant of This Rare Phenotype: Report of a 2-Generation Family

John C. Carey¹, Janice C. Palumbos¹, Chad Cox¹, Michael J. Bamshad²

¹Department of Pediatrics, University of Utah, Salt Lake City, Utah

²UW Center for Mendelian Genomics, Seattle, Washington

In 1978 Finlay and Marks reported a multi-generational kindred with an apparently previously unrecognized autosomal dominant disorder consisting of cutis aplasia of the scalp, minor anomalies of the external ears, and breast abnormalities. The syndrome, now referred to as the scalp-ear-nipple (SEN) or Finlay-Marks syndrome (OMIM 181270), includes other less consistent features such as syndactyly of the digits and renal malformations. Since the original article, an additional 14 reports have been published in the medical literature, mostly of sporadic cases, but one is of a second large kindred. In 2013 The University of Washington Center for Mendelian Genomics and colleagues around the world identified a gene causing SEN, i.e., KCTD1.

We report on an additional two generational family with the typical cutis aplasia and ear findings, but lacking the more typical mammary abnormalities and having additional unreported features. The proband was originally evaluated at 7 months of age for multiple congenital anomalies, including a distinctive cutis aplasia, an abdominal hernia (Spigelian type), and a Morgagni diaphragmatic defect. The boy also had laterally prominent and protruding auricles, delay in weight gain, and an expressive speech lag. Family history showed that his father had the identical cutis aplasia and also a branchial sinus of his neck and a unilateral accessory nipple. An older sister of the proband had similar external ears, an ear tag, and renal hypoplasia, but lacked the cutis aplasia. The father and sister both had normal intellectual development with the sister having mild speech delays as an infant.

The combination of the distinctive cutis aplasia accompanied by ear variations in two of the three family members is suggestive of SEN. However, none of the three patients has the breast hypoplasia/aplasia and, in fact, the father has an accessory nipple. This finding of a supernumerary nipple in a syndrome that also has breast hypoplasia is seen in the ulnar-mammary syndrome and may represent the spectrum of developmental alteration. There is one reported patient in the literature with cutis aplasia and a Spigelian hernia (like the proband), but the report lacks detailed information and precludes knowing whether this case has SEN. Thus, the abdominal and diaphragmatic hernia may be variable manifestations of SEN that have not yet been observed. Molecular studies of the KCTD1 gene are currently pending on this family. We suspect that this family is part of the variability of this uncommon condition and reports of additional cases and families with mutations of KCTD1 will help delineate and define the SEN syndrome.

SMC3 Mutations Cause a Variable Cohesinopathy Phenotype

Matthew A. Deardorff¹, María Concepción Gil-Rodríguez², Morad Ansari³, Ian D. Krantz¹, Frank J. Kaiser⁴, David R. FitzPatrick³, Juan Pié³ and international collaborators.

¹Division of Genetics, Children's Hospital of Philadelphia, Department of Pediatrics, University of Pennsylvania Perelman School of Medicine, Philadelphia, Pennsylvania

²Unit of Clinical Genetics and Functional Genomics, Departments of Pharmacology-Physiology and Pediatrics, Medical School, University of Zaragoza, Spain

³MRC Human Genetics Unit, MRC Institute of Genetics and Molecular Medicine, University of Edinburgh, Edinburgh EH4 2XU, UK

⁴Sektion für Funktionelle Genetik am Institut für Humangenetik, Universität zu Lübeck, Lübeck D-23538, Germany

In 2007, we reported the identification of a de novo missense mutation in SMC3 in a man with features overlapping those of Cornelia de Lange syndrome (CdLS). Over the following years, we screened several hundred additional subjects with CdLS-like features with little success in identifying additional SMC3 mutations.

However, over recent years, several advances have facilitated identification of additional individuals with SMC3 mutations. These include: 1. Wider availability of clinical testing for SMC3; 2. The utilization of NextGen sequencing panels to screen multiple CdLS genes at lower cost; and 3. The widespread advent of clinical exome sequencing for undiagnosed patients.

Using a wide network of international clinicians, diagnostic labs and research collaborators, we have now been able to identify 16 individuals with SMC3 mutations. All of the mutations identified are de novo, suggesting that any SMC3 alteration resulting in a congenital phenotype likely has a relatively pathogenic effect.

While many subjects were ascertained with features of CdLS, these patients demonstrated a wide range of features and variability. Like mutations in SMC1A, which encodes the integral binding partner of SMC3 in the cohesion ring complex, SMC3 mutations typically caused marked intellectual disability, but less congenital malformations and growth deficiencies than seen in CdLS patients with NIPBL mutations. Notably, some patients had very few of the facial features that are recognizable in typical CdLS.

The only recurrent mutation, p. (Glu488del), is located at the junction of the arm and hinge domain, a location orthologous to a site that contains common mutations in SMC1A that lead to a CdLS-like phenotype. We will review the mutations identified to date, correlating clinical features of the subjects and speculate on the nature of mutations in SMC3 that cause a CdLS-like presentation. We will contrast this with mutations that present with cohesinopathy features that are markedly different from CdLS and forward hypotheses why these differences are noted.

16p11.2 Microdeletion Syndrome: Expanding the Phenotype

Mary Beth Dinulos, MD^1,2, Stephanie E. Vallee, MS^1,2 and Joel A. Lefferts, PhD²

¹Department of Pediatrics, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire

²Department of Pathology, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire

Advances in microarray technology have allowed for identification of a recurring microdeletion syndrome involving chromosome region 16p11.2. This contiguous gene deletion syndrome is primarily characterized by developmental delay, intellectual disability, autism spectrum disorder, epilepsy and obesity. Congenital malformations are occasionally reported and include primarily cardiovascular abnormalities and vertebral/rib anomalies.

We report a newborn male with multiple congenital anomalies including bilateral radial ray defects with hypoplastic non-articulating thumbs, large ventricular septal defect, moderate secundum atrial septal defect, asymmetric number of ribs and abnormally shaped scapulae. Radiographs confirmed absence of the right radius, bowed right ulna, and bilateral thumb hypoplasia. Physical examination was significant only for anomalies of the upper extremities with no dysmorphic facial features.

Microarray analysis detected a 751 kb deletion from cytoband 16p11.2, including the TBX6 (T-box 6) gene. The TBX6 gene, which encodes a transcription factor, plays a critical role in important developmental processes including paraxial mesoderm differentiation and left–right patterning. Loss of TBX6 has been suggested as a contributing factor for congenital anomalies associated with 16p11.2. We theorize that haploinsufficiency of TBX6 may contribute to the multiple anomalies seen in our patient.

Review of the medical literature identified one prior case report of an infant with 16p11.2 microdeletion syndrome and bilateral radial ray defects. In that case, the newborn presented with severe respiratory distress secondary to a left-sided congenital diaphragmatic hernia. He was also noted to have an extra thoracic vertebra, thirteen pairs of ribs and hypoplastic non- articulating thumbs similar to our patient.

In summary, we report an infant with 16p11.2 microdeletion syndrome and multiple congenital anomalies, including bilateral radial ray defects. We theorize that haploinsufficiency of TBX6 may contribute to the multiple anomalies seen in our patient. We propose that our case expands the phenotypic spectrum of 16p11.2 microdeletion syndrome to include radial ray defects.

Gillespie Syndrome Caused by De Novo Mutation in ITPR1

Sarah Dugan¹, Chelsy Jungbluth¹, Tom Shuey², Bill Mize¹

¹Children's Hospitals and Clinics of Minnesota, Minneapolis, Minnesota

²EssentiaHealth, Duluth, Minnesota

Gillespie syndrome (MIM 206700), first reported in 1965, encompasses the rare association of aniridia, cerebellar atrophy or hypoplasia, and developmental delay. Familial cases have been described, including several sib pairs and an affected mother and child in two cases. The condition has been attributed to PAX6 changes in some patients with variants of the condition, but investigation in other cases has failed to show an underlying cause.

We report a patient with classic findings of Gillespie syndrome and a deleterious change in ITPR1, a gene known to be associated with spinocerebellar ataxia types 15 and 29. The patient presented at age 17 months with aniridia, hypotonia, and developmental delay. By age 30 months, he had developed ataxia. Genetic testing prior to exome sequencing included normal oligoarray-based comparative genomic hybridization, 21 CGG repeats on FMR1 analysis (normal), and normal sequencing and deletion/duplication testing of PAX6. Brain MRI showed hypoplasia of the cerebellum, particularly the vermis and superomedial cerebellar hemispheres, enabling a clinical diagnosis of Gillespie syndrome. Whole exome sequencing identified the de novo change c.7640_7642 in ITPR1, corresponding to the amino acid change p.K2458del in the ion transport domain. ITPR1 encodes type I inositol 1,4,5-triphosphate receptor (InsP₃R-1), which complexes with homologous receptors to act as a ligand-gated calcium channel on the endoplasmic reticulum membrane. InsP₃R-1 has three primary domains: a binding domain, a regulation and coupling domain, and a transmembrane-spanning ion channel domain. The gene is expressed throughout the central nervous system, particularly in Purkinje cells of the cerebellum. Thorough expression studies in ocular tissues have not been reported, but there is evidence for expression in the lens epithelium as well as in smooth muscle. SCA29 is an autosomal dominant congenital, generally nonprogressive ataxia associated with intellectual disability. Cerebellar atrophy can be seen over time in this condition. SCA15 is an adult-onset, sometimes progressive ataxia; intellectual disability is not a feature. ITPR1 changes leading to haploinsufficiency are thought to be causative of SCA29; those associated with SCA15 have been in the coupling/regulatory domain of ITPR1. Although eye movement abnormalities such as strabismus and nystagmus are frequently associated with both conditions, structural eye malformations have not been reported as an associated feature in either condition.

Our patient expands the phenotypes associated with mutations in ITPR1 and suggests a genetic cause for Gillespie syndrome. Although more affected individuals with ITPR1- related Gillespie syndrome would need to be identified in order to draw any conclusions, the fact that our patient has a mutation in the ion channel domain of the gene is interesting and may indicate further genotype-phenotype correlation for ITPR1-related spinocerebellar ataxia. The mechanism by which a mutation within the ion channel domain of ITPR1 would lead to aniridia is not clear but may overlap with the PAX6 pathway via altered intracellular calcium propagation.

Thrombocytopenia Absent Radius (TAR) Syndrome: What About the Thumb?

Alison M. Elliott¹, PhD, MS, CGC, Judith G. Hall^1,2, OC, MD, FRSC, FCAHS

¹Department of Medical Genetics, University of British Columbia, Vancouver, British Columbia

²Department of Pediatrics, University of British Columbia, Vancouver, British Columbia

Thrombocytopenia absent radius (TAR) syndrome is characterized by thrombocytopenia, bilateral absence of the radius with preservation of the thumbs (Hall et al., Medicine, 1969). The presence of the thumbs distinguishes TAR from other disorders with radial involvement (Holt-Oram syndrome, Fanconi Anemia) but can also be seen in Roberts' syndrome.

True intercalary defects are rare and it has been suggested that intercalary deficiencies might be a part of the spectrum of longitudinal deficiencies. In 1976, Swanson summarized the American Society for Surgery of the Hand and questioned whether intercalary defects existed: “all ‘phocomelias’… have some terminal manifestations.” The purpose of this study was to review published series of TAR patients in order to further characterize the thumb with respect to hypoplasia. In the original series of 60 cases, short, pointed distal phalanges of the thumbs were described in one patient (Hall et al., Medicine, 1969). Goldfarb et al. (The Journal of Hand Surg, 2007) reviewed the clinical appearance of the thumb in seven patients with TAR syndrome and concluded that the thumb was relatively wide and flat with decreased function and was held in a position of metacarpophalangeal flexion across the palm. Radiographs were not evaluated, but clinical observation suggested a normally sized thumb. Greenhalgh et al. (J Med Genet, 2009) reviewed 34 patients with TAR syndrome and described the thumb as abnormal (hypoplastic or proximally placed) in all 34 patients. Houeijeh et al. (Eur J Med Gen, 2011) reviewed 14 cases and found seven to have thumb hypoplasia. Although the thumb has been described as present and functional in TAR (Hall et al., Medicine, 1969), the hands are not considered normal. Distal radiographic abnormalities can include carpal anomalies (usually fusions) and phalangeal hypoplasia (e.g., middle phalanx of the fifth ray). Characterizing limb hypoplasia is best achieved through radiographic evaluation, which is not available for all patients.

TAR shows complex inheritance with a null allele and one of two low-frequency SNPs in the regulatory regions of RBM8A, a gene encoding the exon-junction complex subunit member Y14. Y14 is involved in nuclear export and subcellular localization of specific transcripts, translational enhancement and nonsense-mediated RNA decay (Albers et al., Nat Genet 2012). The variability in radial longitudinal involvement may be related to spatial and temporal differences in regulatory networks within the developing limb. The 5' UTR SNP introduces a binding site for EVI1, a transcriptional repressor. In the mouse, Evi1 is expressed in a transient manner in limb buds with homogeneous mesodermal expression but none in the apical ectodermal ridge at day 9.5. By day 14.5, Evi1 expression is limited to the perichondrium. Its pattern of expression resembles some Hox genes (Perkins et al., Dev 1991). HOXA11 mutations are found in amegakaryocytic thrombocytopenia and radioulnar synostosis (Thompson and Nguyen, Nat Genet 2000). The temporal and spatial expression of key developmental genes, regulatory elements in addition to the involvement of other modifiers within the intricate complex networks of the developing limb may explain the patterning differences. It is also important to consider the effects of movement and the vascular/osseous interplay when evaluating limb development.

Cranial Neural Tube Defects With Radial Ray Anomalies

Jane A. Evans

Department of Biochemistry and Medical Genetics, University of Manitoba, Winnipeg, Manitoba

Despite its rarity (∼1 in 83,000 births), the co-occurrence of neural tube defects (NTD) with radial ray defects (RRD) occurs with a higher frequency than expected by chance. In the Canadian seven province study of 2575 NTD cases, 23 had RRD (0.89% or 1 in 112). This report highlights the findings in ten cases with RRD and anencephaly, iniencephaly or cranio-rachischisis. RRD, usually unilateral, included radial ray aplasia (4), absent or hypoplastic thumb (3) and preaxial polydactyly (3). Cases usually (70%) had additional major malformations, with renal (50%), cardiovascular (50%), body wall defects (33%) and costovertebral anomalies (40%) most common. Fewer than 25 such individuals have been described previously, usually as single case reports, thus this represents the largest population-based series to date. Although this pattern of anomalies is rare, it is clearly heterogeneous. The association of omphalocele with cranial NTD and RRD has been previously reported, with several cases also having diaphragmatic hernia and/or other midline defects e.g., imperforate anus or orofacial clefts. Two cases in the Canadian series had omphalocele; one with anencephaly, polydactyly, horseshoe kidney and preauricular tags; the other with craniorachischisis, left (L) radial aplasia, hemivertebrae and a preductal coarctation. Two others resembled cases reported as “encephalocele-radial, cardiac, anal and renal anomalies” having craniorachischisis, encephalocele, radial aplasia and absent right (R) kidney; and iniencephaly, hypoplastic thumb, multiple vertebral defects, tracheoesophageal fistula, lung and spleen anomalies, absent Mullerian ducts and bilateral renal agenesis, respectively. A third case with anencephaly, absent thumb, hemivertebra, hypoplastic kidney, imperforate anus, absent uterus and ambiguous genitalia had a karyotype of 46,XX,r(13).

Phenotypic overlap with other conditions such as VACTERL, MURCS, Klippel-Feil, Schisis association, DK phocomelia, XK-aprosencephaly and axial mesodermal dysplasia is obvious. Like many of these disorders, our cases appear to represent examples of a primary field defect occurring in blastogenesis. Three additional individuals in our series had a cranial NTD with RRD but no other major defects. Although these could be coincidental findings, it is more likely that they represent milder forms of the same field defect. Only in the last two cases was a vascular insult a potential cause: one had anencephaly, a lateral body wall defect, absent L thumb and R arm and an atrial septal defect; the other had a severe Poland-like anomaly with craniorachischisis, L radial and index finger aplasia, only eight L ribs with rib fusions, unilobed L lung, malrotated bowel and pulmonary valve atresia, though these too are probably just different manifestations of a primary blastogenic field defect. As with other field defects, cranial NTD and RRD occur in causally non-specific conditions including chromosomal abnormalities, single gene disorders (possibly those in developmental signalling networks/pathways e.g., SALL4, GLI2), environmental exposures (e.g., maternal diabetes, valproate) or other mechanisms. Regardless of etiology, defects in cases vary considerably even between familial ones. As the central nervous system (CNS) malformations are very inconsistent, often involving unusual or multiple, sometimes non-contiguous, defects, or different anomalies in affected sibs, should a more generic term be derived to refer to these disorders rather than the CNS malformation specific ones currently used? In other words, is it time to lump rather than split with respect to CNS-radial ray associations?

Further Confirmation of Type IV Klippel–Feil Syndrome and its Marker Differentiating Feature, Sacral Agenesis: Report of an Affected Mother and 3 Offspring

B.D. Hall, MD

University of Kentucky, Lexington, Kentucky

In 1999 at the Smith Workshop JV Cabrea and myself reported a mother and daughter with Klippel–Feil syndrome and sacral agenesis or what Raas-Rothschild and colleagues had previously reported (J CranioFac Genet Dev Biol, 1988) in a single case as Type IV Klippel–Feil syndrome. They noted six previous singleton literature cases (1929–1974) of Klippel-Feil syndrome with sacral agenesis. Legius et al. (J. Craniofac Genet Dev Biol, 1988) in a follow-up letter reported another case. To the best of our knowledge these eight singleton literature cases plus our two familial cases are the only reported cases of Type IV Klippel–Feil syndrome.

The affected mother of our family has since had two additionally affected children, a boy and girl, whose features are similar to the mother and older sister. This family firmly supports a genetic etiology, probably autosomal dominant, for Type IV Klippel–Feil syndrome. Molecular testing as of 2006 for genes know to be associated with vertebral (somite) development had not identified a mutation and the family has been resistant to further testing.

The main clinical features in our Type IV Klippel–Feil family are short stature, short web neck, low posterior hair line, wide short chest, kyphoscoliosis/lordosis, reduced buttock and leg muscle mass, contracted hips, no palpable sacrum, flexed knees, and various degrees of talipes equinovarus. Radiographs of the spine showed a mixture of hemivertebrae and fusion of the cervical spine, scattered hemi-/dyssegmeted thoracic and lumbar vertebrae, total or partial sacral agenesis, and contracted pelvis. Potential problems include cord compression, lower limb paresis and/or contractures, urinary and bowel incontinence, atonic bladder, and complications of urinary tract obstruction. Affected females because of the contracted pelvis are likely to require C-section delivery as has our adult mother for all her pregnancies.

Most cases of sporadic Klippel–Feil, Types I–III remain without gene identification whereas some familial cases have shown mutation of MEOX1, GDF6, and PAX 1. This is also true of non-syndromic sacral agenesis except for mutation of HLX9 in autosomal dominant sacral agenesis. To our knowledge our family represents the only familial incidence of Type IV Klippel- Feil which indicates a high likelihood of a future gene mutation being identified.

Congenital Diaphragmatic Hernia Is Genetically Complex and Associated With Many Chromosomal Abnormalities and Syndromes

Robert J. Hopkin^1,2, Amy Ross², Patricia Burns², Michelle Baric^1,2, Emily Partack^1,2, Foong-Yen Lim^2,3

¹Division of Human Genetics Cincinnati Children's Hospital Medical Center

²Cincinnati Fetal Center, Cincinnati Children's Hospital Medical Center

³Department of Surgery, University of Cincinnati

Diaphragmatic hernia (DH) occurs in approximately 1/3000 pregnancies. This contributes to 1–2% of perinatal death. In 10–15% of cases a genetic syndrome or chromosomal abnormality is recognized. When there is no recognized syndrome up to 25% have additional malformations. We retrospectively reviewed 309 cases of prenatally diagnosed DH. 103 affected fetuses had other malformations or genetic conditions. All patients were evaluated in the Cincinnati Fetal Center. Data collected included major malformations, genetic test results, and diagnosed syndromes. Postnatal information was included when available.

Results: 13/103 cases were associated with chromosomal abnormalities including recurrent deletions at 8p23 and 15q26. Genetic syndromes were identified in 10 cases with Cornelia de Lange (CDL) being the most frequent. Other syndromes included Goltz, Fryns, Beckwith-Wiedemann, and suspected Donnai-Barrow. Heart defects were seen in 49 infants, 21 of which were ASD or VSD, and aortic arch anomalies in 6. Complex heart malformations were seen in 22 cases and associated with a much worse outcome. Recurrent malformations of limbs (6), kidneys (6), genitals (4) were seen as were omphalocele (8), imperforate anus (2), and holoprosencephaly (2). One was affected in 11 sets of twins, including two sets of conjoined twins. 1 of triplets was also noted. There were 5 families with recurrent DH: one mother/daughter pair, 2 sets of siblings, and 2 sets of cousins.

Discussion: DH is a severe but relatively common malformation. In approximately 60% of cases this is an isolated, usually sporadic defect. However, there are >40 chromosomal regions associated with increased risk. The chromosomal abnormalities include many rearrangements that are not usually detected with free fetal DNA testing and can be missed on chromosomal analysis. High resolution microarray is recommended. At least 29 non-chromosomal genetic syndromes have been associated with DH. Of these, the association is strongest for Fryns syndrome and Donnai-Barrow syndrome. However, because these are very rare, they are not the most frequent syndromes reported. CDL syndrome was the most frequent in our series, and should be considered in cases of DH if chromosomes and microarray are normal, especially if there is IUGR. Heart defects were the most common associated abnormality. VSD, PDA, and ASD were seen but did not change prognosis. Complex heart disease was a poor prognostic sign. Several other defects were seen with high frequency. Of note, in this series the holoprosencephaly was not associated with chromosomal abnormalities and in one case was associated with consanguinity. Another important risk was the association with twinning which associated with 4% of DH if both conjoined twins and more routine twins were included.

Next Generation Sequencing of Individuals With Turner Syndrome: Bicuspid Aortic Valve Vs. Tricuspid Aortic Valve

Paul S. Kruszka, MD¹, Seth I. Berger, MD, PhD¹, Nicole K. Banks, MD^1,2, Caroline Frier¹, Yonit A. Addissie BA¹, Ariel F. Martinez, MS¹, Rachel Hart, BA¹, Carolyn Bondy, MD²,Maximilian Muenke, MD¹

¹Medical Genetics Branch, National Human Genome Research Institute, NIH

²National Institute of Child Health and Human Development, NIH

Background: Turner syndrome is caused by complete or partial monosomy X, with an incidence of 1/2500 female births, and has congenital heart disease as a cardinal feature. Bicuspid aortic valve (BAV) affects 30% of individuals with Turner syndrome, which is higher than the general population prevalence of 1%. The etiology of BAV in the general population is largely unknown, with only a few known genes accounting for a small percentage of disease. Most cases of BAV are initially asymptomatic, but are associated with aortic dissection in women with Turner syndrome and become prematurely calcified, requiring valve replacement during adulthood.

Methods: Using cardiac MRI and next generation sequence data generated from exome sequencing performed by the National Intramural Sequencing Center, we are testing our hypothesis that genetic variation, on the background of haploinsufficiency of X chromosome genes in individuals with Turner syndrome, contribute to the high prevalence of BAV. Women with Turner syndrome were enrolled in the protocol Personalized Genomics: Congenital Malformations (11-HG-0093) at the National Human Genome Research Institute.

Results: Twenty-six women (average age 44 years; range 23–68 years) with Turner syndrome were evaluated: 10 participants had BAV and 16 had tricuspid aortic valve (TAV), 24 women were 45,X [50], one was 45X [13]/ 46,X, i(X) (q10)[37], and one was mosaic for 45,X[40]/46,XX[10]. After evaluating genes known to cause BAV (NOTCH1, GATA5), only a predicted pathogenic variation in NOTCH1 (c.1826G>A; p.R621H) was found in an individual with BAV. Next, genes in the pseudoautosomal region and X chromosome genes with functional homologues on the Y chromosome were evaluated and found to not have differences in variation between the BAV and TAV individuals. Finally, an unbiased approach comparing genetic variation in coding genes was undertaken; 47 genes had an uncorrected (multiple testing) P < 0.05. None of these 47 variants are known to cause congenital heart disease. The most statistically significant predicted pathogenic finding was a variant (rs2142661) in the gene RIBC2 (P = 4 × 10⁻⁴; uncorrected for multiple testing), which is involved in cell cilia function.

Conclusions: The few known genes associated with BAV in the general population do not contribute to BAV in our study. The possibility of a hypomorph in the pseudoautosomal region or a gene with a functional homologue on the Y chromosome contributing to BAV is also not supported by our data. Our study suggests that BAV in individuals with Turner syndrome is a complex trait with multiple genes responsible for risk, and RIBC2 having the highest significance. As more individuals are recruited (increasing study power) and pending chromosomal microarray data arrive, we will assemble a more complete understanding of BAV in Turner syndrome.

Suggested Reclassification of Previously Described Phenotypes as Ciliopathies

Angela E. Scheuerle¹, James L. Weber²

¹University of Texas, Southwestern, Dallas, Texas

²Prevention Genetics, Marshfield, Wisconsin

The ciliopathies are multisystemic, typically recessive, disorders resulting from aberrations in ciliary structure and function. Abnormalities of primary, non-motile, cilia have been found in disorders comprising growth abnormalities (obesity, skeletal dysplasia), polydactyly, orofacial clefting, and renal and retinal dysplasias. A subset of these findings can be found in acrocephalopolydactylous dysplasia (ACPD; Elejalde syndrome; MIM 200995), a clinical conditions for which the genetic etiology is unknown.

Presented here is a neonate who was serendipitously tested for mutations in PKHD1 because prenatal ultrasound findings suggested autosomal recessive polycystic kidney disease (ARPKD). Subsequent genetics physical exam noted clinical appearance of ACPD. PKHD1 testing identified three variants: a paternally inherited c.390+1G>T (splice donor disruption) which is been reported as causative for ARPKD; a maternally inherited c.776 C>G (p.Ser259Stop) truncating mutation presumed to be pathogenic but previously unreported; and de novo c.3089C>T (p.Ala1030Val) of unclear clinical consequence.

ACPD is historically classified as an overgrowth syndrome of uncertain inheritance. There are nine reported cases, including the index cases of a brother/sister pair born to a consanguineous couple. The current baby's clinical presentation and molecular results suggest that the phenotype represents a ciliopathy with autosomal recessive inheritance; in this case, as a manifestation of PKHD1 mutation. Etiology may be the de novo mutation or presence of some other abnormal gene or environmental factor. Literature cases predated modern molecular testing, so genotypic comparison is impossible.

A search of the literature reveals other independently listed conditions with polycystic kidneys, retinal dystrophy, and polydactyly that have limited case numbers and no identified associated gene: 1) Polycystic kidney disease, Potter type I, with microbrachycephaly, hypertelorism, and brachymelia (Gillessen- Kaesbach syndrome; GKS; MIM 263210); 2) Polycystic kidney, cataract, and congenital blindness (MIM 263100), and; 3) Arima syndrome (MIM 243910).

GKS was described in a single publication in three sibling pairs. The phenotype shows intrauterine growth restriction rather than overgrowth, but is otherwise similar to ACPD and to the ciliopathies. GKS was a neonatal lethal in all reported cases. Previously published linkage analysis of PKHD1 in those patients did not support association with the gene, but direct molecular studies have not been reported. Polycystic kidney, cataract, and congenital blindness were observed in a sib trio published in 1963, at which time two sibs had died in early adulthood. Arima syndrome has already been noted to share phenotypic features with Joubert and other ciliopathies and, indeed, one reported patient had a sibling with clinically diagnosed Joubert.

If it had not classically been considered an overgrowth syndrome, ACPD would probably already have come to attention as a likely ciliopathy, as Arima syndrome already has. Our understanding of the ciliopathy syndrome family would benefit from re- evaluation of these phenotypes.

Deletion Upstream Of The SALL1 Gene Producing Townes–Brocks Syndrome

Cathy A. Stevens, MD, Kristin M. May, PhD

Department of Pediatrics, University of Tennessee College of Medicine, Chattanooga, Tennessee

Townes-Brocks syndrome (TBS) is an autosomal dominant disorder characterized by imperforate anus, dysplastic ears, and abnormal thumbs. Other findings may include renal anomalies/dysfunction, hearing loss, congenital heart defects, other skeletal anomalies, intellectual disability (10%), and behavior problems. TBS is caused by mutations in the zinc finger transcription factor gene SALL1 located at 16q12.1. Nonsense and frameshift mutations as well as small intragenic deletions and insertions have been described, which are thought to result in truncated SALL1 proteins. All published mutations are located in exon 2 or intron 2 of the gene. Larger intragenic deletions and complete gene deletions have also been reported. We report a girl with TBS found to have a small deletion 125 kb upstream of the SALL1 gene suggesting that loss of a regulatory element may be responsible for TBS in this patient. C.H. is a 7½-year-old adopted Guatemalan female with imperforate anus, rectovaginal fistula, lipoma of the filum terminale, vesicoureteral reflux, renal dysfunction, multiple failed hearing tests attributed to an auditory processing disorder, asymmetry of ear size, broad thumbs and finger tips, and retinal abnormalities. Echocardiogram was normal. She has an IQ in the 70s, sensory integration disorder, and significant behavior problems. Microarray analysis revealed a 605.45 kb single copy loss in chromosome region 16q12.1. There are no genes within this copy number loss but it is located 125 kb upstream of the SALL1 gene. Sequencing and deletion/duplication studies of the SALL1 gene were normal. The parents are not available for study and there is no family history information.

There is a report in the literature of a child with TBS and a balanced translocation t(5;16)(p15.3;q12.1) with the chromosome 16 breakpoint located approximately 180 kb upstream of the SALL1 gene. Sequencing and Southern blot analysis of SALL1 were normal. The authors suggest a possible position effect in this child. We propose that our patient's deletion upstream of SALL1 results in loss of a critical cis- regulatory element located within the deleted region. This could also be the mechanism of TBS in the previously reported patient. Disruption of long-distance cis-regulatory elements have been noted to result in several human diseases. SALL1 may be added to this growing list of genes which includes SOX9, GJB2, FOXL2, PITX2, MECP2, and others. Approximately 70% of individuals with typical features of TBS are found to have mutations in SALL1 by sequencing or deletion/duplication studies. In those without a detectable mutation, the reason could be misdiagnosis, locus heterogeneity for TBS, or mutations in other regions of the SALL1 gene not accessible by techniques utilized. However, our patient and the one previously reported suggest that some individuals with TBS may have the condition due to deletion of a critical region upstream of the SALL1 gene. Therefore if a patient with the TBS phenotype has non-diagnostic SALL1 studies, microarray analysis and chromosome analysis should be considered.

Sudden Death in a Newborn With Oculodentodigital Dysplasia and a De Novo GJA1 Missense Mutation: The First Report

R.D. Clark, MD¹, J. Park, DO²

¹Division of Medical Genetics/Pediatrics, Loma Linda University Medical Center, Loma Linda, ‎California

²Riverside County Sheriff Coroner, Perris, California

Oculodentodigital dysplasia (ODDD) is an autosomal dominant disorder caused by a heterozygous mutation in the gap junction alpha-1 (GJA1) gene that codes for connexin43, the most abundant connexin in the heart. ODDD is characterized by microphthalmia, short palpebral fissures, distinctive facial features, syndactyly type III (ring and little fingers) and neurologic symptoms such as spasticity, weakness and dysarthria. Cardiac anomalies and conduction defects have been reported infrequently.

A newborn male with syndactyly and unusual facial features was born at term after an uncomplicated pregnancy. He had normal growth parameters and excellent Apgar scores. He roomed-in with his mother after delivery. The clinical diagnosis of ODDD was made on the basis of a large posterior fontanelle, small eyes, short palpebral fissures, faint eyebrows and eyelashes, small nares, hypoplastic alae nasi, and syndactyly of the ring and little fingers. His neuro exam was completely normal with excellent tone, a lusty cry and a good suck. He fed well and was discharged home with his mother. On the trip home from the hospital, his mother found him unresponsive in his car-seat. CPR administered by paramedics on the side of the freeway was futile. The county coroner submitted a tissue sample for GJA1 analysis, which identified a novel de novo mutation: c.20T>C (p.Leu7Ser) that was predicted to be deleterious. This is near the NH2 terminal end of the protein in the cytoplasmic domain.

Channelopathy-mediated arrhythmias may be responsible for 10–15% of Sudden Infant Death Syndrome (SIDS) but there have been no previous reports of sudden infant death in patients diagnosed with ODDD. Van Norstrand et al. (Circulation 2012) identified two rare missense mutations in GJA1 (E42K, S272P) among 292 SIDS patient samples. Paznekas et al. (Am J Hum Genet 2003) described a patient with craniofacial features of ODDD (GJA1 mutation: I130T) and recurrent ventricular tachycardia and atrioventricular block. His affected father had a sudden cardiac death at age 43 and other relatives had arrhythmias. Recently, Quick et al. (Am J Forensic Med Pathol 2014) reported the death of a teenager from cardiac arrest after being thrown down at a wrestling match. He had ventricular tachycardia, torsades and ventricular fibrillation. Two missense mutations were identified in the GJA1 gene in frozen heart tissue: c.G605A and c.G717A. Loss of GJA1 function is associated with lethal arrhythmias including idiopathic familial atrial fibrillation. In murine models, complete loss of connexin43 expression leads to arrhythmias and sudden death. The association of sudden infant death and ODDD in this patient is the first report of this outcome. Other anecdotal reports clearly establish a relationship between GJA1 mutations and arrhythmias in individuals with this syndrome. This deadly association increases the need to recognize ODDD early and to monitor affected individuals for arrhythmias.

Velopharyngeal Insufficiency in an Adolescent: An Unusual Presentation of Desmin-Related Myofibrillar Myopathy

L.W. Burke, MD¹, W. Waheed, MD²

¹Department of Pediatrics, University of Vermont College of Medicine, Burlington, Vermont

²Department of Neurology, University of Vermont College of Medicine, Burlington, Vermont

Hypernasal speech caused by velopharyngeal insufficiency (VPI) is a common referral to a cleft palate clinic and is usually associated with a structural difference in the secondary palate. VPI is also a known complication of myotonic dystrophy and other muscular dystrophies, usually being a late finding. We report a case in which the patient developed hyponasal speech and signs of VPI in adolescence as the presenting symptom of a rare myopathy known as desmin-related myofibrillar myopathy (DRM). To our knowledge, this is the first reported case of this condition presenting in this manner.

Our patient first presented at age 16 to the craniofacial clinic with a history of worsening hypernasal speech over the preceding couple of years. He was diagnosed with VPI and after a lack of improvement with speech therapy, he underwent a palatoplasty. A formal genetics evaluation was recommended, but this did not occur. During his palatoplasty surgery, he was noted to have atrial fibrillation with left bundle branch block. A subsequent cardiology evaluation identified restrictive cardiomyopathy. He did not have any exercise intolerance and was active in sports.

He was followed by cardiology and had good exercise tolerance and no cardiovascular symptomatology until the age of 20, when the cardiac MRI showed development of left ventricular dysfunction. Five months later he had an acute onset of left sided chest pain and left sided abdominal pain. He was found to have splenic infarcts and a CT scan showed pulmonary edema, hepatic congestion and ascites. He then developed worsening of his restrictive cardiomyopathy and was placed on the cardiac transplant list, where he spent the better part of a year. During his post-operative course he was noted to have persistently elevated CK measurements and muscle aches, as well as MRI findings of myositis, initially thought to be due to his prolonged bed rest prior to the surgery and to the surgery itself. When the elevated CK did not resolve and he did not recover his pre-operative strength, he was sent to neurology. He had a muscle biopsy and EMG that were consistent with a nonspecific myopathy, and on neurologic exam, was noted to have weakness of his lower extremities. A genetic myopathy panel was done and he was found to have an R406W mutation in the DES gene, a known cause of myofibrillar myopathy. The R406W mutation in the DES gene prevents filament assembly and disrupts the network in a dominant-negative fashion. The R406W mutation has been described in both sporadic and familial cases, and in cases with both skeletal and cardiac muscle involvement. In this particular mutation, the age of onset tends to be earlier and there is a rapid progression resulting in severe disability and death. Hypernasal speech is described but only after the appearance of other symptoms.

The vast majority of children who present with velopharyngeal insufficiency and hypernasal speech have a fairly isolated structural or functional difference in their soft palate. The most common cause is a deletion of 22q11.2. However, this case illustrates the fact that in patients who present in adolescence, myopathic processes need to be considered even in the absence of any other neuromuscular signs.

Complete Monosomy Mosaic of Chromosome 21: Report of a Case With Persistent Thrombocytopenia

M. Aracena

Genetic Unit, Hospital Dr. Luis Calvo Mackenna, Santiago Chile and División Pediatría, Pontificia Universidad Católica de Chile

Introduction: monosomy 21, whether complete or as mosaicism is a very rare chromosomal anomaly. Their clinical features are highly variable. This report describes a male child with a monosomy 21 in a mosaic state.

Case report: the proband is the first child of non- consanguineous healthy parents. He was born in January 2014 after an uneventful pregnancy, normal vaginal delivery. Neonate adecuate for gestational age, BW: 3560gm; BL:52 cm and OFC: 34 cm. At birth he showed redundant skin in his neck, hoarse cry, mild respiratory distress and a petequial rash. The platelet count was 57.000/μl. He required oxygen the first 24 hr of life and intravenous antibiotics were administered for 10 days, with the suspicion of bacterial infection that was not confirmed. Cerebral and abdominal ultrasound were normal. During the next months he persisted with a low platelet count, without bleeding. He was evaluated in the Genetic Unit at 3mo, and a broad nasal bridge and micrognathia were described. A cytogenetic study was ordered because of his mild dysmorphias and thrombocytopenia. Karyotype: 45,XY,-21[20]/46,XY,del (21)(q22.1)[16]. Karyotype of both parents showed no abnormality. FISH analysis with probes from 21q22.1-q22.2 region done in peripheral blood also showed two cell lines. One with absence of one signal of this probe, denoting the complete monosomy of chromosome 21. In the other cell line we saw 2 signals of the probe, one hybridized to the normal 21 chromosome and the other to the deleted 21. A fibroblast skin culture is in process. At 6 mo age during a viral illness a complete immunological study revealed hypogammaglobulinemia, so he received a single dose of IVIG. During his follow up he has had adequate physical and psychomotor development, only mild upper respiratory infections and has not needed blood transfusions.

Discussion: the congenital thrombocytopenia with only mild dysmorphic features prompted the cytogenetic study showing a complete monosomy 21 in mosaicism. The second cell line found has a partial monosomy of chromosome 21. We discuss the probably meiotic origin of the deleted 21, and the subsequent mitotic loss of the structural abnormal 21. At least one of the cell lines has a haploinsufficiency of RUNX1 gene mapped at 21q22 region that could explain his platelet disorder and could predispose to a myelodysplastic syndrome (MDS).

ZIC3 Copy Number Gain in a Family With an X-Linked Pattern of Multiple Congenital Anomalies

S.D. Adams, MD, C.E. Keegan, MD, PhD

C.S. Mott Children's Hospital, University of Michigan, Ann Arbor, Michigan

The Zic family member 3 (ZIC3) gene, which maps to Xq26.3, is known to be an important transcriptional regulator of body laterality and gastrulation during early embryonic development. Deletions of ZIC3 have been previously associated with a range of phenotypes including heterotaxy, heterotaxy with complex congenital heart disease, isolated cardiac malformations and, at the more severe end, VACTERL spectrum anomalies and hydrocephalus.

There are currently no reports describing the clinical presentation of ZIC3 copy number gains. Here, we report a case of a 1.37Mb one copy gain of Xq23.6, including the ZIC3 gene, in a male presenting with multiple congenital anomalies. Specific anomalies included a large perimembranous ventricular septal defect, left multicystic dysplastic kidney, prune belly syndrome, sacral agenesis, ambiguous genitalia with no clearly identifiable clitorophallic structure, nonpalpable gonads, bilateral club feet, and subluxation of the femoral heads. Family history was significant for a deceased male sibling who passed away shortly after birth secondary to multiple congenital anomalies, consistent with an X-linked pattern of inheritance.

ZIC3 is highly expressed in both the neuroectoderm and migrating mesoderm, with a specific temporal and spatial expression pattern that is important for proper embryogenesis. Overexpression of ZIC3 in migrating mesoderm suggests a possible mechanism underlying the constellation of musculoskeletal, urogenital, renal, and cardiac abnormalities seen in this family.

A High Diagnostic Yield of Clinical Exome Sequencing in Middle Eastern Patients With Neurocognitive Phenotypes

Tarunashree Yavarna¹, Nader Al-Dewik¹, Mariam Al-Mureikhi¹, Rehab Ali¹, Fatma Al-Mesaifri¹, Laila Mahmoud¹, Noora Shahbeck¹, Shenela Lakhani¹, Mariam AlMulla¹, Zafar Nawaz², Patrik Vitazka³, Fowzan S Alkuraya^4,5, Tawfeg Ben-Omran^1,6

¹Clinical and Metabolic Genetics, Department of Pediatrics, Hamad Medical Corporation, Qatar

²Cytogenetics and Molecular Cytogenetics, Department of Pathology & Laboratory Medicine, Hamad Medical Corporation, Qatar

³GeneDx, Gaithersburg, Maryland

⁴Developmental Genetics Unit, Department of Genetics, Research Center, King Faisal Specialist Hospital and Research Center, Riyadh-Saudi Arabia

⁵Department of Anatomy and Cell Biology, College of Medicine, Alfaisal University, Riyadh, Saudi Arabia

⁶Weill Cornell Medical College, Qatar

Clinical Exome Sequencing (CES) has become an increasingly popular diagnostic tool in patients with heterogeneous genetic disorders, especially in those with neurocognitive phenotypes. A clinical cohort of 149 probands from Qatar with various neurocognitive phenotypes underwent CES from July 2012 to June 2014.

Intellectual disability and global developmental delay were the most common clinical presentations but our cohort displayed other phenotypes such as epilepsy, dysmorphism, microcephaly and other structural brain anomalies and autism. A likely causal mutation, including pathogenic CNVs, was identified in 89 probands for a diagnostic yield of 60%. Consanguinity and positive family history predicted a higher diagnostic yield. In 5% (7/149) of cases, CES implicated novel candidate disease genes (MANF, GJA9, GLG1, COL15A1, SLC35F5, MAGE4, and NEUROG1). CES uncovered two coexisting genetic disorders in 4% (6/149) and actionable incidental findings in 2% (3/149) of cases. Average time to diagnosis was reduced from 27 months to 5 months.

The high yield of CES in our population and its ability to call pathogenic CNVs, shorten the time to reach the diagnosis, and uncover the co-existence of multiple conditions argue for this to be the diagnostic test of choice for neurogenetic disorders and our study suggests that this diagnostic method is particularly suited to populations with high rates of consanguinity.

Clinical Utility of the Mendeliome Across Disciplines of Medicine

Sarah M. Al-Qattan MD^1,*, Saudi Mendeliome Group^1,*, Fowzan S. Alkuraya MD^1,2

¹Department of Genetics, King Faisal Specialist Hospital and Research Center, Riyadh, Saudi Arabia

²Department of Anatomy and Cell Biology, College of Medicine, Alfaisal University, Riyadh, Saudi Arabia

*These authors contributed equally

Single gene mutations that are sufficient to cause clinical diseases (Mendelian mutations) represent the most medically actionable class of human genetic variants, and are encountered by every discipline of medicine. In order to understand the contribution of Mendelian mutations to the burden of diseases that are suspected to be genetic in origin, we developed a next-generation sequencing-based multiplexing assay that encompasses the ∼3,000 known Mendelian genes (The Mendeliome) and tested 2,357 patients referred with suspected genetic diagnoses from virtually every medical specialty. A likely causal mutation was identified in 1,022 patients for an overall clinical sensitivity of 43%. This compares favorably with whole-exome sequencing but offers the advantage of reduced running and interpretation costs. Only 11% of the “negative” cases were subsequently found to harbor a likely causal mutation in a known disease gene (largely in genes identified subsequent to the design of our assay) by whole-exome sequencing, as inferred from a representative sample of 178. Although our study population is enriched for consanguinity, 245 (24%) of solved cases were autosomal dominant and 35 (4%) were X-linked in nature so our assay is likely applicable to outbred populations as well. The current version of the Mendeliome accounts for a large proportion of suspected genetic disorders, and offers a translational diagnostic opportunity as well as significant practical advantages over more comprehensive genomic tools.

Redefining Clinical Boundaries of Mendelian Disorders: Lessons From Running the Mendeliome Assay on >2,300 Patients

Iman S. Abumansour, Saudi Mendeliome Group, Fowzan S. Alkuraya

Department of Genetics, King Faisal Specialist Hospital and Research Center, Riyadh, Saudi Arabia

One major challenge in the quest for genotype-phenotype correlation is the inherent bias in the clinical assessment of patients with various Mendelian phenotypes. Assays that define the underlying causal mutation with little influence by the clinical judgement can provide a powerful insight into the true phenotypic spectrum of the respective genes. The Mendeliome Assay (to be discussed in detail in a separate abstract), a multiplexing assay of known Mendelian disease genes, achieved a diagnostic yield of >40% based on >2,300 patients with various phenotypes. In the course of that study, we encountered numerous instances where the phenotype was very different from what has been described in the literature. Some of the most remarkable examples include JAG1-related retinitis pigmentosa, LMNA-related novel laminopathy mutation, NSD1-related failure to thrive, a EP300-related microcephalic primordial dwarfism, ZNF526-related novel Noonan-like phenotype, IFT122-related severe ocular and skeletal abnormalities, and KMT2A-related absent uterus and vagina with remarkable clitoromegaly. Unusual mutation mechanisms include autosomal recessive TCOF1-related Treacher Collins syndrome and a de novo missense mutation in ATN1 gene causing epileptic encephalopathy rather than Dentatorubro-Pallidoluysian atrophy. These and many more cases that will be discussed in our presentation argue that genomic tools are necessary to obviate the clinical bias that has historically limited the phenotypic assignment of disease genes.

Health Surveillance in Males With Disorders of Sex Development and Turner Syndrome Variant Karyotype

A. DeGraaf, GC, C. Du Souich, GC, A. Swenerton, GC, S. Creighton, GC, L. Stewart, MD, L. Armstrong MD

BC Women's and Children's Hospitals, University of British Columbia, Vancouver, BC

Turner syndrome is classically described as a syndrome of short stature and streak ovaries in females. It is sometimes associated with other medically significant features such as congenital cardiovascular disease, renal malformations, autoimmune disease. There are comprehensive published health surveillance guidelines advocating a list of periodic assessments for girls and women with Turner syndrome.

A range of variant karyotypes to the classic monosomy X can be seen in males or individuals with ambiguous genitalia. These involve 1) a rearrangement with partial deletion of the Y chromosome, and/ or 2) mosaicism for a cell line with sex chromosomal abnormality. In this group overall, the above listed features of Turner syndrome are less likely, but there is little concrete data. Furthermore, in individuals with Y material, the chance of germ cell tumors is an important consideration. Factors such as presence or absence of Y chromosome material containing the testis-specific protein, Y-encoded (TSPY) region, as well as the presence or absence of gonadal cells in gonads have been identified to stratify risk.

Over the years, the Multidisciplinary Disorders of Sex Differentiation Clinic at BC Women's and Children's Hospitals, has assessed a number of males and individuals with ambiguous genitalia with Turner variant karyotypes. This has led us to question the role of the Turner syndrome guidelines for these individuals, and best practices with respect to gonadal tumour risk at various ages of presentation. What surveillances have sufficient yield? Are there subgroups of individuals to define who either do or do not warrant specific risk management? This presentation will first review a series of men ascertained with infertility as their only known feature at presentation, and found to have to Turner variant karyotype. We will summarize the literature relevant to each particular Y chromosome rearrangement found and the literature used in the risk assessment for each case. We will present the reasoning behind and the yield of the baseline assessments and ongoing care recommendations made for each individual. Second, we will present results of our ongoing retrospective chart review comparing surveillance practices and yield on surveillance in our study group of patients with a Turner variant karyotype and male or ambiguous external genitalia, ascertained through prenatal study or phenotype at birth. This will be compared to data of our control group of females with Turner syndrome incidentally ascertained prenatally (i.e., ascertained through chromosomal study done for a reason other than what proved to be a clear feature of Turner syndrome). Taken together, these two studies will broaden our understanding of the medical complications present and surveillances warranted in males and individuals with ambiguous genitalia and a Turner variant karyotype.

Sex Determination in Patients With 45X/46,XY Mosaicism

M. Patel, O. Abdul- Rahman

University of Mississippi Medical Center, Jackson, Mississippi

To understand sex determination in patients with mosaicism for 45X/46XY, we reviewed three cases with this karyotype and their varied phenotypic presentation.

Patient 1: A 4 year-old female was referred to endocrinology for short stature. A karyotype demonstrated mosaicism with 45,X[7]/46,XY[23]. She had normal external female genitalia. Her renal and pelvic ultrasound showed normal kidneys, normal uterus in size and location, normal appearing ovaries bilaterally, and normal adnexa. Bilateral gonadectomy was performed at 4 ½ years of age and revealed bilateral streak ovaries with no neoplastic change.

Patient 2: A 13 year-old male presented to the genetics clinic for scoliosis and a history of an abnormal karyotype on amniocentesis involving cell lines with 46,XY and 45,X chromosome complements. The remainder of the pregnancy was normal including a level II ultrasound, and an exam at birth showed normal external male genitalia. His medical history was unremarkable and growth was normal. Pubertal onset was reportedly normal and an exam at 13 years was consistent with a Tanner stage 4 male. A karyotype demonstrated mosaicism with 45,X[11]/46,X,idic(Y)(q11.22)[9]. At 14 years of age, he was trained to conduct testicular self-exams for tumor surveillance. Fertility has not yet been assessed.

Patient 3: A newborn was noted to have ambiguous genitalia involving hypospadias, severe chordee, bifid scrotum, left palpable gonad, and no palpable right gonad. A single opening between the phallus and scrotum was noted. Cystoscopy demonstrated a normal urethra and bladder with a 1.5cm channel at the level of the verumontanum that leads to a normal vagina and cervix. Pelvic ultrasound identified a rudimentary uterus, a left adnexal streak gonad, a normal left testis within the left hemiscrotum, and a right adnexal structure representing either a streak gonad or an atrophic undescended testis. Karyotype revealed 45,X[12]/46,X,idic(Y)(q11.21) [22]. The production of testosterone and conversion to DHT was adequate for a male infant, but the anatomy raised concerns about end-organ androgen insensitivity due to mosaicism. The paternal Y was normal and the family chose to raise the patient as a male.

Patients with 45,X/46,XY mosaicism can present with a range of phenotypes that are not explained by the degree of mosaicism detected on a blood karyotype. Shinawi et al examined a patient similar to patient 3 and noted a predominant proportion of 45,X in the patient's normal testis. The presence of an isodicentric Y in patients 2 and 3 likely predisposed to a post-conception loss of the second sex chromosome, but the timing of the loss may impact the level of SRY more so than mosaicism and alter male sex determination. Because the degree of mosaicism is not correlated with phenotypic presentation, gender assignment in these patients must be performed with caution.

An Unusual Xp;Yq Translocation Leading to Sex Reversal

Lynne M. Bird, MD

University of California, San Diego and Rady Children's Hospital, San Diego, California

A 3-year-old Chinese girl with an unusual sex-chromosome rearrangement is reported. She was one of twins born after a pregnancy conceived with the help of in vitro fertilization. The father's sperm had been harvested before his treatment for cancer. The pregnancy was complicated by discordant growth of same sex twins, with this, the smaller, twin having increased amniotic fluid despite decreased blood flow by Doppler.

She was born at 37 weeks of gestation and weighed 4# 7oz. in comparison to her twin who weighed 5# 7oz. She had apneic episodes and stayed 5 weeks in the NICU for feeding and respiratory problems. She was found to be aspirating and a gastrostomy tube was placed. She was extremely hypotonic and failed to thrive. She developed feeding intolerance even with continuous feeds, making it difficult to advance to her full caloric needs. Development has been severely delayed, and at nearly 3 years of age, she has minimal vocalizations, can roll over and can sit only with support. She was started on growth hormone at 2.5 years, which has not appreciably changed her growth trajectory, which remains 2% for height and <1% for weight, and 12% for OFC. At her most recent visit, her cranial contour was noted to be trigonocephalic, suggesting metopic synostosis, but no imaging studies have yet been performed.

Her karyotype was 46, X, +mar. Subsequent molecular cytogenetic studies showed that the marker chromosome is a derivative sex chromosome that is made up of Yp, the centromere of the Y chromosome and the proximal part of Yq, onto which is translocated Xp. The net result is Xp21.1-pter duplication and Yq11.2-qter deletion.

The literature has six previous reports of individuals with translocation of Xp onto Yq, producing functional Xp duplication and Yq deletion. The outcomes of these previous cases are variable, but all previous reports had male phenotype.

This case broadens the spectrum of outcomes for Xp21.1;Yq11.2 translocations. Though SRY is present by FISH and microarray analysis, she is a phenotypic female, which suggests her gonadal sex is presumably ovarian. One explanation for her female phenotype could be that DAX1 is present in two copies and therefore overexpressed. The gene product of DAX1 inhibits the effect of SRY on Sertoli and Leydig cells and represses transcription of anti-Mullerian hormone. Overexpression of DAX1 can result in complete 46,XY sex reversal. The reason for this patient's female phenotype possibly relates to the dosage of DAX1. Reasons why her phenotype differs from the previously reported cases will be discussed.

A Case of 46,XX DSD With Elevated Testosterone: An Active Ovarian Development Abnormality?

Paul Mark, MD¹, Maala Daniel², MD, Barbara Fulton, MD³, Malorie McVittie, MS4⁴

¹Spectrum Health Medical Genetics, Grand Rapids, Michigan

²Department of Pediatric Endocrinology, Helen DeVos Children's Hospital, Grand Rapids, Michigan

³Department of Pathology, Spectrum Health, Grand Rapids, Michigan

⁴Michigan State University School of Medicine

Classic female sexual development theory states that in the absence of SRY no inhibition of default (passive) female programming will occur and ovaries will develop. Recent studies have suggested that ovarian development results from an active process. We present a case of 46,XX DSD which may have occurred due to an active ovarian development abnormality.

The patient presented at 17 years of age with primary amenorrhea, clitoromegaly, truncal acne and lower abdominal hirsutism, all of which began at puberty. Breast development was Tanner stage 3 and pubic hair was Tanner stage 4. Free and total testosterone were significantly elevated (4.7 ng/dL (nl 0.3–1.9) and 136.4 ng/dL (nl 0.0–75.0)). FSH (79.68mIU/mL (nl 0.64–10.98)) and LH (32.9mIU/mL (nl 0.97–14.7)) were also elevated. AFP, DHEA-S, 17-OHP, AMP and CAH stimulation testing were all normal. Imaging revealed an infantile uterus and gonads with no ectopic source for the testosterone. Karyotype was 46,XX in 20 cells. 1000/1000 cells showed no signal for either SRY (Yp11.3) or Yq12 (DYZ1). Laparoscopic gonadectomy revealed a left gonad with Leydig cell hyperplasia, Wolffian type ductal remnants, and no ovarian stroma, germ cells or testicular tissue identified. Right gonad revealed Leydig cell hyperplasia, a few Sertoli only seminiferous tubules, Wolffian type ductal remnants, and a streak of ovarian stroma. A normal fallopian tube was noted. Cytogenetic analysis of the gonads showed 46,XX karyotype. 400/400 cells showed no signal for either SRY (Yp11.3) or Yq12 (DYZ1) for each gonad. Testosterone levels dropped to normal female range three weeks after surgery. Microarray was normal. We plan to perform exome sequencing as the next investigational step.

No similar cases have been found in the literature. Due to the 46,XX karyotype and absence of SRY in this patient, we reviewed the active ovarian development pathway, including the literature on patients with gene mutations in WNT4 and RSPO1. WNT4 deficiency causes hyperandrogenism but also uterine agenesis. RSPO1 mutations can cause female-to-male sex reversal, but also causes palmoplantar keratosis and squamous cell carcinoma. This patient had Sertoli and Leydig cells in her gonads, but not the typical pathology or presentation of a Sertoli-Leydig cell tumor. Classifying this patient's findings with the New Chicago Classification of 46, XX Disorders of Gonadal Development (ovotesticular DSD, testicular DSD, and gonadal dysgenesis) is challenging.

When the Parts Don't Fit. Male Or Female?

William Allen, MD

Fullerton Genetics Center, Mission Hospitals, Asheville, North Carolina

JM was seen at 7 2/12 years of age for an initial genetics consultation in January 2000 for evaluation related to “indeterminate sex.” JM was born in 1992 in another location with a normal exam except for findings of the external genitalia described as “consistent with a virilized female as the clitoris protrudes from the perineum… the phallic length is 2.2 cm. The labia demonstrate mild rugations with posterior labial fusion which extends within a few millimeters of the base of the phallus.” The urethrovaginogram revealed a “normal vagina and cervix.” The chromosomes were 46, XY from both blood and skin, with no evidence of mosaicism reported. The 17 hydroxyprogesterone level was 94 ng/dl (wnl), androstenedione was 258 ng/dl (normal 20–2 90), and the testosterone level was 137 ng/dl (normal 20–64 for females; 75–404 males). An initial diagnosis of a “rare form of incomplete male development“ was suggested and the discussion noted that “the appropriate sex of rearing in my opinion is female” and “normal female psychosexual development can be expected… I do not think that this child could achieve function as a well-adjusted adult male.” At 3 months of age surgery was performed, including a bilateral gonadectomy, feminizing genitoplasty with clitoral recession, and vaginoplasty. The gonadal pathology revealed an “immature gonad consisting of an ovotestis, with both immature testicular and ovarian tissue present.”

At the time of our evaluation the parents were concerned because their daughter would not willingly wear dresses, preferentially played with males, and at times would try to go into male restrooms. She told her parents she wanted to be a boy. By report she knew nothing of her prior medical history. JM was referred to a psychiatrist and urologist with extensive experience in disorders of sex development and after an in depth evaluation, education, and counseling, JM was given the choice of being a male or female. By report of his parents, JM immediately chose male and when asked his new first name, he immediately provided his new name.

Possible mechanisms for gender identification and the importance of delayed surgery in individuals with disorders of sex development will be explored.

Presentation of Infants With Prader–Willi Syndrome to a Disorders of Sex Development Clinic for Evaluation of Ambiguous Genitalia

Anita E. Beck, MD, PhD^1,2, Margaret L. P. Adam, MD^1,2, Parisa Salehi, MD^1,2, Linda Ramsdell, MS, CGC¹, Patricia Fechner, MD^1,2, Stephanie Wallace, MD^1,2, Karen D. Tsuchiya, MD¹

¹Seattle Children's Hospital, Seattle, Washington

²University of Washington, Seattle, Washington

Referrals to Disorders of Sex Development (DSD) clinics encompass a variety of different disorders, including ambiguous genitalia and hypogonadism. We report two infants who recently presented to our interdisciplinary pediatric DSD team for evaluation of ambiguous genitalia who were ultimately diagnosed with Prader-Willi syndrome (PWS).

The first child was a female born at 36 weeks to a 46yo, G2P2 mother and a 40yo father who presented with IUGR, hypotonia, respiratory distress, and ambiguous genitalia. The enlarged hooded phallic structure had a urethral opening at the base. Labial folds were fused with no apparent vaginal opening. An abdominal ultrasound showed a duplicated right renal collecting system, no uterus, and possible ovaries in the inguinal canals. Endocrine labs were normal, but the karyotype was 46,XY with normal SRY. A SNP array demonstrated two long regions of uniparental disomy (UPD) on chromosome 15, one of which spanned the PWS/Angelman syndrome region at 15q11.2q12. Methylation testing was consistent with a diagnosis of PWS. After gender re-assignment, he was hospitalized for the next 6 months, predominantly for his respiratory issues which were diagnosed as congenital lobar emphysema of the left upper lobe requiring lobectomy. We will obtain a skin biopsy from his genital region during reconstructive surgery to perform a karyotype for a possible mosaic trisomy 15 line.

The second child was a 7-week-old male born at term to a 39 yo, G5P5 mother and a 40 yo father. His birth parameters (length, weight, and OFC) were all normal, but his extremely poor tone ultimately required G-tube feeding. Concern for possible ambiguous genitalia stemmed from a hypoplastic scrotum with no palpable testes. A pelvic ultrasound did not identify testes or a uterus. A brain MRI was normal. His karyotype was 46,XY, SRY positive. Endocrine studies suggested a normal hypothalamic-pituitary-gonadal axis. Stretched penile length was normal. His scrotum was flat, but testes were later palpable in the inguinal canals. Methylation testing was consistent with a diagnosis of PWS.

Individuals with PWS are often diagnosed in the newborn period because of significant hypotonia and hypogonadism, but frank ambiguous genitalia are rarely described. Both of these children were born to mothers with advanced maternal age, a risk factor for UPD. Maternal UPD is the underlying molecular etiology in ∼20-30% of individuals with PWS, and may be the result of rescue of trisomy 15. Mosaic trisomy 15 with or without PWS has been reported occasionally in the literature. Those with maternal UPD causing PWS and mosaic trisomy 15 have a more severe developmental outcome with additional malformations. One 46,XY male with IUGR and genital anomalies (severe hypospadias, a micropenis, bilateral cryptorchidism and a bifid scrotum) was even demonstrated to have confined placental mosaicism of trisomy 15 without trisomy or UPD of chromosome 15 on postnatal blood or skin. PWS can present with hypotonia and ambiguous genitalia, and individuals with maternal UPD 15 who have a more severe PWS phenotype should prompt consideration of a cryptic trisomy 15 mosaicism.

46,XY Disorder of Sex Development and Congenital Diaphragmatic Hernia: A Unique Case Featuring Dysmorphic Facies, Truncus Arteriosus, Bifid Thymus, Gut Malrotation, Rhizomelia and Adactyly

J.A. Bernstein, MD, E.D. Esplin, MD

Stanford University School of Medicine, Stanford, California

The association of 46,XY disorders of sex development (DSD) with congenital diaphragmatic hernia (CDH) is rare. It has been observed both with and without additional congenital anomalies. We describe an apparently unique case of this association featuring rhizomelia, adactyly, bifid thymus and intestinal malrotation and review the literature related to this entity.

The proband was born at 35 weeks gestation to a G₃P₁SAB₁ mother. There were no reported intrauterine exposures or maternal illnesses. Evidence of truncus arteriosus and ventricular septal defect was noted on 33 week ultrasound. Subsequent amniocentesis karyotype was 46,XY. There was no spontaneous respiration at birth and the infant passed away at ten minutes of life. Post-mortem exam revealed a small appearing nose with slit-like nares, widely spaced eyes and a crescent shaped mouth. Skeletal exam showed rhizomelia and adactyly of the hands and feet with hypoplastic thumbs. There was a Bochdalek left posterior diaphragmatic hernia. The thymus was bifid. Abdominal findings included malrotation and polysplenia. External genitalia were ambiguous with the appearance of a bifid scrotum versus redundant and thickened labial folds. There was no identifiable phallus and the urethra was positioned posteriorly near an atretic anus. The internal genitalia included bicornuate uterus, fallopian tube, prostate, seminal vesicles and epididymis without evidence of gonads.

The patient's presentation is not readily attributable to a well-recognized syndrome. Literature review identified five cases of 46,XY DSD associated with CDH and other congenital anomalies. These 5 cases share characteristics with the present case including CDH, 46,XY karyotype with external female appearing or ambiguous genitalia, splenic abnormality, cardiac defect and decreased life span. However, the present case included several novel features; most distinctively – dysmorphic facies, rhizomelia and adactyly of the upper and lower limbs.

The present case may represent a unique syndrome of 46,XY DSD and diaphragmatic hernia or a more severe presentation of a syndrome represented in the previously reported cases. A strong hypothesis to explain the occurrence of diverse malformations in the context of a 46,XY DSD has not been advanced. DNA microarray analysis in the present case was not revealing. Investigation by exome sequencing is being undertaken. Given the sporadic nature of documented cases the possibilities of new dominant mutations or a multifactorial etiology are considered.

Infant of Diabetic Mother With Complete Tracheal Rings: New Report and Review of the Literature

Louisa C. Pyle, Jesse A. Taylor, Luv R. Javia, Elaine H. Zackai

Children's Hospital of Philadelphia, Philadelphia, Pennsylvania

Congenital tracheal stenosis is a rare finding (<1:65,000 infants) with high mortality, the most common form of which is complete tracheal rings. Among the rare reports of congenital tracheal stenosis are some seen in Infants of Diabetic Mothers (IDM), including two infants who died of this malformation.¹ We present a new report of IDM with complete tracheal rings, review the increased association of this rare finding with IDM, and propose a target pathway for this association.

A 2.715 kg (80th%) 34 5/7 week gestation female was born to a 30-year-old obese mother with Type II diabetes. Patient was conceived naturally and born to non-consanguinous Caucasian parents, with no family history of birth defects or intellectual disability. Prenatal diagnoses on 20 week ultrasound imaging included polyhydramnios, placental insufficiency, severe retromicrognathia, and short femurs. Prenatal microarray was normal. The infant was delivered via planned cesarean section for ex utero intrapartum treatment (EXIT) tracheostomy for retromicrognathia, but the cannula could not be passed. Postnatal evaluation showed long segment tracheal stenosis with 2 cm of complete tracheal rings. Cardiac evaluation showed a normal heart and vasculature. On DOL 6 the infant was transferred to Children's Hospital of Philadelphia for complex airway care, where noted features on exam included normal height, weight, and head circumference for gestational age, severe micrognathia, midline cleft palate, and low set posteriorly rotated ears. Her femurs were now appropriate length for weight and length. Morphology was otherwise unremarkable. Infant underwent successful mandibular distraction and slide tracheoplasty, and was discharged home.

The Neonatal Phenotype of Loeys–Dietz Syndrome

Lawrence C. Wong, Beth F. Printz, Robert Newbury, Marilyn C. Jones

University of California, San Diego and the Rady Children's Hospital, San Diego, California

Loeys-Dietz syndrome (LDS) is an autosomal dominant connective tissue disorder characterized by vascular abnormalities including arterial aneurysms and/or dissections and a variety of skeletal manifestations including pectus excavatum or carinatum, scoliosis, joint laxity, arachnodactyly, and clubfeet. Ocular hypertelorism, bifid uvula/cleft palate and craniosynostosis are common findings. Mutations in TGFBR1, TGFBR2, SMAD3, and TGFB2 account for the phenotype. Yetman and colleagues (2007) emphasized the difficulty of making this diagnosis in the newborn period. Since this report, several other neonatal cases have been described, often with delayed diagnosis. We have recently encountered a baby who presented prenatally with ultrasound findings that, in retrospect, could have suggested the diagnosis of LDS. The purpose of this presentation is to describe this case and review the neonatal phenotype of LDS. The proband was the second child born to healthy unrelated parents, each of whom had normal children with other partners. Mother was 34 and father 32 years of age. Pregnancy lasted 37 4.7 weeks. Ultrasound at 20 weeks documented bilateral clubfeet. Subsequent fetal echocardiogram documented a thickened and restricted pulmonary valve, a bicuspid and thickened aortic valve (neither of which had stenosis or regurgitation), a tortuous patent ductus arteriosus, and a markedly tortuous descending thoracic aorta. The baby was delivered by repeat Cesarean section with Apgar scores of 8 and 8 at 1 and 5 min, birth length of 47 cm, birth weight of 3.07 kg and birth OFC of 35.5 cm. He had mild micrognathia, a cleft soft palate, severe bilateral club feet and striking hand contractures with missing DIP creases on three and four bilaterally. Postnatal echocardiogram showed a bicuspid and dilated pulmonary valve (z score 3.64), a bicuspid and dilated aortic valve (z score of annulus 4.37 and of sinuses 4.44), and a tortuous thoracic aorta. Magnetic resonance angiography confirmed the markedly tortuous aorta, corkscrewing of the innominate artery and vertebral arteries, and an aneurysm of the ductus. Despite his abnormal vasculature, he was asymptomatic from a cardiac standpoint. He fed well using a cleft palate bottle and was discharged at 9 days of age. He was seen at 2 weeks of age by cardiology and was thriving. Serial casts were begun for the clubfeet. At 6 weeks of age, while returning to the hospital for a brainstem audiogram to follow up a “referred” hearing screen, he was noted to be limp and unresponsive. He could not be resuscitated in the emergency room. Aortopathy panel (ARUP) from birth revealed a novel, presumed pathogenic mutation in TGFBR2 (c.1583G>C). On autopsy, there was no evidence of vascular dissection or rupture and no evidence for spinal cord compression. Review of the well-described neonatal cases of LDS suggests that the diagnosis should be considered in the context of an infant with a contractural arachnodactyly or distal arthrogryposis phenotype. Although mean survival in this condition is 26 years, a subset of infants will have severe neonatal complications.

Malan Syndrome: Two New Patients Who Further Expand the Phenotype

Tamison Jewett, MD¹, Chad R. Haldeman-Englert, MD¹, Ashley N. Kaminski, MS²

¹Department of Pediatrics, Section on Medical Genetics, Wake Forest School of Medicine, Winston-Salem, North Carolina

²North Carolina Department of Health and Human Services, Raleigh, North Carolina

Malan syndrome is a recently described, Sotos-like overgrowth condition caused by de novo, heterozygous variants in NFIX (Nuclear Factor I/X; MIM 164005) . Of interest, the mutations associated with Malan syndrome are typically whole gene deletions (as part of a larger deletion), nonsense variants, and missense variants affecting the DNA-binding domain in the 5' end, while frameshift and splice-site variants in exons 6–8 cause the Marshall-Smith syndrome. Herein we describe two, unrelated individuals with changes within NFIX causing Malan syndrome and note the phenotypic overlap with genetic disorders of connective tissue.

Patient 1 is a 12-year-old girl initially evaluated by a geneticist at age 7 due to moderate intellectual disability (ID) and hypotonia. She was noted to have tall stature, macrocephaly, and facial features suggestive of Sotos syndrome; blood was sent for Fragile X and NSD1 testing, which was normal. On subsequent follow-up at age 9, she was diffusely hypermobile, and microarray analysis and bone age study were normal. At age 11, she had developed a marfanoid habitus, and she was referred for an echocardiogram, which revealed a dilated aortic root with Z score = 2.7; FBN1 sequencing was normal. Parents opted for whole exome sequencing, which revealed a c.373_397del25 mutation in NFIX, resulting in a premature stop codon in exon 2 that is interpreted as disease-causing.

Patient 2 is a 3-year-old girl referred at 2 ½ years of age for evaluation due to global developmental delay, hypotonia, and pulmonic stenosis with bicuspid pulmonary valve. Exam revealed normal height and weight, macrocephaly, prominent forehead, epicanthal folds, hooded eyelids, narrow palate, pectus excavatum, tapered fingers, and marked, bilateral genu varum. Soon thereafter, she developed seizures. Chromosome analysis and Fragile X testing were normal; microarray analysis revealed a 68 kb deletion within exon 2 of NFIX. A review of the Decipher and ISCA databases has revealed no other deletion this small in this region.

NFIX encodes Nuclear Factor I/X (NFIX) and is a member of the nuclear factor family proteins, which are site-specific DNA-binding proteins. Haploinsufficiency of NFIX leads to Malan syndrome, while dominant negative effects of a more terminally truncated NFIX protein cause Marshall-Smith syndrome. Common features of Malan syndrome include postnatal overgrowth, macrocephaly, down-slanting palpebral fissures, long face, blue sclerae, myopia, pectus excavatum, hypotonia, and intellectual disability. The two patients described here have involvement of the cardiovascular system, and one other reported patient has aortic root enlargement, making it important to recognize the phenotypic overlap between Malan syndrome and genetic disorders of connective tissue, which may become more apparent with age.

Vascular Ehlers–Danlos Associated With Neonatal Gastric Rupture and Severe Kyphoscoliosis

M. Bocian, MD

Division of Genetic and Genomic Medicine, Department of Pediatrics, University of California, Irvine Medical Center, Orange, California

The vascular type of Ehlers–Danlos syndrome (type IV, vEDS) is characterized by thin, translucent skin, easy bruising, arterial, intestinal, and/or uterine fragility, and, in some individuals, characteristic facial appearance and acrogeria. Vascular dissection or rupture, intestinal perforation, or organ rupture are the presenting signs in the majority of adults with vEDS. Neonates can present with clubfoot (12%) and/or congenital hip dislocation (3%), limb deficiency, or amniotic bands. Children may have inguinal hernia, pneumothorax, and/or recurrent joint dislocation or subluxation, but major complications are uncommon.

A 15-year-old boy had spontaneous neonatal gastric rupture and onset of severe, progressive kyphoscoliosis at 2 years. He was born at term, AGA, with no complications. At 3 days he had an acute abdomen; the entire stomach was necrotic with a large rupture that the surgeon described as “blown out.” He had spinal fusion and rod placement at age 14 and developed postoperative hardware failure. He has a normal face, acrogeria, and mildly thin, translucent skin; he has not had bruising, joint laxity, hernia, fatigue, chronic joint/limb pain, vascular/arteriole rupture, aneurysm, stroke, hemorrhage, muscle/tendon rupture, other hollow organ perforation/rupture, intra-operative tissue fragility, problems with sutures, or prolonged healing. There is no family history of scoliosis or of any features of EDS, except that his mother had an arterial aneurysm in her neck at age 41. Mutation analysis of COL3A1 in the University of Washington Collagen Diagnostic Laboratory revealed a heterozygous pathogenic missense mutation in exon 32: c.2185G>A, p.Gly729Arg. The laboratory has seen this mutation in two other unrelated individuals, both with clinical presentations typical for vEDS. Neither had kyphoscoliosis or major complications in childhood. The etiology of neonatal gastric rupture is not understood; hypotheses or the ∼110 reported cases include congenital gastric muscle wall defects, mechanical disruption, stress ulceration due to neurogenic disorders, gastric wall ischemia due to vascular shunting, birth trauma, hypoxia, and increased intra-luminal pressure. EDS was not considered in any report. Gastric rupture occurs infrequently in adults with vEDS and has not been described in any neonate or child with any type of EDS. There is a report of a 17-year-old woman with an unspecified type of EDS who had an idiopathic gastric infarction and scoliosis. Another patient with vEDS, without scoliosis, had an infarction due to gastric strangulation from displacement through the diaphragm. Neither had molecular testing. Neonatal gastric rupture should raise suspicion for vEDS. Kyphoscoliosis is common in EDS-kyphoscoliotic type (type VI) and Loeys-Dietz syndrome, both of which can overlap clinically with vEDS. There are only three reports of scoliosis with vEDS—two had a missense substitution for glycine, and the third was not tested. Scoliosis in our patient may be unrelated to EDS, or modifying genes may interact with this mutation.

Mutations In LTBP3 Gene, a TGF-Beta Protein, Cause a Rare Syndromal Form of Amelogenesis Imperfecta

O. Caluseriu¹, R.D. Mcleod², M.C. Cholette², M. Huckert³, C. Stoetzel³, S. Morkmued³, V. Lauger- Haushalter³, J. Hemmerle³, B. Brukner Dabovic⁴, D.B. Rifkin⁴, A. Dheedene⁵, E. Boudin⁶, R. Antequera⁷, M.P. Gelle⁸, W. Van Hul⁶, D. Bertola⁷, P. Dolle³, A. Verloes⁸, G. Mortier^5,6, H. Dollfus³, A. Bloch-Zupan³

¹University of Alberta, Edmonton, Canada

²University of Calgary, Calgary, Canada

³University of Strasbourg, Strasbourg, France

⁴NYU Langone Medical Centre, New York

⁵Ghent University, Belgium

⁶University of Antwerp, Belgium

⁷University of San Paolo, Brazil

⁸Department of Genetics, Robert Debre Hospital, Paris, France

Brachyolmia is a feature of over 200 genetic conditions, but has rarely been associated with amelogenesis imperfecta (AI). In 1996, Verloes and colleagues have proposed a new recessive condition characterized by brachyolmia and AI when describing two siblings born to consanguineous parents who had short stature, platyspondyly, and teeth anomalies including oligodontia and AI (MIM 601216). Recently, mutations in LTBP3 gene, a member of TGF-beta signaling pathway, have been reported in four families with this rare association of features (Huckert et al., 2015). In this paper, we are describing in detail one of these families, and discuss the role of LTBP3 in dental anomalies and skeletal development using an animal model.

Our family includes a sibship of five with three affected children born to healthy, normal statured, first cousin parents of Pakistani origin. Two girls and a boy, aged 16, 9, and 12 years respectively, non-dysmorphic and of normal intelligence, developed short trunk, short stature over time (height: −5 SD, 3rd–10th centile and −3 SD respectively), have normal head circumference, normal bone age and small hands. The oldest girl has bilateral coxa valga and S shape scoliosis of the mid-thoracic to lumbar spine which required corrective surgery. The other two siblings have platyspondyly, coarse trabecular pattern of the bones, and lower thoracic-lumbar scoliosis. Oligodontia and recurrent dental abscesses were noted in all three affected members and a subsequent diagnosis of AI was confirmed by dental pathology.

This family has been part of an international cohort* in an effort to elucidate the molecular basis of their disorder using a combined strategy of homozygosity mapping and whole-exome sequencing. A homozygous mutation in the LTBP3 gene (p.Val786Trpfs*82; Huckert et al., 2015) was found in all affected members of our family. LTBP3 codes for a protein of the extracellular matrix which is involved in the regulation of the TGF-β secretion. Inactivation of LTBP3reduces TGF-β activation and therefore diminishes associated cell proliferation and osteogenic differentiation. The Ltbp3 −/− mouse phenotype includes growth retardation, skull and bone defects (Dabovic et al., 2002) as well as enamel defects as shown by our study (Huckert et al., 2015). Our research confirms the molecular basis of a new form of syndromic brachyolmia and supports the role of LTBP3and TGF-β signaling in amelogenesis imperfecta both in mice and humans.

*Support provided by INTERREG IV, Offensive Sciences A27, an EU-funded project (ERDF) in the TMR, www.genosmile.eu

Prune Belly Syndrome as a Manifestation of a Connective Tissue Disorder

Holly H. Ardinger, MD, Caitlin E. Lawson, MS, Janda L. Jenkins, MS, Robert H. Ardinger Jr., MD

Children's Mercy Kansas City and University of Missouri-Kansas City, Kansas City, Missouri

Prune belly syndrome (PBS) is defined as a triad of abdominal wall laxity, undescended testes and urinary tract abnormalities. Some patients present with the typical urinary anomalies (hydronephrosis, dilated tortuous ureters, renal dysplasia, enlarged bladder) with little to no abdominal laxity and no cryptorchidism. This may be referred to as pseudo or partial PBS or as megacystis-megaureter syndrome with the latter being accompanied by microcolon in some cases. Prenatal urethral obstruction has been considered as an etiologic factor of PBS, but is not present in many with PBS and cannot completely account for the histologic findings of the abdominal wall or bladder wall musculature. Causative factors for PBS are still to be identified.

We report a 13-year-old male referred for evaluation due to his history of PBS and congenital knee dislocation. Prenatal ultrasound identified enlarged kidneys, bladder and ureters with normal amniotic fluid. Postnatal evaluation identified massively dilated ectatic right ureter with VUR and little right kidney function. The left side had ureteral ectasia without VUR and good renal function. The bladder was poorly emptying with no evidence for urethral obstruction. He was treated with early nephrostomy tubes and a later nephrectomy with normal renal function. He had abdominal distention at birth with only mild laxity of the abdominal wall without cryptorchidism. Knee function improved following serial casting. His history was remarkable for normal growth (60th %ile height and weight) and cognitive skills, bilateral inguinal hernia repairs, knee instability and significant scoliosis requiring bracing starting at age 10 years. On exam he was non-dysmorphic with generalized joint hypermobility and soft, non-fragile skin with normal elasticity and translucency. Microarray was normal. Aortic root diameter was in the high normal range with a Z score of 1.9. A multi-gene panel identified a novel likely pathogenic variant in TGFBR2 (p.Pro374Ser) consistent with Loeys-Dietz syndrome (LDS).

PBS has not been previously reported in LDS. While the PBS may have occurred coincidentally in this patient, there is evidence that connective tissue disorders, especially those associated with aortopathy or changes in the TGFß pathway, may contribute to its occurrence. First, there seems to be an unexplained increased incidence of orthopedic issues (most commonly scoliosis, pectus deformity, hip dysplasia) in those with PBS. Secondly, there are case reports of PBS and/or megacystis-megaureter occurring with mutations in other genes associated with aortopathies including FLNA, ACTA2, MYH11, and ACTG2. Lastly, mouse studies have shown that mutations in SMAD4 and HPSE2, genes which function in the TGFß pathway, result in megaureter and megacystis respectively.

Fetal Dysmoprhology: A Step By Step Multidisciplinary Approach

Miguel del Campo, Mar Borregan, Irene Valenzuela, Anna Cueto, Teresa Vendrell, Carmen Mediano, Alberto Plaja, Carlota Rodó, Silvia Arévalo, Maria Angeles Sánchez, Elena Carreras, Eduardo Tizzano

Genetics and Fetal Medicine Team. Hospital Vall d́Hebron, Barcelona, Spain

Prenatal diagnosis of birth defects is constantly evolving in the direction of earlier and more specific etiologic diagnoses, thanks to the use or prenatal arrays, next- generation sequencing and better imaging techniques. However, genetic evaluation of the family and previous medical history, along with prenatal discussion of cases with a focus on fetal dysmorphology, participation in fetal autopsies, and guidance for monogenic testing, have proven to be efficient methods for prenatal definition of an even higher number of specific diagnoses. The impact of the clinical use of prenatal arrays has proven to increase detection of pathogenic CNVs (8%), with 9% uncertain variants (VOUS) in 49 of 280 prenatal arrays between June 2010 and December 2013. The detection has been highest in cases with ultrasound findings 17% out of 203 arrays, and was further increased when IUGR, soft ultrasound markers, minor malformations or dysmorphic facial appearance on 3D ultrasound was present aswell. In our team, the array is ordered after an evaluation by Medical Genetics; IUGR or minor malformations are highly predictive markers of an abnormal array, but the structural severity of the malformations is not; when the structural defects are severe enough to make a decision on interruption of pregnancy, samples should be taken and a consent for autopsy or further imaging should be sought. Only after multidscipinary review of the case, including the autopsy, further genetic testing should be offered if necessary, often not arrays. The likelihood of incidental and uncertain findings, as well as the need to consider the familýs views on their reproductive options require professional genetic counseling during the process. The Genetics and Obstetrics teams contribute together at the Fetal Medicine Committee, Dysmorphology autopsies, and the Genetic and Reproductive Counseling clinic. Several cases will be discussed, with an emphasis on the contribution of fetal dysmophology to the diagnosis, its impact on definition of prognosis, decisions on interruption of pregnancy, and future reproductive options.

Spina Bifida Phenotypes and Sex-Specific Prevalence Among Offspring of Folic Acid Users

Calvin Patimeteeporn¹, Cynthia A. Moore², and R.J. Berry¹

¹Emory University School of Public Health, Atlanta, Georgia

²National Center on Birth Defects and Developmental Disabilities, Centers for Disease Control and Prevention, Atlanta, Georgia

Periconceptional consumption of folic acid can reduce a woman's risk of having a neural tube defect (NTD)-affected pregnancy, particularly in high-prevalence areas such as northern China; however, information about the effect of folic acid on sex-specific or phenotype-specific spina bifida prevalence is lacking. We used data from the China-U.S. NTD Prevention Project to evaluate these effects in a high-prevalence region. From 1993 through 1995, women in five counties in northern China were asked to take a daily pill containing 400 µg of folic acid without other vitamins. Among 21,536 women who took folic acid at any time before and during early pregnancy and 56,541 women who did not take folic acid, we identified 18 and 115 women, respectively, with spina bifida-affected pregnancies of ≥ 20 weeks' gestation. Among the women who did and did not take any folic acid, the prevalences of spina bifida were 8.4 per 10,000 pregnancies and 20.3 per 10,000, respectively (risk reduction ∼59%).

Among women who did and did not take any folic acid as part of the prevention project, the spina bifida prevalences among female offspring were 8.8 and 24.2 per 10,000 pregnancies, respectively (risk reduction ∼64%), and the prevalences among male offspring were 8.0 and 17.0 per 10,000, respectively (risk reduction ∼53%). Thus, among women who consumed 400 µg of folic acid daily, those with female offspring experienced a greater reduction in risk than those with male offspring.

We also examined the distribution of spina bifida phenotypes–categorizing lesions in the cervicothoracic region as high lesions and those in the lumbosacral region as low lesions. Among women who did and did not take any folic acid, the prevalences of high spina bifida lesions among offspring were 3.3 and 11.0 per 10,000 pregnancies, respectively (risk reduction ∼70%), and prevalences of low spina bifida lesions among offspring were 4.6 and 8.5 per 10,000 pregnancies, respectively (risk reduction ∼45%). This difference was statistically significant at P<0.01 and among women with female offspring, the risk reduction was even greater for high lesions vs. low lesions.

These findings suggest that the folic acid effect on neural tube closure is not limited to any one spina bifida phenotype; however, it appears that the effect may differ by level of lesion and offspring sex. The findings are also consistent with observations in seven Canadian provinces where the proportion of cervicothoracic lesions decreased in all regions after fortification. Examination of these variations in folic acid effects may be helpful in identifying factors important in the pathogenesis of spina bifida and other neural tube defects.

Pregnancy and Turner Syndrome

Nicole K. Banks, MD^1,2, Paul S. Kruszka, MD¹, Yonit A. Addissie BA¹, Ariel F. Martinez, MS¹, Rachel Hart, BA¹, Carolyn Bondy, MD², Maximilian Muenke, MD¹

¹Medical Genetics Branch, National Human Genome Research Institute, NIH

²National Institute of Child Health and Human Development, NIH

Turner syndrome is due to partial or complete loss of the second sex chromosome and affects 1 in 2500 live-born female infants. Approximately half of Turner syndrome females have a classic 45,X karyotype, while the remainder demonstrate mosaic karyotypes or structural abnormalities of the second sex chromosome, which may be an X or a Y. The two most prominent phenotypic findings are short stature and primordial follicle depletion. Normal ovarian development occurs in the fetus, but accelerated follicular atresia leads to the development of gonadal dysgenesis and infertility. One possible explanation for the increased rate of oocyte attrition is haploinsufficiency for multiple loci on the X-chromosome. Alternatively, oocyte depletion may be secondary to dysfunctional meiosis caused by errors in homologous pairing, as the solitary X is lacking an appropriate homologue. Despite this, spontaneous puberty and pregnancy are reported in approximately 15% and 3% of Turner syndrome patients respectively; however, the majority of spontaneous pregnancies with live born deliveries are in mosaic females.

The National Institutes of Health evaluated 276 adult females with cytogenetically proven Turner syndrome, and as previously reported, five of these women had spontaneous pregnancies and delivery despite 45,X karyotype. Three of the five women are available currently for further study. Two of the three women had a singleton pregnancy with delivery at term and one had two full term pregnancies. All three women have since developed primary ovarian insufficiency (ovarian failure). There is one published report of a duplication in a Turner syndrome patient that included duplication of Bone Morphogenic Protein-15 (BMP15), possibly resulting in preserved fertility and suggesting that single genes may impact follicular maintenance in Turner syndrome. BMP15 is located at Xp11.2 and plays an important role in regulation of folliculogeneis and granulosa cell proliferation and mutations in BMP15 cause primary ovarian insufficiency. We performed next generation whole exome sequencing with the NIH Intramural Sequencing Center (NISC) to investigate the possible protective role of single genes in fertility and Turner syndrome. BMP15 was hemizygous, as would be expected, and did not contain any significant variants in our three patients. We plan to investigate other genes known or hypothesized to play a role in ovarian failure or gonadal dysgenesis in this small case series of patients. We also plan to analyze SNP array data to look for larger deletions or duplications that may impact fertility. Genomic study of women with 45,X and preserved fertility may illuminate genomic regions or single genes with critical roles in ovarian function.

A Prenatal Lethal Phenotype Without Renal Disease Is Caused by Mutations in RPGRIP1L: Expanding The Meckel/Joubert Clinical Spectrum

C.J. Curry¹, J. Fisher², J. Pollard², E.A. Woods³, I.G. Phelps⁴, K. Aldinger⁴, W.B. Dobyns⁴, M. Volleth⁵, S. Schulz⁶, D. Doherty⁴

¹Genetic Medicine/University of California, San Francisco/Fresno, California

²Genetic Medicine and Pathology, Community Regional Medical Center, Fresno, California

³Central California Prenatal Diagnostics, Fresno, California

⁴Department of Pediatrics, University of Washington and Seattle Children's Hospital

⁵Human Genetics, Magdeburg University Hospital, Magdeburg, Germany

⁶Human Genetics, Jena University Hospital, Jena Germany

We report two affected fetuses born to non-consanguineous parents who had the ultrasound diagnosis of lethal abnormalities in the early second trimester. The first, delivered at 13 6/7 weeks, had exencephaly (?encephalocele), absent eye fields, absent nose, a central cleft lip, small tongue with ankyloglossia, postaxial polydactyly of the hands and feet with seven fingers and 6–7 syndactyly, and six toes. Genitalia were ambiguous with a micropenis. No internal exam was performed. The second fetus, delivered at 19 2/7 weeks, had an occipital encephalocele, miniscule palpebral fissures with absent optic nerves, broad tubular nose with single nares, absent maxillary frenula, postaxial polydactyly, micropenis and histologically normal kidneys and liver. Whole exome sequencing revealed compound heterozygous mutations in RPGRIP1L, one frameshift (p.Val548Serfx*9) and one stop-gain (p.Lys233*), one carried by each parent. In 2007 Delous et al reported mutations in RPGRIP1L in nine families negative for known Meckel (MKS) and Cerebello-Oculo-Renal Syndrome (CORS)/ Joubert B (JS) genes. Cystic renal disease was seen in all; 3 had microphthalmia and three had median cleft lip as did patient one above. Arts et al. (2007) reported two families with JS and RPGRIP1L mutations, one of whom had an occipital encephalocele suggesting overlap with MKS. Knockout mice with predicted complete loss-of-function Rpgrip1l mutations have a perinatally lethal presentation with severe brain and eye malformations, along with renal cysts and liver changes. In mice, Rpgrip1l establishes L/R asymmetry and neural tube and limb patterning and interaction with the Hedgehog signaling pathway is likely. RPGRIP1L encodes a ciliary protein localized to the basal bodies, ciliary axoneme, and centrosome or cytoplasm, and interactions with nephrocystin four have been documented. Review of the literature revealed several rather similar cases under the name “Pseudotrisomy 13.” The family reported by Schulz et al in 2005 had holoprosencephaly (HPE), encephalocele and polydactyly and was the closest phenotypic match to our cases. DNA from a fibroblast cell line from their family underwent targeted sequencing of a 33 gene ciliopathy panel. This is negative to date but further studies are in process. Our family expands the phenotype of RPGRIP1L mutations suggesting that renal and hepatic changes are not obligatory features in some severe MKS/JS cases and that the overlap of MKS and JS is broader than previously assumed. This particular presentation with HPE, encephalocele and polydactyly is likely an early lethal, possibly not previously described.

Effects of Digoxin and Delayed Dilatation and Evacuation on Fetal Tissue Quality: Maximizing Opportunities for Research Participation

I.A. Glass, MBChB, MD¹, E. Micks, MD², M. Deng, MD¹, D. O'Day, PhD¹, T. Nalui-Checchini¹, J. Lam, MS¹, D. Doherty, MD, PhD¹, S. Prager, MD²

¹Department of Pediatrics, University of Washington, Seattle, Washington

²Department of Obstetrics and Gynecology, University of Washington, Seattle, Washington

A majority of women undergoing abortion approve the use of fetal tissue for research. The effects of digoxin injection and delayed dilatation and evacuation (D&E) on fetal tissue quality are unknown. The objective of this study was to determine whether fetal tissues are usable for research after digoxin injection and delayed D&E. Fetal tissues exposed to digoxin one day prior to D&E, were compared to samples from fetuses not exposed to digoxin and delayed D&E. Histological analyses included hematoxylin and eosin (H&E) for cell morphology, Ki67 staining for cell proliferation and TUNEL cell death assay. Standard primary fibroblast and astrocyte cultures were attempted. Nucleic acid extraction and quality assays were performed. Tissue was obtained from six 117–178.5 day post-conceptual age fetuses injected with digoxin, 22.5–27.5 hr before D&E. These tissues showed abnormal cell morphology, increased apoptosis, and decreased cell proliferation compared to controls. Fibroblast culture was successful from two of three fetuses. High-quality genomic DNA was obtained in all cases with RNA Integrity Number (RIN) scores (brain, lung, muscle) ranging from 5.0 (partially degraded) to 7.6 (suitable for RNA analysis). Although fetal tissues obtained after digoxin exposure and delayed D&E had abnormal cell morphology, poor cell viability, and variable RNA quality, high-quality genomic DNA was consistently isolated. The independent effects of digoxin exposure versus time delay at body temperature pre-D&E on tissue viability remain to be determined.

Russell–Silver Syndrome: Prenatal and Placental Findings

Joanna Lazier¹, Kathryn Millar^2,3, Sarah Keating⁴, Wendy Whittle¹, Rosanna Weksberg⁵, Cheryl Cytrnbaum⁵, David Chitayat^2,3,5

¹Department of Medical Genetics, Alberta Children's Hospital, University of Calgary, Calgary, Alberta, Canada

²Department of Obstetrics and Gynaecology, Mount Sinai Hospital, University of Toronto, Toronto, Ontario, Canada

³The Prenatal Diagnosis and Medical Genetics Program, University of Toronto, Toronto, Ontario, Canada

⁴Department of Pathology and Laboratory Medicine, Mount Sinai Hospital, Department of Paediatrics, University of Toronto, Toronto, Ontario, Canada

⁵Division of Clinical and Metabolic Genetics, University of Toronto, Toronto, Ontario, Canada

Russell–Silver syndrome (RSS) is etiologically a heterogeneous condition, and most often caused by hypomethylation of imprinting centre (IC) 1 at 11p15 or maternal UPD 7. It is characterized by postnatal growth deficiency with relative macrocephaly, distinctive facial features, body asymmetry, and feeding difficulties. Prenatally, patients may present with asymmetrical intrauterine growth restriction (IUGR) with a relatively normal sized head. Placental dysfunction with abnormal placental morphology and oligohydramnios has been previously reported in patients with RSS. Little, aside from growth parameters have been reported regarding the prenatal presentation of RSS, especially in those patients without placental dysfunction.

We report a case of RSS with mosaic hypomethylation of IC1 at 11p15 who presented prenatally with early IUGR and mild cerebral ventriculomegaly, but without clear signs of placental insufficiency. Prenatal testing on amniocytes showed normal methylation at 11p15; however, postnatal testing of blood and buccal samples showed hypomethylation of IC1 at 11p15 and negative testing for matUPD 7. The fact that the patient had significant growth restriction at birth despite normal placental functioning suggests that the prenatal growth restriction in RSS is, at least partly, independent of placental dysfunction.

Determining a clear fetal phenotype in RSS can be complicated by multiple etiologies, complex gene interactions at the placental level, mosaicism and the co-occurrence of less common anomalies. We propose that RSS should be considered in the differential diagnosis of cases of early isolated asymmetrical IUGR with normal placental function and structure, no evidence of intrauterine infection and normal microarray analysis, and that, due to mosaicism, this diagnosis should not be excluded on prenatal amniocyte testing alone.

Prenatal Presentation of CHST3-related Autosomal Recessive Larsen Syndrome

Maian Roifman^1,2, Hana Sroka², Sarah Keating³, Ants Toi⁴, Roberto Mendoza¹, Aleksander Hinek⁵, Sheila Unger⁶, Gene Nishimura⁷, David Chitayat^1,2

¹Division of Clinical and Metabolic Genetics, Department of Pediatrics, Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada

²The Prenatal Diagnosis and Medical Genetics Program, Department of Obstetrics and Gynecology, Mount Sinai Hospital, University of Toronto, Toronto, Ontario, Canada

³Department of Laboratory Medicine and Pathobiology, Mount Sinai Hospital, University of Toronto, Toronto, Ontario, Canada

⁴Department of Diagnostic Imaging, Mount Sinai Hospital, University of Toronto, Toronto, Ontario, Canada

⁵Physiology and Experimental Medicine, Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada

⁶Medical Genetics Service, CHUV, University of Lausanne, Switzerland

⁷Génétique Médicale, CHUV, Av. Decker 2, 1011 Lausanne, Switzerland

Autosomal recessive Larsen syndrome is a progressive spondlyoepiphyseal dysplasia characterized by multiple congenital joint dislocations, club feet, short stature, limited range of joint movement, kyphoscoliosis and heart valve dysplasia. The only gene associated with this syndrome is carbohydrate sulfotransferase 3 (CHST3), a catalyst for chondtroitin sulfate proteoglycan sulfation. While stature may be short at birth, prenatal diagnosis of AR Larsen syndrome is rare. We report on an affected male with AR Larsen syndrome who presented with shortening and bowing of the long bones at 19 weeks gestation. The mother was a healthy 28 year old G2P1L1 woman with a family history of early onset osteoarthritis requiring joint replacement. Parental heights were at the 50th centile. The pregnancy was uncomplicated and integrated prenatal screening was negative. A detailed fetal ultrasound at 23 weeks gestation was consistent with a non-lethal skeletal dysplasia. The couple elected to terminate the pregnancy and autopsy done at 23+6 weeks gestation revealed normal general growth parameters and facial features, shortened humeri and femurs, bilateral limitation of elbow extension and hip abduction, dilated ureters and bicuspid pulmonary valve. Fetal X-rays revealed bilateral micromelia, misalignment of the elbows, mild bowing of the ulnae, femurs and tibias, small pelvis with hypoplastic acetabular notch, narrow interpedicular spaces, mild platyspondyly and vertebral coronal clefting, suggestive of AR Larsen syndrome. Molecular analysis of the CHST3 gene revealed a novel truncating mutation (c.334G>T) and a previously described disease-causing missense mutation (c.475T>A) in trans, each inherited from one parent. Dermal fibroblast cell analysis was abnormal and suggestive of impaired CHST3- mediated sulfation. To the best of our knowledge, our report is the second case describing a prenatal presentation of AR Larsen syndrome, with findings of short and bowed long bones and multiple joint dislocations, confirmed on molecular analysis of the CHST3 gene. The ultrasound findings can help in delineating the condition prenatally.

Prenatal Caffey Disease: A Tale of Two Siblings

Maha Saleh^1,2, Patrick hannon³, Ants Toi⁴, Rachel Silver², Ronni Teitelbaum², Sarah Keating³, Ori Nevo⁵, David Chitayat^1,2

¹Division of Clinical and Metabolic Genetics, Mount Sinai Hospital, Department of Obstetrics and Gynecology, The Hospital for Sickkids, Department of Pediatrics, Ontario, Canada

²The Prenatal Diagnosis and Medical Genetics Program, Ontario, Canada

³Department of Laboratory Medicine and Pathobiology, Ontario, Canada

⁴Department of Diagnostic imaging; Sunnybrook Health Science Center, Ontario, Canada

⁵Department of Obstetrics and Gynecology, The University of Toronto, Toronto, Ontario, Canada

Caffey disease or cortical hyperostosis is a condition characterized by massive subperiosteal new bone formation usually involving the diaphyses of the long bones. Two forms of Caffey disease have been identified 1) The classical mild, transient infantile form typically associated with fever, joint swelling and pain with onset around age two months and 2) the Lethal prenatal form which typically presents before 35 weeks' gestation, characterized by cortical hyperostosis, short, thick and angulated long bones associated with polyhydramnios and pulmonary hypoplasia. We report two cases of prenatal Caffey disease born to healthy and non-consanguineous parents of Indian-Sikh descent.

Case 1: The mother was 26 years old G2P1L1 and the father was 28 years old. The couple was healthy and had a son who was well. The pregnancy was initially uncomplicated. Detailed fetal ultrasound at 20 weeks gestation was normal and there was no exposures to teratogens. At 30 weeks gestation an ultrasound revealed polyhydramnios, all long bones were below the 5th centile in length, angulated and thick and there was scalp edema. Echocardiography was normal. No hydrops fetalis was detected. The couple decided to terminate the pregnancy and the autopsy showed cortical hyperostosis of all long bones. The fetal autopsy showed a female fetus with X-rays and histopathology consistent with cortical hyperostosis.

Case 2: The couple embarked on their 3rd pregnancy and this was initially uncomplicated. A fetal ultrasound at 29 weeks gestation showed polyhydramnios, all long bones were thick, bowed and measured below the 5^th centile, the chest was narrow, there was frontal bossing with forehead edema, echogenic gut, absent stomach bubble, large IVC and a large and bulky placenta. The couple was counseled regarding the recurrence of cortical hyperostosis and decided to terminate the pregnancy. The fetal autopsy showed a male fetus with X-rays and histopathology consistent with cortical hyperostosis. DNA analysis of the COL1A1 gene and a whole exome sequencing (WES) showed no detectable abnormalities.

The occurrence of the same condition in sibs of both sexes born to unaffected parents is consistent with an autosomal recessive mode of inheritance. However, the negative WES raises the possibility that this prenatal and lethal condition is the result of an autoimmune condition caused by maternal antibodies directed towards a component of the fetal skeleton.

Prenatal Ultrasound Findings of Soft Markers and Fetal Hydrops in a Child With Kabuki Syndrome

Samantha A. Schrier Vergano¹, Benjamin M. Helm¹, Steven L. Warsof², Lauren Grimes²

¹Division of Medical Genetics and Metabolism, Children's Hospital of The King's Daughters, Norfolk, Virginia

²Department of Obstetrics and Gynecology, Division of Maternal and Fetal Medicine, Eastern Virginia Medical School, Norfolk, Virginia

Kabuki syndrome (MIM 147920) is a well-described, multiple congenital anomaly syndrome characterized by unique facial features, persistent fetal finger pads, and a variety of defects involving organ systems, growth, cognition and development. Mutations in the lysine-specific methyltransferase 2D gene, KMT2D (MLL2; MIM 602113), and KDM6A (MIM 300128), a lysine demethylase, have been found in an estimated 50–75% of individuals with a clinical diagnosis of Kabuki (Ng et al., 2010). Genetic heterogeneity in the syndrome indicates that other causative genes have yet to be found. Retrospectively, prenatal ultrasounds of children subsequently diagnosed with Kabuki have not demonstrated any detectable features of the disorder to distinguish it from other syndromes with congenital anomalies.

We report here a now 2-year-old girl, only recently diagnosed with Kabuki, who was initially seen as a newborn for congenital anomalies and IUGR in December of 2012. Her mother was followed by our maternal fetal medicine group during her pregnancy for IUGR and multiple “soft” signs concerning for aneuploidy. An infectious work-up was negative. Beginning at approximately 19 weeks gestation, prenatal ultrasounds revealed a two- vessel umbilical cord, a thickened nuchal fold, bright echogenic bowel, and unilateral pyelectasis. Amniocentesis resulted in a normal 46, XX karyotype and a normal prenatal SNP array. Significant early growth restriction (<5th percentile) was noted at approximately 22 weeks, along with a shortened femur length. Beginning at 26 weeks, the fetus was found to have abdominal ascites, which worsened and developed into hydrops throughout the course of her gestation until she was delivered at 39 weeks due to non-reassuring fetal heart tones.

The child was recently seen in our genetics clinic in January of 2015, not having been evaluated since her birth. She demonstrated the classic “Kabuki make-up” facial appearance with eversion of the lateral lower eyelids, persistent fetal finger pads, growth failure, premature thelarche, and microcephaly. Sequencing of the KMT2D gene revealed a c.3121C>T(p.Q1041X) pathogenic variant. To our knowledge, there have been no studies examining the prenatal ultrasounds of individuals subsequently diagnosed with Kabuki. Although this child's fetal abnormalities were relatively non-specific, additional information regarding these prenatal findings in other individuals will help to expand the phenotypic and molecular understanding of this fascinating syndrome.

Correlates of Craniosynostosis: Sex, Drugs, & Rock And Roll

M. Cunningham, MD^1,2, S. Park, BS¹, C. Clarke, BS¹, J. Gustafson, BA¹, N. Sniadecki, PhD³, A. Karchin, PhD³, Z. Al-Rekabi, PhD³, A. Leonard, PhD³

¹Seattle Children's Craniofacial Center and Center for Developmental Biology and Regenerative Medicine, University of Washington, Washington

²Department of Pediatrics, University of Washington, Washington

³Department of Mechanical Engineering, University of Washington, Washington

Isolated single suture craniosynostosis (SSC) is a common human malformation with genetic and environmental contributions. Male infants are approximately four times as likely as females to be born with sagittal or metopic craniosynostosis and the recurrence risk for affected individuals is approximately 7%. There is a strong association between SSC and maternal thyroid disease and gestational exposure to thyroxine and we have also identified a correlation between craniosynostosis and enhanced cellular differentiation. Using a well curated cohort of 200 human osteoblast cell lines from patients with SSC we have utilized a number of strategies to phenotype these cells (proliferation, differentiation, global gene expression, migration and measurement of single cell strain force generation) to characterize SSC subgroups with distinct biologic features.

Our data demonstrates that enhanced differentiation, reduced proliferation, increased strain force generation and inhibition of migration are key features of osteoblast cell lines derived from patients with SSC. Furthermore, correlation analysis between global gene expression and these features of differentiation have identified sex and suture specific subgroups.

We will present new data supporting our approach as a means to define SSC subgroups, including a striking correlation between sex and genes correlated with differentiation. Using computational biology and our rich data source, our labs continue to define candidate genes and pathways that may provide targets for translational approaches to modify the natural history of craniosynostosis.

Comparing the Features and Proposed Mechanism of 5q23 Deletion Syndrome Involving the Loss of Phax With Those of Three Known Ribosomopathies

Linda Zhong, MS4¹, Paul Mark, MD²

¹Michigan State University College of Human Medicine, Grand Rapids, Michigan

²Spectrum Health Medical Genetics, Grand Rapids, Michigan

5q23 deletion syndrome encompasses a range of physical findings involving the craniofacial and appendicular skeleton as well as various other abnormalities in normal development. A two-year-old female patient was seen in our clinic with a 13.34 MB deletion spanning 5q23.1–5q23.3 with manifestations including, but not limited to, Pierre Robin sequence (micrognathia, glossoptosis, and obstruction of the upper airway), cleft palate, and bilateral talipes equinovarus. One of the genes lost within this deletion is PHAX (phosphorylated adaptor for RNA export). PHAX is critically important for the export of small nuclear RNAs (snRNAs) from the nucleus to the cytoplasm, where they are then used in the process of proper spliceosome formation. Spliceosomes are required for splicing pre-mRNA into mature mRNA ready for translation. In addition, PHAX also plays a role in the intranuclear transport of certain small nucleolar RNAs (snoRNAs), which are essential for ribosome biogenesis and, in turn, protein synthesis.

We will review the function of PHAX in various pathways leading to the creation of spliceosomes and ribosomes. We will also compare the overlapping features found in our patient with PHAX haploinsufficiency to those found in patients with three other ribosomopathies. These include Treacher-Collins syndrome, a primarily autosomal dominant condition resulting from a mutation in TCOF1 or POLR1D; Bowen-Conradi syndrome, an autosomal recessive condition involving a mutation in EMG1; and Miller syndrome (postaxial acrofacial dysostosis), another autosomal recessive disorder resulting from decreased activity of the enzyme dihydroorotate dehydrogenase (DHODH). The defects in these ribosomopathies all lead to errors in the ribosomal formation pathway. We will discuss the possibility that haploinsufficiency of PHAX in our patient also results in disordered ribosome biogenesis, and that this mechanism may explain some of the common physical features shared between our patient and the three aforementioned ribosomopathies, specifically, micrognathia and cleft palate.

Novel Terminal Deletion of 3q27.2 in a Patient With Symptoms Consistent With Diamond–Blackfan Anemia and Riddle Syndrome

E. Alkhunaizi¹, B. Schrewe², G. Stewart³, N. Braverman¹

¹Department of Medical Genetics and Pediatrics, McGill University Health Centre, Montreal, QC, Canada

²Department of Pediatrics, University of British Columbia, Vancouver, BC, Canada

³School of Cancer Sciences, University of Birmingham, Birmingham, UK

Diamond–Blackfan anemia (DBA) (MIM#105650) is a rare dominant congenital blood disorder that often manifests in the first two years of life with normochromic, macrocytic anemia and selective deficiency of red cell precursors due to absent erythroid progenitors in the bone marrow. Researchers have identified at least twelve genes that encode ribosomal proteins and cause DBA. One of these genes is RPL35a gene (MIM# 180468), which plays an essential role in ribosome biogenesis, selective cell proliferation and apoptosis. In this report, we present 4-year-old French- Canadian girl, with a de novo 12.8 Mb terminal deletion of 3q27.2 that contains the RPL35a gene. Her clinical and laboratory features are consistent with Diamond-Blackfan anemia, but also include immunodeficiency, cardiac malformation, Crohn's disease, growth and developmental delays. Given the diagnostic uncertainty of her case, we investigated the current nosology of DBA and explored other possible contributing factors to her phenotype by comparing to other patients in the literature with overlapping genomic deletions.

We performed a literature review and identified 38 cases in total with overlapping genomic deletions. Only five cases were overlapping with RPL35a region and two of the five cases demonstrated to have decrease expression of RPL35a and showed a phenotypic picture consistent with DBA, but not the rest of our patient's phenotype. We noted that recessive mutations in RNF168 gene (MIM# 611943) on 3q29 associated with RIDDLE syndrome (radiosensitivity, immunodeficiency, dysmorphic features, and learning difficulties). Given her overlapping genomic deletion with RNF168 gene and immunological profile which required immunoglobulin replacement therapy, we sequenced the remaining RNF168 gene in our patient. We also examined her fibroblast cultures for sensitivity to ionizing radiation. These results were normal, indicating that the underlying immunodeficiency is not likely to result from a RNF168 deficiency.

We conclude that RPL35a haploinsufficiency is the cause of DBA in this child. The etiology for the immunodeficiency remains unsolved. Finally, her complex phenotype is unique from those patients with overlapping deletions described in the literature.

Novel Homozygous Contiguous Gene Deletion Due to Chromosome 4 Uniparental Disomy Results in Multisystem Dysfunction

Anne H. O'Donnell Luria, Pankaj B. Agrawal, Anja J. Gerrits, Andrew L. Frelinger 3rd, Alan D. Michelson, Yiping Shen, Wen-Hann Tan, Tim W. Yu, Sharon E. Smith

Boston Children's Hospital, Harvard Medical School, Boston, Massachusetts

We report on a 22-year-old female with multisystem dysfunction including severe gastrointestinal (GI) dysmotility that manifested as meconium ileus, infantile achalasia and severe constipation requiring colectomy. She had blood pressure lability with periods of hyper- and hypotension along with a history of four acute neurologic events including multiple strokes. She also developed a central-line associated upper extremity deep venous thrombosis and unilateral lower extremity lymphedema. Additionally, she had global developmental delay with little speech and severe intellectual disability. Her behavior was notable for severe anxiety, impulsivity, emotional lability and pica. On exam she was microcephalic (-3.1 SD) and had deeply set eyes with infraorbital creases and a broad jaw.

Initial negative workup included sweat test, rectal biopsy for Hirschsprung disease, very long chain fatty acids analysis, UBE3A sequencing, FISH for 15q11.2, 17p11.2, and 22q11. She was lost to genetics follow-up until she was re-referred at age 20. Chromosomal microarray revealed a ∼450kb homozygous deletion on chromosome 4 encompassing 5 genes, GUCY1A3 and GUCY1B3 (which comprise the α1 and β1 subunits of the soluble guanylate cyclase (sGC) receptor), along with ASIC5, TDO2, and CTSO. The mother was found to have an identical heterozygous deletion. By examining the SNP probes on the microarray, large blocks of loss of homozygosity were noted, consistent with maternal chromosome 4 uniparental isodisomy.

The significant dysfunction of this patient's GI, hematologic, lymphatic, vascular, and central nervous systems can be attributed to the nitric oxide (NO) signaling defect. The sGC receptor, the only known receptor for NO, is a heterodimer of an α1 or α2 subunit complexed with the β1 subunit to form the α1β1 or α2β1 sGC receptor. While patients with homozygous null mutations in the α1 subunit have moyamoya disease (a cerebral vasculopathy not present in this patient) and achalasia (Hervé D et al., Am J Hum Genet, 2014; 94 (3):385–94), the consequences and viability of a complete loss of sGC receptor in humans has remained unknown until now. The α2 subunit is expressed in a number of tissues including the colon, which may account for the more extensive phenotype seen in this patient. β1 sGC receptor knockout mice also have severe colonic dysmotility and small stature ultimately resulting in early demise from intestinal rupture (Friebe A et al., Proc Natl Acad Sci U S A, 2007; 104(18):7699–704). The other three genes in the homozygous deleted region are currently not known to be associated with human disease

This is the only report of loss of function of both the α and β subunits of the nitric oxide receptor in a human. Much can be learned from this 22-year-old woman who presents with a unique clinical phenotype as the first case reported of systemic absence of nitric oxide signaling through this receptor. Dysfunction of the nitric oxide signaling pathway should be considered in patients with GI dysmotility, cerebrovascular accidents, moyamoya, and/or intellectual disability. These patients may be at an increased risk for hypertension, stroke and myocardial infarction, all disorders for which the risks are modifiable with medical management.

Two Patients With Novel ACTB Mutations: Atypical Baraitser–Winter Syndrome or Expanded Phenotype of Actinopathies?

Michael J. Lyons, MD, Sara Cathey, MD, Frank Bartel, PhD, Jennifer A. Lee, PhD, Michael Friez, PhD

Greenwood Genetic Center, 106 Gregor Mendel Circle, Greenwood, South Carolina

ACTB mutations have previously been reported to cause Baraitser-Winter syndrome, which is characterized by intellectual disability, neuronal migration defect, coloboma, high- arched eyebrows, ptosis, and hypertelorism. Recently, the designation Baraitser-Winter cerebrofrontofacial syndrome (BWCFF) has been proposed to encompass the range of clinical features reported in patients with gain-of-function mutations in the cytoplasmic actin-encoding genes ACTB or ACTG1.

We describe two unrelated patients with novel ACTB mutations identified by whole exome sequencing (WES). Patient 1 is a 2-year-old female with global developmental delays, hypotonia, dysgenesis of the corpus callosum, and persistent supraventricular tachycardia. Patient 2 is a 4-year-old female with global developmental delays, trivial coarctation of the aorta, and history of thrombocytopenia. Both patients have short stature, microcephaly, epicanthal folds, thin eyebrows, short palpebral fissures, strabismus, long eyelashes, coarse facies, full cheeks, depressed nasal root, long philtrum, and overfolded helices.

WES in patient 1 revealed a novel p.Val152Leu ACTB mutation, predicted to be deleterious by in silico analysis. After discovery of the ACTB mutation in patient 1, a database search of patients with completed WES at the Greenwood Genetic Center was initiated. No patients were identified to have the recurrent hotspot mutations in amino acid 196 of ACTB. A novel p.Ser368Leufs*13 mutation in the ACTB gene was identified in patient 2. This mutation results in a shift in the reading frame and is expected to lead to a prematurely truncated protein. Parental studies for both patients confirmed that the mutations are de novo.

We believe the novel ACTB mutations identified in our patients are the explanation for their clinical features. Significant clinical variability has been reported in BWCFF patients. Our patients have some overlapping features with published patients, but they do not have a number of characteristic findings such as neuronal migration defect, coloboma, hypertelorism, high-arched eyebrows, long palpebral fissures, or trigonocephaly. Clinical differences in our patients may be related to their novel ACTB mutations. Johnston et al reported a patient with ACTB mutation p.Glu117Lys as possible having atypical Baraitser-Winter syndrome or a distinct condition. Our patients provide further evidence of an expanded phenotypic spectrum associated with ACTB mutations, provide the first report of a patient with a frameshift mutation in ACTB, and suggest that WES will lead to diagnosis of an actinopathy in additional patients with global developmental delay and dysmorphic features.

Higher Molecular Diagnostic Yield in Morphologically Complex Forms of Autism Spectrum Disorder

B.A. Fernandez MD¹, K. Tammimies PhD³, C. Vardy MD², V. Crosbie MD², S. Luscombe MD², T. Doyle MD², K. Whitten RN², D.J. Stavropoulos CR⁴, M. Carter MD⁴, R. Weksberg⁴, J. Howe MSc³, P.S. Szatmari MD⁶, C.R. Marshall PhD³, S.W. Scherer PhD^3,5

¹Discipline of Genetics, Memorial University of Newfoundland, St. John's NL Canada

²Discipline of Pediatrics, Memorial University of Newfoundland, St. John's NL Canada

³The Centre for Applied Genomics, Toronto, ON, Canada

⁴Hospital for Sick Children, Toronto, ON, Canada

⁵Department of Molecular Genetics, U of T

⁶Centre for Mental Health & Addiction⁶, U of T

The phenotypic complexity of autism spectrum disorders (ASD) remains a challenge for genetic research efforts, and phenotypic stratification has been proposed as a method of categorizing patients into more genetically informative subtypes. Miles et al 2005 used clinical morphology to classify an ASD cohort into three groups: essential, equivocal and complex. The complex group had the highest number of minor physical abnormalities (MPAs) and comprised 20% of their cohort. We report the diagnostic yield for chromosomal microarray (CMA) and whole-exome sequencing (WES) in a population based cohort of ASD children stratified by clinical phenotype.

Methods: The study group consists of 258 unrelated ASD children from Newfoundland and Labrador, Canada who were consecutively referred from a developmental pediatric clinic. Each underwent a dysmorphology examination, with documentation of MPAs as per Miles et al 2005. Major congenital anomalies were also recorded and a morphology score was assigned. All patients had CMA testing, and WES was performed for 95 proband-parent trios.

Results: Based on morphology scores, 168/258 (65.1%) children were classified as essential, 37/258 (14.3%) as equivocal and 53/258 (20.5%) as complex. Of 258 cases, 24 patients (9.3% CI, 6.1%-13.5%) received a molecular diagnosis from CMA and 8/95 patients (8.4%, CI 3.7%-15.9%) from WES. The diagnostic rates for the tests differed across the three morphologic groups with the highest rates in the complex group (for CMA 24.5%, P = 0.00014; for WES 16.7%). The combined diagnostic yield of CMA and WES was 15.8% and was statistically different across the three groups, with the highest rate of 37.5% in children classified as having morphologically complex ASD (9/24 patients, CI 18.8–59.4%). By WES, 96 de novo mutations were identified in 55/95 cases (0–5 de novo mutations per child). The number of de novo changes affecting the coding sequence positively correlated with the morphology score based on MPA and MCA (r = 0.2, P = 0.029), and females in the complex group had the highest rate. Three of 96 de novo variants were classified as clinically significant. These included two loss-of-function mutations in the ASD/ID genes ASH1L and WAC, and a de novo missense mutation in SCN2A affecting a conserved glycine residue in a child whose EEG showed multifocal epileptiform discharges. Two/95 children had significant findings on both CMA and WES. For example, one male received a diagnosis of Klinefelter syndrome from CMA, and an X-linked mutation in EDA was identified on WES. These syndromes have been associated with increased risks for ASD and behavioral disturbances respectively.

Conclusions: The diagnostic yield of CMA and WES is almost equal when assessing a heterogeneous ASD cohort, and until whole genome sequencing becomes the primary genetic test for ASD individuals, there is a role for both tests. Following CMA and WES, we estimate that over 35% of ASD children with significant dysmorphology findings and/or birth defects will receive a molecular diagnosis that at least partially explains their ASD susceptibility.

KIF5C Mutations Cause Severe Intellectual Disability With or Without Cortical Brain Malformations

Laurie Sadler, MD, Melissa Samons, MS, CGC

Division of Genetics, Department of Pediatrics, Women and Children's Hospital of Buffalo, School of Medicine and Biomedical Sciences, State University of New York at Buffalo, Buffalo, New York

Whole exome sequencing (WES) has elucidated a number of newly recognized intellectual disability (ID) genes. We recently performed WES on an 18-year-old female who we have followed since early childhood. The patient initially presented at age 2 years with a history of postnatal onset microcephaly and global developmental delays. Prenatal, perinatal and family histories were negative. The patient's facial features, unsteady gait and frequent smiling were suggestive of a diagnosis of Angelman syndrome (AS); initial test results were negative. Seizures, which were likely unrecognized during early childhood, became recalcitrant to treatment at age 8 years. Medical history was also remarkable for scoliosis, sleep disturbance and severe intellectual disability with absent speech. Results of extensive diagnostic testing, including methylation testing for AS, UBE3A sequencing, MECP2 and CDKL5 sequencing and deletion/duplication testing, BAC and SNP arrays, and testing for congenital disorders of glycosylation all yielded normal results. WES disclosed a de novo pathogenic variant in the KIF5C gene.

There are three previous reports of KIF5C mutations in patients with severe ID. Poirier et al., (2013) reported four brothers with microcephaly, malformations of cortical development and severe ID, all of whom had the same missense mutation of KIF5C. Their clinically normal mother had presumed gonadal mosaicism for the same variant. Jamuar et al., (2014) analyzed 158 patients with malformations of cortical development, and found a single patient with a presumed pathogenic variant in KIF5C. Willemson et al., (2013) reported a patient with the postnatal onset of microcephaly, seizures, and severe ID with absent speech. This patient, who had mild structural brain abnormalities, had the identical KIF5C missense mutation as was detected in our patient. In addition, patients with 2q23.1 microduplication, a region which includes the KIF5C gene, have a similar phenotype characterized by microcephaly, seizures and ataxia (Chung et al., 2011).

The KIF genes encode kinesin proteins which are ATP-dependent molecular motors involved in intracellular transport along microtubules (Sinclair et al., 2010). These genes are involved in neuronal function, development, survival and plasticity, similar to the role of UBE3A. Our patient is the first reported patient with a KIF5C mutation in whom gross brain morphology is normal. Interestingly, a KIF5C knock-out mouse model resulted in mice with smaller brain size with no gross abnormalities of brain structure (Kanai et al., 2000). KIF5C mutations are a more recently recognized cause of severe ID in patients with microcephaly and seizures. The normal brain morphology in the patient of this report suggests that KIF5C mutations do not cause abnormal brain morphogenesis in all cases. Similar to other known single gene causes of AS-like phenotypes (TCF4, ZEB2, MECP2), we suggest that mutations of KIF genes should be considered in the differential diagnostic approach to such patients.

Developmental Delay With Spasticity and Areflexia due to Tyrosyl-tRNA Synthetase (YARS) Mutations: The Expanding Phenotype of Aminoacyl-tRNA Synthetase Mutations in Human Disease

Małgorzata J.M. Nowaczyk,¹ Susan Zeesman,² Mark Tarnopolsky,² Dennis E. Bulman,³ Chandree L. Beaulieu,³ FORGE Canada Consortium,³ Jacek Majewski,⁴ Kym M. Boycott³

¹Depatrment of Pathology and Molecular Medicine, McMaster University, Hamilton, ON, Canada

²Department of Paediatrics, McMaster University, Hamilton, ON, Canada

³Children's Hospital of Eastern Ontario Research Institute, University of Ottawa, Ottawa, ON, Canada

⁴Department of Human Genetics, McGill University, Montréal, QC, Canada

Aminoacyl-tRNA synthetases (ARSs) are enzymes required for the first step of protein translation to occur. Each ARS links a specific amino acid to its corresponding transfer ribonucleic acid (tRNA) component within the cytoplasm, mitochondria, or both. Mutations in different ARSs have been linked to a growing number of human diseases. The cytosolic tyrosyl-tRNA synthetase (YARS) links the amino acid tyrosine to its cognate tRNA in the endoplasmic reticulum. To date, heterozygous mutations in YARS have been reported in the dominant intermediate type C Charcot-Marie-Tooth disease (OMIM 603623), a childhood to adult-onset condition characterized by isolated neuropathy.

Here we report two siblings with compound heterozygous mutations within the YARS gene ((c.638C>T,p.Pro213Leu and c.1573G>A, p.Gly525Arg; NM_003680) who presented with multisystem involvement characterized by cherubic facies with mildly dysmorphic features, hypotonia, areflexia, failure to thrive, developmental delay, liver dysfunction, lung parenchyma cysts, and MRI brain evidence of abnormal subcortical white matter. Both mutations were inherited from asymptomatic carrier parents implicating autosomal recessive inheritance.

This is the first report of an apparently recessive, infancy-onset, multisystem disease reported with YARS mutations, a paradigm shift for the pathophysiology of YARS mutations. Our observations expand the spectrum of diseases associated with dysregulation of YARS, suggest a new inheritance pattern, and highlight ARS as a group of enzymes that are increasingly being implicated in human disease.

A Possible Case of X-Linked Epileptic Encephalopathy

A. Sommer^1,2, M.P. Pietryga² and M. Kekis²

¹Nationwide Children's Hospital, The Ohio State University, College of Medicine, Columbus, Ohio

²Department of Pediatrics, The Ohio State University, College of Medicine, Columbus, Ohio

Epileptic encephalopathies remain a significant diagnostic challenge for physicians and management is difficult. The cause of these encephalopathies is often unknown, although whole exome sequencing (WES) has contributed to the identification of putative causes and probable etiologies. However, in the absence of a concrete disease mechanism, concise and accurate genetic counseling and effective treatment plans are difficult to implement. We describe a case of epileptic encephalopathy where WES implicated the CLCN4 gene on chromosome X as a possible cause of disease.

Case Report: The patient was seen in Genetics at 25 months of age because of intractable seizures, cortical atrophy and growth failure. He was born to a 30 year old G3P2Ab 1 mother after a normal pregnancy and vaginal delivery at term with a birth weight of 4.02 kg and birth length of 53 cm. He was asymptomatic until the age of 3 months when he experienced the onset of seizures. Evaluation at another institution was inconclusive. At his initial evaluation at 25 months of age in our clinic, he had a weight of 8.45 kg (0%ile), height of 82 cm (6%ile) and a head circumference of 44.7 cm (0%ile). The examination did not reveal any dysmorphic features except microcephaly, growth retardation and developmental delay. He is followed by Neurology and has been on a ketogenic diet. An EEG at 2 years of age found intermittent multifocal spikes over left and right central and frontal regions which become more enhanced during sleep, but without clinical evidence, as well as mild generalized slowing consistent with the diagnosis of epilepsy and diffuse encephalopathy. A brain MRI showed marked periventricular white matter loss with compensatory dilation of lateral and third ventricles. A follow-up MRI, several months later, revealed slight progress of gray and white matter volume loss. There were no focal parenchymal signal changes noted. MR spectroscopy at his last MRI did show reduction in NAA in both gray and white matter. A WES study reported one pathogenic variant and one likely pathogenic novel variant related to the patient's phenotype. The variant most consistent with epileptic encephalopathy was the novel likely pathogenic hemizygous missense change c.1601C>T (p.S534L), a highly conserved position, in a highly conserved domain in the CLCN4 gene. The patient's mother was negative for this finding, suggesting a de novo etiology, although the possibility of gonadal mosaicism cannot be ruled out. CLCN4, located at Xp22.3, encodes a voltage gated chloride channel, which had been previously implicated in only a single case of epileptic encephalopathy in a 14 month old boy with a variant of unknown significance (G544R), who additionally presented with microcephaly, psychomotor developmental delay, hypotonia and dystonia. Neither of his parents were found to carry the G544R variant. The lack of certainty regarding etiology represents a counseling challenge, particularly for recurrence risk and prognosis. Our patient is a second report of a de novo, likely pathogenic mutation in the CLCN4 gene, a gene not previously implicated in epilepsy.

This case further contributes to the utility of WES in the diagnostic evaluation of the early onset/infantile epileptic encephalopathies. Further studies in these patients may identify whether these newly implicated genes and variants play a pathogenic role in epileptic encephalopathy and affect medical management.

Veeramah KR et al: Epilepsia, 2013 Jul: 54 (7):1270-1281

Hu H et al: Molecular Psychiatry, 2015 Feb 3. [Epub ahead of print]

Further Evidence For a Distinct Phenotype: Novel USP9X Variant in a Patient With Intellectual Disability, Autism, Relative Macrocephaly, Broad Toes and Seizures

N. Dikow¹, K. Hinderhofer¹, S. Karch², M. Granzow¹, N. Paramasivam³, L. Kaufmann¹, C. Fischer¹, C. Evers¹, U. Moog¹

¹Institute of Human Genetics, Heidelberg University, Heidelberg, Germany

²Section of Neuropediatrics, Center for Child and Adolescent Medicine, Heidelberg, Germany

³Division of Theoretical Bioinformatics, German Cancer Research Center (DKFZ), Heidelberg, Germany

Variants in the X-linked USP9X gene have been associated with intellectual disability (ID), hypotonia, behavioral problems as well as digital anomalies and craniofacial dysmorphism in male patients. USP9X encodes a substrate-specific deubiquitylating enzyme. USP9X is involved in neuronal cell migration and axonal growth. Five patients from three families with ID caused by variants in USP9X have been published so far. Two additional patients recruited for epileptic encephalopathy have been reported to carry USP9X variants. We report on a boy with a novel USP9X variant. The available clinical features of all eight patients are reviewed.

We identified a novel USP9X variant c.6360A>G; p.(I2120M) by whole exome sequencing in a boy evaluated for ID. The variant was confirmed by Sanger sequencing in the index and his unaffected mother. USP9X is highly conserved throughout species andthe variant affects a highly conserved amino acid in the C-terminal region of the protein. In silico prediction programs (SIFT, MutationTaster, PPH2) suggested that the variant was (probably) disease causing. No other possibly relevant variants were found.

The patient's phenotype at the age of 9 ¾ years consisted of ID, hypotonia from his first year, autism spectrum disorder, seizures (onset 7 years) responding well to valproate, relative macrocephaly, broad toes and a reduction defect of the 3rd toe left.

On five of the seven patients from five families published previously with USP9X alterations, restricted clinical data are available. In one patient, an additional deletion 6q25.3 affecting ARID1B was present. Available data report a phenotype in five of them consisting of ID, hypotonia and behavioral problems (4/5). Further clinical data were available in 2 of them reporting relative macrocephaly (1/2) and digital anomalies (2/2); all features are corroborated by the present patient. Two male patients with epileptic encephalopathy and restricted data on the phenotype carried USP9X missense variants very close to the ubiquitin- like domain and in the ubiquitincarboxyl-terminal hydrolase domain, respectively. In contrast, 3 variants reported in patients with ID were suggested not to affect the catalytic activity of USP9X. A possible genotype-phenotype correlation will be discussed.

Based on the present findings and the available literature, there is evidence that USP9X variants are associated with ID, hypotonia, behavioral problems, relative macrocephaly and digital anomalies and possibly predispose to seizures.

Lessons From the First Familial Cases With WDR45 Mutations

Yuri A. Zarate¹, Julie R. Jones², Melanie A. Jones², Francisca Millan³, Jane Juusola³, Annette Vertino-Bell³, Michael C. Kruer⁴

¹Section of Genetics and Metabolism, University of Arkansas for Medical Sciences, Little Rock, Arkansas

²Greenwood Genetic Center, Greenwood, South Carolina

³GeneDx, Gaithersburg, Maryland

⁴Divisions of Pediatric Neurology & Genetics, Sanford Children's Specialty Clinic, Sanford Children's Hospital, Sioux Falls, South Dakota

Neurodegeneration with brain iron accumulation (NBIA) encompasses a heterogeneous group of inherited progressive neurological diseases characterized by cognitive impairment and prominent dystonia and/or parkinsonism with evidence of abnormal iron deposition in the basal ganglia of the brain by magnetic resonance imaging (MRI). The overall prevalence for all forms of NBIA is thought to be less than 1/1,000,000. Beta-propeller protein-associated neurodegenera­tion (BPAN) accounts for approximately 7% of all cases of NBIA. BPAN has fairly distinctive clinical and brain imaging findings and is caused by heterozygous loss-of function mutations in the WDR45 gene located in Xp11.23. All reported cases to date have been sporadic and with a clear female predominance supporting an X- linked dominant pattern of inheritance with lethality for germline mutations in hemizygous males. We describe the first familial cases of BPAN. The male proband was 20 years-old at the time of ascertainment and the first child born to his non-consanguineous Hispanic parents. During childhood, he received the diagnosis of global developmental delay and spastic quadriplegic cerebral palsy. Other medical problems included epilepsy, scoliosis, severe obstructive sleep apnea, and bilateral sensorineural hearing loss. The patient was never able to walk independently and currently while he is able to stand with support briefly, he is wheelchair dependent. There was no development of purposeful speech, and he is completely dependent for his daily routines. No history of regression has been documented. At 14 years of age a brain MRI documented severe diffuse brain atrophy and moderate cerebellar atrophy with thinning of the corpus callosum without evidence of brain iron deposition. Meanwhile, his 14 year-old sister was evaluated for her long history of static encephalopathy. More functional than her brother, she speaks in short sentences, ambulates independently and can perform daily self-care routines. She also has mild bilateral sensorineural hearing loss but never had seizures. A brain MRI at age 7 years showed only borderline thinning of the corpus callosum. The siblings' mother is an otherwise healthy 46 year-old female of Mexican descent. After several combined cytogenetic, biochemical and molecular evaluations, no diagnostic abnormalities were found. Whole Exome sequencing including both siblings and their parents was then performed. A pathogenic in-frame deletion in the WDR45 gene (c.161_163delTGG, p.Val54del) was detected in the hemizygous, heterozygous, and mosaic states in the brother, sister, and mother, respectively. The X chromosome inactivation studies in the affected sister revealed a highly skewed pattern of inactivation of 92:8. This first report of familial cases of BPAN offers valuable lessons. First, while all previously reported cases (39/39) had been sporadic suggesting de novo mutations early in development as the main mechanism, inherited mutations are possible, albeit rare. Second, hemizygous germline mutations in males are not universally lethal but often result in a more severe phenotype. Third, for siblings with germline mutations, males should be more significantly affected than females although the degree of X-chromosome inactivation skewing in the latter affects the severity of the phenotype. Lastly, knowing that gonadal and germline mosaicism are possible, parental testing for mutations in WDR45 should be considered.

A Pathogenic Variant in BMPR1A Results In Craniofacial Dysmorphism, Robin Sequence, and Juvenile Polyposis Syndrome

Wen-Hann Tan¹, Anne H. O'Donnell Luria¹, Stephen Shusterman², Victor L. Fox³, John B. Mulliken⁴

¹Division of Genetics and Genomics, Boston Children's Hospital, Boston, Massachusetts

²Department of Dentistry, Boston Children's Hospital, Boston, Massachusetts

³Division of Gastroenterology, Boston Children's Hospital, Boston, Massachusetts

⁴Department of Plastic and Oral Surgery, Boston Children's Hospital, Boston, Massachusetts

Loss-of-function mutations in BMPR1A, which encodes the bone morphogenetic protein receptor type 1A, are found in about 20–25% of individuals with juvenile polyposis syndrome (JPS). JPS is characterized by the presence of multiple juvenile polyps throughout the gastrointestinal tract. Facial dysmorphism and cleft palate have not been described in patients with JPS, although cleft palate has been reported in a child with a chromosome 10q22.3-q23.2 microdeletion that includes BMPR1A along with other genes [Petrova E et al. Mol Syndromol. 2014; 5(1):19–24]. Murine studies have shown that Bmpr1a is expressed in the primary and secondary palate in development, and knockout of Bmpr1a in palatal mesenchyme resulted in clefting of the anterior palate [Baek JA et al. Dev Biol. 2011; 350(2):520–31]. A more recent study revealed that the mandibular condyle was smaller in cartilage-specific Bmpr1a knockout mice, putatively due to down-regulation of Sox9, which controls the differentiation of chondrocytes [Jing J et al. Connect Tissue Res. 2014;55 Suppl 1:73–8]. Herein we present a boy with facial dysmorphism, Robin sequence, and juvenile polyposis syndrome due to a heterozygous mutation in BMPR1A.

A 13-year-old boy first presented with gastrointestinal bleeding at the age of 6 years. Colonoscopy revealed multiple juvenile polyps throughout his colon and rectum. He was born with a cleft palate at 31 weeks gestation. He had respiratory and feeding difficulties as a neonate, requiring treatment with continuous positive airway pressure (CPAP) and a feeding tube for about 5 weeks. He had minor gross motor delay in infancy that he has since overcome, and his cognitive development has always been normal. Physical examination at age 12 was significant for macrocephaly, a long and narrow face, long ears, retrognathia, and severe dental crowding with class II malocclusion. He had a complete cleft of the secondary palate that was repaired in late infancy. He had hyperconvex finger and toe nails; his limbs were otherwise normal. He had epispadias that has since been repaired. Brain MRI and echocardiogram were both normal.

Sequencing of SMAD4 and BMPR1A revealed a de novo heterozygous nonsense mutation in BMPR1A (c.1480C>T (p.R494X)). This mutation has previously been reported in at least two individuals who both had juvenile and hamartomatous intestinal polyps. SNP-based chromosomal microarray identified a maternally-inherited 128–217 kb gain on chromosome 5p15.33 that does not encompass any protein-coding genes, although there are two microRNAs of unknown function. We hypothesize that heterozygous loss-of-function mutations in BMPR1A can be associated with syndromic Robin sequence in addition to juvenile polyposis.

SARC006: Phase II Trial of Chemotherapy in Sporadic and Neurofibromatosis Type 1 (NF1)- Associated High-Grade Malignant Peripheral Nerve Sheath Tumors (MPNSTs)

David Viskochil, Brigitte Widemann, Denise Reinke, Lee Helman, Joseph Ludwig, Scott Schuetze, Arthur Staddon, Mohammed Milhem, Daniel Rushing, Christopher Moertel, Stewart Goldman, Michael Livingston, Lars Wagner, Eve Rodler, Eva Dombi, Arie Perry, Christina Annunziata, Lauren Long, Seth Steinberg, Larry Baker

Background: MPNSTs are sarcomas that tend to arise in benign plexiform neurofibromas, and they occur in 10% of the NF1 population. Retrospective, pooled analyses have reported worse chemotherapy response for NF1 (9–18%) compared to sporadic MPNSTs (39–55%)

Methods: MPNSTs are not generally responsive to traditional chemotherapeutic protocols designed for sarcomas; however a prospective trial has not been conducted. We prospectively evaluated the objective response (OR) rate [complete responses (CR) and partial responses (PR) using WHO criteria] of children and adults with high-grade, unresectable or metastatic chemotherapy naïve NF1-associated versus sporadic MPNST after 2 cycles of ifosfamide and doxorubicin (IA) followed by 2 cycles of ifosfamide and etoposide (IE), as primary endpoint, followed by surgery and/or radiation for local control and up to 2 more cycles of IA and IE each. Pathology was centrally confirmed. A Simon optimal two-stage design was used with a target response rate of 40% and 17 patients per stratum in the 1st stage. With 4+/17 ORs, enrollment would be expanded to 37 patients per stratum, and 11+/37 ORs would be consistent with a 40% OR rate.

Results:

Both initial stages met criteria for enrollment expansion with 4/17 PRs in NF1 and 4/12 PRs in sporadic MPNSTs, but only one additional PR was observed in the expanded NF1 stratum. Of the 9 PRs, 5 were first observed after two cycles IA, and additional 4 after two cycles IE. Enrollment was slow and the trial was closed before full accrual.

Conclusions: While the primary trial objective was not reached due to slow enrollment, with only 5/29 ORs in the NF1 stratum the desired OR rate of 11+/37 would have unlikely been met, even if accrual had been completed. We observed a lower OR rate in NF1 compared to sporadic MPNSTs, which is similar to retrospective literature reports. However, our trial was not powered to detect a difference in response rates between the two strata. In addition to PRs after both IA and IE, disease stabilization was achieved in most patients. Novel strategies including the evaluation of targeted agents in combination with chemotherapy will be considered in future trials.

Mosaic KRAS G12D and Rhabdomyosarcoma

Nouf Hijazi, Wingfield Rehmus, Julie Prendiville, Anna Lee, Avash Singh, Tracy Pressey, Glenda Hendson, Stephen Yip, Anna Lehman

Department of Medical Genetics, University of British Columbia, Vancouver, Canada

Somatic mutations in the proto- oncogene HRAS, and occasionally KRAS, have recently been implicated as the major cause of nevus sebaceus, both when occurring as an isolated lesion and when presenting with multiple developmental anomalies. Nevus sebaceous occasionally presents as part of a complex developmental syndrome, including, for example, phakomatosis pigmentokeratotica, caused by postzygotic-activating HRAS mutation, encephalocraniocutaneous lipomatosis (gene unknown) and Schimmelpenning syndrome, otherwise known as the linear sebaceus nevus syndrome, which is caused by post-zygotic mutations in HRAS and KRAS.

We report the occurrence of nevus sebaceus in a premature neonate with lymphatic hydrops resulting in intractable chylothorax. Histopathology of a biopsy from the area of subcutaneous edema revealed lymphangiectasia while biopsy from the shoulder was consistent with a nevus sebaceus. Ten ng of DNA extracted from nevus was analyzed on the Ion Torrent PGM sequencing platform using the Ion AmpliSeq™ Cancer Hotspot Panel v2 (Life Technologies), which included all previously reported mutations associated with nevus sebaceus. A heterozygous mutation was identified in exon 12 of KRAS (c.35G>A; p.G12D) in 694 of 2000 reads (34%). Despite intensive therapy, he developed persistent chylothorax, pulmonary hypertension, coagulopathy, hypoalbuminemia, anemia, and acidosis. He died at three weeks of age. An autopsy further demonstrated bladder rhabdomyosarcoma.

This specific KRAS mutation, G12D, has been reported twice before in infants; both times proven through multi-tissue testing to be somatic mosaic. In the first case, it was associated with an epidermal nevus that followed the lines of Blaschko, polycystic kidneys, unexplained growth restriction, and rhabdomyosarcoma diagnosed at age 6 months. The second infant with a proven somatic mosaic KRAS G12D mutation was diagnosed with Schimmelpenning syndrome. The infant had an unaffected monozygous twin. He had extensive nevus sebaceus, ocular anomalies, patent ductus arteriosus, cleft palate, and cerebral defects.

KRAS G12D is also a well- described proto-oncogenic mutation, especially in colorectal and pancreatic adenocarcinomas. Mice with germline heterozygous KRAS G12D die in the embryonic state, which suggests longer survival could only occur if the mutation occurred post-zygotically.

Chinese Patients With Rasopathies: Detection of 29 Known and Novel Variants Using Next- Generation Sequencing (NGS)

Gordon K.C. Leung¹, Ivan F.M. Lo², H.M. Luk², Vincent H.M. Tang³, W.W. Gao³, W.L. Wong¹, Yoyo W.Y. Chu¹, Steven T.S. Lam², W.L. Yang¹, Alvin C.H. Ma⁴, D.Y. Jin³, Kelvin Y.K. Chan⁵*, Brian H.Y. Chung^1,5*

¹Department of Paediatrics and Adolescent Medicine^,, The University of Hong Kong, HKSAR

²Clinical Genetic Service, Department of Health, HKSAR

³Department of Biochemistry, The University of Hong Kong, HKSAR

⁴Department of Medicine, The University of Hong Kong, HKSAR

⁵Prenatal Diagnostic and Counseling Department, Tsan Yuk Hospital, HKSAR

Rasopathies are a group of developmental syndromes caused by gene mutations in the RAS/MAPK signaling pathway. Characterized by clinical overlap and genetic heterogeneity, the molecular diagnoses are often challenging. In Hong Kong, PTPN11 sequencing is the only test available since 2006 and mutations are detected in ∼50% of patients (Chan et al., 2006. HK.J.Paediatr. 11:290–296). In 2013, we developed a NGS-based research diagnostic panel for rasopathies, targeting the coding exons of 13 disease-causing genes known at the time. The panel was offered to all patients with the clinical diagnosis of rasopathies in Hong Kong (including new patients and those previously tested negative for PTPN11). We report here the mutation spectrum of 29 Chinese patients with rasopathies using this NGS panel (PTPN11, n=6; SOS1, n=6; RAF1, n=2; KRAS, n=3; BRAF, n=2; SHOC2, n=4; MAP2K1, n=3 and RIT1, n=3). The clinical features of these patients are analysed and will be presented in greater details in the meeting. Patients with known pathogenic mutations are studied for genotype- phenotype correlation while novel mutations are assessed functionally using Elk-1 reporter system and zebrafish modelling. In conclusion, the NGS panel has achieved an additional diagnostic yield of 12% in our patient cohort with raospathies ascertained clinically. This panel will be translated into clinical use soon, with the inclusion of two newly discovered causative genes (A2ML1, BRAF, CBL, HRAS, KRAS, MAP2K1, MAP2K2, NRAS, PTPN11, RAF1, SHOC2, SOS1, SPRED1, RIT1 and RRAS).

Prenatal Onset of JMM Leukemia Associated With CBL Gene Mutation in a Patient With Noonan-Like Syndrome

Resham Ejaz MD¹, Christian R. Marshall PhD^2,3, Riyana Babul-Hirji MSc¹, James D. Stavropoulos PhD³, A. James MD⁴, David Chitayat MD^1,5

¹Division of Clinical & Metabolic Genetics, The Hospital for Sick Children, Department of Pediatrics, Ontario, Canada

²The Centre for Applied Genomics, Ontario, Canada

³Department of Pediatric Laboratory Medicine, Ontario, Canada

⁴Division of Pediatric Hematology Oncology; Mount Sinai Hospital, Department of Obstetrics and Gynecology, Ontario, Canada

⁵The Prenatal Diagnosis and Medical Genetics Program, University of Toronto, Toronto, Ontario, Canada

Germline heterozygous mutations in the CBL gene has been reported with a spectrum of features ranging from Noonan-syndrome like condition including developmental delay, reduced growth, facial dysmorphism, and café-au-lait spots to subtle findings. Germline CBL mutations also confer predisposition to juvenile myelomonocytic leukemia (JMML) during childhood which often spontaneously regresses with the only successful treatment being bone marrow transplantation (BMT). We report the first patient with prenatal onset of JMML associated with CBL mutation detected through whole genome sequencing (WGS).

The patient was born at 34.7 weeks gestation to a 31-year-old primigravida mother. The pregnancy was complicated with an increased middle cerebral artery Doppler studies, hepatosplenomegaly, and small pleural effusion detected at 28 weeks gestation. Cordocentesis revealed normal hemoglobin level and thrombocytopenia (29 10⁹/L), requiring two in-utero platelet transfusions. Delivery was uncomplicated, his birth weight and length were at the 50-97^th percentile and the head circumference was at the 50^th percentile. His features were mildly dysmorphic with hypoplasia of the supraorbital ridges, low-set and posteriorly rotated ears, full lips, redundant nuchal skin, low posterior hairline, cardiac murmur, hepatosplenomegaly and cryptorchidism. Blood count at birth showed leukocytosis with blasts, thrombocytopenia, and vacuolated monocytes. Bone marrow biopsy at 1 month confirmed JMML. Echocardiography showed dysplastic pulmonary valve stenosis.

A RASopathy panel with 11 genes including the CBL gene showed no detectable mutation. Suspicion for Noonan-related syndrome remained high and thus WGS was done and revealed a de novo CBL mutation [c.1096-11_1109del]. No compatible stem cell was identified and thus BMT was not pursued and 6-mercaptopurine for cytoreduction was started.

To date, prenatal abnormalities have been reported in only a few cases with CBL mutations and it is still debated whether the CBL syndrome falls into the category of a RASopathy, or represents a different entity. Our patient broadens this Noonan-like syndrome to include dysplastic pulmonary valve stenosis and cryptorchidism and prenatal onset of JMML, a hitherto undocumented situation. It also emphasizes the utility of WGS in establishing a clinically suspected diagnosis.

Molecular and Cytogenetic Evaluations of Atypical Anterior Segment Dysgenesis Syndromes With Cardiac Involvement

R.B. Hufnagel¹, M.B. Yang¹, M.E. Gray¹, L.A. Krueger¹, T.J. Jaworek², P.L. Bender¹, P.-W. Chiang³, X. Li¹, T.A. Smolarek¹, H.M. Saal¹, R.J. Hopkin¹, Z.M. Ahmed²

¹Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio

²University of Maryland, Baltimore, Maryland

³Oregon Health and Science University, Portland, Oregon

Anterior segment dysgenesis (ASD) encompasses a wide spectrum of developmental abnormalities of the anterior ocular segment, including congenital cataract, iris hypoplasia, aniridia, iridocorneal synechiae, as well as Peters, Axenfeld, and Rieger anomalies. An increased risk of glaucoma, amblyopia, and blindness lead to early surgical intervention. Syndromic ASD, such as Axenfeld-Rieger syndrome (ARS; MIM 602482) and Oculofaciocardiodental syndrome (OFCD; MIM 300166), are caused by a variety of genes, predominantly transcription factors and cofactors (e.g., PAX6, PITX2, FOXC1, BCOR) critical to the development of many tissues, particularly ocular, neuronal, craniofacial, dental, cardiac, and renal. Here, we describe two cases of atypical syndromic ASD. Case #1 is an infant female with bilateral congenital cataracts, posterior embryotoxon, iris transillumination defects, small optic nerves, nystagmus, scalp alopecia, dysmorphic facial features, micrognathia, cleft palate, delayed dentition, and atrial septal defect. There was no family history of similar features. Next-generation sequencing and copy number analysis of 14 genes associated with ASD were normal. SNP microarray revealed a 286 kb deletion on chromosome Xp11.4 proximal to the BCOR gene, including the 5'UTR, promoter, and 200 kb of upstream noncoding DNA, and excluding all BCOR coding exons. This deletion was not detected in either parent or in more than 4000 microarrays on this platform. Sanger sequencing of BCOR coding exons revealed no pathogenic mutations. Fibroblasts from skin biopsy of the proband and her mother were cultured for RNA extraction. BCOR expression, measured by qRT-PCR, was reduced to 70% in the proband compared to her mother, consistent with haploinsufficiency. To our knowledge, this is the first case of syndromic ASD caused by a noncoding BCOR mutation and is distinct from X-linked dominant OFCD in both the craniofacial and trichologic features. Case #2 is a 38 year old female with congenital glaucoma, bilateral iris hypoplasia, congenital cataracts, hypothyroidism, and mitral valve prolapse. Family history is consistent with a highly variable autosomal dominant condition that includes isolated glaucoma, iris hypoplasia, aniridia, cataract, hypothyroidism, congenital heart anomalies, and cystic kidney disease. Whole exome sequencing was performed on two individuals, revealing a previously unreported truncating mutation in FOXC1 [c.313_314insA; p.Tyr105*] that disrupts the alpha-helical region of the DNA-binding forkhead box domain. This variant segregated with disease in four additional family members and was absent from an unaffected member. No mutations in other ASD-associated genes were found to explain the phenotypic variability within this family, which spans the entire spectrum of FOXC1-associated diseases, from congenital glaucoma to ARS. These cases highlight both the similarity and variability between the syndromic ASD disorders OFCD and ARS, as well as the diagnostic and testing challenges. ASD should be considered in apparently isolated congenital glaucoma or cataracts. The additional risk of congenital cardiac and renal disease should also be considered when ASD is noted. Combined molecular and cytogenetic analysis may be the most efficient approach to establish a diagnosis and complete a systemic evaluation. Finally, noncoding mechanisms for BCOR and other ASD-associated genetic conditions should be considered when a molecular etiology is uncertain.

Mutation Spectrum in Patients With Suspected Isolated Ectopia Lentis

Teresa M. Neuhann¹, Annette Stegerer¹, Angelika Riess², Edward Blair³, Thomas Martin⁴, Lilian Bomme Usanger⁵, Rüdiger Kläs⁶, Arjan Bouman⁷, Kristiina Avela⁸, Trine Bjørg Hammer⁹, Alma Kuechler¹⁰, Olaf Rittinger¹¹

¹MGZ Medizinisch Genetisches Zentrum, Munich, Germany

²Institute of Medical Genetics and Applied Genomics, University of Tuebingen, Tuebingen, Germany

³Oxford University Hospitals NHS Trust, Oxford, United Kingdom

⁴Gemeinschaftspraxis für Humangenetik, Homburg, Germany

⁵Department of Clinical Genetics, Odense University Hospital

⁶synlab MVZ Humangenetik Mannheim, Mannheim, Germany

⁷Department of Clinical Genetics, Academic Medical Center, Amsterdam, The Netherlands

⁸Department of Clinical Genetics, Helsinki University Hospital, Finland

⁹Kennedy Centret, Genetic Counseling Clinic, Glostrup, Denmark

¹⁰Institut für Humangenetik, Universitätsklinikum Essen, Universität Duisburg-Essen, Essen, Germany

¹¹Klinische Genetik, Universitätskinderklinik, Paracelsus Medizinische Privatuniversität, Salzburg, Austria

Ectopia lentis (EL) frequently is a symptom of connective tissue diseases such as Marfan syndrome where it is a major clinical diagnostic criterion. However, it can also be observed as an isolated finding. The correct differentiation between syndromic and isolated EL is crucial for the management and surveillance of the affected patients, especially regarding cardiovascular complications. Main causes for isolated EL are mutations in ADAMTSL4 (autosomal recessive) and FBN1 (autosomal dominant). The latter also are the major cause for syndromic EL, i.e., Marfan syndrome. Rarely, mutations in other genes, such as LTBP2, can cause an isolated ocular phenotype including EL.

We analyzed 19 patients with clinically suspected isolated EL for the underlying genetic cause. 18 patients were primarily screened for mutations in ADAMTSL4 and FBN1. In patients without identifiable mutation in ADAMTSL4 and FBN1, an NGS based gene panel containing 11 genes associated with EL was performed. One additional patient was primarily analyzed with the gene panel.

The most common cause for EL identified in this cohort were mutations in ADAMTSL4 (11/19); of these, 8 carried the common founder mutation p. (Gln256Profs*38) on at least one allele. Three patients carried FBN1 mutations (two missense mutations and one single exon deletion). Both missense mutations have already been described as causative for Marfan syndrome. Other molecular causes identified included mutations in LTBP2.

Molecular genetic testing in patients with ectopia lentis can facilitate the differentiation between isolated and syndromic EL. This can significantly influence the management, especially in young children, where a discrimination often is not possible clinically. Also, it can show up possible ocular complications such as glaucoma in LTBP2 mutation carriers.

Ocular Manifestations in the X-Linked Intellectual Disability Syndromes

Natario L. Couser, MD^1,2, Arthur S. Aylsworth, MD^2,3, Roger E. Stevenson, MD⁴

¹Department of Ophthalmology, University of North Carolina at Chapel Hill (UNC-CH)

²Department of Pediatrics, UNC-CH

³Department of Genetics, UNC-CH

⁴Greenwood Genetic Center, JC Self Research Institute of Human Genetics

Intellectual Disability (ID) is the most common disorder affecting brain development and function, with an estimated incidence in children of 1–2%. Genetic conditions that cause intellectual impairment are often accompanied by recognizable syndromal patterns. The vision apparatus is a sensory extension of the brain, and persons with intellectual disabilities frequently have coexisting abnormalities of ocular structures and the visual pathway system. X- linked Intellectual Disability (XLID), which comprises 10–25% of all cases of ID, is the most extensively studied category of genetically caused cognitive impairment.

A causative gene has been identified for 122 XLID syndromes and a locus has been regionally mapped for 24 others. No causative gene is known for 41 clinically defined XLID syndromes.

While comprehensive ocular examinations have not been performed in all XLID syndromes because of the limiting need for sedation or general anesthesia, many ocular abnormalities have been well described. Some XLID syndromes (e.g., Aicardi, Cerebrooculogenital, Graham anophthalmia, Lenz, Lowe, MIDAS) are widely known for their ocular manifestation. About one-third of the XLID syndromes have significant eye or ocular adnexa abnormalities that provide important diagnostic clues in identifying these syndromes. Findings such as epicanthus, hypertelorism, hypotelorism, synophrys, abnormal slanting of the palpebral fissures, blepharophimosis, and an unusual appearance of the eyebrows and lashes contribute to a distinctive appearance of the face and are helpful in diagnosis. Nystagmus, optic atrophy and strabismus are among the more common, nonspecific, ocular manifestations that contribute to neuro- ophthalmological morbidity.

While a causative gene is known in the majority of XLID syndromes that have ocular manifestations, many of the syndromes that have not been mapped have unique ophthalmic findings (e.g., abducens palsy in Schimke; choroideremia in Van den Bosch).

Four XLID syndromes with major ocular manifestations are notable because of male lethality and expression occurring predominantly in females. In each syndrome, there are other somatic manifestations: ectodermal defects in incontinentia pigmenti, Goltz and MIDAS syndromes; skeletal anomalies in Aicardi and Goltz syndromes; and CNS malformations or seizures in Aicardi syndrome and incontinentia pigmenti.

SRD5A3-CDG: A Congenital Disosrder of Glycosylation With Extensive and Variable Eye and Skin Involvement

P.G. Wheeler, MD¹, B.G. Ng, MS², L. Sanford, RN¹, H.H. Freeze, PhD²

¹Nemours Children's Clinic, Orlando, Florida

²Human Genetics Program, Sanford-Burnham Medical Research Institute, La Jolla, California

Increasing numbers of congenital disorders of glycosylation (CDG) have been reported in recently resulting in an expansion of the phenotypes associated with this group of disorders. SRD5A3 codes for polyprenol reductase which converts polyprenol to dolichol. This is a major pathway for dolichol biosynthesis for N-glycosylation. We review the features of five individuals with loss of function mutations in SRD5A3 focusing on the unusual eye and skin involvement. Patient 1 had horizontal nystagmus since infancy. Eye evaluation found moderate optic nerve hypoplasia in one eye and severe optic nerve hypoplasia in the other with moderate myopia. She had marked short stature but no evidence for hypothyroidism. She has history of febrile seizures and global delays. Head MRI at 9 y/o showed left colpocephaly and hypoplastic inferior cerebellar vermis. She had diffuse ichthyosiform changes, palmoplantar keratoderma, and fleshy prominence in upper left sacral area. She walked at 4 y/o, and first words were at 6 y/o. At 12 ½ y/o, she can put together 4–5 word sentences. SNP microarray found three areas of homozygosity most pertinently at 4q11q13.1 which includes the SRD5A3 gene. CDG testing found elevated Mono-oligo/Di-oligo ratio and A- oligo/Di-oligo ratio consistent with disorder of N-glyosylation. A homozygote nonsense mutation was confirmed in SRD5A3. Patient 2 is brother to patient 1 through non-consanguineous parents of Puerto Rican origin. He had bilateral iris and optic nerve colobomas with horizontal nystagmus and high myopia. Head MRI found a hypoplastic inferior cerebellar vermis. He was diagnosed with hypothyroidism at 6 y/o. He had diffuse palmar plantar keratoderma and a fleshy prominence in the upper sacral area. He started walking at 4 y/o but he is awkward and falls easily. He says 2–3 word sentences. SNP array found same area of homozygosity at 4q as his sister, but it was not reported initially because total amount of homozygosity did not exceed 10 MB. He has same homozygote mutations as his sister. Patients 3 and 4 are female siblings, 33 and 34 y/o, with intellectual disability, ataxia, optic nerve atrophy, and nystagmus. One sibling had kyphosis and history of medulloblastoma. One sibling had cataracts. Mutations in SRD5A3 were found on whole exome sequencing. Patient 5 is a 10 y/o male with intellectual disability, epilepsy, optic nerve atrophy and nystagmus. Mutations in SRD5A3 were found on whole exome sequencing. A small number of case reports on SRD5A3-CDG have documented other visual issues that may develop over time including cataracts developing in the teenage years in three affected siblings and cataracts and retinal bone spicule pigmentation in two adult brothers. SRD5A3-CDG appears to be the only type of CDG with structural eye abnormalities. Possible explanations for the eye abnormalities will be discussed.

CRANIOFACIAL I

Mandibulofacial Dysostoses Are Caused by Defects of the Ribosome or Spliceosome

D. Wieczorek

Institut für Humangenetik, Universitätsklinikum Essen, Universität Duisburg-Essen, Essen, Germany

The craniofacial anomalies observed in the mandibulofacial dysostoses (MFDs) result from defective development of the first and second branchial arches. The craniofacial anomalies comprise downslanting palpebral fissures, coloboma of the lower eyelid, hypoplasia of the zygomatic complex, micrognathia, and microtia, which is often associated with hearing loss. At least 13 different types of MFD have been clinically described and this number is likely to rise. Treacher Collins syndrome (MIM 154500) is the best-known MFD. Two subtypes of TCS, TCS2 (MIM 613717) and TCS3 (MIM 248390), were described in 2011. Heterozygous mutations of the TCOF1 gene, which encodes for the nucleolar phosphoprotein Treacle, are causative of TCS. In a mouse model for TCS, Dixon et al. (2006) showed that haploinsufficiency of Tcof1 leads to defects in migration of neural crest cells, which result in severe craniofacial malformations. Treacle is assumed to be a regulator of ribosome biogenesis. The two genes POLR1D and POLR1C were reported to be causative of TCS2 and TCS3, respectively. These genes encode subunits of the RNA polymerases I and III, which are both involved in ribosomal RNA transcription.

MFD type Hutterite was formerly published as a distinct type of MFDs, but it was determined that this condition is also caused by mutations in the TCOF1 gene.

MFD Guion-Almeida type (MFDGA, MIM 610536) is defined by ID, microcephaly, choanal atresia and cleft palate, congenital heart defect and esophageal atresia. In 2012, the EFTUD2 gene was shown to cause this autosomal dominant condition. The EFTUD2 gene (elongation factor TU GTP-binding domain-containing 2) encodes for the U5-116kDa protein, a component of the major spliceosome. It occupies a central position within the U4/U6-U5 tri- snRNP particle. As thumb anomalies are not uncommon, Nager syndrome (MIM 154400) is an important differential diagnosis. Interestingly, Nager syndrome is caused by mutations in SF3B4, also encoding for a component of the major spliceosome.

Burn–McKeown syndrome (BMKS, MIM 608572). The craniofacial phenotype with narrow palpebral fissures, coloboma of the eyelids, high nasal bridge, and an expressionless face in combination with choanal atresia, hearing loss and congenital heart defect should be an easily recognizable constellation of this condition, although there is some overlap with TCS. The molecular basis could be identified very recently: biallelic mutations in TXNL4A, which encodes for one of the eight proteins of the U5 spliceosomal complex, which is critical for mRNA splicing. Oculo-oto-facial syndrome (MIM 610332) is allelic to BMKS.

Mandibulofacial dysostosis with alopecia is caused by missense mutations in the Endothelin Receptor Type A gene. This receptor functions in a signaling pathway which is essential for the establishment of mandibular identity during development of the first pharyngeal arch.

The molecular basis of the MFDs types Toriello (MIM 301950), Hedera-Toriello-Petty (MIM 608257), Bauru (MIM 604830) and Zhang has yet to be determined.

Most genes responsible for MFDs are involved in ubiquitous processes such as RNA transcription and splicing. Future studies are needed to understand why such processes lead to such specific phenotypes and to discover the genetic bases of the remaining, unresolved MFDs.

This work was supported by the German Ministry of Education and Research for the CRANIRARE and the FACE consortium (BMBF 01GM1211B and 01GM1109B).

Human Craniofacial Development: New Insights and Growth Metrics From High-Resolution 3D Imaging of Normal and Malformed Embryonic and Fetal Specimens

T.C. Cox, PhD, S. Borgeia, DDS, MS, K. Yu, PhD, M. Deng, MS, T. Naluai-Cecchini, I.A. Glass, MD, MB, ChB

Department of Pediatrics, University of Washington & Seattle Children's Research Institute, Seattle, Washington

Understanding the genetic and developmental basis of craniofacial dysmorphism and malformation is critical for optimal treatment of the patient and necessary if interventional or preventative therapies are to be developed. Key to such understanding is a detailed knowledge about normal craniofacial development. Implicit in this is an appreciation of both the timing of growth of different parts of the embryonic face and fetal cranioskeleton and the natural variation in these processes, in a similar way that accurate growth charts inform postnatal and prepubertal patient management.

Much of our current knowledge about human craniofacial development has come from careful visual observations, detailed histological sectioning, and 2D X-ray images of either naturally or medically-terminated malformed conceptuses or normal specimens from a limited number of archival collections. Comparison and extrapolations from animal models have also had a significant contribution despite morphological and functional differences.

With the availability of, and continual advances in, high-resolution 3D imaging modalities, we have begun to systematically collect and image facial tissue from terminated conceptuses with informed parental consent. Initial efforts are focused on specific growth metrics, particularly of the different components of the cranioskeleton, but also on more accurately documenting the timing of key events such as secondary palate development and facial suture fusions, and changes in cranial base angle, which are all believed to impact both normal postnatal facial morphology and dysmorphology. Where possible, all conceptal tissue is staged using one or, preferably, more methods including foot length, ultrasound determinations and other medical records. The typical specimen age range available for scanning is between 54 and 130 days post conception, although occasionally specimens are obtained outside this range. Most specimens are being imaged using microcomputed tomography, however occasional younger specimens are obtained and imaged using optical projection tomography. Although most of the conceptal tissue is classified as ‘normal’, we are also providing quantitative 3D data on malformations and to date have exquisite examples of fetuses with Apert syndrome, Opitz GBBB syndrome, Trisomy 18 and other cases of exencephaly which reveal new insights into pathogenetic mechanisms. Finally, stage and tissue specific gene expression and immunohistochemical studies are also being conducted for comparison with animal models to better inform translatability from animal models of malformation syndromes.

Identification of Forkhead Box (FOX) Family Members Regulated By Hedgehog Singaling In Non- Syndromic Cleft LIP Pathogenesis

J.L. Everson, G.W. Heyne, R.J. Lipinski

Department of Comparative Biosciences, University of Wisconsin-Madison, Madison, Wisconsin

The Hedgehog (Hh) signaling pathway is a conserved morphogenetic regulator critical for brain and face development. We have previously shown that in utero exposure to the Hh pathway inhibitor cyclopamine from gestational day (GD) 8.25 to GD9.5 results in cleft lip with or without cleft palate (CL/P) in the mouse. Notably, these clefts occur in the absence of the brain malformations that define holoprosencephaly, arguing that this model replicates “non-syndromic” CL/P phenotypes. We have demonstrated that the morphogenesis of cyclopamine-induced CL/P involves a tissue-deficiency of the frontonasal prominence-derived medial nasal process, which gives rise to the medial aspect of the upper lip and primary palate. Here, we aimed to identify the Hh target genes dysregulated during the initial pathogenesis of CL/P. To determine the spatiotemporal effect of cyclopamine exposure on Hh pathway activity, we first conducted a time-course by in situ hybridization for the conserved Hh target gene Gli1 on progressively-staged embryos following drug administration. We found that cyclopamine exposure results in robust downregulation of pathway activity in the frontonasal prominence by GD9.25. We then performed DNA microarray analysis on microdissected and pooled frontonasal prominence tissue from cyclopamine- or vehicle-exposed GD9.25 embryos. Network and clustering analysis of significant differentially expressed genes identified the enrichment of genes involved in ventral-dorsal patterning, neurospecification, and epithelial and mesenchymal cell proliferation. In addition, the Forkhead box (Fox) family of transcription factors was revealed as a broad, downstream target of Hh signaling in CL/P pathogenesis, with nine members (Foxa2, b2, c1, c2, d1, d2, e1, f1, and l1) downregulated in cyclopamine-exposed embryos by microarray and validated by RT-PCR. Interestingly, mutations in human FOXE1 were recently associated with both syndromic and non-syndromic forms of CL/P, as well as cleft palate only. However, a systematic evaluation of the Fox family members in human orofacial clefting etiology has yet to be conducted. Our study highlights specific candidate human disease genes by identifying multiple Fox family members as novel targets of Hh signaling during the initial pathogenesis of non-syndromic CL/P in the mouse. Examination of these genes in phenotypically-correlated patient samples is a promising avenue to further elucidate the complex etiology of human orofacial clefting.

Epidemiologic Data Support the Hypothesis That Factors Causing Nonsyndromic Orofacial Clefts May Also Have Detramental Effects on Brain Development

Arthur S. Aylsworth¹, Stephanie Watkins¹, Robert Meyer², Ronald Strauss¹, Jeffrey R. Marcus³, Alexander Allori³, Luiz Pimenta¹, Katie Harmsen², Barry Ramsey¹, Robert J. Lipinski⁴

¹University of North Carolina at Chapel Hill

²North Carolina State Center for Health Statistics, Raleigh, North Carolina

³Duke University, Durham, North Carolina

⁴University of Wisconsin-Madison

Orofacial clefts (OFCs) are among the most commonly occurring morphologic anomalies. Previous studies have shown that children with OFCs have impairments in speech and language development, memory, and processing speed. The cause of this is unclear. One possibility is that they are secondary, due to abnormal speech mechanics, hearing-loss, anesthesia, and/or social stigma related to physical appearance. An alternative explanation is that these deficits result from primary errors of brain morphogenesis due to the original cause(s) of the OFCs. The North Carolina (NC) Cleft Outcomes Study includes 559 nonsyndromic (NS) OFC cases of unknown cause and 6,822 control children without structural birth defects. Classification used abstracted surveillance data and individual chart reviews to exclude those with chromosome abnormalities and teratogen exposures. NC state end of 3rd grade (EoG) testing data showed that, after adjusting for covariates, children with cleft lip and palate (CL +P) were 1.74 times as likely to fail to meet grade-level standards in both reading and math, compared to controls.

Table. Odds of academic failure in both reading and math on 3rd grade EoG testing*

*Model adjusted for maternal education, public pre-K enrollment, race

The finding that CL+P cases have the most significant cognitive deficits is compatible with the hypothesis that some of the unknown genetic/environmental/stochastic factors causing NS OFCs also have detrimental effects on brain development. Growth of the facial processes that form the upper lip (particularly the medial nasal process) is dependent upon inductive molecular signals from the adjacently developing forebrain. Model animal studies show that the critical period for formation of cleft lip with or without cleft palate (CL±P) is immediately after that for holoprosencephaly. The critical period for CP is later in development, after forebrain patterning has been established. Therefore, CL±P should be more likely to occur with forebrain anomalies than CP. CL+P involves a more severe tissue deficiency than CL, suggesting the possibility of a more severe insult to the developing forebrain. It should be noted that these data could also be compatible with the hypothesis that anesthesia has detrimental effects on brain development.

The ability of epidemiologic studies to identify potentially causal relationships relies heavily on the specificity of case classification. Previous studies have grouped OFCs in less specific categories or have not appeared to exclude cases with the 22q11.2 deletion syndrome, which frequently presents as “isolated” CP. Our ability to demonstrate this differential effect may be related to our stricter emphasis on classifying cases with respect to potential confounding factors.

CRANIOFACIAL II

The Pre-Natal Phenotype Associated With Mutations in the ORC6 Gene, Causing Meier Gorlin Syndrome 3

Stavit A. Shalev^1,2, Morad Khayat², Daniel-Spiegel Etty³, Orly Elpeleg⁴

¹Genetics Institute, Emek Medical center, Afula

²Rappaport Faculty of Medicine, Technion - Israel Institute of Technology, Haifa

³Department of Obstetrics and Gynecology, Ha'emek Medical Center, Afula

⁴The Monique and Jacques Roboh Department of Genetic Research, Hadassah-Hebrew University Medical Center, Jerusalem, Israel

The Meier Gorlin syndrome (MGS) is a rare autosomal recessive disorder, characterized by the association of bilateral microtia, aplsia-hypoplasia of the patellae, and severe pre- and postnatal growth retardation, occasionally associated with other medical problems. Defects in genes encoding the Origin Recognition Complex (ORC) subunits are the cause MGS. The ORC comprises a highly conserved set of proteins which binds origins of DNA replication, and is instrumental for DNA integrity. Mutations in 5 different genes, including ORC1, ORC4, ORC6, CDT1 and CDC6 were found to cause MGS. Considerable phenotypic variation has been reported, with no clear genotype–phenotype correlation between the different gene categories. The ORC6 gene, responsible for MGS3, encodes the smallest subunit of human ORC. Only a handful of families with ORC6 mutations have been described (last search of the literature indicated eight families). Using whole exome analysis, we identified homozygous deleterious mutation in the ORC6 gene in fetuses of 1st cousins of Ashkenazi Jewish origin. As early as 11 week gestation, ultrasound examination revealed abnormal nuchal translucency, associated with evolving other distinctive findings including severe intrauterine growth restriction, dislocation of knees, gracile bones, club feet, small chest and small mandible. Hitherto, relatively mild mutations in ORC6 were reported in lesser affected patients with MGS type 3. We expand the clinical phenotype associated with ORC6 mutations to include severely abnormal embryological development.

Of MEIS And Men: MEIS2 Plays an Integral Role in Palate, Cardiac and Brain Development

Elaine H. Zackai^1,2, Margaret H. Harr¹, Matthew A. Deardorff^1,2

¹Division of Genetics, Children's Hospital of Philadelphia

²Department of Pediatrics, University of Pennsylvania Perelman School of Medicine

In 2010, we reported a boy with a cleft soft palate, a large posteriorly malaligned VSD and moderate hearing loss with a 123kb mosaic deletion in 15q14. It was predicted to include only exon 9 of the MEIS2 gene, lead to an out of frame deletion, and likely haploinsufficiency of the gene in the altered cells (Crowley et al., 2010). Follow up evaluation of the boy revealed that he also has mild developmental delay and significant autistic features. Subsequently, several reports have noted small numbers of patients with microdeletions that involved the MEIS2 gene region, with features that include submucous cleft palate, bilateral cleft lip and palate, VSD, mild ID and behavioral and/or social problems including autism spectrum disorder (Johansson et al., 2014).

This data implicated MEIS2 in cardiac, palate and intellectual development, although it was possible that these copy number alterations could be affecting the regulation of one or more nearby genes. A role for meis2 has been suggested in mouse for both midbrain and cardiac development (Paige et al., 2012;Kondo et al., 2014). Somewhat notably, knockout alleles for meis2 have been generated in mouse cell lines, but to date, no null or haploinsufficient mice have been reported. However, meis2b knockdowns have been performed in zebrafish, which have noted cardiac [looping] defects (Paige et al., 2012). As a result, the role of MEIS2 in vertebrate development has largely been suggested from human patients.

We recently evaluated a 4-year-old girl with a cleft palate and autism using clinical exome sequencing and identified a novel, de novo, c.1004T>C, p.Val335Ala missense mutation in MEIS2. We initially had some hesitation regarding the pathogenicity of this finding. However, MEIS2 is a homeobox domain-containing DNA-binding protein and the Val335 residue lies at the C-terminal end of helix 3 of the homeodomain; the exact position that has been demonstrated to bind to the major groove of the DNA binding site.

Therefore, we predict that disruption of DNA binding by the p.Val335Ala residue leads to loss of MEIS2 function, consistent with loss of function caused by microdeletion alleles and, subsequent a consistent phenotype. To support this model, Louw et al. (2015) also recently reported a female with cleft palate, septal defects, aortic coarctation and autism spectrum disorder with a three base pair deletion at a very similar site of the MEIS2 homeodomain (c.998-1000del; p.Arg333del). Molecular studies are underway to specifically test whether DNA-binding is altered and to what extent by these MEIS2 mutations. These findings emphasize that MEIS2 should be considered in the clinical contexts of isolated and syndromic cleft palate, cardiac septal defects and intellectual disability/autism.

Gingival Hyperplasia Expands the Phenotype of Temple–Baraitser Syndrome

Nuria C. Bramswig¹, C.W. Ockeloen²_, N.N. Kasri³, M. Selten³, J.C. Czeschik¹, A.J. van Essen⁴, E. Passarge¹, H. Engels⁵, T. Strom^6,7, H.-J. Lüdecke¹, T. Kleefstra², D. Wieczorek¹

¹Institut für Humangenetik, Universitätsklinikum Essen, Universität Duisburg-Essen, Essen, Germany

²Department of Human Genetics, Radboud University Medical Center, Nijmegen, The Netherlands

³Donders Institute for Brain, Cognition, and Behaviour, Department of Cognitive Neuroscience, Radboud University Medical Center, Nijmegen, The Netherlands

⁴Department of Genetics, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands

⁵Institute of Human Genetics, University of Bonn, Bonn, Germany

⁶Institute of Human Genetics, Helmholtz Zentrum München, Neuherberg, Germany

⁷Institute of Human Genetics, Technische Universität München, Munich, Germany

Temple–Baraitser syndrome (TBS, OMIM 611816) is a developmental disorder characterized by intellectual disability (ID), epileptic seizures, dysmorphic facial features, and broadened thumbs and great toes with hypoplastic nails. Thus far nine individuals with a tentative diagnosis of TBS have been published. Four causative mutations in the gene KCNH1 (OMIM 603305) have been identified recently (Simons et al., 2015). KCNH1 encodes a voltage-gated potassium channel homologous to the Drosophila Ether a go-go gene. It is highly expressed in the brain. Here we present four unrelated individuals with TBS with marked gingival hyperplasia from a patient cohort with ID. All four had normal birth weight and length. Muscular hypotonia and developmental delay were noted in their first year of life.

Individual 1, previously published by Rauch et al. (Lancet 2012), presented at the age of 9 ½ years with severe ID, short stature, hypoplastic midface, tapering fingers, hypoplasia of the great toenails and gingival hyperplasia. Seizures started at the age of 1 ½ years. Individual 2 presented with facial dysmorphia, gingival hyperplasia, nail hypoplasia and broad great toes, and severe ID with a normal EEG and no seizures. Individual 3 is a 4 ⁴/₁₂ year-old girl with severe psychomotor retardation, hypotonia, seizures and incontinentia. She is unable to walk and talk. Her thumbs are broad, but have normal nails. Individual 4 presented with short stature at the age of 14 yrs. ID without speech development, gingival hypertrophy, tapering fingers, hypoplastic great toenails, and generalized joint hypermobility were present. She developed seizures at the age of 9 months.

Whole exome sequencing revealed in individuals 1–3 an identical de novo mutation in the KCNH1 gene at 1q32.2: c.1070G>A; p.Arg357Gln, genomic coordinates 1: 211,093,374. The affected residue is highly conserved and located in the voltage-sensing transmembrane helix S4 of the KCNH1 channel. In individual 4 we found a de novo missense mutation in the KCNH1 gene at 1q32.2: c.1465C>T; p.Leu489Phe, genomic coordinates 1: 210,977,506. This mutation has been described previously by Simons et al. (Nature Genetics 2015) and is located in the transmembrane helix S6. In a cell-based assay in HEK293 cells we found electrophysiological evidence of a deleterious gain of function in both mutant KCNH1 channels.

In summary, we are able to widen the clinical spectrum of TBS by showing that gingival hyperplasia occurring before antiepileptic therapy seems to be a prominent feature previously not described in TBS. In addition, we show for the first time that marked hypoplasia of the great toe- and/or thumbnails, one of the main characteristics, can develop over time. Thus, we show that the phenotypic variability of TBS appears to be greater than previously assumed and present clinical overlap to a thus far unrelated syndrome, the Zimmerman-Laband syndrome (OMIM 135500).

Expanding the SPECC1L Mutation Phenotypic Spectrum to Include Teebi Hypertelorism Syndrome

E. Bhoj, MD, PhD, D. Li, PhD, M. Harr, MS, L. Tan, PhD, J. Hoffman, MD, H. Hakonarson, MD, PhD, E. Zackai, MD

Children's Hospital of Philadelphia, Philadelphia, Pennsylvania

Sperm Antigen With Calponin Homology And Coiled-Coil Domains (SPECC1L) encodes a protein involved in microtubule stabilization. Mutations in SPECC1L have previously been found in two patients with oblique facial clefts and two families with Opitz G/BBB Syndrome segregating in an autosomal dominant manner, but no other conditions (Saadi 2011, Kruszka 2015). In this study we performed whole exome sequencing on 24 families with syndromic craniofacial findings. In the two families who had been diagnosed with Teebi hypertelorism syndrome (THS) we found deleterious variants in SPECC1L.

The first patient had been previously reported as “Patient 1” by Hoffman et al in 2007 as a novel craniosynostosis syndrome with features of both Teebi and Aarskog Syndromes. The proband was born with hypertelorism, a natal tooth, two-vessel cord, hydrocele, left preauricular pit, micrognathia, protruding umbilicus, a shawl scrotum, and small hands. He had sagittal and coronal synostosis repair at 3 months, required surgery for bilateral ptosis, and a VSD repair at 4 years. He had some initial developmental delays, but is now in a regular classroom at school. His mother also had hypertelorism, ptosis, bicornate uterus, preaurciuar pit, short stature. By photographic review her father also had hypertelorism and ptosis. The proband and his affected mother were both found to have a likely pathogenic c.1260G>C:p.E420D novel variant in SPECC1L in the second coiled-coil domain, where previous mutations have been described.

The second patient has carried a diagnosis of THS since birth. He was born with hypertelorism and a giant omphalocele causing pulmonary hypoplasia requiring tracheostomy for the first three years of life. On exam he also had a slightly short upturned nose, thin upper lip, and bilateral preauricular pits and short stature. At eight years he was diagnosed with autism and pervasive developmental disorder. He was also found to have an idiopathic dilated aortic root. His parents do not have any facial differences or significant medical issues. The proband was found to have a c.1198_1203delATACAC:p.400_401del novel variant in SPECC1L, also in the second coiled-coil domain.

THS was first described as a new autosomal dominant syndrome in 1987 and subsequently a dozen families have been reported with a range of overlapping phenotypes. We reviewed these 12 families and identified shared features with our patients. 22/22 (100%) had hypertelorism, 16/22 (73%) had umbilical malformations including omphalocele, 5/9 (56%) of males had shawl scrotum, 6/22 (27%) had natal teeth, 6/22 (27%) had preauricular pits or fistulae, and 3/22 (17%) had cranial suture abnormalities.

Phenotypic features in the patients summarized here also overlap with previously reported patients with SPECC1L mutations (Kruszka 2015), all patients had hypertelorism, three patients had umbilical hernias, two patients had craniosynostosis, two had bicornate uterus, and one patient had aortic root dilation. We would like to emphasize that as at least two patients with SPECC1L mutations have a dilated aortic root, and a significant percentage have craniosynostosis, therefore these should be screened for in the appropriate clinical setting. As a result, we propose that SPECC1L testing is warranted in any patient with a THS diagnosis.

Genetic Modifiers: Executive Function, Adaptive Behavior, Craniosynostosis, and Hearing Loss in Muenke Syndrome

Colin M.P. Yarnell, BA¹, Yonit A. Addissie, BA¹, Donald W. Hadley, MS¹, Maria J. Guillen Sacoto, MD¹, Nneamaka B. Agochukwu, MD¹, Rachel A. Hart, BS¹, Edythe A. Wiggs, PhD¹, Petra Platte, PhD², Yvonne Paelecke, PhD², Hartmut Collmann, MD³, Tilmann Schweitzer, MD⁴, Paul Kruszka MD, MPH¹, Maximilian Muenke, MD¹

¹Medical Genetics Branch, National Human Genome Research Institute, National Institutes of Health

²Institute of Psychology, Department of Psychology I, University of Würzburg, Würzburg, Germany

³Department of Neurosurgery, Section of Pediatric Neurosurgery, University Hospital of Würzburg, Würzburg, Germany

⁴Department of Orthodontics, University Hospital of Würzburg, Würzburg, Germany

Objectives: This study investigates executive function and adaptive behavior in persons with Muenke syndrome using validated instruments with a normative population and unaffected siblings as controls; and assesses the role of genetic modifiers on behavior, craniosynostosis, and hearing loss.

Study Design: Participants in a cross sectional study included individuals with Muenke syndrome (P250R mutation in FGFR3) and their mutation negative siblings. Participants completed validated assessments of executive functioning (Behavior Rating Inventory of Executive Function; BRIEF) and adaptive behavior skills (Adaptive Behavior Assessment System; ABAS-II). Next generation sequencing was performed at the National Intramural Sequencing Center (NISC). Craniosynostosis was assessed with cranial CT and hearing loss was evaluated with audiography.

Results: Forty-four FGFR3 mutation positive individuals, median age 9 years, range 7 months to 52 years were enrolled. Additionally, 10 unaffected siblings were used as controls (5 males, 5 females, median age of 13 years, range 3 to 18 years). Of the 44 affected participants, 35 had craniosynostosis (83.3%), 29 participants (67.4%) had a history of at least one craniosynostosis corrective surgery, and 23 (63.9%) had a history of hearing loss (usually mild to moderate). For the General Executive Composite scale of the BRIEF, 32.1% of the cohort had scores greater than +1.5 SD, signifying “Potential Clinical Significance.” For the General Adaptive Composite of the ABAS-II, 28.2% of affected individuals scored in the 3rd–8th percentile of the normative population and 56.4% were below the “Average” category (less than the 25^th percentile). Multiple regression analysis showed that craniosynostosis was not a predictor of BRIEF (P = 0.7) and ABAS-II scores (P = 0.7). In the sibling pair analysis, affected siblings performed significantly poorer in the BRIEF General Executive Composited and the ABAS-II General Adaptive Composite. One large family (n = 7) underwent whole exome sequencing with results pending.

Conclusion: Individuals with Muenke syndrome are at an increased risk for developing adaptive and executive function behavioral changes when compared to a normative population and unaffected siblings, regardless of whether craniosynostosis is present. Our intention is to use exome sequencing results to explore the effect of genetic modifiers on behavior, craniosynostosis, and hearing loss in Muenke syndrome.

NEUROLOGY II

Pontocerebellar Hypoplasia Type 7 and Other Rare Syndromes With Microcephaly and Disorders of Sexual Development

W.B. Dobyns,¹ D.K. Manchester,² Frank Baas³

¹Department of Pediatrics, University of Washington, and Seattle Children's Research Institute, Seattle, Washington

²Division of Genetics, Children's Hospital Colorado and University of Colorado Health Sciences Center; Denver, Colorado

³Departments of Genome Analysis and Neurology, Academic Medical Centre, University of Amsterdam, Amsterdam, The Netherlands

Several syndromes combining microcephaly and disorders of sexual development have been described, including several with additional brain malformations. Well known examples include X-linked lissencephaly with abnormal genitalia (XLAG) and campomelic dysplasia. A search of the Dobyns subject database (Lisdb) found five patients with severe genital hypoplasia excluding those with XLAG. These children have at least two and possibly several other syndromes.

The first is pontocerebellar hypoplasia with disorders of sex development, now designated PCH7. Patient LP99-206 had the developmental level of an infant with no language, poor head control, limited movement, no self-help skills, and poor sleep regulation. She was unable to sit, walk or communicate but was able to feed by moth. She had no secondary sexual development. Her seizures were well controlled. She had no history of regression, but her developmental level was so low that regression might not be noticed. Chromosome analysis showed a normal male karyotype (46,XY), and at surgery, she was found to have a uterus, fallopian tubes and limited ovarian tissue with a single graafian follicle. Brain MRI at 11 months showed a normal gyral pattern with mild atrophy, moderately enlarged 3rd and lateral ventricles, severe hypoplasia of the corpus callosum, mild brainstem hypoplasia and severe cerebellar hypoplasia. Unlike most other types of PCH, the vermis is not relatively spared. Summaries of ∼10 other patients with PCH7 collected for phenotype and genetic analysis show very similar abnormalities with a wide spectrum of severe genital anomalies.

The second is a novel syndrome of progressive postnatal microcephaly, agenesis of the corpus callosum, profound intellectual disability, epileptic encephalopathy, chronic diarrhea, and ambiguous genitalia. Patient LP99-092 had a known normal male karyotype (46,XY) from prenatal testing, but had apparent female external genitalia at birth, and the onset of seizures at 3 hr of age. Within the first week he developed severe diarrhea that caused metabolic acidosis. He also fed poorly. He went on to have chronic diarrhea, poor feeding, poorly controlled seizures and multiple episodes of sepsis. He was managed with gastrostomy feeds and parenteral nutrition supplements. At 13 months, his OFC was 42.3 cm (−4 SD), He had severe global developmental handicaps including no visual behavior, no self-help skills, no ability to communicate, and diffuse hypotonia with increased reflexes. His genitalia appeared female, but with no vaginal vault. No testes could be palpated. MRI at 7 months shows severe microcephaly, enlarged ∼1 cm extra-axial space, diffuse simplified gyral pattern with thin cortex, 1 × 3 cm interhemispheric cyst, enlarged lateral ventricles, complete agenesis of the corpus callosum, normal brainstem, and mild vermis hypoplasia. The closest matches are single reports of “microcephaly, jejunal atresia, aberrant right bronchus, ocular anomalies, and XY sex reversal” (Keegan et al., 2004), and ambiguous genitalia, microcephaly, seizures, bone malformations, and early death: a distinct MCA/MR syndrome (Megarbane et al 2011). But neither appears to be the same syndrome.

Pontocerebellar Hypoplasia With Sex Reversal Caused by Mutations in the TOE1 Gene

Chitayat David^1,2,3,Veerle Eggens⁴, Christina Fagerberg⁵, Maria Kibæk⁶, Susan Blaser², Almundher Al-Maawali¹, Komudi Siriwardena¹, Frank Baas⁴

¹Division of Clinical and Metabolic Genetics, Department of Pediatrics, Ontario, Canada

²Neuroimiging, Sickkids Hospital, Ontario, Canada

³The Prenatal Diagnosis and Medical Genetics Program, Department of Obstetrics and Gynecology, Mount Sinai Hospital, University of Toronto, Toronto, Ontario, Canada

⁴Academic Medical Center, Department of Neurogenetics, University of Amsterdam, Amsterdam, The Netherlands

⁵Odense University Hospital, Odense, Denmark

⁶HC Andersen Children Department, Odense, Denmark

Pontocerebellar hypoplasia (PCH1-10) represents a group of autosomal recessive neurodegenerative disorders of prenatal onset. Patients show variable hypoplasia of the pons and cerebellum and severe motor and mental impairments. We present two cases with PCH with sex reversal caused by mutations in the TOE1 gene.

Case 1: The proband was born to a 28-year-old primigravida mother and a 32-year-old father. Both parents were healthy, Caucasian and non-consanguineous. Prenatal ultrasound at 19 weeks gestation was normal. Delivery was at term, spontaneous, vaginal and uncomplicated. The Apgar scores were normal and the birth weight, length and head circumference were at the 50th centile. She had normal female external genitalia. There were feeding difficulties requiring nasogastric tube feeding, but she was gradually transitioned to oral feeding at age two weeks. At 4 months she had hypertonia and delay in acquiring milestones. At age 12 months, she had complete head lag, was not able to roll over, sit or reach for objects. There was no social smile, she did not recognize her parents and did not turn her head to sound. There were no seizures. Investigation showed normal eye examination, VEP, and EEG. Pelvic MRI showed normal uterus, bilateral fallopian tubes and ovaries, a blind ending vagina. Brain MRI showed hypoplastic corpus callosum and significant brain stem, cerebellar and vermian hypoplasia, and atrophy consistent with PCH. Microarray analysis (Agilent, Oligo Array 44K) was normal and male (46,XY). WES revealed a mutation in the TOE1 gene and Sanger sequencing showed compound heterozygote mutations [p.R73S and p.Y231X] with each of the parents being heterozygote for one of the mutations.

Case 2: The patient was born to a non-consanguineous Denish parents. Their family histories were non-contributory. The pregnancy was uncomplicated and delivery was by cesarean section for slow progress. Birth parameters were normal and OFC was at -½ SD centile. The external genitalia was female. At 2 months she was noted to be hypertonic and with minimal fixation. Brain MRI showed hypoplastic cerebellum and vermis, hypoplasia of the corpus callosum and delayed myelination. Chromosome analysis was normal and male (46,XY). Abdominal MRI showed normal kidneys and adrenal glands. On surgery there were no gonads and a hypoplastic uterus. At 6 months she had feeding difficulties, her OFC decreased to -2SD, she had spastic tetraplegia and clenched fingers. She had minimal eye tracking and nystagmus. At 7 months she developed infantile seizures. Mutation analysis of the TOE1 gene showed compound heterozygosity for two mutations [p.V173G and p.H319Q] inherited from her parents.

Little functional information is available on the TOE1 gene. Since the gene mutation is associated with gonadal dysfunction, it is plausible that TOE1 mutations cause a defect in formation of the bipotential gonad. Further studies are being done to delineate its role in gonadogenesis and brain formation.

Clinical and Experimental Evidence Establish a Link Between KIF7 AND C5orf42- Related Ciliopathies

Reza Asadollahi¹, Justin E. Strauss², Martin Zenker³, Oliver Beuing⁴, Simon Edvardson⁵, Orly Elpeleg⁶, Pascal Joset¹, Tim M. Strom⁷, Dunja Niedrist¹, Beatrice Oneda¹, Silvia Azzarello-Burri¹, Michael Papik¹, Alessandra Baumer¹, Katharina Steindl¹, Albert Schinzel¹, Esther T. Stoeckli², Anita Rauch¹

¹Institute of Medical Genetics, University of Zurich, Zurich, Switzerland

²Institute of Molecular Life Sciences and Neuroscience Center Zurich, University of Zurich, Zurich, Switzerland

³Institute of Human Genetics, University Hospital Magdeburg, Magdeburg, Germany

⁴Institute of Neuroradiology, Otto-von-Guericke-University Magdeburg, Magdeburg, Germany

⁵Pediatric Neurology Unit, Hadassah University Hospital, Mount Scopus, Jerusalem, Israel

⁶Monique and Jacques Roboh Department of Genetic Research, Hadassah-Hebrew University Medical Center, Jerusalem, Israel

⁷Institute of Human Genetics, Technische Universität München, Munich, Germany

Joubert syndrome (JBTS) is a ciliopathy characterized by developmental delay, oculomotor apraxia, breathing abnormalities and a distinctive mid-hindbrain malformation on brain imaging known as molar tooth sign. It is genetically heterogeneous with >20 causal genes identified to date and its clinical and neuroradiologic findings overlap with other ciliopathies such as acrocallosal syndrome (ACLS). However, craniofacial manifestations of macrocephaly, prominent forehead and hypertelorism have been distinctive features of ACLS among ciliopathies. Here, we evaluated eight families with ACLS-like craniofacial appearance and other ciliopathy features by whole exome or targeted Sanger sequencing which revealed biallelic mutations affecting KIF7, the ACLS gene, or C5orf42, the JBST gene. Given the known role of KIF7 in primary cilia and Hedgehog signaling, we wondered if there was a functional connection to C5orf42 which's function has remained unknown. Therefore, we also assessed C5orf42 by evaluation of the primary cilia in affected individuals and in ovoRNAi silencing in chicken embryos. Consequently, in addition to the clinical overlap between KIF7and C5orf42related craniofacial features, we found abnormal primary cilia in the C5orf42 patients and evidenced its role in craniofacial development, pathfinding of commissural axons and neural circuit formation in developing chicken embryos.

Congenital Nephrosis, Cerebral Ventriculomegaly and Heterotopias—Expanding the Phenotype Associated With CRB2 Mutations

Anne Slavotinek¹, Gray K.², Rodan L.², Tan W.-H.²

¹Department of Pediatrics, University of California San Francisco, San Francisco, California

²Boston Children's Hospital, Boston, Massachusetts

We recently reported five fetuses and a child from three families who shared a novel phenotype that we term CRB2 syndrome comprising cerebral ventriculomegaly, echogenic kidneys and greatly elevated maternal serum alpha-fetoprotein (MSAFP) or amniotic fluid alphafetoprotein (AFAFP). Renal findings include tubular cysts at the corticomedullary junction, diffuse effacement of the epithelial foot processes and microvillous transformation of the renal podocytes, similar to findings in congenital nephrotic syndrome, Finnish type. In all cases, deleterious sequence variants in Crumbs, Drosophila, Homolog of 2 (CRB2) were identified and inheritance was autosomal recessive (AR). Studies in Danio rerio have shown that reduced crb2b function results in loss of the slit diaphragms of the podocytes and the nephrosis is considered to develop from an inability to develop a functional glomerular barrier. All individuals in the initial report were ascertained in the prenatal period.

We have identified an additional 11 cases with likely CRB2 syndrome. A 5-day old female had nephrotic syndrome with multiple renal microcysts and massive hydrocephalus, a large atrial septal defect, probable Scimitar syndrome and a Mullerian duct anomaly. Testing for CRB2 mutations is pending. A review of 10 cases from the literature (Reuss et al., 1989; Jolly et al., 2003) showed that the phenotype was relatively consistent, with additional findings comprising pericardial effusion (2/10), postaxial polydactyly (1/10), hyperplasia of the choroid plexus (1/10), duplication of the right kidney and ureter (1/10) and absent horizontal fissure of the right lung with incomplete rotation of the intestine (1/10).

Gray matter heterotopias were prominent in one child with CRB2 syndrome and there is overlap with Galloway-Mowat syndrome (GMS), an AR condition characterized by microcephaly, hiatal hernia and nephrotic syndrome in which heterotopias and neuronal migration disorders have been described. Two families with GMS had mutations in the WDR73 gene, but the condition is phenotypically heterogeneous and further genes may yet be found. We hypothesized that some individuals reported to have GMS may have had CRB2 syndrome and found one such example in a child with periventricular nodular heterotopias, splitting of the central canal of the spinal cord, a significantly raised AFAFP (20X normal) and renal echogenicity (Palm et al., 1986). We conclude that the CRB2 syndrome phenotype overlaps with other conditions that have early-onset nephrosis and cerebral malformations, such as GMS. The high AFP measurements and the renal histology may enable clinical distinction in these cases.

Jolly et al. Prenat Diag 2003;23:48 -51; Palm et al. Arch Dis Child 1986; 61:545–548; Reuss et al., Am J Med Genet 1989;33:385–389.

Re-Evaluation of a ‘negative’ Clinical Exome Reveals a Novel Cause of Severe Primary Microcephaly

M. Manning^1,2, S. Schelley², H. Shamsedin⁴, N. Ibrahim⁴, M. Hashem⁴, E. Esplin³, F. Alkuraya⁴

¹Department of Pathology, Stanford University School of Medicine

²Department of Pediatrics, Stanford University School of Medicine

³Department of Genetics, Stanford University School of Medicine

⁴Department of Genetics, King Faisal Specialist Hospital and Research Center

The diagnostic power of whole exome sequencing (WES) in Mendelian disorders is now well established. Wide- ranging estimates, however, have been reported regarding the diagnostic yield, and the difference in yield between clinical-grade and research- grade WES has not been fully explored. This issue is particularly pressing when faced with a “negative” clinical-grade WES because it is not usually clear if the causal mutation is missed due to limited sensitivity of the capture, sequencing, bioinformatic analysis or manual curation of the resulting variants since the standards for these steps can vary greatly between clinical- and research-grade WES. For example, it is often assumed that because of the high quality of clinical-grade WES, a negative result reflects an inherent platform limitation that may necessitate the move to whole-genome sequencing. We present a family with an apparently autosomal recessive form of severe primary microcephaly who tested negative on clinical-grade WES. However, when WES was repeated at a research facility with expertise in autozygome analysis, a homozygous intronic mutation in RTTN was identified.

Three brothers (ages 4, 7 and 8 years) presented for evaluation of growth retardation, microcephlay and intellectual disability. Parents were a consanguineous Yemeni couple. A form of a recessive or X-linked primary microcephaly/primordial dwarfism was suspected due to the severe growth failure (Z scores of −5 to −12) and extreme microcephaly (Z scores of > −8). All 3 boys demonstrated marked intellectual impairment and were nonverbal. MRI performed on the eldest boy in Egypt was significant for pachygyria, few sulcations, and shallow sylvian fissures. Diagnostic testing included normal CGH and primary microcephaly gene panels.

The RTTN mutation identified on research WES was confirmed on RTPCR and leads to complete replacement of the normal transcript by an aberrant transcript that predicts frameshift and premature truncation. RTTN encodes Rotatin, a centromere-associated protein, which determines early embryonic axial rotation. Only two families have been reported to date to have an RTTN-related phenotype, described as diffuse polymicrogyria with mild microcephaly. However, our family clearly expands the phenotype to include extreme primary microcephaly and growth failure.

In addition to further characterizing the phenotype of RTTN mutations in human disease, this report also serves as a reminder of the limitations of current genetic testing and the challenges in calling disease variants from genomic sequences even when the causal variant is present at the sequencing level.

Mutations of the Phosphoinositide-3-kinase Regulatory Subunit PIK3R2 Cause Perisylvian Polymicrogyria With or Without Megalencephaly: A Paradigm for Constitutional and Mosaic Mutations Causing Common Malformations of Cortical Development

Ghayda Mirzaa¹, Andrew E. Timms¹, Evan August Boyle², Jay Shendure³, Jamel Chelly⁴, William B. Dobyns¹, Renzo Guerrini⁵

¹Seattle Children's Research Institute, Seattle, Washington

²Department of Genetics, Stanford University School of Medicine, Stanford, California

³Department of Genome Sciences, University of Washington, Seattle, Washington

⁴University Paris Descartes, Paris, France

⁵Pediatric Neurology, Neurogenetics and Neurobiology Unit and Laboratories, Neuroscience Department, A. Meyer Children's Hospital, University of Florence, Florence, Italy

Polymicrogyria is among the most common malformations of cortical development (MCD). Bilateral perisylvian polymicrogyria (BPP), the most common form of regional polymicrogyria, causes the congenital bilateral perisylvian syndrome, featuring oromotor dysfunction, cognitive impairment and epilepsy. Genetic etiologies are collectively rare causes of BPP and typically occur in clinically recognizable syndromic forms, the most common of which are 22q11.2 and 1p36 deletion syndromes. To date, only one gene—RTTN—has been associated with isolated BPP. Here, we report on PIK3R2 mutations in 21 children with BPP. Polymicrogyria only affected the perisylvian cortex or extended beyond it with perisylvian predominance. Eight patients exhibited BPP without megalencephaly and thirteen had BPP in the context of the megalencephaly-polymicrogyria-polydactyly-hydrocephalus (MPPH) syndrome. Across the entire cohort, mutations were mosaic in nine patients and constitutional in twelve patients. Overall, patients with mosaic mutations were less severely affected.

PIK3R2 encodes the p85β regulatory subunit of the phosphatidylinositol-3-kinase (PI3K)-AKT-MTOR pathway. Twenty patients had the same recurrent mutation previously reported in PIK3R2 (c.1117G>A; p.Gly373Arg) and one patient had a different de novo missense mutation of PIK3R2 (c.1126A>G, p.Lys376Glu). In addition to de novo constitutional and postzygotic mutations, we also report on a PIK3R2 mutation inherited from an affected parent, and a family with parental germline mosaicism. Among the mosaic patients, we observed substantial variation in alternative allele fractions (AAF) within individual samples ranging from 2.5% to 36.7% of reads, equivalent to 5–73.4% of cells analyzed. Levels of mosaicism in blood-derived DNA ranged from undetectable levels to ∼26% of reads, whereas levels of mosaicism in saliva ranged from 29 to 43%.

We therefore demonstrate that constitutional and mosaic mutations in PIK3R2 are associated with a spectrum of abnormalities of cortical development ranging from isolated BPP with a normal head size, causing the congenital bilateral perisylvian syndrome, to the MPPH syndrome. The phenotypic variability and low-level mosaicism challenging conventional molecular methods have important clinical and diagnostic implications.

THERAPEUTICS

Single Gene Mosaic Overgrowth Disorders as a Model System for Oncology—From Dysmorphology to Precision Therapeutics

Leslie G. Biesecker, Marjorie Lindhurst, Kim Keppler-Noreuil

National Human Genome Research Institute, NIH, Bethesda, Maryland

The discovery of the molecular basis of mosaic overgrowth disorders of the AKT/PIK3CA pathway has generated an opportunity to understand cancer biology. The AKT1 p.E17K activating mutation is the sole cause of Proteus syndrome. A range of activating mutations in PI3KCA cause a range of phenotypes designated as PIK3CA-Related Overgrowth Spectrum (PROS). These discoveries demonstrated that a single gene variant, due to a somatic mutation, cause dysregulated growth and an increased risk of tumors. These somatic overgrowth disorders lie at one end of a spectrum of genomic perturbation, with the other extreme represented by the highly mutated and rearranged genome of metastatic cancer, which is also mosaic. We consider the former as simple mosaic disease and the latter as a complex mosaic disease.

The identification of these simple, single gene mosaic overgrowth disorders is a convergence of clinical dysmorphology, genomics, and cancer biology. The AKT-PIK3CA pathway is one of the most commonly dysregulated pathways in malignancies and major efforts have been undertaken to target this pathway for cancer therapeutics. An attraction of this pathway is that the mutations activate the proteins through constitutive signaling—it is easier to develop a small molecule inhibitor than a small molecule activator. In spite of this attribute, a key stumbling block in the targeting of this pathway is the ability of the cells to adapt to the inhibitors and develop resistance. Given that cancer represents microevolutionary environment due to genomic mutability and instability, the question arises as to whether resistance is attributable to the genomic instability or the inherent feedback attributes of the pathway.

Therefore, single gene mosaic disorders of this pathway provide an opportunity to test this question. We have probed the AKT-PIK3CA pathway using a small molecule inhibitor and have compared the results in Proteus and PROS cells to those from cancer cells and the result suggest that the simplicity of the single gene disorders allows manipulation of the pathway that may not be possible in tumors. Additionally, I will discuss the distinct therapeutic objectives of somatic overgrowth as compared to cancer. Mosaic somatic overgrowth disorders are a powerful example of how the study of obscure, extremely rare deforming overgrowth disorders can answer important questions regarding one of the most common maladies.

Pi3K-AKT Pathway: From Isolated Cancer to Syndromes. Metabolic Characterization and New Therapeutic Perspectives

L. Boccuto¹, L. Cascio¹, C. Chen¹, X. Lei², K. Jones¹, C. Skinner¹, R.E. Stevenson¹, C.E. Schwartz¹.

¹Greenwood Genetic Center, Greenwood, South Carolina

²Biolog, Inc., Hayward, California

The Pi3K-AKT pathway plays a key role in regulating cell growth and its abnormal activation has been associated with a wide spectrum of conditions, ranging from isolated cancer to complex syndromes. In previous studies we analyzed six cell lines from affected tissues of patients with mosaic gain-of-function mutations in several Pi3K-AKT genes and clinical diagnoses including Proteus syndrome, CLOVES syndrome, and Klippel-Trenaunay-Weber syndrome. In all samples, we were able to confirm the constitutive activation of the pathway by the increased phosphorylation of two specific residues of the AKT protein after cell starvation. We also investigated the AKT phosphorylation after stimulation by FBS and the metabolic response to growth factors, utilizing the Biolog Phenotype MicroArray technology. The results indicated that mutations affecting the Pi3K complex were associated with reduced AKT phosphorylation after stimulation and reduced metabolic response to growth factors, while AKT1 mutations caused increased AKT phosphorylation after stimulation and increased response to growth factors. In order to validate these preliminary findings, we employed the same combined approach to screen 10 new cell lines from patients with segmental overgrowth, 8 of which were tested in blinded fashion. A cell line from a breast adenocarcinoma carrying the PIK3CA p.H1047R mutation was also tested with the Biolog MicroArrays.

Our results showed the expected AKT hyperphosphorylation after starvation in all cells from affected tissues and confirmed that mutations affecting the Pi3K complex and the AKT protein generate opposite responses to stimulation by serum and growth factors. These findings might be explained by a negative feedback from the activated Pi3K complex on the interacting RTK receptor, leading to lower affinity for the growth factors. The AKT1 mutations apparently fail to induce such feedback and therefore lead to an increase of the pathway activity in the presence of growth factors. The characterization of the molecular and metabolic profiles of the eight blinded cell lines allowed us to separate the patients with PIK3CA mutations from those with AKT1 mutations. Interestingly, the cancer cell line showed a metabolic profile more compatible with mutations in AKT1 than in PIK3CA, perhaps indicating a lack of regulatory feedback inhibition of the RTK receptor, compatible with a more aggressive phenotype.

Also, as before, we tested several compounds that are supposed to inhibit different steps of the pathway and noticed that low concentrations of Rapamycin seem to partially restore the normal metabolic response to growth factors. In summary, we believe that our combined approach can be helpful in 1) identifying the mutated gene in the Pi3K-AKT pathway, 2) understanding the effect of the mutation at the cellular level, 3) suggesting and developing potential treatment approaches, and 4) following up the metabolic response of patients to pharmaceutical protocols.

­Targeted Therapy for a Novel Neurologic Disorder Due to Mutations in ATAD1

Rebecca C. Ahrens- Nicklas^1,2,5,*, George K.E. Umanah^8,9,*, Neal Sondheimer^1,5, Matthew A. Deardorff^2,5, Alisha B. Wilkens², Avni B. Santani^3,7, Addie Nesbitt³, Ted M. Dawson^,8,9,10,11, Valina L. Dawson^8,9,11,12, Eric D. Marsh ^4,5,6

¹Section of Biochemical Genetics, The Children's Hospital of Philadelphia, Philadelphia, Pennsylvania

²Division of Human Genetics, The Children's Hospital of Philadelphia, Philadelphia, Pennsylvania

³Department of Pathology and Laboratory Medicine, The Children's Hospital of Philadelphia, Philadelphia, Pennsylvania

⁴Division of Child Neurology, The Children's Hospital of Philadelphia, Philadelphia, Pennsylvania

⁵Department of Pediatrics, Philadelphia, Pennsylvania

⁶Department of Neurology Perelman School of Medicine, Philadelphia, Pennsylvania

⁷Department of Clinical Pathology, University of Pennsylvania, Philadelphia, Pennsylvania

⁸Neuroregeneration and Stem Cell Programs, Institute for Cell Engineering, Baltimore, Maryland

⁹Department of Neurology, Baltimore, Maryland

¹⁰Solomon H. Snyder Department of Neuroscience, Baltimore, Maryland

¹¹Pharmacology and Molecular Sciences, Baltimore, Maryland

¹²Physiology, Johns Hopkins University School of Medicine, Baltimore, Maryland

*These authors contributed equally to this work.

Whole exome sequencing (WES) is widely used in the evaluation of patients with undiagnosed neurologic disease. We describe a novel recessive monogenic disorder in patients from a highly consanguineous family with severe hypertonia, seizures, abnormal EEG, and early death due to respiratory failure. WES identified a homozygous nonsense mutation, p.E276X, in ATAD1 in the first patient and the variant tracked with disease in multiple family members. There are no previous reports of ATAD1 mutations in humans, but loss of ATAD1 function in mice was recently published and the mice have a similar phenotype with severe seizures and early lethality. ATAD1 encodes Thorase, an AAA+ ATPase whose primary function is to mediate internalization of post-synaptic α-amino-3-hydroxy-5- methylisoxazole-4-proprionate (AMPA) receptors. AMPA receptors are the primary excitatory neurotransmitter receptor in the brain. Regulation of excitatory neurotransmission by activity dependent insertion and removal of post-synaptic AMPA receptors is thought to be integral to learning and memory. In the mice, loss of Throase decreases internalization of AMPA receptors resulting in an increase in post-synaptic receptor density and excitatory synaptic currents, ultimately leading to excitotoxicity and neuronal death. Given this mechanism of disease, we hypothesized that AMPA receptor blockade could be a beneficial targeted therapy for individuals lacking functional Thorase. Indeed, treatment with perampanel, an AMPA receptor antagonist approved for the treatment of partial seizures, improved neurologic function in both the mice and patients with ATAD1 mutations. In conclusion, we report a novel neurological condition due to a nonsense mutation in ATAD1 that could be amenable to therapy targeting the downstream action of the ATAD1-encoded protein, Thorase. Further work to determine dosage, timing, and ideal treatment protocols are needed.

KCNK9 Imprinting Syndrome—A Treatable Disorder?

John M. Graham Jr., MD, ScD¹, Neda Zadeh, MD², Lena Sagi-Dain, MD³, Stavit A. Shalev, MD^3,4

¹Harbor-UCLA Medical Center, Cedars-Sinai Medical Center, Los Angeles, California

²Division of Medical Genetics, CHOC Children's Hospital, Orange, California

³Genetic Institute, Emek Medical Center, Afula Israel

⁴Rapapport Faculty of Medicine, Technion, Haifa, Israel

Barel et al. (2008) mapped a new KCNK9 imprinting syndrome to chromosomal region 8q24 in a large Arab Israeli family, and demonstrated that the disease is caused by a missense mutation 770G>A in exon 2, replacing glycine at position 236 by arginine (G236R) in the maternal copy of KCNK9 within this locus. KCNK9 (also called TASK3) is maternally transmitted, and encodes a member of the two pore-domain potassium channel (K2P) subfamily. This gene is normally imprinted with paternal silencing, thus a mutation in the maternal copy of the gene will result in disease, whereas a mutation in the paternal copy will have no effect.

Exome sequencing in another patient of Caucasian origin revealed a de novo mutation 770G>C (G236R). This patient was born with congenital hypotonia, cleft palate, and episodes of spontaneous clonus upon awakening. He had a normal MRI and EEG, delayed development, severe feeding problems requiring a G-tube, decreased tearing on crying, and neonatal hyperinsulinism, which resolved on diazoxide. Dysmorphic features included upswept anterior and posterior hair pattern, dolichocephaly with bitemporal narrowing, short philtrum, tented upper lip, V-shaped cleft palate, prominent maxilo- alveolar frenulum, small mandible, medially-flared eyebrows, tapered fingers with prominent fingertip pads, bridged transverse palmar flexion creases, and prominent creases above both heels. Features in the Arab Israeli family included intellectual disability of variable severity, hyperactivity, severe feeding difficulties in infancy with dysphagia of liquids and dysphonia with a muffled voice into early adulthood, generalized hypotonia, weakness of proximal muscles, elongated face with narrow bitemporal diameter, and reduced facial movements. Dysmorphic features of eyes, ears, and face were noted, with a narrow high-arched palate with full or submucous cleft and dysphonic speech. This newly- recognized cleft palate syndrome with associated congenital hypotonia may be a treatable disorder.

The 770G>C (G236R) mutation fully abolishes the channel's currents when functioning either as a homodimer or a heterodimer, but the reduced outward current through mutated TASK3-G236R channels can be overcome in a mouse model by use of the nonsteroidal anti-inflammatory drugs flufenamic acid (FFA) and mefenamic acid (MFA). We will report on the early results of an off-label clinical trial with MFA, a non-steroidal anti-inflammatory drug used to treat menstrual pain (Rx Ponsel), and discuss the natural history of this disorder in untreated patients.

In Utero Gene Therapy Rescues Microcephaly Caused By PQBP11—Hypofunction In Neural Stem Progenitor Cells

C.E. Schwartz¹, H. Ito², H. Shiwaku², C. Yoshida², H. Homma², H. Luo², X. Chen², K. Fujita², L. Musante³, U. Fischer³, S.G. Frints⁴, C. Romano⁵, Y. Ikeuchi⁶, T. Shimamura⁷, S. Imoto⁷, S. Miyano⁷, S.I. Muramatsu⁸, T. Kawauchi⁹, M. Hoshino¹⁰, M. Sudol¹¹, A. Arumughan¹², E.E. Wanker¹², T. Rich¹³, F. Matsuzaki¹⁴, A. Bonni⁶, V.M. Kalscheuer³, H. Okazawa²

¹JC Self Research Institute of Human Genetics, Greenwood Genetic Center, Greenwood, South Carolina

²Department of Neuropathology, Medical Research Institute and Center for Brain Integration Research, Tokyo Medical and Dental University, Tokyo, Japan

³Department of Human Molecular Genetics, Max-Planck Institute for Molecular Genetics, Berlin-Dahlem, Germany

⁴[1] Department of Clinical Genetics, University Hospital azM Maastricht, Maastricht, The Netherlands; [2] School for Oncology and Developmental Biology, GROW, Maastricht University, Maastricht, The Netherlands

⁵Unità Operativa Complessa di Pediatria e Genetica Medica, IRCCS Associazione Oasi Maria Santissima, Troina (Enna), Italy

⁶[1] Department of Anatomy and Neurobiology, Washington University School of Medicine, St Louis, Missouri; [2] Department of Neurobiology, Harvard Medical School, Boston, Massachusetts

⁷Human Genome Center, Institute of Medical Science, The University of Tokyo, Tokyo, Japan

⁸Department of Neurology, Jichi Medical University, Tochigi, Japan

⁹Department of Anatomy, Keio University School of Medicine, Tokyo, Japan

¹⁰Department of Biochemistry and Cellular Biology, National Center for Neurology and Psychiatry, Tokyo, Japan

¹¹Laboratory of Signal Transduction and Proteomic Profiling, Weis Center for Research, Geisinger Clinic, Danville, Pennsylvania

¹²Department of Neurogenetics, Max-Delbrück Center for Molecular Medicine, Berlin- Buch, Germany

¹³Institute of Infection, Immunity and Inflammation, University of Glasgow, Glasgow, UK

¹⁴Laboratory for Cell Asymmetry, Center for Developmental Biology, RIKEN, Chuo-ku, Kobe, Japan

Human mutations in PQBP1, a molecule involved in transcription and splicing, result in a reduced but architecturally normal brain. Examination of a conditional Pqbp1- knockout (cKO) mouse with microcephaly failed to reveal either abnormal centrosomes or mitotic spindles, increased neurogenesis from the neural stem progenitor cell (NSPC) pool or increased cell death in vivo. Instead, we observed an increase in the length of the cell cycle, particularly for the M phase in NSPCs. Corresponding to the developmental expression of Pqbp1, the stem cell pool in vivo was decreased at E10 and remained at a low level during neurogenesis (E15) in Pqbp1-cKO mice. The expression profiles of NSPCs derived from the cKO mouse revealed significant changes in gene groups that control the M phase, including anaphase-promoting complex genes, via aberrant transcription and RNA splicing. Exogenous Apc4, a hub protein in the network of affected genes, recovered the cell cycle, proliferation, and cell phenotypes of NSPCs caused by Pqbp1-cKO. These data reveal a mechanism of brain size control based on the simple reduction of the NSPC pool by cell cycle time elongation. Finally, we demonstrated that in utero gene therapy for Pqbp1-cKO mice by intraperitoneal injection of the PQBP1-AAV vector at E10 successfully rescued microcephaly with preserved cortical structures and improved behavioral abnormalities in Pqbp1-cKO mice, opening a new strategy for treating this intractable developmental disorder.

PRENATAL PRESENTATION OF MALFORMATION SYNDROMES I

Prenatal Presentation of Genetic Conditions: Insights From the Amniotic Fluid Supernatant Transcriptome

Diana Bianchi, MD

Tufts Medical Center, Boston, Massachusetts

The amniotic fluid supernatant (AFS) transcriptome can be studied using the cell-free RNA that floats in the amniotic fluid supernatant¹. In prior work we have characterized the amniotic fluid transcriptome of euploid fetuses in the second trimester and at term. The AFS transcriptome gives detailed, real-time, in vivo information about the developing fetus. Using samples that are matched for gestational age and fetal sex we have compared euploid to aneuploid fetuses. The aneuploidies studied include trisomies 21 and 18 as well as Turner syndrome. We have also studied structural abnormalities that include myelomeningocoele and twin-to-twin transfusion. This discovery-driven approach has led to new knowledge regarding fetal phenotypes. Our work has shown that each condition has a unique but consistent pattern of gene expression. Further knowledge regarding specific developmental pathway perturbations can lead to novel therapeutic approaches.

As an example, we showed that oxidative stress was a significant pathway difference in the fetuses affected with trisomy 21. We have uploaded genes that were differentially-regulated between fetuses with and without trisomy 21 into the Connectivity Map (www.broadinstitute.org/cmap) to identify candidate FDA-approved therapeutic molecules. Drugs that have a high efficacy in negating oxidative stress and show low toxicity have been selected for further studies in a mouse model. We are using the Ts1Cje mouse model of Down syndrome because affected males are fertile, yet cognition is significantly impaired. We can therefore use wild-type females for the treatment experiments, ensuring both a normal intrauterine environment and normal postnatal nurturing behavior. Treated and untreated affected pups and littermate controls are then evaluated using a variety of brain studies that include analyses of gene expression, histology, cellular proliferation and migration, and neurobehavior. In parallel, we are also preparing for a human clinical trial by analyzing fetal brain growth using quantitative fetal magnetic resonance imaging (MRIs).

To date, we have identified significant phenotypic differences in Ts1Cje embryos, neonates, and adults as endpoints to evaluate the effects of therapy. Following treatment, we have also shown encouraging, statistically-significant improvement in some neurobehavioral tests in adult mice. Our results suggest that (1) there are specific, but different genome-wide dysregulation patterns in all aneuploidies, (2) there are significant abnormalities in human fetuses with Down syndrome, and (3) that preliminary data in a mouse model suggests that treating genetic conditions prenatally may have therapeutic effects.

¹Zwemer LM, Bianchi DW. The amniotic fluid transcriptome as a guide to understanding fetal disease. Cold Spring Harb Perspect Med 2015; 5(4). pii:a023101. doi:10.1101/cshperspect.a023101

The Tragedy of the Failure to Diagnose Arthrogryposis Prenatally

J.G. Hall, OC, MD, FRSC, FCAHS

Departments of Medical Genetics and Pediatrics, UBC, Vancouver, BC Canada

In 2012¹, we did a retrospective study to determine how often arthrogryposis was diagnosed prenatally when prenatal ultrasound studies (US) had been done. Assuming that US studies of the fetus became a routine part of prenatal care in 1990, we collected 107 cases of 4-limb Amyoplasia (the most common form of arthrogryposis) that had had prenatal US studies between 1990 and 2010. To our surprise, only 26.2% were recognized prior to birth, in spite of having complicated pregnancies (70%), maternal recognition of decreased fetal movement (60%), severe longstanding 4-limb contractures in utero, and multiple US studies. At the time, it was suggested that prenatal detection of arthrogryposis has probably improving over time, although the data did not suggest that in the 2012 study.

Consequently, we have tabulated detection rates on a different set of 70 unselected cases of 4-limb Amyoplasia born between 2005 and 2015. They were ascertained through correspondence, attendance at the Seattle Arthrogryposis Clinic, or attendance at the national AMC Support Meeting. This more recent group indicates that still less than 30% of cases (20 of the 70) were diagnosed prenatally in spite of repeated US studies.

It should be possible to diagnose Amyoplasia in the early second trimester since the onset of contractures has occurred by 12 weeks. However, it will only be possible when attention is given to limb movement by the ultrasonographer. Examination of limb movement is not yet part of the routine prenatal US procedures.

Among the 20 cases which were diagnosed, only clubfeet were recognized in five fetuses in spite of severe 4-limb involvement, while one other case had only the clenched fists recognized. Among other cases, two pregnancies were twins, one had gastroschisis, and six had oligohydramnios, therefore these probably receiving a more comprehensive examination. Thus, there did not appear to be a trend toward more accurate diagnosis of arthrogryposis (represented by 4-limb Amyoplasia) over this most recent ten years.

Only half of the 20 recognized cases were diagnosed prior to 24 weeks. Because of the failure to diagnose arthrogryposis (which occurs in 1 per 3,000 pregnancies) in a timely fashion, families and physicians were deprived of the opportunity for choice concerning continuing the pregnancy and best care for mother and child. One third of the fetuses with arthrogryposis will end up with moderate to severe intellectual disability, many with microdeletions or duplications. Without recognizing the presence of arthrogryposis, the families are denied the possibility of appropriate diagnostic studies such as chromosomal microarray or exome studies, or better definition of the arthrogrypotic features by MRI studies to define muscle, lung and CNS development.

Possible in utero therapy such as increasing fetal movement, maturation of fetal lung and gut, or early delivery could also not be undertaken.

Families deserve accurate prenatal information about fetal contractures on which to base decisions and planning.

¹Filges I, Hall JG. Failure to identify antenatal multiple congenital contractures and fetal akinesia—proposal of guidelines to improve diagnosis. Prenat Diag 33:61–74, 2013.

The Prenatal Presentation of IMAGe Syndrome

J. Kaplan, MD¹, M. Bober, MD, PhD², S. Vergano, MD³, J. Bennett, MD, PhD⁴, B. Solomon, MD⁵, M. Deardorff, MD, PhD⁶

¹University of Mississippi Medical Center, Jackson, Mississippi

²Nemours/A.I. DuPont Hospital for Children, Wilmington, Delaware

³Children's Hospital of the King's Daughters, Norfolk, Virginia

⁴Seattle Children's Hospital, University of Washington, Seattle, Washington

⁵Inova Translational Medical Institute, Falls Church, Virginia

⁶The Children's Hospital of Philadelphia, University of Pennsylvania, Philadelphia, Pennsylvania

IMAGe syndrome is characterized by IUGR, Metaphyseal dysplasia, Adrenal insufficiency, and Genital anomalies. Due to the elucidation of a molecular etiology in 2012, gain-of-function mutations in CDKN1C, IMAGe syndrome is becoming more widely recognized in the clinical genetics community. Typically, these patients are diagnosed after birth, often not until adrenal crisis occurs. IUGR is a strong prenatal clue to the diagnosis and appears to be universal in patients with IMAGe. We reviewed the 28 previously reported cases in addition to 10 cases that have not yet been reported to determine what additional prenatal findings can be seen in IMAGe syndrome.

Five cases (two previously reported and three new) had prenatal findings other than IUGR and oligohydramnios. Three patients had short long bones seen on anatomy scan, and two patients had increased nuchal translucency. Only two of the five cases reported maternal serum markers, which were abnormal in both cases: one pregnancy had elevated unconjugated estriol, and the other had elevated hCG and low PAPP-A with a 1:72 risk of Down syndrome.

Although there have been very few cases with reported prenatal findings, it is clear that there may be prenatal clues to a diagnosis of IMAGe syndrome. Increased nuchal translucency and shortened long bones along with the finding of IUGR (typically severe) should alert the clinician to the possibility of IMAGe syndrome. Abnormalities in maternal serum markers may also raise suspicion, though more data is needed to determine if there is a clear pattern that could suggest IMAGe. Based on the findings seen prenatally thus far, IMAGe syndrome should be in the differential diagnosis when considering Down syndrome and/or skeletal dysplasias.

Prenatal Findings in Beckwith–Wiedemann Syndrome

Jennifer M. Kalish, Jessica Ebrahimzadeh, Juliana Purrazzella, Sara Reichert, Jennifer Richards- Yutz, Amanda Treece, Nahla Khalek, Eileen McKay, Arupa Ganguly, Matthew A. Deardorff

Beckwith–Wiedemann syndrome (BWS) is a mosaic disorder due to genetic and epigenetic alterations on chromosome 11p. The heterogenous phenotype ranges from isolated hemihypertrophy to classic features including macrosomia, macroglossia, omphalocele, and organomegaly. We investigated which features led to prenatal recognition of BWS in a clinical setting at the Children's Hospital of Philadelphia Special Delivery Unit and in a laboratory setting by the Genetic Diagnostic Laboratory at the University of Pennsylvania (GDL). We further correlated those findings with the molecular causes of BWS confirmed by prenatal and postnatal testing. Among the testing performed by the GDL between 2013 and 2015, 23 prenatal cases were evaluated for suspected BWS. Of those cases, 11 were positive for BWS and their molecular diagnoses were: loss of methylation at imprinting control region 2 (LOM IC2) (6/11), paternal uniparental isodisomy (pUPD11) (1/11), gain of methylation at imprinting control region 1 (GOM IC1) (3/11), and CDKN1C mutation (1/11). The most common indications for prenatal testing were omphalocele and renal anomalies.

We found that patients with LOM IC2 were most likely to be diagnosed prenatally because of an increased incidence of features recognized on ultrasound and MRI. Placentomegaly, omphalocele, and renal anomalies were the most commonly recognized features for those infants diagnosed in an ongoing pregnancy. Concern for macroglossia was raised in several cases but did not result in BWS testing without additional features being present. Elevated maternal serum alpha-fetoprotein (AFP) levels were noted in a few cases and retrospectively demonstrated in several additional cases. Those with BWS due to GOM IC1 and pUPD11 were less likely to have features identified prenatally. In at least one case that presented with an omphalocele, prenatal BWS methylation testing was normal, and postnatal testing identified a causative CDKN1C mutation. At birth, a subset of placentas from patients with LOM IC2 were examined and a mosaic range of pathological findings were found, including the well-described finding of mesenchymal dysplasia. In addition, several cases of extravillous trophoblast proliferation were identified, which has rarely been reported in BWS. Regardless of molecular cause, cases of BWS without placentomegaly, omphalocele, or renal anomalies are less likely to be noted prenatally. Prenatal diagnosis is still important for these cases given the risk of hypoglycemia at birth (50%), the risk of hyperinsulinism (5%) and the increased risk for tumors. Overall, we recommend BWS testing including both methylation and CDKN1C sequencing when prenatal findings of placentomegaly, omphalocele, and renal anomalies are noted. In cases of a mildly elevated AFP on quadruple screening, further assessment for BWS is also recommended. Additionally, given the slightly increased incidence of BWS following IVF, (primarily due to IC2 LOM), careful attention to features of BWS on prenatal ultrasounds in these pregnancies is recommended.

Identification of Embryonic Lethal Genes in Humans Using Mendelian Approaches

Fowzan Alkuraya^1,2

¹Department of Genetics, King Faisal Specialist Hospital and Research Center, Riyadh, Saudi Arabia

²Department of Anatomy and Cell Biology, College of Medicine, Alfaisal University, Riyadh, Saudi Arabia

Identifying genetic variants that lead to discernible phenotypes is the core of Mendelian genetics. An approach that considers embryonic lethality as a bona fide Mendelian phenotype has the potential to reveal novel genetic causes which will further our understanding of early human development at a molecular level. Building on the successful approach of using Mendelian genetics of birth defects to understand normal morphogenesis of individual organs, we reckoned that Mendelian forms of embryonic lethality offer a window into the essential genetic components of early organismal development in humans. Specifically, consanguineous families in which embryonic lethality segregates as a recessive Mendelian phenotype offer a unique opportunity for high throughput novel gene discovery. We have studied 24 such families using autozygosity mapping and whole-exome sequencing. In addition to revealing mutations in genes previously linked to embryonic lethality in severe cases, our approach revealed eight novel candidate genes. A founder mutation in one of these novel genes accounted for embryonic lethality in three of the study families. This novel gene has been linked to vascular permeability and we were able to identify another mutation in a family with less severe phenotype consisting of hydrops fetalis and persistent postnatal edema.

We also studied four consanguineous families that segregate the earliest known embryonic lethality phenotype i.e., inability of the zygote to undergo normal division resulting in disintegration and death. We mapped this phenotype in two of these families to a single mutation in a novel gene. Consistent with the published phenotype of the mouse mutants, embryos from female patients who are homozygous for the mutation fail to undergo early cleavage with resulting sterility. We show that our mutation abrogates phosphorylation of the encoded protein, a critical step in the initiation of preimplantation mammalian embryonic development.

Our study represents an important step towards the systematic analysis of “embryonic lethal genes” and highlight the power of new genomic tools to tackle this important yet difficult-to-study phenotype in humans.

PRENATAL PRESENTATION OF MALFORMATION SYNDROMES II

A Novel Hypothesis for the Etiology of Birth Defects

Jones KL, Johnson DL, Xu R, Luo Y, Adam MLP, Vaux KK, Robinson LK, Clericuzio CL, Chambers CD

Department of Pediatrics, Division of Dysmorphology/Teratology; University of California, San Diego, School of Medicine, La Jolla, California

In 1978, Josef Warkany coined the term terathanasia to define the natural screening processes that reduce the incidence of congenital malformations. Juriloff and Fraser in 1977 described the concept of supportive terathanasia. They showed that differential mortality, or terathanasia, occurs following thyroxine administration in pregnant mice. Without thyroxine treatment, mice have a birth prevalence of cleft palate of 25%. However, by treating the mice with thyroxine, 50% of cleft palate embryos and 3% of normal embryos were eliminated, resulting in a rate of cleft palate detectable at birth of only 12%. It is possible that prenatal exposure to a drug or other agent could have the opposite effect, i.e., a drug that is protective for or reduces the normal rate of SAB could result in continued viability of pregnancies that would have otherwise been lost. In this scenario, a decreased incidence of SAB could lead to an increased prevalence of structural defects in live born infants (teratovita). Evidence for this, however, has not previously been described in humans.

We have prospectively followed 493 pregnant women with rheumatoid arthritis, psoriasis, psoriatic arthritis or ankylosing spondylitis all of whom were enrolled in the study on average at the end of the first trimester and who did (n=335) or did not (n=158) have prenatal exposure to etanercept, a TNF inhibitor, in the first trimester. In the etanercept-exposed group 33/335 (9.9%) of the pregnancies resulted in at least one fetus or infant with a major malformation as opposed to 7/158 (4.4%) in the group with no exposure to etanercept (Adjusted Odds Ratio 2.37, 95% Confidence Interval [CI] 1.02,5.52). There was no specific pattern of major structural defects, and biological plausibility of a direct teratogenic effect is not supported by evidence that TNF inhibitors cross the placenta in the first trimester in any appreciable amounts. In contrast, using survival analysis accounting for varying gestational timing of enrollment of the pregnant women, we noted a non-significant but reduced risk of SAB in the etanercept-exposed group compared to the unexposed group (Adjusted Hazard Ratio 0.47, 95% CI 0.20, 1.12), based on 27 SAB events that occurred sometime from the time of enrollment to the end of the 19th week of gestation. In this study, we could only observe SABs that occurred after enrollment which was on average at the end of the 1^st trimester. However, assuming that any protective effect of etanercept for SAB would also apply to earlier SABs that occurred in women who we were not able to observe, we hypothesize that the increased risk of major malformations representing no consistent pattern in the etanercept-exposed group could be the result of a decreased incidence of SAB among malformed fetuses. Although the likelihood of proving this hypothesis in human pregnancy is doubtful, these data raise the possibility that differential embryonic loss could occur following etanercept exposure in early human pregnancy. This represents a novel mechanism for the etiology of birth defects and deserves consideration in further studies.

Prenatal Diagnosis of Encephalocele—A Single Center Experience

A. Ananth, S. Hintz, Y. Blumenfeld, J. Hahn, L. Hudgins

Stanford University School of Medicine, Lucile Packard Children's Hospital Stanford, Stanford, California

While progress has been made in the treatment of neural tube defects, a significant proportion of patients with congenital encephaloceles have neurologic disability. With the advances in prenatal imaging techniques the diagnosis of encephalocele is being made more frequently prior to birth. Given the limited available data in the published literature, counseling expectant families in the prenatal period regarding neonatal outcomes remains a challenge. The purpose of our study was to identify prenatal factors that may be associated with adverse outcomes in cases of encephalocele.

The Stanford Center for Fetal and Maternal Health database is a prospectively collected database which was established in 2009 to improve the quality of care and outcomes for expectant mothers and their children. This database was queried to identify all prenatal diagnoses of encephalocele from January 2009 to February 2015. Clinical information, including maternal characteristics, prenatal and postnatal imaging, genetic testing, and fetal outcomes was obtained through chart review.

We identified 17 cases of encephalocele. Genetic testing was performed in 15 cases and consisted of karyotype alone in six cases, while nine cases also underwent array comparative genomic hybridization. Whole exome sequencing was performed in one patient after birth. Genetic testing was negative in all cases. Fetal MRI was performed in 13 pregnancies, and demonstrated additional brain abnormalities in 76.9% of cases (dysmorphic hindbrain in 53.8%, holoprosencephaly in 15.3%, and agenesis of the corpus callosum in 30.8%).

Pregnancy outcomes were available in 15 cases. Seven pregnancies (46.7%) were electively terminated. Of the eight liveborn infants, seven were born at term. Neonatal mortality for these patients was 37.5%. Three of the five surviving infants developed hydrocephalus and required ventriculoperitoneal shunt placement prior to 1 year of age. Follow-up information was available for four patients. At most recent follow up, 2 patients have age appropriate developmental skills and two patients have global developmental delays.

Our study did not determine clear markers for prognosis in prenatally diagnoseds encephalocele. However, a large percentage of cases were found to have additional brain abnormalities detected by fetal MRI, highlighting the importance of this complementary prenatal imaging modality. There remains a high rate of complications in liveborn neonates with this condition. Larger studies are needed to identify prognostic features that can be used in the prenatal period to guide counseling for these families.

Prenatally Diagnosed Aqueductal Stenosis (triventricular Dilation) and Postnatal/autopsy Findings. Report on 100 Cases

Rebekah Jobling^1,2, Noa Amir², Ants Toi³, Patrick Shannon⁴, Karen Chong^1,2, David Chitayat^1,2

¹Division of Clinical and Metabolic Genetics; Mount Sinai Hospital, Department of Obstetrics and Gynecology, The Hospital for Sickkids, Department of Pediatrics, Ontario, Canada

²The Prenatal Diagnosis and Medical Genetics Program, Ontario, Canada

³Department of Diagnostic imaging and Department of Laboratory Medicine and Pathobiology; The University of Toronto, Toronto, Ontario, Canada

Hydrocephalus is etiologically a heterogenous condition. Up to 66% of the cases of childhood hydrocephalus are the result of stenosis of the aqueduct of Sylvius (Aqueductal stenosis)(AS). Most AS cases present at an early gestation with fetal ultrasound/MRI finding of dilation of the lateral and 3rd cerebral ventricles and the differential diagnosis includes infection, hemorrhage, chromosome abnormalities and a variety of single gene disorders including Dandy-Walker malformation and L1CAM syndrome. In many cases the underlying cause cannot be delineated making genetic counseling difficult.

We report our experience with 100 cases presented prenatally with triventricular dilation on fetal ultrasound/MRI. Of the 100 pregnancies 64 were terminated and 36 were continued. There were 12 cases which ended in still birth. In 49 of these cases pathological examination confirmed the diagnosis and in the rest the aqueduct was not found occluded or there was no pathological examination. Of the confirmed cases of AS 15 were isolated and 34 were associated with other congenital anomalies. In 5 cases intra-cerebral hemorrhage was identified and none of these were the result of maternal alloimmune antiplatelet antibodies. In one case fetal infection was identified as a cause. L1CAM gene mutation was found in four cases and in all of these cases the mother was a carrier. Chromosome abnormalities were present in 10 cases (3 cases of trisomy 21, one case of trisomy 18, one case of 45, X, one case of mosaic trisomy 22, one case of an unbalanced translocation of chromosomes 6 and 7, a duplication at 2q11.2, a duplication of 15p13.33 and a 110 kb deletion at 2p16.3). In most cases an underlying etiology could not be identified.

This large cohort of patients with AS provides insight into the maternal characteristics, presentation, associated anomalies, etiology and outcome of this rare and anomaly.

Direct Measurments of Somatic Mutation Rate During Human Fetal Development

J. Bennett, MD, PhD¹, M. Dorschner, PhD², C. Hale, PhD², D. Hanna, PhD², W.B. Dobyns, MD¹, I. Glass, MD¹

¹Department of Pediatrics, University of Washington and Seattle Children's Research Institute, Seattle, Washington

²Department of Pathology, University of Washington and Seattle Children's Research Institute, Seattle, Washington

Advances in DNA sequencing technologies have revolutionized molecular diagnosis of children with birth defects. However, almost all genetic testing in children is performed from blood-derived DNA, despite the knowledge that the genomes of hematopoietic cells are not identical to every other cell in the body. Because errors in DNA replication occur with every cell division, and the number of cell divisions that occur during the first trimester of human development is enormous (∼20 billion), post-zygotic “somatic” mutations are probably common, and definitely under-ascertained. DNA analysis of blastomeres from IVF embryos, tissues from autopsy specimens, fibroblasts from adult skin biopsies and blood from monozygotic twins have provided estimates, but the baseline rate of somatic mutation during human development has never been measured directly. This is a critical question, as it will provide estimates of the amount of human genetic variation that is missed when only the blood is sequenced.

Using a unique biorepository of human fetal tissue specimens, we have sequenced DNA isolated from multiple different tissues (brain, heart, liver, etc.) from the same human fetus to identify discordant genotypes, which represent somatic mutation. The University of Washington Birth Defect Research Laboratory (BDRL) contains tissue samples from > 200 human fetuses recruited from women undergoing elective pregnancy termination in the 1^st and 2^nd trimesters. Since major birth defects arise in the first trimester, we focused on the earliest gestation fetuses in the repository. Of the 28 fetuses between 12 and 13 weeks gestation, five had at least 8 distinct tissues, including placenta, for DNA isolation. In order to focus on the mutation rate among genes most likely to cause birth defects, and to limit the size of the sequencing target to obtain the deep coverage needed for calling somatic variants, we restricted our capture target to include only genes that have been previously reported to cause any human phenotype (∼4,800 genes). We sequenced this target in 8 different tissues per fetus in three fetuses to an average depth of 300X. The rates and tissue distribution of somatic single nucleotide variation (SNV) among these three fetuses will be presented. These results represent the first attempt to directly measure the somatic SNV mutation rate during human development.

Multiple Malformation Syndromes Associated With Very Low/Undetectable Maternal Serum uE3

Laurie H. Seaver, MD¹

¹Kapi‘olani Medical Specialists and Department of Pediatrics, University of Hawai‘i John A. Burns School of Medicine, Honolulu, Hawaii

The association of steroid sulfatase deficiency and Smith-Lemli-Opitz syndrome with very low or undetectable maternal serum unconjugated estriol (uE3) is well recognized. Androgens produced by the fetus and placenta are the source of maternal serum uE3. The differential diagnosis of low maternal uE3 can include gestational dating error, threatened or missed abortion, chromosome abnormalities, primary endocrine disorders and other multiple malformation syndromes, some of which may not be reliably diagnosed based on ultrasound abnormalities. The purpose of this report is to bring attention to maternal serum uE3 as a presenting feature of multiple malformation syndromes associated with adrenal insufficiency.

Patients 1 and 2 are sisters born after term pregnancies complicated by low maternal serum uE3 (0.11 and 0.13 MoM) and intrauterine growth restriction (IUGR). Fetal karyotype was 46,XX in each case. At birth, normal female genitalia, mildly short long bones, hyperpigmentation and hypotonia were noted. Severe neonatal salt losing syndrome was recognized in Patient 1 on DOL 10 and treated with hydrocortisone and fludrocortisone. She has poor postnatal growth. Patient 2 was treated with steroid replacement by DOL 2. She also has significant postnatal growth deficiency with more severe hypotonia complicated by infantile scoliosis. Skeletal surveys in both girls are normal, except for scoliosis in Patient 2. Cognitive development is normal for both. The diagnosis of IMAGe syndrome (Intrauterine growth restriction, Metaphyseal dysplasia, Adrenal insufficiency and Genital anomaly) was made and confirmed with maternally inherited CDKN1C mutation.

Patient 3 is a male born at term after a pregnancy complicated by maternal uE3 of 0.04 MoM. Karyotype was 46,XY; STS FISH and 7-dehydrocholesterol were normal. Ultrasound showed bilateral clubfeet. He was SGA with craniofacial, skeletal features and genital hypoplasia which suggested the diagnosis of Antley-Bixler syndrome 2 (due to cytochrome P450 oxidoreductase, POR, deficiency). Cortisol and androgen levels were low, ACTH mildly elevated and he was treated with hydrocortisone. Subsequent growth is excellent. The diagnosis was confirmed by molecular testing.

CDKNIC is previously known to be associated with control of fetal and postnatal growth, but the mechanism by which mutation in CDKNIC causes malformations is unknown. The function of POR in steroidogenesis results in steroid profile suggesting combined 21-hydroxylase and 17-hydroxylase deficiency, and so can present as both 46,XX and 46,XY disorders of sexual development. Its role in skeletal development is related to cholesterol and retinoic acid metabolism; a phenocopy of ABS is associated with fetal exposure to fluconazole which disrupts cholesterol biosynthesis.

Early recognition of malformation syndromes associated with adrenal insufficiency is critical to avoid further complications including death.

Predictive Prenatal Ultrasound Findings for Mild and Severe Osteogenesis Imperfecta

S. Parekh, BA¹, Y. Blumenfeld, MD^1,2, J. Taylor, MS^1,2, S. Hintz MD, MS^1,2, L. Hudgins, MD^1,2

¹Stanford University School of Medicine, Stanford, California

²Lucile Packard Children's Hospital Stanford, Stanford, California

Osteogenesis imperfecta (OI) is a relatively common connective tissue disorder that can present in the prenatal period. Classically described fetal ultrasound findings include short, deformed long bones and undermineralized skull. Distinguishing between the mild and severe forms, however, is often challenging which can lead to uncertain prognosis. We performed a retrospective comparative study to determine whether there are specific fetal ultrasound findings that can distinguish between mild and severe OI.

Following IRB approval, we queried our prenatal and postnatal databases to identify all cases during the period from November 2004 through September 2013 in which a prenatal diagnosis of OI was suspected based on fetal ultrasound findings. Postnatal outcome was abstracted from medical records including clinical and genetic diagnoses.

Of the 21 patients prenatally suspected of having OI, 5 had mild OI and 6 had severe OI. Nine had other conditions including thanatophoric dysplasia, kyphomelic dysplasia, and cleidocranial dysostosis; one did not have any postnatal abnormalities.

The classically used prenatal ultrasound findings of short femurs, fractured femurs and skull osteopenia were first reviewed. Short femurs were reported in 40% of cases of mild OI, 100% of cases of severe OI, and 89% of cases with other conditions. Fractured femurs were reported in 20% of cases of mild OI, 83% of cases of severe OI, and 11% of cases with other conditions. Skull osteopenia was reported in no cases of mild OI, 67% of cases of severe OI, and no cases with other conditions.

The most distinguishing ultrasound finding, however, was short upper extremity long bones reported in no cases of mild OI, 100% of cases of severe OI, and 89% with other conditions. Thus, the ultrasound finding of short upper extremity long bones is statistically sensitive and specific enough to differentiate mild OI from severe OI and other conditions (P=0.002). Additionally, the mean femur length to abdominal circumference (FL/AC) ratio in mild OI (0.1922) was significantly greater than in cases with severe OI (0.1213) with a p-value of 0.001.

In summary, when applied to our cohort, classically used ultrasound findings were not sensitive or specific enough to differentiate between mild OI, severe OI, or other conditions. However, shortening of the upper extremity long bones tracked with 100% sensitivity and specificity to severe OI. Additionally, mild OI cases exhibited statistically significantly higher FL/AC ratios than severe OI cases. Using these two prenatal ultrasound findings may allow for differentiation between mild and severe OI and contribute to a more accurate prognosis to inform pregnancy and delivery management.

When Folic Acid Fails

Roger E. Stevenson, Caleb Bupp

Greenwood Genetic Center, Greenwood, South Carolina

Dick Smithells (Leeds, England) first reported the protective effect of folic acid against neural tube defects in 1980. Over a decade intervened before this effect was confirmed by the UK's Medical Research Council (1991) and Czeizel and Dudas's study in Hungary (1992).

The South Carolina Neural Tube Defect Prevention Project began in 1992 promoting the use of folic acid supplements to prevent the occurrence and recurrence of NTDs. During the first 20 years of this program 1046 NTDs (1037 occurrent and 9 recurrent) were identified. Folic acid use in the periconceptional period was determined in 611 cases through interviews with the mothers. Among these, 165 (27%) of the mothers reported use of folic acid supplements in the periconceptional period and these NTDs were considered folic acid failures. Of these failures 163 represented occurrent cases, 2 represented recurrent cases. There was an enrichment for spina bifida with other malformations among the folic acid failures (40.4% compared to 25.2% of isolated spina bifida). This enrichment for associated anomalies was not noted for anencephaly and encephalocele. The 43 folic acid failure NTDs with other anomalies included trisomy 18 (9), other chromosome aberrations (7), amniotic bands (3), OEIS complex (4), holoprosencephaly (2), facial clefting (5), cardiac defects (4), and other anomalies (9).

Folic Acid Failures by Type of NTDs

The majority (74%) of NTDs that occurred even though the mothers reported using folic acid in the periconceptional period were “garden variety” isolated neural tube defects. The reasons for the folic acid failures are not readily apparent but comprise a substantial cohort for further investigation into other risk factors and prevention strategies.

Smithells et al. Lancet 1:339, 1980. Medical Research Council. Lancet 338:131, 1991. Czeizel and Dudas. New Eng J Med 327:1832, 1992.

DISORDERS OF SEX DEVELOPMENT I

Disorders of Sex Development: A World of Uncertainties

Eric Vilain, MD, PhD

UCLA Department of Human Genetics

Disorders of sex development (DSD) are congenital conditions in which the development of chromosomal, gonadal, or anatomical sex is atypical.

DSD are phenotypically heterogeneous, ranging from minor genital malformations (hypospadias, cryptorchidism, hypertrophy of the clitoris) to genital ambiguity. In the aggregate, DSD have an estimated incidence of about 1%. They can result in severe consequences for behavioral health, fertility, cancer risk and quality of life. For families, the birth of a child with a DSD, and the accompanying uncertainty about future psychological and sexual development, is believed to be extraordinarily stressful. Recently, the debate over clinical management of DSD, in particular gender assignment and genital surgery, has intensified; yet the scientific data on patient outcomes have remained very incomplete. Major obstacles to optimal clinical management of DSD include gaps in understanding of pathophysiology, impeding precise diagnostic categorization, along with the absence of prospective longitudinal studies of health and quality of life outcomes.

Such conditions have historically been difficult to diagnose, especially at the genetic level. For example, for cases of 46,XY gonadal dysgenesis, once variants in SRY and NR5A1 have been ruled out, there are few other single gene tests available. We used exome sequencing followed by analysis with a list of all known human DSD-associated genes to investigate the underlying genetic etiology of 46,XY DSD patients who had not previously received a genetic diagnosis. We were able to identify a likely genetic diagnosis in more than a third of cases, including 22.5% with a pathogenic finding and an additional 12.5% with likely pathogenic findings. In addition, 15% had variants of uncertain clinical significance (VUS) that may be reclassified as literature evolves. Exome sequencing allowed a remarkable and unprecedented level of genetic diagnostic success in this cohort, especially considering that, for most patients, all other endocrine and genetic testing had been exhausted. Early identification of the genetic cause of a DSD will in many cases streamline and direct the clinical management of the patient, with more focused endocrine and imaging studies and better informed surgical decisions.

Can the External Masculinization Score Predict the Success of Genetic Testing in 46,XY DSD?

M.P. Adam¹, M. Shnorhavorian², L. Ramsdell³, P. Fechner¹, R. Su⁴

¹Department of Pediatrics, University of Washington School of Medicine, Seattle, Washington

²Department of Urology, University of Washington School of Medicine, Seattle, Washington

³Seattle Children's Hospital, Seattle, Washington

⁴Department of Urology, University of Wisconsin School of Medicine and Public Health, Madison, Wisconsin

Genetic testing is currently judiciously applied to individuals with Disorders of Sex Development (DSD), thus it is crucial to identify those most likely to benefit from such testing. The External Masculinization Score (EMS) (range 0–12) is a validated scoring system used to evaluate the degree of masculinization in an effort to standardize and succinctly communicate the phenotype of individuals with ambiguous genitalia. An EMS <7 is considered ambiguous. We hypothesized that the EMS is inversely associated with the likelihood of finding a pathogenic variant causative of DSD.

Patients with 46,XY DSD from a single institution evaluated from 1994 to 2014 were included. Results of advanced cytogenetic testing (ie chromosomal microarray) and gene sequencing tests were recorded. An EMS was assigned to each patient according to the team's initial external genitalia physical examination.

67/140 (48%) patients with 46,XY DSD were evaluated during the study period and underwent genetic testing beyond initial karyotype; 31% (21/67) had a genetic diagnosis made by gene sequencing or chromosomal array. The average EMS of those with an identified genetic cause was 3.1 (95% CI, 1.9–4.3; median, 3) compared to 5.6 (95% CI, 4.8–6.4; median, 6) in those for whom no confirmed genetic cause was identified (P < 0.02). The most common diagnoses were complete or partial androgen insensitivity syndrome and 5-alpha- reductase deficiency. Other less common diagnoses included NR5A1-related sex reversal, small chromosome abnormalities known to cause undervirilization in 46,XY individuals, 17-beta hydroxysteroid dehydrogenase deficiency, Mowat-Wilson syndrome, and Prader-Willi syndrome.

46,XY DSD with an identified genetic cause had a phenotype with a significantly lower EMS than those with no identifiable genetic cause. In 46,XY undervirilized individuals with an EMS of 3 or less, genetic testing is much higher yield.

Expanding the Phenotype of DSD Through Molecular Testing: Unexpected Ethical Dilemmas

H. Byers¹, M. Shnorhavorian², P. Fechner³, L. Ramsdell⁴, E. McCauley⁵, S.D. Lundy⁶, M. Adam³

¹Department of Medicine, University of Washington School of Medicine, Seattle, Washington

²Department of Urology, University of Washington School of Medicine, Seattle, Washington

³Department of Pediatrics, University of Washington School of Medicine, Seattle, Washington

⁴Seattle Children's Hospital, Seattle, Washington

⁵Department of Psychiatry, University of Washington School of Medicine, Seattle, Washington

⁶Medical Science Training Program, University of Washington School of Medicine, Seattle, Washington

Improved molecular genetic testing for individuals with Disorders of Sex Development (DSD) has resulted in the expansion of the phenotype of several known conditions. This expanded knowledge has lead to unexpected ethical dilemmas for parents and clinicians. We describe two infants with molecularly confirmed DSD in whom parental wishes have conflicted with physician recommendations and state law.

Patient 1 had completely normal female external genitalia at birth. Masses in her groin were noted during the first few weeks of life. This prompted a karyotype, which resulted normal 46,XY. The patient was initially diagnosed by urology with aphallia and likely dysgenetic gonads. Gonadectomy was recommended. However, at the time of surgery, she was found to have biopsy-proven normal testes and therefore testicles were not removed. Referral to a DSD team and further evaluation revealed homozygous P.212R mutations in SRD5A2, consistent with severe 5-alpha reductase type 2 deficiency (5α-R2D). At last evaluation, the patient was 6 years old and fully identified in the female gender. Her testes have ascended from the labial folds to the inguinal canals. Her parents would still like to pursue gonadectomy.

Patient 2 is the first child of healthy parents. No genital anomalies were identified at birth. However, at 10 days of age, bilateral groin masses were noted, and referral to DSD clinic was made. Karyotype was normal 46,XY. Extensive evaluation revealed a novel missense mutation (c.1210T>G) in NR5A1, which encodes the SF-1 protein, and is associated with sex-reversal. While streak gonads have been clearly linked to increased risk of cancer, it appears that gonads in the setting of 46,XY DSD due to mutations in NR5A1 may carry less oncogenic risk. Her parents also would like to pursue gonadectomy.

Gender assignment in DSD is determined by diagnosis, genital appearance, surgical options, potential fertility and need for lifelong hormonal therapy. Those in favor of female gender assignment for female-appearing individuals with 5α- R2D or NR5A1-sex reversal argue that orchiectomy prevents virilization, leads to a more stable gender identity, and removes the small risk of testicular cancer. Those in favor of male gender assignment argue that orchiectomy takes away the ability of the child to decide gender as a consenting adult and the possibility of fertility, with some experts arguing that individuals should be assigned a male gender from birth regardless of the appearance of the external genitalia. We hope these cases help illustrate the complicated decision-making process of gender assignment in severe forms of DSD, as well as highlight the urgent need for clinical investigation of predictors of adult gender identity in individuals with conditions at the severe end of a clinical spectrum.

Concurrent Evaluation of Multiple Parameters for Informed Decisions Regarding Gonadectomy in Patients With 46, XY Disorders of Sex Development

Catherine E. Keegan^1,2, Richard J. Auchus^1,3, Daniel Teitelbaum^1,4, Vesna Ivancic^1,5, David Sandberg^1,6, Elisabeth H. Quint^1,7

¹Disorder of Sex Development Team, University of Michigan

²Department of Pediatrics, Division of Genetics, University of Michigan

³Department of Internal Medicine, Division of Endocrinology, University of Michigan

⁴Department of Pediatric Surgery, University of Michigan

⁵Department of Pediatric Urology, University of Michigan

⁶Department of Pediatrics, Division of Psychology, University of Michigan

⁷Department of Obstetrics and Gynecology, University of Michigan

The diagnosis of a 46, XY Disorder of Sex Development (DSD) presents many unique challenges for individuals and their families. Patients may present as infants with ambiguous genitalia or as adolescents with primary amenorrhea. For patients and families, the diagnosis of a DSD can lead to several difficult decisions that may significantly impact the child's future. Notable among these, and considered controversial by both clinicians and patient stakeholders, is the decision process related to gonadectomy. There are few comprehensive reviews including a range of DSD diagnoses that can inform this discussion. In this report, we undertook a comprehensive review of the literature regarding gonadectomy for a broad range of 46, XY DSD diagnoses, including gonadal dysgenesis, sex chromosome mosaicism, steroid biosynthesis defects, and androgen insensitivity syndrome, taking into account multiple parameters, including germ cell tumor risk, steroid hormone production, potential for fertility, likelihood of gender reassignment, and patient/family preferences. We also evaluated the roles of imaging, tumor markers, and gonadal biopsy in monitoring for germ cell tumor development. Our results suggest that multiple parameters need to be considered during the decision making process regarding the need and timing for gonadectomy in an individual with 46, XY DSD. Importantly, identifying a specific genetic diagnosis and thereby knowing the natural history of the specific disorder are key components that can assist with management recommendations. At this time, there is a limited role for following tumor markers, routinely obtaining biopsies, or relying on imaging studies to monitor for premalignant changes or early tumor development in those at an increased risk of gonadal tumors. Based on the available evidence, we suggest a shared decision making algorithm that includes tumor risk, steroid hormone production, fertility potential, and parent/patient values and preferences to assist the clinician with counseling regarding gonadectomy in individuals with 46, XY DSD.

DISORDERS OF SEX DEVELOPMENT II

Too Big or Not Too Big? Follow-Up of 33 Cases of Neonatal Clitoral Enlargement

Antonie D. Kline¹, Michael Setzer³, Marcia Ferguson¹, Amy Kimball¹, Howard Birenbaum², Maria Pane², Sabah Helou²

¹Harvey Institute for Human Genetics, Greater Baltimore Medical Center, Baltimore, Maryland

²Division of Neonatology, Department of Pediatrics, Greater Baltimore Medical Center, Baltimore, Maryland

³Genetic Counselor Training Program, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD/National Human Genome Research Institute, Bethesda, Maryland

Enlarged clitoris can be detected at birth in term female infants or after several weeks or months in premature female infants. Etiology is generally attributed to increased androgens from maternal factors, disorders of sexual differentiation with virilization, and several single gene disorders including some specific chromosome disorders. At our hospital, genetics is consulted for clitoral enlargement, by standard measurement¹, as the primary presentation, with an underlying cause not clear in the majority prior to discharge from the nurseries. We undertook to follow-up on 33 of these infants to assess for etiology and any additional findings through their primary care providers.Of 33 female infants with enlarged clitoris, 67% were full term; of the 33% that were premature, all but two presented 4–13 weeks after delivery. Of note, 12% of the pregnancies were products of IVF, and 24% had prenatal exposure to progesterone. On genitalia exam, 15% had rugae and/or hyperpigmentation, 94% had hypoplastic labia majora, 30% had abnormal labia minora, and 21% had vaginal tag. Other findings included 12% with IUGR, 21% with sacral dimple or crease, and 21% with cardiac finding. All had normal female chromosomes. Of those tested, 24% had elevated testosterone and 18% elevated 17-OH-progesterone. In addition, 9% of mothers self-reported a larger than average clitoral size,and up to 6% of sisters and/or aunts,thus 15% with a maternal family history. In terms of outcomes, 32% of the children still have clitoral enlargement, with some referred for surgical reduction, and 3% still have vaginal tag. The enlarged clitoris has resolved in all but one premature case (which needed surgery) and in all “familial” cases. Those old enough to be pubertal have entered female puberty normally. Interestingly, 9% had premature thelarche, and 3% had growth hormone deficiency. Other findings include 9% with overgrowth and/or obesity, 9% with small stature, and in those above 36 weeks gestation,9% with developmental delay including one with severe autism spectrum disorder. Regarding specific diagnoses, two patients have 21-hydroxylase deficiency, with no other endocrine disorders, one patient has Beckwith-Wiedemann syndrome (BWS), and four patients have an unknown underlying syndrome; thus 21% with some underlying genetic condition. There appears to be an increased incidence of abnormal outcome in our cohort. Overall, enlarged neonatal clitoris may indicate a possible underlying condition. Mechanistically, known endocrine causes with elevation of androgens comprise only two of our cases (6%), although reportedly are the most likely causes. Although 15% of our series likely have an underlying single gene syndrome, with only one identified (BWS, with a known association with enlarged clitoris), there was no measurement of androgens. Likewise, in those exposed to exogenous progesterone, there was no measurement of androgen levels. 21% of the nonsyndromic nonendocrine cases had initial elevated testosterone with no subsequent etiology found. Isolated transient clitoromegaly has been reported in premature infants and is thought to be related to transient hormonal change in immature adrenal glands². The fetal zone of the adrenal cortex persists through term and circulating steroids could cause virilization³. Development of premature thelarche, seen in 9% of our cohort, has not previously been reported in association with clitoromegaly, but elevation of androstenedione and leptin has been noted in non-obese girls with premature thelarche⁴, so an androgen influence with unclear origin is possible. Long-term measurement of androgens in these cases is likely indicated, as well as follow-up.

¹Litwin et al., Eur J Obstet Gyn Repro Biol 38:209–12,1990; ²Zimmer et al., J Ultras Med 31:409–15,2012; ³William et al., J Pediatr Urol 9:962–5,2013; ⁴Dundar et al., J Invest Med 61:984–8,2013.

Delineation of Genitourinary Developmental Defects in 118 Infants With Cloacal Malformations

Kim M. Keppler-Noreuil¹, Kristin M. Conway², John C. Carey³, Catherine E. Keegan⁴, Leslie G. Biesecker¹, Paul A Romitti², National Birth Defects Prevention Study (NBDPS)

¹NHGRI, National Institutes of Health, Bethesda, Maryland

²Department of Epidemiology, University of Iowa

³Department of Medical Genetics, Department of Pediatrics, University of Utah Health Sciences Center

⁴Department of Pediatrics Genetics, University of Michigan

Cloacal exstrophy (CE), or OEIS complex, represents a combination of defects consisting of omphalocele, exstrophy of the cloaca, imperforate anus, and spinal defects [Carey et al., 1978]. Persistent cloaca (PC) or Urorectal Septum Malformation Sequence is characterized by a common cloacal cavity, including bladder and intestinal elements, without exstrophy, and usually imperforate anus [Escobar et al., 1987]. Both CE and PC frequently have defects of the genitourinary (GU) system, including anomalies of the internal and external genitalia, in addition to gastrointestinal, skeletal and spinal defects. We analyzed clinical data from the NBDPS, a multi-state population-based case-control study, on 118 infants with CE and PC born from 1997 through 2009, including pre- and postnatal history and exams/testing, and frequency/types of associated anomalies. The purposes of this study are: to describe the spectrum of the GU defects in CE/PC, and their relationship to the other component anomalies, and to discuss the potential abnormal embryologic development and pathogenesis of cloacal malformations. There were 118 infants with CE/PC (ICD-9-CM code 751.550) classified as having CE (N = 50) or PC (N = 68), either isolated (complex sequence) or multiple (CE/PC  + other unassociated major defects): CE, isolated (N = 41, 35%), CE, multiple (N=9, 8%), PC, isolated (N = 28, 24%), and PC, multiple (N = 40, 34%). CE and PC had similar associated GU malformations. All had ambiguous or incompletely formed external genitalia, characterized in 46,XX cases as absent genitalia; labioscrotal folds; fused or small labia; absent to large clitoris; and vesico- and recto-vaginal fistulas; and in 46,XYcases as splayed labioscrotal folds; absent, small, bifid or duplicated phallus and scrotum; and epispadias. Internal genital anomalies included: single to absent ovaries and fallopian tubes; absent, bifid, bicornuate and didelphys uterus; absent to duplicated vagina (in 46,XX), and urethral atresia, and absent or small testes (in 46,XY). Kidney and urinary tract abnormalities (hydronephrosis/ hydroureter, cystic dysplasia, agenesis, pelvic kidney, hemibladders, duplicated collecting system) were found in >60%. Ninety-three/118 (78%) had chromosome analysis (normal); 46,XX in 96% of all PC, while 46,XY and 46,XX occurred almost equally in all CE. Overlapping findings of Limb Body Wall Complex were found in 5/41 isolated CE cases. The multiple cases often were classified as PC with imperforate anus and other malformations, including CHD (N = 20) and laterality defects (N = 2); 10/40 (25%) had features of VACTERL association, such as esophageal atresia/TE fistula in 50%. Cloacal exstrophy and PC likely occur early in embryogenesis. Associated GU findings, like didelphys and bicornuate uterus as described, may be due to abnormalities in fusion of the paramesonephric (Müllerian) ducts. Previous animal studies suggest a defect at the caudal eminence where abnormal cellular proliferation between the umbilical ring and the cloacal membrane may interfere with normal displacement of the cloacal membrane and normal septation of the cloaca. An underlying etiology for CE is yet to be identified; previous candidate gene, microarray analysis, and WES on blood have been negative. CE in concordant monozygotic twins supports a genetic basis. We hypothesize that somatic mutation of a gene(s) involved in early caudal development may be causative of cloacal exstrophy and related cloacal malformations.

Urorectal Septum Malformation Sequence—Unusual Presentation in Three Patients

David D. Weaver¹, Ceciley Starkey Casselman¹, Nayan Srivastava¹ and Luis F. Escobar²

¹Department of Medical and Molecular Genetics, Indiana University School of Medicine, Indianapolis, Indiana

²Medical Genetics and Neurodevelopment Center Payton Manning Children's Hospital, St. Vincent Hospital, Indianapolis, Indiana

The urorectal septum malformation sequence (URSMS) was first reported and named by Escobar et al. in 1987. Since then, there have been numerous reports of this condition and the recognition of a less severe form, the partial URSMS. The full URSMS is characterized by lack of perineal openings, ambiguous genitalia, rectovesicle fistula, malformed bladder, distal colon atresia, lower lumbar and sacral vertebral defects and renal agenesis and/or dysplasia. Death is normally from respiratory insufficiency secondary to lung hypoplasia related to oligohydramnios. The partial URSMS differs by having a single perineal opening draining a persistent cloaca and by having normal lung function. Here, we present three patients with unusual presentations of the full URSMS.

Patient 1 was born at 38 weeks gestation with the Potter phenotype, marked redundancy of the abdominal skin, distended bowel, a small and phallic-like structure with no urethral or vaginal openings, a well-formed scrotum with no median raphe, no palpable testes, imperforate anus, multicystic-dysplastic kidneys with hydroureters, distended bladder, malformed sacral vertebrae and imperforate anus. A G-banded karyotype was 46,XX. She died from respiratory failure at three days.

Patient 2 was delivered at 34 weeks gestation. His external genitalia were most unusual with a markedly enlarged penis, no urethral opening, bifid scrotum and no palpable testes. Other findings included bilateral severe hydronephrosis with dilated ureters, distended bladder, rectourethral fistula, massively dilated sigmoid colon and imperforate anus. A G-banded karyotype was 46,XY and a comparative genomic hybridization was unremarkable. This child survived his respiratory distress with intensive assisted ventilation and is now seven years old.

Patient 3 was diagnosed prenatally with the URSMS even though there was normal amniotic fluid volume. The parents were told she would die shortly after birth. Following birth, she had no respiratory distress but did have a patent umbilical urachus, ambiguous external genitalia with no urethral, vaginal or anal openings, no perineal raphe, absent left and dysplastic right kidneys, right hydronephrosis, a bladder and vaginal communication, a rectovaginal fistula and lumbosacral vertebral anomalies. A G-banded karyotype was 46,XX and whole exome analysis showed six variants of unknown significance. She is alive and doing well at this time.

All three of the above patients meet the diagnostic criteria for URSMS. The first patient also had the prune belly syndrome, the second very bizarre genitalia and the third survived because of a patent urachus, allowing for adequate amniotic fluid. These patients demonstrate that patients with the URSMS can masquerade as other conditions, can have unusual findings and do not always die from respiratory insufficiency. They also illustrate that if a newborn has very unusual or bizarre external genitalia, one should consider the diagnosis of URSMS.

Spontaneous Ovulation in a Male Adolescent: Report of a Rare Ovotesticular Disorder of Sex Development and Pathogenic Hypothesis

Stephen R. Braddock, Katherine M., Christensen, Shannon G. Farmakis, Sarah Starnes, Cirilo Sotelo- Avila, Jacqueline R. Batanian, David P. Dempsher

Saint Louis University School of Medicine and Cardinal Glennon Children's Medical Center, St. Louis, Missouri

Disorders of sex development (DSD) relate to a disruption of the unique and complex process which targets the bipotential gonad and can result in discordance between phenotype and genotype in humans. While generally rare in the population, when DSD do occur, there is frequently genital ambiguity on examination. The purpose of this presentation is to describe a case of ovotesticular DSD presenting with spontaneous ovulation in a well virilized adolescent male, and to postulate a potential mechanism for this rare form of DSD to occur.

The proband first presented at age 15 years with a rapidly enlarging left scrotal mass, which was solid on outside ultrasound. The left testis was described as enlarged and the right testis “considerably smaller.” At surgery, 30 cc of blood was drained from what appeared to be a hematocele within the tunica vaginalis. Initial diagnosis was “testicular rupture,” however final pathology revealed the lesion was a ruptured hemorrhagic corpus luteum. Another specimen from the left testis showed ovarian stroma, immature seminiferous tubules lined by Sertoli cells, and Leydig cells. Karyotype of gonadal tissue was 46,XX. Lymphocyte karyotype was 46,XX with negative SRY and Y specific satellite probes.

On examination at age 17 years, height was 68 inches (31%) and weight 185 lbs (90%). He had a modest amount of shaved facial hair, normal axillary hair and deep voice. He had a normal adult penis with Tanner V pubic hair and a hypoplastic scrotum with soft 2–3 cc testes. He reported normal sexual function and male gender identity. Laboratory findings included normal testosterone (539 ng/dL), and estradiol (27.3 pg/mL) with elevated LH (10.9 mIU/mL) and FSH (16.4 mIU/mL) for adult males. Buccal smear karyotype 46,XX with SRY and Y-specific probes negative. MRI of pelvis identified normal prostate, seminal vesicles, epididymidi and small testes but no Mullerian structures.

46,XX ovotesticular DSD is uncommon, accounting for less than 10% of all DSD and is frequently associated with genital ambiguity of some degree. The pathogenesis is unclear for the majority of cases but may be related to mosaicism or genetic mutations. Most cases are SRY negative. 46,XX ovotesticular DSD presenting with male phenotype and ovulation is rare, with only two previous cases in the literature. One of these had confirmed 46,XX/47,XXY mosaicism. Historically it was assumed that if SRY was not present, the gonad proceeded along the “default” mode to an ovary. This alone would not explain 46,XX ovotesticular DSD, particularly with normal male genitalia. We hypothesize that in both males and females there is a retained ability to differentiate the gonad. These competing pathways are not independent so that when one is activated it must also suppress the alternate pathway. Key regulatory factors required in this mechanism include RSPO1 and WNT4, which stabilize ß-catenin signaling. The interplay between these factors allows for development of both testicular and ovarian tissue in this patient with a non-ambiguous adult male phenotype.

NEW INSIGHTS INTO OLD SYNDROMES

Lateral Meningocele Syndrome Is Caused by Truncating Mutations in the Last Exon of NOTCH3

Karen W. Gripp¹, Katherine M. Robbins², Nara L. Sobreira³, P. Dane Witmer⁴, Lynne M. Bird⁵, Kristiina Avela⁶, Outi Makitie⁷, Daniela Alves⁸, Jacob S. Hogue⁹, Elaine H. Zackai¹⁰, Cedric le Caignec¹¹, Valerie Cormier-Daire¹¹, Kimberly F. Doheny⁴, Deborah Stabley², Katia Sol- Church²

¹Medical Genetics and ²Biomedical Research, A. I. duPont Hospital for Children, Wilmington, Delaware

³Johns Hopkins University and Institute of Genetic Medicine, Baltimore, Maryland

⁴Center for Inherited Disease Research, Institute of Genetic Medicine, Baltimore, Maryland

⁵UCSD and Rady Children's Hospital, San Diego, California

⁶Clinical Genetics, Helsinki University Central Hospital, Helsinki, Finland

⁷Children's Hospital, Helsinki University Central Hospital, Helsinki, Finland

⁸Neurogenetics Unit, Centro Hospitalar de Sao Joao, Porto, Portugal

⁹Madigan Army Medical Center, Tacoma, Washington

¹⁰Human Genetics and Molecular Biology, CHOP, Philadelphia, Pennsylvania

¹¹INSERM, L'Institut du Thorax, Nantes, France

Lateral meningocele syndrome (LMS) (MIM130720) is a very rare connective tissue disorder with facial anomalies, hypotonia, skeletal anomalies and lateral meningocele associated neurologic problems. Lateral meningoceles are severe dural ectasias protruding through the neural foramina. Fewer than 20 individuals with this condition have been reported and clinical overlap with Hajdu-Cheney syndrome (HCS) was noted. We performed exome analysis and/or Sanger sequencing on six samples from unrelated individuals with LMS and one with HCS.

Whole exome sequencing identified heterozygous de novo mutations in NOTCH3 in four individuals with LMS, similar mutations were found in two others. All mutations were nonsense or frameshift and clustered in the last exon, thus presumably escaping nonsense-medicated RNA decay and resulting in premature protein termination. The predicted abnormal protein product lacks the c-terminal PEST domain, which is responsible for protein degradation. This truncated NOTCH3 may cause gain-of-function through decreased clearance of its active intracellular product, resembling NOTCH2 mutations and their postulated effect in the clinically related HCS. This contrasts with the NOTCH3 mutations in CADASIL, which affect the number of extracellular cysteine residues and are postulated to result in a neomorphic effect. While we identified a NOTCH3 mutation in each individual with LMS, the single HCS patient, who previously was found to lack a NOTCH2 mutation, did not have a NOTCH3 change.

In conclusion, heterozygous truncating NOTCH3 mutations are the principal cause of LMS. Patients share a recognizable facial phenotype, skeletal and neurologic findings. LMS overlaps with HCS, but these remain distinct conditions resulting from different NOTCH mutations, despite their postulated shared mechanism of prolonged NOTCH signaling.

Germline Mosaicism for an ACTA2 Mutation in Moyamoya Disease Type 5

Chad R. Haldeman-Englert, MD,¹ Tara H. Stamper, MS,¹ Anna C. E. Hurst, MD,² Stacey Q. Wolfe, MD³

¹Department of Pediatrics, Section on Medical Genetics, Wake Forest School of Medicine, Winston-Salem, North Carolina

²Department of Genetics, University of Alabama at Birmingham, Birmingham, Alabama

³Department of Neurosurgery, Wake Forest School of Medicine, Winston-Salem, North Carolina

Moyamoya disease (MYMY) is a cerebrovascular disorder caused by stenoses of the terminal aspects of the internal carotid arteries accompanied by surrounding fine arterial collateral vessels that resemble a 'puff of smoke' on imaging studies. Sudden onset hemiplegia and epileptic seizures can occur during childhood due to transient brain ischemia, whereas adults often present with symptoms related to intracranial hemorrhage. MYMY is genetically heterogeneous, and the classic findings can be seen in various genetic conditions such as neurofibromatosis type 1, Alagille syndrome, and microcephalic osteodysplastic primordial dwarfism type II. There are also six susceptibility loci for MYMY, with MYMY2 (RNF213), MYMY5 (ACTA2), and MYMY6 (GUCY1A3) involving an identified gene.

We report here on a 7-year-old male presenting with stroke-like symptoms that included right arm numbness, confusion, facial droop, and speech difficulties. Brain imaging (MRI/MRA) revealed irregular narrowing of the bilateral distal internal carotid arteries (ICAs), proximal middle cerebral arteries, and proximal left anterior cerebral artery, with additional ischemic changes noted. He subsequently has undergone a bilateral craniotomy for encephalodural pial synangiosis at 9-years-old, but continues to have multiple transient ischemic attacks with persistent ICA stenoses. ACTA2 testing showed a heterozygous mutation (c.772C>T; p.Arg258Cys) that has been previously described in two other families with MYMY. An echocardiogram and MRA of his chest and abdomen were normal, yet he complains of frequent abdominal pain. His asymptomatic parents tested negative for the mutation. However, per parental request the asymptomatic 6-year-old brother was tested and found to carry the ACTA2 mutation. His brain MRI/MRA demonstrated changes consistent with MYMY and remote ischemia. Analysis of the parental buccal cells for the mutation was also negative. ACTA2 encodes for alpha-actin, which constitutes the thin filaments that regulate the smooth muscle cell (SMC) contraction occurring in response to the stretch from pulsatile blood flow. Mutations of ACTA2 have been associated with several conditions, including thoracic aortic aneurysm and dissection (AAT6), MYMY5, and multisystemic smooth muscle dysfunction syndrome. Mechanistically, ACTA2 missense mutations in MYMY appear to have a gain of function effect, leading to increased vascular SMC proliferation and the subsequent stenosis or occlusion of the affected arteries. Other organs that depend on SMC function may also be affected, distinguishing these patients as having a SMC abnormality rather than a connective tissue condition. This should direct the clinician to evaluate for symptoms related to SMC dysfunction (possibly explaining our patient's abdominal pain due to abnormal peristalsis), and may be a possible target for future therapy of these disorders.

Genotype Phenotype Correlation in 7Q11.23 Duplication Syndrome

Colleen A. Morris¹, Carolyn B. Mervis², Myra J/ Huffman², Shelley L. Velleman³, Bonita P. Klein-Tasman⁴, Elaine Tam⁵, Lucy R. Osborne⁵

¹University of Nevada School of Medicine, Las Vegas, Nevada

²University of Louisville, Louisville, Kentucky

³University of Vermont, Burlington, Vermont

⁴University of Wisconsin-Milwaukee, Milwaukee, Wisconsin

⁵University of Toronto. Toronto, ON, Canada

The 7q11.23 duplication syndrome phenotype includes macrocephaly (41%), ventriculomegaly (40%), seizure disorder (19%), hypotonia (58%), chronic otitis media (25%), aortic dilatation (40%), and chronic constipation (61%). The recognizable facial phenotype is characterized by a broad forehead, deep set eyes, high broad nose, long columnella, short philtrum, thin upper lip, minor ear anomalies, and facial asymmetry. Severe speech delay, Speech Sound Disorder (82%), Developmental Coordination Disorder (74%), Specific Phobia (53%), and Social Phobia (50%) are common in children with this syndrome; 35% have ADHD, 29% have Selective Mutism, 24% have Oppositional Defiant Disorder or Disruptive Behavior Disorder-Not Otherwise Specified, and 20% have an Autism Spectrum Disorder. Intellectual ability is typically at the low average level, with a range from severe intellectual ability to high average. The classic duplication includes 25 genes and is 1.55 Mb. To begin to make genotype phenotype correlations, we have evaluated four individuals (including a mother-daughter pair) who have partial duplications of the region, ranging from 618 kb to 1.13Mb. Each proband had a clinical microarray performed before referral and we performed real-time quantitative PCR with probes spanning the 7q11.23 commonly duplicated region and the single copy regions immediately flanking it to determine the extent of the duplication.

Individuals with smaller duplications had higher overall intellectual ability (IQ) than those with classic duplications. Only one of the four had Social Phobia. Speech sound disorder was present in two individuals, including one whose duplication spanned the entire classic region except for the most telomeric gene, GTF2I, suggesting that GTF2I is not involved in the speech phenotype. (Deletion of GTF2I is implicated in the cognitive disability in Williams syndrome). The facial phenotype was not present in individuals who had the normal two copies of centromeric genes in the region.

A fifth individual had both a partial 7q11.23 duplication inherited from her mother and a duplication of a gene telomeric to the typical duplication, HIP1 inherited from her father. That child and her father had a distinct phenotype of hypotonia, ptosis, epicanthal folds, and weakness of the lower extremities, as well as IQ in the borderline to mild intellectual disability range, suggesting that HIP1 duplication is a novel condition and that the daughter has a blended phenotype. HIP1 deletion has been implicated in the more severe intellectual disability characteristic of individuals with Williams syndrome who have longer deletions.

Supported by NIH grants HD067244 and NS076465.

De Novo Nonsense Mutations in KAT6A, a Lysine Acetyl-Transferase Gene, Cause a Syndrome Including Microcephaly and Global Developmental Delay

Arboleda VA, Lee H, Dorrani N, Zadeh N, Willis M, Macmurdo CF, Manning MA, Kwan A, Hudgins L, Barthelemy F, Miceli MC, Quintero-Rivera F, Kantarci S, Strom SP, Deignan JL; UCLA Clinical Genomics Center, Grody WW, Vilain E, Nelson SF

Chromatin remodeling through histone acetyltransferase (HAT) and histone deactylase (HDAC) enzymes affects fundamental cellular processes including the cell-cycle, cell differentiation, metabolism, and apoptosis. Nonsense mutations in genes that are involved in histone acetylation and deacetylation result in multiple congenital anomalies with most individuals displaying significant developmental delay, microcephaly and dysmorphism. Here, we report a syndrome caused by de novo heterozygous nonsense mutations in KAT6A (a.k.a. MOZ, MYST3) identified by clinical exome sequencing (CES) in four independent families. The same de novo nonsense mutation (c.3385C>T [p.Arg1129*]) was observed in three individuals, and the fourth individual had a nearby de novo nonsense mutation (c.3070C>T [p.Arg1024*]). De novo heterozygous nonsense mutations have been identified throughout the C-terminal half of KAT6A in at least 13 unrelated individuals (Tham et al 2015). Common features among all four probands include primary microcephaly, global developmental delay including profound speech delay, and craniofacial dysmorphism, as well as more varied features such as feeding difficulties, cardiac defects, genital and ocular anomalies. These same finding were identified in seven patients and are being further investigated through a patient-initiated registry to understand the full phenotypic spectrum underlying KAT6A mutations. We further demonstrate that KAT6A mutations result in dysregulation of H3K9 and H3K18 acetylation and altered P53 signaling. KAT6A has been implicated in stress responses in vivo and in vitro and has been identified in translocations in acute monocytic leukemia. We show that treatment with Doxorubicin show an upregulation of KAT6A with a specific nuclear distribution suggesting that it plays a significant role in chromatin dynamics in response to cellular stress. Further studies to understand the DNA and protein targets of KAT6A acetylation and the consequent changes in RNA are being performed in fibroblasts from wild-type and patients with de novo nonsense KAT6A mutations. Through a more comprehensive understanding of the chromatin dynamics resulting from KAT6A mutations, we hope to understand the underlying role of epigenetics and acetylation in regulating development and disease.

Expanding the Phenotype of Patients With PIGN MUTATIONS. Three Unrelated Patients Presenting With Hypotonia, Intractable Seizures and Congenital Anomalies

Reynoso FJ¹, Horton Harr M¹, Kesewa A², Kalish J¹, Li D³, Hakonarson H³, Bloom K⁴, Burton BK⁵, Bucher J⁵, He M^{4, 6}, Zackai EH¹

¹The Children's Hospital of Philadelphia, Clinical Genetics Department

²Perelman School of Medicine, University of Pennsylvania

³The Center for Applied Genomics at the Children's Hospital of Philadelphia

⁴Palmieri Metabolic Disease Laboratory, Children's Hospital of Philadelphia

⁵Division of Genetics, Birth Defects and Metabolism Ann & Robert H. Lurie Children's Hospital of Chicago

⁶Department of Pathology and Laboratory Medicine, Perelman School of Medicine University of Pennsylvania, Philadelphia, Pennsylvania

The PIGN gene encodes glycosylphosphatidylinositol (GPI) ethanolamine phosphate transferase-1, involved in the biosynthesis of the GPI anchor. GPI anchoring of proteins is a highly conserved process present in most eukaryotic cells. GPI-anchored proteins perform a diverse set of functions including roles in signal transduction, cell adhesion and antigen presentation1.

To date mutations in 12 genes involved in GPI anchor synthesis have been associated with human disease2. Maydan first described mutations in PIGN in a consanguineous Israeli family of Palestinian origin; children presented with multiple congenital anomalies, severe neurological involvement and early death3. Three additional cases have been reported since then: two Japanese siblings with hypotonia, dysmorphic features and cerebellar atrophy4 and a fetus with congenital diaphragmatic hernia and other malformations from a consanguineous couple of North African origin5.

We report three unrelated patients with mutations in PIGN and diverse clinical characteristics. Patients 1 and 2 both presented in the newborn period with hypotonia, intractable epilepsy, facial dysmorphism as well as gastrointestinal, genitourinary and skeletal anomalies; including brachitelephalangy. Patient 1 had the additional findings of metopic suture craniosynostosis and left choanal atresia. Patient 2 had an imperforate anus, uterine didelphys and a small splenic cyst. Patient 3 with a milder phenotype manifested as hypotonia, developmental delay, minor dysmorphic features and early onset refractory epilepsy. Two of our patients were diagnosed via WES; Patient 1 had compound heterozygous mutations (c.1674+G>C affecting a splice site/c.2679 C>G a missense in a highly conserved residue). Patient 3 had novel heterozygous mutations in PIGN (c.932T>A missense and c.806 -4_808delGTAGGTT affecting a splice site). On patient 2 we did targeted sequencing of the PIGN gene after Pt. 1 was diagnosed based on the clinical similarities, resulting in the previously reported c.2126G>A described by Maydan in the first PIGN family and a c.287C>G missense. FACS (Fluorescence activated cell sorting) analysis was performed on Pt. 2 and Pt. 3, which demonstrated decreased expression of the GPI anchored protein CD59 on skin fibroblasts adding to the evidence of the pathogenicity of their mutations.

Our cases expand the phenotypic spectrum of the previously known clinical consequences of mutations in PIGN adding the findings of brachitelephalangy, metopic craniosynostosis and uterine didelphys as features seen in patients with this condition.

The Distinctive Cardiovascular Phenotype Of Myhre Syndrome Includes Both Pericardial Disease and Restrictive Cardiomyopathy

A.E. Lin,¹C. Michot,³ V. Cormier-Daire³, P. Matherne,⁴ B.H. Barnes,⁵ M.E. Lindsay²

¹Genetics, MassGeneral Hospital for Children

²Pediatric Cardiology, MassGeneral Hospital for Children

³Necker Enfants-Malades Hospital, Paris

⁴Divison of Cardiology, University of Virginia

⁵Divison of Gastroenterology, University of Virginia

Myhre syndrome (Laryngeal–Arthropathy–Prognathism Syndrome, LAPS) is a rare distinctive short stature syndrome due to specific mutations in SMAD4. The radiographic defects, characteristic facial features, hearing loss, laryngeal anomalies, arthropathy and intellectual disabilities (ID) create a recognizable phenotype. Cardiovascular (CV) anomalies reported in ∼50% have been reported as a “broad spectrum.” We report an adolescent with Myhre syndrome and protein losing enteropathy (PLE) due to restrictive cardiomyopathy (RCM) which may mimic the elevated atrial pressures of pericardial disease. We analyze the two largest series of Myhre syndrome,^1,2 add new cases and delineate the distinctive pattern of CV involvement.

CASE: We evaluated a 17 yo girl with short stature, mild ID/autism spectrum, tight joints, subtle unusual facial features, mild hearing loss, and hypothyroidism in whom Myhre syndrome was detected using Whole Exome Screening (WES). Echocardiography noted normal valves, and CT with angiography further defined a 5–6 cm long segment of mildly hypoplastic abdominal aorta and narrowing of the proximal celiac artery; pericardial disease was excluded. Because she had hepatomegaly and protein losing enteropathy (PLE), cardiac catheterization was performed which confirmed biatrial pressures consistent with RCM. Her clinical course is stable with diuretics, thyroid replacement, and monitoring for possible cardiac transplantation.

RESULTS: After excluding duplicate pts, we studied 42 literature pts with a SMAD4 mutation and this new pt. CV anomalies of any type were present in 38 (88%) including simple shunts (26%), left- and right-sided obstruction (37%, 5%), pericarditis (19%), and restrictive CM (1,2%). Major structural CHDs were not observed. Additional pts include an 8yo girl with RCM (who died after transplantation) reported as a poster at the 2012 David Smith Workshop, and three patients with pericardial disease among nine clinically described pts without molecular testing.

DISCUSSION: The pattern of CV anomalies in Myhre syndrome is striking, and includes left-sided obstruction, especially coarctation and long segment hypoplasia, pericardial disease and RCM. The latter two distinguish Myhre from other syndromes caused by mutations in the TGF-β signaling cascade such as Marfan or Loeys-Dietz syndrome (i.e., TGF-β vasculopathies), Alagille syndrome and rasopathies. High quality diagnostic imaging and catheterization may be needed to distinguish RCM from pericardial disease. Both pericardial disease and RCM cause elevated atrial pressures, and PLE as noted in our patient (more commonly associated with the univentricular Fontan circulation). They must be distinguished from each other, as both are treatable, but by different modalities. RCM is rare, may occur due to allelic mutations causing DCM or HCM, or with inborn errors of metabolism. Pericardial disease occurs in MULIBREY, CACP and Cantu syndromes. As the prototypical “co-SMAD,” the SMAD4 protein has been shown to integrate diverse signaling pathways, including canonical TGF-β, BMP, and Activin signaling amongst others. We hypothesize that the involvement of alternate signaling pathways may account for the expanded spectrum of CV abnormalities in Myhre syndrome. The co-occurrence of congenital and acquired phenotypes demonstrates that SMAD4 is required for both developmental (CHDs, aortic obstruction) and postnatal CV homeostasis (muscle, pericardial and valve disease).

¹Caputo V et al., 2012, AJHG 90:161; ²Michot C et al., 2014, EJMG 22:1272.

NEW SYNDROMES

Chops Syndrome: A Novel Genetic Disorder of Cognitive Impairment, Heart Defects, Obesity, Pulmonary Involvement and Short Stature With Skeletal Dysplasia

Ian Krantz, MD^1,2, Sarah Noon, MS¹, Andrew Edmondson, MD, PhD¹, Maninder Kaur, MS¹, Elaine Zackai, MD^1,2, Zhe Zhang, PhD³, Katsuhiko Shirahige, PhD⁴, Kosuke Izumi, MD, PhD⁴

¹Division of Human Genetics, The Children's Hospital of Philadelphia

²Perelman School of Medicine at the University of Pennsylvania, Philadelphia

³Center for Biomedical Informatics, The Children's Hospital of Philadelphia

⁴Research Center for Epigenetic Disease, The University of Tokyo

Three probands, who had been collectively followed over a period of 12 years, were all referred for initial evaluation due to a suspected diagnosis of Cornelia de Lange syndrome (CdLS). All probands had strikingly similar clinical manifestations and were felt to represent a common unknown diagnosis phenotypically overlapping CdLS. The shared clinical features with CdLS include developmental delay/intellectual disability (with language being most significantly delayed), short stature, failure to thrive and facial dysmorphisms (arched eyebrows, synophrys, short nose, posterior rotated ears, long philtrum, micrognathia), small hands and feet. However, these probands have additional clinical features including obesity and severe respiratory manifestations (subglottic stenosis, laryngo-tracheomalcia, tracheal stenosis, chronic lung disease) that distinguished them from CdLS. Their facial features were not typical of CdLS and appeared coarser with age and the skeletal manifestations (brachydactyly, vertebral abnormalities) were different from those seen in CdLS. We have named this new syndrome CHOPS syndrome (C for Cognitive impairment and Coarse facies, H for Heart defects, O for Obesity, P for Pulmonary involvement and S for Short stature and Skeletal dysplasia). All probands had been extensively screened for underlying genetic diagnoses by high-resolution SNP arrays, multiple gene analyses and evaluation for biochemical disorders (including mucopolysaccharadisoses). Due to the previously undescribed constellation of physical and medical features that were highly conserved among all three probands, we hypothesized that this condition represents a novel genetic disorder and was likely caused by de novo mutations (lack of involvement of any other family members) in a single causative gene.

Using whole exome sequencing, we identified missense mutations in all three probands in a novel disease gene (AFF4) that plays an important role in regulating transcriptional elongation. Transcription elongation is a critical mechanism by which gene expression is regulated during development and has been extensively studied. Somatic mutations in genes that control transcriptional elongation have been described in cancers but to our knowledge this is the first human developmental disorder caused by germline disruption of this process. To evaluate the effect of these missense mutations, whole genome expression arrays were performed on patient-derived skin fibroblast cell lines identifying many developmentally important misexpressed genes. Since the role of cohesin in transcriptional elongation has recently been proposed, this newly identified diagnosis may provide valuable insights into how cohesin functions in transcriptional elongation. This adds to a growing number of developmental diagnoses caused by disruption of normal transcriptional processes (collectively termed “transcriptomopathies”).

The clinical features and natural history of CHOPS syndrome will be presented as well as a brief over view of disorders of transcriptional control.

DYRK1A Haploinsufficiency Causes a New Recognizable Syndrome With Congenital Microcephaly, Prenatal and Postnatal Growth Retardation, Speech Impairment and Distinct Facies

F. Quintero-Rivera. MD^1,13, J. Ji MD ^1,13, H. Lee, PhD^1,13, B. Argiropoulos, PhD ², N. Dorrani, MS^1,5,13, J. Mann, MD³, J.A. Martinez-Agosto, MD, PhD^1,4,5,13 N. Gomez-Ospina, MD⁶, N. Gallant, MD⁴, J.A. Bernstein, MD⁶, L. Hudgins, MD⁶, L. Slattery, MS⁶, B. Isidor, MD⁷, C. Le Caignec MD⁷, A. David, MD⁷, E. Obersztyn, MD⁸, B. Wiśniowiecka-Kowalnik, PhD⁸, M. Fox^5,13, J.L. Deignan, PhD^1,13, E. Vilain, MD, PhD^1,4,5,13, E. Hendricks, MS⁹, M.H. Harr, MS¹⁰, S.E. Noon, MS¹⁰, J.R. Jackson, BS¹⁰, A. Wilkens, MS¹⁰, G. Mirzaa, MD⁹, N. Salamon, MD¹¹, J. Abramson, PhD ¹², E.H. Zackai, MD¹⁰, I. Krantz,MD¹⁰, A.M. Innes, MD², S.F. Nelson, MD^1,4,5,13, W.W. Grody, MD^1,4,5,13

¹Departments of Pathology and Laboratory Medicine, University of California Los Angeles, California

²Department of Medical Genetics, Cumming School of Medicine, University of Calgary, and Alberta Children's Hospital Research Institute for Child and Maternal Health, Calgary, AB

³Kaiser Permanente, Fresno, California

⁴Department of Human Genetics,University of California Los Angeles, California

⁵Department of Pediatrics, University of California Los Angeles, California

⁶Department of Pediatrics, Stanford University School of Medicine, Stanford, California

⁷CHU Nantes, Service de Génétique Médicale, Nantes, France

⁸Institute of Mother and Child, Warsaw, Poland

⁹Seattle Children's Research Institute, Seattle, Washington

¹⁰Children's Hospital of Philadelphia, Philadelphia, Pennsylvania

¹¹Department of Radiology, University of California Los Angeles, California

¹²Physiology of the David Geffen School of Medicine at University of California Los Angeles, California; The Institute for Stem Cell Biology and Regenerative Medicine (inStem), National Centre for Biological Sciences–Tata Institute of Fundamental Research, Bellary Road, Bangalore-560065, Karnataka, India

¹³UCLA Clinical Genomics Center, Los Angeles, California

DYRK1A (dual-specificity tyrosine- (Y)-phosphorylation regulated kinase 1A) is a highly conserved gene located in the Down syndrome critical region. It plays an important role in early development and regulation of neuronal proliferation. Microdeletions of chromosome 21q22.12q22.3 that include DYRK1A (21q22.13) are rare and only a few pathogenic single nucleotide variants (SNVs) in DYRK1A gene have been described, so as of yet, the landscape of DYRK1A disruptions and their associated phenotype has not been fully explored. We have identified 14 individuals with de novo heterozygous variants of DYRK1A; five with microdeletions, three with small insertions or deletions (INDELs) and six with deleterious SNVs. The analysis of our cohort and comparison with published cases reveals that phenotypes are consistent among individuals with the 21q22.12q22.3 microdeletion and those with translocation, SNVs or INDELs within DYRK1A. All individuals shared congenital microcephaly at birth, intellectual disability, developmental delay, severe speech impairment, short stature, and distinct facial features. The severity of the microcephaly varied from −2 SD to -5 SD. Seizures, structural brain abnormalities, eye defects, ataxia/broad- based gait, intrauterine growth restriction, minor skeletal abnormalities, and feeding difficulties were present in two thirds of all affected individuals. Our study demonstrates that haploinsufficiency of DYRK1A results in a new recognizable syndrome, which should be considered in individuals with Angelman syndrome-like features, and distinct facial features. Our report represents the largest cohort of individuals with DYRK1A disruptions to date, and is the first attempt to define consistent genotype-phenotype correlations among subjects with 21q22.13 microdeletions and DYRK1A SNVs or small INDELs (J Ji et al. EJHG, in press).

De Novo Loss of Function Mutations in Ribonucleoprotein HNRNPK Cause a Novel Malformation Syndrome Characterized by Unique Facial Dysmorphisms, Skeletal and Connective Tissue Abnormalities, and Intellectual Disability

PYB Au¹, J You³, O Caluseriu⁴, J Schwartzentruber⁵, J Majewski⁵, FP Bernier^1,2, M Ferguson⁶, Baylor-Hopkins Centre for Mendelian Genomics, Care for Rare Canada Consortium, D Valle³, JS Parboosingh^1,2, N Sobreira³, AD Kline⁶, AM Innes^1,2

¹Department of Medical Genetics, University of Calgary, Calgary, Alberta, Canada

²Alberta Children's Hospital, Research Institute for Child and Maternal Health, University of Calgary, Calgary, Alberta, Canada

³Institute of Genetic Medicine, Johns Hopkins University School of Medicine, Maryland

⁴Department of Medical Genetics, University of Alberta, Edmonton, Alberta, Canada

⁵Department of Human Genetics, McGill and Genome Quebec Innovation Center, McGill University, Quebec, Canada

⁶Department of Pediatrics, Harvey Institute for Human Genetics, Greater Baltimore Medical Center, Maryland

We report a new malformation syndrome due to mutations in heterogeneous nuclear ribonucleoprotein K (HNRNPK). We describe two patients, one with a de novo frameshift (NM_002140.3: c.953+1dup, which extends the exon by one nucleotide), and the other with a de novo splice donor site disruption (NM_031262.6: c.257G>A). These patients have moderate cognitive disability, vertebral defects, connective tissue abnormalities, structural congenital heart defects, and genitourinary abnormalities. The facial gestalt is particularly distinct, characterized by long palpebral fissures and prominent eyes with ptosis, a broad nasal bridge, simplified ear helices, and an open downturned mouth with an unusually prominent median crease to the tongue. A strikingly similar phenotype has also recently been described by Hancarova et al. [2014] in a patient with a 2 Mb deletion at 9q21.3 encompassing HNRNPK.

Heterogeneous nuclear ribonucleoproteins (hn RNPs) are nucleic acid binding proteins with diverse roles in regulation of chromatin remodelling, transcription, RNA stability and splicing, translation, and signal transduction. Many hnRNPs have been either directly or indirectly implicated in human disease, particularly for cancer and neurodegenerative disorders. Recently, germline mutations and chromosomal microdeletions affecting HNRNPU have been linked to epilepsy. However, HNRNPK is the first hn RNP gene to be implicated in a Mendelian congenital malformation syndrome.

The identification of this syndrome was made possible by a new online tool, GeneMatcher, which facilitates connections between clinicians and researchers based on shared interest in a candidate gene. Therefore, this report also highlights the evolving paradigm of “reverse phenotyping,” in which further characterization of phenotype follows the identification of genetic variants.