Experiment description
1. Experimental design
1) Authors, laboratory, and contact: Xue-Lin Cui1, Patricia Soteropoulos2, Peter Tolias2, Ronaldo P. Ferraris1: 1Dept. of Pharmacology and Physiology, UMDNJ-New Jersey Medical School, Newark, NJ 07103-2714, 2Center for Applied Genomics, PHRI, 225 Warren Street Newark, NJ 07103-3506. Contact to Dr. Ronaldo P. Ferraris.
2) Type of the experiment: treated vs. untreated comparison.
3) Experimental factors: Twenty to twenty-two old SD rats were use in the experiment. The intestines of paired pups were perfused with 100 mM fructose (HF) and 100 mM glucose (HG) for 4-h or 20-min, respectively.
4) Totally 14 experiments were done.
5) A common reference was not used for all the hybridizations.
6) Quality control steps ( 14) Miller RA, Galecki A, Shmookler-Reis RJ. Interpretation, design, and analysis of gene array expression experiments. J Gerontol A Biol Sci Med Sci. 2001 Feb;56(2):B52-7. (15) Tseng GC, Oh MK, Rohlin L, Liao JC, Wong WH. Issues in cDNA microarray analysis: quality filtering, channel normalization, models of variations and assessment of gene effects. Nucleic Acids Res. 2001 Jun 15;29(12):2549-57. ( 16) Yang YH, Speed T. Design issues for cDNA microarray experiments. Nat Rev Genet. 2002 Aug;3(8):579-88. (17) Yang YH, Dudoit S, Luu P, Lin DM, Peng V, Ngai J, Speed TP. Normalization for cDNA microarray data: a robust composite method addressing single and multiple slide systematic variation. Nucleic Acids Res. 2002 Feb 15;30(4):e15.):
a) In the 4 h perfusion experiment, the intestines of the paired pups were perfused with 100 mM fructose (HF) and 100 mM glucose (HG) for 4-h, respectively. Totally five replicates were conducted. The RNA samples were extracted from HF- and HG-perfused intestines (n=5 pairs) then labelled with Cy 5 (HF1, HF2, HF3, HF4, and HF5) and Cy 3 (HG1, HG2, HG3, HG4, and HG5), respectively. At the same time, one dye swap experiment, in which the samples extracted from HF- and HG-perfused intestines were labelled with Cy 3 (HF1) and Cy 5 (HG1), respectively, was done and used in the data analysis.
b) In the 20 min perfusion experiment, the intestines of of the paired pups were perfused with 100 mM fructose (HF) and 100 mM glucose (HG) for 20-min, respectively. Totally four replicates were conducted. In two of the four-replicate experiments, the samples extracted from HF- and HG-perfused intestines were labelled with Cy 5 (HF1a and HF3a) and Cy 3 (HG1a and HG3a), respectively. In the other two of the four-replicate experiments, the dye was swapped, that is, the samples extracted from HF- and HG-perfused intestines were labelled with Cy 3 (HF2a and HF4a) and Cy 5 (HG2a and HG4a), respectively.
c) Three self-self experiments using different samples (HF1, HG1, and HF2) and one capture primer alone experiment were conducted to eliminate the false positive genes and aid to analyze the data.
7) The goal of the experiment: Fructose induced the expression of GLUT5 mRNA in the small intestinal epithelial cells. But it is not clear about the mechanisms involved in the phenomenal. In this experiment, we hypothesize: the expression of some genes must be changed after fructose perfusion for 4-h, these genes may regulate GLUT5 gene transcription; the expression of some genes must be changed even after 20-min fructose perfusion, these genes may trigger the GLUT5 gene expression as early response gene. Our goal in this study is to find out these potencial gene.
8) Links (URL), citations
2. Samples used, extract preparation, labeling and hybridization
1) Bio-source properties
a) Sprague Dawley rats (NCBI taxonomy);
b) The rat was anesthetized using urathene. The rat body was kept at 37 oC under light.
c) Descriptors relevant to the particular sample: the jejunum of the 20-d old rat pups (unknown sex) was perused.
2) Biomaterial manipulations: the jejunum was in vivo perfused with 100 mM HF or HG for 4-h or 20-min. At the end of each experiment, the jejunum was immediately cut out and snap frozen in liqued nitrogen, and then stored the tissue in -80 oC freezer.
3) RNA isolation: A RNeasy Midi Kit (QIAGEN Inc., Valencia, CA) was used to extract total RNA from the jejunum. The brief procesure is described in the Total RNA Extraction for RT-PCR and Microarray above.
4) Sample assignment to slides for labeling and hybridization:
S1: HF1(Cy 5) vs HG1 (Cy 3); S2: HF2 (Cy 5) vs HG2 (Cy 3); S3: HF3 (Cy 5) vs HG3 (Cy 3); S4: HF4 (Cy 5) vs HG4 (Cy 3); S5: HF5 (Cy 5) vs HG5 (Cy 3); S6: HF2 (Cy 3) vs HG2 (Cy 5); S7: HF1a (Cy 5) vs HG1a (Cy 3); S8: HF3a (Cy 5) vs HG3a (Cy 3); S9: HF2a (Cy 3) vs HG2a (Cy 5); S10: HF4a (Cy 3) vs HG4a (Cy 5); S11: HG1 (Cy 5) vs HG1 (Cy 3); S12: HG3 (Cy 5) vs HG3 (Cy 3); S13: HF3 (Cy 5) vs HF3 (Cy 3); S14: Cy 3 vs Cy 5 capture primer.
5) Labeling and hybridization procedures and parameters:
The 3DNA Submicro Oligo Expression Array Detection Kit (Genisphere Inc., Hatfield, PA) was employed to synthesize and label the complimentary DNA (cDNA) with fluorescent dye: Cy3 or Cy5. The reason why this indirect labeling method was chosen is based on a previous study which demonstrated that the method was more sensitive and consistent than direct labeling method or other indirect techniques like aminoallyl labeling method and MICROMAX tyramide signal amplification labeling method. (18)Yu J, Othman MI, Farjo R, et al: Molecular Vision 2002; 8:130-137). In the primary experiments, three, four, and five g of total RNA were tested for labeling and hybridization conditions. In the official experiments, four g of total RNA isolated from each sample were used for labeling and hybridization, because the signal to noise ratio is high at this concentration. In addition, at this concentration, all most of the low expressed genes are detectable and the highly expressed genes are not fully saturated. The procedure was conducted following the manufacturer's protocol. Briefly, the first strand cDNA was synthesized using a specific poly T Cy3 RT primer or a specific poly T Cy5 RT primer in different tubes for paired samples. Combine the Cy3 and Cy5 reactions in one tube and precipitate the sample in ethanol. After briefly drying the pellet, resuspend the pellet with 40 l of hybridization buffer. Subsequently, apply the hybridization mixture onto the array slide and hybridize overnight at 55 oC in a humidified chamber. Following the hybridization, the slides were washed by placing them into 50 ml conical tubes containing 2 x SSC, 0.2% SDS for 15 min, 2 x SSC for 10 min, 0.2 x SSC for 10 min, and 95% ethanol for 1 min all at room temperature, except for the first wash, in which the temperature was 42 oC. After drying the slides, the Cy3 and Cy5 capture (connected to dye molecues) reagents mixed in hybridization buffer was pipeted onto the slides, and then hybridize at 65 oC for 3 h following slides wash as described above.
3. Measurement data and spesifications of data processing
1) Raw data description
Slides were scanned for Cy3 and Cy5 fluorescence using an Axon GenePix4000 microarray reader (GenePix 4000A, Axon instruments, Foster City, CA) and quantitated using GenePix software. Scan parameters: Scan power: Cy 3 : Cy 5 = 100:100; Laser power: Cy 3 : Cy 5 = 1.15:1.76; Spatial resolution: ???; Pixel size: 10 or 100 ???; Pixel number: F pixels: 50-100; B pixels: 400-500 ???; PMT voltage: 600-750 for Cy 5 channel, 650-850 for Cy 3 channel.
2) Image analysis and quantitation
For the image analysis, the density of the spots was detected using Gene Spring (Silicon Genetics, Redwood City, CA) software (GeniPix 4.0) (19) Axon Instruments Inc. (1999) GenePix 4000A User's Guide. Axon Instruments, Union City, CA). It is commercially available. Description of the algorithm and all the parameters used: R = red channel signal = F635; G = green channel signal = B 532; M = log2(R/G); A = 1/2xlog2(RxG); User-difined parameter f = 40%. For the complete image analysis out put of each image, see the NCBI's GEO database (www.ncbi.nlm.nih.gov/geo).
3) Normalized and summarized data-gene expression data matrix
For the normalization, the following two steps were taken.
a) When measuring the density of the spots using Gene Spring software, the ratio of Cy 3 and Cy 5 dyes was adjusted as 1 as a globle normalization for the whole slide.
b) For the gene expression analysis, the results (midian of the density of the pixel in one spot was used) were normalized by a LOWESS (locally weighed scatter smoother) software provided by the Center for Applied Genomics, so that the results can be compared among the experiments performed at different times. (20)Yang YH, Dudoit S, Luu P, Lin DM, Peng V, Ngai J, Speed TP. Normalization for cDNA microarray data: a robust composite method addressing single and multiple slide systematic variation. Nucleic Acids Res. 2002 Feb 15;30(4):e15.).
c) The following rules were made to decide the genes that were significantly changed: 1) Eliminate the genes with high density (over than 2X of background) in the 3DNA capture probe experiments; 2) Eliminate the genes with density lower than 2X of background ( 21) Lee HM, Greeley GH Jr, Englander EW. Age-associated changes in gene expression patterns in the duodenum and colon of rats. Mech Ageing Dev. 2001 Apr 15;122(4):355-71.); 3) Eliminate the genes with change over than 1.5X in dye flip-over experiment; 4) Eliminate the genes with opposite results in one experiments between dye flip-over comparisons; 5) Only the genes significantly changed over 3 in 5 comparisons in 4 h perfusion experiment and over 2 in 4 comparisons in 20 min perfusion experiment were considered as reliable and biologically important.
d) The final results discribed in this paper were presented as Means + SE and the frequency of gene changed in the HF-perfused intestine. About the image, raw, and normalized data, and other detailed parameters, please see the gene expression data tables derived from the experiments in the NCBI's GEO database.
The design for the dye labeling in this experiments is as the follows: 1) 5 comparisons were conducted in the 4 h-perfused HG vs HF experiment, in which in 3 comparisons, the cDNA samples synthesized from HG- and HF-perfused intestines were labeled with Cy3 and Cy5, respectively; and in 2 comparisons, the cDNA samples synthesized from HG- and HF-perfused intestines were labeled with Cy5 and Cy3, respectively. 2) 4 comparisons were conducted in the 20 min-perfused HG vs HF experiment, in which in 2 comparisons, the cDNA samples synthesized from HG- and HF-perfused intestines were labeled with Cy3 and Cy5, respectively; and in 2 comparisons, the cDNA samples synthesized from HG- and HF-perfused intestines were labeled with Cy5 and Cy3, respectively. 3) 3 dye flip-over experiments using 1 RNA sample extracted from HF-perfused intestine and 2 samples extracted from HG-perfused intestines were performed to eliminate the false positive results induced by the dye labeling technique or other unsolved problems. 4) 2 experiments only applying 3DNA fluorescent probe, which is connected to complementary capture sequence, were performed to eliminate the false positive genes that were low expressed in the tissue. The procedures for synthesizing the first strand cDNA, labeling the cDNA, hybridizaion, and detecting the signals were performed following the manufacturer's protocols.
Keywords: other
Less...