Marieke Joosten

Application of SNP array for rapid prenatal diagnosis: Implementation, genetic counselling and diagnostic flow

We report on the validation and implementation of the HumanCytoSNP-12 array (Illumina) (HCS) in prenatal diagnosis. In total, 64 samples were used to validate the Illumina platform (20 with a known (sub) microscopic chromosome abnormality, 5 with known maternal cell contamination (MCC) and 39 normal control samples). There were no false-positive or false-negative results. In addition to the diagnostic possibilities of arrayCGH, the HCS allows detection of regions of homozygosity (ROH), triploidy and helps recognising MCC. Moreover, in two cases of MCC, a deletion was correctly detected. Furthermore we found out that only about 50 ng of DNA is required, which allows a reporting time of only 3 days. We also present a prospective pilot study of 61 fetuses with ultrasound abnormalities and a normal karyotype tested with HCS. In 4 out of 61 (6.5%) fetuses, a clinically relevant abnormality was detected. We designed and present pre-test genetic counselling information on categories of possible test outcomes. On the basis of this information, about 90% of the parents chose to be informed about adverse health outcomes of their future child at infancy and childhood, and 55% also about outcomes at an adult stage. The latter issue regarding the right of the future child itself to decide whether or not to know this information needs to be addressed.

Genomic SNP array as a gold standard for prenatal diagnosis of foetal ultrasound abnormalities

BackgroundWe have investigated whether replacing conventional karyotyping by SNP array analysis in cases of foetal ultrasound abnormalities would increase the diagnostic yield and speed of prenatal diagnosis in clinical practice.Findings/resultsFrom May 2009 till June 2011 we performed HumanCytoSNP-12 array (HCS) (http://www.Illumina.com) analysis in 207 cases of foetal structural abnormalities. HCS allows detecting unbalanced genomic abnormalities with a resolution of about 150/200 kb. All cases were selected by a clinical geneticist after excluding the most common aneuploidies by RAD (rapid aneuploidy detection). Pre-test genetic counselling was offered in all cases.In 24/207 (11,6%) foetuses a clinically relevant genetic abnormality was detected. Only 8/24 abnormalities would have been detected if only routine karyotyping was performed. Submicroscopic abnormalities were found in 16/207 (7,7%) cases. The array results were achieved within 1-2 weeks after amniocentesis.ConclusionsPrenatal SNP array testing is faster than karyotyping and allows detecting much smaller aberrations (~0.15 Mb) in addition to the microscopic unbalanced chromosome abnormalities detectable with karyotyping (~ > 5 Mb). Since karyotyping would have missed 66% (16/24) of genomic abnormalities in our cohort, we propose to perform genomic high resolution array testing assisted by pre-test counselling as a primary prenatal diagnostic test in cases of foetal ultrasound abnormalities.

Erythroid defects and increased retrovirally-induced tumor formation in Evi1 transgenic mice

Aberrant expression of the Evi1 (ecotropic virus integration site 1) proto-oncogene has been associated with hematopoietic malignancies in both mice and man. To determine the effect of enforced expression of Evi1 in vivo, we developed a transgenic mouse model utilizing the murine Sca-1 (Ly-6E.1) promoter. Here, we describe the generation and analysis of three independent lines of Evi1 transgenic mice. Transgenic animals of two founder lines developed normally. These mice did not show any obvious hematological abnormalities but showed a significant reduction in the number of bone marrow colony-forming unit erythroid (CFU-E)-derived colonies. This implies a defect of normal erythroid hematopoiesis affecting relatively late erythroid progenitor cells. We also show that when newborn Evi1 transgenic mice of these two lines were infected with Cas-Br-M MuLV, tumor incidence was greatly enhanced in comparison with nontransgenic littermates, indicating an increased susceptibility for leukemia development. Interestingly, analysis of a third founder line revealed that all male progeny consistently displayed severely impaired erythropoiesis with major defects in the bone marrow, spleen and peripheral blood. Taken together, our results present the first evidence of Evi1 disturbing normal erythropoiesis in vivo and provides evidence for cooperative potential of Evi1 in tumor progression.

Large-scale identification of novel potential disease loci in mouse leukemia applying an improved strategy for cloning common virus integration sites.

The identification of common virus integration sites (cVIS) in retrovirally induced tumors in mice provides a powerful strategy to isolate novel transforming genes. Applying virus LTR-specific inverse-PCR and RT–PCR combined with automated sequencing on CasBr-M Murine Leukemia Virus (MuLV) induced myeloid leukemias, 126 virus integration sites were cloned. Using locus- and LTR-specific primers, a nested-PCR/Southern-blotting procedure was developed on genomic DNA from a large panel of MuLV-induced leukemias, to analyse whether a particular virus insertion represented a cVIS. In fact 39 out of 41 integrations analysed this way appeared to represent a common virus integration. We recognized six previously cloned cVISs, i.e. Evi1, Hoxa7, c-Myb, Cb2/Evi11, Evi12, and His1 and 33 novel common insertions, designated Cas-Br Virus Integration Site (Casvis). Among this group we found integrations in or near genes encoding nuclear proteins, e.g. Dnmt-2, Nm23-M2, Ctbp1 or Erg, within receptor genes, e.g. Cb2 or mrc1, novel putative signaling or transporter genes, the ringfinger-protein gene Mid1 and a panel of genes encoding novel proteins with unknown function. The finding that 39 out of 41 integrations analysed represented a cVIS, suggests that the majority of the other virus insertions that were not yet analysed by the PCR/Southern-blotting method are located in a cVIS as well and may therefore also harbor novel disease genes.

Types of array findings detectable in cytogenetic diagnosis: a proposal for a generic classification

Types of array findings detectable in cytogenetic diagnosis: a proposal for a generic classification

Benefits and burdens of using a SNP array in pregnancies at increased risk for the common aneuploidies.

We present the nature of pathogenic SNP array findings in pregnancies without ultrasound (US) abnormalities and show the additional diagnostic value of SNP array as compared with rapid aneuploidy detection and karyotyping. 1,330 prenatal samples were investigated with a 0.5‐Mb SNP array after the exclusion of the most common aneuploidies. In 2.7% (36/1,330) of the cases, pathogenic chromosome aberrations were found; a microscopically detectable abnormality in 0.7% and a submicroscopic aberration in 2%. Our results show that in addition to the age‐ or screening‐related aneuploidy risk, in pregnancies without US abnormalities, there is a risk of 1:148 (9/1,330) for a (sub)microscopic abnormality associated with an early‐onset often severe disease, 1:222 (6/1,330) for a submicroscopic aberration causing an early‐onset disease, 1:74 (18/1,330) for carrying a susceptibility locus for a neurodevelopmental disorder, and 1:443 (3/1,330) for a late‐onset disorder (hereditary neuropathy with liability to pressure palsies in all three cases). These risk figures are important for adequate pretest counseling so that prospective parents can make informed individualized choices between targeted prenatal testing and broad testing with SNP array. Based on our results, we believe if invasive testing is performed, SNP array should be the preferred cytogenetic technique irrespective of the indication.

Prenatal SNP array testing in 1000 fetuses with ultrasound anomalies: causative, unexpected and susceptibility CNVs

To evaluate the diagnostic value of single-nucleotide polymorphism (SNP) array testing in 1033 fetuses with ultrasound anomalies we investigated the prevalence and genetic nature of pathogenic findings. We reclassified all pathogenic findings into three categories: causative findings; unexpected diagnoses (UD); and susceptibility loci (SL) for neurodevelopmental disorders. After exclusion of trisomy 13, 18, 21, sex-chromosomal aneuploidy and triploidies, in 76/1033 (7.4%) fetuses a pathogenic chromosome abnormality was detected by genomic SNP array: in 19/1033 cases (1.8%) a microscopically detectable abnormality was found and in 57/1033 (5.5%) fetuses a pathogenic submicroscopic chromosome abnormality was detected. 58% (n=44) of all these pathogenic chromosome abnormalities involved a causative finding, 35% (n=27) a SL for neurodevelopmental disorder, and 6% (n=5) a UD of an early-onset untreatable disease. In 0.3% of parental samples an incidental pathogenic finding was encountered. Our results confirm that a genomic array should be the preferred first-tier technique in fetuses with ultrasound anomalies. All UDs involved early-onset diseases, which is beneficial for the patients to know. It also seems that UDs occur at a comparable frequency among microscopic and submicroscopic pathogenic findings. SL were more often detected than in pregnancies without ultrasound anomalies.

Leukemic predisposition of pSca-1/Cb2 transgenic mice

OBJECTIVE The gene encoding the peripheral cannabinoid receptor Cb2 is located in the common virus integration site Evi11 and is associated with hematopoietic malignancies in mice. To determine the effect of Cb2 overexpression on hematopoietic development in vivo, Cb2 transgenic mice were generated. MATERIALS AND METHODS A Cb2 expression vector was constructed containing a Cb2 cDNA fragment cloned into the 14kb Sca-1 (Ly-6E.1) gene. Two transgenic lines in which Cb2 expression is controlled by the Sca-1 promoter were generated, and the effect on hematopoietic development was studied. Expression of Cb2 mRNA or protein was studied by RNase protection analysis and ligand binding assays, respectively. Leukemic predisposition was investigated by injecting newborn transgenic as well as control animals with Cas-Br-M murine leukemia virus (Cas-Br-M MuLV). RESULTS Although increased expression of the Cb2 gene was observed in hematopoietic tissues, follow-up of more than 1 year did not reveal any hematologic defect. Interestingly, infection of newborn pSca-1/Cb2 transgenic mice with Cas-Br-M MuLV revealed that significantly more transgenic mice developed leukemia than virus-treated control littermates. Because these studies provide evidence for the cooperative potential of Cb2 in leukemia progression, we wished to identify genes that may collaborate with Cb2 in leukemic transformation. Our study suggests that Evi1, another common target for proviral integration in mouse leukemias, may be overexpressed in virus-induced leukemias in pSca-1/Cb2 transgenic mice. CONCLUSIONS The data indicate that hematopoietic precursor cells that express high levels of Cb2 possess increased susceptibility for leukemia development and that Cb2 and Evi1 might collaborate in leukemogenesis.

False Negative NIPT Results: Risk Figures for Chromosomes 13, 18 and 21 Based on Chorionic Villi Results in 5967 Cases and Literature Review

Non-invasive prenatal testing (NIPT) demonstrated a small chance for a false negative result. Since the “fetal” DNA in maternal blood originates from the cytotrophoblast of chorionic villi (CV), some false negative results will have a biological origin. Based on our experience with cytogenetic studies of CV, we tried to estimate this risk. 5967 CV samples of pregnancies at high risk for common aneuplodies were cytogenetically investigated in our centre between January 2000 and December 2011. All cases of fetal trisomy 13, 18 and 21 were retrospectively studied for the presence of a normal karyotype or mosaicism < 30% in short-term cultured (STC-) villi. 404 cases of trisomies 13, 18 and 21 were found amongst 5967 samples (6,8%). Of these 404 cases, 14 (3,7%) had a normal or low mosaic karyotype in STC-villi and therefore would potentially be missed with NIPT. It involved 2% (5/242) of all trisomy 21 cases and 7.3% (9/123) of all trisomy 18 cases. In 1:426 (14/5967) NIPT samples of patients at high risk for common aneuploidies, a trisomy 18 or 21 will potentially be missed due to the biological phenomenon of absence of the chromosome aberration in the cytotrophoblast.

Whole-genome array as a first-line cytogenetic test in prenatal diagnosis

The main goal of cytogenetic prenatal diagnosis is to inform prospective parents about the chromosomal status of their fetus. A chromosomal aberration usually causes an abnormal phenotype in childhood. Prenatal whole-genome cytogenetic diagnosis was for a long time dependent on karyotyping, which requires time-consuming cell culturing, has a limited resolution (5–10 Mb) and is dependent on optimal harvesting and chromosome staining conditions. Nowadays, genomic microarray technology allows whole-genome testing at a higher resolution and it can be applied to uncultured fetal material, allowing shorter reporting times when compared with classical cytogenetic techniques. Genomic microarray testing has been recommended for routine postnatal cytogenetics in cases of intellectual disability and/or multiple congenital anomalies1, and for prenatal diagnosis in cases of fetal ultrasound anomalies2. However, its implementation for all indications in prenatal genetic diagnosis is still under discussion3–6. The main arguments against offering prenatal array testing for all indications are the possibility of detecting: (1) CNVs causing well-described clinically significant anomalies not related to the initial indication (unexpected diagnoses); (2) CNVs associated with a variable expressivity and heterogeneity of clinical features, with an as yet unquantifiable chance of an abnormal phenotype if found prenatally (so called susceptibility loci (SL) for neurodevelopmental disorders); and (3) variants of unknown clinical significance (VOUS). Such findings may complicate genetic counseling7–12 and these issues raise further questions, such as which outcomes of genomic microarrays should be reported to pregnant couples? Should they be offered a choice regarding about which possible array outcomes they wish to be informed? Is extensive genetic pretest counseling in every case necessary and feasible in clinical practice? Since this new technology is already present in prenatal clinics, rather than debating whether array testing should be performed for all referrals of invasive cytogenetic prenatal diagnosis, we should instead be discussing how to meet this challenge. In this Editorial, we review the current international status regarding the use of array technology for prenatal diagnosis. We discuss platforms and testing resolution, indications, counseling, (possible problematic) findings and the decision regarding what should be reported to the future parents.