FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2987842391> ?p ?o ?g. }

• W2987842391 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 Class III myosins (Myo3) and actin-bundling protein Espin play critical roles in regulating the development and maintenance of stereocilia in vertebrate hair cells, and their defects cause hereditary hearing impairments. Myo3 interacts with Espin1 through its tail homology I motif (THDI), however it is not clear how Myo3 specifically acts through Espin1 to regulate the actin bundle assembly and stabilization. Here we discover that Myo3 THDI contains a pair of repeat sequences capable of independently and strongly binding to the ankyrin repeats of Espin1, revealing an unexpected Myo3-mediated cross-linking mechanism of Espin1. The structures of Myo3 in complex with Espin1 not only elucidate the mechanism of the binding, but also reveal a Myo3-induced release of Espin1 auto-inhibition mechanism. We also provide evidence that Myo3-mediated cross-linking can further promote actin fiber bundling activity of Espin1. https://doi.org/10.7554/eLife.12856.001 eLife digest A mammal’s sense of hearing and balance are helped by so-called hair cells within the inner ear. These cells are named after their long, hair-like protrusions called stereocilia, and mutations in the genes involved in stereocilia development and maintenance can lead to hearing loss in humans. Damage to the stereocilia caused by excessive exposure to loud noises can also have the same effect. Stereocilia are full of filaments made of a protein called actin. Other proteins called class III myosins and Espin are also both required for normal development of stereocilia. Mutations in the genes that encode these proteins can cause hereditary deafness in humans. However, it remains unclear exactly how these two proteins (myosins and Espin) interact with each other in stereocilia. Using biochemical and structural studies, Liu, Li et al. have now discovered that the so-called ‘tail’ part of the myosins contains a pair of repeated sequences that can each interact with an Espin protein called Espin1. This interaction allows each myosin to cross-link two Espin1 proteins. Espins assemble actin filaments into bundles, and further experiments showed that this cross-linking interaction between myosins and Espins helped this process, which is linked to stereocilia development and maintenance. Mammals actually have two related copies of class III myosins that play overlapping but slightly different roles. The next challenge will be to try to understand the differences between these related proteins, as well as to try to uncover the roles of other forms of Espin in stereocilia. https://doi.org/10.7554/eLife.12856.002 Introduction Class III myosins (Myo3), together with class IX myosins, are two special groups of the myosin superfamily as these two sub-families of actin motors contain enzymatically active domains and thus are regarded as motorized signaling molecules (Bähler, 2000). The first member of Myo3 was identified in Drosophila photoreceptors and named as NinaC (neither inactivation nor afterpotential C) (Montell and Rubin, 1988). There are two paralogs of Myo3 in vertebrate, Myo3a and Myo3b, both of which are known to express in vertebrate retina and cochlea (Dose and Burnside, 2000, 2002; Shin et al., 2013). It is believed that they may play partially redundant roles as transporters that are crucial for vertebrate photoreceptor and stereocilia ultrastructure maintenance (Manor et al., 2012; Mecklenburg et al., 2015; Merritt et al., 2012). Myo3 across different species all contain an N-terminal S/T kinase domain before their motor head. The kinase domain has been reported to regulate the motor’s ATPase activity (Komaba et al., 2010; Quintero et al., 2010). The tail regions of Myo3 from different species are less conserved. Drosophila NinaC contains a PDZ binding motif at its very C-terminus capable of binding to a master scaffold protein called INAD (Inactivation no afterpotential D) (Wes et al., 1999). Vertebrate Myo3 tails share a conserved vertebrate specific domain referred to as tail homology I motif (THDI) (Figure 1A) (Dose et al., 2003). The THDI mediates binding of Myo3 to its cargo protein Espin1 (Ectoplasmic specialization protein 1) and allows Myo3 to transport Espin1 to the tips of actin bundle-based structures such as filopodia and stereocilia. Once tip localized, Espin1 WH2 domain promotes the elongation of actin protrusions (Merritt et al., 2012; Salles et al., 2009). However, the detailed molecular basis governing the Myo3 and Espin1 interaction is not clear. Figure 1 with 2 supplements see all Download asset Open asset Biochemical characterizations of the Myo3/Espin1 interaction. (A) Domain organizations of Espin1, Myo3a and Myo3b. (B) Sequence alignment of THDI of Myo3a and Myo3b showing that there are a pair of repeating sequences within THDI, which we term as ARB1 and ARB2. Hs, human; Mm, mouse; Gg, chicken; Xt, Xenopus tropicalis; Dr, Danio rerio. (C) ITC results showing that Myo3b-ARB12 (C1) as well as each individual site (C2 for ARB1 and C3 for ARB2) can bind to Espin1-AR with strong affinities. (D) FPLC-MALS showing that ARB12 and Espin1-AR form a 1:2 complex. https://doi.org/10.7554/eLife.12856.003 Espin1 was first identified in Sertoli cell-spermatid junctions (Bartles et al., 1996), encoded by the gene Espin. Later, shorter spliced isoforms of Espin gene products (Espin2B, Espin3A and Espin4) were shown to be expressed in other F-actin rich structures such as brush border microvilli and Purkinje cell dendritic spines (Bartles et al., 1998; Sekerkova et al., 2003). They share a common 14 kDa C-terminal actin binding domain (ABD; Figure 1A), which was reported to be necessary and sufficient for F-actin bundling activity (Bartles, 2000; Bartles et al., 1998). Besides the ABD, all Espin isoforms contain a WH2 motif which can bind to actin monomer and a proline rich (PR) region which can interact with profilins (Sekerkova et al., 2006). Espin2B contains one more PR region and an extra actin binding site (xAB) at the N-terminus (Chen et al., 1999). Espin1 is the longest isoform in the family and contains a stretch of ankyrin repeats (AR) in its N-terminus (Figure 1A). The AR of Espin1 is responsible for directly interacting with Myo3 tail THDI (Merritt et al., 2012; Salles et al., 2009). Recently, it was reported that Espin1 contains an AR binding sequence immediately C-terminal to xAB, and binding of AR to this sequence prevents xAB from binding to actin (Zheng et al., 2014). As such, the actin binding activity of the N-terminal part of Espin1 is auto-inhibited, and this AR-binding region is named as the auto-inhibitory region (AI, Figure 1A). The authors also proposed that Myo3 binding can release the auto-inhibition and increase the diameter of Espin1-promoted actin bundles (Zheng et al., 2014). Myo3a/Myo3b are known to co-localize with Espin1 at the tips of stereocilia of hair cells (Merritt et al., 2012; Salles et al., 2009; Schneider et al., 2006). Importantly, co-expression of Myo3a and Espin1 can further stimulate elongation of stereocilia in cultured organ of Corti hair cells (Salles et al., 2009). When expressed in heterologous cells, Myo3a gets enriched at the tip of filopodia (Les Erickson et al., 2003; Salles et al., 2009; Schneider et al., 2006). Mutations of human Myo3a gene is known to cause progressive non-syndromic hearing loss, DFNB30 (Walsh et al., 2002). Also, transgenic mice with DFNB30 mutation undergo age-dependent outer hair cell degeneration (Walsh et al., 2011). The jerker mouse carrying espin frame-shift mutation suffers from hair cells degeneration, deafness and vestibular dysfunction (Sekerkova et al., 2011; Zheng et al., 2000). Mutations in espin were also reported to be associated with non-syndromic autosomal recessive deafness DFNB36 and non-syndromic autosomal dominant deafness (Boulouiz et al., 2008; Donaudy et al., 2006; Naz et al., 2004). A prominent phenotype of the jerker mice is that their hair cell stereocilia are uniformly thinner and shorter, degenerate faster than those of the wild type littermates (Sekerkova et al., 2011; Zheng et al., 2000). These genetic findings convincingly point to critical roles of espin in stereocilia development and maintenance, likely by promoting the assembly and stabilization of parallel actin filament bundles in stereocilia (Bartles, 2000; Sekerkova et al., 2006). Here, we discover that both Myo3a and Myo3b contain two highly similar repeat sequences in their THDI region, each capable of independently binding to Espin1-AR with high affinity. The high resolution crystal structures of each of the two binding sequences from Myo3b in complex with Espin1-AR not only reveal the molecular basis governing the specific interaction between Myo3 and Espin1, but also allow us to discern the Myo3-mediated release of the auto-inhibition mechanism of Espin1. Based on these structural findings, we predict that binding of Myo3 to Espin1 can cluster Espin1 and thus enable Espin1 to further cross-link actin filaments into higher order fibers. Consistent with this prediction, we demonstrate by electron and fluorescence microscopic studies that binding of Myo3 to Espin1 can further promote formation of Espin1-mediated thicker actin bundles. Results The tail of Myo3 contains two independent Espin1-AR binding repeat sequences First we analyzed the sequences of the reported Espin1-binding THDI regions of both Myo3a and Myo3b, and found that the region contains a pair of repeating sequences in its N- and C-terminal halves (Figure 1A and B, denoted as ARB1 and ARB2 for Espin1 ankyrin repeats binding region 1 and 2 as detailed below). Using isothermal titration calorimetry (ITC)-based quantitative binding assay, we found that purified THDI from both Myo3a and Myo3b can bind to Espin1-AR with high affinities (Kd in the range of tens of nanomolars; Figure 1C1 and Figure 1—figure supplement 1A). Inspection of the ITC-based titration curves indicated that the binding stoichiometry of the Myo3 THDI and Espin1-AR clearly deviates from the value of 1:1 (Figure 1C1 and Figure 1—figure supplement 1A). We thus hypothesized that each of the two repeat sequences in Myo3 THDI may independently bind to Espin1-AR, forming a 2:1 stoichiometric complex. We verified this prediction by gel filtration chromatography and static light scattering experiments (Figure 1—figure supplement 2 and Figure 1D). In the gel filtration analysis, addition of an equivalent molar amount of Espin1-AR to Myo3b-THDI resulted in a complex peak with a smaller elution volume (Figure 1—figure supplement 2A). Addition of another molar equivalent of Espin1-AR further shifted the complex peak to a smaller elution volume (Figure 1—figure supplement 2B). However, further addition of Espin1-AR did not change the elution volume of the complex any more, indicating that Myo3b THDI is saturated by the binding of two molar ratios Espin1-AR (Figure 1—figure supplement 2C). To accurately determine the stoichiometry, we used fast protein liquid chromatography (FPLC) coupled with multi-angle light scattering (FPLC-MALS) to calculate the molecular mass of the Myo3b THDI and Espin1-AR complex. When mixed Myo3b THDI with saturated amount of Trx-tagged Espin1-AR, the fitted molecular weight of the complex peak (~109.5 kDa) matches well with the theoretical molecular weight of 117 kDa for the (Trx-Espin1-AR)2/Myo3b-THDI complex (Figure 1D), confirming that Myo3 THDI contains two Espin1-AR binding sites. Next, we divided Myo3b-THDI into two fragments, each corresponding to ARB1 and ARB2 as shown in Figure 1B. Both Myo3 ARB1 and ARB2 bind to Espin1-AR with affinities also in the range of tens of nanomolars and each with a 1:1 stoichiometry (Figure 1C2 and 1C3, and Figure 1—figure supplement 1B and C), indicating that the two repeating sequences in Myo3-THDI can independently bind to Espin1-AR with comparable affinities. The 1:2 stoichiometry between Myo3 and Espin1 is consistent with a previous finding that human Myo3a THDI lacking exon 33 (exon 33 mainly encodes ARB2) can still interact with Espin1 (Salles et al., 2009). The overall structure of the Myo3/Espin1-AR complex To understand the molecular basis of the Espin1/Myo3 interaction, we solved the crystal structure of the Espin1-AR/Myo3b-ARB1 complex at 1.65 Å resolution (Table 1). The structure revealed that Espin1-AR contains 10 ANK repeats (Figure 2), instead of 8 as predicted earlier (Bartles et al., 1996). The repeats 2–9 each contains the signature ‘TPLH’ sequence at the N-terminus of the αA helix, so can be viewed as the canonical ANK repeats. Like shown in the recently determined structures of the 24 ANK repeats scaffold protein ankyrin-B (Wang et al., 2014) and the 9 ANK repeats RNase L (Han et al., 2012), the two ANK repeats capping the two termini of Espin1-AR do not contain the ‘TPLH’ sequence (Figure 2—figure supplement 1). We believe that these two non-canonical ANK repeats capping the two termini of Espin1-AR mainly play a structural stabilization role of the entire AR fold. The 10 ANK repeats form a left-handed superhelix with the αA helices forming the inner groove and the αB helices forming the outer surface (Figure 2—figure supplement 2). Clear additional electron densities lining the inner groove of the ANK repeats allowed us to build the bound Myo3b-ARB1 peptide model with high confidence (Figure 2A). The ARB1 binds to Espin1-AR in an antiparallel manner, similar to the binding mode between ANK repeats from Ankyrin R/B/G and their targets (Wang et al., 2014) as well as between ANKRA2/RFXANK and their targets (Xu et al., 2012), suggesting that elongated inner grooves are common target binding sites of ANK repeats in general. The ARB1 spans nearly the entire inner groove, covering ~1230 Å2 of solvent accessible area. The N-terminal of ARB1 adopts an extended structure and binds to the C-terminal of Espin1-AR. The C-terminal of ARB1 forms an α-helix and binds to the N-terminal half of Espin1-AR (Figure 2B). The amino acid sequences of Espin1-AR from different vertebrate species as well as of the mammalian paralogs Espin-like proteins (Shin et al., 2013) are highly conserved (Figure 2—figure supplement 1). We mapped the sequence conservation profile to the structure of Espin1-AR and found that the residues in the inner groove of AR are highly conserved. In particular the residues in the ARB1 binding surface are essentially totally conserved (Figure 2C). Table 1 Statistics of X-ray Crystallographic Data Collection and Model refinement Numbers in parentheses represent the value for the highest resolution shell. a. Rmerge=Σ Ii- <I> / ΣIi, where Ii is the intensity of measured reflection and <I> is the mean intensity of all symmetry-related reflections. b. Rcryst=Σ Fcalc – Fobs /ΣFobs, where Fobs and Fcalc are observed and calculated structure factors. c. Rfree= ΣTFcalc – Fobs /ΣFobs, where T is a test data set of about 5% of the total unique reflections randomly chosen and set aside prior to refinement. d. B factors and Ramachandran plot statistics are calculated using MOLPROBITY (Chen et al., 2010). e. CC* and CC1/2 were defined by Karplus and Diederichs (Karplus and Diederichs, 2012). https://doi.org/10.7554/eLife.12856.006 Data setsEspin1-AR/Myo3b-ARB1 5ET1 Espin1-AR/Myo3b-ARB2 5ET0 Space group P21 P2 Wavelength (Å) 0.9791 0.9795 Unit Cell Parameters (Å) a=72.74, b=71.14, c=76.88 α=γ=90°, β=96.88° a=39.74, b=68.78, c=173.45 α=γ=90°, β=90.04° Resolution range (Å) 50-1.65 (1.68–1.65) 50-2.30 (2.42–2.30) No. of unique reflections 93433 (4625) 39636 (5866) Redundancy 3.7 (3.7) 3.7 (3.8) I/σ 18.5 (1.7) 7.7 (1.9) Completeness (%) 99.8 (99.9) 94.9 (96.6) Rmergea (%) 8.9 (91.6) 10.3 (79.9) CC* for the highest resolution shell e 0.866 0.878 CCi/2 for the highest resolution shell e 0.599 0.627 Structure refinement Resolution (Å) 50-1.65 (1.71–1.65) 10-2.3 (2.38–2.30) Rcryst b/Rfree c (%) 16.94/19.11 (25.77/28.64) 22.32/25.34 (26.74/30.90) rmsd bonds (Å) / angles (°) 0.006 / 0.795 0.010 / 1.113 Average B factor (Å2) d 23.2 60.5 No. of atoms Protein atoms 5374 4985 Water 378 23 Ligands 30 0 No. of reflections Working set 89061 37660 Test set 4345 1925 Ramachandran plot regions d Favored (%) 98.9 98.4 Allowed (%) 1.1 1.6 Outliers (%) 0 0 Figure 2 with 2 supplements see all Download asset Open asset The overall structure of the Myo3-ARB/Espin1-AR complex. (A) An omit map showing the binding of Myo3b-ARB1 to Espin1-AR. The Fo-Fc density map was generated by deleting the Myo3b-ARB1 part from the final model and contoured at 3.0σ. The Myo3b-ARB1 fitting the electron density is displayed in the stick model. (B) The overall structure of the Myo3b-ARB1/Espin1-AR complex. The Espin1-AR is shown in cylinders, Myo3b-ARB1 is shown with the ribbon diagram and colored in magenta. (C) The amino acid conservation map of Espin1-AR. The conservation map was calculated based on the sequence alignment of vertebrate Espin1 and mammalian Espin-like proteins shown in Figure 2—figure supplement 1. The identical residues are colored in dark blue; the strongly similar residues are colored in blue; the weakly similar residues are colored in light blue; the variable residues are colored in white. (D) The overall structure of Myo3b-ARB2/Espin1-AR complex. The Espin1-AR is shown in cylinders, Myo3b-ARB2 is shown in ribbon diagram and colored in dark purple. https://doi.org/10.7554/eLife.12856.007 We have also determined the Espin1-AR/Myo3b-ARB2 complex at a resolution of 2.3 Å (Table 1). The structure of the complex and the binding mode of Myo3b-ARB2 to Espin1-AR are highly similar to what are observed in the Espin1-AR/Myo3b-ARB1 complex (Figure 2D), directly confirming our earlier conclusion that the two repeat sequences in Myo3-THDI bind to Espin1-AR with similar binding mode and affinity. We tried very hard to crystallize the Espin1-AR/Myo3-THDI complexes without success, presumably due to flexibilities of the connection sequences between ARB1 and ARB2 of Myo3. The detailed Myo3/Espin1-AR interaction Since the two complex structures are essentially the same, here we only describe the detailed interactions observed in the Espin1-AR/Myo3b-ARB1 structure, which was resolved at a higher resolution. The Espin1-AR/Myo3b-ARB1 interface can be arbitrarily divided into three regions (Figure 3A). The first binding site is formed by the repeats 2–4 of Espin1-AR and binds to the C-terminal α-helix of ARB1 (Figure 3A1). Two absolutely conserved tyrosine residues (Tyr1267ARB1 and Tyr1268ARB1, the double tyrosine (‘YY’) motif; Figure 3B) and Leu1271ARB1 in the next turn insert into the hydrophobic pocket in the N-terminal of Espin1-AR (Figure 3A1). In addition, the hydroxyl groups of the ‘YY’ motif also make hydrogen bonds. Mutations of these two tyrosine residues to alanine greatly decreased Myo3b-ARB1’s binding to Espin1-AR (Figure 3C, Figure 3—figure supplement 1B). Similarly, mutation of Leu110 in the Espin1-AR hydrophobic pocket to a polar residue aspartic acid decreased the affinity by ~10-fold (Figure 3C, Figure 3—figure supplement 1C). Furthermore, the carboxyl group of Asp1264ARB1 makes hydrogen bonds with Asn69 and Ser103. Mutation of this residue together with Glu1263ARB1 to alanines decreased the affinity by ~10-fold (Figure 3C, Figure 3—figure supplement 1D). The second region is composed of Espin1 repeats 5–8 and binds to the middle-stretch of ARB1 with an extended conformation (Figure 3A2). Both the side chains and backbone carbonyl of Leu1259ARB1 are involved in the interaction. Asp205 located in the finger loop between repeat 6 and 7 forms a salt bridge with Lys1257ARB1 and a hydrogen bond with Gln1254ARB1. Mutation of Lys1257ARB1 into a reversed charged residue glutamic acid, together with Leu1259ARB1 to alanine substitution also decreased the affinity by about ~10-fold (Figure 3C, Figure 3—figure supplement 1E). The third region involves inner groove of the Espin1-AR repeats 8–10, which is highly enriched with negatively charged residues (Figure 3A3). The two highly conserved Arg residues at the beginning of Myo3b-ARB1 (Arg1251ARB1 and Arg1252ARB1) insert into the negatively charged pocket (Figure 3A3). Mutating these two arginine residues to reverse charged residues glutamic acid decreased Myo3B-ARB1 binding to Espin1-AR by ~15-fold (Figure 3C, Figure 3—figure supplement 1F). By analyzing the sequence of the ARBs from Myo3, we found that there exist more positively charged residues in addition to the highly conserved Arg residues at the further N-terminal end of ARB2 from both Myo3a and Myo3b (Figure 3B). We anticipate that these additional positively charged residues might also be involved in the binding, as there remain unoccupied, negatively charged surfaces in the third region of the Espin1-AR/Myo3b-ARB1 structure (Figure 3A3). Indeed, substitutions of the more N-terminal positively charged residues of Myo3b-ARB2 (Arg1282ARB2 and Lys1283ARB2) with alanines decreased its binding to Espin1-AR by ~10-fold (Figure 3C, Figure 3—figure supplement 1I). Figure 3 with 2 supplements see all Download asset Open asset The detailed Myo3/Espin1-AR interaction. (A) The Myo3b-ARB1/Espin1-AR interface is divided into three regions corresponding to the ‘YY’ motif (A1), the ‘KxL’ motif (A2) and the N-terminal positively charged residues (A3) of Myo3b-ARB1. The residues tested with the mutagenesis experiments are highlighted with boxes. The side chains or main chains of the residues involved in the interactions are highlighted in the stick model. Charge-charge and hydrogen bonding interaction are highlighted by dashed lines. The electrostatic surface potentials were calculated using PyMol. (B) Sequence alignment of Myo3-ARBs showing the conservation of ARBs. The conserved residues involved in the binding are highlighted with solid green triangles. The variable residues involved in the binding are highlighted with solid blue triangles. The two positively charged residues in ARB2 that are not resolved in the structure are highlighted with unfilled triangles. The sequence logo beneath the alignment was generated using WebLogo (Crooks et al., 2004). (C) ITC derived dissociation constants showing that mutations of the critical residues in the interface invariably weakened the binding. The original ITC data are shown in Figure 3—figure supplement 1. https://doi.org/10.7554/eLife.12856.010 Comparing the structures of Espin1-AR in complex with Myo3b-ARB1 and Myo3b-ARB2, the ‘YY’ motif and the ‘KxL’ motif are essentially in the same places (Figure 3—figure supplement 2B and C). Despite the high similarity, there are still a few minor differences. First of all, the C-terminal α-helix of ARB2 is shorter. The interaction is mediated by a hydrogen bond between Asp1294 and Tyr144, instead of the more extensive interaction observed in Myo3b-ARB1 complex (Figure 3—figure supplement 2B). Moreover, the positively charged residues in the N-terminus of ARB2 cannot be reliably built, probably due to the high salt concentration in the crystallization buffer (1.6 M ammonium sulfate). Nonetheless, clear electron density can be observed near the negatively charged surface (Figure 3—figure supplement 2A). Indeed, substitutions of these positively charged residues with Ala weakened the binding (Figure 3C and Figure 3—figure supplement 1G–I). Furthermore, the involvement of more positively charged residues of ARB2 may compensate for the less extensive interaction in its shorter C-terminal helix, thus resulting in a similar binding affinity to Espin1-AR as ARB1 does (107 nM for ARB1 vs 53 nM for ARB2, Figure 1C2 and C3). Binding of Myo3 releases the auto-inhibition of Espin1 It was reported that a conserved region following the xAB segment of Espin1 can interact with the N-terminal AR (Figure 4A) and inhibit the actin binding activity of xAB (Zheng et al., 2014). By comparing the sequence of AI (aa 496–529) with the consensus sequence of Myo3-ARBs, we find that Espin1-AI bears high sequence homology with Myo3-ARBs (e.g. the completely conserved ‘YY’ motif, the central ‘KxL’ motif, and the N-terminal positively charged residues; Figure 4B). Thus, we predict that AI may bind to Espin1-AR with a similar binding mode as Myo3-ARBs do. We used ITC-based binding assay to test this prediction, and found that Espin1-AI can indeed bind to Espin1-AR, albeit with a more moderate affinity than Myo3-ARBs (Kd of 1.32 μM vs 0.05~0.1 μM) (Figure 4C1). Fully consistent with our structure-based sequence alignment analysis, substitutions of the two tyrosines in Espin1-AI to alanines greatly weakened its binding to Espin1-AR (Figure 4C2). Given that the AI segment (aa 496–529) is immediately C-terminal to xAB (aa 462–487) of Espin1 (Figure 4A), one might envision that the interaction between Espin1 AR and AI can conformationally mask the xAB’s actin binding activity and thus renders Espin1 in an auto-inhibited conformation. Figure 4 with 1 supplement see all Download asset Open asset Biochemical characterization of the Espin1 auto-inhibition. (A) Domain organization of Espin1 showing that the Espin1-AI in the middle may bind to Espin1-AR at the N-terminus. (B) Sequence alignment of Espin1-AI from different vertebrate species, and comparison of Espin1-AI with the consensus sequence of Myo3-ARBs as shown in Figure 3D. (C) ITC result showing that Espin1-AI binds to Espin1-AR with a moderate affinity (C1). Mutation of the ‘YY’ motif to alanine greatly decrease the binding (C2). (D) ITC results showing that Myo3a-ARB1 can still bind to Espin1-1-529 (D1) and Espin1-FL (D2) with a sub-micromolar affinity. In contrast, Myo3a-ARB1 binds to Espin1-AR (D3) and Espin1-1-494 with comparable strong affinities (D4). Panel D3 is the same as Figure 1—figure supplement 1B. https://doi.org/10.7554/eLife.12856.013 It was shown that a synthetic peptide encompassing the Myo3a-ARB1 sequence identified here can stimulate the actin binding activity of xAB (Zheng et al., 2014). Based on our analysis, the most likely mechanism for Myo3a/b-ARB1-mediated stimulation of xAB’s actin binding may be due to the release of AI binding from Espin1-AR by direct competition of Myo3a/b-ARB1 binding. We designed biochemistry experiments to support the above model. If Myo3-ARB1 can indeed compete with AI for binding to Espin1-AR, then Myo3-ARB1 must still be able to bind to the auto-inhibited Espin1 but with an affinity weaker than binding to the isolated Espin-1-AR. We obtained highly purified N-terminal auto-inhibitory fragment of Espin1 spanning from AR to AI (denoted as Espin1-1-529, Figure 4A) and the full length Espin1 (denoted as Espin1-FL), and found that both proteins exist as monomer in solution (Figure 4—figure supplement 1), indicating that Espin1 auto-inhibition is intra-molecular in nature. ITC-based assay further showed that Myo3a-ARB1 can indeed bind to both Espin1-1-529 and Espin1-FL and with a weaker binding affinity than binding to Espin1-AR (Figure 4D1–3), consistent with a partially blocked Espin1-AR binding groove by AI. We also noticed that the ITC titration reactions of Myo3a-ARB1 to Espin1-1-529 and Espin1-FL are endothermic (Figure 4D1 and 4D2) instead of the exothermic reactions between Myo3a-ARB1 titrating to Espin1-AR (Figure 4D3), further indicating that the binding of Myo3a-ARB1 to the auto-inhibited Espin1 is not a simple direct association process between ARB1 and AR. To provide further proof, we truncated Espin1 from the C-terminus just before the AI (i.e., aa 1–494, denoted as Espin1-1-494) and found that Espin1-1-494 binds to Myo3a-ARB1 with an affinity similar to that between Espin1-AR and Myo3a-ARB1 (Figure 4D3), indicating that AI is indeed responsible for the decreased binding of Espin1 to Myo3a-ARB1. Espin1 binding sites in Myo3 are critical for the filopodia tip localization of Espin1 and Myo3 Myo3a is known to localize at the tip of filopodia when transfected in heterologous cells like HeLa or COS7 cells, whereas Myo3b alone cannot tip-localize as it lacks ABM (Les Erickson et al., 2003; Salles et al., 2009). However, when co-expressed with Espin1, Myo3b can bind to Espin1 and localize to the tip of filopodia (Manor et al., 2012; Merritt et al., 2012). Similarly, Myo3a lacking ABM can only tip-localize when co-expressed with Espin1. Deletion of the Myo3 kinase domain is known to render the motor in a constant active state in promoting the length of filopodia, thus we used Myo3 constructs lacking the kinase domain in the subsequent experiments (Les Erickson et al., 2003; Quintero et al., 2010; 2013). We first tested the role of Myo3/Espin1 binding on Myo3a’s ability to tip-localize. To test our biochemical findings and to determine the impact of Myo3-ARB ‘YY’ motifs on Myo3-Espin1 interaction, Espin1 transportation (i.e., tip localization) and filopodia elongation, we co-expressed various Myo3aΔKΔABM (lacking the kinase domain and the ABM) and Myo3bΔK constructs with Espin1 in COS7 cells. Since both Myo3aΔKΔABM and Myo3bΔK cannot tip-localize by its own, we reasoned that when co-expressed with Espin1, the Myo3 and Espin1 tip localization levels will determine the intactness of their mutual binding. As expected, both GFP-tagged wild type Myo3a (lacking the kinase domain and the ABM, denoted as ΔKΔABM) and RFP-tagged Espin1 localized to the tip" @default.
• W2987842391 created "2019-11-22" @default.
• W2987842391 creator A5011160073 @default.
• W2987842391 creator A5017613292 @default.
• W2987842391 creator A5020062987 @default.
• W2987842391 creator A5021163747 @default.
• W2987842391 creator A5043887869 @default.
• W2987842391 creator A5051730015 @default.
• W2987842391 creator A5065075576 @default.
• W2987842391 creator A5071037763 @default.
• W2987842391 creator A5077511640 @default.
• W2987842391 creator A5084588183 @default.
• W2987842391 creator A5086300460 @default.
• W2987842391 creator A5088019634 @default.
• W2987842391 date "2016-01-07" @default.
• W2987842391 modified "2023-10-18" @default.
• W2987842391 title "Author response: Myosin III-mediated cross-linking and stimulation of actin bundling activity of Espin" @default.
• W2987842391 doi "https://doi.org/10.7554/elife.12856.021" @default.
• W2987842391 hasPublicationYear "2016" @default.
• W2987842391 type Work @default.
• W2987842391 sameAs 2987842391 @default.
• W2987842391 citedByCount "0" @default.
• W2987842391 crossrefType "peer-review" @default.
• W2987842391 hasAuthorship W2987842391A5011160073 @default.
• W2987842391 hasAuthorship W2987842391A5017613292 @default.
• W2987842391 hasAuthorship W2987842391A5020062987 @default.
• W2987842391 hasAuthorship W2987842391A5021163747 @default.
• W2987842391 hasAuthorship W2987842391A5043887869 @default.
• W2987842391 hasAuthorship W2987842391A5051730015 @default.
• W2987842391 hasAuthorship W2987842391A5065075576 @default.
• W2987842391 hasAuthorship W2987842391A5071037763 @default.
• W2987842391 hasAuthorship W2987842391A5077511640 @default.
• W2987842391 hasAuthorship W2987842391A5084588183 @default.
• W2987842391 hasAuthorship W2987842391A5086300460 @default.
• W2987842391 hasAuthorship W2987842391A5088019634 @default.
• W2987842391 hasBestOaLocation W29878423911 @default.
• W2987842391 hasConcept C125705527 @default.
• W2987842391 hasConcept C169760540 @default.
• W2987842391 hasConcept C185592680 @default.
• W2987842391 hasConcept C24998067 @default.
• W2987842391 hasConcept C6997183 @default.
• W2987842391 hasConcept C86803240 @default.
• W2987842391 hasConcept C95444343 @default.
• W2987842391 hasConceptScore W2987842391C125705527 @default.
• W2987842391 hasConceptScore W2987842391C169760540 @default.
• W2987842391 hasConceptScore W2987842391C185592680 @default.
• W2987842391 hasConceptScore W2987842391C24998067 @default.
• W2987842391 hasConceptScore W2987842391C6997183 @default.
• W2987842391 hasConceptScore W2987842391C86803240 @default.
• W2987842391 hasConceptScore W2987842391C95444343 @default.
• W2987842391 hasLocation W29878423911 @default.
• W2987842391 hasOpenAccess W2987842391 @default.
• W2987842391 hasPrimaryLocation W29878423911 @default.
• W2987842391 hasRelatedWork W1485247088 @default.
• W2987842391 hasRelatedWork W1976626972 @default.
• W2987842391 hasRelatedWork W2002167899 @default.
• W2987842391 hasRelatedWork W2039079887 @default.
• W2987842391 hasRelatedWork W2068215542 @default.
• W2987842391 hasRelatedWork W2096246147 @default.
• W2987842391 hasRelatedWork W2102789738 @default.
• W2987842391 hasRelatedWork W3083875756 @default.
• W2987842391 hasRelatedWork W3105969802 @default.
• W2987842391 hasRelatedWork W3204034302 @default.
• W2987842391 isParatext "false" @default.
• W2987842391 isRetracted "false" @default.
• W2987842391 magId "2987842391" @default.
• W2987842391 workType "peer-review" @default.