Identification of novel deletion breakpoints bordered by segmental duplications in the NF1 locus using high resolution array-CGH

Genotype-Phenotype Correlations in Neurofibromatosis Type 1: Identification of Novel and Recurrent NF1 Gene Variants and Correlations with Neurocognitive Phenotype

Neurofibromatosis type 1 (NF1) is one of the most common genetic tumor predisposition syndrome, caused by mutations in the NF1. To date, few genotype-phenotype correlations have been discerned in NF1, due to a highly variable clinical presentation. We aimed to study the molecular spectrum of NF1 and genotype-phenotype correlations in a monocentric study cohort of 85 NF1 patients (20 relatives, 65 sporadic cases). Clinical data were collected at the time of the mutation analysis and reviewed for accuracy in this investigation. An internal phenotypic categorization was applied. The 94% of the patients enrolled showed a severe phenotype with at least one systemic complication and a wide range of associated malignancies. Spine deformities were the most common complications in this cohort. We also reported 66 different NF1 mutations, of which 7 are novel mutations. Correlation analysis identified a slight significant inverse correlation between age at diagnosis and delayed acquisition of psychomotor skills with residual multi-domain cognitive impairment. Odds ratio with 95% confidence interval showed a higher prevalence of learning disabilities in patients carrying frameshift mutations. Overall, our results aim to offer an interesting contribution to studies on the genotype–phenotype of NF1 and in genetic management and counselling.

Atypical NF1 Microdeletions: Challenges and Opportunities for Genotype/Phenotype Correlations in Patients with Large NF1 Deletions

Patients with neurofibromatosis type 1 (NF1) and type 1 NF1 deletions often exhibit more severe clinical manifestations than patients with intragenic NF1 gene mutations, including facial dysmorphic features, overgrowth, severe global developmental delay, severe autistic symptoms and considerably reduced cognitive abilities, all of which are detectable from a very young age. Type 1 NF1 deletions encompass 1.4 Mb and are associated with the loss of 14 protein-coding genes, including NF1 and SUZ12. Atypical NF1 deletions, which do not encompass all 14 protein-coding genes located within the type 1 NF1 deletion region, have the potential to contribute to the delineation of the genotype/phenotype relationship in patients with NF1 microdeletions. Here, we review all atypical NF1 deletions reported to date as well as the clinical phenotype observed in the patients concerned. We compare these findings with those of a newly identified atypical NF1 deletion of 698 kb which, in addition to the NF1 gene, includes five genes located centromeric to NF1. The atypical NF1 deletion in this patient does not include the SUZ12 gene but does encompass CRLF3. Comparative analysis of such atypical NF1 deletions suggests that SUZ12 hemizygosity is likely to contribute significantly to the reduced cognitive abilities, severe global developmental delay and facial dysmorphisms observed in patients with type 1 NF1 deletions.

Genotype-Phenotype Associations in Patients With Type-1, Type-2, and Atypical NF1 Microdeletions

Neurofibromatosis type 1 is a tumor predisposition syndrome inherited in autosomal dominant manner. Besides the intragenic loss-of-function mutations in NF1 gene, large deletions encompassing the NF1 gene and its flanking regions are responsible for the development of the variable clinical phenotype. These large deletions titled as NF1 microdeletions lead to a more severe clinical phenotype than those observed in patients with intragenic NF1 mutations. Around 5-10% of the cases harbor large deletion and four major types of NF1 microdeletions (type 1, 2, 3 and atypical) have been identified so far. They are distinguishable in term of their size and the location of the breakpoints, by the frequency of somatic mosaicism with normal cells not harboring the deletion and by the number of the affected genes within the deleted region. In our study genotype-phenotype analyses have been performed in 17 mostly pediatric patients with NF1 microdeletion syndrome identified by multiplex ligation-dependent probe amplification after systematic sequencing of the NF1 gene. Confirmation and classification of the NF1 large deletions were performed using array comparative genomic hybridization, where it was feasible. In our patient cohort 70% of the patients possess type-1 deletion, one patient harbors type-2 deletion and 23% of our cases have atypical NF1 deletion. All the atypical deletions identified in this study proved to be novel. One patient with atypical deletion displayed mosaicism. In our study NF1 microdeletion patients presented dysmorphic facial features, macrocephaly, large hands and feet, delayed cognitive development and/or learning difficulties, speech difficulties, overgrowth more often than patients with intragenic NF1 mutations. Moreover, neurobehavior problems, macrocephaly and overgrowth were less frequent in atypical cases compared to type-1 deletion. Proper diagnosis is challenging in certain patients since several clinical manifestations show age-dependency. Large tumor load exhibited more frequently in this type of disorder, therefore better understanding of genotype-phenotype correlations and progress of the disease is essential for individuals suffering from neurofibromatosis to improve the quality of their life. Our study presented additional clinical data related to NF1 microdeletion patients especially for pediatric cases and it contributes to the better understanding of this type of disorder.

Frequent breakpoints of focal deletion and uniparental disomy in 22q11.1 or 11.2 segmental duplication region reveal distinct tumorigenesis in rhabdoid tumor of the kidney

SMARCB1 is mutated in most rhabdoid tumors (RTs) developing in the kidney (RTK) and various other organs. Focal deletions found in patients with 22q11.2 deletion syndrome show breakpoints within clusters of segmental duplications (SDs), and those in some RTs show breakpoints in the 22q11-q12 region. SDs are known to cause focal deletion mediated by non-allelic homologous recombination. The present study identified SMARCB1 alterations in all 30 RTKs, using SNP array CGH, MLPA, and sequence analyses. Twenty-eight tumors had a total of 51 breakpoints forming focal 22q deletion and/or uniparental disomy (22qUPD), and the other two had compound mutation with no breakpoints in 22q. Twenty-four (47.1%) of the 51 breakpoints were within SDs, and occurred in 16 (53.3%) of the 30 tumors. The association of breakpoints with SDs was found not only in focal deletion, but also in 22qUPD, indicating that SDs mediate the first and second hits (focal deletion) and the second hit (22qUPD) of SMARCB1 alteration. Of the 51 breakpoints, 14 were recurrent, and 10 of the 14 were within SDs, suggesting the presence of hotspots in the 22q11.2 region. One recurrent breakpoint outside SDs resided in SMARCB1, suggesting inactivation of the gene by out-of-frame fusion. The association between SDs and focal deletion has been reported in two other types of cancer. RTKs may be the third example of SD-associated tumors. Thus, the present study indicated that RTKs exploit genomic instability in the 22q11.1-11.2 SDs region, and 22qUPD caused by mitotic recombination may also be mediated by SDs.

Emerging genotype-phenotype relationships in patients with large NF1 deletions

The most frequent recurring mutations in neurofibromatosis type 1 (NF1) are large deletions encompassing the NF1 gene and its flanking regions (NF1 microdeletions). The majority of these deletions encompass 1.4-Mb and are associated with the loss of 14 protein-coding genes and four microRNA genes. Patients with germline type-1 NF1 microdeletions frequently exhibit dysmorphic facial features, overgrowth/tall-for-age stature, significant delay in cognitive development, large hands and feet, hyperflexibility of joints and muscular hypotonia. Such patients also display significantly more cardiovascular anomalies as compared with patients without large deletions and often exhibit increased numbers of subcutaneous, plexiform and spinal neurofibromas as compared with the general NF1 population. Further, an extremely high burden of internal neurofibromas, characterised by >3000 ml tumour volume, is encountered significantly, more frequently, in non-mosaic NF1 microdeletion patients than in NF1 patients lacking such deletions. NF1 microdeletion patients also have an increased risk of malignant peripheral nerve sheath tumours (MPNSTs); their lifetime MPNST risk is 16–26%, rather higher than that of NF1 patients with intragenic NF1 mutations (8–13%). NF1 microdeletion patients, therefore, represent a high-risk group for the development of MPNSTs, tumours which are very aggressive and difficult to treat. Co-deletion of the SUZ12 gene in addition to NF1 further increases the MPNST risk in NF1 microdeletion patients. Here, we summarise current knowledge about genotype–phenotype relationships in NF1 microdeletion patients and discuss the potential role of the genes located within the NF1 microdeletion interval whose haploinsufficiency may contribute to the more severe clinical phenotype.

Palindrome-mediated and replication-dependent pathogenic structural rearrangements within the NF1 gene

Palindromic sequences can form hairpin structures or cruciform extrusions, which render them susceptible to genomic rearrangements. A 197-bp long palindromic AT-rich repeat (PATRR17) is located within intron 40 of the neurofibromatosis type 1 (NF1) gene (17q11.2). Through comprehensive NF1 analysis, we identified six unrelated patients with a rearrangement involving intron 40 (five deletions and one reciprocal translocation t(14;17)(q32;q11.2)). We hypothesized that PATRR17 may be involved in these rearrangements thereby causing NF1. Breakpoint cloning revealed that PATRR17 was indeed involved in all of the rearrangements. As microhomology was present at all breakpoint junctions of the deletions identified, and PATRR17 partner breakpoints were located within 7.1 kb upstream of PATRR17, fork stalling and template switching/microhomology-mediated break-induced replication was the most likely rearrangement mechanism. For the reciprocal translocation case, a 51 bp insertion at the translocation breakpoints mapped to a short sequence within PATRR17, proximal to the breakpoint, suggesting a multiple stalling and rereplication process, in contrast to previous studies indicating a purely replication-independent mechanism for PATRR-mediated translocations. In conclusion, we show evidence that PATRR17 is a hotspot for pathogenic intragenic deletions within the NF1 gene and suggest a novel replication-dependent mechanism for PATRR-mediated translocation.

Republished: Non-heritable genetics of human disease: spotlight on post-zygotic genetic variation acquired during lifetime

The heritability of most common, multifactorial diseases is rather modest and known genetic effects account for a small part of it. The remaining portion of disease aetiology has been conventionally ascribed to environmental effects, with an unknown part being stochastic. This review focuses on recent studies highlighting stochastic events of potentially great importance in human disease—the accumulation of post-zygotic structural aberrations with age in phenotypically normal humans. These findings are in agreement with a substantial mutational load predicted to occur during lifetime within the human soma. A major consequence of these results is that the genetic profile of a single tissue collected at one time point should be used with caution as a faithful portrait of other tissues from the same subject or the same tissue throughout life. Thus, the design of studies in human genetics interrogating a single sample per subject or applying lymphoblastoid cell lines may come into question. Sporadic disorders are common in medicine. We wish to stress the non-heritable genetic variation as a potentially important factor behind the development of sporadic diseases. Moreover, associations between post-zygotic mutations, clonal cell expansions and their relation to cancer predisposition are central in this context. Post-zygotic mutations are amenable to robust examination and are likely to explain a sizable part of non-heritable disease causality, which has routinely been thought of as synonymous with environmental factors. In view of the widespread accumulation of genetic aberrations with age and strong predictions of disease risk from such analyses, studies of post-zygotic mutations may be a fruitful approach for delineation of variants that are causative for common human disorders.

Non-heritable genetics of human disease: spotlight on post-zygotic genetic variation acquired during lifetime

The heritability of most common, multifactorial diseases is rather modest and known genetic effects account for a small part of it. The remaining portion of disease aetiology has been conventionally ascribed to environmental effects, with an unknown part being stochastic. This review focuses on recent studies highlighting stochastic events of potentially great importance in human disease-the accumulation of post-zygotic structural aberrations with age in phenotypically normal humans. These findings are in agreement with a substantial mutational load predicted to occur during lifetime within the human soma. A major consequence of these results is that the genetic profile of a single tissue collected at one time point should be used with caution as a faithful portrait of other tissues from the same subject or the same tissue throughout life. Thus, the design of studies in human genetics interrogating a single sample per subject or applying lymphoblastoid cell lines may come into question. Sporadic disorders are common in medicine. We wish to stress the non-heritable genetic variation as a potentially important factor behind the development of sporadic diseases. Moreover, associations between post-zygotic mutations, clonal cell expansions and their relation to cancer predisposition are central in this context. Post-zygotic mutations are amenable to robust examination and are likely to explain a sizable part of non-heritable disease causality, which has routinely been thought of as synonymous with environmental factors. In view of the widespread accumulation of genetic aberrations with age and strong predictions of disease risk from such analyses, studies of post-zygotic mutations may be a fruitful approach for delineation of variants that are causative for common human disorders.

Identification of recurrent type-2 NF1 microdeletions reveals a mitotic nonallelic homologous recombination hotspot underlying a human genomic disorder

Nonallelic homologous recombination (NAHR) is one of the major mechanisms underlying copy number variation in the human genome. Although several disease-associated meiotic NAHR breakpoints have been analyzed in great detail, hotspots for mitotic NAHR are not well characterized. Type-2 NF1 microdeletions, which are predominantly of postzygotic origin, constitute a highly informative model with which to investigate the features of mitotic NAHR. Here, a custom-designed MLPA- and PCR-based approach was used to identify 23 novel NAHR-mediated type-2 NF1 deletions. Breakpoint analysis of these 23 type-2 deletions, together with 17 NAHR-mediated type-2 deletions identified previously, revealed that the breakpoints are nonuniformly distributed within the paralogous SUZ12 and SUZ12P sequences. Further, the analysis of this large group of type-2 deletions revealed breakpoint recurrence within short segments (ranging in size from 57 to 253-bp) as well as the existence of a novel NAHR hotspot of 1.9-kb (termed PRS4). This hotspot harbored 20% (8/40) of the type-2 deletion breakpoints and contains the 253-bp recurrent breakpoint region BR6 in which four independent type-2 deletion breakpoints were identified. Our findings indicate that a combination of an open chromatin conformation and short non-B DNA-forming repeats may predispose to recurrent mitotic NAHR events between SUZ12 and its pseudogene.

Dissecting the clinical phenotype associated with mosaic type-2 NF1 microdeletions

Patients with large deletions of the NF1 gene and its flanking regions (termed NF1 microdeletions) generally exhibit more severe clinical manifestations of neurofibromatosis type-1 (NF1). Here, we have investigated the clinical phenotype displayed by eight patients harbouring mosaic type-2 NF1 microdeletions. These patients did not exhibit facial dysmorphism, attention deficit hyperactivity disorder, delayed cognitive development and/or learning disabilities, cognitive impairment, congenital heart disease, hyperflexibility of joints, large hands and feet, muscular hypotonia or bone cysts. All these features have previously been reported to be disproportionately associated with germline (i.e. non-mosaic) type-1 NF1 microdeletions as compared with the general NF1 population. Plexiform neurofibromas were also less prevalent in patients with mosaic type-2 NF1 microdeletions as compared with patients carrying constitutional (germline) type-1 NF1 microdeletions. Five of the eight patients with mosaic type-2 deletions investigated here had 20-250 cutaneous neurofibromas, but only one of them exhibited a high load of cutaneous neurofibromas (N > 1,000). By contrast, a previous study indicated a high burden of cutaneous neurofibromas (N > 1,000) in 50% of adult patients with germline type-1 NF1 deletions. Patients with germline type-1 NF1 microdeletions have been reported to have an increased lifetime risk of 16-26% for a malignant peripheral nerve sheath tumour (MPNST). In this study, one of the eight investigated mosaic type-2 microdeletion patients developed an MPNST. We conclude that patients with mosaic type-2 NF1 microdeletions may also be at an increased risk of MPNSTs despite their generally milder disease manifestations as compared with germline type-1 NF1 microdeletions.

Structural genetic variation in the context of somatic mosaicism

Somatic mosaicism is the result of postzygotic de novo mutation occurring in a portion of the cells making up an organism. Structural genetic variation is a very heterogeneous group of changes, in terms of numerous types of aberrations that are included in this category, involvement of many mechanisms behind the generation of structural variants, and because structural variation can encompass genomic regions highly variable in size. Structural variation rapidly evolved as the dominating type of changes behind human genetic diversity, and the importance of this variation in biology and medicine is continuously increasing. In this review, we combine the evidence of structural variation in the context of somatic cells. We discuss the normal and disease-related somatic structural variation. We review the recent advances in the field of monozygotic twins and other models that have been studied for somatic mutations, including other vertebrates. We also discuss chromosomal mosaicism in a few prime examples of disease genes that contributed to understanding of the importance of somatic heterogeneity. We further highlight challenges and opportunities related to this field, including methodological and practical aspects of detection of somatic mosaicism. The literature devoted to interindividual variation versus papers reporting on somatic variation suggests that the latter is understudied and underestimated. It is important to increase our awareness about somatic mosaicism, in particular, related to structural variation. We believe that further research of somatic mosaicism will prove beneficial for better understanding of common sporadic disorders.

Internal tumor burden in neurofibromatosis Type I patients with large NF1 deletions

Neurofibromatosis Type 1 (NF1) is a frequent tumor suppressor gene disorder characterized by multiple benign tumors and high risk of malignancy. Internal tumor burden is a major disease-associated manifestation and can be most adequately assessed by magnetic resonance imaging of the whole body. Approximately 5% of NF1 patients have constitutional large NF1-deletions that are generally associated with more severe clinical manifestations. Here, we investigated whether these deletion patients also have more and/or larger internal tumors by assessing internal tumors and their total volume (exclusive of cutaneous and subcutaneous) in 38 NF1 deletion patients (including eight mosaic cases) and 114 age- and gender-matched NF1 patients without deletions. The incidence of internal tumors was significantly lower in mosaic deletion patients (1/8 = 13%) but did not differ between the 30 nonmosaic deletion patients and the 90 age- and gender-matched NF1 patients without large deletions used as controls. Neither the number nor the total volume of tumors per patient differed significantly between the latter two groups. However, extremely high tumor burden (>3,000 ml) was significantly more frequent among nonmosaic NF1 deletion patients than among NF1 patients without large deletions (13% vs. 1%, P = 0.014). Thus, as a group, patients with NF1 deletions do not exhibit a significantly higher internal tumor burden than NF1 patients without such deletions. However, deletion patients can frequently have extremely large internal tumors and thus demand special attention.

Tissue-specific differences in the proportion of mosaic large NF1 deletions are suggestive of a selective growth advantage of hematopoietic del(+/-) stem cells

Type-2 NF1 deletions spanning 1.2 Mb are frequently of postzygotic origin and hence tend to be associated with mosaicism for normal cells and those harboring the deletion (del(+/-) cells). Eleven patients with mosaic type-2 deletions were investigated by FISH and high proportions (94-99%) of del(+/-) cells were detected both in whole blood and in isolated CD3+, CD14+, CD15+, and CD19+ leukocytes. Significantly lower proportions of del(+/-) cells (24-82%) were however noted in urine-derived epithelial cells. A patient harboring an atypical large NF1 deletion with nonrecurrent breakpoints was also found to have a much higher proportion of del(+/-) cells in blood (96%) than in urine (51%). The tissue-specific differences in the proportions of del(+/-) cells as well as the X chromosome inactivation (XCI) patterns observed in these mosaic patients suggest that the majority of the deletions had occurred before or during the preimplantation blastocyst stage before the onset of XCI. We postulate that hematopoietic del(+/-) stem cells present at an early developmental stage are characterized by a selective growth advantage over normal cells lacking the deletion, leading to a high proportion of del(+/-) cells in peripheral blood from the affected patients.

A highly sensitive genetic protocol to detect NF1 mutations

Neurofibromatosis type 1 (NF1) is a hereditary disorder caused by mutations in the NF1 gene. Detecting mutation in NF1 is hindered by the gene's large size, the lack of mutation hotspots, the presence of pseudogenes, and the wide variety of possible lesions. We developed a method for detecting germline mutations by combining an original RNA-based cDNA-PCR mutation detection method and denaturing high-performance liquid chromatography (DHPLC) with multiplex ligation-dependent probe amplification (MLPA). The protocol was validated in a cohort of 56 blood samples from NF1 patients who fulfilled NIH diagnostic criteria, identifying the germline mutation in 53 cases (95% sensitivity). The efficiency and reliability of this approach facilitated detection of different types of mutations, including single-base substitutions, deletions or insertions of one to several nucleotides, microdeletions, and changes in intragenic copy number. Because mutational screening for minor lesions was performed using cDNA and the characterization of mutated alleles was performed at both the RNA and genomic DNA level, the analysis provided insight into the nature of the different mutations and their effect on NF1 mRNA splicing. After validation, we implemented the protocol as a routine test. Here we present the overall unbiased spectrum of NF1 mutations identified in 93 patients in a cohort of 105. The results indicate that this protocol is a powerful new tool for the molecular diagnosis of NF1.

Identification of genes potentially involved in the increased risk of malignancy in NF1-microdeleted patients

Patients with NF1 microdeletion develop more neurofibromas at a younger age, and have an increased risk of malignant peripheral nerve sheath tumors (MPNSTs). We postulated that the increased risk of malignancy could be due to inactivation, in addition to NF1, of a second tumor suppressor gene located in the typical 1.4-Mb microdeletion found in most of the microdeleted patients. We investigated the expression of NF1, the other 16 protein-coding genes and the 2 microRNAs located in the 1.4-Mb microdeletion by means of real-time quantitative reverse-transcription polymerase chain reaction (RT-PCR) in a large series of human dermal and plexiform neurofibromas and MPNSTs. Five genes were significantly upregulated: OMG and SUZ12 in plexiform neurofibromas and ATAD5, EVI2A and C17orf79 in MPNSTs. More interestingly, two genes were significantly downregulated (RNF135 and CENTA2) in tumor Schwann cells from MPNST biopsies and in MPNST cell lines. This study points to the involvement of several genes (particularly RNF135 and CENTA2) in the increased risk of malignancy observed in NF1-microdeleted patients.

Different sized somatic NF1 locus rearrangements in neurofibromatosis 1-associated malignant peripheral nerve sheath tumors

Neurofibromatosis type 1 (NF1) patients are at increased risk of developing both benign (neurofibromas) and malignant (malignant peripheral nerve sheath tumors, MPNST) tumors. Molecular data on tumor progression are scarce, and few studies have compared the NF1 locus copy number in these two tumor types. To further explore the role of such NF1 locus rearrangements in NF1 tumorigenesis, and the likely disruption to the associated genes, the NF1 gene region was analyzed in NF1-associated tumors. DNA from three MPNSTs and one neurofibroma, excised from three unrelated NF1 patients, were analyzed using an NF1 region customized array-based comparative genomic hybridization. The somatic NF1 inactivation mutational mechanisms associated with MPNSTs appear to be different from those in benign neurofibromas. Interestingly, the MPNST-associated deletion breakpoints did not involve the paralogous repetitive sequences that are involved in most germline NF1 deletions. The somatic smallest common region of deletion overlap, however, was restricted to approximately the same ~2.2-Mb interval that encompassed most of the genes deleted in NF1 recurrent constitutional deletions. A number of genes in addition to NF1 on 17q (centromere to 17q24.2) may be involved in MPNST development. A larger study is warranted to confirm these findings. As NF1 patients with such germline NF1 deletions do exhibit increased risk of developing MPNST, these present findings emphasize the likely role of at least some of these NF1 flanking genes in MPNST pathobiology.

Investigation of gene dosage imbalances in patients with Noonan syndrome using multiplex ligation-dependent probe amplification analysis

The RAS-MAPK syndromes are a group of clinically and genetically related disorders caused by dysregulation of the RAS-MAPK pathway. A member of this group of disorders, Noonan syndrome (NS), is associated with several different genes within the RAS-MAPK pathway. To date, mutations in PTPN11, SOS1, KRAS, RAF1 and SHOC2 are known to cause NS and a small group of patients harbour mutations in BRAF, MEK1 or NRAS. The majority of the mutations are predicted to cause an up-regulation of the pathway; hence they are gain-of-function mutations. Despite recent advances in gene identification in NS, the genetic aetiology is still unknown in about 1/4 of patients. To investigate the contribution of gene dosage imbalances of RAS-MAPK-related genes to the pathogenesis of NS, a multiplex ligation-dependent probe amplification (MLPA) assay was developed. Two probe sets were designed for seven RAS-MAPK-syndrome-related candidate genes: PTPN11, SOS1, RAF1, KRAS, BRAF, MEK1 and MEK2. The probe sets were validated in 15 healthy control individuals and in glioma tumour cell lines. Subsequently, 44 NS patients negative for mutations in known NS-associated genes were screened using the two probe sets. The MLPA results for the patients revealed no gene dosage imbalances. In conclusion, the present results exclude copy number variation of PTPN11, SOS1, RAF1, KRAS, BRAF, MEK1 and MEK2 as a common pathogenic mechanism of NS. The validated and optimised RAS-MAPK probe sets presented here enable rapid high throughput screening of further patients with RAS-MAPK syndromes.

8p23.1 duplication syndrome differentiated from copy number variation of the defensin cluster at prenatal diagnosis in four new families

BACKGROUND

The 8p23.1 duplication syndrome and copy number variation of the 8p23.1 defensin gene cluster are cytogenetically indistinguishable but distinct at the molecular level. To our knowledge, the 8p23.1 duplication syndrome has been described at prenatal diagnosis only once and we report our experience with four further apparent duplications ascertained at prenatal diagnosis.

METHODS

Additional material at band 8p23.1 was detected using conventional G-banded cytogenetics in each case. Multiplex Ligation-dependent Probe Amplification (MLPA) or Fluorescence In Situ Hybridisation (FISH) were used depending on whether only DNA (Cases 1 and 4) or cytogenetic preparations (Cases 2 and 3) were available from the laboratory of origin. The extent of the duplication in Case 1 was retrospectively determined using array Comparative Genomic Hybridisation (array CGH).

RESULTS

Three cases of 8p23.1 duplication syndrome were found (Cases 1 to 3). Two were de novo and continued to term and the third, a paternally transmitted duplication, was terminated because of a previous child with psychomotor delay and 8p23.1 duplication syndrome. Case 1 was ascertained with a hypoplastic left heart but the ventricular septal and interventricular defects, in Cases 2 and 3 respectively, were found after ascertainment for advanced maternal age. By contrast, case 4 was a maternally transmitted copy number variation of the defensin cluster with normal outcome.

CONCLUSIONS

Our data underline the need to differentiate 8p23.1 duplications from copy number variation of the defensin cluster using FISH, MLPA or array CGH. Cardiac defects were ascertained by ultrasound in only one of the three duplication 8p23.1 pregnancies but were visible in two of the three at 21 to 22 weeks gestation. Our results provide further evidence that both deletion and duplication of the GATA4 transcription factor can give rise to a variety of conotruncal heart defects with variable penetrance and expressivity.

Detection and characterization of NF1 microdeletions by custom high resolution array CGH

In 5% to 10% of cases, neurofibromatosis type 1 is caused by microdeletions scattered across the entire NF1 gene and various neighboring genes. The phenotype appears to be more severe in patients with NF1 microdeletions than in patients with NF1 single point mutations. We have developed a new method for detecting and characterizing NF1 microdeletions based on a custom high-resolution oligonucleotide array comparative genomic hybridization by using the custom 8x15K Agilent array format. The array comprised a total of 14,207 oligonucleotide probes spanning the whole of chromosome 17, including 12,314 probes spanning an approximately 8 Mb interval surrounding the NF1 locus. We validated this approach by testing NF1 microdeleted DNA samples previously characterized by means of microsatellites and real-time PCR methods. Our array comparative genomic hybridization provided enough information for subsequent long-range PCR and nucleotide sequencing of the microdeletion endpoints. Unlike previously described methods, our array comparative genomic hybridization was able to unambiguously differentiate between the three types of microdeletions (type I, type II, and atypical) and to characterize atypical microdeletions. Further comparative studies of patients with well-characterized genotypes and phenotypes and different microdeletions sizes and breakpoints will help determine whether haploinsufficiency of deleted genes and/or genes rearrangements influence clinical outcomes.

A severe form of Noonan syndrome and autosomal dominant café-au-lait spots - evidence for different genetic origins

AIM

The clinical overlap among Noonan syndrome (NS), cardio-facio-cutaneous (CFC), LEOPARD and Costello syndromes as well as Neurofibromatosis type 1 is extensive, which complicates the process of diagnosis. Further genotype-phenotype correlations are required to facilitate future diagnosis of these patients. Therefore, investigations of the genetic cause of a severe phenotype in a patient with NS and the presence of multiple café-au-lait spots (CAL) spots in the patient and four members of the family were performed.

METHODS

Mutation analyses of candidate genes, PTPN11, NF1, SPRED1 and SPRED2, associated with these syndromes, were conducted using DNA sequencing.

RESULTS

A previously identified de novo mutation, PTPN11 F285L and an inherited NF1 R1809C substitution in the index patient were found. However, neither PTPN11 F285L, NF1 R1809C, SPRED1 nor SPRED2 segregated with CAL spots in the family. The results indicate that the familial CAL spots trait in this family is caused by a mutation in another gene, distinct from previous genes associated with CAL spots in these syndromes.

CONCLUSION

We suggest that the atypical severe symptoms in the index patient may be caused by an additive effect on the F285L mutation in PTPN11 by another mutation, for example the NF1 R1809C or alternatively, the not yet identified gene mutation associated with CAL spots in this family.

Somatic mosaicism for copy number variation in differentiated human tissues

Two major types of genetic variation are known: single nucleotide polymorphisms (SNPs), and a more recently discovered structural variation, involving changes in copy number (CNVs) of kilobase- to megabase-sized chromosomal segments. It is unknown whether CNVs arise in somatic cells, but it is, however, generally assumed that normal cells are genetically identical. We tested 34 tissue samples from three subjects and, having analyzed for each tissue < or =10(-6) of all cells expected in an adult human, we observed at least six CNVs, affecting a single organ or one or more tissues of the same subject. The CNVs ranged from 82 to 176 kb, often encompassing known genes, potentially affecting gene function. Our results indicate that humans are commonly affected by somatic mosaicism for stochastic CNVs, which occur in a substantial fraction of cells. The majority of described CNVs were previously shown to be polymorphic between unrelated subjects, suggesting that some CNVs previously reported as germline might represent somatic events, since in most studies of this kind, only one tissue is typically examined and analysis of parents for the studied subjects is not routinely performed. A considerable number of human phenotypes are a consequence of a somatic process. Thus, our conclusions will be important for the delineation of genetic factors behind these phenotypes. Consequently, biobanks should consider sampling multiple tissues to better address mosaicism in the studies of somatic disorders.

Characterization of a 7.6-Mb germline deletion encompassing the NF1 locus and about a hundred genes in an NF1 contiguous gene syndrome patient

We describe a large germline deletion removing the NF1 locus, identified by heterozygosity mapping based on microsatellite markers, in an 8-year-old French girl with a particularly severe NF1 contiguous gene syndrome. We used gene-dose mapping with sequence-tagged site real-time PCR to locate the deletion end points, which were precisely characterized by means of long-range PCR and nucleotide sequencing. The deletion is located on chromosome arm 17q and is exactly 7 586 986 bp long. It encompasses the entire NF1 locus and about 100 other genes, including numerous chemokine genes, an attractive in silico-selected cerebrally expressed candidate gene (designated NUFIP2, for nuclear fragile X mental retardation protein interacting protein 2; NM_020772) and four microRNA genes. Interestingly, the centromeric breakpoint is located in intron 4 of the PIPOX gene (pipecolic acid oxidase; NM_016518) and the telomeric breakpoint in intron 5 of the GGNBP2 gene (gametogenetin binding protein 2; NM_024835) coding a transcription factor. As PIPOX and GGNBP2 have the same transcriptional orientation, we postulated, and then confirmed, the existence of a chimeric transcript. This transcript, and/or haploinsufficiency of one or several deleted genes, could explain the clinical severity of the syndrome in this patient.

Absence of cutaneous neurofibromas in an NF1 patient with an atypical deletion partially overlapping the common 1.4 Mb microdeleted region

The majority of neurofibromatosis type 1 (NF1) microdeletions in 17q11.2 span approximately 1.4 Mb and have breakpoints that lie within the proximal and distal NF1-low copy repeats, termed NF1-REPs. Less frequent are patients with atypical deletions and non-recurring breakpoints. NF1 patients with gross deletions have been reported to manifest a more severe clinical phenotype than NF1 patients with intragenic mutations, and display early onset and extensive growth of neurofibromas. It has been suggested that the deletion of a neighboring gene or genes in addition to the NF1 gene may modify the expression of the disease, particularly with regard to the high burden of cutaneous neurofibromas. Thus, atypical deletions partially overlapping with the common 1.4 Mb microdeletion interval could prove useful in identifying possible genetic modifiers in the NF1 gene region whose haploinsufficiency might promote neurofibroma growth. Here we report a 20-year-old female who has an atypical deletion with a proximal breakpoint in NF1 intron 21 and a distal deletion breakpoint in the ACCN1 gene. The deletion spans 2.7 Mb and was mediated by an intrachromosomal non-homology-driven mechanism, for example, non-homologous end-joining (NHEJ). Remarkably, this patient did not exhibit cutaneous neurofibromas. However, genotype-phenotype comparisons in this and other previously reported patients with atypical deletions partially overlapping the commonly deleted 1.4 Mb interval do not identify a specific deleted region that is associated with increased neurofibroma growth.

Phenotypically concordant and discordant monozygotic twins display different DNA copy-number-variation profiles

The exploration of copy-number variation (CNV), notably of somatic cells, is an understudied aspect of genome biology. Any differences in the genetic makeup between twins derived from the same zygote represent an irrefutable example of somatic mosaicism. We studied 19 pairs of monozygotic twins with either concordant or discordant phenotype by using two platforms for genome-wide CNV analyses and showed that CNVs exist within pairs in both groups. These findings have an impact on our views of genotypic and phenotypic diversity in monozygotic twins and suggest that CNV analysis in phenotypically discordant monozygotic twins may provide a powerful tool for identifying disease-predisposition loci. Our results also imply that caution should be exercised when interpreting disease causality of de novo CNVs found in patients based on analysis of a single tissue in routine disease-related DNA diagnostics.

Germline and somatic NF1 gene mutation spectrum in NF1-associated malignant peripheral nerve sheath tumors (MPNSTs)

About 10% of neurofibromatosis type 1 (NF1) patients develop malignant peripheral nerve sheath tumors (MPNSTs) and represent considerable patient morbidity and mortality. Elucidation of the genetic mechanisms by which inherited and acquired NF1 disease gene variants lead to MPNST development is important. A study was undertaken to identify the constitutional and somatic NF1 mutations in 34 MPNSTs from 27 NF1 patients. The NF1 germline mutations identified in 22 lymphocytes DNA from these patients included seven novel mutations and a large 1.4-Mb deletion. The NF1 germline mutation spectrum was similar to that previously identified in adult NF1 patients without MPNST. Somatic NF1 mutations were identified in tumor DNA from 31 out of 34 MPNSTs, of which 28 were large genomic deletions. The high prevalence (>90%) of such deletions in MPNST contrast with the =or<20% found in benign neurofibromas and is indicative of the involvement of different mutational mechanisms in these tumors. Coinactivation of the TP53 gene by deletion, or by point mutation along with NF1 gene inactivation, is known to exacerbate disease symptoms in NF1, therefore TP53 gene inactivation was screened. DNA from 20 tumors showed evidence for loss of heterozygosity (LOH) across the TP53 region in 11 samples, with novel TP53 point mutations in four tumors.

Copy number variations in the NF1 gene region are infrequent and do not predispose to recurrent type-1 deletions

Gross deletions of the NF1 gene at 17q11.2 belong to the group of 'genomic disorders' characterized by local sequence architecture that predisposes to genomic rearrangements. Segmental duplications within regions associated with genomic disorders are prone to non-allelic homologous recombination (NAHR), which mediates gross rearrangements. Copy number variants (CNVs) without obvious phenotypic consequences also occur frequently in regions of genomic disorders. In the NF1 gene region, putative CNVs have been reportedly detected by array comparative genomic hybridization (array CGH). These variants include duplications and deletions within the NF1 gene itself (CNV1) and a duplication that encompasses the SUZ12 gene, the distal NF1-REPc repeat and the RHOT1 gene (CNV2). To explore the possibility that these CNVs could have played a role in promoting deletion mutagenesis in type-1 deletions (the most common type of gross NF1 deletion), non-affected transmitting parents of patients with type-1 NF1 deletions were investigated by multiplex ligation-dependent probe amplification (MLPA). However, neither CNV1 nor CNV2 were detected. This would appear to exclude these variants as frequent mediators of NAHR giving rise to type-1 deletions. Using MLPA, we were also unable to confirm CNV1 in healthy controls as previously reported. We conclude that locus-specific techniques should be used to independently confirm putative CNVs, originally detected by array CGH, to avoid false-positive results. In one patient with an atypical deletion, a duplication in the region of CNV2 was noted. This duplication could have occurred concomitantly with the deletion as part of a complex rearrangement or may alternatively have preceded the deletion.

Overlapping phenotype of Wolf-Hirschhorn and Beckwith-Wiedemann syndromes in a girl with der(4)t(4;11)(pter;pter)

We report on an 8-month-old girl with a novel unbalanced chromosomal rearrangement, consisting of a terminal deletion of 4p and a paternal duplication of terminal 11p. Each of these is associated with the well-known clinical phenotypes of Wolf-Hirschhorn syndrome (WHS) and Beckwith-Wiedemann syndrome (BWS), respectively. She presented for clinical evaluation of dysmorphic facial features, developmental delay, atrial septal defect (ASD), and left hydronephrosis. High-resolution cytogenetic analysis revealed a normal female karyotype, but subtelomeric fluorescence in situ hybridization (FISH) analysis revealed a der(4)t(4;11)(pter;pter). Both FISH and microarray CGH studies clearly demonstrated that the WHS critical regions 1 and 2 were deleted, and that the BWS imprinted domains (ID) 1 and 2 were duplicated on the der(4). Parental chromosome analysis revealed that the father carried a cryptic balanced t(4;11)(pter;pter). As expected, our patient manifests findings of both WHS (a growth retardation syndrome) and BWS (an overgrowth syndrome). We compare her unique phenotypic features with those that have been reported for both syndromes.

NF1 microduplication first clinical report: association with mild mental retardation, early onset of baldness and dental enamel hypoplasia?

NF1 microdeletion syndrome is a common dominant genomic disorder responsible for around 5% of type I neurofibromatosis cases. The majority of cases are caused by mutations arising within the NF1 gene. NF1 microdeletion carriers present a more severe phenotype than patients with intragenic mutations, including mental retardation, cardiac anomalies and dysmorphic features. Here, we report on two brothers with mental retardation presenting a microduplication of the NF1 microdeletion syndrome region detected by array-CGH analysis. Main phenotypic features are mental deficiency, early onset of baldness (15 years old), dental enamel hypoplasia and minor facial dysmorphism. The breakpoint regions coincide with the repeats, and the recombination hot spots shown to mediate NF1 microdeletion through NAHR. A screening of the patients' familial relatives showed that this microduplication segregates in the family for at least two generations. This result demonstrates that both deletion and duplication of the NF1 region, at cytogenetic band 17q11.2, give rise to viable gametes, even if only NF1 microdeletions have been reported until now. Our study reports seven cases of NF1 microduplication within one family. Similar phenotypic abnormalities were present in most of the individuals, however, two displayed a normal phenotype, suggesting a potential incomplete penetrance of the phenotype associated with NF1 microduplication.

A previously unrecognized microdeletion syndrome on chromosome 22 band q11.2 encompassing the BCR gene

Susceptibility of the chromosome 22q11.2 region to rearrangements has been recognized on the basis of common clinical disorders such as the DiGeorge/velocardiofacial syndrome (DG/VCFs). Recent evidence has implicated low-copy repeats (LCRs); also known as segmental duplications; on 22q as mediators of nonallelic homologous recombination (NAHR) that result in rearrangements of 22q11.2. It has been shown that both deletion and duplication events can occur as a result of NAHR caused by unequal crossover of LCRs. Here we report on the clinical, cytogenetic and array CGH studies of a 15-year-old Hispanic boy with history of learning and behavior problems. We suggest that he represents a previously unrecognized microdeletion syndrome on chromosome 22 band q11.2 just telomeric to the DG/VCFs typically deleted region and encompassing the BCR gene. Using a 32K BAC array CGH chip we were able to refine and precisely narrow the breakpoints of this microdeletion, which was estimated to be 1.55-1.92 Mb in size and to span approximately 20 genes. This microdeletion region is flanked by LCR clusters containing several modules with a very high degree of sequence homology (>95%), and therefore could play a causal role in its origin.

Detection of copy number changes at the NF1 locus with improved high-resolution array CGH

Neurofibromatosis type 1 (NF1) is a common autosomal dominant disease caused by various types of mutations in the NF1 gene. We have previously developed a locus-specific DNA microarray for detection of copy number changes at the NF1 locus by comparative genomic hybridization (CGH) analysis. The original array contains 183 probes pooled from 444 polymerase chain reaction (PCR) products. In the current work, we have used 493 probes derived from single PCR products (200--998 bp in size) to construct a higher resolution array with a smaller average probe size for molecular diagnosis of NF1. This has improved the average resolution from 12.6 kb in the previous array to 4.5 kb in the current version. The performance of the newly constructed microarray was validated with 14 well-characterized NF1 mutations for CGH analysis. These mutations represent deletions from approximately 7 kb to over 2 Mb in size. Using this array, we examined a total of 55 NF1 patients for copy number changes at the NF1 locus, detecting deletions in four of them. These results demonstrate that a locus-specific microarray constructed from single PCR products can efficiently detect copy number changes at the NF1 locus, providing a simple method for the molecular diagnosis of NF1.

Type 2 NF1 deletions are highly unusual by virtue of the absence of nonallelic homologous recombination hotspots and an apparent preference for female mitotic recombination

Approximately 5% of patients with neurofibromatosis type 1 (NF1) exhibit gross deletions that encompass the NF1 gene and its flanking regions. The breakpoints of the common 1.4-Mb (type 1) deletions are located within low-copy repeats (NF1-REPs) and cluster within a 3.4-kb hotspot of nonallelic homologous recombination (NAHR). Here, we present the first comprehensive breakpoint analysis of type 2 deletions, which are a second type of recurring NF1 gene deletion. Type 2 deletions span 1.2 Mb and are characterized by breakpoints located within the SUZ12 gene and its pseudogene, which closely flank the NF1-REPs. Breakpoint analysis of 13 independent type 2 deletions did not reveal any obvious hotspots of NAHR. However, an overrepresentation of polypyrimidine/polypurine tracts and triplex-forming sequences was noted in the breakpoint regions that could have facilitated NAHR. Intriguingly, all 13 type 2 deletions identified so far are characterized by somatic mosaicism, which indicates a positional preference for mitotic NAHR within the NF1 gene region. Indeed, whereas interchromosomal meiotic NAHR occurs between the NF1-REPs giving rise to type 1 deletions, NAHR during mitosis appears to occur intrachromosomally between the SUZ12 gene and its pseudogene, thereby generating type 2 deletions. Such a clear distinction between the preferred sites of mitotic versus meiotic NAHR is unprecedented in any other genomic disorder induced by the local genomic architecture. Additionally, 12 of the 13 mosaic type 2 deletions were found in females. The marked female preponderance among mosaic type 2 deletions contrasts with the equal sex distribution noted for type 1 and/or atypical NF1 deletions. Although an influence of chromatin structure was strongly suspected, no sex-specific differences in the methylation pattern exhibited by the SUZ12 gene were apparent that could explain the higher rate of mitotic recombination in females.

Whole-genome array-CGH for detection of submicroscopic chromosomal imbalances in children with mental retardation

Chromosomal imbalances are the major cause of mental retardation (MR). Many of these imbalances are caused by submicroscopic deletions or duplications not detected by conventional cytogenetic methods. Microarray-based comparative genomic hybridization (array-CGH) is considered to be superior for the investigation of chromosomal aberrations in children with MR, and has been demonstrated to improve the diagnostic detection rate of these small chromosomal abnormalities. In this study we used 1 Mb genome-wide array-CGH to screen 48 children with MR and congenital malformations for submicroscopic chromosomal imbalances, where the underlying cause was unknown. All children were clinically investigated and subtelomere FISH analysis had been performed in all cases. Suspected microdeletion syndromes such as deletion 22q11.2, Williams-Beuren and Angelman syndromes were excluded before array-CGH analysis was performed. We identified de novo interstitial chromosomal imbalances in two patients (4%), and an interstitial deletion inherited from an affected mother in one patient (2%). In another two of the children (4%), suspected imbalances were detected but were also found in one of the non-affected parents. The yield of identified de novo alterations detected in this study is somewhat less than previously described, and might reflect the importance of which selection criterion of patients to be used before array-CGH analysis is performed. However, array-CGH proved to be a high-quality and reliable tool for genome-wide screening of MR patients of unknown etiology.

Molecular diagnosis of neurofibromatosis type 1: 2 years experience

Our experience of providing an NF1 gene diagnostic mutation detection service as part of the U.K. Genetic Testing Network (UKGTN) is presented. A total of 169 unrelated individuals suspected of having neurofibromatosis type I (NF1) were referred for NF1 diagnostic testing over a 2 year period. Mutation analysis of the entire NF1 coding region and the flanking splice sites was carried out, and included the use of a combination of FISH, dHPLC and MLPA. Possible disease causing mutations were identified in 109 (64%) cases. These comprised 88 different sequence alterations, of which 57 were novel. Out of the 169 cases referred, there were 102 patients with reliable clinical data, of whom 78 satisfied the NIH diagnostic criteria for NF1. Within this better defined cohort of NF1 patients, NF1 mutations were identified in 61 individuals (78%), showing the importance of clinical selection on overall test sensitivity, and highlighting the problem of full clinical data collection in the audit of routine services. As mutation detection technologies advance, facilitating direct sequencing of all coding and flanking non-coding regions of the NF1 gene, the development of an even more cost-effective, quick and sensitive diagnostic test for future testing of NF1 is discussed.

Genome wide copy number abnormalities in pediatric medulloblastomas as assessed by array comparative genome hybridization

Array-based comparative genomic hybridization was used to characterize 22 medulloblastomas in order to precisely define genetic alterations in these malignant childhood brain tumors. The 17p(-)/17q(+) copy number abnormality (CNA), consistent with the formation of isochromosome 17q, was the most common event (8/22). Amplifications in this series included MYCL, MYCN and MYC previously implicated in medulloblastoma pathogenesis, as well as novel amplicons on chromosomes 2, 4, 11 and 12. Losses involving chromosomes 1, 2, 8, 10, 11, 16 and 19 and gains of chromosomes 4, 7, 8, 9 and 18 were seen in greater than 20% of tumors in this series. A homozygous deletion in 11p15 defines the minimal region of loss on this chromosome arm. In order to map the minimal regions involved in losses, gains and amplifications, we combined aCGH data from this series with that of two others obtained using the same RPCI BAC arrays. As a result of this combined analysis of 72 samples, we have defined specific regions on chromosomes 1, 8p, 10q, 11p and 16q which are frequently involved in CNAs in medulloblastomas. Using high density oligonucleotide expression arrays, candidate genes were identified within these consistently involved regions in a subset of the tumors.

Detection of small genomic imbalances using microarray-based multiplex amplifiable probe hybridization

Array-based genome-wide screening methods were recently introduced to clinical practice in order to detect small genomic imbalances that may cause severe genetic disorders. The continuous advancement of such methods plays an extremely important role in diagnostic genetics and medical genomics. We have modified and adapted the original multiplex amplifiable probe hybridization (MAPH) to a novel microarray format providing an important new diagnostic tool for detection of small size copy-number changes in any locus of human genome. Here, we describe the new array-MAPH diagnostic method and show proof of concept through fabrication, interrogation and validation of a human chromosome X-specific array. We have developed new bioinformatic tools and methodology for designing and producing amplifiable hybridization probes (200-600 bp) for array-MAPH. We designed 558 chromosome X-specific probes with median spacing 238 kb and 107 autosomal probes, which were spotted onto microarrays. DNA samples from normal individuals and patients with known and unknown chromosome X aberrations were analyzed for validation. Array-MAPH detected exactly the same deletions and duplications in blind studies, as well as other unknown small size deletions showing its accuracy and sensitivity. All results were confirmed by fluorescence in situ hybridization and probe-specific PCR. Array-MAPH is a new microarray-based diagnostic tool for the detection of small-scale copy-number changes in complex genomes, which may be useful for genotype-phenotype correlations, identification of new genes, studying genetic variation and provision of genetic services.

Analysis of copy number variation in the normal human population within a region containing complex segmental duplications on 22q11 using high-resolution array-CGH

A previously detected copy number polymorphism (Ep CNP) in patients affected with neuroectodermal tumors led us to investigate its frequency and length in the normal population. For this purpose, a program called Sequence Allocator was developed and applied for the construction of an array that consisted of unique and duplicated fragments, allowing the assessment of copy number variation within regions of segmental duplications. The average resolution of this array was 11 kb and we determined the size of the Ep CNP to be 290 kb. Analysis of normal controls identified 7.7 and 7.1% gains in peripheral blood and lymphoblastoid cell line (LCL) DNA, respectively, while deletions were found only in the LCL group (7.1%). This array platform allows the detection of DNA copy number variation within regions of pronounced genomic complexity, which constitutes an improvement over available technologies.

Microarray-based survey of CpG islands identifies concurrent hyper- and hypomethylation patterns in tissues derived from patients with breast cancer

Maintenance of CpG island methylation in the genome is crucial for cellular homeostasis and this balance is disrupted in cancer. Our rationale was to compare the methylation of CpG islands in tissues (tumor, healthy breast and blood) from patients with breast cancer. We studied 72 genes in 103 samples using microarray hybridization and bisulfite sequencing. We observed tumor specific hyper- or hypomethylation of five genes; COL9A1, MT1A, MT1J, HOXA5 and FLJ45983. A general drop of methylation in COL9A1 was apparent in tumors, when compared with blood and healthy breast tissue. Furthermore, one tumor displayed a complete loss of methylation of all five genes, suggesting overall impairment of methylation. The downstream, evolutionary conserved island of HOXA5 showed hypomethylation in 18 tumors and complete methylation in others. This CpG island also displayed a semimethylated state in the majority of normal breast samples, when compared to complete methylation in blood. Distinct methylation patterns were further seen in MT1J and MT1A, belonging to the metallothionein gene family. The CpG islands of these genes are spaced by 2 kb, which shows selective methylation of two structurally and functionally related genes. The promoters of FLJ45983 and MT1A were methylated above 25% in 18 primary and metastatic tumors. Concurrently, there was also >10% methylation of healthy breast tissue in 11 and 5 samples, respectively. This suggests that the methylation process for the latter two genes takes place already in normal breast cells. Our results also point to a considerable heterogeneity of epigenetic disturbance in breast cancer. This article contains Supplementary Material available at http://www.interscience.wiley.com/jpages/1045-2257/suppmat.