Silver–Russell syndrome (SRS) is characterised by prenatal and postnatal growth retardation, dysmorphic facial features, and body asymmetry. In 35–60% of SRS cases the paternally methylated imprinting control region (ICR) upstream of the H19 gene (H19-ICR) is hypomethylated, leading to downregulation of IGF2 and bi-allelic expression of H19. H19 and IGF2 are reciprocally imprinted genes on chromosome 11p15. The expression is regulated by the imprinted methylation of the ICR, which modulates the transcription of H19 and IGF2 facilitated by enhancers downstream of H19. A promoter element of IGF2, IGF2P0, is differentially methylated equivalently to the H19-ICR, though in a small number of SRS cases this association is disrupted—that is, hypomethylation affects either H19-ICR or IGF2P0.
Three pedigrees associated with hypomethylation of IGF2P0 in the probands are presented here, two with paternally derived deletions, and one with a balanced translocation of inferred paternal origin. They all have a breakpoint within the H19/IGF2 enhancer region. One proband has severe growth retardation, the others have SRS.
This is the first report of paternally derived structural chromosomal mutations in 11p15 causing SRS. These cases define a novel aetiology of the growth retardation in SRS, namely, dissociation of IGF2 from its enhancers.
- Silver–Russell syndrome
- H19 enhancer
- molecular genetics
Statistics from Altmetric.com
Silver–Russell syndrome (SRS, MIM 180860) is a genetically and clinically heterogeneous condition characterised by intrauterine and postnatal growth retardation, dysmorphic facial features, and body asymmetry.1 Thirty-five per cent to 60% of SRS cases show hypomethylation of the H19 imprinting control region (H19-ICR; figure 1A) at 11p15, which is believed to result in reduced expression of IGF2 (MIM 147470).2–4 The underlying causes of hypomethylation include mosaic maternal uniparental disomy (UPD) of chromosome 11,5 maternal duplication of 11p15,6 7 and possibly gene mutations of H19-ICR,8 but is unknown in most cases. In most SRS patients hypomethylation occurs at both H19-ICR and IGF2P0, but is occasionally restricted to H19-ICR or IGF2P0.9
The H19-ICR is located within the H19 promoter and is normally methylated on the paternally derived allele. H19 (MIM 103280) encodes a maternally expressed non-coding RNA while IGF2 encodes the paternally expressed growth factor IGF-II. In mice the endodermal enhancers of IGF2 are located +10 kb from H19 transcription start site; skeletal muscle and mesodermal enhancers are located at +35 kb and other mesodermal enhancers at +120 kb.10 The interaction of these enhancer sequences with the promoters of H19 or IGF2 is modulated by CTCF for which the H19-ICR contains seven potential binding sites.11 CTCF binding to unmethylated H19-ICR prevents interaction between the IGF2 promoter and enhancers, while methylation of the H19-ICR prevents CTCF binding, enabling the enhancers to interact with the IGF2 promoter. Naturally occurring structural anomalies in the enhancer region of H19/IGF2 in patients are very informative in testing the hypothesis of complex imprinting and competing enhancer regulation of H19 and IGF2, and ultimately help to define more clearly the roles of these genes in humans. We report here for the first time SRS patients with chromosomal structural mutations in the H19/IGF2 enhancer region.
Patient 1 was referred for SRS testing at the age of 2 years due to prenatal and postnatal growth retardation, characteristic dysmorphic features, congenital anomalies, and failure to thrive. Patient 2 was referred at age 14 weeks with prenatal and postnatal growth retardation, typical facial features, and severe feeding difficulties. Subsequently his sister (patient 3) presented neonatally with the same clinical features. Patient 4 was identified after karyotyping due to recurrent miscarriages. She had prenatal and postnatal growth retardation and atypical diabetes, but there was no clinical suspicion of SRS. Informed consent for the investigations performed and publication was obtained for all patients. Patients 1 and 4 were recruited via the initial diagnostic testing, while patients 2 and 3 were recruited into the research study ‘Imprinting Disorders Finding Out Why?’ (IDFOW), approved by Southampton and South West Hampshire Research Ethics committee 07/H0502/85.
Molecular genetic investigations
Karyotyping was performed on cultured lymphocytes using standard techniques. Fluorescence in situ hybridisation (FISH) analysis was performed on metaphases by standard techniques using bacterial artificial chromosome (BAC) and fosmids clones. Genomic DNA was extracted from blood leucocytes by standard procedures.
Multiplex ligation dependent probe amplification
Multiplex ligation dependent probe amplification (MLPA) analysis was performed using the methylation specific MLPA kit ME030 version B1 from MRC-Holland (Amsterdam, The Netherlands) following the manufacturer's instructions; 250 ng of DNA was used for analysis.
Genomic DNA (2 μg) was bisulfite treated using the EZ-DNA Methylation kit, according to the manufacturer's instructions (Zymo Research, Orange County, California, USA), except that DNA was eluted in 50 μl of elution buffer and 50 μl of TE pH 8.0.
Pyrosequencing assays of bisulfite induced C/T polymorphisms12 for H19 and IGF2P0 were designed using Biotage Assay Design Software v1.0.6 (Biotage, Uppsala, Sweden) and performed essentially as previously described.13 Briefly, primary amplification reactions (50 μl, performed in duplicate) contained forward and reverse primers, one of which was biotinylated, titanium Taq and buffer (Clontech, Mountain View, California, USA), and ∼20 ng bisulfite treated DNA. After 45 amplification cycles, 20 μl single stranded biotinylated PCR products were pyrosequenced using Pyrosequencing Gold Reagents (Biotage) according to the manufacturer's instructions. The percentage methylation at each CpG site was determined by AQ software. AQ levels >5% were indistinguishable from background fluorescence. In each pyrosequencing assay, a bisulfite treatment control was included in the form of a cytosine not in a CpG dinucleotide; this would be expected to be converted entirely to uracil.
Primer sequences were: H19 pyrosequencing: forward GTATAGTATATGGGTATTTTTGGAGG, reverse CCATAAATATCCTATTCCCAAATAACC (bio); sequencing GTTTYGGGTTATTTAAGTT. This amplifies the region chr11: 1977625-1977877.
IGF2(P0) pyrosequencing: forward TGAGGATGGGTTTTTGTTTGGTAT; reverse TCCTCAATCCACCCAAAATAATAT (bio); sequencing AAAAGTTATTGGATATATAGT.
Array comparative genomic hybridisation analysis
Array comparative genomic hybridisation analysis (aCGH) used a custom 8×60 K Agilent array (Agilent Technologies, South Queensferry, UK) designed to densely cover known human imprinted loci. Probes were designed using Agilent Technologies' e-Array design tool with the following filters applied: Tm filter, similarity filter, and catalogue probe preference. Across the region chr11: 1674000-2979000 (hg18) probes were tightly tiled to provide maximum resolution with an average spacing of 50 nt. The remainder of chr11 was represented by Agilent aCGH control probes, which provide a genome wide backbone to enable efficient normalisation. aCGH was performed according to the manufacturer's instructions. Briefly, 500 ng of blood genomic DNA was labelled using the genomic DNA enzymatic labelling kit, hybridised to the array for 24 h and subsequently washed. Arrays were scanned using an Agilent Technologies' DNA microarray scanner G2539A and data extracted using Agilent Technologies' Feature Extraction software version 10.5.1. Duplications and deletions were mapped and end points determined by Agilent Technologies' DNA Analytics software version 4.0 with the following settings: data were analysed using the ADM-2 algorithm with a threshold of 6.0 using the DLR error model, with fuzzy zero turned on and the continuous moving average applied.
The clinical features of the patients are summarised in table 1. Cytogenetic analysis of patient 1 showed a normal karyotype, 46,XY while MLPA analysis showed deletion of H19 and part of H19-ICR (figure 1A); the peak height of the deleted probes suggested mosaicism, while MLPA analysis of parental DNA was normal, suggesting that the deletion had arisen de novo. Microsatellite analysis showed reduced dosage ratios of D11S4046 (paternal:maternal ratio 1:2.3) consistent with mosaicism (figure 1C). Cytogenetic analysis of patients 2 and 3 revealed a pericentric inversion of chromosome 11 (karyotypes 46,XY,inv(11)(p15.5q21)pat and 46,XX,inv(11)(p15.5q21)pat, respectively), inherited in both children from their phenotypically normal father. FISH analysis showed a split signal using BAC clone RP11-534I22, while fosmids W12-1875E5 and W12-1660M14 hybridised to 11p and 11q, respectively (figure 1A,D), mapping the 11p15.5 breakpoint to chr11:1862216-1921555 (hg18).
Patient 4 had a balanced translocation (karyotype 46,XX,t(1;11)(p36.22;p15.5)). FISH analysis mapped the 11p15.5 breakpoint between W12-1875E5 and W12-1660M14.14 Paternal origin of the translocation was inferred through karyotyping of the patient's mother, who had a normal karyotype, and the patient's daughter, who had the same translocation as the patient but normal growth.
DNA methylation analysis showed partial hypomethylation of both H19-ICR and IGF2P0 in patient 1. In patients 2, 3, and 4 IGF2P0 was partially hypomethylated while the methylation of H19-ICR was in the normal range (figure 1B).
All four patients were analysed by aCGH. In patient 1, aCGH showed a 58.8 kb deletion (chr11:1918222-1977026 (hg18)) encompassing the endodermal, skeletal muscle specific and part of the mesodermal enhancers, as well as H19 and part of H19-ICR including the seventh CTCF binding site, but not the two repeat blocks (figure 1A), though the log intensity difference suggested a mosaic deletion. In patients 2 and 3 aCGH showed an 8.8 kb deletion (chr11:1918312-1927132 (hg18)) overlapping with the mesodermal enhancer. Deletions partially overlapping with the above deletions have been registered as rare variations in apparently normal individuals (http://projects.tcag.ca/variation/?source=hg18), but the parental origin of these deletions are unknown. aCGH did not reveal any deletion in patient 4.
We present three pedigrees with disruption of the shared H19/IGF2 enhancer region and separation of the mesodermal enhancers from IGF2, one by deletion, one by an inversion breakpoint accompanied by deletion, and one by breakpoint of an apparently balanced translocation. IGF2 expression was reduced in mesodermal tissue in patient 4.14 Previously, the small stature of patient 4 was unaccounted for, but we now hypothesise that hypomethylation of IGF2-P0 due to the separation of IGF2 from its enhancers causes growth retardation. The deletion in patient 1 may truncate these enhancers and moreover reduces their distance from IGF2 by 50%, thus potentially affecting the local three dimensional structure of the DNA and hence its function. We therefore suppose that the mesodermal enhancers are non-functional in all four patients. It is interesting that all four patients are growth retarded, but only patients 1, 2, and 3, who harbour deletions including the sequence chr11:1918312-1927132 (hg18), presented with the clinical diagnosis of SRS. One interpretation of this observation is that reduced IGF2 expression from mesodermal enhancers is responsible for only the growth retardation seen in SRS, while other genetic or epigenetic defects cause the other phenotypic traits of SRS. Alternatively, given that SRS is a disorder with pronounced clinical heterogeneity, it may be that patient 4 showed an atypical presentation.
It was striking that all patients showed hypomethylation of IGF2P0, while hypomethylation of the H19-ICR was observed only in patient 1. This observation suggests that IGF2P0 methylation may not directly depend on H19-ICR methylation in cis, but may instead reflect the promoter activity of IGF2P0 per se, and be established as a consequence of its interaction with distant IGF2 enhancers. Disruption of the enhancer region represents a possible explanation of rare SRS cases with hypomethylation restricted to IGF2P0.9
The effects of deletions and mutations in the H19 region depend not only on their parent of origin, but also on their localisation. Deletions or mutations of H19-ICR described to date are apparently without phenotypic effect when on the paternal allele, whereas in maternal inheritance they are associated with the overgrowth disorder Beckwith–Wiedemann syndrome (MIM 130650).15 16 Thus, we hypothesise that a paternal deletion of H19, such as seen in patient 1, may be of less clinical significance than the separation of IGF2 from its enhancers. For the enhancer region, no data have hitherto been presented in humans; however, mice with paternally inherited deletions were 80% of normal size whereas mice with maternal deletions were of normal size.17 We predict that detailed characterisation of the IGF2 promoter and enhancer elements in humans will uncover further genetic and epigenetic changes that predispose to growth retardation and/or SRS. Moreover, we confirm that IGF2P0 hypomethylation is found in a proportion of SRS cases with normal H19-ICR methylation. Therefore, in agreement with Bartholdi et al,9 we suggest that IGF2P0 methylation analysis should be included in standard molecular testing for SRS.
MLPA was performed by Judy Boe Rasmussen, The Kennedy Center; FISH was performed by Viv Maloney, Wessex Genetics Service. Jette Bune Rasmussen performed the graphic work.
Competing interests None.
Patient consent Obtained.
Ethics approval This study was conducted with the approval of the Southampton and South West Hampshire research ethics committee 07/H0502/85.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.