Identification of genes with specific expression in pancreatic cancer by cDNA representational difference analysis

At the dawn of the transcriptomic medicine

Progress in genomic analytical technologies has improved our possibilities to obtain information regarding DNA, RNA, and their dynamic changes that occur over time or in response to specific challenges. This information describes the blueprint for cells, tissues, and organisms and has fundamental importance for all living organisms. This review focuses on the technological challenges to analyze the transcriptome and what is the impact of transcriptomics on precision medicine. The transcriptome is a term that covers all RNA present in cells and a substantial part of it will never be translated into protein but is nevertheless functional in determining cell phenotype. Recent developments in transcriptomics have challenged the fundamentals of the central dogma of biology by providing evidence of pervasive transcription of the genome. Such massive transcriptional activity is challenging the definition of a gene and especially the term "pseudogene" that has now been demonstrated in many examples to be both transcribed and translated. We also review the common sources of biomaterials for transcriptomics and justify the suitability of whole blood RNA as the current optimal analyte for clinical transcriptomics. At the end of the review, a brief overview of the clinical implications of transcriptomics in clinical trial design and clinical diagnosis is given. Finally, we introduce the transcriptome as a target for modern drug development as a tool for extending our capacity for precision medicine in multiple diseases.

Emerging Cancer Epigenetic Mechanisms Regulated by All-Trans Retinoic Acid

All-trans retinoic acid (RA), which is the dietary bioactive derivative obtained from animal (retinol) and plant sources (beta-carotene), is a physiological lipid signal of both embryonic and postembryonic development. During pregnancy, either RA deficiency or an excessive RA intake is teratogenic. Too low or too high RA affects not only prenatal, but also postnatal, developmental processes such as myelopoiesis and mammary gland morphogenesis. In this review, we mostly focus on emerging RA-regulated epigenetic mechanisms involving RA receptor alpha (RARA) and Annexin A8 (ANXA8), which is a member of the Annexin family, as well as ANXA8 regulatory microRNAs (miRNAs). The first cancer showing ANXA8 upregulation was reported in acute promyelocytic leukemia (APL), which induces the differentiation arrest of promyelocytes due to defective RA signaling caused by RARA fusion genes as the PML-RARA gene. Over the years, ANXA8 has also been found to be upregulated in other cancers, even in the absence of RARA fusion genes. Mechanistic studies on human mammary cells and mammary glands of mice showed that ANXA8 upregulation is caused by genetic mutations affecting RARA functions. Although not all of the underlying mechanisms of ANXA8 upregulation have been elucidated, the interdependence of RA-RARA and ANXA8 seems to play a relevant role in some normal and tumorigenic settings.

Transmembrane 4 L Six Family Member 5 Senses Arginine for mTORC1 Signaling

The mechanistic target of rapamycin complex (mTORC1) is a signaling hub on the lysosome surface, responding to lysosomal amino acids. Although arginine is metabolically important, the physiological arginine sensor that activates mTOR remains unclear. Here, we show that transmembrane 4 L six family member 5 (TM4SF5) translocates from plasma membrane to lysosome upon arginine sufficiency and senses arginine, culminating in mTORC1/S6K1 activation. TM4SF5 bound active mTOR upon arginine sufficiency and constitutively bound amino acid transporter SLC38A9. TM4SF5 binding to the cytosolic arginine sensor Castor1 decreased upon arginine sufficiency, thus allowing TM4SF5-mediated sensing of metabolic amino acids. TM4SF5 directly bound free L-arginine via its extracellular loop possibly for the efflux, being supported by mutant study and homology and molecular docking modeling. Therefore, we propose that lysosomal TM4SF5 senses and enables arginine efflux for mTORC1/S6K1 activation, and arginine-auxotroph in hepatocellular carcinoma may be targeted by blocking the arginine sensing using anti-TM4SF5 reagents.

γ-irradiation from radiotherapy improves the virulence potential of Candida tropicalis

AIM

To evaluate if radiation used in radiotherapy can cause changes in the virulence potential of Candida tropicalis ATCC 750.

MATERIALS & METHODS

C. tropicalis was exposed in vitro to identical dose and scheme of irradiation would be used in patients with head and neck cancer. Some virulence parameters were analyzed before and after irradiation.

RESULTS

Colony morphologies were irreversibly affected by irradiation. Increase in growth rate, filamentation, adhesion on cell lines and phagocytosis process were also observed. Overall the irradiated C. tropicalis cells became more efficient at causing systemic infection in mice.

CONCLUSION

γ-radiation induced important changes in C. tropicalis increasing its virulence profile, which could directly affect the relationship between yeasts and hosts.

A multistep high-content screening approach to identify novel functionally relevant target genes in pancreatic cancer

In order to foster the systematic identification of novel genes with important functional roles in pancreatic cancer, we have devised a multi-stage screening strategy to provide a rational basis for the selection of highly relevant novel candidate genes based on the results of functional high-content analyses. The workflow comprised three consecutive stages: 1) serial gene expression profiling analyses of primary human pancreatic tissues as well as a number of in vivo and in vitro models of tumor-relevant characteristics in order to identify genes with conspicuous expression patterns; 2) use of ‘reverse transfection array’ technology for large-scale parallelized functional analyses of potential candidate genes in cell-based assays; and 3) selection of individual candidate genes for further in-depth examination of their cellular roles. A total of 14 genes, among them 8 from “druggable” gene families, were classified as high priority candidates for individual functional characterization. As an example to demonstrate the validity of the approach, comprehensive functional data on candidate gene ADRBK1/GRK2, which has previously not been implicated in pancreatic cancer, is presented.

Transcriptomic analysis predicts survival and sensitivity to anticancer drugs of patients with a pancreatic adenocarcinoma

A major impediment to the effective treatment of patients with pancreatic ductal adenocarcinoma (PDAC) is the molecular heterogeneity of this disease, which is reflected in an equally diverse pattern of clinical outcome and in responses to therapies. We developed an efficient strategy in which PDAC samples from 17 consecutive patients were collected by endoscopic ultrasound-guided fine-needle aspiration or surgery and were preserved as breathing tumors by xenografting and as a primary culture of epithelial cells. Transcriptomic analysis was performed from breathing tumors by an Affymetrix approach. We observed significant heterogeneity in the RNA expression profile of tumors. However, the bioinformatic analysis of these data was able to discriminate between patients with long- and short-term survival corresponding to patients with moderately or poorly differentiated PDAC tumors, respectively. Primary culture of cells allowed us to analyze their relative sensitivity to anticancer drugs in vitro using a chemogram, similar to the antibiogram for microorganisms, establishing an individual profile of drug sensitivity. As expected, the response was patient dependent. We also found that transcriptomic analysis predicts the sensitivity of cells to the five anticancer drugs most frequently used to treat patients with PDAC. In conclusion, using this approach, we found that transcriptomic analysis could predict the sensitivity to anticancer drugs and the clinical outcome of patients with PDAC.

Distribution and clinical significance of tumour-associated macrophages in pancreatic ductal adenocarcinoma: a retrospective analysis in China

BACKGROUND

We aimed to characterize the localization and prognostic significance of tumour-associated macrophages (tams) in pancreatic ductal adenocarcinoma (pdac).

METHODS

Tumour specimens from 70 patients with pdac and inflammatory specimens from 13 patients with chronic pancreatitis were collected and analyzed for tam and M2 macrophage counts by immunohistochemistry. Correlations between tam distributions and clinicopathologic features were determined.

RESULTS

Immunohistochemical analysis showed that tam and M2 macrophage counts were higher in tissues from pdac than from chronic pancreatitis. The tams and M2 macrophages both infiltrated more into peritumour. Both macrophage types were positively associated with lymph node metastasis (p = 0.041 for tams in peritumour, p = 0.013 for M2 macrophages in introtumour, p = 0.006 for M2 macrophage in peritumour). In addition, abdominal pain was significantly more frequent in pdac patients with a greater tams count. The survival rate was much lower in patients having high infiltration by M2 macrophages than in those having low infiltration.

CONCLUSIONS

The tam count might be associated with neural invasion in pdac, and M2 macrophages might play an important role in lymph node metastasis. Higher counts of either macrophage type were associated with increased risk of lymph node metastasis, and the M2 macrophage count could potentially be a marker for evaluating prognosis.

Membrane Proteins Involved in Epithelial-Mesenchymal Transition and Tumor Invasion: Studies on TMPRSS4 and TM4SF5

The epithelial-mesenchymal transition (EMT) is one mechanism by which cells with mesenchymal features can be generated and is a fundamental event in morphogenesis. Recently, invasion and metastasis of cancer cells from the primary tumor are now thought to be initiated by the developmental process termed the EMT, whereby epithelial cells lose cell polarity and cell-cell interactions, and gain mesenchymal phenotypes with increased migratory and invasive properties. The EMT is believed to be an important step in metastasis and is implicated in cancer progression, although the influence of the EMT in clinical specimens has been debated. This review presents the recent results of two cell surface proteins, the functions and underlying mechanisms of which have recently begun to be demonstrated, as novel regulators of the molecular networks that induce the EMT and cancer progression.

Involvement of annexin A8 in the properties of pancreatic cancer

Although Annexin A8 (ANXA8), a member of a superfamily of calcium and phospholipid binding proteins, is physiologically expressed in a tissue-specific manner, recent microarray studies reported that ANXA8 was also ectopically expressed in pancreatic cancers. We investigated the molecular mechanism of expression of ANXA8 in cancer cells and its functional role in pancreatic cancer cells. ANXA8 was diversely expressed in human cancer cell lines. Expression was enhanced by treatment with 5-aza-dC and butyrate, and correlated with methylation status at CpG in the promoter-exon 1 region. Inhibition of ANXA8 using siRNA in BxPC-3 cells which express ANXA8 at a high level elevated caspase-3 and -7 activities. In in vitro invasion assay, inhibition of ANXA8 using siRNA in BxPC-3 reduced the numbers of migrating cells, and down-regulated HIF-1α mRNA transcription. Overexpression of ANXA8 increased the number of viable cells and BrdU incorporation in PANC-1 cells, which express ANXA8 at a low level. Expression of ANXA8 was induced under conditions of nutrient deprivation, and overexpression of ANXA8 showed resistance against serum starvation in PANC-1 cells. In a promoter assay, co-transfection with the expression vector of ANXA8 and the vector of a reporter gene containing the promoter of HIF-1α enhanced HIF-1α promoter activity. In contrast, this effect of ANXA8 was inhibited by administration of BAPTA-AM, an intracellular Ca²⁺ chelator. These results suggest that ectopic ANXA8 expression in cancer cells might involve an epigenetic mechanism. ANXA8 might play an important role in calcium fluctuation-mediated HIF-1α transcriptional activation and cell viability.

Expression of TMPRSS4 in non-small cell lung cancer and its modulation by hypoxia

Overexpression of TMPRSS4, a cell surface-associated transmembrane serine protease, has been reported in pancreatic, colorectal and thyroid cancers, and has been implicated in tumor cell migration and metastasis. Few reports have investigated both TMPRSS4 gene expression levels and the protein products. In this study, quantitative RT-PCR and protein staining were used to assess TMPRSS4 expression in primary non-small cell lung carcinoma (NSCLC) tissues and in lung tumor cell lines. At the transcriptional level, TMPRSS4 message was significantly elevated in the majority of human squamous cell and adenocarcinomas compared with normal lung tissues. Staining of over 100 NSCLC primary tumor and normal specimens with rabbit polyclonal anti-TMPRSS4 antibodies confirmed expression at the protein level in both squamous cell and adenocarcinomas with little or no staining in normal lung tissues. Human lung tumor cell lines expressed varying levels of TMPRSS4 mRNA in vitro. Interestingly, tumor cell lines with high levels of TMPRSS4 mRNA failed to show detectable TMPRSS4 protein by either immunoblotting or flow cytometry. However, protein levels were increased under hypoxic culture conditions suggesting that hypoxia within the tumor microenvironment may upregulate TMPRSS4 protein expression in vivo. This was supported by the observation of TMPRSS4 protein in xenograft tumors derived from the cell lines. In addition, staining of human squamous cell carcinoma samples for carbonic anhydrase IX (CAIX), a hypoxia marker, showed TMPRSS4 positive cells adjacent to CAIX positive cells. Overall, these results indicate that the cancer-associated TMPRSS4 protein is overexpressed in NSCLC and may represent a potential therapeutic target.

Using Representational Difference Analysis to detect changes in transcript expression of Aiptasia genes after laboratory exposure to lindane

Molecular stress responses to pesticide exposures represent an understudied area of cnidarian transcriptome investigations. The organochlorine pesticide lindane is known to disrupt normal neuron function. Cnidarians with simple nervous systems are recognized as sensitive indicators of water quality, yet nothing is known about cnidarian responses to lindane. Sea anemones (Aiptasia pallida) were exposed for 4h to lindane (20 μg/l). Because anemones have neurons and lindane is known to target neurons, it is anticipated that cnidarian stress responses will include changes in transcription of genes associated with neurons. Representational Difference Analysis (RDA) was utilized to isolate differentially transcribed genes in the anemones exposed to the pesticide. After two rounds of RDA hybridizations, 148 amplified fragments ranging in size from 150 to 800 bp were cloned. Sequencing and bioinformatic analyses of 106 clones revealed 56 different gene fragments. Virtual Northern dot blots were used as a preliminary screening tool to identify the most responsive RDA products. To further characterize the specificity of response, additional anemones were exposed to a series of lindane concentrations (0, 0.2, 2.0, 10, and 20 μg/l). Northern dot blots were subsequently used to develop expression profiles for selected RDA products over the range of pesticide concentrations. The seven most responsive RDA products represent genes with products associated with neuron development, immune responses, and Ca(2+) binding/transport. The resulting expression profiles illustrate that these RDA products exhibit various degrees of concentration specificity with some RDA products being significantly up-regulated at 20 μg/l while other RDA products are most responsive at concentrations <20 μg/l. Results also demonstrate how RDA can be used to identify potentially important biomarkers of organochlorine exposure while generating new hypotheses about important phenomena such as endocrine disruption in cnidarians.

Human tetraspanin transmembrane 4 superfamily member 4 or intestinal and liver tetraspan membrane protein is overexpressed in hepatocellular carcinoma and accelerates tumor cell growth

The human transmembrane 4 superfamily member 4 or intestinal and liver tetraspan membrane protein (TM4SF4/il-TMP) was originally cloned as an intestinal and liver tetraspan membrane protein and mediates density-dependent cell proliferation. The rat homolog of TM4SF4 was found to be up-regulated in regenerating liver after two-thirds hepatectomy and overexpression of TM4SF4 could enhance liver injury induced by CCl(4). However, the expression and significance of TM4SF4/il-TMP in liver cancer remain unknown. Here, we report that TM4SF4/il-TMP is frequently and significantly overexpressed in hepatocellular carcinoma (HCC). Real-time quantitative reverse transcription polymerase chain reaction (RT-PCR) and western blot analysis showed that TM4SF4/il-TMP mRNA and protein levels were up-regulated in ∼80% of HCC tissues. Immunohistochemical analysis of a 75 paired HCC tissue microarray revealed that TM4SF4/il-TMP was significantly overexpressed in HCC tissues (P< 0.001), and high immunointensity of TM4SF4/il-TMP tended to be in well-to-moderately differentiated HCC compared with poorly differentiated tumors. Functional studies showed that overexpression of TM4SF4/il-TMP in QGY-7701 and BEL-7404 HCC cell lines through stable transfection of TM4SF4 expression plasmid significantly promoted both cell growth and colony formation of HCC cells. Reduction of TM4SF4/il-TMP expression in QGY-7701 and BEL-7404 cells by stably transfecting TM4SF4 antisense plasmid caused great inhibition of cell proliferation. Our findings suggest that TM4SF4/il-TMP has the potential to be biomarker in HCC and plays a crucial role in promotion of cancer cell proliferation.

Long noncoding intronic RNAs are differentially expressed in primary and metastatic pancreatic cancer

Background

Pancreatic ductal adenocarcinoma (PDAC) is known by its aggressiveness and lack of effective therapeutic options. Thus, improvement in current knowledge of molecular changes associated with pancreatic cancer is urgently needed to explore novel venues of diagnostics and treatment of this dismal disease. While there is mounting evidence that long noncoding RNAs (lncRNAs) transcribed from intronic and intergenic regions of the human genome may play different roles in the regulation of gene expression in normal and cancer cells, their expression pattern and biological relevance in pancreatic cancer is currently unknown. In the present work we investigated the relative abundance of a collection of lncRNAs in patients' pancreatic tissue samples aiming at identifying gene expression profiles correlated to pancreatic cancer and metastasis.

Methods

Custom 3,355-element spotted cDNA microarray interrogating protein-coding genes and putative lncRNA were used to obtain expression profiles from 38 clinical samples of tumor and non-tumor pancreatic tissues. Bioinformatics analyses were performed to characterize structure and conservation of lncRNAs expressed in pancreatic tissues, as well as to identify expression signatures correlated to tissue histology. Strand-specific reverse transcription followed by PCR and qRT-PCR were employed to determine strandedness of lncRNAs and to validate microarray results, respectively.

Results

We show that subsets of intronic/intergenic lncRNAs are expressed across tumor and non-tumor pancreatic tissue samples. Enrichment of promoter-associated chromatin marks and over-representation of conserved DNA elements and stable secondary structure predictions suggest that these transcripts are generated from independent transcriptional units and that at least a fraction is under evolutionary selection, and thus potentially functional.

Statistically significant expression signatures comprising protein-coding mRNAs and lncRNAs that correlate to PDAC or to pancreatic cancer metastasis were identified. Interestingly, loci harboring intronic lncRNAs differentially expressed in PDAC metastases were enriched in genes associated to the MAPK pathway. Orientation-specific RT-PCR documented that intronic transcripts are expressed in sense, antisense or both orientations relative to protein-coding mRNAs. Differential expression of a subset of intronic lncRNAs (PPP3CB, MAP3K14 and DAPK1 loci) in metastatic samples was confirmed by Real-Time PCR.

Conclusion

Our findings reveal sets of intronic lncRNAs expressed in pancreatic tissues whose abundance is correlated to PDAC or metastasis, thus pointing to the potential relevance of this class of transcripts in biological processes related to malignant transformation and metastasis in pancreatic cancer.

Pancreatic intraepithelial neoplasia revisited and updated

Most pancreatic neoplasms are classified as ductal adenocarcinoma because they show a ductal phenotype, making a ductal origin very likely. The duct lesions that may give rise to pancreatic ductal adenocarcinoma have been called pancreatic intraepithelial neoplasia (PanIN). A classification system for these lesions distinguishes between three grades of PanIN. Molecular studies revealed that PanIN-2 and PanIN-3 lesions represent a distinct step towards invasive carcinoma. While high-grade PanINs are extremely rare in the normal pancreas, low-grade PanINs are common in individuals older than 40 years and may be associated with lobular fibrosis and intraductal papillary mucinous neoplasms of the gastric type. This disease spectrum has also been described in members of kindreds with familial pancreatic cancer. The natural history and cause of PanINs are unknown. As PanIN-1 lesions entail little risk, while PanIN-3 lesions are high-risk lesions, it would be of interest to target PanIN-2 lesions, which can be regarded as the starting point of progressive neoplastic changes that lead to invasive pancreatic ductal adenocarcinoma. Global gene expression analysis identified several differentially expressed genes which show enhanced expression in PanINs and may be used as potential biomarkers to facilitate diagnosis and therapy.

Tetraspanin in oncogenic epithelial-mesenchymal transition

Members of the L6 family of membrane proteins, a branch of the tetraspanin superfamily, are overexpressed in tumor cells from many types of cancers. However, direct evidence of their oncogenic activity has not been previously shown. In this issue of the JCI, Lee et al. demonstrate that overexpression of the tetraspanin superfamily member TM4SF5 in human hepatocellular carcinoma cells causes cellular phenotypic changes that resemble classical descriptions of epithelial-mesenchymal transition (EMT), with some unique aspects (see the related article beginning on page 1354). They also show that these TM4SF5-mediated effects trigger tumor formation when these cells are injected into mice. The study implicates TM4SF5, for the first time to our knowledge, in EMT oncogenic pathways of cancer progression.

New markers of pancreatic cancer identified through differential gene expression analyses: claudin 18 and annexin A8

BACKGROUND

New markers to distinguish benign reactive glands from infiltrating ductal adenocarcinoma of the pancreas are needed.

DESIGN

The gene expression patterns of 24 surgically resected primary infiltrating ductal adenocarcinomas of the pancreas were compared with 18 non-neoplastic samples using the Affymetrix U133 Plus 2.0 Arrays and the Gene Logic GeneExpress Software System. Gene fragments from 4 genes (annexin A8, claudin 18, CXCL5, and S100 A2) were selected from the fragments found to be highly expressed in infiltrating adenocarcinomas when compared with normal tissues. The protein expression of these genes was examined using immunohistochemical labeling of tissue microarrays.

RESULTS

Claudin 18 labeled infiltrating carcinomas in a membranous pattern. When compared with normal and reactive ducts, claudin 18 was overexpressed, at least focally, in 159 of 166 evaluable carcinomas (96%). Strong and diffuse claudin 18 overexpression was most often seen in well-differentiated carcinomas (P=0.02). Claudin 18 was overexpressed in 51 of 52 cases (98%) of pancreatic intraepithelial neoplasia. Annexin A8 was at least focally overexpressed in 149 of 154 evaluable infiltrating carcinomas (97%). S100 A2 was at least focally overexpressed in 118 of 154 evaluable infiltrating carcinomas (77%). Non-neoplastic glands also frequently expressed S100 A2 diminishing its potential diagnostic utility. Immunolabeling with antibodies directed against CXCL5 did not reveal any significant differences in protein expression between infiltrating adenocarcinomas and normal pancreatic ducts.

CONCLUSIONS

Claudin 18 and annexin A8 are frequently highly overexpressed in infiltrating ductal adenocarcinomas when compared with normal reactive ducts, suggesting a role for these molecules in pancreatic ductal adenocarcinomas. Furthermore, these may serve as diagnostic markers, as screening tests and as therapeutic targets.

Adrenomedullin is expressed in pancreatic cancer and stimulates cell proliferation and invasion in an autocrine manner via the adrenomedullin receptor, ADMR

The current study investigated adrenomedullin as a potential autocrine regulator of pancreatic cancer cell function. Adrenomedullin was localized in the neoplastic epithelium of 90% (43 of 48) of human pancreatic adenocarcinomas analyzed by immunohistochemistry and was expressed by 100% (8 of 8) of pancreatic cancer cell lines analyzed by reverse transcription-PCR. Pancreatic cancer cell lines also secreted adrenomedullin into the culture medium as determined by ELISA (5 of 5). Exogenous adrenomedullin treatment of Panc-1, BxPC3, and MPanc96 cells in vitro stimulated cell proliferation, invasion, and nuclear factor kappaB activity, indicating the ability of the cells to respond to adrenomedullin. Treatment of the cell cultures with an adrenomedullin antagonist inhibited basal levels of proliferation and nuclear factor kappaB activity, supporting the autocrine function of this molecule. Furthermore, increasing adrenomedullin levels by gene transfer to Panc-1 cells increased, whereas adrenomedullin small hairpin RNA silencing in MPanc96 cells inhibited tumor growth and metastasis in vivo. Adrenomedullin is able to act through at least two different receptors, adrenomedullin receptor (ADMR) and calcitonin receptor-like receptor (CRLR). Reverse transcription-PCR and Western blotting indicated that pancreatic cancer cells expressed only ADMR but not CRLR. In contrast, cells found in the tumor microenvironment, primary human pancreatic stellate and endothelial (HUVEC) cells, expressed both ADMR and CRLR. Small hairpin RNA silencing of ADMR in pancreatic cancer cells blocked adrenomedullin-induced growth and invasion, indicating that this receptor is involved in the autocrine actions of adrenomedullin. These data indicate that adrenomedullin acting via ADMR increases the aggressiveness of pancreatic cancer cells and suggests that these molecules may be useful therapeutic targets.

Differential gene expression in patients genetically predisposed to pancreatic cancer

BACKGROUND

Nearly 10% of all pancreatic cancer (PCA) results from genetic predisposition. Although abnormalities in sporadic PCA have been described, little is known about the genetics of heritable PCA. The purpose of this study was to identify novel genes expressed in patients with a presumed genetic predisposition or "familial" PCA.

PATIENTS AND METHODS

We defined "familial" PCA as patients having one or more first-degree relatives with biopsy-proven adenocarcinoma of the pancreas. Using a PCR-based subtractive and enrichment procedure, representational difference analysis (RDA), pancreatic tumor cDNA was reverse-transcribed from pooled poly(A)+ mRNA from six such patients (tester) and compared to pooled cDNA from five normal pancreata (driver). Tumor-specific gene fragments were identified and confirmed to be overexpressed in familial PCA by comparative RT-PCR. Six PCA cell lines, 11 sporadic tumors, 5 neuroendocrine tumors, and 3 chronic pancreatitis tissues were screened to determine the specificity of these genes.

RESULTS

Sequence analysis revealed several sequences of unknown significance and six genes previously described in neoplasia/carcinogenesis: Apolipoprotein A4, CEA, Keratin 19, Stratifin (14-3-3 sigma), Trefoil Factor, and Calcium Binding Protein S100 A6. Screening of cell lines and pancreatic tissue types showed varying degrees of specificity for familial and sporadic PCA. The APO-A4 gene was up-regulated in familial PCA.

CONCLUSIONS

The pattern of frequency in all screened tissue suggests that these genes are associated with conditions that produce significant desmoplastic responses and are difficult to differentiate from chronic inflammatory processes. Apolipoprotein A4 is preferentially expressed in familial patients, suggesting that the importance of fatty acid synthesis in carcinogenesis be investigated further.

Specialized DNA arrays for the differentiation of pancreatic tumors

PURPOSE

Malignant tumors of the pancreas are frequently indistinguishable from inflammatory tumors arising in the context of a chronic pancreatitis with the use of conventional imaging techniques. Thus, cytologic analysis of cells obtained by abdominal ultrasound, computed tomography, or endoscopic ultrasound-guided fine needle aspiration biopsy is required for diagnosis. However, the reliability of cytologic analyses of pancreatic fine needle aspirates remains unsatisfactory, with a diagnostic accuracy of < or =80%. The purpose of the current study was therefore to develop a novel diagnostic approach based on expression profiling of biopsy material using a specialized diagnostic cDNA array.

EXPERIMENTAL DESIGN

Previous gene expression profiling studies were reevaluated to design a 558-feature diagnostic array. Minimal amounts of residual material from pancreatic cytology samples as well as surgically resected tumor and control tissue specimens were analyzed using the diagnostic array and a newly developed statistical classification system.

RESULTS AND CONCLUSIONS

Our diagnostic approach resulted in 95% accurate differentiation between ductal adenocarcinomas and nonmalignant tumors of the pancreas. The diagnostic array, in conjunction with conventional diagnostic procedures, is thus suitable to significantly improve the reliability of pancreatic cancer diagnostics and can be expected to become a valuable new tool in the routine workup of suspect masses in the pancreas.

Gene expression profiling of microdissected pancreatic ductal carcinomas using high-density DNA microarrays

Pancreatic ductal adenocarcinoma (PDAC) remains an important cause of malignancy-related death and is the eighth most common cancer with the lowest overall 5-year relative survival rate. To identify new molecular markers and candidates for new therapeutic regimens, we investigated the gene expression profile of microdissected cells from 11 normal pancreatic ducts, 14 samples of PDAC, and 4 well-characterized pancreatic cancer cell lines using the Affymetrix U133 GeneChip set. RNA was extracted from microdissected samples and cell lines, amplified, and labeled using a repetitive in vitro transcription protocol. Differentially expressed genes were identified using the significance analysis of microarrays program. We found 616 differentially expressed genes. Within these, 140 were also identified in PDAC by others, such as Galectin-1, Galectin-3, and MT-SP2. We validated the differential expression of several genes (e.g., CENPF, MCM2, MCM7, RAMP, IRAK1, and PTTG1) in PDAC by immunohistochemistry and reverse transcription polymerase chain reaction. We present a whole genome expression study of microdissected tissues from PDAC, from microdissected normal ductal pancreatic cells and pancreatic cancer cell lines using high-density microarrays. Within the panel of genes, we identified novel differentially expressed genes, which have not been associated with the pathogenesis of PDAC before.

Comprehensive proteomic analysis of human pancreatic juice

Proteomic technologies provide an excellent means for analysis of body fluids for cataloging protein constituents and identifying biomarkers for early detection of cancers. The biomarkers currently available for pancreatic cancer, such as CA19-9, lack adequate sensitivity and specificity contributing to late diagnosis of this deadly disease. In this study, we carried out a comprehensive characterization of the "pancreatic juice proteome" in patients with pancreatic adenocarcinoma. Pancreatic juice was first fractionated by 1-dimensional gel electrophoresis and subsequently analyzed by liquid chromatography tandem mass spectrometry (LC-MS/MS). A total of 170 unique proteins were identified including known pancreatic cancer tumor markers (e.g., CEA, MUC1) and proteins overexpressed in pancreatic cancers (e.g., hepatocarcinoma-intestine-pancreas/pancreatitis-associated protein (HIP/PAP) and lipocalin 2). In addition, we identified a number of proteins that have not been previously described in pancreatic juice (e.g., tumor rejection antigen (pg96) and azurocidin). Interestingly, a novel protein that is 85% identical to HIP/PAP was identified, which we have designated as PAP-2. The proteins identified in this study could be directly assessed for their potential as biomarkers for pancreatic cancer by quantitative proteomics methods or immunoassays.

DNA microarray analysis of pancreatic malignancies

Pancreatic ductal adenocarcinoma (PDAC) has an extremely poor prognosis. To improve the prognosis, novel molecular markers and targets for earlier diagnosis and adjuvant and/or neoadjuvant treatment are needed. Recent advances in human genome research and high-throughput molecular technologies make it possible to cope with the molecular complexity of malignant tumors. With DNA array technology, mRNA expression levels of thousand of genes can be measured simultaneously in a single assay. As several studies using microarrays in PDAC have already been published, this review attempts to compare the published data and therefore to validate the results. In addition, the applied techniques are discussed in the context of pancreatic malignancies.

Panel of genes transcriptionally up-regulated in squamous cell carcinoma of the cervix identified by representational difference analysis, confirmed by macroarray, and validated by real-time quantitative reverse transcription-PCR

BACKGROUND

The Pap smear is currently the most widely used method of screening for squamous cell carcinoma of the cervix (SCCC). Because it is based on cell morphology, it is subject to variability in interpretation. Sensitive molecular markers capable of differentiating cancerous samples from noncancerous ones would be beneficial in this regard.

METHODS

We performed representational difference analysis (RDA) using paired, noncancerous (normal) and cancerous (disease) tissues taken from the same specimen obtained from a single patient with a confirmed diagnosis of SCCC. Linearly amplified cDNA from normal and diseased tissues of the original patient and seven others were hybridized to DNA macroarrays containing the candidate gene transcript fragments. Real-time quantitative reverse transcription-PCR was used to validate the macroarray results.

RESULTS

RDA identified a candidate pool of 65 transcript fragments up-regulated in diseased tissue compared with normal tissue. Forty-one transcripts were found to be up-regulated in diseased compared with normal tissue in at least one half the patients by macroarray hybridization. Eleven of those genes were selected for real-time quantitative reverse transcription-PCR analysis, and all were confirmed as transcriptionally up-regulated in cancer compared with normal tissue in at least one half the patients.

CONCLUSIONS

RDA using tissues from a single patient identified gene fragments confirmed to be transcriptionally up-regulated in SCCC both in the original patient and in seven others. The confirmed genes have a variety of functions and also have the potential to serve as diagnostic or prognostic markers.

Microarray-based gene expression profiling in pancreatic ductal carcinoma: status quo and perspectives

BACKGROUND

Pancreatic ductal adenocarcinoma has an extremely poor prognosis. To improve the prognosis, novel molecular markers and targets for earlier diagnosis and adjuvant and/or neoadjuvant treatment need to be identified. One of the key techniques that has been developed to achieve this goal is DNA microarray profiling, which is used to identify the mechanisms of deregulated molecular functions in pancreatic carcinoma cells.

OBJECTIVE

As several studies using microarrays have already been published, this review attempts to compare published data and crossvalidate the results. In addition, the applied techniques are discussed.

[Necessity and usefulness of bioinformatic methods for microarray data analysis]

Data emerging from DNA microarray experiments are usually difficult to interpret. While the level of expression of several thousand genes can be measured in a single experiment, only a few dozen experiments are normally carried out, leading to data sets of very high dimensionality and low cardinality. The computational analysis of gene expression data makes significant usage of machine learning and statistical methods. Nevertheless, caution should be used in the blind adoption of these methods, as this usually leads to an over-interpretation of the expression profiles. The following presentation provides an overview of up-to-date principles of biostatistical analysis. A potential application for the analysis of high-dimensional expression profiles of prostate cancer is given.

Expression profiling of the influence of RAS mutants on the TGFB1-induced phenotype of the pancreatic cancer cell line PANC-1

Expression profiling analyses were used to elucidate the functional relevance of RAS proteins in mediating the effect of TGFB1 on the transcriptional phenotype of the pancreatic cancer cell line PANC-1. Despite the presence of one mutated KRAS2 allele in parental PANC-1 pancreatic cancer cells, RAS-dependent signal transduction remained susceptible to stimulation by EGF and TGFB1. To analyze the impact of RAS proteins on the TGFB1-induced transcriptional phenotype, we used PANC-1 cells stably transfected with a dominant negative HRAS(S17N) mutant or with a constitutively active KRAS2(G12V) mutant. TGFB1 treatment of mock-transfected PANC-1 cells led to an expression profile suggestive of epithelial-mesenchymal transdifferentiation (EMT). Profiling of the HRAS(S17N)-expressing clone demonstrated that induction of endogenous RAS activity by TGFB1 is required for the development of the TGFB1-induced transcriptional phenotype of PANC-1 cells. The expression of the KRAS2(G12V) mutant by itself repressed transcription of markers of epithelial differentiation and induced transcription of several extracellular matrix-associated genes. This effect was not enhanced further by TGFB1 treatment. In contrast, transcript levels of genes associated with proliferation and cell cycle progression did not appear to be the primary targets of the synergism between the RAS- and TGFB1-dependent cascades. The introduction of the dominant negative and the constitutively active RAS mutants induced partly overlapping and partly inverse effects on the TGFB1-induced expression profile of PANC-1 cells. Additional mechanisms such as the induction of autocrine loops and the use of different RAS isoforms or alternate, ERK-independent signaling pathways may be involved in the interaction between the RAS- and the TGFB1-dependent signaling cascades.

Systematic isolation of genes differentially expressed in normal and cancerous tissue of the pancreas

BACKGROUND

There is increasing knowledge about the genetic basis of pancreatic cancer (PaCa). Tumor suppressor genes (TSGs; e.g. p53 and DPC4) and oncogenes (e.g. K-ras) have been shown to be involved in the development of PaCa. However, the extent of chromosomal changes (gains and losses) implicates that many more genes may be involved in the multistep progression of PaCa. Identification of these genes is essential for understanding the molecular events in the development of PaCa.

METHODS

We assembled public and proprietary libraries of more than 4 million expressed sequence tags using newly developed software tools.

RESULTS

We identified a total of 249 genes with specific expression patterns in normal and cancerous tissue of the pancreas. Of these, 27 genes were found to be preferentially expressed in normal tissue of the pancreas, while 222 genes showed significant upregulation of expression in PaCa. Of the 249 genes, 232 (93.2%) were found to represent known human genes or putative human homologues of genes characterized previously in other species, while 17 (6.8%) represent putative new genes.

CONCLUSION

These genes may represent a valuable source to identify novel TSGs and oncogenes involved in the carcinogenesis of PaCa.

Gene expression profiles of microdissected pancreatic ductal adenocarcinoma

In a search for new molecular markers of pancreatic ductal adenocarcinoma (PDAC), we compared the gene expression profiles of seven pancreatic carcinomas and one carcinoma of the papilla Vateri with those of duct cells from three non-neoplastic pancreatic tissues. In addition, the human pancreatic duct cell line and five PDAC cell lines (AsPC-1, BxPC-3, Capan-1, Capan-2, HPAF) were examined. RNA was extracted from microdissected tissue or cultured cell lines and analysed using a custom-made Affymetrix Chip containing 3023 genes, of which 1000 were known to be tumour associated. Hierarchical clustering revealed 81 differentially expressed genes. Of all the genes, 26 were downregulated in PDAC and 14 were upregulated in PDAC. In PDAC cell lines versus normal pancreatic duct cells, 21 genes were downregulated and 20 were upregulated. Of these 81 differentially expressed genes, 15 represented human genes previously implicated in the tumourigenesis of PDAC. From the genes that were so far not known to be associated with PDAC tumorigenesis, we selected ADAM9 for further validation because of its distinct overexpression in tumour tissue. Using immunohistochemistry, the over-expressed gene, ADAM9, was present in 70% of the PDACs analysed. In conclusion, using microarray technology we were able to identify a set of genes whose aberrant expression was associated with PDAC and may be used to target the disease.

Nuclear annexin II negatively regulates growth of LNCaP cells and substitution of ser 11 and 25 to glu prevents nucleo-cytoplasmic shuttling of annexin II

BACKGROUND

Annexin II heavy chain (also called p36, calpactin I) is lost in prostate cancers and in a majority of prostate intraepithelial neoplasia (PIN). Loss of annexin II heavy chain appears to be specific for prostate cancer since overexpression of annexin II is observed in a majority of human cancers, including pancreatic cancer, breast cancer and brain tumors. Annexin II exists as a heterotetramer in complex with a protein ligand p11 (S100A10), and as a monomer. Diverse cellular functions are proposed for the two forms of annexin II. The monomer is involved in DNA synthesis. A leucine-rich nuclear export signal (NES) in the N-terminus of annexin II regulates its nuclear export by the CRM1-mediated nuclear export pathway. Mutation of the NES sequence results in nuclear retention of annexin II.

RESULTS

Annexin II localized in the nucleus is phosphorylated, and the appearance of nuclear phosphorylated annexin II is cell cycle dependent, indicating that phosphorylation may play a role in nuclear entry, retention or export of annexin II. By exogenous expression of annexin II in the annexin II-null LNCaP cells, we show that wild-type annexin II is excluded from the nucleus, whereas the NES mutant annexin II localizes in both the nucleus and cytoplasm. Nuclear retention of annexin II results in reduced cell proliferation and increased doubling time of cells. Expression of annexin II, both wild type and NES mutant, causes morphological changes of the cells. By site-specific substitution of glutamic acid in the place of serines 11 and 25 in the N-terminus, we show that simultaneous phosphorylation of both serines 11 and 25, but not either one alone, prevents nuclear localization of annexin II.

CONCLUSION

Our data show that nuclear annexin II is phosphorylated in a cell cycle-dependent manner and that substitution of serines 11 and 25 inhibit nuclear entry of annexin II. Aberrant accumulation of nuclear annexin II retards proliferation of LNCaP cells.

Gene expression in neoplasms of the pancreas: applications to diagnostic pathology

It seems that, lately, every pathology journal has three or four articles documenting the discovery of another genetic alteration or describing global gene expression in a series of cancers. Although these discoveries provide insight into the biology of neoplasia, it is less clear how they can be quickly and efficiently translated to patient care. This review will use neoplasms of the pancreas as a model and show how recent discoveries of genetic alterations and gene expression patterns can have a significant impact on the diagnosis and even treatment of tumors. Emphasis will be placed on applications that are practical and useful to the daily practice of pathology.

Exploration of global gene expression patterns in pancreatic adenocarcinoma using cDNA microarrays

Pancreatic cancer is the fifth leading cause of cancer death in the United States. We used cDNA microarrays to analyze global gene expression patterns in 14 pancreatic cancer cell lines, 17 resected infiltrating pancreatic cancer tissues, and 5 samples of normal pancreas to identify genes that are differentially expressed in pancreatic cancer. We found more than 400 cDNAs corresponding to genes that were differentially expressed in the pancreatic cancer tissues and cell lines as compared to normal pancreas. These genes that tended to be expressed at higher levels in pancreatic cancers were associated with a variety of processes, including cell-cell and cell-matrix interactions, cytoskeletal remodeling, proteolytic activity, and Ca(++) homeostasis. Two prominent clusters of genes were related to the high rates of cellular proliferation in pancreatic cancer cell lines and the host desmoplastic response in the resected pancreatic cancer tissues. Of 149 genes identified as more highly expressed in the pancreatic cancers compared with normal pancreas, 103 genes have not been previously reported in association with pancreatic cancer. The expression patterns of 14 of these highly expressed genes were validated by either immunohistochemistry or reverse transcriptase-polymerase chain reaction as being expressed in pancreatic cancer. The overexpression of one gene in particular, 14-3-3 sigma, was found to be associated with aberrant hypomethylation in the majority of pancreatic cancers analyzed. The genes and expressed sequence tags presented in this study provide clues to the pathobiology of pancreatic cancer and implicate a large number of potentially new molecular markers for the detection and treatment of pancreatic cancer.

Use of DNA arrays/microarrays in pancreatic research

In recent years, enormous technical advances in experimental protocols as well as robotic and bioinformatic techniques have allowed DNA array/microarray technology to emerge as the leading technology in the field of functional, disease-related genome analysis. Multiple applications exist for DNA arrays/microarrays including comparative genomic analysis to identify chromosomal imbalances (Matrix-CGH), the study of mutations and genetic polymorphisms, and the study of gene expression (expression profiling). Expression profiling is the most widely used application of DNA array/microarray technology and allows to measure gene expression of thousands of genes simultaneously. The present review describes the basic principles of expression profiling analyses and outlines some applications in pancreatic cancer research.

Characterization of gene expression profiles in intraductal papillary-mucinous tumors of the pancreas

The molecular pathology of precursor lesions leading to invasive pancreatic ductal adenocarcinomas remains relatively unknown. We have applied cDNA microarray analysis to characterize gene expression profiles in a series of intraductal papillary-mucinous tumors (IPMTs) of the pancreas, which represents one of the alternative routes of intraepithelial progression to full malignancy in the pancreatic duct system. Using a cDNA microarray containing 4992 human genes, we screened a total of 13 IPMTs including nine noninvasive and four invasive cases. Expression change in more than half of the tumors was observed for 120 genes, ie, 62 up-regulated and 58 down-regulated genes. Some of the up-regulated genes in this study have been previously described in classical pancreatic carcinomas such as lipocalin 2, galectin 3, claudin 4, and cathepsin E. The most highly up-regulated genes in IPMTs corresponded to three members of the trefoil factor family (TFF1, TFF2, and TFF3). Immunohistochemistry performed on five genes found to be differentially expressed at the RNA level (TFF1, TFF2, TFF3, lipocalin 2, and galectin 3) showed a good concordance between transcript level and protein abundance, except for TFF2. Hierarchical clustering organized the cases according to the dysplastic and invasive phenotype of theIPMTs. This analysis has permitted us to implicate several genes (caveolin 1, glypican 1, growth arrest-specific 6 protein, cysteine-rich angiogenic inducer 61) in tumor progression. The observation that several genes are differentially expressed both in IPMTs and pancreatic carcinomas suggests that they may be involved at an early stage of pancreatic carcinogenesis.

Gene expression profiles of pancreatic cancer and stromal desmoplasia

Gene expression studies were undertaken in normal pancreas and pancreatic adenocarcinomas to determine new candidate genes that can potentially be used as markers of the disease. The characteristic desmoplastic stromal reaction of pancreatic adenocarcinoma greatly hampers expression studies in this tumour type, and usually necessitates time-consuming tissue microdissection for enrichment of the tumour cell population. We show that fine needle aspiration of cancer provides a fast and efficient way of obtaining samples highly enriched in tumour cells with sufficient yields of RNA. Using Atlas cancer cDNA arrays with 588 cancer-related genes, we describe gene expression profiles of normal pancreas, bulk pancreatic tumour tissues and pancreatic tumour aspirates containing more than 95% tumour cells. Analysis of bulk tissue specimens revealed differentially expressed genes belonging predominantly to the stromal component of the tumour. This contrasted with the results obtained from tumour-cell enriched samples. Several genes already described in pancreatic cancer (caspase 8, TIMP1, CD9, IL-13) were also differentially expressed in our study. Furthermore, we found dysregulated expression of genes not previously associated with pancreatic adenocarcinoma, such as Rac 1, GLG1, NEDD5, RPL-13a, RPS9 and members of the Wnt5A gene family. In summary, we present a panel of genes newly identified in the pathogenesis of pancreatic adenocarcinoma and demonstrate that fine needle aspirates of the tumour mass are a convenient source of material for gene expression studies in tumours accompanied by desmoplastic reactions.

Identification of genes induced by inflammatory cytokines in airway epithelium

Epithelial cells lining the airways are thought to play a prominent role in respiratory diseases. We utilized cDNA representational difference analysis to identify the genes in which expression is induced by the proinflammatory cytokines tumor necrosis factor-alpha and interleukin-1beta in primary human bronchial epithelial cells and hence are relevant to airway inflammation. Hybridization of the subtraction product to arrayed cDNAs indicated that known tumor necrosis factor-alpha- and interleukin-1beta-inducible genes such as B94, Zfp36, and regulated on activation normal T cell expressed and secreted were represented, confirming the success of the subtraction experiment. A 1,152-clone library potentially representing genes with higher transcript levels in cytokine-treated human bronchial epithelial cells was generated and sequenced. Sequence similarity searches indicated that these clones represented 57 genes of known function, 1 gene of unknown function, 6 expressed sequence tags, and 2 novel sequences. The expression of 19 of these clones was studied by a combination of Northern blotting and RT-PCR analyses and confirmation of differential expression for 10 known genes, 2 expressed sequence tags, and a novel sequence not represented in any of the public databases was obtained. Thus cDNA representational difference analysis was utilized to isolate known and novel differentially expressed genes, which putatively play a role in airway inflammation.

Applied genome research in the field of human vaccines

Prophylactic vaccination has made an essential contribution to the improvement of human health over the 20th century. However, we still lack efficient vaccines against major human diseases such as malaria or tuberculosis. Today, the design of therapeutic vaccines referred to as 'pharmaccines' is actively investigated in order to treat diseases such as cancer. In that context, novel ways to rationalize and accelerate vaccine discovery are needed. A series of advances in the fields of molecular biology and computer science, have greatly accelerated the rate at which candidate vaccine antigens can be discovered. In this review, we will present and discuss how applied genome research may facilitate antigen discovery and the design of new prophylactic and therapeutic vaccines.

A cDNA RDA protocol using solid-phase technology suited for analysis in small tissue samples

cDNA representational difference analysis (cDNA RDA) is a PCR-based subtractive enrichment procedure for the cloning of differentially expressed genes. In this study, we have further developed the procedure to take advantage of solid-phase technology, and to facilitate the use of RDA when starting material is limited. Several parameters of the PCR-based generation of cDNA representations were investigated, and a solid-phase based purification step was introduced to simplify removal of digested adapter-ends and uncleaved fragments. The use of magnetic particles increased the speed of the method, and also eliminated the risk of carry-over contamination between iterative steps of subtraction and PCR amplification. The modified protocol was evaluated in monitoring differences in gene expression in (i) a rat system consisting of livers with and without growth hormone treatment, and in (ii) a human system consisting of normal colon and colon cancer.

Use of representational difference analysis and cDNA arrays for transcriptional profiling of tumor tissue

Representational difference analysis of cDNA (cDNA-RDA) was used for a comparison of the global transcript level of tumor of the larynx and the corresponding normal epithelial tissue toward the end of detecting differentially expressed genes. Overall, some 130 gene fragments were identified. By sequence analysis and homology comparison, they could be put into several groups related to (potential) functions. Apart from genes whose overexpression was most likely a result of tumor growth or dedifferentiation of epithelial tissue, a lot of genes were isolated that play major roles in signal transduction pathways or apoptosis or act as oncogenes or tumor suppressor genes, in addition to new, entirely unknown genes. Moreover, some cDNAs of known genes were identified that derived from unconventional splicing activity or other transcript modifications. All identified fragments were arrayed on solid support and used for reverse Northern blot analyses. The use of preselected RDA fragments as targets in array-based profiling experiments circumvents many of the problems encountered when dealing with large clone libraries.

Use of representational difference analysis to study the effect of TGFB on the expression profile of a pancreatic cancer cell line

It has been shown that TGFBs, their receptors, or downstream targets show genetic alterations in pancreatic cancer. This study was designed to identify transcriptional alterations induced by prolonged treatment of pancreatic cancer cell lines with TGFB. The TGFB-responsive PANC-1 cell line was treated with 10-ng/ml TGFB1 for 24 hr. cDNA representational difference analysis was used to generate subtracted hybridization probes enriched for TGFB regulated genes. These probes were hybridized on gridded arrays of cDNA clones containing genes differentially expressed in pancreatic cancer. Twenty-seven distinct cDNA clones were shown to be TGFB target genes. Eleven genes were upregulated by TGFB and were associated with extracellular matrix composition and formation, including genes usually transcribed by cells of mesenchymal origin only. Transcript levels of 16 genes were downregulated by TGFB and could mainly be classified into markers of epithelial differentiation and genes involved in the transcriptional and translational machinery. In conclusion, a 24-hr treatment of PANC-1 cells with TGFB induced a loss of epithelial and a gain of mesenchymal markers. As in other tumors, this epithelial-mesenchymal transdifferentiation may be of general importance during pancreatic carcinogenesis, and may participate, e.g., in the development of the desmoplastic reaction or the acquisition of an invasive phenotype of pancreatic tumor cells. This study demonstrates the usefulness of cDNA RDA and gridded clone libraries to study the effect of signaling cascades on the expression profile of tumor cells. Similar approaches may be helpful in the context of the genome project for the characterization of novel genes. Genes Chromosomes Cancer 26:70-79, 1999.

Screening for genes up-regulated in 5/6 nephrectomized mouse kidney

BACKGROUND

In diabetic and nondiabetic renal diseases, glomerular hyperfiltration is believed to play a central role in the subsequent progression of glomerulosclerosis and interstitial renal scarring. To identify genes involved in the process of hyperfiltration and hypertrophy, a polymerase chain reaction (PCR)-based subtraction method, that is, representational difference analysis of cDNA (cDNA-RDA), was employed.

METHODS

Ten-week-old ICR mice were 5/6 nephrectomized and sham operated. After two weeks, mRNAs were isolated from control and remnant kidneys and were subjected to the cDNA-RDA procedure.

RESULTS

We identified 10 known and 9 novel genes. Among 19 clones, 12 clones (8 known and 4 novel) showed 1.5- to 6-fold up-regulation by Northern blot analyses. The remaining seven clones were rarely expressed genes and were barely detected by Northern blot analyses, and their up-regulated expression was confirmed by Southern blot analysis using the PCR-amplified representative amplicons. The known genes included kidney androgen-regulated protein, major urinary protein, lysozyme M, metalloproteinase-3 tissue inhibitor, chaperonin 10, cytochrome oxidase I, epsilon-sarcoglycan, ribosomal protein S3a, G-proteingamma10 subunit, and splicing factor 9G8. All of the isolated known genes have not been reported to be up-regulated in the nephrectomized mouse kidney and suggest the possible role of androgen action, mitochondrial functions, matrix metabolism, cell-matrix interactions, and intracellular signaling events in the initiation of the progressive renal injury of the remnant kidney. Furthermore, cDNA-RDA facilitates the discovery of novel genes, including two kidney-specific genes.

CONCLUSIONS

The isolated known and novel genes may be involved in the pathobiological process of initial hyperfiltration and hypertrophy of remnant kidney.

Strategies for the detection of disease genes in pancreatic cancer

The present review summarizes our strategies aimed at identifying and characterizing genetic alterations occurring at the transcriptional and chromosomal level in pancreatic cancer. To study transcriptional alterations we have used a number of techniques including modified versions of differential hybridizations and cDNA RDA (representational difference analysis). These approaches have led to the identification of more than 500 genes with differential expression in pancreatic cancer. To study chromosomal aberrations occurring in pancreatic cancer tissues we used comparative genomic hybridization (CGH). This allowed the identification of a number of chromosomal regions containing putative tumor suppressor genes or oncogenes. Genes isolated in both approaches represent potential new disease genes for pancreatic cancer and are at present being characterized by individual or serial analysis.

Isolation of Human Transcripts Expressed in Hamster Cells from YACs by cDNA Representational Difference Analysis

Gene isolation methods used during positional cloning rely on physical contigs consisting of bacterial artificial chromosomes, P1, or cosmid clones. However, in most instances, the initial framework for physical mapping consists of contigs of yeast artificial chromosome (YACs), large vectors that are suboptimal substrates for gene isolation. Here we report a strategy to identify gene sequences contained within a YAC by using cDNA representational difference analysis (RDA) to directly isolate transcripts expressed from the YAC in mammalian cells. The RDA tester cDNAs were generated from a previously reported hamster cell line derived by stable transfer of a 590-kb YAC (911D5) that expressed NPC1, the human gene responsible for Niemann-Pick type C (NP-C). The driver cDNAs were generated from a control hamster cell line that did not contain the YAC that expressed NPC1. Among the gene fragments obtained by RDA,NPC1 was the most abundant product. In addition, two non-NPC1 fragments were isolated that were mapped to and expressed from 911D5. One of these RDA gene fragments (7-R) spans more than one exon and has 98% sequence identity with a human cDNA clone reported previously as an expressed sequence tag (EST), but not mapped to a chromosomal region. The other fragment (2-R) that had no significant sequence similarities with known mammalian genes or ESTs, was further localized to the region of overlap between YACs911D5 and 844E3. The latter YAC is part of a contig across the NP-C candidate region, but does not contain NPC1. This two-part approach in which stable YAC transfer is followed by cDNA RDA should be a useful adjunct strategy to expedite the cloning of human genes when a YAC contig is available across a candidate interval.

[The sequence data described in this paper have been submitted to GenBank under accession nos. AF117641 and AF117642.]

Detection of differentially expressed genes in primary tumor tissues using representational differences analysis coupled to microarray hybridization

The identification of differential gene expressionbetween cells is a frequent goal in modern biological research. Here we demonstrate the coupling of representational difference analysis (RDA) of cDNA with microarray analysis of the output for high throughput screening. Two primary Ewing's sarcoma tissue samples with different biological behavior in vivo were compared by RDA: one which was metastatic and progressed rapidly; the other localized and successfully treated. A modified RDA protocol that minimizes the necessary starting material was employed. After a reduced number of subtractive rounds, the output of RDA was shotgun cloned into a plasmid vector. Inserts from individual colonies from the subtracted library were amplified with vector-specific primers and arrayed at high density on glass slides. The arrays were then hybridized with differentially fluorescently labeled starting amplicons from the two tissues and fluorescent signals were measured at each DNA spot. We show that the relative amounts of fluorescent signal correlate well with the abundance of fragments in the RDA amplicon and in the starting mRNA. In our system, we analyzed 192 products and 173 (90%) were appropriately detected as being >2-fold differentially expressed. Fifty unique, differentially expressed clones were identified. Therefore, the use of RDA essentially provides an enriched library of differentially expressed genes, while analysis of this library with microarrays allows rapid and reproducible screening of thousands of DNA molecules simultaneously. The coupling of these two techniques in this system resulted in a large pool of differentially expressed genes.

Expressed genes in interleukin-4 treated B cells identified by cDNA representational difference analysis

Interleukin-4 (IL-4) stimulates B cell growth and differentiation, such as inducing mature B cells to switch to IgG1 and IgE production. To further characterize IL-4 effects on B cells, we used a sensitive PCR-based subtraction approach to isolate genes expressed in IL-4 treated cells. Our approach combined an adaptation of the genomic representational difference analysis (RDA) method to cDNA analysis with a physical separation method (magnetic bead depletion). This cDNA RDA technique allowed us to perform subtraction on the relatively small number of highly, characterized, purified B cells that can be conveniently prepared. In the hopes of removing genes responsible for general cell growth, we subtracted cDNA made from lipopolysaccharide (LPS)-stimulated B cells from cDNA from LPS+IL-4 stimulated B cells. Two rounds of subtraction resulted in greater than 100-fold enhancement of expected IL-4-induced Cgamma1 cDNA. At that point, we cloned this subtraction library and analysed 154 randomly picked clones for sequence similarities. From these clones, 37 individual genes were obtained. Most of these genes (30) could be functionally identified by sequence similarity. These included genes encoding Cgamma1 (1), cytoskeletal components (4) and products involved in DNA replication (3), metabolism (5), signal transduction (4), transcription (4), translation (6) and transport (3). Only 7 genes had no similarity to known sequences in the GenBank, EMBL or Swiss Prot databases. One unknown gene (designated Fig1 for IL-Four Induced Gene 1) and one gene with homology to the human transcription factor E4BP4 were confirmed by Northern blot analysis to be induced 10-20-fold by IL-4 treatment. This list of expressed genes in LPS + IL-4 treated B cells may shed further insight on the action and mechanism of IL-4 stimulation of cells.

Application of differential cDNA screening techniques to the identification of unique gene expression in tumours and lymphocytes

Development, differentiation and cell death in all organisms are mediated by tightly regulated programs of differential gene expression. Furthermore, changes in gene expression profiles are responsible for tumour formation and tumour progression, as well as for many other human diseases. Thus, the definition of distinct patterns of gene expression is not only essential for understanding complex biological processes but also leads to the identification of novel targets for therapy of various diseases.

Identification and cloning of differentially expressed genes

Only a few of the methods currently used for identification of differentially expressed genes take advantage of the fact that (near) complete sets of cDNA clones and sequences representing all human and mouse genes will be available for high throughput survey of gene expression. Accordingly, strategies based on hybridization of complex (cDNA or RNA) probes to cDNA microarrays, either on glass slides or on chips, are likely to become increasingly more advantageous. Recognizing, however, that the power of these methods depends upon the availability of such resources, strategies are being pursued to facilitate completion of the ongoing efforts to identify all human and mouse genes.