TPA primes α2β1 integrins for cell adhesion

FEBS Letters 582 (2008) 3520–3524

TPA primes a2b1 integrins for cell adhesion Mira Tullaa,*, Jonne Heleniusa, Johanna Jokinenb, Anna Taubenbergera, Daniel J. Mu¨llera, Jyrki Heinob b

a Biotechnology Center, University of Technology Dresden, Germany Department of Biochemistry and Food Chemistry, University of Turku, Finland

Received 11 July 2008; revised 24 August 2008; accepted 7 September 2008 Available online 18 September 2008 Edited by Lukas Huber

Abstract Integrin avidity is regulated by changes in the conformation of the heterodimer and cluster formation. We measured cell adhesion by integrin a2b1 (CHO-a2) to collagen at short contact times (0.5–60 s) by single cell force spectroscopy (SCFS). The adhesion increased rapidly with contact time and was further strengthened by the addition of 12-O-tetradecanoylphorbol-13-acetate (TPA), a protein kinase C (PKC) and integrin activator. TPA also improved the strength of adhesive units. Furthermore, changes in membrane nanotube properties indicated better coupling of integrins to the cell cytoskeleton. We conclude that in addition to increasing integrin avidity TPA strengthens integrin–cytoskeletal linkage.  2008 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. Keywords: Integrin; Inside–out activation; Collagen; Single cell force spectroscopy; Membrane nanotube; Cytoskeleton

1. Introduction Integrins have a central role in coupling the extracellular matrix (ECM) to the cell cytoskeleton [1]. Their avidity for ECM ligands is regulated by a process called inside–out signaling or activation. During activation integrin conformation is modulated from a low-affinity to a high-affinity state [2]. In the low-affinity conformation the integrin heterodimer is bent towards the plasma membrane and the cytoplasmic domains (tails) of a and b subunits are held together by a salt bridge [3–4]. The binding of cytoplasmic proteins to integrin b tails disrupts the salt bridge and triggers straightening of the ectodomain to an upright position [5–6]. In the high-affinity conformation integrin heterodimers are straightened up. Receptor clustering that leads to locally high concentrations of integrins is another mechanism of integrin avidity regulation. Protein kinase C (PKC) mediates the inside–out activation of integrins. However, the details of this activation pathway are not well known. PKC-a has been shown to directly associ-

* Corresponding author. E-mail address: [email protected] (M. Tulla).

Abbreviations: BIS, bisindolylmaleimide; DAG, diacyl glycerol; DMEM, DulbeccoÕs modified EagleÕs medium; ECM, extracellular matrix; PKC, protein kinase C; SCFS, single cell force spectroscopy; TPA, 12-O-tetradecanoylphorbol-13-acetate

ate with the cytoplasmic tail of integrin b1 [7]. In addition, PKC may phosphorylate focal adhesion proteins like talin [8–9] and filamin [10]. These proteins mediate the interactions of integrins with the actin cytoskeleton. PKC can be activated with the phorbol ester, 12-O-tetradecanoyl-phorbol-13-acetate (TPA), which mimics the action of diacyl glycerol (DAG), a natural activator of PKC. PKC activation by TPA increases the avidity of collagen receptor integrin a2b1 for collagen by two mechanisms. First, it induces clustering of the receptors in the absence of ligands, and second, it promotes conformational activation of the integrin heterodimers [11]. The methods commonly used to study cell adhesion require long time periods from tens of minutes to many hours. During this time various signaling cascades get activated and focal adhesions maturate. In order to observe the short-term effects of an activity modulating factor on adhesion formation standard assays can not be used. Here, we measured the effects of TPA on adhesion of integrin a2b1 expressing CHO cells to fibrillar collagen I. The first minute of contact was studied using single cell force spectroscopy (SCFS). Adhesion increased with contact time and by TPA addition. The coupling of plasma membrane to cell cytoskeleton also strengthened with TPA addition. Reinforced plasma membrane–cytoskeleton adhesion may promote cell cytoskeleton dependent cellular processes like cell spreading and migration.

2. Materials and methods 2.1. Cell culture Chinese hamster ovary cells (CHO) were stably transfected with human integrin a2 (CHO-a2) [12]. Cells were cultured in MEM-a medium (Gibco) containing 10% fetal calf serum, 100 IU/ml penicillin and 100 lg/ml streptomycin and supplemented with G-418 (400 lg/ml, Invitrogen). For the SCFS measurements, cells were trypsinized at 37 C for 2–3 min. Trypsin was inactivated with phosphate-buffered saline (PBS) containing soybean trypsin inhibitor (final concentration, 1.3%). The cells were pelleted at 200 · g for 3 min and transferred to serum-free, CO2-independent medium (DulbeccoÕs modified EagleÕs medium [DMEM] containing 20 mM HEPES, pH 7.4) for SCFS measurements. 2.2. Surface functionalization with collagen Structurally well defined collagen type I matrices were prepared as described earlier [13,14]. Shortly, mica disks (Ø 4 mm) were glued on glass coverslips (Ø 24 mm) with an optical adhesive (OP-29; Dymax, Torrington, CT). Buffer containing 50 mM glycine, pH 9.2, 200 mM KCl, and 100 lg/ml collagen type I (PureCol, Inamed, USA) was pipetted on the freshly cleaved mica surfaces and incubated overnight at RT. Before measurements the samples were washed several times with PBS.

0014-5793/$34.00  2008 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2008.09.022

M. Tulla et al. / FEBS Letters 582 (2008) 3520–3524

3521

450

p=0.0331

400 350 300

p=0.0017

250 p=0.03

200

p=0.0431

150 100 50 [37] [53] [221] [216] [22] [47]

[37] [56]

0

2s

0.5 s

5s

30 s

[22] [13]

60 s

Contact time (s) p=0.0353 p=0.0017

240 220 200 180 160 140 120 100 80 60 40 20 0

p=0.0065 p=0.1645

ns.

+T

[221] [216]

[79]

[53]

[44]

C

C H C O H -α O 2 H -α O 2 -α + 2 TP +T C PA A H C O H -α +B O I 2 -α +P S 2 1H +T 5 PA +P 1H 5

T T -W

C H

H O C

[99]

PA

[139]

O

3.1. TPA increases adhesion of a2b1 integrin expressing CHO cells to collagen PKC mediates activation of integrins [16]. In several reports, TPA, a PKC activator, has been shown to increase a2b1 dependent cell adhesion [17] by stimulating the conformational activation of integrins as well as integrin clustering [11]. To examine short-term TPA mediated effects on cell adhesion we used single cell force spectroscopy (SCFS). CHO cells were chosen because although they express several integrins, they do not express collagen receptor integrins endogenously [12]. By transfecting CHO cells with integrin a2 cDNA the specific

p=0.0768

CHO-α2 -TPA CHO-α2 +TPA

-W

3. Results and discussion

500

Max. detachment force (pN)

2.4. Adhesion experiments and data analysis for SCFS The measurements were conducted as described earlier [14,15]. Shortly, plasma-cleaned cantilevers were functionalized with concanavalin A and stored in PBS. Trypsinized cells were pipetted into the BioCell 10–20 min after trypsinization. Where indicated, the cells were treated with 1 lM TPA, (Calbiochem, Merck Biosciences, Germany), integrin a2 function blocking antibody P1H5 5 lg/ml (Santa Cruz Biotechnology, CA, USA) or 1 lM TPA along with 10 lM BIS (Bisindolylmaleimide I, Calbiochem, Merck Biosciences, Germany) for 10 min prior the experiments. Measurements were carried out without further buffer changes. Single cells were captured by pressing with the concanavalin A-functionalized cantilever with a contact force of 400 pN for 3 s. Cell was lifted from the surface and allowed to establish firm adhesion to the cantilever for 5 min. To measure cell adhesion, the cell was brought down to the collagen matrix with a set contact force of 1 nN. During contact (0.5–60 s), the piezo height was kept constant using the AFMÕs closed loop feedback mode. The cantilever was withdrawn at constant speed (2.5 lm s 1) over pulling ranges of 10–70 lm to ensure complete detachment of the cell from the matrix. Usually, ten force–distance curves were acquired for each cell for contact times up to 5 s, 1–5 curves for contact times up to 30 s and one or two force-distance curves for 60 s contact times. The force–distance curves were analyzed to find maximal detachment and small unbinding forces using in-house procedures.

interaction of integrin a2b1 with collagen could be studied. A structurally well characterized fibrillar collagen I matrix was used as the adhesion substrate [13,18]. Single living CHO-a2 cells were attached to a Concanavalin A-functionalized tipless cantilevers and brought to contact with collagen matrices at a force of 1 nN. To measure adhesion strengthening the contact time was varied from 0.5 to 60 s. Thereafter, the cantilever was withdrawn until the cell was fully detached. The force with respect to the cantilever position was recorded and the resulting force–distance curves were analyzed (Fig. 1). The maximal detachment force corresponds to the maximal cantilever deflection during retraction. Two types of small

Max. detachment force (pN)

2.3. AFM setup for SCFS measurements NanoWizard AFM (JPK Instruments, Berlin, Germany) mounted on top of an Axiovert 200 inverted microscope (Carl Zeiss, Jena, Germany) was used. A CellHesion module (JPK Instruments) extended the vertical range of the AFM from 15 lm to 100 lm. SCFS measurements were conducted using tipless V-shaped silicon nitride cantilevers having nominal spring constants of 0.06 N/m (NP-0; Veeco Instruments, Woodbury, NY). Before each experiment the sensitivity of the optical lever system was calibrated and the cantilever spring constant determined with the inbuilt thermal noise method. Experiments were performed at 37 C using a temperature-controlled BioCell sample chamber (JPK Instruments).

Membrane nanotubes Jumps

100 pN

Maximal detachment force

1 μm

Fig. 1. Analysis of the force-distance curves recorded by SCFS. A representative force–distance curve for a CHO-a2 cell recorded after a 5 s contact time at a pulling speed of 2.5 lm s 1. The force–distance curves were analyzed for three types of deadhesion events; the maximal detachment, the single adhesive units (denoted as jumps) and membrane nanotubes.

Fig. 2. TPA strengthens the adhesion of CHO-a2 cells to collagen. (A) The mean maximal detachment forces for CHO-a2 cells ± TPA (1 lM) after 0.5–60 s of contact with collagen. (B) The mean maximal detachment forces for CHO-WT and CHO-a2 cells (5 s contact time) in the presence of TPA (1 lM), TPA (1 lM) and BIS (10 lM), and a2 integrin blocking antibody, P1H5 (5 lg/ml). In all figures where indicated, the cells were incubated with TPA, TPA and BIS or P1H5 for 10 min prior the measurements. In all figures error bars denote the standard error of the mean and the number inserted [n] within brackets denote the number of force–distance curves analyzed. The results of a Mann–Whitney or t-test are shown with brackets above the bars in all figures (ns, no significance P > 0.05).

3522

3.2. TPA improves the strength of single adhesive units in CHOa2 cells When analyzing the small unbinding events of the CHO-a2 force–distance curves (see Fig. 1), TPA was observed to increase the rupture force of the single adhesive units (Fig. 3A). The force of the jumps in the CHO-a2 cells (33.4 ± 5.4 pN, n = 16 cells, 171 force–distance curves, means ± S.D.) increased by approximately 30% with TPA induced integrin activation (43.3 ± 12.7 pN, n = 16 cells, 181 force–distance curves, means ± S.D.) (Fig. 3B). The TPA induced increase in the force jump size was indicative of either integrin clustering and consequent lateral adhesion strengthening and/or PKC induced reinforcement of integrin–cytoskeleton linkages and thus improved force resistance of the adhesion. In CHO-WT cells this type of unbinding event was extremely rare and thus impossible to analyze. This emphasizes the role of a2b1 integrin in the adhesion of CHO-a2 cells to collagen. 3.3. TPA increases the extraction forces of membrane nanotubes in CHO-a2 cells When a cell is detached from a surface membrane nanotubes (also called as tethers) are extracted. Nanotubes form when the lipid bilayer is pulled away from the cortical cytoskeleton in thin tube-like protrusions. In cell detachment force–distance curves membrane nanotubes are characterized by force plateaus that end with the detachment of the nanotube from the substrate (see Fig. 1). Here, membrane nanotube extraction

A

CHO-α2 [171] CHO-α2 +TPA [181] fit CHO-α2 [LogNormal] fit CHO-α2 +TPA [LogNormal]

Relative counts (%)

25 20 15 10 5 0 20

40

60

80

100

120

140

160

Unbinding force (pN)

B

Mean unbinding force (pN)

unbinding events were perceived. The first type was characterized by force loading prior to rupture (also called as jump or force jump). This was interpreted as a rupture of an adhesive unit that was coupled to the cell cytoskeleton. No force loading preceded the second type of small unbinding event. These were interpreted as the unbinding of membrane nanotubes. To trigger PKC mediated activation of integrins, cells were incubated with TPA for 10 min prior SCFS measurements. TPA treatment increased the maximal detachment forces of CHO-a2 cells at every time point from 0.5 to 60 s (Fig. 2A). TPA did not increase the adhesion of non-transfected (CHOWT) cells to collagen (Fig. 2B) confirming that the observed increased adhesion is integrin a2 dependent. The TPA induced increase in CHO-a2 cell adhesion was inhibited by bisindolylmaleimide (BIS), a PKC inhibitor, demonstrating that the increased adhesion was PKC dependent. CHO-a2 adhesion was decreased to that of CHO-WT with the addition of integrin a2 function blocking antibody, P1H5 (Fig. 2B). In addition, we performed conventional cell adhesion washing assays using CHO-a2 cells at both +37 C and +4 C. Adhesion to collagen I at +4 C clearly increased with TPA addition, in agreement with TPA induced inside–out activation of integrins (Supplementary figure). Concurringly, TPA induced adhesion was inhibited by PKC inhibitor BIS. As cell adhesion measurements by SCFS were conducted after a TPA incubation time of only 10 min, TPA induced changes in gene expression were unlikely to have manifested. Indeed, two separate studies show that the cell surface expression of integrin a2b1 does not change within the first hour of TPA addition [11,17]. Therefore, we conclude that the observed TPA effects are by the direct and fast activation of the kinase function of PKC.

M. Tulla et al. / FEBS Letters 582 (2008) 3520–3524

50 45 40 35 30 25 20 15 10 5 0

p=0.0095

-TPA +TPA

Fig. 3. Preactivation of CHO-a2 cells with TPA increases the strength of the single adhesive units. (A) The force distribution of the single force jumps for CHO-a2 (16 cells) and CHO-a2 + TPA (1 lM) (16 cells). (B) The mean force jump size for CHO-a2 cells (±TPA). Force– distance curves recorded after 5 s of contact were analyzed for this and the following figures.

forces were analyzed from the force–distance curves of CHO-WT and CHO-a2 cells with and without TPA treatment. For CHO-WT cells the force distributions of the nanotube extraction did not change with TPA treatment ( Fig. 4A). The average nanotube extraction force for CHO-WT cells was 27.2 ± 3.8 pN (n = 17 cells, 239 force–distance curves, means ± S.D.) and with TPA 30.5 ± 9.2 pN (n = 15 cells, 197 force–distance curves, means ± S.D.) (Fig. 4C). In good agreement with our results is an earlier report showing that the average force needed to pull a membrane nanotube from CHO cell was 30 pN (with 3 lm s 1 pulling speed) [19]. The extraction forces of membrane nanotubes in TPA treated CHO-a2 cells increased remarkably and had a wider distribution ( Fig. 4B). The average extraction force for CHO-a2 cells increased by 39% (means ± S.D. for CHO-a2: 30.1 ± 4.4 pN, n = 14 cells, 130 force–distance curves, for CHO-a2 with TPA: 41.8 ± 15.4 pN, n = 15 cells, 126 force–distance curves) (Fig. 4C). It is well established that the membrane nanotube extraction force mainly depends on plasma membrane–cytoskeletal coupling [20–22]. Therefore, we conclude that TPA induced integrin activation reinforces membrane–cytoskeletal connection in CHO-a2 cells. An equivalent increase in the extraction force was not detected for CHO-WT cells lacking this collagen receptor. 3.4. The number of membrane nanotubes is decreased due to TPA treatment With the strengthening of the integrin–cytoskeletal linkage a decrease in the number of nanotubes is expected. Therefore, the number of membrane nanotubes within each force curve

M. Tulla et al. / FEBS Letters 582 (2008) 3520–3524

25 20 15 10 5 0 20

40

60

80

100

Rupture force (pN)

Relative counts (%)

CHO-α2 [130] CHO-α2 +TPA [126] fit CHO-α2 [LogNormal] fit CHO-α2 +TPA [LogNormal]

25 20 15 10 5 0 20

40

60

80

100

Mean unbinding force (pN)

Membrane nanotube extraction force (pN)

40

0.20

-TPA +TPA

p=0.1669

p=0.0011

0.15

0.10

0.05

30

50

0.25

CHO-WT [239] CHO-WT +TPA [197] fit CHO-WT [LogNormal] fit CHO-WT +TPA [LogNormal]

No. of nanotubes/nN

Relative counts (%)

30

3523

-TPA +TPA

p=0.0128

ns.

30 20 10 0 CHO-WT

CHO-α2

Fig. 4. Preactivation with TPA increases membrane nanotube extraction forces of CHO-a2. (A) Distribution of single nanotube unbinding events for CHO-WT (17 cells) or CHO-WT + TPA (15 cells) and for (B) CHO-a2 (14 cells) CHO-a2 + TPA (15 cells). (C) Mean membrane nanotube extraction force for CHO-WT and CHO-a2 (±TPA) cells.

was determined. Deadhesion events having the characteristic force plateau before unbinding at distances between 2 lm and 9.5 lm were analyzed. The number of nanotubes per force–distance curve after 5 s contact time varied between zero and nine. To obtain comparable numbers for the different cell lines the mean number of nanotubes was divided by the mean detachment force for each cell line. TPA incubation decreased the frequency of nanotubes by 30% for CHO-a2 and 25% for CHO-WT cells (Fig. 5). The decrease in the frequency of nanotubes together with the increase in membrane nanotube extraction force indicated enhanced adhesion of the plasma membrane to underlying intracellular structures. CHO-WT cells, although devoid of collagen receptors, contain other types of TPA responsive integrins. Due to the use of collagen matrix the effect of TPA was not manifested as increased cell adhesion of CHO-WT cells. However, the decreased number of nanotubes in CHO-WT, like in CHO-a2, indicates enhanced plasma membrane cytoskeleton adhesion by the same mechanism.

0.00

CHO-α2

CHO-WT

Fig. 5. TPA enhances plasma membrane–cytoskeleton coupling. The number of membrane nanotubes at distances 2–9.5 lm for CHO-a2 and CHO-WT cells (±TPA) was calculated for each force–distance curve. To allow comparison between cells the number of membrane nanotubes/nN was calculated using the mean number of nanotubes for force–distance curves and the mean maximal detachment force of the cells.

Integrin associated proteins are often multifunctional. Therefore, activation mediating proteins may also link integrins to the cytoskeleton. The interaction of talin with the cytoplasmic domain of integrin b subunit is understood at the structural level. Association leads to separation of the cytoplasmic tails of a and b subunits and, thereby, to conformational activation [6]. Recent publications indicated that filamins and kindlins are important for integrin regulation [23,24] and showed that the filamin binding site on the integrin b subunit overlaps with that of talin [23]. Thus, filamin may regulate talin binding [23]. Talin, filamin, kindlins and other proteins may regulate the conformational activation of integrins. Our measurements indicate that in concert with the activation of integrin extracellular domains, integrin intracellular domains form connections to cytoskeleton. Binding interactions between plasma membrane and cytoskeleton regulate cellular functions like spreading, migration, formation of cell protrusions and endo- and exocytosis. Our results suggest that the TPA dependent increase in cell adhesion and spreading is not solely due to integrin conformational activation and clustering. In addition, TPA seems to prime integrins for adhesion intracellularly by improving their connection to the cell cytoskeleton. This enhanced mechanical coupling may assist and accelerate cellular processes that are dependent on the dynamic function of the actin cytoskeleton, such as cell spreading and migration. Acknowledgments: The authors want to thank the expert technical assistance of Isabel Richter. Clemens Franz is acknowledged for the fruitful discussions.

Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.febslet.2008.09. 022.

3524

M. Tulla et al. / FEBS Letters 582 (2008) 3520–3524

References [1] Hynes, R.O. (2002) Integrins: bidirectional, allosteric signaling machines. Cell 110, 673–687. [2] Luo, B.H., Carman, C.V. and Springer, T.A. (2007) Structural basis of integrin regulation and signaling. Annu. Rev. Immunol. 25, 619–647. [3] Xiong, J.P., Stehle, T., Zhang, R., Joachimiak, A., Frech, M., Goodman, S.L. and Arnaout, M.A. (2002) Crystal structure of the extracellular segment of integrin aVb3 in complex with an Arg-Gly-Asp ligand. Science 296, 151–155. [4] Vinogradova, O., Velyvis, A., Velyviene, A., Hu, B., Haas, T., Plow, E. and Qin, J. (2002) A structural mechanism of integrin aIIbb3 ‘‘inside–out’’ activation as regulated by its cytoplasmic face. Cell 110, 587–597. [5] Tadokoro, S., Shattil, S.J., Eto, K., Tai, V., Liddington, R.C., de Pereda, J.M., Ginsberg, M.H. and Calderwood, D.A. (2003) Talin binding to integrin b tails: a final common step in integrin activation. Science 302, 103–106. [6] Wegener, K.L., Partridge, A.W., Han, J., Pickford, A.R., Liddington, R.C., Ginsberg, M.H. and Campbell, I.D. (2007) Structural basis of integrin activation by talin. Cell 128, 171–182. [7] Ng, T., Shima, D., Squire, A., Bastiaens, P.I., Gschmeissner, S., Humphries, M.J. and Parker, P.J. (1999) PKCa regulates b1 integrin-dependent cell motility through association and control of integrin traffic. EMBO J. 18, 3909–3923. [8] Litchfield, D.W. and Ball, E.H. (1986) Phosphorylation of the cytoskeletal protein talin by protein kinase C. Biochem. Biophys. Res. Commun. 134, 1276–1283. [9] Hyatt, S.L., Liao, L., Chapline, C. and Jaken, S. (1994) Identification and characterization of alpha-protein kinase C binding proteins in normal and transformed REF52 cells. Biochemistry 33, 1223–1228. [10] Tigges, U., Koch, B., Wissing, J., Jockusch, B.M. and Ziegler, W.H. (2003) The F-actin cross-linking and focal adhesion protein filamin A is a ligand and in vivo substrate for protein kinase C alpha. J. Biol. Chem. 278, 23561–23569. [11] Connors, W.L., Jokinen, J., White, D.J., Puranen, J.S., Kankaanpa¨a¨, P., Upla, P., Tulla, M., Johnson, M.S. and Heino, J. (2007) Two synergistic activation mechanisms of a2b1 integrin-mediated collagen binding. J. Biol. Chem. 282, 14675–14683. [12] Nykvist, P., Tu, H., Ivaska, J., Ka¨pyla¨, J., Pihlajaniemi, T. and Heino, J. (2000) Distinct recognition of collagen subtypes by a1b1

[13]

[14] [15]

[16] [17]

[18] [19]

[20] [21] [22] [23]

[24]

and a2b1 integrins. a1b1 mediates cell adhesion to type XIII collagen. J. Biol. Chem. 275, 8255–8261. Cisneros, D.A., Friedrichs, J., Taubenberger, A., Franz, C.M. and Muller, D.J. (2007) Creating ultrathin nanoscopic collagen matrices for biological and biotechnological applications. Small 3, 956–963. Franz, C.M., Taubenberger, A., Puech, P.H. and Muller, D.J. (2007) Studying integrin-mediated cell adhesion at the singlemolecule level using AFM force spectroscopy. Sci. STKE 2007, l5. Taubenberger, A., Cisneros, D.A., Friedrichs, J., Puech, P.H., Muller, D.J. and Franz, C.M. (2007) Revealing early steps of a2b1 integrin-mediated adhesion to collagen type I by using single-cell force spectroscopy. Mol. Biol. Cell 18, 1634–1644. Kolanus, W. and Seed, B. (1997) Integrins and inside–out signal transduction: converging signals from PKC and PIP3. Curr. Opin. Cell Biol. 9, 725–731. Rosfjord, E.C., Maemura, M., Johnson, M.D., Torri, J.A., Akiyama, S.K., Woods Jr, V.L. and Dickson, R.B. (1999) Activation of protein kinase C by phorbol esters modulates a2b1 integrin on MCF-7 breast cancer cells. Exp. Cell Res. 248, 260–271. Jiang, F., Khairy, K., Poole, K., Howard, J. and Muller, D.J. (2004) Creating nanoscopic collagen matrices using atomic force microscopy. Microsc. Res. Tech. 64, 435–440. Sun, M., Graham, J.S., Hegedus, B., Marga, F., Zhang, Y., Forgacs, G. and Grandbois, M. (2005) Multiple membrane tethers probed by atomic force microscopy. Biophys. J. 89, 4320–4329. Dai, J. and Sheetz, M.P. (1999) Membrane tether formation from blebbing cells. Biophys. J. 77, 3363–3370. Titushkin, I. and Cho, M. (2006) Distinct membrane mechanical properties of human mesenchymal stem cells determined using laser optical tweezers. Biophys. J. 90, 2582–2591. Afrin, R. and Ikai, A. (2006) Force profiles of protein pulling with or without cytoskeletal links studied by AFM. Biochem. Biophys. Res. Commun. 348, 238–244. Kiema, T., Lad, Y., Jiang, P., Oxley, C.L., Baldassarre, M., Wegener, K.L., Campbell, I.D., Yla¨nne, J. and Calderwood, D.A. (2006) The molecular basis of filamin binding to integrins and competition with talin. Mol. Cell 21, 337–347. Moser, M., Nieswandt, B., Ussar, S., Pozgajova, M. and Fa¨ssler, R. (2008) Kindlin-3 is essential for integrin activation and platelet aggregation. Nat. Med. 14, 325–330.