SEARCH

SEARCH BY CITATION

Keywords:

  • congenital diaphragmatic hernia;
  • isolated CDH;
  • array CGH;
  • EFNB1;
  • 8p deletions;
  • mosaic trisomy 2;
  • prenatal diagnosis

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. Acknowledgements
  9. REFERENCES
  10. Supporting Information

Objective

Congenital diaphragmatic hernia (CDH) is a congenital birth defect affecting around 1/3000 births. We propose that a significant number of isolated CDH cases have an underlying genetic cause, and that a subset of these result from copy number variations (CNVs) identifiable by array CGH.

Methodology

We have designed a custom array targeted at genes and genomic loci associated with CDH. A total of 79 isolated CDH patients were screened using this targeted array.

Results

In three patients, we detected genomic imbalances associated with the observed diaphragmatic hernia; a deletion of 8p22-p23.3, 14.2 Mb in size, a 340 kb duplication of Xq13.1 including the ephrin-B1 gene (EFNB1), and mosaicism for trisomy 2.

Conclusion

Using this approach, we detected genomic imbalances associated with CDH in 3/79 (4%) isolated CDH patients. Our findings further implicate 8p deletions as being associated with CDH. The duplication of EFNB1 further highlights this gene as a potential candidate involved in diaphragm development. Mosaicism for trisomy 2 is a rare event and unlikely to be a common cause of CDH. Further investigations of isolated CDH patients by array CGH will continue to identify novel submicroscopic loci and refine genomic regions associated with CDH. Copyright © 2010 John Wiley & Sons, Ltd.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. Acknowledgements
  9. REFERENCES
  10. Supporting Information

Congenital diaphragmatic hernia (CDH) is a congenital birth defect with an incidence of around 1/3000 births (Torfs et al., 1992; Skari et al., 2000). CDH can be anatomically divided into three main hernia subtypes: a posterolateral ‘Bochdalek’ hernia in around 70% of cases, an anterior ‘Morgagni’ hernia in around 27% of cases, and a central septum transversum hernia in around 3% of cases. The vast majority of hernias are left sided (85%), while the remainder are right sided (13%) or bilateral (2%) (Torfs et al., 1992; Pober, 2007; van Loenhout et al., 2009). Next to the diaphragm defect, the lung developmental anomaly is an essential part of the phenotype. It results in variable degrees of pulmonary hypoplasia and postnatal pulmonary hypertension, which account for the high mortality and morbidity in survivors.

CDH occurs as an isolated defect in around 50% of cases, or as non-isolated CDH for the remainder in which additional congenital malformations are present (Stoll et al., 2008). Non-isolated CDH is associated with abnormalities in a number of other systems, including cardiovascular system (27.5%), urogenital system (17.7%), musculoskeletal system (15.7%), and central nervous system (9.8%) (Stoll et al., 2008). It may also occur as part of a recognised syndrome for which a single causal gene may be identified. Examples include STRA6 in Matthew-Wood syndrome (Golzio et al., 2007; Pasutto et al., 2007; Chassaing et al., 2009; Segel et al., 2009), WT1 in Denys–Drash syndrome (Devriendt et al., 1995; Antonius et al., 2008), and EFNB1 in craniofrontonasal syndrome (CFNS) (Wieland et al., 2004; Vasudevan et al., 2006). Alternatively, non-isolated CDH may be associated with specific genetic loci, including 8p23.1 (Faivre et al., 1998; Wat et al., 2009) and 15q26 (Klaassens et al., 2005, 2007; Slavotinek et al., 2006), or with a clinically recognised syndrome of currently unknown genetic cause such as Fryns syndrome (Fryns et al., 1979, 1989; Neville et al., 2002; Slavotinek, 2004).

It is still uncertain whether genetic factors are the sole cause of CDH or if they merely provide a genetic susceptibility to abnormal diaphragm development in concert with other environmental influences. The dual-hit hypothesis has been proposed, which speculates that the lung abnormality is a primary developmental defect, firstly affecting both lungs before diaphragm development and subsequently affecting the ipsilateral lung after defective diaphragm development (Keijzer et al., 2000). Chromosomal abnormalities have been reported to be present in as many as 30% of individuals with CDH (Thorpe-Beeston et al., 1989; Howe et al., 1996; Geary et al., 1998; Tonks et al., 2004), and it has been postulated that these cytogenetic abnormalities can provide positional information about the genomic locations for CDH-causing genes.

In view of the relative rarity of Mendelian pedigrees in families with CDH, isolated CDH is considered to have multi-factorial inheritance with gene mutations, environmental factors, gene–gene, and gene–environment interactions that contribute to exceeding a threshold in susceptible individuals. A number of candidate genes are involved in the retinoic acid pathway (Montedonico et al., 2008; Klaassens et al., 2009). Previous research using animal models has demonstrated that environmental factors including a vitamin A-deficient (VAD) diet or exposure to teratogenic agents, particularly nitrofen, can disrupt the retinoid signalling pathway resulting in CDH (Wilson et al., 1953; Thebaud et al., 1999; Babiuk et al., 2004; Oshiro et al., 2005; Nakazawa et al., 2007a,b; Noble et al., 2007; Clugston et al., 2010). Further evidence of the involvement of this pathway comes from knockout or mutant mouse models in which genes involved in retinoid signalling are shown to cause CDH when silenced, including compound null retinoic acid receptor (RAR) mutants which were shown to have a spectrum of VAD-like defects including diaphragm defects (Lohnes et al., 1994, 1995; Mendelsohn et al., 1994).

For the majority of CDH patients, particularly isolated cases, the genetic cause remains unknown. It has been shown recently that copy number variations (CNVs) underlie many genetic disorders and that a subset of all monogenic diseases is caused by CNVs rather than mutations. As the causal genes for isolated CDH in humans remain unknown, and given the large number of potential candidate genes and pathways, mutation screening may be of limited value at this time. One approach to identify the genes involved in diaphragm development is to study genomic regions recurrently associated with CDH patients. The technique of array comparative genomic hybridisation (array CGH) is one such method which has demonstrated the ability to refine the critical regions associated with various disorders, including CDH (Klaassens et al., 2005; Slavotinek et al., 2006; Scott et al., 2007; Shaffer et al., 2007).

As opposed to applying genome-wide cytogenomic arrays which have an average resolution of 30–50 kb (Miller et al., 2010), a targeted array would enable higher resolution coverage of regions of interest, and as a consequence be able to uncover smaller CNVs in relevant genes. To identify novel genetic causes of isolated CDH, we have therefore developed a targeted oligonucleotide array. This array is designed to provide coverage of genomic regions recurrently associated with non-isolated CDH in humans, supplemented by candidate genes associated with diaphragm and lung development. We hypothesised that by applying high-resolution targeted array comparative genomic hybridisation (array CGH) to cases of isolated CDH we would identify causal genomic imbalances, refine genomic regions causal for CDH, and identify candidate genes important for diaphragm development.

METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. Acknowledgements
  9. REFERENCES
  10. Supporting Information

Patients

A total of 88 fetuses with isolated CDH, which were managed in the prenatal period by the fetal medicine units from the University Hospital Gasthuisberg, Leuven, Belgium (n = 72), and the Hospital Clinic-IDIBAPS, University of Barcelona and Centre for Biomedical Research on Rare Diseases (CIBER-ER), Barcelona, Spain (n = 16), were analysed using the targeted array. Samples from Leuven were either stored genomic DNA extracted from cultured amniotic fluid (AF) (n = 49) or frozen AF cultures from which genomic DNA was extracted following thawing and re-culturing (n = 16), or genomic DNA extracted from neonatal cord blood (n = 7). Those from Barcelona (n = 16) were uncultured AF samples which had been frozen, and from which genomic DNA was later extracted. Only those patients in whom the absence of additional major congenital anomalies was confirmed at birth were considered eligible for inclusion in this study. All samples were collected with the approval of the respective local ethical committee and in accordance with the Helsinki Declaration of 1975 (as revised in 1983).

Targeted array design

An array was designed comprising 15 000 (15 K) oligonucleotide probes in an 8× 15 K format [Oxford Gene Technology (OGT), Oxford, UK]. The approach for the array design is a combination of two methods: (1) to target the genomic regions recurrently associated with CDH in humans and (2) to target candidate genes from CDH animal models and proposed pathways involved in CDH (see Section on Results and Tables S1 and S2, Supporting Information).

Array comparative genomic hybridisation (array CGH)

Reference genomic DNA samples were derived from healthy males and females. All patients and sex-matched reference samples were labelled with fluorescence Cy5 and Cy3 dyes in dye-swap experiments. The CytoSure Genomic DNA labelling kit (OGT) was used according to the manufacturer's protocol, with the following modifications; 500 ng DNA was labelled instead of 1000 ng, and consequently all reagent volumes were halved. The DNA denaturation step was performed at 98 °C for 20 min, and the DNA digestion step was omitted. The subsequent incubation at 37 °C was performed overnight. For DNA purification, spin columns provided with the CytoSure Genomic DNA labelling kit (OGT) were used according to the manufacturer's protocol. For concentration of labelled DNA samples, either Clean and Concentrator-5 spin columns (Zymo Research, CA, USA) or Amicon Ultra-0.5 spin columns (Millipore, MA, USA) were used according to the manufacturer's protocol. Hybridisation mixes were prepared using the Oligo aCGH Hyb kit (Agilent Technologies S.A./N.V., Belgium) according to the manufacturer's protocol. Hybridisation was performed using SureHyb chambers and a rotating oven (Agilent Technologies) at 65 °C for 24 h as recommended. Following hybridisation, arrays were washed manually according to the manufacturer's protocol using Agilent wash buffer 1 and 2, followed by rinsing and drying using acetonitrile.

Arrays were scanned immediately with a DNA Microarray Scanner (Agilent Technologies) at 3 µ resolution. The tif images were visually checked for artefacts. Signal intensities were generated using the Feature Extraction software (v.10.5.1. Agilent Technologies). To visualise the data and perform analysis the CytoSure Interpret software (OGT) was used. A global lowess normalisation is applied by the software to correct for the dye incorporation bias.

Data analysis

The CytoSure Interpret software package features a circular binary segmentation (CBS) algorithm, where thresholds were set at ± − 0.36 from a log2 ratio of zero, and at least five flanking probes within a segment. For a region to be called as aberrant, it must be present in both datasets within the dye-swap experiment. To complement the CBS algorithm, a second threshold-based method was also applied to the individual probe log2 ratios. Data from the patient labelled in Cy5 and in Cy3 is combined as follows: X2 × Y2 = Z (where, X = individual probe normalised log2 ratio for Cy5-labelled patient data, Y = individual probe normalised log2 ratio for Cy3-labelled patient data). The calling threshold was then set at five flanking probes where Z > 0.016.

Therefore, by applying a threshold of five probes, we achieve an average resolution for detecting imbalances of ∼80 kb for target regions and ∼16 kb for target genes.

The CytoSure Interpret software package (OGT) was used for interpretation of the CNVs detected. We have previously characterised the CNVs present in our male and female control samples at ∼200 kb resolution provided by the CytoSure Syndrome Plus v2 105K array (OGT), as well as created an internal database of common likely benign CNVs. Data from the Toronto database of genomic variants (DGV) (http://projects.tcag.ca/variation/) was also used to aid the interpretation of CNVs detected. Any CNV detected in a patient and also within regions in our internal benign CNV database were removed from further investigation. CNVs detected in patients that were within regions containing at least two reports in the DGV of being a CNV in normal individuals were also excluded from further investigation.

Confirmation of imbalances was achieved by the use of a genome-wide 105K array (CytoSure Syndrome Plus v2, OGT) or by FISH. All genomic coordinates are based on the NCBI 36 genome assembly build.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. Acknowledgements
  9. REFERENCES
  10. Supporting Information

A custom array was designed to target genomic loci and genes associated with CDH. A total of 19 genomic regions were selected covering 225 Mb for which there were at least three cases of CDH in humans identified from a literature search, including the study by Holder et al. (2007). For the target genomic regions, the average probe spacing is one probe for every 20 kb. A total of 75 CDH-associated genes covering 17 Mb were selected following a literature search, complemented by using aGeneApart (www.esat.kuleuven.be/ageneapart). aGeneApart uses a text-based input to data-mine the literature and identify genes which display an association. For our purposes, the search term ‘congenital hernia of diaphragm’ (LNDB 170403) was used. Candidate genes were selected based on single gene disorders associated with CDH in humans, genes identified from animal models of CDH, genes whose expression was dysregulated in CDH models, and genes involved in cellular pathways proposed to be involved in CDH. Probe coverage included a 5 kb flanking region for target genes. Average probe coverage for target genes is 55 probes per gene (range 17–95 probes), with average probe spacing of one probe every 4 kb. Target genomic regions and genes selected for this array are listed in Tables S1 and S2, respectively.

A targeted array comprising 15 000 oligonucleotide probes was designed to provide high-resolution coverage of genomic regions recurrently associated with non-isolated CDH, as well as candidate genes for diaphragm and lung development. Eighty-eight isolated CDH patients were screened for genomic imbalances through the use of the custom-designed targeted array. Nine samples (10.23%) had insufficient quantity of DNA for analysis or poor quality DNA for which the result could not be interpreted. Of the remaining 79 patients, 76 (96.2%) displayed normal results, that is no CNVs or only benign CNVs were detected. For three patients (3.8%) genomic imbalances were detected: a deletion of chromosome 8p22-p23.3 in one patient, a duplication of Xq13.1 encompassing the EFNB1 gene in a second patient, and mosaicism for trisomy 2 in a third patient.

Fetus 1

The mother was referred because of the presence of a left-sided CDH with herniation of the liver into the thorax. The observed to expected lung to head ratio (O/E LHR) was 21.8%. No other abnormalities were detected except for a missing 12th rib. Fetoscopic tracheal occlusion (FETO) was performed. The gestational age at the time of diagnosis was 27 weeks. The child was delivered at 36 weeks of gestation, with birth weight 2300 g, and deceased soon after with all clinical features of pulmonary hypoplasia but no other abnormal features. Autopsy was not accepted by the parents; however, a postnatal echocardiography was performed for which no cardiac defects were observed. The array result showed an interstitial deletion on chromosome 8p22-p23.3. This deletion was confirmed by a genome-wide 105K array (CytoSure Syndrome Plus v2, OGT), which also refined the proximal deletion breakpoint. The deletion was shown to be 14.2 Mb in size (958312–15160490bp) and is shown in Figure 1.

thumbnail image

Figure 1. Chromosome 8p23.3-p22 deletion; red dataset displays the 15K custom array result for the patient labelled in Cy5 and blue dataset displays the 105K genome-wide array result for the patient labelled in Cy3. The 105K array result determined the telomeric break point of the deletion. The location of the 8p23.1 deletion region is shown at the bottom of the image

Download figure to PowerPoint

Fetus 2

The mother was referred at 27 weeks because of the presence of a left-sided CDH and liver herniation. The O/E LHR was 22%. No other abnormalities were detected. FETO was undertaken. The male child was delivered at 38 weeks of gestation, birth weight 2585 g, with CDH and no other observed abnormalities. The child died of pulmonary hypoplasia and pulmonary hypertension. The array result showed a duplication of Xq13.1, including the EFNB1 gene. This duplication was confirmed by a genome-wide 105K array (CytoSure Syndrome Plus v2, OGT), which also refined the duplication breakpoints. The duplication is 340 kb in size (67899816–68240036bp) and is displayed in Figure 2. Parental (maternal) samples were declined.

thumbnail image

Figure 2. EFNB1 gene duplication on X chromosome; red dataset displays the 15K custom array result for a patient labelled in Cy5 and blue dataset displays the 105K genome-wide array result for the patient labelled in Cy3. The 105K array refines the extent of the duplication. Genes are displayed in the blue track, where the black arrow highlights the position of the EFNB1 gene

Download figure to PowerPoint

Fetus 3

The mother was referred because of the presence of a left-sided CDH, without liver herniation, and moderate lung hypoplasia (O/E LHR = 29.8%). No other abnormalities were detected. FETO was performed within the TOTAL trial. The male child died postnatally of pulmonary hypoplasia, with no additional congenital malformations observed. Autopsy was declined. The array result indicated a high level of mosaicism for trisomy 2. This result was confirmed by a genome-wide 105K array (CytoSure Syndrome Plus v2, OGT). The average median signal intensity ratios for probes on chromosome 2 for the targeted array and for the genome-wide array were found to be 0.35 and 0.39, respectively suggesting 60–97% mosaicism. FISH analysis with a commercially available centromeric FISH probe (CEP2 SpectrumOrange VYSIS, Abbott S.A./N.V., Belgium) on cultured AF cells determined the level of mosaicism to be ∼88%, with 391/445 trisomic cells observed. The array result and the FISH confirmation are displayed in Figure 3A and B.

thumbnail image

Figure 3. (A) Mosaic trisomy 2; red dataset displays the 15K custom array result for patient labelled in Cy5 and blue dataset displays the 105K genome-wide array result for patient labelled in Cy3. The entire chromosome 2 is shown, with the duplication apparent (B) Mosaic trisomy 2; displays an image from the confirmatory FISH analysis used to determine the level of mosaicism. A centromeric FISH probe (CEP2 SpectrumOrange VYSIS, Abbott) was used, where three or two fluorescent red signals equal a cell with three or two copies of chromosome 2, respectively

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. Acknowledgements
  9. REFERENCES
  10. Supporting Information

We designed a targeted array comprising 15 000 oligonucleotide probes. This array was targeted at regions recurrently associated with CDH in humans, and to potential candidate genes involved in diaphragm and lung development. A screen of 79 isolated CDH patients using this targeted oligonucleotide microarray revealed two causal imbalances and one likely causal variant.

The finding of an 8p deletion in one patient adds further evidence to this region being associated with CDH. 8p deletions encompassing the 8p23.1 microdeletion region have previously been observed in a number of patients with CDH (Pecile et al., 1990; Howe et al., 1996; Faivre et al., 1998; Borys and Taxy, 2004; Shimokawa et al., 2005; Slavotinek et al., 2005; Lopez et al., 2006; Baynam et al., 2008; Wat et al., 2009). In addition, deletions of this region are also associated with congenital heart defects (Claeys et al., 1997; Devriendt et al., 1998, 1999). GATA4 is considered to be the gene responsible for the heart abnormalities observed in patients with 8p deletions following the identification of GATA4 mutations in patients with congenital heart defects (Garg et al., 2003; Okubo et al., 2004; Reamon-Buettner and Borlak, 2005; Rajagopal et al., 2007; Reamon-Buettner et al., 2007; Tomita-Mitchell et al., 2007). GATA4 has also been previously proposed as a candidate gene for diaphragm defects given the diaphragmatic abnormalities observed in mice heterozygous for a deletion of GATA4 exon 2 (Jay et al., 2007). However, no patients have been reported with GATA4 mutations and CDH. Interestingly, our patient was an apparently isolated case of CDH, with no heart defects detected prenatally by ultrasound or postnatally by echocardiography. To our knowledge there are no reported cases of 8p deletions in association with isolated CDH. However, we cannot fully exclude the presence of a heart defect in our patient, and additional phenotypic features of typical 8p23.1 microdeletion syndrome may not be apparent at birth.

The transcription factor COUP-TFII has been shown to interact with FOG2 (Huggins et al., 2001), which in turn modulates the transcriptional activity of GATA4 (Crispino et al., 2001). COUP-TFII and FOG2 are located on 15q26 and 8q23, respectively, regions recurrently associated with CDH in humans reviewed by Holder et al. (2007). Furthermore, tissue-specific ablation of COUP-TFII in mice caused a posterolateral Bochdalek-type CDH (You et al., 2005), and FOG2N-ethyl-N-nitrosourea (ENU) mutant mice displayed diaphragm and lung defects (Ackerman et al., 2005). All these genes are proposed to be involved in the retinoic acid pathway, and their co-expression in the mesenchymal cells of the pleuroperitoneal folds supports the retinoid hypothesis and the mesenchymal hit hypothesis which have been proposed as potential mechanisms of abnormal diaphragm development (Clugston et al., 2006, 2010; Jay et al., 2007). It is possible that these proteins act together to regulate transcription of target genes downstream of the retinoic acid pathway involved in normal diaphragm development. There may be additional genes related to diaphragm development present in the deleted region. However, with no evidence supporting other candidate genes in the 8p region as being involved in CDH, more research is needed. The exact mechanism by which 8p deletions cause diaphragm defects is unknown, and the influence of GATA4 deletions or mutations on human CDH remains to be proven.

Mutations and deletions of the EFNB1 gene have been linked to CFNS (Twigg et al., 2004; Wieland et al., 2005, 2007), in which diaphragmatic defects have been observed in a number of patients (Vasudevan et al., 2006). CFNS is an X-linked disorder in which the phenotype is of greater severity in heterozygous females than in hemizygous males. The mechanism by which defects in EFNB1 manifest themselves is proposed to occur by cellular interference (Wieacker and Wieland, 2005; Twigg et al., 2006; Wieland et al., 2008). As our male patient displays a previously unreported duplication of Xq13.1 including the EFNB1 gene, we consider this imbalance to be an unclassified variant. Unfortunately, maternal samples were not available to determine the inheritance status of this imbalance. While the syndromic features of CFNS are seen with greater severity in affected females, as CDH is observed in both males and females with EFNB1 mutations, it is unlikely that the cellular interference model can explain abnormal diaphragm development without the presence of somatic mosaic males. The exact function of EFNB1 with respect to diaphragm development is unknown. EFNB1 acts as both a receptor and a ligand in a tissue-specific manner during embryogenesis, and the resulting forward and reverse signalling may be important for diaphragm development. EFNB1 may display a degree of dosage sensitivity in which duplications, as well as deletions or mutations, affect the function of developing tissues. EFNB1 has important roles in migration of neural crest cells, a pathway proposed to be involved in CDH (Davy et al., 2004; Davy and Soriano, 2005; Arvanitis and Davy, 2008). A single report in the DGV finds the region encompassing the EFNB1 gene to be a rare variant in normal individuals (frequency < 1%) (Shaikh et al., 2009). However, this finding has not been reproduced by other studies of normal individuals and, hence, may be an artefact. Furthermore, we have not observed a similar CNV in over 5000 cases analysed by an array CGH in our laboratory. Duplications of EFNB1 may represent a susceptibility locus for CDH, or may be a rare but benign CNV. Without the identification of additional isolated CDH patients with similar duplications, or further imbalances from studies of normal individuals, it is difficult to fully establish the clinical significance of this finding.

Mosaicism for trisomy 2 is a rare finding with only seven liveborn cases reported thus far (Robinson et al., 1997; Sago et al., 1997; Mihci et al., 2009). In a recent report, Mihci et al. describe a patient with mosaic trisomy 2 and displaying mental retardation, multiple congenital anomalies including diaphragmatic hernia, and dysmorphic features similar to Pallister-Killian syndrome (Mihci et al., 2009). Our patient therefore represents the eighth reported case of mosaic trisomy 2 in a liveborn and the second case with CDH. The finding of CDH in our patient adds further evidence to this abnormality being part of the variable spectrum of phenotypic features associated with mosaicism for trisomy 2. Imbalances on chromosome 2 have previously been observed in CDH patients, reviewed by Holder et al. (2007), although our finding does not make it possible to further confirm these specific regions as being causal for CDH. However, given the rarity of this finding, we conclude that while CDH does display an association with mosaic trisomy 2, this finding does not represent a common cause of diaphragmatic hernia.

We have taken the approach of only targeting the regions recurrently associated with non-isolated CDH, as well as targeting CDH-associated genes. All patients had a karyotype performed, as well as FISH analysis for the common aneuploidies (21, 18, 13, X, and Y), which revealed normal results. FISH analysis to exclude the presence of isochrosome 12p which causes Pallister-Killian syndrome was also performed for the majority of patients revealing normal results. Only 19 patients have had a genome-wide array performed which revealed only benign CNVs for these patients. Therefore, we cannot rule out that submicroscopic imbalances outside of our target regions may be present in some patients. Additional research on both isolated and non-isolated CDH patients by array CGH is likely to reveal novel submicroscopic loci associated with CDH, and will refine those regions repeatedly associated with CDH.

CONCLUSIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. Acknowledgements
  9. REFERENCES
  10. Supporting Information

We identified chromosome abnormalities causal for CDH in one patient with an 8p deletion and in one patient with mosaic for trisomy 2. In addition, the duplication of EFNB1 further highlights this gene as a potential candidate involved in diaphragm development. As a number of loci which display strong associations with CDH are below the resolution of conventional karyotyping, such as 8p23.1 deletions, 15q26 deletions, and 4p16 deletions (Wolf-Hirschhorn syndrome), the continued investigation of isolated CDH patients by high-resolution microarrays will eventually reveal the true incidence of submicroscopic imbalances as a cause of CDH, and whether imbalances in genomic loci associated with syndromic CDH are also a cause of isolated CDH. Gene prioritisation strategies and next-generation sequencing may also prove to be an effective method of screening a large cohort of isolated CDH patients for mutations in the many candidate genes which are associated with CDH.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. Acknowledgements
  9. REFERENCES
  10. Supporting Information

The European Commission funds the FETO programme in its sixth Framework programme (LSHC-CT-2006-037409). The EC also funds PB via a Marie Curie Early Stage Research Training fellowship (MEST CT2005 019707). The Fonds voor Wetenschappelijk Onderzoek Vlaanderen (FWO, 1.8.012.07.N.02) and the Instituut voor Wetenschap en Technologie (IWT/070715) fund JDP as a ‘Clinical Researcher’. KS is recipient of a grant from the ‘Prince Doctor’ Fund, Faculty of Medicine, Chiang Mai University, Thailand, and is an Instructor with the Maternal Fetal Medicine Unit, Obstetrics and Gynaecology Department, Faculty of Medicine, Chiang Mai University, Thailand. R.C.M. was supported by Marie Curie Host Fellowships for Early Stage Researchers, FETAL-MED-019707-2. KD is a senior clinical investigator of the F.W.O. Flanders. Part of this work was made possible by grants from the IWT (SBO-60848 and 07/0715), FWO (GOA/2006/12), and Center of Excellence SymBioSys (Research Council K.U.Leuven EF/05/007) (JRV).

The fetal medicine team (E Done, T Van Mieghem, L Gucciardo, P DeKoninck, D Van Schoubroeck, R Devlieger, and F Clause) is thanked for the clinical prenatal management of patients in this series.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. Acknowledgements
  9. REFERENCES
  10. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. Acknowledgements
  9. REFERENCES
  10. Supporting Information
FilenameFormatSizeDescription
pd_2651_supportinginformations.doc122KSupporting Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.