Structural basis for competitive interactions of Pex14 with the import receptors Pex5 and Pex19
2009; Springer Nature; Volume: 28; Issue: 6 Linguagem: Inglês
10.1038/emboj.2009.7
ISSN1460-2075
AutoresChristian Neufeld, Fabian V. Filipp, Bernd Simon, Alexander Neuhaus, Nicole Schüller, Christine David, Hamed Kooshapur, Tobias Madl, Ralf Erdmann, Wolfgang Schliebs, Matthias Wilmanns, Michael Sattler,
Tópico(s)Insect Resistance and Genetics
ResumoArticle5 February 2009free access Structural basis for competitive interactions of Pex14 with the import receptors Pex5 and Pex19 Christian Neufeld Christian Neufeld EMBL Heidelberg, Heidelberg, Germany EMBL Hamburg Outstation, c/o DESY, Hamburg, Germany Search for more papers by this author Fabian V Filipp Fabian V Filipp EMBL Heidelberg, Heidelberg, GermanyPresent address: Department of Chemistry and Biochemistry, University of California, San Diego, La Jolla, CA 92093-0307, USA Search for more papers by this author Bernd Simon Bernd Simon EMBL Heidelberg, Heidelberg, Germany Search for more papers by this author Alexander Neuhaus Alexander Neuhaus Institute for Physiological Chemistry, Department of Systems Biology, Faculty of Medicine, Ruhr University of Bochum, Bochum, Germany Search for more papers by this author Nicole Schüller Nicole Schüller EMBL Hamburg Outstation, c/o DESY, Hamburg, Germany Search for more papers by this author Christine David Christine David Institute for Physiological Chemistry, Department of Systems Biology, Faculty of Medicine, Ruhr University of Bochum, Bochum, Germany Search for more papers by this author Hamed Kooshapur Hamed Kooshapur Helmholtz Zentrum München, Neuherberg, Germany Munich Center for Integrated Protein Science and Biomolecular NMR, Department Chemie, Technische Universität München, Garching, Germany Search for more papers by this author Tobias Madl Tobias Madl Helmholtz Zentrum München, Neuherberg, Germany Munich Center for Integrated Protein Science and Biomolecular NMR, Department Chemie, Technische Universität München, Garching, Germany Search for more papers by this author Ralf Erdmann Ralf Erdmann Institute for Physiological Chemistry, Department of Systems Biology, Faculty of Medicine, Ruhr University of Bochum, Bochum, Germany Search for more papers by this author Wolfgang Schliebs Wolfgang Schliebs Institute for Physiological Chemistry, Department of Systems Biology, Faculty of Medicine, Ruhr University of Bochum, Bochum, Germany Search for more papers by this author Matthias Wilmanns Matthias Wilmanns EMBL Hamburg Outstation, c/o DESY, Hamburg, Germany Search for more papers by this author Michael Sattler Corresponding Author Michael Sattler EMBL Heidelberg, Heidelberg, Germany Helmholtz Zentrum München, Neuherberg, Germany Munich Center for Integrated Protein Science and Biomolecular NMR, Department Chemie, Technische Universität München, Garching, Germany Search for more papers by this author Christian Neufeld Christian Neufeld EMBL Heidelberg, Heidelberg, Germany EMBL Hamburg Outstation, c/o DESY, Hamburg, Germany Search for more papers by this author Fabian V Filipp Fabian V Filipp EMBL Heidelberg, Heidelberg, GermanyPresent address: Department of Chemistry and Biochemistry, University of California, San Diego, La Jolla, CA 92093-0307, USA Search for more papers by this author Bernd Simon Bernd Simon EMBL Heidelberg, Heidelberg, Germany Search for more papers by this author Alexander Neuhaus Alexander Neuhaus Institute for Physiological Chemistry, Department of Systems Biology, Faculty of Medicine, Ruhr University of Bochum, Bochum, Germany Search for more papers by this author Nicole Schüller Nicole Schüller EMBL Hamburg Outstation, c/o DESY, Hamburg, Germany Search for more papers by this author Christine David Christine David Institute for Physiological Chemistry, Department of Systems Biology, Faculty of Medicine, Ruhr University of Bochum, Bochum, Germany Search for more papers by this author Hamed Kooshapur Hamed Kooshapur Helmholtz Zentrum München, Neuherberg, Germany Munich Center for Integrated Protein Science and Biomolecular NMR, Department Chemie, Technische Universität München, Garching, Germany Search for more papers by this author Tobias Madl Tobias Madl Helmholtz Zentrum München, Neuherberg, Germany Munich Center for Integrated Protein Science and Biomolecular NMR, Department Chemie, Technische Universität München, Garching, Germany Search for more papers by this author Ralf Erdmann Ralf Erdmann Institute for Physiological Chemistry, Department of Systems Biology, Faculty of Medicine, Ruhr University of Bochum, Bochum, Germany Search for more papers by this author Wolfgang Schliebs Wolfgang Schliebs Institute for Physiological Chemistry, Department of Systems Biology, Faculty of Medicine, Ruhr University of Bochum, Bochum, Germany Search for more papers by this author Matthias Wilmanns Matthias Wilmanns EMBL Hamburg Outstation, c/o DESY, Hamburg, Germany Search for more papers by this author Michael Sattler Corresponding Author Michael Sattler EMBL Heidelberg, Heidelberg, Germany Helmholtz Zentrum München, Neuherberg, Germany Munich Center for Integrated Protein Science and Biomolecular NMR, Department Chemie, Technische Universität München, Garching, Germany Search for more papers by this author Author Information Christian Neufeld1,2, Fabian V Filipp1, Bernd Simon1, Alexander Neuhaus3, Nicole Schüller2, Christine David3, Hamed Kooshapur4,5, Tobias Madl4,5, Ralf Erdmann3, Wolfgang Schliebs3, Matthias Wilmanns2 and Michael Sattler 1,4,5 1EMBL Heidelberg, Heidelberg, Germany 2EMBL Hamburg Outstation, c/o DESY, Hamburg, Germany 3Institute for Physiological Chemistry, Department of Systems Biology, Faculty of Medicine, Ruhr University of Bochum, Bochum, Germany 4Helmholtz Zentrum München, Neuherberg, Germany 5Munich Center for Integrated Protein Science and Biomolecular NMR, Department Chemie, Technische Universität München, Garching, Germany *Corresponding author. Institute of Structural Biology, Helmholtz Zentrum Muenchen, Ingolstaedter Landstr. 1, Neuherberg, 85764, Germany. Tel.: +49 89 28913418; Fax: +49 89 28913869; E-mail: [email protected] The EMBO Journal (2009)28:745-754https://doi.org/10.1038/emboj.2009.7 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Protein import into peroxisomes depends on a complex and dynamic network of protein–protein interactions. Pex14 is a central component of the peroxisomal import machinery and binds the soluble receptors Pex5 and Pex19, which have important function in the assembly of peroxisome matrix and membrane, respectively. We show that the N-terminal domain of Pex14, Pex14(N), adopts a three-helical fold. Pex5 and Pex19 ligand helices bind competitively to the same surface in Pex14(N) albeit with opposite directionality. The molecular recognition involves conserved aromatic side chains in the Pex5 WxxxF/Y motif and a newly identified F/YFxxxF sequence in Pex19. The Pex14–Pex5 complex structure reveals molecular details for a critical interaction in docking Pex5 to the peroxisomal membrane. We show that mutations of Pex14 residues located in the Pex5/Pex19 binding region disrupt Pex5 and/or Pex19 binding in vitro. The corresponding full-length Pex14 variants are impaired in peroxisomal membrane localisation in vivo, showing that the molecular interactions mediated by the N-terminal domain modulate peroxisomal targeting of Pex14. Introduction Peroxisomes are ubiquitous, single membrane cell organelles with a large variety of metabolic functions. Most, if not all proteins participating in peroxisomal biogenesis, collectively called peroxins or Pex proteins have been identified (Erdmann and Schliebs, 2005). Mutations in Pex proteins are implicated in diseases, such as the spectrum of Zellweger disorders (Wanders and Waterham, 2006). However, molecular details of the functional activity of peroxins and their involvement in these diseases are poorly understood. The membrane-associated protein Pex14 has been described as a central component of the translocation machinery for peroxisomal matrix enzymes. Because of its direct interaction with the peroxisome targeting signal (PTS) receptor Pex5, Pex14 has been proposed to serve as the docking site for the cytosolic receptor–cargo complex (Albertini et al, 1997). Pex14 sequences from different organisms display a common domain composition, consisting of a conserved N-terminal domain, a segment of hydrophobic amino acids and a coiled-coil region (Albertini et al, 1997; Shimizu et al, 1999; Will et al, 1999; Shimozawa et al, 2004). Mammalian Pex14 is an integral membrane protein, with its C-terminus exposed to the cytosol and a potential for homo-oligomerisation (Will et al, 1999; Otera et al, 2002; Itoh and Fujiki, 2006). The recognition of one or multiple aromatic WxxxF/Y motifs in Pex5 by the N-terminal Pex14 domain have an important function for docking of the Pex5 receptor to the peroxisomal membrane (Saidowsky et al, 2001; Otera et al, 2002; Choe et al, 2003; Williams et al, 2005). Pex19, a mainly cytosolic peroxin, has been proposed to serve as a translocation receptor and chaperone of newly synthesised peroxisomal membrane proteins (PMP) (Sacksteder et al, 2000). It recognises a composite peroxisomal membrane protein targeting signal (mPTS), which may involve different regions of the PMP proteins and—in contrast to PTSs PTS1/PTS2—cannot be represented by a simple consensus sequence (Rottensteiner et al, 2004; Halbach et al, 2005, 2006; Saveria et al, 2007). In contrast, binding of Pex19 to the PMP Pex14 depends on a small central region in Pex19 and a conserved N-terminal domain in Pex14 (Fransen et al, 2002), which does not contain an identifiable mPTS motif. Because both Pex5 and Pex19 bind to the same region in Pex14, it has been suggested that Pex19 might have a function in the peroxisomal matrix protein import (Fransen et al, 2004). Here, we present the molecular basis for the interaction of the N-terminal domain of Pex14, Pex14(N), with Pex5 and Pex19. We identified an F/YFxxxF motif in the N-terminus of Pex19, which binds to the same site in Pex14 as the Pex5 WxxxF/Y motif. The three-dimensional solution structures of Pex14(N)-ligand complexes provide molecular details for the recognition of conserved aromatic residues in the Pex5 and Pex19 ligand motifs and show that the two peptides bind to Pex14 with opposite directionality. Competitive NMR titration experiments and quantification of both interactions by isothermal titration calorimetry reveal that Pex5 binds with significantly higher affinity. The Pex14–Pex5 complex provides a structural basis for the docking of Pex5 to Pex14 at the peroxisomal membrane. Pex14(N) variants, which impair binding to Pex5 and/or Pex19 in vitro, exhibit a subcellular mislocalisation in vivo, suggesting that the binding interface contributes to the targeting of Pex14 to the peroxisomal membrane. Results Competitive binding of Pex5 and Pex19 peptides to Pex14 The N-terminal region of Pex14(N), comprising residues 16–80, was found to be resistant against exopeptidases (data not shown) and selected to study the interactions with peptide ligands derived from Pex5 and Pex19. For Pex5, three peptides of different length (residues 116–124, 113–127 and 108–127) where chosen. These peptides all comprise the first WxxxF/Y motif, representing the strongest Pex14 interaction site (Saidowsky et al, 2001). For Pex19, a peptide comprising residues 66–77 was selected (Figure 1A). It contains the sequence segment (67–72) that was shown to be critical for Pex14 binding (Fransen et al, 2005). Figure 1.NMR titrations of Pex14(N) with Pex19 and Pex5 peptides. (A) Schematic overview of the domain composition of human Pex5, Pex19 and Pex14. The N-terminal domain of Pex14 is coloured green, the binding motifs are shown in brown (Pex19) and gold (Pex5), respectively. (B) 1H,15N correlation spectra of 15N-labelled recombinant Pex14(N) free (black), and in complex with Pex19 (residues 66–77) (brown; left) and Pex5 (gold; right) (residues 116–124). (C) NMR chemical shift changes (Δδ=(δ15N2+δ1H2)1/2) of Pex14(N) in the presence of saturated concentrations of the Pex5 (gold) and the Pex19 (brown) ligands. Secondary structure elements are indicated underneath. (D) 15N-labelled recombinant Pex14(N) (black) was titrated with Pex19(66–77) (brown) to the point of saturation and was then cross-titrated with Pex5(116–124) (gold). Download figure Download PowerPoint The binding of the Pex19 and Pex5 ligands to Pex14(N) was first studied by NMR titration experiments. Unlabelled Pex5(116–124) or Pex19(66–77) peptides were titrated to saturation into a solution of 15N-labelled Pex14(16–80). Changes of amide chemical shifts were monitored in two-dimensional 1H, 15N-HSQC experiments (Figure 1B). Chemical shift assignments of Pex14(N) were obtained using standard methods (Sattler et al, 1999). Mapping of residues that experience significant chemical shift perturbations onto the primary sequence of Pex14 revealed that both peptides bind to an overlapping site involving helices α1 and α2 and the connecting linker (Figure 1C). The binding of Pex19 to Pex14(N) is fast on the NMR chemical shift timescale, as judged from the gradual change of the NMR signals in the HSQC spectra on addition of increasing amounts of ligand. In contrast, the appearance of new signals corresponding to the bound state during the titration of Pex14(N) with the Pex5 peptide shows binding in the slow exchange limit, indicative of a tighter interaction with the Pex5 ligand (Figure 1B). To confirm that the binding sites are overlapping, a double-titration experiment was performed, where the Pex19 peptide was added to saturation to 15N-labelled Pex14(N) followed by a cross-titration with the Pex5 ligand (Figure 1D). The chemical shifts observed at the endpoint of the cross-titration (molar ratios of 1:1.5:1.5 Pex14(N):Pex19:Pex5) are identical to those seen on addition of the Pex5 peptide alone. This shows that the Pex19 ligand is replaced by the stronger interaction of the Pex5 peptide with Pex14. The binding affinities of the peptides, as determined by isothermal titration calorimetry, reveal equilibrium dissociation constants (KD) of 0.47 μM for Pex5(116–124)–Pex14(N) interaction and 9 μM for Pex19–Pex14(N) interaction. Longer Pex5 peptides bind to Pex14(N) with higher affinities, that is, KD of 0.12 μM for Pex5(113–127) and 0.07 μM for Pex5(108–127) (Supplementary Table 1). This suggests that the residues flanking the aromatic core motif of Pex5 further contribute to the interaction. However, irrespective of the length of the Pex5 peptide, the binding affinity is considerably higher than that measured for the Pex19 peptide, thus supporting our data from Pex5/Pex19 binding competition experiments on Pex14(N). Three-dimensional structures of Pex14(N) bound to Pex5 and Pex19 peptides The three-dimensional structures of the N-terminal domain of Pex14 in complex with peptides derived from Pex5 (residues 108–127) and Pex19 (residues 66–77) were determined using heteronuclear NMR methods (Supplementary Table 2). The core region of Pex14(N) (residues 23–73) adopts a well-defined conformation, whereas N- and C-terminal flanking residues are flexible and do not display a defined secondary structure (Supplementary Figures 1 and 2). The structures of the Pex14–Pex5 and Pex14–Pex19 complexes are well defined by the NMR data, which include 130/112 intermolecular NOEs for the Pex5 and Pex19 complexes, respectively (Supplementary Table 2; Figure 2). Figure 2.Solution structures of the Pex14(N)–Pex5 and Pex14(N)–Pex19 complexes. (A) Stereo view of the backbone atoms of Pex14 (residues 20–76) in complex with Pex5(108–127). An NMR ensemble of the 10 lowest-energy structures (out of 100 calculated) is shown. Secondary structure elements in Pex14 (helices α1, α2, α3 and the helical linker connecting α1 and α2) are coloured in green. The peptide is shown in gold. (B) Ribbon diagram of the lowest-energy structure in (A). (C) Superposition of the backbone atoms of Pex14 (residues 19–76) in complex with Pex19(66–77). The ensemble shows 10 lowest-energy structures (out of 100 calculated). The peptide is shown in brown. (D) Ribbon presentation of the lowest-energy structure in (C). Download figure Download PowerPoint Pex14(N) comprises three α-helices, forming a three-helical bundle. Helices α1 and α2 are in an anti-parallel orientation (Figure 2), whereas helix α3 forms a scaffold diagonal across the pair of helices α1 and α2. A short helical turn is found in the linker connecting helix α1 and α2. Secondary chemical shifts of free Pex14(N) indicate that the secondary structure of the ligand-free Pex14(N) domain is very similar to the Pex5 or Pex19 bound form, with the exception of the helical turn in the α1–α2 linker, which is induced on ligand binding (Supplementary Figure 2). Overall, the Pex14(N) conformation in the Pex14(N)/Pex5 and Pex14(N)/Pex19 complexes is very similar, reflected by a low coordinate r.m.s.d of 1.1 Å when superimposing the Pex14(N) backbone. The size of the protein–peptide binding interface is ≈460 Å2 in both complexes. Comparison of the secondary chemical shifts of the free and Pex14(N) bound ligand peptides indicates that a helical conformation of the Pex5 peptide is present irrespective of Pex14(N) binding, whereas for the free Pex19 peptide, only a small fraction of helical conformation has been detected (data not shown). The binding site of Pex14(N) with Pex5 and Pex19 is formed by helices α1 and α2 (Figure 3A–C) and exhibits two hydrophobic pockets, which are separated by two aromatic residues (Phe35 and Phe52) (Figures 3, 4A and B). The two hydrophobic pockets are flanked by several basic amino acids (Arg25, Lys34, Arg40, Lys55 and Lys56), leading to a highly positively charged protein surface (Figure 3A). In contrast, the binding surfaces of the Pex5 and Pex19 ligands contain negatively charged surface patches (Figure 3D), suggesting that charge complementarity is a key determinant of the observed Pex14(N)–Pex5 and Pex14(N)–Pex19 interactions. The electrostatic surface representation of both complexes exhibits a three-layered arrangement, where the two positively charged helices α1 and α2 in Pex14(N) are flanked by the negatively charged helix α3 on one side and by the negatively charged Pex5 and Pex19 ligands on the other side (Figure 3C). Figure 3.Electrostatic surface representation of the Pex14–Pex5 and Pex14–Pex19 complexes. The Pex14(N)–Pex5 and Pex14(N)–Pex19 complex structures are shown in the top and bottom rows, respectively. (A) View onto the ligand-binding surface. The Pex5 and Pex19 ligand helices are shown as transparent ribbons. Blue and red colours indicate positive and negative electrostatic surface potential in Pex14. Positively charged residues in Pex14(N) surrounding the binding interface are labelled. (B) Ribbon representation of the Pex14(N)–Pex5 (top) and Pex14(N)–Pex19 complexes. The aromatic residues in the two ligands are shown and labelled. Pex14(N), Pex5 and Pex19 are coloured in green, gold and brown, respectively. (C) Surface representation of the two complexes shown in the same orientation as in (B). Blue and red colours indicate positive and negative electrostatic surface potential, respectively. (D) Surface representation of the Pex5 (top) and Pex19 (bottom) ligands bound to Pex14, as viewed from the Pex14 interaction surface. Download figure Download PowerPoint Figure 4.Structural details of the Pex14–Pex5 and Pex14–Pex19 interactions. Ribbon (left), surface (middle) and ensemble (right) representations of the molecular interface between Pex14 and Pex5(108–127) (A) and Pex19(66–77) (B). Left: Pex14 is shown in grey, the Pex5 and Pex19 peptides are coloured gold and brown, respectively. Residues showing intermolecular NOEs are shown in stick representation. Pex14 side chains are coloured in green, peptide backbone and side chains are in gold (Pex5) or brown (Pex19). Positively charged Pex14(N) residues are labelled white. An intermolecular salt bridge between Pex14(N) Lys56 and Pex5 Glu121 is indicated by a dashed red line. Labelling of Pex14 residues that were altered for mutational analysis are underlined. (C) Sequences of Pex5 and Pex19 peptide ligands used for binding and structural studies. The peptide fragments used for structure determination are shown with grey background. The Pex5 and Pex19 sequences are aligned based on the structural analysis and are shown with opposite directionality. Smaller Pex5 peptides are indicated by the lines above the sequence (see Supplementary Table 1). Pex5 Glu121, which forms an electrostatic contact with Pex14 Lys56, is highlighted in red. Conserved hydrophobic residues are highlighted in green, the two critical aromatic residues in both peptide motifs are marked with an asterisk. Download figure Download PowerPoint In the Pex14(N)–Pex5 complex, the Pex5 peptide adopts an amphipathic α-helical conformation and binds diagonally across helices α1 and α2 in Pex14(N) (Figures 2B and 4A). The conserved aromatic residues in the Pex5 WxxxF/Y motif, Trp118 and Phe122, are deeply buried into the two separate hydrophobic pockets in Pex14(N) (Figure 4A). The aromatic side chain of Pex5 Trp118 is stabilised by stacking on one side with the side chain of the highly conserved Pex14 Lys56 and on the other side with the methyl group of Pex14 Thr31. In addition, the NMR ensemble indicates a salt bridge between Pex5 Glu121 and the Pex14 Lys56 side chain. Pex5 Phe122 mainly interacts with two aromatic residues (Phe35 and Phe52) in Pex14 that separate the two binding pockets. In the Pex14(N)–Pex19 complex structure, the Pex19 peptide also forms an amphipathic helix that binds across α1 and α2 helices of Pex14, similar to the Pex5–Pex14 complex (Figures 2D and 4B). However, in contrast to the Pex14(N)–Pex5 complex, the Pex19 helix binds in an almost opposite orientation (Figure 2; Supplementary Figures 3 and 4). Because of the inverted orientation, Phe75 at the C-terminal side of the Pex19 F/YFxxxF motif occupies the binding pocket used by Pex5 Trp118 in the Pex14(N)–Pex5 complex. The Pex19 Phe75 side chain is stabilised by similar interactions as seen for Pex5 Trp118. In contrast, the second hydrophobic-binding pocket in Pex14(N) is distinct in the Pex5 and Pex19 complexes. Pex19 Phe71 interacts with Pex14 Phe35, Thr48 and Phe52, resembling the recognition of the corresponding Phe122 in Pex5. However, it is less deeply inserted into the binding pocket (Figures 3A and 4). The preceding Pex19 Phe70 residue contacts Pex14(N) in a neighboring region. In contrast to the corresponding L123 in the Pex5 complex, the aromatic side chain of Pex19 Phe70 is stabilised by a stacking interaction with the guanidinium group of Pex14 Arg 40 (Figure 4A and B). The orientation of the helical axes of the Pex19 and Pex5 ligands differs by 17° (Supplementary Figure 3). An exact opposite orientation would not allow efficient interactions of all three-aromatic residues in the Pex19 F/YFxxxF motif to both hydrophobic-binding pockets in Pex14 (not shown). To independently confirm the observed opposite orientation of the Pex19 peptide, we performed paramagnetic relaxation enhancement experiments, with a Pex19(S66C) mutant peptide that allows the covalent attachment of a paramagnetic proxyl group. Binding of the spin-labelled Pex19 variant to Pex14(N) leads to line broadening of the amide proton signals of Pex14(N) residues 40–56 (Supplementary Figure 4C). As spin-label effects strongly decrease with distance, and the relaxation enhancements induced are sensitive in detecting even small populations of conformations with close proximity to the spin label, a Pex5-like orientation of the Pex19 peptide can be excluded. Specificity of the Pex5 and Pex19 interactions analysed by peptide scan To investigate the structural requirements for the binding of Pex5 and Pex19 peptides by independent means, we identified critical residues in the respective interaction motifs by systematic mutational analysis. Pex14(N) was purified as a recombinant polyhistidine fusion protein and incubated with a cellulose membrane containing a set of peptide sequences in which each residue in the core binding motifs of Pex5(113–127) or Pex19(66–80) was substituted by all other 19 amino acids. Binding of Pex14(N) was detected by monoclonal anti-polyhistidine-tag antibodies (Figure 5). Figure 5.Mutational analysis of Pex14 and its ligands. N-terminal Pex14 was tested for interactions with variants of Pex5 (A) and Pex19 (B) peptides. Peptides comprising systematic variations of Pex5 (residues 113–127; ALSENWAQEFLAAGD) and Pex19 (residues 66–80; SQEKFFQELFDSELA) ligands were synthesised on cellulose membranes and incubated with purified His6-Pex14(1–80). Bound Pex14 was visualised immunochemically with monoclonal anti-His6 antibodies. Spots with reduced intensities represent peptides with reduced binding affinities for Pex14. Amino acids that retain the interaction are shown by their single letter code on the right, where the letter size indicates the relative contribution to the interaction at each sequence position. Green and magenta colours indicate hydrophobic or polar/small residues. Red and blue colours display negatively or positively charged residues, respectively. Download figure Download PowerPoint A common observation for both peptides is that Pex14(N) does not bind to any ligand containing a proline within the core binding region, consistent with the requirement of an α-helical conformation for Pex14 binding. As expected from our structural analysis, the aromatic residues within the Pex5 sequence are most sensitive for substitutions (Figure 5A). Trp118 and Phe or Tyr at position 122 are strictly required for Pex14(N) binding. Glu121 can be substituted with glutamine and methionine, whereas aspartate at this position, for instance, impairs Pex14(N) binding. Replacement of Leu123 is restricted to other hydrophobic residues. The mutational analysis of the Pex14(N)–Pex19 interaction (Figure 5B) confirms that the binding is mediated mainly by the three-aromatic residues Phe70, Phe71 and Phe75. Interestingly, a version of the Pex19 peptide with an engineered Pex5-like WxxxF/Y motif, Pex19(F75W), does not generate an efficient Pex14(N) binding motif. Modelling of this mutation onto the Pex14(N)–Pex19 complex structure indicates that binding to Pex19 F75W would require a rotation of the ligand helix to accommodate Trp75, which—in turn—would lead to a steric clash of Pex19 Leu74 and Pex19 Phe70, with the C-terminal part of helix α1 of Pex14(N). Finally, solvent-exposed residues of Pex19 are only exchangeable with negatively charged or neutral residues. A similar preference is also found for the Pex5 interaction (Figure 5A and B). This indicates a requirement of negative charges in the Pex5 and Pex19 ligands for binding to a positively charged surface in Pex14(N) (Figures 3 and 4). Mutational analysis of the Pex14 binding surface To validate our findings under physiological conditions, we designed a series of single-residue mutations at the interaction surface of Pex14(N), which are expected to interfere with the binding of Pex19 and/or Pex5. Each mutation was introduced into a full-length Pex14 expression construct for localisation studies in human cells and into an Escherichia coli Pex14(N) expression plasmid for in vitro binding studies (Figure 6). We selected eight Pex14(N) residues (Thr31, Ala32, Lys34, Asn38, Val41, Arg49, Phe52 and Lys56) that are involved in the formation of the Pex14(N) binding surface for Pex5 and Pex19. We changed the residues into either alanines, leading to removal of specific side-chain contributions, or amino acids that are expected to introduce either steric clashes (A32W, F52W) or repulsive, charged interactions (R49E, K56E) on complex formation. The tertiary structures of most of these Pex14(N) variants are not affected by the mutations, as judged from the similarity of the NMR spectra of wild-type and mutant proteins (Supplementary Figure 5). However, this is not true for the A32W and R49E mutants, which are less stable and/or aggregate in solution. The Pex14(N) variants were tested at a concentration of 1 μM for their ability to interact with the Pex5(113–127) and Pex19(66–80) peptides (Figure 6A). From the structurally unaffected variants (thus, not considering A32W and R49E), the mutations F52W, F52A, K56E and, to some extent, K34A decrease the binding affinity to Pex19, whereas at this concentration, the binding to Pex5 is less affected. When the Pex14 proteins were tested at lower concentrations (45 nM), only the F52W mutation did not significantly decrease the Pex5 binding affinity, suggesting that this mutation selectively impairs the Pex19 interaction. Figure 6.In vitro and in vivo effects of single-site mutations within Pex14(N). (A) Left: Pex14 Phe52 and Lys56 are critical for Pex19 binding. His-tagged Pex14(N) variants harbouring the indicated mutations were expressed in E. coli, purified, incubated at two different concentrations (1 μM, upper rows; 45 nM, lower row) with immobilised Pex5(113–127) and Pex19(66–80) peptides and detected by anti-Pex14 antibodies. Right: The location of the residues used for the mutational analysis is schematically indicated on the Pex14(N) structure. (B, C) Full-length Pex14 mutants show various patterns of cellular staining. (B) All single-point mutations lead to partial mitochondrial mislocalisation of full-length Pex14. A Zellweger patient Pex14-deficient fibroblast cell line was transfected with plasmids encoding Pex14 full-length proteins and analysed by immunofluorescence microscopy using antibodies against Pex14 (green, Alexa Fluor 488) and against a mitochondrial marker protein TRAP1 (red, Alexa Fluor 594). Most of the cells expressing Pex14(K56E) and Pex14(F52W), shown as representative e
Referência(s)