QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online December 11, 2006
doi: 10.1242/10.1242/jcs.03321

This Article Summary Full Text Full Text (PDF) Supplementary Material Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Related articles in JCS Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Brown, C. M. Articles by Wiseman, P. W. Search for Related Content PubMed PubMed Citation Articles by Brown, C. M. Articles by Wiseman, P. W. Social Bookmarking                         What's this?

Probing the integrin-actin linkage using high-resolution protein velocity mapping

Claire M. Brown^1,*,, Benedict Hebert², David L. Kolin³, Jessica Zareno¹, Leanna Whitmore¹, Alan Rick Horwitz¹ and Paul W. Wiseman^2,3

¹ Department of Cell Biology, University of Virginia, Charlottesville, VA, USA
² Department of Physics, McGill University, Montreal, Canada
³ Department of Chemistry, McGill University, Montreal, Canada

View larger version (69K): [in this window] [in a new window]   Fig. 1. (A) Analysis of a computer-simulated image time series of a flowing particle population (20% of the total particle density, vx=-0.1 µm/second, vy=0 µm/second) superimposed over an immobile particle population (80% of the total particle density). The average spatial correlation function for a lag time t=3 seconds is calculated from the original intensities [raa(,,3), left] and from the immobile filtered intensities [r'aa(,,3), right]; note the small flow component correlation peak that appears on the left side of the central correlation function without immobile filtering, this flow component peak is all that is seen in the function after immobile filtering. Contour plots of STICS correlation functions raa(,,t) as a function of lag time for the simulation above without (B) and with (C) immobile population filtering. Contour plots for a simulation of particles undergoing pure diffusion at 0.01 µm2/second (D). Bar, 2 µm.

View larger version (81K): [in this window] [in a new window]   Fig. 2. (A) Velocity maps of -actinin-EGFP and mRFP-actin in a MEF cell plated on 1 µg/ml fibronectin. Analysis was performed on 100 frames imaged at 10 seconds per frame. The color-coding for the velocity vectors is universal within the figure with blue being slow velocities and red being faster velocities. The scale of the velocity vectors is different for each plot and is referenced on each image with a 0.5 µm/minute velocity scale arrow. Spatial scale bar is 5 µm. Pixel size is 0.215 µm. (B) Sample STICS correlation function showing that flow-component peak tracking is quite difficult without immobile population filtering (arrowhead). (C) After immobile population removal, a clear displacement of the flow component peak can be observed (white arrowhead) and tracked to reveal the direction and magnitude of the velocity. This filtering removes the contribution to the correlation function of the large static features, i.e. the actin filament structures. Bar, 1 µm.

View larger version (119K): [in this window] [in a new window]   Fig. 3. STICS analysis of a Quicktime movie from fig. 1A described in Ponti et al. (Ponti et al., 2004). Scale bar, 5 µm; velocity scale bar, 0.5 µm/minute. The data matches very well with data calculated using speckle tracking in the paper, revealing faster dynamics within the lamellipodia at the edge of the cell and slower dynamics in the lamella. Note that the analysis was done on the Quicktime movie and some data compression may have occurred relative to the raw speckle data. Also, the pixel size was estimated using the scale bar from the figure so velocity estimates are only approximate.

View larger version (41K): [in this window] [in a new window]   Fig. 4. (A) Velocity maps of -actinin and actin in a CHO cell plated on 2 µg/ml fibronectin before and after treatment with cytochalasin D (200 nM). The image time series were 200 frames at a rate of 5 seconds/frame. The analysis was done on 100 image frames before and after treatment. The color-coding for the velocity vectors is universal within the figure with blue being slow velocities and red being faster velocities. The scale of the velocity vectors is different for each plot and is referenced on each image with a 0.5 µm/minute velocity scale arrow. Spatial scale bar is 5 µm. (B) Median relative magnitude (black bars) and directional (gray bars) coefficients of velocity vectors after treatment relative to those before for -actinin with inhibitors. A value of 1.0 represents no change. Results were similar for actin. Error bars give the s.d.

View larger version (44K): [in this window] [in a new window]   Fig. 5. Protein velocity maps for CHO K1 cells co-transfected with an adhesion protein-EGFP and actin-mRFP and plated on 2 µg/ml fibronectin. (A) Velocity maps in protruding regions of the cell for EGFP conjugated with 5-integrin (96 frames analysed), paxillin (51 frames analysed), FAK (50 frames analysed), talin (55 frames analysed), vinculin (30 frames analysed), -actinin (100 frames analysed) and the corresponding actin-mRFP velocity maps. The color coding for the velocity vectors is universal with blue being slow velocities and red being faster velocities. The scale of the velocity vectors is different for each plot and is referenced on each image with a 0.5 µm/minute velocity scale arrow. Spatial scale bars are 5 µm. (B) Plot of the median relative magnitude and directional correlation coefficient scores relative to actin for each protein in CHO cells. Pixel size of 0.215 µm. Error bars give the s.d.

View larger version (86K): [in this window] [in a new window]   Fig. 6. Protein velocity maps for 3T3 cells plated on 1 µg/ml fibronectin. (A) Velocity maps of EGFP conjugated paxillin (analysis on 121 frames, 5 seconds/frame), FAK (analysis on 200 frames, 10 seconds/frame) and talin (52 frames, 9 seconds/frame), and the corresponding mRFP-actin velocity maps. Insets are a 2x expansions of the small white box in each image. Velocity scale arrows are 1 µm/minute, spatial scale bars are 5 µm. (B) Plot of the median relative magnitude and directional correlation coefficient scores relative to actin for each protein. Error bars are relative error in the velocity measurement.

View larger version (56K): [in this window] [in a new window]   Fig. 7. (A) Velocity maps for talin-EGFP or paxillin-EGFP and mRFP-actin in CHO-K1 cells on either low (2 µg/ml) or high fibronectin (5 µg/ml). The color-coding for the velocity vectors is universal with blue being slow velocities and red being faster velocities. The scale of the velocity vectors is different for each plot and is referenced on each image with a 0.5 µm/minute velocity scale arrow. Spatial scale bars are 5 µm. (B) Plot of the median directional correlation coefficient scores relative to actin for talin and paxillin on low and high concentrations of fibronectin. Pixel size is 0.215 µm. Error bars give the s.d.

View larger version (25K): [in this window] [in a new window]   Fig. 8. Model of a protruding lamella illustrating how differential retrograde motion of actin and adhesion proteins can be regulated at two levels in the linkage. The actin (green) and -actinin (orange) are always concomitant. Slippage occurs either (A) proximal to -actinin through its interaction with one of the components of the linkage complex (shaded molecules) or (B) proximal to the integrins (blue dimers) perhaps at the talin-integrin linkage. (C) Protrusion can occur through polymerization generated forces when both levels of the adhesion complex are engaged.

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

Twitter