Simone Heidemann

Pyrosequencing-based DNA methylation profiling of Fanconi anemia/BRCA pathway genes in laryngeal squamous cell carcinoma

Fanconi anemia (FA) associated genes [FANCA, -B, -C, FANCD1(BRCA2), -D2, -E, -F, -G, -I, -L, -M, FANCN (PALB2), FANCJ(BRIP1) and FA-linked BRCA1] encode proteins of DNA damage response pathways mutated in FA patients. FA is characterized by congenital malformations, chromosomal instability and high cancer susceptibility. FA patients have a 500-700 times higher risk of head and neck squamous cell carcinoma (HNSCC) compared to the non-FA population. As DNA methylation comprises one of the known gene inactivation mechanisms in cancer we have investigated the methylation status of 13 FA and one FA-linked gene in order to assess the role of FA in sporadic laryngeal squamous cell carcinoma (LSCC) tumor samples. Thirteen laryngeal squamous carcinoma cell lines (UT-SCC) and 64 primary laryngeal carcinoma cases were analyzed by bisulfite pyrosequencing. DNA from buccal swabs of 10 healthy volunteers was used as a control group. Promoter regions of FANCA, BRCA1 and BRCA2 displayed recurrent alterations in the methylation levels in cancer samples as compared to buccal swabs controls. For FANCA, hypomethylation was observed in 11/13 cell lines (p<0.0003) and all 64 primary larynx samples (p<0.001) compared to buccal swabs. For BRCA1, 4/13 cell lines (p=0.04) and 3/58 primary laryngeal cases (p=0.22) showed hypomethylation. In BRCA2, all 13 cell lines (p<0.0001) 4/63 primary LSCC (p<0.01) showed hypermethylation as compared to controls. In conclusion, we show recurrent alterations of DNA methylation levels in three Fanconi anemia genes which might contribute to the pathogenesis of LSCC.

DNA-Methylation Profiling of Fetal Tissues Reveals Marked Epigenetic Differences between Chorionic and Amniotic Samples

Epigenetic mechanisms including DNA methylation are supposed to play a key role in fetal development. Here we have investigated fetal DNA-methylation levels of 27,578 CpG loci in 47 chorionic villi (CVS) and 16 amniotic cell (AC) samples. Methylation levels differed significantly between karyotypically normal AC and CVS for 2,014 genes. AC showed more extreme DNA-methylation levels of these genes than CVS and the differentially methylated genes are significantly enriched for processes characteristic for the different cell types sampled. Furthermore, we identified 404 genes differentially methylated in CVS with trisomy 21. These genes were significantly enriched for high CG dinucleotid (CpG) content and developmental processes associated with Down syndrome. Our study points to major tissue-specific differences of fetal DNA-methylation and gives rise to the hypothesis that part of the Down syndrome phenotype is epigenetically programmed in the first trimester of pregnancy.

A duplication in 1q21.3 in a family with early onset and childhood absence epilepsy

Early onset absence epilepsy (EOAE) starting before the age of 4 years constitutes a rare subgroup of the idiopathic generalized epilepsies (IGEs). A strong genetic component in IGE has been suggested by twin and family studies. We describe a boy with absence seizures starting at the age of 9 months whose parents both had childhood absence epilepsy. A 192‐kb duplication in 1q21.3 was identified in the proband and his father, encompassing the gene CHRNB2 coding for the β‐2 subunit of the nicotinic acetylcholine receptor and the gene ADAR coding for adenosine deaminase, an enzyme responsible for RNA editing. Both are candidate genes for seizure disorders. The duplication was not identified in 191 independent IGE patients (93 EOAE; 98 classical IGE) or in 1,157 population controls.

The Variant inv(2)(p11.2q13) Is a Genuinely Recurrent Rearrangement but Displays Some Breakpoint Heterogeneity

Human chromosome 2 contains large blocks of segmental duplications (SDs), both within and between proximal 2p and proximal 2q, and these may contribute to the frequency of the common variant inversion inv(2)(p11.2q13). Despite their being cytogenetically homogeneous, we have identified four different breakpoint combinations by fluorescence in situ hybridization mapping of 40 cases of inv(2)(p11.2q13) of European origin. For the vast majority of inversions (35/40), the breakpoints fell within the same spanning BACs, which hybridized to both 2p11.2 and 2q13 on the normal and inverted homologues. Sequence analysis revealed that these BACs contain a significant proportion of intrachromosomal SDs with sequence homology to the reciprocal breakpoint region. In contrast, BACs spanning the rare breakpoint combinations contain fewer SDs and with sequence homology only to the same chromosome arm. Using haplotype analysis, we identified a number of related family subgroups with identical or very closely related haplotypes. However, the majority of cases were not related, demonstrating for the first time that the inv(2)(p11.2q13) is a truly recurrent rearrangement. Therefore, there are three explanations to account for the frequent observation of the inv(2)(p11.2q13): the majority have arisen independently in different ancestors, while a minority either have been transmitted from a common founder or have different breakpoints at the molecular cytogenetic level.

Isolated trisomy 7q21.2-31.31 resulting from a complex familial rearrangement involving chromosomes 7, 9 and 10

BackgroundGenotype-phenotype correlations for chromosomal imbalances are often limited by overlapping effects of partial trisomy and monosomy resulting from unbalanced translocations and by poor resolution of banding analysis for breakpoint designation. Here we report the clinical features of isolated partial trisomy 7q21.2 to 7q31.31 without overlapping phenotypic effects of partial monosomy in an 8 years old girl. The breakpoints of the unbalanced rearranged chromosome 7 could be defined precisely by array-CGH and a further imbalance could be excluded. The breakpoints of the balanced rearranged chromosomes 9 and 10 were identified by microdissection of fluorescence labelled derivative chromosomes 9 and 10.ResultsThe probands mother showed a complex balanced translocation t(9;10)(p13;q23) with insertion of 7q21.2-31.31 at the translocation breakpoint at 9p13. The daughter inherited the rearranged chromosomes 9 and 10 but the normal chromosome 7 from her mother, resulting in partial trisomy 7q21.2 to 7q31.31. The phenotype of the patient consisted of marked developmental retardation, facial dysmorphism, short stature, strabism, and hyperextensible metacarpophalangeal joints.DiscussionFor better understanding of genotype-phenotype correlation a new classification of 7q duplications which will be based on findings of molecular karyotyping is needed. Therefore, the description of well-defined patients is valuable. This case shows that FISH-microdissection is of great benefit for precise breakpoint designation in balanced rearrangements.

Chromosomal mosaicisms in prenatal diagnosis: correlation with first trimester screening and clinical outcome.

Abstract Objective: To investigate the outcome of pregnancy after detection of chromosomal mosaicism and to determine the correlation between human chorionic gonadotropin (free β-HCG) and pregnancy-associated plasma protein-A (PAPP-A) levels from first-trimester-screening with pregnancy outcome. Methods: In a single-center, retrospective survey of the results of prenatal diagnostics performed between January 2000 and March 2011, we identified a total of 40 pregnancies with chromosomal mosaicism. Clinical characteristics and results of first-trimester screening, as well as the outcome of these cases, are described. Results: Out of 40 cases, 21 were defined as confined placental mosaicism, 10 classified as true mosaicism and nine were not classifiable cases. Nuchal translucency (NT) was ≥2.5 mm in 8/30 cases with respective measurements. PAPP-A levels were ≤0.4 MoM in 9/26 cases, with respective measurements, two of them being newborns with growth restriction. Remarkably, in pregnancies of all four children born with severe growth retardation, <3rd percentile PAPP-A levels were below 0.52 MoM. Conclusions: Our observations show mosaic pregnancy outcomes to be very heterogeneous. Nevertheless, a combination of low PAPP-A and interpretation of chromosomal mosaicism might identify pregnancies at particular risk for fetal growth restriction.

Maternal uniparental disomy 15 in a fetus resulting from a balanced familial translocation t(2;15)(p11;q11.2).

Prader-Willi syndrome (PWS) is caused by loss of function of the paternal region q12 of chromosome 15 containing among others the SNRPN gene. In approximately 70% of PWS patients SNRPN is lost due to deletions. In approximately 28% of cases, the function of the paternal SNRPN allele is lost due to uniparental, maternal disomy of the whole or of parts of chromosome 15 (UPD15mat). One to three percent of cases show imprinting center defects with or without IC mutations, whereas disruption of the SNRPN gene has been observed only in few PWS patients (for review see Horsthemke and Buiting, 2006). The phenotype of PWS is characterized by obesity, muscular hypotonia, mental retardation, short stature, small hands and feet, facial dysmorphism, behavioral problems and hypogonadotropic hypogonadism. During embryonal development, some of these typical features such as facial dysmorphism, hypoplasia of the corpus callosum and hypoplasic genitalia are frequently present, other anomalies such as short femoral length, polyhydramnios and abnormal fetal heart rhythm are rarely seen on ultrasound (L’Hermine et al., 2003). We report here on a fetus of 19th week of gestation with maternal uniparental disomy (UPD) 15 without any obvious internal or external malformations, peculiarities in pregnancy or family history. Solely a balanced translocation t(2;15)(p11;q11.2) indicated a possible risk of a syndrome related to UPD 15. The 36-year-old III gravida (I para) was referred to prenatal testing at 15th week of gestation due to advanced maternal age. The woman had one healthy child. Except for one spontaneous abortion the family history was unremarkable. The pregnancy had been uneventful. Alpha feto protein concentration in the amniotic fluid was within the normal range (0.9 MoM). Chromosome analyses were performed using standard GTGand fluorescent R-banding techniques. Chromosome analysis on cultivated amniocytic fluid

Conflicting results of prenatal FISH with different probes for Down's Syndrome critical regions associated with mosaicism for a de novo del(21)(q22) characterised by molecular karyotyping: Case report

For the rapid detection of common aneuploidies either PCR or Fluorescence in situ hybridisation (FISH) on uncultured amniotic fluid cells are widely used. There are different commercial suppliers providing FISH assays for the detection of trisomies affecting the Downs syndrome critical regions (DSCR) in 21q22. We present a case in which rapid FISH screening with different commercial probes for the DSCR yielded conflicting results. Chromosome analysis revealed a deletion of one chromosome 21 in q22 which explained the findings. Prenatally an additional small supernumerary marker chromosome (sSMC) was discovered as well, which could not be characterised. Postnatal chromosome analysis in lymphocytes of the infant revealed complex mosaicism with four cell lines. By arrayCGH the sSMC was provisionally described as derivative chromosome 21 which was confirmed by targeted FISH experiments.

Reply to letter by Dr. G. Corso

We would like to thank Giovanni Corso et al. for their “Letter to the editor”. In our study, we screened 97 individuals for CDH1 mutations. Screening revealed two missense variants in independent families. The first alteration, A592T, was classified as neutral. The other variant, S838G, was detected in an unaffected woman with a family history of BRCA1positive breastand ovarian cancer without gastric cancer. Our index case was tested negative for the familial BRCA1 mutation. In the past, the variant S838G has been classified as potentially deleterious [1]. Searching the AlamutDatabase (Alamut Visual 2.10) revealed contradictory rating of the variant from likely benign to pathogenic, whereas the pathogenic classification probably also was judged by the paper of Risinger et al. [1]. In addition, in dbSNP, one will find this spectrum of rating (rs121964872). A request to Ambry Genetics for the reason why they judge the variant as likely benign revealed following answer: The serine at codon 838 is replaced by glycine, an amino acid with similar properties. This amino acid position is well conserved in available vertebrate species. This variant has been seen in several families whose clinical histories are not consistent with hereditary diffuse gastric cancer or other CDH1-related cancers (Ambry Internal data). In addition, this alteration is predicted to be benign and tolerated by PolyPhen and SIFT in silico analyses, respectively. Based on the majority of available evidence to date, this variant is unlikely to be pathogenic. Therefore, the cases of breast and ovarian cancer in the reported family are most likely caused by the BRCA1 mutation. Unfortunately, we had no possibilities to test the segregation of this missense variant S838G in other family members. This is why, we cannot exclude that this CDH1 variant could have a modifying influence on breast cancer causing the lobular histology. The clinical management option mentioned in the letter of Corso et al. should be regarded with caution if no clear deleterious alteration is found in the gene, because the clinical consequences for the patient not only psychologically but also physically are very harmful. Due to the present evidence regarding the missense variant S838G, we presently do not see any need for clinical intervention. However, we agree that, based on the literature, there is a connection between CDH1 alteration and the lobular type of breast cancer, and this may fulfill the criteria of an independent syndrome.