Assembly Pathway Selection of Designer Self-Assembling Peptide

Subscriber access provided by University of South Dakota

Biological and Medical Applications of Materials and Interfaces

Assembly pathway selection of designer SAP and fabrication of hierarchical scaffolds for neural regeneration Yuyuan Zhao, Rong Zhu, Xiyong Song, Zheng Ma, Shengfeng Chen, Dongni Wu, Fufeng Liu, Songying Ouyang, Jianguo Zhang, Seeram Ramakrishna, Xiaofeng Zhu, and Liumin He ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.8b10764 • Publication Date (Web): 12 Jul 2018 Downloaded from http://pubs.acs.org on July 16, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 44 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Assembly Pathway Selection of Designer SAP and Fabrication of Hierarchical Scaffolds for Neural Regeneration Yuyuan Zhao1, 2, #, Rong Zhu2, 3, #, Xiyong Song4, Zheng Ma5, Shengfeng Chen2, 3, Dongni Wu1, 2, Fufeng Liu5, Songying Ouyang4, Jianguo Zhang 7, Seeram Ramakrishna2, 3, 6, Xiaofeng Zhu8*, Liumin He1, 2* 1

Key Laboratory of Biomaterials of Guangdong Higher Education Institutes,

Department of Biomedical Engineering, College of Life Science and Technology, Jinan University, Guangzhou 510632, China 2

MOE Joint International Research Laboratory of CNS Regeneration, Jinan University, Guangzhou 510632, China

3

Guangdong-Hong Kong-Macau Institute of CNS Regeneration (GHMICR), Jinan University, Guangzhou 510632, China 4

National Laboratory of Biomacromolecules, Institute of Biophysics, Chinese Academy of Sciences, Beijing 100101, China

5

Key Laboratory of Industrial Fermentation Microbiology, Ministry of Education; College of Biotechnology, Tianjin University of Science & Technology, Tianjin, 300457, P. R. China

6

Department of Mechanical Engineering, Faculty of Engineering, National University of Singapore, Singapore 117576, Singapore 7

Center for Biological Imaging, Institute of Biophysics, Chinese Academy of Sciences, Beijing 100101, China.

8

Department of Chinese Medicine, the First Affiliated Hospital of Jinan University, Jinan University, Guangzhou 510632, China.

#

Zhao YY and Zhu R contributed equally to this work

*Corresponding Authors. E-mail address: [email protected] (He LM), [email protected] (Zhu XF).

Keywords: functional self-assembling peptides, nanofiber hydrogel, assembly pathway, neural regeneration ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract RADA 16-I self-assembling peptide (SAP) has been modified with various functional motifs to improve its performances in biomedical applications. Nevertheless, the assembly mechanisms of designer functional RADA 16-I SAPs (F-SAPs) have not been clearly illustrated. The main problem is the difficulty in preparing completely molecular aqueous solution of F-SAP. In the current study, we demonstrated that different procedures for preparing F-SAP solution could result in the formation of different conformations and consequently micro/macroscopic morphologies. F-SAP was molecularly dissolved in appropriate solvent, such as hexafluoroisopropanol (HFIP), as evidenced by random coil conformation characterized by circular dichroism spectroscopy and morphologies under transmission electron microscopy. The monomers were induced into monolayers when F-SAP solution in HFIP was adsorbed on mica as observed by atomic force microscopy. However, nanoscaled filaments containing β-sheets dominated in F-SAP aqueous solution, in which case water acted as a poor solvent of F-SAP. Furthermore, the results of molecular dynamics simulation implicated that water facilitated F-SAP aggregation, whereas HFIP inhibited it. β-sheet assemblies formed in water exhibited high kinetic stability and did not disassemble rapidly after the addition of HFIP. Our study indicated that selecting the right assembly pathway of F-SAP required for targeted functions, e.g., delivery of hydrophobic drugs in aqueous conditions could be achieved by optimizing preparation protocol in addition to molecular design. Moreover, hierarchical scaffolds mimicking natural extracellular matrix could be fabricated by directly electrospinning of F-SAP molecular solution in HFIP and biodegradable polymer for applications in neural regeneration by promoting neural differentiation, neurite outgrowth and synapse formation.

ACS Paragon Plus Environment

Page 2 of 44

Page 3 of 44 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Introduction With the advancement of biomedical nanotechnologies, functional nanomaterials have been designed at the molecular scale from "the bottom up". Self-assembling peptides (SAPs) are supermolecules that can undergo spontaneous assembly into highly ordered nanostructures in response to certain stimulations

1-4

. RADA16-I

(RADARADARADARADA), which is a class called ionic self-complementary peptide among numerous SAPs investigated up to now, has attracted great attention. It can self-assemble into stable nanofibers of ～10 nm in diameter and consequentially form macroscopic 3D hydrogel with extremely high water content5-6. Meanwhile, significant attention has been given to the modification of RADA 16-I with functional motifs for extensive applications in 3D cell attachment, migration, and differentiation in vitro, promoting tissue regeneration in vivo and hemostasis in surgery7-11. Considerable efforts have been devoted to the self-assembling mechanism of RADA 16-I type SAPs in the past decades5-6, 9. An identified mechanism was proposed by Zhang et al.: RADA 16-I molecules assembled along the molecule stretched backbone into antiparallel β-sheet structures and consequently into short nanofibrils, which then elongated into long fibers via an end-to-end fibril-fibril aggregation mechanism5. However, Cormier et al. documented that the neighboring β-strands within β-sheets structure were parallel rather than antiparallel when they utilized solid-state nuclear magnetic resonance spectroscopy to characterize the structure of RADA16-I nanofibers12. The appending of motifs showed influences on the parent RADA 16-I to a certain degree, but the β-sheet structure still dominated the secondary structure of RADA 16-I type SAPs8-9, 13. Most of previous experimental and molecular modeling studies were performed in aqueous medium basing on a hypothesized prerequisite that the RADA 16-I type SAPs were well dissolved in water at molecular scale. However, a completely molecular aqueous solution of RADA 16-I type SAPs has in fact seldom been documented in the existing literatures unless under acidic conditions (pH 2) and short fibers already existed in the starting materials due to high propensity of RADA 16-I aggregation6, 14. The proposed mechanism thus probably reflected the reassembly or aggregation of RADA 16-I type SAP short fibers ACS Paragon Plus Environment


rather than molecular assembly, which was still a key scientific problem that needs to be urgently addressed. Generally, the aqueous solution of SAPs can be obtained due to the net charges of amino acids under acidic pH conditions despite of the existence of short nanofibrils. Changing pH value or charge screening with salts decreases electrostatic repulsions, causing peptide aggregation and finally forming hydrogels or precipitations at isoelectric point15-18. This finding implies that the assembly/reassembly of SAP can be induced by switching from “good” solvent to a “poor” solvent condition19. Therefore, processing methodology can be employed as a convenient approach to affect the pathway of self assembly and consequently control the morphology of the nanomaterials20. Hexafluoroisopropanol (HFIP) is a solvent with high polarity that is well known to produce stable amyloid protein monomer solutions by dissociating monomers from protofibrils/fibrils or break up of preaggregates21-23. Thus, HFIP can be used as "good" solvent to molecularly dissolve RADA 16-I type SAPs. Given that no completely molecular aqueous solution of RADA 16-I type SAPs but only short fibrils can be achieved, water can be regarded as a "poor" solvent. Therefore, HFIP and water can be utilized to investigate the assembly pathways of RADA 16-I type SAPs. Different outcomes of the assembly process are expected to depend on the preparation protocols. This investigation can provide insights and comprehensive information on the self-assembly mechanism of RADA 16-I type SAPs. On the other hand, investigations on the self-assembly mechanism of RADA 16-I type SAPs are important not only academically but also for practical applications. RADA 16-I type SAPs are increasingly utilized for manipulation of nanomaterials for biomedical applications via combination with natural macromolecules or synthetic pol ymers via various methodologies. Numerous preparation protocols have thus been exploited to fabricate SAP-based composite scaffolds. Electrospinning is a fascinating technique to produce fibers with diameters ranging from 100 nm to several micrometers from both natural macromolecules and synthetic polymers24-26. Incorporation of bioactive SAPs with the polymers prior to electrospinning would provide a simple yet promising strategy to create structurally and biochemically ACS Paragon Plus Environment

Page 4 of 44

Page 5 of 44 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


hierarchical hybrid scaffolds, resembling the complex nonlinear extracellular matrix (ECM) with multiphasic materials. SAP stability and assembly are two main concerns during electrospinning, which, however, have not been well addressed until now. In previous studies, we performed RADA 16-I modifications by appending various bioactive epitopes to its C-terminal13, 27. Particularly, we designed RADA 16-IKVAV that could promote neuron differentiation of neural stem cells in vitro and neural regeneration in vivo9, 28-29. A molecular model for dynamic assembly/reassembly of RADA16-IKVAV in water was proposed. In the current study, we first investigated the assembly pathway of the designer RADA 16-IKVAV, herein defined as F-SAP, in HFIP (acting as good solvent) and water (acting as poor solvent) with the aim of thoroughly revealing its assembly mechanism. Then, F-SAP was co-electrospun with poly-L-lactic acid (PLLA) to construct hybrid scaffolds. The conformational changes during electrospinning process were investigated. Our study would provide useful information for interpreting and refining current theories of self-assembly and electrospinning. Moreover, we hypothesized that the new scaffolds prepared by the integration of multiple bioactive cues and architecture of hierarchical fibers would provide an environment closely emulating multiple facets of native ECM.

Materials and methods Preparation of SAP samples: The functionalized peptide RADA16-IKVAV (Ac-RADARADARADARADARIKVAV-NH2) was custom-synthesized by American Peptide Company, Inc. with the purity above 95%. HFIP and nile red were obtained from J&K Scientific. The peptide powder was dissolved separately in sterile water (H2O) and HFIP with a concentration of 2% (w/v) as stock solution. Then samples of HFIP proportion gradients from two different initial solvents with a peptide concentration of 1% (w/v) were prepared with different adding order and mixing ratio of H2O and HFIP. CD: spectra was collected using a Chriscan (Applied Photophysics, Ltd.) spectrophotometer. Far-UV CD spectra were recorded from 190 to 260 nm with the



cell holder temperature controlled at 20 °C. Samples with a concentration of 1% (w/v) were diluted 30 times with the corresponding solvents and tested. Spectra obtained after buffer subtraction was corrected for protein concentration and smoothed using the Savitsky-Golay function. AFM: Atomic force microscope (AFM) scans were performed using a Multimode-V (Bruker) operating in ScanAsyst mode in air. Sample solutions were diluted 3000 times with the corresponding solvents and dropped on a freshly cleaved mica surface. Samples were dried in an electronic dry cabinet for at least 12 h before imaging. Height images were acquired using a silicon cantilever (Budget Sensors, Innovative Solutions Bulgaria Ltd.) with a nominal force constant of 0.4 N/m and resonant frequency of 50~90 kHz. TEM: Transmission electron microscope (TEM) was characterized on a FEI Tecnai Spirit (120 kV). Sample solutions were diluted 1500 times with the corresponding solvents and applied on a micro grid and stained for 50 seconds with a 2% (w/v) uranyl acetate solution. Samples were dried in an electronic dry cabinet for at least 12 h before imaging. Fluorescence probe: Nile Red (NR) stock solution in the same volume was put in EP tubes and volatilized completely to prevent interference of NR solvent, and make sure equal amounts of NR in each sample. Then, samples with 1 µM NR and 0.1 mg/ml SAP were prepared with different adding order and mixing ratio of H2O and HFIP. Fluorescence emission spectra were acquired on a RF-5301PC (Shimadzu) spectrofluorometer with a range from 600 to 720 nm. The excitation wavelength was 550 nm and excitation and emission slit widths were set at 3 nm and 15nm, respectively. LSCM: Laser scanning confocal microscope (LSCM) was utilized to observe samples with NR. Sample solutions with 0.1 M NR and 10 mg/ml SAP were sealed on a grooved slide and scanned on a LSM700 Zeiss LSCM immediately. Molecular dynamics simulation (MDS): The initial linear structure of F-SAP


Page 6 of 44

Page 7 of 44 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


peptide was built by Molecular Operating Environment 2014 software. The initial coordinates of HFIP was built by Chemoffice 2010 software. The GROMOS96 54a7 force-field parameters of HFIP were sourced from the Automated Topology Builder and Repository (ATBR) 2.0 webserver (https://atb.uq.edu.au/)30. The parameters and charge output of the ATBR 2.0 server are considered to be reliable and is widely employed in our recent MD simulation studies31-33. Subsequently, the atomic charges and the charge groups of HFIP were corrected to achieve better agreement with the GROMOS96 54A7 force field parameter set34. The detailed MD simulations were performed as described in literature32,

35

.

Herein, two simulations systems (i.e., pure water and pure HFIP) were studied. Firstly, three F-SAP peptides were put into a square box of 9 nm × 9 nm × 9 nm. Then, 3342 HFIP or 23599 water molecules were added into the box to obtain the pure HFIP and pure water systems, respectively. Water molecules are described using the simple point charge (SPC) models. We first performed 1000 energy minimization steps to relax the simulation system, after which the thus relaxed system was equilibrated for 100 ps by successively using an isochoric-isothermal ensemble and an isothermal-isobaric ensemble. Finally, under different initial conditions, the MD simulation of the product under different initial conditions was performed for 200 ns, and each atom of the simulation system was subjected to different initial speeds. All of the MD simulations were performed close to room temperature (i.e.,300 K) and a pressure of 1 bar. The values of the radius of gyration (Rg), root-mean-square deviation (RMSD) of Cα atom, root mean square fluctuation (RMSF) and the number of hbond were calculated using gmx gyrate, gmx rms, gmx rmsf and gmx hbond programs, respectively. Secondary structure analyses were carried out employing the dictionary secondary structure of proteins (DSSP) method using the gmx do_dssp program31. The representative snapshots were made by visual molecular dynamics (VMD) software version 1.9.2. Preparation of electrospinning scaffolds: PLLA electrospinning solution was prepared by dissolving PLLA (MW=10,000, Polysciences Ltd. USA) in HFIP with a ACS Paragon Plus Environment


concentration of 18% (w/v). PLLA/IKVAV electrospinning solution was prepared by adding 1%, 2% and 4% (w/v) RADA16-IKVAV into the PLLA electrospinning solution and stirred in an ice bath. Prepared PLLA/IKVAV electrospinning solution was fed into a 10 mL plastic syringe fitted with a needle with an inner diameter of 0.23 mm. The scaffold were fabricated at an applied voltage of 15 kV with a voltage regulated DC power supply (DW-P503-4ACCD, Tianjin Dongwen High Voltage Power Supply Plant, China), and at a feeding rate of 1.0 mL/h. The distance between the syringe needle tip and the flat aluminum collector was 10 cm. As a control, PLLA and RADA16-IKVAV electrospinning solutions were electrospun under the same experimental conditions. Characterization of scaffolds SEM: The morphology of scaffold fibers was observed under a scanning electron microscope (SEM, ULTRA 55, Zeiss, Germany) with an accelerating voltage of 5 kV. Before the observation, the samples were coated with gold by a sputter coater (Jeol JFC-1200 fine coater, Japan). The diameters of the fibers were measured from the SEM photographs using image analysis software (Image J, National Institutes of Health, US). XPS: Surface element analysis of samples were conducted by using X-ray photoelectron spectroscopy (XPS) (ESCALAB 250, Thermo Fisher Scientific, England) with a focused monochromatic Al-Ka source (1486.7 eV) for excitation. XPS survey spectra over a binding energy (BE) range of 0–1400 eV were acquired. Data analysis was carried out with Multipak software provided by the manufacturer. The BE scale was set by carbon only bound to carbon and hydrogen at 284.8 eV. MS: The analysis was performed on HDMS-Q-TOF (Waters Corporation) mass spectrometer (MS). Electrospun PLLA/IKVAV scaffolds in an appropriate size was dissolved in HFIP, compared with RADA16-IKVAV dissolved in sterile water, no difference in MS was observed before and after electrospinning. Contact angle measurements: The contact angle of the electrospun scaffolds were


Page 8 of 44

Page 9 of 44 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


measured using an optical contact angle measurement system (CAM-PLUS, TANTEC, Germany). Ultra-pure water was used as the testing liquid. NSC isolation and culture NSCs were isolated from the whole brain of embryonic day 14.5 (E14.5) female C57 mice (Guangdong Provincial Animal Center). Tissues were digested with 0.15% Trypsin/EDTA and incubated in 37 ºC water bath for 7 min. Then 10 mg/ml trypsin inhibitor and 30 KU/ml DNaseI (working concentration 150 U/ml) and 15 ul of 1M MgCl2 (working concentration 10 mM) were added and mixed for 1-2 min to neutralize the trypsin as well as digest the released genomic DNA. Isolated cells were obtained after filtration and centrifugation. Primary NSCs were cultured in Dulbecco’s modified Eagle’s medium Nutrient Mixture F-12 (DMEM/F12, Gibco, USA) supplemented with 1% EGF (Gibco, USA), 1% bFGF (Gibco, USA), 1% N2 supplement (Gibco, USA), 1% GlutaMAX-I (Gibco, USA), 2% B27 supplement (without Vitamin A) (Gibco, USA), 0.1% heparin (MCE, HY-17567A) and 1% penicillin-streptomycin (Gibco, USA). After subculture, passage 2 (P2) cells were digested with StemPro Accutase (Thermofisher, USA) into isolated cells and then seeded onto PDL (poly-D-lysine), PLLA or PLLA/F-SAP substrates at a concentration of 4 × 104 cells/cm2 in a differentiation medium, namely, DMDM/F12

supplement

with

2%

B27

supplement

(Gibco,

USA),

1%

penicillin-streptomycin, and 1% fetal calf serum (Gibco, USA). There were 3 samples in each group for testing, and each sample was tested 3 times, the final data presented was the average of each replicate in the analysis. Immunofluorescence detection NSCs were fixed with 4% formaldehyde and incubated with mouse anti-mouse βⅢ-Tubulin (1:1000, Abcam, UK), rabbit anti-mouse GFAP (1:500，Sigma，USA), rabbit anti-mouse MAP2 (1:1000，Abcam , UK) and rabbit anti-mouse PSD95 (1:1000，Abcam, UK). Donkey anti-mouse IgG H&L (Alexa Fluor® 555) (1:1000, Abcam, UK) or donkey anti-rabbit IgG H&L (Alexa Fluor® 488) (1:1000, Abcam, UK) were used as secondary antibodies. Nuclei were counterstained with DAPI. Slides were viewed with an Zeiss LSM 710 confocal microscope (Zeiss, Germany). ACS Paragon Plus Environment


Neuronal cytoskeleton 2D construction Cytoskeleton of the NSC-derived neurons, which were labeled with β III-tubulin, was constructed in two dimensions by using the Neurolucida software (11.09, MBF Bioscience, USA). The length and branch number of the neurites were quantitatively analyzed. Western blot analysis Total protein extraction of cell lysate was performed according to the manufacturer's protocol (Beyotime，China). Protein concentrations were evaluated by a BCA assay kit (Beyotime，China). Equal amounts of proteins were electrophoresed on 10% SDS-polyacrylamide gel and transferred to Immobilon Pmembrane (Millipore, USA). Membranes were blocked in 5% non-fat dried milk in Tris-buffered saline/Tween-20 (TBS-T: 20mM Tris, pH7.5, 150 mM NaCl, 0.05% Tween-20) for 2h at 25 ºC. The following rabbit polyclonal or mouse monoclonal antibody were used as primary antibodys: mouse monoclonal anti-βⅢ-Tubulin (1:1000, Abcam , UK), rabbit polyclonal anti-GFAP (1:1000, Sigma ， USA), rabbit polyclonal anti-PSD95(1:1000, Abcam, UK), rabbit polyclonal anti-MAP2 (1:1000, Abcam , UK), mouse monoclonal anti-GAPDH (1:4000, Abcam, UK). Primary antibodies were incubated overnight at 4 ºC in 5% non-fat dried milk in TBS-T. The membrane was then washed three times in TBS-T and incubated for 1h at 25 ºC with goat anti-mouse IgG H&L (HRP) (1:10000, Abcam , UK) or goat anti-rabbit IgG H&L (HRP) (1:10000, Abcam , UK), in 5% non-fat dried milk in TBS-T. After three washes in TBS-T, the protein bands were visualized by enhanced chemiluminescence (ECL) detection reagents (Millipore, USA). Immunoreactive bands were detected by the ECL detection system (Bio-Rad, USA), and densitometric values were analyzed with ImageJ 1.8.0 software (National Institutes of Health, USA). Relative expression of each immunoreactive band was calculated by comparison with GAPDH. Data analysis All data were analyzed with StatView 5.0 software and were shown as mean ± standard error. Results of multiple experiment groups were compared by one-way ANOVA followed by Fisher’s PLSD test. Differences were deemed significant when ACS Paragon Plus Environment

Page 10 of 44

Page 11 of 44 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


P < 0.05.

Results and discussion Given their high propensity to aggregate, RADA 16-I type SAPs are seldom molecularly dissolved in water although various dissolution conditions have been tried. RADA 16-I monomers were observed under extremely acidic condition (pH 2) due to the dissolution of the fibrils caused by disruption of the hydrogen bonds. Short fibrils, however, were still the predominant aggregates in aqueous system6. Organic solvents, particularly HFIP, have been intensively used as popular agents to dissolve amyloid-forming

peptides/proteins

into

monomeric

states

even

at

high

concentrations21, 36-39. Therefore, investigations on peptide aggregation in HFIP and water mixture can provide useful insights in understanding the aggregation pathways and the consequent outcomes of peptide self-assembly. In this context, the self-assembly of F-SAP was conducted according to different processing protocols as shown in Fig. 1. A series of F-SAP solutions of the same concentration were prepared in HFIP/H2O mixed solvents with HFIP (H2O) content gradients from two different initial fresh stocks.

Fig. 1 Schematic depicting the preparation protocol of F-SAP solutions in HFIP/H2O hybrid solutions with different ratios. F-SAP stock solutions (2%) in pure HFIP and water were diluted into 1% solutions by adding different percentages of HFIP or water. F-SAP solutions (1%) with the same HFIP/H2O ratio (50%) but with different preparation order were obtained.

Circular dichroism (CD) spectroscopy was used to investigate the changes in the conformation of F-SAP prepared by different protocols (Fig. 2). In H2O-100% (Fig.



Page 12 of 44

2A), F-SAP stock in water, a spectrum of typical antiparallel β-sheet characteristic was observed: strong positive and strong negative bands at around 195 and 216 nm, respectively. A shoulder-like band, which was assigned to α-helix structure after addition of 10% HFIP in the H2O stock despite the preponderance of β-structure, appeared at around 206 nm. With more HFIP added in H2O stock, the shoulder in the H2O-90% solution developed to significant negative bands at around 206 nm in H2O-65% and H2O-50% solutions while the previous negative band at 216 nm became a shoulder. However, the positive bands located at around 195 nm did not change with the addition of HFIP. Qualitatively, the CD spectra suggested the presence of considerable α-helix along with unordered conformations in H2O-65% and H2O-50% solutions in addition to the β-sheet structure. In our study, the presence of significant α-helical conformation by adding 35% and 50% HFIP into the F-SAP aqueous solutions was similar to the results of previous reports on amyloid-β (Aβ) peptides Aβ40, Aβ42, and Aβ4322, 38, 40-41, which was reasonable because HFIP was well known to induce α-helical conformations in proteins/peptides42-43. In freshly dissolved HFIP stocks (HFIP-100% in Fig. 2B), a CD spectrum with a sharp negative band at 200 nm and a broad minimum at around 225 nm were observed. Accordingly, F-SAP adopted predominantly unordered conformation19,

44

with considerable

α-helical conformation. The predominant unordered conformation in HFIP stock can be interpreted as an inhibitory effect on the peptide self-assembly of HFIP44. CD spectra from all the three HFIP/H2O mixtures were found to be qualitatively similar to that of HFIP stock except for a clear red shift after addition of water into the HFIP stocks. The red-shifted random coil spectra were documented to be related to either different conformational states of F-SAP or to the formation of small, oligomeric assemblies19. The sharp minimum appeared at around 205 nm for HFIP 50%, which indicated of the possibility of β-turn structures, highly distorted β-sheets44, or predominantly α-helical conformation41. In this study, we speculated that the red-shifted CD spectrum might indicate a transition from random coil conformation to an α-helical structure due to the preferential induction of HFIP for helix formation. Interestingly, H2O-50% and HFIP-50% showed different CD spectra although they ACS Paragon Plus Environment

Page 13 of 44 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


had the same resultant HFIP/H2O content and F-SAP concentration, only differing in initial stocks (Fig. 2C). F-SAP in H2O-50% showed a certain β-sheet structure, which, however, was not detected in HFIP-50%. To illustrate the effects of preparation protocol on the self-assembly outcomes, we directly dissolved F-SAP powders into HFIP/H2O mixture with the volume ratio of 50/50, in which case, as shown in Fig. 2C, F-SAP presented a CD spectrum that nearly coincided with that of F-SAP in HFIP-50%. β-sheets of SAP in water were documented to be stable across a broad range of temperature and pH with high concentration of denaturing agent, urea, and guanidium hydrochloride45. Meanwhile, β structures were documented to be sensitive to hysteresis, and once formed, they exhibited high stability and would not disassemble on an observable time scale irrespective of addition of HFIP19. Therefore, we proposed that the presence of β-sheets in H2O-50% could be due to the preparation using a stock solution of F-SAP in water. After the addition of equal amounts of HFIP, these β-sheets did not disassemble completely although qualitative transformation into considerable α-helix along with unordered conformations occurred. However, F-SAP, either directly dissolved in HFIP/H2O 50/50 or prepared in HFIP 50%, prevented or minimized peptide self-assembly into ordered structures to a certain extent.



Fig. 2 CD spectra (A-C) and AFM images (D) of F-SAP in HFIP/H2O hybrid solutions prepared by different proposals. The scale bars in the AFM images were 200 nm.

The morphologies of F-SAP prepared by different protocols were observed by atomic force microscopy (AFM). As shown in Fig. 2D, F-SAP self-assembled into single fibers in H2O-100%. The length, width, and height of the fibers were measured to be 232.5±93.2, 23.8±3, and 1.8±0.1 nm, respectively. The nanofiber width was higher than that of RADA 16-I reported in previous literatures, which was expected as the effect of the appended functional tails flagging from the self-assembled RADA 16-I core5, 12, 46. The fiber heights were slightly higher but still comparable with that of the RADA 16-I nanofiber previously documented by Cormier et al. and Zhang et al., inferring the existence of double β-sheet layers5, 12. With increasing HFIP content added into the H2O stock, the fiber length continuously decreased. Meanwhile, very short sticks (H2O-65%), even fragments (H2O-50%), appeared instead of long nanofibers. RADA 16-I peptide has a hydrophobic side and a hydrophilic side. According to the mechanism previously proposed5, RADA 16-I assembled into sandwich-like antiparallel β-sheets with the hydrophobic alanine in the middle and the ACS Paragon Plus Environment

Page 14 of 44

Page 15 of 44 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


hydrophilic arginine/aspartate on both sides, which then aggregated along their stretched backbone via hydrogen bonds. This mechanism also suited the self-assembly of RADA16 I-derived functional SAPs9,

13, 46

. Hydrophobic and

electrostatic interactions, as well as, hydrogen bonds contribute to the stability of the β-sheet bilayer. The HFIP added in the H2O stock could interrupt the aggregation of the β-sheet bilayer elements, which acted as building blocks for long nanofibers, as evidenced by decreased fiber length. The disaggregation of the β-sheet bilayer elements from the nanofibers would disrupt the β-sheet structures along the nanofiber axis, which was responsible for the decreased β-sheet content as indicated in the CD spectrum above. The unchanged width and height even when the fibers broke into fragments indicated that the sandwich-like β-sheet bilayers were less affected, and thus, certain β-sheet content remained as shown in the CD spectrum (Table 1). In contrast to the single nanofibers in the H2O stock, clustered aggregates associated with dispersed short sticks were observed for F-SAP in the HFIP stock. The aggregates became smaller and fewer in HFIP-65%, in which case 35% deionized water was added into the HFIP stock. The aggregates disappeared in HFIP-50%, which were replaced by short sticks of nanofibers. HFIP was widely known as an effective solvent for Aβ peptides at concentrations above 20% (v/v) but promotes Aβ aggregation at low concentrations22. Nagaraj et al. observed Aβ peptide aggregates at high HFIP concentration in HFIP/H2O mixed solution under TEM41, as well as, in pure HFIP solution by AFM38. The nanofibers in HFIP-50% showed comparable fiber length, width, and height to those in H2O-50%. However, nanofibers in HFIP-100%, HFIP-90%, and HFIP-65% have heights of ～0.8 nm; half of the nanofiber height in HFIP-50% and the solutions initiated from H2O stocks (Table 1). Bagrov et al. documented that RADA 16-I monomers in acidic medium formed monolayer lamellae with approximately 1 nm thickness when they were adsorbed on mica14. According to the analysis above, the nanofibers in HFIP-100%, HFIP-90%, and HFIP-65% were probably mono-layered SAP aggragates formed during sample preparation of AFM characterization.



Page 16 of 44

Table 1 Fiber Size Statistical Data of F-SAP in Different Solvent Environments

Sample Name

Length (nm)

Width (nm)

Height (nm)

H2O-100%

232.5±93.2

23.8±3.8

1.8±0.1

H2O-90%

171.3±70.5

23.5±2.6

1.6±0.1

H2O-65%

140.9±42.1

20.0±3.6

1.6±0.1

H2O-50%

81.1±43.9

18.9±3.0

1.5±0.2

HFIP-50% HFIP-65% HFIP-90% HFIP-100%

79.4±34.8 56.1±17.9 52.2±22.0 55.8±31.1

18.6±2.8 17.7±2.2 18.1±2.3 18.2±2.5

1.5±0.2 0.8±0.2 0.8±0.2 1.0±0.2

The drying procedure and the substrate surface were found to have effects on the self assembly behaviors of peptide during sample preparation for AFM6, 14, which could not reflect the in situ morphology of F-SAP in the solution. We further utilized negative staining electron microscopy to observe the morphologies of F-SAP in HFIP-100%, HFIP-50%, H2O-100%, and H2O-50% in which cases they showed quite different morphologies as observed under AFM. During the specimen preparation, embedding of the biological macromolecules in a layer of dried heavy metal solution could provide protection against the collapse of the specimen due to dehydration. Thus, the sample preparation procedure would not influence the native status of the samples47. As shown in Fig. 3, F-SAP showed nanofibrils with uniform diameters in H2O-100% (Fig. 3A) and H2O-50% (Fig. 3B), which was in agreement with the morphologies observed by AFM. However, no fibrils were observed in HFIP-100% (Fig. 3C) and HFIP-50% (Fig. 3D), which was quite different from the clustered aggregates and dispersed short fibrils as observed by AFM. Given the corrosion to the supporting stage by HFIP, holes were observed in HFIP-100%, which were similar to that of pure HFIP without F-SAP (Fig. 3E). No holes were observed in F-SAP in HFIP-50%, which might be because of a lower HFIP concentration. The mono-layered aggregates observed under AFM might not actually exist in HFIP-100% solution but formed after evaporation of HFIP in the process of sample preparation. The random coil of HFIP 100% confirmed our hypothesis. Our hypothesis agreed with the observation of Bagrov et al., who found that RADA-16-I monomers from


Page 17 of 44 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


acidic medium formed monolayer lamellae with thickness of 1 nm when they were adsorbed on mica14. F-SAP monomers in HFIP 50% were induced into bilayered nanofibers as observed under AFM. The differences in F-SAP conformations in HFIP 100% and HFIP 50% might be responsible for the different AFM morphologies. Meanwhile, HFIP showed quicker evaporation than water during sample preparation in HFIP 50%, thus the water-rich solvent at the late stage might induce F-SAP monomers to self assembled into bilayered nanofibers. Interestingly, nanofibers were observed when F-SAPs were redissolved in water after HFIP volatilization in the HFIP-100% system (Fig. 3F). Such nanofibers showed the same morphology as those when F-SAPs were directly dissolved in water. This finding implied that the water could induce F-SAP to self-assemble into nanofibers without titration to neutral pH or addition of PBS. In general, the RADA 16-I type SAPs form stable β-sheet structures in water, which remains largely unaffected, although sometimes, they may not form long nanofibers45. Our results confirmed that nanofibers already existed in the aqueous solution of RADA 16-I type SAPs, and therefore, care should be given in the investigation of their self-assembly in water. The shifted random coil spectrum in HFIP-50% was related to the different conformational states of the F-SAP, which formed dispersed fibrils during HFIP/H2O volatilization19.



Fig. 3 TEM images of F-SAP in different solvent environments: (A) H2O-100%, (B) H2O-50%, (C) HFIP-100%, (D) HFIP-50%, (E) pure HFIP without F-SAP, and (F) F-SAP were re-dissolved in H2O-100% after HFIP volatilization in HFIP-100%. Scale bars were 100 nm.

Molecular dynamics simulation (MDS) is a very powerful toolbox in modern molecular modeling that enables us to follow and understand structure and dynamics with extreme detail-literally on scales where motion of individual atoms can be


Page 18 of 44

Page 19 of 44 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


tracked48. In this study, MDS was performed to investigate the conformational transition and self-assembly processes of F-SAP in pure water and pure HFIP because F-SAP adopted disparate secondary structures and consequent morphologies in these two solvents as indicated above. The time evolution of the secondary structures of F-SAP in both H2O and HFIP were shown in Figs. 4A and 4B, respectively. From Fig. 4A, the initial random coils of F-SAP in H2O immediately transformed into α-helix structure as the intermediate state and finally into some β-sheet structure at the end of the simulation. However, F-SAP in HFIP maintained their initial random coils, bent and β-turned structures during the whole simulation, with seldom aggregation of β-sheet (Fig. 4B). That was, the conformational transition from its initial random coil into β-sheet structure was completely inhibited in pure HFIP. Hydrogen bonds between the residues play a critical role in the cluster formation of RADA 16-I type SAPs. More intermolecular hydrogen bonds (Fig. 4C) were found in F-SAP in H2O than those in HFIP, indicating that F-SAP in H2O exhibited stronger intermolecular interactions. Interestingly, the number of hydrogen bonds in F-SAP molecules in water was twice as much as that in HFIP. This finding implied that F-SAP might form bilayer structures in water whereas it formed mono-layered aggragates in HFIP. This was consistent with the results observed under AFM. Two structural parameters of the root-mean-square deviation (RMSD) and radius of gyration (Rg) were first used to represent the conformational changes of F-SAP in these two different solvents. Figs. 4D and 4E showed that the values of Rg and RMSD were displayed as a function of time. Both RMSD and Rg values in water increased more rapidly than that in HFIP at the initial stage, and they both essentially approached stable values. Meanwhile, F-SAP showed larger RMSD and Rg values in water than those in HFIP. Accordingly, the conformational changes of F-SAP in HFIP were inhibited as compared with that in water. The theoretical results from MDS implicated that water facilitated F-SAP aggregation, whereas HFIP inhibited it. Such a behavior was qualitatively consistent with the experimental observations depicted above. Therefore, both experimental and theoretical results demonstrated that the supramolecular morphologies of F-SAP ACS Paragon Plus Environment


could be precisely controlled by careful selection of the kinds of solvent and the preparation protocols. The insights into the characteristic dynamics of supramolecular interactions provided an efficient approach to selecting the optimum assembly pathway necessary for the targeted functions.

Fig. 4 Evolution of the self-assembly of F-SAP in water and HFIP by molecular dynamics simulation. (A) Secondary structure of F-SAP in water as a function of time; Y-axis represented the residues of F-SAP, and X-axis represented simulation time in ns. Snapshots of F-SAP at 50-ns interval were shown in cartoon representation. The secondary structure was color-coded. (B) Secondary structure of F-SAP in HFIP as function of time, (C) number of hydrogen bond (H bond), (D) Rg, and (E) RMSD of F-SAP in the two solvents as function of time.

The hydrophobic interactions involved in the assembly behaviors of F-SAP were assessed with the solvatochromic dye Nile red (NR), a kind of lipophilic fluorescent dye with the fluorescence depending on the local solvent conditions19. The neutral hydrophobic NR could dissolve and fluoresce intensely in a wide variety of organic solvents but not in water. No fluorescent adsorption was observed in NR aqueous solution (H2O-100%), while the addition of F-SAP initiated a very broad,


Page 20 of 44

Page 21 of 44 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


low-intensity band centered near 610 nm (Figs. 5A and 5B). Meanwhile, a light-red solution was obtained when NR was dissolved in F-SAP solution (data not shown). These observations indicated that the NR was present in F-SAP aqueous solution. However, the fluorescence intensity of H2O-90% solution with F-SAP slightly decreased compared with that without F-SAP. The decrease became large after 35% HFIP was added in the H2O stock (H2O-65%). Significant differences between the fluorescence with and without F-SAP were observed in H2O-50% solution. All the NR solutions in the presence of F-SAP solutions initiated from HFIP stock showed significant decrease in fluorescence intensity compared with those without F-SAP. Interestingly, the fluorescence spectrum of NR acquired in H2O-50% solution was identical to that in HFIP-50%, although the initial F-SAP stock solution was different. Such significant differences obtained with and without F-SAP suggested that F-SAP might assemble into certain structures with a hydrophobic structure in the hybrid solvents with HFIP percentage larger than 35% of the initial stock. In these conditions, NR molecules were encapsulated in the hydrophobic structures of F-SAP. The morphologies of the F-SAP in H2O 100% and HFIP 100% in the presence of NR were observed by laser scanning confocal microscopy. Dotted filaments were observed in the F-SAP in H2O 100%, which suggested the non-uniform presence of NR along the fibril axis of F-SAP nanofibers (Fig. 5C). F-SAP in HFIP 100% presented uniform fluorescence (Fig. 5D), which was similar to that of NR dissolved in HFIP (data not shown). NR binds to the exposed hydrophobic surfaces of protein/peptide aggregates49. In the study by Stupp et al.19, NR molecules could only be included in the hydrophobic core of the cylindrical PA micelles at 10% HFIP solution, in which case β-sheet structure was formed. According to our previous study9 and the analysis above, F-SAP assembled into bilayer β-sheet structure with hydrophobic alanine in the core of the parent RADA 16-I in H2O stock and H2O-90% solution due to the large hysteresis. However, only one methyl in the alanine extended into the hydrophobic core of the bilayer β-sheet of the parent RADA 16-I, resulting in a significantly smaller space and tighter structure than the PA investigated by Stupp et al. which had ACS Paragon Plus Environment


a C16 alkyl tail as hydrophobic core. Therefore, NR molecules could not easily enter the hydrophobic core of the parent RADA 16-I but could be absorbed by the appended hydrophobic amino acids in the IKVAV motif (isoleucine, valine, and alanine), as evidenced by the decrease of fluorescence intensity. Another cause of the decrease of fluorescence intensity in the case of F-SAP was that NR was absorbed on the hydrophobic alanines of the exposed mono-layered parent RADA 16 (Fig. 5E)49. Given that F-SAP and NR could both be well dissolved in HFIP, we presumed that NR bound to the exposed hydrophobic surfaces of the mono-layered aggregates in HFIP 100%. The NR would be adsorbed on the hydrophobic amino acids appended from the self-assembled RADA 16-I core in water. Delivery is a great concern on the administration of hydrophobic drugs. One of the highly desired characteristics of an ideal system is the ability to host hydrophobic drugs in an aqueous environment. In this study, F-SAP could be utilized to improve the solubility of hydrophobic drugs in aqueous solution, in which case the drugs could subsequentially be incorporated into 3D hydrogels. Meanwhile, hydrophobic drugs could be integrated into the F-SAP by the interactions with the organic solvent, which showed potential applications in drug delivery system, or could be further incorporated into tissue engineering scaffolds. Therefore, the present study shed light on the potential applications for the targeted delivery of hydrophobic drugs in mammals via preparation methodology and functional modification of SAPs.


Page 22 of 44

Page 23 of 44 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Fig. 5 Hydrophobic interactions involved in F-SAP assembly assessment with the solvatochromic dye Nile red. (A, B) Spectrophotometer spectra of NR in different solvents with/without F-SAP. (C) Confocal figures of NR in water and HFIP (D) in the presence of F-SAP. (E) Proposed molecular model of the interactions between NR and F-SAP in water. Scale bars in (C) and (D) were 250 nm.

In addition to self assembly, electrospinning has been a powerful method to fabricate continuous fibers with fiber diameter ranging from several microns to 100 nm or less. It has received increasing attention in regenerative medicine and tissue engineering over the past years because of tailorable properties of the electrospun fibers, such as orientation, porosity, and mechanical properties50-51. The naturally occurring extracellular matrix (ECM) structurally consists of micro- and nano-fibers52-53. Therefore, hybrid fibers of hierarchical fiber diameters could be



potentially fabricated by amalgamating the two strategies of electrospinning of biodegradable polyesters and self assembly of designer F-SAP with the aim of close mimicking native ECM. Given that F-SAP could be molecularly dissolved in HFIP, which was a popular solvent for elelctrospinning, hybrid scaffolds could be feasibly fabricated by one-step electrospinning of F-SAP and biodegradable polyesters. However, an important concern was whether F-SAP would degrade during the electrospinning process as a high electric voltage was applied. Although SAPs were previously electrospun, the concern, however, was seldom addressed54-55. Therefore, we first conducted the electrospinning of F-SAP in HFIP, which resulted in irregularly shaped particles with the size ranging from a few tens to several hundred nanometers (Fig. 6A). Few continuous fibers with diameter less than 100 nm were also observed with nanoparticles adhering to the surface (white arrows in Fig. 6A). Molecular weight was an important parameter that played a critical role in the morphologies of the resultant electrospun membranes56. In the current study, F-SAP had a molecular weight less than 2500 Da, belonging to oligomer other than polymer. Thus, it was rather difficult to form continuous fibers. Electrospray ionization mass spectrometry was utilized to assess the peptide before and after electrospinning. The as prepared F-SAP particles after electrospinning presented almost the same peak to that of the F-SAP source (Fig. S1). Therefore, the electrospinning process showed few influence on the chemical structure of F-SAP. PLLA, a classical biodegradable polyester approved by FDA (Food and Drug Administration of USA), was electrospun and the resultant fibers exhibited a uniform diameter of 1.65±0.12 µm (Fig. 6B) while hybrid scaffolds of hierarchical fiber diameters were prepared by directly electrospinning PLLA/F-SAP mixed solution in HFIP (Fig. 6C). There were a few ultrafine fibers observed with the diameter of tens of nanometers the PLLA/F-SAP hybrid scaffolds at a F-SAP concentration of 1% (Fig S2). More ultrafine fibers were observed when F-SAP concentration were increased to 2% in the electrospinning solution (Fig 6C). Statistical analysis indicated that PLLA/F-SAP fibers had three diameter distributions: 999.4±140 nm, 321.7±50 nm, and 38.8±17 nm (Fig. 6E). However, ribbons with a width of several microns ACS Paragon Plus Environment

Page 24 of 44

Page 25 of 44 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


apparently dominated instead of the microfibers when F-SAP concentration was further increased to 4% although abundant ultrafine fibers were observed (Fig S2). As analyzed above, F-SAP was molecularly dissolved in HFIP. The addition of F-SAP significantly decreased PLLA fiber diameter by acting as plasticizers57. However, F-SAP of a excessive content in the mixture might influence the spinnability of F-SAP/PLLA solution. Since hybrid fibers of hierarchical diameters were well formed in the PLLA/F-SAP electrospun scaffolds at the F-SAP concentration of 2%, the as spun fibers were used for the following studies. Interestingly, no particles were observed in PLLA/F-SAP electrospun scaffold, but ultrafine fibers were detected to twine around the microfibers on their surfaces (white arrows in Fig. 6D). Such parasitically adhering nanofibers combined with those dispersed finest nanofibers were speculated to dominantly originate from F-SAP self-assembly after HFIP split-second volatilization during electrospinning. PLLA, on the other hand, acted positively in F-SAP self-assembly despite that PLLA was highly hydrophobic. This observation was important for expanding the applications of SAP by hybridizing with other biomaterials through the methodological processes. X-ray photoelectron spectroscopy (XPS) was used to investigate the chemical structure of the fibrous substrate surface (Fig. 6F). In addition to the peaks for C1s and O1s at 284.8 and 532.8 eV, a peak located at a binding energy of ～400 eV corresponding to N1s was observed on the XPS spectrum of PLLA/F-SAP electrospun fibers as compared with that of PLLA fibers. The appearance of nitrogen on the surface of electrospun fibers indicated that F-SAP molecules were successfully incorporated in the hybrid fibrous scaffolds. Although F-SAP molecules would be unavoidably embedded within the PLLA fibers during electrospinning, F-SAP molecules of certain quantities were enriched on the surfaces of the fibers, which was favorable for improving bioactivities of the resultant hybrid scaffold. The incorporation of F-SAP on the surface of the fibers was also confirmed by the improved hydrophilicity of PLLA/F-SAP hybrid fibers as compared with that of PLLA fibers (Fig. 6G). The water contact angle of PLLA electrospun fibers was 105° ± 2°, which decreased to 92° ± 4° after incorporation of F-SAP; the difference was, ACS Paragon Plus Environment


however, not significant.

Fig. 6 Scanning electron micrographs of electrospun SAP (A), PLLA (B) and PLLA/F-SAP (C, D). The concentrations of F-SAP in the electrospinning solutions in (A, C, D) were 2%. (D) Image with higher magnification of (C). (E) statistical analysis on fiber diameter, (F) XPS spectra, and (G) water contact angles of electropun PLLA and PLLA/F-SAP fibers. Scale bars were 2 µm in (A) and (C), 10 µm in (B) and 100 nm in (D).

In the current study, the PLLA/F-SAP hybrid fibrous scaffold could well mimic the native ECM as evidenced by the individual fibers of hierarchical diameters and bioactive signaling cues, which were documented to influence cell behaviors58. Neural stem cells (NSCs) were cultured on the as-prepared PLLA/F-SAP hybrid fibrous scaffold, and their differentiation was investigated to evaluate its potential application for neural regeneration. As shown in Fig. 7, NSCs differentiated into neurons and astrocytes on PLLA uniform fibers and PLLA/F-SAP hierarchical fibers as evidenced by Tubulin and GFAP positive stainings, respectively. Neurons showed long neurites and branches while astrocytes showed a multipolar glial morphology and long processes. PLLA/F-SAP hierarchical fibers showed a higher neuron percentage (Fig. 7B) and a lower astrocyte percentage (Fig. 7C) than those on PLLA fibers and PLL-coating substrate as indicated by the statistical analysis, suggesting that PLLA/F-SAP hierarchical fibers promote neuron differentiation of NSC.


Page 26 of 44

Page 27 of 44 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Fig.7 NSC differentiation on various substrates in vitro. (A) Immunohistochemical staining of neurons (red, β III Tubulin) and astrocytes (green, GFAP), Cell nucleus were stained by DAPI. Differentiation proportion of neurons (B) and astrocytes (C) on different substrates were statistically analyzed. Scale bars in (A) were 50 µm. n = 3 experiments, *p

Assembly Pathway Selection of Designer Self-Assembling Peptide

Recommend Documents