Subscriber access provided by University of Sussex Library
Article
Dystrophin hot-spot mutants leading to Becker Muscular Dystrophy insert deeper into membrane models than the native protein Sarah Ameziane-Le Hir, Gilles Paboeuf, Christophe Tascon, Jean-François Hubert, Elisabeth Le Rumeur, Véronique Vié, and Celine Raguenes-Nicol Biochemistry, Just Accepted Manuscript • DOI: 10.1021/acs.biochem.6b00290 • Publication Date (Web): 01 Jul 2016 Downloaded from http://pubs.acs.org on July 3, 2016
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Biochemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 28
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biochemistry
Dystrophin hot-spot mutants leading to Becker Muscular Dystrophy insert deeper into membrane models than the native protein
Funding Source Statement Supported by grants from the Association Française contre les myopathies (AFM) telethon; GIS BRESMAT with the Conseil regional de Bretagne; and the European Regional Development Fund. SAL was funded by the French Ministère de l’Enseignement Supérieur et de la Recherche.
Sarah Ameziane-Le Hir,1,2,3Gilles Paboeuf,1,2Christophe Tascon,1,3Jean-François Hubert,1,3Elisabeth Le Rumeur,1,3Véronique Vié,1,2 and Céline Raguénès-Nicol.1,3*
1
Université de Rennes 1, 35042 Rennes, France;2 UMR CNRS 6251, Institut de physique de
Rennes, 35042 Rennes; 3 UMR CNRS 6290, Équipe SIM, 35043 Rennes.
Corresponding author Céline Raguénès-Nicol. Université de Rennes 1, UMR CNRS 6290, 2 av. du Prof. Léon Bernard, F-35043 Rennes.
[email protected], +33 223236113.
ACS Paragon Plus Environment
1
Biochemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 28
KEYWORDS Atomic-force microscopy, Becker muscular dystrophy, dystrophin, Langmuir film, molecular homology model, protein-lipid interaction.
ABBREVIATIONS AFM, atomic-force microscopy; ANS, 1-Anilino-8-naphthalenesulfonic acid; BMD, Becker muscular dystrophy; DOPC, dioleoylphosphatidylcholine; DOPS, dioleoyl phosphatidylserine; DYS, dystrophin; MST, MicroScale Thermophoresis; PCA, principal component analysis; SUV, small unilamellar vesicle; TNE, TRIS NaCl EDTA buffer; π, surface pressure.
ACS Paragon Plus Environment
2
Page 3 of 28
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biochemistry
ABSTRACT
Dystrophin (DYS) is a membrane skeleton protein whose mutations lead to lethal Duchenne muscular dystrophy or to the milder Becker muscular dystrophy (BMD). One third of BMD “inframe” exon deletions are located in the region that codes for spectrin-like repeats R16 to R21. We focused on four prevalent mutated proteins deleted in this area (called R∆45-47, R∆45-48, R∆45-49 and R∆45-51 according to the deleted exon numbers), analyzing protein/membrane interactions. Two of the mutants, R∆45-48 and R∆45-51, led to mild pathologies and displayed a similar triple coiled-coil structure as the full-length DYS R16-21, whereas the two others, R∆4547 and R∆45-49, induced more severe pathologies and showed “fractional” structures unrelated to the normal one. To explore lipid packing, small unilamellar liposomes (SUVs) and planar monolayers were used at various initial surface pressures. The dissociation constants determined by MicroScale Thermophoresis (MST) were much higher for the full-length DYS R161-21 than for the mutants, thus the wild type protein has weaker SUV binding. Comparing surface pressures after protein adsorption and analysis of AFM images of mixed protein/lipid monolayers revealed that the mutants insert more into the lipid monolayer than the wild type does. In fact, in both models every deletion mutant showed more interactions with membranes than the full-length protein did. This means that mutations in the R16-21 part of dystrophin disturb the protein’s molecular behavior as it relates to membranes, regardless of whether the accompanying pathology is mild or severe.
ACS Paragon Plus Environment
3
Biochemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 28
Dystrophin (DYS) is a long (about 175 nm) filamentous protein of the membrane skeleton of muscular cells and it contributes to the resistance of the sarcolemma during contractionelongation cycles.1, 2 According to the mutation type, protein expression levels, and the functions of the expressed proteins,3-5 dystrophin deficiency leads to either the lethal Duchenne muscular dystrophy or to the milder and heterogeneous Becker muscular dystrophy (BMD).6 Dystrophin contains a central domain accounting for 75% of the protein and composed of 24 spectrin-like repeats, R1 to R24, which are coded by exons 10-60. Composed of triple-α-helical spectrin-like repeats, this central domain allows interactions with membrane lipids, actin, and neuronal nitric oxide synthase (nNOS).7 In particular, the fragment between R16 and R21 contains the nNOS binding region8 and a part of the actin binding domain 2.9 This fragment is coded by exons 42-55 and includes a hot-spot area (exons 45-53) where about one third of BMD patients have mutations. Among these mutations, in-frame deletions of exons 45-47, 45-48, 45-49, or 45-51 are the most commonly encountered, and these deletions cause substantial variations in symptom severity.10, 11 Previous work on deletions in this hot-spot region showed that mutated proteins had different stability and structure, sometimes leading to reduced expression level.12-15 To explore the relationship between structure and symptom severity, researchers combined biochemistry experiments and molecular modelling to study DYS R16-21 fragments deleted of these parts (denoted R∆45-47, R∆45-48, R∆45-49, and R∆45-51). Molecular modeling showed that R∆4548 and R∆45-51 created hybrid repeats at the deletion site, reconstructing triple helical bundles, whereas R∆45-47 and R∆45-49 produced fractional repeats which prevent native triple coiledcoil construction.11,
16
The recombinant proteins can be ranked according to their refolding
dynamics, from faster to slower: R16-21 > R∆45-51 > R∆45-48 > R∆45-47 ~ R∆45-49. This classification is in line with the rankings of both severity and early outcome of symptoms. It therefore appears that in these cases, BMD severity was linked to the structures of the protein,17
ACS Paragon Plus Environment
4
Page 5 of 28
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biochemistry
with a hybrid repeat being less deleterious then a fractional repeat. In addition to this precious information on molecules in solution, an important question has to be answered: how does membrane binding relate to dystrophin’s functional activity? Certainly, at the molecular level, the function of dystrophin’s central domain is linked to its lipid binding properties, which seem to be required for sarcolemma stability during the contraction-elongation process.18, 19 Moreover, the ability to interact with lipid membranes was found to vary along the central domain depending on lipid charges or packing, as shown from investigations performed using monolayers20-23 or liposomes.15,
22, 24-26
Indeed, Langmuir monolayers are planar systems that
allow for control of the lipid packing through the choice of the surface pressure, while vesicle size affects both curvature and tension in liposomes. In this context, the present study focused on the analysis of four DYS R16-21 BMD mutants (R∆45-47, R∆45-48, R∆45-49, and R∆45-51) in membrane model interactions. By comparing the lipid-binding properties of these mutants with DYS R16-21, the work highlights the relationship between exon deletion and membrane interactions.
MATERIALS AND METHODS Protein expression and purification The DYS R16-21 fragment from residues 1991-2693 of the full-length dystrophin (Entrez gene ID 1756) was produced as a soluble protein in an Escherichia coli BL21 strain and purified by GSTrap affinity and size-exclusion chromatography, as previously described.23 The DYS R∆45-47, R∆45-48, R∆45-49, and R∆45-51 were coded by R16-21 truncated of exons 45-47, 45-48, 45-49, or 45-51, respectively.17 The four truncated proteins were produced as inclusion bodies in E. coli BL21 and solubilized with 0.1% N-lauryl-sarcosine. They were then purified
ACS Paragon Plus Environment
5
Biochemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 28
using GSTrap affinity chromatography, hydrophobic chromatography (in the case of R∆45–47 and R∆45–48 only), and size-exclusion chromatography. All proteins were concentrated in a Tris 20 mM pH 7.5 buffer, containing 150 mM NaCl and 0.1 mM EDTA (TNE buffer). Protein concentrations were determined with spectroscopy at 280 nm using theoretical molar extinction coefficients. ANS hydrophobicity determination Protein surface hydrophobicities were determined using 1-Anilino-8-naphthalenesulfonic acid (ANS).27, 28 A stock solution of 0.39 mM ANS was prepared in pH 7.5 phosphate-buffered saline (PBS) assuming ε350 (ANS) = 4950 M-1.cm-1. Proteins were prepared at 0.5, 1, 1.5, and 2 µM in PBS, and then ANS was added at a final concentration of 30 µM. After equilibration in ice for 30 min., sample fluorescence was recorded between 400 and 600 nm on a Fluorolog spectrofluorimeter (Horiba Jobin-Yvon, Longjumeau, France) at 350 nm excitation wavelength using 10 nm excitation and emission slit widths. Net fluorescence intensity was obtained after subtracting the fluorescence of the protein alone (without ANS). The maximum net fluorescence intensity was normalized relative to the protein R∆45-51 at 1.5 µM corrected to a value of 100 arbitrary units (AU). For each independent experiment, the initial slope (S0) of the normalized maximum fluorescence intensity versus protein concentration was calculated by linear regression and was used as an index of the protein surface hydrophobicity.
In silico analysis of protein surface properties Molecular homology models of DYS R16-21 and the R∆45-47, R∆45-48, R∆45-49, and R∆45-51 mutants have been previously described.17,
23
These models were used as starting
conformations for molecular dynamics using NAMD,29 and representative conformations were extracted by clustering and used to calculate surface properties. To compare the surface
ACS Paragon Plus Environment
6
Page 7 of 28
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biochemistry
properties of all of the structural models, they were aligned on repeat R16 of DYS R16-21 with the alpha-carbon root-mean-square deviation. The PLATINUM program 30 was used to calculate each construction’s molecular hydrophobicity potential (MHP) on the representative conformations in the molecular dynamics trajectories. The molecular structures and potentials were all visualized and rendered with the VMD software.31
Liposome preparations All lipids were purchased from Avanti Polar Lipids (Alabama, US). Other chemicals were from
Sigma.
As
previously
described26,
a
2:1
molar
ratio
mixture
of
dioleoylphosphatidylcholine/dioleoyl phosphatidylserine (DOPC/DOPS) from chloroform stock was vacuum-dried and then hydrated in TNE buffer at 25 mg/mL. Small unilamellar vesicles (SUVs) were obtained extemporaneously after sonication at room temperature for 2 min., with half-duty cycles using the micro-tip of a U200S sonicator (Ika Labortechnik). Dynamic light scattering with a Malvern 4700/PCS100 spectrometer yielded SUV diameters of 50-60 nm.
MicroScale Thermophoresis (MST) Proteins were labeled according to the manufacturer’s instructions with a Monolith NT RED-NHS-NT647 labeling kit (NanoTemper Technologies). The NT647-to-protein ratios were determined using photometry at 650 and 280 nm, and were between 0.9 and 1.5 for all proteins. TNE was used to adjust the labeled proteins to 100 nM. They were then titrated with a serial twofold dilution of SUVs going from 12.5 to 380 nM, with 1 mg/mL of bovine albumin added to avoid sticking. After 30 min. incubation, Monolith NT hydrophilic capillaries were loaded with the different solutions, and thermophoresis was measured using a Monolith NT.115 instrument (NanoTemper) at an ambient temperature of 25° C and 5/30/5 sec. laser off/on/off times,
ACS Paragon Plus Environment
7
Biochemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 8 of 28
respectively. Instrument parameters were adjusted to 50% LED power and 40% MST power. The data resulting from three or four independently-pipetted measurements were analyzed using NanoTemper’s NT.Analysis software (version 1.5.41) using the “Thermophoresis and T-Jump” resulting signal.
Langmuir trough experiments Protein adsorption at air/liquid or lipid/liquid interfaces was performed in Teflon troughs in order to determine the amphiphilic and lipid-binding properties of the five. Surface pressure (π) was determined using a tensiometer (NIMA, Cambridge, UK) using the Wilhelmy plate method as previously described.23 Experiments were performed at least twice. The interfacial cleanliness was controlled by the surface pressure stability (π = 0 mN/m) during 30 min. or during compression. Air/liquid interface measurements: the proteins were injected into the TNE buffer subphase solution at final concentrations ranging from 0.0001 to 1 µM. In this lipid-free experiment, the surface pressure was continuously recorded until equilibrium was reached. This surface pressure end point value was plotted against protein concentrations to determine πmax as an index of the amphiphilic character of each protein. Lipid/liquid interface measurements: monolayer lipid films were obtained by spreading a 1:1 DOPC/DOPS mixture in a 2:1 chloroform/methanol solution (v/v) at the air/liquid interface until a desired initial surface pressure (πi) of 20 or 30 mN/m was reached. The protein was then injected into the subphase beneath the lipid monolayer at a final concentration of 0.03 µM. The surface pressure was continuously recorded until equilibrium was reached, i.e. at the end of the adsorption kinetics. ∆π represents the surface pressure variation induced by protein adsorption to the monolayer from πi to πmax.
ACS Paragon Plus Environment
8
Page 9 of 28
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biochemistry
Atomic force microscopy (AFM) imaging As previously described,23 mixed-films were transferred to freshly-cleaved mica plate at the end of the adsorption kinetics using the Langmuir-Blodgett technique at a constant surface pressure. High-resolution imaging was performed under ambient conditions using a PicoPlus atomic force microscope (Agilent Technologies, Phoenix, AZ) equipped with a 10 µm scanner. Images were acquired in contact mode using silicon nitride tips on integral cantilevers (ScienTec) with a spring constant of 0.06 N/m. A series of AFM images were taken from different areas from at least two samples prepared on different days. Object heights were obtained after statistical analysis of one representative image with Gwyddion v2.41 software (http://gwyddion.net). The heights versus surfaces of the objects selected by threshold were plotted. The resulting plateau included the majority of the objects and was used to compare the different experimental conditions. To determine object size heterogeneity, the percentage of objects for which the surface was smaller than 3000 nm² was calculated for each image.
Statistical analysis Principal components analysis (PCA) was applied to the entire set of experimental data. The general principles of this multivariate statistical technique have been described in detail elsewhere.32 The active variables were: ANS S0 hydrophobicity; Kd measured by MST; πmax induced by proteins at the air/liquid interface; surface pressure variations according to the initial pressure of the lipid monolayers (∆π20 or ∆π30); number of objects (N20 or N30) and their heights (H20 or H30) in AFM images according to the initial pressure of lipids; and the proportion of
ACS Paragon Plus Environment
9
Biochemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 10 of 28
objects having an area smaller than 3000 nm² (A20 or A30). Statistical analyses were performed using R software (version 3.1.2, https://www.r-project.org/).33
RESULTS Protein surface properties and amphiphilicity We evaluated molecular hydrophobicities by studying ANS S0 and surface activity. ANS is a small hydrophobic and anionic molecule which becomes highly fluorescent after binding to nonpolar grooves of proteins. The use of ANS allows access to the global surface hydrophobicities of proteins, and can also reveal the presence of hydrophobic areas in small pits which could be related to the presence of flexible zones. The ANS S0 hydrophobicity of full-length DYS R16-21 is significantly lower than that for each of the four mutants studied (18 ± 1 AU, Fig. 1A). The hydrophobicities of all of the mutants are similar to each other (33 ± 7 to 35 ± 3 AU), although the fractional mutant R∆45-49 has a significantly higher one (57 ± 8 AU). In contrast, amphiphilic character is related to the presence of large patches on the molecular surface and to the ability to expose hydrophobic amino acids at the interface. We used maximum surface pressures (πmax) induced by protein adsorption at the air/liquid planar interface at identical protein subphase concentrations to compare the amphiphilicities of the five proteins (Fig. 1B). They all showed similar amphiphilic properties. However R∆45-47 had a significantly higher πmax than the others (23 vs 20.5-21 mN/m), which suggests that higher protein amounts cause a stronger lateral cohesion between proteins. The deletions in mutants induce structural changes that are not always restricted to the deletion zone, and these restructurings affect the repartition of charged and hydrophobic patches. In the absence of crystallographic data, molecular hydrophobicity potentials were calculated on models based on molecular dynamics (Fig. 1C). R∆45-48 and R∆45-51 had hybrid repeats at the
ACS Paragon Plus Environment
10
Page 11 of 28
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biochemistry
deletion site, and their global structures were less affected by the deletion. R∆45-47 and R∆4549 had fractional repeats at the deletion site: the coiled-coil spectrin repeat was not conserved. Nevertheless, R∆45-47 kept a mostly elongated structure, with a flexible zone and hydrophobic patch visible on both sides at the deletion site, with a new hydrophobic patch observed at the R20/R21 junction. This mutant’s amphiphilic properties could be explained by the appearance of such hydrophobic patches on the surface (Fig. 1C, dotted circles). In the case of R∆45-49, molecular dynamics yields a branched structure very different than that of the other proteins: a centrally-located large hydrophobic patch which is only observed on one side of the molecule. Comparison with the interfacial results suggests that this hydrophobic zone is not efficient enough to place the protein at the air/liquid interface.
Protein/lipid interactions using small unilamellar vesicles We used MicroScale Thermophoresis (MST) to study the interactions between dystrophin and liposomes, and the original time traces and normalized fluorescence binding curves are in the supplemental materials. This new technique is very reliable because it is not based on local modifications but instead on global binding-induced changes such as assembly size. The sensitivity of the MST detector allowed us to use as little as 50 nM proteins, and to reach a maximum protein/lipids ratio of 1:250000, which is necessary for high (up to 2.2 mM) apparent dissociation constant measurements. The curve of each protein’s fraction bound versus the SUV concentrations (Fig. 2A) showed a sigmoidal dependence on the lipid concentration, with a stable plateau for the higher values. This allowed us to calculate the dissociation constants using the law of mass action. Each protein binds to liposomes with a significantly different Kd (Fig. 2B). Full-length DYS R16-21, with a Kd of 2260 ± 272 µM, bound only weakly to them. All the mutants have lower Kd: R∆45-48 (826 ± 65 µM); R∆45-49 (572 ± 29 µM); R∆45-47
ACS Paragon Plus Environment
11
Biochemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 12 of 28
(276 ± 26 µM); and R∆45-51 (65 ± 10 µM). It therefore appears that the full-length protein has a weaker interaction with high-curvature liposomes than any of the mutants.
Protein/lipid interactions at the planar interface We studied protein insertion into lipid monolayers in Langmuir troughs with different lipid packing due to two different initial surface pressures (20 and 30 mN/m). Figure 3 presents the surface pressure variations (∆π), AFM images and the number of objects in each image, while figure 4 shows the height of objects for each condition and each mutants (Fig. 4B) extracted from the image analysis (Fig. 4A shows two examples, other conditions are covered in Supplementary figure S5). The heights are close to or smaller than 2 nm which is the width of repeat R1 obtained by X-ray crystallography34 and the width obtained for other fragments by modelling, and these proteins are filamentous and not globular. This allows us to say that the proteins, with the exception of R16-21 at 30 mN/m, all interact with the lipid film without uncontrolled and tridimensional aggregation. Whatever the initial surface pressure, the mutants always had a significantly higher ∆π value than full-length DYS R16-21, and repartition and protein protrusion shapes also appeared differently in the images. Both surface pressure variations obtained and AFM images analysis were used to compare the binding and the insertion of proteins in lipid monolayers. At πi = 20 mN/m, the mutants could be grouped according to their insertion ability: R∆45-47 with R∆45-48, with ∆π = 11 and 10 mN/m, respectively; and R∆45-49 with R∆45-51, with ∆π = 8.5 and 8 mN/m. Nevertheless, in each group the AFM images suggested different insertion forms. In the first group, image analysis of R∆45-47 showed fewer objects (2049) and larger surfaces (up to 6000 nm²), while R∆45-48 exhibits a higher number (8389) of smaller objects (mean surface 1500 nm²). R∆45-47 objects were worm-like filaments (mean height 2.21 ± 0.27
ACS Paragon Plus Environment
12
Page 13 of 28
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biochemistry
nm) with some aggregates. Meanwhile, R∆45-48 objects divide into two populations: a myriad of tiny objects just above the lipid film background (height about 0.43 ± 0.13 nm), meaning that those proteins were deeply inserted, and a few bigger objects (heights about 0.9 ± 0.27 nm). At a molecular level, R∆45-47 seems to present greater unidirectional protein/protein association properties, while R∆45-48 shows a higher tendency to disperse in lipids. For the second group, AFM images of R∆45-49 and R∆45-51 showed more objects (2485 vs 1988) and bigger aggregates (max height 2.5 nm vs 1.7 nm). Therefore, despite similar adsorption-induced surface pressure variations, the insertion modalities were different in this group, with greater insertion efficiency for R∆45-51 molecules than for R∆45-49. At πi = 30 mN/m, all of the mutants were able to insert into the lipid monolayers (∆π > 0), unlike the full-length DYS R16-21. Indeed, as we have previously shown, DYS R16-21 adsorption induces lipid reorientation and a relaxation of the lipid film, thus decreasing surface pressure (∆π < 0).23 As with the insertion parameters, AFM images showed drastically different patterns when comparing the deleted proteins with the full-length one. Whereas big aggregates (strong height and shape variations) appear for the full-length protein, the protrusions in the mutants are homogeneous in height and shape. R∆45-47 and R∆45-48 had higher insertion abilities (∆π = 3 and 4.7 mN/m, respectively) than R∆45-49 and R∆45-51 (2 and 1 mN/m). R∆45-47 and R∆45-48 present objects of similar mean heights (1.6 ± 0.17 nm vs 1.58 ± 0.15 nm), but the R∆45-47 objects were more numerous (5297 vs 3706) and smaller (97.2% vs 77.8% having mean surfaces < 3.103 nm²) than those of R∆45-48. R∆45-48 showed higher insertion abilities and more efficient auto-association than did R∆45-47. R∆45-49 and R∆45-51 had a low number of objects (1100 and 1329, respectively) and this resulted in lower surface pressures. The R∆45-49 images showed scattered objects (mean height ACS Paragon Plus Environment
13
Biochemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 14 of 28
1.4 ± 0.29 nm with 98% of surfaces < 3.103 nm²) and some aggregates (max height 3 nm), whereas the R∆45-51 images showed small objects with homogeneous sizes and heights (mean height 1.30 ± 0.2 nm with 99.6% of surfaces < 3.103 nm²). R∆45-51 thus showed lower insertion properties but higher dispersion abilities than R∆45-49. To conclude, increased lipid packing affects the insertion of each fragment, as revealed by a decrease in ∆π (Fig. 3). The mutant fragments are always included in the lipid film, and their heights are lower than that of full-length DYSR16-21, showing that they insert better into the membrane.
DISCUSSION In this study, we investigated the interactions between dystrophin R16-21 mutants and membrane models to delve deeper into the characterization of BMD-like proteins. For this purpose, we compared the behavior of previously-described DYS R16-2123 and four mutants that lead to Becker muscular dystrophy. The membrane models used had different lipid packing, with either high curvatures (SUV) or planar systems (monolayers) with two surface pressures.
The full-length DYS R16-21 had the lowest ANS surface hydrophobicity and the highest (2.2 mM) SUV dissociation constant. These results suggest that hydrophobic forces are involved in protein/liposome interactions when in the presence of high membrane curvature, i.e., low polar head group packing. In contrast, in monolayer, head group packing is increased.35 At πi = 20 mM/m, the insertion of DYS R16-21 was significant, but drastically reduced. At πi = 30 mM/m, comparable to physiological membrane pressure,36,
37
only head group interactions
were preserved, and the protein formed heterogeneous aggregates.
ACS Paragon Plus Environment
14
Page 15 of 28
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biochemistry
Numerical parameters were extracted for the wild type and mutant proteins from in-solution experiments, lipid-binding experiments, and AFM imaging. A principal component analysis of all of these was conducted (Fig. 5), highlighting the difference between the full-length protein and the deleted ones. Indeed, the factorial map for the first principal component, responsible for 51% of the data set variability, shows a complete segregation of the full-length R16-21 on one side and the mutants on the other (Fig. 5B, 5D). However, PCA did not allow us to regroup the mutants into sub-sections. In all of our experiments, therefore, the four mutants displayed different behavior than the full-length protein. The ANS hydrophobicity and the amphiphilicity determinations showed that R∆45-48 and R∆45-51 are similar, as both recreate a hybrid repeat at the deletion site. R∆45-47 and R∆45-49 exhibit more extreme behaviors: R∆45-47 has the highest interfacial activity, and R∆45-49 has the highest ANS surface hydrophobicity. For these two proteins, the native structure was not preserved at the deletion site. In particular, R∆45-49 formed a double coiled-coil “T” shape rather than the classic triple coiled-coil spectrin repeat. It also has a hydrophobic patch in the center which may encourage the adsorption of the ANS molecule. R∆45-47 kept an elongated form but also seemed to have new hydrophobic patches which explain its high interfacial activity. This raised the question of whether these characteristics measured in lipid-free solution might directly affect protein/membrane interactions. Furthermore, does lipid packing play a role in the interactions, as it does for the wild type protein? To address this, first remember that all of the mutants showed better lipid binding than did the full-length protein. Whatever the lateral packing, R∆45-49, with the highest ANS hydrophobicity (57 ± 8 AU) of the mutants, showed a low ability to link to membranes. The absence of membrane curvature and the low surface pressure allowed its insertion along with protein aggregation. As expected,
ACS Paragon Plus Environment
15
Biochemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 16 of 28
the increase in surface pressure limited insertion, and the aggregation remained. To conclude, R∆45-49’s atypical three-dimensional structure seems to strongly affect its interfacial activity. R∆45-47 was in the group of mutants with the highest lipid-binding properties in vesicles and at both surface pressures in flat lipid films. Image analysis showed its homogeneous repartition in the films, with protein/protein unidirectional assembly at 20 mN/m, and more dispersed objects at 30 mN/m. Even though R∆45-47 is filamentous like the wild type, it formed a fractional repeat at the deletion site that modified its hydrophobicity and surface charges, and strongly increased its lipid-binding properties. R∆45-47 and R∆45-49 are associated with severe Becker muscular dystrophy.17 Our results showed that the mutations directly affect the hydrophobicities and three-dimensional structures of these molecules, increasing their potential to insert into lipids. In lipid-free solution, R∆45-48 and R∆45-51 were quite similar, but surprisingly they displayed opposite behaviors in lipid interactions. Among the mutants, R∆45-48 was least efficient at binding to highly curved membranes. Nevertheless, its lipid-binding abilities increased with planar lipid films at both surface pressures. At πi = 30 mN/m, it was inserted into the monolayer and formed dispersed objects, completely opposite to the behavior of the fulllength protein. It may be that R∆45-48/lipid binding is mediated through head-group interactions rather than through hydrophobic ones. R∆45-51, which like R∆45-48 is a so-called “hybrid-repeat mutant,” was the most efficient mutant at SUV binding, but the least efficient in monolayer interactions. So for R∆45-51, accessibility to the hydrophobic part of the membrane seems to be important for allowing membrane interaction. Moreover, at high surface pressures this mutant had a very low insertion capacity (∆π = 1mN/m), much like the wild type. Other studies in solution without lipids show that ∆45-51 deletion has little impact on protein stability and refolding.14, 17 ACS Paragon Plus Environment
16
Page 17 of 28
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biochemistry
According to a statistical analysis of the clinical features of BMD patients,17 ∆45-51 deletion is almost asymptomatic and ∆45-48 leads to a mild BMD, even if both recombinant proteins reconstruct hybrid repeats at the deletion site. However when it comes to lipid packing, we have shown here that the R∆45-48 and R∆45-51 proteins have contradictory behaviors. Of all the mutants, R∆45-51 was the least inserted at high surface pressure, and was the closest to wild type behavior. This emphasizes the importance of protein behavior at high surface pressure, and may point to a new insight into the molecular mechanism. Indeed, in a previous study23 we suggested that during the muscle cycle, DYS R16-21 should be able to insert into the membrane at low surface pressure (elongation), and should be pushed out at the polar head level when compaction increases (sarcolemma contraction). R∆45-51, which roughly follows this schema, is quite asymptomatic, while R∆45-48, which is strongly inserted into the membrane at all surface pressures, leads to BMD. Another point to be considered is the interaction between the fragments and the intracellular partners. Indeed, R16 and R17 belong to the actin and nNOS binding sites, and the last part of the dystrophin (from R18) has also been demonstrated to influence the stability of the dystrophin/actin complex.8,
9, 38
In all mutants studied here R17 is truncated, thus potentially
leading to defective actin or nNOS binding. In addition, the insertion depth of the fragment in lipid membrane could also disadvantage the interaction with the cytoplasmic partners. In our experiments, R∆45-51 is not strongly inserted in membranes comparatively to the others mutants, this specificity likely preserve its linker activity with intracellular partners and indeed ∆45-51 deletion is almost asymptomatic in patients.
In conclusion, both the deletion type (either hybrid or fractional) and the lipid monolayer insertion ability are relevant to the assessment of the functioning of Becker muscular dystrophy
ACS Paragon Plus Environment
17
Biochemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 18 of 28
R16-21 mutants. In the end, whatever the length of the deletion, and whether the associated pathology was mild or severe, all the tested mutants of this hotspot region of dystrophin had disturbed lipid-binding behaviors. Our results suggest that, when searching for mutations with the lowest impact on dystrophin functioning, one should consider their structures at the deletion point and their insertion abilities into highly-packed membranes.
ACKNOWLEDGMENTS The authors thank the Association Française contre les Myopathies telethon for funding this project; Anne-Elisabeth Molza for help displaying the surface properties of molecular homology models; Cécile Le Floch-Fouéré for help with principal component analyses; and the BIOSIT Structure Federative de Recherché en Biologie et Santé de Rennes for circular dichroism facilities. Juliana Berland helped with the English.
Supporting Information Available at http://pubs.acs.org - S1: Description of the studied proteins - S2: Molecular hydrophobicity potentials in two opposite views of the most representative molecular dynamics conformations. - S3: Molecular electrostatic potentials in two opposite views of the most representative molecular dynamics conformations. - S4: Original timetraces of the MST experiments and individual binding curves in normalized fluorescence. - S5: Graphs of height vs surface of objects in AFM images
ACS Paragon Plus Environment
18
Page 19 of 28
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biochemistry
REFERENCES [1] Pons, F., Augier, N., Heilig, R., Leger, J., Mornet, D., and Leger, J. J. (1990) Isolated dystrophin molecules as seen by electron microscopy, Proc Natl Acad Sci U S A 87, 7851-7855. [2] Deconinck, N., and Dan, B. (2007) Pathophysiology of duchenne muscular dystrophy: current hypotheses, Pediatr Neurol 36, 1-7. [3] Anthony, K., Cirak, S., Torelli, S., Tasca, G., Feng, L., Arechavala-Gomeza, V., Armaroli, A., Guglieri, M., Straathof, C. S., Verschuuren, J. J., Aartsma-Rus, A., Helderman-van den Enden, P., Bushby, K., Straub, V., Sewry, C., Ferlini, A., Ricci, E., Morgan, J. E., and Muntoni, F. (2011) Dystrophin quantification and clinical correlations in Becker muscular dystrophy: implications for clinical trials, Brain 134, 3547-3559. [4] van den Bergen, J. C., Wokke, B. H., Janson, A. A., van Duinen, S. G., Hulsker, M. A., Ginjaar, H. B., van Deutekom, J. C., Aartsma-Rus, A., Kan, H. E., and Verschuuren, J. J. (2014) Dystrophin levels and clinical severity in Becker muscular dystrophy patients, J Neurol Neurosurg Psychiatry 85, 747-753. [5] Wilton, S. D., Fletcher, S., and Flanigan, K. M. (2014) Dystrophin as a therapeutic biomarker: are we ignoring data from the past?, Neuromuscul Disord 24, 463-466. [6] Liechti-Gallati, S., Koenig, M., Kunkel, L. M., Frey, D., Boltshauser, E., Schneider, V., Braga, S., and Moser, H. (1989) Molecular deletion patterns in Duchenne and Becker type muscular dystrophy, Hum Genet 81, 343-348. [7] Le Rumeur, E., Hubert, J. F., and Winder, S. J. (2012) A new twist to coiled coil, FEBS Lett 586, 2717-2722. [8] Lai, Y., Thomas, G. D., Yue, Y., Yang, H. T., Li, D., Long, C., Judge, L., Bostick, B., Chamberlain, J. S., Terjung, R. L., and Duan, D. (2009) Dystrophins carrying spectrinlike repeats 16 and 17 anchor nNOS to the sarcolemma and enhance exercise performance in a mouse model of muscular dystrophy, J Clin Invest 119, 624-635. [9] Amann, K. J., Renley, B. A., and Ervasti, J. M. (1998) A cluster of basic repeats in the dystrophin rod domain binds F-actin through an electrostatic interaction, J Biol Chem 273, 28419-28423. [10] Beggs, A. H., Hoffman, E. P., Snyder, J. R., Arahata, K., Specht, L., Shapiro, F., Angelini, C., Sugita, H., and Kunkel, L. M. (1991) Exploring the molecular basis for variability among patients with Becker muscular dystrophy: dystrophin gene and protein studies, Am J Hum Genet 49, 54-67. [11] Nicolas, A., Lucchetti-Miganeh, C., Yaou, R. B., Kaplan, J. C., Chelly, J., Leturcq, F., Barloy-Hubler, F., and Le Rumeur, E. (2012) Assessment of the structural and functional impact of in-frame mutations of the DMD gene, using the tools included in the eDystrophin online database, Orphanet J Rare Dis 7, 45. [12] Anthony, K., Arechavala-Gomeza, V., Ricotti, V., Torelli, S., Feng, L., Janghra, N., Tasca, G., Guglieri, M., Barresi, R., Armaroli, A., Ferlini, A., Bushby, K., Straub, V., Ricci, E., Sewry, C., Morgan, J., and Muntoni, F. (2014) Biochemical characterization of patients with in-frame or out-of-frame DMD deletions pertinent to exon 44 or 45 skipping, JAMA Neurol 71, 32-40. [13] Krieger, C. C., Bhasin, N., Tewari, M., Brown, A. E., Safer, D., Sweeney, H. L., and Discher, D. E. (2010) Exon-skipped dystrophins for treatment of Duchenne muscular dystrophy: mass spectrometry mapping of most exons and cooperative domain designs based on single molecule mechanics, Cytoskeleton (Hoboken) 67, 796-807.
ACS Paragon Plus Environment
19
Biochemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 20 of 28
[14] McCourt, J. L., Rhett, K. K., Jaeger, M. A., Belanto, J. J., Talsness, D. M., and Ervasti, J. M. (2015) In vitro stability of therapeutically relevant, internally truncated dystrophins, Skelet Muscle 5, 13. [15] Sahni, N., Mangat, K., Le Rumeur, E., and Menhart, N. (2012) Exon edited dystrophin rods in the hinge 3 region, Biochim Biophys Acta 1824, 1080-1089. [16] Menhart, N. (2006) Hybrid spectrin type repeats produced by exon-skipping in dystrophin, Biochim Biophys Acta 1764, 993-999. [17] Nicolas, A., Raguenes-Nicol, C., Ben Yaou, R., Ameziane-Le Hir, S., Cheron, A., Vie, V., Claustres, M., Leturcq, F., Delalande, O., Hubert, J. F., Tuffery-Giraud, S., Giudice, E., and Le Rumeur, E. (2015) Becker muscular dystrophy severity is linked to the structure of dystrophin, Hum Mol Genet 24, 1267-1279. [18] Manno, S., Takakuwa, Y., and Mohandas, N. (2002) Identification of a functional role for lipid asymmetry in biological membranes: Phosphatidylserine-skeletal protein interactions modulate membrane stability, Proc Natl Acad Sci U S A 99, 1943-1948. [19] Legardinier, S., Legrand, B., Raguenes-Nicol, C., Bondon, A., Hardy, S., Tascon, C., Le Rumeur, E., and Hubert, J. F. (2009) A Two-amino Acid Mutation Encountered in Duchenne Muscular Dystrophy Decreases Stability of the Rod Domain 23 (R23) Spectrin-like Repeat of Dystrophin, J Biol Chem 284, 8822-8832. [20] DeWolf, C., McCauley, P., Sikorski, A. F., Winlove, C. P., Bailey, A. I., Kahana, E., Pinder, J. C., and Gratzer, W. B. (1997) Interaction of dystrophin fragments with model membranes, Biophys J 72, 2599-2604. [21] Vie, V., Legardinier, S., Chieze, L., Le Bihan, O., Qin, Y., Sarkis, J., Hubert, J. F., Renault, A., Desbat, B., and Le Rumeur, E. (2010) Specific anchoring modes of two distinct dystrophin rod sub-domains interacting in phospholipid Langmuir films studied by atomic force microscopy and PM-IRRAS, Biochim Biophys Acta 1798, 1503-1511. [22] Sarkis, J., Hubert, J. F., Legrand, B., Robert, E., Cheron, A., Jardin, J., Hitti, E., Le Rumeur, E., and Vie, V. (2011) Spectrin-like repeats 11-15 of human dystrophin show adaptations to a lipidic environment, J Biol Chem 286, 30481-30491. [23] Ameziane-Le Hir, S., Raguenes-Nicol, C., Paboeuf, G., Nicolas, A., Le Rumeur, E., and Vie, V. (2014) Cholesterol favors the anchorage of human dystrophin repeats 16 to 21 in membrane at physiological surface pressure, Biochim Biophys Acta 1838, 1266-1273. [24] Le Rumeur, E., Fichou, Y., Pottier, S., Gaboriau, F., Rondeau-Mouro, C., Vincent, M., Gallay, J., and Bondon, A. (2003) Interaction of dystrophin rod domain with membrane phospholipids. Evidence of a close proximity between tryptophan residues and lipids, J Biol Chem 278, 5993-6001. [25] Le Rumeur, E., Pottier, S., Da Costa, G., Metzinger, L., Mouret, L., Rocher, C., Fourage, M., Rondeau-Mouro, C., and Bondon, A. (2007) Binding of the dystrophin second repeat to membrane di-oleyl phospholipids is dependent upon lipid packing, Biochim Biophys Acta 1768, 648-654. [26] Legardinier, S., Raguenes-Nicol, C., Tascon, C., Rocher, C., Hardy, S., Hubert, J. F., and Le Rumeur, E. (2009) Mapping of the lipid-binding and stability properties of the central rod domain of human dystrophin, J Mol Biol 389, 546-558. [27] Cardamone, M., and Puri, N. K. (1992) Spectrofluorimetric assessment of the surface hydrophobicity of proteins, Biochem J 282 ( Pt 2), 589-593. [28] Alizadeh-Pasdar, N., and Li-Chan, E. C. (2000) Comparison of protein surface hydrophobicity measured at various pH values using three different fluorescent probes, J Agric Food Chem 48, 328-334.
ACS Paragon Plus Environment
20
Page 21 of 28
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biochemistry
[29] Phillips, J. C., Braun, R., Wang, W., Gumbart, J., Tajkhorshid, E., Villa, E., Chipot, C., Skeel, R. D., Kale, L., and Schulten, K. (2005) Scalable molecular dynamics with NAMD, J Comput Chem 26, 1781-1802. [30] Efremov, R. G., Chugunov, A. O., Pyrkov, T. V., Priestle, J. P., Arseniev, A. S., and Jacoby, E. (2007) Molecular lipophilicity in protein modeling and drug design, Curr Med Chem 14, 393-415. [31] Humphrey, W., Dalke, A., and Schulten, K. (1996) VMD: visual molecular dynamics, J Mol Graph 14, 33-38, 27-38. [32] Jolliffe, I. (2002) Principal Component Analysis, second edition, Springer. [33] Team, R. D. C. (2014) R: a language and environment for statistical computing, R Foundation for Statistical Computing. [34] Muthu, M., Richardson, K. A., and Sutherland-Smith, A. J. (2012) The crystal structures of dystrophin and utrophin spectrin repeats: implications for domain boundaries, PLoS One 7, e40066. [35] Brouillette, C., Segrest, J., Ng, T., and Jones, J. (1982) Minimal size phosphatidylcholine vesicles: effects of radius of curvature on head group packing and conformation, Biochemistry 21, 4569-4575. [36] Marsh, D. (1996) Lateral pressure in membranes, Biochim Biophys Acta 1286, 183-223. [37] McDonald, A. G., and Tipton, K. F. (1996) Oscillatory NADH-oxidase activity of horseradish peroxidase, Biochem Soc Trans 24, 43S. [38] Henderson, D. M., Lin, A. Y., Thomas, D. D., and Ervasti, J. M. (2012) The carboxyterminal third of dystrophin enhances actin binding activity, J Mol Biol 416, 414-424.
ACS Paragon Plus Environment
21
Biochemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 22 of 28
FIGURES LEGENDS Figure 1. Surface properties of dystrophin R16-21 and mutants. A) Surface hydrophobicity (S0) measured with ANS. Mean ± SEM of 2 to 4 independent experiments. Statistically relevance of the data was verified using one-parameter ANOVA. The proteins marked with either * or $ have values that are different from all others (p < 0.05), while those marked with # have values identical to each other and different from the others (p < 0.05). B) Maximum surface pressure (πmax) induced by protein adsorption at the air/liquid interface. Mean ± SEM of 2 to 4 independent experiments. C) Molecular hydrophobicity potential of the most representative molecular dynamics conformations as calculated with PLATINUM software. All structures were aligned with repeat R16 at the left. The hydrophobicity measures go from hydrophilic to hydrophobic, shading from green to yellow. Circles are region of particular interest.
Figure 2. MicroScale Thermophoresis (MST) analysis of the interactions between liposomes and dystrophin (DYS) R16-21 proteins. Serial dilutions of 12.5 to 400 nM SUV DOPC/DOPS were incubated with 100 nM of DYS R16-21 or with mutants labelled with Red-NHS-NT647 dye. MST was measured after 30 min. incubation in hydrophilic capillaries with 50% LED power and 40% MST power. A) Individual experimental points of fraction bound and the binding curves fitted with the law of mass action. DYS R16-21 is dark grey R∆45-47 is green, R∆45-48 is red, R∆45-49 is blue, and R∆45-51 is purple. B) Apparent dissociation constant Kd (mean ± SD) calculated by the fit.
Figure 3. AFM topographic images of dystrophin / lipid transferred monolayers, number of particles, and the induced surface pressure variations. Mixed films were obtained by injection of
ACS Paragon Plus Environment
22
Page 23 of 28
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biochemistry
0.03 µM proteins in the subphase under a DOPC/DOPS monolayer at either 20 mN/m or 30 mN/m of initial surface pressures. In the R∆45-48 panel with a πi of 20mN/m, the inset shows a different contrast to emphasize the presence of deeply-inserted small objects. Scan size was 5x5 µm² and the Z-range 5 nm. Quantitative image analysis was performed using Gwyddion software.
Figure 4. Height of objects in DOPC/DOPS/protein monolayers for DYS R16-21 and the four mutants. A) Gwyddion software was used for statistical analysis of object heights and surfaces in AFM images. Graphs of height versus surface show a plateau where the medium heights were determined. As an example, shown here are R∆45-48 (red squares) and R∆45-49 (blue circles) at πi = 30 mN/m. B) Mean heights ± standard deviations as determined by the statistical analysis discussed in A).
Figure 5. Principal components analysis (PCA) of dystrophin mutant parameters. The active variables are ANS S0; MST Kd; πmax; ∆π20 and ∆π30; N20 and N30; H20 and H30; and A20 and A30. Data set variability is shown in parentheses for each principal component. A) Correlation circle of the PCA for components 1 and 2 (PC1 and PC2). B) Individual factorial map for components 1 and 2. C) Correlation circle of the PCA for components 1 and 3. D) Individual factorial map for components 1 and 3.
ACS Paragon Plus Environment
23
Biochemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 24 of 28
FIGURES
Figure 1
ACS Paragon Plus Environment
24
Page 25 of 28
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biochemistry
Figure 2
ACS Paragon Plus Environment
25
Biochemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 26 of 28
Figure 3
ACS Paragon Plus Environment
26
Page 27 of 28
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biochemistry
Figure 4
ACS Paragon Plus Environment
27
Biochemistry
B π max
0.0
3 ∆π30
S0
H30
2
∆π20
1
N30
A20
RΔ45-51
R16-21
Kd
RΔ45-49
-1
-0.5
PC2 (24%)
0.5
A3
RΔ45-47
PC2 (24%)
H20
0
1.0
A
N20
-1.0
-2
RΔ45-48
-1.0
-0.5
0.0 PC1 (51%)
0.5
1.0
-4
-2 0 PC1 (51%)
2
4
π max
N20
H20
0. 0
R16-21 ∆π30 ∆π20 A20
RΔ45-47
PC3 (22%) -1 1 0
0. 5
Kd H30
RΔ45-48
2
N30
RΔ45-51 S0
-
RΔ45-49 -3
A30 -1.0
-2
-0.5
PC3 (22%)
-6
D
1. 0
C
-1.0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 28 of 28
0.0 PC1 (51%)
0.5
1.0
-4
-2
0
2
PC1 (51%)
Figure 5
GRAPHIC FOR THE TABLE OF CONTENTS (TOC GRAPHIC).
ACS Paragon Plus Environment
28