The 3rd Scientific Cornelia de Lange Syndrome Symposium was held in Chicago, IL, in June, 2008, in conjunction with the Cornelia de Lange Syndrome (CdLS) Foundation ( national conference, bringing together clinicians, basic science researchers, physician–scientists, allied health therapists, trainees, students, families, and individuals with CdLS to review and discuss multiple aspects of the condition. The symposium was supported by the CdLS Foundation. The publication of the abstract book was supported by the CdLS Foundation and the University of Chicago Genetic Services Laboratories, and the Harvey Institute for Human Genetics. Continuing medical education credits were provided by the University of Chicago Pritzker School of Medicine, and continuing education counseling credits were obtained through the National Board for Certified Counselors.

The purpose of the symposium was multifold: to update knowledge of the broad spectrum of clinical findings in and natural history of CdLS; to improve understanding and basic knowledge of behavioral issues in CdLS, specifically self-injurious behavior and autism; to gain knowledge of clinical management and counseling issues related to CdLS; to understand and gain knowledge of the molecular basis for CdLS as a cohesinopathy; and to obtain knowledge of animal model studies related to CdLS and their applications. The overall goal was to provide a multidisciplinary forum for discussion of relevant findings related to CdLS and the cohesinopathies, working towards management and eventual treatment for these conditions. The following publication of the abstracts presented in the meeting will help provide knowledge to the scientific and genetics community about CdLS.

The presenters included:

Antonie D. Kline1, Dale Dorsett2, Miriam Gordillo3, Ethylin W. Jabs3, Joseph Hylton4, Douglas Clemens5, Jinglan Liu6, Soma Das7, Matthew Deardorff6, Mary Beth Bruder8, Laird Jackson6,9, Bin Zhang10, Jeffrey Mildbrandt10, Shimako Kiwauchi11, Anne Calof11, Akihiko Muto11, Arthur Lander11, Julia Horsfield12, Chris Oliver13, Marco Grados14, Ian D. Krantz6

1Harvey Institute for Human Genetics, Greater Baltimore Medical Center, Baltimore, MD; 2Department of Biochemistry and Molecular Biology, St. Louis University School of Medicine, St. Louis, MO; 3Department of Genetics and Genomic Sciences, Mount Sinai School of Medicine, New York, NY; 4Pediatric Dentistry, The University of Chicago College of Dentistry, Chicago, IL; 5Cross Keys Dental Associates, Baltimore, MD; 6Division of Human Genetics, The Children's Hospital of Philadelphia, The University of Pennsylvania School of Medicine, Philadelphia, PA; 7Department of Human Genetics, The University of Chicago, Chicago, IL; 8Pappanikou Center for Excellence in Developmental Disabilities, University of Connecticut, Farmington, CT; 9Department of Genetics, Internal Medicine, and Obstetrics/Gynecology, Drexel University College of Medicine, Philadelphia, PA; 10Department of Pathology, Immunology, Neurology, Washington University School of Medicine, St. Louis, MO; 11Deparment of Developmental and Cell Biology, Developmental Biology Center, Center for Complex Biologic Systems, University of California, Irvine, CA; 12Department of Pathology, Dunedin School of Medicine, University of Otago, New Zealand; 13Centre for Neurodevelopmental Disorders, School of Psychology, University of Birmingham, Birmingham, UK; 14Department of Child Psychiatry, and Kennedy-Krieger Institute, Johns Hopkins University School of Medicine, Baltimore, MD

Abstracts are provided in the order of presentation.

Natural History of Cornelia de Lange Syndrome

Antonie D. Kline

Harvey Institute for Human Genetics, Greater Baltimore Medical Center, Baltimore, MD

Significant progress has been made in recent years towards the clinical and molecular delineation of Cornelia de Lange syndrome (CdLS). This has allowed expansion of the clinical spectrum, observation about natural history into adulthood, and revision of the diagnostic criteria, more inclusive of milder cases. CdLS affects numerous body systems, all of which undergo changes with age. The gastrointestinal system is one of the most commonly involved, often with the greatest morbidity. Medical management recommendations have been clarified for patients of all ages.

Based on 58 individuals of adolescent or adult age, there appears to be some evidence for accelerated aging. Although variable, a distinct pattern of facial changes with aging is evident, and many individuals appear older facially than their years. Twenty percent have premature graying of the hair. In the gastrointestinal system, 14% have had Barrett esophagus starting in the late teen years. In the musculoskeletal system, 90% of a subgroup of 20 adult individuals was found to have decreased bone density and 35% to have moderate to profound osteoporosis. Three of five patients who underwent an echocardiogram had a small pericardial effusion, none with abnormal thyroid function. Two patients underwent procedures earlier than typically expected (cholecystectomy at age 19 years and transurethral resection of the prostate at age 39 years).

Behavioral issues increase during adolescence and also produce high morbidity. Eighty percent of individuals report some degree of sleep disturbance, in whom a higher prevalence of decreased nighttime sleep was found in those with gastroesophageal reflux and self-injury, and frequent nocturnal awakenings were positively correlated with anxiety. In the aging population, 70% of individuals report some degree of self-injury, as well as aggression in 40%, anxiety in 31%, obsessive-compulsive tendencies in 17%, and depression in 14%. At least 38% were found to have autistic-like behaviors.

Approximately 65% of patients with CdLS demonstrate a detectable mutation in the important embryonic genes NIPBL, SMC1A, or SMC3. Fifty-eight percent of the aging population have been found to have a mutation in one of the three genes. Of these, in NIPBL, 71% are positive in the severely involved group and all but one are nonsense, frameshift, or splice site mutations, and 50% are positive in the moderately and mildly involved groups with 44% missense mutations. The genes related to CdLS contribute to control of gene expression and embryonic development via the cohesin complex, involved in mitosis, DNA replication, and DNA repair. The premature aging may be explained by faulty epigenetic mechanisms in the regulation of postreplicative DNA repair, as theorized recently 1–3.

1Peng JC, Karpen GH. 2008. Curr Opin Genet Dev 18:204–211 (Epub March 26, 2008).

2Ball AR Jr, Yokomori K. 2008. Bioessays 30:5–9.

3Liang B, et al. 2007. Curr Biol 17:1432–1437 (Epub Aug 9, 2007).

A Drosophila Model for Cornelia de Lange Syndrome

Dale Dorsett

Department of Biochemistry and Molecular Biology, Saint Louis University School of Medicine, St. Louis, MO

Cornelia de Lange syndrome (CdLS) is characterized by slow growth, mental retardation, and structural birth defects in limbs and organs. CdLS is caused by mutations affecting proteins required for sister chromatid cohesion (Krantz et al., 2004; Tonkin et al., 2004; Musio et al., 2006; Deardorff et al., 2007). Current data, however, argue that the developmental deficits are largely caused by changes in gene expression, and not by defects in chromatid cohesion.

About half of CdLS patients have heterozygous loss-of-function mutations in Nipped-B-Like (NIPBL), and some 5% have missense or small in-frame deletion mutations affecting the Smc1 and Smc3 subunits of the cohesin complex. Cohesin is the glue that holds sister chromatids together, and NIPBL is required for cohesin to bind to chromosomes. The leading idea is that cohesin, which has a ring-like structure, encircles the sister chromatids to hold them together.

We discovered Nipped-B, the Drosophila homolog of NIPBL, in a genetic screen for factors that regulate the expression of the cut and Ultrabithorax (Ubx) homeobox genes (Rollins et al., 1999). Heterozygous loss-of-function Nipped-B mutations reduce cut and Ubx expression. The same mutations are lethal and cause severe sister chromatid cohesion defects when homozygous. Heterozygous mutants, however, do not cause cohesion defects.

Reducing the dosage of cohesin subunits has the opposite effect as Nipped-B mutations, and increases cut gene expression in the developing wing margin, leading us to hypothesize that cohesin inhibits cut gene expression, most likely by interfering with communication between the promoter and a distant wing margin enhancer (Rollins et al., 2004; Dorsett et al., 2005). We hypothesize that Nipped-B dynamically controls cohesin binding to regulate cut activation.

To gain further insights into how Nipped-B and cohesin regulate genes, we used chromatin immunoprecipitation with tiled microarrays to map their binding genome-wide in cultured cells (Misulovin et al., 2008). Nipped-B and cohesin co-localize genome wide, and they bind preferentially to active genes. There is also significant overlap with RNA polymerase II binding. We identified nearly 500 genes that bind cohesin, most of which are actively transcribed. About 100 of these genes bind cohesin in only one cell type, and there is a 20-fold higher probability that there is elongating RNA polymerase on the gene when cohesin binds.

One of the active genes that binds cohesin through its entire transcribed region encodes the steroid hormone receptor (EcR). Recently, Schuldiner et al. (2008) discovered that EcR is underexpressed in postmitotic neurons that lack cohesin, causing an axon pruning defect. This suggests that cohesin directly facilitates EcR transcription, which is opposite to its effect on cut.

We currently theorize that transcription unravels chromatin to fit into the 35 nm diameter of the cohesin ring, and that cohesin then has multiple effects on transcription, including interfering with enhancer–promoter interactions and facilitating elongation by RNA polymerase. We also posit that Nipped-B dynamically regulates cohesin binding to mitigate the negative effects and facilitate the positive effects of cohesin on transcription.

Deardorff et al. 2007. Am J Hum Genet 80:485–94.

Dorsett et al. 2005. Development 132:4743–53.

Krantz et al. 2004. Nat Genet 36:631–5.

Misulovin et al. 2008. Chromosoma 117:89–102.

Musio et al. 2006. Nat Genet 38:528–30.

Rollins et al. 1999. Genetics 52:577–93.

Rollins et al. 2004. Mol Cell Biol 24:3100–11.

Schuldiner et al. 2008. 14:227–38.

Tonkin et al. 2004. Nat Genet 36:636–41.

Roberts Syndrome Molecular Mechanism: Insights From Biochemical and Cellular Phenotype Analysis of ESCO2 Mutations

Hugo Vega, Miriam Gordillo, Ethylin W. Jabs

Department of Genetics and Genomic Sciences, Mount Sinai School of Medicine, New York, NY

Collaborators: Alison H. Trainer (Newcastle University), Fajian Hou (SMC, University of Texas), Hui Zou (SMC, University of Texas), Hülya Kayserili (Istanbul University), Flemming Skovby (Copenhagen University Hospital), Raoul C.M. Hennekam (University College London and AMC University of Amsterdam), Maria L. Giovannucci Uzielli (University of Florence), Rhonda E. Schnur (Cooper University Hospital/Robert Wood Johnson Medical School), Sylvie Manouvrier (University Hospital, Lille), Francesca Forzano (Ospedali Galliera, Genova), Moritz Meins (Ruhr University of Bochum), Kalle O.J. Simola (Tampere University Hospital), Annick Raas-Rothschild (Hadassah Hebrew University Hospital), Roger A. Schultz (SMC, University of Texas), Lisa D. McDaniel (SMC, University of Texas), Norio Sakai (Osaka University), Keiichi Ozono (Osaka University), Koji Inui (Osaka University).

Roberts syndrome (RBS) is a developmental recessive disorder characterized by growth retardation, craniofacial abnormalities, and limb reduction. Cells from RBS patients display a characteristic cytogenetic defect consisting of premature separation of sister chromatids at pericentromeric regions and the long arm of the Y chromosome. RBS is caused by mutations in ESCO2, a gene that codes an acetyltransferase with homologs from yeast to humans. ESCO2 as well as other members of the family have been implicated in the establishment of sister chromatid cohesion.

To date, 26 different ESCO2 mutations have been associated with RBS. Most of these mutations result in complete or partial loss of the acetyltransferase domain of ESCO2 and only one missense mutation, ESCO2 W539G, was known. We have found that this missense mutation as well as a new ESCO2 missense mutation also located in the acetyltransferase domain result in decreased in vitro acetyltransferase activity. Using RBS cell lines bearing different ESCO2 mutations we have found that the decreased proliferation capacity observed in RBS is associated with cell death. In addition, we found evidence that RBS cells present a delay in the entrance to mitosis. Transitory delays of the cell cycle in RBS cells might be related to activation of checkpoints due to alterations of cellular processes that depend on accurate sister chromatid cohesion. Our observation that different types of mutations including the ESCO2 W539G mutation produced equivalent cellular defects suggests that loss of acetyltransferase activity contributes importantly to the pathogenesis of RBS.

Pediatric Dental Findings in Cornelia de Lange Syndrome

Joseph Hylton, DDS1, Richard Mungo, DDS2, Rochelle Lindemeyer, DMD3, Antonie Kline, MD4, Douglas Clemens, DMD5

1Pediatric Dentistry, University of Chicago College of Dentistry, Chicago, IL; 2Huntingon Beach, CA; 3The Children's Hospital of Philadelphia, Philadelphia, PA; 4Harvey Institute for Human Genetics, Greater Baltimore Medical Center, Baltimore, MD; 5Cross Keys Dental Associates, Baltimore, MD

Cornelia de Lange syndrome (CdLS) is a relatively uncommon congenital syndrome that has a characteristic phenotype. To date, little has been written about the dental findings in CdLS. The purpose of this study was to evaluate the most common dental findings in CdLS as reported by parents and by examination. Due to the high prevalence of gastroesophageal reflux disease (GERD) in CdLS, subjects were examined for the presence of lingual erosions of enamel. Intraoral exams were performed on 39 subjects with CdLS who attended a CdLS World Federation international meeting with their parents. Additionally, parents and caregivers of 129 subjects with CdLS completed a survey on the dental history of their children with CdLS. These surveys were completed at the meeting as well as mailed out to families who requested participation after solicitation through the United States CdLS Foundation newsletter.

The age range of subjects who received a dental examination was from 1 to 42 years of age with an average age of 13.5. The results of the examination showed two subjects (5%) with lingual erosions of enamel. Other findings from examination were a low maxillary frenum (38%), delayed eruption of teeth (56%), and over-retained primary teeth (15%). The age range of subjects whose parents completed a survey was from 1 to 49 years of age with an average age of 16.5.

The results of the survey showed the incidence of cleft palate was 10%, with submucous cleft palate comprising 24%, and surgical repair 12%. Regarding dental treatment, the survey results showed that 65% of subjects either required sedation or general anesthesia to have dental treatment completed, and 25% of subjects have had four or more sedations or general anesthesia for dental treatment. Eighty-seven percent of caregivers report that they help brush their child's teeth, with 36% reporting their child being somewhat or completely uncooperative. Additionally, 49% of subjects are reported to at least sometimes store food in their mouth after eating.

For dental providers who treat patients with CdLS, it becomes important to understand the oral findings that are more common to the syndrome, such as cleft palate, submucous clefts, delayed eruption, and over-retention of primary teeth. It may often be necessary to treat patients in a hospital environment due to complex medical histories and lack of patient cooperation. Future dental studies for CdLS will continue to prove of value to parents and dental providers alike.

Genomic Consequences of Cohesin Dysfunction in CdLS: Identification of Dysregulated Gene Expression With Phenotypic Effects

Jinglan Liu1, Zhe Zhang2, Dan Wilson5, Mani Kaur1, Matt Deardorff1,4, Dinah Clark1, Eric Rappaport3, Michael Morrow5, Stephany Tandy1, Ian Krantz1,4

1Division of Human Genetics, 2Bioinformatics Core, 3NAPCORE, The Children's Hospital of Philadelphia, 4The University of Pennsylvania School of Medicine, 5Xceed Molecular

Our laboratory has identified mutations in the NIPBL gene as a major cause of Cornelia de Lange syndrome (CdLS). Up to 60% of individuals with CdLS have mutations in NIPBL, and another 5% are due to mutations in SMC1A or SMC3. These genes encode proteins that are involved in an important cellular mechanism, the cohesin pathway, that plays a critical role in sister chromatid cohesion. CdLS was the first human developmental disorder linked to this important cellular pathway. A growing body of evidence suggests that NIPBL (and cohesin) has a role beyond sister chromatid cohesion and is in fact a “master switch” that regulates the expression of many other genes. In order to understand the exact mechanism by which alteration of NIPBL or other cohesin members leads to the structural and cognitive differences seen in CdLS, it is essential to identify the genes whose expression is directly regulated by NIPBL and cohesin in the human genome. We hypothesize that through the identification of specific genes that are dysregulated by NIPBL/cohesin dysfunction we will isolate candidate genes that will be causally related to the individual features that, in constellation, make up the features seen in CdLS. Identification of these direct effectors of cohesin dysfunction will be the first step towards developing novel therapeutic strategies for CdLS. To address this issue we will perform an interrelated series of genome-wide studies including mRNA expression array, ChIPSeq, and microRNA expression array. Thus far, we have finished a stepwise expression array study. Array-based expression profiling was first developed from lymphoblastoid cell line (LCL) samples of 16 severely affected CdLS patients, and 17 matched healthy subjects as control, and further validated with 6 additional samples. We identified a unique list of 339 genes with NIPBL ranked as the most significantly changed gene. Target array analysis using a panel of 32 genes has successfully classified 114 samples with different clinical presentations, and 23, 10, and 3 gene signatures were established. The genotype–phenotype correlation of CdLS is reflected by the identified disease specific profile. The listed genes may represent the genetic fingerprints of developmental pathways that are dysregulated in order to manifest as the various CdLS phenotypes. In order to differentiate direct effects of NIPBL, we will conduct chromatin immunoprecipitation sequencing (ChIPSeq) to create a high-resolution map of direct interactions between NIPBL/cohesin and DNA. We are currently testing the specificity of different commercial antibodies. Combining the expression data above with the ChIPSeq data, we will be able to identify direct targets of NIPBL. At the same time, preliminary experiments are being performed to evaluate the expression levels of several developmentally important microRNAs in CdLS LCLs and fetal fibroblasts. The results of these three genome wide experiments will allow us to establish a molecular landscape for CdLS. Identification of these NIPBL direct targets and understanding their functions will allow insight into the pathophysiology behind CdLS, and will hopefully provide the initial proof-of-principle for potential drug targets or treatment options for this disorder.

The Diagnostic Laboratory Experience for Testing for Cornelia de Lange Syndrome

Melissa Dempsey, Soma Das

Department of Human Genetics, The University of Chicago, Chicago, IL

Molecular diagnostic testing for Cornelia de Lange syndrome (CdLS) began at the University of Chicago in July 2006 in collaboration with Dr. Ian Krantz at the Children's Hospital of Philadelphia and the Cornelia de Lange Syndrome Foundation, Inc. The molecular test in our laboratory initially comprised full gene sequencing of the NIPBL gene, followed by the addition of full gene sequencing of the SMC1A gene in February 2007 and the addition of deletion/duplication testing of the NIPBL gene in July 2007. To date, our laboratory has received over 200 samples for testing and abnormalities have been identified in 36 patients. In addition to diagnostic testing, prenatal testing has been performed in five families.

Full gene sequencing of the NIPBL gene has been performed on 176 patients with mutations identified in 34 patients. Mutations have thus been identified in approximately 19% of patients sent to our laboratory for this testing. This low percentage is likely due to the wide spectrum of patient samples received for testing, many of which may have a questionable diagnosis of CdLS. The majority of identified mutations have been novel and have not been previously described. They occur throughout the NIPBL gene and include a wide spectrum of mutation types. We have also identified 16 patients with “variants of unknown significance.” These are missense or intronic changes in the gene that are not clearly deleterious and have not been previously reported in the literature. Intragenic deletion and duplication analysis of the NIPBL gene, performed by multiplex ligation-dependent probe amplification (MLPA), has been completed on 49 patients and two deletions have been identified. A deletion of exons 13–14 was detected in a 5-month-old with growth retardation and synophrys, and a deletion of exons 46–47 was detected in a 5-year-old with characteristic facial features, developmental delay, and conductive hearing loss.

Full gene sequencing of SMC1A, the second CdLS gene identified, has been performed on 50 samples sent to our laboratory. To date, we have identified only one mutation in this gene in a 2-year-old with characteristic facial features, developmental delay, intestinal malrotation, and kidney abnormalities. We have also identified a small number of “variants of unknown significance” within the SMC1A gene. This result is in line with previous observations that mutations of the SMC1A gene account for a smaller proportion of patients with CdLS compared to the NIPBL gene.

We are also collecting clinical information on patients whose samples are sent to our laboratory by using a standard questionnaire. Clinical and genetic testing information is entered into a database that is being shared with Dr. Krantz and colleagues. Our collaboration with Dr. Krantz and the CdLS Foundation over the past 2 years has been very beneficial and rewarding. We have communicated with Dr. Krantz about unusual findings and have referred patients in whom no genetic abnormalities were identified to his research study. We look forward to setting up additional testing for CdLS as new genes are identified.

Genomewide Copy Number Analysis in Cornelia de Lange Syndrome

Matthew Deardorff1, Maninder Kaur1, Micah Berman1, Laura Conlin1, Xiaowu Gai2, Juan Perin2, Tamim Shaikh1, Hakon Hakonarson3, Laird Jackson4 and Ian Krantz1

1Divisions of Human Genetics, 2Bioinformatics and the 3Center for Applied Genomics, Children's Hospital of Philadelphia, PA and 4Drexel School of Medicine, Philadelphia, PA

Genome-wide CNV analysis is in the process of reinventing the field of cytogenetics. It has revealed significant degrees of variation even in normal individuals despite using very high confidence thresholds. The use of these same arrays for gene discovery in rare Mendelian disorders promises great potential for finding causative abnormalities. However, it proves to be even more challenging as lower confidence thresholds are used to increase the sensitivity of identifying small causative changes. Several troublesome issues include identifying effective strategies to validate large numbers of small variations and how to effectively screen considerable amounts of candidate regions for mutations in sizeable patient populations.

NIPBL, SMC1A, and SMC3, three genes involved in sister chromatid cohesion, are mutated in 65% of patients. However, we have yet to elucidate a cause of CdLS in 35% of patients. To facilitate identification of genes that may cause CdLS, we have used genome-wide CNV analysis. Illumina HapMap550K SNP arrays were used to screen 269 CdLS probands for whom causative mutations had not been identified. Copy number analysis revealed 8,932 potential variants, of which 5,191 were 5 SNPs or less. Computational and manual curation of these variants was performed to prioritize potential causative deletions that could represent additional causative genes.

We have identified a number of large chromosomal abnormalities in individuals with phenotypic overlap with CdLS, but who clearly did not meet clinical criteria. We also identified deletions in five patients that included NIPBL and were validated using MLPA. There were no deletions that included SMC1A and SMC3, consistent with previous findings of only missense mutations in these genes. Several other genes implicated in sister chromatid cohesin are located in putative CNVs. Because no MLPA kits were available for these genes, we used a combination of FISH, quantitative PCR, and MLGA (multiple ligation-dependent genome amplification) for validation.

We have identified a number of potential candidate regions and are in the process of validating copy-number variations, determining if these abnormalities are de novo and are utilizing novel methods to pursue mutation screening in these large numbers of patients that have no known mutation.

Parental Report on Development and Coordination of Care With the Educational and Medical System of Children with CdLS

Mary Beth Bruder & Cristina Mogro-Wilson

Center for Disabilities, University of Connecticut, Farmington, CT

Background: The CdLS population has different educational and medical needs. The CdLS Foundation and the Center for Disabilities at the University of Connecticut collaborated on a study to increase the amount of knowledge on children affected by CdLS, their development, and their experience with the educational and medical system. The information gathered will inform parents and other providers about this unique population about whom little is currently known and establish a baseline of knowledge on the range of services children are receiving, and if these services are meeting their needs.

Methodology: An online survey was completed by 224 parents of a child age birth to 21 with CdLS. Recruitment was conducted via newsletter and email in June to October 2007.

Results: The average age of the child was 9 years, ranging from 2 months to 20 years of age. Just over half of the children were female (61%). Almost half of the children were diagnosed with CdLS at birth (42%). Seventy-two percent of parents suspected that there was something wrong before the diagnosis. Within this group, most believed something was wrong because of their child's low birth weight (62%), small stature (58%), slow growth (57%), poor feeding (52%), and small head size (52%). Parents most commonly noted that their child experiences gastroesophageal reflux disease (63%), feeding problems (43%), hearing loss (42%), and dental problems (41%).

Almost half of parents (45%) noted that in the past 12 months there was a time that their child needed care coordination among different health-care providers and services. Of those who stated that care coordination was needed, 28% reported that they did not receive all the coordination that was needed. Overall, parents indicated that they were not especially satisfied with the help they have received in coordinating their child's care: 32% were very satisfied, 27% somewhat satisfied, 10% somewhat dissatisfied, 10% very dissatisfied, and 21% did not know how satisfied they were. There were only 13% of parents with a child in Early Intervention who noted that staff worked closely with medical professionals who were treating their child; 21% of parents with a child in preschool, 14% of elementary school parents, and 7% of middle and high school parents noted this coordination between school staff and medical professionals, indicating low coordination of care across the age groups.

The most commonly provided service that a child currently receives is special education (84%), followed by speech and language therapy (78%), occupational therapy (65%), and physical therapy (63%). Parents responded that they are happy with how their life is going (44%), their family likes spending time together (83%), they enjoy relaxing with family members (73%), and that their family has the best life possible (34%).

Conclusions: Parents' reports of medical, educational, and family life will be presented. Children and families affected by CdLS are complex and varied. Discussion on the range of services, needs, and experiences and the development of useful tools and guidelines for professionals and parents will be discussed.

Database and Worldwide Communication in Cornelia de Lange Syndrome

Laird Jackson

Department of Genetics, Internal Medicine and Obstetrics/Gynecology, Drexel University College of Medicine, Philadelphia, PA

A database containing minimal diagnostic observations of Cornelia de Lange syndrome (CdLS) individuals was initiated in the 1990s. It proved useful in initial reporting of a relatively large CdLS cohort that provoked interest in the syndrome. An expanded database has been utilized recently for pursuit of data related to the causes of death of the individuals; including obtaining autopsy records. The original database was transferred into a locally held database at the Children's Hospital of Philadelphia to subsequently accommodate molecular and further diagnostic data. Simultaneously the database, without molecular data was converted to an on-line resource. It now contains some 1,555 records, most of which are sparsely populated with clinical diagnostic data but have pertinent data relating to the cause of death of those individuals who are known to be deceased. There are nearly 400 listed deaths now and approximately 280 of them have listed causes. Many of these are simply reported by the parent and there has been insufficient follow-up to document their reports with either actual medical records or further specific information obtained from the parent(s) by telephone interview from a medical professional. Some 40 autopsy records have been obtained or are anticipated by mail.

The National Library of Medicine and the National Center for Biological Information has expressed a preliminary interest in working with our database as a potential model for reasonably rare diseases that could bring clinical and molecular information together in an accessible format for further research. Collaboration would allow storage of the database on a secure server (it is on a secure server now but not as secure and reliable for the long term as a government server) with dedicated professionals to spruce up the format of the database and contribute ideas to streamline its use and access. It may be possible to file the database in several iterations for different uses. For example, there could be a portion with open identification of the participating families but limited login access to registered members of the international CdLS Federation. Such a portion could be used by the Federation members for contact and membership surveys.

A potential price would be that the registered members would be asked to voluntarily contribute medical and laboratory testing information on their affected family member which would then be held in both a de-identified portion for research access; and an identified portion again accessible only to medical members of the Scientific Advisory Committee of the Federation—with registered identities and affiliations and controlled logins. Our advantage would be the professional web-site help, which we could obtain; plus the use of dedicated and controlled server storage.

Mice Lacking Sister Chromatid Cohesion Protein PDS5b Exhibit Developmental Abnormalities Reminiscent of Cornelia de Lange Syndrome

B. Zhang, S. Jain, H. Song, M. Fu, R.O. Heuckeroth, J.M. Erlich, P.Y. Jay, J. Milbrandt

Department of Genetics, Neurology, Pathology, and Immunology and Medicine, Washington University School of Medicine, St Louis, MO

PDS5B is a sister chromatid cohesion protein that is crucial for faithful segregation of duplicated chromosomes in lower organisms. Mutations in cohesion proteins are associated with the developmental disorder Cornelia de Lange syndrome (CdLS) in humans. To delineate the physiological roles of PDS5B in mammals, we generated mice lacking PDS5B (APRIN). Pds5B-deficient mice died shortly after birth. They exhibited multiple congenital anomalies, including heart defects, cleft palate, fusion of the ribs, short limbs, distal colon aganglionosis, abnormal migration and axonal projections of sympathetic neurons, and germ cell depletion, many of which are similar to abnormalities found in humans with CdLS. Unexpectedly, we found no cohesion defects in Pds5B(−/−) cells and detected high PDS5B expression in postmitotic neurons in the brain.

These results, along with the developmental anomalies of Pds5B(−/−) mice, the presence of a DNA-binding domain in PDS5B in vertebrates and its nucleolar localization, suggest that PDS5B and the cohesin complex have important functions beyond their role in chromosomal dynamics.

Insights Into Cornelia de Lange Syndrome From the Nipbl-Mutant Mouse

Shimako Kawauchi1,2,3,4, Rosaysela Santos1,3,4, Martha E. Lopez-Burks2,3, Michelle P. Hoang2,3, Leonard M. Kitzes1,4, Taotao Lao6, Mark S. Lechner6, Benedikt Hallgrimsson7, Jeremy A. Daniel8, André Nussenzweig8, Arthur D. Lander2,3,5, and Anne L. Calof1,2,3,4,5

1Department of Anatomy and Neurobiology, 2Department of Developmental and Cell Biology, 3Developmental Biology Center, 4Center for Hearing Research, and 5Center for Complex Biological Systems, University of California, Irvine, CA; 6Department of Bioscience and Biotechnology, Drexel University, Philadelphia, PA; 7Department of Cell Biology and Anatomy, University of Calgary, Calgary, Alberta, Canada; 8Experimental Immunology Branch, NCI/NIH, Bethesda, MD

Cornelia de Lange Syndrome (CdLS: OMIM#122470) is a multisystem birth defects syndrome with highly variable clinical presentation, ranging from subtle dysmorphology to conditions incompatible with postnatal life. Recent studies have indicated that heterozygous mutations in Nipped-B-like (NIPBL), the human ortholog of the Drosophila Nipped-B gene, are causative for at least 50% of cases of CdLS. Nipbl and its relatives encode cohesin regulatory proteins, which are thought to be involved in sister chromatid cohesion and DNA repair. Studies in Drosophila have revealed that Nipped-B also plays roles in developmental gene expression, possibly by influencing long-range promoter–enhancer interactions.

As a first step towards analysis of Nipbl gene function and understanding the molecular and cellular etiology of CdLS, we have developed a mouse model for CdLS, using mice heterozygous for a gene-trap mutation in Nipbl (Nipbl564/+) mice. Over 1,000 progeny of the original chimeras and their offspring, spanning 5 generations out-crossed onto a CD-1 (Swiss Webster) background, have now been studied. The majority of Nipbl heterozygous mice are not viable, with approximately 75% dying within a few weeks of birth. Those that do survive are significantly smaller than their wild-type littermates, regardless of gender, and this phenotype is stable across all generations. Other common defects involve heart (primarily atrial septal defects), bone (delayed ossification), and craniofacial structure. Craniofacial defects, significant in the entire population of Nipbl564/+ mice that survive to maturity, include upward deflection of the snout, foreshortening of the skull and a significant reduction of endocranial volume, which are all distinctive craniofacial features of CdLS. Neurological deficiencies were also observed, including defects in cerebellar morphology, impaired hearing, and behavioral abnormalities. Thus, Nipbl564/+ animals show multiple similarities to individuals with CdLS, including both morphological alterations and disturbances in neurological function.

In order to explore the molecular basis for the phenotypes mediated by reduction in Nipbl activity in Nipbl564/+ mice, we performed microarray analyses on a variety of embryonic tissues. We found significant changes in transcript levels for numerous genes in microarray analyses performed on MEFs (mouse embryonic fibroblasts; >30 transcripts) and embryonic brain (>700 transcripts). Importantly, although changes in transcript levels are significant statistically and reproducible over large numbers of individual samples, most changes are relatively small in magnitude, usually less than twofold. Included in this category is the Nipbl transcript itself, which is ubiquitously expressed and which was found to be downregulated by ∼30% in both adult and embryonic tissues derived from Nipbl564/+ mice. In contrast, we found no evidence for elevated PSCS (precocious sister chromatid separation) in any cells derived from Nipbl564/+ mice (fibroblasts, embryonic stem cells, and splenocytes), suggesting that cohesion defects do not occur, or if they do, are very subtle. Altogether, our data support the idea that the phenotypes observed in the Nipbl heterozygous mouse, and by extension in CdLS, are the consequence of subtle dysregulation of the expression of numerous genes, which results from a decrease in Nipbl function.


This work was supported by NIH grant P01HD052860.

Modeling Early Developmental Defects in Cornelia de Lange Syndrome Using the Zebrafish

Akihiko Muto1,2, Trevor Hoffman1,2, Shimako Kawauchi2,3,4, Thomas F. Schilling1,2,5, Arthur D. Lander1,2,5 and Anne L. Calof2,3,4,5

1Department of Developmental and Cell Biology, 2Developmental Biology Center, 3Department of Anatomy and Neurobiology, 4Center for Hearing Research, and 5Center for Complex Biological Systems, University of California, Irvine, CA 92697

Nipbl (the ortholog of Drosophila Nipped-B) has been identified as a gene in which heterozygous mutations are associated with at least 50% of cases of Cornelia de Lange Syndrome (CdLS). Nipbl and its relatives encode cohesin regulatory proteins, which are thought to be involved in sister chromatid cohesion and DNA repair. Studies in Drosophila have revealed that Nipped-B also plays roles in developmental gene expression, possibly by influencing the strength of long-range promoter–enhancer interactions. Our recent studies in Nipbl+/− mutant mice strongly support the notion that widespread transcriptional dysregulation underlies the developmental abnormalities in CdLS. However, the precise targets of Nipbl and the mechanism(s) by which it regulates gene expression remain unclear. To address these questions, we have chosen the zebrafish (Danio rerio) as a model system in which gene expression levels are easily manipulated by injecting morpholino antisense oligonucleotides (MO) and/or in vitro synthesized mRNA. Zebrafish have two Nipbl orthologs (designated zNipbl-1 and zNipbl-2), and both genes are expressed ubiquitously during embryogenesis. Predicted amino acid (a.a.) sequences reveal that the N-terminal 200 a.a. and the C-terminal 1,800 a.a—the latter of which contains multiple HEAT domains—are highly conserved (66–80% identical) among zebrafish and mammalian proteins. MO knockdown in fish embryos, using MOs directed against both zNipbl1 and zNipbl2, causes a variety of developmental defects, including abnormalities in heart and gut development, which overlap with abnormalities seen in CdLS. In Nipbl-MO-injected embryos (zNipbl-morphants), migration of cardiac progenitors is delayed, leading to cardia bifida (split heart) in about 15% of morphants. In addition, morphants exhibit reduced expression of an endodermal marker, foxA3, which correlates with later mis-localization of visceral organs (liver and pancreas) as well as abnormal gut looping. Such abnormalities of visceral organs are also common findings in CdLS. Consistent with the observations of aberrant morphology of heart and visceral organs in zNipbl-morphants, microarray and in situ hybridization analyses of gene expression reveals reduced expression of other endodermal genes (gata5, sox32, sox17, and foxA2). Although these endodermal genes are all known to be regulated by Nodal signaling, expression of Nodal genes, as well as the Nodal mesodermal target genes no tail and goosecoid, are not affected by zNipbl-MO, suggesting that at least some endodermal genes are directly regulated by Nipbl. These results suggest a mechanism by which Nipbl regulates endodermal development and a hypothesis for the etiology of heart and gut abnormalities observed in individuals with CdLS. We are further testing this hypothesis by examining endodermal gene expression in Nipbl-heterozygous mutant mice. Moreover, we are also examining functional relationships between Nipbl and cohesin on gene expression by simultaneously manipulating cohesin expression in zebrafish embryos.


This work was supported by NIH P01-HD052860.

Cohesin-Dependent Regulation of Gene Expression in Zebrafish

Julia A. Horsfield1, Sasha H. Anagnostou2, Jimmy Kuang-Hsien Hu2*, Kitty Hsiao Yu Cho2, Robert Geisler3, Graham Lieschke4, Kathryn E. Crosier2, Philip S. Crosier2

1Department of Pathology, Dunedin School of Medicine, University of Otago, New Zealand; 2Department of Molecular Medicine and Pathology, School of Medical Sciences, The University of Auckland, Auckland, New Zealand; 3Max Planck Institute fur Entwicklungsbiologie, Tubingen, Germany; 4Cancer and Haematology Division, The Walter and Eliza Hall Institute of Medical Research, Parkville, Victoria, Australia

*Present address: Department of Genetics, Harvard Medical School, 77 Avenue Louis Pasteur, Boston, MA 02115.

Cohesion or “pairing” of sister chromatids is essential to ensure each daughter cell receives the correct number of chromosomes during cell division. Sister chromatid cohesion is mediated by cohesin, a multimeric protein complex comprising at least four subunits: Rad21/Scc1, SA/Scc3, Smc1, and Smc3. We identified the cohesin subunit rad21 in a zebrafish forward genetic screen for positive regulators of runx1.

Runx transcription factors are critical for determining cell fate in many cell types, and maintaining balanced levels of Runx protein is essential for normal development. Deregulated expression of runx genes leads to cancers and developmental disorders. 12-somite zebrafish embryos mutant for rad21 or morphant for Smc3 lack hematopoietic expression of runx1, and fail to express runx3. Partial failure of blood development in rad21 mutants is rescued by microinjection of runx1 mRNA. Significantly, loss of just one copy of the rad21 gene caused a reduction in runx1 transcription and reduced expression of neuronal genes ascl1a and ascl1b. rad21 mutant embryos also have neurodevelopmental abnormalities, consistent with the idea that neuronal development is particularly sensitive to loss of cohesin function.

Our data raise the interesting possibility that CdLS may be contributed to by a reduction in RUNX gene expression in embryogenesis, as a result of reduction in cohesin function. CdLS patients present with neurodevelopmental, gastrointestinal, and skeletal abnormalities. The development of each of these systems depends on the proper regulation of RUNX proteins which are themselves dose sensitive in function. In addition, our findings validate the use of zebrafish as a model to study chromatid cohesion defects and their impact on development.

Extending the Behavioral Phenotype of Cornelia de Lange Syndrome

Chris Oliver1, Jo Moss2, Lisa Collis1, Jenny Sloneem1, Kate Arron1 and Sarah Gorniak1

1Centre for Neurodevelopmental Disorders, School of Psychology, University of Birmingham, Birmingham, United Kingdom; 2Department of Psychology, Institute of Psychiatry, London, United Kingdom

In a series of cross-sectional, case–control, single case, and longitudinal studies using survey, direct observation and behavioral and psychometric assessment methods, we have sought to document the behavioral phenotype of Cornelia de Lange syndrome (CdLS). The current behavioral database contains assessment information on over 150 children and adults with CdLS with a number of participants (n ≈ 60) being followed up over an 8-year period. Behavioral data are available on eight other genetic syndromes and comparison with these groups reveals a profile for CdLS of low mood (possibly downstream from anxiety) with a comparatively high prevalence of autism spectrum disorder (over 50%).

In adults there is a significant increase in levels of impulsivity, in contrast to other syndromes, and approximately 13% of adults will also show behavioral correlates of clinically significant low mood. In our observational assessments of 54 children and adults with CdLS in a case\control comparison study, clinically significant self-injury was not significantly more prevalent in CdLS than a matched control group but was observed in the majority of participants with CdLS. Other forms of challenging behavior were significantly less likely to be seen in CdLS.

The Behavioral Phenotype of Cornelia de Lange Syndrome in Relation to Autism

Marco Grados, Siddharth Srivastava, Colleen Landy, Bennett Clark, Jeffrey Duong, Antonie Kline

John Hopkins University School of Medicine, Baltimore, MD

Background: Cornelia de Lange Syndrome (CdLS) is a developmental disorder characterized by growth retardation, mental retardation, and behavioral difficulties including autistic features. The aim of the research is to “characterize the behavioral phenotype” in relation to autism.

Methods: Forty-one children with CdLS (5–17 years) were evaluated by: CdLS diagnostic checklist (Kline), Childhood Autism Rating Scale (CARS), Vineland Adaptive Behavior Scales (VABS), and Dunn Sensory Profile (Dunn).

Results: The mean age was 11.4 ± 3.8 years, and the CARS score showed that over 2/3 could be classified with a wide range of severity of autism features. Based on the CARS, there were 7 (18%) subjects with no autism, 17 (41%) subjects with mild autism, and 17 (41%) with severe autism. There was a significant association of higher autism scores with lower VABS (adaptive functioning) scores. Autism features were also predicted by Dunn tactile sensitivity (P < 0.001) and auditory sensitivity (P = 0.001) measures. Among physical features, autism scores significantly associated with the presence of 2,3 toe syndactyly (P = 0.003) and proximally placed thumbs (P = 0.04), even when controlling for age of the patient and adaptive measures. Both of these features can be associated with limb sculpting and a dose–response effect was noted: low autism with no limb sculpting deficits, medium autism score, and highest autism score with either limb sculpting deficit.

Conclusions: A subgroup of children with CdLS will have autistic features. Autistic features are significantly associated with older age, increased auditory and tactile hypersensitivity, and presence of limb sculpting deficits.

Cornelia de Lange Syndrome and Cohesin Dysfunction: Insights Into a Novel Developmental Mechanism

M.A. Deardorff, J. Liu, M. Kaur, E. Loy, D. Michelis, R. Feldman, E. Kauvar, D. Clark, L. Jackson, I.D. Krantz

The Children's Hospital of Philadelphia, Philadelphia, PA

The cohesin proteins compose an evolutionarily conserved complex whose fundamental role in chromosomal cohesion and coordinated segregation of sister chromatids has been well characterized across species. Recently, regulators and structural components of cohesin have been found to surprisingly cause specific human developmental disorders (collectively termed “cohesinopathies”) when mutated. Mutations in NIPBL, the vertebrate homolog of the yeast sister chromatid cohesion 2 (Scc2) protein, a regulator of cohesin loading and unloading, are responsible for approximately 50% of cases of Cornelia de Lange syndrome (CdLS). Mutations in the cohesin structural components SMC1A and SMC3 were also found to result in CdLS as well.

CdLS is a multisystem developmental disorder with classic features of characteristic facial dysmorphia, upper extremity malformations, hirsutism, cardiac defects, growth and cognitive retardation, and gastrointestinal abnormalities that display a wide spectrum of clinical severity. A mild form of CdLS has been consistently reported, however, it had not been clear if this is a distinct etiologic entity from classic CdLS or truly a mild manifestation. Molecular testing for cohesin genes has resulted in the identification of individuals with very subtle features of CdLS bordering on apparent isolated mental retardation. Mutations in another cohesin regulator, ESCO2, have been found to result in Roberts syndrome (RBS) and SC phocomelia. Roberts syndrome is a recessively inherited multisystem disorder with craniofacial, limb, cardiac, other systemic abnormalities, and neurocognitive dysfunction. While there is some overlap between Roberts syndrome and CdLS they are easily distinguishable diagnoses. Other developmental disorders have also recently been found to result from cohesin dysfunction.

This session will (1) review the mutation spectrum and genotype-phenotype correlations seen in CdLS, (2) outline possible mechanisms by which this disruption leads to the variable phenotypes, and (3) review novel directions in understanding additional etiologies for, and the pathogenesis of, CdLS towards the development of potentially novel therapeutic modalities.