Spider Silk-CBD-Cellulose Nanocrystal Composites: Mechanism of Assembly

International Journal of Molecular Sciences

Article Spider Silk-CBD-Cellulose Nanocrystal Composites: Mechanism of Assembly

Sigal Meirovitch 1,†, Zvi Shtein 1,†, Tal Ben-Shalom 1, Shaul Lapidot 1, Carmen Tamburu 2, Xiao Hu 3,4, Jonathan A. Kluge 3, Uri Raviv 2, David L. Kaplan 3 and Oded Shoseyov 1,* 1 The Robert H. Smith Institute of Plant Sciences and Genetics in Agriculture, Robert H. Smith Faculty of Agriculture, Food and Environment, The Hebrew University of Jerusalem, P.O. Box 12, Rehovot 76100, Israel; [email protected] (S.M.); [email protected] (Z.S.); [email protected] (T.B.-S.); [email protected] (S.L.) 2 The Institute of Chemistry, The Hebrew University of Jerusalem, Edmond J. Safra Campus, Givat Ram, Jerusalem 91904, Israel; [email protected] (C.T.); [email protected] (U.R.) 3 Department of Biomedical Engineering, 4 Colby Street, Tufts University, Medford, MA 02155, USA; [email protected] (X.H.); [email protected] (J.A.K.); [email protected] (D.L.K.) 4 Department of Physics and Astronomy, Rowan University, Glassboro, NJ 08028, USA * Correspondence: [email protected]; Tel.: +972-8-948-9083 † These authors contributed equally to this work.

Academic Editors: John G. Hardy and Chris Holland Received: 30 June 2016; Accepted: 8 September 2016; Published: 18 September 2016 Abstract: The fabrication of cellulose-spider silk bio-nanocomposites comprised of cellulose nanocrystals (CNCs) and recombinant spider silk protein fused to a cellulose binding domain (CBD) is described. Silk-CBD successfully binds cellulose, and unlike recombinant silk alone, silk-CBD self-assembles into microfibrils even in the absence of CNCs. Silk-CBD-CNC composite sponges and films show changes in internal structure and CNC alignment related to the addition of silk-CBD. The silk-CBD sponges exhibit improved thermal and structural characteristics in comparison to control recombinant spider silk sponges. The glass transition temperature (Tg) of the silk-CBD sponge was higher than the control silk sponge and similar to native dragline spider silk fibers. Gel filtration analysis, dynamic light scattering (DLS), small angle X-ray scattering (SAXS) and cryo-transmission electron microscopy (TEM) indicated that silk-CBD, but not the recombinant silk control, formed a nematic liquid crystalline phase similar to that observed in native spider silk during the silk spinning process. Silk-CBD microfibrils spontaneously formed in solution upon ultrasonication. We suggest a model for silk-CBD assembly that implicates CBD in the central role of driving the dimerization of spider silk monomers, a process essential to the molecular assembly of spider-silk nanofibers and silk-CNC composites.

Keywords: spider silk; cellulose nanocrystals; cellulose binding domain; nanocomposite; biomaterials

1. Introduction Silks are produced by a variety of insects and spiders and some spiders spin as many as seven different kinds of silks, each tailored to fulfill a certain biological function [1]. Dragline silk, used as the safety line and as the frame thread of the spider’s web, is composed of two proteins, each with a long repetitive sequence flanked by nonrepetitive amino and carboxy termini [2,3]. The repetitive sequence is characterized by stretches of poly-alanine domains that are interrupted by glycine-rich repeats. The poly-alanine domains form β-sheet crystals, whereas the glycine-rich repeats form less crystalline segments [4–6]. The interplay between the hard crystalline domains and the less crystalline regions gives rise to the extraordinary properties of the silk [7,8]. Dragline silk has unique toughness and higher tensile energy to break than any other common natural or artificial material [9]. Its strength

Int. J. Mol. Sci. 2016, 17, 1573; doi:10.3390/ijms17091573 www.mdpi.com/journal/ijms Int. J. Mol. Sci. 2016, 17, 1573 2 of 16 to weight ratio is five times stronger than steel and three times tougher than top quality man-made Kevlar fiber. Due to its superb mechanical properties, dragline silk has been the focus of intense research efforts [1,10]. Despite the extensive knowledge that has been gained regarding the structure and properties of dragline silk, its production remains challenging. In contrast to silkworm-silk, the isolation of large quantities of silk from spiders is not feasible. Spiders produce silk in small quantities, and their territorial behavior prevents large numbers from being raised in a confined space [11]. Therefore, production of spider silk proteins through recombinant DNA techniques has been the primary path pursued by researchers. Silk-encoding genes have been cloned and expressed in a variety of heterologous hosts [12–15], which have allowed for production of laboratory scale quantities of silk-like protein powders. Yet, with few exceptions [16,17], the material properties of these cloned silks are inferior to their native counterparts. One limitation has been the lack of molecular order in the recombinant silk proteins and their tendency to aggregate in vitro, bypassing the native refolding and assembly processes [10]. Assembly of silk proteins into a solid silk fiber is extremely complex, and replication of the spider’s native spinning process is a major challenge in laboratory settings [18,19]. The current work addresses this challenge using a different strategy—the production of spider silk CNC bio-composites. The technique relies on the specific binding capacity of cellulose binding domains (CBDs) to CNCs and the structural properties of CNCs. Cellulose is a product of biosynthesis from plants, animals, or bacteria, while “cellulose nanocrystals” refer to cellulosic extracts or processed materials, having defined nano-scale structural dimensions [20]. CNCs are exciting biomaterials that are relevant to a number of potential applications, including polymer nanocomposites, transparent films and hydrogels. CNCs are mainly produced by acid hydrolysis/heat controlled techniques, with sulfuric acid being the most utilized acid. The crystal extraction from cellulose involves hydrolysis of amorphous cellulose regions, resulting in highly crystalline particles with dimensions of 5–20 nm in width and 100–500 nm in length for plant source CNCs [21]. The mechanical properties of CNCs is impressive with the Young’s modulus and tensile strength of a single crystal reported to be as high as 150 and 10 GPa, respectively, which make them useful for the reinforcement of polymers. In addition, the rod-like shape of the particles leads to concentration-dependent liquid crystalline self-assembly behavior [22]. Noishiki et al. [23] found that CNC-native silkworm silk films had breaking strengths and ductility about five times greater than those of the constituent materials. The authors attributed these improvements to the flat and ordered surfaces of CNCs, which served as a template for the assembly of silk β-sheets, a process that usually requires shear and elongation stress. In nature, cellulose is degraded by the concerted actions of a number of bacterial and fungal organisms, initiated by cellulolytic enzyme(s) or microorganisms that can bind cellulose substrates. The CBD, a separate, nonhydrolytic component, mediates this binding. CBDs have been cloned from different organisms such as Clostridium cellulovorans and Cellulomonas fimi [24,25], and enable adhesion of the water-soluble enzyme to the soluble or insoluble substrate, by bringing the catalytic module into prolonged and intimate contact with the cellulose surface [26,27]. In the present study, we compare synthetic 15-monomer-long dragline spider silk derived from the native sequence of MaSp1 of Nephila clavipes, referred to as “silk”, and the synthetic 15-monomer with CBD fused to its 30 termini, referred to as “silk-CBD”. Unlike silk, upon sonication, silk-CBD dimerizes and these dimers assemble in situ into microfibrils; this mechanism appears analogous to the formation of silk fibrils in nature, which proceeds via the C-termini. Furthermore, the effects of silk-CBD dimerization upon the mechanisms of CNC self-assembly of the CNCs in sponges and films are explored. The general motivation for composites made from silk and CNCs is to produce materials with characteristics reflective of both components; for instance, silk-cellulose composites may provide a strength and toughness profile that surpasses either component. In this work, we observed that silk-CBD specifically binds CNCs and confers molecular order which is different from that of either the silk proteins or CNCs. Silk-CBD-CNC composite materials may be useful in a variety of medical and industrial applications, and this preliminary research is a necessary step toward the end goal of harnessing the attractive properties of the components into a composite with superior properties. Int. J. Mol. Sci. 2016, 17, 1573 3 of 16

2. ResultsInt. and J. Mol. Discussion Sci. 2016, 17, 1573 3 of 15

2.1. Protein2. Expression Results and and Discussion Puriﬁcation Int. J. Mol. Sci. 2016, 17, 1573 3 of 15 The synthetic2.1. Protein spider Expressi silkon and and Purification the fusion spider silk-CBD genes were successfully expressed, and the resultant2. Results proteinsThe and synthetic Discussion were spider puriﬁed silk and from the fusionE. colispiderusing silk-CBD Ni-NTA genes were chromatography. successfully expressed, The and pure protein yields for thethe expressionresultant proteins ofsilk were (47 purified kDa) from and E. silk-CBDcoli using Ni-NTA (65 kDa) chromatography. were 60 and The 40 pure mg/L, protein respectively 2.1. yieldsProtein for Expressi the expressionon and Purification of silk (47 kDa) and silk-CBD (65 kDa) were 60 and 40 mg/L, respectively (Figure1). (FigureThe synthetic 1). spider silk and the fusion spider silk-CBD genes were successfully expressed, and the resultant proteins were purified from E. coli using Ni-NTA chromatography. The pure protein yields for the expression of silk (47 kDa) and silk-CBD (65 kDa) were 60 and 40 mg/L, respectively (Figure 1).

Figure 1. Expression and purification of silk and silk-CBD. SDS-PAGE of soluble E. coli proteins, Figure 1. Expression and purification of silk and silk-CBD. SDS-PAGE of soluble E. coli proteins, stained stained with Coomassie blue (A); and Western blot analysis (B) using an anti-HIS antibody. Lane 1, with Coomassiemolecular blue weight (A); andmarker; Western lane 2, total blot protein analysis of the (B )control using bacteria an anti-HIS (E. coli antibody.transformed Lanewith an 1, molecular weight marker;empty lane vector); 2, totallane 3, protein total protein of the of control silk expres bacteriasing bacteria; (E. coli lanetransformed 4, total protein with of silk-CBD an empty vector); expressing bacteria; lane 5, control sample (empty vector) after Ni-NTA purification; lane 6, purified lane 3, totalFigure protein 1. Expression of silk and expressing purification bacteria; of silk and lane silk-CBD. 4, total SDS-PAGE protein of of silk-CBD soluble E. expressingcoli proteins, bacteria; silk protein; and lane 7, purified silk-CBD fusion protein. lane 5, controlstained samplewith Coomassie (empty blue vector) (A); and after Western Ni-NTA blot purification; analysis (B) using lane 6,an purifiedanti-HIS antibody. silk protein; Lane and 1, lane 7, purified2.2.molecular silk-CBD Quantitative weight fusion Cellulose marker; protein. Binding lane 2, Assay total protein of the control bacteria (E. coli transformed with an empty vector); lane 3, total protein of silk expressing bacteria; lane 4, total protein of silk-CBD In order to characterize the binding capacity of silk-CBD to cellulose, adsorption/desorption expressing bacteria; lane 5, control sample (empty vector) after Ni-NTA purification; lane 6, purified 2.2. Quantitativeexperiments Cellulose were Bindingconducted. Assay Adsorption/desorption experiments are commonly done to test the silk protein; and lane 7, purified silk-CBD fusion protein. apparent irreversible adsorption of CBD to cellulose. A reversible adsorption process is defined when In orderthe tovariables characterize characterizing the bindingthe state of capacity the system of return silk-CBD to the same to cellulose, values in the adsorption/desorption reverse order 2.2. Quantitative Cellulose Binding Assay experimentsduring were the conducted. desorption stage. Adsorption/desorption Therefore, in a reversible experimentsadsorption process, are the commonly ascending branch done to test the apparent irreversible(increasingIn order toprotein adsorption characterize concentration ofthe CBD binding in tothecellulose. solution)capacity andof A silk-CBD reversiblethe descending to cellulose, adsorption branch adsorption/desorption (decreasing process protein is defined when experimentsconcentration were in conducted.the solution) Adsorption/desorptof the isotherm overlaionp. Reversible experiments adsorption are commonly was seen for done the purifiedto test the the variables characterizing the state of the system return to the same values in the reverse order during apparentsilk protein irreversible in solution adsorption with ce ofllulose CBD to(Figure cellulos 2) e.due A reversible to the mech adsorptionanism of processprotein adsorptionis defined whenat the desorptionthe solid/liquidvariables stage. characterizing interfaces Therefore, [28]. theIn in contrast, astate reversible of irrevers the system adsorptionible binding return was to process, theobserved same the forvalues both ascending CBDin the and reverse branch silk-CBD order (increasing protein concentrationduringas evident the desorption from in their the stage.non-overlapping solution) Therefore, and adsorption thein a descendingreve isotherms.rsible adsorption branch process, (decreasing the ascending protein branch concentration in the solution)(increasing of protein the isotherm concentration overlap. in the Reversible solution) and adsorption the descending was seen branch for (decreasing the purified protein silk protein in solutionconcentration with cellulose in the solution) (Figure of 2the) due isotherm to the overla mechanismp. Reversible of adsorption protein was adsorption seen for the at purified solid/liquid silk protein in solution with cellulose (Figure 2) due to the mechanism of protein adsorption at interfaces [28]. In contrast, irreversible binding was observed for both CBD and silk-CBD as evident solid/liquid interfaces [28]. In contrast, irreversible binding was observed for both CBD and silk-CBD from theiras evident non-overlapping from their non-overlapping adsorption isotherms. adsorption isotherms.

Figure 2. Adsorption/desorption isotherms. CBD (solid line), silk (dotted line), silk-CBD (dashed line), at different concentrations were allowed to adsorb to cellulose (Sigmacell 20) to the point of equilibrium. After equilibrium was reached, the highest protein concentration (1.2 mg/mL) to cellulose mixture was diluted to allow desorption.

Figure 2. Adsorption/desorption isotherms. CBD (solid line), silk (dotted line), silk-CBD (dashed line), Figure 2. Adsorption/desorption isotherms. CBD (solid line), silk (dotted line), silk-CBD (dashed at different concentrations were allowed to adsorb to cellulose (Sigmacell 20) to the point of line), atequilibrium. different concentrations After equilibrium were was allowedreached, tothe adsorbhighest toprotein cellulose concentration (Sigmacell (1.2 20)mg/mL) to the to point of equilibrium.cellulose After mixture equilibrium was diluted was to reached,allow desorption. the highest protein concentration (1.2 mg/mL) to cellulose mixture was diluted to allow desorption. Int. J. Mol. Sci. 2016, 17, 1573 4 of 16 Int. J. Mol. Sci. 2016, 17, 1573 4 of 15

2.3.2.3. CompositeComposite CNC/SpiderCNC/Spider SilkSilk SpongesSponges SpiderSpider silk/CNCsilk/CNC composite composite sponge sponge formation formation was donewas asdone previously as previously described described [29–32]. Purified,[29–32]. concentratedPurified, concentrated spider silk spider protein silk was protein mixed was with mixe a CNCd with suspension a CNC suspension and then sonicated. and then Thissonicated. procedure This hasprocedure a two-fold has a effect; two-fold in addition effect; in to addition homogeneous to homogeneous dispersion dispersion of the CNCs, of the the CNCs, sonication the sonication process inducesprocess annealinginduces annealing of spider of silk spider proteins silk proteins by accelerating by accelerating formation formation of physical of physical cross-links, cross-links, such as initialsuch as chain initial interactions chain interactions related to βrelated-sheet formationto β-sheet [33formation,34]. After [33,34]. sonication, After three-dimensionalsonication, three- porousdimensional structures porous (i.e., structures sponges) (i.e., were sp generatedonges) were via generated freeze drying. via freeze drying. SEMSEM picturespictures ofof thethe resultingresulting spongessponges showedshowed thatthat porepore architecturearchitecture andand alignmentalignment differeddiffered betweenbetween thethe silk-CBD silk-CBD and and control control silk silk sponges sponges (Figure (Figure3). 3). Silk Silk sponges sponges had had 30–100 30–100µm μ poresm pores (Figure (Figure3C) of3C) irregular of irregular shape shape and and with with no no particular particular orientation, orientation, very very similar similar to to the the CNC-silkCNC-silk compositescomposites (Figure(Figure3 D).3D). Silk-CBD Silk-CBD sponges sponges featured featured 300–500 300–500µm leaf-shaped μm leaf-shaped pores aligned pores inaligned a relatively in a consistentrelatively directionconsistent (Figure direction3E). (Figure Similar characteristics3E). Similar characteristics were observed were in sponges observed from in nativesponges silkworm from native silk producedsilkworm usingsilk produced the same using conditions the same applied conditions here, which applied were here, attributed which were to the attributed parallel arrangementsto the parallel ofarrangements silk fibroin crystal of silk flakes fibroin [30 crys]. Thetal flakes composite [30]. silk-CBD-CNCThe composite spongessilk-CBD-CNC possessed sponges ~100 µ possessedm structurally ~100 alignedμm structurally pores (Figure aligned3F–H). pores (Figure 3F–H).

FigureFigure 3. SEMSEM pictures pictures of ofsilk, silk, silk-CBD, silk-CBD, and andcomposite composite silk-CBD-CNC silk-CBD-CNC sponges. sponges. (A,B) CNC (A,B sponge;) CNC sponge;(C) 100% (C silk) 100% sponge; silk sponge;(D) silk-CNC (D) silk-CNC composite composite sponge (75% sponge silk (75%and 25% silk andCNC); 25% (E) CNC); 100% (silk-CBDE) 100% silk-CBDsponge; and sponge; (F–H) silk-CBD-CNC and (F–H) silk-CBD-CNC composite sponge composite (75% silk sponge-CBD (75%and 25% silk-CBD CNC) on and a magnified 25% CNC) scale. on a magniﬁed scale. The glass transition and degradation temperatures of different silk and silk-CBD sponges, as determinedThe glass by transitionTMDSC analysis, and degradation are shown temperaturesin Figure 4 and of Table different 1. DSC silk analysis and silk-CBD of the 100% sponges, silk asand determined silk-CBD sponges by TMDSC gave analysis, Tg values are of shown 140 and in 172 Figure °C, 4respective and Tablely,1 .and DSC degradation analysis of temperatures the 100% silk of and279 and silk-CBD 283 °C, sponges respectively. gave Tg Interestingly,values of 140 the and Tg 172 and◦ C,degradation respectively, temperatures and degradation of the temperatures 100% silk-CBD of 279sponge and 283were◦C, similar respectively. to those Interestingly, reported for the naturalTg and silkworm degradation silk temperaturesand Nephilia ofclavipes the 100% dragline silk-CBD silk spongefibers [35–37]. were similar The changes to those seen reported in the Tg for, characteristic natural silkworm of thesilk amorph and ousNephilia domains clavipes in amorphousdragline silk or ﬁberssemi-crystalline [35–37]. The materials, changes such seen as in silks, the Tg result, characteristic from increased of the amorphouschain interactions. domains The in stronger amorphous the orinteractions semi-crystalline between materials, the chains, such the as silks,higher result the te frommperature increased required chain interactions.to induce a phase The stronger transition. the The lower Tg of the 100% silk sponge may be due to a more disordered structure, as seen in the SEM figures. The elevation in the Tg of the 25% silk/75% CNC sponge is likely related to the significant

Int. J. Mol. Sci. 2016, 17, 1573 5 of 16 interactions between the chains, the higher the temperature required to induce a phase transition. The lower Tg of the 100% silk sponge may be due to a more disordered structure, as seen in the SEM figures.Int. TheJ. Mol.elevation Sci. 2016, 17, 1573 in the Tg of the 25% silk/75% CNC sponge is likely related to the5 significant of 15 presencepresence of CNCs, of CNCs, whose whose crystal crystal surfaces surfaces may serve serve as as a template/nucleation a template/nucleation site for site the forassembly the assembly of of silksilkβ-sheets, β-sheets, as as seen seen in in the the silkworm silkworm silk-CNC comp compositeosite films films by byNoishiki Noishiki et al. et [23].As al. [23 for]. As the for the elevatedelevatedTg of Tg the of silk-CBDthe silk-CBD sponges, sponges, it it has has beenbeen well established established that that CBDs CBDs form form types types of dimers of dimers in in solutionsolution [38,39 [38,39],], and and this this dimerization dimerization factor factor likelylikely also plays plays a arole. role.

FigureFigure 4. DSC 4. analysisDSC analysis of silk of and silksilk-CBD and silk-CBD sponges. sponges. Reverse Reverse heat heat ﬂow flow vs. temperaturevs. temperature during during TMDSC scanningTMDSC of silk scanning and silk-CBDof silk and sponges silk-CBD atsponges 2 ◦C/min. at 2 °C/min. (a) 100% (a) 100% silk sponge;silk sponge; (b )(b) 25% 25% silk/75% silk/75% CNC CNC sponge; (c) 75% silk/25% CNC sponge; (d) 25% silk-CBD/75% CNC sponge; (e) 75% silk- sponge; (c) 75% silk/25% CNC sponge; (d) 25% silk-CBD/75% CNC sponge; (e) 75% silk-CBD/25% CBD/25% CNC sponge; and (f) 100% silk-CBD sponge. The arrows indicate the glass transition CNC sponge; and (f) 100% silk-CBD sponge. The arrows indicate the glass transition temperatures. temperatures.

TableTable 1. Glass 1. Glass transition transition and and degradation degradation temperatures temperatures of of silk silk and and silk-CBD silk-CBD sponges, sponges, as determined as determined from TMDSCfrom TMDSC analysis. analysis.

Sample Glass TransitionGlass Temperature Transition (Tg ◦C)Degradation Degradation Temperature (◦C) Sample Temperature (Tg °C) Temperature (°C) 100% silk 140.5 283 100% silk-CBD100% silk 172 140.5 283 279 75% silk 25% CNC100% silk-CBD 138 172 279 255 75% silk-CBD 25% CNC75% silk 25% CNC 174 138 255 269 25% silk 75% CNC75% silk-CBD 25% CNC 163 174 269 231 25% silk-CBD 75% CNC25% silk 75% CNC 178 163 231 237 25% silk-CBD 75% CNC 178 237 2.4. Composite CNC/Spider Silk Films 2.4. Composite CNC/Spider Silk Films FilmsFilms of silk-CBD of silk-CBD and and CNCs CNCs were were prepared prepared in order order to to further further investigate investigate the effects the effects silk-CBD silk-CBD on theon materials the materials and and the the role role ofof dimerization. Simila Similarr to the to sponge the sponge results presented results presented in Figure 3, in SEM Figure 3, SEM cross-sectionalcross-sectional images images of CNC of CNC and andsilk-CBD-CNC silk-CBD-CNC films at films mass atratios mass of 1:5 ratios and of1:10 1:5 (Figure and 1:10 5) show (Figure 5) showdifferences in infilm film morphology morphology related related to the topres theence presence and amount and of amount silk-CBD. of The silk-CBD. CNC film The shows CNC film showsa atypical typical layered layered morphology, morphology, whereas whereas the composite the composite silk-CBD-CNC silk-CBD-CNC films appear films to be appearmore aligned to be more alignedand and dense. dense. Film Film alignment alignment was was explored explored using using a polarized a polarized optical optical microscopy microscopy (POM) (POM) system system equippedequipped with with an image an image processing processing module. module. TheThe top images images in in Figure Figure 6 shows6 shows POM POM images images of CNC of CNC and silk-CBD-CNC composite films, where the bright and dark regions typically indicate ordered and silk-CBD-CNC composite films, where the bright and dark regions typically indicate ordered and disordered areas in the films, respectively. The bottom images in Figure 5 show the processed and disorderedbirefringence areas images. in theThe films, CNC films respectively. show the Themulti-domain bottom imagesorder typical in Figure to CNC5 show films, the whereas processed birefringencethe silk-CBD images. films The show CNC uniform films showalignment. the multi-domain The Abrio 2.2 ordersoftware typical (CRi, to Woburn, CNC films, MA, whereasUSA) the silk-CBDprovides films a showvector uniformoverlay tool alignment. to analyze the The proc Abrioessed 2.2 image, software where (CRi, the vector Woburn, azimuth MA, is measured, USA) provides a vectorand overlay the standard tool to deviation analyze (SD) the processedof the vector image, direction where gives the an vector approximation azimuth of is measured,the degree of and the standard deviation (SD) of the vector direction gives an approximation of the degree of alignment Int. J. Mol. Sci. 2016, 17, 1573 6 of 16

Int. J. Mol. Sci. 2016, 17, 1573 6 of 15 uniformity. An average of twenty measurements per sample (same size area) and corresponding SD Int. J. Mol. Sci. 2016, 17, 1573 6 of 15 valuesalignment were calculated; uniformity. an SDAn ofaverage 44.25 wasof twenty obtained me forasurements CNC films, per 13.58sample for (same 1:10silk-CBD-CNC size area) and films and 1.33correspondingalignment for 1:5 uniformity. silk-CBD-CNC SD values An were average films.calculated; of These twenty an SD ofme values 44.25asurements was indicate obtained per improved sample for CNC (same sample films, size 13.58 alignment area) for 1:10and with increasingsilk-CBD-CNCcorresponding silk-CBD filmsSD content. values and 1.33 were Furthermore, for calculated; 1:5 silk-CBD-CNC an we SD qualitatively of films. 44.25 Thesewas observedobtained SD values for that indicate CNC the films, filmsimproved 13.58 appeared samplefor 1:10 more transparentalignmentsilk-CBD-CNC when with silk-CBD filmsincreasing and was1.33 silk-CBD for present 1:5 silk-CBD-CNCcontent. (Figure Furthe7). This films.rmore, may These we be qualitatively dueSD values to the indicate particular observed improved that approach the sample films to film appeared more transparent when silk-CBD was present (Figure 7). This may be due to the particular formationalignment used with in this increasing work (i.e., silk-CBD CNC formulationscontent. Furthe castrmore, onto we hydrophobic qualitatively surfaces)observed orthat may the befilms related approach to film formation used in this work (i.e., CNC formulations cast onto hydrophobic surfaces) to theappeared dispersion/self-assembly more transparent when of the silk-CBD particles was in present the films. (Figure 7). This may be due to the particular orapproach may be torelated film formation to the dispersion/self- used in this workassembly (i.e., ofCNC the formulationsparticles in the cast films. onto hydrophobic surfaces) or may be related to the dispersion/self-assembly of the particles in the films.

Figure 5. SEM images of CNC film cross-sections (A,D); silk-CBD-CNC composite film with 1:10 Figure 5. SEM images of CNC film cross-sections (A,D); silk-CBD-CNC composite film with 1:10 weight weight ratio (B,E); and composite films with 1:5 weight ratio (C,F). The upper images are at 100,000× ratioFigure (B,E); 5. and SEM composite images of filmsCNC withfilm cross-sections 1:5 weight ratio(A,D); ( Csilk-CBD-CNC,F). The upper composite images film are with at 100,000 1:10 × magnificationweight ratio ( Band,E); theand bottom composite images films at with200,000×. 1:5 weight (Scale ratiobar = (500C,F ).nm). The upper images are at 100,000× magnification and the bottom images at 200,000×. (Scale bar = 500 nm). magnification and the bottom images at 200,000×. (Scale bar = 500 nm).

Figure 6. Top images present POM of CNC and silk-CBD-CNC composite films with increasing amountsFigure 6. of Top silk-CBD. images Bottom present images POM ofshow CNC the and corresponding silk-CBD-CNC LC-PolScope composite analysis films with performed increasing on Figure 6. Top images present POM of CNC and silk-CBD-CNC composite ﬁlms with increasing theamounts POM image.of silk-CBD. The color Bottom wheel images in the show corner the pres coentsrresponding the orientation LC-PolScope of the sampleanalysis alignment performed (20× on amounts of silk-CBD. Bottom images show the corresponding LC-PolScope analysis performed on magnification).the POM image. The color wheel in the corner presents the orientation of the sample alignment (20× the POM image. The color wheel in the corner presents the orientation of the sample alignment magnification). (20× magniﬁcation).

Int. J. Mol. Sci. 2016, 17, 1573 7 of 16 Int. J. Mol. Sci. 2016, 17, 1573 7 of 15

FigureFigure 7. QualitativeQualitative comparison comparison of oftransparency transparency of ofCNC CNC films films comp comparedared to silk-CBD to silk-CBD CNC CNC 1:10 1:10composite composite films. films. (A) 1:10 (A) 1:10composite composite silk-CBD-CNC silk-CBD-CNC film; film; (B) CNC (B) CNC film. film.

2.5. Spider Silk-CBD Fusion Protein Assembly 2.5. Spider Silk-CBD Fusion Protein Assembly To summarize, interesting behavior is observed in the composite materials when silk-CBD is To summarize, interesting behavior is observed in the composite materials when silk-CBD is used, used, including an elevation in the Tg, differences in internal structure and morphology, and including an elevation in the Tg, differences in internal structure and morphology, and differences differences in alignment in the films. As mentioned above, CBD dimerization may play a role in the in alignment in the films. As mentioned above, CBD dimerization may play a role in the observed observed Tg elevation and in driving the differences observed in alignment. Gel filtration and Tg elevation and in driving the differences observed in alignment. Gel filtration and dynamic light dynamic light scattering (DLS) and small angle X-ray scattering (SAXS) analyses were conducted in scattering (DLS) and small angle X-ray scattering (SAXS) analyses were conducted in order to better order to better understand this process and whether it affects the properties of the composite understand this process and whether it affects the properties of the composite materials presented in materials presented in this work. this work. Protein solutions analyzed by gel filtration and DLS before and after sonication indicated Protein solutions analyzed by gel filtration and DLS before and after sonication indicated increased increased molecular weights and diameters of silk-CBD assemblies upon sonication, compared to the molecular weights and diameters of silk-CBD assemblies upon sonication, compared to the control silk control silk protein (Figure 8). The pure silk solution was eluted with a peak corresponding to 140 protein (Figure8). The pure silk solution was eluted with a peak corresponding to 140 kDa (Figure8A), kDa (Figure 8A), whereas the silk-CBD was eluted as two major peaks, one eluted into the void whereas the silk-CBD was eluted as two major peaks, one eluted into the void volume, and the second volume, and the second eluted as a 230 kDa protein, which may correspond to a dimer. After eluted as a 230 kDa protein, which may correspond to a dimer. After sonication, most of the silk-CBD sonication, most of the silk-CBD protein was eluted into the void volume (Figure 8B). The calculated protein was eluted into the void volume (Figure8B). The calculated molecular weights of the silk and molecular weights of the silk and silk-CBD proteins are 47 and 65 kDa, respectively. SDS-PAGE silk-CBD proteins are 47 and 65 kDa, respectively. SDS-PAGE analysis under denaturing conditions analysis under denaturing conditions supported the molecular weight calculations (Figure 8C,D). As supported the molecular weight calculations (Figure8C,D). As reported previously, proteins rich in reported previously, proteins rich in glycine and alanine can migrate anomalously during gel glycine and alanine can migrate anomalously during gel filtration, often resulting in overestimation of filtration, often resulting in overestimation of the protein molecular weight [40]. In addition, the gel the protein molecular weight [40]. In addition, the gel filtration was performed under non-denaturing filtration was performed under non-denaturing conditions and silk polypeptides may adopt a rod- conditions and silk polypeptides may adopt a rod-like elongated configuration, leading to size like elongated configuration, leading to size overestimation, compared to globular protein standards. overestimation, compared to globular protein standards. It has been established that spider silk It has been established that spider silk proteins form disulfide-bridged homodimers [41–44]. Gel proteins form disulfide-bridged homodimers [41–44]. Gel filtration of the native proteins under filtration of the native proteins under reducing conditions revealed 260–320 kDa protein monomers. reducing conditions revealed 260–320 kDa protein monomers. In the absence of the reducing agent, In the absence of the reducing agent, a shift in the molecular mass to 420–480 kDa was observed, less a shift in the molecular mass to 420–480 kDa was observed, less than a two-fold gain in molecular than a two-fold gain in molecular weight [41]. Dimer formation likely leads to changes in the protein weight [41]. Dimer formation likely leads to changes in the protein conformation, resulting in modified conformation, resulting in modified protein migration along the column, as compared to the protein migration along the column, as compared to the monomer form. The silk-CBD dimer may act monomer form. The silk-CBD dimer may act as a nucleation site for higher molecular weight as a nucleation site for higher molecular weight assemblies, which were eluted into the void volume, assemblies, which were eluted into the void volume, even before sample sonication. even before sample sonication. After sonication, an increase in the proportion of higher molecular weight silk assemblies was After sonication, an increase in the proportion of higher molecular weight silk assemblies observed, similar to that previously described for native silkworm silk proteins subjected to was observed, similar to that previously described for native silkworm silk proteins subjected to sonication [33]. As demonstrated by DLS (Table 2), both un-sonicated and sonicated silk formed sonication [33]. As demonstrated by DLS (Table2), both un-sonicated and sonicated silk formed homogeneous solutions with 3–4 nm diameter particles, whereas un-sonicated silk-CBD had 55 and homogeneous solutions with 3–4 nm diameter particles, whereas un-sonicated silk-CBD had 55 and 260 nm diameter particles, which increased in size to 96 nm and 2 μm particles upon sonication. 260 nm diameter particles, which increased in size to 96 nm and 2 µm particles upon sonication.

Int. J. Mol. Sci. 2016, 17, 1573 8 of 16 Int. J. Mol. Sci. 2016, 17, 1573 8 of 15

FigureFigure 8.8. GelGel ﬁltrationfiltration andand DLSDLS ofof silksilk andand silk-CBDsilk-CBD samplessamples beforebefore andand afterafter sonicationsonication (sonic.).(sonic.). RepresentativeRepresentative gelgel ofof silksilk ((AA);); andand silk-CBDsilk-CBD ((BB)) proteinprotein samplessamples beforebefore (solid(solid line)line) andand afterafter sonicationsonication (dashed(dashed line);line); SDS SDS PAGE PAGE analysis analysis of of gel gel filtration ﬁltration fractions fractions of ofsonicated sonicated silk silk (C); ( Cand); and silk-CBD silk-CBD (D) (samples.D) samples.

TableTable 2. 2.Summary Summary ofof DLSDLS resultsresults ofof silksilk andand silk-CBDsilk-CBD samplessamples beforebefore andand afterafter sonicationsonication ((nn == 10,10, averageaverage± ± standard deviation).

SilkSilk Sonicated Silk Silk Silk-CBD Silk-CBD Sonicated Sonicated Silk-CBD Silk-CBD diameterdiameter (nm) (nm) 2.782.78± ±0.04 0.04 3.8 3.8± ±0.11 0.11 825 825± ±401 401 55 ±± 36 259 259 ±± 3030 96 96 ± ±8 8 20482048 ± 691691 % volumevolume 100100 99.999.9 0.10.1 32 68 68 64 64 36 36

2.6.2.6. SmallSmall AngleAngle X-rayX-ray ScatteringScattering (SAXS)(SAXS) ofof SolutionsSolutions ofof SilkSilk andandSilk-CBD Silk-CBD SAXSSAXS measurementsmeasurements ofof thethe silksilk andand silk-CBDsilk-CBD solutionssolutions analyzedanalyzed beforebefore andand afterafter sonicationsonication (Figure(Figure9 )9) demonstrated demonstrated a a form form factor factor closely closely matching matching that that of of infinitely infinitely long long rods. rods. The The silk silk samples samples releasedreleased aa veryvery weakweak signalsignal withwith apparentlyapparently nono structurallystructurally orderedordered subunitssubunits formedformed inin solution,solution, neitherneither beforebefore (Figure9 9A)A) nornor after (Figure 99C)C) sonication.sonication. In In the the case case of of un-sonicated un-sonicated silk-CBD, silk-CBD, the thesingle single and and rather rather broad broad correlation correlation peak peak (Figure (Figure 9A)9A) indicated indicated a nematic liquidliquid crystallinecrystalline phase,phase, withwith aa correlationcorrelation distancedistance ofof 26.426.4 nmnm andand aa domaindomain sizesize ofof approximatelyapproximately 8080 nm.nm. InIn otherother words,words, threethree subunits subunits are are in in positional positional correlation, correlation, and and the th rode rod center-to-center center-to-center distance distance is 26.4 is nm26.4 (Figure nm (Figure9D1). After9D1). sampleAfter sample concentration concentration (Figure 9(FigureA), the 9A), peak the intensity peak intensity increased increased and the correlation and the correlation distance decreaseddistance decreased to 24.6 nm. to Namely,24.6 nm. theNamely, nematic the phase nemati wasc phase denser was and denser more subunitsand more were subunits in positional were in correlationpositional correlation with one another with one (Figure another9D2). (Figure In nature, 9D2). the In highly nature, concentrated the highly spider concentrated protein dope,spider muchprotein like dope, that ofmuch the silkworm,like that of is liquidthe silkworm, crystalline, is whereliquid thecrystalline, main silk where protein the constituent main silk isprotein likely toconstituent assume a is compact likely to rod-like assume conformation. a compact rod-li Thiske conformation conformation. enables This conformation silk protein processing enables silk at highprotein concentrations. processing at Specifically,high concentrations. the molecules Specifically, seem to the form molecules a nematic seem phase to form in the a nematic spider glandphase andin the duct, spider with gland the long and axes duct, of neighboringwith the long molecules axes of neighboring aligned approximately molecules parallelaligned toapproximately one another. Liquidparallel crystallinity to one another. offers Liquid desirable crystallinity properties, offers such desirable as efficient properties, spinning such of as molecules efficient spinning as large asof silkmolecules proteins, as bylarge allowing as silk the proteins, viscous by silk allowing protein the solution viscous to slowlysilk protein flow throughsolution theto slowly storage flow sac andthrough duct the as complex storage alignmentsac and duct patterns as complex are formed alignment [19]. Afterpatterns sonication are formed (Figure [19].9B), After the silk-CBDsonication correlation(Figure 9B), peaks the silk-CBD were fitted correlation to a two-dimensional peaks were fitted oblique to a latticetwo-dimensional with three subunits oblique lattice along thewith a-axis, three formingsubunits a along rod center-to-center the a-axis, forming distance a rod of center-to-cent 30.8 nm ander three distance subunits of 30.8 along nm theand b-axis, three subunits forming aalong rod center-to-centerthe b-axis, forming distance a rod center-to-center of 54.98 nm. The distance alignment of 54.98 angle nm. between The alignment the two-dimensional angle between subunits the two- dimensional subunits was γ = 83.2° (Figure 9E1); sonication resulted in the addition of subunit interactions, which led to their two-dimensional alignment. After concentrating the solution (Figure

Int. J. Mol. Sci. 2016, 17, 1573 9 of 16

◦ wasInt. J. γMol.= 83.2Sci. 2016(Figure, 17, 15739E1); sonication resulted in the addition of subunit interactions, which led9 of to 15 their two-dimensional alignment. After concentrating the solution (Figure9B), the lattice parameters changed9B), the tolattice a = 32.4parameters nm, b = 45.8changed nm and to γa = 93.432.4◦ ,nm, and b namely, = 45.8 morenm and subunits γ = 93.4°, were and in correlation namely, more with onesubunits another were (Figure in correlation9E2). with one another (Figure 9E2).

Figure 9. (A–C) Radially integrated solution small-angle X-ray scattering intensity from silk and Figure 9. (A–C) Radially integrated solution small-angle X-ray scattering intensity from silk and silksilk-CBD non-concentrated and concentrated (conc.) samples before (A) and after sonication (B,C); (CBDD,E) non-concentrated Schematic illustration and concentr of the silk-CBDated (conc.) hierarchical samples before order (A in) and non-concentrated after sonication ( D1(B,,CE1); )(D and,E) concentratedSchematic illustration (D2,E2) samples,of the beforesilk-CBD (D )hi anderarchical after sonication order in (Enon-concentrated). Models of the structures(D1,E1) and of self-assembledconcentrated (D2 silk-CBD,E2) samples, subunits before before (D (D3) and) and after after sonication (E3) sonication. (E). Models The orange of the unitsstructures represent of self- the CBDassembled moiety silk-CBD and the subunits green and before blue units(D3) representand after the(E3)crystalline sonication. domains The orange and units the less represent crystalline the regionsCBD moiety of the and silk the monomer, green and respectively. blue units represent the crystalline domains and the less crystalline regions of the silk monomer, respectively. 2.7. Cryo-Transmission Electron Microscopy (Cryo-TEM) 2.7. Cryo-Transmission Electron Microscopy (Cryo-TEM) Cryo-transmission electron microscopy (cryo-TEM) images of sonicated protein samples support Cryo-transmission electron microscopy (cryo-TEM) images of sonicated protein samples the gel filtration, DLS and SAXS results (Figure 10). After sonication, the silk protein did not support the gel filtration, DLS and SAXS results (Figure 10). After sonication, the silk protein did not form any orderly structures (Figure 10 upper images), whereas silk-CBD formed 100–200 nm long form any orderly structures (Figure 10 upper images), whereas silk-CBD formed 100–200 nm long fibers/microfibrils (Figure 10 bottom images). The liquid crystals and the longer structures revealed fibers/microfibrils (Figure 10 bottom images). The liquid crystals and the longer structures revealed by SAXS and DLS were not detectable by this method, as they form very thick layers that are removed by SAXS and DLS were not detectable by this method, as they form very thick layers that are removed from the grid when preparing the thin film (<100 nm thick) required for cryo-TEM imaging. from the grid when preparing the thin film (<100 nm thick) required for cryo-TEM imaging.

Int. J. Mol. Sci. 2016, 17, 1573 10 of 16

Int. J. Mol. Sci. 2016, 17, 1573 10 of 15

Figure 10. Cryo-transmission electron microscopy (cryo-TEM) images of sonicated solutions of silk Figure 10. Cryo-transmission electron microscopy (cryo-TEM) images of sonicated solutions of silk (upper pictures) and silk-CBD (bottom pictures). Scale bar is 60 nm. (upperFigure pictures 10. Cryo-transmission) and silk-CBD ( bottomelectron picturesmicroscopy). Scale (cryo-TEM) bar is 60images nm. of sonicated solutions of silk 2.8. Silk-CBD(upper pictures Assembly) and Model silk-CBD (bottom pictures). Scale bar is 60 nm. 2.8. Silk-CBD Assembly Model Based on the gel filtration, DLS, SAXS and cryo-TEM findings, we suggest that silk-CBD forms 2.8. Silk-CBD Assembly Model Baseddimers onafter the solution gel ﬁltration, concentration, DLS, SAXSwhich align and cryo-TEMto form liquid ﬁndings, crystals we(Figure suggest 11C). thatAfter silk-CBD sonication, forms dimersthe after hydrophobicBased solution on the spider concentration,gel filtration, silk protein DLS, which domains SAXS align and are tocryo-TEMmore form prone liquid findings, to engage crystals we in suggest further (Figure thatinteractions, 11 C).silk-CBD After leading sonication,forms dimers after solution concentration, which align to form liquid crystals (Figure 11C). After sonication, the hydrophobicto higher molecular spider assemblies silk protein (Figure domains 11D). are This more molecular prone order to engage likely ineffects further the interactions,morphology and leading thethermal hydrophobic properties spider of the silk silk-CBD-CNC protein domains composites are more, forprone instance to engage by driving in further CNC interactions, alignment leading(see the to higher molecular assemblies (Figure 11D). This molecular order likely effects the morphology and tomechanism higher molecular presented assemblies in Figure (Figure 12). 11D). This molecular order likely effects the morphology and thermalthermal properties properties of theof the silk-CBD-CNC silk-CBD-CNC composites, composites, for for instance instance by by driving driving CNC CNC alignment alignment (see the (see the mechanismmechanism presented presented in Figurein Figure 12 12).).

Figure 11. Silk-CBD assembly model. (A) silk-CBD in solution; (B) Concentration and dialysis of the silk-CBD protein solution leads to protein dimer formation (C); and (D) Sonication of silk-CBD FigureFigureexposes 11. Silk-CBD 11. hydrophobic Silk-CBD assembly assembly domains, model. model. which ( A(A) ) silk-CBDallowsilk-CBD for in infu solution;rther solution; interactions, (B ()B Concentration) Concentration creating andhigher anddialysis dialysismolecular of the of the silk-CBD protein solution leads to protein dimer formation (C); and (D) Sonication of silk-CBD silk-CBDassemblies. protein solution leads to protein dimer formation (C); and (D) Sonication of silk-CBD exposes exposes hydrophobic domains, which allow for further interactions, creating higher molecular hydrophobic domains, which allow for further interactions, creating higher molecular assemblies. assemblies.

Int.Int. J.J. Mol.Mol. Sci. 2016, 17, 1573 1111 ofof 1615

Figure 12. Schematic illustration of spider silk-CNC composite. CBD drives spider silk molecular Figure 12. Schematic illustration of spider silk-CNC composite. CBD drives spider silk molecular ordering in two levels. (A) CBD’s ability to form dimers encourages spider silk alignment; and ordering in two levels. (A) CBD’s ability to form dimers encourages spider silk alignment; and (B) (B) CBD’s ability to bind CNCs drives directional alignment of silk-CBD-CNC composites. CBD’s ability to bind CNCs drives directional alignment of silk-CBD-CNC composites.

3.3. Material Material and and Methods Methods CNCsCNCs producedproduced fromfrom thethe sulfuricsulfuric acidacid hydrolysishydrolysis werewere providedprovided byby MelodeaMelodea Ltd.Ltd. (Rehovot, Israel).Israel). Standard biochemical reagentsreagents andand SigmacoteSigmacote were purchased from Sigma-Aldrich (St. Louis, MO,MO, USA),USA), unlessunless indicatedindicated otherwise.otherwise. 3.1. Design and Expression of the Recombinant Spider Silk Proteins 3.1. Design and Expression of the Recombinant Spider Silk Proteins A synthetic 15-monomer-long dragline spider silk gene was designed from a monomer consensus A synthetic 15-monomer-long dragline spider silk gene was designed from a monomer derived from the native sequence of MaSp1 of Nephila clavipes (Accession P19837) and then cloned consensus derived from the native sequence of MaSp1 of Nephila clavipes (Accession P19837) and then into a pET30a vector, as previously described [45]. In order to produce the spider silk-CBD cloned into a pET30a vector, as previously described [45]. In order to produce the spider silk-CBD fusion gene, SpeI and XhoI restriction sites were introduced to the 50 and 30 of the CBD gene fusion gene, SpeI and XhoI restriction sites were introduced to the 5′ and 3′ of the CBD gene (Accession M73817.1), respectively, via PCR, using the following primers: CBDSpeI: 50-GACTAGT (Accession M73817.1), respectively, via PCR, using the following primers: CBDSpeI: 5′-GACTAGT ATGGCAGCGACATCATCAATGTC-30 and CBDXhoI: 50-CTCGAGATCAAATGTTGCAGAAGTA ATGGCAGCGACATCATCAATGTC-3′ and CBDXhoI: 5′-CTCGAGATCAAATGTTGCAGAAGTA GGATTAATTATTG-30. The SpeI-CBD-XhoI construct was then fused to the 30 of the silk gene in the GGATTAATTATTG-3′. The SpeI-CBD-XhoI construct was then fused to the 3′ of the silk gene in the pET30a-silk vector. The pET30a-silk and pET30a-silk-CBD vectors were transformed into BL21 (DE3) pET30a-silk vector. The pET30a-silk and pET30a-silk-CBD vectors were transformed into BL21 (DE3) cells (Novagen, EMD Chemicals Inc., San Diego, CA, USA). Cells were cultivated in LB broth at 37 ◦C. cells (Novagen, EMD Chemicals Inc., San Diego, CA, USA). Cells were cultivated in LB broth at 37 Protein expression was induced by the addition of 1 mM IPTG (Fisher Scientific, Hampton, NH, USA) °C. Protein expression was induced by the addition of 1 mM IPTG (Fisher Scientific, Hampton, NH, when the optical density (OD) at 600 nm (OD600) was between 0.6 and 0.9. After approximately 4 h of USA) when the optical density (OD) at 600 nm (OD600) was between 0.6 and 0.9. After approximately protein expression, the cells were harvested by centrifugation (10,000× g for 20 min at 4 ◦C). 4 h of protein expression, the cells were harvested by centrifugation (10,000× g for 20 min at 4 °C). 3.2. Purification of 6H-Silk and 6H-Silk-CBD 3.2. Purification of 6H-Silk and 6H-Silk-CBD Bacterial pellets of 400 mL culture were re-suspended in a 5 mL solution of 20 mM NaHPO4, 0.5 M Bacterial pellets of 400 mL culture were re-suspended in a 5 mL solution of 20 mM NaHPO4, NaCl, 10 mM imidazole plus Complete™ protease inhibitor (Roche Diagnostics, Mannheim, Germany) 0.5 M NaCl, 10 mM imidazole plus Complete™ protease inhibitor (Roche Diagnostics, Mannheim, and sonicated on ice for several minutes in pulsed bursts. The soluble and insoluble pellets were Germany) and sonicated on ice for several minutes in pulsed bursts. The soluble and insoluble pellets separated by centrifugation at 15,000 rpm for 10 min, at 4 ◦C. The soluble fraction of the proteins was were separated by centrifugation at 15,000 rpm for 10 min, at 4 °C. The soluble fraction of the proteins filtered using a 0.45 µm syringe filter for faster protein liquid chromatography (FPLC) (GE, Uppsala, was filtered using a 0.45 μm syringe filter for faster protein liquid chromatography (FPLC) (GE, Sweden) purification on a 5 mL HisTrap™ high Performance Ni-NTA column (GE, Uppsala Sweden), Uppsala, Sweden) purification on a 5 mL HisTrap™ high Performance Ni-NTA column (GE, Uppsala pre-equilibrated according to the user manual. Sweden), pre-equilibrated according to the user manual. 3.3. Quantitative Cellulose Binding Reversibility Assay for CBD, Silk and Silk-CBD 3.3. Quantitative Cellulose Binding Reversibility Assay for CBD, Silk and Silk-CBD A binding reversibility assay was performed, as previously described [28]. CBD, silk, and silk-CBD proteinA binding concentrations reversibility ranging assay from was 200 performed,µg/mL to as 1.2 previously mg/mL were described allowed [28]. to CBD, adsorb silk, onto and 30 silk- mg ofCBD prewashed protein concentrations cellulose (Sigmacell ranging 20) from for 30 200 min μg/mL at 25 to◦C. 1.2 After mg/mL the protein’were allowed adsorption to adsorb to cellulose, onto 30 amg series of prewashed of dilutions cellulose was carried (Sigmacell out for the 20) analysis for 30 ofmin the at desorbed 25 °C. After protein the in protein’ the following adsorption manner: to cellulose, a series of dilutions was carried out for the analysis of the desorbed protein in the following manner: the most concentrated protein-cellulose mixture (1.2 mg/mL protein + 30 mg cellulose) was

Int. J. Mol. Sci. 2016, 17, 1573 12 of 16 the most concentrated protein-cellulose mixture (1.2 mg/mL protein + 30 mg cellulose) was diluted to ﬁnal protein concentrations ranging from 1.2 mg/mL to 200 µg/mL, followed by an additional 30 min of mixing. After centrifugation of the diluted test tubes at 13,000× g for 10 min, the concentration of the bound protein was assayed using the Lowry method; the NaOH in the Lowry A solution elutes the bound proteins from the cellulose pellet.

3.4. Production of Composite Silk/Silk-CBD and CNC Sponges The purified silk and silk-CBD proteins were dialyzed against water (dialysis ratio of 1:100) for 18 h, during which, the water was changed four times. After dialysis, the protein aqueous solution was concentrated to 6 wt. % using Vivaspin ultrafiltration (VS2001, 10,000 MWCO, Vivaproducts Inc., Littleton, CO, USA). The concentrated protein solution was mixed with 2.5% w/w CNC to yield the desired silk:CNC ratio. The mixture was then sonicated and poured into a Teflon mold. The mold was placed at ambient room temperature for 1 h and then frozen in a −20 ◦C freezer. The protein-CNC solution was then freeze-dried to generate a sponge.

3.5. CNC and Silk-CBD-CNC Films Preparation Silk-CBD protein after dialysis and concentration was mixed into CNC suspensions (2.5 wt. %). Concentrated silk-CBD protein solution was mixed with 2.5% w/w CNC to yield the desired silk-CBD: CNC ratio by weight. The mixture was then sonicated and 15 mL of silk-CBD-CNC suspensions were cast onto a Sigmacote® treated glass substrates. The silk-CBD-CNC suspensions were dried for 48 h under ambient conditions until a dried ﬁlm was achieved.

3.6. Scanning Electron Microscopy (SEM) Samples of the sponges and ﬁlm were mounted on a metal stub and coated with gold (the coating thickness was 5 nm) using an Extra High Resolution Scanning Electron Microsopy Magellan™ 400L (FEI, Eindhoven, The Netherlands).

3.7. Polarized Light Microscopy (POM) In order to investigate ﬁlms orientation an LC-PolScope system (Cambridge Research and Instruments, Woburn, MA, USA) mounted onto a Nikon Eclipse 80i was used. The LC-PolScope image-processing system has the ability to quantify retardance/birefringence, and indicate molecular orientation. The molecular orientation is proportional to the angular shift in polarized light that occurs as it passes through a birefringent sample. Abrio 2.2 software was used for image acquisition and analysis, background and specimen images were captured under identical conditions.

3.8. Differential Scanning Calorimeter (DSC) Analysis A TA Instruments (New Castle, DE, USA) Q100 differential scanning calorimeter (DSC), with purged dry nitrogen gas ﬂow (50 mL/min), and equipped with a refrigerated cooling system, was used to study the thermal properties of the samples (~5 mg each in Al pan)Heat ﬂow and temperature were calibrated with indium. Temperature-modulated DSC (TMDSC) measurements were performed at a heating rate of 2 K/min, with a modulation period of 60 s and temperature amplitude of 0.318 K. Aluminum and sapphire reference standards were used for calibration of the heat capacity. In TMDSC, the “reversing heat capacity”, which represents a reversed heat effect within the temperature range of the modulation, was measured and calculated.

3.9. Gel Filtration A prepacked Superdex™ 200 10/300 GL column was obtained from Amersham Pharmacia Biotech (Piscataway, NJ, USA) and calibrated using a molecular weight markers kit, (product no. MW-GF-1000, Sigma-Aldrich, St. Louis, MO, USA). Markers included β-amylase (200 kDa), albumin Int. J. Mol. Sci. 2016, 17, 1573 13 of 16

(66 kDa), carbonic anhydrase (29 kDa). Samples were applied to the column using 20 mM sodium phosphate buffer, pH 7.4, as a running buffer, at a ﬂow rate of 0.3 mL/min. Protein was detected with a UV detector, at a wavelength of 280 nm. Samples were collected at the column exit every 0.5 min. The molecular weights of the respective samples were calculated using the linear column calibration curve.

3.10. Dynamic Light Scattering (DLS) Hydrodynamic radius measurements were performed with a DynaPro MSTC800 instrument (Protein Solutions, Inc., Charlottesville, VA, USA). Fresh protein stock solutions were diluted in DDW to a ﬁnal concentration of 2 mg/mL. Samples were centrifuged at 13,000× g for 10 min, to remove insoluble particles. A 1 mL aliquot of each sample was placed directly into a quartz cuvette, gently mixed and analyzed every 10 s, at 25 ◦C. The estimated hydrodynamic radii were calculated using Dynamic V6 data analysis software.

3.11. Small Angle X-ray Scattering (SAXS) Sonicated and un-sonicated protein samples (6% w/v) were transferred into quartz capillary tubes, which were then ﬂame-sealed. The sealed capillary tubes were centrifuged at 6000× g, using a Sigma 1-15PK centrifuge equipped with a capillary rotor. Solution small angle X-ray scattering measurements were performed using a state-of-the-art in-house setup, described elsewhere [46]. Scattering images were radially integrated and the form and the structure factors were analyzed using the X+ program [47]. X+ is a user-friendly, comprehensive, computationally accelerated software program for the analysis of radially integrated signals of solution X-ray scattering from supramolecular self-assembled structures. This system can address both the form and structure factor contributions to the signal.

3.12. Cryo-Transmision Electrom Microscopy (TEM) Samples of silk and silk-CBD (4 µL) were applied to a 300-mesh copper grid coated with lacey carbon (Structure Probe, Inc. Supplies, West Chester, PA, USA). Samples were blotted in a controlled environment held at 25 ◦C and 100% relative humidity, and then plunged into liquid ethane using a CEVS plunger. Specimens were equilibrated at −178 ◦C in a Tecnai F20 transmission electron microscope operated at 200 kV, using a Gatan 626 cooling holder and transfer station, with a TVIPS F415 CCD digital camera at the Department of Chemical esearch Support, The Weizmann Institute of Science, Rehovot 76100, Israel.

4. Conclusions A unique approach to harness the structural alignment and cross-bridging of spider silk proteins to CNCs for the production of silk-CNC composite materials was used; this new approach involves the concentration and sonication of silk-CBD, which seems to influence the alignment and morphology of the composite. We propose that the higher molecular order observed in these composites is due to (1) CBD dimerization (and higher order assemblies); and (2) the binding of CBD to CNCs. These new composites may be useful in a variety of applications, including medical ones. In the model depicted in Figure 12, CBD induces spider silk molecular order at two levels. First, the ability of CBD to form dimers and to mimic the nonrepetitive spider silk terminal function encourages molecular ordering towards the formation of aligned nano-silk fibers. Second, at a higher level, CBD specifically mediates silk-CNC interactions, guiding the formation of structural and ordered composite materials. Further work is currently underway toward exploring the mechanical properties of these composites, and whether the differences seen here, e.g., morphology and thermal properties, translate into improved/altered mechanical properties.

Acknowledgments: We thank the BSF (2010456) and the NIH (EB002520) for support of this work. Int. J. Mol. Sci. 2016, 17, 1573 14 of 16

Author Contributions: Zvi Shtein and Sigal Meirovitch, both supervised by Oded Shoseyov, conceived and designed the experiments; Tal Ben-Shalom performed film experiments; Shaul Lapidot contributed CNCs; Carmen Tamburu and Uri Raviv performed SAX experiments and analysis; Xiao Hu and Jonathan A. Kluge, both supervised by David L. Kaplan, analyzed sponges by SEM and DSC. Conflicts of Interest: The authors declare no conflict of interest.

References

1. Vollrath, F. Strength and structure of spiders’ silks. Rev. Mol. Biotechnol. 2000, 74, 67–83. [CrossRef] 2. Rising, A.; Hjälm, G.; Engström, W.; Johansson, J. N-terminal nonrepetitive domain common to dragline, flagelliform, and cylindriform spider silk proteins. Biomacromolecules 2006, 7, 3120–3124. [CrossRef] [PubMed] 3. Ayoub, N.A.; Garb, J.E.; Tinghitella, R.M.; Collin, M.A.; Hayashi, C.Y. Blueprint for a high-performance biomaterial: Full-length spider dragline silk genes. PLoS ONE 2007, 2, e514. [CrossRef][PubMed] 4. Simmons, A.; Ray, E.; Jelinski, L. Solid-state 13C NMR of Nephila clavipes dragline silk establishes structure and identity of crystalline regions. Macromolecules 1994, 27, 5235–5237. [CrossRef] 5. Hinman, M.B.; Jones, J.A.; Lewis, R.V. Synthetic spider silk: A modular fiber. Trends Biotechnol. 2000, 18, 374–379. [CrossRef] 6. Parkhe, A.D.; Seeley, S.K.; Gardner, K.; Thompson, L.; Lewis, R.V. Structural studies of spider silk proteins in the fiber. J. Mol. Recognit. 1997, 10, 1–6. [CrossRef] 7. Hayashi, C.Y.; Shipley, N.H.; Lewis, R.V. Hypotheses that correlate the sequence, structure, and mechanical properties of spider silk proteins. Int. J. Biol. Macromol. 1999, 24, 271–275. [CrossRef] 8. Rising, A.; Nimmervoll, H.; Grip, S.; Fernandez-Arias, A.; Storckenfeldt, E.; Knight, D.P.; Vollrath, F.; Engström, W. Spider silk proteins—Mechanical property and gene sequence. Zool. Sci. 2005, 22, 273–281. [CrossRef][PubMed] 9. Ko, F.K.; Jovicic, J. Modeling of mechanical properties and structural design of spider web. Biomacromolecules 2004, 5, 780–785. [CrossRef][PubMed] 10. Kluge, J.A.; Rabotyagova, O.; Leisk, G.G.; Kaplan, D.L. Spider silks and their applications. Trends Biotechnol. 2008, 26, 244–521. [CrossRef][PubMed] 11. Lewis, R. Unraveling the weave of spider silk. Bioscience 1996, 46, 636–638. [CrossRef] 12. Fahnestock, S.R.; Bedzyk, L.A. Production of synthetic spider dragline silk protein in Pichia pastoris. Appl. Microbiol. Biotechnol. 1997, 47, 33–39. [CrossRef][PubMed] 13. Fahnestock, S.R.; Irwin, S.L. Synthetic spider dragline silk proteins and their production in Escherichia coli. Appl. Microbiol. Biotechnol. 1997, 47, 23–32. [CrossRef][PubMed] 14. Miao, Y.; Zhang, Y.; Nakagaki, K.; Zhao, T.; Zhao, A.; Meng, Y.; Nakagaki, M.; Park, E.Y.; Maenaka, K. Expression of spider flagelliform silk protein in Bombyx mori cell line by a novel Bac-to-Bac/BmNPV baculovirus expression system. Appl. Microbiol. Biotechnol. 2006, 71, 192–199. [CrossRef][PubMed] 15. Barr, L.A.; Fahnestock, S.R.; Yang, J. Production and purification of recombinant DP1B silk-like protein in plants. Mol. Breed. 2004, 13, 345–356. [CrossRef] 16. Xia, X.X.; Qian, Z.G.; Ki, C.S.; Park, Y.H.; Kaplan, D.L.; Lee, S.Y. Native-sized recombinant spider silk protein produced in metabolically engineered Escherichia coli results in a strong fiber. Proc. Natl. Acad. Sci. USA 2010, 107, 14059–14063. [CrossRef][PubMed] 17. Stark, M.; Grip, S.; Rising, A.; Hedhammar, M.; Engström, W.; Hjälm, G.; Johansson, J. Macroscopic fibers self-assembled from recombinant miniature spider silk proteins. Biomacromolecules 2007, 8, 1695–1701. [CrossRef][PubMed] 18. Jin, H.J.; Kaplan, D.L. Mechanism of silk processing in insects and spiders. Nature 2003, 424, 1057–1061. [CrossRef][PubMed] 19. Vollrath, F.; Knight, D.P. Liquid crystalline spinning of spider silk. Nature 2001, 410, 541–548. [CrossRef] [PubMed] 20. Abitbol, T.; Rivkin, A.; Cao, Y.; Nevo, Y.; Abraham, E.; Ben-Shalom, T.; Lapidot, S.; Shoseyov, O. Nanocellulose, a tiny fiber with huge applications. Curr. Opin. Biotechnol. 2016, 39, 76–88. [CrossRef] [PubMed] Int. J. Mol. Sci. 2016, 17, 1573 15 of 16

21. Rivkin, A.; Meirovitch, S.; Shoseyov, O.; Kaplan, D.L. Bio-inspired elastic composites of nano crestalline celulose and cellulose binding resilin. In Proceedings of the 34th Risø International Symposium on Materials Science: Processing of Fibre Composites—Challenges for Maximum Materials Performance, Roskilde, Denmark, 2–5 September 2013. 22. Šturcová, A.; Davies, G.R.; Eichhorn, S.J. Elastic modulus and stress-transfer properties of tunicate cellulose whiskers. Biomacromolecules 2005, 6, 1055–1061. [CrossRef][PubMed] 23. Noishiki, Y.; Nishiyama, Y.; Wada, M.; Kuga, S.; Magoshi, J. Mechanical properties of silk fibroin–microcrystalline cellulose composite films. J. Appl. Polym. Sci. 2002, 86, 3425–3429. [CrossRef] 24. Shoseyov, O.; Doi, R.H. Essential 170-kDa subunit for degradation of crystalline cellulose by Clostridium cellulovorans cellulase. Proc. Natl. Acad. Sci. USA 1990, 87, 2192–2195. [CrossRef][PubMed] 25. Goldstein, M.A.; Takagi, M.; Hashida, S.; Shoseyov, O.; Doi, R.H.; Segel, I.H. Characterization of the cellulose-binding domain of the Clostridium cellulovorans cellulose-binding protein A. J. Bacteriol. 1993, 175, 5762–5768. [PubMed] 26. Tomme, P.; Warren, R.A.J.; Gilkes, N.R. Cellulose hydrolysis by bacteria and fungi. Adv. Microb. Physiol. 1995, 37, 1–81. [PubMed] 27. Bolam, D.N.; Ciruela, A.; McQueen-Mason, S.; Simpson, P.; Williamson, M.P.; Rixon, J.E.; Boraston, A.; Hazlewood, G.P.; Gilbert, H.J. Pseudomonas cellulose-binding domains mediate their effects by increasing enzyme substrate proximity. Biochem. J. 1998, 331, 775–781. [CrossRef][PubMed] 28. Haynes, C.A.; Norde, W. Globular proteins at solid/liquid interfaces. Colloids Surf. B Biointerfaces 1994, 2, 517–566. [CrossRef] 29. Li, M.; Lu, S.; Wu, Z.; Yan, H.; Mo, J.; Wang, L. Study on porous silk fibroin materials. I. Fine structure of freeze dried silk fibroin. J. Appl. Polym. Sci. 2001, 79, 2185–2191. [CrossRef] 30. Li, M.; Wu, Z.; Zhang, C.; Lu, S.; Yan, H.; Huang, D.; Ye, H. Study on porous silk fibroin materials. II. Preparation and characteristics of spongy porous silk fibroin materials. J. Appl. Polym. Sci. 2001, 79, 2192–2199. [CrossRef] 31. Nazarov, R.; Hyoung-Joon, J.; Kaplan, D.L. Porous 3D scaffolds from regenerated Antheraea pernyi silk fibroin. Biomacromolecules 2004, 5, 718–726. [CrossRef][PubMed] 32. Svagan, A.J.; Samir, M.S.; Berglund, L.A. Biomimetic foams of high mechanical performance based on nanostructured cell walls reinforced by native cellulose nanofibrils. Adv. Mater. 2008, 20, 1263–1269. 33. Wang, X.; Kluge, J.A.; Leisk, G.G.; Kaplan, D.L. Sonication-induced gelation of silk fibroin for cell encapsulation. Biomaterials 2008, 29, 1054–1064. [CrossRef][PubMed] 34. Hu, X.; Shmelev, K.; Sun, L.; Gil, E.S.; Park, S.H.; Cebe, P.; Kaplan, D.L. Regulation of silk material structure by temperature-controlled water vapor annealing. Biomacromolecules 2011, 12, 1686–1696. [CrossRef][PubMed] 35. Cunniff, P.; Fossey, S. Mechanical and thermal properties of dragline silk from the spider Nephila clavipes. Polym. Adv. Technol. 1994, 5, 401–410. [CrossRef] 36. Guess, K.B.; Viney, C. Thermal analysis of major ampullate (drag line) spider silk: The effect of spinning rate on tensile modulus. Thermochim. Acta 1998, 315, 61–66. [CrossRef] 37. Magoshi, J.; Nakamura, S. Studies on physical properties and structure of silk. Glass transition and crystallization of silk fibroin. J. Appl. Polym. Sci. 1975, 19, 1013–1015. [CrossRef] 38. Xu, Y.; Foong, F.C. Characterization of a cellulose binding domain from Clostridium cellulovorans endoglucanase-xylanase D and its use as a fusion partner for soluble protein expression in Escherichia coli. J. Biotechnol. 2008, 135, 319–325. [CrossRef][PubMed] 39. Craig, S.J.; Shu, A.; Xu, Y.; Foong, F.C.; Nordon, R. Chimeric protein for selective cell attachment onto cellulosic substrates. Protein Eng. Des. Sel. 2007, 20, 235–241. [CrossRef][PubMed] 40. Jackson, C.; O’Brien, J. Molecular weight distribution of Nephila clavipes dragline silk. Macromolecules 1995, 28, 5975–5977. [CrossRef] 41. Sponner, A.; Schlott, B.; Vollrath, F.; Unger, E.; Grosse, F.; Weisshart, K. Characterization of the protein components of Nephila clavipes dragline silk. Biochemistry 2005, 44, 4727–4736. [CrossRef][PubMed] 42. Hedhammar, M.; Rising, A.; Grip, S.; Martinez, A.S.; Nordling, K.; Casals, C.; Stark, M.; Johansson, J. Structural properties of recombinant nonrepetitive and repetitive parts of major ampullate spidroin 1 from Euprosthenops australis: Implications for fiber formation. Biochemistry 2008, 47, 3407–3417. [CrossRef] [PubMed] Int. J. Mol. Sci. 2016, 17, 1573 16 of 16

43. Askarieh, G.; Hedhammar, M.; Nordling, K.; Saenz, A.; Casala, C.; Rising, A.; Johansson, J.; Knight, D.S. Self-assembly of spider silk proteins is controlled by a pH-sensitive relay. Nature 2010, 465, 236–238. [CrossRef][PubMed] 44. Hagn, F.; Eisoldt, L.; Hardy, J.G.; Vendrely, C.; Coles, M.; Scheibel, T.; Kessler, H. A conserved spider silk domain acts as a molecular switch that controls ﬁbre assembly. Nature 2010, 465, 239–242. [CrossRef] [PubMed] 45. Bini, E.; Foo, C.W.; Huang, J.; Karageorgiou, V.; Kitchel, B.; Kaplan, D.L. RGD-Functionalized Bioengineered Spider Dragline Silk Biomaterial. Biomacromolecules 2006, 7, 3139–3145. [CrossRef][PubMed] 46. Nadler, M.; Steiner, A.; Dvir, T.; Szekely, O.; Szekely, P.; Ginsburg, A.; Asor, R.; Resh, R.; Tamburu, C.; Peres, M.; et al. Following the structural changes during zinc-induced crystallization of charged membranes using time-resolved solution X-ray scattering. Soft Matter 2011, 7, 1512–1523. [CrossRef] 47. Székely, P.; Ginsburg, A.; Ben-Nun, T.; Raviv, U. Solution X-ray scattering form factors of supramolecular self-assembled structures. Langmuir 2010, 26, 13110–13129. [CrossRef][PubMed]

© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).