Correlation between large rearrangements and patient phenotypes in NF1 deletion syndrome: an update and review.

About 5-10% of neurofibromatosis type 1 (NF1) patients exhibit large genomic germline deletions that remove the NF1 gene and its flanking regions. The most frequent NF1 large deletion is 1.4 Mb, resulting from homologous recombination between two low copy repeats. This "type-1" deletion is associated with a severe clinical phenotype in NF1 patients, with several phenotypic manifestations including learning disability, a much earlier development of cutaneous neurofibromas, an increased tumour risk, and cardiovascular malformations. NF1 adjacent co-deleted genes could act as modifier loci for the specific clinical manifestations observed in deleted NF1 patients. Furthermore, other genetic modifiers (such as CNVs) not located at the NF1 locus could also modulate the phenotype observed in patients with large deletions. In this study, we analysed 22 NF1 deletion patients by genome-wide array-CGH with the aim (1) to correlate deletion length to observed phenotypic features and their severity in NF1 deletion syndrome, and (2) to identify whether the deletion phenotype could also be modulated by copy number variations elsewhere in the genome. We then review the role of co-deleted genes in the 1.4 Mb interval of type-1 deletions, and their possible implication in the main clinical features observed in this high-risk group of NF1 patients.

Pigment epithelium derived factor drives melanocyte proliferation and migration in neurofibromatosis café au lait macules.

Background: RASopathies, which include neurofibromatosis type 1 (NF1), are defined by Ras/mitogen-activated protein kinase (Ras/MAPK) pathway activation. They represent a group of clinically related disorders often characterised by multiple Café au Lait Macules (CALMs). Objectives: To determine, using in depth transcriptomic analysis of NF1 melanocytes from CALM and unaffected skin, (1) the gene(s) responsible for melanocyte proliferation and migration, and (2) the activated signalling pathway(s) in NF1 melanoma. Methods: Classical NF1 (n = 2, who develop tumours) and 3bp deletion NF1 (p. Met992del, who do not develop tumours) (n = 3) patients underwent skin biopsies from CALM and unaffected skin. Melanocytes were isolated and propagated, with five replicates from each tissue sample. DNA and RNA were extracted for mutational analysis and transcriptomic profiling with six replicates per sample. Mechanistic determination was undertaken using melanocyte and melanoma cell lines. Results: All CALMs in NF1 were associated with biallelic NF1 loss, resulting in amplification of Ras/MAPK and Wnt pathway signalling. CALMs were also associated with reduced SERPINF1 gene expression (and pigment epithelium-derived factor (PEDF) levels, the reciprocal protein), a known downstream target of the master regulator of melanocyte differentiation microphthalmia-associated transcription factor (MITF), leading to increased melanocyte proliferation, migration and invasion. In classical NF1 and melanoma, but not 3bp deletion NF1, there was also activation of the PI3K/AKT pathway. Pigment epithelium-derived factor was found to reduce cell proliferation and invasion of NF1 melanoma. Conclusions: Melanocyte proliferation and migration leading to CALMs in NF1 arises from biallelic NF1 loss, resulting in RAS/MAPK pathway activation, and reduced expression of the tumour suppressor PEDF. Activation of the PI3K/AKT pathway in classical NF1 and NF1 melanoma may facilitate tumour growth.

Microarray-based copy number analysis of neurofibromatosis type-1 (NF1)-associated malignant peripheral nerve sheath tumors reveals a role for Rho-GTPase pathway genes in NF1 tumorigenesis.

Neurofibromatosis type-1 (NF1) is associated with the growth of benign and malignant tumors. Approximately 15% of NF1 patients develop malignant peripheral nerve sheath tumors (MPNSTs), underlining the need to identify specific diagnostic/prognostic biomarkers associated with MPNST development. The Affymetrix Genome-Wide Human single-nucleotide polymorphism (SNP) Array 6.0 was used to perform SNP genotyping and copy number alteration (CNA), loss-of-heterozygosity (LOH), and copy number neutral-LOH (CNN-LOH) analyses of DNA isolated from 15 MPNSTs, five benign plexiform neurofibromas (PNFs), and patient-matched lymphocyte DNAs. MPNSTs exhibited high-level CNN-LOH, with recurrent changes occurring in MPNSTs but not PNFs. CNN-LOH was evident in MPNSTs but occurred less frequently than genomic deletions. CNAs involving the ITGB8, PDGFA, Ras-related C3 botulinum toxin substrate 1 (RAC1) (7p21-p22), PDGFRL (8p22-p21.3), and matrix metallopeptidase 12 (MMP12) (11q22.3) genes were specific to MPNSTs. Pathway analysis revealed the MPNST-specific amplification of seven Rho-GTPase pathway genes and several cytoskeletal remodeling/cell adhesion genes. In knockdown experiments employing short-hairpin RAC1, ROCK2, PTK2, and LIMK1 RNAs to transfect both control and MPNST-derived cell lines, cell adhesion was significantly increased in the MPNST cell lines, whereas wound healing, cell migration, and invasiveness were reduced, consistent with a role for these Rho-GTPase pathway genes in MPNST development and metastasis. These results suggest new targets for therapeutic intervention in relation to MPNSTs.

NF1 microdeletions in neurofibromatosis type 1: from genotype to phenotype.

In 5-10% of patients, neurofibromatosis type 1 (NF1) results from microdeletions that encompass the entire NF1 gene and a variable number of flanking genes. Two recurrent microdeletion types are found in most cases, with microdeletion breakpoints located in paralogous regions flanking NF1 (proximal NF1-REP-a and distal NF1-REP-c for the 1.4 Mb type-1 microdeletion, and SUZ12 and SUZ12P for the 1.2 Mb type-2 microdeletion). A more severe phenotype is usually associated with NF1 microdeletion patients than in those with intragenic mutations. We characterized NF1 microdeletions in 70 unrelated NF1 microdeleted patients using a high-resolution NF1 custom array comparative genomic hybridization (CGH). Genotype-phenotype correlations were studied in 58 of these microdeletion patients and compared to 389 patients with intragenic truncating NF1 mutations and phenotyped in the same standardized way. Our results confirmed in an unbiased manner the existence of a contiguous gene syndrome with a significantly higher incidence of learning disabilities and facial dysmorphism in microdeleted patients compared to patients with intragenic NF1 mutations. Microdeleted NF1 patients also showed a trend toward significance for childhood overgrowth. High-resolution array-CGH identified a new recurrent approximately 1.0 Mb microdeletion type, designated as type-3, with breakpoints in the paralogous regions middle NF1-REP-b and distal NF1-REP-c.

The spectrum of somatic and germline NF1 mutations in NF1 patients with spinal neurofibromas.

Neurofibromatosis type 1 (NF1) is a common inherited complex multi-system disorder associated with the growth of various benign and malignant tumors. About 40% of NF1 patients develop spinal tumors, of whom some have familial spinal neurofibromatosis (FSNF), a variant form of NF1 in which patients present with multiple bilateral spinal tumors but have few other clinical features of the disease. We have studied 22 spinal neurofibromas derived from 14 unrelated NF1 patients. Seven of these patients satisfied the diagnostic criteria of NF1 while the remaining seven had only few features of NF1. The latter group defined as FSNF harbored significantly higher number of missense or missense and splice-site germline mutations compared to the group with classical NF1. This is the first study to describe NF1 somatic mutations in spinal neurofibromas. Loss-of-heterozygosity (LOH) was identified in 8/22 of the spinal tumors, 75% of LOH observed was found to result from mitotic recombination, suggesting that this may represent a frequent mutational mechanisms in these benign tumors. No evidence for LOH of the TP53 gene was found in these tumors.

Germline and somatic NF1 gene mutations in plexiform neurofibromas.

Neurofibromatosis type 1 (NF1), a common autosomal dominant neurogenetic disorder affecting 1 in 4000 individuals worldwide, results from functional inactivation of the 17q11.2-located NF1 gene. Plexiform neurofibroma (PNF) is a congenital benign tumour present in 30-50% of NF1 patients, which in about 10-15% of cases, can develop into a malignant peripheral nerve sheath tumour (MPNST). This study aimed to characterise the NF1 germline and somatic mutations associated with such tumours by DNA analysis in 51 PNFs resected from 44 unrelated NF1 patients. Germline mutations were identified in 35 patients, of which 21 were novel. Somatic NF1 mutations were found in 29 PNF DNAs, which included 9 point mutations, 5 being novel, and 20 tumour DNA samples exhibiting, either loss of heterozygosity (LOH) of the NF1 gene region (16 tumours), or complete or partial NF1 gene deletions analyzed by multiplex ligation-dependent probe amplification (MPLA) analysis. The type of NF1 germline mutations detected in patients with PNF were similar to those detected in most NF1 patients. LOH of the NF1 gene region, as identified by marker analysis and/or MLPA, was detected in only 20/29 (69%) PNFs, compared to the >90% LOH previously found in MPNST. This systematic analysis of the NF1 germline and somatic mutations associated with PNF development suggest that in most such tumours neither the NF1 somatic mutation type, nor its gene location, is influenced by the underlying NF1 germline mutation. Evidence for LOH involving the TP53 gene identified in the PNFs is also reported for the first time.

A novel mutation in the NF1 gene in two siblings with neurofibromatosis type 1 and bilateral optic pathway glioma.

We present the clinical and ophthalmological findings, genetic analysis, and therapy of two siblings with NF1 and bilateral OPG. In genetic analysis, a heteroduplex profile was detected in exon 4b of the NF1 gene for the affected patients and mother. Sequencing of the DNA samples identified a C > T nucleotide change in exon 4b (c484CAG > TAG). This nonsense mutation resulted in a change of glutamine to a stop codon (Q162X) and is a novel NF1 gene alteration.

Somatic alterations of the NF1 gene in an NF1 individual with multiple benign tumours (internal and external) and malignant tumour types.

Neurofibromatosis type 1 is a common familial cancer syndrome, affecting about 1 in every 4,000 individuals worldwide. We have carried out NF1 gene mutation analysis on DNA isolated from 25 tumours (dermal and plexiform neurofibromas, malignant peripheral nerve sheath tumour, MPNST), obtained at post-mortem from an NF1 patient. Macro and micro sequence alterations of the NF1 gene were studied by dHPLC, microsatellite, RFLP markers and multiplex ligation probe amplification (MLPA). The underlying germline mutation involves a deletion of exons 2 and 3. Of the 25 tumours studied from this patient, characterised somatic mutations were identified in 9 tumours, these were six small deletions (748del T, 2534-2557 del 24bp, 2843delA, 3047-3048 del GT, 4743del G, 7720-7721 delAA), an insertion 649 ins 73 bp, a non-sense mutation R1513X and a single splice site mutation, IVS4C-1 G>A, eight of these represent novel sequence changes in the gene. Evidence for loss of heterozygosity (LOH) was identified in DNA from 7 of the tumours. Each of the tumours analysed contained a different somatic NF1 mutation, indicating that each tumour is the result of an independent somatic event. The somatic mutation detection rate in this study is 64% (16/25), is one of the highest rates in genomic DNA reported so far in a single NF1 patient. Only 68 characterised NF1 somatic mutations have so far been reported and so our data will contribute to NF1 somatic mutational spectrum of the NF1 gene and will be important for understanding the molecular basis of NF1 tumorigenesis.

The heterogeneous nature of germline mutations in NF1 patients with malignant peripheral serve sheath tumours (MPNSTs).

Malignant peripheral nerve sheath tumours (MPNSTs) are a major cause of mortality in patients with neurofibromatosis 1 (NF1). We have analysed lymphocyte DNA samples from 30 NF1 patients with MPNSTs to determine their underlying constitutional NF1 gene mutations. Mutations were detected in 27/30 (90%) of these patients. NF1 mutations identified included nonsense, missense, frameshift, splice site mutation and single or multi-exonic deletions and with no obvious clustering of the mutations across the gene. Fourteen of the mutations represent novel gene changes. There did not appear to be any relationship between the mutation type and the level of clinical severity observed. Of the 20 patients with high grade MPNSTs, seven patients had small (<20 bp) and multi-exonic deletions and three had small insertions (<20 bp). Several studies have suggested that NF1 patients with a constitutional 1.5 Mb deletion of the NF1 gene have an increased risk of developing malignant peripheral nerve sheath tumours (MPNSTs). None of our patients had a 1.5 Mb deletion. Larger prospective studies are needed to ascertain whether there is a different spectrum of NF1 mutations in NF1 patients with high grade compared to low grade MPNSTs and of patients with the 1.5Mb deletion, in order to determine the true frequency of MPNST in this sub-group of NF1 patients.

Exploring the somatic NF1 mutational spectrum associated with NF1 cutaneous neurofibromas.

Neurofibromatosis type-1 (NF1), caused by heterozygous inactivation of the NF1 tumour suppressor gene, is associated with the development of benign and malignant peripheral nerve sheath tumours (MPNSTs). Although numerous germline NF1 mutations have been identified, relatively few somatic NF1 mutations have been described in neurofibromas. Here we have screened 109 cutaneous neurofibromas, excised from 46 unrelated NF1 patients, for somatic NF1 mutations. NF1 mutation screening (involving loss-of-heterozygosity (LOH) analysis, multiplex ligation-dependent probe amplification and DNA sequencing) identified 77 somatic NF1 point mutations, of which 53 were novel. LOH spanning the NF1 gene region was evident in 25 neurofibromas, but in contrast to previous data from MPNSTs, it was absent at the TP53, CDKN2A and RB1 gene loci. Analysis of DNA/RNA from neurofibroma-derived Schwann cell cultures revealed NF1 mutations in four tumours whose presence had been overlooked in the tumour DNA. Bioinformatics analysis suggested that four of seven novel somatic NF1 missense mutations (p.A330T, p.Q519P, p.A776T, p.S1463F) could be of functional/clinical significance. Functional analysis confirmed this prediction for p.S1463F, located within the GTPase-activating protein-related domain, as this mutation resulted in a 150-fold increase in activated GTP-bound Ras. Comparison of the relative frequencies of the different types of somatic NF1 mutation observed with those of their previously reported germline counterparts revealed significant (P=0.001) differences. Although non-identical somatic mutations involving either the same or adjacent nucleotides were identified in three pairs of tumours from the same patients (P<0.0002), no association was noted between the type of germline and somatic NF1 lesion within the same individual.

Molecular evolution of a neurofibroma to malignant peripheral nerve sheath tumor (MPNST) in an NF1 patient: correlation between histopathological, clinical and molecular findings.

OBJECTIVE: Neurofibromatosis type 1 (NF1) patients have a 13% risk of developing a malignant peripheral nerve sheath tumor (MPNST). Many MPNSTs are histopathologically complex, with regions exhibiting features of the original benign plexiform neurofibroma (PNF), of an atypical PNF, or of MPNST showing varying degrees of de-differentiation. This study analyzed the genetic alterations associated with this pathological heterogeneity in order to identify the genetic processes involved in transformation from a benign to an aggressive malignant tumor. METHODS: A histological and molecular analysis of a single MPNST tumor that was subdivided into three histopathologically distinct regions, a benign PNF (region 1), an atypical PNF (region 2), and a high-grade MPNST (region 3), was carried out. Tumor DNA from each region was analyzed in conjunction with the patient's lymphocyte DNA. Somatic mutation analyses included loss-of heterozygosity (LOH), MLPA analysis, NF1 gene sequencing, and a microarray comparative genomic hybridisation (array CGH) analysis. RESULTS: The patient had a germline NF1 splice-site mutation. The NF1-associated LOH analysis found that LOH increased in the three tumor areas, with 9, 42, and 97% LOH evident in regions 1, 2, and 3, respectively. Additional genetic changes, including losses of TP53, RB1, CDKN2A, and of several oncogenes and cell-cycle genes, were found only in the malignant MPNST (region 3). Array CGH also identified genomic gains and losses in DNA from region 3. DISCUSSION: This is the first study that correlates the histological and molecular changes associated with MPNST development, confirming the significant cellular and genetic heterogeneity that poses both diagnostic and therapeutic challenges.