Background

The Silver-Russell syndrome (SRS; OMIM #180860) is one of the imprinting disorders characterized by prenatal and postnatal growth restriction, relative macrocephaly, body asymmetry and characteristic facial features. To date, more than 400 SRS cases have been reported since it was initially described by Silver et al. in 1953 [1] and Russell in 1954 [2]. The clinical and genetic heterogeneity of SRS make it difficult to define its clinical diagnostic criteria and genetic etiology. Several scoring systems for clinical diagnosis of SRS have been proposed [3,4,5,6,7]. Most recent proposal suggested that a patients can be considered to have likely SRS if at least four out of the following six criteria were met: (1) small for gestational age, birth length and/or weight ≤ −2SDS, (2) postnatal growth retardation (height ≤ −2SDS), (3) relative macrocephaly at birth, (4) body asymmetry, (5) feeding difficulties and/or body mass index (BMI) ≤ −2SDS in toddlers; (6) protruding forehead at the age of 1-3 years (Netchine-Harbison SRS Clinical Scoring System). About 10% of SRS case had maternal uniparental disomy of chromosome 7, so far about 60 SRS patients with matUPD(7) had been reported [8]. The imprinting genes on chromosome 7 are believed to be involved in the pathogenesis of the syndrome [9,10,11,12,13], the specific causal gene(s) are yet to be identified.

Five cases with segmental UPD(7q)mat in patients with SRS phenotype had been documented in literature since 2001. The only mosaic segmental UPD(7) that was reported by Reboul et al. in 2006 revealed a 7q21-qter mosaicism in a patient with SRS phenotype [14]. Identifying segmental UPD(7) in patients with SRS may help to narrow-down causal genes and regions.

Here, we reported a second case with mosaic segmental UPD(7q) mat, the patient presented with the main phenotypes of SRS [15]. We compared the clinical findings involving mosaic UPD(7q)mat documented in literature and the finding support the notion that imprinted genes on 7q contribute to the pathogenesis of SRS, even in a mosaic status.

Case presentation

The patient was a six-year-old boy came to hospital due to severe developmental delay, short stature and mild dysmorphic features. He was a full term first child with two apparently healthy younger siblings. He was delivered via cesarean section without complication during pregnancy or delivery. His birth weight was 1.91 kg(<−3SD), indicating small for gestational age. His parents were non-consanguineous without family history of congenital anomalies. The paternal height was 162 cm, and the maternal height was 138 cm, the mother was not investigated for SRS and the height of the maternal familily members were not applied. The patient was reported to have failure to thrive since birth. During his initial hospital visit at the age of 6, his height was 91.9 cm (~ − 6 SD), weight 9.5 kg(~ − 6SD) and head circumference 49 cm (~ − 1SD, thus a relative macrocephaly). The bone age was delayed. Mild dysmorphic features were noticed including triangular face, moderate protruding forehead, relative macrocephaly, fifth toe clinodactyly and irregular teeth (Fig. 1a). Although the obvious body asymmetry was not noticed in this patient, this patient can be clinical diagnosed as SRS based on the Netchine–Harbison clinical scoring system (the first four features are present in this patient) [16]. In addition, he had normal electroencephalogram and electromyography, the growth hormone (GH) stimulating peak was 5.08 μg/L and diagnosed as partial growth hormone deficiency. He was subsequently treated with recombinant human growth hormone at the doses of 1 IU/(kg/d). The treatment lasted 4 months and stopped due to no significant growth response. Cytogenetic and molecular analysis were performed.

Fig. 1
figure 1

a Patient at age 8 years and 3 months. Dysmorphic features including thin and short stature, triangular face, moderate protruding forehead, relative macrocephaly, fifth toe clinodactyly and irregular teeth. b Growth curve of the patient: standard deviations (SD) were calculated based on the standardized growth charts for Chinese children and adolescents aged 0 to 18 [33], according to population-standard charts, the patient’s growth improved but remained ∼5.5 SD below the mean height for age

Full size image

Method

Chromosomal microarray (CMA) and cytogenetic analyses

DNA sample was extracted from peripheral blood lymphocytes by using Lab-Aid DNA kit (Zeesan Biotech Co, Ltd., China). Genomic wide single nucleotide polymorphism (SNP) array analysis was performed by Illumina HumanSNPcyto-12 v2.1 BeadChip array. And the SNP data were collected and analyzed by Illumina Genome Studio and KaryoStudio software, cytogenetic analysis was performed by conventional standard GTG-banding at 400–550 band resolution.

Microsatellite analysis

Eight highly informative microsatellite markers (short tandem repeat, STR) spanning the whole of chromosome 7 (D7S2552, D7S506, D7S510, D7S517, D7S672, D7S2410, D7S2504, D7S523) were selected from Genethon Genetic Maps (http://www.bli.uzh.ch/BLI/Projects/genetics/maps/gthon.html) for parent-of-origin analysis. The forward PCR primers designed for each STR maker were modified at 5′ terminal base with carboxyfluorescein (FAM). All fluorescent PCR products were analyzed on ABI 3130 genetic analyzer and GeneMapper (Applied Biosystems), the size and loci of each STR marker was assigned manually by identifying the peak on the electropherogram.

Results

Uniparental disomy (UPD) of chromosome 7q was detected by CMA, spanned from the region of 7q11 to 7qter. The B allele frequency indicated that the UPD region was homozygous and mosaic (Fig. 2a), and the mosaicism level was estimated to be about 80% according to Conlin et al. [17]. G-band analysis presented normal karyotype. The STR makers analysis revealed that the UPD was maternal origin (Fig. 2b, c).

Fig. 2
figure 2

a The scatterplot representation of SNP microarray revealed the mosaic UPD region from 7q11 to 7qter. b Electropherograms of the D7S2552, D7S506, D7S510, D7S517, D7S672, D7S2410, D7S2503, D7S523 microsatellites in the proband (bottom panel), father (top panel) and mother (second panel). c The results of microsatellite analysis, the Arabic numbers denote the PCR product sizes in bp, D7S672, D7S2410, D7S523 loci revealed the UPD region detected by array was maternal origin

Full size image

Discussion

We reported a rare case of SRS due to mosaic segmental UPD(7q)mat. This is only the second such case ever reported [14] and the first case detected by SNP array. There are more than 60 UPD(7)mat cases documented in the literature since the first case of UPD(7)mat reported by Spence et al. [8, 18], we summarized the clinical features of typical SRS features and UPD(7)mat cases in Table 1. The body asymmetry phenotype was only ~ 39% prevalent among SRS cases. Relative macrocephaly (79.2%), SGA, BL and/or BW (61.6%), prominent forehead (61.6%), clinodactyly V digits (57.0%), feeding difficulties (46.5%), were the most common phenotype in the UPD(7)mat cases. Our patient presented with at least four common features, thus meet the clinical diagnosis of SRS (Netchine-Harbison SRS Clinical Scoring System, NH-CSS) [3].

Table 1 Phenotypic comparison of typical SRS cases and SRS due to UPD(7) mat
Full size table

Mosaic UPD(7)mat was a rare condition. Monk et al. did not find any evidence of somatic mosaicism in three UPD(7)mat probands in lymphocyte and fibroblasts using both Southern blot and FISH analyses [19], subsequently, five mosaic UPD(7) mat cases have been reported (Table 2). Karl Y et al. found the first case of trisomy 7 (T7)/upd(7) mosaicism by amniocentesis, and the fetus showed likely-SRS symptoms with SGA, low-set ears, prominent forehead, triangular face, obstructed micrognathia, clinodactyly of fifth finger bilaterally [20]. Miozzo et al. used haplotyping and cytogenetic-FISH studies characterized a newborn presented SRS phenotype with complete maternal isodisomy 7 (i7) and trisomy 7 mosaicism [21]. Complete maternal heterodisomy 7 and trisomy 7 mosaicism was also reported by Elisabeth et al. [22]. Patient with full UPD(7) mat abnormal cell lineage and normal cell lineage without trisomy 7 also present SRS phenotype [23]. Reboul et al. reported a patient had UPD (7) without the mosaic trisomy 7 [14]. A summary of the clinical features of the five previously reported patients with mosaic UPD(7)mat and our present case is listed in Table 2. Collectively, these cases indicated that SRS phenotype could raise from mosaic UPD(7)mat status.

Table 2 Clinical features of mosaic UPD(7) mat cases
Full size table

Five segmental UPD(7q) mat cases associated with SRS phenotypes had been documented in literature since 2001 [14, 24,25,26] (Table 3). The involved region would span the whole 7q and the smallest region was ~ 31 Mb to 7qter (Fig. 3). Clinical features of segmental UPD(7q)mat were compared in Table 3. The most common were relative macrocephaly and triangular face (5/5). Notably, psychomotor delay was present in the case reported by Eggermann et al. [25] and Begemann et al. [26], but not in the rest of cases. Additionaly, The relevance of 7p for the SRS phenotype currently known are maternal duplication of 7p12.1 (including gene GRB10) [27, 28], These findings are consistent with the notion that the imprinted gene(s) on chromosome 7q are the causes of SRS phenotype [9].

Table 3 The clinical features of segment UPD(7q) mat
Full size table
Fig. 3
figure 3

Schematic representation referred to segment mat UPD(7q) cases reported (UCSC genome browser custom track, hg19)

Full size image

The only previously reported a mosaic segment UPD(7q)mat from 7q21-qter mosaicism by Reboul et al. in a patient presented with SGA, relative macrocephaly, triangular face, severe growth retardation [14]. Our patient had only slightly larger size and presented with very similar phenotypes. The mosaicism level of Reboul’s case was not evaluated.

We detected and confirmed the mosaicism by both microsatellite and SNP array, the formulation to calculated the mosaic level proposed by Bliek et al. [29] by STR maker may be not accurate [30]. The pattern of B allele by SNP array offered an alternative and probable more accurate method for mosaic level assessment. Correlation between the level of mosaicism and SRS phenotype should be performed when more such cases are detected.

Complex and segmental UPD could resulting from either meiotic, mitotic, or meiotic and subsequent mitotic abnormal recombinations. Interestingly, apart from full upid(7)mat mosaicism which was likely result from mitotic non-disjunction and subsequent trisomy rescue, our case present the existence of normal cell line and UPD(7q11-qter) cell line, indicate the mosaicism may arise from somatic recombination [31, 32], further study would need to prove this hypothesis. Meanwhile, mosaicism could be tissue specific, but no more research for other tissue of our patient was performed because the father did not gave the consent.

Conclusions

In summary, we described a second case with rare segmental maternal UPD(7q11-qter) mosaicism. This was the first report of UPD(7q11-qter) mosaicism detected by SNP array. Our case support the notion that there are genes control postnatal growth on long arm of chromosome 7 and indicate that ~ 80% of 7q11-qter mosaicism level was contributed to the SRS phenotype.