Identification of Critical Regions and Candidate Genes for Cardiovascular Malformations and Cardiomyopathy Associated with Deletions of Chromosome 1p36

Cardiovascular malformations and cardiomyopathy are among the most common phenotypes caused by deletions of chromosome 1p36 which affect approximately 1 in 5000 newborns. Although these cardiac-related abnormalities are a significant source of morbidity and mortality associated with 1p36 deletions, most of the individual genes that contribute to these conditions have yet to be identified. In this paper, we use a combination of clinical and molecular cytogenetic data to define five critical regions for cardiovascular malformations and two critical regions for cardiomyopathy on chromosome 1p36. Positional candidate genes which may contribute to the development of cardiovascular malformations associated with 1p36 deletions include DVL1, SKI, RERE, PDPN, SPEN, CLCNKA, ECE1, HSPG2, LUZP1, and WASF2. Similarly, haploinsufficiency of PRDM16–a gene which was recently shown to be sufficient to cause the left ventricular noncompaction–SKI, PRKCZ, RERE, UBE4B and MASP2 may contribute to the development of cardiomyopathy. When treating individuals with 1p36 deletions, or providing prognostic information to their families, physicians should take into account that 1p36 deletions which overlie these cardiac critical regions may portend to cardiovascular complications. Since several of these cardiac critical regions contain more than one positional candidate gene–and large terminal and interstitial 1p36 deletions often overlap more than one cardiac critical region–it is likely that haploinsufficiency of two or more genes contributes to the cardiac phenotypes associated with many 1p36 deletions.


Introduction
Approximately 1 in 5000 newborns has a terminal deletion affecting chromosome 1p36, making it the most common telomeric deletion in humans [1]. Individuals with terminal 1p36 deletions share a common set of phenotypes that constitute the 1p36 deletion syndrome [2,3]. These phenotypes include typical craniofacial features, cognitive impairment, behavioral problems, seizures, postnatal growth deficiency, eye/vision problems, hearing loss, cleft palate, cardiovascular malformations, cardiomyopathy and renal anomalies.
The distal critical regions for most 1p36 deletion syndrome phenotypes have been determined to reside within approximately 4 Mb from the 1p telomere [4]. However, non-overlapping interstitial deletions involving the proximal region of 1p36, starting approximately 8 Mb from the 1p telomere, have also been shown to cause many of the phenotypes associated with distal 1p36 deletions including cognitive impairment, seizures, postnatal growth deficiency, cardiovascular malformations and cardiomyopathy [5].
Some individuals have deletions of both the distal and proximal regions of 1p36 [6]. In such cases, the additive effects of haploinsufficiency of genes within both of these regions may account for the observation that individuals carrying larger 1p36 deletions are more severely affected and exhibit more of the features typically associated with 1p36 deletions [7,8].
Cardiovascular malformations and cardiomyopathy are among the most acutely life-threatening conditions associated with both distal and proximal deletions of 1p36. Heterozygous loss-offunction mutations in PRDM16-a gene in the distal portion of 1p36 (2,985,355,185; hg19)-were recently shown to be sufficient to cause left ventricular noncompaction [9]. Similarly, a heterozygous loss-of-function mutation in ECE1 (21,543,740-21,672,034; hg19) has been identified in an individual with patent ductus arteriosus, a small subaortic ventricular septal defect, and a small atrial septal defect, suggesting that haploinsufficiency of this proximal 1p36 gene may be sufficient to cause cardiovascular malformations in humans [10]. However, the other genomic regions and dosage-sensitive genes that contribute to the cardiac phenotypes caused by 1p36 deletions have not been clearly defined. This paucity of information makes it difficult for physicians to create individualized medical plans and provide accurate prognostic information to patients and families affected by 1p36 deletions. This is true even though detailed information regarding the extent and location of an individual's 1p36 deletion can be readily obtained on a clinical basis. Defining the cardiacrelated regions and genes on 1p36 will not only allow physicians to provide improved medical care, but is also a prerequisite to understanding the molecular mechanisms that underlie the development of 1p36-related cardiovascular malformations and cardiomyopathy. Elucidating these molecular mechanisms may lead to the development of novel preventative and/or therapeutic interventions for cardiovascular disorders.
Using clinical and molecular cytogenetic data from individuals with isolated 1p36 deletions defined by array-based copy number analysis, we have identified five non-overlapping critical regions for cardiovascular malformations and two non-overlapping critical regions for cardiomyopathy on chromosome 1p36. A bioinformatic analysis of each of these cardiac critical regions revealed at least one positional candidate gene whose deletion may contribute to the development of cardiac phenotypes based on human studies and/or animal models. In some cases, haploinsufficiency of two or more such genes may contribute to the cardiac phenotypes associated with a 1p36 deletion. This is particularly likely in the case of large terminal and interstitial deletions that overlap more than one cardiac critical region.

Ethics Statement
These studies were performed under research protocols approved by the institutional review board of Baylor College of Medicine. All clinical investigations were conducted according to the principles expressed in the Declaration of Helsinki. Written informed consent was obtained from parents or guardians on behalf of study participants all of whom were minors/children at the time of enrollment.

Patient Identification and Accrual
Molecular and clinical data from individuals with isolated 1p36 deletions associated with cardiovascular malformation and/or cardiomyopathy were collected from four sources. The first source consisted of data from 33,566 de-identified individuals referred to the Medical Genetics Laboratories at Baylor College of Medicine for array-based copy number analysis. Only data from individuals with isolated 1p36 deletions and an indication for testing that included a cardiovascular malformation or cardiomyopathy were included in this study.
The second source consisted of data from patients with isolated 1p36 deletions confirmed by array-based copy number analyses who were accrued from a group of individuals receiving care at Texas Children's Hospital in Houston, TX, USA and by selfreferral. In these cases, informed consent was obtained under a protocol approved by the institutional review board of Baylor College of Medicine after which clinical and molecular cytogenetic data was obtained from a review of the medical record and correspondence with the individual's parents/family members.
The third source consisted of data from individuals with isolated 1p36 deletions associated with cardiac phenotypes who were listed in the Database of Chromosomal Imbalance and Phenotype in Humans using Ensembl Resources (DECIPHER; http://decipher. sanger.ac.uk/). In each case, representatives of the contributing institution-where consent for the submission of clinical and molecular cytogenetic data were obtained-were contacted and given an opportunity to provide further details of the phenotypes present in their patients. The final source was a literature review conducted to identify additional individuals whose cardiac phenotypes were caused by isolated 1p36 deletions that had been molecularly defined using array-based copy number analysis. Reports of individuals whose 1p36 deletions were not interrogated using such techniques were not included, since traditional cytogenetic techniques may miss copy number changes on other chromosomes, may fail to identify complex rearrangements involving deletion and duplications of various regions of 1p36, and rarely provide robust delineation of the boundaries of the deletion.
Individuals with complex rearrangements of 1p36 that included both deleted and duplicated regions, individuals with 1p36 deletions associated with unbalanced translocations, and individuals with deleterious deletions or duplications on other chromosomes were also excluded from this study.

Defining Cardiac Critical Regions
Individual cardiac critical regions on chromosome 1p36 were defined based on the smallest, non-overlapping deletion present within a single individual with a cardiovascular malformation and/or cardiomyopathy. In cases where both a minimal and a maximal deleted region were defined, the breakpoints of the maximal deleted region were used to define the critical region. This approach minimizes the risk that a cardiac-related gene located between the minimal and maximal deleted region will be erroneously excluded from the critical region.

Identifying Cardiac-related Genes within Critical Regions
Data regarding each gene located completely or partially within a cardiac critical region were downloaded into a searchable spreadsheet using GeneDistiller2 (http://www.genedistiller.org/). This publically-available online program allows the user to access information on genes within a defined interval from a variety of online sources. Information reviewed to identify cardiac-related candidate genes included: Online Mendelian Inheritance in Man (OMIM; http://www.omim.org/) reports, interaction data from the Search Tool for the Retrieval of Interacting Genes/Proteins (STRING; http://string-db.org/) and the Unified Human Interactome (UniHI; http://www.unihi.org/), and phenotype data from the Human Phenotype Ontology website (http://www. human-phenotype-ontology.org/), the Mouse Genome Database (MGD; http://www.informatics.jax.org/phenotypes.shtml) and the Gene Ontology Database (http://www.geneontology.org/). This information was augmented with manually curated data from recently published manuscripts cited in PubMed (http://www. ncbi.nlm.nih.gov/pubmed).

Results
To map and identify dosage-sensitive genes and genomic regions that may contribute to the development of cardiovascular malformations and/or cardiomyopathy, we identified individuals with these phenotypes who had isolated 1p36 deletions defined by array-based copy number detection techniques as described in the Materials and Methods. The location and extent of the 1p36 deletions identified in each individual are provided in Tables 1-6 along with a description of the individual's cardiovascular malformations (Tables 1-3) or cardiomyopathy (Tables 4-6). These deletions are also represented graphically in Figures 1 and 2.
Using this data, we defined five non-overlapping critical regions for cardiovascular malformations (Table 7, Figure 1). For clarity, each critical region has been numbered sequentially starting with the most distal region on chromosome 1p36. The first critical region is defined by a de novo ,2.4 Mb terminal deletion reported Table 3. Summary of individuals with isolated 1p36 deletions and cardiovascular malformations identified from the literature. by El-Hattab and colleagues in a patient with a ventricular septal defect, an atrial septal defect, patent ductus arteriosus and a rightsided aortic arch [11]. The second critical region is defined by an ,3.0 Mb interstitial deletion identified in Case 4 reported by Kang and colleagues in an individual with a perimembranous ventricular septal defect and a small septum secundum atrial septal defect [5]. In using Kang et al. Case 4 to define the second critical region, we note that the The third critical region is defined by an ,7.8 Mb interstitial deletion in Patient 7 who was referred to the Medical Genetics Laboratories at Baylor College of Medicine for copy number analysis with an indication for testing that included tetralogy of Fallot and a bicommissural aortic valve. The fourth critical region is defined by DECIPHER patient #1634 who had a patent ductus arteriosus, coarctation of the aorta and a small ventricular septal defect caused by a de novo ,2.9 Mb interstitial deletion. The fifth critical region is defined by a de novo ,3.6 Mb interstitial deletion in Patient 8 who was referred to the Medical Genetics Laboratories at Baylor College of Medicine for copy number analysis with an indication for testing that included coarctation of the aorta.
Using this same dataset, we defined two critical regions for cardiomyopathy (Table 8, Figure 2). The first critical region contains the PRDM16 gene and is defined by an ,1.5 Mb interstitial deletion in two siblings with left ventricular noncompaction described by Gajeka and colleagues. The second critical region is defined by an ,3.0 Mb interstitial deletion in an individual reported by Kang and colleagues (Case 4) who had dilated cardiomyopathy which coexisted with a perimembranous ventricular septal defect and a small septum secundum atrial septal defect [5]. Kang and colleagues indicated that both cardiac malformations spontaneously closed but the dilated cardiomyopathy remained and was still being treated with medications when the child was 5 years old. While this history suggests that he had a primary cardiomyopathy, we cannot rule out the possibility that his cardiovascular malformations contributed to his dilated cardiomyopathy. However, further evidence of the role of this region in the development of primary cardiomyopathy comes from Patient 16 described by Arndt and colleagues who had an overlapping ,8 Mb deletion associated with early-onset cardiomyopathy with transient heart failure whose only structural heart defect was a clinically insignificant ventricular septal defect [9].
Each of the cardiac-related critical regions we identified contains one or more positional candidate genes whose haploinsufficiency may contribute to the development of cardiovascular phenotypes. Genes which may contribute to the development of cardiovascular malformations associated with 1p36 deletions include DVL1, SKI, RERE, PDPN, SPEN, CLCNKA, ECE1, HSPG2, LUZP1, and WASPF2. Genes which may contribute to the development of cardiomyopathy associated with 1p36 deletions include SKI, PRKCZ, PRDM16, RERE, UBE4B and MASP2. The cardiac-related phenotypes associated with each of these genes in humans, mice and zebrafish are summarized in Tables 7 and 8.

Discussion
Terminal deletions of chromosome 1p36 are the most common telomeric deletions in humans and carry a high risk of cardiacrelated medical problems. In one large cohort, 71% of individuals with terminal 1p36 deletions had cardiovascular malformations and 27% had cardiomyopathy [2]. The most common cardiovascular malformations seen were atrial and ventricular septal defects, patent ductus arteriosus, valvular anomalies, tetralogy of Fallot and coarctation of the aorta [2]. Among individuals with cardiomyopathy, 85% had noncompaction defects and the remaining 15% had dilated cardiomyopathy. Similar patterns of cardiac anomalies were seen in the patients with terminal 1p36 deletion described in this report. Individuals with interstitial deletions of 1p36 also have high rates of cardiac abnormalities, although their exact incidences likely vary from one region to another and are more difficult to estimate based on the relatively low number of patients that have been identified [5]. Due to the high risk of cardiac-related problems, most children with 1p36 deletions are screened for cardiovascular anomalies at baseline and followed over time for the development of cardiomyopathy-often having several echocardiograms even if they are asymptomatic. Although this type of careful monitoring can help identify potentially treatable problems, it also places a burden on families and the health care system which could be avoided or reduced if an individual's risk of having these problems could be estimated based on the location and extent of their individual 1p36 deletion. This type of risk stratification will require a better understanding of the 1p36 genes and genomic regions associated with the development of cardiovascular anomalies and cardiomyopathy.
In this report, we take the first step towards addressing this issue by defining five non-overlapping critical regions for cardiovascular malformations and two non-overlapping critical regions for cardiomyopathy. Each of these cardiac critical regions is defined by an isolated 1p36 deletion identified in a single individual. Hence, they represent genomic regions whose deletions are sufficient to cause cardiovascular malformation or cardiomyopathy. The size of these regions vary with the smallest being 1.5 Mb (28 genes) and the largest being 7.8 Mb (175 genes). As more 1p36 deletions are described, it is likely that these regions will be refined and some may be subdivided into more than one region. It is also possible that data from additional deletions will reveal novel critical regions on 1p36 that are associated with cardiac problems.
It is reasonable to assume that each of the cardiac critical regions we have delineated on chromosome 1p36 contains at least one dosage-sensitive, cardiac-related gene or regulatory region. Although one or more positional candidate genes from each interval can be identified based on their known function and/or their phenotypes in animal models, haploinsufficiency of only two genes have been shown to cause cardiac phenotypes in humans. Heterozygous loss-of-function mutations in PRDM16 have been shown to be sufficient to cause left ventricular noncompaction and a loss-of-function mutation in ECE1 was identified in an individual with patent ductus arteriosus, a small subaortic ventricular septal defect and a small atrial septal defect, Hirschsprung disease, and autonomic dysfunction [9,10]. Loss-of-function mutations in the remainder of the positional candidate genes identified in 1p36 cardiac critical regions-DVL1, SKI, RERE, PDPN, SPEN, CLCNKA, HSPG2, LUZP1, WASPF2, PRKCZ, UBE4B and MASP2-have yet to be identified in individuals with cardiovascular phenotypes. The identification of de novo loss-of-function mutations in these positional candidate genes, or dominantly inherited loss-of-function mutations segregating with a cardiac phenotype in multiple family members, would serve to confirm their pathogenic roles. Experience with other chromosomal regions suggests that these types of mutations are most likely to be identified in genes with particularly high impacts on cardiac development and function like GATA4 on 8p23.1, ZFPM2 on 8q23.1, and TBX1 on 22q11.2 [12][13][14].
De novo loss-of-function mutations in cardiac genes whose haploinsufficiency makes a more modest contribution to cardiac  risk may be particularly difficult to identify. Loss-of-function mutations in such genes are more likely to be inherited from an asymptomatic parent and may combine with other genetic, environmental or stochastic factors to cause cardiac-related problems in sporadic cases following a multifactorial inheritance pattern. In such cases, animal models may provide the first evidence that a gene within one of the critical regions plays a role in cardiac development and/or function. Not only can animal models be used to effectively identify low penetrance phenotypes caused by haploinsufficiency of a candidate gene, but they can also be used to explore cardiac-related phenotypes-including defects that lead to embryonic lethality-that only become apparent when the expression of an individual gene is reduced by more than 50%. Deleterious changes in genes that make a modest contribution to cardiac risk may also be identified in asymptomatic individuals from the general population. One source of information on potentially deleterious changes in various populations is the Database of Genomic Variants (http://dgv.tcag.ca/) which catalogues genomic variation among ''normal controls'' and population-based cohorts. It is interesting to note that exoncontaining deletions in each of the positional candidate genes identified in this study have been reported in this database (Table  S1). This suggests that the cardiac-related phenotypes caused by haploinsufficiency of each of these genes alone are likely to be incompletely penetrant or may go undetected in some individuals. This is consistent with the incomplete penetrance for cardiovascular malformations and cardiomyopathy seen among patients with overlapping terminal and interstitial deletions of 1p36 [2,5].
Several of the cardiac critical regions we have identified on chromosome 1p36 contain more than one positional candidate gene. We also note that large terminal and interstitial deletions often overlap more than one cardiac critical region. This suggests that haploinsufficiency of two or more genes may contribute to the cardiac phenotypes associated with many 1p36 deletions. Future studies aimed at understanding how adjacent 1p36 genes work together to impact cardiac development and function may provide information which can be used to design effective therapeutic and/or preventative measures which can minimize the impact of cardiovascular malformations and/or cardiomyopathy in such cases.

Supporting Information
Table S1 Exon-containing deletions in 1p36 cardiacrelated genes found in various control and populationbased cohorts catalogued in the Database of Genomic Variants.