SemOpenAlex |

SemOpenAlex

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2983455466> ?p ?o ?g. }

Showing items 1 to 63 of 63 with 100 items per page.

W2983455466 abstract "Full text Figures and data Side by side Abstract eLife digest Introduction Results Discussion Materials and methods References Decision letter Author response Article and author information Metrics Abstract The cross-talk between dynamic microtubules and integrin-based adhesions to the extracellular matrix plays a crucial role in cell polarity and migration. Microtubules regulate the turnover of adhesion sites, and, in turn, focal adhesions promote the cortical microtubule capture and stabilization in their vicinity, but the underlying mechanism is unknown. Here, we show that cortical microtubule stabilization sites containing CLASPs, KIF21A, LL5β and liprins are recruited to focal adhesions by the adaptor protein KANK1, which directly interacts with the major adhesion component, talin. Structural studies showed that the conserved KN domain in KANK1 binds to the talin rod domain R7. Perturbation of this interaction, including a single point mutation in talin, which disrupts KANK1 binding but not the talin function in adhesion, abrogates the association of microtubule-stabilizing complexes with focal adhesions. We propose that the talin-KANK1 interaction links the two macromolecular assemblies that control cortical attachment of actin fibers and microtubules. https://doi.org/10.7554/eLife.18124.001 eLife digest Animal cells are organized into tissues and organs. A scaffold-like framework outside of the cells called the extracellular matrix provides support to the cells and helps to hold them in place. Cells attach to the extracellular matrix via structures called focal adhesions on the cell surface; these structures contain a protein called talin. For a cell to be able to move, the existing focal adhesions must be broken down and new adhesions allowed to form. This process is regulated by the delivery and removal of different materials along fibers called microtubules. Microtubules can usually grow and shrink rapidly, but near focal adhesions they are captured at the surface of the cell and become more stable. However, it is not clear how focal adhesions promote microtubule capture and stability. Bouchet et al. found that a protein called KANK1 binds to the focal adhesion protein talin in human cells grown in a culture dish. This allows KANK1 to recruit microtubules to the cell surface around the focal adhesions by binding to particular proteins that are associated with microtubules. Disrupting the interaction between KANK1 and talin by making small alterations in these two proteins blocked the ability of focal adhesions to capture surrounding microtubules. The next step following on from this work will be to find out whether this process also takes place in the cells within an animal’s body, such as a fly or a mouse. https://doi.org/10.7554/eLife.18124.002 Introduction Cell adhesions to the extracellular matrix support epithelial integrity and cell migration, and also provide signaling hubs that coordinate cell proliferation and survival (Hynes, 1992). Integrin-based adhesions (focal adhesions, FAs) are large macromolecular assemblies, in which the cytoplasmic tails of integrins are connected to the actin cytoskeleton. One of the major components of FAs is talin, a ~2500 amino acid dimeric protein, which plays a key role in adhesion formation by activating integrins (Anthis et al., 2009), coupling them to cytoskeletal actin (Atherton et al., 2015), regulating adhesion dynamics and recruiting different structural and signaling molecules (Calderwood et al., 2013; Gardel et al., 2010; Wehrle-Haller, 2012). While the major cytoskeletal element associated with FAs is actin, microtubules also play an important role in adhesion by regulating the FA turnover (Akhmanova et al., 2009; Byron et al., 2015; Kaverina et al., 1999, 1998; Krylyshkina et al., 2003; Small and Kaverina, 2003; Stehbens and Wittmann, 2012; Yue et al., 2014). The recent proteomics work showed that microtubule-FA cross-talk strongly depends on the activation state of the integrins (Byron et al., 2015). Microtubules can affect adhesions by serving as tracks for delivery of exocytotic carriers (Stehbens et al., 2014), by controlling endocytosis required for adhesion disassembly (Ezratty et al., 2005; Theisen et al., 2012) and by regulating the local activity of signaling molecules such as Rho GTPases (for review, see [Kaverina and Straube, 2011; Stehbens and Wittmann, 2012]). In contrast to actin, which is directly coupled to FAs, microtubules interact with the plasma membrane sites that surround FAs. A number of proteins have been implicated in microtubule attachment and stabilization in the vicinity of FAs. Among them are the microtubule plus end tracking proteins (+TIPs) CLASP1/2 and the spectraplakin MACF1/ACF7, which are targeted to microtubule tips by EB1, and a homologue of EB1, EB2, which binds to mitogen-activated protein kinase kinase kinase kinase 4 (MAP4K4) (Drabek et al., 2006; Honnappa et al., 2009; Kodama et al., 2003; Mimori-Kiyosue et al., 2005). The interaction of CLASPs with the cell cortex depends on the phosphatidylinositol 3, 4, 5-trisphosphate (PIP3)-interacting protein LL5β, to which CLASPs bind directly, and is partly regulated by PI-3 kinase activity (Lansbergen et al., 2006). Other components of the same cortical assembly are the scaffolding proteins liprin-α1 and β1, a coiled-coil adaptor ELKS/ERC1, and the kinesin-4 KIF21A (Lansbergen et al., 2006; van der Vaart et al., 2013). Both liprins and ELKS are best known for their role in organizing presynaptic secretory sites (Hida and Ohtsuka, 2010; Spangler and Hoogenraad, 2007); in agreement with this function, ELKS is required for efficient constitutive exocytosis in HeLa cells (Grigoriev et al., 2007, 2011). LL5β, liprins and ELKS form micrometer-sized cortical patch-like structures, which will be termed here cortical microtubule stabilization complexes, or CMSCs. The CMSCs are strongly enriched at the leading cell edges, where they localize in close proximity of FAs but do not overlap with them ([Lansbergen et al., 2006; van der Vaart et al., 2013], reviewed in [Astro and de Curtis, 2015]). They represent a subclass of the previously defined plasma membrane-associated platforms (PMAPs) (Astro and de Curtis, 2015), which have overlapping components such as liprins, but may not be necessarily involved in microtubule regulation, as is the case for liprin-ELKS complexes in neurons, where they are part of cytomatrix at the active zone (Gundelfinger and Fejtova, 2012). Several lines of evidence support the importance of the CMSC-FA cross-talk. In migrating keratinocytes, LL5β and CLASPs accumulate around FAs and promote their disassembly by targeting the exocytosis of matrix metalloproteases to FA vicinity (Stehbens et al., 2014). Furthermore, liprin-α1, LL5α/β and ELKS localize to protrusions of human breast cancer cells and are required for efficient cell migration and FA turnover (Astro et al., 2014). In polarized epithelial cells, LL5β and CLASPs are found in the proximity of the basal membrane, and this localization is controlled by the integrin activation state (Hotta et al., 2010). CLASP and LL5-mediated anchoring of MTs to the basal cortex also plays a role during chicken embryonic development, where it prevents the epithelial-mesenchymal transition of epiblast cells (Nakaya et al., 2013). LL5β, CLASPs and ELKS were also shown to concentrate at podosomes, actin-rich structures, which can remodel the extracellular matrix (Proszynski and Sanes, 2013). Interestingly, LL5β-containing podosome-like structures are also formed at neuromuscular junctions (Kishi et al., 2005; Proszynski et al., 2009; Proszynski and Sanes, 2013), and the complexes of LL5β and CLASPs were shown to capture microtubule plus ends and promote delivery of acetylcholine receptors (Basu et al., 2015, 2014; Schmidt et al., 2012). While the roles of CMSCs in migrating cells and in tissues are becoming increasingly clear, the mechanism underlying their specific targeting to integrin adhesion sites remains elusive. Recently, we found that liprin-β1 interacts with KANK1 (van der Vaart et al., 2013), one of the four members of the KANK family of proteins, which were proposed to act as tumor suppressors and regulators of cell polarity and migration through Rho GTPase signaling (Gee et al., 2015; Kakinuma et al., 2008, 2009; Li et al., 2011; Roy et al., 2009). KANK1 recruits the kinesin-4 KIF21A to CMSCs, which inhibits microtubule polymerization and prevents microtubule overgrowth at the cell edge (Kakinuma and Kiyama, 2009; van der Vaart et al., 2013). Furthermore, KANK1 participates in clustering of the other CMSC components (van der Vaart et al., 2013). Here, we found that KANK1 is required for the association of the CMSCs with FAs. The association of KANK1 with FAs depends on the KN domain, a conserved 30 amino acid polypeptide sequence present in the N-termini of all KANK proteins. Biochemical and structural analysis showed that the KN domain interacts with the R7 region of the talin rod. Perturbation of this interaction both from the KANK1 and the talin1 side prevented the accumulation of CMSC complexes around focal adhesions and affected microtubule organization around FAs. We propose that KANK1 molecules, recruited by talin to the outer rims of FA, serve as 'seeds' for organizing other CMSC components in the FA vicinity through multivalent interactions between these components. This leads to co-organization of two distinct cortical assemblies, FAs and CMSCs, responsible for the attachment of actin and microtubules, respectively, and ensures effective cross-talk between the two types of cytoskeletal elements. Results Identification of talin1 as a KANK1 binding partner Our previous work showed that the endogenous KANK1 colocalizes with LL5β, liprins and KIF21A in cortical patches that are closely apposed to, but do not overlap with FAs (van der Vaart et al., 2013). We confirmed these results both in HeLa cells and the HaCaT immortal keratinocyte cell line, in which CMSC components CLASPs and LL5β were previously shown to strongly cluster around FAs and regulate their turnover during cell migration (Stehbens et al., 2014) (Figure 1—figure supplement 1A,B). Inhibition of myosin-II with blebbistatin, which reduces tension on the actin fibers and affects the activation state of FA molecules, such as integrins and talin (Parsons et al., 2010), caused not only FA disassembly but also a strong reduction in clustering of CMSC components at the cell periphery (Figure 1—figure supplement 2A,B), as described previously (Stehbens et al., 2014). To investigate this effect in more detail, we partially inhibited contractility using a Rho-associated protein kinase 1 (ROCK1) inhibitor, Y-27632 (Oakes et al., 2012). In these conditions, the number of FAs was not affected although their size was reduced (Figure 1—figure supplement 2C–E). This treatment was sufficient to diminish CMSC clustering at the cell edge (Figure 1—figure supplement 2C,F). Interestingly, at the same time we observed a very significant increase in the overlap of KANK1 with FA adhesion markers (Figure 1—figure supplement 2C,G). Live imaging of KANK1 together with the FA marker paxillin showed a gradual redistribution of KANK1 into the areas occupied by FAs upon ROCK1 inhibitor-induced attenuation of contractility (Figure 1—figure supplement 2H, Video 1). These data indicate that the organization of CMSCs at the cell cortex might be controlled by interactions with tension-sensitive components of FAs. Figure 1 with 3 supplements see all Download asset Open asset The KN motif of KANK1 interacts with the R7 domain of talin1. (A) Schematic representation of KANK1 and the deletion mutants used in this study, and the summary of their interactions and localization. N.d., not determined in this study. (B) TIRFM images of live HeLa cells transiently expressing the indicated GFP-tagged KANK1 deletion mutants together with the focal adhesion marker mCherry-paxillin. In these experiments, endogenous KANK1 and KANK2 were also expressed. (C) Identification of the binding partners of Bio-GFP-tagged KANK1 and its indicated deletion mutants by using streptavidin pull down assays from HEK293T cells combined with mass spectrometry. (D) Streptavidin pull down assays with the BioGFP-tagged KANK1 or the indicated KANK1 mutants, co-expressed with GFP-talin1 in HEK293T cells, analyzed by Western blotting with the indicated antibodies. (E) Sequence alignment of KANK1 and KANK2 with the known talin-binding sites of DLC1, RIAM and Paxillin. The LD-motif and the interacting hydrophobic residues are highlighted green and blue respectively. (F) Schematic representation of talin1 and the deletion mutants used in this study, and their interaction with KANK1. (G) Streptavidin pull down assays with the BioGFP-tagged talin1 or the indicated talin1 mutants, co-expressed with HA-KANK1 in HEK293T cells, analyzed by Western blotting with the indicated antibodies. https://doi.org/10.7554/eLife.18124.003 Video 1 Download asset This video cannot be played in place because your browser does support HTML5 video. You may still download the video for offline viewing. Download as MPEG-4 Download as WebM Download as Ogg Effect of myosin II inhibition on KANK1 localization to FA. TIRFM-based time-lapse imaging of HeLa cells stably expressing GFP-KANK1 and TagRFP-paxillin and treated when indicated with 10 μM ROCK1 inhibitor Y-27632. Both red and green fluorescence images were acquired at 1 min interval and displayed at 15 frames/second (accelerated 900 times). https://doi.org/10.7554/eLife.18124.008 To identify the domains of KANK1 required for cortical localization, we performed deletion mapping. KANK1 comprises an N-terminal KANK family-specific domain of unknown function, the KN domain (residues 30–68) (Kakinuma et al., 2009), a coiled coil region, the N-terminal part of which interacts with liprin-β1, and a C-terminal ankyrin repeat domain, which binds to KIF21A (van der Vaart et al., 2013), while the rest of the protein is predicted to be unstructured (Figure 1A). Surprisingly, the KN domain alone strongly and specifically accumulated within FAs (Figure 1B). A similar localization was also seen with a somewhat larger N-terminal fragment of KANK1, Nter, as well as the Nter-CC1 deletion mutant, which contained the first, liprin-β1-binding coiled coil region of KANK1 (Figure 1A,B). However, an even larger N-terminal part of KANK1, encompassing the whole coiled coil domain (Nter-CC) showed a pronounced enrichment at the FA rim (Figure 1A,B). The KANK1 deletion mutant missing only the C-terminal ankyrin repeat domain (△ANKR) was completely excluded from FAs but accumulated in their immediate vicinity, similar to the full-length KANK1 (Figure 1A,B). A tight ring-like localization at the outer rim of FAs was also observed with a KANK1 mutant, which completely missed the coiled coil region but contained the ankyrin repeat domain (△CC), while the mutant which missed just the KN domain showed no accumulation around FAs (Figure 1A,B). To test whether the exclusion of larger KANK1 fragments from the FA core was simply due to the protein size, we fused GFP-tagged KN domain to the bacterial protein β-D-galactosidase (LacZ), but found that this fusion accumulated inside and not around FAs (Figure 1—figure supplement 3). Since GFP-KN-LacZ and GFP-KANK1-△ANKRD have a similar size (1336 and 1400 amino acids, respectively), but one accumulates inside FAs, while the other is excluded to their periphery, this result suggests that features other than the mere protein size determine the specific localization of KANK1 to the FA rim. We conclude that the KN domain of KANK1 has an affinity for FAs, but the presence of additional KANK1 sequences prevents the accumulation of the protein inside FAs and instead leads to the accumulation of KANK1 at the FA periphery. To identify the potential FA-associated partners of KANK1, we co-expressed either full-length KANK1 or its N-terminal and C-terminal fragments fused to GFP and a biotinylation (Bio) tag together with biotin ligase BirA in HEK293T cells and performed streptavidin pull down assays combined with mass spectrometry. In addition to the already known binding partners of KANK1, such as KIF21A, liprins and LL5β, we identified talin1 among the strongest hits (Figure 1C). Talin2 was also detected in a pull down with the KANK1 N-terminus though not with the full-length protein (Figure 1C). The interaction between KANK1 and talin1 was confirmed by Western blotting, and subsequent deletion mapping showed that the talin1-binding region of KANK1 encompasses the KN domain (Figure 1A,D), while liprin-β1 binds to the N-terminal part of the coiled coil domain, as shown previously (van der Vaart et al., 2013). Sequence analysis of the KN domain showed that it is predicted to form a helix and contains a completely conserved leucine aspartic acid-motif (LD-motif) (Alam et al., 2014; Zacharchenko et al., 2016). The LD-motifs in RIAM (Goult et al., 2013), DLC1 and Paxillin (Zacharchenko et al., 2016) have recently been identified as talin-binding sites that all interact with talin via a helix addition mechanism. Alignment of the KN domain of KANK with the LD-motif of DLC1, RIAM and Paxillin (Zacharchenko et al., 2016) revealed that the hydrophobic residues that mediate interaction with talin are present in the KN domain (Figure 1E). Using deletion analysis, we mapped the KANK1-binding site of talin1 to the central region of the talin rod, comprising the R7-R8 domains (Figure 1F). This R7-R8 region of talin is unique (Gingras et al., 2010), as the 4-helix bundle R8 is inserted into a loop of the 5-helix bundle R7, and thus protrudes from the linear chain of 5-helix bundles of the talin rod (Figures 1F, 2A). This R8 domain serves as a binding hub for numerous proteins including vinculin, synemin and actin (Calderwood et al., 2013). R8 also contains a prototypic recognition site for LD-motif proteins, including DLC1 (Figure 2B), Paxillin and RIAM (Zacharchenko et al., 2016). Based on the presence of the LD-binding site, we anticipated that KANK1 would also interact with R8. However, deletion mapping revealed that KANK1 in fact binds to the talin1 rod domain R7 (Figure 1F,G), suggesting that KANK1 interacts with a novel binding site on talin1. Figure 2 with 2 supplements see all Download asset Open asset Biochemical and structural characterization of the Talin-KANK interaction. (A) Schematic representation of Talin1, with F-actin, β-integrin and vinculin binding sites highlighted. The KANK1 binding site on R7 is also shown. (B) The structure of the complex between talin1 R7-R8 (white) and the LD-motif of DLC1 (yellow) bound on the R8 subdomain (PDB ID. 5FZT, [Zacharchenko et al., 2016]). Residues W1630 and Y1389 (blue) and S1641 (magenta) are highlighted. (C–D) The KANK KN domain binds to a novel site on talin R7. 1H,15N HSQC spectra of 150 μM 15N-labelled talin1 R7 (residues 1357–1659 Δ1454–1586) in the absence (black) or presence of KANK1(30–68)C peptide (red) (top panel) or KANK1-4A (green) (bottom panel) at a ratio of 1:3. (D) Mapping of the KANK1 binding site on R7 as detected by NMR using weighted chemical shift differences (red) – mapped onto the R7-R8 structure in (B). Residues W1630 and Y1389 (blue) and G1404 and S1641 (magenta) are highlighted. (E) Structural model of the talin1:KANK1 interface. Ribbon representation of the KANK1 binding site, comprised of the hydrophobic groove between helices 29 and 36 of R7. Two bulky conserved residues, W1630 and Y1389 (blue) hold these two helices apart forming the binding interface. A small glycine side chain (G1404) creates a pocket between the helices. S1641 (magenta) has been shown previously to be a phosphorylation site (Ratnikov et al., 2005). The KN peptide (green) docked onto the KANK binding surface. (F–G) Biochemical characterization of the talin:KANK interaction. (F) Binding of BODIPY-labeled KANK1(30–60)C, KANK2(31–61)C and KANK1-4A peptides to Talin1 R7-R8 (1357–1659) was measured using a Fluorescence Polarization assay. (G) Binding of BODIPY-labeled KANK1(30–60)C to wild type R7-R8, R7-R8 S1641E, R7-R8 G1404L and R7-R8 W1630A. Dissociation constants ± SE (μM) for the interactions are indicated in the legend. All measurements were performed in triplicate. ND, not determined. https://doi.org/10.7554/eLife.18124.009 Structural characterization and mutational analysis of the KANK1-talin1 complex To explore the interaction between talin1 and KANK1 in more detail, we used NMR chemical shift mapping using 15N-labeled talin1 R7-R8 (residues 1357–1653) and a synthetic KANK1 peptide of the KN domain, KANK1(30–60). The addition of the KANK1(30–60) peptide resulted in large spectral changes (Figure 2C), most of which were in the slow exchange regime on the NMR timescale indicative of a tight interaction. In agreement with the pull down data, the signals that shifted in slow exchange upon the addition of KANK1(30–60) mapped largely onto the R7 domain with only small progressive shift changes evident on R8. To validate R7 as the major KANK1-binding site on talin, we repeated the NMR experiments using the individual domains, R8 (residues 1461–1580) and R7 (residues 1359–1659 Δ1454–1586). Addition of KANK1(30–60) induced small chemical shift changes on the R8 domain indicative of a weak interaction (most likely due to the interaction of LD with the LD-recognition box on R8). However, the addition of a 0.5 molar ratio of KANK1(30–60) to R7 induced large spectral changes with many of the peaks displaying two locations, corresponding to the free peak position and the bound peak position. This is indicative of slow-exchange and confirms a high affinity interaction between R7 and KANK1. The KN peptide is the first identified ligand for the R7 domain. NMR chemical shifts also provide information on the residues involved in the interaction, as the peaks in the 15N-HSQC spectrum pertain to individual residues in the protein. To map these chemical shift changes onto the structure of R7-R8, it was first necessary to complete the backbone chemical shift assignments of the R7 domain. This was achieved using conventional triple resonance experiments as described previously (Banno et al., 2012), using 13C,15N labeled R7. The chemical shift changes mapped predominantly onto one face of R7, comprised of helices 2 and 5 of the 5-helix bundle (talin rod helices 29 and 36), as shown in Figure 2D–E. Our recent elucidation of the interaction between the LD-motif of DLC1 and talin R8 has generated insight into how LD-motifs are recognized by helical bundles (PDB ID. 5FZT, [Zacharchenko et al., 2016]). In the DLC1:talin R8 complex the DLC1 peptide adopts a helical conformation that packs against two helices on the side of the helical bundle. It is becoming increasingly clear that other LD-motif proteins bind to talin through a similar interaction mode (Zacharchenko et al., 2016). The surface of α2 and α5 on R7 forms a hydrophobic groove that the KANK1 helix docks into. A striking feature of this KANK1 binding surface is that the two helices are held apart by the conserved aromatic residues, W1630 at the end of α5 and Y1389 at the end of α2 (Figure 2B,E). W1630 and Y1389 thus essentially serve as molecular rulers, separating helices α2 and α5 by ~8Å (compared with ~5–6Å for the other bundles in R7-R8). The spacing between the two helices is enhanced further as the residues on the inner helical faces, S1400, G1404, S1411 on α2 and S1637 and S1641 on α5, have small side chains which have the effect of creating two conserved pockets midway along the hydrophobic groove of the KANK1-binding site (Figure 2E). The talin-binding site on KANK1 is unusual as it contains a double LD-motif, LDLD. The structure of R7 revealed a potential LD-recognition box with the positive charges, K1401 and R1652 positioned on either side to engage either one, or both, of the aspartic residues. Using the docking program HADDOCK (van Zundert et al., 2016), we sought to generate a structural model of the KANK1/R7 complex, using the chemical shift mapping on R7 and a model of KANK1(30–60) as a helix (created by threading the KANK1 sequence onto the DLC1 LD-motif helix). This analysis indicated that the KANK-LD helix can indeed pack against the sides of α2 and α5 (Figure 2E). Interestingly, all of the models, irrespective of which way the KANK1 helix ran along the surface, positioned the bulky aromatic residue, Y48 in KANK1, in the hydrophobic pocket created by G1404. This raised the possibility that mutation of G1404 to a bulky hydrophobic residue might block KANK1 binding by preventing Y48 engagement. We also noticed that S1641, one of the small residues that create the pocket, has been shown to be phosphorylated in vivo (Ratnikov et al., 2005) and might have a regulatory function in the KANK1-talin1 interaction. To test these hypotheses, we generated a series of point mutants in talin R7 and also in the KANK1 KN-domain, designed to disrupt the talinR7/KANK1 interaction. On the KANK1 side, we mutated the LDLD motif to AAAA, (the KANK1-4A mutant), while on the talin1 side, we generated a series of R7 mutants. These included G1404L, in which a bulky hydrophobic residue was introduced instead of glycine to occlude the hydrophobic pocket in R7, S1641E, a phosphomimetic mutant aimed to test the role of talin phosphorylation in regulating KANK1 binding, and W1630A, a substitution that would remove one of the molecular rulers holding α2 and α5 helices apart at a fixed distance. These mutants were introduced into talin1 R7-R8 and the structural integrity of the mutated proteins confirmed using NMR (Figure 2—figure supplement 1). The relative binding affinities were measured using an in vitro fluorescence polarization assay. In this assay, the KANK1(30–60) peptide is fluorescently labeled with BODIPY and titrated with an increasing concentration of talin R7-R8, and the binding between the two polypeptides results in an increase in the fluorescence polarization signal (Figure 2F). The KANK1-4A mutant abolished binding to talin (Figure 2C,F). The S1641E mutant had only a small effect on binding (Figure 2G), suggesting that either talin1 phosphorylation does not play a major role in modulating the interaction with KANK1 or that the S-E mutation is not a good phosphomimetic, possibly because phosphorylation might also affect helix formation integrity, an effect not mimicked by a single aspartate residue. However, strikingly, both the W1630A and the G1404L mutants abolished binding of KANK1 to talin R7 (Figure 2G), confirming the validity of our model. Finally, we also tested whether the KN-R7 interaction is conserved in talin2 and KANK2, and found that this was indeed the case (Figure 2—figure supplement 2). We conclude that the conserved KN domain of KANKs is a talin-binding site. Talin1-KANK1 interaction controls cortical organization of CMSC components Next, we set out to test the importance of the identified interactions in a cellular context by using the KANK1-4A and the talin G1404L mutants. We chose the G1404L talin mutant over W1630A for our cellular studies, because removing the bulky tryptophan from the hydrophobic core of the R7 might have the off target effect of perturbing the mechanical stability of R7, and our recent studies showed that the mechanostability of R7 is important for protecting R8 from force-induced talin extension (Yao et al., 2016). As could be expected based on the binding experiments with purified protein fragments, the 4A mutation reduced the interaction of the full-length KANK1 with talin1 in a pull-down assay (Figure 3A). An even stronger reduction was observed when KANK-△CC or the KN alone were tested (Figure 3A). Furthermore, the introduction of the G1404L mutation abrogated the interaction of full-length talin1 or its R7 fragment with full-length KANK1 (Figure 3B). Figure 3 with 1 supplement see all Download asset Open asset KANK1-talin interaction is required for recruiting KANK1 to FAs. (A) Streptavidin pull-down assays with the BioGFP-tagged KANK1 or the indicated KANK1 mutants, co-expressed with GFP-talin1 in HEK293T cells, analyzed by Western blotting with the indicated antibodies. (B) Streptavidin pull down assays with the BioGFP-tagged talin1 or the indicated talin1 mutants, co-expressed with HA-KANK1 in HEK293T cells, analyzed by Western blotting with the indicated antibodies. (C) TIRFM images of live HeLa cells depleted of KANK1 and KANK2 and co-expressing the indicated siRNA-resistant GFP-KANK1 fusions and TagRFP-paxillin. (D) Fluorescence profile of GFP-tagged mutants and TagRFP-paxillin based on line scan measurement across the FA area in TIRFM images as in panel (C). (E) Widefield fluorescence images of HeLa cells depleted of endogenous talin1 and talin2, rescued by the expression of the wild type GFP-tagged mouse talin1 or its G1404L mutant and labeled for endogenous KANK1 by immunofluorescence staining. (F) Quantification of peripheral clustering of KANK1 in cells treated and analyzed as in (E) (n=12, 6 cells per condition). Error bar, SEM; ***p<0.001, Mann-Whitney U test. https://doi.org/10.7554/eLife.18124.012 Figure 3—source data 1 An Excel sheet with numerical data on the quantification of peripheral clustering of KANK1 represented as a plot in Figure 3F. https://doi.org/10.7554/eLife.18124.013 Download elife-18124-fig3-data1-v3.xlsx" @default.
W2983455466 created "2019-11-22" @default.
W2983455466 creator A5004861283 @default.
W2983455466 creator A5013584337 @default.
W2983455466 creator A5014194467 @default.
W2983455466 creator A5018685938 @default.
W2983455466 creator A5024109873 @default.
W2983455466 creator A5025238408 @default.
W2983455466 creator A5030030700 @default.
W2983455466 creator A5058146964 @default.
W2983455466 creator A5061174255 @default.
W2983455466 creator A5078115532 @default.
W2983455466 date "2016-07-11" @default.
W2983455466 modified "2023-09-23" @default.
W2983455466 title "Author response: Talin-KANK1 interaction controls the recruitment of cortical microtubule stabilizing complexes to focal adhesions" @default.
W2983455466 doi "https://doi.org/10.7554/elife.18124.022" @default.
W2983455466 hasPublicationYear "2016" @default.
W2983455466 type Work @default.
W2983455466 sameAs 2983455466 @default.
W2983455466 citedByCount "2" @default.
W2983455466 countsByYear W29834554662020 @default.
W2983455466 countsByYear W29834554662021 @default.
W2983455466 crossrefType "peer-review" @default.
W2983455466 hasAuthorship W2983455466A5004861283 @default.
W2983455466 hasAuthorship W2983455466A5013584337 @default.
W2983455466 hasAuthorship W2983455466A5014194467 @default.
W2983455466 hasAuthorship W2983455466A5018685938 @default.
W2983455466 hasAuthorship W2983455466A5024109873 @default.
W2983455466 hasAuthorship W2983455466A5025238408 @default.
W2983455466 hasAuthorship W2983455466A5030030700 @default.
W2983455466 hasAuthorship W2983455466A5058146964 @default.
W2983455466 hasAuthorship W2983455466A5061174255 @default.
W2983455466 hasAuthorship W2983455466A5078115532 @default.
W2983455466 hasBestOaLocation W29834554661 @default.
W2983455466 hasConcept C151166631 @default.
W2983455466 hasConcept C169760540 @default.
W2983455466 hasConcept C20418707 @default.
W2983455466 hasConcept C62478195 @default.
W2983455466 hasConcept C86803240 @default.
W2983455466 hasConcept C95444343 @default.
W2983455466 hasConceptScore W2983455466C151166631 @default.
W2983455466 hasConceptScore W2983455466C169760540 @default.
W2983455466 hasConceptScore W2983455466C20418707 @default.
W2983455466 hasConceptScore W2983455466C62478195 @default.
W2983455466 hasConceptScore W2983455466C86803240 @default.
W2983455466 hasConceptScore W2983455466C95444343 @default.
W2983455466 hasLocation W29834554661 @default.
W2983455466 hasOpenAccess W2983455466 @default.
W2983455466 hasPrimaryLocation W29834554661 @default.
W2983455466 hasRelatedWork W1980943505 @default.
W2983455466 hasRelatedWork W1987704865 @default.
W2983455466 hasRelatedWork W1991143970 @default.
W2983455466 hasRelatedWork W2102283558 @default.
W2983455466 hasRelatedWork W2971075995 @default.
W2983455466 hasRelatedWork W2979646326 @default.
W2983455466 hasRelatedWork W2996317077 @default.
W2983455466 hasRelatedWork W3128989310 @default.
W2983455466 hasRelatedWork W4225324471 @default.
W2983455466 hasRelatedWork W4285094362 @default.
W2983455466 isParatext "false" @default.
W2983455466 isRetracted "false" @default.
W2983455466 magId "2983455466" @default.
W2983455466 workType "peer-review" @default.