Subscriber access provided by University of Newcastle, Australia
Article
Nanoscale Structural Features in Major Ampullate Spider Silk Christian Riekel, Manfred Burghammer, Thomas G. Dane, Claudio Ferrero, and Martin Rosenthal Biomacromolecules, Just Accepted Manuscript • DOI: 10.1021/acs.biomac.6b01537 • Publication Date (Web): 06 Dec 2016 Downloaded from http://pubs.acs.org on December 15, 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.
Biomacromolecules 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 43
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
Biomacromolecules
Nanoscale Structural Features in Major Ampullate Spider Silk Christian Riekela,*, Manfred Burghammera,b, Thomas G. Danea, Claudio Ferreroa, and Martin Rosenthala a
The European Synchrotron (ESRF), CS40220, F-38043 Grenoble Cedex 9, France
b
Department of Analytical Chemistry, Ghent University, Krijgslaan 281, S12B-9000 Ghent, Belgium
*Corresponding author’s e-mail address:
[email protected] KEYWORDS: Major ampullate silk, X-ray nanodiffraction, hierarchical organization, skin/core morphology, nanocrystallites
1 ACS Paragon Plus Environment
Biomacromolecules
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 43
ABSTRACT Spider major ampullate silk is often schematically represented as a 2-phase material composed of crystalline nanodomains in an amorphous matrix. Here we are interested in revealing its more complex nanoscale organization by probing Argiope bruennichi draglinetype fibres using scanning X-ray nanodiffraction. This allows resolving transversal structural features such as an about 1 µm skin layer composed of around 100 nm diameter nanofibrils serving presumably as an elastic sheath. The core consists of a composite of several nm size crystalline
nanodomains
with
poly(L-alanine)
microstructure,
embedded
in
a polypeptide network with short-range order. Stacks of nanodomains separated by less ordered nanosegments form nanofibrils with a periodic axial density modulation which is particularly sensitive to radiation damage. The precipitation of larger β-type nanocrystallites in the outer core shell is attributed to MaSp1 protein molecules.
2 ACS Paragon Plus Environment
Page 3 of 43
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
Biomacromolecules
Silk fibres are adapted to specific functions in the life of a spider. Indeed, dragline silk originating in the major ampullate (MaA) glands is used for a variety of purposes such as structural fibres in webs or as a safety line.1-3 Due to its unique combination of strength and extensibility4,5 as well as ready availability by forced silking techniques6, the biological dragline silk production line has been extensively studied7 and considerable efforts have been put into schemes for producing artificial dragline silk although the ultimate mechanical properties of dragline silk have not yet been reached.8,9 Dragline silk is often schematically represented as a 2-phase material composed of crystalline β-sheet nanodomains of a few nm size embedded in an amorphous polypeptide network.8,10,11 Mean field approaches based on a 2-phase material and its chemical composition have been used for modeling mechanical properties of dragline silk and guiding artificial spinning developments.12 A key requirement for close mimicry of biological silk production and designing silks with specific functions is, however, a deeper understanding of the relation of hierarchical structural organization and macroscopic function such as mechanical properties. Indeed, bottom-up molecular modeling of a 2-phase model with random β-sheet nanodomains at the nodes of an amorphous network with entanglements allows reproducing dragline stress-strain curves.10 More refined modelling show that an optimized toughness requires a lamellar distribution of β-sheet nanodomains in the amorphous matrix13 reminding the nanofibrillar morphology based on nanodomains stacks deduced from X-ray and neutron scattering data.14,15 Nanofibrils obtained from recombinant or regenerated silk proteins provide in this respect interesting self-assembly and network models.11,16,17 The contribution of larger and spatially localized structural and molecular features to functional properties will likely be addressed by bottom-up modelling as computing power increases. Indeed, dragline silk shares a skin/core morphology with many functional polymers18 which has been attributed to two proteins secreted by two regions of the 3 ACS Paragon Plus Environment
Biomacromolecules
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 43
MaA gland.19,20 Nanoscale imaging based on transmission electron microscopy (TEM)20,21 or atomic force microscopy (AFM)22,23 relies generally on embedding, sectioning, staining or labeling protocols which does not allow probing nanoscale features and their relation to functional properties of whole silk fibres. This holds also for
the transversal core-
differentiation of the MaSp1/MaSp2 proteins observed by fluorescence labeling.23 While X-ray microdiffraction allows probing the hierarchical organization of whole fibres, even during forced silking,24,25 it is lacking lateral resolution for locally resolving nanoscale features. Here we explore the hierarchical organization of whole dragline silk fibres by X-ray nanodiffraction techniques providing nanoscale step-scanning resolution. We have studied MaA fibres from an Argiope bruennichi’s (called here Argiope) “bridge thread” which is anchored at two or more points supporting the weight of the whole orb-web, even in the presence of external forces such as strong winds (Supporting Information Figure S1a,b).26 The particularly thick fibres were collected in nature in order to avoid any influence of forcedsilking protocols27,28 on the microstructure of the fibres and associated skin layers. Materials and Methods Collection of silk fibres. Bridge thread silk composed of principally up to ~11 µm thick fibres was obtained from an adult Argiope bruennichi (Araneidae) spider living in its natural habitat (Supporting Information Figure S1a-f). A minor fraction of the fibres were significantly thinner and more flexible or even undulating and partially fused with the surface of bridge thread fibres (Supporting Information Figure S1d,e). We deposited a fibre bundle on an X-ray transparent and low scattering background Si3N4 membrane (Supporting Information Figure S1b). Single fibres were separated by pliers from the bundle and deposited on the same sample supports. X-ray scattering experiments were performed within a week after silk collection. (Supporting Information)
4 ACS Paragon Plus Environment
Page 5 of 43
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
Biomacromolecules
X-ray scattering. Synchrotron radiation (SR) experiments were performed in transmission geometry with the X-ray beam normal to the sample support surface (Figure 1a,b). X-ray microdiffraction was performed with a monochromatic SR-beam of wavelength λ=0.094 nm which was focused to an about 1.5x1.5 µm2 (fwhm) spot at the sample position. X-ray nanodiffraction was performed with a monochromatic beam of λ=0.083 nm which was focused to an about 170x170 nm2 (fwhm) spot at the sample position. X-ray diffraction patterns were recorded using a single photon counting X-ray pixel detector with exposure times up to 2 s. (Supporting Information) Results and Discussion Silk microstructure revealed by X-ray microdiffraction. We will set the stage for Xray nanodiffraction by exploring first structural information accessible to state-of-the-art Xray microdiffraction (Figure 1a). We performed a mesh-scan of a fibre bundle with 2 µm stepincrements. A pattern selected at random shows a fibre texture with Bragg peaks corresponding to the orthorhombic poly(L-alanine) structure of MaA silk29,30 (Figure 2a) which is also observed for minor ampullate (MiA) silk25,31-35. Its equatorial intensity profile can be fitted by Gaussians of different width for the (hk0) Bragg- and short-range order (SRO)-peaks (Figure 2c).24,35,36 It is generally accepted that the Bragg peaks are due to crystalline, alanine-rich nanodomains while the SRO-scattering is due to a glycine-rich polypeptide network with local order. The average dimensions of the nanodomains (L[hkl]) along selected [hkl] lattice directions derived from the Gaussian peak width37 (Table 1) compare well with L[hk0]-values from other spider species24,35,38 but not for L[001], i.e. the particle size along the fibre (c-) axis for which a species dependent variability was observed.38 The alignment of the nanodomains along the fibre axis can be quantified in terms of Herman’s orientation function (fc) based on the azimuthal width of the (210)/(020) peaks (Figure 2d) as fc~0.93 (fc=1 for perfect alignment, fc=0 for random alignment; Supporting Information 5 ACS Paragon Plus Environment
Biomacromolecules
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
Equations S3-5). Higher fc values of 0.96-0.98 (in the range of the high-performance polymer fibre Kevlar29)39 were observed for Nephila silk obtained by forced silking at drawing speeds of ~20 mm s-1.24,35,36 The correlation of drawing speed with fc for Nephila40 suggests qualitatively that the Argiope fibres were spun at an order of magnitude lower speed than fibres collected during forced silking experiments. The unit cell parameters of the poly(Lalanine) lattice were determined as a=0.970, b=1.082, c=0.680 (a,b,c: σ=0.001 nm; Supporting Information). The intersheet distance (b/2) of 0.541 nm agrees within errors to the poly(L-alanine) value of 0.527 (0.014) nm.29 A very weak peak with a lattice spacing of d=0.935 (0.005) nm is close to the calculated (100) reflection spacing (interchain repeat) (Figure 2c, Supporting Information) and has a similar azimuthal width as the (210)/(020) peaks (Figure 2e). This peak -called here (100)*- is not allowed for the poly(L-alanine) P212121 space group based on antipolar-antiparallel β-sheets30 nor does it agree to an alternative unit cell assuming a random displacement of neighbouring β-sheets along the aaxis by ±0.5a.29 The (100)* peak reminds a diffuse equatorial peak with a 0.65 ±0.05 nm range of lattice spacings observed by TEM for Nephila silk which also does not agree to the P212121 space group.41 The non-periodic lattice (NPL) model attributes this peak to an incorporation of GGX motifs (X: variable residue) in the polyalanine chains of nanodomains resulting in a loss of absences in the (hk0) plane due to aperiodic intersheet spacings along the [010] lattice direction.41 The peak observed by TEM does, however, not match the lattice spacing of the (100)* peak and suggests rather a breakdown of the 21 symmetry along the aaxis. One could attribute the peak in Argiope bridge-thread silk, however, to a minor polyalanine fraction with a modified chain-packing due to longer polyalanine stretches than the major polyalanine fraction.42 This implies that there is no currently no convincing argument based on X-ray diffraction for changing the P212121 lattice symmetry of the major polyalanine fraction. 6 ACS Paragon Plus Environment
Page 6 of 43
Page 7 of 43
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
Biomacromolecules
The nature of the meridional (c-axis) so-called “S-peak” is likewise not well understood (Figure 2a).35,43,44 Its ~0.42 nm lattice spacing does not match an integer fraction of the poly(L-alanine) lattice repeat of c~0.68 nm and the particle size of LS~15 nm is more than factor 3 larger than the L[001] value derived from the nanodomains (002) peak (Table 1). The S-peak has been tentatively attributed to crystallites of a 2nd β-sheet phase with larger chain periodicity due to the incorporation of GGX1GGX2-45 or GGX-motifs43 into the MaSp1 protein polyalanine stretches as it can be indexed as (005) reflection for a c~2.1 nm axis. The small-angle X-ray scattering (SAXS) pattern shows an equatorial streak, a meridional lamellar peak and a weak diffuse halo (Figure 2b). These features can be modeled for semicrystalline polymers and Nephila dragline silk by a nanofibrillar morphology.14,15,46 Indeed, the Guinier extrapolation (Supporting Information Equation S9) of Argiope’s equatorial streak suggests cylindrical objects with 6.3 ±0.2 nm diameter of the cross-section (Figure 2f), similar to the value of ~7 nm for Nephila.14 The Argiope cylinder diameter scales approximately to the contours of the fibre derived from the L[hkl]-values linking thus atomic and morphological scales (Supporting Information Figure S5a). We note that ~30 nm diameter nanofibrils obtained by AFM from regenerated B. mori fibroin11 relate to different nucleation process and protein molecules than the biospinning process.7 The angular profile of the streak in meridional direction can be described -as for Nephila14,15- by two Gaussians of different width (Figure 2g). The narrower Gaussian agrees to nanofibrillar objects of about 100 nm length along the fibre axis in agreement with MaA protein head-to-tail self-assembly mechanism. The broader Gaussian showing variability in width (Supporting Information Figure S5b) has been tentatively attributed to crazes.14 The lp~7.6 nm lamellar SAXS peak (“long period”, Abbreviation: lp) reflects the density modulation of nanodomain stacks in laterally
(ideally)
uncorrelated
I (Q ) = k | ρ c − ρ SRO |2 | Pc (Q ) |2 Z (Q )
nanofibrils
with
a
scattering
intensity:
where k is a normalization constant, | ρ c − ρ SRO | is the mean
7 ACS Paragon Plus Environment
Biomacromolecules
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
density difference between nanodomains and connecting SRO segments, Pc is the form factor of a nanodomain and Z is the Laue interference function.47 The axial length of the SRO segments is estimated from L[001] (Table 1) as roughly 4 nm assuming a homogeneous distribution of stacks composed of about 4 nm nanodomains in the nanofibrils. We note that neutron H/D contrast variation suggests glycine-rich SRO-segments with smectic order.15 The nanofibrils appear therefore to be the ultimate fibrillar units in dragline silk, similar in lateral dimensions to amyloidic nanofibrils.48 The diffuse scattering halo in semicrystalline nylon 6 fibres has been attributed to liquid-like interfibrillar correlations14,15,46 but contains in dragline silk other scattering contributions from nanoobjects as discussed below for the X-ray nanobeam results. X-ray nanodiffraction reveals skin/core morphology. Our approach for probing lateral nanoscale features in single fibres by scanning nanodiffraction is shown schematically in Figure 1b. The highest lateral resolution was obtained by 333 nm step-increments but we also probed fibres with up to 1 µm step-increments to explore radiation damage effects (Supporting Information Figures S9a-d, S10a-g, S15). Composite diffraction images of two fibres at the highest lateral resolution are shown in Figure 3a-f. The images are composed of pixels differing by the displayed fraction of reciprocal space. We selected either the Q-range of full patterns extending from morphological to atomic correlations (Figure 3a,b) or only parts of the SAXS range around to the beamstop limited to different morphological correlations (Figure 3c-f). The fibre appears to be highly homogeneous when displaying full patterns corresponding to the poly(L-alanine) lattice (see below). A fibre texture is maintained down to at least 107 unit cells within the probed volume at the edge of the fibre (Figure 3a,b). The fibre diameter is estimated from the image as 11±0.5 µm in agreement with the largest fibre diameters in the SEM images (Supporting Information Figure S1d-f). The lamellar SAXS peaks form also a highly regular distribution (Figure 3c,d). Displaying pixels covering only the equatorial SAXS streak reveals, however, a strong skin layer which is not related to 8 ACS Paragon Plus Environment
Page 8 of 43
Page 9 of 43
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
Biomacromolecules
nanofibrillar correlations (Figure 3e,f). Several strong streaks extending in the core along the fibre axis to µm-size (Supporting Information Figure S6) are attributed to refractive scattering from the walls of elongated cavities observed by TEM for Nephila silk.20 The overall contribution of cavities or crazes to the intensity of the streaks appears, however, to be negligible and cannot explain the broad peak in Figure 2g. Additional surface deposits in the form of a protrusion and a sheath partially enveloping the fibre are observed for one out of seven probed fibres (Figure 3g). The composite diffraction image of the protrusion reveals a heterogeneous phase mixture (Figure 3h,i) with bulk scattering resembling viscid silk glue composed of glycoproteins (Supporting Information Figure S12a,b).49,50 The surface layer corresponds to a lipidic phase with a characteristic d=0.42 nm reflection due to a hexagonal chain packing (Supporting Information Figure S11a-d).51 The d-spacings of the strongest two reflections in chain direction correspond to natural triglycerides.52 We analyzed the scattering of the central fibre in Figure 3a-f along a nearly orthogonal line to the fibre axis (Figure 4a). Indeed, a broad equatorial SAXS streak is observed in the surface layer while the central part of the fibre shows a narrow equatorial SAXS streak and the meridional lamellar peaks suggesting that the two profiles observed in the microbeam SAXS pattern (Figure 2g) are due to a superposition of skin and core scattering. The profile of the lamellar SAXS peak can be described by a Gaussian with a width corresponding to an angle of ~660 which translates in absolute values into a correlation length of 2π/(2.35 σ) ~ 7 nm, i.e. on the scale of the diameter of the nanofibrils (Figure 4b). Nanofibrillar skin. Nanodiffraction reveals an azimuthal distribution of individual nanofibrils by their equatorial streaks. Indeed, the azimuthal intensity profile of a single pattern can be separated into eight Gaussians (Figure 4c) with an overall angular range corresponding approximately to the broad SAXS peak observed for the fibre bundle (Figure 2g). In contrast, the SAXS pattern from the centre of the fibre corresponds to an overlap of scattering from skin and core (Figure 1b). Indeed, a narrow Gaussian with ~6.60 fwhm is 9 ACS Paragon Plus Environment
Biomacromolecules
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
attributed to the nanofibrils in the core while a ~360 fwhm Gaussian is due to a broader distribution of multiple nanofibrils in the skin which cannot be resolved (Figure 4d). We note that the scattering observed tangentially from the skin layer is enhanced relative to the skin layer scattering observed in the centre by different path length (Figure 1b). The absence of Bragg- or SRO-scattering from the skin layer suggests laterally uncorrelated nanofibrils. Indeed, the radial intensity profile of one of the nanofibrils revealed in Figure 4c shows several peaks which can be modeled by scattering from individual cylinders via a series of orders of a J0 Bessel function (Figure 4e, Supporting Information Equation S8).53 The experimental curve is smeared by scattering from neighboring nanofibrils. Better resolved orders can, however, be observed close to the inner interface of the surface layer (Figure S8a,b). We determined a cylinder diameter of 78.0 ±1.5 nm by linear extrapolation of the reciprocal peak coordinates as a function of orders (Figure 4f). Streaks from different locations of the skin show a variability in diameters of about 40-140 nm which was also observed for nanofibrils peeled from the surface of B. mori and wild silk fibres.54 (Figure 4g) The smallest diameters are obtained from positions close to the inner interface of the skin layer where also “crossing” nanofibrils were observed. Well-separated streaks from individual nanofibrils observed at this location (Supporting Information Figure S8a) support the suggestion of a thin, less densely packed layer at the inner interface of the skin layer.20 The lack of intersheet hydrogen-bonding and chain-packing correlations implies peptide chains with weak interactions, probably in a helical conformation as suggested for low-crystalline Flsilk42 which has been proposed containing 31 helical (GGX) and β-turn (GPGXX) repetitive motifs providing elasticity.55 This skin morphology would provide for an elastic sheath surrounding a more rigid core. SEM shows undulating fibres with
