One biopsy core was collected into a vial containing 1ml of RNALater® (Ambion, Catalog Number AM7023), stored at -20°C for 24 hours, and then transferred to -70°C until RNA isolation. Renal tissue content in all cores was verified using a dissecting microscope and any non-renal tissue was removed. Total RNA was isolated using the QIAGEN Rneasy Mini Kit (Catalog Number 74104, Qiagen, USA) and was quantified using the NanoDrop 2000 (Thermo Scientific, Wilmington, Delaware). To further assess RNA quality, samples were assessed by the Agilent 2100 Bioanalyzer (Agilent Technology, Santa Clara, CA,) and samples with an RNA Integrity Number score cut off >7 were deemed suitable for Gene Chip Microarray Hybridization. The complete protocol for converting total RNA into biotin-labeled RNA by Affymetrix Human Gene 1.0 ST Array hybridization containing 28,869 gene probe sets can be found in the manufacturer¹s protocol for the Ambion Whole Transcript Expression Kit (Catalog Number 4425209, Ambion, Austin, TX) and Affymetix Gene chip Whole Transcript Terminal Labeling Kit (Catalog Number 900671, Affymetrix, Santa Clara, CA).
Extracted molecule
total RNA
Extraction protocol
Total RNA was isolated using the QIAGEN RNeasy Mini Kit (Catalog Number 74104, Qiagen, USA) and was quantified using the NanoDrop 2000 (Thermo Scientific, Wilmington, Delaware). To further assess RNA quality, samples were assessed by the Agilent 2100 Bioanalyzer (Agilent Technology, Santa Clara, CA,) and samples with an RNA Integrity Number score cut off >7 were deemed suitable for Gene Chip Microarray Hybridization.
Label
biotin
Label protocol
Follow protocols of Ambion Whole Transcript Expression Kit (Catalog Number 4425209, Ambion, Austin, TX) and Affymetix Genechip Whole Transcript Terminal Labeling Kit (Catalog Number 900671, Affymetrix, Santa Clara, CA)
Hybridization protocol
Total RNA samples hybridized to the Affymetrix Hugene 1.0 st chip according to the protocols provided by the manufacturer.
The gene expression profiles of the biopsy specimens were compared according to the following groups: Group 1: Normal transplant kidney biopsies (n=12) Group 2: Non-specific IFTA (N=17) Group 3: TGP DSA-/C4d- (g+ptc > 1) (N=8) Group 4: TGP DSA+/C4d- (N=9) Group 5: CAMR/TGP (C4d+/DSA+) (N=14) Probe intensities were extracted from Affymetrix CEL files, normalized and summarized at the transcript cluster level using the Robust Multiarray Average method from the R/Bioconductor (http://bioconductor.org) oligo package. Array quality control was performed to remove arrays that significantly differed in initial overall probe intensity. Expression values were analyzed on the log base 2 scale. Significance testing was performed by fitting gene-wise linear models using the limma Bioconductor package, and genes were ranked for differential expression using Benjamini & Hochberg adjusted p-values and log-posterior odds of differential expression (log probability of differential expression versus probability of no differential expression). In order to obtain the Benjamini and Hochberg adjusted p-value, the false discovery rate (FDR) was set at <0.01. Gene set analysis was performed using the limma romer (Rotation Gene Set Enrichment Analysis) method. This analysis generates a p-value for enrichment of three categories of gene expression for each gene set: probability of enrichment of up-regulated and down-regulated genes, and enrichment of genes with expression level changes (P-value was set at < 0.05). Gene Set libraries were taken from the Molecular Signatures Database (MSigDB) and new libraries were generated by mapping HUGO gene identifiers to the University of Alberta, Edmonton, Pathogenesis Based Transcripts (PBT; http://transplants.med.ualberta.ca/Nephlab/data/gene_lists.html) gene sets. The following core PBT were analyzed: * KT: Kidney specific transcripts (n=63) * IRIT: Injury and repair induced transcripts (n=228) * GRIT: Gamma-IFN and rejection induced transcripts (n=50) * QCAT: Quantitative cytotoxic T cell--associated transcripts (n=25) * CMAT: Quantitative constitutive macrophage-associated transcripts (n=72) * BAT: B cell associated transcripts (n=50) * NKAT: Natural killer cell-associated transcripts (n=244) * ENDAT: Endothelial cell associated transcripts (n=114) * DSAST: Donor Specific Antibody Selective Transcripts (n= 25) Analysis for changes in gene expression of Gene Ontology (GO) categories was performed using the GOstats Bioconductor package to calculate Hypergeomtric p-values for over or under-representation of each GO term.
