Paul S. Meltzer

Classification and diagnostic prediction of cancers using gene expression profiling and artificial neural networks.

The purpose of this study was to develop a method of classifying cancers to specific diagnostic categories based on their gene expression signatures using artificial neural networks (ANNs). We trained the ANNs using the small, round blue-cell tumors (SRBCTs) as a model. These cancers belong to four distinct diagnostic categories and often present diagnostic dilemmas in clinical practice. The ANNs correctly classified all samples and identified the genes most relevant to the classification. Expression of several of these genes has been reported in SRBCTs, but most have not been associated with these cancers. To test the ability of the trained ANN models to recognize SRBCTs, we analyzed additional blinded samples that were not previously used for the training procedure, and correctly classified them in all cases. This study demonstrates the potential applications of these methods for tumor diagnosis and the identification of candidate targets for therapy.

Mutations of a mutS homolog in hereditary nonpolyposis colorectal cancer

Recent studies have shown that a locus responsible for hereditary nonpolyposis colorectal cancer (HNPCC) is on chromosome 2p and that tumors developing in these patients contain alterations in microsatellite sequences (RER+ phenotype). We have used chromosome microdissection to obtain highly polymorphic markers from chromosome 2p16. These and other markers were ordered in a panel of somatic cell hybrids and used to define a 0.8 Mb interval containing the HNPCC locus. Candidate genes were then mapped, and one was found to lie within the 0.8 Mb interval. We identified this candidate by virtue of its homology to mutS mismatch repair genes. cDNA clones were obtained and the sequence used to detect germline mutations, including those producing termination codons, in HNPCC kindreds. Somatic as well as germline mutations of the gene were identified in RER+ tumor cells. This mutS homolog is therefore likely to be responsible for HNPCC.

Expression profiling using cDNA microarrays

cDNA microarrays are capable of profiling gene expression patterns of tens of thousands of genes in a single experiment. DNA targets, in the form of 3´ expressed sequence tags (ESTs), are arrayed onto glass slides (or membranes) and probed with fluorescent– or radioactively–labelled cDNAs. Here, we review technical aspects of cDNA microarrays, including the general principles, fabrication of the arrays, target labelling, image analysis and data extraction, management and mining.

Gene-expression profiles in hereditary breast cancer.

BACKGROUND Many cases of hereditary breast cancer are due to mutations in either the BRCA1 or the BRCA2 gene. The histopathological changes in these cancers are often characteristic of the mutant gene. We hypothesized that the genes expressed by these two types of tumors are also distinctive, perhaps allowing us to identify cases of hereditary breast cancer on the basis of gene-expression profiles. METHODS RNA from samples of primary tumor from seven carriers of the BRCA1 mutation, seven carriers of the BRCA2 mutation, and seven patients with sporadic cases of breast cancer was compared with a microarray of 6512 complementary DNA clones of 5361 genes. Statistical analyses were used to identify a set of genes that could distinguish the BRCA1 genotype from the BRCA2 genotype. RESULTS Permutation analysis of multivariate classification functions established that the gene-expression profiles of tumors with BRCA1 mutations, tumors with BRCA2 mutations, and sporadic tumors differed significantly from each other. An analysis of variance between the levels of gene expression and the genotype of the samples identified 176 genes that were differentially expressed in tumors with BRCA1 mutations and tumors with BRCA2 mutations. Given the known properties of some of the genes in this panel, our findings indicate that there are functional differences between breast tumors with BRCA1 mutations and those with BRCA2 mutations. CONCLUSIONS Significantly different groups of genes are expressed by breast cancers with BRCA1 mutations and breast cancers with BRCA2 mutations. Our results suggest that a heritable mutation influences the gene-expression profile of the cancer.

Rare Structural Variants Disrupt Multiple Genes in Neurodevelopmental Pathways in Schizophrenia

Schizophrenia is a devastating neurodevelopmental disorder whose genetic influences remain elusive. We hypothesize that individually rare structural variants contribute to the illness. Microdeletions and microduplications >100 kilobases were identified by microarray comparative genomic hybridization of genomic DNA from 150 individuals with schizophrenia and 268 ancestry-matched controls. All variants were validated by high-resolution platforms. Novel deletions and duplications of genes were present in 5% of controls versus 15% of cases and 20% of young-onset cases, both highly significant differences. The association was independently replicated in patients with childhood-onset schizophrenia as compared with their parents. Mutations in cases disrupted genes disproportionately from signaling networks controlling neurodevelopment, including neuregulin and glutamate pathways. These results suggest that multiple, individually rare mutations altering genes in neurodevelopmental pathways contribute to schizophrenia.

High frequency of BRAF mutations in nevi.

To evaluate the timing of mutations in BRAF (v-raf murine sarcoma viral oncogene homolog B1) during melanocytic neoplasia, we carried out mutation analysis on microdissected melanoma and nevi samples. We observed mutations resulting in the V599E amino-acid substitution in 41 of 60 (68%) melanoma metastases, 4 of 5 (80%) primary melanomas and, unexpectedly, in 63 of 77 (82%) nevi. These data suggest that mutational activation of the RAS/RAF/MAPK pathway in nevi is a critical step in the initiation of melanocytic neoplasia but alone is insufficient for melanoma tumorigenesis.

Vascular channel formation by human melanoma cells in vivo and in vitro: vasculogenic mimicry.

Tissue sections from aggressive human intraocular (uveal) and metastatic cutaneous melanomas generally lack evidence of significant necrosis and contain patterned networks of interconnected loops of extracellular matrix. The matrix that forms these loops or networks may be solid or hollow. Red blood cells have been detected within the hollow channel components of this patterned matrix histologically, and these vascular channel networks have been detected in human tumors angiographically. Endothelial cells were not identified within these matrix-embedded channels by light microscopy, by transmission electron microscopy, or by using an immunohistochemical panel of endothelial cell markers (Factor VIII-related antigen, Ulex, CD31, CD34, and KDR[Flk-1]). Highly invasive primary and metastatic human melanoma cells formed patterned solid and hollow matrix channels (seen in tissue sections of aggressive primary and metastatic human melanomas) in three-dimensional cultures containing Matrigel or dilute Type I collagen, without endothelial cells or fibroblasts. These tumor cell-generated patterned channels conducted dye, highlighting looping patterns visualized angiographically in human tumors. Neither normal melanocytes nor poorly invasive melanoma cells generated these patterned channels in vitro under identical culture conditions, even after the addition of conditioned medium from metastatic pattern-forming melanoma cells, soluble growth factors, or regimes of hypoxia. Highly invasive and metastatic human melanoma cells, but not poorly invasive melanoma cells, contracted and remodeled floating hydrated gels, providing a biomechanical explanation for the generation of microvessels in vitro. cDNA microarray analysis of highly invasive versus poorly invasive melanoma tumor cells confirmed a genetic reversion to a pluripotent embryonic-like genotype in the highly aggressive melanoma cells. These observations strongly suggest that aggressive melanoma cells may generate vascular channels that facilitate tumor perfusion independent of tumor angiogenesis.

Mutations in the human Jagged1 gene are responsible for Alagille syndrome

Alagille syndrome (AGS) is an autosomal-dominant disorder characterized by intrahepatic cholestasis and abnormalities of heart, eye and vertebrae, as well as a characteristic facial appearance. Identification of rare AGS patients with cytogenetic deletions has allowed mapping of the gene to 20p12. We have generated a cloned contig of the critical region and used fluorescent in situ hybridization on cells from patients with submicroscopic deletions to narrow the candidate region to only 250 kb. Within this region we identified JAG1, the human homologue of rat Jagged1, which encodes a ligand for the Notch receptor. Cell-cell Jagged/Notch interactions are known to be critical for determination of cell fates in early development, making this an attractive candidate gene for a developmental disorder in humans. Determining the complete exon–intron structure of JAG1 allowed detailed mutational analysis of DMA samples from non-deletion AGS patients, revealing three frame-shift mutations, two splice donor mutations and one mutation abolishing RNA expression from the altered allele. We conclude that AGS is caused by haploinsufficiency of JAG1.

High-Resolution Mapping and Characterization of Open Chromatin across the Genome

Mapping DNase I hypersensitive (HS) sites is an accurate method of identifying the location of genetic regulatory elements, including promoters, enhancers, silencers, insulators, and locus control regions. We employed high-throughput sequencing and whole-genome tiled array strategies to identify DNase I HS sites within human primary CD4+ T cells. Combining these two technologies, we have created a comprehensive and accurate genome-wide open chromatin map. Surprisingly, only 16%-21% of the identified 94,925 DNase I HS sites are found in promoters or first exons of known genes, but nearly half of the most open sites are in these regions. In conjunction with expression, motif, and chromatin immunoprecipitation data, we find evidence of cell-type-specific characteristics, including the ability to identify transcription start sites and locations of different chromatin marks utilized in these cells. In addition, and unexpectedly, our analyses have uncovered detailed features of nucleosome structure.

MicroRNA Expression, Survival, and Response to Interferon in Liver Cancer

BACKGROUND Hepatocellular carcinoma is a common and aggressive cancer that occurs mainly in men. We examined microRNA expression patterns, survival, and response to interferon alfa in both men and women with the disease. METHODS We analyzed three independent cohorts that included a total of 455 patients with hepatocellular carcinoma who had undergone radical tumor resection between 1999 and 2003. MicroRNA-expression profiling was performed in a cohort of 241 patients with hepatocellular carcinoma to identify tumor-related microRNAs and determine their association with survival in men and women. In addition, to validate our findings, we used quantitative reverse-transcriptase-polymerase-chain-reaction assays to measure microRNAs and assess their association with survival and response to therapy with interferon alfa in 214 patients from two independent, prospective, randomized, controlled trials of adjuvant interferon therapy. RESULTS In patients with hepatocellular carcinoma, the expression of miR-26a and miR-26b in nontumor liver tissue was higher in women than in men. Tumors had reduced levels of miR-26 expression, as compared with paired noncancerous tissues, which indicated that the level of miR-26 expression was also associated with hepatocellular carcinoma. Moreover, tumors with reduced miR-26 expression had a distinct transcriptomic pattern, and analyses of gene networks revealed that activation of signaling pathways between nuclear factor kappaB and interleukin-6 might play a role in tumor development. Patients whose tumors had low miR-26 expression had shorter overall survival but a better response to interferon therapy than did patients whose tumors had high expression of the microRNA. CONCLUSIONS The expression patterns of microRNAs in liver tissue differ between men and women with hepatocellular carcinoma. The miR-26 expression status of such patients is associated with survival and response to adjuvant therapy with interferon alfa.