Prenatal identification and characterization of an sSMC and proper genetic counselling is a challenging task for clinicians. Although it is regarded that about 30% of all de novo sSMC carriers have associated abnormalities, prenatal prediction of the phenotypical consequences is problematic. The first study evaluating the overall risk for an abnormal phenotype in case of a de novo sSMC was based on a large number of amniocentesis samples and found it to be 13% prenatally [13]. By utilizing molecular genetic methods such as FISH, Crolla et al. confirmed that, excluding chromosome 15 derived markers, the risk increased to 28% [14]. A study of 108 prenatally detected cases of marker chromosomes collected from 12 laboratories found that the risk for phenotypical abnormality was 26 % and it was reduced to 18% when the prenatal ultrasound was normal [15]. They reported the highest chance of abnormalities in case of a ring chromosome probably due to the largest amount of genetic material necessary to form the ring appearance. The risk was significantly reduced when the marker appeared in a monocentric and non-satellited form, while it was higher with bisatellited dicentrics, acentrics or isodicentrics.
It seems that somatic mosaicism is often associated with sSMCs. Chromosomal mosaicism is one of the main difficulties in prenatal diagnosis. It represents the phenomenon of the presence of two or more chromosomally different cell lines in an individual arising from a single zygote. The main mechanism of mosaicism forming an sSMC involves the maternal meiosis I. or II. chromosomal non-disjunction error followed by incomplete trisomy rescue in the dividing pre-implantation embryo [16]. It is predicted that meiosis II. segregation errors occur more frequently then meiosis I. errors and it is strictly connected to AMA. It has been established that ovarian aging is the most important factor not only in aneuploidies but in the formation of de novo sSMCs as well [17]. Chromosomal mosaicism in CVS and in amniocytes is well-known and occurs in 1-2% of CVS and 0.1-0.3% of amniocentesis samples [18]. The differentiation of the cells and the tissues begins at the early post- fertilization stage. The distribution of normal and abnormal cell lines in the fetus and the placenta depends on the stage and the mechanism of the differentiation. When trisomy rescue occurs soon after fertilization, the mosaic formation regards both placental and fetal tissues, when it occurs at a later stage (following the separation of the fetal and the placental compartments), the aneuploid cell line can be confined to the placenta, to the fetus or both. The prenatal study by Graf et al. reported a total of 61% rate of mosaicism in sSMC and found no difference between the groups with or without phenotypic abnormality [15]. According to a large review that studied 3124 sSMC cases previously reported in the literature, the authors found 52% overall rate of somatic mosaicism. Non-acrocentric derived sSMCs were more involved in mosaicism. The authors emphasized that in the vast majority of the cases there was no correlation between the grade of somatic mosaicism detected in the peripheral blood or in amnion cells and the severity of the clinical status [19]. In a recent survey of 143.000 consecutive prenatal diagnosis the frequency of overall mosaic sSMC was 69% and the risk of confirmation in amniotic fluid following mosaic CVS result was 33.3%, suggesting a high rate of confined placental mosaicism (CPM) [3]. The main indication for the invasive procedure was AMA and ultrasound anomaly. It seems that sSMCs derived from chromosome 16 are relatively rare and it was found that 91% were mosaics [19]. sSMCs can be associated with uniparental disomy, either in complete or segmental forms, as a result of trisomic zygote rescue [4,5]. Mosaic trisomy with UPD occur at a significantly high frequency from chromosome 16. However, chromosome 16 does not seem to be involved in imprinting mechanisms with clinical consequences.
According to GRCh38/hg19 chromosome 16 has a size of 90.4 Mb. The proximal short arm of the chromosome contains a several copy number variation (CNV) hotspots that predispose to deletions and duplications. The chromosome 16p11.2 duplication syndrome (OMIM 614671) represent a continuous gene duplication syndrome with genomic coordinates (GRCh38:28,500,000-35,300,000). The typical region is an approx. 600 kb genomic duplicate/ deletion from 29,5-30,1 Mb associated with developmental delay and obesity [20]. More distant starting from the centromere is a large microscopically visible region of 8-9 Mb in 16p11.2-16p12.1 that was reported with developmental delay and autism spectrum disorder (ASD) [21,22]. Although most affected patients show different dysmorphic features, mental retardation and behavioral problems, the wide range of phenotypical spectrum refers to incomplete penetrance and variable expressivity of these genomic abnormalities [23]. Moreover, while patients with a deletion of that region has severe obesity besides developmental delay, affected individuals with a duplication are characterized by reduced postnatal weight and low BMI [24]. Thus, it seems that the phenotypes of the duplication carriers mirror those of the deletion carriers [25]. From centromere to telomere, the proximal part of the short arm close to the heterochromatic region is prone to CNV formation. The whole region can undergo duplication together with the heterochromatic blocks forming visible, unusual G-banding pattern [26]. The centric euchromatic region of chromosome 16 is in close proximity to the large block of heterochromatin and this centromere-near region colocalizes with an euchromatic variant (EV) [27]. The EV that is mapped to 32.0-34.4 Mb needs to be distinguished from the potentially pathological duplications found in 16p11.2-12.1 region [27,28]. Euchromatic variants of proximal 16p11.2 are not associated with phenotypical consequences and can be mistaken for the more distantly positioned 16p11.2 duplication syndrome (Figure 2.). However, the overlapping genetic position of the two genomic regions can make the differentiation difficult. The differentiation between them is not feasible during the conventional cytogenetic analysis but can be distinguished at molecular level using FISH or chromosome microarray analysis (CMA, aCGH). Recently published studies indicated that the novel microarray methods such as CMA combined with molecular cytogenetic analysis is particularly effective in the rapid and accurate diagnosis of sSMCs or copy number variations. These combined protocols are especially useful in identifying rare structural chromosomal aberrations prenatally and in assessing the prognosis of fetuses carrying such abnormalities [29]. It is worth of noting that although aCGH identifies the whole gene component of sSMCs and the other underlying chromosomal anomalies too (except for balanced translocations) in one test, it may miss to diagnose sSMCs formed of heterochromatin and the cases with low level of mosaicism [30].
By reviewing the literature and the genome browsers we found altogether 3 cases with copy number variations that were reported in the same chromosomal region that our case. In the Decipher v9.30 database one patient (356289) with a dedicated 1.75 Mb heterozygous de novo microduplication was indicated to have a pathogenic abnormality, another patient (356289) with a 1.54 Mb heterozygous de novo microduplication a likely pathogenic and a third patient (402189) with a 3.1 Mb de novo mosaic microduplication was indicated also likely pathogenic. We did not find the corresponding references in the literature and the morbid genes in the abovementioned chromosomal regions. According to the sSMC database by Liehr [31] the centromere-near region of chromosome 16 contains altogether 96 cases, of them 38 cases (39.5%) are without any clinical symptoms, 16 cases (16.7%) are associated with clinical findings and 37 cases (38.5%) remain without clear clinical correlation. It is interesting to note that in the cases without any clinical phenotype the rate of prenatal diagnosis was about 47%, in cases with clinical findings it was about 19% (3/16) and in cases of uncertain findings it was approx. 81%. It seems that most patient with a sSMC and with clinical findings were diagnosed postnatally and presumably the symptoms of the newborn made the genetic investigations necessary. At the same time in the cases with unclear clinical correlation most diagnosis was performed prenatally underlining the difficulties of the prenatal assessment diagnosing an sSMC. In terms of breakpoints and the corresponding chromosomal regions none of the cases in the database were similar to our case. However, one fetus (case 16-U-39) was diagnosed by amniocentesis with a mosaic marker forming a ring of the proximal short arm (r(16)(::p12.2→p11.2::)), that resembles the most in terms of the chromosomal region. The data of the clinical symptoms are not available. Albeit, the distal breakpoint of that sSMC was involved in 16p12.2 and the shape of the marker was a ring chromosome. It is also important to emphasize that supposedly the sSMC cases without clinical consequences are less likely reported in the literature.
In our case we identified a marker chromosome of 4.29 Mb derived from proximal 16p11.2-11.1. The exact position of the marker’s endpoint is uncertain as we did not have appropriate FISH probes to the heterochromatin of the q arm. Furthermore, array CGH cannot detect heterochromatic targets. The 16p11.1 and the proximal 16p11.2 bands are gene-poor regions of the chromosome. According to the Decipher database GRCh38/hg19 the chr16p:31,699,804-35,989,873 region contains only 20 genes, presumably pseudogenes or transcript variants, and only two of them are OMIM genes, namely ZNF267 and TP53TG3 (Figure 2.). The Zinc finger protein 267 (ZNF267) gene that localizes in the middle of 16p11.2 (GRCh38/chr16:31,873,806-31,917,356) is a transcriptional factor. The gene modulates the gene expression and functions as a negative transcriptional regulator of matrix metalloproteinase-10 (MMP-10). The ZNF267 gene by inhibiting MMP-10 might promote liver fibrosis through diminished matrix degradation [32]. It is demonstrated that ZNF267 mRNA is up-regulated in liver cirrhosis and may be a risk factor for hepatocellular carcinoma [33]. Tp53 target gene 3 (TP53TG3) was mapped to the proximal short arm of Chr16 and is located at GRCh38/chr16:32,673,518-32,676,128 [34]. It is one of the numerous TP53 genes, those transcription factors that are involved in cell cycle arrest, apoptosis, DNA repair, chromosomal stability, and inhibition of angiogenesis. The TP53TG3 gene has no proven phenotypic gain of function effects described so far in the databases. The function of the other 18 genes in that region remains unknown.
Exact identification of an sSMC is especially important in a prenatal situation and the time necessary for the diagnosis is of great importance. Utilization of a clear algorithm and diagnostic protocol is a valuable tool in the management of a prenatally detected sSMC and can prevent the unnecessary delays during the diagnostic process [35]. Regarding our case, the characterization of the marker chromosome before the molecular genetic era would not have been carried out properly and in many instances that pregnancies would have been terminated. FISH with whole chromosome paints was very useful in the analytical processes to give information about the origin of sSCMs and might predict the euchromatic content of the markers. [36]. However, precise genotype-phenotype correlation can only be determined via chromosomal microarray technology. By applying SNP microarray analysis, we could exclude UPD, determine and specify the gene content and the region to be a harmless EV block. Our statement was also strengthened by the notion that the detailed fetal and fetal cardiac ultrasound examination did not confirm any malformation [16]. The child now is two months old and has not shown any sign of somatomental retardation or dysmorphic feature. However, we emphasize that those clinical symptoms such as developmental delay or ASD can manifest later in life. It is important to note that another prenatal case of a 16p copy number variation was diagnosed by our team recently. Genetic analysis of a fetus showing mild bilateral ventriculomegaly and partial dysgenesis of the corpus callosum (by ultrasound and MRI) at pregnancy week 20 revealed normal male karyotype and a microdeletion of 1.385 Mb of the short arm of Chr16 by aCGH (Chr16:32,542,904-33,928,095). The genomic position of that microdeletion is inside the region of the child with the sSMC in the present case. The association between the CNV and the corpus callosum dysgenesis is uncertain, the pregnancy is ongoing now (unpublished own data, Figure 2.).
In summary, we present prenatal diagnosis and molecular cytogenetic characterization of an sSMC. Exact identification of the marker chromosome as an EV enabled us to provide proper genetic counseling, to allow informed decision making and to avoid the unnecessary pregnancy termination.
Figure 2. Schematic representation of the pericentromeric region of Chr16 showing the centromere and the neighboring genomic positions. Figure indicates the known region of 16p11.2-p12.2 microduplication syndrome, the position of the euchromatic variants (EV), the present prenatal case of an sSMC (chr16p:31,699,804-35,989,873) with the 2 OMIM genes (ZNF267 and TP53TG3) and another prenatal case of a 16p copy number variation with fetal findings diagnosed by our team (Chr16:32,542,904-33,928,095, unpublished own data).