Comprehensive analysis of loss of heterozygosity events in glioblastoma using the 100K SNP mapping arrays and comparison with copy number abnormalities defined by BAC array comparative genomic hybridization.

We have undertaken an extensive high-resolution analysis of loss of heterozygosity (LOH) in 30 high grade gliomas using the Affymetrix 100K SNP mapping array. Only 70% of LOH events were accompanied by a copy number loss (CNA(loss)), and of the other 30%, the distal region of 17p preferentially showed copy number neutral (CNN)-associated LOH. Combined analysis of CNA(loss) and LOH using MergeLevels analysis software predicts whether the observed losses occurred on a diploid or tetraploid background. In a side-by-side comparison between SNP and bacterial artificial chromosome (BAC) arrays, the overall identification of CNAs was similar on both platforms. The resolution provided by the SNP arrays, however, allowed a considerably more accurate definition of breakpoints as well as defining small events within the cancer genomes, which could not be detected on BAC arrays. CNN LOH was only detected by the SNP arrays, as was ploidy prediction. From our analysis, therefore, it is clear that simultaneously defining CNAs and CNN-LOH using the SNP platform provides a higher resolution and more complete analysis of the genetic events that have occurred within tumor cells. Our extensive analysis of SNP array data has also allowed an objective assessment of threshold LOH scores that can accurately predict LOH. This capability has important implications for interpretation of LOH events since they have consistently been used to localize potential tumor suppressor genes within the cancer genome.

Identification of inactivating mutations in the JAK1, SYNJ2, and CLPTM1 genes in prostate cancer cells using inhibition of nonsense-mediated decay and microarray analysis.

We have developed a simple analytical method that increases the efficiency of identifying mutant genes in cell lines after the inhibition of nonsense-mediated decay (NMD). The approach assumes that the spectra of mutant genes differ between cell lines of the same tumor origin. Thus, by analyzing more than one cell line in parallel and taking into account not only changes in mRNA levels after the inhibition of NMD, but also comparing mRNA levels between cell lines before the inhibition of NMD, the vast majority of false positives were eliminated from the analysis. In this study, we used Affymetrix oligonucleotide arrays to compare mRNA profiles of two prostate cancer cell lines, PC3 and LNCaP, before and after emetine treatment. As a result of our modified approach, from the 14,500 genes present on the array, 7 were identified as candidates from LNCaP cells and 1 was identified from PC3 cells. Sequence analysis of five of these candidate genes identified gene-inactivating mutations in four of them. Homozygous mutations were found in the synaptojanin 2 (SYNJ2) and the cleft lip and palate CLPTM1 genes. Two different heterozygous mutations in the Janus kinase 1 (JAK1) gene result in complete loss of the protein in several different prostate cancer cell lines.

Molecular characterization of the t(3;9) associated with immortalization in the MCF10A cell line.

The t(3;9)(p14;p21) in the MCF10A human mammary gland epithelial cell line was the single cytogenetic event that accompanied the transition from primary culture to immortalized cell line, suggesting that it is related to the development of the immortalization phenotype. To study the molecular consequences of the breakpoints in this rearrangement, we used a combination of fluorescence in situ hybridization (FISH) and array comparative genomic hybridization (CGH). The 3p14 translocation breakpoint occurs within BAC RP11-795e22, which accommodates only the TAFA1 gene, a novel cysteine-rich secreted protein thought to be involved in cytokine signaling. TAFA1 is expressed in normal breast tissue, not in MCF10A, and shows differential expression in a range of breast cancer cell lines. The 9p translocation breakpoint results in a deletion of approximately 4 megabases on the derivative chromosome 9, which includes the CDKN2A (p16) gene. Array CGH and FISH analysis demonstrated that BAC 149i22, which contains the CDKN2A/B genes, is also deleted specifically on the apparently normal copy of chromosome 9, making MCF10A null for the p16/p15 genes. The exact extent of gains and losses of chromosome regions resulting from rearrangements involving chromosomes 1q, 5q, and 8q have also been characterized using the BAC arrays.

Overlay analysis of the oligonucleotide array gene expression profiles and copy number abnormalities as determined by array comparative genomic hybridization in medulloblastomas.

Combined analysis of gene expression array data and array-based comparative genomic hybridization data have been used in a series of 26 pediatric brain tumors to define up- and downregulated genes that coincide with losses, gains, and amplifications involving specific chromosome regions. Frequent losses were defined in chromosome arms 3q, 6q, 8p, 10q, 16q, 17p, and gains were identified in chromosome 7, and chromosome arms 9p and 17q. Amplification of a 2p region was seen in only one tumor, which corresponded to increased expression of the MYCN and DDX1 genes. To facilitate the analysis of the two data sets, we have developed a custom overlay tool that defines genes that are underexpressed in regions of deletions and overexpressed in regions of gain, across the genome and specifically within regions showing recurrent involvement in medulloblastomas.

Overlay tool for aCGHViewer: an analysis module built for aCGHViewer used to perform comparisons of data derived from different microarray platforms.

The Overlay Tool has been developed to combine high throughput data derived from various microarray platforms. This tool analyzes high-resolution correlations between gene expression changes and either copy number abnormalities (CNAs) or loss of heterozygosity events detected using array comparative genomic hybridization (aCGH). Using an overlay analysis which is designed to be performed using data from multiple microarray platforms on a single biological sample, the Overlay Tool identifies potentially important genes whose expression profiles are changed as a result of losses, gains and amplifications in the cancer genome. In addition, the Overlay Tool will incorporate loss of heterozygosity (LOH) probability data into this overlay procedure. To facilitate this analysis, we developed an application which computationally combines two or more high throughput datasets (e.g. aCGH/expression) into a single categorized dataset for visualization and interrogation using a gene-centric approach. As such, data from virtually any microarray platform can be incorporated without the need to remap entire datasets individually. The resultant categorized (overlay) data set can be conveniently viewed using our in-house visualization tool, aCGHViewer (Shankar et al. 2006), which serves as a conduit to public databases such as UCSC and NCBI, to rapidly investigate genes of interest.

Defining the expression pattern of the LGI1 gene in BAC transgenic mice.

The LGI1 gene has been implicated in the development of epilepsy and the invasion phenotype of glial cells. Controversy over the specific tissue expression pattern of this gene has stemmed from conflicting reports generated using immunohistochemistry and the polymerase chain reaction. LGI1 is one of a four-member family of secreted proteins with high homology and here we demonstrate, using GFP-tagged constructs from the four LGI1family members, that commonly used antibodies against LGI1 cross-react with different family members. With the uncertainty surrounding the use of commercially available antibodies to truly establish the expression pattern of LGI1, we generated transgenic mice carrying the LGI1-containing BAC, RP23-127G7, which had been modified to express the GFP reporter gene under the control of the endogenous regulatory elements required for LGI1 expression. Three founder mice were generated, and immunohistochemistry was used to determine the tissue-specific pattern of expression. In the brain, distinct regions of glial and neuronal cell expression were identified, as well as the choriod plexus, which is largely pia-derived. In addition, strong expression levels were identified in glandular regions of the prostate, individual tubules in the kidney, sympathetic ganglia in the kidney, sebaceous glands in the skin, the islets of Langerhans, the endometrium, and the ovary and testes. All other major organs analyzed were negative. The pattern of reporter gene expression was identical in three individual founder mice, arguing against a position effect altering expression profile due to the integration site of the BAC.