Remote Magnetic Orientation of 3D Collagen Hydrogels for DirectedNeuronal RegenerationMerav Antman-Passig and Orit Shefi*

Faculty of Engineering and Bar Ilan Institute of Nanotechnologies and Advanced Materials, Bar Ilan University, Ramat Gan 5290002,Israel

*S Supporting Information

ABSTRACT: Hydrogel matrices are valuable platforms for neuro-nal tissue engineering. Orienting gel fibers to achieve a directedscaffold is important for effective functional neuronal regeneration.However, current methods are limited and require treatment of gelsprior to implantation, ex-vivo, without taking into consideration thepathology in the injured site. We have developed a method tocontrol gel orientation dynamically and remotely in situ. We havemixed into collagen hydrogels magnetic nanoparticles then appliedan external magnetic field. During the gelation period the magneticparticles aggregated into magnetic particle strings, leading to thealignment of the collagen fibers. We have shown that neuronswithin the 3D magnetically induced gels exhibited normal electricalactivity and viability. Importantly, neurons formed elongatedcooriented morphology, relying on the particle strings and fibersas supportive cues for growth. The ability to inject the mixed gel directly into the injured site as a solution then to control scaffoldorientation remotely opens future possibilities for therapeutic engineered scaffolds.

KEYWORDS: collagen hydrogel, magnetic nanoparticles, neuronal regeneration, 3D platform, aligned scaffold

Neural tissue pathologies resulting from either physicaltrauma or neurodegenerative diseases detrimentally affect

the quality of life of many patients worldwide. The ability ofmammalian neuronal cells to spontaneously repair and regainfunctionality is limited, presenting a critical therapeuticchallenge for researchers.1,2 Major efforts have been devotedto the development of supportive tissues and scaffolds, aimed atcreating modified environments for promoting neuronalregeneration. It has been shown that guiding and directingneuronal outgrowth during the regeneration period canenhance neuronal repair and recovery.3,4

A fundamental mechanism of neuronal guidance is thesensitivity of neurons to extracellular topography.5 Our ownand others’ previous studies have shown that interactions withplanar surfaces decorated with organized topographicalelements even of nanometer heights lead to directed neuronalgrowth patterns.6−10 However, in order to develop a supportivemicroenvironment for neuronal regeneration in vivo, implant-able 3-dimensional scaffolds are required.11 Various strategies,utilizing nanofibrous scaffolds12,13 as well as different types ofhydrogels14−16 have been examined as platforms for neuronaltissue engineering affecting axonal extension and cell spreading.Several methods have been used to fabricate scaffolds withaligned fibrils as cell-directing cues including electorspin-ning,12,17 microfluidics,18 strain-induced alignment,19,20 andfabrication of prescribed patterns within the 3D constructs.21,22

Strong magnetic fields have also been used to modify fiber

orientation.23−25 In all of these studies, scaffolds were designedand fabricated ex vivo, prior to the interaction with the cells orimplantation site. Facilitating scaffold alignment in situ is still acompelling challenge.We have developed a 3D scaffold, based on a collagen

hydrogel, combined with magnetic nanoparticles. In the body,natural collagen forms highly organized fibers that profoundlyinfluence tissue mechanical properties and cell organiza-tion.26,27 Collagen hydrogels imitate the native neuronalextracellular matrix (ECM), exhibiting biodegradability, weakantigenicity and good biocompatibility characteristics.28,29

Moreover, hydrogels express mechanical properties similar tosoft tissues and exhibit high porosity that facilitates nutrientsand waste. Importantly, hydrogels possess the ability to beinjected in vivo as a liquid that gels at body temperature.30,31

The use of magnetic-responsive constructs for tissue regener-ation in which various matrices are impregnated with particleshas been investigated for bone repair,32−34 stem cell differ-entiation,35 and tissue engineering scaffolds.36 Alginate gelscontaining magnetic nanoparticles (MNPs) have promotedendothelial cell organization.37 MNPs functionalized withgrowth factors improved the growth of nasal olfactory mucosacells38 and PC12 differentiation.39

Received: January 11, 2016Revised: March 1, 2016

Letter

pubs.acs.org/NanoLett

© XXXX American Chemical Society A DOI: 10.1021/acs.nanolett.6b00131Nano Lett. XXXX, XXX, XXX−XXX

Here, we report on a magnetic gel-based scaffold that can bealigned in situ dynamically and remotely in response to apathological need. We embedded magnetic particles in collagenliquid suspension and allowed the collagen to solidify under apermanent magnetic field. Application of the magnetic fieldduring the solidification period leads to controlled aggregationof magnetic particles within the gel, forming “magnetic particlestrings”, which aggregate along the magnetic field lines.Furthermore, we found that utilizing this method with differentmagnetic particles affected gel complex formation dynamics andallowed us to achieve alignment of the collagen fibers. Once theexternal magnetic field was removed the particle, aggregatesremained embedded within the collagen gel and gel fibers

maintained their induced orientation. Both particles and fibersmay serve as topographic guiding cues for directing neuriteoutgrowth (Figure 1).To better control gel complex structure orientation and in

order to achieve maximal collagen fiber alignment, we examinedmagnetic particles of different sizes and compositions. Nano-and micro-particles with different parameters were mixed with acollagen precursor gel solution (collagen-type-I, diluted by theaddition of 10X L-15 and brought to neutral pH by 7.5%sodium carbonate; detailed description in Supporting Informa-tion). The liquid solution was then placed on a tunableplatform between two bar magnets and allowed to solidifyunder effective magnetic fields of 162G to 2110G (Figure S1).

Figure 1. Schematic overview of the experimental procedure showing collagen fiber alignment induced by magnetic particles and magnetic actuation:(A) Liquid collagen suspension was prepared containing neurons (orange) and MPs (red). (B) Suspension was plated on top of coverslips andallowed to solidify either spontaneously (top), or under the influence of a magnetic field of 255G set by two parallel bar magnets (red-green bars)(bottom). (C) Zoom into collagen gels. (Top) control gel reveals random distribution of MNP aggregates (red) and random collagen fiberorientation (blue lines). (Bottom) In gels solidified under the influence of a magnetic field MPs form aggregate strings aligned along the direction ofthe magnetic lines (red particles) and collagen fibers align as well (blue lines). (D) Neuronal growth was followed for 7 days as neurons developedneurites in the direction of the magnetically aligned cues.

Figure 2. Formation of aligned magnetic particle strings within the collagen gels. Two gel concentrations were examined: 1.4 mg/mL (top panels)and 3 mg/mL (bottom panels). The effect of one magnetic field is presented (255G).(A) MPs (1 μm) mixed into collagen gels (1.4 mg/mL) andsolidified with no magnetic field form random aggregates distributed homogeneously in the gel. (B) MNPs (100 nm) form magnetic strings incollagen gels (1.4 mg/mL) solidified in the presence of a magnetic field (255G). (C) MPs (1 μm) form magnetic strings in collagen gels (1.4 mg/mL) that have solidified in the presence of a magnetic field. (D) MNPs (100 nm) mixed into collagen gels (3 mg/mL) and solidified with nomagnetic field. Aggregates are formed and distributed randomly (lighter aggregates are out of phase aggregates). (E) Magnetic strings are formed ingels (3 mg/mL) containing MNPs (100 nm) solidified in the presence of a magnetic field. (F) MPs (1 μm) form magnetic strings in gels (3 mg/mL)solidified in the presence of a magnetic field. Scale bar = 100 μm.

Nano Letters Letter

DOI: 10.1021/acs.nanolett.6b00131Nano Lett. XXXX, XXX, XXX−XXX

B

Applying relatively weak magnetic field, around 255G, hasresulted in more uniform distribution of the magnetic strings,therefore, it was chosen as the actuator magnetic field (theinfluence of higher magnetic field on cell alignment is discussedbelow). Two gel concentrations were examined, 3 mg/mL(Figure 2, bottom panels) and 1.4 mg/mL (Figure 2, toppanels). For both concentrations, the embedded magneticparticles aggregated into distinct line-shaped strings, whereas,without the external magnetic field the particles werehomogeneously dispersed within the gel (Figure 2A,D). Inthe low concentration gels (1.4 mg/mL) magnetic stringstended to aggregate at a lower Z-plane and did not disperse in3D. In high concentration gels (3 mg/mL) multiplaneaggregation was observed which is beneficial for 3D scaffolds.Therefore, we based the 3D platforms on the 3 mg/mLcollagen gels. Next, we have compared the effect of magneticparticle size on gel alignment, focusing on particles at thenanoscale, suitable for tissue engineering platforms. Usinguncoated MNPs of 50 and 100 nm (hydrodynamic diameter)has demonstrated better gel fiber alignment for the 100 nmMNPs (for both magnetite and maghemite cores). Notably,functional coating of the MNPs (with dextransulfate, starch, orcitric acid salt, Supporting In formation) has led to less fiberalignment than the uncoated MNPs. To note, embedding in thecomplex gels micron size magnetic particles (diameter of 1 and3 μm) also demonstrated nonsufficient level of fiber alignment.To characterize gel complex formation, we have used the

uncoated 100 nm MNPs (magnetite core, average dry diameter

of 10 nm) that have resulted in maximal fiber alignment. Ascanning electron microscope (SEM) image of a control gelcontaining the same MNPs that has been solidified without theinfluence of an external magnetic field, presents no preferredfiber orientation (Figure 3A). Under an external magnetic field,the fibers of the particle-containing gel tended to alignaccording to the particle strings orientation (Figure 3B).A confocal reflectance microscopy (CRM) of the 3D

collagen gels (illumination with 488 nm, detection between474 and 494 nm) was performed (Figure 3D−E). It can beseen that the control samples show a random distribution of thecollagen fibers, in all dimensions (Figure 3D), whereas in themagnetically actuated gels, collagen fibers align along thedirection of the magnetic particle strings (Figure 3E) (FigureS3A−C,E and Movie S1). In order to obtain a quantitativeevaluation of the effect on the fibers, images of magneticallyaligned gels (SEM and CRM) were compared to controlsamples either without actuating magnetic fields or withoutmagnetic particles. Two-dimensional (2D) fast Fourier trans-form (FFT) was calculated40 for the SEM images (Figure 3C)and for the CRM images (Figure 3F). We have acquired CRMimages from random areas of the gels at different depths (notshown) and demonstrated a significantly higher degree ofcorrelation between fibers in the aligned magnetic gels than inthe control gels, solidified under 255G with no MNPs (anaverage peak value of 0.06 ± 0.01 [au] and 0.03 ± 0.004 [au]respectively). In addition, using the same FFT analysis, theorientation of fibers of the magnetically actuated gels was

Figure 3. Collagen fiber alignment induced by magnetic string formation. (A) SEM image of collagen gel (3 mg/mL) containing 100 nm MNPsshowing random fiber orientation. (B) SEM image of the same collagen gel solidified under the influence of 255G magnetic field, showing partialalignment of collagen fibers (white arrow) correlating to magnetic aggregate string direction (red arrows). (C) FFT analysis plot showing fiberorientation corresponding to indent depicted area in image B in blue (no particles are included within the analyzed area) compared to image A inred. (D) CRM image of collagen gel (3 mg/mL) solidified under the influence of 255G magnetic field without MNPs and (E) CRM image of the 3mg/mL collagen gel containing 100 nm MNPs solidified under the influence of 255G magnetic field, showing fiber alignment (light blue). FFTanalysis images for areas depicted in D and E are shown in corresponding top right corner of CRM images. Scale bar = 75 μm. (F) FFT analysis plotof same depicted areas in D (blue) and E (red).

Nano Letters Letter

DOI: 10.1021/acs.nanolett.6b00131Nano Lett. XXXX, XXX, XXX−XXX

C

compared to the orientation of the corresponding image ofparticle strings demonstrating closely matched peak angles(Figure S3D).To have a better understanding of the gel complex formation

along with fiber alignment, time lapse microscopy imaging wasperformed (NIKON TE2000 with a mounted Retiga camera).Collagen gel samples, with and without MNPs, were imaged

along the entire gelation process, starting from the liquid phaseand up to 25 min. During spontaneous gel polymerization,collagen nucleates and elongates, creating the hydrogel matrix.This process is usually random, leading to tangled fibers of nodiscernible organization, as can be seen in Movie S2.Embedding the 100 nm MNPs in the collagen without applyingan actuating magnetic field has led to the formation of

Figure 4. Measurements of physical properties of magnetically MNPs containing gels. (A) Rheometry measurements of complex modulus curves forcontrol gels, with or without particles, solidified under the presence of magnetic field, compared to magnetically aligned gels. (B) Turbiditymeasurements of gelation kinetics for gels with and without MNPs.

Figure 5. Neuronal orientation in magnetically aligned gels (A) Confocal z stack image of a leech neuron grown in a magnetically aligned gel (3 mg/mL, MNPs of 100 nm, a magnetic field of 255G). Neuron is fluorescently labeled in red. Background is a single CRM image from stack, showingfiber alignment. Yellow indent shows neurite following aligned collagen fibers. (B) Corresponding 3D cuboid representation of the neuron withneurite tracing analysis. The long axis of the cuboid represents the direction of growth. Red arrow represents the orientation of the magnetic particlestrings. (C) Elongation of neurons is measured by the aspect ratio between the long axis and the short axis of the cuboid for neurons in magneticallyaligned gels and in control gels. (D) Histogram of deviation of orientation angle Δθ binned ±15°. Deviation of orientation angle is defined as thedifference between the cuboid representation angle and magnetic string orientation (depicted in green arrow). Statistical analysis of the distributionof Δθ is not normal or uniform (one sample Kolmogorov−Smirnov Test). Scale bar = 100 μm, *** p < 0.0005.

Nano Letters Letter

DOI: 10.1021/acs.nanolett.6b00131Nano Lett. XXXX, XXX, XXX−XXX

D

dispersed aggregates of MNPs with no effect on fiberorientation. It can be seen that aggregates can move freelywhen in collagen liquid phase and stabilize during the gelationprocess (Movie S3). Applying the actuating magnetic fieldduring solidification presents a movement of the forming MNPstrings up the magnetic gradient, leading to flow. As gel fiberspolymerize, the MNP strings are no longer free to move. Itseems that the orientation of strings movement together withthe collagen flow at the beginning of the gelation phase lead topreferable directionality of the newly formed fibers. FollowingMNP stabilization, collagen fibers keep elongating, retaining theinitial orientation in a template-like manner (Movie S4). Tonote, gel complexes with less particle movement, such as thegels containing microscale particles under 255G, demonstratenonoriented gel fibers.Rheological measurements (detailed description in Support-

ing Information) have showed that our gels are fairly compliantand correlate previous studies of neuronal growth in hydrogelsas well as moduli of nervous tissue.41−43 No significant changesof the physical properties of the gels and gelation kinetics havebeen detected following the addition of MNPs to the gels(Figure 4).Next, to examine the magnetically aligned gel as a scaffold for

directed neuronal growth, primary neurons were seeded withinthe gel, prior to the gelation. We used neurons of a simplemodel system, the medicinal leech (Hirudo medicinalis), thatcan develop at low density cultures allowing for tracking ofsingle neurons and neurites. Neurons were dissociated out ofthe adult central nervous system (CNS), similarly to 2D cultureprotocol8,9 and were incorporated in the collagen gel precursorliquid solution. We allowed the gel solution containingmagnetic particles and cells to solidify under an externalmagnetic field (255G) for 40 min. Once the field was removedand the solidification process was complete, growth mediumwas added (conditioned L15) to allow neuronal regeneration.During the outgrowth period within the gel neurites initiated

from the cell soma then branched and developed into a

complex 3D dendritic tree. After 7 days of growth, neuronswere stained (α-tubulin, Santa Cruz Biotechnology, Inc.) andimaged by confocal microscopy (Figure 5A and Movies S5 andS6). 3D data from confocal z stacks was extracted and analyzedsemiautomatically using a neurite tracer44 (ImageJ plugin, USNational Institutes of Health, Bethesda, MD) and a MATLABscript. Then, the representation of the spatial arrangement ofeach cell was plotted as a cuboid (Figure 5B) and the aspectratio between the long axis and the short axis was measured.High aspect ratio represents an elongated neuronal shape andan aspect ratio closer to 1 represents a multidirectional growth.We found that growth within the oriented gels have led tosignificantly elongated neuronal shape with an aspect ratio closeto 3 (twice the aspect ratio of the control) (Figure 5C). Toexamine the effect on the neuronal directionality, we designatedthe orientation of the longest axis as the direction of neuronalgrowth and measured the deviation angle Δθ from magneticstring orientation (detailed measurements of single neurons arepresented in Figure S4). Notably, the growth of neurons thatregenerated within the aligned gels was cooriented; themajority of the cells exhibited highly significant correlationwith the magnetic line orientation (60% of the cells developedwithin the range of Δθ < ±15° compared to the magnetic linesand additional 12% of cells exhibited moderate correlation ofcell orientation of Δθ < ±30°; chi-square value of −16.88 df 2;Figure 5D).To reveal the nature of neuronal alignment within the

magnetically actuated gels, we have examined the neuronalgrowth in several gel complexes with different gel concen-trations, particles, and magnetic fields. Neurons grown in gelscontaining aligned magnetic strings but without fiber alignment(for example, 3 mg/mL gels with particles of 1 μm, solidifiedunder 255G and 3 mg/mL gels with MNPs, solidified under620G) have demonstrated less elongation and less direction-ality (Figure S5) than within the aligned gel complex (aspresented in Figure 5) suggesting that the induced alignment ofcollagen fibers is key in neurite pathfinding.

Figure 6. Neuron viability assay. (A) At 7 days after seeding, viability of leech neurons in magnetically aligned gels was measured using a live/deadassay. (B) Electrophysiological activity of a single neuron after 6 days in a magnetically aligned gel. (C) Schematic representation of a neuronillustrating the measured morphometric parameters. Numbers indicate neurites, green arrow represents neurite length. (D) Average neurite length ofneurons in magnetically aligned gels vs control gels. (E) Average total branching length of neurons in treated vs control gels. (F) Average neuritenumber of neurons in magnetically aligned gels vs control gels. *p < 0.05, **p < 0.01.

Nano Letters Letter

DOI: 10.1021/acs.nanolett.6b00131Nano Lett. XXXX, XXX, XXX−XXX

E

To examine further the scaffold as a regenerative platform,we have assessed viability. After 7 days in culture, live and deadcells were labeled (Live/Dead imaging kit R37601, LifeTechnologies). We found that there was no decrease inviability within the magnetic gel in comparison to the controlgels (Figure 6A) similarly to MNPs effects on viability inprevious studies.45,46 We also used a standard current-clampintracellular recording to monitor spike activity. Neuronalspikes were sampled for several neurons demonstratingcharacteristic spontaneous spikes as well as action potentialsin response to 2−10 nA stimuli (Figure 6B). Moreover,measurements of morphometric parameters of neurons withinthe magnetically aligned gel demonstrated a similar neuritelength, less branched dendritic trees and slightly shorter totaltree length, compared to control cells, within gels with noparticles or magnetic field (Figure 6C−E). Within the alignedgels, neurons presented an elongated and directed growth withless branching as can be expected for efficient neuronalpathfinding within an anisotropic environment. Such branchingformation and shape maturation also have been observed inother 2D and 3D setups for directed growth.10,47−49

To further validate the tunability and efficiency of themagnetically aligned gels in directing neuronal growth, weexamined mammalian cell lines. We seeded PC12 cells andused the collagen gel as the differentiating platform.50

Following treatment with nerve growth factor (NGF), PC12cells have developed neurites in three dimensions within the gelmatrix (for detailed description see Supporting Information).After 5 days in culture, cells were stained and analyzed. As inthe primary neurons, PC12 cells grown in the magneticallyaligned gels developed cooriented neurites (Figure S6A).Growth orientation analysis revealed that 56.5% of the cellsexhibited a significant correlation of growth with the magneticline orientation, within Δθ < ± 30° (Figure S6C; chi-squarevalue of 9.739 df 2). Within this range, 34.7% exhibited highcorrelation (Δθ < ±15°; not shown). Additionally, we haveused these cells to examine the response to magnetically alignedgels of low collagen concentration (Figure S6B) demonstratingsimilar trend (Figure S6D; Δθ < ±30° chi-square value of 8.313df 2). To note, within the low concentration gels, orientedneurons were observed more clearly in proximity to themagnetic strings. Because these gels are more diluted andformed thinner fibers, neurons have relied on the particles astopographical cues.In summary, we have developed a method to create 3D

platforms based on collagen hydrogel combined with MNPs, asscaffolds for functional neuronal regeneration. Our methodallows controlling the alignment of the magnetic gel-basedscaffolds dynamically and remotely. By actuating a relatively lowexternal magnetic field, the magnetic elements have aggregatedinto magnetic particle strings along the magnetic lines withinthe gel. We have demonstrated that these strings have served asphysical cues for neurons that developed in close proximity tothe particles, leading to elongated and directed growth pattern.Interestingly, the movement and alignment of the magneticparticles during the gelation period (less than an hour) have ledto the alignment of the collagen fibers as well. As gelationproceeds, following MNP stabilization, collagen fibers keepelongating, retaining the initial orientation in a template-likemanner. We found this process sensitive to the formation anddynamics of the MNP strings within the collagen.We have shown that in gels that combined the alignment

process of both particles and fibers, the neuronal branching tree

was significantly cooriented with the scaffold. On the basis ofthe neuronal response to the different examined gels, we haveconcluded that high-concentration collagen gels mixed withnanometric magnetic particles (10 nm core diameter)demonstrate optimal conditions for directing neuronalregeneration.Recent studies have focused on controlling micro- and

nanostructures, scaffold properties, and orientation, includingthose of hydrogels, for enhanced regeneration.15,51,52 Forexample, embedding metallic nanometric elements has beenfound to be effective in promoting cell−scaffold interactions,affecting growth as well as the activity of excitable membranecells.53,54 Usually, these methods require ex-vivo manipulationsprior to implantation. The ability to implant a scaffold directlyinto the injured site, then manipulating its orientationaccording to the pathological need, is still a therapeuticchallenge. We propose a method to overcome this challenge bycontrolling gel orientation remotely post implantation duringthe gelation period. In our method, the MNP-containingcollagen can be injected as a liquid directly into the injured siteand orientated via a short-term application of an externalmagnetic field. Such an approach holds the potential to fillcomplex defect shapes and to be aligned according to thepathology need. The ability to program the magnetic fieldsallows designing different microstructure orientations of theregenerative platform. Our study, together with recent advancesinvolving magnetic nanoparticles55 and the penetrability ofmagnetic fields into human tissue, open future possibilities forremotely controlled scaffolds for tissue engineering applica-tions.

■ ASSOCIATED CONTENT*S Supporting InformationThe Supporting Information is available free of charge on theACS Publications website at DOI: 10.1021/acs.nano-lett.6b00131.

Set up, phase images, confocal z stack, neuronal growth,PC12 alignment, and methods description. (PDF)Collagen fibers align along the direction of the magneticparticles strings (CRM Z-stack). (AVI)Time lapse imaging of collagen gel formation with nomagnetic particles or magnetic field. (AVI)Time lapse imaging of the gelation process of collagenembedded with magnetic nanoparticles with no externalmagnetic field. (AVI)Time lapse imaging of the gelation process of collagengel embedded with magnetic nanoparticles, under theinfluence of external magnetic field. (AVI)Z-stack confocal imaging a neuron within an aligned gel.(AVI)Z-stack confocal imaging a neuron within an aligned gel.(AVI)

■ AUTHOR INFORMATIONCorresponding Author*E-mail: orit.shefi@biu.ac.il.NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSThis work was partially supported by the Israel ScienceFoundation Individual Grants #1403/11 (O.S.). We would like

Nano Letters Letter

DOI: 10.1021/acs.nanolett.6b00131Nano Lett. XXXX, XXX, XXX−XXX

F

to thank Dr. Rachel Levi-Drummer for her assistance with thestatistical analysis. We thank Prof. Aryeh Weiss for hisassistance and advice with the time lapse microscopy. Wethank Matan Tayar for contributing to the morphologicalmeasurements, Chaim Gartenberg for assistance with gelphysical properties measurements and Smadar Klugman-Arvatzfor technical help with Rheometry.

■ REFERENCES(1) Siemionow, M.; Brzezicki, G. Int. Rev. Neurobiol. 2009, 87, 141−172.(2) Chiono, V.; Tonda-Turo, C.; Ciardelli, G. Artificial Scaffolds forPeripheral Nerve Reconstruction. In International Review of Neuro-biology; Stefano, G., Pierluigi, T., Bruno, B., Eds.; Academic Press:Waltham, MA, 2009; Chapter 9, Vol. 87, pp 173−198.(3) Schmidt, C. E.; Leach, J. B. Annu. Rev. Biomed. Eng. 2003, 5, 293−347.(4) Kim, Y.-t.; Haftel, V. K.; Kumar, S.; Bellamkonda, R. V.Biomaterials 2008, 29, 3117−3127.(5) Huber, A. B.; Kolodkin, A. L.; Ginty, D. D.; Cloutier, J.-F. Annu.Rev. Neurosci. 2003, 26, 509−563.(6) Xie, C.; Hanson, L.; Xie, W.; Lin, Z.; Cui, B.; Cui, Y. Nano Lett.2010, 10, 4020−4024.(7) Britland, S.; Perridge, C.; Denyer, M.; Morgan, H.; Curtis, A.;Wilkinson, C. Exp. Biol. Online 1996, 1, 1−15.(8) Baranes, K.; Chejanovsky, N.; Alon, N.; Sharoni, A.; Shefi, O.Biotechnol. Bioeng. 2012, 109, 1791−1797.(9) Baranes, K.; Kollmar, D.; Chejanovsky, N.; Sharoni, A.; Shefi, O.J. Mol. Histol. 2012, 43, 437−447.(10) Mahoney, M. J.; Chen, R. R.; Tan, J.; Saltzman, W. M.Biomaterials 2005, 26, 771−778.(11) Johnson, B. N.; Lancaster, K. Z.; Zhen, G.; He, J.; Gupta, M. K.;Kong, Y. L.; Engel, E. A.; Krick, K. D.; Ju, A.; Meng, F.; Enquist, L. W.;Jia, X.; McAlpine, M. C. Adv. Funct. Mater. 2015, 25, 6205.(12) Wang, H. B.; Mullins, M. E.; Cregg, J. M.; Hurtado, A.; Oudega,M.; Trombley, M. T.; Gilbert, R. J. J. Neural. Eng. 2009, 6, 016001.(13) Christopherson, G. T.; Song, H.; Mao, H.-Q. Biomaterials 2009,30, 556−564.(14) Li, G. N.; Livi, L. L.; Gourd, C. M.; Deweerd, E. S.; Hoffman-Kim, D. Tissue Eng. 2007, 13, 1035−1047.(15) Nisbet, D. R.; Crompton, K. E.; Horne, M. K.; Finkelstein, D. I.;Forsythe, J. S. J. Biomed. Mater. Res., Part B 2008, 87B, 251−263.(16) Zuidema, J. M.; Pap, M. M.; Jaroch, D. B.; Morrison, F. A.;Gilbert, R. J. Acta Biomater. 2011, 7, 1634−1643.(17) Yang, F.; Murugan, R.; Wang, S.; Ramakrishna, S. Biomaterials2005, 26, 2603−2610.(18) Lee, P.; Lin, R.; Moon, J.; Lee, L. P. Biomed. Microdevices 2006,8, 35−41.(19) Vader, D.; Kabla, A.; Weitz, D.; Mahadevan, L. PLoS One 2009,4, e5902.(20) Voge, C. M.; Kariolis, M.; MacDonald, R. A.; Stegemann, J. P. J.Biomed. Mater. Res., Part A 2008, 86, 269−277.(21) Sarig-Nadir, O.; Livnat, N.; Zajdman, R.; Shoham, S.; Seliktar,D. Biophys. J. 2009, 96, 4743−4752.(22) Dos Reis, G.; Fenili, F.; Gianfelice, A.; Bongiorno, G.; Marchesi,D.; Scopelliti, P. E.; Borgonovo, A.; Podesta,̀ A.; Indrieri, M.; Ranucci,E.; Ferruti, P.; Lenardi, C.; Milani, P. Macromol. Biosci. 2010, 10, 842−852.(23) Ceballos, D.; Navarro, X.; Dubey, N.; Wendelschafer-Crabb, G.;Kennedy, W. R.; Tranquillo, R. T. Exp. Neurol. 1999, 158, 290−300.(24) Dubey, N.; Letourneau, P.; Tranquillo, R. Biomaterials 2001, 22,1065−1075.(25) Dubey, N.; Letourneau, P.; Tranquillo, R. Exp. Neurol. 1999,158, 338−350.(26) Glass-Brudzinski, J.; Perizzolo, D.; Brunette, D. J. Biomed. Mater.Res. 2002, 61, 608−618.(27) Lai, E. S.; Huang, N. F.; Cooke, J. P.; Fuller, G. G. Regener. Med.2012, 7, 649−661.

(28) Uitto, J.; Eisen, A. Z. Biology of the Dermis: Collagen;Dermatology in General Medicine: Textbook and Atlas; McGraw-Hill: New York, 1987.(29) Fratzl, P. Collagen: structure and mechanics. Springer: 2008.(30) Lee, K. Y.; Mooney, D. J. Chem. Rev. 2001, 101, 1869−1880.(31) Macaya, D.; Spector, M. Biomed. Mater. 2012, 7, 012001.(32) Xu, H.-Y.; Gu, N. Front. Mater. Sci. 2014, 8, 20−31.(33) Panseri, S.; Russo, A.; Giavaresi, G.; Sartori, M.; Veronesi, F.;Fini, M.; Salter, D.; Ortolani, A.; Strazzari, A.; Visani, A.; et al. J.Biomed. Mater. Res., Part A 2012, 100, 2278−2286.(34) Meng, J.; Xiao, B.; Zhang, Y.; Liu, J.; Xue, H.; Lei, J.; Kong, H.;Huang, Y.; Jin, Z.; Gu, N.; Xu, H. Sci. Rep. 2013, 3, 2655.(35) Shimizu, K.; Ito, A.; Yoshida, T.; Yamada, Y.; Ueda, M.; Honda,H. J. Biomed. Mater. Res., Part B 2007, 82, 471−480.(36) Li, Y.; Huang, G.; Zhang, X.; Li, B.; Chen, Y.; Lu, T.; Lu, T. J.;Xu, F. Adv. Funct. Mater. 2013, 23, 660−672.(37) Sapir, Y.; Cohen, S.; Friedman, G.; Polyak, B. Biomaterials 2012,33, 4100−4109.(38) Ziv-Polat, O.; Skaat, H.; Shahar, A.; Margel, S. Int. J. Nanomed.2012, 7, 1259.(39) Marcus, M.; Skaat, H.; Alon, N.; Margel, S.; Shefi, O. Nanoscale2015, 7, 1058−1066.(40) Ayres, C. E.; Jha, B. S.; Meredith, H.; Bowman, J. R.; Bowlin, G.L.; Henderson, S. C.; Simpson, D. G. J. Biomater. Sci., Polym. Ed. 2008,19, 603−621.(41) Georges, P. C.; Miller, W. J.; Meaney, D. F.; Sawyer, E. S.;Janmey, P. A. Biophys. J. 2006, 90, 3012−3018.(42) Flanagan, L. A.; Ju, Y.-E.; Marg, B.; Osterfield, M.; Janmey, P. A.NeuroReport 2002, 13, 2411−2415.(43) Cheng, S.; Clarke, E. C.; Bilston, L. E. Med. Eng. Phys. 2008, 30,1318−1337.(44) Longair, M. H.; Baker, D. A.; Armstrong, J. D. Bioinformatics2011, 27, 2453−2454.(45) Yanai, A.; Haf̈eli, U. O.; Metcalfe, A. L.; Soema, P.; Addo, L.;Gregory-Evans, C. Y.; Po, K.; Shan, X.; Moritz, O. L.; Gregory-Evans,K. Cell transplant. 2012, 21, 1137−1148.(46) Bock, N.; Riminucci, A.; Dionigi, C.; Russo, A.; Tampieri, A.;Landi, E.; Goranov, V. A.; Marcacci, M.; Dediu, V. Acta Biomater.2010, 6, 786−796.(47) Yao, L.; Wang, S.; Cui, W.; Sherlock, R.; O’Connell, C.;Damodaran, G.; Gorman, A.; Windebank, A.; Pandit, A. Acta Biomater.2009, 5, 580−588.(48) Shefi, O.; Golding, I.; Segev, R.; Ben-Jacob, E.; Ayali, A. Phys.Rev. E: Stat. Phys., Plasmas, Fluids, Relat. Interdiscip. Top. 2002, 66,021905.(49) Baranes, K.; Shevach, M.; Shefi, O.; Dvir, T. Nano Lett. 2015,DOI: 10.1021/acs.nanolett.5b04033.(50) Krewson, C. E.; Chung, S. W.; Dai, W.; Mark Saltzman, W.Biotechnol. Bioeng. 1994, 43, 555−562.(51) Yao, L.; O’Brien, N.; Windebank, A.; Pandit, A. J. Biomed. Mater.Res., Part B 2009, 90B, 483−491.(52) Koh, H.; Yong, T.; Chan, C.; Ramakrishna, S. Biomaterials 2008,29, 3574−3582.(53) You, J.-O.; Rafat, M.; Ye, G. J.; Auguste, D. T. Nano Lett. 2011,11 (9), 3643−3648.(54) Shevach, M.; Maoz, B. M.; Feiner, R.; Shapira, A.; Dvir, T. J.Mater. Chem. B 2013, 1 (39), 5210−5217.(55) Polak, P.; Shefi, O. Nanomedicine 2015, 11, 1467.

Nano Letters Letter

DOI: 10.1021/acs.nanolett.6b00131Nano Lett. XXXX, XXX, XXX−XXX

G