Changes in gene expression profiles concerning renal ageing

Age-dependent changes in genome wide gene expression profiles in human kidney tissue remain highly elusive. In addition, neither the compartment in which these changes occur nor the role of specific pathways is definitely known. The primary aim of this project is to identify gene expression patterns associated with renal aging on an individual, high resolution level. For this purpose wedge biopsies of deceased donor kidneys aged 20-80 years will be prospectively collected. After RNA isolation and amplification human oligonucleotide microarrays will be used to determine the transcript levels. The experimentally obtained microarray datasets will be subjected to various statistical methods and bioinformatic analysis in order to find statistically significant differentially regulated genes between different age classes. We will annotate the genes to functional classes on the basis of publicly available information as stored in large biological databases. Then the genes will be subjected to coregulation, pathway and promotor analysis and expression profiles will be linked to clinical data. In a sub-study we will dissect glomeruli from the tubulointerstitium and perform separate microarray studies in order to identify the compartment in which the changes in gene expression found on a whole tissue level are predominantly located. Results of microarray experiments will be validated by real time PCR on selected candidate genes. The possible consequences and clinical applications of our study results are diverse. Clearly they will provide a deep insight into the physiology of ageing of the kidney. Even though an "anti-ageing" approach is not our primary focus we cannot exclude the possibility that pathway analysis could provide ideas for a therapeutic intervention in this regard. A correlation between late allograft function and selected gene expression levels should provide the basis for a "micro" chip or a PCR based kit, which could allow the assessment of the biological and not only the chronological donor age of organs (maybe not only kidney but also heart or lung) before transplantation quickly. In a second step the data generated could lead to the discovery of new biomarkers. Bioinformatics analysis will provide a list of potentially secreted proteins encoded by our marker genes. If their presence could be confirmed in serum they will allow the establishment of an easy blood based screening, which again could be used in transplant medicine but also in the general population to stratify the renal risk associated with exposition to nephrotoxic agents like contrast media or other drugs. If our results obtained in the kidney are reproduced in other organs such biomarkers will be of interest to almost any area of medicine.

The primary goal of this project (determined by microarray experiments as "genome wide gene expression" approach) was to detect changes in gene expression in the context of aging of the kidney. For this purpose biopsies of deceased donor kidneys aged 20-80 years (model of "healthy kidney") were biopsied prior to implantation. In cooperation with Univ. Prof. Dr. R. Margreiter (Department of Visceral and Transplantation Surgery, Medical University of Innsbruck), Prim Univ. Prof. Dr. R. Oberbauer (KH Elisabethinen, Linz), Univ. Prof. Dr. A. Wiecek (Department of Nephrology, Endocrinology and Metabolic Diseases, Medical University of Silesia, Katowice, Poland) and Univ. Prof. Dr. G. Steurer (University Department of General, Visceral and Transplantation Surgery, University Hospital of Tübingen, Germany) a total of 200 samples could be collected, 91 of which were included in the study. For isolation of RNA various protocols were tested. As a measure of RNA integrity the recommended Agilent RIN value (RNA Integrity Number)(Schroeder, Mueller et al. 2006) was used. It was shown, that for comparability of microarrays rather the similarity of RNAs than the absolute RIN value is critical. It was decided that samples with a RIN value of 5-8 for this project (scale 1-10, 1 = worst, 10 = best) can be used. To validate microarray data the expression of certain genes was determined using real time PCR. Therefore different qPCR protocols were tested and optimized, resulting in the establishment of a preamplification kit for TaqMan primers. This made it possible to amplify the rather limited amount of specific cDNA derived from biopsies by a factor of 100-400 with 90 % linearity and reproducibility of the used TaqMan probes and primers. The expression of selected genes showed a good correlation between array and qPCR (83% agreement). Furthermore it was tested whether the material was to be collected by means of wedge- or needle biopsy. Due to a significant difference in gene expression between these two methods (probably because a wedge biopsy contains predominantly renal cortical tissue, but a needle biopsy cortex and medulla tissue) an SOP was created and for the study only material obtained by wedge biopsy was used. Despite these efforts a primary analysis of the array data showed that expression profiles of samples from Poland differed significantly from that of the other centers. Thus, these samples were not analyzed any further and a total of 77 biopsies were included in the study.In collaboration with emergentec biodevelopment GmbH in Vienna a centralized online array database was established to record microarray data. This database not only allows storage of huge amounts of data, that naturally occur with array projects, but also offers the advantage of decentralized use of raw data for filtering and analysis of gene expression.Project 1: Gene expression changes associated with renal ageing. The aim of the first project was to find the association of altered gene expression in donor kidneys using microarray technology and correlate it with the chronological age of donors. Data of 77 microarrays were preprocessed, normalized and divided into 5 age groups (group 1: 18-29, group 2: 30-39, group 3: 40-49, group 4: 50-59, Group 5:> 60). After quantile normalization (Bolstad, Irizarry et al. 2003 3957) of arraydata, differences in gene expression between the five age groups were analyzed by ANOVA, SAM (Significance Analysis of Microarrays) (Tusher, Tibshirani et al. 2001 3896), and linear correlation. For SAM analysis +/- ? units were set to 0.012, resulting in 124 significantly differentially expressed features with a false discovery rate < 5%. ANOVA was computed with an overall threshold p-value of 0.01, resulting in 103 significantly differentially expressed features. In addition, each of the