Cell cycle-dependent nuclear translocation of endogenous PTOV1 and flotillin-1. Quiescent serum-starved PC-3 cells were stimulated with 10% FBS for the indicated times and stained by double fluorescence for PTOV1 (red) and flotillin-1 (green). In parallel coverslips, cells were also stained for cyclin D1 (green). Scale bars, 10 μm.

Depletion of PTOV1 and flotillin-1 by RNA interference suggests a dependence on PTOV1 for the nuclear localization of flotillin-1. (A) Specific depletion of PTOV1 and flotillin-1 proteins attained by transfecting PC-3 cells with the indicated double-stranded siRNAs. Forty-eight hours posttransfection, PTOV1 or flotillin-1 levels were analyzed by Western blotting and quantitated by densitometric scanning of the signal (right graphs). Normalization was done against signals for mitogen-activated protein kinase (MAPK). As controls, nonspecific (scrambled [scramb]) siRNAs were transfected in parallel at the same concentrations. (B) Confocal immunofluorescent images of cells transfected with either of two distinct siRNA duplexes targeting flotillin-1, and control scrambled siRNA. RNA duplexes were cotransfected with pEGFP to identify transfected cells. The majority of cells transfected with duplexes targeting flotillin-1, but not those transfected with the control RNA duplexes, have a markedly decreased staining for endogenous flotillin-1. (C) Effect of the depletion of PTOV1 on the nuclear localization of flotillin-1. Confocal immunofluorescent images of cells transfected with siRNA targeting PTOV1 (80 nM) and control scrambled siRNA (80 nM) are shown. RNA duplexes were cotransfected with pEGFP to identify transfected cells. Cells transfected with duplexes targeting PTOV1 have a markedly decreased staining for endogenous nuclear flotillin-1, analyzed by fluorescent staining with anti-flotillin-1. (D) Effect of the depletion of PTOV1 or flotillin-1 on the nuclear localization of the reciprocal proteins. Cells transfected with pEGFP plus different concentrations of PTOV1 siRNA (left panel) or 80 nM Flot-1 siRNA 1 (right panel) or scrambled (scr) siRNA as controls were analyzed by fluorescent staining with anti-flotillin-1 (left) or anti-PTOV1 (right). Nuclear localization of flotillin-1 or PTOV1 was scored in GFP-positive cells (*, P = 0.008; **, P = 0.0001). At least 200 cells were scored for each condition. In panels B and C, images in the red fluorescence range were captured at equal voltages; scale bars, 10 μm.

Cell cycle-dependent protein turnover of endogenous PTOV1 and flotillin-1 and of GFP-flotillin-1. (A) Cell cycle-dependent nuclear translocation of GFP-flotillin-1. PC-3 cells transiently transfected with pGFP-Flot1 were serum starved and restimulated with 10% FBS and fixed at the indicated times, and GFP fluorescence was captured by confocal microscopy. Scale bars, 10 μm. (B) Association of the nuclear localization of endogenous PTOV1 and flotillin-1 (Flot-1) and of transfected GFP-flotillin-1 with cell cycle progression. PC-3 cells were treated as described in the legend of Fig. 3, and the cells with nuclear PTOV1, nuclear flotillin-1, and nuclear cyclin D1 fluorescent staining were scored at each time point. To determine GFP-flotillin-1 localization, transfected PC-3 cells were treated as described above (A), and nuclear GFP fluorescence was scored. At least 200 cells were counted for each condition. (C) Cell cycle-associated variation in protein levels of PTOV1 and flotillin-1. In experiments performed inparallel to those described above (A), whole-cell lysates were analyzed by Western blotting with specific antibodies. Antibody to β-actin was used as a control for protein loadings. (D) Cell cycle-associated changes in the levels of nuclear endogenous flotillin-1. PC-3 cells were subjected to serum depletion and stimulation as described in the legend of Fig. 3, nuclei were isolated (see Materials and Methods) at the indicated time points, and proteins were extracted and analyzed by Western blotting with anti-flotillin-1. Loading normalizations were done relative to total protein content as assessed by Ponceau red staining of the transferred membranes. The bar graph corresponds to the normalized values of the densitometric scans of the bands.

PTOV1 and flotillin-1 are required for normal cell proliferation. The effect of overexpression of exogenous PTOV1 or flotillin-1 or the depletion of the corresponding endogenous proteins on the proliferative state of PC-3 cells was assessed by double fluorescence for GFP, anti-FLAG or anti-HA, and anti-Ki67. At least 200 cells were scored for each condition. The results shown are the means from three independent experiments for each condition. (A) Mitogenic effects of full-length PTOV1 and the two separate PTOV domains of PTOV1. FLAG constructs were transfected, and cells were analyzed 48 h after transfection by double fluorescence staining with anti-Ki67 and anti-FLAG antibodies. For mock transfections with empty FLAG vector and untransfected cells (No T), only the Ki67 staining was scored. (B) Mitogenic activity of flotillin-1 variants. Forty eight hours after transfection of the indicated plasmids, cells were analyzed by double fluorescence for Ki67 and HA reactivities to determine their proliferative capacities. Nuclear (N) or nonnuclear (Non N) staining of the HA reactivity was also scored to determine the capacity for each variant to enter the nucleus. Significance for Ki67 values relative to mock controls is as follows: *, P = 2.97 × 10⁻⁶; **, P = 0.0001; ***, P = 0.19. (C) Overexpression of GFP-flotillin-1 induces cell proliferation. PC-3 cells were transfected with pEGFP, pGFP-PTOV1, or pGFP-Flot1 and analyzed 48 h after transfection for [³H]thymidine incorporation. Induction of proliferation for each plasmid, corrected for transfection efficiencies, is expressed as counts per minute (CPM). (D) Effects of the depletion of PTOV1 or flotillin-1 on the proliferative status of PC-3 cells. Cells were transfected with pEGFP or pCMV-HA as controls, pGFP-PTOV1, or pHA-Flot1 and the indicated siRNAs and assessed by double fluorescence detection for Ki67 and GFP or HA. For cells transfected with pGFP-PTOV1 and pHA-Flot1, fluorescent staining for Ki67 was determined for GFP-positive cells.

PTOV1 interacts with flotillin-1 in mammalian cells. (A) Coimmunoprecipitation of endogenous PTOV1 and flotillin-1 (Flot-1) from PC-3 cells. Total cell lysates were immunoprecipitated with anti-PTOV1 antibody or control rabbit immunoglobulin (IP: CTL). Ten percent of the volume of total cell extracts (Input), the IP, and 10% of the volume of the nonprecipitated fractions (S) were analyzed by Western blotting (WB) with anti-flotillin-1 or anti-PTOV1 antibodies. (B) Flotillin-1 interacts with the second PTOV domain of PTOV1. Total cell lysates from pFLAG-PTOV1, pFLAG A domain (dom), or pFLAG B domain transfections were immunoprecipitated with anti-FLAG or anti-flotillin-1 antibodies. Total cell extracts (10% of the volume [Input]) and the IP and nonimmunoprecipitated fractions (10% of the volume [S]) were analyzed by Western blotting with anti-flotillin-1 or anti-FLAG antibodies. Immunoprecipitation analyses with control rabbit (as shown in panel A) or isotype-matched mouse antibodies were consistently negative and are not shown. (C) PTOV1 and flotillin-1 are distributed in detergent-insoluble floating fractions and also in nuclear fractions. Cells were lysed, and crude nuclear (N) and nonnuclear fractions were extracted. This last fraction was used for sucrose density gradient centrifugation. The fractions (5 to 40% of sucrose) were precipitated by trichloroacetic acid and resuspended in Laemmli buffer, and the entire fractions were analyzed together with the entire crude nuclear extract (N) and the pellet of the gradient (P) from each experiment by Western blotting with antibodies to flotillin-1, PTOV1, uPAR, and actin. (D) Immunocytochemical colocalization of PTOV1 and flotillin-1. Cells were fixed and double labeled for PTOV1 (red) and flotillin-1 (green) endogenous proteins and analyzed by confocal microscopy. Confocal images of the samples were also analyzed along the z axis with sections of 0.6 μm. (E) Nuclei were counterstained with propidium iodide (PI). (F) Immunoprecipitation of flotillin-1 from metabolically labeled PC-3 cells. Cells transfected with pHA-Flot1 were labeled with [³⁵-S]methionine, lysed, and immunoprecipitated with anti-HA, anti-flotillin-1, or a control antibody. Immunoprecipitated proteins were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and fluorography. Scale bars, 10 μm.

Subcellular localization and interaction with PTOV1 of HA-tagged forms of flotillin-1. (A) Association of epitope-tagged flotillin-1 variants with lipid rafts. Nuclear and nonnuclear fractions were obtained from PC-3 cells transfected with the indicated variants. Nonnuclear fractions were further fractionated by sucrose density gradient (Fig. 1). The fractions (5 to 40% of sucrose), the pellet of the gradient (P), and the total crude nuclear extract (N) were analyzed by Western blotting (WB) with anti-HA. The same membranes were reanalyzed for endogenous flotillin-1 (Flot1) partitioning and gave results similar to those of Fig. 1C (endogenous flotillin-1 is shown only for the HA-Flot1 transfectant). (B) Nuclear localization of epitope-tagged variants of flotillin-1. (Left panel) Cells were transfected with the indicated plasmids, and 10% of the crude nuclear extract (N) and of nonnuclear fractions (S) were analyzed by Western blotting with anti-HA antibody. The same membranes were reanalyzed with antibodies to uPAR and to flotillin-2 and found to be negative for plasma membrane protein contamination of nuclear extracts(shown are the HA-Flot1 transfectants). (Right panel) In parallel transfections, the percentage of cells with nuclear HA was assessed for each construction, as recorded by immunocytochemistry. At least 200 cells were counted in each case in two independent experiments. (C) Representative confocal immunofluorescence images of transfected flotillin-1 variants stained with anti-HA (top) and with propidium iodide (PI) (bottom). Control reactions eliminating the primary antibody gave negative staining. Scale bars, 10 μm. (D) Coimmunoprecipitation of FLAG-PTOV1 and flotillin-1 variants. Cells were cotransfected with pFLAG-PTOV1 and HA-tagged flotillin-1 variants, lysates were immunoprecipitated with anti-FLAG antibody, and total cell extracts (10% [Input]), IP, nonprecipitated material (10% [S]), or precipitated proteins from a control antibody (data not shown) were detected by Western blotting with anti-HA and anti-FLAG antibodies. (E) Coimmunoprecipitation of FLAG-PTOV1 and HA-Flot1 in nuclear or nonnuclear fractions. Lysates from cells cotransfected with pFLAG-PTOV1 and pHA-Flot1 were separated into nuclear and nonnuclear fractions and immunoprecipitated with anti-FLAG, and coimmunoprecipitates were detected by Western blotting with anti-HA and anti-FLAG antibodies.