Peripheral blood from SLE patients was obtained from the Department of Rheumatology and Clinical Immunology, Charité University Medicine Berlin, Humboldt University of Berlin; peripheral blood from healthy donors vaccinated against yellow fever was obtained from the Berlin-Brandenburg Center of Regenerative Therapies (BCRT), Charité University Medicine Berlin, Humboldt University of Berlin.
Treatment protocol
A total of 50 ml peripheral blood was collected in Vacutainer heparin tubes (Becton-Dickinson, Heidelberg, Germany), and erythrocytes were lysed in EL buffer (Qiagen, Hilden, Germany). Subsequently, granulocytes were depleted using CD15-conjugated microbeads (MACS, Miltenyi Biotec, Bergisch Gladbach, Germany). The CD15-depleted fraction was stained with a CD14-fluorescein isothiocyanate (FITC) antibody (Becton-Dickinson), a CD16-APC-Cy7 antibody (Becton-Dickinson), a CD3-Vioblue antibody (Becton-Dickinson) and a CD4-FITC antibody (Becton-Dickinson). Using a FACSAria cell sorter (Becton-Dickinson), CD4pos T cells, CD16neg monocytes and CD16pos monocytes were isolated with purities and viabilities of >97%. After sorting, the cells were immediately lysed with RLT buffer (Qiagen) and frozen at -70 °C.
Extracted molecule
total RNA
Extraction protocol
Total RNA was extracted using the RNeasy Mini Kit (Qiagen) according to the manufacturer’s instruction, including the recommended DNase digestion. The purity and integrity of RNA were assessed for each sample using an Agilent 2100 Bioanalyzer with the RNA 6000 Nano LabChip and amount was checked with a NanoDrop ND-1000 spectrophotometer. Contaminating genomic DNA was removed by an on-column DNA digestion step (Qiagen).
Label
Biotin
Label protocol
Labeling and hybridization of total RNA was performed using the GeneChip 3’ IVT Expression Kit from Affymetrix according to the manufacturers instruction. Briefly, after total RNA extraction, 100 ng of total RNA from each cell sample was reverse transcribed, cDNA was extracted and used as template for in vitro-transcription to generate biotinylated cRNA. cRNA was fragmented and 15 µg cRNA hybridized to each of the GeneChip arrays. The HG-U133 Plus 2.0 GeneChip arrays were loaded with the hybridization cocktail, hybridized at 45 °C rotating for 16 h in a hybridization oven 640 (Affymetrix), washed and stained with streptavidin-phycoerythrin using the Affymetrix GeneChip Fluidics Workstation 450.
Hybridization protocol
Labeling and hybridization of total RNA was performed using the GeneChip 3’ IVT Expression Kit from Affymetrix according to the manufacturers instruction. Briefly, after total RNA extraction, 100 ng of total RNA from each cell sample was reverse transcribed, cDNA was extracted and used as template for in vitro-transcription to generate biotinylated cRNA. cRNA was fragmented and 15 µg cRNA hybridized to each of the GeneChip arrays. The HG-U133 Plus 2.0 GeneChip arrays were loaded with the hybridization cocktail, hybridized at 45 °C rotating for 16 h in a hybridization oven 640 (Affymetrix), washed and stained with streptavidin-phycoerythrin using the Affymetrix GeneChip Fluidics Workstation 450.
Scan protocol
Arrays were scanned with an Affymetrix GeneChip Scanner 3000, using the GCOS software version 1.4 from Affymetrix.
Description
Purified CD4 T lymphocytes isolated from active SLE-patient 4
Data processing
Data were analyzed according to High Performance Chip Data Analysis (HPCDA, unpublished - Joachim R. Grün) with the Bioretis database (http://www.bioretis-analysis.de/) using the default filter parameters for decreased and increased gene lists (description of database c.f. open access paper Menssen et al., 2009; PubMedID: 19265543). Chip data included in the Bioretis database were analyzed using the GeneChip Operating Software (GCOS, Affymetrix), version 1.4. Microarrays were globally normalized and scaled to a trimmed mean expression value of 150. Quality checks were performed according to the manufacturer's recommendations. All chips of one SLE group were compared to each of any other ND group, and the following parameters of absolute and comparative analysis were included in the Bioretis database: expression heights (Signals) and mean, median and standard deviation of Signals of both groups, call for presence of transcripts (Absent, A; Marginal, M; Present, P), p-value for presence or absence of transcripts, log2 value of fold change (Signal Log Ratio, SLR) and the fold change as mean values, call for the significance of differentially expression (Change Call: Increase, I; Marginal I, MI; Decrease, D; Marginal D, MD; No Change, NC), and the p value for that call. Additionally – not calculated with GCOS – t-tests of log Signals and SLRs were included in the database. For each present transcript the significance of differential expression between the groups of arrays was either calculated using strict Bonferroni corrected Welch t-tests between SLR values of Experiment group vs. Baseline group and SLR values within both groups (the latter always giving a mean SLR value of zero) or more than 50% of non-parametrically calculated Change calls (Mann-Whitney U test, GCOS) have to be in the same direction. Compared were Grp1 vs. Grp4 (24 comparisons, that means at least 13 Change Calls have to be in increased or decreased direction and/or p-value < 1.3856E-08), Grp1 vs. Grp5 (24 comparisons, p-value < 1.386E-08), Grp2 vs. Grp6 (16 comparisons, p-value < 2.286E-08), Grp2 vs. Grp7 (16 comparisons, p-value < 2.286E-08), Grp3 vs. Grp8 (12 comparisons, p-value < 3.048E-08) and Grp3 vs. Grp9 (12 comparisons, p-value < 3.048E-08). Significantly differentially expressed genes, were filtered using both default parameter sets of filter criteria; these are a combination of four different queries. Filter criteria were developed with various data sets of GeneChips and validated with the Affymetrix Latin Square dataset as shown in Menssen et al., 2009. You can find these validation experiments in BioRetis without registering, using the public content and public login. Click on view single results, select any existing Analysis beginning with first 3 letters "SGU" and click on "Next". Click on "Chose an existing parameterset" and select "JRG_Increase" for increased or "JRG_Decrease" for decreased probesets and click on "Fill". At the bottom of the site check the box named "* use Bonferroni correction" and click on "Search only" to get the list of significantly changed increased or decreased probesets, respectively. The Affymetrix Latin Square dataset, consisting of 42 chips in 14 experiments with three replicates each were analyzed in BioRetis with all possible 3 vs. 3 chip comparisons (one direction).
