Figure 1.
Transcriptional abnormalities in oncocytoma and chromophobe RCC.
(A) Genomic regions that have significantly higher (red) or lower (blue) RNA production in renal oncocytoma (ON, n = 10) and chromophobe RCC (CR, n = 10) relative to non-diseased kidney (n = 10) were identified using the comparative genomic microarray analysis (CGMA) method as described in the Materials and Methods. Plotted is the resulting t-statistic obtained for each chromosome arm. Only the most significant results are shown (P<0.005). (B) For each gene on chr 19, the average log2-transformed expression ratio comparing oncocytoma or chromophobe RCC to non-diseased kidney is plotted relative to genomic location. The red circle indicates the location of the centromere. A smoothing curve was fit to the log2-transformed data to highlight regions that contain a disproportionate number of up-regulated genes. (C) Genomic regions that have overall increased (green) or decreased (red) DNA copy number in renal oncocytoma (ON) and chromophobe RCC (CR). SNP-derived DNA copy numbers were computed as described in the Material and Methods. All tumor samples were obtained from female patients except of the single male sample indicated (M). (D) For each SNP on chr 19, the average log2-transformed DNA copy number ratio comparing oncocytoma (n = 4) or chromophobe RCC (n = 3) to a pooled normal reference is plotted relative to genomic location as described in (C).
Figure 2.
Somatic chromosome pairing in renal oncocytoma.
Representative photomicrographs of tri-color interphase FISH on renal oncocytoma touch preparations. White arrows indicate large singlet or proximal doublet signals. In all images DAPI counterstaining is shown in blue. (A,B) Labeling with 19p13.3 (green) and 19q13.31 (red) probes. Image area of the 19q13.31 signal was quantified across multiple cells (n = 25) in the same image plane. Area mean and standard error are shown. (C,E,F) Labeling with 19p telomere (green) and 19q telomere (red) probes or 19p12 (green) and an alpha satellite probe for chr 19 that also cross-hybridizes to chromosomes 1 and 5 (red). Inset highlights centromeric pattern. (D) Schematic representation of frequently observed FISH patterns. (G,H) Whole-arm chromosome paint (WCP) for the chr 19 p-arm (red) and chr 19 q-arm (green). The inset shows a normal cell. Schematic representations of the paired chromosomes are shown below. Dashed lines represent chromosomal regions perpendicular to the plane of the image.
Figure 3.
Overexpression of EGLN2 in renal oncocytoma.
(A) Relative expression of EGLN2 as determined by gene expression microarray. (B) Anti-EGLN2 immunoblot analysis of whole-cell extracts prepared from oncocytoma (T) or patient-matched normal tissue (N) samples (lanes 3–8, top panel) and exogenously expressed EGLN2 controls (lanes 1–2, top panel). U2OS were transfected with plasmid containing human EGLN2 or empty vector alone (MOCK); EGLN2 appears as a single band of 45 kDa. Anti-vinculin immunoblot of whole-cell extracts was performed as a loading control. Asterisk (*) denotes background band. (C) Anti-EGLN2 immunoblot analysis of whole-cell extracts prepared from renal clear cell carcinoma (T) or patient-matched normal tissue (N) samples as described in (B) with the exception that a longer film exposure was required to reveal the EGLN2 protein in normal tissue. (D) In vitro translated Gal4-HA-HIF-1αODD(WT) and Gal4-HA-HIF-1αODD(PA) were treated with or without cellular extracts enriched for EGLN2 and mixed with in vitro translated 35S-labelled HA-VHL. Reaction mixtures were immunoprecipitated with anti-Gal4 antibody, bound proteins resolved on SDS-PAGE and visualized by autoradiography (lanes 3–6). In vitro translated 35S-labelled HA-VHL was also immunoprecipitated with anti-HA or anti-Gal4 antibody as input controls (lanes 1 and 2) and visualized by autoradiography. (E) In vitro ubiquitylation of 35S-labelled Gal4-HA-HIF-1αODD(WT) treated with increasing amounts of exogenous EGLN2 was performed in S100 cellular extracts devoid of VHL or reconstituted with VHL. All reaction mixtures except in lane 6 received purified ubiquitin. Reaction mixtures were immunoprecipitated with anti-Gal4 antibody, resolved on SDS-PAGE and visualized by autoradiography. * denotes uncharacterized modified Gal4-HA-HIF-1αODD; IP: immunoprecipitation; AR: autoradiography.
Figure 4.
Decreased HIF levels associated with hyperoxic cell state.
(A) Normalized densitometry of HIF-1α protein levels (•),HIF-1β protein levels (▴), and HIF-1 DNA-binding activity (⧫) as presented in Figure 5B of the Jiang et al. article. The large (•) highlights the 3% oxygen HIF-1α protein levels in both figures. (B) The densitometry data and oxygen-concentration data presented in (A) were log2-transformed and re-plotted. The summary statistics of the best-fit line are also shown. (C,D) Relative gene expression levels of HIF target genes in clear cell RCC (n = 10) and renal oncocytoma (n = 10) compared with non-diseased kidney (n = 12). For each gene, red indicates increased expression, blue decreased expression. (E) Relative expression of BNI3PL as determined by gene expression profiling.
Figure 5.
BNI3PL is regulated by an EGLN2 mediated oxygen-sensitive response.
(A) BNIP3L expression level was measured by qRT-PCR in U2OS cells transfected with plasmids encoding EGLN2 or empty plasmid (MOCK) under physiologic hypoxia (cyclical 0–7% oxygen) or hyper-oxygenated condition (21% oxygen) (left panel). BNIP3L expression level was measured by qRT-PCR in U2OS cells maintained under hyper-oxygenated condition with or without CoCl2 (right panel). BNIP3L expression was normalized to beta-actin and expression in the MOCK transfected cells was arbitrarily set to 1. Error bars represent the standard deviation between the normalized value versus the MOCK performed in triplicate. (B) The experiment was performed in CAKI cells that contain detectable levels of BNIP3L via Western blot analysis. CAKI cells were transfected with plasmids encoding EGLN2 or empty plasmid (MOCK), lysed, equal amounts of cell lysates separated on SDS-PAGE and immunoblotted with the indicated antibodies (left panel). CAKI cells grown in hyper-oxygenated conditions were treated with or without CoCl2. Equal amounts of cell lysates were resolved by SDS-PAGE and immunoblotted with the indicated antibodies (right panel).