Article Text

Download PDFPDF

Epigenetic signatures of Silver–Russell syndrome
  1. Sayeda Abu-Amero1,
  2. Emma L Wakeling2,
  3. Mike Preece1,
  4. John Whittaker3,
  5. Philip Stanier1,
  6. Gudrun E Moore1
  1. 1Institute of Child Health, University College London, London, UK
  2. 2The North West London Hospitals NHS Trust, Northwick Park Hospital, Harrow, Middlesex, UK
  3. 3Department of Epidemiology and Population Health, London School of Hygiene & Tropical Medicine, Keppel Street, London, UK
  1. Correspondence to Sayeda Abu-Amero, Institute of Child Health, University College London, 30 Guilford Street, London WC1N 1EH, UK; s.abu-amero{at}ich.ucl.ac.uk

Statistics from Altmetric.com

Request Permissions

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.

Silver–Russell syndrome (SRS) (OMIM 180860) has been recognised as a clinical entity since the 1950s; however, a molecular genetic cause remained unknown until 1997 when maternal uniparental disomy for chromosome 7 (mUPD7) was found in approximately 10% of affected individuals. The focus then turned to the identification of causative imprinted genes on chromosome 7. Over the next 10 years many of the potential candidate genes identified on chromosome 7 as well as many imprinted loci elsewhere were excluded.1–13 However, in 2006, Gicquel et al identified a major role in the aetiology of SRS for the chromosome 11p15.5 region.14–16 This work, and a succession of comparative, detailed and confirmatory studies, demonstrated an epigenetic signature of hypomethylation at the 11p15.5 imprinting control region 1 (ICR1) in 30–60% of SRS probands (table 1). Here the H19 differentially methylated domain (H19 DMD) regulates the methylation and therefore respective maternal or paternal expression of H19 and insulin-like growth factor 2 (IGF2) genes.

Table 1

Summary of literature on hypomethylation status in SRS patients

Hypomethylation at ICR1 causes epigenetic dysregulation of H19 and IGF2, ultimately resulting in reduced expression of IGF2 and potentially phenotypic growth restriction. Interestingly, the degree of hypomethylation was also found to correlate with the severity of the growth phenotype as well as additional SRS diagnostic features. In comparison, patients with mUPD7 were clinically distinguishable displaying a noticeably milder phenotype.

Since then, the literature includes a number of detailed clinical evaluations with the aim of developing a clinical scoring system to help characterise specific subgroups of SRS caused by mUPD7, hypomethylation of 11p15.5 or those with an unknown aetiology (table 1).31 Nevertheless, due to the broad clinical spectrum of SRS, strict adherence to a scoring system may not be completely reliable and any clinical evaluations must be supported by molecular testing.30 In addition, hypomethylation of ICR1 is not exclusive to SRS as Guo et al32 also observed hypomethylation of the H19 DMR in one of 24 small for gestational age (SGA) placentae with concomitant biallelic H19, which could therefore account for the growth phenotype in a subset of growth restricted cases.

The role of the placenta is of great interest to researchers in fetal growth and development, as many knockout mice studies resulting in aberrant imprinted gene expression show effects mediated by this tissue.33 The collection of placenta from all pregnancies is not routine but would be of profound interest in the investigation of the aetiology of SRS. Data from Yamazawa et al on paraffin embedded placenta tissue from three retrospectively diagnosed SRS patients showed that the degree of hypomethylation correlated with smaller placental size which in turn correlated with reduced IGF2 expression levels.25 The role of the placenta in determining SRS phenotype may be of particular interest due to the following observations. Gicquel et al14 observed hypomethylation in leucocyte DNA of both monozygotic twins who were discordant for SRS but in fibroblasts found hypomethylation only in the affected twin. Meanwhile, Yamazawa et al26 found hypomethylation in the blood leucocyte DNA of the affected twin only; they suggested that the twins described by Gicquel et al may differ due to fetal circulation sharing, which permits transfer and engraftment of blood stem cells between the affected to unaffected.

The accuracy of molecular diagnosis of hypomethylation in SRS patients seems to be dependent on the method used. Southern blotting, which is both labour intensive and requires large amounts of DNA, is quantitative but does not analyse individual CpG sites. Some patients shown to be normally methylated by Southern blotting have been re-evaluated as hypomethylated using newer techniques such as multiplex ligation dependent probe amplification (MLPA).27 34 Methylation specific multiplex ligation dependent probe amplification (MS-MLPA) is the current diagnostic tool of choice as it only requires small amounts of DNA (20–200 ng) and can detect methylation changes as well as copy number variations.35 Methylation sensitive high resolution melt (MS-HRM) has also been used to discriminate allelic methylation patterns.36 37 As a result of the varying sensitivities of the different techniques reported by various research groups, a true picture of the incidence of hypomethylation of ICR1 in SRS patients and the degree of the hypomethylation is questionable. A standard sensitive, diagnostic methodology needs to be agreed upon and implemented to compile an accurate database of hypomethylation in SRS patients.

Unlike transient neonatal diabetes mellitus (TNDM) (OMIM 601410) and Beckwith–Wiedemann syndrome (BWS) (OMIM 130650), where an extended genome wide hypomethylation syndrome has been proposed in a subset of each patient cohort, there is no evidence of aberrant methylation elsewhere in the SRS genome. Rather, the epigenetic change is spatially restricted and that strict hypomethylation of 11p15.5 is unique to SRS.21 24 There is a single case report of an SRS patient harbouring a duplication of the ICR2 only, where the maternally expressed genes KCNQ1, CDKN1C, TSSC5/SLC22A8 and TSSC3/PHDLA2 as well as the paternally expressed sequence LIT1 (KCNQ1OT1) were found.20 Mosaic mUPD11 has been described in only a single SRS case to date implicating the ICR2 region of 11p15.5.23 This could be due to a double dosage of ICR2 being non-viable. These two case reports suggest some involvement of ICR2 in SRS but, due to the small numbers, the significance of this region is not clear.

It has been suggested that assisted reproductive technology (ART) may play a role in imprinting disorders.13 38–40 In a comparative study of normal and oligozoospermic patients, hypomethylation of H19 and hypermethylation of MEST was observed in patients with low sperm count which was suggested to contribute towards the SRS phenotype of children conceived using this technique.41 Kagami et al13 described a girl with SRS conceived using ART, who showed hypermethylation of MEST but not the H19 DMR, suggesting that methylation defects in SRS may not be restricted to the 11p15.5 region. Interestingly, a screen of 54 naturally conceived SRS patients failed to detect any epigenetic changes at the MEST locus.42

Despite the lack of success at identifying imprinted genes responsible for SRS on chromosome 7, evidence for a role of mUPD7 remains consistent at around 10%.15 With a further overall 30% (see total table 1) of patients explained by hypomethylation of the ICR1 on chromosome 11p15.5, this still leaves a challenge to determining the genetic aetiology of the remaining 60%. Chromosome 7 is still a possible location despite the intense molecular study to date. There are now even better analytical techniques such as CpG arrays, suitable for in depth methylation and epigenetic analysis of the entire genome. CTCFL (CCCTC binding factor (zinc finger protein)-like), a testis specific protein, has been shown to be involved in the establishment of methylation at the murine equivalent of ICR1, but was shown not to be involved in hypomethylated SRS patients. This leads the way for other genes involved in the establishment of DNA methylation at imprinted genes, such as DNMT3A and DNMT3L, to be investigated.43 Recently, Spengler et al screened the zinc finger protein 57 (ZFP57) gene for mutations in SRS patients, previously identified as a cause of hypomethylation of multiple imprinted loci in TNDM, but found no evidence that ZFP57 caused hypomethylation in SRS patients.44 The next level of epigenetic regulation may turn out to be a combination of effects on chromosome 7 and 11, perhaps marked by a specific chromatin signature.

References

Footnotes

  • Funding SAA is funded by Wellbeing of Women for this study.

  • Competing interests None.

  • Ethics approval Ethical approval for use of SRS patient samples in this study was obtained from the Joint Research Ethics Committee of the Great Ormond Street Hospital for Sick Children and the Institute of Child Health (1278). Collection and use of the IUGR matched cord blood and placental tissues and normal control cohort was granted by the Hammersmith and Queen Charlotte's and Chelsea Trust Research Ethics Committee (2001/6029).

  • Provenance and peer review Commissioned; not externally peer reviewed.