1. Volume 65, Issue 3, Pages 904-917 (March 2004)
Gene expression fingerprints in human tubulointerstitial inflammation and fibrosis as prognostic markers of disease progression1 
Anna Henger, Matthias Kretzler, Peter Doran, Mahnaz Bonrouhi, Holger Schmid, Eva Kiss, Clemens D. Cohen, Stephen Madden, Stefan Porubsky, Elisabeth F. Gröne, Detlef Schlöndorff, Peter J. Nelson, Hermann-Josef Gröne 
Kidney International 
Volume 65, Issue 3, Pages (March 2004)
DOI: /j x
Copyright © 2004 International Society of Nephrology Terms and Conditions

2. Figure 1
Immunohistology of control (A, D) and hydronephrotic kidneys (B, E, C, F). Monocytes/macrophages were rarely demonstrated by CD68 in control kidney (A). In peritubular capillaries and in hydronephrotic kidney with extensive fibrosis (C), but showed a strong signal in hydronephrotic kidneys with an inflammatory infiltrate (B). Lymphocytes with the CD45RO antigen were seldom seen in control kidney (D), often documented in hydronephrotic kidney with pronounced inflammatory infiltrate (E), and rarely seen in hydronephrotic kidney with extensive fibrosis (F). Alkaline phosphatase antialkaline phosphatase (APAAP) reaction on formaldehyde-fixed paraffin-embedded tissue.
Kidney International  , DOI: ( /j x)
Copyright © 2004 International Society of Nephrology Terms and Conditions

3. Figure 2
Morphometric study, semiquantitative immunohistochemical data defined the degree of tubulointerstitial inflammation, fibrosis, and tubular atrophy. The quantitative analysis allowed the incorporation of these data into a cluster dendrogram. The dendrogram of the x-axis reflects the relationship among the samples by separating three groups of samples: (I) Con, (II) HN group with high inflammation and low fibrosis (Inflam), and (III) HN group with low inflammation and high fibrosis score (Fibrosis). Con, control; HN, hydronephrosis; inflam, inflammation.
Kidney International  , DOI: ( /j x)
Copyright © 2004 International Society of Nephrology Terms and Conditions

4. Figure 3
Cluster analysis of 375 genes. Cluster analysis was performed on log-transformed values of the fold ratios with gene CLUSTER and visualized in TREEVIEW. The samples are represented in the columns and the genes in the rows. The name of each gene is shown at the right side of each row. Genes that were present at higher levels in the examined group (increased) are shown in progressively brighter shades of red, depending on the fold difference, and genes that were expressed at lower levels (decreased) are shown in progressively brighter shades of green. Genes shown in black were not different between the groups being compared. Genes shown in gray were not expressed. The cluster dendrogram of the y-axis shows predictive genes extracted from the whole gene set. Genes are grouped according the similarity of their expression profile. The degree of similarity and relationship among genes (and samples) are represented by the length of the branches (the shorter, the higher the similarity). The color-coded matrix and the dendrogram of the x-axis reflects the relationship among the samples by separating the identical three groups as known from clustering immunohistologic data. Grouping of genes with similar expression is reflected by the tree on the left side. A cluster containing 31 genes showing at least 3-fold difference in expression relative to control samples was determined to provide the greatest expression level of separation of control, hydronephrosis with dominant inflammation, and hydronephrosis with dominant fibrosis.
Kidney International  , DOI: ( /j x)
Copyright © 2004 International Society of Nephrology Terms and Conditions

5. Figure 4
Confirmation of the array expression pattern of nine genes by real time RT-PCR normalized to 18S rRNA and cyclophilin A mRNA. The left panel shows the array expression levels of each examined gene, and, in comparison, the right panel shows the expression ratios to 18S rRNA and cyclophilin A of the nine confirmed gene. Y-axis shows the average of expression values of all samples analyzed in one group. Each group includes different numbers of samples as already mentioned above. Different ratios are due to the different expression levels of the analyzed genes. RT-PCR, reverse transcription-polymerase chain reaction.
Kidney International  , DOI: ( /j x)
Copyright © 2004 International Society of Nephrology Terms and Conditions

6. Figure 5
Real time RT-PCR quantification of mRNA expression of the nine candidate genes in a series of renal biopsies grouped by characterization of level of inflammation (I) versus fibrosis (F) or mixed morphology (I/F). The graphs show expression ratios of each gene to 18S rRNA (A) and cyclophilin A (B) in the renal biopsies. RT-PCR, reverse transcription-polymerase chain reaction.
Kidney International  , DOI: ( /j x)
Copyright © 2004 International Society of Nephrology Terms and Conditions

7. Figure 6
Gene expression profiles in correlation to clinical outcome of available follow-up data. Real time RT-PCR expression data of the nine genes analyzed were compared between patients requiring renal replacement therapy and chronic stable or remittent course. All analyzed genes revealed a higher mean expression (marked as crossbars) level in the poor renal outcome group. For IL-8, MMP-9, MMP-7, urokinase R, and integrin-β4, differences between both patient groups were statistically significant (depicted as asterisks, P = , P = , P = , P = , P = , respectively). The numbers next to the individual data points identify the specific patient samples defined in Table 2 (A). Dendrogram showing clear separation of patient groups based on gene expression (B). RT-PCR, reverse transcription-polymerase chain reaction.
Kidney International  , DOI: ( /j x)
Copyright © 2004 International Society of Nephrology Terms and Conditions