|
| Status |
Public on Dec 31, 2011 |
| Title |
P304M |
| Sample type |
RNA |
| |
|
| Source name |
Biopsy (liver tissue)
|
| Organism |
Homo sapiens |
| Characteristics |
tissue: pancreatic metastatic tumor (liver)
|
| Biomaterial provider |
Hospital das clínicas – Universidade de São Paulo
|
| Extracted molecule |
total RNA |
| Extraction protocol |
Total RNA was extracted from pancreatic tissue samples (50-100mg tissue) using Trizol (Invitrogen) according to manufacturer’s recommendations. Aliquots of RNA were cleaned up in RNeasy Mini Kit spin columns (QIAGEN) including DNAse I RNAse free (QIAGEN) digestion step for removal of genomic DNA. Total RNA integrity was measured by the relative abundance of 28S/18S ribosomal subunits and verified through micro fluid capillary electrophoresis (Agilent Bioanalyzer 2100).
|
| Label |
Cy5 (Amersham)
|
| Label protocol |
All the steps in this part were performed protected from light. Labeled targets were generated from 5µg of cRNA from 2º round of amplification modified with 5-(3-amino-allyl)-UTP. The sample were resuspended in 4.5µl of coupling buffer (0.1M carbonate buffer pH 8.5-9.0) combined with 4µl of mono-reactive Cy5-esters (Amersham) and 2.5µl of treated-DEPC water. The reaction was incubated at room temperature for 1 hour.
To stop coupling reaction 6µl of Hydroxylamine 4M was added and incubated at room temperature for 15 minutes. Labeled targets were purified using RNeasy Mini kit spin columns (QIAGEN) following manufacturer’s instructions. Incorporation of mono-reactive Cy5-esters was verified by NanoDrop spectrophotometer (Thermo Scientific) measuring OD 260 and 650nm. The total volume of labeled targets was concentrated to approximately 10µl using Microcon YM-30 (Millipore) following manufacturer’s instruction.
|
| |
|
| Hybridization protocol |
Labeled targets were resuspended in 1X buffer hybridization comprised by 100µl of Formamide 100%, 50µl of RPK032 4X Microarray Hybridization Buffer (Amershan Pharmacia Bioscience), 1µl of RefOligo (100pmol/µl), 5µg of competing sequences and treated-DEPC water in a volume to complete 200µl of total volume counting the labeled target volume. The hybridization mixture was denatured at 92ºC for 2minutes and cooled on ice for 2 minutes, and then it was centrifuged at 13000rpm for 5 minutes.
|
| Scan protocol |
Images were obtained using a laser Generation III Array Scanner and the Array Scanner control software (Molecular Dynamics). Slides were scanned with scanning resolution of 10µm pixel, excitation wavelength of 633nm (for cy5-labeled targets), emission filter 675nm and PMT voltage of 600 Volts.
|
| Description |
One microgram of total RNA was extracted from tissue samples using TRIzol reagent (Invitrogen) and the Rneasy Mini Kit (QIAGEN). RNA was reverse-transcribed into cDNA with a T7-(dT) oligomer to prime the first-strand synthesis. DNA polymerase and TS (template switch) primer were included in the synthesis of the second strand. After phase lock gel phenol/chloroform cDNA extraction and ethanol precipitation, double stranded cDNA was converted into cRNA. Second round of amplification were done with 1µg of cRNA. This cRNA was reverse transcribed using random hexamer oligo to primer the first strand synthesis. The following steps were the same as previously reported to obtain the purified cDNA double strand. To generate the 2º round of cRNA, we performed a in vitro transcription in the presence of 5-(3-amino-allyl)-UTP combined with a mix comprised by rNTPs (A, C, G, U). Five micrograms of cRNAs were labeled with Cy5-esther according to label protocol. Array platform was hybridized with the fragmented labeled cRNA for 16 h at 42 °C. Slides hybridized were processed using Generation III Array Scanner and the Array Scanner control software (Molecular Dynamics).
|
| Data processing |
Raw images were analyzed using ArrayVision software (Version 8.0, Imaging Research Inc.) and consisted in images in gray scale, corresponding to Cy5 channel. The procession analysis is required to extract the numerical data. It involves an array template (or grid) automatically aligned to locate the position of each spot in the array, and if required a subsequently manual adjustment to obtain the best possible alignment. After estimating spot location on the array, measurement the spot intensity is made as well as the background intensity based on the area outside of the spot.
To remove artifacts signal from our analyses, we employed the Artifact-Removed densities values (ARM) calculated for each spot by ArrayVision. The ARM is calculated through MTM Density (Median-based trimmed Mean Density Value), which represents the average of all pixels remaining in the spot, after removing pixels with density values that exceed four median absolute deviations (MADs) from the median. This procedure removes the influence of image artifacts (e.g. dust particles) on the density estimation. The ARM density for each spot was subtracted by the median background surrounding the spot (distance of 2 pixels with 3 widths of pixels).
Since each microarray platform contains two replicates of each spotted cDNA, a total of four replicates measurements were obtained for each spot in each patient. The raw data of each spot was compared to the median of all spots in the array platform to filter low values intensities. Only spots that presented value above or equal to the median were considered to further analysis. Replicas that presented median equal to 0 were discarded. To calculate the average of the spot among the patient sample replicas, first we normalized all the spots values considered in previous filtering using QUANTIL method to remove systematic biases, in order to compare across array. Thus we calculated the average value only for spots that presented at least 3 replicas measurements for 4 validated replicas or 2 replicas measurements for 3 or 2 validated replicas. To be considered as expressed transcript, each spot value had to be presented in 75% of samples in at least one histological group. As we were working with different batches of spot cDNA platform production, we performed a batch adjustment using the algorithm ComBat, in order to remove non biological bias.
|
| |
|
| Submission date |
Jun 21, 2011 |
| Last update date |
Dec 31, 2011 |
| Contact name |
Vinicius Maracaja-Coutinho |
| E-mail(s) |
[email protected]
|
| Organization name |
Universidade de Sao Paulo
|
| Department |
Departamento de Bioquimica, Instituto de Quimica
|
| Street address |
Av. Prof. Lineu Prestes 748, sala 1200
|
| City |
Sao Paulo |
| State/province |
Sao Paulo |
| ZIP/Postal code |
05508-000 |
| Country |
Brazil |
| |
|
| Platform ID |
GPL3985 |
| Series (1) |
| GSE30134 |
Long noncoding intronic RNAs are differentially expressed in primary and metastatic pancreatic cancer |
|
