|Home | About | Journals | Submit | Contact Us | Français|
Using an Affymetrix GeneChip® Human Mapping 100K Set to study a patient with a late-presenting, right-sided diaphragmatic hernia and microphthalmia, we found a maternally inherited deletion that was 2.7 Mb in size at chromosome 18q22.1. Mapping of this deletion using fluorescence-in-situ hybridization revealed three deleted genes - CDH19, DSEL and TXNDC10, and one gene that contained the deletion breakpoint, CCDC102B. We selected DSEL for further study in 125 patients with diaphragmatic hernias, as it is involved in the synthesis of decorin, a protein that is required for normal collagen formation and that is upregulated during myogenesis. We found p.Met14Ile in an unrelated patient with a late-presenting, anterior diaphragmatic hernia. In the murine diaphragm, Dsel was only weakly expressed at the time of diaphragm closure and its expression in C2C12 myoblast cells did not change significantly during myoblast differentiation, thus reducing the likelihood that the gene is involved in myogenesis of the diaphragm. Although it is possible that the 18q22.1 deletion and haploinsufficiency for DSEL contributed to the diaphragmatic defect in the patient, a definite role for DSEL and decorin in the formation of the collagen-containing, central tendon of the diaphragm has not yet been established.
Congenital diaphragmatic hernia (CDH) has an estimated frequency of 1 in 2000 to 3000 births and has a high neonatal mortality and morbidity [Torfs et al., 1992]. Several genes that cause diaphragmatic defects as part of multiple congenital anomaly syndromes have been identified [Slavotinek, 2007], but the genes that cause isolated diaphragmatic hernias are almost unknown. To date, one nonsense mutation and several sequence variants of unknown significance in FOG2 have been described in CDH patients [Ackerman et al., 2005; Bleyl et al., 2007]. Chromosome aberrations are found in up to one third of patients with diaphragmatic defects [Howe et al., 1996; Enns et al. 1998; Holder et al., 2007], and array comparative genomic hybridization has therefore been undertaken in CDH patients[Kantarci et al., 2006; Slavotinek et al., 2006; Scott et al., 2007]. These array studies and published reports of patients with CDH and cytogenetic aberrations have led to the identification of numerous chromosome regions that are likely to contain dosage sensitive genes required for normal diaphragm development [Holder et al., 2007]. However, with the exception of FOG2, cloning the causative genes from these chromosome regions has proven difficult, perhaps because hernias show substantial phenotypic heterogeneity, ranging from classical ‘Bochdalek’ diaphragmatic hernias in the posterolateral diaphragm, to anterior retrosternal hernias and hernias of the central part of the diaphragm. These different types of diaphragmatic hernias have been considered to be genetically distinct [Ackerman et al., 2007], and as Bochdalek hernias occur most frequently whilst anterior and central diaphragmatic hernias are rare, case collection for molecular studies can be challenging.
We report on a male who had a late-presenting, right-sided diaphragmatic hernia and unilateral microphthalmia. This child was studied with the Affymetrix GeneChip Human Mapping 100K Set and a 2.7 Mb deletion at chromosome 18q22.1 was identified using this array. The deletion was inherited from his mother, who was reported to be phenotypically normal. We mapped the proband's deletion using fluorescence in-situ hybridization (FISH), and determined that the 18q22.1 deletion contained three genes: CDH19, DSEL, TXNDC10, with the telomeric breakpoint of the deletion likely to interrupt another gene, CCDC102B. Although this deletion was inherited from a normal mother, we chose to further examine the three genes in this small deletion.
We selected DSEL as a candidate gene for the diaphragmatic hernia in the proband, as DSEL is involved in the formation of one of the principal components of collagen, decorin. This paper describes the 18q22.1 deletion, our results re-sequencing DSEL in 125 unrelated diaphragmatic hernia patients, and our evaluation of Dsel as a candidate gene for diaphragm myogenesis.
DNA samples were obtained from probands using two protocols approved by the Committee for Human Subjects Research at the University of California, San Francisco (numbers H41842-22157-06 and H41842-26613-04). We used 23 DNA samples from diaphragmatic hernia patients recruited through UCSF and 96 DNA samples obtained from the newborn blood spots of children with diaphragmatic hernias through the California Birth Defects Monitoring Program. The clinical features of 96 of the diaphragmatic hernia patients have been published [Slavotinek et al., 2006] the remaining 23 patients had non-syndromic diaphragmatic hernias [Slavotinek et al., 2009]. We also sequenced Dsel in DNA from six patients with anterior diaphragmatic hernias kindly provided by Dr Daryl Scott at Baylor.
Array hybridization was performed with the GeneChip® Human Mapping 100K Set (http://www.affymetrix.com/products/arrays/specific/100k.affx) using 500 ng genomic DNA according to the manufacturer's instructions. This mapping set has previously been validated for the detection of copy number variants in patients with chromosome aberrations [Rauch et al., 2004; Slater et al., 2005]. The results were analyzed according to the Significance of Mean Difference (SMD) algorithm designed to detect copy number variations and parental studies using the same mapping set [Delaney et al., 2008].
Genomic sequencing was performed with a BigDye® Terminator v3.1 Cycle Sequencing Kit on an ABI 3730 machine (Applied Biosystems, Foster City, CA) as previously described [Slavotinek et al., 2006]. We sequenced CDH19, DSEL, TXNDC10 and CCDC102B coding exons and intron-exon boundaries in the male patient and the coding exon and the intron-exon boundaries of DSEL in 125 patients with diaphragmatic defects. Control chromosomes were examined by restriction enzyme digestion or by genomic sequencing.
Section in-situ hybridization on murine paraffin sections was performed as previously described using digoxygenin-labeled riboprobes (DIG RNA labeling kit; Roche, Indianapolis, IN) [Chao et al., submitted]. The probe for Dsel was generated using primers: F: 5’gagtgagtgcgtgtgtccag; R: 5’tctcgtttttgtgtgcaagg and Dsel expression was examined in the murine diaphragm at E11.5, E12.5, E13.5, E14.5 and E16.5.
Exponentially growing C2C12 myoblasts were plated in 60 × 15 mm Petri dishes at a density of 2.4 × 104 cells per ml. When the cells were 75-90% confluent, growth medium was replaced with differentiation medium containing 2% horse serum (Hyclone, Logan UT) and harvested on days 1, 3 and 6. RNA was obtained by standard methods (RNeasy kit, Qiagen, Valencia, CA). cDNA synthesis was performed from 500-1000 ng of total RNA using the High-Capacity cDNA Reverse Transcription kit (Applied Biosystems, Foster City, CA). The expression of Dsel (TaqMan probe Mm01294533_g1; Applied Biosystems, Foster City, CA) and Hprt1 (TaqMan probe Mm03024075_m1; Applied Biosystems, Foster City, CA) were assayed at days 1, 3 and 6 using real-time PCR (Applied Biosystems, Foster City, CA). Quantification analysis was performed with the ΔΔCt method using Hprt as a control.
The 13-month old male propositus with microphthalmia and diaphragmatic hernia has previously been described [FF264; Pasutto et al., 2007; Chao et al. submitted]. He was ascertained at 13 months of age when he presented with respiratory difficulties and a right diaphragmatic hernia was diagnosed by a chest radiograph. We have no further information regarding the type of hernia, but we assume it to have been relatively small because of the late presentation of the diaphragmatic defect. Surgical repair was successfully performed. He had unilateral right microphthalmia and the diameter of the right globe was reported as 9 mm on computerized tomography imaging at 13 months of age, with an estimated volume of 0.38 cm3 [normal human globe size is 3 cm3 at birth, increasing to 6 cm3 at age 24 months; Galluzzi et al., 2001]. The propositus’ parents were reportedly normal, but a maternal aunt was described as having anophthalmia, although she was unable to be examined. Sequencing of the STRA6 (Stimulated-by-retinoic acid-6; OMIM 610745) gene was negative in the propositus [Pasutto et al., 2007].
The second patient with c.1516G>A, predicting p.Met14Ile, was a female of Hispanic ethnicity. She was born at term with a birth weight of 3033 g (appropriate for gestational age) and an anterior diaphragmatic hernia was surgically corrected at six weeks of age. She presented with respiratory distress after her initial discharge from the nursery and the late presentation also implies a small diaphragmatic defect. She had no other abnormal physical findings.
In the propositus, the array results showed loss of heterozygosity and log2ratio results consistent with a deletion of approximately 2.7 Mb in size at chromosome 18q22.1 (Fig. 1). Arrays with the same mapping set were performed on both parents, and the mother shared the same deletion (data not shown), but cells were not available from the mother for verification by FISH. The father did not show any significant copy number variations. We did not find any other significant copy number variations (CNVs) in the propositus or his parents and no other family members were available to be tested. The 18q22.1 deletion in the propositus was verified using FISH, and the centromeric breakpoint was mapped between BAC RP11-1069K2 (not deleted; chr18:61,828,786-62,017,616; numbering according to hg version 18; UCSC Genome Browser) and BAC RP11-246I7 (deleted; chr18:62,306,291-62,490,328) and the telomeric breakpoint was mapped between BAC RP11-105L16 (deleted; chr18:64,478,284-64,627,510) and BAC RP11-22A22 (not deleted; chr18:64,596,566-64,777,467; data not shown). The deleted region (chr18:62,306,291-64,627,510) therefore contains three genes - CDH19 (chr18:62,322,301-62,422196), DSEL (chr18:63,324,799-63,334,947) and TXNDC10 (chr18:64,491,905-64,533,333). CCDC102B is located at the telomeric edge of the deletion at chr18:64,616,550-64,873,406. The sizes, domains and expression pattern of these genes have been summarized in Table I. CDH19 and CCDC102B have been reported to be located in CNV regions (see Database of Genome Variants, http://www.tcag.ca/) with multiple entries for each of these genes. These CNV studies do not definitely eliminate any of the deleted candidate genes from consideration for the etiology of the diaphragmatic defects, although we assessed CDH19 and CCDC102B as less likely to be involved in the pathogenesis of the diaphragmatic hernia due to the frequency of CDH19 and CCDC102B deletions in normal individuals.
We examined the genes within the deleted interval at 18q22.1 for sequence variants in the propositus. We found no sequence alterations in CDH19, TXNDC10, DSEL or CCDC102B on the non-deleted allele of chromosome 18 (data not shown). We then sequenced DSEL in 125 patients with diaphragmatic hernia. We found one novel missense substitution, c.1516G>A, predicting p.Met14Ile in exon 2 of DSEL, in a patient with an anterior, late-presenting diaphragmatic hernia (Table II; Fig. 2; numbering = initial A of cDNA= 1, DSEL Ensembl transcript ENST00000310045). No other tissues were available for further studies. DNA samples from the parents were not available, but restriction digestion for the p.Met14Ile substitution in 200 Hispanic control chromosomes was negative, although it confirmed the substitution in the propositus (data not shown). Although the Met14 residue was highly conserved, the alteration was predicted to be benign by the PolyPhen website for the prediction of the functional effects of non-synonymous single nucleotide polymorphisms (SNPs). Another sequence alteration, c.4445G>A, predicting p.Asp991Asn, was present in a patient with a diaphragmatic hernia and trisomy 18 (Table II), but was not further studied as the same alteration was present in a healthy parent and our model for the pathogenesis of the hernias predicts haploinsufficiency. We also found c.2301A>G, predicting p.Asn276Ser in CDH patients, but restriction digests showed that this nucleotide substitution was a polymorphism more frequent in panethnic controls (5/84 chromosomes = 6%) than Caucasians (0/180 chromosomes). Comparing allele frequencies between CDH patients and controls from dbSNP, we found that only the A allele from the SNP c.2646A>G, a synonymous SNP resulting in p.Gln390Gln, was significantly reduced in CDH patients compared to controls (p = 0.0038; Table II). However, in view of the difference in ethnicity between the dbSNP population (Caucasian and African American) and our CDH patients (mostly Caucasian and Hispanic), we cannot conclude that this finding has real significance, as the difference in allele frequency may be influenced by ethnicity. In addition, we had no second cohort to replicate this finding.
We examined Dsel expression in the murine diaphragm from E11.5 to E16.5. We found only weak expression of Dsel in the murine diaphragm muscle at E13.5 (Fig. 3), and we did not see reliable diaphragmatic expression at later or earlier time periods (data not shown). Our studies continued up to E16.5, several days after diaphragm closure (E12.5 –E13.5 in the mouse), as in some murine models of diaphragmatic defects, hernias have first been noted after the time of diaphragmatic closure as late as E15.5 [Yuan et al., 2003].
To implicate Dsel in diaphragmatic myogenesis, we tested the hypothesis that Dsel would be upregulated during myoblast differentiation. We examined the expression of Dsel during the differentiation process of C2C12 myoblasts compared to a control gene, Hprt, which showed no increased expression with myoblast differentiation. Our results showed no significant change in the expression of Dsel with myoblast differentiation (Fig. 4)
We have demonstrated a maternally inherited, 2.7 Mb 18q22.1 deletion in a male with microphthalmia and a late-presenting diaphragmatic hernia. The deleted genes have been summarized in Table I. CDH19, or Cadherin 19, type 2 preproprotein [OMIM 603016] is a calcium dependent, cell-cell adhesion glycoprotein expressed in neural crest-derived cells that may be important for the development of Schwann cells [Takahashi et al., 2005]. DSEL, or Dermatan Sulfate Epimerase-Like [OMIM 611125], acts as an epimerase and enables the formation of dermatan sulfate from chondroitin sulfate.[Goossens et al., 2003; Maccarana et al., 2006] TXNDC10, or Thioredoxin-domain-containing-10, also known as TMX3, is a thioredoxin that catalyzes the formation and folding of disulfide bonds [Haugstetter et al., 2005; Haugstetter et al., 2007]. Finally, little is known about CCDC102B, or Coiled-coil domain-containing 102B, and both PubMed and OMIM contain no references to this gene.
We chose DSEL for sequencing in a group of 125 patients with diaphragmatic hernia because of the importance of this gene in dermatan sulfate biosynthesis and decorin formation. DSEL acts as a chondroitin-glucuronate C5-epimerase, converting D-glucuronic acid to L-iduronic acid, and catalyzing the formation of dermatan sulfate from chondroitin sulfate. The main dematan sulfate-carrying proteoglycan, decorin, is a leucine-rich proteoglycan located in the extracellular matrix and expressed in the diaphragm of mdx mutant mice [Cáceres et al., 2000]. Decorin binds to collagen types 1 and VI, regulating collagen fibril formation and the stabilization of collagen fibers [Schönherr et al., 1995; Danielson et al., 1997]. Decorin is upregulated during myogenesis and C2C12 myoblast cells that overexpress decorin show increased myoblast proliferation, whereas antisense inhibition of decorin suppresses proliferation and accelerates the differentiation of C2C12 myoblasts due to an enhanced sensitivity to exogenous myostatin [Riquelme et al., 2001; Kishioka et al., 2008].
Our DSEL resequencing results showed one missense substitution, p.Met14Ile (Table I; Fig. 2), in an Hispanic female with a late-presenting, anterior hernia. Parental samples were not available to determine if the substitution was de novo, but it was highly conserved in other species and absent from 200 Hispanic control chromosomes. The missense substitution was also not detected in a paper that sequenced DSEL in 113 patients with bipolar disorder and 160 matched controls [Goossens et al., 2003] although the ethnicity of these subjects was not mentioned. However, the p.Met14Ile sequence alteration was not predicted to be disease-causing by software prediction programs. We also found no evidence for a role for Dsel in myogenesis in C2C12 cells, as expression levels of Dsel were unchanged during myoblast differentiation when examined by quantitative RT-PCR (Fig. 4).
Although we have not shown that the deletion of DSEL is involved in the pathogenesis of the hernias, both the patient with the deletion and the patient with the missense mutation had late-presenting diaphragmatic hernias. Late presenting hernias are defined as diaphragmatic hernias that are detected after the neonatal period and they may constitute a separate phenotypic subdivision of diaphragmatic hernias due to diaphragm weakness.[Numanoglu et al., 1997; Elhalaby et al., 2002; Baglaj, 2004] As decorin is involved in mature collagen formation, a deletion of the entire Dsel gene could result in reduced decorin formation, with a predisposition to later diaphragmatic weakness because of abnormal collagen synthesis. However, the gene appears not to play a major role in the etiology of CDH, as evidenced by the few sequence alterations observed in our patients. It is also possible that another of the deleted genes is responsible for the hernia, or that haploinsufficiency for more than one gene in the deleted interval is needed for the diaphragmatic defect. Re-sequencing of TXNDC10 in 25 patients with diaphragmatic hernias also returned no novel sequence alterations [Chao et al., submitted]. In addition, the occurrence of the deletion in the propositus's phenotypically normal mother must be explained. We hypothesize that the diaphragmatic hernia in the propositus may have been due to an additional contribution of haploinsufficiency for an alternative gene involved in diaphragm formation, unmasking of a recessive allele, or environmental factors.
Finally, this paper did not investigate any of these genes for a role in the microphthalmia in the propositus. However, TXNDC10 is expressed in the eye (Unigene; http://www.ncbi.nlm.nih.gov/UniGene/ESTProfileViewer.cgi?uglist=Mm.268041), and therefore this gene may prove to be an attractive candidate for studies into the etiology of the microphthalmia in the propositus.
We report on a maternally inherited, 2.7 Mb deletion at chromosome 18q22.1 disrupting four genes in a male with a late-presenting, right-sided diaphragmatic hernia and microphthalmia. We investigated DSEL for a role in the pathogenesis of the diaphragm hernia and identified one amino acid substitution, p.Met14Ile, in an unrelated patient with a late-onset, anterior diaphragmatic hernia. However, a definite role for DSEL in the pathogenesis of diaphragmatic defects has not been established.
Anne Slavotinek was generously funded by an R03 grant 5R03 HDO49411-02 from the National Institute of Child Health and Development (NICHD) and a K08 grant K08HD053476-01A1 from the same institution. Hatem Zayed was supported by a Postdoctoral Training Grant in Medical Genetics, 5T32GM007085-30, to the Division of Medical Genetics, Department of Pediatrics, at UCSF. We are grateful to Dr Daryl Scott for providing patient samples for the project. This publication was supported by NIH/NCRR UCSF-CTSI Grant Number UL1 RR024131. This publication was supported by NIH/NCRR UCSF-CTSI Grant Number UL1 RR024131. The findings and conclusions in this report are are solely the responsibility of the authors and do not necessarily represent the official views of the NIH or the California Department of Public of Health. We thank the California Department of Public Health Maternal Child and Adolescent Health Division for providing data for these analyses.