| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Molecular Oncology, Markers, Clinical Correlates |
1 Departments of Pathology,
2 Surgery, and
3 Oncology, Johns Hopkins Medical Institutions, Baltimore, Maryland;
4 Department of Pathology, University of Texas Southwestern Medical Center, Dallas, Texas;
5 Department of Surgery, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania; and
6 Roswell Park Cancer Institute, Buffalo, New York
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Experimental Design: RNA was extracted from eight well-differentiated PENs and three highly enriched pancreatic islet cell samples (80–90% purity), and examined using the Affymetrix U133A oligonucleotide microarray. Microarray data were normalized using dCHIP (www.dCHIP.org) for identification of differentially expressed genes. PEN tissue microarrays were constructed from 53 archival PENs for immunohistochemical validation of microarray data.
Results: Sixty-six transcripts were overexpressed 
3-fold in PENs compared with normal islet cells, including putative oncogenes (MLLT10/AF10), growth factors [insulin-like growth factor-binding protein 3 (IGFBP3)], cell adhesion and migration molecules (fibronectin), and endothelial elements (MUC18/MelCAM and CD31). A total of 119 transcripts were underexpressed 
3-fold in PENs compared with normal islet cells, including cell cycle checkpoint proteins (p21/Cip1), the MIC2 (CD99) cell surface glycoprotein, putative metastasis suppressor genes (NME3), and junD, a MEN1-regulated transcription factor. Using PEN tissue microarrays, we confirmed the differential up-regulation of IGFBP3 (70%) and fibronectin (22%) and differential down-regulation of p21 (46%) and MIC2 (CD99; 91%) in PENs versus normal pancreatic islets. IGFBP3 overexpression was significantly more common in metastatic (93%) versus primary PEN lesions (60%), P = 0.022. Fibronectin overexpression demonstrated a trend toward significance in lymphatic PEN metastases (55%) compared with primary PEN lesions (24%; P = 0.14).
Conclusions: Global expression analysis provides insight into tumorigenic pathways in PENs and may identify potential prognostic and therapeutic markers for these uncommon neoplasms.
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In contrast to the more common exocrine ductal adenocarcinomas, little is known about the molecular abnormalities or genotype-phenotype correlations that underlie PENs (reviewed in Ref. 8 ). This is likely a reflection of the small numbers of cases that are accessible for molecular studies. As a result, diagnostic and therapeutic approaches for these neoplasms have seen few advances in the last decade. We have previously demonstrated the power of global expression microarrays to identify a variety of novel tumor markers in exocrine pancreatic ductal adenocarcinomas, with immediate translational potential for patient care (9 , 10) . We now extend our gene expression studies to well-differentiated PENs; these studies document aberrantly activated pathways that may impact on tumorigenesis, and they identify potential cellular targets in these uncommon tumors. In this study, we identify nearly 200 differentially expressed genes in well-differentiated PENs versus enriched normal islet cells and validate a subset of these genes at the protein level using PEN TMAs. The overwhelming majority of differentially expressed genes we have identified have not been described previously in PENs and may serve as potential therapeutic targets and novel tumor markers for this disease.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. There were four males and four females in the cohort; the median age was 46 years (range, 42–55 years). The median tumor size was 3.6 cm (range, 2.0–6.0 cm). Two patients had concurrent lymphatic metastases; none had associated hepatic metastases. No patient had an associated endocrine hypersecretion syndrome. Tumor samples were collected within 10 min of surgical resection, snap-frozen in liquid nitrogen, and stored at -80°C. H&E-stained sections from adjacent frozen tissue were prepared to confirm the diagnosis (2)
and assess neoplastic cellularity. RNA was extracted from PENs containing >80% neoplastic cells and <10% necrosis on frozen section examination.
|
RNA Extraction and Hybridization.
Sample preparation and processing procedure were performed at the Roswell Park Cancer Institute Microarray Core Facility, as described in the Affymetrix GeneChip Expression Analysis Manual (Affymetrix, Santa Clara, CA). Briefly, frozen PEN or islet samples were crushed in TRIzol (Invitrogen, Carlsbad, CA) by using a Polytron homogenizer (Brinkman Instruments, Westbury, NY). Total RNA was then extracted from the crushed tissue and cleaned using RNeasy columns according to the manufacturer’s protocol (Qiagen, Valencia, CA). The integrity of total RNA was confirmed in each case using the Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA). Using 5–40 µg of total RNA, double-stranded cDNA was synthesized following SuperScript Choice system (Invitrogen). T7-(dT24) oligomer was used for priming the first-strand cDNA synthesis. The resultant cDNA was purified using Phase Lock Gel phenol/chloroform extraction and precipitated with ethanol. The cDNA pellet was collected and dissolved in appropriate volume. Using cDNA as template, cRNA was synthesized using a T7 MegaScript in Vitro Transcription Kit (Ambion, Austin, TX). Biotinylated-11-CTP and 16-UTP ribonucleotides (Enzo Diagnostics, Farmingdale, NY) were added to the reaction as labeling reagents. In vitro transcription reactions were carried out at 37°C for 6 h, and the labeled cRNA obtained was purified using RNeasy columns (Qiagen). The cRNA was fragmented in fragmentation buffer [40 mM Tris-acetate (pH 8.1), 100 mM KOAc, and 30 mM MgOAc] for 35 min at 94°C. Fragmented cRNA (10–11 µg/probe array) was used to hybridize to human U133A GeneChip array at 45°C for 24 h in a hybridization oven with constant rotation (60 rpm). The chips were washed and stained using Affymetrix fluidics stations. Staining was performed using streptavidin phycoerythrin conjugate (SAPE; Molecular Probes, Eugene, OR), followed by the addition of biotinylated antibody to streptavidin (Vector Laboratories, Burlingame, CA). Probe arrays were scanned using fluorometric scanners (Hewlett Packard Gene Array Scanner; Hewlett Packard Corp., Palo Alto, CA). The scanned images were inspected and analyzed using established quality control measures.
Data Filtering and Analysis.
The 11 .CEL files (3 normal islets and 8 PENs) generated by the Affymetrix Microarray Suite (MAS) version 5.0 were converted into .DCP files using dCHIP8
, as described previously by Li and Wong (12)
. The 11 .DCP files were normalized, and raw gene expression data were generated using the dCHIP system of model-based analysis. For comparison of global gene expression profiles between normal islet cells and PENs, the three islets were designated as "baseline" (B), and the eight PENs were designated as "experiment" (E) in the dCHIP comparison software. Genes overexpressed 3-fold in the PENs versus normal islets were then identified by defining the appropriate filtering criteria in the dCHIP software (mean E/mean B > 3; mean E – mean B = 100); similarly, genes underexpressed were identified by reversing the filtering criteria (mean B/mean E > 3; mean B – mean E = 100); Ps were generated using the t test.
Immunohistochemistry.
PEN TMAs were generated from 53 archival paraffin-embedded well-differentiated PENs, as described previously (13)
. Each cancer specimen was represented by one to five 1.4-mm cores on the TMA, and, when available, adjacent normal pancreas parenchyma containing normal islets of Langerhans were also arrayed for comparison. In summary, the 53 arrayed PENs consisted of 44 primary tumors (9 with paired lymph node metastases) and 9 isolated hepatic metastases without corresponding primary tumor; adjacent normal pancreatic parenchyma was arrayed in 43 of 44 primary PENs.
Slides were deparaffinized in fresh xylene and rehydrated through sequential graded ethanol steps. Antigen retrieval was performed by citrate buffer incubation [18 mM citric acid and 8.2 mM sodium citrate (pH 6.0)] using a household vegetable steamer (Black and Decker) for 60 min. Slides were incubated for 5 min with 3% hydrogen peroxide, washed in TBS/T [20 mM Tris, 140 mM NaCl, and 0.1% Tween 20 (pH 7.6)], and incubated in appropriate antibody dilutions for p21(CDKN1A) (1:75; Dako, Carpinteria, CA), CD99 (O13; 1:400; Signet, Dedham, MA), IGFBP3 (1:40; Santa Cruz Biotechnology, Santa Cruz, CA), and fibronectin (1:1200; Dako) for 60 min at room temperature. Normal saline was substituted for the primary antibody in control sections. The avidin-biotin-peroxidase complex method from DAKO (Glostrup, Denmark) was used, and slides were subsequently counterstained with hematoxylin. Assessment of immunohistochemical labeling in the TMAs was performed by two of the authors (D. E. H. and A. M.). Loss of nuclear p21 or membranous CD99 in >95% of neoplastic endocrine cells was considered "negative;" similarly, labeling of cytoplasmic IGFBP3 or membranous/cytoplasmic fibronectin in >25% of neoplastic cells was considered "positive." The >25% cutoff was used based on our extensive previous experience with immunohistochemical evaluation of TMA data (9 , 10 , 14, 15, 16) . Because normal islets also express IGFBP3, cases were considered positive when the staining intensity was greater than that in normal islets. We found low to moderate basal IGFBP3 in normal islets, and overexpression in PENs was assessed by comparison with this "basal" level, a task greatly facilitated by the availability of adjacent nonneoplastic islets in 43 of 44 primary tumors on our PEN TMA.
Statistical analysis was performed using the Analyze-it software package for Microsoft. Two-tailed Fisher’s exact tests were performed to determine the differences in expression of candidate genes by immunohistochemistry, and P < 0.05 was considered significant.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
3-fold in PENs compared with enriched islet cells, and the complete list is tabulated in Supplementary Table 1
. The list of up-regulated transcripts included putative oncogenes [MLLT10/AF10 (17
, 18)
], growth factors [IGFBP3 (19
, 20)
], cell adhesion and migration molecules [fibronectin (21, 22, 23)
], and endothelial elements [MUC18/MelCAM and CD31 (24)
]. Similarly, 119 transcripts were underexpressed 3-fold in PENs compared with enriched islet cells, and the complete list is tabulated in Supplementary Table 2
. The list of down-regulated transcripts included cell cycle checkpoint proteins [p21/Cip1 (25
, 26)
], the MIC2 (CD99) cell surface glycoprotein (27)
, genes involved in DNA damage repair and genomic stability [O6-methylguanine-DNA methyltransferase and GADD45 (28
, 29)
], putative metastasis suppressor genes [NME3/DR-nme23 (30)
], and junD, a MEN1-regulated transcription factor (31)
. A literature search of PubMed9
revealed that the differential expression of several genes in this study has previously been reported (either singly or through global microarray expression analyses) in other cancer types, in principle validating our approach. A second PubMed search using the gene name and either "islet cell" or "pancreatic endocrine" only yielded two genes (p21/CDKN1A and CD99) that have been previously reported as underexpressed in PENs, by immunohistochemistry (32
, 33)
. Thus, the overwhelming majority of the 185 differentially expressed genes reported in this study represent newly described tumor markers for PENs.
|
; Fig. 1
). Table 2
also lists the labeling pattern seen in normal islets as a comparison. The denominator (i.e., total numbers of cases evaluated) is variable between the antibodies because of "dropout" of tissue cores during the TMA staining process.
|
. Fifty-two cases could be evaluated for p21 labeling on the PEN TMA. Overall, loss of p21 expression (i.e., <5% nuclear p21 labeling) was seen 24 of 52 (46%) PENs (Fig. 1B)
, including 21 of 44 (48%) primary tumors and 3 of 8 (38%) isolated hepatic metastases. In seven of eight (88%) lymph node metastases, the pattern of p21 expression was concordant with the corresponding primary tumor, i.e., three of seven matched lesions retained p21, whereas labeling was absent in the remaining four. In only one case, p21 expression was present in the primary tumor, whereas the matched nodal metastasis lacked p21 expression. There were no statistically significant differences in loss of p21 expression between primary (48%) and all metastatic PEN lesions [7 of 15 or 47% (P = 1.0, Fisher’s exact test)].
Fifty-three cases could be evaluated for CD99 labeling. Intense, diffuse membranous CD99 expression was seen in normal islet cells (Fig. 1C)
, whereas as many as 48 of 53 (91%) PENs, including 39 of 44 (89%) primary tumors and 9 of 9 (100%) isolated hepatic metastases demonstrated loss of CD99 expression (Fig. 1D)
. CD99 expression was concordant in nine of nine (100%) paired primary and lymph node metastases on the PEN tissue array, with loss of CD99 expression demonstrable in eight of nine matched lesions, and retention of labeling was demonstrated in one case. There were no statistically significant differences in loss of CD99 expression between primary (91%) and all metastatic PEN lesions [17 of 18 or 94% (P = 0.86, Fisher’s exact test)].
IGFBP3 and fibronectin were up-regulated in PENs compared with normal islets on microarray analysis (4.1- and 3.9-fold, respectively). Forty-seven cases could be evaluated for IGFBP3 labeling on the PEN TMA. Immunohistochemical analysis demonstrated weak to moderate granular IGFBP3 labeling in the cytoplasm of almost all normal islets (Fig. 1E)
. Using the staining intensity of normal islets as a baseline, 14 of 47 (30%) cases expressed IGFBP3 equal to or less than the normal islet cells. In contrast, intense, diffuse overexpression of IGFBP3 was seen in 33 of 47 (70%) PENs (Fig. 1F)
, including 24 of 40 (60%) primary PENs and 7 of 7 (100%) isolated hepatic metastases. IGFBP3 expression was concordant in nine of nine (100%) paired primary and lymph node metastases on the PEN TMA, with eight of nine (89%) cases coexpressing IGFBP3, and one matched primary and nodal metastasis demonstrating no labeling. There was a statistically significant difference in IGFBP3 overexpression in all metastatic lesions (15 of 16 or 93%) versus primary PENs [24 of 40 or 60% (P = 0.022, Fisher’s exact test)].
Fibronectin is traditionally considered a stromal marker, and expectedly, intense fibronectin expression was present in regions of peritumoral desmoplasia (data not shown), whereas normal islet cells were negative (Fig. 1G)
. Of note however, diffuse, membranous labeling of the neoplastic cells themselves was present in a significant minority [11 of 50 (22%) cases; Fig. 1F
]; the latter included 10 of 42 (24%) primary tumors and 1 of 8 (13%) isolated hepatic metastasis. Another notable feature of fibronectin labeling was its propensity to be up-regulated in lymphatic metastases compared with the paired primary tumor; thus, of nine paired primary tumor and lymph node metastases examined, five of nine (55%) demonstrated labeling that was restricted to the metastasis, whereas the primary lesion was negative. There was no significant difference in fibronectin overexpression between primary (24%) and all metastatic PEN lesions [6 of 17 or 35% (P = 0.55, Fisher’s exact test)]; in contrast, the difference in fibronectin expression between nodal metastases alone (55%) and primary PENs demonstrated a trend but did not reach statistical significance (P = 0.14, Fisher’s exact test).
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We have performed the first global gene expression analysis of PENs and compared this to the gene expression in highly enriched isolated human islets, using the second-generation Affymetrix U133 platform, containing 
22,000 human transcripts. We have identified 66 transcripts overexpressed 3-fold and 119 transcripts expressed 3-fold in PENs compared with normal islet cells. In addition, we have immunohistochemically validated the differential expression of four genes in TMAs comprised of archival primary and metastatic PEN lesions. A search on the National Library of Medicine’s PubMed9
using the gene name and either "pancreatic endocrine" or "islet cell" revealed that only two transcripts in our combined list of 185 genes, p21 and CD99, have previously been described in the context of PENs (32
, 33)
. Thus, the overwhelming majority of genes we have identified represent novel differentially expressed genes in PENs.
A particular strength of this study is the use of enriched human islet cells as a normal control, as opposed to bulk pancreas, which contains <2% islets of Langerhans. This approach has permitted an accurate comparison of neoplastic endocrine cells with their normal counterpart, instead of a plethora of nonendocrine tissues. We are aware that this analysis may have generated a minority of "false positive" differentially expressed genes (for example, genes present in tumor-associated endothelium or stromal elements), which would not be expressed by enriched islet cell extracts. Nevertheless, by judiciously parsing the gene lists using known gene function and cellular localization and, more importantly, by validation in tissue sections, we have generated a rational expression profile of neoplastic endocrine cells.
The paramount importance of tissue validation of microarray expression data is best exemplified by two transcripts in the overexpressed gene list: fibronectin and melanoma cell adhesion molecule (MelCAM or Muc18). Fibronectin is classically considered a stromal marker (9) . However, on immunohistochemical analysis, a significant minority (22%) of PENs expressed fibronectin on the neoplastic cell surface, a proportion that was even higher in nodal metastasis (see the discussion of fibronectin below). On the contrary, MelCAM/Muc18, a cell adhesion molecule otherwise present in several cancer types (34, 35, 36) , was not expressed in the neoplastic endocrine cells, although its expression was up-regulated in the microarray analysis (data not shown). MelCAM/Muc18 labeling was restricted to the endothelium only [both normal and neoplasia-associated endothelium (36 , 37) ], and this transcript was presumably "overexpressed" in PENs due to the rich fibrovascular stroma typically seen in these tumors. As stated above and demonstrated in multiple reports from this group (9 , 10 , 14, 15, 16) , global microarray analysis of "bulk" tumor tissues is likely to yield a subset of differentially expressed genes that are due to peritumoral stromal element, rather than the neoplastic cells themselves; therefore, careful parsing of gene function and immunohistochemical/in situ validation are critical to avoid "false alarms" in microarray analyses. The ensuing discussion will address some of the genes whose differential expression we have validated in tissue sections of PENs.
As stated previously, we confirmed the differential expression of four genes using PEN TMA. This includes two down-regulated genes (p21 and CD99) and two up-regulated genes (IGFBP3 and fibronectin), respectively. The loss of p21, a cyclin-dependent kinase inhibitor, has been reported in a large number of human neoplasms, including PENs (25
, 26)
. In a study of 109 gastrointestinal carcinoid tumors, including 42 PENs, Canavese et al. (33)
found a low level of p21 expression in most well-differentiated PENs. The authors used a "p21 labeling index" and reported a median labeling index of 1.29% for "benign tumors" (range, 0–20.48%). Because the scoring criterion was different from the present TMA-based study, we cannot make a direct comparison. In contrast, Choi et al. (32)
reported loss of p21 in 40% of nonfunctional PENs in their series, which is in excellent agreement with our own findings (46% loss of expression). In our series of PENs, we failed to find a statistically significant difference in p21 expression between primary and metastatic PEN lesions, and p21 expression was concordant in 90% of matched primary pancreatic and nodal metastatic lesions.
The MIC2 gene locus is located in the pseudoautosomal (pairing) region of human X and Y chromosomes (38)
. The protein product of the MIC2 gene (CD99 or E2 antigen) is a 32-kDa human T-cell surface glycoprotein involved in spontaneous rosette formation with erythrocytes (27)
. Intense membranous CD99 expression is considered a sine qua non of EWS-FLI1 translocation-positive primitive neuroectodermal tumors/Ewing sarcomas (39)
. CD99 reactivity has been increasingly recognized in a variety of other mesenchymal tumor types, including poorly differentiated synovial sarcomas and rhabdomyosarcomas, epithelial malignancies (gastric adenocarcinomas), and lymphoblastic lymphoma. In the pancreas, besides pancreatic primitive neuroectodermal tumors/Ewing sarcomas (40)
, CD99 expression has been uncommonly reported in solid pseudopapillary tumors (41)
. Choi et al. (32)
had previously reported absence of CD99 labeling in nonfunctioning islet cell tumors. In accordance with our microarray expression data, we found intense, diffuse CD99 expression in all benign islets, whereas 91% of PENs in our series lacked CD99 expression. The mechanism of CD99 down-regulation in PENs is a matter of speculation. Some studies have suggested a role for the EBV latent membrane protein (LMP1) orchestrating down-regulation of CD99 antigen via a nuclear factor-
ß-mediated pathway in Reed-Sternberg cells (42
, 43)
; however, this is unlikely to be the mechanism of action in PENs. We found no significant correlation between site of lesion and CD99 expression in our series (89% of primary tumors and 94% of hepatic and nodal metastases demonstrated lack of CD99 labeling).
IGFBP3 is a member of the family of insulin-like growth factors (19) . The insulin-like growth factors, their receptors, and their binding proteins play key roles in regulating cell proliferation and apoptosis. IGFBP3 is the major carrier protein for IGF1 and IGF2 in the circulation. IGFBP3 possesses both growth-inhibitory and -potentiating effects on cells that are independent of IGF action and are mediated through specific IGFBP3-binding proteins/receptors located at the cell membrane, cytosol or nuclear compartments and in the extracellular matrix (19) . For example, in some cancer cells, IGFBP3 has proapoptotic activities both dependent on and independent of p53 (44 , 45) . On the contrary, elevated serum IGFBP3 may be a predictor for progression and recurrence of breast cancers (20 , 46) . We have recently demonstrated up-regulation of IGFBP3 in pancreatic ductal adenocarcinomas using the U133 microarray platform.10 The commonality of IGFBP3 up-regulation in both exocrine and endocrine neoplasms of the pancreas suggests an important, possibly growth-promoting role for this protein in pancreatic tumorigenesis. On immunohistochemistry, we found low to moderate "basal" IGFBP3 in normal islets, and overexpression in PENs was assessed by comparison with this "basal" level, a task greatly facilitated by the availability of adjacent nonneoplastic islets in 43 of 44 primary tumors on our PEN TMA. Using this criterion, 70% of PEN lesions demonstrated robust IGFBP3 overexpression. Although there was an excellent correlation between IGFBP3 labeling in matched primary and nodal metastases (9 of 9 or 100% of matched lesions demonstrated a similar pattern of IGFBP3 labeling), we found a significantly higher overall proportion of metastatic foci that labeled with IGFBP3 (15 of 16 or 93%) compared with primary tumors [24 of 40 or 60% (P = 0.022)].
Fibronectin is a 430-kDa dimeric glycoprotein that exists in two forms, termed cellular and plasma fibronectin. Cellular fibronectin is the major cell surface glycoprotein of many fibroblast cell lines and serves as a ligand for the integrin family of cell adhesion receptors and regulates cytoskeletal organization (22 , 23) . Several reports have demonstrated that secretion of this promigratory molecule may be a key event in cancer cells during the progression to a metastatic phenotype. For example, Bittner et al. (47) demonstrated overexpression of fibronectin in a series of highly metastatic uveal melanoma cell lines, consistent with an important role for focal contacts in modulating melanoma cell motility. Similarly, Clark et al. (48) reported that fibronectin was one of the three consistently up-regulated genes (along with thymosin ß4 and RhoC) in pulmonary metastases arising from human and mouse melanoma cell lines. On the contrary, peptides that mimic the cell-adhesive region of fibronectin were shown to inhibit metastasis, indicating that tumor cells must interact with molecules such as fibronectin to metastasize (49) . In our series of PENs, we found expression of fibronectin in a small but significant minority of primary tumors (24%). Of interest, however, lymphatic, but not hepatic, metastases of PENs demonstrated a higher proportion of cases with fibronectin expression; in five of nine (55%) PENs, up-regulation of fibronectin was restricted to the nodal metastasis only, whereas the primary lesion was negative (P = 0.14). The expression of fibronectin in the majority of lymphatic metastases of PENs also presents an opportunity to potentially use fibronectin-conjugated radionuclides or immunotoxins for the treatment of refractory metastatic disease (50) .
In summary, we have performed the first global expression profiling of well-differentiated PENs and identified 119 significantly down-regulated and 66 significantly up-regulated genes compared with enriched human pancreatic islet cells, using the Affymetrix U133 platform. We have validated the differential expression for a subset of these genes using TMAs. The raw gene expression profiling data are available on request and should serve as an effective "launching pad" for future studies aimed at elucidating tumorigenic pathways in endocrine tumors of the pancreas. These gene expression data also have the potential to allow identification of new cellular targets for therapy in these uncommon neoplasms.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Note: Supplementary data for this article are available at Clinical Cancer Research Online (http://clincancerres.aacrjournals.org).
Requests for reprints: Anirban Maitra, Ross 632, Department of Pathology, Johns Hopkins University School of Medicine, 720 Rutland Avenue, Baltimore, Maryland 21205. Phone: (410) 955-3511; Fax: (410) 614-0671; E-mail: amaitra1{at}jhmi.edu
7 The abbreviations used are: PEN, pancreatic endocrine neoplasm; IGFBP, insulin-like growth factor-binding protein; TMA, tissue microarray. 
9 www.ncbi.nlm.nih.gov/PubMed.
10 J. Iacobuzio-Donahue, unpublished data.
Received 4/30/03; revised 8/13/03; accepted 8/14/03.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-Methylacyl-CoA racemase: a new molecular marker for prostate cancer. Cancer Res., 62: 2220-2226, 2002.B. Blood, 97: 3596-3604, 2001. antisense oligonucleotide. Mol. Pharmacol., 55: 396-402, 1999.This article has been cited by other articles:
|
|
 |
|
  T. R Halfdanarson, J. Rubin, M. B Farnell, C. S Grant, and G. M Petersen Pancreatic endocrine neoplasms: epidemiology and prognosis of pancreatic endocrine tumors Endocr. Relat. Cancer, June 1, 2008; 15(2): 409 - 427. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E.-M. Duerr, Y. Mizukami, A. Ng, R. J Xavier, H. Kikuchi, V. Deshpande, A. L Warshaw, J. Glickman, M. H Kulke, and D. C Chung Defining molecular classifications and targets in gastroenteropancreatic neuroendocrine tumors through DNA microarray analysis Endocr. Relat. Cancer, March 1, 2008; 15(1): 243 - 256. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Di Florio, G. Capurso, M. Milione, F. Panzuto, R. Geremia, G. D. Fave, and C. Sette Src family kinase activity regulates adhesion, spreading and migration of pancreatic endocrine tumour cells Endocr. Relat. Cancer, March 1, 2007; 14(1): 111 - 124. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S Fontaniere, J Tost, A Wierinckx, J Lachuer, J Lu, N Hussein, F Busato, I Gut, Z-Q Wang, and C-X Zhang Gene expression profiling in insulinomas of Men1 {beta}-cell mutant mice reveals early genetic and epigenetic events involved in pancreatic {beta}-cell tumorigenesis Endocr. Relat. Cancer, December 1, 2006; 13(4): 1223 - 1236. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.-J. Byun, I.-K. Hong, E. Kim, Y.-J. Jin, D.-I. Jeoung, J.-H. Hahn, Y.-M. Kim, S. H. Park, and H. Lee A Splice Variant of CD99 Increases Motility and MMP-9 Expression of Human Breast Cancer Cells through the AKT-, ERK-, and JNK-dependent AP-1 Activation Signaling Pathways J. Biol. Chem., November 17, 2006; 281(46): 34833 - 34847. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Cozar-Castellano, N. Fiaschi-Taesch, T. A. Bigatel, K. K. Takane, A. Garcia-Ocana, R. Vasavada, and A. F. Stewart Molecular Control of Cell Cycle Progression in the Pancreatic {beta}-Cell Endocr. Rev., June 1, 2006; 27(4): 356 - 370. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. C. Manara, G. Bernard, P.-L. Lollini, P. Nanni, M. Zuntini, L. Landuzzi, S. Benini, G. Lattanzi, M. Sciandra, M. Serra, et al. CD99 Acts as an Oncosuppressor in Osteosarcoma Mol. Biol. Cell, April 1, 2006; 17(4): 1910 - 1921. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Cozar-Castellano, M. Weinstock, M. Haught, S. Velazquez-Garcia, D. Sipula, and A. F. Stewart Evaluation of {beta}-Cell Replication in Mice Transgenic for Hepatocyte Growth Factor and Placental Lactogen: Comprehensive Characterization of the G1/S Regulatory Proteins Reveals Unique Involvement of p21cip Diabetes, January 1, 2006; 55(1): 70 - 77. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Gronborg, T. Z. Kristiansen, A. Iwahori, R. Chang, R. Reddy, N. Sato, H. Molina, O. N. Jensen, R. H. Hruban, M. G. Goggins, et al. Biomarker Discovery from Pancreatic Cancer Secretome Using a Differential Proteomic Approach Mol. Cell. Proteomics, January 1, 2006; 5(1): 157 - 171. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Buchholz, H. A. Kestler, A. Bauer, W. Bock, B. Rau, G. Leder, W. Kratzer, M. Bommer, A. Scarpa, M. K. Schilling, et al. Specialized DNA Arrays for the Differentiation of Pancreatic Tumors Clin. Cancer Res., November 15, 2005; 11(22): 8048 - 8054. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Maffei, Z. Liu, P. Witkowski, F. Moschella, G. D. Pozzo, E. Liu, K. Herold, R. J. Winchester, M. A. Hardy, and P. E. Harris Identification of Tissue-Restricted Transcripts in Human Islets Endocrinology, October 1, 2004; 145(10): 4513 - 4521. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. E. Hansel, A. Rahman, M. House, R. Ashfaq, K. Berg, C. J. Yeo, and A. Maitra Met Proto-Oncogene and Insulin-Like Growth Factor Binding Protein 3 Overexpression Correlates with Metastatic Ability in Well-Differentiated Pancreatic Endocrine Neoplasms Clin. Cancer Res., September 15, 2004; 10(18): 6152 - 6158. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Kunapuli, C. S. Kasyapa, L. Hawthorn, and J. K. Cowell LGI1, a Putative Tumor Metastasis Suppressor Gene, Controls in Vitro Invasiveness and Expression of Matrix Metalloproteinases in Glioma Cells through the ERK1/2 Pathway J. Biol. Chem., May 28, 2004; 279(22): 23151 - 23157. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||
