Abstract
Background/Aim: A deeper understanding of the complex molecular pathology of brain malignancies is needed in order to develop more effective and targeted therapies of these highly lethal disorders. In an effort to further enlighten the molecular pathology of brain oncogenesis involving the her-2 (erbB-2/neu/ngl)/N-ras/nf1 pathway, we screened the genotypes of specimens from various types of brain tumors. Materials and Methods: The studied specimens included 35 biopsies of four general categories: 13 neuroglial tumors (4 astrocytomas, 2 oligodendrogliomas, 7 glioblastomas multiforme), 14 meningiomas, 3 other nervous system tumors (2 schwannomas, 1 craniopharyngioma) and 5 metastatic tumors (such as lung carcinomas and chronic myelocytic leukemia). Screening for most common mutations in oncogenes her-2, N-ras and tumor suppressor gene nf1 was conducted with molecular hybridization techniques (Southern blotting, dot blot and single-strand conformational polymorphism (SSCP) analysis, respectively), and was confirmed by DNA sequencing. Results: Gene amplification of her-2 was observed in only two cases (6%), namely in one glioblastoma and in one meningioma. Screening of 3 hot spot codons of the N-ras gene (12, 13 and 61) and subsequent DNA sequencing revealed mutations in 19 biopsies encompassing all categories (54%). Screening for mutations in exons of the nf1 gene by SSCP analysis detected a novel nonsense mutation in exon 31 in a unique case of a glioblastoma biopsy (3%) taken from a patient without neurofibromatosis type I. Conclusion: Activated N-ras appears to be a major oncogene in brain oncogenesis, exhibiting the most important role in the her-2/N-ras/nf1 pathway.
Despite advances in therapy, brain cancer remains a highly lethal malignant disease worldwide, with a median survival period of less than a year (1). Hence, further understanding of the complex molecular pathology of brain malignancies is needed in order to develop more effective and possibly targeted therapies.
According to the Knudson model, genesis and progression of tumors require the alterations of a number of somatic cell genes in a stepwise process. In the last decades, molecular genetic analysis revealed that oncogenesis results from accumulated mutations in two major classes of cell growth regulatory genes: the oncogenes and the tumor suppressor genes (1). Most of these genes belong to an intracellular signal transduction pathway or to cell cycle regulators.
The most well-known oncogene implicated in the development of brain tumors is Neuroblastoma–rat sarcoma (N-ras). It is located on chromosome 1p13.2 and encodes for a membrane bound GTPase switch protein, called p21Nras (2). Normal p21Nras is involved in the control of cell differentiation and growth, existing in the equilibrium between its inactive and active states (GDP-p21Nras versus GTP-p21Nras). Activating point-mutations in hot spot codons 12, 13 and 61 of N-ras gene result in protein products that inhibit GTP hydrolysis constitutively, leading to a prolonged growth-promoting event (2). There exists accumulating evidence for the involvement of N-ras in some types of brain tumors, such as glioblastomas, gliomas and neuroblastomas (2-4), in addition to other cancers such as leukemia, thyroid tumors, melanoma and colorectal carcinoma (5-8).
The rate of GTP-p21Nras hydrolysis is increased by GTP-activating proteins, the most important of which in cells of neuroectodermal origin is neurofibromin (nf1-GAP) that antagonizes the Ras function (9, 10). It is encoded by the tumor suppressor gene nf1 (chromosome 17q11.2), mutations in which cause neurofibromatosis type 1 (NF1), an autosomal dominant disorder characterized by predisposition to uncontrolled cell proliferation and tumors of the nervous system, including gliomas, neurofibrosarcomas and meningiomas (11-14). In the literature, somatic inactivating mutations or deletions in the nf1 gene have been detected in about 23% of patients with glioblastomas (14, 15).
Stimulation of the GTP-activating proteins, and therefore of the GTP-Ras pathway, is also effected by activated growth factor receptors with tyrosine kinase activity, such as human epidermal growth factors 1 (her-1/EGFR) and 2 (her-2/erbB-2/neu/ngl). While alterations in the her-1 gene have been frequently reported in brain tumors (16), less clear remains the role of her-2 gene in brain tumorigenesis (locus 17q21.2), although it was first characterized in rat neuro/glioblastoma cell lines (14). There exist only few reports that overexpressed or activated protein her-2 may be involved in the development of brain tumors (14, 17).
In an effort to contribute to the understanding of the molecular pathology of brain oncogenesis and the involvement of the her-2/N-ras/nf1 pathway, we studied 35 biopsies of various types of brain neoplasias for the most common mutations in these three tumor-related genes.
Materials and Methods
Brain biopsies were collected from 35 patients without NF1 who underwent craniotomy for brain tumor surgery at the Evangelismos Hospital, Athens, Greece. The patients included 18 males and 17 females and their ages at the first diagnosis ranged from 26 to 77 years (median 63 years). Six of them (17%), aged from 58 to 77 years (median 67 years), had a first-degree relative affected with a specific form of cancer other than a brain tumor. The specimens included biopsies of four general categories of pathological diagnosis: 13 neuroglial tumors (4 astrocytomas, 2 oligodendrogliomas, 7 glioblastomas multiforme), 14 meningiomas, 3 other nervous system tumors (2 schwannomas, 1 cranio-pharyngioma) and 5 metastatic tumors (4 lung adenocarcinomas and 1 chronic myelogenous leukemia). The metastatic tumors were analyzed as controls. Each biopsy sample was given a code number.
The mutation screening was conducted blindly. Total DNA was isolated from brain biopsies with the use of a NaCl extraction method. DNA samples from all biopsies were screened for most common mutations in the 3 studied genes. Mutation screening in 3 hot spot codons of the N-ras gene (numbers 12, 13 and 61) was performed with a Bio-Dot apparatus (Bio-Rad, Irvine, CA, USA) using hybridization of dot blotted amplified DNA sequences of N-ras gene with 7-8 biotin-labeled oligonucleotide probes (19mers; Oncogene Inc., New York, NY, USA) per codon according to standard procedure (18), i.e. mutation detecting oligonucleotides in addition to a w.t. detection one (Table I). The results of dot blotting were confirmed by direct DNA sequencing. The screening for the most common mutation of the erbB-2 gene (its amplification in multiple copies) was performed by hybridization of BamHI Southern blots with a biotin-labeled 1.6 kb cDNA probe. The screening for mutations in all exons of the nf1 gene was performed with SSCP analysis of PCR products in glycerol containing 80% native polyacrylamide gels. The candidate mutations were analysed by direct DNA sequencing.
Screened mutations at 3 hot spot codons of N-ras gene. Mutant codons are named after the amino acid they encode for.
Results
The results of the blind mutation screening of the three studied genes in the 35 biopsies of brain tumors are presented in Table II. At least one mutation was detected in 20 biopsies (57%). There was no statistically significant difference in male versus female patients for whom mutations were found (χ2=2.32, p>0.05).
N-ras mutations were detected in 19 biopsies (54%), encompassing all four general categories. More specifically, they were detected in 9/13 neuroglial tumors (69%), in 6/14 meningiomas (43%), in 1/3 other nervous system tumors (33%) and in 3/5 metastatic tumors (60%). Interestingly, more than one N-ras mutations were detected in 9 of the 19 biopsies (47%): two mutant codons in 7/19 (37%) and three mutant codons in 2/19 (10%). Codon 12 mutations were found in 8 biopsies (of all categories), codon 13 mutations in 7 biopsies (4 of them were gliomas) and codon 61 mutations in 14 biopsies (7 of them gliomas and the rest of other categories).
The incidence of the most common screened mutations in the other two studied genes was much lower. Amplified gene copies of her-2 were observed in only two cases (6%). They were detected in 1 of 13 neuroglial tumors (8%) and in 1 of 14 meningiomas (7%).
In a unique case (3%), a novel mutation was detected in exon 31 of nf1 gene in a glioblastoma biopsy. In codon 1948, a C5842T transition resulted in a change of a Gln codon (CAA) to a stop codon (TAA).
Discussion
Several sequential genetic alterations appear to be required to direct cells toward malignancy, namely the activation of growth promoting oncogenes and the inactivation of growth inhibiting tumor suppressor genes. The major causal events of oncogene activation include either hot spot point mutations (such as in the case of N-ras) or gene amplification (such as in the case of her-2), while inactivation of tumor suppressor genes occurs mainly as a result of loss-of-function mutations (such as in nf1 gene). Cumulative evidence suggests that genetic alterations affecting components of the cell growth regulating signal transduction pathway her-2/Ras/RTK/P13K sum up to 88% in brain tumors such as malignant gliomas (19).
Mutations in her-2, N-ras and nf1 genes detected in brain tumors and reported by codon and encoded amino acid or stop. Fam: family history of cancer; amplif: gene amplification, mut: mutation.
In an effort to elucidate the oncogenic role of key components of the initial part of the her-2/Ras/RTK/P13K pathway, namely the her-2/N-ras/nf1, we screened various brain tumor types for genetic mutations on these genes. Our data support the notion that activated N-ras is a major oncogene in central nervous system malignancies playing the most important role in activation of the her-2/N-ras/nf1 pathway. N-ras gene codon 61 in particular appears to be an important mutation hot spot in both neuroglial tumors and meningiomas (mutations found in 7/9 and 4/6, respectively). The important role of N-ras in brain oncogenesis has been previously reported in neuroblastoma and glioblastoma multiforme (20). In addition, N-ras mutations were detected in more than half of the metastatic brain tumors in this study: two lung adenocarcinomas and one chronic myeloid leukaemia. Activated N-ras has been previously reported to be associated with both types of cancers (6, 21), however it is unclear whether the somatic mutation events occurred prior or after metastasis.
Interestingly, more than one mutations in the N-ras gene were found in 8 of 19 biopsies: 2 mutations in different codons in 7 samples and 3 mutations in 1 biopsy (13Ala, 13Asp/61Lys). It is unclear whether these mutations represent genetic events within the same or different cell populations of the studied tumors. Similar findings have been occasionally reported in the literature (22, 23). It is possible that genomic instability in a parent cell population, induced by mutations in genes important for cell cycle regulation and DNA repair (for example p53 gene), may lead to multiple gene alterations in the daughter cell populations (1).
Hereditary predisposition to oncogenesis may not be excluded in 5 patients with positive family history, although their ages were quite advanced (58-77 years). Four patients with N-ras mutations in their gliomas had a first degree relative with tumors previously associated with N-ras activation: cutaneous melanoma, endometrial carcinoma, colorectal carcinoma and lung carcinoma (7, 8, 21, 24). One of the above mentioned patients with glioblastoma multiforme had an additional mutation in nf1 gene. That was probably a somatic one, since the corresponding patient had a mother with colorectal carcinoma, a tumor not previously been reported to be associated with nf1 genetic alteration. The fifth patient with positive family history had no mutation in the N-ras gene but instead amplified gene her-2 copies in her meningioma and his father with salivary gland adenocarcinoma, a tumor which amplification of her-2 was previously reported (25).
Our findings indicate that activated N-ras is a major oncogene in brain tumors, while the roles of her-2 and nf1 genes seem to be less important, although their protein products closely interact with the N-ras protein in the same cellular pathway. In accordance to our observations, low incidence of overexpressed her-2 protein and amplification of its gene have been previously observed in gliomas, while moderate incidence has been reported in meningiomas (14, 26). The low incidence of nf1 mutations in brain tumor biopsies is in accordance to previous reports in gliomas, meningiomas and primitive neuroectodermal tumors of non-NF1 patients (2, 4), but at the same time in contrast to a large scale multidimensional study which has reported nf1 somatic inactivation mutations or deletions at a moderate rate (14).
Somatic mutations in oncogenes such as N-ras are the most common activating lesions found in human cancers. These mutations are frequently associated with poor response to standard cancer therapies. Despite the genetic complexity and the pathologic heterogeneity of brain tumors, major components of her-2/Ras/RTK/P13K pathway may be appropriately targeted by molecular and immunological therapies. In this context, the frequently observed hyperactivation of N-ras could represent an opportunity to develop new targeted therapies in order to increase treatment effectiveness and decrease the toxicity at the same time (27-29). Combination therapies aiming at multiple molecular targets of the activated signaling pathway including her-1, N-ras and P13K might be more effective than those of single target.
- Received May 23, 2016.
- Revision received July 7, 2016.
- Accepted July 11, 2016.
- Copyright© 2016 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.