| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Molecular Biology and Genetics |
Departments of Oncology [B. R., J. J., G. P., R. H. H., S. E. K.], Biostatistics [N. J. B., G. P.], and Pathology [R. H. H., G. P., S. E. K.], The Johns Hopkins Medical Institutions, Baltimore, Maryland 21231, and The Eppley Institute, The University of Nebraska Medical Center, Omaha, Nebraska 68198 [M. A. H.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
28,200 people are diagnosed and die of pancreatic cancer (1)
. The mortality rate is the highest among cancer types, in part because of the asymptomatic nature of the disease in early stages, a lack of sensitive and specific diagnostic tools, and limited progress in development of effective therapeutics. Better knowledge of changes of gene expression that accompany pancreatic cancer may suggest new screening tools and therapeutic strategies. Several genes that are overexpressed in pancreatic cancer have been identified by subtractive and comparative methods (2, 3, 4, 5, 6, 7, 8, 9, 10)
. Initial surveys of these tumors and cell lines by RNA-based gene expression analysis have been reported (11, 12, 13)
. A series of genetic changes within pancreatic ductal epithelium accompanies the development of precursor lesions, termed pancreatic intraepithelial neoplasia (15) , some of which progress to pancreatic adenocarcinoma. SAGE3 was used to compare the gene expression in short-term cultures of normal pancreatic epithelium to cultures of pancreatic carcinoma cells using data from primary tumors to filter the data from cultured cells. SAGE technology, developed by Velculescu et al. (14) provides a simultaneous and comprehensive enumeration of gene transcripts of a given sample in a quantitative manner, whereas it also provides sequence information that is used to identify the differentially expressed genes. Here, we report differentially expressed genes identified from the comparative SAGE profiling of 294,920 tags in 10 pancreatic samples.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Panc-1 and Hs766T cell lines were cultured in DMEM supplemented with 10% fetal bovine serum and antibiotics (100 units/ml penicillin and 100 µg/ml streptomycin). CAPAN1 and CAPAN2 cell lines were cultured in RPMI 1640 and McCoy’s growth medium (Life Technologies, Inc., Gaithersburg, MD) supplemented with 10% fetal bovine serum and antibiotics (100 units/ml penicillin and 100 µg/ml streptomycin), respectively. Use of different medium was required to minimize the variance in growth rates that would otherwise be exaggerated with the use of a single medium.
The cell lines samples AsPc1 and PL45, and the two primary pancreatic ductal adenocarcinomas were described (19) . The isolation and validation of normal pancreatic ductal short-term cultures HX, H126, H48, and H110 were described (17) .
SAGE.
Total cellular RNA was obtained from pancreatic cancer cells, 90% confluent CAPAN1, CAPAN2, Hs766T, and Panc-1 cell line cultures, using TRIZOL reagent (Life Technologies, Inc.). Polyadenylated mRNA was purified from total RNA (Messagemaker; Life Technologies, Inc.), and cDNA was generated (cDNA Synthesis System; Life Technologies, Inc.). SAGE was performed as described by Velculescu et al. (14)
for all of the pancreatic cancer libraries. For the generation of the two normal pancreatic ductal epithelial cell line (HX and H126) libraries, MicroSAGE, a SAGE technique modified for limited sample sizes (20)
, was used with a slight modification. In brief, total RNA was prepared from HX and H126 using TRIZOL (Life Technologies, Inc.) instead of direct mRNA isolation from cells as described originally in the MicroSAGE protocol (20)
. A modified lysis/binding buffer was prepared from 1 ml of lysis/binding buffer [100 mM Tris-HCl (pH 7.5), 500 mM LiCl, 10 mM EDTA, 1% LiDS, 5 mM DTT] (Dynabead mRNA direct kit; Dynal, Oslo, Norway) by addition of 66 µg tRNA and 10 µg of BSA. Total RNA (5 µg) was dissolved in 1-ml modified lysis/binding buffer. mRNA was purified using Oligo(dT)25 Dynabeads, and SAGE libraries were constructed. As part of the CGAP (NIH) SAGE consortium, all six of the SAGE libraries were arrayed at the Lawrence Livermore National Laboratories and Washington University Human Genome Center. The SAGE library data were posted at the CGAP website4
as part of the SAGEmap database (21
, 22)
. SAGE data of other tumor types were obtained from this database for comparison. SAGE libraries of two pancreatic cancer cell lines (AsPc1 and PL45) and from primary pancreatic cancer tissues (91-16113 and 96-6252) were obtained from earlier efforts (19)
.
Statistical Analysis.
Partek Pro2000 (Partek Inc., St. Louis, MO), Cluster and TreeView,5
SigmaStat 2.03 (SPSS Science, Chicago, IL), Access, and Excel (Microsoft, Seattle, WA) and R6
programs were used. For all of the statistical analysis beyond the initial description of datasets, SAGE data were normalized to tags per 100,000. A subset of the 10 SAGE libraries was obtained using Cluster and TreeView programs by filtering to require each tag to have one observation in > 3 of the 10 libraries, which produced a total of 6,245 unique tags. This filtered subset of data was used for all of the additional analysis. PCA analysis (23)
was carried out on the filtered subset of data using Partek Pro2000 software to graphically plot the three major components and Cluter/TreeView to list the genes of each component.
RT-PCR.
Constant amounts (1.0 µg) of total RNA from pancreatic cancer cell lines (CAPAN1, CAPAN2, PL45, and AsPc1) and two normal pancreatic ductal epithelial cells (H48 and H110, which were derived from pancreas of a 16-year-old male and a 17-year-old male, respectively, and cultured short-term; Ref. 17
), were used. Normal duct samples HX and H126, which were used for SAGE, were not available because of the small quantities of such cultures. Reverse-transcription was performed in a total reaction volume of 20 µl using Oligo(dT)25 primers and the SuperScript First-Strand Synthesis System for RT-PCR kit (Life Technologies, Inc.) according to the manufacturer’s protocol. The products were serially diluted and used for subsequent PCRs. Primers for corresponding genes were selected from mRNA sequences obtained from GenBank.7
Optimal PCR cycle numbers for each gene were determined by empirical identification of a range of cycles that produced exponential accumulation of amplified DNA on examination of a series of reactions comprising 25–40 PCR cycles. Aliquots (10 µl) of RT-PCR products were separated by electrophoresis in 1.5% or 2% agarose gels, depending on the product size, and were stained with ethidium bromide. Primer sequences and numbers of temperature cycles were: S100A4 (32 cycles): sense 5'-CCCCTCTCTACAACCCTCTC, antisense 5'-AGCACGTGTCTGAAGGAGCC; TSPAN-1 (36 cycles); sense 5'-ACTGTCGTCCAGTGCCATGC, antisense 5'-TAGCCCCAAGTCTGGAGCAG; CEACAM6 (32 cycles); sense 5'-CCTGCAGATTGCATGTCCCC, antisense 5'-GTCCTATTGAGGCCAGTGCC; ALG-2 (35 cycles): sense 5'-GACACCGAGCTTCAGCAAGC, antisense 5'-CACCTGTGCTCCATTCCCTC; and glyceraldehyde-3-phosphate dehydrogenase (29 cycles): sense 5'-GGCACCGTCAAGGCTGAGAA, antisense 5'-GAGACCACCTGGTGCTCAGT.
|
RESULTS AND DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
. A total of 294,920 tags were sequenced, and 86,658 (28.5%) were identified as unique tags. The average number of unique tags per library was 8,666, with a range of 4,127–11,711. Transcripts found three times or less in each library comprised 87.1% of the unique tags identified in this study, but this low-abundance class represented only 39% of total mRNA mass as judged by numbers of total tags. The number of unique tags that matched genes was 77,746 (90%) after elimination of mitochondrial DNA sequence, repetitive DNA sequences, and correction for the estimated SAGE tag sequence error rate (6.8%, attributable to sequencing errors; see Ref. 14
). These parameters were similar to those reported in similar SAGE datasets of other tissue types (19
, 24)
.
|
. A more informative means of displaying the key correlations is by scatter plot (Fig. 1B)
. Scatter plot analysis visualizes gene expression profiles in two dimensions, and the Pearson correlation coefficients yield quantitative comparisons between the gene expression profiles. The relationships suggested by hierarchical clustering were in broad accordance with those shown by the pairwise comparisons of the Pearson correlation coefficients (Table 2)
. As expected, the normal ductal epithelia HX and H126 were very similar in gene expression profiles (Fig. 1B
, D1 versus D2; Table 2
, r = 0.981). Such a resemblance between samples of independent derivation provided a measure of the reliability of the SAGE technique. Cell line Hs766T had the highest degree of correlation to the normal pancreatic ductal epithelium (r = 0.935 for HX and 0.930 for H126) followed in degree of similarity by the cell lines CAPAN2, CAPAN1, and Panc-1 (r = 0.797, 0.748, and 0.515, comparison to HX). This relationship was also seen in the scatter plot analysis (Fig. 1B
, D1–2 versus CL1). The two primary pancreatic cancer tissues, 91–16113 and 96–6252, were highly correlated with each other (r = 0.980), as expected. Pancreatic cancer cell line AsPc1 had the highest similarity of expression profile to the primary cancer tissues (r = 0.770 and 0.778 for 91-16113 and 96-6252, respectively). Scatter plots confirm the larger degrees of difference among the latter comparisons (Fig. 1B
, D1–2 versus CL2–6 and D1–2 versus I1–2).
|
|
Validation of SAGE Using RT-PCR.
To validate the differential expression of candidate genes identified by mapping of gene identities to the SAGEtag results, we performed semiquantitative RT-PCR analysis on two short-term normal pancreatic ductal cultures (H48 and H116) and four pancreatic cancer cell lines (AsPc1, CAPAN1, CAPAN2, and PL45). These two additional pancreatic ductal epithelial samples, which were derived from different individuals, were used because of the limited quantity of HX and H126 primary ductal epithelial samples. This validation thereby also served as a test of the generality of the results. Eight genes from among those up-regulated (Table 3)
were selected and assayed by RT-PCR. Differential expression was confirmed for four genes (S100A4, TSPAN-1, CEACAM6, and ALG-2) as shown in Fig. 2
. Interestingly, the degree of differential expression between normal ductal cells and pancreatic adenocarcinoma cells detected by RT-PCR approximated the differences observed in their respective SAGE tag counts. Lack of validation of some genes (keratin 19, claudin 4, basic transcription factor 3, and adenylyl cyclase-associated protein) could in part be attributable to the statistical false discovery rate (18%, see below), the incomplete current state of tag-to-gene mapping, and a lack of uniformity of gene expression among samples of different patients (a test of generality).
|
|
We used a four-step data reduction algorithm to identify genes differentially expressed between pancreatic cancer cells and normal ductal epithelium. First, we tested for tags exhibiting consistent differences (Student’s t test) between two groups of samples, the two normal pancreas ductal epithelial cells (HX and H126) versus the five "non-normoid" pancreatic cancer cells (AsPc1, CAPAN1, CAPAN2, Panc-1, and PL45). A total of 669 tags that had a P < 0.02 were identified. Among the 669 tags, 376 tags were up-regulated, and 293 tags were down-regulated in pancreatic cancer cells as compared with the normal duct cells. Second, we retained only the tags that expressed differences of 
10-fold. To calculate fold difference for each unique tag, the average of normalized tags from the cancer cell lines was divided by the average of normalized tags from the two normal duct cultures. For the tags that had no occurrences in a sample set, the arbitrary number 1 was assigned for this calculation. Third, we filtered out any tags not expressed in at least one primary tumor sample at a level >two tags/100,000 (this was performed only in the identification of overexpressed tags). The latter criterion was applied to reduce the possibility of identifying genes that had experienced induction because of cell culture. Finally, we set a cutoff to draw the greatest attention to genes of higher expression levels, requiring expression at an average of 12 tags/100,000 in the five cancer cell lines (for the genes up-regulated in cancer) or in the two normal duct samples (for the genes down-regulated). This algorithm identified a group of 86 tags that exhibited a robust overexpression or underexpression in pancreatic cancer cells (Table 3)
. The genes, tissue-type plasminogen activator, cathepsin H, and CEACAM6, which are known to be up-regulated in pancreatic cancer (10
, 12 , 13)
, were confirmed in this study (Table 3)
, validating our current approach to identify differentially expressed genes.
A permutation procedure was performed to estimate the false discovery rate. There are 21 possible permutations for which two libraries could be considered the comparison set (corresponding in form to the two ductal libraries of the original analysis). We evaluated the four-step process for each of these assignment choices. A total of 330 tags emerged from these pseudo-trials, which included the 86 of the original permutation. Thus, the average number of tags produced by these trials was 16, a conservative estimate of the number of false-positive tags to be expected under the null hypothesis of no real difference between pancreatic cancer and normal ductal expression profiles. The false discovery tag estimate of 16 tags is 18% of the observed tag count for the original analysis.
A potential source of bias could be the known variable GC content bias present in most SAGE libraries (27)
. The following approaches excluded this bias as a significant source of artifactual results in our study. First, the GC content of the two comparison libraries for each of our 21 permutations (above) was found not to correspond to the number of tags produced in each permutation. Second, the set of 86 observed differential tags retained a spectrum of GC content (Table 3)
. Third, the genes that failed to conform by RT-PCR did not represent a skewed GC content as compared with those that were confirmed. Thus, whereas one must consider variable ditag melting and the resultant enrichment of GC content as an uncontrolled determinant in analysis of SAGE data, this did not appear to exert a major effect in the assessment of differential gene expression.
The NCBI SAGEmap website8
provides the X-profiler program to reduce SAGE data (21
, 22)
. Using this additional tool, we also identified genes differentially expressed in pancreatic cancer cells as compared with normal control tissues or cultures of pancreas and other organs that were available in the SAGEmap database as of March 26, 2001. Several genes in Table 3
were consistently identified as differentially expressed using X-profiler (Table 3)
.
Biological and Clinical Implications for Pancreatic Cancer.
A notable feature of the overexpressed genes identified in this study is that nearly half of the genes comprise secretory, cell-surface, transmembrane, and tight junction protein coding genes. This could correspond to altered cellular attachments and cell surface architecture, resulting in aberrant cell-cell interactions that are a reproducible characteristic of cancer cells. One reason to explore such alterations would be to develop new therapeutic strategies. Another use is suggested by the detection of secretory proteins such as HE4 (a putative ovarian cancer marker; Refs. 24
, 28
), PSCA (a putative prostate cancer marker; Ref. 29
), and CEACAM6, which suggest a use as potential diagnostic markers. Indeed, one of the markers, PSCA, identified in this study was subsequently developed as a histological marker of pancreatic malignancy (30)
and is a secreted protein detectable in the serum of a set of prostate cancer patients evaluated (29)
. Interestingly, a group of ion-homeostasis related proteins, especially those specific for the calcium ion (Ca2+) such as S100A4, S100A10, Trop-2, AIF-1, and ALG-2, were identified as overexpressed. Another example is the major vault protein, which functions to produce multidrug resistance in a cancer cell (31)
; overexpression of this gene in cancer is reported to predict the response to chemotherapy in several tumor types (32)
. A group of genes down-regulated in pancreatic cancer was also identified (Table 3)
.
Differences between Pancreatic Cell Lines and Normal Epithelium: The Question of the Minimal Deviation Malignant Profile.
The Pearson correlation coefficient (Table 2)
, hierarchical clustering, and scatter plot analysis (Fig. 1, A and B)
classified Hs766T as closely resembling normal ductal epithelium in its expression profile: a "normoid" cancer cell line. However, differences in gene expression between Hs766T and normal ductal epithelium might suggest that this cell line contains a minimal set of changes responsible for key features of pancreatic neoplasia. On such a comparison, SAGE tags mapped to genes encoding protein translocation complex ß, regulator of G protein 5, nuclear phosphoprotein B23, MKP-1-like protein tyrosine phosphatase, tumor necrosis factor 
-inducible protein, catenin 1 (102 kDa), RAD51 (Saccharomyces cerevisiae) homologue C, guanine nucleotide-binding protein 
5, BCL2-associated athanogene 3, and 21 others were overexpressed in the "normoid" cancer cell Hs766T by >10-fold.
PCA detected groups of genes that could represent cell line-specific expression deficits, that is, genes that were not expressed in one cell line but were expressed in the others. For example, 116 known genes lacked expression in Hs766T but were expressed in all of the other cell lines. Examination of the chromosomal locations of these genes revealed that the cytogenetic distributions of these genes were concentrated in several regions and did not have a random distribution. For an example, the cytogenetic locations of 13 of the 116 genes deficient in Hs766T cells reside between 214 and 263 cR3000 on chromosome 11 (33) , which is closely associated with the fragile site of chromosomal band 11q13. This "regional dropout" of gene expression in a single cell line raised the possibility of a homozygous deletion, but none of the genes (10 were tested) were absent from the genomic DNA, and these down-regulated genes were physically interspersed with expressed genes.9 Possible explanations of this regional gene dropout in gene expression include regional gene silencing by methylation (34 , 35) and regional chromatin structural changes (36 , 37) .
Gene Expression versus Known Genetic Mutations.
Profiles of genetic changes are well established in pancreatic cancer cell lines (18)
. For example, homozygous deletions and mutations of tumor suppressor genes, such as p53, MADH4, MKK4, and BRCA2, were known within the cell lines studied by SAGE. We performed group-wise comparisons of SAGE data, including p53 wild-type versus p53 mutated, MADH4 wild-type versus MADH4 homozygous deleted/mutated, BRCA2 wild-type versus BRCA2 mutated, and MKK4 wild-type versus MKK4 homozygous deleted/mutated, but no distinct patterns emerged. In the Student t test, the number of expressed genes that achieved a given P cutoff level appeared to depend primarily on the power of the comparison (i.e., the number of cell lines in each arm of the comparison). No other pattern could be discerned with the dataset, as might be expected from the small numbers of cell lines available for comparison.
We may offer the following summary and perspective. The progression from normal cell to cancer cell undoubtedly involves stochastic alterations in genetic composition and gene expression; however, selective pressures related to the process of tumorigenesis and metastasis result in the accumulation of common sets of defects that contribute to survival and spread of tumors. An unbiased survey by SAGE analysis identified a candidate list of differentially expressed genes (Table 3)
. This gene set likely includes genes of which the deregulation contributes to tumorigenesis in the pancreas. Such genes may be robust markers of pancreatic neoplasia and suggest new targets for directed diagnostic and therapeutic approaches.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Supported by Grants CA62924 and CA72712 from the NIH Specialized Programs of Research Excellence in Gastrointestinal Cancer. 
2 To whom requests for reprints should be addressed, at Department of Oncology, 451 Cancer Research Building, 1650 Orleans Street, The Johns Hopkins School of Medicine, Baltimore, MD 21231. Phone: (410) 614-3316; Fax: (410) 614-9705; E-mail: sk{at}jhmi.edu
3 The abbreviations used are: SAGE, serial analysis of gene expression; PSCA, prostate stem cell antigen; CEACAM6, carcinoembryonic antigen-related cell adhesion molecule 6; CGAP, Cancer Genome Anatomy Project; EST, expressed sequence tag; PCA, principal component analysis; RT-PCR, reverse transcription-PCR.
4 Internet address: http://www.ncbi.nlm.nih.gov/SAGE/.
5 Internet address: http://www.microarrays.org/software.html.
6 Internet address: www.stat.auckland.ac.nz/rproj.html.
7 Internet address: http://www.ncbi.nlm.nih.gov/GenBank/index.html.
8 Internet address: http://www.ncbi.nlm.nih.gov/SAGE/sagexpsetup.cgi.
Received 4/25/01. Accepted 11/20/01.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. J. Kreitman, R. Hassan, D. J. FitzGerald, and I. Pastan Phase I Trial of Continuous Infusion Anti-Mesothelin Recombinant Immunotoxin SS1P Clin. Cancer Res., August 15, 2009; 15(16): 5274 - 5279. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. Riley, K. P. Engelhardt, J. W. Saldanha, W. Qi, L. S. Cooke, Y. Zhu, S. T. Narayan, K. Shakalya, K. D. Croce, I. G. Georgiev, et al. Design and Activity of a Murine and Humanized Anti-CEACAM6 Single-Chain Variable Fragment in the Treatment of Pancreatic Cancer Cancer Res., March 1, 2009; 69(5): 1933 - 1940. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Kurio, E. Murayama, T. Kaneko, Y. Shibata, T. Inai, and H. Iida Intron Retention Generates a Novel Isoform of CEACAM6 That May Act as an Adhesion Molecule in the Ectoplasmic Specialization Structures Between Spermatids and Sertoli Cells in Rat Testis Biol Reprod, December 1, 2008; 79(6): 1062 - 1073. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Ramachandran, T. Arumugam, H. Wang, and C. D. Logsdon Anterior Gradient 2 Is Expressed and Secreted during the Development of Pancreatic Cancer and Promotes Cancer Cell Survival Cancer Res., October 1, 2008; 68(19): 7811 - 7818. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Feldmann and A. Maitra Molecular Genetics of Pancreatic Ductal Adenocarcinomas and Recent Implications for Translational Efforts J. Mol. Diagn., March 1, 2008; 10(2): 111 - 122. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. D. Robinson and G. K. Smyth Moderated statistical tests for assessing differences in tag abundance Bioinformatics, November 1, 2007; 23(21): 2881 - 2887. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Hucl, J. R. Brody, E. Gallmeier, C. A. Iacobuzio-Donahue, I. K. Farrance, and S. E. Kern High Cancer-Specific Expression of Mesothelin (MSLN) Is Attributable to an Upstream Enhancer Containing a Transcription Enhancer Factor Dependent MCAT Motif Cancer Res., October 1, 2007; 67(19): 9055 - 9065. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Ortiz-Zapater, S. Peiro, O. Roda, J. M. Corominas, S. Aguilar, C. Ampurdanes, F. X. Real, and P. Navarro Tissue Plasminogen Activator Induces Pancreatic Cancer Cell Proliferation by a Non-Catalytic Mechanism That Requires Extracellular Signal-Regulated Kinase 1/2 Activation through Epidermal Growth Factor Receptor and Annexin A2 Am. J. Pathol., May 1, 2007; 170(5): 1573 - 1584. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Jimeno and M. Hidalgo Molecular biomarkers: their increasing role in the diagnosis, characterization, and therapy guidance in pancreatic cancer. Mol. Cancer Ther., April 1, 2006; 5(4): 787 - 796. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. C. Cunningham, F. Kamangar, M. P. Kim, S. Hammoud, R. Haque, C. A. Iacobuzio-Donahue, A. Maitra, R. Ashfaq, S. Hustinx, R. E. Heitmiller, et al. Claudin-4, mitogen-activated protein kinase kinase 4, and stratifin are markers of gastric adenocarcinoma precursor lesions. Cancer Epidemiol. Biomarkers Prev., February 1, 2006; 15(2): 281 - 287. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Chen, S. Pan, T. A. Brentnall, and R. Aebersold Proteomic Profiling of Pancreatic Cancer for Biomarker Discovery Mol. Cell. Proteomics, April 1, 2005; 4(4): 523 - 533. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. P. Landowski, B. S. Vig, X. Song, and G. L. Amidon Targeted delivery to PEPT1-overexpressing cells: Acidic, basic, and secondary floxuridine amino acid ester prodrugs Mol. Cancer Ther., April 1, 2005; 4(4): 659 - 667. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. B. Prasad, A. V. Biankin, N. Fukushima, A. Maitra, S. Dhara, A. G. Elkahloun, R. H. Hruban, M. Goggins, and S. D. Leach Gene Expression Profiles in Pancreatic Intraepithelial Neoplasia Reflect the Effects of Hedgehog Signaling on Pancreatic Ductal Epithelial Cells Cancer Res., March 1, 2005; 65(5): 1619 - 1626. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Aguilar, J. M. Corominas, N. Malats, J. A. Pereira, M. Dufresne, F. X. Real, and P. Navarro Tissue Plasminogen Activator in Murine Exocrine Pancreas Cancer: Selective Expression in Ductal Tumors and Contribution to Cancer Progression Am. J. Pathol., October 1, 2004; 165(4): 1129 - 1139. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Thomas, L. M. Santarsiero, E. R. Lutz, T. D. Armstrong, Y.-C. Chen, L.-Q. Huang, D. A. Laheru, M. Goggins, R. H. Hruban, and E. M. Jaffee Mesothelin-specific CD8+ T Cell Responses Provide Evidence of In Vivo Cross-Priming by Antigen-Presenting Cells in Vaccinated Pancreatic Cancer Patients J. Exp. Med., August 2, 2004; 200(3): 297 - 306. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. L. Kominsky, M. Vali, D. Korz, T. G. Gabig, S. A. Weitzman, P. Argani, and S. Sukumar Clostridium perfringens Enterotoxin Elicits Rapid and Specific Cytolysis of Breast Carcinoma Cells Mediated through Tight Junction Proteins Claudin 3 and 4 Am. J. Pathol., May 1, 2004; 164(5): 1627 - 1633. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Koopmann, P. Buckhaults, D. A. Brown, M. L. Zahurak, N. Sato, N. Fukushima, L. J. Sokoll, D. W. Chan, C. J. Yeo, R. H. Hruban, et al. Serum Macrophage Inhibitory Cytokine 1 as a Marker of Pancreatic and Other Periampullary Cancers Clin. Cancer Res., April 1, 2004; 10(7): 2386 - 2392. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Koopmann, N. S. Fedarko, A. Jain, A. Maitra, C. Iacobuzio-Donahue, A. Rahman, R. H. Hruban, C. J. Yeo, and M. Goggins Evaluation of Osteopontin as Biomarker for Pancreatic Adenocarcinoma Cancer Epidemiol. Biomarkers Prev., March 1, 2004; 13(3): 487 - 491. [Abstract] [Full Text]
|
|
|
|
|
|
N. Sato, N. Fukushima, A. Maitra, C. A. Iacobuzio-Donahue, N. T. van Heek, J. L. Cameron, C. J. Yeo, R. H. Hruban, and M. Goggins Gene Expression Profiling Identifies Genes Associated with Invasive Intraductal Papillary Mucinous Neoplasms of the Pancreas Am. J. Pathol., March 1, 2004; 164(3): 903 - 914. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Koopmann, Z. Zhang, N. White, J. Rosenzweig, N. Fedarko, S. Jagannath, M. I. Canto, C. J. Yeo, D. W. Chan, and M. Goggins Serum Diagnosis of Pancreatic Adenocarcinoma Using Surface-Enhanced Laser Desorption and Ionization Mass Spectrometry Clin. Cancer Res., February 1, 2004; 10(3): 860 - 868. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. A. Iacobuzio-Donahue, R. Ashfaq, A. Maitra, N. V. Adsay, G. L. Shen-Ong, K. Berg, M. A. Hollingsworth, J. L. Cameron, C. J. Yeo, S. E. Kern, et al. Highly Expressed Genes in Pancreatic Ductal Adenocarcinomas: A Comprehensive Characterization and Comparison of the Transcription Profiles Obtained from Three Major Technologies Cancer Res., December 15, 2003; 63(24): 8614 - 8622. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Sato, A. Maitra, N. Fukushima, N. T. van Heek, H. Matsubayashi, C. A. Iacobuzio-Donahue, C. Rosty, and M. Goggins Frequent Hypomethylation of Multiple Genes Overexpressed in Pancreatic Ductal Adenocarcinoma Cancer Res., July 15, 2003; 63(14): 4158 - 4166. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. D. Logsdon, D. M. Simeone, C. Binkley, T. Arumugam, J. K. Greenson, T. J. Giordano, D. E. Misek, and S. Hanash Molecular Profiling of Pancreatic Adenocarcinoma and Chronic Pancreatitis Identifies Multiple Genes Differentially Regulated in Pancreatic Cancer Cancer Res., May 15, 2003; 63(10): 2649 - 2657. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. A. Iacobuzio-Donahue, A. Maitra, M. Olsen, A. W. Lowe, N. T. Van Heek, C. Rosty, K. Walter, N. Sato, A. Parker, R. Ashfaq, et al. Exploration of Global Gene Expression Patterns in Pancreatic Adenocarcinoma Using cDNA Microarrays Am. J. Pathol., April 1, 2003; 162(4): 1151 - 1162. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. A. Iacobuzio-Donahue, A. Maitra, G. L. Shen-Ong, T. van Heek, R. Ashfaq, R. Meyer, K. Walter, K. Berg, M. A. Hollingsworth, J. L. Cameron, et al. Discovery of Novel Tumor Markers of Pancreatic Cancer using Global Gene Expression Technology Am. J. Pathol., April 1, 2002; 160(4): 1239 - 1249. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
