Abstract
Gene expression levels are regulated at many levels. Integration of genome-wide analyses for the study of DNA and RNA provides a unique tool to detect genetic alterations in the cancer genome. In this study, we generated and integrated DNA amplification data from comparative genomic hybridization (CGH) and serial analyses of gene expression (SAGE) in order to obtain a molecular profile of gastroesophageal junction (GEJ) carcinomas. DNA amplifications mapped to specific chromosomal regions and were frequently seen at 1q, 4q, 5q, 6p, 7p, 8q, 17q, and 20q. Using SAGE, we obtained over 156,432 tags from GEJ adenocarcinomas and normal gastric mucosa. These tags were assigned to UniGene clusters. Chromosomal positions for overexpressed genes were obtained to produce a GEJ carcinoma transcriptome map. A total of 123 genes was significantly overexpressed (more than fivefold; P < .01) in one or more SAGE libraries. This gene overexpression map was integrated and compared to the chromosomal CGH ideogram. Several chromosomal arms that had frequent DNA amplifications showed frequent gene expression alterations such as chromosomes 1 (15 genes), 2 (9 genes), 6 (6 genes), 11 (6 genes), 12 (8 genes), and 17 (13 genes). Despite the relatively large DNA amplification regions, overexpressed genes frequently mapped and clustered to small chromosomal regions at early-replicating (Giemsa light) bands such as 1q21.3 (nine genes), 6p21.3 (five genes), and 17q21 (eight genes). These results provide a comprehensive tool to search for DNA amplifications and overexpressed genes in GEJ carcinoma. The observed phenomenon of the presence of large amplification areas, yet clustering of overexpressed genes to relatively small loci, may suggest a high organization of chromatin and cancer-related genes in the nucleus.
Keywords: Gastroesophageal junction cancer, comparative genomic hybridization, serial analysis of gene expression, gene clustering, transcriptome analyses
Introduction
Gastroesophageal junction (GEJ) carcinomas have the most rapidly rising incidence of all visceral malignancies in the United States and Western world [1,2]. The majority of GEJ carcinomas are sporadic and exhibit various levels of DNA ploidy and chromosomal instability [3]. Comprehensive DNA copy number analyses using comparative genomic hybridization (CGH) have been widely used to characterize the DNA alteration in several cancer types. CGH can demonstrate recurrent DNA copy number changes and map them to chromosomal locations [4]. The development of serial analyses of gene expression (SAGE) technology has enabled genomewide unlimited comprehensive profile of gene expression in a given cell population, representing the entire transcriptome [5,6]. This method has been valuable in studies of several tumor types including adenocarcinomas of the colon [7,8], prostate [9], pancreas [10], ovary [11], and breast [12].
Analyses of the human transcriptome map have shown clustering of highly expressed genes in chromosomal domains [13]. Chromosomal arms and bands are known to occupy specific locations within the nucleus known as chromosome territories (CTs). The positioning of a gene(s) can influence its access to the machinery responsible for specific nuclear functions such as transcriptional level and splicing [14]. In this study, we have globally explored the genome of GEJ carcinomas at the DNA and RNA levels, and mapped the DNA and gene expression changes to chromosomal positions, thereby generating a comprehensive genetic map of this deadly disease.
Materials and Methods
CGH
CGH was performed on 18 xenografted carcinomas that were generated from fresh tissues from surgically resected carcinomas of the lower esophagus (n = 3) or GEJ (n = 15). Xenografting was performed as previously described [15]. We and others have shown that xenografted tumor tissues provide a pure source to study gene amplification and expression similar to primary tumor samples [16,17]. Mice were examined for tumor growth, and neoplasms were harvested and frozen upon reaching approximately 1 cm in diameter. High molecular weight genomic DNA was prepared from these frozen xenografted tumors by standard organic extraction methods. Histologic confirmation of the xenografted tumors was performed on cryostat-sectioned slides stained with hematoxylin and eosin (H&E). The CGH experiments were performed using a mixture of fluorochromes conjugated to dCTP and dUTP nucleotides for nick translation. Hybridizations, washings, and ISIS digital image analysis (Metasystems GmbH, Altlussheim, Germany) were performed as described elsewhere [18]. All the CGH results were confirmed using a 99% confidence interval. In each CGH experiment, a negative control (peripheral blood DNA from a healthy donor) and a positive control were included. Based on our earlier reports and the control results, we used 1.17 for DNA amplifications and 1.50 for high-level amplifications (HLAs).
SAGE
High-quality total RNA (500 µg) was extracted using RNeasy kit (QIAGEN GmbH, Hilden, Germany) from two dissected GEJ adenocarcinomas and a pool of four normal gastric epithelia biopsy samples that came from four patients who were referred for endoscopy for dyspepsia or for screening prior to gastric bypass surgery for obesity. The two tumors that were used in SAGE had closely similar pathological characters. GSM757 is a GEJ adenocarcinoma (T4N0M0, moderately differentiated). GSM2385 is a GEJ adenocarcinoma (T4N0M0, poorly differentiated). All normal samples had histologically normal mucosa confirmed on review of H&E-stained sections. Importantly, histopathology examination confirmed that none of the normal samples had any areas of inflammation or necrosis. The tumors selected for SAGE analysis were estimated to consist of more than 80% tumor cells. All samples were collected after obtaining informed consent in accordance with the Human Investigation Committee regulations at the University of Virginia.
SAGE libraries were constructed using NlaIII as the anchoring enzyme and BsmFI as the tagging enzyme, as described in SAGE protocol version 1.0e, June 23, 2000, which includes a few modifications of standard protocol [5]. A detailed protocol and schematic of the method is available (http://www.sagenet.org/sage_protocol.htm). Two thousand clones were sequenced for each case by the Cancer Genome Anatomy Project (CGAP). We used eSAGE 1.2a software to extract SAGE tags, remove duplicate ditags, tabulate tag contents, and link SAGE tags in the database to UniGene clusters using the recently reported ehm-Tag-Mapping method [19,20]. The resulting tag libraries tags were compared to UniGene cluster and to the SAGE tag “reliable” mapping database (ftp://ftp.ncbi.nih.gov/pub/sage/map/Hs/Nla3/), and statistical analyses were performed using the eSAGE software. We have only included tags that had reliable gene hits. Tags with multiple gene hits were considered nonspecific and excluded from further analyses. Significant changes in levels of expression (P % .01) were determined.
Quantitative Real-Time Polymerase Chain Reaction (PCR)
For quantitative real-time PCR, 20 primary GEJ cancers and 13 normal gastric epithelial samples were collected. All tumors and normal gastric mucosal epithelial tissues were verified by our histopathologist (H.F. and C.M.). The collected tumors ranged from well-differentiated (WD) to poorly differentiated (PD) stages I-IIIa, and there was a mix of intestinal and diffuse-type tumors. The mRNA was isolated using RNeasy kit (QIAGEN GmbH). Single-stranded cDNA was synthesized using Advantage™ RT-for-PCR Kit (Clontech, Palo Alto, CA). Quantitative PCR was performed using iCycler (Bio-Rad, Hercules, CA) and threshold cycle number was determined using iCycler software version 3.0. Reactions were performed in triplicate and threshold cycle numbers were averaged. Gene-specific primers for 11 genes (LGALS3BP, PPP1R1B, HSPA5, TACSTD1, ANXA1, TOP2A, S100A6, S100A7, S100A8, S100A9, and S100A10) were designed. These genes were chosen to cover different chromosomal locations and to be representative of different levels of gene overexpression in SAGE data. The primers used for RT-PCR were obtained from GeneLink (Hawthorne, NY), and their sequences are available on request. The results were normalized to β-amyloid precursor protein (APP), which had minimal variation in all normal and neoplastic GEJ samples that we tested. Fold overexpression was calculated according to the formula 2(Rt-Et)/2(Rn-En), as earlier described [21], where Rt is the threshold cycle number for the reference gene observed in the tumor, Et is the threshold cycle number for the experimental gene observed in the tumor, Rn is the threshold cycle number for the reference gene observed in the normal sample, and Rt is the threshold cycle number for the reference gene observed in the tumor sample. Rn and En values were an average for the 13 normals that were analyzed.
Results and Discussion
CGH analyses demonstrated complex DNA changes in all GEJ carcinomas (Table 1; Figure 1). A total of 164 DNA amplifications was detected, with 54 HLAs in different chromosomal areas (Table 1). These amplifications were frequently (>45%) seen in chromosome arms such as 1q, 6p, 7p, 8q, and 17q. Our results are in agreement with a similar CGH study on GEJ carcinomas [22].
Table 1.
DNA Copy Number Gains and Amplifications Using CGH on 18 GEJ and Esophageal Adenocarcinomas.
| Case Number Gains and HLAs (in bold) | |
| 1 | 12p13, 17q, 20q12-q13.1 |
| 2 | 1p31.2-pter (1p34.2-pter), 1q , 3p14-pter, 3q13.2-q24, 5q31-qter, 7p, Xp, 8p, 8q22-qter, 10p, 11q (11q23.3-qter), 16p, 17q12-q21, 18, 19q, 20q |
| 3 | 1p32-pter, 5p, 6p, 7p, 10p, 15q21-qter, 16p, 17q, 20q (20q12-13.1), 22 |
| 4 | 1q, 2pter-q14.3, 3p22-pter, 3q21-qter (3q24-q26.3), 4p, 5p, 6q22-qter, 7q21.2-q22.1, 8p (8p21-pter), 8q21.1-qter, 11q13, 12pter-q15, 12q23-qter, 14q13-qter, 15q22-qter, 17q, 20 (20q13), 22 |
| 5 | 1q, 2q31-q33, 7p, 8q (8q24), 10q, 12, 15q24-qter, 17q22, 19q, 20 |
| 6 | 1q, 3q21-qter, 5p, 6p11-p21, 7pter-q22, 8q21.2-qter, 10q24-qter, 17q, 18, 20q, 21 |
| 7 | 1q21-qter, 2p, 3 (3q25-q26.1, 3q27-q28), 5p12-p13.1, 6p, 7p, 8q, 11p14-qter, 12p12, 12q14-q15, 12q24-q25, 14q11-q21, 17q, 19q, 20 (20q) |
| 8 | 1q21-qter, 2q, 3p24-pter, 7, 8q, 10pter-q22, 10q25-qter, 12p12, 12q14-q15, 12q24.2-qter, 14q11-q21, 17q, 18, 19q, 20q |
| 9 | 1q21-qter, 2q31, 2q33-q34, 5p, 6pter-q24 (6p), 7p, 8q22-qter, 10, 11q13.4-q14, 12, 13, 15q21-qter, 16q12-qter, 17q12, 20, 22 |
| 10 | 1q32-qter, 3q13.3-q23, 3q27-qter, 7q32-qter, 9q, 17q, 20q |
| 11 | 2q (2q32-q34), 3q, 6p, 7q11-q31, 10q, 12, 13q13-q21, 17, 18p, 20 |
| 12 | 3p21-pter, 7p14-q31 (7q21-q31), 8, 14q11-q24.1, 17q, 20, 22 |
| 13 | 3q13.3-q23, 6p21.2-p21.3, 7p, 8q24, 12p12-pter, 14q21-qter, 19q, 20q |
| 14 | 4q11-q13, 6q22-q24, 7p21-pter, 8q23-qter, 11q11-q13, 17q, 19q, 20 |
| 15 | 4q11-q21, 7q21, 8p22-p23, 12q14-q15, 15q25-q26, 17q21.2-qter, 19q, 20q |
| 16 | 6p21, 7p (7p13-pter), 8q21.3-qter (8q23-qter) |
| 17 | 7pter-q21 (7p11-p13), 8q (8q24), 10pter-q22, 12q21-qter, 17q (17q11-q21.2), 19q, 20q, Xp |
| 18 | 8q22-qter, 9q, 10q21-q22, 11q13-q14, 12p, 17q22-qter, 20 |
Figure 1.
Chromosomal ideogram showing DNA amplifications and gene overexpression alterations in GEJ carcinomas. Chromosomes are arranged vertically next to each other. Chromosome numbers are shown on the left-hand side. DNA amplifications are shown at the right-hand side of the chromosome. The scale at the bottom indicates the frequency (%) of the change. Gene expression alterations are also shown. Overexpression changes are shown at the right-hand side of the chromosome. The length of the horizontal bar correlates with the number of altered genes in a given chromosomal location. The scale is shown at the bottom. The highest gene clusters are seen at 1q21.3, 6p21.3, and 17q21.
Global analyses of gene expression using SAGE libraries produced more than 156,432 expressed tags. Comparison to UniGene cluster, release January 2003, identified 26,633 unique SAGE tags. The most attractive feature of SAGE is its ability to evaluate the expression pattern of thousands of genes in a quantitative manner without prior sequence information [23–26]. The genome-wide transcriptome analyses for significant (P ≤ .01) high expression changes (≥5-fold) revealed that 123 genes are overexpressed in GEJ carcinomas as compared to normal mucosa (Table 2). The overexpressed genes (P ≤ .01) included several genes that are cancer-related such as S100A proteins (A2, 6, 7, 8, 9, and 10), heat shock proteins (HSPE1, HSPA1A, HSPA5, and HSPCA), protease inhibitors (SKALP, TIMP1, and TIMP2), and proliferation markers (TOP2A, TGFA, RPS7, and IGFBP7). In addition, there are several keratins that are specific for epithelial cells such as keratin 4, 6A, 13, and 17 (Table 2). Sequence data from our SAGE libraries are publicly available (http://www.ncbi.nlm.nih.gov/SAGE/) with Geo Accession Omnibus nos. GSM757 and GSM2385 for GEJ cancer and GSM784 for normal stomach. Our quantitative real-time reverse transcription (RT) PCR for 11 overexpressed genes confirmed the SAGE data (Figure 2). However, we were not able to assess correlations with histopathology data or clinical outcomes because of the relatively small sample number in the pilot study.
Table 2.
Gene Overexpression in GEJ Adenocarcinomas Using Serial Analyses of Gene Expression.*
| TAG | UniGene Symbol | UniGene ID | Locus | Tag Number | Expression Ratio and P values | |||||
| GSM784 | GSM2385 | GSM757 | GSM2385 vs GSM784 | Pval GSM2385 | GSM757 vs GSM784 | Pval GSM757 | ||||
| GGTTTGGCTT | UQCRH | 285761 | 1p33 | 2 | 30 | 34 | 5.90 | <0.01 | 6.50 | <0.01 |
| CGCCGACGAT | G1P3 | 287721 | 1p35 | 2 | 57 | 62 | 11.20 | <0.001 | 11.90 | <0.001 |
| TTTCCTCTCA | SFN | 184510 | 1p35 | 0 | 27 | 27 | 10.60 | <0.001 | 10.30 | <0.001 |
| TTTGCACCTT | HMGN2 | 181163 | 1p36.1 | 0 | 9 | 17 | 3.50 | 0.073 | 6.50 | <0.01 |
| CTCTAAGAAG | C1QA | 9641 | 1p36.3 | 0 | 28 | 22 | 11.00 | <0.001 | 8.40 | <0.01 |
| AATCTGCGCC | G1P2 | 833 | 1p36.3 | 0 | 53 | 18 | 20.80 | <0.001 | 6.90 | <0.01 |
| GGCTGGGGGC | COAS3 | 352407 | 1q21.3 | 7 | 108 | 78 | 6.10 | <0.001 | 4.30 | <0.001 |
| GATCTCTTGG | S100A2 | 413843 | 1q21.3 | 0 | 2 | 46 | -1.30 | 0.741 | 17.60 | <0.001 |
| CAGGCCCCAC | S100A11 | 417004 | 1q21.3 | 0 | 27 | 32 | 10.60 | <0.001 | 12.20 | <0.001 |
| CCCCCTGGAT | S100A6 | 275243 | 1q21.3 | 2 | 83 | 57 | 16.30 | <0.001 | 10.90 | <0.001 |
| GAGCAGCGCC | S100A7 | 112408 | 1q21.3 | 0 | 0 | 75 | N/A | N/A | 28.70 | <0.001 |
| TACCTGCAGA | S100A8 | 416073 | 1q21.3 | 0 | 0 | 284 | N/A | N/A | 108.60 | <0.001 |
| GTGGCCACGG | S100A9 | 112405 | 1q21.3 | 0 | 0 | 334 | N/A | N/A | 127.70 | <0.001 |
| GGCTTCTAAC | SPR-2A | 505352 | 1q21.3 | 0 | 0 | 27 | N/A | N/A | 10.30 | <0.001 |
| AGCAGATCAG | S100A10 | 143873 | 1q21.3 | 1 | 48 | 95 | 18.80 | <0.001 | 36.30 | <0.001 |
| GGCTGGTCTG | MGC4677 | 446688 | 2p11.2 | 0 | 18 | 14 | 7.10 | <0.01 | 5.40 | 0.016 |
| GAAACCCCAG | IGKC | 306357 | 2p11.2 | 0 | 2 | 33 | -1.30 | 0.741 | 12.60 | <0.001 |
| GGGGAAATCG | THYB10 | 446574 | 2p11.2 | 6 | 88 | 104 | 5.80 | <0.001 | 6.60 | <0.001 |
| AGTTTGTTAG | TACSTD1 | 692 | 2p21 | 0 | 29 | 56 | 11.40 | <0.001 | 21.40 | <0.001 |
| TTGTTGTTGA | CALM2 | 425808 | 2p21 | 0 | 19 | 6 | 7.50 | <0.01 | 2.30 | 0.207 |
| TCATCTTTAT | RPS7 | 444012 | 2p25.3 | 0 | 16 | 0 | 6.30 | <0.01 | N/A | N/A |
| TTCCTGGTAG | U5-200 KD | 246112 | 2q11.2 | 1 | 22 | 11 | 8.60 | <0.01 | 4.20 | 0.177 |
| TAAATAATTT | HSPE1 | 1197 | 2q33.1 | 1 | 43 | 14 | 16.90 | <0.001 | 5.40 | 0.080 |
| ATCTTGTTAC | FN1 | 418138 | 2q35 | 0 | 34 | 12 | 13.30 | <0.001 | 4.60 | 0.030 |
| TCACAGTGCC | FLNB | 81008 | 3p14.3 | 1 | 26 | 2 | 10.20 | <0.01 | -1.30 | 1.385 |
| GTGTTAACCA | RPL15 | 74267 | 3p24.1 | 2 | 13 | 47 | 2.60 | 0.258 | 9.00 | <0.001 |
| CCCTCCCGAA | MUC13 | 5940 | 3q21 | 0 | 5 | 16 | 2.00 | 0.274 | 6.10 | <0.01 |
| ATCCTTGCTG | CSTA | 412999 | 3q21 | 1 | 3 | 87 | 1.20 | 1.132 | 33.30 | <0.001 |
| GGGACGAGTG | TM4SF1 | 351316 | 3q21 | 1 | 26 | 32 | 10.20 | <0.01 | 12.20 | <0.001 |
| ACCTTTACTG | TFRC | 185726 | 3q29 | 0 | 17 | 5 | 6.70 | <0.01 | 1.90 | 0.286 |
| CCTGGTCCCA | SH3BP2 | 167679 | 4p16.3 | 0 | 2 | 33 | -1.30 | 0.741 | 12.60 | <0.001 |
| CATATCATTA | IGFBP7 | 435795 | 4q12 | 1 | 41 | 50 | 16.10 | <0.001 | 19.10 | <0.001 |
| ACTAATCGTT | RPL37 | 80545 | 5p13.1 | 0 | 18 | 6 | 7.10 | <0.01 | 2.30 | 0.207 |
| TTCACTGTGA | TA-WDRP | 175596 | 5q22.1 | 4 | 96 | 108 | 9.40 | <0.001 | 10.30 | <0.001 |
| GTGACAACAC | VDAC1 | 404814 | 5q31.1 | 2 | 18 | 31 | 3.50 | 0.080 | 5.90 | <0.01 |
| ATGTGAAGAG | SPARC | 111779 | 5q31.3 | 1 | 65 | 75 | 25.50 | <0.001 | 28.70 | <0.001 |
| AATTTGCAAC | H2AFY | 75258 | 5q31.3 | 0 | 16 | 9 | 6.30 | <0.01 | 3.40 | 0.078 |
| CAGAGATGAA | HSPA1A | 75452 | 6p21.3 | 1 | 7 | 43 | 2.70 | 0.461 | 16.40 | <0.001 |
| ACCCTTTAAC | HLA-E | 381008 | 6p21.3 | 0 | 14 | 19 | 5.50 | 0.013 | 7.30 | <0.01 |
| AGCTTCTACC | HLA-A | 181244 | 6p21.3 | 0 | 0 | 24 | N/A | N/A | 9.20 | <0.001 |
| GGGCATCTCT | HLA-DRA | 409805 | 6p21.3 | 3 | 61 | 87 | 8.00 | <0.001 | 11.10 | <0.001 |
| GTACTGTGGC | CLIC1 | 414565 | 6p21.3 | 0 | 2 | 18 | -1.30 | 0.741 | 6.90 | <0.01 |
| AAGGCAATTT | TSPYL | 278479 | 6q22.1 | 1 | 21 | 3 | 8.20 | <0.01 | 1.10 | 1.153 |
| CCCAAGCTAG | HSPB1 | 76067 | 7q11.2 | 0 | 13 | 94 | 5.10 | 0.019 | 36.00 | <0.001 |
| TGCACAATAT | MUC3B | 129782 | 7q22 | 0 | 22 | 0 | 8.60 | <0.001 | N/A | N/A |
| GTTCCACAGA | COL1A2 | 232115 | 7q22 | 0 | 29 | 4 | 11.40 | <0.001 | 1.50 | 0.396 |
| CTGCCAAGTT | ZYX | 75873 | 7q34 | 0 | 11 | 19 | 4.30 | 0.038 | 7.30 | <0.01 |
| TGGGTGAGCC | CTSB | 135226 | 8p22 | 3 | 38 | 2 | 5.00 | <0.01 | 5.40 | <0.001 |
| GAACGCCTAA | DPYSL2 | 173381 | 8p22 | 0 | 25 | 8 | 9.80 | <0.001 | 3.10 | 0.108 |
| GCTAATAATG | SULF1 | 409602 | 8q13.1 | 0 | 18 | 5 | 7.10 | <0.01 | 1.90 | 0.286 |
| ATTATTTTTC | RPL7 | 421257 | 8q13.2 | 2 | 117 | 38 | 23.00 | <0.001 | 7.30 | <0.001 |
| TAAGTGGAAT | YWHAZ | 386834 | 8q23.1 | 1 | 28 | 7 | 11.00 | <0.01 | 2.70 | 0.482 |
| CACTTCAAGG | LY6E | 77667 | 8q24.3 | 0 | 12 | 29 | 4.70 | 0.027 | 11.10 | <0.001 |
| AGAAAGATGT | ANXA1 | 287558 | 9q21.1 | 0 | 56 | 138 | 22.00 | <0.001 | 52.80 | <0.001 |
| AGCTGTATTC | CKS2 | 83758 | 9q22.2 | 0 | 17 | 11 | 6.70 | <0.01 | 4.20 | 0.041 |
| TTTTCTGAAA | TXN | 395309 | 9q31 | 1 | 29 | 27 | 11.40 | <0.001 | 10.30 | 0.002 |
| TGCATCTGGT | HSPA5 | 310769 | 9q33.3 | 3 | 60 | 29 | 7.80 | <0.001 | 3.70 | 0.018 |
| TCCAAATCGA | VIM | 435800 | 10p13 | 0 | 3 | 29 | 1.20 | 0.532 | 11.10 | <0.001 |
| AAAATACTAG | DKK1 | 40499 | 10q21.1 | 0 | 17 | 0 | 6.70 | <0.01 | N/A | N/A |
| TGATAATTCA | MGC14697 | 171625 | 10q24.3 | 3 | 38 | 10 | 5.00 | <0.01 | 1.30 | 0.860 |
| TTTGGTTTTC | RAB22A | 21431 | 20q13.2 | 0 | 90 | 6 | 35.30 | <0.001 | 2.30 | 0.207 |
| ATATGTATAT | CD44 | 306278 | 11p13 | 0 | 16 | 8 | 6.30 | <0.01 | 3.10 | 0.108 |
| TCTTGTGCAT | LDHA | 2795 | 11p15.1 | 1 | 51 | 19 | 20.00 | <0.001 | 7.30 | 0.020 |
| TAATAAATGC | TTS-2.2 | 118463 | 11p15.5 | 0 | 31 | 7 | 12.20 | <0.001 | 2.70 | 0.150 |
| GGATTTGGCC | RPLP2 | 437594 | 11p15.5 | 25 | 316 | 164 | 5.00 | <0.01 | 2.50 | <0.01 |
| GCTGGTGCCT | THY1 | 134643 | 11q23.3 | 0 | 2 | 16 | -1.30 | 0.741 | 6.10 | <0.01 |
| ACAGGCTACG | TAGLN | 433401 | 11q23.3 | 1 | 0 | 28 | -2.50 | 0.405 | 10.70 | <0.01 |
| GTCTCCTAAT | RAI3 | 194691 | 12p13.1 | 0 | 18 | 10 | 7.10 | <0.01 | 3.80 | 0.050 |
| TGGTTTGAGC | NDUFA9 | 75227 | 12p13.3 | 0 | 5 | 22 | 2.00 | 0.274 | 8.40 | <0.001 |
| AAGATTGGTG | CD9 | 387579 | 12p13.3 | 0 | 10 | 53 | 3.90 | 0.050 | 20.30 | <0.001 |
| GAAGCACAAG | KRT6A | 334309 | 12q13.1 | 0 | 0 | 15 | N/A | N/A | 5.70 | 0.011 |
| GGCAGAGAAG | KRT4 | 371139 | 12q13.1 | 0 | 0 | 29 | N/A | N/A | 11.10 | <0.001 |
| CTGTTGATTG | HNRPA1 | 356721 | 12q13.1 | 6 | 123 | 57 | 8.00 | <0.001 | 3.60 | <0.001 |
| ATGTAAAAAA | LYZ | 234734 | 12q15 | 5 | 75 | 378 | 5.90 | <0.001 | 28.90 | <0.001 |
| ACTCCAAAAA | RNP24 | 75914 | 12q24.3 | 5 | 116 | 6 | 9.10 | <0.001 | -2.20 | 1.831 |
| TGAAAGTGTG | HSP105B | 36927 | 13q12.2 | 0 | 28 | 11 | 11.00 | <0.001 | 4.20 | 0.041 |
| TAATTTTTGC | GW112 | 273321 | 13q14.3 | 0 | 0 | 211 | N/A | N/A | 80.70 | <0.001 |
| CATCTGTACT | PCDH20 | 391781 | 13q21.2 | 0 | 0 | 19 | N/A | N/A | 7.30 | <0.001 |
| TTCACTGTGA | LGALS3 | 411701 | 14q22 | 4 | 96 | 108 | 9.40 | <0.001 | 10.30 | <0.001 |
| TACTAGTCCT | HSPCA | 446579 | 14q32.3 | 4 | 121 | 32 | 11.90 | <0.001 | 12.20 | <0.001 |
| GAAATAAAGC | IGHG3 | 413826 | 14q32.3 | 12 | 302 | 389 | 9.90 | <0.001 | 12.40 | <0.001 |
| CAAACTAACC | IGHM | 153261 | 14q32.3 | 2 | 2 | 38 | -2.50 | 1.720 | 7.30 | <0.001 |
| CAGGAGGAGT | GRP58 | 308709 | 15q15.3 | 0 | 1 | 16 | -2.50 | 1.032 | 6.10 | <0.01 |
| CTTCCAGCTA | ANXA2 | 437110 | 15q22.2 | 2 | 27 | 29 | 5.30 | <0.01 | 5.50 | <0.01 |
| TACTTGTGTG | SDFR1 | 389371 | 15q24.1 | 1 | 23 | 5 | 9.00 | <0.01 | 1.90 | 0.762 |
| GCGACCGTCA | ALDOA | 273415 | 16p11.2 | 1 | 0 | 52 | -2.50 | 0.405 | 19.90 | <0.001 |
| CCCCCTGCAG | MSLN | 408488 | 16p13.3 | 0 | 105 | 5 | 41.20 | <0.001 | 1.90 | 0.286 |
| ACCGCCGTGG | CYBA | 68877 | 16q24 | 0 | 0 | 46 | N/A | N/A | 17.60 | <0.001 |
| CCCAGAGCTC | HSD17B2 | 155109 | 16q24 | 0 | 16 | 3 | 6.30 | <0.01 | 1.10 | 0.548 |
| AACTAATACT | MGC40157 | 270232 | 17p11.2 | 6 | 108 | 14 | 7.10 | <0.001 | -1.10 | 1.287 |
| GAAACCCCAG | HSA011916 | 84359 | 17p13 | 0 | 2 | 33 | -1.30 | 0.741 | 12.60 | <0.001 |
| ATAGACATAA | C1QBP | 78614 | 17p13 | 1 | 24 | 8 | 9.40 | <0.01 | 3.10 | 0.378 |
| GATCAATCAG | CCL18 | 16530 | 17q21?? | 0 | 16 | 3 | 6.30 | <0.01 | 1.10 | 0.548 |
| CTTCCTTGCC | KRT17 | 2785 | 17q21.2 | 0 | 2 | 53 | -1.30 | 0.741 | 20.30 | <0.001 |
| AAAGCGGGGC | KRT13 | 433871 | 17q21.2 | 0 | 0 | 73 | N/A | N/A | 27.90 | <0.001 |
| CTGTTCCGGC | PPP1R1B | 286192 | 17q21.2 | 0 | 0 | 10 | N/A | N/A | 4.00 | 0.050 |
| CTCAGCAATG | TOP2A | 156346 | 17q21.2 | 0 | 4 | 5 | 1.60 | 0.700 | 2.00 | 0.050 |
| GTGTGGGGGG | JUP | 2340 | 17q21.2 | 1 | 32 | 28 | 12.60 | <0.001 | 10.70 | <0.001 |
| TTCGGTTGGT | COL1A1 | 172928 | 17q21.3 | 1 | 66 | 41 | 25.90 | <0.001 | 15.70 | <0.001 |
| TCTCCAGGAA | CGI-69 | 237924 | 17q21.3 | 0 | 21 | 7 | 8.20 | 0.001 | 2.70 | 0.150 |
| ATGCTCCCTG | LGALS3BP | 79339 | 17q25.3 | 0 | 0 | 18 | N/A | N/A | 6.90 | <0.01 |
| TCTCTGATGC | TIMP2 | 6441 | 17q25.3 | 1 | 40 | 23 | 15.70 | <0.001 | 8.80 | <0.01 |
| CAACTTAGTT | MLC-B | 233936 | 18p11.3 | 2 | 19 | 29 | 3.70 | 0.062 | 5.50 | <0.01 |
| CCTCCTATTA | RIOK3 | 209061 | 18q11.2 | 0 | 2 | 22 | -1.30 | 0.741 | 8.40 | <0.01 |
| ACCCCCCCGC | JUND | 2780 | 19p13.1 | 0 | 1 | 19 | -2.50 | 1.032 | 7.30 | <0.01 |
| AGAGGGTGGG | DNAJB1 | 82646 | 19p13.1 | 1 | 30 | 36 | 11.80 | <0.001 | 13.80 | <0.001 |
| CCTCCACCTA | PRDX2 | 432121 | 19p13.2 | 2 | 28 | 15 | 5.50 | <0.01 | 2.90 | 0.178 |
| CGAGGGGCCA | ACTN4 | 443619 | 19q13.2 | 1 | 1 | 39 | -2.50 | 1.613 | 14.90 | <0.001 |
| CGACCCCACG | APOE | 169401 | 19q13.2 | 0 | 22 | 51 | 8.60 | <0.001 | 19.50 | <0.001 |
| TGGCCCCAGG | APOC1 | 268571 | 19q13.2 | 0 | 34 | 56 | 13.30 | <0.001 | 21.40 | <0.001 |
| GTACACACCC | CST1 | 123114 | 20p11.21 | 0 | 1 | 19 | -2.50 | 1.032 | 7.30 | <0.01 |
| GTACACACAC | CST4 | 56319 | 20p11.21 | 0 | 36 | 3 | 14.10 | <0.001 | 1.10 | 0.548 |
| TGTTCCACTC | ENTPD6 | 438431 | 20p11.21 | 0 | 26 | 4 | 10.20 | <0.001 | 1.50 | 0.396 |
| AATGTGAGTC | C20orf129 | 70704 | 20q11.23 | 0 | 20 | 17 | 7.80 | <0.01 | 6.50 | <0.01 |
| TTGAATCCCC | PI3 | 112341 | 20q13.1 | 0 | 0 | 62 | N/A | N/A | 23.70 | <0.001 |
| GCCCCCAATA | LGALS1 | 407909 | 22q13.1 | 2 | 38 | 26 | 7.50 | <0.001 | 5.00 | 0.012 |
| GAATTTTATA | BZRP | 202 | 22q13.2 | 1 | 22 | 9 | 8.60 | <0.01 | 3.40 | 0.295 |
| GAGAGTGTCT | TIMP1 | 446641 | Xp11.2 | 0 | 2 | 20 | -1.30 | 0.741 | 7.60 | <0.01 |
| ATTATCCAGG | RBM3 | 301404 | Xp11.2 | 0 | 18 | 0 | 7.10 | <0.01 | N/A | N/A |
| CCCCCACCTA | PLP2 | 77422 | Xp11.2 | 0 | 23 | 5 | 9.00 | <0.001 | 1.90 | 0.286 |
| TATGTGTGCT | SYTL4 | 376981 | Xq22.1 | 0 | 18 | 0 | 7.10 | <0.01 | N/A | N/A |
| GTTAACGTCC | RPL36A | 432485 | Xq22.1 | 2 | 34 | 17 | 6.70 | <0.01 | 3.30 | 0.112 |
| ATAGAGGCAA | MRGX | 411358 | Xq22.2 | 0 | 19 | 10 | 7.50 | <0.01 | 3.80 | 0.050 |
Analyses were performed using eSAGE 1.2a software [19,20].
GSM784 = normal gastric mucosa; GSM2385 = gastroesophageal adenocarcinoma; GSM757 = gastroesophageal adenocarcinoma; Pval = P value.*
Figure 2.
Frequency histogram of gene overexpression in UGC (Upper Gastrointestinal Carcinoma). Quantitative real-time RT-PCR analyses of 11 representative genes detected in SAGE analyses (LGALS3BP, PPP1R1B, HSPA5, TACSTD1, ANXA1, TOP2A, S100A6, S100A7, S100A8, S100A9, and S100A10). The gene expression in 20 primary GEJ cancers was compared to 13 normal gastric epithelial samples. Gene overexpression was considered at a ratio ≥ 5.0. The Overexpression fold was calculated as described earlier [21]. The vertical bars demonstrate the frequency of overexpression of the corresponding gene in the GEJ cancer samples that we studied.
We mapped gene expression alterations to chromosomal positions using the University of California Santa Cruz's November freeze 2002 assembly of the human genome sequence (http://genome.ucsc.edu) (Table 2). We Compared the DNA amplification regions to the gene expression mapping. The chromosome mapping of altered genes revealed clustering of several genes to small chromosomal domains, suggesting a high order of organization of the cancer genome. Some chromosomal arms that had frequent DNA changes had also frequent gene expression alterations such as chromosomes 1 (15 genes), 2 (9 genes), 6 (6 genes), 11 (6 genes), 12 (8 genes), and 17 (13 genes) (Table 2; Figure 1). Overexpressed genes clustered at specific chromosomal positions such as 1q21.3 (9 genes), 6p21.3 (5 genes), and 17q21 (8 genes). The gene expression profile indicated that although the amplification regions are often large, the expressed genes are clustered and mapped to small chromosomal regions such as 1q21.3 and 17q21.2. This observation indicates that the majority of genes located in areas involved in chromosomal amplifications remain highly regulated and only few critical genes may be deregulated and overexpressed. Although the DNA amplification is one mechanism responsible for the expression changes, other cellular mechanisms of gene regulation are often involved (Table 2; Figure 1).
Despite the genome-wide chromosomal instability in GEJ carcinomas, DNA gains and amplifications mapped to specific regions in the chromosomes such as 1q, 6q, and 17q. When we reviewed the DNA changes that were reported in 73 tumor types from 283 reports [4], we observed that DNA gains/amplifications map to chromosomal regions different from losses in most human cancer types. Thus, the DNA alterations are not randomly distributed but have a rather unique distribution over the chromosomal domains.
Our analyses of the transcriptome in GEJ cancer demonstrated clusters of overexpressed genes in a number of early-replicating chromatin (Giemsa light bands) chromosomal domains such as 1q21.3, 6p21.3, and 17q21.2 (Figure 1). A recent comprehensive study of the human transcriptome map demonstrated a similar clustering of highly expressed genes in chromosomal domains [13]. There are growing evidences that chromosomes occupy discrete CTs in the cell nucleus. The compartment for gene-dense, early-replicating chromatin (Giemsa light bands) is separated from the compartments for mid- to late-replicating chromatin (Giemsa dark bands) [14,27]. A novel theory for explaining gene expression has been recently explored where the transcriptional status of genes correlates with gene positioning in CTs where dynamic repositioning of genes with respect to centromeric heterochromatin has a role in gene silencing and activation [14,27]. Therefore, it is possible that the effect of DNA changes on gene expression alterations may not be limited to their respective gene copy numbers but also to the overall impact of the massive DNA amplifications on the chromatin repositioning in the nucleus. The existence of other regulatory mechanisms such as upstream gene regulation should not be overlooked.
Our results provide a comprehensive tool to search for DNA changes and genes that are overexpressed in GEJ carcinoma. The presence of large amplification areas, yet clustering of overexpressed genes to relatively small loci, may suggest a high organization of chromatin and cancer-related genes in the nucleus. The impact of massive DNA changes that we detected in GEJ carcinomas on the nuclear organization of the chromatin and the repositioning of genes in CT requires further investigation. Moreover, the transcriptome data provide us, as well as others, the opportunity to develop functional and cell biology assays for particular genes of interest that may serve as prognostic or therapeutic targets. This is expected to add to the overall understanding of the biology of this genetically complex and deadly cancer.
Footnotes
This work was supported by a grant award from the National Cancer Institute (1 R01 CA106176 to W.E.R.) and by the Cancer Center and Research and Development funds at the University of Virginia. The contents of this work are solely the responsibility of the authors and do not necessarily represent the official views of the National Cancer Institute or the University of Virginia.
References
- 1.Blot WJ, Devesa SS, Kneller RW, Fraumeni JF., Jr Rising incidence of adenocarcinoma of the esophagus and gastric cardia. JAMA. 1991;265:1287–1289. [PubMed] [Google Scholar]
- 2.Locke GRR, Talley NJ, Carpenter HA, Harmsen WS, Zinsmeister AR, Melton LJR. Changes in the site- and histology-specific incidence of gastric cancer during a 50-year period. Gastroenterology. 1995;109:1750–1756. doi: 10.1016/0016-5085(95)90740-8. [DOI] [PubMed] [Google Scholar]
- 3.El-Rifai W, Powell S. Molecular and biologic basis of upper gastrointestinal malignancy gastric carcinoma. Surg Oncol Clin North Am. 2002;11:273–291. doi: 10.1016/s1055-3207(02)00004-2. [DOI] [PubMed] [Google Scholar]
- 4.Knuutila S, Autio K, Aalto Y. Online access to CGH data of DNA sequence copy number changes. Am J Pathol. 2000;157:689. doi: 10.1016/S0002-9440(10)64579-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Velculescu VE, Zhang L, Vogelstein B, Kinzler KW. Serial analysis of gene expression. Science. 1995;270:484–487. doi: 10.1126/science.270.5235.484. [DOI] [PubMed] [Google Scholar]
- 6.Velculescu VE, Vogelstein B, Kinzler KW. Analysing uncharted transcriptomes with SAGE. Trends Genet. 2000;16:423–425. doi: 10.1016/s0168-9525(00)02114-4. [DOI] [PubMed] [Google Scholar]
- 7.St Croix B, Rago C, Velculescu V, Traverso G, Romans KE, Montgomery E, Lal A, Riggins GJ, Lengauer C, Vogelstein B, Kinzler KW. Genes expressed in human tumor endothelium. Science. 2000;289:1197–1202. doi: 10.1126/science.289.5482.1197. [DOI] [PubMed] [Google Scholar]
- 8.Parle-McDermott A, McWilliam P, Tighe O, Dunican D, Croke DT. Serial analysis of gene expression identifies putative metastasis-associated transcripts in colon tumour cell lines. Br J Cancer. 2000;83:725–728. doi: 10.1054/bjoc.2000.1330. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Culp LA, Holleran JL, Miller CJ. Tracking prostate carcinoma micrometastasis to multiple organs using histochemical marker genes and novel cell systems. Histol Histopathol. 2001;16:945–953. doi: 10.14670/HH-16.945. [DOI] [PubMed] [Google Scholar]
- 10.Argani P, Rosty C, Reiter RE, Wilentz RE, Murugesan SR, Leach SD, Ryu B, Skinner HG, Goggins M, Jaffee EM, Yeo CJ, Cameron JL, Kern SE, Hruban SE. Discovery of new markers of cancer through serial analysis of gene expression: prostate stem cell antigen is overexpressed in pancreatic adenocarcinoma. Cancer Res. 2001;61:4320–4324. [PubMed] [Google Scholar]
- 11.Hough CD, Sherman-Baust CA, Pizer ES, Montz FJ, Im DD, Rosenshein NB, Cho KR, Riggins GJ, Morin PJ. Large-scale serial analysis of gene expression reveals genes differentially expressed in ovarian cancer. Cancer Res. 2000;60:6281–6287. [PubMed] [Google Scholar]
- 12.Seth P, Krop I, Porter D, Polyak K. Novel estrogen and tamoxifen induced genes identified by SAGE (serial analysis of gene expression) Oncogene. 2002;21:836–843. doi: 10.1038/sj.onc.1205113. [DOI] [PubMed] [Google Scholar]
- 13.Caron H, van Schaik B, van der Mee M, Baas F, Riggins G, van Sluis P, Hermus MC, van Asperen R, Boon K, Voute PA, Heisterkamp S, van Kampen A, Versteeg R. The human transcriptome map: clustering of highly expressed genes in chromosomal domains. Science. 2001;291:1289–1292. doi: 10.1126/science.1056794. [DOI] [PubMed] [Google Scholar]
- 14.Cremer T, Cremer C. Chromosome territories, nuclear architecture and gene regulation in mammalian cells. Nat Rev Genet. 2001;2:292–301. doi: 10.1038/35066075. [DOI] [PubMed] [Google Scholar]
- 15.Hahn SA, Seymour AB, Hoque AT, Schutte M, da Costa LT, Redston MS, Caldas MS, Weinstein CL, Fischer A, Yeo CJ. Allelotype of pancreatic adenocarcinoma using xenograft enrichment. Cancer Res. 1995;55:4670–4675. [PubMed] [Google Scholar]
- 16.Yustein AS, Harper JC, Petroni GR, Cummings OW, Moskaluk CA, Powell SM. Allelotype of gastric adenocarcinoma. Cancer Res. 1999;59:1341–1437. [PubMed] [Google Scholar]
- 17.Varis A, Wolf M, Monni O, Vakkari ML, Kokkola A, Moskaluk C, Frierson H, Jr, Powell SM, Knuutila S, Kallioniemi A, El-Rifai W. Targets of gene amplification and overexpression at 17q in gastric cancer. Cancer Res. 2002;62:2625–2629. [PubMed] [Google Scholar]
- 18.Miettinen MM, El-Rifai W, Sarlomo-Rikala M, Andersson LC, Knuutila S. Tumor size-related DNA copy number changes occur in solitary fibrous tumors but not in hemangiopericytomas. Mod Pathol. 1997;10:1194–1200. [PubMed] [Google Scholar]
- 19.Margulies EH, Kardia SL, Innis JW. A comparative molecular analysis of developing mouse forelimbs and hindlimbs using serial analysis of gene expression (SAGE) Genome Res. 2001;11:1686–1698. doi: 10.1101/gr.192601. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Margulies EH, Innis JW. eSAGE: managing and analysing data generated with serial analysis of gene expression (SAGE) Bio-informatics. 2000;16:650–651. doi: 10.1093/bioinformatics/16.7.650. [DOI] [PubMed] [Google Scholar]
- 21.El-Rifai W, Smith MF, Jr, Li G, Beckler A, Carl VS, Montgomery E, Knuutila S, Moskaluk CA, Frierson HF, Jr, Powell SM. Gastric cancers overexpress DARPP-32 and a novel isoform, t-DARPP. Cancer Res. 2002;62:4061–4064. [PubMed] [Google Scholar]
- 22.van Dekken H, Geelen E, Dinjens WN, Wijnhoven BP, Tilanus HW, Tanke HJ, Rosenberg C. Comparative genomic hybridization of cancer of the gastroesophageal junction: deletion of 14Q31-32.1 discriminates between esophageal (Barrett's) and gastric cardia adenocarcinomas. Cancer Res. 1999;59:748–752. [PubMed] [Google Scholar]
- 23.Lee S, Baek M, Yang H, Bang YJ, Kim WH, Ha JH, Kim DK, Jeoung DI. Identification of genes differentially expressed between gastric cancers and normal gastric mucosa with cDNA microarrays. Cancer Lett. 2002;184:197–206. doi: 10.1016/s0304-3835(02)00197-0. [DOI] [PubMed] [Google Scholar]
- 24.Barrett MT, Yeung KY, Ruzzo WL, Hsu L, Blount PL, Sullivan R, Zarbl H, Delrow J, Rabinovitch PS, Reid BJ. Transcriptional analyses of Barrett's metaplasia and normal upper GI mucosae. Neoplasia. 2002;4:121–128. doi: 10.1038/sj.neo.7900221. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Hippo Y, Taniguchi H, Tsutsumi S, Machida N, Chong JM, Fukayama M, Kodama T, Aburatani H. Global gene expression analysis of gastric cancer by oligonucleotide microarrays. Cancer Res. 2002;62:233–240. [PubMed] [Google Scholar]
- 26.El-Rifai W, Frierson HJ, Harper J, Powell S, Knuutila S. Expression profiling of gastric adenocarcinoma using cDNA array. Int J Cancer. 2001;92:828–832. doi: 10.1002/ijc.1264. [DOI] [PubMed] [Google Scholar]
- 27.Dundr M, Misteli T. Functional architecture in the cell nucleus. Biochem J. 2001;356:297–310. doi: 10.1042/0264-6021:3560297. [DOI] [PMC free article] [PubMed] [Google Scholar]