Discovery and characterization of new microsatellite loci in Dipteryx alata Vogel (Fabaceae) using next-generation sequencing data.

The use of next-generation sequencing (NGS) technologies provides a great volume of genome sequence data even for non-model species. The development of microsatellite markers using these data is a relatively quick and easy process. Dipteryx alata Vogel (Fabaceae) is an arboreal species from the Cerrado biome and is considered an important plant genetic resource. Here, we report the development of microsatellite markers for D. alata using NGS data. DNA samples from four individuals were sequenced using the Illumina MiSeq platform and high-quality reads were assembled into contigs of the D. alata genome sequence. Microsatellite regions were identified using the IMEX webserver and primer pairs were designed using the Primer3 software. The amplification settings for each locus were optimized. Fluorescent-labeled primers were developed and used to genotype individuals derived from three natural populations of D. alata. Fifty-four microsatellite regions were identified, from which 27 were elected to primer design. Among the amplified loci, 11 were polymorphic, with the number of alleles ranging from 2 to 10. The expected heterozygosity under Hardy-Weinberg Equilibrium (HWE) per locus varied from 0.191 to 0.807. Genotype and allele frequencies for all loci agreed with those expected under HWE and linkage disequilibrium was not significant for all pairs of loci. The probabilities of exclusion of paternity and of combined identity were equal to 0.993 and 5.65 x 10-8, respectively. The markers developed in this study are useful to several types of population genetic studies with D. alata and, eventually, for closely related species.

The retinoblastoma gene is involved in malignant progression of astrocytomas.

Loss of chromosome 13q occurs in up to 50% of human astrocytomas, suggesting the presence of an astrocytoma tumor suppressor gene on that chromosome. To determine whether the retinoblastoma susceptibility gene (Rb) on 13q14 contributes to the formation of astrocytomas, we examined 85 tumors for loss of heterozygosity (LOH) at the intragenic Rb 1.20 locus. LOH was detected in 16 of 54 informative high-grade astrocytomas (30%), but was not detected in 12 low-grade gliomas. Deletion mapping with flanking markers on 13q revealed that the Rb 1.20 region was preferentially targeted by the deletions. Tumors with LOH at Rb 1.20 were examined for mutations in the remaining Rb allele using single-strand conformational polymorphism (SSCP) analysis and direct DNA sequencing. Mutations were detected in exon 8 (1 tumor), exon 24 (2 tumors), and intron 24 (1 tumor). Rb protein expression, as assessed by immunohistochemistry, was altered in 3 of 9 cases with LOH and in 1 tumor without LOH. Our results demonstrate that Rb inactivation contributes to the formation of high-grade astrocytomas, and therefore implicate a second, known tumor suppressor gene in astrocytoma tumorigenesis.

MTS1/CDKN2 gene mutations are rare in primary human astrocytomas with allelic loss of chromosome 9p.

Human astrocytomas frequently have allelic losses of chromosome 9p, suggesting the presence of a 9p astrocytoma tumor suppressor gene. The MTS1 (or CDKN2) gene on chromosome 9p encodes a cell-cycle regulator and is deleted in approximately 80% of astrocytoma cell lines. To determine whether MTS1 is the tumor suppressor gene involved in human astrocytoma formation in vivo, we have analyzed chromosome 9p allelic loss and the MTS1 gene in 30 primary astrocytomas. Deletion mapping demonstrated 15 cases with allelic loss of chromosome 9p, with all losses either flanking or involving the MTS1 gene. Direct analysis of the MTS1 gene, however, revealed only a single missense mutation in a high-grade tumor that had lost the second allele. The low frequency of MTS1 mutations in primary astrocytomas with allelic 9p loss suggests that MTS1 may be more important for in vitro than in vivo astrocytoma growth, and that another 9p tumor suppressor gene may be involved in astrocytoma formation in vivo. Analysis of the MTS1 gene also demonstrated two intragenic polymorphisms, one in exon 2 and one in the 3' untranslated region, that can be used to assay allelic loss directly at MTS1.

The putative glioma tumor suppressor gene on chromosome 19q maps between APOC2 and HRC.

The frequent allelic loss of chromosome 19q in human gliomas suggests that 19q harbors a tumor suppressor gene that is integral to glioma tumorigenesis. Our initial deletion mapping of this gene localized the common region of deletion to the distal long arm, 19q13.2-13.4. To bracket the putative tumor suppressor gene further, we have studied this region in 55 gliomas, using loss of heterozygosity studies for 11 well mapped, highly informative microsatellite polymorphisms that cover this area: D19S178; BCL3; APOC2; ERCC1; DM; D19S112; HRC; D19S246; KLK; D19S180; and D19S254 (from centromeric to telomeric). Twenty astrocytic, oligodendroglial, and mixed gliomas had deletions affecting this region. Of nine partial deletions, two cases maintained heterozygosity at APOC2 while showing allelic loss at the more telomeric markers, ERCC1 and DM, while five cases maintained heterozygosity at HRC but lost the more centromeric markers, D19S112 and DM. Nine cases lost the entire D19S178 to D19S254 region. Three astrocytic gliomas, including one with an interstitial deletion, had terminal deletions of 19q13.4. The minimum area of overlap shared by the interstitial deletions is between APOC2 and HRC, including ERCC1, DM, and D19S112. These findings suggest that the glioma tumor suppressor gene maps to an approximately 8-cM/5-megabase region on 19q13.2-13.3 between the proximal marker APOC2 and the distal marker HRC. Among the DNA repair/DNA metabolism genes on chromosome 19q, ERCC1, LIG1, and perhaps ERCC2 are within the common area of deletion; XRCC1 is centromeric and is therefore excluded as a candidate.

Analysis of the neurofibromatosis 2 gene in human ependymomas and astrocytomas.

Ependymomas and astrocytomas commonly have allelic losses of chromosome 22q, which suggests the presence of a glioma tumor suppressor gene on 22q. A candidate tumor suppressor gene on 22q is the neurofibromatosis 2 (NF2) gene since NF2 patients have an increased susceptibility to ependymomas and astrocytomas. Using single strand conformation polymorphism analysis and direct DNA sequencing, we screened 8 ependymomas and 30 fibrillary astrocytomas from non-NF2 patients for mutations in the coding sequence and portions of the 3' untranslated region of the NF2 gene. Only one mutation was detected, a single base deletion in NF2 exon 7 from a spinal ependymoma, which had also lost the wild-type allele. These results suggest that the NF2 gene may be important in the formation of some ependymomas but the NF2 gene is probably not the critical chromosome 22q tumor suppressor gene involved in astrocytoma tumorigenesis.

Molecular genetics of pediatric brain stem gliomas. Application of PCR techniques to small and archival brain tumor specimens.

Brain stem gliomas are pediatric astrocytomas that histologically resemble adult supratentorial astrocytomas such as glioblastomas multiforme (GBM). Our molecular genetic studies have suggested that adult GBM can be divided into two genetic subsets: tumors with p53 tumor suppressor gene mutations and chromosome 17p loss that occur more commonly in younger patients; and tumors with epidermal growth factor receptor (EGFR) gene amplification that occur more commonly in older patients. Brain stem gliomas have not been studied since biopsies of these tumors are rare and extremely small. We investigated the molecular genetic composition of seven brain stem glioblastomas (two small biopsies, five autopsies) using polymerase chain reaction (PCR) assays for chromosomal loss, gene mutation and gene amplification. Four cases lost portions of chromosome 17p that included the p53 gene. These four cases and one additional case had mutations in the p53 gene. None of the cases showed amplification of the EGFR gene. Allelic losses of the long arm of chromosome 10 were noted in four cases. These results suggest similarities between pediatric brain stem glioblastomas and those GBM that occur in younger adult patients, and confirm the utility of PCR-based means of studying small and archival brain tumor specimens.