Gerhard Isenberg

Talin binds to actin and promotes filament nucleation

Platelet talin binds to actin in vitro and hence is an actin binding protein. By four different non‐interfering assay conditions (fluorescence, fluorescence recovery after photobleaching, (FRAP), dynamic light scattering and DNase‐I inhibition) we show that talin promotes filament nucleation, raises the filament number concentration and increases the net rate of actin polymerization but has no inhibitory effect on filament elongation. Binding of talin to actin occurs at a maximal molar ratio of 1:3 as determined by fluorescencetitration under G‐buffer conditions. The overall binding constant was ≈ 0.25 μM.

Talin anchors and nucleates actin filaments at lipid membranes. A direct demonstration.

Platelet talin nucleates actin assembly as we show here directly by using rhodamine—phalloidin labelling of actin filaments, Nucleation by talin still occurs after reconstitution into liposomal bilayers. This is also demonstrated directly after protein–lipid double labelling and light microscopic imaging. Talin, thus, is the first actin binding protein for which anchoring and nucleation of artin filament growth at lipid interfaces have been visualized.

Interaction of the 47-kDa talin fragment and the 32-kDa vinculin fragment with acidic phospholipids: a computer analysis

In recent in vitro experiments, it has been demonstrated that the 47-kDa fragment of the talin molecule and the 32-kDa fragment of the vinculin molecule interact with acidic phospholipids. By using a computer analysis method, we determined the hydrophobic and amphipathic stretches of these fragments and, by applying a purpose-written matrix method, we ascertained the molecular amphipathic structure of alpha-helices. Calculations for the 47-kDa mouse talin fragment (residues 1-433; NH2-terminal region) suggest specific interactions of residues 21-39, 287-342, and 385-406 with acidic phospholipids and a general lipid-binding domain for mouse talin (primary amino acid sequence 385-401) and for Dictyostelium talin (primary amino acid sequence 348-364). Calculations for the 32-kDa chicken embryo vinculin fragment (residues 858-1066; COOH-terminal region) and from nematode vinculin alignment indicate for chicken embryo vinculin residues 935-978 and 1020-1040 interactions with acidic phospholipids. Experimental confirmation has been given for vinculin (residues 916-970), and future detailed experimental analyses are now needed to support the remaining computational data.

Analysis of filamin and α-actinin binding to actin by the stopped flow method

We ascertained by the stopped flow method the overall association rate constant, k +1, of filamin and α‐actinin to fluorescently labelled filamentous actin of ~ 1.3 × 106M−1 · s−1 and ~ 1.0 × 106M−1 · s−1 as well as the overall dissociation rate constant,k −1 of ~ 0.6s−1 and ~ 0.4s−1, respectively. The overall equilibrium constant, K, for filamin and α‐actinin to actin deduced from the relation K = agree well with published data.

Structural aspects of vinculin-actin interactions.

It has been shown (Jockusch & Isenberg, 1981) that vinculin (130K protein) binds to actin and induces actin filaments to form bundles even at low ionic strength. Here, we present structural details on the vinculin molecule itself and on its interaction with actin. In negatively stained preparations, vinculin appeared as a globular protein with an average diameter of 85 A. The ability of vinculin to form actin filament bundles was confirmed using shadowing techniques and gel analysis of sedimented material. Analysis of vinculin-induced paracrystals by optical diffraction and computer processing revealed their structural similarity to Mg-induced paracrystals. The lateral position of vinculin on surface-exposed actin filaments of such paracrystals was demonstrated directly in electron micrographs and indirectly by labelling vinculin with ferritin-coupled anti-vinculin F(ab′) fragments. Polymerization of actin in the presence of vinculin-coated polystyrene beads did not result in an “end-on” binding of filaments to the beads. Rather, actin bundles were laterally associated with the whole surface of the beads, from where they radiated in a star-like pattern. The growth of actin filaments onto myosin subfragment-I decorated, vinculin-incubated. fixed filament fragments was not inhibited, as was shown directly by electron microscopy and monitored viscometrically in a nucleation assay. These results suggest that in vivo at the site of an adhesion plaque vinculin may link actin filaments together into a suitable configuration to interact with the plasma membrane.

Peptide‐specific antibodies localize the major lipid binding sites of talin dimers to oppositely arranged N‐terminal 47 kDa subdomains

Using ultrastructural analysis and labeling with polyclonal antibodies that recognize peptide sequences specific for phospholipid binding, we mapped the functional domain structure of intact platelet talin and its proteolytic fragments. The talin dimer, which is crucial for actin and lipid binding, is built of a backbone containing the 200 kDa rod portions, at both ends of which a 47 kDa globular domain is attached. Peptide‐specific polyclonal antibodies were raised against three potential lipid binding sequences residing within the N‐terminal 47 kDa domain (i.e. S19, amino acids 21–39; H18, amino acids 287–304; and H17, amino acids 385–406). Antibodies H17 and H18 localize these lipid binding sequences within the N‐terminal 47 kDa globular talin subdomains opposed at the outer 200 kDa rod domains within talin dimers. Hence, we conclude that in its dimeric form, which is used in actin and lipid binding, talin is a dumbbell‐shaped molecule built of two antiparallel subunits.

Vinculin, talin and focal adhesions

One of the most complicated connections between actin and the plasma membrane is the focal adhesion, a complex of proteins and lipids that forms at sites where cells attach to the extracellular matrix. A major protein component of focal adhesions is vinculin. As with other components of the focal adhesion complex, vinculin illustrates the apparent redundancy of proteins that mediate the connection of actin to the plasma membrane. During the past 5 years there have been several studies examining the role of vinculin in cell function. The purpose of this article is to discuss these findings and present an integrated model of vinculin’s role in the cell. Vinculin associates with talin and alpha-actinin via its N-terminal region (Burridge & Mangeat, 1984; Wachsstock et al., 1987). It also self-associates to form head-to-tail dimers (Molony & Burridge, 1985; Johnson & Craig, 1994). There is further convincing evidence that vinculin contains an actin-binding domain (Menkel et al., 1994; Johnson & Craig, 1995). Vinculin is also a ligand for the focal adhesion complex protein paxillin (Turner et al, 1990). Furthermore, vinculin binds phospholipid bilayers non-covalently with an apparent two-step mechanism involving both electrostatic interactions with acidic head groups and insertion into the hydrophobic domain of lipid bilayers (Niggli & Burger, 1987). It has also been shown in in vitro lipid photolabelling studies that vinculin as well as talin directly inserts into the hydrophobic region of lipid bilayers (Goldmann et al., 1992; Niggli et al., 1994). In addition, vinculin interacts with phosphatidylinositol-4,5-bisphosphate (Fukami et al, 1994). Therefore, vinculin binds the lipid bilayer directly through one of several mechanisms and also binds actin directly.

Kinetic determination of talin-actin binding.

Smooth muscle talin prepared from chicken gizzard binds to skeletal muscle actin in vitro. The stoichiometry of 1:3 for talin:fluorescent labelled G-actin was confirmed by steady state titration and viscosity measurements under non-polymerizing conditions. The binding constant (Kd) of talin and G-actin was determined by continuous fluorescence titration and gave a value of approx 0.3 microM. The association rate constant of talin and fluorescent labelled G-actin of approx 7 x 10(6) M-1 x s-1 was ascertained by the stopped flow method; the dissociation rate constant was calculated at approx 2-3 s-1.

Interaction of NBD-talin with lipid monolayers. A film balance study.

Fluorescently labelled smooth muscle talin like native talin interacts with negatively or partly negatively charged lipid monolayers. This was measured in time/area diagrams using the film balance technique combined with fluorescence imaging after double photolabelling of talin and phospholipids.

Determination of the affinity of talin and vinculin to charged lipid vesicles: a light scatter study

Recent experimental findings have demonstrated that both talin and vinculin bind to phospholipids and insert into the hydrophobic region of lipid membranes. Here, we show that the light scatter method can be used for measuring the affinity of proteins to phospholipid membranes. Large unilamellar DMPC/DMPG vesicles were produced by the extrusion technique (LUVETs). We have used repeated heating/cooling scans between 15°C and 35°C to ensure protein‐lipid interaction/insertion. A molar affinity of talin, K = 2.9 × 106M−1 and of vinculinK = 3.3 × 105M−1 to lipid vesicles, respectively, was determined from the plot; light scatter signal at 380 nm against protein concentrations by fitting the term, In(Io/I − 1) = A − K × c to the data.