Introduction

Silver–Russell syndrome (SRS) was independently described by Silver1 and Russell.2 It is a clinically and genetically heterogeneous disorder that is characterized by short stature of prenatal onset, body asymmetry and a triangular face. Although mostly sporadic, in rare familial cases SRS is transmitted in an autosomal dominant or recessive, or X-linked dominant way.3 Diagnosis can be difficult owing to the variable presentation of phenotypes. Maternal uniparental disomy of chromosome 7 occurs in about 10% of cases and is associated with a milder phenotype.4, 5 In a previous study, hypomethylation of the differentially methylated IGF2/H19 imprinting center region (ICR1) on chromosome 11p15 was found in approximately 30% of patients,6 using Southern blot analysis of an HpaII restriction site close to CTS3.

The same ICR is hypermethylated in about 15% of patients with Beckwith–Wiedemann syndrome (BWS), which is characterized by pre and postnatal overgrowth.7, 8 This suggests that SRS is genetically and clinically opposite to BWS.9 However, in 60% of cases, BWS is caused by hypomethylation of a second differentially methylated region at 11p15 (ICR2), which regulates expression of the transcript KCNQ1OT1.10, 11 This region has not been found to be epigenetically altered in patients with SRS.

The ICR1 is flanked by two oppositely imprinted genes, IGF2 and H19. The expression of these genes is controlled by competition of their promoters for transcription activation from shared enhancers located downstream of the H19 gene. The unmethylated, maternally derived ICR1 is bound by the zinc finger-binding factor CTCF, thus acting as a chromatin boundary. This blocks the interaction of IGF2 with the enhancer and promotes H19 transcription. On the paternal allele, the ICR1 is methylated and CTCF-binding and enhancer-blocking activity is abolished, thereby allowing expression of IGF2.

The ICR1 encompasses seven potential CTCF-binding sites (CTSs) that show allele-specific methylation.12 Based on the observation that only CTS6 methylation is correlated with the expression of either IGF2 and H19 in normal human embryonic tissue, a key regulatory role for CTS6 was proposed by Takai et al.13

In previous studies on patients with SRS, methylation patterns were determined at the H19 promoter or a region close to CTS3 within the ICR16, 14, 15 by Southern blot analysis using methylation-sensitive restriction enzymes SmaI and HpaII, respectively. Here, we have analyzed the degree of methylation of the HpaII site close to CTS3 and CTS6 by quantitative analysis of methylated alleles (QAMA), which is a bisulfite-based real-time PCR methylation assay,16 and of various loci located from 300 bp upstream to 1800 bp downstream of CTS6 by multiplex ligation-dependent probe amplification (MLPA).

Materials and methods

Patient information

Patient Esr1 is the second child of a healthy non-consanguineous Turkish couple. The mother was 22 and the father was 29 years of age at the birth of the patient. Pregnancy was normal, and the girl was born at 42 weeks of gestation by cesarean section. Birth measurements were normal (Table 2). Body asymmetry was noted after birth with the left arm shorter and thinner than the right one. Later, a shorter left leg and foot were observed. The girl has 2/3–syndactyly of the toes, like her father. Motor and speech development are normal. Body measurements were low but in normal range at the age of 2¼ years (length 85 cm (10th centile), weight 11.5 kg (25th centile), OFC 46 cm (10th centile)). Further 19 SRS patients were comprised in this study. They all showed severe intrauterine and postnatal growth retardation (<3rd percentile) and at least three further signs typical for SRS according to Price et al.5 All 17 patients indicated by ‘AC’ in the patient ID are a subset of a group of 51 patients previously described by Eggermann et al.6

DNA isolation from blood and sperm

DNAs were extracted and purified from blood and sperm cells using the FlexiGene DNA Kit (Qiagen, Hilden, Germany), following the manufacturer's instructions. Semen samples were obtained from patient Esr3 and from two normal volunteers whose ejaculate was considered to be normal on the basis of WHO criteria. We used a swim-up technique for sperm separation as described elsewhere.17 During the incubation, the motile sperm migrated into the medium, whereas the immotile and sluggish sperm and the particulate matter of semen remained in the pellet. After the incubation period of 1 h, the medium, which was enriched with motile sperm, was withdrawn and the sperm cells were used for DNA isolation. Semen sample of patient Esr3 was within the normal range with respect to sperm number (38.5 × 106/ml) and all other tested parameters such as motility and morphology.

Methylation analysis

Southern blot analysis

Southern blot analysis was performed as described previously.6, 14

Quantitative analysis of methylated alleles

Bisulfite treatment: the procedure was modified from established protocols.18 Genomic DNA (1 μg in 30 μl) was denatured by adding 3 μl freshly prepared 3 M NaOH and incubating the solution at 37°C for 15 min. For complete denaturation, the samples were incubated at 95°C for 1 min and immediately cooled on ice. The bisulfite solution was prepared by dissolving 8.5 g of sodium bisulfite in 15 ml degassed water, adding 1 ml of a 40 mM hydroquinone solution and adjusting the pH to 5.1 with 1000 μl of 10 M NaOH. The bisulfite solution (0.5 ml) was added to the denatured DNA, mixed and incubated at 50°C for 16 h in the dark. The DNA was recovered by using the Wizard DNA Clean-Up System (Promega) followed by elution in 100 μl water. Subsequently, 11 μl of 3 M NaOH was added and the samples were incubated for 15 min at 37°C. The solution was then neutralized by adding 110 μl of 6 M NH4Oac, pH 7.0. The DNA was ethanol precipitated, washed in 70% ethanol, dried and resuspended in 20 μl water.

Rationale and methodological details of the real-time PCR-based QAMA assay are described elsewhere.16 In brief, PCR was performed using a 96-well optical tray with caps at a final reaction volume of 20 μl. Samples contained 10 μl of TaqMan® Universal PCR Master Mix, No AmpErase® UNG (uracil–N–glycosylase), 2 μl of bisulfite-treated DNA, additional 2.5 U AmpliTaq Gold (Perkin Elmer), 2.5 μ M each of the primers and 150 nM each of the fluorescently labeled probes H19met and H19unmet. Initial denaturation at 95°C for 10 min to activate AmpliTaq Gold DNA Polymerase was followed by 40 cycles of denaturation at 95°C for 15 s and annealing and extension at 60°C for 1 min (ABI Prism, 7000, Sequence Detection System). PCR primers were designed to amplify the bisulfite converted antisense strand sequence at CTS3 and CTS6 lacking any known nucleotide polymorphisms. Degree of methylation is measured by TaqMan minor grove binder (MGB) probes that bind to their respective target sequence. Therefore, the amplification of the methylated and unmethylated alleles is monitored independently in the same tube. The binding site of the CTS3 and CTS6 MGB probes cover 1 and 3 CpG dinucleotides, respectively. We used VIC-labeled MGB probes that specifically hybridize to the sequence from the unmethylated allele and an FAM-labeled probe that binds to the sequence generated from the methylated allele. The improved sequence specificity inherent to the MGB technology, which is widely used to discriminate single-nucleotide polymorphisms (SNP), allows reliable discrimination even of single base pair changes.19, 20 The cycle number at which the fluorescence signal crosses a detection threshold is referred to as CT. The difference of both CT values within a sample (ΔCT) is calculated (ΔCT=CT−FAM−CT−VIC) and is a measure for the degree of methylation. All samples were measured in duplicate using the mean for further analysis. The relative prevalence of either the methylated or the unmethylated allele was set to 100% in case only one fluorescence signal crossed the threshold, indicating a relative absence of the opposite target. For precise quantification, a standard curve consisting of bisulfite-treated, defined mixtures of SssI-methylated and unmethylated PCR products (CTS6) and SssI-methylated and unmethylated genomic DNA (CTS3) is implemented in every run.16

QAMA primer and probes:

H19.3fw: 5′-CATAACACATAAATATTTCTAAAAACTTCTCCTTC-3′ (7872–7906); H19.3rev: 5′-GGGTTGTGATGTGTGAGTTTGTATTG-3′ (7972–7997);

H19unmet: 5′-VIC-AATTATAAAATCAAAAATAACCACA-MGB-3′ (7941–7965); H19met: 5′-6FAM-AAATCGAAAATAACCGCG-MGB-3′ (7948–7965).

H19CTS3fw:5′-cttgcttcctggcacgagGGGTTTTTGGTAGGTATAGAAATTG-3′ (5592–5617), H19CTS3rev: 5′-caggaaacagctatgacCCCCATCCAAAAAAAACTTAAAC-3′ (5749–5771), H19CTS3unmet: 5′-VIC-TTTATTATTTGGATGGTATAGAAT-MGB-3′ (5665–5688), H19CTS3met: 6FAM-TTATTATTCGGATGGTATAGAAT-MGB-3′ (5666–5687). Methylation-discriminating nucleotides in probes are underlined. Uncapitalized nucleotides do not bind to the target sequence but provide a tag that has been shown to improve amplification and facilitate sequence analysis of PCR products. Primers used for CTS6 amplification and sequence analysis are as follows: H19nDNAfw 5′- CCATATCGGGCTACGTGTCT-3′ (7696–7715); H19nDNArev, 5′- GGGCTGTCCTTAGACGGAGT-3′ (8273–8292). All numbering refers to AF125283.

Methylation-specific multiplex ligation-dependent probe amplification (MS-MLPA)

The SALSA ME030 (Lot 0107) MS-MLPA mix was purchased from MRC-Holland (http://www.mlpa.com). The probe mix included in this kit contains 27 probes specific for the BWS/SRS chromosome 11p15 region. Of these, 13 are MS-MLPA probes, containing a HhaI recognition site specific to the differentially methylated regions H19/IGF2 ICR1 and KvDMR, as well as several other sequences. In addition, 15 MLPA probes located outside the BWS-SRS locus are added as control probes for copy-number quantification, as well as three MS-MLPA control probes for complete HhaI digestion in the MS-MLPA reaction. We used a total of 400 ng of genomic DNA for each sample tested. After 16 h of hybridization at 60°C, samples were split equally into two aliquots. The first aliquot underwent ligation only, whereas the second underwent ligation plus enzymatic digestion. The ligation, enzymatic digestion and PCR amplification were performed according to the manufacturer's instructions. PCR products (1 μl) from each tube were mixed with 1 μl of internal size standard and 20 μl of deionized formamide and were injected into an ABI-3100 genetic analyzer (Applied Biosystems) equipped with a 50-cm, 16-capillary array.

The Genescan project was imported into a Genotyper file containing an appropriate table of categories and two macros for peak calling and data tabulation. Analysis of data was performed using a modified Excel spreadsheet, which contained algorithms originally created for BRCA2 MLPA analysis,and was obtained from the Clinical Molecular Genetics Laboratory, Regional Clinical Genetics Service at St James's University Hospital in Leeds, UK (http://www.leedsdna.info/). Data sets for both the test samples and the average of the normal control samples are imported, corrected and internally adjusted in the raw data worksheet, and dosage ratios for the probes (equivalent to dosage quotients) are calculated and listed in the analysis results section along with a probability value (for details, see http://www.leedsdna.info/science/dosage/REX-MLPA/REX-MLPA_analysis_User_Guide.pdf).

Results

Patient Esr1 was referred to us for BWS testing. She had body asymmetry but no sign of overgrowth, omphalocele or macroglossia (Table 2). Using a quantitative real-time PCR-based assay (QAMA),16 which determines the methylation status of three CpG dinucleotides at CTS6, we observed ICR1 hypomethylation. This finding pointed to SRS rather than BWS. CTS3 methylation by Southern blot analysis was judged to be normal. To analyze whether these results reflected different methylation patterns between CTS3 and CTS6 or might be due to false interpretation of the results of Southern blot analysis, we determined the degree of methylation of the HpaII site close to CTS3 by QAMA. We found similar degrees of methylation of the two CTCF-binding sites, although CTS3 hypomethylation was slightly less pronounced than CTS6 hypomethylation. This suggested that methylation analysis by QAMA might be more sensitive in detecting ICR1 hypomethylation than Southern blot analysis of CTS3.

Therefore, we used QAMA to analyze CTS3 and CTS6 methylation in five patients with SRS showing hypomethylation of the HpaII site close to CTS3 by Southern blot analysis and another 14 patients suspected to have SRS but who were judged to be normal by Southern blot analysis. In all five patients with CTS3 hypomethylation by Southern blot analysis, we observed hypomethylation at CTS3 and CTS6. Among the 14 patients who were normal at CTS3 by Southern blot analysis, we found CTS3/CTS6 hypomethylation in five patients (Table 1). The degree of methylation in patients with hypomethylation varied from 1 to 16% and from 2 to 22% at CTS6 and CTS3, respectively (Table 1). This is significantly lower than the methylation observed in a panel of 50 DNA samples from normal individuals, which was 35 to 55% at CTS6. CTS3 methylation ranged from 38 to 50% in 13 normal individuals (data not shown).

Table 1 Results of methylation analysis
Full size table

In a study by Tost et al,21, 22 H19 methylation analysis using bisulfite sequencing was significantly impaired by a frequent SNP affecting a primer-binding site. To exclude this possibility in our study, we sequenced 558 bp around CTS6 in PCR products obtained from genomic DNA of all patients showing discordant results in QAMA and Southern blot analysis and could not detect any SNPs that might affect probe or primer binding.

To confirm the methylation results by an independent method, we applied the MLPA ME030 BWS/RSS probe mix to the SRS samples. This probe mix encompasses probe sets to determine the methylation status of five ICR1 loci located between 300 bp upstream and 1800 bp downstream of CTS6 (Figure 1). Here, we ignore the results obtained with probe set N0213-L9999 (covering CTS6) as we found that probe binding is frequently impaired by SNP rs10732516. The remaining methylation-sensitive H19 probe sets showed, within measure accuracy, similar degrees of methylation as determined by QAMA (Table 1). Hypomethylation at the IGF2 locus (7175-L6784) was less pronounced in patients showing hypomethylation at CTS6 (Table 1).

Figure 1
figure 1

Map of the H19/IGF2-ICR1. Map positions of probes and primers used for QAMA and the probe sets used for methylation-sensitive MLPA within this region are plotted; (1) 6264-L5770; (2) N0213-L9999; (3) 6265-L5771; (4) 7177-L6786; (5) 6266-L5772; (6) 7175-L6784 MLPA probes are described in the MRC-Holland ME030 BWS/RSS probemix protocol. HpaII restriction site was used for Southern blot analysis as described elsewhere.14 Nucleotide positions are indicated in base pairs according to numbering of AF125183.

Full size image

As transmission of SRS has been known to occur, we tested ICR1 methylation in sperm DNA from patient Esr3 and two normal control individuals. In all samples, we found complete methylation of CTS3 and CTS6 by QAMA and almost complete methylation with H19 probe set 6264-L5770 and IGF2 probe set 7175-L6784 by MLPA. This result is consistent with a normal sperm-specific methylation pattern as published by Kerjean et al.23 The remaining H19 MLPA probe sets did not show consistent methylation patterns in six sperm DNA samples from normal controls (data not shown).

Methylation sensitive MLPA probe sets within the KvDMR (7171-L6780, 7172-L6781, 7173-L6782 and 6276-L5782) and IGF2 regions (probe set 6269-L5775) revealed an unaltered methylation pattern in all patients when compared with blood DNA samples from normal controls but showed complete loss of methylation in all sperm DNAs (data not shown). The degree of methylation of the probe set KCNQ1 (3553-L2919) was the same in all patient, sperm and normal blood samples, suggesting that this locus might lack parent-of-origin-specific methylation.

Discussion

In a recent study, Eggerman et al found epigenetic alterations at 11p15 by Southern blot analysis in approximately 30% of patients with an SRS phenotype.6 In this study, the methylation status of an HpaII restriction site close to CTS3 within the differentially methylated H19/IGF2 ICR1 was monitored. Here, we show that hypomethylation as determined by Southern blot analysis could be confirmed at CTS3 and CTS6 by QAMA and at additional ICR1 loci by MLPA in all patients studied. Furthermore, using QAMA and MLPA, we were able to identify ICR1 hypomethylation in 5 out of 14 patients (35%) that show signs typical for SRS according to Price et al,5 but were judged to be normal by Southern blot analysis. On the basis of these results and the detection rate as previously determined by Eggermann et al,6 we can roughly estimate that about half of all patients with a typical SRS phenotype can be detected by methylation analysis.

It is interesting to note that patient Esr1, although hypomethylated at CTS3 and CTS6, does not show a typical SRS phenotype. This patient presents with prominent body asymmetry, described as hemihyperplasia, without any other features of SRS or BWS. This is distinct from the observation made in previous studies,14, 15 in which all patients with H19 hypomethylation had pre and/or postnatal growth retardation. Hemihyperplasia is an abnormality of cell proliferation leading to asymmetric overgrowth of one or more regions of the body (OMIM, no 235 000). It is associated with an increased tumor risk and can occur isolated or as a feature of overgrowth syndromes such as BWS. In both BWS and isolated hemihyperplasia, hypermethylation of the H19/IGF2 ICR1 was found.24 In patient Esr1, the opposite genetic alteration, ICR1 hypomethylation, is present. This suggests that body asymmetry in this patient is caused by growth restriction of cells of one part of the body rather than overgrowth of the opposite part. Therefore, body asymmetry in patient Esr1 is best described as hemihypoplasia, thus representing a very mild phenotype within the clinical spectrum of SRS. In some cases, it might be difficult to clinically distinguish between hemihypo and hemihyperplasia. However, the classification of patients is crucial, as an increased risk for a broad range of tumors such as Wilm's tumor has been reported for patients with hemihyperplasia.25 In contrast, it is not to be expected that hypomethylation of the H19/IGF2 ICR1, as present in patients with hemihypoplasia, is associated with an increased tumor risk. But this has to be investigated in a larger group of patients with isolated body asymmetry. We recommend ICR1 methylation testing of patients with isolated body asymmetry using QAMA or MLPA to distinguish between patients with or without increased tumor risk.

Detailed clinical data from the patients who show hypomethylation by QAMA and MLPA but not by Southern blot analysis are listed in Table 2. Among these, patient AC28 shows the lowest degree of methylation at CTS3 and CTS6 and the most severe pre and postnatal growth retardation, suggesting a correlation between the degree of hypomethylation and the severity of the SRS phenotype. Such a correlation has previously been observed in the studies by Gicquel et al14 and Bliek et al,15 using Southern blot analysis to determine ICR1 and H19 promoter methylation, respectively.

Table 2 Clinical data of patients who were judged to be normal at the HpaII site close to CTS3 but show CTS6 and CTS3 hypomethylation by QAMA
Full size table

To analyze the possibility that the ICR1 epimutation might be transmitted through the male germ line and thus increase the risk for SRS in offspring, we analyzed the degree of CTS3 and CTS6 methylation in spermatozoa DNA from patient Esr3. We found both sites to be fully methylated, consistent with a normal sperm-specific methylation pattern.23, 26, 27 These results suggest that the epimutation is reversed in the male germ line of patient Esr3. However, it is possible that the primordial germ cells of patient Esr3, who is mosaic for the epimutation, are not affected by ICR1 hypomethylation.

Several reasons might account for the observation that QAMA and MLPA are superior to Southern blot analysis of the HpaII restriction site in SRS testing. QAMA, which is a bisulfite-based real-time PCR assay, enables precise quantification of DNA methylation. In Southern blot analysis, cases with only a slight hypomethylation might be missed. This is supported by the observation that four of the five SRS patients that were judged normal by Southern blot analysis but were hypomethylated by QAMA show a moderate degree of hypomethylation at CTS3, whereas three out of five patients that were diagnosed correctly by Southern blot analysis show almost complete loss of methylation. The false-negative findings by Southern blot analysis can also be caused by insufficient HpaII digest incomplete DNA transfer or a high background in single lanes mimicking an altered methylation.

A prominent role of CTS6, as discussed by Takai et al,13 is not evident in the present study, as the degree of CTS3 and CTS6 methylation is highly similar in all samples. MLPA probe sets located about 1.8 kb downstream of CTS6 gave concordant results with the QAMA analysis in blood DNA of all patients, but the degree of methylation was inconsistent in sperm DNA, ranging from 32 to 87%. This suggests that the methylation imprint might be less stable in this region, at least in sperm DNA. However, except for the CTS6 region, we cannot exclude the possibility that SNPs that are observed at high density in the ICR1 are the cause of the observed inconsistencies at the analyzed MLPA loci. In summary, we conclude that MS-MLPA and bisulfite-based methylation analysis of CTS3 and CTS6 are superior to Southern blot analysis of HpaII close to CTS3 in detecting ICR1 hypomethylation, that ICR1 hypomethylation occurs more often in SRS patients than previously thought and that ICR1 hypomethylation can underlie isolated hemihypoplasia.