Localization of multiple melanoma tumor-suppressor genes on chromosome 11 by use of homozygosity mapping-of-deletions analysis

Integrative genetic analysis suggests that skin color modifies the genetic architecture of melanoma

Melanoma is the deadliest form of skin cancer and presents a significant health care burden in many countries. In addition to ultraviolet radiation in sunlight, the main causal factor for melanoma, genetic factors also play an important role in melanoma susceptibility. Although genome-wide association studies have identified many single nucleotide polymorphisms associated with melanoma, little is known about the proportion of disease risk attributable to these loci and their distribution throughout the genome. Here, we investigated the genetic architecture of melanoma in 1,888 cases and 990 controls of European non-Hispanic ancestry. We estimated the overall narrow-sense heritability of melanoma to be 0.18 (P < 0.03), indicating that genetics contributes significantly to the risk of sporadically-occurring melanoma. We then demonstrated that only a small proportion of this risk is attributable to known risk variants, suggesting that much remains unknown of the role of genetics in melanoma. To investigate further the genetic architecture of melanoma, we partitioned the heritability by chromosome, minor allele frequency, and functional annotations. We showed that common genetic variation contributes significantly to melanoma risk, with a risk model defined by a handful of genomic regions rather than many risk loci distributed throughout the genome. We also demonstrated that variants affecting gene expression in skin account for a significant proportion of the heritability, and are enriched among melanoma risk loci. Finally, by incorporating skin color into our analyses, we observed both a shift in significance for melanoma-associated loci and an enrichment of expression quantitative trait loci among melanoma susceptibility variants. These findings suggest that skin color may be an important modifier of melanoma risk. We speculate that incorporating skin color and other non-genetic factors into genetic studies may allow for an improved understanding of melanoma susceptibility and guide future investigations to identify melanoma risk genes.

Copy number variation in archival melanoma biopsies versus benign melanocytic lesions

BACKGROUND

Skin melanocytes can give rise to benign and malignant neoplasms. Discrimination of an early melanoma from an unusual/atypical benign nevus can represent a significant challenge. However, previous studies have shown that in contrast to benign nevi, melanoma demonstrates pervasive chromosomal aberrations.

OBJECTIVE

This substantial difference between melanoma and benign nevi can be exploited to discriminate between melanoma and benign nevi.

METHODS

Array-comparative genomic hybridization (aCGH) is an approach that can be used on DNA extracted from formalin-fixed paraffin-embedded (FFPE) tissues to assess the entire genome for the presence of changes in DNA copy number. In this study, high resolution, genome-wide single-nucleotide polymorphism (SNP) arrays were utilized to perform comprehensive and detailed analyses of recurrent copy number aberrations in 41 melanoma samples in comparison with 21 benign nevi.

RESULTS

We found statistically significant copy number gains and losses within melanoma samples. Some of the identified aberrations are previously implicated in melanoma. Moreover, novel regions of copy number alterations were identified, revealing new candidate genes potentially involved in melanoma pathogenesis.

CONCLUSIONS

Taken together, these findings can help improve melanoma diagnosis and introduce novel melanoma therapeutic targets.

LOHAS: loss-of-heterozygosity analysis suite

Detection of loss of heterozygosity (LOH) plays an important role in genetic, genomic and cancer research. We develop computational methods to estimate the proportion of homozygous SNP calls, identify samples with structural alterations and/or unusual genotypic patterns, cluster samples with close LOH structures and map the genomic segments bearing LOH by analyzing data of genome-wide SNP arrays or customized SNP arrays. In addition to cancer genetics/genomics, we also apply the methods to study long contiguous stretches of homozygosity (LCSH) in general populations. The LCSH analysis aids in the identification of samples with complex LCSH patterns indicative of nonrandom mating and/or meiotic recombination cold spots, separation of samples with different genetic backgrounds and sex, and mapping of regions of LCSH. Affymetrix Human Mapping 500K Set SNP data from an acute lymphoblastic leukemia study containing 304 cancer patients and 50 normal controls and from the HapMap Project containing 30 African trios, 30 Caucasian trios and 90 independent Asian samples were analyzed. We identified common gene regions of LOH, e.g., ETV6 and CDKN1B, and identified frequent regions of LCSH, e.g., the region that encompasses the centromeric gene desert region of chromosome 16. Unsupervised analysis separated cancer subtypes and ethnic subpopulations by patterns of LOH/LCSH. Simulation studies considering LOH width, effect size and heterozygous interference fraction were performed, and the results show that the proposed LOH association test has good test power and controls type 1 error well. The developed algorithms are packaged into LOHAS written in R and R GUI.

A genome-wide homozygosity association study identifies runs of homozygosity associated with rheumatoid arthritis in the human major histocompatibility complex

Rheumatoid arthritis (RA) is a chronic inflammatory disorder with a polygenic mode of inheritance. This study examined the hypothesis that runs of homozygosity (ROHs) play a recessive-acting role in the underlying RA genetic mechanism and identified RA-associated ROHs. Ours is the first genome-wide homozygosity association study for RA and characterized the ROH patterns associated with RA in the genomes of 2,000 RA patients and 3,000 normal controls of the Wellcome Trust Case Control Consortium. Genome scans consistently pinpointed two regions within the human major histocompatibility complex region containing RA-associated ROHs. The first region is from 32,451,664 bp to 32,846,093 bp (−log10(p)>22.6591). RA-susceptibility genes, such as HLA-DRB1, are contained in this region. The second region ranges from 32,933,485 bp to 33,585,118 bp (−log10(p)>8.3644) and contains other HLA-DPA1 and HLA-DPB1 genes. These two regions are physically close but are located in different blocks of linkage disequilibrium, and ∼40% of the RA patients' genomes carry these ROHs in the two regions. By analyzing homozygote intensities, an ROH that is anchored by the single nucleotide polymorphism rs2027852 and flanked by HLA-DRB6 and HLA-DRB1 was found associated with increased risk for RA. The presence of this risky ROH provides a 62% accuracy to predict RA disease status. An independent genomic dataset from 868 RA patients and 1,194 control subjects of the North American Rheumatoid Arthritis Consortium successfully validated the results obtained using the Wellcome Trust Case Control Consortium data. In conclusion, this genome-wide homozygosity association study provides an alternative to allelic association mapping for the identification of recessive variants responsible for RA. The identified RA-associated ROHs uncover recessive components and missing heritability associated with RA and other autoimmune diseases.

Chromosome 3p microsatellite allelotyping in neuroblastoma: a report on the technical hurdles

Pinpointing critical regions of recurrent loss may help localize tumor suppressor genes. To determine the regions of loss on chromosome 3p in neuroblastoma, we performed loss of heterozygosity analysis using 16 microsatellite markers in a series of 65 primary tumors and 29 neuroblastoma cell lines. In this study, we report the results and discuss the technical hurdles that we encountered during data generation and interpretation that are of relevance for current studies or tests employing microsatellites. To provide functional support for the implication of 3p tumor suppressor genes in this childhood malignancy, we performed a microcell-mediated chromosome 3 transfer in neuroblastoma cells.

Identification of ARHGEF17, DENND2D, FGFR3, and RB1 mutations in melanoma by inhibition of nonsense-mediated mRNA decay

Gene identification by nonsense-mediated mRNA decay inhibition (GINI) has proven to be a strategy for genome-wide discovery of genes containing inactivating mutations in colon and prostate cancers. Here, we present the first study of inhibition of the nonsense-mediated mRNA decay (NMD) pathway in melanoma. We used a combination of emetine and actinomycin D treatment to stabilize mRNAs containing premature termination codons (PTCs), followed by microarray analysis and sequencing to identify novel tumor suppressor genes (TSGs) in a panel of 12 melanoma cell lines. Stringent analysis of the array data was used to select 35 candidate genes for sequencing. Of these, 4 (11%) were found to carry PTCs, including ARHGEF17, DENND2D, FGFR3, and RB1. While RB1 mutations have previously been described in melanoma, the other three genes represent potentially novel melanoma; TSGs. ARHGEF17 showed a G1865A mutation leading to W622X in a cell line derived from a mucosal melanoma; in RB1 a C1411T base change resulting in Q471X was discovered in a cell line derived from an acral melanoma; and the FGFR3 and DENND2D genes had intronic insertions leading to PTCs in cell lines derived from superficially spreading melanomas. We conclude that although the false positive rate is high, most likely due to the lack of DNA mismatch repair gene defects, the GINI protocol is one approach to discover novel TSGs in melanoma.

Detection of copy number alterations in metastatic melanoma by a DNA fluorescence in situ hybridization probe panel and array comparative genomic hybridization: a southwest oncology group study (S9431)

PURPOSE

Gene copy number alteration (CNA) is common in malignant melanoma and is associated with tumor development and progression. The concordance between molecular cytogenetic techniques used to determine CNA has not been evaluated on a large set of loci in malignant melanoma.

EXPERIMENTAL DESIGN

A panel of 16 locus-specific fluorescence in situ hybridization (FISH) probes located on eight chromosomes was used to identify CNA in touch preparations of frozen tissue samples from 19 patients with metastatic melanoma (SWOG-9431). A subset (n = 11) was analyzed using bacterial artificial chromosome (BAC) array comparative genomic hybridization (aCGH) of DNA isolated directly from touch-preparation slides.

RESULTS

By FISH, most samples showed loss near or at WISP3/6p21, CCND3/6q22, and CDKN2A/9p21 (>75% of samples tested). More than one third of CDKN2A/9p21 losses were biallelic. Gains of NEDD9/6p24, MET/7q31, and MYC/8q24 were common (57%, 47%, and 41%, respectively) and CNA events involving 9p21/7p12.3 and MET were frequently coincident, suggesting gain of the whole chromosome 7. Changes were confirmed by aCGH, which also uncovered many discreet regions of change, larger than a single BAC. Overlapping segments observed in >45% of samples included many of the loci analyzed in the FISH study, in addition to other WNT pathway members, and genes associated with TP53 pathways and DNA damage response, repair, and stability.

CONCLUSIONS

This study outlines a set of CNAs at the gene and regional level, using FISH and aCGH, which may provide a benchmark for future studies and may be important in selection of individual therapy for patients with metastatic malignant melanoma.

LOH analysis of free DNA in the plasma of patients with mucosal malignant melanoma in the head and neck

BACKGROUND

We have conducted charged-particle radiotherapy for mucosal malignant melanoma (MMM) in the head and neck, using carbon ion beams. However, even with the use of carbon ion radiotherapy that is characterized by high local tumor control, a significant number of patients develop metastases after therapy. We conducted research on the assumption that, in MMM, the detection of loss of heterozygosity (LOH) from free DNA in the circulating plasma may be of practical use in the diagnosis of recurrence and metastasis.

METHODS

We took blood samples prior to therapy from 17 patients with MMM in the head and neck, and extracted free DNA in plasma. Four types of microsatellite markers were used for LOH detection.

RESULTS

LOH was detected in 1 of 5 patients (20%) for D1S243, 2 of 5 patients (40%) for D6S311, 11 of 17 patients (65%) for D9S161, and 1 of 6 patients (17%) for D19S246.

CONCLUSION

Evidence based on the irradiated tumor volume suggested a tendency for the group of patients found to have LOH in two loci to have a larger mean tumor volume than the patient groups with no detectable LOH or with LOH detectable in only one locus. Of the 17 patients in this study 4 patients had recurrence and/or metastasis, and all 4 of these patients were found to have LOH in at least one or more loci for any region. LOH analysis of free DNA in plasma may be useful for the early diagnosis of MMM.

Extranodal natural killer/t-cell lymphoma, nasal type, in a patient with a constitutional 11q terminal deletion disorder

BACKGROUND

Most cases of constitutional 11q terminal deletion disorder are children. Malignancy is a potential concern, as these children reach adulthood. However, since the majority of patients are young, their risk of developing malignancy in adulthood is essentially unknown.

AIM

To report the first hematologic malignancy [extranodal natural killer (NK)/T-cell lymphoma, nasal type] arising in the trachea of a patient with constitutional 11q terminal deletion disorder. Our patient is the oldest patient reported in the literature. It is of note that this cytogenetic abnormality has not been described as a recurring abnormality in extranodal NK/T-cell lymphoma.

CASE REPORT

A 36-year-old male with congenital psychomotor retardation was transferred to our hospital. Pathologic evaluation was diagnostic of extranodal NK/T-cell lymphoma, nasal type. Staging marrow was negative for lymphoma, but cytogenetic analysis revealed a constitutional deletion of chromosome 11 at band q23 [46,XY,del(11)(q23)(c)]. This abnormality was present in a subsequent bone marrow specimen, along with an acquired abnormality, namely an extra copy of part of the long arm of chromosome 1 translocated to the short arm of chromosome 14.

CONCLUSION

Patients with 11q terminal deletion disorder who reach adulthood may be predisposed to develop neoplasias by virtue of the constitutional deletion.

Genome-wide loss of heterozygosity and copy number analysis in melanoma using high-density single-nucleotide polymorphism arrays

Although a number of genes related to melanoma development have been identified through candidate gene screening approaches, few studies have attempted to conduct such analyses on a genome-wide scale. Here we use Illumina 317K whole-genome single-nucleotide polymorphism arrays to define a comprehensive allelotype of melanoma based on loss of heterozygosity (LOH) and copy number changes in a panel of 76 melanoma cell lines. In keeping with previous reports, we found frequent LOH on chromosome arms 9p (72%), 10p (55%), 10q (55%), 9q (49%), 6q (43%), 11q (43%), and 17p (41%). Tumor suppressor genes (TSGs) can be identified through homozygous deletion (HD). We detected 174 HDs, the most common of which targeted CDKN2A (n = 33). The second highest frequency of HD occurred in PTEN (n = 8), another well known melanoma TSG. HDs were also common for PTPRD (n = 7) and HDAC4 (n = 3), TSGs recently found to be mutated or deleted in other cancer types. Analysis of other HDs and regions of LOH that we have identified might lead to the characterization of further melanoma TSGs. We noted 197 regional amplifications, including some centered on the melanoma oncogenes MITF (n = 9), NRAS (n = 3), BRAF (n = 3), and CCND1 (n = 3). Other amplifications potentially target novel oncogenes important in the development of a subset of melanomas. The numerous focal amplifications and HDs we have documented here are the first step toward identifying a comprehensive catalog of genes involved in melanoma development, some of which may be useful prognostic markers or targets for therapies to treat this disease.

Common fragile site FRA11G and rare fragile site FRA11B at 11q23.3 encompass distinct genomic regions

Fragile sites are specific genomic loci that are particularly prone to chromosomal breakage. Based on their incidence in the human population, they are divided into rare fragile sites occurring in less than 5% of all individuals and common fragile sites being a constitutional feature of the genome of probably all individuals. In this study, cloning of unstable DNA sequences, which have been previously genetically tagged with a marker gene, was the basis for defining the genomic localization of the common fragile site FRA11G at 11q23.3. Mapping of the fragile site with six-color fluorescence in situ hybridization (FISH) resulted in the precise genomic localization of FRA11G to a 4.5 Mb region. The chromosomal subband 11q23.3 harbors both the common fragile site FRA11G and the rare fragile site FRA11B. Here, we show that FRA11G maps 0.8 Mb proximal to the genomic region previously defined to be affected by expression of FRA11B; thus, the common and the rare fragile sites at 11q23.3 encompass distinct genomic regions. The region of FRA11G is known to be involved in somatic and germline recurrent aberrations, and it is conceivable that genetic damage resulting from this fragile site might contribute to clinical phenotypes.

LKB1 mutation in large cell carcinoma of the lung

Germline inactivation of LKB1 is responsible for Peutz-Jeghers syndrome, an autosomal dominant disorder characterized by benign hamartomas of the GI tract and an increased predisposition to certain cancers, including lung. Acquired mutations in LKB1 are rarely observed in most sporadic tumor types except for adenocarcinomas of the lung where up to 50% harbor inactivating mutations. In this study, we focused on LKB1 mutations in lung cancer cell lines originating from large cell carcinomas. We identified a novel 1.5kb interstitial deletion within LKB1 gene in H157 cancer cells. Homozygosity mapping-of-deletion analysis (HOMOD) analysis showed that the deletion is accompanied by LOH of one parental allele, indicating biallelic inactivation of LKB1. This deletion results in an LKB1 transcript lacking exons 2 and 3 and a predicted in-frame deletion of 58 amino acids within the kinase domain of the LKB1 protein. The truncated transcript was expressed at relatively low levels, and the truncated LKB1 protein was virtually undetectable in this cell line. To determine the impact of LKB1 protein truncation on its function, we examined AMPK-alpha, a downstream target of LKB1 kinase activity triggered by low energy stress conditions. Phosphorylation of AMPK-alpha was attenuated in H157 cells treated with 2-deoxyglucose, and could be rescued by expression of an exogenous GFP-LKB1 fusion protein. Therefore, our data suggest that LKB1 function is compromised in H157. Of the four cell lines and six primary tumors of large cell lung carcinoma origin that have been evaluated in this and other studies, LKB1 mutations have been found in three cases. These results suggest that, in addition to adenocarcinomas, acquired loss of function mutations in LKB1 may also be frequently involved in the pathogenesis of large cell lung carcinomas.

Inferring loss-of-heterozygosity from unpaired tumors using high-density oligonucleotide SNP arrays

Loss of heterozygosity (LOH) of chromosomal regions bearing tumor suppressors is a key event in the evolution of epithelial and mesenchymal tumors. Identification of these regions usually relies on genotyping tumor and counterpart normal DNA and noting regions where heterozygous alleles in the normal DNA become homozygous in the tumor. However, paired normal samples for tumors and cell lines are often not available. With the advent of oligonucleotide arrays that simultaneously assay thousands of single-nucleotide polymorphism (SNP) markers, genotyping can now be done at high enough resolution to allow identification of LOH events by the absence of heterozygous loci, without comparison to normal controls. Here we describe a hidden Markov model-based method to identify LOH from unpaired tumor samples, taking into account SNP intermarker distances, SNP-specific heterozygosity rates, and the haplotype structure of the human genome. When we applied the method to data genotyped on 100 K arrays, we correctly identified 99% of SNP markers as either retention or loss. We also correctly identified 81% of the regions of LOH, including 98% of regions greater than 3 megabases. By integrating copy number analysis into the method, we were able to distinguish LOH from allelic imbalance. Application of this method to data from a set of prostate samples without paired normals identified known regions of prevalent LOH. We have developed a method for analyzing high-density oligonucleotide SNP array data to accurately identify of regions of LOH and retention in tumors without the need for paired normal samples.

A key event in the generation of many cancers is loss of heterozygosity (LOH) of chromosomal regions containing tumor suppressor genes, whereby one parent's version of the tumor suppressor is lost. As we develop a better understanding of the molecular mechanisms that generate different cancers, a description of the LOH events underlying these cancers is forming an important part of their classification. Generally, detection of LOH relies on comparison of the tumor's genome to the normal genome of the individual. Unfortunately, for many tumors, including most experimental models of cancer, the normal genome is not available. Therefore, the authors have developed a hidden Markov model-based method that evaluates the probability of LOH at all sites throughout the genome, based on high-resolution genotyping of only the tumor. They were able to achieve high levels of accuracy, specifically by taking into account the haplotype block structure of the genome. Application of this method to a set of 34 prostate cancer samples allowed the authors to identify the locations of the known and suspected tumor suppressor genes that are targeted by LOH.

Combined genome-wide allelotyping and copy number analysis identify frequent genetic losses without copy number reduction in medulloblastoma

Detailed analysis of mechanisms of genetic loss for specific tumor suppressor genes (TSGs; e.g., RB1, APC and NF1) indicates that TSG inactivation can occur by allelic loss of heterozygosity (LOH), without any alteration in DNA copy number. However, the role and prevalence of such events in the pathogenesis of specific malignancies remains to be established on a genome-wide basis. We undertook a detailed molecular assessment of chromosomal defects in a panel of nine cell lines derived from primary medulloblastomas, the most common malignant brain tumors of childhood, by parallel genome-wide assessment of LOH (allelotyping) and copy number aberrations (comparative genomic hybridization and fluorescence in situ hybridization). The majority of genetic losses observed were detected by both copy number and LOH methods, indicating they arise through the physical deletion of chromosomal material. However, a significant proportion of losses (17/42, 40%) represented regions of allelic LOH without any associated copy number reduction; these events involved both whole chromosomes (10/17) and sub-chromosomal regions (7/17). Using this approach, we identified medulloblastoma-characteristic alterations, e.g., isochromosome for 17q, MYC amplification and losses on chromosomes 10, 11, and 16, alongside novel regions of genetic loss (e.g., 10q21.1-26.3, 11q24.1-qter). This detailed genetic characterization of the majority of medulloblastoma cell lines provides important precedent for the widespread involvement of copy number-neutral genetic losses in medulloblastoma and demonstrates that combined assessment of copy number aberrations and LOH will be necessary to accurately determine the contribution of chromosomal defects to tumor development.

Mapping of candidate tumor suppressor genes on chromosome 12 in adenoid cystic carcinoma

Adenoid cystic carcinoma (ACC) is a common malignancy of salivary glands, for which the underlying genetic mechanisms of tumorigenesis are poorly understood. Prior studies in ACC have identified deletions in chromosome 12. To further characterize these changes, we performed an extensive LOH analysis in 58 ACC using a panel of 28 microsatellite markers. Results show 66% overall genetic loss. Three markers (D12S1713, D12S2196, D12S398) are contiguous and define a 6.84 Mb region of deletion at 12q13.11-q13.13. Two other markers (D12S2078, D12S1628) are also contiguous and define a 4.5 Mb region of deletion at 12q24.32-q24.33. The three remaining markers, D12S1056 at 12q14.1, D12S1051 at 12q23.1 and D12S1636 at 12q23.3 define smaller regions of deletion. An analysis of microarray gene expression profiling data available for ACC shows several genes with significant transcriptional downregulation that map to these areas of genetic deletion. This combined genetic and genomic analysis provides several candidate genes to test for functional tumor suppressor activity in ACC.

Complex CGH alterations on chromosome arm 8p at candidate tumor suppressor gene loci in breast cancer cell lines

Loss of genetic material from chromosome arm 8p occurs frequently in human breast carcinomas, consistent with this region of the genome harboring one or more tumor suppressor genes (TSGs). We used the complementary techniques of microsatellite-based LOH, high-density FISH, and conventional CGH on 6 breast cancer cell lines (MCF7, SKBR3, T47D, MDA MB453, BT549, and BT474) to investigate the molecular cytogenetic changes occurring on chromosome 8 during tumorigenesis, with particular emphasis on 6 potential TSGs on 8p. We identified multiple alterations of chromosome 8, including partial or complete deletion of 8p or 8q, duplication of 8q, and isochromosome 8q. The detailed FISH analysis showed several complex rearrangements of 8p with differing breakpoints of varying proximity to the genes of interest. High rates of LOH were observed at markers adjacent to or within PCM1, DUSP4/MKP2, NKX3A, and DLC1, supporting their status as candidate TSGs. Due to the complex ploidy status of these cell lines, relative loss of 8p material detected by CGH did not always correlate with microsatellite-based LOH results. These results extend our understanding of the mechanisms accompanying the dysregulation of candidate tumor suppressor loci on chromosome arm 8p, and identify appropriate cellular systems for further investigation of their biological properties.

Genomic deletions in cell lines derived from primitive neuroectodermal tumors of the central nervous system

Extensive genomic deletions affecting a variety of chromosomes are a common finding in primitive neuroectodermal tumors of the central nervous system (CNS-PNETs), implicating the loss of multiple tumor suppressor genes in the pathogenesis of these tumors. We have used representational difference analysis, microsatellite mapping, and quantitative polymerase chain reaction to identify and verify the presence of genomic deletions on a number of chromosomes in CNS-PNET cell lines. This systematic approach has confirmed the importance of deletions at 10q, 16q, and 17p in PNET pathology and has revealed other regions of deletion not commonly described (e.g., Xq, 1p, 7p, and 13q). These data highlight the prevalence of hemizygous loss in CNS-PNET cells, suggesting that haploinsufficiency affecting multiple tumor suppressor genes may play a fundamental role in CNS-PNET pathogenesis. The identification of specific genes and signaling pathways that are compromised in CNS-PNET cells is crucial for development of more efficacious and less invasive treatments, as are urgently needed.

Fine mapping and evaluation of candidate genes for cervical cancer on 11q23

We previously showed that loss of heterozygosity (LOH) at 11q23 is a common genetic alteration in cervical cancer (CC) and that it correlates with extensive invasion of lymph-vascular spaces. In the current study, we looked for allelic loss in paired normal/tumor genomic DNA from 121 cervical tumors by using 20 well-mapped microsatellite markers on 11q. LOH at one or more loci was observed in 81 (66.9%) tumors. The deletion patterns in tumors are complex. However, at least three LOH islands could be defined between D11S614 and D11S4167. We also genotyped 11 CC cell lines and analyzed the results using the homozygosity mapping-of-deletions method. Five of the 11 cell lines showed continuous homozygosity that extended through 11q23.3-11q24.1. We used a candidate-gene approach to screen candidate tumor-suppressor genes (TSGs) that were localized in that region. Intragenic changes in the entire coding sequence of four candidate genes (RNF26, USP2, POU2F3, and TRIM29) in the region and a proposed TSG (PPP2R1B) centromeric to the region were evaluated. The expression status of USP2, POU2F3, TRIM29, and another proposed TSG that is telomeric to the region (BCSC1) also was examined. We identified previously described single-nucleotide polymorphisms (SNPs), several novel variants, and three rare SNPs in the five candidate genes. Decreased expression of POU2F3 and TRIM29 was found in some cervical tumors and CC cell lines. Our results indicate that a major region of LOH in cervical cancer exists within a 3.6-Mb stretch of DNA on 11q23.3-q24.1 and that somatic mutations in RNF26, USP2, TRIM29, POU2F3, or PPP2R1B probably are not important for cervical carcinogenesis.

A network of clinically and functionally relevant genes is involved in the reversion of the tumorigenic phenotype of MDA-MB-231 breast cancer cells after transfer of human chromosome 8

Several investigations have supposed that tumor suppressor genes might be located on human chromosome 8. We used microcell-mediated transfer of chromosome 8 into MDA-MB-231 breast cancer cells and generated independent hybrids with strongly reduced tumorigenic potential. Loss of the transferred chromosome results in reappearance of the malignant phenotype. Expression analysis identified a set of 109 genes (CT8-ps) differentially expressed in microcell hybrids as compared to the tumorigenic MDA-MB-231 and rerevertant cells. Of these, 44.9% are differentially expressed in human breast tumors. The expression pattern of CT8-ps was associated with prognostic factors such as tumor size and grading as well as loss of heterozygosity at the short arm of chromosome 8. We identified CT8-ps networks suggesting that these genes act cooperatively to cause reversion of tumorigenicity in MDA-MB-231 cells. Our findings provide a conceptual basis and experimental system to identify and evaluate genes and gene networks involved in the development and/or progression of breast cancer.

Frequent somatic mutations of the transcription factor ATBF1 in human prostate cancer

Cancer often results from the accumulation of multiple genetic alterations. Although most malignancies are sporadic, only a small number of genes have been shown to undergo frequent mutations in sporadic cancers. The long arm of chromosome 16 is frequently deleted in human cancers, but the target gene for this deletion has not been identified. Here we report that ATBF1, which encodes a transcription factor that negatively regulates AFP and MYB but transactivates CDKN1A, is a good candidate for the 16q22 tumor-suppressor gene. We narrowed the region of deletion at 16q22 to 861 kb containing ATBF1. ATBF1 mRNA was abundant in normal prostates but more scarce in approximately half of prostate cancers tested. In 24 of 66 (36%) cancers examined, we identified 22 unique somatic mutations, many of which impair ATBF1 function. Furthermore, ATBF1 inhibited cell proliferation. Hence, loss of ATBF1 is one mechanism that defines the absence of growth control in prostate cancer.

The 11q terminal deletion disorder: a prospective study of 110 cases

We performed a prospective study of 110 patients (75 not previously published) with the 11q terminal deletion disorder (previously called Jacobsen syndrome), diagnosed by karyotype. All the patients have multiple dysmorphic features. Nearly all the patients (94%) have Paris-Trousseau syndrome characterized by thrombocytopenia and platelet dysfunction. In total, 56% of the patients have serious congenital heart defects. Cognitive function ranged from normal intelligence to moderate mental retardation. Nearly half of the patients have mild mental retardation with a characteristic neuropsychiatric profile demonstrating near normal receptive language ability, but mild to moderate impairment in expressive language. Ophthalmologic, gastrointestinal, and genitourinary problems were common, as were gross and fine motor delays. Infections of the upper respiratory system were common, but no life-threatening infections were reported. We include a molecular analysis of the deletion breakpoints in 65 patients, from which genetic "critical regions" for 14 clinical phenotypes are defined, as well as for the neuropsychiatric profiles. Based on these findings, we provide a comprehensive set of recommendations for the clinical management of patients with the 11q terminal deletion disorder.

Deletion mapping suggests that the 1p22 melanoma susceptibility gene is a tumor suppressor localized to a 9-Mb interval

Loss of the short arm of chromosome 1 is frequently observed in many tumor types, including melanoma. We recently localized a third melanoma susceptibility locus to chromosome band 1p22. Critical recombinants in linked families localized the gene to a 15-Mb region between D1S430 and D1S2664. To map the locus more finely we have performed studies to assess allelic loss across the region in a panel of melanomas from 1p22-linked families, sporadic melanomas, and melanoma cell lines. Eighty percent of familial melanomas exhibited loss of heterozygosity (LOH) within the region, with a smallest region of overlapping deletions (SRO) of 9 Mb between D1S207 and D1S435. This high frequency of LOH makes it very likely that the susceptibility locus is a tumor suppressor. In sporadic tumors, four SROs were defined. SRO1 and SRO2 map within the critical recombinant and familial tumor region, indicating that one or the other is likely to harbor the susceptibility gene. However, SRO3 may also be significant because it overlaps with the markers with the highest 2-point LOD score (D1S2776), part of the linkage recombinant region, and the critical region defined in mesothelioma. The candidate genes PRKCL2 and GTF2B, within SRO2, and TGFBR3, CDC7, and EVI5, in a broad region encompassing SRO3, were screened in 1p22-linked melanoma kindreds, but no coding mutations were detected. Allelic loss in melanoma cell lines was significantly less frequent than in fresh tumors, indicating that this gene may not be involved late in progression, such as in overriding cellular senescence, necessary for the propagation of melanoma cells in culture.

Circulating DNA microsatellites: molecular determinants of response to biochemotherapy in patients with metastatic melanoma

Although biochemotherapy appears to be a promising treatment for metastatic melanoma, its impact remains unpredictable. Microsatellite markers for loss of heterozygosity (LOH) appear to have prognostic significance when identified in primary tumors and serum and/or plasma from cancer patients. However, their association with response to systemic therapy has yet to be assessed. To determine whether microsatellite markers are associated with response to therapy, serum from 41 patients with metastatic melanoma, drawn before the initiation of biochemotherapy, was analyzed for LOH with nine microsatellite markers. During a median follow-up of 13 months, the overall response rate for these 41 patients was 56%, including 13 (32%) complete responses and 10 (24%) partial responses. LOH was detected in sera from 12 (29%) of the 41 patients. The response rate of these 12 patients was 17% (95% confidence interval [CI] = 5% to 45%), whereas that of the 29 patients without LOH was 72% (95% CI = 54% to 85%) (P =.001). All statistical tests were two-sided. The presence of LOH was statistically significant and independently associated with disease progression (multivariable analysis, P =.003). Circulating tumor DNA markers may be useful in assessing prognosis for advanced melanoma patients and their response to biochemotherapy.

HEPN1, a novel gene that is frequently down-regulated in hepatocellular carcinoma, suppresses cell growth and induces apoptosis in HepG2 cells

BACKGROUND/AIMS

Examining genes associated with human hepatocellular carcinoma (HCC) by subtractive hybridisation, we identified a novel transcript, designated as HEPN1, in non-tumorous liver. In this study, we aimed to evaluate HEPN1 gene expression in HCC patients, to characterise and to explore the functional significance of HEPN1 in vitro.

METHODS

One-step reverse transcription-polymerase chain reaction (RT-PCR) and real-time RT-PCR were employed to determine HEPN1 expression in 23 paired (HCC and the adjacent non-HCC) liver specimens. Sequence analyses were performed by bioinformatics. Transfection studies were carried out by expressing HEPN1, V5-fused HEPN1, and green fluorescent protein-fused HEPN1, individually, in HepG2 cells.

RESULTS

Significant downregulation of HEPN1 (P<0.0001) was detected in 22/23 of HCC patients tested. Gene HEPN1 maps to chromosome 11q24.2; and the predicted gene product, a 10-kDa peptide with 88 amino acids, has no homology to known proteins. When transfected into HepG2 cells, HEPN1 reduced cell viability to 37.5+/-2.5% (P=0.001), and induced apoptosis with typical morphological changes as demonstrated by microscopy and Annexin V assay.

CONCLUSIONS

Our data show that HEPN1 is frequently silenced in HCC, and that exogenous HEPN1 exhibits antiproliferative effect on HepG2 cells, suggesting that silencing of HEPN1 may be associated with carcinogenesis of hepatocytes.

A 2.6 Mb interval on chromosome 6q25.2-q25.3 is commonly deleted in human nasal natural killer/T-cell lymphoma

Natural killer (NK)/T-cell lymphoma is a special subtype of rare malignant lymphoma that is more prevalent in Asia than in America and Europe. This newly characterized haemato-lymphoid malignancy is highly aggressive and frequently present in nasal and upper aerodigestive sites. Several studies have reported the commonly deleted region of chromosome 6q21-25 in this particular type of lymphoma. To refine the smallest region of overlapping (SRO) deletion for localization of potential tumour suppressor (TS) genes, we performed loss of heterozygosity (LOH) and homozygosity mapping of deletion (HOMOD) analyses on 37 nasal and nasal-type NK/T-cell lymphoma patients using a panel of 25 microsatellite markers, covering the 6q21-q25 region. In all patients studied, LOH was detected in eight (89%) paired-sample patients, while hemizygous deletion was detected in three (11%) single-sample patients. Combination of the LOH and HOMOD results defined a distinct 3 Mb SRO on chromosome 6q25. Quantitative multiplex polymerase chain reaction analysis of 10 sequence-tagged sites further refined the putative TS-gene-containing region to a 2.6 Mb interval between TIAM2 and SNX9. Eighteen known genes/Unigene clusters and 25 hypothetical genes are located within this 2.6 Mb region, but none are previously identified TS genes. These results provide a framework for future positional cloning of novel TS gene(s) at 6q25.2-q25.3.

Regions of allelic imbalance in the distal portion of chromosome 12q in gastric cancer

AIMS

To define regions of loss on the distal portion of chromosome 12q in gastric adenocarcinoma.

METHODS

Microsatellite analysis on chromosome 12 was performed on 19 human gastric cancer cell lines using 77 markers, 71 of which were within or distal to 12q21; some portions of this region showed extended regions of homozygosity (ERHs) in 10 of 19 gastric cancer cell lines. In addition, microdissected tumour cells from 76 primary gastric adenocarcinomas were examined using 13 markers of interest implicated by the cell line data; 70% of these showed allelic imbalance (AI) at one or more markers in or distal to 12q21.

RESULTS

Mapping ERHs in the cell lines and sites of AI in the tumours identified three regions that contain putative tumour suppressor genes: region A is located within 2.8 Mb between markers D12S1667 and D12S88; region B, within 1.9 Mb between markers D12S1607 and D12S78; and region C, in 0.74 Mb between markers D12S342 and D12S324. Fluorescence in situ hybridisation (FISH) analysis in two cell lines confirmed that two of the ERHs reflected deletions, not amplifications, of D12S81 in region A and D12S340 in region C. FISH analysis of marker D12S1075 within an ERH containing region B in one cell line showed neither amplification nor deletion. AI on 12q was not associated with prognosis, but was associated with ethnicity of the patient.

CONCLUSIONS

These results identify regions on chromosome 12 that appear to contain tumour suppressor genes important in the development of gastric cancer.

Analysis of DLC-1 expression in human breast cancer

The chromosome region 8p12-p22 shows frequent allelic loss in many neoplasms, including breast cancer (BC). The DLC-1 gene, located on 8p21-p22, might be a candidate tumor suppressor gene in this region. To evaluate the involvement of DLC-1 in breast carcinogenesis we studied DLC-1 mRNA expression in a panel of 14 primary human BC and the corresponding normal breast cells as well as 8 BC cell lines. Low levels or absence of DLC-1 mRNA were observed in 57% of primary BC and 62.5% of BC cell lines, respectively. We could not find any correlation between DLC-1 mRNA expression and deletions at the DLC-1 locus. Transfection of the gene into DLC-1 deficient T-47D cells raised the DLC-1 mRNA level and resulted in inhibition of cell growth and reduced colony-forming capacity. Our results indicate a role of DLC-1 in BC carcinogenesis.

KLF5 is frequently deleted and down-regulated but rarely mutated in prostate cancer

BACKGROUND

Previous studies mapped a region at the q21 band of chromosome 13 (13q21), which is frequently deleted in various human cancers including prostate cancer, suggesting the existence of a tumor suppressor gene at 13q21. The target gene of deletion in prostate cancer, however, has not been identified at present.

METHODS

We examined four non-neoplastic and 18 neoplastic prostatic cell lines or xenografts. Homozygous/hemizygous deletion was detected by assays of duplex PCR and real-time PCR. Expression levels of genes were determined by the methods of RT-PCR, real time PCR, and northern blot analysis. Mutations of KLF5 were detected by the approaches of single strand conformational polymorphism (SSCP) and direct sequencing. For the detection of promoter methylation, Southern blotting of genomic DNA and restriction digestion or SSCP analysis of methylation specific PCR products were used. Finally, an expression plasmid of KLF5 was introduced into prostate cancer cell lines with reduced KLF5 expression to investigate colony formation for cell growth.

RESULTS

A 2-Mb region of homozygous deletion at 13q21 was detected in the LUCaP70 xenograft of prostate cancer. This region of deletion was further narrowed to 142 Kb by a hemizygous deletion in the NCI-H660 cell line. KLF5 was identified as the only complete gene in the smallest region of deletion. Quantitative deletion of KLF5 genome occurred in six of the 18 (33%) prostate cancer xenografts/cell lines. Each of the six samples with deletion also showed loss of expression for KLF5, suggesting that hemizygous deletion is one mechanism for loss of KLF5 expression. In total, 16 of the 18 cases (89%) showed loss of KLF5 expression at different degrees. In contrast, mutations and promoter methylations were not detected in any of the samples. Functionally, restoration of KLF5 in DU 145 and 22Rv1 cell lines significantly inhibited their growth in vitro.

CONCLUSIONS

Frequent genomic deletion and loss of expression as well as cell growth suppression indicate that KLF5 is a reasonable candidate for the tumor suppressor gene at 13q21 in prostate cancer. Mutation and promoter methylation are not common mechanisms for the inactivation of KLF5 in prostate cancer.

Deletion, mutation, and loss of expression of KLF6 in human prostate cancer

Kruppel-like factors (KLFs) are a group of transcription factors that appear to be involved in different biological processes including carcinogenesis. In a recent study, KLF6 was reported as a tumor suppressor gene in prostate cancer because of its frequent loss of heterozygosity (LOH) and mutation as well as functional suppression of cell proliferation. Loss of chromosomal locus spanning KLF6 is relatively infrequent in other published studies of prostate cancer, however. To clarify the role of KLF6 in prostate cancers, particularly those that are high grade, we examined KLF6 for deletion, mutation, and loss of expression in 96 prostate cancer samples including 21 xenografts/cell lines. Loss of heterozygosity occurred in 4 (19%) of 21 xenografts/cell lines and 8 (28%) of 29 informative tumors. Fourteen of the 96 (15%) samples showed 15 somatic sequence changes in the KLF6 gene, including 7 that changed KLF6 peptide sequences, 4 that did not, and 4 that were located in untranslated regions. Expression levels of KLF6 were significantly lost in 4 of 20 (20%) xenografts/cell lines of prostate cancer, as detected by RT-PCR and Northern blot analysis. These findings indicate that significant genetic alterations of KLF6 occur in a minority of high-grade prostate cancers.

Chromosome 8 genetic analysis and phenotypic characterization of 21 ovarian cancer cell lines

The short arm of chromosome 8 undergoes frequent loss of heterozygosity (LOH) in ovarian adenocarcinomas. Fine mapping has identified several distinct critical regions within 8p which undergo rates of LOH of 50% or greater, suggesting that there may be more than one tumor suppressor gene located on this chromosome arm. In an effort to refine the location of these putative tumor suppressor genes by homozygosity-mapping-of-deletion analysis, we have analyzed 21 ovarian cancer cell lines with 19 polymorphic microsatellite markers from 8p. Eleven of the cell lines (55%) were homozygous at every marker, indicating loss of an entire 8p arm. No smaller extended regions of hemizygosity were identified. Refinement of these 8p target regions was therefore not possible, but this analysis did identify the ovarian cancer cell lines that would be most appropriate for microcell-mediated chromosome transfer to complement the hypothesized mutation in the target tumor suppressor gene(s) on 8p. The 11 cell lines that had undergone 8p LOH were therefore characterized for colony formation in soft agar and tumor formation in nude mice. We identified four cell lines (JAM, OVCA4, OVCA5, and OVCA8) that were hemizygous for 8p and that formed colonies in soft agar and tumors in nude mice, making them ideal cell lines for chromosome 8 or candidate gene transfer.

Mutation analysis and mRNA expression of trail-receptors in human breast cancer

The chromosome region 8p12-p22 shows frequent allelic loss in a variety of human malignancies, including breast cancer (BC). The tumor necrosis factor-related apoptosis-inducing ligand (TRAIL)-receptors TRAIL-R1, -R2, -R3 and -R4 are located on 8p21-p22 and might be candidate tumor suppressor genes in this region. To evaluate the involvement of TRAIL receptors in breast carcinogenesis, we have analyzed the entire coding region of TRAIL-R2 and the death domain (DD) regions of TRAIL-R1 and -R4 for the detection of somatic mutations in a series of breast tumors, lymph node metastases and BC cell lines. Overall, we detected 1, 11 and 3 alterations in the TRAIL-R1, -R2 and -R4 genes, respectively. Although functional studies have not yet been performed, we assume that most of these alterations do not alter the function of TRAIL-receptors. Additionally, we analyzed individuals from BC families for the detection of TRAIL-R2 germline mutations. One alteration has been found in the Kozak consensus motif at position -4 with respect to the translation initiation AUG [1-4 (C-->A)]. We further studied the mRNA expression of TRAIL and the 4 TRAIL receptors. In BC cell lines, a strongly decreased mRNA expression of TRAIL, TRAIL-R1, -R3 and -R4 was found, whereas the expression of TRAIL-R2 was only slightly reduced. In breast tumors, a 1.2-3.6-fold reduction of mRNA signals of the 5 genes was observed. No correlation was found between the expression level of TRAIL and the receptor mRNAs and clinicopathologic variables and between the expression of TRAIL-R2 and TP53 mutation status and loss of heterozygosity (LOH) at 8p21-p22. Taken together, we cannot exclude the involvement of TRAIL-receptors in BC. Our mutation studies indicate that DD receptor mutations occur at low frequency and are not the primary cause for the altered mRNA expression of TRAIL and TRAIL-receptors in BC.

The random development of LOH on chromosome 9q in superficial bladder cancers

Allelic loss on chromosome 9q is a very frequent event in bladder carcinogenesis. In recent years, efforts have been directed towards identifying the postulated tumour suppressor genes on this chromosome arm by deletion mapping and mutation analysis. However, no convincing candidate genes have been identified. This paper describes the development of chromosome 9q alterations in multiple recurrent superficial bladder cancers of ten patients and shows that loss of heterozygosity (LOH) on this chromosome is almost never the characteristic first step. The regions of loss are multiple and variable in different tumours from the same patient and expand in subsequent tumours. Moreover, the regions of loss vary from patient to patient. It is concluded that even if 9q harbours a bladder cancer gatekeeper gene, it is unlikely that the gene will be identified through LOH analysis alone.

Defining the region(s) of deletion at 6q16-q22 in human prostate cancer

Deletion of the long arm of chromosome 6 (6q) frequently occurs in many neoplasms, including carcinomas of the prostate and breast and melanoma, suggesting the location of a tumor-suppressor gene or genes at 6q. At present, however, the region of deletion has not been well defined, and the target gene of deletion remains to be identified. In this study, we analyzed 44 primary prostate cancers with 16 polymorphic markers for loss of heterozygosity (LOH) by using PCR-based techniques. We also examined 23 cell lines/xenografts of prostate cancer with 38 markers for LOH by the method of homozygosity mapping of deletion. LOH at 6q16 - q22 was detected in 21 of 44 (48%) primary tumors and in 12 of 23 (52%) cell lines/xenografts. Two regions of LOH were defined. One was 7.5 cM at 6q16 - q21 between markers D6S1716 and D6S1580, and the other was 4.3 cM at 6q22 between D6S261 and D6S1702. Whereas no correlation was found between LOH at 6q16-q22 and patient age at diagnosis or Gleason score, tumors at higher stage appear to have more frequent LOH. These findings suggest that deletion of 6q16 - q22 is a frequent event in prostate cancer, and that the deletion originates from two distinct regions. These results should be useful in identifying the target gene(s) of deletion at 6q.

Defining regions of loss of heterozygosity of 16q in breast cancer cell lines

The loss of heterozygosity (LOH) of chromosome 16 was assessed in 21 breast cancer cell lines and two nontumorigenic breast epithelial cell lines by typing microsatellite markers distributed on this chromosome. In addition, dual-color fluorescence in situ hybridization was used to metaphase spreads of these cell lines using chromosome 16 paint and region specific probes. Eleven of the cell lines had LOH for chromosome 16, two for the entire chromosome, three for the long arm, and six had LOH for restricted regions of the long arm. The results supported evidence that there are two predominant regions of LOH, 16q22.1 and 16q24.3. The cell lines with chromosome 16 LOH can be used for screening candidate tumor suppressor genes at 16q in breast cancer.

Identification of candidate liver tumor suppressor genes from human 11p11.2-p12

We have previously described a functional model for identification of human liver tumor suppressor genes in which human chromosome 11 was introduced into rat liver epithelial tumor cell lines via microcell-mediated chromosome transfer, producing microcell hybrid (MCH) cell lines that exhibit suppression of tumorigenicity in vivo. Chromosome deletion mapping studies identified a 950-kb region of 11p11.2-p12 that was retained in all suppressed MCH cell lines, suggesting that this region may harbor one or more genes with liver tumor suppressor function. In this study, we generated a comprehensive transcription map of the 11p11.2-p12 liver tumor suppressor region through examination of 142 expressed sequence tag (EST) markers among a group of suppressed MCH cell lines. Of 142 ESTs examined, 19 were localized within the 11p11.2-p12 liver tumor suppressor region. RT-PCR analysis of gene expression for these 19 ESTs among an index panel of suppressed MCH cell lines (n = 3) identified 11 potential candidate liver tumor suppressor genes. Examination of candidate gene expression among six additional suppressed MCH cell lines reduced the number of potential candidate genes to three (stSG30184, stSG10014, and stSG29748). Northern blot analysis of suppressed MCH cell lines and derived tumor cell lines suggested stSG30184 as the best candidate liver tumor suppressor gene. The 3.7 kb stSG30184 transcript was expressed by all suppressed MCH cell lines, but expression was extinguished coordinately with reexpression of tumorigenicity by these cells, consistent with a tumor suppressor gene. Subsequent characterization of this EST indicates that it is a novel transcript with expression in a broad range of tissue types. Further characterization of the genes identified in this study will provide a greater understanding of their role in the molecular pathogenesis of neoplastic liver disease.

An 800-kb region of deletion at 13q14 in human prostate and other carcinomas

Deletions of regions at 13q14 have been detected by various genetic approaches in human cancers including prostate cancer. Several studies have defined one region of loss of heterozygosity (LOH) at 13q14 that seems to reside in a DNA segment of 7.1 cM between genetic markers D13S263 and D13S153. To define the smallest region of overlap (SRO) for deletion at 13q14, we first applied tissue microdissection and multiplex PCR to detect homozygous deletion and/or hemizygous deletion at 13q14 in 134 prostate cancer specimens from 114 patients. We detected deletions at markers D13S1227, D13S1272, and A005O48 in 13 (10%) of these tumor specimens. Of the 13 tumors with deletions, 12 were either poorly differentiated primary tumors or metastases of prostate cancer. To fine-map the deletion region, we then constructed a high-resolution YAC/BAC/STS/EST physical map based on experimental and database analyses. Several markers encompassing the deletion region were analyzed for homozygous deletion and/or hemizygous deletion in 61 cell lines/xenografts derived from human cancers of the prostate, breast, ovary, endometrium, cervix, and bladder, and a region of deletion was defined by duplex PCR assay between markers A005X38 and WI-7773. We also analyzed LOH at 13q14 in the 61 cell lines/xenografts using the homozygosity mapping of deletion approach and 26 microsatellite markers. We found 24 (39%) of the cell lines/xenografts to show LOH at 13q14 and defined a region of LOH by markers M1 and M5. Combination of homozygous or hemizygous deletion and LOH results defined the SRO for deletion to be an 800-kb DNA interval between A005X38 and M5. There are six known genes located in or close to the SRO for deletion. This region of deletion is at least 2 Mb centromeric to the RB1 tumor-suppressor gene and the leukemia-associated genes 1 and 2, each of which is located at 13q14. These data suggest that the 800-kb DNA segment with deletion contains a gene whose deletion may be important for the development of prostate and other cancers. This study also provides a framework for the fine-mapping, cloning, and identification of a novel tumor-suppressor gene at 13q14.

High-density allelotype of the commonly studied gastric cancer cell lines

Gastric cancer is one of the leading causes of death from cancer throughout the world, and studies to elucidate the genetic defects found in this type of cancer are growing in number. Increasingly sophisticated techniques and the sequencing of the human genome have had an impact on the scope of such studies. While the use of tumor specimens remains popular, more emphasis is being placed on cell lines as model systems where specific data can be directly combined with results from other studies. This article describes a genetic survey of the most widely used gastric adenocarcinoma cell lines. The allelotype at 351 polymorphic loci in 14 cell lines was obtained, and the results from the 4,900 polymerase chain reactions are displayed. In addition to confirming loss of heterozygosity on chromosome arms 6p, 7q, 17p, and 18, additional deletions on arm 5p and the pericentromeric regions of chromosomes 1 and 10 were detected. Areas that might contain homozygous deletions or amplifications also were mapped. The rate of microsatellite instability was quantified and shown to vary greatly among the different cell lines. Most important, this study serves as a genetic scaffold for the integration of past and future studies on the nature of the genetic defects in gastric cancer.

Defining a common region of deletion at 13q21 in human cancers

Previous molecular genetic analyses identified a region of deletion at 13q21 in a variety of human cancers, suggesting the existence of a tumor suppressor gene(s) at this locus. In our earlier study on prostate cancer, the region of deletion was confined to a 3.1 cM interval between D13S152 and D13S162. At present, however, no known gene located in this interval has been firmly implicated in cancer, and the region remains too large for gene identification. To fine-map the area of interest, we established a contig of bacterial artificial chromosome (BAC) clones, narrowed the region of deletion by loss of heterozygosity (LOH) and homozygosity-mapping-of-deletion (HOMOD) analyses in different types of cancers, and tested a candidate gene from the region for mutation and alteration of expression in prostate cancers. The contig consisted of 75 overlapping BAC clones. In addition to the generation of 47 new sequence-tagged-site (STS) markers from the ends of BAC inserts, 76 known STS and expressed sequence tag markers were mapped to the contig (25 kb per marker on average). The minimal region of deletion was further defined to be about 700 kb between markers D13S791 and D13S166 by LOH analysis of 42 cases of prostate cancer, and by HOMOD analysis of eight prostate cancer cell lines/xenografts and 49 cell lines from cancers of the breast, ovary, endometrium, and cervix, using 18 microsatellite markers encompassing the deletion region. A gene that is homologous to the WT1 tumor suppressor gene, AP-2rep (KLF12), was mapped in this region and was analyzed for its expression and genetic mutation. In addition to low levels of expression in both normal and neoplastic cells of the prostate, this gene did not have any mutations in a group of aggressive prostate cancers and cell lines/xenografts, as assessed by the methods of polymerase chain reaction-single strand conformational polymorphism analysis and direct sequencing. These studies suggest that a 700 kb interval at 13q21 harbors a tumor suppressor gene(s) that seems to be involved in multiple types of cancer, and that the AP-2rep gene is unlikely to be an important tumor suppressor gene in prostate cancer. The BAC contig and high-resolution physical map of the defined region of deletion should facilitate the cloning of a tumor suppressor gene(s) at 13q21.