Local Actin Polymerization and Dynamin Recruitment in SV40-Induced Internalization of Caveolae Lucas Pelkmans, Daniel Püntener, and Ari Helenius

Supplementary Material

Supplemental Figure 1. Drug-treated CV-1 cells on coverslips were incubated with SV40 particles labeled with the fluorophore fluorescein-X (FLX-SV40) for 3.5 hours at 37°C in the continuous presence of the drugs. Cells were incubated in medium of pH 4.5, which quenches all extracellular virus signal and analyzed with fluorescence microscopy. Representative images are shown on the left. The fluorescence signal of 38-65 cells was quantified from images with low exposure times to ensure a linear range of intensities (12 bit 1-4095) using Openlab 3.0.4 (Improvision) and expressed as percentage ± standard deviation of the signal in untreated cells. Nys-Prog, genistein, STP, LatA and Jas all inhibit uptake, while OA and vanadate stimulate uptake. Scale bars, 80 m.

Medium version | Full size version

---

Supplemental Figure 2. AF-SV40 particles were bound to cav1-GFP-expressing CV-1 cells at 4°C. After 20 min of incubation at 37°C, cells were fixed and analyzed with fluorescence microscopy. Representative enlargements of the plasma membrane are depicted on the left. The extent of colocalization between AF-SV40 signal and cav1-GFP signal was calculated and presented as the percentage of plasma membrane-bound AF-SV40 signal overlapping with cav1-GFP signal. Treatment with LatA or Jas does not influence the colocalization of AF-SV40 with cav1-GFP. Scale bars, 2.5 m.

Medium version | Full size version

---

Detailed methods

Virus binding experiments. Purified SV40 was radiolabeled with ¹²⁵I (Amersham) using chloramine T. Unbound radioactivity was removed by gel filtration and sucrose gradients. A specific activity of 3 10⁶ cpm/g of viral protein (corresponding to 38.5 cpm/virus particle) was obtained and virus retained full infectivity (1 10⁸ PFU/g viral protein). To standardize our experiments, drug-treated or untreated CV-1 cells were resuspended with trypsin/EDTA and incubated at a density of 10⁶ cells/ml in R-medium (RPMI 1640, 10 mM Hepes pH 6.8, 0.2% BSA) with or without the drugs. Saturating amounts of radiolabeled virus (10⁶ cpm, of which 5% binds in control situations) were added, and cells were rotated end-over-end for 2 hours at 4°C. Control experiments performed on confluent CV-1 cells on 6-cm dishes or on cells brought in suspension by either scraping or treatment with 5 mM EDTA showed that binding efficiencies were not affected by treatment with trypsin/EDTA, that receptors could be saturated, and that unlabeled virus could compete with the binding of radiolabeled virus in equimolar concentrations. After virus binding, 50-l aliquots were taken and diluted into 1000 l ice-cold NT buffer (50 mM Tris-Cl pH 8.6, 100 mM NaCl). Cells were pelleted, supernatant was completely removed, and cell-associated radioactivity was counted.

Internalization assays. Purified SV40 was first labeled with Sulfo-NHS-SS-Biotin (Pierce), before being iodinated (Biotin-SS-SV40-¹²⁵I). We estimated that ~360 biotin molecules were coupled to one virus particle. Biotinylation had no significant effect on virus infectivity (3 10⁷ PFU/g versus 10⁸ PFU/g of viral protein). Most of the radioactivity (90%) could be specifically immunoprecipitated with anti-biotin antibodies (Bethyl laboratories) or anti-SV40 antiserum [H. Anderson, Y. Chen, L. Norkin, Mol. Biol. Cell 7, 1825 (1996)]. When the virus was treatedwith Tris(2-carboxyethyl)phosphine (TCEP) (Pierce), only 5% of the radioactivity (background) was precipitated with anti-biotin antibodies, while anti-SV40 antiserum still precipitated 90% of the radioactivity. After Biotin-SS-SV40-¹²⁵I was bound, CV-1 cells were shifted to 37°C and at the indicated time points, 50-l aliquots were taken and quickly diluted into 1000 l ice-cold NT buffer. Cells were pelleted, and the amount of internalized Biotin-SS-SV40-¹²⁵I was determined as described (2), except that 2 15 mM of the membrane-impermeable reducing agent TCEP was used and protected Biotin-SS-SV40-¹²⁵I was precipitated with Protein A beads and anti-biotin antibodies. Background values (less then 10% of cell-associated radioactivity at 4°C that was TCEP resistant) were subtracted from the results, which were then expressed as the percentage of initially surface-bound Biotin-SS-SV40-¹²⁵I that was internalized. Data represent averages ± standard deviation of six independently obtained values.

Drug treatments. CV-1 cells were pre-incubated for 30 min at 37°C in R-medium containing 1 M latrunculinA (Molecular Probes), 500 nM jasplakinolide (Molecular Probes), 25 g/ml nystatin (Sigma) plus 10 g/ml progesterone (Sigma), 1 M staurosporin (Sigma), 100 M genistein (Sigma), 1 M okadaic acid (Sigma) or 1 mM Sodium orthovanadate (Calbiochem). Drugs were present throughout the experiments. Drug treatments did not result in a loss of cell viability, and the effects were reversible.

Infection assays. Drug-treated CV-1 cells were analyzed with I.I.F. (see below) 20 hours after addition of virus for infection using monoclonal antibodies against SV40 small and large T antigen (PharMingen). Four randomly chosen fields in three independent experiments were analyzed for the amount of cells expressing T antigen. Data were expressed as percentage ± standard deviation of the amount of T-antigen expression in untreated cells.

Single cell uptake assays. CV-1 cells, cotransfected with DsRed-N1 and the appropriate cDNA were incubated with FLX-SV40 (1) for 3.5 hours at 37°C, washed, and analyzed with live microscopy in a medium of pH 4.5. At this pH, the emitted light from extracellular FLX-SV40 excited at 488 nm is completely quenched (1). Transfected cells were visualized in the red channel (DsRed). Images in the green channel (internalized FLX-SV40) were obtained with low exposure times to ensure a linear range of intensities (12bit 1-4095), the signal of 20-35 cells was quantified using Openlab 3.0.4 (Improvision) and expressed as percentage ± standard deviation of the signal in nontransfected cells.

Fluorescence microscopy experiments. Cells were seeded the previous day on 18 mm Alcian blue-coated coverslips and were either transiently transfected with the appropriate cDNA using Superfect (Qiagen) or used directly. Transfected cells with relatively low levels of expression were analyzed 12 hours after transfection. In cotransfection studies, cells were analyzed 24 hours after transfection, when, according to control experiments, 95% of transfected cells expressed both genes. In some experiments, cells were fixed and processed for I.I.F. as described previously (1). A mixture of the monoclonal antibodies 4G10 (Upstate Biotechnology) and PY20 (Santa Cruz Biotechnology) in combination with Alexa Fluor 594- or Alexa Fluor 350-labeled anti-mouse secondary antibodies (Molecular Probes) was used for immunostaining of phosphotyrosines. All microscopy experiments were performed with a fully automated Zeiss Axiovert 100M microscope suitable for recording of Alexa Fluor 350, CFP, GFP/fluorescein, YFP, and Alexa Fluor 594 signals. Images and movies were acquired with a cooled charge-coupled-device (CCD) camera (Hamamatsu) and processed with Openlab 3.0.4 (Improvision) as described (1). Images of each channel were acquired separately with 500-msec exposure time. Time-lapse intervals were 500 msec. Trajectories were obtained by marking the position of moving virus particles in 50 consecutive images. The maximal range was acquired by calculation of the absolute distance between each position on the trajectory respective to the starting position. The maximum distance of 18-37 trajectories was taken and represented as average ± standard deviation.

The amount of overlap between Alexa Fluor 594-labeled SV40 (AF-SV40) signal and caveolin-1-GFP (cav1-GFP) signal was quantified using Openlab 3.0.4 (Improvision). Signals were obtained with low exposure to ensure a linear range of intensities (12 bit, ranging from 1-4095) and of areas in the cell where structures were well resolved (periphery). The cav1-GFP signal was subtracted from the AF-SV40 signal, resulting in nonoverlapping AF-SV40 signal. This was subtracted from total AF-SV40 signal and the resulting overlapping AF-SV40 signal expressed as mean percentage ± standard deviation of total AF-SV40 signal in at least 6 cells.

References

1. L. Pelkmans, J. Kartenbeck, A. Helenius, Nature Cell Biol. 3, 473 (2001).

2. S. Schmid, L. Carter, J. Cell Biol. 111, 2307 (1990)

Movie legends

All movies were recorded with a Zeiss Axiovert 100M microscope, equipped with a 150 W HBO lamp and computer-controlled shutters and filters for separate or simultaneous recording of GFP and Alexa Fluor signals. A 100 Plan Apochromat 1.4 N.A. (Zeiss) objective was used. Images were captured with an ORCA-ER CCD camera (Hamamatsu) using 2 2 binning (camera pixel size: 9.9 m), and processed with Openlab 3.0.4. (Improvision).

To view these movies, download a QuickTime viewer and the brightness of the monitor should be set to the maximal value. Movies are best viewed with a gamma correction of 1.0.

The real time (relative to the first frame of the movie) is always indicated in the lower left corner of the movie (hours:min:sec:msec).

• Movie 1
  Live fluorescence microscopy experiment recorded in CV-1 cells having bound AF-SV40 at 4°C and immediately transferred to the heated microscope stage.

  Image size: 10 10 m
  Recording: 0.77 FPS
  Playback: 20 FPS

• Movie 2              Movie 2A
  Live fluorescence microscopy experiment recorded in CV-1 cells having bound AF-SV40 at 4°C, and incubated for 10 min at 37°C. Arrowheads indicate dynamic virus particles that become trapped during the recording.

  Image size: 10 10 m
  Recording: 0.43 FPS
  Playback: 20 FPS

• Movie 3
  Live fluorescence microscopy experiment recorded in CV-1 cells having bound AF-SV40 at 4°C and incubated for 20 min at 37°C.

  Image size: 10 10 m
  Recording: 0.43 FPS
  Playback: 20 FPS

• Movie 4
  Live fluorescence microscopy experiment recorded in CV-1 cells, pretreated with 1 M LatA, having bound AF-SV40 at 4°C in the presence of LatA and incubated for 20 min at 37°C in the presence of LatA.

  Image size: 9 9 m
  Recording: 0.44 FPS
  Playback: 20 FPS

• Movie 5
  Dual-color live fluorescence microscopy experiment recorded in CV-1 cells expressing caveolin-1-GFP having bound AF-SV40 at 4°C and incubated for 20 min at 37°C. Arrowheads indicate trapped SV40 particles in caveolae.

  Image size: 10 10 m
  Recording: 0.45 FPS
  Playback: 20 FPS

• Movie 5A
  Dual-color live fluorescence microscopy experiment recorded in CV-1 cells expressing caveolin-1-GFP, pretreated with 1 m LatA, having bound AF-SV40 in the presence of LatA at 4°C and incubated for 20 min at 37°C in the presence of LatA. Arrowheads indicate mobile, virus-loaded cav1-GFP spots.

  Image size: 10 10 m
  Recording: 0.22 FPS
  Playback: 20 FPS

• Movie 6
  Live fluorescence microscopy experiment recorded in CV-1 cells expressing caveolin-1-GFP for 12 hours, incubated for 2 hours at 4°C and 20 min at 37°C.

  Image size: 10 10 m
  Recording: 0.79 FPS
  Playback: 20 FPS

• Movie 6A
  Live fluorescence microscopy experiment recorded in CV-1 cells expressing caveolin-1-GFP, pretreated with 1 m LatA, incubated at 4°C and 20 min at 37°C in the presence of LatA.

  Image size: 10 10 m
  Recording: 0.79 FPS
  Playback: 20 FPS

• Movie 7
  Live fluorescence microscopy experiment recorded in CV-1 cells expressing GFP--actin for 12 hours, having bound SV40 at 4°C and incubated for 20 min at 37°C.

  Image size: 50 50 m
  Recording: 0.07 FPS
  Playback: 20 FPS

• Movie 7A            Movie 7B
  Dual color live fluorescence microscopy experiment recorded in CV-1 cells expressing GFP--actin for 12 hours, having bound AF-SV40 at 4°C and incubated for 20 min at 37°C. Arrowheads indicate virus particles that induce polymerization of actin tails.

  Image size: 10 10 m
  Recording: 0.07 FPS
  Playback: 10 FPS

• Movie 8
  Dual-color live fluorescence microscopy experiment recorded in CV-1 cells expressing GFP--actin for 12 hours, incubated for 2 hours at 4°C and 20 min at 37°C.

  Image size: 50 50 m
  Recording: 0.61 FPS
  Playback: 20 FPS

• Movie 9
  Dual-color live fluorescence microscopy experiment recorded in CV-1 cells expressing GFP--actin for 12 hours, pretreated with 1 m LatA, having bound AF-SV40 at 4°C in the presence of LatA and incubated for 20 min at 37°C in the presence of LatA. Arrowheads indicate dynamic virus particles that have recruited actin patches.

  Image size: 10 10 m
  Recording: 0.09 FPS
  Playback: 20 FPS

• Movie 10
  Live fluorescence microscopy experiment recorded in CV-1 cells expressing dynamin II (dyn2)-GFP for 12 hours.

  Image size: 20 20 m
  Recording: 0.54 FPS
  Playback: 100 FPS

• Movie 11
  Dual-color live fluorescence microscopy experiment recorded in CV-1 cells expressing dyn2-GFP for 12 hours, having bound AF-SV40 for 2 hours at 4°C and incubated for 20 min at 37°C. Only one frame for the AF-SV40 signal was taken, which is merged with the subsequent dyn2-GFP time-lapse series. Arrowheads indicate virus particles on which dyn2-GFP spots appear and disappear.

  Image size: 10 10 m
  Recording: 0.49 FPS
  Playback: 20 FPS

• Movie 12
  Dual-color live fluorescence microscopy experiment recorded in CV-1 cells expressing dyn2-GFP for 12 hours, pretreated with 100 m genistein, having bound AF-SV40 for 2 hours at 4°C in the presence of genistein and incubated for 20 min at 37°C in the presence of genistein. Only one frame for the AF-SV40 signal was taken, which is merged with the subsequent dyn2-GFP time-lapse series.

  Image size: 10 10 m
  Recording: 0.49 FPS
  Playback: 20 FPS