Structural and Functional Analysis of the Actin Binding Domain of Plectin Suggests Alternative Mechanisms for Binding to F-Actin and Integrin β4

Structure, Vol. 11, 615–625, June, 2003, 2003 Elsevier Science Ltd. All rights reserved.

DOI 10.1016/S0969-2126(03)00090-X

Structural and Functional Analysis of the Actin Binding Domain of Plectin Suggests Alternative Mechanisms for Binding to F-Actin and Integrin ␤4 Begon˜a Garcı´a-Alvarez,1,3 Andrey Bobkov,1,3 Arnoud Sonnenberg,2 and Jose´ M. de Pereda*,1 1 Program on Cell Adhesion The Burnham Institute 10901 North Torrey Pines Road La Jolla, California 92037 2 The Netherlands Cancer Institute Division of Cell Biology Plesmanlaan 121 1066 CX Amsterdam The Netherlands

Summary Plectin is a widely expressed cytoskeletal linker. Here we report the crystal structure of the actin binding domain of plectin and show that this region is sufficient for interaction with F-actin or the cytoplasmic region of integrin ␣6␤4. The structure is formed by two calponin homology domains arranged in a closed conformation. We show that binding to F-actin induces a conformational change in plectin that is inhibited by an engineered interdomain disulfide bridge. A twostep induced fit mechanism involving binding and subsequent domain rearrangement is proposed. In contrast, interaction with integrin ␣6␤4 occurs in a closed conformation. Competitive binding of plectin to F-actin and integrin ␣6␤4 may rely on the observed alternative binding mechanisms and involve both allosteric and steric factors. Introduction Microtubules, actin microfilaments, and intermediate filaments constitute the cytoskeleton of eukaryotic cells. The integrated organization of the three systems is essential for cellular functions, including mechanical strength, adhesion, motility, intracellular trafficking, and cell division. Network interconnection relies on linker proteins that interact with multiple filament systems (Fuchs and Yang, 1999). Plectin is a large (ⵑ500 kDa) and versatile cytoskeletal linker (Steinbock and Wiche, 1999) and is a member of the plakin protein family (Leung et al., 2002; Ruhrberg and Watt, 1997). Plectin has been shown to interact with several types of intermediate filament proteins (Foisner et al., 1991, 1988; Pytela and Wiche, 1980), actin filaments (Seifert et al., 1992), and microtubules (Herrmann and Wiche, 1987). In addition to binding individual cytoskeletal components, plectin crosslinks intermediate filaments with microtubules and actin microfilaments (Svitkina et al., 1996). In epithelial cells, plectin is localized in desmosomes (Eger et al., 1997) and hemidesmosomes (Hieda et al., 1992), two structures that anchor intermediate filaments to the plasma membrane. In *Correspondence: [email protected] 3 These authors contributed equally to the work.

hemidesmosomes, plectin binds directly to the cytoplasmic moiety of integrin ␣6␤4 (Geerts et al., 1999; Niessen et al., 1997b; Rezniczek et al., 1998). Thus, plectin provides a direct link between the integrin and the intermediate filament cytoskeleton. Mutations in the plectin gene are present in patients suffering from a skin-blistering disorder termed epidermolysis bullosa simplex, which is frequently associated with a late-onset muscular dystrophy (Uitto et al., 1996). A similar phenotype is observed in mice upon targeted inactivation of the plectin gene (Andra et al., 1997). Hence, plectin plays an essential role to maintain mechanical strength of both skin and muscle. Purified plectin forms dimers and exhibits a dumbbell structure ⵑ200 nm in length, as observed by electron microscopy (Foisner and Wiche, 1987). Analysis of the plectin sequence predicts a multidomain structure with three main sections: a central rod domain, flanked by N- and C-terminal globular regions (Figure 1A). The rod domain is predicted to be mainly ␣-helical and to drive dimerization by coiled-coil interactions. The C-terminal globular region contains six sequence repeats. Repeats one to five show sequence similarity with plakin repeat domain B, while the C-terminal repeat corresponds to a plakin repeat domain C. The crystal structures of plakin repeat domains B and C of desmoplakin reveal a common and novel fold comprising four and a half copies of a 38-residue minimal structural scaffold termed “plakin repeat” (Choi et al., 2002). The C-terminal region of plectin harbors a specific binding site for vimentin and cytokeratins (Nikolic et al., 1996; Steinbock et al., 2000), located next to the fifth plakin repeat domain B. Plectin association with intermediate filaments is regulated by a mitosis-specific phosphorylation of Thr4542 (Foisner et al., 1996; Malecz et al., 1996), located in the ␤-hairpin of the fourth plakin repeat of the C-terminal plakin repeat domain C. The N-terminal globular moiety contains a 220-residue region with sequence similarity to the F-actin binding domain (ABD) found in dystrophin, ␣-actinin, spectrins, fimbrin, and other plakins (de Arruda et al., 1990; Hemmings et al., 1992). N-terminal fragments of plectin containing the ABD interact with actin (Andra et al., 1998; Fontao et al., 2001) and regulate actin dynamics in vivo (Andra et al., 1998). Thus, plectin ABD is functional. In addition, fragments carrying the ABD bind to the cytoplasmic moiety of the ␤4 subunit of integrin ␣6␤4 (Geerts et al., 1999; Rezniczek et al., 1998), and binding to integrin ␤4 prevents association with actin filaments (Geerts et al., 1999). This type of ABD is formed by a tandem pair of calponin homology (CH) domains (Carugo et al., 1997; Stradal et al., 1998) separated by a linker of variable length. The high-resolution structures of single (Banuelos et al., 1998; Bramham et al., 2002; Carugo et al., 1997; Keep et al., 1999a) and tandem pairs (Goldsmith et al., 1997; Keep et al., 1999b; Norwood et al., 2000) of CH domains reveal a conserved fold built up of four ␣ helices. The two CH domains of fimbrin’s N-terminal ABD make an

Structure 616

Figure 1. Structure of the ABD of Plectin (A) Domain organization of full-length plectin. (B) Ribbon representation of the ABD structure composed by CH1 (blue) and CH2 (red). (C) Stereo C␣ trace of the ABD in the same orientation as in (B). The sequence is numbered every 10 residues. The exon 2␣ insertion is colored in green. (D) Structure-based sequence alignment of human ABDs from plectin (Plec), utrophin (Utro), dystrophin (Dyst), the N-terminal ABD of fimbrin (Fimb), and the second CH domain of ␤-spectrin (Spec). Secondary structure elements are indicated by boxes and are labeled as in (B); the position of every tenth residue is indicated by dots. The positions of ABS1-3 and the exon 2␣ insertion are indicated by boxes. Utrophin and fimbrin residues previously implicated in direct interaction with F-actin are indicated by green boxes. Molecular figures were generated using the programs MOLSCRIPT (Kraulis, 1991), RASTER3D (Merrit and Bacon, 1997), or SPOCK (Christopher, 1998).

extensive interdomain contact; hence, the ABD adopts a closed or compact conformation. The crystal structures of the ABD of utrophin (Keep et al., 1999b) and its close homolog dystrophin (Norwood et al., 2000) reveal an antiparallel dimeric arrangement. In these dimers the N-terminal CH domains (CH1) and the C-terminal CH domains (CH2) from different molecules make contacts similar to those observed in fimbrin ABD. The crystallographic dimer formation may be due to domain swapping (Liu and Eisenberg, 2002) and suggests that ABDs can undergo a conformational switch between open and closed states. The two CH domains of plectin are functionally nonequivalent, as observed in other ABDs. Isolated CH1 domains support binding to actin; in contrast, CH2 do-

mains show very weak or negligible affinity for actin (Fontao et al., 2001; Way et al., 1992; Winder et al., 1995). However, full actin binding capacity is only observed in the presence of both CH domains, supporting the role of tandem CH domains as a functional unit. Within ABDs, three conserved actin binding sites (ABS1-3) (Figure 1D) have been implicated in direct interaction with actin (Bresnick et al., 1990; Levine et al., 1992). ABS1 and ABS2 are in CH1 while ABS3 extends over the N-terminal ␣ helix of CH2. In order to understand the mechanisms of plectin interaction with F-actin and with the cytoplasmic tail of integrin ␣6␤4, we have determined the crystal structure of the plectin ABD. Using biochemical and biophysical methods in combination with structure based mutagen-

Plectin Actin Binding Domain Structure 617

Table 1. Summary of Crystallographic Analysis Data Collection Space group P212121 Wavelength (A˚) Resolution (A˚) Measured reflections Unique reflections Completeness (%) Rsym (%) ⬍I/␴I⬎

a ⫽ 33.6 A˚ b ⫽ 79.3 A˚ c ⫽ 82.4 A˚ 1.5418 32.0–2.15 (2.23–2.15)a 42883 11486 91.2 (61.4)a 8.2 (31.9)a 16.1 (3.9)a

Refinement Statistics Resolution range (A˚) Unique reflections work/free Rwork (%) Rfree (%) Number of residues Number of solvent molecules Average B value (A˚2) Protein Solvent Rmsd bond lengths (A˚) Rmsd angles (⬚)

32.0–2.15 10591/862 21.2 26.1 243 153 24.2 31.4 0.007 1.30

a

Numbers in parenthesis correspond to the outer shell. Rsym ⫽ ⌺|Ii ⫺ ⬍I⬎|⌺Ii, where I is the observed intensity and ⬍I⬎ is the average intensity of multiple observations of symmetry-related reflections. R ⫽ ⌺|Fobs ⫺ Fcalc|/⌺Fobs. Rfree is calculated using 7.5% of reflections that were not included in the refinement.

esis, we demonstrate that binding to actin induces a conformational change in plectin ABD. We show evidence for alternative mechanisms of interaction with actin and ␤4. These observations are discussed in the context of a possible allosteric competition. Results and Discussion Overall Structure We have crystallized a fragment of the N-terminal region of human plectin, residues 59–293, corresponding to the ABD. The structure was solved by molecular replacement and refined to 2.15 A˚ resolution (Table 1). The crystal contains a single molecule in the asymmetric unit, and interpretable electron density was observed for all residues. The ABD is formed by two CH domains (Figures 1B– 1D). Each domain adopts the common fold observed in other members of the CH domain superfamily; they are built around a conserved nucleus formed by four ␣ helices A, C, E, and G that enclose a hydrophobic core. The 310 helix B and short ␣ helix F cap the N-terminal end of the parallel helical bundle formed by helices C, E, and G. The CH2 has an additional short ␣ helix D in the CE loop not present in CH1. As observed in other tandem pairs of CH domains, each CH domain of plectin is structurally more similar to the equivalent CH domains from other proteins than to the companion CH domain. Superposition of other CH domain structures reveal that plectin’s CH1 is more similar to the CH1 domain of utrophin and dystrophin, with a root mean square deviation (rmsd) of 0.43–0.50 A˚ for 47 equivalent C␣ atoms used in the superpositions.

In contrast, the rmsd for the same set of atoms in the CH1 domain of fimbrin is 1.10 A˚. The nearest structural homolog to the CH2 domain of plectin is the CH2 domain of ␤-spectrin (rmsd 0.46 A˚), and it is closely related to the structure of the CH2 domains of utrophin and dystrophin (rmsd 0.61 A˚), while structurally more distant from the fimbrin CH2 domain (rmsd 1.06 A˚). Both CH domains of plectin show the lowest structural similarity with the CH domain of calponin as indicated by rmsd of 1.22 A˚ and 1.19 A˚ for the first and second CH domains, respectively. Thus, plectin CH domains belong to the spectrin branch of the CH superfamily (Bramham et al., 2002; Stradal et al., 1998). Despite the high degree of structural conservation of the CH fold, the plectin CH1 domain shows some unique features. The N-terminal helix A is longer than in other CH structures (38 A˚ in length), with four extra turns at its N terminus, and projects out of the core of the domain. Our construct includes a five-residue insertion, HWRAE, contiguous to helix A, which is coded by exon 2␣ and subject to alternative splicing (Fuchs et al., 1999). The side chain of the His makes a H-bond with Glu108 in helix C and is buried by the aromatic ring of the Trp. The last two residues of the insertion and Ala86 form a 310 helix, not present in other CH1 domains. The backbone of helix A and the AC loop downstream of the insertion adopt a similar conformation to the equivalent region of dystrophin and utrophin. Thus, the 2␣ insertion has a limited local effect and does not propagate structural changes to other regions of the ABD. The approximate overall dimensions of the ABD are 67 ⫻ 40 ⫻ 35 A˚, not including the N terminus of helix A from CH1. The CH domains are arranged in a closed conformation forming extensive intramolecular contacts that bury 1550 A˚2 of solvent accessible surface. Helix G of CH2 packs against the cleft formed between helices A and G from CH1. Additional contacts in CH2 involve residues from the EF loop, the short helix F, and the N terminus of helix A. The nature of the interface is mixed with both hydrophobic and polar interactions. Polar contacts include six H-bonds and two salt bridges formed by Lys73 with Asp264 and Asp267. The side chain of Lys276 in helix G of CH2 projects toward the CH1 domain and makes a cation-␲ interaction with the aromatic ring of Trp165. Comparison with other tandem pairs of CH domains reveals an intramolecular domain organization very similar to the first ABD of fimbrin (Goldsmith et al., 1997) and to the intermolecular CH1-CH2 arrangement in the crystallographic head-tail dimers of utrophin (Keep et al., 1999b) and dystrophin ABDs (Norwood et al., 2000). The differences are limited to a small change in the angle between CH1 and CH2. The linker region between CH domains in plectin ABD does not adopt a defined secondary structure. Clear but weak electron density is observed for this linker region, and its B factors refine to higher values than those of the CH domains, indicating a significant degree of conformational variability. This is reminiscent of the CH1-CH2 linker of fimbrin, which is 16 residues longer than in plectin and is partially disordered in the crystal structure (Goldsmith et al., 1997). In contrast, the linker regions of utrophin and dystrophin adopt an ␣-helical conformation and are two residues longer than in plectin (Keep et al., 1999b;

Structure 618

Norwood et al., 2000). Thus, in plectin the interdomain arrangement is likely to be governed by the large interaction surface, not by specific orientations dictated by the very flexible interdomain linker. The plectin ABD is a monomer in solution as determined by size exclusion chromatography and sedimentation equilibrium (data not shown), and residues in the interdomain interface are not accessible under native conditions (see below). Thus, the crystal structure most likely corresponds to the conformation in solution.

Actin Binding Induces a Conformational Change in Plectin ABD N-terminal fragments of plectin containing the ABD have been shown to colocalize with actin filaments in vivo and to bind directly to actin in vitro (Andra et al., 1998; Fontao et al., 2001; Geerts et al., 1999). We confirmed that the ABD supports binding to F-actin (data not shown). The observed dissociation constant (KD) of the interaction is 22.3 ⫾ 2 ␮M, similar to that of the isolated ABD of utrophin and dystrophin (Moores and KendrickJones, 2000; Winder et al., 1995). The plectin N-terminal fragment 1–339, fused to maltose binding protein, has a higher affinity for F-actin (KD ⵑ0.3 ␮M) (Fontao et al., 2001). The observed variability may be due to different experimental conditions, including pH and ionic strength. Alternatively, the presence of plectin regions additional to the ABD may increase the affinity, as observed in utrophin (Moores and Kendrick-Jones, 2000). The low-resolution structures of several CH domains bound to F-actin have been obtained by electron microscopy and image reconstruction of F-actin decorated filaments (Galkin et al., 2002; Hanein et al., 1997; Hodgkinson et al., 1997; Moores et al., 2000; Volkmann et al., 2001). Docking of high-resolution structures into the electron microscopy maps yielded detailed pseudoatomic models of the CH domains bound to F-actin; these include the N-terminal ABD of fimbrin (Hanein et al., 1998), the ABD domain of utrophin (Galkin et al., 2002; Moores et al., 2000), and the CH domain of calponin (Bramham et al., 2002). The ABD of fimbrin bound to actin has been docked using the closed conformation observed in the crystal structure; it binds to F-actin in the cleft between two adjacent actin monomers (Hanein et al., 1998). In contrast, utrophin ABD has been proposed to bind F-actin in extended conformations (Galkin et al., 2002; Moores et al., 2000) similar to the one observed in the crystal structure (Keep et al., 1999b). Optimal docking of utrophin ABD requires interdomain reorientation, suggesting an induced fit mechanism. The CH domain of calponin bound to F-actin has been docked in a utrophin-like orientation in order to satisfy biochemical constraints (Bramham et al., 2002). The conformation displayed by the plectin ABD when bound to F-actin is unknown. Our crystal structure corresponds to the closed conformation of the plectin ABD that is compatible with a fimbrin-like interaction with actin. On the other hand, a conformational change would be required if plectin ABD binds to F-actin in an extended conformation. We used differential scanning calorimetry (DSC) to assess the effect of interaction with actin on the thermal stability of both proteins (Figure

Figure 2. Conformational Change of Plectin ABD upon Binding to F-Actin DSC scans obtained for F-actin complexes with wild-type plectin ABD (A), plectin T74C/S277C reduced (B) and oxidized (C). For each panel curves 1 and 2 represent melting profiles of 18.0 ␮M plectin ABD in the absence and presence of 18.0 ␮M F-actin, respectively. Curve 3 is for 18.0 ␮M F-actin alone. The peak corresponding to plectin ABD or the reduced T74C/S277C mutant shifts to lower temperature in the presence of actin. This effect is inhibited in the disulfide crosslinked T74C/S277C mutant.

2A; Table 2). The ABD shows a denaturation profile with a thermal transition maximum (Tm) of ⵑ64⬚C. In the presence of F-actin, the ABD is destabilized and its Tm is reduced to 59⬚C, whereas the Tm of F-actin is increased by 1⬚C in the presence of plectin. Thus, plectin ABD undergoes a conformational change upon interaction. The strong effect of F-actin on plectin ABD thermal stability (ⵑ5⬚C decrease in Tm value) indicates substantial conformational change in ABD upon F-actin binding, which is consistent with an induced fit mechanism. Thermal destabilization is not a typical behavior for actin binding proteins. For example, there is an increase in the thermal stability of myosin and tropomyosin upon binding to F-actin (Levitsky et al., 1998, 2000).

Plectin Actin Binding Domain Structure 619

Table 2. Thermodynamic Parameters Obtained from DSC Scans for Plectin-Actin, Plectin-␤4 Interactions

Plectin ABD Plectin ABD T74C/S277C reduced Plectin ABD T74C/S277C oxidized F-actin Plectin ABD ⫹ F-actin Plectin ABD T74C/S277C reduced ⫹ F-actin Plectin ABD T74C/S277C oxidized ⫹ F-actin ␤4 Plectin ABD ⫹ ␤4 Plectin ABD T74C/S277C reduced ⫹ ␤4 Plectin ABD T74C/S277C oxidized ⫹ ␤4

Tm1 (⬚C)

Tm2 (⬚C)

⌬Hcal (kcal/mol)

63.9 65.7 68.0 69.4 59.1 61.8 — 59.0 59.6 59.7 59.9

— — — — 70.3 69.7 69.2 — 63.9 65.7 68.0

115.0 111.2 110.9 125.0 — — — 116.8 — — —

The absolute error in Tm values did not exceed ⫾0.2⬚C; the relative error in ⌬Hcal values did not exceed ⫾10%.

To analyze the nature of the conformational change linked to actin binding, we have created the doublemutant T74C/S277C (Figure 3A). Thr74 is located in helix A of CH1 and S277 in helix G of CH2; the distance between their C␤ atoms is 4.1 A˚. The introduced sulfhydryl groups of the mutant are at optimal distance and orientation to form a disulfide bond without disrupting the interdomain orientation observed in the crystal structure. Both Thr74 and Ser277 are buried in the interdomain interface; therefore, these mutations do not alter the accessible surface of the protein. The formation of the

interdomain disulfide increases the electrophoretic mobility of the ABD and covalently links the N- and C-terminal fragments produced by trypsin digestion (Figure 3B). The T74C/S277C mutant behaves as a monomer in solution in both the oxidized and the reduced form, as judged by size exclusion chromatography. Wild-type and mutant proteins show the same secondary structure content as estimated by the far-uv circular dichroism spectra (data not shown). Thiol group titration of the reduced mutant yielded 0.3 ⫾ 0.3 free accessible sulfhydryl groups per molecule

Figure 3. The Double Mutant T74C/S277C Locks ABD in the Closed Conformation (A) Ribbon representation of the ABD structure; the view corresponds to a 180⬚ rotation around a vertical axis with respect to Figure 1B. The side chains of C74 and C277 were modeled forming the interdomain disulfide bridge and are shown in detail. Position of the trypsin digestion point between R121–M122 (Garcı´a-Alvarez and de Pereda, unpublished data) in the CE loop of CH1 is indicated by an arrow. (B) Analysis by SDS-PAGE under nonreducing (left lanes) or reducing conditions (right lanes) of the T74C/S277C mutant in the two oxidation states: disulfide bond opened and formed. Tryptic fragments ABD-Tryp1 and ABD-Tryp2 remain covalently linked under non-reducing SDS-PAGE when the C72– C277 disulfide bridge is formed. Position of molecular weight markers is indicated on the left side.

Structure 620

under native conditions and 3.1 ⫾ 0.1 groups per molecule under denaturing conditions. The third sulfhydryl group corresponds to C207; it is located in the AB loop of CH2 and is not exposed to the solvent. The inaccessibility of the cysteine residues engineered in the interdomain interface is consistent with a closed conformation for the ABD in solution, as the one observed in the crystal structure. The thermal stability of the T74C/S277C mutant in the reduced form is slightly higher than for the wild-type (Figure 2B; Table 2), possibly due to new hydrophobic interactions made by the introduced sulfhydryl groups. Formation of the disulfide bond results in an additional stabilization of the ABD that is reflected in a ⵑ2.3⬚C increase of the Tm. This effect can be rationalized in terms of additional stabilization of plectin ABD by the covalent crosslinking of the two CH domains. We propose that the engineered disulfide bond introduces a constraint between the two CH domains and prevents opening of the ABD without altering the structure of the individual domains or their arrangement. Compared to the wild-type protein, the reduced T74C/ S277C mutant has equivalent affinity for F-actin (KD ⫽ 23.1 ⫾ 0.8 ␮M) and displays a similar thermal destabilization upon binding to F-actin (Figure 2B). The oxidized T74C/S277C mutant, with the disulfide bond formed, has lower affinity for F-actin (KD ⫽ 59.3 ⫾ 10.7 ␮M), but its Tm does not change in the presence of F-actin (Figure 2C; Table 2). That is, locking of the closed conformation prevents the thermal destabilization of the ABD linked to the interaction with actin. This suggests that reduction in the Tm of wild-type ABD upon binding to actin reflects a conformational change that involves rearrangement of the CH domains. We termed the final conformation “open” to illustrate the disruption of the CH1-CH2 interface. The locked closed T74C/S277C mutant retains the ability to bind to F-actin, indicating that opening of the ABD is not essential for interaction with actin filaments. This suggests that the conformational change observed upon interaction with actin may occur subsequent to the binding. The oxidized T74C/S277C mutant prevents an interdomain rearrangement and permits a distinction between the two steps. The two-step mechanism of plectin ABD binding to actin can be interpreted on the basis of the previously described models of ABDs bound to F-actin (Figure 4). In a first step, plectin ABD binds to F-actin in a closed conformation compatible with a fimbrin-like model in which ABS-2, ABS-3, and the interdomain linker region form a continuous surface interface with F-actin. In this conformation the ABS-1 is hidden by the interdomain interface and does not contribute to the binding. In a second step, actin may act as a wedge triggering the ABD opening by alteration of the hinge angle between the CH domains. Disruption of the interdomain interface may lead to dissociation of the two plectin CH domains and exposure of the masked ABS-1. Direct contribution of ABS-1 to the interaction interface requires a ⵑ90⬚ rotation of CH1, while the conformational flexibility of the linker may allow the correct repositioning of ABS-3 toward the interface with actin. This final stage may be described by a model similar to the proposed structures of utrophin ABD bound to F-actin (Galkin et al., 2002; Moores et al., 2000).

Figure 4. Model of ABD Opening Linked to F-Actin Binding Surface representation of plectin ABD in the closed conformation observed in the crystal structure (upper panel), in the same orientation as in Figure 1B. Residues in equivalent positions as the ones implicated in fimbrin binding to F-actin are colored in green. The ABS2, ABS3, and the linker are in the same face. The observed conformational change is illustrated in a hypothetical extended conformation based on the structure of utrophin ABD bound to actin (lower panel). CH1 is rotated 90⬚ around a vertical axis to orient the ABS1 toward F-actin (viewer point of view). Disruption of the interdomain interface would allow the orientation of ABS3 in contact with F-actin. Residues corresponding to the regions of utrophin suggested to interact with F-actin are colored in green, and those coded by exon 2␣ are in red. The positions of T74 and S277 are indicated.

Interaction with Integrin ␣6␤4 Plectin N-terminal fragments containing the ABD interact with the cytoplasmic region of the integrin ␤4 subunit (Geerts et al., 1999; Rezniczek et al., 1998). The minimal ␤4 binding region in plectin has been mapped, using a yeast two-hybrid assay, to residues 36–302 (Geerts et al., 1999), which roughly corresponds to the crystallized ABD. The integrin ␤4 cytoplasmic region 1115–1355 is the minimal fragment needed to interact with N-terminal fragments of plectin containing the ABD (Geerts et al., 1999; Koster et al., 2001). This region includes the first tandem pair of FnIII domains and part of the connecting segment that links the second and third FnIII domains.

Plectin Actin Binding Domain Structure 621

Table 3. Affinity of Plectin ABD to Integrin ␤4 (1126–1355) Plectin

KD (␮M)

ABD ABD/2␣ ABD T94C/S277C reduced ABD T94C/S277C oxidized

24.8 37.9 57.5 42.9

⫾ ⫾ ⫾ ⫾

4.9 18.4 1.7 2.6

The binding affinity of the plectin ABD for integrin ␤4 fragments was determined by isothermal titration calorimetry (Table 3). All binding data were best described by a one to one stoichiometry model. The insertion of five residues coded by exon 2␣ does not modify the affinity for ␤4 significantly, suggesting that the AB loop and the short helix B of CH1 do not participate directly in the interaction. The effect of the interaction between plectin ABD and ␤4 (1126–1355) on the conformation of both proteins was analyzed by DSC (Figure 5; Table 2). Binding of plectin ABD to integrin ␤4 does not produce any change in the Tm of plectin. A small and consistent increase of ⵑ0.6⬚C in the thermal stability of the ␤4 fragment is observed upon binding to plectin ABD. Therefore, interaction with ␤4 is not linked to a detectable conformational change in the plectin ABD. To further evaluate the role of the interdomain conformational flexibility in the interaction with ␤4, we have measured the binding of the reduced and oxidized forms of the T74C/S277C ABD mutant. In both oxidation states, the affinity of the T74C/ S277C mutant showed the same order of magnitude as that of the wild-type. In either the presence or absence of the disulfide bridge, the mutants showed a similar behavior in DSC to the wild-type protein. Thus, the plectin ABD interacts with ␤4 in a closed conformation represented by the crystal structure. In contrast to the interaction with actin, individual plectin CH domains do not support binding to ␤4 (Geerts et al., 1999). The requirement of the complete plectin ABD for interaction with ␤4 suggests direct participation of both CH domains in binding. In addition, the domain organization may be essential for the correct presentation of the interaction surface. The binding interface may be formed by discontinuous binding sites in each of the CH domains or by a single site formed in the interdomain area. The crystal structure of the first pair of tandem FnIII domains of integrin ␤4 reveals an elongated arrangement of the two domains (de Pereda et al., 1999), also observed in solution (Chacon et al., 2000). A missense mutation in integrin ␤4, R1281W, has been described in a patient suffering from a nonlethal form of epidermolysis bullosa with pyloric atresia (Pulkkinen et al., 1998). This mutation disrupts binding of ␤4 1115–1457 to an N-terminal plectin fragment containing the ABD (Geerts et al., 1999). R1281 is located exposed to the solvent in the C⬘E loop of the second FnIII domain spatially close to E1242. The R1281W mutation is not likely to produce a general alteration of the FnIII fold. Lack of plectin binding to the R1281W mutant may be due to the loss of a polar interaction with a positively charged group from the ABD. R1281 lies at the edge of an acidic V-shaped groove formed at the interface between the first and second FnIII domain (de Pereda et al., 1999), suggesting

Figure 5. Interaction with Plectin ABD Induces Thermal Stabilization of ␤4 (1126–1355) DSC scans obtained for ␤4 complexes with wild-type ABD (A), ABD mutant T74C/S277C reduced (B) and oxidized (C). For each panel curves 1 and 2 represent melting profiles of 18.0 ␮M plectin ABD in the absence and presence of 18.0 ␮M ␤4 (1126–1355), respectively. Curve 3 is for 18.0 ␮M ␤4 (1126–1355) alone. No change was observed in the position of the plectin ABD peak, but a moderate temperature increase was observed for the ␤4 denaturation peak in the presence of plectin proteins.

that this region may be implicated in the interaction. Interestingly, plectin ABD has a basic belt around the interdomain seam and the helix G of CH1. This belt is surrounded by acidic residues (Figure 6) and has good shape complementarity with the ␤4 interdomain acidic groove. We speculate that the plectin-␤4 interaction involves interdomain surfaces in both pairs of tandem CH and FnIII domains, and, therefore, binding to ␤4 may induce a conformational lock of plectin ABD. Biological Implications The ability to interact with multiple components of the cytoskeleton makes plectin a highly versatile crosslinker. The N-terminal moiety of plectin contains an ABD

Structure 622

suggests that binding of wild-type plectin to F-actin induces rearrangement of the CH domains. Conformational changes are not observed during binding to a cytoplasmic fragment of integrin ␤4, providing evidence that binding to F-actin or integrin ␣6␤4 occurs through different molecular mechanisms. The alternative modes of interaction of the ABD may explain the cellular localization of plectin. For example, in epithelial cells the cytoplasmic moiety of integrin ␤4 recruits plectin into hemidesmosomes and prevents association with actin filaments (Niessen et al., 1997a). Interaction with ␤4 may induce a conformational lock of the ABD that would prevent proper interaction with F-actin. However, the locked closed ABD mutant retains the ability to bind to F-actin with an affinity in the same order of magnitude as the wild-type. Therefore, the competitive binding of plectin ABD to integrin ␣6␤4 and F-actin may involve both allosteric and steric components. Experimental Procedures Protein Expression and Purification The cDNA sequence corresponding to residues 59–293 of human plectin (EMBL accession number U53204) was amplified by polymerase chain reaction (PCR) and cloned into the bacterial expression vector pET15b (Dubendorff and Studier, 1991) using NdeI and BamHI restriction sites. The five residue insertion coded by the exon 2␣ was introduced by PCR. The absence of errors in the sequences was confirmed by DNA sequencing. Proteins were expressed in Escherichia coli strain BL21(DE3). Cell cultures were induced with 0.2 mM isopropylthiogalactoside (IPTG) for 3 hr at 37⬚C and pellets frozen at ⫺80⬚C. Upon thawing, cells were sonicated on ice and cell debris removed by centrifugation. The proteins were purified using a nickel-chelating affinity column (Amersham Biosciences, Piscataway, NJ). The His-tag was cleaved by thrombin digestion and removed by extensive dialysis against 20 mM TRIS (pH 7.0), 150 mM NaCl. The purified proteins include four residues (Gly-SerHis-Met) from the His-Tag at the N terminus. Integrin ␤4 fragment 1126–1355 was amplified by PCR, cloned into pET15b, expressed, and purified as for the plectin proteins.

Figure 6. Molecular Surface of Plectin ABD Two views of the solvent-accessible surface colored according to electrostatic potential: blue for positive (20 kT/e), red for negative (⫺20 kT/e). In the upper panel the ABD is in the same orientation as in Figure 1B. A basic cleft extends along the interdomain contact, and it is flanked by acidic residues.

composed of a tandem pair of CH domains. Plectin interacts through the ABD with actin or the cytoplasmic moiety of integrin ␣6␤4, and binding to these two partners is competitive (Geerts et al., 1999). We have solved the structure of the ABD and designed biophysical and mutational approaches to probe the different mechanisms responsible for the ABD dual function. The crystal structure of plectin ABD reveals a closed conformation with an extensive interaction surface between the two CH domains. This is likely to be the predominant arrangement in solution. A conformational change in plectin ABD linked to F-actin binding was inferred from differential scanning calorimetry analysis. A plectin ABD mutant containing an engineered interdomain disulfide bond still binds to F-actin although it does not undergo a measurable conformational change. This

Crystallization and Structure Determination Crystals of the fragment 59–293/2␣ were obtained at room temperature by vapor diffusion using 0.1 M TRIS (pH 7.0), 15% (v/v) 2-propanol as precipitant. Protein at 20 mg/ml in 20 mM TRIS (pH 7.0), 150 mM NaCl, was mixed with an equal volume of precipitant. Crystals belong to the space group P212121 with cell dimensions a ⫽ 33.6 A˚, b ⫽ 79.3 A˚, c ⫽ 82.4 A˚. Each asymmetric unit contains a single molecule corresponding to a solvent content of 33%. Prior to data collection, crystals were transferred into cryoprotectant solutions containing 0.1 M TRIS (pH 7.0), 17% (v/v) 2-propanol, and 20% (v/v) MPD and were frozen by direct immersion in liquid nitrogen. Data from a single crystal were collected at 100 K using a Rigaku FR-D X-ray generator and an R-Axis IV⫹⫹ image plate. Data were indexed with DENZO and reduced with SCALEPACK (Otwinowski and Minor, 1997). The structure was solved by molecular replacement using CNS (Brunger et al., 1998). The CH1 domain of dystrophin (PDB code 1DXX) (Norwood et al., 2000) and the CH2 domain of ␤-spectrin (PDB code 1AA2) (Carugo et al., 1997) were used as search models. The rotation function search only showed a clear peak for the CH2 domain, which was further confirmed in the translation function. The rotation function search with the CH1 domain of dystrophin did not yield any outstanding solution. Once the correct translation solution for the CH2 domain was fixed, it was possible to find a clear translation function solution for one of the rotation function solutions of the CH1 domain. The model was refined with the package CNS version 1.1 (Brunger et al., 1998) against data to 2.15 A˚ resolution. The refinement was started using the search models, in which residues not identical to

Plectin Actin Binding Domain Structure 623

the plectin sequence were mutated to Ala. After initial rigid body refinement of the two independent domains followed by simulated annealing, the R factor (Rwork) was 37.0% and the Rfree (calculated with 7% of reflections omitted from refinement) was 43.8%. Extra density corresponding to plectin residues not present in the initial model was unequivocally observed in fobs ⫺ fcalc sigmaA maps. Further refinement included cycles of conjugate gradient minimization and individual B factor refinement, alternated with manual model building. The model converged to an Rwork of 21.2% and an Rfree of 26.1%. The final model includes residues 59–293 of the plectin sequence, the five residues coded by exon 2␣, three additional residues at the N terminus coded by the expression vector, and 153 solvent molecules. The main chain torsion angles (91.1%) of the nonglycine residues lie in the most favored regions and 8.4% in the additional allowed regions of the Ramachandran plot, as defined in PROCHECK (Laskowski et al., 1993). Only Thr211 falls in the generously allowed region of the Ramachandran plot; similar torsion angles have been observed for residues occupying the equivalent position in other CH domain structures. Design and Production of the T74C/S277C ABD Mutant Identification of residues in optimal position to engineer an interdomain disulfide bridge was done with the program SSBOND (Hazes and Dijkstra, 1988). The double-mutant T74C/S277C of the plectin 59–293 fragment was made with the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA). The correct sequence of the mutant was confirmed by DNA sequencing. The mutant plectin protein was expressed in Escherichia coli and affinity purified as for the wild-type. Formation of the engineered disulfide bond was obtained by incubation of the protein at 1 mg/ml in 20 mM PIPES (pH 7.0), 150 mM NaCl with 0.1 mM CuSO4 and 0.1 mM o-phenanthroline (Lee et al., 1994) at room temperature for 4 hr. Oxidant was removed by extensive dialysis against 20 mM PIPES (pH 7.0), 150 mM NaCl, and the protein was concentrated to the desired concentration in an Amicon ultrafiltration cell (Millipore, Bedford, MA). Sulfhydryl groups were titrated with 5,5⬘-dithio-bis(2-nitrobenzoic acid) (DTNB) (Riddles et al., 1983) in 20 mM TRIS (pH 8.0), 150 mM NaCl, in the absence or presence of 6 M guanidinium hydrochloride. Correct formation of the disulfide bond was checked by SDS-PAGE under reducing and nonreducing conditions. Alternatively, the mutant was subject to limited proteolysis with 1% (w/w) Trypsin (Sigma, St. Louis, MO) for 1 hr at room temperature, and digested samples were analyzed in parallel with undigested ones. F-Actin Binding Assay Binding of plectin ABD to F-actin was characterized using a cosedimentation assay as described (Moores and Kendrick-Jones, 2000). Briefly, polymerized actin at 8 ␮M (Cytoskeleton, Denver, CO) was titrated with a range of plectin ABD concentrations, 2–120 ␮M, in 10 mM PIPES (pH 7.0), 0.2 mM CaCl2, 2 mM MgCl2, 1 mM ATP, 50 mM KCl. Samples were incubated for 1 hr at room temperature and centrifuged at 100,000 ⫻ g for 30 min at 25⬚C. Pellets were resuspended in twice the initial volume of SDS sample buffer, and supernatants were diluted with an equal volume of SDS sample buffer. Samples were analyzed by SDS-PAGE, stained with Coomassie blue and quantified by gel densitometry. The dissociation constant and stoichiometry were calculated by fitting a hyperbolic model to the plot of moles of bound ABD per mole of monomeric actin versus concentration of free ABD. In addition, binding of plectin ABD to F-actin was monitored via changes in the intrinsic tryptophan fluorescence. We established that the tryptophan fluorescence intensity of plectin ABD-F-actin complex was lower than the arithmetic sum of the fluorescence intensities of plectin ABD and F-actin alone. Thus, binding of plectin ABD to F-actin causes partial quenching of the tryptophan fluorescence of plectin ABD and/or F-actin. We have used this phenomenon to measure the affinity of plectin ABD to F-actin. Six micromolar of rabbit skeletal muscle F-actin (generous gift from Dr. Reisler, UCLA) was titrated with increasing amounts of plectin ABD in 20 mM PIPES (pH 7.0), 150 mM NaCl, 1.0 mM MgCl2, and 0.2 mM ATP. Tryptophan fluorescence intensity of the samples was measured in a MOS-250 fluorescence spectrometer (Bio Logic, France). Emission and excitation monochromators were set at 330 and 290 nm, respec-

tively. Fluorescence intensities of actin alone and plectin ABD alone were subtracted from the titration data before fitting. Differential Scanning Calorimetry DSC experiments were performed on a N-DSC II differential scanning calorimeter (Calorimetry Sciences Corp, Provo, UT) with a cell volume of ⵑ0.33 ml. All experiments were performed at a scanning rate of 1 K/min under 3.0 atm of pressure. Prior to measurements, plectin and ␤4 protein samples were dialyzed against 20 mM PIPES (pH 7.0) and 150 mM NaCl. Alternatively, experiments with rabbit skeletal muscle F-actin were performed in 20 mM PIPES (pH 7.0), 150 mM NaCl, 1.0 mM MgCl2, and 0.2 mM ATP. The dialysis buffer was used as the reference solution. The reversibility of the thermal transitions was checked by a second heating of the sample immediately after cooling, following the first scan. All thermal transitions were irreversible under the conditions used thereby permitting use of only simple thermodynamic parameters and terms to interpret the results. The thermal stability of the proteins was described by the temperature of the thermal transition maximum (Tm). The calorimetric enthalpy (⌬Hcal) was calculated as the area under the excess heat capacity function. Because these parameters can be obtained directly from experimental calorimetric traces after subtraction of the chemical baseline and concentration normalization, they can be used for the description of the irreversible thermal denaturation. Isothermal Titration Calorimetry ITC was performed on a VP-ITC calorimeter from Microcal (Northampton, MA). Fourteen microliters of plectin solution (1.4–1.7 mM) were injected into the cell containing 90–180 ␮M ␤4 solution. Reversed experiments where 150 ␮M plectin solution in the cell was titrated with ␤4 (1126–1355) solution (1.1–1.4 mM) were carried out as well. In each experiment 20 injections were made. The experiments were performed at 23⬚C. Prior to ITC titrations all protein samples were dialyzed against 20 mM PIPES (pH 7.0) and 150 mM NaCl. Experimental data were analyzed using Microcal Origin software provided by the ITC manufacturer (Microcal, Northampton, MA). Fitting of ITC data has been previously described (Freire et al., 1990). Acknowledgments We thank Robert C. Liddington for essential support and discussion, Nuria Assa-Munt, Carolyn Moores, Kenneth A. Taylor, and Clare McCleverty for critical comments on the manuscript, and Heidi Erlandsen and Ray Stevens for access to the X-ray facility. This work was supported by grants from the National Institutes of Health. Received: December 2, 2002 Revised: March 12, 2003 Accepted: March 20, 2003 Published: June 3, 2003 References Andra, K., Lassmann, H., Bittner, R., Shorny, S., Fassler, R., Propst, F., and Wiche, G. (1997). Targeted inactivation of plectin reveals essential function in maintaining the integrity of skin, muscle, and heart cytoarchitecture. Genes Dev. 11, 3143–3156. Andra, K., Nikolic, B., Stocher, M., Drenckhahn, D., and Wiche, G. (1998). Not just scaffolding: plectin regulates actin dynamics in cultured cells. Genes Dev. 12, 3442–3451. Banuelos, S., Saraste, M., and Carugo, K.D. (1998). Structural comparisons of calponin homology domains: implications for actin binding. Structure 6, 1419–1431. Bramham, J., Hodgkinson, J.L., Smith, B.O., Uhrin, D., Barlow, P.N., and Winder, S.J. (2002). Solution structure of the calponin CH domain and fitting to the 3D-helical reconstruction of F-actin:calponin. Structure 10, 249–258. Bresnick, A.R., Warren, V., and Condeelis, J. (1990). Identification of a short sequence essential for actin binding by Dictyostelium ABP-120. J. Biol. Chem. 265, 9236–9240. Brunger, A.T., Adams, P.D., Clore, G.M., DeLano, W.L., Gros, P.,

Structure 624

Grosse-Kunstleve, R.W., Jiang, J.S., Kuszewski, J., Nilges, M., Pannu, N.S., et al. (1998). Crystallography & NMR system: a new software suite for macromolecular structure determination. Acta Crystallogr. D Biol. Crystallogr. 54, 905–921. Carugo, K.D., Banuelos, S., and Saraste, M. (1997). Crystal structure of a calponin homology domain. Nat. Struct. Biol. 4, 175–179. Chacon, P., Diaz, J.F., Moran, F., and Andreu, J.M. (2000). Reconstruction of protein form with X-ray solution scattering and a genetic algorithm. J. Mol. Biol. 299, 1289–1302. Choi, H.J., Park-Snyder, S., Pascoe, L.T., Green, K.J., and Weis, W.I. (2002). Structures of two intermediate filament-binding fragments of desmoplakin reveal a unique repeat motif structure. Nat. Struct. Biol. 9, 612–620. Christopher, J.A. (1998). SPOCK: The Structural Properties Observation and Calculation Kit. The Center for Macromolecular Design (College Station, TX: Texas A&M University), http://quorum. tamu.edu/spock/Manual/Manual/Manual.html. de Arruda, M.V., Watson, S., Lin, C.S., Leavitt, J., and Matsudaira, P. (1990). Fimbrin is a homologue of the cytoplasmic phosphoprotein plastin and has domains homologous with calmodulin and actin gelation proteins. J. Cell Biol. 111, 1069–1079. de Pereda, J.M., Wiche, G., and Liddington, R.C. (1999). Crystal structure of a tandem pair of fibronectin type III domains from the cytoplasmic tail of integrin alpha6beta4. EMBO J. 18, 4087–4095. Dubendorff, J.W., and Studier, F.W. (1991). Creation of a T7 autogene. Cloning and expression of the gene for bacteriophage T7 RNA polymerase under control of its cognate promoter. J. Mol. Biol. 219, 61–68. Eger, A., Stockinger, A., Wiche, G., and Foisner, R. (1997). Polarisation-dependent association of plectin with desmoplakin and the lateral submembrane skeleton in MDCK cells. J. Cell Sci. 110, 1307– 1316. Foisner, R., and Wiche, G. (1987). Structure and hydrodynamic properties of plectin molecules. J. Mol. Biol. 198, 515–531. Foisner, R., Leichtfried, F.E., Herrmann, H., Small, J.V., Lawson, D., and Wiche, G. (1988). Cytoskeleton-associated plectin: in situ localization, in vitro reconstitution, and binding to immobilized intermediate filament proteins. J. Cell Biol. 106, 723–733. Foisner, R., Traub, P., and Wiche, G. (1991). Protein kinase A- and protein kinase C-regulated interaction of plectin with lamin B and vimentin. Proc. Natl. Acad. Sci. USA 88, 3812–3816. Foisner, R., Malecz, N., Dressel, N., Stadler, C., and Wiche, G. (1996). M-phase-specific phosphorylation and structural rearrangement of the cytoplasmic cross-linking protein plectin involve p34cdc2 kinase. Mol. Biol. Cell 7, 273–288. Fontao, L., Geerts, D., Kuikman, I., Koster, J., Kramer, D., and Sonnenberg, A. (2001). The interaction of plectin with actin: evidence for cross-linking of actin filaments by dimerization of the actinbinding domain of plectin. J. Cell Sci. 114, 2065–2076. Freire, E., Mayogra, O.L., and Straume, M. (1990). Isothermal titration calorimetry. Anal. Chem. 62, 950A–959A. Fuchs, E., and Yang, Y. (1999). Crossroads on cytoskeletal highways. Cell 98, 547–550. Fuchs, P., Zorer, M., Rezniczek, G.A., Spazierer, D., Oehler, S., Castanon, M.J., Hauptmann, R., and Wiche, G. (1999). Unusual 5⬘ transcript complexity of plectin isoforms: novel tissue-specific exons modulate actin binding activity. Hum. Mol. Genet. 8, 2461–2472. Galkin, V.E., Orlova, A., VanLoock, M.S., Rybakova, I.N., Ervasti, J.M., and Egelman, E.H. (2002). The utrophin actin-binding domain binds F-actin in two different modes: implications for the spectrin superfamily of proteins. J. Cell Biol. 157, 243–251. Geerts, D., Fontao, L., Nievers, M.G., Schaapveld, R.Q., Purkis, P.E., Wheeler, G.N., Lane, E.B., Leigh, I.M., and Sonnenberg, A. (1999). Binding of integrin alpha6beta4 to plectin prevents plectin association with F-actin but does not interfere with intermediate filament binding. J. Cell Biol. 147, 417–434. Goldsmith, S.C., Pokala, N., Shen, W., Fedorov, A.A., Matsudaira, P., and Almo, S.C. (1997). The structure of an actin-crosslinking domain from human fimbrin. Nat. Struct. Biol. 4, 708–712.

Hanein, D., Matsudaira, P., and DeRosier, D.J. (1997). Evidence for a conformational change in actin induced by fimbrin (N375) binding. J. Cell Biol. 139, 387–396. Hanein, D., Volkmann, N., Goldsmith, S., Michon, A.M., Lehman, W., Craig, R., DeRosier, D., Almo, S., and Matsudaira, P. (1998). An atomic model of fimbrin binding to F-actin and its implications for filament crosslinking and regulation. Nat. Struct. Biol. 5, 787–792. Hazes, B., and Dijkstra, B.W. (1988). Model building of disulfide bonds in proteins with known three-dimensional structure. Protein Eng. 2, 119–125. Hemmings, L., Kuhlman, P.A., and Critchley, D.R. (1992). Analysis of the actin-binding domain of alpha-actinin by mutagenesis and demonstration that dystrophin contains a functionally homologous domain. J. Cell Biol. 116, 1369–1380. Herrmann, H., and Wiche, G. (1987). Plectin and IFAP-300K are homologous proteins binding to microtubule-associated proteins 1 and 2 and to the 240-kilodalton subunit of spectrin. J. Biol. Chem. 262, 1320–1325. Hieda, Y., Nishizawa, Y., Uematsu, J., and Owaribe, K. (1992). Identification of a new hemidesmosomal protein, HD1: a major, high molecular mass component of isolated hemidesmosomes. J. Cell Biol. 116, 1497–1506. Hodgkinson, J.L., el-Mezgueldi, M., Craig, R., Vibert, P., Marston, S.B., and Lehman, W. (1997). 3-D image reconstruction of reconstituted smooth muscle thin filaments containing calponin: visualization of interactions between F-actin and calponin. J. Mol. Biol. 273, 150–159. Keep, N.H., Norwood, F.L., Moores, C.A., Winder, S.J., and KendrickJones, J. (1999a). The 2.0 A structure of the second calponin homology domain from the actin-binding region of the dystrophin homologue utrophin. J. Mol. Biol. 285, 1257–1264. Keep, N.H., Winder, S.J., Moores, C.A., Walke, S., Norwood, F.L., and Kendrick-Jones, J. (1999b). Crystal structure of the actin-binding region of utrophin reveals a head-to-tail dimer. Struct. Fold. Des. 7, 1539–1546. Koster, J., Kuikman, I., Kreft, M., and Sonnenberg, A. (2001). Two different mutations in the cytoplasmic domain of the integrin beta 4 subunit in nonlethal forms of epidermolysis bullosa prevent interaction of beta 4 with plectin. J. Invest. Dermatol. 117, 1405–1411. Kraulis, P.J. (1991). MOLSCRIPT: a program to produce both detailed and schematic plots of protein structures. J. Appl. Crystallogr. 24, 946–950. Laskowski, R.A., MacArthur, M.W., Moss, D.S., and Thornton, J.M. (1993). PROCHECK: a program to check the stereochemical quality of protein structures. J. Appl. Crystallogr. 26, 283–291. Lee, G.F., Burrows, G.G., Lebert, M.R., Dutton, D.P., and Hazelbauer, G.L. (1994). Deducing the organization of a transmembrane domain by disulfide cross-linking. The bacterial chemoreceptor Trg. J. Biol. Chem. 269, 29920–29927. Leung, C.L., Green, K.J., and Liem, R.K. (2002). Plakins: a family of versatile cytolinker proteins. Trends Cell Biol. 12, 37–45. Levine, B.A., Moir, A.J., Patchell, V.B., and Perry, S.V. (1992). Binding sites involved in the interaction of actin with the N-terminal region of dystrophin. FEBS Lett. 298, 44–48. Levitsky, D.I., Nikolaeva, O.P., Orlov, V.N., Pavlov, D.A., Ponomarev, M.A., and Rostkova, E.V. (1998). Differential scanning calorimetric studies on myosin and actin. Biochemistry 63, 322–333. Levitsky, D.I., Rostkova, E.V., Orlov, V.N., Nikolaeva, O.P., Moiseeva, L.N., Teplova, M.V., and Gusev, N.B. (2000). Complexes of smooth muscle tropomyosin with F-actin studied by differential scanning calorimetry. Eur. J. Biochem. 267, 1869–1877. Liu, Y., and Eisenberg, D. (2002). 3D domain swapping: As domains continue to swap. Protein Sci. 11, 1285–1299. Malecz, N., Foisner, R., Stadler, C., and Wiche, G. (1996). Identification of plectin as a substrate of p34cdc2 kinase and mapping of a single phosphorylation site. J. Biol. Chem. 271, 8203–8208. Merrit, E.A., and Bacon, D.J. (1997). Raster 3D photorealistic molecular graphics. Methods Enzymol. 277, 505–524.

Plectin Actin Binding Domain Structure 625

Moores, C.A., and Kendrick-Jones, J. (2000). Biochemical characterisation of the actin-binding properties of utrophin. Cell Motil. Cytoskeleton 46, 116–128.

homology in the F-actin binding domains of gelsolin and alphaactinin: implications for the requirements of severing and capping. J. Cell Biol. 119, 835–842.

Moores, C.A., Keep, N.H., and Kendrick-Jones, J. (2000). Structure of the utrophin actin-binding domain bound to F-actin reveals binding by an induced fit mechanism. J. Mol. Biol. 297, 465–480.

Winder, S.J., Hemmings, L., Maciver, S.K., Bolton, S.J., Tinsley, J.M., Davies, K.E., Critchley, D.R., and Kendrick-Jones, J. (1995). Utrophin actin binding domain: analysis of actin binding and cellular targeting. J. Cell Sci. 108, 63–71.

Niessen, C.M., Hulsman, E.H., Oomen, L.C., Kuikman, I., and Sonnenberg, A. (1997a). A minimal region on the integrin beta4 subunit that is critical to its localization in hemidesmosomes regulates the distribution of HD1/plectin in COS-7 cells. J. Cell Sci. 110, 1705– 1716. Niessen, C.M., Hulsman, E.H., Rots, E.S., Sanchez-Aparicio, P., and Sonnenberg, A. (1997b). Integrin alpha 6 beta 4 forms a complex with the cytoskeletal protein HD1 and induces its redistribution in transfected COS-7 cells. Mol. Biol. Cell 8, 555–566. Nikolic, B., Mac Nulty, E., Mir, B., and Wiche, G. (1996). Basic amino acid residue cluster within nuclear targeting sequence motif is essential for cytoplasmic plectin-vimentin network junctions. J. Cell Biol. 134, 1455–1467. Norwood, F.L., Sutherland-Smith, A.J., Keep, N.H., and KendrickJones, J. (2000). The structure of the N-terminal actin-binding domain of human dystrophin and how mutations in this domain may cause Duchenne or Becker muscular dystrophy. Struct. Fold. Des. 8, 481–491. Otwinowski, Z., and Minor, W. (1997). Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol. 276, 307–326. Pulkkinen, L., Rouan, F., Bruckner-Tuderman, L., Wallerstein, R., Garzon, M., Brown, T., Smith, L., Carter, W., and Uitto, J. (1998). Novel ITGB4 mutations in lethal and nonlethal variants of epidermolysis bullosa with pyloric atresia: missense versus nonsense. Am. J. Hum. Genet. 63, 1376–1387. Pytela, R., and Wiche, G. (1980). High molecular weight polypeptides (270,000–340,000) from cultured cells are related to hog brain microtubule-associated proteins but copurify with intermediate filaments. Proc. Natl. Acad. Sci. USA 77, 4808–4812. Rezniczek, G.A., de Pereda, J.M., Reipert, S., and Wiche, G. (1998). Linking integrin alpha6beta4-based cell adhesion to the intermediate filament cytoskeleton: direct interaction between the beta4 subunit and plectin at multiple molecular sites. J. Cell Biol. 141, 209–225. Riddles, P.W., Blakeley, R.L., and Zerner, B. (1983). Reassessment of Ellman’s reagent. Methods Enzymol. 91, 49–60. Ruhrberg, C., and Watt, F.M. (1997). The plakin family: versatile organizers of cytoskeletal architecture. Curr. Opin. Genet. Dev. 7, 392–397. Seifert, G.J., Lawson, D., and Wiche, G. (1992). Immunolocalization of the intermediate filament-associated protein plectin at focal contacts and actin stress fibers. Eur. J. Cell Biol. 59, 138–147. Steinbock, F.A., and Wiche, G. (1999). Plectin: a cytolinker by design. Biol. Chem. 380, 151–158. Steinbock, F.A., Nikolic, B., Coulombe, P.A., Fuchs, E., Traub, P., and Wiche, G. (2000). Dose-dependent linkage, assembly inhibition and disassembly of vimentin and cytokeratin 5/14 filaments through plectin’s intermediate filament-binding domain. J. Cell Sci. 113, 483–491. Stradal, T., Kranewitter, W., Winder, S.J., and Gimona, M. (1998). CH domains revisited. FEBS Lett. 431, 134–137. Svitkina, T.M., Verkhovsky, A.B., and Borisy, G.G. (1996). Plectin sidearms mediate interaction of intermediate filaments with microtubules and other components of the cytoskeleton. J. Cell Biol. 135, 991–1007. Uitto, J., Pulkkinen, L., Smith, F.J., and McLean, W.H. (1996). Plectin and human genetic disorders of the skin and muscle. The paradigm of epidermolysis bullosa with muscular dystrophy. Exp. Dermatol. 5, 237–246. Volkmann, N., DeRosier, D., Matsudaira, P., and Hanein, D. (2001). An atomic model of actin filaments cross-linked by fimbrin and its implications for bundle assembly and function. J. Cell Biol. 153, 947–956. Way, M., Pope, B., and Weeds, A.G. (1992). Evidence for functional

Accession Numbers The atomic coordinates and structure factors have been deposited in the Protein Data Bank under ID code 1MB8.