Metabolomics,Unknown,Transcriptomics,Genomics,Proteomics

Dataset Information

Gene expression analysis of relapse and no-relapse Wilms tumor

ABSTRACT: Wilms tumor (WT), the commonest renal tumor of the childhood, is originated from pluripotent metanephric blastemal cells resulting in tumors with triphasic histology, composed by blastemal, epithelial, and stromal cells. Histological subtype and tumor stage have long been recognized as important prognostic factors and used for the current therapeutic approach with blastema considered as the component with the highest clinical impact. Current clinical and molecular research is directed toward identifying prognostic factors for both purposes, definition of the minimal or intense required therapy for successful treatment of children with low or high risk of relapse, respectively. In this study, to identify molecular prognostic markers predictive of tumor relapse, samples from 26 advanced (Stage III or IV) blastemal WT, whose patients presented (13 samples) or did not present (13 samples) relapse, were interrogated by 4,608 human genes immobilized in a customized cDNA platform. This analysis revealed a set of 69 differentially expressed genes, where the nine top genes were evaluated by qRT-PCR in the initial WT samples. TSPAN3, NCOA6, CDO1, MPP2 and MCM2 were technically confirmed showing down-regulation in relapse WT. TSPAN3 and NCOA6 were validated in an independent sample group. Gene trios containing these two genes combined with one of the remaining three reported a reasonable capability of discriminating relapse and no-relapse WT. Negative protein expression of MCM2 and NCOA6 was observed in around 62% and 72% of 34 independent stage III and IV WT patients, respectively without association with relapse. However, a significant association between MCM2 negative staining and chemotherapy as first treatment was observed suggesting that MCM2 protein expression is implicated anyhow with chemotherapy treatment in WT. Twenty-six blastemal-enriched advanced stage (III or IV) WT samples from sporadic and favorable histology WT were received from Children Oncology Group (COG) and used were used for microarray experiments, from which 13 samples were from patients without relapse in 5 years of follow up. Tumors were taken from primary nephrectomy at the time of initial diagnosis. Blastemal histological component was isolated following protocol described in Maschietto et al. (2008). Only samples with >80% of blastemal cells were used. Total RNA was isolated using TRIzol (Sigma) according to manufacturer´s instructions. RNA quantity and integrity was assessed by spectrophotometry (Nanodrop) and microfluidics-based electrophoresis (Agilent 2100 Bioanalyzer, Agilent), respectively, selecting only RNA samples with OD around 2.0 and RIN>5.0. A two-round RNA amplification procedure (Gomes et al. 2003) was carried out for all samples. Labeled cDNA was generated in a reverse transcriptase reaction using 4 ug of amplified RNA (aRNA), random hexamer primer (Invitrogen Life Technologies, Carlsbad, CA), Alexa-3 or 5-labeled dCTP (Amersham, Biosciences) and IMPROM (PROMEGA) using oligo-dT7(24), following manufacture recommendations. Five µg of test and reference cDNA reverse color Alexa-labeled aRNA targets were competitively hybridized against the cDNA probes in a customized cDNA platform with 4,608 ORESTES representing human genes. Dye-swap was performed for each sample as control for dye bias and used as replicate. Pre-hybridization and hybridization were performed as described (Maschietto et al. 2008). A pool of RNAs obtained from 15 distinct human cell lines as was used as reference. Signal intensities capture and analysis were performed as described in Maschietto et al (2008).

ORGANISM(S): Homo sapiens

SUBMITTER: Mariana Maschietto

PROVIDER: E-GEOD-22696 | biostudies-arrayexpress |

REPOSITORIES: biostudies-arrayexpress

ACCESS DATA

Similar Datasets

Project description:Precise regulation of transduction signaling pathways genes is crucial for correct differentiation of kidney cells and perturbation may lead to tumor onset. Wilms tumors (WT), or nephroblastoma, originates from the metanephric blastema cells, which were unable to complete the differentiation, resulting in a triphasic tumor composed by distinct cell types, blastemal, epithelial and stromal, where the former harbor molecular characteristics similar to cells of the earliest stages of kidney development. In this study, to identify key early events involved in the disruption of correct cell signaling during kidney differentiation that can drive WT, we developed a cDNA platform containing genes from signaling transduction pathways. By using bioinformatics tools, we defined high conserved regions between human and mouse orthologous genes and assessed the gene modulation in blastemal cells captured from WT and differentiated kidneys (DK) in human and from four temporal stages of kidney differentiation in mouse. We found 18 genes, whose transcriptional level of the earliest phases of blastemal cell differentiation was recapitulated in WT. PAX2, FZD10, SHPK, WNT5B, FZD2, CRABP2, FRZB, GRK7, TESK1, HDGF and IGF2 were up- and MAPK9, PIK3CA, FRAT2, ITPR3, CDH6, HIPK1 and TIMP3 were down-regulated in WT respectively. Hierarchical clustering based on the expression of this gene set grouped DK from both species, human and mouse, and discriminated them from fetal kidney and WT in human, and the earliest kidney stages in mouse, validating the model proposed by this study and reveling a transduction signaling signature of WT. High robustness of this data was detected since expression level was tested by quantitative RT-PCR in the initial and independent sample set, with 75 and 56% of agreement. Agreement of 62% was also observed in protein level assessing blastemal component of an independent group of 137 WT. Protein expression of CRABP2, IGF2, GRK7, TESK1, HDGF, WNT5B, FZD2 and TIMP3 was also characterized in human fetal kidneys revealing interesting aspects of kidney differentiation. This study identified key genes modulated during kidney differentiation which may play a determinant role for WT onset. As far as we know, FZD10, FRZB, HDGF, MAPK9, FRAT2, SHPK, WNT5B, GRK7, TESK1, HDGF, PIK3CA, ITPR3, HIPK1 and TIMP3 were not previously associated to WT. The strong connection between nephrogenesis and WT highlights the importance of a detailed characterization of modulated genes belonged to signal transduction. Identification of gene expression involved in kidney differentiation arresting might be imperative to point out early alterations that drive WT onset and consequently are potential candidate as molecular marker. Here, we characterized expression pattern of genes belonged to transduction signaling pathways in blastemal cells during kidney development stages in mouse and WT in human. For that, we established a model of inter- and intra-specific hybridization to assess gene expression modulation to identify key events that drive WT onset. Dye-swap was performed for each sample as control for dye bias and used as replicate. In total, 24 frozen favorable histology WT and differentiated kidney samples were laser microdisscted for blastemal cells. Three kidneys from CD1 lineage from Mus Musculus were isolated from embryos (E) and postnatal animals (P) at ages E15.5, E17.5, P1.5 and P7.5 for laser microdissection of the blastemal cells in all stages of kidney differentiation. NA were purified with PicoPure™ RNA Isolation kit (Arcturus Engineering #KT0204) and treated with RNase-Free DNase Set (Qiagen #79254; Qiagen-Germantown MD USA).

Project description:To obtain insight into the molecular basis of ductal carcinoma in situ (DCIS) and its progression by infiltrating surrounding tissue, we have performed cellular-based gene expression analysis of pure DCIS and DCIS with co-existing invasive ductal carcinoma (IDC) and compared the histological and molecular aspects between these morphologically identical lesions seeking to find key genes involved in DCIS progression. For that, 30 samples were evaluated, 4 non-neoplastic (N), 5 pure DCIS, 11 DCIS with co-existing IDC (DCIS-IDC) and 10 IDC. All samples were laser capture microdissected and RNAs were amplified using T7-based methodology. Microarray technology was performed using a customized cDNA platform containing 4,608 human genes. To classify the 4 sample groups according to molecular similarity we performed comparisons of their general expression pattern using ANOVA test (pFDR<0,01) followed by Tukey´s test (fold>l2l). Among the 4 sample groups, as expected, non-neoplastic cells reported the most distinct gene expression pattern. However, among the 3 groups of neoplastic cells, in contrast to morphological aspects, pure DCIS had the most distinct expression profile when compared to the other lesions: DCIS-IDC and IDC. Additionally, by comparison among pure DCIS, DCIS-IDC and N, we identified 147 genes potentially involved in DCIS progression. Unsupervised hierarchical cluster based on expression profile of this gene-set could discriminate DCIS-IDC from 60% of pure DCIS samples. Keywords: disease state analysis Fresh-frozen human breast samples were retrieved from the Tumor Tissue Biobank of the Medical and Research Center - Hospital A. C. Camargo, São Paulo. This research was approved by Ethic Committee of the Medical and Research Center - Hospital A. C. Camargo under number 587/04 and a written informed consent signed by all participants. All patients had presented at least 5 years of follow-up. Thirty samples were evaluated, 4 non-neoplastic breast samples (MN), 5 pure DCIS (stage 0; DCIS-0), 11 DCIS with co-existing IDC (DCIS) and 10 IDC. The non-neoplastic samples were obtained from perilesional mammary specimens from patients submitted to resection of benign lesions. All cells types were laser capture microdissected using PixCell II LCM system (Arcturus Engineering, Mountain View, CA). The total RNA was extracted by using the PicoPureä RNA Isolation kit (Arcturus Engineering # KT0204). A two-round linear amplification procedure based on T7-driven amplification was carried out. Amplified RNA was then used in a transcriptase reverse reaction into cDNA in the presence of Cy3- or Cy5-labeled dCTP. HB4a normal luminal epithelial mammary cell line (O’Hare et al 1991) was amplified following the same protocol and used as reference for microarray hybridizations. Dye swap was performed for each sample analyzed and used as replicate samples. For the raw_data (lowess_data), see Web Link below.

Project description:Unravel the mechanisms underlying brain aging and Alzheimer´s disease (AD) has been difficult because of complexity of the networks that drive these aging-related changes. Analysis of the gene expression in the brain is a valuable tool to study the function of the brain under normal and pathological conditions. Gene microarray technology allows massively parallel analysis of most genes expressed in a tissue, and therefore is an important research tool that potentially can provide the investigative power needed to address the complexity of brain aging and neurodegenerative processes. One of the reasons that account for the resistance of AD pathogenesis to analysis is that clinically normal subjects may exhibit considerable AD pathology, blurring criteria for distinguishing subjects with normal aging or AD. Here, we analyzed hippocampal and cortex frontal gene expression from 32 subjects separated in individuals presenting, 1) both pathologic and clinical AD (definitive AD); 2) AD pathology and normal clinic (pathologic AD); 3) cognitive impairment, without AD pathology (others dementias); and 4) no cognitive impairment, without AD pathology (normal individuals). Our results show that based on gene expression profile these individuals we could verify similarity between the definitive AD group and the group that only had AD-type pathology (pathologic AD). Specimens of hippocampus and cortex frontal used in this study were obtained at autopsy from 32 subjects. Neuropathology, specifically NFT and NP, was assessed in accordance to the Braak staging and CERAD scores, respectively. Based on pathologic and clinic criteria, subjects were categorized into four groups: 1) nine subjects with AD neuropathologic (Braak = IV / V / VI and CERAD ≠ 0) presenting cognitive impairment (CDR ≥ 1), termed “definitive AD” (dAD); 2) five subjects with AD neuropathologic (Braak = IV / V / VI and CERAD ≠ 0) and without cognitive impairment (CDR = 0), termed “pathologic AD” (pAD); 3) nine subjects without AD neuropathologic (Braak = 0 / I / II and CERAD ≠ C) presenting cognitive impairment (CDR ≥ 1), termed “other dementias” (OD); 4) - nine subjects without AD neuropathologic (Braak = 0 / I / II and CERAD ≠ C) and without cognitive impairment (CDR = 0), termed “normal” (N). RNA isolation and Amplification. From total RNA, a two-round linear amplification procedure (T7-based protocol) was carried out for all samples and for a pool of RNAs obtained from 15 distinct human cell lines used as reference. Labeled cDNA was generated in a reverse transcriptase reaction using amplified RNA (aRNA). Equal amounts of test and reference cDNA reverse color Cy-labeled aRNA targets were competitively hybridized against the cDNA probes in a customized cDNA platform with 4,608 ORESTES representing human genes. Dye-swap was performed for each sample as control for dye bias and used as replicate.

Project description:In normal skin, interactions between melanocytes and keratinocytes, and between melanocytes and the basal membrane are functionally relevant for maintenance of homeostasis and epidermal melanin unit whose deregulation may trigger a continuous proliferation of the melanocytes. In order to further elucidate the molecular events related to cell-cell and cell-ECM interactions in relation to the biology of melanomas, we analyzed the expression profile of 64 human melanomas, 21 primary samples and 43 independent metastasis. Tissue samples melanomas were obtained at Hospital A. C. Camargo. All patients signed an informed consent and the study was approved by our internal ethics committee. Tissue samples obtained by surgery were snap-frozen in liquid nitrogen and biopsy samples were collected in RNAlaterTM (Ambion, Austin, TX). Non-tumor areas were removed by handy dissection and only samples represented by at least 70% of tumors cells within their natural stroma were selected for RNA extraction. Total RNA was extracted using RNeasy Midi-Kit® and amplified by a T7-based protocol. An indirect cDNA-labeling was performed as described by Cox et al. (2004). We used dual-label system in which two RNA samples (test and reference) were separately reverse-transcribed, labeled, mixed, and hybridized together to each array. Each sample was hybridized in duplicate, with dye swap between sample and reference. As reference RNA we used a pool of equal amounts of total RNA from 5 human cutaneous melanoma cell lines (SK-MEL 05; 17; 28; 37 and 188, kindly provided by Dr. Alan Houghton, MSKCC, New York, NY, USA). We used glass arrays containing 4.800 cDNA sequences (Brentani et al. 2003) were prepared in our lab with the aid of the Flexys robot17 (Genomic Solutions, Cambridgeshire, United Kingdom). The list of immobilized genes can be obtained at GEO, accession number GPL1930. Pre-hybridization, hybridization and washing were performed as previously described (Gomes et al. 2003; 2005), and slides were scanned on a laser scanner (ScanArray Express, PerkinElmer Life Sciences, Boston, MA). Data were extracted with ScanArray Express software (PerkinElmer Life Sciences, Boston, MA) using the histogram method.

Dataset Information

Gene expression analysis of relapse and no-relapse Wilms tumor

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure