Figure 1.
Initial characterization of the OS-9 protein.
(A) OS-9 expression in various human cell lines. Equal protein amounts of total cell lysates were used for SDS-PAGE and subsequent Western blotting. For each cell line, two independent samples are shown. Endogenous OS-9 was detected with a polyclonal antibody raised against a peptide corresponding to amino acids 600–667 of isoform 1 of OS-9. (B) Protein stability assay of endogenous OS-9. U2OS cells were treated with the translational inhibitor cycloheximide (100 µM). At indicated time points, whole cell lysates were analysed by immunoblotting. (C) Effect of hypoxia on OS-9 expression. For hypoxia, UT-7 cells were exposed to 1% O2 for 24 h prior to Western blot analysis. To determine any influence of HIF-1α on OS-9 expression under normoxia, cells were incubated with the prolyl hydroxylase inhibitor DMOG (0.5 mM) for 24 h. (D) Protein interaction between OS-9 and PHD2 in vitro. For co-immunoprecipitation, U2OS cells were transiently co-transfected with the plasmids pOS-9-V5 and pPHD2-His, lysed in NP40 buffer, and subjected to immunoisolation with anti-V5 antibody recognizing OS-9 by its V5-tag. OS-9 and its associated proteins were separated by SDS-PAGE and analyzed by Western blot (lane 2). As controls, samples of untransfected (lane 1) cells or cells transfected with a single plasmid (lanes 3–4) were loaded. Representative Western blots are shown for each subfigure.
Figure 2.
OS-9 shows no effect on regulation of HIF-1α.
Total cell lysates were used for SDS-PAGE and subsequent Western blotting. To generate nearly anoxic conditions, cells were exposed to an oxygen consuming chemical system or to 1% or 3% O2 for 4 h to generate hypoxia. (A) U2OS, HeLa and Hep3B cells were transiently transfected with the plasmid pOS-9-V5 48 h prior to the experiment. Lamin A and actin were used as loading controls. (B) U2OS cells were subjected to ER stress by incubation either with tunicamycin (1 µg/ml) or thapsigargin (0.5 µg/ml) for 20 h. To detect HIF-1α under normoxia, cells were treated with DMOG (1 mM) for 4 h. A sample of DMSO-only treated cells was loaded to exclude unspecific side effects of the solvent. (C) U2OS cells were transduced with lentiviral construct pLKO.1-shRNA-OS-9 (shOS-9) mediating a stable knockdown of OS-9 expression. Control cells (c) were transduced with plasmid pLKO.1-puro. Representative Western blots are shown for each subfigure.
Figure 3.
Cellular localization of OS-9 and PHD2.
(A) A lectin gel-shift assay was conducted to test for glycosylated proteins. Total cell lysates of U2OS cells were incubated in the presence or absence of the endoglycosidases EndoH and PNGaseF for 6 h at 37°C. Digest products were separated on a reducing 10% SDS-PAGE gel which contained concanavalin A co-polymerized in the top layer of the separating gel to retard mobility of glycosylated proteins [56]. Glycosylated OS-9 is indicated as ‘g’, deglycosylated OS-9 as ‘d’. (B) Detection of OS-9 and PHD2 in the nuclear fraction. HEK293 cells with and without transfection of the plasmid pcDNA3-OS-9 were separated into nuclear fraction (N) and postnuclear supernatant (PS), the latter containing cytoplasm and organelles. Western blot analysis included BiP, GAPDH and lamin A as typical marker proteins for the ER, the cytoplasm and the nucleus, respectively. (C) Detection of OS-9 and PHD2 in the cytoplasm. HEK293 cells were co-transfected with pcDNA3-OS-9 and pPHD2-V5. For hypoxia, cells were exposed to 3% O2 for 4 h. Cells were treated with 50 µg/ml digitonin and centrifuged to obtain a cytoplasmic (C) and an organelle fraction (O) and subjected to immunoblotting. (D) Isolation of cellular endomembranes. HEK293 cells were lysed mechanically by several passages through a 30½G needle. The postnuclear supernatant was processed further by ultra-centrifugation to separate the organelles (O) from the cytosol (C). High salt treatment (1 M KCl) of the organelle fraction produced a wash fraction (W) that contained dissociated peripheral membrane proteins. For immunoblot analysis of subcellular fractionations, cell aliquots were normalized for cell number prior to loading (B–D). Representative Western blots are shown for each subfigure.
Figure 4.
FRET analysis of protein interaction between OS-9 and PHD2.
HEK293 cells were transiently co-transfected with the appropriate ECFP- and EYFP fusion plasmids. After 48 h, live-cell imaging was conducted by confocal microscopy. (A) FRET efficiencies for each ECFP- and EYFP fusion protein pair were calculated from 25–45 randomly selected cells which displayed various fluorescent acceptor/donor ratios. (B) Scatter plots were fit to a single-site binding model. FRET efficiency E is defined as the percentage of transferred energy relative to the energy absorbed by the donor. Additional details are given under ‘Materials and Methods’. For random FRET, cells were co-transfected with plasmids pECFP-N1 and pEYFP-N1 expressing non-fused ECFP and EYFP. (C) Fluorescent images of co-transfected cells were acquired separately through CFP- and YFP-filter channels using a 40× objective. The fluorescence intensity is visualized in false colors on a color bar from low (blue) to high intensity (white). Marked by red arrows are cellular areas of very low fluorescence intensity in the CFP-filter channel and of high fluorescence intensity in the YFP-filter channel. Each experiment was performed at least three times, representative data are shown.