| IntroductionPancreatic adenocarcinoma is characterized by aggressive tumor growth and a high metastatic propensity, resulting in a poor prognosis with a 5-year survival rate of only 4% in the United States. The pancreatic carcinoma genome harbors multiple amplifications and deletions, pointing to the existence of numerous oncogenes and tumor suppressor genes (TSGs) driving the genesis and progression of this lethal cancer. Understanding the genetic events may serve as a useful adjunct in the screening and treatment of pancreatic cancer.In the last few years, the development of techniques has allowed the genome-wide detection of alterations in copy number. The initial approach used was comparative genomic hybridization (CGH), which is based on competitive in situ hybridization of differentially labeled tumor DNA and normal DNA on a normal human metaphase spread. Derived from the concept of conventional CGH, array CGH increases the resolution of the distance between the selected marker DNA clones for the practical genome-wide detection of chromosomal alterations from～20 Mb to less than 1 Mb. Recently, an array consisting of 32,433 overlapping bacterial artificial chromosome (BAC) clones covering the entire human genome has improved the resolution to 80 kb.Nevertheless, the primary limitation of these methods is that CGH can detect a net gain or loss of genetic materials, but it does not identify situations in which the loss of one allele is followed by reduplication of the remaining allele. This limitation is overcome by using high-density oligonucleotide arrays to measure single nucleotide polymorphisms (SNPs). SNPs are the most frequent form of DNA variation present in the human genome, and they are evenly spaced and stable across the genome. SNPs are emerging as markers for high-resolution whole genome allelotyping that allows accurate copy number measurements as alternatives to restriction fragment length polymorphisms (RFLPs) and microsatellite markers. The Affymetrix GeneChip Mapping 100K high-density SNP array (Affymetrix, Santa Clara, CA) allows the analysis of 116,204 SNPs, distributed evenly across the genome. The array is designed for the genome-wide typing of SNPs to facilitate detailed mapping of DNA copy number variation. In a previous study, we presented an improved algorithm implemented in a software package called Copy Number Analyzer for the Affymetrix GeneChip Mapping 100K array (CNAG), which improves the signal-to-noise ratio and optimizes the selection of the reference. Here, we used this technique to characterize the genome-wide homozygous deletions and amplifications in pancreatic cancer cell lines in an effort to find novel targets for the early diagnosis and therapy of pancreatic cancer.Material and methodsPancreatic cancer cell lines. Twenty-five established cell lines derived from pancreatic carcinomas or their metastases were included in the investigation. Sixteen cell lines (AsPC-1, BxPC-3, Capan-1, Capan-2, CFPAC-1, HPAF-II, Hs 700T, Hs 766T, PANC-1, Panc 02.03, Panc 03.27, Panc 05.04, Panc 08.13, Panc 10.05, SU.86.86, and SW 1990) were acquired from the American Type Culture Collection (ATCC, Manassas, VA). Five ceil lines (KP-1N, KP-2, KP-3, KP-4, and MIA PaCa-2) were obtained from the Japanese Collection of Research Bioresources (JCRB, Osaka, Japan). KLM-1 and NOR-P1 were from the Institute of Development, Aging and Cancer, Tohoku University. T3M-4 and PSN1 were from the RIKEN BioResource Center and European Collection of Cell Cultures (ECACC, Salisbury, UK), respectively. The cells were grown under recommended culture conditions. Genomic DNA was isolated using standard protocols with a PUREGENE DNA Isolation Kit (Gentra Systems, Minneapolis, MN).Single nucleotide polymorphism array. Array experiments were conducted according to the manufacturer's recommendations for the Affymetrix GeneChip Mapping 100K Array. In brief, 250 ng of total genomic DNA were digested with XbaI or HindIII restriction enzyme, ligated to an adaptor, and amplified by PCR using a single primer. After fragmentation with DNase I to a size range of 250 to 2000 bp, the PCR products were labeled, denatured, and hybridized to the array. Finally, the arrays were washed on the Affymetrix fluidics station, stained, and scanned.Genomic PCR. To verify homozygous deletions, we used gene-specific primers to PCR-amplify genomic DNA from the cell lines. Simultaneously, DNA from a normal person and distilled water were included in the PCR reactions as controls. PCR was performed on an Applied Biosystems (Foster City, CA) GeneAmp PCR System 9600.Quantitative real-time PCR. Gene copy numbers were determined by quantitative real-time PCR using the ABI PRISM 7000 Sequence Detection System and SYBR? Green PCR Master Mix (Applied Biosystems). The standard curve method was used to calculate target gene copy numbers in tumor DNA normalized to GAPDH as a gene reference.ResultsTwenty-five pancreatic cancer cell lines were studied with the tools of high-resolution SNP arrays and CNAG.Amplifications in pancreatic cancer cell lines. Twenty-three regions with known genes were amplified in at least two cell lines each, and the size of minimal common amplification was 146 kb at 13q22.2. The most frequently amplified region was found in chromosome region 12p12.1-12p11.23 containing the KRAS gene in six of 25 cell lines (24%). On average, the number of candidate genes within each amplified region was 4.87. Of the 23 amplification regions, eight loci were newly identified, including those at 1q42.2-1q42.3, 2q36.3, 3q25.1, 5p15.2,10q23.32, 11q21, 13q22.2, and 19p13.12.Homozygous deletions in pancreatic cancer cell lines. Twenty-six regions of homozygous deletion with known genes (inferred copy number n = 0 by HMM) containing at least two consecutive SNPs were detected in at least one cell line. Every homozygous deletion locus was verified by genomic PCR. Of these, 20 regions had only one candidate gene, four regions had two candidate genes, and two had four genes. On average, 1.38 genes were involved in each region. The size of the minimal deletion was 6 kb. The most common homozygous deletion region was 9p21.3, found in 11 cell lines.In our study, seven novel homozygous deletion loci were identified at 2q22.1, 6q22.33, 9q22.33, 20p12.1, Xp21.2, Xp21.1, and Xq26.3, with one to four candidate genes.LOH in pancreatic cancer cell lines. LOH mapped to chromosome arm 9p (22/25, 88%), 18q (22/25, 88%), 17p (21/25, 84%), 8p (17/25, 68%), 13q (17/25, 68%), 6q (16/25, 64%), 3p (15/25, 60%), 6p (14/25, 56%), 22q (14/25, 56%), 9q (13/25, 52%), and 12q (13/25, 52%) with high frequency (>50%). The most common abnormalities involved loss of portions of 9p and 18q, which occurred in 22 cell lines.Conclusions1. Using high-resolution SNP arrays, we were able to study genome-wide copy number alteration and LOH simultaneously.2. We could identify narrow amplified and deleted region to find new putative oncogenes and tumor suppressor genes by using the Affymetrix GeneChip Mapping 100K Array and CNAG.3. The novel amplified or deleted genes discovered here might become targets for the early diagnosis and therapy for pancreatic cancer.
|
PDF Full Text Request