| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Department of Bioengineering and Institute for Medicine and Engineering, University of Pennsylvania, Philadelphia, Pennsylvania; and Department of Biomedical Engineering, Boston University, Boston, Massachusetts
Correspondence: Address reprint requests to Daniel A. Hammer, Dept. of Bioengineering, University of Pennsylvania, 3320 Smith Walk, 120 Hayden Hall, Philadelphia, PA 19104. Tel.: 215-573-6761; E-mail: hammer{at}seas.upenn.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
–5). The interaction between a cell and its substrate has a specific, important ability to control cell morphology via traction exerted through cellular receptors onto the ECM (6). However, little is understood about how the initial interaction between a cell and extracellular matrix results in the generation of traction forces during spreading.
The initial contact between a cell and the ECM is mediated by the adhesion of the cell to the extracellular matrix via integrin receptors present on the cell surface (7). Integrin-ligand binding is followed by the extension of pseudopodia from the cell body. As the cell begins to flatten against the substrate, it forms additional bonds with the ECM, rearranging its cytoskeleton to form actin bundles and focal adhesions (8). To spread, a cell must exert force against its membrane and its substrate to allow for pseudopodial extension and the stabilization of nascent contacts. The resultant net force is a balance between the extensional and contractile forces exerted by the cytoskeletal machinery (6). This balance of forces and the resultant degree of spreading, in part, controls cell physiology. The size of cellular extensions determine the number of adhesions between the cell and ECM molecules, thereby increasing the degree of signaling initiated by the ECM to the cell. Additionally, protrusive forces mediate migration by propelling a cell forward. Likewise, cell contraction is essential for migration, as it allows for the release of the rearward portion of the cell as the cell moves forward (9–11). Cell contraction also aids in the remodeling of the extracellular milieu of the cell, as cellular forces have been shown to align fibronectin and collagen fibrils (12). The balance of extensional and contractile forces that mediate processes such as cell migration and ECM remodeling are thought to also control the dynamic changes in cell spreading that occur when an initially spherical cell flattens against a surface; thus, an examination of cell spreading should inform us about the basic mechanisms of cell-substrate interactions.
Several other studies have investigated the process of cell spreading (13–16). Recently, Dubin-Thaler and co-workers studied the dynamics, but not the mechanics, of cell spreading (14). For very short times (
30 min), they tracked the spreading of fibroblasts on different densities of fibronectin. They found that fibroblasts tend to spread anisotropically, and that there was a lag time in the initiation of spreading that decreased with increasing fibronectin concentration. They also showed that over this time period, the extent of spreading increased with fibronectin density.
Although this study done by Dubin-Thaler et al. addressed some intriguing dynamic features of the early stages of spreading, we sought to extend the work by simultaneously studying the entire time course of spreading (up to 4 h after initial contact) and the mechanics of spreading during this period. To this end, we used traction force microscopy (TFM) to measure the forces exerted by bovine aortic endothelial cells (BAECs) during spreading on flexible, RGD-derivatized polyacrylamide substrata. The mechanics of endothelial cell spreading are of particular interest to our laboratory not only for comparison to those previously published (17), but also for a better understanding of endothelial cell adhesion and its role in blood vessel formation. By using a well-defined substrate containing the RGD peptide fragment of full-length fibronectin, the synergistic effects of cryptic sites along the full-length fibronectin that are revealed due to cell adhesion and contraction are eliminated. In this way, we have designed a simple, well-defined system to study the mechanics and dynamics of endothelial cell spreading.
Our study of the forces generated during endothelial cell spreading will allow us to address a key issue in cell substrate mechanics—the relationship between focal adhesion formation and stress generation. Techniques similar to traction force microscopy have been used to quantify the forces exerted by cells constrained to micropatterned ECM-coated islands (18) and the traction stresses at focal adhesions (19,20). Balaban and co-workers largely conclude that forces generated by a cell are wholly attributable to the size and number of focal adhesions, whereas Tan and co-workers report a bimodal relationship between contact area and force, where force and contact area are related for areas greater than 1 µm2 and those adhesions smaller than 1 µm2 appear to exert large forces that do not correlate to size. Similarly, Wang and co-workers have shown that force generation is more likely attributable to nascent focal adhesions, which are generally smaller than 1 µm2, at the front of migrating cells (21). In our previous studies using TFM to investigate the traction forces exerted by well-spread endothelial cells, we found that changes in the density of ligand on the substrate lead to changes in cell area, and the total amount of force exerted by the cell on its substrate is directly correlated with the cell's area (17). Our previous results, in combination with the results of others, imply that the force/area relationship found in our experiments is coincidental with the expansion of focal adhesions as area increases. The study presented here directly addresses this issue by measuring forces exerted by cells during the process of spreading, before stress fiber and significant focal adhesion formation.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Before all spreading experiments, cells were replated sparsely on tissue culture-treated polystyrene dishes. Eight hours later, the cells were serum-starved in DMEM supplemented with 0.5% penicillin-streptomycin and 1% 200 mM L-glutamine for 15 h. After serum-starvation, the cells were replated for experimentation.
Surface preparation
Coverslips (No. 1, 45 x 50 mm, Fisher Scientific, Pittsburgh, PA) were chemically activated in preparation for covalent attachment of polyacrylamide sheets using the method adapted from the protocol described by Wang and Pelham (22). Briefly, the coverslips were passed through the flame of a Bunsen burner, coated with a thin layer of 0.1 N NaOH and allowed to dry. A layer of 3-aminopropyltrimethoxysilane (200 µl, Sigma-Aldrich Chemical, St. Louis, MO) was coated on the coverslips, incubated, and repeatedly bathed in distilled water followed by a 30-min incubation in 0.5% glutaraldehyde (70% aqueous stock solution, Sigma-Aldrich Chemical) in phosphate-buffered saline (Invitrogen, Carlsbad, CA) at room temperature. The coverslips were rinsed thoroughly with distilled water and air-dried horizontally, overnight, in a fume hood.
Synthesis of the bifunctional linker
N-6-((acryloyl)amino)hexanoic acid (N-6) was synthesized using the method described by Pless et al. (23). The N-6 copolymerizes in the acrylamide to form a reactive polyacrylamide gel. The N-6 contains an n-succinimidyl ester that is displaced by a primary amine to link the amine-containing ligand to the polyacrylamide substrates.
Gel synthesis
Gels were prepared using 5% acrylamide (40% w/v solution), 0.1% n,n'-methylene-bis-acrylamide (2% w/v solution), n'-tetramethylethylene diamine, and ammonium persulfate from Bio-Rad Laboratories (Hercules, CA). Additionally, the gels contained 54 mM HEPES, 20 µmol/ml of N-6 dissolved in ethanol, 2:25 volume of carboxylate-modified fluorescent latex beads (0.5 µm Fluospheres, Molecular Probes, Eugene, OR), and 1:2000 volume of n'-tetramethylethylene diamine. The pH of the monomer solution was adjusted to 6.0 with 3.0 N HCl, the solution was degassed and a 1:200 volume of 10% ammonium persulfate was added. Twenty microliters of the solution was dispensed onto the activated coverslips and the drop was flattened using a circular coverslip (No. 1, 22-mm diameter, Fisher Scientific, Pittsburgh, PA). The entire assembly was turned upside-down and incubated at room temperature for 60 min, allowing polymerization to occur. The circular coverslip was removed and the gel was rinsed in ice-cold DI water for 10 min. This protocol produces a gel that is 70-µm thick, with a Young's modulus of 2500 Pa, as described previously (17).
Covalent attachment of the ligand to the gel
In these experiments, an RGD-containing nonapeptide (with a sequence of NH2-Tyr-Ala-Val-Thr-Gly-Arg-Gly-Asp-Ser-OH) was covalently linked to the polyacrylamide gels. The gels were removed from the DI water and incubated in a 50 mM HEPES (pH 8.0) solution containing peptide. The gel was rinsed with DI water and the unreacted linker capped with ethanolamine. The gels were rinsed with ice-cold DI water and stored at 4°C in DI water.
The peptide density is measured using a solution of 5N NaOH to hydrolyze the amide linkage between the peptide and the acrylamide surface. The concentration of peptide released into solution is determined by comparing its absorbance at 280 nm to that of standard solutions. The density of peptide on the surface of the gel can be approximated by converting the overall concentration of the peptide within a volume of gel to a surface density of ligand by assuming a 1-nm penetration depth into the gel in which a cell can bind the peptide. The density of ligand on the surface of the gel in these experiments ranges from 150 to 1500 mol/µm2.
Cell spreading
Cells were pretreated as described above and replated into Leibovitz L-15 media (Invitrogen, Carlsbad, CA) supplemented with 10% calf serum, 0.5% penicillin-streptomycin, and 1% 200 mM L-glutamine. The cells were plated at a density of 5500 cells/cm2 onto polyacrylamide gels and immediately placed onto an Olympus Inverted Ix70 Microscope stage (Olympus America, Melville, NY) in a chamber maintained 37°C for observation. Images of the cells were taken periodically and later analyzed for cell area and perimeter using Scion Image software (Scion, Frederick, MD). The light path contains a green monochromatic IF-550 filter (Olympus America) to minimize light damage to the cells.
Traction force microscopy of spreading cells
Traction force microscopy has been described previously (10). Briefly, traction forces were determined based on deformations in the polyacrylamide substrate relative to the relaxed substrate as detected by movements of 0.5-µm beads embedded in the gel. Phase contrast images of the cell were taken periodically during spreading. Directly after a phase image was taken, a corresponding fluorescent image of the beads embedded beneath the cell was taken. At the end of 4 h of spreading, the cell was removed using trypsin-EDTA and an image of the beads in their unstressed state was captured. Using custom-written software, the bead displacements within the gel were calculated, the cell and nucleus outlines were drawn, and a mesh that fits within the outline of the cell was created. Using the bead displacements and the material properties of the gel, the most likely surface traction vectors were calculated using the technique described by Dembo and Wang (10).
The overall force, |F|, exerted by the cell on its substrate, is an integral of the traction field magnitude over the area, 
where T(x,y) = [Tx(x,y), Ty(x,y)] is the continuous field of traction vectors defined at any spatial position (x,y) within the cell.
Immunostaining of focal adhesions and stress fibers
BAECs were plated sparsely on tissue culture-treated polystyrene dishes for 8 h before 15 h of serum-starvation in DMEM supplemented with 0.5% penicillin-streptomycin and 1% 200 mM L-glutamine. After serum-starvation, the cells were replated in standard media on polyacrylamide gels conjugated with an RGD-containing peptide. At various time points, the cells were fixed in 2% paraformaldehyde, followed by permeabilization using 0.2% Triton X-100. The samples were each blocked using 10% goat serum (Sigma Aldrich Chemical) and stained with either Alexa546-labeled phalloidin (Molecular Probes) or murine anti-vinculin monoclonal antibody (Upstate Cell Signaling Solutions, Charlottesville, VA) followed by AlexaFluor 488 conjugated goat anti-mouse IgG secondary antibody (Molecular Probes). The phalloidin-stained actin was visualized using an Olympus Inverted IX70 Microscope with an Olympus 60x, NA 0.7, phase objective and images were gathered using a SPOT RT Slider Camera (Diagnostic Instruments, Sterling Heights, MI). F-actin polymerization was quantified by calculating the average fluorescence intensity of a population of cells after background subtraction. Focal adhesions were visualized using an Olympus Inverted Ix81 Microscope with an Olympus 20x, NA 0.60 objective and images were gathered using a Cooke Sensicam QE camera (Cooke, Auburn Hills, MI) (generously provided by Dr. Michael King, University of Rochester, Rochester, NY).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
|
 | (1) |
). Additionally, based on the experimental data in Fig. 2 A, an early inflection point is evident, a feature lacking in a simple exponential function.
|
|
) and as seen in Fig. 2 A, the area increases with increasing peptide concentration. To better elucidate this relationship, Fig. 2 B plots A50, the half-maximal area, versus the peptide concentration. When plotted on a semi-log scale, A50 and the concentration of peptide are linearly related and can be closely fit to the equation
 | (2) |
) that spreading rate is independent of ligand density. We find that
 | (3) |
In addition to calculating the area of the cell, we investigated changes in the perimeter of the cell in response to spreading. Plotting the changes in perimeter with respect to changes in area on a log-log plot (Fig. 3) reveals a power-law relationship of the form
 | (4) |
is the dimensionless perimeter of the cell (P/P0), 
is the dimensionless area of the cell (A/A0), and 
and ß are presented in Table 2. On a log-log scale, is linear with , and the slope is characterized by ß. The physical significance of ß is illustrated in Fig. 4 for two distinct limits of spreading. First consider the growth of a cell maintaining an approximately circular outline, a phenomenon that can be referred to as isotropic spreading (Fig. 4 A). In this case, the ß-exponent approaches 0.5, and a quadrupling in cell area corresponds to a doubling of the cell perimeter. This is in contrast to the spreading of a spindlelike cell with many thin extensions, so-called "anisotropic" spreading (Fig. 4 B). In this case, the exponent ß approaches 1 in the limit of thin extensions. As the pseudopod thickness increases, ß falls between 0.5 and 1, with larger pseudopods leading to smaller values of ß approaching 0.5. Therefore, by investigating the relationship between area and perimeter during spreading in this manner and quantifying ß, we have been able to quantify the mode of spreading as a function of ligand density. Data regression indicates that at higher densities of ligand, the average behavior of a cell is isotropic, where the cell spreads while maintaining a circular morphology, as indicated by the values of ß in Table 2. This is in contrast to the behavior at lower densities of ligand where the cell spreads anisotropically by extending numerous pseudopodia.
|
|
|
 | (5) |
d is the damping coefficient, and f
is the steady-state value of the shape factor. Each of the four data sets in Fig. 5 is fit to this model and the resulting parameter values are presented in Table 3. Notably, the model is a two-parameter fit where the oscillation amplitude (0.09 µm–1), the natural frequency 
and the phase shift 
are conserved over all four ligand concentrations; only f and d vary as a function of ligand concentration. The oscillations in the shape factor are prevalent at low ligand densities, but do not exist at higher ligand densities. The negative changes of f with t are indicative of periods where the area is increasing more quickly than the perimeter, which would be consistent with cell spreading to fill regions staked out by earlier pseudopodial extensions. However, increasing values of f versus A indicate the perimeter is increasing more quickly than the area, indicative of pseudopodial extension; the oscillation in f versus t indicates that, at low densities of ligand, there is alternation of pseudopodial extension and periods where the cell body fills in the area between extensions to create a rounder morphology. The damping coefficient increases with the ligand density, indicating that higher ligand densities either, 1), prevent substantial pseudopodial extension before the movement of the rest of the cell body; or either, 2), display a nonsynchronous pseudopodial extension among a population such that the average behavior results in what appears to be relatively isotropic spreading when ensemble-averaged. Based on qualitative observation, it appears that both scenarios contribute to the lack of synchronicity. At high concentrations of peptide, the pseudopodial extension form at random times in the cell population over time, and cells tend to spread more isotropically than on lower densities of peptide.
|
|
). Other studies have indicated that the relationship between force and the cell area is actually controlled by the size and area of the focal adhesion complexes independent of the cell area (19,20). To further investigate the origin of force generation, we examined the forces exerted by endothelial cells during spreading, before focal adhesion and stress fiber formation. Cells were plated on polyacrylamide gels derivatized with varying densities of RGD and observed during spreading up to a period of 4 h. Traction force microscopy was used to measure the forces exerted by the cells on its substrate during spreading (Fig. 6). Even at very low concentration of ligand, cells exert significant, detectable traction forces early in spreading. At the earliest stages of spreading, these tractions are uniform and directed inward, at both low and high ligand densities. As the cell begins to spread by extending projections, stresses are concentrated at the pseudopodial tips (Fig. 6, A–D). Despite the outward extensions of the membrane, the results presented here indicate that the basal surface of the cell pulls inwardly on the surface in a state of contraction. This same pattern exists in all the cells observed in our study, regardless of the concentration of ligand. Additionally, strong tractions localized to a given area of the cell are not necessarily predictors of the formation of later pseudopodial extensions. Likewise, early pseudopodial extensions do not necessarily indicate the positions of future extensions.
|
 | (6) |
|
), we hypothesized that the generation of tractions would correspond to the development of stress fibers. BAECs were pretreated as described above by preplating and serum starvation. The cells were then plated on polyacrylamide gels derivatized with 1.0 mg/ml of RGD-containing peptide. At various times, the cells were fixed and stained using Alexa-546-phalloidin.
The quantity of f-actin was quantified by measuring the total fluorescent intensity of a population of cells at each time (Fig. 8). The total amount of filamentous actin increases during spreading and peaks at 3 h, and then decreases to a steady-state value by 24 h. As illustrated in Fig. 6, force generation well precedes 3 h. In fact, detectable forces are exerted within 30 min of plating, before any detectable changes in actin polymerization, which we observed to occur 60 min after plating. Additionally, qualitative examination of the images of Fig. 8 B reveals that stress fibers do not appear until 2–3 h after plating. When compared with the results presented in Fig. 6, these data strongly suggest that cells are capable of exerting significant forces before complete actin polymerization or visible stress fiber formation.
|
). When the total force exerted by the cell is plotted against the cell area for every cell over all concentrations studied for t greater than 3 h, force and area again appear to be linearly related (Fig. 9). The slope of this plot is 7.1 x 103 dyn/cm2, and is statistically similar to the relationship we found in our previous study of cells that had been plated for 24 h before force measurements (17).
|
,11,19–21,25–27). In some cases, these studies conclude that the forces exerted by the cell originate from focal adhesions (19, 20). We investigated this assumption by looking at the time course of focal adhesion formation relative to the time course for traction force generation.
Vinculin was chosen as a marker for focal adhesion formation (28) because it has been shown to be one of the first intercellular molecules recruited to the sites of focal complexes, precursors to focal adhesions (29). We found that focal adhesions appear at varying times during spreading and depend on the density of ligand on the substrate (Fig. 10). Before adhesion, vinculin is spread diffusely throughout the cell. In cells plated on the highest density of peptide, vinculin clustering occurs relatively early, between 45 and 60 min. At the two intermediate values of ligand density, vinculin clustering appears between 60 min and 2 h. In cells plated on the lowest density of peptide, there is very little vinculin clustering at all, even at 24 h. Clearly, the number of vinculin clusters and timing of the appearance of the clusters is in part dictated by the density of ligand on the substrate.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
). Observation of the full time course of spreading has only been possible when the light is extremely minimized by limiting the light intensity of the bulb in addition to using green and neutral density filters in series. This longer exploration of spreading dynamics allows us to correlate the dynamics of spreading with force and ligand density. Using time-lapse microscopy, immunofluorescence, and traction force microscopy, we have found that:
Several studies have reported a relationship between cell area and ECM density (2,31,32), but here we report that ligand density increases the rate of spreading. In a separate study of cell spreading recently performed by Dubin-Thaler et al., it was concluded that in mouse embryonic fibroblasts, ligand density affects the lag time before cells start to spread, but the rate of spreading is independent of ligand density (14). Our data confirms that the initial rate of spreading is dependent on ligand density; however, in contrast to Dubin-Thaler et al., our data show that, in BAECs, the maximal rate of area growth varies significantly with ligand density. Because Dubin-Thaler et al. and Giannone et al. both waited for the onset of spreading before observing and measuring fibroblast spreading, and only tracked cells for 35 min, it is possible that we are able to observe changes in cell behavior that they were not able to detect because of their limited observation time. It should be noted that the differences in cell behavior between the study presented here and that of Sheetz and co-workers (14,15) could also be explained by the differences in the cell types investigated.
The changes in cell area with time revealed a monotonic increase to a plateau in cell area. The data was fit to an error function, which successfully incorporates all of the features of the curves and fits all four data sets (Fig. 2 A). The error function itself is interesting, in that it represents the integral of a normal distribution (Fig. 11). The derivative of our data may be thus interpreted as a normal distribution of some cellular event over time. We can only speculate on the nature of these events, but we suspect that it is the binding of integrin and ligand. At early time points, the binding of ligand is minimal, but increases with time as the cell spreads. At some later point during spreading, the majority of integrin molecules are bound and it becomes less likely that a new bond will form, thus leading to a decrease in new integrin ligation. Thus, integrin ligation may very well give rise to the changes in area observed.
|
,15), as they tracked changes in cell radius at evenly spaced points along the circumference of a cell. The power law relationship that exists between perimeter and area indicates that cells on higher densities of ligand spread isotropically, whereas cells on lower densities of peptide are more likely to spread anisotropically (Fig. 3). Dubin-Thaler et al reported that isotropic spreading is most prevalent in serum-deprived cells (14). During serum-deprivation, membrane receptors are not occupied by the growth factors often found in serum, and therefore are more available for binding with substrate-bound ligand. Likewise, when the cell encounters high densities of ligand, as was tested in the experiments presented here, the balance toward soluble-ligand bound receptors and substrate-ligand bound receptors shifts toward the binding of the receptors to the substrate. The number of substrate-ligand bound receptors may, in fact, determine the threshold between isotropic and anisotropic spreading.
Our investigation into the dynamics of cell spreading includes a comparison of the changes in cell shape that accompany changes in cell area. We defined a parameter, called the shape parameter, f, as the ratio of perimeter/area. The shape parameter is a measure of the approximate number of pseudopodia per unit length for a population of cells. At low concentrations of peptide, the spreading behavior of cells (f versus A) oscillates between periods of area increasing more quickly than the perimeter and periods of the perimeter increasing more quickly than the area. This oscillation implies that cells on a lower density of peptide, spread by a cycle of extending pseudopodia, followed by a general increase in the cell body in concert over a population. At higher densities of peptide, there are no oscillations and the shape parameter indicates that the cells are spreading isotropically. Perhaps the increased ligand density acts as the damper represented in our model, preventing significant pseudopodial extension. The comparison of obtained damping coefficients may indicate that at lower densities of peptide, the cell is able to send out extensions more readily, since the membrane is less tightly bound to the substrate, whereas at higher densities of ligand, the cell is so firmly bound it can only spread in a pancakelike manner. A similar behavior occurs during cell migration (29,33,34), where a cell can be so tightly bound at high densities of ECM that it does not send out the extensions necessary for motility, yet at low densities a cell sends out filopodia to sense its substrate and the filopodia become the foundation for the extension of a lamellipodium.
The oscillations seen in Fig. 5 occur even though the shape factor is a result of an average over the perimeter of each cell and an average over many cells. These observations are distinct from results recently published by Dubin-Thaler et al. who reported that over a small section of cell perimeter, the radius increases in stochastic, transient extension periods (STEPs) (14). In that study, the steps are described for individual sections along the perimeter of a cell and persist for <1 min, whereas our study shows that over the cell population the duration of the oscillations last over many minutes. For the oscillations described here and that of Dubin-Thaler et al. to be representative of the same phenomenon, this would imply that the subcellular oscillations of the Dubin-Thaler et al. study are somehow synchronized between cells and perhaps within a single cell. Such a cooperative behavior was not revealed by the single cell studies of Dubin-Thaler et al. (14).
Cells can exert forces early on in spreading, before any significant stress fiber or focal adhesion formation. The magnitude of these forces is comparable to those exerted by the cell at later times. Our results are the first to show the presence of traction forces without the presence of focal adhesions. Several studies have shown that small nascent focal adhesions are capable of forces greater than those measured for mature focal adhesions (19,21); therefore we cannot rule out the possibility that there exist small nascent focal adhesions in our spreading cells that we are unable to detect. An additional study done by Galbraith et al. found that adhesion between an ECM-ligand-coated bead and integrin receptors gain the ability to exert measurable forces upon maturation into focal complexes, as marked by the recruitment of vinculin (29), indicating that mechanical force applied to the cell stimulates the strengthening of adhesions through the recruitment of vinculin. In contrast, we show that the initial adhesions formed between the cell and its substrate are sufficiently strong to exert measurable traction forces before the detection of vinculin clustering. Future studies that will lead to a better understanding of focal adhesion formation and traction generation should include additional cell spreading studies that monitor simultaneously for GFP-tagged vinculin and traction force generation. Ultimately, the systematic monitoring of real-time recruitment of fluorescently tagged focal adhesion proteins to adhesion sites in conjunction with the use of TFM will lead to critical insights into the molecular mechanisms of traction generation.
In addition to exerting forces early in adhesion, a spreading cell displays an organized distribution of forces. These forces all appear to be directed inward toward the cell center, despite the movements of the cell membrane outward. This general pattern of traction forces has been reported before in migratory fibroblasts (35), indicating a ubiquitous feature of force generation, likely controlled by the fundamental nature of spreading and contractility in cells. The key difference between migration and spreading, however, is that in spreading, all portions of the membrane tend to be extending, and there is no clear direction for the cell. Thus, the lamellipod-pseudopod asymmetry commonly seen in migration does not seem to exist in spreading; however, the modes of membrane extension and force generation appear to be similar.
Previously, we showed that in cells that have been plated for 24 h, given adequate time to spread and form focal adhesions, traction forces are proportional to the area of the cell (17). Knowing that focal adhesions are the points of contact between a cell and the substrate and that the number of focal adhesions increase with area (36), one might conclude that the relationship between the force exerted by the cell and the area of the cell is, in fact, a relationship between the force exerted by the cell and the total number, or individual size, of focal adhesions. However, the results presented here indicate that total force exerted during spreading is proportional to the cell area before significant focal adhesion formation. Therefore, it is possible that vinculin clustering is not necessary for traction generation.
The existence of forces before the appearance of focal adhesions or stress fibers is intriguing for several reasons. These data may indicate that other techniques used to measure forces by assuming that all forces originate from focal adhesions may be neglecting forces produced elsewhere across the membrane-ECM interface. Note significant force generation has occurred before focal adhesion formation, even when comparing to systems where ultimately strong forces are exerted at steady state. Clearly, this suggests that individual integrin-ligand bonds can transmit force before clustering into focal complexes.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
The work was funded in part by National Institutes of Health grants No. HL64388 to D.A.H. and No. GM61806 to M.D., and a Whitaker Graduate Fellowship to C.A.R.
Submitted on October 11, 2004; accepted for publication March 24, 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
2. Ingber, D. E. 1990. Fibronectin controls capillary endothelial cell growth by modulating cell shape. Proc. Natl. Acad. Sci. USA. 87:3579–3583.
3. Asthagiri, A. R., C. A. Reinhart, A. F. Horwitz, and D. A. Lauffenburger. 2000. The role of transient ERK2 signals in fibronectin- and insulin-mediated DNA synthesis. J. Cell Sci. 113:4499–4510.[Abstract]
4. Gospodarowicz, D., and G. M. Lui. 1981. Effect of substrata and fibroblast growth factor on the proliferation in vitro of bovine aortic endothelial cells. J. Cell. Physiol. 109:69–81.[CrossRef][Medline]
5. McAuslan, B. R., V. Bender, W. Reilly, and B. A. Moss. 1985. New functions of epidermal growth factor: stimulation of capillary endothelial cell migration and matrix dependent proliferation. Cell Biol. Int. Rep. 9:175–182.[CrossRef][Medline]
6. Lauffenburger, D. A., and A. F. Horwitz. 1996. Cell migration: a physically integrated molecular process. Cell. 84:359–369.[CrossRef][Medline]
7. Hynes, R. O. 1987. Integrins: a family of cell surface receptors. Cell. 48:549–554.[CrossRef][Medline]
8. Burridge, K., K. Fath, T. Kelly, G. Nuckolls, and C. Turner. 1988. Focal adhesions: transmembrane junctions between the extracellular matrix and the cytoskeleton. Annu. Rev. Cell Biol. 4:487–525.[CrossRef][Medline]
9. Maheshwari, G., and D. A. Lauffenburger. 1998. Deconstructing (and reconstructing) cell migration. Microsc. Res. Tech. 43:358–368.[CrossRef][Medline]
10. Dembo, M., and Y. L. Wang. 1999. Stresses at the cell-to-substrate interface during locomotion of fibroblasts. Biophys. J. 76:2307–2316.
11. Galbraith, C. G., and M. P. Sheetz. 1997. A micromachined device provides a new bend on fibroblast traction forces. Proc. Natl. Acad. Sci. USA. 94:9114–9118.
12. Tranquillo, R. T., M. A. Durrani, and A. G. Moon. 1992. Tissue engineering science: consequences of cell traction force. Cytotechnology. 10:225–250.[CrossRef][Medline]
13. de Hoog, C. L., L. J. Foster, and M. Mann. 2004. RNA and RNA binding proteins participate in early stages of cell spreading through spreading initiation centers. Cell. 117:649–662.[CrossRef][Medline]
14. Dubin-Thaler, B. J., G. Giannone, H. G. Dobereiner, and M. P. Sheetz. 2004. Nanometer analysis of cell spreading on matrix-coated surfaces reveals two distinct cell states and STEPs. Biophys. J. 86:1794–1806.
15. Giannone, G., B. J. Dubin-Thaler, H. G. Dobereiner, N. Kieffer, A. R. Bresnick, and M. P. Sheetz. 2004. Periodic lamellipodial contractions correlate with rearward actin waves. Cell. 116:431–443.[CrossRef][Medline]
16. Wakatsuki, T., R. B. Wysolmerski, and E. L. Elson. 2003. Mechanics of cell spreading: role of myosin II. J. Cell Sci. 116:1617–1625.
17. Reinhart-King, C. A., M. Dembo, and D. A. Hammer. 2003. Endothelial cell traction forces on RGD-derivatized polyacrylamide substrata. Langmuir. 19:1573–1579.[CrossRef]
18. Parker, K. K., A. L. Brock, C. Brangwynne, R. J. Mannix, N. Wang, E. Ostuni, N. A. Geisse, J. C. Adams, G. M. Whitesides, and D. E. Ingber. 2002. Directional control of lamellipodia extension by constraining cell shape and orienting cell tractional forces. FASEB J. 16:1195–1204.
19. Tan, J. L., J. Tien, D. M. Pirone, D. S. Gray, K. Bhadriraju, and C. S. Chen. 2003. Cells lying on a bed of microneedles: an approach to isolate mechanical force. Proc. Natl. Acad. Sci. USA. 100:1484–1489.
20. Balaban, N. Q., U. S. Schwarz, D. Riveline, P. Goichberg, G. Tzur, I. Sabanay, D. Mahalu, S. Safran, A. Bershadsky, L. Addadi, and B. Geiger. 2001. Force and focal adhesion assembly: a close relationship studied using elastic micropatterned substrates. Nat. Cell Biol. 3:466–472.[CrossRef][Medline]
21. Beningo, K. A., M. Dembo, I. Kaverina, J. V. Small, and Y. L. Wang. 2001. Nascent focal adhesions are responsible for the generation of strong propulsive forces in migrating fibroblasts. J. Cell Biol. 153:881–888.
22. Wang, Y. L., and R. J. Pelham, Jr. 1998. Preparation of a flexible, porous polyacrylamide substrate for mechanical studies of cultured cells. Methods Enzymol. 298:489–496.[Medline]
23. Pless, D. D., Y. C. Lee, S. Roseman, and R. L. Schnaar. 1983. Specific cell adhesion to immobilized glycoproteins demonstrated using new reagents for protein and glycoprotein immobilization. J. Biol. Chem. 258:2340–2349.
24. Katoh, K., Y. Kano, and K. Fujiwara. 2000. Isolation and in vitro contraction of stress fibers. Methods Enzymol. 325:369–380.[Medline]
25. Oliver, T., K. Jacobson, and M. Dembo. 1998. Design and use of substrata to measure traction forces exerted by cultured cells. Methods Enzymol. 298:497–521.[Medline]
26. Shiu, Y. T., S. Li, W. A. Marganski, S. Usami, M. A. Schwartz, Y. L. Wang, M. Dembo, and S. Chien. 2004. Rho mediates the shear-enhancement of endothelial cell migration and traction force generation. Biophys. J. 86:2558–2565.
27. Wang, H. B., M. Dembo, and Y. L. Wang. 2000. Substrate flexibility regulates growth and apoptosis of normal but not transformed cells. Am. J. Physiol. Cell Physiol. 279:C1345–C1350.
28. Geiger, B., K. T. Tokuyasu, A. H. Dutton, and S. J. Singer. 1980. Vinculin, an intracellular protein localized at specialized sites where microfilament bundles terminate at cell membranes. Proc. Natl. Acad. Sci. USA. 77:4127–4131.
29. Galbraith, C. G., K. M. Yamada, and M. P. Sheetz. 2002. The relationship between force and focal complex development. J. Cell Biol. 159:695–705.
30. McKenna, N. M., and Y. L. Wang. 1989. Culturing cells on the microscope stage. Methods Cell Biol. 29:195–205.[Medline]
31. Gaudet, C., W. A. Marganski, S. Kim, C. T. Brown, V. Gunderia, M. Dembo, and J. Y. Wong. 2003. Influence of type I collagen surface density on fibroblast spreading, motility, and contractility. Biophys. J. 85:3329–3335.
32. Truskey, G. A., and T. L. Proulx. 1993. Relationship between 3T3 cell spreading and the strength of adhesion on glass and silane surfaces. Biomaterials. 14:243–254.[CrossRef][Medline]
33. Palecek, S. P., J. C. Loftus, M. H. Ginsberg, D. A. Lauffenburger, and A. F. Horwitz. 1997. Integrin-ligand binding properties govern cell migration speed through cell-substratum adhesiveness. Nature. 385:537–540.[CrossRef][Medline]
34. DiMilla, P. A., K. Barbee, and D. A. Lauffenburger. 1991. Mathematical model for the effects of adhesion and mechanics on cell migration speed. Biophys. J. 60:15–37.
35. Munevar, S., Y. L. Wang, and M. Dembo. 2001. Distinct roles of frontal and rear cell-substrate adhesions in fibroblast migration. Mol. Biol. Cell. 12:3947–3954.
36. Chen, C. S., J. L. Alonso, E. Ostuni, G. M. Whitesides, and D. E. Ingber. 2003. Cell shape provides global control of focal adhesion assembly. Biochem. Biophys. Res. Commun. 307:355–361.[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  N. Tymchenko, J. Wallentin, S. Petronis, L. M. Bjursten, B. Kasemo, and J. Gold A Novel Cell Force Sensor for Quantification of Traction during Cell Spreading and Contact Guidance Biophys. J., July 1, 2007; 93(1): 335 - 345. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. A. Smith, H. Aranda-Espinoza, J. B. Haun, M. Dembo, and D. A. Hammer Neutrophil Traction Stresses are Concentrated in the Uropod during Migration Biophys. J., April 1, 2007; 92(7): L58 - L60. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Sengupta, H. Aranda-Espinoza, L. Smith, P. Janmey, and D. Hammer Spreading of Neutrophils: From Activation to Migration Biophys. J., December 15, 2006; 91(12): 4638 - 4648. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. H. Wu Endothelial focal adhesions and barrier function J. Physiol., December 1, 2005; 569(2): 359 - 366. [Abstract] [Full Text] [PDF]
|
|
|
|
) 0.1 mg/ml, (
) 0.01 mg/ml, and (
) 0.001 mg/ml. Error bars represent mean ± SE of nine cells. (B) The area at half-maximum, as determined from the model regression in A, was plotted against the concentration of ligand on the gel, indicating a linear relationship between the log of the ligand concentration and the area of the cell. (C) The maximum growth rate, as determined from the model regression in A, was plotted against the concentration of ligand on the gel, revealing a log-linear relationship.
